From ad92592adbdd357b933e27ba57f9931ef480422e Mon Sep 17 00:00:00 2001 From: Morten Hjorth-Jensen Date: Mon, 13 Sep 2021 14:05:16 +0200 Subject: [PATCH] new update --- .../Results/FigureFiles/EoSfitting.png | Bin 22202 -> 19175 bytes .../_build/.doctrees/chapter1.doctree | Bin 875820 -> 877046 bytes .../_build/.doctrees/chapter2.doctree | Bin 448795 -> 453021 bytes .../_build/.doctrees/chapter3.doctree | Bin 169316 -> 169118 bytes .../_build/.doctrees/chapter4.doctree | Bin 727679 -> 1148848 bytes .../_build/.doctrees/environment.pickle | Bin 188006 -> 172244 bytes doc/LectureNotes/_build/html/Clustering.html | 52 +- .../_build/html/_images/chapter1_11_0.png | Bin 6370 -> 6183 bytes .../_build/html/_images/chapter1_136_0.png | Bin 0 -> 16030 bytes .../_build/html/_images/chapter1_13_1.png | Bin 9592 -> 9562 bytes .../_build/html/_images/chapter1_183_1.png | Bin 0 -> 18622 bytes .../_build/html/_images/chapter1_199_1.png | Bin 0 -> 25198 bytes .../_build/html/_images/chapter1_201_1.png | Bin 0 -> 97061 bytes .../_build/html/_images/chapter1_203_0.png | Bin 0 -> 33523 bytes .../_build/html/_images/chapter1_210_0.png | Bin 0 -> 10633 bytes .../_build/html/_images/chapter1_218_0.png | Bin 0 -> 22611 bytes .../_build/html/_images/chapter1_27_0.png | Bin 13506 -> 14221 bytes .../_build/html/_images/chapter1_3_0.png | Bin 9496 -> 10355 bytes .../_build/html/_images/chapter1_61_10.png | Bin 25483 -> 25394 bytes .../_build/html/_images/chapter2_245_1.png | Bin 0 -> 9236 bytes .../_build/html/_images/chapter2_316_1.png | Bin 0 -> 10353 bytes .../_build/html/_images/chapter3_25_2.png | Bin 5148 -> 4835 bytes .../_build/html/_images/chapter4_114_1.png | Bin 0 -> 70740 bytes .../_build/html/_images/chapter4_116_0.png | Bin 0 -> 28657 bytes .../_build/html/_images/chapter4_122_1.png | Bin 0 -> 92305 bytes .../_build/html/_images/chapter4_57_2.png | Bin 0 -> 11509 bytes .../_build/html/_images/chapter4_57_3.png | Bin 0 -> 30170 bytes .../_build/html/_images/chapter4_57_4.png | Bin 0 -> 30569 bytes .../_build/html/_sources/chapter1.ipynb | 366 ++--- .../_build/html/_sources/chapter2.ipynb | 110 +- .../_build/html/_sources/chapter3.ipynb | 387 +++--- .../_build/html/_sources/chapter4.ipynb | 1032 ++++++++++++-- doc/LectureNotes/_build/html/chapter1.html | 415 +++--- doc/LectureNotes/_build/html/chapter10.html | 144 +- doc/LectureNotes/_build/html/chapter2.html | 241 ++-- doc/LectureNotes/_build/html/chapter3.html | 333 ++--- doc/LectureNotes/_build/html/chapter4.html | 1186 ++++++++++------ doc/LectureNotes/_build/html/chapter5.html | 64 +- doc/LectureNotes/_build/html/chapter6.html | 104 +- doc/LectureNotes/_build/html/chapter7.html | 92 +- doc/LectureNotes/_build/html/chapter8.html | 88 +- doc/LectureNotes/_build/html/chapter9.html | 88 +- doc/LectureNotes/_build/html/genindex.html | 14 +- doc/LectureNotes/_build/html/intro.html | 14 +- .../_build/html/reports/chapter4.log | 61 +- doc/LectureNotes/_build/html/search.html | 14 +- doc/LectureNotes/_build/html/searchindex.js | 2 +- .../_build/jupyter_execute/chapter1.ipynb | 504 +++---- .../_build/jupyter_execute/chapter1.py | 286 ++-- .../_build/jupyter_execute/chapter1_11_0.png | Bin 6370 -> 6183 bytes .../_build/jupyter_execute/chapter1_136_0.png | Bin 0 -> 16030 bytes .../_build/jupyter_execute/chapter1_13_1.png | Bin 9592 -> 9562 bytes .../_build/jupyter_execute/chapter1_183_1.png | Bin 0 -> 18622 bytes .../_build/jupyter_execute/chapter1_199_1.png | Bin 0 -> 25198 bytes .../_build/jupyter_execute/chapter1_201_1.png | Bin 0 -> 97061 bytes .../_build/jupyter_execute/chapter1_203_0.png | Bin 0 -> 33523 bytes .../_build/jupyter_execute/chapter1_210_0.png | Bin 0 -> 10633 bytes .../_build/jupyter_execute/chapter1_218_0.png | Bin 0 -> 22611 bytes .../_build/jupyter_execute/chapter1_27_0.png | Bin 13506 -> 14221 bytes .../_build/jupyter_execute/chapter1_3_0.png | Bin 9496 -> 10355 bytes .../_build/jupyter_execute/chapter1_61_10.png | Bin 25483 -> 25394 bytes .../_build/jupyter_execute/chapter2.ipynb | 255 ++-- .../_build/jupyter_execute/chapter2.py | 95 +- .../_build/jupyter_execute/chapter2_245_1.png | Bin 0 -> 9236 bytes .../_build/jupyter_execute/chapter2_316_1.png | Bin 0 -> 10353 bytes .../_build/jupyter_execute/chapter3.ipynb | 397 +++--- .../_build/jupyter_execute/chapter3.py | 312 ++--- .../_build/jupyter_execute/chapter3_25_2.png | Bin 5148 -> 4835 bytes .../_build/jupyter_execute/chapter4.ipynb | 1188 +++++++++++++++-- .../_build/jupyter_execute/chapter4.py | 680 ++++++++-- .../_build/jupyter_execute/chapter4_114_1.png | Bin 0 -> 70740 bytes .../_build/jupyter_execute/chapter4_116_0.png | Bin 0 -> 28657 bytes .../_build/jupyter_execute/chapter4_122_1.png | Bin 0 -> 92305 bytes .../_build/jupyter_execute/chapter4_57_2.png | Bin 0 -> 11509 bytes .../_build/jupyter_execute/chapter4_57_3.png | Bin 0 -> 30170 bytes .../_build/jupyter_execute/chapter4_57_4.png | Bin 0 -> 30569 bytes 76 files changed, 5646 insertions(+), 2878 deletions(-) create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_136_0.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_183_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_199_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_201_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_203_0.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_210_0.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter1_218_0.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter2_245_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter2_316_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_114_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_116_0.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_122_1.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_57_2.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_57_3.png create mode 100644 doc/LectureNotes/_build/html/_images/chapter4_57_4.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_136_0.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_183_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_199_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_201_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_203_0.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_210_0.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter1_218_0.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter2_245_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter2_316_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_114_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_116_0.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_122_1.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_57_2.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_57_3.png create mode 100644 doc/LectureNotes/_build/jupyter_execute/chapter4_57_4.png diff --git a/doc/LectureNotes/Results/FigureFiles/EoSfitting.png b/doc/LectureNotes/Results/FigureFiles/EoSfitting.png index dfbe67cd59e806eae792b7494aa052645a40d056..ad6242d0a35cbefc836811f73c7601471d51e9ad 100644 GIT binary patch literal 19175 zcmeIaby$_#*EYHU1(kA%5|V<7NJ}?Lw~BPPbayHSh=3>|ErKA@-JODTcXxNgHgQ+Wy8Yqzf<5C~i;NzvyB1iCN+ zf%fqVCj8AKRdPf47mtIOs)M4nv4gXooe@G#&%wsR+QGt9pUTO|&fe78ij4{WVW4{H z;9z6V%gk)~A0J?{wliUFjwp(Oi(IvlRI^7QunAB*qLE9$b6;>H*cKHwY=!8@aiLv4+O&8~E&DuXByKzx-LT)zT!gZ~s#=VILnk z8)m<5Zxx_Nj7bfDyz0%I&`|$D6pjAL&(H6W{4y01iQMA5CJcW$?R?WmR8+Kn;{n1S zemU_BaRYu?m58{D`leI;=>PxXe=|8vR!(S)H6b({F8tzo7W5Ave*ExZ*?1^C zhH?$VKJ}vQ$2nI#^9Dv*-;* z=ve-jYcgYxws=r=AwWPYf0vOlusfFJnYD4MzbFX)?oCzv z-Q27@S^bJZv$*q5(VSz|$(VKVT4Cy_U8|g?W8W34GRucWW1L|Zrwg9iu5%sb+b!|T z>WqDTef-CZ7>Lk_h^OmKQ+9K$f`Pj zn5OLM*Joxln2kI=T{w8Au3n3Q|L40_*%fpQUze%35%MR;7aJERjB|5FFk#Fk8!-8%gSMo=#qh8U+P? zZ{8N1)f9?~COPhuATHTVcplg;#5&s@j5#YleQML<7SQloFUF0ijZ)p~&iC)NOpALL zty5R5``0{Mmn?kSj0H?jEQc>WQ*!)RH0~C#ylj?Pd2F^`abP8QI4Sbv3Hs5b?}6#d zmy$>%Rcg@Vq4yOZf88ixSmW{QitL3TY%?l+ZC1rbi-!bnmmyTf4a@lF)QaCww z-SKMMUFta+nK=JbbjX-T7)f?F{;;dtH%p*mjS4v+Iq zr^@Tin>Q?0quh@kJ<1-kkn5mhWV{Cf^YtrnOKWR%ctltjlAK)5(C{v7gW@qO3r>xm zZR}^yo}Kv?G3pcw#=oAPF)7HXrsBhuA5beAt5DR)Ec@Q1CQy`{i&-=xqNb+QlN)uL zHbR>>#-J%i7-2b5NF!9`ryzI~T3$Yut`mP%SWAmhO>KnkA+M-(3-PaCw_k}j2E{hN zr#yfE*!T0x3Bk|bzBz@4DS2+JNCe?n|C%b>vl!0z=dq7VN>BfAe#V^OeIbJ=-f4Gt z-Rl!uozRg~?vD@^YhdrY7&}1w6j>6Yb8$@4((-F@T;NrEtU+pey58aD&DuKW%_gDq z$k(qu9?}OteeoiwiJ-IB>*Y&BMUA<+hLrq!M8S2)OfRkwNPBHL_-5v$F5W{-4_nNt zjeq=HeDLZO(K-E?SpoYAsAAiE37-3YMwE!VzTVi(L#JmcJ?PGt#+3v^OrAIuUw&`qw~K-uly-4ZajZ9-jhq%%RThZ z+WzKF{Dg#bOCN$vz%BICC*AYyh%ZPHGsT+Zmde2L9Ujk2a0o;goOwQa(9kKAzctTQCnxC_bAk-Nf> zyu7@Vq6Xw)?$}I93mMUR6{Q!w+cmjby)XD`tmazMW}@$%9+Ji{=Y)pdEVLYH+#Apd zZZkF*Ew)Yi@dHUh@*6gQ4R0*OKa1C?3u3+_zGJk+9xezgb1oY%NbW-$=ki*D!u3sr zj#s37Ebmq=HnU5?dY-@&jLMyXfgcg~x6V&wN=~;8A({`5IyNRFsysqvm`ECh&bO`& zXdTd4aPrYBmv`KYh;b9^Kd7jbiQpc!I-u)NkR@wb?n^6PFC`t!ox|4l+{;Qz`e?k^ zDN}NOOq!aRX?%XVP<*;gUsW`pl+>zPuq@(surA`XFONsgd!>55O~+=ffD1N&%l_Jk zjP|${KLtK#bVh1;hDKi0@HM=5a?zECsKny&Om7OQ>{%TB4wI z+j{Ki7|5Vi`lTiJcD5=pJ3~1Ad*{Bn^)^uypsR0sL6K@SWUOMaXRU#)D zBA&>WcD8-Xu%$-D1 z%ua0;H`llSd{?zdFrHlJmakrCr2+SstPO=vRyKsu@MPBwK;}%VL5lKl!*ltuf+n%gB$J$9O z{QC7!+hvMtbu1c7_&r|NpSUwBTU)I*7i_?)DVdca&J!cQyR0*JZ$E-37I$DPW2pLu z=)Fe#uBHyZBiVUF@N_5Mn&A5N>h&=m{=-RTmudf$$^%F5^Bu~?A-2jT9!k&98Uh9W zmb9!&+Z<^kr>!>_?_V{E!#D$+CU73DBlP&1er_KPSLV(y0xm`lTTJ3E)imzNL zWMgG~#gr4(v!eeTVTKt!K=f4E!>Nb^A1AkG(jj)QkUX}Q#7XRc(nDUpo$1Dn<5)ZU zbg0y)OA6lK6yhsWJjWEg-LUNEPd^`wyGNsTceY*~371u^+Q5Bk>IEys{>Or+RHSul zck^Cx)E7dRQRGUtIF_Ec3o=t zP31MA`+3Cg^uB7n$x8eAvvI~W&T6|6Nrgzsi>o>{x)T|>k~3uS=6w;Q5My>N%qsH% zd5gjeuOUK4H&Q?*GR$WHi!1Nnz59mKvi2*FJ)ysU1iQOuWxd9UaC^}4gy!-^K-#G! zxlF`*{ZU((Hl0e-7dxZ!j&)Qo@Z!bSAtTM3*MvndaR>1K{32TB%zNu58NYO+ zx1BAY)+NlLNS3AgFzw26{8KuFDGu~)l58i-(L5D4?pMr7|2PidQ&B%raXY$!>g7U= z^@%af`-wi4?bXL<2L(-OI$L_be!aKSKkzabnmTr0@rY<6dOJEgZgkbfF8U9>y@IrX zRW~ORP+gwgtDph)&Gqis-WR?4IqJkwy>F3qGx8;mGY;|oULD)!XMWTi&tbryrJ_^ZbA-tYz#a!hb)JGPUn=xH;o_bN%7vx^IdLg-LSKR1dhTSG$u){yyVr z@bIztxJF1_mv0k<+3pfiTi?KLty!P$LjMy~#+~JR7}U2jo*T;@+;5PlM#9M-B*PTg z7p9%@@$Un2?UVDjwBXxs^_E=0$+vGC)ZYCV{HTm!3;l1euDcVY&*`a-V8Ykadz|!x zwzCFl?4&k*4>h{!0e%W!CDQGoyfqfW;8!%fA9u}9kT-6>bTXY5`}={S7+kAxc)}o`^5rVQeHQn93q&q-ntaYofMD<5ySFv_MM1ylWK`o85O#8I=dhM-zhmo5Ed z`B<}QJ_00C=lN5I7<`r4$yCd`RMp}wN&-tG%*TJ2LzU}Ng0^_b@F-kok?0s0C{!pY z829}8w~Nm)l3;C(Sg%zak&NbirRlEa{gzT_R#CzSfY8|P#p~DTsP4Jye2|8$u2JJh5(c&0H=UhN$?zn4b`ld4>FDSZ^7D6-=J4M8LOU67EC|*4pjd!H z*)zOjJWEC%$7#Luo&VC=6>R)RW$di46fS?Wo|~IvHE6`4@H*6U+h6@nd2!g3>+_b4 z$;-;(z%b;p=>K*$$DwmZMn<%O%pi!{?0-N zEh8f?H_|1}b)E#WuDF2#&Bgga6$zgcgUCCaJ3>P7wY5)1t+-kvm}z-Du`KsMQ85bnBYaORW&JU>s}MbhYLG=^d=ko#Fj^#Lds4<+eMe%FLqL` zx$JF!Byv$UN+GCO6kMzILzN$sm32y1c17mpS8E)n{W;G`g_LTK34BgYJb#~zAO)p2}fm@ zd>5XRhiYZK_=)3UPd~+`W%;L=WZ@Y{+!K^VbC_xGD&N;@^)b~fbB`b8Qg7A`6`n3vLP+U?{~2|TG` zFixYzB!1SnM@-TNG4OFstcW562FUmcHU2~X;3nO+g}1X_YHu10t|O7P^-`Xul&SD_ zFVxrTQ~J1yRZH!-Ze5nhL1>kUvsiiPK>4GWZI2fefDD-@|7SMNH!qhkt*`dmV{#Hl zBZ#%YgAnY$ET^EUGc+MtEp=Z~X1vYX6fi zRnXSH?h=-YRz@`G2j{fyss|1ZZ~g<9D|Kl~|C zcMb{2OwJ7XK~h2{V1tv^@Z^9Z1g;?>{H<-ZyIuix`(`o6;_vScA{IJb(kY_bS~_KT z?T7p_Z*R~Lg`ozZ$C$)JmyZ-#-=n_nqU7ltU_??9N<>koFC4+NCK=lqG5HArVI`D|3M*|z zWwqAai`hy3=g#_%dhWy$JBm5W89z)`M9q6v^SDM$Q`&(pH^8A`$&!b0f>-h2=4gK- zCK4biZTyDG0@sY^a3QYo6LoZQG8lQK2r8tb0`qFxa>GhOZ;>_V3tL1ueT*NBnA8nC zKu&x|-MQ`isU%}-_`=hyEx~T|zjJZ9bx7s+aGxvj9uqpoT9Rg;ui(`#2*K?N)ct15 zQ4m-TYcS`CiZJ-G%LrBX{d?E@?)+Kp7WMiYq%jf4m_!>{Er}L7_XrtaB#a+WlMs`) zo#@DLLRDs3-cemcUUK?@0`C=S^fa8n)ZU`F%-|4Y*;JojdC9aqIR7bXw{81M z)0??^Lz*?%qg~(lGU(sRyb;v!9iG)rt&M5Y`)mr|n+h1@{CxH}?{ingBQ`cDgC;;d zFsNpKTPhhE8j_9U2k!I*h5IV2CKC}tOiWBRnllinS2keBfT)voWrn;kHc-~8y1nld z)3kgQC!dY80AEG3OJqtyoj>dHx8q|sJ9~Rp9N**P<2*OH|E&B5keOFUOTNstMzEQ* zlMym$Hh;K9KbT{HwbGw~F^Q`3TTPTEb1h*o2h$A!XfE5oX(M?ZACi!e;E-{DdW%iu zwqp6gFJGD>xTeMNM78J_B$XtV`b(;=SVrwA};!{I2jitv2s;2zB%*2zfRR?5&TL(RJRu zc{5wT;Tp6rc<&RJc)Gf}LPJBJs;j$o+r6)QEG2bSB8cn`0|QRkm{+$Q63LS%a!=`W zM@NH4-GMBvt;L9DS~5CRSl9>lvuIS93g+X1fq|Yhm8|3@KozusV?1BV%lT30sSaQ0 z(8FGxyv}erzuw}qFs>xrzUyoTv!i!_!Q9Fv)&%d zzluFA@dc|ZEkfSa)wdF3en%RSJF-T`=af$8vN}?0r0QD z=yPQ-|C{jq`sM$b?F?+PH_PUw#FOAQGE!>GW#ke1rx#jrINJcgZPTWG&Az3Y#c#Fq z-R`az>;Y39*H_HmSpUWe<#yLr*UL8M!!@j?HmB-Fy)Vv2J$Bmu#PIBV%%W#vnvZi| z*MmY0d>R^yVe|FvogF#`hWa(9JfIDMA@mOny!YUNUwS$NBAiJ}+Rp9~8X6ja*&O|b z4`0Gzg;Rr+_>LB;iWI+UT@&88ge7$TTer5>S2k0&zKo%9shxzw7r+On#h_2qj9$GT zk~a3niJ*eO-c6L2qE+T(sm^KCa-jff~9yviD zR&QMNX!*lB4#nv2*=rg?I_G$YJ;-myxWatS>!&q)7??OOGn1JfUjFuP#KA*06&g-K zYv1~~(90IVkC|YJ9{DL;xGCs%gdOYQwsUnn`qR81v?vbq{)5asRR*9SlX;>EkRkT7 z7B)GvzFMJ#WS6Ou?KQf)k8j+#@zP3%EnO09gp0B>lsv5 zqlt5x(-qi?+RDDDhBM8LMbATL)k)6u4!SlVO6Z2Rt1<9dtjGCbd&4^DLKyxXW+gP6 z{ob}dYZ`d3o2c$Ge9F)Fs;@t6`q4&h(n=4)psf(}LUqP@NWx_1I6uE49#YNnD(Z;t zGYPcC!5`#bVB&Ay)P;!`{Hcb~2_(GAdijq0`NqZhM!pImVA7($3uaBmlr|;QnwOeo zhwAC=q!<##o%)otsV5TEFrK;``j`0?T2EB+aM4p!3r)6W1st8`q2oln2hG7i@uah& z%nL8z^b>><5bkL5IIrV)9haq6fGMwHM=h&Y$DGs7)GvY64tgxA?;;PDmd;Gm{AI=3 zQG){PEamAb0Q%Z(`IJ0USAgswEKX;Zt(&#agPN>7*Tp-2K@}ZEaA(7j02C=V>c%D;DRgIA zx&)TuuUk5)Q8P+I7)D!_`lEjO(p1X z5}W`0neCQg-V=v)&3LCrY;ouzG?Is>=YsS42?mK!4zm_-5eH0{audDEt6R5jZADa_ zzMY*lv?`e+9QW9f+uxYj2uy}P;_WezAYLX|Bb=1=o5b9u`7_}f1h>3I%j@YSe0Uue z5yH?L4^_BxK~ZF6ifAR3gOR#m;ZYdOy)WUe;+f-cQ6*xkVvr|@{NNIzf)4W0O>Yov zC1qu2e?(|gGcbGv4bhtY9(A`q$-Bq%3nxSSZFE`HoruQZ4H?Y{CqU)h%dT@b2T}bt z6nBt+USVKuPX{n>x9XgC|Gl3iiWqk_!P0>H^(Dw?Sd|$4tPO4!(`Uu0+iZL{X2n%f zQlgrnd<@dXclqtv8pB19Fd%)vB z=JRu~P?L}KRp%$f7kYv%1#J?HOYCnceSo@-DlO%;<^b`bXP=V0h1A-^k@{4X5$85x zlY$lT&rKSAYJex8r2t3Kwd(>jEi|xAsOdRG^@0{6Xsr=zwQ3`h!Y>h~0|&IOh698G zOwW@zHe2Yj<)S%rKw*pX*p`Bg3h{91^5q=2Ju^sjfUbUUMcqPfk#KO}GU&vKxc>I9 z$s~tB81}#V=8`!RGg48Zv}XZ=>>>pmqh3i6_lq`d#h*x6h=s9C{_{ z-@DvHG&DB48RSt@3%1R$9vY4b+Bt>l5@Wv7qn+RWu<~<>LDVBKsNn-}n=?(zhaKK_ zsSH_YRLSn{-AxB~9y}E?7V`+Hj0Drm1KN~O1)(7c17FgBG3oA~Ryd^HQYq!49$!^5 z(cOXXLJb8OKW$f{R$#_uE;KA+;2S;ZR3m%7rOPCaV*RvpVOhOnr5 z#)l8;f|?<9IjH<~eT46WzkwS8u`XK~JCX+*HZX8GfUMu)kUWXzw2TVxh~*nVZAwk{ zcY4{D8mu~o>rHcZ23NCLAC>TlU^%d(zJ~?h6M2sZ?0CY$*)fE9vTS5+{P`|zsl!q& zFl5ro%7l0B+(C@GZ&Z9F=L`SyM;4@eH7K?O_hnH#8DVH*!sfiL`713AvXogVWeCs) zHvL?=c1jdj*MiV76P_h#Mu!W(5cLCm+RU>p{jeAmgVd}nl$Hw0bx&ufu#L^8T7?tj zEEKl^JW8?=pVR7O{KY9Ppi{Zl<&~94;DaEUe+PW>=PRqP<4S+X=g+mi_=3%f3IbyK z`qcLGfAqGd8)Rd73rZ%E{SHGgu!3^klqd=~aES@AD=?i5Eu`@+l(X6%+@;o@4YOVc z?UR_z7zYyexBPrSz@d4SCp&=pT_NtHYUM308KL<=6@a9alpWa0(_1csm7aW{^3yOe z;QPle3_h*iHX7>fXV=BPs=Jz0 z`H_-5Ym=t?wd>r-N%X|&yZ;8HD`$BgFQuSn(oOT(vl>4njmvzb3CJX@8if{mzdzp# z4G*8T;f{(U0wG^Y>Nj00IUd%%d-vK7=0)qRUyx^+{2tNC<5;;)5&Yh--7=DsF{D!9 zJ22X!OrKQbU-kVLs17~XWO?B&BBG|hOkUG}W%Bq3{tJ_rjqP@JVaHt??dWc|4q{>7 zjpK_IsN6QLoO^qLsu&iU;iJC2p@+)0j^;9p7ySU~Js_-r%rs~U;yC*@SY$onYE93@ z1xt(mTv72Afr>6zX!P~%89sv$*oZ;b6jm3XY6O)_! z>)+ONeBYtb;0{&A*8->MYnxXD&=p z7aBd#PpuM)i@%<1RK){Ch=fpz8z=;wGtwk(3c5dbPR?zRc%pL*njZb{pmjU@N&%$l z4{4cSt^mb`Vin(^Ijr9LouNpuG%RKB>YvZ4&4bSFw)+xAlD@+wg&GMK8zG^k(7_T( zNy$mTBh&NI){S#^tYFo+FBf&G+P*`e=|5z5_E^GbA#cI|gz0Ra$GgVv*3{j{#P1&W zhx-~@3i`Fq)!cY#aCmF1#86l^jeG7MbxJOZ=(7S+3tIu2;b#&CC`(ohV6N@BEv;JAdg6$wRMH3ebw*PIIf-~bj#LnV?>e`;%M zqiXSm7usX5;^6F}P{Be{dQwyvy|ex5qQ8_Zxd2Thdooc%4H{G|Y9dsje*XN10{_+} z=>7n?=2#3pLiIfDv3zL6EV@3Zig2X;PVZB}5tW+8IbWyi;M|zx?ZIalv0|s6RM0!2 zCAw5?67T8My{(pKtoLuAbPLIo5%rLkUS&PGc##hyw&@^2oVDbiB`37uIMrRF%gj$&!zLJuBw(8;_{pYJH=HK%CP9uP7$-iw?^gfXE+R)aX z-l!FRIM869%h>MTQHR0L-qq&0Drnx%0Q^Zsjwv#v=U4jAxixQVlaYRE_^>74Tl<|h z5y{mQs8+KtU?ek*fd*5xZhad|OU+;?Lw)kv*K zPS@*eBbd=zaejvnvSQF>6clI$1Y#@P54PL+Hf(z2xI(^uZO$kfZ<(y|t>{YfJ4_fh zqovGfndwZJf$)5Ka{YeN{vFO?q4`HQLUtAkScM@ILAMH2hfRqlv}ao@Iu}tWXT_v1 zzFV3aatni&$(q@q&g61$Uu8h!UswE{=r4#vy`J*hQatD64rv{QK z0t?*|hkQIN#6K30j40Dr2yg(Qu%oJBrRW|!_y)-s1^?6|KWyLbPj5*~FDGPqNkHCM zZ(HFX)VsPy-OUff25bdFnf2|BJg+l1baeEe>FM=zcCLw1NYl$8_%Roe&8DS`u(Ox4 zt|}kn2i`l%R(u=kD7qIIk0*5DOJ@<#bJ<)+qya`j}Zf ziIuT)D)>IRDJeYh#rVUdAbM%@IpQ{_1##8cnl%8y1z@RW0=dSyO|4jvrm<$fV$+Xa zD)ijahMoL8b6Kh??ingV>~3U{g&)l^drQF%o|>8h{NhvTd9n=~!Znc&xrIi%zGDBi z=)52d155tn-qL6hhEOc$3n)zsYBAKXnS!{iC@+t`LC8Z64p6}vZ@f?~OG&NIM=j0zFIt2lMKC;j@I*1SGFOSReR zi&59x24^X5&oWR8cF+s`m~V@2n*IG{X`)I9XeUxRKR^E<@(hE$U;KCxFMRc`cXIlr zIdYjGO&^y#U0XE~I*H`sWTd{o?SfO;!doAqf!E!+*a(+pWqHS{vP>?*eN+4U`pR z;QL%$2jx0CT3T&MQePm@s1T8nk$_Iw07gKL?A$G1U9~{j52T#*VIqG?k1uo1+M>q2x1KSCvAKMFEAt_ZVv&s8*v<(0v}E9=o;^dx@;T4v zwleVve>U1L2DmJ^y!CpL=UFkEb z^`lq-^U@%(gQ*FV!t!f5XGcgz-+-ACI2f&i1`4YhK%NU&MWgx|A$VtJXYI%5otYCa z!W;B?7UqbkAO3Q^7knjtEWxGlVJyhrEh2#h zQu6ZmLHS5Zx(f_|EvO4T0(C6S<4U!p7pEUiD{c|{B)VTJ8vWdC_kx&Od+Bck|GlOd_VXT&k^pc)t8&@o7q1}aIfthHviw=6)$YqMHUwV=xv zp*0?uaAx43t}c2Sh57&_XpLOFt-g6*cFxio{X2!uH)q6~2yT<TrLe$AhW_eIsw6KTD1n?>x_A#vqBFiw^lW>2)n2CwA3PN*v zE}!o)FMG3^=qwiM3c`SZ_afh}4to&?k6Hwc_OA zp@WuSfs#C;miNu0{U~jRpsQRZkB}kZAjl@Y{rSsjyK7nC=7ta^@!)F<%}3?*b?+Ry zmzY3W05N%b1{hJ&D(>n5om(?bvBfD05r+>NKdwtfe-cJzBhwJjL_Ji!9^YFL{GBrI zB7m@@enRXC%cjp^n+ik+2;N(d`}*Vs)$`teueFwUzC27EbFy%h_VY}6!IQ2ioSDq^YCQ;nFmG|QYM4URlX zHq)05TU*EI`|A(uY}~U|861|7zxxv1Zx9k1qT+U&90HT58GLot^z&(LShjr}=+}|A zT9j1w?kax?2e$q0WvUSDZ08GPe@1J{db7W+U`+xV4(WK8(&zQ4-Jw&A!omK>%7<~A zg|`=IsJoxTtj1i97^!Pa7g;_0btSmxvZ|j${vDnlv^lXHo^SmmWxxcojRA0?Y5kF4 z$Ebox7=ce`wlH|s#H%j7Z#)A33`$j>NX%X&Ug@7xR8qze*8LXyN`&{3g(}giQ{WB~ zY5E)D^IW1Khqpg-L_%tcmEqm1KI0!IS(2Z#R;46g`V7tagMkwUdn7KWMH6pG`1F&# zzE57Ceu9AoUuNkqtdM6>MKYpE9IyaY3wG{<*>hB7g@jzVHdd4*EhtOMFh}j`Hh5jI zZ~D|*yW;p}IfZ{JSV9E$;UYk$H5@du-KG3SN&>35G2?xNT)bfHhY#0Q_X+NGL~{k7 zogC}{(C!%=Y_=>KYs@$8L7-UwA(QxqC5K0PwU>-L;w+7qrVViFs7!S3kv_uynZ<95 zwE(UUjQ0-us?Uimoa9s=Qg;s*_#q&^PWIPk;P{v6HDRE<{=^AHJ5P8M|3^f3lhfhrh;X1v>sxcXU}*oi*towVwq1 z%Z8Ww%bc`Xwe0<>r5G`;Rl2I*dT%(ZfjhxOnlC1IwCn^L>g-^Or*0)7(x=d3=n2SY zG!j9V1*gZ7y zjmA~O`aR@DJQFQ5JW_}~ojYZ_yX3CFd@Z2iyQeO-Cl>1T;XmgSyH~w;Pks^FmdGe6 z-Rcm!;0K#0P|)M<8v>{UbGf!N;t%QRQ`6E6`hF>ar;hEvq37DXu&0dznebPjG3^`3 zYaaKwUbpmJE?PKKcD=n<<6Sq`9d!v$@OnuFaQ|iH{(a^Jq;nv)G~JedpaN;BCj}Ly z(1kQZyNu=nro68l=7caHI408=OliH%ciq~yM8I&3L2o5or1+PDtn%#}WYyJoVvA=# zw+~{ln?!)?rbR!-s`F!vC^+bXn+&ooo#~>hxFu(D+z_wP*Y4u@84qSNT{^42xAt%M zwY4B{ZgA=&U5pIv#!+|^%V0rDOb0mPb|eLeG11UC-3}aC^hL?;B)l??cmUQHAM{W8 zSD?0j@qQB+PxM^n*?aO+m=YSR#ob7ADxmy7nI97$~i#O9IX`rxzRZa6I%tKz*XJ$_F zZ^@_ibE|R&9^y{-N)wI6388L57B63}kBz1uEMGztggH)JqNIw}-)1B4Wtse|iad4?bP$OqGOD+Co7f)5TQ#c_fY zE&1fs?od;sTw+X;;GtU3p`bRAxVl~{jzGt-tUp+)-kj2c6DWCk0e*g{r`~z4H6l5m ze74#~SJyL%N%de*e!{B@hqV9g|^le5Mrc&i!pHm-;(mITc{ zU^I*x^!E|=OhS2jmx{t-!s?9rEC!3?>rR4SHzBCXEm?dC_*lWJZo!#Wav8=RgK?Sa z=<(4>@b(eWNh-f4?mIM9Kfi1D=u)vEPgU#ZeAW66c2FKg#AihSGfwm2J4*d`emyit zL)GBL-teiqnZ3rgHL&kou)hyq7XqU;3mOp=Au$t2VR$_uxDNA5Jx%N+qY4l!Z?!H zELPC$x^?}AaIMeFceIXoBvl-sywU%du<>eS!MTPVlFy)^9i!LrpTqP{!NUgPh*d2r#A$vHJ`uTG_%)*mF~m%npM z!Db-*c%4tnUF?yEq>29U`zWnsIH;kZh%mboM43S=loc~pVnoKXp42thn5~v0eyPV= z7$@SsXiuTsx4+}ezj94j@DB@)^oO84+LVuHC&6~*$!JAQ)mL(ngLghA`0_sDuJpy^ zggB0&xlBdbJ2*M7TZ6RAlc8fwODN4mVr;${^k2Lw-(4NF>u;XI_xF>u@X6;YBa-Lq zkDM05?PpcV3-+vVqAx4BP-&G)eClO>SG@SojG~T42!2Z*mKx7|zU>rYpWfo1*A<9- z>HG8g@vc90sx^CvrEeRPB1BK$%L24TkO(K|B6N~-azaL|$^uP$k|VXc>6Xn%o5#Jr zj>bF+>SLdDa^Ji(a@FAtTapgQx`|hC@mR`a;dqO3iW26x&!2Dnn}F8Ur2<eWH@KLaBHJlbPEdc-bI(64Ot1|^-d-gPj~A|--#b$OPCBMdyvM`G63 zV8}y&{)IX?4mFOWfC-x4F^zWb@b%e4F1B3v*K75=FFUedkuQz2YRX24WSH0xUtid9G1lrKRLyV_D z9)QefilRNF9UOLDt+W2A@lPEy-JC2XzK*r{$nZ{?**sHLcB~l*wO_D#leQt3^v~M_ z{^g`U>)KdRpOYMUmMEU{KNs+5Ekdo*BHF!Dv`b4Z(3>C?3x;WgQKzAgcb7nivjZCN zk;^8WQUoI|lCtX|#&uzkS(lcUj3Xd@g7-)%4jrBmd4%df|Ic0`k+Cn|+4kE^;39IC zxp9AX)D&f__&W(6j+BRHTI0H`I5V{=JVbCC)5i|iWMLWpzkd-$`78vc?sUCBN>qo# zyECDC|Fiow;;K@imB}!c_hB6+v49)QBbbhnso^yNg|opu2p8vjf2EjWxHn@-6Ye z8H+|09Zw@5t>Kt|Vk4<_x!#i>83RMBz5t(zQO*jfTZLX4MOGZ3R2t#{Q_gKOSxvXH zN4}ZKwpS&1F`lPahdDzlbmTwfz#I7P9i{=SW#ccUGm|TX14ohXD~<3n zYL|aSojYD!T>Os*{?<_i{+R#U37hZ9d&j3In3t)TA22dDOxdst1O){pCne3g5Acx% zed%v~%hx0L6V5a1RC%G)m2$8nkO_IkfW0^BnOWN2ZjS}!!7vObf=t!3I;yOnBDT@N)T#y+t9G{v&0@AaTQFW6g1#jU$91zu)<(Tf*3vbK z*7d^|2X@I>A_OmpB($|D;Cv|`yfuLZG$CvC%zrNQ`d?@CDI&td;~a>Oy$9g1zUPLX z8IKTz07$7|158@V(IbXeJp`yH3iUWwYr=x-fIXvV%!w4z+Kf3cprb-R_=z~w>OI^8d-nyY?=0nTR#ff|s-<ahbYIhKB{_DU`v_xZ%UC)1wY=W6+C;)HD6^pFm>lW+oHL?fP0#)gTD@b(Ms+-TGN}Vs8(n}f`vzhN=Ya~ zGyiro4ZPQY6*?}ji<60-Of7if2o^YScr9_^&m~Mu^y?H9@{*E%Tobu|`MQ7a3QZFQ zQ`M;*IFp4@feCT`<-Q#6oGZL=gcoHmK^gG4#vBxTww2P9w?N1xhZmGU=zz||sUUR5 zzO}VA>belipq}psbol|6dGqhzQ(RR?xZq|0H4DySMH+j@Q+M;C=H~Lr;Y=`EF{}!L z7?@gCR@T{th4?*Alt1vV>kr%kAt52V8xxd1;O&5JvXRZo05WFzRwLwO$YEH zuz=^Vs@PNZ+6vIo9G$M50r7|g^0E;;5ik`OEE#ynEnHsoyVOL`DlcF)9ebSZ4G`r- zZ5+TvkFDMS1@I*t=+*oLdB;3E&o7^w7akw=vL%WC?Kd&SUQoJ{eyl5O;Ae)HP9dbk L$@s02Q@0{U~!Crf83Y1Bmy{6s2!T8jhCmR4pCW_b zuql)1gTHv}MV0LptPJfPb!-hFGCKCw=2rIRCc5Mf2DWx4R+j8c;D1JPV|#mRJ6>jH zi~qiW$;#GCDh$BdV4{t*seTRRC#@R``R2pFVRgIuMZ*7gknU)6;MB>1!R&Z4x*|UoX^*KFn|Q zbTpcCL2+=__t^STAE&UxuYK-yUng|LM1pwm@xc{bcw>)ND;M4P|GqW@k5d-z&nu+k zxk9!^vQ9RH9tgc;;<;Gu)}?7=;#!y6f@Uh_8a#z2G+kds(#oxa3wSInEFeFA{IFgL z;bt@F#$E4ElzR8hM`=C?zRWbq&xF+~EDd_1Ue#XAyWXylJ!p4DP!qql_Iix^{BSvl zx!81!{(N_;<@#&~q4MGW#`$u|?`ktGj0MdnLn`{wA^4Ha_1s*sP_u8b$;bJq4ZrTH zoPVFY;R>%}mMuQsvO1}UiKyjOMell)n&C;)lY^sf2G7NxO83_DroRgdH-7nfJo0MM z*~k*AyWrc~h>#NV*x{Pc;Eo&~j$qFb6%ko&xcUUCK5RyHzB`*JQqBML>C>WZRd#+p z^{ZDQqobel6dBl!`t{GY$8hoSM`mZG4Gn1^6Y6&8xk`nZWo17tX3ANek5?n391Vo7 zx)5v6rxz~I3O%LVYC|S8Z-@%(&bc$?8kTD(J}N2o{CZafY3-O%F5Rp;(QMf=1&83u zO7Mcb)yoT#3;S(Q#$sovojA@e7?K;+qaJl3=FauE2hkU{QN2Z`Ln2#(`&`t zv-Gul^N0Jo+f8wyR!e51>C)I)hnd7mbtX->ql(?haG#zv!|75&w}%_^Q&G^+mZyx8h4ywH79MUmG5iw!NU!O@T9KWs&vo$~ z)D9f2Wbj|DrhWSS8Jn267ZHPXJX8;Qy-*fsxUQMk< zFHArhJYZ|9&e+5RA?4Svq^zWCt=)55Vhq06*cO$_!=gx2=k4F3dlsBr<2aS&flIjLe zJKmN*%DuFELX3?5)zvTNU3Ok=@9Yo?dGKG%o>0zvKD-0d=lpPY0miXZK4YoZ8xzjK z!ExAzPoz?$%cGe4)A?k*$YJk^zLAmLt}=qzME(1V{W<58K5n_2>kiTzYP7(@KVwq{NiOh8dl(dfj)ifz>&WyZap0q`n!+=At9 zZNB`+Qf33jTHbFIRTC_XW*uB?sgyK_PlC(y^UNF9Z1pN8%9zcJt=V*2jZIAxgdySK z@~dk<54W&WYQ>bF0XSJlOMYZ5`kfNGy1FRO z0-G%hZ>A^0NQj0sTRbO9zU(MIXLQ1?sSEW+WY%76M1@q|ox3`njZqWxDGJN>N4K_7 zmrpz_93X0sWcJ-2N6YC}OoS)RCtJ>a%+52{KHW5j!OjnB7Vf-nZ~5sNa$nNfEIK?X ztsr~xVrD+7Mc(DTn(h6hIdt2R8ALxVablnL_UH3zMPbie5tVy?{P=`J;I#jc9^aRV z@sPHl1kOG^J^gakM$$?j_X3jk-gH@S9A}VGlC$sEuaBx75Bj@*GE|*Un@LJaQoMZW zpO8Q-7EI7|yw=mz-3=or_X?*JhsQ9({)DOe+2lJ%uiFb1w&J2zJ9c=ddN{dxTOaT)VN^@Nh5qK6Fn!&Nq+?;8&fQrWCXt^4B*@c4L}6{Zed z*5q@vdC!S`hZ;F&t%KT==gA(`p$r9;aXSSy-)n0rvvxNcz3=ucWRmy^vr6`eX0V}0 zpYiZQY8!OP(9dN5#8uh59bljc-M44CF!J%mYt%T{jD1$pDN>H%Fu@+llB4C~iUKEK z$)2~Xzu&O?2OWe_wbcI~RQ%z=!Nx{!#8`GC3`iLN@l#B}OLFGAQ|zv;F8-U{64Lw| z>48D7gWEGs-ByWtS1TTTtprgiDOAU~FGNCjTkq7Zn_k`@M=Mnxcjp!<=jExMk>aac zZmPFWd@rlbg4z$dhGLryq4BLs;HdZ zQCqZ!MIWWOW@j8HSw#3y^xe(HM|E|i#&QcS>o&Xz7L#GzP*Q z*Aco_cQMarFmr!*bJEXukWkD|P5lqGRFqR*d%nksg@=X~ly zv9z?rsBA4HblJqn_?)56_2R01X4CdG%QZbCV<=zk8D!yR-u(G1m%xg7 z)wjmR1U^qNGfGRp_4cApn_aar-&fS-FD|B7SxAYrPVC<2{0K>2TXCpD34OM{o%;kz zM7Xv#=J$$V7Y-Lvb>@;T?1!YNNeiI+3 z)J6R{kt&wjb&K74=2si&;P~${B{Neoig`gQqYgou#y9x=((sVL%L3-LQt z9D^=XI_&ZEXP!A%YFNI7`BYuUy+Qg`scf_QuA7KNxlge@L-TF&CcEctU_#Pgl!~d` zKoKUu_r8h6UovoUIv4rlm}-}+-}unvw1|lIYlUHET0yGunG6S%zb~M<>QLkCIrb(n zG^uigA>WFBgvP-7&oxUjzD2u1<<|40*7VYzr$xR_RnrCU5VC`e&CORcqLYkv#`}<- z_+dSN-j3<%^aQ@w1ukZrz|S3Ya`pwexhqpfo^9cjeh~jKIqmz~8%deHrH~480Gs3D z)SMlsG;G^e0yY!R|9zJh{fp0Nauy)YeK#aI?-f>(ejh`f>kyF z6Zv(kl>JXdxrg20_JSuwD!O1Xx6q-;{YDXhkHS3VBc_IjyIke`Yh!Xyq9u*&T~gx#F!zUZ|YgiZVzzq}ts%6-=)yC&O97>5={#`p2GG%#WgV-|R< z2e6vDqOPPt>#?eNzj2y6q7uL>mOgIVbU}N@op6FR-QO?re_}Soj&bi(w_XY zbhX<6dy9lOOw4^7WIJos$M-n@wH5&>z+s=f-TVg?;x=#Dkp@QC5N+A${B z@E9j&#nJeT>r26p0bTmK=8DWrq{*$vX}K@6U%@)DJ5Q4$2@Q$T5A8@}(OOt)HNF+* zW@mTo?Z;Riw-S;=vPxow?vF&dET+BMbSYr4h#AY8k~^+Tuv)rBx3mXLD#u&^pKT;R zpu3*CiAqR3)zMi3kSPoVKi*nFt>xc(1SypbC8os9pe?!hjcKNS=9HS31ex*Vqr&by9(Gw-dckZQ}rMhLRmG*AD=Z9%-nr>KLI^|8llBC{K>9hlv*nh`; zV0=Kqw<9F@b!Zz##o=Y6Ffuh%`m^{M-}91xdaAWV7GebFnG|H-{k<_Gp>NZM8tPPC z{?o`AavH85Ly;?2z4f8(<222P%`pjrq2E5BJDh+1J8Gj}eQSnmaq5AXpS7DsE=rAE zLRpK-?MawgWGW7+Xfia!{5e}^GP=1xK(=A;rKBpCe9a`}4jy~yRGJtYYj2Mds=FO! zFlyanlDZu(DQI4_;xy=A9zp?cJxQ3hvb0=Zk8hAuS0`q!zohM5x2qX<2k2go|47Qz z)YLyPPJV z=xIuXL}6r)w6iV9^&-S{S=DOQf4dPqPGO=FoNqTPyvx5c@@whOQ&Fcd@fA2_I|=6%#zP9~c712kN&F#Uf~swWo_ACO0v9(UmabbF(KdHy zC6zb(GXuY6py5^@6^9xFFV06)H(GT$?V8#oasS!R$QuqFGpA5nap%PY&+c8a2KoC5 z=DyKTYL?T|KSE%mG1vZy2_fRfA|nf%*)s%l0=d1vUMv?*~CntxB`vMY#|A66zf3dDT! zkg&$_K))*@gM*XPX2ycsri)s|e5yFW=gGy%aZjNhTkPe+!`)-EaktLZt`S+2hrNo0 z<;$aw7$n^W%dK=-d0Lf+M!0x*iRI@)w@aVDewF1XOUug(zPvXl_Z_}x7`Ln|I$Hs@P$zwqDfKTX0JFrM+$dSq8KS9{p38%oU6>kmzvtaCNG;o@^x z{hla*kvsNYS{hBST^+3vAYK-aTck&i9zA{fba7a!I_}q*&(|)Mo(bGvvK!sji;C3ttvQ@5(TdYeDq;A((XC*CF=wy$Z1W2b zTx{*)dR3E`O!aCw-g(Zt?3mBH*`fF#qnl_MvBk>9P2DdTsq`A2m>AVM)@ZmLxN@_Y zjmtnr$HvMU))&;7@Z|)rUY=|$$l2OHlZv{%D*ckUMVJDol%clZ)8|2tqkH~o(E&X3 zh|%SZ$ZCz5xuwBW4&o>LF1yxe~@~Yvrl-_F&$ffbSd| zzzi9gkWZgp`v(OH{mx_;K*YttkxCYHzq(uri{W>ELFRc21$VQ*k`n0;3XC(1%AX7> zsU{3+lRgNi!1X)$_4{{hXQMSumX@rWL|#SD54&{8=szKZc2F>z>ub?k&4yAC*;&>y z&j-Qv-YdbPBAdO6zAzG36fO6%cr~lK@Xk(&NJbU55@PB>hTZ-VGn%V&cP&#f(Dt_`vb6J{jZ6pOvg-xD`p2=0f~BPw{Cq z1}RqD-L42?`e>Gyo9i^%oqw?lKN1cjHL1Pc?x{T+CnE5Tr&BPr?>eM}9eY;IA)9fn z+c-_wNL%`3i!ek{P~%o8>T#+OM!TT6r@Fe-NN)l3`bbz`A9q0TrnaI73(8ZN@QNv_ zP|%;U;TgauZe-bV0@6}usPb>1(y~Ny3GFxbl7jeNnodiP1#b`EL5fRC{;D$XFa<

s z$2R730RNHytij$hzaSTx28|?K_T%~O-}{97)L&xhm~;wB&8Jo))5%{h<13I$i*#C$ zqh6``cIhvB-fwtbf)(sa5^%lT(Rz6Q_U+r7^1i-4M&&r;eZ?80@plbaEP+RxrxWQ8e^=^_rDl#qj)fD|~l;Cc{ zf3ARIOxtK~V0-;qY-h>WqbaK&VxWI(G-%=BA(*R}zij1sLj)AxwW1DEN)C=6KtPNH zd&zHb@Pj#*+RZtTLf|^4rl!ZRZG4}YiprF;I~ky6uIiTDW`YANok|8Vv&c=7vg+7!y%Dpt#ZRdt!4JaF5YQuy`5w%wA5MI*K`nev|MT6B~alf|5Amx_;?+K)DWLarx zqcYV8PK~c=cVE`l)~Kkc5{hRnE0Q@JHfs+3HXv6x|Iy!t&KuII1J)QWR`Sf7^= zK8n5jM3Tux5c@g-TCOwtx$J{@kMr-2=rC^~fhV0=Otw)`Q6M@xmKYDgfINti@RT3K zWGYThPM`-KUo<=<2H?=LnNPADA0NMb`4UKVMYe12i;IgT{#&AgL@{g0Dk?^CFGPW@ z+x1KDGo$+rL&8rPf?s_xVz`TeIg|jSm!1X$OZW0`vz`iTo`$v|lbqlIO%IQK__%;T8pSppl*!Hd^OasHSA1|O%LQoq_0b|;(A2eYS0 zX=?<9dIJbv*F$#Q(6_eUa`rc=S>o@P+V=h)Q)V>0ZaZBc-p7Y(pU*Em)VrbGZehpB zB_e?05)VGQ%qzy5Lvl{hvdaAY%4{{$51_bvJsj7;lDk!6;;%%IIPabIGw-$S6aYzy8x_PnV1L@gt!lJu}pSmo~(^puf`UhwyvU(Rywt^2Fn2u)boL$b}92~JIBCvc0PwUh6q7IeMrp)Dzl4sA-<& z$kCa>P1|jBuhQ*CXvIZD+&)^q!6Pf=5iyez_}t}(B4vw%hu4%rr$>M_GcmI!Nt^X| zaByJeq0(Ex{LG|S@nd|PxcWyd8>kJ4^^pho)Hy=i8d zTV^7852rJ=`^qM-as~M%^rWbN&S5}xO0=7KH{5tiQGWelLdLd?ho=NW3>+e-c9Skp=tiZrk z-)C<-{P5p|3`bO(Ja^wv% zY2?>^2!AH1>FO2e@AEn>Cm;-UB|?+3l4a{rFHK2B5022^B_`=+K@*&_4JOedNry4v zf1TM3$L%N$K1vtObgAKo3d8Tj`@N*C?KlkM`^IaDFK-dvP&`LI-|QNlUkN}9K6G$~ z?fpHJ?_TG;P${=%re1gFXyf(wM_Z2&Y^kC$C{5aC6#QF%?wL{Y32+mNO%SlbhS_aPbXVOG;JMRj_7> z$p{UX`DA0UL60)X->w(?25a3v;qVAw%PstKu40kV0O@$47G~q!#lk=frcf^cW#4oN z7&WSKy}*w@rb>)TT+U39F-d!anCmxxSJ1(RsR6P)@T?B=+;t0!`vYhuqco8l3U^p# z1eTdUSVSHpQWRl6$9?_gZz!|Q8;JmtgeNBUMQeFnFK2AsD~_{|OqX4!;;0 z8ls}6KI&y3IBdm{bKBWrV`G~}l=&A>HS4qKluS)mNJ>bg0Qw42eq_{)IahagcXmz= zjGv!4cg$tj7F3ZPL_5r!J=&gy#@Je`O>g2&_&iw6_MQ669yyr6NV+L~QsU#|!2-+O z3`|W;(T2=jart48EQ3M^U|^(;rezV)(T>9|kzHVNvugL_qmqLvD-Z2l?j8(EOvLo9 z{-c$-YdjRrMfBA0DGm{F#OWo-9eEX;6W-Hnxs;IoFcz}7dAI%1kf9lu|Xy0 z=GabNUOsHb-Sp2jlX>~m-ScXUsc1Lue^Fe)$XP?r-ui)AMRArT1Ctp8#WSc2F> zt35q&bwGOB)v+URMtr#>yQ}Y}EV_r+ley>Yap2-~{muT0FOjA8{%qC4b@lM@$Nl;n z%jd6DTfyKE2)N7%oSmI;JY zG&pGy>|eTe_z%|(yo3@Fwg(}ljtBEr>O#hYDWRs@wI3nds{g4Daj7A#t(E0n>(UZQ zaV@e{U>Rd9$5BKQ%$F2a^qip9-;x|=K}LV+d9C-wUzF{Ddz(!Z>~*e#M8wF+qjovx z$3SP!gJkG4>E&cubNYUnGObP{uG4w!7lQFz^_RT|EiElM;A8+L&+BWK2Lh&`r}MhhnF4T+X*gd-`A@RNQ@d>5sB)i%xvb;N|2DBbp8LfU^QLHKMK z-=A(F<7$u;z*9z94fIfmi4;&-62FZSxAb|qJMbLXDQv(H9!L^!28g-}R89$mJZfp< zaG>FZL(}8!$wC?};n#>>Z-S@VF`44t*sAAMNZ}mOhar(y%*cEk_N3{Sl9g@Wb8tM86CCOzz;n`+=GhX~VL#B@GL$BAhRd%>TFLdt`93al>2k;%8;I!Im7 zBD-2s-uNh&8i&)y0c`j%yl%P@(nnTOsLM=%wG_Ix&dM&ilPUC8^gdqv(O^Tz-RS73 zc9HT^%!@FiXCR3yh`!@w_ALKOpM}U0Zn@QwuRsIb8=D}@xykw=r+`QoJ7d+5_;9!I zAh3s&4gmb_-__HyeTGqNe(^$8J0*|mlPnmq9fVG|VM(*YX=E+h1;k*WGx zV6p*R*}8+o(O%yUeBaDylyS>OLh9+|cB1mE!SmXkYn#nRVe5-~#hZ{Dq zKc>sf1cnm)C89rlbWGs3Rb0yc-Mm{qnE#t>A=!y8fJ;DKaaG-!3X-o@(T`Ju2R%Xs zI70LYLvXUhJ8E6=kyV(-i-d(I1VtaL~%qvJcTErWL)q4**1Prwj00>Kc&Ja6^B`uOm;U9#Y7 z*h7K7qTA{Zh426Lp3Xv^_TfY_d3lYXHay|?cjN`JEWZ15F!E@b|0KEiA7;IQ!#)wfrn2Fg~swv2$a$_ANZ!A?z3c zJdGsqzQF0!{)Eh}u``^Kkl%>~X!*oIp_3CB`r$8;wPIiPhsz344a4)FlB$qDzE0QQ z(AKt9d{9s@;0V#9NuR)XDS)I#oe9Va{!`|eS`8kF*Pkp!MFlJ!JM1I-o_o(}PEg!L z9Ef!^*4)-T^t|X+seedbjgHbQjrGsp?wa&&snWOKcD@qWI=DMP{V(C|c7;a)#FJ9@ z8wa3ig|+E6?K$9JVKoAHB+=k;#}77*{{TCh4|9*>$y^W55$i!Q)O!AmOPc8Ow^aW{p*F{aN44+U?~F25;v-t?%N8 zzTB;Xh=LBbI@IcLlF{V>LkP0c5tjA)cb5YE@a5IUJ<^;x;+g?oS2GSBJ2Y~J6>LKd z_D@UVtwH>@$l0C`_j#&ifhHrF2!DWvc749L178ZakGVA=GSq5Q)=p2#IQzirM~+U0 z{Z>GZERfQKX>NqS8H7J3J^fpD_Dg^~@bK`E9zWLMif4_LcDq`OCYehYH^H)>NYz`^ zxF387&3vy&rqa>WtgD*Hl$z`EopdaWAq$6U@(U#Q&?rWgpNv=xrM1M+rN+T5ldXr(EP9@o zBs=R?03M%KToveOyP&sMpZ{r+C?HRodWe< zHPfoMUV00USGJlQ6#`#rt)H;5#zAWX+?)D)9Y@zUy#c&9R+oiQH zCfR)=$oE56=Zjz zAsN__K7#LpBj$WMBwnP-IV&krU-NbPtziE{aPtb@>mBrEdNzN|y?(rsYaD!%<-KD7 zLY%4aX6*(9IJmhLK70@s6GMLW>J{Kb$wOw2d(&+IO#x7ErBx2N#r^d}L&W#*gY|hJ z<-k=}{Vwc?$jB;S0a9u35CAZ2*7MKv0X(` zl|9i+V12WW6?uwaLTk6+DcPV;z?48yi~v{KcYmkv;`LQE4tu))#WkX%BNG$d+vPkp ztL^E4yGp`!(!|^xK+JD`WF9fVb9D?xzT{9dRp-iWVq%h>nF&g@u{?G(aKC9~Wz71z zAsah8Lgjj#nR1;=8SQ+gzr9Up=Ck>J_pkvBcS&axMCkcG)fl?L_UJEZ!U?q{@pPP% z&Wa1AhHJe4n{OlxnI$_dqX0BJb#DjIB*3OiU9`#iTf-UV3-v&pcJ@0ywuOtg4*!a` ziN&*qCGv`j<2F{ZBGvpr<}+DnS-5EEsIPq+w@ZU9^<;=LRfv1;b%KY$O07n!e*$Hg z`Xe(-OR$Xu_k@g$j6f}M`Jl8Acx(TPw-0W>a9QGZWebe2w%z44kpUwz77hiDXHewCrLj%Kjaf}_D*}s6AQSU@*V5|^YH4%2D|gI{$i_^A+Q#-{%ewS_&>GVLH+E%02(}v zte{`deTgQphu)-s(?7qHygROSmEmTXk_X!_ZGMJ2iCt^o({l+}u#;)topnP8Ky2O8EWwK+^<~0u&i7{o}QkDoB!Ai zdm)R!$O-%;&aV??L6KyJIq22@Y5tS-tK)`lslMM1D=9c+59}rx{ZxVVdC12RlbL0dn~Wu2O))TT`jY2t4<760)NC)ZsmuRIdGL>q^qciFL3kr!35mQ_z{>Xj)n! zg&Tg-N{R52pTAb%7Gl3M9@5w-{P)pfFVu*x(ybBwNFGn_p(81FM11-wt6LpHh;xng zf_M=gF~M<$yx_5=g1aWU=bb2@2mOP`p!hv0seiuC6@h@m_&aSJ9+W3Pjq}P)wKCv1 z(9|?{+TZ3BmiNkAe##`6I3eA;ALh|XiLePo7FtNM=lzvr&qiOoA28VfZ)%dUqR1+h ztP(1bRK2#AGn0d>z$d?OWmd23NdiOZEh#r?fJ$+p%tsOs)e6f*S_cN0Z99Ac&nJzD z3v*JdQ)B%D_rGyc4LKUKMQbI5j-@)+irmJE!==B9(yEzB7q=&TJ@w5?#wq#qSn+yT z6cj_!mBnGmyg^UlN~?x{+V6<64#BkaYGS{-@IQf6x@I+zvwWeyS2zb%MDF2s5MF_b zCBrOt%$fHMt@2aCb{PS$T|K|bu&Q%VBRgB~8_~ zoenxpxrk@p%iD17FTccokE|E8(~i!KMrr4jS%MX=0FJc`a67|=p{_{UlWvB>4;mVK z0fr*2VZeTJAE+Khs@~re@u(X$K7hWPJhPnpo%qFSyO|#r4)8*V%SSMR|Y8hf62U zM*l#1i0P3Kspks|K>!A8!0EpyC(D6~sf~>da4R?H?@kuM6GoD$qDXaR%b2s9Vrgt( zoT@4DE0O7urzW9U6OAG?24_@_rbi9|uwfLCVqa<2#yBpxM}vR}>V18ngr)|J0l*%p zKU--7k-Be4V;uh#X&kdXR=1|NesD3*VkF=lBFp5-E1a(dw zO%kM3SM!!z%piMp>BC;XjsaQ(Q`IWA90wgt<3n!lPnnFF$(`520Ey(|dAIT*LkX~& zL~%w4#|JX`YDHH0z@Brt-pb+u5X1fYv`M$R=M=aYR%ZWa9!mKY=c=tG6&oYSKB69NGj4hZmq ztb-H1VkC4!Xt&Y-8G;>N_(hjKhB#kAocMnVq1No2}&lT#>` zR0K(M97@1xr1|8ZU+%R7AW#FBMhci+!0X}Dkr)Gjnd_9S!w*l?~$r0 z318;fjvkDJu)~bx z^)5ORFSbPXzN~~ffdY-;_Xw;d5K+=mK}4ZvW;OzcAFvzUBIV7w8t&&oYd>a|MO!_q z(7E}#9ZTxW@yEtOI9{Wr?zy9w2hmWnhac&Up#0ee^Wfy$LHg~rPBRiR8o^?~T#y8< z(wKT*b@Kq-?^i7Vv@}A)^67srX0y=9c9PerrNegj*6;ywc@Q_P<52#P2iKCBO=t z-_Zb4X!!LrZm2;Wie&&rnI}KU{7~Y}BIj695!oI}N;IMuFPwIZIxd?qJ%qn};cU2U zV+LgQ5U|MQ7Wh+NA5#!LYe-2ija&cv0Xq;GhxB`6>(*VBt*+KTWYD8$VALICo`4}e z1#JOOpFK;huTRptUB-u#SU31Mdg`Mysu*6YD@sfi88)N3Ko!RP=HFGH_)N8IO(=lP z<9`8;^d3O2S0@{KWvT!)=?tZc0w}wH16WYl)rCs(0BQwDN-`r21J}yfRc@-Dmh}z< z6mNKUC}Hz!;R|mKPr9nQ*K8AurA4Zv9{9$y#+P8-^Fi81jCAhHF>Wf%Z+#Gf;lTgt z;x7n?>F7(3yB#ginbfg<`;d4*WicZ ztTSEV!q@O)W{VBpHu?TE>-H$j6I24fwguta0VkU6godLt(7TRc|B2&a2PxI%0leZf zp8ePQt*mRs^VPG|Fu^W3gWZFV*U0#y{fs95X;29uRuq2m0}?W4#OL-hrze3F1r@L5b&u zvb{@ZK%o)&ne0>`7k1A@V|X^I0@lr$^y<~wq+XcKUfBej^`fveO(bm`7L?hV4yymfQ5RDgk-&wU*V!V0b70yp0Q`ZHzj{R77#nb zfR;)^h*TyrVrTsbhiOal`}!d^Jj`BX8tLm>A1*c>++G9m(y%w?jl4V*0v9`g$pYSv z1E#UUs>1WZqoD3wl}0Y9cfj-B4j?ZG+>FsqWTg*UEr6F*N$%Hko+k$AYqjp10_C`S zDe39H<$6O@xR@05uLB5`Fm!YT^KP^MT5~7a+IDda@6J>FJOIA&N*`Wan>0iDW zL+I(~egS)lIw!Xvr|jQC$?lIeuK1ARVm4r7e2R(L zVOo|{sxDHW2yk@N@ZN3_SxQb$c+`8;qEG}RLU}a8xsV`nhmO>&>L3aHL7)Tf*Bls5 z6&pZ$L zfr7~01|VA4*x0fvmK8Ig-a`(=WU#^hn(i{(uc`)puxg%@_(XD+aTfi$M73oITeRj_ z&Upqa&?Wa?mk`s8>&~X=ia~SqJB$5)xxqWrchDn1DITgnmh+kR6@GXmXi>5p)0yZ0 z6PzgeI9HLEVmWBscCoD z95A~_jj5tR%D`I%s)T67ufOxytwRps+h@@S^LSW_88njhmJA5aEPV;EH~jMI4B)1K zw9g&ZlSC;r!QTSWzJ7&u#Z5;d)UDvUiKI*I3_?J=#Y|QMcn4aUf0k^k`c1WV#`Db{ z9*XojjzJq?VzCUM51_i|D65hb5z!n$E!AJ_5?(L>vX`2T*Q*Sk*YrEtg{G9nsZV~% zprl_&Y3nM=edP@K$N$nB<4z_zKE5Z0Rd4#;04QS3Ol%cOLJxAx3zkQkLx^+f+m?*U zQ+NYgyX~7(Q!Cv}t%hs1Vg+H#LcAOhlGoNLl#aA8l9ax=cHs8EI=-~r#XIAeAP`YH zr77i1^2WDc86Ntr{HnT4L*zE7|4Wd)4)s}widJd(^7|)0e};VfG$WDoo|=Zvmkx)` zKMVy96=9mcAXLMT?{2((wSFFS0ATvrh`K&i=vSsP6LE6;p?5$v<13hl41z>DDquKi zoR7_4&cBrur7rs{3(JWN$lH^>5ImB8F%dtw2(gN$gDrmn*iu8pT?#Q7-)cQY0MjoG z!A|34z)$ypry)ACpX;`MzSsj0M8wVa(RT^j4X)W;n_VL*NsJDXGLEro|AOr|PIlNn zYzme3(mCq~*w8B>P=@Da6<2ti{M?diK=ZAE0Z!AGAu6Q|J*i-$<1%;ukYREoXgDy^ zVsCv@)yOwzn0X_+Wx0EfJpF>PRq>o6LZn?YssFnN7rO0p=oK;;U!wFI3knbzZ1Q^F zxaUV6PK%hA32e?W2cS#|pr8#-sEVx4Y;tEv(8@1NH?}V`Pb#Xue zcfiEiFE!n7a8|U%0mkPx+puvkty~&1T>K&-#43DLBmVZAymsP zpd~1ed0N6!lJ!d&kQ_xK)yE=FPxWbncSMyXQvTizUR8Jd5S06mMTg^|=n@k#UdW(1 zMFAV4+J|Q=1-mk;z;F$?9zJ*r!Bya=y)RvCCEr+Y=YnlV_*+O*+KFJctAGEhGH?p& zSVr)xQrlH}Hbmp6K0AG@o;G|G*CLTaEIhD37i(={)%@;Dm6S+&8gQ{V#RB;0cgFxI zaA;jBMvDneCT}8=8D@Lm<~x@QV|Z#GUV~o;#C8<24U0wNEt>?H{BNACI1_`B&=yyl zC4WnJ24AKCPt_b8obL1mGZ$A8=F3Z~XvBi@nNUCdRKIzw?&%nq5op*)2sTgHtZ#fF z@r{bB(@8*&Mn2Z(kwjcGE2iYwH}7rXm=V+$z|A(gj(ph^$?k(&sHsbNXP%k_HjNL) zvr@7X2y#e9*8ocdCGa$0nwgp{pXh1*x!P3X;40O&KyOx@^P@0vuiA}ELVCPq2Czx} z7I&e)UnM%XE6EdOcWp|%&tDdOv*U1Gd{6b7Kf4#T`8n5!fZ|LBSS^GOqHY!EMTJmO z=BBZ2SS3%~>N*-zjWOizX1Z1*2{bVQrPi9z$vkO^1vq;?Hk(Q@M!)rd0L+Kenj+F$L@a0%pXs^`|OlsI;DdnQ(GWfT) z&+B9M=x=_f0uTj0ZQ+IBSldRaTjD2gcy`fNshK!%>Dk5y{61*!$%Wr2pj*r0?BtAJ%sKoHl%ld>DP&D$pM zogq)*{K0X}{>l}JZjQtKAAboWU`;wkPq2`kpQ-aEp?0+{E7Rd(qw4#=?m;v(tnU|O$)Fo=cPixYRU z9TdF~n)0D&A-($3=}W*k=LGCK&@Bkq77rga)%q4H+@oAq{7iWHB2G=s$#utQH+Rx!X0KU4M$$IRuFap(C8Oi-MMsdwwhw= ziSos8AVU@?(wnZVlzyk95D;0I@jJzDn>0Gt89+A{iEp=io{ zZw5G6|F~XQf|dw56-`CO^^oGWC!fYRkM`I5h)pJgdPX17NT5g?tPh&IWKzMMS5D4^fmtjFD+oUa53wT?;1R7e-8|lU_jhZ;TWOf!DB9)zk?292l*2r9ZrX z{|s&%iT@WTNk~ic@BNP^k%aE4W!=7V_thWfS#&Aby;!=>s2MCcqS3p6I0;%ufXhMR zKhFVzW69L$lIQ77j;u|Ml!gWg+?MmNF9Ddi+d+?x8JE?(w1EM&jEoGZVE6*G^gk%K zQ56ecos`Yu+^kn};^Tn0E)uqdDc?=b#aDWgDvU&F|aa*m(4YX2sseaFwbWY29_!(ud?p z*h$y^-Y4YMCAJD(u6T(s>6M(NDJn7dJDh|s|J@f1DA^+T=R$bf2+6}e?TC$1;F0~#e31e=H9`=_J!qnNgJGH zNYH{9CD_G+hJmkG;vvOEonOWAZHhIPoS!O2%0;8GO+Xym82Ic!jciyF_z~p*-k}s7 zbE{u_+SXS!>KFayI{-PtOwvSe9S(-RA%6pITo8#?!2oCpe4B<+UkEPP{)otI85RAZ zEmHmT83uXD4SD+o;k!>C(kpsEP-X(Lwl+uL(!4*d^yuZ6@V3e{Rg`BK+S<0v?H5%9 zYQ_owUM2=ETV9B+=CbV1Dy{db&W5S8%6|QdxEVPadF0+6ft>^w9!1iFK0+c+fYA{H z?^DTE(4+31FnOZSB5LY5Z<-MFsnR3n%+R*Ne^+N{0M{h2p!TDJIAdKJ@k%Xh$@pa^ zP0si7OELIuO?Jc-2T-?2=5R_RPriW0kPYzq;=tGG?#D1$eMayzq~mu0efrqAu=Ssx z9BSD3St03K(qsZXz;WBp0OIG|PefOdIlX4vChEj6<`i8+$>r(7Z+y}Sh7 z;rLrx6kAyN7MA6E^g06tAGcLWM#b=ku8)+-#WO6vt(c=D@uDKFv@aoDJSGn1 zMQc5}7`t?7O&T_^%Wvb*7o-yJ^FmM~{8*?w$lpW>TSe1@)s;n%sE@pESLCt~XkS8H zwRfS8j=DoaE4c&tdFHyvgp9rJ0h{?ZGPHkMFhf@Y72WN&#@J#Y8 zOqa0}&YtiSd4T?DK%DPFYU1-5PPfaEtQBm8Tcap84?e)RVM}0&wCw_W8e=jDKTFNK z9J+0mVjXp=HhE}y+dPS{&Z@NtI$1J~IyzRkt$W)`|Oqk5^W zlq@tvxEsk)2{)Xy7eFirCoz#(Th{@{V)crTI1@^L!4$8Mh!94f1bdsBPb()huEeo`N=xz6>Rz#;uHX4ix{84o;Whzt6Gul*uXwJUi@6CdLb_nf^h}Y0vEyRa zzLQnd*;x;h;-keS5{btIojh>z;Y5tvakbN_Ig}%cga8HUJMvy=g^#j$NQ24-H`;4* zfP~$WpMehgnQhJ3iU%)tuj-4=Nbvx=Tgb8ru80&>Wx?==acwTZlp{5~Ycl(e07oItq z1E#?QfhBo~%a_q&_w@Qza{0q1nc_HqUiGPhh$>sa-3$fDzwG*nY=2LR&$QDcS;?%8 zI}PNiQOKzgk<8+n8j(6Hy8=h3u|1)jo+;8WSW3L1YH5Rf{P+Bm7shbFTTV7`hhqOU^&VXEZL7y5nEmVG=ARL8`7OPYjQ6lqTS*OiiS3MP&73?GE9=p%zAbF1$~FhuRk|Z(hnUz{M6T) zIlG!83+s(NUjt=kN>~!L846EaeO=h;Mz78vCT%Y4H$z2vu6H(*+vHD^MESkvvEF7$ zl}0~7QRc38Lm?`kKM0RE-QA}jT6rKE**70U38?B5FBFxRsZEVh*#o%Pf1H(<7B3vM z-)dR^v(GkcR^U;Vku^LVkwPOzQ0Eo-J=y^h7w6sJ{Lt>-bCTW+32&l? zwy@|6C(yCjC0Kh+7h{{@%AstumJvIXm| z!seEHnMSIMS%MCR`N9T-N`vBhk3$NVnKeHW%Juwe4mRY1gK4c{efEf-9egY@G7`G< zb$?4o^{ac35$08cWi!7~fRB}YQ{>EO6K8B?B;*q~rn-#ldeX`FtbYCY57T>HH&~8_ z-R3A-TfvWqsSJK^^!RO>f#}RHEX;wX%~{vMxa8#N#j=?vN1LXkn7DWg>AK*4KMpu` zqdGISGGU|L1G{R{2b;E@XBX;hz-tt-5g{uOib8B<&}XZ&AtS$fD(ipP7}Ce49{sRj z{4dFH-op|)zGCH@HQiz%n%^aEU;&AO_%pC`Y1*A&NQJ%Tx=Rf=r z*82ZP&v77WQfbQWf;O~YSNYk#xb#np9HH`#_-DH4Vd^&G+U7bP z+50tcO+wJrR{|9?noMSNfc<(aYza+AoVUKsw#JC4L$*Tn{i6Zx5J{y{Yv%HSB;NKJ zknS_}J2WFX^P#oH(W-tBvB(O58D>)`Uz66h*!>18*BUhUpOgcPJh=^)DhL3208vy9 zT^Xva>1SO{NAxp8miq-nIfh=k@Bj~jR#Qs{10SMfgKSLw!2fSKXq|@h)1$#u1=LXa z4Gy`69YxT8r@;Q#H0vFZKT_-hAc+R(#6Jvjrl07#w{!QHg7?n9buM24=?_d(1|T&; z2u>~_Fu`@2`}I0>?euU8K@KHa5@KRz*RFl9n0HsE{_8r>&awYJln(Mt=%uY*MURYF zBC@w<(umXfm1Z6QD`_w1zlnUU^_&ZV%SfF#k$rv?&<_M4KsbjF^Q+53yrCMu%SYoL zrWxld*~)!cqHdj&mX1bxr!QZg0fY=J;~YT$>~2c;mcV1O+1klGhS2i&wRX@P8eb9A zaIpiP5gW7i(YVyqh^FzI`&G5IJ30M5`@VH!9{VHx1GKRqssf_;14HAoKP=B$cx~Rl zs5kSk{}*TrG;Y4vLQzlW)?*A1L7bINUDHR!hY6rsPF9Rvc+$| z{Wc0L3KHxD@PB5RZDz=)r>H49C>-zpTEbwt#w$3qvXa3%52wy9yj?H@qSYb?WN;~; zhGP#b<+cuUUztxntON%Kf9S6vb`M;xaict|N8Q>r&y=z0zlW)$mc|Unr#uL&+Et(j zkRw-fU$fOQV~LOhwZMQsR|)(x|0(hjThARY+cbXb79f@O6_yH*I^hc#aHR6Y_xHJf z1y~k{8Gz0qMh5Ad!z#Irw5)2RxmI+l4~ic@`o6?Y3m_?rT0gJtG6*<8)}JDarMf4dqA83dMb0w6K*O65P09z9fdmD?M zAx2lP@+VCH znoL@R5hHojZ#oyho?tF+*xUL(4HI%48d^%SDZdsJS4Tl07sUTz5xcPmpveGDzJd6n z)dy#hkVm`SIXuJ;BK-ikXZ4-nD1#3OMe=1K5vxh}IYxa6We>mvP&$bFq;>oCZ0@6o z*jOg}srLBh)>fpGhY~lGbIi)hgK7$j|0JzL&U1fw(Fcde zU$nL!6Y~4aCK@S7UUOYPmy3twL+0>AaWlk&*jQ4HseMg{P zWJC}>jDo5}QBe_8iq+nW$DkxaunuRe=9VO~Kx=K%0e9~^g8UJHBhSF5%xmM6f@XQx6 x6%~BgW4)aI-jnk1j$2VzL*i)uyL{E1{XOFx7D-Z;0OZJ`4E0Pg#k%%){|VPmO1%I8 diff --git a/doc/LectureNotes/_build/.doctrees/chapter1.doctree b/doc/LectureNotes/_build/.doctrees/chapter1.doctree index 90c4c8184c2c1ae7fb3e8b3126d55211bed1f409..2da43f30b2371f3819b61192fcede45e51324c2b 100644 GIT binary patch delta 186867 zcmc${3*770RVRwSoRj37lP1kMCry(y&6g%^ny85+OR{93G-%?*%cE#_ z670?c!wjQX3ye`{B%AKrg$j@>lB;|k3md*fpdJ@#)O zd%1J=k!_nRK|Z#A^vtC-a?A7A>oYgzO2-3SccS-+>fCXMFJE;jcNd`If$itrYw7H_ z-n2XNr;`wntDCzMYj+-PsebA#di^e7VK<&k?>8oCaDKge_1B%>4b8EcXh95kxo|t+ z>!sJd?1uA(7HgofHqIiz)B4lbQRmN(f_SnUJLc~Ct5+8Q>%acI7lGyvU4PHZ&ZndC z7>r|XU&~qSfJuucR-9^~xjWT#PqWSQ>+in?yR-{X!ig~lq$OH7n#`l~Yo&VY_uP2n z_1n$|0~_bpw_S@}-qp3pjCP~B?(AyO?&c++`r?CNbnW8q&hGAdb?FXJ_;;6X1*ClW z(%s>+#k8Z_;){-9Imy$zjWVowm6j3qiMSS^cd^K4FjMT9__wt{nt0Z zoewzoig(=O+y__=#IVXzTtV-<+hJk@9s;&&s=`(j`MeWUxs$>yZ&!aE&87JWbUT*^rmZa-@Jb4 zgAc5K_@>*=6mwtmHS2RXW9z@Z>67b^zu;rJ7pz}*^IhQchi?Ah(epzuzr6m=7p_m0 zySKdMRQbiXKD_gS+!}p7vi|S4-gc(E!L7gOPN3ZGynUnS=-W0*?+0#wDR{o|#kYXx zciwg{cz)`|x2%8Sr5D%LJ6;G%AGu@m2Hm~m+G}0|PN5e(QaT1XIZG$1H?!ncw^Ucx zv%4=Hk94xbgU8jUIsn$dX?R4?$u1|Vuh8WL-KRU9c=3#GTDkL{^RGPEy6zjEeEprjx|4f+%kblX;U7GDe(2>lt)-VfcA|WF zecQ|4ajN{Km+hS@|J*(AJW+n%iRa5}{oZKj@hv^)Uv)J68hP;YlP|mPD_8COac=M9 zPhEZ2nt9-TN6#Bcxf_1Rsq$aL?>trhb>wZQ%1@vt-`})O^N*c+&tczjs(c1tpDKR} zKaKZp;`I1Q;n7p=A0$t=zet@N|ML2?)CjcSNb>p0DYo|>{KC$eC*M;n?(9r=)(_qM zlH7EC)z|;XL1FWCBYb)NCtv^er`7!AD^J&4T0i^^??2V%(t7f$H*RV+gZ7UGeG+-_ z=Bsu-us_xZ)?@ZV$At&Ej~o;>UpK;+*7xymJXLdf{W<>ar)n;(f2Q=7gPPn2*3*co|)oiJ&DUYA5xV*l#`S#NlUp!rLY5nn*z2D;URab$nur+Fq43Q76zqwro z*5?zQ`*J_K&OQHSH*8M0(IJsnUA6wy6F0B#=v?z05PQ@AcpaU4?E%KGbzX7>j2E6$ zH-q2i0?yub!}IRGGA*4hl}k4`SN-fGm)2__igX>#+ z`P?I$&iDN6BR4v6JdM5tfe+v$pzyw(IVv6btiwRyu4->)oBpJ-VHtN7tkAyTCJOcm6vM=j07fmi+I&D);1C zPSEq;by)l1_48-~K1cs$Z~ay^p(@j@zF6%(wj5&N*Tq1KY>gJH$XvVH{PT#CYwk71~9RU9D?_GWO zy7uk*xhM9g<_FI1Deph~;`Q{-YtKJ%i1Cgm|L3=tcP@S43PZo;%`aQKZx(lcVmtH= z8+-KKhuuDKcJJdiU3K@)2hMKKQg8%)=baay39?SUGv4{sG3m$s*Y`d=Iq1Fl+MI!( ze)w%qtNFo4(5KZr@4G(nw3-XweKR-#@A~d<*g1doR{LCgH4u-Jq^_U%ZvGjxx#!kj z`R;pmZhy8uuXzh}$J6<7&jIK|F?R1*fBJiF+d-ZUAoo`GfYLjUagLPEDWvPxk3M?a zPU+c@{`jqL`A10E2k1W=z!x69{~rO^=f!>ml&JI=1r2QNMOk;gus{8vM zJymAk_Ks8Ko8SKCQ{~@%``)SY``)oWQGVZv=gVvMo&3&^UA;LsFTDS7!yM4`JMa8r z?$X(7_7&KXs`=o(>j$2I)@PpR?OcD&`dyuSfhB_N{aFS!>yJEf>-8H;Sln0Ktd={*l! zIMPptdhGfie$}~jGjF%K^J|Xv*v}kJOZeos|M0tWJKwfHA>XzZ-&?)-ZBLtLu;<_V z-h03`39xtL0ec@i1o*b~_3zWJ)a9Y~MY;E$E-;WpUCYzwcqur^^bh$=MD;+ zFVOYLfA`_Dhwl%n)}Q&vn>UXev`-wP?Njm}{{7eHau?QLe)2BxdDD+S3_joW`@QylWkO^j_c>-XPz#{`f~f zz5l#(*R@*?f9)q%&xY507QFxVQ{-uk(oa8rC5%7&X(2b+U|c+~5RYlPcDp)<=i^;l zR*&8uOz@xn$DcZrd()QgH?4p1=YM7YdFM_0bpO)FKXSU_!~e%ePggwg3s0V|c;pww zi*I_`)Hz>WfAJUdxsL%dUvY!8IY)o%aN^#ye(fjjxZ~*^PH)P+^*cXtTkgMXC+a|I zPxSxFCuXN9c-Jre$mxoQe)&VED@LCzpRV}aCx7;I#UK32)A;<2U;U}mEk6EhzW^$L z^qfDUC4BPd|I?YB+`a2}{mxzB^K-xc3y057e)=~q}X zhjVebU%LP0QtsN$x$RzWt!KZ9p8xDoll2FF^VR2%HvJpc7yqlV^U8bnzWDlU?pwe6 zzrOuk47!0U0kGYWf4F)5+TVhA{_VL#ZE@!Lo2mcHLoCpA{ovo+bY|0Kt^Mfk^*5Y+ zJA3lF|Mox*=nBBh&g19S^mp!Fe0)RGi5rUZ)xC#qzfjJ-2M~Gx);j#FL-5DfxBbo= z_j`lS4adLxJMac)voDWY%|IFK^8WAs*iP<~n~^Sl@@Yc=mG$+%cTeuOzzDCn=pL;8 z-yV+g$@RN`Pq|W`v!A&q_t|Ib<9)`wLZ7|*H#`r@ef8No)qfw}ClP#Z*!l6_=dT2U zTz&SIy+8PqGcR91@An@#)UhY%efSU1BbcZD;Dh_mmyfwVHjvi8^oOs?k?Ti4f268U zz;I7}%Mr|mn!kSPp({|ce)x~>U!Q&Ify1Nh7~tIgQMS*@y?^}1-D3*&Z}z8m;OEcn z&6M1&Irmw}x#v%yQ{+7SCzG8|?yvRHmy-Z@>HS`90;M7jDfxy#DlO?*gCy=Vw2(|Gcw#a|U>8e*bgAl>lG)`Bv^HH-P&Z z=NNGP8=vQoAhzmR`~vh01ozy&K>XH8mQR5EF~Y$tZe0L3e(9F=wO@MaX+Y~s z(K7(Sj{3rvfPekM$;-kq=DPFe^l6aa`18B21ajuf#QFt)ekHyZU-s^J`U;j-I2^!w@`%HPjqyvdO5ae(F%fimZ zcWmCB(1>7ifBmn}ogX^ISbl!c^BwDN_{u{&@3^qFS>Lh#ldsgC0dSyFcYc1m-dmk| zqWiD@HG9wOHxgJ?#uQ-==q-c{<7ey!Dv$SaVu_woOb zLv!6DdJYBZ80x~_yMFoVmtCRjhPBujBk-1#q+~ccjfGY$1 z+4FO`uiLcP9F|8T?#tbo$_{Aob64eF_R*K@+ce;e*oUW7MNX`QYAPoXt_WpFYIB zcZ<6>y*hXE_HuUu_wlQ9uRK87d~GJ@!Zo?k?lJ0f9ZH0_5Ab_{*pcERp=ZlHb5 zl>7V<&f)SMDu_*+$!N zt}6_G`k*+}>dozZJlmSydCbG1$eo5cIB$$C%8L)I$SZU8g1t|DIQN2+QPKUHt2dF_ zQ;UA}i;wo&t2-HTU`g57UKUV*;m?gw@o z^YA=4!Htm~1DSr}Nk}wv9nv6-b{-wDH`GUa{+&S^$!^_P2}!-eq7sFlaD| zK^m4mo*SSW0N-R|ZIg5&Gv4H1ZIZio?~6li6x_G_azKMMitpXF@&V!F2hO)0FfjN} z1KUKOhhexsis_pGZvE6XmsTJ_>{Qy%1G_!I>|{PRcE=+RWO#v%yuk+$Tg**2#d!j^fSpcC^$PffO&m<7&$!Am%ia z+aUq159~7Lc#`g8ft)il&?oa?gC1`<)9ejp2NM%b?$@>P7);9kjNA`qcLYW=?~lzW zz7Nnm9>rVC4TqZywm6%vMw>p_{>$CTJlv;mo7iNGEDLZG#Jga+!75Dvr<;X7Bo5RB z=CSx4Yp4nFke{arfR;&E2&+Y<}d6eZeE_|Fg|=YVVvF@;TvN@$8%NG~N-#xsUB> zwcO>^olh6sXY9@sH2f19Uuh{Wp9tGMa&{|xhpTY=LGf7eAHKiM{a@Ra=^Z%auDmjr zucS0iP<-Y0PoISsf9wC(YUTuq&pr08uj$SxFxQ{E=0fXVSJiE|%~pI5h2ahVT;=z# ztNZrUSLbeA{r^nFYF`6iQKy^|T`hW4fIykO2C&tEZI)9{BX5V{x=knge7jlhr z4}f(5%khfqH;Dx2#ltgUws-q=xx&uZUccA5F88XP8+Z1;_4?dD;CJmFYLCrnxiP60 z_iwGjjj?cOC+we=n+97WWba#U$o zpS$%Ozt0A9{oa@V(={*N`_%JuFTVZ3P2IgB^Zf_Tlvi%b`&kfxuM0Qk?%wc+-k5vC zj(7durykt7bMJR<%)M<>xbwHJxxDvb>df_fU;5?TOLo3Y?A`DKSKYSvz$bIxy$W!C z%q!g{9yBn+Br+X`5sVTk%!ftB>$G^4)2X55cPC-B+0?sgt7FR?(ybw!+UTkUt20yN z)=X@mnjnw$mQ(ZQMmD1OiPr6~Z53sM(rm=H*d}j#o+0}byjWnWtOoir!h%%;UlK4@ zSoRW3%^SWoFOY;!B(X(^)2hO&m8jq}QQB+rLTOTC*y^aP8FZDcwRB~|S4UNKiqLj- zDlAxA8C4xWP*^S&!^T2Q)QM7qi^5VSy93k`Ne)pbd9~g)mqSfzIB8S1R~^PygSnb- zOMV=;#qrct6KWAPXjU1r0@0D0(|Wn-OLZ|R8Mrltn+3a9kj4dmE)M0&vb-c`k~}U^ zfhjZ$K2UHWumxr@4OGFVO-5!&T8Owj6Vmmx9J#Jww0e>sbP}qN&RVO)EYAz6P7m#H zk!NdAjp3@@l%)v<@p^~@DVEUmW)y7;o-xd;GO-||vEG!WVFO#5i={{={8BFTy9H^W z`YPsk$9svEUcgvSyhYgmp-76?C#%QNWTZtS?2_ zXMA)X!U7ShLrx}9G4p+N%+27UF~|>>iQLe%-q0G#J$OVdql()Y4*X%AW|mezoK`Kn zV%Ke!j4LxK1nb*(rJ-6LCVo5yJ@Sb|1V&-j4dbc5#p}T<43^t6IamE6?XxAn-^oYC ze6av)sy{>{HHkWXK_{>v$q(m5Q3T828-Ba*qiLiDMz7Ds$-G@mX1)p=<9Y#2mk{1l zjo!4l)UilHi2}ClEfqMU`irW}+U=|~p?meBNG!x<0a^uaBURWkZP-T2d9+|xaC$&H zbY;Nnl@?eNT$pf0ebl#gn;!#4Foy-RgQgXoYtSAoBQ!=c_G~&vXK$MN|H9k z=`~2@EVf~_^0b1ud3s1!Y{8^Z(&p2exS}DuI^atBvfdbCG&e*MS8Syfo^Jat3NLsX zn>5L=K+$?h4sAABOg&c7Tdh8ru{uqcL?+~_%bKR;btWGKgHqP(c>PpNS&DUqhN<-F z&>kp^RmWTk&U%DDZ?$zH=z5*5sV`RLsf5$X zxCQARZFy)WMPrm^@lwRk=UtD<*k~-bYPf4Qdte7K8bg+XmO;pRi1zUj?5p`tLRU`R4$?D~K%)1`T{&-b%}UgP|dZ;T7%vQ;4(l#9?^vQsFf z?4X5Jv#dFwv^17u5z>-ASwi{{8DldNDGDK2 zjv~5Dib{mioLkq5LwqSlltVP}VcmmN`sXB-rwu|^uASY!_of;%8 z;hm5&hd{YkB8=d4u12g7IPGV)rZ8z~P$IAhqNYL6wR=e)S`0$77}6DnPt7D|daPo= zd6-*Rcxf7{k{DyXDeMN8yi^lS^Hp>}aBh+Aw-^&DW<*pY+fy9uK-X){;XXR0xKW!g z(Z0_SOhm)B(&x*onG+Uu#OsRkfG;-}6vBx@#ZzbKTOO-^mg9Z_r z9BNL>)RIdeV9pU$a5l6CoabasxLv7+^Rh-1@=F)#j_RI95rov{j6P9kN4BC0 znL6wheU%8*K?6>zeI{NmrI{%>mOmc%>zHV8CO5B6z}9Opp16peHc*%4eAl;(DS>H( zkA_$`)JSnS@ge8k1*zZFHvxP;Tf`NJfDpySnA~f1ozt@HbOB!s|rkzx4 zR5@a@)qJKnzTvg!iDIR-?jr4VA%%1Xi44w|unPALd1zrSKW3Urm8djuHzb2)IW6^R zcEFfyR1Pu7!Dw|!*KjWGwUh3&QqEdK+DZqh3@{&e%p!9ryXXE20Yp11fTkv18In9DfS7 zXSqsKBnHDue%>KN&np#L6hs$73aYRuxuk_uDlV{u5sXD0XF}MS2sqPDX1eTPNiwEa zwE^3x$vjrAv}rA+McSefj5Ua***8tR)hTF_)>0-|!j)=W(z4Muv*d9`XWD|-EeB0Q zs7Ok!S>|d>rKjbm#2BR}!(m|e?PTV*1=H(flLaqg5m2B0h|%eSH$k0BN%zuW$gY-J zf51dcK+8Cb*C!J`Yfnp^u?Z9kSh-sdt#zM1+ zFDpJVsPsd$;C7?U#74--9r3QGMg@JYx}MF?%rZK#^Frt`c`eTjxC$*S8q{pOYPtDh zG%Pp4X(`)rgh zQ8D;tS+Lt|*-dd?7a*+aqXAtv+ZNm6aeP)C(GbI=LNZUnhSDrnn!GZbK&!HQWgCtj zc>(`VwDZsm|+Dyolb=qgJ5!Gq( zi>jf}ld{SwgOrU}Y0zBAO}9JC{M~ZDGPooIMdB*Kwax&gn{3-H(GDMZgLb5~Xn&&e(MZAgR)jGwh4D(Ph{(J=>kMpf z)YC?6x3rN?wd(l;&ueZjA3d!m^h!0Jo|LL%-yV&fmZPw2OHgN3q0^n90#ik|^-aa? zFuc$Z!THZOdC@B{3a@zhLa8>ZHD72V9uYUyI#c!Prf0Jnt=XNC64zRsP^H;_-ZN9? z1|Hw;M1_`^O`D@5p>Fkw*-9*>6OrTZn z$;VB&%K}1!u!@vh@F+i_SbN~|9b(|)NLPnY-UjzHJc|V{sXC#YPI!JqY4y=4cAH2& zk##ew6p9cL0L9HBeRvfKczfg)~q2Ov@4vXSbY>~X(?aF+s#n$>*x@n z%R_^1r=v-L-5Y>JxYvv+SE7DTWGgB`fX$?myB7A~MoZXs;qh41o8&q9Xr$Pi& zPdYG^YS)$rF4WA#6%?^}*;|^xN^u#sQSbPzsaa;LL4hv{NHwWK^fIU?%ZxPZhBF(n z&GvXQsJ8+?%LjS~3G~reBt~K(T1JEpPLHK6s&%>44@96LLn)N~<{;`h8*x&jh->N{ z7!!(3Ob%$j1@0g1w4SzDk3^J-JzzYR8XGhXZi>QOCA+RCs!~+~lvg;$Rv1_CLRUZn zhM`);u!ZE2QY>KAglpgvq@h&?j4gCL3zcW0If+VBUTOs0Od6C}6`T<>g`I?agC>X{&|6W29z(%F{L4)6%f?D4oduGI5m7j6i}#7idg zfm=m-8v>Z&h^lrPV%iX)U{McxNedd0BhhUQ)W%|-;sa@dxTCf(W+k_!Y!uR%4V+aa zjDvPj7+Vn8i}Gfw4cwU}MI?%A-b|H(X`Mwyv{)d}SnrcqCbf#iQd`hhZ4#^y zxUQKZ*)URMB#vrqHdWdqtTG;IP~Omr&0-143zR|DEq7M8QqtDk2D#AFGNU54r)S}` zqF6+lrD!}MQPM#YXVAvc>CnN;G*K@Kq@CD`!pgqnm#~=Rs!ok>c_F6jE1cE}c~{4= zaPFlGV8#<=mzb&c&~7UX%v9)tUAH7svTZ)oT0OPbBY_o3D+6a_>KcyDhQYGHS}C}; zi(EpN!G*D6(@@F{mU`+ls)u`>Wvw&!pjN(4C4q)7sx6gNINNBd#AGD|l8}t8xKrqi zr?$ek#wN;K5LE?1LM%Ycu4Y_M|TH7;1H5clXZfx5U3AUOAMR*`PX%TMNc41oR zDPTvm7WI5smj(mZU9fO#+17D<(56hdF9;MT`Z_b8r=|839io&CykKL$7ne0f6q{o_ zwrXRY=D3s&0!0zaN}a+Ti_VALxPpi*_;I4FOODfRqBXFaR$d0xb&jcZYpek!76nnF z4%0KKOj}ri9_SVqFQ5j=k>nhKrwI66BWimYsR-7vsr4pY5Tv?kN#Jr61SM<;$MLMt zk+BIviD*K{`NFtgNT8&{nTb)x!L?9hwb=7p;4J5$S$8EaEVI5j*2XPDXN@8yxlB|6 zH`;L|>So#io)8Qp>9Jw5dS-IN<&q0H6xC^lA#Iyxns7)04`7>wlgfy$@#eTuobaY0 zHqjL`^h6))uq@vhvH5$Op z0@I34C;g~0rZj_MgIPvP!?nbsZ(>7>j8)Ur$Crx!e4pWF^HVPGfdG zMJ>&oP~^xj)q+Xb>J_6R7lC~iDn`v^R%5O`B}6M>7?kzKEk&59lRjZI7-s}ESucU} zRzd+)li`7mdzxCqWsT;mDO&FkipGT~BwHNI77Cm}v9zl~*rEwsTre2CLcvr`wCHXe zmwL64y(rC8a$`aF(^8c-S}i<;R#FkC8zM8Q7a%Kk!9N0Hl}4VEdTyZ>O2Be1lLp3* z%uO>n39HP&Ni13%$c&2zFvWCdSWPYtlR^zR(T5V(>j4jAltmH_{J!c!Era>B8;a*m z5U-r-9dJmDcn2QWd8E0({pxCAsDoaJXr?R(5-9|k)3$LOHh76v#p0w)w-~grESE$- ztS;*z%?x=8d=;R)rAVlk-CgFY=%5Z;c-e9u{A*>Z^#AOli;x;TwCcM0%R z##vEFyde(GsL6;=8}O=CDwUIAIQ2%|IYO$A+m{k36H4>KU^vG*5jc0tPnVC|SxzQAv9vZh%dxTTtiEjooXBP5UF!Z4NkeYz4QLn|a1ge)cQ5X0%k z0L%;-@a-gDpmcL3(cRIs*z1nunNfF|Ud5Cui;l%7ZC)H8;oRyVC5D2^T~2}Wji@*2 zHu^}xw8AOD4TnS(ZcYaYIX8PfqMXNuRz}sja&_)b;X)^&iFDEB7)X*OF)_xX&6nWT zI4GCBo?otOE~c#-C6=0L!>m!H9BA3iNOWF=NUtDOJho*jazDfyd95aUVvPYSr0Eg9 z9Ly{i4LUU~2KFM@%tf;;=FL_Dr{hL{fy41ctazDeadiT(%HGK7gM^tbRCfd=pO~)i zv3AqcL7b$PU-2cfwSb2G{AAKiB^V~XA<-71a=ql*SR6I0a5Xd=1;iugNjz*8>p?bC zodQ#u=Gz1x4wb;smXu!%iVc6RFJVfzjJ%9P)_~HHmJ4==8=@y(g2)C2Wn&LWK)D^?8V)rGTr|dIe+&)k{z_Uu@VDR*@%A zDPNw%JTlFMq`-=Y4FQSVN z^5twfD>YQDF>$>zQ;IqZnwbsJX>q1Wy*#D`OLEw^VAqXU8IDzJmXzn=bb@3aU7}sy zMeE>ll@5s&hRaoFG)c1}Q5s>jVX=c-!*DnYOGAk?y7*!o%xj3|qy?!l?Y10lR-g77 zv}m>RH)2_kUhrdRnzrqeBctQ$pPKM7mGGs=7Ltq2I9(0a0;1WeVFosx;xSh z;Mh`xMR%1CJ%!-&#-d$nYZIY?Fb&*`mrkkJ5>X>Xvkro=SX?H>S(m6crv&Q<9^WVv zV2Z8!Xh=$Zy@sv6Hp6_$ zVu@8R8W_cU4`zvx1Zt1FSt&K)g-CNP&ou ze07CWknb;MfcpekpkuGcFqYD4)%>yA9k%pIBM5?OJSOVCT>wS2sP|lhHj4IiVH63(U zEZZ!4z~Y^iGIVD5;ywa5e6MR7j!&Q{;Y|^8ZVZQmY6vkc5C-UZe6<#dji|2<+bu>> z$qJ7YM~k7CcZ$HqT7p}CX*LCMOL@|rp_zz{WqE-OjPV?E#Rfr;AqumGNDBnXjYlh- zue*-pHzmZ5m+5$+(2G_p8aHdxChuiJJY%GqBLP!1bNzr(l4?0mfm0caE#8NlDV#;3 z4C;t<07IeN(K@M*n_PKDE1oc8gi-)S-8ry)$3ml9uhf;OI4UD?-BITY!)S7=L7s?; zWX8}~hFO}#S)r6+Y{jJ&RIYTI1%~dFBy}k-YAtIvnfO&EF&tnmEL2^DH5MJeIpzz2SW+@=sn172 zw-%1EA-Jc|^%;kf4KF`LA%v~FVSUcxI8+@n<KE0k;l873QeG z7MuMJX@m1d&zC89Qcn7~B(zy_tS#e=2*K9wE#@=dXO)GOA65%Z+GCncH6WTHSdCiQ zR~g!;pq84V-TqiL!!9Fx2nV7OI_(t=0hniYljs0d!*#aJI!lD;u1Ixry z3Qw0=3x#WaG6Fu7ti?_GA>41ry)X^~L374vwWmofg<-X66>E(^+|;r7m`44~WTkqOsCurRqW(xxkcKZ&_q% z*lYKI!ZQqkuXLtEwAP}gjMO!w9?wG*%@Bwec(_<0KqxJiVBnToQDYchCWXkt;-qIV zl4?vNU-rC)?@u&)ISX5L5yZZ9XrWmn(zFPe!WXd6(c~%!%^3x_J#T{x^&DzT5+T-0 z#1Qs9er6bGqtWS=rcS$Vj2NYf_S4BIjJs+>?Jgu9ogr?3<$J4uV;QUx1hCw|@`E8& z=)eL9gN?g|{EV6_YF5*DuvrY7S_Q`4TtpvQu_%qGxd}t(30E*C+&b(tfqYg7!_KxZJP@YQ|jQ( zP(nCPQ3zj*$iqW7J|*^Jp-Ci8becF zyPdURXl!KppJ?825Z)n@}NrQ2gd zbjN<1E8svQe&7HHTtS5ZWH7O}yCaRhwQ^hxBTl zsCN2Z1Z)MZNLp=>3NZ`3e9t5qBR}FQ%rxv*R^us{b*I&G&9nrR(!HYr^DM19z zb_gO}cwC2Vgv%2iS9NAO>+@Z{V1;#N4$@m_TBl9gQ9x8; zYRh<4H{_Ppo8j;*Td+~1FIO=bGG*DdVQEzt{X98q0s%6dIiX3F@-p?vMolLos8}6? zkFgB`{2||9guXs-kZP-fsG1+dy`*NrAY2OvQsykM%(t9;o(vYSM^U&AJQ0~C67HnS z)S4RI&zEPm#}~Sza&KB6BM8J44B0?d5hBZJ8ub(bA?-}8H2fZLTcK>?W@8ZT?oaJu z&@mUY;mm6xtX1uyW*sh-=}-!26gVZ!Z0Q!~G;EICew)Tgj&DAp^4WTApfBx)EizCRS_QpdX5tWxG{{QOvNGKs_0oZR*{UX z594}`orlTPCE5d+?Mcf?5*F}Tk4TeziCr3D*25;4J1cq~xDg|w(}^1jDgs|Z%FD8; zWU!}Oj5)18@Lkt-d$PDBmh#P|sVrb&C3Ox~YL*59q$@d5-e=QaE;NU7-JsVb`tn zx;oj%rcx2oE8M)7G{nYAkAOo_s)qt8h4s8x>y29|!qIS{(1Ad>1_38cp0>+s!cH-< zQ#O@?ClZo6%yiun-~_3GY>disltME^(};dAoyW~Cs)N*sE}ClkV&+U4U)1B0nbw8? zZ_yp*X|K|9x|LR~xLj3CU8(9*Ny;ZstBF>|MIup~UC$lWCOu$NNK>gzBDJDL*|AaX zmOwOuZt-Y&Al53e%G7DNjXFxwOWH~;N#;{OUoI5mVUlRr0W~!(0n>34Tr|eEs5}`NNI`+GI9hh zT`oPTt>G#IBI?4blg9Y~WHeyLP;>~$l;&|cgo{I~UM^E!i7igSl{kZnjjm?|C6J0X z%1n`H@k)=K)_iX&SuNaEodlxUI2)4%h+B9Bq`1rFNV8R8DlQt)%q<64RI*3yjLJ%b zc#=^?l*|Cx#%cuoL_?#8qbTXAs@aV4>a?L4O>k2m(Lz|{hF;w(lPTdme_>sE8Hw zc{mXha@Cw{Z>ns;sCxOJDagG!UhsiySxhI2}UjBg4@idRhAtu`Ybjp)Q}z;4+jb^6+NCDS9^@XjL9%=!ys*h zm~ks`N#+>eWfQc5*>)s)en&Gp-be%c2aTKcIXdhuM~j9anO-Rhp&1A4GBygPXug1G z7>4zM)45(xJU@UcTp!#srX@iru#!Td-9@{DpjemaS_-hK%yMNT5B++D9fF&EN>_*w z@L%C@yf9X)c}WKz#ZiGS2AtaCfF|Qn4sEea35CTmJhe=*V#t`C>WypxLYbCPgrq4{ zsH0MS5LuNPL`UO^<;^x>+$^qi7-4}`Tn3h7HY^ocLWDV0Ef7HMR(YxxTXTH@uRuDG zZndf)LrP8astr6t#U+te`+eKh!JQ4Ng)!CWGJRr6+GBJ%Y9>%M6V)tMJ0Q2JI1c1V zb>xqyv`Ucm7@U33ESxr>VNt;oe&CmiBI25IzdH~I!+2@-Omp1AYX#D7G#YdY;{S|1 z)On!R*ck&`5IdV86RfB==R+jjClC zV`!=SKy-1}&G+Pbqd|d7CxyUr8k0iLSOBZfLz2jA0%xWM*o-d!;RWsXi(_iSnhkVD^mxhuiB~$Us-!C!Hj9yQ6-5i}K(=U?HK8`h#Vk^(U84~m z#c`xxSE{Dv*aN9R;xH8xKsvPfwORs=ZY!V(C$CHNfV-3*5-FX)@avN^98 zWquMM3@&6XH$-AT6b-B0@EetRq1c*2g27L<5j7WdBwt9n3y{PM!ZRe4wVJI`G0=N4 zv~XgW8uW%pKbg%;vL1mmGMu-dp@XJKr%x}1fDVUx4Ub(tvq+(Mx{jA{9-+yU#^Zuq8KyShB1{ld zTyT|uqUL(1ht-)bzBJ)hmw*~pplQX~OzbV19Gcg~s3TE2@C~9X2+k%o+8GyX+2&$M z*ONM@EgLqI&}~~uD=A2urg4AX4yP*F?v-GfnCJ6472zR@)Q5Fd-RJ?SNP*13X1!6Y zD;?D@>kE3y%T~9E=^RCqLIY&csR6}^v`S|kmNu;hjfHteTv4zH?v{g|9*q_qp#?7J zv-Sd{323cy8aYM`g5F`VZ&7I(9C(@H4-nu}LWhpj$(yhiaYlry;~1ua%en2f8|{ee z&uzwFp`z6)5X3^uK%CJBInOdMVR-`OXOZlPO@!=M^98cC5-830c{JlEj6jzef2nqR zx*?QNs@^now;gZ)+RF%^_6zpHtwbP~r>Y@PFb8RUP<13W>2kGVGa$tlq#MM`S_=Fg zBHgAjXAZLEaZk24X^$e=V}Kx>6qKaKhEikW6HAQ#d51D)y*BJg)kjb_DC zUD7q)DK{w*gxFdjOGWWJAOWe`64g@4c15xj`@q|(CLo)gSPThksW0b6nS?RP2N7;o ztq$1**~f=ViNy4UDGnR(CYQX_@I(c%`jfctCjJUoU3Iy`IT)g6m`L(&62WE7HxS?f z3pso)k{!S3w+UAr6-lsj=n3L@Wsvo*E;&3Oq23q)1yR2FI zo=FV@Z&|=%1=&+2x({-Z>Ke%O$J_mZQ=>cN5~OB^cA~lM3P>!?1Gx=t#vfZQJW)XY zzfXfiX)EpXW>W3~ogp?dkm3hoHFn%6RR!OXW>q%+ONBC ze-d`A)a#Q{BTRdN+r?Pkn}QVkh<0(PyP!ZSWPjOMfK2xxx2)sc-aKRp)6n{0|1$YO zzCwZI5+0X@yuav{3no7WS&OqK_?g9kqpK@~ZhD$A9*qXQ7Kl`)-P97B!f22*bP?qs zitB<7SyX_UFuSZZ`M#&o3=c+C zEsHVcb4yPgq6LYvPn28=nm`b4+sCI22EHMbC${^>t z7)*ip#R6V=3B>4(lrE+1R^A4&jtU*GS}Z=!tFYG*1P=VXq|S{DmFfvY(Hw$b>q%gaH;a#AY~fnoQ8K#V<-av%uyf2UGY+82V|_GI+#o z^L!LHz~CjloJ|oh1LRN8wY2G#XvEXNjaK2XIne;S zmXtG8ra%g&H{K+t)?=~WFhS~W4QI1f1eNpIFirF0mVgSw!hUKWjy{9v6AYHG$f2CT z(~i^B2_h;GBE9+3>U0~;8o>^#)mNT>Z2oT_a~I$TU~h_s`;U7iE_d(dw_YKm>?F2%@b*)u2A81X8OO(W)vuX(}GIeNdVQ?dSLZpL@1@ zvpwjM&i|b8e-_Vp`F_8j&-?v)f9Y-Y!{7efKMX(T8T-L60Z{kXy!{va%lCfq2VVc~ zd$QjN_ORW{;fufh{lof;Z}IQ@!uR~-zqfn&(Kl~?{7diH6+ix^x4&!u@|Qk&_zp|* zy@!`S__`0IXrhA8xD z=IS?i82`(cUil#3;9;EqtB*W!u%rK%^78vW_rCw>@#Qyu?=Sqm9|1D{pP%<_1r>~Q z@Na43nO2C{lZ^)Z+4fO4_^MIkKXHRUw-@Fo?ia+o3DNO-JybqKXQKghaWwB`P$pR zaeMj7+aEJt{`luMS?0@M`P`=8c6s>^KDX&B*30gZ?dyBU#ZUez`{ncld|C{Pu6SFMsu;hc7?$euw|8 z_ww(5Zd3VhzU;pHhd%#fedXKU{srp=fAg2#|32^Sx6=H{?*s48{#!5S55MV^Uwm7> z{dWA#-}}rD``H)1!<~M#N1N||nPHFKeW$no{QFtBEQmtQ{o+0TEYFTVX?qh$$&;PprF?b<9`q-DBeSetv z@a662Hhm>}`8QzG`(yv9-Tub0k6wPux4jX+>)pvJe%IT7dVTrJpFDhbU-_?ZFTdqO z@w+~MRlFloy$Rm!-#4ymRINq|d$4Kda$F{_K1I=grGceQwiVYhV7$ z&u#iox|i?yrab_t&ATh*|NRAiLa~3*;l18p9sf=L@*N)mZGQ4+d$?@g{JIa`9q(WF z!P`GNz5EP3eE-BhFn{hc`pLKR%ddKQgMIMsF8Z}^c^&@A=>@_HV!ZxlbOx`)R-M`+?8(@iX3S`Ul^<>EC_6{KV&;RsVsP18jP~zyJIX zynM&Yd$;*_{=t{O@x|Z&!Mh>!_kHlT{^@VSe)muP#HaS`{pkxM!|#9r?HkPSZ=mlt z-?S|gzVwsiAA0$VUwHFl@4o!M`l*{pc7dm77X0x#n0(|7|KlILdc(i>jR&6Z-|>a- zPTt$YrN95*eY_7p`nq@T4dP$_=Ff{h-jt>Azu*6-v;6&*-<%cjKY#2yqZ?3b{Q2vz zSdl-SJzyI={tG|wmj02KFMau=*+@2j^u*73&dIO7_niN& z9|6y~bN`g*JVDQS?}P zh}XJ!Q^nhSKkt?c6r>1@QlUn3>Tgw1YIX)jS>7lz=jyI8GzZH7csus>BY(}zt3rI}z#t7jDfl}1DLa(`< z*Qa8{WCIt^t|7@jzv-k)rCtx0dwsl3n6WG8n{gVPSe&OhAz^}#=T8dMBxNjKBR`b* z@Qz#rVnkP}q_GmyZ7Mxndj_Ifk)v9AVARmD!Cb6dmWDssgQZxBby^?G&1KCyjkqRT za-X+?SY$z~)uBCNQe|5avuSGA6k+BaJGC)j6YB?d}$oneqV zGw^*9GoU&H7l#HdBB&AUNH7z$nt>kLyhjI{MGtvT*TkMQa~4dtrDg<1M4ibkD8};G za4>vMJY@iz88v%;jLeM$tyxlrezhcru+9Mw@1=Mmn`8#hx5mO-I6b2WU?6(*=*W|I zmb)PB$-}SUJ?my=zp-061l!#*y75LGh5aTNdGyGX#mQ~?0W1vSka}xAOS@65se8OB zkaFe9O|3YDlBCzQBgtl5`|Jj~{V_Ta>G7Eu+dzDkWO2%sp7-GEP0|6{)m*WLA};ud zRQA54KG>@0Aw?;kLX{9!ZDP+=P^_hq_}x#>8cU+n3|7_^n0K#GaM_p)6tQD$!|lgo zOfB=?fV8!?(E_vc02 zgWTMQa!7E$0Cynbk^L2z2Sh>%Jx>xA4o>Ho+$Y!lDC?9(@_BxNTH79uV1tw`x=(N( zO^)??v-Cp@_WrTLEnFgrc)SGZalvp@xD&{Y3sN1u91#{Y>FpP!$`v>S)(zP+en5f933C0>1+iI;esg>OUet{8a0xZ>D4){ zu#An9Sihpv;+C2THCZY1&&eeS!8k9WT^$Qo+{{xQnpLBW;&?Kj>Y-^o8!QSaL%~n& zh2?W1R&}PIt=e{dL16YlYG3Y1mea#piVd9~B;e&4{QnYzeUnXP3IH##(l1~_B_y4@S+HCNil%J9~#Cpg1KpiEMXX=Yis^D|NR zaeQFmRAH31EF@`o>^aoaVwbAq-Em9dq!nDC7=NUu5t_^3G8Q$Yg#Bl&(}Xk$yI;%Y zX78M|+W_ze=u;c3kWco#K}^}q2v9K}H%TK`G@SWp(UKgy>ZzVik!?2hPeUirEs-vv zXc5E>J!^IfO3853uXnJg%nq$x+62>@rRFq9vb%+(1WAVKZy}z^46O8wn<`C zJC+DEtsDN6UM!VB%I%gU9I1si54cs1j(r{Nfq<$Q851;31E zVGsYzpv*y$ZE=yYLZKEdLBgrCOsC~k@k;hiolne`xIZIH$C}w^xcdl@3y-SneYnw5 zkqElZ>4{*VJ#v>Th7Mc|5F9RsLReb@YwnvBDP=vYB8^-lj&(t)2aTu*pNrAyZk(I5 za#U`14_{y#7OT@X+#*X23Q&nAEE?`8eTfNC{;W@D)4>Q!aH1SU?i!eXoDIVCR1F94 zAhk+gN9i8bTd;2PQ01-_F#41Z+>Cv~d$1|nWkPkph8sxY{Hj~ebSXx&QEdbMR4Erv zxE}KAaol2eBVX-6-0_%|?O8J6Y?tIx0?*whSI@uL=r!iXWhGt>qnZQIs`Bu@m7Ghh zy3IRy@$P06>8Kc%Y`X9b{p`?6ySsXQT=F_BMxaE%py+?AoqB%Dz54lf%DWz*GDsf7 zTr>1b-rI$bkh2Yyxvsc)koRhyO@7_x)LKx|08WmOBaOg?&JBB5i_j;fq9~uwm9FRy zC{}H0`B~H=HPlMa$!lo@O%A3pVvt@J?O}Ka?0PSDleyn0aK}RE#!PODbXo7Sx^MWI z!XER6f}Nznvky9z2C0o+63npPHW zpV!4X5@Z{(wQ*SQ%8kSx!mB<;;@08l{O&vS@o<}a?n2680iMrrOO7&i*I=hJcwRn-|N@eNHrZYU4-2CFl@t1WNKQ%LY!qn*hx0CBz?9}IPT+i1# zSArJvIWcvalys(?BM(SLmL#^sWt3Rd*$El^S)yd%(p1jsG45JSGNyNBRNSq@N(eyt3f8|6iJxU?eW zxQ3##<(#Y#HyVdTXO<2_F!m!i z9JL};{1l8z%+ugA7-sNc=fFl-!WPSBe!WL|d)zH~Gnew+k}icWgXT&(thX?~*E(vy z0Z*a@4M=h3S4urRPNC3Qj^$bRQ6$VKh1EtRg<^1rS@X$BL6CfsBZ4_8R056pknC5clJoM^qt%bR$;-xzROVH`+4&~?KeTLVT)2Xm#{ z3?HiVTVP%ZeR_!CZ#)UBMvwF|!C!B|^tj%j882B=hRSqoRJ`4aXU}5oWaEyaQEnbE zy@$?{)*BK?=h|ScS{D3e23A)Pmcn^cOHFTru-YRJZXSb0V=WtqB(n#m2PhJpJFRh! ztEaFv$Ts`jjfRW!8fRMqvsZ#=fh!yl_nFeYe%Jx;g>0o<)*yv-tjvlqpsy4mwsgGZ zteCo|+w8vPK~u)}x3hF0E5YU#J`H<+gDj1C9SSMr_FT7ubN-M&ds1(ShDFiwtlYP! z&Z8(~Qw{kw+^p8}il=L&{_IUA5R3?JGV0WvC}Iag?4-x~KB|+*Bo9lfhM?c!ES;=4 z2M)z9T-`ik*M))%Ye=;Pa*_}Yj+SeuYUKMT*TP6ZRvg)&emGic`#qGN?QyE3cg{z+SolM+Hh^$>IS|pnuL0Bw4=+-unNEDOHRpq z|L2*QJb&Jq$+s9hO6ljDn0#tP`{fO2;JN$dOqV{|soFQJG}lb%_Fl@BqyU>(Cv*eE zv_=f;__qIO2(POvW*cm{y^lh<0q+GP2{55;TCgPv;{-GB=j+s)2797@bl;C&*Ihb$ z&g-z?+*kgiu0Zp`Vnj;7S#4|nWK)mV!7FY9BJyJuj*VbAZV3(tm;zCfu#F`D-kBVI zYy3R20>7^VuO$MU?!uc{<-GVh09l$$mZ+LN!H-rNV)kz1$D`7_L1`6?y(H&b?@rl* zX~CRI02|?kg>kX&YYIl8Tsybv95UtV6vE|rE%UxFoY#4v17jKQ!mR6Jg@TiGhmH>J ztm2{ZMG|uj_KYXsy>a^~d{eE)9_F#vr&-f>J6nl+22JJ}kLd%<#==!F2maLx(efT1 zcS5MeQJIKrybRe!i$VKd;c#{MElkXBETFG)5RWQQu4zfkh?IknwQbOlzX$R6H zpB7pEWQ5Q{k0kAOQ;0x2NjqW(A|FA51sRr~I-*!VOvW&)130H7Si-$-Z%-Z6->&OJ zwa(3Gxn1205L8NI+TyXjPjBVztP+|mTtV{yd%sg6ig<)j*)Y!42pDp27NnxN^6ES? z;RPh4jqmo9;Ol!q&s%Tyb+;9IVy8R60I8wfx|x-;Sxo^|INd%RERJ|Sq~s!Ai`|Vf zHc?TCld(kKH+i=bpfokuQ2QgAPNi-690iqRFSaa9 zW=4~sS>LZHY8z8@gQ^6c?3zRs-C9&ERz$hf5AeX%#?qmP_z0}NBdxnq{XWXwbL$@3>3<94B4p3 zNyM8h=uU9L&sf}egHCUrjnA$q4uBVSQR{$nr)rf1ZMPR9kLAM#(OB9H+L5`XtWW0~Ol^uK$LGnRwcD-2g zXu9T!2sCe~x(5#xiRoML7K(igwTG4Ko(Jh-2Ajb~nTrb&)Kz)ea1{o>32w{8N_6bk z!WEROe0H=F!=J9DC|l8VEkJ{1fc_@k17^Vic2lMSTa}n9O{WQV-xXv?PdMAKJplTc zC$bl5utX-0(VQMz#9%i}pyqQ=xG{^8Q9x7M6&)cfB?t-6gVS7{MWC_8h%ltp8p^O* z%ry?NpzwGf5djH%7*IIj3bz*DRfKD-i2MvRkX;aAinmJLnf6y8cU{%^f>lo*oX96Q z_=6kRV&xL7q`LURASjD6?OAz*;?#Y9Fz?rMu$&8` zl!2+4o@YHEIQuJOs}g`vDI7>JXRJWfL%uD6H>)d)Nj>RT%d~lqruZ7=IzJ9FSkc3|bE2a)kjz+g!(eQi5X` zS}W&eduSV&HyE&R%~SA26#PI{T?r*tsFdRgH&8eDaIt&09Opsm%|nfL12gU7Fr3MC zgj=9ZJ{ejX`L7kVy>lIM(EMy71O%&XuhnOpWxLeo*iZdS%?~9$?Bm_tm@$btV49(V zEF-6Bnu^sZyGWh&LU4zGqe zn1kH8aeS-dBN_qrc8eA9>b_JkRn806+1$5ltG)XBT{PUvaTD}(3>=KA!#rXzw5|DC zGkSJ_Es04h-pwIbjuz5p19DSg+vEgp8jpk1(NOs?;n=Yt>|1jKcUkX%R^$WmhG_6(D{#IvX3^V=b>P#Wm5Tg}9 zgx!GzFlx*;NYAJS2ipV0T7+iS=vx*mP}_RQoJ<6u>RoXkYZ%zOi4l|iN@PPE=6Wk* z$$%TNxWxONS6VDoOP>4MFR^8wtD}(aWq7FuqveuPe0+?)$Y??^7o}Ko;)b(Cb@qI8~n3@=9SgIlOs&QayE;|y+0jObWp#Imxo8~* zHkjsk1gGqvt*RjRFzjs%x>hkpcFKR8<{UzvDN3 zR;>8z-~6$E^7daq_}kn4Uw-*BZ~y#XdHIju{G<0`kbm^S+uL7#`3s*s1m(w9-u}s- zfam_mPkN-$A(0SL3F6G{5yVPy*iQRm}GGum8nSX_F2}C@<6amd|{bD{ClfQQ;e>;|SMck{EC@B25(oy+t2;L>uJ`Via{P1F9kdlqU0+l% z6U{Vz%(Mi@8+{&h=&J$Td>JIPrxO5_~fQN;0=4g8yRT*s1(Ek zR^vHYQ>sAZVVYP92B^XWGj|E@vC2Kq%Tv)pdxVcbJg`|sJE)X4x(r2e=Rq(cXxO9u z3QY7y4sM6kD-J7x;3(@Z5LBf)) z&peRUQP4p!-ROf8?1OF4J)2Yq1m+v7sY{-FDJGn_#o)b!%0P4?wh577SaKB0&1!Z!*mDMi zEiksBEz=QTv<+8I$%-A))%mQ0UaM*?7&^@~=sEv&@}x4CPBzn!*(zgTZ=lb5gC>b^ z-V>XE*#Nj3Bt_XUDsw(zPjg4M-c*hOG?}=QD+j^$a`ky-1Xp$d4z??^Z#F^&?akfphdfcF?obEOLa#N}V)4Ll{PPpI2@rgbULL2sg5u>E`U%)?v`_3U0GR6Kw{xQqRQGMkIm)^tD&wuB6_-K?pu4a zVMk5B6X)l6R>$+siirRm<8ZMYkXgSmkNL8kml1tr?(QnU^%tMHElg)18(_+IKqzrH z=amJv)}sbJE4vdA0zd#}Eby?>EYob~`Q(un`J_LO$0H{xD@w9)Od>W%!>3Tn3Q2t! zwm1HH+f}MHDh|iYt3=;OU62h79AZ($Cp3S8Fhg;r6S zVhp#WF}t5OnvsAn(+iSt3aJAkuW~EoC?i2Hhz9oH<8i*+U?Fvh!FjtJx}yfVAUe6Z zS?Li_8wWLnQ9U>!J%>Q;o)3)=rQt-&_opMX#U*_`9|GEq)|qd(K%D@F-0Stf%7D;6 z-}f^3FkSqxD$#q*BK!G)@ZQ){HW-UCNJ<$h(>TRvQXTo#ra8&`b$nzN>D(HyCEj#P zQu7j&@ktA7;?l&%;)CNg>?4IUr16 z!?J1TU_BU{jJT`A+RzepenOV8lYr&g395{^&PhbZ0%QWo!4z>jdR>pd$rJ+tkA z^9ttqNRlzIk!5MZ04wm0?$H#5E5e)eh{Yhk&RyvRTzy;)mbqY`L9Z0+3t9lm< z8BhlaPl*BDP!=f;tS$@#5EtN+AZ`h2Y6W(?KdXm-$yAV(IamKIE^QY8QSxjLv}&G0 zxZP=+n_EQIeia`N4zv*jpX48a^OGNdpk0;`q<^{&%xPvhEFwh&!5Uy@Yb)wk`~+-> z+m_ET494&Q0eXdkfzp>lL`nd@=%rFj=XohK^se?BX_trrAUJwk+1>D(O&q2o3s_KN z8BGHym>?&RP^CNmzepcmxa=b^mm2RH-^v`O6!uM0A8y_%@53IrYR46~OR(2Oy0lOABt zd^8b72i(0~SxOj#)lzB~fQX1Tg9Fo3&B0SSPUU5nJ_fMlZPq(5ZE8qg2uG$N_(JOW z&Ig(Q zZp6?=k981>wBBqcQ`x)eQPF{Vl@6sKGQc`CbO2rfC~1=mdKUu`pm6aJ@}|Z!NTuCK zl~k5;32k}?DYGqga)oUk?qSoab2IN25d+pPA|5FKpHaS9q&$DPV{6|%R}WkrZ}4Fh zrNvSY5Q|b=i*bzBpmc4v70(R&3TJ_mraklxnk0^t-r$E2J^W&$mxasOP>~qogv|Sa zHjQW05#8u*1bWW0BAn0>#jjXG&rnnc3w5`Wk!_knZI7=B175^wCKL~$lZJye89iuBd-bN= zQxN20*L4=1RtUo=>0lO8-r-xK!!UGG0cU_|$xSJ7l9i7%{Nf@XmMZj{Dv-(t>wbah zvT(=Sf&{Lj(h|aq6v$(AE+ySL9kENf9%_D97Fe9av42XthM6L0SI{M^vyR%KENpN)m58;|{v)i72TBs_xiQIEDY)#+YDO# zrWLLL%rn3g0f4;vy_zlj_%3Q>Pho7>D91%M6NU%iwsj-JUR(hzL``#$9ypL*aEH4S zytNNVbwDUUzi^fjaL}H2FElE9cOKkWQd$wPGRpP7 zs2m@rZ1d_&B*_M5=UTAe=T2dldVf*N-R>Ib0CxAOMQ3*2?8HB*aqjzPdioF&?QcOeZ?wDsbzA#A9v@b2&cecJ!e> zXzKxS_+#Kwc=SB0eZrPWZysH~lOsUu^?eAijrF(&ox}qI=Rh(A*&VVnxr%lN9bJ>g z(ENVQhq7no7Oe=nIZ5`$n37>w))`C^u9>{dAPCn)80bi(jq^Pyabv5BX<=PLh{wp4 zXV5LA5%c8eRhGyGWHq#gd=r3ksuuGur5|c=3U}247QqGaprk^- zz1*M;%b+{kfS`f`@35+(&O?G6VWH9nzV6iY7Q}&ML(mv7V~ouWXlslC5hRa61qFry z*ZH$pAx^$VkVT_?@Ci8!0LZXNriQ>{U~LGH_ergT;(@7N+x5B>u`J1ydZB8*h-QSF6xYZl1TY z;kyUG=*Cq+c%($+jopzjDjY8n6}Q^zhU#j3S_KAFEvcy6Y)(hdHKAc#(H*{D0HspH zt-)i98Qys8kFe~B2%uv_bion7wW|U;32E1sENk8R)NwQ@{eXV zJmA;G<%{dwk5Ebh5z%9WScNXW2hf@jL7nIZ4)ojv@9wvCDs5o4DKHZ3^Pl9nvuTDWuA&ea}vaV~b7_5SJga%b}QNhVVO z++W$I?d06PoUY#4wS#R48KKR3cPU8|5R5pLV1 zQF62$jW-2nX4WGnSN&F5}2*UH`CZ_E+I z+?%ncdZtg1ne5Gq&Ce__^(=v^16{K)VwuxLo_iR&`{uclC)T0p#By0%93qhSr9oF; zx{YZ!5Pj4?Hv2nNvOq*qq^p+j6%?%peJbm@SFp7TQ*-NX6dHn%3O_0fGj`@>*E>~% zqS#JTbdw2WsDiaLXkaoD*(O1f0LH*85I|ez`IC(F_F9&DWnsgd6%5qt+0VcFS5CrTUmiP9TF^BO0&KFhJeSK3&`ZhSt3&?H5f8|G<{koJxno-pJfCgVCrl<^pByJA$_8|)2Qd|1a&m)<-33Rnr|-o@XkUwq z;MgJ72D>IBMKaVFbT$vV?j2gpQjN3EByQ{DogTLfG^h5y*9dD_E?m?L^_9L)!%dql zj~4B~Fr#ReybR42c4_wn-x1`*B%}%F&^cC?M1El7c_e}b7+715C(eC?HfS8E0)B>! zrg@o+VwaJkJOL|DJj6^FoqP_!WE}jo5SV&B+rDQ2$S3Dt>wd{OGaUNTp@K=$;EU=` zr=IWu9oQT$3pzdPLc~E)seq965UX-(p5w*1L-WZR2?3|v!6L^MXg$th32IbEQ~XCo zHifw8a*a(3VV}fwTNW|f#jW8p5IQ0Vo^Rt@NDDlxk_9)-J8f{_G&S+O^9CZQfslQi z#RaUKKL1n;s_nQwB3Ct=-Ixk#sA&`Kq@BL8)`x@>jgak$8P8m3RCyAieE9M64p!Bxx0E>@St@V6W!kMsl| zp!??p=ox%Yh0YW%D8vY(bD3?03|5oTb~~CFsDOzHl@lJhK?V{I&{QDKA6Zb#kRIex z7zqsPT)!*t$5UNX*{p(=b;BAa!=Yd~%vK~f%Hygon|jdto)(pHYlJz0u}g%_y74^U z$gynIb7-pLUJ#Ns?HS4ijX+{a-ktoiN@61Cl{=tTwz7ZK5ElxG;x-nI$@c2G1ody6 zYEj}q5-PQ!cw*BOEevj89!cT2wiPURkwpRrT$q&23ZZ=R8}FfUTlZNRWma3-_BF#Q z=({qgpe5;!-fvQ}If~{~9E|hpwM#=vQ>d+?I_5FBt11`Yrqbc)-67tNr!qu?>zS&2`^g>l%n2$-dm1h+XxP90 zTo3yMq|nnXCz5Tm02&|Go0m{*4uNV?!AZzX?>>%j!-ji)i58Z2paodf9?X?;s(dKV z+p)e4c9@z^)oXtsxSv)13IPxx`v=(|o7u@x^c+zGs~DW61qXUdh^N%f>IS`-zz`$j zYfCmbSE!E}gLR(mi^2giISxk3xCWVQfepB}45ld|sMdSvYvvFy{wR2-6K{YWv>a?` z=Vm|#KOkvmfd@zc)nIn*P9~Gb$B-5ZTG1A3hd=L=u#4e~zCQ8%`otSq_YXsJ8%PDc z6z;Jwz@eX-PmW7lKBmo#;;n%E7u@&tifkZNJ0Q^aD_N)fY= z82lyZNWCT?yz8ij0~HTPzym2asb&cWOfm7^_5%`eQ%DM`2@1lQW>3mv3S|$XQeh8= z?ixnhc_{7ZW$A-@(S^XH_{idIl!+!w-vW3p-LyK}0h$TD+|Znb(ZFqrZ5*{oVRT4z z?m))Vb;*dXCC_<0fg4;&SnZn`mlbQ<*v}vcTiBvU*B*<+jReczkO2?)!=(4?V7=9i z1(4RgAe1EMnZ%u@YqPr}uHrx<%^-`&mbO8bIh6;RK>>KU%&N=5ZGIO5<7pr5fi;z_YyS}QIt&$xzCf}u=y44-0}wQ|9fK@^ z28R!3rt7Vyb4GQQ<^Y&sRt_`$iUh(y&6$8~)PBGxhHIMi`%OJewOJE_ho7Gm3&1bYgrT0pi0$KN~_xb@bsWYe&%pH6ZR zfaZp+2Y_$mR{lt#;dFoi#I46PYhCWiq|}Sy1YvDw__kD4&zE@-!If}MxM{878}AXp z$k*nL7h_33*bspPVkY}WEG0}fT|ob{+hb9N4-kV#NRA;iv4w(EE{sBOc=mD~44p8k zHK0>>0Dy8k8*&hRtFVx&G=XuMERZCAR|8s_!F_F-^x)tMa+t~u1SmzD8e9U~vC@l! zEtFdi?;=Qg#g|GisneROHAjKPrA>E`p+RM%kPEv3X6yUw?tC8Q;mIigZ9M>FvWJ{1 zi+91yn9*n9GI_G+WzM~VMd*4b-TMlK!_!)f0@)3TznmK@ueyB491oZkLzoC_j;u5B&Ou<3Pk5X+7m>bDXv_11{iqS$r;hi<`0i}$MS%j#`VT_)M79NQjYJoh}zR_ zwXGjuzLajPeWO@BZ9#J~!Xm8nlB2NLj{*0tU~qcMj}_#iP&g1k?~wcwWO%n9i|D1(_Hxixe! zZy=8gXqoY(AmrmnHUjFw)mrPFdXfOKN769Nhm!8kG-cc%+-fue7DxwOkZKM8P?LSS)9*xY`9aUd)MbiNbAob_EFA*k)| zYw>i0y1qXn-N|`aK--hdMZ6TzgSb86^hCBNuBb%N*Q;}P(N8+p+*k1i81xQ+Z!eAk zvf3o1d5Gi#zU`ev5;iZXWdlJruzC_UoearBd{s`kJzoo$VCGW8RTuUE(0g6H+VUc_ zM+oK$vz>1E=i?+wGY08lmIL44JVQ~b$yv66%*k|J<4MV4a>#AQVg(sctKrbc{0%oS zi1hI9bpb(Gx^RO!k%=$7X~-rdni$Y!$`8y#Ffmdz^narEV zdJS{0MGX#%b)2|kfe8ckJZQ&Wa?;0%je2=?5HE@vh>)$jIT7Ut~FK}PhOTZx` z=h5_{FmE7Ay19c%m9L{MSzM?&zF-r?q#m#bz6P|6;0-O zIup9pD;K}G6k-v!H24#&E{Z4a zSK(boFMc8|;*?eiNIuHbf!R9g0)?X>@{k?dH9p6)CMQO?(N+|Fjp<|5x{nEH{oZj0s}VbhP1Q<8IU73F31nSh*m`IQe$xJ+=l9;gmc%9uXPs=IJT5JDQ;H zqfJs35JwEfhxDjB;QJlRI@ic`T<;-E8oPnA*g3an$3UTY>FFx~GA0RhojC}d$8SSb zgeW%lY;kXaERPK7ibX__&enuLL&ABx)Kql!QxgZy=o%{v*5$RPkmILY?;0P6a38d> z`BW;bshMcOJX7cT9>9s79C;0w!cRUFN#N+W0V2cfpUFuJwy|PM3 z7!cPxr%gVBM8k>lOxJ}ogqFAzr8zH5^VI~-pb`qsYodlZ5&f@O~_wrgK5 zg*0u$Fo)FoStKzNod5$xR2NJ=0;y(85yJY6(XPI~jV^nKU+R`m$Ai*9(jw|(a915b z4?r78I$q`(6V`D6e(@rVo`b1+&_gr)(pF%;+UG=gD<$@%0;gX_lyZB<8D4PBV>3F2 zriip7qhbGAayb_6A#(QaB369}Nl%dXi84r_slsN_p=K(qf!D0LuRD%`pa%vzM!d#^ z-#J_KeOFx3<|m<(1zxJ=Wu2y5s5_~UNvoTT_c$;^K$;zB zgN;}Zi{#xUbJNUYAVI2JsHoYoS?RaeTj<(VK!>!ri_B(g?x3t{uZ8LP4uaJn!Os~8 z(LAY7QY@~EJo`{prv+A2H@)|#Ly+^)_E=6XhHY@ z7QEzP+MZj(M7*O6{uPmM2bBX0l@jUAh7Zg!j1TL|(ZBDp>x zZeYWQJD5x}IlV2D(!Af;zF|!Ec!hYEZmuHtAaR(a1H1`><8aGUYy+>+C>|h!nGNGx z-rALkp3s_2ogr^XkP}3QZR(%)n*^-$Aish;KMJC_PxTnU3;n*Ovd@o}f*RSt&O zb|Aoq7Q=*k78no}%6D!fNk=Fex&pofGnB5POQ8}xh9z68VK#fc;&bN6fIYViq%k@s zERw1y6q-B8sfeH$J=`1buH7?%X1ox{@7Wt zD3u!p0!D_D!%!o=0aGNCbp+iJHc<};-38IY?r_9b-Q$S@ws>|1V2`N@QFAx%?kxb^ zMYK{PZ*)iY51X3Et+zU-V%jXG2F4PIO4;X93E8LacIM-bnTs;g-QI6dDW)(G86Ks= z^+62+DHA$PZc!cGRny7_Gp0jBm(Ud%-_nf}#M3Pk2z}r<`}vd~teTBiLVL(-4$P-^ zG`NMIGc(8xM9&0qAt1PL3_ATK=jL2hfY|`bVG??Y-ZXQBidyBn3&TNsnMsS}x z3RQMIsoPE^w{#}eo?>pj$(9bmP*qd(L<@BY{bP3IfqQX^q@yJrmt)m&0N4pAGIW~P ztHZQoTpAl5Q9}- zQW*u5+uDxNyHD?}%b6RonH!qod9A51;>ga3)k?P!0n$ht0JpP)jIVZ`G2IURVCJ|7 z-PzZ!5U4%CR0>?|sIEgO|25 zEXV#o&fYzCetf;}>hGJ2dXm$m-E&DGX*o&g<!1obm}p zj#L1mFVuofHex&UgIt!Kg4FX}gQE1}U*kI%4x%%}13-ub z$_|?g!Gxwn3Z~vr=*|~}Envs*csFgD^_@+Krh?O{n2-yBS@gUv}kz~B*CRorS*|ZHCFa>bE;!|I{*sc0Y!iBQ6~9owWv=41yn_9kAS>V zyFFlg%=;5C?l#BvoYb4CeaZkrQ(lrrJkb_qYC>q!AGwA&mb>$n>(hkRqpW3X*kIfU=c8e(_+`49`& z9c1$sAFdm@=%=R@rVs*Xp-^|)NO0d=t*sND0td=RS|d&`6}%ddhdV+A4XIFgFr$DjGp@xR-zDz4*kUf z&j0d&%_a<`#&LZr9PHIESs=P@A#AAOCYTKB=WKO}@A|EV(039%dtm#b5@>D$)MpQz z3oWdu;59PmJ0xXxuE3VZ*2Zgq8v@lNk}^$$B=TuAfcTYyvVMV@hIHceZ9qUg*=1%s z!&tizTwR|Xo&<{^r9vm0l}aeb=@j4I(eiR`Q8m<#k+QtZ2D^R}fOgNmLp#`5AU>Uk z74)aWGdSu5n4%rnhL5iW%u7cS3F@dNLE z$LpUjv|U~PF>F`n|NZV$+tp{BF8|&~Uu?Sqo6c`*yZWh*fAJr_``vFJ{jKkH{=~lt z#N}W6jsfcLE1!C5{?_}y`J3#Dsuz&X-z2m>~y~q0I`=9wO z0G$8WZ@v5GyWjb~{ylF$|9h@4&(-H1JN*_AvH#Ww!_pTV`GeW&-8Wtj{)I0(^q;lg z{)aDm*!tZY^^W-Puy6CIKYP@VKJjRM_e%zI*WZ6VaQoXH@BQ`d?SK591A?LIkN&`$ z_3rfByvdLM?hm|u^Bug)4_2!8>JPnr7oj}-p|`*D)_&y!S@|>GsPFLUcW?ElAJZp< z^8et6-|Ba7zx}JE{-6DwZ~yUkzWd!@daFPB<*$6+_66_m#fpCSZ-4Hz zeDvo(ed1sK%KKk?|HEH@|I36+t-ah{1Eq?s|Cx7hi zSHJb$>kBV__5DBmvA2KgGe3ISZqQ}u+c|+c~ z50Cry<@(uI0-wuY|G}U9(vQPH^Wmj@^2vPWFJ70#fBn(ntKaj1|Lw27Id8$|p7{5E z=aV^n_5FYK5563H(F1?&Yj3~sR(zianqPk-zvzJ<2>1W;_j#cIe|`C1`l1JZICsDD z+duHn{p1__2Cw*o*E9Rc@4AIP{-vLMIu_t4M&r=MQxpO__o??3tWs~`Qf z?>%cDE{y-;b1wz|pFe%?|b{XcOU)w_rA6t z`^MWh-+sr}K6}v~$xol`*WXXyc>DLi^)&G1{ow}zj_>`<+wS$v>(u?kXa4##M$SL} z@&EcWZ$JL-?PtICeNJHb&wqNLPfp+;{kgaN4gcEfHN%(pzy8U$ACB|>PyEwwKl&4& zJ;INEhqeCx7k~fTzxtm0{cnHaZTI~S_^bNE%E#Y% zDc|_|`yXZBegh8i;Whl?XMXjWbMyb*?%Z&1fBc8v|J^_N_Ji-OzxYEx{h3Mq3ljs} zuYb<)oPIkoVDss@vcC2H{{I-B|M7gtC;sjKH8S99yH9+`|1JOar=B3H|_~p~A+^mB!#Zr%z&m9ol-Lxf>n^|VOAIQ514W5X$fo6Jien0U(Q@DOXD9U-3g0J)KrBiya z(Ikja1J?Jj4$?*`aeW_yZMM0glD9wU-ow$k6}I7w0xIn3588i$- zFm7^>Vs|@FmjUo9?Xz~ZeB8+?3aPj&CZuHLQJvJ3^VZE&(Dgx5+ZDlh=RFZi+FmiX z!oVI8VGS2lG_(+F5#|cHt)}WapV$dM@$Cq~XSo<;6pKW*1cubze1)1VV7v0XY&}r` zyf4S;8iYzG@iRWgz;T?bBq4|aMIcv>1=jmkyfV9do=*JqV3UQlcAHT?wS}Wm6sYB| z_dq&~!Tw=&9&@4HGdSp6kVi^PWC*9hZNzl?jq>NENtT;+O&5_b`y;qaEmaMwzB*BoVlB z>Qk4pv)VTN94>yiH}92ZO?UPp^1SH}c%Yv_9_cr``+S0;1E5+^$Y)jRf9}zA7`5yZ~_9 z$Ad(o=F!82mD@PT$^+=}*5?gwtd6R1GD_Rpph83no1KIlgn5*yRQK1|4UjpS1)+l? zYTU#q4Wk~;mqp~15$sEwreaWrb{^=r{QL51Bm}Lr5vhxxZw1HQd+23=%+1oKIMCH* zkT#q6kjWmE%It>~M#j2nj$S%wGj#A&3#3S$VHd%H1k|Q4cvf8kGO@yoyTC%+p@1lr zk|_n)(57wT!)$DaC_c7@$x7VLSbm0&)21F7MZ8@APB^ojtIxoUfQp-Y@t-> zT*tItuTa!%bICB{+&k1!YJ-o&fxP|UCzjCYKXrBt#OK;$E;!LNcr;KFJdMS(RwX><6i)}Bf zOU14u80|;sHp1^Ox*cseG7>ioiuA2a5c0At4#g}`h6;Qc#54(6 zZ2?Y|@P1nw4k#Rq`-TJ2ymSP!Y4vBuj}CsMMx_~ki_fcYk7#SLpG9;8AWZ|m2!5w9mcb++)NW? zP89V-w$81_u1=z(Ap?;ADBJNQh#yh%NSR3v)9nbIaf>XBtO!}e+!}E6k&$KE6^51w zf`E(ogy$JJm=zOJ%aKF^FJ|kX!`Iow)K#djc9v6B(Y1$(NM=|TYTj_ys|HL}&U5LC zM8Q$iaE3108|eZPx?HU5MK@gv7?}(Q9QKUgT4x zhQ(Dmd6xlJM%TJH-@>iFiq&0fV#MuX=Y;l}u}|QF0D@O5gKwq@p3r(__}3UcVNsGe z$4i;{45@5HfxAs72n*)sQ@*peIJ3?AAxSQHcY>2#Qp;|t>=0CI<%$8gZgktPih9{= z8&=#GwgFwo2|&fQ8-gVrJ9EjfG|iEO)aP?QKzEk$UaPIg(^L2j?KF`buU9b)eU2<% zCm1w%VK6mSWgI&IDHTC+Wmy)|dZx*ABZ)!=jx|2R!HUH!KnlEEP65_w7Of+YjlKt9 zw{2OVdut-sEX1_{|Ig(TJas2cCpMxP1oWcEPP{%xCV$OJ)ZqI*$dC0J#75-}2C3Z| z+MkgHp{ejn$^}jA9ldtWV8KgiY}dB(u)+Q2uAzD^(})F_!;CO{KP4!2-VHbGc1b)H zJmBqRD{L~QC)x&cH!rup21_sZZb_HzR6k#%k}kr#JVCGL%5br`cZtoQ*#I`J8}4w_ z9)N1*85sYPO!QX%6;0hp$31x zIgkL0aISSL^W((C4)|v7fbl%BGxXu=mgbo@^84hJFJ(B<*bU6YHxE8zBp{swtdMiQ z)5V?yROEpbhB?v42%qjFJ{5=aY0+$~Q&~K4Evtsbg;8!+!thw1XK)TN6$oV^UIlz% zHOR4=@2UFKz`eox-~Do`Dg$<1USRL%{qClRtaEnl_#oC8SX8*4B7Q^rC6tc52k;+*Ytw zJSg@Kqg&a#aFuFrz&@(hhtvjVOhyPBYP3N{N#la<_!l2mtrI+X{Th<{@O73}=L|L; zyK7@6*M|rw)?SqR+ipo^upd)R(Bb6G$ZjXj4Dhlf=rib$!xv;qDxM>bFR&EkJlG)= zmdCX)QfmIHfLqI6ntQkaor&fw zGC~@~M_gD-Zsn2aKkO?Vh3=JjJ>OD`zQ8ulIA~A&vcBAR0_?ISLAWUiXh{#wcAeU` z>fTEJici*Tfx{b8lJ5`602?l~KBT+2FAqC#UpVU$6KJ#SwRsZ~lAfKAGJtK#c%vof z+>&cB^Tj!);M&=M+in9P^yXGqoDYkgsP>}JvGr?|J+XA1>z%8l?G#toXIKbJ12u15UdCTG4Z>uc{ zzM~z-ZxK$TPY{^{S7=eu4PcUu4!p%y*`U!JR~0B2ib!Of!yU*%YE8I%2y2j!{MwUn zz1P`uyQMY|M|N&w*xXImTMV}CY705;G=rTfo$j(jK%CfdWna%IQL~99f>_CBVYRvI4n)rfx`L<)^Ee;XuT7gjF!`N!9Elk z_zz_g$GL8m6kfJaw^B}jvz zFK&Zr*OCt6H{^iMv_fC7EAZs&awO@6phyVb0XNUStLrw5F3hg&0U=H$uD;TrIvCFe z1Gus9gm=eC2%8d7?U7Xsr2kBom3jB1B+(K09hZ9zFKDOgmaRM%P~Cl~7nJ}&SVQRI ztvVJ1TvRF!6Ae@@YB}&$n?RwJP&=Vra~ihOmhv16zwNK*&9S^KtO6W@;5aW8jxB9GD0jd8}MCnKBvd?GWYv0#H}?Sj|Z zPQC`BdUU3-u@*R=N`V04uFkZuX6&xHQWH9$I>4Vck~Y_-*KyWY;vtC^x5j8h@QQa!&+GoaY7gYNIo z5v*mB*QixP3W9gJ8h6OM?_I-TQ#18k-kjtetj?y9KfryG?{&rPc~u_3JZuW3~XjMMQPkkC^v#^8o%za&C`kIYL~M#1FSn0uM!5+V{i| zZdWZdIJ&o5T*8hN#3|UIxdl0{3Y;PHD!_n6TG**R3b26k&O|+BHwf|+Qk7v6{umeM zXl|JjMG9Xjs#mdc+u`f`p_$nSYz3Ywnyr}4@MsGX!g2-K050H6<|2mZX?hk zTxX@u2z&2!Uz4<+NI+tNQf#+E)S5SB_3W>`g_pRjS}LGpr}@#xS1Cr9m#773a^6v- zKI=e-d>mN^ zU_XmKi91@|_p*1xFCtEy1#OHzyqjL$ZCSGcMg98j0&4wcW(1zFGa4NU<^dYv2sXcD zJm_^8jkJEFi9<}W zDx0uK6(;n~J*D%sMq-g*D4lqK{r~6#cjt&~+&frnx(&72AFbp13FDA9r{opL4KZu59Z4KTWeJ51NpD1qzlnryd+s$K<95%N>LCxX`3uDP)n@lW_HL_W*`f;aiihFZ6C0^VM2st-^ zqywBiadnR{619T2R|g17$n#((xfX3pLWn?hjcoGs)J7@TgEp)E$_|zpd3g06LYOy@ zA?xwE&#MIw0#0WHARAB%I0k$$(bjbys{<>Y4lOE_H2}G@1QcaE)Njq!rpTLlyC0!G zXdr{5c~u#~q}E8Dxp!E>=8nO@lL6hX5ZI*FcpSFo9=!B6&<%%Oh$iNU?IFYwg0&Dr zmIbY)Rz-)GpPA4HLptHvO zJ|=V7OO1H)*rGvB1ha5Dn63CC1Z2pmLuH+PJx_Nz_fY%40FBH`DQSUCVYFuJ%^4^j zqKklZdi{i@FVc-N0|*2WgtMf^7$Tt$(>a3D(IrB|-J@Nphsus&-Xkw;aaX(sIeUQj zFbI<^V1hh&T8PKZeZa>xPK*#AyfMs@m9t`zKnR>{t{>Kg7hXHG6awXJm(Ye>XRJyq zo?4KA-+_h*P}a;cc6UKjB}?F>4*D>kU(G+kW%(;q8>T^|$C_YI7w{tNuRSG_;~SW~jR`zRD?o{Pt&)Ff%`as$~k##n+dbk~zI=p&0~BJwAO zWE?)l#Mr|!SYBy4vVX*s!-IccF}Q;iLNT1&xdz#hRCXi7VeY`V0Z)oX88gMs}@ zA5(psV29jWWg++~$UCtuAYE&J0KZ&WS#HI82eRZ-4t$Ptlxp#_%B{5-0D$xLd9Pqs zy6W?OLkUtECXu4=0;5h>_V+dJ_#iHV}s3gxTXD=zRiR)-3Lst&0bQ!M+ zHNy7_gwg4wB&4ua63}sN^D)iqU=x#csI(_CEDX0i>_h=G1?UYZ_;_sv_}pfTad+R* z&$h5lb~@z5O!4vnnx=6C5J+85kaOYk#T~FRc^Ewjc87qOKcYNk0;uRHnY<~PdqLO) z0y&EEAkW68u1@maWreAsha6nPqy~M4>;Wz0lHpDZF=#QdljsF!hn_v&a~9>(k_X*# zAtWb-^ku_$j#dgq+KSsOFl3)jWX(ef&Uqqu06Z>^b}=h=_gA>NIbzwnc9t!NndfxZajydsN6yYS7we9Sf8+Ti`GlNO0cVI=JW^?@1wv*B6+zTCO^^aVgOayP`;tBp@5Ptxj_KOMeW( zec5Ep(>3EBrJ`NsDj*jMc5#5l7191soKBF8zN}&M>r~XlY+Z|hygXRo4UqHfU8_dv zu7{^O?UPJPW;6&nLhQP&lfYg-2k+6k_e=#o0_?yUBe)wwepI$PQC#jUI5!ua;1Trf z?Yo}$7!i>u*aSiA4g$AN8i`zUQ`t@g1bxtF=-W!z%@ZmJW2R_jCgGpBx`+lB}V|CqCOWYi;eg zx>k_0bb*R{a;?(?Q;Z1^xTd$8w} zoLslfPdMr?g@T100uKIIH024v$LLt*S4Uy@*sw+wf2JJpN_)<}bDpWfIklXLupMIj zG2F6J-$8e0!V_+hh@q(kjmCg2W9U_8mZW>JfrKfOBrb2)1E^AR#LSMyJ@RS{+HbEw z#D{PcAKKvc*nO=4Nxo;7ZLRlYM^&WhG8X>56#Lj;HdR2(A!(+6e0<JY17ia;FXC$3O z`l1DE9x9)A$4ck}jH-l6;owd)0DOpQWQDY(zkM=WWIxc+%4E*Hy^d(&jN~h@1DfUm zQ`ajR+7xm!!80zR>S@7jZlv2L?L}~8VJpuCxWlW`85N-sX$S^!tzg9|N&M0Wo%bMi zL|{;nGr?vmKxZFJi@*S7%@DGK9*#|t{PU1py6g^n%Wmaj$NSZmC16TmO?Yos)t3r?P!Ot){+) ziib=*z+ciDMY0BRX|f~%(gT(NOMpYXNdRHQTuZ3%jJumXQMXkL>BL7^R>}MY*Dx-_ z1L?JO2J#N^(pYq`j&IM#0b&x_t9a#9#~qpbdV)&~8R*LP zS~9GBAVYqE{Jk^vJiA%BDcg-{FRkusHnJ}=oj%}0|2l1aNFKfIqj! zNoZ)c+*p@6jc+4eT^R93g9e0x9N!1go*_A!WklD)&}{cuE_Y}&K^eZEL`{hx6k*@P z&=}1M-um5>@R`&@X=2}3COykXC?5gIX&4J=-<{;R#9hLTpo3bnQ+kHB@y>xjX{x?* zlp~xH?UFQ$7)_U!OfqA(mu_z5yDMQEpxsNr!@chG^9X6WV+sT=nrTU=389wc&L=fG zFk$c%MPmk&d$gipd34SIplvd28!i0=ZrFAhAzIl3&CoD-er%{1in?G1@>x0M62W&Y zXeCT(x5x#w_$t_0iFm3sP@>-NkN!ozL-)&30L*X;xvRe*>a=R~>v7iCm(b7fZZx0O zooVAGYeWfHv()BFGPeI%txM=YF;%kLF2j_ZXj7yi!r>0P*Xgj{jH?uuUmE~tLjZSl z&$rtMIx~TsK(A@TBt6iFwwTcA18l~jLW^$zzJq(M1|@cBbuBk84rmyz57#yIteKm9 z5Rw!~`OBCpsT!?UQ$yf~dM_|!zP#3XkCn0qkKH9`XUS(7({Hs~lhe6D1BZD#aXO7# z8sHe9`FnKxeh;&I2vs&VAh!1~MqEs7lWlP>Z(%;%WmW{zgp9`9TRKbJ{X{ z;?{nUI`J?2EXyb->z1Q9;iXrRL)+7ak_fF@Ep(9KBw9u@xgRplj)Tn#;&{g;Nk+x- zg{FT%Fv6`oD<3K%(XBWHHAqZGiEO{&>!bf8OR%)i&o-}R1g&0GO@kF0ZGbmuvG4OFl$QY};rIQqg`Mgvrpp7T)H>@PE@S`dQDcB?#Vru?|3aUed* zKE>BwxDPU&9mwA3ZH1QeS%ZoxW%z4r#{o)_@_NzpBEcTFmyRL<(@G8A%I1I+CYClhg!W^QCnT|BBl^G9%?cP)e$CEE>}l(BW0Gz7Jf)1Kd`YeYVu z5gFX{CwixX^sIeu#he-ccNY-0&z2jJXsO1Q;S!I19)DUusEW8?4V+qt;X5% zM@HL^xalIUoPDU4n`qfkH6Lo=8Mkf?$3YV2$zTlj~$q$3#~_0hhbvxc1F~BLx>64kqDvR&3@=c9BGeVw z9B3yNlzAojkSMOUH<$@;Rl)oto zl`r2o*ug@=uJy@Q8{j?=!Q=f`*1+W7T zUqw70D{#twxGq$Rlr}gNt@kwWTEMc4?oE~jWt2ZdeqrqD&ub46f%dJuLSl>oF%8Jt zSkVw|uqP2|4$LcPgW{f(YR^31q?>X+NT*RgM(4hPXk|W@vfOz!Bj+m+Np>nKL4{5< za7jNh_aQ2l72vG$dbZMJW@tItU%bsVvbRTn2N{(1U_ED~oL3ooorX)=ca2&lXaz0E zV3IcLQuHcud+d7k?vFlP-L|H@gea8La9(3oag9p!9(!bHNZ@JZu`zmkIEyg^mU~2@ zp_fVL3(NFNce9qWg|LS!#J)K9UAyZbHVBgFjpyLMIAn_`M_ySUT|vI#qxGuS^%wQN z?N?p~y605Bll2WxI(e2=ub=TSMAus)+?ksHxWE9=13zD1aXP*PJSZhcT{p*RMI5va zwl;*kV-{F!Le%sXSSzz3Z(gkz5Fs_~lV=X{!1Fth&{)XM61vprz{$^tk|ga_iG$H$ zD4IjQ)7d>_ST#a0&$a{!(>2LeJyi5E(588d69>j|teaWQtodQa>XJ4fPTnCk*bCP8 z37}Lo^u?RqX^5a4Gf#~INtqG(^xXz48Q6)~BA_DfAu#41;{zB#;aUilHez06AtfF>HVGb%;H|i~y-D#t2ZXMOwH&W|``GO($g=RElY;ZS!oo z=MzRH4UEtccm2&PC(B!>5KyaKjB1r`VijNS4Yhv~HP#69-9$i_=t zF>z??O0SU<);69EV`Q` z?u8lNs_M5F6$bZ=H}|T4Mq|q_{H!w%r2ohgYY%|U$y|+us$?hk_@HM#>A><4#wJwD zA^9h?qW;MrR}YY|<0XDn4|yvHX$~~v^8W@TS{{h&u}Mu<7RX;pm@|=m4eRUb$%?O##C^035`+EBGZ6Hppz2MD8AHm#u)an>Jm-3=coO zKx<66;rZEFYE)p&TY=nlyvaL*0qEoqph$?W3SgQX;~+ls{xOb)J&MW}{Y2%RYSBI8 zuNZe|B8;$4c6>nBzzDN~5pMkeZ^B!$Ik_Fh2_3KFitKL99e+RvP)IDWDr!B=!FFM3 zNC%X%9(qN~!G4{iE1Hqp2%@>Xkjh1tyV0R@tk|K#-?Uyu}3_4&<| zMvz`XO0w@Q{t8;BkY9~zow9CFuRO%C;x;L$b!NK@1w6RJ$za$AO!i&{IcIlFb8jBi zYGrbk)e=NrY(N8Ggv!SRkeb&^JVSfg!E>fn(y$#+t=?-Ik%W%cXa>YihrmItPN7ZqWB*PfToMo|{Ep$f7yB9GN0C zQ~R8B_}wC^c=!E9{r>O`8s1|pLS?;=OAm) zez)b$z!s%$Gw28}i+-}Q%PtMKJob!$lN?PskvH+mU?Uz-oG=+*D_A=vL%${OGti{Z z0F!_TJj0wGVm)S_!XcA|4sO{YROv~0tTyHB)4&cv+M}!Jpp9m2n^c*#jpEp0AX3H} zA4YnhF#IukB!yl_A9&~m5rO`zgS3=#F>uhw$o|3{9vZuapqQ;8IgP_mG=H0^^);op zw}J2w20S(ZuJfXEauTvKgIdThj+6$<4I{d+Sr| zHiMvIb=zbO#S+kqkvv4mb3!*)nzO4RpnLm(grVrPH+q|=hVz|ZRyHJE*Q8#-2i9i6 zE^EQ0H3L+JrJo{5s(AwNLQq&Hg?rv{{q1s)DZ|WWkLo4>&)(|FzIH6kC!Io2XyJ?$ zIcH&W&$Q-$(%ftzGs$P-y+VQSIlQ$S4CLMr$2DxW$K2h*`2<1Uu zfW{ZXsQPN$mNzfU5Ml+{R1mN(_!k#e92VHE*2xt}yeQOu!WKRRg1G@f-DCLk4Zu~w zig{utHJ_8i3qF08{UZ|d*L6rHByk7faskIU;r+ok_V(QM5#-|(;OU@JhrUk_#>o$> z4(eoj7`-Gl1i^b3T{GH;TAh}M*eK#%(?q%2i0%f+v%ob8x8wtsJ`amxMyqTQUZqZUBZw(~e=MaCqEQCtM`g0%(=aRX?0TV$_HF~nO)6zGp-sua$c zDX&%U+|<@_MQf{K889UW0`kfvn>?fSrFWP|4ucn?9<)HX0hxKG}fk+A+xAOe$L0T8h~cTc7?wxs3>dyE3f zR0CDoasym?2$XAqintFPcfAwT^}z$Dn=~hILyJgf+TauE3^tC1O~le3lpT+QCqjgVa3y1uwu~0`cW(%L193uk5*+w zbbTY`zI#y(#wnV7zoTOexUgISLtM$+4oGsjO*+17z<&xd8GhdsU9bTSuD#qNMz#wN zA`8M?l@3*LYc&me-AAKMOqxjUIbPQ72AyvQRW-tVzuiLr z@&wP8vpOP82W6`Uw{<>Kd~AVAt3AIc}A|9OR@e#Xu;xVq$irE4(kb;@4y0XZbEdT z@d$Nzu~0TQ&ZJhN1%;Y(Y;2)58AG|T8x_VY!QDdy(%gyJIbIHMj^~7A38LJs$NWg{ zE|FD;zQaMHR+$8g4-l(zp7AtHHV)!cR4H*s20z$^+3q&7U&a?w#B3YVtap{aB>4Eh~ z(J8~5g8fo8adn`_6;RxsR^MY=B2r%75icyEUzR>ReglRv@gzJhQ}Q7BE8RfgA2&KG z2tptB>LRQk^_EQ~%A2qw6t*fO#2-NnD%i1e+%m3@Sg?UC*s{0rm2@cXn(P)nMrVcM zKO&IjB&(>URUh<^L$Bj^Xdz8Q1)jmvDZc?^e!YE|jon|50~WRCA%dDY72+4ldsX+Q z79^L2TnynPOjWZ;Q0BK#v@11{@e~BoUJ3J)RZgAM1~$^)jqL+k`I?GWH<&oO zeYfjjDUqO*cSeK^1^Zbe2*-7N(t)v7ET!nXL)!Ris%EGDw)YGODh5yL7>O%K4ucxI zxV-dqgRW1<2Orh92gLkZb$zYC60*eGF3Wl1CZLj4OQfTpZh33&tRrkinLSz`ea3rj zOBV?oEk~HbyBdUnfJd`dpntla(yBW>nrc(U8pT}MzO@VN=JX);oDA3-5g6=v%+=;O zg#qR_t)z1PRqc7Qrro$}%Ik7RjiC^Yb5Hl-HH3uB0n|T%N)zCM7~rD1yMzugyX^u7 z!HBABY2&%_c!R`AN-$ed5XVlL)slzKCE(c4{=%v!Wulg@#39{)!185JtlQ^ffjmJT zmbuKDGr{dxZx0^cUB`z_pHubWh`EJ9TeAeNX9Y2u*zPFYLrLKGClHOVH48iqM@hCe zIi2wp=ETa)s!l^AZXdUsr*YO}2;!RgDa?IU98rW%#k+d~25@j%A(%E=~qbKO^VaTOd-nT2U* z%^U7v!*C*@A&f81Fx*z#TXc7Jx8tI$3%4$^reqQi1LCATr=|T&&WCMj@LaRm#!9I@ z2)&haB8&RFlVhCaiuXh4A-NN2NBrne`VEC_+v7_2msOQ)bAYB}{`mSNtd?JC?kRxn zIKhKfQOdnTBe>hmh-GS=LK0*{7`#KQ=aqfr3z{i7cr0l^|4Wc)7fd0)v#8E2i#krcr7%f(#eA9vsP=c=@EU{>4YX!$BIPOfzy)_|Zj(ZzE#UU>! zo+aq?K-~s$K{uZl;4j1iBRP3j%QbHN17RWQO# z^+}mGvb@n_n8?m$LSrDzJ0Z~}xNq|3X(;N=RY}R9_agfd+KaH2Bb+kmQ)3$b>5^(S zlm#CI=AkPn)946;Ed) z_o0HBRhdNZVqVG*#f7uu=;{gIo-= zwc3PSG)xU?`!-^_CvFxY-cy^P)d=~e#XyjvQDIjh2PK)*qCTG)mNTVImJA0>kSX#A z$`kkMDvRXLnGrD!#A*jSg#;KrM+gBi4VX?e8e}Fxc2~*`Wj03=J0VlsOB0lBJO$!wM95x?60)M>;3|o7G0y zy%WI`N>|v@S_&oxGLfrt1CWs#3m(a-xr4;pZKMkuG@?E$1kR4&M8PW)Sba?%7?G7B zt}#Of3Uy|Kh1i;9u^jkJOw1ru#6qTD_jZ(=Raynlro{1RJIW%8Z!`w1zUdac?4Clz zabe0xb7~W5>7CaI=hG1YxtG5x(F_1$k=A<3M>hyXiBYxfMHAUoQs~T*Oy%kJYI40Kf8a6lWW^9LuI*y?T*3Mli z*}({*nerghT8?J41Fi?5m>o=WbYVN7X@budvKc$2NVpN#@MFzVi87GZ@Lw7 z!wOh#l?;*RmkK5fy;ELrPzo{n%iMkrqz_e)l2X7a3Hfy$(PI`qZbWN z=Mlii8PZDk>N|J~qy?-Gz?tT=L#+5ovf^_@E4Lxiq{lX=oRtG_0|g?X%p7Di6J!k6 z`^)tvL&9KN9z^kWmq9qDVpPb!m2hzbY&IqkGD5=FmGP0Z5K>HYuCis#MAVQ$Y(khr zH4AQ_22WP9NUIynddLT_)dom>Dm0%-3SO-}IK?jN!>4Q$9ivdnFzq219IoUqJiOn& z_&Lw{PEQMNfS#HbK6>4^J^fm9g+JbXpXa-}x_a9DX+V%B@1{=RKW%Y+R0rgk2>CvNKZyfgKSsS6)C;*H@Y ze0S={uDwfr;GrYhSo@!)MwdQt8`{|6m;YIcx$p}|e6j0DrN6g4gYWvlf%>kLd`nUP z>bp|HNP3)<)h2tf+;<-8htxNu+|w%Vj1T!l`rMO;R~4W8 zrjN}%R22u(#Qm|@!$P$>aCY< zZ{%^4r&6ytkG^@nki)P1NUCzJqZdv~`uZOQ_~WNue)#Thdmi=Bjke#k9zOb`sc%nx z)hq6?A+&#eHr$)TxBgh_QGnbnzwt4w zrq|r#LH6URZR&*&A6FB~`Mr;0ap5&tO)orG#j&YZw?CVrE~F|CpMs=sh93AUwzfjr+(nVrMKMH$HV{nFM(@3yQz7!G z`@Uo4*N?Mbx>5N%uYTZ{Qb8*9&TDm_`|2Gx)qU=ZTk3x0;b%XQ`j3}>?)IvG`d3q} z)Gyyu^{?Eh`sWUB`ef?Ush>No`S8YH0|4vCZmRR#k8i2-^3|XGwbbCkrO(`6=}Uec zV6=aBQ>DLpqtee@{h?n^ed0puGuLW;&;Np2|L9ojgV#p++^>fZA2Of6_|*@ZAG&P< zPM6A8{zmFsFI;%hLH$!8#lGl~!*6~HOW^D+PW1;5ANWn+<>IU3$z}c45oPx9Gk%fw^DDspLqmRd)cHONe<%24)6Vm7e4<*zj33zPd;+>WuL}HcHzyZb-ekJ!!Q2p z)X&^_d-c12JN4j&6h2QXc+qd3fAom){MWye8lAlCo0DqJi81ppzV4~?e|`B=Z~eDq zYVk;N2)r~o1b*knDSJmd%8o+ z;)}2TlYf)?-TPACdD_n_=Jx`b+5Eobe?9!%?Bjn4@_YfW7>>6eKK9wvmDKpf2kIZ6 zyzIg2$h80R>@axSBd3D)_r3U_eESPteg7LzpL_C=1NEwt%U78{OuasJ>4Ud*`tbP0 z3z^hMjvYRpdcXI_XGMPS>RtVCV%}OQm=aA-o>8DJgJ8+-?iwc{v%|kdzW)2hoxS7_i{>qmLe|QS`H`y z_Rl@xQeS)d>{7d_Uw`*!sn#w1`lp|Rs&zVe*Sk9T^VGZW_vvcC@_FpnCm%U(q_6*c zD!O>dxP0qII<3?>R(d(vM1ON*Ft7jms0e)57gFNk?O#acFBspGV7Pzr3#k{d_uNXa zy~|$u+W#x{h37x`{tv(4It$3@PP=Tr=kTf*ojm@ffS=!e`6l4!#YOctcV6oy zHuzhyv!G(1-_6srkIWhV{HCYR>*0ql{6DD&A9L{e^-s^~>+$KwSD*M|>YW!;+ZP`` z^50UA($^Gw^;z(itlxg$8E?tCcK_JrJL=!>ynILf``T&o^Z)nBpI=CQ`tnl=3jU`L zCnqN#I(v8Ze>*w(>T_S>?SlvMxhHSD^y#}Sgul445H9V`2K4%`j|<^d_no}r@QV9R zzUsoKj|;)R@8runQ-Aut6aT){KYHcy?)Xx({=!$@H0#$7)hB-BtdaM>@}7t4eJ?)w zU@AqtG8rel9DesDC*OIAc;%h{6jWbyfqO`afTK@-;UKIbMz(5c{VE%@YDoP5=df)4ACq@K7_ zA>a1s3reZ?-mQ>NKXk&PkmQ(nBTk6Jg$GkFyV`nrvRFQR9H9#zK79WxPFjzD_?7{> z(_#EeuRMQZe(jC%`tV`)@JZu|dzW$Jc3nYAS9=+-4dj4B*R6-K^FNZ(5a&jegM>*&B@3+6v zU%=vfWqtLfPn>+qg-hI{x9!Fs{mPSS>P?TH?Z);)&%Z8OaE~6I{Hl}rZBN0*1vZk0 z4*dfBMFC zbJ<+oqyEaPPhNkg*8kI2pS&jZt@n8F3$H%;<~u#8y#^0Hb&m%>{hE_Mc>JDyGf(BK zpSVXMuX^3d_yTwMZWcnoM_%`3p8cPH-AVaE>enBAMrZ#gAHCz*uYKLwTKJV~%kPd0 z?eawelE&fDH=JZHJaBmH8%|z*__o)dT*2Qz{)UqRUOxYZlh>!N3=ZG$EB9S~^os9X ziJjnzKi99o9zWIHCy%rIT1<21SjY>XKAYofpWOE}*7~0um(YWMa_b^~=-tH|m7ah0 z+{fn%NTHlW^Ge!w{?~=$C%66ZWf#8jruIL6r}p1y@6A7c#hrefbmeq*zv0P~M1DNp z4lij97XJLo*L+Q90-##|4fVtS^yG4C>?+p@{ca{>HIPhoW*$;+bY**rb)*x_B| z$W$&V6-HQS#1`tpWy}c+Uv6Mm+)L2 zKJt%G+HbrjAK&uK3#SWjy;S43PE}uh+B`3Dn;tF?h~2@956vGudF$bOzwYEk|LA%t z*JSX+tN!itKq@6(eDUG)KX~%Q)xY_=lP5Y~ce{`6OS#4F<`p|PIPbZ(ES@@5=kLFE zwFQseLn?pdP<-D><4}$6d-35%-golvFMZ-J`{Yx{eR4|j_wnc4fBkOz#Nky>pXe{U z{i)X-e*5W@hYvsh^a*p}kM8#D>)wCzc~C&f zWY?vfUwh42pZ@Cax5%E+f`K@skv&g$4}%{(ncRKRT*ac<>G}Jgb6o{BZ(lmMEZDob zXRm(fR!w4^8S+$_>Kp)dIh-dYu&w~+SBWQ z!oRFvS?Zy_eEpS3zcNwnE9!FPcs?$=wd+iSiGBs&y72?v!k#VlE6LuxvUV1>e?^P0 zEbIv`#N?~v<+xI}D0q!a!7+{>p!mK1Qwy)t~KgPG9akO!LQMV{gYP!`|IOZufMV|l4n!^@8Ijez19Q_LtvvP z_Hl4ue^$D=yot! z>VcPxr#hJgF6I?oJzIy<8$wt1S*&O4E(`jc#P)28yIX0v*1!JoL}A!}{o}XDKM%e8 zYrovhqd$4Bna`}0%Xip=DCVqvDKrz($G$Q6huoCsn)%EMe)5@a_uE?jnsW!hxoXdU zdQ)d%YS}-7bK)5s6X$9xKC2QRzw60!TkrPye~7c_TuYy2d9S?l;dXO7lOetHY4>ao zyK@aKKC2QRzw7yVb5mheJ8SBfaiE@S-y5&D@7We}KMuQ_88Sn(JgIM}mv`ys{{O=(Dd}hJAL>fNxqrCuD;KF3{a*7-uDzxpAAah27k~fK;nw@$ zi-)J5e^E^R@Z)DDtUtc-=;86hi|@bq@t58H)D8dU&)x0Gw>)t1eW}Co8p6xrp%+|y z(}lyab#Fdb#h0aC{?9MC_{8&5fBg7SvHnut&ObQ3>cNX=-p=dVa||=o1fK{LiUpI}l&iL!V?We(}Pyt?Bp>^X%p+ z33~nO!%JUt@o)cL_Jb?pMeULFlWUyGo_D(^@63vjP8Gi6l%DnP#EajxU)bsM^sKX9 ze@c`^&=k!()qb_usw+8%ZWp`WVwE#n>Uvo%w;1J!*=(`Mg`-Td*DCZ28PLhq zcB6E|(1-I@pYP9eT>&HoTLO7*Q*cYFaYZi^6M4YXO?$k{TgYhL3xYD3tgK#ZVfAw0 zUtG-><8e9@x2u-tsJVDswKkpBV34g^i%xL|%2;dBT932ik+o=7E1Hr?UP0oz>U5_W zdR?bLW2cxqfzKh62fQU6mtv)OJqhz-&grSm0m%Rn*tG3(4Mact#hM0gcNvVTyK!UI z@_mWWSym9-*}^A1BjJeT8VyCEyQ*JWCWyduzbU#MwLjg;{;V?&Kq*CmA}WF1CB zYGxNa03d0z(|j?X`?GpU=?06=!-;bfvs zH)$4BpG>6Bx#1m}wqq zqQXYRP?t#10}8xnkRtG(b^<$7G@q8_buTtl{Gkk&dsZh6%n;w{b~jZp+5w8ir`1xo zxdKYFM|bNyxSZ+@j7tkS%9>_^+0!3|Uw$H zm5Ht|7GPXmrIDgts&X7V3$!MnLM2c$bYDWxIr8KJACsVE^HnZ1M#L&V25&VM9zPW%f=%{PyII*-~cL*fpj@fn-2HTI&;pl1XD4v)CdDN$jSdYd2PIf!|eY7AyB!&azz(L!}`P zu~#|GG|dfIcH0{a6DMp-wQa9%V2LevLwmPv8)(P^U2e{2fba#yB)Lu230M@oL2(b_ zPrw)$Bto3CN1wRxoVZrQ~3&w$?pL3uh6I@oFxiXY82Z5st;c?C)_~Y(%mp zG-Ixj@H%dA@x)qhv+TK4Ilbz%{l#oC=ojWwKpGYD*+{TRu*p^vVn6{ac$ z9g1hC$+{3QOQE~t`ArDots*bU^DIdw%OE$;2*r_^AMbq{Fw8VA9JNsbK{Zzg3*EHk zQ-~2VZ|ZA90yHjI+;oo1*)u0hwz%c z^tv#iTM>^{x(T#)RIs{!pwzz6CNMZn~gchy;x z$lV_zmfbO1QWEl7 zDQ}vyObwq!S|+)Oq8J3Dg&K}Cy5{jfIWfRRDGn#u#SlxxT2=}|ToZGyP+9ZyPGLhS z%XK;qinTlkzJ|8Xv({GW8f-D1^ZR5@cHnAnl8X_}?}1xZ=t+;0^J|GD?G@O)3sJ|W zkvzNt?i?QswXow2l(IA)hxK|o*R(+qHLegNX$AA zG|Ch>xx1V<sjm?C=)7+8Udy<2JQU3;dREf-m=`# zdtft6MxvOnbipFiSr$9qv?xVR;k1F(%CM0w1SoH~9E>{ahFF43Y1LMcAghICC(gpW zo@-Y-lis9a#pB#QSycTxO3UE2+UeR=#cr)me_V*8n*|jz{e=yx)9q=gxDq5UlsP7u97`Uuxso15HPJv zVA9rl^5!_?NslsC(xY=DwB%P6u&$josOM(qjb*#-Z1o=32!`$P99SMBC8rv|v|-gj zW~Gqn+@gm}RL=m|JI`rVW&}FjJqZ-rQgKelo4VaxZA5O^4U6i~@6l7M4B7}I8iZYE z*;$vB_I6IBJNbGv-#WQ@X|iKAWX`55gL-AP?N-3B!?t<=?`to>lB5&{tYLZO&Ug%( zlYtkSLQ!+*tYMeeL3QH~+7q*pFYS3o?yQlA$t^i)mRA%%m)$6(@iG&XwrkJk^Gemg zS=_3)eX;CMX(N`jX`7wdqmeXWjSjnZ%E01VaoeB>z+Z8bH|ZXby>tC|whDNP$@pQf zQ!iSzW<)kd5FHD8rK*;v4YLGX%6$y(kYd&s>MO&J2qrHwGYGhONgU)QRss^XwQZQy zS^&i9s9*;hC>$NDB#ek!T8=d~-I#^5dXpayt1E6dqkIRuc&Ai0l~U>ESk@zA zvA9%&T zB9R!V^~s%X*VxrucDHN337!v1*0Lb4s|jAIqtFso}DnYyA)Ya%Pt+;~!A!|ptEjD?6hT2wiM&SWEY)DxRrxi{uG zzz&IV4v?e`q^EcNybE;klG0yiSczE-%hhJpSy7A?8bl|bBQr=opG$`1@)BuALS&A$ zh)w4k#9Y)Cnmy!r4rr7VJ*+yE*_EBDS1Tz?Fvm9r?W_rcht=9u2%6xztrCJwcfI8z zfQ(zadS$l<^CoD9}<g|!5U&aJLw|CWbSPXJGAsD*dsY;i-WJXoRl3p4Yca>3d)AjOI*@GCAX-VrO)5?{nZb2_)+3KnZYO<3PBxi29__r|9vn1@i7) zcOhvmx%Uji*+zucA2a1?S@FX`hs-y`E$DJNGwZSB3WS0#U0e{b*RYjQSPJ3+>y*Q7 zXsD*zjmV7vqCg|c1f>Zz~gIPRp^yu-3Ip)HEc?*rB5c<6jEI0T|Z?^SHwjEqbJuQ`(VXF^mkR50V)-?{n(9s|lX4*f`-%9a4J> zx`ymF8FEOE%d%3BM7EvnOqs8$+b=JKgGN~A(xFZF5X6OTd8w{ zGR30?O`dJFz?oERCOt3Zs*P$X7jErR5P)H?2zK|hP#t@FvyF|p8P-ZRu%85@Bf0~? z8#~^lxCHWC435!dwm6+atsr3!8+o%X&v3)yu1*?6#FEv9z%i|$28OZF?#TsDDchiP zCLkpqpw|lE z6P?zxR|Vuw>CwJ)LN~%S64x>uL>xSG7A{3SEp=Z9hnqJ^azJ?NRO6^qVTC#A5Ai z$O>y|HXjx@%eA(4c9mtwNI?JU^+K23LKoz=JeVmhpqwt1quv*mYI>XYmUSCM(yLBu zr5+n-ISyooB@73P&K?iUJu&foKug5}N`}1F(YdwA-a0a*kY7@QSv-0~jzgbl%OSYd z)}3*po}br#%e7rE0Cx@s^2U(c-qeK@2uZVQS=H5S(cfcu!8Zo0YL7l`-MOjQc||mR z-LN{X%{W;Oot|5h`-^jG^-)u<#bCucH}6M7L=B9>h2EyK26k0`4wlJ0tZC=e`F?X( zZ`StRYx?^2`LWg6?7NJ_OJL5%zX>!Kc zqxDeYWNpQTu>g|2X~%(ZnaMMth#Jxc=a=lR+SG$`D_<_#o~H>6RSZFe*o(EM!_dOi z%cpAq10@BPjX;M9Y_}Cn5FL~ZG}4l@gLSu(+w)UBj1zl*)u-u+knJ>Olc?2j{IF)`O=T+vCCXVAARp82phc?61S-a)TZubvZoAb@%8vn!O?jnsJSi28oexCY z2EbvPp~lS$gDv2qS;KaBxjM@bJ|StbyDu0-uU6>k4IIR^NbA$=W~hT9dypq0Ze5NC zu(N;^iHLh_b72-1!*04%k9Qfr-FC%MS#J6~}q26AChm@Wuo@6_G8Lu{v7ucHn4EddTC#mit3syE@83rq<9>_Amva|942v>zcy*SV8 zfH1eB;Imk$eK4G1?j=ZDpn-1C>b^NnG+ee`T)=e@Pwh}2Rv}j|`MPAZ=X2H=#geo% znTU|bUcXEbc_o))#eSh8<$a1BcdP8cnWWoe-Op!(yiE`DAO^Jy6yqrTwBhDwt-7)H z>dbJxgk>(Ifd|_J=Sm%1p#}+fa3#-fqFI<{mz^M6HYR1Fq>tLgAj|<7TaYr#i5o?3 zE|&OmZ1TXIi(y-C1CgteuPsN_1*E%q4`gIYgR8q~btDaf0>@5E743Ss$ zM!gT*w3@p)ZvllP_Z;@8bY8LJ&j%ECi zC5^(ys^ml^+=uCTG?z^$1Z7vLtm!Zsg)nZ}{dKhUmfNnG)}1Bj7}L_o1M>&5_Q8fx zT~v)O%p-+<5x~L|?28Jw(Xy-l@tLWQt9Aq=!+8jP-Ekz8r(rhREN?x#HisK-q;Nn% z73}~L|9+U(DvLsmS@4RuZ24vImPNz$6zHXHG+lXxLV7fCWr3Voy*BUns*IgAMLow^ z_`Q2w+vzoL;5KHw*TR zh1qT!?;~7WIw!W)L!K>^D8Iq3R&YUgVDTGR+i9odmxSrcSN*|u%9G=GQqGS2fz|bC zx3{h(u=Ud#P#OJ!dHP*2Y`_;xi%Qg}6?39pjn%~SxN)l@PzAfuI62-&jpBkXkzuyJ zXXHh*YVyQOunE94m))~^zFetkGyO^P+pNPBRz z{FlZ#Fs4mQ_tXQ!c&qYXAb<}afyliumDnE3S>+tJ5cZWu ztu-g9QN2a0P2fT%PUlP&G^L$7f)KnA43t>iq;A9Au#J>RoJ@=!=vQRj9k>%8fdx!y zp0T=%Mt?ynxDdryZ9+>n+n&wz#wvo_cbvDa63}|7nwXa&qLLnDghinG25?Q%ddbi5giuvL2x3bR&S z7PbKKgB5ivD-oM*w*3uTN{e70Z*E2zySkeZ<$?#INCAc`u%k8d(}gg>ugjBAF7jK; zCDVf}#B6EojF#}?PQdYGH=B!40rq9rrActh4J_tx#Y?U3PE6cU4&o z^2V5^dj$i+3F{#|gymQ#;ULL12X1vbpVLHkTV~QKtEuynwu=V9-Jc6>niOQ%;kKg^ zoD)!C>$nO-uh_^3B`K5X+FB$_WCSYdakstoyiY7$;5C({`VwG%m3&2Qv8V%#9q;Jz>PN%k$DaR}6npuWLi?F1KA=>MP;udQS zC8+tVLW4yWFd6Nn)l6f|vDjm^ z9^*(`K_^!FjOvag@WufFbiUpKO_XbAr8MmP>p?a*<%@<+!FNm-pzvlxi_Vcoz3*>J z0hm%q?_gdyWF$>(*1 z;_QSqU1~dZ)7kB;zMWl(BZZ7cMxp7_?jl3j{TZdq9GQeoS5Hq$id367aG{x?(PFnw zvB>L%)qo6pAz2a`mF$z6#wc?_2DM`r{KyVWo0L?b>r{L1SW&M}<=N07rh~ky7!zVw z2wP%}G-oTNFqY~p7l*O}0!KaH=lJ@hCg;_x+BeL3l`(y{GiR7>WQTEX1ScXLH!Ebj z-UT-$xz#}dBI{Z^vsn*koNl|aO}eIMuas?RuFz`Lo9)9{$qsQ%@8AOIlXN$Pe^hFC zwIxN5B35xmb7BjA7q2-Vl72`QgLwtUk9=0_RAR4Np2yv4+Z?Plf)a$BH6GS25Z%s*qJYw`~*nV<7opv zpBc^wP|ryd-{D9v)WOR_)U8>rq*QlVtD@&6J?wXeStgWfQ#Ly+b#!L2tFm2g+}<@D zyXX|{F2>*JX{t;d|GW)v%d$c!*|D);e>$}d)xZ^r zIl8&m>ayT$+FDP!;lMr}fgYk+U?=F-uRFXE*tk#J)rRaBdmI$Po!BL2ba4&m{c2WVhAm!b zN=|HsG}|X|kq^MSM_Dy-Yvd4Gf|!~#th54K4B%9*SKMKFvR>whAqf`hK^j(am*lJF zF5QQk6}ZEhv&&N`6j5U{I~(l1rY*L}{Wh z6{q?B!gASG*p;?Dm(}Op+0M|pnUSH&dd^w7i0flOIw6{2l;foeXlwOSPqOpiCeJd8 zzZ*1#d9Yg#rP%@m1k`dZO!l}P{jN=}!EQY3OYGDZoIX1ZWJYHzkYdNUwOW3$=XXs8 zWP|NmHJVPynW1K<&FwtR7_N{UqmuPZl=oxC#=*%06-INsFlo5W%8y*}@)y^o139Kv- zq_?RJ5OapH=~hKVv%6A0)sPs1^iYzV%DgGcqdCZ^ zSe`GZCBA9!JuvT)RE>^z&3?4Y6j-q+@?tr&UPMDmi`+6uiy`?2Q120n+?;qIjvh7r z8AFhgC+bRtllL-w$*b))sJJ{+MPXulF=>|jkoLm*(8vRgy~{Wh>8iK}x3%o95zT_Z z4sqc#>@@=%?M_dENePri4H&>GA)*~wy|-=we!bprs<}-BJ7L^yZ(+reLH@_`M}4Ex zlS>Ww9QM^rhTd+=HNA`&PidmJdGM}pSv{Ab=^QxVcR^GL#i*(*yM#FL!`3h+x-_CW z>P42`#w#aONp4bz#O8X&ST$~9=XJiDU&57<4neSzCiPsYzN3`&ZUVQqP%5ybVwo-i ziU)2{-VaPNO|~r}vvejf77R&g+GkpZr0~(q&(AhsDjBRd4N{uTs)&oHW*sG0Y_rWA zsW?K~SLjBiO@TXE&k?Ylk$!zUSLVK3V2j3(wWahJEEbtH#a6}?Tha-tO>D+h#I!96 zJzMKi`&D}hLV2qV7}*68vwg3m2?K2eQ_MyUNkh+qzI(Nsa_Vr|RYwsd z?R8az#WLLh65Yo+L zC)uUl!YWNWUIDDGQ#!&3q&MIi!G%>)$em|IPPG^r1hbjVeUr_F=^X;-3X7VKi1XO# z2m7^{tD9o8D@854NpBGzNgG5cUIb6&SHEXE{-y!P%=iqm>l!$*#&nmm6kEd|M*3hK_A8iq^cIt|5HKCikrp zqfFTz!b-W;fRVr?OM~5ewW;VNE|)oi6Ui=xs02%Fv|V*MsSl%(IvUk_fi{@5yK9lK z6T_1t>}-{QWntk#uQ zVvVABaU9ecZo@{)W_Z>VvcJgZ9w3&HJZ=HJt8Eexw)5Tx-Crg zO)x;(%Z`~5R@kLTeo^z*yE0o*7_e1k$CLG@=(#~ovR7Pv)73Y%NeE#ImVqgxBP9(E zkZP6LQB`^cJ4;z1av9g0BUGod5RLh|1feLlLA1%G?euU&G$7u=xY*LVAd;28#4V0uE|<2w)x=#j;Cb{@vb!l&{V9A+UW@{h>}-uuHF=q z*&fRoy6v8s|Zyc8v90!;y6iLX|s{d9`=y z8$736Bx=NM&u8m0#Pw6>)wwWFc(zX6dix|O^qN{USwINyo`qvVIz0-f-f0YHt=d-B zL}a$j?QAQ{a-kq2|Dy#L%+RorHY7rVg6Af2O7VRpA*qw8zh2jiF$+o|G0bmyiQt;} z9Cjsy_UxMW4xXJo86}RWImk5Nhib~JJThN)P18E{2Q53RX-AmSBpxOe0yaRPGzDdY z;iZ-`6A+m__Tf&Thn;m{HJLW6RY@j|GMi~v(qj;DR4`|>x^4D{!!(_wBxP3lE!Qwn z`KpUn*n7lm;~%;*XKT6HrmOj0;*7Mb)oGov5EBhDiLoacLOfxbUDbw-aLEnJJNSDL zvmRq3mE7I}c>`QjiL|McrjZJh72yxs3iSa^NT7uv6klG!TZKN1@cVFJ=1GF6G`I=k zpWx;3!}JD5PQKo*ohBXxjE0RJtn7|vgoxwqRN+4qC#`Om$%C3k4}4(@bEuFOFj(oj zXv@1m%Yzp@lg%lx9Uz6n<3w*}X8MaH@|jX4n0@n&orN^2T3AL_5=k=aYsV zTM^kTQYek{2!+?TV}4i@D~4*edWZ*H_X1c)(rS{Rk}Nn& zMxF^o`FYD5T5Gx6;<32;;F>Q=g5JfBNCKEAt0hvZ_TV&e8NowJ6b-o`SIox^K_0ra zs2j=(0a#4l$yE!n9%Iaq5i{tw@%gR|$Cw#q$_yzs^dSNro0u{OD}FxjB65G$pLtsk zNlIdjydbzE>+@}TMS)E%tcATzo2bWF)(%wreP`L%x`^4dr$Yw`3|I=CMVaG`4&fNF z?g}Kaa@eUX)RAQpbXIMvLCZsnQ}UGxLiae_j7aNI%J9BQ{mBL_WDTcK%5RF;E(kJq z^*WOLNIQ-RvYPjV6@Ev!n146uBDZ8D@5`Cl(RDh@Vs&autAL0-Z7m%W53S933R{58 z#;h`6w3R-G>*2=R(6`#ZM*lhRFY zevE&yTaf#A`0=dejuBUIOtZXpDk2(sFq)h*E}-j4n8rMs9Y-OFp1;+8bjzuF>{Q2N z3pE;=eszB?sAQ7p|E2T#|1nIyI7IurbCo7`o;Bi>Z$8)Qk-vsY5k;Kb&pWKv65@^H zHwQc~?C7GugXD{_@LCVf?8*`W%t*~5X3~C{b59-2Q9&4SwILbQHycwj7{^0eTkOJ$ z7wq($v7F|nw#F0NQppJru)G;LWS@?UH5LjsNn{)gIXAOn5)8B&@`X{)6C^jkRvbyl zj#)^m8f|T$u*9$$^y7$G(0O*`hx^Ky&s5zo2i|m;4eDz_)X6^RD3PUB^QEX+X2bzg zHb$9gUdT9eHsiB{+PGS&7S`~M>W0rKPDm<@!WXCg?VKwU+g3RmNr~DOMaYrrFKtfU z%obTk$k$c)c?ocnT06E@$cjzGpG^v|q?h^@U!dv>zD|?UfEhJrW~9q*A8Sr?@`yjo zAlsooT9Q0<+I!U@^)#79u3Auc?0Qkm3YwdBr<2Nbva zN;Rm1IJ_wF_89!6(1Y8E>7xr)WSEFUg(0<(_hD<%r?4!m^Cme^Wd{4LZf(~M-umR`DK1gTk z=`<|qd7?1#ias|9ax@zbqwD58yUUi|ukiz)T-DVyOjgA`9Oq}c z{bCkm5DqXS^Ky=2s8TkF`GOdaZH}kse90k4WN|bZ))J{Oj=Y-Jtcyl*u2rig!bU2n zm6i(;J5e~K;R$hXu%JpvNQT-z*rW;MH`qgcWrPgBl6DoA70I15+qWyM32w)2gwq^e zs$l4_EWbc3je$cLb|oa3@_eld+eT@qXEuwjM~vY;*+kRgsxY2--3_lS+>TRX=(QYn zW=>VEZcF85I6;_fJ|n12c2Vl{d`=|c%pNL3OR5`7hA=k@A|&;FXe|2ua8S=wKt0w) zVgn*Z)0GxgX4*JyL8n@(taIa1r-cWJLM0-ySwC2XTWL-=YqHYr1Ur&U-1DrFMOrMe zP{*waGawa`K`6Z?lG$#5(o?x+zCZ1P-Y+Uzypd%)lY$M`UUSbj23cM=xv4YJV42i| z1(Z*?UvoJK8VJ`#ax%Py5V4G8K_MJ_W|k2pF*#m{{6r5n>&}oY!yH5ILw$WPSp)U*ySb*d%N8}`vE=OiJzZw}cx4E71wF+Ge7E)~` z30kqe_3N!dHCJB~v1K-wOL$1vy~PBA3J1q$c1P2Li2 z)h{7Kx`)(veKb}=frU>f3B%>!!qKGi-auwz34yQj8YiBP+yJf`3~TP#-jU;<-)S^6g%8qn{g$oXo!wR$-47z$dR~P!r?YJlm zp$po1%Yk~oT-%bl-rjFZy{XglHMk423u`&gL3*Nn`S95V-XPt5a@HYRD^2>&?L1 zO(&}ye+*a~HU>&bQ=k$eH*}xot3`3X6cYncvxaSFZnof>MKF<<6B9X3f=!18V4DPH zM4?kCPm#tNBHG+b%bFwcg}zI4Mey&5h#em25w93C$!{~Wl^9d1SJV(Wk|#p+!e^%9z9G$U!$M>8X797nMO!3mJ!!rV)Z3CRZ`bdrSxhY}L1 zLyI9Hu?Y|$F~u=A5Q9Vbej}ZtQ}&SiJl}WsdG_cW&HUegZ}0nif3Krkv`6hZM=NjF z7~F7^?r?mX^he{lIOho>X1AdG$!N_saC(8lU_*g?L8E346PjvHExE|HY{l?|gC9*D zs%vvO=cI91r#rEZrUQdg?5Wd|)&pL`q0Zr|X8K~cixj*MgF#76WhWkQHw>TOS_vqI zVGa=9NM0LsSWOltMsO(bMfWgc z&)XA}BQ+JGzEB}L@b=g|Fn7s&}-&D}~kJ#}{zP6VM$CZ2Vr zyWWWnLErfZBAU)xe3XW#kLq-`>Q#!(yWpVSoAY$A)|87{v5~d~-4bcj(CoPW+cX@X zn{sBz(ykWiV?#gxd}R~R{*KyM5m)%YG|gfK9|%K1ttl5eB^21 z^vqOrO2k#P958hN@j|{s@rh6$5tQ3lx|;C75Lvux0VB3cFS#ZzC2DB`!Ed7c3CFRW|-V} z@HG8K4?J$q7RT*v^WEakH;luLdGjs+BrWv6fey1Utu5_Fk^*fJ6eFW0RLN^@~r?sSs9P$IOz z2m{04(f4-CL0wy(3xRH7>C3yV-$`UWZQTzuFYwyi=;(z)!WnWUFz;Pe^Yqd;p*DRW zA_Wn~BroNw1-PC^#SX#%YEV})g2QIBw@s!5pQ*I)W&mWv_hoNpTD3KL`fyTf%!6dE znhN-QTyA8q4jj;6Ot3?x3S~F9aJvNw4vs;Nnh^9gi?eDf76R?YEC)+n#8&}VL`Bfe zn{f)diiK$+Ux2Cx=r`biW3EB{)qp*KwTy9We zn6SrSwit%Hy`5WVxf0oSTP!Rfz^T6GaI+i^vpwLYNo-)9+q4q)^=gZalJF1W7y0T2;XuX`4i>@HgpiXN`5S4qKyVebtsn-h%A~^Q53Zd?F$?W(-;ENsiR*U`J*_t@d6+` z?qnRauFDP%CxfnZg9yciYth%S1Ko&lbnPj}?(*)T!Z_ z2M9=Smn2D>znpAMVo=Xy8u?MFtl4vH+uogt(mWb4Jsfg;qA*LQObrzSrP83IX|Q7i z!U@%E?uawkgAz{Xz?cw=gPIgr6q144aIpexpRLByPERO4^Otc{_;29?fnclDFjX6a zxu*uEh*VUXR1+=`4h-AlT-($Z)kThLPIC@ zOVMGS_U1Ug@1>cfJJ_k?|8_c+24)tpp_?ifwjAsSNdwiNB(#NncUN z$r$MfL`f+)$d=lUXv78gdRAtp4PTdWXUByyZ)-6m*(|CqpQ=efXV%NTX)fC0jiwk;!CEEmurrd(eSfRNfwRi&S2M?+;0Pt6=IHxyOF5R%zLpd4I?ixd9$( znl#FgEjwGCaDI@@O!%N1O`@u!q!aOqRV)MPWT!92Sp;NmJGqW`ieS!etdpWxX(hNM zm!nx{%-aB)a>*CyHwJ;qu$j{$e)t!pS$~?TQFJ#8zKPgupaktzRmw(%IpdCkFh&Vx zz2UGjgYRWW>##X;LQP=Do9Kha1QDRyoWnZ}^wMHCXrAXYrZWQ>$pzBc4)A=*K|zn> zdMm?VUI@CYS|c*43B%bCFunSqIm30pp%tg4YPCd8{Vm$fm^390QdKq2+zaZYNo3q` z^xZli@c?82z=3MBj|Om*X@twks%KonZj%y34={fM>(0TDoT&DLSxtdTrrTx$W49La z6cUx9H<70@NiL+R6db@C;P;@W{J7cGv}TcyfRU?W^CmmhE(`rYs2=7j?A)(6IGprE zn57E3tJufRnQ1j_m2#~!Vag7tyXlVUgM=_F1e64m%d?GIDK!{*Cy8NM&dJhrKuZ+M zvS2wH;$VksJ1(n=s;NOZOnWL)#O@*o1-RZv2LUBG3AvO;ZA|K&Mo%OAjBTE#i>^qB zCoBCXRZDVxs!S)vL0JOz0RX_jrD~F`H3TA0DTEtsZOA8Lbik#h0c0rYo{(D(1Z%yHJzF>5vSXmNOnba-wg-;Ha-X*I^ zP=qSNWB2z#w!*WKx`RqpvW-e205p0zN;UFjPslT&%IUsyCI%b2F4#jpDW0?moe&C6 zDP*3^dZU0*lxYucyOF3*Yy>CCoNSAA?5=2i!k0z2nuK{)47f_U$UF+Rq5Oz%6uNw) z?cz=IuA?jS0Xe0KLuWdrGa^r^q4YSCb%ni2n<8eEKrbxLM*U%;+?`5pXil)tCL&RR z1uM$5gPkD&2;ww98%NX5WJaqtxUmx!(XcDpod~Mt_40W#Q>5!jIXVX=HMEC)HIj3u zv{H`=bn58;s?wIJMzzt)lWtGfJ9Q-N8qBDZT)#80HNx7Aw@I!llXhz}EpKDX)3BI# z6?2`6;sF$w&nQ_T3L1j2n$U-OqmxRG+rayS9stw~;e6BA(!~ z82T)~Ogezn^xJvaAJ6a^z*dIteh`E;eI+{#k7pu{`903$-~>_z1Dbmxq<^c=m>tW27IIk!s-5e4{GtibA7v=H_z+cz3=*V z-2qHMBG__pT?UfzVA2}C2MgyGL(anTI$t=39)#^X_$pT9<@8=>!&1MAw%?2OJjfi& z_~Vf>=;5}ma{k(UNxu^fB#Ho%V$5{BjlD((P^k^+bgG)b5^P;gN3qm2Xy!;CavF}oeU2&h`f^nKosk_WA`EEQ1+w;zKZocW^ zdSJivO=cCXy6HSR=Bo8ooNtp$m2IwI6q+?H(JB_%L1S7fI(l-Wk>z+llj&z0#nD8t zHIzZcpLV$zmgeJ)Oeaj$$$5aS)Z!%%d>VPIq9s<2HnIZFfBcwq)%x{dqlk63kd%gS zgPexiNnczLebubc8J5&=xWiFCp{h1R&dLNNm&K&5PNh9#SCmfnG(}1Qq&}Rqf?TIN z@kXLyx~{e(R3`wo0wLClR^|5G&C`BmR0?}3tsx4{{#a}}nxJ3H2Vl7gxrGt*D)w-x zCiSyi+&h**RTb`K!Gcm!Jn2lZ8Yg{qXtp>$)oQ~zu9(Q|SW{wMhqp$J^dP89N|mbE z)+b^Js(qL0%9TA+QtW2uB3MLNXMY?nhT(hCglm*n2sw&E6Eh`0PPTNr4fK6+yLXbw z0=JPbdj0U+o-j60!%*U|O9%VGg5ObQ19}7go{XkUJYc7^%_u-cD!FiyV`mM{?aB!Q zcn+qiR?`jEBv2Ls1u2Q0`Iu*$-k#R9wbWqR9ZctPlGhJ5jlfVBLeuKFJ(>!Nn##CH zSE0!$egOOvIU0_o6@FrBy!oj&CHQ9!K<0ui7o5q*!xA4ZQIY5{X-n~CKfLbJiBh^+ zb4F)kEk0HnzIvgdhx>Ud9A@TB)k}ASPM5FQYNsL-x73=5Yzykt1P}$`Znq_AURLwK z{NxD+8ke0*c8lI}RUCOnVv*c^uwJp<_UMvI47*=uhNN0NqDbg&Oeg$@hE`w-b zT*wZ=zm?4B&3@J~wACi?Uy6(~67BS9QpxKLf>R<3a=ZzMHJ#~b+Y2;O&IF;%1|yr8 zDm3_1D(#C+_zVzroZC?{p%-J1&EcSdm~22R63%e6U1$hdwOJUuu$$vprNq0TT2-ll zjA>F%&Os%ClTXh*N!GbIhs%jeDuUobRGW#Kdzch)WQ$bUo_C~GW7Q~bzc!Z7Wt}N{ z1kd3BSZJu|ctV1drs+u9+$p!eM@4(Ffle}xNzhAWlcABCR)v&b&47d^>+y0U*DTeG zCQwLcUYx@w;>B@Elw~el%yDuvRlr#-D2;`F%bDylfKtoVatX9J{;9^rJzc@)sx)9N z^|c&qxk59Ol%qZ2#Nf8QCsl>1!htIZuTNz=9|bMb1eC@A98OBmbWVz(Azi>VTMpD8 zo;EU!hyFmXtaPSBJ?D=ASd#951T4=%nIT9xW{;(u)6FM+>i9sIriW|k0 zXStI08kyH!DblF(Q=&>UAvGj~QGJG*L2hC)us)xCBtN`*>U1*i3P zt5Qy~iW(oxL@nc%Ld|}Rj!nUfY^!CO6IUOW89SXVi;b@-YC!9GiWRRX5hHby^Nc!nXDU<;raiNfS8|K?*_5|IZD)(IGr<<9 z)+nkwnz?(dK;~PZh$d@xFy6%|eG;N{rUWo=w8{mdX;Q4s9bQl0p^wugUtJb$9!EOh zQQc)P>#t-~X&UpUD7iCqJHoI5X|lVo$HYt+AWR?Qa%R9#>l59HR@v2`#8h@-XXz|k z9u38GNSfFa@pO_d+Ugyq?jXTJ0FG4CjgxX~l$T>m6Q>B*%Xy|>fkq5UrbaT?ionx#jY>r4jDB6IU2!qhQ%dr6$fgm2Hf$nvT4|p{y4RaE7F+jku>p za&E?%7o0V^Ibj@yZlcv>!rc)aNF?S}}n!)2&egCim+eXDe(njvLz-~|(aiT%&YLzhy?JG4wQ z!Go(cskoi2(y!KCK3iE+7+XdiOLY$|l;lZ8lLoP1+tCeC1Nc9HGF4AibhQbSqAk+! zIwo$Vmxl*pGCgr7>Vq5@t|D<;xzdmOgl=*ej{%L&Q=qYndY}$1m!a>gwWotgEgA03 zS|(Mudm7X96v>f|Nkro*8it@uxJa{8(jjLB8g@IKfjUQIJdAHXjts%u;|YpkI3;s{ zxs3+U7QkG>h7yHfXQ;%dy=pb><+}B-D9s#n#jSBsJ9zy0iM+YNuE6#{3R z5FPTJl)EhZqF^o<6blsTt^n56syD=DVlp0u2Q^IRP9iu>0CVc`N`j-nB)tCdD51%M zA~3y3RhNUEQAMr{M{QRQhN^Zio$gmW(di@?PKVrSyH4{wt2M_%Zj!anY5$mq=c`bV z1{HNU&9NC*#twQRq2iw8l2s&!^(i0%092YQfrJ6;YXbNu`Q|uf7066ml1QmAoOq$nnYKJ? zA#P^c0a&Kl8P;W$%Iq|JvW?l3b}dj)Vr0|BS_%GmMzU64zMqCCl^ zoV*P@B{^VfYN3HW1mk_n2Y9BP6yXg`_pxvDPS>a{CBguv2Eh&+2|%4ylN>VaY+M68 z`haUmRSvd;sh<@Ef%L*Aq1jxaDBG9j_5#_pLz@i#Pr&z-hzuM7z1U&~5qO!56b)=n z0L8%0AcvciY#iuOBFl|Vc`AAZ7qjV0)!~yt{yR%Yj;d-YRO|Fh|KHNXN-Lgr1#!XI;`dz=5hXxX28N?; z6YFl~+yXck_BgoC@r?sa6dg;l}J zMXAzn6>Ehhqtyyg@q%D#4+n5R27~urBetBGuljN>6%bpd@jaIDI@RgpHev#YSB=Op z5@Xo~L)L82{{>2mrBf=`a|TNFEydM*n;eRVx_O@ke3zXUrgD?b_$)PXnY2w?M(j=h zP_>x%48x!S8k)0e!3nmT8!9u;dkP_TA32yukEIi!+9g|5VK`=+tyxHmP6zf@l&NIv zWhEec1+Xmy-GemEHf53q2cRpRZ3(GKv~ONkZE?R3mnXkhv^6C=T_{r&s3ay-!(Gj^ zmC7W)gkLtH|8H*FZGzqE*w!Oj}j8<7S`MGA3mAzhD7My`sRl`_31U7?!{ z#dMP4KtJfVH~bTF=C4KdSZXLwN^(++Ci_9?RRb+5E`uOmq4*G^2~h=JXJoJxvwM6s zXD!HtVsju-E1c>~r|Tn41D9yZ6%~sxDTWiV8lPsnv`^uBE=?l6MkgmpT`B?gLen2033YwEFERa5B?Vs~sZ*4) z14SvPgM;}b;`E6%GCKuf?aV1f;9?mKM{=d18loIvpeiR+&Ku5%VLz3ws`;ru?H{$O z9`D$t(l+;mcX+zho)8JgqWv@}#WS(2OH<3jlq@ELAjtLPaG~gX2K{KrmTmK)8O?!Y zCqB!Je43Ui_rx@F3Vs!a$zCVP&m>xL2UXS`!Ewk%I!YrZ zn682Rc@ksNjhQbmiSa?-)o^muH0twD9kxM)Z;Wkm9}57KuN&0+WDFM`4YGx{ra3>DL8Bb7hQ*o``)YM5&Wjl#T zkcW$|Cs3kFI464?vIM6^8GM7YOf%-^paRn;gN)xrCi+cn9AW*TN=MB^Co>Qp*zAr8 z9}T$#yWnJ7ImcWB6rtio?1@Pas5ji}{O-n}A=qD?OLxS{L<&?rzWli2>q!0P zSSZ?kJ}_8D#&FGL^01ub8hMYmM5m&YA{$MKg=mvkMjkfn1ow>B1+ zW>C-?Nr8rmzDv$)&UjzuAr>wue45MFTXcRjV_W{bUU#|&3Y{F4x?Y#fNixhR8f^uH zd^J_+jyj721Yqnl8}D>`!XpsHCBykDHI5hHmj%PwV62OZKkon(~TU7|1$Y&C;WJEGoG~yAx&)Oinc&cc%o2x6;cy5V^v~_DuO@vnkS?DoMc*bB%bF!PM4MBXW z0s6@?Ssa1}b9j+CMt+KCdu~kn^xp8GJ4w*Rb4@z+28I#T)h_CT)5jnx*L>gC9B*r@?mQt=CI1$d}hWE^JV ziIiK)(4KjVVw*`N*l7!?Y0Hm0!W6fwNjegp@RGZ!ioP*$9Q}?X!Jr3#f=FgdKgD27mJtUbwnbe z3Vf?PQ_Ju_O(K>I73+0ewxSN2{ZVvWsE0xQ9(Q=Vlq$P3U9XpA8x$M!!O5`~jM%9` zp=xuvY;~pr;=6Xgtp5Osud&H9Vg3 z1-;tNk7;|FZ}IbrTnkp1c2UVX;V4dlnj9p7RWaC{G=NC;+5r?Qnoy%qhy$j#r{ET2 zqbpJ11BLEj+UC;b3Cw*8$IIPfp$i9sab(I=!9OsJ6lA?LjwvG5*HmxM4=;sTa~7F( zoK8PT2?M0dHK@!bchTMsyDGx00gZ>!;~0+01WFm9mvAOI;B}KjN(hx)oFC3cfo`s$ z3en~erURW!F%4g^&&lDj$3Ak_aB_6JrxTuJLYhC631=FbuT$m2F1TLSsi5xg>TOpg z1sb~Q2pFvAANl4n$M#$??FBV`Tl?Mc2?w$~FLHj*8J z(|I&tr{!X4=#bqKyn{%u5Ksk|TL@wM6?8;;=Zh|HyxtsXOxx9Nmh~!B?W38TlYDJa zlW`?0W-9!6n4WPSWzOVXiK46011>xYdw&~|3W|uMRZ1ks4P{gx2!aYf+-Uyax6kYG^1?L)00RK z-R*!ZN}D3|LlU>EKK#GrT(l+Iig9H$s8@NaSJxTUWH+rbiVMs&YM)~O2ID<~->HEG zsSZdF$$OI^BGqMI8O-U-7*|pzGXx0`->Br2iGNayi$O;u!n%d_RJ2pLgSHf10t|GD z#o3|4PXG}ETq?t5=QwJ~gPzi*oDN6AF)De2&2Y;v#c8^ZtvN&u^KvKV5KD>{AE)XO zdm_>qm2CN<>i442-Fh6pM_B~pNmNaQ+rTEUaS482_H-swE#anQb3}`_L6-#+v{Zmj z2%$BSwlrN6rgo4_IOaK75V2L0Ei{Y2K#nIVn}e;1=>pI;*nnWWXIzpt5b_S*-U`>F zh|vPh4=oa`y1X37;}WzyZTLH0CQ%dYbshjX$*tREXl|fU5i`+hs_i0^a+NP+3tWTY z>1byFcV(BaJJDOk$($6rcG6zWHr?=c_J~>w)EtvWj&`y$xr&Hvvyhu@K$X}Y^Lagv z5CM)JT^fajfone zvi&}5N9YyZov4y-+Mk`xp$Yah)nv;~jpU{?NL7*yPcfcG&5MHvVxf>!VFqO6aORXt zr8J18ahoe;JtZknuX@xW*e*yhr&heKfogXK!C99{aW#c0Nva*L=Fy?cSEa;pJRwI% zluwybA(E}QGi;-wMU!?_n|Bd-&6QCj#ggcmv12mN4#H_sL(o8B1c+JPDJz~ZnN!43 zZ3d#xOeqTU8tFvj8QMNnsaiRi=f+8?T`Pwo?5yhWX6uxSapKU*9j27*qaktA2OU1y zr$UmSE)AwyNlK0yZbXmcZF8_-MpOqKqZ$-QP~g_g%%BPEgVT)lx}dC{qV_IlWMb0x z^;oqq0fR#|>uyITO^_T#?eZ)K+k=#|xrV|7*J?Rcp-S%248HAZwbd182$QFWK@I6lz*PdYk?l&>d>KA0hL6`mf2$Y?UFi7<+h&C$5wr^Sra z^=F3Nm_N??W5u>sSNK$1hS}MJqe37Xh`URT)TANvZqQR8VwATNo@yylfZ@5|Xad{B zGIbvm339)iDm3kbaTsS@-#mn1WGy~D^^^Tv+SATAhn()R_x&C+5Ub;iH;oq)Y0|G1 z>wX~(EHAuRREB&7>w+=^t1E#v}h zmdX@B2HE5q+(;a`oq~Hy=SA8(55dU$_b`BbT$z>mwRqovy7kuGIOu=mAHj)33OC?zEor_7MLu(*VaXJ#eu& zd8e8&PoHAMuF9)JlA-Dd(-+s6xwlwxOo46zma0E<;zEjKqQ#c;N1*V ztqez^j*}$Id9X5)0_OX;lpytLW;StV#G0QB=(01@@mFJUJ#RzQPS4b@iyRi9Xw)~+ z3cg7%N%Cs5+ATB@ZKCeatI?dV7@hlk82LaYa&e{~Dp%zo87`N_SsE)MjtPK!+Nb_f zK&eZCP+IG!oorPUU3FTGC2i3LRY&9*rBVZ{tsE3GQ%__{>-`q6Vva&F*p#|8UnbR? zOv3Ik0-3cPNkwebhd%q1gxkm12b;>=FOBWxq~P>PEt|+R^`xyi^%dN%WT4q|c^mOM zqGZ}A-PRsX5w^6F(MCXA0TPEJ4Z{o2vQI)jjh@vC0GQgr*q@osDz;gmkSkEpGCSpp zF$wN8Hj=E5^IRC-CF6{DOo4J7zTC;IXP$2-5tVf8N;Vvfiyk8ig#*FS@;E`w{B*H8 z1A#Pbft+_7&|(D^5tx!7FsaeOQLS>=NUh6>N4J*GbWHIXeV4&`sBT{+2fr&^WE))Y@XuBTd_G zcg-Hi_8q~hQdb9+wif7#;Jv9RQ@VC4iJD~A$*LYyo3qBUU#kLPg-be}%V0uU^-?n1A_*63;? zsQdrYNuM`uzZIZ!J0 zbX4oNlM0xfvvHqPkV`I+f=Wz-njoEFV?o@{)Oj3!l76ty$U*|3qa@?<+Q-;qS`EaK zj8AM%PbZXxZ$?}U_oJ0emYfMGw+6dTMI31}@PJSmVdezT0ZzUV!L7m7(*_kM?`uur zmIAeYBjD>+NtK^P!92>BVU&dzH;E8)VrteaP3FU#OCMBBFSZWJ30=_chU)?l2MBW@ zOagg(a_ETwqLGYsBws>S0>ebJR4Rj9U7=(>J2GSWjTl?a#C5{{lOL`Go+z*S7VWrQ zEj8vjo(@C<@ITn&Y*s79xJ;f^A|q)pw(~#@xt0zLW_$Je1B71bg;tmjw^$FDy*7Y( zzBm{;!D_kR>y1Cul0ijSsLi*@YL*#*2Ylh(wkL98DHYq7F+w?HK>IM?M-s7!<*--q z*tsHLSc}8bok{veCg2odhUc@23VIUkAdB$J2RETLdL;cp3N~ z!vHx284fp+P4EfC{`gz}J&c={zKntc19*B;%}BxkxbG3z|0G zgkQG1T#Jy+CaOilt$H|iWCM*l0cp*kbG?jwC&@^uw+Y>!6 zJbP)xX)er#EUrQYTgM((;oFO2c)u_u;SbKnXJfg|QWRqnkHPdYtfAD{Y?Ubk$Fxq@ z45t9H8&5BcSl$R<6h{98xuL`T&jWv^SrwWpN^|A8P9F~ZS`$`}c8K>lnHG%P;JPUm zZRqNG9FDj$VswwvFt2h&x$EF$KrQ@D55f;3Tt{)_DmR8HuF@5=#RMDxLjg02l`0M$ z1q4Gj5spH(=SU<$$wb21)Ugdx1!%|8?1%z0IBX^`&5AQKUU;_`NZQ&SIpI`WStmaO z&T<0%OKFs=W8X(c`EnP2wv`G`_PW@3Xty#$@#Q)ra6Qmc`293ka0_zFl=`F^#T?=s;DBybOIq{!}HkC zm#Yszw>z15C_YoK)#zlYR1!c4;dFPSIo&p5;10*6%m}_}g@xB50yf6uYN0zyMGBdS z31H$gC>*8~V*Z9~plpp4;O*uw%B3EN1vs$>s}ai}_#Hjd(T$~a(qEHEtuUmca}12? zL||(JOoMj7jBMNO=XE9C1MG1G#7TKHt;5C6&WRNn*wI{-0fRxe2Q$21fx%i}K|)3L zO_)Z!;oz0{fZ=EGNMeDE^usAK6GX&ps;9#2oxsU{$cv$NF3p(XR~Au_=Qcz%Rx23c zRNvlJw4~7O4WXogvN2^&5x_7rV7ue&;n^5bn`#glZ`w>60XsZbW8IP+ zuupSA0{KER+z(7*p2Z**=4nB`*lGIOe(8SEX1z=pTdh9&aK%x*PWV ziUMB?1fgfeN-@-%mPkaLr-wRRG1U~XtZr#AO9f-N&Z+aJGFF0o2rGrHoV1Uth;UCA zR65=RAVn=VlCW`h5K0HPAFtB2VMf$&3^q-2AkBa%aD#Efs-wt&TdG|xm_}w-35wo@ zuXh)rqE>x~SK6-ZxO8=(8*y(&eh=RobM(R zl}djUk5;&9lSkngu=dY#SOD{WIJ|e4p|@E~cr6vBd*xxcz$joy8kY6$Y*_YCY%)p4 z1;iri2j}J?0}l|XIttW#Ia#48Z?_C|$FOAM7eX@-+UX)5y(>{iJjfT1&y4!1UD4GgAipf--tzUa)L50iv@_wDv|EI z5R?$T!RN=2DONNM7Aq#MAVjj#YJN@ym*0?n9Y610zHhKU zwYZ*jp4z%m;_+?kRQwz2v8T0eYLTtWr6;a3KKh$SF16l!zF~jvITsy!vH8a3(pRoB zzP@qJRo2&@Z?M01-u1_BU4DJbWyaMZ%Vox^|79teV)`Ldj!L?t$@Q+czAMGR=Y#3jEaP67LF0d>l<4?<_=h#lHyaAz>lh%vQr`6hq+`Z=KCu?FyvEGs^u^NABY1u){^U*$uil*l{xowSSKfH> zrc6ltqqRjd-*MlKEuwX#b?%GM;_tfm!ub`(hd#3QKFj;fT_pbWxVn)*G86yP02T3_)`dZYDfBKH;m_he64iR z{#2)oM?SuGnep+Dqf!3#Y-w-(tF?`Tavc5@it@9wrMm2|*RHc7XS11%;a{(vXZ_T! zO0JY|wUWk1|9b7w^^L0!?w?b}6OXLjd~&mUezsG^jjIjcch_z@vGUckuU>7_l?SWY zEjr~(pIm$VXw=N|aw%-)aN4jLR>FAdQz&&{mzbaLWVugF>(gs@9(~DD%!_u!SS*J9 zAJfHGTODk9!}eg`E9J~@e`f6=i$&Q$3psg%N%Iu?nr;j)y6-4e(f$RJ{R+E z<#M&E9!{g$WDzVr7xOV%+v_%jFiv%wu~}JY%P%r>`;o^s8$h zSl=(j_|<<{yWJ8wB;o3Edit7s@4WOHE5zOPk86#6;xH(l`p32BTR2nP>E)o6m_}V|O-Dr8^ zAr=4b*VevhR&}v%J^oKSJ#xX<*M8r!er8{f7!Q1FZD9GxE|C}Qbi$cy4E399@<9nv zAfM$+ha~*ve_r#PCE;dGzI{mQ*tgcg2c=$=_a}!W%zb;UVnw+ZP4JK3UOR8&zqPkr z`<*rN{9~&F^e*Ft-&woda)t5Hf7=(9J)?=Uz1__4LT<(5Md zE|&WhSr=csy4cUci&z6(e8%4-+17sXvn*F=L(6LqNxi7r0jU>Nd&eOOr;o3756bq(kFVW;NwM1TyUY7T@4V{{N_hXj zt&#WNLfmrY<|e-@)pD4qT(7#KF;`D5(q6P# ztJSn;NyKk=3~1Kh}PF?0({DmZzEjjAKu(kyqV(-QF3qI`1}z zi)&YLpS@Guo3At8{p4Ek7f9Khz0pIY41cobwBCH(VtatlPyb}?1HVA%X3_XVg#Oo0 z*TjqWRj_&`>i(ubIM-%T&9tYC&p)+x>=vnwuciNW9)7$QKg{b(1>fKxw<7>om{WjPsU{||P_zj~+HUw_^G zL*jqzPj~YE4-vOnKX9Fq{W)>1}= z{@y|)tgl~o#Otp!;#R`3KcB@G=(9X}_VO7Qo=Lun*v3DdOOVD_&L!Lj6=&lp@gi&=i>e)1oDUx*x|>xy zxH$YVg0XZCk?_@H#62sGb0Kw+@dr;Qu0E(VJ5_tdAp#eR)=txGUxpW}-D?j?c*$|1 zwx=X-J5D_7l6|sHV)&jiEpy7EVxF0Omf+bF-uwCAd*;#FIjFk-}-t*ZIrwMxtjCvNzoOK=IT$ z556jGJpWoEeaU?|f3!?>!6iFG2{PVv1M$LrGH@Y1aU(%ng+*bv#^A1m%S;zz@cdK6 z%pgt?FF#1m>rWA}eg;>=hI-vc6-N;LCQJ)WUpjtJoi@O)IsTPqlt+1+lOTMDVlihK?yH)5~TGfha`NNllc4r3GaR7H*U5Z zH@;BVS$TG~%}TFZT$p!HGQ@MNZ&~(%k#X$|)q#t+-+I+^ZrnXKrCm4hO%>5Ci|n@M z(-k}I@|+vb{Hm9DzGZ#C7~}K*gCH%pojv8^KKu*v-&_&$he-MQt;BP!&$)3q6*JEz z&fC2Mh|k3s>)d$A4sgx-)3*#2n;VSr*5_fy-M+77j7y$RnCE}9?YA1k;oVMs`;Er$ zKA+&M_{4y>AK!VhB{%c};`Qrq*e~~uW$XffxJxdxlll!e8m-%iU$)}YoJg0hJX+oC z&fM%HPMu@@@L_fJzjV8_TV?uSka*tFFPLSzT>6|@gw3lk724aB;Sljk>m!GW=ia$0 z{^~Gs%hAV|lE1$r`BM4emAAHrt8};VrZDk7>r;oxzw19&QW~cGV<%26N`BrcWAxEA zn{h0+b_b}bi`!%a{(oL*QC zSDtR5Y-i59%(7wBUq@VJ44-bj*f{-C;_2r7XfYuYFC|{Np4*@4&SC7AcIA9>C+FNv z#=C!Z{2D7h7i09p^*!_G%$MuLgI23@So>V_Yuh#1D$L9FGimu4@&rkHyxp^UpOS;vpd9TD{@}6@a>(Q>!crL> zwt2XGWJFwGwO&D*Yt;G`W z2PbRikP`mElt?a$WLUPF8Z}S4cCIX^HH66w(&8YxYhb9a&up)8vpdY z;|}9Rw-cuq*a~f@5r2&|UVA$%0e9uqJ9-1Dh}WeOewdo#5I&LJL!|M1(g5FWy|IJZk!D+88`nO3 z?uBQb_uE9{yyxwQ^wj?kAU(sEAG!79CYooZHvdc>i01h;c_5kx9b~`yoXd>QzK(d? z`iaxap}e91dBy9AJFS)$oj4#@@Q{Qv?4E2Y^#(^nK8>jW1XHRMT&HqEZ_n<8Ac`KHSj~BOs^~d(te`SwalxS;@`r)r^ zi+Id<DiGKyyIeyl=`p_#Wa0>+^12oidj^w4DtSE=CbP7rVCp z-n~Wt-S-lov6c=i^)>&#EpD^aXFmKs;`!NW=uc3KyADbCoIfR8RtP{X?8a~X>B_0`9e+yPeC!QoQ7)HWw`|BuojaMXt_)|; zPjid)ZHLv?@82%(R&9@eka(eG{e$~zYb++f$99!E*WOP4!JCcm{24K|;?wv0=Ur>L z*?9S%6aRTuIgg%Q&TAfoRs&)dUD$tcPZyqf`Gc@4SpVa&T5-R#U98O={qesb!q)Y3 zZ`n7XA|HkI?-{qO4k-D|cItC)G5+?W#L#-~EsKGPKc$ahKR-auqH#_iBIkL3NxXDN z&f=Q%<-a8Ov&xDdBI^SmC!W6|%NIOp*#3%m{mNJ7EsXeU;(x7tWxn-yf4%yaVdrr& zd*>&J*Q~r|ro7{m#0_UH*xieQZEV`Mj7SU{rzI^CY zD--_vpCZ1y{=ohYSuD_hw5#&B?{vrmw-_(^3?Uwr@GGCe72?4|61F}|$W{ngl;!wC zd&kwa4-r5Qe%W~9Z_c)n-u3AAz}Q;Q@}DC*$Nt5v#pTksZ&}`$?jfGBwFc~(j^}@# z_`b#R?}udY=F=-Fm$$@?w;W-P9ibK(tWnE_^Rr*vTlQ~$k$9iQx=tOk8h=dRC;KOV z3x)ADn?hgz@E)PY557XY#pu5E$kUDbmx;Go?T1PK#wS;$-}%lX!DE4Cp?ypFm;Uha zl{dBqigfoGV3K?>(|%B)>|m8{>Z+jlMDqQNWG#-8w zYVV(G#AVCh-tY}v1V6B=q`9wdOMDk){NNj~a2+7b^G)J)*3TT0@Zx_aUV2c%SNt)i*M!O1Qyy3gVZ5QoNb;`Kyd&CXK zW#7XJ_e*C>`}Xe<%4Q)BZi>aQ9WR<~Z1}%lh8LTa)Pzca>^aO5>AH5H0H~ zZHs05#LAHsizpj&{JOxy+w<=}sZ=KD6d1jbHx}ah2sO zi`SPQ7a5QIh*-z(|MnvsJhoQ$kmU${^i`JU-iMX_@#Wi(8(04)an;fPFk8U<@}oZ| zE;ek|BhRobe?R%$`(A$E7Q*x1A^BM+?>~PNqPFGtuq}S~%X@P) z`lc-7+|2@+O|V%O;|1p&Iezpw^dhU}U!;Gl@?!pY5!MO{bb);kN+1kyXqj3 z&;2=Z9vUmTtFac6e*;g93x0|t+iH$iAAP~{2)p{_#m2jdBc_&X-Hjq7{24Roq=tN(s;DP{3- zk?{cwnrpMqe$((-k8Jkr^7n&!cHdi9Z*}UnWQbY$a<9-`owu6i;$5A$`^izW4{Y{D zx9)m!GP*MOya&8sP;027|DY$U9lJ}YfsQ?XG`}Z}g+R4lxaZ47?bx5<_06u`EeKtE z{Ah10&3ld>v7bR4D&FTDQ`Nzgs;6XyX(GU<8xURG{VPhaJ9`J*) zvLUK{UTF{dr#B|8Hs9K4_csQr3`t{Qqdh)Yz|#k!@rAb>f9Cv_jY11Abp-_V+E51f zEnLoUfsy|2lx1$#W@qO9-^zB8D|Uu=SQgXll6N?keYd5ZL<5}KY@uiGZK3TwZu^ys z_r79EUJelE&2L!g=A8j@-i|K#Rloi87m9M7zx{>7x~#P8-4{EI(U)`xVLLvW8lU{_R+vSHM{Br594jZHI{rF1Z7dMZJwiqs; zR_QBkWriUsw)?_(V^YWeR$Va%9B6d9LSr%RqQZFG#)$6K6}dg#xbdumxo&*Kf8t_k zFhmgbwl18yWvhn&zvg%4N`BkTB7Mai43|p}qZ@!LIPtOX?rntIDg{i+J#{nn^y3ZG zwcQ&IWCiWBArHC&zG;|^vTUfvhB%Nn6eIKGkxS$C@m3}OKaqC{7F>!O3CE%TK zKzpL0jL$_*JTrsFt9QqPjUhb6(Xd$P>RqfciZ_MeNm+@j>gvU4UvgXbck-&?e(wya_3EFFtUq>OBY&vGc5JsrrB`43bSglDPiS5vcxaw z$<4%H`uKu%vd&%w9oarAn6I-{xmg_Fd|Th`!hb95ZRwMrF)G|kjLPE`@Gj{Y#ozQ& z#bMpwBQBoD@E-9Zlla|#>O<~4ZOHFD)1Q71KJ}?Rcr+|mFggGco`uU${I@NGJCF7M zyS6vHcvW43hTExwK3>Buiq~q-UT2SN#k=10>@VI_eCvOHn7gxkZ=tB!%R7hr$O{$= z1hzMn29mgo68+ZW(g<-4SLNj^_<7?;-TBiKpy40BqVvjoS2{e(&J(eQH1~N&vm}~-u<6_iB>=K zM61f<;y+7#>uWn-@w9S}6K%io(uJ@8;I*B$zNzHe3UPXKiPkH_7 zPwo8p)tJ0&kN+xL!F>929>4ujx3p(I=l8#{^Uhc9FVFeLBqMKj)%Ab=%bkyrXz{(D z*7?SdyL#tG*Y)$0U4KZl_~zey)b)2f82|t7M2jQ`dIhj8O8D`vB|6Q+V}AX)?91#= zhc_L4IratnZ1vtee=!~7`ZNF4%U^r_{p`y=@%p#Fj+J+i&thdf|L&?2-n;*kFIcuY zbiVUN1p(Xr$~S#{|LRvCulFZQec!2LU;QNco=b)sD*2qYW4&Z0fAmt7y#DPS`$eT5 z_W2X3m+td#y}d(Tn!T|scRKbl?78RP^^Q*G-qq(*AJwM&T$9t*kZ+mCw3$EbjWu1NT zoj*NuoxSxXjxgq{I`{8g{n$r0TR-yj+4}ttKI-1{e>$`E=CAI2`hV+e-Rb2ApFUg9 zKYO;m>}xviEC1iyUGg6v?2>Q)&W~}IJpTn>d)eWypg*b|{%7bXcKE|>a{tYr{Lp4z z(SI7-P`VMvPzjfsAbo0~br?UI5=$G0j^8efkULnB4-Q1ARt~=eU;q92c=QYvG z9wyn|(tk7%0)73)cXdAZWtTGVS$yxCufPARUjDLI-}%B9)6ci>>hM?hg9k(QaN~iE zzj?{V_WFZ=we#o(&Yyo=bwFRg?VCEEdlk?(Ck*`h4d2xHdyjv;$x0snm(Dlc`RGCk zdE2`?@4NHSh5qqpzq#`*FZk$JzqzBoWS{=-Gy3#*-qRUBuIu_U-`i2(*BkTj>N~yv zV)*ECOXNr2+xhMLA9(SzzyCX6s(7L zJL1y{-c0HhbN66))UX6Lv8=p?|ki%>HE6p|K@jtRlWMavsyU5>h%w;=ED|l zXgdGqw(4u-2O!A4+wJSueP8GATxdE!boIN>s`#7T7gqfI*Z!T(|9J2IpS*Y?;Kesz zfA8hj@BXRBmgc+8kBa{ZbNvnfYe#w0iwoXi;a|Vz{hc>nzv}%R6kcce(iOe^{hcp- zS`Tib=MTTX^H$&yzI18r?{?nyw9?=9cRK^Pe)-Y`?_Onw$GKUe=0CVyq8IYgE9UIY zgu0hH;8NgEe?sSz9}mFY;eQ^6e{t_#biMfJotNNW|KQ(xG;|NA^g>^HMRVIm^`H%S zZQkqu$f?oT4Me6-z5Yu-(0R@E<_9`|FIPxcuJB-td^N;OEV@x$~#H18@y=rr(2iQ@nJKeqVT0=fmg! zp?}z!yy9nWpZ}*HKmYZ={_z*C?l1XAo$tH%3(u(i{<~jT`}5o9!F##?krzMk&6_9w z{Sy!TmLGal{3Go3AN^3rcv``m<(U6480`OnyL%KoS^8Ih7~W&_lxAP|Bj6Z((lZKv z-H&uW{R*1DHI+Z{f-!jh_kN_)yLa#NuHXIhFSW$*fAi=m9t`pG-|~+;pMUTE7jiF} z^~)e!Nl)}f=zPk*x&Z8y{Zy0_g<|3CSnJKwl7{a^4eI`zG)&w5tn9}4dj zy_x^#U-5GgI=+s$G;!DHJ$e<^)fc|)`h$P-<=r>F^lPw)>#zLX&TFrK@LzU5^NM)e z10(pZJ1sr#&{pVce>pWP{W z^R|+I^nZ4K{@&FuyzQDFz5286z3uw_zu5U}4}ZO#_^hHbn z?m2M$ok!I}js0Kle8rPh{o7yeTtE5A_y6n8mtH=(nEY4YDGr-M`<0G!Hzoui?E8PE z^EY1k7oLN`dSmk0&)r1jx9@&FJ#P3d&prQ5zuI~4)fMsFdoCCw@!a*`-*&$D!LN`K z`*U9hc8Z5WlZ@@=i`SmY<-Sr>-8bpA%AH9~iv~jax9&%{m9{taKxO2MtlG|Fp zo{`8jTZa4ggf1^|0yL#VkFW>jv^%I3~p8b=q-}K2J_nQ0fd-Q~l8b1G-kE5>s;p2wC>hTjmMSu8Poo|GSZX3Sy zrw0w+J6-?k|J^yg>b>96+dp{C;79&l=S%N@;Kl9Ce!F8{{o>lDA9(KiH-EeH zMOPoV>HlkARQQwMMO=UB?{tVyxRW%!{iJirdqVNcLelQk_?8FlL(KT%l+jB7ZjU)R z?oYkq`Oo+dozu&&-~6Tya{Y$R%m35$5B@QvbDsV=a{XnW`|+Q61zy8PlZW|}=biwJ zvIn^8J3r~u9$KqEe%$$w23LLS!%LxWO>VA@z7()}fmyGw-~H2Y>*f1i@$#?wi+A7s z{(>r=ar5+XLgf0*?90CF)4MG3<=u}q4RUpFas6k%^W)xl{p+uK`8S>iLu@0BW@9ol zi6C_ADHjn68!W?eY?`rz4F*9QTf3n&rOLrF)!C6;Nc-6wG}(#ut~)8lUFl#o^6_BP zpu4*0I_s*t$ENaFpYW^UU?dGzy;Zs;(-p5$1AB=(V}e4LO3*7#nsg2)wy#wbVhJF& zXoB7+tM4tfQbN$Z-oOOdpnZ&`QH8>|PruEkFUX{?Ww;pp-~LJs&T58oOq z^oYNFhurB8J~)lKK`tAbXRD`uQ-;T^&BFm`C=A!9>M%dgKnl#axoQgH>{^MTK3)w& zR?NALgfm`DQAD+@6`r!SgEkV`Igzcc;D`_h5SWmO1zIq!YlXsg1s#&cY|0P$HFQbr z4tNPZVyDuOzv&|deSEUhJKi(Syn?c1gMh9`I!RA~1rjbO!Y}#5V!Ky+%S^|ar72T1 zJ#mSUvuun?aesVF8WPP61{7pQw8E1;(9>wj3IsFN^yJkRK@b?XnL9Iav#!^Ja$xt@ zHSZfGo-De4!Kf!{x-^c;)2j5hYYhn3a%R&q-k0+RHJDE-EcvK;@fq1 zCXc7AKC;Ey7|qVIJs6NI8_1aknrsVUfL3MbIP(D+oS1Uwq+@BXkQgnk`6#21%4$z1 zrfr(I(6?ewSnUQqUtIVhL8bGaO(D@Hv;Ao9n|4MK*1F+0y%AZAQ_Wx0A?Km5Z`~Gj#v?uc+b#+pyW7^7D@=EOPXGzqK` z57pfLBHOMpM@N)+5C$B92FEfOZzil|TNPjWE?OOEj+^!i!qc3EKFb$-q6k+tE2D-W zil9+bRZ(d9wp*;;il&u&<56W;-JvW5uB(V%vtkJl)!~ljX}u|@kvDW5 z=9ugPH&p%HF}j4^MKl_w@2~{glN^U`xquZIfb0|3ss(-kqY!)bVm3H&T;2BQ$UcQ< zO^%R>?{jKxZo2w5naCO1rol=Cap>IW9vDv0ta)qn;sKh-peVlH^?DH~u!T0rnD~)M zh=?f5U6!JdFp8=JL77nodY!0FmWopCW;q^NB#2B|h(@xG&R!bN#CSZ~3ai0+0Kdm0 zF3jzw%b>{CpJ2>Tabm>D7UAM8g9JX=*yJRg(=U27~Y(=;~k_%dVHJVQ)QHhE|^q zX0F?nFS;U#je#)M9Hd&NV|hHsn#%9hsh$ibx(2eJ+ucDI_kl__#4H5+r`k1}7bpqh zOz>a~b2`O`!(o!N)BcUAh5M&zJeu*b?-r2qKpqz;_zMFUj%j z(31Q>1Q9WVL`slE^*9uC^g@`8(}qMR$31SuqukAAil~6WwWRY#?J{zDnrWcIk|SsZ zBF^oG=J>TDf>`HRnrQS8`c;Io8VI4{)Bz{ZoR}W`U^3kd^iltK zP)ltcR+?L(v>1SBP`f8N0wk=ruH4rGF6v&ixY|=W!6xYtG;;M7-StJ(n(ynd?za2d z8CMl`tovoH)|?rS-Hf5;0cdYE7~ef`QHz3f8j;2EI2xRm+xQ~TbOdd*^)|Ejv1Ch_ z!H;-hzi?7eks3!_dxhldG0hjGQhbt8a3R0Vdwl|NS9ADyts98VFmlrG$i`sJ_up z{6oLk65`rpN?sAe77P00J>JH9jH!?XQFSrC*4O88w-rv5c(LWlN)Z`)!8?iw(#9Y~ zOEmfzvrN=tml&~#TD)LcK@vDr21)3K=j))XB~UFZMXYqu7@gy@?R2xBPuy8ornyBy z%+UfD6{lLjw%zJs_cUE&n+fIe^J5obGEzH}>$=xFIPD6Qk&^ksEGp2B9l@fMK(34_ z&Fyx}2%a{meS>7+a9i36$eCXbr#z?%wKv`9j?xo&-qB`9iO@Ff0Z736KB&~?r=>YI z`7pA!Yy=v2sC78d>_iz8#`=UHxXf>iGm(O}qzl?&ESB2qK+BkBO$GzdL^4+@NLe+5 zsjMdvn)zE|JnZvRCOd?S;cCG19BP9MFP-kq5o<&(F=M@2)P`V2MW4Bh7Q=xWR?c#1mKQF5Wrml)~{w@g}%JqzdEWViDIk{(KJNFtUZZYLK_ zrL1y8;@D+a6g2cGwrfqi^EO`3v`sShk|hzOJxM*vCu(&lL3n$D7yEgHnuAcuvb49L z>BSmO9y2-;3aAjyR{3nRsN8%rn^8nCKsgo^)jd7%1I|*pi}iCr`Af4S&@Y|s`NJsAFL5{ zQj}$1I*{z#AmL0{A#GIEh3)FqApE@_6NuihQ$PCNnDRFF?oH>4ChmmHW7oH!_3KM(Ye2 z=j@3B!bIh9uk8eFH>OGA;1&!)ECi0;Fd@fnn1O{Qx`t~BGLUGchjOC`r`ddrjHRVy z%=Tqg)&eG6yxZYK@PjTM4-wSdsd0Z`&qwXdM6$_2BsdKl z2_x_}_bLs?b_j-I6AshjsLPq2NJ(CAreYtyjFQtG4~?&lNm7~CCerFtb`A~;bKN3x zJ~!ypA?yN_)?GI4tuAi7)>V~mB+ZcuCqWE-GN_FKoQUN?)BH9Lt9{fPUE+-XI6Mj) zr0F9>1`3=5!(qJPJj@JqwA0v1CY%tS!;oZ9A9Prx+#N5jc)S$(^FUI2qa)~DR>fc< zB>h#5h7BnA5k%x16lOz(P{7nGQ;BP<77IL`p5`o_H2NH8^My6&!P(fYOZk)?#eHu! z9mRZ1Q$^{>6aREFI5~+<^pwZ+kr(H3lCW5;9mottk8X%2XA1G<%B>fmU$@t&yvuOM zQ-mNzgklrIFNlCaGYo6&ZR?gjgQLijk%w}lG^V>?R%qxj^mJF$b&X9z#0r+1pfVuD z3uHdzm12tyg915`-UeN%vqg!R>kW=;;?~+3GmwE+&!F>`W(nA+ z2(3opjXwb9stmQYR~&G08EuwZ3Ewgqf@G2;*sc+*m*aH4SQSQVRva@QJDXUTELsq# zR^4%u$F{<0-N>#ICSJ7zi24kIX{n5uEEmeTMLdYAs-zm#BE4z2PEPSfb@RV~u$%bf!uwz_4O5$WsQJMWg-%#0|G1 zpZ07WPez^we_fIhimsHPP6j6_$&P*yQuGq7!f=_j{kh#AXJ7?Q6D#U+R`aqjosrpy zUVGUd@2|USa*fS;j7XYi*!zspo^Yua#XY`QR#;YpkEBl+5uH#{wp+5oYLX_$)vg+n`S6$B_Dw1U z&@}|;&Vt=Bgm4!NSP|7+PK{G{&W=$RE*I5~eljb1+Oj*IDG@k@6k;{x&0U~^bpai5 z<;=2s4|Sz}TkdGos*6M*qLij!qPoHC*d6l%7{O>OxBtW25h$LQLQKouTm z=KE~UpzUV$L^MDhJ8c#Z|1TGY<-kyvzwh3N0?=|UJ?MoHR>hfsqZ zS*k>UvWra|*=5%YhN)X66vBCX%nuMJ23W?~&bsb2N6x2p0NU}OXnmP$T36xf*n+G| zMp+`a9T?bbnja8X*q9ldFKd`n+CaJkN8o9bbLYN$1Mo;|&??^qb+zjvt&)qYLGL&? z`Iu^$Sy0gB=~PWuFhz68IkmiAl#EUy)%08=phCac&jxx*a3CLxpM+|I%)2*@7lXA4 zX40xMRuVT%kLBxyzy?ier`BrJN3tzb^yb11up}Rrhy9M8Wv9u>OOyRP*40v_aX&6? zp3UW>NlftkB=eVJdMpnobm2wW8L*OGbkOH(BQa7(T2A83$4+wwk?Fw=l%_2g@$MAN z0gXUh15`4t!Y>_4511pdFry;h@6wy$2dD(%0je#$Do=RI9ZdQ{BJ3w?E>9gL)DD$; zI|6_qh0~Lx<@+`GUrd4kj_m!=*+^rEMB;@GD#2{3fqFDGTwFAQ<5Ah*NLSCq_1JW+EwA>%EN7ObYwi!j z)#}o>IYe~A`B2PC2-e3#alvc_d9wC4=49Dyrwzu8eY3PAE|2!}N(Sj`E?CVuZQ7Mb zAZdG)E*iP|e50c}ilRIM;KOboqJBE%$>U8lnQ<#f1r9YB|H9NSpieC0-y`wR7nE*J?)Gg@|D56in z8{L!L3+BTPP6&SjG(64^2c#f5*5oMVc;E7pAAR+D>*`>^6 zx*sl+rKZUT3=DKjb=A2u&7vy9=ca3HM`vo9U&3m;4*^n`qlCwem&mTE4sdQ`3}Ft4 zeAaM!b4(ZbCa*bgxy(oTf&z4`JwhBTW&I{Cn~}9cek;hq3vwSp5`FD zEYiEPI&Ax%4U*!0_Fx=Kyx$x&Bl4iLlF%QI_YMS1hs8$0$RS=ed3WdOiyf@3xqM)Z zXwcPfzqge}#d{B50sxZ)c3L^+{iuZU8@#Z8$=+?h6ZTwN5KlXF|)^uR4+ z$d_pLCXG~FIXSc4Qf0T3)g>CX4~b7WCvVr?j>nsf9Z8M8z`dZAvIqWvVWq;&alA;~ zW-;v^<@L-?yN&ED+Vw~hcX8I;ooKIhj+IY>|9i=sQG;V)5vN|@K%h$>N3R86FL)NOEW zHbUFQguLyns;0^*PpG+}XTrtF>~7%{+%-c{Q0xh5w`OuvXB^2of5IcTTq^Uj-7R5v zseI(I5us`jqu_|Ep7+76`y@6e6S4?wu@RQVK-0%*pKo z3>=N~#bs(XK{Ov&$xNxxl218$ocIl_iHUvd|EIQ2W<6?RC7je|OL=Gch@9p(s0X1R zj*Cs!EJyPVxGVv|vD`5x+@w2WBCJerMq$H?)5iJ!WdnTo1-{dDI7gXn3Aio8)GzPw+3WnjsR05$eyd{L9mu}Fj|JACJUsS z(>L=z=4R6wp|Kk$!{J~g0umj=POwrPI&4b<0q1{4Yt6C+o%XZ5G^Tce`qYheY6eR{ zDZt*4NCSq=ESWCK3?^Y?9+7B|+{`Pv)*->PjE-DoVJ;yvRR#G>nA2oDWg;Az@UT-L zH=OlOmA9NJ{n-eN$l$aMletQ1<1S#S-Uf=%+yOg_1iEigZhg_ZZw-pwj>FKDWL{S7 zDYh}~t-mZ4$X*(%##G^UPcK_P*kR3%1%#JF*d||!t0f=08;Vc>R-HmNba43x5}oqn zFox{vcxcO}n~iOwf21Mn{-x5ndR z_s|Nk6C7=W4Qug|v^60yVGJhI?Ky@(9?@i)MZhS|;A|%FfhTOZ(!{jkDVr>f5!D;5 zOmgAuPyzf=1|cSUKCAn@fHOs_ZbdoSig4~!kRF`?avv$(ZkJz&Y=+7}g@ zuabo8A(30$OtXAmHy%XuQKX1M%er#oXJ|&@K?@o11CzCF0)d3nmh>f^7D7(9YmBCf z*bl%#1y2-#v(?NlCQz6hyY8{i^>WC2%k+4TqJw#12$T+)lY+_6g%D8PBhoCVkdQ7B zqTFdirg!#|X$!}5*6h$@yxisvS}#VK&Iw7~U#W*hKRaTyDI*ySOd$DwHHpAm<+ARf znOm0cd0p_>kHb@*)1KcB=Q`geN1Y26)?qg6(PeE;>5$pFHSTFUrdL-o?Wps>#5t5H zc;w8|-PX6zE`LEeO|p0FUhFP4J06ju-%=#L-tCT+JchNmm_nag+_eRb!NP6b%V+Cy zvtPKSgdp*VT$W)ugRED!4B$V@YU`S7SeETd4bbm!n8Y8Z?_RC@uR@5X6x5B@nc!r;$lqVnm8uNe)DbGHi3eoma4_&X8V{law+c z_hWAskf@tMqHfUWHh_yfEayFB5(c~35Ce}^x2={$q?jK&kcs4)Y;=KUs!P~S0IV6a zr;2U)xnz_Z&5w+AZ^>Fl;_s)aJx??;48t=J%em;tJoXmi2DAM0ZOC+CBUJI$ZfJ0uU4C2>s~$9rhP@L_y09E)-Nk%!TJeR@0wadbqEHC> zfV<#oZ?X;{Gcp#_hD#bN+~#9+g&z)cR#YxX*f}Z$F$PXY+eUeu>PHzt;k7dd3`jpc zO3Rj(N@w8sip6LJ7&oNE=aXtUC3XJAwi}*p?jLrH%IWg%ZWzcDl zg!9K zSW}u4_bkW_piBfl_ToJAG-2KwK~~hWck4l=M!|^86dIk60V|gv4YNy!6zO@0z(-Wb zq$e;iX1^A~5W%{DK@o1n5O9bQS#?^I&lxI7nDG)DvLgxf-X8M!i4s%9Dw^UlISfO<7Gi(!!koQ zg<7|pidR=iWbf-kY^@R@ZQFf96b#B8o&HgZl}4J<0HeA`kr?cD4X>}dep~~|iK8Lc zbbR_8@7JPC%Fh2K?0Q^tCeYx#SxEy@1JNI2OxwA%zUBu1Jg6S?Eq>7 zD^^(*TW=%4DZt6-IrX$w{nRRUtC3GFw~%s7g2lphv%_8wcO%Kkn&F{1AW^^&)Om=| z&B_xJ84&L*NpgusyuuBITG{m>iE(Uc`j7)>Vgwmah%XDeav8R-O&e+7b5+wlO$-QH zrZX`2+-9wAIkJ+)FQaaG#=~} zfM$=tsLONsAAG;tk+Nr-ZUBZ>s(&tkaNsYaGf2kOX-m5 z9p>5b4+?^l23xr6Dko3{iDqoG$+C#b-DIGYGFBX9-r2 zGtU6~Vtq}D<^6EgisT@#ax^9>H-g1d)jf2coFFkW5stHRmJxBIH8?*3Zby#8$>-HJ zL^A_xwm3Lssjm4IFG5&U0%~HfX&OQOT}_XtBd|VODg!E=eoBTG;U1Sv<#i!zouBPM zSdEVw^bk!LRK8&J3l9oOlf{PD6~M`I$m$-L5e@q?1xIx;9^o2xoN)_rz(P`D-f!vL zkpeUhw4+yweJ#-lst;(#2*|I}DZ&To!eXeYGcLjgf&+7S!V9W2W_v z)KXQooNWw;KViO{nWuS3)CQ1xAvM?-`=vmNE4{k~?rC}&)nI}krz%`H7#C=jnfAct z8t!jSMHE>U5pArS0S2aJG`_ImBtY>6*6fL5tKy6tL#TI#7*XNl3vYEj)-!z@frBCf zKyS`XVp(4HKG^=W>~o4N^dYGT$(TV3IU0OvT^JA*@XvGzD+PFFyQU1w%C{4BS`LyOwY`K` zzHuVc#UxdM!^O#PR7Q}cDXk=oG*#_x>7bXw3OxpEhdcaezA9qV zRx2m-ZxWSWiVb(HK(aPnGH@5A?!Ib8Luda9ZM}C>%M1Z z(}TvKU;u%fE}xuCt!)UaFpHa=XMw}xm}`s-BM69*qah0waMxfzwPC{o3oED;VES<@ z$P2*`bcr%j`h6nnc4v_CKOBx$k78>eEX|o3A`NQW3|8Y-Wzh@~fraH8+uQ{-gtL|k zUmN?%Kz-)I891@1z906^E7D_*#vEqBLb*wFq+1tyzg8o0UYJ}R(CLMQ)B-)BRO=Fu z3`6Y$D{=;m7qVaQ$E=|$U^4}ZiPN(n4;M8$D{1Puvo2=}ff8i^WDjb8YyS)21!i#& z0?$(JHGW;6N6W~yw4t{(AoFGdBdY=-t7b({OM5+3-5`w0c4LGIFxyR| zt&VTcJ%ZVDZW0{m#Ws!szd6Q{dYW!XcSLJ+>9u;JUlt-sP(`1ggJ0+D4!t!5H^(J} z(8JztttVo041tJ7jyIfujiokU)}k>)Pmq&jkeY^6909h24V(a3Lca2rl!S+CBREqc ze{mMw3AWJ)1L4HDrUEVpjz7`^XP-_AZ88N9NgLSnef*GO!nldpenW;JU;}S^y+GGS3jxAL)%n zoBhBT742l0g-+coEffKMD3CVuoz`p9o0uLFgg!vbsI_$qUkfX@XfK}eB@I*{a&5xt z=R+E3pVk6tL823naORQ*a$*1OHGmt3nhe|%rZ`bE2>Y>PP`PK^H#223ryCEhrSx1# zHe?Glu%`~GKoZn*OWZR7IN>9jj%G=UU~Rt~9~aSy&r~vkxHC$Y;E?E!CiSOSuk{F; z*G5acV!|$EK0Uq$O^;l28V1&dNhQN!Ln8GqhiJg&~#bE+pQT2Vl*{)Is83+xHwnr4U zO$m-ezXvJv9dBtT_Lx9`QbKxgGg&oYZn6Qw0s3JaYII^OVhN7$TH|KC0C}I@1dk>K zL^=LMQ&u`Z?8m#1wk@m&X=4IX3_*nr&)_`t$dtoZd%7jLZlGAYn(gpm9!S`nP`iA9 zPDgOPBTUqZ6t7(1QDy+I`auFHHp%L`b;*?X2%rzHoYe)45NC9{=GQ!iY}N~WXA%gB zpao(^DUkXDK_f|X;Pws>go?I(ZdsYs^Qvl_1S!FkU)W2-CaM@J9zryW%t;$YJ$HT6 z+7)_&9GxMs*M_l7W3yQa$!E9e^E~7mlhMr)te$KSNM}U0)%{s+O}kYl49hvXN*lDd zie<2MEA#*zAiLYaW^hisKnYxybP?rLcpOdZmLDvHN@{?3u=ch&K?d|XTCFe8I=I}X zGD)^@pu&BdZV=-VjxA=9Yyt|8fSrMG(@-U6bYV`ahiSRlWqp8M8;jy%KVAxZZv^X@ ztKyO?vGK)6JdfqvO-2%=-&uh7ZKn%(05~admqcF8QG$XiIq6*RSNhsaCUH`1MQ~wV zdL)E=&6DiH@KEJ~p+*)3H=zh5xLjf~QguEkRShmhxK2CISfqcd<`al+#8c4K?Ve17u8@QoizdLQC1FjDG@j`3DwjmD_n7=+Opx$QFrPjEw zZZ0RVzP0+I#gI*lt|=r?W{?NjVWBkx-cLF$9CyQrW#Td?yKuxq081>d&)SYQrx$um zN=S18wh1uWRlmjYZ3R#U!6arT)JugWM;yA)#2zFQHu~r!Cf2AoO@a-?_em}Gd|Ct# z$bqEwLh})9w7{osm!GUE4F|lq#P|!riC^jB&2Hg>EqawD#0-C%kXWay67 zH>5={D2Doc45mM{GJytCjy|ooWKv4Yeo7O&kv#0h6X`r*o6rJSQWgQ9Wf-yF7aP{B zO$MwM!0&cMaSIM@BQW$d3xp&sn>huMTM97YTF)hIZSL?US6YCBk@~t1zW-v}ppwI$ z9Gy26EnNfmMr?(elzOL}G4_`rb=5%Z4aW(2p4WDyG4a53HvwegdnUS!(@g?46=DFi zx(!Y}9vE@MJ>0w_OoD@`uDC>H2?YUuT$^K3cZWUAm~`?>zo-RS+xjt5r$#YcnNMHhO-+{(JN$BlM_yeqj| za^L4*J!%t7Nb9YWhnQClNz*v=1sC$*fZPv z(q*BpATvnw)MPpvz~uuu#`>z&Z?@^qrKD}U)p;7uM3qJmWWBq1?vhZ4%pQ#{v8w<~ z+3`Ln_JJms{$%bK$Qd@Imhgbe@-ct{N<2(R{o@x~hlW1M0VAY8znTw4GWEruCAbESy~SL+exn-9l4%O^9pK?;j*Xg1R$| zb@1afB7^g083N$wQY-!#je;&I1>)s0A~)EQFXL4LOa2O0s^|< zUtsq>t>coj>Gf9mAvDo|a8JP|ErS}H0R>G!xC9mZBEHTsI97o?(a;u-J7J_w>O3*E z_1OcWAb)_k$rShCV3&!p2ae?CavLtUm*6NxRW#iIpTv>6m|N_|-~r8hy@r`V4B6J> za$1KxCeUHBhmF_@i{o@63%0aN9Wbk$<&*7VtlL_zva~Qn19xE;riB+o;{za=pOvN6 z9YkU7E<)fut)v>IdE&@fd#hhA*4ikyccFGhBei3^gs8LB>q;31Ip~ zEjC0+UjaJahR-4(Jgj@xtQG+aYDjV*DWTaDgyxVa0L`wB?Q)t>esrdKHst7q-XI-% z;Sh)n5Yiq2I`>wX4Dr1OJbmAXJD=Na!vVX~O_!G3lTP?-)jRU1xire+6_o>Xp8#KB zbV*{6h_*iItzHdJ@~k_a>V0=84_<&1Sv6wfmx@*Gm^uY8<3 z`aX%e9#P1VDk^{>12+q9RL(0f&`L6qC^M+gc5a+QxKP;OINnX?_)?t855w>boQh&z zxx4O#QVEC1g$@c=C)SB&xzj1#%=6m6u-h*oIN!p5q6n6!$3wmth=@@M0GF~JqVag_ zHgNl8S||iEZ4beb1B&v@%c;|V8+&JF-{Kg-Ed*>qn#2i+1DYjzLaU}N#SENEvG>yw zqFA_?Gj~OTDY|ehpCE-Dd%QZdQ*5J77BVCr8OU1$NZqhjZ9wjb$S7(aQkPrLxKD8= z0T-1iG{6vqBmta=z~?y8d*$2+%Q-L}LTedOiyVXyw!3A&=fPYMV&CP002_*^)(x^9 zT*h`&aU@fr0(XIf{^eS=oWcF4_4>rJS_WvxYL-F&x$D9)dt_j&0d) zKgvi%#o$;HN0>aUW?}B1v0*bEPkp%B$rZU5Ikcsv7(_4Gtg9{?Z+YZ0F=eHUA83Lh zso>Y+^A$tKwCg=!&EG`6m_pSUV2y3f>p+XGgB7Io=3sM%LxiF=Slg+)8kj*9m0$~FPgh`m*NWIg0-)Yq8Gk1C~ zibIhiMT7v?V3Z;~1PJh45Rf9Gh=O2=pduiTp9K|B-%~97zH_JD-Pv6d-h2DUoxSJY z^PTUU^PTU0=W7RCGh;RIC5pr=?r=@$u$glvd9~zM#u5^f1~^Pd5H`1LQ?f`y3+m8R zp{*_IWEzP16vdkg2>a8keN$vI#4jYI87eM=SV5^9teAGUxe%fnAOZ&|iNoD=*mDpk z36$pqFIY{hnTyvwR@*5#TM zX{*KtQCpKc+G2Ut>1%~NAZez-_6nB7YQ*PoR#nop$q`~Uk?F8C7n5rXnM~9WtosTX z8Q+w(3Xsoe-O#MZ;j?P5`I0bF+JZqOG&r2bCOGz*E0i@0UrdiYNSYj`Y}4Zdr=bpL zsV!EjD3fMnsYInIO_ghurcrIm`XI z`B+TH7XYFV1EK+cQBSih)7k^#5E#cYR+pw4ft?x{gN<>04IHR4UY!UW71CM{d_Lo4 zekb6Kk*;#P1s)CRY$4<*o4iRuz1;8$!EIWId_3*o7}!uo137TLmB`F|xx}Ds!L~c# zPzEy)cAj)O6Yw%6VO}$YUoEby8laaEw$5=g49RH~hDfjr#y=*rTRG~ck09xa%} zO4&?Dk|=5QYNbL1LG1>l;5VflTD3gc5(v{mh&v-q)Ko@QkV8y>DiaN;Riw$5l^V?v zi@98ugrZrM$4)}R7Le6ERY$s>sU&^2WJ{Y5)SMNG#Hxc%+GHtbNJ9kzDt^qO6*s*? zPuKz(Q!<`f#}$KUHf=~5D_T7bT~aGBX;lWwQMM(GGB>jOxE)>%np5vT>v>GX?bQoh5ciI-H!GbZs=ha4QmU+1vwTqb+exXq;crj@ zswOkUoUo5CK>&rt%}4W{unE9T!LH;-r&6Vizz^YCL4{4wmqE^gP5Tt>QYiW?8 zG}9w05%XRwO}e=(rX>^QY;{JD=9E8RE?Cr7=w%(=9C`vS!S;#VN}Eg zev=eFY!<0boyanq7F}57BF&_~B=3}RR6$&HxZ4(36h=sO0-H*=%>r5~u&ja=4rxYI zP`NTif*U0`e#VLpskBva2v62}6zZIz5j1Me;BILN)k@NIIT<74 zf@B;V$sz8mB${%9ONyKCG1QUa&nAv}0{WIs6evPg2xq>{Pc*Nrb_!y5v5+@eTM*;b5;q-!TvP| zJV{Dpj-<<6beO>pSe1p1Mi30z(x%d@OIbB(wT=qg9XYK_3rUKUc@#$!j5-X$NH5h- zLvQNF7y0yLw4*U*eFCrpg^k#nic=HsXM^E_Q>n1}=*Jv{Q;S)@5DLNg90F zP_Pkogsc!B9A@^KHls7=)r1^*6J-zO9Rf4N_-3kLxazuU`c=sDR8tKnz%n9N`&5EL z3{E{z%L5^K)ZY>L?8XqNk8D(}s&Ck84O`oswuVyyttzKY$-$A*%P%+0O{uyoTuPiV zwNw}xzjZQv)Qb@1ygpP;0x5A8-^ z+)CCc@PAZx0>xlfCW_>vR--5rEbV{;O}5=kvVfAoGzrTl6V6& zM95yGag{8!QWU%kLI0sBMiW+v#GzN!+e#l;%c~@ElZPF>BI27R89alh*$duU(+FZ~HBF%Ztj~ovecgqoP z%MQ3(=FC(UBGp@lDPWzAg48*cl_<2;N~;moN2YWrHs7i77#!kI+&yL1xLXasBo+kY zgC%PK6G8**g#{Ig=ZJzcn9FM|#;xGpoC6nUlgTUOD172 zDG

u>%?hgQ5<;pUAub?-{5{rUWrmV87>=^l9+hQ@JuCv)!FU+b6#?Uxts3E}bpG zPCo|rkE|u(wS{GdYB6GSPN~CsWI2Oi>5#nwVeF|&nY5rVt>#oapr{#aU`ew=tTo8l zVjx0+q!V_lq*;{vrya;9V%iKpsWCMy8E--qs#!u+Pqal@%SyK}p3)@5Hi+v_luPZh zyJ+CItGbL2_&F&14OyWhngDl8Pq-n?LSzK!`cn|^5u$?jY*XcRN7Le3rkL=U(iHgL zy3}QRCeduwGwqJO;PFV>MhSQs3B0*dt**#`qjN)&Hi23`2Udro&B(;`Hsir)x*+%G zsd5v((>iw{3ThH}zL2wr!)`JGVI=h_NZq2&J1P~EsXmpr6`V$EscBG(a}l*3LMhU2 zO;jGsl_+qsF{RbfBB|@7RSprXXm>24%|g;3ix*M}DQmW(%WY~`+$CU$;MwFYkYrNj z3D*(egg|4ji$ocV&2EfY%-)bpB6UxB!n))XS@WBX8f7_@HI{>tXsB*4f=t(_X2ni5 zSixh}F1)0-2=KCei3!%9g)f2l@jzWtiwIz`stm=n0_A{73PT`$MKc*&18jX|Y1{5^ z@WF-BDe!_WKHr)zXb-l3!QbVnI>8 z+|eZq+XDWMx0$-%92}Ag4p&iY*^5T5dB$UD)-r@>PB{i<7 z1Sm3Jhj0fy#TOY$UWy-UiyLC0#~{{vU?en_Rw9%JHw~8?$(+@i6m%=9C?wh{mdJ=Y+zxna z`ik773yBKCR>ah5`Yn)BMKzVwHlkoJ(v~B&mc~a#+6wR#MY2Jo8Z)WNkgo$=qdYor z)0Md(coIN3;M|8oS43 zB%4ie3ou05*;LD6q)KYAj6)J4Z`!Nis}+$`{ z(P?hClf^sY7 zY5}~f4H>6Ikuu0rA`n7!K(SUgi5N^i2I<-a2}e-m@#o7WUt8>I>0sBV|qa$^ppLmj^*8 zL&B%r47pVxg4JWpPmwuCEo)QpOF}g`=1VG+9_<6wu6WsPg@D2kBpVD{f{0KhRhhRc zma1A#2wW5KbB=@yvXLdcWf1UEKB14QhEv8mZ07RjipHgNb|~;#GPO0SjI14#L0YtC zxX|Q7e7{Cq@HIpsGRUv{Dn)Q>gls73jLS^gM0#YftOcS8lUKt}R_fNNDXS>hwx`m5 z%C3YMa(PMYNI=%KD4Fz==1K$h)?ly}2&1)yTQk5(V<@*1%@f zuF?DSkwB&0YG-QTS_=E{aszZn;K`RG(+-Wn?Miq9jXKD>en=DJg;4x*5&WN$pG*@an)GL6|GkjwwW}QSgq_%#f=S{(PPnw!uEt)=Yot1X3`|Dfc{lqPBaanj$z6l zwODGYYDI2_q!v^-B{BspDzVlf3Bb1|MMBzyDUe+JR8pVRg^~s`tZynSvW#A@^H!v0 zPr6!;dcm*2DfJ69X05@V6r?jcRfKdI!+FSb2||ESoy(J6*dFA~ZEu;Z*MnM$4=Pa~ z`2VCyi!^Bw>)qw4d_gY^)Z006(_d*rxH3e=8^J{^LuxDbsI;Zc2EC4uMD2#A2=8Ilk)AEnpXFEStwxP}P{(+I26Sa#Mwc`}6h$Eg#s{$vIjtzl<+3#=7YG6&$g1EL%MD~x zEHXKxa*NPVgbY9(NG@XbCgqf+E_bU8bx#}IfKnx?8-_erl)@%S3!X}LnLO+d_^qLe z0UT-skT=Cylz>4g;VlTG&ax`x3xHui)l}u9b+^kFYQ$@rc-k)plUl7wR%I#h2nQ7i z_!L$(!2}aq*32}6Nq4m)F1yr02rE${ClwI4k`0x-8kR!5w}K30O?H>wBZ=z5_O!*P zt%)>sQQC!gL^GFfkYMgBGz{T(&8ITDGvQVo)ZQ(TEDocg%3E9%DG(PUkPIv%sCYmV z7^{1lARS5UV0^3U@~t>%OglrVRw(3?>NHpi!$ON9t16e%QnHvRD-6DPDGmF)rqypL zYF#3Tf6zl-Fl7Q1#o*i$jFrWKQfQD3HH);beh3T!?W>tmYP=v)>cN)bjw=G-|EGr( zI)-@Jke-s8DZi@Y5Tz6f8R%Ld5u-N)rcv;0b_OfCDJx|B%=+{($QKZ)`cd#WWJCdP zXp_h4^vlc@5>&Wp@J7fpo-QR1Ot&l&DTDReWi!G7ffbaNDO8e})eQ{^8}PVdQb*Cyl&WJ&7$oU%#qjCXb+}ZBtPd8i@?ek zuT>TCfEg6lVwG0_xi=M%Q#Q;mDw1|*ER?fF+oCF>DYw2Lw7U%OZR;kj?V#r7%34bn zP}bO_a|$I5aT}X1qzVGh1!pG7#C+zmcXlPV8s zO`41nP{V^7*bBKg#8!h+kcf05A;<&g6{_+O_FbgZ;I~&5C}k8R6C$N`pFP`(Lx!7# z04$$P&_`1)i6oz-P#!30I}hR4aWExWG!~x&q-m6(O4*KMbilrP8oO-gZLrfLgf2Y!V~LO~Rg$!!{8zSZ0G^WSwfFvw9SXe zen@BlnR^Tn-KLE@EM~K`1r7mW5D1h?;KEjuMS(4aqJ=5& zP!}XBpb-!z9J!Jp0Y<`n*exg+3`y{RZ$rMSyrPwnLHI|x9xGYIRl6U2!VKWI7^#aX zC+xw+b`p{pLGo6SJrwnrbP;z^D2+gt3ax=^s+^t-cy38#BS~vTT4$~$3pA@vp&D63 z8sb{LJ{5vsM=NC85Evunn4(jPmHgp&-ctmZPrFnAO6YW3R?CAwlRg?&^GOQu3_%9e zxD%4j_9H2!f|?{!gXE8ie7?$$QVq}plS*I*CCLBLu8Cy5U`6o8uY;)L5Ia!*8mCyn zht!BBFbI_!V$jQ00_~103Q?-9EJUPPbrKKoen>AGbU{+G29?jxn#*Q1V60^%RWYdL z%Rad>2O*ngXGAH3kDP^6OHoz~zuoLK$Lz*7aC4~`)SZxOMX9P5%a8ytY6ydrR)vfj zN{NATS0WW*$CD^l!D$>aEC{uf7}A7*dv+qx_B));sc^uWl|+TIdfOHzNvRHKqnGJI z)PPJE{#Z0sucrhy*eciU@nX?#m5Cu&CqIx?Ccx)ZQB4@6VYSF#tcuh{RW6_qK$>Tx z%Vr1dcAYAh>otf!HyDllNG>N<@`EB-+nEebk=jPk?k`OFVJBMvww!X5OW|s`7MQB0 zgC(;fk!jkrkamC!d$L6rsEjIpi;W+#W)lXXzy*&4c@^x0of#0zkpil!Qt@*jwl^i< zv0mk;App8AY#Ut_$T8djB{H}}w|jP(rc9a5w{rqp6QrOdq~)H{gLb2BFoI77gjL!h zMhLQ8q=GuFT4H5V_bMWZ1X)%WgdP*b_qI(21AC>f9V%2O7_WDA( z)j@`sOc*xAwIGR*u*pQ^A0`8%v~I8iWWXrxvvoYsd$|dMne~uHxeba?*t9w|xlF2E zFzQHX#2f_!67pVxJQ3}dELfEaK(7SZBlHSE1AHyOsu2#8AOjY_M?e&;OCcFD<1?UA zxREm}82^P5Tb_bBmC5CDk=!p!=z|q~#R*%~QU{1X17A9aGp_YG& z-&7XqATw4fnn{Abz^sLcSIO8H$Uw0N?i4 z8Kwh{!5*>GRnS3D4&?JK8o|K`bhGv%W2$o)a>a5OF(ybxY}K@E$O$P`Z-LppP zA!t*^HiYNqVER0epIK!GIgC`J?1VtO3WYde&ZMUpk9fVjkutR$5Y!KT-!jk#BJY|! z1gpdoB|$torSwmM*52Gop=ev!F7Z1M7O9nL;NuYW=pjbgmyP({9sxKW*J}mj{hlPd z!DlI%LjsOtq&umIrJE5*^VzJKT57V9XsN2zdO%#3fmr5_fnQ|W+yvP$Xb(WnP8CQ! zXuqhen_;5{ALo1_BxvVlWCM9gx0@ZvRnf>;AmK73kMd=}(-n+E2!IH3z#)f`CWOjB z%x)FMa0($&v7N8SjMsLNVZzhZ-y`e)PZ%C;VXiS{Yq{uv|+eJEgF=rrCL3b5$NeRqq zaNUYD!2_nAH#+$c)1rkWgAy6Uiz)b?KwZ$xN}G^+$tzHZqk>=!5@eI0p_M`IK!GI@ zkf+qecHX52LXAQW>Xx-i`2~EjMAC+wWARQxLP-S5hN_ZHYuyS+aSnOtAY7m(@`l<< zU@W>^UnH|Mx>6dG-vQ)(b?f7ZZT)ooih`hF5Y(0l18KLb~ zsSDBNB6P{XWzz-MokCp<|9p&;b&t{Lp725s z!Jofv`ss_fXD!06>B7He8~sOux0ViF%DuK%#<_-Gbq5^tvYozjs76=U#;W z=JC`AsZHh^hu{I+5g9x-3)21!{^eOknajDKn^m;s3htJX_TbFY>=j%-=Vvp2j=hTerCCK|Mcg$49>$?i#Lb-5^vnCXTULz@!#|4n zu33Q2ym|z-@C)3l;H~b#Fsxpkzl}Ly=;R9iBnR8MonBSOPvC5)i5h;x@OC#kd-dlv zd@rjA9RYfvj@!m}ZXY2$bSOC|iV)tshFS*4n8DoLKAbyf=2H#)>{-CZo45-Vp|fFj zvfe(wM;M25Hq7*on)r#!uq(D7(h)D0ZJ%+(OaHw3Tf>93a&fPZfWx62@iJWbA9=#d zJ!edcPZJ8oQiV_{fjD*HSM>H__n-x$72DvyS?{c>A2(?)+^G zx0~K`5x$PYc?MCJi2m?9TwJmsh0JqW_vFR+=Ec~-LnDwM%_NZ3y)%HGxqm<)FZa4K z>sa_6&a~)q{N2@mp0!X?|Jbx!`0sVQ&-t-{^UvYxe&r_vsDIv0KXwJ)#QwaUVbO3K zyb||~j{>5fy%HxmN73PL^jO49Acm`O>ktTYdhhqH!fQ~}<@Mix0eC&&icU2?c0XB} z4Cf(QIbNQOHzB8EHokeX5uXf0#`6~aqM5h$2t<38Vm4i!tQIDhR}1vYOSsizcYI_; zDZ3);aHx;2IXsDZXf)li< zj9&={Kpl9DE82N*IN%FX(mdx-O`moR_q`QZbjKm_nsax|$!p&F*I?4m>UHUVpVw4p zGY$yqR?388$lE7_Z0@sohx{~H{|d283a0?5- zN&#vcNO=a^8nH+!T9|1Zy?~5Vym;wAMq-UmW5=8_(pfv^;~E=vaH{X_Te(X5rCYh8 z@e|LwE_Kpb*THFrsparp0+EIO{e@gUZTu?NGj=B${PooB-O20Lu>F-EUld#D@D8Gm-%7;RXa8PVfU-??QR~UhezYwjPv!522_=ps*f}Kq=hEeG>Eb zpnT$!gXwdPKw&*(LzzBsKX)Am%l1HELlEf^5Uhu65X_+b`T$o&-+TZ@=CeIei-({t z7y-q4I0EYAhk!^YA7bU00fF^k1j32m z;GRzZ>cB)7N-xM!EY z9(a;VF1r?u>w4-7^v1K69oJ>AYlrPzwTnLODK0g3bN8*UbTQenj{U}fVT>LWIk%`s zuY8*OHs?X6#f`($ZmB7>Mfbr4oqMZ;U%>1BT{FJU6For>vkq9uy}PH3?RO^i*`-pM`Q4d2!Q@lXc;cOst4k*Ha&Me%Z+e2 ze?6q0zCW|4wq8$8DrYlh{1-cgc6SyT-&M zLkvQaSx-B}PY=dq!D*SB{HC`*58p_RmcHO<{J4*zKa!k5fBlVy`qu*t;M0scH4pmE z7r5H7NEiKcyXe!8PT(Jpo@nBG+_B%y?(a=VVco}^km*nTfcr%b=bMK# zBTvnok#zsu+RqMVjG0h*wJCZx1JwtUT{#&ecU8_ya0{NK%cmva~jVe76~ zm>e#R^p*Cku@tE1+>owOFT@$9SA>=qIpfjZJoM#E5EXWxv0Q*<)dNxC!M_h?Evlv$ zpSj#USFOyM%f;CF^VRz0nafX^tJX1RE#HJ)H(#ww&swgYtJdphE#HjYHD9e;zP7G6{w*26Ib6BAcS8(bRG^@Yw%}vlhS&*RV z%@eqH*|8di>#e5`d=?k&c*QUn!v*JYZpn%M?+N_3m_Tz#EN`=B2FvS9?r-G>7;Mrk zD7ink7B4Qdb`~Xw4H)qhL^A&eXL2megpn_F80KsP2$&QPGYLR ztHLMyykz?SWD=)8I0-+7^FN)g&zvsJ*6yCFKd`A&Q~cB>%O?(N*XrAZ zujODL9NGZ?_sjwAuM)9%Fu;HL9}Vy!Ng^|u_0+K<;YV<5)AX9ncvfLNxTY(M%qgvY z+(pLf`wH!xlE`0X&f2p6tf8IodU`sBTNZ&5rmLlS_w)D_V|p#a#kE?7^>)=X>fm6A zN$vDQSK}=Vb8B~l;+o^8UDx9Kdw0waSXh+1@he$H$Xxud-T3vhie9)5-!r3Vg8s_& z_>D7)R@0>$@H7-L3!0~8Bf==Er#I8MpM%j|Y-swfFXDIi?%;>!+{c-quh@gznHx~C z@T_bQjQyK`BmPGY_6T(N!zCxsJNM$b@kh`g0J|Nqo&EGT_ToFS?{=H{E`9u$aS6H; zU3dzXdV;>-%ed%>dqD~28bW`Be&);g);?U0mNBF&z5<5yUQ9KiAFdp!1+E(F@5Q%Z=jqtW*qIeCzZXAcq#{Vy z-@X^OVPBZ9Ip2O5aK?jMu=%4~`BW`VwNVGyORN`G!!#;X6nl@i2ZdcCb59 z2N_DaYk13G_X8ep`|44!o9M}};YZQiR`4X)V*Oxl*uZu4ffYRYyeJ*X)EY&}IgPIe z9OzqC@=ip!AnJc@CGYwkUYvvUxsTwt&%$!Yqd>RAeu&0-x3O+XFOc!d)x6W^MaDm0 z7yjRB-sZV#oif2QV43-99hl&qGHjpe!t0OYxyjk%NR5GZ0vy23yY7_K66%=~=$qE? zj>az2v#ECGkTQFxQ84s}Pv9qGH!e`|%WHWj0ZLu+VZrQ?&l}&s-{h=*NI!e^ZMxxU zd=35XQ#hZ#_yzok-RIOdFdjy=Ar(EOr$2ubZwK~}9x)uS7C!zo{zC5#{h$ea>KXjW zSw+8J&pW^>y5;o=`lrY6w5%djaq?UE)3b^k8+d0xQMV^7Ji0x(^Pd}d*f?RB-4S}# zMwqcRh5@ObH|zmnpl{sB`wT`HP!HgT-f=wduHh|oVIBR4@8UG8NCdXoCttuXVHK?s zOwgV0!MAA$hhdLvVb4wh#S zdnsQx93OU7Lsw$%Hjl7CQ}dZ0 z!pC}Mk@ZKg3}+T?_%YrAgi*KFu|Sf&Zzl+Xm(lP17=H-+HZ-ztiTL=T0S1{}F2!LY zH<{h%Pxj^G-hlk`CwPW4esIui{(jf4_+@ zr+@HEeCh6A-nC(M&ohl(pMKy~@IpJ-orr_*gVm9aGL>U z-Oq4G4-EQ2rv96M2A}9IWb}i^IP-J-rddUrU*I`lf()IoAohj@dmX=-gK3NmA;S-O z`Q*69II;T0MC)j+O!%cF63bEGRuPVF8v#V*x$Q@A9~|;ODE_b@CfPnyR()G z@DsF>F|G;mlRiJVWZ1yF^o1*kopaUu;R-^89X)?N-AaOF;lPQ3jQ7+^VlzUBG4p=m zP68)bOX@OOb_DT#Y@2BTK6&ji8;5+d>!TB2MbP6`(@Yis{aP_jD6lC8zlk27yHzu& zZ8dfIlsQAeV;%p3BX7)8KWMCGbnIP_51Ut;`9+X}~ zTsn3O>eG7aCT2V~Fndtq&S4MsP6yW#S8}lZCd3Ol`|0IJ5_k6Q&=2BjzdMq+aYoT< z`Wr_PJE5qH!#yl;P=qMSaWqlpUxPJr!z3*7! z7h@lEk^h^i>yh!$=xdK|Vb8nA$4ohis*+#;-oC5~UJBW`oYVR=5$<-gE2IRQ4Hx7r$)n_r? z1=t)_0R{>h!hSF!OY zd#|#zCRf};_^|Joci%F#iSq=_+ebK8AvMXjhwz_-&ayVpaLe!r)*FwT*{)Em2agU> z>v|AwzL)Tf{l43k-*vmfdw$r|uNgpnS0>hjN6m~Y(*$)l^l&D6d+3eN5UZyx`-#&r zY>8zz+C?m}(EIKuKHa2HE11{9w1J_CM|>A$qPe-01%_E z?!xT@8h;SBDa#%tyt5j({y}2TtfJ&YBSor*iBBR7(I^tEwwR7gOh5225#nHHG9BTZ zNejLPpUc4=^y7l@Ge^5{*HiHEY7D#Z!(kWBw9pUyi#Q%&FfE?SdbDRaJ;#EJ*i6cy z?x(@e@>u%jM~DwF+S28f2ZwO4v!F@u&lT&zHOzz6J*P9)gKh`ubH7gfWBkFP_@{PY_Z1s3(Z$u^(D` zw7udRgo2^%8GsR8ol3BZdRX544Wcmqn^7#0FFzgsCc$jSnSI{Rhj{*mzT-)PkNu`M zxvVEWs3$x{6vqkc2njOtx5hdk!Pkc{2`jyQ>ylIEs^xi_5MZ0;tHoZl0E8ho7Ti74 zux_**Ol@VBjM_?H@l8TJt{z41m2W~bUZxplV%``c*MFOtm@hm_oKD|(DR()2&9lU| zAz8#f9(;2~hgc7K9lGrW;u;S2wsrT@Z=OJhz7L$E`#;CZ zY;wsLiE9TXUClf1CBhCxUG#r9M1h6==trgJ4*tWJ2m#1;kN%w4$~ltW_!hBgaq5XH zkEVb55^?le#)Th!N!}X*ov`ix!OcgGg%&Y0Ibma3C0H-4qHp=>a3)y~R&*cqxIgOw z)A7~Q=f6zkIrOs!VOd_iZi4xTKK_Tq&4VIrhmDas|M^4W##x1{>6d;)>=`M%<;O&Z zDa7CtvGFH_n3F+$+`Nhjc6@)>s~vRs9pb3PNWV-!eUMm9uUxn2_)}Rg9LiX3J_u`Z zKV3RVcsQSD#u8aA-aAML8LP$LhRt1$hTh25uqDy~9Y=r;2)Azv;i5|!gen#9g{+exs z$uWfTE7l->IIQ{_E&By=0DG+ml=Wl}ed6oHwLF$YbeWlc;O9$Er2pUR#J8|zc1C&8 zHG?g_b|6z3}d9Gf( z`Jcp24z)l<_IG2Q9T5fwgL3CP#N~&G)x`?{M+(408zOoED2`tfA%qnAnCxIp-jd#A z&$(Ng$B*Tx-w=0U|IwwQkMaP1&^}|c)c0d~<&go|-EZ%17d(6T=hry}0)`n$wA3~zYrmxI<1(zf5jaz5A{86NA& zkL+v_o%P`5;gEGj^#Aw+anI_-j+v{56%C2;90M`s?P1f49rVBcNIZqXFPaq|>k-Gn zZkzRBcH7fu{)u>%g9#m7Ve_3K^eqgCzOc!9@UorR7#!D+abWeVKjuITm(h_W*snMi z`j&qalDW0q?=L;Ruf16U!w2dHTKQMvjqx*PjRa_KPW>;C@nR0PRth_9Q2Pw}hW{eA zV`n(f%Hf;qceFB@p&YvI3^Z!kpF;~4e*I-+)-Aui{?ugyGgLbaPq;pj) zg*klej`PuLjI>;Fw$j4gI9pvn_} zbpUgTow4gL`VjWd*n|^JD*Q};kHd3v;GZtBtG~;7pFU#|@2fo4r>OtE5-jFbF$L3Z zS7pXC4(mys(?4Czdya$YyKU&5^s`HO_w;VzgYo9kya{^pcKn)IMT|cHyv?jSGaDd- zc-J;vbk1JSG-Ix6I0tLlF|1m2(yJZ@14EHPg;S*eK=AG!-Z7Xkw&_u@Y=hR98Q_e_ z%mz+Bu!5%;AFbNO=)im!p?x@@f*Ri~^sVas+ci%Q5M@xi>j>U8v#@1s)HB~Y;_EP! zGmBQN=53u9vG?Zb%{!}kTjr|7Sg9Ag(D=?V1QWbXvl?fN)U!urZX@+Z7h40#9$P1) zIX-S-Fnj4;YkBK22?M3CS7$?^_parg4Cr-fiuFY5WN4~uu(r5n)6}Mc!5Z^29cRqd zp&?;eTwQau#l@^6xP9Vj{JYrTmJv7^d-cpB#$G)v8rZ9cMMzR*?A0@i7<)A|!z>OK zBHfO3?bWx9^rLI9-ankX$gm5y(R()XI@oQl-M2h-0=@lsSQy>^IWEAyXR02Q3@p`y zl7XSRU($8-VeHiXqM@03P=fl0tklpL{inY!5fMX-dXsYb$-JvM<4fJMCxt$36K^Y+ zLl1~}U3+78SaqqJ-nWS-o~zdTn|Mmj3G>y;p2|COq!xS{Uptkzg`+^VhPg#%wg<^p zj-tm-;T=VPiqDgCmeQ*fyb~}$s%m#sQDFRg3-3t!uUmL0t{xr6VK#^sGbm!DqM8TLm7Gn({0Y=2RnWwG3cARM ztE@WG9#=jbaaAwi{SEsfIQ8vYvf-n|)$84JimUEF4Zc>_x@U;1V?A+I>g)zgo*?2~ zvg~V!q}Edpqi&1njE09xJ_h2`lVaXU9PB%XhWWSV#=N}qguamW%zT(nuafe%V4SxZ z!h)ZuoVUMshrR7)IE+cbyM6raSt|vom&uItAww$z$AJqIgC^gerj$Gr=vTTk$zwf& z0b@kQC8zTq!;bVY-0esY{m$vU^7xUSSxa(1O>G5#m{UDH#pExx@DsF1sL0r=&Rw7)W2*uH3^}u4ejKVe zXD9Eo9IWFxBzpcdcujnddmI*8L5sF6THW7%U3=AFidvop>G?Y(FMZsXM#gG%gO1(C z^uPacy!GKh3tyt|Rr9XGzSNr#)|1|ZOkbhlt>lb9KGF((uNJUjHycccTPKcAlWNhs=)0;_7jZ2d@{riKBw`nd|On+<)tPQ>X%z)Kx zoZ;|q*r7KE?L6Hq&{GcH_IW|47eI5RlNVyaLaDgE>g4TZ6(P)bV;DwD7u_636}u6c+ka(F6Ti8FRG0$$8CknG3Zi5w`Fv>2~w$0d(q|c zU2wS+U3P%~7`0W~f$JU#3?A|{ p=?)&U_jd;m8HT%qhxEtY!9&|vX7DKBwe$;T@ZP7tbtW(Q{{gCTOqBot delta 183627 zcmcG$dEDz(btsCQbCR5sf#l>QK*n! z{UFqw!PqfQt-RC?(DBv%sTZtodpUH?>a9mktiGAL6hc578 z1vvWP=!Vft)=w8NDx^mI>N5}AvHsW(FFtn3eMd&2J$Z12{@wL29C4dEBWQHd=mjHq zbf)sGs5iRqKIB5ZpE}sdg(|HIM_zcmHab(=!c{v+?~Tx*;)&JoANlEPkB@D4>VrY{ z^;6T~sVLYQ>(3s;t~?F~ekz>0cMqnCb87YSda*~18$TqyJPi{ z=UoYYz35`<_^GiIPEUm+{rKv4FVEb3s;3)c#{mdmcnYx54^JiIaC9mFJx)NC-@o|I zI{;)d0iEibR;R>yIGQ@ggXu6#G+#gEYJE#H^i!*wF2PS62gv>DAOdv8nm?XK!AiUQ zrj>iuv#&h9X*|$E?bL?k)%z~NPaf}UfgYR+qWyN~R~bCKO~a1L*}_5h;8Z}yQ~b1o8M zcji~W@cQGa`}T^@*ofH($+deza-ip}H&^bz0qbIm$>-@=+KASfXB4dih_`Cr`pmVT zp|g87LVD_aLVBQ&txC^3Z>3$-;<+5W`Wz`gfbu*qH>ma{ZuS3O*@Zs0Q7fzNvym%5 zGdfef<<2cF0;;XWJ1;t&+UDT%&+YjqF1V0xIzYmOYwgeRc|`T4=Pbp8qI&MzePnl> zOKJGOTEw>;Aozb-dB3akh}DbswZXwfe*pZ9&3}Affvi4q_hrz5@_swWU0?mF)GeEt zIeqe#7n#q7R*$^(g;!qzJ?mMA=5K}l{I#p!J9!BN>P>!Zb@bXBpck!9U;FPsLEUoQ zuR~{6*xk2*zp3YZbno{g#L3m4K4*2P{9m5?)d4)nYF z;SHyO_PgoEjrRM%^&3t2E7#wiIt5hW_0IzZ?WQv~oSH`dslhntM?o;2I>D)l7V7>B zfm%Jj!fpod-5s?&-&EY}6}zhV_zJq^=GDVDTne4ubaarhTYwE~%$IG3bMj>yCT#75 zdkXa>`(zz#c>uiM9(vgdumF1L0qPI7IS4_5s z`p0|TdluwNKYljk{Xg-M{W|N@H>XanUi;$L0npZRzO35a>u!s`%1bV~DD{QZ>IYxH zZM9@Sw(}E8t>50h;^b5%e#sPux#OV zN(Ug=9T%Nk<+(S2Ry#Dl>M)Jhm*+=Qw_LQvmRnYztbK65aQYP|R}a?Tei-ly@u9P;qvNhO=G9RwYOZft6uiZs?VNTJt944^%e<%daFm;FSzKd_uaDkEB=;~ z4@RTAMw=bg+$H#yi`J)(Ui_S6_dze*Xsm6bPo59w%4;#uf8L;jzPkEW>(*zV2hVda zjW^T)JS#BJ_3xg%;_CC`8nr2CwS()o+t;6i>)uPCJ5KCVv7yYs8} zT|EjzH+T_(Sh}wTKQ^8s=NaDW3Hi=LtXy5Kyy-CD50tZ*`&j4x!!_*g8xI5SR@a9C z|E8WbgJ1qntHU+cFFOm+fBCx))|i9Bjx_B}&cj<0|L3pxi&W~SBO7S~kG1z3scTkO z_86cb#oi)?9ogDxkt2ZbXR7z^&g)I9WBuw*D^RmBIJtUj|7Q;ZuoE|J+8o%JOmy|& zKu;+b!bs`HK?EmPf2_azFyI%4_wE3myN`TVEKL;jcV_irZ4oyfgrlikS! z7o~m=u=7(#k;|(O?7sNzBkRV~NA9@sOWXY1ga6hJucDs$PNfn-?7{NU$@(z*dbyM{##&B_ug^hH_xJNci^|i zcS5P_k8TMHcubQAL1}B`*Hs_f#r^P(H3A0m`SsgBpSlLx9?z4D+5ZC_d(qLY%^#ckVVLS5N=cox5DwgQhptfdp-FCuhg-?(F9zGt_K-Bz0z!Rp~x zzk9WQ^`oibHXAOh?p5n=D7kuv!F$m3>X+`lbC1`OYi&5!+?wL ze-8A8)y?<67`kKi@LTV``uP`cX{`g@y!ZaXiE~K2ef78Z-wDwdXwrEAzV+ND0P~rh zu$6faD6F6do(l;VK=H2+wa4N;wRQVJRClfJd)p0QS3N)V-+Jp?Z+T`?w%dmn zX#YRncK3gDC~vp`iu>Mv`ag=|T^B%c$2*W6Za_d9fEo9|JIMdgu%5gC!UGEcNaV&# zyS%zL=TCt?sB4~leD#~D)R!*avYEcLdgnVI-7maqy<6D8HUz!+UGLp_y$d< z`1HF!vR`=Hdmh`byMFp4uoAuZKDZCsM)`thH=cHh+c#4fhemuyU}=N{WJkxpN-HP+Ls zXFvRr)Lwyk{N(EI9{$W>z+XMGzUA!x&2z;v)xBh+5>M=`sskObo__j9;HYkpBOh$v z`pmd?S4(PoA#`g9Ir`jPl?H~jnbFItd;LAN>VD|nEB2JvzJ@#Y=u0kLZTNVD`Sz9g z=$*%ZZI?F>JbCYez?1 z^R@ZO4+p8=I5d*|HkTY4$p^Maa(Q*H!ULCVn-uT=`42;>Z*CL!&6V|$C-w`><5^JH zk(|%&Qu6@l5D4`b9{)I$`nT!lh0S9_|7Sk^ z#Qy6~f8pBI^0A2*iB>VJGLo%;F~!IPbR_YZejx!Yjt zpaMD=B?ko{{k!k?Oq)ZC`N@soU-I$?!GQ@>-H)S;E(R^Xzs;eYCBEO`>Mwu8JB#vn z{^nz6LE_JU>@3JDeyemAYda9=L#t1J;miAlU;5(joK^P|U-}$?>`ccVEjzOR z#V`L2lqwus|KRPH+_w7G?|g3eH*{=kGQR%1KROF?;`je-1KHjlzProH-6dH0$~Dk2 z@7R{|m>%0PPycMM*6NY3zHDEeJiNN(4+g0h-?@I``gkgpTKM`S$M;^}|Ktz9_G3`$iDO%f z6)Mvm@_oJZywA;rMt0RB(`W+mb$3`|k`$q`)xik2^ zn)id3K#eee{EMm7cenaH@!hj~yM1;0H||XRV7sdwx%lC3SKnQ|`Gt#9h*YxCuA!YlGyKenrS4|GEQH)Myt&0{m3pZ#xz^PpHi{qIK@2)|Ij z;kU4TG}}hzfB)9Ic0echN4{^Gtv>eOU$T1S>GRQa_qYG^KC&&Ep8oc|=cDP0?_GX3 zG`$c_-+uZgfMz=`byh9cA-PNnU^V%)k zTL#e{J9j1>|L(7+u6ykGd1UL%>L0JZ6A}Ej~!pXw0ha~ z&|@p+FKz>W|M+_!+y0$;?D!TZKK-YaYtM^%b<>|o(ATz!-&q>_SXYn#S>Z6A7yLPV z;v77;uQq~``^&vaIe_6Se?HyAu>Q)!mz-QJ{^Gt1pn$GeUHvzEKEeSculdW(puo^} zhx4t!R4y=_(qB>FpkRaM+|m5kzbYR_^Aq0>&PU2;zYnjj`kV8O=7oRF9LDf}{q=3< z!|;v2CV|Uv{=wA$#=7NP79Mmt{%-w5=jaVkbO8kT56&XuqdzE|dS;BfQu?+H%3X7D zpTWmLl;QSwTzmBR9_aqn!+-m-)T5WJnO{z&uOF$S{{#Yw<_9kW!g;28Yi&pDb_Cb` zyG-iD$2ac|D#hDZpZo_5`t^f@NJl#jKfb!_AMZ_l{IV^j`tjB8{$u6ZbK37JRp{$m zD0Zgv0E%Dyr`$e@hyV4q>&}bf%^R-yV6!*IsB`;g6)l+<6enJoFOi9su0hCm;FyVCjSx^e`Pb`WeHV4~K4bOcIvKqvQc?(!R8eZx_RN_}k$ zD{u_YtWO_-p0j=n1f95MFG_u&hr2F@e&Vbi-hVMPKDCc}muUwz!_GQR{p}$Z9zgr* zOQ1V%d1f3t+U3al-pipqWqSa{sY?OJx0we$x_&h%U3(6~E$R9l@Cv$R8_kZWA3&2o z2Aw{K`R48GYLCyxe@yJ^)K86{prEN z)9c6KORh}0SAg)`%||wRaQ*F5(DlchZN%0U>&D-wZeE{04c&axcmq3AW#6}>^EchS<}~!W)Pq;7-&RYVUZ1%YdiSQ_{mW(7tbhD{V6n}y_4Iwvjq5vZ zgYI8iqxPiGkeh^4#|OGP>dQSVqQqf1$d&m~xl!%1<4ieI=DKLE!A86+j1sxs&YO|s z=L$2XG1eFZpVr4Ew%s5I6W1paQ5fY}E9v&B4m)F*e!I~vqgB@^lp3vEkL7c$G0LNk zBXa3w(2)n&K$h`=+=-^~02xwrtA)!QIh2QC+E?JPm%-@JYqR<4qLRtfLnbs-yXW@D zOj4Z(1e=r>0l_cId1g^r7#w5P7lu|sTn0IZ!^5XCwYhAjM1V~!AfLh zu&gz+8QW`y1+nLMlN>yObLmCXA2&+5m^D~lMon*mNPV{+Sa_q&@MX%>715o`MFGiG zt?ax`8oEUCMM|Y5Q?ZbwIqz`IenO35$>bL#+iwlaBP*_X#csqnFBubooE~kVFPZM=rGEddV?)?;Q>JC@T}hIFo+Rx z{=|wKeS@XgbR{0wI$aBv>yn2Wuq5&(s9)ppi7HnzSxF^XDlAoIvkh^!G)R_3Vq!O( zdCO@IC!s)D8Hr7BO-Sg1L)&?`53^*6u#lOFu@k+8)Jw*&H|Wt~U|A+yp8H;oEDiX< zq!sl^!D;F;ZPgpuQfyNK8}xJana+q<(;ZgpRv2JZ9_GMga0#0*YQlEeWu2WeLcrJz zZg^$iiAM{jR&FR%ufP;tj+#(~SmOAY;rk4n&oN$uiCB)o=Bi$!rTiK&hJK;UbZD8IdRuWv_x{19_@fbkK6n4wYn?Wlls zc!cKqoFP)6X{In7RuzV&9KCI#Lb}h|%*2orhAYF1VbUdX{o)u!nka)X0#nS0^PwRy zEp$>hvprg9ana0n;%wqdGMQLfvsJUo8kj(i5%XT<9{j zoEVdK-GssNtxk|eQIhCVG#m>Io2X)kC>S}K*C)O-T_%PMEBT=nH_Kr?Z8F1|gpOg8 zijz`*KzQxn84YjffmoPFLup z%v9#h0bj9(NvX}KW2V5TgM8kQN?xu^wL+nu%i)$8;3zl^F&T}slmLd`@)zeP?^+&62($-8Vm9UV}t`fPc?Z-QJQ^P zH0;q#j^`31_CyPj)N)HfI@Kvl@)45{>begvhmD>y$2qQ%!G=9Kl4-n=*K4qe;YQPF z`Yi+}s0PdvKm|k!o596#ExYKTMpQ-|wh~4vkjfyj2c5tWn!~9_g=K2e>&{@mmDGi5 zwVx{`jMSmHer}R$l*YxZCbXLct7BAkU#u^wCc^}imX;w}DvLJNjuQq`lgSg5#>XZ< zM9Gpt5lp5yi6>aw*66&KZwjSJCR2mGnp5sJ^s36GaWX-Lt|N^-jNoZTswwSkQ3%zB zPEQc3PjhZhLflFyVm{wYS5crm>!=VjOR}O*IBD#HzEP*#>DNbrUu#e4*{l+cm{L4P zIgT+ZX4H!K0_8*<5vAuuInw(RVX&yyr4|Q>_6Z;?oqC%rCIw=wx8S_cwe_aSk2|tH zm@Rl;a-%M)cKwLT)p^m9jL@XMsRR~9;s28QOa7QH+BZ5hahy0>&1-E z$Y5#DTls}Lw;9x)ceI3~O7^ml3>ykV(#3AYHi!;7oNArMn5`-_lb@H-x(AnX9L+N9 zkP+D_T^q%PhBndAtUDPt%1M;R)LAZv&8v(!brMde8BZ{VO>~5C4bvd&qalrUh%8UB z{k+1UX|G?$8P6#7S&_*b&E>SkQfOF$YeYDO4FSbQ{b^(i<4k?Vri0vEt+HIsaTB2> zbYsWlrbU%?OZ~>MQseW{IMdS{CTn=v4zG%1dEOig{Ak8X1rcRxR_sYiC9B$4z*Q?` zu|feg*<&VP50Qi#Gnv5r(k*M_MvXDGPA4cWs*7M&oN0jxOqZOL85O9pD6~VcLIlm9 zJEB!aVK1QvYA>4=vspX6kScXm(cLJ5(`iht*e0wZRx&Nr(MiUvx6%wpY=dXF z2&^JQou~<5zO5R|j?*Zyw(KxR=YD-egu5@|Gwrl6jrL1GGUDilcetyQP5-t9-x0CVXYl&lP}SEif>U%$%tA z6lL1a~Y+*6`TGsD)XLZ8tXp{M8#V2aFJ z8EL46ir@C83Yfy86BuRAA`D)UdR~l+1zX@nxtwt=c-Zm#vlcMdQ9}~DWt?%AMrYpb zGabe;7E+_bXW2n@QS%%1IoqgJRLjwn8d#|oP_F{)Oo1@@K$vP&V=~#u#{@QbRTW^S z&WMtdZGi>k^$BFx5vVR0zi#w*9D(YhnhP@Qp3_iGDvRZm`;S-tKTvnM*iiAmi?}Xy z(n(x(mP29{8n}?f=IM6aFE4Tn zJ%@GM46)4d%>W<8@uJM=Gza{ZYC59lfjgUZeQ$v~1ksj+VmXw0>0vQvF&>TgG=Vh> zz=ZY|DnIMVnVNztX>?+lSu2hw(iv>jr&=HaMa0UmglO0CCqB=fSz_j z&n@?=!JL|CwXmBDWp(CvI!G>WH0+YBwnA+lG`rnuTD52bW*DNV;`~%4a(wAp?XjSQwZNmn5e$17zY2yp^8A=#Vk`Od$AaIuSUo<{%@>aTZAcJjF>b3i%Qr zbcWMLiBU!vFHVaZj(6*HZBm}m20;%8jUewUHtfm-?CF?Y(ZQhQTHMqSc8LdJcLpK| zb-tnDzGLSLdRFLE`x29k(Ryg~a6xdYji5T4mGe2l$P8#@6sUu+Kw3U9yqO6^&bw#|qj6p`!I4Hxi!S}rbBYS81=QId3Xmf2@@X2Bqd z0!+`m2|S!aHxEl=xEwe|%hq#~sfKoYF5B*9np~C5=2~vKIvKXJQcV?>(FAXJ22&dg zE?jE@m$_fUc?spfBA0o^FEy)Oe&YFPeWnybVG7q|_nbN|l^B*d0QTi%+)76hqoDx| zEUuD=;?mg6jD#Gbk!99bFrpGUlVYogmn>PUc3Vp-Q>aTtiXJO!r{oBJRa1-JK%UNg z5%nv0DLotxwWc}9k7t8QAt#p!vC4pn1ooO_x|FNkiGdC{jjNQVNj1|0)*6ts#w-)V zGV6$#FsBzvSo8Y5#%CgXPS$+Ue*?Hec&*og&T0o@uF&YjRFzGzEPwdEh~>`rAa4ToAB9m zt5?j=y1iVwgE4I1YV2E(H%{vNy}MEVy`-G4JQGUEpUum$y&l(C%K9=N5^2> z=%@Wgr{5mKiknB`vW1hqveH$Nd{LYj{f<5$MxKONE{GxIyG)kU(74t^bRpjr>Vrx* zk>re4;0rVm?@Esy0+TaQc)o?284YP=S!2oKWo+Rts`JX+#Ot$hJ8@iLT4FffZh)|g z!wQrB(6->jO!o#JrcJH!tQ`v^Z7Awcsp{Tr+V5IbJ|ihP#)uke#sT|FRB2(x4DN)5 zkz{5hJu|2~Jf;LXXIGq(GB9Z_uosmHyo}M1E71JL+TqGUgXN}RZO-|gqUBs_q++qv zUG#c-$IjYFgDiNhZ~WYA`sS zB!Myz`U6YLDwA|S#O9$GxpG0mm;SIkuXo&ehu~W|6)XCqL00Mw5<=@|`tGDVq?_ed ze^_nT+wf@7^C+%TbGg!B*3vPGvK+(nBBl<*fjCuLRMbZpI95h7$rjSQn3o-PHo&l& zAxd?M^jw83RRe9VHEYRihEIysB~q)AdIuR1+2Y8+^Nu;2X&i#lXf3W6YSX!c73xG& z5fVxql%#e&AnI@$)wzJ5Vs_xuElq^0RiX@>xl$TuhrOOawy@DCVw_Rdh{ZZ%r->W} z=oWm-LX0-sB*WG?t}EW6l(Uxok?DtOrz}*hJU`$}r0!`52s+PUxKhJ19wLQ3bi#2` zrv}t%G^fWiG|pzF8G=M&E6<@*V6o3Ly>`o}kgZ9hNOXu{S)P}tZbpkltmW4|pfb{2 zJfEw1t)j^!9D58&5`qb*iDefxCzDbJ?6K(tle=LWZ!E}J$--kV9t5Ff&hruG`28S3 zBs9tb^^(pFqQG_Ik}A`o?aPs`d-Ylo`!IT7C^6{R~ z$+2YugbV7xrc=|MR&OLP)ya6KbCoWacZtZZ%d}n8Of_a_HAnU|DxQ(ND9Q4GSIK0= zb#hFN1HMK}#_?h?v%n*ZqB5Loh9J()pxIVIsoR$ejej0yjg%@1i|2gJv>WjG@@F=)=_-Io5KGhF;DxMyBHov(h+h$ShLxm9Uey z&=8|EH4|5}eHg@_=^4)9`ATKb;fQX2MB_z;4T?Th%FUE!WnN=OT9Xg(Jc;uZ1p;$D zuD=|0K=Uqsh1|MI|U|pMX1!hVZ(|$;t=%_T1F&yL4{n$ zG!hO6Y#0b&{3=kuI?Tto+d||8Ox4_4U1?E$YRHXi&3H>*+5$ySNZ%p(g0xnHEl z<;=X?t@$%$m{E(-#1zQ@XR^J1zKDVNJXfs-wJ=J=UXQ^=XA~)lgQH5lUMLWv0M0h- zEU*abNPXrklUyIk^%ibxSST@CKHUy^rmq5nGL3c00oijWwR{4%TLiios@x)@*cl8q z#hOmIG1>REXowC5ZeB-}tm|R0-1A4Q9{1F+5cK3iOojuSpTxk6?^a6iOyCsRapHym zTz?gD9cx+1@IhMYAT-!*V0a)jn`jgdOC1vFG}Dra(x_)Uk~ub6jlpaH zceG5Y0XwMP;M+D@p7_m%mB~&9>8ZfA#PNbJ1tnMR%YL@rkercOwzJcSsud$%720EO z-1W7k+T6^c%ffuk28ygl_M@o4WXRFT4Qadr~V~9BwK4% z%MR;$wG39SWSfF!@KZm_+u&?XCWo#1qSen2fu^#BEF>$q2HdHsHC6iwP< z#ar5gW<$*cLU&R0XFaQs?KI#ds1&h&zbx5tAzjS%YDrC)BpDD3bZMICaI5+ z=RiNlT18$+o%D>NOusg64hKS&jgKBf zG?#Xi;kY`F&Do;eF_m_$P_5DV?!3TS>1kb-CsAn2&D1Q~GyyEljVu`fhi zNjB?jQn5#|SO{#SH^Yjxd>;X7wL)=Btda*O`^1;*6LWhJvYWeK_FLPR;g8A<{s5iCv$7B{iE7 z$`FQ$Y$(WasLK<-!qrN)Gy?%czDDMN9vr0oGT*Z@=pqsnL}ez)Vu6nWnP2({IWn7J zPYYpxJ}9g6XwfM*&3FKI5S+G}j#&9@ zR?)4nM;7eXcqWe3B;D%xPCe6(hiyB9#6C(!xFc(#Q-)h3eb&kAXkKxNC0!D`ypPp^ zhLT7-KXd|IU-;QXSA-ed1>QeKn?a^wM+h}u3}m3|{UJ(#&BhcO$j~Ffku zXTU?G*9J{Qd!eFipA19G#^XY#HRt+vo`!w9mH~DSZMW61#!=Crxt#df<_rljCj%TO zfk{UTaO@f7QP~bksMc$zeUy!u3OF%g(oD-}Hy9@`_K`>)jgu&f`BGdfMRq498!)|a zTwkFEb{HTC$GSvYaOo`EsoAq~lx-?=JV#P#4VGy|Nk&FicBh$TZQ#sCuIaJPIGi>T ztrw?dk(m*$(pjjH9YvXjAdpgB#IhQ2h38_Fomyf~lNLS+Y!gl-^HDUXvKqp7DbiK3 z%t&F?1zNJKncv5)Mv<)*$G)c6?I}9u2ma9Lv;Ht0(&=4~JXDaoWQ}7s2(&ve1kq_+d4Q2YQHT zVZV!`xmtzb98DZWhfIDS1D|XlLibJ5k)8j?QGc;GubcUR1k^`M0{W+etaoQL4 zakiy+%p%>+3@YFx5zCr2F_MbSMS*SjSwYk-Frb3#kV{~*vP8vgXFB=36-8kS?)xmY zC{Afb9uO*AAUf4qz74~?O%$rC9A?4U4o;(CWrAAmj?_dbqMBERksRWA(*+L4d_u>7 z0($O>e0o@t2kdAn%^FF;aWks#>BEVWr8U?Nf(-D#;&xiARUE3`o~vy;8#=gcht(j; zbkMrlB)HPh7>8(*1&8Hx-l*2bO1Z=Ywmg&tFC2EWb+19wK=_an=b~0L(cvuLvO|Pt z<*_sNMUhc3kdwj=E26I?*zsYw^%^k)EH{tqzr;YPn>N=BJ^< zBq+~Mc~&6pelPT}hF4Ouz#*k6x}Ptl$v)7UGjx#D7b-Iwm^!WaX?sv}JYJ6%mYB=d zu_@UCp){Y$cvX{Z=Nm??2m`AiVwDW($_p3Amt`R+fypZYXIv~7u^N)Fb`C+hIlhpt zPKSj`MHzE;S99#BB#)?`rD(vGb9tiVIqE|54A$(l5EzLjYCp2Zb69DaBi1$*AK^gS zPPMP2@q$ES9T5WVq-(9(VPOxRFs@%I|I}$#_5`9 z2t7?~&iPIw7iesmi50&QcNPc^Vrc!P?i6waS?61=gmavtSkE@~Q8=K=!%Q)(6j?@t{upZ23E^yN(em?2;z$rC|_k@`4Rn#G#nf4soC8@fZ zPUD@n8BS_tiYF6B6ooQc8zi1rPwT^Ftxu@|Ht&{8qX3H(0d>PWUIakkU}K6|lJCewAL0+c`QfJ1vHYt6g0PNF&W z&>Mzc>z}gZ<@nf)Z0l%GRv>g02iiInJOUUY5l%QSk{%tictR zwiogwB&SVjsL08bs(U5~j{0HW10M^h03R?*(Qu1L!+0pT*ur#LgTX)mUwyILShnKW zx>8oM9kelUg-YHiyWl8qTAcL4X-iomQwx~a=|Lll6zVW;VLpMmt~Hq!Bg}Qung-V$ zt}D$eQ_e^i5~EV@c9Wrxb%8Ze>FJ1Cu*h&h!^8ZDa&QXvYY{jjCl;M2gQEnX)}DE%Xwctw}L4PgiD~<7-YE1*{P2p z>%}9-3otaNRW1(%RYl`Y$6N+U2vd>Y>r}K_mC>7q36An7;3V9cN^xo0V@8(9^Nftm^in>1KolJzuKT9-cOoNv4NQG!Up8TGV*TXYVAf}B?C=P)ec9VbEiearNX zglWV+IF6fkl-g(lyKXw?wK_sV#Eq~`#yxmM4mwi7$)@Qk9@XGsEuTeE&&LXw9rouE zT1e|jlk7~@3@->>iL#e1Il_i6!UPN=fOHbZ7-!SKF#+)>Jg`J=sHZ0jxHTX71B{Ht zMJDtDmY%nPZ9zC_t}uojy+QMJ%7JwY+izEh z`q)g;xH0mEL9U*l#pP&>=Q`#LDAO6%&uL0ElpF%)s}U>&4y9&Ww5|GV7af#`NnKbQ z%cYquFSBN+Y)=PZYXWmQ$~4QQ)APDHF$*YdqP@}#B#}%cwLhygFshk#%N+}u)6po~ zw+)SKC%Jh$Ec@OJ^G8a)hjy}=aR%gpAn9g%kTv_E-807dSqw5?K>A_7k&oCs>!Vy1 zW};jc)t5BK#G_Hk?$qW&mH^p}t!YS$LR^8zJb{BugBrtknM}dX;yFt%W~f|05Aqnb z3KF6)fPn;6v<(`Ou#l$cr`5@nHe6kC z)I3^VPV-VQs>>uua1unH2Kl*q!!7d@v*0E$AxJ=xSG;s5W?G|8Iia*jBA7Br`s^ff zBG&lg7`RXRTqq=^z5q{LhR>nZs4(X}r8$%eOd18}9Ekvq284-3N>&NWBMT8YcU;Ek zf~?c!TxvQPx0Q^^hDzHRfYr(s?B3>d5ZIMX6qp!R2Ww^;QbFTsE`c+G$4A+5ra9z` zc2dlPbYCISl1j+qR&$=~+c=MD88xrW1&S#3^DVv!QmqO>&Yd?Kp@piNn{GPehSdXS zAWM+(*kJ1UJn+CJR%LW>kSH1vFA3I28RY8r$e6L2aZ7LD1yM?y4Pl~IOs1H4AO+LH zK^BjQI`s+2XVMpa!*7yonF{N%*yvj>TE=^6r!WR#%aN-YQ_%uPlmR1x>?et`WI?m? zy=6bBUvUGWt&doVmcLr6KK(+c9?Sq_B%(XHNkX;5?%$jr8EzedM3H%^peTK~lY6Gt_J};t80Yju*8*^D-Xt{ZcnKG~>_wyi~IY+}JJSvx(Wx6_D zYILIE>|n@dL0&RTa|GN5xu0%Fn<{mwD`ux72D~3`5@dRFz0JTzk!nwbNDt6jb{e-B zc?QxFDKpO|!wFq4mC=!I3>d~{wHk24i`^VZzZ9oYGBWw;X8C0qkQ2=1V_U7$q_R}) zR;R4aw4UAEWL<4?uL@*=AFIVpiqSO&*y%{xLXJ!0KJaPkjhHE-qX-=0#f)HO13IeA(|yHWC}ofZ-kii!LaHq*rLY>B z*x0S-RSN@Q6P+n_8NF6y6HzDuUo|MzIN9k;n|{1NTV*0(6$fTaLYP^YR9B zc*XjO<%{e4q39aAif-5_# zZK}4fK%GU^*DVMI$TpXh(Djc`pv!jduRZm$3G~n4qgC7Ve)P)q!hs&&FK*HH$|>}L zvp`uFDxCP}*}Ov654cb|^_6Wpwm%HU&-b|l(5F18d2Yko*Dw8zi?P)IcZjWr8g9NL zvY}yf1MavF&B5=ju{`HMtwU4)+dlM}RO;DVQ-9;(d!zhosJ#i>xux~gKL*fI=){R< zpFR8QDv)XTIO@%Dv_%)^CiU*Fq=Om;v2(zUD8DAs9+uxcTETG(BIt zt!@h!ZueLM;XB=e$JJx7@UNLc>$5=LnL&4NUeBQ)-<|K?96ECZd^B}KaLRr5`iY-{ zeqpD$`IPDg(Stt?ojKv2%ZkkX)zDT(G=hqcF^$2KCG( z__MikzH`y?6M%1j4?Z{i+AG#y`K61l8E)UTV0WU;U%C0*?}O&M01VXFT++Pt*!w^M*cR`C_?+$=|BJn@gdp4vvEA^(HYTJ)3HCuIy6S( zcoql8gjqcsbv&ygxm6`lU<<@}?ZqMsf|^2mF65fL-$(>WF8c{uk+7g61ASgDp*_}U1Mux~4~lqJ#>_aEO_(C5@PN zP69bnq&IKk{Lt<#`BJ}HS`2w`bCl9mKIXH)f7b{r4+k`j0dEq?jqp*n+MBo3i6=?D zf(dSxia|*=^$IBjSrdcXyfBn-a4ct2w%*pI5i!efZEz{1Z+9#w!RI}hCu2|d<`OcA z8h%pLC5JCs9a30!z#SrRrIgN4c$l?31dsf_qswfJFgn_yDr0QK)q`5j0}TrVJ2QIV zikS}lyabrbqoj`H1%UtVO(Iy;lv z*{Fo56jAYYf?iTZ&I(#xld5qHObvr3xN3_KEj>46u!xD17Puan55S$pmLbD5z0hy0$fL`&>J~29)M%ZDnaKmn6k)Z zyl&b`kIEFcG^3#ABy}BI1ciaDOA)~$Y<}kT8J_1rUPWdM1Vyy6_7q3@T`?>=AaSI? z#niCQir~U8vuI{nGGa=I>%lcqr%Ie{h)5ely5Oj%xEL1UUOE|;(lRZJ@WLN1nu&%X zh~ER!OS?;uFHi_Vkimk<6q8Mgu)i2i2ksPHIT>Pu_Mn65RJ}CBYBe6w9Bd?uNr^H_ zT@%qoKBN}#f?z^$9k)=JO@&bg_=Oz7Af?$b-+)1+yG2!nJ`j=rpSCyqnO-0!#6-h1sHcGCe)*3x>{`_^bb&;8u@ zb^Wf+WWj<(_pH~$%o0VOyB73WA^zrW7#HWKxQ8Cvu-qp;uzflSCN!%koz>y?)ZpW6 z5}v!oa9wk1ybV2br{TIc4I_OSWbYofA?%pUWW|VfCoJ2I=YbwoCVAn2d zTdz5j$ws&{#xccPr%tdp;z=A)rtGOLJxYx&;*3yw`|4Kt97vBA3#T59`E=JU=puGnjh@wGh5r&byt)`X6@%MbJvXpAnZj-<1ts8!2vChPjx}p$o=|@%@BM zcd{5-n6_Bs6@AYnPw-eu;0-NM{gu0%NobxP5P?_{$>*2e1eE*EEjx%iX>DJwM7Yy7i~>#fn6Aca7b9 zSbG*MeP>;fG2gw`pt?0SxVVYn2`%Q_wz`8;O_7IU=2<8Ailb~M5lY==!Yx*~DYypa z&Kf{d!iV=u3;uPdOs}E}D(K*HZ8WO{cM$SCO{fX>9s~_zk$81t<7#Emhsax&+kR;0 z)p2%rl#2Jv%Jem*$9cVUbcgdOi>y(@r&p73{Q1>gCRsCoqck^mu?)i#N ziNpver|JXup~BoHkh){(wjs2yYxibL8rUjswx)%$722H?$`7!Ayekis#a;*h%5Cpa zdS!2U4{%X_d6eJggZ*7R*84YoWxvsbISW{@WbpFC_Ce&Pe*)j~1zzi#E<3tY%lp-M zJrDV+3|S1s1X75yY_+RB{s4=tu#w!+8p<5(0$nWcM=;cWaGf$osPQeSgakN9zM^D#?50f2q z+(Z5U?>cz@mRIO+bRGP^yFgp4iUc)oLc6BkrzgyvUgZS@H^Ir!(CS&v$d%GS<14xv zT}SyXkV?bnshFo0u>l^)UL^azPdo|qEhD&pWWv5XwLO87xp}&<9Mk_nGRS;%JFJF@Ye z$D|jnZqiC>fqk;VQuju}gyqOaAZu&f81u1o^-k_odo*7vv0NXlfZ)Xx=K|D}x9A8X zp-sbTQYVI`2R{FUBA;^G>!)bfnXzA=k$h1a=m7Hl$6eVDq4kr;j(o z#Uj%q19;M6F`+>(NC;L5^-@AAGoK^jUBd0|Lf7%_7{Q_@PjZmqmzhxpxUCvjExVu& z;61=13G&r1>(0me0}D^Pb0uIgG?R6o;}nI$9!51M^CHxKzCo=YC3$$>h7+hTQ+^|Fl7BaD60IFd`1!&? zEw%TTw~opR+KG2j}Zh`okq0VvxSZWLJ;pLX5 z<8A+-E+1P}+2t0TtG28i$a2_i>&v!N8$G_X_n9soO{-V^x>uhSO)95Mx$q9D!rihg zJ7ifFa1)>c&~R8S+Sj0@^CMxD=~slI+kH%?Ma%CR3^ZX&t7BnW!NxcZGvkR~ybar?Ey94@abTKSeV)?PPGx-ybztR?|Z~A9vJpVv?sMC4BQL zapA0uc^Z`X)j(&0d5|1zPVPeLB3SpxI8=iZ+}mY0kYbIP(z4$ctX2${13^pIg8`%* zL2;!wEQC^!JHm%!atNtbv3Y+lH!u%~a2R;(9;?*AsY<3?)GM@@o9KJAJw6JO@2*(i zU9hypm^;2cMlZr6M4H)yI|96{H&EA~<3nSY<-(p|gEJ;RC#4~?Szdv)`C|bC14imI z$2S4`BVK_4AigG~RpKK--n=DVQ+;z38$ILH^Hol^+uQDu0K*9Z)gkJGl$LRaR=3^G zEytJXrbYqohpPrsSUWPbr^drFuRdWB8u$FDIr+Rd!9u$Zzotj<{&h`)rL@DHF77J( z2og4Ul;i*72m8BttoLvF%6`p*{YEwaOcmuo8F7W!)Q)~=@M7;TUC&n^>Mt0QlW=Ix za=Y2@ZD#LhulVehG<7Ho4{Udr;3W^@o9Js&wiOkN1qO;BM+RUPG+;oRiHU|LSNd+t z^!xR-Byeb)M|yeMe$dpGQ5FQlAzWHbP);S7hndaPk6qxVXZIX&)k=l#sZ3zmEub8K zGMNA>>Jn}DfnE;Q>DvUzFckX5{sHq zOB!5J)S}E+7J7DPq@h3XFw#$HOd81O?&{1xcEIq zEzFIfj3l03g}qi?%sgdZn0qZQPxmGi+c}CJ<2*s%sAL>hswU9u-(&c$H9YH2>q^5( z@@Xhh@Vaehz$RHG_@tlwvGo`V#6BK2K4PbWeU0N@B+Z;D1`5m3zTLa%=DnqlTPCzm zo?AW04Ce1zt-&9-@CR~zy~P)txq|I-weMQ-@jIYzQNW^7Z862~QX-uUrNG7kWI3T( zdF}`uuH8Z?>WNS|lx1S{cnanSCk3Ygr9-~IitSnnog;uyMIKIXe34}z6njE~&L=G* zS6{Bo0sLX`ZJopJ9zQl6H86_g5~hc?7j+-;=X9=jGwfJp;(nB5lMj1!}`!;9vI zL(df+)OB74z@w~*2*FQj5rO+nyX~c6D((f%#pnIaAi#anjW<(rL5B_}__i5c35*OO zh-c5!sx#-pr#YLb&I{6uT~7sjb@#qs7J8DA**~KM{2J(YlPrnN_0+k;<;;$`8u3fB z(t;j62>n)qJ32cvg`z0hrt}hj!)zZ6qR!J8X_AkKTik74Q-Kqcom>|RNQ2%kr}pp) zdv3bj3xJmp{#cVwZeS6*e;3+gqSj$R6q$H{hAkLM`S4l`6TRGs@QD?-{6gGz3ASKh zlHr6MRI~X*0WcJE8Xk0`s?j7Z`0=q1tDSgbMETUq8-A40<=(^ghfg$n3<9|F=PUG! zF2YWj5kubMWSS2Y`(`)16kg!D;$An-qwk>WOL3_%SW>Aj4!EHis_@%cQGBP3$L>W^ zdvr|-Yp2+O(VbK2bjalH%P|R$Kb>l%lb6zASyG@E= zuo-FK9JR6McO*|y&?if6++HXwA&jn~6Ijxk{dKf}_k$FP?7K|S9}q7mypwQ;3oF@0 zLrR9=@0XT(p5lfsBU{%A6Y2KRE!wNSG!*Po4XvQmNs*q4%|3Oba-KbmroCrNjuw;N zlCgiIriC>pv~HnTCDBKXJ-7lxIk1wN7(*;f#J)hW$UX1?6`{@H>(J1Kg*WS2$>6O? zm@`S2@nO-E1Bf@$hDW%m#@7-Wxx}MtT5%$032hX3gwnf^@$1Z_q`tl*iAcd>rYWDDHDaD3-zb++s(y?- z9`TTZ3+ybLsQY$*Dl#LSH?eu%Ze09YJ#D%8XJ`?C@^>`PSc@i8K zZ=OPq9rQbOSYP121J87u1XA@BkUqVF^K=SgeWM>n1YV-^`84V`T9(HGTKBzTa%I#Z zdDKlg`8*Fejv|~r;R<*UUe8odDF=#TGKy}n7@;%r=7)M5B9;<_-+G;YRefV&etdhm=g0bcprj&T5!7 zng&qTz?F5NeXu;-M)lV97nE;IM1jLwve4BMj^j?M+uK-dG)^pc296;mGkE8%7+-M+ za&?+#rdho8+;DUI(;U#1e7Yg0Ic8r47+Ec>B9yoLoctx?GGyvlBNov6EJNq@G{20g z-Ee#zC9aQ2JzvZt1MGnLuKB9M^J>e94OriJ~p^0$m9qkL~ z)v!(d_#}fZEM=66=2s|#R))saI1hd_>vpf$xGjA+;Y_xf+S`#hH15UZrj?dwq7iL3 zIKb5tonXubb0lJT53d`h4Ilt6?uJdK`0E@uP>!!P_o67Q9Gr`ZX+%eq9s{Oq^~M2W z6P+S}33h3D40l&Q?xEx}`uZ%6W)%SMPYO`!oH+=vHuE#(0WDQANA@sems&!Bv2kyk zYZf5o=?AF(m5AE|ouzO#kTw?LszJ}#H4mvtgpW>Wr#B@>k3TO!p3;|s)4Gei; zSay#Ne26WeYDACiE4)!OF1-vzV0;uozFcHro-`(5x*2B&&W6+JIpqX26^x>OzO9fr zwZlD1Jq?=Kc?j)<&-w;W;I!n63J-=A9!cGo(wH7J(fzWoR>GlwZe zb#`Y8EkTt(OcVu2s${;)^00*4?D_dnsu%hMipz#~W4;4U(iI7vnz&-a?OF651?JfX zj$&fGd*R%xW!F1aHTO(mj&z^UGavv&Ka90vNb2T96%nS&`$o>=j-h<_3|Gp>NJU(7-f0%AVM;!vB{-QiFoh21hn44n01c86Unx zA%vkX4G0Tor!iwXA8x3ponSm+riYdpn($1$Vfy2kk_QW57;n+JO*>-88{+KI31xMO z3k}0yOS{Jrj??=+|68AEznmMjP#cI8tjaRDK_U=Z`&CG3UQTWRNhC%cW-d{-TkXr6 z*Ov3ZOdX{}HJepqXq63!%R}#o-yJNyd((sJCO^B`Ra@HusTzC_Wxz6QmC;aU>$C|D*j{m z=fC)^`tn0<`TM_7TYguZ@!$UJhu-oZ`079ZM!xYw$>o3gvtJ~ae^2pOzyJAv{ZD?^ zAOAJ7%ir;Den8%T=2t#Rpa1ZG@#$at?0yBJuCG7;L%;9SFMR&)L80BC8vF4v1^E!Ecjs3A-`VkKP`@ix@{ro$A<&*Q(uYRME`PHBO{Qi|s|Mrjn z^NU8t{mUTB{LycG{OD)@VThIw$9zt{PKKc!e(CeS^Sw`hi2Joa{rcRF$=6}^Y`-2K z;rJ!%=U2Z~qWmc9=Z8rKU;W{q`w(mX$#11h|Ks2Ok$B~YfAS?h2>TkKA+U17TM56s z+duyKm;d_j{NDF}-OqpV>mUBzkDlr`$|m^=|MIjyl+k_nX+QIC;Kl#+NBLF}y8rSZ zUw^B``|gAMNO%MBMStk?zxw@8fB(;Y`C@(?Wc2NifA2Rx2CeN+eAJ)*;Kw)k2fs^r z|K=xu?1S+B(_h>Vf%t#=r@qC1@D~g5=f235e*9=ZB=`ShtlGD4w)^sM@89Ch{sD*$ z|2iB7=nZ_`rVuM$L%SMM#lIW>^qc?AZ}r~(<=^x9Z~e@tfAp(=_RDMeq4ex0KKL(T zX21K}q{(0XMyd5@|H_X=rhnnDfP(AifBECBe^*HSqwDeC|K?Ns_ImtVKfWG+`SYLs zt03Y1<)8Zyt$q6C&p-2Dewu#p>mKJ9zR8F9%Rlh~zxyaZ_l-31H~7OZK@flAo2UQw zQ~vHBe*Q~;?b9#(`04+LzYg-qpZV*b{;i+*(Z2%!^+k;O{qOuKsP6u~f9`kw{?C8u zzXGx9-}&eeYlY`tSYy ze{lCpIQ2a@9P+S;ClT zf8l@n^!=ZrvHQP%A|HYJZaxBK-d!**vf$LbERC%OJOmlo*=1V}`+d%X9N{KAjwR8j zd@9GPmMgLCFPtFzT?M+7f-m-CunQ)bZq0o0(87SJ)I#?~dOy5MM;6KDv5y}MvtxHq zHqlq$lP{(~p$VBn=}L3^%I%QB>7ftea41+*aUjvs_kQX{?1DbQ;mnM9RUZV|qNlb4 zJnF$Z!T=~UbM@ghrpV_`b2tj3noHrGS4GW^&Gw+?s@v`OthW{Fwc6>)Xz3|aH1};= z#&!`&gc8{jF}<=&EEF6rA*Uh*WuDGAj>k)P-QNVG=4w<@cH_-?Bzy4ZWcwKWcc3^^ z_uF<^6BUM_K$Vz*X{l!_7c5{~EoT5XL!d+7HRBP;BAqDo2dC)DV6oZpczy@NP07wB zp8^A21CnLEUQgVGYk^pQEgXsOdrU`xG)@Ap4$ME)G@^cKET^(X>=evug?&w318SIE ziChg0q6!c|x1iqa--IXbZVtb_RE8z`Z9hPdzQ-cuC2X zvn7!G<(_?__Uz>ebQ;D@3>;rUS+Qw*&e+wh@Ib8D=IQGUwih>p8Oe2xsUxRGIgH2k z<_e%D!Zo-VzZ_F~gn{rzL5E4b3S4Xf=L^U}5EDV?Pl;&&Czk+N>sjT2k2wRd8q9S1;j|w1XXfJ+c>z0_ ztn&ze4Dl%w?Hw?T1P=gg2LhoUEWDyLJMjA}e_xK|z66;NSUOesnhSW=3k$+HkWF6D ztus*M1tY`lj^g)BMAf+_>+4P+r55mabKu(`wBmJ9Kns}dp*)-I8-!tFK{Wbu21+{P z6)2ao5_x>`r`bx=?ZVII9$G&_$?xGp72^%ZEEHaNWIj~BDF7{k4g}TOw0|ehrwxcu zbo~gS!vUsIdsYT4je$V8}P{O zA;=-enIW6&u}NzPpwNLTN}bmR=3~JkgP&+3Q)m(1HE5MUAPIk>hJay#3Vwr&WXlV# zfP>uTZ4``tn=nlT z1J2Rq3E-N0bywH=W-g)&u4P7lWeR(aUOaFGVSucr8{zcEsQjy*OO;;=w{%KN#u8oH?kYt`kOT$FH?6tbtb&g85p@w z#GFOwND*Jo={$~mr?MJ|*@)z5U)TuP1BK;$6R-UFRRqPT z8QIF;qgi66W%ooKNR<$uAjUM%p-K5Y&uF~A9$FC35+tf#@@4O%0LBXN_*qa4zd z?+2j`EMm)=rX+O4f#UDk2qlA#gr8LyL>x)r2-{?SxkJ|?XJHKpR{2>8C3t*s9}K3j zc(F4w?Ln0JkeD2OIQ#{l=5q?^4oAdp4(Ihetl+hkdGy+1OoyT^zg|uzb%pDqjNa|V zPhgAo@7tw@D^`nE7HCJ=J>d}{WD9tu3XMCkpeV})zSx0Cnpnp_Y|Qz^x+fHR*d>O3 zf&d&+bZJDH0oE_FfbM8*7@uPE zj8|gqWQXzr2=k*Cz{SsyAyb}5K!)r zjv?r#Pk6xj-2niVqzQwO1-cdknfR3zb@tGgkEGBEf7vhBLXUE;ORa;wR>ZxhA=c!& zK4G`bNfqHJK~8yj8T^-kHil{Pu#+1;u?M|D3Sw^XoKJ6xCRL@5%%^w0Z0n-5Tg_eA z8*`1F@{G(u?DN*^5e(cgCZIVt$B}~0-sG-6JjphL2lE3tr{QQcn$Ik|OCUh0?ArxM zswXy$*+_04(8t?MMblAd$3rdxSSy8tb+|pAcj28xXP)Tj`GORnHcFjq*8>QaR*iji zTF&)hCjo;oyU8eor`+%dL5*;a<_WM-P|JCQ_Q5VAz#EfbdLSmQ0GSK0w83X;kO&4C zApF&C=iTIAXAosXcL=dKAQfQjSCFTylaag|=)zk^s9PSijGqR`IiX~ZBYcIw0-dY^Gmo41=TioinZFCZ4O)4=g&`IIguqH69>JQvKS7_><0~~gXdqg; zLZgwU`8ryF$kLdmheZypN~gLe!>& za&3V1F@Xw(12f)DB#*r6`H3t!=Jth`c=aJ@esR>lk$0H>$kpjKl6J10!L0Hom~YiO zMk(!pSEO1_bAoZ3F6V@g>=^=Qkr!a7sH6k}=-D<`&Gt)WkXppxDU>@Y`zVS_dc>Y? zuD}J(4pw+40XP_*fSOaCy@L*7JTcS{yViLfriwGSkl$fBkvw~c67@BOMn7qW5UBDb zWv+Zt$KF0zvCEla{b-x^g~65X9oPG;zcXGp_6lNz&|vQN zFTxDV$HA9U`~V2vhM8N7Yp4UT%q8LGWo;WhEwd-9O$;A!<)2=XQG){15brw{!L8+R zEm$sn4C*U&Zu~?m{H+=Qz4YBNr)usP6|axN$IRqS)>N=?b7_+oND==QIMmi#OZe;>|Zvf9NfH(Dt#&iVPeCPP`n#=>3%d8JktRYSTP4 z8Ez-~CftB!{4(T@5Xy3bgb1J~F5>jo(9&GJQN7k^oF}FUl(Z7+1EWCTE3(yl3=mO> z$`X-U=&PLb`NPAaviVv{P~u7?&p;+1vU?yBE7UskT$nu;)`#G*JL^~U#5c1~0-6(u z(O#VhD3DU_L<5;u*9=HNu|&SDV>kih{eS^>S8!z?rh#kl5jlvgI8W`egsNFujA1u| zlE6B2Et(D)Qo_pk<@X!-{TEeOEe#n=?drtvkrA1^)pvJyg?;5(Z(7Ypo>`(-`gVL0 zXfD=1-Jff&5P9qLE{8)>t){;d?%YlS4423eMhJcpw?DS_gH_6dU6*mT zEB$&i$JBpwsX*=6=l)5(Vr7@W9pHFwW`=uL#}t^M*7qYYwZV&I^fw^r-bh!q=dr8ID?Wqfgrcl=#2)*6ca3*~M+=30L=X(qW zu$2GK&Xr{m4H4pdzA$XWf>ne7DiQBKIx#>fI$yld25kg!h8=f`E&FyDO^~lv+>QWE z!&ZZNLC(Bzu(R`7#G?Q-LT8QJC`ZhkZ1Ar)T9Z8H-g=M=6b5>hW;Dg zK~g^IS0Wdua~;bx!&J5^2pwn{*-Z;I!@occPqK739mGF zbO`pk?m?oOU2)t4OXuK-QwOb`kN*dru!u-^SCI>CSfK9SHoi_W19Ec6O8%(n9U+ZfLb@yHJTCDS4H0=>S4yQvDIw(GG1pAd4P z2w;*%oPIcrLgXY)4|6h63*;ApDfMH(2pCFVln2>f)Z-#&AV!x?EC46(j*^Gfbx!7V ztcleacV@GP`LAi_L!*wAB|AbhfXoZ3Z`PnaqGLctbA02>9YPi~K5>HA#pwl8v`(jN z$L+nRq$rk&j;Vb-E-CcXE{Gnt%bHErTAI0h)gB=vNyJ9d(^S36vN(pDbbuO_Y) z1Q;sOGnL|@mSkzW_W_v&9VPruGI6oR@4h(z!rXE6kNf#93NvmiEzHroV0d1b5KA|+ zkoc!pW&=cE9+zDz+12p^(@4lsiv`*ZVPHO<#L~Y!+p4O7VHKB{Ca?u^6}0qj`^+d^ zkXr>JU3!n?#>G&K)#y;&UJ%`LC)k!xSz*QO5x`Z_p;cco8W246v@P`)> z(tjp>I9Gi7)OjltGf@EkAJE!%_3WO^JyKEA=%`^nE zs3dH1TRyw$GPKzBZ66EMcZD`zAzna>=8KgNqc7F42ORAku#b9Ur+?&UDo>Q2B@J#e zP^pLK>hLi6rPV~aRW765cV!PBVRmJp^Ffp?%Ogt6qxh>fSoASVD)#YrbZAw(zT>m)zjOY;P9CRPhjD?3WB|)x*MdT%rGcYDhQ1g_f8=a1ZWPu!|G)*?T%XOn7LD1 z=aUDNZhL3E!tx}b6)@oLeXt&muK%duLLA{CYR5j?)sssGFJ4ahR#k1aj@=nL1JzNT zA)m|~_9h(HXgJ%tkl~>b{xE_#HE=2kV&>GQJSJRDJYNv;`8_L{t+RIE+oyNEGa#ghr3Vo-y!>1%#5!aS5AmmQsu;93 zqfuwn>l5u?!IGd?yQ1v_>R7pA_SXvKK{!Zy3hd=#xLOa0CmGEgRIWn@=l|7FW%5<} zrOd?OVO%bCTSLOk!Pn1fXueXq^?!{dR&&3T)P?H$MG%TA2Nwsn`X zzd)1Q+D}zQ8n!9j;M2s^(mOgZWsiKz~{)LcOC{dRe-XS&>!2`@7-~Dovec+iYA?P5byfeQOw9O zXbFDz(<{7UP6vaD#@xVM;3JoCwkZ%IuQXt{AyA{bzO#ggYV=M#kIoa2WSA&goSDq4 zRAFZw$wMFm4#3nb)TI@jqWTyf(7phPrc)Y=nP5A?0h&8URBb9pc>pCUK1p#hS`@+z zAOB%2B)!SviQwl~@q7Nt6;plr{teX=`|2X*i#*J7>>gk{27DxHDk2 z!Y^#@+lFWGM;nU6Lk3;=6qw(z7$NKPy+Zd{vOa3E-ps*(6m*9uTo5xU&)B+)ftpR8 zJoS>*vXdmv`W$x(%x$JcW`XJ!@y2ccgqY#ngw)Zfe6DcU>V=D@5NUO}&?LB5x{j`g zeh--GJ%|RvGqA*e+!29R*Wm@u6Dm&NUeidzs{CL>Diqwbi*ae1Iwpl0GTwP z7KzmduI_fpa?E)Z0|F_POqSkAophKOAyMfXa&;aOH6hYd@OBbW@t+ z$>}mBsXRBPRom-(mYxzo3jji?KsS*sA7wsj!()M{WVCzc_Nx_vS9TJ%0L_42ZaNTE zLwX<>BVsD6HULHdpD(fb4DldVfZd*%8!`30Ty;PS2ie>}{eK14v^KrkbkHFHQhCe~ zdS>Vcx{Sd);sKK^ycX_syklsJPQxSoD9|4fTW8kH&k$OE^?UbHnT}q$&##^*+2K z4gcX2z`dRCCLZ-5MWtDm69fj~;Bq(iRvu%0jR3L2e%Lgm8S$2LD^Ix$UQE9?$Lkq9 zq=w%xQ>iI1K^RGE*Hl(-HsCONs8^@aUI9DEh-tbX+AJkL z?@v#{vF=wAy(<N~I(#0K5kQN0VjG;B_T^3f4 z4_l<9`?rAGEwAvxAXa>{cR{svAWrKv8LHBue(Se90xmN zL_+M;;ywwtss*D7eb74f+TFtbbb%M=ihUo3sDb7WIC{pnJD~ruDLFU}BtM*c@Ud$b zKn37+1F7D;j_5#YFGk-JLECuKa1|=;EZm1+Dih2_W0-eQQ67N@^!{pKVHaE18QP>7 zFPNte(5_>#p-1l(vPNg9$G6&h4>%;)S7tt0Vb5meqk{_}8%!4y z3KU5$&y1HJFIhX-jxCxoh`TfKa>*5XUh@gsz;HsknkR3?pUv$c1z~F+EA0i%NeCYG zju2O}#CX8)p^$j9E<+df`f;YbGShvrPjCBtH{px#KMs%H$!tX!qeCu+no}0^{&lABuddPQ$-8%(6_6}RIo59jWR=GhSJdQV`A=Y!_$nc_7HoL(@W{k$Bzgf_& zxjvsO&>~uvPPsCGBzO{D{`>m0L>Elhllr`1z$UqZDE7Yj!|aCq)7TTw3)`(J2@OO# zNyp1{Rs+a-?kpaH#AJ~0UttWZfSrsIVkog=Zp&u?^=?>%MFSf51h1sK2>`K zqRPQ@n+ROc+??#F3-1O7&yz|U-tz=Ny2j|r^S)(^21_?bp71(hOyd_P5t`?k504WcEWBIHpj~W6t z)Xl>mwTPs89T-XwxF-)D3!Pu(YnhLfX~2*Szi)Be-q`NaU9?r%+yh%b`LP=U*rB41 zdz+Ou9z7Sj?>L`>3+@gRw@QtUG_=aPecVStx*#E#B3cv-`tCqBf&7D71?uer??EWj z3gQmH`QShRxZF^uL$X9*_{sNyxT4DdOR)!ln}}#C-Rj!p^q^)19%A1SG7<;aTMm4F z*QPl{>lB!@FJN-5dl3%C5?IQsdpSFly-ux86dN^zy5syZaK?>4O6@J=DmRl&44nCJy3%)P z1ON57KX%}5+-MNr!^zXbj_eJTCvgj->W;hg;2jLG7Xq=}CXk{pU?yItoVoBx2f+0{+Rwa_{2l)U0awx*z7FTm-4(|e;H*K0E;DnG7=2nEaqM{c z%3?Z+w$L6R4f8GTl!H@0mHp*ZvTM9|o&}f2#Cn76&Ey)yU>RH{BCfmc^wnO zf4fw^UpgCxK1)03HPbYUfRYcEOz5}$p^Ij~7(|V-V@GO0hwG;1#E3EM1_i*xT7{!87#)~;e zC;MfL!HPnIvUs5P`{c9M>H=~MveEtG2>b~=9YF)O_eAu?-@&Ty%)GJfy7$Y7(%CLp z=%>J!r%(|72|=kz1_Cg^QZM*puT?haU|{GdywD81nf!f3IJ6|d`|0@8{uz0P>miPy zfbO)9?pK$INQftqjDem6z$WAQbdh2t((81~8G>nyRG6o_%E6AAzAfl{Kxzx`xaPwJ zEu*I0d3XW+Eb>cO-w2TV)f=(MiE)A66r3IR@~m(E)D~yz$aLicZohg{4aj%ZK7?gG z46Zx|at(Br^Yb~Y>j|zTl)5IUkW97}J??3U-z!LAv_LxWeko6x;l%|sE8uxWvC-ea zr;7CmO`-^-F(Dks@G;Asl^RWn7nj_Iu7#|T4rWYz4Y>z8s4{t8-lMh=C;0@qrwk;m zlUtI`Z7<#BwmS?^A;c6 z1muwVJQn*4%(7GqhIJ%vYi$Fu4_|(-t6utxKR8y^90EzdNYxPTgN}Xmcw{EHVYaOM zyke5P;F;k{K56dn0$bA(rSDGo?HvP(d~=G`%2P&!h~Yi`G9Y84!=a$8BinTMsg(vf z*jOZi!&Py;3CAXt9c^<;H9rQ9$eV|-gk%9jc=qbNh%zm`nEPXa=XjkVNB36FkNf5R zfWrCkKzw;;K}SlxKA^vNFyB42#y)2I4g|=9pmcy$TArC*v|2L?=B?s#MEEs+DTUvx z8YSqZvicB2Wh(hlT#Uo@o-+eyfX~-fqH(7~*O@$(bjqMIJwl$mkziEr%LXSSVBmo; zaImEzAo!Tbtys+&@)=49|1loS>r!?pkQ)GSG$D`!53+|8u4fp_g@M=jZfuII5LY&DaFG?nX^iCftdRAQuV;qDyle>s~ zLwQJ^W~yQ`*pOq2^xc*xQ7A0MNifSMxy~9GDx>P|M@&wr(}@pW8rzio2yyIZiaW*F zk=b4*Veek5u4%AoW{{PQ!FK$#Y~%JJqvJKO?dADYgNV{-%zUo=tXU&y!Nl)T0$G`W zKCbXdZ%-X&^zc<&*fQJj7sNv3aLIgF@_3_%;vW{EklaHAFspTpOdP4H*`&#eGsK7T z3L+{1BsD`7Jtnt_mNpn#K&-%dz50tjq6BD3lo(Qf1`cuy`4I%D_w5afCaa(i;TB8> z1_&hEHf4M%tvJ{zNhAWH*vzoY*0YLcMIw;< z0D^Z^3$!#KVhen;o(sf+ZI%T=_K=m7qA=&T#%Q*h>t0nqVJCapFxgZe&c|#wKNQl@ z`VjKNvw|ec)*TqLfRJBXL>SY5^Sk;r!uh8!sQVQz2If5uQCErY+oCPH|)_ir}!$RdLMJ0TfV36*`DFS2d9eryVXj(rXJC0 z)Z1C}Qhy?H0Y*OU$MPw)8E&Ro0J?2BV)uzYmVR}8y0^NtiQPA*gAa|+PkAGM9x z!A~_7{%8N^@BZp%zxw&*TldV5+;d-`h6h{Lla1 zuk;`Mx_ayCb zo+r`uXY19gSFhgX|NkvS`Sg{H^`t9b1>)cE!57_(`%O>%InXEZ@4avR{ZDy7{(G!O z|L!T`H=k7A@BIa$e&-{%)pFSakQz6FwKe z`TEa%(7sRq%@@jl+s*epM+{NM&4-_R8S_aGzxdi$-AcE2UQhhgooB5V7kk?WU;73^ zAin5c$@lx<>mC+h-F)>M2>I!kocfL%>K*7P$H!{)*m?ao5^ch|cdORTkG_%kzWbkW zxcP?ZQ_tMod@}KE5C4YSf97WUDa6-3Jn1cOBL3SaUHW(a!^@w)`6BkIcM^Z+_WBQc z&ZF0V?afEF0t3$^(86q!X`h7Q~xB&WZ2;RN_mYbmDE}u^;Ee zZ+izi8mfG~`{y31%%d;V-B0;6LcBw$&ph#sdfNwI=OEzj&dDFL>rg|CvvE+nq1CdCR+q&w0!m;imbrr~WGOB~Pw}^wi59a3b%&Ue;$2 zuOx`CedhMH{@S}=_L;=n?>zmrFRj#%zngd$@&4QTJa~!6wTFe>yydfq@48>i^>FQ~ z|KYO;_0HE`YslPuF-5${!QmSz;xFDIKJ?^zF23Ml(f7OAs~KWLeE7DWINW?2KHmB8 zy~6)8`{Ke!EJj%3Up~3;FaN%m6#lO{;y*t9Coj3{Z>|#}K|K3tPyEJTec|LkdFJMe z>qPjLm!9_|2cl;`PBCA|6E)&b?%f+oy~`7S=i$fqe#5ssbMvhY;u|0N)3td+fp`c0 zbTj(KPhf6-Rv=z|eIvi@mqF_NErIyySM{I09X*t1Zw?=P1^v|Dd|2-XBl;hR#D9H< zu%5kr^w!P$KAU*|{ZH4!^%9%UA-?rRfBO8-#b9kcdv6^0;O9Ov4*d4#5?}PHFTZ!y zzjE)Y&%OWAtKRGASLdUI=_<^S^Ld7bzR zABV5(bgrHtZ@jtDiLbjLoBGfgB(P6YID$3>pPX;qa@0=T@pueMLKKaAk&2`ac9(|A z_O#_}+S#`!n=I{!ox_T;!q8;~nYFV3&}M2KHQlY$)k5uc=gp+i0XTLPAyzpVueW3< z*64=Z?{E73{bkLpDMU|o2l{_ukcUAXVdME+sEeIg>F)aDiJGQPxuKi;`FIk!p=3|C zo87+I_4(O!weGoG-r$WH@3aq-kMYhi)HM==x%#_ zz^tvdr&Dv_Wx_olB+J2W4(Z@&+DiupTkW;;d9|@5b5c{mr2r=e7{bY9JzK7Zt9ron z1AAScSxYxD_cU-8-IbY~;Qb2(F5CzDtA&SMt_m&0wp0YiR6bAE`@xP~43pk+kk{5D z%_&cH5Wv=#`3$H!CMZ0~SJd%z-G=A6Zrg{B&MiR7Z7N%Uaf*U%gj0KM^~Zy)+R}Mh zGNv&dTGK;wZ?4NeNS?l!rSL3Idac|zsI8pot7Dh7)}*pS=KEBp`%pKxQ7(p%a;93# z5qf0>enxi1^=_<;D_;ZQgrjwyvp?XR^R*MuNl7{YcI1hrsh+Z6IWirYwhbIR!Z z1>8!vJKEKnE5t(Tz^!tm=uLos=cl|>yZQ^YUTK`x1Sh&q< zFa~KmL%rJyAd{`hh20Rl4VD7WYOAqvZP{}a5dK4ZHkbpjm`1(=ukRvFy9}-4nZn@` zl1ctjlbpB}EbGl#VsIb^oadg|K53)gd5yC5?rL>z;D03P4Ar*m+k?8m`{&xRC@z~= zq|~mW>NR_l!TxI?STI*@wDysiw&YS-9jFzjFhimh)iXI_yb zy#EriM{OQLCZ^ug{6O&hruASnkvV zz0KPrY0o5a2F+#BvT4f8VM~_pcjn>$0?C~{rwdAT<&n$p3qWTPglAtc@{tq zakmc#+XxKTatY60M>sL3{c`HiT2$}HyKSQZC@0@oflyHj0{q^ zD?MFmHqtn59c#53u!%ki6-0Ssc3b0JyQ6PSy(7qAS_QX)RDmy3Z z@l@7_GdE!y{h(K<{z@tLy(NYlemU>$A=#L3>-hrGs_Wr?Kj}N`p63cfGpx00S}-)S zz+8;V)3Q=vupkT$z06SP(bgC?w=hPCp_hv>CWhOxdbn6qzZsvwG7hxXwzjGvjw?|S zj9xZq_K8EE1D@)*PAeRdI`2(ZYs++c0t0eBm6_hDR2FdyhR~y|y;t=f+hL8M)WisV zvSYRP`}v`s9TGt4hNF$z?m{|oqXwfz+3`WzSCB8?cjxKGG7t0Z1Tkrc)t84ZM9{Wf zcBqdRO<{G|SjG_8ah%*mz_mU+&G!epZVrWk0*nc}UBejz79i~aXk=3Xg?9x~X`utp zaun`8|D18D#ToNJ8J*5iF<_y&rX+L&X|>C4ssn^KSWhIUus2oVV%**8-BXPmq;i;-BPT}BS?vsP-6NPEO zx|;HE0Grl~Mwyqc)?UqzXJ;CMuYbg=(>v2yjKQDW%#0QwIW0r!ug^hU@SKGn!dd3D z7J`;k6g)d09xOng=VQoPFjf@Pau2l#Wv-}fevWF)+*44D4-L+1U<4nEi&SHHW)TGW zF|!VgJD#2(9BphXy~{8c=w2?DS$@#!TZXUuq8^WR1fpUZxFiL>iEV5Aq}-FfszU10 zuP7QbEa26}W!79sIcKJ9BcnmkUO^{vOS@7>iI=5RyM89CMaENI_XK_;OijGAiz#?k zfN&Mo^~Ls>$V(>V8IKJ01y(>E5ZPw4;jF)tJMt-9`K@$b*hem)7RMa|u~cl&;*&rn zhad22YY8XgdPjnWU(U+W0>1EX$pFm=)(%3#Met<{CpVM$-FW+TEgWt+}xmVj53@7l%abX6Dbmq<;<;Fgeh- z7*To~Ov>G^4%|s6Uh->x$=ZX>ew(IOVecd>*qXK2iDHORvEB4lZG*+@nO}_tW#4G7 zD0^1$u&?-Km-$sp9E|<1v-&(qm6! z5{y5!Ug*Hq&hXRR$6kTIE;2`kwYHiiL(B$?wgOlvb@lhb`GIPYS|1Wi(}fd13VfH|`+YXjl_U2RD(ZZXUJv&A}W3E*j@x zVF7XP{*iXpqI+@#Z`(N<8sD==&SX?uAK2v^&LvTWU%}>FNfDws8WBu+jpiV350w)Ic9>>E9aIb z@DRC#K2%&q-NNq+l7RJr2q{Znh(i!b7z-8{1F{ckpu9&0!f4$a8Vsp0(P0<2SJSB2 z$y8~g9YJE89P+hWT-4^Fw-uUBt*K*W)1f4=kIw2oYPQc{QAt?mEyjTvmXaGai)Clh zNmW$^UjNv&nRs9taz$Ddx8MMdYtZBpT-(jM5|n5{?sfcbLT`0uF+u{Lv!-2x1}YN# zt~y9m%T>U`80sVhCMfwyG`Wg>DeHP*?bq~q6LxG11lSew_@p*uApyiX<)M!)Y<)eS zO~>o5J2~tYz@Dz4W)6r7cDamM()&n6$F$GA%B`IQvrVg+opMckbWSWfwKT(rRh-)< zH5u41@KwT9n;WQ@V8-1gwgJ15*f%-LGiSHk>J`ks3|AC0@P-q9+=tB#}uE9>4~?q+9ey4tC(ja~c!&v4*j?O%`?#JE0X44$wg3WXetR#USBCUsz8Oyy+&as!%%H-2jYptQv zjp0?Bx=@Uuo#UyjS=$1@lle5B#zuGDO2DR352FZ>;0kgo(kdeZajcIT?GT72xp5X( zJy1ZVD}JI*(n4RW2W#pWYl~|`R*Y_TkYrD%FIpL2u1n+DV@qQ=Eflrs=)B!nGNZNf z8b&bP0lLcR0z`!liIs6AFr?0uT|R%oTHn3u0aHbYT#0rfW36>xVpZbjhd_=AGS|1G74wZ4;d`0zT zJLpsD;`VV`YreZj9KCGCj=pE?}Y-#3e9#bWR;NscT-)daQyp6kVBa&R0$n$Gg0u(#-mo_RxY(`nLGUp)j%Q?L2C3C1IWG%L*xuosx@ zCP#nT=`%qeyi~0`9gm=JpmM!yFTeMF7-fBapqPhY*6Y_tc~Ev%+!So`Yh(nbwtyij z`Dz0)^*=ZOyaReqtlgb4Z}C zBf|nK&A58*u2^yrq!37UCTx*)-JDSm^$|AQD{ESuTEebnahNR!ZFfmeVq+(&BgW^g z%k&H?jw678E#Ou3EualFdUB9?)`I*^!TSAVwbZb7OXtmfbm!k$YgEgYBhWq~i)2 zjGK1S>FV6rNIHxodDE8N?Mm%aY*$)y3;IfD@6HZdaHvmIFP$bVABi5*i?p83@y)gLn*&;^DBtK;bS17zm&g=&efz);T$- zocZm!Bd6?#ja>tva}76{EE_<r1<_xsLr7mNcok-6#~L!@s6i zD}A{x3eDAeKQdYG_SI5A#L?a9=?KW#iW2xui2XFx3@HIRPrDAaX!1##fq=d5ltoiM z9j}Yo)@Ve@W&$~N=-|a#w`1RiaslI}%PmD-no~h2!KsLj0y#0A(H5%HLxt69uw*mXhoN~ae9h&iej|a>u zK2@BfPFuen$h>5eR?F31gRI+_lqZ)7DOVLKhg5CUXcRK2fV0iI=8@cSt_}D_#pu9m zZhF&J05Kge-lXOx*1IX)vRMQw*R2jy170<30hupV{dM5Ol6E>AFcrsXMx)_WzFFlI zqQ$jgRV_xhkE3o3W=X52OH;zv*ahZ&V+d!*eDZ?nao}N0hk*DUZ zu{zaiLN92o9P%Xbvl_eEch&K%UbL)a(a(5Dcx|VOH98b*XPp-Eb%j*zto@Mhh;Um$ zP(y?lOdU=}K*t+Z-{03Edvx(($80XoaCqL)l)|jr7|#}10}M2?ig*={-7EDVWO%`b zn46p33PNOv5P1*-IXDnBvW-DTwM;qyvni{_(NsM4jN63>xG^<-g;9IE+Biwy#*&Vd zw0a3PT4T@=^s~2N>`T!z-rG7IC6g(s4S($TusHCpO|AyB!C^&CPp+{{`E0p$ z+T&&tXa~t~Cg&seuyVXuoNM|XZ}hxhhj+{dz0UpgQmmaJO1U57eJ|9uOo-p+|GrQeo08AOI*(a)-Z+T`>NMgN2btbqi~fEB^zrA%IUZY zi16df)QbbqZLzJz)^B5N>S8tDVOaystb^vk8RERzHU>41Pm5!a4$tUi4)#STA>3djT4oFt6Fy|@#%oMJVqa9_Q%#5&gqFuhHTgT(N zs}u^YzjO%)yX5UA)Qv%_&2>(8OBC%E$Jvl1TY%r_TmAFQQZ7}rAq8!GNH+OtQzv!3 zS5VcG({^edff%(rJxlJHDf?;5=+>LklpE>2vjR&_z(tpvK9WpxE8uJ{8RU(seRj0w znq3T}u0N;X;W0s9XVcOnqJiFi@Oq@{l{mHreRa;lHdv zfo)71yRBr}qS9+`Epgu+clMG~DqPZ8U^UoV;4#JCrMW4)l%P9U{O@-3+>L{2uWY(b z$wc9-=K1oe$5oYSTahL#$_fS#gQPf48jX2SJ5#1q!N$SRV=`{G5-?tR^l3b8VV2KG zeiQUJoxvDZn4@hi1?BCq1pQq$MuRo5>gBvvV~Pv81s%1QrnC%pP-~s6;tjlKfcYN} zTH8&M1@`hhHFi)@sjVTv8WnRy-47je&yG75k{e;n(<=@{eC15bN(~WN30|{ZgQELX zQ6+qylty-TDvK)?5QR(oJ=hvfVgJqLOOuf=v*VyPS_cSL-!F7dYkMY_HPjOM9{Jeq zj@F}2QIs496GgU5Z8~%BB!**_3KZHm{M|;#y2Tg|8L}E2EFbh4h@g|*Y!cAWb=>yK zjlVdPSucymHHpr*?SiUVjkGkYxN%vyxl?ncW^>Y5ti7Sv*q)=E8fakLTUelka7}{L zY;`Dar(k_mlgQnsQv)7;#|Q_J@ zPTSI?j9z;^x}EN8nq~%uc4qwL)N4gl0 z^kM_-?%*$i2yFM@S{yQz>5jXKL8mE+DZeZMXvmO@gkgsRJ2MYrv2Xn-6) z6EwfUTIB*^j*F2djFTYL7*LVj8foJofOX2!9b>&_8!)vl#jbRoCGAt4QV?K0mJ9Hk z)b>?M^{BT7hIuA^IXr>(VNiCnT#Gv28xPub@CO@7Wi?f{j@{0^ilzLJ)*?qYH>K!X zCCf6Gk__O(3B^GXqSe`y%xecSTy(>uSfz;Ha3P59bUocC@!S7d<;;_FRu zv#RekBUO{zS*{Ob8N~8K!6w6o4i#kztNF=RpL4ZyjINyc3Qd{*?9e$vR&9X2;AOLs zLvy!k)OzDyKBqToIug|w+sbA%Um6B!XF>vAqP{%G`(WycdZ!3ujZ}CV@^q&hIrqh? z!za}=+ksfRG}f}y*N)ui5RW%oiJH$BtdU4{=FEZO6At{UISXi$3@DHGOVPPZTe~Kg ziKAi!sWAA>^0s`=j_~ttO?fGs_^2_S{Bk%~CLsuz^kNlIdNJ1)6Sfl?k;phmHt&5_ z>cvOo8CEh+yGKMxX745Uxg+DK#E)1i*Q&xZ`QVa$O%|YIKJ3TwAuJ9Uur4*$6C+^;f>ca) zL$NBy^@d|j7xFPMJU^V)O^LGB_1I{W@~+pYz)+p_mXPfMfpw~N6BS%7R5&n=tr5j$ za2ze_dNLdM3ldaivDXt4cQ%9=b)%@*V^K$Lfwmz`-t7%YOsLT7Xs&|hLswq&05d=_htvWSM1Li%ihBZbrXXdgo z*SKby?|K#@);+{)unFDNnnfh8nm%-W0pjRSpe9QxhaG(pp`>s~UCNzs!ny3A-PfLm zd=`YM`_bIgs{vcn8ipdu?UNN7{Q}B;Iy-Bcw2h>u;5sL#nP84*-3GI)wfO?H**WG} znU|FKRFl*mW>)Zq+hom#ULY;X@X{Pn%cU!njcR?_uWH zH~M^LJ8XlLj~fV9pGMsj62j+P)R){g3D)*7y^N9AZqAH(T2uV9HXK3ujP~304ioA9 zCK3VcaOkgBbTkDmTfFv#aVl@Lp`ul_aIvU4Q%G||)FbpL5(&gAPh@K5kyS7g7t@u& z)!bFRqYTnAM;0207({iHNQVY891I1CW-jc)jIjBF^1#MwPIe2{U7iYPcDd*FzD{*n z#Emu+66E@!P4%(Y)O#IZ5)KJ1bgNufjd+1D=E9Fb@1!u)g2WQ|4m)TdOR5QTVIafU zwh(4(C`}nFNhZs7x`(}&ead*do<$d=>$3VnBge;zN3g+kPA#MjlTLp;ET975c=T4K zN~H{pma(hZrP_$D%jf2(DvwoZz^6(pu1J`}gluZ!WS~t*gU*VgXDq-7AB5`8Zlu&Y z^IStEy8*cK>j_&@hy*bFuDiW7AeiB2w!2U=Bh?4j6Gf*)oZ?g z7M=c_($mJ4n)E3?%BGOen#C$7^mkq-IOwAU;R=;J90aX!ye$NV!mu5W2b|af*VRuE z$X@kCDlB2Bw@(2WqRm?OvVuxP1mmcn=|b!xmkTTDjjLtb30YCL=WMz4#kp4%)j@_G zp|5v4vr|2)bhg?m;tbm4tt?V2@5z8> z>5#j#ja^8RKG$+f_aZ=uRpiF!y%<8EQaHJ?=}%%*;uOdd(~VX}Ki`qf^$5%?Dwu>B zvzsIZyB|`Elp5?|Dtn};5;sJNeOXH}<3xfU+NL_z#!?@EBq++n+x}8-ckDI~35o%^ zX;5PkaIv*DuU%r0qt?il5%rJpb0?2iAwN~{MQKKBrw0WYfpG>TU-+y3yQo!BXoE%!4PYyDNw+g(8`Y>g-SXqLKAhvDTSI!2L~C zjizAI%NR+kYO;rVG+~!ma<0gQ;s*S&q41O8Tqrs^Fj|exT!Yfl*x+f^#?Z7^n5+yg zn}TZ&u;0mKre1WIh5_jat_>$Pr|zsPIALVkgvPrWB>UQg?@o70axTLKX7W6n`CKh+ zO(eQ|Sd8~EsVtT;v*opZZD25*ahfOL)F1Kda1e59MQtECp7Q*mi%9q8JX5zPhrOJ& zyv|S!Sw^nw3d6~dkSgVL|GM0cjsRqcnQ21QsA-(I?l~(OQ%+fxJr|KVf4Q;{Dv7be z%skPn$>W?_j1(me`tc+ZOlnplinuv0maBekNgfAfM#Vu?tg?`t#j6Q?S5}u%Z<}2g z2UD)6L+hjFrcDuE(39c54$~OERYEe4DSPU~Th7V6E@XoS=cfYTk(t$yiMs1RhdFcbZ4kz*lRkWt<|FB069)35{sb5ts34E zJt;*H`+(^2knYm0@i}V_ItL-=1=}`hq!Hrd*h}4p#d|LAAOw{RE&+^t8{tHp(;TU7 z;dS1{n#DKgpsh-EA-RGLdlPu(xE=x&ZPz&};O6ZHwQc9Hb`8mPH%*~FXGuF)9#Vh^7@EuWw~6_u=N{)^rx{#x13$DfyK_{WcAc@2=2Xwcd+L> zy`!j%$MRGjX6k{9imF%$=tmN{J*y5m2Lhx8NjY{Zty6X&7^akWK|ckRQOucUlH|r3 z8%;ABtOkBG4^H(>n<<<9xYCEK;gYK9^H$0hV+66lm#3CsgU(#IdS>Sl1QizmFVu}$ z0lT6);;Ej%(v74j@-W_6_1ri@_CREZp5p>d?!x>gm$v_Zm(JUQ6YgrqL=BSY)&4yHe!5~it2W?9pDywR`QYM=|wH-}) zQSY-|%`PfvF~G=lU-(F&Ee}&6G1~iLL=~6QVhnTdl5WE6cRX^mBbdYLd<7&N0I8d1 z8>xDyOvc8jwL7h*OX%~hOL-7!TX5u?`Nqk@Ola}sT=KBw)4K7gA&Dely7q|QP}+PX z@uQ|VIUQ4jsh=RFw-M|GKk{l0HRKmZ7;|paoJ}Djol4`1Dw1B>S(p9@4&Kzh+^@u~i|Iq0cqJ&_iVLp<*{_ z1;bx7(;O>N!2Y68pdA7Hb>w;gUuCMY>61vC?x%=wH{C5X%^ZnLH~ws+ z^fzKwksjF{@l|l_k%Nq{R{llIseYHUXUCD#ZLP0_+d&@GcvrL4VBe-9f4MmJwleHO z$Eq{F&UpZ%8qK06XL$Co1fO$WV8_jg19Og=HFYsXZ?Pb3iLS+r_|~jo>jpMfqp&{L zhD#|wImjMoHr0Zbq20bDle8=9=RT*<9A!)r&sfZTJB-Y8i|uSnS6`AD>xTwQ9Ze`a z30KtPphW6mTTa&DET6+lng*<(1DM3ROuWX>J8hafeStxFpSj0wu>tctZuw=P17XUK zxN}VyOTN#ZRy6dAEOba{@N$)RIMe4EnUo`%>#3T;h-52MoT%Es;1!c->CYF~!SBMs zcxyKhhE*uYg0fY62`+hKtmU=6z{Pc;1wk^(SWdjwPUERXB5*M9HXHg`o^NKQV**PS z)fhH)ZGrBdE-*ZSzKuRTFP7`G=Qg>(n)e|71Gqb_`;QT0*o`Ovv0B*kO+);cC6G>scjD=b*1F;!`3?RyUJoK~=XKM3PG1fpk&xe_^OAPT=fhr`*Wz);4R~^8ug}!FJ>1wMwwbpE0V1Bw z$z|A9*U&jic8tGj){aMSE7yUeEptVvCe?~87jQ5b&vSOj&JBIr}aC=i&<| z#I8ZGpDOFB({P3pfKTfQ#SC(til*^;m-LTay``?&Jx_|(J-)d+$s|RwZawri+=)AO zmhfa4!7z#ylpLy9Qs)#cx@$UDi*0E^L24*e7R@Zd3J4IKlm@gi-K&$cC^pFpNF>2o zNM#Gl-i%dx!-lwF2E4}gRcmCQ>Rlwgl6VdqE5A3Pq1o8oXt9-cW@PK!w`E0*WF8)1 za5U5M2Nxc=I-)EB#aDDk4Y8_%Ff2=By0Y!=%u`5VTNeg?Fj11>B|b58@`MzgR`plZ zdEQmDWqW#9GaQ?2^pRf14SG6AIVhAu*qjXNX0z!J*gX$z?GaN(ZY|*@_$;^kdKlO( zd(&ZuJ&rcC+%2|y%2iIA&9;K&jkJd6+pSkUw!iiRlB*_Er0hMa9LPlt ziu9cj__EDrnq{p8r#OrHC>?fu%_JFUUS-S8#$?jO*3zzTUXE4-bPRMRN6K!YI`Os& z@_NW56-)zjX9A$F!w-CFz+(Cn2gyuwnr517^s;t386HgJ(uLC=xu(InSvBA>W613> zfKv7D`QUkjc_^#sE|X;j%xXv2F;RXvpkWesn6ur!cyYTsx?UZ9x{VcKwnbXk5PRmT zRQDW$g7q%TvWu>=-1S8-WT%;j4OTi}S#y-ndh2;~^6JQJ3lnf=b^tZ{zQoOBD3TU) z*g_BNVKms+&`Rh8y52c1Gg0bQ%5?3+4yN6<&+%neQyVg}%v$B=%4umGg;hRbvD5GD zyeelT&x3-_tQzb*wkd2*4Opt5I^B`H@yxE*&}P93CU$Lvk0)#kOrC)bFz&Y3ISs09 z;Vwl46t+$jt^4s{(_^8av}|>1%#+N;1!I{~9+iCf@P?)%dCe=Sv11YvE7|WwPxUplU-fYhXzbKJP^7f=y#@^K4SDqQ#d9F5oQ;sO*Y$%G-LXJ!I!1#vvaf9(gVTIpv@q39ISJ~q-t?V_t|Zn=%KdWK22kiWaFZ*=A)3#c`-*|c)O?Cq{S=8L%0GiOuU9EP@X0M_2 z$j%N38zr?)Je16;tu8|^iJY8z{YhiUyV&thPu{rdfvvC<_w0Tk`A`w@l~%-CVha;C zdba`HlKIeR&xF+D@=nV$xA|;1fn@;I2#phBRuNx%)l8!BlFz*;YuQS_I4iJ_?9QZq ztANxvGpWVP?y=PJ;pKFMyiWU6(VJE=3jL@i@igZT$N40SgvB-xS>do#%h6;fgNky* z1kblehbA<-JJU|m@y#S?WaBAG z7kUxS#iTmdH}EHv5&&{}%REcDtlwM56HMQSCT`_@KI2f9Va$dbbUM+o=@)DC<7 zJ-=+IWk=;MlOkn&uu<$1lWt&RX4AmD7%tLA|tB-vUYrqqEIgL!Vhx?M}P4 zJ%v=E>`*u zb=YyxBD=U-DiBgzk~agXrSMV})7;)YFBg&P0MO_rC2b~h%ViB+g|;5dQ*K#O5UAQY zEp`MLS=xg>zcX;!f|)hOBd$7e6GuJg>TcL^8?H2uMt)udRTY?S%XOOx&mw zRUTZ%*1DGlr%fLh8T-5aOicRjnO;x!&Uh9_2zIvZ_5_ADew>{)QZLP@VtTzmDHb2l-7zT6Vp8}7?$5|sK{Mul4;bWFtx{^K6mq7U!4vZyv?WjY1VIq z!@hlFM125y1l0UTZ8MN$2O@G@jsytVo)x)$qI$Li+cdsS_rWtB^C#xWe zV6sqcdN?Q7t8oKE5!-HztB%&vfvuT~tIE}<=$I)lXT>5uwPF#5&9n6xgg|j|5!Qo9 zX}2ltbv<|5v76#)JOU|l3XE|X>L1^BR!H8pz@w}lE*Wzd5k2`Y+^H0TkS?|>RUQz`b7(#qTVS({@0SXPO{we zwxG@|RdZ5pPV08vU+v3jk8jEOc{&-LHrX;(`^aR?%JZhZ&iAK|w1g?_(qE-V{5Hd} zJ>J>AxwEyAV?%X1(ahbnb{Mo4LvS9FW?h10w^BK4tq;Z4lgwMX8n^qGY*BU9byQk& zaL7lK`8khMmmLkU%{sM@dwDUR)V6FMoVRhdU>zoPCX23GO_%vrNG4>FI0spkA`|es zmTxT=K;SpoobB0OCxH~(Nok(U95n+QqT_b?eIvCICP+|cZit)k@rQy!yvM9x{;pvP7$L0;I=U$GfpxC85 zPIu3IAn#;#YBb^AgJ_M=H;qH&Vskn?L3K?>3qR%)AG0(80_YJ%f663D92O#sjpqszfY+662ad05>)S-zaXyJi&H zjMW|#Ao}s#uB^9*9cYd6hZDV@s=K+&d6;7HEMWM3^rXS1)=A^AyD(h3Sj@3aI_$xr zuIrp1o4r}DYEH>=wo}Q)nJceU{8d^;`+0Rutw~BbO${0lJM%nPL$k0xo&fb?UL)sE zA{%t1PK(;@dxKOiB9G>qZND|W^iMrzJ}_(n_NueAUxzVoJ{N(pOm;5p2_?k`Lt@2e zl*oz4B$;QUFB+ zu-yYE47e~Y@Z7wvUR2i-_Sl_VSrF49JX$a{j*y~T2~pmjV-Ib!-fB>7W(CQ>KynD@ zlUBL&tAb|Q*)Hylmz_ECyUEhkz0w~*2R_5vcQ_tSdJ}ERt;5!aLpdw5$coNhu}2Ko zb&SC>Y=Ai;mlty#h#=^Q_O*a$0_AOtqwTTM7Y6+R?1nx=;T@b#kZ3$r4qbsQB6lUR zrU5Lmibh;B>ZFUR*$q6C4dk@68i@-3XaI1>eLrT&{sxqMF+BO~c?wQ4aO9ms?R?_o zf@8oJzT{CFWc58K+qz&TklU=CLaWy7H|LF{r|>99+BtI9YRv3%dOYoS;0v2&2Bu@2 zPTYX9+w9SGh6m~3Y#FXyL%5|U*v*;N8Fztbh))~>glDIAO7&GdYif*a50RqFt)j&t zF?&&{raGe1Es`Qt!6jwqx}YkXj&L5g`yQa7WgU<=8jgRu6!D#-a*|NkytO>1^Sa(I zheh`o&3jO91qZHd&^w68*0xN}rHj)NDvaY%voX~sdEPpMcL_*@$&Yu!tTpE;{_O8B zS9JT5Zy#sP7zR5b{Io~r8YG$HdfrZ><(zj0#)i_v7IkWarcu^+No_t`shz~t;!$Ch z@m$KbJH8FyJg{g414(ijGOfo^7Y6EvZ1=~dH<2<5f{RKku2~zhuz{8`SIl5?p z)^>#2GK~>R1eSj^bX{^A_=~K)O7pSM&eNmVK*%5NCkqB(bv(vk(%#wjfcNxs@7w~Y z+fzY=i^pwH@jD39Q8I)AL%F@$T5G(#RH2x2?x{P?Ry~&T(E0 zM%6V0w1Oyp(L7pP+1T&PR8v|c1}ho#{VNdCxc)Eg!kGZo-EBKt|6);O1Guq6I*s8+ zQ_}IcG1|n}S7bvOUdt&v?YeY6!e4s@CT zZF1C(g@IF}NuIkcaXNj8V>8`Eu0_7f=39dmygd-brnjYxQ%RgvQ-KF&+9+fNSNzM zKHAFitKXM@$@}6DJoV@P+*chJkNju+KK$ujU-rIt#@~$p_?lN9vovtG?~7Fk0$xA) zyF7beJj5scYhFix7ERab$Ntr7kCpsqQf7SZ*Wl92clCe%p#sk*-R!@Gcs=pcPu+a@ zhhFyDSN*l?58fF6+h;$%Fym{!mG~)wc+cx@rJLS!^Kbq;;&0#obUj?9n||fn3Fqne zyzV86Nx%8S1b4@|`B0O1_XEYGk8EDh1A5QR54@ky9~S(U?;zASzqFJ`Rgd2J-mm!^ zZ@iO#>=OUQcM$a_)#TkDAU^%hU;EhSebWbs>66a;_t*2j{EuCm=8f@>T+8{)n;%N=yyE76`_321h~51DcM=itq1Qc8rRsOz z`$DTdDEQ~Ti%_2a@JrkL4d0C#|N3o>-}FH8@+V(7`@sqMV)?q8zxUmccl=)bp7`+V zuIi1S{kXQ<>%WKi5b+03t`PCNFBJKpLelpV%~!qRx!XH^*>gAUufBr0`I+w}Uj2BX zAN^k93y9Y|_r$w=8~ut0YFhUS{j68L{a$%*eePCR?B@Uaz8AL|#LIq!pq^Zw^rbH< z@3nW|^!>!E?>zl^*HRx5_gen^-M{<;#PfFu<+=B|*1MpydHF$N_L>jj?HPa93uS!X zbMO6~-}&S_pLg@}ze{{I@p;ePmnHkV4`s>j{`TJ`zWfdmJ$L)6L^oggL&RTwbM&H% z#qW5uM}OPRXZ$_ljl}u6M-?ll4=;*B{vILzCkpxMo8P(?^4D(*d7uUMXd$;Xzt-zk zXzhoIFM9f=wYs^!t;7dDc11=%LiFz6+k>J$`Xi8qi=MmJ5})y-FMc@9A0@tm_=P7w zoW(zWp!xc;W1iK6mq9{W!6F(uu$R<2dnO zf1DG$cL|gD=rXMFh>%ma^*jfM0PvJFw>+5fB1^RyLr|t>#eJFUvYw!N)PZ3`U z<-OP6_DSpJdw&Lf(t1g?9_^F&M%m?KRs5!hRYVy-OEmA7cs+dJ-@onImy~#`nfi~u z^%d5gFS}RJea+Oz3i@9^NL*0DZG%7e$TyElUEaL^=ZGJ>^Q}*8_P@1!q2V5MqM!K? z@qKUp{{M{o`_bEKJ$PGw`0HON>HA-Q^SeJ!q))%Jkel!q2=dM^eVh|N`U}K~`2O3r z#^L7h5757GymqLk9=XK*%HDnM7YXLho&WP%;Qe`_`lZKv-7npI!!Hr9z4OaAuld1G z`sBMG`X%C9?>zmdFS)y$A zcmFT?%vN|G}?7jg+~){Da>4 z(+@BI&b#mVCqzyV?|lBYAwKDmuikO`}_ksm*H(c1>^{!fH;=jktf z=~e&x|CtzKL2+&H2Y38)53l;AcR%-kAwGDA_|p5A{j6WZWxx3Lvd`Ra=tqA2N9ZqK z_G!=j;!B>ugKqNkzefC(JH!t>|K1P(^|#-7{$}~>&;x$(>2^r}+sq>#X%# z&tFFf9B$^nMf}jikJk&_m4BPKbN$J^p8P*Pdit$!@;Co8!g<-V|MA7`bUO&#`QM2y z@#o)gD{Ojy5cnq#3;2)E-#n`Ni;vfrH#qyQQPx!+3 z#qmr3w0r9A>wb^;)mPs6^qbfGA@SxnT*vD8OYyt-g8#dY+6j(mj!(TY{*ZVJ@fG*h z!_PdBto-p8hJv@g;hvQFFFbYgJ3su&PkWVr@4R1m{^oc7ka*+WkNhj*Gl^&AH@xV5 zxV`JY^sg~S4sP%Ifn@nde)ENb<-1>x@9*4^Z$A9{#OH8-mRtP%e?$B#acA?NxW&~I zZ}E3u{s|AJtMFsp;$7V0n+W2o-f%l{eAUgT{t@vNZ~m$m-|J(GncInj_?CM&{b2D) zzHq6px_R3l6Q6y*#OvX@8u{iwCf@r0vG?vVuk7nt7<+yD&s|nXzZa9=o)`6W(EUf=#ap386E-}}7p^FGhl zIP9zR=C6G6ahZRH9O=`~?Ny@n-+uBz{pLwM7zxlcs76S z(-)Xm0Db){GdSSC?~i@u>7`u%_`Q47cm=56x~T5wK4f+9Ew2#u4?h3q?|S;)7oQL{ z{^)l-{U7hWNIv&?uaf7V{_dxref78dANOecZ+`aa*S`J@{OtMfe$zL9_w(QQ^-sU^ zv&knflAm}ilGm~3yV^H*K6(Db-}5wlFM0ku-}Cg1g_sLxJrN{`RNudhf-5 z`1wy4v;GsG|4=yFz6xjE|Le^Xdwz9yzp-%kNd>{L9?7TSi*GPq{jS$uEN`9WU;6xq z0@Ul0-}C1`zU22Rv3dUc%+r7Qt;O$!XH8grZ}!zc$v*x5_dfgc?=Gmv*7!sGllI~l z9#8M}1~vQj-|{-Uc>S5j2KA5mr{DIx`qF29_4EJDKM~*c`v1MLdikx-KSqB?=fyvM z9PBlW`Ov{Wlq&xGW4XWh`OiQ9Q{l<@xWlDC@)_~HZ+|`tPj67shY$Df{^B?P(8qo7 z*GNx4@ZOg`<%4JG>Gyry2Y>PYLFVzn?|qd3edrrszLNL5@kf62`ERpNKl=P%$xr#0 zK6)!YipTq(aymcrJD*nf8}Ue;-}=H&eMLTD>o0%TlltC^%@;oX4f%av_|O~jI@tTa z`^}^OSHJKnx$|#&@kjpo^FR1KPo4L^_xb<)-A~{D{N*2c`V#*8@BGM9AAkH8ou?oA z)}OrR_qb7C>lc6BPySm!`C~u%$W@Xb`vUIp=aSp~UcExplkU?!(|QET@3bHRwEf<% ze5KkzWw}nD|BdcbwECm>54}|mOzsOC)%(XG|H!w!`rU9{X0K1_LqB`-?eF~R=aN7E z@{zsTpz7ewz4@bGc>dRWPe1tre?A`BTmS1Dcx2Bv<>}A8_^V&ziT$S^|HS^_kH0yq zzxb(7?D@L~Pk-aPe(s%nwH-bG(r@|<_1^zeyDrv-r6$ zJaU>JAKr7CpTkMqEhPT7`t+;6vCQ;4Mo&NgO)q}wYh3LA^Wlr#y$axe`tQHF*njfs zLcY!b{;nIrCq?k*zh*tne%Gsn^L2vystxq*!LBOO_4P-Cb-n&@3Y+{i5?AE0fATw? z|NKwA_pd$wx#`mpkO#pMU$$JcXa| z(LeJuPm>qF_me+L{_NA2Ui|PEAFtK(ANbjC6#x9SpMCm^-}K`5fAJm!Ke>Kzeg4fB z^88ozgV&(vzy5(Ym(skt{BLlPeO(I%9YvpXp$+u>)t`FLeg4n>(xeXwVDG@?{^}`(Iyt^}Qc?|9^k+{11BXf97|5d;t8U&wurAiSPg7)BpdW z$;%tT{>tv&om22npa1UfdH9*^z+>(Br2$DjY#8wNMe*7Rxh zMS=cNn*JT|&-lrz`XIKwE)6{&|M|E5y61oSr{DX|=RfAW|K%6gKk(8F=;xjHf8rKh z-ZBFU*sZ$?1Xc2?L(iol@sSSv@pW7tw3Ry#paB4K)95%JK~-xoJB^g(K3{hFgH^}v z2PPfA)X@Y>eRx^A1L9H+=Md?f?su3-7&{h3qJpBwomU$WZ@7p*x8BXYtR!)7=@Req zB5zd~+2bG(&Gz^9YUCac+m%7}Y;AnlRm(|w|Jfv|m!Q3BBi~ds_h1i{k+{%Cd)pqi z@pZ24x0tiiwNtE5lVQ6(k1!oe^!go%@Qo!I1p!8B90RzClU4=tJ@!CtWjorwcO)Zxi4j zaP7zA)#Y;8W;p9p?Wsx^r0R+#p)u1!Hh&2D#&KUsT9PokH90365PD!Dr1uj)I8t|) z3#Gb-meXvd3Ku3CFmlmpnU-0&5VNch)QosrQ39mu>jUfWi@W(wuiGKAWd&81rFla1)y=E|# zvd`9ItQjivD@+cNtCQZ}P0zlOLvQRB){&}0^4@`ZrfN+Nu=*LNi{k{ND*S{!O<2ft z=mjlEYb?kK2T^{oF=0?N^h@51vP!B>CsL^@Nxt5lW~3%0NuHtC^5(kSjQUNJ>>x8C z2=)RN8mm>*ltpo6n4{P`?;-nqyI_>bpVwG4Nf$Z`(z$?=OPqG5ZofN+Cjk`a@ZPyi z6HV;^(^FkG3ZY5D6H)4WT|vLhMnt#93cu1YT4CSx^A-V36={>I8Yot zC#ObmJ<4)e$Tyf43-2d+yW;t~+Ma>e!zPNP%iW1$cwg}cr@QUM9UL=GSU7uFclCNo zK3de)u8OOoEm&1anrdR`PRs{I3bNxtF)}3wU3QzaJp3LQh9LQ%t0*#CSus|(sdPGL zsn1Phm_ke(jk<{&u9gU4ZNt4f$@7slY9%?N*TuZma(V|xEG8ZDn}geFlPps;p5yY& z5R>}uTC%HV30_Ay>p-ic#$4$R@7ARrRtv`@h;ZrWbwAg5o*Ts-{NDL|cO>DKFoCKQ zOtQ;%b7OfqsAV~TL%Ky?&Or^|4ssDRBrHyNA_(eSi&gyhD)jyGsMY$Ex_j${mab`M zb=1OE0lH8YFBfG=Ebw8FRCUW7?WkIu66~1Tr^_j(CvzU_bcI=968FA7@fF;t_%yp^ zRW_SlWMjGLZx?*gJ4I@r=*N3VOw+No2#)yg&b}k5DH$E+RW__)31hL72}Z|O6}5H} zmY(;zBRJc%oBJJSWX+~(pKK!!0~5Y4u7zd*BdemSu8^=|2}ii(58>E$=*(_;AA{jR zw+~5{sg2M=a}|Qic^X>|E$>Bwyc@S1QyUs72r_*ux#aL;!A#fvRxVcZ!Jccy242qN z9oAk`*Y}CEhkMyN4Nk3(tF2BmnAGG8AzApwp|KGC;I7TnxMuwZoo=(ik;yDROT^N| zyXm(CaXP_8yDS@vWtetWGsO(<0{-kc=qID1KN7{ZRpKsuTA;Dx7Zekg*F_K0W@U4_ z9`L43{mv2oAP8Gb6IrI&&KkC8VL^r%-`_i4hBJ*87Rb=5s$#ZlH#4=%LWcneX2S!x zJ+*SLVy&!=&w*LY;3mSjr_3w2&2ec%>g_P>Liu$zpVKV7z6Q!Egc%97)oo(ZMC4w# zNEn+4iEoeYtS;WonT~L5$H(ze!%FqbzUUv|iql@N z3`>nur=UBLPGvX%34^VJjF$#Nng~40IU@#n_v{^naT=bS`{uJbaVy(vrOhUyU`zxd z3PD(V#&)+)j}E51R%47jjZ#~>C1!hC!$nTm9>=>;=ge{O01) zwVv)S+q>UeV=28JZ6D`1%+p}O;sqS28VOr9Lpwmn#MXusVmXwHC-81!E<6-!>Ej$O zNn5``xWanXy@nL#i*kPKc*=UcjxJJFHIx9GD_N`7dp&J(lMd@XvBIRU4gK-sp_D~Z z=CGUwkVA^917q9{a<|=yatFatRahh#dNv>kSmO;ZS6LmvjcA2X%GI^AT+ASfvg{x1 z0Mb9BJHh)NUVlN1K;$ z5?UbaBkFm*-j&$r<6Z~uSC}82nC5ymVQA;na#pITfkx%v$FHmnedUkxWK%-x68?0u zT_2JEV7_gOldxgevICD@>up>_Ap-_6qyGK5jTh6yJ--{T2esuF&5T+1Ugdqzwm;bE zpgSWRaXZ!a=kp4A;Pklat(JQm){gmk=d3s^w72s{&m6fpo;G*({o7Er9le9SUF)s# z$1mEOLKCI^6*5Q%8yd}zFY7m2KbQ9HyI+v8g_*@TzxxIB@1IHSX~C&ucKA=CYG-+^Mlc?4gW@OlYZ#ozGe&ap0D z3bVe^<;q3e1tQkh+YUlJSARh+>vToBv;IUavf+%EhH&>Jp=h5lyVH0#u&3b)3b^}# zUpCjFc%78+I5nGiIoig`0W3%FL8efmSd9kR=lB9g%7`2=v`E+{X%n7Dn zq~$7H1Y5>&nVHfE8&(Q5ai8gzv2~GZ8af98bmj_1UaxDfyE~oRp+&bCK4rJNX*y#Y zDaXiHce}Z8lbfXHTY0iF#bak!FSfG47OFAjW@XflD^Azvs=OD1K^Pw9V@q1j_i?49 z47?bk^5iZy;bF{2$UR)QzxC7$Z60X82SnW7AZZ zyL2OBW|X{vY`+ys4u((L{S-Dyy6q7)$)nbP!zWMzqXk)J((u zgirGGF^5#kFtO(G-~!Rhn5bCjnUAJZ zPi*?B8(q)CqYQsN%oQ%_?($>~ncV@ZPBe_Yl`e)0w`1&tn_>)U)Iji@6BTT?;bn7PiZZK2T|4E+W(twn$aP_KYVkMGah$Y? z@=a;9>v`0$0_6v%1@;f2SdUg3^jL@WMe=5YAUw{TBOL553#B@aJDwbxz24-sIP(jc zk}|nFR1z5nCw%&3joVv9nSg@&?TH7b@hR>z5==L$k$jj{%c)eRVFjo)smx&79P1DN)bRA-5(t zgf=b~rC(}N4rG(ZCM-={*q$x0-w#5yyBY3?iN5GrY^WA6wb)F0y@0=Pr(%o&*rIY) z8#|uv3NOxhws6!V=gdyxAibFBK@&1L_PR%r9+fRo&3g3lb{(vdu|nuL%hLg49ob8wS!PhNO-O~x5G~GVYsNET!aWZ70R2R*VPL%G+ zxioq-QcV^XUS>iMJM8)Tb<~?E7}Tx1yQa0vNL^oz zbxU4O%GL{Z%`SByyTLBjb3y2>ub4qrT8S|Xh!yT+QYCbq5oKMZ>{TR@mmnh@N;M=0 z`4&lJQK=qD5<`%R)a;UO$xyK3qz&piOr&C@V6aQwdWkar1F&r;eLN_H`ahjN0W6*O6-Av*d?J5 zn>o*qq~mOR0T%A;S~dp?(n#GWHe=&R1tU{t2}{j-qfOYDDK5FY8{Vx?_N+IsRDF^q z+rn{95J)2$Y46pfNJfaf^=+QQyp0$4=9Dnop)ucj&{QaQv`Xev2;Mp}oB%Nly`-Rn zb=4-D9eG+`ld&*8uo8)x24l|3p>8~1>RPx)zIl5=X(WSn`3LASpGIjaI&n0X_o-ZEcW%GgM13Iln&zDo&{jq8`Y(rcDs zLyPCwZldKj9G=(oxaL|irGo~uoaNdmn8Ot1x{V#pi(w*4n~-q*5aP692s^D)dR*ux zX3bHqD-@HtDN7wSy!~k~+fF{3UVVE0HibWltfe`z{80J@dt?|o@H?b$+Q^%{%#ghO zdUmwm?2n+und?)sqH}9LmBb_RT2Ez&)`O3b)^fg1m}~+_dmDF*_t=nBag%n7vk2e zlKytRaDuM8-RHY18lU1c9q-tk%$P94PDmzG&d~&qq0oAB;(WgXdtUBzE9TvgeCaZa z-KGEL3tplnB%ncwEPAf>X3-$O?u2w1I-~s(zORc?mi?<_=>WhK4-0i4wFZAfq#*RL$~>Phynp_mpJ&^(pD~<{Fi)5?0?{lDo!2X{DA|Qu#1~X=;YxjkM)#RFmWk zcw;(Ko)Us-cjQn+7B=2dADkFpv(nDJJ}^`-Fr~Tec@qn2b>Nvu>pwsS7S?i?Nfp?q zo?lMyb=-F4d*uN3EM$m#i^8^y%H-A8hspeSfj-v~Lf!0Qi?rVmC=9iS$|V%x4ga!a zKW2u=R_J}ZJOoF(#MM^i__`};Vs*GJ4|gP6%;CPdoh}BYw!2IVJnmv}cbdO15>fY{ zr>E@UcxE3~=_7Q~6wzndEkCM67^?s?uIE=JT^@~Lvw+6Cqd|xNUf4V&tJ;Pca<^iz z5TOr0**_k^RyxkdvKdcrQv)TMe!tkmR6HHb`VJ{cFy4a`e1>}Oq2~wN$19pG^TsRb zQBsKq84roRy9P7x_l9vr2NknE9ID_c!TLFk!pwYmN~Za=N1@GebRaoYp7Qqau-g3AL$B<2@dk1Paf)raj0<@Pk!Sjc(3U_IfDHoL$ z_2wyaEsSPA0HqwV- zHZF7AgW$o}*^sDRPq=rt530aCsj~YVtDH%yuXcG=6=5FJmcQD*7$-(t_ z)QWIi46Y-|Xw#n7(L`gU_+b_1&9sr4SQnHfdnK3BbVm)vH0z%PIF|@FVm*YlH0=wm z2dsA#yMC4)s^k0w;3<=twO3SIfG+Q}S`9ZUZR{dd@K&dbx1V^$IpmhvWgVXI7IW8d z#E|B3IAzLec)j>=fl0k0f%lWEG4O?CME5!%(flB$UYC$%l|)(nW|CrV8bPG)HnkoM zqVY;*c;gr#^h$7N^tsUnaVagDe|4utd~OVHVhY=}y8 zBy45+fmm|Q`M6a_^+kX#;w+&clyQyBozW!aAU8z67_L3kA7CwK>L)Tj`L0xU$Gy1U69hA7bW@7FK!CMr zSKK!H$hi*Eb>$E+_;SdM<*;r@eMFtZ;dQd;$)f0GH%?g;@@C_d%W8;Nc3_kKaA?Ep zx@*N|!m7!fP9Q8!2li- z=P2mKsLCA&0)ku5p#TY0*)TZZ;F#trW>`hXg+7LRQsn)l@Ay1I$N!$>cl(f zN!lk4+tYR`pn9^8`&4aiTM61`HZk6uW?Q%_uX66}55;j!uY`8cSzEhgOnDs4iJI$} zsf#uu=10>!p3Mm!KB;axZGWLJ_h)i4yp0c`Bb^}BmMo!u$`P>6sICELZG($*DMVI> zIB>BKcd~w+1aOIpr&VLR3%I@QH=BJ^GU<8=Q~MQK8Z!x@46-yRZu=K*br|<+Wy3k%?AEBurk>WkYns-yThv5(?e^fCJAI1)s^5VhjI>!PzkO|JdR zBny?*rrwIByoI9Qcu({7&9jI|&P(U$(Gux|)P&9l{&mGS@JsY0i%6DQ1AIS^VHt$# z9X-!yv?S}}oX}lpH+$#RRm@Q)6t9JzQo{~A)k=`4@L8o3RELCK(Pp^E#dWE9 zzUaz=cZ0rfLJejRk8UcljkwS@VZqqCr@=xaC(-6G8dgvRF3f3uE@vZV3H5RUAd;0g zTRn|t+n63}&^$L!=iXpZcCE>E%$M~MegR=474m|vM581HI2HL*iWH55>#r-S2xsC7 zDcOyl!TPb+aZtP-HA|DWqI5f5I)OmLGm1y>OJes_NKrvlv>WL5NS`oCH+LJ@(KU%S z3XCEeK()PAVu*0fv-zr3N2yw)pxH9gYE^Y&!P}10sAx73#K34u}ONOo?k zbXtgoX)~vC+h&tt5~pzW$cLmcN?RuiAznqWt9}SUti zOM3hR)r|Jr&H0g}K%WRW@uLX^pAP#UNj{Fr%liL2FmC@ghxi*ZPdN4kI)j-a7H=rZnx#0|CE#Vx4sHkckSZc2FB>H_z0ZOb3Y%?jqU?_ z@ApQAToL)-Akfg8XZwc*nw$VIVwl*)6c#Jwx$UShO7|S}Zkej=wWbLd`%PxFolRo3 z0({NoyEVFpR~<&?fN)(3o$^C8+V_T!)cej2kpgf-NOoJF=$8n2sP)}!sksCimABa* z_EbGE5kMO;(1sd>N7Se81qZNK68*}c;TkV|FVe3!sn7)z6(}Nd zj`XrZ2{7)&ki!LCCa3ICXH^NCYk0`AGvB%QV?)a_j4QfrLj@bq0mbTS5%(Zcs%cR~ zGV|@U@+OA{(HNJ$cd6lF0rtWgF=Yajcqc8u@!~pT@6`{tSKKVB?Vea=3)w6|W};Du zBy3tSk+p1y0C*UJxO`xqyP#U^k{L%H?#*hrVPzT&I)gCp>@`$K5y|NJ0i9*o9-q}C zJpCSW+~EnIo~8o&~+2`x6pLIrnb1)|16?&owB(jy4Z zNlqdy#!bU)gYEjVtd<%;-f?yZoc@7Bd!L~P=N^CZ)BWnOvaw79d1h_l{}%4> zy50*=+rY-|M>5QzYGygNvDb5yo0p9*Lz5p^(wS;8083#&RU3W$i2{PlZ7unLo|Eo_ zI#WRTR#>6Bu-@ETuGkc=+cnzanaFX|2;FDd!Y>>B11I)t*%s4} zcnj$97T^{vg?e~iy6{bfJ~>H>G6-62E2L=Cjg;;g{jJf&9Bq+Fuuxt!cuAtDj$f7q zJXM)mlu}lI`Dm`&y|^$c5Tjleu1M4edFX@N{O#?s8&cyqEOQjkADI3%h&*m3NV0jT z2H`fpiChpQE4S+Sf|&fOGupXx0p98(!v&ZM-wL_Is>Itg zuk9oLbYK!DGxBtB_Sx;6W}J))7-7WKQ)$(ajqF_AyJxj1!fnv!ynFv=OX*_EPQr^j zKu>Cr6`QUxxXOdYL9=QC4bhX&dgS12TW2>Sqv_SOF2(88ZEY&mR#Q-;3X5MVmPm9a z7eRp)Br~qI5J6YOL0(?mMZf6T*39*iox65;PLh3Tc;!JJ8?f8kQ8C+~Kl0@c9Dd_i z^}rZ9ksUZiE}3)CO(ogV56A(e-UMFNQ51Bi(P@Crj!i*RIVI%oo|6q0@@>(|JD_wg zO_qW@&7cT(;<(+4lM`{-IO4{bm5UOFQ{|Ot_xZ6tSto+J+`K7&%{iD>87U_Vux%oS z)Y96H&IrMs37oz>crhlB0d-ODWXa9)C}fkUE=BUva@aSh;fR54$hOI3ZmPGrvp_tm zw%t*&jwr{=$HjyIJqiXP{Tddkq!OAYOvg2MdMKa{eLK1}?T}1l>`3A`uNEXU8*|}R zj&>2{#i?mFOVsk(?w#p)nt&Gwl9!&&jj_IBt=v|W^ERBri?+^9&#N10HawLyg*CPr z>T9ME*eM;fJC0P_?i6jmh?pr1tFT<$ndi4na2@zBhA7y*) zWJcLagbKg9ftqn1WX~0}p1Wro3vU2Bf#U@VeBOAvC}_&+U!ye2g&Kx~GRY3dr=qo> zIR_hHlU0qfG6|2D)MgA#)FCnN+SvahlwLV*d5kyb`FR6`So1Ecc~mtn3eHj&-fo9< za(zMU@Gtk+KY`zEq}+)`4q}nzv7`_7BXP*%R9nFKNOmTyH~ynYdRuGk;(#KNU2Qh*_PD<8w0*rIiOJwhsggSIL{SZSQ5_t^gE6wF zTWh$X`eI$osA`DnUIkRz%CJilA!QD3=Vi}tc8*iTF4T~HwZAErJ~%`M&#MwS>)U-v z-xtSv=s*eeL~m}*X%$T^^{8LZ==V65wPgvy?rkElq}g4>&9H;?hJ-XcpS7mxo~5CR zYK7?5uqW81hIl5+GdEC{D~f3*%qf;8rjqdID|fQR zrl!Y@c3&Ge0urUdmFUSSqp)G7K$g|>`J^|PM-YvbCIc%RjxDDr1QrdV5S@_fdoCBkG zHO>Z^t@h}`@IzFq7ijEx0&L+9^gJ_;AQpzOZkC0HjU_=<$aK5GVxqV~!EFXNx2;&# zVpC-*4|*O7+sY*)u|67fq$QQL&YliKtnHL)xJfy8R`AIvELUkDx+!%QNwjHVIEgVt zcNOWwzBH(hmwVrs_BHH80>^dq18Cf48Kg}^QcKet?P>bbwhMUjl}q(@xlkHPV=HIa zQ9~2GZB5iJ)spN$m9$$zmXtDMRFGqmKcF@lkR-gCqLRGq#Gy3qi(mvx1`5c^fR%$JXMO4we@rlTMbQCE*=S)+#H=#b}~JO>e}WifU$4}gGu=GrfFxVtyyvp zU&ED1#z%8-%I5l7HhObeALeVYqtG3_=I1R~D7bfmhtkkS%|Y7nxq7=i&<7| zYxLXQ3ou<@oldujH0n+l=L;crhtj&Y-Kc$bd9ZxF1i|bSi|)c#_o#O%R@aaZU4vhZ zC8`KEVM$7Z$udX?q{!wRg0W3|W?N^46ss>=aTZtRt&B>fMn{Jg?Q)$hLtVSAzSX)b-~bp;`=bO?()HQ9$jno;^23W_SF@|;i)*8HRYy$j#p^c^1H63r>Q z$X`+WlIk_~2mA2e>Rp0mrrqBC(pDqK=sdsr5DEe)oq!K`;lclUmpnL`nHBCeQ0?1= zhShBQ?g_eij_T`2e|NFI^-^d0pi!dt`>0M7lPM}M-8av$;@%#+nHCiY68d$xo-k-~ z@)$@sf^MB00WdheAz5a4?gFx>lqd_qz1P36m~x@rMn_i8v#sr5kV5;tt2 zm(W!?sa;Lt=>i$L4d7M6`43qku`%jL>)g%fcePk{f++@feCPU(L8%t?5r~1Est2Md zTDs-5c-h;MM(b;wG-2eDk%#V+LumkD^tq0kX=@?-BG7<70HJxP6aXwjC!@b4? z4&dK%gXZz0mUGYwM}ZV@qc5?cc31VI|GPhk%pK!19=1380mae#_gqxUofvaLD|l}S z>y1KJA7leblH-F3t36o(Pf<H%>T^>1sGT2LVPnXv`1M{V>Y~G-+9N z+Tl$H%*y#h1*e}1#)Vf(0vyaMW-;Y&xMy52fRwt^=S`(RCwNuWjx;qN{kz?Dpk2zE zI=VWn$S5>&`BO71Y_!l1(hA1S4BEgNrVHS_3!A%Pf`+^Pp!@?Uih3Oc;Z?+VMA;0Q z%g{h?ZK_|6MnA{?8lxu@-Dvt{)7hzK_a5!cV2X0Nzt8=kM+`O@D4hOjLDQJ4SQ*HU z4xRg~5Ww`P-|2wKY^+Y%v_LC-p5I4mxPrV?$qqb2R-9Kb=kDbNpw-Ja3H^oF3GYh zlud2!g;L}+I20@E?(8_v!FbK5a=>lqH8F&Lx3ZdFZaw`hV3#lDG6zewI;QD@hs$B! zO&CfKlrT7=-jkUYloD&2)HmY4q6G}PhTA&Pb zzCYMqd!sWkAlju2`Z?1^3_N=<-VxiZGV+?}md`lFIMUmGB^IJE-0!v;i8ZhRuR}8C zWIoL2W^bamiPXx@Jcr{v`Umu~-d;@ctXau&C?gR&N0Kt0J4$?t*%EQUZ?u6KbsT0L zZxi@FSIHfOI?8T|mDx4H;S06PNgc$uY9h`DTol5oWkxli>0O4W6R5t?v7Fd7Ojy>b zygb94v_G%}tux@oY0ZjrFf7VU=@&RvZ$7NxOPegIIp?++%z<9qJlt*yBbmj3E)*DiZmMe~M(Li(``F+RON;~*yYi5jr zvtwy07it8s$wo3}`=qnc+fG|hoB9?!Rl|rdF@#1Yphi8mLY9aST_MfyV-zMdZ|l`c z^aa3UlEh4@Q($5arPy^~9+VIyiHm@_&+veuC^J*N#UiK!?8gg+)USyrGlEw08;eXN zW6UiLan>h^zSl6y$hSrjCHDz*KPhXnFvl?Sa-BY8_V@A%lS{de9emL%Z-+?fON5u4 zi4|AdM2k8MWoiwjE5!ETrSP}O!-y>t?UA#)23{}SmaPE&S=gs>oqAnb@|jh#g6WB2 z@c4o#^thY)eAY3y*av6? zQI&hShk~O4}9+>YE-w~bcU6#N@7>2;hlWB4=N=`afTv-Z?7qcGVDRzhGENs>IOs>H9-oj#JnnrO^ z)l`mYf(|e|5;m4?c!RyN7WNxm7n+`>^*3ZVp7Ps9q`>;74<@DNci&-<3qFvrQ!AzyMt;HohEwR z=!1AfhMc*r`f+F(H;IV75y$;>0o0N(br7bmq;n-H7EPqZhMFylC z`-vZCccd($oTlolRoh(PR(`?&v$(sj4eo@}ErB#jBoFt$FHgv{9b}oB0-O+L$Ab3x z+d`TlyeXcW^G*NL*KzCn`FphcZxo*3j1Z%KQmqmltO5GM@?AyB} z>SR1-zTAaQZOEg32qu$9V`kMV;`h&`qk*LufZ<(Kd;bk z+nJW@uM6WJR^SJ|HF$>ZcW;l0iR)Xfm{%<5!La`pf_`-(%IftL-A24UNW;53d_KH} z&*tSdqJ1Qvsn*NZ?(W^9#O^-v_7U{_aTnq3-DUIvVfkgFd(@uY!B?g)g z)fnArBdPYcBJ^sCHq;iwfP?#B3%zGHXvi{|U+<{S?1EM*$V9{rJ((-Nr_pbD2o1o$ z^!k{HH}92h?%o-BwC39ku6AbbfrwTw8jnAMw0=MhSlgqhaO%NLvxQ^RxqHQEOgCq5 zaXyf9tH8jV5Xs7TauAg*agxd%}a<~(x4OeasI^ z-f=L%EFFjkQj}Bc3A>PmWOkNOGP$m3gIM2Z{@3Va=L=6yAtumbswH;?7^;!(U38>^ z^1>2>E}IH8IiuZ4_j$s@o$JsS7lYH;S73DC-GxK=A{pEa9h)AB`TLGi)~B=6V2p0M zi1TjeBy4m(JBt=*_-RIoViEeV@Bepp~TIJo=Oj-1d2o6(-MBBq17`XL!f8D z5@`NFc1(d#iIJcAVjSDS7L_(ma#U<|o$i_FVK!X}!TZJaTlQW$2G@sbLB7GX+nwr; z`GGYXu4DxE0Mq6)bW4Vfd+hcx?R|+wFy1dgal-f{h*pF|A(Hn=L0%pxdNl$tx(a$b z^Zolaf1bh&z3p{2a?`a1xhD2#2~G@aO3(qPU76dEU&~}Wl*lRiN=WaR%ZX9_j8OfQ z8LLR&*|WKxt2q@nYi~MBE_-d}B66v5$u@YxriE1&8)(Chqa(5%vT`x553lVpw}CZX zrcOx|ycY*?HrMV$?CW56<}kc4>w-};KHn-fL#f84PUZXrhhcO#R~MGeK=QwY(#SZ0 zKNF4zb_1o0^B#`T&~BmI4U5vso1?(n$bhqoscyE}@&l zc2RuH8yQ^Qa$}Z*c(n`{M`E~|N$VS_FT+bKS*h;!MJL8lyYva7#mZ(aQbYVo62zY47sIKdVbNRSt7xtCm!?|zAmg;c*MZ}pyd1;m1<^0Sp z=(9_aqjEqCzI`!FR+sT_jfgCaqGY;Q2euY&1tV!Mm=595*(ZJWz!NN~kUSCgOx#`Q zPn3k3&5BbUP}>ytaZcuK_mgz7rsFZ0rE^}aL+W%gJ3XR_8i!~W<@(P5(C%UsgM5fv z0=lpX?ANunxDBgWfm*GSKxN1{GT=R-m`TGSu@5qYEE^t1DE17O%=Knzc_=;JHji70 z`t5LIJnEI0w*oBC~jfXalp07Kx zZRUd4Ia|tTFzV{Zu+s(z9A{DXyyb?BTsrL@uV}N#K0(|COXP@X?y-N$Eho(48rZ#A z4_lWkpy1pxYxb_1kSxUDw4ICLH3k25D`S+hzrieV7dy%MQ9!20R5m5YD!bV$I@KH| zHWyF^C+<;13)+K#GP_B(+~jg4xZskiDu3Q6j$s}rFlCm>F)f2h&W_}e96c&xam2}4 z&B-pc487A-P|kE8^^Z(D-WZT;ph_Wah+C?o)70B2C{wkz*D=uvpe3Y+Zx7ea!%I!!EQ%x8q~W-`(kEO9yS(r zCT>Z0lObY}w}8mk_pzEl!{zZLXmNW=^GZlxRiVnS?tgzXTGuP@bRc)Kd6jO10`U0% zN{X?*1NbTCXfk}A6TZ=X;I9t-4oCE$9bRcc+}@tY`S^a$o$>lk!^tft-QGvrY_kwe z`3YKHxY)N;FeaV1e)ml~C*O=TH@!iv;$xVt%a2VUw~r>&V?OW$1&fbA>5pT(&JS_% zZo3cs)0b0>@AUi34_}FDV>%K&Lerh*=yrX)L`P!1)oy#;IUCOwg9Dpe=gYfy<uP>D$Dh43_gRkYQ2bC@f4est+pcM7)rWz> z_|+KJ2l^hL0<%yO`v_BdwrtP)QkMpqPIqUf_!0FtXn?! z8rqp9n9jBnSDy&1swVu2obPE^n?u;fvt+qzx7yZ;h@RinhZ<8b+gvHfPK|>ts?+=| zF`Z>N4Z!Lft&KPdRZUxo;@(?ETrafxm!XAuK#R9-$XXh%Ge_r<%%kZX0sK7lI+4f5 zB{;##3Exe45SWV`rEr}j4b50IM~eZ;sXMDDl(eJ_4E)PWBR!Ei^8uwd@eC?lq#I@` z#YdJ5I&~e64t+_zcDbr$NR60Q$@&u2hvltK)R+r9oI5Dyj26_Vnar}n#bMz;3t0v@ z=DGtl3`dYsEWC`B1_6iJT!v7`3-F2RXtbp*kAxzT-%!)WB+-E+4vnB($$`U3+nVF- z0k>JS)BX&ih+FWME;Mq0aWwUmM%~;7gr9L61;fD-i4iWjJ$v1RoEkk)92e_eZSzzD z?WAc?iuIKjNC|2Fs_Mp4*y`QwnM}|YmYpg()FIXpXEZb$Gm1cj3`FTdhCFBKWen|# z!r%0li((up*k9uL^t6=XQmVlJJgLbLGj~8jeDx2W+#xn!%(%1*c1yY70y>V$TooKW zuH){gh)U0O#SKuo4kF5kE9m`JCFdkLCw-Wo_ul`0__1@6s^@v1cYoh!`95++IR6b) zFu~T!c3XMi6VCJ%wvtP?9DtINz@H_PoL^%vQ;)e2zs&&Z1HTS<|?Yn7Pgu& z7~}vM)HuDpjx(1WB%`i?L0NTGqWxsX5D3uOP*vfNGsNYXRZk6{V~uI{%4oZ8b2o)GPwuvw2OUk+kgHNM z0|ouBn;XzDV$h5&+^)o%mMV@e0L&{@x-MrxnL(IfgOxQji1V^|bSa&-Rtvg~BG?QC ztT{MEDAX#1N!3F&l}bIV7lEzfutZC>l%=W(A#AA4)Pi9{%G_{P&DypPV1Z6cQEl87Epapxio&VR-%q7D(d*30POPL*rw4C%Bk1v9@Yo%ZCZI6^w> z;Sy`M)&dmg3}o4I0VbQYfKsFhrwHrAx|v_P##=&HK=%k2Z|(edt7LrMkipqACLXrq2*CRpbrEle25I zL8}**4@w?LcWptqhV^Q_dI_$223Dc&CI1t$PYw zuN#&%9q=7@QsKUNZh)j^wX%fQ)6GP%5+^ua)9&fh zMvsh6$=mU;--SS4%7)dkft!~-DQn2mIt9h)xHCVp?bJh%a`d^NL@yxd!-Yr0nV{*g0n(G zqr%N79N?hU=+FtlWw#g*)O7)QG@xiSRLU+L7#d?TnI*t#oVIccVJ~K`7cnOoSz2jj z(=tYg)AY6iSz6XkCn$pg#$BCU3GH`Aq4&o*Eqs%GFt*nzb0*s<#UgRF&feEJb5)wI zX_O(G9ahG)+mow?9m*&`>SWDKCf5i?TxwZ68sRefVWE?%W3KjL;0!e|_cN=Grs*c2u+oX>L<8tUXn2n_K2vX zQ0q1eDtUwUcsXT;)pav2Rt@u0D_831G}OQm3AZ!DPLuPyS-YCjDSd%ZO`Z<`1Sd_2 zQ7@bjJ(LlacU{;4AYoxe>?vDeHc`yy+if`X*rRq^DpFFWfm6!Z0AZ1*05++ll8mXV z)pUdyC7N~NU=)VqlH7`2T9)9HiCCam)^b#*P#9`00G^FximG%_=Z(}90{CQ7CIlJ9 z^ZL-#bxI2zWSZ`!DT7&Sv{>pcD%FiPoAL+%ow5;w78iRN&9p0yXhK!Wq!_;@YH4OW z1_$daAOyW*C{mQck3e>pr<2nqwfb7AYw-*zzo%qq8bgUXlR*fO&z|M+QuTHTUX%I@2*8yjZDDzbhCoy zyMu_SWJSP(r2+^t&IT$*0nsRKf)8C)TeGJtL9E277zAe%HcchMgc1Lpgcn^19sxmE zJ4yiZY*+j_uQKC;x4f%lQVA+nS6eoOr<}@FLI)|L?Lo8&C`Re3~}rp1ckXPh7$aux+-O+ z+eq5%tj$;{1?p7<1tpdCNRYz)K@AL5FKo0)I2cA<=B!sx(?*5@Q3!81uATUt z4p*H{{Ye3AXuU766}sZHTQY6H5$J4y@60y?Ksbt&3f);)&RKWdv3_8P`0fq9GVGJ>5K0Ypd+E^jaO`I(W>u1?w z-)KyZO$QHetjSbG0L3rZcGn553FUwpVMxxX?}mbYZ&iQSamo$BjjA^Jz*`)6*)XNb zs?5qLrS}wZ%g*`??yf-*N!c?Eg~^>V)bzE2T<`Ar;~Gytg^eznk!kZ$SumY1+3a+s z73UIqOVX2Ts%3hG%%0Q9?bI;dH&Pb1mTUsEPUeoXYTVgV26@F6$I>RRG$JacrNzk& z5lU`>g;eKcq1-UI4KijxC)ixWW$&f6a`=E5OO?2#Rgbrujk+HJP?mz6_U3(hlfPv! zq*9hNuCX$MyXdo&2UfZn>1zNq+!z8Zvg`@P`-MV-RT;8zPr&Rk^|J+E+%}LYdYPU| zlh7s7DLLGwu`v~th|a)m(v^?6d^MT62H+bHQ*)*{CY7tH=!P+0wmaLNp03kjlx+a*1X>CWI(d&Z;y9aiXsDv6rxtWT8m0x;Mk{Q_wT7aJWEe6_Jghc%>ke0r#iv9_Y;uMy(5L z5O>u9(1{t^G(_eJkb*eS%V=QLM%(^yLvD67G`UK&GjtaM7MU_=ZMO$HGuPC^Ivc9H z`~#)g9|dkKVrLo(MFV)AjbtFJP9(cjo3%v%=~xJKRT>*t^e`0#E>xj@s~?MWQb5{_ z(kaFgiv}`)cnL6S0RuE=Ysr+h1ct1lUjPW9ORwMz6@{TbRHl>NGRr1Zje2;1;(EST zFU~LWHyl9n+N=)a^8rMQ9Pta59^s*^224o^&poHYY6gOlzQl zTa?!#>USD3li3l>r2TMq%A1T;l|HP?X4<+gg5NC-%F>G(EeT_8=%OiQtD)iac8|=# zs&z2Sm|LNYJqwmyeS46_m8f4=VK4S%J)K}5*vF}8u;lJHtI)kxV@Xfj2doG-@xFdQ>;YjyfP6j@Q*0dpQO;TJYtB(M2DqA$Rb%OB>WqPfT zRz-~l!SW6SJB2S`VX7{#uAfmf>^?Q+tik)eN3pq96AG7*Rozm^5TKu{#1O*(%x4{t zP8el_tuP$hY$4>*q8Svl8n)1f(KQBl)ELHVC&=Im;z?F|rLHCn^j1Tsm$%R{&9IfI zn4FP3j z^_x2Yu8J7*IupDX?Is+|Q^3Q-9v;?dyJ^h7im3?@dUZL~ZkyDMp^iw2u%%RQ+dIP& zTTB4Y#R?4vT;>$GE34Y-tQi1!zPLu-EEZgKkGls9dOwoabX@tgr=JmY0%cA`^_;FQ zldE{EW_7LPu(-PQ0#zKj=X!(MY$K#GH=2C|V^Zgm=A^3LNgHKKg~l1Iat1@ckt#Rs zLwTB_vt67?5w%h+YYVPI7ia-UzrgvRE3QTm@84B2*ni3iP1f61q@qcK7l;^&fZBuT z=t`p~%LA=TU+qE%PbGjH*eO}8(N^DD?P-#pX54}}!wf@*9U<18@^Xl4&3beNPq7qH z17UZV0^Spy@8%t}2a)RoY9QCz<>0hP=(YyO_-h*q%~;n}R};YxxykZ`GrsHqH~ArYJq z^4CK^*clEAWs3_SBY>>7^?Z(mzh2G2K%7XoLpHcLz=J0u?`R`#b=Tl3aCA3gGP#Rd ztqYEv#e`q(ilzb?mAo2F7i2(Oss!mk$l5TlstiyZ;A2}?Te7m00pMn#K|fW`#Oo~> zClm^wsh!bl<<23K324)mo}r{{s(SJU!m~p(z*rM{TEyf~7kEa>dQqd6 zt2B+8r_@Wvs+kszU}VlQ$T#wIW`HOhSZTY;K+T*kCKYMG<29fTjw7s7OIccG#$L6+ z=`+*k8ftA8*XaS57{V#u&~e8j+FBrpNsEj*Oww=4uyuQfjmomkn1xNYI7>Hy7*}u0 z-wxGFUTsuk>2=g?SpWf;ig?z9Al-H$?D27iMmm)aq|uID>=U)TK9N>v8mUm3qboM2 zs-DT`Ouhhst~@~#2c07`bl23q9^E%|lT00%TspJ{E~Jj6HPi0;=xEE*pjAj}R6-_) z4viFqn)ahLt|FJ4_lAhSx9ED(Hdo?gEU3?ZdB%_y^gL5?o@JlG@^<#`Boz0 z2^I>S0Fa@mR4if%Smab+8M9PuLwzeoTm4YI>k*{}ZHLZwh~p5@ z+SRRav8Kwj+z2TvbOVE!Ru^tq1Du(rm5FJ;Z&u-<64fp;VPLsn=|;gJ?+-OU-3_W$E=yf&* zjf}?`<;>+w$t{P2Y9*Tj3dXQ(p$z3}gQ{edKq`dSdQ@kips%;611&(70XsWw80bSy zd#>b0;-mpk=NM=-N^~KG$Oc)vR_Sya@Hf}Oq)nctcMhoiHg;l z`4rI0iovX*5}=e^C>?KAHTIO+H*~r49z#m*%p);JFQ|1Q*G-V2BUUSbN0WV4n*tE> z05ez_KrReV1(j)r8=7#3QZsTaHdzH71MrU~=jtFCOMs1;m@p#D-3W?M27Sq3ub%VU zV{j%cv@@K;E4L#=l45CdfDY6&I8+T;&O^(r9vfp1sskJhT%B5plH~?2xY0Xnv^;M0 z*MNoOf{mo3h2B33XG=x0lxRoexd5wl>x^!r4{qQ(Hpw}$2T}k4FE&$cm-!M-vCMzGm8Cg&S1b zeqBz@^kkhDD|fW11YkKV?zC)ZLAa$+Q$yIIO5s$tLgiX1=9V|zEh7emD;36*EDug;le=Eix_B`yooqhM(I{nlbJr?Z*4sjLu=H= zfWF|*Mu9NUY=wGsY>>36@`H-JX9bu9JY9!9q?w62m5mB#SIhck4b}1`8lD;lzuC4& z2_py18M$(av6%-JR+j`4QiK6yB-5|LzYfXl;;Ba8X;a2I%RrxqMsnF8CT&w+Obt?S zWJGVrlmLLRR16UW0CUh4*~>qK#qI zYgDLk5Rt?Uz@<{^cE)2#RddKbl2%y!g<8}BxA~}Jh?I40HypoM_z`h3C0u?CWj6q# zV|ficYcWKv)_AGvZCHX#HfToRp@NE=vszCgYVjr*Ll~MqeW#jYQeBfN5AaHa&6(@4 zOj)wH2vF@oM#ChXo?M5G=sL1Yr*BYtT~*W-NURi)G>3hyvJ>uf-KIgv6ssonI*!tn z0k=V0I-ay0A%h--2g)deM@5J1)dmwtcmh#%*RD;a)BTPf$qQO$j}0*7xc5ck09B3| zX@?@+LvUVO0r*;jxtq5d3-VGPHeW>X6bu?JZ^d;i?yhN4&_5dtc6TgY4k}OiGztT&oC5$JJxPNv$?F8&B3QVipMpoVK*J5iBW@+{?1 zrezVv2=7aU(vuA3VbnvIR2JY7X{xP=#;|>U|C9xh4IzuNi}WA ztHR|>68?sm+??OiX*4ZB)(!dqIqxloGrnY!O4t<%#DgRnAsTl{19sAtUbREN%+SanR^K^B#iV~~n*9k`Nl)^4*mL|h@G)Tz^@TENy- zm6T}eLfzL1rnx?{p9T6}&JnE1ygoC7lue zIUh^jvY_>xKEqgkpDXDtd#IN;9?0Cb-W0x8RGal#=o1|4rY5*lV?h{sbj=(@wG z>Sr4cHeS%mf+c?nn7xr+#TgFgEjo&^Aquu-k&j%f)_F&*A|%JMDIvE{$y9Au*ql9I zv2~MzqdII_0JTF1w$F*^8-E2lUPoW;0-TnrIE;luJ*HBr6l2WUe@Xzi&rF0WRxe!YWh9>VnR|ssa2No$=RKG32~u;06>0r8{l))ja?p(#Txqf;Wp;K3UWYi%PFwhjC;z z6P950h{7f!M{qmKA=Tml$DRCuW`Zp-x&be+SMyI z0Hz$QHZ4UeW%K)3wo$AIo~^MIOeQlG0Zg9* zpk2ZTulA~tORni*8{KTS*~TS=q8h^QH9Pztfn4Cud9@~ZD?2NIyobLx8*ORf$!Un& zuwdgd*`sX<=!=@Zz74shsD`bXEk%pbdZ1m8F|A<9X)lx;F-sy6rK(Vv1kLd0~OXEbVUH*k1>=k_;8x zR6NP$oW-DUX4-ALsBWJ^&TDma8C9(wR^^7Xwi%y~F_aP; zOe#r$o@{_p5&SkMi6+L`Fl>lDo6wu8NafpAH%FDsjw7E|y&{E7d&5!ovUN*n4e^;{ ztiR|Opk|ur{{xM&mhIQPquih+K>bIy9_Q+2=Q<@;6M~M=CPzI zL<6K?lC>kEqm~@N#_o=}EyldK7?p~QEgs1OeG?@mQ-VgrW>9cFs)OKF0G0MI1GNI6 zaX*Wc$FMBJ@Qcie2%Bn#6YhSZR;Rrf;51D0DQ-xW8Vm)^3yO`4kekU=t5$>O1oBQs zQVp9I2Lz=?i%D%YI0sENAiOIurMA2|VRaM$N|D~d*9Br5r%#3*rCwCgRM^XLWOb-E zfDeeR3U=R2r&2O0kzul)!=_O~>$OTak!1$M4gwk23XIJiM>}7vah!r-dS;u?0ms5( zOI!;A16CD77un+>m_I{e&PQJYLj-B8VINfjDIh82bw-{?m=scU$F9j*;aoqmkR#-u zPL*Tu4FVQ+KVdS}4Oq77c43`J&_f|bE`kO!0=xpEsobg*oCW%f!2~o?VVUGOfDI}| zWGtB1<6_87EJYWQwi1_IDiWzQ1v__3>?%_)9f8{!^i<9k?vi#I;TmpO<@#I^xsjnL zfFd`K48(P$s{mG**BAr-LJpcr1mCn1MR|f#C*x{0mIicFnhxqyQ5p$j3JBwJ`1H1t zTxSoJ!q_2WVp2)O>S=(4fIS^YuRyZ~MiHd#RdQuH@MjCK@p$s+FO?nnIJ#Mw$)Kxu zI-#6Tg*^5i0-Y_IA^a19Zm2%w@wNBTVOrig2HFk7&ps94bLwy=S4FZl3uz`XwV{jCP z0(QPaLsbt1JIICy{C(tW?kWI&Sqr%{kzkOj0I7g-0RFq?j6~`oQGbUy9PU_zG}?Wc zFn`q;xvH2r&;li-)YGwCd!T?9a!;mJBIjKjExs2+pg_PCJS~8vffNA(2mk=XYcW3z`gvanL1>{Muz%uCVeu(3 z;(`Pa;Yc#sY_^0%)W}?mG>P4qbAU`Ydf1CgMcgm2NKqEQzpv>DK}|SlsN*Fxwb6nJ z!8q{Fu!cOlozB1|SDOsLd1V~%*>02#xgOla2T5&8Oq(ty_SIByYIO@h7>Ez6JF;D+ zS{TK(G!razJRvLX!<0ZLl-nIu(uy!7f3|AX4+aI;#kxvWO?cPm!=cY)2_x66BZ^2l z#Sq)6;73v$r9-FfjbI#y2s4x_vU~-uJjsf~PrHJ04VQ{$nsK+_|gpz?jPZr4Luacr*iTL>aXdTqgBNTXI@d?M(j zSogMsz=M(l8-_B4tVb##Xsl*~c`RZ=&0^*v8x`PoOEy^={yj_qYkd+R20-RT+9Rc{ z)#>CB^HQv!@FWWN<-=@iNIL@1BIw*0vC#xJ#E`P8H5e+|H1MBVSs}tD8KjYD!fGf0 zcM*Q9nxGe%x&b}}Yn8g*?k8g2Vgg-*fYo}bf!P3Pj(%TA%S2_6OTz8(=(K`c2VJ#U z8K+x2g76KGf$4@;R>ZxYc(ZJ!LbQ}_#`BEud9W+|uKV!)h zDEiypvR;4g>Gq3G7w*{ctY^KvnV9-BU3edUE8Hdg^Dn|br}598qqm$sQg}c9dPU&_ z`1j=_0{*6`M1A4*cJ4^wB!0~uwLR3{`mEfsoGiD$aM1z2{$^q?@yQ+h{U1DM=hlA| z-q>Gw)&c(RA$iF;h0pKcAG}9;`PS6v_xmmse!qy9e}BO%c{Tr%?=QT8c=8JV?q4k) z0~h@IvS;wECYe zc!<4cr#&c~ZVwURQHkRYdU{e_*Yc^mZ#m3=upuG0PyKA?Jr6EivD(yA4>p#Y`d?k}{K`TD0xIe+pO3)>|186JO~l(~_|AAhlM%hp$qc)V)FW4q+7Yvc6gGn}R# zY~ZxA{OH1c65`IW90wkQ9B-eYK4-#qBlXG07OWED<6{ZP$*U_fJfZU+C^W zv%Vk1k@xiD3l#CdOy@67$~d#k7k|C*5(1yXB;gPI2Eu=1rr@c!Hx_*IHw)UW&pIe} z#9zi$*g9eVXOYvc@$8v;@Ph*w0+;Xl;)ZFpI{DxH#6m%`bDzL@YVzl>(8PYbaFEyh z7DM{r!Bg?jQOOzp>fbF~$p7_;g{`|!V=c@HE6%ow5c~L_{C44bV&B12ckewU*~j1f zWl&7z-Dc^%tE!k|PWM#orwv_Y=v3;tDaIG8ath5Ie|g z|FrPYrHx$pZGT>P0ny%saC*6oiH$&j-HxXgQoA>>6Nk)8pIW$kCl3f6^x>D2@02Y4Y(xL>*AQft_|**Gos%y4*+KqS1lgEJ2!Sr-Z=!b`kx2Mk z7RW1ijXQLf`pI)h%~`0ml3gO<=K zLB2xr9bWGD&9Z z_<{?_0)e`UQ+Mjo2S$y%y0uI_x$e;P)-p9tMzek8u4j|qk}O?*>4pJ0nYGl6Ko?CK z{qjrsTNWjkpM_d0(TRjR$Fp{taOE4j$gG3_1FB@x<+H!sUs5q>{M)QeoG`2k*F${OB_<9gBABCQM&^ z5t&(9U%{M}|MNwpT=GtV@F@R*KTEeB%0+Xr-1WkCZowgMLV>%@QTwyRd%(&4BqzCN zlftHVKGZ;L`Kgu~4~X7#$ft;MD1*o4ikmy(AjGM&0(-%kFQ zmy-64a>;q?Wn@%x^@!KJ3iCg`3`1+Jej+dLVKTUp*R*hvO@zDp@->FLCOtVdR&G)3 z&`PSCa5#1ambMA$+GX=SSCBVi&p3PYnvamBab4yY{pb-=C3({(91=&#*RN}hE02)} z=Sd@0W| zU3PfGJWwny?&RMgCzmBxilwZs<5y0m{bjsFL0-73ytwS}sF0}_R_`xO1^Il*0po^3 z=B(tdQI*Dd>y?R)sWk^PaQDd z#n&|CtBJb~^V*LQ^mE2*e&L}tMhWY=$UzmX1RawIg0I*J|B;sy*lzA&ml^ei$k+?IsXMCsa`5wKD!d+ zKF4o2k*}5folw}(h1{8wF8*se(zRzzxv*!#7)-7-lQ(Q&f^GFJW{jOvGgX}-T)DtP zZjsERJZ2@2O5S~TO8yO2@?Yl>UT-7yl8XWuO9Aqm^9Xl5vi(ZQRU+Z)WSgnu z)MR^Y5Uu$SLGmzhX4DP*4};?3De8)z4w7#m4*0K_9g*{NMYuXoSMWKL&hTHca$c0Y zR>i1s3AAUZWAn=LNuHZlZeDd7YJ&OO{ z^U2$n?%l}pA8*2({hu>6Q%zXjdjv+e2UxN`)*!Wvz5=@OddUlcx-;vbT_XF zw^}uSio9maUj;e#7oHZ1Jua%n`LCqbapr%LB40&pJ2JP;67OHf_{7$WkBB_?9}$(~ zdmG=HdSi7o7T(1_FGGHaI69x}OP-pd5lk{Cwq8Br%8j`4AIPqk@=ICre!@JT-<`H? z>w5OKJo&t>(GkzE$WyVWKK0T>(3RKb$-4+5J)iUE-Z(|X-|>)i7yqsj>D$^Jv2KsW zrONBYeaqjIA0u8mpXozCpJpmpr1)Hg{QA-_Jp4hQ{weqKq%viwi_5ODgitUpZ#+%hY)X4vm5Fa@rZdM;T!oR&i z7M4GTL=U%WaV=;H7;hnFy!iNl-ly0?p~ z8=6F%%=zBB(XsO49?2{a*B+fa=LpqWhptZM=e&q)?sgrW9UcWyrGCeYu-*jMKRV32 z|BjRq&z})->4r7N!K3^q|Bie!flr7i+g%uI*1RSIiW+Yf|8gkMizMZ9hcdQ)rXDQ;(+4>?^Oih5Qsj+&RBx4!v}(0%(~# zUPiue>%R(u>@VDNbhH}L@0)sKGRb*yI(}w8SKa+nG$RA?r~Uy7?xk;TtmL&XC*8#N zW-96X!y5TFkMiGmIq8{4C|H(>Uv5J9$ybnXn@hO3$Pj@EDp8jy=rn@ z;%|Bt`I@DD$JULS!@PLlh1usMdFASXH!!KhKK_=Iq?XusY}8RxPi8s_L--QO%F)-5 zyU!u6IyQG8Cl2gbU75xU&p+Qvo=*_AW5U9-WZNj&FHezuNm!I@^+eY&tue73q8vVBtg1G5K#JubjsVLa6wYg!DQ52XBMh!qc~r=Sluy6V^_i zeD{Hkm6^Vej}JC)+JsB>jidvTjk@u%H%<<2JKjXPB};d&Yb;zPDsP4^?BfDU@hTC$ z?TPI)f76@DD<;>8&ZP0~KE|isLI&nB`s!QA5KObHjJVM*Q@Wj87B`q#CU3o+ypsSk z?OVyK*Vcqv-%7q}%P$2H_ZR+CV04`Kyp6oz+NlwfhBL^jQ8FjZE#3&Xlfs7I3e~Bf zR^0C4;q!-a6GFrLay*up9-(TpEsTs-D5=WnA`CEeW#M`ED=jzXV^H1BBR?hnXncQ~GMpvBpJXajCtht^zSMM4-;Y`(LNjBqz zvt{gr^9TNeL-N1|9~>-~5B`?8h`;?4kr?=cx06RtE&p&~b@II) zE<$6cpQ~m3MV}%+O5oG``bEW_$&;y0ei#3%&yjB-Zp1}?j?U%IPpolU-PjI%nzU}I z3i9kPl*QutTAtjx{h{@5u6!1r7?J}o*dY4UG&{M6fABhD4}bY*$rF;7ZK%w@PME$x z#^3f?EYus>ja-2EXCM~0GB6RhvMIhBo8p7>o8og@&s%NS@zVIj=gHSgmOitgemr!C zjN~gD1Ue&MXblG)(?rb87y5|4!aBkGS$>XnG!h2I5D*jQ!^`qqg9! zUs-?Dx%n&Py%J)}@wp@7q3~KIFd}ZbmweySzT+DPr+BhBcw9KSt?pM({M{PiKK|2R zBVR=9J1!n}`RaY-xwC4(ecvHJ%3t$!^52Q8=1W1mdyU!ZxGaAIDhu0jKK{+ARqpO@ zk}oEr$Hh}~^f({BA9wxeMsfe?e)0;UD2lr}kM`WRRssmc(SuKM^h`aP89n_oUGxG@Ve-symM>ipahtD3q_e%`yZ-`Ruo#Hz)sxE$vgb(1tiMb2tOMbLA zFjlMLf1E6C*>ggie7pGiPsfU%y{{)PSe^AV{LlL}`Iv+_G@t*S$ES#R?-QhDi*m%8 zo??BEf9kE#c*Y^?T`RwMg8Yz#xNbh{&kd%C#A_G-*nddt63>Y7i!Wc0?zv;?l?mV5 z{y@Hq_k0T~iSqBr_YybF=l{@`r-(=VhrcHkTk0eEtD|DxVcI6;|MU0j_Kc7GfecHC z7tLo(?0xp6AxDfq^GEW+Ew3Fhe)UwtEL}VG>ZAhv(La&*5pS8#{!n^~X2kxcKf@j8 zE`j}&b_n|O>JDjK@~=X9YYK4s) zPop94n=g>=Lu)L@OV?e0A>Vw?(hoNZgsamf{2%`UJ@jh`sZR3q8PR;-U*q$`6a3OS zQq4RTH=HAVi{y`+5ONFBo97YUwIF3A+xhzm=~`pVmMB;ow~VcgKjeS&Q}R**a->P= zRrA>0O-eYRu1#!*($hp#2?VgA+b}|q(ME-5kqmuSWjrLD0j^IJh2qjzl53e&cA^oOv+?wFEqby<7qdD36){us4x z4vf4?&zCwSpP3Qz#>rUw7`1Z#1=2l|r47X62?VA5R430+^51{9G)H`l8c8+v2v)6I zmZZlB;xS6xh99H&`*%rO_&R#^=1KXZ3hd(FxLbNaLh_OerI(IAcJhz!mM#%+Fc<5b zkSwiVR&G8hWwxA@JohxtQ5TOW#gCo5`#I9>TMo<78p4zMh0@*pH+Jo~Kr;UOZJR&- zw?m+=UNWW@B^ndXKH0!`$E6_r(Bd4znDFEXE@RW8iu8BxlP*E7=~8N!NVlP0mYI5q zFMn$E%F!+E8ZkA_!O>ok?eHgST&e%a~?7 z+uBFg%{CZzmM7h;<|P+no{Aqk`KOjKPt^+k?rV1_w!7C6&i1ZzL@0jD_O6rv?8U3o zbo}>p@6LHV-G^&@MJBUPHt?P9-D$#ut21=;<2I)Pu6u_U1lK)YyHC1nUKdZ(@1E)6 z*(cLoJp1G|LT`?z@9dN5ex7|Y)6cVyru%vJ$#g$E72>M8_wT2>3#$_jP{PrazvOc1 zBgDOmxrgf4cc#1eH;nXUOAnx>)*Ze?Q@~FZ(|5)@-?~QlfP&wrl9~v73adN+siV?a z@1&J~R!O%?h^Oa^aL*UkIIV7g7qZgBThFJ(6?O+s-$G>4$B(b$%D;w{-b?JK=ho}Y z($4ie%;z-HOSfJ*Vk#Rk-TM$X_1dJ#R#cpHuY};{^M2yyDWcJ-^}|<5pIEXAyk}3X zqAA2jPd6O#@-=EJ%^RL8{S|>vaa){vgqDDt-T&NL9?4rA$-9G`5RcnxeqWQ$e z)6M2aO9o&6)3%-bb8nZN&;P=E&Y_j}xTLT2|6jCdOx4|lzL*nMe4LpdRy-cID>uGV znwBip#|<~-VR_Aaq(_MXGri6v8fz_GXZYv+lk{!7HxLQ~{?sksHWRNNk&egvENT1Q z(f~q2@0Vk?gVW>hmU=ef^uA3vednL0fwM~gZ<|Q}p7%(VLfPV!d#&kjn{q3T_e!M_ zT;MjJbKjpGxT|YR<4tSL`JHj+on>vg<}RsY>z_w+?vD&Fu}ZI2Zn{F8Wuoztf00J` zNA=R>vo8?mD(B9!*FUl*?4t7jWR^*RvYd{WPZ#V9d?PE&_Frzik=>(OlPLTD zr2qI|-Yva^9vf@_)_$6j@|OAilzoPNI`$!HWb3$%^KPBeZ=C}MkXK(p^AC;!IJb~JY z;a}H%L8>7BEVS0aH5iD8-nZ5YPbm4vJ}N!7AiRI-!ClLI?eyZpj;R+OSFSidCe;#4 zyHulvW^KA1=0$_-#r&;zNhc5!Kb`R2rQ)}L5{ji=Dp9Y5uQz;B`u@pRDEoHv#!pFK zJ^5*5XW9O5(%YwgdvMo|-TcGGqGOq8L4j_GwElgD*Pla z3p=JN@wjqy19b`XAl=V@_Y2Z1<}rHq7oi{i;$|ZJ-4~@t#&r-r z1jX%-zbM^}zr_aIJ5}uClX>;7m218veI7wPqMCb$d*c4JVYu4LKl`%u_NCu%Y~@p5 zk=lu8si)6B#2?qFf3M<~zbf_3BNV!a*trSepT8Xp1x12CcsdTZSPbWGfn)$ zuS+i^BynEyVd(+>9X}#>tvA;$E$>;~{!lf+?zOck)M|gBt{!)tN4t;lroD zCw=ZbLNRjiYv02x`Qawi>hHrqvntKhLi&>LORdDOHzAxhEWBSm8my)Jrv|HF4%~V! zD|YJ^cGRt@Nej%37_@#|LhNDZPRho)d#6T==+G@#1h*a;@xGiD4!nxX6jN_aI(FN{ z742l;Tt5CII3C{cBk4N{<@~~4ADLn^8b7cPF7xFd!H)OlA4?NU2BFwloeQim7xJ#3 zNO$sA{RETS$xg2dcMjJ$8rYG|5TD}Ye*ObcFikgvU^9H2$jle*o_DV?!mNR}d}eEJ z)C8ST^YZ(Dw4Pu1Vd)2mm(AyQ=C{-QMChjKPjQl2I=!)mYay1zTW2cw(Efd^eRG=S zpZb~9HIG^xitV3cD1LAg>L34Hdigx+y^ml~`urw@uY5#$34!9pMQKW+>6Tsj%_CC1 zg!sYyrg%cPR*msmyYkjwNUxME{c^LWcydN2qOiu}mn`pnRC+UkPqCk`dw6}n{OM8Y zZzSjO@gGVz__8eBy2b~xz~kq-t;-s57G0tdXA%GXN7k`hx&JY#RU&~y)f|5GkEi*G z)90m+OCLG#T!HD#IG)-LR)KAdn(3jelU}&m+t1bTKmWD#g#_D}0U5-e{`h0-yJqos(xyc6;kixFc)?mFMon;H>%Wh> z`imNIQQLFp)C-eEZN>Em7>*_XF_-D%c*Iysl*4T{CLN=zcf`ZLQDADW^V zZ49hr@rtbxPMn(|ZnVP@4@|u_=`sHOlEwQaskyQ&5&P#h(vOM7`y{Qo!W=p}MIr_u z?p|2@ri6GIC*Esb#_`*vi|?5FifAF%(#r%bvB#e;rw=Uj@pM*J}*gm1hpuTJCVtD%U2EF3>=Ptf;%Xdc2_AQ}$2Tr|s zurRvjaTDS*+{&-cT|7+?505JKFvq`V$Kv}YUqzRWljkkIe)5xWS=+d_et z2hGj}DKxeGm3tTa!(~&u(aNP8NTicKzT8cp+pMJH#pQfu`PeaEK9x-$JGR_NXB+8e zGheMNA8A#W6X|7m^e5ZRW;%s0rE2%+ayOSx=0KG!wNvTla{Sy}zm{&);*EGY-AXr> zFT0@>FDFv*%Z@L{OU>%?h)ue8J^};sl~q0`PV*n!yLjX7;oM3e*th5=UOrQ4^Q6+l ztN8!1Z}D1UD7NOV^(+7@py!)=7n$d*ukpk-KYnA+hBt6|8L<_(V@{N|uC>w>_0*3d z!JLj;jhr}1ifwyZ2&c9*(Nk+yE2t+71$3@?gx6z7p+s;D#Lr3Xr{|F)@Xa$AXfVjF+Q+sKPPa_Qnti^WTq z1G)6FI3T9Gv72w@mgCFoCemmcEnl-NKcP7HrkQWaQs$I2w=Fz!)m+<-?SYVH=27w(Zh+$%z7>6Pa!$H{3%NiS=Z^a)0;VL4W=x^Y=SpWu{g4aXp%3ysuLN^ZsmwNB)@ zk!`zRC&oHIYc}RR-g9D?Aal%3ivO;`nD|AFuih@%&wEcSzGfIS%S%RY$fvBhnN&KG z4RI-kqGYL9vklbjpw+4~7NQ|KXvzDvtWU3~=hcOvDio_nBI=R?$tYS8dn0V@D8g(c zu4vT!btYu;X}uwxIUMl$;<}tco=Y?KexskUxicO%%eHHMojq*RXKd-fkZ$K?o%{gV z`ZO9<#aXLF%9)B2VEX2$FK-q8=Np*Wc30~&`b@Qqx2hF>v=p;CB&Ra=w9eY-hb(>& z-Z6S=T5Z;vcjUA2LEBN+D@rk6P#Li*1}fTP^qF-gT`1&gw)-kizeWvH3JvE?E2S-=)PPaj zFKPv0(|WB*tAiL`Sy8LUPi?zy{DY>bwTtiQ=!Q@0F;qO1EH*@bg^H8*m<=Z6EC>X( zLgG!Iv1ZrToRxqKX$4eVD`V#B@q8l^u{d2pbKrVx z^Wx?Q~Q!!rFqYegP8TRKuyMUczH78ojJmU<`T zFQUeEV7>dTtXtKI)1|5z=?a})doR{h>;G|F!(tnQV3KiRyzn} zJW6EuX~;bNzC6{}XcU9Cr&4xD25NQ0mQJTbosP$0XB72TsoAjmb1F4*u4`#GZDNL& zOs3;Y%PU9@ksBmb6{HFX_>GK9mQKcVMqe?V@Ky6pUAj_IX;QtxfU?xo)HP!lFkG}CG=s9f~Jw&vZbX`ei_@!BxxGi)su!$y%}pFB^#0>Dr87MASUo& zkt(Io9#S%Au;Vv!9uFOs^}_+Z)2K}OI;N()&zXZEODP;pwJ1x$;7r*DeGOL&r8e*R1R;@ zDCP8%@VSySw}2eWMjil%cd3=AH^VH)zhu4mT44n#tzowAU$QKXKXBnx-lh!ju=LtP6<&m0Bfop%`>XFKj`&LuCK)xIL8&6>i9s zWoOG_QfgxMK)h23R8-2mJ5tp9Rl~HbucjHMX~ZJY(}zmYa!-lmrCgfoG^;Gj^sEt6 zuj1)Ynt?46oUxTi#5uMek5B<8k)yksHj~pA3nghvxQysO3^3m-Qt8x~6X-kV$${6Gc z$yx1g494&v)2W4!5=h;TYErb;jEP@BZWaTU{!|I+=&~kDkIQJy31{2pK-aZ{^0>!h z%tgaUtJ<_FwS!Cnd4@W5xhp-0b3Pim#^eF5wHtTEVnbh56V)d@nUsdA6pM5)=~1yd zWMat}dzpwV+NF9(E#syN9IMh8k$5)imlc9`uUhMmxy#K8MMYXFWGIpqYdu*3Sx21Z zTCI-Rp*I((1X7p5>!#7h zG3ySLV@#sOB&kR!L}U51Wi@GJfr~Y}S&yNgOv}_Nof4^+BeuR1Ifnw-P_`elmK>(O zvS_cPrVZq1>0qD3vhBecY1F2|=-7V_Nzz#i z{o3Q}%XT>{{(F2?M5_Y!4ryAy-YL414op2A#%8FwvAlL3eZDZC`WN5PvM5Mn;R2;` zCz(Uai%))LHviom6yf#kVl$z{bhyFKZXOz zS6X}4zfGMRR_U609*zmA{yVT#8GC1BaGVCSL}icnpUOZj;z>=+U` ziyp)^8BXgEEl8vtk0>NxY7weI;Oxe^wCZezZ_j!eb6g=fMy~N%Gu$I^i@bjC4|Dyi zia)8j+ z7Eu+S>ubzW-^kVp$cLKM8S@DxYHQJ<7a&1{beGrKUuF@@aRisY!cBm>%$1*~kmSyL zT)u}x#J4&TomOH@m0koa)nGkkj9e*udUy>A+q1DAs=pjN!~U>L&|t?p_C9o-m}t(l z_}~%-(WFO;-_8$D>HEZ-`uhX%?H8*gWN5y=!9xHbBM(rQ&~msl&gIS! z{p(FuQRgE(pn``GjQ8b|lAL{x{6>4_+gsGi5muL>h@)TOuWJviEhPdZ@A$BCMAuoB9FU$Y=ce{R1a$k*r1V?nr{XoJ{j!TY7o- z(ZR=M(HXex{zAq3t%%aS@uiV^Z09l}liT_HIDKw?fLk8KNO6OTo>g$Za2}c-Z2rPN&WqySvb`)1$Tb0(ah?qT?_oZT zQId0votniHZ&N7LXjz_brr}_pdTA*7Z>J&r-m*+)+^aH|^B=am zI!l>eoq~*L*jw*Vr$%WC$Xw%Uspl7cev4Pelpuk#Z+u?(fiVT{?4;IsCCSNfWrZts z@@`kXReWlUs1>=9IF|YL1IFg`IecR_nu{P0#4{ z4X^Y{C01fTclUPXZhc&w3IvM6S%1|d9N?roip}x`_>8au-S6N7WNL9O*EYK)sZ-YC z@%Wyz*Z5o*5xtqSwRRZG%CN-?nd~wtntWb=tKi-nNg`;6KvjSmwvD1KML}ZGd9-w4 zJy`sSUEQNuk&kxgwTOr{W+&;!yh}_;N$o;TN!q+r%e#01;c;om%+qpve|ar_ zh)D;Fb&)|2KLevywBR7*gtLrg_bkz(Kh*2d(R-Vp7_5cWhI~U1n)t1hWuKSip4qVF zH;-O-%$GFc$e!&Alj?HN2g#43ApdFrKrGSpcNS=pF!#i-OCO8agXiHb85L~9>)(|>7XEBB_7o&6 zRJJPZYbGq>tR>@3sB0<0*FGz9o@2D-G$UPyymT^@$dC~2yF}RDOL1kfg>xO>*(s#F zdr^7)imS$t3DO4O4YE`Hc7>a4Zl4d7$BTfkaE?7~kZDNov(+fdHr=6)BxIzr@p9^6)0STgI!nAj9q*|zsa{mhC1DTW{F zerah(=;o+4oalf2>qREsj3g4Uj$}AMo{H!iN-%xsw8lua|d|iPG7@1YgNF#t60# z)0wDz`%7Mm^L$#-?v=)k=w9R78b7}HLx(Lg-n}+PXpZ@p}U>u@J%)cAPrG(34AD=iCvS_0xwJu_KOT_I*?Hv==;G$&&zsR}tqzf%p8_JWoX-x}g)Ez*_6?QK**1BUOQxQjL!! zoPog-t9){rTCzH|ehX`DzrE|nzbX?okU}Ly-2*md0#<3QoISYw6bXbb8c-L_3(73p zC&!E32B3;4y1pkRKZ|Am1BqeJvR4F_WWI)2@mh?*zbQ|B5^{h-l5TP47(4&c+sx-S zA%$7?h)2^I}!_&@#NYrCSY#;4soRwmo z_jlYr24!2f$Bjr(F)RR3UGIX5_84)>GK2ea%z)+$g4aVvAF#q5L-F&SG7p>a{b}ju zjj*Y>`mG7q+xc$Ti7cax4csutnfj4hJN1B_x1PMt5I&Spn1m2T8DWMsv!2vmE}m9P zkg6+Ec)a6BVv2PsvgD%c z^|XG<*E(j)GZ)yAwyFJlr!Y8?pZ2r;V=hPibKKNo1in6ufuC9T8}@e7EWiZtotuu-A0-ckcaJqDGGq z3q{MwcYuKYntyaEYd`@1Eyc+CG zeDiUwZaKZMNXFcLB~#C>1V6Z}#KpWJoSCrR@}jszjl0qZlY}B+Sbcz4Y(BHuH8l1@ z-|jl_i&3;@_B9dU|3KqRm+G*hi$Z1ibG3!R z5a)}U6%Pnf+{&o^lKG$A>fZFTL>B|4=s1PA17guK-~=yF$L@dk(D!aDiE=xcrXt+c zu2T(&d%$dCMVC5Dr3m$F)VDXvRlGCMX4|WT1AC0;OB3M@lN+J`T*5h~PPN9^TSqM` zaC{Nu<4hd$AIt4Zf!t%+1Ytf?^wQYw<v+8uw=BR8{L>nX<=6(Y=lKxH`45kZ3nfH%wYr>Bf;QhvusSOp7n66J#vz9# zwjZM&yJbSkQbo8v>L~6~+so7Y;smN8{++hviT9KzvRxIT*_$_y#Eh^B_5nN56?nzP z0_p;SRwi~o#mCExkz^haIwN_dP%DW7s{Mr}7F{k2Pl|@&XIJ&KnX!JgQi3bD7X?)# z#1*!0n&Gp}@op-jl6;fg>a3A%7)uo=nc>^t%Ib$nWnNU7zH+%znJs}3QDh3N0Y_1x zqES?qivq#Q6I5heDT>Vz+B1S61=<0%)5=S@kf%aK(Ue2qJ1^izIoy7QkiZ&qX0uCd@;b)RrE5+KbH*8H;L3lv}RFaLLuZ6A*lX37LG zl)x3@XxE2ju{Nj7zhKN8GYeD-q_R6$ve}7J#mpOJ8p|#mXloj~Dh1KkPhusv2HU~v zqF*q*z#Xk4*cWWQ$Sy#7~OWv%hN<5+@S(rmX0L< z-X|Tye?RAtZ4`@zs%N@~u;u&q4%6{QSindEJor^J4ED7BIfY}0_FR^gF)BV*Z;RgA ziG)S(p&dHiNA_coW~ zo%2RtC!$ zsEo#G#695jSbLZf{KI{#oh=Gd;xi7SYtB|o`4V}-cFZFxXx3!jCOzN$LH zs(6IuXpcvecmrtzBj2#Q+ZwjDeo~xEY-Y zI?4n1+_S(tF*e+oNVw2?hTkW+_i}?COC3t`NdR>VHnPJ722D%v7ZCfkLQBT3f5C$G z-FlB3T+jE)bF3-{cYD9P^-ISVjj)xdk<;$UTrFkXoX44-L4VA0_{WgDy*jl={1AF^ zseTGr8v~Zatuo{g9>DJB((cigSi%D+dKT0l<`o8 z9(}qsrZ`|on9g%=UMqU9*EGj)5{Sys-`SJQ@~>zff-DoKpMY2cGw};XmB1o0Bg)6) z%yIA6WS>?Wg335%F@W*XpzF@@9+);G7OEm-n9AZ1H`1RphpQ+O39ME;>~+x)W*mY`4`OJ4eraZ@NB~ zbbVjeQZ^Y;vsG#)XYTXH`0oeuzGBo;&0_Njd*6*ulo}p;dSH}Y7xf+K@8fnqb9P;k zrwe`-EGB=Oe~4H2tx>OMim@DwX|yR3PcuC+o3x&_w;D3<1qhHQb}raATWPHLNA=2s zL>ix`x0UTJ7Uyr|B(L)&b3*MRI#P$#NTy#~|LnwbkCy2Nzv%)XbE=PlUs~wmjX^vB z1U2eAR}6_ep3$6obL%s=Z(#PQwI)@x^fZs83bYA-+6p{Sue|#_XJ?#WLN1k>Rb>X0gj}!NP<*J@!^AlGWBkGHSiT4N z#oe)@9G-H1lb%5)QWh6>l8ZiEdGn0{s%HucUK*i1`v8Um*zzO@)%W`oEye4>=eIBZ zg#1sPg3t@=)uF66>ONw7^NumQDPw#qZvqh1@W&$ep%7@=y~98}J#gi6zvrO#X8X;% zmisx}DDG*dUOHotMTMVD$MC40r7i}S^t(3N#OY+N zxPnDS3iq0!9_E?2wD8gucS!>yr00LWAnijfvsi`4fd}%afAt7sA9aNVY;>*l2zP=# zde}7?KoY`DK{_dJJl}TFisz#i@tIghfdqo{{i^h46s8Uyv^{!_frJYS8uyrJ5;Aet7+FLsqe>={tg&cLqh3>_*3fv*K4fnD=H_v z$=XQZ&ZtHlUZH53W^Q)XzZYqHwdxPW3U+7_m#cleW7wgdwVoW3*Qd%p=&wfH}n810|RP9p)~=w<$bxU z=M+ewl-M;pdUW%pUa9Aty6IHUF>#cARa#ts0Vk(rN`cGusby6pbr z5F>jbi*RQd_awif;wA2ney>uE(7a02uws%Ij;PxIPP{H+T+oLbSy30Ik*(H_7MW}- zA>F^ZKT}BSkyfA6Su6iL?aDEZ5Lh@fK)CK>x0iue;cYM^aP_(f@;1bWtDULvBQNPE zAW5{g+ccdX%)MlFN8qm#PML{Zm9OJtMy~tL?6o_Sv(a+%#jT~`aekIiS1&h)-28$H8sIL$yS#}hV zQS-HTd+K>(%oq2}lg~Z+T*24|Wq0%jz5D}2^=;c2<>CY{DSwM1I#_g|0qpqucirE= zYkHy`U~@21>|sL)4m^o2j8A%(qlTnSFI_X0`u-BZ?91a7WHh6GdBeo?4T<4IYm;Sl z)~N?j2kKK(Li4D{eq&o|{rAWnOWa6RLl9LnPFH16WZI6v<$~nhmC3G(JI$=6w9mf? zTd|TU>#J#kut4i4Tr+ya^77mh2YUW;c<#1NtbhcO~n$fk{- zq~ilY%MCp?&_nK@(QwvIUi}%7Jnw<0+{+gVgfIg{?`c*<1~AphW^39abyx0rlKSBR zQvtgl*fly+^}qhXS*spf9ae!SMBM@31M9dSO~G6KjeK_9JV)7h)5~MdS)_R-AQwYc zR~3e$u1V()r2a!aUI72xfFV0W*mVzEx0NRlbJa?|d!u`cPGb5%142pxEew?7QzZUe zc3&IpF+$`vH~f3fK^RG$Je*k;@`h~ zgr(XU!}GztcptIx`_SD+-UTFGU(kw0(vAh|#c_E!!t^BAPOLUdIKh-RnP-n2w&o-$@!eXuv&kBL)lF5isWFo;o3fAhtJMfKtud+-_Jq^Cck73 zqxaqH55I{0Q>|p6#~)3~M_~F3ip$=oQ+6hl;}poz-ICrrZ5#vos~f*iI37ceARGt^ z#7-vji}NNq6}4l0zn{lx0}byMoTghqJX34($VDI$-~qR?7~ZA;u z4*bGN`B79O1LdM(E`-FYt<`ABdkr}&Y&d__NkXwK@mX%i= z8*a?Lbj&O)F>%{{IC>~Z?q*|rqgj=jA6#xd3;ZJPg_2e<2yL6QG!KCtFecZiv;=b5 z16P-EqPV;EAFyaS-a3h$1gh=F@a{&nYQG>@NpI@w_58jC)Ek@#YqF1@xwt;B2m&l zQob^LJ=QB{WMMpF8j+cTvVe;kKt2&XFOhP3_>WAiUz1LCTCLOp)k3|X__3Sax$mcw zX!#bom%+Y-t;~1fLdW`g?@LHAywzT~Glbe~{{_S2y#||+~&Vry=CyRwBD9+ zup7>3UES2+V+ra_kvQt(;)jL9VTQ_uOx|9q3gMX!O(63`a00%-K2-$!=U%zL((Ex2 zTyrVV`z;<87hAE6U}ZKpdrfyfbisPGMNIGZ5Nx%gQV>#_?;EIJ|B;IW(3&J!N4uG7 zKL`AkBc`M|kqt32cSuh2HcxvjiSwxd#+0#z1xnP5x&AYv|2)AdkY~wBg4gy$&~n^j zv-9w=vRw1p*a2<>M}-^dax9>TYlZ+$j9whRgL<=)bP%aId+m|9s>qDcMew|Xp-gy~ zxM{-MauZOeais(bzq9q^@j6K~VOIA-&l5$T=B4gej7@oVb91b{W6n4EnqWufE#Fc^pslA`Nf2E4?P7Zz~P0lRh^LB zrIS3|uEc!$uxR@D864wk7S=QKHt0PQFQWa(X@vC#!DOTL962G#LAvHsmzB;iCIKTK zd>;HcZko%vBEeLbfh4+NRt zl&Y}`Q$lL=wtOD0EIZXraDCmRqSHjvdCbVcmxjv%>5u2ZBP~x5SuT4y$h*G5pFuyy zepCWJ+R{&+jiv91j6^;)I5xZi^nM05gV*AFvT0Yu)F4>%gp zEflo}`5~V5HKXAhgU@uD=09-~RI?q4ZB?$}*0ez9RdkRA+zh2>-Wc_v(1+@cY!>3} zm)0-!=b8r}o$034`r~SGDCtiM?P+oK7RkP#(*@j5`3rdOqDL8I1!8!(R)yT+&k4bn zIufrYQaE8h3~?6O{q=<|EXu~feBcNQyow11m1b0k7oFIA$Z9>e3G+bBkMCGLI(}lz zHqJr5%R!V>5>Ys_P+E0B%jFA1qkJa?oA%1B?2(>ezm@Lv8p94J5f;V5nLkncb^Jv#zj7b0c=BUYmn znAgqfrIx~?Cjd|ks21z=I+bWKgoraSfaKTc1}^3Im+1U>rg&N7ANU~As%Jk>Z1dpp z^7eF-W(-dq$O7SWg>!OAjMwg1flz3pGnkjJE2uM|)$Duy z84{&>algB3jmB$udCxIGj#+)H!rotlY7jH#hLLt7(^{d8mj}^Bvfi9zs&V=PT z_yJ{K^~+csVj}^&8a~_Omx+*7m2Bd@!o9@ z$PmOZh?iSVm7tiy$K)z48ckf0OSemH4I^c? zra%Chi9(U!uF`j!IU0sJtWy18z%7BPsR}FC3b3CE86Br5{IUXIvM{V`bk#rz=_H7k zk5BL#z-VyZxD4bwkSgwe{nG{$22VQb#t31QB|SB8dRFkc?8=6y1r1oM$wL#pIq$9JWsF_~nYnK6Lx+UE^l zUjB{OeJU=A*z%rPK{7^NyDIV81=E+eG*DDV8Z?5?E zyO0i4#J~`l9U^N|3U7t$oU!*IwPmb8FO5wqa`6F2FmLftr+PV%*+vmAgRi@UQ;r{b z)1AOZmbi6HFxsQbkk&6{K*rnI(50%S_3r>CaKgg(C&x_ZcoER!^J}J3`umT-WKerJ zqdopmaxQ~s<3?AkaszS*MdaG!G^uB%0_wpjJVxyD=7d4NKf#=$q(!l@lG6fMliV)l4=JdjVmw`UUWZ3y^cFy2o zIK}s@2hiov4347ST-truzg|)KgI;Rc(}-Y!bCt)-!2TBCt&^EDE_#U=b@$GZMk+OLKdt>X*i=sDYU z0l9$<&u9F=QrlJq+43yeifD9zB~f^hvwYqP4^V7jDC1TaW*N*BBExiHq7u zR7y}XURs6g?J7Th`JYt7FbwKX@B!QHMgS-98pD1XE*@w+V2GAc78=!%47j@2sMRiz zpP*(+HR_O1&gIKr_uB#D`dX4*+eluQta75e-0gc=A4_Ff;-ooYU3?_^#qF4*{W6`6~aFf+RwP$+^xYeQ%v zudyIf>xLpLxI7nV2!q3^`qFCpP(N4_0OITS3PQ7ZDiZGPG)%Fe$J3yQC1v3o4$`q5>P8|@Cs03w4*mgpiG1JX z9}+ZqC~1$Ad9wc4V&_dwqPf}?@5saT+! zoaBxq*IE$Bk`4HRjcc?W8k2J4Qg7D@D3FgD%4_B{6aof=O`uNDESBWFQq@p!l;&d> z%iB#w)G{%b`1_3IbAn^1HtmqQiCJ^*qsMDUKa)HsZb>^WBwH4oy?zG_#x2s%>q7GY zCyp$!6ubk^izH9UM#4t}Q+k$%Dj<3P3b5`(bDi5A!+Kd|%W<+By#F2Y@?VP=&83FW z*5APRzr;a7XGmUP#vl)hJxwEDL6?*Q@hV$!I;y~{WsB02j4G$@(gZoNRADbf1m7%K zF54;yK%hL|r15Zn{R8_&YJBzuZ<=|Dn5e$W7&=jsUCa%mL7jm6Mr2F8=D|?f$LCz1 zG&wV)9bJysx%iAxh)E3oEK@R8a9Y6+jG}}S1P+k@AbJohk!(UC-P?m1NIW&cA2qNX z>R!PtAwj$3l*D#uSlyMB-%rD3*TDt{n$D0COT5Ta!)^U#@MqUN&P!xyGY&Kr1}*_5p0rj!y<@cLQ}61eZVdyI3&a0mrZL!un`i} z;vtsr_o;r~0g9xOW*x$ej)=ojP&$gx>U1ro}%8}{5CQ_i9r1kQ?_0ca7 z=U)CjS3(KyGiT`c1^j#Bb5xsq`V3cvhJ>xkm&x$%mt{9a9_Brc+HJ9z9__wI}S@>g}WgB@V4dkqK1 zen8;50psIN8snhILK(Kx|!0`Di`_j?Z2HS1Gsg7oeaS)KmckIA6UjXsL&IQP*Apont@#<~nFI0Sf9P zGa6E&~^6(}Pk&B-yN2PG9nQ#aMP!{^|XsYZ z2i(BWb{ilFzm(Zrb`W(Q_!;RfkU zT9XY&r9Oso{nEuyH`>$^>|GcKUet_6?e`p4^cCWhD zYY#fvCrSaVJREM2zm?G#)u!DMk)mIpImd5b zXxUMm65T}ty81tm%^_L=Hbb50NrEK*8u$7!Pw?4IjlJ$wb+*^r5s{h~-4q0rP1jYK zcE6skHyA9@HNj-7&8nReqjEuZ>(1LK3n5lME^Q@smHqixNh2QUa`2!|8?0!r--6H?M&lRD^4py>s+Nk7d=hGM6^ z89?q67xukBVNQsJpKiC?=q}e7E%u$jo{w{?$z})WxKl;+J5*T>gwgp|@3Z-cL*7?w z!>xE2OJL3U9K+C{XAT&w+6a))b0mx zk0Ke2cJY2az!G@AVE9U70bCbYZOX`=q{ItB&(!y6Fqvo8`bL-6=LH?@chevflQED((YQ=)c z?U3_`7=y{IMn-Z(>-<+&cv?7`@b?8BF1@!E%cTKtdD+lNss5y~4n*U)`!QO8Iu{+> zy~E#ohPerKx_(;XeQl$TIQ<>MGwrpjDxkVkEE3amv^!`YZ9|Q-S`yET2&zr+)^r3p zJx`Y0-_M0tA)jRVVwH*kKvcx|t*9tRZ3q}cL3Sq)1m>EMWHSxKRj9E>ZiCrde^R>Z z22c4E-YH;gF5~!?R)x@XICb@vxP4GC){NdS=7av7Z(ton`vOMATqyqY>}NOxJK8*< zwJuhs`{nDMKdWsoQE%85dJ9n{BOPz5d6B0i?9NLU=w-v7J;_HO?0Qb2GkPOsUV+U1 z8nd)ZXsjTJNP8Lj_JRht3lcbN$nGZB9`~5Q7&A)H<(j|%l={3MZ&y*t`rh-HYo20< z0skE((_@K(nGbsMQZaI+Q0+`;H|ZZL*zRwCJ$;S>ue)4VorIk3l`w=zfiN*8)$>Kr z-S#QfOIfB83`mNR0_Xt6_d1iYVF-{!Bme;0hS^jb8%Amt>H$G^2CLrz&^O-W0t!^%8aVaiB2ZI@4{6+xNh0dq*7ayqjetVik zX}XHmikrfn&oGz?E^0E(OTO%EK3EO^9)QbAaM{6L@mMlk(M z=T9G6!DO?Mv>_8rbG5;l)fuJlbQgzjWiSpet5boWE9I73Ds5jFPa;Op6B2M&(Q;eC zY|LXWk|d%3re6H29}A_&y+HH%Pe}e4KDH{apL_WqGXmL?*|Is-d0=1ma={tFW63QlC8S|Qh>mFxPO69-tJQ#=nUu!{F4 zw_&tAJc1O8hI4ET zi9vspkLD7EUyi)gN_HZc--1w+M_hEaFZ3M-HW;92k>-%^xj&YLWnj2U2)&oSPaoJkf(z)fSorc_^m^2Pt*kIE1KM-a ze-;dw>|{c}3P5%STRbCA<5a5|>2TBAWX?XvID~@-TkdtGT8{b)GC8+sqj|bP%K@34 zz+WHfsl#NX@FhR0H8(U^ssGJvDtJu!$!9ktzx~WXryI;(l6uO9Qcr%D3SNLv zn0!GB2p~=z>1vQSbY{cnRGv-|Cvq_EBvay!u6_e1mKUQm@4R0chSJZ#vemDaiNj|V zc~B$DKRq{a@4sx|t^U#6Twe*?ic)VG&)CBil+z~d;Q;ZEUuWv&esPl7et-47a4** zrI45vY!sC2`rT+UmHz0i^*Dmj_tg0w#=U`zyY;0n(d`y&w+GH&WJ`JWz$1#C znha+b;66JWPgx=+;W$0c;m+SER7Zh%-3DDb_ZZpRI$pDRM>Z%{Sjs!x>FnbaHxy`B z97+9WcU26vo9FX zceZ7F_RhW*vL#VxHErKb!IM5l0ve2Kf3I(So`0n`Yp#WXO;Q+i5-4EBXbVrY-_Bl4 zCVjwW+RtpM&A8cuhE_)nJA<#ZzRG+yC4R~vP^Ipt)`odU-B5e>D?Cb=QxR6|J>T*H z^rP#~Kic6?j$zvmzGun$>c0C0G*c>gW!Kwf6Fl{u+@6wKrzXaZm%CP+B3ZbAyz^Fr zRbx4_I5Yxy!^jJ9**)ERseWKw^n4*^>`b+f%!;;4fiIj+!}Wkt3t~vJ*+3pblF8Et zXjihUv>`POin(<9_aYb&p0@n;sxawio9=MIznL%wP$#&M1hMEK^s(93AiX__bLpJ{ z!bXl%i9f65qgFxserS*MXY75%hM1ZMAd&cV_HO}yg)~>mo_s6HYp6gLygPUTQy4+n zKz~MMYyU!`OKwoNk=ZUR589}eX`lRN#|>`iB5n%2nfYv zE+6gi2Rn#vVo$%WQ9CnjUN`0jO-+c}yvT(C``_OWbmE)l!bbQ0!X z>n9M)pHlPu`PBXX8kQV#Kczc#hp&3y<)SzB->c(whWGKYi^x0C?+BDNcyKSfc);X| zVD#n-Ls#8X!_h#(e8k~w{W;^jdltoLHdfVHE_%2^EvYJq7nu){Xi2{Cch1R{s`n)aeln} zHH*OPPoT`)GXu>HqSi1TJh{vJz{~cTKek>-5!{ubS2zbZoQFUhm1g$MLFYvGx(W}j zM|JfSaBmoGhmfXCF&jxe!{zIet|iTYOh>ou_u8T$Myk>4{7`GIcxGa_q4}{)B77N& zY0hx$-S8p`b>yEKch5oZow6{|HP3)aD<^1Mfi?|fec5mHVDlK*DQL)nrm4>hN7t*K znhccGFAa=B+tp2c>8lI2?sDVuUK}*PHy7g#?USbujSHXyS40~!hSJPswi>j(IMMhzPe~v0`lqppq`1L1albP{+#EL__ zcf{ybM8=beQG`64)X>%tu+fg&b$Teu;PO$mi)$)PrYyE9>N`*pg(p#|RrOA|V5JL>Bn> z)^mAPeXLSXBUYWdn8{?SI+?>aI;g0I9q`;u7$KI+1_}+OcnrSW`D~yZ$!SuPYHie~ zbdH;Ks_}YUTrwF`#q))?BqEG_@aFdn9+>wBx8AZqnscZxu?jm>Ila`F? zygFSLQ7&@`ZyOvoWU3}H0})-ONmF*zIqcPW8t@1$ z%47vZf3XZtjES=*heQccMm{&5H8_J9=xSJ88d1>a{=7S!QkftwH9#X$joG64$y5<~ z&P=2_=^k=K^YLwWQTRO`$M{WsxzgJm#EHFLPz#D{m^*5Ivx74cjvz@pvs39BNwh5=6bt zCOjF4FeY<|(I+Ym!=%fM*jXw~AqC59S|&V}1^cXt(tJN1DM?X-kd+R>>DEi!`NBkG z7^V&-X+<5i!>%po92Ut&-TsC$qR&;bBdVg@P=)t+eo!tpBC=M@9#ZFJFn3j?nwEmP zF2Q^c$Ht@yjV`Ol6oCDBQtpvThtdeQRfvF9bmyf?l|Ah#He~6#+))flwaJD$4sihD zZ+dkGOxMz=x>77!Xmblt*JKdxm9U!hE2E)^o<685f|_dBtwu=fEP~G=N>0^P8_YMW zx_ZNr$$J9!j9H_1`eH!^+<(1cS++W^^_ohd@dmv5j1z-dby}N|DlK+LDpZo&qOf4` zD6AkNNzoarILwHqq%9`Ih*2gFj^P|9rA@75b)!OOEgG}q7EQITP}Ii8CR1vwS7r$` za4?c&5>!9f&IM9SW}i&%h%tS*~ncN?YsDhTVH*25BgcHM7kvtz)P56V2x+@YLA8WL@WJLu&L-s&a=|iES zvf-}hA(%nPo@Uh=4V6;znlhAa@8k9nW4u8jhDIp1af?=;sMN4V0$iyRF!>|d~>sG5B9s@ydG@>0A+r0{{&ot!r z)Z(!rwNEz=&-IwUX?Be!^>PKh;h52|I-DL;`lPlBXG2Rd2Vs{X;tq|6Q}TSoZ!KFR zQn@5DmT(7>gZ1>t;Bb5t%!N{ekOf8?;^e`VE|5|B0)?P{LR55>O9_iiQwakpv#mOw zx2PbdHPdFKdAM7KR54LgQq&{dqH)9@f;2I%&;~*_nL^)h^CB5XDwCY`haD0!W0ONPk1WIR1zVD65x1W>5uNY9*}cpmfh5^xe2xlkmf9R$fI6^SHTW z8x|R@7$~cu<==@|k zhg6|Y_*4i3sKXMHEGSF`brKO1l1X1#Dz~RXO`j|rFau+fJ&-a|gfnXTcdVNlj++OXZ|M$9r@+;43linIn1 zSc}X!xq>=>NUX( zw(LYPZKD^f3<#}ebkV`4jIjt>F`73ff;nj|pvVuovXjl2ZlnxC4jQF3of9g0_}&Gj zQEe%$gfZ1bkX{$;(90&vc2zloa2#QsDu-aDzHrbbMfB65V%j>cPeuYJ*sxSBn!$7e z=cOtF49bjSR@!`H5(`q4j(S!rm~~MclEPK7L1oE$y#Dg&M003J7taSL@>onnUh_~L z=HM2Q!i%8ZgJbCtlUfmSp!ef;G{zgC<9+5F9{ohx1UC;vx(m8eUT=6PU$mBo$J`}@ zUy*k@N8)w2P6sdQtXXf2A{;_S9ydfBIIV@CS*P7{5pgkBuh`w*@n*t~fZJn5k(~~Y zr8SfbnwFyYaajfz|*9Jhqj zawQ#9BID2LYIcKJpGhO6TM)WDc{T=Br#=d=(VW|!7%J-ET90| z*1&iO?*F)z1J?Miq~5BBsI@Rivu7h|SjZu!p(qrs*WH8Wq%t`e%>~`zi9{ljO8bVj zBMy5xGnT-)E#iz>f_C^PHbE8$N8lfC*qDPNvY|#qbxqxDfgv(3 z8<0ak;fy*%Rzy%N*y3e*1+xB%dq_Dvs*b6QWxS0$D%R4lORcdh#6@{Y637i!)#}=a zB4agcnH6jp7G%~8xXbTdR8%&~{7Wp?BnSZ#F)E5x+JX0_yoOS7lyXI`!{g2924UbEz ziW?DtFN|F&3mqp;@xdU@L#lMc7t7U(iAFcm*>=yY%4#8J)zq<=R4eLji%EOd7fKH5 zt7&hgnH)@~o#DZ#MOFrHIPAi)OzlF@s2Hps>a;OP27JFbs)g|kh{lf%Y1AaGXke6B z!UkFh!0WgOw!9n?jg2ZR6}2oh>~&h|m1>m^>zpjhUHWkY4vXGkFmLmN`7lFvV9jXZ zZ9Hzajx_P2Vd6Xt`ZQn>=~y4u41yle!6lqtV=KIJMu7y{&EIEK(T?hf2U4=cCUA8J z9qlzDclr*F@ACULq;2ZW@5%QU~uqx@G|N5A> z+4$(Fm-Lj;(e`215na*`2QlAZ+Jpoqf_BQ^a?$U78I48f_Kd+6Kov!ZJyp&hbtzOT zSyLVuEu;$w!cN~{Fs}5D2F2MV))tFb;x?L96~wF0>R$0Q0LQS9rQG_+nyu>^4;;lw}e^y-~5+B2RuC^f*i zbs{N-xUManh*iv@WHBmB213%T4Vp|?p;l$FWYPwb$hc|&F)sCxs#ED=J}&_khAfD6 z5d=e}kT-{tRj8L?o!-E0HFgA`4Th4=9BwQ@knXsvs-t_+24&<#39iG2qKI(efhkpxr0yb>YCU^ZEYF{H6xx7n1+VU^2}8CA+~3k!US<58x|qJB)GKrkhD zc>)q?L75nVQ#b+)AwDz$%<6QA4{LOS!lDQ=*(er7$SugSIF*5W#yuiY+I2sID8==3 zI24j13Q@5Mi5g7zaKQo{P~H-%rZQedN}jj5~Y2JOP#kdMRyRb@ITlBIJ>#DPF0GPyV#gGjDARH#>AB53Q>Q-gH^ z9r>e~ip52U#F4JLJn1nz!p+9u)d68%OlPCbu%mR40ayl_q=?3#@xX^m4Bpt1HP_*h zod{=e+lP^k2c~DYTgr(&u&kQ&Hk=bUVZm{z5>LUltKc_Jj>#OJ z3M2^PT2qH}B8*wXb&*OnGCAbPXZ4U|OazpbbP0o@m3$m{F#}8CW^EgJTHQ6Y^UTvKsdYB&tfjLL7#4h^eZ$LP!>fE z_@YM~o+urN9OMi$00o@J)HyncpjiQr!FVN-$c1y_No)WxVx*U1QNo!PR}E!MvT=mx z)+peUqj1OK@rF8|2K_V&SQ`sqWE^IrO$q<+9PQ--1F3kW?xJJ#X*?E-Md^+f5=9tE zcv`mp5iCB{5lk7V-py#PSluX&!%s3{D&WEjeCG&Aaf1Rv)nb84=xC9t$Wv6-QhtvE zu@o4qR=p2R4~8R}m@`5LHb~L8CUzG9(8fj%?q&PfD>oVQVL>?=X#5Ar)bWNQ_SB8?C4x_511&pGUCaZ_Y z%w~a3cUYGhRv=1wM&X8ZREtYzBxRqC7E;y|`8+(86ghZsrm*@>PNaMZm0ztJ4CHKd zEP#55z8<8Mc1tJgFx)D)?uAR_#pX2Pzl(H;NaFFww203PNi?oyC>Crc7@i|UHX<8` z<70Bz_Q%ZGtZEWbEHb*3C5LP1YVza|T$EkLDtgy6inyrhfE)(QWnAG#>jy00Gz^1? zf-ydk9&^X#YFNeMyJ#KSD}&vY&0w>yr)FG0M z3gH=tMU|>eWX*K0+7m&(dn6oCuj}~I0HhaJ#JaS^nn=Z?g-|%9qyG)&O@W-h?kyM$ zT{kt>Kes!3J#J#Wq#93U_v4T3ZS>>Y^kX+3xS?@P_741UY4%P$FKK;Gogcxa;`_ci zZ~o#PNX@LZaqhSL*V$D0Ys87f_k`37cg2Mb7b54Oedz=}-ErwOBEUpusEL4*v@LNhi*}j~4JCB>_Z0DON{nBdO}f6{xd) z7xYq;LQS|hx`3;s@l7o@lJ94(ho-*ETX5>?kB1EPPu_y%qL242eWP%M>7Qm~M*n6` z*b(_ZX;!aR?Pf0zZdZbX*}v6!J!|Q-j);ItPFi%@0;V@Na;l$Bp+9X?Mv$c;P{tKI z`Fn+&i`RFYHl^OhKZe(LCEa1mvX^h7Ug#697Dm}pmu;H5a5sP3Jl>ub>FwD>iN7m2 zfm(VEeQR5Yxv3~duDWfmcF5U*M&bb6Av)Ny`p`0^{3b>ri;|$7^gF-bHBnr zlX~-7zJdi$U-rJ3@_dcImrY};T(yUP>5Qbe_V70hOoPmX_^JHsaLH z-0S1J@uRNepF0EGar?M6P%rG`?;|ik^qR1hQ~0v>n0%7!si(sK$^Y+T2>0f>?*Vpf z?s*Sz{K?MJW!>Cb>ptc^!0dATQP%+Ps0AvxEgnq43ZJZBvJqC7V2o67-c?DV;MA!4 zJNa-?>sM}6E3iqb6dJ8WDUmCYovPN@qH1VCH8@46Wm0V`i!P=4di({xhRB>r{o@6`eBJ78yX%=X znSSh0?9|PW^cJ(DzmiI0-7h9m$=!C`vb~>NPK~_C_xJ6l2WDAz=jPU&ul%tyWLIvU z`umIg2Y3Vf+G968jWSm1(9-!5;?~W(Vnb`FI}a`BJ2n%^L^J2n)ms2X@JLzq({5Fm`2ls*CDtEFyrg`RDwV(~w%Qsh{(&?M@kad1lI{ zSNNZ!zhQPq4!|onQ;(+RFQWQ?!MFGQv(*8IFaVFtmr%z`|JF z4TF<28^*P-@*g2|Z5W?tVQlG!!O58o<9}Y`A0i^>P(14bDanEu>IQKUn=>24zyZFN zdiD+UNx6;77ue5>-GDebvjP3_0DmWOX&c6uSQuaGhQZ013F9r|`gZY4*y7*lF3!oB zS^N}g`$7Jb#Di@Zm$5J&=!U_`nGNIZ-|%lDUTDME!NT})Hw;eBY#1Bg0*v3cVO+t& z_-!`~PR?u?<=^qYIKbP|qP*$Jnq5UL{5{`H^l#xZl~=Jf@U~D_|DM0TSHYkBp1+nj zb&i5({{dSK05B7b1M^?qSaWh_V?Fsc|6XG2me%Cj$>OznOV{M$#neCN9scM%EQHiu@9^_|O}ayuWk0v2wLpHA zh2Gdgh5yXY_Fdlk;>)f2w@}KW`1TthQDohHak11B&4+e%|A$7 zw`J}PcP|?DSK>6{;?}Wa@d$mqSZljz(J4n|)@4M%_gY{>A53N(K!7uI09p7R|K=qR z&7_qMm-a+!(A!+v&a*pUAKF5l>l7>{@ax^l-=t`?L3|>n;>Ex4<2>Tck7(oT-EC~P z+jv8)gUqkk$}gGSmzQ^vvRC0xba-c5W|)d&QWL2}y_Pz^l0x93+V1Q%_-gta_N17- zp_NRnd7u9&WBd4+^(;H;DEm^*fpQiT(5bBFfwIE`aGLXed&p*-C4aT0HxDp9ln?w5 z|J=UQwlY+D>Q;IlNvSLE<@d|y9Gl+MH&ysMwmm|+b#C(8xV0zwJ$7ejXhc9{4)Xib z2Yh3nyM?op!C69`{4if|MD#@I5ApXCv5&y{)7>~vlhd33?BIMJjSrZmBjZs&DLK9I zA^(@%)hn@r+@K%ecxye&ED3~Y<=&{Q>WLJIf-h`VW?{A zfBDz&h`T;wgl_L1p^^54Cmfy1_kXexS~ESH*JKZKj7U`~yR*)z4aCAGZ|n5j?#x`K z5D8hvXg)T1elk{zbuSXxy;vHsQhWnHI}((~J1@Fis+FEE-8fnt-d(NsSVYPn2o_P% zD;AzUMIF6x+r^g47aYa>V?X5(!M`P>;&j^+y(wY;M^M7LOBXi!-fvODU$;`R=jThe zR6ApFM1*k4WeZINvGClvmzVkHb}cW{Q+4InI+YXWexj*5lMp`HT;&O=(B%uy@=ntS zv-U6NjMEG<`@|F@|0s&t4vLwIUeWi&KWsPjLz04jyE*l9{`|K{hSR~PoZIs}N}N0A z;xRRJY5yp1iKjg}9UheCE2x(*>lYJ|bGf1o*`u8i^_)wYzTEFSOxgWk?w1jTIm#~E z(ZA*}WrI8V*Af@bQTE{-{rba{J^u3k^~6`^D0?BFxB4(;@1x7!I!D>luIT>*fp&v? zmQs?x^ZSoz_nT^5*`FqeXFg(1J=J?o>3`Q5fydjUc4XV%$0mAlu$rn=sUh#Y02WK- z`I5O8%c;n&{+I7Kr#<&LlzyZP7pKLq_9pVj=O%LMQwBkxuV2S7Tnlx~!DWl@*ySH< zhwQyUQfC%&38Xy4f_#5FC%w^p`wN*~dR ze$XU%ksur&fx5LD_30J;qAzqRH_>PvG4r*cTC1dDh22^E0JdVhB};X1^iKM1#Ttb~ zv3`YYZT}jnM8^C(mbgy9#5Fe&xgr3;U6?Rd^9xPnjY|A^7Mx%cSZ@L~s}oBn@|Uietv zQAY)!dMcEXsr(9Dn3FDj)hi&W6@Gz$N)8F$`FDqr(~9r+9>$wy45PEH%$uGs%!zxg z{isiJ24;l;)bfbnvtPMjA)h)eBDj=z>m$bEVDGWG=O3N1IPho~*o zQeb+QOoh%9DjjA@V=`u6*V=R1DKgOnY^kXxU82#8Z$j;K!(vR^j zWsq2sZ7aySkZ38$dX%IUWW2p|mF#H9X5d0=$Yvxl8nRiHFl;2PA>%#TuBTTwQM8I| z20U6t#wN8?WEdnyZpmrhM*XhqHH)Smzn=dHkN8t-EdQjZZn=^FRn9A;hMc&Gzn4v8 zy4m?nzH{JDGlzB+bMH@6Tij(^bA;ce2ecz5`*DZ&et2;;6KS) zGc|fM|ICGi)nz5|JN41 z@4OYVjS{6sB11Hn6&oZQ<&vgKrh=v9iWSO@GPO)4)2bBPSXEwI)u&R+v@)quD$z=0 z3N@Qe>+B>OB@(q3w_Z!3%##C{T&n0&_Q5FUJkOQ)gd`ATl+%`t4$Gh*AR!F%{8@`zCshOTPMo5jx9%t;r$na-&n z-pgNqH1WlcAd8E-$YPr5eD5ors+;Ym%#bv zjP^_>9`3-s-ark1mp?&VZ(!C`yv~1!KXgJ*P;03xpCFe||M3JVBA%WD;GZ7iJ6M3s ztgU>HUq=!%)6U~m&-Q34a|$iCU%qYkj(HF!u&#kSpzM!9NjCs zFSK;8*(>nHw9d7%Tkz$+_vzkPmi?=trBdi)+x)I!O8XVTB*7!7>#rA_fuEmKg4^3~ z1Yx9CuEo^%uMzCvq#alcvxqNTE7;9RJAUQSiz(mN1Q5G1kWiLs?HPLnduAm4&mO@D zl45%WQlWs03Zo>Qx@WH-#N(M;_?W4U-w@o|ek05+n8#fwa3hHUu!swQSxoP~PVi?Q zFViZYp`N^6P#DM>KuIcdyR@V~bR_Hxo5y;bmS_H|c`T0-5jUvMQS33fZw z=xu@ml9;+rXPeWmTl{Uo?F8>{t-Akaq=-8O7qPF*NIrb0;I5{UcftO91Zxgcsqr3xlQ?&d zN}s<+An8@on5z}rq;e=lb5c535yFV{gB{f>ik7QIdR_{RlKlBsG0-2AIwpd(cJ+{-AM};Q$JZOJPue< zSU+yzAbFvCG4<-hf~(jhidjuAKS6j6o3@Mo?EDjj+|OQEfW>L;Ny4)aj~$Osm7fx< zUoy?$c6Z*PlZ5LJ*EbhFDmanKJt`3Mmdyp?k&}gMI9Sp>zGA7cuwqt^uY%-#S|OYL z@0HZ;ONC2$nl|V*rFUaT%lZap!=cW5TyPf8Hdo2pmI+tWJ%f2!c_s`RWkIYCq<{81= zGm>6ET{t)+Y1>NSqcf65pA(#iq*hBfcwpmt_YC18V#)VfE!0x3I5FGtPcqu!l{eq^svGrjC6<@Fg~l`RvCpa6eElFLtkhhrH@RDFP^BrwX%R#R_Oy7J-aFEAaPTlq!H;Yldz%YyZtF>|r4T~1x{bAk1Qo<%up0ME!k=u8zZ1T!N5VsW|tfM7B8*vo>i z%t*TG6~VPLlH$J*G>PSAW&}^a=G4-&Fd+Lm3}8wF_VTHCWxZv6clf?Zun_jbK4rY?U|P$0Iq zdQiw|PP;c8zZUG{@&3Y~Mm_kN;J97eem21TF~75g@a9pczaiMLxXtyhWBcSUX6o7B z2tMB~j~@o7i+>A#rUi03DfE_LJAtq1HDxOY0B0O$+ngLuLF@jV;F9@-(lWOIT4w1I zKu;eLWwOqQiY=|_@G*4=Gxc(F2%lIn(X>IvQB>svfoE#jp9QP?wX^jDGn8_NlMqvl ze+cd~Osn=jw(55_I#W>Xj`h1f5|K~TMhZX|Pg&=a=Pc}@0{NuMzLSookNa&W2F|Ei zspF3!Pnzk&Ahz^Ai@s}Uokj0HhP;%=yPPKE_0(Mz-h81x(jf$VM4MPn?KqCSh)0Ws z2FNq;BOf4dZ@&?HEzEj-@*+|>u(wS&o%7);ikZ4o%rsflUI0ui{Jqo<7m-_gl{Fx-v*FbJCy>;@hqIgY z=@UscG2hA+2E}jgl>X4tS`m8`JoY3MJZX-Cmz+cvd(}vHBq5rkVEttB(*$a!R}yOd zuLS}Eqc^aWym=l@rY$+>TT999eO@aj_p+?p${btCVHR!2)UC@%97(rXX*Ptnjf$R1 z-rRnpe`qf3?B(R18A->SMs7t?tK#E+hl(B67TH{P8d>2H+gsqaQ>xR++uCpR4}F|l zwvxO8e=}u;B@QYr2-KmKO^U!ClY&N<_RkM>*+gQbnM=?(49oU?Do%D10Kh-pi{=Pe1Xl zb?#OB9c#~3yYk}>)xFJDegt-Ob1Z1x1~OdDuS}}=1R8^%%g7(rEkA#dQ5gK+Xbcv$ zmqAW9%%@Jjn>=w|c7_)3ZENp!5VXo@3|_gF+&K@Ep*2&k-A|s?caDvjDVuDqnR3zJ zI$f=y4(%tOBOL9VFR(dQ>Vw6 zliw!>O6{EK)f3{B19y@`#Don7`fKOCxWh_)cspr5p(8)p&O$BODEctn*FH?+m0ku& z&KG&QTh5v`a_3}>un4t{+%LMD^!I(e)q*{(7Lb>+bzQ?inzqK}WQ^EoYurUow(RIR zRamJ5&ykC!uD^#on;`DTS5U6p4~*|8Kg+%{JpR=E`S3^Ndfr*o@eh;i!D1KN6q0g1 zOrAlIcBa~OoE$4VRQ)h{o<8@s4UyZU_xp#*Ly+SE`UXm}Y2ML7iUn=7GuGJCW5>zS zwQ?4rk3L2w zsBMpt&+sm$p{yk`FH*k8N%irxkdO<>4*FWfFexM_xsBk#$H~$O*Y-k?BUa|!(2eI_ z7SC&`gS6cF+E%xdoSb$Au_wvVCEx2+0mY2ylMLw2LGK$Z=?+OY8R!W6lfl08f6Xt;Fs>o{M3EUJ9_Sunq12eY7Vpet?@=S^&_>l|Ig$;mmZ zl`~D8oE+;ER+d|NgxNvqF5#U)-T5PO(-JdX4VMvM)D+AZMVK~sj`QDQYd2HrACtp0 zAf|nsoR8JMAN+(on`-`q{336X)4E$(_+#Cz?P8Tq*iNgWAWg$<|!=Hz1NU`yaf7qj9U9saue?d zbbFXNdDTzJ)0sJW7YptQ2eo4Uf|L5j<}pKySGxOzoG%{jZa63B`|XCeHw;b=)9^)8 z-})I@?^ZES6;yipLqot8gFK!08TihTQu4bjYHf&150LwpT;}XG(`d3Gb}^l~s4W_K zge`v=<$jaAn|E0YFek%GZTU61^H>furNMqjo&L+CS5lW8BscH+^M?y4!$I<5;ycck z%Iygj>iy1}4;>_lfv0C;w2BhGg>&Kat+8z%L7!p^KJBDF`xd#bSHbVRMQ);wf0NwM zv7zWLr~DoH#A&lDT)@t%EQCGb`OAp+=cv8@dosLYZYT@-DV~a0N-T77Pcq%eSpPuo z>(z)uf57Q%=^O>Ql>IbAAq4v5(YMLVJ|TfMdA?9lAcQ!!eoy6qkE0wU#_`|uTabsfH@)#p1;Iel6PdcLd6W>efYE}KS? zV_Y_yb6b;bv2IiE{F!V*0_*CYR5%AT-zELTRWAA*fCb}U?~>;;3&stGPW{`d!h4wF z+g+{omy`1a;LdSx<>Xx0t_c0t<{f|f7fkB=X4KzeS{W_>!>vJTPwJc4>hE(=C;pB6 zPXfQp`p?O_ySsi)&aC>U#@;7i;ZZNXORl3n|DVJ{;kVctUTuMyW?eZs7rN+U$MQDk z%4y$p=c9c$QU4kseoY*5^%VTvlkN25{UCkXsW|)r`Nn|2-8~ZU!W6F}M8sk@JrAel zS&v=aDJ`HLI7FU92;9sxnn#LF*HZP>&_%D=v9zu@W>nm_pTx|XfZSHCvsZB?ft1!R})$uagn=c z_rLg^P6tkR|08D+w+`%_-Teptbu?6~GT|CxPZvsWzTN?`(@nV_CRY;prAhHbP8Js> zN;gmVMdH5R7`?Y(<@Ch9A4@mP2*gJVZzPB(n1RL5)ZZ5fJv{uu)Iq)ezodWw?%Nj% zZ#E3q5cd40kf@s_)_Z!O%|0e{dvb#%Y*DIW2D3-tSL>c0iT{p@SU@_>^>^LMo6 z0g_tNt)B~!!MCOSUg4QRz#R+4v#8i&A$$BhmIbha${vrNS>a)h=_@?6#l$V_E5k`K z+V7sgxtcHM9CbXuNT^=Y69S`li!ntzXDKdvp4LGhh;`K8ycN_Ni-osxket?jvq`M> zn@#Izzu6>;)qWoyyM1#s?1Php8&2q1bS=dyzwes^!uL-Wu4NHo`kT>x&jQ+0_x(b< z=h~c3ch51~@!W76lxvxA8SmF^1bcv?_Ae8j1q559$En~TrX;3HOAmgi7wIv2aO(3X z3&B#gH<6_*0JUeeu*nNh-gQud2dJ%Q32$ZJu3rno zrcKWazQ!gUSR>rSzTtyA*|!5LU=?=hI^orvq&0v*E512N^jGU&5}ZW@ysg%7@M#ar zj`hMTc}pJX*`f^_g`1#YrCZb)nDuOn9`I5(ZxqT7Q~2$TLJjY~<|-T)3qLh5tL9bI z^J3u!-dpXGZA$B|7RJUI-UCce62Rg|65$QK09_~*>WFzhE_d7w=V+Nw(yJi5pPVp9 z>1Sj@(_tFVutjT`vdT6^ch%gYAI937x0Vg>l>Mn1Im@zW6HRj z@b84hH#ZYI=1zE#7-O4A9S~(4VfoMkt5aIVaS^RlTmTHyde?Dn-~C+BYxLUuC!CfzK2a9}sxAg;8i zZR&j|GpN%{p8nZRjk~AP=LoG3$oMGTa|@)@e=HRC6X+v`Z{%$iK1BS$x9izLpAXc0oo3cATF(rMk;8Dln91=$DgqB*bu@5$1M;*grP`ulM&P zVBfo)p;_ynlYpN!2}e$|`)7`Kj{|#vY54Tu$4@$AnjN%LcbJ8j6Lx=VAUHYgftb47 zB0QEike^ixb%zz(X1xn=iUpjfPOu5pun{}niCrU4RcylB+iyf3g~0NK?ZQ>~yHzu% z=yc6Lu?uxP-o6%ueU#EEln?BijR)h%a8DN=ITnw7)IldUiWU@NU#sSCyM$8w-Kv@M zaeHeKyFt#pmzatNUYcFwx4qzK4lp%NA9u=ZjW7AB6Me#c?eYX(FT4O|T<`jXYkC1= z(EsNgz`p7iUcv%ndM*?Yo^x1OCkMDh-i_GtL1C1IMO!7`#d&1}jsGzOgUVLhfjd*3 zBLv$|w1b(e9Do?8?}vm1lEeRND0NskI1jqTsb*Ms5y87;ia1Yr-MmCR0N;-D>BrY$ zAe;Rf{kW_TkL~p1%X~a8p&u8~uTA=K>1lX;iGEy2KQ5pj+Yv1;`vv;(6D=M;q#sTC z^|N?ThhE@|o389Q0VV`7lXcwLEY74@OCjcyBK_346ZH) zPgnM7rZA17i$TxDpoY7wfRT$q$;F`KVo-50Xt?T`<#fe#_hm???=tIK9k-9}N2cSZ z=HstsUlY$O=#FH1?sB>#nV$R6$@q)uxgB&zGClWa1^%LYF8dPK3D-0DaAkjnI97UZ(`Gs3?t R%zhQ^u=A?al`-M){{uPGBHRD~ diff --git a/doc/LectureNotes/_build/.doctrees/chapter2.doctree b/doc/LectureNotes/_build/.doctrees/chapter2.doctree index 5622196e5181e36853a70061a243a6aa91e55ba4..d7dad109b1e1c5b5c63d1cf23cd61f551aec672b 100644 GIT binary patch delta 44017 zcmcG12Ut``*KlSocV%G_gheR|3n)#2-K8tl-j&$JhDsDHSTMFI)~FE+UUBR#*4QPw zNsNUg8jTt?iAjvzByXC=l4dmOe`e;kMNRU3|NHqoa_`KkbLPyMdS_Sc4P5nUpvx4o z{DnO(KZ>Z2lw*hyBRjP16h+QWL$xHkAqcgul1YY?il(P?&w2|k72g}&5$W$LOpyAx zy0I5>+F$Uh_}bNM6l>xaUg%8jwg98!)QKxEA?N4*3q|$F##pW^<7{A0^fn0 z80pR&mx`F$8xTH@NDeY&R(J;`BJ`zHSu;S&3iXhRf+Nv&z_dx#L);<%e6T}>CZb;z z2>ue0JyzuB+J`g$QpyXnNkc;3!6r;gjH969LBa@3_L#9%ij3TkCu4%lI8zFaT3}qr!Sfl|R`GpQI5KYK zd7H6xZc@0EUf&&UsAyHc2S)p(ENdA2{V=8l-iL{;78;H}WQCD-$GQXH-Pm0!(!x0T zoucm9Ao@xZtSfX-v((;3oYFz|7{FyFm}Cc`QdyF@VwJro!Y+c8Z4IT5Itg|;g-Aq#g*zGCn9)@=V#NB`BL)Dbpc)WNI1{>r+vg znul-)f!y7Q{H63p!*O?kl(hx}UT)-$I!He>$^HL@Mdb)3WWj&wB>ewt`-;Oh!nvrRkYiuaikgR+HV62YLLkt-0=o1 zq1SvWR(D86xVa0t-VkNFAJ!@p9K)v~Dd!`Ex=N^12hrG}l) z$O-x=i}R5hcX_=^PEglwj75O7uxkvOC$o9jwM`Ygl5V?b4sD$Mx+}&j>HaQ4yFh^e zscTOgdP~a7O@s8&o}J`e%JZb0yE-=p@8YWMlJ4Z@aTJoZ4Rt}^ah!@Ly*@+~3B-7$ z{6{M;Glt~X_fPzSNKhosboUTc$7Z$|(XT(kUZP|h;4j_EJCFTEQe!2uNp14oXf_%U z%c0G2182zo0V61cHZFLy9ry}7xC69wM(mbP5lLQq;QAonjW;@cWfZtZ+o(^>7(A1PbLJm397eV7 zp9|#+EOa9MaBvsANF%j8hjf5K${c8XK?j7_F{%txJ_Yf3OGArKMZ}Or81G@FM|-3z z!=fO;gWCH*vPZN@Q%1P~WZ&>f6jF8^1rkfCAd<6#1PlIL!=X?b@{MxH=R7HE>UQoU z^5`qAn^_|U;kz0g@NO`&J_XVWkkJ>890sqf5|D>fIUzu+ z$V!QeCb*%olKX^_6wU%HnGl5}0H;Cz;|W`7ehzZULYtjQB3_veASV9OD^&6CU%BRb zn(7%?cv3n$$&QX!2$Rb&zF^??yCKuNNQb7on=VjH=89m4zEh*5u2T~MmBot-((b8^ zd=-qr(d`FEwFj4NNyixO(^{j)z$vO)VRuNE)@P#6)>mL1W)8#Ydf0=RaSE*I zs};DT3WHoHKjdYjxK!NyS$*j%4PwEw-O=$yubHLH*=z8~YHYY81%GLT1ONuEu1GSvwPm__)Il6ba9t7c}%5>`Q`X(HD=TS3r6Gbd6|u6 z?vP9R{f_D?(9R2|?Tm0cdD?TRpxN@b!8>?kq}{nLDR1~EHDEATWYKZ=rU<$+SI@?_KyL@3S_%bjUexv%0( z-bPWewM{x&UH}m{duKK%NHa43Z6hr@;GhsqDPD;v8`i&`^dvRC=)7 z2Je(4D>M;J1`?Ze=5=?-dwa!9j)jN1GRGLZvbWFBr!|(oTR9UgXY7_sWvkNha#wJk zNVyR-XdRNf`K%_@q+(4({83hJ;Pu65RYdC&w z3yQk72l`2e)_q+Q#-JIrPOW%4$vNv88Q5UN@kH!vWDOCFK*i@Q62cjE0O_|3C*nvI z3b_|wr-Wd2r3|?n;yALjaYGDls>k)(*o}b#rM_39aW@r2h3Mx2O2t?=rBkl78LrtR zd;Z}~>(DqUZ*yz80Up_0W_-=a+;O&%58Pkd!f`9!*y4#{AeXl}Q~L7_$YagJf7^7l zNwVz@kR01!pkZX)ziFfDNSlnjz#M1b_89!$vvBnKZj#pR2?M6)9qllX)1Kd5W5~r; z?6`!!W_VvqZ|QNVN1T$%9RTlIJ@>DL6+N2}%R z6>m3DvcZ#}XMG04)$gRs4CkN91odSUuM;Rz&UYNhgc0yFNMCx_CZp_qHxb>Fh1NSb z1ahH}ipnPr4q|ADsJc{HN$s%V5S6Jz2{;_XPy(ci55pkx0$TNKHy*0S^oPlM11aat z2GU=LDYcGn4<*(o8*_UFQot(w+4PXaW%1 zq-{st(MV~@2NKPvlU=fV9$03c4(K@A8-R>izB4m(#Y@NR2yb$SbrT^#dh^3j7+q7Y zPfod02R@!nVYL52?E4N8^t}$$;iE*A z-9ufz_sK?-giZ-JNuy3~loRwpd*6o zfr0#mLN9kIdzG;ZE}QQJVhmi+={D36pcoJb?zO?6u^q@+bfGOK4wXZoubNGz!fMoiy9=Kt2@TN)%jR6s5+mj-TK81`AXI9Oh>-uc_<9tciq{nzSsE0EG@VzIi+*&mFd!> zfg>AUuK6VpcX48R|Cd%eM6)#ES7%1ee;Y>CVas6?oG9~tk8z^>@b~ecSoR-j_ze$o zt}F797X2|3ujGR5_XE*JY095zXr(mweu$F&A!NTLXIuV^QnCmB83>CD9Q)v(T_BrJ zrD7gL;VaeIT9wD7U(`wguxEx4i&iq@tdzcg&=I|a;r)Qdi-AHl=p@pgaY$!l8~>XxGyU`_=8q?})U0f&0# zy~pJ`Dw=cTNn1Qvlny^>F2dYR;|MU)!4+34UyES%mgEV@R{kSm3Q}`9+z@I2OF~Qo z)STpEv_nqO2c^yyP%uD%G7Y^!RtxAGV=+}RuyB+oUC*pDS2Icx0HEK?h*H+IKOkpC zm^}N>^^_J4@s_e)GLSkVx`wBC^4gulj6qfzkUN?}wi-}AfLJ7%LgK5TGmuGDPM_y2 zLv9TR!K)QfgAH_jyzsrJvpysvT~Rl@$CEa=u$rF(M&TFmhpv&xEpMkM$!P$^yhuhi zKxVVt1g8ZUlzGE8Pm&!d*213Pjj<@)_%6_c1&Tb$)ZQ)uWQP$wFrML7Iqk{Dm*koo zayLyc*RM>5?RCbgH;VLYLc!!ZG&&stitc%6%2Kg z=R_9%3@@dI(Z!?k^M)-hyd$Ru7L6%hTzC&s-DiON<>uNfc1wJ3H%qSFl3;P1@j(-i z?Kc3=9^F5$@OOAA{Dc0vPyeXAA3$nO;Y0X4d(5Q8h0dW$r~GfSd0nB+l^J}LaCH&9 zq_<1(^Q>M_6CFc-#99j#p%AjR7C5LMxm*iP!hT*f4wF(p@N}pn=IPu0P(OqldNJ>6 zNNfS9f~Ek^88aG{sBuQ1w+ zHgl3;0naWZtt}`U9%sW)1gR5@yhydW$d|u0sEa)Ds~R@tLZ|VhUkKBluDg*bA{Y8+vp&V!;#y+_18ub=%Ne$7K#fk%J*3<>IUi90Mz3RYOw6CH(aVI4eF|w+A#i$ z?n-VoK=b}T-BPFIpI*GxS8>a$UVrD7j^(jvF~YTLuJZ$n21s-=x!(ufQhv$o4`%gP$-PK z-K|8*&+%s>N)&K;4Ynvlic-O?WFhlYQ4iP6T#t>K9+rmSuFY#OcZG$WHK>AIYlvFQ zRF2F>$c*s&HTbmQ6s1c;@006ksDS3mYDJ@~3gqK-)CeHVg@csH}#QkaPvV9}eb+lOsrXBKj!kh9HDAI>y@bjSh( zKOF+7qpzCxY1P<>d$wv{NBwLV2q7y8X@%rfj#O{f%+EFkBrCbp7!{~sn9<_48W>x> z$(eN2P=Q04_Tj$HaKB6xt-xc;pv#>CdB zgF<`N3DUL=I)GO(l}Kt^&0Bi3MP>NDH`|FJ_1kHl*18=Uh+Tb{3!G>N{WJJUXQ+~Z z_CW0FLn7Lv6y!=ObCJ#TVog^iwlpVu#z1u4+a7&@qkWibktxH4XcW)Mf9?QNaxF*oYLfXPI*1FYhJnPt6D(nj zrhL*Mi%qE%&@l>nm4Toai0h07Q4A({nM#o7Cxil&md0b$K&>>qNxj|$UB`QU*rtH{ zTn)+_{x@AwuqF6$>B>PZ=7ZC5*9$w+3lmgM=&>Y7*Z;Xb^E9gi7#RX0~@rP7cxx{No_> zGd@m@;!l1Vj5gwPW*+onhXB%X@_7Ne%;5p#Od%NU8yy?~PK&C?dFkd3fdb3wZySPk z;=h1KSw+Hd^ni|SE)E5xzsPq(Q7SyZDryCe8ioiC^5reM^>9tsg5l^Mj`iinP4;E0 zs^oDIAjA^m2n0t4NR4q|-9sZZ>s}cFu?lzeCD(Ic*LtU$lRKD6i&1DL?gy;sNst0x zHhFbFuVVEhBS)hc)Q|H$J6hv=XEYjv@HAiEr2bMVSonB^7y0rETfqx!9Sf9@N0+(j zQ?tfS4FN48cgI4UT12HO)^PM62dy7}2(-}h$j9SVb2x5|N0Sl$LY8z5xYmNg|2z@e z9Awh&*tB<~yH;T9E9f--6&TS*mN7|V)O!-D3#U42vIRS`YBCCi&Ld|E=-^V*@pv-r zdKF;o6aZQjAXzX49icEd?9xc-tJh6M8xc;e$y{ny3$Yn$Oz~}|&kn1~q>^u@qXQ_F ztepYzzRNS`ZrnAZ2WvCSqO3_+$l9n%mFP?tJ#1OanKz)+f=0OzEgTK?j_+(AeekC+D zc(Uy=50g@fvmGGT2jLfL@o15#L8w?WnX(dhmUh2yB+D1U_ITe3u;w;@hz}`Uh-Sg4 zCuALrJYZ;$iu_60I&hY`5@_Q^F0Da13Yc!cpaxFh18bq{V$@M&N-gL|`)xys;6o}4 z>{W}c*8|ecbuf5?p*?{L_wJ=rkJ^h|UJpajh3nv4QLkG39LaPY3M;|6a&!+3rK%R8 zF-(NX{)UP)l#C<@Hh-vqG_W<}3bo)xZvTKyWcM-XTCA`gNIEoy?xw?QFtuB_47t#e zU#OI|*+lv*g|e3%#(T6>mm|0ao(~qKXF1NPqF6=#fL1fB(Jd>qnU}cN3NYGgGI#}4 zp$__v6SUn*h#RY!9mw{TipNBAH7x{G8omm3Ala+XYxq;J1wb=MtxeER)3c*RngD99}EfZbI)e11oW5-YQ zVXVR7p8kCGCvG2wk-KLg#B|t)Y9d_6pO+}ypAA{oYueW#J`QM&b;#5sFukZlhB?q? zIR#Wq?{;sYCXiM37DV*NesqrIxbYT@n@|qbCzJGeM-w;x9fXi^0M{qapC324q1uWo zMLs=%wi!oqq#~}g$D2AVCWL(PE*g&}Q5kSzgO^ySrFxn-j@I0>U z+ec8jbCzRPCGw|4>)r#C%8c&5hcYm7a7GSC=c8Z)NJ1HH(aw;C z$D#8w9tS_TL`IwdbDZ(#yB7?Ujc&7C!1p|Xq5ue921~D2a^(pM#5jPAJOTaO8M5O9 z%sYXUb>kTyLEtc}uniwUcs;`v@jOWtaRi=3t1ylU;6u`G)s;pi8`JM&lx)llU>=el zz&ya?-ee7n8*|IY=n-xcz~lZV4NC30!`PdX=Q47e>9hvt)PJ0X*;7FPkGa2UQ08OA z`x7+4IFYj$&ov-BKViDR^9jnscoxWo=tz20$8$YIML<(@!i0rzI(v{N{PU<|l468J%I-r0rskj2ep)FTXXQ%vK zPAM}nUPB9neYCC=2Xer`W;$UC2QjMu#yB6bSbo#P6`i8h95vnyeE*4))%T(E-P{D zM>HGZ%0L$U3vR-g<|A5^G*n4qZ-FT*xq6l4#w{4&RR;0~3P&&^(SH~s&0_}(fCr7} zjcy}<8LvHhq@vKtG!;XY9Oclo;7WG8gRUZ6vo`M|uV`T2n3mi_U*quFyjgT?3&(2W zNaZgOYamZ=>hRi>Kk4|Z8mt^6euWht+)}1B{0>8vE5AYP&Y=u!B=8SFZAtq64!xd& zwB~mRuK60A0s`-g0q|sV*p_6_A7~iOrBfW(5dlu@45P;q(MxGj0A)o68d_u>G41W7%GpXDJ%$2-1Ug0EN)|jp@iI3h zQlo1+aphUKos3N%Q0wK#@J9Mdt-lIkcco7}h_DUz+v@NMQk;P0P-iCX1>DHhrw)tI zHAq=7W+&1-K3S(~^%zD>bh7=efd7Mop|6Dl|HLFJEPNc%B3^;$7Zh&Lja)O}`Y44m zvXR5?pg~+6K7NiPgR0?OfOap}5;)$1O9iR}FtyhxbSM8ibD$M%Ko2VP;#0x1m}=(1xxtPHebovX`m^ zUy!LmxCnj04fX}`4aTXEVpeYVHHv^n4H~3D*2lpxuaP0FGx7@J3mqdYP)lEsi5A=& z9?bAxaKqOK!SkH5cz9M7iwuP+hes%m!tp_@<8#)mI22D*B)Wy+`nZKoB4t7&5gQ1;nT)h zLCkGu2C;#KDMhp0W5AYb;nG z4v)dTX%=J00`@NoyBe{$SrzOBu{cbXa5@&Jz#{89D^ssY93E2z`)-_0pB`4+0;abz z_7y8M`PYe;4X2qN`|%J2=Zp{j&2DRn4evv^ZSeE=gIfjb_k)$)*8a#_@&D$*s-*wo z_7&`tmiCOX#@bTiZT95kr1*qXdy3UsVt0#=jkhM);_cRWYmzNKDLyI54Yp)st?{Y$ z)YN#`nT@q3CnZ}G;*;qXt(LN+#wI6QZSg4ymfpSH7PC!Qx*PjH*|hZ__X5O5>H!nY z2>pVexjoiA_?g>d5MN=hJkYT)5pTh0I=P+-@o_roodVHPP1#bbq|C&>AztN#V0PSY zhzB4%J-Die@1{<{x<>L*8u;mS@_8Db1&FMpSdfn0|E-IhPRGqeeCBUlZnoH@B&#(gG0|?fC)pEHQWKerSd@R(Sa5`t z_~g_S>LOaoVvAK!{!1>RGs4%;c99c*=OO`)f@Zi7VS|OOH5e@9&*pfioJ0n4ZV0Rp zU2lOG(kvJOm=?XP8l4Ye$onnvY8+wV@#~dVIEpN2jniQr@9S0&w-f*y*)=5#TjMmE zrytoxFwP{njTVYC+u+}EZP_Bhi@O1wYE)~vPNRPplcu)A-(mc! z<@qsbI;`1hXH=A!w0@OZ_9>SCqnOma34C&tn`^PgCMLw&l2Yx77E6gWHo=;jlmh># zJH^;yQ{q$OZSgjXB_=U8B{?MpP%RV$WcCCIS}AtB1>RFr6O)q@ARJMUEjB(SCDopk z450Yf)cB<2)Wj4P!T{5nl$c;mv8GxeBBmsOYbPWp$P5!wOHvYS_7qUW9vg47rqGta z^Z>OW9>!ZCCW21(M0-+vsx3(-v!=wS0B$_xm<+|RB_t^-CM8&tQ|t-JlpzFN8%S0P z2CQvXJ7<_;OSaillaqRv0890sReiI$!MkE7`o@d|jgxA_Eu`7UjTwX+M>CCEyhX7K zXd_!C$t{6maFbB=q5QP5yMYC0k--{3rr5$QiK1{rXl8H&cEEz!K(#=kTMEtP#a9b& za|5f}VtL7#>QL@@tFFl8#ZEY|Q)+@e*`5TY0~Y+9>TYHK{+GRt(`iPXcQ3a*v$L6E z`J2urfb8mw7a0%Jh7nkJ5E>VJc|#_4!EXPyA9$+^{;e85_czV*<7YO@_kL2F<;nlE zW?2azG%(N8{Ew{YgV&Z=lj6Y}Xk8~l%aGra6Ie|{GfJ^1@w|AuU4~l0$5~!-N_+yi zrG@$?QzTA(HBII33GtfMQIrjE>q zvPn#qm9@twQ|)LYR#lWMi<+AonW|{Fmr(Q2S6(dk#?)b8#)rN^ux5p(e<02NBq!Nf z5rAU?1Z~d78lT9*43H$ot8dA&5D2JpU`b6$VHI5hwWx7Su*S=9yIro|loWd+r-o5V zg6w0dNvXW3DXB25Weflh!ao$+#&Nic@riO(LX?&(5Ckhi!15)$I#N<$IL!Hg9l1Y& z;bACDIr6e`j<$qEE;<=ZCX*yt6_Vumc-fPa6QSdxB}++$j)lJ^B_{#}FysPgCbYlQ z_+&01IoT$=8$^FAH(pAjU6u!zMPrFdt8Y4RvYsjxkYl0iAu4O?I}rIdFXvp<>Et+C-b%?$aS6qC6TRY z12-c}2jJIBRDVfU>p5Aipk&psnnARZBykOtNTWF%j1&|HP-|GL8O@?#iW1otav78u z10E#1tj^UmAJOVos}t3jvJsT%qWQeuSC!f*NzuGRt5vN!xEYnW0=(o_lcMu>)k`(^ z(%gd9C$COk4@y8u<;{+TTCKfkey%yF=2-EnL-3kZVo3^bxYVzCgJW;3q0l$(;q;Ar zCVk@`MO{6G`U3br_#bmcaA~Dwq|)&j#ZI=4JFgPP)t;Cj2Mic5%FQIzYKJ@}a_Ci1IV2^-%gQ9AB&jVaS!n?Y zbes!HC#M1;WyM>xt_%tG5~TyxaiEiJrj{*PscoyxDu)9a1t>4M#5yaeWrzp5rx@7f z5+;MDatV{I*2EIZOAeblJ1SOy2q_mLMW=$s!A9c(Eh0D^wY<`0^NL_)XJ*u_K;L*7 z=v#6}&_U0EW!&g!r0iv!q%5CbdKnbJZ-($QzeC4qJm-$XpQ8^LPqq(;PIDVAoB&#UKrW1j^-meh z)@-jifmcqz!T5s^X)1xkO%q_LQ!azYM9>uf8N$!B)t{uvYds0a8QnwKxZN1a*FEiz zRJ9*F4o$*`F|HNLH#HAUQAr(_r{G!$M})E+CeLa3Ahw3`Ji{!(r^aujB}h3PR+pS6oGE7k+hkrIFdUPAH#)An!2=)M~AYL6Bl0v4=M!p zbyrycDy{$2S*-q>&BC7;m-2Eh38e>|Ox@d@*|8i-1i*q7sH-#5(x;Kw$uK~s0mXf~P6V|pS^E+k$7#3%{|D`% zTC|p{(gEkff)VVGu3d#Y!KNuS!jd(x8&^qet8osl)N6KG5B6Gv>(K4hHMoFc(PK5} zS3N?rwKx+XY_$y@bXWD@T0Ddz)5B{dcO7#mrro9Ws1`f^1*gDj>;)U)KeYll%GSeO zSg@PmwGk>jj`Z6Ib`2-1HbUbHugk(@IC;1c4}cWcqm6=N+6*^waUA-fI+Sk4aTJkl zC@_N$Qm|WXfk>)h)2~cF{;dn_Q_B(yx8m{6aPoK*94{$X5Xc*NUKIp(sBXK0aA}(p z0@Y{zcG!22%XXaMQSJ0?8#JOJ?xP)A1N>nJ{s~{>Wxq!L+zC4`kfLpA=q}YA9i_W4 zLij#MyHED+hK;QI#A}b{mi_kN-Po%h-)(DAQ*%o&x!EydFI>2QE%o>gaTHm*4=61p zejj98>e1G%c5Z13@bW$!0yygqz%smrgzbmTO`f6@w7h@p2e(pq_EX3{ci@~Vi2dHe zO(-JstXCAooVOujD&z~4+}iIjaa5`zXWTnDS4G^WAm$ywnN{RmIG{uPkReh!eGP!> zGzi#->ny|?e8NEpHCMqvXYp|SeLZ#rtD%N4*-W2G^3pS?G{R#K$ng;wk!|PU?OJM{s@} zAjUn-;K&)!`dRXl7_x_6S~xk2T>$Cohd!Gj{0EeDani?|0((dKeQGkI51=_i;S z$1dVZgi9jWK??WNYK?RyHChU0^4(>e=~~96ETCF(r|Nr!O5z+!u3$J$n`h8^H&p^KXD#GkPD#_8WLF!lzhi z=tKU{F&$l?>}P-lkkfKGIa}$q8ZMk$fA>fG`lB`=eMgh;hg#%GvC-C)K5K%cV`k*CE{}~TK z4a@VO3;*h8t!5tlj7K6oHIfgVx8G5WY9=S|;9OTJk~vQqT=**^WbU&2S3CtBr1@~EAaf5g?Kiv(pNeE5hxGbgW#yRo zJMNC~C2j_K$^(7N%E<9~z zP%`IfrZUQXf?su}Di089muCtgG?%u(y|n_Ii?MI7kcsdhUmb7 z@!&u&gTqsz_*Yer1pzLDBV7d{jv^>SYb-g#jV?kFg|kWRHW#3gE9N5?ArEhl;?v5Z zA_Fjyb48&AewP{w1`~PYXHf{mN2Az0r;$N$NAHpr1|bJ9_@MEmK_mUqAgnN6jbeUv zIf~8Z$f0ULBgcv>)dVPKF}L<*4 z7vb5_{3L!Eqx7hwp-~7udogGgM|k8KIjV;;pGUdsbrqxd zI2zOymLfcu@okSs*5@L`MPuCwl=N zI2$<`&FT@{%*XL#v;e1dzvK+q^!hqk*g!}_Uph`V5R90DsDqBXS=Gjoal%pKU$oeP zg%3auc!=p0Jr=WNbF1(n^rAWOLMCZvQ{SWSP(bv)5F2MW<#+30wYu>ocWOuv}=$gVgBry(45+I7m5v6W|Fb|=oOt9&_UA3#U zt{O?g5QI9Dyfh(zkYu4goZnzd+(`x=otLBt5x8>$R&$+6+Z17koT3kEc6%zcsLl=e z4F2;f+{xxt;Rn~(xN5T*U-qH?iH2(V9X~V_#v=IQ6BI~AYd8+vN=i1=FCYMOl#==9u%W490WCNBp5k@2UPJl8s;kGKWJbWA2Oqgi&iedFv zJ%-l4&3&{Ei7A%S94Z0Vj^V?fPg-eB zuuf~iUGQm5DfJ&5v_zy)J0Tf&ih*X%kNU4^C)8Gt`g;Z`g>6kfZYOlZtz&rT&;Yz( zj{vlIWIbBM);FRn3lo7vXbm zw3N~YkZzrXA$T2QV9M2mF5TZ*fbZ}bQXuiXfP(OwPa~13T?BvQy%1*Yr7^5Q>X^tK z_b*)p80vzt>EP^OS5*-b+f!&`yuyurNj7%BZkn;Dbc0ayV+^1D%}^~R*L0Ka!XV>c zocaN$HZ9tw!(we!=pmfMuCZ*)MLc^t3wI5Ol^goho~nnDPkIV-aX66BZ&_F8Y9t49 zg+Z<@LNb8y(H|x3!!*UwXWY5b2%ozKNyvd84EfSL0Jp+e}V06$bj=8z}W*sUUHWiN$k`e9! z$L^{1t}PMjq5fPNjjq9?py>jTeZ#FF!3EUy&=_YV)PsC7N_ZP^Sz!;3)=YAFw6Ml? zVcZL}-my&Lqlogsh60 zLiaiK@0|N5Z|RU&^KSVH#A6pLZ*o_4P+;NX*fdEX2tL^wITa#Z_Y#P7b*wa0E|~^e zz@-RNgeD5K<2!~3Y*nGy48EY-95Gv&tHHj-RACUO>5bc{=#XQYCcLjO^E@Ml;XW#q zGGmv}%9=1)Bb~s`vdFsMM7!ypAyl(9Gs#Oc1lTySvMQbf3eus?Z%<)$ohht2t^qt?=CtB zQ>y7n0>?uxfUKI! zRAQ_T&J~(tcbgh<>VbjL62a!t_#>4OfM=8nRzRaAGU*_4xj=*$NuLLQBIgR8{EywDAsfIv#Px6j9asRPd7z{* z$g+@eWF^d62!pgyvZ$CcO;o!w7|ltk^O(Pu_fh`sI;za;o zT@_y2t#Fl%dBZdt8=N;oL)IyM&*Q&TI=9mjx8D-@+Yo?Vbc*`rhb{7xyv1WqO%( z43m~@5hBcTo#^uC*y(3c^~{B{o~kZ+gfV-IFvRuccvjEFRJe^a*s6twwp(G^JT0ED zXkFS`#l07Ex<#C>!y6i1(Hjsi*T(ZjsYh>UbhW|Yu6sG%Zc3;2;p?}-+~QC?Ur%Yc zT_a&F=`1Jtn2Xx6T@wYtvkSfiqUik-{YH50Ci0B8)d;qP*-=tS{MMC zXDCQ4IJlDQ4lu=t-5>`poY)2ZGR33o+7JnKpjUQ7@Bk1Sef6&*-7#nn{5A(lARUoQ zHuwS~d{_ZI6G*i;g#$PtfkmAzPZ65;31<=Rn!vjBuEgjNM$!~rswj5|ZD8F4cu~^@ zy(Lt_^z`@V4-W8=jT${Ldw3Bs}YLiT-oc;!RCZCHd5cS3 zhvgUNvDdOi%07|<0^GguxjKQnmKkJ6T+=&^1YlowEX|k z6`jfcm|l*mTq&&CY(>H|58+Qua5{wVD&LOUl}w=@d_DO?IEwH&Uh`}j;w$b&?H_;_ zog+OTK=&hq`#cnyp<7Pi>koxS6ij>k_J6T91EUrn$FRSIdk8)U%Fc5M!-hmM^RZA~ zy}Ym{f~#^_D1bb;zB~>4C9+Oq-V+F+khlB^%*$0WSk5t%-=7Gr%Uij0bjtQA|HNey zC19()>CQ~(Mxiq{iNP*&0G;Ytt9ZBSX8+Qm!v&Xk51gO)e24B2nrPqu6exGpexXY0 zoA`g?&<OBoXmGA*RuFN#qyV2R%&@n+oC~mX|Aj-oT)lD$R!OhstWtW`7vyxcNpZ?gT z8NBmeem$*uI8f16t%q4;(HWNkRcK<*x+GJ2@C!XI8(jG~_MdfupIrJtu0Z8da`+u5 zw&F_9fevy}etIX4)Mz94kj!%~wf)pzl!OuXTbc$RkZ(VADZw8k@ozIrK67b+ZcrZV zc3c;7`7>x#ACMnDbLn04DWPrQ_om=*(|H#&8FAjlMPygL@btX%E>Y+KQ^v}FRLW~` zaq)UfQ13?j3OFI39 zi@RMPH3OwpLt7pc(T5k`N%!Md!yCr z4I^2BqTf^TjdjChW=(Omu|+cL8k;5aExlf~#D5(Ime&$R1Pf!&Rj;d)dddf?H!4;2 zI#J<#Z1-quw{4ZH zpgSf1O?!%3ii<48g9}Dm^2g;38!R++$^W)gKo3E zVi6^$EHh~pA})8z@*%ZD#d%e-ZiI?ux-7b(`r|a%vJ+D1P|)LdO^p-K5_AiYAee?h z9l~2AZ#0|+2x=`D9P{dm83+faP>tb1?$#6EkdrW_x|;}ipM{Gvy#rIYqpQ&}gbaxg z<8cxd9|=F+Y4k0$RsE&Lz(SeDg9veI6|QeaiYuyQ&5jaV&@8I%6RkXNk+Ro7jgV+@ zFYceh_l2XyB0PkZ+*IucwIchI74^k91?k)RVltlm45Xe7#CjA7iVREX8ybLANKvUe zUT=}>ZQ&OD^cZn7!fR9b;-*)G(*4Ue>{lkbI@-pG6$tHKViVhuPVr(otfS4ff$ZJf zs_>u_#LG#d4WCWn$33RlL^nViVHZcrXppK*65*41XNPbsNZc`iph_0#&4S}j-eQE4$M7SS9(?@ zJ}Ke@+&q;ZYv_>5<|GSJMS9Kdic~R|g4kgl`EGXH-w9NuZ+Yt+^!;e!*?RWd(rYTwzPqM^D_&nz}G#g0XC*!g~ zO&NSITbz!+*T9g^7d+wq^~Pc+iblOHSZ`43ZRF)9@I$O3nc5!iX=>G!sX`6yC}=9) zLD;t;Yo)#oNl9}t0q%3lYa<4bkDH5uu-~+&4Q$pvZXvn@8i{NnHdWy4vfGBvaQBwt ziwb;POK}Kl=L`>RB{l^3)DFONek-O2({E=hu?#M1WBQe}R`nxmT8owFWvWjHR6<#$ zLF(r=Vhf(>uhkOVBTT{g<%aw&;1@sDoXP<*;3(}z%D*+#0`swJ?Zt+8A1gW9M>0Ez z+3*0x(f|SQeH`aHi0ut9Xk$wC?aGv*@y$`vRcwUNWp1VO-NapR-5Rx$X;D$6(u^;Y z_qvNVj?@AD`3(5f9^glpx%|t-*i%eZSoZEI_Q1Z4SoqbmY)EFmE~Y40Z~~5vNt zEF&4Jv-DlL;#|OFh925WGiYfq@dD0n#GH+;FW}aUnB49SIN9W3Z?S-asTVELN+Vlj zUmvkK#seC`kQr9L)AAtH$i~dg6JIxuX~YIhqZ{#SG!LrAkdffW5L`*${$h@CI>(&a zh!3rP=&vbg93b|`3v>!D<)z!Nv0FgUK-kR!B&wzkAgeiMpt!a=-UDjVUA%8JH5^cV zQ!uV%+#P?75#iY2M~&nk0~rgYsU+jaiuGj#a(Jv*fiFA5caMX{3Jhq7uO#;IP=}ly z**#ti#D6+t8z(@cRInXWCx|uRmy*+*1fHD;y#z<0589P?d<8;dC8$rod2x?YT#8%K zQEQSo4Iw+ZJ3|bE$DzsMFlL>?=O9!}P#J`kGt<>dbljc}!(TLm>4;}=<>F?F zmvI?Wj%2(FxHCxWSD{m#kwzEATjC8)AT&!X0Sr)&wnx@rKba-w8V_@w4$3gVkE4^XuIDE&=sp$=GtSw*qI) zx|b8&QMg2eo%Yyt))yTltfo#7aUXo#RI*fD?mCSZUMO-=ctbktmFuqoUh~Q8E8*hv`CLC?mDmJ3oWL_yF)-zZ=chCO!>h%4GE168 zhn;6vi&6LsjS%vkx>7)+UFjSuauUh@1ZsQh8fbKIMv@vv$_n+6ih?8Y{d8j63?ce9 zd2zki2_CEz9Ep6iNz8{&DRQ=eW8EgRUWaOBkR~KvJ|uCQcn_Cnu#mtQuBHs( zXFr;%nB`>6cJU}hqJgZO%-I2hjPeY&tw92Jf=^8%OjN_2;8^%b20ueGL}kE5Iey;- z`r@-XSr?cr(r+)QdzQ@GBjx~@f5LIo2{d#s47FuSw$lB?2|Thw^jCh;9I$AlAGN5e z_!|`>e9KiUlb&8}k1d(ZC711oiF&O}T|M62FQ$;#aDyAUw;wwFTA9qEeH@xaCp$z> zj8ilD8T*glW(EKc^KsmKTZA+At(b^#GW38LiC<)LOpC6m>a-%C9DuOXimdwx+Dj|4 z|6Oqn&@ioaXxygc;1WX?`R*W$V~aCcSL$A^zH=2i)*TX$BRnIMkAUZa^jfgJd-I6s zMlSz91lPyb8oDUA*;uq{-hgK_c^7!_nCMnrp=ZAV8$fOzg%Fpu%HT_WJ1W-D2?7=@N~~UKNXOslI8fgD z5jP~+W_aL-N+HdTh981G(GzZ~CnVy8xKT^pJpn^F6$U9Yx$u!ls$^|F38VHW+`eDD zqO&gxL|LcA`v}{!*g)EzMf!X!E`cQ9RcV^V#&TCQDuA$WV^jhb6HlFiIlMm0OqQL6 zIlMm0hYUR@5}HMugpTDT2PguYh~t=!16pLFZwv&k>ZY@6U_nRkGl| zNh3M@IozrXs8q_gn&NW1kN!f8LwFk(wT(>v5=23YBL7qsnM}SQUNRodVqN6>S-gw9 ztH)v~?ePiu`6{%lCvr{J&Ss+qcRj+_ z*Tji{z-F>qzTDP#Ul$L;XV_)m!U$={4W;AqA*XJL_ej$pMED`GY49~f`)uBw$9)4W zyjeD%!q^9?YD{uK8R3)W{*+T|-uwBG+uw@c8;i4< z3_{DN4^N;CZ6-Z6<8ljCD-G^EINvgcJo!}&sIIJ3>hj2h^`e_w;VX?j2ajOZVG{0h95wce&U4D`XP*g6^tKA z*+aniM#I4SK?U7fQ)XbDh?NImHyp*=G9WmdaM9-YgpW;`vSZp;U9MVjP)8zQNGcbuDMH znzEp+mIYN@%}@uS?Zo1$md7#K)c}F^U0NXcx{cQ^I?BpKEN%u*_`1#NW@x8k@Y;pj zoT)h#jC*c|E-Hr1;|C1`K!je;t)j?G9y+7wT?7=O-<0k!DW)+yPMHkN5&ZDl-P(o# z5>wp}2P=TK8iq9N-IT1TZWs=rkwJzqnvC-_z?H>tAGRR?$+BR(LdLD;#egy{>q*yp z8M-O>vidKo_(e4gEp@#2&`;!=cpF-f6mLTp0F;Fo;3jl$gFKD#Ay>RXQ8%*R*FcA2 zzxf(6;gMJq4pdz5F~rj!%A+Nq;Saq+rP#;>v%$Uk=%!k9DuttuH4T1brLVzXKVZ{k zv%%D8R%Ek6s`P{~YRh2qec52I{GCtWUO$lj%F4y1-w?(F+Z$nnD4rmmNKw#H!V2(cG?tN1x?QylEnM6RhFi$Z z07Hf&#m`_uU_{RlLsNLX=5L4~d;Jam@SRg|fT5{gW1b_AG-(~zIwadwXPj)Ab7pm| z07_nCbcp-WA!ZaZD9|t(vY0Q)zHx!onkiIM{1E9-8>(5Ov~4nz@H*s&I)>V~S~D7UospUZ8DeyN>9}uekf8;o)4UspJ%cho*pTN;i5f4l z7#i!X#WHlp!o1L6S88fKCeL{WYBH~m-4;V2lHZ|fyahU?piM(x963T=4>63zU7NAs zJUY}6jQeu+>uUz>O3sEFW-3S{!T@R1GmuV)8Dc0Bjj}pwE(ezq(yp#yKs`%jG4#d5 z#|-PAKWemPa6ary=F$JKD>*E0AoS8Bq8hl7JNw{B=|^=9r}12_+TMB~bup7#5~Ha$ zx0$Y;*8MlUbu+GihYt-zk;1)JYG?ZV0I9Sv)we*aCa1WAiP8^Ip!U z6FMjawTe!0x%XHbZfJq`H{;*)>~U_S@O9xqINfE5G{A2mL>R2l*cL_@GF>f}R+PUR z`8>jqk!qnsn*l|mM&%DEwu~H;Kf1V}XgJ;19Z@topQ;f(R@R5weZ;7u{(1cih87f$ zCz%Nl)BE2AeC_@PTvKbqw`+cceLE$0KU(2El zK0b2cb+rgT6$Ahe1)!=xnbSbpm)R-iAMK=L$w_kmnS=@B9;g(wkR1m z%6RdgaqDL4l>BG(YQQ?M@K@Ss>iZVTCrj*l`?4)A*-*{=pYH=&Wd0i@Zh7!?aw3{_ zz;v9G;q5jevPZhD*UR~%iY?vKZP2=v0zV6+N@B-Qyh%v1A(-s`7FJLu#45uNpU$O0 z1;fEV^9EWhMK4+Ej)(-tTmp!SsvE01tx1okGoEukXdc>}dSa?)6;_I>Pvtx7S-&~^ zcH9&mr;PtX;rBarqijgBbIlrp5*AOBBlTHF8du+tVYFD<<`-MWK;tfmF*I?l)_XB+ zdQBOQzSZ#@*5H~nXAQ1Nb7wzfEtc{$D?D|@Z%!zrx(WdUa5fkow9#Z!`Hgi$3YO7~ zf@vV@sxJqv<|VPvuflKA(%M?vUJP+>r&>bpXO_mo$jyyhk2N$Ac{o`fYiO=7Q*%Od zSec4>nTnq-6D*g;!VcNaID-+x;0Y30E`f!BPyNnnSN)DaWc4H#t04uIa%m4%j>c(Y+~fS*sWLp{6Pi7C+3eFlSezVNhHT^ z$Zc$K>oqj5fBw*kT%$>rUd4HXCfX;t^~xVUP(xNVd=8&l&*39&DYU>g6f@kE&6Zjl zsZ!Kd&PHPP{Yfp;bM3p=%lbdCZ~Q;mSBFr=z7JGgo@?IRJ&x=7k#k)U{JLKwL#1&FmpZuxTQ%}H?Sxk2 z8|35oDb3KE&X`(d7|!ETu0m-G(lpbMLX&g^#C8@}^D>{+H~^MUmZ4Z9Vu~Kg0!39m zENfA=Ar-RtqWJsI;?uIRVT3agf`yhQRf7@DVZq3WKg>o!`7B+3z%sHS^a38Wj!x+& zpa!*0&!&cKD0N0NLt}VMZDy!xR3d8@dDPU9r4NHVhe`gKS;+h|wX1GL89KDajG3N{ z=jg~VP}ArMCt;d_8Fiqi=h|uLcxG z{M#9VP?%$67lQ?n9iv>Tp=ZHj;RI02=bu>jZmB;?tDIH+q>t+Jo-MfzK*N7?R`q}K zbx?2myC+uo9Tj%1G+e)IC-b@)+SYo`*FpdInMJQU@S`lxKhExe4=x%d{g+R&>ns)3 z@|h>uJzF}Rj5W&N1c8gfuRsOh(qyt~5lnzA7X1-fm^>*5WTT_Cv1A>F3xg)z5P;{w zl#GtFb6c`Qv$?+7p;;X(WhnU^s2bKwpMwAb2Dan}VQ{Lv8>m_)(};bCw(xOlKs9lq+k z^96PdN^!!vF=g4)c-4~)?#9=soBqGHu0A%3Du~}b;=m=XwGd3&Lh1DnP~l=MQev9M zf)RIbn+@t8pb-NNXxTJ;rZJFU)0zteD^}X@jHb1g<0Ht26zdiQw2&T(NKayc_z|NK z5_2X}(-=u(bY|Xuyd(W*E_*XO@ArE%``){^H@judE`tW%qT3<56e?sKps{u8F^*`=E_3+|FLy>K`Bj5;58Qhj)SkId+UL9xPQCb#=wy;!XvHk=Pk zSD13oelXBV)k}=+hlj;hQzc`*-3w_b_r2{+!xFKRy(RLj{6P(e^k`hWn(zSA2(i$N zqMhMEb=wZ$W}E0Wc@pn6<(~(kCp_edFmni=rLV}tDlpMVy0`j zuq`Tfdg%LG14MoQPkLMG{-$z}4uX-vLe)y##%fi0@}wg!H3spKiPD9|nB+gEUh zQYsvpB4xPo&tN+~$kT&xZ6t8(Lzp-V{o+S20KfiyuipLd99)df$WfrQH47CejeZVm z#5kdOXsW6S!v6-IhpH0266jXwF_)D6U*I}7r*LW-FJ)E_!MCEw@`kwC3&UJcLma0} z+~Sm(W2ux8PitnJGQ$^PA}SJ|o9RuOUb(5rKt|N!`J?#98%q{k0(@Gzf5^GOfyb{n z@GC7fuyqEGP{V<*wK(vVp+*sL7kz~To7n6@s5%pziJd-n8MmcdEzZ|BzQ${7xAMRA zm%yskQW3{bj(GQE79Ce$0r{cS)6Vq}-2V-3%ZpDv1kX)-{NBDCxe7Z(Ms;FM!=sws zAH&u)&Z0Xx<*T~d7d5F#HPJwmY6ACLX--YhoGL#afw{A*J%b81n1`h|;jCLBYp+9@ zye|vIQ*Z{AC0X1R8jvq%VNTR_U(e#ODyG)EJ=fsul$H+#mb>PVy>MA1fK0$&B_{6nIn62)62o}Y(eIf?DKA>h15q7GNbPA!Qx zoGzUf65nH1IXA49NTj(vwqXsHhi}`oZfh5IfU~SV60scuW-}<80oe@3W*|0$uo-~O zjoaL?&5hdJpv{fh+>qVeSWcdQrc^Fqx{V8%Ym1X6?o9Se;?87`5O*fKk+?J2WyGDy z?jr6?mfp{*$aWIYZKjMd-5Kp{5Dq zLbS_tLquKV+hU>|DW#&u%3F03HdnOh$@0IXDhLfCr^g7v@@}t=WLyiOvXU;}?KQ$I z{)j47>L}-Rt4?ymkx6dm-#egWXmMe`zQYF=7ljWl7(R61Yf0Yn&gyl`0|O)h+2qim zTD6`e6g`H&hl5b7(Rsn%TnBKSQtlD&Nw zvI@*kzGC%c>w*vS*4-BDYnnv zZK%?K_yGxk^#_v0I$JS3CU&*SKh`;qt7G}@tQ`4qG{Lc0My+GygE4in4U;u7$ZmP0 zQO3#7j{J4qf#&X*m9Hx%yRKln{CeD8)L*v6pE8f;fKiy( zn+f*vjD#5EHJ|6r1BBXgZmF02W<3j1V%SCAp zo`AoXZGrNIw1pn&4bCTkM$l}*Ze5I3i_XFF(fa}Af4Z6=R9#+^76pI%WW24tmt+UZ z+4U#m>H@W#PppxWl*QqI@_+_?5Kb4!-9J&V{AOkm&J##kBDTp%Sr(Ko*U!p_Fj^`!SEg^x6PU1k@nAx1Fo!|2j{XfGih^jU&5%A@hg=B8?DmbsiOaCw8SHQ zxUrJPr2;9~AXbxqYSL6b-=q`AQgQ?3{!N#A&}B9&1{jn5eKTG5lV%$*I>uxllaIG( zgpSQ`osYi~pH}lQk&SrgN)J9OE9tz{dnR_dpmts49W-k=Dxry=R-|*kY2; zB-fQsx2fS}HZey$Cf{ju+=JP@{4JCLb^B9onM%95p@8z+Z8H!~Hjy&}P>%IIL#^_< zncn_5!Z{{r7j5thCQ=ZM?Ur9OtQ`4b$7vLNw_zm8k!N)}r6lNMNOpiMcV4IGkckB- zP%e3~GtDYU4wjd_oQOw0hiR)WEp;l!=X94Hz<)y5OzOI<8qah+i11?0XG3>8UPDb_ zlf8Rb&|*2LM+X%~W$ZD+Zudx}FskulIj?5{M=Zjp4M^o}dwq-mJfW*yZr(>TmOQY} z1AN^?>}|kSK7!r?xJvIA6d?S7tCZdMON>P_WI05n<*#_t?Dj7wsZgp^HBV^eekBwN zYSG7veq-r_THWr6xA7H*59EDSa>D%V)53g&XHu$cx$7XkNsJy;hVfF7)yGmfxi}6j zrNxka|LTOc%DF=`AuW}RyP_b3bqX&+F-{44KXI3LQ6I$w_=1q!qC zL;9m{6lUXwCMfb$yvsueb3AIhKNYm^h8?JY_WAH86}7nDy}i4x?KBRe3c96l;BAXNggxE8NX(p)J)?lbBl;N-r> zJe^ujE_>}JYAZ*SLRT>}Tz+>n<;Uv$yV3V>ABlBYpO$&gA5#ktlt{q>=za!|^G1E- z9%F_8oQKtGW8&~wiG@|`{L`AeK63upO%$o!HJpp*d62(0ZZd_~9|FUq@zm4EWfSPr zdi+VReN^+%!abmt1MMpBSPGT+K2@a+J|WSDIDd-8`aVTsP6t-mIVn!wFe!~<(Wdvl zj3zgriGe|}a<|EO_&O!4Ynd$#?f1%vp#vApPLPi-swLl;>>qHIn`x=8_WN?xH(Zn& zq6kSIg(Kv)6TIcwZ#2YDRp}H;gC#tuFJw=N0GWNJY{Nm775yOHa%zHsOPG9Vs;_)# zY95X=;ABoqR&eC0)5>wC3W7`z-vdFk@hb98Pwx%o63X5__6{gjtlJ|J5( zBVCYJVIMCD5;>MD9{L29pI^`x1=OUZ;8FwP+YE?n1JUf7BsW;p3B|}|i{kNpNj|j5 zrTUA%SyZiAQ?5(~qMGs$5`}Ae$purkLT^&GD30v33TBHDZ^Gg{!1G%bC)*N?5L#ye z<^D@V0Y>J)2L5I)x8p8eG8&LiS;=>zyyV-F3(XsZsim- zl`)$tf3Y$P-u+gwLS(IKq!uDrSFVd(u&yRB-nFV7$R%ajC`7KcIv#B_il>m3(YPMd ztaTfxTuG-)n=AeCe%B znzH{UhRPGRQKQUzD<2{td{D!wwys6AQ(#(8zxS6$7`Q#?eXUqS!sVc^k%xqL3XKwXYT50#Klkt2r zi%>l}BJh)K(Ri7ejR>r7YZB%Ionj;p?CpTPonu=J*s|aDV70RY<}{MdmpgB74YX`X zV88|4x8K31IMONk&mB|r#Oj^>@Xuy8VA`?Uf=0nJo$zkjC@%gskCp5=7uV3tBco53E6076LCcT!y?|G~RC=;LPz zHy*MxgfRKfLn-J71)=}pSCxXPs#@ObuP~bPUMPB~;QjJmRt3BsM-mmhKR$`VK^EEW zZU7nL(Ge98%a4)@Ig{VdQ%EnqAB7vJq`4pT(-HX?{Ne|lX% zf=}sGoKUmkS1im^Zhd3{!1)hl9enyV0FO3;AAHmk_p~t1{VH%S-}tcufx7~0nU-xC zIMU}0sQ(_%_l1xE>~4h$CC{x zFWy@AoP1kJz=Ir3#t^$cqs@rwMJwMbU;e@_w>WO0aARi%ET~(r`doGE!=Km3xQ>-E zt|QkxGY-{}%f8@ToWLd1TF1)!D;mV)g0rD=m$R`71$|H--hVclk;O7&qxO^zV62yA zfPB=C=D~ZdbE9-(((Wv%e(Ky*J&W;s^UEa_a{7FgMssLw5B98+fbw6y+Ktc>rZv=NLiVzK!uN z{FjklUVSkEx6^Q*QxY}~z2pZh+g?gk7%#q58xPWmAoGt)#Xw43iasdGip!tTSL%Ap zudr^14a48PvfaGQhg-x4MohKXO45=aEohPa_KyyZYZr`B6*h+T zt^#u?yPW)oHw+w_-<=G9&-@hyfA8Miz~3q65Lgd6dvCrm15EQBCNt&8Ka}kK*vBRrqSy zr;jW&ze|pg1U#vUrpxh6mGpKjF?;CTzP5M2ONppZc<2(|KASg8`sw&5Qv z2RD+pKWQY=S08E^2q%_cPotQA?&EiBBq0^5YRRy1rJIY zNZv4^S~QQ^cB;X)9curQ>u5AOza%ICZi51L7;840&SzAks%d zmtbu?uQCept=Z$H;z2JD8$B2vW4es%C8J3}&t^^>RuhdNyDFnhp!>Bl>W2q$y$6xk zywE9{qNN*cMivK^thAX;;OMFX1Ec|!doC#4{%lq`*=I&C;uTeCX2sY)ALb^OSO?C6AJ)E3sgDs5q*WP1woB9FY$ zU9W@OW(R0l63LNIu^5!6aWQI6hP)#L1U^q&cV8>|pfdjK->CbkQQZNGx@!k$fM|Jcef>qQAn3U0_D<3j38rg$)&(#cI@>P8V<})_=azFqXAB<{|KEWslDE;ap zyN?0AdsVu|I2z61kPvd`q2NP`V~`E) zR>1D>V^9jE1e;`9|3eR*P#1k`!e@M0g9>_HgZkW;c6+sz7FDXa=oM3CRpk>}4Hf<` zx+u8M?8)7`+mcekZAqy;y~E)(E!;+$C!vx53yy2h|7Two5h^(D3EzL`xb92INJiMp zkM%d?O*@LietxXOA=mBTVE{3XfxY}_)bqKQpjiz94jqq>{toom?DAucCCyKb&(`}o zro+851w7B(gd;WbV{>V8DHX&x{HVJxN<;7Bc7AMF(9Mr_R3zI89@&nxaH0%&%*#em zq)$r}i3jUopx|!#*dU7xX@COUp%2j;h`je2HoJdFhu#cNq?+YeJ80@V{K!#B2qfp~ zqb_EdYgPtjhRr(5>pB)?SB_Wt@fCtKI>23$2{TW;!;dX};QfAVkAzgoL*8fyiOEI- z;Q?wl!6ytLieI$}st-@LLNG7~?ZrPB5P+yhE=r(yWvijd@5n{<@Lf-Y;5^h@Wgx@U z52DD?Jd{akj8z6ACU&)3Khbq&wsw~^L{SKb`O9T*wL#J3Ok-%8VI-n48VH$ed5N@= zaXd;_K~nJ z`?KW{QZ*mdLQfbc(mNm22f(0wlun;?$Hp4ur3M@B@Z`^Y^dSxk;O0JFEX3p50o3oU zopjV7lG6^g1}rwNrBe|H!|^fV1h6JR^4sg$7PkkdY8n6zpKX09OP5SaI-oPSGh;_! zZ7LD>4xmGD@@_}u4f)#K&4A-Y{5qj%JXA-tb`H=xG_tc3sw-zFdy`$Ap#C{;T2Xk4 zPC;334i+LQ7%XcY`m#YMX5VX_(Pg|cfSlOjunx9;FaH@Lz2_5g$Z7Ql}Ow9znckk5M{7e%5)JF5c}JzwmJe#6#4c4C0c>cxsi zChh~L+R_Vsfx`k>vs%&{T17M|>j666Tdpw=Bc?uR3Pqx87-3|49~d}msR;$>RhpvS zTJt5X(U3Mz#mc1dOXwPIOSu=Ju8cAHyf64jTk=g`G>Sr~zV~%mTwnLPmr*prLjzg3 zJf>-))JNS{&@MbFkT3Vn;%vM32hoscJt5VDC*$jXQ>Wy1jumkgQe_ruC=Z@nowOc+ zCZV^;=>d9!N*;)Q#e29p(hJes=n%;p1ogd#guaTtS0OBXoMsTn@`!gpOXFRJw^qZv z67LK`+wrwP7VhqHmBWj`64yw*B9uWP)Do!%Ie)9Y*YFti-CBD==I zc!@2OcA`w}Wb<&qX;qCIwpBHz=Fl&7*6xTAFr30g)p)zPIZ{{Adlb5k##5I0Xd34h zKN`wBo=kiV#d1h2q04w^sSd{DtC5TAh3cfuXjGXd=pLXY%Ay&R6~?_`G%VKQa#j$Z zwJmfo?_V5a(0GJC;KEN5G!D9^532DnHE^6Bre=&o2NsQk^GUzW%Cau5$Yt$#R0p9u z)XE{`aw&9K5167pXW!FM?^JU)nt%c|w4oDG6t)JjCTcaJk#iH#5qyW4(8z#$wfmPz z;BGh?R_dUKbF~ng;G`fn2&p**)Qu)_Q=nS_Fg+v$hm>%h2XyFE)D@t79MDk*(L7`K z1`KFR-7<~UJR3Z@PfSB0$UHiT1-MZ`;2LZ>$NIId{csXK11-T*sCGs8P0qjjOi*kJ z=|2-SqY%pX5@g}wL2T1;-z=Cf;0-}6>R5l#3^qK7jSP^37cI6#t8sBZq{P0Ut$T z$to~ouQK!yKMvvpn+fw_U?Wy1M}HLUs9Gl$8q~VE5S>6T5WiI@$0u^KRrO~@xgdh>;V#vIb6qlbEozDjs`J6z zmc^h}7dU_Sv6O_mkTdIGs^-{iAZNy2Gz4{-Pmkc&$>|SpHPUtN$Wmk`lUJfS=shaFg;G6ksGiygog(vA zqYn6dbvh`pH1$!s8p;ue1)A0g zEAk-tF{{!gOtSxjz(%6LR-u|IaYU$MIIkwY$$&c zb{LJu13^Dn$uL*Zh~1YDBVUY1bHzuKprdGqlA;f4-oKBcg`QdNlkcNoMDu7_&f0W# z1lR?-UB^%s#_KrGb?zk}p)^F3fW7Dwtz;zp6LcLPWLWT^?iWg09EYLVC#2*!G#<#) z_K#KgT?+qJgVQ}sibvOhVeI(XJ{*dEWl-xuY76Cr4oNaTzc@x_Sn0bOw=;z4X zdblpG6hej^M-9+Ha{UAhl>kHw^_{U$V88(Pu#;#eM!}@uE0jaBPU*$yc?!AA2_dY0 z;zHONS@QWQXn{8fv;3`7=m~Ohy7)6XU5hhliMctaYZk(0<)r3Woi6h%I*2+kI&0+b z1{+gXBrkq}`kMQ5>V6^ID;%L!wTh*(1Kj@SP#cVfQ6?=wA67d@zCwpEo*cqM%i;^L zDl#*Kh8DYJphiELJiP$c04%yUPKr7UwaC#6D2c582319q$s{Md+0-lk4O(Ok&Kz;Cy5NMQjg~Iz1`TYl!r5-i$(4p}~C^FZ9Nupup zhrT9>Y`cgSR0NE<1o1{gW-7M1jOOZD#2keqL9duAs2g-Y6R$v*Hklm0Lc^qDB3SdgT%ZPMar<|(X32mXv8 zk`{(Cs}_=0S0VWD6n#*u*1LvgRLJ`E8jQF+v5s7aDu*m+03_jt7CYQ6Zy-4RxgwOs zrhnaJ04CV_FEkCIZRF@}7{G471&MuB^GwJZ_A3}^8&_!?iM@@MaJ+nUmZ4A>J;Mf% zc9#Ol#5<@4M~G!-p6Yqi@s|6@DF2FRS(wqlU*3wtp|o%nD2ArT?Y8VmUu8;$n{&e@Ozx1;3|<6pQN}U=7lwsO))a@w5j^STOcf^JFjcjSNbg3gd3QjhOx-`B#h*ZgLTO@1|$|f zVN!h^S%{y6wFp)s;iU+Ll8IIruq;V{^D}iwJ1?AxLu;}zSSZ=*g?B)Tpn75ynP7&J zDcJNtCAl*ddy}2ha6{CX(E)I@H_lUV*uk0?jSMMYTEG$u?xZrhWWgDzNX6oZX_$<5 zXrjv0*9SY*YV78NZ&arCm2i#~SNBsLQ8C3^)o^pK4KMP3K7iIITFs(s@%Pz zaFnJ`(jq9SGDiA6l!RnZ8FcJ$8>tgU{=3%v%v4_;+EA^XVp|rfmb{~%gop5O_ zwk%>PH;^+MzZHkSL3nO0zOcD80mtLTwRj)ZG8ubUnM*kx#-EQ zb24tCkN|@t$=Idw_P615^ho7B-G)b0;O(>SF=PEihQr<`a<6!eX5 z9y)yHrz&MSu2hH9;c}#=dec3y^bD6R z)t+JxcO<2yxiZo+Qqt&-n4StxPjaNDq}p6*_MSbx7qHD*x@$Y^zp=UNLn7}BPLDGx zx*=NCe%>aTqqfH;nU`{0&JouDrR(-tGkVn7I zA-x;m?djBY>`CFuKUn~|9Ng2U)K z21A^BuIudmH?9-rE^mqlB7Bdz4m?Q17w~pH$>Mfgb6iG2Z1#VvIrxIo-&tG0H+uo$ zJkWjE0>|O7a5hLI;VmH$!-%ydggX_1jpA$u1PU>ZiUY#gtZmUqt!5QD(_7&`adSlu zW7D3qu`bndn-if8ei6`Q`wTmo(*}E6pYhD+pB?$M4Ze!;P+$qc>$yAg#o^D}nP0m? zQ*==H|55bn(g;3`>fSxkp5#c&NVlh@q=YAyrho%xxSUR#lg6e**sgb^xokF9MmW4X z9T3mbI0!cM)9fyr%kFZ7ho>Z^Ix}1;Ham-GfSKxYIMX1Qg#-6=7lcRfb;=N^Tq&-U z)C`+59O7WA)1ICJq7_hTMk%BzLx9=R(o*axPUh-BmXe;4;z-E=5apikbfj>ORG8h7 z;z~_%Q5DnEY-#B!X)2Y=1v;msfG|*(>gcdB5MVlh47fO$iuMeL&5>g72}6T<9dU4} zW|Z(m)gs<@%Ac8oy>VlBgR+Vx40iB_f~!W*Oknp;rMSHOdg;B>SdlVl$<=aqv!{A9 zomr_F87p(DH!CDBuvXku?{vnPm6fT(O6tw(j2BZcBrRV|hxBwiSQCl_3i0=Fh$1%j z@Bg#sdFF7W!H1`Y8#^8^)TDU2kjlrt#KBl#b z|FHeRIkP`1<3s9(=Z|QcsO?k7q zQdq^>(rxy11V!Bo; zm$OvW$G|6r6^n7SWibDQ9x_ES1dOO`&`82*n=x?Itiy}tp-#F(Wdkm6<7&Io(wxA? zR+IF7vu7~61YMp|kq_fRN72AAhdhfy&Hr#t8yaO65E9Mc>eoR;pgD@xeyX}o~W zjPx{(*si`o2cUnUC}Ya;ZT~19m>UMTsGA~o$1b0j^|8Icc^A^ zy10s578jEW{Uv*Y=HgTtrl}QeOQj@WS!V`xo&ESLw0)%nz?(Uvots19=v0j7w56n| z99;?u^uSKVc3@eRO$KE{rKD%1rjg4p<8`IrfQpys^{iD8%v}`2>IQLYC2^)Q4b(`Z zJDlz)dJXF}lcre|Oi`-XLMek9V+<~vuDP1#BQC9OU3#6^HDlVD5!C2n@OjNwo#14= zx=VHM&}$XuV@h=>W>n*fOK&xrw;R3Gp}Cjl7PLNW-y51gl9kmM9IEv0@F5RWWNQ?2UOidt;u--k3)*SEs(9M-f*_h8h@Bz)W0q+J7jH zV^=E-CM0SLf@r8f)2ZEHfWeY3jRCyT_)G;uVO6G3jp3!-9c^k`g<+opPKUOoM-Hd1 zg3bX{RghC{s^!6~YM$MuSRV4g$JLMl7E;ULa+NA^1nNuUpal*nq0%Z-Q0OiiKj z$6M_O3|7#}U{FD4kgCSFR5dt4^GVMrrMwg&>AD>?D?mr87D87+=K#wE+^ne?X{?B9 zykr%@tDKsVzHuwCH&zDr))sUya-f-M_H@eY|8Jq;nc<#c$aLnp!#(%&hkI^VW|)ZJ z;O^ZQa1=c9UV!DWu7mJ$B@6!^&Zde zI;=(y#%M&vUL65D06CWFI)LpDlXIi-S9oWHzE6h_ zMDP51_37&!aClb7Cqhk-v`e|L8ti@!(s=Lmi}_$vdL zt)Krjn-M42~B?=R=Uq66bdPCH{$8M3FTh=MA)bzL{U~O>wNHsi z=hZ5TRZI)YGLfwAhD0)UF)J4>Fo*iG^w^tc#oph5vM@wO;-P)vc-oeieiaTUt;xwer}t6>Rl8@aq17o$rI2B|mK z;2dDU_8Q+q=Z z?Rr4)C%>$R_6Q()-o>9Z-hlhk6g{}+G9X(0>lnOX;znG9ytWb7(Fu=kgx~~(^l)B> zeFr`g7cHo=%BsbMb(r_p2G@c}YV7G>b z2Wf8XKDrZEL-;3-_Y--v3-^Tc_!N)K+^sv+=H0MD|F{kd$7E@_UR(Riu@hr!G}EqH zG&|n6=rF*oWb8hi1`vB9q9|m|}FMS7RseH!h5Xd44x8WcK*MVYCsrL@zSTfFE@TPC^Djm`3 z0Dtql7(Q%gs;gv?hj4clxm-tvERR)V5CW&ENiS+7rLM*uhRAk;Dz2Q9JQvM6{o?na z22PMM@8M($rhWc*G>ZpRO_!IDOI0r>9Ho=`^_@O}Sr-0C!RR0uqSOCV3M|aWLQJsR^ zzkG^Ez=jXgi7Y$G3<25%xZgR6SHsq6jBdc`G0cEZPBSF0Q_$W4Ov6SgDLDnBGcA>S z8aIFxS7)0CnfVOP_uzU&2UC*E*i=Vn)b=c1MpLwze&>m3+%y0>daEKi`GsZ@_suVG z8N$Ca1>r%~eu=x$BxNPVvXYGd3i<^A&H4(@*D;y9cK#Z7pjkAUMS5Zz1p@|kpc&_Z zQE4O}U2Hm!4|_F>W!2b(ocRH>+Y_ya*K4M4=w5pPufVuNEVEt*^741MD@{>ex4#4P zHY5>!gg|%I@9|NDUZqT0K%QS0_yo*$5$Bl4arR?k`B97Cb=I8llZ&_gT5(|w)!mr50QVn1;@t@EXPVLnJvj;=LuTO* zV_7d^DbpzSlY(FH3>}x`orXwT=PI21G7?^X4&kM%(CzIfZLi@x{QGm@r>?=-16>Ik zP!g_#TSFG@2>o>>6wBSej^X0Ypt?M`es>f6g4@si=qBz4->TN-->SZT3xZfoT~@5K zx1hgCuggzU*S-xTF=WQCFd$IiXMV+@xTPmN{x&G4FoSQ4sA5xZLthId5RfbvHRC9H zHM)a`AUwD(A4E&R9@Y*df%kBCb7@`XJLBr|5YRr+fW_S6lY4j-!n5k~IMUGoW_A4W z@A!gw1!uC1Ga(oL(0$;+A9w=ZqyV=3sks|D@h4t__k+#g!zRnx3T)lS{=!`l{sauC zo+HF(z!-3oo8`EM`w3a_0D7EHi0z@~VeW|!@fL({Fn!6O$C@VY*^hA*gdf&r1DuC- z$&n|xgeEm7v&PYex^T5%fo$^e-?+Yx&5T|DDX6S5pjiv8LZAp+5r7Z#q+Dz_vRJrH znCgj61|cELcOg;^M)CV&;oQC@NTL@802dZnzC!}_AZGqYNXS9BLmV4|ju9Au8GflC zj6!{h{ckn|m6hx62F3BOu=<6lmED)DG6~5FCOxEtOP_<&UKEOP-#F&rXLVww)xu4< za4BRQpPJ6rSSvx|oFp`ZHVv8t*Dh>LT>(DvhcCL(aQ&8C`0hWna913gvRcYD4jak5 z%0fH5kqf@30~FRNUcxf-i8$t>$AOxjum0VL#cUL77OEorbsQauTE5koD{XL*w=e;r ztISgLLv(y6js+7d4pnV*HO`}hzsl9|*AW=20C$>?Fd1X7cos>$$oEx+zHp^Lp07|0 z=NESQ3Yn1l)mLb)pumG&OG-ZQ6Pl6zenKZ5z#OWnzc7X~2qArT39)cQSmBZvAS6~m z-Vgv4&ZMy~#IsO$6?spf@PH!Ih<>q}KqESJ5%<$-LU)9QQw7;2nBypcova834<8=S z<^sdrQVpS+j>PXbLJM?Epx+cML{S3D|8F2xeUpB=h48j`KF5#K%&An~v{2ziulM3v zgCj5f&iFp1!Sz{gP~k%Bk=7Ni<(*MTHEC;K_T6f=@rkwQNl znE;)6C@GE-QgA{74HVYSE46~a+%`%WOp&OKO6zDgBCF~Mqj2K{9z8$Q2o#4J5iKl6 zxFhAwzn|)vz|OzSiv{r=$+B3X2|O;x3Q_P7>k7>Pmk)GCdmw&OSLml8Qf_b=5Sb7s z^u_Z$up;7xd;=EkKQ_e+!zcy&yemOyjkoF)RF=^;fJXbjq5rB_ij;Z+9P~Ywz^3Q; zGiDv)odiZW<_<~{%y7Xk3AaJ(P#oi|P3T5W*o0Tj*Akd_{F1;|fQs$9;pWad$s<7DS5c+vZ^%%8@Ty6mOFVC-UKw@!Wc!~fYZ2H%u#wQn28E+OlAEXMQ zUUfJ%45WaXv`*7Ry9cKUa}c)GV==dZOY;PGH<$1#!nyVM3dZGhAs#iSu&Lu-Kpw@E9!VF=XlA@IKb}cftK78_H4}}cj^n*y+(4qhEv9> z$=r@iX#99mJ-t6fvzTP~a>r!-<}0n}lSo9i;HH@DFyL?5LJV5Tb!105TjqecN#w;G zAsLWZg&L@pPUn1%umi?n^?1O4qA};jcVEpDe4%r$r*)m^43n_P2aA?EHPx#6gv@Cq zJfe8CQM}VwH}E%&g<;S~(~8LG=_?H6PwZM@k->w zt^DRXc~NsA&m5A-yrM=T?=BoiE6@|K7Q#~;lgLL!*9=hbjR5y6Ed`4Z;G|59-zdPB z0?p2uTsS8Y+CKkM;946YO#4!xYAvC^&v2kyXp;j>4;W4r4_Q zcLvkH;cniEhU+)DDE4gzUaccUy`dDD-7bmZ^uD|B2fj`VoM}ady{J_r$$U|OF?b@6 zZcfbvOrhT9PkBvzlE_>4rY<_+@h%XZzJe;GpS2I}s#oFcu0ntFHBR^oBeavmZVXP_ zG-=gMIDvoX64#!sC{aq{5h007eAHc+7}`Uah5eJ*k@E1`YGrZL5veCgt;2~UllVNR znE{D~laG4R)^7taTybwHGJwHxdYc<^^4ui$EstfP0S@H1aGNCN?)e3JVT%d`ySWdi z>y^ZB*!anlF2Ma)f$$N=gOa$r|NXK`9^ek^C!9n<_T+D zD99ePM?B##z6vuqmG%n*jAj};n6XLVV0MeHmr^@Z2MKHOO{hAU{of?52Mc>?iZ+5Y zBQEogjz!>+LNXsp)GAhu6-cs*g%TW)%tJT3Rw68!jRPzbG&{gO4sJ{+$NcyVI2;`3jk;zqbW#WUXc{d z5yEkSj)-7xukgy!DvS|bG?gr_I- z7By&;Uhr9?z_*uj%EgS*PU^j;m#pP$LOEW~OO`!aD;eoETKL+$pVRGABo@7{!{2;e zXl?$8!;f*eb#GyW>QhRvP8}m;A$&HOwYxF{ly$^ZH?N;Kk1Hg19P99yz01ZykiL`5 z+uB7>x&Zg>@d81(l8uGBH4~s+n#t*jKwXKnm?*#rMjM-t!LAU^rAeco0UdRvN~HZH zp})%I+$5n9d~c_4=`>mRfRe5GNP=tQbuQ|~w5oufI><1Jq|icKOJ03Hkj$N;n{3Av zftW|xSS5}C99oH`Q+4>NslqZm*`{~$7_P;loxuYsQnl-3vU{4)8?cz87y#Vhr0;a$ z1YG%2HeSdfMKc(0X3#fh2#3&KonW4c(i|7fhw)eul^;gpWxyj3s&J*|WB9y^=t?ID{NLp3T);D-ZZq3&2^C`>vn5tdM7S{LC`YvO=e>ze2c*lN>BE;5urx zDF&XnQurNb0~P(emxa}44z{#GZmfdXl}-Lw1v6v?L31wL%@b+uY9R-ZSW`J0XOK~Q zjj$h2bkN2{%xl3D$2qi$qwi~o2vewy@Y4!?)(SCrfsTpBkuz_>Xl!dWReQ%{Nup4lY4>h)3ztL8pb9ACfqR_D=>Oxz6P*rF8PO|>;R z80+5HJaYo)Fka_yZ()TZKHYm7KwHjR85aRc}Q%x58v& za|&OBxn-c$HF%dZct>Mkbp%H1?Tu*w;ntV5-VT49+l4gz3D_Ne(uHi@t`8(mY!|NK z^C^0hgsX2^O?|!-EDZo{q@y5d%@WS+Xo%=M`DQ1~O*l?2uAItOV@~Y^N$1IjyC4Js zjBRSYWYCh0ws!7@VG{}`XZAqP5|_&Q#!fm`IBC0AIEcejxdZHDR$_!b_X(#FZkWnC z{)VLeeqji#MgW@-^22_iCGMCiw;Nz5Jr1zq_CFvTrN}f6c6&##SR2wU5aTy$&;PDt z@jJp_-tgW0ybpys&;6{UB=vcpb-Xk~E9c-;P0)YxQCeO7KF@!9DX0(JxL6s!DT5Cv zdg=$el7_wX-=K#t&HYUwPM!Q2YU|4bhm{N}7~T8jg5d=VW_u0pH@two7M>@$1Efgu z^GCw#p^l`IqW%kpji8r~f?n_)+ia7Hb-UEm#QomKLX0g_ff>;H{rB!Nts&5>W7R8} zbrVk`CHI9I;F1&J7FqQ!)4{3F{o1V|DeGklb^AULK1A?MCG!`yA$1~^jq&?D0)Kg* zy!uGEu7ZmnL;U=~1AOnX06*-)-KqBz)hPqqQ>1~6U2x)yOfAd{8&^Yc3vs1Nq#H7|)lt}DGknN94i-jP z(@A;`SH2snX~Sr;1=9@5hBkx;g6ThQ2riQ;#RqqWPcZ-K1E&sY&ua;jzSfG_Htqka zC3NWrx0))A!Mo$of`LT?$?G4%($A~@rslp0isMyi8J@mfxCGiM_v$Fi3U~)+UIh>G z^dq4cc{R%9Q$=O7mSZx{J`rWwm+?RL>`e8Zaf7uA)1LP!ODnCCcSwV}WUtlETZ=SjEqL$utnM3a}{;yP>LJ8dyIwZI2`45(Wje8+3GQ>FOhZW3S4;``XHU+rL0$!t?_M!2)U)_*%9j47<%T?V zNA|=4rdphJ9u9DN=v#ZBDG@Z_9}|`iG1_VOzHP-}BYAx>1@sjlQZz;Cjw~7lHeI zW*UU>YfggB!_^Fbes0>KCg7J-=tjm5r%X*~7TboEoor;VyOmZEuaR}9O?@lSFTQ5ex1)=^5;2IDcuyvgD)4uYWLRzJu8H&?p5Lob-#$O<@T-0x;P!LVBzEIxmT zR2}X-(%V#k(x%=Yrd4GiX%{BesQ7&#@2Sb6>f%asx(ixeXra@^7IH~Z4e>v2^IK|& zB7y~NTdq`B3pc8NpTPX8!3`cd&PtEa=Gi~BG>dBZ zL@&Zjk~aq~Q#eyk+>KMy*&67|M6noWQL#3wcwIAkDruP{CaXvblSCJ8`y3?iWHFi| z(N)n@(lHsNLW)W?@R~`=^1&{*l0`TKJ~W-pRefXCo{n3M_%0B=+>h-d90;34cBY7} z@vL-y0N9o)dP81=RM24(6Gb0%-+4}kXvb?k(ZbWjAqpB04apE&lP4)+7QE*<#R8QQ zN?nCKb&9pgiA1p)eS_5z10%rGC#8zCtuFCZe9?deJUgX}vAVQ1>EaeWMP=isp7@V5 z#1RyOTFBpJumP)=59*5#ag7Xi_%9+uzWkP*&O$EfJa=-X7ud9uIVBCKASvmR}DPKgKTn+ z*q7>H_!NNnWs6xf&)5-w;9VQ+WP2`Xt~80ya^e10ypZd+khmI&J!y*Sx1*8R0B^{k z{zt+Z!*6mgBy}5uehPSPV{tNmM+ehxB;=P%u1v;IMFdPK`0hhjY!8 z)@r0B_8#C}JGgE;$blYWhAPgdr`Q#in<n&vZWRDMq4)&*9#? zmpBVBp{#Zi)>}7rW^eHV4rst?pH0uh8Zb{>)dz3_$d*3hKnkWlwn#6jVy%7!VpELM z8nD&vn|*bzmY2nK=4K7pSfWV-wtj8heMmEuQqh(BiCxW|I93OaMV9o_xg6*xzJv=5 zx)n0rtow9M1!UJNFx3Yns-Xd(Xc*mJTwMiEZopQ`f6^6jL#0l@SZ2)KTVE65q~P)f ze3iV`>)@j@iF;j4P!LG*>tZ?H;R)|E2HG$%pb=k| z9&n-THXx*jPWNT<9`^QJP-PqW`b`)eZ8MngxF`JKTxgW=0TDHo|3;lF88S~CZT^WX zbwyDs(1-v^1-KiQiA5OS$>a;prwsHg#J~WcCWbw}g3VB$u*14Y9Hyjjr7Tuee=ic( zK-Q4;5QE+!V7AaK9;&M@0sSkH*u`Q`6;2y7d1JBYONyi6$9_15J7kF%0lU9hEEpFp zg#bU(#gD_!Sqi-Gk>q9KQm>9#tiIcmGs{E=%6FTVi(ewthile{{Jlb4rlfG6EOun> zo0ZUsjm%=9_8M$!0N8MZHum)`o36v;|thQIfp{LNZ{{ za$nd2U2uLDyD{ATU<;TJznINWXVr{X!v(WTAo*pRco+B2<~!uSZ5L~y63P-j;cTp- z_9yu}#1AP}-gn?M89QMd0tt4m^TbZ@7FE`bo#1JBem0H#)|Xz@7;stc4ZDFoUTYM# zg$X0Rd!eS*lB_*qJAl)H9y;s+3EK<9ae&a_2=!q&AV;-S2_yE2+Ho$xqLFdv{EFh` zeK4fK*C~5;HWd7t@o0WPv{P@RPij z9DY1x{JYT7lv1vES9}rS&>Sj;_^yYGk?Z8}gZ7s+jY7$!!{U2@#BauYqcta)M_Ky4xQWKSsDSmDdLM&S8yh*$1F%XJBa9#RCl5DRn0Q z`am2_fpnF|nxwN+9L9Z2gl~!_05*K1Irt;3U}W}3Vix)3kyw?Sr*8v#!wez*V}SP} zb3PI?HBbO~{cXuwA&;HMpJy=25>OxpzWVv}6VbcMyd2&7vqQ3T|h0J{)O3_QqdXT(MHNsqo@Gn`{R4tCe60CM^o zhW!VX0Hngq<1EbFwJfe&`4^y^mdBKvdJdF>EUL+OI!`6&6@4ilf=|wJS;c3OwqJ`o zloWl?!lj>wX+B3#v+xCRp`Jz0*M^Y`7sL>nN5%YT5aWi}+7RIqE}?`h{uU%aisD^2 z;8FVrk|p1X-xom%lce? z&-d+%JPJjO5vsX;FF{Pldva+nPi9}%9eB%S@mv36loEorw#1M|7GHr*^RryzW!e70 z_0(o|jClPB4dz%bufWT>ba2oBPGCRzQ5*x=w6J9JPkL-1pZ+8s!VhzKhn}l3QX=lv zpCRIU=kb$J0p#wlFn)LR*RTKr+vxG6^s3I$y$TX=n?fGnL%w-UFO&Cm@v2uw9AreYE9jn;77EpCW`7&pmdZqS5G{+D>0rf4N~+N%jx)H`-d{0R5VV@>VtU&WT_ z70NXeZfvvtCWg^pT34YZw*fkuK_SZ{)&M=VO3OQH2OQ{r^^RB_;WEy6#a%HJEoEgV z;eP>Ana*87Nxugu+dNRp?!l^x0zPvOhAc-sz@T4E^Hy#Oz@kf5Lw*CSvmRKde-kq( z6B-N4NYd}nZIQ^m8DJPD9O3!74jiR`vvpPqUp1H-}hQ3z$M9jJ@HuX`G4aUW(Z zy&BTKhTH;xqL`2oA5W@05L3BcMO-iVX&0L3t*OP+K1Qv07FTM7L8+laqT0fM8&e9N zic;0AI#lfie|#WL!*k$_0X^Zo<)P-@RX2^9pvU?aKjS}&&SBGp+`*uKKnpd7pSzReNQ0v zUuno%Q~$q37aX^xs`)I^F|T;C0%Y1S4G5f-8-8jF4#o~;p)mF%7pf$M&6Nr`#^$}Q5vxi6nMtan zGus5XqpL{G5Pkqgr5-!IDqOUINnR}pejU$WvQWSuRiz?$(jY~XGkm3&=#%-b;*+&J zDGK};ib^rD0HV>xzcBgR50BJeRqp>MCP%|~j2$Y(@QK8a6 zDVx0cHH`4ytuCcP#@~U`0NkxH^KllI`!X59A$q~PHRhcz`xN45(mGJGKaW#V4Jilx zGZ+M=Rav<4eK?N+{~*YLP})`uf)ww&UqhN>8V^Frs%ny*UEv<4Ys0Gnckv|`gQaGs z@PS3%4VMxt@I4X4IDmjWItH-99Yyp&e%UZkD&Uq1l(-YB@YMdF1f0X zG!pwZ;oY}ATB?PEo3NM|qg&3G%!!t!Xjn-xfK~4~STke54!%v8M;XYuN5My%NAraM zayV9MS-B2e=31v-IGGV z49IGPw8n$|`@{7E15l1-muL70tbVfN09ZFJere>u;t|6JjSe4HJfi63Vd3OtJZn{d z$4fbty_>VeWzr%^%7)1k?0VM)Pb~T`DPu_0M9EV9KR{#?x=~G7tM8L2Mc@fd*s4?h zc}?F@#1x0=P4K=#KbHPyQqljOU6cwVfGBm5kh9deKnx!#c zRg!KpgxH}oP-+u?DlnBb@iVDXHcazb-6kHCyvd<7sh(N2*JLA_)gZ$Izh_ ztx{xzo-cTEcfx*!r#KoC)^cBP>c&%jVU|cBxbTg^|aJH&sM23`(&o|)_N6}iC`xyBXo89H1S z#bq_BO*=i5lb zbt0zdlzdQB;3w7Q>d%+E807J`m6 z{u%VLHDy0e*uR662jmf*SVyv^lN7{SB>4A|reyVX*qu)3D201q)b=hs#|?It8?5?z zy|^|CF&;LK{I7oAAjwNExLVC!)LHsLBCUH$wcLYxO0ft(H$NPa9)6}O_?M>te6a0D za6tVKj(W;v?RCu`FX=k|C#PKgJLi;t?oRI+s%Dp#!!5S(pWSZBj>^EFp56u?DR3b(Tmnn4OSh-P zjlc9tRd`o%DQ}N=xT=%=^ZW^p|Hkp1m8eoBknez-YC$yI^lM9jn>OL5d0wP+2mJg8 zyX@29WH-69E9g_;7H22ipbuBY(u>&B;f8#EBdDqyE7S9M+LGX+-E`ry6MohO*01NS zmO{0Tj3)#2KCCdM8S7u{&3KsK?Q2u}USM%_G~8`4Y#sClc^oZALEGnn=I4?E-E-cO z;Cr^#q^zyPu25t<51pH_(7#_tZ{3WHS}(Q0t(&pnYk+{fUEN@FA9fpZNrw$mcU;_z z?g3a`29;QRuCh@|Fi+$fl{RB%0e6|@~6t)KsRaAg;+f>2keD$Po)>2mhA=rFR%OjUb+I&`~|jL(EJ6`#4WAV zldS6x+b^}FS+t9G>YNn@@9mdfMyQaNmTkxlV~Wt*Gkogn=!Gwkcixfad0uOs`ZnIV^qab&OU+XyN-* zg8l!sb#)<8MM1cG#VPSxlu?vr5Ja>Y+4@+}4;0>OoG=i=f=~+OBAQELJ%ou5mi&g4*c8sMOM>av=WAo|Ao^y`{@j*cX1RCf&dfJw_HfQ&&N)SU zbwJ3ZZ~P&Bv+VIo9ugPT4+i;?zAGNnc?*YqN_$HLX2o0mwcpBjK4s*Sm3sRSh3upW zHXbhy(e>=31pCM^UCw#6V+5nSlzOhbw3h0H18_C+k7_xx5URPA0-RJJVppCLZcrq0fEel`x<>CnI!2cwiIF+#?8A@lrO}o(( zrv}Szb0wD7YBRc14O0D_?%;-TST-ldI-NNZv6>h3#47G}&0i-yxk4egUO!HC=pOLx zy1Ax+I~;w~SE{GRY1!gb)%M-2zFt?)OJLaF>q^K1oOJHoOHjhT?@G^iBgYRdsWZlI zl=#n#!{uZ7`OeFLph=&SSM=MmCtTC}{S5NCyVmU!G;B}93|9E2U0wg0`m&Fb6Py6d zlCd``sHJOMS&?o+aHR2}gB$;QmXnV9nMrD~?B$+xc{?*7rh;BOMThK=C)0$r0qcvrJ&09KQ zw?H%Y>=%U}1lXqXJM}^ucODtZlZ1`0|auDNfagD{skuc7J$?+*|-p-j7jSfkY*+`xD3tz z$OQZK=lI8&Y+Vbog~@KzeE2St4H&WeAtqgL!aoVpcDA6o4x99J)eIGMe8C&#vnxv6 zP~w6T^GZxBF{{L+5_3vSDKVqOgc9|tt+5oR+xe%Uv(XV~jb#ScIff~?ZZk~5)yL)S z2H1YaDZs*vQ-B?1oC1vR_c6da7-tva>+`k*1F?r61>zn?%n*HK*XGak!yfeXrdewG E3-dM>CjbBd diff --git a/doc/LectureNotes/_build/.doctrees/chapter3.doctree b/doc/LectureNotes/_build/.doctrees/chapter3.doctree index 677298c6a162bd72808c65472eb04db2f4cb0f83..29981a4ccaa202b2452cc135a9e1e4b74296c1db 100644 GIT binary patch delta 11007 zcmb_?d$i+bb#J63WF~YVnVHN?o_8RSk|;!yCD{@sA$s^Fzb#poU05L7l59(sC0Uki z7wQ$DrK<@^m>Ez*(!5Ak0t5;d!l5aTN<>|w)q+6I))+flhT1Wl1UNF#=BMM{);gD;0||Iu)YVZF9XgLFlxO2)RZ%N{{Npx>T|%Mm{ap zN2AhgFsosTIB1p#p{U7&w8gUt+wV)hk|P=^tXPxfg5I;oxyoEjBuu{B7H4iLZw>g? zY@W|_yZtF&(EGea&Wo90yR;0=Stspl&AQLhZI`qxajr2aQ8xO`LD@|qNv4M&Q)0?E z?WwNQREltNxRKJw2(uu`je&%O65o`2tTn8q`eWAWb)%Y9kSf_)LBfTMP!sdX zUYML@BQBatO%*R|K{H|Wgk{>+8b~sgN|C6a^l@@F8!9as*az=aRC$aQ=AC@Q_9Zv( z%!Y}qOO`CJS;?Aw9ba&K&CQc-)-5(ljWSr=9r5YDG>alm(!Ew*;>4U>N{a=1=pltl zt2x3GjxVGrYL-CCDIX`erELvmve5_3{zys-gOrGpu3Bm5ql}nz&3-;@^|Es-ZF#d| z*O^LnD&Liiq9$_SNe;y^Lf~V&MMzDmP;yZ#lH`udOfphWoDb!(gz!S-X?~K`c_rd< zCO73wrHKiG-^R_tbjfve+!qL8QO`^!Qrm0QB|U4UqO>j$Q7PgSrbd`q%ucYjj%neP z(@BnCk8Xu!ZA50E2=Uey@cAH_DdFR>lM$#+oDvDvMS4ltS{+QmG;tBCdzmE zL8|3MwM0-+%29cl?V~LOpG3-+qF5CkuuGPV#;E5c1Bjn5<8U$u`^{wl1~IFn_$abmZr0Fzku0VztJj~nww-r zGqun(U$!g6(yW9zjcQ>znbd2}7)i|r%~PGqAB~fe%^T%P%fsxUS^Ix-z}-=S=#i>$mq<@VBK+U2cb-{P&H2I8KVDkI?3nPN_0fpZ4^e7-GM zv{X0bs>oVaB}i5BR&U;;f+DWtB~PXbEsMr1T`r2usGq2bJYh7f7GK3@ij5nJnpsX! zR_?cy-jp>HSdYxQW)lrHmWgPu+b&pxw(nFY>7FU7K}BUEfncY&Pt}v2XO?KE+3r~_ zyxGfwoPtelvP!a;V`p-#snC^)%}9oZO-x+qAhRml4>fwwF_UE;nc;q$p37}(iPUSB zRVwmPSk#JROH<~J1+OBa_7ohdHH>DVUvLWzE2L(26BigWR}#8JkgPLeV zI30S@>G(y&j7J1;CM?~WW^0)S!}*JOxuO`HMkZy|;&2RET0LN8&j8t};dmf@nNl4C ziA{1_q{fOl6{N4zd|=qAG_aa2clma1?97ldh>VnJC*Aar{Xdnl?MKUwXeVzmbFD3-g^naqV@hHsNwgdwOv??Fgj6nQ_K=lzwjQ;cK3&&y zj1tyoWHuP~=1y0Tmnmb;qNH=VDL7CNXk z_tC{nWCk^AX;XvB zg@P`qra|kX=$l-ZBO--PX9g-&T_iM>UUZ8Kr5aHawnf^NIz#e?pbR>_t~Hbr^+v_? z0(@Gz?ts;Vfdg`m$jGqsG`W0{3)_h~_O zDq6)^(&aE`@ttRq=Pe-&)wN;UDJ!1S*Neb_5|yUHCc(;*)u!qwxhj-pDl%H z7L+eNE6sz>DVcDyfv8Ab(~+u`auuDQ^z^1R>GF+}b>44n9+@Jd!=dPa&@7bfR5P?~ zQ1!~Il3b?BU9999iE_tDpjrU>4MEI~+V#FAE>QtB=boDsRhFVlEvrJZ9nZAABRfIh3C2}frOOAviP0|KNrc5pV5K^S-IU~l&sm-< zW}9{;U$QBwPO`Eo5-d)0p-m}7Vx|nu>=H2LM=OGGSEM#!YEDY&vw9Z(vxqJs653`jI63Hpvi zRGiogW@b;gagw0)Dw`Se2~+~Q&zDBMnRcXh9-sGmLQ<@NF!8%WHOhBebv9dNnqI0_ zoNA>>E-`ciB%yoWpwntwj?~Z3<};MbwsMF>&j#t)2v>SdUFwsW4rpMiQw|~BUK8bv z{@8LHQ=>W18F5pomrobDF|Qa6lU^1;85U3=BxT~&0?Gv>4N{_Ld7VOqY}wlhsi5bz zL6PPYMJX|@=T{A78+2!lNw@T(tcR05Q6z?FITR}WAoo)JVVW*WOP7kG%+QoTw+fQ< zPPh7!!e5k@rCGgEGUx4M0SMGAp2^WDCGk&RmjdrCkIid4SdVvDYsvjMNJdgXZfb2Zja8 zG(hpn5q-JIkfKqjWveDKky7FeGZU(q$3v+>*X2%L(K#h5rue>REt8nsp!qE0&&8=Q z#l9lpNKYV(yqjg%Xwp3ALB^N)xXeRi>;Ffhg`;g0Dv?o11b8syG30 z%o0H%LDUq`4MxHu5<@2;39H^=IZ2qHc~Bg)J?qnW)A1CPWtx?Q1Y(`_XT#GW_89MiF8kzVj8yBu?Qnr9|D zL$Touv(1J@7eH5p(e5-#lg*6UG4tghr*mE_6Q&YowObYKk>nOetzNq36~-u6q!Ln? zskCxThcFRZ(}r|@p+>sez=*1UnlDbGc0&sXiG+GTq+L{$TA3c6gODu`IZ z8c0ED99DQqrm{IU(@l@07R{kt*y4k^+7{%93k!-8MpKe=&75PZdUf0YK9e7{%PVVm zgwxE)il)0wTENLgyD&#e=}AGd=`q#cHJmPY_5KlB03)~N&t<);EY|0!maGX2Dw;O1 zBsFcVhS9AyO&8n(M!Hd?4l$4cM}B{rjp$+26L({5KNE8G2IhmF=JZ)&s&&o6vaT$! zAm`HET9a*dS+Ru`$PUYg#j+ZR9!bsDbsfwMR=GQ+%rLzvXlO|zwpc|QfmTSiB-V0t zJ{Q$0EJ`*qYT$G`#7I_+R6$iV%|Xk=1TkcJmQMHQxsWeLc^V9CY$z@TyQb8bD&hp? z(FpY{uWbii%8;{QkOj64y|K<>{dpn9%A@o+KdKHS5R|S!az1U~c6N}Bc-c$_IkJT0 zbw{zhImZlTC8X!`wo&O$W<^s#jC!R!C+AY7)KAMvrsTC!XfqQ|@MWzVPzLS{s-BTZ z64NHyPZ*s0(W&&n z8TPbRHw6xir_j!mzJ_)A)TWBqwJQ>`s(68o^hXtB8e*MA(MUEm z2N8-&e;nckdYZRFCO|SX2HcQTb0M}EQVUQi*s3sF;<(sA$4tlzVmPQ9ogqRv_#DeC z4cwZeW-j7$X1%)@6$C`3g<@8j^_X-eSokzcI8&f;)CCO{>vXE#EGefrJ<1t=D*)!7 zfZ@xSO)=G+P4Z~8AQ``nWKnieLKej&nG2#yw!fT&be1Ud%?=pjI;Bu$i*$}Fn;emw zW?Lw@D59t)ig-BTp@|6!F1LDUrn(rb$WXzCA~R0Sketh?iL4Rnq1hurlZLRWAdaVr zR$cFPScV37bp;G{5!Q)xom3_=Zi5RB8i~+EXKCg^#qj2nv>jj}$*aLaE-X?LJ&k~p zsp>Vg6b)(>GEFRsCfdYMYJqm`R<;aMY@hZd0xgAn0%IDTLNLi!R|^e0rp@${Ax14y zEX|{+*iCXql^40O#AFO-v6bVTG+cEal#HkK{2Jc?P=!jOhyUkd9-WDf`B3(rdPDCzoPENe1H-Go(uyzg!;> zDne&a(}+^#x;7OS#guKNvb8K-rbmWYwMomwCaqDc4?=2vwDg#?o`g8YQH<<^M273 zyIf}8q1y9MroE;gHFU3wq_m{2CJMAeMJ1gbCC5xQm&cn;54CKdL6%VrRrH2!k8K-s zdNJxn{b?Ol7$rgGN~Nj;I>Scf^u2KD*ve+PIBNWiz+!Y!h)u@ znVcaE!2pr#n9V7;>+uUrC#ts=bw5#MCz_WGl7z7sunlG0m?5E31sTux#!V0Dhh~~) z(WXCY4iRbl3QifO3N4(k1n9D`z}q9GYxua6BiJ#n;%3vryL~5{?Dc$eH8F*mHdzH1 zYeH4Q$JErxQHrAL zbA<7W!MxQ|qfWQic8B1C#vAobJ-`8XcXe->GXpYPYzGO0H&nmcWrqT|dajxt-OYCx zgc}Ic)(CCMLNZ|{mb6{UQ7Nn2R;xDF)g~Rzp{=A8vKcZ3oVcG@M3&-IKlUbRH}!DQ^rCdCnoLblw2RQ!=|1dyq2tS-K5w$#oix zRJ*U`-|+8@Q7CP=GdG=dij-Vk|J{O>8cL;Jp!ilkfBnX<{{;Fu6kqyHs1IMe|M*Su z6~`~Q_MG^(o1vo{fAdY~TTuL;Z-Y|tGk*`=w)*Y5_%A*Hy*$4EAD}yTJ%&$1>G;79 zLRZ93KMkSqTldG_DRZtJ_ujt$th4%;4sYFm{Ci)xHg+$E&IK#o{t&>N|83}l z@Ll`k_niyp;@eh~w>}2c+xH#lCis#4$Jy&Iiy!$8bT$0w{^R(&iG5$U&pE4qO&RQtNUt=is=k30)~8G@hPf83*94Zh+(dk*TY z9@*s}{EDBJw??d%nOh5fT^!w1Bn0*8y_niuS@AHtjDeki4=c)VR zdtL>-4!ZPANc#s+aZ4ohZ2aV%zzQ$?0D9FHbBz5EdJj0kGc2**8qj#PwWoHrc78nk zN9a3HCH~|+(B;S9=3Wn}Ywp8O@9d;<02~R_KYQBgABx$3hVn7}&rk#KJn;a8!%y4W z$FpzaIq(*Oy#v6vvhiJ?f{w-~egv_5-nEi&>F}nN=xdMPc}U_*Fh#c>Icg+0DE0)wAIzAoz2e%$=|7V&3@Gxo{PNzj82s z^4E|CeREr-eg1#nb_wwI$LGNrfLbfRw*No96mo4VfLv|if}aC5E`A)kVZ+@EBNsw1 zi|_jebOUsD{OmWNH|~4qB`-VabZ1=@<5xk4;x|17aq+$HgvpIhTnfKyAN0CShSzQ~ z9Amc`UcHN9c{%qMgmJdL&L=t zYNZ@_vrD;%f%lvZe{WN{?_Cr>_(u5goyYYOcV7?RaT>|J0hnW(X^o=a1mCwqT3P5| zeBeg-_8r<9_N^O%xliSMgazh4m2@iyPgkUszK`zG_x!lQ!=HiaL-9TV?3+CVYTN7L zcM9+u_dUL|VetBI1vm*{YXRU*eBt9zX5;N?_}KY--~X1|*Z((h*2XKU@I&XrokQ^x zH^JA#Z#@)$?;FtX?fZ|nZCmcmhfep$|ML6r(+A;qZ?W98#qys%v(55{J1nL6>04ka z{_#`Lb?^mz4BZSJUVS4sLzf)BfcM9f$?HSS9`v+eFm_IWg7DEpC!3H9U)jG}{Q9@I zr9OOUBq-gg@5*!w`gvB`82WgLSp zU6EGB`tD<(SpVb7ip0LBc9sL;|K~C2*3%fjbR6POVT2E#f$?1@Aon!JYv2DOy?*h2 z=;qTe3*j4S_J3$#S?}XB)$^OBeo0b9Yy4#EoA36^H9v|O*7qr}a+u?O$e(&K` zBX{i1bM`&IW9~maymGT^c9Y(7`1o%>b>+tA{~Wpuf*y!J`75B|FYktG@W(c{`@^>^!nY# zc=%8ZnxaD+fBYcyMF_s~$g245d+wa`PL_?eos%0+J_MZy!_1L&b^67Jp5#XTY z?;zVq=17cw6jD!u9{ebXj>Z~vWH)H%$Zuie|MoB_Mqhgbs;;Gb=u3O&N7jcP9XWaEKi^>uwpe$O)~r`5O|$LNP^svg{7{rfFdcaff2S?J36H;+OV{KBqvf4#Zx zp&jNIj>JFs42WcaS{Z)Vo1bin-9_3G`;iUk&G4lcpS1Xq9cf1{-aM>bB*1fH>#7Ss z2D0rg1-A_I*Pn%=y)^izIFOo?+X64y5x9l>z+Xatus5@b6Si@ey$Cma47y{lyosxC z%}|H5MR7cUxIin`(In@+sr#OwO!-olo~zJ64nX(QLxCr8J| z(V(v%)m+yebk|jKMe3ebIgc7++a5d5u{uiz}cIUaj^X2W>I%%WNdhmG{KtGHhzX`qp{{F@5UL+pe0>2!7c70^)t{8(s zCvf{a;kWFC_gu2sww!y(X4_IfwC7|#gfH1{TUHG5g?|cN3SAK2{nYyFZ=Zr*eKzPV zHqiIN0Sq-Z9=Hwu#@Uqz!NCmgUwvGC3HZ1Qd`xfI>u4iAn7jI9bwa}vfLI>B3SNC) ywP*EtIk=xF4_Cb9HD7t!zhZ^FF8=G|pw-`$2QX`SL!b+M0{%JFjemIp9{n#@5K0mN delta 11030 zcmb7q3$*K2d1j;pk{h^$`#AUJ0qG5pl%Nn1yN$&zf#mMvKi z+mJCCI#XDL0676Qgl85dKub#@&FQ2-19S$~nzA4*LuZ&07J+7F3ap{L>A=iv+vlEh zZhE`g@mb49d++c6_pkrG_c`zV$d%vu=#?LT9J`ffQVgCvw*?guI?bLthmp6k z?D}E*S+k?l)cN6t9Yf%Iztw^s+(|ut?Io8CR9>pbsZzn7aNtu_n*ChU22U*pd}YZD z*Iv)(`eApOs4n!{Rb$W`gD{xwA!hRm+hQG&8MX~k1^v(UGmJ4(yKJ{!=5v& z#JNDT9jc})PM)cWsTDTXQX^)(==4Q%QleV{>K0LP&@y`_Lb4RLw8jj&EUI{qMtjRB zKqIrq4qeuwhzd$sWyV$8O~+G^NUAz{MJ#D9l8cmjYJq07X^rfnqR=WbqUmI(SbAp9 zI7yCXGdx;Svy>aI;<>1eT45ALp`0FjK{M*~Qd$m0^y-)z_uK~2=`@FmH!V9%w-u)8 zjC{bb?;=i6>&F@ z7={YfrB=D9m2-W@>1jir$)Stsph0A1jH!YbQw*wNxS`T#qoAM>v{-iVg^_dEh2D`O z%&JkPRDlZ%7N)Xc+L!$9bZP4bw~Qr=C)zxpo_ZM~WBawL!58LJc^UGuDor%1nCj1( z5`x96qMR4iVJz#mfCpBcjk{`O@}s%M5k{spthju)K~E>1_)%5X+hX{%Tn)J1aMrve92Qe>9k<_v}UY0n!d*qAdz z&tDN-Ym(`VJEJ_Ugh8i5n<>qU$0Ci^4UX_aZyH2&b7mHdo*SluI>T3OA_(XLrgTJY zPywbQdt}?Ng?1K+BL$0ioa}Ucgjre>N+5uz3;m+w*D}LwW9A_BjU4hE^9Y+TMyo+ij`Rw>tr+=F)-Y3~TiwsiJBtB}-CB2AU1;e9M72-_{!>H&a#j*ttm%#&}q3@Y@!Qi z)Sg81d?kegCtINWj3{G9sKrCzSQlUoiRY_a+mv#dtHm)ox-fm+PU!zLGP!lIPiZzrtabN_buu{B!~>YL&;Gma^)V+I8DS zW8egy@A-ZYt&kPM_B1&wnFQV~M`)(ZPrEWw^_#d7Y~q_GMjYm#QqSMg zl+mZq&?~73(&WP>_4I647`TkHTxe825NXF*;H)n-1n@YqbhSAu$@etmMr%Ig=S+|1+9Tlrr1Ee{!TOhe{eB8yOUfdj(>HDnPZ zOKnFYvx03zUDh}Fd|VL#8DCm-%s{D6%bWVpWHbJpa-*(m*CMy8p!Q-iYKp1xs+*ai z@u*vq3RYUdGD5aR71JHLUv5r}iuLr&cC?&cr25sP)FPi6`Oya{m@h?j zD$?@kB0N`5tI->#O5a&05@jStQ`f=#4$_#GXs17~FhL+T#!=wozR$A}J5KkS;i5hT zd8@(DOvx1Z30n8<9M;0>A=hc=dOi{7m=rrfQ+ju;e!EB!{V`q}8k`%`D>5gG@hIl3 zEP*d{+7=jr<_dVAY2291Ta6CL9<#Ap*6~!1BGl4&WMiN>iC~S3(w^U8@^U^#$Ac*SZkLkO;{VJphQLKjFonxnH(S8H}U z6peypOfs@z+e?8>V?Nz8_>ON8*|$dHr$z`+`P2t}lx|^m zNTACRP?FLs4u-w*5~Wd&2eqsd@&+kjA_&WnYfyDnV{)#Y#{&r?xC)YD9VA7A1XN-M z8B%4k7TeKU>@z(3nb5TA<9r>DHgH8Tr31tP}*75!bU0;7j@A0W#pwBato_7)@t~| zTpC1mLH4PN1M2ml69rAseR6^U+MI?e4xN-r)a;efH3GrTr8wH6TU-j2jr4*Ar5pTI zec($yEz*=}M(r3Pg*twVW7Cm`D>an!7ed@N$zIH%ptKitt`~OoXr?29)kS+~*agj} zua?J?V#8JCIWAGC{!MXxhO=OtxY0b$so(JaZMn%c4yxtJy;61YbN&CVASFTli}U7+neQz+XdJP<)J zx1z3Z)wAWL>b9!oRJk`?`k+ns!ilb1F3NFC1~^d=m(w`jpzW03WwQEUh43Mn!v^(^ zjJO%GIVuNfdk%81Zd(!;l02qbYA!?qv7N_s!sd)Z;0H!wMx!H}c9?iP?%?_u3uXl9 zhAZvZK%J49LF3R6K!s{qWaLe}39m2XyoLqo=@=xWG2Qd90XtHi;Yt}hMiJ~c>+mw` zddPI?TjOb)sZg}lnRJ&-M(6#mrB!`r0lKe}Tw(1|ELgm)T0|+8CurNm0<#gs3oJ`h z8lf#wmorkJ^P);lUduGW+eVyB)|pbuVQB>D4qdg$mYWi%rYODa9v(6)5`quBMpSpBw9@D@1=FI+vX~ur zsdBm>45bDEIx(`<*>qwK=){EFe4$gNRghZDO>jI)O+aUb<}KSYOpLA-CAHer$Q-Zp zv|YrjBGR)(S{zAZu`pKh_7e4?DNE)MxyzCTx)7BSk}vmJFY?J8y-qNggG^M%Rr;AF z##-fuz}Drv)dw-!1hCJXNaDUaC=} zgmRr({fq3rPfVdBzcr)5p^`SYbk+=MSrx)3nq{VHr6wMM7t^H?S#6I&yZH3W2|<=4 zn@4BL&@<&0xTpyDRTFfsB4PKK%+SswLS3ahbb&;Md0NVsIfj(9V3>=Ss8}+1ln+3s z$#oHTTu0SLDoz`wGR?UY#=r)JA%O;QQ2-%aw=+PULCDd-b~BZ&cT%jWjc|>bPFPK2 zoK!qSvqq*ru~w}t=&&VXF$8@b?`kN%6srspTY|(#lr!l!!yzU3%c0uv$MtzD(q%TD z4hbXhi=e77!zNAoc}bOfa#>dA4R$(n(|$LD%zNGP%$Z?LY!=6(MyHOI7pUJF(ORT0 z#!bseb)uQUts3pBudy>7WUstd?2;CXwLy#<6nI}ZuHjiEKc{lwa4F<~?Yy+z6Ro~L z3h`oWy5PbnBrTHS%2^u^n$<2FmFBdmx&$+wl0GL&?nVdqMpL56ic8XL&WRMFENs@fTqD1V-I z(hG6sRzYzoG2@I9q!A^I*gTFPh?q4T(`q);=)B^1Rgvxop0t!jy6Kp@-*W@G?Su_| zCB_QK4Z$dFk0oSk;Sy-8am^fIA~C8u10;nmB0{9x5~bo}rWMPAW}%FbZ<(M7HJor! zWkv)Z$p|}d&!(lRw6M{*N@i)hYQ%_?r7UoF(8=<8x_}V6Mb4RW9=uyh)(b@0k5>}H zq06S?fs0Mvws@BWC)_3`SJ5DHL5E5=Mk&zomS`;G2Zn(Tg}8_ltQcn9DK_*&G84oC zmennW$E%vQzIX^Et!!J2#Aj_QlxDd(y3F=mw9EF*fGzua#Hw-@#Vn*PtId^c)hZSh zngn51k#Tp54p9oU*>g5pl*r67pPmm3K57hdpa&$gKFBX<*5b>tATdl%NQq&F&$THm zgJw)j

+Cm|2@^&C1JugebsWjAf=;m`x+bjf4!2besw??1L*3lBQY{VkMyF03CMY zXn;pNP4`AFQ$?yggU2}L*9@~T^3j~k`z4wP%{I-J%#7t1yKBaoR;(g}!8ngn4NtZe zBbwUXM%ay)1OYmy=|Y{BTNPm-;BmuPB7C=&KyV%4d*oV7BlI;OPnh7k@? zu~=Rr-Psfq#)uKl7@nH&;2No_x+!XbxW(qO-UnBW%{rN7GUz-YId=f=@Y1PrUPD2D z0HUy)zR4l*^Z z%5&c3B!+9);3A4OhQ9AM$x4WOp03eEGn`5Wa5=7(kAm6+EqC=iI1o53q!6eqXk79Q zEh9FT)^u5HH?6*<>P;2lJ#h8NmL<=zki4$;fOI2on$DwKWyOfNXL!}->xyE785f)s~G?{>J&VDwCbzl`N&|ya#_{YncU0G(=MqNRfiStF!`1XZfl+)SMc zCV@%mSgPh4XbLG-HHv5WJUBmk^I?UE9BC%bT5?dQIfBKf?MvRA{rs<0`O@6Y>8FXHsL6pl5?sAjX2j%3r>i0a^6 z!I&2$78zIam1#rPgRu}#o2I;u2hbx+27}fZXW^rD!B$C>1yM=qc9%`z>9CgBtmDLi_y_|M!4nnM)a-;+JAR|0M0 z=5I?}Zj^60N?03bGJBohhJFma;s&+r4<_O4+$=DsvvULYoZ(MqvzJEQ(V*821{3$b z{^fVv*FSUjMUcs>Z-TFV@kJK?5sT}Fx18L)E4hC6xwl-FeEm((xxL44g3nx*Oy2_C zn(QF(yVp-_@^=XQ_U+Tdx4;ii~CfwWIi@b&Oh7tqww@SR(l%V{_Rkh{m0}AlTS!zI(;}zYdMIY`=K8w{a>#UkQH& zfD-6c@N1#f#Sou)75pN|*lv7;_BHz(ADIFh-`>74x$}1Tic5ihlJUd9yFPq7{AUpS z!BfEJ3dy;9p`G0rxpnVHJbXC}Kf2Y-)j9ZQ(C0UMC$G5=dKQ>ZYI(SIe*Fu1_yOo^ zhwIOIbZri(e`%a=0jHNF z-20)Or2RtZmIRYw>DV1xVSZy@m>(bFTl>NsnQES1`($#i20O5SbhFp4HzSV!^pF}X zUtWiA0kBO_99fy{^F1=P&-X9t@NdHJ-k0R(hrI6Jr#LbNC~j}x^vsX-Zx(<2X!6*7 zAVB}S0?&{A`|a!OZJEITu)ns~fM0j)3){8-u)p@p{k1nY;ddST&)c=X+F$!W`)hZq z@V`Fx?d{sP_SY`?)AfF*6H$Xdee7r3wV!O)j$itx2kQRezR*Xe{?}$|EA^!tF3;cl zQVYHfg0I>+V!5k#*2=u*8Hc>C+BxE^XmUdvP9=|Z;8#LdCAIHE*rfoPyyh2BesA#` zaPbtJ-q~d5>9?;{|uzclUYlVTnnycf&78c5Z@iO{#B#`tTjw-9c?QaTPGn$z#tr zpAI`ad*6K%^j!!Z?d*Eg3&5}Bmv4sNe`NT9Ti`Rvf4Lj_ox{1^&yuH-kGuuSt*6#W z@Y(l3DEzjKj*jmfG6&4=yay7G5K6DQK6%z#p$FiHE(Z2aLca#`*;;=`*>rYxPdsoc zX}=A6$Fa|@)qdUZ&vv$fdfnk>0QH*R+)0XWhwg#Dw0>V9S$+b#65gl2{a)xU_#0bs zzj-foGyD{g;C}Mi+JP#*v6G0q(7X0QyI1yahJSGp%tpv>Y=lgP4};tSHt=pcwfDq* z&=(;1#F>p6Pn=2q(>tLa7_Oa&fnRuL6O82HZeZp25~zMiunTmv`V5Y^+L-cKu^H0 z-)y{*?oC@3?BScd{!DW9d!P@mL2DLAR<=l&3c@9Qpx+Pp51dz}sdyYKxW z^x&FkZQCO+wr`v#KMqwP_&sOWwmtrDu6^cVUfIO#{ht8DAKPlF{YhvBKeFBNUhV+- zv9rl@9)@1BM*Tf?=GZ%4`Hs!MhF!9I!$+?@egwC_;oT2I&xGMGZ#R6ff4~MTuU+bY zKLUN=*tgDZT)+eCP!*F=@ z#<1bl$=Pp#Jby3*fpR+GpM>%W{Up>loKODfN$8_ThIf4%`qbfY6JF7GptnA4>Xm;B zz5QuZS3CvG54KzD0#*)$1a@0T@t2-j2l?l>!hQbgqrcw}W? zzz4q%U4AKi@|unIPF|CI`3KNrhr`V_U;KOM{-;g7>xV!!2Sgi`{v+rEPn&xAkAb!h zP-~5YPVFbqJCDo(wVzJ@`X?aG9^f|I{^X}nxSm=Iy>s|F`_Zua8T1G|xF(^04iw;D z1AO?!$?yLhdin8lhfuKmlb=Ir0NXG*vI6`KxarIHKJ*Ld-YfUG%ixEugmNb z&9;v@xjZoFTdrC6G5G^;{nb_Qe?6Ui&V|2xF!f9izWllHUth@TtNX0*FCMUZ{D9RS zaQr)t!N0hG1o~-GxeG49KRSRV-W#FolF!`;E0=;EXK!#f9KfIjd*UAWuP%MA23m(< zWBqk4_;1nS+2phL!l#jc3!baP8{qZ#O;@bHpMP@wy+&6z$ZFWX{?7YeJp8qH?!w54 a4TGBjWO#PX1?cGIyYP>pMuOc3kN+QXyWBef diff --git a/doc/LectureNotes/_build/.doctrees/chapter4.doctree b/doc/LectureNotes/_build/.doctrees/chapter4.doctree index 1bb3dd684ce2d7c07b6071642aa3cdf563b471a3..9e619ece3d720412c4f6706466521483a4ce9a2f 100644 GIT binary patch delta 467800 zcmb@u3Ap4|buJpo)9JSH&~6O4ZQ3K)R@MsVYfp@3q%ndkz0u>Tmw+xsSgieZ?EDcvo=q<5y&mdp?mXAe%4!+=(;0zj)b$ zSHzt&4!-%q|Ju3p*8EK`b?$R8xNvUPa!x#UV%Quv?QwJHJh}Vi)3M8Emmv@Da@XGm ze}DC=A36N~ww<3m{BE3>!uQ`g@i1cUYV3X2ww-(PFFY9a;obDfmmfa*BPR>jJ?tRP zWzKyC=iUO$*x5y{z6W{ZafGd_>xYDYboEahe*dv+w%>=(`^e$-hpv?_8vORVXw><+ z<3_*oy73|9d$0Sk!|$!@9pvE+(njyQ=@my8axP>W7=GyKt8cjFGSvBMB$co3s;57E z_+z8FCvM!Op8tb~S9bsD`PLq;@}e)1C154{rJK!|z|Y=|~5|oZ5Xc-rOe-KmPQN8+U&WKXc3&yMKni9DU^2Azl^3jw5() z>@Ku)EWtm=;xJjH4lfDUf5=X_ssr`E_l-}!?#X+cE9wYr_U^}to1cD9oI=pv-cozJ z|3LV+^e$TJ0o7wK-F-Ix(%m0TJdD16%R}y9-M!u0$Xmm)!UgN(?WZmtY4>}{yU=$X zOZ#W!tBxZmQ+J_1aV){lQk8wi59Z{Pn+vX-qR+yWNGLA~g>uO2QK38?&=UmxVkCo8 zU%Y7bySKCF(62-+To+nL{b0RcL^sd=?arCa+aEn~^>K7@?Hu|KTh_O9yZ60pxj$d{ z5sK@NUbcI|m7jPJgDVm1kv<=iBfHnV{CAMscE9rS!Ywbp4EFZk{2hfoH{hMOEaeq% zMNs0hNQ;Txe}3qr`#-{uaL%thY@Q;*0nXFw-P$80n!oH&t%=$G*u?IyK9aa#hy;fn zH^h*88R|wuL=xK$aoOhCSD$#n?!R(Z!9d$>Ieg$K|GPQ~Nf*niXRzwj>n*ouQb zU<=v(VCt>NQy`ApcMHOG_r_pFUL5|+oO`_YhRX{*H_MsPbb7Ug8sr~n|K3FoImOQrif1+vnjiuN&oq&-w3%nPv{#Yzw;3K zrLENUU*0GFja^DwA4hR}6}@Tmsxo>H`sc@zJtAZ1Z$#2NhI;pdGKv1{v2-^msbdH> zFU+86is1F}A=q?;g#5-=x_h{5{&w~L# zUU%H{k}h5_S{nKMA>Z2(c7^)g$IbX%bprXr zapFDTNH^L>{n?EwmpT5$#fNb)eABHQD zcy12u@!`Wq#dGh|&DXphdHT6Wzlow}u80g#_^fpx0c7{qR`QzruK14cc#z2Y-u>b4C~oxb+r7)b8@X>+@!yQRa`y}VS>%;Zf6c#RC)t8$ zzHic+z3xeG|Eu>(xC9>a9`9A~<*{oz?qt|&E??bgPMc49uZC;hD|hAK`>#ECzP)$p zYyRoopZ;9@>ARN?WUe_n?Cz^SeIHEv1E0PNY3+XO(=R_&0)nG_s#k2{*YDi4d*icb zPZYv?XT1Bb*nF*j^5)&jv-lZo@6v&U_sn6WArI|-;MtFyc;DXr_e7FD1Cmz8Hz0df z^E`ujZo^5)jc4QV_(4G6Tn~%-?%KKP>2LlrgCHLb_v5z1{qRJ1 z)JwykBYW_Z(H@*dl+7FaCr_{Uwn2FlbI8Px!}I$aQ9Lr${_N$$OI{7W0zcoz4(!o* z1Nelk8^Xt9O}sKO$CF*`bIoJ+=INjQ-0AAekM0Z5z|J?l6@2l&EBy1?@Xz7ouZLgX z@2%nQ_eJUfdq=+r3D>>1fk#*PY_x?hKpu-kaIlL%zPF24I=8{1b~n%VPrme9?dCIo zb}Ql{r~U?JyuiD><}AXX{}^rX4G_X8PrUk2wtsWQ=F4Z1A3Kif9k(HOAv+u70p$K; zsJ?z1A{<90-40Y|E=2|I|Ai;MEA)S7yuZ03^adz!@b|YP@4n??z7N#QyLEH>=@Vy= z%VG&4-fN5;_k( z^L=+Ae}bNR^va7wyL>6~2x3Q!T_1|}XFhs3z@wXQycBs5ee}x6={>rs-GjV;|JUj5 zNLzcoId?C@AaA+ESRc9<`QCl%@B@tS<~8>r@4Dd90~nG&j?K4WNFDjeC5FE99D-jk z&RfqN@_H~^^M1rVZssy-1|QNB8H-Fd98n$49S>^yh}q0v>q^ zK{xS_?A*Bd%{YRg>6dP9zyIWYn>z{Q;So5U-SS6nd8hhKsC$BeVN?;+$a^i!7@=dp(n`GRq-efUszaOZa(M&5YLsC%3L zCzlY{?(O6)dvd>V^S>WKUUAF-8-)YI`A^5Q^~|R^z(WKXrEJKoJghbn_Mgc{zH|PN>zU!>Rya+`nVY38An2X{sh|J*?r^S?Ab%xdu&ADZJIiA+XVr^ zb9bNzQHi~s@!JUq?%6u!OUGp3HNoGU7ZJ3gh+q>ub$Mmum5^(X2@ArI?U~;wA%B$? zFOCiV1JiB>PhI}%jq$!qPxh(zA#>tCak5+DiB}U33@E$48~lP``~lsvHyk znBR8zv-#~8p)WtZANm~G%QwQYxPIrt(8u110@&k1p96I}8puBL$(zv}a_V#Yda(tl z_fWdu-ky}N%ZFO@xy|Yp^!_s!6Ya$v@4EJh&1Y@_gzveCI`SSzsdx75?ZxDqSKbO# z|9FWx|L-{oo;e zd~?s8=x zXae~_IP<-m>tDZf^5#oGPiMS$Z$5#er#G*?554`Oft=3fWA~xu&Bt*FrB0qh?~1m9 z_9znO6Q@2Z0WJg=5ZQ*U^Af6cbpo8tQ>(Fz0~G92IY{&d7O^l zbU5N*^EiW2=->#aNA7Ot15U4F(X$Bp__vVRn?8Iv?v@fx#@ka7dzjHXk2Tz_Cu7B`klvaIk^SW1{|A@lPZSuFJ+0F|5{=!4(tqA%X--6!{ zedBPpEhX^FJc7OtIrWDZ@C&m1GY(Af{}ajb#rwPdGcUgUKu-<`AdtFg^GiI+Y(B}O z0{Vx0od4nGYzqDAMWccgP78qQB3Ey7mZL)+v6KhnA8~hB{p3C9>4-Es(nq=3yhudf z0kdx!}O>;0ODJ|8)C|JCQ~4saG4el)oq{*znI?%zC>L0^j8ziF1xJK^ImGU(gT z`>)$j@GSp67OtAp|A;ViDa%PJQ-b(M7d||F#v~*FSYQz-O=C zygt8k&n*{I9as}@vw8f#pFD$pWh=j<^qVJiG=AKqQFZj|mmqqugx-yOHrn!IXrto9 ziE9p%oAc(|DWmrvL$leILvFbQhw6GnI9sg)HTwHH`Y?jL zVH0ayLS5144z*xQ2x|0>23kN)z4Mxj&VD4}x7VQP{hlqqr%qgPVC~+y>3%25qVK$B z^R+)ddFKrm5AX-yiGCdY_$6pR|0?wTA?@M9j*PtJ4;J>*kD{+ekYBs>!tMY7c^|>w z62ijzuR;F;x#jmS;_Fb;e{wTwp=Z$Nwj2@wDDMyT`4j zag*(${acC`E%0y$Dw`izqi2xjmf!I|91dLE+}}fu{lUVI5CZ+B9{N%EzFoo*f-NTy z#WOy7>*lrrn%wLR(CEreYf5v+v;&2aY=8k!MZ^DluKXmD7 zKI9xj2ks&>L|=RA;}^+j^Ls<|9P;y9$^6Va4(UI>d3c20i+=ow7>?XMPYjS9UmByodFr`~x%!z2dJpo~Tdschj}Pge+q`g!%II@@oB!PA%~MEX zE*t`4sGpyr6mrD|3AT~*d86PnNEV(uvbH04kMj7;Gc!~`kh7cSYaswRySZ%vdFa_| z&p%Bt->)pt2a))B^L=^C_1O&2+)L)wg6XVBYCm zcv9XNE@MCM`Gf5uvYW52(b5G&`fosA^{s~dubYSeClpzpy3~-|8zI)+TkgREBUgAa zmmB0uCmuqcxzs?@H=+4c$52J60;HqIuc-Fj;EB(l{#U;ny%#PW?$i+pY}q`CjzNoqr9t;f(h$5ofnJp-y}qJLY*f^zVE#`g`cf>n=SJ zS~`Xh4CSx91sx;E&6~&HhTe(ZeBEy0^i7-3em}}@{>%5Hj1bHx;20s4sG4#jxf1wXv7b)Jay$#Ji|H11n zI5qzRlK9O_-j1qBb&s~XdEeV1aIQvN+;f%pKjDGL^o{m{o#ZC=4)nIu-hSSH)0Vdf zH_dmTFFL6HJ-D}k2%0|fkOXiblv|$p;yb|Hop^J|$Qkc@uG@pAA9x(SdgqOBErO=G z$H8zQPwX-D#O8fZz-in2g+8&jnC>nl-B0XI6rJT;x7>uod>g3s$tThGT!sGpbs=i~ zZ`83rd0p5`a8Sq2@7J;K_~7B@JiVRr-wRuB%%{+gJa7Qc9@dy6IQz=UbZj>JPkO%x z<39$!I_7x%r1#}#9)3UiVl=)_^z9?HVUqz;+Pvr~^vs66e&@RVs`So&Z^6?W?SY*$ z$04%^B0Ue8o&6B{S_JvVo@RgJy3GqdgvL&NV_y;nnJbuO<3lKep13~hB{{vHOB0_z zlpRoh@k3|}h0nuq<;d-WsR8i(RyBwp{bAHRf!=%lx3QZqx&G*G9&{z}e|N~yo!1{K z(7$&#q1JGGW5~BGLeyUr`&KypTfY0J(PzH8T+8rK+&+rg(;YyF(-cI42+tcKz*?7NC=eVNa=9|~B1- z!?$rBue;&sdAxDoy0%_%DCcW7vFmqUh@9UibHLJ3wb;D%`ke=MVmCyt<$=w!*YDgP ztz_Ie9P5BAG$-HYO7I(Y)+qn&tmNTiSMsK}9?t&qZ+9inJbr2iMbSrZh!p(M&Ci|Q z`N00K@FP@-cRqjTt=k74xoPJ;&wn&r`B6+b^6r0q(+&e^5lG|SDwfjS))S-K6Sb;N4vLA6Sq&h~uHAQf@Xn%30%61V#8 zeluUZP{K$tBq>>9)-e zgjh12VVJy}OBR~b%#yO^ZZKOho@^_aB~cWH_4^FIo{z`;Xt^LpjY6MI;Z}m3Byx_P z5E@HL^xAW&zgqdppwXVy9Fh*QX_jHMO20f@$m)v0F?F0Y=B)`~d5#m{qoLfKj8jVn zD`sd>NcS6l zV?{BP6UerXspWDRlVH`UPBCG*$wJ1f<*K7}dBLi!bds(4T)bMKl^A9^V$-u5L`Tw% zSzUI=)gIYCTmvRbf_3&M_3^^sK{;xmro857Qx4b6_&(1q47(tdV!?1U zw9ADBPn2S-;ewK5bZf3k>r95pl5Mjnh7!QP0?%OeoRgp_(oV_QMZ1_z z8Evq%W^UYFcGUE!k?)t3N;R#?V5~Y~vKNsS#r(Q~U_D)pBl*mYS0B2LTj{5k3A|)LBNvtBWS(=HX~9_xX?C{I!Dnc>9){&4Cu#b_awVfmo>L=Q)7U5_%XT71 zu{F#q4`W!N5_fXFwY$*aa2ia(wHmdwLjXhxm<}=!^wtSqNO-AUNPH-sAn8zH77}`tj3$ux>JL#wR_cSwH8Ukbt>basA5b# z(WA_Is?9L$TEUZ())Yizl}GW)j5MucA4FH3DGLxm|cC{GtlVk8xF z)tSXRBsJ^}oI5AHDj(0}Q&XcS; z8BoRYx|>pJ)m}vb&uPOx%Zp}#vl%NWRpP~TVjQPaRC`v}8qvOaGQoT~p2A@H;Cvc=^YjARD zaL-+jAH`EPr>Sax)vg!ytX~GN%!SUzvgQP@aSU#i%gw?lmJLnyL}!U{b?mTU3ds3j zKqbxGtY?7luFb$@=SJWY;%T$)lT;gP+8qox1uULmOa-Ph2Vi;{!)~m#p5IDgG5rx#xYD_y(&1Nm0|}RmG74;ir*~S zWxPU#r$DKJkr)-KqZO=xO-l6|-I;ktisMp&G2u8^kL@G|p66A^YY8Hj0#}j+cHSaV z+yWfod_Koqc6L-)8#T= zb>xIj#uhk;a|U3Q;!VFKI>pp#wW0;KI4Dc2S&z+F;C>UrTv`uSdC8j9)ifUC@P0De zDGoFrS64WRS$zl2cwPq+v&_>)-WCj^;L65IAIC?l4AYr~Iz1U@S5kZ~3>0#;akBp5&T@J}PA93{xW5 zaWgpztWvh8_-xw7RWHDbOIQFOn(4e$o)dnZ$bwT`5dGQQuvR3~aanU^f&ZN+^c-I9 z&q)?@^RN&M%zD$Uf|{7U#9XdbZ9DIY;CcnpOib5a1vGTlo3z;eQmHGeHd7zWT74=n zRN~`SAstJWx=IdKyr7G#3CY>=DC>^#7!I2kAMp?{HH~~GlbU7YX*Vzyg>pR|E+}YQ z5Ux#v#$;g>AxvxKX-&M}1!J|x){I|~}Bs^;^! zmrN$vfNGDbMF+G`?((u*04J9rJ2RrW8XJMZv}}UF@U+EbGJ3b+Rx_SHk;EXY7rkk} zU)E>c;mB}vOGf1+VFdOCEDD}TBrtW9wf*rLgvND+I2Ct8Cz0(WI*Dm!)exuPU$xW- zUswb+oX*J!1Pnx|-gQ{fa$(^t#uch~WyL1w-m(xR>_rP)Q?e>o(m`7orIU*~9R?;4 zo^v*su6#b@X>!o6r$e8@FH&t4tid|;*%=`qnKX{u8o%faTHz8Bm2{)#Ou?=X7pxg? zFu76}qC=1>FqI<}Kb}l>R)ez3Vf2h1@IkvRg?56kN^Fh9APB&+mG~6b=TdLQb8X3B zMwP6=GKuy;Nf{8l>2f{Z#nz^p9js=fT2HBjSqdj}Q;K1j*g&0ukHD=3W}244fTwTR zpk4-OtD_CEG{g(V@q}C1VF*U#^r1+CVvhLI$Q0)E0<*#WddW_IK-fJ zkD?l;QytCIBbwvn)LQHEN_$?9L7?Ut*_1sN@njkdykF&1p;jYUhc(C58bl9DE4!M| z>F!cj;N;A5RV>XglkUJ&2fZFHbvi8Tghw@5SW@}rVv>^tis{!xM=o=ViJXdJLNE!L z>L}Kf!wf5_1Z%n&pH`}iwvd{Pn?zu#y=AzskPNTdWKnUoKIjFJ9ad&_V`3I1h?Pfv zR`nKGvQzAWO@-*HU4ck-DTvEXaZwmFz_=yZb`cBhY_jP3F{aDpGipXCcuk_8%l1;; zMSfKfYSnIIJE1GE zf+3r%YvBDGkStMo*B@p3{JaIKmt37Y&9ntDvC9 z99CKf<*TH#C4XF`32?tT$u(nzl;(l8^p+il#PVLg1ewu3sbEQ8OvOc zE2FjtmaPRr=M?f(IVA;*#h4?^b7Hg);Zlp!@s zsaUz#D)dPcl!*gpHJ#7N#DvZ)C-q#n1Xgjn7%RPHoCkNDWadK94gJE9VzI?~>;xnY zW^~mICd;ztJF=MQrLwNx>)JDeP&9o zT8Vy!_QhtOp9RUqqQea;Txi0nxoW6H!z{7*(qyKyR%$4?Q>s0dm2jA~pKO#8eo13o zADpPtFt|=iPifMeC1}0UHDpm2+arO?mKG3uXLKpY2@%TCad{DbE5!gH5#ZXc5^{bt zhC$?pEOw%2B`)hi(5j>*JIxz{1J4Q5^XS$GggF(s>4t-Z6Hd5cF8u9lyw=sl?#Qi9 z7p)YR0vm1@qGuGjn;J8UYxA7GG{jUf5PH>0AnG~b&@ag)jwGE*HH2yWHaE(Jkd4Aw zlRd~r*&Bmfcj-_J-EN7O3~!lr4Oi~Bv0gy>fhZ5F>EW}vJ2yjxk zeDvIbRAR0@SA$Rq^yLDcab>$(sbx7vP|QeQ_?Rp<#C$S05sLLhvXG5%fP5VPP226NkMRCk{%8AfCY3Q$uez!JjNg@O5mAsy1#1LF|FJW_qNX``wY_@W{f^( zuLsJq6AlD{g(>lb;u(oLF=?3zouFnp8_XMo!(pg2N>tmS=(^g{ucpR|$Bw&@Kyb0C zQnE5suVK{M1Af`DnvfS7SwJ^fH{Kjpi5%FL1f1!N-|58*hAzSYd1jasCUh}aYxOia zSEp5f8pbyrwb=l-h=tgVgPE2UB&k_zK*nhtb83*wjna@}8*y(kDTdn2&&v(l0;8kn z8{q$a$T0*irFE5z07-`=uNX|Kjvyh*=>5hd$1eauF*5O~(5g1QCLkzlu<6SM>|NI7 z^Q(43j{{~SX9b%`dj_dOygC<_%p@m65Zf#?CmG;xK$2h~!5xwihd_$QO>^bgDowNf zl|f8>tkYXoWzI>djf7xOJ$hB{#8RCymyoH3VcMk;mmW3yNw=2}JL@_g&oYDAN_B>o zLsgh!-2jCOMWm~~R|wB`*FM(p^u4*dvZ)vxAJJ~^DQhmY6UQK=;~u5856^$Bz$`~H zGp^5t^~{+q8N+r{wo>5DvBdRgyg-s|Y>^BjU%3`w*+B!Ja~dE9CX>rhbIDre#~B@CEspLktau{5 zq}pqM6abfpZ*~^)fjDNQt`r^%BE&>o)bi+li(&2)#<3MVXWYr&6c&|G@;(Y z>Zx(9kRu>GfaH#$O|yU}=#`3T^7fJ%9j8p5_A6Ob5+s7=M)rvlENiak4YQOl{yLN9GU*UcemUrc~vZ^I*LS z)TC9dGh)|d2Mq$?-6q7xQ-X56*1SO&cxM>vmYeLVu<9z!WS3}rD@=;V=|sS;C^tx9 z4ZTpb9JW;SwpYen3nDsHM*9PRAyetOtnNlPfzo@W@O|Ov~wN{YWrr0s~ zUv`Pd6Qz#TfP6D(7GyOO=ky-cs4v?6S&qrYgGo*#^jI~WHpa7FGuxY&D+I0jrqZnU zLWJ1E)WLX^Wqe0j6@o=2zy_7T5tVsT_o$?_;=M(y>*q16MQFX@C>xsWK19M~ZMAMU zE6bK@7r@MR7tIb}c>pkC1xq6h8#fm%7PD~mL`)Rztl=T%&?dejL};M*=40X z!IOaRHs(S%-5~`GpW+;?vKd$KtymIpk!7*JoaD+J-O1s_Op0OgRVg{6n_R1!&*93* zv@v_N)Qn_?l7$qg-dLLAB1=(b(3;TT%f|JIYI01|c5AtWpDB2ycz>1XO8GULY`g7B zj_OO(q*Eu^PG?Nyyq>n;xgkS!@)^l(fU#!*XUR}~*EA-zj4(Ij3_pnHxM98`#5&!1 zOG?lrgtvO7l?oYlb zEW)%pq*7584B~|ngmTiD$~t*h?vF``w=}Wc<l54RaxH5anbe-XLNMBUFh0+?1{x z5*d}~C^4l-nms3%6$s}B(R{G5R3=z+dr@u52~z$9g3WM(ab8$NgF+mdcOWWO(u!?G zwLjPy!?se(gd45mm=G^TY%X!_&7^U^Yw9T-Q)x=M*}_%MYcg6qTL}BbSi!i zt7ck3UKfi(m(rT54pmzwtd#*~Iffz>AQ(#lR|AGF0DytbYh$C>8`f%&i6~Lbpk;{? zK5`1>Zn9W*f!S);flyCsP>_ysfj_N|7`C9SbC8zyg)CG*$reOUV=2B|4a;(09)ZuC zEQ-N|h6J~s3E&itQD#1O%{M0HOh7hUmrbtg9nVgS@Oq$)(stsPM5xScn8LV*x{vbx|9D_5yyA zD2TdNaRACP3a~X<*b0VTFmojZ2&5cl%+7339%8D6<76T}0$>_y1=FP*FX$msw&eSb zLBn7=)s!LI@j{+bl_nKmCm?N!afMZpFw%xGvvsI{&5XY33=){dlaRBqSbAg?N2~6# zB|_?zE!m|`Lr~hSy=Zu~n9)%PR^rW;RHvD25-TfReB|4t2Vh;Vnw@l2E!ht^97-dJ znLpHJkppksFFVAVRy;Gc%)5EFk)=g`1X`XbtUCEV6e)ZYl41<6HDh7*Tgw-~yM^__ zpv=~*)rtcJXe$&(aXM4ntTQX*oiv+g>ADXEAT#G0b1#Wa8&azR=}NcA zV^xWk=cBa(sKtWH+qwzxs8KA--L{uqjz;Rtha6ZO3AL_{`_f=Vc~OKmr^@Q8wONHSjI1O^PXJF ztrCgBY%yu2`Ff8u!?9$RT!jWJ=^J1v#Fvp(XF4J~LC)1rK;6F2MFeHn5tRwZIWB|n@k6RjjWeg}fL9y{}M_e~}y^yRapQ~}^(U@N=17GOp}GKt|ZVFz|U6Ep;?S*ljY4R$TI*d>{7g;`Z@ z0PO}vznO6|LT}lS5*59occH#FN@}byYE^`6iJ9|@2m=PRNKUDEBG!efPq@d0?uhYn zR!tb`000A6W%PI!qdO!h7FPwI1}DM|PlQmGg_#CrzrKnU*A}!8OjaQ%z=D%#F1a}0 zbr?%scu)qi<^j$x+UaB#_H)VG!@R(bgdq=Rx9PMot0)iz4xJgADWFiuGcjMmbTo0)Js*5Q8WM$e++)Ywitts0?LN4;u<16N`U z%U3!Q^p$|<0N2UVwQO?<2-bp|FYyIcZ1t1lc8HO9LvY38v@Z-ZiPqGyF-w2|%&Udl zv@qIP5gWNVFfssLa_+d`3yoq;BKP*steqImXaO6EP^y@D zny=+KQU#|Lz~DK@beFQ5_lyGL7exS`)%3KD^I{r2oftr1A&+EoIN>rwwZivTipxV{ zBXJ9EnfH_tt~4Tqr0lPIWvJg1WetbgQwp%@rPgg^oHz+MpJ+EiI}nyNWEF6LoNYiY zwQ9O>((<;EZE;|6l{71)y};(UTq-t=wW{MP9jk>l&f@`~h;gxuO+aRn&_f6pq#yyU zpRJ39VF5y`b%j+etkeM$A&-irA1YWqR-VV3le+8A!fXa$+|f{-O&n?sDW?D;&!?cKtcp}(PSAz)Sc6A~ z+#Z~!kon4Pj;-{wTpT(>>eE7J zJa5es%~`?-O|?=f(J{v`7S=Le?O1ZjAC>x6%U-f&R+WNSed++i8C~vnN^*>6W>C-~ za_g`pk|dyVzDPswMzfyq!@=ZONnLbG1{syyY=bewqh=Ko=IrpONn(L2LZN|$N*c=w z1e3_GRJ|F3lKEnSZ&RtA%Izua=&5lB~F_Rq@OVDEy_5``HY+8Vgj->#`(xCn#LBR)*1p*-J zX$E3}8Qv1fPlCxDg22fioY_A|8 z)LO0F+O)q1N~*~^rz*?#K<^t#2x^p67-FqFI%=uO5EvsHt$Y`1sZgemx23SKF-lM! zel#`Z#F#69>o@u-dy;FLWqB>urB>J`!S$g4iiA)a6%Nr9Dv;wUIZrxZ!i{_%_Q06?^)P~ln9^z{`Mxr1$xGV@ zNP=HfT^x{sb+Utz`A}MW{(G9BgH~e6BnpAnwk!zP2xyroVsv|Esbt@VC|IRywT`J{ zK`k@(q1Z5-U>;57yQVkK=K70L3C!F;n^i4^uH)6%s9neooaJBwR;rQZc}s8CqTpd! zn<=Yx*wE*67mAQz8+_Jnrs1E=)b`+1*ykS91f#4ZQ|7yAJePAS5sJkRwfbI6uQVzd z*f@JH15&EVbg-8V@k*5}jhIr?H2^1!P;bbZvL?V|pft03nqP$eU$kpcYSa+BrQT9& z_*|4GMX35v5lt)Otg=ZZg5mb@+^oc9+y&GE0H$x+)uQ30_h7p4KI9Vs?vY# zA06+Lu1Mh}W)&E$i?@~x1kZ$18TA)UDPyl@&_BrH5)O61Saw*=CiV8*@Szga^o(R@ zU0mj&$ljlqha=v?6O0W}SB`?BPz~&&KJu!AtQhO~YO*siayCoogXSy^=%z6<33)Y} z%yKm%z|@L9lNM7k9;2sX*=VdXWKUY;xWq6K$VQq>h>7xI!nJd$xYjMl_$G8u5=%L@ zT61C4i*+eDB^p$(W~q8TLGfXii)zz4SygnWZf-KG>A?UxQvnl&=+>AB!$8)&PF;bs zLb~3H8~bzGAr(EwLD8>N>Nu2~r?epIY;?OnIEhZ8j!WHBBpwpvt==;lgVttjKXp-51HY2i_%1t{C$&u1{ovgQ< zc-wDu@-ZltXPQu3P5Z>6R45pP)*trj^JT|wq;;GmMBlXgnqMixP6LALutFj3K(Ypk zHdREZ0Zdy)BJ6(cSYfG_6~yGc!}jOFaFE0>oRhKKYN<}yq{$N<07ql>a%-LIHw%Er z4d#gmI4m6ihgQs~%@=e*EOIIzW3dt55lR81sL+OiOF~;9#xd|9)nREAFO{SPM0YiS zMI53C&9cz(T2oel1Eru>5c=a*9)NZ-If~6O@P^qA8+7r7-YEs*8uvy^1`lXmDpvIw zJuWQVT6(}pcB|F{Yl-*c0P9Lr5^trZ!)}`Ek!@ZvYtymY8ZxzM5g)k8U|iPckSp4? zftbLiV_H!(XhP0Ur9`HLb%#?*Y}UF%F2zABUpt>!k*To7BMV3s*gp>Huw<&iu$5Iz z#m!>3WR6{G0Zk|w*TnG!9d@?zTmjls03s%4LUN2R)Wp>7lv{qWM>UC<;sC4#{f)Jy zDla>@BNK54Kqcs5@Etowdwta#J9)C7&^iHwGxj{2Ee$exrczkbMhu$Q47JiCmQk5) zMDyKHa$G;tn$Co2tz2+uEX7G)0$BGJVrtn|7Ys2SL-P&6z`m-Oy{NCKz7Svb`F1Hb z;N$(61psw^#ScBaHHs$-{Z%^b4+#4UdQ`GujQAC4=f~z+$BMP)4Ck{B!?ZxAWUWUj zbJ=RN^F28A3pj@zS}-Z59hB7>T^3>z!+50GsX$s5Uldo>I@ZsVR$-DO1H4Ji_|-&^ z!9?rzQr$P2bvjH$A$!ON&?u!qje_jzQGY?$Wnryn0ZsR;N=L7D>C7sL@v|hK8F3|- zn#3kEvXNw}C5f6R7E~cgr%FoFtP7KJITjSn!H`a7lo22$=@QwmF$eGszXz>>tv4$D zREkOYvm<6(WpZJ+G&Hfhm;?9|jLFT5!d`iFk}dOlh?pR81%ST7+O3kuoiv*AF_0=k*Isa^$~0NE}W5Ry)(eZVfJPGIVl z8YHmoq?$3OK4eUFe%bbwj1M&jQ(8Nu3^8W4Z(21d;cHg4m5J-ss#eLR$Z0bmYI|W6 zpkT!o1o(W(szD=(Txt&}N}>Vssu~>(n(6F#6M8PmgbWq$ur64l+zd5nkd{mKbHS|I z96Jz+B`PW%;!85L@}_cy!D>x5b8c)lC22Xrb~tFcz(#pOg|I8^YT^O4T0(NMS5tJr zZRJ)>QmQs>NaMOBbVK+RFw0QQ2(jPVUM&#<7dF?I;fT+?u(O)eB%V<-0aWQAg3k1) zb{$HTsaQ5>tn;y@r8@Bt%}R$&quoZ*6h?fq0)0pDA8q9cXdZQvV|At^nhZ1+)GM)q z7h{?%t0c4ds8~)ZklC86{%WQ+;UAGs#fP=}w1{>2z19HeLSBH_p<%&tCXnfh4dJlH z$gML`S3^q6K~qWu>vpwZAvPez0{k{aQNmoV#DOOBdsXsU71G){x2!|j3BgO$XaLHp z51?|dKtxr9o{6v=e*je5wp`wyLx_S=fOC5}JF}QfN!M6IEEeLyQN@!xU`8qpS1BV3 z#ri46*?Zpu6mq|LKu%Nkel>Gg1~~WvX4F(UOxYC~e6ydAFlGJ-rks^TeV|3*n^0fx z0nEdm(zfRE-l7ax6vTEy3Lv|v+W-)4y{;Xe+3?7U-bBldKOBLDf3%(4gTS@AXX&tKUl|j_50o@Z(?EuDu$T@`ZGjO5~ zn_AW-M`46YLx7fO#slBk1kX%Vve2}pT8TnJSB8=R2WMbRGNNG3A-#p-1oTBGrL=5x zTrOLwDkU4L?pi8n&v*{{5i(?ZR>_h8*a_)aE!zRz#@az6Z);4kU@Kfj&{GiK1*$_8 z*P7^63}~5$+zV7b$=&+Tm_M&B0DN^JQ$f&T2i(8qf>Yf(ik2JxuE5VZ{xsxf6tq0ZOEdI?+|4k;WP)>JW=G+d@A@FkH|<80pqLFIdKDw(Q7>_18b^AK#- zpbn9TH)Hgfg-OsT=eFHpt+?i51sv>@ktxSml`W2|sYTRNk{9wNco7AH@0qSFYekSu zHdma$`%|t0{l2*+p;}?DmzN9T9+^+^%|Z)I2w`W0DfDf_3p}<44ev}_m0Z0AFO~#!dDx8$&5Z3A z2d#-)C_|NXY2-PnGfQ-&5Pq$W0>f^arfH8M%xKqx6(EO#QyasJ5K?(cL@jynhJ(~D z4xzfEji53CEs%X;R`4=-Mo3n*~5sXS5#&x`X(FZmHo_L3@cpC-w5+dsV|5gRw$Zr|XASVfl?HTM z@=&GldzQZ@@}p*V7`{dUcFs@-3zD}fooCW zp#*(LgRlT^W#TFeA?z%h>cPui86Pij*-XZ<(AQ5<-BAK4h`Bgm~?{phWcpE9@k4EDZysMdLj^cP&At7?`N*4tXR#sS7n}e!} z5yeh{z)dGR8)w5f-Jz&fHk)z$xzFZ0Fo_nuhyq4CuS1~I%Tr?$Z%rWe1WXZ{)%Zd{ zgn3H9m7($guj-`^+ipS81}g{ZMP0)!2wDZY`SmRDwD3iNhx%_TL+-uz=PqGEg<>Dl zaOm*`TyN3#LQM_d(`ZZS09F@EvPK~+P=s%?@ZmdyfHOfi!D1rtlf-xd>m4ZJt9OgA ztuxNI>+;GbA&3j_g>P37xi~$I(S`82unjv{lc3=VXc2M}&{R5vCUB)kURyGY2_wB|3=NT(L)j}f&3JWVMvI9wuW6}xHJQK|RrSCQ zWSn+yQeUZE%$z{+IZs$Kxo^z|jUJtbHw!C%SJT$E?M$3Xr>wPYhv!pnidDt&Lep$l z7a(Lx$7ES<$xg}QQ>j#1EV!Cd2O4ON$S8R+CmWhZG*af!&TE>hC-8Yd4Zs+fmbZ|@ z>WG*cR*Shg9m9I1Va+JiCN;3Fnb82=_Q-ksI()Osm5G7^R?9BKIq-ObPTSdZ%GFxR zvOx9J^mMr{0GyqY=rWFFBAcY`35EP^?1v2>wcpRQCJZM4$LT--&>EG)U0X^el;jq{ueaV~-xQUUk>FT2AxeFyP zEjdc8-JU;Y%U=K7Z^eQ}rJ_s2;-gmZ<$I3?|JANSGx<MIE)D()j@E8>%cdr3mvcI|b3=+xqCO(B?Z+VyG)n+R<6h

)5~lpq*6_)0 zHuSGr;GYjx!!8BH(twLri*TxKZrs@vcq*dkQzFyp$r#?OZ%xLm&FYDi zm;_c@d7#Yt9SqChy94A#yfzL4b-Mq=x08?WaOQ$oF5qTHzPpk|UdaW*-&6^LwsP*$)iOFHh)OZ= zK0Wd=dd^gwp>JS+$zyTZ?NQ3l;cn!$0naB&78+)7Au?ph_lxzT+e*c>+KOa)n?Y)s zVp5zpxSU5My?2=AB3;6CXA>Dg_esrhuk-pvU}TpvUydbHeat= zZq0|i`g^5rrMSmcv>7LBpBkP>qI=RoOxBJ7H z2OswLrn=Frq;n57@=|XQ+YIp3XIrL>x6uAFoX+y3AE+4i;)UfM#9&(5Hx61l z0?s(`@O)f$n5m^^d@CKy(d;uR0#_S&4tRVv4|NGq8GjTs`1SQe5=XxkzVPcNU4)pE zO%XZnk0b%JxK{q@V6?~u#kWwhOdSSy*mL0)U_X3kR^-3S$k#MqD<-+ssE&{1E+NVA zRRqVtc{$Kz2+ zZ!7g38@OFhqx|2ut5e)v{b!BBEj{j-u-3}4oiXMyujTT{yw~-LV%k@woEA^bUkuhp z4A1dBV3)cJY0?gF&7%ms@2XO`yS1qvv~OP9+3(DOT-B%Jw1Ylz#Px087Rr{qMfhj( z16tMp`aysHpP%%9)d8J7-eHFs<_%sxu0Z@2=KX$D1b2n}35KNk@^Ko6dD+@3Ng!LA zd(J}JNMig@x_GzKevik0qy43m+}-c}jjExbLGq9Y5x~1HmQT{bz*Fsf%qwCGd+-x3 zkstS&oo2#oFOQd7aUA!?8>h8iI?~zYhp!;UTERi-Jk4hG$tGm$;v1jY-8!S`Nz)|y zCD7|;3^mrza5tMdY*b(Qw+gwuWZ&Ifr<%9(pUUCE8u0xrpEsPSt@p?_3z(yaAe+PY zq-;NR7hdzxi_sk}R@cw!1lL%?H<^8B&y3EBM+yEMdt%Wp@*Qpx`)$qh`(+Fxm7@Ql z@5Z%0%+ou{*~S-slequtw|h2UyuCdxb>&u<>^|!V z)_m9W94!jvA~P??yO?sG6?KjQ&k`<$__0zDGkM?l^~_qQO}cWl#M=bIr@*Jb3vjWB zF{}JHm9t13AF@QY5BzwTcQWHNww4hjH@Rtxl)o`Kh*U z^@clS#Uvl(}wW;Jp}v|b-)`)qPISLMXd_B>t0&N#`;RT z$?P3|*C%E3bmo2WXVB}lc3h4J#=+%IUu}Ey2?^H5QCF<3y<7i#UN4#LhJBsKA-OJ( z_ww~EM|%>Qm-;;o-|RHWU7APpy4vN(&HYSs1I-`kAY8)?KDc2pt!cS?8v_V2wiYY* zpA9q%Rc*V69(g`1CAEFuvxCwVil9b|g^%}RyXb^P$!GbNXU3Zv2w~{yD?NM0jUv{I z=q52M%r-f*9?@3X@oR6XR-bDgKQB(bq+FiqI9aE1S?wp-q_;QJ31TATgzAC&=Ll_b zqX2KB;UT0d;S#N*%VilB6d@ z=KJ~_)8YJiE3jSGT)pygmXnUljD9M??)xxG(W3eV-@I0G7M)zK66qyvoyKtcRVnVd zVd!W<3&#ShZe9 zG#rhA={-$#G;H<#GwsRU5F=xSX&AZJEZ|G}gTy)9yt$OLkbMsPT~*T;+}fF!zU6xs zXt`4M@5|!4m167S-8j*{j&rbib`gK#cb^5fKJC;8_j_;Xs%2gqjCJ`SsYM&%F>kxS zf7K5kP1?&@o*cEiQ!7vF!?T3q{mDF&-s&!bmza&hQtW4OU$aGxp_G8Dy_4i^<%l*n_oD~JjC=u+GfGnW8Tby~DIw-DTg! zn_Eu^9$#Fasal??xTj9(_dNb>67A8m9J{A2YU2$W&hs{Tu=ggYP`q$(kRf++q1HU= zGx#DONX7;WVR4_vEy{<@z4-$vUCIlbis?(xrWd_@aJb-y@w9n08=qCh_flEqRZ>Ph zZC1CVtvYoa4P2jh*6-E2M`r)xIEg_CGZbFOQw=}3>v41YT;(|w+~Z?)-F}SkDuiRJ z|1|6-F)T?uBXsY`jlDUP2td4{7gc&112Do3xaO)vbl+ zwGK{`Ne@Tl!+&QhRKG~Re#*0WZh@+BZ=z)nP@nn+R{Gu0RG6%z;9DMJl?wdmslD7A z%h0QSl|Sk|GR{|euu064)em>|=}86)DS5&jO|OJ1wUBkb^Gy{$EJ?9Cp2e$vIY{tX zn>tG%R#!Up6&v! zkl-KgW0gDYUA)WENzh9K7kuOh*_@qjFyg5D!yFm#CdlKHIi@Uf8}{LG@`28w{E)^Ue%CjFoh+=J8Uc;%$iDZkz?s1hUFy7tF3u$SyZ;828h;dZ3xV zkKLW!SupUo%}qz05?Dx+-NoXtS?6H~hSF*Sh?#5EozgdVnzh#>!(A44ciyd?@jfG} z3D%ZcdGcHO?pVKQAFt`QTfS)|ZuDCaRq8p~CZ8o&G~-R%71C7#*a>C%@!C6KzUI)1 zN{j5X!GN63X<{f;k-3L9r03V$15~1~|GPdg0%yXHuY81Rzf(8So^{=LDfCD8Ezt?H zuD$c9KDP1=lhj>u>f0TsbM%YKjBP8-t+L$u*GId7nkqjqW-3cZeISu;9a(>aYS3Xs z#qM%g{wC1VYN?s2owd0XFnjZ3_XDk{c%CkExihU$-aYT)^93V6q9(LpDA1OvYuk!z z;Dfd)&sk2`I3=fwU(bdoX7ucQv-KVk4mk`{fEj2Qm`;MJy7DN{`4`=zq6(GS$(9Gep zxW0{9DL$20^&U>=Qh?ElP7pYm`e1&KpO*|4(k;(9u9O}aNO_y_G>eY1U91)A#+kuM zJEA8@Js-yYx1;{NN?M6jsOFg7=1py%uT@05mu63oYpCHY3 zYc|5}M-Po^u3e3}kJVLhZo-abbalD7F!KFRgv7I;pS=ymK1xV%XH>;nzG=^-2h8ZD zH1PXDZbdqRfM78~Vb~{oXFp!Aw3P8YEcN~UoShBnvsuIUsY#e9ozm$lEeZ{SSNC%- zwgrtx=BPf+w--%fnPg$5J7HTI1s(rWMX5Cu8c*51=?bR0^3DcwX0ngT4;AKP8r*bw zDEdd9KG}xqZG*cWH|BhYnUF=fm#)~1{EUziE?e*l#*BZmos*-DqfrcFU zAe!1oy49hFGo>zk2HRes-|-eROwFf^cPkzrmwia@k|369hZaUq+_-7ZoDEQZQB$O4 z+FZ(rX;Jc=_~FHFniCP{FjXAmJ)V7_Xg*K(JXOu=C#+cT1bNG;)b)MN?!mK355*SQ zCQ3Mez4tCN)2RHY9<`CGlsprk(IR}j)8NoYqYd_u+h{LdwciyOFKMG0E3Z@Q9qQ?L z%{BF1r!jZn5&LDY4x;Us3Z@bZ-Uo<7I6lVq zDP058N5kY`dUQVJH>R_?rcs(YbiKvc=i><;m`Q}^e97-RYty6nXocwWWV@ELXj0J1 z){3F^*f@R6azwEI%)07YThMd(Qx%w)y?3#3=M9tI=bQC$&!)*drI^*(=OMWZ&s1dU zCR5AXZP^bK+*JWidmesK4$7f`-1{+cE~!%R?IZs7<^zQ8FVlhxq zc10+*M|$rIwfQ-$Q~Y4R@LT{%;0*=SK*zC&SqH--BnZIi{D=f-!io>iKuk>)c2u)f zo_=m$Atu1YUpmL#;ab`vtrbX0-!%OMgwp8P&F)=-$a2jKa7c~?TTRkjOebi5RK0Hx ztQQYEHp;Z5|0$MzDem-l2i8J!lTT(nM|wyd z0YN|bnsqA`WZ%e3(AF-;pUxe>1vH^#$-paK4oD`wpnYoBPr1~*XKl+cQBhLC zBD;E{o4v!F+f(t1nKXwcy`T@1XC?C*>XfjIFQ_$fq8V_AGC@KSvv9{~cq*G%$Nf459yRDB}U|nuj^G zSsh}UmtoFxDnYURkToXDoNzG3h)knJJE1~$W1L+L!P;bdMNIONS5(*RVUc=Jua9@8 zv!{jm5C#ef0InXz`FG>iI$Jm)7sBpz2>VM@^L^GJtJN1jMlFv4RWX+^YyJ(Hj#Q|< zG!+Lmd;-w-q7i<{7%qey)5yTv3^8>z1l@ktF^nOLWe-MbJo0`nsORM-JbGQ+j6YAjo$*A#$Pt zGlzE+lCjf8hYwYmx8XiJr|+Ki>ux6;%u5nJ4$H@#pKbIqc!;+PC0TZ#Z)q92L{pyw z*f|YmvbP%F&7$XnEo1hr`O=fNuj>~v(Lvy*`Yp-5Ze4z?P@Sz>+y(K@?i_SX?d3UsoHSkfBAswOBxr~_XEQj|7M&8qFDN0%o_YLq~;5v{8j=~9<2eB zbzsEZ+bgv*GnhOK0Wgh&Zr5>O6OBnQu8xe9MGKpiyuS;L?cLX)V!C$dU#CyhZeM!W z8B~Bdm&XM?GyD$ci{@0^o7+gZz9nsrZ+j?myhy$uk$Pb2mLk!-Y5;o3Fgb{HHSZkV zknx>%P7O}he!8BSsEorslG2;AaXXXWoM1>nZ`-ctj9Zswy7|txJ|?rn=rG-RX?^00 z(ng2@B~}THjC_H&YirzwVlc{$@?Gu6bSX$3LOuzM%Lx;29$BPwYvdd>^8Lh0iGrk| zsMF>wPvad~PbNtRdD@=%Jm%KRaMm9Aj;L#ZT5 z9Kc-QJS>+xMy*->2KhVj2KUom4c^^!8mb7De=092Q*{Iu(#1kQ3eoY)iVD8&WwXJ2 z=Yf(KNogIyJ_l;_abZ$p1j9AA3{gZS>1ToZE&6-a&&TUe%@+IwJ81!24EOH0XfD@+ zw%{ZS_jP#vLjP3R1XpY4&X)p>!x}D)C*RyD$!<=CbVwOV!O<+Iix_MK?*dFDj0Ti! z!?tua>-*yg0i#ar{X6abenL=Dd`ZO93jxy9wZ{3lxaL1NRTSKkv+!{5T(~(1h&pPWN$>LFMkO_0>U(<4 zQ40j+6LMm0OaRqNdf0G_HUpvAKjG+Jhm*7d{>XivjF&jEYwN;0PYkO$3~gBM_WnD< zV`3QT8+_b63^1Zyr^oVlLQoU%omO;Az{yJIMp4(LU>@^gd993tS3ekX_=Ly5p~ALi zp96rs?S6+EANl)D)VVc2j;i!OR8v>#Q+)LpCohp%#06L-_BsAuCF2qnYcluK;4rfB za_sat+wCowZ%Tl57vTg(U`$TU6DW4t4v4(kz*M=eLo&c-D56=Fc*?*bq<=4O zW0tm&$zTW97ydo5#U-dsktcug$wfx^gjZ7z;0t^1G2+H(Y#bFJ>8w5Gf7Rq4(#5;N zDN0O0BUqaUs4K{0lJoxDue)_yI50v(-cHM!Hc%`bO%#*B?`n>$63a0Nol?GKPh5>wU5G)pD_fgOvV^ zn38+#X}bds@8xv~MBCyaWT__ID?J$l5vd)WnvZZ54A%VnvZ1UkT*o&<)~0>x(vb&D zc+vdx3l8>n#-9U|AAsAJ_x5;LT~GKS-rYW2o8UFbiG&o#8+ZlbWnPA2<|{-6d<81S z`hk(WB_fYRC~|+UBV=w@ke5Rk9l)DYlGj}Rv;m@_Cxl@8Xp6407l~k*(7uYbHD;7s zQ4dU2nHRp(<7ciYI%)FNE)1Ni!5DS9c$YFwTVO8DIN0lRl>xd54?}~e6XgoM3!?>=npQDY;9?44X7z{c-WQb4V? z&iFy-98@t18~_(2*N?IK5h-=W#(T8onmp5>cZUBeABefn9e=G^XlMCs3!|L$lTx}9 z=X{S&imGpS0GyasveMP-df*ok+{cJ5U&lM*BMvM<=ILr_Qm5*?X$RepKJ*TAj z>^-wR?PBqV2M_4NK_)@%dQGO9EL*j`sjPOA_;( z@Ft{adM}UwOzLR(8HJ|cs0b19eNqyIKzr{~b8cqG1=$#ocrf+Qj_&eY#!vxdgJ_`#SO~2FXhSB zEntH3sFLr$*^E!G&{HsPwyzJ=#`Ohe(i@?C9PDQ{K;tEMf#BKTt-gsvgP#CrwHJlo z$xi+q#ajNnyuAs>RR8Kd~61Q3EIH|$c-V@(ta%(nc2co(pV5bZN`=K(H|ThcDF@cma#ofa^S zp%QlM9o}fPWDN_GBPuYzJ>U51~G!>td*kHS=wxM|$?Dr`t+x&_U~6PJ;jtmV<+Is?)zS}Zdqh6(s^-fy$E&9{hYIUc8Ji(~lkuBZDVHF?wr z%I|8N5WX5}40mz--+E}|YNF$s0=s%H$6&STAr_oj2UcSYa0X1=;ySt8Iukba3P0qY z6tH&}NPFUZ%L6ed zlp49^1rk3{@7RTKTM3(21~oX^yZ_!};Kuu3VaF;ufmf<<4TUevaym6Xi z0jtJhMZEf6d&}@j`utB2=nuPJhp|JRp4Rzz6+Q!HohVl24e$ME*M^D|%H24a5s78h znu+eGyQ=tR1n+u&1L=ad4rZ!&$pq}8uCx+Lj3--MAi}84SYjB{b3M3DUWIBD2V;IdiT>P&-9Miv>g3DqZnE}$;N}lQy^vCDqJ9Ir!BE~1bm)Fx4`kOQ9tr9+^uTD z!*Fz*)Pd8lMMi2xuFtb{80m>vX=E+s8M5JFzH}P#@f;htWv(9L4jK=RzSq|F=>b!s z;ce>%ip;*a?nDahe*p{j{D7P7MFJhjN=XKK7RA3s{dh?DABURo0I_eEV1JIF_t0iY zhzFNQDx;*HZ{|y71&Xj$jTNR!9wb=(FbX*D8<1{}-6Yio+J>XR*(aVkGD>uqb?EEinl&3T8a1&u;F$scCmlxBGR7Gg}EcI2=R=0Q^b{9Q}Izw+g>xx@5e zZMWLerQjn77V6Sk_C?I=wFsvxj1%9;LcOmkDgTV2%pIstWg4lTzMPC=hkAvW7qR;i z`d5h!CsCEa4|+ejcr{D+1QKZkhzcYc;`>`t=PA3K#boj3mHiPtDRXq!n3BG5r!sJt z?nh)M9t0zs9tGLFm7fd1WT*^FL>IJ0V6~`{@eD!R%ja+yH%KF50ix;*v!HwDhN<}GXaBWd-rs{OAM;e~5wxR%lPM$?XghN&* zK_2^8Bc0-|myh6`hCYTLDfeBkB9L&x0nvHz!S4;It`H3|A0AprB9T6J!}?_5HhP0* zoe+cnS|q$RhDY(|A$EeWKHRneQNUc?<%vO=$Tv1z{(F?*Jx-2|CxfZj#J}}xaY1ER zKXt3b0dsN8gW@{3z{hXr3De{2!)mz%F6U%nHBo_u8NEwzkA`)kz@Ah2tCjD z_WmPdThMTvyvx?G1VO%j+5wJ+%Ujbw)$v6gwwP$64K2X-iinu`+s{w-+N7-_2d3G< z`!bfo%K(3Bh(3=?ltEZJen!Zu=R~pr8V zk{{?U;5jkB(QrQxL!OJkiN8RN&%QP(y=yJ`#yE7Z((r@9k$WxI0e~UWXDOVedc{Go zBZ`6#VktMr-?sy+1*KJR4(v;NZ!j;((*!uqYTOR#{wqVK`GTeHxQLh9FD~-^E@Z|w z0Nrkn?Jj#n2CrC{#XTPv!8h{xMQ{Dx0gUiV@rxjzraW(b-)i}CO-tXy`#|R;n!Lr2 znxm?p(>L4u;2)g=t+*8Ml0HYQ6MJJlSL~_?g`+v;>yM+kINCs&@c#M1Wv$HQkDygi zxtA0PRP-bATFi5fUa$+Izak~l?|WJHlCai-2GZocu}XNOF;=?>8oUBL@urnkmcc}G zkit(0{mwv-J-4LC#=!!0u3z`-HZ^H<`@01eU~Eq$s0A4M9%(DgeU9tD zx=9@Q(9?E%_*(NHOeJ|$weZkoT9#w_Z&Ismo0wrnf|+1*eeMpRq2q-t_^?!8D^|j5 z36C6c#V@g1VT|xm%qU8>3EwwO6?C-kfFx5_8WT|dEgX1oW6pc#>fULWVojF#!fULY zIlZyOjPvnqbH4aZCt%VY7IHl~Y4$NnBe=V|ChmwMp&UX{EaR31@2)&&<2Y?Lz*OV7i zt1pz^OQzK~>FXPLh*ap0%$!zCPQqPd3#aY|{`T#z$AY&V3GeLTMVBq4m5v{Ngzmrm zUee;vppIms!)=&o(fgbaCipUAjnjmovr4@ho>4pQj*fU2Z!rv|tzR|r^?*YBr7`wY zJZGI7j{r*jrP#>M22Y3mU&kB4=@?8O(Y4JtcJq=L(8@`;mf#Ja?hg72Niu;INnz6g z)S^x8;X8St`Tu;O;Y9aK{M?-2(V{nuTnNEn?|aNIZj>fzb4x&@jX7uPof9 z$S!Qg|1I!(%Ebjzda0pY+~S92Ns??!Hz@Bv0sN$lb5zRAH6?xvx?PJ1>s=X5q>VIK zkvAt4FsdX_Q5bJe_`GB%SbQrI^niyqfwyZ->P4C!64=~@mtAin-76P?*<_Fqe>XqA z;AuksGyxX5K+NEgqMJv8B13rPE8_XEtLP&7i2yN{tLts#chmGfuHU2 z5jPukDJniHzoE8*w)3TCEZa(OT1|h|RA4y$yNF0o1%C1{=y+Uj@SlnpK46k`>Cq%E z4&t*YFD}gE2r6VT<;om+@u}z}Rf=l5i`tRH11z}{MiG{Aa?*Sp8!1F@aX98 zvbL7vppMF-8J5OVpn+uVpy!ddmd1+>d0N4 z4*pP(EWll=Fe@lJ;0L}b%EBKQ)z>FNEoWC;Al&qNDdQU+V~j@c>@}XkBmT7f{lSfQ z6Xh`DTFgq}q9|6iMb%gE1%Ve_;ihtO*zSE08LSywgE$nYNEaY2xssOm{p}?siV4r5 zY@*TClX`XNudf6b)sL!Q*aX(2i%mVL{%FyMt=GfbFD&E1b?!F-ml8*>@@Stv04(bY zdo_~?+hzJCiq?|Kzp?&-8z(<-Yv@5Mbc`mnYy~Sw=*O-+Yo8mI>^xicf{+@I8h9W| z))NAKzZFvb7lP|{FkMbfJ7{MfMsBa`?kW(EK7Wk{X+~938#oSm@8lul_ik63&0=F_ zc_%^?89vWI*-eYjqg78(zEu2nZhzDl()8j;|61>-#|Ctnr^45&s|uE!4Cmas6hJ2M zKkWc^AW|ykN`1!R_8Z>`X)9R74VpW`ojLDt`15t{JGPnFC{AE{l~XG>np9iX*1pQ? zdkahJ7Yc!yEAxL6XU+R`=dt|+6%CTCM4>wpV924yn^`L^wIfB8^)v-v) z2De@LkQHqn0n%#nXCJI^d{$c~MaFk&p5l*!XwC;?b>}!K_?o<*KxEW%u2$@h-x(#o z0h|rZCAlFl(wM_77@#g6QLuAVw7~8qm{4v69mt(OYJv%AhV>rbCnWEo!~!P`;WHRR za7|;Ynjf(=7D4J5z#_B2MDUW`&t|aHu;Aq!BS9%khhIFu$wICAoqZt)ygs~m*v(M0 z(G7vH{bp>Wft6tgqNd}I0VuoVCV61P@3AI^{6$q601K!`9ti0*f zc760MrDYiDu|z~uHv!HGu;^h5VMoEPG%Qh~*A%rNrFf0T?3D3clL}tUWle!+Vb8&a zM@?AZw`WmQZC-Rt_BBE}6`8YI>G^|`qy2jc*0Q?0h4c62tr>HRRiQNGTw{hGX3eIb zdiLiMu{ImPm=eO|N)hxm>jv7PH?DZY3%Y-}YPq2he-66+_hD4vrt{n?kfkXaz1rEeZ0bXSUV`a@aL_PAfK=6&HIeH zx~)Rtsiwc%BF(A&W>CM&*8qY5(tY3SfRH|_%KW}}&iUF7@wZBfiH1X_Cy+hgE2urp zmXHS^?1SkKC{znbPZ}JECvd0?=15O6x{gP8dK+YKX-O_8EQ`c%Ixq2<0Q{6f9#?C%uJ9!g{)G|IO$oK913nd{ZVK4V4bQqYcw-X%*Q&`f70qG_O|dCzW)>} z731^SKfbp+#3%jkvmVwvG80N@#eF1$p_KE&x+8MCBNLCU3kwgnV|6OAN|Xq)9Z|hh ziwdY{*^;~(iw!LAwqiRM0^wUB2YYVUc740bn7<<|7(ECB6Atv1-Vb8nB?Bgp2%lkT z!ID#omWihXV;~sVuxmM;MjC$Z?F!W>99CER~q;v48Zpg`44p$`aK0C#Q@a#-UhlC1LRmi_N!~{VeQU zOHR}?&$Wf`INX6+=!YwIM<`cUEUxQ2B{$$(#qI@lm?;t32)1otui%!nWtb$Pk>0%S z;VDK($XRvy3hHU|UD}9*92Y-HwoRu+&uIP8q#&ds_q~j2{+;1J%v<#R;_*I+EQb{? z#vw4PV>XkL>||{y_mO$#WF@z-%IMQ>T5!< zZ=hLTjWZV9p2gqE#He*;U_7=WQah^Jo$m&3Pzb}`#|gPk7O>Ec7wr0pkYkQB6q*uf zfI#wKb&P96#rWJG&;ff;I_QS{G!0=rag|d1?am`x=e7s~tj<{E*obBTK7xfDC6fY$ z?fgENUSuEAOqN`TH5+hMf}s2OGoxzD+Y7DLMMm7ZFmAT8<7IqZ$K3@xgU?yQ<`gHx zfOQ~}2)?@O_8!sQa5WY_=74`XEs5vi0zj8^FqlZf3WzHqQFKJ0V!av=d4z|C<6)^v zDdv^NaGL%0yScMqJYs*w_#R8(kK*ruf;9E*)g4KY!dlS@Sc^LV%QNz@`KRERS+xiA zK<8_X?!n^`r4wp} zT|x?m1jk7tkwCn9T`004FEFo-7ezGeL4dV$_ls?5hcFkS#}j9Sbo7{x!|+*2CwL6Y zyIG?{kxQBM`lW7dW7KpgUN#A~R*blB?Ynk`%l+rQP6w7Bh&L})Qs6vu3P9d?gXApx zgn@h(h&9_UAGl!Qt9W3hwB)l2&rW<(Btr;omL5JI_(yd7idQ&`zQ}=7XFWh(Mv%gn zFwGj~=U5^DOFRr61J@o}jNasfyHML-a8y9{(Qq)sDrJXxyiIZj{-a={uHOVU8IWzF zS&!bi-p81+A!dLI7_Niw1FQU&cdXY{j<`_D)WOsj`EGp9CG! zV6sqCj<675w6{IIG2wH5s@S9ka6&r}C~{+)Xheauq0@*3`y|Ue{u0ah`^lQR&Bt3f z;`;%k1KE%DbVJTs1sIHhBBXAxpAVCAvU|F7Ap5j8r!92s7o9y}^@d!31wd3IoNm*# zJ-8-tQQdw(c)M#D*98(TC!S2xeQmxtw!Mpva64S`aOQ#bP>_cWsFZ~z*=8ytvNlKY zbcm}Zk~XMv3vs`Co@Bszh`I{k7qS5B#`c>Fx4f>T(EE$$@+$^bec}&8Sjb1p68wSh z*t4K@rJzIYH&zK^(1^9Y1K~&v?uZ+sCDK@UpYZ1{jtAT}i^@Jrdt?3VJ`1zV0HemT zMiI`25w~cIbn0UUn&L zks~)-xuZZ|K6r*42wn%WU-|aL*FmUzHW&{}NO3QXtdZWcuNo~KEbV}SaUN#8*KLk% zoKQ5>BgDt$^3elAup$?ZqxI}QIZoEP(7zaY*=G0%IIstPk98T$UW$FYsB=l&d-VkQ z$%EK^tBA+{OTy!ud;yS9J@_C3N~jl|=axlCeg2z>AW8_ln1gmYg5mGRVS^jBh8ruQ z?kKKe+&R_^OOv=mJLexM8A0FGH6*1Riih9XelJGs1V27?Wc=-~qP`t(ve%t=`9SB+ zzDr}?paQ+#V_-`x5!107h+_P?$kx->QYw!I<}*Y3I?^v?@QA4fcJo>3xx)uNC}tZ4 z=X^?Qc|!KfF-npp_Yn;~CqifLYrlp^9_92x7g0 zwpzI|3kVFza1yEhFTRg4{s#+LoW+Po<*Hd4HP90d3&sOk6Qrvf6WRMcfLf?EUD1 zik`-X)7G4=;AZD7XHzYZA7{fx{vR)^Ob zw2h&~={bz9f)Y++cL^FCxXUR^=P@?bD(F30RS0j=&heA>9~`10ooB+siEOA3~ohLvwF zUtncO!k#|kTEjMZZX3b9WL>ebYc{c~1K!Ug03pq;VY%0K4H)Mggjce{4_o2}Wk(h~ z;AMpsZ1{?Vp_ktNh={v=y~ysk zmrxz2%x1qp=;VR28nW6R=&khLXZuwcBb5sE`mesDE)-M0ar?K~s%+}?B^i2hq|m|r z@jcwm0ci~9$cust6z2A3f}aR$k@jaA=hR**J{N&*)Y#N?v0f>=Ns4Y-4-d z5;)h|36H4Zz|e6j_Z`1t6#uExpjW+o8K2cu!ONB^*wtqrKZG!d?xg6&Ivfw9c)7o*x&urwRiT zxBx@w6t8Qoe6z?qV)U{?L89bSu5>}Vze-v4wB=_DPoQQG<|Je6a$!bd)d9?l$@ z^ao4mx}4Yj{Fe6{Y!L39w1+_%+K?|g`(gH0y5@%ml+Op{D4xMqJ;ZBlNWrIv^*o`0 zAkiC9KLm>)msCWGgT(#;v0oml%DtI8nOU4<^%CgouY*Nm9IgpU$2LTmS6Cw9S#5r+ z4nO-zs#BlXW8Cr$;|b!KAk+Er(Gkn~{~C%*0C{ZU@p^h7THO5^8_%*J>=tAxN1Fp9 z#DUSBr zZ8;Ex6*~0{)ViwU;8QpsV-#bdr_GNg$%JVEQ9_nIyE&4D?xgCy)yBRa&jN)V(s8rO zFgq)*ImcsvYckr1TdCh8WIez$$S=LafWQF6%*SKmo@LY1PeR|^O=L7TMv5&UCcGf< zLSQu^>@rtK@6+{IVA|H6575isuvM1)w#7GT?B>hz;pKZ)RoPy^r^EfdK%|A>CZ< z*TAx1GG%Fk3`nomeKmt&L}So_zd$*5zG&-oQ5pfIo>!g6oHRU(nAt6%HSH(}e}~<) z0?zEmnVU0&KLQ?;Yp~{6Bf=EaOX_og?TvIrl%Wv{eQIt{zu(RTu|!tdB zZuN$bAU4OREBt{?JVAVri{6qDZcUKE(UeCA^d^Qxm{oIvV=^6JORPTwLW9WkYCnlb zrP@Y={(FT%!)Z?XK?6w(Eb&noD9sgq@%#HB?*3vm zM`%qrR>>=bsc*>|FkUroc=51_G`z-%Xg;;Uc8OwQ=y!gU-(vx4s7*viOiB_FK?OOq zHqT@=F*?kuAC4d>!AdV>wc_aI+9)|8J@1q1{ww#WMkTn39`I!>@lmcKa=aR$PtI<@ zFvE^~ltsXP#Qn+;~Cip%c^#G+dK;E&zwrO!gOv zTw+Qh*3I=IzQ2p-W7u&neQDXB^z>W{{?7~Kd5OZY`G6zj=FHwi{e5ginkT*=5~I=q z#4=RyCvFbxS%DqyVwlvTd7e@mSa{8~lfufFCH8Xo9L$rMh7%ROk zV{e8nq4G5OhB`gi(1(o+`z>2koNz-kF(UHi=~FH=F#Mw4wL4s7SU1`QsOx(d>e$IX z|Hj9~bwqhy?Q9W7Zn#-iSdrCz5za`zs|=w6*?Jj>Qt^;idSJkWf}~)`r0aNpI6Lf+ zeqjnYF9Rgn0RvtO!VEZgDU8_hh1I@tx^5$+pZ45e{zsNYtR=t{SqDaIQ4*aA|F2 zq`V=Y?>$;GAtK}i?`oEU-{=htNon)LYefSSx&ihmy7;^Ay7kbTc&xkUkna@NC_px* zze5s!Jc##giG2iM&iCj=#s*PsytT#`t9`8x3Xb9mvuLmg4l!2HoOh24_U_*O4xxjy zxcx2=dH}HQO%Qx22pIZ->h^bbpFJ13Fc0^w8;?wzeb$>*Si`S_K}N3B6-!RD{BUp1 z4ZZ2A?%6+Qgo0WLXdONSEf2xo_{))KcBdg}yCF2S>Q2$?9=VbS_T9e}3|qe}3@@A0v-ChcME8|2y4JMy>q^dfdU;2w`V z5k{5w1>OGY+&XfG;=-i1$O3agjxYG*BI^#w7yH0&Jq`;qau}Mp1yBw~k6oxe*TOEc z%&p%*1Wc6rpLif&1NXr5#-Vpvs$B-bVM8re66sj5HkL(!@1YRolbd`(;0@IUlGAbN$Q0`4h8Ab=jc?!JS~jHcNLB}MfyDFoaY9w9#%F9S1)QFatmDtui*gS+ z9d&^cUqk6;<4XmN2#1o1^7j1aC;QLn+;cfu}k4(>~guvmT$--LIKCftt(HThgf7iM7P!~cc{k|51RwY3E8m14+QNLhM0lPgD4CQJMbbLm1`x-@K!0P z zhT%&Gx65I04IrO_NGstM*WaU!@9lcR3=L7fR~z_?3Wk}e`(hEK zHv_G779TCR*qTlu;oKsp6zD)(KMO*_0L(H-4@hj0s1CX-r^j1#N$|(LU>f>t8}OX~ zxqSWfg2jOG?Kd8H{*n*XD%n^Jzi?D1U=4V~LYCQhzWFe_`ms z3J7nopiV;+pO&yNvA*A^Tnv`|GS8|Yt9$Ie*x{R9nuR0K;>sw2IlNn{v%c3B!KdGL z2$ccv^?fZ$SNcTVghL&XwEY`KGeow)^aa9FeIR>R04DT&o6r(|r__Wg`1NYfvsbbu zN2s_Mx0ll~-de=Zd1%;GXme1QCm%kLn4?ulC=z}pxvs&NM0^Ob=5ve@GdfVaMt%Sg z0~OdN#1Js;>Tk#fFdta-I&06+7sJIS^uG}wxm$kZ%A`N^yy}zt9wW`u`P(&t@;uJ| zEnyU{Q^DxQrgb>+iH-KY@0F2!D7}@AWHIBEVZFiryxHi<8VDXEQnB9#^V}sEsMs_` z^ds`gld;%%ZTuAK*f3q*vN&5y=Z^cY^K*Xyi2>N@n5`E%yS|`01EUncQW%gZ@;eLOH8sUDOe?>W%VB-G{?ma|FeoO2?CXlj84>{O2h*QIvY z-&)n+Q~VI*Z0XejuYq!LehzfahkAiza{LJN6SIy-n!0O#kA+^tZZ7%9iKY?KLveF`VYw#~d`gJ+;PU`;NJi_q zb%;|AGMuKrdN&P3cFtbbmGh-R%0!_fCmAm$ih;FtMEO&4EmQ9egMw_Vx` z7UwjUb^!i)Vmf4QE-!43@o0M9Q*`iXG%oBPn|vzf<20o(pr$UKZ4KDR3$HbBZlfQt zflWTz8%GH8a2wbu1+(Ot_;i?%FhYo00!aMnP<+@mDQMnsYDrk4k}D~&>=vO?eUXbt z%FcijR0{A=!*rBkx0NOz$VrG`OmBD0&2u6b^Ld8y0*4!Z4gut0trz;Z~>r5IF4h-E>Ot% z;2fMdcAOxzi5)wMlQ=#kj$;W$)F@E5zdt@z_ zQ#n&!?}(t2Qa7MEf!I?iqN{eirTAD@r-@B$#d3_MR023uxMpH!4~3E3W5SGCk#d84 z)B(qNk#hcq-FV-8v6{c|2Qk+Hm7T|X44?3T?iT+S9 z!2_gKt0Rutpk{oZC{9B;kQAMDq6t?H5eckhtxKWMebW*}6P=@O@#V*Le%|7B`V%2+ zai+cIhw+*~%is}nbgkw+PxjRwkgDb}TD*7#AAj%qo5ek=)LTd0rtXYCVf_H-`zabvXdmnh;2GAw$dZ zAa$up@kpqr?_#E8V&hMnqAY)Vk~Gg=@JG;taqbB5uG>tmiP&sK*6FlV8MnaRtGXMi zW99fNTRRa9qL(OF#q2tmklt+%E~B?E)Nln`QX9I@U~#25J&8Lid#^ZR9cE`4>nUhR zUfyX6K$%=uhF0)wOCkK%hQ1QER5K1p>DSmkHJu&AjrLx`k7GaAk} z&%$|n{_k19N_Nvk>1=6iif#|iFvVl}IV=juQc$gRsVrOa^;`As`LTL)MF0m~ox6md zzJMGl<~w0j!V~@H6K?P$^sYQ&rUPWRqB*SsWM3$zv(qO zkjuDU&Yf1yy9tOFP2Y?Fp?fo1AF>BT&&b@H_g$79I^XZOjZFKry^n9b%3-kb&Bn+0 zNgSPbTTWY1)YOhzerjr4QjD$|J`3e0e{$FI*2V4i52$evfhKRP4okf;`>pq@aR%8l zX(IZo&rtXdiQ2g>Cgo=UDDHUNW()%a_G`d zAdWw_Q(5jEI?gis)T7*8^QcG95C{1`CTPF@1scV9^Xxz-ul-u>yw)uF1c>5X^@4{S z+Fv(zR8MBD&t}2&VfS2D|9r zBqRg0N2Q9Whr{Xdt$3?iz3KkPrhLR4GL^yEh-;SnY(D;Lt<0)$Zdqu!jooWE7)j{v zS7(cW)`>oDAJ1%L>}0?Y`Oe2Zi|L<0$iEm?efBrcBX6$C){1z;NNJIGJ zs*zG=1-~LGEGh4&x1mmCi2$bXXn`ENYvHz$$7YD7TO>*PU)qqKL?7*FmBJlT5NDHa ze{TkDyVyCd9`jM_cp90wbBwoPIZGBjBdTgQNtmBg(K7+Rr>MX|Y6ZwQcA)Ne z({Gei4aJvP1-98GXj0@Dn3jb0MklQZ8cFz5lxxZ~aUi!{HZ%YX|-W2xa zE3G;&-9yhYSa(r^N%%RYXu&08C&xFn&`vCg-^#SIQzxw((qXCO?WN=HuHU962wJ7@ znW2^X212Cq=8E4ZpQ-wX7i4yyAs>OWzNaNCTDOM(TOPbLy7{2ICVea35PvIThWw84 zV%(e=!SDv%{`SdwQlAma?ZrAG1%UH`C*5h^1}{a=W=^M-df;4BcQC8f%LajhGk<$a zAGe}-+POFZVM2v=2;8oDzBj zhLrhj)faPHSPjZ)U%b}|X#dmXcsUd`KAp0Hl{+~szEX>k;Viw$vz|w6l-%h)oyFw1 zY&3@7KMJCJq%s0b63$l#g{oU$(SJS&FP#|K>G=}-UqS$`{blP!LL)jJho~aGe$tST zw_<&y8h-Hdcv6lD1EF*2o&nS6Wa`5Po^2FRt-i}b58=$r#-$AwkA;gPZZE- zsSyj&IXA-flDQd(&fjupk$nfpA=#P5yi0F>zYo^Is$)VftvVEbujsW`8j@GqWg<@m zaN^czDTG{<&P9G~n=*Wj{>5c+hT861wZo>VvIpQptr~n)_vNy7a0xNP&$!PD|?(E>N!qOy{Y_yP`LE z_?PuPV(Rpc>T5KT8|gMQ(#T0~LN^{owIENjY(**mBH%Y}nglj-E|c=tyKXz{O0Zyl zF#|y~x(1e|S2li{1Lxb!q&YEtB65(nd$#7)thNis#T z^~AXBJx`raL&&xd#lwlaA~)tq-g!UrV_0L3#D@I%(h@dl-9uZ~qxf0=7_AscU`FA$ z(_ry%fTHvTB#T7Gy?vISQ?EwtjrSI{&Z)uD!HHCzv4*9k=v>pKF0%&z1Hxe2c!%iz+F}2A47GFP(497*|S;r8yFy( zDV^_@2k_NuH`IoCb98Z7R6zb2lZpw}juT1CM(TuR#LYzWx&U6~GrVa%$7TMJIIRSJ z>wz}^1VnPbL^1Kqve#2x!77`*KBdf(h8>?K&JjCjyZnd<7(}7s8rsY^H82d@tfs72upwDCn zsZJ^*H=5nU$X{Gr*fsC`dT#;5K3!dyWTP&AID7D-Y`S&Zy{R#qT~B1 zEe6@?EZg`%BiY>PZ)v-#xH;!1{cRUqfABw}wl2foB&DhsND0JI}=z}uXBpS@tetysP5^2pxW zZLD~fHT?tGK+ZFOZue?JqsoNK$R74hh{O;O@dlA|6E=z;HpbzWk}H(vGi-mK;_xi= zzW8|R!#Zu(u;&JPS4ie6fTFX|GdY$IFNm3uDV7Vs#-5wSp_OJ=UOgl0Y*Ex04XFMV z#4Ln+7RoIj&6ef--Kd!NrVZY+k&4x7)iD8GkmCkoPh7N-V3>Bg62_zTsEb{9xtiq7*+@+6P7?*14nl#W&gMZ0r@j~DFx)2^F1VKjaI>GS0OrNU%rw0i4oj^ljipc<; z68L{KH-qa$=mPLk?``3PKvZ}07P;;@qDJhBT*wo>pNIbExcv4I{E+E`C5XoJ^Qwbn zjMB_RW%ud)IAe;9Jxc+Mymy0Rnnerry&gJDHk)Zql{&g$P-$+(G7RLr+StvqV?mAA zR{NrLUwlGWyfFtsx|nqlDO}&0JFwLiL`3djw&v)E6e9W1lC2n?Y^_>&fyxJkarSdC z0G_9!yne)|a^t(%ay@9{LyJU;>VRPB0Oq5Z~A5>Gfmq%+B|(42iwG}Ac=245wf2{v@?@Z}7aXZVx{Z4U@DzlSZgOZf z+_l^M2@iJ;=hvAD%j^;kMh_21(gHj*M=xD#T$Mh%ooF={@k-4j+WJpl2%e`!>tvbS zj*AIHtKlqy;=h5#!1gEkE1>CUnF-vsVBD|P5yAK?3SH07=GULv=OPh#?D4T0@v2rOM-n^Ii zrByJ@YEdNzCiX>#al44}MBJoZ_Qg66m7>8qqS+FllX_PEk_vr*)tV)4Tf`NfY#d8f z1YG|J@qCC!Yo<0d1F`T?W`mib(afRWdv^_&c<>(H_G8G?d;LJ2TZIU^zb^4D1T%{M zLR@~k?A>bnhVh2~5i)e##hXG|Io`KKj=Ri%QiE;cXh*>LI6FT5Rm?KkL?U-_^Svon z>J?A#X4FkW*$8o;I9a#(7{#mkmG%8Md{Wp|It!5_zGkq~iN)I1K|qB3xjBk9E)p zdWPj;@lNF_s)?UJ-cGSSY^xns^&*=Ncep;clS|W_oOu3{p^uFID!OERj*Z2pJI}Q6 z{6Q8Bc4vKm`{(lTjuzd^bFbduHXjW%;;iYKV*Q>@ndGm z51d@mi7{g`j~(j_s#2crm{v4!Pa8cVDa-YK*zdW7b_&Jw)TN994Ys@p(i+1(7X3U6 z>BiWm&k9e*;q{I1jY8ofaCw=%q-}hWeV6o!L9H*;b`bSqkm7w@lfNxZXP3`PRkzt& zezN!+G^FF-pmQJI;b01G-}dw_ciWDC;_618ag?W|;7BND0%OP-RngrR8N(`#A<=t> zxM%uJ*Xb8Bj@c$~E|@@9>x=OVvW^%9)B&X1-6U0D_v1|pg|;W=tVs|{Eez;IJtHTJo?tgDuaJ#yUEPnP4l!&c`; zZ&T-u%G;-G_qA$zgxn3ACPvzn0{P6 zdbj(4^$F`NzX=HxdW5C?^qnwzAnUFOVW1OBb+8R9xLHuZLh-Ut}gO>edk9-9vBMe7GeXIS9>w>yuL|A zeIB#n(N$P1euq_}`c-let6l!yPHx<_%khMcNGS-09zDN{;BL|%pht*S>!qfj1g7VA z019D)aCeLPCk70SS)QOw=sIX;Ryse-%v0*%Ub;2=r#S>Kz8RdAx>g zR=y|g{7r@cf*a4B43+G92?$W2)nX*OJdva<9Ur?Z8wEC^VwE$z{0+8#4MC-~D9}MF zj0P>Gd!A`14<4&;y32!i?hO5}ek-!Y_>w=K6PC_M^4i(-=q0i@Br%L&i6@h*)XYp4 zx7v2T3CiSc0m>~3{$t1UK<-56kk)I^Ngdj&k@kc)FBElbh<@E!wa4=~pq`1BiA2VT z#YlXzFd4UKkJIke;a)~&@|=~4PSOvf3ZdU=1E#SRmkr9f?%fpu85*v;WDbo52hyfv z3Rs2Iw&AZ(Sl6K(*`tswb2&<@SC~373o6xM{&||+-KDVw#(U3Dr1bz(Et{+}@jM+4 z%60ab`eV7ksor>cSA!EXlC4d%T?|g-cIr>icL9muSJBl7ER(gk*@OYHdyR(-A*7VR zp4$Yl05B4CF6Zv-xN~3B1N%3gv7-00Lu`Q?p;>lDYiHbhjgxvVKy#vgBT_!PrLz8j zgq`h=SX*kQ)#yJC8-J@8^T8H#-4c7+{@yUNuy-142SH^o4XRZHMPf%7fZLnz2TUYM zr0F(t9MLnG1u0f*ZRCW%)x_Me&N*h^N6J^mlDKs6bgD4MGC9nr<}pz;V;qgH0y9vq z@rIJjz+U^V^(m_mwzcv5w#Vdww1jekc@Gl9-nkNaK4L|KN4l~Qxx4S^P<#VU+wBFi z$v_-V%>O{6prOZYg;ngp5?VX0%gJRXpo6ZBr{@b{nXOqKZs)Eb%6AFPX=%IZ@1-J- zY_BJm&T67a#S=B$Stuggz{J8gBd=k>1q6&VN>Wt0#Z&O@^uF!fui9%visou+y30qw z0&Dd0?iX=U*VL0;X?Ba>H?NRap4^CrSC?en^&>0t@8uazSR6-W$dx}&>9~g2v1u=J zj|hFlM&mm@ljoV`ueb7CO(`?vC{{yvb{wCdYVd*tGZv{O?m>*s0aR^8S!;J0A6Ai9 zHBfmbQz2k=$4UogW6@j?h7;7->B7Sb&x6v`UU&8-8b7%6>a_9oFIb@Gt0u~#_MUu zCd)QAY~iu2F$3xiQr(3D$TmFv?)uT`eE21sgzYfPeIzv>Ju&#URO-K>P%61#JnwyM zEz9LbIYt`bh2yw+g}BmqxVv#}v;#s8i}9A`fIy1VD+HmqRy^4pBiBQNy=l7RZ1+~9 zTayn(@V}dHRS)%Bn`sU0R{g=L4Pq+Fv2xpFvvNg-bdDx#Z z19Wq&uSgw<;M|OIm>wnEL@5)Ro`yFYevIWs8v+x3d$M!sO;~6rjls#-Ho9WS6-Zuo z@Y7bu8N23-d=QxEzcp)&dUlPqO}|;aPW#uo-8>n%vkuD9u;*bla|!ASV6M;o~Yro`? z^bVj{cOpbMR@H{5>p>JY5)KPp{24kyKx?UNP6WARWB1eXtNf$v!l~4%xG7jKWJ4b= zM6a&U#df<(^g%jl)k3&9K)5&%SDAafL%}#n$6|$ipzUwo{!vPu-x7Ksp;BC9jf<(h z2Y-#>>U3VAZEZ(IE8oFR1beXh#~RO~btI*o)a2#i1?^>taY{I3#Zz&h3qCg4;`MBn z`{ww=J;o<+XZL>tDHY+clm33+25~Yi8MGt#U)Z!r$e0gR%L*lRO@a%lyQ8^LqN!8Y z+s_9ENAmTdmaeiBa-aC3<^4k(|Nn zUs}pHs!_(L?w+LPXuO=o!(xR@-eF zxcRwgD(mNwr$~)i8n+D0p?)L;X@g!TxVOY-+-J-?x@kIVX0=G27Lc&VCap0t3dekX ztpaV~BWfx)avh$%&d=xxRZ5fGwjSQw{CRZm8}^;KKw!9=G_DLR-U;0N^K|wN(tL!h z>cFzEqP{|G$ar+N+7(XIoW2<9aIUa5eH`6@qbgcnTOEHXjLCm;4DxhSyp!p6c5wYg z?tZoE#QyUl48t2FM;84BVLe8K;-oQ9N0xwR67JFZ#9;3kt?rdOaCzF^oZ$#&Mm>2p zH>(dKFgV9Mu-e*c|2h`EV-=Mu2V)nM%>>nIY+e~;G_uE%2DH5eeL|pnAyxW4{aIHW zNEuwRJr!uIUC?rb_x@#b71X2N{+&!$_(2?m=tq*nZx{(V_sjURIDGgShYaz(@|b@U zRY1v>yTYHcJ32C5bn{u5$V{CN16K|$u~X1XjFTeDdnoUezdob^yKiJCYO%Y!Wxg>VYojif{wMB{+h>^YnNPce(F7mzEKnvR%NnF zN9!%h4=$D9G`_yw%VD=!2qWn|-8kHBn^kbC0ntjAx|b-UMG55hT>^H~R55f?NyHZM z@hu>ummPW$;LPW3$s}r-@0#Is{K!bY@$%niaz=}RSw~)PIZOKOG7zGZ8xxO*Jch}` z<*4p+S2A0TlkoFl7CRNEA=Q_sZFYwLQgaMw4~P3|pi|);vo3rWyq~NfiAQjq-mcR- zFUfk3MFka2S#tUxZ8;ifFZ7=`(_|m%%|}WCeFi7Mtso@m1@;@uj5&j(n$Ox-rWY|Q z%&P<8^0q=JZ1bI*9`!do2h_Y%W&&wi_g5V^-N~!GJqn-B z_&0HP+ zhp@DKJbkhKN+(JuzTVBg7d`b+EBiFQX&>)~lx@ubrHuTJ7#qESRRx<#Wmwf7DJsX< zG3fAQFpy(5m9Y?xB-{clsq8lN^*@_u+fUNVU45>D z;a%N?OCNOu?g&qsEYO8!yO8mg_CES2{Mc^dzUYQ@!k)h9;80|L)aR$Ox$RW}t!qS* z_c{>emOkf^Vo35j(M6ot-q>aM#?cclAh@=ET=iAE@v;X(g7adq(x3KU`L6XHN~&@X zS>Zwe5!S3(QxDsm;X15<&Ajq}ygFwUSo*W4ulPB$xY{Ai6FwnwrP6Jyf|dn* zL=DLO`J~QqdK2k57Sf3tzUXcGU&J%L*erf zU8_!swvkjR8B3d{gCEd90}^^*+h=4zcS0axo$GU!zEBUbUBe3Ty_spJHYf>3-FpKj z#Yr?(_GM%VGIH?qy8HX%dqDR$DgaOjiJ-c8dF{o@TG09Y|5&Ns7+lxSjv2UDuRU)} z1fLox@i+W8gw35b_9sx4)A0Ihq{DTKR^r%bM;ZnHltzLpuwGh*nSJv8TTf{okE(8+ zBuP+9t^R41re0W;LxwSJm`3i1N-aNqj-Jo$C*ST^!ZsvpIvzs-CVe z{w;y&5F^#aHc(QuvsHml1 z*2=KNw>ram5~s>mL2!2DUjLwUT*(v(!jAK%jXJK5>smigyS=R=FNDN@hAEObKNfqO z-J+nDtVSD_$LU@1mb}-n3bjxy>1HzwK-+cmw;Uw^3}8VOe3Y`jf$iH7Q|?cDI+~}6 z{TKtC`3wehGA4_AT|Ct1^m6Q$B4&(cdg=7AkhUJ1HcBdG<&alh38+o23%yb_VYbb9L#HCEI z(ALlUE$itFBXd=xGP-{E^$DX@ zXWV*S4<3%V=VxSr)A0{pu_t1h9gi)=qyAS2ua3Z{a!=L+d$a>4gY% z{^O;b9$1r)dTzDFfTT#Upe(lZL!qvC`JYku-nNtuIZCV~yQh5frIl5<=X`UJaL7!0C8m-XHL`Hs%7= zAsUvGXjq1LeCsC+u+Icj(kZFdPsO~4ZyTa0@lBKMolK2Z{Cn;`LY=mYtuVW}a0GBD zc%Ryn^jjmtIdhr>t&9rosVKTot<0Ln7p3|pX27n_FecBJM~4Ccm4!nYr@J-am?mJ?poWW6LbEAw(+` zX-`vo!V$Qp?IDhq?pF+Au3;d?TI6a{nt^tNcv*pb#DhckI4sY z*%UiMogE=m_m|y1&2;rQB9Rg})YfO`+hM8#B)?~)<09iG9Acv+Wd-ZB|HpU{ws~UJ zw$i+ujQb2z4#8#JK7c(vR|2Onn>HGM7VdGHp4NEbtn-HZPEp@<^;&ul8vEh^|D%J= zqEG+dFw26lFidl1V^6xhb0)6w0U-R^Fx;W1FvCD=R^0kmeyRHHT3gI zEYy$Bd-oZz(&R&G0F_;cw5yN6Vmw%by{WE;eVkaHt+Yta`#`Y@$(h0KVlYaNtKSV# zLU|tp8xW})>0v*dV7#lyM_kl*w8LFWYK*+K7SKDEPR3sUujxY)d{)TvfzG-L z<=qOcJ&V03E?^QLXSFd&nwIL`P+=FCvgP;2pR1X$;e(mme5-*`|EP}(e^q~U(D-oP z++%rVED`!WSve^obKF~h%>+92;(Ga(bu=Wf^{I8Bgwm#A7oHwFM${v$8Gti?R5cB&P zP%;c@dxN)hnvi%QLhPVW;SVEECbHOn5F`?Q5CnCF1Vg;;s9ZR*gawC{d>B8EJ=SYa zTYhc5PfqK$Zfn2H%Y=I27oP0$GQ4`)l-p7?8WoVUHd-?O~0@7E6?7xkbo zY@Ent%)ZVO%C7VLM;H_lnaYC<-`>!HiMLoAql^IAV`J9*qgFH9WQr|dk+u~-vxR+Z zoS^xpCc+<9s zb|pLgIWO?)UD|G+*%^78c&$asi-=m)8>pXZnvffK`_z%j`v~bBHP7*0$QchPk{E%+ zG1DD`t&S;Jf3p#)DL#iU&_emk!I^omc2<)w-ER2UXvy+fkTa$q*b^Lfv+M5(m*4F_ zS$_#f25}pvKgojTw^ zy^l_)v;DfO!&VZH;!84zmL?0Ja9ZMmN`wx$aXd_p{%(a0KC|XO{!oh!sM$%&yzxW{ zdnY)kzKgI84h$&!1t#NBLAzpIEkCyAgVUHg4O_d&%Cp8`+LzA3JBvXnJNLQS8Gx$2 z`NviNnO*6yVeo?IGqUMoUQBK46HzMe3H1{;v03f9Kcb%jYrQxV`nT%YcCt zbFn;R@#2IyU0+fD>}W@#4&E92Nns)a+W~0u(vFnfac5Ju9yWXZwjD>!L}Mk0=&=)8 zd~xH0J8QO_<(Q8IZLh|zX1cuMP$IBE?JFeMs zZ`*x{)UHcKh^}JX??rX9*)*EkYn}21E)3EQ&gpzP67P66)iNiY-vtz;V=>1ke*B9- z;j+gSMNKX(M@!LPRnxL%bi{qX*bJ$yNZfVNI`negg^=X;HBy&%UW-K zSIe>l1ol#jbn?sK?R}8&t~em9!~6yx``b^I_Td;VmCCc11RgU-k+3Z;5 zM%H@$NFVGnsMs{coIwU?+8V+-$inQtY~{`sFPHzBu7`|NNF6BvuQ{}wNzvF0m-Td> zl-XUT!GPV$Pw;Zyw-$q$)w-tMAxWDTKYTP7cCdW2axQ2j$$bz|0%G}mLA-kB?O$+} zZ6$wAdi&i4v~OP#p$eByXskf675o^!PtEVkc9On;Mz1Nm>SB z@GM$(e+1G8&w_Mo?8X{|IYCHLnvZ|{x<9@ABPW_@z3ZB)f#Vi#*K8p|3fJov%;HDm zGQO^A?m@wzzP5*$>Ng!iXQ-B+w;Q+o2x9S5uA=P6!DEni0$B=DdV4D^`z4B#juazn zGrs$H|Pm7qfy!y+JSFU z$qCucXTFYif?pth@SX|Ml8*A`^Fs^yXdZ=WOnP7V>30Gf$cwZ!ItuV&#%}wBs^_r4 ztItwl6JxWytQ!RN^3PvK<-Y2Ylf`7~O292$9GW{IVw*)kZo- z+Tz@I{(foRY;Rk%e#p6qQND+(h9rbx3uFoN-MoZU-F4oCby=GmKfA+3k^`|sT#3N3 z357>BIK&Y_-zVV;zO+a5CpDZuM}AN&-Gvbe5j4b6OyaBasEuo#2W7Ic{p>noTW?qq16bX4!`_t1DK z9f9Q>3>oLZ_poY2{(<6OE#HGhxS7(T@lA0Lejo)B65-XpGt z3A6<4jl!V(k88Pb%qtQd`-PFn1!UR7alTIYFJ9=+e)U4+eMou&J85t@+ne?j2?3%T z@3B2`eau3iov*s18SUN3ObJ$1U=2u?bM8D$oyT^Y-kRJM7|d9~qjL~(tqdZosUW1j z$pv4W8F;+yBvDi(=TJW$%0O8X3BaVp{bg$K#ZYq){;!XkW5eq0xuHG>k#g~8N;UQPe!wRYjy>++7`qk-tWaXfOM9#D!Xw) z3HbsqO??4wc}65Un~D3k(%+kzydC3u=z;~JaFFI$s2Xb#JSJv26sE#p%CI}+S#D`! zTZfWsJzbG%=P;*#$OzBQ6~RPSuL0fGw*sE2xu;0RXwT^#rNX9X2vL%bd*Xf>9JeCq zoctdkLiAZoB2lu~Y+4~H^=}}+iP5C-pFqBRI#1D6vkt@=#>{DmzKv{($sV_yZq(x7 zLZ1;^GFCl7O&6R`m(v}1siWVuMjQH0IA_Vu+;jQZirNnjk~eSjkOx2BLt=@v=)xlF z(&J$UkxLSWSYID!-Sc`HSl-hSl1UM%)ob^|?gYg$+T^2FXY}ln{}P$EoidtVSCu);KkIP@Zws1e z+PzX!j@XFzS!0WhIi0l03}2p|Ed09X6pY0)A&7VYOAG-`gIDPnjL|Y z4;^g=X<|Zv-bXx&>gLzFQ}LWKd#cLHC8HbiwY+u0R8G{5*p*#QBonSw^zqI0<2|vG z0op+C%GvnTU;|1pVcfhs1ogNhSRi)T`)D_wg0ID*>CzQaqTbstfP}~=RZvxoFn0w5 zB-)asI{k6mT{uIKIYtmvc&kfE=^g`YGJ#o2gHZ#BE zC-bTWL5XeQ$w3;(xZ1vO{%F9U5~cBn>DAwnwMF5%K!`*l^GP{0DLynrtDGgwpZve8 z@x-{Y3V0{U9sLB!)v#Dz?0I!^5pO*F%YwEq96~V-^(hfGoPf`Wq&p?Xz5`h)SVOS$ z5F3NRXOM*CzVHNWvOoW+oFzmUJAR%doKd*5FK-S{b&fs-|1hH?LYpYz6%9!znxyn_5@8Y0K@mB zcbay^|7@t;8&mD?r_=xH-}~qH|MZ{yi+}51i2wa}{~Y`O?!WjK|3l%w|DV5%|7(8f zzxeO}`9F{U8~);d{P#~+sJO@PzdSKZ{pa{!@@xE#|KRVp#0h`LznT8e$^Y{2{I~yU z=U=w}%QQLvb5Z=i{{MCE>dH} z{`3A-I-GcV`1NE1>##Y?2!Qt#b+)AOTwQH!4SlsgZ`@Oyy_Fx?De#idw75_v* zCs~upq*H6;i{XKIHgxI5n0lGAH$O#^WGp4075?qGBY&wsM}>kblCD6w#x}QfM&e5< z!%1>oBG`;coi*A42@_C99i+NuuXitwZZfMEoCx)bmwkqYZ<*PT^ltumX;F>6<+*5K ziewY5-)kg^1685^4Xd@Vwcf>nuEjH6mBw8ZVPEprN1dsWoH67NaS98z$DF~cgaDpf z$~g7ELKrrmN~y@_vjuhN00*lRW?#*u2M{mXlO4bhpXTvMC!t5VdRe5vy-cRp>qZjM z_Jf2-nvtItzZa848xiMQoj4V;&VEZRSGswTS6! zEgn#QY@H5dCJ8N25F0K1`#3RI)Pp#d8T{UmGtnqGEga%}$ZORgC15-I`vjC)QM_LW z?DM@7{m(+N9w6&GLY$e7SY~j#uVw)BKS!kQ(R-sGTf*o)7RT@0lRt6*FwMy;gKJi( zhmRu`CM3V!*RwNyJ_Jz&Tp4r!W>;_RCXzz^%9RF9Pk*fV<$8Jh{(4;+^_<|Y(LnYf zvY?|{P^y{T0k2p&(KI@ok~YUn;&>af2UD~?ujFOfU&uFtupbw8)raLZMSQH9yL;&m zji_w^GcKRe(6Y))S3B{7spj#KrF6`D-jCFEL<67&&)+o@jWdKgOdO@3VZ&}jdNG_# z>Z;YkUq6>ynqf|q8oa7TdZ{_ht}g~7PRz`eQxG!x*EpRbQ#Ew`nP{zoX`)salxLB% z*$|Qy$Fb@uiX#4gTNjff-+Ieon7ER{@K=k0iYcoh?O`XOye+n!Y}|zae*07pr!twF z@9CB^m)}8?YCf+QQ`ANGy$En|H=~s&a3I#?NQOfguh)m0%J_P_|1fO1n|e8~9RmZx zMct9zQZ`hH)M8`*&pGl zk10BZ(`&LBu7`q)L~%-;oiyr2BPUaI*X8kFPA`1Kd8HRc$;0y6o%>=XPGR_Tn_-C- z1KqD!q6mA#+3qp`w=ipNgcY{+o>u*GSg3tfyB$e(xkHqoJX;J2Te9vy_BeojM_b;G zoN2b~i%TU*I}LAH=)rqERg}6O;C%DV9vzSgCzr;3S7Hgs4Po|tx_oW#I(TnT_IxEi zW*W;3fBn_$6Bf}-0aU~w=_l(Ry*6Mdz68s7eFDWAG*7?rMs#`rW+Ekos{mM7Jnk9fvF9guU6doMKO6WUe< z#&^Blc2u=z{e8-?jEwFaNfe4QVl)=x-++1gDO%S17Ou&20iA1__wgNZu>d69{V9n> zj_j0#=%{EI%XBhPl;PjGz*`{`woIxfkw&sHx8Z^26tl<6i{f_EPA7R1-|OCuhF|h- z0*BRB6GcdzWqjLtiL+LPILAy|vMxzI;j z%J9eo)NF$j`|w_#;K%{|&XW#JehZn4eiox85+udb_gv5ZMqxi5>?M}dTe{k=JVkp= zGNtiaUZc%{XQh2#ZEpurxru;}voU%?vPlHI#S&$hJ1x=APBsz)UAQvu$tdoUXQB9G z1%f|sLecGT{VfehV@xFPJ3{WA{V6`51W58u%eeVzx#2qmymqXc9~TcVS50*0Xgr=Zv~kY2hs7Io{bY2VVA!r@XN zG5&)%kMwBp1^#A$KE7@))7^+WAMSE5Kk6k@B%-|h1^LnJerp=VK0^At*Z&0Lp3tB+ zk0)zls65}T7Txbkq)^!HA8s)(rVuzCpt8{GA6~3v;EC0;?4CHw+ov;E^Thz2yR|58 z7Fa;QmRI_} zgCPz6)0-WBleEaktSP9|ird!k1~hVSoBB<=G7afd!hA1cPL?GTf5GUAFdb}s{5VfS zSv&waX$p5xNZN#TJ834tVA%qfNg^1z33HXLydrsz*~1}kI=BeLDcM54egz_cwEWbymM24mc-^=Wdz;ao ztM}=1PDgXKZ1?)R>A&eN!GMf3(Q!`C4IA`4`3v*gdUM29-~Nf1_rI^$ruZv(F0!(r z`-oB=Gz&9ke=*Q?f>jXVHgu-Uui^g5nESc`=654iohjs89IdrZ^x+2n4h+7Xz!swF zJ)QnhmT||5d^C_6+erCakFMs2-PoQ&l@e1F-KE%5g!b(fkQT}~`?KqQo%w+j;VOY&y9+{^pb0sJv?% zdDfp={jJ)91j>~$JwDpsQ}(fD&TpSsuZ_%#=j%U=*I;80Sd+m8a&cXkdHq#`N5Utjp}c|rIW)HPnFa93{$mX z?DKvry40I-xW9FZ(5#-0KYbX#d$5>y>Xj1AyUJ~52S!OReB3ko_bRzlmn+}*C!jHx z$7iihEhSN2oz}rW-&E7a2{7XxWuQcV)Mn`41SXFd4Bql(ZW9*ockfh2QT=MbIQuT` zZm+pwV3Kcwd%rFW)s=tNy>b%ABd=H&x*c?E6`Jj4p`_k<8r{!=39v_2~{fv_MWHR zh}4pyAJ0F>EKeSRVQ=(K;=%cdab&|8Gs-GSlJ~!eUz&Gd;XXP!M?X^ByHfB$Zq z`G!zl_A=+D%kkYDNZ(wkJ#Xsm!jfB2y0AL^i0-CZ5v}#TdwKbp`?ck{=r{_7OSfNa zeLExEIs-*QY2CHk!)o^o?R|AAjN#{?oMW$>9}Y#{^cr@D(_&w1jg9_Et>yGnz@S~o zec2tMNmsh7yNta~bxlw~H|4D-WwlQ^M-xU`BgqzE*F6B}<(iUiMP9nwwtn%)WL)&x zes|TEW`a1qPJVmUj3F(fd4DkRG=@&{w1dxL2juXydtmVD@ewTVOJ<>F-V+O&1v9jo zIr`fZCa#r%Irbs|f%au>NM^H?nQ#QxhwePk<-E(;Lnm!QRNd=RA1 z$@-R#kIBN4n%n9Au5`VH6?p?$Mv8L%euXc(MPs?Q;qUYb?K#n;l3jgTU5&$G#F0Mr zriXXgQobZPZfk=Cr+$3@b?zP+Q}*B=jHx=Nr+a2Zc=u5`N1sRPh#ryf0Z)*g(9@;HifX$9Aa z7}JlehrE`!TTjl%Oh=xmlyBtPR#n#?KDjr3b~B1BP^FkOizr^IA}xkL+>~17o{+-w zwa*SNE1f`P$#5Oo*kd~w$?BUOxavtZPhoF#b^K#>b-zQQp?fjp#`PWMU{ItoL?&cn z_>Zn$udj}l^H*y%d-smLaj(HL)B1fsShU?ye9g>HyI@lL+@0K6D9=Oe&Gq!P+*hXB z6&+72Mx8*<1#j^@OVqZoyWb|qMp@XM;=88jqm0Ho(|R@H`FEcC z^J{hYmEOhJw3?0SQX6HBr&|5mL^SViy@9oAH1sXC0M@ztKh8+{f;_q2{^ogB{}+-p1uJo<+fUvh!X>VPx+Pq^`5; zY0pFYBb|8U&(B1CHfPEV~)A^e!7d^N}A_tRY_;M z>kw*_pF1nQ1!>r7KI-dU4kMhzYQ=Y7cc8=xvU0B;|kZz>I z8RD7Nb$-)U);I?vvr!rGZCmm_iDZ}Ag0D$_Y^@~JQu6ff@~~lho98*{cfWI050@i; z`0Mn1^~nIwWk>IK_Gs!tR~r|FRR;O2aP{Zi4+{Glmt9N(tyU83*4I~uxxvPyG;_w! z;Ss7?h8X6c;sDG=4d0~9jQf(>i`lMF_rPzM5RvUk*q}Pi> z$`rl-q9hCs9dp=j#g}w4OUy0CyP;L)PYaDrrb}aDA6%7u_N6XYJTbQ(ak_i;*D{85 zg+x|%SI*e#jP}bJWl-a5wSqZA5cjTK(?fIqo?fLSSY<;JS?1lxJWzB=3CJBd%jtP? z-ro!2^zwnZqr@6LifI{uKRchrltWz}Uk~}>?}W0@ikr*UeIXV|aSJ}h&yB0k#LDYk z;XYd$(bDO7HtXE{#d%EW4^EiH4M0(s0VUFI1!R72mVpm3ODHl4ag#*iI&mU$j$*-1)vboBQ_o zy^JRN*Uf2(pVQyml#yz5L6n;BMj-PeAF<^b6YOHqds@5aZfDLW=Zl2fWWHa>VXN`% zIPIl$PU_O;QgI{dZMq)Al&Dk8k~|i?!D6 zFybdTznggc=E1;!w>!J(wz(Bj25C|5z(EgZrCOIKf3}{NTDsiuo_&7OfBliF`?i$1t?Vx8U&$GK(pb{U9l}4_`0ZZ_Pu-;=d0svIpa3 z*Oc<++#Jr>^AWui#CEs+xmTv~+PeMOMuG+B7>XUgv#=wc-Kbfv?C~M|jOYE^yWQKH zJIIUI<=NeQxU+(44v?k&MqzqtdGqww93EsSX7NS{;5mdY;B-Jxo8rm+=Idp$mhM$k z=g40Mz0xJ9!_vCdcy%A2miu7xoB_>x3^#ieCt4vt!Fr*rh;O)d`3!qU)&-kf$ki7~ z-eSMbP1%#}dx{2zQIS0Otz)xl>_4U3t9OEjw7I0MuXvzVm+ND&0bw3&mDWAai&wy( z?ylNta_C*%IyF}Jt*Ae<9;H1D8wN4{TJ2=7E$SvB&F-CjII!2PPgJ$<;A+mty<88X z?Z{hy+pEQEsLyYsU*8{93ki=#uP={Xc^Bwrvb*zCX=7)f1hZ@N6%+bNm~6+mG0^X4 zjOL4Q(A^K}qCI-8JIm)4W}Q+`$bku@ZEjnK%)7}+7+g^~zUK;BoA$_4ci!n(`Mno& z0PVeZdBy8{{hOV84B|%ts(>(YROnAc4)?MpMW{Qs#UETH|9EnbzNGi&;ozE;UbqRq zSM&2u^KpH1!drEcb8P%_hcVrYIIO8+JB)@HGHC(ijr|HkKyBX9a+1K>6XQHNX3E)L z{FrpD^G`$|w5i>#ZsrYJtp=$sr*%}ir($`X!^_#fPk|(Vzu%r_K3_yv$dn@6XZJ(8 zkT0T-?hh$!|g5s2ZbNrC#b}`cZq= zJja0Bl081fYrTsbg)y2!{i+WF@WYt4bw8zzj_>q3OEzFj{c%!>KU=`B`7!PDI^~ngt&zvCb_AkLZ!wR|DpKl~aCZbc?MM{@r_obx~ z5}W(#(~gH{mIlgryJ@-l$9(UTar@a02ZLkyjaza;s#Rl394|t?ae$^oE!k#-YfN7l zD$V<2>*lEH{YLA2+CAIXzT3Hl#KHHfA&d2lgKz<6yqrp_wz}C3^vUKJcsElsH%$g% zFE5(6CQi^j-`87&Rm$hHws-TWIGm@$0pl4ApTD5J3U2q8(h#N8_cf1)%fX&)7e5c< z^iZL{-;ai|O}ffrU#XK2ycM^4AT@V^5&}ADrpI5p*#0$pBc~@Iz1Gp|XwbFmreQ=1KJ+RPtE}M5CBH58YoOKIY_?iUSyZ_Zz z+pRsFdg19{LgCs=aT5PuOr6Pss$ANvFUSD`(hXHQ0R`!Grhp>7bP5x1{SR_iojT*M zDs!j8T5l8a1hSBjdsq8cm7!ZFD79IlFQQ~m$(7xmuIx=<-@J26c1f_>TMAMNlHpiE zI(@)^^z*({R6oK5`HR1WS{RM-w6DzCboV!pj51QaJ^aG(?6PYVr#hO%I{WJdW3d$= za1&sVrT6F+wd#OYSwSa^=N{nCqo6rn36!DynzgyH!CJ&?XBu&0^WNKLn5UQC1@wl( zd(G``QUq5^9A7?%Mjl)LyDM!e!`(XLF?Eq@usPtX zHwk${hSK_E%u)BI$nEO-!UAO?y5^oiRrv0zYZ3JNaF@pZnI4$5!(b7|9Y^7ZW{4Dc zlO@$UBd8S3%t$F#s1@{9(BG<0;~rv1YC=w&!NK!sS(X`bLVqsz@E5nzISXG^2$J!^ zmKaEVuBSz;IY)}P`e3vSa~EW$Tpe9)#-%5A0&vsvRMb*P84H>D*Pl^(%HCwn>mINgIh|xmF zV+zE+&_O;0pE4L?!ac9a&a>wXa?UM}CW`z0bZ-2sxA%c|C;4Gz)67z+tVJ(XF__^8|JZOU-ZjLC8@#K4qK~~# z2gr8HZqFGivqr(6B^6kVZb3Hz-$?yIW^NbD&W91SsN>6Ji#dQFsP9o{`2M2Kj!J8R zyVe&Fjn}}mzbzz{%h)+@-}}vlloir`Qi73pu%W7kP|pwZ7@;c%hByGKuC38tWF_f@?4VOS9NDgNrfpx{0OIyoLF$@8_3}FYt>_w$dY~lktk$!P zewu;}ltLDezCa#)VmekE;fOi@bVxfD*6>rl4t};KmGhlG?i5 z^(Tt-r8(>(V^227tL*w(DUL+se45Jbwy!+1j(}#Gr<&U&PeP<)1_|viNq$4cLS(NO zVjLVKQ+O;QyikO`cq|R5F7{id%+Z7`vYSRKUCMy>>vI#gJ~BcOuw~}R>>VG*&jW&c z)PN~W#))bA@oy9wVGc7zT3FgQFRvj~CnEpU$7SQnlg|uHR}X6^|E|D?f~WnqeeNi~ zx*dg-*?V|M&XHirQncazHpOsrXVSg_X8j%_ku}%6s z?w;bZN)tyw=guq0feVw&0;08BeYD$`qcfC|tM>~LS;_hVrl|na@cn{vG zv}6ljomhhnhid6eW%`GFouI)x@lWfMpTrx0zB~u1%E<`hANMnI1&WbTyBX4`z`*Yo_vsJbV}RguAia4OXW3>;k+-`Raa zXNMUAm!dZ{0!ezLY^>;Jlufd z0DxLq;u5D|7)K9Db=w&(kIQd^4X3T7Zyh=ihH74mJdPc0)}0=%cfTYHWgdZsk|n}b zXFlo+{h3ae70R5+Z-)6XW3}>H?v5%K{{0apr*94s`sgFWj*K_zLdfQLYBx0_R94=X zDEOvQ{{V0QAb!U8`{mEdw-+JQg#NQCPMc3+U7_6Fo|!qlQ{z5)%X#$yq71B3Xm=dg zRCfq(B>`H$H4-+`rxG?*1#KAxm4uE&oT*$9SjwcA|o? zAH`BcaRvP8km&1&$){#qZB#CZKfteM>7d=qT^;88PDw`T4^VHv2$F9$j8r>UKaPL5IUoT7s8@a0w z@4M#^+cxI!9m`8{-hU`)M$UMx!tcA_U*zOc^fILKeF!f!Hwqa4ZFk9Uy1UNNDIHjq z9|IpV9C9mXd2-W!x4gf7CQF_%f7K6fj%E%)^w01NB4>DYb9Wu&htU3 z_H2SiA-=Qz3*s%lnf>((oNjdV>;Ptu)LSEO&N6^kk+7uA%OCW%nAC#}q%IoRsGV~X zPbPO49;A6M?9yvCig-TiI)A4g&#=A2ypFdETV)&bG z^=?@e3^)D-{o?mBA>JsPqUi`iPV>=tp)W$gB#5+`&rGN~`+ZGi zbC(|?YDiuBT&Nnd1_nvX#OXF27MlHvkf6QV)i%Bma}oGw-shZuV#t>xCWqdOFhWil zg-4^y80^3~rh+PS(r-MAU_|M(dIX-%PM~&W+z{&=OQvY^BX!ozwLf%J9w(W;_dq7O$n$L3p{x7Btn zFNQ_9>ihn@9GPU^AcES-kWHE`XuFSbhMvP6$9|}?Ijp*We?`{I`Q}Z;%lCVNcf%b) z18v6REmbuy`Y`<^lVbHGY`wVMq0LDbgqg*9@urQg7936&tZz03P}beX6tmB7Tu%G< zZub$y_IlpZErZb*tw%%Z|)&=f&%d?*=}b-C<(&Jj6P@x)?!}d5}v_B9%IYgKWBKyaD9X7>)=N85{Lo1 zR>RA4%+LwjQsbR>7nUWN_v-STvdpL<-of><0Gct=tGCM#31H3J>_R3gr7ZU-kAyN9 zsu_-l4Q9l*%MMZWc`ehz8WZKhi1I6Y{uryB3c7~4JuSWDZ|bpf8Of#iL_Facd+!sD z)C}<^%@?E=o;R=6r#y}y z6L>Xp^vwc`AAtET%<+m%p~+pQOQc=6euK+^!~PR($VDeb*s(Ey5T*=I=)Kto_iAGf ztfNY6>MaY$P2GMnAe(mCDhVPtEK>|5Ne>FWTT$OpTf?ajC1bMVN;>^Ui8d9H4R#o*cq1_gq1h-K7XqdkG5PutCV8=Dcm- zLHv#_Nv%?MS6J-*7Z?O1-t{^tF?*1i`+z|dV7i{(EYoh^) z*ll*g`uW?g=^p$}>{2if6{b5=0gv#wbyP$z8lDa6t^-P^Ezg-q=TkKm0Z-V39NDU| zi~BhGE2qz~s&Ah`ck(`UJEsm3n*HJ%H{7Ki$ZPIdVTj|yFs2m-P_M%krappD4P)ka zsK3Zq!SoT^?P$2PN{*rP8P1ROaiT8EszrMl6f9*55ank~Ge&qFc=gx~AkAl9U_-h8(5p2!`;ym`G_+VO;~T(}3MQ}d3q+8aCJ5dE$jkq>v@9Jluo zV_zUAtu=&(`oh_3QGfuDBEODE5_I7wpm5fg(M#6&y#WH|Iq$OQX)82+^YcV!OPBCI zgYA0{P60D7;3si||ipB&^#S8_+Rj@kgE9 zR1@!&T7!2B=w^QI{qtE4g#bEl)uXrl`MHgk@4QQA>Y9f# z;=G8{5uQ(-86YD+-e%=|sIhstJddN&;o5SLFx1W@wpeQC*|$F$@UeUx1w^d@aZ;~y z4~m@has9r}H6UNY$fl=y)yTl%B2~MQW*tXMW#qfpIzv&X9!J1w(u0s`5LE6HsR7qt=;h_CRDxo(jC6yB9%6LCh;Il^>2tz1pvIq4^3F$;pzkUHFecB zTtqp}WTOhoW1r*FguKB^A}gF3WBpaAqtrwmfYIkyT|f^ZiV(%6{%zi4&FF@wO#HeB zWamYodf1~c{FX#P0RsdO31SK`evDtJz)*R&m%eXr?;*W0%{QlsdPZKtVw8eJ!-FW( zQb!?Y>e=FfNsen8#>r1>+;;Hy87NH3l3J}8YaAKqT%RJqp9q}7k9{p`-04EDKkq~S zIS-&3tp)wEGSh$p%nS2c+~2wn@w=C4P8kx=Ivm{dQ=Lzb>yGqS)vM```)hMoMPXHd z9ERXctT)1Sl2J=Mf;@XvqNM8eOf%Su2E9e^4Mw)-?ezk!0K4Y3Tgu#G?Zn<@9-C_T1UNHW(l4a`y>2bSU$i6XfxnB zDZC&s(n40OQqvx>{n~H(8yoU*%@+}QytP+J;q~v`F?AhT^A23rA=Cw7wOC=dSSyAfimhoyggk1`grlq41V)=?(be+>a<@S zCROKLVUr&EYh(ztJvZ+V`Pp{K|De+tBkutNO1)f{jh#6bK%8b-Y^4S=u87VVtY0Z< zD(&a@k~&=cU3qXbOqyBkAV~9bZ3Yn0q_MOJwPv%S`oC7*M6l#*3%#$AKX!y>?@G{t z6j_~Jh)ARMJ9R0%WWRssa{@@ExNvJQla=Ndu%(((01?gGfu`@CdpS%1@4O|_Uo&Ao zM%<9!!RNoabJ@P2*}7a^`eIH!<5x{(g3W$E{xzhHeg;*$_96v&Nl(h5(WGdfBDrb> zcZ{m;XI6!Je)|ajvmF{5EGc$2RJElw z(B=yDeR;UT*(dA`qdhTmb)K^1{$d{6qGGW_TE_F$KFT)^2f)terA0@c_n6=AIh)w` zKC(-J)d7y*C)C&Gk+^iu5eDAN_9-E3Jk#E!oc7PJ@lrhCElAA1^JiyIMB)jkVAqX} zsi5I*ntYj>N|>6SIm$dexebXu$|3CN%sIL+EM$ibelB(Ya`(q0*M0~e1*WfgRKq~$ z7@uADP>xH9a+x0RRF(Q))7U-=1)*sTZ!f3$K}Bwt39Rt#gJ`+}OMRZ_+o(D+;#zLC6|J_(9F0 zD~aZ=(IUMI+19UE-0ZV{brY{|>$Bd4z9}AdOlo<=S3nLmtk2C4B2@Q+>SW2PqPl%*K3*O zKaYzyUe)og1I>#G^Dr9*8|x7^8Bky&3dumG-}x+q<^D9N{&eHle0<*2%_p0$neqg& zQTp0r^E4fQNtv+tz#A#Z$Wxv~cuLKzJH9!U&H!+CTDn#M*(dXLoGF-i;x)ok*h z*Ut)-LV+C;6uYE$R15}v7Qg>tjt#v7NO@GfrsI5R6h#y6>?08pFBh7OkPL8QR~S(} zBn9Phkx7?q|GgitkXRb);^f}^OP3hO$}ChCl;0lFVEY4{OJVyB_gdzm3FlPCE*X5i zujDU>dDM2>BFbzM^lDpUW=gpxWdb)yITsN6n zX|MG&sDI(wO9b-ZrEwS{sp=zO7clE_E81q`du5kO_KsjCnDxg`cmb`_Rgnh!hf--KP#WJq0Tmx7AeF^#c zA&W}UyTk4tuMpNPGRONP?)|eO$cCRHc_`+#?G2sGU9QlBr2*$0rAM}?p&DEm=Z6ET z$hdp*$u1X1*niL>c?(7UbtcY;>ybQz+(b`a+AeMfm>p<=7V7E@9bme3C8Wy3YBBGT zYT<-V7PwSvEK0db3~Xxq?nXX0pt??TlT!$SvLfwh)s3&17!(6WpuF z;X9vztIIH7&RHT$s;AdUC>fu!Wp-E95az-wx5use7P3JSZo{2>FjjoRHOUVRK+#lt z+l9$(uCa)YkC3$@5uyWqXwmN~z^|Xc2FUXUM*`7J$lp>5UF`+`K=*|wOQ_F^NIjK z@U!$4p&lmWXlRN0^p-(;>VlE5eP_o|Anb|M#_n<5(1lgMKGS2!dJaf`XPY*+Am74w zjCC{&3KXow_{g0tF!XpHcU!QEli$VN!o&GGB{Wn*|jb`fvZ2f=?3D<wa1>CZQ=xe7!#eEA)DrM_{QJ--l4h5>2gD= z7@cCR<6#M9?!Djp5gYUBAtNZhTza_(AGJy#&2ti`FRycW57xn7iZekIUn(rMEMJnXFF>Q1K`O z7^X86sI|W&9Y3EDOYYBUSOR%&&!-=+%(A*(1m4RPvr9tl7;O?Wc*Aag?$-~cxr^Nm zNMi1^hiIR%I6AUXRg4?YqleuYASHb*a9ZgMbh=vX`I^@SYwpXO5hpo z5LR02nuw%{@%uSP?y@0aG#`HE6U~vzvYHv6BbY3Jb%>dk>Ef8S?p9;UItn7e< zM#sEDr+Fx4cOI<>z78|HeW)^(k6%=z1|Z`C8Vq732$Gb7x~Rf3ij?6p#Ii59T9}35 zU_PtB0M_Y4Ac>JLvuj(x^>;f#6Yet81v|y)Dd9+B`quOzRNkfbZ}VC6d**JNhQV)u zo^s#<(GY|&v8xZ%Ew+;|`|H72Z;pNj`Hnq3%DZ-Z#srv(vN5D>-T<1r1r%CDj{ zeDL4GT0swKUGE9%r33a?i`^?^+@K*+)pYX_z2QR-vo-6 zAXXo+A5w=@m+1!J0i9}Kuq<29 zm0>y{+c>UerOAk`TVqr@xgey8knuh)FCZzO*yw^CxF1Y2)d z@3M8O#WY|m#902mba-t~LO*ci{SH|Cw|NWUX;o;~OqM+tVD8rkRL8Sa;}yu>UwjLI zi)4;-nMg@)^gDMrzfE!4aq!{1;h#600jn2~Km|B7KkS$5a7g=9=Is!sGn7X7iity! zH^zQ~m-ENfwlGk$i{Ncgp2vmu+!zroI4^$a3V%QHEoHF{+W@I_vJO7_x#XQkdDD*NvJ zu1HIMb6)d2u{}GT4Z88JykT{#`4z6-5myG0Gb=A99WTMDs8t!d-h@ajAd((46q6Lc ztrs3R0<+9Zl0!5n>W--2I+6(w?+>Ex9z5M5HT)<;)(gKFjEnP=w@PjY zeco091oWnH9gS@Y1yL>-i%Wh9W=Tioy!E;&t}-ZlRb?HW4j5ie#SiP(uL8vq+|1(y z8}6+8HaE@)+r*95?#$yjC%|6_GGc@Pz>Z`Ww041gz!`xM1KF({ySQ06Z%BF zD026$YO7}VeaAyq1Sc|vpWB;wCSHq5LIhE|r{!1rL-+}*>Kq(Z@^T~|1d2PTmJhB9 z8bG_|(Ezmwd!yd^?v+emRNm+10@@;yh&t^NJncCf^0>~rX8Q96?Bl1jcmp8gq}&Xp z^!ga%ERH-g)jVIap_)KQ@`Am;EgrM|kG~xgd)Oj6a&aH<@Nyh~u;KSeSd;bj_qL|WHAr@2e;BWU3sLD|| zLDGFZvz1KmNJQL?qud66jq1{JqqTf(@1jl3`-yz*beY+5gD*(Fe9omQSuGKlAm-In zi$HOVxf^e~c76R*G@x|CgJuC~SS_%CVo%=X2A%+xi3bAaWbU-SANBqDb#!@`_v`k` ztw?w4G(0^`hrDs3`9y888m!VSk_>RG1(#{-hn5kFAdhkoGu>#!PSxuF+j=fbPoV90 zXW<9%a{@%DQJj{KeW6Rj8(E5rg8D2%GIxc9AWaiz6k6)!fI+x|6t_RW*E;MD>UCE( z-|QSzVDndJcSf3msRly8t|L-_demm_n5tIU>lRn0(*%O4bKL4`_wOk>Zb47_QRnx^ z!*R8P1BF_V4|K!W{K^N;5SeIKy9IUfPMB~E;i=9*^?!bX>Oo=Ntk9$qj6PoFDmsw; zxqBDH$RP#=AyK8V>3rh?Oz!ut6%P_(m&3S5%I@KLg*k<-(5D~Ug$42h`y^#6VcOC3 zAS5doHK-Pg{SE?5I?*!wYPuCq3!h{1)0|?sL@A2yuJf_~~-d z$5lT**`Y2cpE{`KTeBW8UWE&v!`GL5Ru@&H7JCGX3(wnwYKpAzetUU$aWcQGfK|Rv zj8l*aYZE3P^U$dwAYg@78JOLNe~l4w_yuplIGqlkw;;R`-T?c!bwsF(3_f>4H?WnE zP{8D`ScRMA;4th32{J>7rMm`5&H0UBq}VMMj&m3MzTW3$ILFP@3C3#Z`(fp4+J2WT zeP7ltavSyy>~}cRME)&W4+uljx}KI0tpI7$8?o62h^P0=S&ixZeo}`{xf*vWxcHLV+HR*9ho7>Wx=#f5XGIX{N`TS$6s; z)DN5+eHSRC3XDz>Sh;z(>>9q4L`E=5p*pJ32t*0PCpH zpt$|+->9eP@7pK$FtkGd(kbVfomPDzFJe^;?*BXE9k<64@|=`LRDmDkU(0d%8Ar$S zy*ymuzHzNE#km2=<7+ysgKT+S6&(B46>o*csVxJ#8x3fxb-D(L^Y~WhCpW|B>%Xg> ze}=~K{hO7oK9k3ycwIX@X)|V7-Gj%I-~gYkcL;fu;u%3t%6Ia+%mTUK#8m9g{Zia6 z^TCMCFh2LQnPXqDX`ZD~)$RsnpKJQ4wMTco1I3{{o5dpKPrNu{)UY&~)D`Tb3L3}n z>FYq#aeQ-bkNhZ zMU)DoTU>sruY(CK3+(GjApL*_3wOHkuW+&lzB`&56OhsCxov+MQ0lvZS+^Iev+9kw z1GwQ3)zNNCa>eH$3k%1MhXP!kaSxvWo#VSgbkr$-Jy!M^=k$Eaz#9Wg)oIYNJ|3IZ z0wdu#_-Li(y8+dCNa>?sZZ}&UGvO-ZP+n~6+Xm%J_ETv=(VWa8ju(orKRXDGA$VtM z$}Lu)Yexo$`Am6KM{4_qE#iio3D~_ooxLi|<1r0?2O%gw%^GNfVqNS@S76x3Ty|5Y z-_%$CUV|ZnZ;fS;1Hc}#Qa5i=0EKC?!!id{YihMDA31qD%|Z zN8}CkgE^G_Q@4Lmn;qZX3I)mlSv8<)TCZ656HkV7My5`D!`O#5fd4tmb*g1SXudBX zDl~tz9T&VRfaggKEl&Dz_Z6om6Z%C9_k2QkcHMSfP=2w9_; zp|CsDfBLQ0C)cENlJg^W*oG5?#Ixz!Vf{XL?s#KXj8df*g#6D-r@an{`P;Rl>yKnO zxn0LDAAQGwc^!$u0=W6Tj@A}J&^w}JvX9ksNqj&5J>}5jUp89^47(X@mtzy$5Bb6f z-1K01elXl5EK!_=921(C8GQIy8ztGon+}RB)hll|FHZ-${^~#!&Rb&hibd(T71w4) ziWpd^6Y0O}?1$|qK>i`UCQAe@&f%&eZ)5m{Y-Sq=Eh%g_$CnJZ5jw!}fmw^Va2@c_ ztRA(#eJ*T=Gk8RXyS*^GDy#POOKbA@yRp3Z1(I5yUcmcGDbdbMFnLzj>JFB@)kv1A zkRglti+K*kepT%Y*DBGRr7L9PUQ49Fz%}(4LbThmb9f{}Gat`4@wuREN?*k`ZdtQn zJ%)7kBsB&g8A-j50uxHuhFGp`H3DYpl-v|t+hdckJp6prU$<;p>d3FaZcZ-0lUdAw z5Z6uUZGcrdF(-DY9xtn8<*#aN&Y010jH$m)FBlb12PA|Y2L25=tvoVrc5c_cD85y_ z1gg(k5+BZL5ZKEAae+bXA6j@xc?c1BwUmZibvi**502QrEe<`ky{~`UMQYDOe&zNb z4grnrar7E?2XD(j zxG=hU4ZPvHJC^$S1uefWman`JouzNY}A zw0&V#k1haT==cH1Zs7u6Ee@4EB9Z;oct-REll``{&3t6e-Nc_0-$ao8BxGF=Zq&aO zVK?)wFh8LY`e!)l{mXjq)0ti&(hHUUtb)5dZ|x_P00x465cIQWC6Eo8PZCcPGP$6r zBt|wnEj0!ZEa%mi)zFHt&(ufrTJ=BYz3~-!#_k3Jl+3q5IU#pC;LbAu+K}-qjz|wN za@HbJT*IAxI*`-gfw;h&5a7ygp?v6Drzwd+r=E^fEi3V9^!j%^-(c`bw#WNh_{I;1 zZ!57OpU+JkRL0zlg?+w~Kl2M~(D7f)Oyb}{0e!Na!BK`wZhb3U1vit~m0f@&E)kZgJQ@#@x*q!S zK8o>So)0B1Jn)}RdWw*nYMkU^_@1~jg&i_yrOWoQ|HM>s6v%CKSB*-%Xjjbi68KDnqlx%d1Mg6{Z5!m6Khr=gd6QDcy8AE zBBu+vva1QshMP61oU{nt0CbH~);7`S6iDAGt#US)qPzX42i1m(?pDt!T)^LW{;~DB z%=-1O$iR|7U6gCJy}9?gIB!Cet}|Q!WWBWjBf67!reUt+O~ZkKb0by1EcVM3T+j=C zL6N*DqKFf*OMviW%^uOZX0tXI>Ba*`v$zN6Ti4Rh`U^fqN6#<=44@LUjF8=E6oOQ5 z`FFbmW}%)s^O`!l3rxyPs^HEM&NxCW^q3vk++<4-3iHFP$KG8(FLa<;&XgY99pi#n zB@jJ68jC9043Q$I62X9**@Wf-taM54+>8@>*U1sz8MrWS>|~O?{ndDY+~L^5M}M4z zX8YW>svn3t@!bu6JRvl)+}<7F`xy@CcJMEVBqcE6APi9?PG`_8oBP*BSsg#~^Z6m| zcCZ*3GU=8zLeiQi@LR;%YPCI<_w*lb&r(ZDlPn##@%cy&;C^5_;%zKPR$YB{S-Fkf zx}_?x`$*Dis4(3zhLnfRUhNS6zKvW!HwLeolAfmy902ABNT_ng#2vbdx9jV=kD`-t zzLV%~AU@YXIt{=ZGuynLkJCw4b_-BGQ47=KyJnlS6j5t$k3J+%azH;Qd0_ z2Jj48M*zz39e-{Kn(p3xWcT-I*bo9df%PTQx8mj)AY;P)4(HV>N%Dq`-M++p0Hm~M zy|OX^LZV>7mE!E@pWnI7yL=FF^Q<4k`<7s@wnYdGM?E*dq%9Qk4fwyedSZ}f;OS8H zQAniuS&GC(v6`|E;jC8MUj4prGxRuqA5-M57_-Y2JMSA*)bbIzbkt|9EHN<3D%#(9 zqJiz~RZtC6oCTU#bJ{BcXUejiH3T`-<6P@1R5j}tGB~xEk5ZEgiVT41vhXq#Y+`)d~Ug@dIwn^{Ua92KK(5km{*{Gm6Z# zShDf204R;TJIGBv07TnIBDQi>ezl5Jmb6Qj2g4`EoMN#~FE$uU15}5)**EkELFT=8 z681)1zT`CcLSnq?=CuSFZ1;c-Yp+iy?u=BkTKZmwA7e8^^ZLW-I8H2@6oGm7aIQcv z^G<**vk+-E*bP`~5B5eN69HE%;j0ICH|6(i+RrKsgcC<@F38NiPOI+< zb5w5$X~C>H-fKu4-twyyHa+8z)hijcirpG)uk;)Y=0P3Ex1f4KMZz}XO2nslc-{Gh ztr{c~eFH+`vtx)KjLY0SInVfyP*9wVi~bdCPUr-NpH1We6-5K4{kZ)@xs9~-_r%>vF&-S^bIMiEEDd`G3N}_ zr*^a`ial_TP8;x_y_b1a zxGlmh=lr?*^s4;Ktmp%z5@)S~7^ul{oVa4MAs*U{8xr)}*SyQpl)$0$@XXf-B6(=L zG>_TsmnkW>CB*eyxcWfur$@(VqL^A>wjI zu~Ch;##oW6X5GB{nJC_2s_NEk^b)A?$%t4|LkD&Q^H;XbFjD2Q)R=yaup{P!Bkf}jho9^3q5 zA$@Oo6Y{t}JUc$kFwWQz*BJU@dV!az9XY!{2+K~&@%A?kDR=yF3tYGHy$XKtmf1Iy zVfOV)ohLp%SuaBT-lA!@C15swF^Axk#>?^%YsN~;t+naxSj?pTgk@VU{9~CRjV$7^ z^hqbMY=*#(UR*_(@{4dX3a$(L{u_{Q5GdmF9)aV^o2y6Ife&%(;V&;|ppLBI5fjs1 zl8%tYRUfl6|Iy551Uy#ew`WhUX9f>g37-KB*CTN(<>@%SyJ6#p?<+*^pEw<}~ahXWrWodkSd-Yc-*|E?X3?(WH3donb|kknG?sttCV zqZ%t9*`#-6yx|bN^Gv}deu%s=!&!Xb!hr1Q?(cl4ZknL+ZGPHOtX(Pfa4RfnzDNPx zff3ZbB3cIjl;X49TF1IHy%eP&BC3{Byudr5F}?HyIZ|!dKB0hGxUl?~aVm6%I7TCW z`Bda$myFxnzV7wQm^2qyijiK)lg0f;Fc$v+E_B!!)H_)#5V|>UN)J^lJkaCuAhw+N z{;>qoNcY=uMEnA&k%ME?VEQJFx~8Tm4ps$$Q}7V>58Ogmc$d$8qweHxY;^-Xc={cX zW43zp2+0lGhI2z-@*Xga$qc+_d^XT7L?(;IB=TjJTfRp9bv>lX0mU^S$FHB${)0x< z!v)%BGyM`3xp}@xgj0~FTyAwk~OwUOfEg~0aY`B<_Fs2VVxYN#8 z%t0cU+5xYnh%LSXKf#&!Qa_a-Y?#6u!C$OOmDir$b~pE}3Zr~@mG2kh^<)ZS^?EpN zl5Yp}h&&Sf`l&!!bLbizev8fKd4?!IcIrzAmDGJ)>W=UP>LJ2gcn9~^DQ}KW@DM*f zX2I*ub-#0e7affWxR;e&0O-2QznjQibw$ilh3GGcI*m@q#vPJE-`1@^et{IYW5?K# zdl-;*0azbTTDM4R*kda8c=@7$rSyh}Qdqy{TUoq(uS^g_K_%sJ4$J7gO>c-1-W;G0 zC9OLgm5Trg%IrjA!a#ohE*!U}LChrRr@aOHqc7cJ7P~-)_P=xRK=u2JhB5h;C|EHKSzxM-CV3MKQf`W7+^Hsz9>A|*OaO~y0#DoOi zW{JZiUO#F1;jng^kVT##NlJMf=*hcrg%MmzVegi^*XL!4xwBPt+1^4qPhPMzBt$}1ep&j{H={Hc8=Qa9Q!;(_>F9`gHle$~$tLD*LM z+XFj3XCaFRt{pdQ5wG9ZvWpCG)`#BTBs-^!60_W@=z?sdxGGalPAqbH_~#xD8DJLD z_LE@Z>GX`G$oVsvPnYYWmN8~r+t*@3_4_55nDhAeD``WCZ&AYTMQ($%ush}t@c5ef zw2-M3SV!&8q|ksLA`WODXhviQLfWv55?O({3g2|$GY}aOoe?zIX;&???uGWgdN79l zWxKn0Y5h6AY^pv{#Gr3zNPZUK$rnEw#H0uBA_Wrbg#tgHQL~@(8shSKknPROK4-W}wb-xjYv?i!0oFH4 z+br+$dhrS^g7iZoKBpzvzQ#s(!|V()*){MQ?e&@2J=}AnIY*?p6D+D+JPV@4QIrn{ zqM?qYvwFzF6uD@y;j=D_j7ZJvzmedIv`5 z?kxIAVYHi)qpavga*r#Pm2}eHuMd6DPV~BsBif#5sNBF)+#x-&ZNdB*t?wkyklQv> zhIe!S(!L9PQnrt`vNOrcFF&;dbvs0mw_J}b!eNJ<*c!mu1h@iNM4ay%VV(5T^~rC{ z3=19@lq)i5@dW5ojm{{>GThhTmcYS-^vNw?bOA5ENB40+=d#u1VZF7G58ssmVY6KF z4|w|i5AgK&M?8Hz!0i&Jc9F`=2KKofV!XLI zYY0znZ9l$0Z4Li?m1MA7v_cf>oXPrZdEUc>u71!?dsHsu{32>S?GRYAm2kLyw6pc2 zS*XS4a=l_AzySaXJ{%qg6Bf&>BTHxkFu-eXf{t~TR~F%G1?p#iCgq0sBjBi+S98V7 z^I}%SmYC{@*Q(+lghs>MJ9n$qkhe&9WC%@i`r zH;vf?MB>Bu8GZ=?U^myiUfE|0qPVIgQwj#mCZM7ESq-nu-9SNa?%S4yfUcTBxbxcO zjdCKE=YcCZLVAl>fUO7OQ-A4QNF*l5y=)Fw7fL-Ic|1?$dFcD!IsCRFn=D2$Ss;ck z9=PZ`g6%)TkoRiTyq=r>@?oahktvY zN(`k;ouZ2!u2!jRXc#%Bt7`4TfH`O1nkqvMg}zMvxDxAlw0@A@hZ7d!Gs=heZQEOzkhkuAI$U$wlC(MW=ih#CZC_&Mfr^Z86ecn z#j?higt!(Cilkrx&nv(;cf9zpKf313zq6HdUC`+;`TB=|B4a*yYPhv7Zeb%=?La4( zl#shZVb4bv%*FM}su>hPTuG9va9e;yRA6OmaO%Vk&f=OlcYzatF*FtIX`>jE7f@Z}=MVd(gl<*b7=qGd1{ zEYI{RfvY-S5KjB^Yqf@Hsa2oCPTx!UVY}85Y#p@1+q?~rNI-q& zfb2BgfALn9=Lj zGKL+)IV4R5gP)Ll(Byd-CvP+C&_RU+Tr9H4)_QWwitkvOQnhqMa1@sN$3_N@;m1*# zKyKKbVlF&}j>akXI%%(kNVq~i^ndIZG;YC%%m$b{Tp8t@AMuH7R+W#Y9ncZ#v}nWx zX|2LhCv8{m;n7yH$tao$6ayagHKo|lVwT^?9Sbu4>f_YmR&*)%cU$ZInx7EyVpxCT z;7xQpYN9b=?TpdSbmM2?^_6DlgfM{gdjucMiKZYSlX<*$+yI)s5ZF*aM+*@ zT)hxFEjUYnb-pv7jRytW3nU9qp=OFkTm@tTS(_8aRF2xB z^!hU&(2$$>r?J?>e_bR8xI4_(9JAX(cS=q7kE`nnfdC81Y%pX?Kv{8#yUatE8aW2! zjHLq~$6rzsCOQ4K1UqOyw)kygO1&0y0h^&9@Cqc}J?+Ya`v+w-M(fxg;PTrFgw&YDWut+?a@`fOxJuqwbZs>>{ozrXKuY{byOuIANxZc{`No$tSI z{wbvoq|(jh000~Sj?To;&Vd>@_c&muS4_qbm9Gwx1poM{#I+agGwUC}b3`z~NFR?H zMtI)rBE*V`k|VoB?OdQEsk|v#?MoOU3ohG8OFFgDt)G?!@0XN6)D><1Qn+oICCGaf zrQ}+Gpa97>;uYiXsP*~`;`GwivZL@O;4#-N!RsH&0Mdi8DuanRaFjwSyN4=J`vkAs z7HopkbfMKc;vhpqvPFOWN_Zy_Z=WqZp&CQ$9db~qn|7$;Py2IO{CQPWe|i|0zH=Ga zJChhNib6oU0S9q<_?+mNaB%?wV|5IXs&T}mWKA?{)MA0N#`DKPZZ~x=E_s>i=8N#H zc&+2i2bWDMm<<|Q_rrh@SPjmaJZ>WqkSy6fVe03!KR9=LYpz@gFOh&L*poXS!G%B6 z7+6J?+kKjE6cpRC^Rr3kd28Fii%gp?+U9`SM4Q$89rv*zQ~k4xRjmdy zRrJxYMBEQ!Ax%g1R%L2(o~jOmKe=bt*9A~r)0jOVoQtmm=mHctl$|l01%M}!^>Wm~ zCe=kKJLm;GF(K5Jxh4Q{G`Oh_n14Y<^#z@7O(UNRgaiasEH-QoDn39N5#0O~%W}mn z*|{S^43f!I=lGK!k3c9XkZ#3_5O8%sE7$IplEn6aLFa|Xnc-q-&|l6@mZEhaxeU3A zC1NyK-@KSa#JNaQOWGfmAV`o9);QbH5ZW@EdVmm)oF18!TSfBjHZ+PQr|VLGZT#mE zXNj=SR+kPL$q6hKHn<9WrCHLjU8tw)d#Kb?c-wekHpUHPrz~c5-@_mE{=ilj#;*B0 zNKW1365uTOQI{3u+lE?!s|FLuf88$g+)D-%c2y~Iyt5?-w_60qU!>cGRKe)>vz^^> zRcloO+ve(gdn5ACz}VbF5ZOrSW_T*W61zbjvD#uuwrA6@-N?Aq2gL6jIB2nlW1>61 zb)T+TW9eF2Poi6*7IIYq{(}v=WS)rA-F{wtQcoRsRQSsi4!hG(;h1 ze2ILsz~XG_sw^Psg{TULJYdV7J7)=-Y&T4y>4XcS`@jB zz}Fm9;2g6jD~t9@93Pt}OvepqgaR^a<>`cT z;j!64d;VHsEPT5cFN6tkV6zG>YO?W%sTEfn|5LCGu4Ojj4clQJhU` z@#sWq!5dZc8231UJXB(XDHDz%sx%iL@}-e!YH?)>xPcNyr4l933e2JsyxJ?^Otc&O z1e}$MV;_h!v_!O*XbRfgxdhCr8=Q`PHzJ^O(PCjB#}IlKg`X`yVP{Rv23;|nGZXGm z%Z&r15iZY4p_%BecAGxDEKv<=#^lN^+;WDJ83Vdoih>!tyKweLq0Fo{cTSw3KunLY ze)U0{i+L>4o@GZ3Y9fwYOM^^F)GyQDCmRhdBiuA11(&=rxDs#aP^us zCFKtG7F>D4Qk1bh2jwb)*n@*bNRu0q9Q1qbF}5ZFMY=w&=VE=60*nbuvtSoRLHtK! zi~~f`meuGSWjL@SHe?BFDw3}=fx4en05uPYPWZw^^3#sW0!FwGPqj&L1Utfu*wt1b zniw=~2;BsVlE>x2E+`wGI=H|@b$fCJAo^$IVz4+%76uu`(d~!5Lq0Wtwd9t;8(56W z^;W~+eZ=)9)u)+eqmKJm_y~bZm~A5jwI+lAh2tybX$C$5z|s~=>2zQai-OTH3U`@o z#^82vf}J8f$k|*zkhoRl3EOQqaJQQDw>c^XD&{OLcDjHIBl z`?}+@=;O4k)8k6_ap&T4hduK}4gR#EuO4pFHOiJP*qbF9Oi79x95a#DJm>>;gn^W6 zP%o9DA&GU>jFXm(AoQ1UaZsM*;P2%G3r8vos+~eOW58O)IzUsL)jMTZN>p4zKU-SR z1=bd$o8ViV0$&8UCjOBa9`)dl!*mgS1XApX6EJ+9D1e0x=ygZYO|uJ(X4j~yS0<_I7%!YRghXVM?%_Mu8EICL7E!!xaa8xKk=)(N=b7Onx|YObgh-R{hyU ztubvjQW~7iJfc&=fR1JA)p^6`M9^bsl%e2dP{ZLju3mbGE4TYZp~VY+EipxWdaEXa zjF@Bcu4YQ?OAQ0+)wXmmoVe^lbW&}Ipr{G%y1>tjXsC<=VywUb4bV-5;9V6R+POu5 zqmxd$2c8#P05s4dE^xp*L7bA$;+LRV*F`_f%5VDmi!DJvrH>FKRWCJl(a!Wzlo$GqWa9*3X$07)m z)f`K!Lbl8I0Xf!Gm);b1u+C|2IA-NBygy0}*pvu{t4ao0b#MyeF(;KXSk7%#OrCzb zd)le_xeQL`Yw<}aG7WdBB&;@g@xy$pH){wo>cC}tyr`r(%*^eM(?CHjAljw^ZnmgX z&V>YTze4AoG4%Y>1)OiUJzXF3a6TQuCEDF_1%sN_1fM+1G|U@|Nyrz6>6F9yK)@Gm zMyC=SOk&O;7XL)u?X{1_DXDI}0HjePKBzYU{j5}TjpHSm&SC%ss@j|lBktR6U`a$_ zuP!p_QGS#~{0(g%GJF|lJ=irOT&6@v;d#c#aB;-bBU~J@0Hu~Bc&_$`QX>ejFC6%@ zi4iq*29j}`JnhD$a);vc2>SwcD}sAM9eWJMbPTq`P>yGLc>HFu5Aa3l1zV3cl~fx? zk^MgAkf3k09A+rh6|GQ?DFP9>hT7|;;Y7OQsg&WAY_7sP*+n(i0qbbVAQ+VFn3ZG? zAlW3w-JEvCpu-@SEM~#Dl&&`e6(PZ^g5j=f<2rpoYL+CwBh-n-0FW>YNDUhWdUnj* z-D-F;0skwk1u&IzN^dYg+wB-6vvzn66~b+G8xC`$yi1s)SL9mGR{?MVP{t%zGF1Yu930a{(GTydu~Qgg@xscR zT{e6Dj^OMg_7C3HSwJYcltB*;4&7!eY=X`98P`0!4K`chMryhqV*9}NNroq^n=7?a z@UesSh;Tt_GBhko5OT-Dg<3d@ol|;VL*tp$O=7VG_cV^;I2oS!jwYNOWD)$otwMd& z<3trPtX*G7RcnQKt;!Fo%PsXGIrmK0}$a~NM!z2`STmrFCKdfmYgZPC+ zrC_gO50B8BS_0m3GApKXhJ*MvH$8MCAWwO z5E*%nN(nw#Ld2Vor8;FF0vO$3I%&Azq~G!P#L$Sc1FjqJDowdkj2C%OsSlI0aI=Y+ zQuJ_9P+uT#ij^B?Qei=HM7eH=(Nz-p;Zdcoh~boJkX2 z(_5wq&K9{?D?8PwUx{gxcz$7vjcgnzQe&kssY|`gKy`bgz|flrRWBf)z<)ied4bqQ z19lYNOrY3Bh=L735l%{IxZ(UY&e;-C2SyQJ;kyPJ5@NgKESwec!1hn%Bv99e+MdC* z%ot3J)Hnddlk5oim>GA?<}T;Q#cFQh7%UpTS`~EO43{7DK^Rh-0Qk7H7%%MuK-~nm zRJlQq!ujE1BDDnGLnq9Z zE6ms?w_Ub!vuJumrK(Y?j#;YK992^+5P9HK)L4vMVT$2hU_lP20ly25=#i~e!l^IS z#7#^Se!@+)5|+bpxtL+()C>si0{|PfYXigI(-+ck>KR8g7Hdzz#RKkEp=B>U<`n-J z*Ku>r-$zsft17*6C^MJ{VsV79cF;-06Lx^ zc_xT_8@PcbDsMX0qW>dIL(-Q5Px$KLvk{eYQHEIAxMe{HM zI8ynHR>+;kkrHNg?Aa!CkcICbynz)Mer%29vD<2d`Z++8_d2s>uVY*Yz)P*C*2OMt z)`&xnD;y)F)Ha}Bl%BU)w8KNnwe2D08+_eI;WTlm9i2cFPbiHoIJKCU`O7|KKRt+w3OIa=VLlO@VIZsal+Xx?<uiLGMi#w3WwWg|8&krg4p%PW z=4(=|GLxEdG?xQaOEf#HDs!)|6Pp7QDdKcaXm)9k!nat&e6>A=U><;azC|tys3ob) zCVhaIoDc2eCa9DZHaFz_F&xGe+77N_H-P1At|{XBy7?xk_!6@o;q;^6^&c4LO!xtAGK`9xlJIrHO!4fLV2YzsR# zHC3xjY87XM^9Q1GaptV|6to&I`g?%&!6ty23{tcBl9jnu1(Ag@ctya4Cai#+s?}`z zeO@UuEqNL~r8tA9K&UU6OKULXmz zU@w7A2X2LVAT2F84fm52JgOatK1zbhr6q#J1|~d=V^I?{05fXkNu>zT5JhwGJgAMPbR%Dt>N=x=KBQ#TJ${H7mYyr-WayA?> zfTy_x?echPwf?+dcMW>^3R{ZNjjWT+G~r{K41-atRR=vg7_og*f2CrO(Brxzw__9V zh%gfp6ts`mfY;SSEXObl>d!*C16P7*B?0~^5ALH{g1@1bVvTepO*5d)jU(!R0@7wF z5dtAYw~`6^DoJVx{uXBqhp?fth9ea&F_{X8_yF;t%(ra~T>h5<2kDy93*)S?qBf%f z1-`#lGzITp#BMACOtMvN0OSssk)Y`W8m`UdiEky%XeBXTC`JQzp(fnr<5I>0nK+FTw%A0sIt@XA}gd z7DH0}Y|awEWeNZ}=1R05>E z4CbdaE(z>hFyeOrlD=LlfDeXC2}|&@Dqu1gglR$3!fwPGC4I2usQRE{PC{)gm|Bi! zg&_?n>I^8pfT3;dVg!R^8r<;gBG8FQT3n#O{2&ekP1;!uh8oj;3TL*6$F?_8bElw9 zV*;IrYht?)0l!_kf&d|kuJTJbY-)@45C~l19H3G9rC7D?v*U<7k_E_*D;*o;u)#(! zM`M&%=sE6^>#7h+)~g61M6hvh3KmJwu)wXbQ0v)Uiwfd1U5MH22?bdL5mR#D8N$S@ zVijcF<^>eS?HB>;%tAAGddD{6&&kF zsJ7^7`9-CP=)xqQ>QB1dQYg;bZCflc*5LHgnDU1EUC4k)zk2)y@&I(WLH_37E;#TSG{gH76N!&$_~l}sh?sy4gmcRGet+r8_1QK zBkU+M#Tea*0jiU(Vn{dJusye7aXnAXH)&)jE}f$~BZiq(cx^HT z!k{$iDWE*7CPF^e^b^a(T_s$faa|C8PGZhkx1}`%gd9(uQ$Fv;@jMvvW~CBkV2PWp zR{KoCOp0lDiWCNmroOn%SDUiWVp`R8moLVPo{bbszL0xKmBR{S()%^H>G#4p2zr zR$&kaNCkjH5dYI1kCRyKVt`2lnmFK>)Uf7Ev~e}&2blrw?#2ouVccBA`|gSA0V<~2 z7HL}WxPw{6RYZ6h6bUwJznUr6KnS{h+noZ^kEaYzdU3oRFy|jrxYNr+Uqx6OxD*%} z$GO18SY`uJ#=%m_mcrxyPHkz^1@8j&Z`SEsAmIq##!U`~n4qpS1lpJ~H6o&5G_?~j zi|7gM0Ts`vURTQD810tC`lQq8*QIs4e~_4T090Z40>f8TC^Sw-rcEeLTEe(P4MfR? zV8Sy~r;EYo<(Pntgb#SGbFE`r&MMqD7ZQSF-SEE5W4k9;P(F3(HZF5y{pbtL_Tt|R>G&RUgb1DMEq$>P1 z5x+g70q09mCe0c4=n6HV#|91{IKKuCRRVW~cb35YGGAZDYYaV&hQLqVk7c`X{(zQN zV%5GZ8#GFZek}pdq}B{W6a5Xq0dD96UwOfs3FdJnrpmYn;OkR{5)xW23TjRq>dM73 z&VGW3q{b0RCAk2^d&CWJC1o~Aa1FR#0n}$)7%wJpB>?`+G``?^1`LnN;p7U543grc zSWQiehzzL}C{BboDel(nK$r#QKWob~C4@>8W5bRVGYDpTHQ>N96}BHAh{jz(FNz)V zVhm6E2JK8ufDz`*8$)me%OB3k;ml$R=+;6G*KdpEU@U>g4B#=Yt}Lcvl;+i>x+sK( zi0$$Ssmc&jZknxz`&FgqYy%HsA!-1h=wyaeQ3X0+fvKiAM>!b+q;p3M;n32V5pA9h z2mVd$N#Sh|UNSE(e+oeaF)773qOUa}4^I?nHcwH@{uB|dAlzW)nE?`J% z^F;}a8+~6?L&pPoyx1F74QQ@%1FqJMZXVvL>3%T*xUTX93^3JE6YPq8vD6n6br9J5 z(%p#910!-22Bby{x25f@WUEMxp$NyU+#rNLO*-IzJF?l(55t0!qiRS;6x<#Mg5T9x zXCMy?!$qzONQC?}??{Ob;J)T?GNC2HW6A~qBe?Xq6QFO95WMDUgsY)GLb=_UPPjcR z0}ZcfUfqD!84A5<8+S@}Z~+3WNz#Bf#@X#meK`PyCsfpfO@L_ZsRCXD(*@V9wS>Lb z)?{v0cObZ&HUK6|&~k)p8*psZBtkF%{pe2kgY#^$HslyLn}k1l8?Nhy(@KI(y2}V^ zF9mop*7rd>Dycd+wiiOnh#N(45%{ZBc3D#poDtLp)w<{~s4k2+?c~g0G1J;OMe#d( zE;xMR%)~jE)@7i~q+5U>iYEleC^VddmTg8S%M+(!C?{aa{PvS=YlrRF9U_sgIBLk(X zEVvM{Firt+$pL)K5Q3O-?=1*|(;yhC0)&I@*b9y3DBsCM5CSzAPdIl@!rWO0izNVv z+x00N`CAb{gDSI6h ziRmFI6x2$CqHr`awmWEVYlf_qyd3XQlx85~Io#1oC?1l$12`^&zYj`?wVWRehNUL> ze>ihy%{UUdo{D%;%(Cfpd}(uJr?6st(Lgv`HjYwZ3v>WHVZ5AAi&0!C8Y|o)l3bKx zfm);#(-KS*&=st!k&NeT%pg1RAlRW0#w9eoo2G4v2OOy2q%nyk!iFoBU0xvS8Rz|m zhm9CEHaUhDoy2g{z|t^%4Zpq;zpxJ`xG2kutvzSU+HFyJn zy%N+4RKQy-;QY<&X~5GP_jOa4&isHy?0Cv4@Kpdd)G^#L;3v?<4a2A~V?47- zxP%E58;d#|&{1QDXR~EIAbC}X>HR0AGH z7etiMe*v9e1i03soapCc70v~ySqy`juPzLjDljl`&D|YR zwRBZS1DArGM%7krq(UWBzul0^#T49S^O#xM%m|?KK5FDf@PV3~!~yez2uXvCxa7B0 zIN)D71zeu72)`|Y%o-rgHD%^#4O9LqX37K-MHrm1s0E-Tb1auZ83bAY9Gy-%_}^ja zbh=~}jL`#l>4oeH2=CPr`~aq7s=~)|6Kpm{tV4`6u}6q%YOd93$HIuZ?~Z}CjvxhbmHw9^K4W_39147#$-hTGjv7dNB}@K26y zO3_^&))Az(7-LH%1C}igbmmM3ry@>7&I1S;FoS#;OKv8M_AAK%-LDUD7>yuJ!#4oX zgCHfu`;_SO`<*#BflvTrk*Q{F>Y`n^^B7{upHtHe;0R-Rvv32C)8?$ZmbH$P5|NhN zF$iT#y*`3`nGVv<2EExXu(hguj%xBen#+z0K|5Ta@$Lu~0>Htl$wION{GfcM5Xu`% zuP}|2!KwAac)#+W*FEeB2>XtntvkJ&U)3&o_u z*lt{bwIk*;1qefF#omV{%=gkbZ-P-W?6jch!5A(Jd7#{6dO>*51KQR(4|K)`vAow) z%md29T^3cU;&9eyRga-pSYBZdcETgoqmV90}$B4I6)DOd0^nj5R?Nh zW`U)*i>nt(n7AVgpg}0Ec$02|$rRkO`B*Et#F-Q;z#J;kZhu_^7-VD`h+sVF+ZWjex1>Qn1>mTXL;;gP<84hp?hk#P2qGk7Mx=VY zS%lj#H3!RwhLHP`0izsgsSTl|77g1ToO+@L8j2z_Xig;?Zg1c@lL8-NNweDvb?kDh z2~3Ht`4BO49Ll8ifSd~UmhmFjtkr5f@FO#zp$c}q zIk_ouoB`Q^&B#DAoR?6zB~Gb~DYb^{G$RxfxWQn>?St+f#xdqeU(4eL8L^1rA2I_6~c)ejsna zxW%csQERFR>{Vdf8z)NhmVt~0Uo9<+&5db`b_mSOiCL)IZbdnd@h=M$ zflSQlD@P)?9y+N$@1#_F!s8zIvPo_2zoMk!ne-gv&vT7G&%~eW58rk2I_dfN-^-;X z{=Uo)W3Bzp_7+x znX(se>FJjc*XX}T6K4|_;F(%$-*)O_H1TPx^_s2w@4NDZSt_Jf2TaS@C)wn6Ce4kEvkFFy3Q zdWl;OO5>>=;zn!BOykIz8SfMm1-wwNI*E(*OOIO)=`|;D=|L&I(n;KA)i#vokc)7x zN^}3$Pn@sMT}0}jbj(`4ZbLc|H<8{+=h&(l(V^GJjP`C^f7i+9oxHC71nCz2x2Qu$ ztnb;7j*}zQgEBXpaSQRl z8EL-xX~awMNWbg2^~NK&?o|duuME{Ob?dvg5)+Hna=xis7X7Yg5cZvK#)r|y-+Kn( z+_Id%Pu*USB(Ad@)vxdnH<{}8;7Z*r`s;5esGW4KSjisH#@`|y_c;CYN39oI*bPZN z+e2J^*(#0G>b~1c|3&C;@er5lmwSkt_ZQ?*5AniP_1@woo_0`Oj4rd>xuF0>ACWzy zO}^Y^M-RJA}ij}2oXJM+b+_1GpMGH1Q<03rQ~g7rfEH*(hNtoBW!34!ohA2CPd;Z>biW9NR$ z+1H*Wg*S=coF;-Ba=y*_OjF97rT0{FcSg8Sw`ZUXKl2fXbS6Vw_E-hce_60Lt>aAv zxiL%BcEpVjV=npAEYZRLS9-Emh?SmT-28jV`V5Qpp7SYr1S(W(5Z`w`ecMj{lXIVMx?;JMf!h=%y%CoHRvwqqXHYb{OVLe?Y z-fa0D-JTm3&BKRiawte@U3(@GLE!hb%Kcs}_$ zYRv3g-#hU2>-DeQL44=bPoGUZY_a~_eBCV9g$KW`pI0T?#FH;DQ$M8NRo$t^#Rt8P zYW#VXc+y#xi}i;ZgyX<@?Znoh)7HnHTG&J)NZhcU&vmPO82*6Qkt?n(zQ&_8hpdGYDl3}1I=v$ePO)a9EU z;+I5cyDqI&U2>~L_q_hwJ>pHoZ2NU}^?LE(*H68w4_&ps@B(vbexd&NL*k7)-;57q zIK6K~+;zsIfB4Tsq<-;n%aiq`Oq_f4h5H&{cg}oR|Iv@hBbIlV1u^xxyeh~Gb?cZ2 zZ0#4<%VXji%SSc{eQZpG&M45W6QX*?qc2a0SDf+a%?i=Nqg9iym1VsRPc^5+k1f_8 zn2mxD{rPj^ovW{AtK6|5-n{40Vcn&{LfRHO_JiFfSxu?CjdGcO#bNS@^@0m`*Mj=0 zQIK8lIYeV?mBMLbME}NfV0&M?NhGa;P%#Ml{4nt<&rpRu?+NwG~!BpJop0QUDm;cW_7PIrl|2350I0NqdKP>UjO=C z#4-J=cM*)`CI3qy?*7FAiDAI^?&WZFFCvDPPyR1?ee(4O2%~ZDc?mH-%kl-Y zajjptP(N`GA#8oY$b0X^_M&?ToAp~O75u`-chiH^PwpWaJBb`#39m+g{;J<3Zn6G+ zlQ?lN;X1mKJFWMpyHDC4H~QPq2isQZp3FaZFHzdb93NL`#9=+~3PL~Q(FL#E_voKr zN!-8Ztqb(CUQGn`M_)zc_B=nVzxdU}eP=vM-FHTU7wDJ2hIrtNw|?>(qHnb8?%-T& zxOFYpzjBhemwfHZ{_}C>KfU={;zQQKMaJxm58d-R;$_hN>ixt`*BLth^;HY3QrV;P z_m=d#9w5%w@7GbIxegV1iJpGphUN4d-`?%bRdUe8(Cdk3ldr#MTk9_}X02`Ow`u&=D@@TJX#KJmic{8oY} z^t0bWG|2C4H_o@s;+b8;{N4et@9mmT>Ti83aq%hlTZtE0EdQ}NXYWG?B!!%h{5BCL zpLns^Esww0=oT~Q%%caqzBlI&8adx~>X*MwcrBJK-SVal3Q;?L&i)?8bn%wog`oAs z#irG8LVwab2zlq5@nKjE?|%nz)7A;2Y@53BR0`rmJ)fN!`L&PUv5LH{m2E+pQ)WRX?sj_hlp9M%k9LM7waE?FX279+*hB|bJM-&=&2777wF6%U^jE0Su=Br zdj6^eNaU4&KwM0I>hWfWZ6|$fmE>xg+)ea^{?_*pC-nDzks$O--iI@^cWg@ep7#;W ztyNm5r%v7ae$3Z@yh%!aKXJoa9Y67YLOLjsTR%X=tzX)d$htj?I_}PJYgIQ|)zAZ@ zWp&#h5^?h1x7+{U%tXy``NCC8tzO5n_E&#MyvOp3%_@3#e^+-_a>(Y94-($3CttEp z0dM*cmY@qR*)_dhb8=Vy$(QUg&hd4(GHY+`$>rP+6F=U%X@4%~egwVcGIKFC{NyT^ zoAkR+!RkFg=#f*z*c#a&^zuIZ>`hZ~8w(Y-_Q<#`pXmwCl-LIo`0UF%mlKle9e%&pNnf(P1!qP{AyFIg(&#EfWn~y(8{JHhdFWJ1ef7?fQJANey z#re)Bh>vYOvcH06Lw{pc)DL`PSN@Sp^y@x_zI$Z5;qdjrPZ1~abydu@x7Tw!b?;{g zhxO>=FFjLFcjthM^gs9vagqL$KP70(IlAW~I|rd3UgiFH{mwrlBG$)Ws^2FNSL&bt zGa`F@bAr;lP&K{!aa<`}zahET{W+%d)5%#LgduhNUl8iAmFDXn#F38hG2)53@Dbve zsrq|Ldc$AhczsXOum4NpwR@j5{|Z~HJ?}jIbHp3=J~{XE#EbVn`K!+p*BeiE6=N;m zJuQCQUmKgdmtSh?_{%TV3lCv)xBb=fa^0-1`wQ3w?|pRk!-Rguqpv_xBo7|M+dpjlZ*M&VN`X@Wvy$_g{z;IKVe* zcaZP{|ANW-qnm`!{|<5I(fwJTZo!>>nDz6UWHNj1Mm+-|vilhi+dk9hhS*;l@g9p!dMp+x$5KOk;-tXA9J z0$T!`azN7`z4{-wwQJRQ`%Oz}yZJ4JO$j~g--sKyRSKsK%3U3JtmbcTO7s{1M)*-# zV^o<+zt*P4_}EqYm;aD>yk+YJNB1fHtA9keFi9JoVqPzNbEWjB-!}Gu`ZIonOJLWJ zFhX8%)R^ZtruCLpQTL>!KjFtj;UKC0kBM6>f4E8N+dsx7^>&u{(C>3w4(susV4Ly+ z-SfkpO0VU!r_#hviMLoU{NfpvmVAeBfTjQFzi|2ZcmG8=Eq`z1ed7g^7tb*}d)Jv#W*_g#byrCm`h~xM<>cCw`pbVoa0jLS@Gpp` zTO*sKlK(?62T8r=f6%Yb+9dTq|3h5$7%g$HPpAKr@aZr3CGkw_a8n|e{tDNI2Q zze3h`Z<6}UUlBJSAobd_$(yaOHHEy!S;@g|h?%pwnmMJ1pNQ5yBJD53hu%d#)$(N1 zUBR4Q-?6DQXIsew)z7TMA8(R+-v+6FwUV0kbDN}o`z-R6mM^Y`)mjzz46D1IK%Q;2 ze)}@h-23)r`Ujpw){cJrvNNZaGu8IdRleU}4Kc{yLy*kYetG@#1bLnH(6QaYvRBGU zlH95g=EC!KlDzuYD9AONlJp!R{Rb7~Z$62FaGRur?;pD1*T^-tN&dIaCWB{G78PCPqv-Mdz;_LEh1Z#kyFX^V_+?U&bo zwgt8L=q4%oe9~`y%P~`Pk;qriCvSMHLhM{CeR)%I&pAxq{#f!mmH6%^dHV(AjlYKc z&o{~6aDo&LY6NrSoO674JX#HgwMwn)rvB2S@xC%{)?kzz5d-%?Iw9s5BJM&=lX3^ z%2cC0b=Xy-3-v3mB#-D9UrAoOs*>4d^D9Yl>wt8BcqPeLt(WiW_nvh1JFg;dI3uA? zT}9RpN{G9fe7g16riA|RYVujXMzh#A$-np-Qdkwl@WVW;Kjm8T8a!KR;ab`EsL1oL zC7Tv9cR8Brh?Ft2Fjb0N>A{}&6E~1=w^;AI+*FA>FV`=*k$n5=tJ$27-blXcj7JaN zM0W6KRh)A+BtO2Hyvky|$4LH4Y{&kDhQgixW+r~Wjr@Q1KIz(#-92k*ZylVr?qkUG z)(4H6u5HLAA3SEscUV7q`JN$n(JGA(UcSBP;B$A;G0!X3-dS&=(+0`Cf2#krgS`CI zF$amOnupKG7qgn(i526*F#qxVh)m`Qc$}eZjdr3T?pUhB7;ZwXL-74 z(#}YG;uX8Py*Kk)-Q?W*u-VCnSIMjm^dGoks~p`Vp5w@;A5`|Yaab2=Q_38udoo=c zr~*%3tbgTcIFU^ASniDq!Bh20<2+|gG-OF z4cYDEpRYWPyp??B_PgaP%;TSPPOM%mT%l*5PKGVFoqE&L$-HGtHr`y*AN1o{w_%#l zR&8ok{aH9=%Op{wOnr8*KRKIy&nAqw~`+Hx4dN8`gfbsTECM*;~Cw4 zp3&`VZ92F@w`hm1vG2B2Zvylhk@gB9YyWTuWQn zF8AriUi;+&+@GSc1LqsLqP14#DXhA8p^2!~h+Zo|&MOah1!zj(^EVt*C4NCXv44EQ#gGKb)v*a=TmMqCwa;xUCuZksH zssDZ!ldoWueed$}y(}5Gv`i`UT5erRzb``Wde|e4-KJf8DY0GS)AQxpFVUUMHZ-kH z<;m+YJ>>JGfGDQjUS(Isewmr})mv?&mDirXcyFF-zx;1`jtk_S$n)U>$ywgB%G0^Z z^Ibbl^{)L*Wmr^}2RBJQr9_st4v<=u$QvyWZ<4}9v{S#eGQ54S7Onjfz0#bzj_pic zYpuP~BKYCFp-g5h#80<1^(WhT8*Rv~>WOv5{!Fs6Zj3zlh;Dy3e8R5z58@e5S-YcJ z`N;h4on(l-@G5fv9KLG%GV5upG|$5?#%0#2o2%p};0JV-X~v(pN`I+hy?pD0A-AS{ zu#H1?QnoPq*VE+cQ0e!z$m9BNwxDulBirkDj-Q^?CZoH@Pu^7t@>iKQZyvd|$#+`u)$q8(hY|Wz z?~o7dd33(+lCWD?f9rhxmlEm5qg8R&l3x`Uo(X^RGV%uf`cISR=^yEmQS1FznLY}x zJzlt{<95h%Py1i3l4##{x|a94BKloJSO72Xk!|94xBKh2jrwliuYAFFfH+EUYoC0p z_5IuV9_WMnjp6}bj^oV@-);IU(@z6KP3-z;P46^U9^7xNc9^brbkM|C~8d;gv{@PpX zS)Fc1{l*Cy)32G3&m^9BwW;%u*B>hGwBe(x99FVu!-pp1udG{FZ}u;-zJ0g)D?yAI zn@T>}vUS49Xl;<3Zp3f@F&dFFvo{-Y>vOxQoVZ&5%PE<(p168@xLvJ3V@7@!9&KO0 zW2dZZFRy2*M;7E)wi5euwEQi38?FG2x|*3?xXLVXwSLYcvtZb!D_EQ zuw18Zdl8OFwjSMIFtdyAc`?r8E!XVs?Z>BfQ+f1i{iiP`GuB79I~QL^-y)Mo_osAv zh@CzvxNK8$_LoB0_RH(iKa+P_=}l7W4wbuQF&B?#zYIplqpQYSEBt!n8I$pgErR}# zLLMApbobYGE4`9~27Kt{WX*c#HO5HXy7QWSV>xjT4rThQV!!4IPd%-JcV44^;U3a; zkkqBWN#3s0GP&*EzE@^_Cxj$+r_r^y_7w2%?0$+`BBTK zH#hnxwFBhP=!I94C&({tXY;qKY`*Z;171fqfA(tf%Lw<}*tg&KvjgOi&9Ckw9pry( zXY*q-8#BP<$3Hya^*w&o`mbI?cJ=>$4Y{;F?%Ky#87?~Sv@)!>rT%yKlg}czt~I;t zsn;6g&g`YlLW{KNJqoVf*&T&F(xI`TKpvO2E4 z|E|ID%Xig43gBHSh(AV1u@j%rDf_j0wSVZiC2HalEN#Zq>!ZxQyIJ0BRKzsVq!fNU zJriticck3zBdLi0{sur1On3wX9pN|d;NEk^xmZ^f@s6=LQG3cs=?*+S{mu$+L8_aN zFT`KuaPJ{SoGbLGSM3Z=-usNF>z3f5!^@k&UCB0X7KQ*V^81DW_pB?O;o$PBa=C5p zZi+R4aoBhVZ@cZ}WR>#0PuVU3-n^qeR0qn*7`cY-aJncv#eH(4GeNeCI~rA`y0@_B zN&}P*36Jn`Z(VG}B~ zFeiduA0Q_2FXILTzJAA>AODgiP>w6Jo(Qs!4!G=7O_dtE2h(1mUnn$y<2*}&*3UDf z!bxwl?*LkDlpi)-+|UCEVNXkq`J|F>1~6fd>?mdGb>}3~N+dk9L}E6}vJC~qPwuJ8 z1Kg7h&{-kXX{M5{$wJmrfSOW)(7JTbMOu`H-Fa{xQKNxoDq0aH^q`x|ByCh1tkL75 z0?3@|G~#Ziy$zRrTvDgiMRAT6<*WAb07PYta&RF#?Rc{xIx6;7EjliVO4aUkwo6qI zvnwLR3WK6|O{Q9=++{wz%((lkOY7i;qS_D{P=d$jQ&6M2SvhYe;vO^p)Wlb57>N{B zp`LQ3lYL~2a)B)+-t#g>#x8!SjXmoBt<3WOi_rg%GRyzF8UZK~*zl8NGGKRuIza;N zzS|`-BL^e9dBtHbmMfv5!rjfnfovJS@oQd(I;TgSMJI0v4&Dqviz|b1CX{Oe2$Fu27u> z<)tD<8P%qFidD9l_ogedz1S=IW?qAW)WBe3+G46v4ljhFlTjUVCmT#7*Lf=C8p>Tx zEfhx%kjf8}SvemX6(SW+uuzx;hYFLj7pFmIDYi(5RUU9f1zBPm>9kOHsP%A7@o}JM z7P?C@+0A>4l}fshF4nZX%$GW14cwA6KnxjM7^u~KK0Vgz!{DIQYWPP=$~#NB!{u5@ z2Kb|#9C4Clm#4knECvqJHd3cSvkvUMX{(X!N+WP_7zJwQeZ%}hDU5k$s>*gJSVnAY zN@JG^j~tW537dAgn{BQ)OAaRXUbx#y`%^J{(FYDGdkft$wvCPcb8EpdXuKv_ce~!` zXDXwftO)Z)q}0#Z#gW}LExG~q%oL-JkXFuQL)CH7<*5M1%ppp{xnS$W>2V#PQEai$ z7=~vc^U7*t+SQj^6Qx%J-AL6X$wj)Z*_ols=dH&os@JAw=H`aeXFM zG?n({Vhc9KcwNbHykcisO%GLGL>L?6V#_s9N3?UGC~T$T8L=q)0Uk#u+yhDg17Nd| z6iaQvBR9llzE*E#GXR72Xsw#Sg@>UYwA&^u=4vuN@G@?(sJVNrBR&SLU%ruv`e$O) zCClA}ttdC@H9J2NR1l>D;Ws!1Y8>743;CE0%q;;5{3+vPGwzIxhM|m295{ufD^VXP zjdm>AZnXl_czap*%kyH#9ZHU*!YHXK!{X2$8rou7u_{!?UbR4BW^;=SGwwE>aj6jP zj=Vj8evqBnnSsq#g@zSCH}6V=nzNeb6E@HbvBIPj;w3lj^iS#7D43oJ9W@CS5D{!> zQC}fC>qj$_Mt33t%ux*nitU1`I0h+(16Z)601#P@jS`bWr|n{cXaNQ^#Viw-GL?4E zA8BW^c2{l!D8@my?f|)TY8;Y)xLglwF`9O<(;*+Ob6~hHwnKe>A%aEGnX2~yyq*b7 z3hhv@U*t+s6ZF!+8e2xDIa#c`D}`PtFJ{#_-?xKJ(ij?Q!fC6?%|u~Za11b!WVB3W znDjaPzBJpQSlKmBw>_b>+XeEQB$u$$?pf0rcGHd!L)B&@U%y&N%Y{FLx5PC$9Z zhH?VnPazjuXe))$Jf+&p%Xmzl4bq*uJyV$hgFB;6Yn`!IaZPPX!WpPCorTb=t4yUN zXT(Cu&Cp6cFpOtvW>FVtYn3>i5v(ra6 z48XY63Y95RZf#0~4=>7irfyADt0OSTgj)86nvQrToms6=Dh_OpKsTGMwWfhtWl@Y~ zs<{eiiH3Deb~dz{+~#cxShKygLUEbr!tG{q)GdQJXt5Zvt-^xIX@lszFNLXQg%x}y zIzI);W7E-8XlJPDlmOAoMy5_%yU~J{=IY~~FFBfGXmt_*p4BilV=7i+iBzi>XboDN zyEjjT7QWepg|xT2 z=!&j6Ca0 zQiHh~_BjW!N_tuqf=n7S>8P)SSeeem%EQEPxa>sgN^6pBJEnQJ40da0Gu8EV<8v7_ zU=zMrF4>)}va8-uCsdpj9UADh9aE}M4g2eUuqgKD(=hLtNli*C(Eu+4f1`UoPWLK} zyzESj8|7A)>wzf@6zDQ9DfUUUO*4TR>*7l72BXfGg?HKym2yXBe#RKAfxDj%b!HO*&Nx{*=1p6pEKstC+#+$9{*TP zI{2jK05WPnqX9#)gmDsT0!NN-)3ljv#UNX%JJW8;hKdgg zp*A27%k%IsEd+;Ie_QOQ6K#neV3}w-Iy4>TX@5loChXGH42qy6a20CPNT|5*JEuLT zG!guHEe2LLdp`h2!Z`s=I0(_adq78XG&3lyZ zcUhglV9|tP8%yjWoCBhNB-vBbF0J5Fu!J&Gf7LZrO3B6~*;*=?`3>VG(g)+J26S95 z9dE|V)`ISL@!2 zd)k<^n7YmH3kb0L>%CzqmgGY=s+4Awd4#PcfS9X<5@IkI?03XujAGOFTFL9kjg)a* z^yEQQ#{?!(Cfas^drzG;!)`B#&F2+g%^lP-R3_W*DRO$|r2xU%q!@15bY^*;qZ^AR z(C*t6Hr(N2+2AZ0s|-fPqCKwi{+!g!k@HS9hu$6pZ z*NA{5K2P&@k?uHnni~kye!58wJN;shALOE`;-t`0#4$hV#v0Aac+edgCb+wnP?!0* zD~S1CDD-EHw>E8b6h#@?!Fx?fm05WbpEi_&yI9FZoPeCJjm5=un3}tqgIq+jdE7o0 zlv~`m(W!DXwlxXoIUp8iQvMlLX2FCOn+5^=oROO$kWYuw@m?)B%{WS)1Z+WCOEhWQ za>&vyUf^;Jn6_08lQVP*hS*5N1ozqZV?v=~MsLd0fCOb+~ar(4eKI6%z z*jRNY4f0jCkk1qgHaXhp3ZX^B%cK^)O`4?6C;nva9n-QCalhxH_WPhqb|x7!UTp8+0gC^n>vIhA7uE!8b$F-OE@ z$;YYz$IKB*E{DTdQ&aPWUc=2v{ZTmS9NK%uO4+fXGG)Mjca!rjXkPpCYNDMFVR$5* z!l)LcMup^{lA@wDz%0{?9L+mQw7@lbPMMwttL0X1%4-ZA3$qQ;CU*xBwKq>zdaXiK ztt&o0>vqz$F%CpPW?Tu`!2md^4mGaL!_Cv3k+rAMx+*EN8Vs03w2Q^uk3G(1gq z`9*`K$|Jc9j;EoJ_D94-*<0%Ny(LeaNtGj&z@(7wyM+3}7M*lmDSj5CgiJTmPft5y zz&Y=F8_`BF3Cwq}9@VHN*oiq?q8^UAxK=%_v`3R^2Roa-AlT@_j1D$@MS$vdx%R9v zD1lr-#yYVO;(-aq0`Kn_o0V8X&dq$?STmgUa*Hv;DR$niMKp<#uq%?q#(dmqxq`d7oRva@STZ$CI@`5Us>#>^#y@rRhM>R9i=d8fxP$e|l%9D5 zeV^ZLb4mF^L2i`BUQbf=%r)>c_}YzZFIrJR@5bT8Dl-HhWkHw#ns_GUu)Nch%k8SR zlKglPi&o;XamHUrfvZL;4IKOu{16e3yI)*jFCQIO%l=BXnoBNny_&x~1{88!j*r-= zSQIt)P;^$*S+c`tYqpfF>~7TJ&P7O!vu<~zJ62}cKZ>>m%u6koUbyLkgLT@Xk!H9m zX3`N+^J#6Enh|el=J5(zKR%lKCWU1(Q^T~GOxshM6bqzTP->5oWpAFEO*0gzxH1zl zDlekZo`<9S!*-N*PFX4=&`h%F1S9t-B-Be$m!|ec?QRT2%Q1h_UJYfc^oVIzJD!w3 z`2Ui09ZRlSLG*)IkVINah>XZWz>X|3A_>^&*WL5HaJg)k=ZyrnZ}$m(-Ueg+9qmm; zO`g6kT2!di9a^A?AjThLWkgTu)!_tSE&o6P6N5&oA_zrqu61`1Na%d3VlXl;61AUfJou*Z4?ZSzv3>4kmb_W7F{ZR1LX92KD<-39e4NO9yfabWi2@2T z1dUIfjuEo_BS&==&c)rWE2Md|Veaq#Gdw>?&_HEJSrTQf07U}ydnwTMy+KT?U&G39 zYW{rP>Q+Kj#~0Fc3w@8npo|!4y&`ZpFlf1JxOvNkgO;0x{z6ygkdX$6WSQRUWZx5F zOA_Vxuto%)wDkDEGVUD`m4;P}hNRF(n|&R~OPi8A?C$n36(I-Y@m1aq0V!$hA%)0y zPKZ9ar z{_sg7R-Qs4^Qks$FK>;=3drCyKHK|K&^p9nK~`T6F&ABdYDRIk8B5st_f_pul)!IX zwNjr6mcQ)lBRl2=Ha54Y^)7k|?hxNXLP$d72cmozS^19(aIylt&=0H$W8E(Cxo&$0YsZeA1+RR*YWQ)V=$pNNE^9KFD%5L zPw^!}J)rjDEdMGO%S_gb^O0_acFh-twkn^;aV!Y}(@orP@e`47B9CNgO;*a&qUaGi zjafnhb_#P<8+Kw%^jyY|FqW-*O!E;(5 z)$7)DR7A8pJtAChFs%RAv_9Bx<1FLy!{O`A&)_H#L~?X$x4i?9s?$XvS_?u1E~o|J zePt@KPPasN9i9=lM@;?FD{m*hShwuV^0+?{{${>eOz=%jA5Ic>ZQrx5gxHDJiCzYE z^?gUW_0N`zO|M>nPB5KTxvJlvX4-NlPYx{KH#}69QBE%|Fx-47c7K*}@bVXHUi#-> zPmp7J23WJp7LnIv7sOg4_~iXJX%PnFtOq22w}yt_R8wW}3&j3l)&{*~qVR!gT}g*d zY%XMH5@GP|uD=RHAb(S>aF=#^-c}jk^y=6olGAsH!h^V;FC-sp-l=}#?(u{?zy2aW zy9*({3^I;7pKaxmfWytfRAN1a2xi}jCE-eb!GEkFMEa7w%n%J9rjhqw+Eks@Rjw67 z_QizAkac{ww%@4(%mb0Vy5p++hHwMA-*SkJnWI~TnSq^PB=`RKi5LSz#3gS>G!|%> zyKf9BRylFmFiqs1HnpMI1Bqs#K_?3WQwKouJo^=#b$q-k zNE(c_pPEfhiymB|-Em|kn%7fZxMhFanzJv@d%1->eJ>z6M}t=&N2deQTssy!@=&X< zDEUYb_8SXj&I@_P7XH;E0M+uxEZiXES9rB`_V~VsDpn?|H)iGIKDFKc^wHj~-i6qI z8F`&QH*o|2s|s5`I~6Xzi647Mpw&JZv@)Q-;vWe67@iZx6c>eK_b89Gb^k(a1hh`= z5|8QZveEDYtRQr3oE#>;xn67?PkpqWWL%A(0?yJv3*@q;5$Qtz8&n|&`L}I$d5Ma? zE$J5|T-15^n(vW58yMhrN1o{(2g71z%31H*3nYF{saU;}e6~N!D3$Xh$5#l_yw@|t zSl-y;p%du$*lb}si4=bKS3idecC&LuMdd4Uo}nD8SS}!t0~E*mZ)Fmsu1u$Zfz`M6 z+&G@_^{#gWx|g>Cl_0Ey?9{#vn|v%eFt}j6kOu=@{|M1EF~mnBWPw7u?3tdBIe?aR z01(}lvoD5$K&{{^^#s#ZkR<$y;vFI>^9H-f*Kfy=&lrFGdJ;O-T>k(hw!+w!hn%Vc z0Fe$*Tm*i9-4lRJ91)n&Nn&4bZ#A#yYOqZZVcEmB(ClhGJQjr-ilPtp&Z-X-iJ+g{ zS+9D!VHfpp_)Xf^UdB>CeTr$kYJSP!YSc#Lg@}d3(c5$g_KU|5HFg=>bnF<6QQU9}zONBxku#5P{xm~!9CPcZ z#AH*+giPqahja@T7W`hOb3xe39iv3VB zL@=wQ&F8|@#)H%3?Reja_JaPb@vC=`4FWyx8f}l-$17TTv+ni{HLjZ=w^HiS?;WLX z5S!lB4@BEt9XedH07PCzflfhB9lf+D%acr{!e5_ag9t6lhR7w(8VcvV*1M^_pD~k?>45clb;XoyPAo3pwP7 zC6zR)%ya9V*Oqd`0OGIj%zT-J>zjfa@9%X`Vu^emdL$ZgNL~dBDer*-^s%xZFz9^L zH&5Yq6-~SHv!qKOU9?^!KiN=bsctSnFPYxsWT>uyoY$;CW%PtzcOsD+f+FTNTUxqC z)9SO6?kr0_ests#n7K}GH>xT6z(6P+H`UqrGS9dVUN~1!{R-pG#1Juq@C^D{8R$TI zl=(Fs${-ecz@Kql75^!)cNmNo7VDdyGe{#}v|ioTD|hsYqNw9$sJ=ipd7CSOmXit6Gv-7PP;~ z=n}#7L?h~FL+@(dM^riR96n|!Bzd5V>(CHzXP!Y@rvikzn&XxIizJG91fYD~kTD#Z z9kh5QCS=D~(2Yfal!uL*Lk&@OhQsL^3*OV>uM_SL1-kL_4*}2*iKu?sW|8qFP0<>- z?~6nbktSQ9yX!va(@B;+%?$68W^SZCthJYUV#cHfgRzQsDkXjhdFR(0VE1VAyf|chT2>;DOTG#fCMR^XX?@FX2QlJ?ZR~I#3au`o_26(#r zmD$)GMT*V}CVr#IwQrskP#~uUfm`~vdipd4Sr+f{RA_)0u3TZRP#{)rOGpa_2Oax7 zp8OdS-Cs(s|Cob;v?t#<``?V7F^LwA52f+gQVstwc}s5w(GcC~Ciz(I3*!N#_^AA& zJO~wSE*AexA9oqC0UQI*o27<~w97KJ@JenHZk6@}h4`=TRi=-+ID+ic36Vf4 zb1RM^DcvaLvb)32k?G=%;1CX!M{eVVgh^xiFi==(yfPMzsp|5=p9(FwLCJjw%1`bP z@Mf^vjef6334y#{|6LW8@OlXEY-S?Ww)_GsvDrueS4y7rnmtezR^uE-+Ttph|Tkj5a!R0gYa}l zBW6gTw6@Fmm>L$tjT$u?lzSEhfU+D1STj%|K$DF!dU*r;9c_{1G;a}rQXg$U4D-!8m{MHPT~l zd7J_baLJPwW{?b&JXEY6kGDRQsEpIctal+-JV;0`utKt4nzTEv1bkv$qvlTkqI0sy zlw3-vc)4pga~FhlNwHH<)^nx-6}8+;&EP;!pnaI2SiYf3vVE89@2}!P{g&yt*@FqL z*m&JMk$mH1RaYs@UhXsxxHd{#`}u06k{zDYci7ep?3hno`y9KrwaFF8`udxHc`o?e zfjeBSk|#$td`L?jvJYN_>AmN%qd@b$N7A5EepAR6V6q&C%Q;Qa(wZ& zx02d8-fq@IbQr`irTx1vv4v2&_Q(&=LU9dN!8t`i4i) z2xbk$up>YB^LTy=W)(#2@x9z%*T?T28Vx+i0XwCEIPeu>3$G5#z|n#Bi9FU|K2dEs zZqV68uLD3%ZW~o(A3Ol67AuH{b}sQ1R1k9Nb|wZRSRxia@W>5+&NH{%^I~AnmZ$fP znRvJfH<~Kx0@!8FuXEzJ-XU0O!)3E2X8@OWo=G!l@1S6wxfwgqRsyXLf1xA;dC@@s zo*XD;aMrf@R+6TK@j(t%uA&V`|89L~^K%2P>f@@SFt;Nb!B@^*@9`airN^O&+LjN$ z%RzU-TfA>qeR5hWogLF=xVmk9cf!$AN$WN-3C`d#d1(ykclrF& z3FhPy(`pKNJoo%1iFo+Ui=7SJ4-^mCVTLGZjTU)$FMJX6u~%O2vRKmt;1~GOEca%} zu4LqEt@`l3;nFvEjrIa5><9k5O%DFK9th^6iyD*vq7H4xjp!E2x29^!4@3+MCY$^n z$zPrGO&a1aM{)AUG~^D1xc6{>fZ|CWcjM<|WkA=^rV3M6s(+6H ztq)Oe+2M1fv?WN_mL@nfs4L}l?^UXM8EESGPdMiJMx*WvR5*P7!m|#wBm!a-D{R1I z-i|8dj<3(A4I##{oD0xk;K!v`s1#TVkbFUUYqPk?&vy5h)(;g;@Ob#~0XvefIl!bl z)YH2`v2BwAu$YcV5uE_%=T^fhC-D4dqQ7pKc_V0wCP#Ae5g=x?w|yy>37<=qd@ax; z5=IZn(Z2eqY4Ro7FtsRboTazpU5IMmOw8{NU=Uyn;aq1KH`x6dSKsXHI3k%3pISB! z7;puTvjt12c*xU*US-jN92<@~_*JHkn-ZZ(%6o4L+KvUAycisXL-L`#r5L z?j4>NJav}Z2@^~E=(`&W9Ez7%XZ=OQL?jM37{kQ-UOb*ym6dn>WG-R--b8Y;_x!`2 zNPbHqOnqw96R~!VEY@Ep-S8&q1-~?;xAl&aDSq8i&>-SdZbu=!k}zoA2igX>4SUQ< z{`JIo#+pW!97wI@wBc5&lzSpC-k8_PptA_zyy9eDh9El!!eG)`_gV9K!c+^1gcp<$wy}xJi zLJfjR8gHGb0@u~EH~C$W(NLcat%zJ3OjQ@h#|r7q^zePU(EP%{iwx@ZF{Equ_SguI z6USYDsa~-q@b=eiaxq^yEkb<&sAt>i5m-aFj{w;T(hASj}Dmy)$RJ&DFb{9!DPd*J^I>3WI$lu1m$E!um z&YDL6P+A<5k9tUn6DY5a2mNB8pOg*C@VP5Dq=^83>q|kT+Muql!`WvYbPD90e?C=g zIVdo{JSf}VeGM=5QTdha^J+zRigSfX-RY}m$XQaS$D_apFVgWdyX)&(VQ>j3;|Kr( zqI~XYY^m?;yF%DE6ct{iEuJp!ddo|OXs(6|>E{5t*aV=8n6&$5^1dAK$(<2K_-*`n zvh;?gTM;%8GA6sa*O=t8pat|lxfHu^OU|VBBNGZu$v?<+zW~vyel=Dzp-o-iLURiW z@}mTL@Hu>F9spsN1ePm&NZaN6w)oJ&NnBb{lLZvWiveUy7t4GEqO4F`}x>M9#Ie7xB|2z)4}b4 zaLZgFi2g=v?|P#;?Jev9+|!}^jxm!xucA=ANU@T|M{m%d$jxb>r5=dbDo>?0uBVE2 zZ#56|`$a7Cu-zt47)A)Q)g>y~c)QnVs?e`JBa*tLtndD9>z0$*(AzNni_)6*jAY2> zy&es7gTwOyf>Qqz6n7PF4_F>s{@cMRv-zMiVi;3n3`{%ys(`} zR_BcEh9;cu&vI>mBV3oeZx>dsBj^hdq}5dnG^TebM4#Pj0A>?Tn!x^1aYrG%k$wg0$cB(& zn+>a7Rxqi~JNkSL9v4>$2)qcY5Ts(5y30KGV;;r}NLX0;S7&PriUuee?h(C5O&t|% z%W3obVoHszFK@wTduCTJJf3(YaJ6C}Gy)-nb13(hhP-!0W8R$G7w)l&(7g6uPS%G_ zW)tFW$5cERW^-sz0N%+wJgqM#pcK_YXZev1^QHfC=#oYd=twtm9T}E%-Fmoi@~muW z0Zrcd>g7=1FFjTUTE|g4<5q#ow>es9?NV(vTRd%AjBP7c~EM z_XuCwRf!&w{6tSI`yO*Oo(K{`oq(L$?^^^}Hn@5W^*h?Gc{8pn_#+uVF z$|aw^PMZ<;RdRXbWwmwB3ld*2Oxpi`Y9rQ04~^QXYb z!#kb#(7|^!DCh5XEA=0ltzHeIyf9Z{B_#*+zQUH~Ru^&UB?I-;gT2s?fI0xBY3|WI z!HYrn&nCwas!v*uZTMq*!s+n&gxZX7yY2z!es#`pcq@NmM1Kd>mUB@4<2tR25hKus z1-||Qb+=7KzY(sq*Ovzgk3B$mGEKJH%f0sHFjof#5av^=+u{Y!aDtzs97H;1m1d3y z(U2Tj#W*K@;7T@YY4fdAYv^a#PeI}S$bbV#P(up=7iyi;)4b4Iph0S z#S&x>7a(@hb;dC$#nlQxB?>u}qsNP{T4JDv8Phn)6OdFhT`tPXTGxmC?(;fG1>pJr zN$9%$QLB>T$k$79&}%}%%;PXr7_Df@UP)O?lve`Xd>2y5V&$5Hf?TVW2$s5TLrLy# zvT8V&c8yt_PW%9c;mAYj9dsaipzwNxm z2!V_n#jmRTg(qP`W9XZcgZl`BWk?VQ`S2(wb?MhjJfn-52(!}D&F~gqKs08clFvpc zSxc#%U&K1UrC3}~Yy@!hNE~jsKlK~wTwesz-lu(S+14Xpefz^2#(EI@%tW#wM*oOr z;EgEO75uotSQ~zjgbMLZ8;QrfyxsE2wqY5GM~M zG8geH2)^He&eEpB26R8#Q)aC-IBzYV`UPP+3C}{9bPD#7W~G$sfDUVQXc3w4L{28_ z%>^Jz)Xr=Jp68>LaGxjg`jYbPQHDh;lX_+xAiK*8GBR)w}F5ZbtixUhO)eXr^V}5 z!6@)gF?GCac!5D%#s=PD=Yap=U#K~uKtEtW28WfS^PrSuk45|3$8l{rkRHETkSh$- z2^xWPG++vv2<9WJ_Ad6do4R45>iNnXu1#uWfz=(Dn?TPv{Yg7)`1DrX2~}aQ&|P)n z8VE2{*vgReMI*m|nlxNN`4<`fcx^zRbB(6=V6a!V)2Q=dMb7hTee(@pKNm)8Ku*y` z(E6X3O}b#lBbE>x#NB_iTLD)$5HU~RTS@i%m6ZoGIn+c3D^Qf4m0&P6!sC1b`tWC* z38$XV-RK8`OWm)|)s1e>_cBV~XK{5Bm|H)_n!i9|4&{{YZPf)^od0se2&^A$p~Tg} zlP#dFD_B+VyE$rZ6lg#|rzi>G>|bSw{ARJie`1WrFNPh2QZf1 z15zS9Zav-EGt-?XDyybr05p`?&+@A_jPR`v(&l6sE?2T|mqdvyz>aJrl7824{Yd>41mm4PrDx$;Us zhxlGx6-rU15+14Y3KP-Kgr*&au!1TRg~Y=2-+E}Vt9Fk%LH`ag$;h4dLVI&o1<@4z<`yva1ufq z*`~sHkcIQOU-t~US%>)qCiL+jHLQX;t2UtRaDE7E5!^TwZ+b9Yq0hj?d+NC*5MjI||vz$D#&4BuT0kaaE-{Vq&Si5|1%Ts1R2E7lId*a#$S3!F3 z9z{l3F=4+Hke7LZ`8eBkLA7jg)4?Wioj0%wI%&B%%X4GTc(21hVY%vHX7`z3wqOLF z@g8FjI_Zw6oi`3#K>LmeBAOzKO~M~$rE4u#LO<9kT~?j)dCz7QRV_5|&?LtRdjSml z6DgygCi0+)jJ1kYQ$2H$deXS11_Q_mK}7CZCcCI+t0Z=JE*WBkW5~G=E<(*Oc-2073ApKq?~By3xc}WgQ@;RIYg0D2 zJ)1kvBNm|G7;*WLRmoJ@WxR%r5$7iPw*JK7+(DVIq0)zy7qRZ$m^Kf>sV0}J0A?`nN!DztXeymX z-prVi^J9if&R}m590PAJ@+B-_8L^^Rg0ineQ2N1q*QYsg>F)iogBqy-T=)i8n$kNH zt%T__?J`BqLD^;<5ASlqdu*nQmzTe16kd_H`Lie#^~wBAWic!8{mZ&PKrci@%)xCx z(5JzmUQo-5Fz`)-h=gRQD(gJ~=kLcBc=kpy8AxdJONylCRWg82IjxhjEb2gGQnEMI z=QUD*BY|@Jfdro=()p7H6K5N;#U9^KeIge9o-;}0X{uAWg8WhXc@Q3C_WRC{p#C}( zlM8H>)aKy0avC;Pa{r8X)cp7o$PPr3kKtI7<2EEoC*ICyo8K{T6SbX)$5akPT+e2U z^2v4C!(R%JjnX>XHL4ED*k#+(>_#jytW0Fu?{{431LQrVJ6P`T9T(Hw%Lj5iWblN6 zRL?3h@Vb4&D@%MR9}pIVmB8~0wf%I)*svGVdrRy-9_g!pdYkWmK@o-s=ezPaV*F%| zM;~FJ*r=yS4Zb6}ioN4YKJFL#P=9zLa61@u3TjeDa`A_{KUlYR zlM51|C@-R9=A#x;H1JQrUhTkCUJK-igJg3W#`GEx))WczN3oms_(E#pxjz4T2It9- z{zn9rxkkJe%Bja!6Z@YH>=7e^Z(}fmrQq>pNgn6QFb6IWOc{x$^5q(6MKpAna^!!|&Q6MY0*CbhF`u_2Xk|e%z z4H$uPgHWDzw!D;&&F{D``~(s)&vQ~PAqL`_%aA=ynAG7%j~#^HG>T<`gR$nxxI)$17 zSakDM3Fs zF)PH(^Nd|_gB$jpXv+mHG6~~N>4?pf&@4biKH#Z6&8OZ={*e!|z&suFat*Y`;3e8J z@U#am_@|?6j#zIcT+vUmW_Q??o^tM=i@LBKGe+k+UVPjfGz}LKD^ovFMLTL z&oArlq+WJeN@dDNe{FF>=7X4c=0$n|zqb?g30~~i#qxZ~1Mqgdjt^2`->7>h-gGYf zjCp|$__T{khrUnb)LR6*D%9O^&#P&Ei7y!W;mQUvs{eV|_W&I2UzM>V(ASu|9rAxY z!Ch9*l{zI5F<#)5^>cXwWdr(R-sLR2bt|XL7Yat%FHE-QTmJap9nlidRS>_9HVrk$ zrCRgp1}&HZ2w^Hi*!+5(p%p%?8hO0d&#=y9nriZkny(jX6S55T4O%+4y8K0KC#cDf{$o;ztD7S8z!gVRzFT zNaJR9bQh;YAe*aC4grSjx# zM(;mzSW)&}5ErA|WEFbFf^n}H!PL}%IY>Xas-#w-t5!VlKNtBPd;1a5X{YCpV}922 zXt{c~B>DaYz4y(YWZ|0&YBHvJ&lUy~ZO{Y{DS&QYDO1OG_qT5&p|ksar8B($>o#Wh zwe^HoaP|rw^V}O>@r4O z7)JyiH9L@7fTjSAgA`vbNPZqso!nO<{iQ<5yFlstIhE>`+>S>KTM?XHZTmU|^uZ!k zyvlp+IVQegH&TIRk_RmkP;K$(Ch@VuvFe0Ztnz|wT5K_^Gpr3 z*SDdPS4s|TzJT{s1Y+SHqVk?6ggyX|OuOzQzi>Txg_#egmLyCmy()BIaHGhjPSgri zRWnx>QuWKd{@Uiz#R9rZ5^uDn$FE(+?k%>FA%BUn3`qB*JyIyNftivUS%RKys~r>U zksXu=DV^y#@#m*JRKf?`ffY> z6}_vT8^$jII2zM;3A<2ky}ijo*h-YemAO@Xr%N*R8_zmyYZ@V*_es0WY!&oA> z7n8wma9;=1k~Fjh^2z5cmh)}9pl6~`^p=VpjZRtjKHJb;`y2?Wk07ytYhZJDZ9rfL zQ#xHeADzLbvu+k2Jns-pXBFnEI@VGNbf1LNa){8uUDw}zd2R}ID~xXun;)%5rzcl_ zbSD-8G$wmo6ds8WM7K$U%(FZVJth+M!Tx>aw`MBInXP1Cl+>K#Zd-e;K=Bw}wYpxO z9S{;hg}FCHXlSMmgV*P(>5lihInFv%pXULaY<^LhU795rvYK03P)30k;r-n1{wWbQ z;nToiSv8BMMY?%`^ygTaBy<6y0(c1MNWgdejwGApc%qepDC0p>NTWXe{-!;nVe8Nr z@JN=l2K5M`E-9Orjv_|;Jq&5S`t0=2cwdj&gBxm+&dH{I?1I!E;A&BrV_vEO%VX~R z{S~p_w3{zXa8xen#XxkkNg3P^*VL$MEW117P)hm6TA#K#p=w}U?uD9xCxz4YucLrs zU}zrHbMQARc~&8we}6T^o%u!Kv!faAm?NP1>(yW`BYBzz*SpQ#Y}JJq3yqA)&I;Nz7!J zr!eR1(tZ@$^X$~Ho^*xw{EQtYzB7*|?VB04H8^j|QdJkcp2Zl6_CVb$;~@_BE9(7C z(LnEK5y5?bZ}~dGUY{~#1U4y|Emr&FT}hAF62LP2-7$L=FN~xg**Q25y1X1g{S0r5 z02IqTsT^(S8%{4RB@k$N4QlV$3RsB{-S5U5DA~84X^Ut?(PDg7=kvbmCT#jS{Wrjm zxTK`NqR!r(*el>sIx(>Ot-z8f&LbJo=cl_|!p^*JSz+>qyzrE5&1`VwxG8-*%OR1A zMg`E~FLqzE%{!i*1L&dKz1llf*Mrwn*b5`dF|~l^sej6#sN1&KX)+Y38t{Yqs{H$A zdk%|ZzBzOR4%j5>@v3WM^;DM6;*HQe!-lTF074F>N$ocS$m}1Y{V^Ls#(4TfwpaKOql%bwNpDJf+R~zy=_K8lso2oWC z$9z?e=+2F@WJVsD(#KGm^BN?VgZJEVl|@)6Wx#28PZI|%g=eilHiiqbAG~Y1zjT9~ zU|wXbqX7kPJmom~63C8Xj*IMCntB6!#txd!=STDafbBDfaVKGZ7mh9}6UYSvFA1#W zJ9Qbpf|lyQi2b8Z>KFcIs3Vve1p{0!7$4qXy-q)QeEc)x7wp4<{Vu^CxC`74Bkb7m zs;4lko%SRx8vwycCBy~wlnl;RwYk5ht6@;zH_pBv%L^xFrFm z{dw(F(u8_0Db3^Rh>tx4uK4N&H_5*=F*~{w8r}43;CMBNB8o7{Ej*j{nDtIEiV*V3 z-745d5+XMduJ4l<0*;yUE09L4RB-|Sn^NeBszUzL0Unb%)8JIaQimN4V}$8}TqfHG zjJg-`oS~NTj>VTeo)F5QCSx3|!0|Bx|H{-q69YXkvsgdJJ(gfnBL7TEq{?MsKOVT{ zmbc4O-oe2}2N!I?6sY0XAII?O2*j}eIPIZgpDxdCKt`ja0In=AqH|CT7{Eby6LRKX z*qE`PF;pjUer~*TbFYq!fO7A(9czmL>BH_1rg*UH@M=oO6Tr1p&Dv;d<;%^NDyNoTLP=QHe9qx=yl^%Af;eOTyLdUT5HL#cB#CYwKeFMgQCyGAvleF>(-O;7|&tNx|c zN}j>5)BRKe&0rRYMWWmk_@Z&fe)SE8ra5)*-+(6<)ShusVZ~a4zRlt}y?wiGPI@Nr zYq))IpQAux3;5VU-Y6XN94YX>67ylvR}J2A032t3V3IJcT%~=J8}sIKxL>#<#rQS) z4P8308_8dw?SgGJ=K-!W*a7FMhkodaVU<$CTq7b2pbdk4T8Y8tUxL7!w;_1SVrhEf zM&K?pSicJ2`O}Ej7@QCT_s0ib={tEVn6C40i-=(}rPS~F{eD4%<@^Z{>OK2GISsW5 zEcgRm9t)jG-K~s5X{pdWLf`T3_Q$S5)b+WHv?lx&{hC6X!Rgs)%ykv+$s;FVmszXs&I z8FbfT@P|Icz>ez3-DVB3bhX--zNyS{OJig zn;Bof=E`Emc|7NCv)SL;$bM7sGkrfC>5;beeFTad6JYJ2=Hn#xQGg&Inj`k$F!q zhLYc)Gj<&=4^Z6G_Z9y2bdQag$8THqpgdWB#=pCrK0r%H^~TN^3Ol<`(EYTqE&&I= z2%d}bWrTHlPp*6b5vk8iIBRAG%lw|v9drV9Zu4C-oFL8qs19quwJgGUo5bh46`Kfl z+DUDswCo9*+W76pj!^{~FykOWRoJd2f5UeIuJ4gh1n{+WYOI9{CPFVoGG+ zU3SXoYf=~#MU1y9)PD_RD@It=J{hprOk^knhYTWm1$ul0Kf`CnkAl8p${(Q zfr7=ng#8_aW^VZYUL*t5{cHW} z*qlCtue$|bz{}5PxbxNKox+kn%YkzExE&xVL`mm>b!=l373=zd0bina&e|s!Vjn~m z?*hMBrSp)@rrbVWjQzSrk8!Xf)_#U18=ma~<~Szf>)+f!vtESxz%mZ{@zUbGs7`CU9TqdFE`NQs+`BfxV<-gU##tJ&IkgSc9f_>4s_;Hgcv3;)UnIfL}E{Ru^q?y#T`v@j<=PiN5G1~QZPNhN>i}f6g;*j(10(Ra2~(*AC|R1K%-iKx_;hWuNzc7_ z^3`<}ydPAXB2k5f)tdJBj|J{j3+cMbA`#IeRdPD27<H~aR5z>KTO9MU9?awibgE~$VZ7@$B zSwpwa`o+x;Uji-5t6s~oLe5C2}V$jjFcyta4vM zLFEgpcJn@2qzq$vWwO=f>RyCp<-R)Y!TL<2sIBds^b=)Vn`OXimkS)huEqgXFIGM1 zr`vAsl2siut8ic^@($KI>vs&@cKM_%7j_#{iZl6MTlgv+zHJ+hx9qW5t$Pu>-ZL}r z$|dIh*6ZPLOkqXbLFyQ+gxRg<%ZBAuSNn#6D9WZffT9N686V!|5{oaBJkVCi4eZuw zQ;9u4pjk7iC{%7T=zs5)9c81JXAjju{)WhhG`=2W(D&EI%4t6y<95_RWc7#R9r;tQ z74{*(T9uj?ZlY)K({Q(v=eNGy-&_6+{%HGnOJ(KBBTlN^e>s@$ z!4BG%0*D$$@6TNB58#)8DtsR!zJRrpU_%N{F^=ww$ieO!r-Pj?T~DUT^6|v#@_9!} zm(MV5Y;r;FVPgte8)*OV73kvg2hcBWRm1Y_0TzUL+6!cs z!85-Dw2n)hH$U0G<+ZQ!63f*eXwmN(xKwZNv8AZzw<&9QqE_~sP5waovVw2FyR84V z9qZ}@>d6)RN+ss<_Y!7stkWHd#K5u`p|n;oZ|l!7{irPTS{Gag7yL)P2WgPoNri4; zS;5|0#dReQMnoXP${VY)F73EfF~Y9E>we`vQKdeZ1{Oi&B6tUXTuB)9qtg=5>mnWw2E+4-J&_C6z#Vuv zz+5rLb>INx0m*c;NeWo9Zzk-fRylv1Kxbs4p!Sree$^dQ4R(g1cw_Q0*h3K8uTtqU zf;Jx+A3S;?{_Ct#*4E&*_zL%~RkmN7^~&rCs67c(j@z(FM}$p7hTXHjyP@QN_ z&I?)lb5g=~h|%9(_7DB%6ojq8y0pVuZHh3?C<2~Spb9TpKk(&oswD9sQiT1jYnYFL z?&C(00yY-8!5B5={?+~bBj1SZ;hnbqcY=Zz7NB+E#g}+9u~{=EzBxGaCTFh4?($51 z3Kk+XV+!OWoJ zfk{$?Eygum%Taz|U$7)11kVS1u$8Xfp~0f>m?Mu~yhMuZ&U z^?1|27FOS!X|~DqbqqAEujR}3A5CYnttys9(GQ{lQcBYb5D*Zh>`5gcHHtvz{(3H7 zjBMl>IWi(&ym(jPoW0kYGwyd~u@UIER1H!HY9Im$a^lTK`)c?Uw^k7Ho_yeKR+(au zWX9nD)iCFMp|sBFX>ZxH#AS6DeszjwxhEur0``!R?lCjZcW{|-pxOwt7*qM**4)qD zc@94~`BE?kKZI|hT?|I!N*J36LFxsk22hAlG40y zzPod{^UG*58p8;dP>S%{D`(+R+4#mNh&sh5%Hh-M46vW}$MITY>iqV%GSOCI+OyFv z%x|wbyh(WQ+YxD>XT-CG>yQ?^^W9?Uu)%k{Lg}LBxRetaq`!`#r6Q{z4i>_5?>lyd zSe;C6Y2D=(^D*gAxv~Qq`{AfnoPSEerDM% z@sL2HB;v39SSce#v#lyJ511lMxiOE?9)q@-{aq3!+hrLB^x5(z#ScUH>2WqAnw&skn9JvY zjU5dnr!U((jq8hnQvH>hQ8L5^%MQ=AO6G@~s!L3xCD+ZU)X6O3$Hp=9kO5S-U8rdq zQ@be4kfIHf=@70@0)h0fy}AFzYiwRgS0PoQ8Ir4C)Xv>@E~mo3_6wG`ScXz5Jvi=} zGWK6<6qHQfp;WF2@|yA2W78b@9T-_gxDoD$GjRR^v)PaTaDu-l<3|FkZ+nRkR}3fj zjs|!yh8tvm_eTkPM}4Oi1TBJ9+@Lw9;U>u*GiPJtU}45@fe+AVNh<6rOxOjqyZl+g zd)abiN!`^yVxqVSp;yx`Pi}|EbI;v+)x!$FKy>E%IxER;e>}!d{AN`0ju^g)>x92`u3-bxxG{k|)P7;*1?$j_IT->H^r{_g1xLif#U{?c zga4tW`#kd4ST-&>Al)3AK2fZ9-cc;Dzm@$HMu~k{tmG0-RI%|2FUtfe1;;#Aj-*Jf zq>-=WHgIi_M?Zi7P26ac!X*=Eplc&n6cl=UD5w2=rHlKt@InP*Ju97D8BBQs{g1)m zCa%QLyWYIE0snm$@7uq@w$c;eu`S>u^WjSLV&5 z3s(h8xm!9a6uNNk@=PBdXFx~tlSVu@&*GIJ@Is`{W6L7HbimB=Ov$Gkuy$Sr{Uqt`UQOu#V+8Z^ zy!UD*M;)kGjmXNgXW)$|?b_g~R4ejTxv_q;KQ@`MCOcpJk>JuWz}RekisWc`QGQl0 zddj@n^h-*#o}UIn+UxQXU%LBpBF|iTJZ!1t2A=4!o!RX# zEk}NNL{xFH=&&~#QN;Ia<3oTIY>~_#hz9GWxo|pfC237l>UI%k{3 zSiKWmX0v~LfjTShU5O_0Tj@OXeBP_cBN9GndU7LNveF#k|ktWM=azIgZSMP8l!NK0UzzVH(=BTBzmcu99e^9zZt z>BYd%1U;=!A3rS#T@UHh7A-QviER_{TNY05OL!i3@@^Ftuq*aD2Xg`u5$jF^C=2kJ zab4&n&Ybw^e=_mg{B@0Tqn)1@ z44#x-7(=&IF1vxjxAU3kxwg!(VG*I;jIfE zqJfBdunGb&0~nq zn#5fO$qNqCI@W3o$qEYxRa zdKw8Skj)+bV~&~N#R*YI_{5!0gVX*ZUki_B6cfwe;%{W~gS}ZM?_luq~wL>B(L-tYfyyc&d%#^a%3kl>8kMc5V3w4xFIdlpFkj(xZp( zqgnB4r{)?(=m!ltBFi40zimcc6PcJSv0>aaz3+zh%By$ssY&m!h=1y* zTi6sx#W30$Y>T4-#^BCR3Ch~X1abEUf6FaNKns98pakO?(NV+`cbN8v)3X9}k*6YS z46nK-(aLm@j=e-)`@`q<*01Ov^jf6-xgoP+qb)zL+hpVy$?vfJHJoUTx;y#2cw)_5)1K(`&#Va zK{$tzd=+94A?xjQ#y46fa z1<2)6@Epm`t}>nD=ZVnfFcW0tXci^jlFK&2Hpb3p!26oW=o`PL*SAvcW^*inn`Q%@ zLTQXPk%fKT)aMQth4$}6Mvpe?F=oSicU$l2S8SCIrS2`!xVLCYAJvg7uPRz3%T*+C zvELo$o3R5d4r4*8$N+3NS#k&>H#jDz+t;0e$v!W3h-cI7X6Vi*S#NiJcjLlpQ?;Cj zsoL?pyI+&}GtPg59u3(Yxb3&dz-e?ArE6_Jc7%T0_1-WoqVy}EbENZ4J>%5SW?CLk z{6-<(H#1#|DY?Ig(p+jnn9>V(Yu{*C>FIq~zlhkl?l!nbd*+2p>sCD&A72T?kh@N9 z)!-+*7pgz)GO-VgTNBJ-D8)g|EHuY*K@N6t$$XmD@SjzU_Gnzgs7abs@j{O7iqEkF zO78r95B8J$w1=eIwc*7W$)TJdpgzty*l`^u7X$Cxhqywl$lU~78nLH~C0dTk4uD$! zIh^`)P2j7QPiC!96*iQ|&QP?bvoI*QMlw*+o8~0=;evA<;L{(&1{FKkh}&Kr3+nzI z@^{QTmo3<18muEY`3x($7WR?6Rb}G?G}si)`L)|uY-ERnzg#9Pwy@-WQ_J7mFXHsz zOGD7v-^*z~nJtU9R3ROPCQ#?COG6EV2IBKs&(`N-iXvj2W90h}S7azg12I_$YZkqP z@GD-M%(;RgLCYyXMZv3V_;bRAl7JNm>OBqjyor2Z+k;0s^1rt!_={^P+#SM*XP!eV zkH=Mb^1jc4pLwt6J5R}Q010&xsRULMN*gWPVk{1;{sA-tF|y}o%Y{R`-qBZ|PZuo1 z{n{ty>g7FDz@7}@f@Tm2ZWG`*<~YOLw3=oPhen~evh2&+W=gqM-v%#8% zp}|Tf$z>~_cIx?S#DMnVy+6Pvl%eDJMa>+KKK#rN;|rdg?XEz8p=a_K=0+0LtRgBrh1Uyf5&icAeIRA0WAbPsnJ!elTT^cN1iv(>J<6DQx(dVs1T2e>7} z<{GRl#*0RObx1!HLTAiy_!05gJ zo{P*!Ni-wB)9yE;`$HN1{=#$-`7NW|1lNf4U2VrLMEJ{WRpR@4b};<9?C;;YJuH4U zbE5M6v0T1eHbAGjQ(%aK6y}ZkDW2M+9;86|Yeh#sL~Ge+J$V%N=4uL;`cQ%N-6$*ftyCkv_8ubLGfhaQNTMhfzU{ z+#(;Ie{e6b_1`MRV%rg^le-Ik`-Cl-GGw}aIQxf9SuC3iYU3wo`;Lp{c$4Hi=HWgE zkx2ii=|~V7XRrW-*9LVrYZrpLNf*zG>fZ?UaB$F6NrWFS^ZH)4WDP)KXWwppQxh9{ zCwT>AP_|;Q!dI#2jj)k=lptLEV)16n+OEVEBWYfaByrhrUl62o8UW!BrbGAB#6A8x z0#EZ#L&%i3DPv2P3yZpko9)rlUg4rYh4V6{Ep-1 z=`<&d)_r~RiMp{o8r$0k;os&VU8`4fgw>9`2onRE_%J?~pI7L3;49v2sP0cTz%&}u z&2J-hhhO3Ki+%lTH@?oJs`m5#Ic@-_H@G8390>rir#tQ`3}-wsjI!**dWf%MYZA_t`D<{S5Ed$aWA$HWxm`?~y14r)XZ z28C?-nA9WIZ6<|H$jiH_UldAD_$P^3j1`vF10#yo>aD=;>ur35SUIIP)wU%zc^J-P z0D;@>_xMI(5J4{p^mQFaZY;ahZ5u%w?L-xSy+^|XWK<(}=Sd~BZ^djh1@GQtO$t%s z-6XhIo13qtNeitIQWVhN-nq{@wzU|z0raH9bfs|lCHBbt2__t#N8U2v7pFM@CEO-T zQ*;tB=Ls8J?Y$X!l}L*&t>DUjUMu*{l*m z6qtRVuxb}6x(t5-;`uxByj6+OcoH^lnb9AG($A9*(nA;TpVmh!>l6Ete=xaQoS<2+ zcp21K-1;KGy<|9Ho#>$ES~>&aMvfQz>wrIk(?W zn{X7i1m1>X(Qd&7*|h9YQpBikAN~BsiO!-ybC?DQVWhM24?5wc6>YK%a)_*d<0*Ky z2A+ujIQ)US-^sCBq^h%;5~NbX^#)I{2WnRXupPH$;P}QwV7YJ1lbpoLlCLbVRqd~uTO3fTIto~wIbmZT!d%r_7Ay5J=SdU_Z#zf zfXz}+rB0cehZGwfHq+&OoNnIY!X61`Y8H`1Uas;QkD&dSjMqKPg-*|Fe*_KG&DMoK zBqeuy8Ek|->?jDkS3Z&@DiN<3P7t6(gi+6#@~w42tPp0*=gG*NYu@W zy)kOj(sn(6C8)J55)esrAwsuBAC74Pfz=9B)2D6YnE6N(D5MgU`vaS zH8LL*8Kcb!*oSR|dl0}?2s!w2-ZwNJKIa0sx5Mi5-p8nZUu-1>Dv%^MG9WcG!6iIm zWA5krHo(2F?&k)Ve&2K*Q1QQ*Vb7!)P1E1VzZdT`_U(CGcz~6_iLg&9@pHHR7K(R_ z-ABXzY%ampM*{u;pxl@8YK)w|AMOi_4a?q8^nweP`CJ4*gJfLQBICTsr<|uJm43CS z)x&I-V@Qmz2YjfFZE~4TyIBRm>hIy)Rf*~yZR6HlhvY~i){ka6KNH?5qDyy}XpN&y z_-vjFcKh$a6PDkZCX@ykYh&~+`>O*g(({7}VCgt;BNZp-dx}dL?l{2*Nh=uO1a!=_ zQk5Y+4^>Q}j$7Qdt2PRMxioP8bV#J4%g(+HsxqN|-yEXX(6t2!5v+D9k}Ut!@FjZn z(JMhN9D3gT1|oztiV;*~SP2-~-ooKb_J@pq@?}gHquvSYtFBAQ1H1qIZsbT85Aji# z`Sj8~5ux~S<^J!@a_e4i`PV$66gY&Zb{CKd{Tn(d)m*2yI-1|SXkv_ju zM;P5xlrI{-1Pl+x9^r+DSCEka8~vgYSKt;f6xx2R>n>yrzN7r;EPO>HIr9jBIxO7A z#x8HTzfG*dtXT8?e2!!ei$Mp`osa<$wJ&!JdcVQ)mdv0N8UMu}F%DL@=_qWevVC6e zk5{x@i0YuVV0{o2`EYNY_p_wJcEf+Y>ISH7vWXK;=ZKs-an#7{mDJO z>#6r~^iYvw%K2)f^;xm=>}-AG9RD9K=N}oIh8U-ZG$@-TDA=4QJ{L>Qr}rNRv+V79 zIp_<4bf@<%Zi4pI9`!5m?yvY-Hs;AZ3v>JbKM|7Lk3F@1^`G(sIqO`inSj3jtT zne%-!oR*22)sIhj56>@;I3tJgD75PNJ`7c{fq<vQYM^7vGZVa)+FjF6^ZCbIDQqWUe+ryFj^KStngNHnF8aXi=;*6&mXLYCT@TPe}TM7OMM~{yPJl z^V)h}%BV>%4o7>s9-y-7md-P?IZMCqTf7u5c$xemvJ{1YFaTkuIw|(cL@B+*0IFU= z!dMMF=i@n|mPJ(XI&Z9<7A)|)bE&H}>Ap zu9{R2h}JpF>xXGn7l3AC9wz6sL>AS*dN&UltZ=ehSCDO}ec0Bb^UDpbH(5W;q6j3K z6*kxXAHxF*D}Do{5W8`Ixqz`)Gt}D-8Pi-5uX&4L| ze*QX^cgA-I5E9BU_5%#+F5oONX}IAYK?(U$v3NS^Q<9HfG1X?NYyc#T|I9)*kd!p0 z=%My9MOQ~`QkRc?KHl!4S+j$~R=c#>-?5c=wbc*B^SuS$I_9Tl|AW01FP?nH71nol z+d}kOJKBk1J)qesllz^hhlli#pm1s5%Jl}8;;L=4n7P$x7G%?(4rq>F+n+VLWtOMp zV01s#_a2P4NAKFbb{;l$`X<@8#grx({g<_U{6=Tnvp+HI9l-GpQ;NEHzIRxYJp<3v zO>p6VynvR`MV!lIJ5+FgY|wAGwlr7-8#jueWbV|8FaN$Y)X18!{IocHT;;{+Xg54) zx&|#LmCd7?02xr%)DooL8MWCy@p|rrs^<*gNIIJR4iHeWCS$ff;$}NaorA-Pg7w-9 zd$zDULL9LfM0xjxv<@Ki_)<;>lAKZtT6TSwstDx-`k!e%`}y{(zZ zMism6h&@eU|6@X*3;9X5`YkcOX8!kmo#r2(UAR)auli2wF1}fh(*Zw|y_9d=A)A`l z>>RlcPranpPA?>+1#cpm^XnCW%Z7>Po1^G$WqJ7~2|X*DZY3?F7>yt?SWuV|tOlcZ z8%Ugm6r|>DiFwgf@0^BQasL4+2yHLgTbq31#Dtt!5oxTa+1T1HZuTLz&j@vRvFh&U z3wo;{VCpH6u)-ZA7JNZ59$QZ-;BL>iS*4h25!I`I|2ND^6rivY4(K5u(XB0Bs`pkf zi}H$eFi9<6lKt~JG3E~B_T|1IXo@WSz#l!DU>k4PvuXJ~lb+phS!1Afkk6fo*cPh0 z=axU3I|mgHzZ1hw-!co-n`i;HR|a3r8$Ax@%UU8jv?HP2+^S<&&M|V1k2+S?PkLQ*e_MjeSWhArd z0Nz+wgn>}Wl*w5phP#Y7SgB~}INJBQ2pxkaL*Z7@aCg|h1sDpVTa-|mWWhYPfB~89 zON2mlaR{+{xk#1IW`^@N27nuTQg5sv1OUs=c4l`4CnGflHxdQ=5P!b`zWb3!`CtR- zawHAtnd%J0d^f z>U6VvA0qE0O`=$Cs|P5D;{G`_l7T`Q@eM&t2l7VOZKue5%(~x+nm_jIJ2|6~inxYU zT{pZ_0__Su?&lj*^^3tx_dLGG-CMGDn1477N)&)=tt7rt^qu-HFxY%p-(e3vYT^L} z34u)JCpzo%^95x1)3P?R{??yUb3bgc5um;kpDRNoOfr8QmZsj}#BMaOU$|0XNHjM| z3u1|!^SD2v;o1v!QMSgnC0G~z%g;`*om1*~`$XE5TO7(S5rUIS{I<@K)5|&E_u@9s6V zW9G9PRz>aN=8N#F9TBG9DL`g?3T_gF>rCbE6SeHZE$EVnocZDH62{?W;(-Mf6TB5( zTikajJl_XL6aFT;YEtgw6Xt$gK1!(od~hZCTNbf}Ro{Nsd*)UIeIsvPvIjL4wmBb* zj4}&{A&IO#MOj`VZu4AcF|r_h5h1$l8a%LWL1K4i*Vf|$fR5O#4==9wb4U3?Tzf<{ z|BMI2Uuih=#q}&6E%jLc>;+3lnD3No(}THLFYb=h?DsVuThdA`Cf{}1>vbAVInH}c z7Rrs7zzKcbn1v#4@PO(F*m@f*Q_e@36{VvJR0C}wyG@e@?O!2C<9{zt<}e}un5SID zpWFPMmqM<6#B?(pYWXadHjk-^+RmZ3zPIQf0x6sQ`4NJJtS)`rYCp5JfUeYxwNk0tW&BUz~I!6z;f&(1^T|5;Dq zN9+b+rH0e}PMXwxOJG0r20x8W4o!|9URjEtYl~rL=e#krsHZUEH1CfA-Xl+o2`wPx~Y42KilDQk3xPpBcGp`4r z{zr?<)FU}oRu;gQfJWttvFuE-P66nM5391ea+{UB73NVeejI_%oe)+Wx0vH@KWqz9 z$6ngi+Rc^gE`PcdwXUxwtUqru8+}J_xF(c6y`yc@qw0w(nDWb}DItdp|JlJh%-EGe z+d@f${bd=m;I}a)!ud{qS}hU%%H+SiSBM5}6=Mf@)AO@*xqnxfm8Jne9Xm9h zkXWAWK6Qns!45v1+!7TkRMosqwyUu>{p#TgT5y0_=nZ$7v-fVVBoLq)Y$Ko7Z~CVo zJPn1KsN`AxHO)1{R0@^b;}WFtD%xn>Mc#M5ymJQU{&Ws`Un#D)ZAF~!j%dXUM?~e$ zN}2>Oc5<3<{ATjUGcy1Ns|GfT415tn`fB=NrGBTW0~Tj_-lOGpCX$)v{@$Uv)w_{C zd@+Fybyf_CBd!`aS|WZ+J6hw()a#^m_TU1@nkVPu>hCLwa^b!fxoWERNNgb2!JiFY zCb7+9NSGre?SuS*87-2{j!Jf9NYb6QqeT+#W$TxqprMcVIb!~fkHr#WG91q6)~Q^e z%Dcx_JYwpptC}ib!T3*>Nqu0d*aBXWa}Y<*?|>L6=&6M^r_S0b7YfqYHi*Q>=Y8}2 zsroLR?AqUgc1s=1>x(YZj*H9j;qin>~{lCHx8?AC z`Thpgk-SbPkjF=(M$d;ZF3E;J-%HF20yb{{Mrn6=Y|f!_(P#WS%3`Q$CPM_&E~nBY zavbr_$ea?kN4sThRI7XB;4kL^1j6-=E8w+|M$8`-&~)JV`H0o%{9vT$31+4M03a|V z2oj3>H`N)Sv`}-Y_S~R3FDd{b}I)MFSIF1bMl3#o4~_?GcOl?mY~ns|?HQx7%OqlOzp2K^4PCG$M22K>d~ z3nAq<+TnL-4ACEC?t;OBj`+D#+_z7%8ssP7_pb`#e7G26Es?P23p4Zg8o<#P{RV1q zx<_h?b0SuP*ZX=-xe@|$Na|`3Ex`pMTSrJA#}IsiBK$?VfzgS7JYMM|<#@B91!Y5O zaCL6ABr;}Z84Upvj=6(9P^-tm2R(C@?Bt`%8mPSJk;0KYCnqENEy z3EqiBnZe*c7n|oWAlHeii}(HDQ9zEcq~K{DTpDkNLEt~IBxqfG@d;To5nu+mgPrl7* zHPn$W&f(6o232?_(v!0e)Jo-RsaHRTL^f<+m&=G_d6c0j+%`ezVy{W%@$(S`C>!&x2a#~W+TVUuRlr9GVX-2t|X=~Zrq&>A1Driq^>Gq14v zfB_8BQOO9Z2ADLEa54}1fFQ*r(*0fygg=4z5P@th-XC}`|4A*h4=Sw3B*vd_HV;;tzr1XK=RfBBwthXZj zl&#yEG5MtxPp{Ln`^L)cpr5MKd6CR@3;%H7KB*HyIG0H*=$nBOoa>1BonQGGIXjTXP z`_5`86E{yuf3V0Ywn#|4xiu!0fJZ(zpcPwNQ(HHK;D+wuCid-#kY3WL^JKsCnUSCM|zC+o9~xp;JN z*6t1Wa?|w_XvgNQd;6+afxP$%+2;)rqTjAy*+)95Pb4?<)0#536PeBVx}~ng+DfCa z%g+3xn@$68qpvox)We@Da^xK#B%up(=rseu>*1w9J_#@WJ=4#rxj->hVX^e3dTru% z9YScB@X%Y7crhr8-?{OXa7XtCvwRY&jq>!-p%-hQfuYjYu(Z)Wx9JAtv@>x25qOat zaxLPFkfvq-8Ir-2YC#rML`7Fr3g=fU012J^_AW24b4qy=wB0EYr6|5u$-kTz2keIX zE?cuH5Rc`*{0=CgZz51sRwR;oWs5?V;{&-l#CRKe*~@-nsxHnbD7c3o%uH*!Sb_P( z|1*JJLUo}1!p;R&PaF>L=fuYGi-nQhLgr-kMD)4($w3e?#8W{sIuK^emoqsK^FuRh zJsuC?#xNg}&cCi_QG7ot_52#6N6yT+6rl)5A(<}TxlTt+gM8?tas#pPWOSYQXQj)V zaGSFA*nYrt%lmFyMU{>`)5H?>L`6a0%j8bL^cQ0`Znf`Sb^EZB72vJdq2=O-4;Rw1 zL#$z=Evky@`taAg_Hc(5GW)bf=`Hya^z#R0@r>0T`6+AL3^DAoe(e8Rs0~{n6pe3O zMAG`YDCw(F2Z;@q{MvWLUa0YEY6uO0n)DWtNc#qz_thTa^k^k^cR8+4be`S(L-8hUz|@!|np{hi6M z7FEytwksC(cv>e`d)=aWc_5D=FNmzm{c$*ENZlgy@&4(Rko7n3EQlyrGTiUF-!bi= zVZ+}%5yzBnA5XhuXzKciLAKlPJqMg(9XJZUy2XW-++md>mRRY)f<9(Yi$C>;h zf70>?HeHmBk3q9Mk^))NYSnfw94{|E3m?aO@S~+94evf9*7=D|o+nN`WhOXqWh{Ma zT+rVRLQ!<`-2MeTlnlRC4trK;70|D~MNcxnoSD2lO!eO5v<7q}5ZwW1WcI(z-_c>V zW>OQ-pQ3^QqRTCw`qP9EC&Xm7Vv_ zOzw03ttv5f3Bumb#0u*MSI#vBZO1|3qds z_iq7h)Hum;S1Wg%EGvz6&a!~c1nrc5oP!f??|R&RjdFqM(bcN;h4M{pXahs@P2~mm z0QyS6$m~q34zsy^o%{#0LXz2b%u$hL@J2sa>v)5k-pB~5*&jF3$daf-BL^z|IOGlb8;?KCyRmtlKktovzlM%&-8h1B z%=n5T$vn^;TxthJfNyjk9w#0lU{Sl8UwqPp&N>pv$IZ@X745{dLrbB;Xh59&+$aB>Ib}-AvmVnrvpQ+fnkI! z;PL9?;{`H0{op?|OguAg^0In9;V_bNx3ceiFSb->W^5u~9&cofBSEkf zIApvHFZ7RKV??L#CBHN;?HTU47uMT)f&SiCTs{&*3pEK-u$2*L$C(XA^Oq8il$*vq zE>R9Z#~pn_tl88S1xY3c&sD_yLq8~WhJ)1LJYKjBb}!7QBwV*}V65S=+T}lB0`u3h#+{Q0R*=HlOa^Amfo}QJb zpVk?l#rhtbzl=W5N+N0MZnO9IJ-1P!d3_=HQ1x=)~^b=b1&Yi8_Sk?9zgnpTsLKN21nART3}Ug7fxJYa34n@$+D{DYt3iz;uZm1*yn$c6HEkw8==Fl~cpkInjVvID`vaW@ z(FtTQ1uyyHFsQy9FE{QNFUBQS4^PtuoMhmmx3Rr`cBf=hBp`x#h>EX;v!7Dw-fj9J zj_UO({g%^yH)b4pzHrzJAV!gAnl6#?2^9QT%HNwaxDL$Jv`V^1Q2rzStrpBFCqcjbZms()R-bos`3N;hM1o5#p)xV@(k6(S5zVjmWvQ2Z^s`{9~iUiyL&8X`9m z3)A2XeD5wjTEzu$p#;FMPXINIvqc+Gg`=Q?uF@8|=@o%#OG1f!luzj^ag*EGOP9g9e zSU=!2+s3V&+Ma5SwC&S_$8(C#br?$y1BIy7)W^7Pu8440^QJPo4cJ7e(WMC8!>!=? zH!jA(#sZpmnzF>-==Kj<17);MZYwml^?d#N<)H)b{whv>R3=qv8;4W}tPT0pUosx{ zfe8bJt$m1ce!_YV60EZa!4BCrAKznRu5DOo!CyCuzs%e5qcmnlIhNH9rybAN?rLp~ zbHp8>f!bpdhVv6#7|{+B)Bxfr{ck>>KXWatrA6WOiNdg)D%$|z6I3EVKR~$`37Op0 zIQxY2e&64Y5bQl3(oO%IFWuNuXnDPA%EPu&*KEV%*OGl8e#pq-sj#`Zu<|SY6inRA6Mbn7 z>Z9`bwS|FcKiGe8$^CS>Y-(z~L%@4S5!U&-*3BW9_SWdpEX8qyF=y&S}yA(0KL7Fwyxil2@0Zvn_58cssUS4+(bV2R)zC=IhP5 z6lY$5eErk<;=CdOaq5wQF2){Dz4w_{=IgwD?lAGhRypjG+f@}xXuPZr%U!gdpxh{d z^f{)Nm}bvDM%t&{sVAwk&MhCR>)-yPhLMy{gO))DR-<%-7G8srt23bkIEb=M|K*1tD`lQ5@|cQ59;}M%COFh zm~k!99F6>Psr<+DTkpQ#7f|VVbcF;sM9GxM+}0j8?j)Un<@r?ME(u~y1b6n9po!fz zNA9#dJsGB+5VZmS_5l3`(nVjoCsCewci@5?(~}%Dtp*wReU{O#-|_FZy;HG;D(YUv zY8^HdT8THczL~UGgwyACSb9qmQ!e$)z_}ciALorkqoR59J7^J_17aFmvK{e2a>1kvZ^$qI`0L@Ja2)`v6G4t<*sB!n z!nUXB0s}jTAh`3adJ)84RSsKuBn>S3h){*U76=Am!SsHn=L?{J#8zhoW2up}i#70ou-&>8yzJ9HDqhU|c)F4= z)zY1v-zRw^e4cmm($bjebA5eIU&4B47|#4VAkv59FfbDltoT1@5CIFLVt0#w^MN0-lMF9gw{r6oSBCInAHVV&2~NMk z=7`Mhw`f$vTZ=5nCERJ3cY$ForTxiW_paZ^`WifvcKHrK03w2QdsFwE?y0`&3<5eN zM;%hh9p7JdzaNZrnqH%m8;Z()pDCt24+#?ALfJxAZvwzii9094h5kj}_!khRAHfdw z`^*meWR6}4ysh0ZU!ZB<-~XOJF(3MnH^ii?d-ewHN8RmNuEzSfoR{&vbpwX1f>v)I zZ2*P1Oxmzq>!9|fXu460aThKD?ggsmc_#wniHm-epA8QcG~o|aLuShM&}WT_nDdb+ zhg+s=;?DMpuPbeyf0-7gV25OP1+JyWybsgU0w>YjVN;q&^AClNL3SFHr>Hp*pIhz( zwvIoGv%_L; zt4<74#yyvj@F#Il_$lBBKH-|f<`W1Fe%9S2SA5POFqfKCP*AXrL?Ar8F27@#13u5~ z&q1P|x1)(dx#exX1v%&|v&S8s-zjfL+;Di}ZRkydG%n1P-5M-=BK*ZV{Kdf~deyh1yixnDTj&ZZ zEv+%t!K@%Q*wE6B47Qgeqob!%CFGoL*TmV1kFbYd1$`0N9e(J&pU<&#xruG=kHWbg ze!LR@AS90en(o*@RxHYs6*Dtse0VT;)t2mS!>{MR^H-l;fZqy>f0Y8ymfuWA2q-B8 zNvtn!^(OaUK0HLyjvuBN#xR>eQUXTifb)XoUTlu;%Q@9BHhkkyLb+I% z#u~@E_4nb+DTGpv*_rt8A^Kw5V=!PL8UKw@CtM%$K+XEXVqHdK_bBi58+|y*Grm3( z`+Mt{>`v;&`v4H~>q6SSc(D!^D&M zanl1-p%0ifvL$E&MB{7K4PoeZQ-hF@_tmr}@T)l`>>WC&>idRE6UMou>w(7<0>{7o z`-Q2xAA|oJ9Z3CxV#n~8f<(4khA#_bF@6vH4Q1|^Pcd2B_ftRtOcR;bMy;!7E3Fe| zScJn~1mpLxvcGWJAlQcEq?{JATkPA*vq4=$JQ&#?9Y!jNcIXLojsOW)S6lW3ze`n4P8*7xf$vG?_B~E(W*wn4$GFuHRG^Lef7>$M? z7cF4^CZieEr={~s?*zLQyztMzGL4YFj*Yc_96oh1ZD_yLBGw#RUwK%{+C7JWqLS+v;33ldsz@ePt_dzX=qVdu{M^^G|!^@}a+aZ|Vac z$s4B^+gvPnfZE+Xlf%GqAAZn+ksH8s7dso)+{(4m|K9gnpm#T5cK%1wnQSYnWl{8l zXh7`#Cjr? zRhApL$rT_mk|53UeGz+(1vqgWhM%atPDuJQ;aL_3dH4O>UuM#J0*%}vzTIe<(U;6^ za1YmL=--zgs(Xs_0`dGn9ICfv1sizrXS1&_14Ls?nOGCHnpowOpJ%^-!P7QTfr2HHAeeA8y^@GJeZ(xTycxL?V2+zZNDq^< zo{jD%bdH(3AEvB@;PsriaS$Q5)<9N;e{4N%yWffCyC=cJN}R+A2j6p%eA!h;M|Rwf zS(m80(#o*oLC$cOwajZSF0f@tI)b`4Tv=g-MwgLxuVIbf$+aRopa_YHRjS8*{v4sN!7r0;WcpgQxcPSrF6Z){+sNeg)-mb?fcdNrjOPvkOu5=ZaUHv zGWUlq))%O;V{k;Ec9G(~en6!t)?AAB=DD3SU1J3X~cKvQOy2;`C!o`-~Po0yiP8VdR~gP#T>9*Mpa{rT(vU0S&s zg&l{bhkM`gPq16ak%%6^fM?XflCQKm-OJ+}S!f&tV`?rbZ+FBkzDld%^8xVUSkH`c z-2rM#?~n8EywLof281+xtK#QtU`T0m3X8fAKpP#i%(TAySRVFw>TaX|q~gHkPHC)J zJDqD#u>`zv@21=YcDXME=1o+m;cY(aH^0yyXJU2cb$C)-PL?t<-Obdo$HXYA~d8RgunAb z@c?w;w$=AE?SaKjL67;I0@>vI;)q<++Bxz2&@;9UX@KcAn~Ch75!`!wEe2PIMtyUW zo@i~%CUG#}FG8=3t~tMVPFN*#r}W9sWGV|)k!f9(nVJ!tt7Q0^ zc9AO{X`;Grj*xfIb9zRR6kRcWa(EJmZJw^*e3$R;RaUl$uLoRD@5ir-81LNyd^n)oq`9m?D-i>B(J(yRx2=v?bdX(}VRv-^Q z=gG(0=#xJGI>+c0iR$XyDX`}(yM=r{q$Vtq3@BP0n-s@>A0p$IzF;%D^iEd+F#V>m zK%m#$yVqq@e%qh0rMD2Uo;!+w*F*3+9qgLz1oEn1+WI~FZfSn;{~u0ww?sgOpy8dF zG~{2KR@S#Z^5?J^-|X2@sBSAeWTNSGY##VNOMF&Yy(m)=0=B-lx&GC4LIYYI<~jn^ zpbKf(6!7_RROHgA4#7%3HI-Prpg;~T!a{yglaV6y{h)dzkC@x^scYM>T9*s;Z{Vg6 zc$xkFDql$y&cWPx-9U#rHk!-L5D+VzbNJZ6hE+uQjd#xx|L<^TAO2ObkiW?mq9Jio z@G;zguayL;UFhb!_Fz{ArhGb|525p*fOFQK#;4l--M zN0qJE98M_3-Mok+egIQcrVTikb|P_Psj(5k-XufzXm?=fGoJ;|XQoG1rZ45Z>o0Fn zMq2sfP>x4XC?&b;Z&1~(-9@CJ$KAi{`@dTW`hrRmx{eJzQg`O({ZPIXbM1fp%ak&7wR4wo_0t=NA-3^612joumQ9Me2x8CO6N!!UZR@u~y zK~2Q>_h)9HSsAy_9cBc_vF-amaL4wxMz9SW9Q}vk58O|=JHbX~?g1X3$&^8E88r;l zi9<@JPk`!W^yzwAC-D}BxIkZHwk+k_^Ub1D3Q^UUYDf4T$c4%?fdx~D)O)*eDptzd z)yK{pHzxIt*E2pNpT7t6ZOWN377M}5Pxw57P($S7G0Yx9gCDzJBV^6j$2{{AEFyGe zSDTb^a$gY!fZ>b`BiB)uLWTLwA)C?;sT`jX%P}z~eAOyCsTYzdJNUqyK7{t}ONm#h zpBBAe?KQ9dKsDNFubjqM)3uaUl`pf?T@tVR-TRHaFYKUnzxm(Hw5ic~3Uf@u!}UiF z=Iv|AT4gVFJ6sEMG;3h!E<=Y`=|8cNK~tv2H+m2AWHC0ej6X96dkS_6jXTtN#NaBQ z%70Ul9CAN;CMz7Z!GrYyaN99CEwrbP2EQ9rJ~DzR|9I={?uA# z2>I4IUkN<#?dB?3a@0Wrt|bN^mk^_$a()P#%`vM(gQ;>^CmiJEJ3%Q2cqd6$`Gakb zgyI&*q7DUm5t*u+{Cw%(C36T-;C~`+sS_meA&=Ek?JBveEUCPe{`pI9V8vRTs-AtP zBr`;nLY2d&;z$5~9z1X$*^MER2?*DRkQ@EGto-W^WZ{lu%xf>b3FrFvFZYcTu~X#& zb3tw(7x5OvO6`59PNUaZ??-k%412^2W2Zl&quGfm6=^$ryijU<#Nb4mp2EHlx#{3b zh>Y)Kgawj-v6>P%kCn=I=L2tKLN{Ln|13?3Qtaw-L?zXg;os=d+(LxvIq1{C?RFW& zUfeV%Pdnw3$_@PSzKnx)1>+BhTmwkRN;re=>l;HM%F$bDn_UOFOpq(a&7?u|4u(gg zUg|-p@QXMXt@zW?tUQb`a=IkD%Ra{=%Pt&(k&xj1NyA2D)P8XG?9}UE0dr3R7J3!` zMx zJpL80WB{K`OXe_F)E~FO!W9G;qs;{!j@v&QcY20f=0KeiwfbF*`>0+@)KRH^01l3o z!L2&^_c}fOI`m80@CQvApfN#*YTy8+t(<(``$OkKG!Fx}WP;@Zz1pF8#^0~Zo`JdOW%I1wC>EVL z%(IkZth-p^Td4^g=K3>9Vy-x@lY8AxNTY_Xm7>`Zb$p}|D^3KF)`MfPgsCoeF|n}6 z{8H#9!PG|@-CKrG&M)Opb~dagH%xi~um4Ou0jhJH5(J;&&gHQjbEG!XhtMXz(fGx2 zNCrO{)9m2EfOULdz?y`ihVvcqiZ6tuqe3K7d~F|-`8U)}13uqQd=wt`k#Apimz0SU zt7QjCu|9kD_SZ~wuf%!7Z+p;hf7Dm~jtm+tc>s@$m~aaa$#fm`KJr zzk!VH1xaV(cUlhNjo>z@@Xgr4Mua5=`8%lg#LensDnPI8;4 zjWjR%uXf_0-F=a&+RtUE7=!w8)2{G}UCc?ci%A6=VYPi<9IJg;Dv#$C7HwU*DgggI zPRs@|Lm4cOexCvQ5hxP799JlM&x!(Hg?XyHGa^5Uy|-G#{m0k=KQ$W;+4Qttou6a@%y-sG(#z=f4-rd_RbV4%zy@yYDTY!_(pOtX~f_R&ah)n13%Ce9+Qln(T$xM5}=xbS?1 z2FxH2<@SkYdxNr5{m{TeL<3~e}92BPmx5pvmZRUE>-n{Riq_{^|Mf1P4s9@Q^)Zsl8pzG( zlL2ce81ATSuKB7xvE{})pQ0D(psfsKvqD4TOz@$Is1~o7DLp0Pvpqp7TH;vQKh8g0 zeh2_t*iz+RQM7%{*3qc&B)XZ&1whF>j2G%Iy0-@YzZR{Xi-SGNDi2o=tPnw}yQ($< z>J=%(nFX)$n_sQ?{kAT)lznYRR2g2aC*3lQOpWQBX~tqIY4>W@ntjW4@nnM9Be$Q} zA*0$11pw=h%t3RHzfURQr+M%4!YIf`)gk}Rracbh=XO7aBs0D?^+IuwhraG;1Ik6K zJ`C!7lgRVC?zQiR^vm?w(+Vqg_+WLr$w$wYiS7=2`%&SKE2!s;%0|AXs)|kg+pR4n zSNQ5MiiAAnZW_U&y8HTM)sG_=Sbx=f%KjVp>5F~NRCZwc=HG)wc#x>o9zi?_*?Xm6 zfQJy)=e%#qMz;GJLc9vXtE#m$h!??MEr9n{Ju~fFehRe%b58EZX?0a<)#Kq#lxMAC zS^QW%m8d>eDfXKDTdq0!Z({eXi0g9PZ|ePHG1yT#Ft{*Gc)LHJwKXe;!|~8AXyq60(^Tp0JIztn3w@$>ba*J0qRWMx5#rj-2O) zI0kk5y8TG!wA*QhVX?ax9uSeUi;EWansy(ppLR2vEx{J&@oImH3y zI7y@=brdJiH&B9&v)=UuQbmclIw-Qe;HT}qJQE%%#r_~zHk8geEQc1UP*P~$)jPi$ z=%2b}Nj^6X) zac!s!)zkau$LrVq+fL7W|GUSNZxW>7#kCK1nMq*bR_zUbb$mTOpC83=`uXbr?T;Co z=9Y|ZRr9!S#vZ!VPPSj&9Qf`!yn?pAN+>+I!MC}hm=Bqr>F{2L;D8jFcvB!aKwdE z6xZK4SHWRA#NX>tyXN+9)y}t-4WdT>>g8q|U|}alif1!Qdg|)+blM0OcK$O99A-Cw!>+8Jdo{;&?hST~cOAt)d)}sFZCgPeu;yDsL8d|*bs})g z65XF)`*#`D%@H*Cc6?pFI!H*jng(*jVAAia^5atpiH7rAaiHW}G9=UiRg80eH z-fQIT^n0fw@p>pOb}?hig`;; z{Ca`djnGxCarwfJTk<{7)n@e+AO9SB%j~*;INtL?>|~#w4}@2AtDX(mq-5PQO>} z5T2FhT($LbzpJwRKs<+@ogHCZzO^E4IK^9(E#znV^@ibQ`2I{z{waeV{aOm^e?0ES zC43$}zv3aeW>Qq__*efs4Xx^#@*2mhbvHn+7Qh^TwXT#+L{#F+5a`6eKptu9g$)#M z85?TluOzO9zC9=KU{geozWER^2VIcpmu!PVvU@JaM(?Hx%)fCos>l!(V@uv)`|c%Y zKsSBh&aMB4v}F98+%lo4AIbC+H#-d^t%^5#*hhr`>@>~E7d%!!yV=s5} z`ey=-;eC$UAei=pxL<_S!B--|<}i2Wq!9`Nrl)=m`i(Qg^=Ym?M;h;Z!p5Q>PMbV{ zsq|dbQ3e$NFn|+XjKNd0eYc9fda#w)cPdfDLatz z+TMLApb#J2)uZigM<7%>oFxfR*_fGeEGHKl=Ff|vV2WGe{JSf)09$P%Wd>u5667J! z5L-Wo&T%`fi0dEqQ<#H$IrQ%1>^!N%GV9|E%}9n7y|dzipU|3C#Yw^={{_bTV?%Xh z`c&7kfvOH$v7uqq*BfZHLCqHyy=z;4#xanExVZmymKKlhAnIH&uU`_qxFpgYNHiRg(f%62<7xhugtR!c3s0_jGE zynxTmRrzpr9|arD1Y^g{?Lsy{Y7{Sr_#3rH@lCDpDPEg0-`>&rQx3%WL?JkYD&uLr zN{3X$4tjuWDKCvU3xD=$Jc5q)8z?@GKthz7mwrj!ks0mx>u!vei~M-32gbVMMs*7n z^Q){yG2#pMP{;Sj?4?=Lg^MCA;vW!urx&tD{F90wuAN-%XM7kS6c>?xUUWR}lI73=6ojH{ z@UVuM1f$Z|Se=#is*R)n_Z)fi@f4D#Kj@?0eGaP7W!@`NkvH;SfjV$0eBc466^AOd zuUYUrGBN3iutPt-i<`E&lkeQFcUp(1-c)-o+U7K6y_eIUu9ZrE&u5mhLmtDDVJ3|b zkV&Rp{2CaBRx{?$?EXCwP_F)t?nII{$j+Z0Hz#vboR@6Mf=3B~NfG_mB|{`QjssyhzpevjbZ0>E1i&GL`II&Kb`@O+-q;zS{Cj;@LT6q>Gw#wAr+3l zXA=KA1TejwP+Wk&({S{ppuYR-W1P=1E-wipRlHsi--t@D5zqfo;sGY26ZmH-`X_AY z0OGwH8pnP{|N3ViuZG*O>=5iz417r~8!s67jxc5o!lV9gA}7BNx8y$PGWSfZ_cqaW z;G0tTikpE$Ven^AUY|PkhIxm@GxvC14)>K~#I3*0#ovf~gJs`u_jrId{7r?1FTnH^ zV@aSLfA%f(t_h!;7|=)caDTCo5q33JO)Kv&_Gjb5FTQLVTJ3p0Il21^UGwp_Sx0nM z(v{kYd0iasK_6^L-mUXplYZb@^SDFv>5Tgk^2eTdzrh!jonjTO+s%EL_A5S^V-p0c zfPo{+9vmonkiMma^!s`~hPKrn*Ao<+nfv1Ck9gGD{j=a*Vb8;eytQ|sbcD?p2WN@= zm9*S!P;U(geS=d5Hz7*CWxJiRFh22rmN=8>x@+Gt2*U0lOv1+(FR~`=)yBoDo2ti) zZ(vG2AV_Fk2Lx-z&vw>t|8jjNmkBblUw&OqPyU#h1zd~Q~_+ffiUGT(8eC52$IUTilD-r{LW_%4y z22savZWe`+^Fk33oj2g|_2-{-9Ai+c7PqIdPqVKg^2@QXmk+}{kG}+;BH@Pe%%doC z=Ccx-cKEKjn0OQVwtvaX>0e0l$+TkQ{XIq=ZXC|UK_ehY7_9-^r(k8KpMJq{|ET$j zrHgP(D4!04q|6p+iUqsHbumqjqqCgU?E#=bS(g;QyG#51e_U+l<5pR=?uCt~x4Ot> z!rNuKFPrh7=D9aqI-^niiC*7na~$iB+m>{@`l#Kv7JW{*<@M5d#1P2?h1-4;{AT|R zoCfdr**z91C+w>hNF-$^rTn_zJ!B)uB5*)sR;nTgPbtLsvQ^Y0Ra^lPoqj>p-kVE^ ztl6LZFn*$q5DhKmeTtLi-Nsw}@yw`~^8Ui`FsL-_4;^g0%4SvVL{GQiEWRZ?e^*c- ztkY(39m+u83CrEr8d5kPbV$zvRC)hoK7KR9CAt85FAR$^_hV&*Wubx_i_ejN`!n|d zMuGI%Nb;MCRP7GSg0kO5@8*8#ty{_(lM+7b+vg@qChTqyyDRO#xJ3O;jEjk_LPBSD*4N-;`6Myms%xuwJ*sC8&?v{^2`&#i?&> zA?)F>ad)D-=x_t{lImD9NDm&McpjGy8-yDN)j|z=`?;i9(Ae;VvhkLqJuWki&R`?BH( z3c{EhkUfqJ<;Nba3LoMhgXFBkdr-Ygl+%=v<~b%*<{6z}RvSlsvG7Ousug`4E~M+*BR2#pVJrSeWsFDt(VvuK$Pzg7WE^XKdKgvynC4xo6_emd z^>I+a*o*hrmz=AcS7REyyCqLYlcnT!M=!`GMw?_V_BOuX(T6^vB;&51BRqwBd6nO& zy517>Zxpx)K>)gUu?f@=6|PI1fuZ?lZ8RI*_NIm*Th^r8>SjD&?*RUZ4{S!U_AN?? z!(ic=O-mR=un4j3KJk=AqlYV>_j%vy#(m$e8%>?hUUXx1 zd5`JZzeiq~?m1kq`mq7GUjHmxPQ$?YvG&A@<=`M6KE==0KhQjC+GPO;$*e+KL1Wjv zXs3FkxjU&T%}VW={ zE1mtkMJ&4cg?GFjJZ~Rm6Z;);dv>DTI^yL@ir%qX)E7ih!K$kcTrJ;qPX^*%Z$V}8 zwea{izVY#isMdR39UNTD9i++O*c|acfL`amPV4u?)U`NYr;b(0xUHG6W-X4Zt44)X0h;f#vO=I*zfSc1a~>$wSbiUw)qkI| zY&r{u`cTzknf@xyR5yP*Xg4lwCUOee_#=$RDHq23i{ul+TTl(^kBr)B!2pyYFjZPw zh_Sh^zm*x^!+-oC=jycQ62yh*#^L%*iWZ&;ssC_@hy$$#!H|?R@{1w;8SAFL+A>rN z_VH7F91X4l2CpwbE|UiqKccV(LKB+|=e6fM^;lM2K9QRW0x`I#FPV7+^c@q@%kZiO zW8Vyi5_2dWq$u}BCA8B6k4_V$ZUJ+MGs`PKxJ;0o;@A1x^bP%o!zRXF$CP~freeqH zC7!yske{0MbCr*u5GMdvDu|KHbmjeVG8IfTi!mNj)n9HP2%S?8^bMLG&aHLpOh^@_ zX5|6w63?fLyTbV>;baWT4LH?FX?Pd5#@=hi0jHHJHUFb6NRHrws4wNozPOF{IiB0F zCQ7D%TsGlwz;re-&8AD2-M?k@8$-!^KgnlgKP--5u<7{Rf^6ZRtn8J?d5l46*u*-3X==tvcZ69@!#Wl8Y@g#wOQzZ+Z4sQJ6|Azr~>7k zCM?M4<6*YLd?!CRq9X{2@x1LZ(UEw{H2le(XR3YK5*flh_PRjziKKlgLL(>mqma!q z%O8i2K|dL>IAbxIMIfQ3N46)@V=IRIU$EQlxgoaHq-fHa+oos?>Y;XMKKn^XrQ%Xw zMyy`lJx*hX%KGjs8?Dr%u0Js7J^sT|Iy9B7EK1Fzq)6I>k%sn#`Po)bLiiRv@1kgU z$E6jMb*+gt)CtSZ{J~yB6ej-G>>`6=R^0)#U;t&6D>b5>93VczZmxJ}l_}#MP2_4$ zjclycpfcRk?GGgrPnLbw?o=3#PB%e)JG9r=g5m>N|I7=-#4^&3#=F$Iud0A~5`lgm z9rSbXF^{3^qFCIS)1Rg!A_BEfGJ+57MUpGnvKZz{aF{F_Qp^Q51-E2TYU+a2Eb z%fb0H=|EXcN@I&QLwumBWmd$HHYAGIQs92wYT_U%rX7NKShI+ zYz{jPaRFtY-g0ktJBJbbK8i0#rS)e2>&-j(w6tamc8c z+vO6FJZMDqf4hw4p*_kPdL7&=`fkqeGr18sZ~H^%#}i(;vkI$uLT}OXz#wlrjF@Orlh5VZ*tsg$ZvKeKW#bslUIWy&7zFBE}WV>t)932 zE^da6<%%S6EHwTj6ML>H!gIM}`1WsPX`hZl!}ph5}wGW3Ol+=f-{VE%;>qTrt$p z9uQ#6j4scs_HCUX4jp|ksO8=^I>o%QNoc_9`+1+HTNu2BErOs#%~yY4b&nRE2gx#a zv3GnP`@CQ;rhYE_tlz&yQ;?LO9-8k-m-NDPX2*-1^ltFVRIjW|{o@a;cYqH*Jl2Qr zMZFWT4{s)oK8|v`vk+joXtXFwB#~e1L7UnyI1kU$a6|fwq1Q}&Q(bhzxc4N3-nmm? z)(9*xPHy%iQHXKsYLczir3^K|^%Tus+L{L0yPR-?A#O3b&`xkn8>p3!e*-K96}!9NK44UL@^w7l9^MBR z$7bJ!y|%mdBS-k7IfNI)nA}z^X(PyMT@B9@`IBtY$w%vcI~8rmVU4_0Be`Bxv_KvA zaF%U)wn)_G)K;>uxCcnD2R%1)dB4;cQ6tpxSJ2*$E$DnX-DB4`1Nd>@m)B4)j@Nao zsmc1VANa2#oQvs=eG2%Fw;pOa`iC}^d8UxtvOI<#@^-W}1}+>n+ftsGKe+CyUuX1@ zz;Z7fpY%QOH6TJzMsl5m_Lcm_7X4HwA6H@I>i(^75ATJeWTYClppc7c)=hi!BsX#zbtEI~H5qNh^B$Wn z%xpgM0hsHVO|SVL)TtfH5?18yr*dhJ&05}2Fa-h)O4eYF&cC5Ouuprp5Ow~=+c=~+ z?yhm%a<#bM+u&jRf`66kKcCO69p-APUQn};&OBpe$yFZ3^O)N~RKWrZh)nL`c7Tz@ z#hw05-Orji9*E@3c|$1fcGpaQUh#bFk_!~D!)1{_cV9X=@~LU6Bq)-{1B9o~r9iw* zD334N`p3DT{zgip;aICT>pVhwYRSggV}$16o5(^+d-8^sqlqGYKRnnZ9>o;?hh=v) zj;ekUBwvxeB2_8h`eMu$TpEMQgTnk_3nhQail_@d`CXEsY`V1&F1kA0SMerz_C40t z?6%H-b{gK6BWzhPPpyLYnV>UfnF?!&3-E4Vlz5Zy-$MxTKQxbSUG8ho`!p-g;mN)B zz{(O*g^>i6OTD6$f{rn7TDI44z(34DXpaQ-@3?sTPW0cG|DYMRW%G zI9#vgPx)o()EuV=1|v+feTHmN<;U1?A+CGRr_X$J9C`;vxG zyOikN^ZfoDiB;GUtU#>6@d#$|NzB+YJsqitnI)H>7zAW@pII=a`0w*AVv39wT&A?I zwuGqPWV-`CFD3JdqzsC%=_7Ux{zU78Cb-7j@kfS3IhGbpPtP8mYa=dlJ!=FNo)CtI zmT13Oy9Tkp_Gd0X`9EB*^w?w2`9%Eob?Sd*eh)09m>22um*z2hy-xo4#Ugq+NnhkX zO+*3vy*N@>aKJIH+syf_mxaD6^SMz^$k5B8iQ~wd7TDJv*If8Rp|KU`Blj#1Kr9SN>-2kM#%d zvnqevbCQgb|QL`wj2PnEWFuPl*1dOP=NZ9?0{&HPRbDqSS^o+v5Pk zzKKT_`w0)%^l_&t-!Q!&!vqG!r>W`EhdN(*{h}zR2PS%hKKnreLjH_=l7}t&xv$A5 zH0WPqa1@Wb(w*t4324(z#;7;`^0!05cwE=eVi-mtl8xqhj5$?yy6Z&VY?+~H z-o{#;Ai}+}pSQoyZ3$gG&M}U^xz|`Hk3&acN76&#aTdOoR7iTmykSt2R96thO_kVM z01ETpzYYi@_!0~!j{wQF-{Gv#+e;Vxc0%tJQ|;gH(3nU(`^YxelWUrhuncTeZT0$gJW-nnvla+yP z=TO;Qt9bGj^gu8y_RFCkAGf34%aA?KJyJK*>&h{hJ@Y6#Sgot42ByAr`KKZGTTyXX z7j=6!s>|EeqYgu3{GE->`o6_4;|!e~d_XdAQ77^YE7U6|67w-xSM6Wla6ekwIXRSTtwv>%<*r7<)C53ar_hQDoIs|hecs7jEi0c(@*Jh(y`c5a z7p%h6>8J9*j#oet+w_Ys5SUMUue&;jHya~#u$sRrAEiHkeKDSDN;6C~H<3Q~_hZ)k zQ&V3lZeb3GU^PGUZ5&A9`{H~~`*$=q4qSvw3QHYRB2(5O=0=)qOP^pOzV@F3;TCTP zwdU9Us}Rf%n+#pmMc+##q~{0VujuW+!)`1+|9mVdgJfnHI|G59L0~k8Yeh>)X(t%t zt=2Sc4tDZ%F}n5pF+HjUS%K|J-M7iH;DIL(nti5^stxjcG9|g*H$vVjP~4) zXF8B!xZ2YuMx1gXKL#0{PQvpOoXpj~gKxHd`&!lDXq?7b5}kwD++K37z6{q^T)Feg z^04g5CWH%M4q#YN>$%TJX7n6PySsihMu_$H@WAGKXj<*J-HUC4&W5bR zE;baX)AO?uN-I6e`0}v`%>0opA?J_4c;O%Z{p3n#EJY2JsrM}h;Znm=Z&YqlICp2i z_7`QwoaW;ybV!`5`|YKSI({ggp?ps^<4NCIEV>iuI=Am!zU(OU><*|#<={p!!=Vjc z?dyV0JKuJp)!=snf&Ph`hKjNk?wvYH((8?bhv$_Y6_2|_$sYeA5jHo8s^}m>r4LG( z?>Pf>r9HpHd_RlqB3T_TBPLLtafHhF-AHG?dpi_t*7tKwr zqUoU=_jl$mU}3e8SrRgsZRZK;RXHA9Y+fRY)q@@1aU-mg@*waypecNK@IjN7*-K$~#hkiI%^lM`E?D!S#`U>?9$p6A+S3gOe^Q0Kg8jzfbriMdYH>V$IQ|(U z4E6Gm^@{Hu$o?cX6<7m8(Own7e1CiksXJ*D?9bX;mYsKyz_|jKO3IghoqwFUgg-4# z_LSIkVdcght-c27{VlBa7vkl#?ccjkMQ^7EpYQA!^HjCR^o)|ZATvZ6KYzoNBTd~p z_nAzx2o!TbDCxzrrmsR`*s&6Nxj@%KV z{sml#bX3nm^zY)?r}F{)xY_s5Caw5*DH^+LBaK%~**imDa6;CR?fGu)*%V0HJOtA< z!s!T>Bz*QmAK9`K&it3J^r(e{RabflcV^%KF6urmWc0VvxX+$XH-B%9wzsHHlp86# zPkazZ=jjTW#3&>Sc_k0c)y5oij9nb(45?}2#Szf{0S;nYE;M96Wh>8xt5+fovFwaG zO30#eWcPk+p&RIQ-|;*Sw>papr9J!9OLB+g&pjx-ti(<62vzz5&A0?TsASal;@YBW z5sF4X`2(Ptr;Ax$HNYOtpukUXop>|$z(JB-@QgAmg zj0$-)jylL2t?3P3wuBXeb@rkU1)v3g_S)k`eBt>G&=MZ67r10jkNbU~K};Cl58OEj z_jO|0o;`4!I4KzQ9*goKrO)iQZ;yv~dA&cnPNeVecZeUuYA$j*ckD#S{ciL}t8jCE0101mdl=C-ECnWq009RmsPY zmI)M^AOFTY3G)RsIJ_4|?&j2jxp33l1&gpgd`Ou3w5 zgwVf#ki{wX<7qMIEm0pV*!1OmFjp)dYcBa(8hslwD~sH>9_F%24(NYB?K|UE!s*B4 z8!BP^w1#T>$4&g?!iDRe-v-63FYD{D=a{AwmeqHc;}a{F4>1_6g(l_oa(a4|iJEYh z>0%-u^#U(D*}7~`H^s1Uw$Sk225+z6HWK)VQvmr}W*|KwAqM zW*7Xar~1v1bq@LWl{H=^(j95$_&Wxxsuv(D?bk(STpZ17D~ZHhOXhg@ECP9R%yC3* zLUG}`>#2@y2sYIE8?I=>eM8T)xtT<*iHY+UcQrZyXpKq^t^mK%zy5O zm}9lIm4!I?-m!@^maWu_u#O6>fzL)?Zan+-uw$9|);YpVpDgJo*S?|AUexR3O<$Cg zP|6evyf&7l*GqlEjL`R`{0ChYb$HcErwYHTbhBS3j2h>`48MZ=k_)}Gi@yL{*Pnbf zqB32g6aVOVV+F-QFwBN396M8Wr*h+Hm1K`yY3Xv4WMuB#~mfMK0V- zm($Q5aq%%vb&K{xD=Fq#6EyQ1+Pl#MjL_tBSe+3#N%HeHdnYpZ3%)ttDR0I^B-1>$ zZxA4}-*M8ImQeQs6WoWVl5CXavoX>2HSjV22kLvj*mHZ!x58d?KUun&qKP(&12NS&v zxUGA0+o_(D3_(Ve`Eh!33^-BB<-;TdpZ-du>l!Y|UH)m%RedX(0xJ|z+;!)Z!f#W~ zs^e}W>)~fPpY$iqUifmu{exxKcpsQ#HoyGqKmXmUQ#;BvfJOt=llkOyKt~ZFRsh7X z%2>U#%3Qd+`#`@1@vxll zeMJm!{5j)DHOaf5LQ{63>J#BP5sS-`Z(D|Rp=gAcZrLTWl5=MzKI6Y0hv6}9)?tiD z;K&H<=$Gfj7C;t<-BskmEfxP(o&rZ;-86l?Ovce8@ATc?emH?~#~%{UFE9T#(nPt* z)q5_hQ5=lbYAvB4VM-ezp^gffd9r1~(V&tdFEmr}p!`F9(@P%#XPE<;smlEQ z8i*6iUjL9ac%{gG4g63p$u?k|)As-_nYSkU?DMml@=?Cy`h_Ww_TZucG4b+VX+!*j zq)%vTiiV8oy#_{6Mml;~g0o;r}A$PiSg8?HwzsobZbIXM5_T zT%$YVfwEoaP-w#OG}de~2*A}p0FGj`5Kh%qBYF$)U=+C#~)H5rT^!&C$!$(h` zOF91a5-$wxsm)cFxW$t*)H9Kpwr-~RKa$R4Yf*I#!WZNKkUUfo2?CM~%#d{qYLat5^@HcSB8z2tfVmBtPO3dpOvh^LlVF;Wj+T zR+OQrkL4juTrj0%-(JIK;W@jbu3a5*`{&}f>r7}e!A3U1X4YT9;X@6com zZ(l4Kd3>35Bw_v>rCA9ATiL)ntm6C@*Pkb?OxYt+KMl8`zX^Ha-NZEBFAm~EJrxQ5 z!wj=}GWXsz?e8`fCmU&-UzQ9*_NBUr)C&^%W%RBaK3wYr%w->f7HK|Ib0+ez(3%p6#tT?0$N9 zSKa-^D=%wNZHWT)+oFm3*6emv?=U49@7oQHoABCk$A;gVz7Wp<27Fg@5o~rMZV<#` z=Kz~u5j{WvHg<2fzY2bLS(@Hx4MniX4>X>`r!nB~7y#>Ap&?VfG6;Ga|DqcWSS3I6 zs76_!E!ltWv(`#(dtxub0(cAe6B@BKmFLv+ZlhNFT%G}liZ4a|u!`w%dKzNfo~`ZS zP-Um;Qk%Lb9@d?Y=oZKB@S#0r9kf;&KqtX&nkwFD9d?1mnC0pI^G-;XKflX6J(w6v zCd5?}KJF{7L6d%V8p<8g?VTS#(R0By)QL9GUC;xptYnj$>+A7>&j-W;&_zxh zhtOt-a~^6~(F1%@Zpwwv#tNG74dGLB7yE8`Y#u_E5iq97WM@(zU2mGefk{Y{gz zREr+bKqxD;$P3tq`33EXkHCdg@qGD2aFQuW+>$?Q5b5Qm8@k#-2Y%ypjh~$5;=ikw zIk6`p!m`Du7$`Lx0hV0>!Y4`L^>QEpKOD_Hp2A?T!))E7Dd_`I2F#0KDIWXWPmq}` zwDZZZvj+T+J$8bJHTAo&tNw(Guz9IJGH2qZKswj#Tm3+$vwwX+|71w=9-NM$2b0)B zXPzF7Z7wTMuYBAKlgzvmEI7}`*PdelZ>e`bo^RT z1L~3T^5$2w%i&V^Hqu4va9uypDSr~~Beg+s{kwK{ap%r?KYyPzd2N?t6fCFWef?c} zn>}R4#Rc2=qK=du=(Zi1760+e?RHt7PKx%`x->3KE0+FR&U{inD2}{o1(P~E_QC>0T+wY8RX)> z1RT0M?+Le?#_?`Uu};0#JuQlgNJPYHzYhi&$ou#aGu$sYedY5-o?K}#1Q)nj%{w7b z4%CRcTFryd^#$5mBK$f2Pe>#NIsEQe4P*PT}D$S0UsA`(wss%EgE@ZoZb!;TEeo@w_{(e zI$V#pWycX?aq4a{f4I@})Cln620`UsHagF?h%eZ8OMRaJV9L)n z%^m2L;-PHPFzZv^ermZby1UGP#DUBK}!uaEno*!D6ZkvH@5+#oUy3p2jd zX%K*uw^t-+&o>a6`+jrST=rV_`vTuq?~5>)E~n@Ygb0Q5_^MVu1uh9UDqz<)N8y=y zj4D$_3E!sglN?>EDQ3KRmEG(!TE;;LnieQLTHmUZF6);YFfNxfrO)c8I1KG%Sum8J zmY(%K&#vVOd*eoGxyI^mPBR%eu4ya0&Ur<^eh2}ta}{xNsdHi3U{5?S`F9T@Tr?;` z-#r}p_)K`-eI1bwE>iCR>t_Q@o&SM{D|7k!dF1ZMVOMI+Rr}{q@+(?n-)B2{^}#hT zdNv_Rs1|VOn$7P8Ki|gHR z*{^VoGwyD+^3FVL>z4oCwj53?o}k;&8`}S|_Fmo1{k7X8E5}n+d;h(uA;Qz@31?+m@%8%ZCxsu2-?pn`T}T` z?p<;_Z=++vGyYR$g6=@%-!LH~&VPa`q~}rB4+~ZM$_h$NaXB)Z7Uf`CEz&vIA{jG-?~|P;=H8U*b7dc9Pto))&ZU}<`0l<+*?I>(8P+WL1fkK!U8$a~HSna}S+lrudXP@J( zjvp_oq7!$8FDM%6@H!w>j{dRO46iqS1aU&`{b1pL`-+aXXWZ)8yoFFjy;tpNSBc4= zQD*!{Ltk&f-*5B*e2Rlj5O}SLWub&Gs9$|EVcdKfug+JI2kmbI_q^JG!+R5p3W05B zw^&18@Mf8!`_k5J3OC@M#5Q1ebP4Th{dFA(*wAQ@5f5tEl zUmN``haz)da9Eo}x_zuODr|ceoAdF30jvwKzXn zyM=n>q#U8E^;J{G3T4Zp99=d_vDrbYEXsV!oR6gs8y(jG zlZ3tGC|p6;4n4<}Pw2n-QqbQH7$+Lf$Km_gQ)hN_dMN0ZJieYv_^_pYogy@?ZT%e# z35^`W)POKzCQ?Si@k3w@36^g+KvX-0JF5kP3WEXaoKS+1mKYW%FRjTBl6?$M&=L<} zk|9eD`o;NG>%ZgY(P^@RvdtnsxmyB^pJoS2Z)b;GTghKxa9R@$X1FM;mxPxA0)3ku9oM>UULl%l*ms0j_318%pCSm~7SP93beS4WMHNCkXxA;A~d}qS1P{)93E@X>mNs(9l}jJC}BOM?vm7Tw%b5>K!u8|5fFo-@F{aj+SY^ zEpcYZ;)0`em9fd_?pxO7sBPNCR(Eg5o{~-0a&K%-8xlxq5h03!rQn5X!Cd5O zz{v5{543xs{y2{_kFK+?@W=C5Q+}HM2%G?0@Df_X(j)pW@LBabXXmVdxfS1VZY{%z z1WXDEeuSOyMC9|9W^ICI7kD(-ZEhuF_LSiM+b&e#sE_L0q&Z+z2AVYN!g3hBwH@Gi z%fllCEwxcaYU?tAn!Cns7HGCwsO1eaW6w#^PwSrd6KcuE)$HHM#oyMg8{d6Ys#5#+ z>p25$2rV$$FR$B&J`*B+a4O+Oc7RDOsdnsn5?J)PeG*d(+AI_TRSz`(5mc^j@CZ1m z#yisbqJI`7S?=8Zd_Ep-xY;oFVK;lvpDLc)Bp$qIlH^0B9T8db?TI>d+wJ;3-O11H zW(11c?yc=#PX69{dU0r_Ltr3KnMSV#{mwf#st)rXBtmBB{)p1Gnns;HsyPe$J~Eo_ zSvM^PtbI1sXW|Av*MrSnqcZP|atKjjGx25qtEmzS!Tk>Ko;u@5*)1Y`f=f z?BldEckn6L73qL53oCHJpY{kim(PRNN#ms$W3caopcJxtTG4eY#vERUIUcunv)tY; z;fMI>N}a0ww(pggS9UV;$1`&GXA7GO;k|EhL(%tS+V2#M)YdHCv!9`!?o!G>i6s>$ z8``c?$Thf@?%hKlCJtum!7nlb%}5(Wdg&)Xr5tfg7XyuJ`Gy^$LJ*3tP--A3FNW@F z($WJR+cATI`^7H|HQMITwtlI<^Qwl;XMr{A0F}!)ZaqLH*cpDdGa`1z+COVKTob7! zJJGS?gqD>n^DGKN5MJUu)3o9yT8GF6LYB4$Z%PkSq!HE93$bZPdd&M@IzdGo&Q_DYoaNXtMil7Hn|H$k56WZbNnC`P z5Bqs*it_k+yCcF}Txz~-{oe8?IO1M#y*IDTl=5;_0eu02*yk;oHA?xG{Br;G*ZT4q zOwyH|mtiL{xUvQY)4j zu&vf@I3psNAll>VMejc8`{gBi>&G_j+XqYJ6=?;K$dEzo}*huuT3*q(`AVak8_O(BH!x29DJL<>= z#v=J0$z2DYmeK8eM*V<<1V3(|v*L)77NwT-O5z{h1_?*Q+ z(D$mTxKUc|xC%hsg(G@Y{@*2 z-N)V@NYCIz5th3EXq*+3rR472&P+t^tlyJCHEx8+J5zxXo8L?01oA}%+a+=PyURSw zH$i@qxT|5A6PF%uUVrD$6<+@lkbh`L2ZdbX>P7R_3qTo4S)xn5k@3H`3POnvJXqVV z)?R@~Kl%DXEIV;Sd?6^e_U=ryA;JrZT=k6Y2YkNnp|GE^gAwQ`|FJh0Zw@tK_%8J z$`ZK|T)!(fE(Z!2$ICCYlVmFPc}oIO-cw@1@YzLyyf@}AFWvM_5 zNBG7=`*=0cal#j?HWX>cVf3*U$F|0Nq;Zvxv^_90 zVyw6H2Xuh`czna}^_Yd6#4PS-ASJ_iOGmT|Wz=W84afi%vkX~==lfQBt&lFj1-{H7 zTpDWxenGn?G`~h zgx+IQb{2{Qk^H)Aul8!(SQjKNTyrXBePcdgI|Y>uunGUKSGguk?82{SJgXpurO(+F`@AcLIF)WApIASY+yhRoZ#rJbz=+yISsk+s-x~xa0YG zg7ZXcn`sc^zC&M*%u#hb1V`!90dBXxKKnwWbb3u;!%BBd_yd*>xUc;rI}Ftkqs}qSu$W`5(D#KVOmY06>rONPy_YT ze&_+FKwS@zbJ5Cw50-${e6wRH+}NYW!1-?D>#58BP;4Jl3p#Md9Zi4XPE31(?-o?8 zi#MVWtdm4s_$h(j*5g3Qw0f)WfaV-KBR^wPLAskfuH16wHrD4Nec*u*YY z*Ps`hD`z^5c<;(%bh$iF+s}%Xas1#ps4tPil!sd+q z;rZncTMK+`m=%FNDy-5D%9sl<}Q9#0I!Fd2S7o01$>lT=szR3 zV39$vQpyFY=j??3fW4}>j>B<2HpI{T)b~SV6uV3&%>Z)+#hXZZ?~DO>@&);Q$(Z%p zSCf$22<^wFbifu>$ZGj<9+0mU?`t8{yI7YtONFJ+pRz<%^smGD;D(Lb>WG>#^yR&%)Om`cG%`++%*TE%x~FZR^MlR2b5?4sU}Y<~`BA>&~e2`dr_AQe>AS^9Z%=rU`B8| zM_U;RrFGoig7`K=9&lN4%j6_e+hC13Fn_aR#w3_W&YNvF?2_YXCfV!q+Z!s%Q8AM( zfpB4_8lGSrxZ6IaFXNcoGVI3C?<;$PX541c*=ApZSy`Bc$%!9b0Qs9ap|{Iqb40ge zli$;GIAsLQ4|c*=_b>a6`?u+)$69}YrWxE`5BnW;4$y)K6==U(g_%#Nzpsxs91MS5 z?O=o`C&>pTH-dNPH0+$VDR5|&-Rr^$Uk~x=blkDm)8D+M8pRxuoxyucE?Kd=yiF#Q ztX_3^%s`_%h_frXt?=cVxy{KYg>9m_MnEPk*EUNvOIFJ;-TMxr$7^B@IRxGOH+UXf(-eEENb+gAv}WZ7gBnBBE6}74 zsu_EuCb+nW4PIi4o4`X%U#M~ceS@~Hqr#v2V7Mo0aR>z$x$q-<+>M(cj}(Ar2IOH> zWMA@uAtM90l*{wb zg^_S14jHz!ZL}rO1+}?6V9|O?^+wNwweJCrLR?BzFCu*z4EQFXNPHlP81Va*Hpa!bHo*!UJmEyV}NCa{ZJpLJGFAKVsUsr7Vc$& zZfL2jck8zD#pYgZqtu)e6`w?m#&hx6V_1xetD7FaPFfi%3U1OV(lIF$vWfMD3$_=} z;?Fdg=w0#t0f8C#8Up=jm_!!(Uu19BGlukX5OhTqaYYtdTPc^l!)LJf^667Qu(ojR z_QMtH__*#gm;8#iC-!LF9`Yv-#%uBRTZzYyD8+Ei-s=9@TE%^o>Dh*1OUAwO zQD^Bq@!#lAQvGKb)TB^l!P>WHd4fe4lXxye{Oi{)&dQKoE+00i@OmV zt)K#3SNT3)5v&f*8l$3GkFl5p=w1j38)a%w$)zgXxY-o*7v562&GS1rt_l)(XuaXv ztIcTFiXZa4C!T>Jc`p0efLbci3cWnB3cro!?M;79q<7i2f1ZI|XQ$cD@Mq`Qb=Hv| zL--+2C0rHb!+ggW;im2Lu2GM-cI0X9q=L3!gsEiyT4r;%x6)1pQuPD3_pBSDP@D|; z8#%(l*>ldOM6cX{+%fM8@T3~t75iH59d^6i!`Bs1JL|D2g2HV9_C3j~d-^cxQ)R8@ zO+q-KYA7Er8J%LF^}^vhTacl+A=uq(Je@y(2SzpBIcg}OUQX(c%cPDkm^g3OmpF2G z=#X4i>M_aUWrGH#=s5!D+>(riw99kz1O+g!)Q}Qn>a}NiCQa+Loy$2<_wZ4QMNi%N z*gn?;{je_9Iekbq7ovG-m!UG~g7x?GD$fe8-~lc!*CUrVTvNyC5q+in{&*x=v{|aT z3#@)mG4KdPO* zMJ>}R@B9pP@w(FciZ~wEbL~snMk3XxcP7Q1+&!8Mwo$ZCRQFs`MwlO4sGpOd?@V#J zt?{6-+wsR(h~14#9HSTm7txB%swIXy_a9i&$@6ui`AuK*Kp5?320L=XjnW@dF`lJF%b>b&o;bql&brb(S zi-IG_>vjC%fQWBhV4_92%V0`%t#4*ia4oj%U1kWhkRQbB%&VwZX8VONS+$&>2Syb_ z@HrBq-zf3LYgC|Cku~8=!yxxIMy}w5R=$QqhBS^~LtfABK}Ey1gqCT09e=$>$uDJ7 z@bH`L^U=c9L2VGI^77-H^el*-BaZI&;m}Ecxi2(2pPF*(aM5Qc#)_Wgh9|A|b$2Uv zZ8SWaj|wu5a;(~1ayo9io9*K*&n_$T(lZe6RL@1W0+r{;y7|B={gl6l?zXA?THl{t z>`hx#XQT9{A$jeL#5)cv(Bg0HH&Ff*1>k}$1vZmoeLA3wce*goslMD3FWW*mzw^t> z^wD0xhHgu)GaP(xLb$<4w;>6^c&(dZuu_1GuyL(Yf%N3>q%J_x(qSEJ>4?@1UWF<6CJR21@3TsOv4WS3cr$)i{Ddcbg$}Yj zI!HMx^sJ$GtP@EM@p^Yn=Y<9P@;)UeMK=P9{5?GuXDp8ph&w8`;`G6Ov0+w=m#T%; zN2gP&Yisw^eCDnnk@cO=xJuffp?+6yI6DoE=gSwVyjWV~tUnk^NDqWZ<(py{t?h2W z2>RqBS{qJfdb~}G|L&YUnAntqD}X!wXc$`ln629jj>yzRT>#QIn)^9>sUo>wgEXeH z$!b+6(!g;u6#9-Xbawywl1%kw9(n-!que(Qxot)<^zsXH*-@r_>ZMGE- zGz16?=q{(NC3f(3mFsf>88W2CZN=($3fkiIX1OQa_gK)L6Ig zkY#$Q5_j5u3Ce3Tm&fyF`pO-KcsmkxFEt^4zXluv_Ites#{vu4^U?W5o;x!&20eWP zG_~o-rNWZVEx-0!q3{a|6eec%sRg8JMfe#DZ3gbKvWUJkar5bTE^jwZ#+ZTfC2*gC z9#LnKR+YX|yd}A3sjLS`Y5FL5#9G!+m5#_{q&rjnXJ0tJGcsS7vQMvZiBN%UclIV| z3-IUagJj&BTjZqQc@JV1Skv-2&y^;B`91VCZkZqkms{8X5=$3`N@02)i8rU z$VSMh{bf6Z%??FA`b5#VK(Cf$Z~pM}w>2Q62PouhM?FWzSl;wVgOe8$2+87+|5a#| z6}|W7!^I{ax_IzHZ?qmP9x@0GM0CM9JF);!FOCI_`L1WB>koue+BZ+lQx>_PkQAc@XP{R39EMyjKrwb)ke=M}31y-4GD-EJdoR1ejx=O((+|34ED{&uo8N>;^jW_LW+o?sgo=m}RlXZu zpkt&B3gKkzhR5Bl`OJqH5&Q2n9avtIj)%XUG0nF)M8Y`~)CiGC24m$oeUNkqg0~18 zj`tb2@G-6;TU%(nDq$dLZwTy~U-6P^kZ{?Bwt!-Zen`F-#GhM<)r@50-?lHgAu@*n zf$2+yAt&?ur>-3Y`J#_%n4tz%*^M8U_pMPQ$qRkvx?Op0qJ4pX#O03-98fX`zj;|? z`_ma}!*2Nu$8A`?!nG-i2!viKv<>Y&7NQh@e@838<erxUQl z?-TFla;Y|r3m3PL<0^P8Mz@SR@6>!I>9A^thk>h=m61DVpOOM4So0YdSFI>g_i7Em zmZ6yF!(s{ao%Qp`VA=!AzB>5bBlw}5N`A+X(yX3xpQvnUxfrtdj+EY`&9r8r0V~t_ z?#XdrWI9#dcNic=8*H1$102~S%Bk)3K)qA2C5giqeL~6+NO4*QfT?&4DzHPe5+ua# zyI!9!j@z(VkHXDwj{e4q`^EOfHGsz;4dT7M$tRe)1VM{7;8>wyhK{4l%{FnxE>lH& zc)7S#I47cK?4m0KRUg{IwS|Sr-7vA|h=0EH>ou!Gfn%oP$=})onGx>dz8ahplN+T{ z^1596^sTpXjZ+!_yS}eiT3q8N3^mlH+a5rRV&|R+qlcZ(-!M46DfPpA3sje)t5$3M z`}ovva2t0x%*)T1!^dw=rBRNxaWXHO_s~k*Bs+L9TDSr~pVFxbKlMBTCA5b$eYVNg z1R_r^i)XR?o{s!+`&`dGovS{dFyZCO}_F<8& zIr9TCc?8wt+^FxsKv^jqR`4zTORrVYeWG)hS8EkcZQyHV@&!ifbVBAAIT(klP?)q* zJgOyhtFI%XoNdaJEH?CU!s(8Cvkf}4{NU25?L&{1RM<)uKt8_QC(By?rGLer#w7HT zNLT=r4fh(A*vh}I&xNPRs-*2ZlN$YD0U6)wY{REY)w?saU zm#Moa_7zo?S_po`xOomZ1A@eN5vq!`)yCI-uxU%@>5Z#r8c%2%68eWI%$!e-Uhr@G zp*7Y?Zd?`06urazGkYB<;PFBJZ6mE^6`e$s2eJ)OZ-(WXJu-h@q+Nl1PRZvd?k-kn zNp~Em#u4@&&FALOEF{8~2q+o*$zu#DW!d)%G6-QNG?D&x61YmXB17f;Nk zK<&EgSJ{G~Jno--1rXf7C`IfA`y=vv>eZjOexKOg>B}5N4?#nxH7XoEcK0yLedUT@ zSm3j##c^VzAzO1yNLxUU8>GPt`PjqR&0d^u@mq6CsNZM&y$<@Ui??luzPQs=4rB8x zAfJ*YtGInN?+uCcZYo`;ns48?Lvw`UqQYd1s3wB8AHrqL0~OWq0-P4u>rJC_=gTky zWpxCV`7Jc{wuCZkr7eQ;&;0@mB-*Qg8cIV`ASZ8oq^(H(|7tCjgXG;j8p=JfYr{!b zv%2B1qgJQ){5QA3mij>%6dQiPi2xgb$B6B(I)QkH_qs=;_)z`zJ}xrNDCj2GG}%aO z*{_aL;*0UOj5qyYfI)gNf5W&=>^ZgoQ-ki8f6Cln~) z&1mq|iQ1#0qu8Ly5Or5{5;6`e#LVcu_0wf24ygr>@W|`#PLuEI1xPC|y^Z^TvOQSs zYnbV&=j(iP7>9I!NRJOAiGhX##~F1<=p6})``C1A?e z>HeW%Ao;@*F!)a8~Nb(`@8zv2D%%wn|ioxWs*+C1+_f9-i# zMM1lS0@4ErDy-jOucp+4S3i^s5c3O^#&+mV^qXn_+5q(Qc@FsOfITmITJ`Q{W^R%C zOjGPQdC{*{H=y)qcYcr_m;-Lm{0+S^-u4^Y@sykPkOM$~zY_kNa)UKLbI&T}+#cuJ z`e4y{Yae^wVX^{CpL2;U+W)*4=PH^AZq+a)%9As*Ty&!Acq+G_JI?<*j}!;;YCC(s zvhsq7uikmiE_~B`6K1XB4a&ypiKV_t$yJ920eQ1u)Hfye7iW(x9N(0E2$`n8zA|TlhhOi!fP^YvT{KR>iEAlR ztra&*s-=dXL2AW;_v9!iz0bn+Q+ke%3MlPeYE4a&f2BO^C*p6U{y3N_AkS(h9mo2r zHwG%d)X(?Z`rB^Pf8ox1ik`}(xx9N&w^{ui<9?r?fe@V^@!%kF=GWwEP7u z>2o&TMPA;lwsZ#+^2Dd)%S~~=1ASH7SK||u$l~pZ)q6q3pLRbUeJ#Qi3I$U+wziPM z)Lxj@dSWC5DgvsPk&e=?zAVzG4C#s88PWkq7j|TD#_d%1G;@Gulx?xmRaMPT z0QKIUC69A(8SG2-X>iH{flDVORm%v$E+@gj5>F0HR0JZbVRcmeMB49LQY(vduD>ieDAm>}^f^>(t!i`8uw@bmVNKVWWd z3BLoO=1uiiS;nwY@L0huO_lxzT{Y}1CjAiNHqtT(;b*QrDdrjn}(K=E7Xq3 z4c)byT2nEz1%dZ=G`2{>vbd``T^hd)P7y|tnf%qrbvlqG(MX@WujH5_9y*;mT@%%6`vLB;WOsZVZ($ zY&?G0q)uwEMO;gO@Qm$V#J>#ybyAIFbS zh5iT>wORAOWfQE-4Z~pa2Ip;>7b_b-d<%b44eLX?o-7791raZayHQ+&x)y-&p>#95 zlBFZHLQ!1qh=|{ViPIVRwoQQf+OIghCWnvFc;BBJZa?jqe8j12z+Ryy)GTkV2p-FQ z6IC+96s=&Qct&yf7Pxy6Dauzlyi4j+3=3SR@`f;?2`zzW=KX5#Mw*C-!2v#p`Yt!( zMc)pnt2A~i#*opb0-B&iVb?}dVzCeC`_Dv|h=>*Q4pHF&HSINsL}q_4J(_}+t8z+i z+_2=Qqku%93X;Re);7h;)ArDQy;p=5*`j(>#3I&U^1gUt_d)0Gy=~ssS!IOIwPrlI z?e#^zhjcT>=P@`JeNW%XYc$bkL3Fx$tIw&f2HcqGG3*kO-l0q zS(`ALsbbF~iZ1w)8Pv>X5Vi=3vt1DY6X{L^SBV>9s(?ESn(_a)k>!KG$G!J{2x%72@U1nKY-2&&>q|R zuzk#!=E1z^9f?|%Ed^QZaGb=@4s)fGN@bo3X%i@tuW#2FB~?3PdDK? zvoKk#^5%Rz+Mm)#ET7`*E&wj39c+7`rN>m-cJzzCj1KB^)gx0M8M5&Zat)>kFc zooF5O>s{-;5(-A%9iR*Z9!8il>xdH}G?Bv8a42U~jzXS)Co|&#z`~Y#&I!-;%y0kR zS7u^2eD?ZT47;s0_7kH?=Ollafq<_pHF5+?mwG=))@F*AEHi!5Y z@!RCDm0fQHu2;n7iF#swzaxb8GO-x72oAxpc|xGx`ZW+|22bitw$VlsM1pt^6-2sw z@DP`0^R3swsB4e`ld-sOqANfi*IS>@1UsLl;sOASB-uHVF(nVO-91Eg1|tN^kCoo^ zg%4$*(bJG;?k5h%^oBN~3u(ob{-SKewk3VXUJ4;f>fJ`?k{W~C-BrDvEQ5k0DQQng zn1HS@qC!nmEbqoSjG-tzbTxW?PlSXsz(!)j;qJu+7-TW(*q*!+?Z^F(sA2Y{NS*7i zlT`M98-w8#u>yGxAzto!7|9CvhDczGar&I@&dM(SG9Q^c(E@6#U)nDV%ji{pHz!bO zeVyRnr+wq!>le4lWA&$LSJ#%xms{qj;-j&pOw z@Srr7R{PQ7I>Gz`7KL0|;mHD$Wj!Y_D!a_LYU=TqKQrsN>I7Q1>mWF1w&jIu7FO%o zGS-?X3akYy^YOgt(qH@S(|<`%do`6 z(FJvv)JWq=)CP4DmnkfktEh$G;tTVxLT^dVOl?HG4#x3-+>(JtAM)rW%csuIyq%#8 z4qSKX@I-+K_UHt&fcCdx+hhY;m4Kmh@?c-Dy4af{&I$ z)c7pjIEJ|wJFEdDlfSnP4mbEB)(+1f+wc90L-A`_rtuu&gfdM$)e}}K5#$M_F*#v67sCBc z@$i$+@`TAO)1{3o3lsI)3>)H(H;9X8oSrNejvBL+9f-bV5%;sSkPmT_A?3K-2w`rL zDsdNtDi-~agY3RPznM?DMv^!U*?6VAgmonS31dOPpXf`J?8qDK@HHCYp0}PKey{RE zN>|N9=RjQY$NTIC0G<Y8 zv)$7`z*UFpyX9u0{NDSc&IlC4j1$XFhTQ@aJrOnjkJ zQQR2BrI7|^X2WY zyPVr@wxB?Mr+uIc*Yo~Z?*9<87n3XZ<+;3En|E}#bQ3~9nIZ*^uKmH(JEpJ4}@E6Mk$ljG_)dyAI{fW(W8BV>0 zw9b`s&b!Cm38m}fbA6(7a2v4dH*!a2xsy~YXxWy1U-5ulfkg~91HpRcj&iTfneO9Z z9JSux=!Gb|(K!Hlu>5|G@l{Tj7w(^5_=h0YaU^zJshVIl%R#fYc&n$c~3?ZUq; z*V@gd{1IjV{rS&zy&PzVNW022v$$uh$707DPCC%m${n<+RO0TmzD{XG0BsK$d=Jjv zl&~X=N?Raqh!4Wdl60Ye>uc0Sf^`St%Rvm^X)-`lq*?(xz|N(-Je-(GX80dsRQ53UbWDf@(Xpt&O&JnE zsZMpAb4fB_OaJ*VZ@Cw@YV$r9?aCNm%yWy=du@GUV6WfCFP6hj+i)sgTZE9lt(b4u zN3t=bi=aNxCsNlJD>(_doNU5A`J9MEd_r|2o`<_>`Y{%ILuR}2f^syJ8R2rA_B||V zo7;N5EBuc*R_hj*cDMC6xzZDNU4$4Z9_s(NdY7Kp`8BO`Z%_ey3#l7X0*Orvq;6?f zwGocb_MnjxJB||{lQ;((l}bLwc6>-t^q;z>Yxqk8SMf%JwtPUf!LE~e`xPEI6`V=h{_ zqqO>yUl0-pMhIUM*pDw-rOFQ$gb=&-&ezdfIK;Ef_k!fog2%=ExF5)+F^H?Lv+9Fu zG9mA~g%BgYu<+X-Wz)B3TY^IO&3PLqib*$SblMvtcP0sYZz>=vDbPw0VhR*aQc312 zW|tvSatvpQST|rA$Tyv;;XXC984=RE;CkNV#1RZ&{Rs%$8R_%VxFEE{#}kyP!x)7> zKDNkiJwbn0&k!v5s&#!6hF9<6*FZt~d^w;6;oPguN$|{PUMyZTLOyKNH{=rmR47(! z(Eeha5olQE0x~wem|bwgw|SMa3p{Jg+m8dgreUYbv(HDw{H%$NWin9pPJ^DFLRmSjSODOVQ9K;oJ0TJ zuh#s0?gGsF7w34^TY#uFpFb-J7^_ZYq1(sPZA&Z<68PuHhLtt1OWG&W(fF$C`UENs z$62#=B;-8M>&*+gZ7mRz(Hc7c%kM~#Avy;gP+0|!jsMLjq$(zJG_Pw4;6}m0g0k-I zO_(?s4-u=?YGH$G$Mb3Sp6&x!QXyu~fT( zcWdI=0!CT(gdfpxx6LwuaG6O7|3f(<+At*YnHGeLsU}j@UrhDYOI}v5>bhP*ldtM&C3PmGp zV5oo;x6}G)xxsN&@3C*$eO4{vcemH2L`uRdW${#q1Bp_$P%Y#<5o|oLl#W&7hwLfA zoShK~WVXtu#MZDSuU@Cfso!W?F^?Gwf9#r}v8E;C?TG9w5oJvBFOM7pfu#2V&v9jQ zc_V&2Z+GYGg$zH%PC_vJ;Sl98%t_^~DZTl4F^-mLO~Mk4wwGy{jm=@Q{av6OAPOO` zFr1x?6%mX?hh$uXF%%`j!X$)<4iphs)C+OPa+IVB!zvTVVW?%UUf8Q@y;waq6mweY zjZoaV;Zx6Ok7W)gMs2NyvuvqY+9byV3*Id6uW5gg4c(uvC?;R(r9ZbRrW!}4k6eiD z%$*H)-zyH4^(53lxhElaZcoPE-@4{`V=Bb)lGqDwosP$2#HQs)9nxVgH1RxLfoJEW zkDNW1@Wk(z%e8Rc9oC>^Gxd-#gziT;rAV8F0YrF_3g&@8?wU2u`i>)U{`DK>Dw>nB z4o!gkDjT?hTOfo2H=+RK?sj%&;@p~+0 zcSO*M#yMy&W_@U|dQq>0sJ%sKQ8;W|>!^>mF{HeEGbf1|E`3be-^Q|{xg9Mw*ik!> zq1xf;={vBcLBXD40h*g>YVu^Mc{G0Q@+7NX8xd}Q~4%qun<8Wci9EBZK{u7#Wp zJ;Bg8Dy6y;Y8*ySVCGt>!=6zAfJsO{0vDXZJrBbR60M;jG}(@40{ol6L4rQ-SJ?wW3q2HN>eW1(0ZUdn&z4;H+}%&f-BG+n zO_5%)*V0dc~-4EuT(c9@3WRqY~MovLR2-_k?}F)sS86@Ky;D#%d1uuXsBJ z3?=j{ex&a~JxA76;8zF#lV!5phH6#Ur#T^+B2+c6+EEUPrsd_1V%I?UG`tNF17*$) zMNBi*l6h?VS0SVg+8@yxYGrk*>(^8Fk+HZ+-}qD7EFSO6eKvjEa2PW^eb1w&3| zd`Bl~Itw+Li?L0tpY#^cey{&Ae?@Wx5= zri&(F4$ZClBoi%TQx30)m~5IExP%I%bm{^aT6u**;$dVjwPoR^-=tu`Z(3%?pv(ER z%3Rm!oQ1Eo4@o6Sj#qcBBep|jVI-~XL&sk01nk^3akk%S!H$cs1Ilo11$-JlfXY2d z=XN%aB5flEP)F#3?@y2$Z}$rma7sjLk|4)Yb}EHuJG*RW9Ue#RzR6PLM}Ei>cY>#s z%>Ix_D~V&L9u=f;HK+Jw(d-g2dpD=To%;a5-~!xP^8CXQe;dg}Au_8)nq53kFHFhV zo9YZ2DP$cdLsU#{6kc|y14)P)rV-{^A?@dY@gGTSV{+B!4~Y~$Qt_5c9}K)AiOm5v zR+dMo{gY(TCX`T`5aNlL<7A+}&AD-dw>D(j@sMw#^S$R=v=!;IcA|i5IB!^9)ogUV zxbump;8)$UfD=tq&|<>3&eJsB-~z-@wAXf2EX9Gu)i0G$43e0U@ z&P?q_YghVQb`0jH^#STQ4+f?Wc_-qpJ2}bq{(9iK9;DTkzbv?eWwk8xz8r_ugchAq zrjB+34aM&nFLvj7vX`P5$#`iI^IEzYEv(CgF!k%z!ZJS|4o~Tf4()(w#^XPDQIGdK~26+D3QAk_nILl1;p5LbAR7Je}b0(Ijz7hEKg z5Cw^8nHP;H)PiLX5+P$lB5sH&9hl>x9M7299O|CDr%EWa>$ce{nM4$OjBQMEYH@zt zUVsz%OkuT~%dKFfe79PJ2m;P?VuONpv7$c9IiDze79>&DBBz=I5ynY?^I6fR;sk?X?rC|_&l7L z5V(OwK?wAd9h~=W3-b!UNmL2>`y#jjBa%fp-Coxo zbHxWKY6r**#89}X(nMV8;?bZ$xE^Cpz-h@EwAi$@jC>8`l|Sc^)7mS<#~+QM65swFJc4I`5R}U z#E>L53U?F>3du2c1f3xCLt2y>PngR;cgYH%Jt}zpQ%o-5*M%b)7 z(RF&!(%zuaaEP_T>xJ;Yx`zg^Jr2gcIVBwbY@{CWf9~>C$!gEBJ;pJ^)yI7#x^Mh~ zYndNCeG;|ehd%h;N|)oM5`!p?;BG|_q!?P?2B);deqob-Fj7SM8@BWoF3l{RdjnPw zP0Nsre&X1AzF$^R*wwl^3Pr~n6|6?VZWGZ8O2XCR3+!bRK=ugX9Di>NqPRT`>=E-K z|BRMFq=O z4*)X&x40voI1;RPT#h9qoVHzDYzk$b4Uuai$_RUg$o8`A@nin6mD1`og4YmYMHlfU&hk- zmqcY?+f<)~KXLj78G zVj@)$8N@8V6_X2=iSGH?)xfRSG!~UfJ-02~_=s@IDN=O6$Ujjk1LTV$L!?2#OCi7R;{EMx9JIUN$W0Hy>dO$XDz&fq5wgpVGrW)tysPGOV{0?J zc!u@3&b*X^nscQtYGj^|Kqvvd01;Ud9ZyI0T{gwrY}Gs48{s!+8sTVYlCda^9^C2% zzmI`gvJdpwD7<^zoe&o+g8un-puctrLvV7aB~2QI3H?FlXU582uWj4mkspz&$>M(A zpU`=@OG`7co#<{{NnlPG@Uq4&21lpk)N2GmkY!n-=rK1+u@HFZvG@m10_%Yh@_`h- zZu$e(EL1$>fgu;_0yoAbz*x!o^pgEc%T2hIf9ROBsx_eH1|ZaW@m7DG9p)dkCO28HP{~lPH1ozeTcP4BI;DiNJ^q zT!G8UIN&;r!U8RyfhExFksDj_$wK6SO|}G*dtK*JXGjHr!x*6f1P8jmq@M;(QZ)sg z!GV-~_rZ!|M1Jtyl7l6~54;@#d?O@i8z*HaY#G~}xz13F%u8rUZ_U7o>S*^o5a-Avb!xTF)M}uO?CG}%+jX=I~W(mH4()I>EaBqka(a!{o;aw za>r8@hF&MNhv$E~KnvW02-iLVnJK-^V;2L(o4@mOWUX8d>}Cv}F4Hs$^Y@ja7-gqW zTE43^DIlbTzrAY38i8S6mi9uO*6NY^?k;N{_FS~2>O^3G%<7-l=p!u<2qR2aZnf<< zz^EylOH?YFI#0(BHAEAdm656^qad(mx&rja()Vo{{DcWWy2noww>%OM*6$b%E~O~k zcw)^JA=iVqLKsd19?n%X)QQ@uy|;&(qk*#T>yCNj?Oc39-iDFkdRSzj8ySoKu}Jan zv~aZS%QYqIdU>VWbl4cz*+Wa0DZdmE0)lcbF;UWGh^j*^+C`mojE3!THJ( zL{1uWgkyj>(s`bkL~+;2;bz8n7E#hS_Z}Z?>4AjbJpjQm)0;2fd$}BC<8p# zkf#Xlwe9XGWSG~GlEgSqN-q21147Zr8tovk-`IPVT<+@`Jl`=4|MS1j$USO&O+s^4q4gn?QOo&>-X42kw zNCE=Da0dCUKEWVqsO!)7o2Pz*wRrLDz-UD-E9fl)C_6BaNN$n2##|Da5mPR@zHiuBVzA#^)bE>b0X1)aOcZzb2kUApn${cN(tGS(5 zB`MGyF7hJz5aT`~#HDK2vtT)C8jYkfMp}i@E{?f4LNquk-&enqaw?z_;4V-E_@h%c z9(QE24H}CYY*lzBU1*2sg-N~09Vu)OKi;(yg5R=TH3IqdEoP1%to4dhkjuFD4|ib% zdqQ7+|3g9G-jDshlI36i%m4WQ{OrenzwQ6zw?+7SKl|CgXv6>RuSqfE-->_s{o|j1 z|M;i);}8Gzw|`Ll@A%9AQGD>{zxeOIKKs4wuVsJzkA9o|{XhB5zxg-B?;r8`><@SJ zpZ%?$W&dkNWWOr3U;jk?kALIu{`}8=`qS!{zyAyC`vt!${_(GW`Y+Wl|K4BA{^1|} zw)o9Y|F-(&um5M_cc1!O{|mn2=hMqMTI&198edwd_@mGMpx}P@r2em8{}fce{0BeZ z{dg6h#rI$TaCi7mfB75#&+pU0v&L_<@yY++uX6wLpNb3qir4&?@9Q7o|Nrc7#(DcF zgZCfDK^OeC_+$LrUwlPT<9~3h9^bEv$WxoV+MoW;@#m+%Pw{`e_{Wd>_S?4oZ^yET zethq5{!8=wzxh}Hp86~6fBBpL{O^DDx04Ykf@;c3?|Vk#x>Fx=*k{weuGNYx@97pXo&+Uk$c9z|45j}_q_ocpCWBRi9jg^Tw zSMwVevIpm}Im)A9i!cM2`uh|idh30-y{G14QqxUn<_^AvSD&Ws@k2mrR$fR);-CCe za8utGT#Q4K8jaa!^`s~pn(hG^$K*Wpg%WpO`3 zTbhZGtD2+MrO$`Hdc|&$NP)nxKRUgu-~OpKi!b(c$4#JzgXYy6=(X>cC{P_F)3uU^^|*bXS*8UkM&l% zwV40PDT%u~;T&;?N0c3i3iAoUb2%rg>kG>uVn~ekZsOEsNotf}-mt7WGZMUGYFhu?>tjJ1D}E zSaFv=TPn*0*_w5_2&u+eJ+_Z4P<-J<6_Q04Zg8ely%C?@O!o@McMc=1V;IJVyohkt zjNZntR2=h`%fTIR%vpaE^Kl(1NQ$ghHrAg0Nx-m?y+-O5<;ee-k%VJar+S%T|>1*)pFXZL(QkXcSg;gfG)7qtt4 zSAFI+FXsLCN%Fb2cn}@K>4GzN|o0Oeld#mYa zv5f^&hcyGm&jpN9oZD})pxc)m*_xw*E7{$P)y$7%5qzw;40saW;@RH^p~&6w0i-T`$RuyN_GcjO%ANA zW~VFKPtpzX(Fcfn>h;#z-yRvuO?##^RQ?fP*1;4K+8S9yC{@0BssqJ6GyM?3)hsTk5m9>Mc@ z2YghJ>8^P}JbGzg@>C#B#@FA2L0BlPF{mb!u`bc+99;CMuja*vCT_$UiX;%}5*Bg7 zwrZ3<_uk4}(yFpqbUWD8%lWfGG0T{GYzxWMX?w&%Jt+1;;>YMEtX31(PQHP8n0X!0 zNU)G6Bod%5oe$vopupOz~io9m{Y`@S_&( z-0AW<2UFiZ-=67X7s8_Lz~F?Ku2-lr6I}CkD|68lk3u(bwbC)mBz5XVl$$4;6powK zsoxfJW8W)iW_6xv$J9y2B}-M2#wj4E*NaG4H*c9h^lND@U@b| zn(yHt8#;HuBjSluUco6;-LU=f3Pio?&(~QmpB)FWk$dz%PuB)|t7z}(fi z(o6Fe)#XW7bc!T!qHWY$*I18ke?}bg>qcr(5G;G@Gxqjw(PIV1W9r`mCXw-)iTep} zug6C(+?8b0#gCGez&9HkpgqZy}>2)ZU?=EW5bh;0X7WrOYC1-1o=>7dCrF1@im zOg;ao^$Hx)r~ZtApbnIMkde<24pT))!zE8%Wi))+32*$&qd)#C;u;#_Y*B4B8(~b3 z;DSNw1sJR|XTs4Lhj_IYpaTd=Rr$ehK|U+;U>wL9BAQ0Mdao#M2e&p9UEyOw(C z(y-+|q{AOtm-no7Wyjp$YRqUm)7Vq=NL-(X&1=K-2t(@;IZmA4dq3>96D`4>f;b|c zk&vpllZJ!w?v6%A5L?<8SIkH_xANj}5f2#*EG$|Wbfw39dAeyhxi;$lxqk7SaYgz#F-{$A zPe24KfPo?e7?e>w&D_L81}dz5h9gtpYBF$_OppsaV75}@`V-U?uh_ffx|5|O+zxDn zRBwoE)NPb}U*+zB8;x;!dfJ1BuqasG8TTYN=lcm!wCh*5-JSIxukL}5-&f~jGsP@o zNWL+~9k;8yxm^JovfAnwq|6M8sJx^YwzFNi^Exm)2w~M!jsNVthv+o1y!GM;5WLHi zeOko@!Fv?u$5eT5xeBq$U|>`S#a>MteRWvvOssM&B(qnOEGlOyT^kGck$AFgoC4!h zO)d+UPMdACPP|1Rc^7nT*4uaQQ7C286rMrr&d}SvV~WLkdCS%1GvukA(u6V!hKZ2Z zI_{J2=F0bOrVD2_DVZ16(oL;%3YP0`SHoiv2@{U<1CnOw99P4^Clmu^yKWqGlZYrrj)W!<^&l$5X%Y z&NO#4;0L)QxzF|9E}_t({cK$VB{SE{SZx$Lb*_?LW=7bD@qB2t;LCV8fD_be^cOOJbf_36qQu1_99-ZO{fG{`mEcGSFDdoU+s zrfVKj(rjnp4d*}D-;>L z{N(#%^`K)*7pFB4MB_?R!LxyG?fv)`EY6$b#P-W156ceWF^wB|LVHgs8#^h zmHc$sxjTCY5oJxoiPBmx8NbJXbhoe07w_F)EUsvgP2<$FIRlKBf8F}ZZlmtuE9M*2n*{c5va z0UWW35??xu^??EV3ew&6ma6so1%6nwi!})L^vCj1$@tE!$OXf%$}{_6-cMrkvX-hA zd@s?_YTB(v=Y|q2$@h{EA3iFWkai7a;CLTE*mwPQjJD@rGaBUTeb_OjkpIjo{*7-Vx{OA!6xAF7NuYvELUsmiNPgD_3 znYh__X?hbWs1tQfm&@DzCq0| z$l${daac%Wn63!T^iI=;Wt8=r_b;mwaN|ip$C^v{2Qm}6#?tmHfp)KP0}mQ;Jzmq7 z?P=PitUREG3{W|>m4;Z?DcJV&9lIMQ+Py{WA1*y$#`Q4~skY*z^nhAm%V_<+GZ)wa zQy2@DkMnm>vv^(+Q7#yO=1>pa%y=a4RtR<|KXgAWFaZ~@01KJmk4c?aX>0!+KxI~G ztixhk4e(y*o`J&m-%;V2i({+PyTcRa@QWg3&$Ih&PMu_s#iQz;%v(8un(HRWypRj7 zrXS{SoOD%P!c!Xa-y(^2UD8BEdG5%wKAv_#TV9XN`Gl6tZ1pWw z>|K6q=GVn~&uYj>Oe0MS^4|N9U31U|scF?+?|9L)u#XT-UrB4Za1r~?w^SsU+_=%b z8vu&jH0HI2;z7Ic%<9cwX?{!v#%K686C9NXSeoDE@k_|BIfsGy+sv$C`$mgyIBPwK z!-sL@Sz&^W?FDO2mwXseSm6h!)Jud2O-A9S)!)fGgwTMV5VX3ICa})=kc+vCEVea5 zjd=H^w)5m$Yf2f;oYMHp^|;92n@+Y4x69+a3{WF%D7xb+1;iUPp@wGEhH$soLRk_` zZ}DY%-1&JpeAGScLqe#oUgv|}7v+|A_n=5%X(UD@1y8rZQMeQdq4`}uYWVvSEtkuY z;lb6}Zvt$+tHb*sk`)taz!)bZ1o-{iGuG?~nvHIgSzkX|vKQB*x_^W=h8kr%_MpJo z8%+~G7vCg)w^4N0i*f>v;K2h5wS&&gzP$Sk2?1#RoXnyN!gT{%IXYib-n!_3RYBM* z{V^(@@p{5A@u17wU7pa(^nk&+iNNDpq`|x|bmb0fffc-m7ok%W{~plpur1Ed!cWO< za2zfRvgJ}vmj%;kmObizbab#pcj~YS;*}s^vOI_Mf zYpq3qye%Ms_D_G`VeamS7SPnh`}%r#*y}~3->J4wkXq%Uy|hb1psEiag?mhevDAuS z+vvO4U=5JFN)N8vx7{8I0BETc9bDEuCBvNgSZS9LaeZY8zj^98nv1iJarO0-?^)1F z#fhCUjww&)Z<+nDtcMT8wI*-e<8c$tFt?POt6x;Ng@5;2=CJv|Dpm(4PDWmO^i+{8 z;T6m6?Iz>&leW{+21MmGC&my{XblE*AQu}-NIt8z@&kXqL+5^fn|BHNxLwa| zSZDj#G>BuFv(xRKl`?2X^F~b{r~GJV!J*MJ_f2#MvM|JrpGq|l1Pt9!pYM{1g8a1TMl@?%GueR*i^|pjyz{b|bW5WB zZ!@AMkL3O~da$_WL!O`W*K4&@(0{QIiSa@wBXuhP6u)K(HK(X*yOQtu?tUwpCA_*z zssujgZ0_EvKqU)e!N%Om1S#)xrn`*<&eh02J23iqhq#nXsAL6ZrCYWJ;2)*S(U5n} z<-XvbLE0s-q`$dvhah8eziyG?dMKI8SQl^W2~5lTvSI86-u1n8K1a+h^gN3b1zRo9 z+)z28NXv&fza1o}R5tPHkoCCn4*BhbcNxG#)N1?H%|F7I;(-ZledVL5Je3r4?6_xG_OlnmmJY1y|gZbY!Kq*R!h^<=$95{WXGK zmbx3t`a$qm@Amp2Q+8{t3Y4L4#$Q$1J2I9qXWPAk>tznhqZXP5OT1X{&mHoqR@(fL7FIh+kxHc0XiD?=; zGpaeU!w3Dj8L=xMxtr6O7y3)rw@@#K$%E!9v)Y)Y0^qKuK*RbtRL03sQX2Hup}y`9 zO-+-C(Cj>)-zDOD6_TgY7XM!pnq|@=zp!`ecgxA@?L8{>o@tuPssW zY(06=y(XcbCcZbp1px?~)4So8Q0KjylntM<`_aOKe|{bKZp?R4Hjh zM@Cncph0i&-Uh~8(B>&eV{u!*N+v$;q*26Fe~}iLaIN<)*ei#;jIFmQ=)pq$s;*=kJE)$K)(+cwLNP)ArHqcu?A2{sS0@1B0;yQ9# z#)f01qkFsCT@8syBl7zmnI>8Z7fIxEMbac)Lk{t-gE`H3Ggt|=@rsfVQ8Ml6fFXmy zHSTHBcLCN+;ov%PQS7bNg}Q(r7zQ8MVb_=L+#_RAViB{r8R-cP92oDmuZLZ35@eWb zhOtCf+bfY!h~4c{I&fQYaiYre`P6P2+EMAaGK+Q6o$vO9fgth~Cwg4;;pHOHhEn-F z628fbW3R=khR1=Xt-*XC-HZS5U4^HEamTr8X|)gA%|bx5;C47l+O$-TEEsgx<~+|p z!WroOT*w7%Zy2FAm-zwL3ZtX-2&)@O~~yJHicQOE0XmQ<6z(fF-$4Bf*XhxKU8#xWy=EHcV|vkIRF)ardF4bc^4 zc7KgP^O3o!OCAAyFzWJp*4zGK>PesX9PE8wB#?3A)sCl!(_{?2#C;>;$;C{oJqIT5XZ~0_DbN!aowcPH5X$~YqVij-MOLRJlQ8~yaeJ2=~B0to52ts(Qh9i{c_bq z9d`2eB2@(CUxX*7o{9051+vR#qb%Wr<|Nm07J7EqNd6}6r$O&FF#WUJ zi)4+kX4?Tk82Nse>Up|Vcl=NvN2A`KoxC#LQ1Mp}mb;l-sJ@;ii98FKIt7+Qe6h$> z3kU?(qp+ajEi4Uql7+gsQ=5iDFY5k%#GYCq5%5gym!FI7rOGKZ50<=NTz$?AtI9Lnhb zz`XcUqV1~Qs_DFQ8D0(0QVE1P{V7wJco_}rzt!C*ttU;MQqO8~ey`I~ zRE81*@A-pl=V0@#jLmPB`bBM4~Z<{eNCS*s)<u3xU@Lr# zY8xwXHYravRvr(G!tMxQ{?W!&WaESYv+bINq)Ib$t~KP#e%T_L7vvvh_b?fgf*3_F zJo=1R%~xl0kj-tz<8j=&m+0J4va;Ic)3?E$o1SzEQ0?H3wIL9*dGzz(1BENeZ?Vo; z-eN5t_uKtV9{Y9keJq_oc$MDEFTEB`$^g=1>}fJ>TdM1AX>Sj@eo_yed^DD5!nbF^ zo?4{XL}M%1-Sfm(CVGDM+}?(|kjN5?H1^u+9{fe!dkN>C~J$=}UbhV)84LAzdD|}r%$);5!dHcH;K5 zV1?t_KCc#Un%7*dZ{#d0`J1AhwIO%to0-1f%E#;Wenvf|h@e5uGj`wrj+5_|i6&!ki<|u3sCE6SF3qk-d zK*&~p=QZ86ottj3f#O@dCk9GZ25CKY|5mAU&pRwxs$@(BH;cA!u~t^yD8xsQ9k2ZE z@zuZX!m}@|>qjCi$S-YY^PoO5@apP!$HNA^P5NQdUsBx17x7y_h!rnAKup5@lauyn zuNDY&iEmiD4DYe%nDdAc+xA=OLW+Z2EOq;EpY#o|VxbYM!%#(OT=lx!ts)WJ)zc0^ z&=AoxvB5m<>%?yk+Ye=uR>}C%D&1a856J1cRoW2gvA!0-U4%Kp(qEGgvNvfsZN;~K z^+E^9jL?H0GrhbO(MEZNkTR`tqFe+Fe{KxsdW}-pTUR+nZn2;}YAm76X;`%5v*Mti zj}wGJs{)Fn52xygCp_f2i*!zN4oUv@RPvpAIzK8#(B}*46E&{3-2#*oahvRTZD-XO z;L0gyETGg`W7L)PLKnTN#g%(Mg*JO=9j_hxXLlwJFyO%6e8e1+jkQee=QeCh?0~nPU3B=ZYxqEE9{c-<39O!I$f-d9~MdyIY zyEyJWsDCu?EHf`4V8-4X66yPx%2u_jmi|?t>UDec;u|GwYfwE32$7+junh zStC6=r>@xssT#qdG+fKTkPu&tm=;E|?G00^uYKq655feU!!HpY?knJ33kQ3*wv5N% zw(@-;OV`%1=L|!vF8Fps)bq35-E<&Lg*8NccYm-;04JBd#&BMs8vY9?Z8^pL7dAVC z90_zS8}tRtpgwJ}R9d)&`u2n97cN1aPBXITUMiyN zI^nL%{qlGEbVDQ|rf^KdTwB@j(MU2~cI+I@h*ap#SY$ueH6km|NanzrvuKAfM#OaN9!KKDCP39- z1M2OtM7)}1ORX>ErhT5WUE68}qB1c?*bl0*b|2$ai!PqFNYWRh6HmE+QCOIEzSdT} zQ;JjlP7JG~mY1Zz;t)i{HV2vYb~HZn?zDh7;0$ALC`6uUK8~&tg^lRnG2Jvk1xv4& z9qXm_%FQuS!DBE9H+J@cpc^^ zMc(v%=nrq=nad0Jzy_I(Du>YL!-_@QPnDQMBIDq9c#j@1Y_r2wN7s^0FoJYl722de z$7X*$(ar97w988{-!K!%*U$3%@V8ZBd;DxE!X~K#9N}y?zF)bwU$tF*k$}ddE*%1W zkyv+ZF8!(9MxXDb6OG2<+ia}YEF-7*dKPc_?%Pm;Ak>4Q|HlLWKRb^@wpu5~VrB0b zGutW~Sm*PQ$+4tA>uHp4yMset)icPhPx7(Eh5y~kQJnPF5aY>7S=G(`<8}nI)>}Y% z3T3mB4;LKtavOf+27O9HJ!)c`&im5h>f&~u$2k@``P&%^{*yR+9=lL$e%>K z8}j$9DeOmvNH19!Ng3t6!jh_e1)JrT-O5n${PD6}e;?mhi1BXFR?Zo0p>GAYhYZi5 zG&F)4>P{=0`&SFT+$aRXq*Y`2RGr-mo21GciS_%F5L-I~i3z%0eZ#%>2!ndUOj4yj zCBc|26yy3OXbDEen(k>@kA*U7 zH|EgLnGuNS?1;?XY@gX!@JRXdz(0f}r@{Si`S2HSSz(!s9WE0ToCAgIY3WwT(;0p~ z9U3p@dTE9D`Fa$$@K%fs#w+<~@d(|OMWf%fA(^4Q*a5kcPS*%XOjqoY%D!eGpkOBP zO|t;*V~b8w!zYSQ$%wymR4hO*P|nl61obKzAW zK@^cJY?-)tk=Hh6qKf`F5(m_gUNkp@8_2CVPwcf*IybGRuDDA&bP?P8WX)eVk#2dj zuI}Hvx?rX@)G-J8?%Z$RtHwP%gv0Ts&n(z^CkF#D(AdNo3_vr zoGz>9%eVFk*M3$bnk*zt0lMno1}Rl@Kk=4Rh`}nsEQ*uZ?z~W|O{4T7(rkp$nZfvs z4_0#!1QtkDpoaG})fg##qI7VJeIiKmcw-Rg0?(;LeK!DZd}E7f^O zt5@WI^qaPg5C)kmkx__esAyqoZ(%o5*Y=*Cf)#V?lw7@^bmzEJidJ@(=-`loVvz>! zX7PI1*#I~q>GZo7xnNp_9>KmTZpnTldluy^hBLWFSC2lp1Mn+-Uc|_XN>{koi_Gn# zh-9WjNaTv&w7p(wgLzu*VH@1qxL!Bq=}~JtgMwCy*PD(XnPIgVL|WA^UCub6YJ;@4 z-dti_jZINOz;=9yhi0vr5o$K0D;y~_r)e{Z%^qKPtMniA%)nk99WFXMW6ma-X=bfD=&mJck%;iMK4_GJ+I2>-n;==o8>3F$Z z#}N?%5^5X>`(t@sUe{}YGGAzt6c6b!#a81^{v1KgLuN(fJh@@nq|jgZ4Owy~ItS_W zf?5Ul5v!lW$FKXf{q)VDQ15x1l4pGkNQ6pjnl!xF7x+<@9YZs|S4ngISrKuamqv)W zao~iAb!QcbwccSPsUQ6b!HUZ4AN3W3CcaM^7m|GG0A z7p7+Pygz$1DACott)KUyr)PLl=ggj0H^f=4H{V?Zd-3K|ejTGcg|ENn@~VBy{cD4O z8LZis#5HsCn!x^D*IEG%&jYKU3h^nD#I1gpU9eLkwi)_EdOsceQ;98LO-hv8)>867 zR)+));c9MCQk=(w7o31rU%#(oaWViQn^rzQtu8#GsLoKgs%TW`K_#1QL~LRiA;PQV zj0&30_V0)BRWU~GW%vHrMDGq~v!wbGvNDqRSt6c;IxjBU}Z~nt? z3Lq;EY^k4xX2TPlFSth0I1%#+yUmC1HeVM{RY+YJwQ&>S$dgi@f5|1|P@>kY4`v6C z7smnVf^nHCC?-ApVIgrp#`;=Gd#ZCswp9+e8E0oaA1ghbHl`c>)* zCS5rW^>1Byau+^FLAwWd3uE)3Jjn+%rrjC}U`EP(WT)mE{~aOd^B1*_nb!bc;+cKh z^_duuUIS#EDTNSxhsEd|Nc1P3R3IZx?%rMyK_5bn=QsxDPS;Ew;xUUH<0|~9`UzNg!_5~&5RZT z^j#lA8&uMG&ysXGt*&!ihb%xi6+6-rhJk_$W75NRsH+EUKlzNBTX~r>r$e=kJW#Wd zhAn@6)}*n4MCt`sP1bIZC2*rqhitD<93)GVGYEd#=hn0@bu&>fvw01Id|X%Z|`v*Yj+sh6C4kz{A_&FQIG|5f05Qv?2o!6C!9EZ2N}F|A29>v(CcrxS~(@FAr`0scitZni{Wb7{Y7!jM90d8NL>wgZ3)Sjpvzrmw zS)rmW=wcBiDEGw)qld`(v|&VWyM7Yj)PBe*Eosq_C@F)b(&L($FXxRwozDe~3HPJg zakYE7`>)_vgb$7r;@LkKm+~O@I&h+%~*mk-Y>_ri?F&#dt7R{Zw zH-8FkZ@he^ERW5aS)qm=&4oeHo%S~!^49nn|{uO?|+ zucbTe9F}{CZ^`ysCdK(QJbhva7oKJM8`Hk9}tNb0UpGf4sq#n&Ye$Q+>3-nBE zog;i=I8G&5_0Xu*^ao;L;n45o7Z21a%ZHS-R;nViBm>|r^z^@*9(v3WzsLc-0x&iZ zD(SKZy#0@Y)1DD~ao(}=L9!A27Ktv>d+&sHHPT9DocQ1Oz~25IOlPVNI#0Xgd_J)T z)N+Um(mVCdyc(d4wS*hj@cQU@ei;}GPr%1dkP1D$&LH7UAd6;Y-G#iuMDXy`9m4uJ zu^M`azO+*4U1YwXFeE+-1=~TWk9p~4>+W|1ugt@cGr#N2M3$0HCj@~HiJs&qfv5T} z!~P+l6E2@o10J8pPjK_yd1vwTcc!|d##IuXHdqmnE+I1yN}^O}7nyv5SoX9)E8MS7 zpYbrttBW07Q2~0gUTHL7ht$rp?t|JaV&r5gL)zA}avVA*+GAC#L8tUNt$|nl^ z$TwfZ3U{uBz4+*1?4c_cCKUit7ymDhJ3e2 z_}bq!B3>-k0gh4|PXJ{UNhBR6UHo92M{jTNqVI!_rc0avj7nf%M!PwPGr}SGaCOuv z`nnH}rfVMg`*GV|RCLUYA1cz12&JWGRwQI-C_tnrzCm7LPmvgh5^^18t5I@_ð4 zq5jI!;|=hGGzt~_4W694zt=Hwb*P*NzSoIixT@v${q(>X5}5%mSWDW&Q;Qg14L5EU zF%|nE#t-A1dF6*9+3n6JV5~1TD4z~`K;P;D?vr%i+==JiIUFgnFufgr+;6#(em~BE zc^R?B+}Y|GE(cSC*p}6X28tOYP<0hpSg;~X3|j6jl3S;=0(Cc_A}VBVDv^BT9J$?j zeKAoqU4Ib9o7~oJf*HFNVE8TJ7hGMvprLU0K~W*0A&N0Zj|6v%TKAr?8F!T)9|4G4 z`#5r{bqrta@pXV4NwNa3j^H2Lyz}UtF@8y#3Qfx;gb15XYEX3JmIKk)h+V|?JU?Rn z+dfda&7VNax#Ot@?3^&I_C^x?-RpUVLrJBqj&}LcTu!KjV;y9(J>J+OO>7!);@kJj zBQ#CPv3^JgwC{@3k-8eo^gr#Q-2ujyxIaIa7K}O+t;~@&Q-O^w1scatUjM@;0iB49 zh^RC~=b|+G1ApuS_4-~IYt0qBeO1u%`$l0I5XEQ#psTU`C z1UAS45pW5Qq(a~2Tbu}2+K5rI0oY2vUh0DZ+6|K?u2*XK+10#W^~Qy{u4j~|5TOiULjclNoeOXUDFK!0(wYB|rhznAFovqGY^So3*WVc>(+>xGjL|K6ggPrPcY@4^}aYrYPe1$ya%OnPRQ^P?x^a+0^4$fuW}`=wZZ15HsPe|+`^;m#UeTMbIz z->Kf7s4u7*Z^1YLmhIWzVonT}($(x$qDY5mU!LRzV8P!aAH{ z=nqV6EaO?>5@$G=^%F8fSvwRnMium=9fR`4USoO~U|KrlMqh7pEqV*EMR#5tkI!)` z+gPb51dTvUDV9)n)Lygc39?xoP67HZ<3K*3tHS!&_@__GX<%9jm{|PE;DDBbAQ&>S z)pm_uRVLVEaER$LDvgfTKAY0D&9eZTEa6%PMotggVxC9+RdMgDR62T5iWl^%FNaNT zkE;jYFq_(XntS==600r7w(vr0H&$u{S*5pXNzySs?TU)9D*jw?XCsF@2;Y`9R!9C> zjo8aIdq1qTdXrlt(jl7~yq0PX{{B3;7FTX^eSk~09sj_;dRL-VlZm~zLW!`RXjpzw zy6LmDU6$Qq>&&VA@z&S;m{%eW=jrbAeAxLSGvR-Z{JN}_`MJwc!b-anM?IUtVJ|Ym z>N;9Nq(pox0%Cr(c~R{Bll{4i(w(qr^mm0LM+&b*)3SQ!M5-2um*|SXZW-tAh1m6` zj?V&T>23#jx%yF)bXm6Dxit1Mtpgb+s6P#JuSZV4k_d(yEd2S3IM@5ud3x?>R>HHH zj9~1>@3OALv{=2-)$?cz==vC8tRW4pmf*(rf&36Is0EpBymW%G_g@|w6N(rVDYDZ< z(tHVitBwe$k%k$u$=>3#DY@n+5>#8YN3`P;d-?4S zQnj4NdbDvE^a9(Ij;F&bJ|Wx0kR|@u4}m#a)lKR!+i!%3Id{V2q*TbP$Isd2TO)c& zf-EV>>&@~D^7#5Ao)-x=b*@nagSBCS8HVHtduq8qyk`7ukRR6mpltLJF3-tB?LyYH4hlAqmR${2la#uD_UmuP{a zyBp zp&b5yMXgMAA=~ff_E06#Iw_`puSJ{lWpr z_nd#M=lVEE++lIYs!LW0*WXtMvIm5SB}7iRNM7}Lp8{f6)5Svt93*yA^s@mYNx$X{ zfwr6?+J}o@enH-n7RBX7JvEVKuD?6P3BuxUk8$S;fq372?i4y~6xQdT_DnI@kcCI` zxm2Jo{{02D^uE}wnyuv>+G!GoQrio#yS15srxP-uJEr|TQo?ci-W_LbqD)+$@y+ft z#i?QwK|(C$EBJ%KU+?2ZnxMf0Oi;UBvEY2rgtbj{Djx{tZfuO+ zaC|eMlm&J*esK!_bu>kCN%g^DD}WS`E}HO%r!_D?U5?kcz~fPqbvm&2 z-;+I+LdFxDU0Ll@w6^-0APZ^Ki`w! zs}pVBd+^gwm?2}`m`YR{jL%GEj3V0e0{Hk{@1JE5SmZ9TqgLB`K>Obawh;Jq$})9}d5 zTGnxEC%&MN$E?wm49Hj77XO=LLt(B8u${s@IA$Q}{A*Gw)#d$ItCGj5dgQ z9PZ`ebQrft?7~V&ql{EwnnCO8L=XahRARYFvagN*SlmlH8l`$oVsrC<_v+AKl#Gcn zwj-4fPDg}y2~LF1e5W19qaIfPc#*pRQ1a*MhOVk^3x-MuW~G}X<@oAeT((#u%hmmJ zhl#Z=;Vr?gJqrbq~4bRPr`76XasX=EwDCLz3DaOl{mj+uoYajN>5F!dPaoQ2>&!ha{Q1H z^4Lz%sMHC3%#k5@ICa9iOhk`vSVr}$VtLS81^ zkVlH0>?>Dl6jf)tYd94KCf*P&7>qTG__=n%Jq2AQ_HH3c*435Y&AlXjdpc_9Rpb%C0pQ%g%qeRsA173I4!Vjfw(c<&Q^!_vAYy^UEAs( zrc1G_s{oO$x~;$0^UVWT7JJ+MDj>#{^$3`;gyh$Ch15o*6wnROukv4i-K3X{fp5qz zr1U+zGu=h}ei!$LiX*xkxadtqhgAZ8_;ktnvm9UFawBJgI|B^go!5|-uZtOkG$fwG zCW|BTUxX34n%C(6&eT?3uO*N>QQfR)NJ}AMbBc9c|c-F$tFZgXCcCa z>M2!DsUV52+U4YQR!t;BbaZutwy1tQu?=R{g=lF68J)N%2fpmmjwg2}Jc`_^L58Aj z9Dc&^U4x*#n&EE!S6b!i4d5~qFC8GSLy*iqF^(xLw&s;% zh+&H}4LsoP>m%(eOJ0W0SjkU{9i8|rUKJIk8pl%?w|+r<{{|L&`Frp_O!^xWq(eoX z9(It#Tu+t;3f&1_ZbJ#$jXu8+M2(7j6x$h0?~cpm@3rlvl%;ZolXy^;P=4hqcg2{< zzM_=g@2D&A_eeMc$f2p$hkF^z((Aj4&IFmQQ*55kcPY?h&dB1y_;rGB)Nl7Sjb!MZ z&96D^PdD3Ojr3G^1mI$@IsT1Y$s?G5_ZB82x=q(x7%I3q?5JIIH*XeV6zp`J``l3K zm=C3EWkIu4$alvmdRHObS9yH9E|3#bl>1bUJee__>WujpYI4+w!KM0`8p~Q^1M!D{ zxZnJ8!?Rq5KgAhy(@EU&5XJvhnJzWXfUC`7x) zRJh?mmM85gpp7BJ;mf4(#6G;Z{v{i%wn({bgn5Rf(=TL%s8c31KquU=M?K~2$bOgG zNPufhN+kQ-Nys@}%zbf$8w{%`6gS>0)u3QBW#bd=eRgU$&a*Br1SNm924rm`rP~h) ztZVxCmd6*k8lb2Ua^d}OK)cYlgO+_iM2bO?Jx)kQkZ}9hrwEEg31;(enw0 zq>e?%*zwZn{%WN|A-U!K9NbqjkH5UYD>b0;@kFljM(i^07}*ohk>MG)3XFoy3N02a zI7*Hqr>XJtPl4WE4HRvz77MKuRgfUhu-A$lTrgU800D9v$YO^Mtb2~LA&X~M{tg7s z67G#XUnT5!>oO0?mPzCaP=quTK423G=rXI&fEt_}#?V;;q|bIjNVXBBpJ@4@~wQGS4JG{-$ZB0>}LIT)Aff>PYBq( zW-u{R*y=z?D6|FRl|otz+yw$^wikzoQ;x$sDAPv{QSs~kZaxNs_TGXptYGIuQPS$gSSkZK_3?!)W>FOzawGPj&b^ax9^eQLUvbSEf zk^N_cK8>n%{W7_f%YZ(@TsorUtI>~#^7%oMqfhvwzrRT}k_N;95ID*@tqchHPb)Q5 z(}a+|=gLNuTX&N&z3k83+M5w-Slya~KzWerK2IgoX2=_6e@E+(QWa*j;__B8$J%D~ z(1J-AwT-=u)ry2dnmpA%i*;~Iq$r(?!`zqZ%3z@K@em9BVHgBgA}fw1Z(Ss>Q{=8w zYF!`eD+okzxqfUu*&Ael7Xf!qqSAYG!;z6b>aICW3MG4RGkOI=>%N)-N zP};y#S}~zkZjYNbSdG+myY9z{{89%a$YH>|yi@VJcPL1*VU^1Z;p~vB(kU*8I^Lp9 zITj^l(JOjuCA4ftC)Ru^C7AFC!TvTR!+(C*a;FuQjv!lHKk>}kaR5;0^|VN7!A}^s z?3x()J3WVo8Gvl`2!UFY-!TV!6?Dc2wtY(&GbVhu2fdy~l9!C(XkJ!Pgjo5G6~JI4 zviNcPkR>84#kOIUt7Qh;B7R@2{8xVaYrV&bA}((&0nfRTkxqh87cs-w9Wv?D5j}a) zC6$aTlqdZ>b%@n-V1?F?M;W(5PU515FIR+VeZl~Qqnq!6(IAayVzP2hy_y1<%C(3Q z$Y53P*J!>k?7-FW`=x%;x4L1^h!b>43KeQK5)}7k2n9Q2bj$Zjy>S%uQ#z|idtlIQqI-b(S^IJB9xg-MC*RSa?p{ripgmOig>s0&sf0g1xe z#vQt7s*Zzu|GhH^FBb`Bg9$c&kg*QfSZq2!@%*KRLR~;wh+PK3gTfrA(+ZebE1~Zp z0zY({ZYZRU>;-Ycs=;#-gG4_$ZeSnw^z|u8Y8)^$LJvp7PGTp43hiL2(qSrSrOM%g~C98$~w`Yk8^>S=M!AqG%Z9`R-ZWe zuqxVLP7^6GKy=7y={7fSPNHpz2{$t6kJaV)<`p@@E058>ODiwFT#x4iFpTw&wETGX z6nn200PMa-Og@eN-b8H>`tL0h!z|{K2809X?OBkdQvU4Ep+ybqJDJWW_ujk>4zGAT zD$9ZdoC_d&>Zt^t4&oDAer|)WxRP=VB^iMsn#H0|%fh;MUr`&re&M;W^OKM0LMPBK z0bwr1s3mjQ4Sl>iTgRvpvc8w-^{dnOSieg9z!3YFkptpaiHDcUs3h?rxMp?vgy&Fw zX^FcuvVd9nilkXm-UFrUCFCT|dYxz>K6yg%6arc)0{z!c->uW?!tcbs`ht&qaydu= zFan3GkT$RpV0R{D+?04rJb$_2vIA?$<0{-dB}W?8?89|YOeEVLV|+u7YNW5>J+#r? zB9jxk9TxP3osRsS!}W47_Wqlc`h%tCLYvtJ?mxM-iw96DhaUMl=uyDW5@0|!KIdGs zdG(H{Fb#9ZaE!jFr}qh{Z?D&Pi)cmu9=AzU3U}PK;jzOpz)9pYp;-=ke!My{-vgLpXYG{FM3sTY?=K)C{$_#=}w53K+Y`FNRNjh|Eo((iOQj#W%q;0=Ah z-`srbx0clJp*_Lzdc@RmdIkV1MPE>~)d6UJ~IA^Ty{q2uJd zM2q|Ha_u$5^uxXBEOdLPuIe}E1XJqW0q(j!Ru952h7fGioq<$LDBZmPUv2T6_JW-X z7F?iZ!bmZ4RLFM(t1(2eQwtB+u_53Vec$sUK5?$ondqggKD|2}5HVWBNgpEy7yuKz z(qkwhDk47fP}2Wh@n_HWG8$N#GP~6;Puy-c`c~dQ0;5b%rwWF_<8m#0!SC~28K$s{ zq<9>^Qa!^-Ym1}$@=%vjkVQh-^!~i%XDjN4P9x!H8m0iV;Nv%xFR};Ngpfrfx~bR8 zwuk>Pf|TuUz18F1d~U3__0u&X;RglwSrrabz*HZSjecI$y;e$gshY`OEns3Yy@#fG z#@Q9`leEA|T$GH)5{x_K)_zALRD{*60->4$?KF~A9lSUB2gLFOKxMo;%&kb3{LBzB zS&RU^+XE_t$Ip2r7`N$_d7O@!0+QY53Y5d}-EwDRkFSlExoRI(1!T5_XkXU!6iOmc zzi$EDV*uhfDNbGrA>Ka^chtgcFA*Y0=>ojI{PQBf>-!`P!dRIbIH$Q3WMN}|Nm9Q- zGyL@VVU7nIw0&CbTmMTVaVfKQ=tNjHthC|H;9$?@Vb8wYhW~HwH`}Wl=35GvU8e!& zWE#WuoCTl~mh1&YRVXn@0?BdzC)-^-0rQyUpODy6;2#v!)xm+4Xt}#-f$w7BaKfG; z!rChUJJRezcZxJEe!Nc4w=Pk(Ih%(Cq*TXv{+S#c2K5Psd1$_G`=vRt*s)+Tm4R>c zAEc#}r;`+je2Ki$r<~K73pyH`;dE(jJ^0U?Am5y>=xCe^Pc@bwgt8>$cn#?rfNI0> zf;E4wW$OMdI$v10x4=45w$Glvxr$+BziY4G7JVSp1z9LHsomL9C=;iyd4X!~-g%7U zXYe>cs70jWIzIIR^wfQAwz!+0==5)mGcxZXSl4dVYXC8T<}PS;@_UUvAv2fXH`{k_ z!?Yoh7~DMD{!z^RQ~b?6$;A6X0M$(2!nFy*COjTFK%aZbrVTcxt{6-(_GPz@`Of%? zo6krY)n2=E8~#Sk)`fn{F!i8eam>*E7e}AV@S7$&UFD>g3}f!XtQ^8&=?Pw3ko@;a;KyD zp~@Jp0kdHlC?;9z#VWr1lXsw4(SRjQX+eat5hZ$JZ&0(lX6lQChhcdo&EK z{vjw{eNpb6h$i>lUgbTb`+`I#rXf4Hn1iJbvb$VwbclnTgUfPOa*kqpTp<^V-#kaLWyhy7Ge_b2kmlcG) zw{JZOU)1cpFZC_op|9`W6#9IXX;GoK7d7h^$N;9! zns$`YZr8eOf?uvFuX2=bGx}cC+s8D{%c*Ufo@D>7pf2Zh5H`;S&Q!eBBy-=v6doZ1qxV>*fnJu-CQyJ=&yoaLwYS@vv`7x)YwqB zeEb1hs8kC!^E06krw$+)E?T~u`g%Cya9wJE0X+e#oLtm`%@a#6erhr@CWR!$n)bN1 z9;azZRqQnVwbu^Ig@4{?@WJtC37iv%U@tFl1w&w5Ui{amH-}gW!#}GZ8BBGb!vE9! z&H2+unHBr*rvP-6Pr29J@zPfvQ3V~5B)XAwRs?Y&PtK3cy!5n*i;Af8G1p%-RJP9x zu-2E`Wo-+qV8G5BQejuaE=)=#L=-ihRK2cg*7gwOd@$d1!`2ScSLsUmx93_Kv)Ft( zjzMD$BkB>93s`eqBeCNQKT7C{N8*GH-YA+Z0i{JRWq1yNFrGKFH&Xez((dw6m z=^)CR&rS=NNPWlWwM%*V6rhno?1C3JdqKtJ{k4)7(IBBt7 z*qJ@a&{xVd@dgW~g)4k~xYlF=4+f<~*z@ZrIiF!}5PphDSns>%{j*&2Por2Huv|Za zEf0x3lI$sROQog^WIk$1|Cq5%gI-ccilCn^*j9bNR(gRvsrB`avU`zRE#-CvAn5FS zS>>gf8O@X+o`tU3NIVE24?hmg2O6KR_wNB_^f{mov$FXF=v)Q!z4vwq3i!x`4LX+8 zP{PZTb7Z5Y!H!f$b{ny77_~nO^z-G4qE)*^Oxy_?l5eoY`|EJ!4+gMU zuUFhUaJaMRdBu=yMTB0Gh$4I${5&Qs<7yb-7N2(2uB|}`ANF258fPRN^>JA4$b0%6 zh4mw5Ek<)4Z_r&||M|jTz*syvBQL`nSmQ82kQ1?Y-mh2&8mc4Z@R9cVVk$rKjT4F* zbws#_L|k62_)_Yic*F|ix*$9QyARy4 zFnA6t|QnmjZp~o+$m7dY$it15hu3h$STTE)%03QD*u2&g39ZDt!<|w2KxY}?x(`&ezbJ)>?tVWn?1#DTxP}B-- z(WR2cO<4-~+SBx5?Z)#oM@E%?F%9P;>O-EMHGd&DoHwSE_9Nl-Haq1XUSYA}MX`R;U(a}T9Pgc9rM%H1q{7!PZ06KEDv4aPYM>0O#cUyA--*GQmMjRn=*pq;{w|< z6@MbN)0^(OVQZS0gT2dy+8hK4|3sPRADE^v}(nCqX`)B zER!trgEO=P+qzc+0I42Zz-C!9)dLnH#9Ryf=R3dTU3KT47*R;>$ia{Rq;5|J4|HBv zH=vn6rK@JbYs1l?w~jsai(kF)AZn@BmWx&dV1gzq8JXJFD1P{yq$6j>K=F? z!cWI{{KgkVUAo^6G8p}nOXlWt4R=Q5WR;3R2}Ji}kU zeS<1&ReSUG8@gFj7JSOHkYC%vSUHQC0MNg`qIAchFy+eecH^WAT5SNyxYT=z(7}al z;WXnDba%}THk+(s>gd((6b=^I^;;pM7?|*1G@}Or0^JAW zgpsQgvW$sss~N!kCz}~M*yBsb5Gf+&R1qf2Ty`vmH+l#96=tYFUn`qT{9BiX8`GeqQCA#Z3Cs zX~=mi+R>rRko&Yn)W>w<9T|ZW4{#uUV}5^~f<_GDvwE)*rg<=-&iNo2+X%5hqWH^& z7!)bUe}(C)Omhwx-7>pxgYt-zuD5I^|B_Iie_t8X41q`^jXNH@C;Nie)aRV&#+r*X z2VUB$G;$pEavb%B^fxDJ&XMCCsWr)#Vhu=Hpf-$ zSZVM3i9^#T5RehQR_4Ai<^GO+sRzjR4%7z?jTU2f=K=W$I##yJcR!zy#PQDi^|zlK(e3uzv$_u20hLl{AgErwAw6T$jyDL@Jtqf4udF@tyuJNu zBu}*LN;!$Bgwh2!NKv@&?~cg&uZRJkQnxl% z1oZFq2Y@;dApktZB7qU5)o)3@YoHAm#v71N?20ZyLiF<)5xE^~cxb7gu1Lzgn~vGN z5bP{B`z^tIm*(#PBH>=}podMIf@*K+1^Tdh?H|6I-Knquux7Wq56CBdL%0^6ka87P z&kLtAyKOokoD5F#Uo-Qyl<&a4!O1xA&l#bcZ~LmXcdvy7Ax+~~h~>Ap4O@p#KQIi? z%us>#25eW0o=1BPfd(tp0V|sfe}D8@eefAqOX$?4|LFes1r0K!O`e&9QOlv90kf|u zDG8|ND^<}3mu zEQUqKi-I=l`Bv@SZ?*TBSow$Nr)PEQ{0M~h7VlC75zQXQsd`e*DIW7*3fH0%gHwJ1 z1eOi@FX_cFoe1-=zvc73dwESL`Y9$a_wVxqhv;(ft~Gc^Tf}$&n*+pp*8)O_#{J@q z{MskJTZaWPUN$#*jQz*XBU|2g=1nBCmo7a%H;Tgauq7!f;R?4kocLrebVG_w6-Bm;Qc)7zI zaYo3P!O;t~hcAee{q6Sl#ky{?g4$$o!?ao^J4n6Qhs;!)T>%hgJw?ckMwhdU-3UDr zXXCD7(!1kc9geRaTq^2DVn3QhKu_Pll^!rm5mtvbonTAe7c3VtzwW#n=qjg?mKHYp}m)E+Y#Z^kgl=c zvk*R+tY^t@I`@)jy(@JG#;WjlHBp-!07c=L5{M>r zE?0E^q43oq38RYB6YrBBvDf0U2YX$9qwgDotq8NW0UE@+YmrFudL@0h|DBI#6y@Sz zA`FZ!66xu!an0v$4n3(6+&~v`ptTsqX>tc+4ZrOy*hv^|P22}OO-5xuxOdLXvDJA%# zJU>4P$2WAFh7~@O!hHj)!IE@fPygNdIJOS7)m^+Kd)+k>;j|c4{yn<2gyb;*q-JPQ zHGjBn(}MO2;N0yR@{IEJzJ>i6wy=m#&aWD{jUV|D7YSuAZZ@0sVNoq2Vy}l&HKYp? z96~74Y~r1I1Dr^_g=Zg5xA@3#44JlnpuyuKBtD?oZKky@JY5lp&3aD8=s3#2?F|L@ z7f-GK9|Sp{X#kEw9%0o$X!IxtM^D5vhu}C$}t#wAjzkR~AJI(GI4zio636 z|9WQE_V>Jpyjw>99RMFn3J{#3zDN6=U~Vfehlcb_X3IdyT4cF6$1M>JmV@7tGhTzE z$O=|+&l&tBYGVnpm=ywhF%y^bV8LKQro|TiP|HJ`IJbFoGcHa7;}{^bA#IME`*yZ; zUHWZAaGoYU9DpVt#~!lmV~+$(|IUcpn;TvOxe^FV(9Ym$%*dyNF-a61z>vPCX&mQC zh(kA;`;8geWs2ZtbOA2l5=&r0H_8??7&zrBsZ8|oAveMj<+W+2wSi1jo4?EXj|{wc zU+QSX@Q79+_;dyjn_e%6(l$x%F68!6q+5WNSd$=Qc{Oe`Q^T%<^b$t21J)ea3&%LtdJ1zz5bLfir*QsYM0Y-J3|rM^-i-?zcP6 zkz@*#EUbNaj?zQE|9q;1)bV{fi1Y3se$vmUd?dd|LcJ%jH!qHaA!4vwQ#+-E%pLa+ zioOqNZ}U#ENVT8xNDabLz$#`FJUW)2rLh3c6pD)*K$Q?vTw!zvy9$wv1hdbEJ$A)t z>f8EwF=i@Q0McXP&xT?`Yqq?wZ%F>GcNIympSOF|Oo$&KR7csFLL~|2zfc{cBN)5% zW9fFGurz^=ej}o^+a8O9tXxl^Sl2#H_m$EtmVboBko1_yihM4v={x%1xLx;4Vj#4t z@=0C4q5oZqSR`#h%eXTa{wp*cKHMSu8^+}f#j>Z1s@%R*r<@>>gW!qb4^IeoSshI* zQ)_M}`epkSSB)l!0cOzdpii?|q&@Wh@e7c0gyZZ3aQt;nq3>|l`ZSu#lZwC(P&C%8 z&0yv z&q5ZQEN_1Gl{dtpcyBf$9dV?np3}I!=~+MNc5U9svvRqe{23XXP<@zZRzocTAMgS0 zy9$Oull%1m>^5I%4z$hc3GV!@^$(kLY&R@Ggj>`F>?8*>(`yMtdfO}y&H}mQMGc73 zMMN%qV;c>FVh#D4B|JT{iTU>q@JYi$e|7|J#v!dw=Cis?16U!U?iV&iV5lau@^H@l zO*{0iCP*FH{>_o{p9j47q;$e;6r650X2r>HFzTdJwM;;yWOqTRy{Ji6X1r z%3#^v&%*jPtk2Wchd>9GNu*2H>x<=(e)#1Bo6l*&h?L ztHY#EVMjQCihvNTsk2cq5toR>DN>=>7rdb|@Y<#3X=_Bb5i9SZsAw)zJ-76b9sgW{ zcu_o?orl;>66zSeVY;~Qzgf|5f4dLOI8)qq?h)QeMod6`A?iI z|0t$6bTQytD8Y2atOo~X6g-zDRt*`3%Sh~aXSQ4C*6g4YUg8&&1NziJa0@AcaM*)- zlmK*w?amV3Xqarvko{G!*gv$KmiC(861holQTd`~*vG7lqFJ zTP7G`*kfLe;r+h)o`GWuC}u?@W7%B^LCmvHYBPTkT){%I20lZhRTMe3y`vZ%CHwpC z%y#~f=+5;s2jvmIdy9Yo>3l(+_v2)ZpI)sa-hWT0)a~znV^a^954Jcr{HK%_Lmd2&$g7!YV))T9v7(^Mn@l;@>!nwe^DIfnp(44Sfef*9Bbj0{(E91wIw&7 zE?N9pfL5kxVW9OD3FCQmr)A`t9`=P+twx4#BrS_@7ZwRJ3b~^U^Ols1b;5rE(53fg>N-HFnqq`);1$p(SOj9SF?z`E{oi)n%CodML?yl999EKXQnir@8nhl zhdhBa!{7`0Q6&{ach~beLDb88;|gCRey$5-!c_eOYyN5eu5MddvWj`%s6u=%xvp)I zac7XA#Wrt+{8LS0R(#P->oS;^4&sTy7Im)7LHcCEoF8wF+ZId(JEa_RMwz;5rSng!z|}n3E0tysf|7 zBQvSZWk=i`2>U+OZ+rmfYy1U|ua1Y$oum^E*f~@1uq_doR3YJ>{Gc^E_)#BT{}aCWHLmO6Xqe^Q_Eqroy0Y{kxO^MsI;eDPp4TY z$M6kTua0V0rr+9~&y5#eQk0$&5iSq-o<{RDdzO`2!{6P->`^lAPLc(;9O37$tF#om zzhe0hFx{Sd46Bg38_No~Xt|g0@gr1F)pUceONEFz4W}<5`nwc2UnJ8HY!g^!Pr1m=*1HG^Au3dLmZ>mDbbnUvUy7v1n9h-O{qvk+d z3tkwA0|8GJ=44J4)H6i{5j+qD&t?7tull`fEf+CJ27KZ7?B3`3^{(psy}$Q;U7yQ# zKsff8#(#TOfNyeHt{FnSCFO`1Hsl7AYs6EF98lXYG^)VOFc4Rjt0#z0Usm5O7l)CZ zeGZSBDJvCIPMF{ddhecZdm6!0Rspl*hUF-Ew{QKM9+ol~fBKR3?!0r`uZ%~laTKut zAF8@|I`!6~=LU=;qSJRuenalEZdLl!l$w|?!T_Wh?dl>9r?XIZAj=p@O_kFMO&vU5w`X*c(hMl|e1<(@5JYEtGov9vYY?60@2Uh9|#~ zJ&2Urrn&k&*nC$TJ8wH`(EB>-;ph{h2Pn9U?h@tC@e8nKKSWoSd{`Gzw7sq+GV1fI z+A8s|ig=(F#*u$R=uZ|;4K7~c1UEeRRdKhCAt)PLmGi3l{JJ&e+*#Yyi=*LAIdssoH+<9^+`i8;05yUNM>?Gw?Y^!^Y^|{4$6u8;QBDbRzWI#D<7?0 zwTeva_k^mYu5s8I>6~nZ{04O!gmxn0+$a>U$RalJezu%Pc*p=jjGoo7h#~^%NPwF| zO#%lw23IFtK58`*bD4)4bR941)yMiJ7=-6k!+oI7^fT!4;F8=rA$+0`C4B*~rrz8V zPI^HZdFz64|3)B{7o2HxcpQY|GJp;ZhlV764);kjTDO=PEZ_X8OT%DPGRZ~hKm~?n z3~(^!yE%^H&FFnlZsA;Z6s;EAM(cc07;mT>?_@^ zEmr;rYm1bH%T2}PDT-FdXkHm3y0i}`b+7sYqv4`l?3cV<=G0>-k`sE-dW;0Fbz=yQ z3+i+mx^S7M4&x1`DCa~|j%4@-cc(;b18Oby`ABz*QJC;Zp@W_q+>%%U?>rrl(x=*M zvCYsG7Wv!-WAqxx1S$t)nj{wktu=3!6>Jx8&QXt_N zHeQB2zHq?Boi0?iJx^^f-Lprg>kv^~;2B^O1ko~OlWVK&yvuClxc;w_~S0u`d z#+^Wmh^%0jM$@>at0h8-z&yrz&qn9_8SRGc+esWizRzdS^}>$mCt?D|?(>n0eG)+! zOMq*UZA&QX4W zK;+r)=!akq%Up!PZovOa&c(Kn{XvjF(1F%3A0ei@-A><845p6$FvlUTCFo+lnlDr< zP>#nz8Z-Bv7VR#5VssA6)yrBxxEFvs_JirA^E<0$NQ@62Nz9Z$+S+VtWSS0)!x+W@ zl|du{9MjQpn=Ag^V;XX^R9Jmr0x4?khDJZcL>`By^@y=5`8;o`MVh!huHKmJ2Ym1 zS3Fi(Hoi~D1iAU(zlXrzlO`M%Xpm$Dw#JAifEGjuN1}<;X&1I&-aBlrt+Lzxh+*Ys z{S>OBX}|Y&fjFNBi5NbTVvq>Iv*QGod|*XxXnn?=#l0+H0TRuGE0b^uED!4v=9rm1 zL9#OD|I672BUuOyvQ(Vb8?e4xTqNf$*qz%2z)U@Jn0z24QY6XICSXNV_t%b$f{0BfQSr;v)x<&%nd zsS@Ee;eg374f?vpQu~lxA(9v_#N&=c?F?Q&dI(SZI+f*Ld?&xp4m5L5Fz^HIYCA6} z=j%ytAIHZGUdQPLtZd|Sycg52H?wt4haB9Yj}rh*jF`h!zKF29&keSTI8 zmfjzvs~L76X_*L^q|W(zrYhbS!N@lV{I4C;z{h7F**YFb_&kzM_w&~yw3x0vq?>!P&DqcW>sjB)I@16lj)+Zt^*oc>vxHt~!rdi9F zmLQond17EC3yL~XmF(~nRUO2m4bv)54JLo7$a}A{$9hZ^h7y>zk&4!F&fa6D+<9BPt>RRfV-!YhM$>D}0-aujhLOR0hG)`Ev z-AMAEg)8#om5iWR;bx)ObR+|q894Chj)5d{FUAYinN{Bk;D$$*xrec+>4nV5ln}nW-{Lc>xR2y4m14@PHO!6`It{xfzz+;#}`>i#9#RN zo$~+nmo1L$*K?Sl7wL{C_``R9`VYVPSO2d2SAXSK{sDSPfB&cP&whE2NBmj*?@T}Z zyVDPUfFFMMkAM1Y{O{ox|6%;WkN@E7@Bie#{IA5n`Az%$Ha_9cp1-?0{;z-ISDydo zlX(7i{Gb2$JNmEx^Z&{>fAZZQr$79yAL!Fx`0e;V{_%JJH2vY<`t|3J{_v;qPrm!F z(;xo$w~6on*WdUz@Hf7B{@T~S|K0tswC?lwzx#WC`UBwy|I?4Z{_X$xSp1N5Xf1?LYg0{j2}ykAL#V z+3);_`Zc^jdir67zZ$Rc!}NN1{EVmjfBg8nd-}uQ`DXX^B7TdS{`7Zu)wjQR^#AQ= zuktVPQj~x61L;?P<=g2s{Py%3{(a*Y-{qHI!*?^~n}6xI{-WpoPyVle^n?93zxpqK z`8ON&vu#p>worxf9tR4cuT(fPyVAH&|e?VpZ)1C-mAal$MfI+*$;l_ zn_vBVf0>Wx5C6iC=O6z!zj#RJpY7xM_RoDj|Kb#Q-ZsG>RYBb*_+7V?aKlr=py(yBs0sZLr5~MHGzo+6yX8EIg^CM~y z!@%`jL6j(;`u#urY5hyz^@96#|1bQ~NB#3h{x<%)dj6Up)p(TmANlv@j_39_f9c7^ z8GeKBRp6&T@xJ@(r*Hn?Ut^d*qJOW*9zU&r`BA=q?mzjb`EUK?zxeOyU;A(W_iw)Y zA5P!=_HUvGJ6mgunT?+TBghHkyFdEVzw#e^MmccjbU^3V#gebsw8IcCB@w0WF6=aY zt)AAp0bnb6*k||-*6}G{t=4lYxvWn)^1GArxqe=n=yHQ@*uQSr(|LEf&mfy!qa58| zk9k@^F1%Zu9AqM-9p0WX8;z{RAS@JP@?+CgWig&BMKI^`#1_G?V+i~p~= zw;mEoe4g$GaSn34`REmg9Rm&m{N`s5Ag#Oun!Uj<9*h8@jG(R|It8OcMuS2En`(R^ z)7Hb`x<@21lzJ^;iJb{O_#=VBNAH&N zM}r}9XliUY8U%N2iP>SlIuT?i5<9UYj=UsSMmStZ zQeY(?J!4?CnFwW*?p}xZ>z%0T^oBe-Y~mvOCTE)moJQMaN63Y2cAxz*CT3OFZmZdB ziH`?I0tW}jnJb5#{O6yoj`+My<+-E;_Lf4%_ll(No$221a!Kk#uEP@VNH-73x>wkw zW{Bir`4BzFS>-gH_htj5iyBs}8ghy12xy(po(+kJmPkVj2LTnzHPd+397sW|xQgw< z6A9ipDO}mOs@$-DpOHpYkOI9Ad%kEo8=C#dtet0$Ry4R^UmQ+xiq_R}^r#T|gzrDn z64BsY>L9ZW271|ZhO}ZsdBt*{A|-`gHS+jooFFjlpjjIk4+Fs7w+^;&E*s9!LR^R_T{T|;qk%6N(9mIREYCX|wlM$mNjhjYj zP5NP%O-kFzFXt>TLweoHSD1ZrW01xebEmypX8O#ZRGj^%sN>H zZVWSi**wOboJ{%~1aqHuyx~iAvbLB{doZyhe{WeAb2hJGGzR^B)(79&Zz{KBKvL6v9Kd=eVt0Rfff0I)T zkrcJ$BO?cuXO#+?|vgSK3hW!IBx0lv7e!lAm$5BxWXAuShU8xqb$;%Zg2lu5y8Dx$E#1->8H6+~PI1id|3YCwO!CdR}f8a4F)9_9%4E z==s?P`KP#Swl670#HX(^`=?A1CKbiw0vX z-I{j?V+}JG7O=2^_(4fKgmqy8ou2^!5tXiPYCR7{c_1FQg}AKKk%Fhi%N#Zmd@W-h zuI67TDKKY;#?KLK@2@+VT-Jdi(k&|xOWl_g<)T;nb(vHfD3&WDOwSIZk9 zq`e6^klBd?T0Sr>m{=Of!`+BL7H53S1V{+`*N4aY^W5y!%e_?!Jn!Ll(-070%0w?8 z0sVZmPi~SQVt`*GUk4!p1hPY*yyOEjfcG}Pq+qYy-YeUF1{D$j%f9h$yHSC88u`NS z=e42dY7wkm3wz7mo%RDPj^bjH)7ae|Awr*P57Vy+OM^B6ylM7)eAaW!^_+*cT8s?r z&gYA?$A#A9blK!^A#Sl5O-?3cx@?z=tV)?#4}ldYA#Z zU!*NNixbxydb9`?sDB9?N?tbo8CT5V_>s>J}Rx~=7KcY|abiFP3=&5--K>k49H zpXc>tZNl~K>>Kif)d`|MP$M>X!96@@H3jF1cjKWzF-A1S1MYl8Di)gMmZWudu{eGu{_If(H{u@HWx?Y2pJtXTgsX=gkG zgJ+j=!tVJ;Z7#P1^cdI!p3)_P6`GeiWOAPnw2@9m;baYz%ZwK;P{-ZV55*;RzjKlu zC`37kOUQd{w>xpFiv(3Q_wXpH$&xiIbVpC_3hg-hR zb@TG-CdHaQa>qP)6Y(ou$we-Q$T3DKD4U2d7xY>$YcFO#FOWrnZ(d!mY>-mZR@sFR z<>>y`mWy>^;V*NvZ|xqyAfS*$__F}q#y*c6YA@iS$rvA;vBDu}-p;3WAHD5qoIZqR zUpG%|5Az14Sdh~-&DAqFpUUgLb7B6e^PrEXPt{Y`*!!Z}C4P(hX#}G} z8nr+Xqd<5&-|S-Dc|JbI!X7K=W=}^y2UPsHcc!m{yE5awvtFQ~TZubot3F@*c<)xh zc)kHtN}O&THs5gt`U^pi=Mi+^#WihaJ%_y#op`g zcyF%s7&K5LC?DH+6g!5`q@{h2Z}nb7K6#%8bGUZ0+Md6PCLL{FYcxY&6@ZKqyZGRm zO(MOPfWaFrGmg&895Fre8RMQru}OvPgJsTA(GV||z=T=#aA?3-j5T}2N2Ex&pV`mV z3|(F;{{)D|cupr62SlnM90lD&VA4Tm*3SBQR}Dx&S6I83B_xl9p9)xa(~A;bOm6qc zXM=USdRf6oT8?2fO1BueUgPq(Qa;gKo|v!4?9$%PRD}Jx1Fpl{1ev=ZPkRqj9`Shp zSQZNYWyGX`e-2Ow;`B3vw_1Ulg6JV@EN*vl`fLf(E21#Qp@UL@{%Zr-A!veRn3;s!gL7fMYLvkY)~*>D+d?g8^fKpg1pw zU1ICdMn4|)(Fo3o6}ZdG5F*Sei75y=;b5Hf`ZHWE`7>N<+MGI7pPPd7n+_`7;Keu= z9RZMhD(k!)e1qdu7yypXbV2Fq<=wwGJ4n^x-bAoedUM_z>vYz>F|pV-_TkAmP`la7KFFIksq0h_3rFhmIcE4rPtB2r zI2~vou->i^6Qm!}Oq$D~dpWaTx7}^?w*4oyxYHk+sP{z#KoKkhdu-J##QrH4#t6wM zC?W$%re7sI03czPd0B>8lBP_1c?0;_DtY)8&q+~p9W)+D+y-j~oC^i4q1i}_<&!4i z3zd%$J_BYe3g%S^Obcp&!Fz3D8q=$f1q=vUGc9>;vjh|XK;A3%oLH6MT@A<3p~w>v&-A{xjehf(iq=K9`X(NSO+9ciPx z0iFR_BP1@U8SQG7YsKmTq%qQoQ=~}i1v(o6niCOJz2)>Xm8Q{-Wy1kyclvN#!-y{C z0|Z1n8#X%ODSa?2uJd?*smuzi)?RRq86BN8=7k)eXsD6sRK4r>(UD(eaeQ~zbG>f& z`Qjp)xpOc2*(yqu?Wnwieq9jL8AQ(5%MOv-k9$g+;ytEatZ3Fb@JY73ZGqY2!e>?Z zfNfcr?|))5?p6}>Uh}bDK#WuhwIX4!9;1yxy}hn<`CzSMuD4;#w0msL1W zNsMVb8&jsV>F^PpxGTo(5Qg%G#mNIe1}8!ilU+9RR7tyn8GdG0Z% z?*rubG_p*b7(=Gomxjbsc{vrR;h1!QDtkAJBI*iFiN@_!@dD~3(E6|tAf=aB;91WT z^1Ztd`)f)mMVdZA9gE%$WI(qIUHA-%`Qw9wJP6y|QOLADpKI@++uf5KhAM)5O6F|K zTzdpWBc~c|I4GU90>nw2$7r<(m#o`iw@J>E_Gh)VpGdvGKR~>&KF{rKtL6LU^N|+} zfNHyJOZ(?Uz{q^qycf_Ya&>{F2bI`LU94jPvPOte;6~Ms&8b6arj>8nde-NgETtS(IuMct>EMGUbaWD_E4heqFulB~4!_yXw_^`L1?c z2%i*%SRs%iuKu_#$+=2;nrfz**?>KHiSn9+1A2f zbvG1h;&fJ^!p4$<2_DhkF1|WYi|khrt0|6{rsQbOkc~^3+O!wOer`lnE8YpR8D}5#u7+dj?tkr`LjLhb_ zTsHR$p+7fUcl)Uk#iGp744e((1K+c0N(yRXdb{cEd4)iSHo0gFa#&+^xhv+E`=y$a ziMk6y$!i7#0gOwSor4g$oxYVp4gs96JyqUgvss;Yx96LwY;qYS@!?9bd-fQmYo924f4TIj*o&B3u%aC-Z*-$s}1@&g`lnwRhmGr1*T#szSCeRi*y1hP&LJZvycK1+t0N~$M}&1FLur3FZJ1SmUziYr>nP{s zSA`ywW$k%aJ+Y!YYWX`36##&f!q6pP=-*0g~Y?AiInw6cczHz8Urog zLZcjO&Y66?9W67KuGXmtWh`QKGo){n{W+M*CUD?adp*G-UGbdw} zuoC1o4YYYrOfDpa|tndmRjn2&o0Hs z;tE}=O7UpydvOeVM!a%9FHv3ZI)VVAs=M7=iLPX2&r~;WzBq~L{w1S>=B~)=eGs=u z+C}P4MkriSg;O`9ObZD|vT@UG>$*$rIulc#Bn%we&%mRNjr(lS@@deD>6g0(WRO93 zc|+eUrA^+_nP)irM%tf!t1g0b~ zrz`Pj_8;Z}^v-b@DQpJ59}E%}hghWY@N3HH=4Y~+#arR4+3xFhOz+9%$?ogbD;d$hf&$U8YIJC}_6J)m=5=n) zzICp4u*EFTM{XVw{%AcD3&$KBmEmV~#$yRL3Znb~dtR6GWoBPq48xB079V^UBrkHa zE3PhrNwy0po_DiZOwu+dV_dmP)mg-6?bX?zi~Zhw@I4O3JA}dL@u&31x_VEo(N)Rw z-TWA1o`87k>(#z_g6LV{$X5awrvvNu zWQ)(!{+wP4dK);E#xpIl)pIw1X%rDxb!cEW_SV#mBSGl<@bF-Q$w&dBhv=@QXiWW; z+gm&Kh6H=|c^r5IsH($3C8arP3v3A{qUHtF=XJ@u)*E>s_h`S5qp-mf{q=AM%0<>sM)$ys4i{tJegp=f9d~y3lRlZe7uN$L?q2=vw73v4 z;xCQ$HLwz6bGx(cVsoqDM#7PaT;fQ$KBqx8^>n5N6}V9e?3!$Pc-Eu3w~eNP-hap1 zMV?M4p%o11={8-*H|yS`V@@MYv^^IeiM}R}v~Y5S8{kP41HIcalhY)S%D2F;0jq5z z&GL3?v)?bObKbM}i@NTbmUkRIlZ@8N9*~@Nv-ncYquE&wM{<8KvQ9qKSJHXC>6~$~ z$5)p%D9r*>)YnnelF^Z)tK%XiwyC96n5Io=^Cgk&wm``D9VNN-4PtsAFC&fealnN? z-t~p>;J7}&n8r1k6~2SZa4u=^yPlUu2K@>N4l67k*vkj>Y1fAh_bvw66<(_h z8JH+k;LAB5OYI4+jv@Juo5|W+T=nT=41-`nl=QJgGaDpgB*bO@ndG4Xc_@s_VNTN= zNo2;bcR(U8AF0DS!g9MMnwZ=Te3tpfByvRPv8hC&&3xW>4JNg343AL-^YAbv>m|5A zy|;QIh)Iqimz<#t)0`A-`2=7u6G5NPgeRMVRTJQTYCu)@_7TD87;aWm$-Z-Qkq1zW zE}hTmXjM(kycaY9tk}vs>E5%zo0`1K{vMtk{lO{2Y>zIUGpxsxa>lJ=q5*M4O7~p# z5aea-W`jUwc!1+P*8*hf1VqOM!&8i$9lgVvK*`y{5p1XK3B;Jl%69AR`u<`?-s5A= zYNj?kbHvdxd#dwNnvc!B=WVAc^%sraWgM5m(=IsIa|C|tN8WhA7VZk>E6NkZS85s# zO)GiW&MjdURP%ycJ(0(Ush?}qQGbr!x72kAMI^h0VbdTl9q!H^U|-{DtDl{7C<&&I zPX&f;EhOSygns=Z!MXtdgMN1^^z1W`@3=W`HLo_s#@DBrAv+(u3qFbn8%;H?K(jxf zzlk6%)QAI9Y0Pk0l57s9Q7HeClvD;k7b#QxZroP9kZ3XK@Vsy#6_ zC+@~-alpoKL$loEYJN?pfK(dGEwgS;FFq?4Ifg*YXE(baVJ}19(!1fRS1?f6dC9oP zCl1d_X&6WqMsYf>{0obc`;BV&1!)Dl@Gbf(@O6)SGJQjag**XGIKS^#a9{5jq@Q7I z6uXNs^@jw-y*V`Qce`6d@UGXudV)=~T(zH~INdMi+=(#_0?JIupF8(p_-VOuuT7*{ z=_Nwo5X$N-v}3J72F{oPI{6kh{W}rG`{S+b7BVs>ReuDTaX`Svt262~QJWq;AOMAk zX=}fxu~`9x=^ZD%^7THG=?As(llaq!clEIjx;KV-plY z=p9@yok3N*^D*JbwVm!u>RMxUea#w>9o|6zTIsVlu5Ncbokz=_CNWva)+$ik=nl;@ zTHB@gmf$NXBcU&RNg5@?{}|*KYgKt}9-|IGt*N|D8NHriUppWOWIbD^aGQw|P{XDr zTA~tUU`XYWSMMLdEvuwrQ1w~O4ZO8wI$bvpu-U3JW&x?YVdej(;9CH{GY$w9W_`$vj?~32P7})SFb z1g!r&Oqe->+8l2_doSMGfqC6-N@*(CnAqGM@(;^2oey zL{DL<9&}^$Y!GhSUXQq0aYJ&(Y}OUzCS3ALEhk%L2_6`_pO^=LDIcn~#fLe14SfoK z&X66$E{RRBGq3!OnwKJ~_okk4;Yd+u&U`^6E9kJLZj;m!QAupUZ z(VhYxJ(Xjz%$>$y`i)21nk3BcO~kyfsGY0R(*Yp{4M=2gB8~uoVrL2`qI#A1WukhQ z`dF7~tX|YzwaoVdF(jmwIVy>CHe-l%9`+2ZKV1=J^BqUENNT)RJi@D}<{?q6cr341 z9J%&xKDS2yB@MPCtx+egl`&U>g7+x|PcYd-e@ez>t3kM^q_>%xrNO;lL3*H5&yfxa z&5xy4zubBf;Kt z5!ZSO($UF?!?z2kQ9>mWD`;|wCF zSQb?+a=El~Rd7UnY?JlAp^FDH`qHZdxb((Y*0E9!v||KFSj9txTL*{P#D*X$~%nQ8EqN~b9ofH#dQev(|+!#Jt%R$#cr?z z3>40cpQqUbRA-<}Ponx{)&4%AoaKUv^eEkJWSJU~kG5>1cZ%*-HW-Hr9Su@niR2m? zT=BAhYp8I}TwFJ+BDS18%JylpFtuV8z+jQ?kZ2OC?j4Q@?{7p(Yg`KoQkTv64M%@A zf$7$ZBPRz*D?xg}z{=D~cVKRxXP3kd&QA(vbjMOGWhc}B}3rQ4g(EazMiJ}Kuc>-(d0q#yaeognHrA|dl34;l~l#Na*@WL z?l--6Y@}g7akP$ze0XWtvM3?#)%RJY$eZ(Nf;Y*_34Os{oeT&lxz!=G_xGxt$5rCGhdUh7Vh)azE7LO_LlYJ& z#Z1yw8K1gjmpTlIc|*>x1VNna*(SblQ;Wq^M$UMY zYjb{AXLBfuSm;R?bhCA5^@0ssc5aMy|3R8vfh&R|l`VN=Z^76ph%5!Fz`zo<9&SlJ z&{7HXCHjGppB?F=CFq2>n1p71<}?}DUMtbykBKvG9oyM0I4~DZ&+Y_NuIVMo-Gt_F z?&$jUS;h=QsZXH*yCm9Ck7~<*kkmogqwuKGPF`)XH6PFS5cB=zF-eR2>3+vK%j@Pl zFsz1`&q&B^!HM;>TfVksaT*r|oWeCQW{Psht0hjSZAPY2*&1yMnrF4+H5~3mob8R4 zKRb)cZ!JyZ&sm{~U8HP2$U!+4t~H!3k4zc&I@=y;Jgl`o@f52{5@wX>tSp|V3OFp7 zT(iy4FQhnG=Bw zu%nmJ{mP5A2K~)_Bqr%)$+gRP33XCY?cSF$aC#pU0jL<;Nhm%Q9{_-&ZMqt7>$i88 zyg4f!LIp?-WWGZ0XWpVzc;u6qk{m=g7_5f@WcG|lw@ z^nW3qsW<#k(d@<=*!AEki!B)BgEwZ}asQRO%CiBo89M}+3 zAYKALk^p>K!l?jgBQtMiw2UyCR5!x=gYbC?tHygT@DCJ>8-&LuFoP26^Kz@p2UMDf zU6pwqp#n=q!_H{v_6pj5M2;cVysIsdP|*|OPqwTp`?1Svy?xpeIy1h_`n*TKMq^Y_m3KD;dN|FFaS_>Gj*!qVG10zHU9!f1 zbjt9(=^(96SI3&UtSB32U}BJ)4ZHOS!Ha_AhcU0x<%hKzeOWZ70@n(kX;72%VL-$u z8gtx=dmZsxVvg&+JQFCF6>E@)`FxUOXOCh^GZGESuy)%w<$A|n!7LbbLdb0H2snd; ze|EuqF4fl-@dVe`054osLmdMvczlRs+E$;4kL)%0Jlh*Itegp~R6Gd{TD3$X$@@CZ zcR>pDHG8c-x}8m{`xn^_K* z7rlwRZ!|=F@ctHu;N^~-^tr4em0_T|BCcc^oU)hHUzrsG0~RzT98>7hXE^)O-2#?< z2SoDm2;IjF_?2z{Tr_Z5<%vKZkZ2uGIWp)mTpQ|W`Yj_~7el(-xYPqnb%D!8_Z@s# zVbu1*Vt$VV!i>1&(~IixRubcYIsn~_QC}z_0yFNW&1ZDx3?Tp;^bBJYIL)0L1`UoN zC{_>flrI@qvLp?Vp;#}2?a$`G8A-guY0m4&XlmGcLY+nC?dXbjEyR$NXW9CMM*7kR zjo`I|{&$yR_}~ShVNP?}YJI04dx%5Xj2FCNt9afV{l2*z`{s2aLK<_a!5~QDQ`h43 zTO)w=4q~u%(YDArq6LJFYnI`tXvtwh@$$-j2ylz25eEpPUys|-rx<_PS~d-bH+6-rhq&|B^;I4nvo(y1 zJRLMj_=~Q{qXU%%l1pdZi7s_IKZvLJ3Y@FlDNs%)_^UBoKL(5hk1{JHZ>W(PQbSDIZVkD! zXa+MYfOzWL8m=fnAhp@0CT>Vm7N0zv2h58ax!H`+2HDXA*rfoK$+7_Rpu5J?`FSog zjw-mN35DF<*{znbUmm5dx|D3jiF*H<^@lAY32f=S3ajFAR8JIxoQ~m+Sl1eygg7_b@>w$bxKL-Ep94)uxU6ae znekXCgt@P-t<{LpG8H|O#)K2!6JCN3W;f({hQpc|rO%VF)Y8Vv^x5;ZRjL{=i{KPB z&c()vd44=VOx$OjdFkC(*=b$sfL50xx*9jo8T1Dn47NP8w8O1cpL7NAyzw4hk^pDK zrMZSUKZHA@5Ew2Gvh0KvatJ>nF$3)PHc286)*OU)1B1?NCq@#jo^$NqPkg$E$yNm= z*f{RnxxNq)13E>Z0@8@Jp>plzTU0U4(iLM#$rQkL9ZSqaG7UsoFlf1dH2GYP2yj{o zl0<9F<7G7K!8NcyQ*)P_#;5{0v?c;oyYo6A{BAlTC5UKF3Vmyx)GR+v3P}tmcWPO@lLsGyt@NjlF@wBt<-Uy|$egjo?AUN~2UT84j`H(ZX=%hu}2A-1Y zb|&q2{t6CJAA-pQ{f+gU4nJ_pB5 z|A1A2J+0D9`f@dH4ARlsS#xWSh@52W)+nI1d`{*fh>d~(R=}>7PfIqQX*x1RU>+Hs zNUSK>Xahk|hYTn6GVCrcz(nu%{FdH}e7n27(G5KWIZ>7G?u>34_z~}mRBa~zVd%3K z{Q^KMU5@?f1l6?$t}CV+$1}gnCRc-v&u}s%Y}!|J3Y7MSn4T0wZ1ft$`1TU8E9wF1 z1(REnKvrXj=CeHxGVzQ|=$&*J@JQmcIxniG3b}?2n0`*6+}w@gaQ~n{9Y&mACXtAM z$n^IJ*bFT~SZZyy^j0xdY=Z_2SJ5mxN`Nzw7W$xz~yy( z3|Xr_IJ70o;3Dj=Clkxy1M(66=+7Zbx{QcUmq|#3wz|ccx1Zk7j)AAt;Hvaz6uO+8h>G9uqO* zj4i3Z3}0rlMl2TDGxQ;g9uE-j7t#)++G*>?lmjLI#(H131X5Jan8AjFRMd2k8{ey9 zi)M|s0_q3aJ`<(?e0YkFx8%e5XmUX`W}}~e;yXDth=|dksrIhoo4n;yVgCeLT{K6IZOe@#*8-ZJv)GQQTB`Rl+;8I?L_ zM>L^3U}n2&Gx#ij$!UUz4S%;%RV0DbHfCEe;U7IDiMB$4%FEZ886pTHW=L(}y~JsD zbMTZSv%^g5AUJm*#ih&2u`hs@gY}*IQ=7?E&4V`boJ?8v1x@_T_ZkqK6F%_iM6}R7 zkb}(D_tBxGK9k7d0n7XYwLqZ8jL)X*Fx-0=z76CE?EoQaA(x_ZvJ#of)rpb{w>zA# z!PQ^fn>qA=^=1Z_edj1>Iuv(*huhsgy&~=NqbD1kmlV+STH$&Jj)b;5MiC>6kIu z?lGegT!fSk@S^nlmZNk5Hr)zPMUi-;Z*BU7D!1%=DCEs2)H8E8i=rh*im2&vVr1 zf<_Or?x59hf5d9v8kGEFq8utMQHVmkMc%w_U~@<4x}6R6SRYJHhd70oEG0R2TQDHG@(j zMHt}S+C8ibdiPsyHBd}}C^-f$R;RiLxHzX-|7C@MDX+u=6^@WEARkzV0+3YpfH(L) z5{}7fRMDSjo0*{4?dO_8-I}^lPqDG~sz89Nc!=h*&{!+sn`m{|_zqh9$b=0U1fwO`!Equyb{!nS= zCNjbg?g30|?=+Vn^gF5JDt(Bjg*%>B+QVPxd!?juBFJ5o z_A%plLU*>$4&49%h7IB&uYekc;v^X7`m3k5!MUqAk^_yaBWESDnbJ!Y3jS9i@|h|T zXCy@iM#AoRpokr;5|g&z?p|KgmCBt?D@?uM1u49{3gmUALr16;RV1{p2U286AOd|w zJT0eA=Mh{e+9E@%-7>z8ljRnHhA+|F3;b*=CO9HvF6&ne38^IcZ#R+k)=iJt45Iz+p!+F#nq z<*>TE*x0aWFxuuu0hk4tpeILyGCgl_8SOj@-HhAk>o&nq;2F|Jo|%`WHyoDbLkvXR z>+iN#*3h3i7H2tI|Jv}~>UDqzbhyrTtQq%!Ms{YXtCc8Cw5NtR(s)fvQi5E$}Ov%CfpRU~U=LJVsW+aC^KA_At-AIuCH&bV_rK*-i z(GRjAwb--*rIfnpfubUHjdc9=T*iHkY`kpb$h>dF4TW>|UTe1nC|%oiVvp)JL3RqQ&M9<4faetR_ozRY5Y1o(>FuD9~GCtMfNt| z97HLIr0=T)O?cvq6sw2jlW#__(J;6HLsI-XL8@VK-lsddZfnx<#tT#X-p^+xctX%v z3DQwxpHBYCDa}%e5BgmL+Pr`_Q)iRBU*^WC~4{iFSG{oQUIjzq42E!|w2Sfei26 zjoB+5nw2rE+3a}b*3swnUc1P$5DBJA5&{bQ@>MLc*3PtuyP#t$Izt5@(-b|v&4E1iJ)7-s`+Wif** z#yc#?hkw)=6UBQ;Zqo#Li0xBE?6VGv_4w4h{%W08Izk5X-E)rv4r~G>QIfy^{&+pV zY?qy}MDX!#0Lmwbj`;Hsqr`@WN^x2GBhsm;%izDHVwZ^L{qHbx=2Ln-k1K^)?WP~; zLjPp}KMT?a!i+ z0w9ai5CZ8c`+>kJ`aE&R6V!q*HhAc{^W9INCwNQLV>@`)$VL*LuJ5}Lg1AY%#==fR zzDGo&`8wchW?ibAp!hGGo##S&NxO!BamN0{hWRDz^342}V?|I{5zDH;bb`7qt8|T= z)$tX`)CJW$9icx;S>l&nyqmq;8xCQ9PAhGDL*~FG_4U(PTluSS!_0?nAfS-LlWLO! zDpNF;s7qoX1)M>^n$LPn4w!fpTq@X_l;={JZJLZgf zz~1`ZWCgrF{Oi-c%G8V}#zzUJzhH8bpkDabe7W~%%b5K;h;?>?oDyTQUv+#%LOej$CAlw+=yG8@&FUA}XV zpU1E2Gz-0R*arLe>Q#H^4 zm`c64PPh~OjsvimN<5NZS8(JH)4RL@%zqCxm=CqOd*e8ESp=^6Y!ErGw&F#;J;yJw z4v&cA>nn23PK2!IYtuAbr~Fl`M#<{uxESvlrsH@Q-0A=J-@3|3a$j^m}^#D*k!+)J!}e`1LpAk0-V9X_Q*-moC=sq+_Ly zr)(@Nll2n$vV(OE&GXsN25*?Ac|ms;;U=#wb#-mz$Nm)g%A=l61Aw3ia7Ud>b`giC-Wa)}BBN zhL1AylZamJ?#jK6G;!(g&G)`F9_p}se(7Wsmy^xhXy=(z#?^=gmQe6Ra!~gj&TCuX zDadO(tXu?DfQM)YD+BA%lx{s6N)j!rev;jeA4kiQOU#-`E^c7+dJiV1!?*A~NZ(d2vdv&dVBWa--NDy+9o_OI?5^5$yPoQomTSG;56|_&<>h<|{U)nB|Go7% z9@`NwCByI8U-e#9Hbopz(dUd6%s4ka=F1;4;%NpvMK10{xbSMdceghl;`b4gI1lR8u^^`jW>^88)zMEFUD*%so z1O!%8vj%q8AK9(MK1S3q_N-`70|)J_vi*t#ME^VwXiTgA%NH+yZb9BDFAuG-nGLL4nx^_D!s z3mJ%6!9g=|d0RGDz#+rFD)NxU@*4}L=K2PJa+ir#G z_OrG7@~vElM^4ZslahBUx-9Flbkq|aCGvax`ew2c@0DmsbW($Nesn$MW7fd32@<7OS;Kg8Yf z*=&#zq1MDG{Lm^0$ltriv~u%vkeu(-A*#nCJ;t1;RnqO`Z_4HU3$6cRwGx)lSlsDG zKyDN6O=Az(Aw>JdWACrC=^5ZY%y^g;38mEnqfBL8tk4=pzi&f1`xYIP=~ZK^(NO#@ zG|D}4gjY1#(249Ehq+3J^}Tw%I{o!gEcmxK){KWmFeD~EtRak!?#88f+qskKN_AFp z#mU_omOgdYMT4@30PTR2i6k4sS#IRA81@~ib!%5Xg?mzE_vAp877=@y2QulM(fs&xEj zxFovuGwH)tlFaMBy(HC6^%fQ17kKXU{kuMAA5)(rEv4QD zusp*-)3%!(aX*gCzKi`<~QmslY&e*ez zhn-bBPyCj1S8Rcm`2F*%h+gX6Z7RsYU@WhG^?IofOoUVjD+Pz`z&l{&dMSbj>hrkU zpM-mu=kjh|xb*}1*~4jh%OB=PNxdZ$hgsJ*L#u8do$QfwF^)f=YN@jD=Axc**4;G5{qwLCjr{eHgCpyAbZ}v+Rwu`(;UA2UI^H57R>2H2gW(>pIpf2Oy!bKPo-a&^dni z%$TYl0Sk5m{T6r=$h&0K`*lSZ znDgzq8n_5%bN!gH)AwCce zT|idluIBe}x??c(UKaZ?SH^aGz$=u$+y%uW!}xfGdgG+v-nSrUP+mK|;xG6Bwbglo zzZhx!={8DE!JD~fBY*#7an7@R63E7s>Hm1I4ASGie7#{dz;g;XJfA*2s7IGHya6kR zFNVmo3Z@et_Mdq1fIvV#^KveX-J{sMlXV9s0dg4Qg-$LUFSzly)ov#SGVWM4~SMm!|Ry)~a2wqkV%Br0BBK6k&a$Z(q zx}ojWKH;2bXqX0{qQ~>0IErP!*Kreu`w*SQ>kaBiTDjcRI6Y#y@9eSZP&Q8nVS%(NN7zRsS0y_klKp)IHs=5DKYo)L6~r|c zHa&myud1%k{T)`j$MX*wNejPa6y#68H{TXoEem=gEB$S?fH??LW9JZyHTMMS_XsTE ze)|B55vLm?aB=DKUZL6C?VveQ=`@b$+C6V4+51u-v**u$R8l;Xv8iS=1+zSuejl~0 z;#wcPx6vs^?72lK<+`gRE8P#)yYR35O8lXE-$lGB$eP z*XRjhTVDso6iW?zTK9$o+HMtpu5k{hR$+MHgDBp{S>@>cJ>AeIms{NZaLC$fPc@*M z@CXc^<;#bi_CvB<-6|1U?eaK%PhR5qOQzx>{#LyT4XeZL(zFc|1Sd&44YbCUmo0E* zNHLUV4^*0o?RX_2u7l~$-Vrf~d>eJU#nt^9UQqS$xvf*6rv1V1e$hRdL1g7cp2r)y z-@n6%6wH5PzU{+u`R~kW$164{^i2KgF>tVv`*+kikDpO`OKsSLKcKn7&u+pWIoji) zcZ%ITk6^?`-_I$M?{C0J-#Epm(t6j^>q{^1Bj^s#zXm#l&FBT?rxyoKjK}X2EVMRI zxL$sA9A;R3F$AW6P^pJW-0467OjA#lbh&D@y!QYicwKJvVQbHLX|H~ZC>zJ{`Unej ziSZh>x9lf(*=#XPx{yCwfrqn;Fwb1*HJBCyto3j=vaBK*tKgh$%4?rk*HK2-+0|O* zO7K;}SYU4M_HwF&$3D0mXMD;X_V)Rpa76HHO4Vx4G>bx9BwW?I&oQK}_;p}D3`MG# zf`tC=l$5I|_sX?iMa@JSr`zfM$(hq6c)Nl>DGzc(Kyi!akht^E;>jC6XETehMcqbKYzlQ`5rQ>0{Hl>#sXru-Mwo zC8CKwC-z<`I(LxTr*eis33L+k~pFZVHj;P6e?&uU(O5lY?wE6o4~?cQm| zd#pCa18N*#&&xkf=jpiN83UH`^#_sctM7yJ{%fuZ?S?G?YX<=Qx|T$8*>u9@>dzks z0Emx-t{2^R>gTtPWHUkS zg~keg=Yf0PN#6HpWMS1x0`gv;>3SVvbQjE)#a5#{@1Ct5HZWdc zpBy^xJ7Q^Zp#u?WuS|7gn>TIpm9JWPcJVTbQ%0>2QxIo5l=BpAuZ=Mk!Q@r|Gq$r7 zWw{m^!E$8H@!>XIm{25e(A_89^0K1hlf6Op0WH-kEAsY8I4R|-c(cFMmRgOyrQ`1! zc_?~oplf*kEjXhyu)oST1D;j4*|k8m<=;el;`P7uQ;d>Fu>K%UTRiBh)wSx?pVvYE zZiY5ozcY9Md{9qnb{v3Da?d8%V84_L;A&??AlB2i3>zm71p|h;XSFtg@0h)#F|C~3 zNkl9H)Lq(r(HKU&8eDb5xq#Vy2r*;wQL>ElAMN6yFi$}jbI{y%W6?gO;gr@--Vkt(u0AAI(&$cv*ktA5TYV|}KI9Q5@DrU9dUjpK9G(8LKkROt z5Pku_yh4EM+DZN}=p%u$Afi|&yNwWS5UnCw1gaAO5pkV;eh{LgqU`I-XCqz!9tk&% zf>xX{-cDLEgMxze;_&S2cYRrscfT$~(+7`bYsKeecQnf9x<{Ka+dozXkqJX&pT8^5 z56!OlD93n`qHBFd>&e0+ec+{axwo^sM4Z@W)2iNT_XGn>uQ&~fv8QqlHl=6KkdC=2 zYro4Od4nZbAsDPt=R81M(`*wgB^v7G9A03AXVPm}I6QRYkCF}_BK(}2{nx{_4{0`; zLLBOr(T^jdR0$lEG(pM9b|`Bs*|8~nz{t2xcI(dzgCN&~2mI}?uXi*)c8|Zl`VZ54 zvcU*^;mz&E6r{TH&qg@&%@YTboQKptZd?v^&dga`l zbF+dC^lh%DD4mez(`2T3907oDlYT3#q2$}&6o}A(MtHq2aD1(Dq8Zae59F=nIgft0 z6lNkPcd027H3O??&iCP~ZBo$`^7)7owKrh`oIY9)9IpaM-%3Rv^ae9jt-kcHySXU#e_z zYr<_=-1E-3=`3<}p9%OO<_l&V(bBeb3CMup6{PRzIu!>4)iQ&ka)iFIpZf3DR@`|e zu}1(gSS+Pik2G0F>Ip%SfsuqK|K0s9j_<{SnLc8zPVgGvRJ`6&x!Qq$&+a$dO|ZkB z>D`Of%j5ietKC-enR4-oo9U{w%xnc(rtqe`_r)D%h=Hv>X||)GXT+QlV*hV0Kt?tD zfLVriJS?9zkTStjRliCT=Ef$S+ADA_e4&~`oDnQ+7uz6cEiSgr&8pkh!Eda3mAXem zY(z(x)B{tK=a*%4wae%wK_UqCJEJS10Vtc&HE50+>?41<%jL46Jgyr&$~+(3a6>}h zirCCYH=f(gmny21r-`8z5_g6A9H>(5edR4|uP5_UmmfMPVWs-fa5oVr`@p_yD*&g; zCb<=zt!ok_B!Vo+BimmGL6y5A|6oI$Z1xe*w!gHw66JI5jc+kiR3b%wd+~x8L$Qyi zh0gbrKZ^E##pG>Pr4&aK{uox&OSGSCjVjtTXJ6|#Mh;t zU}nA$7*LO+@ryk`{EkVDtpL@T z=cB@RHubFm{(=xkGJw?}`ReAux;o9cQgFU>?j9Ar#n*z|(n7bh+f@aSesQ@E{uUGm@OO)bCd2wBa+}6#7G_O* zQMV%Zn!#u$OY7NC`o`zXmpXQNF9}%|G%#l7-VyYXvaX-!@5nwFtfVHfFTR%>QB)#M zqeJkNxcq~*RV+Ly*YV0(K2k}4oT3-DS?a5FqSQHkclUF?JDLOv>&!4#`8^{KKW`cIoS2dYL6$5j?)-4^m%TsmOX`Rjng0FN_Y ziHP*|cSpFwXaY7Q>;IOP>aA@|K@)PKa>=$PQ=mis-H>CG>&jf08OHNL!Ya-&$fg=h z7J%E$e7d_KB56^*i~E)Z{t)J>`CtQ~PcRUX8#^Kbvw)$iOZR(9v%&M8RLZSS#bDLh z#_Ig!oGm38BzBTaeX^xHUEN>?e-TYQefFKiRk?^p|!65w*NsN2JeA2jPlCCpLsMh7Nm#(*6 zkw$g%!bx8U0TdFL%`gt zVt)1rs(*KcGq)x5E{8~C8e{H%`Oj9{=5|)$1hLNth(~!r!{q$V?@6A1aK{%QRly&m3imJSN6{CNDS4OZa8^H$N%d7% zJ{y(C4@m(FO@~pCa^rx#=a#|$a2($IR!eNt%CkrOe4~F!=p7s&XQ^zrp^@0}&P`6h zYYO;|1e_};I#~R*yK5klJS*4nuPtnpv& z+iTyDs}!hXE@Gz>J8*!sIM##hxu|loL#PW3_%=k$-t65oY%NOHQ;22LlAC94G*Et z$pfqhk1(DzE)!QZxUa;_whb8(!E}mS^7l8AiPK4Da|Ct$Cn0>amvl#Y36f7*2pUF0 zC+b~!Z_+Yq>vhIL{dgBMW3OrO-^|ss=8+4)9?Q+&cd5valDPKc8@?VV_tQ?2F=}@a zi{vut(SkSS`qR#y9{}iub>yQ5fzqh9218FNF3%~9A@=~52k9u#85)D*Re9YTmFd7p zW=EMEAAQgVQCLTq%N3kaF9HxeO+-G} zW0}~rJ!BmDNZhkmI9%!F_fx+eBt$Ce`z4xPWflmOcAtg~Hcfa;7`*?j6e<2m!)d$7 zC};ObedGz`_ZJ6sN3I6#fye5PTK2*&0n`8xm9KFejHPSpM+}q1ABu?}RTN}ub&}S6 zMsGhl$s&mH56|!yN=I>*`h(K`vf~detaLI)7Y%M*JoA4+7v*?hoD_LMRFHViYS2*!y~ll_8e@DA}0L~+r9aDLgI$QJL1SjdK)NWcO! zhX}KMJZ?Zo>#c66;M<6_hU*up*x}sNzoJ`xvI6b31GDuzFTV{~g@zIgZ2VkM+sxpo z%U=P7C^9skTPI}Gn2IvyY0~Ez*sqJh=v(YP$M+ReQl-HiE)P68 zrB9uGTaMh;H?KJj<3fHY%+MLYY6Co@c(S@35h60uXdyl@Xi)C{)^O>wM%d*}7hCx* zo1aa+x7k>KWUW}9r!GT2rXww8B@f~0Tl17>4YmAsZ*V_e7Z;%kwq^^{uKjr!b$v(0QUNyifs<2YIR-2Fv|4;nXU%W}N9 zbQ&)@-5P3j?UiV(Pf!RLH+TQFfmbZGXjHR7F(z?U;Ow7-LP$hdx4{A=K5?bOhfSk8 zkBx*IN?I^s+;T^R7m7xB0e|&F-Q2J~z|1?xc%|FO$9QAZppeoaqo@KX3C2ra(`Z_f zish{g?_U8;S#Q#>!W~f#9ryl$84L(Y zb3vzz`kp5%m{9aqST@h?bI&(LAd; z3To(Mo^Mu1M3|z>&$ppOT^`hDYmaWEqM;AN90H2dfDs-}jZ*N^&U6!bhJgfx)_xei zg;J{to8pReZY1~@7EvSQ_Uuw~%$Vk{ z-DL_zFC5R^U5l1dLDHnQ= zyLCWKTc|hhs~r}-8<|EI#Mu84?{9v|3P@LDiG!em@Y~8Rd8cl_$kmB}Hg1NGVGS|2 zIvtLhFr8N0;1#bx*vLpHCx0$>yr*duk51;2m?$d{dBYHa@dMq8}c;I5JL& z*rmbzkym7-gVbqOjhmWjf5OGH?-??wz}My~;GPkcJd zxt*xy+1#@CWz*Xmg-rI6JRT~X`4#=>*=n)4d2Qcigl2Vy6M)uRzdzM|X;%I_f%4{& zbi4MQ5KXAB24$@@kPE<{9oTnAp(QE%@UV*`)Ruok1sUSee}^V^9hdg_N~~a1-j$^~ zu&=!qs)u}xNT1wVz4`Nk870v6fN!Sv^_DFr0?V~HTB+}0RbTSMaKmJ|5HGv;TbnkF zRplD_WN9fB#1;yM3r;dZyS=&PlWXJy5?9|R3TfC|K#RGabx>Izo9Vn}BV0JgaWmt5 z_3u&qEDo1nUSL@%Q1qDgqrT?cL1-DP`8fWF&S%3BisFL2g#4Z+!J4G)Mc<21l;`zp zAyEeDq4mCp2E~f;dcT!O8_BOPRFgLABz^=014VG1i|Nl9==Cg$59a`sJY<6MBEMAy z%R$Rp_jq}xT!61ltut(UJ{fb$ zTh_<%)mfj#MleTaoo@$W0-Jnde}sXcUEjCdT&{0#)x{_M+C7^Xj`u*WR5P^gs-LVK z?q~%#Y1xXsIrw3=0L)*&mfjw{c|9`A0!~`n8y`|EAC_vh@!_o>&bFjEt73;)$OtFU z^K2&oQ(thp{t|XyPzz%zx3jx@SkXww5>_EeF#oxpAo0zkk~^a=|Hc-@sc7w#AIlSK z!&(>8RPcBY@~5yPjnl(E_iU{2k#(^$S$@E$4*AUJ&fqcWM*?!{q%21o_@Ix?j&_3xmOE@Tzw<|Ca7V;DD`gsnt=Q42B90mjeU_n_tbvZW8k{ZBC^xz zx<|tW2X?53oTIS0V9g>RgG>ibt49}CZ@CHsAMeUQ__iE&l6|~FYkT6Wex(ngVPg?- z*bytiT;X+X?rEt?cXsRv){2!MCCJEM95P!;6~gOjz^=go6g12(uLt4`+97)fLiAo! zJQHQO&BkKyIxlZ^XhcVS3KExIlhuu^`P{8|uKE;*=cG-u-9J1Wnm`K&YBHFQ;R!a} z`4-j*cHHU?>-a*0r7R%egq6RpHJ0*8>R{ej9nsZ2t%A4TBQ8*LG#Zo%N+@%oPH`gP zJI8H-ueJB4n4eSRJW_Q!TtcUh&J?ybKi>;pHivDXCsouoQ!ns({k*e}4z?KbiV4Av z1?W6EwByh!{Wi)({%ibAC#Y+1gN=w_f3ML-u63ML#n*Vd8OI00Y1mYu?(~L8ft;%R zWKrdh?~@e(#q)?Np!@lg#4<*Oe14z2G-DPdN-R3 z9HJ0ME34&O<%(WoNIe1JWYlAgSf#G1#4E}LO~XWSM(+G_U3dmX(H4u#UEv&W+A{@% zJ=rB}PG;Vj%|bXOr@@hh57v&h!!4nxI*<2=aLA`QYTMK9#=Z-SsCVcZd3U>&Q%AoE z$H#Yj^waJ2^#K!%Ih)M$fCdFp;#ejoXnEjRbQ6r7JNYL9NVqX$v$M*TN&;vng&wGn z531R$kCN(Ks@N#JVHoH+AeZ|9ap}jQ?NEcL2t+z)TZ1=oKd4)!@6qZ0Ajrm^m9O<$ zKB&!~sNz$t*sBNPmW;2GYHzE{Gu&@7T+ol<{Del=!)w6UDvT>KGeXCUc}b;&Yw+#a z9=1M?JzB)Y};(p_1)#IUGi{xOA2qR?A^P)iOTNe zO_b}_*+I76v3ny1G96ue(DxG*cOX8OenZ^a+Pg~EbSYf!&8@QOYiE1y0*twxwb;oR?sD5z+S3n1VcnO_TtYQ0f_HLu&C(A6$cbWollo&k9czHj?Qi zO=I&}pnT#mnedDZws>3wPyKwz&>KkjE^k}ehl-M;ByX6pL?{)NG{X}pRg~o`OMvur z7Pi^CWJY}V=z?i0l-ATuqJQ~ay=q^R;^#;HpeX5Ug0J9q+e+m#dv40j!e4e7-Q|R( zDeHbXI-0+J=qd%;1@Cr)Q0jS~(V`A7q@)S^WfMS`W=|*-!*EdZ@-h1Hq79t_%|wd5 zmss$4JJZhk!1ZIzejy2Vl{10Hvde3&Oi4)_MOpkSq9NV&bVB)dCS#4PW=IK|gXO^y zh*~o7>jeM8&5Nu1Ro`j4*1jtrXr=rte-&DK|# z;~po>pYP!C7@PG&NQkIubEH_%Sy(!ZeQ;f4^Aw?5XLo@07ATqI_m%~bn#`;eKf4hx zf>bY{)IQ-vt6kiJq@<7A&7Mk4h6>6YQ$GE`WBGYntW*de_&#ts9ojtM1J!|H*LP4t zy!J-}Jj~mOzR4QFnz>|9f)qNM34t*qrS5>N#INtiGmC{oy?aiukd=@6MBl73axsv>9zz%WqZrdr0jwS> zSkti;1iU01mpr^^i;M0NaJY}F%AKFv_XV*f-8rmw=y(KgBT7%?2YE2}T|DADs`&ls zH~wtwkDJD&oPb}L7cx2Tp=&;)z47*~E9Z+~N90kzf$^`2h1GYHcGP>s07+@{iwtS6 zie%l%g7e0;QIH`1BNzFPwf}(kEf=iV&1G0J%pxUrk7T&(zUV7g1GAnUZq~m+$wz*=h%2OQm+YpaA z4cl+rBirO%c$#(e>+lEw(j?tY#yJ@a^WEj=nB!`1#8naiuQ4YVeVX$Y6-mp<8*(Kd6n3YD^7{6ctvj3!=#*&S-ZM|*+C2}UpnV%Zt}#t z9}gpjbvz9$m<^*VC)FOmJ?8KH>#!y0M1@rAX3zpV*fA{XSVBSG2A4ss)_kI7Hp6JB z{Km0YOI56{KlDGI2hQFCdc2t3I_%1Rbm@(=-D>FYVi}gskNBjp&8(^kJiV35l~@? zCYI0!wx!?fjzGgZ&J62DQOPc%j=L9QyB64zUZnZA-udV(R8y{A@% zFJcs}u$%BMAUQyv262p zhJM|7`QSt&-Jb{5Iy0|}i+QDbnu{S*D*lrnfZfdi4tXjc3X{$&r8XCI{9?Cn!z2Qj zwK7&%v;4;n^R`b+jrt=-8YvrJ0P_WDT6_BY6c+EeH{=~QaKLqG(EA;QHR*m$Xw8Sq zf0b&%X~>LEN?W#tvpyTSxW31kl^Xz_?{}LE7bC;+oNn)N;FZRC|8i89FrASa*xt>1 zKVejH$PTEB9H0<=*eCCHRLpR-kV3-)tIiv9mAn8;pco zpJ`s};^86?i;2CgI{|hiU^h@o-N0>)St_jH9GnWlBAQ#u)l4Jfveb;8zq@9(C$n%b zebo2Ffpbze{8o9}{3w}hzhUY`BF}FmdTvxh4F_`5+OO&xAi%EPkh&-jUU+y07-wgM z?R5LwiJ#PVwkrf>neoFnvAY2!be749({eRZDLX!^hX2g{H0NFi|I%9E2n|CN&+T&{ z=YyrWPaM~F`tIijv#UEHdo^GQZ~?G=am`n!!2G@x%D#P^sCl6YjToWVgzOe`h~kDO znhQAFjfnd3=l_b?$XpKP=G@kh8e%l2b6+)E$H*{5yQyTh7zPUcy?xp*4F9cCZZ zy*Yb7Ow3rXUL7?nsTkLWF702w@y@B<`>xfyJ}2HS^4j8sl>(RFKG$>tJ+lCS1I+Iu z5^bRw=JC0k_u-M5OVu&^rFA=bx0GOn5#iz^V&@2jcMs0^yS7dTEQ5<1g(aEQobV)` zB`Q+5p8tppkNC$s?}ANU4$_AGnyvPHn3Au7P~_1c`6bV{-u@Ma{hDU6`R9MXD$4K7 z{jE_F>6-0?o8)gDVL(n}upXOo5N=!La!2Fxing<8027WuZ%17V>%CFu%~@kpzCUR3 z>3)s)XiFhqrk&Jz7K4))`QeH7W z*8#_(zCPPG(Cg7}QsQTV=&Z8AzjYVADsfy7&G;HtzB{rW^;=Fglum4Q`WOA^lwtFR zCH>ItcLUhoqyOb3HydOV>ar8<@6goI-cO7B>vMbkND=3DyrerE#Eb&0iRQwM5px&r zOin@XL~J-q>H2*cV~T3jQ0T=)edMm&f+!DjEEx_-FRN>q{B*_%QyR)OBv#7dO{?6u zT|WaVIE!JE!!9%x63?`4dVUETk4F${fgbPRV2!ntkB<%0fPo7NzDgHG$mP%cSEx6^ z%mSlH$>m@42C9-CTj|YOZ_#4E`|LX9+I1JbuMV<6fffA8%wA1zPJ^-x-@~S0h_-Vj z+&=2v9dT@F12{xo_463j2au5R6P|vCiz*!UXtooMwEcJ-@4aH!k}rlhTwI>nVFx8l z^%GWvPlSt$e(nkgWZ{WUMgKagt0(Vel90 zHt%$-ch-n~lQ?dOmMz+yxRK`Bx%(TU#{km?3V7Qz`u9Zlr`_)IQM3?_OUSV((D4Z@ z5gg3#DrZKI2ch=2xLVt&+ui!N*Rbr)sa5}~Z&$uj6oWN00_t@}0Q4fI01O0{1>zP0 z_I?8IzfVKvL{rWn->SlZ{j#6Q+JGcfudT3IDC>iF#p(%D&0}?jOZ2Y%rj>Br z1ut-ZFit6LK+*iTA%rb@^>u%QJQNolv}2pK8MlwqC$YG)P6@06bZ~Ku6HDqyy|pWv z6*sEIW+&Hu5cm4#R&Zr1-dyxJ@2|e3@&tk9Ktm8xog|#TvxdaiFv;Z6G);UrE)Xz+HUMhY$|J$5`IhaHIU3<(CcgIr~-V4swV$G)_px5w> zxLYXrTWy7nw1VSsm0atE>~(t&&)VmJb1(iYu!HW^xv(>>+jnUwbmSg5!!{fs)i`A5SEx5sg5g+l(Ka?zauXp!!c``V{t#$&}c28;mAbdh`XN3qXLWPNbwuUtwA)v-8W6) z=sx#TaQz+?Z;!Jz^1h8@zEWYeSwqRr|MICA5ba8rdvfw@W5oy4-m^WtO7U6z+KnQU z<26m8e;SbD31^uhR}?U4oZ{%c6S{-eO*WPVoe$%aAt%IjY_d%zmp#hRP#RI^xB=i5_k@Wutw4(iuf8pG&^IFKoQV4vC>x*3xjGrW|qvwxy7VOxORm@-!#B zAxwk_U_B9Ty6gg(&j}5U_p&1>8x}jeh4yEk)H%)@+Y6K8RT7^LUJ5C^x1yL#S2s~& zbvx#uKW+`58nxfSna#6!gIOzndo3>B+(;BN+}4Dj-7;THU*&*nOQGThM=n|p=kJciW3cS|(QWe7cQ=HW zZuR`ITY0X~y`>w5dqCeJ)Yu&timTJL{`I!y=e3zMFlKMjHY!yc*6DMy3H?6tOhY9x zGOhIuu*by|k|*|=1-jkbn!?7oc1xpQOyEYibjd7a=Fpv#^DtZQonr||GhVqW3iWhg zviH>EfnDDZ%Fcrf2+Zhr3D*^ zZ!MJT?(>+!`+5bnIj_v|cM=$%R!FFht4RWFz5&x!<)nK$^$K*F13!uWv51m#gD29JQW+r_*#I-);>JAxB3*5ZdS5wE0c=$0|uuOy9E|wA&3l2Q%Q-?`kF_3v@ zRYmJP22@BRg0_13?X9`;_P#F37=bOAIZ~1zNZ%ak8rZDPEmc$VL3NuuS8!$#SF67$ zEr1?h*2MfX`9yn;<_0j*h%-i(*Od4~+4#ef$&%a%DJ4_Hy`R!xRv;Mns*eo&}NIHHwy-4;#w=DSOj794*T`Y^@lt?I%;KYPu=HzQ5mSz%jj$GyH>klr)3KS4PH; zH2>}cPyO^&gdyriO0f{1gs_tt%Yx0l$-t+sD*PJt;ISoW3U{7%<{$8U#4uEW{?V0%aTm{3qkt&x0vE9w>XMasjaGE zwIb&-f?mnQPaX845mUBQ78=U$bag))!=^%-1)p%B3JJ5=4*6a!p&ZGhi8H5ggNJ#L zneOVZ-v%e^$($^yJTcj6|Dj69={tI&n(k(>(&cKKe2*xJpPp5qYX^1_F~si6>pUjB-Wa%k zAy;*3R$7$rg(O|;5TJL^jQ^gZm}mz~_zg)s}&Hx4(VwY+5K(f5{Fl%_Jp&=g*39_E%nnsS3Q8+QF~n{>Ag)y?t)H|2!SV z_3K4`i5;7=#-YX$rr{f#;Kc8C;5zVpb`m~EoCfIK-o**dTmX4F^?>Cc8}6B7{;9Jx zfsa+bNZX`XA3M~_cyDJy>I9m`{x%A|y%)R^)x?1bEb^L<+jcC_O*kJ_qF-OP$~{Dn z11Fsa$Ft>)_x-#79o5;B2tvT+K;TFK8kpk|6LIP957@%8tEk5#R`k1{>k`=8E8EJ! zsP!PBiad`Zv~St7*!?BrF~^M&!M>7}g}Xp{$fQ35I);bFa|Gi2^KrTBf<0@(R;=bs;5s%I$a{vAGiN~Wq z#$V~Ipw?MVG<|`KD87|MQX5K~e>tan2IURo(KT+KcvT$gHBzy1K4`#aU zWDRi~1cTrM9!=QnWXm6RBY5ym{X^pcK8Rh?2kz11&f!;T_D_K0ru}As{Pv0Q*Yo}& zrk?){^MYZ{f(b4@Zqv{LDP_F?1MF#Bw(GVI@DHerDnX?yU%Tz?J{>=YOwFU+ycEGm1QV|)3tpZ{fv{*?N5t)NFwJN^k*@V_%@sR=Z3$W*fmO%Q0CC#)>VWY;sQ>DhI z(ugwfoAwo3dVEu@2&iNU1e?UME;uFD>tO&oThQdALd#3xp&xUPz2{{DrVX#m z!)Nf#jB~1QncvYbW>CdLV3H&^7}G@62e>|Pk*hkWwmnCxQu(cT$E<|04kR^J zpPN}-Yguz;sAf(k=Z$}L0#}4DQ@qhxSJi0lPp=rbl_aY? zmuYOifU{!&1N8C+7PU+F0hz05a4K@u+q6E=c+2?vx{%&??|@nn+IJ2yBu@TfeUD;O zK_noloM9atS0HxTc;Voh_G2}eewcW3B#15nZ%&sMDA$VvVX zz0I4qD|O(5v?oohOW+fn)KlGl9yBaF?W*DLQnMw&zSAYEhYYNbknye^axAF1Ovt~^7h&`3tjJ(wNkZnj3$-h*WhRNS-9}Bw;Y!;U3 zPkhqRf1q~ecHhVA0ZyW>T7Ua#!*rNt4KJjq1ZMX=EsyU0&a+W?lTeUMU1y|BE&k#odbGg68c{42YHo44YRWj7~jwl2!OAacJ_@ZOc<_iTZ zyY0Dd#lO?@-^tlse(e@+6`>TvH??gIROmpH_=*!yj!r!(7u4qbr*k~0-moi&Y&}8! z=VnjW2}(#F(Zi8sVKbJ^T-=TkzJwVNb>k*fe79E@)Q3=aZ;JW z`DlESt-L3H>E=mvZ>VD2jc~}W)=p2>Tn!}=?{!SrMdEcm%HccA5nJ$%WyDe2P5J>d zhINhcChs3_;==lPgCn1KIAFq)-k>5uF&O2e`|X9q4XPW?;HZq+eP}4+Vt1+XrYi$$ zoyX}XPs_^0$H%fOlpF*=@etVMitxXc+U%A9&sUdDDqTGxPmD06hKLS{yd0wkz9u|r|5vC0aB)XZ_{5*1;((JkK)HYjp8X$rO|AgIfTmQZK zHZWI$*u^MbbeZq_%&u@8FociR#s)DOI||Pk2NrBCV#j));7e`+qs(7 z`l$BjPr64hRQ$^(a>jhNkn+e9FS=}!z3%eD+4+bIMrX`-PFCL^5A)&j1Y~qKkrxZK zmXTqj@Lveg>me;bdg)MC=i$IO%5Rbj^=08vdoNX?1h7GPuGlY>?fQ6@hi$5lqK^f` zG8g<#m7{{h&4%t4J4m9QTdU=+gck9!pz>1|DUS#StFI9MpDwKme90(0vDn z_}Qu=GCc2Y9kcm$z|<r#Afiht+3o1dMoOd-=CW2 z>vPR<&(^I!9p%UUafoLin*B0iDA5>=UR472ug91i%687%qCG#B2;ZO24QkdUgmkil z^P!`zceKD?sR5zJe+V8KR)N5en*!0Rpt_xw@1+U)sBPaT;Dx#Xq{53qYw2|zj%5JU4<)L!ih;`Pq`0H$+;#vbi z)%hCHW*GKxjDC;9(>6I+8JGI5B=#Lx&7;12BP5b%eO-14m7bT{r@`#TN2x1)nD-4mbcj` z_Oe;-20G5iu|#wEj@|x;QO#ZAYC!B@dBNWtcOO(`bL6b^WTR$z=%RfRcR^u&yM@(# zp8|_->O@oTkGzE-t+6Q6HlikPDO&z#plSgovI()?LhX;TgKX^akVjiU&MT4l$ z5ZeQ?dP%Hiss-3=)~8bfK}#Zt%<9W*y!zd(-FY%ZI7KAd6W&Qy@o-%Yw$F1dzV53~ zLKA6YihJX>t>sLgYE93<8e>*{o)@uYzC0p9p(Pf3J4k-Ew-ivp<2fmkXFn@_pPn;U z8HaCJB#GtdS!R>?p5pI-u^6)2jwyg(O*L=lv+;`^W-0&Ga#C!iZy=06{vfu0ffekC z7T8vcJ=UZ}-ENVL*M-f{Vkg9jm;oHQui>HpUuEFkP@fi4n!X`ZB?z4{w2oIIQ$L^B z4dK~B%Qf&?Q9CB!ti%l(CdlvuAC-oo+k}1co^wf1%kq!HfDWHqouQVYXKhjH=p~$T zopUg7HVz{;7Vk~l6e6FVU#IBsGJY!6-(tSPLWmh*Z8Rh^ybAsPu)j^f1gml$G{Nrs zX6CDMWvKWBn54nAaRy!`7xRNn4KMxxwpG)-tq9;x8G@ia@nq4 z;@clwkIQ9B*?FRI2+u$ZS2?!V2S|LYvD~K#{ zWt;jIebv)Q?(6J8R?YV=jh-6YUi#AO@WQncPpAwJ9`=t^oMe9__6)h|#tb7YT4sej zwr~avQd818R$~Ye)#Y$NnSH2vgy5=UdkBroBI-Qu{$G(@G%>q*V%B(-?6=STN$6Sc z0s+zTM?5`QXje=e?xFJjlPx9@tp~=UN|fSWKj2i}<0*l-2R#BtC$)1P*Y(xy|BUrl z1ytRO*sXJ3V4w7j4E*%OS;7j{#(wCiKm4U%)B5zlh)rS)3R)}#hdFz27 zhhuXGv}d|PQ9;<{C-5X6WpjkfKokC@xgZa7OWn^^3fz{r9y%r5Sv%MiI1xWm?2p;7 z%D!AkS1m+37w{)i75wknyI}sa1BXV+$bye1XXiLzp`35R$%T02vPev?YrQXnf1^Ot zuD+*s+|ay6QK+Z6XDC-U#4lr>wSt|>%$k?B;ZBuc@U~oBkdPV%MoR|W*el6fKrRe&dnlex-2@Vhr?{M<61}5zAE6`UuBgxD^4)TH7c8) zm3+|;&YvG%6k002sa*j*?`eJTTBGLW0gS>Sv+Q->vXqN|WB4 z@>Est+$)Nr(V0t}v;~4R{$Dg$T%ZvB=p+pdq8=tp_ZfG#s$I-#>TcA46PV#kC_)bRqE2;jcLL<2-Dtu+#f19_p`tv^XVwY2kW{*x(`|tX*oZ!S&#;D#6Py zMA6^9sH|Qt5oOEFP5PL?29${JY3;d4UpIFVPry}S$zOp2mXai&8)rfFVZlplQ*Lkh1{)C2QU%FiA=q_ zXIaXfRJ}IZ{USz)rOQP|Hq+wBl_Y9g?_rs#m|ga0#B$&F*)$1Z?syFCeMXS0*aMzv;wdc5)S88?Tk_5` z0S;D@eqRZTbK5kOR@t^X>Y20I5$NkzW_opI?)3h$h&Q5NSAL(5g!P`;G~AB!Cq(9g z&v~`qo;Rg^3}rx56{H|6`?=j!55D_J2S!nj)m@+UKDxgzm_om#eXEAGqS=8gpgu@( zYmJ>0XH?(8sbo_KSxC~J^Nvdo%IEmHA1|loqhj*bUyCp7SArRJ;@PX5V>Leg4GcWKD083}%XQ-TjU|s##u-z`}EDfUB#C7G$Pk=v;T{Fa@!`vFg~VrtJkB#tt-D z;5>WJiAL)F=&1)jldaAbZ~gY#X)u|VXQ`0=V>cUcvR)CTcfwW{J&(pAVi zv-40iwqsFthOMihJq(I=3p+^($g@ke+NF% zD}(&_$E=Zpdu~A-lRsR>46QpzYPEkn2du6OnH<({U_KH=iPE__snYSakesS>Q;Ylm zHIJt0I!JiQX~zn!H+5zgv#!27*2A_t40$4Z92}q*dFP3-r$gFW+^m_p)wIKEP47~D z)4p|rJV8=eXM2B+F3N@bw=p!UO6exr-wyy`n&T6*B2bh=OQ)oFj0Lw;{XFaG`q0QVakFZ`gDH1)SzwI%H z28H?#mxFM_>2ZEz%U$&4qe^m0P^2V3Ty}fdl;K}k_=^2oZe>>~cwUN*I(PI25Pi>n zzA}6!Y@WcruQ2j`%iP|ncb!aRsx#+^=j17E*bPx^_F78AO+uLKY#=6MV*~-zUAX_| zvs_);S2FFEMX2j4$v!XS4^aJ13Dc^FVD~E*09`#~T)l6^06ddG;Z>gx&v(a9Cvgun zhKsu4bBz=S70mxSpuX?Gv$^?j`8E8@>RB~DSSB9?)AJfh-`)1YPJ4iJ$8k+x2#Qqd z*?;T+Hxw^Vy<}gOY^Hw?I`Vk$e>CS&&1{`N-N%o|6j{Foaw*}!^J?AeyQ>~Yux5Dl zBgUo&ciKzrQ`%qszVS)YWZ&JLA8J^To9z3ct;%n{FGmb2DEulmI*y7kd~ZcXMNr{E z2=|N<6jxsJn4_MnUbOK?d>XojU+c6Dh{xyaX!j3AzHSlkUoJ1Jbw6%yf4ttn7^S6- z$}Hh`BG0*r5o>CTJ(+Xj6TV^+`AIJhq=E`w4O7fhXQybse zcild=H{jM?LyGxvV7vEkHHUPjg|&7h-{}>Tw5h9PzOtMG-IROy@`e6*QkK;N7{T72 zVmvyY;B(a3)uj$1!roH|f5lo+-R5NPSD#cu2rHr9s>8OyQhm1h7e}r}(?-2#;tB_K zu9b-1@rQ9Wjr1FQG#dBOCp17ZJcF!L`jW%3J-8k%T*Et5G{iAwv+K!!tiJkUP4kMp z<`;`({X0^E+tw%@zcnJF_McsOTR#+MsvW`c2rv2eFB-M_yCJ%f4W>BfLE<(+-T%RPt5-ztx&YGWCj-^y96hg+L0?R$ zxcrpfByu7@OPT=tK-CErk@0>j#V0R(; zV_lR_=$?p`IICl6J2M0+@`y#i&|O=-vCzC*adCqyy@t?L@S=ls5B~Y6ZF`QHLv4m_ zA%O%+cv%mFyoHp2Ri2M|z6=xIsUB@6Qgh-y;86bEX0ylLRGz;845CVh&$78IMQ+YZ z2U5U#4JF93?PDpw_AZpWMLyi!_+=67rlZYHkT06vn%LvsHq$X79@S_|ap`#|H^lYb35dvG&fI)D!WHQn*oSRY-BNlfw+^M#ii`ra3LlsM9e2Ok45D~>(@O{c69 z=)1}J#KJ^J&E%9s7j+R@&6SrTV-%&X`;*8szOe3o*rErk($`^*aIaczyjZ|gOsZ<1 zo2z)--S?j4Mkv&pz-V0h=|o^99b~V2K@ON0SY4@~4p_OKJT#*qoAHZ#i%$`*PKCg} z%j3QEwuI82?fy#1^(aCDN>k+Y%L*H=?(lA%3gY|ezZ4}>6rCYRv{QjUfPW@umM1{u zopL&Ybi)Gt{OX}o>!Gb?b}u*Uhhy9XTA@Bo9*}DIM-sH9=r{EsLvfdRZVQtgW1sjh zLooor0qLF6Hd9reu9o26N>=(#@Rcj?gKxnXkxDzgD<4}YHKXcC$smK$1TwQX;Px`H zr3<ghuYZ)i!0 zge1#qwO4Ysy&bVkE4h7;B1OLais+ozz~(4Zews7R@IUhP*OtWRNMVtqgJQDslr;y-4xBrgJOh8yDY zV}V}>c(#wS3WGgT?U}tz+*rh`SX{)cy|o-)tO3wW+${2MkGZhusr>n}dRgcsO;bytY) z>$@pnG%p53<#4g^H>dnkAQEr2^b7O zBag+JRi7tm%L27T7FRF=dgm^dl{m=hmi2>}Sqe;>x-aH$@)dy1^!g5Wln|QF59qRc z+k|WoC+Agosb{mmg9Y%z(^lpT34?Ld!>jX;&}Sg)0d5?eL*kqkC*9Jl^fvN&V9hK1 z&5gEazJF+wNl^9p+REBqb)em)*q4P?C>AXc@e>B#I_UG-uV zA)Xbn+4zidJs>l&N6XZkPMFvVd8 zKkxq7_z(9$FqQpQW&z&M6NqE`TOZcVN?PjsQ+u#@;fyWoJ=5=8f){}2LZIc}@ z8%EJ024dO`iUwaS^j;|ErJwS#uX3j^yl&z2&16|+eB3c>7}*$+UIq zn!TK`(ytlPsDR2v8Xf)L-LdKMwleu>$BPkP8^(?@p74;*Yc8&9n~!gY+}YRJl&~;A zE;?Cw58e?z+sK6FvcV-%OeV_xq@Qw+twPG^6x`oI{4Lz?9*M6|xaT8_=eyjfe zxfrbCC&W{-zxV{vA&>M2P z$0Qbw%O5sbHg)XJh&H@Fxtf4Rg%fdk`wB9iLge0}cd%~6JFo{h547_WP^x}?)@lyY z)D0PI($$AHb45wj-!B+rfs0pxhj(|x{2)bmQ`>UAt3#heD!e`$lgIRGRO>9A(;_yPu46TcpOs znJv_>24FsTTF-QYunpRl zNbYS|O^Pj;Eu=I<+BL;9!A;25=S^(5R#{GRXhB?6?TCY8mykIBVpU7*RnQKi#xkGl z^ruc{g7+hi-fsp?F5~{jy0#AS$Qhxt%o|lgRb`hqjG3TC!b!uU!eVW2{#7zwYbkMz z&p0K@_UoYx71=^t_uRLt~{w37DCyibsA@TMX#?K*p>3Siaa@8o}ESuc7ed$F&>RLYZQKzfuMjDZD^{Pox3A?XI z4F>x%&0O`qneiFeqJH;<|5}s<0Y9UyC{)g;VBFVkp4H%xZ?RG4?j85Olrl`po29sJ zwl7cgs5^haD;x!3vxq4wDnRd;y`?D+&d{`R~8#U!U9 z<&y4t0lk62!NxA+_xjG!XC9MH=B#g|B z2ZEE23E_ah{7Ud2MH4s0M1I3m1;*miUBAD=h=^4$Y2Bm0-EDLvq-UdNClu~3Ko(}j z9xO50m@SlC<_7TD5Xvc(iGx2zDqP|K?-1|r`4AhjlIP#rLN($uc$mv>Gpa@{A=p6T z`1maw`y^k8O*d^PX32V_YBzfMO=YqJCECai+h-RmD>nfLAX`c=nYu)R6dYEv(l6-L zDThjA0B?Zjcc3^{J5SHRx%508P}AWed%J;c47TiMl2_v0U-G3cDdHt_1FDOR~ge?V>;5ZMM_JL)%eM#CW~-|Id~(q z0m0i)tRsw2uZLSUH@H(Kw({=bo^+5qIvmD3pQ6$@sqnpLozDCoyV;TJdEFL&ybJs* zx%9Zb&%EZJRwhwp$#h`&JE*9nPKEq2koo;vo!Is@Kz?6XHrMCXaj?(2ZRKc3GEZ#z zRysE30-&3ujkE2 z6BYD)mF8wft8@ro&*&8iZ+8bz__07f$4ZRtrI%*&=Gkx3@J$(Ag9*+*!?TgW;$Pkn zMcMn#o>JwwWbd5Du#%OS%?&=Q6WN$faJqI^A^e@rS@w4Ks~jSU=k!tx?iy?QDO1g6 zTgo@X>}%DV^W*NiHYssefBn9%BjAVf6tD{F*;sLq^5qs0yMLL$(Cv^~ zRsFrf-44q(=DQ9n$_G#843c=(z~7SJ;{qTW2RjW;o40$JFqtK+9#78~6i)`5eznmW z<4DIh)xFDxIE`1FQa97EB4k6ZdwIK(tK$pz_698=CT{unfGW_aZnz8SCikIS=R=^J z$WhCQ&136k=2;RsA$j9(uW7$X#&K$T?Bkt;TWxFG>(8BM8#;{rV%LPggFmK?>TxRsK;VR)E>XyNRlQ=K=-#oI%D

hFE@)Xb8j0+qp2(bBe^?jNoCEig&x&=* zfzPV2A=$_}U7zRxhd~qwCQI~C)}t~2KkBB5AIicAU>`y{T%B>%cmYNXAW&)0^<-ZY zymD?NLD15#4>Z0;8H%Sz_8~sT{QOK|FV(ByJ%Gi;>{~pVx zU26Eq+}`};NbTI;zSw~Bv(G~pb=kef2rK_>DXywAWc%kcbDr3EJD=9LZWHwIP1c{W zldFsJi(x6DZkhM4umB9*6)NT(gqQ2t#8o68tQ%7p6vIQnMG%dObe`rfT+FZX&(S8I z-7(zGd?gpC6ho+it78%(ApKI|M0*$tuHj@r}qZj@OSW4U-DAwW90%bX0*kc`#ZnI-^8x+%-_SvKi;@Zh`)Hm z?msH+J+c1=+Ih(muFD}8iE$Pzx=_3OxNnWaXNP$w(?5!GrSEWFf1D8A{G%CWT~y;cAb+Q7<(R(Fi}u#m+vIu;^WjC%vF4s$=xCIcLG+ zu%^y#ARx;4fOLhH#jTX{66qOG4GM4BGWkw)RXWuVcWYI$2^>a7iglyl5R4^w0XOn|7GD z26aBWN6!(c<^Nse1OhT2~to4NCW~%Cx~z(fyt>42vP43M9NHw+>EY zI}roQO%)8@#IcuSF|ekKr3prvcwXIFR?|!Ja0@Z}-16%QFZ=9y>UWJN_N(Ch1@^mA z0!j4G3ly(znLe#duzpFG$_`mxCqoGK-R=^D>m___Wpw%GwzBAcedeEr1@Q#G#k(|teGFrVy90l@3V~?Z2KK` zniEIsB4WzH23w+NF!_{6NaR9;<5kRiZdxc&eM7LHVfiYm7Oy}q5J>|&?rq4<%Ydvl z1VhaN^x~Evv@NsYE#2yn1m0b-7D%?QFg0I=Gz#7R`LiBl6iBv)#!n7>FU#Fz=M1TY z{u#z^=~;;CQ#V;BBnv*d<-(SAj~Cs;DdpZ#Z?f>on0vj9Ql^Ya=9x>LXQIg-o_J?` zW6H4)`x?(sKaESd;#B+>6aFqR(5%Qwglxy}!yNBgUvHW1H-{|j0x!)pb4o}g?!s`^ z;8sdZoCNy3z}OXVWDZc`jS6>MBpKA|Ygc8-3K;0DMgqT^7dR%$oaR|*mmNNOytVsV zQHhkt;B;_$I>2r70Bzb0<)-h5=YEk=#3ha zRw9t$-$=z%Q1vV~xy!$*X05rKCz1`|6lJ~2M9-Oj(x}-V=oCZf$sg`;Pvtv3dIx9v zc=I;r*duPy8>8UOtFe{e5v}`1Bu#g5X55Ws+DvXK1s>?SeaS4wZux3@oVr{+=>TO+ zyT}s4NSbt%8Bim_Npi2X(oiQzERNO#lv^zwy;`t303~h`WJ4~#5zxN%PzdVx>giPg zz}DIcjgMr^-al^Z>|5!nXNT`VeDriT>In$_I{_3%CCjLX{0P@NJAJjNcNwYr^0 zc|F#|1hl2SRgeiJyT50sl8_ldnXM*d&dPyNbW)R-^BOIY1$*Q&bW4==q%VHX+{NL4 zsgO*Pj9S>Eo(KJg9~>A}2W}hR!E`m_EbL#G!4SnD6cr*l8$zsY^cRAS7axKj|BwgU z_0KSn#^|o4-dr98qI|Z*L zcO~J%+*{CZc49S@($>P0XULyP;5}K*-UtDBr4xPz^nv-2{vmTtugHJ^V*IEAeP=D3 zaK7NT+-IzAgV5%|g8F72%jfDT?1~j(5wV~k!W1KN5u94FYB=X&iXGI zB*14&lo`HiFu+b~dqpYc*u!S`317L9D^38J!0%8)NNRe5a#f>0pJ^a2&o%w~1+sh2 z*4PbdlO0~AJ+oER>ss!Swr|gk^%`1XK)A5CaFUq?WE|}T0jTgDqn)<3+PEkC+;3bW zTZ~KD2*>R}5yueeL&hffF9!to^!-%77jfgxwV?*jCry;e#p%hkzoY{)$>*&UR!L}9 zxPQ>^_M=>6CODh@B~(=Y3|esXgb%aP7R@T@5b&+-J8P=v=Ze$un-PonZKUtL_Ba83 z1WTd}P(VdK4zqKg(V)UedISC)*ssQp0Uj|#s3?jgeU;Dk|5z z9+5#j5i9DMVUJTX8ph_kjNLQNWLWI&!}TEd5ghAw?{AsMxUENy%m7>z-kiZenoyuZ zJaEo=BsuMKw-fR`SHFs@Ct~Lb3s0VGMPFxBeaK+{-;iP7-#%6tMNbjsqg^J?(Vf

O9|I5Q)BqQsZ?_}We*z6LeNT=finau^JS!ll!#%98e zTA2773>NcyJxg+E3UrA>o^4E?ltx?1U+sD9hUwcXoDXjkAe_`dlk~dVo@P|S>c?mh z@%0a}w!wqYVd5F7X5Ps7(b&SO6--|eEcC=6>n37b=~Hkcc0aMW5W6tv-|#HJI`_JU zqIKtF^kI}+rq zy2E}-X*ydAev{V$Iw-Rj=QAK@`?t^sjNRMZU}#JTT88B{gh=PdrPQW)7cb6_{Zx)} zz(OP|1=ZAUcTC!CY7qfc8d0rz*i&O~IAao>dv=#4u!ZhvCSA!qWkpaA{$@ct02u<9 zW;WXoCwZ@gUivLvem6yWOT=e#cJ=OQjO!!C`9O4KrgglB`2Fab7QKIS<2yQ%r(q4A zPbwdM+r;=~4w1%!uQ~;I8l>iz#$0c*2Gxp@t2Z44l^qWhN%d}#(3fvWg`S4B;zpRp zafnLxXN8$u!x@=ba>VBn;8Vq%d&tc3`L9(wW7i9jgKV@Rf8~cz3#~&YvyNmf+!jId zkkW{KyRYx5cv9}}eWfDgUvo#a4_ar(!U0k$ST}EhXyec^MXTHqZfv>kYRiI0;d6v! zBov$9NQZ_kP!YG!d@FK=?o9aVd_q<*SMznhA+zx(SPo1aO->HG^BP=TXWV`aMbc(H zLqs<|M1y>LhB_wd_#ZktPkON8qTL?{(6?b^bySV?IfAfk#@lclm^u?^*|L|UP2IkIL20f1h?P<* zzcchV@De?ArGGZ4559>R;pxA4#x0RJel=!zr!Vtw-%kTK!lTUrE?4Jd7n`+ba7W2o z{~T1aMF4&k{`gXM#3N`tvYTWMryFb@)Q13ndjdwb#ITQi-yOpN2Z-prK|Ho{$wPE9 zop$Vcz6x~c060?yzhElixXm%gd2SK35M42=$a68D|Kd-4-^L3MCZCDhHo9|2v7v1K zu{=GHqM8*SI^e4}lwzxwAU%Lp`I27-lsAI3zx|4hP3?j7kzlLBRmx;7bKnOe=o0cZ zRTqD(67a7aKG0CI#bY)O9FR|KBEc^3-7tVj9@FvLKRg@}Ad>|IuF1$9_o~6&<31IA ztT-+q30Zxr27WDqC1AsHf&$uoN`aD7Sba2ly3PK)feN|wyXjn@RHZ+tc|&l6?3XY3 zcRk!;NtGP1k|az3xI?1YK&Ps zM|B!32{;S^OD1=D+8Dq~yLlqwwFQEgNa`zXSeKqW(IkgU7Dwkrj46m}q$2zm5zDOr z+{~@>#`yl&H(NNS?yNX9Xj?%Sjka`_70 zM}hq@|Dr#{@h%T{8$Bk+zVKmSWO?w1Rx3v9dc*V~b|}}~>2LI3$=>&MWwWj5FC^PH zYEJ{=akZ!SLV4AP|9qBH=~yXpO6U9pej;uFm*mz_BzqZRvu@majajsWJR${}7x}BW zBp0i(chW7kk?l`L+AeSK|pA5T<7<& z*E>=KVhgD@rQm4ZDjX9I!fR< z`6XoJb4+|oQ}dvC)L&ZRxhZLy7xt${jj!Oj>^Tf=57TnwNxNDSI~n+e7d@Olvn1yJ zg3CAY(NM(aUQ4^v5KNudbzv!;aG%+F+|X#?ss@tq@ z#{*^n=vw3HBoTK`0#Vh?Ai*=vu+Hc=@()4@=O@Df*3pBm7Nlh? zrNs8wS(oQ3V{^M=f0~L+9wC*{43V4dsi9bRzKZV*fj?#$yd_VF^?l!fByNB;okWBB zhb(+>rXb+7@eN6X3VXu;VH&Iip-tbn5W<=?acu7=DNO{7)|BRq5 zw_=z*j<-T1cJR2K{ya%uDZI|iY2&lJAC?WEi(?)~VsjaXJpa@jtnMRj?rK03G(eHU zTNq9oD8-68`Hf-ci(P6-1Ci(av6c8v##c-}Oxz(-QuVsb1Y~BhBfEGarNciU{duDD z#2~Oj3ZqN>2hoDHw_`s8O#ZEoAQ)gfLNSdF=n7)qx@VDT7&h8*vwuM)7`n8(z;tu=*-WnW?^aEFIUbg!n^n#teX~*pKZ`Jwg+m71iYM%e~_z z4UfCzS-i1^O?YQsFjKbip7osyxA1}0#z%6f!TL1%>OPzi0F}vTV2B?Zh zHhE<1mLc!CZQ{H2K4l(%Yr7@x>b6y5Iv| z$#K{3C7saZCK(BVKKFUJbTCOQE(!P=AOr2e@`>0Pp4wUB_O6f;3oAY+RYqVm>?$^) zGqOZw9!*phylRKND)-slxwO(ZK7`g_&C|K{`l`u8ABnW&Co2^a8q_ z_u0o7qDyec;rKHxdt-1v9k80;rNK!-MfsEs$|L(MObaj6BOQ+Fc^8rzAFI*5L3vro z>eb_dp&fFbZeVJDikpn+(_nkFdi0;bM4$xg)FJYJ`=AymJ=N&2l@8Wj#xGy-ob|=1 zRY}VvZinwsdAvH^v50l6G8ZE;_FPphq6v#%gbU+PL;K*R#0YZYXRf9vP@Rz#+YFX^ z8@No?9q~CPtP{a4`dm%qPY{Kt44f33Cu+kTKv5r>{Gh|oXK{>W)%_@6D4}N2bb4Zb zIEK*;i49*PHER?33oji8t50&&sKd=j`(GeIT+qmBQ5C?MxZB+y-*?6-Y+F&BM3^k* za5l}c2qxcO*Vj*KWvO}MBPI0T0qyd&_g+x&o+}3tq7Qvf>d~(8G;#U6T1X|X(?{vD zIN(H`(`)}(G5DKy0yc`O31dchHkuk{Zxtd9><5X?FG~jjnRWt0m&yYO?v%t`-M}RBbMCh@ZHw-AY0}cO^ zmp&_Qhl@3jhqCJ8p2QlP4p&M~2v6a8OA+Qd24vj{6`XRKnz1ezyL4|FwHQJZ%x@{2 zKlOYt49|{P)>EEpm@DyfiV;?gx$k);{&}P9^Gi32(#_o6+$WurzP_(dbp6>2y{&AX z#arie*G}g>+U@tH9GQI%xWF}=)fuLWkvvk@cytf30wfg@Ksw~^yQ{R8ppjP(2wb`S zYt}7AJidk1FLuY-*c6kBIxLt!7}edxLxgAZ(E?f^y=sxSuR*t$KEGuZ(38CJ5IoEA zKtQTJZMSMmT?B!xM>)C9uj}}Dm3h6CQSWtRB<$7$LX~Jl&klE+Is~^?>3e$`8a4;W2^2el*ejjV$8_2v)ii`5?AQ;aR&{I%Z8TB7+v7s|9dN0p7pfENpg zOR`!>Bvx0yL2_o=$NBX<17>KvTr0Zq5eg3Jo_&4bAHEN_XV)`3206E8VRcK4Yh!n< zboe33s+IYJbDePE_|o6F+F;EybuL_!v9jA`>M3;(!M-WGo9$)Elt3nb?s_5+<5G$; zW(xL8k-l}U;TWs&rIOy`!*w1sn749<(K+x(UJT`n#tT>PW3i$-qtF6_`G%P`{5%@x zPBD)6`}WN}W9e!p&4^%s5sC=SKaV*RR`@Uq=b)S_Kj@y}VL&MMb?umFW;TLP^IChr za1!W#6n=-thZb&%$I&q&)9|9e8UEAfcuvB2+`@>&-mkkx?B&hX0I-m$Hr{%L=@obT zSGf}^N_F}*>;|HtbhKpyV_N*&P*qC6Cu|-!5GtD#jk9lt zD0eMb%6%%ChBiIIz-Oviepf)t(5C@xk8XF(nDd(a1xA>h8JqtSy8n8IJM3?QuM+s5 zsgX#yGTg6+9RS}5uw8FZt%PivVn{n-u#%U}9w`a(8GQP76SOyYRQ$%yD>n+74#6L& zb9>);BBn^g$MOChls=ILw-NQGP(fLyHr;}p`Cy)8Ai=}x)F1INkV8;!dhcu5A zZm?<;Rm}HZw*IP{fg*vihh=hnJJ=8Ytph|Z8Y!p-%&?D%>jhEn<8gaO96O*lpW#Dg zKLFRAxS^%+uBxG?B!Ussl-vebGK&BguHTokVIIcy#4C{z9fGTtwiR7b-~hbnfB&#~ zRvq=AEC3kC23J5*nUCp>mPQEvwA)qlw_J=MRepe}N)<5D8Nw^I%GWI$C_e-zr^FmW z!}VPG?&ksp_A8Agu21HoJyj0^_ks{VJ7b~vqB`mH>d?(6-tAN;y@qn|2KaHgU(!pQ zMNhi~0S!+)j`97S8o!`cC;Go|;0obtIS_|Wt+w!`mvB7Qot0I;(ABOLB4GOEq%%&a z9i#OkEv8;4R9PTOS$)WlcIWb3e-s!_>MrvByC5kUoDIC$b>2Q^pnn5Bq^Pzm_c}G3 z4Iq^P4%@<+(?DP906;+~YU~x^h_S5DwH0OvVoe1I#QeBJJ;$W^+2ndnzdo?*IF13( zO`PGW=t$2a;6;|UDeJbnaFuhrx*_?YkJ0v=;*t0|!*@n&uTtu0=J`B)uTl`6yTf_F z`6Rxvqr3H50qEo`HpE%py+wctVtM1lt$pWVx%r{)kwrldKfvO;WT{EVC^*$O#eip` zQW0cTU!UDJQjjMUpRcWiwk`7=%^sQ$mM)oX(V(R)800y=b%=eu@T2-}7j<;~Gd0>^ zXOmj+cEg7+9s2p|iRnpyKOLaVLzJd#AO0yKWCc&&pd}*#y+#uRXJl*W7{nKqi+OuC zHglHrs@+54RL;pG1p@vJu7sMKuYkl8%3-O!++XZhc488l3l->35EXSIx>O_i7IaQS zJ#!+SynyDkSo$I=h3tyVN@wMF0noFzA5G0c_nyZ1IS|!-&mc#f%7;NN59>x)3(806 zXv#D|4i+BorSEGG=rh@On-c8T7Gfb1U8ir8omRC+JV5uDEHOJX3-it6GSSteB^FSz zF`%`Db5T`^hrfS%aeqrKo}Z5cmdm|(fOfpx`hP8NYcEWf*L)Ov{tPHaUmho7blzJ| z7s}>SW~tA=r$VmJ^o4g3gKm&`N)R6LGpPaG+>~tE&F6G~xaAwqA2?A3Ni!0^oe}S& ziCgx6{|!hKzug3fx{CTI7Qf9I7KgJj^f9ga;Y(kh!gGw?Dfr^%(@i!h%JqR6AVbzr&JmlKq?3RMTi_q6Ef}5(*w5I))AGN-rysKz@HrkRm zvi@r92ESI<>K3D()8$!$DLf?PEt?k(?hXX>GDxZS@5;eJ=lT6;z0Y;-vX!(INAc*3$(GV1we`vyISX*AC3=Ff%3=yONtmMQ}%+ zYpm!GmBpl;UdONS{6eHi#^&YiGalfhkLqUbvVA!7+TZ&1q^=^3bdy~CGB=C7N=&2c zoIZ=M4BWYtEJUl2eB#T}jgS&Nqoe^bK0_+2~Cn{O9j4HYk5t^bu>MlM=KUieoR@ zZhH*Il7&9>+$$oODV!J)O$ZfJF^pi;AOzYOT1aXu_zEK<(-J=Suqy%6FiUvzICspd z7`J+Y$VTsJ4sv!K$`{0Kn(5Wv8Ymp2Y4aQIUi-+m9X8tBID+8MFY58yY?PP*7U>jq z?(1!~H^2F!De8eIu}Qze#j%aP_U8Od5QfCZ>6U+zLzK>VgV>=y55{hOpq?YLrzneP z6bMxD+LU;2eM0gOWNTb9wyXA*@Czo*)v%@{(#d>Q&pqXdWOcYR(nFKYuX^(c*pdvL zMkm8S0Q;(N7WXQHZ0DDpQYsKxqoQuVPi&^cA;po>#NVl|GlL`brfLY1kk#d;bM>ccs!d7fUu6FaG8M8y63e zn6;3>)Ir>PyCHzG?)kLq2Yh$Vn*e_$_LsORero@l_8Q=e6)y0Ro5yUlq?Vi27(~SB9N;DN; zySR)(+1btYiP{!rfEIwL_Lrl=YdiZeepwn$2wK4TgdTNzQkyn!&QG5s%@mgfd^gI{ z{{=H-jC{%Fd!>av{*QTC%XA;CvDV#z2ns8%2e~A7kc8fN$~gKx{p^fd z`BHHEwrJM-DkaK8gei+xN}I|B6M}0WYT9!?eSj&+ z*C}IvpT1c}Ub@pWGuFE|bex^{(j$^pp~`N60fz{V{BGItDNmo_HG9vJN^KyZ=F9sK zs@%rrC=A`VHRWFJK5aG)wFwv;@T5~&rJb5@g z4JIpaNE*;bbLjMae!C3vtS7MB+Yb~`Pg{L!zen||%8nBS>Q-jq>v>6;GX&u_E|D;E zTFjoMS{cPpQcY5p3TQke*pHjMgfmdR$B6JZguO0Z`r2J8ouIvPbJ1V1M7QIE1<7mO zw>~Q3xVovPzMsJ62|jU?%yKdddsTCO$MQr-IGO%E)spNZnc4kJdGAze&w!%gW~+0G ztV!&CVg{gr(Ky5OKCJ)QKe8=7S#L#p$}Ai@_@yMz*P zx_019>o>}s`JFFAdb7o$toSiat_cmUoNIAY3pX>3IUU3_MJ;YFms+7eQH5UfYExM@ zJLV0{iadqj@N}4-CqM2n_hl_r)K4=~d%NbTK>i!rL*D{pm@T#qWYcn1 zzA^XL!v=Q%aE;#r>3_S$E6RyI;tM%Z*TR<>T!4!Z@)`(;k~lErValGQ?M)J9;BMVlq=}W4t_50&Yu3P>iTSjks zKKZ+*j+#v<+$t26dA>zJq1)ff;84F`r@MneBj$}!mgRu(_HXKVXqS}$CCHl2V$?Or z`>)a17jI{8ZM4LBbq1c%mpn_ctw54VG_3qPwd%3Oq;@8@5=ich&gRT%Yz2hNe=WtK zu1rM2*kYd)YvWtwIt+@`NIyn?XkIP}QQxoBX5&~SesM(`P*_@Aw7+kJ$?hl?i;yiw zcj$L981$USphF7p0-!tEe~`buu>1f&H%!lmGqnW6aUS#H7;_snGSL5h7-upd8Mu4! z1lv+ZzAtuj^>X|BdHV{`&D{1OS?pI9mJHZgw(9ZyP2?cTzzgWP509|5ISy3VL7OX2t*lzN(%vpZTxF5@>KSkGmf=+n3(6Ja6NJE zqT3ywr+hjRHPe+*tLVsFE`4w;>@5Xi9-uG3{Oq@3OYTt+zBSPUHHD>r82ZJsiq`qPJ+Ce1hg zg;!ry7&WN2F^j@g{R*Z=9EJOIQ&efZzCK&my!bITQ_4CcG}O zd)@7MYLocNS3oVa+Z}GI>kqHp#Z>mQfS6Y;f!e(b+b7A@^(VR~Zn75@8bsPC4FM*j zv(b=5nYk?9VR|`SYflBz(cUR1J09 zpGI4_T^iUaf;#)jrp8HLiyy3Fl{0QkWko$mS=WRTQ^euAHD-NGvw*i;+8dd3n$(2@4!WPf?RQQ`6Ul#py(T!#V zh$s61PRomjWq-mS*i=G)QIxJW{PFm}>aP)sR%(GiCvTkA!c6e>>7=U3W@?RTny%Ha zp)Yg(*y;xn%FDQ%gO5cQyNKe0-A1=2(`4-}6Vz`@IbK)kld0D!UEKzz_E77llC zJ2{iS?DEGsWXa$gG||F+PRliwjC`@felux6yWZIoNLa;^u>cUE!;F2F1#0u}u6ig{ z6iFq1v|WMhw~2DjdK&A2-McC|lO%g2cMSQBQd!34RM*qFCy5OMU<9%kWHMJ`ND_2J zZox&6?Um&3-pWohL9DScoL-j@VcDfUvQ)`Z{}Z;7!uKAd6Q5t@&n;#O`P(NR&kYnR z*;9e?;89l_0^5d7y8&wBYuhFyg6tOYz41AKUKJ(7U5pIyu9kLmkM9Z;Wi`A&Zid9EL<0Ge6PN5jbkE$=SXI$OcUkPzaGA^quMFR>Z0nSBM%7@KQ3IPzmySR@XEnl_6lVgXnZHN#M9`Z{NUnc?l z89cv%%NJ9tS3mlRRYAYlK7+)Nagic&D1(&m-j(LtJxv0*zq<)e8XMT*1BZYV7-#dNk7eWT z2@~Ii^EAPhqU20RyW7_!X*3|cElG^tmdka3Nh*=Z8G^{~+w7x#rPCb?O9DVjqYG#-+h60c`6ydx9%PMVayd)|%UJu)(m zsqWV;k4WXrv?W}{rnNiMbR#14q0hyj=ybVp!@6)8b;-#ZM|QpRnzkT zGh(!x=TCU1y8HYZ(wTmS|6-xEjc$?Q{(Y%Jqg0 z5W$~09f!i^g%$4mb~d**WDG3IYx7qh^s5lrmk+=SwDxVbx99J@ytg3(fE?G3H7fVf zyn*s*c^86v-J~?wChm*4ltyTTInO5why8;O(Hti7~zW1J+mFo zHebEt8wf1i5#YEsdH3_%^0za7e(|RxN3qAhyIs{1_dnbq+5E_?CSg8QQ;LF1`UTC5 zk=fV5i+hzdFw3~hx^;*HkK4v>!%dCVQ3eTHL}Ojt_PnPqBsZ6JmmnT<)j&*9Ej%WZ z%A=C)?Xsdt4uq?J-atRvKZc&%=O!6o&^t!he;sE%LK09A(?6zbO3Em_fY)`}rPG!q zMUW_OG|BHqp)tu#MQja{JCPYb)R+EI8=%;XBVI>?|5f zPS~IaI{Ng&Y0<%6J-T1O_XjKovY+BbKoV4oeMk78Q14#|XJB@lKr~K~vwyYR_t-no zGY|bEJmnIpm5Q(n6|8(tqE1*cS^dGHi$l8!7rrOn61_|9Rq`N>Y4!H71%+PS!&&`? z4?|mfws?(3eL{9&#J7R^k><%Aou(|_2FjLZ8Wl7BoVu+&FZct_=f>Z~)* zf?QKc=FSxV2os57cPQh(du-x;856PRhx*_|9Pin z*|dfra6i@XH#@Jf>wSbso802;cu9$YpAiNYlkq0u@(^TKS2=)!;OE}5(s-L5*qP|s z{TOZHJUU!}Ou6yq&Vb@ny)zD&f0eFu$P1%t)<+E1RL>(8ky~);fo7Oef1p!nI|BXt8yZRch6r(IDYRkCu&Jx|D6&*o>f_qPqq<^U^q z1SM#p_P>UcS6fTfV_=4&a)fcf@yp zZ$n&qlkLhnw|T3eAH6O}_2FUfG$T)U%$IKOPAIm5pm?v_3%`Cby1l54EFYJrF!BDh zVN>@IEic4Vh~{EL$2W{LPv=)|UIl18-+ta)l^}He?SWNsRX(ez_GUjxp=BHz~8c-yWdxCM+_JEXT-c__C#$7Arty7m#F4vewb>BhM>G{TlPm;XCr65oAR;9dLuhZ zdYqmUX1X2>x-rc_1_?uvaDFz5)?iXObq-eE(OuiOv>teGkoBLqwr$3cmp?5kyKN2g03|2!$A)z{ai7(AKb>H#mhKc-d>mB6`A9HI-H(+f?m-G zKiQRZ%;M#AUiRl@+%hzN@iOUAUm`s6K7W=hH~E_{n(K9$DiQt-radu@gTur=bhr=^ zW#dJio?#$j6p4D=Cy+v|%l*s1!H)d?Rp{$?Cqwks;mf$&a8S-{;qpgt{?<)+2lio8 zzVUOPmy*kc9fs5?mjE93?Qi_^G^xPH#LCg(ZT0V;{D}dM?rV9lGw5?cF*p-~0K(o8 zSU&xY>pARd%gfh0^zNin)1QCG@9^ZJ#QUY1xsM)QOu6%~#Pd^b_q3pa0cv?z1}NkM zB|bjc+#MxZqkQv(uICuDk zC$O3}gFJrUA0%O4X%7C9B1I3)DqZ30W{i3x?p-k`w=Z*ZyOSE@B(%}r5<-Fo`Aw2i z(9xLi_r=_#F+Y7@laW!f6)G>Y0Ndq}?gI^b#z&^p+HDpoLv|w6;P&5Vo>or~U!ZTG z4W}+}Nxuy__bc?*uu3{W(b3(#N;99mWB#x?Txaba<&Kug3r>pIig1d?RL zD2v1J!a?Jh2g+#}@-koCJLDF&^Y+3muu%IKYoQ#qtM1Hw$JgH-B4hSlhT(EM%lHno zVZIw4zpHsX?G@kKRM!*WS2lt7D$phMdrJsaS*PGeMw{-ij3gnIcNPPzy)K*h!avFo zmuGye#B|SzHNvTPa%If})}g8TW{3@o8@;c_x1?0>P(nB5_pEt!-qG#_Fc)F_ee7oT zcl!YKd6@H2JYF921)wpU!}(CRhb<2kj=@R6NsbV>bDWShdo85ARjh~O;u`~ z7_I1VF9(lPT0j&ZE^=>~wU+Bd)Q06Q-J>FHZ>7%IzsBNY6~R{9CKE8Q$M5ThTklj- z^NamswL7o_``{`sZ{wLW?0f}U+RD5QZxEo@-v(sZ_8pZ{=Pzj`cWaflq_=HP+b^=- z3v!_XxzP#;e)k^+Q@Wes@#Q@a%H0{Y{-j=9 zfBg}>6{Si+7Al@LI_Az_8}|br7&W86p}HH!^^_}Tgqbd(^9y73&%>}e`h36DOMvh& z5Cs|JfZu5(?C>kRHI#vm@?%!jbtv%{VE4NV9dlN3Lo!-ly6W(X`zK#rzKeG!0>#>K z0_g}3qPO`IFOlLjTfwLCpDO&OaMR9EgFE|~P9L}Rw`xyqIh*zK`XGHqoSoEg>m?sA zg>&$g-6WGIiL;uSY=p#A{?uE`Zighf2822=&jnHi-Nhw%W?mLhx^ow#An_iAG(q`a zF-#{?FY87ODMi!#OsND!f@qmh{r0}mnea!l_=tdZC4L~Keqz^aP!(GLC!`JW&d_@W zO^FAs=tlt@PZi>t{=Ip3*e7Um>hLIQ@b=cZ*~GfB-q+7m;E3gj@epO5iMXD7wu#o%BF3LALsdk8z2t{)pwPh*u|OL)x|#@XUd%a?#*Dx8oG9(!76V zdoWC#6WXEZ^UNFmIswS=voRNJFJfYqXAi7+Bpqye7aN<`DM6yT`M~w193Jy`u|gjG z;8<pBv=A zwRq>2rrpq}W1|1Y71|cIRtMUp_3x}V=%?oi5|Oq|0oJ8HY!Nlt!rBvF@A`8eHueLT z%=6GPmtiPo1yzC`4~ZED*D^2<4?Ie@jU_)q2CG2iaJL^eJ0*{K1`5QQmmOr6SIc^d zfFM8<`dbA*L-39-*!ZdV(0Q9eAiAq{pj<$<&tvBJ)DQv;KW@AmUUa!{;R z)$kM!$v>j@+bzmj^;b8mvRICAN<@KO`fIvEHR_p8$`K-Y1&}vp*5y`%D076Q*W63@ zsoldjqIV$yBL@y273kB&qf8z9ICR!JQ^Di{L&)F^N2@VBK#lzg|tw(Oj3 zc7^`1B|@_YoNRs?^S9kij&i&h_K{7F!)Ck_c{1+;CZ?u0m~f1HwZC0?hcN|W7e@jI zMpU~Syanddpsp|QIa=6OrZ1x77@JV&OtvBC9`UT23El+isY>#Pj zc+WL2pO5V=hj6CGYdpD8bE;0)M*n2fNyL@nX(G7yT{aHTl^Izxl4-X50b&yMakO00 zVH!>xHGmB)p1=+lPn}A?rb}AOM=NNO$ExW$!raXduVTL&xMs95!d(UO_Ku8JMQK(h zQvt}3HcMci%x?g>4_H6E2T$z{YA=QLHR%r${{WidKp*pNh<4$oUQe(|8XS2PARr=C z%TCw9CwsxAc9!#=H#-YOeno=3(a}9+dH(7sz(6eP;3zs5SCGQ-M6>ez`4B z%>AOJ()^lk!o8*=AwH=20_#`5MfPF%u>FbpX!4ulp8#lfZWLc%=263a*_jd0PRP z_DwSr|GhstV4%Cwlf;~6Q z0A9(TrbFbfz)|BL-oCj2X(&m?3lWHZOV+FX`gFH04?!uYO2xezut*cNAy!YQB%Yt( z248|}yz~T@nd2|b64!L`b?hQFfWeAG5NHzyAJw}PU5j9h?18o`jYfA-0iSf>#Q5ODJC9SD4t*RRtOz|Z)=6|gP45YWutlS6yQv#}xMu!svmm39C3=TmEFx?>cdtD1^ zJ}HsYQ?HpneUID3_|y5~f$)ji05c!$rpwrIM}Uhh3hzox{G3%chC%#}{o^1x``6Mh zM}fXddGaO_Z4(VV^yxDg#jC5;NUW=daDG=;-UE&%)Yp!imd> znfl}(wv4VW^Nzd2YleaTb8i~{7#-`RcpBRTskuoT4)yx_;Z(Di+89B$=P^DkI}f$n zQsz;f5kAG{wH1d;&k9posOFX+*=_Xh?E`)dhg6l;&%WJq&vS0~VuEIo*loUMzU8C# zx;s-nsvp{kI>#I3*$dUqQfT(+;0RYzAD~nkkV4n=cLOx~j5a|;6jD^29FMWDtkW@G zbh>2e31?G)nt4)fE?Aj{s%bebB6g(fVh(#(6jl#Z*Es7KE?;KE3W4oKjaT3Nv26fk z(r2`h2`3Q^FB+PcDzsOi4URR|1;s7L;{R8k!tDl-hNZMl-wVGKrx-6ZMBw^Q0CGjn z7^hQaT?~}i3u^m~&nNELObn? zSgl&}H{Z}23ONssxGSDIHBg*=%b#*sQR^1BUdSA7fwVcdPv=7~?)dnYvCzj&us-=4 z|G0};smz8=eGs8j`uSLhugqr%nb$=&ZeD7wvrW#9Am?$?&204S>V*A>)Y}C) zs=sip1SJr1qZ??=?h^gH{&rz;IzD`bl&&+?D|~54>0B$k=PtZebpo`}B9U#|>-GA+ z+Sxn~z3gg3{XRdiwN0?NL2ftbA)KQfWi7?EgpgPOm>wwSnS_yd^xrz?&WLY^jb_jSnC~3)>C(S@xNy zUo(+EPw+o_qF1Ixe;TW$qyOwziPNDCEt1O?6k6^$V!^Sl?f$zQ*E9Eco_GT3+Y-T3 z#7-uD%SfZ?Z3}GAAKm$=H^!X^r!!c2&c0hANL6+FlwV{<=9?TK-3p`=Lz~s8_km>o zHQI|(U~?qadozY{aJupAiyy^9mi_&MGJ2{*DDR*oVZ4rmz|0t{JPy(^FfAIPWEGU% zEi(Tv-5#)EM3KRU;btxOf!b-o3A8g%G;v9pZ|w~S{&|0>7aqO<%oW!Cdisf-5<7^^ zp6Gak7tqBfz!6cX6NXKO@zt?1*`gGTcEH*lb2F@DyRXz!1tfLSK87PWvfi-3OI_dS zOGwyv;_F>vxeOrj%!P1qucsZ2xJ6ADR-RaUAg#a|^zh zcp;BD#iW+sUWx!^eC9r(2=6qH<)evTKUh4BMqg${&y4#bkkTgVTn0L8pO)AupFNtq13 z)8cOqDVmTMtvB!kK0vu~AoRJvipZZKI9jjx^=vj50&^c{?@0$1-BFqm8=hAg0|8LbbaWLZ=4=(+!UhQ3QS{Pw|sFugw=-E z8Y)TNQPE6|+p_yh)dEDazD})YQ<6!#=sylzwZrld>SgZpsLm&%R{Vk-$)fx4Vu!fM zLRQ~|gqKy5k>ckbm#o~P(;sXBlQyY%eo62295yjZyBqr!j;TjejCD4rPvgB5Y`5_Y zZ!3Jfr-i~K`vO7F#!2Y%GQh9=bE)HbVh?K+VTGpCpws7jk&kL={rYMJ)?8EQa!}wp zas&A<>{#{77>&c4zAz6b24;*|OyA#+bOy=cl>rYD+f_mDu*YLkpR~Ortr!lF!C2aA zpPMI3t5|`I)jBf$LjU*nhtW{ppgmmjAS*uzJ3)+voj6*N{$P>%S8pn_J`Sij5Y{4@$ z58E$C0a55LS(*@=%`!=e}Vco}Zo?92?fTnnXiaE3LZaZKLddf3A8;EQD(9Wzh{3gA+<1-D+Q~#o> zm$wkFz13b8HoKP)h9we%W665NEWkOuj>55Wc6*#u`8y#b8GL2K_fOT+Zc(5ds8)NR z8{!Z_cT}h1qg01;Si@k}&BXJK->NtjGGfsW5&E}mwoke%`WJB=+;wJ$dFGHFx0%Ua zApHseeyMBtJNM&OOFIUk&W)6qee)KJoot>nZEaJ_PKqnWJF9$NO?~&gL?D@7uq)ah znPqrzH^5BZGm( zfY`Tuv~`)A=@P&pZ#6bT8t22+)w3wy%Ry<0nESgI09+WNczzDD@x%$q)YB=xw4%eH z0A5f@rB)kdffBVm3_T|Uo|uh3Q^wO{rC)Z~2FukxOAvLpA3%?e-o*roPN$L&4O9d8$&g zuI(Ug@_&< zE0Jp1l^T?SWU+`5M90Q2`W)0)DO~_MUb#^ZLG_7`SSu2a3a!TX+PvF*#=#u(srimf z(2O39?4HB{CD!6lNE>@=#=*<6nqYrVU?`6;P6*YmedVCFUiluFjwnH5fqlUPwYk@t z^La|+DR_3#cXob4j&J7QYZ3*&49?RW<9PUCz;nN$x(K|~QIRb)A$JTDK4Ja@x-Tl; z;5{ZSGXFhxyom0+ESSFUhwcRn3$j>2ceC=Haqr7JcN(u(6fJ6CT@Eu=?z8(8Wg2vv zkL%{&;wN<EUsFDe|0ap!$hgq)zRFJCbb)*JaBJ8KdD=ut6|6-?4dxt90}&v+|L|X>FZ0|N!0gTKthcX4ps=~mjDOCDA(@( zcLQRjp!$;l0v1#Emle4!2Zo0nAHFMnyUfA0y7a*Q(%18i-vgyte%fO8dR2Jzv2?8a zDdF(ep`RUmxz({s)H8kE@dQDxJFGyxYll4tpi#qu2Y!eP?&%pHwrN zDw7$jr!9~vu2$au?N|MeUK7`uHLM%^dX-o0opbF1 z{8Bu~L3ObP{S#%=&R5J`$OyoCJ^0Tfj=B{&m&Q?}5{U2i77L&ERbD?oae2sUrQ&1G zpKZ9r(BnE{jl~TK$`nTeGRvx}9=8q>!RjqQx_O^*ByYfPc)rZZpz&%F)iO}p~~~u8DIXB#AED@1}%gM$Kx3sUNWlTFnETGepXZY9;iP&_%(GoK3ZUlb5H~S zZY?MW;GdgFeSJclbb}T!WXLL>%)vgrh-^`X;m+D_0(5CHjr-zw_l;#;RbP|zD*I|D zF}0)rHC%i#FINwVpkj@+Z-2UVU@q159;#Zq3ofqzqYw0@!+7~tU%74s{FKq|n4a9R zQyj|+xZq%VVl?zVGb^SUdXP|wF?jv%{j`uUtyj(nD5MVwv@T4{{qL^$fqu`W`^MW7 zjKvkKp`Wtoy&Ch?>Dgo$-7<-4Ts4Q#5|%ZWHRf+6LPTd~f5_-sZ=+nd`2yv9Lli zN^elW7NhZrg8zf5)eMPB@B%8t2%Fzq4$E1Xl6jcu!v#>dEO&~nTLMU>#@+WtXQ*`) zyKgZbYwCGCl-14np9R4aU$X@mf@fZLzN+u*WdT3*vb{Qs4~iZzVHw=AGV|C$K&8dE z5~-3zOIZyj*RS|Wiq0=S$e$5ptu}Ps3yOx8g$;WZFg%zR->m0&DC|!m?bRM~rw_5Y za+^zz$5Q(BJzLi&CEWN3oQ5wxH{g@5zQN9x$vvH&XZg&Q*u?r@b(NTdw$NvfPzhzj zdQXP8t(~yS+uaQR4Bpw26v5`@y{M2ar@cZzfo`(xjk*Bi@4YSVfx({*uJ~;kw64{6 z(dj0>v#n!e8_AsP-dTg;9KH>J^k4Kz(vBajqYw#IrhYwLsREriSv<0RVuW#XYQ&HJ zatvxzh8_LO2FzaiL4jKHbhFzE}2Rfea3q-_^DKXb! zlEB3XENlY&4JrjC@2kBFp_8qJ$A_?8H2e1X3eWGIm7VhVi68oL_q1tsbZUiQq_-vk zvilxhsF;a5Ca5SI2d@u&$^PSZzZqP~JWI*ZE_kpXgW+|?DLOmrZ-W|swLX~HUX%2% zc*z(~46W#O_0Sw4cG^h%o!vWP8KZq ze2xU#!U#nLarX&V1y}jcauj>EE?IEDT+5${8)>CPXm5oVu)VWf^VFra_rrPJl%G|4 z@jNl=N2^CxG=G2rtnr((9B9j1q{-99CN9APsUnuY7=dUSm?GJGhvlFN?o8Y;_ zGb|^F(=!mm(@^1mdpDAp(M7B?u$?htVLyCOW&M2;gTY{!?In~SBN>(_*I2IasixI! zKg19vxN>jekg9e@StF-_hTpO0fPXm}G_{>zhkIP*2k}u+AUS%UO|wxvlu5jUqaeQh zpkdV7WN8AeA9>@ppw*+k07ul2AWBHsw{UytA5F5|4W3W~lzo81XH)@PAE zif^IqK^Bf6QPF5R5CRk31~w=t&;E>Cy@zM61NDYrqG34eBj$?_68Y7X<{S=gS65%J zX}%ZV!7y(SI5W5+!y%x*e%DK@j9~j=0>Vv5;2w4D+VwIj<6qO zYPVjrQ|9mSB_-$>H?6~^Z-UZMI5FVDv=F64e&F16+yDwdxqq?i7Hc@mG8oIWU$(C0 zAA6CC!-jmVzpcxwiJNUnux4o^UA~WPD3zD45(eZTPJNb9TRPiH2$_=G3`nNUANZ@4 z9x3+K8zqw-%{_c`G~Utx#ZZqYWf6QI22$83LflU$xtldhRZJI?b$17i41I3U;TuuQ zJi75OQt{b_nBn@Va8(m94Z2ID2XbM{!af|5dunOY9RO2AIte&Zw{=igWtO}!s6oE- z5awU}X^z*+Mhjw~^K9N$;O4`2Q`i|BnvweKb(%JHbHe?#COU})&;|cGuWn?(Kz}KO z03U498nOfAR6U-~grjfzI;}zB=sAy)IaQOaqB>vd=}rZY=lv_I9`8W6Gzs)UybQ~N zH2iB<5)6nEVC)zDRb*IN0yB$op6@Ab*u!TZP`-5WCq0U4%@*ahthsf0{A%BUS3O>Y zqD%lcbm3d!ZwBaG-L1a{Y}1|_qKgcMlQMLuiRL?>@6)-uLXyQ@Iz7G}oTHY*(M>>p zM&8hK7|#?2>avI83A^fWca|AzRBc|Mpk)-!~dRxcAKe9aD_=sxB zcVHulTzrCVqwQ9d4jVl9HZM%CmrS!tR%f4k{rl zWnX{gI>2QlY8E88keb>IiYBciVM#LZik4jo> zYJNYD4~Fpt2`*Xxg9`UY!1CvBbGa|j^}<)Pc-)bJvGP;-8XMN$g~Osdxhl39&~_)- zR?c+(>>l?_yc_YleQRR5)3--^cZx3c+I{Xj|Cii%iTLy#+mQmU>sed&&R_2`Z_z(b zi2PBdC+EF>C8$>(W+Xj>W-ea6Jlq_lV;C)bX7>4cwX57-rT16hQ!Y|3h?mZu@>Nj% zW;#zfmGprNar!xG$1%BG1QIL2c#z*J`|BDC?Wk5il%~G!C=U+iV2Qx?JcmMb0TwH4 zcrZ{?aq+)Z8Xq2Ivc+C;gV;{9eg%4K)X$K=}%MVXQpbL{|6E zGG$0C9g;U>_vQzi=WzoofsU{7A26BQgxwNJswq@DKJK?a<1ylG@ZRP1frM)&|HR(C zva_y+Ybqv>hHyDr^eV<1VeJ|+d3;N3KrDX`?Er#*Q04(9)lS>6IYrSis3LNFCmahs*PU$xPYDXLA@4YJ*7}_oqXj%6Gagzg!<(x*a)_I?Q zrhXqPH0e?_Z#li-ih4?7qRP84$vXz;C3cSnCfCo&W%b%?FPFo3aHJ3AIJ3a!Sua0hO6dv_-`>X$@m}w zV=iFEdp2(=gmUrMR7lm~=g8YfL5tT!%C?1`t<2@aDj*TZ7AXfRLG0PHhbjd+fYsLB zv23HOM+!vg@V<9o``I3g^gA8z$ z46kX*=;hH<;o> zy?KBE@X03~HRUfy8e|V$+?^w?$^HJpfB=4&o(_hZ@Ic~hebK}3^3?4~xHLnT=iY-s z1sJ!$OT+xC$7Tag!~F=%sO1fY}FRtf`Fkg1! z+&Hb-o~|MmTzSZY+Ujs@!bD}(wDeC;=2OJ zVEPM9a(%QzH_YUoTyJ-eU!&$i(D%C46Ub9$+~F_PqJ9E>t;~8r6CkiM8IG+&#N(!3I(sc4sSif}Qi=`56ILPmZ*@jQD@XpN_>_hd`~1 zQG0y4o{o3D)Iln^wDVa|%N;F>na7Wukipra0paMf&gZ3_| z=hv9m^>u&AKfv)kS3SVv^u<2Q&K#Bev%`>jy}iTkeQ6=^V8_q>_F+g*2fer0!(0@3u1(E7nSuXM+*a5Y0I|_*q;HfRC$ie*b+m&b2Ysa53l;gJ#`W3%8+VXT^aeUo9adms= zNd!D;_vwiNR`MJ2K=tHwYt-xDY}oVp5jWX({LXJxAu}Ajab-dsMdOINC%5F)?KZt= z9B**{E!9b@;EhST>-ntY`JDWMO?`{n(pLEASB|gpQylvj#vuiLSz}wcavoroSO{Iu zO`cR2ht|+uHuARLJE&#cbKDNNwaIVbs3xsjGP@ckizM`LHojjo@m)870+ro$JuMOz zPS;#nX*JmPCkU_joZMv)rAYaKzY8yRaai)@CFKt>BS_ zx;#&E(QeG&^(&rE1a4ZUGL0^Nrw0)5Ld(#P=YW&I2SU;rPb7 zx0~Gp30I$5IvW<)gfvUJ3%FwyewT^mWhal#YY z?i02m{NGD~>sNZ0n@t|(A6@z+Unr0CZOmo>pJz2#v&un2FkOxhg8lV>`W*YNlPvtG z=jtpB8l&BP)Zf1RveUbYp7h-+F6#jMxUC?_W{gG zsnD7!H^^bz8Tq)M7eDt;4!zSqu)aq(?F|#Uh*;rqQ>{T0J%5um`Zou1XW2yYn|V-X zW})$&im++5MuE9pAqM7Q;wF6MtIg(-Z|ij*ZZ8w+GJ^=|yq54#-+G5|BL#767=To1 zYSP{X(oQs>d6e>!sUGZs^5t4-^?nS_%K0el$`=X$#ba^%+B4*rV7{%7(%#4mmoJZ3 zsP2xAb^g@rNxOmTh+>cu#k3`K53FGyJ)AFk?4L!_7|-?_%FO$dOQ-m4K9b(H?Ux1TBjAzt z;@?$^inn0KTWtDANAvFc9Itg&-m}x?d`4{d-0lQk37J=^O}g;+kpwFBh>zoiFg#2? zO^CB}zA)2jrToWbz~i2$FFYBqt^eyt%UE6FwzJI+2>nm2G`TLO`Sl*XrO(rKITTy! z?~ac=SD*8%{ytQg1zv&AF*k15yjJcBQRM!hCYp8Z9Xj+TXTSfV0vQ~odRyf0(W4dS zScS09-(JI3YYHPe#hwRn%x4L8*=+E zf*kQRhy2gDP^^r*ouHjm2-?n@D+n1RC4Ard)IG|?|dHpay*{rA!w9HIal;rQ4Hgk;v^E&+O(tcRz4T zTp~txmar8*V-sNwdFf?6%ETqE& z$>|H$F0W}k?MqWwU$AXagsasg05M-zqvv-pL3d|>$}Av531u(>DB<_@(+>x#ww}R3 z|APl3ngndbfq2!OulqOfZkLt+HF{K&*BKtTF*Huexk*`1sMYPqrbh~jEi7{-N3=5t2D7N;M4E4f9bR*6h3+Z_i}!_vs%M+d2ZXczj3veXzmK7} zE%<=N;jzS+_Tmm04xx$U$HcxI?mygjH9imDEuvRuN8(NoRNbVjCs@E>=xnmjSNj9c zEkL}FhOK^Q%{Y$F1PAN7-5nn=A1O@jQU=mq4LW~!nGZ<1%WQTh%@^8AbF%;&@_{CC z*IjqVNqFC7>>S5OF3m^uKa6gRkO0CPK2iK;7swNF zGP;%GOm7tl|LXjFsjPsIL*B`um zdVG#Cwf~_Q;~QTzBRNkXV;WQgNqJ=WO68bK@~z{5(EbGpHpR{=gqj3hE4H#z-a5w- zf~lZAvOA#F@g5O^(CWf0?Lt$4=b}?k@5eW)mZl=!f}Rf#&wbJ2x6*R&y8AYtWx51E zoRC>uc9eH{fbq0DtcGgSg<+~vP;dhIc{8zYwF4E4SPj86z~Ai;^t_4kI;9JV^C*19 zsQH*$>X(|@!XwMP`kS&DJF9%dAvRtVbG89wq~LTVkX}4sd!Q-+8cHbx{}7g-TGFWm zLmfGh03hMOHufgN>B1i)Yx3IiyK(G4BNWrEZU-c5-Z}V1><0IEKrbSCU(>U7FRLay zUPz8;BrJ=_%eljoo;y$LZ0Q@8eF*Bd$Pi!%tTy_R-my9M`-_+Vl9(st4sQYX{hboI z&J(;wii;@k#&qyBdB>N){!lNse`2`QAzfM2IgbA8kwNwVjL38QvNzsrV40^PcHw}O zf5y!C*fAgZ?ctw*O zEtduKuk+t8N|Qw@|M{*^q+tm zFGC`f-0ZqK`=Y8t|K$P{!aPn={xDkQDh0Xh{-Bfu8g~yQ^ExY$l@Akf?SOnY!Jb|HVc;npm_;ETS z<~vJ-Fhy;>J^&jjeARmK_6be)RCRBTXhM^f1VIcR3%(=F@qR(MH}8@+Yzq9myIs|j zOj#W48pT)id>M<@Loq*qefmj=U&@qA_uTngD!Ud#V%EaP&zvlSA3jO&?m(8j{C#zb zZ|CX^pwVl5U{3wQfw?wewI?xE^uaLL_eYWWm`*DFp?sqdJKb@E{Xv!`)W;j*c0|bu z)%{)4cdQn91a}R7S?}S{j_j@g{wo<+u|lXY=&EH4Pt&Tsz38bQ#MCp{;WT*YtOoDq zB!KdPG`(&<&#&lyz3&vn!)*G#%z!2gut@rYs{3FzQJ3OwBdH2XxvD$gTE(Obc2AQh zM|sFT#Oud|*`<&}`gJD^qx4!4n=ik`Y%|keroDkVIr!k4fNVsoic14_yAK{&R2(1ii< zVH*FLTZMzQoa5i0JEcZ}Uaho~M_%dbu23dp--F(c9V})co{6#0h=GqS%kBD8sqL!5 zAmQy>p9{%EJ&z@DH`*vhZtrk+Vu*x33&;I6e67}fJwI3t7D-~CBd3iJ-ZemJ5T zed-8D`f%0p@6s{zmJmoXwF2o8z7xfX%~*Tkp=E)vI!_N|7e4dw8x#jQxbsE3?r_a@ zyPV$QQp z8_z-EcO&BoM6S4bI;w)}qF45wY=tEe>KzCfK~`q^*ze!RD`aOi!ZVvp9Tv$-4I(@v zAGr695Jc_sh3xOEg*OH=7dp7W;fHK!@Lk-&@}okj9`*V5tvlv^(>$}U@XiAR_v;Y| zGG3{T65Y4NO=@kdi)01xk&o9@ZE@|c_)P7L%j0)9!R2YtDZ?LQA*}BEM)&KXkUeU5 zrQ?2zqgF&$Uzr^z;#m`*1bKuZ5HMii1mz6Kxag@{qjXhweCYG?aS* zuIzTdJPt-*4bj1`hUW_;+V#~Tte~NT_M@Q=W+qjnFEB6Qgk|&z^Kypa)HxPVYd0RN zUBPG{!N{2bqNkwVJrQ=y#U)&?s5_bJd2U}<75L7hX;xB4Co^(kUHLx1rg++R+rR>y z%hlEPHJ~F;R+iB}!8X7>VfV~`QVGG*($`cdPa%-=FC0>!?g%h1iOu{BDB?4{Hd}N( z$^DG*TWSvv>dCH2H3LXBe7;wjArdpBFt0cYcz@9R=V{})flYcC(Y)tPZ-qJufXb>p z4_RE8zvwLlgB*SPfIk;vn0g63? z5eqVZ_{&WLU~;mrua(SjgSj8BLF!yNK28_Z5fQ4h>*(j) z2JeE{-$eWPuyK;{C z=Yx+`Z!*EEufjL>o>nkMITAf-NV1501AI349SvI`Cf((Gi=h8+^HxH$S^mp(K3Tk7 zKRXK;@NdY;Z|izyDO@}da>ADD`wYtbL*BkT=XGk=JE3KO8l*Z`yrPWj4H9U+PD1oG5n4OgZJ&jxtFwBvP-eQ?4q^6 z*v1&&%iU%Rez6fs^l71B!16c(OsK)D?(fZy;*cQ1-4@2!5cW=o>rD~1E+De!RX;nt zb6Jo{I@^0B-`-sY-d`)gx;;0Pjp&u1wzPZ8eyWz6(R-8?+fTb)js{+3}nyZzp+Z!2Lq-u85U#N#LJUg{as^+1hfSG zkYZ=ba9}OOj>+=R^Y-ZctfRyFLO0^?h(hSIV)_VT7zf{I(eT}S$q~`}ewLdsd`5|W zxMP1(FfCaIbc0K>zX^j!?2~K=4Ex@xF1Ztl5e8E$ z5R&O23tzK@{fz$B-f_KsmGrkwGO3JeyQ20<&7q>&lUh)?l7BdiU)?hUVUf3osp5yi zC!>e9X6&*tuU1QtufOCCS~@NV{p$`_PJrhIqB`RDa=hRog6%B`dRqq#{QdnX{FP6vEASeZ?2wPhF!^Oh=$UYaTa0H2 z3G-i^YK3lRzHcapO-Mx<$C#Z48zcXR2~gjyqn<>EQa|$|f^OZ!&Klu~sF#f*L2I?l zy!y@Du}8`UelQ^geoqH|5R${qQO3~Wa~8fvJ@z?!NysKt8=t&ObqjLflQgUGN1&Yb z-!nPia4=_nCwQiFzndp=p()XcUTy2- z<*g3+INirM4)IQd#f1zPfM;6u%gZJk`8`x`Kk{2o;ZT!nahD~a zfw}WFIooHSkHZDf&jy`*K~|i>@PfZJ2y%FoYY3jml<=63yR8aoH0X2TvMGGuDgAIt z?)G@>KT*jGJ&8}AFFNKMxDBp9oXRtFr1@q6?~^KnYy1TN6TofWB^ju7-Ung$(6Zi* zKytcW-+)OS9inp=xwp>^Wf$Wt6#GMLAg(R@v0sr-m^iOBbGhJ#D&v7vb8bY+Mpu8H zH6wX~Q6z8wgW<$f9gf%~{$+!dLb0D8%MRAWT7H&9#Ip;XW&=Wy+H@u56 z9fun0X1RQ!4Bs#ST*@qIca`hkW}7zKgUhLAGV6My_MSo6HyhkSN~hRY!84C ziU5|cA9IUi_`qs(@)WldR3W$$oxs0GaG~|z-rv+BXl#i{S`vfCJlZ|wdTXB}yFJv0 z9g2Q)c6#yc;NG-6)!Y*2!#_&y6}R!5Br0J%-gkP7z69`(mAMoJ04?N2$ei|&O8_$Mz`Z@T7)eejrL}B1QJKf|9+2utATS7N0mp39?pEj~s!bHl6 z#tXMIziyM7L?^^d<;W4th5$eI`>WMMuASF&;-Yr)1P++Go-Z(~Iaf+mVpP=&o5A(_J}&t>L1{zv=UMPSU`sOKZdCIf_^R|F_+U9q+HIc&NC zpGNf!F_C)emJ9l=CfbS60S?{mz}skifYoo(lf^9jG2wP#zDN%jf-SR!T!J4!?TuJb{x z5%G;-t>ow5>=@Yc_AQ;bGEm%0^z!jXD#fLR(_!h6~RbC=Nph!A7%D&voy6Xw4Q z_kL%Dm%q#Vt(qOec%IBw z-E}kO@Xi!BZ20Qg4S_@qz#s$SG1nksxaAJz@>aZ^(nItZz4+cNDVpe2WBq9Ph6v`M zQ|aw;b}RiWy-Db+lgmW}|2C?zdbZzqIdGAp7z%JgsS7V9yPiwJ%f9^k^0zeZ*}d=m z9f<5{gT*!u&Z8U0s#XYXkLv*Zwyh$`1wdn2TS-em_Ed29<`;nHTq0>F4_;zI- z`ZSm*gBlSZWVa^j7FK#ECIafvT~a6&GI)Dj_p1t^ejDC}r+h7M=}3^+BEMpPw<6_1 zX4o&p{dh~J-JRedW_kZDH{`Ide26wzQ9EzE{>CudaUEIYPbPLoXzWF?n~kgfe@C%L zI{Di#l4?R4QfF4&rZkQZ5NM5}WSpodp-jVNiUdFE9 zIlE1B*@@?9^KAM4YBmoIcDk-o+NHuUKe!!~9J{1XE|!s>G5CGssLX zq=u^X_C>PnKr-i^F;aTy7wW$!XqF@!%1Ry-LCp358;-_JPwgXf+x$kNv@&CkBxHDE z5eKV~qEJNt+agVE9oU0Jrfl$>>;UhWP%e@6l<=2~(=cx!l;(k>Am& zdgS0gaPQXr^alC*cn-OIK9sH6l6ee=bt%^O%|$5s100zXfCkUuEG$o>-J1$h5?&l= z*lw9Oyl z=DBn1R~262=OyiE4$$lV0=zmr*}GqFZ&#tbDBMPBgcDU8FC&R9>6P z$bGR~246eaIsX+9t60{-Q^3<^n{f|)Zj(NO8kkBPg z{+Zfv#FJV9o4#jmMXKv~p@g;(XFUGSFvWU9St>ucwo3gCxySV+NU^D(H6NdANy%(n zB}ILO{zg)@3@@Z`^I7OQo?!TW@!Fq{gfvAyZo_@sDOAfm-%tA;VXKhYaPzvwtoN{{ z-ViR4E_OE;C6Nxw^^O^+a-2SoVIlxfh;0 z-(M}q6;}K}-~O+_cC!|YVGi-Wqo(`wq%dPm%I<_k1E5nH=6?z*oiG!E+gGANhV~mE zp-yRI%%32zE*3Uv@-O9bYLw~+l|bR($GT$F5%QJf?-N&u8kp+UwnJ*Tf!`$@4@Cp^ z@?lcWK54-Ovg6Kb=VGaVukBrmzvj*5X=JEl;1?zKYZ3p!en@S*HPbCBo7-vL$4&I^ zI5yM{r;qya-pt>x^DFm!h^GD=))E4vQiHlW!o6pNxFK zng!>s_G(gG+gNA;Kf1YwSe1I8J#JGMtRw4E%_Mp$^>VxsJSXYAHTe{-u8X0*7Tw@DJ8><3#W|B1fg-N zRQmBlS~mYcmU1(8yWXn*klYbHyopcs3RuUsxDg2W_A2och;Xq_=awxjjRh zx(Z}rMQZFHZsKvv^GiNyyB@R`<7rLQqq>eJaU+VZ#!x6{-&BtKpmrr|U6O4$>Y9 z*~X;r;9i>xAQSAfdB6k|pI^se<__2UA1~#(Ug4i+bR4yquk3$q7~;8{=?Isgn-*&x zOl7^>B5k_)^cQR_0MXE4a+N?-3LRkEuD=@QYgjb}N?Jd3@@J>m{vD^)fzC+U~C zBfIu`!ljc2CVn^GR8Hz*Tk;F{7y8d9z%O=-S$lW*Sdu9pGg9TF<(8SGw;WU)DX+Ud zd3>F$wQE1x*;=_f#x8BO!}7gSX!9k)lFr^N@olgudjxuR_m0lO>$J9jCC^_$i+m;G z;W0(~>W}2!J^-ODrtyCKG1Ut(+do%V3k8D#2ioZ>7PLuaSaEk9h1bBPV&hUJ$L@^l zN6qPu4>tQI%M!7bTTa#u{ug-cpE8~t!}_dD{U=JMN>ZdZdq!}#6RPwU(Q*5CFV zc+t`09UcA@Er*|M7T99B7ZC48t-!nx`uKo6$lwUqsPm6Jy7`%uUO~6S1w;620)Ybk z_qsady~K=XJhsQtS&{)@PJkpMy&Hry_8WgmvNtb!%lu&;w@IZD9~FcxI~bhU@8gdx zD>u>j{WrB{K|a`iwdTzrgLYEEFu@s6T8>mqa25~FfGL1n5wIgQDKG|myG^b5vDe`2 zIO*s+WBFkF`RCAKgp1QdPy}cSWq91I+Z5>ladxs@H~MxI_L0?C5ve!gy!O(eb(8Dm`0Xs6A!0oyKpb*c$p zQl}oDv>%iwACh(~R=Eok#W~4o33ZM5J&4)5?6+*x?6m!a=smZoIT28g>4V(P z*_uaaa!0BWi6AJ@@|C@wQO8kV;VG|e857b5MYG7K zAGlLlLv@w0+B_Mj7($2L{z4{CBq8V&BAegXV}e0^E*_x{35J)-l=Cc6k$Ts5D{I@_ z#r?j7ii$>M=D+u1i7lE$nDczjAMyKQI9gvOtp_%5vF|rMCKuJt?QiSyXGR#By@6zJ z1`%8(4W?k<@d&?^g!O91xB!>F`FZe~x$FZXze>aLwE5b}Sd=a+)UmfwI6*gMF}@Vn z8t{ORJhEB*K&{jb5hsE|y3x(c;$P^+K<=j?H%QS5VfcAu)D=@*9%eu2m+rF}sqFoj z02>+X^=I20(eR(A7A@Rg?qpQ_79%_rgCqv$GRnN-V84#BP~d2&2EC5FbfnCY=aaAD zVeQu!t9`xZd25$4E#_?9To+T{xQ%Vm|2uYJ>zk?0H&rQN#~lmd`6~ar)-{Z{gO=^% zJY3Xx9H#g^wR6kN1H|=zC9;kgEx~2@&u(1JN!IJ?E@En}0oC+>1ZG&k^y=#1-#6kA42NaR`CAI026^VyJleD+aT)g9=iA6_Bq1J# zLby$DsRneM7^XsTHs9Z7vt6;U-oI_s?8vPq7oSq$SRU;oMBhky5!QBkBYXTC;GPky z<-==(5)V0q06C{QZ!elFyJJe~Q*#5hQ zF8KR@mW(M%S7;lx?KeW#Pv^Vmh8=*gN1j|FIr{MD^NSk)N4@r%zVERW(?9!K!$@W+ zGtJ~o0>#k>PDDrTO=QD4s|X)l2Z5Pj7}^iVgths68H*<%qMpx41}+ab{lXBgvdHxg znS}k$8Af=+Kb5fcx_m<=1o>>7>d|%9K+T5XGXni%tXJZwnXQP<=$ds|Eg+hYI}?SI znyiDoKWT(&I6_G3C}ObSuYhj5BZ5j%er;AveE09ObH9-%vY;U6)28a=)P+!ddi z7$6Wy0eIMZ)9)rUC2{}OzunjwO5gHh@}osjdaM^%FDewcE^VSkb=x8#i^mM=Z^TR?%=N)@KT_CPlqkPmgU25wwopW z#%w|}Zc3O>@+Fbey#931Lv44|OD@?yOJ4|I+7T6u*ZvD-(UnbhF}j6Cb7K-3Lv%Fp zPvY|l8BIVA2)$}Y8t?9Rzp+{v;=lwBPS}2QfL6_8fsWAnS9?=7_-yv;c#aEO>_{9`caVVH7kM4u>iS`r>Q)o5wbaZ_ zH}4PYCz|lp=l)|)zcqLsEV{t~@y%Uxj@yZ*oO#?_tDo=e9?v`bHYThLG|+k*~H^T0&3nurEjuoX_9G>#h~J#M<4IcRkV~kSIh2`@7#F5lx3L zD{YY#+4kFcGlEb+f#q_E@g-9=n`7R*58M~UJ?NvE&wUTIXO1tCd4mXuhxpMgm+P## zx);>{6bv_$cq5&kKIIK2s^n|T^PqC=gTqx_@lq=g&}z(02q+9;ewRz@y3M_!A22A) zK6Wza0fI@A4Q7RL62bt6fA-Rbo{UU3z_)sYfqv2)3f4->naxYDcbj+KzdGFkn*tqw z!?DPbfbC%V^n(csskvf~9-O6;>->2NDwd^_5Pb8r9ydSIKZ$oSxIU9J$FPP$yO>5_ zNrO7-7>~haT>8nFU&ycLy(u;vjxX>c52S6yguc^wS#6TT-$4+B?0BJvL+>hb zM;@~ta>K3?SlsG!Yj*f@I=SS*!-DQnQqo@w|CI zHNv{=J$LV)8zY`6V7pYr3#vuvdDaVEpfq#j2f~-%9*VOYLYTO}d69>pH}_t?9w(Bl z>T<3ikq6NPgTeXT9byNAt@gNEba63&n^4d%xLd%CQYSN&38233XkxZqfAZY}f^|}9 zSPbIY9^-EJQPp@>DE9t=1wnb1vNs(<%a|n=A-`}9p*}ICt33I4j!ciSw)-nA8{x;* z{ET>ZhaX!VJ~GmgSdZ!CA_ENj-RBL|FB88k058wk;r$YDU;5C?YX&Sz!9W<0y&cEv zJL>ZIOK*1C`obOM?{tco;EE4_x&Nk**%rQObJaZ@ATGvs#3+srO1pCjumLGh#CJ3N zuMgC=1ysC+t!g~BhQ8r~>OrT}lv3Lq4hY`czGRT!)qb4*!Gwl5{6dDyX~UjjX50_B z;eN+bJP`0auKo4jm(5i>qZNBMbm_H=_cszt(9pyA;|thFfYs3VR^PiWcYkoK;i=|0 z_rZqJw>rEt<@>pZBPb%@Zvbk?vtf`r-xtpWt50}j-tO3D;@_bJ8TLwY4WoNaRdU~Y zo>?SE*MfB3cQ1b6=*1JWAcW0PqhigAB|M}JIa=8nG}Jgl)z{NsKg)p$X+MT_b;jSM zeCP3Z5n1*8EPyg_ywjJunyNu3@lwG#DJG#(!y_4-7{|m#+rB?!`pu)ET22x?8IyVB z?Ty2O*C6BUcY3n@TGh((cU?Xy7$9;-r^$Jfum&S->Xqz434DzCPzpBmy7KK$*&$!n;7%T z$g_4A$i5Tw{N7Zt+442I-_wS;Jgcb;4ytnA%t(|Uf!sKp(=FTqmvy@;>oQ;1+jhhF z*m|z)H#a_MIzOJ>=87>nd_0XXOCxcNa(Z%g{nN%)c+PSP+9 z)q4RMs%?4!%|5`3*+oO&))C(7zJ7x39PbPMV6lc7CzZj*C)9s-z)(os0Cv?*l$2J~em z5&A)iLK#til`ZZlAQt26&4S{`+SA$T0=Goo!l4p+Z#I|(k;5gLm?+F^N!iQ1lMU_QvJ7Z_YLz5LQ(WRI4GIoXmKy^?BRJh23ZWM7218s5eE1z z7w8zQ3TQAtFTfWkv<#0EhDc)zG$XtkJGUgsT}UfMd=Z#i_a?jm+n62T_!opvXcwo6 zhC6)B7w~Wc1=0Kwv;4#M=}(6pcVoSv_&`*<^3*_Qu|PESyQ3oM&0OE?R%6*h!<2j0 z0j&O&?8JG+8<`C%gtsK7nIh8V0e?P!* z?N=>c@d@Ve3#&*r4g*HdJ8i0TeRxS9C%;5C;*3E{jJrSkkYePZAQA+As9jPJ=vOJ>9JMEul z`94wrpDoT>F)?SFw>3bL6g3$&L-EzAp4$Y6*^fka>Ih@jWh0=zq!&ob0})u*Rb z0-*>;SgAFdPSWjVem}#%p*AL#vN&OYXgch?{h>qZN1ymV(#plO#LFd081Pi>6(d~T)S*{KbAr}B67t#DDVa$4Lp9>aLRwgljgSkTYEnwe~`Yrm`;k82hQu10I|eJ&5bHZ z_EZ7EwdXmG!WCUC45)n%nsBb~!R_zYnBc6csbbk*;UY#;BI~pQE?p>{w&z`Z0r`X? zoJib-qx>gMd!^t%QK^yQnAU$-Zi^lFMh>^Ap3LSf_DGIX@#gg19^fO+HY+x5)s`_q zSz)&zQYn;&pL@v-C3&9;YD==)Io%wN|2w*GrPu7DvA;9+4{9P$3hslu>#+oBM?5=P z+)7+p{r($O! zCxQ2d1uRH`B)$6k2~&n1@zAHmCK*iCOmS?G6h-Tbfis z+c&o#wktSf;4XCEJ4FR8WS$(g#EZpQtTjEQ(NIW(DTz9#7nSeSa>* z;*e+&DB8d`p$wq9Bb+>6g8Q4hnwwo}cmO)$d`z5xnSm8M_rFfj?>KoxS~Q=BQq899 zo>{bWNLtv?Vh*=vpZF`!1CJn2h1;~Mp+TJ?&;`bkJ@aKeBD^2E$cde z5qkvINJds#Lj(RoV@>SPzC$?*HL_g_deF&LKn@K^Fog+5Z1cI|pS*!p^@)-#A6?&X zu?GPj7WnaMiXEDNX4>a4vHnCtY!0)1Kbk2K(J$m8>sbaOl3zJW>xA zRBl`mDqS70gOR!r-U>u+1`o&f*TLUpk!3R|Lsl!kyP#9BQ!(||i?k2Nr zP!FX94L3%1&>&3Ck39_5yJWl>qdPybEjV0I-f;bGPB{wjvgt+f-t>M(`*MC!g&9|$ z2n3Oe=1gE?0BIe;&q70j^40EzR#KOMdu}_C2BDfgA#jjfO-#W61C;#Qol1}k#Y}!W z`LHqy)Ur_cHR@v9`H`17g!~ChdR=fo@Y@{PbRua9axKtbVY-FGu68ZVKpA@rbQmU1zz-svCu$74;}}A9B&s^HyIfyYPk-P4+`I|s^3Wis*gWh6j165bo96q1CrA2><04Hn0Z$V zYt1=sGh)S}rLZsy*a%*0WvsW&!;JHc$Df5hh4bBfgii=p(D3P9=p!x#P-Sbwn%P)j z0epmtOK6G9b)mAPa(u`l{2z;DH^a?J@kOfyI^y436T8tcS8EOg*I*m&Ak*e1+B!AKLNrnq6UDLST(q?+?c+-?7vwg09<|F)K*{;vJva zuzm+w_eQ>1`%Fj92weSqFW}Bf8K;9chhL-Yx*67@L4dBDyolujvVzgb=#eYti85oQ z?|@K|d5eaW*mxabc>=%M-43Pr z_kI{N0d8;Q2E^&;kW+-Uwilo4R~O*7qyxW*GmvrGZ3yGpoeB2E^TeT6%vz#z7-cY- ziLez2-#5hyuzAD6HTZ|_R75juVlhP%GVkS5!li+=w*o1=!;Ttzg%O6NMVP+nXDr&T zbcO8-&*BLMgHUgOLCgTBsuBi*6^pv{BX%QakKt^)0MAGB6kyXT{^RTCJp$*ClBR?S zokFHvp#uvITz05*1(w4t{5C9iz3=0Vu7pzSi%Z3LPAkL@ zZZ&R}HYE!~e*45TtzwcS>ZJx}?2bU|cRci5xwnHLaV^=ssr0I{MczUf0DK~%C_7sE zr-LP~R1wvbDEC9D$o5m!Dcv!g_$srS-f5XG;n-+hR!A1p<~~AhQ}DPY2y;5Re?!P6 z72B;Xoo`fQ4{-|^g_s_q;(1itM4I!A60?R(y=|9A; zsc&O5Yukap?ulWc!K$Yra!UYuN&@;9c5FMrQa4a-#lf$Cg?UIIKE5&iDT*KUV=r?= z(scTLAuZ2nL%f{G%*=qqcSN>_a`xT=gP7*#61lQj_i3txG5%T<7_slEOAK>(#y(>B zW#ywr@Au@Uc(4U}t$pxM!W@j314ccbK2I;g<|_G11t^L$6E`J}enQpV*y+J4pT^Tu zMq^@YH&5XqUC`K6pE!X!!hd%a(``6aVbh{ds2-dBb=mBQthV9wnfj`%&lj}PZ^Oh6 z;i_20=hSSl`A~2#y42%sgCdTK+5r^a^O#hr9$*veyc>00CXaq*wFqs4NeWhX2`>Y( zP~t=`rF%Lqn2RrN-zfNC)+b8?~*s`~I5`g}O#KWnF(z8c16kx zl6}=qjL7prPq_=ila1)g&Rh`*%d{Y;fMl~zl2+#>n`w01N=Pi7;u?3y^eeLfp49gt zUbZL~O{e?Ga)7YEa8+c6_G}%3KVa#%^VH z%!NjXO|j3Bx+TycFdJrw*hZ%Z&d;e+nm~Aa0Qh6XgOQ@NI3Bje6Gur?MgNcc;283RDPDLofrB&H5|?bLg>tj8#zIP25DW1^o_|I_%2%vTD!hbt{=0> z$85?HC(zDw{g}m{fyWAX?BMkR4c$=$05YI%3mXuPM&^iOI_6WC(vd#M+0pDJ2`!02OI#n`BE?pk49Vp?AhEcT+5^S%iPN5it`XbfzGKs; zL3HDFN1V^nF|9C-pLw|ra~lkb$bur%!S4?faBC8n0-+9v-Ah2q<1;jMDG9qC{t@ZI zs+U(|Vbej;J+`0HVOHVpqo3g$TRwKExqbsP-aQw=^S+|;ia4kOrc|A|yHCt<*geqm zS9!+`mYRI0ygY8DD~FF_3QW|tfALa3QyjtlC*(p3 zev5aafAUIp32{OaqoJy{~b4z_p3^OuBnC>F6#2%yeF^0D45ex~CSwF1F4&&GXw3zi;P_uq1 zVg7kloY~@J1NVIDKC&xZ@4!CX9~4tO+vnU1mhQ7Z`}M>Alq4hH-XyQ*Szh0n$;JL} zClxfpWwVR3puHRtAxxN;6V>VgF4`54*IlMmkQ`NBksaOd-TeoBM1Obww}22PJ+itByugWl|(_RrmrC1}z^G&kQy zhv|mH)$+ioJzpH-?N!zvm*9vVcE%o2pFkWpR{+vEQ2%sVollGAd|5yC$MH=#3IIu0 z%CWF31*1)H{0+9TV19mL(fFO~7-5BI+1p$d6}Y*LZ%&*Ey+{fo?6ks5 zH(l$~{X2IAyCqSo&Cb;;snSmtEuJqlSqe*t9Z-G@Yi@W0X%?r!tm!B081x;;>ms=> z(+ZYwrD_}g^O*U5hFIXXTsCBi#+Vw8jVPh(t*Cr`&i5-o8V%k|-O494#!O*Q-st3v zRE1#-H1F*;jj3L`vJqA)q`Xwd_%8aG2jmqzMt##PC(kv#$AxJ!gsS{>w2^Urp!sa- z9)^w5ZnUH3Xh7J8d)q|rEEkj445;AwQ&J86{kY1gM5JV0H!!0x%rgLaS<@ouU`%@} zj6B*>qrJo|UT2_4?NcWHtWKVzD?FuRSqnt;=AltwKe>NX* zKy!N+XK>EhX_z!)rqKKH+kD_=&_se}P2l;!Kt5A?Y{|6a1z{V+lonZ=0L6UR?omq@ zIHX#`nX?G^xiQy&5;DAz(L|=ujE@8;=g0+Og}a)&=(J%7McgRWktIDVIwLNJ80x-X zO{hwd&$M0U**m#$3x)z}b?IS99MbKQQ@45NjA1E<=yCvr#c=^4sWi4=!qjh_lHc5<@4Le{&VE} z4c?5&qZU0*DAzWFh1z;XW&t4Q8JX3VOxbpQy%Cp=T-vMt8kB)@ZbY+*1{U=Urf|k~ zo{(%2oJBh%2w&Hbpt%(eK{k-zNW5dp%d=KY3T1;iD+BC-FL)&KO;n(Kp`|DhLJejO;k}$v?fYn z^zVYe7fwPNk7XdrI8Omq(-*6&SGvcoxTOq4Gv|*rm_6^MCswyDZJ+%`rt7x*vggCt zz&DR>wUxWJBC0aBp0_ZQ`#G#k>%KgKi&f|D1e~E!+p^pVk#t5Sy+BK*aUm17Q9a3g zF3f6Gqe1FofW!~73fE5MLRUw9tfKwO?!*aIuF^Z%gYZux@dJJF!9Tq;86VV{mw=w_ z1%*YzbVIhVL$EjLiB^Y*l|Z+s*WUKN*llk5Qeal|S+fJ^=RPl9U}%Pewz0N|g6KEQ zrChns-6gc86T>`_>FCy_e>!cv0;AIkiqX4%TFK@FQ-_nm^&IV4=Cf%;5-zTaa)gCO z?oH=nTxav`EM!rc`$Q?#8WT7g@`+V5Q6#I|k5guN80L4WmMGer^utIg79*s+Jn!QZ z!O%!pI#1Sc!2$amN;QY;WYcy;t4bnH<)sRJCQC(#Ih}J zw&64hS18=sB5Fo3n>1ZwJ8?j^^zC-ScFIVWkK-UlvUmjXWF`$)tJa~1@@e8;z*goo zbO4{`t1fj&TxuF06!%`-j8vrzgA#d zjdifWJtlwOprD!HPd+C7l*TG0D zId6nk32fRiVxPpFX2IvUA+eP??h_Bo^iRAxCz7O00RkD&2>l7a57)u4+iNK&8{E?> z$8uK$NdH2^eX;xWDIQp}J=RyCnCOj%n-R`)w9tYG~)(7qm{=w#2e8&`GD?c z1TUiX1YkXbEhhKzbKX&x&7X&cy{4;%+4)o|EfFqTfp&{wMPVCq?${SUoQ4}F&2yd7 zP#s)E9Y3SDDp|R&PISMF9Lv4U~>@jW;0AK=-Z_xHML=i3@ zUkZ0rvgf1#Ms8XUfp=b!ri~qJ%|SjzxVjV9``g-J_Rt@8F&ZW6C0hfkt|^DHc^pxQ z$#);R>^;o`@gh#G{ZI-pf5n3=vzCP9#bMI6i()mU9rS{Y@*Gt(QQ(3&umb@0*agFD zA8O;?87F-XA`)>k1rsa2^K9X!XqhegX?kKHoOiYlIYHJvv9yb_qQ+&v$yz&UJK>Z%VWzacR<( zrR$29eSL%S&(>hCt?$Kh_vV3RlV@AoSq%DPo(gD1QK}PMp36(f6>e7~iqJ~ILzoWaNp?K;Qw-)v^sIz|U^L;o!Z37Gj!D$0(jdsgi0buGp(wEEg z(%w(rni;}q!0Gplai=4=;srrm=ML4M^Wq)njCSJl^@WA==SQW!~ ze$QT0hQ+34vy+KR7ua@KvzhSmuUd~I|UFHy!F0*aB zg&mi<#xz>omC>E}SiJBTpNUZ5bua`OaOJH!A&%9A^H=N4P?LyL6+@j-_w)WFm)7a7 z4JwN#rBfJW5v7Aj8R3h8`YvUCM_mmfr#*&y5K&NJv5~SlzECjz{=nj3rA1r`InYj4}%cfc7st zq^0#V1I+ky(-Iul3}FJ{1Q)a`*E{-lOL?h##Bdd27f! z7lw1sidVeuzn+Vt0l@4s$E?;L81J<9%j9i8=t;dcSgIO$22=1icblSra8xV~b3N7j zCqZ(xc$|9z67Ns}&NId|0yM#{KZD1|>73N|!*Y7vmX!`?R*-Ib~cp*rD z1#nb@ut01D5-IGR3(I7O;#R}4w{+WM4}qTF(9}9c`OfbF%Nbi z3qntCL}LTqI5Kc7c_QkH=~I$kUAnGNy^Q+?=J+dgFQ|H*9zqM13#z$OnkU1wv;0sk z?V@2RI$ee+I=g5GfGg2ky(BYe3sFbYh!fM@YeAg1-sa+$;4`dOmr)0zqEV@>wnZNbL{BbVV`!G(H*{<7Fr zz%mIPAUodav0t)59JF#MKV>Jp4zNx15EPYW4)ZJUNfe|!24xFq_WQ18KZSVs5h7h?XWcniWihEX@BimXKq#KiPNGFn!n{x}yoFZv{_XUUE6aGX<26bVCn`}8`|QG* zGrKIQdVymke2w&Fh1Cov9Y&p2hvq)$cSBw`27k8 z_~Vd;$>3~9-~jk!H%ov@@ya7i1Sd4Eyubl@>#pQfPho-Zdi5#_3wj;By)?&idL%Wu zg0iPdd0AQ%vqSmPLtQsr7F?Dn^Bs<3urX3oFb*R6QS+5hJc~)}MQ%P+){SYB>);Ru zc|oD~1A}6Zts^{$>h z_hq0{+1=*5Sx(7TC~7SRPF^Ul3fkY!PZgRC(p9$Yr_)BJg^({bTWTMyUi-7xkWH5L ziqKBRo9KR}eOG!aF#ug;x7xqYqlC}0Hx1!~Qq+(af_YSFH#5DtqYzFSlG{wU!JZ{W zzG0zB90!d+^+l9(bthIhB9C&d?mF2T2FYS_6%}LNov)5XCF=zWEJ@(gWXuteq+*8e z6cYiX7oA+H3?U@%&(R4v-584pvYqC*OT-iv8*vkv4yAt1rm#`Rc<=RRLO#!|zMmG* zchTQ{A`SYZ3jj*O65$`bN9nBs%kz?IqSBmMkVAWKC7ix8wyQ8#LUWYW6DP#K_&viMMtc1IA#t zx}R?%pJ{=8SVD&R9IHeOBg?HFG4+Iw)5sNupd(HuZ$Ik^wst~4v|9UAS*(#x61?Ml zxO4eY;d2(f7rRvqE8n#96G5EptDL!hS<1R|bSY)+N2h@yM;ls-I5C#jOlGs-eV~+s zBkC-l8npxRAIY}?Ac;rH!cBfn8mqp=?dxcE#)1joC=B?FU;>Gy8RK|; zG28j#jC6DV@r0uhalQTY!tJGxF`Y4}Ts-XNF}Wj9frv89$L_I(Vshh%OSW?_FE6W- z(1SF8e(piKC%c^!o~ekm&QW=${Y6{1S72D7V;-k@yg&t z97Hg?Wz;KP1xk`9bZ$Hvy|H4=@L9E>K(1WV>c=6GYW%{v1jYnBv2n?$)7D2?^ciz- zlPw~>pEFgrSn5vGNrkV9j377@{B9*1qKHdbyz?2Vb}%d0Oa2|vvH6EE4umG{Y4&Yv zOx6bX2|Wh1tX0)5y_9NHnDNTuuR`SQN4A-kMUXIIA=%##sdBWTa&v$=R%M^)sU%$`x0@*!pNeQv1 z?e24A)H@(1i1pLdQF~^5`XS!f(sqD$nydhu?88vdLoDr9*H7|kR;1~|li~r2hp7SY zu`OgoFc*8THufb;0GN9UI7e43#60aI=;)g)x9;Xf=r@R!s^KQvDKYM0NWCHBmmyRg zPZF@pI1w9q4p@3}d1KZT%^q(1L7!dtsRKkDL$p3e2Kp&dyKUlc`jUONsd2is*&VJ; zG|IuIiA+vAkX5YPssZO`RA&3izUc{oMfiej1y8MML<$8$JH7;YoJoueAyCh;F| z6|3#_JDW~9X3FKh?KbA)z!xrVdJrbnBAuo%OjY(M$0HO|GX{GLWsStc{;U47_}cN_4G7*v?dmy+baJ%H-lZq4l~ z5KB-b2zr2OZWdEpxAF(8(s-=45@}a2O(&B5AVGw>w}@lkPV~N7XC94OF8YC%uX{M9 zQoL3xs8MUcW7)mMRvu*3?s_`!C@jW%_@|X{9@JsUZ_0cvs?T@0-?PGyUbU**;jYl8 zP|{ayhba*2g5MWT^u}BaP>u}Ogq0l}2krrs$7CYdlON=Z&IU8I{XJs8+>8$fMBHR{ zX&=r$+&;AyOZ%ZM%)ln+WWc)VJa0DY$F1J>;|v7p<@-KJqPoJ!u%nOTNY{3=^$!ml z%{yqw!R=L@F^nc(VXq1<$4?o-waz7^h$Gt0$~*O%z}I(h5?Q1i!SiUI#iZZxH`hf! zgH4Mjx=+4*#S!6fM(~!Tm-dh+izf&^#&UuA`wJ)xSGaXZQ9TcdA<{@yy*QTbaZtT* z21PDS^0gxS7>kEOG-QBTCWI-*yt210cW!---2^K0_S=y~KUrkzkJ?LL{1(?tf{ntW z?Q#92l(mGM{}W{eUBh{TmgIzAzq1(bTwdo9{%Wkypgw2!dsoq5fi#_aTeH$cv*R>B{y&+j{%c2CuAq3&-g$#$}`{(G+$2OAwsC36v z(T^&=lyY7)ck8kx+_xSEx8QPZ(@Q%`&=fbGjN{|3E`*`9R?#tZ+pK!OSL zRp-cKzuj3((GJ%gTD3!#YDH77)&akm_Etqq%`8AlEy`j4Wm z;^SAv;Zim4U?Uz$?IpqjY=G&x2|n;s@X~}`k>$C%zse>e($9l-D+Lx!aFTt3n1BZG z4Dp(KSDZs{bZhNAL{1A-+ zwj;^-ih|JWC-)EdgZAjk3g&%sB-Xbf(k*9-MVi>)M5|)fhOWKPoHfXYpNr@F>ci8K zaU6FB1ub(fqqd%R*wgGWwb;3Dl8`z5psz|nDW`5IcQYYN#4YbQOrz5ynJ>k*(%o0~nrefAT zn7X7rlqY`nB0`lDcQwSDOrYYZYfKX~-IRM3t%6M2)F%({H?T!J&BWogdYzmN_DAHi z_lcEwTm<80@3#sQ3BF~{!xc7 zk@vwcTX#=xA7C0Pt@=|=Uh*SVt`R zJJJdmKrJ?{yc0K~e4sd!?L7^aDA>7LEx2N*EDqsi6B8eJfst>gGfD<}HiFz=zWS2| zf~_lS5_vLKOM24^EEWv%mRtZbdi5G6W$1sxMK+NT5D7sm>dx%I*F)cR*Lr4cerG<# zo_MZ-0zS@()ifiC#SpXMU<5h^ifuP@;FeuuZ2RPCIP#hBQf80~TN%)-ubgZR^{(TbB??b?lH^V*Yeg*uvW|99lU7SX8OFw11^eOhZ| z)ps02)0?WNIpj+0H{7khy|OcgGIoyQ5-Z>6qUD(bb$i**+edy_=&vf>>tPjC*J&Oy z+UoW}c)hAI-`S@$)N;ys3ai`AQ!;(Fu53u)z{lW##f`((Ey|+Cg+WokAuHSBq&L|I z`7Ku2IY<9a*{n|JwU6n1v|Vstd_W!%b;6XRWEvQFSLF_iM{z%T%M1O=KDXy59XGwa zT$2&Tqj_JQxdR}*6nBaaDTb#r;xmcC79Mu?^^&>W+3UvxHjj`6$afo{m9&hDPd7|% zY5Sc(kc0Lr2ZU0jC2_PxRCA-s}(JE*W4LO2ebxw!HibjoktM2IkD4Dht%ecMBM&yXo3B?Wd34 zGMOOHa}5#a4S@IyLwHmmjmhhM2irKo8FI6s=P^CG=qsgFQy}29!C@jQrBjyDmcB}~ z?HIl+cMs&}a`o~~y}yq-YnT;#E8bZ~cd6p6{mh&1hGl#2iCQj>$Vj@}6V{n6HMb6x zouV_010kd3Y92b^kZqFZ|>B+0WUkBV_K?pgzk=!-km>03H=|od%YD%1oK+8T>zP@!gtSZRBitmAU(6q1-4up zOReQP_*|dd!;qIrD>y7Zytu!<{AdnCYmf z5$>aXT13b0bxl*4lhnS&4*7kNo=LxcKG)BZZ@F~-;Js|0%|NLf1Hbjx{)=i83kYK$ zvoM_b+-oI(wRyQuE_9&#VRP7;7mot2T-I77FIX5Xk9f->3k6Ft42{N}h_tl35|6|n5LO7OUz4l=BTJU0_gIOFF# z@~~NN1Ac}u=gJQUfyd4srS~`4UM6F)TgUNIMiWHg@vvkfW`}Tqux(`k65#z2WbQM< z3Fgy~@T+#!?7)A%rGpD3;=ECr`!OI<>&+3=y?AjxG3P#JVVn?al0VN8{~>Ytj2(d; zm?MDaHm9Nz(--xF(##oxSlxC?P2;|R!12yS$>n1@S^C0Mz~*OckSS-K!tJ5b}k`UK$~&+fx^Gge7fu`ixP zUlG)vxIPlLdoy6V-odQpfj6;*;)xUJxkR4}WcIJ(28Ih~nKVMR5l=%WH$>p3@2z() zi)}?RMNcUNJ!T9d1qstvG-jv8JTCVBe7G)cmNHBt^cInjfBzJ8bud+hNZvGt7WPDp-)U1IVi*Lrn5X<$_=Y;DR1f?-$A26w^&1 zn7y$lc*SY2zE@)r8_XW;@sda=rUF24sGtE9pyZ&;Z=L!^*%@yMLyQQvk;OJR`KN(J zBfOV+iEKps^)A@eCm>P) zsg|u_J+BbK4kEmdX$Q#e`;$<(0@kRB#^^L=#4H*|b^MMN08Tx@>GI@cj5;c2y-<>r zipBvUveZHeqZi$7&cf}IL$@Hcl$<}!MM|N`*7|PkP#u!o&d`V1A5lEfndE|A@riFQ zAs;>rF;*YvWA=7p1=u`mJ3pAHtAk>3})9i8hj7!M#0v3{?)spoEJd3jNreJyU z3wZGZ-HK)B6IODu{fVC&bvZ-O3qQs6DES!@E=Ig$)@yL2jUmkU^2qM79nx|9Ii<#= zoXtZmzfb$?)$@}>fNriXv=QhUccq=a=8vp=3Cj|}ZW|S7(bKG7y44iO#4wI!8vDKd z@@?0>0qZ>#&rF>B0sc1bIB(NckDb%1j`8JR>7%EsEUeaUj}0Hqn-*@|c1x64qdHf4 zSFc16KtP)NISCVEs%GIL{eZ+a$1O@d^|`*AT~@*PSneX?f->O^EnDch_Fu2vK&#Tz zOO~J>SJe5`Zq~T2um^^Og3jpGDO&UQ+ZpZ7IG7Dne>hP&DhN>)F!763wgWrEczd7q z8yT4qe%K6u)T0#oZY`T>gX1h@m)~C`rX$5%seRf=5&qX13C7bK3zU{tp!V7%8}71kv*^A}`*d|K;CbY#hei z32Xblp>msA@5KWy#Q6xf5~VpK-a*C>quW`;-fcuwKcwa69NWNeJm>I8G@7kESM<6L zeB)55Ksi1YC%U#MNS;r~jp;yoUMaG;RMc{y@=bWB#FLt?l;FLvdrZ#eqNs^-5suCo z*G6?wPi`E|#ggGVbx_Pyh*GQr3|~L5PyPsQ2pVSi3r`P+@{V}s5fEE0xziB9k%q9U z!`xXuRu6=SRi9m!Je;o*hraXsSz^-mikj-0BX;u*_jr>&WorB4`e>X_6J~w(Fw(i1 z$ozi5JNMDQJ4ZnoHzlVlyGGQXMdaEA(zl%H%)SGi*1xVPA7OBv zZVu73L0w!Zw{K*Gl+_k>V9$12+FSf|yIobDU0z?`!{lHjHJ$h{wu1x3)c)hMBk9fQ zzRu^M5oX`{{3&vA-ebKb2`(3shmhT| zfq=1eeZ048 zwwv{1{(O#beJjK3B#OracGC)T86wJ{1v7Q9NAU9Po@SyEWDR(#710G){hff$x&vi+ zy?9)*lj*hNeABFHntildd5x>*g+TcJMxv+eM($`}7VKxR*p*ky32z>E50n6%-lLIN zQ&Nh7(!+Qdl;P5|i5pGah*{~+t9Ho((Mo^KJzR8}J(VP!ORgOp-j{jPV5Bk}hCl@k`vv}1*-K^mOXL~4WL8JZVP2unu8abuX*I$Kt2qPpaQG7DB*0w@pwJQ)kO`dpu2 zABfuop}@j=u0(6=cwhUE0S@Y@U}DA`;5{c_%e?_SGztHhSI1pM!&6mX4ik2G_0Bl8 zGx%eUGeOlJ0Jg3GnSGY!3_H&=5i;nzm0?{(x!DROk%>Gz`Lro{>|(N$xT7I#EKx1I zPlThiS;{FBTlDL_EX&obT97DNeMLKVEBP>X^A58)&DclF>S@Uplm zY?E=tM$K70c^Ieclk>iZk?vY4AU0fb)(d3WkJxU)Ym#ES1v-5OQl@_Etq(zaIG>)g ztb2xkI}^vLY9&9hufW=5X>pB?WOwlHGqN^Zmj3)wbbv~cKL)(5SeMe@=-qN<51V_w16-< zw4M#rVeD=fx#<*x!mbr`5tz%7xrMZzMvK=?&K`?_E6??(^8C3LR)hhx7-p;aH z>=hyu4_*llJ#hAU556R#)6=40(o~rf8cy<3=&IzI2{%}ibJ->HK1K%xjLD{rftQgfzx--`p8Db z5WTjs)2858yM-hfy{>ekyn5J^bf{#i1T`D}pRK;V9IWUSbuy)4gls3ohG8lM026s+>R_O(%*115Y<_%5 z=JA*x2JaFU2VYtz7-02)>9^VYj1rcb6rb3?p}0kpGI(xLvRoSc_E_&+&7z~wrJpyg z7wk;OM28EtMkrc?X^y+olSj>Ze6p+Kvwi@*`q(Tj7NhWOXI*Y!XP_-Dy{mBb7UKC) z-fl!Q#NvKb;sdH%F_22%`6lyjmQXKv%zq%5Ov6bi&-WqHa*obRWgi7O_uS;3t3>H= zzCCWLaYTWfvxM2WB0T?O>N_EEyjOR^XOR ztets*mPXszc|^B)F9h#0QFB0Pv$1#1HLuUf7JVdPayrlL{(3|kjg%a>D&{S680Fhk zveNdfu9NwKnlgy{*ueESe(&g+rM!l8Ep0ahVrdu~`n7(Ifr6a+`J>5C%?9$v1FI1c zQHZoP$bm230%`ZI2(eNrx7K4mqr2^iTM+Tlxzo%>#kdP`(D{^68H!7&$F4cMYkJ{m zB5O97!143rHQUQC`S5acBSn#Kj4t}bTOZHcjNYuvRlMVNAr3|xJBieUVLk$t@OrjJ zdcB1E_bhs5UCh_gr5~)$+gVW~-uwhJ!0h&FDcF`%eI#&r28tnQz<8x!JTAHo9JKNP zo@sr-SryQ*1wr&S{3eloo1yPvBW)9GI+p`TMGG@vdUa~8vj7s?X$Aw$s3Y166C?57 z&hgiS=r4E5RZ+mvuSE|q0a=wMLB;rZQwnBSXoEJxsqSIyPE#`&CdyYA1c@2-VNat9 zKEwau2<;qJiz`3Bp%l?L%&?~%yd2xL;lq+M1y{9lb-3x#oQ#BPYx^MVycwChgKm5a z$tOT{dw3+*_}*g(grjvx##%S)s4D~tAN=P4MlB^P{mV{6S<| zvoNaePNq!1Jiqo_8jH*e0I~!RF&wg1WPE3D4}XLZ=!2NS75SS~cpxBI{121{eg?a$ zLdG=j_p?{J^ul`@k9CM*dwtCvo{x=+PU*=Qk!)$U42Y3I^|&60cCk;lyGpxW{Rj{W z2yA%Sc-EF4`P0kwjfAQ)t`$fO601N!cU6LRBz#1#cL#6~!leX{B8<^fN1Ggvj}rZO zN|(h-ZK%WDhij==Xc3tc>YYszfOl%Ie=CH>d7v7 zddT0c(TFCY3;;8R1B>9aJMIu{#;p|L6(9X?*KDI8DzrcmA1c?E@H7En(=Qq%WT`zY zqVP;>xH&0qwOzjC^TN|_XG}KPsO^@F*mJNLjcZ4GtZ!%?2PB*3wJk_Pw_CtX;HoxS zJ6qWvEVe-ho`ohS)Dcq~J5j-pM*qiru`f=yrV+(Gg#APF+}@u{H0dmgl%6ZanGW4M z2wA2#Z+t`L?$-j4V(*BXN@>eoqoOEsyvAI>n-_Ix++*W@MR*#ukK+g=-4N5!`E%`F?9W){$zL+*g6-LTYc2P; zePJ2=5e2z#H(xJ}la2X$XMA+~M<~n7^MlEI3JaRE##;#l3x_V-{)*fMpB+mn>31kI9?K1N*Q)#M0FU@)obE%H~@22CLE8 zI?hfi*Vdq>FO{sD)q`|=JhvN3zg$%3))n+KxXct-pJAGu9&~YfQcHzGZ{gN-$HR+x z-%JgMF$|5hQ<;#PL+!ek?P7CTV?$UT&NP4?z0j50spyV6QR42llBjSxEu5`%bn_T3 zKPB>5q(kxNoeeBZyd7UUzg1U1d9HMG+0GuhjIBu?UMhgz(4-6#Q4FO;(cY2%WmA zS($_zwkWw04kG@NAboPH`mu@Q_}6~z7yiBTH{X2k^Ve^;fBCo5;OD;gz29hpKk~0p zI{n=D{@L%Pe+j>m{u`ga{mb~~-e^I~u%^!VzqW|JY-yZ(0pa186@o)d{Km8Mb7r)X5$-nqt{V)H~7wMbt|Nf7D z>o@=DKmUW@|NQ=Ml0TLFijn--z^{kj{<-t_|LnK_&+mTu(f9wyAN>}6`PDyt#Y_M1 ze)dGa|D$hz|Nr>;AAbAS{?#x4z#seg|KV@o@qh5gJpQ--9Q#jvUcevx`Ct6@zxj{9 z{KcQVm_Phazx>y~{22fK?XUmx_kZ<=zxMzA!#_X8=k>Nh`pe(`=3oEvSHA!E|M*+l z`~Hle{uDk28h>k(AEo~!{)q?v&p-IZKYX8l_CUM^(|__M`{w(<^yAN4fPZ}Z5B|=V zf9VIm^y9zht@xk*?w3FN?dx~G{OUJ9{`S}ZgD-#a>s#@^E`Iu?cGS^Vfq1ch`9J&e?|$=r>$?|g{rKD2?|%6Y ze_;K+FZQRO+c)9QUhMaO8UEw(yLXyejCK<@gWM(omdA~D{B*-JgfEpwZdS-fNx--)~>FyyD zVz_|MbsG^9_b#X?tjhjS1W`Ar$O$zb?Vfqb55Oes;c?{MND0BaHG?8JQA-yQ8*%x9tSa>)xBBxw5ori z!7+P-pNfHuQf4d1$379+8mj(fVKY4NKlmit`n_(ePPg^T2IulmMVi9DWEKj7@W7`c z%R|$hyFcWzxWDqkM(3?O=c_MU1Iy0H&3AtBXk@;#_p!**&=t;tdn1d&fdA3QBFl?n^`>H2-S9X> zP$5>%mmdd_?1$)u~#nH~=pN~8hexXD=={o`{qMcuU zA#!8&%Z04)#xH^u{>r)g5tp-X{*0>sa-$=@6xkZaU)6u(weCh@{_#7$6p4kxvz8QR z(%WB9fOSnEw)bBncUCtnDd17@WTZXRt`WQE?DIVyZCK(QeloHUI_If>jx;m9jJFy( z==FmqBQJ+9ErHfUD->8=5p!qHmm`mbs;?=;3o37x^RC}S8l3KjBj-8~J`$P7TJQ1Q z3eYu6oR6G{w1=-*!uk_GBVUPp#{F5NVH#{0`f6lj=ncBJerTp7R6yA}c>1g0JO85( ze=TzL9Nma#w&YLWe<;@2k*6b9<$)B>L|z%XMpN6s&Lo=iibIkAsQzLh&A4|bbc0{i zwEE%_=d5o;>Vj&qaq-9NVO)H_SXGlSUzF4azz-jeye{-bjjqw0MiJ;0xF`NZ{HAU ziL8|EEZ6OXe;Zspbz|h8t2?Q<-(V&3hcD&&%&MpB`khVA+W(B)9O`sl`6YKM8E~hL32(7H(4jvQ2CRXPcan??txH zDXgsc`76#ZUW(L)j}-yqw7nVZRaoCid_Qt+_yfgiz3qXjnUT+)&O;1b$PUIIGRXJ7 zCDIVeYI+$+>t18jzxaDJbQGfV_$iIi5;Wn$EDJ0tEZ-q?$p2oZL z9SY1dP0oq8Mpg#Zn*BGC&Ec1e)iURiY?%g*Nb{gXVcFx<{OIDzvE|-mN4Kb zy=8#cIAV~_{jbQq;a4>m7XdpTQPkBH!P6J~Jo2?rb*xnHk{Y|7?m1LIa<0~4;2Gy zGLawX%-=+=E&^2Q((seTfS&$xqZ}cGN3pb{leQL*E(+wiQzMOIWjmEmUFg5vnPHP00+(f z>9)x7$eU!dZ_v#?UpupKAS7j%I=B8VGS#t86E}x%mn|*`qVY?e1_-C^LsiqA+K)x* zoUV^W<~a|KR-JQ>;dWB~sbu|5xW6;;Y&BZq#}Cx#22fgfFba~E_Z zt!OejbSN9|jSgAZ!8e)(DOx-l1ihch1$3W(D*OB*5Au10Ttp-H!xkrWy|}1t2w3>v ze4naGS6UE?&N?B|Ri`iIsTp=UUcw5BU{BMq|Kx{lycrMnCDX)js%j>He%1wP3Yo@Ca2@#(DeCj z@U?nFgNcXXORO{=IS-yK);af`Ev^V{bMF30Wc~szQ4?EGFtHUM;@?y}WiSJ#Dz8`ZmRs9fY48gDdbdukEYUqr_|Q71$^@^B(-pO^Kz`sq``YO-6@f z$8wSF__IYmn;OG)t)qB&0ll*p4==XPj%Tc7Dr+Gd7E`a!d0#_S`YctY_L|mV=cc== zW;xg1Sykn9H&!ibSQ!4Rzryo%%i4$6$43rj2UE#r+-N(RZEjoM+=B~_+uK)0dwQ0) zIkz@eU1W2?X8hY|wQ;q!VT-_y(Fc87w`}?F2K@uKQ_>IjMWg$DI5}cGZSkI_qshJ$ zmG#Q|-q7;~d{7JZUKO)+ld8FLxXt-yQ`HhDv83u;+hPif#! zW$OF#VTC^T7B2`DuW->mT~ex3NdtU&oobFXbIWbRzHwb^DJJmqz_FMS)reA5quijz zs#zgobT_hLu=F)cgCZLpOA8_!wQgj?n}!@2k>22q_kQyn5vqQZJ3EZv#LvzZ8=W(M z8EN5_-ol5~OnTGOSncgPUfH*M_ZPRUz9ioHIz8#sI0QfR4H|S>ibibCjIy-(M8mkL|{Ow%P6W&p*(%o~#?O{|_Oku#(C}N;qH%~kpj<>Qs zye!CuN*`84))lo~;QV5~XmVbhFV;s6Q%7qPhZrYTVK?6sP#0GCK}*C#;Wx`x1KIHQ zF9y^EMK>=LZ>@F;DC(RWuWAavS5xDjpR4*#tFvv9Xb-A&<|@%1{%EmUzh8u5L2X4t z1E@&~=OY)2f3E)5LZk*4i)QCz7l{qwujw}4uQB`ARwF(cTt5;1VX>;`)r%X#0HM$` zAZzqM?XTC1pN7Jrw&H7^hc1|1IEz8;3zvvJbIxlk#=nWR&Rt7HZFqK@np}@>RFKYV zbN+jY=nLboLeY4wQA9(ZYZkAoZZBxa;Q?Bw!9StFx3@V@w20+l{8iwM*BU%1(76;= za`>te6gUx6c&6uz-ot)AuL+AwOX z?ifH#q>lgh3K0%f-%$jsdo=tXXsqsNbB?SMcLmjIUM*U}A1_wx&edXN7_}8v22hO^ zB+0Bb;^X0eD?#IzKB{P>E5d`*vR3?4DEwR-yHTDi?A5$W!mZ&KHGbh|6`beVoZHun ztEvNP{r7sYG5p73wMI9H^=^~&ha4_H-XLm&09~+Ayfr+#U9q^o`TyJ~tQq;uC%Obo zWalHB#CE4|lX!Qyx!q~IO*GYX;yLt%AB5?yL5CEfjZ!APoDXgmFIR7E&!3f2240~N zovP`ywcVND3GUk3&O{EV{P9jq)&63YO%wr(P`rW^pV5SgqEc|PFOH>S7A!ehMq`Wo z$WY9RjYa9Nw{G-`I4^XId665bTeXQB7?BQM#pzz?CK63*!7(_+UH=3viH;xMB5tV) z-R^v6ff#oUbZfi02CDJaK;c?XHs&yjlnNvBUF;H8p(4Plj-hNUos4GV*TseoHCYE^ zR&O+Ejb>t2W8;;*@%`~^^JQ=@BpVy8XvXSGC1chgXvWm3hihSWQ2@4eOs%X-6tj+j zo*&NR7RjLA(*8^n5pP`l4|PlGxgfL&_;JbHm$}Tnut5>bYnEfsXZbs*A?CecbeY9? z{0(pu82@kY5br+?KJs$L|Eq!H-@Ze2=PBR#ms%AZf8VRcC{iHu$d2TZCA&m@`1g9A zbskjUf7$NbzDu+O)nfDJoMlP|`D>X@WB{Src@a$=qm)6q`S7NC{OK%st=R9>zE;@b z#mkD@2RrXlNNPaPj=y`Y_(=7tWd)Mx)7OX{;msP|o+niORm+^qc8fQK@mC2t`Q6PZVHX6DsA*YsFci>Ua^%(i*dyr_b?bcYK-i$tacz_^U88UTgT! zbsy;w-w6MWMC_G8R`cy^6G2K zgPoJd?@oz*p&8%I8)sMjzVm?Sa;`og-W>ieXke({kG(}v=bOupFHDP>;qW)*sI%kH zc>j%MKBuJd=4rQ4pc~|rJpSdZxG7XUb-90XB)Xj5>%==kXF1>L5KZSNT$%Zg8l$Pp zkFPx_-VzQ?b)H%+q$ksOecEy#LdKi%5ZZOU_(8b3wGg2lN5ubxR{0TnSR>TB{P-<5 zh@GK=s*q;Cu*F@WOR1{NIP_Qhw;M%GDBQZ-`QA<96`^b07WNVb4VUiB6k*yo8WFXJgwti{t`FLs7Qt(tO1E6sk&8E?j;+)v&pejBPjmWLLG^5r*+ zPlVp-r`#ts<&G^s{?c2-_rlW)s_t9v{OE5)SLgvhO?-8oWKP@|I;Q(#yuHoe7vs%% zeR;zz;;HbA69s)?gY_qG7ah)r-Y#Ar`i8$#kLXUFSbqG#ZK7&Q=!C}1c>V8w%#1hV zVRnuqs!yx>BVksX_`Muq(qfP5wtp*ud#-ll&C|=+)m}3lk^^;&dHMwAJ_UY$cBjZX z?@0q6yTaaIzQ4_ zzf?EZvnl|Fa>xPZsZF68=h_d7q3S|d&T}6`zyDMOg46LKu|BBQtsfHgp=#$l2Sii# zAo+`f8fFe#e7}eXLAd9BKxili!HIkr)~cjLjmFwYmy3T`Yz%EKTHjnZf&_}i3_N|L z?|2Zds>nOK-AxTn%R}Nd)w@>aPYlO-NL&<(8qNMrQ+(G-o)2>ht9c`qyU4L06c;(e z|A>hB>pm)0B8y1V&}0YQQ)8W1YD64cyOOmpA=i+xi3OwUtj~oR%~Jng^x{ke~LtE^(v-o^(yDZM<7!5 zPo8OC`Lr0qGY#yt3)Q65z_7}A%cIbD;bp6eFMq<91=NI0+wqvlR&80u+M~<)$Xi5{ z3{m41+O;O`K`+4zq#%aFn}3y5mb?)oI~>T2;mh>5)}IE2PdJD)M{+BcfMG^{!jT zL&^;Hc>-kDy&>#nI`C!Dc|82@;Pb;SUE_iJKv z9-!C#xN6p%LX12jxW;&Rb1{%FJ`MLga8%RTK;5G*TUOV6U0A1ucdeFUao<8}R`DquhUjkNuB!eMcOw z-nBY^(n&w-|IxkPtw&_nYG=pu;-Wl24?K?$5MC)l7&Ux^SoJ+jzv_P}Z1T|e5vcmA zZZfW${Fl|vkH3#V)xSvc;pfFa5bx7JnKl=`AP(S>#=rpJ!{FO5h@XY3|5S*nHXPO%{Asn*_hWHwP_3tbEY^l=i`BZ~MHoW?Z9Vy-=nl6OtHtrqfLghqK*Mzu ztMy+$3C1w`Q-lewDpu>cpNg%s3)}LhfOFtyi16+Indk@)FbK|c^pQq#57@lV`Ss7l z7s8{(n)>~JiQPfONcYq2W{}i$4dWhcXms>9Y>Tvx8PtG!DDyG+L>&1xIH(8;@ED&kS4H!D!aX#L~DK zhN2Wv3-<}7(ub^6KVHSGzIZy;OWBTy--sw8TEnU9C^+l^KaTJKiQv zUu|7A2t4qA_Rt9BN)8>eIPyib05;K#=uZs|r4GvUQ3~>O>Mk^w>VgsQF{&&qSp_^d zvX(UK&Jm)kVLvAJ!Pnx6vRsu%9dC^E12r^eq1wa|unWAZw(HP4rpL%@iQ_)&Cqr$c z$Hm|*DQnm$P$lUQ#vd%irft>(>1=wNIDgf+<7Oa}UwQHPmGco;aE8YQn!f507iOu> zIlKmKiT^#9j1kGnMaAynXm+rt|L}E3_8rb1*|2*cIy@Y;d{3L(jx5<7%SNfd?qsxQ zD7tSq7(5%bw$$(LNe%U74h`eaVfpaLzHEbK@5YUHqr*M1wD36NR5fL^q)0M2h0@CZ zkCX*Pv)1l8;gb(8ectIj^174z1zXg=qghAnVniQV7zAt{A z74O@ya9?yZn`&RUas5&}*{JAJynxr8GbmV%m`Ax7eY5~0GkTXhEOz(A2I9%Xu>+$B zF{hG8EIcV3#>3eqZAUg(L9gtr)$C)DBTZuWP_!pDbXfNCh_yQ#9XPz~h}a!V_UV>X zuP28@Gnfy2RwfchySG}1=CTSUX0TOo%s5No<3u!hc!eHHtuS_tB3Jkz1=%SYKZqM! zAjh-mRNrVXNoUA*S7~X{UMJ^L9pd*;C8Dl%_fU)<*-?D6+VBE7zg9og3U21QE}lI^ zp!?D|HxVF7gT%&0hT^^PETuK`NLksGh$s7?AF?qnF_4b-#i1csaipvsJdN&;C9M>q zZ@{WA-ro!zsR)_r+Hi?ubKB0U&GW`f5}nI;Pu16J=4?SEh;S!Xayw$7%vQIQ@^CP^QH zmg(p~jMOF+CYplkizRzgqfmTlP{P9mb&`FeuD-2V61HLQVMwMrXy7cgCdk!02({Xa zO7RS~8(7&m62mj-Ta0N;C8UD31uJ;yVR|N;CqzbYhHCsups0QmaWIa9 zv7~T^bwex0k7Gg(05l9nHY50O@TU)gMc@ zSlAl@qQ#NO&h13maySU`6EcL_=tv(1nydmAqNMa&D&s*z`RJPK%3@#AeJ|uGf2>D1 z@9XDlX-T(^oX3w_^SycFm1YG#RkFQ-fcJ#ql5Mfks}PXndXAbcaFyN6x{3pVUR7va`R-RgMs1% zVAc3VjSoMLhuD<~9|N0~3Tp;VfrfP|1#jIjvifPy;@IT9g0wEgLCpEwMU?W=YAT!YWxY^Ub)OS`dwD- z8dA$zW%!to76SIVN$3y*nTdiCq>D6G7ndA~Dyc9xw0c9JANXlMIzH_o3n!rUj2@2< zV0Mt+)P2PSaBf-}shjYs(R9}k(=S}m&i+Q1C=s;~dCi0E$?${8POhM)} zTiO;(ZFliiyd;^tmbn(Q$6t&!nMQYU&=W&qO_?h~AfYJFs~N{x@O8MfirV8Rwkf+p z(^T8zxk0}nUNjvVeoRs-j{6-9u~4MTX@ot?KLB~N!_ucCBQWIuJij*bbeCK zzG4DMm%`bb=4HZNo40%~P7K{u9>WJ&T*y=7l4`i_UFQqU;wO1xd7?b8O9c(;u69Jo zE{kR1fU~l&onc20;1`@~yxJOtWejgxf54if#d-y=IR;}xBe;xNpk^JdSO&IbK)SlrxRa9kLOy~zk+6#bA#(G6 zXjnJPbKyVqT=;?J@-M98ToC*ro1v{@p3;rvCQv!AH?8snOl6>*28Q)S1%LNT=$~_s zE+u2t*CWr92#I*MpWO{?^WQKk%m6zSeFJR9v($W{>Ea=m_tcN(8P0hxie_E_7DKbC z#I$?<#yk8I&*(&BiF$d{_43lJlT&`ygGanxnUDNkn$2Bex$5hZV|i08qdV#>oFs16PL<)K*_J+PX}P_AF+?p zh#A8H-~JNU3dtU6HV2Zt{?-ybD)`;3O1h(5?>%U+G2AGO!S>ZKVEY^O_{0Nn1ZHqC zZcxln0(pB<*iRn$8Nt{~PP@7Ur&WJ1V9X3oYukzC3J!^YW8^$^ zIWVfunJK8bT(M`-s+zm@OYVwm?pip_@;rWePWZX&HsOw^j|sp<(l|$)$!Z4sX|^HJ z!SZ65SZ&l&1udUsusxHgVTpBkxM&r337`bYzHE!)ok^kAUjnyDpr)wJmc+plT+-ei zKzq$4yDoqMy&WcEFL42)(>_Ka&k`(RSf+4_lOP(Qumjn(w1F_rhN*nte@J9)IG~k5 z>V~vETeZn=o=St5R%@S|OaOCeVT+1Yu_qliW#<%mQ9AAOV=9T zS;I@QG|-j-d?{L~6;og(>2F|K z9+cX^VWjCS@e92u1y1MW2iCx?VrVmA+p>b!d{T(x3^&a#k8zB4iKn1@fIA?>(i0-*pP4^eqv){DJy#eAy7L+?lP#T2-WkLsXQ(iNO zghBETx_KxFPr(toiSrNCLvoL1oLg6k^KB6wQmGzM%;3xjE<~oOtkM=U(##scK5jP^ zqGuHOjy$VXPKJnGN1jP30boyNkh#Hr)J!&od<;s&fGmLKAB@ANml?%erqeK!HV(EbD|E-uOaA784r{b-sH&j6^=m>@-q%#E}xM@cfV5h|A= zftO)NGx~*88}u6zZazz>rIfgCg zcUd9q20qsV6h^dh7|!Hj()T?>G4fxt*BfaCF}SgjZInUpff>VxADZDe`Dh`>0Hw0A zu`G;YLGXrRWAGHC2TBePr}|<;85W+cxDaASZc~M<JQrEe0 z3Nit7d~-l+^B_f?`^;6GD%lfD;*46-VQiGsmICNCSm8=h({WFvn8?Nger_4IwF$HXdyAGED2YcWi6WHuu)NuGkaoi!s>wAS;4+t|ku+_YJIQ>>COd zO;-T%Dy5x)bV+c{WsZf$I1I!oRJcN`!%9X+GwuWdFzK*O_4V_^jl4)dP@1K8bz(jG z3OV7xAlO0UT9r*?$irg7f)u=rKJ<=~t>gkW5J`1l;m>Zk-AoR%7>$JBBJC9EgLWHqY7r z+EDz!jr6}MSBgPRxt%NgC(P6J*3s?VXV{67+qrK3HK3Heoom^6+qsMaZcie7_GoS5 z?2Tn&^!fsf@-s^(xt$C7kBC~EjBupIx1DR5yPa$KsoBoeLVF1VwsRFhg0^#o=WZ;X zwDCqFSn;h~xi?&aaHlIVhocE3dPn=NgA)mxkk|?*%?xi52T-w1dqK% z=u$Zzm~+h=^DP3o+i2xRYH>O!1R4mYT?4YLwEFz=)x|=QOridHeV5 zOWm*}E%wP>mx7a1D@LSNz=t_BBzVshR-^2EM6epGt=>Uw`hr)Jw+f6>blSNohp1g0 zCunEu1aAuO$2v`|#IT%ev8)bcZ+iQ|$hAoJs`lGq9d+qQP$R9I=KkSX{}DQ0Zv(9HCaUmlyQ$ z%piz~xrRHwx}waC#!T;_{jMHcxcE)$$`fU)#ua4-Liv>!k6$?- z0eh=j=uS#98dy~X5dx-|uq_}~YZtBZ3dL_}>ATKAZ9)L1<|9&5<&)?2LGs*abc?U_ z?7-6Iut<;d$joIH(V;I({0^|f$2`nQYbx^v0e zSdJoFN{Y-tpsY%QEBOYA4)U8(LLWA1eQ30{zC?|4jT&ZC?1?!kD%C(_OJxA(7E#hZ zh9HIDy#4&DDR%j4JDJ=zkMX1pHQj*@Y2p5$INohfH%F^e|LjkhPur z%1-xG2c#?+rmGioVinKPbmKq`UVCYy4)&Tf>ezUsNrLTDuvl#vSk!yZo!n$mib8ZK zlj6s2%v@UYFge-m^UAWxjJ1=8M?Bd(G};%lI{Km%^Jtd zqr2mBclBn&U)wqP?iuV>8MIn?{74?PXheU+HkJLOLYrz!5n5J8N(1E*0NW}U9F2AO zM;qN>qA|1+9WDdWZWEg_;ME3>zYkBE$#bVjj?i<(eZa%O^JF@$N zr3$>oK|s&K)DINA#LP`t7}q9dY~o7~>Wyko*LPsLdmuS8FWT29_Y`6+ zu^)*bG&COoG9j$HE0u%Vs!w9Y=+HGfg1J9J?DiHmSBT zX5ir2Ae~G-hqoNL#`8A~;Q8M6JscqMPCVg&2+li8tFDkgh66NM**U*&77@F1@*Oo; zqhe??)tNPlZp0XMo5-fN>uA7?L(?eBW*n-HM^ppZaY0>316^W(G^XLYP0rn&RW-IT zWJXKV%*WilyngA0(qeBw#~Emk&;8}y{Y62Xq0wOv@TN`V40d7le!J%R*a6vPi4mTJlf@;GD z+<=qFjAO)>yME^tMg7i`Z&tNhtxhjf&9bN2-c;Pv{C*;e^fVhoiDTpf zHfK=OL7ruiHMSz6+jiF7*|j@!Ae~hk37LL4h;SqmAHtz^l=oi{syZnsY-Ns>`_U>u z!B5Ngl%VC~w*}BrEAd@j>a00zq5@CUv3EF~8j_g<`ouC$fp+z@PX)766aL?%xKTg zONs|nA4tcdbc!A4RAZYPbTMos6!&^Y{%H;?ku^q#U8FdSBV$UK=|1ukYBCK{qe1JB zqi@=_sQQ6?OQTPoH2yHFNe*?Ncvx|pSDMx2>WOldmKwQnU|7Gq1*s{hmO1wvS;-Af5WpeV4d>=Xa{^-tv4rsH; za~QkFMtDE{&|hHauOgeq2i-FtU_MR{n3o#I;Ma=6VGx>p&=+W``X;ggsA!SWv^>KV z1s);QZ?mJP*CtNetWN7I)>9l4ND&zWC+|sV6wVJ4Ke)$Ip4Z`fJKK{^Mk}lUXE1u4 z5>yPmDLfDccJ2d{*mCuHG9;eSPIrayY6C^6o^~EnY$a+}CRBrDs)1=m5M4PP2-2h} zF|f6%9KpRp3vL6cvZ|hIRLyI==3Fg(4JEL=m=Z9W-V#`IUeTI)El;;MPgbEA$V>oM zjbA$zu+VE3mZ7hdAjp+dnn}QmD!yCMn2gh5SK(&1jaBL4Of$mHBgP$*^h`QaqJw!i zN*X>B<$PItHaTu!8o-P~v^DL?iu}h4^+b9zL^6&9-z*(5-_)9Q7;j8R%&woqhU>tw zutS_htz$9ZU@`{$EyE~D@*{S#@akRpXoPV-ZVxsgadmdoIM42_I15=k7O)>DUnkJ42lqEOb66V^_H!iBJ$e)vvbddp=lEA8$cI1Urdr zI)!|=gqLU!AFhRS1@zjfh(~E}`@ZZRSf$u%MoGWCiM0U^U5vJHH(RlZ==BM)@}Bss zU^e!qkcKR?HHb_|D!?JcOc(a@;BaBK58(;p1O1UN?G`NbXtSJmY9jCCL%eIiMHiD=^unSI;YuEY91$ zC1ze&_?+jICvge@*;AzqGH~{u1hJB~>@W1>P)V8P>6D_S=Xqer>99v*!S?9K*dFC& z$8%cO+_0H%*da=gvsp8Cz=3LhH>+$b&&$fv(!;at@=7V5!I=IFW6I}$_%Jn=FgV^L zKp^WL=ZX}_A9{OhQ5jy52sc`;2qLsZF|FX^E?xyJCV5+~o>;H+gn?UkB@oWh5DLiF z=}EhZ>mCEGilNOcRAoMLkubC?*KGl+&A@U?WdP^OY(TXd@|0yhqO?D2Lh4h^n%o2d zxH5_NXK}p2m~CW(`r{4fb(9%zc>de)5!EJ42qoeTE5?gA7zNxOd+`RVT#UY3fRWFw zK6%G~K)hjv8*fu;rm1s8E-I#%14;(D&9bPiQ+EgLEhJ)i^ZW&*2^O(i=0($t0Hb^3DEX(Pz8PoH8~7-rem z9u#O9_Fq|avcS4f{awDIM3=)Y<92zgKiJ}~iH48R0!oEUm<41Y!woO(Yuk(C9F4%q zKG#8Zvf%kUxvxYgt6w$2PG+c+mhL5wIrq}*WP3%OtQ?-dlc^G&T==?iJ9*SUlCXPe zH;lY}Y5cI6K4i}}hMfrH3t;JDsuXvdj`s8-YC!?D3XroIZ+}oPpjl?oNh0T`zumzH zYdI@l8*W(=bhlIc-vBxr}PpYHXCsniP+nPFIE>KidlLD|@#+xbl z36Q>{G?8;f&PdJ+bVe427gm27T5D`&opi5NBqN!m|S5OI4CH(0QUZEEB zXGX5cw|r;v79S=9fy(Lu2eiE=ZyXdcPs5;lgyipjBK~^Zk=m?#7O?g{op{QFFMs$O zsOk)61>QFT+EL7|x{xIeas<7Voi5hr(`iGCLW+O*k(%FW^6k+Lud&$w;A11=)j3yu zoj9B+>^2ragDP*%cundD{$|^l7QGY$^)+R1?CcWLP65l6D#h69%w?k?ZBA>Mv}-@k zwzw?5pHA(fOhlUCv_A$5dhc*VW3Lv{F!&}3_hMJMu?{X^xvg$EI)IbED8CrxaFK(p za+a+O4zEPq3Ht)CLZC&aF_5^CdAhMI&SYsqLI&N60r_Mc@r)xaz#NWMHZ;7aQ7A~wzTJ`T{;Xr|D3?DY+_d4tfO8 z5Bx~Kj*;e+hhj)i;e1*HH}`>>7cPg*Zw#90#zdz~+O}~DQF?_J(2_i?<+XvJme>CC zGWM#I=cNpi%P1UgKu!?OGc@F(u{?J3{FpCDKH|y>YnZ>#-Nz#tn{y8JsllhJ9j_Vx zAl=o;KA{Cci<_RAi?m_QvGI$~Mu2y?S;w`bgVvs7gA4Z@%kdi{;$09YH$vL`a8oxT z?Jeg?Q+jk5_s8M_)8>IbBjm1Ly0^7s3;bAQ`Z>nc>*{9rwP)|3>sjUD^YIaNKSBOu z)%C<@N!->Z9_w_z^SQ`kdz0$uqj-Uf?`sz7oM_BIWct)6h8N^N(+WNJ+s4M7P^~En z_p>6${i;#Q=15|n^?pjm)A!`YGL3Zq3tub6_rh;SfOeQpP^R-n5iG@J41h#weuJa^ z{j{MQmVpLH5%+wEE!Gt%l%|ZtCOU?7?kj04em@SG$c&7p`6zx4K_h7Z8IBo>mryJh zczKszxd(tk-^(qm+#aS3Spo<-$NHRGU#^;Ee=Sc@CAx$rOKwR8MDXNCMRnGzF|4qy z+Qc`(>U2MRClH1$nAL;s^@#=PmwZEWaz};+(mA?H%)+rwCwX*Q3@JK(7uHUayLO!d z{4pQD)~PddjPW_q>4dHq7uiFYf&P*Ecb6M6WlK%ZU8@muXl%R+6Dvcr$2vsYnBz$t zX~1*wGA}9FI|-PQRpK~T4wqBY6sFQcQQJEuflleg#oX~RK28{Bx7=(=rZpb{sFc8Rsz9_DGHpGcd$-SA;9@?s? z&bu3q6**|aGt_xmKt09zbTHjj!M*LC~Cfc@RfJ%j|-TPo$6ejX&UP0R-t3A+ zGk}4P4&pdu-exEbc+f(wTC5mOO2(0SDa7gJNhfEgl5&wGcaa|5x99@r73;!hIv=|? zTn*yHvIpttWNKdF&F9$@*%UopqPs=p-ZP4%CNq$eJRtILI^%%np;T%=I+_|zA$vSV z*P1{|BFh>%O>R<8#&R~l4e2;nVy`D%H3FRZY~&=dWaTo*fb}P9(0EynH`<4HslRr0fq(- zk;94zBB40%o;Gcs{il*5{QSUK;tA^_Z?#k|QUH^CYL_(HomIDCF1u3rIxU4gv`7l? zB6nkpRo=Xu$h5Ri;=G*5w8ZgBv|K+a^Kv57^6HZ~FaLL1jx#Ooai%3pOR2!$@ws>W z75~Fk9)nX((loI7r1?{=o{CQcC{Bnn|2aq}Up*C{2KMOCpFutOr-42nIh8rOf6=u6 z)l^7-)tYKac~b&v<4=Z7XlRSy`)xwwO}RGVUGTv+HeQ~B_mo|SkjSIBW}&|s=jxHU z=H`Ntv$LhK(drn&C5v`%P87XMiG&Eob1MT zT)3CzP!-b6umNBX!t~%J7u!NG0%2F-QU;i9lB_D`DgfN>XOluq?#OG+9`5@R&Ur#&(zFwE%EJggsK`@#JiPe!M? zCLv|yI|mMl>D`U^MvK&JYJ@ibH^PdAE1J#+PtqPW7`-_18wVd6ge1cWg{KiWhh0XT;3r;_ppN5pSzaw08NHjAPnZ@i1m%WhpqR z2!EUu_L%?wN&c|)igeehrce-pru{!Zg9eGmJWf))zi+wWP&wN_$2=a-f2{5A{5NQ! z+vVcD0F32x5iZWHn8^G!-45dbm@DOEz69Wj*3hLF{sw**b@>+l#+!15{LC;3c^tUI zry+X&Nhy=P^LhGkBqp8h!}6UE4H}@$zO>d$)6tDln?yM20C^w}B;IHr-K)T5X#q=e z6st3D_lr9DCo_sOF(I#MKLsE5vm=$3XZ?7G0U@|XE3}k^HIjlR((SV5z{yLLJC7wH zN?p0F8-61>D4%8;{g~I_#TcAnwO2N+e9`)g5P%O;(ZZ;x(}6)m9LcXvhbGB=&%AiX6DS-U0xBH^ z; z4Bc(cjwV4KqRlYwL8b*a5uL$odUGfV|_ zKWxxk_%loen{B}&bK#Ut1ue<#DZ5Vf5SJWb(El$f?JWB*EA58&ohoU^lYGEXwZ;tZ zld87H^|2do{=L(=`%ZCD&Gd!&-uFilQ9&H!;Y&}ujmnXWui`*P#GIOGFaJ#-6=EJ_;O zf1YAqfhOortQE0jq>}|>y@Sd4fl+9^EKW#+MvftlKu2NYUZUs_E+%y2gLsTHr0GUK z@Nd-hnWu1jAzfpnoffMT zXuuAjQ?9wsI3ydE2=D1hM=8o8R9IxI3pGCC3rGc2nR%2d-?DgnH$O%Z;`{FLkUs$jGNvR+D9M* z0TB!lxQK6Z3??TGQfUOH((wT---qaGFCK56L-mR0BE~Xfp5p8L8OOZuO}ly9$0?^K zUv6B8wTX{(AxW|RjKt4?Np6*MZhd%OyMmuTzmAO#)+fG&l32mqpl9;^UC!GcsG1r1 z?CCt$@9&~)s;lnD{PtzY&d6O*WOC|7YVjfvA`);I&E1AmXYITRThWX*T3kOtIMY}( zkjbX-;zX=XK4DS0tT0fU2&igP?i6CTd9sqoKyL!DYW!lY5Wg;W%JNEMf^yD4u~Kle zgXCr*ma7d}7d-WgNKyl}iNhGgQ9FK~>+%WciOFXJwMxOw4wCnURNFL34Pan65m42p zHt}KSbj-jZnVx>BT)h~Gt(`EGGk6<(-ngCwy=5TRRUtSeTf;Gurw4tMV~N)(J?fNu zFj2*5AXzcAndJEK>)8Sv7gZ_EtJR}^#zI_YA{x4ltW>${R$HpM3zdc}(A4CIiNUmr zrPq22yUfxOuU1O$EVBl%Ue(PfZqWh6IjxX2(v>IC&Asvv-Rf3o_LIS>4$<+Ksxu60 zSxgZ*vAvs(g3B0DKJMpj8qrsD`+SSWo2^t0r+kc)RFvDW-Lu6DhN7OlYtrRK)_6Z% z&P{iL%XI(_Aaw6L9h`CHWWng}#$~GXiotz1LkYU3cwf# zg%uFIa(Wv$mjKwt>p^$H$V-mmBdD{9;iL?yiD+d2=N4g-d_<>57fZGopKH_v z&6Qn+$C7ozFyA_+r*Ytv+`Z!vyzxoI5XeVBhq+QNT_)A+`XU+|mGaG>BFID*YR66X z=z%>o^AwxU4+864doVo14t!;_NaVUVj}84n-C0FZ@^n{;5{;UsB}xH@2fI0dTjOS% z-JuD2Z@2G~w|jZ&%*)3aQ)ix^JTna5%6rAP)BrBs7RR@68qMi(^Y1FSNc7i_HPBWO zh_~HCvHom*`|gprHHK{hV|&|oWAQ$iZ4e$!$}RrJbB^6~j8lks|HuiD;kDsqc2Ji8 z&HXC*T_b78`-L(coOugMO(rOkq}FZ_54-V-UI6EmCST%vQeQoTc(KH4AIqm>A(Rb)gf9k0@gI}-=OxH{&&L9~WRQ3Et4Oa|a*ATn*@rMA zGCRpZt|sDCSw@pC=RE<7|J4@H*KzUo!G`THT6`J#RW#2Ct~W^ff$Bs~KLfzX7M?Sf zJ7*qxR&~H@ilRWdD(9w;i|OZ7X>17>&HP3^%~y%0Dcb8KH&!It3LmXB9_{)pe6wL`Bm#+4 zWqfhL!Y!Iz>m1k~vTQuU{Sz?ezp@zSp+6{A@1c$3l?THS~Idb%`lNVUZT1aexJWg!I2RY$a;4jp~MO&Qcu$UjZ#JOo$ z)YP=n(}qhP57&8n1{Xn(oLhB4=&1A9uvly_k<~n1z$bPzTPinqOh^rT3QM6l_Vq}i@YE{5)MgDQ7wY*xxQWT zwie-AWwoe+G>h5{2wZUkEEiRKku?6bK@)9qJ6a!w7-i?jMQvDeDkfdK;`|jZwU*5IZCyz$K~#G}%%%8OMfm&%`*dRc&xp&JRV| z0cVVnEq|){p(s1pe;$4)$`0cHF@D%k5kDvi#&*mNTaMn8oIj5JayZa_d47iyRbv&~ zFS!rVLexn1fgwYt3o=&ew0IU1IK#c(&D=ohiYq{59*HH55#b3`oGb4cH_4bSQ{wg~ zk#B>^DV;Ljj$RfdUNtM;FM49^p)paxu3bMa+O+5{18r3DQ{Hg{pe$DqQ70bV}QX1R=D;3 z?M>t?S!A`ev~b8LZvr#uv3B0w=pxz1${#zt+#67BXu@cEbilI(NMGre+KraTsNkl_ z2mny%&z4p=mR#k&MJDtSAuz|e;Dy+!#Rb_NAt?)2iZlkA7d9>f@}~!B@yuH4l;N-; zW1-!WqylJiajfA{p(AO=+TkHnwDGHWvs5XV?N;0dP{79E8sOUf;N*J(RH*X2K}9&< zXYqK?Wpbm@4cASBp7FTa`^I46G5cAtZVa2r^VeXg$_3_!rIjFHCe>=ma{0M)6u0BN>>?q`OBUW3+Dl~=e33H z)3BLSbkGUBW>_ajv6s%xeWbOi*{t8_hNVlh-$tt*Y-#Jl8FmW$o8MhqUiZXS_n)J` zzV3p*&Xob+#J9Xu@+e!nG75$ z8AQD?u<)uz(`jtv1K$kAk^@*ktsrKh-Hf=w)CjyGE_O}p9#}81F0W~$FDvk+oxZHX zmu2*&tqtFn?;#(B=wG(aW)AmEzhz{O!3gsQ(3Ur?V1sZ68ClF{%6(3?$wlKcP72<# zhk6-XSPS7FTgIjdx(cW^f|X?Ax{)lXAP*!LuWxU%mNi)`nygh#Ry*}?`JM*Cn@ai3 zC<;jtnd+|zsDnz-F1iD%JzF>E>w_ww-v{=>z(Iw*D&YcL80Inz^+?83|D0T4OE#d! zm0AAhGeu-)Xjz^BR*zbHqO{VY(EE&QnxOD|Te$#&fgS)`R2?||lz z5Dr8n4MIn7$R(zdFYw^X1DL!8C!dKSaOMV`md3J6I$#jSm4b=A>ry(ArUQIe5#=jI zT#ksfsH+sQr3Vh45rloDTf%%t{$%=%{Xf_7DUGo)K7E9CC7lpD4r5r{&(F~No@e~3;VqflU8E9jG9EQAt1@lZVq*&k};$VjM_Mf zgaQL_VbOUOP&A&CvJMZHQY)kV>rk=vr}raT=z)<2wiyARt}b(wY4Bv))hp> z3J2eCD3t`)40ToxGm%S4VjMnes|@>?EHqK}GSKX<5MFH{zxwdxFkgqq!4qa+1IfyP z&G1iS%7HQ{DKVV%1TitJ7}$*b=~jqgD<)uN^G^xnT;+9{K<2D)Zkdm5uat$_UUQ)r z+Mk@&c;3XLcx*K;&ky3k6v`Y(XY1RRP?Uv${FFgyUQU7u_sFVe;9M8xO zY;RfRx0_tkj;=XXnNL2Ascb*}p6#aP?7#y^#Nt#k(O8q)Fc|@Cp6(N zmSw|1=k`o>OJy%-kq4}|;iZ~#J2PBU?z63OahW7xdElAssrYHS< zlqhCFmZpS(tq$uIFi3tLRpGkA=gT_j3}W|>$04VdLEvDfaG2kB>EjSRwx@zY>~T5d z)H28{*?Vw^l5HI5yJBk@fxgSO1|=$Wu2a(N6&x08Xqy|YL0_I@94Z!IGUFLzqI zSgEqkLhX9W*fqiCyeE|hJ%>tbP1uioSTvkwEhS`h6w=1MZ{s3L78%w+j_9K&oS?h9Qt=KrM z6&VpY$3&!r+y+7S%!ft{koQTDOSaLvuwltI(*ow*4xvLvX)%=%kfp=;18mO^_39Ei z`~Ze|1yW{#R6v0~vj8tvmTc1j{_BGJI1!z6&BS#QV)gKWkgPXXjFbt7oOwE$ibCM> zbUn8W9J_0Apv!PQv{bx-H>Sm5mm)@pczctz21aISBbe&iuyPw4>*LLB4Gos)Um4&I zZ56cVEl&Ov9H~AKt8Z&{GY|rbFWF#eh`xZXVZ|E8IGqU9zp{vOE5Jq@vXxWXvt#{) zwms*{DW|k1i%Do(yQ?vDr;^gagl@!63@o04DD5{0d$yHf5boO+WDuqwRN86xwo_*i z<|e-QmEQ7I^+_XUf_aG09d=yeimfOmJWY`&f3~G*v zQP>!h4{{47eKZdiSTv1FQ{CvLmREtyX&Za(iW}vpjydgsMUCc8Z`3dbY39n$(>p=oGRu#Pg_O2=zl{wGXADO8GrLLO2*$%ri?e%CvP>v z4SF(Ev%4pR`DK&QVZX9>k6?cLpcpy5S@Dmu2HHI-K;36bOn` zkpOe!5NHE33n$XygVHVNJPSmpR6fH=!f96!QWfK*baCPyvY=}d@7<(fB(-_)lPQl6$nWd=l8pHpzXd zepZvWOy!gK(@9vVa&`!kZuy71eS3^kLst7uoYk&xY<0_WpSm?{2GX&s!EW~2g-!jk zI(fJ}5?9`l3HD>a+lM~cj^y=2&1ROmUoJ=LJ{c%=@44_IFL3mwily!@Z%?QpR&jVH z&N#s}Ts+|+S{HXvCQN5)Y%w-uB9F)r063sLhKM!K9$K8r3^n;MuAY+Az|q@uV7V)Q zVL)O$hUwBaI=Cnup5_(d0=OK8i&xh_M@f_gBSCV~Z8U6SCP{OB8Ku z{5<*PUCidms$J({)lOSLOVQ!^8~@Iu1Wpmpr-cpSSezKAo)%}YcVKeFP)s>lzJOKr zm!w5(Dlwx=i*<%nsf00V6F<=s_@T?xEYk{hA`BCVm zI8z=PqQ#=4V(4ZgxGE@}IgZ>ZV$UHQWu26VXJ-zY=%hBB(e4;mY=;M44!Ul(g zsB~x?Y=^2ytr|sMmppNi{K}UFb^?niREB&;eG<*e&!O~HVNXylqt`yUt$~0U?)cO> z*V9akaXWHtV(!J905$jG0!Mde0P!lMoq=;{XiLf(3XN|VXv5WRw^lqjiotW0*JW~< zbkZp?C+<}2J}H^I>uVEt$#Y7~ntcWf`;Fj}B7z}bNd}KK^)$e|z6z{t$ zYisP;*GAbT``Yo15^R0wd} zwBzESK*aNpDWS6Q;_?C!hLdVD(WNid%Lh})NF1&fNDaf>q&uCkj=3%sr(-NehHzY% zc>#laS(UebEd=kfg7EE zEZxxK^(4PUtx-gw$fK4mXVeY^j#~KK#oi(}S^22-U7mu;Q#zd@4~H9xJ72KKFPD$+=5NFzC;JDJ;w$Y?m33hMt6g+hKUa5-7RBW z_;P+Vx8Yp3zUqThPNTz}k3P0!&hZP*o^~J<9$4zU^_*$14}baMdP6`HF!OmIu940a|T^}Vf6I~Nv W)VZ^!{c7rS_;G2-xnTCR!T$?k)cocE delta 55381 zcmcJ22Y3|4`>;C+kX;~qxfGII2$uq>mre?V-dpG(orD}D3n?U_DH4hej6Z}64uYT{ zHk7W$f(7gy5WAvaFDPJF)bE|y+nL?F-H1N_kLO`8yEAVuQ+Ia%`oVuckltoxn@#t{ zY-ls&z^D?TQEU=xOY2JK)s@y(G<%v;RyAMMoLteSY5lZGmF4alXU$yaNcWtEMRo2O z$rTL^we>wRGnXt`lCh|=CZoD0v%Xe7bHxy$M^2TvLA8b4t~ zZg%(XnH4!%nR9FE+zZ_G)s@xr>K7BtP#q9A63pY>wGHm7GIw2eVMbMLUZ2Hv4NH41 zu5>SPmDanKxXWrw=ebjwE85Lyt^juH+;iP^?&>*iJ-aqvZq@1$P5uSfG;15K=u42|WYHKrxQX5aR4IltgL}BNmlaxDMr`MFGNe&k@lH*H_!@Vt;$u?q+|_ zYS#dNpKRAhSWPsL8oUw^MQezrdqxHov%0&3O62JkLPbDDYDq;(2@qL@k7Ni*sA9V* zY=w9^gFPeO!~VWs;(sr0e?NOZu)|Pa{GRXP3Q)zHrsz9#tYe_Jcf5=JT@hRXe;*G{ zjUeh`vWy^{Q^kbHF#$-q%rho2A@DNcGWf%OulHWK%(E&k%;O5(VtnvN=w{=C=fiwz zYzp_OQ5vz)2$IyP$@t)lPR+&#`y#uM2dh$Cl=URd44m~&QS$|1HsKtucSqmN{w7>s z=U8vJA03kd_Zqewe{6?iE+Vu}z&o&%;2E17>N(eCGrK1k?QugqL2(mw&!*YefyRtZ z07fcAdj-#d^mOVzgzGIy zp$e{`FS-p9gdeGbegu*3V}Hk`%!0o`8NtLW`ZHyC4yOzjB0T}A)5HV>k~x$(gyVKf zUG1ZWYI4Lv0ri~isZ#_Y)3dNJ0j@FWQ|y@p#268$qfoNY&$B<>MU^pJLz3zFCcTsa zFLY&^0DBI)iiH_wu&EhoX7%S~Bnq`=^`Fnk23k9cTv0?^#sras#X+CU>4LD9O0(AU zR920Bt)DcWe%ZCcRzg4~iJ?RjW-Y>{`^6Xn?F#dB$%%{ghh?pbP~D8<#-K>yAmfxH z@*K>WVMaJEH&J-jtp2LpiHvDNT+YLW`;YRvz`Z6pBV>?#Q1^O4I7cM6pYxabyZm(F zPbTvOP3ChT^Ej!S8N~X6bYu%)0l+~#zZ5J}36U#sdvBqK-Io^ifxm%M>XHm7*7HHp zG1MynXS<{v&&NF+F+BsoYAK`QOVk!YG04)X=M4fH9pD+87%zy=fY;$?BPuLp%wiZe;kJlfyh;jk>{bzo(>+2}Tcz*y~IY&K=#|3;>b>+F}Y2K4y@C zqG!x}A`f$V#;idRgT@{}!eOe2!yd5RXllJ9)-T zY^S0Kv9 zBHDNsPDu|5YXcTb8Konzcghr9{V-48srMq_n(2g9HRvR3+KNC_*oOKsX}0ES0>^s3 zo|P{QITSZLtpl3S29NC)(j}cmTvO13(&~AOV7$=W6jWKetf8W&Iu|<9$;vpw7^I*! z_%xPdJuM4jV@9dGmVNDwtl{LI_6(IFp7RU3f+;&yT`Mg0Y_IAfEXC)`gTfA-s0t3s zZ-k21*Ox82s#)>=sVs#@(8?t%SHK@l9m*@~YZsI*n^j)gP};04hiA%CPsHN2!CL*5 z^bx^1_|OlFT|+wPaeC69NCmQ!pGXuM@v*<#729i}I%t8hTGa-h-v>$J<0ra^^GTUQ z8QuoRMWR?wwzRtFmtJb;SXgqx!gi7jQUo%TdUj(*XyY-5L*@C{Q(`@gsBG8%XmsP8B*@0TF}h) z3^iL9VL$E{iITduVmZwMO@H7cF+$gt(+;7BU>Dn}s?!a`T%CnN!H!qO3N8n9)rN1J z#T2d8hNKXI|L!ad#XolzCYT|a9xEgYT`7_fKP33cSYe-=e0~x9I(RGGBIBG2-&an*Yx17LSNx(%9xEP>EIrQ;7a^V zSD`1ml8Px@i4(gC+q`#S91~B`__uCgwOtmtYLbL&ywKzaI&;{SENt?5^hdHV%8bpa zDZ=H#ix#NQr3jfmcrH(6ylT`k(u7Jg)c2-=!N0RW-9B9y=7ajibYZK{qf(c!%nUVc z=|kI*Cct00<1&N-FCKGs!fZq`N~O@$fAywmOM0cu6R%WV$41hiu2w9*CR6Bxh9=QD z7JLqF#L?}j?~ZCmhK1em%q*dO&~SK0MuaU3vxF}KglRagNJxO6tMY`$ymx3?J8H6p z-G#M2k8<;ct=>mr(+GCj6Xb5;A)SpiTnSHc51WQ-3xuwA4yDb+&>qdr<}fWGg@P>j zFfpY_*yQslsfSQ%f|^Y7B%{a7ao^li=;ecEaZjPf49zb+g+%nWIT}2rSQzGm=lx<~ zo6n;Iy@V<=)Re=E6g4%xUSsX(P*pR>3%!M2W*9p4fk+I+8~O__>Ueq#EjRWN+6aJw zryE@{@da-?R*c{8D?BKq;|=|UJkxA*^dw6b9InEJ&-D|=*^7fTU(l?IC-fJJg+Y`p zD%goX7<_S%!K2}pBtz=ax?00q(g#4k3Q`Dehf+aU>*5&`ABb0uk^=G8f#A(3q~H`dl2Gibk|u`zTN%4k;s&)pI@5f_XVg3bC1Et|&*Uk@JS-(mOyTl-Fq!jZ-^}1k@J~+UF$3BXhq45P0?|~Z2NKn(GZRw%;sNR^D=5-r2YEY> zCHPo8SdO&VI@AVvA`?*GAWYK{); z$C3SzkK$l{>8oCHu5g%8PQiU&u}Kb#86}5uNTj`kVaX`Y4-u@ZEKf!ggkw|$+8pJ4 z&Nw-QgHzBb1PGunjgKWG8v=k0DX3WZ!UCXuD%$5yB^SS!imniTr|bq+3~1_!OT~~_ z{6|;RIh>52$vV_<1LY6WP#+y-2@gw0CHe=Rl*#S!v2?V<@8ND2dfK0dRWw#ns>FJH z26~@mpiJQqH@vQp8w5gNMkYESpaGI~2=wn5pfx8P5<;Ll8@&X>jG`Pg70r~$bVlfO z2^=V8__#a}HL78a0#5(}GQjQNe8JUY}1hljjPRgb+yc`lZadbz|+pj0IbrXZi z4E$qv)DLYiuvWL4p|~C|&PUzR^*YvcrJMk;_G3QkBB1+J*1DUGg;w9o@N&os7t6c9 z<5flI`j$3D=vKl0m=CMqge0G#XQ(Y{3fOpz;d@Ln^j02|=!ErSxU47o2|*zr$Z;6X z18{$PG3qR!&stIJ&R-Zc4hhlPxi{J>*nje27c~1)ABdyB44OR{8>qR-Px$h_=sNV1 z%IQxyupj!w4+{wWTR)U6IPj7F$VG+Lqmb0R=Sv%)q%Cr?kU^o;fsUT=U;WW~AWW6g zl@3vOsa`Z@m}~HXfhZqG3`B3*hqO1k5D=sJd<-Y{5JxuzuNs7s(2(}j9z*b}gV2p| z$L*o3t0#@`7>stIIu)!AFBpQZvDf)irj5eYobscbavfec6eXDf(KZ2F07VW%*+7>w zs6)va1c%u;4Al#0e|zhkaC8teHiv{rymtXg3_41LaVSqwHfS?)Vu49@AoT7C)F`0W zS|N1CV*&}D`S2*@5$vb@)ut^5Kt?BiMLp!gqYTL@{O)L!65|i3IfU{6jva#>klW3{ zmyAK#W_0QByF&+-j14rMV^LRhidx(RmM$hFSb&}#i(Do!bkRdyzR0(Lyk{I5WClq` ze8Vh&+~ZM!ms8;sJ~AHVnxP;{#xWCMX~AuQXY~X$+6;2fv*KK|)B@zwbldY{9|3LYVC}Q$4D$jG39i)pMX^M%@9y9i`&7!Wr+}1)IKy_a zjh5zi<1tfF5!&5>Bmgr6PfkVq(MuMvd#0gn0L!CMhne#SlFfKkoF$-7T9GV1i}A!E zL9%f((U(G(AN(X^zWzKc;d)%Tk?104GG{VF_5IVoT=trYbVP*_K6 z_4i-F@N!5<884KfBlg6Oe%jwU2Pw$akpvWN9LC+r;1fHlje`l)@^aLK`cfzw`g$ll z*bUMEFljMUNaLU@#Z>yqtPxe}aLrZ7fv)6o)cwE^S7N(@rf_68Dd-5?aYxeCk`nj9ccTjp)`^L|7f&R&#I;?M?0CUc$3Oxt}sdQ~Sn#Ce>A9`_QAU(AimG0tj@{ ztI=}%OMbe+k5;2}^aj_+!SxLOC7f7;QqW5}9q3BA6u{QaHRx&dZ7XaYeTbpufWX!x zwdfW5MSr&NLX?c(sDmC%1)0)5@gW1hi2q!OhM|k9(J$ipbueJ@#Q=S&?RBU-4qSw6 zp2jJ$S{K@YC#)XE4i|OO=}PrXw%l5e1`0wEKHY#4;HS+Z7;hE@8$$~y?p}m?;z3s! zDZb4)FT&jxLsy|FnERBj)Ijm}#poj>xG6>C4#o@bv4!G?m!TVV&vyJEhvCL6(9ODM zrwgPoJo-wsnLneWI_GleajNLx=i}w5KE~}MK*cKGIN2J^n%(!f6mI;&RcJg%)^`P3 z*c!5T@vAE!=pM2_)~ykZ;>^-Tcf6+&#oGP(Qyb_1Ipkz?+ydK_m1vY1B;6+PsRdB+ zDpa6jVPO&IaL+2z3hOMT*8rdWD{hHhjY0*~R<_R2X9C-qM*fC1s9B7ZY2-)9MAh9$ z^1P5{Qr(a|@#(A4H2nV6s0JmrLKlrUK?F&2_cf?a5b|)`2Jq%Q-0wQH$9sqJWa=$R z*TXO)Pv&lFzrq~VFV~}kf-q4Lh zs)D!6>S9UEdelB>D?Fpq%bDv@cR{#c<>G$4_9leAcj$gq6;n3AF#CR;DmrsU2%1Zn zdH!_+x}_a@L#CSqy0B5M>C7`su-pPOkLnV&FbY4i4LP!ve>=yF$8SORXK0K3Y@wLW zJ-4lIsB@QAH7nl(_f2qD?yjqAR?fG$wxU>+?FIUG^Q*<$LUa5M<|ki(m2^|9+CuZI zV24N;^{Wfb+7%f!rb*V52tthkXWpZTi&%VbrFML5?Z^ z+J>+kV7`IFXPtn|*_TCSwKk0&eqG)DsHN`yfIH9r-_71vnJh+xU-#{nK=|JV{#3i?&)|Tdecn5? z-bV9bMqtn;{s^BR0S)h;fk93D8Lf7-P0&`KM>n_id!)1r8U={CH;#)4(%J{^jTNqk zf`YOH;iSseNjy;u+GIcJ*cG9Nqi9LyLl7ey(+_DJJ)zZ_% z9oh%&vIm6t#n{pIK&3;7KE`HpLIEMTxI<8X1iyqrKhlipia7;knCiXs&n)QF~Ek}y+Dg;{+ABR}cVg=uND1tEl%hS1T`Qhb*^=r->i`60VER)hxK;Pa?| zSkN{0r9K`2?ezy?-~qQ1mCy!y3~8Xp@JacN2N{b?@t*LYI6%wYO;@VdpY^`3mfjIT zxC46pl9cwSL21xxy6zAiqyAII^=}<(7B!I8@lZlg58*#Gsb%Kva2sfLYwDVt+PJGr zt6-mC=%||VMGM>|H4TGn7FCxIQcK&q7nd$r1jR8WHO?wp!%6CwJCo|?RL-w#s9jLg zkc2ZQ$aS4e5ICk~RpA{Sq)zzt335X6tuQYEk9y|jR26%-CAuoBYZo=R3SHSbrR8OL zd1cviaO^bjRF&5A8uM!xG~nMR$z5>RB-s{{TfmBz&8n?)m*cESa(H~z zf|_}C?s=q);>fGY@cvyzDpk&Dz)ww*!#Z9*tHE90Fspt}=>h;fJ4sF(VGya1VXr5I z&G1#Z8|o_OP!4@b;T?{ka7t+EGD=7XpUyC=e47?k*VI*&E~vc1tt;)t{?4hIP`AiU zxOO!b!_%I**;N)E(8Jl9^EA8vQ*#W79CjsEw1F&+-~YQ^x?zcZ=esN$9M=+-al>It zh`xCiy&Dd<(%xY-{^$;II^H@!h{1byiC7G^(}Wyir+xU)gO`a~gBc7Zv?qVJ;_^990(_eXp2p1~lVdF-I6^0rcv< z;zTc9_`$3G2V@Cz=0KIXa*qu|SIlbTg;%GBG zREqT$cdZWE;1udC4blg==;55k&h6i`@}Vt zcurd2nY~}^Oz@DFZR37%s}Ih(2gK{(5p9*|rX-#5gYo@Wg*=Iy2M7sr(m40x%KFL% zct_M(v#1V-BnE{$=9JdEwU;@~Dn!x>*Jv+wD(yn4%eDzsUG_=tX9!BD%jO>x;iaEY z9P*%;06(uC60^N`C^VD`QTTw^-5yFX8NHb9Cdt4X9`}0j5sp0+fA)Zw=EalBVNFaa zklh00ZPth;Aq)TdpxBE}2`zxPlW(!(*Jgo z{(iTHypAi8q2k zVBqXx6cOdEk_c70&7fN}Bp?p%j++JuX~QEXRJfh)g^NlX0AIbcp~hKXvtTicB&%W` zN52x?JCvWjar7qF^AAu$X@swZ>4D$;UZOsyO3}eHiw>g5qF#O(VlpNy^$-G-# zv_wuTc=F020z9*ihB}))>Cl0W#6DIEuLV@e&aJ@BwqZCm}toeUBtZBijkCO#gVNtLYFhQ zS%HsxPAt>Fg)o{5&ORqrl1FTSt7Dxw3QGm*dgha1)~VpkA{NRVlC5XH&h)0GLRIs0 zJ@XyT@r7F%me<4Z*r_l@^?Ddx&@{n77c{@O0^U0f-3q{5HXUxxAzRUePN(Io1?U;Z z6^Dd07xTKfA}BhXhFT=fJ|nssNe;1pV^Uq0Ln0a~+%6W~u7(9BAvZw4rXWS10t__+9G%NxhUHw2+CRSjLA>tw8| z`WI;O`x9bC&{$QIqj~ixv+7{7yWbGo1)v$Nm@M8dX)O!ki4h8ZTO1rzqY|p(gmQ{Z zs*`oY_EL1q6K{)W1$1RA)cVgh0SB?#zb78I-{7UM+t*6_blH+(KgGfi-7yk;{tBQ0##Yw$zf?0K)7j`912kzmbPU5Oo9f6!A0*>IS zcX{1_JEaDodX>`%ki`53d=#%8DhCIEThP|bp3=Xr(mmH%=dP-(hG}zWO*M7Nv6@S! z`ITRiTUwx{uKBF$nxU6-B^(6}5!b}>hvKb4Z>jn^$z5Qov8v`VEq{CnFXakn@WM}^ zzj+1^{aD=Uy+dchX%>w+Ew1u;bXB%Z#2=j&qhrqaS)VnN=kSp_$sx4GtG@R3NT+ii z&fw}Z;vl;}IDYnw*iDGE0805p9OA>vy`PAs{E<2hJMdF@ty`5zlS2GVoJSskyHWqq z@j_Im>O<#0DP_c}(C<;L>H?=Rx`dezsX%&HF!Lk4YN#9$0Jfw4Q_#`S8`8YS=P659 z%9IG2kSF8P^CEvkBGk=wI{|9YKF}kdi5CRqjvy@z{NUqXi1+YYYTScgirai1UH=sf zl2o8rHz7uc8N5Tl4D5aR?M7-V8tC&l6F~0Cuf=jZ$H1D#?%%+0`!*}6&EJR_lv(Jr z(xj?GN+eS)1)G@VZ^bKxhlv{E$b0rdSr(g^=}Y#*%l1MJawrB2pMDFQeV)Hm40HT2 z6yt4Y#W#hw2!=eE^AH})=G^Z1d)H$7S?XTh(`y2vi z9OIw(Rm?(*sP1UMo#eOxs^4#7xfe+o2j-EBKm1MXftswqC;cww$8b0;UVY*n-DU-k z{tz$M$v}6aH2)#q?E9?zqW&38L%Dy79$t+;`uk5vLrf#|(Hnn>8~7tk32gt{7ulvN z$rrGnAXNb%jU}4>b$TN<@XP*P>ZupFm>0l|xWNm!GkL38JB&}SwhfHJpKMQ^tvq4S z(gt^RIqA#7Ah-&SA#*lVlvYC*!tGpKT31CSLRX$K{= zr|^p)HQ^>CJ%}PBspCaP;&}nm^>E9x&ZkB|Z}u(5g@MwWC<}%KRxd1{`JA~qp{N2v z+No+I%?#?Va@QAcut{q7??T58TI%7|p)c%cE42?mC9M!b-Ay2Y(6yrEvCs69*7Z*5 zPSf;oDH%3Y5b`vT?9&;NnUVM&NlHiX%Lb0RVgq;3fo_M;%2v3GU&9b{NZ{`3j#7Eh zMpX?raW&+?)>vDRZzB$xe_)-3V`Wsq(rtsBh9 zyn;7|OM?IlH;As3>j{`}Mo8BS=&~s4&zD8Dyw*t?PVRZ=FXnhTq^{LSaBDG1hOQ(g zT_h%<1L$_KnB?KltG5!u-W1T0&xJue(iDKNjFP&GU888QCPdL-#UDjU?Sq`~jJ6FI zqNEW5%2q|nj>2;sQh$3kfiSiWEe@$z7=Yu_ye_?h3kpzh5=*4`F*&>iMz$1J7{{S& zBG$%8bA>5ZSTZ_GbIoYq)mh3C7Fa=r$4aSO0O|@$W2MPvh)%~!BZcd&5b38iK_(WS zI#G$*SgtwQMH(5jQ&sNm6e;a-tl^sB)uG2x>y-8gdvTUaGJh+2bro|-4hJE2RUBC4 z34ep|B_i`V>5$)BFeB`4Q`n*x>^P*2jis+_K*Cj%*>E-gfA%-8aU0I%15|c zqBO`1Xx|{IPWZzD$kkP43&5u?awOhEP zxllr*B`-dUHLOqCL0HG-+YSanvKlxUzsef@BnRH{DxC?3U957dDX&RMHEH(h!WHAh>wsyiQJ zF2n)B1=qWz$L+hksC11THQ64{@4+(+epfVpJp;U87v3>INqJzr+S1OtbR`T?g-0*~xJHkz%Cx`p>k-oLx?ob)fd1{}&(sbcosv+7V zd!EDJ?!fQ$l~y45C9ZinaZOz@N3U~DkqrJ+r{j$^dFb^-pI&*&aPT+kuHh-fUF+S| zWU+U#yBb!#;GkiAyt^FUS@U1(DtOOuS(233iF^*_Rfn3WUg5I51SK?4?dT`{Eui-u z_;g>`aqYm*4Un#|zwfUX8o_6|US8w40Ql*FQa(pQEs--wns0`PqGAqDXz4UQZ78AxCRmDc#G?ZCQAMWgSn5Rf+5D?R++ z)8nOdGc9zPAQcJ{9;ML}%f|@Z=D(0lAYo3@WZQzg8oetY3 zwXP!XylQ_dGN58dH{}D>jkv0KC!QHgRW_U(U?Jwo$x`YiC>vgbu|UvpxipYynMJWq zjl}DjD+^PkK*4@hXFpdCnkMB7*K+)yb638qGj5tH4My;Lpi@Xfhz0OAE8wf9Nt4V- zel`udd;6`RW=@B(zn7%YnHurw>C#&Ah)7Ahv^pl2#o$|xNaF$^+|^ug?QF@8-Loa9 zqMmR$+@^;s-ZvS} za!lWJ#vjadc^$%B0ItloiwyWzygne_G7r)I}eZ zZUukE9BGUh)eq)C``(*E(M<*hUYQ@Q!grKI4{)Rv_6WCBXoh{Z8v=8d71aA~DaQ<| z%Utk;IxDDkbEQNxsHf&a|LGblsJrJ$d1g>&=K-zFR#2B$NEv2Oo8dXSk3#h)FGQMa z4R&(P)eWM~k+T9FioX$C@+&2q5WXqL`*Mc1AK#%oZpGMo#oYC%ILEA@^5&aERnM2M zK%ZDaB`uKthrr!Q6b>e>yiQsYPin5wwes?un(D>wrLMV)s^^fMM!FRMZxOB@AUFab zE@|i%#15!|$yjw$5{ye#aHj@dul4?XtKpubq-as>Le<)?iyG|bc!&l<2@UqY7D(-p z5ROBdV7wLH1!vYsWijCd&d|L31e1^iZ{vAesS6w{z_+Uf$MuKDBWhuCv&ahSU@f$K zj0u|H5*A8bFM*MXRw$oXC=E1Yq_P8?9x~4g>RdNl>OWJ(!>@+1{Fv*cSUM~)YQRx1 z4fc^`OTF~HKQFm>dxLbXu!-=Jf+_~!V+~SbxORAu5$XCx5R0H*)~M-*w350@nR_sFIg-N5LyULp1XhHCfwfzUtA1*8Th5gxN=0uK*AEnw!ml1pjh{ zG*Eb5b>i2%;L4ra+!8Q^EE2j)5dS|_>95G}d^#_!oyh(5mo2G>SJ! zf<`78-bOb5FMS;5|nqhpc-@=cZpgBE<(lV@`VrX>5|(s&Bw6&|X2%ttwYr zuyd!B6l&)jVJM)Tf=6$X?h?>Faddv;o;ZB%Mws7Vw=~1JJ<@ugN3Le65*~5ogc(t5 zZB)xg%`jn&PN?Wl#Nm{irS1HdpuZVQw;7+{cQ6oy%1rf9=GgK6&5%7W5*c+p7>N1M zTco>DK)gQYV>yb8aca!NuOa4jRfu|+45eG7G{NqOH<+05e*89=YyoMQC3W=;I*z*?0#MvYbnH-W zqWJ0SFx|LHG0r|Bg^7I9p9GA0%R>cRTpb6$zeDOJ-lyWbH=cHiJEm}Ccg5ogJEft5 zt7Y3xX`*01;&0u+U68t7QduFQrg7*aIOqXrbxnYd?1CA|54nEyKwc801JpDef3#B? zjyK;aO^xwK<7?$}6d~<^8W7@&yQJDs6L|bxBb?eL+%3f+l)(Jc2wQqLjOzSh@s7K} z#NiZbgsIFK+pTOt# zNlDp<2`2`dJi>TW9)~|Q!s-j%utfFkz0%j>hy?0j!&C-HmUeNg4o)ENWOobm;?|Ws z5|5_`hI9*S@uLcUbkluOYM3d>*Y1=02&-`L{V+*h!)4CoBW%K^9AtC8@9^?{ zhdj4-|n?TWaYlIclrUzlR(*)|z2VroVy(3NcQQnS!4PJc{Ff!r|ZF=-6?fOF7; z^ZgFR-PK+})U12{rje$u_@{Io-91Lenz7;bEpqXnI%LIcglh zsZ4iJ~n8*Gs) zUs6}u0Q*^;)irR00{OwqtwZTZ}k465zqid;=uCZE;y zzR=w2_jV)vztZbas)-XMH^SLt|K)4Dmb(6zs*Yf%oDbqYqI9KAVWUF4NS~d!BBCsf zqj_j6HJG0^wNk0V5umJLIx5$`NPKA)|8E92*TrS{0{NJu_Y8i}4N1CmtG7;NB&^gy zrEV4=bS@2G*bdJ7Q;3GZ!w8)zs(~&cGVXZA$Z%TtirbqfH#?9j!Hws z-D<4gqsDsZFmA^?y|LbM=TYg1AUuRa-iMi~hq~egb#TnxLq67pvGo(rz{c9=y6VT) zBRM(%T==Z?F~>t^5Y{~>9UwS~R_GM74uUYCY6Vi@)z8CV48h4hr(TejptIDpnX#K<0R@Cz@l+#`IYO7K-zO8Lx7FkfeU^w7)FdY?xtUXcn20>+;X zKj#m&X#VvI>~TuM=f9Ei@=f0f5rehO$sCR+(x&`jSU3(R-?Vv6st79TMl*c@KK_+= z7y5}T6L28=_-oP*_`VGxYx<7OnaL(_Fy?vUb!m-#A|a!X&{gW&!Ij5gnPw&>t}bn? zU`Qr*!^e)no*DRMD!IUm^`hZ@$EEx2OZ;inz6T@FH7BHWeC#VJgHrgBqgc|-umKfr z387A;xzp9s2;s7G!7~HEYiXkwEe}z*$uwwh(|QxEJOp<+9~>2Qr>fXHh{E_gYK~tx z`P-p}zJ4{)peW`t9n6|6GhBy6F2(%LHME~ZdIk2(3FH>D1O z{dMmn-NFaF8qiyKhg-z{DnsykH(c^Itf#-;jZW0T?a{ZT=irv=jh=q5tIf&jG7X&d zI8=@nWW0E&92Wq}pvo+X^xg34^wFwy1(u_gc1hHg+a#&0kDWN<2ug^i@$q-0_w1ph zo{2A#^Hbqqosjt*)b}BxID*h5+~Yl1>ox%e7wa!%T0!Z%!J$c1ZGdZKoV*;tFY{|1 zDZ(qW?jP?%Fk?KQilZKoE5I2h#AMGF2h7DRVUC+Xoc-QKCa=Lav1-~rXw551 z^=E2uLz9L_tO!s#qsuNEA>bngw%Gq0*SbZO5PMx~5*M!-N~mjPc9oy8-%09Oxz=9p zB$GLUJCktX$I@y9zswzU#RR|R4lSdSI}Kn8h9NLSBBbvrsCQ!q-iKK**dgV zpJH-zXmI(oFQqcnCz-^uy-%{Az2N|y2<*s|vxJeH{f%vGTHN(X#_KZWd^6CSzJi@~ zv#g->sRWP^DQA(UN{5^Csf)FM%JLlEnImzfYy`_2LWn!G+IQ1OP~VWjmDQySobILY z0w_KHhV23|DdFjO4bhE5S(U7&7OA=fLiMpc-b zxG;6e4BLMYrbC;KQ_!wU5@v^~Fsny1^4tG~FhkpREJROnDK;)9dFzYt@B)Uk*7gKCrWT$+Q=Fl_%pnl|k_b`emU)Jq8$lxh|(N55h8 z|29gaou&zg7r~-?^!TPozkaBe2za((byz2q*YP>436ccOK=}N*;2~m0DovN^scHjw z!XBzQehQS32C(Id^TF`GaF0~lndm|4n{*yO`YSkD7=lA=vf3AVgCpvZs;&iDfp#IF zQ!PQslm!qc?rm+^6A@U$?D{EqEn1ms-IqF3WRP8J zm|G%#4qhOjo2dNogYWqzxUc;te?36IGzy3Q8k{2Rwg6Aw7=}nEPc728tp^QtbHFt`f}>c ze+Kuqgrqh*X90;{{WCbVRZtX=#+Dg;i6+G;-xGcv3TK-&7~UBf!Z-@Iz+sp#0TmEs z(49>)zImYM!76ZQ$r_`E3Z40(7xl| zzk{zAhpP1kQ+8>OB+6J-bJ0!L%BRtUG;0rye~a0NkkC97KQG9Y_HyqN-84ym1Rqq*Tfx5~$y0^js4fwp8`)mYvt%O63Q4C9a$hqjvg;fT zu!1_)fsu~o67Wo*$?+~25x{FG|wE6<;%tG zYW-WO3TU3GH&$g{IK?I(7j{z_O=rnhKV=}n^31vrd8PePLdMtw03VFPW9)JuF0#uh z=xJ_*jbAWKkEY`tcJNjBWk%2y6InC{Te3oBw;+6A&Ed}Aa4k$YAPhBs3X`|kzxC&E zN4T7We&T!`?9Sl7O~>{KIUd0;<4ae}(Svhb6(RqEf?YIxO}*pjbcUA$8ofieXl32{ za3lMB2TC4*w@1p+onTqF64qeV(|NUFqqi{5SjdNrX)fwjDK2k^Zw+I(8mUDyxyrWo7KybSe^YJ62mEJJ9)HZp?E)JdbJg?A3Mtvd{C~6mB$iU zn0M$TCOTu`K_~9-^j910Ytf1Oy`1d(Ty1x{)QNi&P&#pc#0pBUnLAaT11>so55J6G z9VyN)cr%^2KcjlHX&yMU+u$5bP&#q{BN5c0T%gR+baUajNp+w@C+;uHux|9wZzg~c z&1B+Uv}aH{whYq5O(Bzsc4veZ{9GPgkR`*WZ)MTE(gyaLc!PK9-UT}qT)LRy1E^%u z2Xu7MM~ymRb7D=nZn%`DGLw?w?epn4ffAaF=^T3jsb>rx<^`xP6(xjh8I9ry3Nq9= zcKBtM(-n8~VuWdsV(%RLlvY$4KZNUuK|(*`uVh&i?3MoP(b&#Rg%57marSbm8J5Zn zb@dp2>3q6B3KnDDcPZ9d;i2>C{_vR0r=zV_Q2P1w$_%$pK!~at-Q_n0dwpxtE$7_TNu9L( zRiBBUEs!V0_=7U()M&yFO_9~T0Y>Nzh0t}W&!lrR{zynjCF77Hxe(oHMPDCRDk6f9 z_}IrqvZr;Sp0Yx!cU&XVnLXtxmPkq80=;jAba=5m$t$bI;=9Fiy(Kp4RliVd^u<1d z=9t+Z#$WZ4bA>iId%m24`@Uk692N)EIux&BR`ixf`!LeBk31P3aUatWsB6 zm4K47tox9wvzdEvNMqO*TuRvTD6|UOL0L3x2V|*XyYX^HxG$8@u&wVWuLv5i)*DMy zq@DulF;-;}L|4CNslu6dDRwp*XsCeD)YH|kg({sIPUqlVCe^E$1W_vfc!s|*}#dOPxTm@m-H2mz@wA~}zMsaTCY`WK8+jy=k=|I4s5w+TQUR@&B ziD@eC6pq`$ad$J~)|LDd?yE-2Ka2fU+g-EdubH21wq6teF!Rp}Dq#dtj9;YG zxT3L~GJ?AB-E%LO1^YU$!E`gTOQ*=akw;Y^83*Ki#L%zHZux17oM;M`Hcc)@cUgcL zCUn5G@8z^O)vq}WwXvc-#(&=$i)&7AJge7l23v8 zW-Qno9Af}o zxu};F3i8HA41FU-#}E;Q1!z;5{A-tf0c&@)`$UgmBZo$Z9xo*4y%OU~uTv zd2+3|S+(%ZlojgG8(%c34n4ZB=gE}(1lv0_n5r~wgr+eD|^2yM&~di z=i#b7awnf9v^r*S|L#O!GMwIsYo8Q5;T~9$+j}=A8=gjr%Fn75e)Ng62e~rNLk${d z=jOxSMwIIpXLnc0F8sp+xfg1m%Yv(g_@#%&k>E`ES_*POhax~`Ud#WTz$RSK}`gPn-eb`AcO>f zm1sKcZwFgDhmw7HaAtm0DSUKwsg_#Jig*uc5fL4RUAgK2#5$y$DV>S!#KaY_6YWViKDg-KHOVcrCHYZF}LbZL@5sC2d`CsgDBfJ_2aw3_7HQBh4A`@T=qp zQGT9v9HaM{Ejb{}56xO3KPjNWd32g)u)kE~(~@XB=g*EDTCyLUho4?4Pq3RnrL2NZ z+?~cvNhEtYXP|n3x^s}v4p7SqJeGbS z>T)RO^Js>wh~}#K4oXNYX^llwr=f0gjyrUORpw!Ds$s*aWjUuNbf?t*rs|}_vKS*l zZ>U`Aw$n$hmv0P$@4_DcMD7x&u)%;*2HRGZIdCDLE*b~FAfm0i>9{O~MM}r+hHadXYx3WX&Pi=vZ?U4dAyoZ?PjA1HP=~YBjv}GvZMShhJ&*?}fUO7q%EYbnws|;SM zKm6g5?+k#I;7jp`8K(Zwe+grDs<}Vpcc=c)*pxvQ14~^7bHZkB8WY~H5gJsTnarv?exyZ&$zhR(U@DeXHCkyhwyHf1k7LS%Z8k8t8nzd#ilF{;t1>w{C;J@EOi* z-M0+>-R?MTyL>HzU)ECTiuGK!{Jve@fZ($nmVB)ajnNu#4hXvcW{2ESu(!|m(>-BK z2*V-YdUsW9?7HIer-T-n2`87rF9j4`mw5r}_`Tl`^~>P}i_U zLl=%5tjD0rW zcn(EuCBb9s;rwf+pR0v;H3q|G&_Q+rRYI<6`V@u z&(SLn$(!J$Lc&oO?L6+cI7b=;bde%jzJFLQ5?VfgSRN#xBRURs^`9_us4Mv=4v#!4 ze-1~7`*A2iuEVpRkjG*93HeSqI^3YVcQ~uxQnM|)V6@-Y$I5~ z5jjji-{j*_N9AG2Q9%0YqoWGIdMU#VHC1~OT{P)WD^rr-$Q}qSf5KX?(z~GwWmvIL z^|U+=5f(`9qGs0-{?e#aUCBQwb>egKZ-RX^5yj|;vtN)OL6fMYVcrhRdt8;H3tFbV zD1VAV{UD^J(0 za_pG=0%|sh@&Om+#)6jjj>|!9(2Y8Vbmi-e4CzY#iJ_%$%k}{A5yFr|X;BSJ+U!%D z_ya0(ZIwq?vLRnR0nU4bk1mo|=Hr3y%hv@!`k~DOzWBa8LOi9C`Jh0Z;`x}sE`7pLOlCpXhzjr0pSC2 zwlF;8w44xw3XN{0JEhjULArAxi?Gi)F+ln8X*oaI8Xk{5BRlYjGjb-%v_hfpA^}Dw zm5vZmg(?-a@o}HP9`YfDw2hB5Yva%S$@2TV1^A0Wwu%5yE7f^cBstk8Yx!rm$qx_^8N82+@vP<9qdRYinxVBOx_qi5yb&?i06)uz7FWfWXL zbpm=IZabS~30*C2FJdjK+a}VVGOMM-^m#SMb!TYE)=d~X{WT4g?(t41F zXS|PCR74%Pph$J#C~m_%D4}WQ?(bll9zKaCPqq&twRPBjfCosxA;%D$|#si&Xyys{6HlIf;e~~lIa3A>v za38b6o%}1Tuc)}q1v>RBydU(16*%!vZQJ-Dyd&GD?p)IGPb|;!!ppyrw&HJ9oqreM zseeFQ!EUK!{Vu}Rt-m>{1nvSH6)Xt;a#3C*AZHJnp-q>#*^P+F_Si%AKOnqEC<)b#A3MorEU#&#i; z(6s;FUvh&uLajHf2WchgVribSD$Cp3|CZ09$*O8h^LGC)4Ky_SJot@mnh-OO&@sd% z%hT#6P<-ETo8);UJ(2FN9w*0Y&D6Xe_%O0fMe};lc!AsE0NX&jKSI1Qz?Ow>ppuv% z5dv*9ITGp?n*(hg{s@n4W4oW<(i-_~ZEN8X&5?9zUB?(vq$;G&{vl*}Ml+X-nu{k# zQs`q#KVJY#(G)o_)VnyN%^l9EDQ7cFre)mtuc$_SNi}}a6t2wYq2yl-t~1)%3Ifon z9&~Wsh4olWXL;Yp#|PS+@GX+l0Ug7_Mv=%_w~UKzwq*P7j0nq)i*2?H zl+=@T4(nzxPJ!SjHd`&nM3Faz*cO^0lI^xsG{6ecQoF4$M?}As7URziiv@3|b(R&n zm7%sgKu7(tk+j)5wlz&ID%1I6sCT(Tn;oTI#68kmtfeNHAqOY(^GWVNe)*}lZE^*i zM_FFuTmm2RYpAL1qopVR0`8z!_XMLml+``e>~tm9eIu06HvP{~n==62)RPANO+E2| zPT&Ik_H+S!ks&G4w%PbB0{;v<=Z6Bw(;jZFB0s9v^+q(H68XIft1&^pL-NcqU2%-kR z+#)XlRlCY-Tub27ilAy;`(h(S1}pd(5tK|5%LK>x1(owE8Wt>bmRHW53oA?24bIY9 znAw8ULN(*HT8G=wR!r>}3`3ipQn%2xT;%ppLhbl=tSw8h$B>FfeMLEKIjCzf3&0yV za7-~?=d=}~m||*uxc$;;+r@9G9iEM|Erdr@eYD}yAt2OM^})<-W4!GHgT_SL zU^Inuu%2@;rWglywaqqzTHMvv11+$E`o61eff-a)H`_qeXaz<0RGDy?o@7fz8?B($ zCz*37B-?tUJFTErB-;i@`_s}=5DrPPxp4awTR-$LML{><8&tz|y9#egvAuwvvm$kC zs%^TNGNd#}@9$bcZB4VqBQO9PPV0yp4BIx{c9&qk;BQ#Eu<-}SzkwU}LNTVFue_kT z7TnfnK&#JgscT)HY1``aC_T%TX-O)j7xQ8xsdut$3w#h=m2FFaM^rkx4y%LU(p~cO zx13{pBM^=4bt%)|CB4k1zgJ()8om*|3{AAQ4%;5W8>YXfG&$h%9(Tir3rM+lIDcX1+V{Q?Y{EM#3mO|SvZNyEX5p@6mWp{7r zZEK5~E=f;oP4%?%1Y_k|qn;*uyE24BHOcX=j22c+!&6fSRKWgoclA7XYHAN>L)|iR zs53dx*$pd_&`eh@D6OLnuoG4loTWGbPu8Z-PemuD-ZBawjTVA38>M63?9+c>r$D`Bg|Q-(Yvb_okEZx(KMg~14 zRmYH~T3kEGrUZaj;PhQ$FXTXY{w^`Lt#%?~FBIGIzZD@h0{S8Z|J@j}2L0Zvd7Pi*WHd z-7=oE@>TAJikkA~5+xN*RZ^13r5haBq9nk@d)7e|lqSG?l8TdDeuv$|$}ezf^6nFD zgyJgLSfwcBa+X}aCYNtu%Zls~hlx97 z3b|y%U|z`}m+sI_S8~YZ8FD#7F3X`Oq%0$sAV~C$N*lN;br1_m4Y|BVE-%AnQIl9* z3IZ;xb=Q;q6iQL=%PuQXh}lc1$xEoYOQ@+!sF_QsiA$(?OQ>l}s98&>NlU0XOQB+K6(iKxh=sq^*ceVig~o=PgbIxfSki|-X>bIS2%*7& zLg6nO96yl1XmDhZ2%*7ol>9}5V?qS{MT27;i4__g?MSS+K&Oh6#0rg#TyjfegM4*f zjScdNaT*(q3i;--8XV+{q-t=GZ;z?LK|Z*o1_$}pvW#ZC7T2k>wW+u|<|2)Yu|3x@v5Zc`G%x x$P9)WTV(90h87ulsG&uABQ&%W+VnPcoLf_aCtVxzZQCI5&k-UXdR<7x{{wuFuhjqm diff --git a/doc/LectureNotes/_build/.doctrees/environment.pickle b/doc/LectureNotes/_build/.doctrees/environment.pickle index a1fb76610b389ca5963a33e2a61922215b107704..2ef5fde0a2241afc058f2daa95dfb28d86557b2b 100644 GIT binary patch delta 7175 zcma)Bdsvj!)_1SreiMWNub(5>5kwF~5w#4IL>fR8Q@rK~BQOG^GlODfC?;a!Blgbc z*3UX>$Fhrg^t|O*X+ot|UQsjCBqb~L^OSZ@$F84et-ar20zJ?3{WJTw_FC(=*4lgR zJ?}g7&Hli%%>ldl$hm&@qEd5lRk_*O;8#>?THr9-Qk)I`4s*3*uGKoN(!0ysXkxUV_XTMRJnIWT>;jx5!%QFjqP}lgZA80K3EF zD6>1tiZHf!S*62fRcDOy@)~|w*QHBV#`y^+`GE9&mUHdW=k_YgX}7fT(7Mf*R=2dd z_|kZF!^5BZD@i#wz;aGW&bh8+8gH^zH96;N#)6cnyw6yO@_pW?N_~E)a(2R-%wJud zpUsq}Nuf&H#8wulG*8;5e3uuh+6s<&v&#yb9?kCrL0@Irv}Np^%BDZgkj^w?9P3LN zS|)ho2W4j|CPbN8?1Wm?!!sYzaqkGGR}FKT8UIL+=wBwzcx@D-7mFq{U64_y>^UEy z2AjtLPwj!zr6sp?m?BM9%zu%w5H+-7Ck+QG-vJz|7j8m{&&g$xYOH+=X+Clsp`me6 zEz^Y(NZ(lqrE>h>phNuhalgj%-r8FKyZ={+=TT{a9uDC|ar zdZYFv&Iqa#&pc0%h-&JVlO3x+NJI1LHKe%pbcAYJa|0otSjQsNIctxScF?+;H1t^C zMw+HIs0md>L8@~HX=ZF#LBp4P8x6}hU8iCG=3uIF$qUR&Ij`!~b=#IO=BI>KM5=xN zvKFEG#&!oNi{H|#sXNv(Ww=YPo`2;fro6mguRizM38rj&U$0i}MAh_TdNpd-6{eJ* z)~mt0kLr}&=k@B})n}P9^nzY(cG&eD4FAAJSw6MSoSV?tOnNQ^LN9QY$_@z?8siQEJBF7n$;Z*Q3;5j)Z~E z-{>tCF8Y%+P(Hn>SFOj|7|!~LT`fKFHnByYdK8$yLYuJs^i)#&3ibDAn!S|jpY-aX z^H#U z4*BLQatgZk0}Xetp98+h7kzKsJc|<|`kC*}GTto|C->jNZ4Ho{*!aUL^6380IbKTO z9KE{sr$(k+G3nJIzieeJQ1r;RzpiA;zlx*O6~E)Y1}Y2QiBuoFn+Yt?g>5nnl!6ky z>!^;!;0$Hv()q9r!nr=ewab?YB3uAhkw2Twbb)dOT;~HAHf5k<*b(Wn2eE3Y!KQRU z2n!`H>^RqeP}Ztb3g_!xJtNpIrYv5lcU{sm?1WEi^)4R8aM|8|M(;BAU@tJmQm1$Q z-jiXP!7f}FJW#=|-2Y;e;kkOfYp;QQ#dN`Pe_Xqx*&CoVt=7A253n^fFE+*%F@XJq zZX&IiIOdWXO6&GO_NgXaj&NNX#2z{Wf7|+2ZzvDqZ8N}1_lJ$A+aQeR=YyZCHW<sAcBtQoxW0E9EXG8Wc?I*@OxfkIv$ z2jlsF4}e>I`#@ODyOAg|Ulk8o{LMJnrx?B*!4D3CIQU)j8Iu4}ye$jD`IUIc=5HlH z8}yBre76(fp{M{OGgn#GdEqVj5?z(c2=$KV6URXY^oZvft003#@@+XVg&UGEEKhWY z)g{9?Y{C|h0zJ-<5{0{P`EL=(n&%ubJoO?+{dFnco_?EJ?pkJ8JC2WWn_Q0_Ov zY(C&oxAJ3dH;Q>R7w3ECxaX(zP0J~(oL^R1Vi;puVz##}+hmwrYPQ-QEyjsw^tC+A2*I_zaaAgAY1HpD>r`6C9E|fw=i+ zxiA2~@9yTpMmVH(-mCGV^WbICAj5g%d*I`mIRWbZ_}*!_&@b_c6LEiLOov#0xDakZ z=vaPdD)b2p9c#1N?FoxabAlw_^szf1!YgvIdJ3Zb`a#dLU_L934MNo2j0lkuvCU)SMyaJNDBgg%c$rCJmdpQi@`5Ab9XI8+74LvcZxForw zEQz} zJ+HGVC6mmH%IswhYImmD@|Fq-11f+KNEU$CB96YG9Mp)dyVIQ#%n9Q zzsb#(;9=8T+RrV&l7%Vh$axIwLWAdK6VF?1gJoa;nn(*biaXzfCwtL5voTz}FdMsy zw`HTRc!4+ST>}q7l~;69JX+|C_>8#9X?e^o018iN0zdWv^sM=kBrhYYb6UIx2?z+| zB-A0ylh6yHT|#e!8zuBXctAp5gl!V~A@ofUWc~;TNf>}|qJ)767YW$i7=(DUoDN3V zBw+}*eF(4A{5%+0d{;Uxy^;usG}9%dH0Nqi6sbsoa^5E)<$Os(${CUc2jx6KLdrQ? zgM#A;DNq5=Bdo!Yc-(N&nl*)yR`3TYP;F5|1Si&3)&7&uF#2LBjW+sW-s1yNj}&NX z(V*a4=N5>sNeWca*Ah|@0YP1A$d{@Z93*I{ipdgE6-y+h;s{6*mNmSEj1ES!rLXJe)?EKYMSpP z?9eo6S`mR*L{>M95{O0AI4$K8Qx)qaq$*SmilolF1zN4&-2x?*7%qsYz5)qptEwfW zR9}#gQaz|aL3vdQHTXn_H2x8S#@9%kBc*^PRW$^REw8fU&!G6Yq@^E(>3@#TB*q}I z2X`f;?ECwOS;~H#21Tf9DPWGC9UY;u)dVA*LIHyX-98D4?vjK=2YO*ZRSuAls?64) z2vX@5sGKcQp!%CND2#rP0xd>Fq@cw5spm&XNiN20w?K=r!YyDsSN-t^TX_S7VEQdo zUx-Nm9~u;--F@Y}SmA6bP+{{mNJcf5Mz=(XzvUJv^aZy-C+V(R@J6A(pI`=~4+`0C z!54)}x8R3@;uic-_)rS8oYy6!<*W(u7gV&KDH77E6-r18wNyfC@E0Yd20tPp-S{^p zq#HjhK#iMqhE?0 zh!kg>d?KfxLGmosIF)BvlEPm(2`T-a6w*`T)83slOA)WOJdK~TFmMY0?IiG6`dUSi z-c@m+AE9xeCyZIns;UOC*r_iqwJH4FQ~0KbP)LUBl<1G;BUyJNZ zotD3)@TXBE`wcDdCmP>tf!=T=g@4fkgW#hS9&j2`!;Yc3Cs1NH15f#B%O0h6hRAvx~n98ox)d~0c#|!(>l4j zIMC|i(8~SJ;?GVn)HUlYOb6ZO(oOuGb5I*Qbg1~_V*Kuyn`wH>Fp|AiXO_t$eQ zsVG^7(?{A+Ui=v>PHeaS(-q?x&D#IB_YbulZ|Y!8n?7GB_Zrt!wlA+)AFSlsfv zIqqN&)t`^v*E!+(DobeR3$FM!@tVEo%#(zHCo+Xg8G`KLOC6bpO&0u;s5M&p7ucvs+ChuJCr zehUA&vBB3?g%8v;mEQN&J$;Vn`S2MRq37dt;AjW3<(}i8Rfo#vSsoR8u8Yrae%n6T zVnnHQt|bQtERMRTgU7~$dAXWLkdnPTDyhNa!OvcV$`m@s9b7i&Od7Ab^p83i@qVAZ S9vAA939?UIC%(WRZvG8ixo0LJk4*vz$tK|uvdQvDAWKL>ULbEshzk)20wN%rWJz{Oc9-2v2;OP{ zD_{wtoW+5nLabJ6snWaZQj{7*r7e|b&WQU-|j5}=&NhA*S9p;107LK z4wtQ|L9H$cbQpYoo8RH{JL;sP6I0s@0v)lkZ^3xAVKG3z-3ClvpITZP=;-fo`MqwS znc39hgMbc~hQ3T~Izfen5W;V7^VhoFYkYxwT53C@TY=W$b_H$=IwxI?DD95u*ROJV z_YKD{EV}z&&K||O7}Igo`C6~__do7+o>Q#LH=T+I-Z%AhEa{dC=3b=E9=u4itd}Ks z)g@`jyst@|lve#$k|2Ft{U(W)Ub^9!^x_SP!QJyG5^3xFMCts3cZoqdy0DYP2kUCy z?hiy{SvLDDo@PoLZrVV4f@Ik(gq#c7ZmuAiDA9RCfA~Rp&U!>B&FMpQF8K2D*%4CP z=fz@LsdCfzF zBm@(i58&m9>o8@1FhY69+B>0ut(r$tgE_u^SosJ4alG8qx`{+2SgAzLr37=^whO~{ zASL+9`YyR-GP_0tP?LA_guuwn!O1WF6$p4kuW8*?{Yi3g#>*fiM*8kzYVgbBNfbExbHeODl~IX4 zVGO$ed>)yjICJC4-yw7Msq3lq`IE-r#@Ci$r5K5iUq2ctbq9^Xm%80VTKBXuSoPL6 zB9(@W!HaKSCDOx((t=mtiKEglpMyC%_Y9Gi{5~yszNd&voi7-JW8VKekruz07R>qE zdqf)Za#}F^!WkmHcRVdGdR#1Ps@ zNHN`c&KNooLCjP#zHbagM3MW5^yXz_=!s|oLka8eu9Q$+ELn*}yz`-M1A$aEgjU3p zyG1EXe25MpiBx*@iZL`gk-Qrr#jP@ihNO@uh}2^)oe^(WZFP%?gUB3-=T+;4{lfHp2%DQl-E7WiI)`>g zNS1ABY(NE#VHZZzPuR1g=yq0@OXq;Hdq>loSmYRbNLux~GWJ#uy^DOqrsvQCHhw5g zg$4(5=~9-pktVZ;#?mTQkwceDrnBYj>KK|!vf0`awQ0jxnv)V|Z*lI3gl+N>+$%RA z%EVS>>{J3Z&|zbxSJ%4P;yk*J%nI{Pwa{wVI|E1~-FQo4{mUanT3E~C>~*F^4HS57P0+(}C14pOVQ#ZIC#nPsw4vzX;x zN9)KIrTWe5V7ii~s?~-m)J_g4)rY3QbZ1Ogt3R3wf@ja*K@Ux%cJ|f`werj9)XF}c z$t&-kLFckNXYoqgOuCBuk*Tw21shwTy6r2XIg)jmnRQk|`t|gtUj|z-M{N~T zN%yc{RdMo5mDI=PU$0h9tb**_Z{?MrRMC~Ja;{owu|n%#&Eu69tx$PWHQh`v$uoFT zp25WN)Id&%DY|kqjSf9>1KkpnoVD2Ib+}u6COFV|d?ts>RLA~t6V1rJMupey^I2ML zO^$jvMY&zfvy|GBv(2r7!NM6V4wr@fWEmaCzF11Hu)L)(WTs_w8GCgp%?a&ZM)6o7 zc7`J>XcpVLoYpW;B25ZiUQQPigoC8X+oJjU4ArfqD@0l7wOiW0MedSQQm(GW$4{g?09a`^l6!)Q}owjn}7WNTcVeK6+M~XSc!QWId)Q zte|GM*KYDR+FUT-c2g_taJE`_stj9FKgvVMJ*&!{J;DsngCY{HEsSKrNHD{8*j?pF znvqS=_vDOF&llTifqNGl%W%+A63JfuiM-vgCmeJbjK}~z4`H@j=?wC19|}YIJnTQD zI_%9Zni1P;v&jycMsj<}Q)0M&NKSUpL5Dz;f%+(;L=-b#V4<$2iMUUXiN#EG)Z1-M z;p3_)7o%6*8*b281J6L=r5U3{eN?N6Iww4;GZ&-!>@M8mTkS5`(M|5vFx8NWUR%8b z-YA&r?Y=sEvk*4h|AC)S?Ne1-$8gnQQ7!J(Fzt4W64jza)$}S!-z$&kRdaZ+*byhp z`EE5Lt6szBe2|`Dmtfc#&JaCtrrXtOZp)6!*ZTHWHR(=56pDt}C(8Q#A@d)98F z#*}1zn8PAWC_K!7;Si=ChTiRhad$~cPiFQln)hTnIj=@a%*05t3+{q!{u-WWDf*aa zMa(ndF=LeM#b~O5aiUr02C+&si7fLQ_56|8T)xM_TLQ%Dvb(W~rm{Ir@EDNS%Q}di zXrd-q6{&i%Tn*W%K4eppE4*;ZT5qbc!<((3SZYhR4%~hYqY{V?A_O`oKzM*(vvC z{*bg+jX)F4?4*Ydh9HCWL3V3`?CKL_PI`^sZui)Iy=&TZ&4b4GHxmQerHUy}H^VuX zb1ExVf2gny(d#^-(Ruj)N#_Tu&XJL3M}47oqKE1=o!4lBH&kKyJSnLHU!a=oZE)}9 z$6jA}0$!tDjFzE@AH63GGOtDki8GRmDS3A70-!;OidY#+5CVcq+OU(B-SxC*wLxR+H;(@;(M0t>9G6ROm%KA+%-LTaD@j-IVt70AK@eP zV;qFUWySsawbVM7=JxBCt9i2%@2hjHaWHEtjb+1A=_FdqWtFm>sdVm;3F91|^{(20 z)0f__-}Qy_hsrOZmld;5Q)woJE4<(x&tIQp;d9Mty1>)--2rU?2^pWLP zQzJ~lIDP+~lKb~tP5)Szk2cdYA{yJ`rzPNYowAx88a3sKOB(GiQ@z`?4qnFi-JYr8gQBk%It~jRg^A{M z;BTpSTh`g(e+PHih!pEJgk%j*X*Aq0w5G|?(CBYkZ>o2!UTyc{s~?*OmLz=o5iaKk zTMA`og|aEb%s!db(#vY;Wz~#SiC*OyS>;Is-bf=$#X7vYj*f+OSgQ9Wr6S|M zReht_wAFMo`%4WiV(^bo6l>o^*AK;yIjkxCOAhM*{uzfgntxhiEej>zL0cj%MPs2w zgf(t+zWDoZr(~CYyU7GSm19Ts2DV! zL*yu8Q{MJOaYi)$k~hlCkvPguvWk{fLjDi2Y5ih!2n+sa0tLp1sn=+Xb~qH25?%t zz!`w;0uBfGGXXOJJ}ICH;2Q#F0sKP1Y=FZja5gi*nF5Xg*eKvgfZGKe1@ONG91ZZV z0=DM>{#xK%fLU;Of$10nutLDG0B;vC58y8Zv;h2rfa3suC}2LoK_#4FJit-`3ji(` zun^!z0gC|svZS3io&flSC@uzgMZgk(#!}v(baSAk_XDzG{JX3%+Zugmz&R%l>Z5r^ zJ7vfh!4Ad1MbIS-%=Hfh#9TMTgKK-V6_Y(5akLeaeUX5ekn076gj9o<$iZVE_-j}n z!-Ib&Ach#0z)La26d7_!b&7$K0>X%FhgQ$Xn74XcF|bwq08WLn#>tR3s!agb^%fF!vf+y z@kar1pZG{X*eBFZl*yeiKqo2(K@}pr%8xJ3buK~LrTBAGIeTpnT{phB^{5`}>vHSZuU~lS{3~5ytuZNX21mBw zLAr#U`Z-NhT}K}nbVhcKnbx(k^Vs|!aBbMDbRf}?w1|coj|_jd!`TE@W396pF5rQ=g&re&T`u-?FxkcMnc>)D vtInt{Y{rAsRgBv}Sd_~%pQ(^tvX7UQukZSe>Vj$Hfml3^Lf7qs&n*8B{A*_0 diff --git a/doc/LectureNotes/_build/html/Clustering.html b/doc/LectureNotes/_build/html/Clustering.html index 777167a84..eb64804f0 100644 --- a/doc/LectureNotes/_build/html/Clustering.html +++ b/doc/LectureNotes/_build/html/Clustering.html @@ -5,7 +5,7 @@ - 11. Clustering Analysis — Applied Data Analysis and Machine Learning + 12. Clustering Analysis — Applied Data Analysis and Machine Learning @@ -54,8 +54,8 @@ - - + + @@ -139,17 +139,17 @@

@@ -171,12 +171,12 @@ @@ -188,12 +188,12 @@ @@ -205,12 +205,12 @@ @@ -286,29 +286,29 @@ @@ -324,7 +324,7 @@
-

11. Clustering Analysis¶

+

12. Clustering Analysis¶

In this chapter we will concern ourselves with the study of cluster analysis. In general terms cluster analysis, or clustering, is the task of grouping a data-set into different distinct categories based on some measure of equality of @@ -343,7 +343,7 @@ exists different categories of clustering algorithms. A few of the most common are: centroid-, distribution-, density- and hierarchical- clustering. We will concern ourselves primarily with the first one.

-

11.1. Basic Idea of the K-means Clustering Algorithm¶

+

12.1. Basic Idea of the K-means Clustering Algorithm¶

The simplest of all clustering algorithms is the aptly named k-means algorithm , sometimes also referred to as Lloyds algorithm. It is the simplest and also the most common. From its simplicity it obtains both strengths and weaknesses. @@ -436,7 +436,7 @@ no-where, don’t fret, the method does get a whole lot simpler once we start programming.

-

11.2. The K-means Clustering Algorithm¶

+

12.2. The K-means Clustering Algorithm¶

The k-means clustering algorithm goes as follows (note in my opinion this description is a bit complicated and is lifted directly out of ESL HASTIE for deeper understanding purposes)

@@ -460,12 +460,12 @@ understood but to help you along the way we summarize the algorithm as follows

That’s it, nothing magical happening.

-

11.3. Writing Our Own Code¶

+

12.3. Writing Our Own Code¶

In the following section we will work to develop a deeper understanding of the previously discussed mathematics through developing codes to do k-means cluster analysis.

-

11.3.1. Basic Python¶

+

12.3.1. Basic Python¶

Let us now program the most basic version of the algorithm using nothing but Python with numpy arrays. This code is kept intentionally simple to gradually progress our understanding. There is no vectorization of any kind, and even most @@ -718,7 +718,7 @@ two improvements.

-

11.4. Towards a More Numpythonic Code¶

+

12.4. Towards a More Numpythonic Code¶

def get_distances_to_clusters(data, centroids):
@@ -1000,8 +1000,8 @@ Runtime: 0.004456043243408203 seconds
         
         

             
-    10. Basic ideas of the Principal Component Analysis (PCA)
-    12. Neural networks
+    11. Basic ideas of the Principal Component Analysis (PCA)
+    13. Neural networks
 
         

         
diff --git a/doc/LectureNotes/_build/html/_images/chapter1_11_0.png b/doc/LectureNotes/_build/html/_images/chapter1_11_0.png
index 8585b7491cb359fdc358093418f34f4a361e7869..65d331dc5e178e89457d6a5492e5d80b639a355f 100644
GIT binary patch
literal 6183
zcmd5=cT^K;yH7yKA_BY1iqb8ESSY%HNC_;W!hkek=v9Omx4m
zS_GtpGK!!P2}(y=gh(erT9B3-w|sZ+y=T8azjM#boO$2(JoBD+p7MKs&vVDz)Ieyr
z#BLM{B?KCtyM#iaUFcP_Pg%CdWdCGgAwG$vQ(0g20?6P^KMQ(+jhOf}x5?52i(p~T}DgJ)2
zqD#4@VC#80C#nL)NFh^Gh_wUXbZ
zSwmx}vI0>F6mTO9ZZ-`h!kzLOzV9kPnG}XHd{c|+3o4l{Ei%JddT6{0G1w`?8H{vrG?{U*CCdPIF$oNZSx4$LsrR!sk5kL#zZ_re
z1q$T`QyINr6M8F=Injuv_ETDhQna5?wsLY>r$<82nymRJ|3pP|>FNwYRl=`(Atq*=xA_K0sn9pRL)f<{#BW8x+|>Q$G(
z(`^0(w35t{6b&6AxQ;_S>E%)E)s8hsIv(j`uIoY6cC4W2&NZ&8u@myavMe%l
z-IPAgTu7;3^~n=G%Ml5^*jZ7?<083KD~opSQfU&d6&>i3R;EA1va|Vz`Lt_sAI5k(
zXT&2<2XvFaNK3Uz&)Z!s4fBJfL7e`2sH7$27?-W)*ZjE#}^Ta%3dh?=+>)oK=-vGQoh=|l|KXCE|y^FT9^sxLmXi?TIMm7|@Fl>0sf1a=@|e
zW@8-hO(HzDe`(PMg+s}fkceRg&1p0}ws$?HR%f@ipW!w~FjOVLBV<(hy~b7D1a%Z8
zS{dcCYo|cVB<&ZgL9Jf_#m=nTKoS0kTQcJ^+;No_2~7GE}{sgMgIRV&R_E3AHMLvVvv&#
zxLjPS132>3KD)2J+|ko=h#E$}L|1S<5ilBcvi314%N8=@ALVKA{XpPqL_IF6^%#;x
zs((si@|5kO&1B|~cA9W2-#qd(=UV+-YzO^R0@`mzd<>&q8d(4|3pel#ptOVD{S$UX
zIido#KEE<+FbBI_gMy+cB&AZ)D^xkZv-*`RY3NxgsoHn6;YKx|QP)8qF}G_c0vtQn
zpN`;L90jyP()d8QO(l+8D|Y4>^$vI%HN`^-!x;u%>YP^QamoBW)V3d-|1H$$Uar_h
zCaGSDDu*yhANU*}F!s>>=oBQbz=?ax;L0KvEx^205)*7ew|a7423QZj&N-jx!A2zIdKbVg(2ccDiXi@T3sgjIFhu$|-i!4YdKtHGwU
z6hNAXdsKfOk%CIe+2j;u?%5ODu*X<|bS1iabfRgK6xgLLF*Rh~DF4wZe#*|U>!5o^
zR;#D*u5%|9fSIIIH1J01dyQM0vLXKd5A7j1uWr1PWCW*e*+E>=rdxa*^!oZaI?Dmt
zXYy1A?2?k03Y7hoF_n^m8GfI6y|?Fh3eXXM&FszQQF8QVWV|(8MNi{e$PywP-&()?
zvjLB;6O^49M&NFJ@(r4aK?64=e}ckA*2}vUj9^ul3j%S#4PfyFQ_n9d_TtkcU{{9t
zrw6UxQO!JMsGXBkoBf;(J*-+}eIT+|je%d4iLD@VyYWenD
zGkG1Mq`(;u88HA6)~4q*0p*5Ep8yxbrKf;%o;f`E`mb^143~yihKmhYLuNcX`$OBS
zxEsd%{f~qukkPHB4}~|7nDYtFyi!`X>H+X@+Id^4mgx30-uXFG$w@iHfPW;J@iRb4
znICPA5L9^U@Cq+E;
zR6y%*RD?wq*KYd_1HU^!{T0zbQd2#jkkaefBkH%p0MI?H)$fUWltWk>n@||9_}@wE
z&JXgSmxCpyw%E%6@+~I$rmOoqF9d!K4lb>^bZhWZvZq_6-V@QF~+g)0eIok
zqOaK2>-HzM{bm;XEpqKv;F*uRY-`Gv*-Pykmr)`6Da7*r-0{t0aKZopBsr<
z6R4@|<}d>HXq5xlxc;WKpUf5NmT9zLuW}o&(oIit!1}Y!UGk2Fl|g)OS@JPbhn@P*VmKZzXtYW6ZE?~X%5)^95>6`gHE4`l%+!{fLAZef&*h@zhX{0p&i+(9s{R$pB~
z`mQ4(%l|l?J5}h^2lsKz?Q^3>
zp8f30B>At{Xlfr&sP++Wcsg9+(HdWWDEA+IWcyg(MS~l{RJw<#`b}u!*>mmi#iS!|
z$+mv(8pgj12<2+W_6oP(5XjrntoCQElx+-o)WvhWYVuXj$FzDiKHpR~&GlJKP4^!t
z%d?oB=BVCW|Cs5KIuvDg`EcaP?3eQ5Ki-p43a!DCvJEX+PoDBdmL~Dy-UHmwdu|q#ncU
zv-e8OpVE9X>gHSRxVcBLIX6*;Gtki$nA<}R-`swMSv_z;JZ$LGFNg1Ev!`=q2OI{;
zIo+f0@XHHBvAWIq?v|3LtLDVlwo+wqMOixp-(lNlgt^64T;B
z(P|lkSKm8boxxG-cSC0g45Q}JN)iE_TT$!~7DqMUtaql!H*p>kQ;#IWAGk?UElwki
z{#zC?)*|>GlteH>S+y-)6DQPsc=F1RzwxQEX)ot9C8LSIxAgnd}adotQjbh>AlDq5kBV~9+MDp~-vXR(EPO1azQT4)DwPKCB
z$3yn~a10+=X5E2a7$t;EMVE#zZ#mGVM~0?&SpKCUuC^sTMj>vBHOW$HRToCyRbP+3
zOY!#3c;)r1(@5VxBgqb5?)EGuKYknNQ|s((W^**C{q*Nmo4}-9m|sem>Glw8(MPFz
zOBUjgKiwaaYw|Ge2(r!C?Kq;cBSBXSUv?DA_|O=>KR`0+Q=is*vhB?C&!CjN_-bqD
z$>#*J^1Sut-Q}4%*MyOhXKwGaT{xT|xcvEz{V)|oLy>lYxn&*
zHz8JWMx*X2y(yi|93eU^Wm?D(-)U~dlW8VqKN|>z32)EUi7*|S-{#L?uzg1tDg_e4
zRHcH>_|!@=?J;xwSX0iArBgbBxmE2(^CJ|t_bq2xz7mHd(aS_mc8jyTckc)kg1$Je
zvFdMp0biW&VH%Typ#`uSe;8K3x>>p6ELCMnzjoq=u=gxmf#iiD^B$mbNoe&OxsKZa
zMc0M!!mIz3$W13HSBI~6(iOJR{q)wtbcg7Pu(renQ^_-Ca7$DydMr!Nms>ynrB(?R
zc4f_`r`X{ZXZin2)%ByRS>*mqwY;5XdQ&@x3O-o?Bo_hx)@QK7Z-3wOGA=rV@fNF<&J==SQzwyuTp
zFIv%!^_h{vi{BkUx}(i%`ZQR7whIU}?S}8q6#8ccMYY%0aQ2_Uf}XY$C8|S@NjlC@
z$aRDAoFLQoLlU0|=tV6gJzADlQ7bNyFTJA{fQQ2OJt8%!O-SH8?o}V9%X8ay{5Qj6
zRy(YqaG0iU9IJs@&z-I14-JJ46l-WVMB?;PcWZcl^;Z#ca8c5E6(fh0Fo8@n;kYM*
zVk3U5*Pzvpc4DX9a~qDx2x_0Ea!dkEU=^J8h);Z%ShG;WNvaP0IDM?6psG%lxpu|Z
ziK(9pE}eCiTof#JYWFg%5Dt}G9EX%VWTY9+uHkpeBam((p)D0bJ72l>PoR
zA@O~K%o(Xiq*1-82OUAvTbeVONV8X2klqltB+y%KlH|TH|DHLfB{BWww2O(GLf}rj
zS3h9##xeW3V(F=Jj1su?4gDDMA`@G#pNb1G7QM$%o^cHPR7Kht_3A%t@=r=mo&dOX
zj&6ZC@!uAUcROhE6WgXusc>2F%9smXR=n%r8*-BHB{&|*MwOS4=4Q4s=H_f_j^it|
pJ-Kb{3NDA?_Okz;Jn!hPl+S+q`nH@Mw`q?8&zqhrJL~%Me*qAD1nU3*

literal 6370
zcmd5>cT`j9woihB^l=zbx-%YNR0NesQOZ~ljz|*^z4y=(dI{A6;~Q2sK1N5CSBK^pY3veRt-)JMaDT{&?%Ith2JR&tCcVclK}pzTe*Gmbs~+
zsF1V}0)Y?(jV@e4Ado4%OL7lC&q6p*C&~NL4mGd|wZwXbMz{uhB1~LEasJp)e~jC~
zaL?cn3^q_*mG@CO=o1=>3puT(7Vw9IDmK_#?Jc1?oA)PzI3wE-1VUKi+l54AW**Gqs1#q}(}TSP{a>JD#3!C2)Qb!r(P&eC0VayiP$XUMw+!kMlzcz
zGKtB2O=4SJfPGS`*p?W_hv}UlkvN7bChkX7e;B_0j94vld-F$T5NNs}vl7E3GA71$
z%Nt9$>5IyW*T?pZ^SA>
zoEIm=w2j^F+7&28V{`_Nbm|J1T6L8X-7y+B?h4#tiRSU;($DM)vxe&u;&jL3(^3uGn8Z1}GAGB~bN3EV#V${BQGNQ`sV
zut4jeGo(Mqibv!8oVle9u1|)H-A|~zB0@B01REJwuK77UQ#U1fIn2HbLhN
z<=H^`$od4YL)#P7>i0Wc3CF)Xh-9@);?x`?s~
z8ednu#M(0p^->}}Q5=PRs;$8BZuq#$$42cOq#k;#O@JpK9a35rV^)H3RrHm7Fqi)6
z=4zAzrxvh>Wer2(Cf}VX`Z3kR#)noy+@>h>Wb2dVY_weK^QJ5}OQaghxQ5Kblu@*D
z3lB@{WBG%KvGhs0;rpE~4)Dwu2Ut=+jV{8lFHCTc@<0BQk63x7118AS2nt&@xHIZ+Z)J$LMQ@!z{@WSXJ
zF!S~;=asR#nW0Y<^@Lc;G8k@6+b@vW*hEy#4Aq4T7ERo{#?dZVXw(5(MOgP{Cp-X3
zzwagI4=lp>Z3;W@Y80veeYG^>JTPd=w%To|5c~Dn-T18uW%Dwi)KPN;
z*PH{LnniuM@AYXRO##KQhV(CeDWUA6^a{v8G*cI4%GOrQ{9uJQ=@x+53kC*Ymw)Ni
z5(yLN5FL`t?v`0`hnDAw8y4lOO~@kCd{yv@Veg(4R$^LO_5+VYI6YEb(^+_Db0u3i
z1|t*(!lyIWT#JddYE%Ps#Eye_eH}6Ia`0hN5iy*Pt4FP&#G*AV@QVd37znHOEoZJ|
z-vm-_ISO>Xw0k%&lLe}Y0
zkAW5L(yl?exL`O9_TSYh5U!?U3q@Gdf}qM$Vil5(oVKFn9_~{DXI;i!EudB8ek2{`
zjM6iQG?i-5)S9&1CSZmW@UZnVaaNj{ndn-s@L}aPrVjvT16#~+#;{l3zhR>O0MMy)
zl2LFtoCVLB1TWs3<_5rB%Ip3xH*N9(rcV)^^%{4*@1cv+#);vCaIfzjCliUeHjo*L
zfvOMa;eom~t2MU;LtzkWYlsWTGX72$)*Onb)A49?%VJ}=??ckz_Y;xKmByV|ilu;z
z;`%9OI^`mBrdjEM#~G5G3#-4h-KiMBkRNHkc7zDlH1`BLqFm;Kmp9a5P20*iW?yLk
z7?68ixK}a)p}ex3*Da&*jz+0aw!H^nbpBdzX`9jUf@|-7_y#)kVu&PopfwyJ=irp>;k3IP}Kndhh#_3Mj73z^?4HNn44MOEB89b+95*i<5>P
zOG3qOT!rr($XQh(&Yq{_Kw$=uBXZ=_3nDL`r%Sm8uro^42-U%8S&~}YdkE|qLU+0r
zPb4UO-C>A3Xx#se=T;PAbCtFa4E((&8ZW^KWm+pi4*^X0ZG29PF(z*1t0;Q^Gzi~8
z@Iuv06zG*rik|#J;Kgn*;q;(dm@Kd+%6!u9du!hF+jsGIrJI{$cZqqM00{Wintj_hY7pc)&AH55kP?_F#anaL!sSqMT&flPs^ageEeS0bQQVPmwwR81B=rW5p*!xqXcUE@Je1z4d^_GV^F5NB2509?
z76SM*9p>+iK>6h_EkK^}mlpt|z#R5^`?uzwsL{+|rY!RZ<~5*51idM8oTn@Vw)S!|
znloq097zE(&+X}3qm?X@k}_ymiyLMUAI*FO(#PwtLRU&zB1FBOs9pFA+9xn>u+?=_B
z4X(*X!IC&ZTp+aV!&|#Hto|b}Re!BMWj=O+;BQ#D{t?wW5siS;p#y~`9XPNDzO2K;
zqaO1orXACmB0&6BiAvhl^tn>X35q~kj0SaKZ<=4k$D5=O<@X(Dd$vYvDx_3FJNRm8
z=b6;5t}c46F5aqE1hj+T5p!s1;*O@JGFcI306OBB9qZP331F6{NAEtt&N!7$aexu-1IC)OXO@1iMm|e43WPTUGT>d6Q
zbZ~fQE-%$D{>w?2!8`2ct!RP^#35T~2|Cr}*`ATEYRDr5Uarn0NNjGHeHKmork4Jj
zP{9i=r31&(ZY3K)qXG`Q9E4*8*M5OUA5QiF(jbvXAp7^^s(M-jjw}9vf
zx`+rqc5H%$5i{eJSU2%BPovmL{c*DVU(^b1NI1o@&ShA`rn+ps&X0b{8URZeSL3?e
ze8cCj%@2R1Bo=b}?X%(%gw-5sY5`2Ty**x%;1GQLXGvMDZV>`-Y;Z^@WjIgTp?!~E25WfOVL9#(SW`FIju7n&rC4ZosFjKHA}dAA%Qf1Dc;3<?_muwQ`+vZ__AV*VvD#0SS4LmEX~6?krGwq#nOS}-aT$0W
zQ2NLJRECC~xdwFX1(WgC>rWbM@vPDmw&EJ(0;-R{6-;p*dm#M@r_rj0m+~Lz&UP6i
z<|@n)6s}jX;0jx$+2Hy-Ai>fI8-S``o*uZRLL7m%H0I!gff!
zui)^=zOm10mG)O`MC=w)2R4?Q&%11Icx48-A|XSCZHlyVrGXQEXyBEQ`%
z(?6b3inYz@InR+W|C$(?74fp7<`IEa>;ZRM
z)h)By=hIJGo;493)j
zE$>qy6KTO-3`x{42Ne|3fG~~4X7j%bbRScx!~`nQpvm!8T$HJF3WG|Q8pZ!*|fpia(xPX`K!Mt%?G^-^Q
zEXxc}?d|!v{o)k0EiI;SRX(c!QlGXc$jF(rZ}1yP(#-?7HPz4c2$P3PB^=+S$MnuV
z>@GcL-LiDCWG~)16lMFB*Wku{PZoRVJX34!lcOcICc;T;*fD$_I(n3>5{^0(
zeb)>bXggD@OxB6p8!5!OxwlJScx}&jObn%HxKNw*FD>!ojws5$hPL=!*^dCr=RtOM
z1Oq6xo+$P;y8KB^q35yLmrjT+4UQKxxmP9dw_26tcs>UwXl!>e5o=P@ikW$-BD8jk
zF_DaN&U1^kg$Z_(1Y2xxoW?a=$LuEm)nJY3}5X&?W)%*9WyVM~Z1
z*3Ul+&X865>+iIGv}5KeuK&bYrIeKTcFadbI7pVG80nMy(8_hr^%wPR;K^9D!uNMr
zIZvYaa6kM4HC&&{?x^sI%lV4nl-n28TqN)sC#jAN*Pp>25nf;Q*Rl7XCOf5uu}hH`
z*^4r1TZHs7&`wo7c6arwclAgzntCWfna^)wHcGCh{+hkrMR4cZRK`*7tF%?CT6s)O
z9Bet4H1IeBu)4HvQHNlR|HtR4HhF;32_jqm&S5jnVqk
z8%gP{4bjJYF
z3^nAphWGr=`RlvB^If{HxtQ5|uV<}wKXtF?34NiiOhQOUh=YSe^6aUiCJqiR8eAr~
zZh&vhGZgT^hlGpLYZoo3rHi|XvjvWtiHoBh)Wy!m^uC*gGt36+AjAv)ao@LgadCu6
z^6}aK_YJ&IXDhz$`05n!5(3AkdN3RuA`0wgM%YW^i1)eR~{*w)1D~?+Nqb9
zqm8bRZC~V#r?;-Rv1ES1xicQCbVHFuT<{yQ_y;t;>#awG&57i+zxW<(MTT?ah3AEY
zhlqa*=N95Nwjg0m3M@*UsfUz$Z)471{)ydEXLo`BZJOR*StQqTt^a$trjY3FKOl3&
z&_`o-o18qlrGmX4mxYC;!io|FK5%pSsmaL+!~<(XLY}s8B9s&ruc@2xvasB7B!q$6
z-@(}GalJE3`69Nr$L;s}`$hKmzaMIb<;We3cI213MVvque(O?*8J_gQdFbe4q|Us)
zwND-OGX`UJJk1w=8&l*qO^Ez04b?JMn^IlnBB=}u3AvuqW}v9ZI`_YkAW=uB^g@U|
zWk<*F@>Kb-7bBQRoPIjxde%S7y|@r2*W!OZ3VmdJx*N5P$_$(7tFTx>kC{b9dKpWm
zE+{a0$t8r@Q-{mqw%?C3rdC#7b&z_;&F(;$VZF6%tIDE}#
zcbHDcEQ*XTy_TidTs|TF4w-5WCy>rZ({i|Ta+m0f#-Yx~r8w+bAJ;Ks@W5?)9Vi_u
z6~Uk;(+Xemji;S$4K_MF`F7FP;sLluXDVbOyZpf$qwk=^a0?T?5%WtL7D1AllD>Ab
z%cN5O^V#G2=2QOC<_pPPj|bESj->4AK6Dzg81al3AzURjU%a5BxXFu0s&$vE_{T@B
zQ(uejfn6>_Ue|j%rl*{~({oMnZcAp{pU6$~N{NglJ)DI4&+JCJJ0oR|9M`h_Z$4I0
z;c}lnr*NoSyIkC=-(EVLZ2lR>CyU&y_D|GOmxw6!SyoAL9q|^Q56E_LcSn~;`36XB
zpOHhoCl`7#OrA?7pMJ7Hy@uaUxz9E-c&%pkjwU7Rul}7NY;hg+b*P8ikCml)NDt)7
zopXxXPqaiUGxlri8aCdPsf=hdfSZLgCv$#LU)HP5A!}Z(EQf_`r3o{0pUHEUg*~P!
zE8h_~emq4mPYBq;H&55bs+&BSAeSK|Lc`-Ze;1Yi0PjCcj+8+{Z%a>Zm{!QRQSRaMSD`
z1z!esJ2UR*xO5e*6(Nf9uSF>h(iIUIt8^9k@j*$qN+7F9$~=uHG^%%3H~QYWYr$e5-mrODpYg)F{>VEG3&5DVIxn2M`g&_S
zyK8f~a*e@E=c~LI+n48(=oO~xk0B5)8JX$WqoaKnG)k=5B6t(x;9xnJ*`6rlO^O5b
z{*m12xW?g5F7L|(B!8mI`>g>=-)Wl@@5=ANdo*wP4e#F}HPQPemC*J2v)GM0CUkVB
z0d34PUazIXNH_1N_H0+0*qafjK^^5qKdxQ#9~XYyK@iBCX;Jqo%EMcEJl-jZo4>aRQD$JU
z6Qrd^qMc;YNm6afxK+Cs{1rr_6BGn#O}2{OmyizaCdyAI0ej`gt`f_pxo*=b#Oq!aiZw=j^3uR|q
z`K&-Xea_@@JrElCo5+P)
z+1d}vk}H~s`hBvc)!84LCv!1zN*w0%iVSa(DLN=jl9FPC|0w6@N8C{fc*go5`y~0mri-RMZiblWK){Y$)SlrRZ1H|+)5Oom
zk2HiAei=|d)Y{2|sJ|#PQShf~%Vyqn6c50s{KTltG}YU7P9Lf)|C@~^s|o3{R%)ot
z?X^^?6cj|bPi8(o6q{4Y72)LTJkGM_f(t=d4Z@3Gj{df#Fj?5;2Z&Np`SMy?RT%Ya
z;>-Q$@Y@=Sw9QufkJosHXs@5_7&6U@*iB0YEYt1~16yT5+IB*MH#x8wP-aiW|9eF7
z*@$9DN1Mx!hoS5&0gI&e{R9ioNEXAok10{Oi(ayYBS%8j#X%E_*O6$?3rxTCQS-Xf
zCq{|;%A74~Q_M5B7Xz|Zb~;p&pILn3o@$PkAAF#@Bfgwa`}^U8?8KGmA!UjXt^RhVze1M0Y-@i+PqPhSNl_3lhL?YV3b8A>$c+N-pI5@9%Whe$TCfha9%nSQU4
zw$57!r|8YNwuC>hdE!)Y6qy;!PlOpe2`3S@rJR92W0sePl9|z)><1y;`?>zLc&Ne4
z>gc-=@6aFxh8GkTeRY0AYb;bjK_@gig7TtvP7@2~iIX}XmDVKplP&*558Ul4u5(@g
z(?k&!cfmI@aeL)^){4WCEAnXkw_#FGI#*2t93Evl@GzcdXgH?zki-KBrz~KF6R&DJ
zY41~LJYKhH&P*?Lr!9bc;>{OfkxqgxhQoAyWqqwLFS=PaI5w1W9t3*g(3p8N3jG|p
z;N}|He`{ed^(v{E*f@(McntcF(6-mg`+(Hjoe7&kW?9jn3*ej_=937>q}#oXZ?z<`
zUG2kqK4xRTQ!`%gcYh};niUR-ci$ysZ1mL&Utgs5-SZ@q(|@y4E2RAmX3vDfhtpZn
zQvN(<_Y1jdx9LEUB8R^5^h}RPFzuvx&Grw+|
zb=zs|a?UbRx~Q?jk#Lo1an1C&*{r+b3I4_WO{iS^ERBM(f1GMT%dZz{RqHpqEeR)G-6JK!7(4Ps`aOIQt(f4RD|nvtZ5u+}Zez!Z9arTojV8^eNU6@;|^YuFh*w&sndQaFVc_7(=5({
zJXG=reDv3Z?8yj~8F65PiY>o}r1o2D{w`E8r=S8pA<%IHn_Y%Q1zGKmglkm4szuaF
z4GE8{nCTb4GO|Om7Of$id))uzo+`D
zI^x+~y*MRla<;TFs{Qsy$-BhMJ>LnxYk6vZ%+}w!aTB&g#5mrdEE>-|nVJDLl;bbY_#LXHKpQqH!e=K;0{|H}p3nUUP}oy&O_l8DYM0b8r99
zf|E+7C{tP2J!zmgf3HwQe`Z1y;7&7-y-ir;p*6HHqeS+Jf*FI=4L*Oa#r3VNg)b_+
z1=4cc4KS{#(xwxp<28*-vk80&4n%JFsL@PYjd;Cb>d~t+$ee_5Z;kkcc!9nTk=(~p>m9ZyS1JOOw8CVe)4OI;A
z5R~^-2+zq87pZzVF=?Au8kL_^S|O94a+I?hG!6!ji5biM`yxRD*Wai-min@Wc2d)RxCDp>|lcW9KA1eWNNqNhbrcNPFry)YJ&*=*)gpv#nVdG8fgd3zDp9$@GjA
zZCEa=q<*jDHBty~Cxy;B!lOOT<{gNCmfD7h78RuQAG8u*$&Sj^7gz{@qn(Pco_Dp7G+=sH;=T
z+FgI`OPj6cOTQ*`>Nbw!LMVHM4`$C+%%t~qb549baJe(B40z;36z<|4M(yTeD(1bF
zJYkV2)_eUun@uP$*=b+elT(=8mg?YSOg^IKXphlo@WG$-hahG;f6~X2bW2N%{-F^?
zMy*7vU*)Q
zJ8eX@s3eH^R?VW@y3ao|$c92#h}A_|WZHWMU*ul(KBBb#EV<1f(U$-
z6`^{I(?<%5hopTKAxaFU#UeieXmkL<}urxY9G^kKN?QnNZ1ClT%dAb&?3DH~KLM
zTeUkwb>tQC+O7%be{_wP>()yaBc-;!!QpWlzJIkCJeQ&k+{P{b*vf0m?6$EndZJth
zq_@->4t}|X3!EX%3OzS=MJU7HN79w?T%Zu`W5!1@Vb|3!AIDXd9u5CV&4EBZHO+9>
zs=ik(c(>cd1h2NE=eolJ;5qf>dk*h&Ev=Mul*o&ikQd1z;S=~T%~=ZGB+6YJFLLWu
z1fWnu(597ie_6Sc!6-K=MDB?Q`mX<4&S!Ve`6~58=(92>L#EQEsb?i~^Hr?kf8RZ^
zn`B{R;U9I$={v~1!o|14(UeF!l*@0NNw)%+VLFe#*8U-B>4$jcWZutecR8sH9A$z8
z@q_9#vj!~WMa8MqrNb!*$kl>uAkV%$ReajP@b?ROTV(%6q?hX8!;P`m){~2=1b(?^E(>)tr`Jk3pbEsb4>2u8FVt(Nl2kTBjH#k*twy@Xe+7
zts>{KGAqIExv+)#p;e=9kX2ikgpr2@ggjxOvt?n;BrOk75Wd60NboW_|ID;k
z8rtn=0Tk+UvARa{XJ{ll11YS7T-gz`9IsnRpdHg+|Gnbp(4t-R%7wd(!{7r=>YwHr
z#R9TDJe1KrpTVqIS$-f3?zv7KktFT~iO4k44k_TDU^9Iyr!d*H?Rd8WuHyuGMy(zG
z8&jWuQn8anYL6EW1lX0mcmC-xNp;iZ_g>DpE5g1wLzecXh+O}xb5u_sU$f7|jpL`&
zAVf0bS}FHjydQBOFqK1N_vqKmcYf+<8*aa@(89OlT`T%jkkk8E2#;#O^wi@s{ZWc=(%=e
z)kVp!?i)KY(K4D=<3ji~c&Tz0TiDVjf8BhcH~n1E-0HdafRv&cX@7Bcy-`ppo_Ct@
zo7+|#ax5Dv-pG2TQfE@6*msyxg@TeZepFGa#d0UJP7wF4;Jm`9!5N))kPjzJYGyHG
za5NLP9~dt8HayrU`y)nN#~4*Ib`6^Zav_w=qXOkcs{muOZk81pU*H-4`>n*17XHw3
z*SP9?`9bz%cNKp@Pt9`)&Gi!u8b;q*9{P~w2b|5?=Gm`_l%;o`B#Lc>$N8!zO;IjK
zR;>pFcMa_(b6!wqu#uZM=#vl~L4IE>u~Z=_zPX4WP>@$ru%IoA{Y>hAo*F3eh5wLiyv!}r3O2lUXn_Ep>
zZqs*L0Z$_61{x%T&TH3dUEZhhC=}teuq=#$MGj56)z3rlCMo8JFnb;(B>ER?eF0cE
zd@{Mpw1ZRyyS0ADL#LZ%`BD+-?$2s&bP+f$>7#CISd-r(PpsqaW1%9W2n)8^%v=?((70klPRaVA@4y5Qi7jbhr>r255{=GNFf{xM;vo_eC&>xh
zJZLzDMu2wzHd`RZ=$8RgR$-~I_cYpfq-ec;pq6#g6k$9j*;7nGK7N|=xql;$&a(P<
zOY7uCDpwt2s;}P3O_<&0b6`2yk{^9hqjn_XHEMkj?@r5rXbpP5SI;faOvG*;Qxx$+
zmno}>SskYJVPu4rghYTrt!0uEjCi}E`kW0DAMugcEys~N4F6{J?TGwG)X|9#HUbhR
z-?Re-Nto4?a0t(k(|*gJnp-c{7dJqe-^!5q%7(|gdUgJ;!JZ6nqfFw<72-fU!2BrM
ze@&X;;Nc+vbF;Qlzdc&z`PigMcZq0nYsKd;Mr}+l-a{JkJ~;0YHTnA?5lrUF?Z%|{3s!6&-lGstZ(iy=PEjpC2c)Mo%(4mk&8X8X@7zJiVH)sHH&Hm
zo4kTqN?aT(2#+KuAeO&3U9f-@Gj$4T9iSBDw`#s|r>A>Kz#O2X4}ecYjS&0-jvqFlaLf6-Ft)a1^G;b)Ne7ikWL0V
zy630;$hEv|IIrYpfoC9wQLxjl5E0SZTPOVciRZe&grbt7xyE>$l7fAB&;$*G@ChU^
zQ+fMwTuBM>v<$76_t1B(z@Y<7gE9!dKGSy`k(BDrSj-35XYTU59tgZiq5xj@BuJc@
zJpuBL199|X@L<>X7gG<%p?anIc+>eg$+dg*GS6TPN044ViR!W+iq3;_=)(^G^zm*t1NyHn#Lnum&Omg{W6)>5>+B>a69TenA
zy{9B$Ax|A8n3M5g^pBA7+<)gbgWoj
zk1CKH@_7S7h?e=TR9Z!2DQCUsGX&g+-8)61BK`
z53#$Xl3ZOsGmnwgKdq!OdpD$6lGwz6!&7NnIHOKZPW~bxYP~krP`c8T3nEGuDS3hT
zRD5<^&8253o5t9yg{Y^Q%*?hwUA_(uCJR6gNji$g<*2#HTl}~-WPh{qqbE)s{QFKs
ztD-^&OOoGx7M2(V6WPc;4#}+qe)sbs>IzTlcYqUbl4C##S`)xyXxt*_q2J!i+~V`xBLXmH`QB6tU3AV&PmIf+bnq5V<&hG+cD
zmT<|6StheB1@=%iQ99{^Oyqp;6uo{TnZ~<5D%$STnzYsqK!FxCPeg%b<<~|n@^8!~
zd17jNygdNDPV(M9UyzzS`dKzVlXA8olT>me14^fr%a`X(Qfo;sIL2g)!O8?dO=_dE
z7j*{P(ZoMr*xaapgdIBup2LfJTr=x_*QMo5S%BbZBPy`@pDkrDU#W}scT0IL}(v;H&(SZifY;mG;B
zbDthOLfen_^&KU}H|O*~!S0V^7kOJ2T^@^h>sgAP_|DcG8q&R_nGn^^v?2?0_t?|-
zyem$@>uP9k5DSi;Re8ZE;mzkIHd{MOVl4r79MGbnSJcIRcwpb30FbnoP9=$E3cC9i
zWon3<|Kqv6*G^`K`g~5umJtY!u_ss43Y_ffMf%@{w^nc)lv1KHKKBbKyYNH(C}8>!
zw%TlPQvsq&L>hO1MoW)m0If3G$20d0kM8ag5Z8hO(IsGz*bdhp0`ZE#dv_i=2Bw59)5}Mysd6=e$xvqUGj;7y<&_l+YUC2)1elS=S*Sg)S5#E&
z%;IMXrqxS~@7qVbV0@S7_LPD=n)3<3W%MW_INqf{MyP>#wY@J&UM|uyg+b9~t^1
z^2&1yoSHV&Z+h?N^w`sOxcYu-GO^|t1F!BMyZCZX%XpLSs(9hQP6woT-WEz)P7MvU
zeR-&24M-rWFeh+q4p;-eSCs@%>!kx{jzDj>XM;dO#UgY
zt^Y_P7s>pO@xrKh3M?sRK5s|)8F71Hk9OYF+0U;gVeu`Xea}+vZ(df-MnzxEYGxP*
zRTzW+*>1y8<90Fj8Mz!mCL>J0xajMv1@|gwieJH}HfRCYi%qDMiCIO8PYEUJYw)ND
z0t<$SUW5Q5A$fRje8^n9*&zB~Bs}R6?i}lFNZ<;r)#dHhP6XcW6+oETh6qedP89LS
z0&GteM`9<9equd786M4K8=Qw*|MCL(czsnj3ok%44w2gG<`jcLjI{5E@C!%a3
z+qvV5g4+*%4Zrq+W4mccvlMIPa)33WX-Cg;6aCFx&R$m#JrH@xGXHByL{(AI{3a#J
zj2LNI%^g-lagDo-$B++b2d4^nm7|4&J{5?2>EHs)DnjER7V9lIp~};mhF6V~)EY#e
zByr52&=8?SRV^!l%x+IjWS>0jJ0C(Ef@6(sDfF*r_A$T~3|J@lhWGlRjBIXr3+K$b
zgE4Kj5~5xv_HV^9DrP3NVs|FF%_n>wg39rMdI}d@$t;SUXG2;cr!Hzf^&_CzXj9AYGBpoj}7UA-&zvjqGe!MYQ4#{baA@9
zGNKP%#-{NN=z+zh7!5WCQr*I0ZIdUYl~T+sb(UD^rvlho{`~n
zYA2IR;q%)KOL1%?IIZ*57W085ZQCXQNBF@c-m;RK5Z}mWSINj@9v>WJOB~OAJkqXb
z>vy!+f?P`h#XcPYy#rS1zHz79&FR66ULiR-Y%}PzBgb|UF@CFo6yNEl0yN26;?8@d
z|B+reAMogUEN&<)OvDU(Gibd5Pnc>X7)qBJL1S@B*&B64xU-jN`D0|wRZu80A#>PgU
zb^Y0epVY>zvCqrRd%pP2blVDczQxq>sDUTUehK38g`$MiqGMsd7@}>0G$RZ
zf^-jU7TG>RgD$&<>OiP-=N1JadfWf<)Mwj!c5$zduM4gMmDp*))2p=Xm$jcV=kK(m
zx75Yja9sWjevRB=M3M**PTwQ7TQEcS2_71}?5go5?#@E!{89|>Y0-;i7^fX^S!MJD^AXK8-$+6{M
zs=i@m$E(TOd)-vU7_2jAz4&%yvd`aq*kYB;oe<
zjW~Q}ELiX0fKiRl7O0C1YQLOr)khKWHGIaPS72MUE6yitxfEeP3;uK#2*&m@w3%F?
z`=HTe&2y*qW~C3tnbCLuC2#H09Z(xK3u_}_-vP_7o8_y$a+-aL;Awf800L?pRSyQw
z&50^$G(zZXO~ZfTmjYRKNr{6o95O__ZRQZjy_vwq!t%as*yIqv&F`Sl*lE`HC|W&N
z;%Ew13uMZjCO(XoE*0)zBwX_XD|NjcO#a@s$4%JE!15Ffxty03dF}0Vohzt@DcMU}
zgF}+AE1>2F{QkEwn}ljO{0njg)1%?nrs202v)SaJM-?T#*DVW~GJk;;@z;hhg2{J(
z(6-8~mj>uNsL8u}kk7F0D}{!1bhb*o^Vx0}MO0VepOR;b^~(@cq-6zokbmRHSpz_T
z#9sFUP%?2UvGzTtW`9Vf7Cj)P*XZS1HKFU;9$OpdeD~^w7BeFwou@NLvNK0bNAq>O
zppuLD@_hCZTdt&3{<`SZRv0xO+nvx?h5e^r#iS98I#}uWRXy4rx}08Dm=mbPahqWF
zmhHyM-ghy2hu3d6C2JSz>_$iVm>a_@ZGCjHzT|peK)X*-Eo^14mCWxmncvRG`m-5-
zOuTa~somw`pKYZn8uPbS{H#9e=
zPdd~`MEUJ!{W*izo7V`A7K1eR4rRSC;sWfZSFrKv0(J9o(DvooiX2j8)Sw;JpX|Oo
zx%C1D>o=0nz{VO*-SBezu$-Kwyv2%$6AMtWSpVc)#OZHgXflCsm5Ry5JefbuJE`r4
zmIX`Ty2Z!uyf9VRO
zkt?5$o0wyG!Xh~zju($_xVpMdxlP(L
z_#MwTobC1f9%&Ie}L?FkDOqbm)EjWd`r6~92(`n!fNTe>$
z&=TNsU(w&zYJz?fOrB0!8VLdB0^tyd$(23elm);Z8XPRIo)pTT$_&O0w10_c|*tj&SZ6HBL
zU(>Z<$AF2Lrmt<}x;U_0A>fukBI;|ue*J2(ubzIQt4nQfZ;y>(Pc!YcS3E5hE&emw
zbe-h|Fs^X`*onYyReBxR#_<^sG#<^=S5NA=rs<~P1P2xigDreFRxDMPl$@L>=P!$M
zlhNaTe}BJs)ytTlM`uG1@$|2Bp^H}=)lM-B3J#gh=59*t|5DHcb~Qls4Z$_!q}%4K
zybW+8t!2afmq52#fb%l&^c#A{j=MA9-c;$lu|KRaE5?(29VDmINx-P#R)9N4L5b{3
zlM8Sj6uI0NubY;u$!Tlr)}ZJ0>o~2IhM(%n^S>`2hM>%#5~pA;qJJ?%*(yhVVQhol
zy#}X6pfvc!hHynEZ;GJk6jxumI>lt0L0BjF+HnI=&7yNT`g8X6NP|)AzKXkf^-IU}
zlzHr$l1i@LMPCmskUJOV(kaS2OB91L
zV*O}2pc#p%UT{zr{23_VAgN5&Owe#M192Ifb*_n#u5MR(7h@l|^d)pSC|_6E-o@rG
z;Gp%dEcC62_Lfk!-IZgZsc%Suu--8l%v9oj^TXQshm~e+$<24!iCCX~|5Gu!boHx5
zNS83!?mK~)TkX|3df3>T;5PUpmY)n;3(YD{iFyEtygxkP5r*f*8ivWp8Au=>X^LNQttT
z3>vSp4hNPptgXRr@-TO9bp$KM?QEUI70hvtFoJO?QvfR^AzE-&2y7@Q|o^W2;#(I{>EhN=m!J0JCGjW?fJP3rhtna`eY5H`Qe8o7q4nT|MUf9M)
zG>E8OVGO*-&f^+GTs8#je>n)z6`Be0BbH==zyE8RV9d!!Fl?#zQ$RNM1#!W+g>luv
zM3nNiQQpg^h<{G8#^oFM?fFq6#NG|a&KT%9-mPTDp!N^$u{N*=0+lK|8(S=1h+Y;N792127(iXT@Hg!zC^^eAey
z#IO$(!*l%bAl`}K-K6@dBIUk$!~Y>hecP|sP
zdYz0Z=73D!C{e_GbY{Jyh(8a38gM`7eNxmPd8Iub;HT1lbpB*Zta66NcW
zZ45i-dlMT};#by!CQ!mm=iTrXn8Uw>MMEkTUSFBU7LHn)M#!77m+`<(mVuc7!yeYj
zaV0%i4lY=8gLODI8^nrlb1lW#NQ5mD8~D9@>HM;e#~uVaS1*nD5(WUAfz2d8DEIsf
zIY;miDOcx!&>oC{Gm6Ifc(0<2yBKSAS~K}?8jRk&|GfbF+NW#ylzk>!V_1fo6-)`(
zv3u;oJv_iij=K`VT3l7M#;eQ#$8|lrEycclgLyg*>oLCXUz0ACMwAdZf>n9^oUer^
zxK_s>t7)2Evlb;v7eO-O65tbEH?
zqjxJ1QLMhl4-pv(fl0pA@_IReWVoMsWrH3Q!k(zU)4H;x-}g<;ElRd-1EclQE5ih9
zTSNmTCflTou!{rEM)CYGNwrt|6jG38gypZS(mdI6Q{Cv~&gRJq&o`?VPjAf9w6JXC
zVFRZ*rvgmBq?A-GpsbEDNPl(U62>bm7`!*i8C-t-^T)1}LkW2kFAhKBvxvKUuJZ}s2)>nhTKwHPVZx3RY`mWDzg
zS(TNQ*jom9bW!kZthkR0{_@ugSk{|)-}wf5XS|0PaHMBJX~ONOAZtkJ&p}tT*zbuq
zQsZj3i2p~WE#-~1MG%vdlh`U}E@ZXC^mBdvNr&{(Tz2i+Vmo(WPlj}u+G!LtH0VHt
z(Iiw{T;<|QqUI)a?Vta#F`x$JD)21+%YO>gz6&JAN>C`@xLsT@lV1YJD$0SVkpek_
zrr%y-7df*)7d2`CBcV3-5%_5uO~hOC?m?7v1@@-h;E(_M_5?u0;jE;C
zM82ifn!$So;*V?3+Cl8_|L&Rk-$~-|Bp8E7Cf$-~mDXdEo7xE|#P%MCfv>RLpVY7s
zJu7TJ;0&%kP;pU+?0lN7nc{mm4pJ}NccAe93v56lcvwR&6=X-ndv$lYbap^e7mr0M
zXg{i~DUs4sOM7K@x;~tKA)K4;=d~Ztj$QJ^!SazL{0Y&XhZxRVAA^0tYy
zonX0?jyR3T@^g!D0YDJJ*ETj74(d`57FBlu4VFOWGl$2Pj!c%%c+1;#Y>IBdBU`Pk
z!dpBT1%#CJqO86UT+n~7hh4A1Lx6;Yg!K-=fcPRYDEi9iaTxhp4&rd`z6DTz2=zpI
z&$Lgw&lbzlFS5&F2*jow@jx_>(*P*mjpoY>M6wuJmyLT&l{{I^ss#{u%6dmwM8v5TpW5+svl@CNc`z<2b)vdfDGI_0)b2s4e<})P3Cfa~
zPJRV@Y?OEXHnu3#DPCx2y5I+4kX7^f;pEo(iCu(fMKdXA7^6KJPbxQ=nVC^{cXwaL
zT1l>U|7%FuQ*I5JgICM70zncmR36gJR5?ArysN&h?l-grQrg%F^+PB3!pw*x7gsNn
zTVP0;E>Pds2Da_lfYi3MFMOtD>>?=?%z_@(bRgm?3AmW|teh`(iVW~#Tp6YT3DR(LDf1v9hxP5BiGN=Sv
zUIhMDriMJp0?knUK7em!Pe)dQ%0(>YiMx}{=Nr0S+MMPh*>4hjuM@mex~DkR2V~1j
zfaY2O=Mx>ShM;f5WkB}S?j074j$lbXBUfrMsAR6LI3E2R0K|>FX3GhyKEXvp#jF0$
zgz_sRHXQiB?ZJ$;0mERC2?KQ8b`^}2nAowKIe6zina|tbyI4wO%4Jy1+vftbCayC8
z7Axz5!Ks6s!MU6H?5}R~#3DKjab3s_Up#tk6}07KR9DA?GZWCp8LDU+g{(FGN%%%z
zAA?T6RMpkhwHVC@rp$3riF$L#7c^$-;vOG&$u$k5jXjrR4Zsz0TNzFTYl%^S**J)bt3aa@m($Bt~Y4F6F+$)~AYSd^E
zpl4#s6$JiD7G{ydDzQUTIM(}KeL}BwhPQu8DP(KN{u46PZ8;I{-ssT
zFE||kXzRY@rotTgU%F=wt{O>Ice{Yr+r%3qrN&4@_vxRYf)%&rpPTnOfD(vqLArtt
z=t1U>pPS-L2sk-ZCp{EPQ-j5#;(lE=`p9b371O~k`;UeOncQ(J*8`L<~k46=ppfS=o
zvzHe~?{^FRaAuEc0rw3Wu1MUcU8$?8s=TKhq7$E(@ZNF6YB1nVpn0Io6fEn_=c8%%
z-e;Tw@rG{W>cQiHHQLrmD+OHcYwpelPc)7=ID-I1pUSISTc{ty>XTdN58AylDdrq%
z=5Jt+47Qq19g>og+)mz*)26o>u(pH?V3mPeA0A#nf(
z@?T36%;0=0&H~pFxBwV1S!R=m^g;WIJ|F>H!k)%qgFze6#oscR&C_FE;oOj&9cV7t1J)aEn6Ui`z%D$g9LrY$bI^XZ3;d@guWOwj0`8i9-Z7yP
zDeL`v6kC`H>PbnJsTIwIK~FVXJvj&oDo*zyrT>=>>i=srP;<%jfY<(@s^3cw^orv=
NQ&Lwfk$?N){{YXv@uL6$

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_13_1.png b/doc/LectureNotes/_build/html/_images/chapter1_13_1.png
index b145ce82a3c0168dd9cd94435a2e21c69993403e..d70ab26eed2c169cb7994caa15171b77cf728c3f 100644
GIT binary patch
literal 9562
zcmZ{KcUV(R({~aebR;66pnz17CcPIGkbr=S^eSrTA@p7Xw~7>jAYF<#0hA7**Qit}
z(EyIu^bro-p5lXWlOXhR=Kv?w-ExE{=TuFTH(SJmIpEl#e){v#&40
zM@34?HaYja`0JU+kSDNW#PePNp3f4iKR
z|Co9YYW;%{%fVjqsrRufmQit=PZ$y-{MyS{K`@Xw2
zfpOAQ?K|I?J0!Q<;=dz{A=Md=ia5YvquzXXWV5|ls9`XZAtZLd{kt-2My8=MEl3t4
zO4M=?5FlBi;AJOj!btS0V2%aWi}yCE1yyfw@(JCMx|jJl;#~q~&~@a#0UkUbq1_o_
z7A9#am_uI%-k|cQ8l!U3?mR42t1N*AD{j4*Xa9MSXOBz23lY?d~xs6xT_KVEeh^(92AZ2NZT`k20)v&
z@aGbh9eVggp&3!qg?_MOzuRWgUYst&VlY(7GiY5NRgttO4V{XVZOJiTB>CS!{DXd(
z@LbITuZ&VKXg0;u3Q4%O)_+P?I4(!q>j$Txox#_kb7{4@cy_GO{A1L+G~dtYDWREL
zq>;2;^}=mcch69Qy<$DDFsU?Q?!q6Mt{APqIH_K_eHM<=kR;W2kM#(xj|C43<=k{J
zSUe2!=$^v`VO_9OLhDI;YEUQrEYG!_neMy*1-l5{hE>-TfmP^t_oqfQM&J#a3V!odgM?vMXzVY)XFsu
zp8h!=%2P!X$b{hAGMzWa7Dsk~?C8|!k8Y48-kQrFG*xe(yCDzd>TfXG+$K%NH1a6X
zD#`Oprti^Wc@^Gyff0;VW!t;M$Xb#8(S)(wk-)QlY)fQ-qFuz*Er)r!`C}}n&{>G~
z^v~d=n7{X5Dw~|A(+P)O
zqA>!_azQ%cVkEFVk>(0>v>(_Z4RY`zUVY_*Ac&Xk;_D!ps+;^n9v^Utkr6`l4uC_d
z#B^VT_A~@y-r%lzJtHB8EU>T>|9fP>nx3nbKAMoI%4I74q
zh1#svz9F=OE%4DI*>XN!*J^2~o0*)9{RR|lYZmhG6+tjTD>_K#${OUaF-rfSM-dW;
z(0(TzAlv%gh3NdvgfZS(|I=EW>x=@UTKRfFra;gW+CUoU{KPGis)T5IB0v$CuGegd
zSCt7*d_%=;9X>lgPN*$kr*Gd^$=+WYNizLyj?Id(7(S0eTh0$-eUO_EI;l|MOrKsh
zOm_{%9F?O@vBA)bk)t(oELFCo+Qvw*YIHgsw-
z{Pij5H$zFjOj{CA&G;D4`Wp)hCACa7+O&y(}-U=55z
z3kco-of;;7i(n7@fHsF_xrCl6LL&rSs?o~OOYdUH$QGn}!rZD3il;!(D}r-NYqw6}
zf~nO2F9Us(Ji@bQ3j8BA;6)D6@{|OA7b3DRvC45eFgFxq_Y-^eE7Vq2Q|*
zTr`A4&>U^}{U@Pqj?~0ch(jv+P12P_&XjnmMyexzXgd#z&yRk~Wd4C*$0{7f?Kinu
zb0ZK4P57mN@*RJJJmYF!=`6qj_LgMwYvL%JYr2
z=PTP5hiylt1v;g@AE%og%UMfp${v7l?vHllLQx6%k&MmX3iI@k?QA%?!lIg=P-$=f
zJt#AC6t6+n;yom?V^4rocQZeK^v*ToC3ECk~Hy!F@FCQHqc<@*jBO`8~JIn7}N%N_Ew#EIL
z`^1y!pkJcG0|Vca({}A^1SeP!1JXHn`TpJF15iAe!q5rnmF)LJ7DqS>p~Cjj{L6JK
zrH#fV;%dcI%jWNN&NQ&TN&mzHTX6+UZdMz@e`C7EqYt&TtfH8H*8XfxRexOqa8(Jw
zWFqd*cL%dArziE{acjU8yha2rcMqFnvL3>mnVDdt*@n>FpM5qARh-S3UbBwI7Hj6z
zf?m^szIj*Pl;t)K3i%bk(zZ@QFGt?#O*i6t#^pRR&3^}T@`w1g#a!wG92~T{GRo-2
zkaXG_?01MmsYGMX4#$M)BB_v>qdFZXcV6s&?c4R$=nWwoig?UUUjx|fpq2ur8tnS6
za>fmo@w}Fpudq&6%ki9ij}9c1UGun{ESGvj%-*r*6
z>@`1KZ;+RcQ6r&i<$lZraB@;xg%ht?sQvkPoo=8^rCbps9u(UVy}U{ODUnkf7UC~9
zpC7Foz*3G6Cjyn_U%*C5gv*hq(XBNcUdgEQW1W%`#Ya<{YLbk@8JQ9z66?dp+lr;5
zKZpyGVL^KZs?pfj%zcAR@gy>c}eRM~mMEim<1lohTtlpM>nO6-Cjx#QlTkgG0J{eR(cqrqX@2
zEcJatM+wi=l7~y@X2-Ds_HNZxf`Ob19u2(v~Y*9m5yOCW;5WA
z`#v8Y&v6f%>11a@t(aT<*Jl*w^%myLwP*RV5YoMoF{^z!Sn3XvW*y5M=GGn9Q|FCSfywvP!PU-J|5fHYe1d`3Z
z9>kgHf@`|(#HwelE^gpfA}Y52)VpD*=KW`;8in{P+)qz+P*?{TVMeqV?HRGiP0r?a-d|ah{LGeh
zxcT50l8--}`BeW2=HFc#lw5MAW&LrG5_xTCy)r^Qbb;;chFIy8Q)lO2mIS#GPcQ*$
z#Wc;-TK}IJ8Y#cH=;Az#Xe}cZx3DJ{>xs@h5LoC*IK5yKSa=EXdT=ez4W`U4T{*;u
zUk1A>_U2cJuCh!@C5kI#`v-{xiCN7)#ztQWC@xcV?ob;`ny4)ew#%snuu7+~b(ntM
zn0YW*HdK;j?le&|#JAPhv%WsFSbT&J9CXCEq?_tV)I@}tiRIh+w1-R}9>9
zwUSoa%&&SjQh3$QgtBUeE8?#Vl-i|su-|d@6Ma3F5!VGXD=PtMnT&RAq5fL@1nfX;
zG6&#bjpSLK@i?PK2L}E%k~AUs$%N%JC2s&u^Zo|4!wN^{wSTCP6OE-NEH?VfAXige
z=|)>Ohv{ko%c}O&2kcVwS`G>uW|bQ@r2)jD7a^ATV5;@;$Kim;A#9_E^$rnV$dj1#
z1&MW^PR*q9G``!m88+dPPj%_~8IhRGadoqUmuD{@ty}#pI$DdH7KBmhPa#W*(dMQ_
zsq7BCB0o(AyAl@=a}=|^a@%vGqYU`W#dy;2!S&=?rV`1g(Q)6X5PWw^OIw>RYL=Hs
z2ijq)<)fS>*=oOi=hX5C$|`C&;FXTm`{6BNX6hDhGtGhy@X7bJ)P%CkqvlB3_nC(O
z-WHB_Q*`dQDB0c4Evt|+V$K{jO%;7d-Y~cX+TCR|d1R1Y)N%3Sew?}{wYPq^PMlc
zG{x+7K%T&SjpvMIVu=EODL5_qxFUWGRvjVBAS45H$VAvpSAC47jBU_Q;y@H-b#tjtpj%eQTb`ug9*uTWJdKen*A0MH$9Lr9Ka?xVclavng6C*S!rOpTan
zC8}>BFZseNPcDKmI-^_<)5W*-p9w5ER$Kn;zPTyk2S(Od0oztY{G1^(JkUmw(=MaU4l${X
z(COQ))nt(<5nXP2J+M!Y?(qy@Jj1y#hT+5y@=qPu}{FmMCjF~uOZAjor72}orrf6
zdUcw9^%cveOD0aUvs(S>#SL^tA61v-Xum`SzNog$s>xFs4yMWp^Y{16SGj}`%jyw~
z|3Fc8R!y$T;aVS@=0kY!0lBE@zeAK*GYUl)Kbu79#Bsn^o#6R0&Cdzd0YiV;{iko|
zv#bQO@5<*<9P4BSi`$(S+Zq2PQ&y~j$%u)Q8P}gE#@DKF$h4}HYoIy^BQxRx&u{I1
z$qc4i(R_z>4=yZ3@t8DTOi7mTj3FMQYL(goL4*hHi7#cyi_PURpcmQ7*
zW+Je_Di3So`AZ{#YPjtaw((j~7y2^G#joy$%qd)dcS$On@jsiP#XcdG7tbws!7shu
z%|2<=bug3vEK9d!NVU&Kja3fjR<3*mQ;4sXdqi-4rkD3kPyfWJhk>S2Vjn1`
zHu`Am4aN;DHw)Rzm#@>6*%V~WDtm4R
z%&3f47x;us4z>holuw3GF?*lqwhpdH^1B}Too@HMu`D`1juA4*YWTC{)`xl#QhCPb
zNW#A&*LSc$^2j#FL^lO$hcKoik9ckrc%M5Rr{_}eOOqp
ziGT&obqEBP7!&ZI(_CPoC?-&Lt1JqK6HivGLn+8>ujW_56%YOuj^6EcENxPgk6Ptd
zjo2#|(BRGHI9nRn9pDL`_B!TBu*ZNoq&B8KVZtQ!r=9$Fcr4xh6#wYe@^`4!1b*FD
z#ryj!gzasvx$gu|y0vLUknn5C-380ZU1p%}XD_PJI@sWOLQLS}hA><4T*R3XV~+eC
zaYI|$S3xhT+1+i5K879ygjgmR&rG}C|Ql+TFEDlPW+
zWL&26Kl25U)+b3zZ3z|6O>ybpGDU=DZ8F7WS>GHwgzA-MSY=
z6VzK@(m4KOInX;aOzZ34NeW9^P|LF
z-kv%R%tvL`r_UxCQ!if+-TLbB@)Cm&K$@&h8+H4k+eaA*=DkeZd(B$|UBMU{8}a%D
z9YNz+mnqsh_>HoGKd+|UOr<$jJwVJR;*!s`G7T~O!Bp~je=vI`b&VFD^^qp`{qDRy
zUZEtX_NQIW!BmBX|77iqSlN^*T4N=mO6sKCX9vu`9TAvfdDwoe_o*9~zZ}fk>sl%<
zwD2T!Qz#OK#WUKhB-;p#JgH?OhZ7T_Uwo|M_F>nq)0HSZY;PV*2zoDC3|IHy*-07#
zAbM|hF*9)M99m0Vd0s!3?`FesEN|+@acGsOXi({gc+h49Cj=d|*lp!66H|PWP@47^
zC$BU^&g`?j@g){-{zyMmFXOf7BnQfbJ4$CRCtPeZraI-BM?EI+PXqpN6w)troQ
zel-8$`+oNtG^ZdiUCNk}hkd?wNLCbks!sK#!Zw$xDBVu9@Ud^Q+Trw?G>}^{Ej=&q
z0D3aOV1{$7zQU;AMIj0Gv%UqJX+{fIxtQQDTVT3hIbt}zU5V=vVBO2o9?%t$08IOt
z6Kj9M%*0aW+ZF
zGF^sYW-vJFd16kxvIDs3&*V5Xcd;F*@>%;83o`0g_8&l^$Ub1a0
zkGXbl(A`~$b>vIF$*n!v1qF&P{HenNfGul1YOcNVgVMTm%`4>Mj1mrWb|~SlWl#L$
z!*|QXjN#9Z`@@HuqaO8RG1j7@tqtB>y#tdN&qco4H6%00xGh4be;qP`n6FN&r=jk;Y;mNV*
z1Cv}`BR|UVAnd7{xYfvYE3~@m*Hs(j1a&Uop?PN-L5#9TwGC>uW&P
zSTD&M!BNB;)fJToLE{9^u8xCjxFF_5YZ!feFm$y6K4aDGlE6vdzMb%DYO;rVFWc<1
z_mhst#eDt(K_7IXYDki$xXt&nm?ig=a`7x%Et&vHo
zX#jV_fcNCl#VR$r(7Go0iZZ7sjQQmrmleU8o{t2e90=fI&!?+}zo^bNA*^#h-^Al3
zit{#XS~LAjbt^x9L+%Pm@YAbt>UNqN98GVh9PEY9h2`wh_@xlUNSV03ubZ&6K7qK^
z`Kh_rDMl;PcHcUF(Z4R=f8O0f&^eKUD$9Y|JWM=9RGT*&nCT)hn
z+5R4Wpp4pxIknG)Pqy@C1KszeG;s65Fy`1q~f4$)<*
zCufX>vJ>Et_^7`_m0o3^H~#7|18VASDk2R^gm-gLWx|in|+`@
zla*WWM?vq_rGYRrfo}uya_vYyH%Quq<#^SzMz%lxYqjh0t*>DUw%xF2S;@C+L<|Vh
zlhN2x`*l7C*mKX&5N7s?6rt7e^6P*B_j~c4aeU<>{R2#EgN~#-uUw4B
z9ANS??a2H#k$!w(U*1%`kn!rE3oE#(Z)0{w8CCBx{#=pqC$}O?gt#mQa&^wptK~6!
zA|;@G$^MpB^dYC^k^<)QZ}#sFM+;t7fN_jSSd~Rb&(JQLRRw&llO
zB+@&>dH2T8T6>1&Gc0w4MOZ(0>qku8Z2GCW{g$zz(2!Yr?FwUg*RLC@4%-V~5?{g;
z98=inB-@oP4r;I6+iX*8G5B-GM7);sO?KBB%xmf|n=avEi&mTlNzq%hZ=cG20@W^w
zNFKv>J(Cst;`>eLy^$H-lx%6iI4{
z+Y}=SeziVXhv@i=knarj`y6Lm4jLNqMG9VZA5WTIn=+9ymAJClbozIQ7;BmBR3_LJ
z{Q9;$)kv;-UmgpHqNx<=nn2VTqk^p1_)!BLi;_dl|2$Af9emE#-1{iV?7P_(a(GRS
zdCBGDpnEL)U*zWr{w1m+-8To_uosq8`#F2=c<2>*VCd^%``872u$43AaP?K9c$%*g
zB2e2#g2Xp}d!B-bgb_kibUB<7hCCE|7xsCM-C%6_1c5jKM;~pQnWJ!31QkNtp0-bw
zA%JLP(bn9-w%2P<;Z#VVKw0!^*BWJh*O{US238$@tAsWwn_y9KtM?c8Puh?nOO=_1
zO?XO*7?>g9O
zpIYM-PnQ)Ddq1~QE3vd1NlRNeWs$xq5!#0lbNPx}R4n7A>w5wQN%MbJS|(n}9Yk7iPH!#G#?00Elhvr(N)DskJkv=SRIqNW+YeaYRRc=0t{A
z!PpOa>~E~R*7Ik8a|M1GGZd2DI7q*Yau~43xm%_R2=Uxe4cM3=%$D}5Q;rqxF%7#p
zpK=uSh3pPt>}T^GVP-yc8yYI!L2)fX&99n!k}@Z*f}>i8RftV}E&qK2^(vO{iWd7G
zNV<0Pg_HwiG4z?Qn`fH+5a-8?yBC)~_*;3W*f&UCL$*>pcI!aU2L@yO`zVa^`l%dsUqIu_ONG6y!VO2E;3L~<43)@T%l7)?*5~)#a~BdV--=Q-<96*uuz3t??Xo-
zE5F&bbA%jQt@%CGzAo3xy_&_J{V#cnPY
zgU%??_E$$-4@eVqp1-|&zeHn`p|Z{$27B
zH^`0P1n>ehn5M0l`%=Ymbw^_J5f3_V03g+c|LJyvV82ja$~0}ZaGZLZznyV<_GNlH
zC@-$_QsjNX8?);frZiCkk@j_`p$5;bS2Q5i+>MMDR^KO5#c*}b`zk1B&9}jJak{dH
z$&hW{m$?=+JEtvy*SD0Q=6WYcbc%s$i}&v)4kwV9&eR{+qqO*(M#JQF*+0$fv14wGx8pgx|~qE6-&YGWWMj>?Qm>TL{JU!|dDznOy;
z*aM&xOCSX=f`0zP5x;xdLIIj@@oyx?*n>1h=$TT4adm+=5I427LwU#ZUfgSTdB2*A
zFDcR5x<6lyI)8+BL^*Ehc#v(VKM;=2agm-n1(#)>1=}$~jlmi^m#Ky2FEOcC=x9%~
zL*D3gCPTagp>0b-5b3QK*79k@i{6fuT5yu@RcOhd<-ckQs^}7)ZvvpsV+2{q=*7bw0K%d`c*o8G%*U?*eIlSt(?~x~-&#+U7TL4P%(OJ3q_O{d2U2jP
zOv*O@|5e{$u5Ym_DSS5Ryya;XbX*wY?3t
z^CQk(ov$p)o~|a3lh^0x4%>XuoLo7pt8xcm1ft53nH~n?t1*&**@XOGZfj5P_rfa_
z4|F?0eOKU?gQWJuY(nlLKdvz6h`G;y)F3O5MKjCj8#0s)V_7oQ&6}Wf73I$ADTmPl
z)&UAVW@~W09M<^E2d+XN%!cEgD>ZNRz}Pe
zkjO@o5h^W#4{Vct03uK3d84QG=2W_jAEh=WcCpYE(;fbdodF}Z?|`lYHDD<
z>4l@FE{acdMS}Wjj(I=uz^VST{${AT%bU}1Eug*qOT>P~O%u?%N$CBgdtE4ZvrQ?S
zO7@X%pF6e1-7xS)`C>EN&~nrdl)4y%V$UUr5bXiR9R1S8M$}FCKD=*itzplO8uWFc
zVol}fceCl-D)m^zbnXg^{7F|_H+AbKy!+xNlDeh`k9yX<3?Gr~7^cP<2yM;PaN7*Km?2%25aN9=jffufHVZZ*c9JUVo&^d%g~U5i9Q_oYA0#0Fq=(Hd08UXW}&CAEBEAtGZhm7df{v^+GuxBRTjqEXU5lCcK!GMU7z?SKT
zFS=xKy{Gf{Ay`ng`8PV!0o9M@Bub%MhX+4^jx-5Iw>p2gE{Oj)3p9eY6&K6XG^{1E
z>yBGDmg|9CN|G25TwcK+GnM2i4v>@pqs_X`m!AiK`?m_$8e^0{bT|}EnMc>(=>;i1
z(c7~FRKA>|$^&7?lk=m&N0nh;fiXa(48YP7i
z8P~jk4G={yx~ws(CU~e2EC|yPa9{pMBYKPFEU2By{k0e+d0Yjg&rCWUowUXt&9Pe9
z?ZtxOAg*1H=5p6vCj;wCgMj2jHL5lg}@
z1+wsKtZx!y)l0?**I&-qTukgL6aa}{MR)tRkl);93H;fg-{D%Ame%~50wHY<$OKKd
zut;D(TYIk9c>Z>->=qCE@S*&amYD`4d}RC+BS>G6j~;=zCUOe0G(;C0OCUdbg!Hi1
z_2W1^kf8ijp+obW_w<1ZifpxJWrqMT8Ps=$`i?-TWNgYPopRmnPiEo&@Q^mY{nXHB-QKx8H!-;!7&YtP7y}|r
zdCRoAg_;HnL!lNaAfuqhs8TGbu8e?2y1W#_rsdhzUu?e#mAaEUOP0MqCNv|fINRc1
zxCik(*KuY=suoEzs;T`e;>pIhp9xVq&?8Ff(0R|}gI^NBjdBLt3?l^G-_#kjb0<}X
zBOhXOIpuE)ZWCOe{|&-0Nj4>g_`+*W7S?SE3=~vPfNdl=b9`oXQ?wc|c!HoXb2rD2
zA44F|@R!GcmC9RO|SNk1zJFy2)
zL{%IsM`}{295tM6l
z5oY6levLmXi>~5WF?c$#Pq)p5FD-6*y8pB9oOA?#1}dgn^D65v!UdlmjR)q4P7kZ0
zN{tDnxfj(s(7_6FBBxKgcr&&uSL{;lX|50tD`p=b=-0R)&$oA$j(dlQi*JQLtYI@vIsOjnXV-l@QSTMg
z7yH;ChQU(&da0Ll*6Gs_aE?X$*_
zJ$DA$V|IV6#^YX0e;4N7t$uyZ>=_o~E08M-fNmWcbRRw-!0=&GL38!V2cG;&#RN;*
zhryX;kc@yIK!eOaXP-aHB4{Rkk+Ux!?r=XVYdP2nb}IBg4g3mJ8~~K{63bY&E0p98
zkky(GQHcUXwE8R6^>fZe#)wa2JV#A3U5U*2cRROz%NZ#I{OmF{kp~mC80^j^_pH+N
zs;2LDkDb?vi8-DTtbVA3Sw@L11?e&od7LU75SM(84@d__|LXYCbYe1vbp?pBeI_7L
zHj;==nFFFfnSN;yPYApLKQ<_mAEKto6|O!)@2X-D9SU44q307oQL~WJ;N-oqZ)M~w
zB~-2yG%cn>-Y~!Wl5E63T1@?nJZ7a;1zgvMRJ1*0VF7jh#H4UXU6wNq)tW3;oVi)b
zll|c+OL@)vW4zdT?qU;(QRLDPi-*T#&$6W=`*28b0<8$zIXQCrXqt~lxY>JOq6PAc
zvDK>{qp8xEhG*)uh~cizFiu^Rexly)UYh!NV6SU>)f|rx^I+juROOPncduw?=gWuY
zXpe-1xk%sL`k>3N9WYbYCNX>dp2U0GuT5qRXZKn`j=oG49jl7`zEX9M!aRvCrATS|
z765195k5ctc=%dQoe#g?%Sw$;t$}5k(9wjEd|F5;xSiF7UwIXIyAAx<;`jR3PI3pW
zg8Ehu7^ha<32+xy8^ZVeS0+M}=^QLA=9ajC1l)^G-ZkB9Crdq~S-xqnK(_K=f6%Bh
zby0I@*weI3Hv{=%X|rh8e)LJY?G3DU-)&lFd1?;D>o>NCIwSlbcDMc`FxNe}L%%!a
zy?M3;+*$CiFv~Y6YWuR4@UH?FUZAt|zcTOb%Z|doB*ptqbg`SeZ^J>fgp&0VA2ZPY
zmGAYnW#Dj5klytdt8$p3xqy`fXZmRU^>5_%w`Id5`(xGfN{|*Yo9Y;Pn3Kq<9WfvO
z$=H>u@81_v(iA|I(gE*Pi2ey*7}INQyVmAg12+w3IZO2_Y1$2;Ns7-A1r7W7iy
z9=OEDmAEiZw^_1-jbe;LbZCPKH&bRi$L0?a4YxX(9F?sTM=uZm>6Lvet8602nRcV^
z_JbPba0U~mwNxR&w4{4#_ZC9)Xs=Y`9O25Zv%MX3@>mBLEc$K_JQ2|7lzURG8X>P5
z_-!%JOZfdyiN7@iu>+z%_YoD;t!VNpo9OUbEv7`MDJJafGRxS?uky-*H#?JblWvm<
zkJ_25tS`dK(weKT1&)9ETC_)cO4Tad^(m}O#7ufHMouf0{b-XM>aXFh&9%Nri|R6{
z$Sx{Y?tZLO!`oN>4qF$_7(MrusMjl1U=!z)yQG%@_=bxzRy)6NS5lM{(CwEha0P?<
zG~wcs`>Li1GK8!fIWc*wPr8SAuC7_j5#?ufm}adW)7;
zXJh)vi<=jClYBuKJKQ2)i+iT;On(Uc(ZXc`f1&_t0NQZmJHbYirnyq(HTu-ZwE#O@
z{jzvbdCSNXJyJAfeKN^~e*FA8j?KZi7ICgA#YVMlUaZT}oTiumvY7XRlr&;=!f%`V
z(~)y_Q&OQ*edOa%&T5A{v?2cT`KcGTrd+_z>O>fLVc5y1P{CJI?p@wP6{9KzHg*QX
zDwk^Ky8&gB4PRPX@}&f5Lmk5$_qDB=+rj=#a$)$n>N|+d;FJCDZw>R4bCUmtX(qjL
zm&Q@c`;4v%O}rU1ZDdgP3xmvO>Fn!(o9$O0@^qZVvsgVf4t>CnMSU|i{#GJHxSApz
zX8M2^iWo(JGx+}ng^<)pZ@^!za1lXE3^qn_(UbJ<-4%m?x|2Gm9`e{FcO1_SS6aBd
zg$MDg;~&QaiD>B;#@wbBE2s`K(jh
z0KYx)Q2y4PQ@Qj7Dp^{h8o$B8^0E$0^13oOgkE)F%BYcp>t|QBYth>Jj_*#f7e|`<
zv0`vVsgA>i7ECgi;Tv`anz#3e^i
zM?1&dOS22l-n~RvNo21wyKEmXl{`8GJFjf_lXD*9=7p*Zs%ZWNl1;uB}qEf
z!=niavz(EYn!9ynBY|SbH7}=1{3-;c+J$~M{$?3-56`m&8|xe;Fs>UG
zV<#pc>GtM*$eQ)?#+rYl_@gufPJu(WDus8ZY{yahTCOic
zZgG=Z(sY^bSTE=G4}^<>`d)zu0BIW7rK3?f>A-!d4#hrGq?%?@Vi>3PMteG=Z%ys9
zCkEMy2LI`4_{oFQfTd&D*mqalGxc`qoW+H_J5U!n#f_Ss@82gB?s}c^lV2@`ok|-s
zX_A+Kj>pUXT+2}l@#Qsp1zFI=?~8n4#J#VHDm-wjT6;2)4bMf2>8D>VNj5|B?o(S7
z_KfALd0XzFc5(4(i}f*OwrY!SWcgp{K;OynlxZf!7UKGv-@Ow6o>)F`3CA&#HL7{@
z7&?zO7wC-6@s;Ir_es%UmAHG-8_k$f%TUe&^2UnqF@_v|&Cb!A(y{#EbvPGZv5~(g
z3Ml1}2q&udoyIAcHi(LgkF30aKd(Gp1T=+t|Dj6^=XE4#J6)=s9ri5LeSs7$jFFJo
zzY44ku4C;F@l$H*E~^ha3d&{Ehl{{M^o36Xf6id`azy*S6L1Ahj0&nbj2tDi&T&M>2<1tZOPV>v=e=ix
zvgA3JFPOBV(r<`C2iF@11CrEB7)h)^TgeojhQ(}nmwSXYBVS*U3H|ZzNu`F>j}8G9
zcFV(dM+TR7S7~7H;ro~-$I
zAz^a1ekh!&UPQ<@-=6bC);}M#M)e#V6&}teN5a5H*vfCCqS%YzxDA~cmjwJTjgc=;
zU6Fc+8BhQMUI~|~uhaoedWh@GNYr0>^~PE4;>FccK2syC5ssyVr|CxeUYNaP^c!aZ
z2UqKI9R?V?wfFOvDi*%bd{1F%ZaO9eSdce
zPz^)Jfhygdw-^v`TN+Zed{x#18`T_#J?W(3(gG!BFpTdZ|M@HU81=-I)(lkj6C|lv
zpWps;Ns7_`sPaNzWN%*;Ygr1aG1%)^()d(kPz^}fggn*EYqjuOztQobUN;5V=^WwD
z*74FH@Uo0%F@+Ge5Y0!C6~E(In!YItdmi&
z4$TW}C=)%pb!_lHLz5xqx~Z5@)5!lRFuh*+W{$!>Cr3AWS^9s{5$~{-4fV=i@3*L0
z2o={vZ2AMO#Zy%1bj`Sw^k~5qL!+-py?K4+i#{D!&Ej*hp4Y%li76HpN}X!6OPF?0%+E8zKGE$lgru%I
z-6)Y|p>ebW-6ipKUU-)E#jN6~Gw9t^bYvCY4Y#9Xrtm~PdTVJn*=WXcy2X_SKZY8;
zraGR*Fu;c8g3{A(@=P!nosm2_^G)fUmNvA`gPi~1`
z=MB|H=icfJM6e>N6rMC|E0o=fFy_ea5Y9dc`?r;>e4E*RS^|
zVLn@vv9BZR_YNq*TQUb0#tgLqDX5BK6#$_s}Nu+m?slt6}*y-yg*=
zt+HCY@QKgG3;U1?N)k%uYGZFJ;2h0kpUO5hu
zB$}4Xho;*CGd5&CQEa$YjP}7jzho(Wma!RilL23W_*L67NJ`i+=&|`fVt;HOM|acB
zkxj(U#dxTb4%Tuj6b3(J_DnxoQu2weO$)D#$NuGK4|kV-pRXwW!CtGsEJ~xe7$Y~f
z6X!d|4!fT7Q4Z&8x^LX*var#YUzCP1<*RVMJP3ECI_q7OFI37fXmkoYkP5u?PZ|o;
zF93B;&}~SnFALdkDKN@@shje{>aGcb*{a;g%500C?KwThcT74dlzm#oC(iO?;IH3-
zYuvUEc(c+L>qN1Y217hICKO4XYtn`qZ^afer(H5z#D)E+&drZ)xM8>#a!3y&#q+*m
zl{$hx_4XqdQy(gtM%P1WK*$MIyaL
zHY1JQH~O5f=dVNkhB&LHFIby*EA2+lVCkWAlP6YFNa&1+psHjSSyFQ7kRC4x=>Bsh
z$WtP^0*j6ACBOR@iBt9u4~nTDDvwikd2;$eO{3ffVFRB{vJo@P50Lg70h*%x(Bb1=
zIbS7M{@8H>{qN|VFG~`XYL(lT%nE9=-1*ZSv2L`ygL}>LK~bI`?)@e929#NnEOWe0
zJxtCFrNVG28&LZBBXpR8XS}|A*zP(C5t_SyOAW^e{5g$D00c2!=-pXf
z%TI***~F}#o@+LWEn&)Tt2=z-(l6OqW%pEI9et0GO?1Q<^jUFY2w-3mn%HY{{HM3s
zMM=@JrL3Zs(TSpRCpY;-!IOwj{%y?YU1~{^LcE4kcKaNlVj<#!frW3axZ9)Jp
zE|bM?CPtxRnCrld0Oi|%X)s1{5>EvXIp1}1d2SLS(5@%qhrNPC;__wgW0u;d%!M@bvK#w!c)6TEgvQ9o8zTTT7hK*bn3!0x7
zcfZ!@qs%DBl>_|CyJ-QSlw--6ScRruF=E`Ddd%UkCl1CBa};=*T-j}4Fwjhy)6oMC
z2qX5vX&SZLH|;$X>YcL(GqX=igOhGbK)G9Q{1l68=c314`&VYwpf-*yxW~^G;uz6y
z<$qh5D`z}kfm4>M48wJXLcl-gkqDc@Uj5{7zAI|Zh$^%Ezw|7u)>eX(t5%yvTj$&j
zQ|Anrd7?H9gXx4cH%A8qTuXMWzo%Ylag8g7wS5ZaQz}*W_bEO^os-^L$eSz2T-(Xt
z=dACRaBCoA>p5n9`*%CM0cSBgNRL}|;o{Q2+|56A6$KaJhftfwGKbK!07sprqkipl
zwk!T+ww>1Jgrds0Gw1iNf%`C*%Kpid(S*~i)0ifjM-Fs7FCMB8Pj?E4&89MS3h;!h
zNW)(XSVZz?q@faB1G&=DP!_b6GD5zgiMEf$i?hVV-(7@&8p+0+#?+ld)UX0cEsX1j
zdwlvL4_sVozY)%Ow|!V2F`LnOY+yW!iv0blM|&@Mapi<6@(guyac;7xP>^bTV0u!#
zc=4n=*EWNuolUsEsEf-=C0O7J5BFBBapW}pC-|QykV*mK7{CrzWg|)gOIqt;>==dP
ziV-0AI@jdt$F67lm=OK0h{QL>Ckdvf;>;VqYA3Ew@2ht-7yLU|p
z(|la9Oy_939pN2qILAgh1&DFP9NPhgbVfigs9j~X2-yO?5257Y&s|4P1GN2h6OrA0
zS&Yb0MG!cc+jJ>9GbWuNfZp|2pI)WLs$h=8UVk9G+oyto6Fp2f9VND0ps-ncZ$TY6
zslf+=0d<$cEzKpV;VxAm?!b@?A-xxERB=E@%3v~;B58)jiS$TF;Q=V?f_w
z?{Mp+m6$LXz=&}G2K#@mZt^X=j~|9be|{~3oMY+6$6@ftVdVn$7;UPDC^N|l7nMsW
zT$Jl@b9=oolaT=`HQkH>+CbV`6=q>JuSR&a%*;`$IB&2clyna1VaA2p1^)13q#z!5
z6&@4KILe@n`Oz|cv~|z`(%tU-2``aja9s`8OXp~@iBg~D+cB1FKvKXMy4A$_`FZ|!
zjy^d`ahynKM`!%|4grb%M8FJGibe-5h6l!EbPA{gqD&+$_#|;4X>PRo)Q6Ch?r1Th
z61sJEupB~3`VP5slV{oeNC(3Qc%t`!XdXgx_!ZtYyiMa1qhLc!8%{raQ6svnr%hu$
zj0N-lWJH`X_|^Wo({9q6?%Dbb#4EIV-?oIsro~-FHx^9=sF!%7w@5pMG#An{0(BIJ
zyK|C~CxwB&DT)npWbyBKg{e${wV43aYc_WY!I2&L*$@^?^U>#vy`bp#gYJGicP`jm
zM7O3aE`yH1b7`2h!2rHKGoVhR*#&TAX^h$9CEU+ZzuP`h_nsh72>wb(v^+Z5Y(Yml
zUll0JDcXC;eQhqo1NZ~<$pGv06fgC)PW!OIxDYYSGt(2T=u^FBo!cprs1xWM?PyD4
z#YZIY7>vQYF*Rl_+5wiJugs*}oZj`(I4KzHj4%3m8tOfb&`PtT@y7qJ+jpeNPQxkc
z{u-S6`j`V@U&b>B^XDGnz^Q0t>pGKWs
zlh}_04#8g^OGwNb(|E4r?#?k;LVkt&?*)iw)k9jEJ4;M4lF-01e#$R;;D)>1`jzx5
mvxI*S3DLHa{y!W%PNK3TLhaU?KhjP?0j5X`M8gHoL5+sTWC`bm$pf*VfB7z_}=bSTw5(GhV1_cxqB
z7sEo@D}L4(0oOpm=y;b1^k!yH`Oj*P#6-UY*ub_ZNiVP8(-i47*Q61o07SqjS0;km83P
zOjL507KYYfj&{XfzpC<2rT2k>*8GVPk&#`cwuU;oy1B1^!idAe!X8^#uyAv8*Ecm`
zve_@ceDUJNw~P$kv6T}j_s2(A+NTN
zRy^nS2F{0#3SYi_xmiy{fcgC8%jVl5f_5scrSowYLpDn1qV$5PTFbw^Mk3QZxE3Y7
zJUl!qPPSVswi+pFVo*
zgr8sS9@69u{aQ9x=@dC(95vx}9uEtUG5{Y(K5c-gvya{{o+i->aR(x1FR>
zyyS;;|u&}2IvZU|c&8-_pMmlu&pg7Eje-e|Buo=JLK#g+ZruVc=H+as*
zTXtLVt4qgmy-PlsbwG@}tX%ESI=z@b)b{7|?SlqZ>Xc8k$$m6RGAsS&n<#?G22xGF
zg@t5~qhgqt9g%a+KtJlE8~KKv)2Aqr>4Pm@LAZDVul!>zE!q&FePSV(`mKTLv)F=y
zHT3iTDnmiS?DWRp-vMEGLW$7V|I(IxPKBQIoDD`~FG`tjq+<0q<#xh%$NJgS2#$FFXN6jh`8285fN8MS&ARNhMX
zve}!QyL?;i%}228ekrciX{`0Zk&=_cp=D(Bx1MR->U*Ur|MKOmhuot_7@?t|NHMWQ
z_R5@`kIl`Jzv}9~fBROmWS7J~_9`+WVspFIa|)JJgTMy!=N~`9LWE!A*HBniw3}w1
zhO<_WHfO0%CPt
zAKu}TCd${rDi)NI8pzdad9hw~EPQs@Il$+#d&!T$a3{LDqPBd4d6k%*^j7QG#_6C#E$Cw!ZY;Xh?bs*&
zoTujYp!^9_H&LjjiHYR?pCiwelp2GnMh2@rX^f2p5Z|)0rYCDLXXoZP4BDJ7T)2P{
zBY21{d^oEs{-Mmb$hr8<9UOMzaGKnuzIoT*n;3$^cl_2kV-^U)C+ZW
zv6cRxW65iS?ERHlDJi6eJHy)WVB+LgZLN`QJuF1U8(WpL2!sCbD>J1^P)oEyMSM0DYa9xOw8t
zVSenoFNW;G_xbl;UI^6gj(b&F+N>b=3F=iN8(v<1e$L1m++el#(JpxqwYv~W;9)}h
z!-?#%Ih({i+sgB^a<{WXY>Sd91B+69G8Gk^&eyWy;$^Pt5aOFG9M!DU>F8ua$JVvx
ze%oCMj(pRC>^v}PRO{J$A79EU=ChCCwD{Q%{y1GjLMZvdBAH$#?TwFDy6&7EZzOXI
z(bN0heHO)~{Gom98k7FI@`X4;jRm6WnI
zS`=R;j@>?YOWuI_9^2h^hi$7;Z1LdXIBz7CK-KhzhZ?u<8FTsI5KMiIxA>fr;w=`K
zw6L7vX*V86blo@CVU3}_VegKRJ8NmF1jb<1DU(Hs_WhYgnaMw&eO4!`38GuyHRxCm
z!ry
zozabC&V_-Kb;r&;x5MLf0gR5jTwD=an-!Y;AQEzI+%$!>VN8)fCKl2$$
zqs+2cvF!NLlIN|zsAcbtyCz|B_DD$U6dfCv7lp-g;Dd@Omm|%~XdD&llB0i2ST*i)
z5zj|y%{$h`*7LFVNrig9^F0LDBSl5Uilb#FfF_T*62ep-ZbVo
z;n@GpV{UG);$V7%*U3r*u{JEmVc3yc-{`R27(z=o&8fyhdT#~)W2F2gQNJwnQtcO(
z%pZ1@CtANC{?wEC*6N0|-(tyPI6oS2JvcnfwbaB85k7i0-|o6!k$k+Od5$c$7#rI-
z-0dNTfAfAg?o^)pl$3Cb&!OB06}9n&u^5l7i=j$r;I6}tg9@n_#w1RWAi{J0A4#g**?NX!Tf9G-O^t8@!(bc9o
z8*QHW(z7G$i&)(44h)2$QJ)G%JF3Tr#EiAE=<{Pk1Jub@-?5*vzUrwGp7P<~s7iI*
z5H=iRxJe@tv%9;|aa@YNsPLK7-G`%W!Rg5!-XS6yEq^}OQIx3x6pL`JNN}7O6LYq{
ztK^AGl)h`B#NheohF8ARk?)!NRZjfkr3AUDjEwTlNZ;|w;~cvsKeyvG7Khcr#`2A`
zKmrl9M$NNrY^4l`e1A#zLkrDY6a!Q>b5?}?^!`%eu3ap2^&|S~9O@WGI!Ajea^j7S
zyp)DXo2b6tgUraNV
z&z)|uW7@;=Hu>C5FJJJ`=YnWBr_mp+GKaMeG|Y9p1POcLXR*dYx=7_hLNa@-w4R~i
zR)@edt5P)t@r@hPcKiVWOM^y*f&K+AS?VTdfS3Hj9k%8WSRq7XpCgM~O^DHvVcHh2
zZnAYv;=o}<-io~7GQqqr#nH9?`+r8haNk77hOK+}>4A#L>kN*bTi>M2WH
z_jTB9Njf7f8b7BpsO+=|tM%C*9IKXENIRCHj271N;_jd)$C%Jzm9L|pHII!?
z?X531oAFatvruiA`j)@rH)x;ZP%gh0h*=(@YePP0_E}nLUB=4D3x2)(CKuvlNzQXR
zTuvB+_TIe!O^APjTlGX~U%n6$+bxp>;n28!%++lQi;MgH>m4qDAn63&P(Isv#K)wh
z4m1e*_Ko(|tv7~A&e$1>rD90rr{Q%iv8VKR`50g;+O3gvlxzv*9;sBQzYZVJGVJc{
zeaT}!T)#cn3SXaeo}Umv5pmCTe{~9enlv=@+`U&vM&>HI&U_Zde2bqyP9@(U*Lp@g
zEG*1vw@bo#f5%?DQu>V@7-JXWaN%!dN^f;E=lmyBxvavGznNH8MeAn?r$MKf;H<)7Bw1252
zbH+-O`C4}@y-F!5xe&dwI;Vjj{~+ksuWc8VC^dd4K+$!m=ExcBwFp-H(8^*&cq=W(
z{fL%u*O>*&{b2{A-^Vp(JnEazEPe0;0AM$&JB~Hf5s0(_E|oSgpf~N$wC;R}hndm$
z^yN!pF)^|AB`ho~q_}tzH`G>}K~_SDreo|Lx#LcN*qQ7&Ak8wyurpr
zGzC-5yuqQ?>PiulP7);Nwg2V0Ub?{e;K7%ytYFXcIG-IZjP-n#OYU$-jWwAUkNMV3X^oaxOUH6xc`YP+qVXUcCiV2G
z%|l$~*9HEiwhJ#*RHP=V+++D{4W2{-X(A(SpmcdfWGqD9cA32@HG4%C#rt9P6^(Rb
z>CItPwml3ZzJKvQuebLf;`y{|F67y+5K5=h%m`T9unrF`7vDxjn!p4r&>%j1yA4--
zU$>;w=Zi_%&iG4D)Xt8HUUpsVc}_L#Tlv3HqlGd1hyGy`c?1nplNm=>M)3#J?3G}#
zwbMOWbl8b|t1EURQ!VmMYFluhj=|!lGB!a#*QD^(zeujaqVo-1m9poxx^-H%KOyY8
zu765ROCTV3{f*|P5RiAw{f5eK6NLAO#q&Ipj_VFAnz}X4^0Pq!oouqd{j|-q8S~gC
zh|uDorT@xJM89{3O)~NHrVlz>aFr3waWkLu`Yd}tZ8IQpKN8
z*RW*PzXxEW>KG0SO-X;H-K)H8dSh{|+b!zv?|yr~{8}lqBaN#VFTVv@yiTYl!hvNwlVh)$-I9*W^gHA-M9;U)qdCPwX3a?1!>ucA
z^h81~MKfjW*2;b36PajW(CkbltklSm!D;=nU#BpJoF|WH3+9FJ@30aeuQk5X)bp8n
z>3UvrS!juDq;W)FO6czb;^a+Wc_A!<>~O^Tw0}nO4E-cVe`lDeY*}OLH_47oObBIB
z4TLDVdOrKh&mt)KG#Z!tVtMbIePdr8sPd&*G`3>
zz!>Ur+|SQE(6MGJSB)FFc3OqrZuNdpV(eRn-q%0ferK-R@`WsITeioaVz^jUhChmL5A4l=wIZXJ_Y8
zkK;8o&3U|0o&5XvZ(n8=WLg8C)4`pMs{Fv;DNi$%a7D_BD0CYemu5F+SWGqsdV4>{
z#s+AU_4f9v^?EBPC=3-C6SYRO8aBMYu{u^B4M7;{Ww|`3Egf{yOA>M&EU~_gfC`FH
zPH2x9(n2j8UKES^l^V`Y!MY%FPDv@L&z|RJ%o-I39G));2`6;Our)0N9{>Jrm%xqx
z_%UXQv55#ap3%lyciQDmGK9nWXiE&cZVco%U}S-a+1eY>bXCh~H<;OKq5;y0MnpuU
z%yC23!GR}Hz>%JZhXitvjSa`Wd-rDN=V4x+xVZ@;6w+nyLRK#;<1ezD=ztQeGlm7$
zx033@q>tj#>bMJ;8q07~fzDCTBV-}Lj)QSG(rFcQ^A|4b=6t;6Hn7Nls@z@CQ%%CG
zTm#t1sDgF
z=yPe@WdbhTMJ)OJR=f;O_(S7kS9HaYXnIUa{Azps*!3H8)-N4yPCF-t#*meFEt#q(
zHy_&yVGCk=#l_up*Ucz^64RG2U)G5>K0oneahU@?aMXT4UGQYHZq#k39hg%z
zk(#9W{JE%}-mFRwE-r3BP|zJg!G!nkuO!GWkoSloXQh3$b1W1{Rz!yUGWw;IS41ri
zw-7G-15
z9D#lHCQ?a>gpZGpH&IsU=W8(?9XfV)_KO!UevFF3t^u$EV4%LUGvw{ti?A5o8||Pn
z$_4~Ju3qHVmc7G9er@E-gW`6720_8WkuOq21{@vsiKjR=Ha3G5&YUdKfrPik6T0qL
z33mz~Gt;^FWrT*{}3RpJHZQ{xtXJA4CVj!GVDsb+TK+BXh9_kQQ#$
z>5%gmr26if-;p^a+w8q)HI4wba~zkYM6=rCl~QapBcg!vU*#gX-_ya5VF6QCgx6-9
z2DgNkG9?apetdPqgHmx}?IJG#>K%_}J&s(D(<7z~X(W>6Sz=wwvOP|-z;o{@8$xRG
zr`r+ZP%Jo=a~X?wP#}2B1}D!v#l)~afBp>jcmMu#@c482i9vHM)>Q#4FhEz26nyDAaAs#&XOBSwX`1aC=6J@Y%$ejJ_1Nk_fJX
z#d1^3itFMmYjlGy;-FRY?3b!#C6?$wmbyxj897=i@garsHXUN*5i3Ei$DVM41J37k
z%Pk5@ni+@yTJYYKfj&z@{MULUMv2@iu1GM0@Hykb`v5f7bQX>b7$8QBPann@vz&y
zG_;UhPYti`#SM9dNOGPhr}j9|2FxA;=@MoiFyFVo=c$ZTwdhMMYvV0^S3PNzTa_|v
z;u&Xo_WBwuVUe{n*e1%%B#)o0ptSkv=(aWO@fm2cha}c(uOsME9+yt}3SY!IbS>pD
zHijwJa9!hGuS98_p*+t~3AdlNCkhll!iB+ci``Xp+m+{6{KBop(Pb#=nkBR9J+zuz
zs5Bdv_z>~-U7|~c_DGRA#ZNLqLd`vM7vXr}&R>cP^~;u1!z(U-GSZJ!rRw-B5izi4
zt$q+qLcr~aK^0e@{?yL6PPj!V8y#4P;N6!rmm4+se$lzKbHyfxEOWv
zDAdJf9AUNcv)Bgy@+`*q8-vJuVz>kR3zT6qc2>91yQ8ex@Vr&A*zT4R6!~9+feJUy
z*nf@IaQV)@e2vg}>wbpGA22mj5vDl@le-C&SJpAPzI`+_?ix
z>}t|@wq`Z<18HR1-4;H#FmBU*`B&T=tw70;tL-hdZv_y>sqRs&bL4U~h0{pq2N!UC?
zm-(xTg)wdP*z|U@afFE#>QVmjO^8j{Heoe2qWF}&HQ(e@C1qsvn?K(AARq59nXgNR
zty;zZtAk2{T@CxJ)*~nx7z~$>AzbX#8yA{eTW2gik0}BI0!FQys45^uQ3|`2pqI0;
zl8MQ(ljjh1)sVx*I{iN^UvD^$3M(^s8ixsLU>FUo_bT_IV;@Jzg(+EEW3fax(Hy1>nrRc0
zL=Cj&&32CxO6KJ?23$=LICx4K6aOGLE9v*M+Ti%b`7G{f=K_cHKIaW2%
zdr=8Euo-qxGLzu4MB@UoAYs*z&eo`0E__ZHIuC20$U@*($C`w7Dc*_q3E9KX4C*=~
z5J^LyO|v5ek5}`Vm2yeYaHKIH7?K416n{PrWjSKq$#YuL@o%7uq;i)d|QMNA+
zZ#_EG$daDzR_}e1er~&sQ?g#R0&_E7{c+LHuIKo9fP`iG3NIjK4+fS8U4%
z`$ySwA3l0>vsE)rB&50lXc^-mei&Q?CeHgqvr`O{HT;|yazVkU&?h>(KQ&+Dg<#Oq
zi+5&_5Kc_ll?|4#2L@FhlKDOoRYnjo?9+*dY($cX59A_|vP=(t&d(^4CopDR_~c$}
z71q==wUI6o*}c73{2F;tQjT@9^@HEH411ltIC2+z!|T}MD`C|I%CM*p_Od<4$n5?7
zoy61=!y_W3g;gEB%3d?PF~1PIAN~f9s`nAB;x1Ul&CSj5z^E?Kbx#w|zWTT=ML(yH
z4y(C_0`wXan4Tg~84?*=QDb_w8|ZG&w&305^8fhp8fq76e2v~KkOM#YA_@vz(SP9e
z?qZh@;8OT`j2M3V`>fR`c~D8KMI|Hvhmj5Tj
zZq;BCXcfM8MvXa^g7d<8ukTgq@{bHb7Pne&EW-*G78Zkgf5Nt8k>trqEra&B1}Iq2
zBpOeA`)L)>{@Xh{_ZS&@b<5ffZi|SVzwoHaS6l4z2ejK$Sv#HvnpwLxO-TYve`fm~
z^AUNV_}=`yP`WjLw!75TG4e${cRuG-wQRnO0ERPhy6#wUI3JG=I|Y^#Nu(N1Z?JGl
zHer&~wts^;&PcIrJ5oMBRYvK^%Rk
z`ZW?CS!>@#5a#}rO8OJ{k+zy%vz(z_`?%N+c>S36uecsddro5+@EoXxTuL+4?ZLfr#Xl%W(9}g+dy3p02Z`)xV#DjMZe5tz
zSmn3hE{-F2vp4jxQ4sknoF(}WIOCx+a(DL&)WeM@e)#J9zac=BE+nNE6huH3bauMo
z$q8%t|Bk6UoVarRn@f9#soe7T!#n$F(u(>Yi2{tT`XZ9*YuEdroi5$k>_Q2EEJ(fr
zjLgilXc!;=(9+=S6d+V@w_{BpdFp4*dhEN82xe
zXsWY5TKZB+sk_)x)1V{4S!D$R$;W8b-|40(1$hgc@PHVNwHV$q0gSUUcjejRzXY2n
zEdzsJd;G&0n^H1%oqBSa3+*3{{(;K_5&Lgi^>gf>;g>s7*Dcv;;}_Vp7mktr>6t&Q
zYk#RQd{*@X>{xQ{2$uWT2m1S8sHz4Inf22O(sj>9L@I@Ew+LUsy^k=uK_$NZR-@oD
z?xu}ColJC-Fk+)(2cOfh??_)`a1S#4L#jMvYjK
ze+sYm$Dyw7sD?F`shsN#PZR;Wt7N0v(lF)rtrr
z`rSK4?o5Dpx|yBu5CF~B!$m~>@TtIPVTlI7EO$^w_f~|3`no#VViyN9sEEWJ8J<&}
zB3a^^QttvU555HJK;LI0kS?*v${(AJ`v$z64717O!fSbl%Cs1h41lo
zf5WUOs#Z7aZ;_$Ph(GACoDi-}wmR+{WUfk}pt^7|Blhev`w8PMl6)1PBY!m(?wA0{
za79Ix*0RE5K9>UGI<`wj6!rC6LIDAp6MilM2XsKN7>D6gG0yLF;)9LDPn)Ofize|5
zQTeoCAv)8Npoyor2nhKUA|mx??(a&}u%oVdW>r>ERF)(E9?>u(DhiGydT#DQb`zj$
z98Z+V&piwoV#%>e@cb9MeFY5gx?42UFkotD8b9c6O*eqQENMW~GZ8!*JdaiArk{a#
zN|ncOXDY8UBC20&yO_<)?v+Q_X!IZ24+3=E*93Vm(2li)gf@uYhma6Tz}beDCprW+
z^d@WGe2`D|FVMQ%5b0LtAR_io(d&_<9-{*NZkl>L8g6%|sK2B^M7N@;?4h9{UP^qh
z_8qKQdbT!)-gXAHgQ85F(D)6w>J|o^R|g&54)Zhk_DN4<=t$lKYsu=?beg}~EqGX8
zqBV4e^7KKv;DA+lcF-g;4MK;a;*C^^5GBO5&}T{-8g*sd$1MU%`017#@oX%?!SZO-
zT5kw950ps7{;=up`q**MQf+PR$rOR7QO+-9+8jHcPtfj#7Z3Fn6lQAJ;um>73F&4p
zrHmXSZw`s=VKB0?8iFei0ixCgOic9Z8c?&9@gZfTd+@|w@c~6Y%^im0i@w@gS8*rH
zR?3M6zW;<|yCp@9S+mnAA_sm#As46}a&s&vgw1#;`uh6Nz=>VAkxM+0t~*GdidlPR
z0SA0G4Q%oL>x8vwD=ycFKYf-&Ry53X4SarrL+!fC?n^*}U$Z_|?!43f@DN}@JLoIf
zz^g4NnmXoe?L46z)O_2juZ6!qvs|Br=WpuQl4gA(R?(u1;I*0cKb|FEvzinIw~aBN
zTptp8y7rN&1R}SpDcZR?2e#Nt8>7Bbx-NhCk#ZY<@hu{w@Ap@k_|dh~cpmd#Kr@7e
zU8lsCK;LxxGxeY0GMWaVvGV>sOTUPX^auMA$-0XxtM&r9Zoy4gIPai6H^3T@`M#;H
z7J*#JffhAlh233rd*K2|4U2me{ZyX{3b3W=Ujq@fUBWx;i{(thlUSCs+D`+zcS%hHA&#{FXH3MEL$xZITExos4%AB1gN?b9jfGHF(Wg6d>Q@xUs5(ktC9!EXNHrumONghYWz`9VR@(>z7
zh6wGnF*fFw7zI3s*z@vh9{VD7WI`$VvS)PdADdD^HB7$C%p3&f3UHr&`}S>eDLvUD
zXc(Dx8GMgsGVR)j#}hBq)B?1*yIhSV49@gubJtXz%LNHbrKvGRU0nEW=i6#c3cApg0ye_zp!FK-XVRp}wplO_c
z0kf=MV0IDp{K#076|WkZ5~n+M(?>38^HaJ9@0t*7c8kwFU?BjrS{JS9(K2FL?cSJ=
zflSkH5m9RN-NTU^8z8ti@;HXA2&Zf9aivj#xgZ#nelNQm$6S3E{z;T?MoH$Ls!9oAsOr<8=
z@~#+tA*|~8F`OX-<5IvC?dcfa6uQSir1*0Y!{Y_Ep#n?~Ao=N#`MMrS(75-0dKVyB
zXDd?5T6rzK73+)y`|4Wrw_5g?_M3H@X?J(X5I~=F78v&|judB8&)w_xrtA=|T}Y-%
zs~E+nC;G|%mfbq2ag%iG25H~L&HUukObF7U;^2eQomo(!6nu`}v{!OC^pFg`_-5G*
zsDOiOKsU`yPdB%69M
zZt3AuH-(@DwHGEe@iGy;)%hvP^y1@CCM`1TW7Lf;&-39HbpF6Iq8N!Au?!i|bro~_
z*o3f}Z6-utrNnwhQdZXCd`Az~_^88(IrvL3pZ^PugSA1loM3t!btMc`>(F`uY!OJR>^Nfq+m{4CHGLk=HG=
z9g#UTqPjV`wvDXEh1}8>%ejz8Qen6BV!l0op-SHhq-(qN(b&O3WkgtbxD=4+U^)g;
z{u&JpK@G5Q&WCf+rI3P#s@w%U&vuf>dNxXC@S#HJYW*QmPfG4vJ1*38Xws@}VnRtq
zPt-M5NA9Wa_}J&9Qa#DvsmmEtR6~QBir)@x(l%^Ao(^OcWO4cjj59LU8042~f@KI6cZhYvY@XddOdg
z41nY@9N>%y1UC^NCg5lR5Ja!Cb!XPf-RrHuNOguiIp$}gQOV|89J$*$jD?i$u
z&#Z{;3U2GlKc4zdus00mo?a5D?nuq*2mqi^81i8yd+Di`SD{96?7x0VVI0RGf5}_Z
zrIP|t<#|EaX13XRK|S+dz&iWifb}il?twT|ke7ecvsAB>Taq5wd^oodov%HH@$|ji
z)ftLG9)I=$)F)dUp+Y%ihuyfW*Ti|zX`SVG^p1?%pHGkU^z}o-!+m;|^!k1%-l3-l
zHTqvZS_B&U0Wko(Pf}7+3taSu(xjpJeSLg~tNA|6o}QiG!>jwGli!kg@ZZ*8@Fe6*
z4=lUzoCac+6KK@2LasbeXF|~j^#`H`0#OXqnURr^yK5uSv7Ck%*r0Ml6VDp%dx|Nk
zsm@ysq$`E=p;5F<8E-!%Q>Enn!WUIOeoi0!p{}22ur)G3Qo&&yF=??wH`fdY2ZD2N1X3~@H%#*G*`J0xpc0zm6;YTOGLvy-d#=K|kUsCYzf_Q!Tj1E45yMVVAD={D7Ojgw5z{YKWK
zoljieL<|{+c4HVF>pG44R-CGHl&p2b`^cV*;ClVQja^;Y5jBtlc9~~b^k_smjafA&g7-xRY^dS@rgT
z=t7V#8oUE|xD
zO27|Dlvw6xJGN+n)&pS0hBATCYa5fLPV2ZsuxiMI*zg0UmzC`V4NFQyj+_)uy
zPeQ5U4!=yGXo6pHNk`BM|I8~3?&Q%A&Ce!FV1be3j{PEcH-Gr}31$Uk0ZAWF*OOrI%zC$Z{7P&LB;IV{NXzv+dXrTjFM`PYmrv=ug!CfU;Ah@rik70u)w1j0g;qtcZ
zeGk>i;}a<|l4s{<^2?J9blXB#5Z$Yg7~j+}3>Dmq|1o`k(hjHk%^EpBGf9Rb!77M{
zx3VvKL39Op7}V!tvbVF)xv)7^hZYrj`}(X0)E&_!Hs;6wX_0eE)iX}9!To)B(rPAq
zW_VE~Fm_#N#JZh3@p3n&D{wa%U6@|lWR(1N{?I=(N`(ME3w{Q_HR?efXoyQa>6~r{
zR{RdD&Ho$km)6Ll8W%I&eN*YDfT)8w27UCKA+r?Sg`zw34qxoi
z^x*-n8+L`P@td{@cvy$cTBH6k;ugx?q_~xML90&NeQZoGE~VHg3ap%YIxc(wgiLEjAMyvWHxT=#9*POV>u`~pp3Q*-y-qk}Ovz++eSwSu6{
zpa-jZm&XM$n5&CpZ*LEGcAdA6ffgDeQc~zpu2=q&Z+GtBUry-xGG|++=2nT~)wJwv
zkeA051y>9vj-%G*BtDpm+6$)$D*y9@iqXp~q5E
zm!aUzR4bGep0*kN9JwA@i(OtHFlT@3k}3kzXv$Jg|GFe0ulKzA-i4VxrCAZCz9
z&0U6+Zg}aQJBZ4__}pv{J+Y~hZvT9zxXpdqlBOpvBYPdG=X!PeHeGiN?Dbl-3AtgO
z@_ler68~1bG*aA;-A2zaMfbJwH5x)fRtPKDwwXEJtS}Too8@{-p4c3*cN{uM27U*V
zRHABsf6{vh$w~nw((7^?Q@v*8M!n;)vx^%(0U
z>{1jA6A#3Yao4dBr8*M0zlS1R2FDuV(nf~3`h%Ow{Ep^-&m{sMQuJE$mQB^@_Jqdi
zy5&ioK-c%cs;xkOF8wPA8!5axT88YPc2$D%jhS<~g{rCBy^XAN-D-@8W-$OyJ`zud
zw)?c7?l6s_wnkFw$b>71sB~3V>fx9mI0lQ{_mM&W<*H$anD;%pd#Y{pDLSt?ZIBGZ
zj+c4@G?8iKFAyJXZOE}*QKY%L$MQd9MG`yBc)D&CUA-+~N0;XwQ~Zn=QW4yup;ejM
z>PeQDWmQLmM;)|{L#FT8P?l+gGVG2)^ZzhC%Td2p`b0Q>TN6O5eK?RM)V
z9sY!gMm>OFZoG%CfES2OYwkk)qV?vxx|YIVT8FapfnA7clmAri+qc+GrH{e9eCY~8
z_(w%vRIj-oUxdorjQ-1_qMsr)+*8PSxO}}tMOa@k($QU&!l-E?PR=ygiJ}aq=6^F~
zhjEQ~S`OlL>HR89Sgn7Lf+bQTF-y&4*xFCIZQEL-kGR&)|M4J0XK{k{O1?_*c7XrE
zbu2FumK{-c0SC%Cy+;j8_vn9_f5xM_`2@92SmvjRte3clC=i7mlsG|ci0aUN8I@EP
z*{R1t*q3|bwr|t*p3jBROvHqQmI?@?N2~k%MQhD>cgioUrn~OO|1{7kFVgSz?|56>
z_tOo$Z|^UN;#zGZxmAm)ZY?~&99*U#bo3KLsSK6J4uG@v(Q+3n9i8C|gwa=c^bd<8
z)~nb_cFD**NM1L+qt=2s^>
zBRQdaLN+M!{*jQ`MH~}H=m?nkL!OSoN6BJiE1Sbw={0j08S$!52a!d~q%HqbYLQfI
zqr*J+b=x?LB6NHQT@c008Q^W^QW|hEZb3
zdQ)P=a%kn(U`+CNdTy_30yqTrU*q^DW8D_~NG(K1*4-+uH=<=3eUFYV1@W9_c{ySA
z9QL3i&5T|tm0#UErkeMK?w3woC9Y4TQfb$o{k^UFpDGuF9>W?3T{TX*4!|<%gY~>aU{`|{fzg_<7SBjh`34gG-%x6y!0hn8N4`W
zXJ^vM!iiILem$uY1lj$HYOmY3eiWbKx{d0B!oDKRXIlIn9
z^Wg#vt&@W-@W`W;QJ2Cj0+EgyjTcu%Ulv=5fAIhEr2Khy{~g#Axp8g?#{Z%d@<5Sj
z)zCm9PZQE^W1fCD!{zCMtzUdgP$lGsi%zjoa%>A&~_8DRu?*!v5ZO5L;mak1-b
z{J8Z*;Clc4kAM0xK=Rkz+MewmGSeR3*ZBIcR{97OfTOa>=y7_H3o`$5rVsq%On1pm
zQAp8M?J4xDIt=ue-07XgMEffS^Yn3n7k~NcRldX8YpCF%Sp?Qae*bJdA|)jy25=fk
zKWX5NQMD+AW)d7|7l9TwgP$!p2ruX<0D7iU-51?k(gkP0djZnU{(oK>fVLk0XM=yc
z4yD(T9?!sruFHz6vyJh;y*A+cI;Na`Y>=&xIXpYan%o$0-v9m-q;hVVq>MOK%-c$GrZN^>s^Gm2|Yc(U9MKU90^szC5&K&5M6ucwV;5GHg
z7sAFX{RAkK0;CEa*#ur~kp6B0o7V_zS*KIyY)kkDa3VoDUWYc^1r<+tjNnEVNt5QE
z8n(yjyPfV0X#B{T8!5F-X9qek6_l;U(AyFnaoH2+z6-|aU*!c6lE-5^JedsaOrUdf
zgQ{o@xNAQM#WkQS`U)Y4Hq(I83}mG*D9zxOm3#X1y{pH6dkaA?=X`7AR>3I-kW_W1
z*_Gm;enTbDWKrz;eK^V(-HQ_cFa#Kk|GZ(Z;5WW{`9n@8TTNTg7!6VZs1)JR(F;iv
z2_aw%CjdYRohj=B|CmLS9jn?61azy^$mK?=^YX%gt=}Bf4as#ov;mW@a)ndw!>8uC
z;5!w4{P-8&!oSU`wYBwnxGOkcGhSuW-@Df$RLw_&c?Cg@-Y4K*vg_KJ`|UJbb?eqG
zw9EqpA~|Rn;7Mt3{6ImGk4*kL0i7T0e|YydCOXmotQ#UE9?haY+7rLS{s}Sp=-)Ps
zp_=dU)XA@3zjidfhCf>{K3A~4O?MsVuTP)2kGvu6LDuG5!0>Qsjn4t5v%Fcfc83=oAfWM=5Ju=Y8f!?Av
z=ocxh6;E8qG3uftVNq+Y4lM>{rlTH
z>cl_*h4)j``w$5G_n9=9@#GtIr5JGjij%bcWQ{@vOnzVY;
zL51#tFT50FHU+4|z=H#)zpwTv0A=5L=EFk{r%f$*bHh&;)#wJDS~KB2h99twN=kT9
z6;M}lXw@Jr%GZiqRFnO)<%z(e5k??#@)67&<(lU($%Y4;Q*H-SetT6}z@MYl6dQuND5+=9bb)C-*$VMQTWjG*(X?d8
zO%qd7*lRenAoYM)=6ezQ6Xu|)XcC(94Zw~CzhDiP6Hfi#5&7e&epO3rFm>3}
zU44l5AeN;l*U^}&CGw!x5Q-nxC;2W=XOX*WQ#
zvoO}>(*@!!!0~i8zTh0j^EzHHC-8!CU%zofOi8Iz}|{h(LQ0si|c_aKs=J
zaA1bruvHQS3v_#b#8T7bkj^I5juZwBMfp`V&(pp20f3|1+cprxSzO4^`3^C`?3}R0
z&&Kxg$B!RJ_mq$<@#ve>27v-%SmUV+yz~Gl?4F;h4>xYyrTg0sG`hUJ+}zOtHX3eN
z!04@xw&FwY*6La}GA-#lftP;?m7vo{*iqL>Nt+-ALTZ_Wt!?=0?PWCcx*Y3!frPLa
zy=Gv>w|+BaC)l2OVPJw60WEjWMpvKc&yb*mpv}H>u+)|T69<1b`TQh(VELMH2f8tR
zl!JqVWw8Gob|8-q_W~-!^Q3O%u0NOSGA1pZ1w48MLlHun&-DLnA*TPwl=W}BYw;YT
Zx?F5p?8?VVc$W}DN=z1+^;paM{{R&Lek}k1

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_199_1.png b/doc/LectureNotes/_build/html/_images/chapter1_199_1.png
new file mode 100644
index 0000000000000000000000000000000000000000..f19e88b164f90fb08abd355a23c2b47a742e10e3
GIT binary patch
literal 25198
zcmd43by!vFw>CT#MO4ZlB@9p+B&4KKq(K_#W+5Q0w4#8dAf3_;($Xpd(#;YO5b0D}
z`Ww^zJLmo0^S$vq|9#iK_OeGakF*kSB
zt>?$bJKs55r#)kJ&PF5BA0~>J0!WoF6u&Ddud%4f7(29CVmoB3slZgOR9In4+D8iZ
zmb4qJtn4DSU-dDiJ}r1B{y_q7NA_a4+(moEn-2U9BH1~YPhM-C?vT^#Zt?Lm=}L&L
zY)I0ftaab3L>2VRlMfD(LoarF5=yfoT4dq01&s+0HP(|po3b|E1#
zCnr|37V=F#!r3YEmuEtJcKL88pHCUoCsE<#4Wr9H1Nfx_S**J=yvxIv~m95UV`5!P*>Vb-51&#P+8w~uRcE7-)ds_
zTzj#Sk&)s0XVP!5uP=3>KWoWtVn2!U%9X7dcCV@>;m*!Z0hiST=amtXwrse^$9v+=
zW>wYI+EzpPy&K&kMFdyw%r}wg5A>!pUpIax`4DvCmbb1VxBXFWWF9{g~=6-RyIC*Mn)|GJlWT3UC4
z0;5UFG+BhkkB;_#S$*-pz{JF~=nz$19hI$*>042VXz73K93&R{|2c}*+wZkm$C7ji
zhB<=r^H~OmhMYZgRaDYCIy!7B-R>rxuV~nvjjvv*7!@ZWf|Y+reG@rTcqe{Xah8*tv$q
zL!(gvi8r_iho*AWiI?<6-9;-IgBo=cTm8Om_A@mnke(a(jc_fbT1*SI)c@nlrvW=8
zH9&+|HEU;a&OD0FZ;UPZcyY%`V7bI$@dT23(Ud||b#3qDNN?b7EWPE!lt*raZiTK5
z6vibcF6;K^kY0a>Yj6`$UIqw7|Cg7TB|K#Bt5JD&d9;HdmlPNBk{6t4GYm^MKX}k_
z(SaLB^uv_97b`WR4#>u6V*1~S26PLw$7{4aUPtZ)`VD%r
zQ7pZSgLx7p5ZjS?^{;K~Lj{j`+_oQQq^EzKvA3N3{A#4kQU$vm5OB`P(a~akx-~pK
zLpB=IpD7PcgH4g}bLJX|{u6W_ET_|9DG(t5=Gak(XB^^zQnDg8Xb
zEv_4>tla0P+SXec9X+ILow09AoLwBOpA1Ppyowy9Yn+HRg6-LmkKXCZnpK
z&J7bqK4*v)uT1>$MYr&958E*0LnNF%m*iH6)mmKCzKpMT;NJV?{}oSqyhdt0c`6+@
zF%#pfa~oT${BAYLbmisGX3m~+Ew|;HGXLaG<@5Wa9f!VR$n!zW!uB-gtFNJnscB9{
zMa9xmTpXpTnVECV*wh*W7nh1n<)gK=wKx&4db)+}QQL+eLENx097^KqQFC31dDf#9
zIra5=D=RA_wa-dtJL9qCLhS5{3Rx=j^z`BBTT580@!DsU0*=`wB_*4yHNRxS$36G0
z!B*J*BXIxsV^DZ{^3i72-fp*7A*@KZ;b4Y+tLS*IC?(*m!JD;RAB@FzZd$)r>!bmK
zxNCoHW|8gt*%Vj~!T!9B+ulM}&5@ptPJXnW>q6?`xpU|CxbIUSM>@u`S#4bj{lx({
zcMrs*!v~F&wZm1oT1=R3(05J0M7*C|?t?I9cAnC=B6=u$+0iGT)Kb$qb^ZI=x-;(J
z?AF3l(^9>|-(>-Q8QB(3bBFCtAcnpr_*XHfPm0ww?BpwGRAxW52)G#}r;)JnIlAub
z>4g!ebag)k9*fTNHBM6lcbl4j_u(OY?CI^gINn+U9w+Lm(sZ)&9&y)wD~L5NWn*d!
zzdZHca_Ae&*xeiY1f6uE#lD;s%=#FR_s@#Q<3
zx_FmPBG*r&agzU8g>(&}8#fEH*sI-Ejk@DMFSvuhN)L>uISF8w3B<$^QMwcs9DmBh
zdheveF|n!y=T3@b*1#tF!BE3uP4os-gn7bj1!jD>2)=!UDWaT~-jU`zOV}A7kw@-V
zfDOHL#kfDBmm;^&7Q3(*zj7MCv&)W8Q9@e|dC?oHPX-T2n
z#pwt+f%uTX3cqmSONKa8(h-?PMf1sW*Jr7EP~2s6o=ZE&cpx16Q6Ebp8mz+&ipOa!{&4~9W
z^ze;N`}l6t7DWrvt4n-eyXKZ;*QheMy0xULrt&oj2O&;^1ap0;HTPz@^BMhkl8diNj{#r5mm4kW2y94Cb&VXK^?MD*A~8Lv2Egbe|mxL
zWJD51DX1}5`B^K@Uaj=n?Y0LYgT(>&(GQ5BA4n~8`r}kxt(CSOKJ%XFK=#|$S~{LG
z$3>k|mygL;5RTvIW4HAN+^6tWjY&3Y+aT&z2p_yoooB!RHiC;j%ngeKeP;o|v5fRT
z%!Tnz-o3_>OKOvr
zVIsubJm-qU6EnU(eQMQ<@z^$;PcsDW#4?D`0(jpZBizb+a5yhP8;
zdqYn>@7{hWHWt0h^~tR+9K>tZ(Ww-ccf;XrM~M9ezp>l|$LMY9ytjrNmvNBP3+QzO
z)rZqx^7(P9Oug0$nWqGY@P)ZDtEV>MT?0x{C$3VW4^!~gJbp&uwOT53
z^D;v;kC|p}7mpj5Jh~x0-RK1#s_DkV-+1l{x_2{qrCMm_(&dy?3{KM_-beQ#ibct^
zVd$jcG~^ZRO{-R)HZ^}!1<>c6Edg&jLx}5Yq%qB^nhq$Br=APQCk?dk(-n9$+I-m_
zAvQ&C(6z8gI+$RE{hjbz1T5pOOKokv_hMX(I3bdH7CpAz4`tf0ffA+HvsmWRTV##C
zb0U&g+Bm)68f_+Pk?ju=R^;4M)Tg=#kkry_4zx3ZEEfE
zq0w0>x`{{L!^#M9_!4$7-`zc1+c{qC(*4YM>Y3abvoGx&U5ajJ5Pw4S`!_7enHV=L
z#Kq!%CQOQ)Z%%f|YHqbD{d+`s=+gLgv!Qu1DBy-s?4LBTmrHe6M$dl()7
zbt}V36SdOCIDzZVc+IySUP=tG(`1Be)1%k^93nC2XCTKI^!*xnxJj{QMC9T_Cy$!{
zxb#OyUqcHit>9*SMK$+)uy&8A#My_kHHxg(L62~~A4qB88dUcWvaWMW9uWsAnq+a=
zP#~$0LcAAb2(e!=#TISzUuj;fy^eTE&QERNJFi`#--rzttl_)cR{jHhic^zsRbnP8|h$;q8x2Yl{jAhtlJg}s
zf`*m%UNspht7|9ZE{>Xz`wrV7hzD^=iH@#r0gu(t{O(*b)|cSQ=FdR(#rfVevhU`D
zxoWP{VN#oWOGWJ*_yvzD6~fDGCxnIm{O}ohZ|HO9@wa!8%o@H-H<7oA1y*XN)_Iwv
z)tw=IMJ4?@n3dmouXf=HQvdD3(-OPwvA>PhVbxP1f}2Fgzlku9AL9eiygB;p+YRYn
zji{PtRxV~-M6%>f=YhV8%8kl>MO)AR0y}Gtjs}Z7)h^vq81Oo1@tWVCIIcfBoH(w-
zYW4MBa;>wav#Yt(8G_hR6%N&3+M-NMPE8FXi!%Ke26gQKRQTkZTlueyZCQU8rY=hj
z-M1Pl>>Dv`(JHGVY$g}2|7da0wp_jYgsKA@Z45i;o@3`q8kiw9n*8+V>%!6c^~j=m
zuH;?0Yb+K@
z^R|wTW#^ifDP=RW%((dYt>w~g=P?^{-{tMEx_hLeJE=}!(h%CnEAf#FY$m49R=-z`
zR9MF^E-ej?j~8~v^P*5xcXVTYecDk#?cu}g3kUe*8?&8n+h_W^y^b7KhD)WwH}mul
zWzLhb;GyR-^6Xn0L{a;`sm_p#7h`km!s)vCn^|#0)6WevjzV=_pi}j*J(@kbd#+#C
zWh~?MrCXS$020?9&(Clhem;ek&qN{z@3kL2;&EQqWYw?Nxf3scyqkQS^UHImW^sPL
z=l5je);HtOSE!MdmF=rKSBq74(O{e--bl+oX)`R%%}mj~VGj`<2bq8MCd5{=_GvLG
zyPo?eYekVmN9;skVBl7tEW2*m#O6Z(RvUBCOHNAU?Fzqk-aA9d6=M8_1M&|+5t^j8
zPI%)rpYufIiZ|X%SxPGqbcd|weJOo7HPF+?Sj4`^_7hSDgy&jfK$jG%JAhCPP<
z7Nu!sQZi{*|HlsyKqMi6;Mh#83G5zx%0@jIuB^rKkbrFWgs5aH3l8}c!O*7>#B)l9
zo+ZP|;$1eSOhSxN0GclVv1K5-iz5aL%W6&IG$|zviGJ5Hz&CiunSmHT4pPIDNZb}d
z67TnR^8@_lZihZb{~YFJe9By_<*~LG2(kew20uH|Psm>2F@(3lWd@wtw@uRI0sp`|
zgO8^OBx*tiFc4d*oa2Y>aOEBc>9(%jhDG~}qAQ<$dhadeBSD^Wi{JR4{=5{VS4i?D
zr0_L#zKncS157~9LGNc}QmV~ee68ddqF>;7;LlQb
z+q&iE<1QGz4+t)1?wv6}6CE)mx;L1|Ym1uCrrx0mGH6=!8Rdvp!YSnXDab(&Rsr-x1rN-mP9U2I`d(zjAOjbV0$><3sk7%Pl35f!E8!#lc(B3O`ukSpyNLK-M&B>pe^78zOrk%K6Wr-ezItHaTVt(l%75
z28tVX7i;h0_-2yaM)T^ion5##>Z{)X(&D`i=L`i*&5d~gmiak^dFYx(zAg=VFz+F6
z_51>WuuWamKOkP2kN@cdor`@w23^@_8hGZ0QLkEs#QLie9Bhq=7qgf3&)myWzxj5eK!GjONS82Mo3{b-^j}!P4h7=uK&w_^m0q
zSUQ&(47(Gl^7_>}SrkqphCy)si9g^~0zZ4xl_~7MJ$qIa9u&dnwC9uiyA1WJ9)j^s
zWHY|eoN{1f5?l6Jt~5>Zr6TDqe)V1p`-_OE3+&sTymYM8ZNj`}GLBPew=Vvql5ymB
zEPG)Gu%Hlq`(6)683c_>=lf?h8{GiAvRdWfg2XuWFT_E4
zx9EB_ERwO
zl(B_;S!2t57fdaNc1zrA!vhbHt5g6fH?@WraCypixgF%9hLx|w;T0}(lbJ*El|$bh
zZpq-WZ~jL9qF45Gm9&7~Fhe6p2b|6yF1WYHPoh(557pBc)3`1-vXIwtUcV(RZ!J8?
z;}iqS`)!UcU>e1i~DaQ
zBwGvr(>&Q~1Rwnuel#_roWwf~pfDJCdUKQ|!od|u;6(3F(=D7r_cR~EtWJ3aVKV9u
zK|kfRSV>E#B!e&Tzs}t(%N-l7S=a`s8(M<@BX{Ml4zt%s1JWQ*^*8Y>5`8(je~tRc
zAAQq9N4o$53ly3FWL0;k_gn9IBE-V5PlYlnkFp4p?q+Ejmzo&j-EWX_Y+~S|SGe`z
zD-rVf{)6`Q=HM$%qt=y{pI)5Su5n(uH!3SDTmE>MlU3(~5|Pm6n}ehCEeq6lpDInY
zgc6xbA{v&!2%IL6;0oY^$BBf^v@@eXK0o-{X7A)wK9sMOHka(BcjeCGo@=xR
zy(h_PdFl&cQZ&6wMGb6?9A!3R)rHnTW9=R7G_WUkljZs&u7*oTOaykfEG$SRIK)$~
z-jbe+s{_gdjX0v$SlS;|+6*oamqvz%Uyp9jcy*o(WfOHrnVFaddV5n^-`+tMNf?7o5a?~|4siPiq3ktZyF*M4C2ouvM@0li@g30{i#3
z<5AizJZk%2h{i@o6|=zEDPnGX>)k&Q%^?p|J}$j&ahSRG@ISDgyRiPR9qReHW!a&M
zJ4JV&e>`1@?&GntqU7d;HGinv(Y=*x`R7e{h<=^&F+Ig4*qw%ScQPd~!D4cId65a1MCNVyt
z(cc+&E28l_u;Zs9EH5v&JUZCbF2|^4JDempcy6FXOOrnYNms_-A%@yqiRr6!Ux)+`
z2D762&^-uhPXi@bd?B8Ji-Q9rt*<<8E@~%C
zKYn{Nw2gzG0Kr!fqV&ChcM7Q9i9gM}bkbNoH|yi9lLpm0+X-sn?J3e}eVrkk76k1w
z5ramCfaYT6OU$Ftl0s{6aFB_SQQ5=81H0ncpx+WoOZ1EnnYDPlG5|?Jn?p$7`AUCr
zd=rJSN|Im_q<4eV_NbYf)oNL(5GrLYtyt)oesiknH0zIGR1v2k^yaTv70DCr>2Z2%
z-x0z&PcnVe5)%0f-%qI=h`|f^?%7K>Ek3_ITRj_R6%pAq`1_-Q0lDw^goApTFCAB<
zJvjqxN=*$jN<9x-vXxFDAKy6)mmIiw?YOuethH=K7P-%akMi1lCl6ObIvmn#l>s^I
ztuNm;lX`ws#@D*mb{{%vF+WO7pFvd>o-Rf6%tZ7vf=3-Vn5tfM9I)f&rt!)Mn88PH
zGS!Rp75sFaeOcpg@(Ss7k;jK<+_7AJ>f3~T6s>uzMI+)_OJ_OT5i8&@8??T$QT}Yh
zj9S=Tb7#zD!u!;z9uX7(Qf=#O3j>P{{1bSAv|H$qOt@K4Yxvl3n%2z=CxJ1#QF1Bv
z0uCbAm$q?lVzbF*J@>07Zenl*}`wF7ZXbnC_ql}}gY7{}^n?cOr8Cv&u!=>i3;UlA?5|sEKKMG*Kr&Ko+
z2GfLJKd%z;D)L3`2H7rk-eYe5vk1Ph5L3H^AK_pC3A3u6USTYk`Qq`>#Bp9~YO0bs
z4u^M01aBLr*s-(T`*+xMfU=oPLWmj%FgUo{p@k#QHhZPEny=VeKo`xPk19^sqk6S=
zql;JyhePS)^LOZ8133yuVb=1A9-g-E)7v={a$NJH69&lEd`AgX*ifGi4h*E@oFTM;
zh{)nXAQ{%r!@9V6nkV4={^y~uwI)0j&bfs-P}uIJeEh(y)`&_Sc4Z*HxRw;>qv%2}M^HinRRAY(Mo5BX#J#=;
zz6LLO;sc_cr>)~a)Q(Wu2IQHMFh%X_y;-PI2Q5smJ|F$5lgS2q3PsA>op=@$slgmUpQBA%U7t^)4F)S(y_d4B)8xV?K2
zXSlIeD-nXL1Pdba{Eb6@^Crj}6rfqh%F4Re$m%Noqn$SW)329tgH;41L|(D?+Na!LuMQfbA+H}4Zj
zD;ry7-3>-d6`T|)(_u{AZ=2_$KT<~#$q<#nIgMpzsyHL(%Q_jo$w80UY&m%G5d4cm
zm|8Bgjvn;vp;0@3xHEypHXJ#1#_=$t_xU=>MF90lL}GJ*N7zj)brD~du2ee(;9B$+
zyR=7>wA1!g5iv2@#|PcVs&;n8M{Bglx6!YPN|b6qV*}{I&jB76P(IQ`FFZsQ(wxCn
z9lQ5Guh`b6m?u5EZ)$295FD(gtQ?l9kjRwP&riIIY?eX^Reg@^djEW_N!|L^2QNM%S=CrQrgMlp!|wzysquSp8>
zvKig1kWgjd+lG~U&OlJ~HKTtLMvLHkw?~=AcMh%Gx%uVHb=ha~ftV|TAKtUe1Ofj5
zJ(g*rQ)qpw9NZfIT)gHPLhxzlNjTmq+4xTfJBv%_JaTBy$p`c5@z%L!(RAQuGhLQ~Q*Y2BUb)&^>
zDyb=jzflWw{AD~!oo{3rP&EbkW-fxe%+{arCI;F?xs3y#TDbq+X1ZpEgQ5NaIG?hF
zY!-%c%RFtTp)D{3MglPU!3HxEDpCEG<@wT`J4=^yv1&yEr%B02@B)u%R
z#xkH;(}q?>;?Pu+yk=`IdMhtZ?M*zLUn1>>g;}wtYDDtqAaZ#8+Q5^_^O~kYm1CKw&y!lLjFgjEvqK0dF>NnCU
z@N{ZHXN!ft3>1+b?G+ts2L}hI1f7+P1Qz(RaU}-2HjL(_CQs^CY}2~%fxa#*WV`3
zkGuUE%qyyA@@u@7+?}^<@v~x{OeOE`vz<8Zo&Dg4Qjhe15t2882CT8aT==x3i
z;T30(jrfUWm8i|lm%=NwjNfg9or%t*+$Fs45r6R#n@4YwA-RBhZ+1e^(;0L9yXSoy
z4SzXLwA)oB*k@H0x{55UKPyVFd`3_3$6=e^N`L!sv$*x^10_tVN=sf``7+n!n)XMl
z@c*FM|3}5wX;i%V)D7Yt!1!D;y{BSm0dt%BKK5s{$wm7f?E4A-Q2OJF)$ZfW3)XtlI7bN{ts~JM
z#|Pgo&tzHb7=~gnacob8cyNlO|3k6pYLB0MZK$9CZtLn}2L?cZQaZXq
z6DkEgkagEO*Nslz8k+{_o~Wv;ljg>V4)ED$iYfTXH5N;kBsj>P>_5sG%bB)F
z?Rxhrcj4sK>LoV$1U_YE_52GLl(ulSCk2@;`?r+$V+;@1nz~vN|I8B9jENPIIkyLIXhgnu)1aII#cuO@U2KJtYWNlAw{p5#ThSZu+K{80s@OuW;Q}L-{0P
z{5G>06oo`%+FCeWGZHrSI+;P`Z-+qZ-3xXwa$`Glw!-a@P?lH&45&IfQ^
zTlcxn0m)>RxZW$u!e`|*nP{9nzpsB%nd9aR)qjV)D}oSn%4+F1REMd>jr}I2zGk`a
z>sjJ%=k$2Y2&U-MR~d10bBUp~
z9@i1%&m{=QDK9>4kZsST*F?dn}sn|F<
z*|qEx;D80ejJ6SU!8D&=d3-1X!A|%WAiL*pz~T&F
zh)KXvMa+e0{B8Z?si69gc7JUBloJZ_{h^b6U9KwY-pV*awmlhe^K$wUNJaz%VxbG|
z8VKP$rlg<1zG);f`hqQFf~vKDL?!}*27vPYHR0wj2JVzCJ=TrCPb5A2Sy-oc{u(2+
zK1SNnCGfx!#41*c!kb`e^AynSU%Sy)w(l<`miY$9TMu&4h0nBji;XyHls^^TFh`x=
z@8ME8#|7P<92K?BzY7Gx&~>!uz*D)a8q=+1y7~P}7)6OT3pBlq(Iq^9C5YG7)I5ZC
z(aaM=+Cp()xj_Lm;*matW|c7X$g^{8RUp>I30{*@KM&8jC`YlP4*r+hU5=*A>x8h)
z$gkDHH=UT?1iN7Xv8UoV1th}DmN?9#Py7J+!*hY3ze5>r6yQ95&@rnx3A%Wd&9v^c$`dn{y
za~7b@g)XCFi`f`CYCh#6a1f~HuNpzJYZGjIO{{SVpaCYvJX9Yc
z3H<=#%nI6E<#=p2yvqv;i-^eCzLpHYX!DHdo8!Z3Da+i^kO$!|%AesbLvR-XCj|ml
zJt&YxC82L)I4>t37@~Fu9=56kHh!-+=(ZLo-3s3q=;F_xh0FMk59?m5>CCOV^x-|x
zAv608J&EsU(H;~Hpd7=@wEiON{XFh8K9#d}P?B6d*GUFSi2dD_%=bpLxXtC495($t%*)gA?_)XSul2D0i+&&3
z*|U^rdRP6m83GPI$RjFU)G!uiFM#omqdw#F1#0v*8Ha_#I&&1H8G>uRd_EN^54;lW
zJ#IvkQ7>re93hzGE{&pnSVMquFQ2>^a&yP5jm3i50zD_7|9cTaufbT)hB3I$$73qHT&L*bgl?O-5
zi-lbf7A%3+q8|G91E6FWAJfeN`kgYW6a5SXmXyY_Uh2lh2YqU+v
zG^Wj{c%G|&T}^p3w@+p6?{Z(2;Kx|VN>}|fUWo(3iDNjwI2KB%&Djw9zo!7P;wC?K
zd9bUoyxQ3)B!N&}(WeSJs5UVPSwd1)Y7QZm)5woY@KAK9BO8*(bq9FhT>`^-JBaDM
zGRj|rk0gWH-h>Lr)x{+P6*!vZwlDK}(=kGep=zphYc)xRr|c=3!wjFpY7FKKL!lAt
zG^N_bbfcCfj!W*-yA*laM6fmF*cxgf(hdg${5~Tt+XY
z^}T~J8%3EX7@4p~k7U}M4yW%nweVM0@KZYv^LL&!Kw&&ic0ps7^_g65tH)6nh1Ep8
zpHFY?me-Z-mF!T2UnqiRtI&>XY$i!njZO0Ho3hTgY^f1kqAnr?XWqv{4JZw#{()<7
zf^5w6y*VJILUZ^(QRE}Y8V$UmQEj`D+w=819YMPSw;@$9B3>LK(KhGat5Li&{|pCyEDcYT-xT34{@#m9qkX`3ou{vc>96F5*cC|b-h+Ngcwu$e@216KUHa@0>U
z@L$9-B$qCw(aT0X2!HYEo#$SE^5Q?bmH!lIa}nO5(pluopKyH$qEQ!ihKH{9Iz_aj
z*|sW1;*X2Ao%ojjs!o95(MDRYpI
z(b86ryMN@D7wvO>wRY2
z28L!MZxMgkJ9J(9g=Ubt?Ow~evYGc$>@3wk5BE2C7GyTW^}7s`$l`Wh9rZ_J^`xoJ
zyxp2M|AZ-bbhGC%G35u$Dd@aBzgL}^c^`I->~T>+!OGsq)S9EUt+n;w#6jl-13mpc
zMWL+-D#F>cd(*C7wi&gFCqQ~UvCJS^RHZVd%FTW6*J^H0`6YYRH}TvC-D7W`rPMUm
z_8@yqCJna#9`^n6zYs!>ZKl9?RgSI%wZ%eP#_9JDF|?m1=(0MH=rW!I=3xQsG8o&-0SJM)R*OIKIa9^K&T3{u`+&8r
z+n!`ZxodB{1#^`^GWH?RlCytUS;$IW-E#wi31s9M=PN>?mjldEO3-Qin9fnfc93s?
z&;|O?n}O14oY13()__Jo2)#CDl4Ev!nKGPWf`
zo16-CB0T2q6_~DatI_97sAPu@ndgl#eMjr@K1&=_Jgf=)1Ir8_y_vqs_k9UND~a?m
zDGZu$4t={Dy*+_F(sNz%2cc-3_VgV=LXICdgJm>outgp%OOfH1^DdyGlat=UQ*e7l
zwDJAA?dF((RZKo9>9tQkj6uZ?sleCd=KSGUq<~&4J_T}})|6uYR>P>X%<2GNCk={k
zK5|2cVB_WtN>b4-3*}uhXY4?(UxCk^@$2TEw|JH&>~_d7Y<-|<_@s0+U)ym751eSz
zug=`5BzJiE0n%etn}<_3O_XbO^o@WvK?(5o+w)^&oviITrQ|_I%Xh}Ot^GkXw0V>y
z)5U?~uc`Cn%qC1vVYmA*plHb6oaRRr{5UO3Y;ljayKiGzaL+?!&@D%s%1S@u>IG!8
z*%zxxe6xTP#9AOj*z7%;+5*1mAMWSsvirlBjqhu&cJ0cCo}bWqHr$aCU4;d%xN6v=58U$*Y{dwK#gt>?;nmo0VI1{8oj`
zVAzQ%x7aOcl2Hs8oLw=WhS3DjmnmMH0=}-Vg_QEx=nTIDtzgw`{__!_%m!|5%x|@2q`D|Iv=*P
zyUd%AZ~+;r4~9WT`KWf!U_uMz2meU3MZZw!zbUjmp5Y!YP(Pg45G>z#-vWF7L>m3H
z=kX*XjA_VI#d;3zYv)HXcQt(hrXXg?2*Y=
z&!ax7<=QUx2-Yz1Sm{)R|1>Ui^#V95RChPQTRge&=9lIann`TyT>NgB$ZT+Ivcw^Ur$U!OSnq6Hl4+A2W
zM-g{3ZVG_wNEU2PYkDjNsyKDNe0;d=^VZe1@dVn!YB|aA0G2~D^83@-;T6)cJLWON
z%HG*zQ|}krD}#IWzw2cCWOXgLU?;~8<>0ut(h^rsk4`~c1Uk8>cvw~~sLa?#;R7~CUPcNH|R)Y@>`
zp)X<{N_kbfzSaAQaz#+D2n2S8@iA%SeypdYPyA3M_dTKWX>UjI>rC0wEX6J$I9#+3
z#6MMnx<8ZEBZ?M~F+v^K&LHTaP!l?|I5aFV6EKs{X>cd7DK#~PvHF>rCzfxve-Ms<
zSO6^RidbV$b!8|M0O8a9{CW5<-{z}ElXg;PLOB4nU=BnnQw~}#Z*(SM><`G&Z*|?V
z8s0h`6nN^hl6$_ydGlE-RlSmJn!Nr|LFSlHP@{nxME}@0vl;IaVg*e%v(R=3Xx=v8
z9z&UxoDplxiGqKP%EX5-&@^0(NL~O-IBOdfkyn@_k(?|Nu#n@;VGh*sT^2Ro?ynl4
z5W))I%_wy$fSpH$;N$}bVmX6L4L=l}Bx*b>x6+$KAN2Pmqz6`bt+(V?VNU31tH;S!
zcGBDg37zdK19q~X0D>p+=@@egpSZu#Xz{$wK0FKcxHmMz*haQ3%{tZ2zjHgw=4TTt
zr26X@e;yH$I`fbvR-c6-VzBL&B;xAF)&4PyyWBY!2CMGR-D0NyWC>Wcj@Y=;@@U|0
zFhx|Zi?iDNTJ&~q9U1ePlipymZ0|vqb|>%N+xw=F5JhPrbaaxx&2h7#EbzItv{lp{
znO#vlwycR<2mAYzw&-9jpgWcdL{oqRr}$ZaT&!G2FRuu%^j0T~ZB<<)fZYFqfOc;>WMB{_z8Px{
zJxy%A{{5`$JbPAqB30J$!4{3j`(OFPVTdH2_6b}e2o6v?m}WME+@TdD7~*tb7H1mx
z)L!i}y3X!ojC|OllQqHlaJ`Kaw+<`=WgCEm(9-li<&P=UtWo#037-vW?_>T&jb+zo
zSGh{gElBFk(X1v%KF}t_D&Sr%_DVt}BI>dgY#q)=o6<$*%1||oQ9e9hg!YL0-XitX
z^!ww5gD{b&o;2IzPtbW-rgQ$35vKUszJ~KyTWYnS#?02VSSCVpIq=TaYQuDhYnz{=
zZnz#~2<`3lLGz0LxZ-Kf=|ERZa}3LHtTF_lgVjJWC|Uc6u#O!+sXu;G6UJQCV;LudU-C{Z+Zy}>Sg7{
zZ`oS8z^D)w!t-Ysity?46-iK1vC0y5nPOmD#CezV%BG;!EgTasusL5OGi_;5E`rP{
zN8O%ReX5O?wV-g3g7FcEVOQmXD3UVOcI1>H*B`svg;HwsHR5}h1c?C|fWjsq2w7`9
z(*ZrbjD<=~=nq*xUEt=Pl)37}0E#khrm+|+?HcGu5+>U8^ess-^^zFF<{B~p=i=qE
zLpS`G)Oy_WzEkjJGXswEmRjS`ov^7{xx=dYUepR(SZCp8X_`1#hq8rW7oJrn@k8cEHp&;-;
zN#i81qi5(4JbV40)DRjW*H0em1=<;)jgTyAII4#6m};jbI%ycU-`lQiC^+6ZKGM(r
zX-Mp9*nR<8&W;nTO(piqp!|8@*km!8p8~?*h2Rx)Rgc0>*l#(|`~RC|o9<}kUnZ5B^ns9I|Hr?YInitvr$W<>ff5
z!^)Fy>{Y)aX^LtN_tq0qzVr62%V-GW9v&uyZ3|Im&mC;+$A)454%}8R9tt0hyon*s
zGu@D~Y;>z5t2NdzzP-&a@f<3q-sqikucwF!_w4bZdqdFpVZK7K+uSW<
zzKgWvmV0$tqe19#-*
z{Of*B>0%n{@0w@t1eeu@|Dk#H$w6ZDEX{UySeva7NY
zG1{4>TIbPq)Vwc#ywL9UiY~;b)pfnppj^9JsEQJPu0TsmD;KTV{#C(^+3Y;MOZNE8
zUL)u*QKJMXnrmSJgQ$q)`)}kmHc7EXz^~TTm5N7}mXG9uqPC;xLww
zv6UpPyVG0&q_t#MjIl2$BqA=;rd2wtXMdJ3#7)(G(9Kfx&rk#wt*+4dTy#_gR43?c
zjdUMOX`@n?#i2*GYVcD9QSACOxqRd=$G_(zRg>dqkO
zP5e5z*z!}5dtKF35Xy4ys;o!MnPpO6rG!x=9Gs@54J)^W-C@-izA0D{0oK5KOf*rs
z)P9)iG24eJHKW$Z6Mo=1SYwikc%2h^At!TtVG7s3`xtN28*ms_SxF}P-0E)a?@9TG
z&)gA1Ql<`!4Hcw=&=0GvTHXc+@Vu{eU)9oL4YL!bJ!5{;LXvPp`y^hSBH65qvpVK@
zO#ra>yi|m#WUxUG;OX4y_gXw|I_h6%S8%;W<_@OFX9vEJ3vJZ+jCho7w+Ws=<@b`o
zpea+seoN6Z<|3Ih`E%O)Iq0=FV<~pqI{mSi*ng#Q;3&T+3CdQ|8qEb>j|V1S8y&)-
z*fV&8lCq;2mm-s6pVV{Erf}fEFE#}E&k>#4`o3fuVE2vNiX~6Nei}tO5?eO!%UA{ahGcGUkz!5Ov93~2eveq6F{ARYcrsod
zbEkxM+l1j?PmkU0+i`rHQvv$|DDf72`Xx!+>FXoC>-XV8jA$B9(dzI7GzSwHZiM=W
zv19cepNk%id0a&WET&asOIF3hGc#PVT0%PD2|M*{o?9w
z8d}Mcy|#m0Sdf6UnCIf6u?SO=uxK6nHqPtehU9xQOMd7OWYO5|kYRS@Wr5;32Z2`<
zX!Aody*(CFVEf$>
zyX^bEy?#}T$ji$b_pTX5B;{i>i4?7%LXc-STf2aSK_#|WEzOTc`5&zgFsaTBg*3oI
z9hhgiFbTDn$mmlgElW0!$;@7qRS_=LW&vHK3K%&I>)Zm!!uZOjiK~W)Se@MW`FqqR
z2KEjY=Qo#ZMFdfu9c<=1eN7g0BoQ?|u<$`euTn*MYanXHi70%pd0ov`@bXUKRu0nF
zzpy=2=cEb>rmb?pX2$MB1IQ8cE-*Sl0nwuEBA{WE7l%_lwCRH8%
zmXbN70Oid<->S+Fm8H~2Ea+Tw@onb}dV$EKi_9+e2`_xFDXdA}Y2*An;&&!*(@eUG
zRIVGFU#x{b(;FB$$X+n!+}fC+8E6pJNNv#LRJbH%7DH{N1Y{DE;MhA|sk|jB~n-w8j0Sw
zZA2Z!FDT1m6BC}b%&qY?c>-E=?qb!=!4FhmOs2teF8gSzNeK0j9X1BlaORuS6i>Do
z+VQR&y@Oa0spq-#0U855=HeCFbw@NSS4^5|5M_*sO-cQJM(o}MZ|FSK#g@lwu47(S
z#V|$_M=gK!ZhC8L*=esofv{#7=i82=2Jcd$F3E`VJ8g)5jir~iGDZVSK1Dc=D`{OI~_H>B)!Nprg$j
z)F0PL=OfKJdxh%k%U3YZe7NI#h$a#*3=Qq=lOO5j*t=9=$hvN`OWAsFzN{iz9je|N
zJPwaZxy8Czt4iGsUSo=VMY)FeUA#HrWThqYAqcE}wZSO}il?Dt8~)Ve@uy0t^^Wld
z!72vCb}x9sbgy&7Ai(utCd?y<@|f8e7F`iW4VVbJuqITj&00MbcY%o}E*HKldc3Fe
zdR12`8^A+H$esCK83QJWa<}w{v)1F(J)Bzkz^0(qK`x{K{;GF4#`k#oPNw16&Iiwf
z&o&?==}R4M@EbsG}GO=^Xnm;p};SghM@y$Q4mQWmuvSJx>q
zwB-b}#!}$c~=EM)EZGMQs2xcLJEt8|=q^dU^h$3DNbmCKK11GodSx`WfN
z;#St=mDr`-~C%bgb=k3Vmn&lWDk78vv1<;g;|2Y`Aa%h>eSp7p){^!NW?`>sk
z5!C_!bD#!MyUTsAsqvv2v@l>MOa9oR0lMakd(LXfRw`;9@wrRg$2)TJc?Zwm(J{gh
zmY71gS#aya-k!^Fe0PL)L?Hsmg1qK%bH>AUQ$x^cg7sdtnRv)>G4B7<$eG7Oxwdh9
zI+Z%b!KsKWov5fp2W4+{%DzP?Mhn?8maH{Hs3^-}>=Mb^49b!s|P
zEMuMJy>93H^DdwFbN;an<9Y7qzOVbbuj}{wf|!7>Cok66jU#P5n)YKCIB+3^FI1|D56dR^tlu
z^{*e2(+2ybH$G_-J)5t%v*c)&G1e6N<_svs`##`Jo!8vo{^^UQPxozlVa^^6Xq~o!
zXkP_39XJLjCJnv%j(_uOz`dEhkhy(}TghYcx;>9k*_5*|_PzKzxuiy^olpZPsPs#6
z>+L?R3d#X&l`p@Aa9S0DbTtL
z#f5~eQ>NMbzlJK;gQ=on{|?_*GjAHUN2sUOnPu<)t)I=7t$8|JV9#5O0$R-+qz&)4
zX)aPk*G5k%@qK{PF)B`^Y3)0?UnF+;$R&)@aS%F#1JCt0uRK{OW6?L<*x%7JXP1TahR)vP=WTrcv!1ibs+`pNk_K^#wTQyK3yjZ%AJAWA0x2u6$|srMq0}
z)xFLio}Mkctg9ft`2=W5QOOuW+k%W%fVrSIznB0ncHN%{r&-$)B{8dRjoQ(|;Z<@nVr{dw)D|y~&BFz$wf;$02
zc?EI|3U4@7TvN-^tzV7Y!K*d^zqeDvJ-G)q^_JZ?$X3)Ef;G!X5*c}5tlx>%S<68e
z!OZ+-6nC;XGAMY;re?8I7(bF4TkhU>rdLfNmL(Vjr7kJW^Yka8&m}rQnbv6RGxoYR
zC!2b>G*9Hpipmnj}TYZD8-|ntmhhD0U7Vkos|&g>Zb@
zSAVes-4|5xFLt?NFuM$G&BSlZ^G=E1?#Q9j2F3eLVWD=>fUc)#1E&YY=JxQ=>l|(!F^Ds$%pep%`C`Ns
zw1TV2;~}zXl2;tj(unbj69wMYcS<=gx0F@{OnMRq*M$f%Cqj4pRBfljyS*ZYyvMIE
zua71Xsv_FEr;4`L^vhh0GBOy6ZfDD%EcasS)l^oSqZ+i^o-3%#Pfq?0N9{8jN6*bP
z
z_=F+pfa9S?K&kx(lXj-HBN4LZ!bsy@rQIg2B{
zNc2{P`~*Y|k_T}ukOJ+~Y|DaXNKuv-ci)-hw&K6oAJyFSRPlVE*1T1p5VpxVX_Z7G
zkt&=%iX*k{y~s!!0vXzrIk~hKnAt7ZRJbN0kqlxCu2x`unVn-uN4`K=(C)o4RbyN3
z9QQEb%a7ViL&Q}>VfK8rL(A_#tK8`<=e&K>x~SP;LGex7@BDD<^HOc_w^+L9@0H3e
ztp1qh5HKkCE;jdf4uAmrV}9QA)l+d#r+o0|H%&g5$E{vkd>Mkx|I>gA$UaR8PNCk)
zUjV?>cMD=v77rj=X{NsyBOPlqRx9wy6C054Sz7kzLGx!djDPZ<59NSUADJnka|zYL
z-Bp*s{=SV_E`RW=2~QHvx$iHk;O+hmuy`i^Z`SG<6U(cTE&LC8&+>Z{LN;>Pc30x)
zTTHZQzAISvTkKrnf%8Rc7tR}=zl)94t}Q#=#+nQ7^{|ZU-IL$@WL@&fIwCPa=<2x>
z?onR`CGY%ElU@m>B;weqi1!mOj-;c=6Z@OLN&)H=Wl?e4B200hI5
z-ag&11wUdkugaVL6B~Gq^h&4pCgt<`6$p{V!+!AoI22H*kLB8OznuR{1
zEI=`h9%|Wu=v+|uFXH4cEACu#Np{;+?donqE(c(XA-8`1OXKVh4M
zF7HhJP&Iu2kHD&c3i!9*G;~8yMX8sO9)2CMO%jBmV+O
z|^FfUrC8YI9pm;
z*0YypqkTr2#fplG>OnFTEps{s6`Vc+g4YXlN;}|Yl$#%@UoNk#+(E|%*LW?@xzvr!
zieb4k$V^BNDAwvhxcmdXjXuv3oXJlm5i&DO?AA~yo*ep}{YLZOP|DviA
z;mmUz1-VSHAt$D$%Ig6owFFRGd2IGgn5PQ%uTm}xp8$|38eR~14SA}mcn_V)&621w0i=esunwiOf4+7G+Yc(RZ(#U*WWU<$=vkJb#-rRXlSSe&Ro%JZOt_R
zOsXGy(F+m>#47gh@w$sV;?q*vfAj?m3iSGQUu$e`PJ+n|(_pKpC^V&I8ok3wg@tk5
z6M%CDt|{y#d-jNH5svMpw6S!d2_x=EK16mLurS;@D{OqYVN6ciOLeKtz
zx(~3iUi(vKB@y^ZPfsuYAVC|D(Gp=ouq@H*bAN6z8ST27*#Z8Utj0!tf(Tzrp+)}I
zjqB#l1+&hrlE2uq)P*Ldu%r4Lu{&NgFt)x8m;G#=G6;6R<(cGKGJ0yQu*eh{&4;#c
z7u5P5^88wN9z2q7L+^>A*GAOBn@fT@Zo@-E&M<%MDyM@~zULXHW+Q*u1@U~az-#vC_*
zUx2N}Zec(zyc2n*AU=V%uHe}-Gaz`CS|1Q)9|+Kria8!J@#)z~oy-zuvPhj$LUMA0
zTPvpV-MbbTy@aHsJ$hARdFji;tO(^MVUPcb@N#MpPuOFWmIHmI>v;aT>V8fP%&)7|
zhaLT}ie15NgI2q2%q_9TrY05&RAprraIuV1$NLa`rDEzZ%P?1^IpDu;Lr6F+I%7%rq8{Qw`K$h!|P<|<)lYl;AU)ke*
zzw23cr!rMb27ZKIwyW&*p=wKnt?}cq=x)@9+DP5#Sk|>V&=Afb>})qxlF5EW#dGfNWWaW`
z_@0MYFPBOLatQY-
z``)$U8-#xvjxf6e0;NGMX95TNyjE_0{sNP?0TT|ER_Od{PQEyaUP??#%7gVdw=@mP
zSpVY#_Rrl@JzfrrRFs!H_g48X_bpK9BO`x-%0f0iN1Af{La)A_9!)WLsYyq>6X14p
zLqrp#V)$02psH#K^NemnnAy_hei(9?m$8v+c9w#hdS@ua50{~Lar%aaXZ7_Pb-Sk^
z`@5ePv-0&wZDE|u=_ijLKW4g@cep0DjkN=b->BEKPC&GWhmzzGodtEy_q#{=J9vDU
waE1HRV@(gxyu5z#Fh6|C+5w~e4?ppaI<4t86)M;#4S#{r(=|R-e8M60U$ntLOaK4?

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_201_1.png b/doc/LectureNotes/_build/html/_images/chapter1_201_1.png
new file mode 100644
index 0000000000000000000000000000000000000000..13fd31349a36b9b40f6d70753bd35a54c52c5129
GIT binary patch
literal 97061
zcmbTe1ymf*+U`390>J|W*8ss?g9i!j?(XjHKDfI(!GgO5*TLQ0-StfVdw<{l&ROTK
zb?;patDEkcma6Kix1Znp0F;#$L4?DB0{{SsVxmIw0KkVz0083X6Et{B-`eOVc){WL
zQ`J$y#@NwC&)x_irRQjCY2#>VrcdN-Wba^RW6ewl{-Gf%l!Jl<-Pd|zNwx6N0)g3
zzkkBk83xN|f5B=2yW$A}v+zV6zdYKyFJTHR-vmNim(O^8Xb@r3Wfw?E-Q0EdckrC|
zxmpSP@6Iy0YEdJF^Z7qR{quq2Fyn{$_j=&{SDZ?JqEFzDKrGS!uhq)lfr05#1@aFc
zJ~$Tfcs|4s@VHM@YH=BlrTuhwZ&^Dh`LDzJuUW5mNM-YSUEJP|IkyxSlRNB^1~Dn*vjP98n2Jj4+>iMD
z16Mj$@SOfprL#Kdj)9huEk2fvZ
zb{!cO&5!>aDGV->)93AKArytA$>W|5K&?_~csNz)crf8tP(TWegjaHZ9G3gx-wj`x
z$>YJcy}sON1A1`j3q$8{z0l2S3o&hi`=`Jc^rJs+n4!@>5zznLYr>!=x2gX28Gl(J
z(A5d6^xO>&QOR2-->PZ*Q`IBKkbM?yVXBod9xwx
zz|BC=i>^GPlad?jXw;(glNf*In&UZ!B0-ZhsU$Po7gy|%xgu4$oMSE7z2%UH{3wU~
z+RSomF;%0+cK*wOU?$k!OPz>W@w3}Na`<#ZzO>XlSNuvL0rYA}hqf{=>8X0+b4x9YP>+GfU}+$l`Y7P@PV^xEeZg$VO`{M_ZUkXeV1DSX`AHAP~CC{$6heMGw1vi0}N{?TU@N
zk()+^a{VEf7ZF_Hhfa|W%dC#90?Q<&XMXFZyK`UR{I9+ADYV+Sz{pU3szfHsO;IZL
z61cAQPM!VMzV)Nb{g-grxojWFU*b}3T+DmTExP78gYwrbG)cQAgrBqHq7>n3ms?~P
zyZu=b({@eHuU&q80YpkHGnT6TG?ah!$wT>}zVy*3{){L-9|@W0_J&SWYFs|wbPi1|
zXbF3{lanfC#$dT_8pl~!IhW3KrbO^X^!1cW_ONsPRd6u}qKHJ8NRC3(=hAKHZs~`l5JC|~QG_~ouzaFKd|bkxMaWLO
zmt_O@{aB^S0RDY`(H%HNzYfy(CX6_zYrvouj_!9|xXWoipKHxqVG1oinA8taiWFk&
z75+l6Q6E#$8+J)|I-@!JS$TQ0KEh*6fQP-6)m3Z;UBLbMIyn)M?=PjYt+TVJ)>f`e
zE|(&uGUalOMmhs+6B82vczm>Vb~Zj+X;#yI-s}na&c)?$Fp(Sms55oBGM}wSz|7%v
z=jqqhSzV8BCypOxuwd~;w8lE@%KVlgvlXRun%pf8wySF?Lby!$
zVqmZy&w5X+O^2es%_(`T^@#UW7H0pe)nom=d{o_yfN@FARVb3XNnDD$?sr!lv2@Hh
z1{18K3nGM8Yv-rgW^f#dXwK#F(L%;aYDQINp8olkqs=ETdJ`7
zkPYU37u~YtX$MbO3)^IJE;C6xUYw0o`i(4Z?2Qo~`xKog4FP7}+Bo{O#A~bL*^ie!
zP&RS>m&@PWcVkmDbhkEtzSMB^M&SEWegC}emtQk8IIP_D+k$?}ID%x(;M1=zK*W=I
za$(&Ax#`w+XeKxJHhmRN#opXrgw|&F@Z?C!f;r9Cx9*$ksrF{))L%x2+B|JRP$Bhkd4CbyFf`vy|p2(@&ozs>=3r%z#zus
zalUc+_=F<;6PbA|#YO}V7&bkjfU8kXXL#Rsb}yi~l1e^$~ZDk7;p}!2?
zBx{x?8f1^*Rd|MAFg&x3Lyb>eFy_|Q;wfFsW2@JgourQLox$tzju%~OP6+JFmlJ_9
zm~gYbt9R~WIN9i8}zEokF9c`q$R7dP0!@Zj2(}8IICg;N@{&bV?>xZ;;u~hK=r7}<54}Ky-WbkWFUcBnPRL5@e>b)yr=!8~RdLZVv=sR&
zqhn`3gj}M1FWlMk{9y#O;qQD=F7Z?=QWBtTqd<r#B5QBdmPklBKHpS3-y
zj{vj-QXqU$)g-xt7EJuMph;Bb6KXo(gut=Xl6@`&d#{`jMax24e
z$?#dKNV{vKLoeYsLy;Nw>3L8ig$3wpoD~8~wiscAb5@8d_H%rKu%|c|7`&LE_v8HxnBRDB$sqm|4QvvHvnYPMXaFUs{?o
z8DqcsA$ZJb{x#&+WH}k=jQiWcEt``)2#D8MYDX(<(khkcL^;wd*qN_fCMF}X-@kaw
z;yAq;vK~^dPdoD_GZ|3m0vO~&?u|tJLy|W`hI>~E;DZ>wzNBKN?d=;x7&+c
zkTvu=b^yO@kIEYr&8E09oJbuQ9K5x;>34GrBOCyYsM_L4d1MKV6A~1ZJ88nZ4bC4Re=NmVq+AM)5jNdLCFeDhG}cl_&TZBq7MCij
z{Vk)9PqoUqq&zhji=2(L&VF4v!m|wNEy$qcLk-fdY`g2jo?kU3bmmjM`x$2Y^-E7)
z3M(VOO;Y!>wTN}?^YuX^ZpoBl+spG|`pX_!-POVP>^bNlQN-LLn)79htq=)W&|Du*
zxL+*sw)FlI~&4V^Ns7EH$ImyqB5eo^OC^Nssr
zj4AXJ+`z&L6o~>!rb5y$M|3BxV`le9*C~bRg~lToz}t7*q+sjFMfY1=LniD-(E3h)
zs)hS5Y?M@g_b9`5b$ja9%hE&Af+7^;f6GA~mRoh{H
z&)?(jX#`%cP|V+9un={S%+Zeqb*IiVMwG|^rR_YwezQMkbJ8G`mQtb`Y)5QgnDnowS4T+%9K)Soa!D42H)nHv=HM18d#kQO<#MMyhUU`e%>iNN}
z)a*p-mIokX&6DePpTynr29ALEdypbUWu~jD>Feu$;@^@*lX2n!V_~uWwy*Aj2VN?<;AY)pb13X%TW0`!}ol2FB
zcStd$?TMLRC+vA<4GQg*%xY~4AQQ2E-YJ{m_Rqr%!5FE>(;Yts`&*cKAvJOQ%y>Wm+
z_^1vVhDtajjE&#Sj#=Uwqy^D%qw*g27d!W0c8B$JKxea#W0WlZb
z(FM;>`t^1s$~s;@mtLPuuX4yAPC-LHJ9Sc8JYGTFtk1PJ;5&GLw)*wb^Gx&#>rHt0d#f=~I*y1v
zQ1Gfpmnl+>+5bASdPhK)Et{p$XowhyCN#MzGA=e6Nh(*Xg|1g01EP_GBU(uC4p8(#
zkoE{AkFB)=E&!65M1Y`uG{tz&R7&2F+f0m-^IGjoZ7wLPG>{K|XL(I0rgFb@sIinp
zP%fb&;ZjxKD4dg*Z6pXQjI2^L0Xl_lA&=){iK2dsE<9|R$?oBW@6ojT1<-=_)FTZC
zCDUvXi%UHg6x~GORH&TwOqth%Mzp_|B|k^m;vquD4+W0+MK%gYi8}))b!P8>M)DxTM$E
z)qSN~wiY|xxMfQcXMR*!AmiXy?h{02!AJpj1I@Qj=EEi#k*Qt4er|h$@Fcz
zvFFK@rxN_4w%Z*5!x=dbUbEXBPGt8hkS~(SmKG9%;&M3~y}Xr_k}6fLCT6!@*|^#r
z3DQsb!T;zrD`RPQZ5*D?uH%v^0!-WeO;?D&uUf2QL6F&z%N!jW1RHP<+S?pD#>@Eq
z6@*g(IJ~zeqE>({6|@5+&~n9lU+X8O1^(OyW;6RgxaEtfo+6f(_JVLTizAhNm#KZW
zFz3}?wm8GlGWQKM%J7T}>qmC0!1qK<2YO>?XD~PFn7hX1{)xh4az7}12
zG^zw&oQX1_VL&=N?n>L@loeg}3`MjQXo(#r#QQ{^s-E#l{Znos?>}5!X|l
zvfGnY=4f|my2GsLN6}U%#AXMh0_i2I;T$i_smiy+LSb{OfWx|G(NP+JESpXE@H&i?
zCQXfugu#qUiL+*zJTO;-T^{sa>5}DD=G`@I#gt_oe|g54dBvJ(N3#p%Yteb9$>pVF
zLLlK`L1vW_oSKLYkZNHYc(3NWoV0sb@!-#oTC#IuYL4KvI=ulA|^-IUTR;}U9)@rDQv*0AqdVoh}FRkehPO5;k2a1_o@4=DT;e+PH
z#Ht`p(f8LgLdyBd+!9S}HM>)1qm@EXEE)haEC;m7M<8qJhR(~k_MGoP+Y
z73!UYg;Q=$#I9_fv~C(*W|1YG(t8yf#YRt##DdbszZ+R9;5Ruxg^hD7-ziJ3ZlqRA
zu^umFBv&<-I=^yGJl}p4#b`*XJDiTX4!OnE&UJT0Qxh{Gbza&|$OGaYy3m`HioKY_
z0Y~~{9|~c>{a;Z1=y+{bbjtcYsLdXqVQw!$#U^m+{czdlTKn&Fpvr>d)bWCjPFOKe
zFP+J`u(0r3uLrt63O`(uZ2RT$8jsU4G!mEH81Y*&v)7X&nRJHe{n_eYo%4>B%kiK~
z1v~@I{!)q^>C0>fpc3nIoHQqb*ky!@63})rhu_I8JRrlG6N*NzP{7Yi(YP<0>X+I=
z`BqLv^ZPxxVJ)xZm@oL*We90arg6@dU9KmZF(Yyz(6KkS_J)O2x-5opm)FAkN4K|$
zcqk)%l*^_aW00i+(qQodlgykRp|(^xEtopwI{diYWsYa<7FS$Cfivsj15i?YM@y%Y
z{)PWs^6#NL)qs-+981f3K_BDl3_ec9?K0x91c4ukHI7dHWItc19)g~nkUKbcCszis
zbc};c*t+FOnJkUNS~p^*pbXK@&q&gf)*@h~V%qBQ6I+v;LZ=TdP57dGvlc#;$x@W_
zjh4XiQ1(}TfmZBx<}=aHw@eY|A{B2N+-WJMsB^SEj{O3sr>_1wvA>OBYwX8UBCbiD
zuWY9XSH2#05sYF~K_tRC8A4r37(>L=-D^k+Oa0MbWhaV{f0`#Ocu9M%?gCV!Fj`(|
z4$O=($58JEK76dzyTSnsHzgR1UfH>}NbOD5!{9C1`aV8m;R1kwVyyT|&NC7(nhdmId!mc8aN#QtFwF=H1OU6XK2inRZ!Z|4rFI
z23>AnH8!<=7MoBcKpB6fl&p?5XoOIzl7!yLmot9`FZxY3zljwsjr08(w^KqMkFvP4
zJLlRoW_YqUFxK84Mm{8tVi}3p4J5#V;#ctYMgpxh5Fivr#v3-Xy`)EOrV^EJ`o<9)
zfay<#2!60iO>WlD-*SObWX_NFG#E4!YsRz;F)809Vz{^r&JM+PE;g}RTS@902Xf35
z5YjrZwNY7(;KazKb4nCp(``B+QX>jhp5gK^f-CAmGko8een*@GtOk~+m#^9k
zw~L)s?sDjEDsvUQYa}yiXpyokA3d*
z{NJ?Xo7*v_Yff*x(LaTSYa9>d&sJMQH0cJ0hvDey=|{)Lu)o_9nSip!^^_3vVHdNDeP&Qxnv;Jy8TMB~lk%pG*qGjsHmsXb@&m2!-EjoCqcu
zEn5!vM(mxv_>)yC^n8(*jPwVeIGvj@$JZM8L*Un|rUH%X}Z`d{E6ER6W(y8MyfpAMr185qm`^uMk}K<+Eu|MgmaE0>S(;*-=79(ufOCYCH{9a_f)P7R0keN^N
z(s?uATK+zX%3x}TD=O=e3e3jV;z+qh9kvsl^OAmrg{+q`eF3z69%0+g@XTp}No~hk
zapll_yeJ%StG`^!eg9Ded=@M&2U2m!k064!jquWuS}kR?{is5xeT-h=_u6u8CX1U7
zq^PjlCXsn|$IvCSINDE8za5{%^GFPChV#qk0r7ey{JB4Nul<{jR2_UV-QOH4c*
z_FT$?O8tl>#-=B*k}^94!73LMFmMsyTy}8?&Uc-lU4NjDyr8JewuWIZe7wGzh3&8I
zr>Q=d_k7bam>BorzSHXI
zg_!k<^u2bDK*_c$;Xg?ch6+#)oW~O47FUr_rfcFuI~8cXBjrqfKEWI{u0re;>eV&Z
z8!qWHpipXNp%kOc;JLsZ=;SGKA&AFts!nIZ0?P9b8xZ=p&*w-_j6rVHSu#<|yo9c}
zi9QD^hV`s4V$C~15WTIi{z*EY;F2%~E5Ujj5&RShaGr#ZHERx54I_tyw_3OiX18CS
z6d$>d-$5&Z-(Zre*BofULK|4|aI1jEx+U}zs{FBg?;j|oBOpI#kjNimja7I20Y~;@
zC$ZhNI7{{9lG*8sT*G0gklBx?YjRTSC)BBF`$VxpognX$uG`~zI~Nzwl}f4nH#Vy!
zA+Y`ci;N5^LHzV72OOmn(x(4^03|n*_PqDECMg4w(wRK(lWWsc{c&y<#!OxhmhLR4
z1Js>A`aw9ER#x?#N|yvY+1^$mQ&p$3Z6Y9;ocRd?)g{%bvc4ZrbvTmX0%psFg=?t@
zoQWY&+_IdPCLMt&TFOQK;94HRCuEC_VWi%9Y(XY@9kG$0KkyG_w(cr
zhY%u|fKYsn*HgP598u|dsMQvd!4sJo{92;@QJ2Y7wcjy-?ew>1Jwa+krZ#qsfsw%w
zriBT!*?}Y{>stn(Q$;eBhi!QY$El3bYR1yNj^Ic19ApZcJsUv6q+;0B`p?E;K|#S5
zkGJ1iDlwXT%KZqU@?8mNbJ7YN82njUKL>}D;u0pK(LRh#{BC!M7JZ>)ff5I#qVM878S^8ZbLF
z)Rt0>ka>Zz)fVJIolq#TF>OY}tJSZk3Hn%jCv`I%
zm)cyv_Xt^Pt^Wd(I56MAt50cBy1%UO=zOD+cOZXyG
zE;9yH$Zyb@9($ZB9n&S(`&FS7H;D6`PDi#67n>2C)~hX>Cnph!iRiGfu%qehRb4fS
zR(pT`LW3Jbh>fSLbz3`|a)n}XsqMjtDj;$sG|@#Aa*=b=;Fp@{IEQnY&WU;hjiq|?
zr}I9y0{-h3?xx$&WL^jW-pVaAeoNzQsR9N1cNJQhcgFU|%|@GttK*;Fzcj*ZJXbuY
zaF`R|L_)(;JB-pu0)ca#E`8O_@FcP
zl17hAXv$OoFLlQZr>%{Xe~+(Nq8{~@?{WQF#iiQdrPp8Mez}nBq+>}&tF)|eadBsW
zC-EJu4Xd@%naL33x)+YcJKiurSJdu_`8)*Qq+fsu#*G(5lyoS~
zejlhsEKs%Yb=4O%0IWW2_9ak?a0@TVq(N%ad3&{sOo&+IMNan+iMJ!Yt?A8R=vezs@Ud-&ZDN=
zEW#}<5`&>7)wmKf!#n0u=gdS+=A_l63`GqqnJItG|GYgxIU+i(rrMoBK}a*^Q7+Zy
z{_5<>_pq;OWDx#zTP?nrhd(58mk*4Fj12CrC<8+V&iUr6O!Z7YpPxdxHciIPuTM^F
zHfy6TPR9%k3^`h@&JT}|b~gu;ABGGz4FxSX(lhDVC{P6*3=kwEYi54J1YX~qR74D%
z?%ix(+W6#)KYzd1PiJ+a(OPPJqd31s4$;30gbFs~MnZ(^nOXUQrm#)pkyRhZPyuIa
zO`z@8CFqpYAj%#x9(`Sc7N>Vlyk!$wl=zwOFOUi%H1@}HCc60^s>vi<2)Ouzo^V2}5
zY$+@Dv?2~`Jd)q)MTHf9-Lop
z8T{_C(q0v0C*)@|oW8wDYFKxSm}BlBRy5zpc-l=r8~dM7K?eHP{&GmoN6m94&$~7y
znw@Ya$S=@c@ftFwYwh2dnh^vI1aX{A0~6a5wcU24oePZArOmoG2S+7i(hSGYYHXRS
zLqJEPgboMnii~y%T%GCmn0GOIm_je-D!1T~){0T%5A!aZ&wxEnt0f+JCm--!P$$z2
z9(sKHay{BU6Xu6Lm>O}@Y*%v2!OTp6>ypC|!vLH>mj$>6uKDH=g&-lRqu
zzZb!Z;STtXNfjL=fmMdN;?)7P^Es?UA69|Q4mDRO=-9hr`bnd1dBfjRM
z_L14G3*kkFzKkdRUcSwc5qyjkIR)@a1A#i|w9JL$V(c9}X$|SE8qLefw7dF+f>y*<
zUTHBTRLn8ud57$LJdTZ5s$K%HrfrxyK(7!F#?u76luG_v*&T~T&{0P{QMauQtlBs~
zDu&+N)QWavD(7rB-9GV{Cv1h}(jICRBW&Mmk+N$2Y!M%8F76&ZlZ|ihQP;CD@A`wr
z?vM!kBdy1M@NmKL_=dG;2B&|L^*Oc9!#V1?>%CDj=DM1ySW`ttN6&Wb?L_01#Gzw4
zivImBJTqtVV)l}CbYXAo^tw@0Mgrvg<`X~d{MMn9U^Q_66CHhITKLs<&5}qSUZFh1
zd74@xN%Oqgx;2;_+azS>rYy&qzI2QbQU
zkI?gbDgE-=9!5iP($2KzQFACN)q+F-!^*GDamMVh_w5|P_%!`<^L0<>;9cfabc@w7
zq^_lL#5&1(gc%2Iu`jr+!)RHq0aC`&;v5y9Arz$f*O)1p7D^*yxcqzr$@9+Ww0;=uq
ztRYDF0;P&mtu8i4^Hnm<_BOmCnDp8aAt5kh>Fg$8+2v%RI$$WC{P@=UXWoDlSU4&u
zEM#;tziYSJ?vLyNfueQ1AHS{FzH}UsH#E7`_Kl@iwT|ZtjoGUvPeo;yO;@o-ORrC)
z4h;SY?0z=6zSwa>l3uuYubG7V?bH_)0bd^Mi1%}LN$H`W&Kckx?We`zB7P64maoig
zHQ5_MFb9Hm+FcCT4}=bqD;A?IqDk?UBoic#o#+Umo_~9K4q#fuFHtkd&~PS9$cSa7!F}g5z{yUv3G-1$~zvAlbgn@^XQGCrRdO0-xThZ~X7+opA(b
z)e|2J9PI!t$R~fYi{GbGMJnbPoMJ_Uku9S6wcPGxTgx?%6gNlk!&AunV(NF*tLkYibJB
zz`$Us-eLgUHf4&LN(<^{B}
zd4gtw2wrUWa9PY*8!Ok0dg3bFO~iqryH`6EEQ@R?rSj5kHWbO^=JUx;^;0v?=1|}V
zrn8ogPw1ymt?3n0l9fiNX6Z)aIrW~*Oq5oI;#jJhj@wk{sjW8mo!;(Ay=T1zrF?3{
zL=J;-wAQDtK{G*}byN$|Z4|2mPzI(F#hA
z_-z=CCTloB1SHohiG8psm+4e!fL>XbvO2=w3tl`fz&s+aFR)vlB{9%_L&L6_wrU`yujq
z+|9I03aw^++1I$=xG)qoZ#iXT+M4}03v|19(XMbIev9H@6PJL^zS}sx<7g*nJ0oz9K
zif<36$w#*TPL@2z0{;RA6C=eQm2Vo5tmI(xg=VKW&$gYZDFxVdf*qaB?dI?22LZ;M
zkU1(<+T~iU%I6tD*B1X4=w%PUew0w~$by26EhaCIS}c=Y1WuR2o|Z_W5wo>rpw~8~
z7vOc+L!hIhbMMpv4^02Eg?!egHsr(D*x2aW8H)dlgv`v$Jnt4w6w8z&QAmV0z=?$6
z;qZS)tf>Mqus#e9Zoy3iTstTKkE@2$ypu_)$QvOfqz?`~5Rq!92(&Xr(0yf`V
zsw856#Q&S`20_
zQPr5Vf1NpmJehEu0u4*_9-81UzNk$={J3eta0L8=kf8Wo3I8VuSy%e;A6)W(2S)$c
zPX0|;pmIh38=U+M!uq#~h3HdI_QAhvhJxyC+d%VY&wml5&w?NSMVp{rUQnQny^5db*0#K|74Magnphr#D&LM^!>pjIK{SzK1(R_X%?c%^^WNn!%8?a
zdi{h22eB5jDo-;t4&m@wt)m^pSH3)xH(K5ynh3W=_;BfZWvqB9cCMvh`I}n$Rh@)Y
z5!Gk^V>FqKaC}fKS$C2e2t_(`j(iEj#ok|yJHoLT~D>&z`=nz4AZ4RKd3)2C>T<KsmZ!FltEV&5$MvQr#pi&F$n#gn+pl^9A$v)RIrREcTiF4;Bq{Z#S_Q&%h|S*SdfZlJ|QY+rD2
zm64hG2o9HOByDDYp@ZFbU=A%?Z>|KkxPVDaF?bSjb#?trK#;}gQ>I*D@9E`x-uX@d
zP7iHyJoE=!UijW$9K2tW0Wm3Rx-6DBKwQkv=R8nOs-J^-LQFKMohYL8s^o}CupKBQ$D@SkjaN4%+6IZgP`_IS@k>_*na;g|=qF#6K^$h1)<7M&qqO$Fk
z^SCogNbS2PF7V|eKbzcaKAUA5SXf7>fc6$ra2@vUnD5@)LMpafuBJ!53knKr_oLIB
z$BnG5#vjlL%CXuLU@Q7jw+Kf8Rs5VzM!%B!yo6NRT4hY)j-mWLArG0AT1SIN@ulVU
znb4La2Fl9FSv6*gz;>qpSWZsV(|dX}WdBU1)cf!b|6@5B^(#ZDR7vs*NMp*=<7kUE
z(!wYv_-OyqP2JA4)7w9nCZ&C$}*yJ=AQ)dnRwG^~vK}
zk3Ns6NJY^;L;TIiymegLC|5drlD&`!8NVVXCS$UH?qZ)Qp>OPWS#n5Po+Yv}!(Sp8
zz;ML;mzg+zMBut5wmx&+DnkB=j_0cf0yC5JhM(#uXy7d*;}+k+x}V&P4R~}j2Ah1;
z8U0xEI}O##E(0H6rk5>zrL_)!86AAjqrPqd83rE<`FlYga?~ROoR-;hRyU$(4!N=q
zt+-`zs16y*!^bTS@_$>ZV1PND=I&5N?DgTgB?G@Vf5g@J8#8u}(BW1@CNmuHW^CU+
zGI-W(cP&ouRD`+d);2;a@{qScBWv%=m2w9%x-7+)(NyM;hldBt;^pP#zg89`Lf%nu
zXhEq~LrOqEaDEHUpkCkW@`F5CZj2@oiTdrZFFjYT)&rKZ-n@)?2Ft+yk!vWY(7OQo
z-4nS4MD=3~DntgH&7W7=t2S%HY^$PA?i^_cq1+?0Ic{16EAG7#*#H32*;6Xg>hjTF
z>XA{FRQ#T@_Hwbz?)i92j2E^HdGR{uHHycI94QQPe9mu&dVJc=~#X+2U*|
z(W)%<_j#!`my2vl2Z>$5LKE4}o8h^$c%9=6fYQ-^?@}C2UV9KR)_5q
zC!7}6Icm_ib=zOx)n%uKp?tGg#qGEFb(3}-hu9-flloa`^H+;B5*#*0#_QJ%BSQ`5
zXE7rz6{V{XILuaCD1FwOMy~fe04EC_AzAG!>l~mR%KT68;b0KK_hI9}RAxZB(sRzs
zG|Ra4gis9K!$Djf2Xdm;+)kPscD!bVt}~(_FbG9TV)4U7ULEbovdY*rUUF7Po;a+A
z_>#Ftr^j5?ATS-|ur~hQ6i^TRnms#jJyI{u#gwPyR3Hz_RQhFKpAAab(1;^
zp8F-1)9<^KtfynWgveTgX(e8n(99LWIMg4xKg_Md$+SKC;lR!pKNN~DL@GqF)Ds1T
zU_w}L%<5{lPuX2{-?@I5QBdu{oXMH(L7y_EPa~Fk*ZDC2tTTN^`@P|n0haL+94I=w
z+HC&G<+IO3gabJ%Rk`64ta?lOMmxX0wmpBcYBjn}MZm(sVenX$WTRWi*mWzF44e7^
z^vi{?55g<-iy?RP!H#FBEMBiC?MTq?
z-wfa|MNDjInw*V|4eWPBqNXm>+#ipcSt4*UGzHrx6tPeZ&=^?8564}wnYB3i9&bXVdGgElFKL%lErPn`>6Rq*#opwbt
z8llUPk9B@Cd@~#jR!n;}S1hfG3V`**?(Kb>zVvuQ^*v{M7rm74qr3wu2#*DS(=bz;
zjOH2%<>6;nHKDu1c?h2b}ByG5e@OSc91
zjR4&1kidWCWn`+|(I>#mi%CCcQ!Vd@ARP8w8gy>$W5(>OttBat4=<0L(szDxSEL+A
zZa~^G`yCS?D-ujxU*{G-J&fVx
zNmD>(!d+~-6;@uyTm;{s%j5)e1y6GkCGjzN=k*~kEpxt+5{;v1AesY#>
z=oRXV%SgyD%Cb-VUzDA7R8?=k?NJ0p1QbvjM7pILlx~pj?(UFM5D=uhk?!u6ZjkQo
zZZ^H|!~2W(J?H$+9rymV8H~*y?6uc^p0&Pn&d=;!xlpO6{ttmjy3d&Rnfx}T$Fx&o
z0PX7&CcBSf2+xGKkFu(zxAi4e<+*m+BO!UBNWt%JFBQNIpYg~D0=qN*DMzgqkCdX}
zx2}e5ID{bouDL0g^lDm;$82{Hxw&1X!IQ+sT7<&9n2s?Ke?Ho_!Xic|H}CV_5mzy-
zCXOcG;^5*`hr=o`A~upJe~4Ug1V?5>QlGL#+o$O!iD88d9YB%lR~%!NV|C3<+Yd+<
z8G0NZBVG>)rYjmHUf;?{8Tx)BYj8RIU8CQR^cvG(hr5!IIU}-?g0`b#x5U6jcnli}
z;-y!E^|B{q5T&>wPodhT(EvTSVl!8yxzJ(6asBAovxWm
zCvuc3{|AYStMg@j6EWw#1VlqVJv4e&S>23&L(b-XZ`Y&~j-ix>JFj6*spa8RmNqrT
zp$PSz@1X@<_GX;>^xW$MrE)tmFXB!UGT{cP_s7$L^sZA$8GO@t4v2ap5F6BCc(2AtbpE?LSIyYojNC!yeOrC
zPu2}Jufm@8)-Cv?GO-5{u2usni6s^tDXWgo7GFb4$*c3GR0>q7kHr
z%=&McnO_qS5P%S>Uu`YBjg3l|ijaVSwodB&E06mtBU+6*fu3-Z?a5*SwJLL+-#+VPBdPlC%}3iQ8EJ}BmOC@9MWACkIgkoiB=!y6XFSfs&stK
zL!t5b3qEYU?ZFsak_#AUkG_%Y-_p4=mr+XQc9t@h!E}fp)9v*_=+J&Z5WqScwKB^;Sl`Z(*DBR)5eQ;=DD9g}L{5cOaWMJ1c$I`0{QLG0)AkSSh+O
z-$OfZmN=_UQrA#PnsUs(GZ$zoHv5xb&9aNP`t=N~I>wNaV;k`geixY8Y4Ti<@o)tH
zk+ASl{X1bXv>jEbL&16QL}(c!(_)RFVPE=(t;$dV6NQDawmB+_SEZ8tqBGVb)SMOL
zmSVlx5>IQn3lz)VSkl)yDy{KBDdVG5h`(0oCiK;f$LkDGb|&&gr^_;7FgY3sx`OYb
z_uZnSH-JeYoGONQuxb1V4d(#9YYC&!f(gF?67VK-zif<&)+AlSGIho#fmKtGFzkbk
z9e>i~D5vn1?wkvjuU3uYJZ{6u_cu|z
zoKh=YC+&9kD)H^)aA)CEY)j0X7Z|~lM6<+7#QtcPaAj^t6U7;$b02Q+N;`20Nc)Ir
zkt*Ta;7Bx*!u^}7uSJ1(WuFDks+yvIe@E~t{t-<6|FkXajU+GB
z@H~L!uvCEtWVYCOhfbzEFrhHLGNT3cJol8O}bRlw+WJYQ*ZRJ!tMWjKkrEx1LRsT{pP
zKLEF1zDPKaXXI?O+|OSF3qjk=O#IoIy_=hx@eGqE?d5R0U_^x-T&LcqmwGS<)Ax+x
z7$Pc#N9lxe-&2&{XJe$jOH*Mpq3Elo<4cW1Tf*t)3Za9^k)My7rZV~sJfq3H7djn#
z%moXLFLEWsjI}Wr^fqBj%-SFl8e1(WM!;*LXt%qg!9qQ)`WBjU8~O&nFWo;Or%aC$
z)hAKE`6Yhv;un>32DW1Cw)2T(_6eAV>kTGbaXHfz&0qmil7j=p^tSNKut_O&nO411qEp+GU+rfq2|W^NUsj
zxiU+ut3x@jW^`DX&FV@#MrGrf#umP&NyN#1d*TFEBT!xb|Io-a;cG`S9$!ETpE@WSwhs81qjKm9+O49
zIE+n~4OkG^R+Xbst-X|-#@GSgcW*fDh|tRl+$ndZG@mk2Uv5jLmkEiwHq^89E6_@s
zGx@7Ewd;>O2KxkYXuG6%SXAzp<7d(+#DvnY4-cD
zHcG$Z!Q-I#Ggr?D+qEurgU8>4zU$;~ekQfN5Gay=xq{KZ(PQ?Ad|x*DRbSWgK8sJ+
zB)+Y4+egF0o5u=u0b9RC6ob^^e-+nDKV0{$=uw4Vy!A%R`I{SIEQXQ+`G+2a}D%+_p3M~va&80OK|x{g?;6L
zS1Y{teJZGuQZWu2UghPCAq2dn0AK^O9@~q(=|B?LZ3eaf!eEg-JvL)UnCfond|P2Tz%}lhbRErtnmV`vfE+1y82~tDuOk#X~$&4d21na>D^_cepN{AArB
z=$dXYW^%ynYGn2LveZ$J>QkCPv=EiC--}yDyrqRk*F(3$kxcP&Go?YKOu*z}baNhr
zSfKp@?D9;1mNYShvO=cy2UH87t!~%;N_Kqu=0W|I_pju}f6)~>Jlz8G|1GCLkA3qo
zDgRPb5dKrWfWmzJOLutrABv2C@z{!^@*i+6FPSY-rg-ROk~!nFMuR7KgQ?`-^1DQ%
z9%Qxg%lnfD!dd(eqmRNbZI^@R9&#-u(o}A%v$7fC58BEzmE}GK3g$xa`)AP<55kIR
zg3v<{=yeJ%IHJUsaj|s8W?Y{wni?!M(fHxMK{1}|rTY(@J0E*Jf`xxDVSxo!NJ1iT;gIBSjY{)R{|fKz5v(5yf#G74JD2MPwO*Z2
zx=w#AeIO>i7nq940IUClNC&y`r~fnFA}Gjnw`K-laDW5HF_e&ypl4~Byy^x_FBu=b
z;gwZYA3w0>q>{OM03YU2jVF|+v$K=QY~sb2;dEi68=gb)zx1K|Rmc6A4)DL(Cll~`
zaG{Y(blWrmo50@Q9s~Zv)DB2!fnE3a_fs=740@d}@VT5n_&k3d%V-q5!2WMRs5={f
z@*_X|aG@G&(&pm$?DLsa7^?`8g}DG)ZtkRfrQ$N9A$)apb!EUx=!>S&1D>4E>*#+I
z-rl7g9Nt^;Kek-~k{^gTFyj09`s!L)r5sNU4hDm*YGq~R@dv7*Uo0}4hKP0`gE;P;BS$vdp!gW-63uH`fEUS(MOu%p)hRn@1M7Z
zGIO;e-lUv;bQtuWktvc!)c1Y@1H*H{Jo%T*;Lh8F(q&r$FTIhWl&3uL6fS@O_|jD(FLeZp$=6Q4)jPp4#0
zbTbWaY(`#T!=G<=zVgOodnB!1_vIFK%}Wx)3mc5_%-O$Kgldd&Mu`5Xo07x|6v%zM
zUW(GPU?xMR-+Z4KElIlWPY}k5Vv-UnW)sT^zxsLRhwsXe29fy{a%1^3f-d1&Hmd)P
zZKhbLx;@6p$!X+rdvn;ly0_Qg=n5(0f^H2H0ivJ?xLe+uDkUm69C$}hPrrCi06$^K
zq{{5YD^#>@%?J`Qe7WoMyWamy8Uq7j&vTK%5aqVgtmO6^KUpF_b*Q3}i8Kjy-3@=B287CA;tT`RZ;s;OytwJ)V*7c-|XD
z3)3GG1cY*qFA4{Yug0HAg{b!}yh(1q_8Za8I%s;Ptx|-U1R2bb$i8+-<~Pw3Dr8Qt
z;W9^RW{DH>(t4&~1i4?TlXhjvoL+XpWLMebkkjFm`s(Ojc@I21;5giX*Sho@A@@5s
zP$Ek8+Rm%}h;L~CbYuUVJ1(-tKyYqCVR*0Xg(
z2pG(-mPuuFwcAym2L?7b7*}8`Ck$JkKLbauZ+LPy(-3)=E}NqqJV%JkI&Hrouz8M2
z)j)BBjV7r-AK}Nb32%5sOv{ucPMTac^$kMGo@#)KUzSR8wY8N`nLUCVM^Hz0A6wvV
zt@cEd|AGq-_5(*U;JceuP2RYvP$1@4A0Id%^D#cG;d3=M59{{U))wS*5WgStFj$B$
z4(1cquO3z#mlOS9T=rJ53WxQr1GfYV!99j*M|}NDn^`}$l$S90hdlh-mO(r@Mg2yS
z&L!tha)%Lq4qH#};v{dL{9Ks=^er{I@1A!aL;nh!1o74kHhnTkRFz+yK47fv)f-R4
zwqZ?)SVMd9VX0`8?acUkfjKyZl}x+u(@6U2$Mof%v;8LomyAkr3zG{e9;s$sRTXCA
zXNx~hzZjhEv@d=ZDwBQ*(^4;`GuycBIvJ}-B|g<+_fs)lRGe(58fVJhl$^o+8(0!D
z;t*4--M=4?ALyYpaq`y|)pb4llIJ~hWDHGdxLg-Y-n2H>4BAqo6LD_zyxO}Ck}hK>
zZO9+|-GZ8ZQCT}B4gEpRK=rU7NQ3YJ1iXX~?nF%APZq8@DA~}0s@7K}d>i+9o!`d4
zbC|!)pLR6V6fVRSsL7MDjFujc7irn-4X-ttQSWf1e=^B&Fs42CfQFCAef}cPXUWd&
zM7T1MFjSdcKPZz|e^B;WDM=&g1P~@w3647Vt4ZXO=3a}^mc=pU3ZBf9Wn95MetEY_
z-(O!bmRr6v=^xAAjOw&fh>Q@2o?A+Yn#p&Q9hxpGI?g#HoIdgA=zkKxLtkQJ2LO+$
zZLOm|-(O5CNDTum*m{69g&r^gzz!@5?x8v>j*hIuBO)G!ie&~8@FpYU*-pA%(UNwU
zi#!YLc22}Z_aB+V^vNA|os5gIY5hHsIN_Q#dNoNRpE8qF5j%H|ii1&2NPje7lIpo2
zeXm&LG>*ltEbF=qLmRyKjWkAy8v9K-3CVk|RU0wwC$@Ha`f(=>`JFiu^ypP@qCc9iQXtnNN7ApTm~Uw?`Ib#9Kh4;&@Qxkc&Hn*w&!+85xB
zv*6+Uv@Su7
zTI+Yh`O|6{%#m!QHP=P6S{)M~zrXu(e!iJF&oyZ1-gtY8KU|CxYS>|sL*KwX6ub>J
zXfKIH>WO18@VxD>6?zwgdbNpXd7;_5cyL;3sgfk^yLoT-sb15bcd|^&b9z!#U)Jls
zq3iIjX>IaAOQt}Onev4L2pY`w1XuiIy>U~uW@BJP%OF)@^r4=U*jp~=IPJo{MpBT#
zT^~Hmn)BfXANMQhjY8w;GXmizR&@d{=P$rZBx^EJZMx9FJ~1)zKq@;pNM5Ha2Ul|=
zH8y%;k>nqN!-?JP(z>%Ej?sw0upd)hLj(MT3;>tez+Ha@N$FWmNEOQTIONFX6Dl=S
zxUxI@99%ol@@pg#Ta(-FHPtDdd;^t|lmjXP(+t`pV#yMYgDgP=$BT_?ygO`_#6)x@RXF4z|gH7(0U&N{NOUGKc1NiWLi@uyrtF6
z9{qgJ{O)?VFVLhjSTRGOBO*C6+#vaz(yOU)>I4f7U2F7vvLd4w_ntOeba|3H*DbZ`
zT&^;6m5x=le?$c{~sii4QqBdu;5sStRdmM#J;2f{lcA?InGFRe9VUiQNMYW-&9h+onq?
z&Vt9}JVjJJO&{Nl3U{A{aLBpJ
zwA0_bdhw|d%(sbt-UtM#VI+hSt^`ICNioL6PW
zh#8sbQsQJ6o<{G@7Qb~TK}R_wLZK&(U6sHz$mM+WA^W(_beytC-37883D9UqYdvUh
z=rkpO`DLochDbVvXMJNJ;qv;r%w|;x0D??rt7rf=P494_5f?c5q)!($uh5!n1|Kj!
ziCbI*v#RUwsrF$$_&h3fak;iia`76RA1ED!PAP~v=7sdjJo=TrZ;EDf)A4k(u6~6~
zs_J}0U(HI&Dc;Yb&^EToqxl3@?R4iX2-mJr1H)z&OL)gwzhR;LaB#_UtxOZx+`H|n
z>5Db+(?^0~&~etpE_=)lHP9~lNcK2b1x$0E{Ho&DSh3`pq!QP`IPId5FjM+fQh^48
z466b1{rin(xKHa&{If4~v3&~RF79StdXPI|MEgq+Y+Nhx+LYW;BUhz#5JIll#;E-S
zDfW33Ydy`}%Grh1{QR`=*UWLO2CI%YLW1o|i!|^fCU2DDz7`^IWV%W`<932H?wl!L
zhR7*3U;Df78&Soxy+<2HG@^af;ndEs!8Ua1I%2?ss_e
zT3iO48>qQY;!+=P5*A(JH5DkaeEEjxsVATQE+D)EretSTXf6)o*|_#}Xi>d{Bt~bZ
zjz_c_e>-V2{R%K61|awkIb3Z$J&>Tl!Q^zLW{3*}vkn04B?PximxC@ScpUAUDV~E~
zaEJ`arza=6;Eoo|Mao?vmf++OZ1bJ@K@FltLYZPwC;QOw{|Th
zXbO%Jf@z@4$2jlvHQ_fE^lBDn7`p&WbTLuL(nM^3v@fhXqkZ1JCBkL)gv&@L
z^v$cja#0+c1s)`SS*bFs_d{G0da{kw>x#k4jTX!`xMXN;InmN60T^Rh$>m=
zuD{4@Thv>=8lI2!#cGOll7+=W)>T0e&fOC(IYLgUN5q2dwuU<_9OmQr`FeMJSLf%~
zWyOl|(OSqxi_HbazM72ypIvcl>)dJgaWds-)BMhIGzv^kz5Nqxrs?4T3@BXdJ*3O3
zz2GwzWg&LQ_ynVE!K0(AowtMHej`gEQ9yp>3(+LC4mr3SM9I{kn>DT*Aku48ce|@K
z)F@;+Z^j<TEhs2BIyq^bn@jxHw9R}XsdsuhuHNZz1c=dK*GMwh`UN)Rt?_&e&-*K~
zrlzI>RP!v1;>7dZ?LC>oa{W%(@!k4^`*u>D
z3wUWm5uMiRFt*W|ZYo+?-_Vet6sZoOyeeyr$(PN$hd2$;wJ-o)ez&5@fE#J9lgFC5~
zrnn8zv`>r<%aZ+iQ&
z#)Lb%&es{o4uwmb3FGCi)6#faL
z{|6lZpIpk{g#LouJmXs?`w!HMbI5-T8BJ|CssA)clb;EM>^=>xA8Y%0BNIv`^Yk*b
z!t4)De)=mRes2HMtt@xGcP&3}-UBz%S5y^D5lmh)^dL3X4}}v)BV=DPN0m_Vaj`)+v(>6v8#+2Qj!;o_YK)vAq~EM@BOaiS{Q6C69ei=^f@C@uF}
z5}G8jlqtq$X|u@ZaUdB{ZmRJ^y77icp;y#-tLu~^L8gPOQ7LF{B$A;|NLMx!9c=S_
zkYp0)O@DQx&jkO2U}LDDHjy^_FUX1}_lX`PaDceEat=oIg}L$oTPTPLx#sUiVLqaY$DcaeeO26&E1
z+OE*(@l#@qFKT`hZpS_t@bK_+jW?mMh_X>JFj6wJJfTCeOg8$RdWd$6Vv*Kq`r8}@
zTGijm48#p2s5Exa?zueUEa&~UqBa|(F?Z>Bh#vgWg%qc%N-bk=cbD%?oN@bxxJiLO
zIv3rrI#0sD^)ztXhQj0pr*o2v1*Kd^3tusA+}#z^_j98|lMn>tH~oj|Pvu|*CnN~z
zsW}}YF4ye!Wf96bC(_CyUiV2atoV8=3u!2;qgG`J;bM-nOx8-oPj*=Lz9bfdqX=<-
zmRt$%z<1H*(3L4C;<6(=Hkbkvsv5iU-ipgNsy9taf^@^P8MSd!q=@m+n5jZS)9BMQ
z8OkHqwZ4t*UC~~{6f@p#q{(|c(C+X6ChIUG0$VFRBJpH
z5mykLuEFVE%W2aJyPmcDqcz|ftL&tbkmM1JgZ#7Hu4-?I;50!xeUd~%8o2QvoEaMT
zaNTzlhjoQJ49=a}=wfg@A#Td6wWPD_XCb~tNO5(aDN_`N@rBo)v!`POXPI*E+t%IR
zI5^Ty``5jPnBIf#33THSvkxC&1_uXk&nLB1lK(V06B84cXtz9`sj>how8qwKH63_(
zTtTn~Zgh14bO!5GP*`|%XQvlTySmL#1pC{@D(^87bp0i9l3X8xU8P_qG83fHyjR?*yiB)8KM0A7Q}8AqS-GyTB}KN
zF|^fokL7;pF|#=pkFz>n&JO(iiM4lujh+iSyER3(OyvX=Cfj>gt6AXN6)_ttXF1@E
zkD}EwkcFRd=_6yS2!#|;RBFG^`=)QnlF~jO%vfrwfSAuxX_vIS$n`fkC*|-k3Qx|=
z4!#hxZ-SmGZu0ZHWjg*Xmh=><`D}&0-;Vl5(U91IR!2F6bA+_Mu?3fKLIxgxmz$w-
z+Ye^Q($GLu7Q+H=8yaA*o>`yft;M+^b6RypaWWqWJldIghQmn2lGOCI(FcK7a=!(F
zuQBC$^{V=g?}Rv^u#j2HN~>eDQAjq^l+n%YR=I3SE!%P^*?TN^S1hTnh{w8tAw6}I#9*w*d#hWgXy)N
z=Y7J*O7XI4D_Eg1nzjNtC&sK42vRzS2zm^%>|1#F^yT*+udar|kiiWs0p0FE3f(B{?NQ)EuS3do~nzkiwxo>%K+NHMnoGP|JLt`!r(@9)%c6(5%MmSBNi|#{H;SvpPwouIBmW^)=DpD%?#gQU
zg%t3n3WeYY@uewX2X;@ECWthz
z8y+uZYxtQgHU>#0@#X&H8@D|pm`^zLoNp=HN3p{mk~oD5ruEvd@!sd@famRy_q
z<~6IU=e17}-t0(`LJO`06>FjIV$$ZfYj{{|%OR7U`d!20W{4L%4VOr;xPax)Uk`Ek
zoEokk>+K$@JU3l&;Lse>8iJ2vvpIDcoE;;muAG&J-FIYc6o<{SA7deJ)r}AlS2EZx
z90;niNU*!Rwn*kaeWcwtF@Lt7ev_I!Z9io5&DW>=ZkJmnS&1BdZ?4vtVszux%a=Jy
z$$j^){e8#qk!dZE4nrfY+zAAT@6QXk6D=v^3Wh*
zT2Q!8eUrgb{l{9pw8-fw?e11%v-U!5rc{zavs63&byQ7+N;4O>BCYUV7~>4RpUF0H
zVts&0Jt>e43W2YD(M6~PSpV!l4~fRt?XsL{&bbB)|DgX*rjEN5WrVJ_{&#f>;Z*sz
zO5LL3qN7wHd!xOemFBmA-{Vk^P27_Jt3sSO+}wGxZnJvbDJz!V;B=5>F)5$&6uXWv
z`ziL6DTBA!r@)Qo0j-k{&^mJ-d_RjF@9=I8Flp*{ig;mQ`C+*xKbN@=Ur-N$wg70U
z`5Xffr8X2dfM56}MiOgRq1zvm)lW+1m7Qc6#PUw_Dn@bP{p)d!-Ll(}H@%bj
z{T;B)o1(sk_RhN`LAE1W4120
zk)02s;id$7ZztS!ER21*9J5E2Rlpj$5+u{rE6~5Uf^64A7}7oOaTKypih?xEKF#;Y
z9=@KmZIVNg
z-89EEYVtN4`YZI{(pm(e={>tYS`o_EPoyX(i(3z$4#FIDeJv+$Z!bUJVKuM6plR{=
zILBruX><=Eq%Y^g!-zXpM#J!F|Me`=*UD6aTBg}eU`7i{1a<{b)uH{{rGQ1R1m)zp
z!v`H0S*(+Q^H9u94`?}+p|Eq7o}9A`8!0<=HbNh%_3h2-CD++-To!;ZO_{yt96Wr>
z!4-d_bhycp5bF87VfpzI$Qb(|VwAnW`W_gV0p=tk#NpJH4!aZU2lK4iv0!+bqh8DK
zU`+&NWmg!ols)%X606$(TLZwUA$6pD_#dw9Vc^Q{Uwg!O;@;`X4ZAFWzRn|(7nd4$cIZfR#+z_#m#yvGeJCa8I?@w9cB?uaz
zTB0P=bZEeqO|M-ot)*vLsB>Fp$ctZZw7b`KefiU3rtVxj4{bX{JLT(#O3b0
zsF=>KRkdvr;s?qB6R4cQ&F%N)J9gVMrlux3hb+lNHZKzK$qhoxZb7C76K%-{qXO;8th@;?L#BA#d_P_Q$YF6YaaB{}|o|rOS3d0KAA7
z`Rrfr@;&u0oKC|;h|dzo5~jK_g~ssPx8V8dE1A+qpEt{QUz+*eP~@T>)?Q?^Hjt=uO~M
z-7e
z->!r?%*4$+Q4px)W=~_RyBl12Nbygb!wF@5RnlnCG$7Cw5pPoaXe!D+!#s(VpW}W8
z%E{8P|Gw&dj{`oLxLoV%g8F+j?1V41VW>m~EVE-41arIk`(gE$Y+rs)&SsQ;^}1m0
zJ=#*J8EiN-yKgX2Z+48|os?p<7a>SY++)moO?LnWROtTx(^_vw%8f99PbT}+EeKm7
zv`;_rhS5oD1b5&Cg0#rMA4LLRanmI*SZCreWqNy`Uk|!6DZ-r{}*3Wt`hQa*Ghd+W;60CtanvEO)qdsbqvry-Nh!Mlex!CH1
zXfl$4K&4oq4NQ0f0s?E|O=nM8L!)xPjWuDy^ll`H)JL7elIr+BMK~1>FR^~%ziY^E
zN{N8j1u6ms6ss73Z#AsqR9se?px+_cg38fQU7NXL6fq(j4^eJ~sM+(%^9r;Dq}X=RVK)CZhC$Y>5k`z*
z;BYAQ@458rl4%Uoi{6|aBkw9^O$kqo4;D&ahH%p5oa8nR-V#u5gs8<%6n)V#*~jGY
z`&uNBRCfX7d036Bb~)G5u-1vkHy#$xc;(+?NUT?<=R=Nt63Db_T*?~$M&EEV?{4<)
zYdJ`>1BxPp?j5m&CJI{aVkJUx@@2$JykfbVRBT1^K|H9;ZEJ;7Et~VzE1JB>u|NFJ
zT+ut3Wx+!{?p=O`&GQ23!=dp-D`yr2JJB>*PRUn@iiYYE<25EmJCpf>wYHg$Fqv&(
zgoBd@9eKpX5qqO(Iy%HB9kQ6sWt>yg^;ll>_$nhLtx_cu!y5)_5siUD{s@v0o{a&
zTnG<;>5|88=WD0$0U-9X>w
zcVQyN(}g6}bQ@Y^v7xR$JC4elnws_pb8#R_h>M3up;Uwms2MG@vp7C{Enw<0)!@ST
z;MRr#TfheJp5byizl4Q_1(yp)7Z--R6NS(dDFP`p49OE#Dtc^gVOP$Ak`LjI1s_z)
zxP-j3LQAasIX79i7+!{IPw8Qv{ay(AfQK(Jk_A^g_~!MifwZ1JU%tnUp>!Q%AHkacmzo_!blkuWV5d#K3@77puM?x5)Y$SJg@q)0xl+4E2k3T)%
zkNOP&%>Vz@k{!qcbLrQiZDi@i7k$p6l)~yjNJ<~qM0JriVw@x28l0aK0%0+vMUE#IYKFFcG`U8K&h6MDDhjh=F|5Voe{mvWFn;X+w7Jk*|z
zS5zm&-j%tp5I?iD*$t_T$Ww|bC!7i%Zp?XNb$WSYO!2C?(nopaABMsfALdGn^_=#|`5wG)n5rWeN09VVCkJD5r{1rWHmAj0^m
zJ}vtM&L=dv$d#j3y>{gC3T}-fq_Q;p%7SXu-3AeJC3fI
z?yo#4X65S}{_qibas%0>9c8=+$2fb81t}W;8~Y;f&6d6yJ+Ddxt(wVI4{c&ZkJ^^|
zFblwv(?8z~QntN443xC
zRPnn6(T;an?vqZ1&1XYKPfQ{?%CLx0-?8iwUM)nZZw3IfLW;PTe*41qFIHnZ)CF_{({|t({!Oz+jkgY>aLGo$(3mN+6SS^v
zdBE=sF%n(4x48cO78|O*Ez_`0M&dPmuDfWY#naH+I}GD@k45kF2;en3zsd)Y(guv7
zX~F)igJLjr$zmeMK1HS2h76{EzHlmvkps=9`Pqe?u0L1-_edM6tth36@ZrjI{aE>+
zo|Psj8|f&}@w7y}yKwlkf$OFOXnoGM{?M%M0fCI~~=F9^xGVGouj{
z6cqMAt@60L9tP=lltDDk*7FBrH%r&TVdQw1{Xs(VLoLBTi;LYW8O?lGRx4~v7ZbN6
zz=!$R6<=A~T=~i4{cc?San==~<9Yp|vdl~3H!gOmYMQ6ecaWg;1{O{Vs{_NZ7)MFR
zTS(gDqiM={nPJrfrQ=oBY({{1
z>;bWShYcDK!jGXd;wI77*8Wwh*L6Zx@P}6QX!K8d#aRM1
zl-kR!J{%U!D4%kCpZ}}u`0X=6VG)sMzyZ7QJCz@#Nk2mlB*BE3LWy=)&~kZsnWLI#
zTsXD%FW2-X#+AL&*QqJxhidG5UA@J-^7(%omvzm}CF&jb*<55Hnq%cgLx#7PP9Zp+
zf8%YIc`YZ4G|Z<;K5kxtP6$RFzCnyyA7B)
z_{(oCD0qJ>^$ZNK)6`h!fA9_r3@oy1IZj?0-P?R8lTNn2J~L^$yHNA;w)2#B{Hvjw
z{N2g!3N@nLsi#GTVHYqB?@lhxC2)WSIM*9iNHCAf39*ij4gmRj1G_)yx!wk&a^UCo
z)@1L%{6pV+@{ukGn|TE|>IH;^!XqQww@0&00Op|69t8yj6w7n~i~V$!1r7izeP&ht
zchy5M-$g_*ATCKHbHxEkbfnTu38?P^z=nZIulocrrv}=fHPL^Y=(h@ikp}D+?0t5i
zvwAYO@8JH9Km=n@MhfmsD!{`}>9+
z-`(2gNA{Wt3Py@R2%T7cuJC{$BB6Ts@F2nmx*w|2n0>+d>Q|k0RLP;!k;}BkPpVP7r+dzq|B@=W{+g&Z@yp?}
z6~NoIglL{LW6g^Et6+Pfl3r5S;c<}fScZJ?z+y*!J&y=6s=CoxXRU%Hi1NK^YPsEO
z^ltBFYhv(Bk4CxX+MDaY%vzgpvCNqhTF``jeF
z=v@{xW>d`gDa^)leZ_z1EHLxO1KAU%qUP>afjUwaxpVFUE`P@G$e?t*yK|*bN9r@%+S;YL;9v7b$%60qV
z?PC{kYGtK<6L})9l|aY$X1;9zz5>yy(<^I;V08ST9lW+$ZLeSI1UM=qVKFYHht#B3S;
z<;xGiwFnqRUGwuvppJ4hPl0NpP)+y`wXZ*kleLcvg@C6WxGebnP7KN1AF3nfl>%4%
z=*^TzKVyHQ<2Pnaua-)(z#`*th8Rr#TE8W5b;85zI4`letxuRYp9>;JcY-X%!+c#r
z52)uPuBorbP|a{Qp{Rfh7$dLvRosy&g333ip}4>sZK=$XY7oJ+wttrxVHX?6$%*1*
zL<$&tapuQK9gu|PjGHEs>(D&M6Mz`9MkyL6>Xwx|&$Qx-@
zQCj;vXFbr0+4uXB_4N>@u}qQuID>8ES$y)C|9^Sesy&agMClSR&cJgbvBG?O{-&dWT;U+(HXuu~p$qzoHs+hQ8G
z)}khthV{nc&>`KVy*n$m7#(SeM^Oq{`?YmmV96a7ixF0lav%1QeRHH^>wP|oD|~^r
z_n`!>#l6>kad309{Na*dA+6*|qNu6WkG_>$0SVQ9#AY3RpKq&dF5fk=TtuaDvyq9+
z*Ub4o2NGIt5~{Q6r_@u&PqeKbkxGca9gL_!6!k%cea)o&
zswSJF$K;D?r0GrSfpS<7E~7Hrk}T*B{T_BBfTWVWx_cK+J9>i|X-AD<(61X3Xocn9W>=^`95>q(oc)NWJYn?ckp-!?siD|SYEs}J
zv0cfsH#3?uk1ZUEYA_)q8!iM3_C+s6?`PSIBi%jDox`qUh5eH0L=RN~yaU
zolo^%!!GwLOU7M!l8Bg`&e#i@^Ukl%P@yiXNWF1!38jbaijGzb!r6-MQ)P~&9lpdAVmZex*h5gf}VAcB7AF-f(ml+V=Fd2MT6K$j9^
z>K9ST&d4!Ar5dETd}mSfxTdDY8YFEWc6zY$0~hgo09qs=B7y~}5s&L_GI`3VqFdHQ
zH*arm{;%s4lNB6vR=h-{?{U9qp=kaT+U$g1kBn1sqjzT&{d5e|&~W;K%7{~9ja1=zLOFhu0#k#i09-o8z~=XcW6%*7v12&6?kiw4k}4soUCk38+8jOhOmCQ%8zo7^#z9=OC7Tb@TS@7##hdg?gKM*p=
zGT+H{m)hCAB^kWbym4|^J7u)rUXq+u^qqb#c9k{mVsnEvT)eX}({h*-)pzW)Qn`BA
zkz}OY%yAE+-di)On?ZoVvVQk?GVPPa$RYPTDmL^l?de7(JHN$>14$Ww5K|iiV;TDP~zKhmz~y@a;Ag1{-1NF^+Nzr^unY#DLY&X&q}hsa%p!yTSQa@fTK!`
zCXhXNt{R8q3KGWOoll@KXc`3YSikY3_1%=(eALvA+bl8{mvX+*oYq-n{0z3E69VKuZyc&Qu%#_giVh67@3B~xUANavN}cXa1N1*3j>@w5{qDOHWf
z`cn8ZBuQG|Sw5Mu@n9$7eV#L$Z0j)mm*n%laMH2ciZ^L(S~O^O0-&bw>V5Y
zURM;90b?>>b8^VRo3yd90a-)YQeD9!t>)mEn2UCY_1@Qjh$OE)TkYWVmXP`;MqF2!
z+K}Z9c|rfVWlM*j02%>mVZueyXow$%Wm>Q7@rvhk^`ilEY7^#EoMv~&;VlVSBaE?6
zVRa@B^60YSRK#!IT(`l=Q%c}4w|oxZjW3`^z1Sa)4)G_nSyoT4PQmHYl@jnXoct)E
z%oNQM9jg|HGv_Nup)dw9`4tDdey3#
zVlJuMnMO`X`2_=xxI{s}LmS+JBI&lL=gM3IcP!Md
zH#R!-v9rqBh6a|_;tGZGpW3qVVHE@4QOWgeI|!dL_s#YvbHIPmvbsG6%lX9or^8F}
zXFoc){8s5`iQ{V~gL#|T?`oktmyDG%zg_Q?gQaGJjm}40z^Y~^gq7S#V<3P|IuR{G
zO=I5%7n%BsoS(?v02fv$-1Z8ayz<6&Dreed@R(F@NMT=#Q|ZwjK@Gzq!*o3-DnvEw
zQvG6oDwJH8zYK?&d*j90g%3_nnf#Oc<8K^zIL9}vlQM=BxwEwhRf88oMz-V|0usiK
zSyLsL)C;!BdAo8c%T!(AQE?sq4t|fUsrG~SY1Fa~6Mi)2!=e=PE*+J*3mGvuIKA7g
zwEJMF%tXo(EmGKj?u+=Xe-`eG!|IMNL7J?bZkzs|8%~(@0rs}E7IxAmUB{dTf@@qO
zfuEm{u7zGuR8;1I^;cN~@t3c$+D?#8bL13)aK-9Q
zIs)fE_u=wb>s*%BKACrJue{p~8X_KGj}Wqmr8tdA@!uYf8@o%P5kMDE=fPS2-HX$gKmX6tj3-Ya8G}pc^QklEn4&h=}58JqfjyEK4(-=I)
z@OX}m#N4FD-_iJ}9qNoIiQieej!VC}zo}?dJ*Z`5x}TajZ5@bkv-y_d*I`A;`h}KRjSDJhJwq22B^VswPD_UMZF#*M5q(~~K|#uj75O0c<|_;z63
zd5C5-=(L%>Y~lltmS-f^(+xtP5%);s%v8W9bQbAUrSK|NDY|W0u{CTG^zCB~lE_(l
zkrr-B4cy00sjRZ(#vHq7_UtDXeb&>8>Cvw+9*rfsksisEv;2obMPRCZ4vH(j#l&E|dNq^Z6iuU6
z1~AQFV6^CQLj%mUo1oURXYxfM8RTcUlLZXA;u`_3=MyyLHg<2BdEyExUU;9I&*N=3
z8)1Kv=9?aMMF30ke3>i~)u`*Kv&W2!Z;I3Az7COs{{NxuEugCE*Kl7@1VutxS^;UK
zK^g(+?vn2AE~UFWmF@;<=?3Wr$wedGtn+8@{e9oL_w0M_Ib-Np1J+!uG3Q$IAMfvd
zo(HZfB;H{{o`PaBI5;XJgJ*QQ-i~VDkzalf
zsCU{gOOEHD*Iww|UnBqyZ1BLm&NE&f6`EZg^=!}0_d^sUVO
zlWU&|^1jeH`mH3qsSxlj(tGW%c*5n7La|
zxY?mnNygr8
z4i;wD!9r9~_KM4JmvNKStiBMz$G?`N9krU7;|E7fTBqwktslPA$OX5XW5&(H?5BYD
z_`$iQShi@kiyv-_%K|ST^xjzueyA%{6MIa8*%bv!TI_Vr_CmPj-WhKAdq%aQ?wEc|
zB-lW@WSj5RH8S(#tyHX0*mp;b1BPcb<`P6aF5fk_)Q9}9e!z{(Pn=H&KaX77iRHJt
zKYE`4otRpET3wd6rVL-W}A4dNymoDS5#AYuIY@c^gcVO-hmjLp?JlI=2V
z_aR~O;%fS6Aq8P&iAA+{aXLm<1Y!r>cZ`TwDY*>&?K{7@-DJ;1k!Lk(pN;kAbjxug
zo?3rx_1ltDue5dN8FZb=>Zu3iVY%nePlpDf5-V&%nM1doV)-+~
zhQZ~05+3(zbB@@#<_VpfQ@qn9<*4@poEN-eD0<9>;ezEEveb=Z+|
zzIOPPWQ#xP{Rw5KS6N0JX79ZB9L*QevE@8vNW)q<=#WbN&!V#v?wY08Q
zB5qu<5>dShq;T4E!S3|V8h&rKVV2oy5Oc}3NG#E)@pQ6$zSn_h-&+N%eJ-}V*#1Y6I#7bG16>fh#G=hoerccYI#4bu#$b=bw%EuznJ%ICq->O5F%;^a{+27n{H%gv9C}2HcV_OJ
z-h>gqFr90+uwUHBrD|V>7eCF<==!&k
zfy6_DO&T4Z?#v%@oqdh&sgr^Tl^WHslhlfgjAXQZP+P}xi6AkZP^j=-{V2Z?H7Pe`
znpA8pw3lgWfXQg{L*9+_TKoBJ+N@n#w@BhQ`J$lkFvIR3Ll}?N12~Ks8KchY;WyDX
z4xPFg8>=`1%9129c_Dq*)_E^OjgZEEYdV&rVqWcYno_+e$A4dlhSD>8oKqFG!HIqp(Lwqtk
z9V^LG1yd_7J<3|S5NljN>Wj04aZ#p`ij41^`^^;_aNk^M
z@BH0tp;6KE;A&yYn5yu4u|lPr{1vM)UD+OlBzmyV4NZ7sXo|8ZuG4kZ0Bqme^PEKt
zy3;BxmO!KN9IB{4N_qt6Hyxq^iD0pf9H_Kgn^W1%-BiQC{ZxZl@e%FrrMa&q3Z$sT
zGa))*)G?kQwhtD1fV9H3jD=Vm4iAAaE%{d+zJHI$*Rzjva<$7d$
z$X;<5Hx+VxXlYO|$;tXP7Yte<@wdhKNm
z*vF$|5cZzEcsX0;%xb5=(f;MFB!0Fh?|1cKaas*N)V2g-G(z0Y0$2sLA7=2Nlg+@$
zh|%pA3)9Q(uLtxf%0{&NTmI11%s%@+YzDjJJz5VJiJM`Biu{ZUSqvp@J%DWTj{)Q=8@OF^ZX`Y
zFTXr;Pa`*MD9)I-p&2xvy+DFgO&r^^RqJ+6Oc)L68sQflAHUk-fC%F
zs$8}5`_JkZ|FY7-47mI=579nfX4Cw)pAVcT-nqAX!OvS~Vuk=ypkH^r+JW!>TkReVdj3CYm4t$3bpQR<&jrxHAJ%5B?YX^#D5Yvi(CA)b>4Ev0>4E5)20S4GXb=s0>ejGVfF=LA|L<{+AhTZCKGQq=>Iws
zS)ri6taZN1!u{o6)z<
zJ|qI^feSa*kA6ChP6uWTD?L41rF*7!E_{hXdP$5_RF7vbTM3LT>rZ8Z1NOLLW{x{N
zem*C&<0$%ROlGb1!nDxM5VQ5OT+I;@9`f!VS@}9Z4ql=9EIlX`=zn)GoM__~*h0lQ
zNh?-M?U793!2>;UL@VAMi~SN}V$zJ1kT#tsmiOwhugVW@ydPMCOi8yG>~}DC-KNIl
z?QQMJshLu&=x%Z2A-(x%77xd0$<(ZyX{Bnv$~ct~8?=)7veDISkBRzm#U;tmx2_bk
zt5Qp~lBW_6F$fYnwhqt329C*Zr)#ful=*V-$z^Yd&(LYt
zTOkBSdgv+=#SS>iT_FpR#AYKRlimLj7G=LXgfWn}w%ru?d1awfNhi
zGHn1T;gNjpm}AtZ31)opsQG%2_tCJt2{Jq0tSOM&>CE#R(%q8+JXzb+K%<96$^eg*h9{?v|HTbA6pFSEH#>2
zT9O%bpRBIV$GQU$4^qH$!Fc1w$H6m}KupF~rz)NF@g$^yENmx|nj13|mh^leDUK>q
zEm4b1idbUJuf0G(kl~0K6Q)qamZe{eOqwX~wm(w1xFwrVf^9^1JSy42;_mI+kMzk*
z#*z_+=F`O^HPQCl%Z+?X8kffNK^!o&nJv=n)^}MjOiWDR)B7gps;9u5@UZohu`+)Deo4C6DGY`pHH9<`qTBfqK*eToE039zt;)v#$mv!p=
zK!mI@d?ZB?g@*Ytc<%neHaX@c!Z&d-$mLe-j|pA!i&8WFl|1PTrf%|lM(C_e$<9iO
z0xZuThZ>{x`kC{At0}79%KpzBYz;J-QrlsV1^GsnjUn(F|Z&W_G-V6lju&Y7EXeTMi@sOY>5v@a{oen-zCX)2KCsd7cTGkGGL_eUsU
z?D@do)MY2r1@HCHuNTZ5&O7UxNNnl1rxu8W^x_4knWG@_wvu2G0nytG`obOFB;%nqS_`p(XU^Bq#7+
z$x0jKY|7qdg45a~jK-!9{oYGCenCs>*sVGlO)uv_preRQ_|hEnb8nQF3!N!G#kYvi
zNjANeH*{KG!u0V)SFrhS>R?)gi^Uq$lOi<(y8212)wEVJ(&;Yjp9G~h_ynOMe0ReQ
zIY?S{Gc+K^>~*>m{Z1iGQ3ZV-!^;u0E`HD;r&Y=AG(rA~S)UipGvGTCO1sfn6gQ;G
zp(1Z|S8kuMCBv{Z|J5l;j5G&<_xs$?F;iu2CnVJovlZO56_1#2TzN1ycB{O~w<9{W
zM@Sc8suIe=*a-PngZ6Abvo(@i+RdZ87gJYN+JfAc!*geY7%fbuYScX*=Uh+C6Mu^y
z!6TKcgv>#XHU}@Ddh0r!NmCfA%x2FpG{WDKd`N3$M!K9DR~r1y-u%+OQOdYz6t3fm
z;0}*c6O?st`pG2|k>q*u&l)F;!o@6Vt=8}pR%0*s6`~Qd$pSPGS@jMi=z-3SkxV`r
zVAACqgJmtQ7-d+r4@No4m&pID3vuMfyBOhRk%3-$=WVIZQ&;I8UeDiBe)3}@?
zj#_Uc;F0kdoc2{e;{xM2z=sCX9{mN;Juc@V%HF#eN3Y;
zdePcKT;$BPYNmsk;e^dXSK2~tE;k*oKCXP&ccbfXbRv?nAcMhOAtE_l
zdy!qW%@kkyWp`77rE>mW?0Uh~fOYvQLefAqG+t=A3$?nkSoSO>bpl7@?0qKPj%1!SsnN=eLo0ZmLslEg1(-Xj~YrFT9DSIOnZ*
z^I&xFXNOauv|z$`QH|m~>d*D1rbzy4%*ahAU6}6Pp{Rgrm=)(+T5Nm#mzc_1ZuKNxTCtu8S2)UmWQWYC29IX0Fd7AySgXK{;Fh9voVM|WltqyCGnEu*v>JK|t5F^lD9
zNcyShL7Hr@p*Ka!l?mBaf{sxcyXMc5OI0Rsj!;e(S#M8cRK;(jjQG3=g}yYqzxbJ2
zXYWTi&Uh!L(WC?}9&*IOKI$34BHAz6zy{ay6hWN(?+#z3Y%^?x-^-pMzwu1Hn0D%U
zFmxS=!bmQDSroKy-$>1bSQw9mF7Xn)MoHDLOczmy%a+BI@BWGSyDC)f{+mLCF0?Zqts^-iNK64-fK%pQQE|39naU
zn^DbgBwXw`%an~8O>tnP3if^2=d
zZ_+wSy*x&Jb#;Bc3W4OaTVGloxMr)@nsI|>fc&Yj6Roo!UeNnX&_AP4q(}uOmh3&p
z`}Ih^0+mr4i*?Xe$;dkOARieP#`2o_&-PgM*B8|Nxs8(nkynQIH_XTPT(M493MDut
zz&|tWXe6r99fPv^R}NXM*Y4{)E!QL57*YG>;3>0v;ix_+wuOes?~U!Gu;z
zdV3#Ec8NsW>O4y^u{=_nj-L=c_#8dU0cp1K>MJd?n;m^;Wb}bHHL~Off~~}i2_!j5
zXSwFB?cfaBoHyL=?1z10so_9Xj+sRA#Pv~t+p;ZYV;JLwOHA0gFVY)2EI{{(03vPDQoC_36;zigu
z>uc<)VHsb$2Jl=7=@myavk=|RWxohLpPHa~ch+zx1vz-|^7+q;W!70r}&bhX0@8{SATnlm2;bnoSn#*SAq
z#_B)J|MWDbVHW39l^b3n@=4fSRCPE+aqh&p;vq};Ig5$1>>{n_Mh`rVWW^>fH)|95%T5my;b=(EV%p75(GGWWeGxW1{x>{hZ2A{G
zGh{EZ0L(31hJ3HK@g_9AiPZPv3gdd?9hW=3aB{SNY3kG5U8SxD2gVc>%gBaOETd^`
zq-!#t7GD+I_#;my?`0NFr!)AU^F1N%&eiT1+}w*n;o%gltdPKOZWsGq8&s!fTjvK8
z=?nEo_K?G;{O-$x@_5co@`O14B?ixyYP!H^DUNPnWwlSBuj&)^MvH+VGV
z!p;&4`b3EBM@~iLQ=(0;!)sM2r{<81{yo~&=(0|%gz#4mxuCKWdGH=W(vNt4*S*@~
ze)Jawr%o#CmWpW6Ny@{#`KcecvZY0ZV#~Y_S!^BKZ4cz*zI+r4Pkw?s&*fFL{kkfH
zSA-BBfq!79`h&HGzGA{fth@aA)UmOVq4cbIZzi|`RLA`WTQ+x~?SRQc1_n(=t19=B
zq-9|-0cAR_KrMlXbOtY;irs1&@lePBx|!X#fnd>}~isN`kO;(@Y#Mh^jyM#Z+JS=1TYC4*~+aNX&XHD0^fqivQ4sSd4b!&#&*D
zRECDBvf)W+(CpTq>K)xJfnFr=@6Kio9jxy^RNt@HZG_iAs6hNg}BP5CNY^;$_ySH<_c&v
zln=&vB&T|vDW~=nH+|*1VYgW;@DoSyc+u=US=I~w?h4b7!FFS)X$6U
zFC%Z^k(ZsPDsq_X1=pX4OrR)yu%$`DHazbgXQSGbG||Jm?5w3|OQqN2f1`x&S$Zapk7=H(Hx&tgAz
zVyk5kZ*qO04_?CMZzmhFk8<;FA8Zfa0+kp8;6o`NZ5&8g#dl}cuNOw=TvwezlTWfZ
z7Px)#wQ!xYtJ1W+zaqCXKLQLo;NbNUP|bLUr$39W@ee(nJ9P-{H6(-y(|ebyXE<2A
zmg?WinI>sY3RbxuAJ5w(8uhJ{
zY#^sBU0^q)GN&wse?H<_RQz?3GiySLhDwxD*-PyC3{SOi;R4@q{G+nLvkh$W?!mXV
zzYb-pZk8#DTWz_Nv)gMrbv~mOm$agSN)=1+HGAH<)K1&-&l?JOG(oDb*?N(%&3cA-hDWltbx%`#nl|*b
zc)i>fFwWuCtLlncgS)ZoMc!Nd+a7DC*QsiaR?og#^9@(Fr{W-oa@kREf5~z=unEBc
z;?nPNUZXDS=n?bw9yuUrIhB4rO7v%E&(QDF<;RuApUd>Hwj%>0#}`?erHDV`!Ft~F
zrf&V_dIf2^(%x2QJi%0T!TfTE{BsJ%-M^4;#)_wKHp=|*Ejbg$uGm6dk)s<9fO0Rk$fXw{5rt0k$}f7E*OJ?$#4J@z-`fRym^&7uwsbIP2Iz(DId11wlJI8-bg2<*w<8+B%E_H8g}!L>}bfCTmM5L$Ki
z)Wg!-(~VW;gkkl;>8Iu(?Lo-z(;A;y*`d97+>K8n{mHcN^WQp!I=Z)bSDB4s3F6`T
z>F*>w%x&noGyIK@Y%F9I(n^L@S8`DX-!HzSlL#&hj&n@k%ZkEUyiCE0?TZ!DD6DYD
zLs!7v`LfELT!A}?XGqy(#8Tg!zo`!Y#$ZBhz%p5McyJF$%ev?F$yVHM*fl)45NPbN
zFDwF}?GgrsiY}b96*o)n6y?S;(=S+4*g_UgB+sPx(DlD!LQ4&m`iC-J39)c?f|YBi
zoT4I9r-PT{qFK2@IjKoYBk_IN
zv^JRM&|}nnCi>1qk5r7WroVuelPcICZ82gkn@UvDpg^Cr@5e(j5OA>T7hih07clMwC^6i`}yR^_%G*E$(l*%cw)aKD1
zJq86_i_?{^$%6?{j1PK_Xgs)q3=>zToG?byyLQBt95{1iZ#QU*P4GEHy^w^BDPfAV
zMn8}5b4}p4(17aIHRVUOTiN2)NAIQSbV
z-KKH7##~*w)LG04I5<3ds(~iq8?ZSB8{`)+UVH;22!tXDA2IJu754)7lVuJG3lter
z?7F;}lj&QvS5~yF9*xZv_q^=h841JIwN5OLTx?6`?*8cI{b0g!-$Lp3SG}FF{l_2m
zgr{l2vpIFx{)BHuY$3qg=~tVDqu<~P&P^g+=B_NjZ7eoY4=pH+Z3L*)cW>dGC7fxa
zMoVvKj|fen0bp|F@tptTl$4!K`Y*8PebP_CQMLzciUsewY9i&gcdQ{3lKX?Hl&rIpeGM|G4};KeqW#zWf_Z
z*uRrQg09N>(*C~yWIbX~z@EAAgRTl7f{Ut3g@7H}5ZXVIMAgk-1r?+l1!|+=kEk*G
zMCm2hS6GQiqsNOvMeqH1iPWcFYk?Q$6k;H92}Cqwb%R5`sJ#<>Jn{%D`teASMd$)D
z%Uk2WbxuNeGb_|@{)UJDQXu^e5dR;dCt`2>gnzV2@c)p-{|+Suno<4}KthQ==}(vy
z75ta`qd@dOr1m*7D7^eZF{>{U%=CfnliGyiK{9avX*3=)fi&>HBNX%pE1}SpO?(2?
z8`BJ?`zx}v;l!ipLV1OeZXo$G)6b3Ci@#=O`T|x3v1!r%!9pEr7p#SvLqt9=QMIce
ziMFk6`?Q##w*Cuv$ZgS|i0GOPEH$CxvKzsRiS@e~^6pN=SFQ){=2=>l
z9RIP&*1PTz`Xv1H_o`f2=!&;AX*o()PuC{L%pR|aQY<_(Z&#P1^pLMk>m@X)grIe_
ztwxjm>#WGz#rbQg)~!F5D=F_JjIv}Yf7nA{!MMm;Q0#T@#$dYke5s~1oHZk(n^Lam
zeJVR$wq$ZlqhYQ%Hl#PRY^-)cC;LQ|r8u%Pr(JZzj)|C$!fx4(g$kg^63)5~D;t@A
z1E&@beB8^2NxDRJjkQ*5WI6kpo#D~Y{{^Zc?G*x`)~YPSRyXb9hMBFn;|4T;r%W5z
z-@dIjZthJV{QeV7fAUV-3M{Pb6UwqJcF>UyC$jnL&l3Ym@fqxT7&Z_&WQe)-yC+$J
zN_}6u=;&^Qy&fzWQm*2)-L$UjNO7WSjrU5eB
zyu1EQ6`Ei2=3Ss<>{euMxV8lP8anKAH@u+|xtCG>hH$j$Lc{Q^Fjt%7tZa145$cwA
z!xM;f=HqNx?4a1UimLtR`s}m8h7C%=TrT6Vu+yVaKzu^~&a+z=R|p5W9mX9t%6Een
zL>TUe2ke#ie15`~9z4lzuDYxqAJ7{JHmi-v%GCYImD)0fnp-Njr(VKHxp9i6vS)jo
zJ>Re^>5c5Iak6AFH8r?9c}bP&h3I;K^0@LN27>!n6{;OtdD0@OIaM@1FQ2x1P^UMip$$>!Qwo
zGs2YJ)|1Ln{-Ugg5xu`45krXFlwkYzyE#$t8K`^}sMk`1V&!9e%OfFwfkFO?u+*|o
zu|W2t3hd2>^HSb1JN4+Un(MlGJennW=I7F&n=OH<@WM|n&
zA-nmwrrKbA)BO>M#f9;D_d*)tRcOGhK)ev-mPpzq>)BWJfu3RHWvcpD&h%
z*AHuljJ6M)92J%j@taFr!E|-Y=SUvy9iLzUT_V2uN{-Amo}(+G&T=4NAcQ%QNBfIa
zoe$h%)hmSEOG|X4$;$>?I#PqRog936q$289v9Ae&XNI&m5^8DoB37c4;|^!vtk?qw
z-9N{Njvlk-a7smN>({4m66n$mWeg}|Hm?(tz5xU=VkNQ4GTlkHw+`Y!*Q8s
zjp;LU9J5UJ>3sIWv{}~GAN}C{8O^3EGzZ-iPIQJAqeXM(+R$$zk9<=DU;3(}>Z>-5
zeQ5mf%Tb47ota~v(YsR;c&mJKpJ;7HLNr|5m*Yq%)vBki7N*N8zkGG$hN?SV@C@b*
zZ=8@zgv+h=^-qyfc)1oUrDm>L7S7mQUsfLF$c^eTeAO8@yh?f!6ZV$+j~w~q-6nj+
zsF_Hr{YEQln#*_Z?wuszvsQiTR5?|v$Ba2s=ZCX|(oYnRwhbM~gG=e^Ch}t$tce((
z!e=2Ocbj!RZ($rQ7+T$2hQJDyT#=n
z>^7dSc$^}GZ3JzUk*jnz^dbg8*>BL54_{+wr2Mhj<~
zV|u@WFgg}iCxtueQy?S|2#nx8+h*G+-NP+*sWuZ@c9Ry+frGh3)y%O-Ors|$Lb2%V
z)aN_dA1c4j=v-*x03?5{KNbR_)f+N-i5)%nIAI_3gTq3ME_)Hqw+1)uKpxD`IMY+G
zpkt+u{FFVPv*IqKT;X*8zm+Pj&HgP_3Ip}=BevlNTjr(=r^ESCk(L!LnTLsYd%
z;%DM*TSt@m9>E~>5a!SlODx^X`n}3T0~v?u9S6e)-F&o*(OJ@@T?bSwQ5BBGMv-v)iYa;mtK@
z(6!DtaI-ZhcKsH;ge9};5FX~+Urz4-2L#6c8wB$A0ZA9ff-VXBDuidn#%0k6%$C>}
z%1LvoQFxhp4&jAfVE1)g^GR1u-(;1~OhO!6~7#wLH-kEt!tuQ`vq{>?}pG4IAyDv#!9
zhBtzrun?D=bkJVSf1tP=!%7z
z8YZ4?G=5y{adrGDA+yQf-*j0d%Ebknl#EQ0{r<%8LtWjps)oj<>T>&7hAl<_Bk7x4
zGj49~7%xxZ)9uw*Zb*M{ZWC$$w>jkEjS5vvHtat8D_-x6CP2@%sG`b)G!Uyu#qSXz^qz&uJ{V`0i$AYoPV;!oh;eoAIdj2
zR!&^1?%WIRI{jWe_O!Lk6dp&U^!d$26I#mmha*CM!aI4{9p5XeDmyv`I`+Bi6Z0lc{BCE7Q%#+^QuWZIR!js4xOa2lxo
zURC;;DRUAR`%(2_>}KO8b#91lweZPiEoql<_wj9cJrG~+9r6pPUU0l&q{U_ic+NWQ
z>6+koSBGoVC5J$cdl#F3(K#5i+>p8ncjkjYWHJ$oS62&yVecGmKVrq-FBN8gk5X5A
zTiI7e;pp-qd8TY9ue3Mf-Fl)ne<9i4gXfTXosgat3KOnTsfIaz?R8)Gl0UFkS#q$t
zO^hujg*KgVaVdKYy*XN4yj^*!>=DDqQ$}7kQb1C*Cx&%hl%d5XQ`kWw+7`1kr^D_F
zAez>*TyX6QoW6Dxbg*3IdhbpmS9Cyb>8})`b(i!DXXPh@|gV+@!EjQOyOTGMq(()rea5pxwy_}
zH~=)oQM2;h;f|zT)P7*9!ID-+#P+;x0^PqP@f*$Ng^EQ45LBW&v=xtg!jB){x-MVv
z#V=eI_HW+OQZ}4lZ7QpD2A;1MPe5KEdFWPHtaYu@YP2$zf3eFbXB#4ARQcWL#CEcF
zC^W2{S%2_gS(4UPzPM{T`>Qq0D?8)jkf$q->a=fYi0tj#=x}`Ji3Ww+C;b{pb_@B-
z+$xT^voGT1yV*sM=^c0%-25PsPv;({jHZdzhL`X!`v+Ja4rLE-8@%#<;pj^45#EFjkLp^pGm~tLvqu4j#H9#n{;5wiHOL`w2=7+Ry!%5KCN77#Ja6D
z?{9QQ4VO3lVMVmzMBA!ntxMmL%%(%$Ufwz=$Q7|B*O9K*eX6^y@_vPo!^0WO=md$Q
zf?GS%`!TFAIZ&7Q*=w2_eLd!C(YuXTTXZQKJTq65#2Q)qnwa?m7eH{h~774^6WJpNq1g#0{
z(XD1V7Eb>J)b;NC|5ozjJQta7Tl-UWB<08V72~KC+q|!@=D0fUgv7#O*IS-y3
z(%#%S3#=*F)Ujo!g(6EU&$}Hcw!o3{-hyI$HAM2eC&B?8ow+&9;fi~X#@6AXxTXbn
zVIc?K)nus%!jn8Y7PF?K?}w9zac5=7K85idjFCV+nP7F1(@T-MU%_Yb#;|%%S-HgRF|Gb#-q}b8B8+PS}VLuz}1o0`Oqv<
zDp*J$Vf?CR_SRs#<|E66>2sgl_-8W2yfbw6cJ?HU@1xIZ0EU~tvV~ypz7W9Tfo4$_^0QI=bZ-{Ke9AF1S}#oif5VsJY3J7t>9-weP@obBfOX{g#jb
zO7eg*wcs_i;q_J*#}aq2&{z4w9N`wTuI?_3yZvWi$YU01YxmyCaUPg70y3}`PJ|>i
zwO0?6s`W!X!(S$^465utAj~#$G1-i=kYxH*HC>{|l?S-Q7bs-J%?f?b49r$8`r*1J
z!b$h~N;+99KO#&>t1!k&eRso;lM!wFd{1n5?dwvNVZm3BVZtKD#R3yGKO5zCM%lkm
zE69gQr?DGAxY1o<>T9x=O(<@tYuL`*azDt?dtWS>J04f68DO93ou7I+5V9g+{#NGfNd$f9`_24N7?7*kaL*Kx
zrFsxMMe%T%hP@fO6K^Wt6aULk@i5Rdv#$-802%#HMRixxcLYA3;5WQV?%!L_DC4LP
zorDe58V!68KNe5ZH7S&P(JjRu3I?2S?v>MT{K!a8M+U(tzh}&$%-|co)-ylKF=oX^
zF+aRz0D1OfnmW`{Oa+Ge(p>8I_196D3Y}HFrB220)SIy^FUx@hx@SN8(C)8y&#nn(
zs;jGWEs|sI?9Kz)tBj=Z?XGW|QH@1!H_1cru%f+xqxeXbqvU9gTO&zWme@iakg0II
z$j7~1zp$voV!v0$mPKz~{NSnR7PiqD-m>pet4lou!KrLH+x@4RO<=TQ
zse=DnE;sBpZ`Mn`H0EZ}>4GUq4Q{y^;QOe9ldYxKvN$$~kB3g@1H?Fl^fkDYt*=`p|%`?+EDn!d6PmYo5fJD*B_?-;%OM
z1B<_cxduj=ay+PydC&P?9}ST73EnQLIqVAcVBN2)>sf|(_JkYUongEP#ohs+EJQfG
zBDv_mKOreqf3BKh2gi>0vNbev#-j+Fh~LVGqRFzHg|^0v%k_MM3;J61;hazF4m>5F
z5uDcgeT{|uKjFuDs_e{I>8RH-Rjthn-nS185g`as%Z4z@ACH)ClmP$pSfFRNgCo_7#BRKzN%1==|d5
z;Tbbw0j+p`;^m)9xAMD^;Oefe^>#0Iz(!37&cE951Q_nA&mOgH4|VQcax`py
zcp35jQ>h{vKJMTy+91r8!x$0hjvfgJ)d0>Ciw$HY2U?Hv)d$9&f`YSgS5
z7YVe-#e*#sAF#ddxq1U?JAm)(IZv~y(;>ELvbdxndCA@~jQ^0|)@BIc+&#a1`M!m5T9F>*?~vIwR&AX+FmU{ps^)
zK=j7h0kStfN1SCY3LPLzdX3qi+s_4KQ>GQNRUBaZJ>6Djt1Yyk;B^3UX2%$569a4f
z>AvYJ-vCLtEa`*Y{l!AdjVEz*3NaNGK?Hk67IX5~Op^jwsr-{^wLi)-LxGZR@6+pO
zW|iy$c@&6$?28>n_|#>>Xe|4lDo1@<4P(9`bWR7(UAT4P1lQEZ_lZmzi0#&>6|F4jTI;z>9ikCcNQ$e}XWyZt64i~tb(N9#=I-vUYLjt0YhP;#5?H3|^A
z`EYfBqLFaG!l@uGNs@Vclyh_uoYZI71^>y$BBU;VJX@Wrv6Ns0R7F^}XW+S6t#JU;chK27>_fC$XDzOWP7TWz9cGn=lhi)l8XFZjQV?%M|&~Q
zNNWCwvsbqmTo$jf)452&ntH5;DkN_X^}bswSzx|?g``~ZMXwAmEC$B-NZQ-=K;kf9
zM{rEnUfCQumyvbAqWFG!u*5jWxo))|bc(L6=;=9W+~*l~t5*xH*7FBp6Z7$0)d
zxP6DRUzZ^``2@dsTyL+gKfR8TqEtnf($X4G;Rkg|lbJF?!+|*18sr)Acxpy-0B831
zAN4wsM5gXkZC7hAL3L74IpG5Fhv`Hf92}gE+0CI0!Ue7Fo$AEbDsQ=VmX6a7$D!NH
zfhZ-Eo;LZCH{qc_Gy%8Zrq-?nq@&;r($Yq+_*`BKZ$*}dveazjZvI^F|Lr}xd*Ze92H_|9r~Ztcyr+n>{%Sv?Fjy#53;Jbc|-In%oM
z;pXOM2h4gk6QNTk&$j1lX+Af#-p7CVm$K}+z}ThM->NdfPo~^vA7A~g76QsD8&5uj
zy8l=RcHsYG0w)*upvwR870-XN&cBX#}p(bcm4Ftx#vyAK7#ebUv
zLT!VCgZ=#cUf+?CleapyXnS2hqo=0_A3;S$MV944T^f(Oa{V3X*U!n%SEx3Uf@yTz
z%d5Zp9Q>DZ?2})kPtT9Th5ATfK=FErnB9tCy*F~yv8At1G%Y>dapL{^_lvEbTr-;>
zYzMd)zfM>9+nq1y>y3RraGe}FM&+EFoo(ikLGkucB8k3R-#q%;!|!3;!ouRAE;B6Q$II?4?KwhYN{*3@T&FuXYgR-
zWa^s
zdgNX9_seaPFusumh~S5He?p(jBh*(cW|Kic^^Sp$AKK8s0ai}N$7|hCFOTL*yyO2ta&Z43xq@qJ
z=Z28~8cF6tnD9s1%MH7KE0gG7ylgJzns=%~zVCeC4mZAUK;lwtM3yAJgW$+a(W&;G
zM6S_huKOK58H!I&Zd&qW2?_alv-7Qrxmv0ftY=j6zzf&cU9CF{VlvXqlrwhxTn3J@Zzw>{ZuxWc1l`{V&U3w_HG_oPo&-{y-dydx91aoA*q11;u!u@g@umM1Gm{i4SfJef9KPb}iGeGSY9a
z$UgmmTcKe6(8N@ziQ~V*PsYSi^oxHS)!NUYlN+YWOe^U;Q^4+L;M}PhOUEQckRdud
zIF#(*C#UFerf0M69(-6DBT|FF$OOWuDx)KEVSiymg7thIV>P>U{^xFK7*H;j)Z~ts
z4>ylVi0PqU`K*65?`gvF?2<-2dyK(&hjSp|efw`-9z+W+@9X0pOSVJk*CE<>n-Xpn
zHh26bv_t0u1f2H4o!Z(xl5g|lj2Vq0_kBxibFExm6lw3DDBF0fYwlZ?oW7g6ZPcE!
zR*GzAh9_l@KfPCL-;}A5#lc*V$NI0S~?Cyu_2BOmNPiP
zk`bJ=L1!2mZ%H9;;3^PTYhj2cuN+Mw;f9QKGEgaCNeOcnA+Yx~+hCsPLd%wlP&}W&
z100BP#t?M?n<_TA$~|S3MbCRb50rn8{RzEcrDQ#1Z{o#q8%^NiTZ*3iX&Yq!P;oZF
z4WSlYTP+d5Mn@L_?tqWZIe>-rjE+WvAp1(w3G&a$y^+KUKxqm_$^3dN=)?W-#l<5l
zbJ(AawpwaD?nLFoRGprkMB2)18Qztyy|xpz=eDGqJ)0o?k8Z~+{mZ$yy#YT)fP1|D
z#f&$M&Bi7FnmjWHV*yn!Jlm?W2?2!If0H9I-WSISt%1=AglH(uc&#nT}#~TCzWVuwSJNk_H1~a&vfWIKw4SVUUEb>e(|2daEEe|
zah388ep7RLj__0B+duP}pW^;k%@@mqwPD-HNP9x^-d!@Y9;|07G6@T1m|S(hruuSq
z9D@ELv@BIbw3IaPS4ty<9PlZ#qt%GgW-cRZ47%3S;ZO#OH}}OAYyIb(zquXj
z?X*MfV^U1c#KyPm&|G>OQ|v>hG%c%t{DTb1k}>({kDdPjezK+OW#mC{JTsCxO3K?o
zF1BoyX;oK5BXgghTO4DkMdVI{pC`&+Q_fuum*;hqV+W!@pYQfosko3j4`r34Nza*h
zld@g(G|z3&p?~~OnJ-8yU^reOAO23Z8Bf?{{t6h5ZgTt30+lv&b_{E6K#`TUWRK`W
zE?C`|ryQR3{iXy=C{TarvpVyYaydM2SO7WQ1lU4(JZ~_*fCa)B2vdgQa~hN#M3PB8
z<*-eMlJx=Z7%a7(N{)RTu2(eWS%CSHO22gf*|x?Sbc^S!f@vBDy7vfjT8N3U&d<_C
zaj;ZxCZDH!FOC&$t0U%%2nlIo_Tp&FB=#e-pM@5%6wChE&lQ37hd#P*J7w8km`K+?
zS~MLnXZ#9?r=To@~gPsargyjhOgSZ?NzrXEdNjDhuG2`B0`y
zEC2U^&)~B^b4^^w-aI55^);!fhm8WqHX+OT$q9u7opX~baV>=>WaHI;unX$u6n`H~
z@~I3)*%CD?@V7C}c@|_bgP;(1Wre5H;t7d&)5i#xH_3Ewt|NiLaD2C^={`7K?RLTz
zjzK=NQeTIPxf0yiZ6Y&Qp8;0@j_Qzee2)y&5$;0laPIzFDkge4S83`gF@iga#5{SZ
z3aQmq;d)9Mx@hSo9bnDP$+EO>}doAJXlRGjf
zy@N-D!;)f0+QRor?_I{FYiJir*jt6pPqgoL-I*Cs?WZDwbvL)31+jF3@Gfdld{)!Rwq92%w*0M+EZC*EF?Nm>aOqC%b9Z+4Y#JJp
z#D=T&cMu}06mC8}M{qW)?3+}(eu6M&e4W!aMM=(3iHqfoRf(sHZ|O}-33A4vrO~b7
z;1$38Q;GcjBwAdxgnSoEZluGETeG4hMO2mMyQjEKpvEC^$XzSj)=Xp~^=a|@Fd?$w
zMqhdfs}}30GCfnaRSn9mO?yR{H|X?#E_pWJEK;yW{Cv+YY(gz;LoH4o7ZDQ?vt6;k$x$6cyt-tY1gHhtDj->041v>cvL3W
zuZy_0T;M{Cx4Br`O@yZvWl>;kE$%u}LW_SES=RcVQlfVt!;5|)b2!2>_~yjh?Yy#D
z%Gv^>vOkW;yZkTMU?U1zy8&ZmxxlfQ4j>JlUY!s%>|8kB93TCbQ7&y+zK2#jS@pi*
z?db}i{fjO;h(?MNRiAo3M+J
zvlaHB8aCV4vYO8NHM$Ow#jNuH+HK!8?rzxWbb_KgWiC6u;n)M`2L}y){&hCrUXc^G
zl>B-$4B<;^Li}Bow57=)=7#|~CdXdeqG{W~U1R0K5b8K0hhn0gq;qtc%Ysz=-Tmz1
zv{Ek$B@Mgj9yEy|r|@|`vVX{cQVk~?!HY1jDYvdmJt;U
z)AFHinoGuzvA8#YZmLoZ=Ph|jYqo?eDaC^MV0m?o2tCa&amJ`Q
zm=HuArHp~ySmK>(ygz94dQkJ9+8r(jtU||+&wvd%HOegtJL%UE7c|ZW>kArgOD>ed
z`zqnYBLU0t8e=|3mnc7E||dO8k_AQ;WZNTp?D9Pem>
zQjn0K5!!#!H#nEft$z%}QxNzVSdZ_oEF7Jc6qJCYDQMn(W(lHIqFaZrBE!W%zUN{y
zC)%z|$c8vu>5ZJKE|U$dX>MwT3aQyCS*lJSV$~gm
zhx+X5ddB#Mj^L|bF&W6Ah;3YP{}nVL-3AFG<}!tz_ufu4IKO6v1SQ<13hK>?s4X<+
z5=7$&pI+D;PReK411sjaoBTj78TH6d6FMnKFK^y0kql<4&Dy)d@G(WN8&F!HzE3GE
z_Ra*NT^)vo#$Ww3E*AZxCqC_}(#f$Ol0$WLP*j|cP&I$AJej|((nF4)BVss&g;H17
zcy&c7Upvq{81NVHkmJF_%`Nl2TG{>{rjY|@j!BLcccb66Fs5R)mkN)R`x|g|wo*wiGw^lffb?>dQc>4Buul
z3LBWwpTR-830>LR>X+@p2jZmivy-t&@5(ocPsWCnp{O6hV`{(Oz
z1?W8L*#wR`b`=@t8!_s!4kR0{9NBPHKKlYvIqxAn
zzQK;-?^xmg&gBsOE;Aj>-2-Y5lyj)_UgH&GYBtYH*M~^@`}?cwI|f)t@E(Wir7gRh
z+J(+s`>V7FsH^jRflM~t0el`TM?_{fN+!z-LJCTjZ)gW4`24rf
zbdc!@8gv}I;Uibbl|#P}*wN3ro+0fgP0@^S3muMoReSG0yC|LH3pRn6^$#7^tm%%W
zjQo=4Iy>Fiy)J}>NFOC--PRW7B62hAeUxsrhdRrjm#ZCUG{sB2VPm{0eqPXB`+ooZ
z%K$@07DoM^W)JzLX@Ci-F>W|mMsqi%CAWQRy`~Ihblx1HQXqBHvhmTvzbT(6fggAJ
zJx{~IEAeUb?!7sVxr&Pg@&!pu5}x-+X)XJ8UJvX(PJ7z;@l~$W5Bn4m7;z#|^n>tT
z7}5?NXv2qmnF`;#78>2)&=cUDnjxoGm*TK#$9&hNttx+-E$W{47%_o{m3R(&Vz5L6
zlT5w`4tIgO&A}v2w==U)VgDc04h*?|M
z)_n9C6O_@9K9jMpP%bNRB#70klhTYuP(r7&Y*Sl{a)no(It{ZJDL{
z&)w{(M{UFr8BNQ5$Gz6&FZ~x$&o=GH-@%+TxzVYw?$wH)bU2d%T{3UE6uIU=R@p`?
zuFo2>JGDG-ierBLSujgN_U-+M*ueLJ7w9-3*gl3*BCOd&YeFP5!b<9p*>H=O;2f$h
zt^Q0AK5^#57^<}!gzHh+Lz<^Qx5(hQFX=4Q)2|9+cOvLF$wC}GcT8Y_$YyO|$aqs)oFn|&ZN
zIu}p4u{N{x(P6>@xtb=~BJ^L~SfQ@b&>UW+Lr$;xQvNoJU-!i2XXjD4H
zx+>f{n4tLnLr&&fFbDpXEp@YI0IGha<>g;FZVrIV
z%$}MnfS-+kb_>~5p5AXTh%Zr5yMOz%RU=h(u@;=9{e
zPg#$Id4!AK3)s9Z^7oz@-m_btC!q4k3&&&qzOZmj(2w8uW36dxD}#~m$~!p3b;nb!
zjLRpf1Q)2`eAnwy`2E~mTn>;aIw%x;j&MVqzv^myAHyCXyd{U7S}=d-VqUBqVz~V?
z^8R)-Z1%8*Y8vpL2G1Z_5?p9jhwwMg4@7AE3;(MW=-&x_loN4>t#JQwIxa+~`{-^v9Qx^Qw+eXux|BFW01AZ5|95ZHw||Y;DTzBEW&gbGjFr&990aRpErp8B?Lhs~5<*ohhy>p9m))3~G?wf`V!F3V=lL?eBk_lVdlbTaGS^`8y
zISxXEu;sQl&SOzeK^pv#P|SM|FpLm^5ij%ZLIF2Er~dsQPNvf!V_FOTko7v!%+tH8
zBbcP&B~qpQpW2LYn4fHeV-oUhu6L5;IF3$v7e6_;kw?H&B`3`yN-=Zxw>EV(hf_}A
z=BX|uBFz%^uhtZ9Ci!3%R>4=8{QI%rubyXe);#)~_Fo9F#M@W)UsD-0{EV{8E&N
zWpbYzj(?^M+ied=A%R{zTp&L>vBX)g$@)l}p<&rmaKIfj2
zxA{!Xy9Y?_Xt9Hfk1@$s*sshi?o(KX_grBTvx)IFck=FcqA#NCHw}9{)~+na;i4qZ
zN3*h_(gWa0)`IA6jGJzW?d?d`wly;=n}!NiJpBjp0~eHsD&dATc&+!fAo=$5Vgn1W
z1iB@4ps-ubkS6hzr|DcA97)NUV)d6uPRaBHU!j^dQ()3-f7lvIIX*wn2K~}crIKK}
ze|o)@=6HQ#{KS+4c|E5MLqo%-o>c(FnjE|Z^bs>4uu4d{@|&;sEFgdy7Ng`P*_&L)
zQBn~S+p)p)_=#qu1ieG@`#;2EsPtp3n%SuvjQR3mXTz9LE#A*42O=gdVz$&(QP_|)
zdHwe@r{C9}i~Q6R8L@WL^Kl?jrVAN<$*=k;t^He#Qov_Iv90`Or+z$VOR-r`@94qT^Uj{}Kbbeyck+^Hz4Ry^)XrNMeB75GUr35Mw0r?VXV$wh++qqB(OmX3Ta
zpAcg-2g?I=IQa6JaU{vy8J35@c*DYV7Lc?LkV0^DE}2y_EHsEWy=-vmz{kJ2PXCl#PUbM{tjfdr)2*`)@;eM=vC59jPn_{a%qU5_pE%3^1
z!!b(*y2_EWY!TmjYJJhooXpau8xW@#2Kj1jf9;w7pK%pxx&N=T|D`;nu779WStp^<
z;0@f!5?dcSD8_~g5uk_uwv`_;PFYm=g!4RdFV0LdC{A?L5X0O(Hlo3RcDE;vk_)BB
zkzNZ?s3|m9`b61Wu0WlLN9v97a=Y#rzdSz9c6r$-)xXJ6=Qmkl(f(
zI}~h@HH$<}9;7+NNW&?L!wyCsaRz84qmkrxfz&4~6J3YfSgH~pPdn!(A
z@bUDWFU)sFgUt~T1ByQRc((BRL-wpp+TdvoW79sRe=R|0a>i-zK#$RNhst&AaVsqg
zOLU9p?O2Hp_Y
zQ|KU61RE#hCLF0n3i_a3hiLG&>HgB5{OkTLRnl)VE@VwQR)(|T8-EAPSM
zaX8knFce4mq12oU^WJgn|HBl^YDgng>wgj%(i;A6M24ljeLP6dmg)9bTT)Fi(q-j0
z1I2Q--9P-JB7V=q&rjKVAT{r4Db+qMc~kG`c@*(C>Pi7&K^(pCJ|+Bk{~Z%UYL8>d
z+@Qtj_LQLzbanj@K8J82J*Qp|IK10O@+=7uU
zv(>6Ck&VG1aA~+@h9l)`!LRGK#3Snk{=9ub$~N1b)-g)p+KyuMa3-2s(Nv;)%Ei-g
z9v?TX_uS6n(OlETI>k`KFX7v4Qck7y%8)j{?OZ`G{_;7e+#&y9Z7_@=m>OPN1GVgI
zXZ$;`*@AnA=YG}nsyyk#z^UYp6+%0%IU~+;Cpe0$AqT5dV15naH=B!)x5^SEKUkV
z3~Kw`^&AWseJX&T8C(zw
z8{VuSXA2w^7UC(Q_t(Y?Y%F3+U7k832Gllem<{G?c#e_G(OE_&dLu77$=8g`yS|OH
zZ=W8Eig{k3Q&&HjZ*EcOi;(7w=6cohm_7&nL!|VyV(bJ&m_Dv3x5`!@FkBcOOOg&4
zP2bisEC|G86CZ_9Ozh0O^z(a;UJviY&E7UOyOhVHs3pz+1xkDm7h9RM9p4_oPr{pvC=N~j2}ie9iyc|=#l6K4r1CGrM=vEj3jHHIU;
zgt*X%7I(JYI5`Cm4vj|n5`j@y44hmS9HioYL{lIDo$pd16l(99_w`96^IfsZ5v}ux
zvO-M#keqHK6#ZVosOpsW*yG?LzE(HKVY)s@Ux%Hb4#J2(5=#PK^{q2zj
z+t=%e67-Yjn$TiV*fL#MNfo8iw#niKkl>Ck_$ca-h#8KE8b@4n?%WKCV6mmU_GJE3
zf)8YzkN51=)LZ_OL|diI*%$J%wj5nq_m1J~6)oj|xAEBFK~9$mM}MBcedfTvh+IYu
zFzE@k-JZ-t8>9vmfh7wMX^HH+RJ9Ih@GNtDsgHyF0&Yegsi(iRc6Y0hBx3+2&Ud`o
znFtLap98O}=0Wu%~p}YR=`l^ptF+C|LV~lBYsF
zM8s<=@7$Ltk$khxA{h4VB$;B`ZLG=ujzGhQ{9^LhW#4k~ICq-^e^ADnaUIkqj_Y{N8C0A_;rv)o%;hmaPbZ$p{GXyl`=-9
zLsb8G1^+r}rc~z@s!fLpsZ|QjYF!4GZ!0GNCLRYE4WU|ET0jAOc#CIcw=FCTOhimf
zCpg!B{P^)}aIpASg;}IQc!TlfokY>zNLKC_CT924C^*Hf)jUG&eq9CPXVnNA^>dU&
z4+t7KP?WZhKg;Xzv`n@(1tdL~63lVjARKXa?NhU%
z6zqi0k50?`&$PInBWg)0o~)?e=>^Ve*k4w|`4ginl{2z;E`S{T#a;2%i%(2Vo*oxd
z%(0%XF*qsBN5vXl+q9E+jcY4Pa8UIzoX4no+p!5?&GI@fcsImN(i6wWaK8d}!>Z5y
z=f;^!_>MSD(7HGCI0G{JFM{ntx%D#`O687Q@WlRQxbO
z_9`mQYP&hSNR1Z~>#V9h;g=>IlV}8RcWlR&3;Sp2e;YfTNmn%Xn
zn6Ipusmr=*unO;W@f@5U37N}(T0cc9%>Ef4(I#pC0&)2HvS!_yz
zdWD|WGbNmRMmTiC0JaM6XcBl;bqpXs$jRWDDlchVec0ZreSBR8Z^RUyWbnEypo#!-
zgtX%)3ZhUH2CBeRDb{1u$=Y^{pb5Sl}M><
z;bKe3&#ZZQz2lopxf?knux&r@Ro{=mF*Cz74(|()$LU*`fZ|l{c*c)xjISBN!d@|7
zRdf?s>uAUo!H3KHHt`cI3jPPs4UW7y4-B+U&9<%gu^av8-*Lc7Vk~`Oo~xJd{!Hn~
z3k+&I9@)0s7www8ANj(F_yM9jn^vy{KSom2$cVJ{@qrsyCy43jLa3B--kT2L69T_6
z=*UU)x-lS)742lI~Ax_JSt6u8(V3y>_OoA)DnQ}(FBQ_yNS`bV)t`PV8sl;=c
zky}(JIxl+Gd2M)$8C(9p5Rv*X8yOC{C)=07J~yyqmOw_C5;z4Y54Fc>q8gk7fdmAH
zMQdt9<`i?m6Yc~-FiH6yZht_TMgFdE^rcgoeY?U%Detd`>bW2uEpaVpuUZ0Nob!eC
zAInH%P#ebMt72-ijXYQ`Nu}RwD)V97uOi^}+`Sm5J3W~7&uNhf_A30cJ4=9Jee<^5
zMHC(DYmu{MblCC?KP#@ZCv3XgvAX-+q`42*2@KfxjIo(q3iTJTcV;PiW-e17^(K0H
zREW6`RXbK04*ThRjz0ObaT<4Ze%KS^=zq?VV|a0)!>e<>zG~xNJLvVL!lIcPYzlhP
z+ivy1u;{clWFq;=q7B+XAcOsVeLq2kR3516l$QPx82G!hldSt(tyr@YmH+OSX`1In
z7#Gnp#2*44S=UGHs2X*488%D7ILsU%-v=%@gWgE8i2@ZwA|fKq8Y>Cl%bRUMPQ&<362*jky(<6(m{_%?E8Qc
zygt{lKQ}NbrQp49H{>?#W?6r?`O&;fsI}oLtuCjS22LvWQ^|1ih&pC!eQ){eB@T^*
zaxiS7z1{TAGR}X`8c|VVDvZH|4w{OhfIS9(JO-C^2*3$W_JO-hct
zY*2!#vyq}SRFnhR2{WxUTx$qqYI(x+wOW`?j`EhW7X=`N7)-^(K<&++
z;3#3q{2A$#8_AozJDw+~q3yqa7g=jxRjGgMD=4i~A4QuTxoR^+o5JGj?hRY<5#kkU
z8!t{J$L{ZJE3WsrkvEBJp+?_7@S12tEd~Q8as35Q)H*zRq~Awm1fY=Y|KZB&>^Ha=
zeW3%*%3jde8ezWmqto{dz{v%Kklv5v@N7wfhOzyL^MOT9Jb(aDm&Y!ouym*BS7GR0
zS3!3|p70j6AO<^tjBQ3i_q$HfNP7#oF@tO%_6CD-Q)U3dyR`+c0l@X-eLyqpy)v-;
z+7%cUU^+R$O1r>`m7pzaCQsQ>h2d=9%Wfp2J9PHU8=
zk%wy^#k~^hs9HciWTvV`7KmL(8#COXKI+vc4V8+g)Cd>FBawVx;p9Ccx8P8sH@+&!
ziD=h1Wr4p7=i2Z@aufWZcXr`#s=s|K`hcp_TNp(`@Yo{%Zo6*n-cni@rB}b`wAD&h
zqFwFT=6ABMu7@Ao7yIKFo+_~z85uEXRD%IPK;P1`f2I9g@<8mXRnw*>*2#37*RoaSs;nE>%38Zp0)MMa{{z;xd|>0Esg>c3p`XLK1RUNUjK@H4<3!LsX?;@KbPb;l
zeaBzrR-7lQ7*7RROtr=IdoM-(XPac6vN@GXj!)T~0j+mW3Z41md0Vnj=zqVHtJ2!7
z&UTG|0$Kl#ha2?vh%Wq>{`M;H-y$w|)7^D}|0Sax>v@O$XFc)C_OkZBlQ2L2O|`>b
zqyFb8i+e97ay+Alt$L?ck^Wn8_tlizkmedNc-CwSSw;W&cSr;+ph#>^QF9I
zEWnOf>Dh&eW`I^!QEjk3HX2=|9T;4NL{HHAi#^TD`V*lB9m4>C`6S@095E}p6#KK>
zyD^>beEZ$V3|z!|?w16ZoyS-9cb+l2FGTlvWAg&UI+yU&<;V6CXUb?0ei|Ga1Q`S*
ziev`h#&qcZ>{4%;BE7&wOjpcy32nT;w|Dyn7YmKpIzSUP95tV^cw5HKJuo;#-|$0T
zaf)6<@&|a~pi;4)MK`aZ5`#=7Zq2KEH7cmbPYiLdyuJz*6W@rv6mOc&=w0F9owPW9
zYhc5>o@RUbEd9OOsFf;zpqDbC5fd@}(uQ2?x!E`ixoL%kY8&{acDO~uc1#d>OvtB7d_lElnX!G
zF~_s=J>#6pukq9|Uo7N}eDS(%rRY>gLg*6@RIz()u4wI4-r!)J_XfSC$Mc?n9u}%F
zR}w6%???hUs+skUVKcK2ABG7i`Y(#h56;Oxpebr{*$2gD*1O_C1!2mR?My3qAAnjW
zZfn8@-xpC8Jf#jnVdh)o$l00_SA%eS*scC&_@R^CI=G{s?8HeWgrXAL4$SdBf7;+f
zi1{WveydPpV+>_O)Y$P*&rh-=ZHb35xeM@U3{2Sw)@HBFlUl!4B}3{f#5@Pw-6Pgd
zJ1dQeSVa~u5asj20jYGc4qFjTrtgH$)J-LlbcNlW@lAi+N+DrdbIfJY%!y&*tw(@j
zG3m)B^u<*X*&2+sO1Ga+K*Rl)O3O3r7~tVu4EBmH(R=fy(OY}HXD>}=D6>e1D2VW8
z*>moXJ87xr$BCSz>cc>d`JuO8*;Zdu6`or+OwR1|q)1`+Q6(pjYPX7<+)0pcxfbXb
z-XNvCe-)Du$US|?=)U({F*105I6}#Pq(aFNZS~BFeBI`yS1L0Sq}aMxgWw9evI*1f
z^rV<)%Pr~w64vgv^8*6#LHWAPT)<8HRJ(yfz;y%);f}!rV)fu4w6XDN$MRDqi48Oe
zFk8*MhoV+1u~J{IcQnNFi`u>qh4{Mi}+HNk>NRbQ?*xGRDZ3
z)ZgEcNDe~N?XDFF{RmIkoMLI>MFMC0gPhm0W~`|TbhuH1)=
zMfx*De0|5~YfBaCPbO*p@RcE=^Lk(}I=|2t%=kn!kN%}J2k9*snUY1)N=NS-&q*QLZJs`i&K{emow@54%_K)8hL)r=?2#OM?nK_4**idSaE{NI`hhc&(Ig@{c
z+cjKYuz??un=2R6-f!hF_+VS^z@v0|KB9SoFAe*U@e)0_&9iuaIY>m_;AGte`@VYb
z{yydUmhgbDxroOo)_$h^oF0^h#O|oX$KPc*hSE^J6!gK(gyBh4
z9V!#Mol!8azpD{$VWa9lEyB0D8t{J4kF3$Rs=D^W^_Af5@nUe_8+tvY=>XoVjA4)1
zMw#`njn+}ib8INwGxP5tp|xB^FkErb5`Dc6ZQ)#Gt4NqoO}EJ_9(8lC9ck+RGYF--
zNU1fsACoFnBd%1CoS8N%uHeV4X=;`i`}b(q_7PFO-$jN*Q58+Y3A&Su>920Wl7F>j
zQY`h|K2ztK85#9{#my5}(U8qv{w|{GY_PuAp-4SBHHkuTzrLoxUmP8u2E))AG?`fP
ze;uP&$y+tioPS$c-EOh{ITB0uoxU>Br`H+kXC5Uo43E!DQd&scY&qZ!TZOPLUIy9R
zIMTX~Dx)>Sg>UnIiJpAej*R-CQ9F@Et1-(xvr=osgAOaZ(ogRB5*gljrb7^L{uFY#
zqP`EL%OyPnw5PXPxz&PFSS`zqflP0twm8S<-!^AFA_S@>QvX?B?*JP5^?vR04A`as
ziDC<6SN9g`qd=VQsS*bWeVyPRXp|fH={M#=T=wTW4;C9az3w!VA!dW!y?x)6C&jK=qH^oG-ebZO;$K2WBJN;hwoVMHI(+^XJ1sXH!8OZX(_Hh
z?zcbK<4m%4-JVU-RL_-y;QNy%IJi8B0~OW=KaE9RouatV-J=7?<^
z=rw;=JI%;4$ybt)+f_yn+uYM>i~@bwq!O!*0fn46p9ivdOz53@>gXpS-o$ahlDbWq+k{n{&Q6gl+0&GM}53SK{J{1R2=DPO`Sg0K943Is3%kLkZ(rg!G3}qT;x^5QCrL8k#2me6O6pVj=?x57b?c1KeCTn>lM+wP9q{_>mvBMB|kgD0KW
zQ2(84*bou<
zQPWgfV*=i~qmnB-#KYmhm{WGH?B$p&?=9_~81lPw--iXISbBH_n9v7i`?)Gi_rnM=
zB@bz1wS})&Wv1QK?`UB`>%%}Qx2ioVks~B5$bXvP&dq-9NY9M*rck6onr^?##c&}_
zO2Q3aZlpemd?R8WOAPC_h&OzdkfgHg`RShPKqE5Anm62tAFT0m%Ln*yn)gK8-#4x{
z;=TD^j5)T7=*TlS7f2X>AG@w+kg!d(W)w#gGDIp(o1$_;*;;O+!mGQ1NMvrr)e=z9
z@g+oTJ;!$Pf+=r%edoST6m`;rz@6cqyy2pEhB$ZD5K^wyjBH#Deo*E7^A=80?4KA%
z&4zHcSpAgXDTsyAHUg1?e$P$eNMn
zcU}s)j@+>xY0mJ5>9lmwX2nGYF@jv@{W%6Y$iac^cNBs=y;eS;%YfFW<5BS&*C2M=
zB~ohYKrW}%O#2GRp*Dehm&INYxP~RqSt>~B~HLauk3ZmSLMW~81z#jMcNgb3q@AH
zriV&EI+eR6ChLg!vm~Fg=d;@=;Z0Q2szgy5|W-UDt(G%KT?|ITctC8{gi+<
z*rF$|xP1GGu^c0jW2%Z)Y*hE~PDvwv)UcBhzaNuq#KpSC32w_)ze^lCz!O_EZkAlu
zxMyG8nDbFNWKp1y>~!F;YXbuzdJ+=RLiQyZh9m;bfDyv>yNdSEG3QdPpDsVg*A#5()NKd4nwuHzvoXS*-f9Cln)M21jNK(SuLlG_GaSP
zZ@66!UVOa2Ip3Qp_W`DSys3Tg0T`ra%2V1oIvQ^D#qfCD#sBuxyRWd6J=I{v9k-2i
z#nCY$EhtbyF5U1$gxDT@km^^JJy`d>Pv$fe7WzTJtI5v_6+S-~S>Z$@QKO&Dbzt7E
zh25*45%~=C!KyTqXO-x|&Pjh56mh(JOC4vDb+~vjr?DGyAMId$rdKgBg(S>QjOUoO^vGeb9EwFY6GOTHKZ0R
zF59UdHYm`16l=XzKGWw3YAzPFC5T_LLtx&q3nM}BlL;jHGGyP-tp$5V2ralN=7!!o
z_&zwo8XB9*Sc%YLp(L{|j5wXt@*iVg!{m&JiTo@9_gAVw3_+~4FJ}w&xsd=
zU}4Z%ClJTZw^A?cSrQvBM>{*7hd13qiK^1GP$3FH7}FHaC7MA~5$WAjrJ*roPe4)|
zEfJqjxN$Y~?A``H!bM19-WoGOj(#x~ES)0_(LQ_d2kp0cdD?v@H+;rUhU
zRdSJ4I*Lfj`Uac2D$^GW8vADtGs#m6W_OL5;^Y_RzU%BxM?@5p42BkPym*hoIbJ4mH7sDYED>r8>*O9iAwa(kHWV;l#;uF8&FZZYNQ>#EM_=O|4t9
zPUB&9m`e3L_AiGm
z(l6Cm%nQ@l^o^~17n3fLAb}iMxi<<5N;JJrBB14Z&61Fl`@^9U$~U{6s|{UnjSZ)Xs}$n~qTMu-&~dwv-z#7A
z_cGz(;kR7-pP^IyNMmdtF^73?IrD0sg1{@m4@Nc3TXt@xo!oe~;>xtEA;hF;-i-Xy
z@urVv43%mPHk7q{gTq7k@diXl)(;!+Pl(JrJg7ux^DZB)cch(UVV8F#%eRE
zQ;YNV23u+5#{Eqm&qT8e{2MU>{Av~717HDH|n0I
zNKb0jK4J#NO=J?|whxMGa?mneQ4s^!EOD|KMXGSL=V|#Tnf}{C53E%Ex5+!6jvmJ-
zQ}r@NDf1#zE(wyyns2DivWg29E8oj`HWEcqdA2
zDX12hn2k?bb7yjfUuaA(x?oIDlSwKFpVWLLLSk1(
z-sXtjtu9n3NajX@bPdp_QBu+VHDmsi#_T={0HrS<|e-RS8x
zIPXh>=le@|__nsTub#K5o~MXNNJgE3Z;)_VC@=fbG=d3v69HGU%w!N3P`^YS9GJYk
zyz1)f6`DRDv!5`3u(u~2NDEs$+qn_{=If8fZ;H~_7gK(IL2mL3igs}^f}2x+Dhu93
zCRJuL?EHk->CsDBrEvAC9W`A(yulQ8$2)O~bAXzUD;4U{_DFZ5&>LUQj-$T|P~3z<
z0+CYqlgn)~>+K)7=AmmC9U=Ht37Q7PhVkrXC{Wz0Hl`DijI_K}_(n`M^os)o=-fDp
zLikjLSRYy4xInLvxK43$(9b(Kb(Ax5``-dpwI$&f^W;So3^2u*I(vqZSl7?O8|+~)
z@5Kx%?Vvw!F_k{Z+>{Ku^2z3p!sFvK!cN===;P#1nHA@k({
z*PY+Ic
zesA~ip1i@;9&(553~SYG3!w!^V3pWVoyUw2s`c?RlqI*N$)U~wH0SSDlDlw>(EavzT&2akrqd1}p6ts!#T|ciY@g@jS@xvxit+2suwcecN7iDFA9s`(-
zpi+u2LIGzS!4on;5nvGGdKr;Li+%ljzglfLLjUl`Mvv>;xObf*kr$^-f9)QZOsPiU
z_l_T8<(|F`RWv;O-V(+eE|nrNPObr>9
zoC>`(TZXL84G;DvuQRe<3sZjNNXLvvGwc7;Z9~4t`R$c!TOAl)xzY$XzTs*e9*ME|bOvUFQPf3%ZXjw`KJv=@0
z)k*=nj1cNcG6PFBV9)!iT9^$ua-ZGYekvzU5ry?I#jJQTuD9$rhJIYX^|#%c4Jb7n
z??q!^I76u_cY8cgOi8tPC?Ubt_=EpgLb7fq+HuPcXJfE+=X$Z{{q;KfvedEhPZay*
zI-U}8@ui84*$|D|THCnGf%JB{3&GbVW1giRFLqeCmZlN+KiRxZfJM_%1{@Cdna4(u
zE$(6{P*dLLL}XFTHJ|yvNPFwBs@A{jciD;{QqmwDA}L*pbVzqeH%K>#lt_1Xw{(Ma
zcPvV}yPLJn&EDJnd!KXO=Q{6m&L3XtqSl(+YtDO)@g1L0T56{RCi5&bDxteGRoS4l
zoUZTR)}_}ltG~ljeD@CL8<~6C*8dn@Ei>eTYLEVkogqB^?e5YiAMocxkG^49V*KTz
zQND(4yMKLe^XC(R!DX%1E_d&*FMF8!{w2vjfWAjw)$IQi61_x#Ff9G;{Ybt49$)s*
zljz)R-u5Ahbsyv1UmxKH;Fw@-CZ{?=QLI)d6q~l63X+TN&yTsmpJv+w2Mev&rx85H
zb3MMKE<{Nmx0i4OzToiEWDOke3@C-hJ_ldW*~AZ5D2f-?Q=ml8%*1bO81_WI=>$aT
zlG4(i5>qf=uU>1T1!#%Y)z$T1nzZx2{^k1d-Ai@;2tS$uYi1L)2|js&mjYbV0x++_oi9OB^M@MYnb2VO4#
z{viYF63DqXqSvov|4Hw`mRU?E!PL;#;Uln0nY7!!L#Gu@O=%`3Ccuf)TtAdp01W5R
zQ)A-b^!Om*PX5$-0PueZ!II6D
z14F64V3iH$%2K#p@3O|Rn8g)}gR9x)@gOf!F%T;IfePCG=OB1hnQV>a-T?d`xS&}s
zwmbqs3Z!4)Zs(4xEIwEKpQVP-L-?9fI_)7YtNA;XIq;nDde`MxDAN8h5~beP=f;;I
zATY@Pv1DJiegAWuh(zGKD2l_Dt`f=R1s^4od2+@?zgavkYXRG5kzB!5wQIW+bvnfm
z(iYEF!5S4RtNq_S~>G*g{c#rzT#lw);P$LWFohe72a5%8NZOQV~NLrbz}fZselbRvlwa>S{UWh%&mZ_k6lE
z$LgKHGf{8inRxJoST?t--agpt(en@)nbJrGi_youbWIQ6i;9t0URp6){JtNF5RXD4
z1uY7Uh!een!-S5mPD7BJVzw#@dKDWrQ)Tk2-Isa9>Z8NLBEn2AAMfHUH0a_HjPo$7
zusUqL7L9m25K)|U+tm9|s${kE-T5Ud@EBKs$9SkLCP;gVpR(N2%>Uc%*A(+L1XN&^
z#}5aOoSxcl?-63MGhVL)pAAJM>>$1&=ErL9fqd}b+=er^eEuqh&G}}aXd&^6{>yx8
zqtTgD+jNQIGoN{Plce%Z9fivUwXB%b1X(JIqT~i2_*Zsv1DWx{V7>VdSI2Z41l?lm
zB$+(+Fx#CUvGJV3sI>=DVNsVP?VUtrC4z=cy&S62qc98!i#fjN&
z)iDE;dZamOf#xjmlP
zqQ-7X!L6}-&%OZg$2J!CyU#yH(Pu|u4kHt9a?Lq}HOXFT+ayhfkGrld?XUHwu6mqk
z?ml|IaDsGEcX{`Sz2~b`5$I1N(>-_IYdu{%yU9R8$(sl&{_{;wmUrmeA3RGVr6Z`l
zE3zppd3f3ovECnb1Wpddq{3sJ&iEI1#804_XiSDLt%QVVYHb*RVQBZlEr7dz^u&C`8=)UT5&|6m7dtJw$eJA$_yP`6SeiUW*-?nf
zzZ|L<0_<^Hapo~dDM*iRSD48S!ltIRxTmAr*u(vBXK~dXD5Ery1{4IP4tHHICAWJXuj7kpu)jq%lS;U1mhJ!{7Ut=Od@uj{XS>sHOW22c@Ugcv^rcs
z2$ABz;4{zSi;lU@na<+h
zXOwb3Ojm3B@w-a)&yF^nes;(9ZA}PRACpfD1w(|VDxM!t6qZGXzQaMixgBob5ZY$0
zGAFYC0?9;`U+u}j=SINBel3!b%;o3Ew5)SmYs^~MUno}f3~8k5xiiaNsvhRD-MwFZ
zvjK;_^_O-&mQT{;+GSjLR+!vW;?9R#&l%9(*JLq2t#b*Ue0`goe&al<#bYyABT@9n
z#Nsw6dmBxqn?utll_O6%CW=l?rh+!uq*$WYNlA&pN%0$J)KBh{2
z22-(uXZuP8^uZsK6LtUR7E^43h;qTxI2wvZ@*4&O+&3!4GzzPY48Y
zlZD&Scx4V($lo<uT;db&!00aicHeU@SUit7`pONeNKQl6~)?x9&D4vx9NBG
z;?NR_C{h_YY1~
z%(IW6bx$2PMC#3Bfn!}S1>Ib>!npLpIN4JJ
ze=ych2nY~WRSN1GU79L8uwfevU}?-C;tGNd#FN-q!y?VjdV%MvP8AIorS;O&Ah6OE|A_e0MjA0vv2AdA
zh^jJ4nM2Vhql%m@)~q2PvmBlRk*h`5!HLasrQmq4+aAIpwVLl|!=7AvPPiGYCz>Cm
z7dVT+lqYC;Oa!%+!=-v+;GAi2V$go%ua9Gwd)u2UttZyIs4Ux#-~QDewi9HI33yw#PVumA|K%a*!qKM##X
zwuTh(jjZ(9q;di(lJGt+bUuas)bA$hqorIFRfF_ZI<~MO`92olRT17+RI>2rJ0!_%al4l4=Zeu0WwMRZ$fm7J)I5GlQRok9P47m#h7o_k0pD^
zqNiH|+};SvRu?d-x-|7y7F|Nk@iZ!)i$r&Ja761wE?*k_zZz*(Pb60-6pSjSO^Z3t
zHotz#UmcqGD5%%)DoB4Fx$aq`-qrV$>!fEILq3gyt`>~bo)IzpQErT0tEWPr%`F0Z
z-&fuIxX=h`Sle%U?mNKvQ-QGZ?Nm!!plsP~qEi(t@u^zl+x=K86ErI}_&bA7s57mJ
z5y?;;PHU}7YgVI0Q|;P1lCsL?w7SC+-o;y)1&e_^jfUFA)QaxzZXiLF0+kT3Z*Icg
zYD%yc6O5S^Cg4)KIJ3(oc9$uP0Dq#)rR84Jl6CCQcmH
zjr$we1Gc%qw@nN#`|>#}RJS>;qgY;f%_1@sSbwo2rSjF0FArck6F{)s?k?dt>qWU5
z#14o|WG+D|QDgfuL4OHfv2b5vIP!-MJPLswFxXYR!O@7vuzrsg??p>?8`;C=KH-(I
z-At+u1tl)Ah*V71bhFUIOhx1^E_Q`MX#Q+a)oW~P1C|7ODeYtQbsmEl(K#ygb}(8@gJ_Kn*7l&#KgzL(W4*mZ9_kptYn
zyec01#66z{zdM`*hiyFraT(S8J#I=byk(kCnG}x`Byll3O0A-f2TSN{+a3zL+rhel
zbf?kf(YLJ#5r(@GDzpN_XeOmVF19blS(srsBucd7kR%6p%{6Zz{!G(7$i!p6T3=63
zPXG)!UCzT{(H7UUz=)}&nHbD01!$YR&R)w#qhy^#bSjEo<>o+
zaDYUHQM5i+H3pxI8u8wPy3blV-L+Kn+JCz-CAwZs-IrmKu)y$rh@9S!^k*+JrR-Tz
zgA*JteIzr~oBRV!&}0TQ9%tU)R-nz$s5kn{DJgR;fdz@H3URIxs{OE@(gtJ0!+8dr
zFb-X=N6mA&FI%IX8HUIG+a~#O?2U}s6b#Pc!B_Y=mNv)xh9gWlqZ8iF(Gu}Zc(?0%
z)DruF0_+Ad`@XVi7Px6I)ErhB734)1$A-OoHR5YxpXaGK{Za-N8U0Y5?M~GZwKFo
zmy!`Q2|sHpW&p?PB!wQ~+*s1(g`<^hjt35#=+wsSX&~
zRVcQJkg5wZO^UFxLE5(#qp>n(Mfax9<$;-r7$lXSR%g$6*$?_(q=TL)Ir5dG-*}w$
z#cytJ91DCAKL8B|Z1R5Y%*LL`W|u~WPE9Tx^s{CVu)RO>L7JL~Wj6h;?D@D&BF?L?
z<$nGkFNNc_LU`#-wIK;F?TEWym38^@%)Fv>y{RRiC*S5!Z{Vo!CtNe8(ow@!IOf~U
zwklsmrgE?ib{gB%Q-ZD?PrS1#d2DX8Z*6o1~^tk(^Yngc-r@vHo6#Lp#L+OtA#gPAnwY=C+${;
zfSZ=CZjaF32CRJ!AMh)U(={5smscaVJ`FEg5##BfUoY9>MRzr0GZI>mOcE6z)a0(1N#CT$hD5kZByjQj=n~9guQ{LoW#>KEGA8(A)?^zLsaF
z%HJJNMcD=kL_)5=W115(7$6ByAg9iHSbWGQxjYd_gRb48egd=G^G`sV6h0_|vM
z+}Y)PB`X=4>6q~AR60NBFcH%#GHh(&OA(T$N59Z&Zmk3Tx=P?+1>&z;ftg&cbCWGX
z)$ih&udU9p`B4~Y0q_H<}ic{m;fc*_#o!4T)1>iQ#a&=74F7D!KUf+fTU&q_@*{#DaInj}iEzbxV
z(EZ~usJPv+s?<8`zg!EIxuuLH5XoJ-Vhf`lJ{d$ky)_Rn)r~|?{53e2e`!391vdJ_
z%_ycvBknwz#u3>wHy^-F`$&!=E&>Lfc%Yqa?(p7}O%MjRekf`*e`x7I8&^Qju9=2E
zI;K#|&E=#-a*%i?A1;pQO#(Y5sOy@TiI2JW;HQnf5E#2q4kwRccC=%XD2T?)y*^LR
zQ^DvYbc1+*8{>8sh93`dt^WuPG8mpD6NlNMibjV0T%f<{qu@
zDGvGH`7bC3+RQz}TCn%P+jLpvc`-Q2KZEUx?b1dll1;KP#{8+c$2rkGnUue?tjC#M
zBi{wF39i6pG};Tz`L%G_iVe?5ykh>$DO)wb7oSVQ!uS~e!5Zn>l2$Pw$2YX5q2DBQ
zX<1l>13CT~)B8*?EBie|z*Oj!){F;71eWX4a54|2?aerWx3po1|
zIl+z=bYU)y4wXCUJ~=aP_Yg2-Oi6QbF+YlGH|kzz=;2=l?E3>zc0>PVZ$lFiGu65&
zgw|}E$;l7%H}4JcjMW>}H*2QxomTr6)z+7v8g)5QPGVNG_rp@hrMrjYatrgc3kNh#
zVXg^nH_KJRJze~?;fv5Tgz*H^bWf)o}`0E2uv6y(2s0PcKEUEc-fed*C#!q
zGA%ktdKLw9FW^tH!A3AA&c!Efgd!vf@hdIMw*Xg@
z52^g)H)7~yZ#>EIcN=;Q2X;JME(GW5C3Na<1yfjLEsv+0#Icw?Gct730TZpTYs`^n
za#AUMbn^h0Gd@4s9BX1`Ql(7q-2Ngct|QW)HK3Ap{93NK)Oe`pk>vcnoc^)d8q44s
zom&wRQ3h&Ommd^wIao*-KFvP{(`dgUh?tJaJkxq8a%Rr?v*c2rKPNZ4Mi3sX;g*ui
z_q~iJ5lNcf?!Hf^zxO~&E+;4=$>*%&5#tLBZ?3uSf}9J6tyh{~WxQ#Ly-v8dUz0eK
z+inw3=1qJQ9fvYeaRi)bJ}3nsaI{!YrcG7MRjNVRp6Sz)4ThJoqTVbQ1l{?s-W#6{
z_7=D*_D$P8epZeWEUet{l8Wd>e#vMkXSBl^69;>Gjho-fWjJ6!DfCKi?2Y0^st9!T
zzgD<@xH6elwY@DEud(|f>?!gKgTQ@J)U;*nvYsdFjO~c438L(lw>!)?Px`v8D`$nn
zyykb>pY&E8#Ioj$9ol`djRN)L59#2w#ENo1l)VmL6}mDqzEa$po8+ZPB(6LJ8P_AV7u
zWsVP__>}PL_iucAdwW3C6NqOq`(kfz0Bk&nuDBci83not9tkXnl82
zJ_(yBRK0{9cNLg1?6b#>?XQP%&7i5j#n#IS^zj^%p4PEv
z^x^u@9--*tU$2O<02e@z7O%g)*&2JgCHx@d%hknN0#b_7)RN2UIWuDo&CHRc%2=~I
zR9jSzGM26eqAkM|kH`$!o0d
zlth7tx2ycq6sNTjM=1V^fa|H7*Lehr;A-F7zNw$Ck4&O1QmlIk9@41vE{&HVZpT-d
zw-q;8WVmFOIu1nqwe(Bvf9i~A6Otpv37%)UK0|t;*VX26m)%(ZnDru2psCVx18OJL
zXsc3E+P-tVm|=E8p||+dh5;p}2B)eW!q}-DW4jSwSSV90t?KSL5EWndf#`&g-NREzvcBZpL1VGyMS*fzi23iwpseos-aN9_1dV38
z74L2N158Xs-~$_t<)AGtN`nACnQHcUZ*RC@;8sJ$jw8`u5tk|s`Pan6-!6~WkAK_G
zmzY3lL8n6t;Nh`Ztvr>8d#44qy?1FBu&kLtQ=#47Y&VD)LqKyY8?*7)XHXlJk+s!i
zYd%vM2m&#TRx83_m$mD5r&MM!^Jy7cQ3;V6-K9LYfgaLe`WfFWiMSh8n5)8u{BeDBX;jO6BN5}
z`$C?g@lb_NoL{_gk_7DO$XX8^2_9>Px5>6Bq=D;tklxW*Mz{?BX)INE%U`r{g
z_z@E&06n`HGd3E|uiU;`!za2TAewrU0U}YiQubH@eOWIdR@o!Eb3Dk?;_r!%_XN-*
zjql|>Xx}p
zsY4f~|19`CXu%6s+TTlyw?dGct7Y10J2O4GPninRo2{ZJST~Qh-8Fl6m`5p9W~nq8
zDc`sT>^0+OY|umA8N5g-7;mdi5e;;#lFryyPBts4=YPaVm*^#qB#ash9Y`%0ek~b`
zzG{+!)DVoG&zsE%=lE87Wxbxn=Qmq}n0I9!gPagJl}!>l#et*Hi|3ziJme98*0cts
z+q^j2;kgDsS~a+1CgHTVa`?nlkKz7raxGm+;lDKVk6Jc<&!+4%d#ga@e2r)
zng5pb(F0L>P~E6spjx$wl!bBBtGZj=(b17MY$z-?s`M;TKYl!t
zC4p-^QP2(M59V#;NUW|~HnfuyPh@6wbYCfaHV-XKaIDhOz1fG<4yEkOU{c+)dbIZ7
z_9tEw-tnOCiqqzvEe5SdOs6tVM(2N`B|Afot!xLQ_r94w#4e1VGoY@5HOA7#CLu_Q
zrsq`DELMb9V2fa@gfXkMmM&DImg=NBqPZ
ztBj(&8<(c0knMWL=@!ZBM7v-gtc-4<5sutwZem3qD?w!AH`qui>{rqSSZ}vGG2H<9
z#9i&THddT*`|i2=g)2C_J$jGpM!5RK_PyR?t4BT5ry-yp1FTW)c7N0}%9aBTu6L-o
zs<;J(Almgtu>?^%bhuBQ!+ZXP#Zo7JK2UmnPBmTG6WnRl2CS3T8zfJu-()X+0`0Fj
z8baOzT4pAI3Zz$QQy!e>mLLSNMy@
znVa?afZzJ;W$i(VX>SEGOyE+Q#1&tTQrd0yVXRpy5d3F2h{?D(vopM~bl0L%+@eJE
zWuKZ2*)j&f_@<{tHVA&V@AxxZMN%{M&2o#(H*^@{NND^O`?x7c&-D-D(>FqTHN{I;
zt|DiPt5fBK+hfe!eWBxCIldKTzM-8qoWKdLmU#Xcu-NDKn~6ZK!}Vq&m>s4tMOPzb09dSG7Kg39t$shZ%p%^-DcUJ0&Y
zg>KU2@9G3Tk5VV@k*bFwrJqS<_-OQh2hcUdy?VQDTD$&<4?1VQ!B~NJN#^TfA#A?>
zjaY0=JTXYNdD!Jj>zNy@l;F2Ib2(qt_F}u4fw~MalYAS+#qEJ#37A`x8RV7VG+ZrSjz+2M&n$WgB`#lX^F9E#=h@Tzx{qTm|rE6_IHrN`MY9{Y|qv1hisDL7U^{DIGv_6e}}K_Lp4f>F+XT8
z$Yu2P37A$#+(ebP9M#Y_RbLUebKz6X`PWNH>z)bFZ?lE;J#Qir;Ok*-8%0ZsnD
zMLLqZ|BiITxYt;*!;^}IB-Pn$qC6_DewuZ^A~#Qn+g-QXp7;B;I&<`P=9;wXSR%i_
z1__K~J#jw6Bc>E~pQDNtWidzmHRH^&xERWAPEzkach#>vnk{LtF_fCXI&3~nmDrSXoj-+a+gGckjSz?Y(?TFq+*irkrHNE?Nld_
zP;MIuDD^JLSu7x=GL>$8$`mxWpyfinieC_LcFAGvZJxKN_-4M@2v8zO_MiLjrncJ%xSMrSHfqK>^z5o@&;~Hz_R+HEYZW$FuU41T9
zh3bb#Eo8%UeT>?~;hUT}YYK+2EBt;xdw
z?B9nHQI&b0p5K;dN&AOT0XRT=jzESh?{m^;{fFwhJHH`cOZ7=IHV(M-M%snaX<$1E
zVNAo~xGf1}8-kHqxfhN1i
zgac2J!5yXx`SI%QH@?(gv$Fw2A3K_wlZh?<&U+LFRTFU91%Syw>2gEqJI)$_u6+YM
zo0+O6ha!W|pI-wl1{u^M-}QubtsYhJY~?jvBL85F48@>&u_2_$RGiSQK&}^xNpA<6SZXZxi#qUHE=8xH(*14Pz+=4A(At5_CdcGju)9+E{J&vpbO^a-;g#@NI5rdu^X~gnK+ymQ-7vm@jhu%jI%N7xr^^`JIV+!
zM{lWJ&qs0@I4AeTluUj}mmkgA8_QSpN5JU0PygjJ-PsO%#IoJlnVnpYRMcoToRNmp
z`OJb^p}=BtzSfqv#PxdbymnM2#LLHLq|88ay2@fOgy^I2Cl)tq36SFhL)d9z(X`$Z@Q;WFZN_iTrvy-($L!V}!eYVQ)0Z>zr+mU$K6vUoO~zuxuJ
zzRhZdE2K9>n^@%9QMpmqAE?Vy)fQMyPWOSjd=Nq(6{LaqzC8{Dn?-ZJ*%cFyW?bAv
z9%=tpv1@FT(#WK_kA=|cEwKl-`|M{ER!jud*pkUYa%*_KK%St#3Xwsmi9}y%Ut}Nh
z=J2xw>@pv66G39dHzEi>R@~DSwyh^IYk;NV}a|%BjrODWkVhl2CFF2+T=?w+6tjXf0
znorKV=m_hjG3QALjO?+Bk|_;3v)JzH!@)Y(yjD2(=C8L2JJC);EI*#6qcZap^(2tV
zZh&`3FiHC14pY?TbC;6x^4BCETSMHIyMzZFx;a<+@3gTW{y{oH(Y$37;vX|*jwsn8
zV$;{|NoB4*xuC@p0`fq$`VEI4;LFnX@|>=;7L&xlg~p8YrI2
zkBR?jdP1*DOzziCS!#wG)0y{-fd<>RG#fIS`Y7V?Zm??%IP2N0_ZAg-?Rt*W$Ajx9
zp30AjD{?h34w4QZe``U2RzM`c4JdWNG46|*#p}-)GT^nu7W>zz|4Vp+!T0|>JP})k
zvbj$7uj((A7wPrHw@kZ_6%tV_?a_Sf%QR<($U3OG4*j|-{+r{uSys?w7uL;_e&(Id
z8a0MA0vNpEApr(Pf69mD$$Cd-O;jk&av
zUsiuZc`ThRnfwXFn7~fv0cZpK48o}n&dvsJz)cp1wvNu_DOB^WwO!NwmKn5-4!YtX
zgMbFB^}{>;tjX;bF;U_x*u+ZbE4*#I-s1#A^bnvRG=^o#?3@R#JbwTeliIeL?Z(+9Q$f&3R?$OiJ)7mE2PJ>Mo
zhVMYK;0tKa0==tRT3RPt
z_}A-Lq5Ja;q~6~^*QpQK4}W#&QD{
z7Eh*6tsaRGbFn4JH{NR#>eMka=wF!nNU2(g2cqV9zwKgNx6jw8!b|O3X`GPm{0O8*
zp`)HhA#q`a0kI~)C9nl9!6I-8Du7Gysi`QDb87-WB)7k--Z?%Z^7(7o@N99&scH@V
zq&i%6V2IVb$r@Rs?nR*x4g8iEJ64679aCE+`bsH2A933rZXFy5uQPH;jEMT6%ZKo!p>E00
zEOh;RFwYKHAd90y@2VK`5}>H)XQ@MhO$@0WCq3SQALF%o2QRn6J=(i)zL%x17{%Ux0c}M
zgDHbzU9AU3LTY|RCK8JwU1UqMIm$kgC?yg=zVjIMm;M6i2Uy>r89gG;v{DJJ1W5q&5F1GKZr^U$R=fj
z%~*F|>@owBto3S-1Q&M$pzil5WEl)fVzXO)U93nAP$bcsZ-Znj8mLB$?kHcN0rEQTk;1WT7l7yz#9qDUN<}-bB
z0#XS=v6czm>8o0AG>zi7)bJ-H;k@}_
zgsd0eqxRlz%+H0p6+ZrKWaihYs4g*(L8&WNQlsPWPMhU?2AS{pbpr8nBO`Tok<V8FQHQPR
zQMjM1i)Ej>aw&kw#sv*Q=hz3PjhP*j7QPm&cd710t6~vvnnSuhLkx`2q9m2Of7vA=
zny&k4!gF;q4Cw90?uRq|e;|>?C#+J%7UEB&yNV<*#a1^r414=(U&@~a)ub&wiE(xj
zX{pu%nF<86jgK_W+rgOv`>+iHAr#)7H%mgu9kc!TdXdP-3He{No=g-d3s;$Y<|>yx
zI^XxDH;DpFzGSy)4}bqRw6sNl5a@SsK-m!x8&){FwYn;jE_iyBuqi)_gpdmFdWOg4
zv{YC>Kl55-1ou)=s{W9@=LKnC}@v2o%s4kwSt=7Hh^
z8CN;9)10is-rVfAnviarz}
zUY%*zy>aXU@0@WH1>g8+^TUHi9;gNHUYZ0{fjx?uiM7>Y>Y;+TW~5n-8j+A$R!83~
zVy39yw77&Ih>64&;6TZDd(7T7r$XAy=KRgpUkR=vobCv_EbNo+xfi9%c7>|Ok;vws
zz~w|UxKl%>=YNfX+
z09Fwnc6Ui>Bs8LR_>$pKEgL@nwSG!n0+*303Y@({gMRtRxL#Hpd2};3m_G#$AFx_(
zx-abwM?KUAn?n12R~%s77*;?91fyBtTpvMA7RME^5DzW>*m0_=%QFN$&cLtxX`0-S
zB!-Qnt^bw}6aT^cj=&j8JbVC@yVb(5p~^*9=I~96hidccKV@X90Ld69!eIH|`pI9m
zN6ZMgjnM31)$s~+hG^t(aR+-g=hteRET4D$H0yt{r_lsFifDb?8p}g+raZXR0+Cov>)cPvi}ZfnB~qkqr{2^MkKh(jiLWfd?G49DQy3j
zr?z(MUw`1{u#B$=625r~pnCpz9jW(DkZ<)VhW9tf<;vUz(39O^*sc@M%A?6te}
zkSe;ltXQGvK%WCagm+)ego(2GK$;4}j_;hU=RgG$-@C5|hlVEkd4_;|F@v2+Wom&+
zcmVI&gA_^y)S5)@3Ag7zm2dFZG*5b?1qX5+P>4SQm=(g`Fsr+;!}6;DEoc@upj6P!
z95qS8TD_OuOLk}V-L4IA!SvtGba|)YwPGfX!b6G4vO!5S7?yV+hE@JfvSwR=Jn0UX
zak;C@f>!yVpB&dA35KLO9~z$E2;sW;z_DEDl7K@38FU$QEFa}q@aKNE>ui(dykR21
zs`1Xdx!5mpE)jHE?#@fVeM)YAzcy*M0`sQ=B=eSE!^
z@#V#O^6K+}rdr^>xUMA<7`&MT(8>DxeXDZxaW*W}^+;Qz95p?D(v@
z=`G1O1b5~M2bq172Y;Y9!#i{3&>RVeJPrR^HVitxh4ZG^geSjpz0gpU83ooxK4H*q
zb!le9^jBvqPRyCU=DTYeG$WTGZFioWEN*E+fhvdA$s8I<4>$B@f-jo_El-=|vjsx=lBl~F06=yU*=A9j7;>`NZqui7c
zqJ~faAyakC`wF7o0Gjh~0g%AK1_nEFYYMLhK>79JsekHp2C1E&Ha{W!vY*t*usvPz
z2HX#__AOrngyr0%;hT`%FnNVqwNJmSeFsL073eByr}|a`M?<@H3zAQ{g;D*_-s;`N
z`KfVOhKdJTQY^@K>y{=cuFzLWTknw`YSx5b^*R+$TziT?7A5Nm-U=x<#PQ?t44s+O
z9wymuV$&;jD5yQz4PcS&Z7h5e&{OL=m86}SxT&&jT{{8$_A`XuN8+&WV^%_N55p_zV)}zi8l0=^s0~rfXWSEj~S-I2H(`
z`|g0%{FA-)DB*2YJOrH`uNNP4Z>Zi_L=?T}5>r`Y_3U!h4wlGk69ubqI_?eYzc5Si
z5hq_h9{w%--Opfw!u7BK&OC9(VuQDl6fMP0H%JuTH;dOYp3g{TXjOs7Z={z_*?1t|A
zu6L5Es&JzX@(v^;V)aDY#!JK(2@oUYX6Bw@8PNNorP-Dgk7*A1auVH&n8frDmM^D0{PW59p=oM9EkABj#iw{^uAH95rCE5*5iY;
zgp2)#svI?j{+e+k=gC{128xlf!e>oS2#8L*`(^9gr+YV)NyK{|r#uU6sn^$QHPIVx
zFs6eC-?%rr>i&FY{;;ZUt|qJ2fCv*xN}BpX+OKb*x5M4B9<@L?>J>X
z$NRyWLG&Blmx{%%u8qmHQQ{KhcN>{mw}>lqq^UByY>~@yl+YOYVr_>)N*OFOm-{j*
zV2et*#lP5I)`f^bWZb3HijH#DShq-jw|V^HESBGf{~g%P$`#6cQ$5IPm^gI&
z0|HCZ`wIj%C}`_{2Lki8%hMh|k}FW~c)(feVieKc)z?=u8ihHX^E4t-b0iD%jYoaR
zgIMV4Df#WWs}j;NpT3!Asf-?@=N=%?d{m*v?hH;bH{_C7kr$6$67YRt2KiZ0vA8vE
zz^WBRxd>`R@TjQR=lQRu3=*+8?eZ2+DU!!BOs8b}l6kSerV03`NKw+Wvg&k&9Fni1
z!Q$scJIAM|gWKHW@Gae$D+q+9}Ah8^A#`bKY4qUxX-O=n^wn|p_#
zM}NQmMXNNaRBGN+HHlG~ZF_D}O{rY;5P6@srNN%!6TwyTFpDQA4=#oct#ZVvP*b|Q>TTfDs^S|l*(!3p=|fJH
z!l108z#7;&Qh4b-(&QUl3NR$vgAO>#qN#i4qy+58Ju}(=&xp+B5t8*!qeF6*w-bYh
zcLAN%)JSyDA~}>IHS1JkxfWts#E{sD1)I;!K+AX2Y=5fgD%AKpdN)TcpONWt-}@1f
z@{Q?lV(>a$naTnjIhu~N2kvNiiSMz6R4bFOqu1mEnsmuCvd{owQ7F@Y<&wK>x)vOF
z9(K4X{;rbE7wiE&e*h2=oJrbcckgn@^JxfwkRBb}K4be1WVqk$jRb^Htk2)(pBql5
zqCh`sen^#*%9h|ZhwEJ7FBlone&+2#)5}HG<#F&j%}x~NOlnB#H8cBiWyjF%{37h9
zxlW}mLUxPN>At|&0)udzxT%wi;MP9KW`py0KH)3#1z&I6ll@i0L3>}QRGQi|g5U1^
z*>m{pwg=_IEFU>}uB}wcb*0_dN~1h>lhloJHO@bptJms{)wsx;+`4}*gd;=Ht;*N4
z)8}RLz*o<@^Z$hi+lUxu;Yjp?_Jyi(*n9tcvgk4mO{`{RzkRy^<frS`~g58h<$oGQ4pzAK>?ou!U3hK;aN8K#NB%{F*W~JaaS5o
z<=U^QT}r8hWQatPlqsZb*kO|rQf9HslsRLTNE1S)WtOQZ37Hd-iVTr?j7Y{>q)b^W
z!?~V!@BO~}oKNR>z8s(Y{FY}u_kI7T`#)Tl`RG=B`Dn~|E7G-X{ii<)O~~jT`rM
zSo4i_9vG6lB5>@hD9^i)xI+K?kpbFYUzVhHMtIZqwAz&rol=Vvo-ucfUsrdsoPHd2
zE8te)rJ@xbmj?;aJ7s;-@7+we>yY(S{Nk70dfX=0(<7gGtZ-7sbDCYH8SGpiUg@a4
zGQ07x(XPF_vg7A%l%L24S<#%Vsg`kEbbPJqtoixBJ@9lb3`4WC;pd7w`B^n-XAcB!
z*s|}BMjg7h-H+(4Cd!1MSq@5hfaF}IU%=F^H;5DtqN9j>|mrqG7&
zv~q>|mj0UsD&j$oCQOX$KKq^%SSfAF**{Qg&c?p0_D24u_?0P(C}FmHukJPO%JJRv
zl0i!GW#8ZdV+JXiy$#X3m(Jg+mn{ei2{j*!qc$2!?n36YDplL+PTT5|uC1Qcr#()z
zB9508#y<=vOYZn_josI-3)~!hVWuUXN**HVcyp_t96v949$ZoruX!ShE^Y}5iq7X5
zJ+*Y3-oo^`#+R;BJx%8rj~*
z&6<4DtZj!a+mEk)EtRVm=@94CKh~Xfc6MHus^lkeD?p8G^Qm*~F;C|Bx%K;`<-8Ze
z6?|xy<|a;)BO`5k^7yj!a~~prUZl}BKIB$*3P(Y_oUuNAAWeC$fhXmnsI&pF7bxx=iYz
zIymM0a4O2KW*)vPKgjTBgQmMEj8NE<_akFucEP~M2VsHy4kzo>3ns?Gi?9T50mt|o(Thx7g_j_~FF
zXH4XDcycLC@E;Wew$w_OdBJ}dUzD!@x4OYu`oD`9|BES^PnGfmj1r$K6~;vVtU_S0
z<}U=V8```7{?7l;UxfzUuieK7*X`cf|0nU7OesiMft7
z+BY~3zIb<=nV&!g1FqJCOGB^e{vn=`gVWTM>7Ub;8K=`!{pTTzBelG~ztnkTx3#qq
zUv}o9#y$K`sTHZr_Zp}R13@jdPMsWHZ{a=PsSgs7Q9r8bTyMqXGTqAJ@K!l_`QChU
z26zA&kxJ66AYOqFB&0cd4R~xL0LJo)ioA1%e-sSsJ<{geamb_&R0R2Xc`p%gG4rdo
zq#j!Ac;O8JO4jn(?QD)wbRf&VPm`1ObCPoN=$UIlzl1k%maoW})fRT;K&c40
zOcXI91U6y-WBulx-&9>jIu4C`Y{pid^7K3s6BBc>*r>x+Dor)yJ}QHxCMh49+sevM
zCsd#L`1t(2eQ0xl&1g&F7_YL61^LlX!R55)adc4d~|r_obzQ
zS&os^SW{APZmy_xQ{tQ}3_Djr7xrmb0V
zRm5j;D4?z7o-z{!tek&Dt-4z-Qx%rJDQ7|zbuL`Ef8Hl}$H8>uKz3HdYGy=2P=(y5
z2bTQ|lfSM)6v@w@KaWaC5XQI5UtKalt4_Sza7~CwVZ~SL*!ksTP6h90hc8rbuxU$I
z?~N&O8Z@Or$C#!iE#I=*G$qQGTGhV`4mCc0{5E=tL~Q^usN>~ThSj=Fy_d>_Dz?EL
z5R1>t6U#(a8v~KC0!ffBCD4P4#VN9il^`p(#aC
z6oV7TY2zurn;WDZ-V2aCJx6GgBPFq&-!t3~$jO=54wRR&vak?4w|DPeb$}2?Z{bx~
zQgxB@Y0jR~(X7=b3jn{;)W2*m-iJD43Bn++Hqe>g743Cti0vaSK$VXZo4kGd_H>A5
zCJu~vK~uXySe?eMukx#sC&=GPDRFT^0Ke2Y>MNnsDlZMKL6`LfAWRWp
zp)gA~Zr((l;s+N=5z5agZC|%Uoi#Jk6O0sxfUGS4FmoIzr?$;#h1uEK3e$y#hKd+J
zy`$vE2;?NnRgY%al_4aD7Z(?S6G@K+Zl>@kUez!%;-qsaO@tp9{_=&^<})TxGVB2#
zqNJ=GFLr*7Z82v{824rliCfUnH6JG@8_^j2ePLk?VkmRx^$Qy<#ju{9o)V*6NlD3G
z7!C89J43Kaw{G2{%X1k$;lG*VHICU{s1HqR{r&xQ`1!$|yQkzR4;NRkYMOWl!Lx@7
zrW#Ni2MuB%IF7Yq`yhHRJ3FI6Q(lp8FM70@Fd*#@PsK~w?SLa00$g7@C<=0mS3zE|
zjg@uIsIR)FhMbTPR8diJ-Vptr@>dpAa`fOe;=A9QfwKl6#(?tXr28jn$3&~$6oq7C
zVW=q+_@J((wGX@^oKm*i
zwr#ssSm+x1)79O5^y90WxcH|4^Yu)ri=}}2+aLX{t*xFUlK7Pc`xCK;-ob`yXP#|L
z`ZX3T!CZ$bMrDBFj73*x0uJFEttrqb=b=WtME~>4(+?`|z__2l3ee3oG#sW_^XAPu
zxFa;lCQzhvX}zOXKfZn-ShfJ+N>Bn}NXq6}w%qyGzu=T^L`Je4I&_E#25=gc->f|&
z5<`ta2SHrI1S2gEpd;tpj)O)Po>OCEHPe-=O&GP3_57Av&tW=ZIltu4NkK*zDACTq
zK2?M3AfV9{6ckL#J(=;!X)^#{m!WYKc(q#AFNGI}D(XHs!G=b20+1uqs8|@^O}Jo8
z8I1J(d@(V5)O+z#Xswt`w6N}{o_up^%oGW7@SCrr+hh;3we`4t%6L}G#lr*w>HA{4
z%A#zLj>X5vtEj1+xu>*lX=!P3A}x$=xg+(_BOx^H+(ahY_^CZAhKQ)0(|RV(D2Km6w0d@s+nPqsWri{wu;;M6+KZz
zZztw8t&R0Tnx?S2nQfg?Q4I0FuUkj*|h%hWWh>hipP)$K3s
zWn^F#V>96>4?7Jc5-iqz&EfNdcPEP!=8HnDiTDTP*C!y78=06!gGFC_gd3$UB#Y9@
zP2Dz|N%>y&Y9+1T#?5_4HEjqhP*G7Sck|Wyb&BAe``XW;PDT!H@?=RLwh&PkG42ne
zV{fK%d9vpi_Vlz6>ocId!m3x5cj?aykI!Uzy?-@`O7H5=!Bx+>31+*lY;ha2+^Q(A
z=cw_1>CO#9ku1lk_!GiGlgg{|TC%2H0~lHq>?5rdG>x~@&LO@6Pf7!9ILz2bh^2qo
z{&Gc?+WEbY^>dr%DcvDQe-NJu;b8#vnmz3Q&6nCj|kdrDeb
zT7F^S+K(lJSBGK9!J_96?x6y&d0X1TlwC(};}>1U=ZSdnq`nd-N!SkpR2oB&i!fBk
zp7?HHWV`?)i=I7LQZ5J&LHRP{Ejv#gUT9V0b)NsCn`Y+7E^3+%7C!%ONvpe4un+~V%Syl{Nzh%qyDOQuhM
z`0&%{s2}?EKa{yw)z*$bX!kC-daSq7S8;Z{x!I$B)e?p_02XUyW#!rn*V2E|H&Alv
zeXJLBj~qFoUFK#})jW3=y1SJ}u5R{Jl%Aemk>iK=+N*=C(CPm<2H=8`QgVg}wRb?y
ztn#nSY;EW+q8k94HoiYAe3dwA!kVO|@zc}O`#Uva?0UDDSj$DnFr7?mLG0R?@%0*(fIlaV0Yg%i4+YqXXz`MBFf%FyyZAe0<&d^~5;O
zrO6&~f>Rz*b|AD5yZQz||Gd6S5eR>|-VT3VkUyWZY-2#5qY<(0=N
zc%5vc;-GuV0c&Wd?^C*Sho`1aq3zQd^e<@WBsG2CV@M%&oz9Rk%pK~jdxy|S)@$z3
zP~cX4izOmQ%#`ivyFH7Dj`Z#0E4#tsB(vn)VXgSPcMB12^C)<4n8+0w(O<4t?`pQg
zW|Z<&C7QhOjz_#bMs0I+B#yo@K`Iztq4lYQvHzA`CtO@+HZiQ9A+7+|uV3#kCq*i*
zEUsl>V9-pKs|SucUO)(fDke8HqWc$f0ls?eZ@$sh&COe7WMqQaMUykKv2VCiRpcFP
z-1@Mi4+fu^hsSZz8!g0dp-l}S!ZZ6E)E_*ec7=tNwHk4~2Fwyu0}N&@Oe-?<&0JYO
zlm>3gJ`v>ue^IH+UOesr{2i<$Bb#tF+DlO>kP;p&vN@x$vS=Oy))2~rVYo{%n>GQc
zQ}i^ca(TfR@I1IIHa;!Kj`f;$KRSIA6U;C2xr&obX-y8HL+!O}mPuKkr>C0{N`mZi
z9cM_*HzB@jT8hh22$0JzPdmox!Au6QON-?G?3Is>hBkLU`sAu2UTPfMjaG-ul*J10RJ?I;TiEYGn-w+nh%F2rF
z4wuXpU<+Jbp;~**O+<{#mDU?Ir+7uqbIv6tiksJ17#%Gi~Mu}Rh-6CDMjx<55QIGWMd*Z0NJ
z%t_cvdr$AqyV{vLdx@k40%CDV$%d|c`iY)=e!>t}2eKS?ct7{xCk3tIjD9y@p(6f?
zp`Q>n{h3?8CF}&wqUzDR<}6%1w9yw{MZqE}JSvzrZ?3MY;_o}l!p_c4fVyM1LqaJO
z&(l|Cm$>yeZQ4{NG>w`ikre?{oS)Jw+|>);U6+=T$w13Ak#fU@AR$r9)77Hw?z5Xp
tf2Osz&BXr)z-7bVsN%>){U3i_WuRDgj_S_4$I;M6
z^UcgNb9_|v+_TS)wbxqjyWV~MKS>CpBI6-LAP`hxAwFpcgqK%BXo{hbxl`ceF)5gNY+{VOE>$RP(m9?R{8N)m9i~6;Jjg5shI}MHL|N6i?
zb1QwC_HPvlppnNGLQ2*U$P-M&ZzM=u0uBV?sUghwLDnH*Ys%S0_9&6(YBV->k~m!h
zB@@f$Qs7NkKbWX~VySno
zD&I$>C;T$OXUFrX%pZQ_u)H9m<82~|)zRzS2|Yy7?@fRwY$S{x5RzEK6Rh=m9o?r_
z4jkR@Nn!CnL;rWJ`JGKcp#LspZqjhSe^>jnNFly|myVw>@qeG8Zzjh0?_pdm4FCT=
zk3w=`20`aj|2gnqs(|nMoy+f{r#u??PTVt-iTfSE#=Osex}-
z<8atf3xvKxg*eQ9erPeGr0-bNf#_ult>eCqSRCWyB1II>*BTlckm%^>(_Z^JuY*1w
zLgjLk29~-#)E|mPcPggrl#gvw$+fm8e`RN9|1LM3aGAUV-A4xFUCblku+uo%o^seM
zYWt+77NSG%uWVlX9F6<*_0=BXHM%6%5f)_Dz8+3+b)L?P@Dw`@`6S!h{knIG={}IX=OqH^b+d@K8soNT2^w><;nTc#@^J98+Ity
zF%CrI;&jiz)U-z!dbt5a;E~dyuH;=(+fr
z3)1D@MC0|Ncoy^4$zRr%OVJ8i3$4Ls%+X*$$T6gIu%_LwU0pV{3A?UL2hxREw@L?h
z_q%BHua6faB~vppT0%)N!X;z)pip&H*Q4R=+}y_QSO&|jvQe$0^jN8?@Uv*;lC<}fV}-AB8b;Qlm0?Is`8zeU;)3w|x*9UdLcWyYH|#xWVA
zs8>54=yk_2J<)CrYA7`vq}6QvvKeNKusnKmFgS~dzZKvtrfKzR^)tJC${VcT2)rb!L>@zUh{oyLx4aB&$=bQac=dtVkn2bj>rmAfJ
zmKsKGZ%8C@`U4xC&-kW>c#>AX_vPX7@%#$UHJgvGFP9P5%@||ZZ7J*ag3V_Os<#_~
zeOtfE%gJdN%8}_G8Sz0pzMfZjoD
zJZV4UTD4OKrqkuDOT8DDS5;2VyCo1?8_Xj0K4M?XdYHQ%#2XkIHh}$+d~ov1dS!lM
z=89RqKxIv)%dcJ1lRO}GCiwNgZJ(;KV8i_X=>q8d5nJ)*@Ekx7JFOuEOZE&xE#X9DVI6*M
zEKM>}@XNE$@V}~)_1p7TwuC1|)r)Wing>(?#L`2aIKRlv|M!Mwk7a0e=(ULJ1EnVQ
z3M~n%1L32Hrx&>GzQta4t(pAS*Z*Jc?f-3Tk6C71IwD9`H#R;(>GtWa_i3&@!R%x6
zT#(oiY
z%*lw(w`jj-w_ayU>s+XuEADVOn3XE6b$W!0?)Uipp!Y)2X@FCYT@YpH$7lCm5(+Jq
zF8%s8-6UCIn_9?Ym>Vn+R+0CokI{*Sj3*Gd0g*-=FR9Y(7RR?G0`RFHe|D)Po_=)M
zR=~epjaOPas_Wkvl*r3#o!+i9QEm;+|NYrMbp1GdYWhL#$?<}-E(^;X^Js)nD*1e$
z`xQeH*x4qKgNBEM8cUtge~WdWh=__>PT7vmF(jX%^m3i3RK#ft3+kgWZ#xit%n9$7
zTrNEO4#tZ3H$)aq9^7&(&VP~`1Gjg)^B5w5l37$xMzvy&i|%k7{Sgv&8BRW7sYRRU
zS~v8S1?^Q!;PnwwM3bS6sf}l`s<%sKzcbsJ5?C`Sj8>eWRDP=7JUrD9pOV(oiOOgH
zMz&JtsI!733D=W%tThihFO?=#JhxYOr-z%DansXAWLq2KA0DGY`sx+ztTELrOxY)r
zuP^i;d3r2$eM`&8xX%M0>)V|Vq?s>CzNBsq!u1-I;6PPSvAl1$J-H^2Nj)(zlE@__
z@-YXcywujXP9pYwaqiZz3=g!{PRP{I`RLOzv2JFt#3$o#Q~&!a7_>NsZvNeo32baB
zWMxCy`;`oD-KdL|Q`^B$k)ToC<_<5yu36pF=@81
zJ0L*BO<-vTW@ZB7;^N0_U@tdK*E&DB`Vohx?^EYS+~sD(RdB2%P{=?;lr;X7Vdq=F
z5X`Ww|1oT<_#jqbNB6j7a(sl=?|*Ez+2cq$8m*t}>mp=vcSXXcW_x+TlPt?TeaMc4xHxQr2Cnich+ZD5E!fUXZ1pi8Bo&Z^XG$iA0EcuJ
zBEhl6sAAp2RJoGmyjxc2`gX<9^EK9mQf+M;vKtzNwKkVpA185q!6;exkexxRF=vbR
zz-=Hws06IDeDx+`ony~_#KrX&m>U1#cbq^n173lrbK)tVIb56WO>sv6uLHW*(Gh{5
zz($CCc)<8hpE%5N8dsaDGF1F0ekArefn71p?2G#Gc317IgCAih;h*c1h-5K@B}}}g
zuaIWX_R8Yp$-i8@Ou_d0#hkbm>V84Q&dy$Z{emAo`LvmAcg~kk@7nHkS2GBor8l2c
zu{=BHX~yFtRp~fCO(~hgZzfgNX<3f+az<6wSGpvYgP`-U>P@%*yOAmnaJFQ<
z@D5bBw8ELU<~a!-b!Et_nPw4ig)uC;>Od61{W8-(6FhD%g`3NyHM1~`H4r9Fr%&p9
zLH4oG7P`2
zTa6b3*K*N>P1DZZmJLj}JTL73!GufRC9PU_H8%X;O?Yc_ZNYk+erc1g3XCeoc8_ng
z{Db-R>0WJdeEhEG>j3YsUq5(D&Ew!xd`87zbKF`f*BFsjy
z&c(4I88hmQry2tXF%ycY>(oLl3pKz1VFX`LD2^t)0N1bd#!~^V;
zLb&J8(7-Zt)R^SnWh0CZDmVA=3$10rbQQd3IJj5K_QwN{Kd}0rk^ob(fTz`5f+k-(
zP4p2=3
z!AFQ<`1U{3dyS1IH&>0kGnu}8P>{^g(NFVTp8tfGYU%dF=&>0iXw#Jowvb)U$J%7E
zEp?)M%dQl@)Syt3QI+!nW|u^!%lbB-_ND(KCWRO7GRf0_fS?<>
zQRtO~C>08L-;>7B+eiJ#DgUBXgbEWUTmAOLcbB_bhTfsZ=iPGz>=@fU_zxIR(z+Le0(AVcfaE{w2!qxx!l1S)V
z$3&l{49n@}@-Q2AeEgS*@N~c{>5$s}&iXV7nb}z}Iz9|se4?Nu?m^pH-IQ{@O3h@C
zBDF&MU!)FyXe3lZ7TBbOzlH4;cuMn+iXwCH`
z;@``6hR!P?&oBk9y0}daE!g9|HoMVGO|qBJLT;P7BbY}MkNz3ezI(5d3G1F1l&#$)
zn25?Y^7gjh#Vj9z`muZ9ldKZ9cClchy#MpgCxe?iAdP-XFSF2YH1!`Q2Q
zZ(1{X%8+ZBuE1hAJZ|@NkLYI9slUn7LkbhC8r0@xIX1wUwVkX5*_!_?Nr1Qf^{9wA
zNrgpjB9|D{QOr~#2k(D9N=sFm84=Gq-1`ccR}3Sg1h>w+)sGkd`uan9
znAU&sPMTwMCtHT}TjGq@DF40N0-3ibnb(N&m$32shw!;`FnuwMh0=I0$a?akvM)Wb
znTD5Y9z5!O#&f$u$b_|ZIp2g6UlSJ({UkI)jCC+h;g{nHJ`P^*>XnDrY18coQ+NCG
z&KBg<8OZ&qnKpeVnusHsN|z;Hft=`4{Oa!0_5aag91lpmk`8&i!rBL#+$wEV5_>e|
zS0sxKux}TQO}x;2eOUUKT20jj%#UVZ-1^UgvaZ8}Ch1bD>;uJr`zRUKL)ZZf@Mxk6
zaeMfRAW#9DFa(S!novzKz)tu;F@vz4F?2Yv8-b0$r8IiGgW^KYK?hBA*V5*%sBWLWF9R~e94Q{Nqn&iVuR}hF
zbkVc`jgO}?t>j_eNV)ddI+tQud7>Aw*t`uhb~Y-$So8gK={1vb7NR>_a3_oHFLa+z
zk=Euo>R&nKQX8N4d%W?Ct_W{1$4AgQn`nz
zkO$9@8;B$L1wO(PsCj!($ZwV){5o}rCGfmaL@K6N;Sp&#H9bRVUyh-1x5w?!&uENX
zOUS=z`FGlIS8KVkyf%5kG@Qa1UOh31H{SNY8u`l-_>=~Iq1hZR^um#?G!*yKtqorn
znZF>AO77MF0!CMZH;#SesdwA$55Mb_O8FewlhH@BN~WB5YAY8)I3N)C$xa(~-R^w^
z(mq{rztT^arl{cI9>T*nDk}ZO9Y0l2K-n~~?Bo^;Ot$@*HZ2so^3z;^v2i$Dh=72=
zVZVcn&Ul0x@;g^9Zx_Km+#t=G0}0NnQvf%7@Fz`pS8wM=p6&q=)an&nkMCUwAG^~b
zx!Y+I;k>ar91k*Ee+{pvV`cSwLR0e$^3$Lnx;go4Fh7*+o2|{U2Jk%8%C!jlOiL?i
z+aX0vB#sw4dbs;vNoqHcDApb2n`;6D0T_onkeA#pI801TV{OJpM!Nu7R37auriS#s
zGMU_SD*vgX?v8L@iC=g4)FnE^f;O+0YCHPJsY5eRV5!D*NJl_x;$ywC2r
zoX+`Exy75$_?Ko72EY^40RE)eBopv5W(ps20nfH^olJTyFYBS8
zplARrq2|WPrXwXoAcFZzJ#l~!1$aIeBgBfWX`%qW%1oPw4)30nX{t**R##X#VqFY3
z1i+$zxC)Yptm-f6$xUZJ%7;;_Op0rx~O?f?iI>vsT%(x+65U%o4JsO4KI
zTCoO$4qv3)GOXTT)T9iKw@SHA2AtVkw-)YFM3HJCvhb65UqmdFigu{}_(F*_=*F3m
zU;M2oSqUeA$GSX$Ae^W@FS$;;7D(%soeNmW;3i0MJ^x-v@;68ke^1T9Pm%W=B8=*iyYatLboA@SvAC~)X6L*sRDTrogON1YLW@u&emY=z5f$`bgVZvWiMOK{
zO#|5YNSR#@_X|vc@r6;euAC)Gwz!+P-cA_QGQ2j`O+)1$kN5fy7$o2x_$Tsw8AX#I
zDz-tMfx^TOW5g3!qnV;5IQ#0X$aji@9tjqY$iI8JzhYEG&l`{ydoz@a21S+l5%N?3
zs9a3@?`JCL8_$fU#lMd@JR)OKnAb}Uy=ZJ4f4Zw}G&WDQz9-3|@+BthIwo3fW=IJ8
z>6y`8;OYusrOIb2Z{>}6)<~9aT5abFYAK_@=Ew!E9mKu8-`6IMKLW|(oY2y$L_~~S
z>8LK5pjjX~-!1iFs;V8D1vlp(#;eJ@{s_DGHJZXuPr_5)nCE`k^vyPrD$=GdDEu=<
zVt1TYOtUQe7#+$1avmha&Q~`2L?tCBbf>G+yQ%w8I>_$7O6ZxH+-@&iZR+iJ)o`il
z9W5{r5Xzof6dROBxD*)c^4|dzQL_df@CucdZeHs(lf0OU@&fW}T=qQyxaHh4!KSwx
zS39UBIA@s1D3)n_$hd5s5{bw)d(ncuS|naYmmj*~=+^z^%^k2Ug78IADFeqlBuRnw
zPM!oR{M^i?0Qt81r#qsK7zy$U!N;y#neSejspw3Kb>8gr@bc`IdHE-@#?N0B;;EYD
z{w(jm0y%v{)5Hl;bZ>qQ7i~;uLF1NvYAAngTCgxNf!j4}&^l@c1C)7--Q(evzRdBN
z>t%;YKL>-U9LMAGosEW|1LLBK>OPB!sjBwM@Tl8tT3gD$W$OoDmR#uSjQbg$Ks1@y
zk$)7t;x;AISI;YVkcRPrtQPM{@K`0Z#Ik0iJlkgx9tu=KT;2*pi*toR0qd*x-
z>7c2u)GZYo0OfFfQ)27eecZ;p<3?n@9``kdcEJ4*6CMgc(St&_mrjqr2l-nI>++|5
zFJDi*cgg+X=8fpTT6#rntjV4a
z$=61HA51H_u|_{n;71YRLRBm6&M-hh035O$U3ni94C`f9(s8sRbHIZFo
zI;vRf1tecGR)ZEFJ~1}l&7H5E-VYKm&CbCMo}SjeUr|<-D|W<(_2rGfTVa^sjX!GX
zlK5;?b~UPxXS7${`>N~mEB?FqAk>Mf@6>^ZYZUr=3SIL1)yb{i(WwY4P`h9a!g#MY
z@kvPe2t&zjen0%C6)e8aD+Tc$1C>FW?h$y6<1TrU>(Zy*nTeAljn=a=gPBpxN;f=y
z{750BS$opaJShcv%QK-HBd4@;u7>YlF-LW3nA;NqCgx#5vH$|%dLhR;Esd7GU0Qjk
zPRm3~M)HnO_~vCFNRP6(*Cpp|;xAci-9i~>-=8Pg=BomxI3LbN{g~HG3!~$Ww^*&T
z^A;c)5S@9UE|q-1dDb!8dWLGiY8;foY&LxFH7EpiF)
zno92i0U(J1wa3g|q;Mu}vkBK#0tA6q({(&qp(ne?mtQCN(!GU=eX;0|M7GKxyPH(L
zh~J+8$~UwfW;@V?pGP|E`j7t9nbwgZySvn(lkR%9boC_h-LlRIM|K&adI9PetbYkm
z{`8MUDzrWNv?JL87XqXcs7|y!lpec^kgwN=soGA@9C1H{APhFE?_m%M-rpKH@YU_;
z=NlLihG3ZVP7)x)xN)?L;V&h$MV0N3X8?c$P=}yPey1U~Y-@FOvxB=lBcEeON2`Jh
zmX7)J&}r=+wYl$e74F-xGMUbT^Ly16ws=XQOSsmo!OoE@l=M^j*jX<@J+NHi0PX=x
zCJGN#q(88S<^lhQnBM`XeB6PJ(#+<3Q%dKnJaW3qh)(Nl}@
zNW?r>LHpxdz}uOhF?5N{k74>w@dZP^7eiCV^{4=E@(go`P}?WrW(78jIhMpE@_3l@
z3-B`hC(Vjt@i%~P-&cDGDai`$Ta%0xo+gKj3bs|zAWcX70NTm#`GxcPR^YWlqv7($Z_X`L<)rr;OiMKHScGY=ZE%$ytW4?8y>0#$382ML<0{6ymH>zWvCjOD{;w6jYj96+}
zW%z4I>Z2aFB>5by5?88`Xc88HKVR4O3&6RTZ7Tj6Z^X@xF}
zE_Hm&FqVjR6DgBgOnOPR8)J}t#Bih^yxKDfkOg9!$X=_pLqEWh$pU`+a4Z}2-lMQ#
z3|HoF_eW%L87$)lXP!B^MBX82>9XG(WUtzXCGN?kg)Y~u_%|?ud8JmTeeQ83EXcMm
zl(%0W`zBRl68E<9HXsed=sH%7(q-$WWw8Dz{wDC=+EHYXp66d&*?)WUtDr2p62Pur
zhY}bAXm$tNEV6>g!b8PVFtV89fy(7}vNMcuqi)%YZ{udo{v3C22EZrMS@#TrlEt{~
zu@QDIuu)-aIQ?EhrajoQ)-6J+M}=YN+}|uH8yc~&Zt_$*?+=Y+CdklJe>7LJ*K?u|
z_m3f15LrpuHi@VBb<^GY26XX{=wY$9Bv<=WGI_m7@cDVFgRIaT4
z(-TB{4%_4Tfvn`+E7{pR`Y-RVkH5)Y7b$D627x~*H+#gwJftP6T#-YJp}|S;14MIW
z_0{(~;YfolIa^=vNySg=f5{MN7f$G1rYoZ`k3hhzfe1T>4f(IGLGxIQWWG&K{8gWX
zX8b|H3jrD?nv8CoZzFHvHSyFl=72yBc4tI(`!#>sOU`(rzu&O7KI6!RLVkyfUL!}`
z4z4XA^B1wllDN68E;;v~3PK&b4LT=4_r5Bigtb3$h&L4-k|32p^F8t$_lB}1=vi_i
zriaq0^C~hTR*Kh%0iBWxBlpQG;U%u9(oC=C(yu2Z+{FrK8l?#q?<+(K@tpUE?shZu
zy?=HO!L*<5f2$9|R(C-^nY2cW=pc@Zk1qo943_w3LQZ5)syl*cQKV3K-=^{|zk-tl
z`O`q5``Ca|pd^*PHU!jDZshQI(AAo7$`HxaSPi
z^@QPa0P^3U*6XX8>%gpJH@qBKWCb-x5Sd+KR)Q^`%hkItn?<9PuDB1^_)fL
zfkfP=`du5?=o%os4IvGSIt95Og5U8RfNklFg@VSF%o#`I=A9
z(4W(R9`j`hF$I@m@ZzH9UZU5=s8QZ8)@wmQ?EfYTI*!^ckVS)736iQ&vDa&5!m&e^
z#WYyM#yA=T~DSekH?MoT4>zR!E9D#u+UIYG+ds{gredzYfRTTvbvno
z8W|ZChlb*_JSBA5{apLiH!QZFW1xQ2d_y`h@Cn4fau#;yL*A!}lERmbzc5>M6zG4l
z@308T6h{p{PJBhObXdw&#tu_S4ED>eMDRd7_ft^3Yu@U~6Ul6Zu96rAMR5KWq8A>c
z0fD$7Q}rREf}PBSy2Ochs^Y{~&yOtrhlhib3(*Q{4GW<>A%JdRWMZ1zxxU&#NJ<6?
zEjIEiw9fXIPRB>G_sEbg{ID&p}CzWQVKgTxf-_
z{2%Spv*I$8JjD58V2Gov9^2|7f%O<{@K53a!V|{c(F-R0T|fH2PDpPy?pE8&+?E?b
zjR?9*r)X=Nu{G+0>+b$fH?SekX*^2fc2KkC?HdKwur8Rhe8f4aM*aIFiwH2-pK}FW
za@X4e>f`~H^^>}K*LWj*aon!#i25j^n1|5MRMZ{4ZCn)JZPHeNEW|nfyXm7lEa?5
zXp=ZD1pIJX#R(2zo)*?tuA(AifmiW(_{#k`K+&eUSx4LzC+!d?=TJlLmBU>69_=EF
zWwu;vd+c7$w&T6L9New;A_a~VQTK}TEBVk8RS8pPMN4>E^}iY!Vc_F%ad}8XNW=Mg
z2vHi7PL&Y<*2&7lZEgG5sd}}R5X?8^Hy(97W6y
zy;J>G9@%%C6sn=d;hGH|5R)|lv5t{uNn=;%vsS=Jp*&rpVW9}eHngr93F~M4
z1$--AegS)-K&TX5#p}MQVaiXYSr)conOg|Ck_Mo))b^>8qK!~GjH6&{gV{%dLq>sd
zu1N_)G_>OnVVAzIPrUtp(ATn05MJ#*yZ}CuPgxmXuP-UM?Ul3G$r&iUI{=Z?Pm4KT
zu!JgZXG@jqxp@{qnzZm$+3W*-ET4=F=5kNG!mhxV&tuQaySvZCpJss?pVwEVG}Tyj
z_;N!ULhlf_Gf`LgkfFD*1;{UI90!~Zvb@;sM?0QBt&G4!M`sJB!$EiK2z8Lp6UZ4e
zG=b1ATwK)o3{?P?!87{#c1kl?-(8QM>Y4TzqhH(M_noK=MW4(V9Qk^VV?&EYp4ooA
zLO6mXmwn+2z!Z}~X2*>STM*=N5H3S&-f=93&l^*tjuha!Y3}A%tb3?{nwRob@{>J6
z$?YDFex&yR)`QZ-M9c0X84nYJ@GH?0m^SVf}?&i9(27yMu%xrqh7m0U%6qbUY*O
z3uU@{YK1vsbG2UEXXz
z*X%t2YC&4V0b*+9@}L$R@t+Lzg8AH`GUw-!ai$TY?<10_hsQCo1=i``D%ZBzHFk7A
zhMiZBlmm|PITa#~c6Wc>2^+On`0Xv>BC&YDAGk&rgqKK)%I4S(f&D0Oj|
z#nW=Sw@~7qmzUS%jjiU#u~QR?6ogbwg`cfE;a-1z7v=JZQwf+IU^Y^WY}mFZF09;M
zs6jC3kSNCPO4AD#DwanaaymK#EYRlPFga>Wl5i+-<@A_2?X2HUL87E7$8!Q?(DI-d
zfy~kxmX+P}Ua0wg^Zms`?H^b><>*6h?(Q)rM5J_1NUXDsMoBd!G9LZ62NmPjq^&|I7ec@G<#^3UVPUl_z
zV)x!>tnK<9wcrZSl!F8ODj^oz3PwVAIG+A3CPIKhiOWsJ!Z#q?48tE;Vb#CMU@
&eKlAWslZhA5>-EV
zufh_feUC!MziTz&(0`l4AaqK%ZUT(Ra9alcpNh%ROsBgpvW{-4ZXoyX)%s|~NrRr4
z8qddWF$;NDwVGN}ADHTwlJaI0k3S11(Izp5{$V!RSFp|s1WuIgS8s>)i>>M9;N9ZY
zPqVodu=x9S)+;R=(p;SElgmQRh&YZHrsMH14c?)Llci#LSq?pT9ab>>gu5TalPHV&5XC|hyCY~mJ^Ai9|Z?`qRB*!7-)i12Tma2b~ooc8WV*?Et
zHsec`4Ksfioow0U4hk3*Gyz8py9VYdCqc3*6yHCkpRWxE{C?j!&?ftmw`ja9`_Pi9
zs3nxf+GFCG6NW)3U`jha3d`Ubx-wf%Xga_M|9~O@k}fwoJ`@zQ#}m|8eBXs%_$}AP$tX{83sz
zweZS5+{2PY$-&M-I@6^&1kx%|7o3&=0r+_5S&P_AR#k2g(4pV15`rlW|)
zWmQfCLX?M};=v&MsITeJgK!nP<}fGAI4RGLK<1bfJ2&Cx`!6}zz98NOkW1x}U_C#;
zqna~WjOb03z|2R-qc2@39GCfc{K%qbol(qHy#Q7(6
zM%@K<(Ccn6Tm!QvU+8pp5m+>)i5{3R7D<(LbikH`LQHJOaeSg1se58{^`l!M
zOrf*oACAxCxW~CiN-dwr8uB!Mf5^O$kh>AbxMts>hC*2o9~z-MwNIykEUHop`C3w
zLF6?76s>~+gOhPSxn-gB(4TW5Y>w`u8$^aA>7_U%o`-s_SRZG;v{Qo+2lTq1``nI8
z#=b8BU8~q}k=9EQ0v)4{sNx8R>BpjusS5F)q>_yW!}tT|)oRW`5gc~%>J-cDk5z(w
z{QOF9Mw#90U^e-vv})(@oOlCPE#__peTPHpImG=00$4nCELTiu2?*k#BtLeIeeaFF
z0n=R^^BfWxvxFHBAZS%|OPtcB$*(EIt4<3DuuT`C3oQmTNF$yZXW*4@NGKipnOMFGcEP^|b2+`pR
zEx*0Pa|{n>Vfs9{L|)$~u@27E*XVp+OFR!AJlNe(zs6IpFl*_EBD0sW45m;lr;k+m#9{caYrQRq#=@PX2G-58N(0o4`6noN}?f-98!fF1DjVe8rh
z`la3MH2ZDbxs{jwQ~xT_p)L1h6%@{=P-gD2J=oab2nmuxYJp;w%_I|6hE8QvxDU@e
z=C5fOrC)hT5x&1-lXmCJzt$*sPm#Np%X@1!oi4vULq7TJ1P|n<5j*O&INh}V#G`|Q
zURg;_$Q*W?zh^FbW@dq2-#Wd>8@UU+V+~Unkd>U5lgW&d3Gx9QIl|4}HH!nc^(2~V
z)WP-K{<^}pL;r~&0AQNRzmaOiz;pahRrMu_CA+0~K3n|{=ygOKq4GzI7BJ*M>DVi^
zVQ((^u=lc43qcZLrz09vluJ!peHs4R$-%o1;
zkuxZbeNy!LoNzDVux=mIUEN*!h?f==HULq!g}L~(+^lH@C4>I&^qz!-GV)lm+U(RL
zD6}{bAeXi9w%F+Bk5N!EUwiLMC#W!0sn|*-fx@`sNV@d%S4c3BP8w#
z9GXF9|1{x1{zoqSD0fTh_~VegNF+}|dRbzpTOj5ob4fBkmVl3N`pJwtp(M{GW0Kps
zVY3!pX=&-~UK?Q>(3P(7!vlOG$uP>g%Ogt1J=D9Ox<GGqbUY5eANL+5?*-SiFj=r
z(0-*{EE{txbY<^5OwW{_<960_^qVulmH0H%-G4A|7L}*_TlxeFV8Gc4m%)}
zF(>ZVCzY$IffoJz*qY~uYlxp~jI^cxuxtKh3y)#@oRh%9RhFI8duW{~7C>>bf?#nX
zS)7s!T_jRARf1{|+PM3eh*|0qvHBQ5G@&*u`5NgxC2GvuWaWt6Uk54Qzn;QwC0i+8
zv)%DY8Blxu+S
z9rLPl2oxADnak>W(3opcAZ&eT11CQ16kp}_(;EMBg|kZ8zzn;yF(*()g`A79{5TFitYnAID%>n&0*sXYrL#bMb^+!
zN0hiK(BE}ti*@{gO`(3`DG<11uC;pk;iC5$n{Wjw#+hVQ7+duFf2srY&}#ctP`1{+
zRyV32Neo%IoAs5DscW;2mO(2EAY}l~fHbj>yB-u~MB_JXvl^9jS3l`%LcJxP|PM!va5oUB*+1EuD6y|@16+{2S+dF=5>Oiqt`#z)!T`_
ze=m4k=vCRtouM6OgWxN`D2N%~i=Cc&T&6Ranpw3ulO7m8lUFiEuuO0`H@C<$a)$_7
zu?ORGLjLtM_RNfJ?{bjR5ds0A!Uxl9ZPWQ%bn`)GJZc7g`S~4BT4^!_|6QKdG^TlE
zy*cDpEV6@xbw17ap^aGjH=FH#th_?v+jlGS%<51Yoh*=JeiUwmy{QAaD)9W)5g+$Q
zPk_w1pzI00{idLy@XL=mLCfC6=-4({J~<>Q7?wz0A1Q!_PVl6$@FjA7I`dlz)6+vB
zWcb+aH`Wi6zLRTY^J`m{n+4H*pZ~hg0=h4-spi}5Qju_e_0}kNQ0$JdWzpfZyxDTd
zZR8Zuj`JW-f*o~>Xpz19z2AMe`nIMgZ%dhsSH0&Q9Y4(1;xjmQ2h=|V8RifQZ}+OX
z)S3*UVKFvm6i`!BgatPg@VTUiB}(0)AccUXwrtXkWU5F3>y9Wd#`l&1#l&tZZ4`#}
zJ1hT6$Wj3o19Y0x0qdV$^>n?~`KQUzE0Fmoss0k@PW({}MWQ2GR!ta|?`B}QBl;_J
zNG(z%&=GqI5If+&Od)*QteB&l1Er~{!(9dAo8M%!F*%n|MVFHQ5R=4vC(Ks1~ApnSi5aOo)Y{P7Z%8QO6<$gs;-?*M>0y
zu5HQJt}E7bgK~v*1P_4z$U^J&eW|_4lc$9CE+VmUpeV>E#rRT*cU=QIFbwB_n&;0jL%7rY=^RtH-3F85pPW(}!P?eMPre1_hS*+dA!B9dv#J
zH=D`Qt!zvomo|znTuJ;6{g?kJlwNCveW-s-JNIj5^iJ}4DGTK9&-|`|>{lf$p$yR7
z$dPxImW4_gy=A7L+W9AO&%))kcQA)q8<2QT_47(JWAD2)3grY>{BhRtD0=doi!}U>
zh%s>>UIEmB*_NB3>Gf{gkPga-YahEO!dswjPoXM6fGnUj2s!APkdJ
zp@+S+vlsg@Zo^Ouk%{%l{UNlvZC_T6skbaI+KZ#Cx$cNTw!yPnCUagrzH1^O11-53
zhc_5U)4HvYxlY(cweB%I>-pUYO_I7pgVtD6>F`~^0##9UbqLxUl7uamX(b{QR)Dro
zsW^#A+Y!a$OXz*^2}b6OewJpGj+rrI-wz7fj9#>7v;#afZhwb9spa#5SKKKW-6N-^sn@&)OutFm`k8I6Dzylu0AJej+caMv?ZIynVFCYh#I$r
zGQY|1*j`=r%6j$g5^u$bYAYs=K9QKYNkHGtp8twOIF4i2Ryq@P_Xr|1f0f2?Atc?U
z0wj;kN0?cX0e}J}o~M<&a1vE!0pi}dxh?nshU$JeJ8VdvZV>wWvYMnbJv{b~Qv||M
ztkRaGxr4C12SR|sEm7-z8evG-2C6KJwSB;g$JGs{uv|~~8laVFPV2oJ_#d5o{se?2
ztGb(hHXrPt*jdZz`O6mYntusX{dXMD037K-6aZ~vEe{66Y0+l6;QRuiRlgq?gZvnYUwcgErEzIu^uciH^$
z#)?Fzb3t)PFUSLd!mtytRl5F8BxN};i7A7xgjD(3t2!7RBEjNpXRGhZ7K%Z#F*tna
zH#xiF7$uzUF;FFgw3v*78mf27{>Se&=-!qq6
z_bv*6*ij2{xM=li$E~aBbHei$-CSmf?xm38HjJ
zQo=s_jjcKANqmAe$b@A*#8QR;xhK{GCxQdO#b27lnc4+fQk^v2xK^CD6(__3gDPGh
z?ilyk4?0#S>R6dDj(5MMKW_s^0AeYECrEVtsn-&ez!x)^_cC{ik4fZb&X#l8LMn5r
zFs?lZS6p&Cg&P@bEFUs$Er%}HczaUr7`#aKl-)HFB8)M(-r5-i2u;ivnfr}0hUhtm
zLi6q21U^BJE|x7q1sD?c8NanBO$Mht%Om!gt^Y$5uLg
z2_A;ekf{MInLV%(Q_;=dmnHwaSAx!L$?YdcocgwRt?$q4#I!MYeZIvDWoYv~*{`kQ
zG{1@5lVn$w9X}2S!ah0`U4X*p;-&aTWejdkUz`Y*?Co&htm!DqFyKA8TBDfQ(TnTZqcC7(GNl<8%Z!;+M=V$H9DCi0zE+hBi;W&IQ(??PMsrM47R595O&`d{GC
zm*y&e3-JoX?%qm`lAIz0sMtsd$f(lHx)Z(+9dB4r{@rjsb)6@HLIWu6x&{A^O@m_*
zKsIM%9Gy!NG?YF7eGoDIW7fnM>l+nYSz~JK$(&IL=wq_rFLvd2!^CD#A|Go*)XAw_
zz!^TZ16S=kvI@@-Y6UG&&HZc9_EPjQv7v5p?UZZC4|dJNkKu*9|Bl%GfwA&pwV2=s
za~Fb~2S^ZOEZyD;M7N2T$vHN#SR+x2H@^n#UtI)-fs?6hryr5M0Z4Q-*`>6W?`3=o
zdrVm`Sp!^>RL1jbzI<&nUFW>K&SJXBtgM7JMf?G8-}y`sW`i=eqUE(@7KV@m^`?FE
z*rY$5xxUiHpja%GFdOYlj1u3UM7VHfccWub5)vyQ1q&S;Jirvr@W-j6ch6np^NOX}
z0B7+^ug`W#D3wEHr$*_-uNqT046uOx%eN#tDj#HC=G$^FA
z4bFj03d;H_M_h|k#s39h^LKHG2z8@Fs?F6FY;A6aex8UKpZ+h#BmP~@k(E7eKG3axov<%*!q?eM34wBC+
znPp5Aoy3gG=yf$$d4+6U!Ae`w$1;n(Mmp<2$$#>%1i#sTfZeV^EH1OSh4+PUv690iAo|C3d&SSPzKJ#)M0QYZ<{>t}x;?Mlgz
zCm+)i-79gQV_+}LGm-b?Eqb_IP`zLG8W4E=)^OwHQyP^rPp?;k(H)mot=F6)@*t)K
z4u-P#Q#vaTQlg1Nu+j0EQ0
zO|rUxlRjolZ>*4e>XhFyn=o1l@9;+YJ-K1ME!*5=nyf@I>7u?iQ_o!f8FxFI+~ftO
zOitN3|5aTG@-)v=Px`j#HHm>cxIb%%CqjJo_$wsTuww`ca>O~>G4-Kb*
z?JvvLP9b3;fgo;sgA_L|Z}rXy1|W&xy7P9ZuAA;a>A0|7z-6?1XdStq#1f$_L;^(@L!ELnq)1GA7D7=x&
z1W;!?mAz!;ydHluwUSdSVGNfXkBQGGC4wcdB>FwU5y}jIr_nA?N|Xv%++{j);pVWH
z^7W6XydnWa+Hfx;1U-4{sz9t)Z{JsYsP99$TjJ=R!o;
zXW2Y!<7fUWwW|T=$B}GiZR*3r(f4b)f1JXtlNI2K8{==Yuos4d^8c9T!6pv|&HK6>
zLolE8efb<)X}s}-z=s4HbnN)-Uaf132RJOHcU+h!oc>MC>2bxh6V*02OQvEwEO!d_
zS|}RN1ugt^uF03qY_igmC#&K_v0(>YK-yCG7a+N3K{d_#YI)?!kkuH4`=o>6K%Q;l
z()VQ7pFu#gqoyMcyG@(Sw|~MJrA=a;EDS*WpFp$XV^6vWm=>TEji`N+Qhu3b95lP<
z8OLjvD1fH>9zPVC^u+t-IPh=@{sPA7>(-}}8#n0o#9dww;GzzoOI|54jl
zM^*KG`(hvpDqV_zl!UaDAfbmY5dU(2HQPx)KR|&v@_-zN8u3X&`#Ub
zA*$x?Q}-mL0@nWLrq}6Hb1gl&{Xj0EckEbG!S{J_3V~jezPo#C*oDzU`*(jZFr2&A
z*NYxQ2?B7|-ny2h<~xYW2jns<0ZP^%
zi5}>iweB{YxFh3+^%9Sx63eP?hYVKFwY2_c&&SFy!Qg(-R;0y$nrFkLIK!R&&PYu_
zS}SoK(&Mrlp`Ab#rKNoL(%bXNOX<21($jpxiCb?
z=IYhm##Aij*IdrgC}@HHo;S!6WMnXP#tSGf=raEdRG;U`ewX^Y60P1vLoZeKG!iaI
zir^(uACzxZY2Zu5UX2*$wK2OoyGB337~zd>_eUwkM{cY4t-E3i)UBx`^Ciy~@Cg_D
z@6vON%_{cnfexogUGO}@RmO%^U+AdWfb{ptLjvUIQHD_UVwA0wSbZ
z;OU^$)1?)BrchsRmXR%I=5@#nuIFcUQIgwm5+QB062o)+q<>bKQUAl;SYF^=Zt
z#eLO^-wn0{4r2&%RD0E&mrqj0Lb$ZaE)jwuGqac4Kadk4Oxw@khx(p<1EI??5`+);
z&lf=_Tik-}!81Yt)63XjiPkej<$bJv$5tsTUvIW?`#o>mJi4$2AG#3ZJk
za}8Y&athS40J;8rWXPB`>`BH1By@Z2haa-ZrSm@ClZO@Zt-p1lHouhN#65_J3L5>-
z!%?^jkX5Ou(?RbyMFF!wO>DF9OFK4`lg1)VQmFm8n)CQpjl7AOopZC^f+H~U75eO$AhFsIFxvfjYSXZWO_8}d|
z;}^_P8b%Sm-k$~szh&)NJX`A4wzwHZhMdDenZ}+cXl#E=<|fnQ&885_V0F2GD!%Xm
z_2vg-ml+pNXY*nAC`?PnS#i2^V3n>3+Z#mczfAApzH%S}fFrVzX!S#i*lK$-*Eer?
zt5ajs@G5JA!$=|U0(o7H75{eu1~2j$?;P5;3w2=#bYG&8?PYhweB-S6wgomWsy
z133QYJ$t|g2ecZPx|XDzOJYl&;q>03KO
zQJmew;)e&knS?soDmZQ!;D4wQwU|F*UTT=Hj=nN`M9Cn5VZUz;s&%-Ax8%-odG-xo
zWu)s$jmG%3;9im2ygTJ0OJVFSX&x70XEVeD)P~Kz*B>G(db5q&@w{RUTD4|Xchh#|
zFJ~$SWx&yd#<7dQFDO@9p-0s2TnO9nalCoos#~iv`rG;0-s4JrutwmA?#Gh*%tk
zs-NZ9+o~K>)m^Pgr$4Q<3Nl3@@_jtKgBF1-R?aJ3Q|(Xvl}tyHqbO;eBA#
z90(Vavi&r3n)bZ+;&Ou)k&>Rm{d_1{V`ynPJ{Dr7gx4lU)ZURAqyOTY8C<4y%5Kh=
zH}yxiC5cMrLKk(f?otx&r&Dx`NsPWhS~H6VQWq%FVUAwRL=8G+18ibJG*vyCU1d8c_{)#jc{d`lFiL~M)mZL(Y8L5KUVqA60C$iL}Yz0=Nx(2cog?GE?d0p<3&bkIpWhi(;uSRI-i_G$qE{COQZ)f(@_&+shqIt
z{Ast0bHy^k(SsppSOKl632vea*Yt4ctRdC3xoQ0PIKD*AE7EiSF*S^@23mG9zU_Aw
zIGtX62?P-~4f}zra`C%bATy`Pom*)BJAJbCSc>xnkEomj%}mt2C^t9peH0dWU5zvp
zxm%po&(2kdH-q$mP2jL(fayP+)GdyDoAzbbWQ&1y)Zn+FCiEe$Q$RKxl2{HkZnzWp1I|Ey+iqzm7~Y=7>TON9w28ozc{|o
z&L6{ao4x65AgkYxe`u62?StR|La4uTfXp^5EAONEJ2ytY4@XjSmphgC*ciAOgNFLD
zWwX;MW<+wlH?`kSlW*0WQq%d1w}BQ*@akrwrCXx9rKB@Y*%$(p)wU=NKjPe00+kp<
zAV_<14VrWJoLth9BLfP#!FAwv_Lee_AHtSQIju
zzBbqPJkyBGx$5-d)_@d25NpWHn+~Q_xQYcO3L`UpHjU*n9-e3aRHeXZ=Xbvuz};QK
z$_&fPLxCO#pU;Z->Ye(?aacASNi&+f>=6!68T_DjFpWPOvr{l>%*3-zj64gppHmPcT
zsf}Q*<=Il$GDQwLm%cp=Et9$(^^VCVWBAd;unF0;3Vo22Cp5>GS62I1#7K#s+zh7niHZZ0!PsH_P0VJ$B|Z!LS7#erJS*izpwcY&A@bgCD#KZ
zq&McZ7fe8k2bcw3y_>?%_%Qk8FPUlLA(mITyh$sjGx!iwcIAfDXpC{>*NC-Okf%J+
zuqvrZIKcj5+i=E3Hux@4ShHZ$a5uk?h1iQKExMHHG_qG^omVUl$|P>4woyx~k}=^QSz7FQ2lrmo$H{0~0C^V|_5CE?7TTG8JVz2o(RAGM@xZ*XJYFtOO>*F=Rem@NFT
z8X1ZzAz6W@3XmpgUY6k%_LhI4>6f>GR+irVyU7%9_eQU&Q%4^Wly02^T&)t@xxe~!
z6!oGv1t?0T{GjW%F{Xh(ccfc?xdLGkBFY_uLDPCj`!@RKl!TdUDVzw4fMPmbcfwgpg&`|-N(t0NX!j=SHj&3L`uwOf0FrPsNu}MOUpBCGlNw40P4Hzxz5DM{do3>
z_jZO{L34N+SE3`hh;g_~7;Ny73&yRg`Ejm}@UjC9Tcn=#^RnF@UvW%5^&fnK=VEu{
zklS{kAWx7o>)z9CJ@x-g!)FLFOD}UkHcfEqaDcUpP|<-H_upBm(G@px2UeBjT5CfW
zswJH#vnaKHEd9MHBQ$q;2z)KkBiZqF=qQmojNCZO|9yn`|+B)`~Y9iQ`$G>HGNo{szi3P;jKmq
zi`XG(vq7ViVpvM*=LFv)C?#M^>Tw-!KeXg!bC+4GKqL~F!p9-s+*~E0Bz17xjvcl=
z&B}6tDG%2W<_sa_8CK-ks0Awh1JF4KZ
z(Ojy6+~?s;2?z_k!CDAvgZDnQJNoXN1;vv~?G(^f
z_VW5f0`=Fq$E)+f8Ujp6qX_Q_TdeF)M0O3z1EfY@Wwy)fDcnN;10$snEda@yLad4_
zY<_q07RZQ_K1|R)pBnz$+kxVEOyqa>!#})0=BKxC-_yK=xZHt>x4Rxe?#kbx;{hjlOp6eSTU2
z$L(qIvZdr$&6Miwn>D&p8_}`m`9nKe*+C4=QlPf?)QPzJ*Y97>Ti<)psAOh;q-9$!
zEmSR&qaGgXZl*Re)^iZ7v{j?WNEjhQa$8SP(uS`0ab}*2U*6q!dJoA|)gU90AUJz5
zy_0-zh;B;51(O$=_Bz3xYHrq-_;c4I<(Iy`Xp-4@VHb=0HPlzw4U&BEz75AxmfgzB
zA1w^h8aB^U5pAGL>w|YCtd-evyO)u(Wa%Xh7+%~rM>O`-8l^Vh4&z@P
z+9|TBQ(
z%*oYj8{lZy!hM!A5ht#d6icNv(|R%xxh24Io@JmY;@-E@07hr{xYzAxGxeLAvTf47
zXv=+_UWkZ6{g;f>ql;4?YRnB3P(DRnD3l3h!W4*@A?KmCmJVN}mz^fmdF4Cx#T>(b
zEe*mIPt*5pXT$A%R+0bMcAokfNywyq@A8fo+c(?4-F(0WZk3Z`rO~Gjo)fk<0=d$f
zr-61WCEAW^Y?i}v&#?_6o%_EVaM_1O6wHQnqJm)10BL#i%lxfR|%_+UFtliJ;
zqd1)dlO2OAh^@X++S4_w((K$wuFyyy{66bPRTegDX2BIfph{mSWx>e*VFGH$=fk9f
z{`67ShMJc1XPpPAC2{^tJtfKF@TXj^KkRHrBlX_X`TsI_H98eVZdjp7`7Isc@j)4OAQ4Ux?6v?s|orLCwnV&P;8VmaX-JRC<
zvi#5kw8^A4^$7>?SJbYeD^IaB{1$qoF~F>JvA+w={O-y7@5lnp46PQo?5pPG`loc8
zkv&~D`%3S9N*U{z-zPhV_SjmBKA#b@bpzC>vQpr3c!4ERBq}f}?S70m8KXeBzS6iZ
z-VB#j9Y5RK{ff>jE7nWf+!y2`LmC2fAn3%|6Q*yu-tV4pK@zD=d#~7
z*O4f~!eRcFhd0ZVCwvA<|8E<_7+cP|%s98N@w*ncIyKq6l4~^~g$1t6mk-(mR?Xe+VNg?}xI^bNohw-J0r
z{EEInh3yqBq2w&%P-nTmsIJl%CBxpIj;6OtXqjadN;Jd04qF(A!tT}9@~%T(k^Kdy
zPT1Y#4JAInLm|%2Cs46mw>4ZJpmHu@s}_p7h!v`va7lbp=2G4F=9R~KY%EYRdkpsA
zJ@{(K_-$T&v%_!dKE+$-_=MAdH~kKZNY-mzLDXdm`kE2%QI^wdbnk|kV`5&S=N4Ys
ztm**_5fnbutO6;oNwd*cbXmp^E~Ol5aZGIYr>OHXvp0_Gern)3(7RfuzZMg(c+=qz
z(2hgZKEBj2;XMPlUn8h%(&0GUn|}XHe^vd}#`hvIO6$`h^tX^O1Q}8)eRJvWrTcPC03;0tklVtcyP9+D
zcY|L_g>lbqdG{awv|@&-KRB9O##!cp@i_m0D3X~O9!z@K1f`1mfGvR>>3QyzAA>Hf
zz00NCIb#|8)zT~H;x@JN1+K_ZUNvLv4vg8?GuIss${#i%3UHyHh+I((u!po%-*#5>
ztCcutgv&Cd5ZpF;j_vaBc0_xPT>JONP7&91q+d{B|D$yCb*7_6kT)K$qic|N{6A-Q
zQCrSFUzZNWc7G+KqjWO{ALdM655!v+d0J$%Pzy(#TkhUAtsHRSQ5Mm+&&!zx=B*b$
zs&-n4W~T8ZugYD<7mz27xn@xVsvM@(_2HA&xBP`7RS#AaDu!zojnaRIxsk+Zso0j}
zx)lVOk$uQ(VoLgLQXU+_?+n1Utl9=CCg*LIfF-G*lA*
zxp+4~E6-!rgA;5^S3hz*OPn^!vB5SzxS$fc?ij{)9`m{Vsdin=r>qGV?zgHrUW*gk
z+`^-Z^~`?T-yAgay3)ocVQv*d`&dvk^zL7ji~{QllBHC`CP~G~hG|`ub3>^>bXGav
zRqaG~5ORr3?Xy&MKU0B4B$d(O1Q{qrLE8T7*b&+p%`O;!$7OUXACc8iGx3>{_u>?pmx-!^APz~4(5Op|nd#z3
z@u3HvHd6uMM~;XanR`$IP7K-&UkwZ%s##2}n2rTV;UPlWFEy9r#ebRo(}$-MfOWx_c|o)!#awScd_k{f?hI$N}1=6
ztEzccemU~nY{+QBt$p8=Fj^&XsZt@1@w?jSCv!^NnPyHmF7|IWewCst*l-N%SUWN>
zF3r=sBb91#Up@v3H7QJe;`8vPr+gEESq->oJ-Q}lud{)H{_1^fO!-ox4_cQaMQZ#g
zQ->c3>6uvF}*!X-qONHp6w)Ha-jFOU~s=68tEj2kxY10SOElLTm?`l@5
zQoa6g0nR&39_4kMjj5rIi+$4oT$1!byY$vFsqZn0ub;}AHr4J%AM5Kqu7<=?c`)Q2
zlnh-#)WC`}PZ6{A@>kb?pnxGNU-#tMzFJrS<)ZzXGj3?|@c#v1R;
zzl~i={&v=XGVA)BS+$K;K?zUeX#4q_c^0AoZ}cV;3>%U4h*kWlNx7I&-gj+=l5Po>
zkv~?6#M^DnWWUqA1)O%4)d?w)K9@fTjG4?O@Umf4m-6Qz?iUvn{3FJ)!Txo_)$BB&
zydJfizfcg2S`HPn^g(`g$>(jb1G2%
zg*_;~Uo)9>eeyA)S61)G{wrVq3q8-OybHzBdThPUy~za5G~8Xt
z(5^Z&Pm-4stDc#*cx2KU>|i5oFi)B&=9?9>j*e!N!F>J_uk4cJ4K!J4`(N+0Za!>l
zvl3oAihZNCeEiw9Jmjqi2h~GrX*!PCfw)@{SiG!Q)H_obRVcXNXmiU5B{P5s{b6Gp
z2NYxtCoEicBHknA5t#USIY#>e7cYHjX-1lC!-cx3z^Lj|&4m})duUK<3QkHLq-jS6
z4s)6{0SZ`|L=5dQvR2x7l=~Jtbyg=Km&aP~QIaQ8DA02RlKc=A4duCJ9$WQ@Z0_HG
zx7f(@lSHh__aD;VO&LS`uC}b~CMefUy;cu8y=Yw~{PxQayK;(39qya=I<{~KNr@Bt
zZLptt##HS;;dkItow?!b7YgzTS;rHd8}|e7`ls7SsxVeH19n$CCvOJ)vUq?mSg6xN4lhv$G64ZUq?^GU65$~FTX%BNKovurI3>m
zrgPtr=?loTf6XDf7birm)a+yC)15~~euEC0tSIb~YCBd+Ubz<}L}z12(>^K(47D+)
zhKWO9M94B_VRA~G#yJ-&5A_ndWJNp)RJeZ#v1^3>eyeiny09!!SyO~gs38*~r;d#;
za-2Ok1C;yJ{g?!c$W|*8%Sh~GL1Ds#Mg}6%+^=)rS>DC|7;t!n#pX@2_sIAbk<24C
zsuvC;Ro~K|=M4~s;^^@4r9$gD_q)TnQ~&XdN{^BSEGE0Vca
zUvs=}-9mo?QNMdd%3Z}JQ%=XUDTYKtwsA>#56*fe#GAV0l(JjoJ3g8FdRXmz0;>!syX}_X@Ex;j1X@qZA7VhS2bY={TuvA}xvY
z-st69>z3CERRksH4)YwlN^G|j*fwVEPRp*f#E=)gp3n+WI&T)FRJQg2R50qx{o2^_
zZ-fQYhy^1}-;;%tLmGt`u2Y=#dvvQ+3)IjOz9k@F1PZQ}0$$8&xOtLIX&=w?(tFMR
zCvv^4Ep>uc(dGCLJNM|KH(IpbW0L(z-PshKz=k$lf$k{@Vb1uf`A2mU3(Z*;snzl?
zF-1*qlH4W8#1y!!T@h?SU~XsT{w`bUUFwfq?)tKTHa2d7sl&2iz0W(t`XQhiautgU
zG-!d~O>91#-wYU&4gHfHeNd3GExCY7EO>gQ;oRqIPh2%!sWK;y!xWt?XaGU7f3CrO0%w+=
zFs?9+gGvr?Z)g4zQ3VqfxNR?v2O}pD+;H|b*vdnvbEn*=soApkfa4DMIViK!ZAJq^
zqLs#zYU=oR1z;fru|9%rgZ1xkridwx3J5XXU35G51A--sW(U
z7vV$4I&PBg{6E>){OI$-1jE@FWbh0LX16zX*i_Uneo>;k!F^1ZgBJC~*{G!&ifQ>t
z;Y2Ebvp$8ogIqyUq0O_i%Sc;BXPGQ-t@AA_+YMq;dX3DIqp^k~?d}GA8qq*NK`l45
z23>s%K1UzBm@%@o5q@zCgdDk}+
zGkwY=xlhy^XtUBv*t0+QSrV4~W8R`M_8YJ;KOa`%J>V^G9ICGdBrBq&1X>8+8UQP0
zJ#+X5z3aM~zPsMmedgN&h{bQfU$Vdk1Xg0{VU>xX^K#j5?VC*EoxZ?h#*V(F%;Kc2
zWfCOSk-JrV;sKw0f$_9MlKg0~dsRmc!Cv2H%++DmBi|wIuv}55Xu>i!zr8E^)89J2
zP1myNU+5VBo!y1){jpW&Z(UA}i{^>-Q-HH_VeCLl18n~Uy`pOY9riso>qM+uxjEp<_nwhRIB4pSgO#!0C!OH+N
zne@oB&f^P&09$YsfViwE0(ReIaFerPASrNAnx+?{VhHwaZt4U|tTuDORw
zuGEtMQFamh61107uqMc(g+1%9R&;WV7V#Xr!zqUNk!-xaIxJG4v+KAZWRqY!E&P08
zr(1X@4QY>-DOt>5xfQ&)d8n4|IUdMgXY#J>KCG~{K4LzJOgE)DyS8nP+4W9a)@
z_5uzcFVwodX;J%-rPHx}cJa1p8M_p>@_-RVO5p~BU#V$N;_`_AKmRlsbN!w+@@u2_
zAXQVxwXwEh6bsOL`%EnMVc-}_jk!g?nVxl_@^5K!K&|{bfq07iMuRPU+5Kr50Mzt{
z$)MEGwR+CECeN|Sx$Zr>jg(POBqO`6
z;h4|&*yxp4{d}8{ySp9he(H|o3dB9@e8rnbj!?XU$Qy7$
zqK>rJ!n|r9w=7<{XEORwrZAO^3zF=Xe*pQ#M~PBB71c?NU!6A&Mue~b^b?}P4dzOm
z!Tm8;ns>LmSxKK}r9R2+DuQdZ;ksn}#dQ&5=)fqD*mo
z>Jo(3?4C2Z?)u|<-TMg)eV>kmAxKkZqFkBiM@DdP;)}r
z!%i>lvC_dsmY5~Pk)i%w>*NK!Yvy#Um*MM$x_5W@~dq*eey+x6QvWVFo>~
zj??hfwy1wtT6oR0JWsuWQw7t#QLXUg5)3wpri)huPs}p_#{jT-*&IQj0U|%jpv{s&
zy;g1rx!AMe>aT=2L+XfA9Nxb3#U(DcMfZ>Y`cCe`;c#&9bm6G8l#l7-<}Rbzl#t##
zzgz>*RlvPjHYkGic>5V+nOKe=&RaIJ!VkSK)bH!5O*?f>nv~Y^mb#OHOq$gp*kHlZ?H}YeV}!w
zHneEaaW)|T427P%}z0JDwM*F2ziXQV(-DUa4!2MZwd
zUf>V_AV#leA&+W=x@i`Z)<)xZzf&clzSE<(g(2w<@Yd-HV9KMcDRkhqI))zXg
zMt_cLKP_%u=9Au!<60J`jTt8X=+uCrZ2jQA(1+-sv;sr%*Dq6I(lf}+i);+N**k6W
zo#f4Qs7Y|b#QVz7HA#x!3EGEB?L=WEjob13wyW$6Duz4N1R3t+Pr&A?N06g^*p==^
zVIe=E&x4pVSOV|DFWs}25eU(}zIN@6`vMtj7l6kC;T_0I4
zc5GHVFXK9|!KGHb7GjY{?f*|dH!9@tw>`vuvP}Zr7-{IK8m+Nk_*`!*9}`iGisV#i
zc(z}q@p5ZHFK5{2{hIKiOC>ZdCzg_@0)N*eS?fd=?4pz&BXcI5tKh}lZ=y8`_F|SI
zm*w1nOy+G@!fm*hiiQZ(Y+m6}M!M}^g-qMSx>H-jKD28!T!ZF3qiY_m)+%+Ape71O
zT)26K=de6*lq4R{jXYwnryN@IN!m+bVC6VeTqw5DJ-=CxHfX8U
z5%jmHyfDC*=-c?#r_9fnP9eL1R^P@Eh@NL2X7P4P4gkdfZMfr-Rok*lO?iI&&;Nw0
zOgZ15Q$RpsSulPNq3ft?km7y*^B!0s`r?voFFtl@e4zj7>0BZg%^5m6T>i`gZlBtu
zC>KS8sup6EO5_Quz2suA$jPAib;;+5f`Cxvg^+vD#%A2Zi^EQIxZq|&Psb`g^46dH
zUR`qGpD*EnFVWL0@&9o}8TIKi#Bmh%dCYz5#eCD4Oz(gDj`Z3VTYG*wo6`^^P}t0G
zlBd;wv7J@nrb4cT3HciRlG$OWylQXI33_q+Dxu0LLA7yDk#hk1B7?VUO$b#+5Ca^Q
zQ~n0GHAgn!`ogr)^)2ghk|={#;Lvf%l{+8y8mXzcav#v;_MX4dCcqS
z9|OMcAmRo7NGsx{;L2hcpaHT|GfUM-A4ZhU!VMyDNIXS8?nZmF7Dv?4CW2C4e>KojPlXnvvcwr
z!R67}?RYDx)x4l?-Z4VQ@o>+&6YLjp@Tt#Y)dGnC$;&w!4s8yBS-)Lnzp83^f|!UR
zns49M5cw@P%Vc4M8-YM~5f!-gx;1xv@zUVB(cjW@BWFH_cx=-nY5;%UTmOuI#}c4V$W7TCP*pab`2oHl!)x;-32&)g&Lbvuh!rNy$zE#c#5>yK4W6D1
z!40G3j*gDf^CQIAT>+f+%pUW+Nrb?5aA=5sV3$HM?6NrFI7JF(wk-80$C&A_UqhXa
zc81C4?0K0don|g=z@sI`92^{4%Jm95Z>H^iP|1VoJyd1U8(sn2Quxl
z^715QWiR}3Zh@S99Y$eEcxwMLczpT+V0vOLgCxlAb`a?7czK-yD;f9$=Cb=R!FmS=
zePVe{Q;Lg;Hk_#5>DXKVtI|A7+2l@JcHx
zFcUX5kTUb?@o%dWe#}~yJ5J=-)bQ|IL|~^9+bt@WXXu1`x#i_FpQsIf$)OnUoeKksSn
zsRJ_P=CZR{Ve&-&nLG)OlM=TRS~Ti6NkJ&&aes#5-bkzHykm)fAIvu2PoE^U6tE?@*952Zl}q0?ZT$ISg%~VbO~V;z1aS#0>>ZA-%4<3Job`T
z&F{vdh$Ok!th{OWz7lyVB#B;|!c6Y|nLwa`+R8Vq^Qyd@zaf6^LzX8FA}0&79G>7?
z%LjUWg-#pODaV%&7nMC5i;X*80NrFJx1_t>wRR+=&JA(2xhqixj%w{=0X7dH^e`@1?s#-
z;Ov7)lnHv?GcGMa^&{J(Q{d39Cw-t+ROAjxS<1f!(qa|+gC#Riq=+FTaJ$~
zr?5y#Nh?8UXrpH8h=ZtA-(sx7YKl76NSTBL{wD3}DlCcwLFCCN-f?}gJK+d^rYVh;
zL%ODD>vM>dFbHs_*49#a?mBwDt#jQyfVX*X-Fhu{_yZr*i-KWo+~%xZKMns?MOQZn
zcvUyGO3yc(9Ie5y^a^?$3Ubse^OTmCBYMP=l9Jf)w6gw8c+uUeo@5w*L@noEibV(|
zW2$Sjv&g9RPt5=Fgs3%yrZqxY)USi1rgP22`PbhfIO@(QveAb)xq-0LG<5ArUp~F&TDB6>UWD`V6%PJ1LU8~sR8WUq2h-Al1YCc=&($Y=rn!dGfQl;!d4Q@1kEZN#^{@my#M+)#_(IjAwvU;WCJ8e4SqEG
zAl~0#F2&`3V23P);$j})lw7G8x4}bHr{m&~Ef3y^JveX~x9eA|glQQjm;PdZ?w!%`
z;XZ6w)=R?$(mb(EzXo$kJZn~~mmnmRm5brFQe(l#ix;Y$M1nzkk_s6=C9$LJ6tQEo
zv19y`b$qZ3f6fqmhfop+hC=tnC+t0N68FDqS$PVv8Rs@Shp1X1Tk6)N>soSo`OumP
zNjehwa&qtj$k^{f_(TQ)xhxNXoTmfB!!w}HD=jaN*SC1fvUWpC$8Dpzx_}T>HqG;|
zbO*(gjP3^aq<>zM|J5V?f0DI;K{9R1Q$eS7T4cjOl(N@~?3acN;2DDyTCBo3Bpf4A
z4nh~QJh5JE&dMT`I>7}rxXmYF#ut!SVg!cRC8j+svtJ^LU`=CvyoZ11HL{ekULy9y
zMy0to1FZJiw!dg;=3*z!E~K006`61y!amWfT{g%#}~)@3q~E$
z?0YQ&^fg)9EIaCzNF+NGgvnv`j+9%(hiHqV^bDYEa4R!vy}UeGcNH-=Z|~>`
zQ(&@&Fg6W2LzCF?j2Pk~ej4j=)v@7o!|_ss$qD>M3af$%6WVMlCP
zy~u?1ak@F+k6bV~U2nOKAqit6;&m=~wB28FqSxSY=Z^wuQ7iQstUwwu>b$%>A=xxB
z2#MgAj*^u0Sy*AD^F$Ky{p^Yga!?Tlhi
zSN3*gAPz&4cUhwvVapB)abGt2)}RRX@|g(^SM(qOT%JOtbzw)@>qP-O6*-*A$tkG6
z^6=rq7$IAx{JP&-$RtOCc7|fS$=)dB8^i_m2ds$7I%o%r>a{8?ufcsABlJ`K!)fv&
zlddbd+7*}M`7Qd~CqTUo=BRot?3>6Z=R0FxO-||^LgLg4`JQy%jX(FOnyJ)`B+!V)
zGtBSD-9}*j7IWL4b#xR?nU~a&jX|_5-1iXLG=%$R;PNg8M(VKX_AXB$!deOmw;%4>
z!{dcjeKXPX*csV~k+gYrqQ)8)0Z^1bC88>9vMTMILdc$`ScP1qf2^ugPJ{CfuE
zv0`pL3&+m`O>=9g;Wj(r%DLSSw$Tk1leLr}8V`EYuV~Lh#M%SdV?Rhyly~d+VK|XJ
zQ|Mk+PIflz+lDYj!X&8?fq&ffE-MdW$P1ldeb^q0I%8H6v^}d5*q%0)I~=pXWkDE&!q1^$AXuvq!KFs?ql!&178tr7krWXG
zh7KJNq`T|eqrP>%_nh;cway>ka=F-Z-+Ny>?kj%R@7lrYs`8ZNjN||SN<{@Z4FCu+
z01!-{B!N$!L|mwWepI`Jw{UiQ=x7d99y&j^wRg6)GGTKycXYC{ef}Rrq6g2q%1Ry?<9)QbRigJHwx+gA<+1n;<
zdI+s_ylm6bC@rvXo+b!
z_xN*&j{Y<49$8{y0Y-=)P1nGgpoPKv$_9
zv7cvr$m_Z^%CJ4*=O-t3uolzs^0f30oAhqG+8w#3?rwGF>yJ|kLr(zz=4jiI%{N8f
zoaK+k80{MlUFGCnhK1dBbKCyZT2oUqIn^GeC6t~wzm_W_UplvB6t{J@jbQ)Tr3ISE
zcWMgk%KK(;y1Hc_A?dB+@Y{ZVehWjT-0bY+fxA+z!^6Y5Mr}Jgad$U`J3^M0-Z;$l
zd07cPeOaoe&?tS842of_Q$q{b=&!9fwIgcmC+e@e%xR8#%>21JZ+v{icRgW3PG4U?
zQQF6IdM5FP3$LfAr*(x1)S(ZN0b8?K$w|A*QeIn49YPJv;gV|#qrUr#s~`D_nkdCb
z&j{77$c{Sqs<;?m27o{P?UxEQ$yGVYgOv}fNBhgIE(3;i31W7+o7M1hye@B)yD%isgH9{DwZc~!bX;AVx)Jsp+XYHqdV~T1
zWKf2`Y*{*#H=R3hh6k#47ypBUUE_3_5*1v%Ro8n{;fLiY_;x|Uoj>FJAsRse5Q>vD
zQRa2y3U~>eE-tRoo1PqG&@iJ#M7o(eMl9u9zb|%}pi;!~1MrK8)~wyh=pX=~3OiB3
zySJ}x{z$GQlIZX2jb;Ymz^_?R7YhFJCYA9XnyZ@-$dz+HEEeG;1>p6fqAq_YnE*)t
zWV&SAwswghJ^hRpXlQzWGCZZy+&cP!BaNphf78t4EUhvPih%7li<6~1`uFe%_VB9G
zNt~kYN13(TDa8Kg;F9_2jijbk
ze_kl|IE>Gfkl5WikdA2tzaI8zF0ibAZZb27OWoGOzgHDLO|77e
zu93$ZZb~}8Tlg6i0l3W5F~)&+6-pl+w@<=V`bUW-0^
zW;hVpLBxBho3J}_I%#mu5$tlpc8Q>LG4Wal{s4b_U#PtiJWvDYkO3tpF~F>4^fOWO
ztReNqKEhB9M}V+^wqvW!RB*Dl#=i%y>6&RVk71@8fij%6?PYVObKE#FKU7RM;pZft
zrXKqX4e@uE36&gpPWgYjQb&>f4DK#7cAPj``JX-vR4V`YtFvJua7Khm?8Y4=-iSmW
z_ApsF$Wj3F6$Kv$uDaK>Q0*erS6_QxO*zk+b2|KGW^IY>mDw;192OBFKc_4V+fZSy
zwQ?AieB{{7h=Y<&AldGZMsjsj-?&M8s)!}$7_DfZ{zp4O1P{w>q&Z8@Eis8y;%l6F
z;m+$uBv;|L6gn`wu3G0M7Bk;P9^Q8`0WY9rM?rkq{EUaOU%z`g)lk|LXp82T?&8v%ZpN@wmjr
z&24(Pw^{5m-|ty;9!33Kx~IU+$A-3p?c?k1O`>RxlBuuXV~k(<9nJW?TJ}5MUkyLr
z311ys_7#7{BE>jTVIOW-ahN?a&p=xZ;x5V_dDI{rvSh3b|*S^
z+=t#NOR57e=fVSpQST3PNPee#Wx9Saugq3&gr_6QVyr(cDJoppPmi9g^R%5^pbzOJjXY8w8>PZHeF7Bb8~5Er^QJo
zG4ooc3m9anN{;+%u<69
zQo_{K^h#Fqg90lF2B3_1EK
zuBaJtiBnQ;eD*xW`g*K)%G&BYiHL@Ns(PE;w_L{Iz#MaSvi(akJwF`Pqb(@ynqGX9
z%FCKLK+8bvVIzv+3M7boW3iJLQ=7=8!xu_3;Xe(1Nu0>`v!d%)F0#i
z@bfF;4>n1Hg9HT-fI8-3=OmdJ#uL6hxzK$BE~x^U^SN|*Mf#uI!9Ds%XbM+^%^6fc
ziTU@detcLDZJKdD$`4S6$Hi_!^Y$Loqx?P(G4`)IG~@cSmzYp0)y%+LkC#~#NtKN`267k<|Cs0R=u&xVj
zp=3H+r^Cp1um(Ln#Q^1sAE4BDPr&+asb-B=@g0%_r4L@K1GOtrlfgC2KldJ;F&fo~
zB>s|bWc#sW*3xex7jJZveL34bM&yW2c*a8?C3y0+&U9)lE;Vo15Gy1jDEhPto}jdm
z&H|SfZy5P&XLS&$>y%YLOb(23uK0uh56bhYR`-GbJ?1T(bLP)wpdwElB&?I!qj
z%th*OJbh4b=Zlzix&Xk^paPoEb0Y7$TSFKbPS{|%`15@131abLbR$)XCOlVXD2|5j
z5Wh!m=5;zw=mQHlq9hv5oiAXL`kZr<{E3s$BcHbwd{l=cD6C!M+O-~95I~##8poL%
z*l1h#wj3Ap8r>8#(sCn&v#db{vNlOo)}&@l8iYH}6#V0Haj1UL
z1*Y)U26t0z@`QrEzc#d{*>r2~p1;>HtGlV*)wb?%ApN4yZdK?5l#mncB7dQPcgXf*^*&>#B2>Gkf;dQ1siga
z#lM9HmHG#A)v<5cdn&TTh}2&xJ2j9bXGWwj-`d88Z{7vqp$H`eZ#Veb#w)&6XHu2i
zw)~tV^z%(5QFbX#<=A8lRcxrkw~=DkT>TI*c99A6=|#F4I;lNQ$T;^9o|n-`p+ylB
zNy=(sx9p#9s%IxYaaaBWZO0U$!j(rxa1is=nsa3ksQam+Qa}F$B4y1HIvq5l$fS+_
zIj8faYM8Zj5ir+*pCHH8uR5*|Ism*3&X7C+4SN(J7W
zMO`F~Bzf;hXe!m16>1F9|A-+%-P^3*(fT|O)G1Wq>$$6p6)c9N1_y*}lu$-r^TjQb
zviNsp>O1_tM|L$-TIb@GvE%(5u|)m|gLZBt-%3bEQD6<8$Uq64*y_ZiE-kfNZ(|v;nH@%CDr3?x?|1$5P%F
z8f)%FQpNgljZ_bJGm-Me8|Su{+$$kFdg2t86G+Ef^t@_ys%82GNeuoAP!Sce^O`j*
zG%X;7kx)RS*fY_(0x2a7hG6n!Zq@8*0ZLR-0?)3AIab7Pavhe<#Syff4WLD84~+;?+8z+=%1V47D@lppoL<}}2e(yG56Sdw?rG&-f{!flj}*x;0+;PYQ>{K~
zgyv=_5iV7FeE(FT4i!w%S2J_ADJNq
z3Ua5&?A+YkBsaQm6?^Zl1qJ=gn=T5xtKpPjvS_@Xj94|u(K(ubDeHbOT13KzqJ_JuoK!4
zw6wGoH{j*ziNiBG0zVlYUmxN5B_qAkE|=C$zBH%r^8XjW#UtQ&pRlTxn`*4Ur
zyY}26_oGF@fDh`SW1dzI$Hw^Z$iaF~V!yDSo#jyWIDf$UP;}b2PG$aR|5MIS2OBFX
zeqs4=dh~_y{#fa1*o-eN&9YC#oFdlHZ;zBB$!KCT%TIk^J1e?s*?3j62*PZOeywmH
z+3$r0IL$+xU#y(>&gAmKh2Z+UqU`j8+Nn?aN4FYmWBjCwEpSSEm!~Tb7THmRrZ`ye5}w
zmJ9bytf=$~w8FBDA3IOhmGN(dr{Th9J|d+g5aZFOFlthpns-%ITV7qbX<;-{>0r7r
zSj_Ps0JkBX(t2A5XK|qB*}Lm~FJgxE7FgxO>Z8fJ$%i3?;v(0SnxmhSV0iqU)$3zFHAvUkE_=nVI+5_CyURJ+*?Kcd*vC^X|&x+(Mgo-4%!~
z8Ic-MZ-#r)4e4@o4z~M-t&XX>byN2zymjj&9kZ
zNNHbvM>`*wia{vfwiZ29iZ}ssGp77Q%rb5f02mT(m-M6pfM_QOHZ%~ffvBwd%JEFB
zuNJG~{F}T28xnuXvnWNgN1Y3@5BoRr_7sps26@%fO8wk&L=jw^q?bz6J8k8fO3F_k
z>*VF_DI$v!?-)91x6l!F55thqW>Qw%8yI9SvoS&4Asl|sVT=!CcH)F!>A9NBK4sWO9K%hZpJTS+AMK=<$oq;;WYD?y2n{nN_;Wk)FBl4>M6
z5&7~giGThXR8B~1PbW_v-9C75Q84gcG0t?>q>V=!&BDT<;$g1Y(JT$tlL|+nsY=!{zb4C`1Fg(QvSwNUD
zCC9#mKkmq#ECYz!(ahcIUQ{dzWTm(7&t&0wMxKI`pzS7pRB|jogroi%Cyc8fBH88ZfRqF6}iUU~fzdWA?8BF^#BfT(giNebbhO+)NKs;j{4*7ceyMYK!LObFT
zXWrr8%*T4!Z|FP6o1l)~x&X|Ap@dVAz{oX!sj#hy!NO^C^A~S|Hu8XS!r|MmNuX3(
zu5RtNoXb}4mcavQQOM{|#nT`a{z4FeHo*W|rM4b6O8A14RQ*W+35bAN75yNE0GK5p
zC^+d+01R<;%aXNF5FyH|&XN562_Xq1K)CY!qncFkU5V(jw{AFqYsGQO?F5xSy=BTr^Y_jgJ2`D8T70K}`kMeL6{
zbv5fW>|uY*&IShs4H$U|6jn{g>(+U?U@9tFpHnj7{t6&+F+>T#YDox$ZJ!d`TwPq7
zAl@=K3D2Wg&XB5ptE{MVZ*_1`*V)aj>!30#i(55GQV%OzT`gENH^7aDE0eDhitO43
za^JwRl9f;ZGb~2^E-67Oj@Nob>+-qjy8ixtZLDldYgg08#zEJ}Ty>33OV0h}jJi|T
zKL4se!SIgSkk%{Dw`>2<6;jP?;DXh1*xQ(mV+{`r>x|}raezbX918Aq@4D}Sd+qLY
zg2T7Qz`+_fdwgxF%!mnJg}K-MhrGP}zX`Yn1F)jX`Q|*z*mE``?s#c=IUG-*iW;AR
z!_qP!eFeI`A$9#X%KLxPgcV89j;}L8THeIl+Q(gDQx6T>g@Pg1V^D)jsp
zKJLGDg8xN_vMmf<1UK|2#WP=v@%|9*n4a-R*ipTfl@OdggA7CV@2h#wU?J)pkAwp1
zKUVVyz?qBPTw=eEQRF!yK^>-edPPK_o1V~4_IGMCQdaAu2R%n7h8+NQ19Ihj_f_Q>
zSmlCMlaB6*3MGGdwDkBg-BdijI#`G=1^)lmM0yQV_y-03u7$!v=N-6HxT!l3uVG?{
z!4$+(b9m6QeUx_9#rFm~dkZ%g{s%R+tN>7tgz&5ao_pLCeG&K;DkQom1o%r*qZG$>
zn#}M=W^gx!(*@SL%&K?*ye0A*?yo|b@-It$-WgK^ny?-L=|)YMAxS#749EFXCk+U2w4w7O8*)o-mL
z!r)^;;P07RIrSB0O2tkyp9f1V)%cC;CB+_%(kCBnWKEtvUO8=3B^{5@v21>&#_fpc3-zC)sFEQWkG(MBwz3|pf@qAS;D;3V={bwz|
z{9Q2rVpIr`U8;KQHWDUF&A*bR-gOzvJeb)rWU1E0Y+cLkciTL_`h#xIvabJBLs32o
z?IY_m^b5kLZMT=0+bOg7Y8yB{#%k#_&-P|n`bvt4(Mm~4z45&PK(<0NH^y&`+wbhe
zPK!A1Pq*>l^L*M(r~P(5?2I%Vjvxyu9h+B90tNZzfUVrT!jJ;}ik3{31l(wqlS7x(
zb`!VsYB0Xj2Ye?rt*AyZ+rcwTVz${>NN%GUSs-DJ@QFJ|BKpAn1U_^JGV0nYJo`os
z)cptr`ut%dA}QAz6>$3eyqSvx_(`YTfnJX69f%Sa?EdsVDIq{
zjRu-|8G?c5g?win0qB;5qwjdrTNCR7|LHFp_)Uwj{%`t2yc1N%z{@Mc%5D71l=#~$
zzwfMsjdEwyYr}%fWf2stS*tt!uQ_8e)6f4E5Q$rxYkOi3uYg>=ex3g5)2FM0rs21r
z`80a}EN0@37N2|X&1P{lN0Ly-x{*)H8PP4__Q^~t1}sY{B^_;5}q4FV{3b|bIFvdZar8
zC>xtnwb+@z)q+V#Nz)n{CX%`NfAAUaD%p-zi{7|-lf&d_;wBWdr5KtJ8yj?VbhJ~`
zuwS$D*{z-iI`ZQ)@T`yDRLQ8F7(T^!j(pw7#>PhZA*Wdwqf6eAAU5Y87sh_1HaMx0t=Z(2$9giaAJGu*z(DW{`Vfq(Vsb&l~66
z#yu<#_68bW;W^JQ{rzI|T~YV5REM|)mW+?S8{>Vl@bL=IIr3GnhHbC7eXmwByXS^}
zhno%uJFD*No#K;?ffU7V%j1=^nF;V(36JOB2r9|dCK|jYQg4y1!>^^ky7J+4YhRz{
zT@{t)aNmuz!AgfTpGJN)nM3D=u^REC+2kW0JV};(xL`T?TAn5N)vH(T@D$kDae;J3
z@6?c^UdwfK`18S&R&m0XIC!ruyp`2Y@;SWYhL1AP;#%-g1}X;4@TvJ9KY}jYbZsI!
zEqNv?Dhki!&i5B6d+DYB$jM3yX>V^IEH-WZWCP3B7`0gQ2M^ecVD?fuUs!{SR2B>_
zvl(cC$7$g4{mm8`rxe`G&kgJ^H?R&Ex*6IIV|W%hU%!R+WIo*$1r2RyUFLYb^4oJJ
z6F5z=Rc|KV=3#ClwHmDE&5m^`KTpdW3+bXnb?Bh~AF$b>Ox{U(()q93eDJ0rP`s@w
KS19}7$^QX4#N{9W

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_218_0.png b/doc/LectureNotes/_build/html/_images/chapter1_218_0.png
new file mode 100644
index 0000000000000000000000000000000000000000..cbac4fd70f9f2d17f9f7159143d4097c1198bfa4
GIT binary patch
literal 22611
zcma%j1zeO{_caFM1p!gIC8R?_Is_?^?p6r}1f&KSN>o6QPLUEpke2RHq`PAn5UGJ-
z$f4^yGu*en|NF)D@`IUqp7Wg8XYIAu-e*EI)s+bGsqt}ea0peDA8O;^TyDU@xwLfs
z8hFPq@V+znCFZ7J=%(Xjt!
z#rgOg{&@kflgo3yj>M`I(8yn}l#N_*a0qT=elFpBN}MzB}Da0?8C-E1dH>-@Jtom6mFYm
zkN-F^sSAb3@n5}X`Am+n&H5d=fhGj8iVL;K%U|D!T;;7@DTL;SkMC@CxXo6mr_Kd9
zMK1e9JcU*AfL<}aP3m+6zj647u7h#oba=iq2fwT&<@mra0s0II@QW&PfB*;Qb=a~p
zXYswmq1&%k(yfszB2i11Hrt=^E;i*&
z)Gsvt8cYvWR!~rg$#U%zQ|eCOp*eE`?Z5sQ%y1Ub)pb9=kS9yZ8)Cn`z!8(x@3ZvC
z%x8<$xgoQ$(R_c_zlz@T_nhN+pSXSYBh^Tx%X*k#--wzBTman`8RaMMmOBH4TmXM#KK<;c~ixvvWxTk4g4&y1)53TH`#cv{V4@
zwVY~P{c7&LKM9zLhrTF#59OG>joBPN|HHddhYQ13{FKbh3Z+jr-wPLX$n>?z$9S)r_%MYX=2XmTpVvdvBrR
zX;O8&1M9@Eo+$IV2fO
zRz|hVd?*+B%CFlwD=oyc{S(dE^RpJ{>F&4{B^Uz%x~AVO
zM`YH=46Nxs8Ih8Z^`nr+-9qncyQfD(P6)zlZ?_M9N-UdZi+M~c-D=ub+*aca%m>Gt
z&fE`A{Zie>jAB^Q`*UcV>a_%IdIp#yCBJxYG#C)eO#8l_07b-R!}S2V!Pp-y4c|>hSSP=hu%D5=@N}sOrwUEgB8Z&o<7BZfi(t!73ewd8($r
zP+0hA$($ehuS`Bq*VEF9#;km8Ek4|P!-dDJR#!uEE0%uN_jETM>EZ)(s9G6LKP%_%
z9+|FLUTK#(%h+3+%rvWmtoontFH_6-)$13T=E?`>l|xVjb`B1s731dlIvS}(Kjo^x
zm4z~=D;iz>3`sHur$Oeh%DPm+8=0`0vVLik;>M#qdanf%9*ag(Vwscge&|ew)*=i4
z)7{$NZwP`$Ci3#23QTaAptet=kDl%b>)K8P1nq&@fbte>>KR!&=+Vtjo3${uXvy;*Uw(?F#|ov{6w(b3xTbCvWnw3ztu+K#TN
zr1|mw#s?`_EL)mq9;_xbR6MYSMD3$y9{5YoK%`@Si<8Sfj+1j0o!Inp04%ev8Jt~3
ze1Do?ce}e7(IdG0;^ZL2&N1OrGT!-emr%Glp2~6I%+2dpb&ZUuB#6MOAD@=Y%LfzL
zwgeHD%qx5#Upd{MH83+vP4n6AobZBGF24Fz;YbAbP7C*~kdDQ!p;!%Ry5rgTNSWL>
zZ?tN6$6#oblfMBjE^h5cV;%a>2m4^O5%^E)Y{qwUNMDq89Wtwo5JC5)Ntq_I0!^LcrBfmM~`9%$T~nsx8y5#gjILUwWS={$X!QPi-P?mnnMfqY9mlW1*J
zA}ekxy%myl)WYCb{7ST7X~_n&vvShJT@7V2MNWwtu3f$a%>A)yBCj4h=@6NLX$F;`
z4I+rfaloK)n+%I^aF2Ay+LgOhK^a3s+Hlu?sZws$Na>?*349hsph2}XNu$}*b&uOy
zsi1oUZ|gSBDjWA3&Ig7y&Z}sNa@HI6WfoqEq67J+vrL>Cz{E^;b$5?fHCC9?M;Zh)
zzwkZm7FofQK}Z+x*CLD7T_pWZJbjO0zRw$4Kk$vO!|MS08sruhs_iWC41gVx4>qx{
zVbgPs8+ADmvGe|OpSOL|$I}7Ae(!FA$5o0N
z#E?o(Jj8f7+j)D7I{BD#|8r@~HGAbuZb`Hc`=slJ`&w8|yeQ^5|GK`)J-awYL
zrwx?#pGD4i&I^x@jpcp+uDL&ZZbn8%hAb$vv1L)ooZrL>@Ibi
z53hPLRxo)q+3)wI&}3Gh-&0mK!ABqb*bK}XftARu=2s%>){MdjJCyNl&nID%{h8y}
z7+?bsh-f-PE(0I358vM~Zt_bls$F$8Tc52&$xH1`+|Da59)2$4otwaYSg^gbQ#ZYH
zcD%8QwZV0n^PPJtLq%p1yGV<3S9oF5f;m)rDmqNXZzOsW!Wr>nQ%>oTbFBp312ZO?j6c6Lp9dEcr5
z7z`Z*fYVFTJe*4KK4iUmKD%q~ecJ%o5^5gPTRbX5)z0(om}TmdaPqKSe)SnDFLRl`
zs*3~D1p#G!{REbkhW(k*QtNY0SOj9)wSPQCY`J4S!!EviG4#}tQZaJ>*N&GIof4B$
zV>br8wrVmM(R(c$om#lo-N5{>L6LtQvawok>J7G~%MI{%o5s_~m
zyFxkpCp&y2xD=)n%dS;3(7MKB_;CdUMGc@wV*MZ?^uq_d(V#}kJ=yl0*l&4xjlp*|
zqvv-Pk^RM}k(MT?YL$h1p8DrEE$^vA#8eSf7zXJ6Y;YFijfi~dWQK)wb>qf&Dy~+s
zqOMAc#r%T&0z1$)Yyd{=edluFHL+UGZFXX8tQ*6U$!=G6Cw{
zTD#LHF*-HHseiu<
zC8q+qYbUkSBREy*V6p}UJQ(;mbFllH1~&Z9y(`Mg5rB&)vS55=?a^v}1z^u=$@kSF
z0qAu2#$FX5ls}WZnD`HV@*8d8G=TVUlaZ5mp&<4%^nd5g-`&fpSn(#ki2ujHJp?46ACbWa9dl3MkbIW?Aqe&K$4*zgeuf=jP|_?d^qE
z4`qHX@8s6k`zVL=aCXrQ?Yy1aS={&Z+pd;tICZWF4c)f-r1-`pz6OePx*_wQ0T>i<
zMA!f!u6RM1>D?i5yjHfmnqRyctwcY#z1Y3GTkJ1Yw0!MLUY>ch@*lhTBcH>(Z{Ltx
z#(tGw!QwIa=;nUWn*7UN=Pok8>0bT+?yi%v2a$}nO~W*o16~8hZ`uCiiKbtZ48#X9
zaK~Zd`yEI?D~2!KTi
z?BkmU4Ptf#N^ajPStr#1w8%1CZ;-#a`S2geR6zm_X8gVKZL}M^Wpou0bZ;U`tsh`<
zNSivt`?dG9VH5
zH&UyBwiY(nP4rbU4)C=u08GV_QtRLgGkpX)$NoD!=x?+q2hCH*FJH_WK6o_q#f04j
z?-OWZ`;2*L(%+8>9+2p{Y%YRDTj!w;Ia%X5_8@`N8mKE
z8CQm(-h(EzuVi>!C0(jnaQTEUfb})Ez{Sn{Hx3El7xPK-ZieL6=EB7Q6mCM5u3vOn
z$uz^G_mePYe=PN-=ErjxKq!ZBHT`T8pbWqm_`T0c~H
zZaphBEP3W>lKxY?D?#tVAXaEk;hN^=MPyw1h??cJ*zUz45WnXRYsi>={rWW~>*xrI
z31pj2QBGvnucFh>57Uo+5c_?0@Js8uhpL>ozB%~%!jLm(qi3G*v#0k{NbDV-PL*1#u1(BuA>I)1MzIKsh#)KcYxrui2SEm*O#
z*S|=Qlp7|fUvIqG%^&8tyt=yTw$m$OzdDjzyH?uGq9SuN82N0xxasW3P?Wv^Be(J1
z*g4z=nuEx}veM#<<(W@+8s=XJjGQT+{QseJQ@0dAYXb}4ifr}NyyJ}~3vHpRGKUMX
zCEhP#2`z(8ZGG&Qoc`YEuYvbln=Gra9lon)V341itAZRa27J`@*?7y`_r`-(KD)D%
z!=ojUCMp_SNF5ydCxorsYsW@qNJm5JYmMu2#K5?++TVmj`$3a@F<1tmybTY1RmG@w
ztz-r-+QO)6=YnX^@x?w0cto@V5*ziIAQHw%EHOe3Q$le6^HV5#P^rS_&;=|2kmov&
z);*de{#dZb19S~!9xS!KK(n;!B^qwya^(-?JgYA2IWniYEVf7J6)4c&pgxXEJxS2B
z!{y+KU+7V-%7T+8%ouT#F=Qgz;i`Ka4}k!dTjhd%@r}QqHE~$iOBj1W5_qQryHhT9
zqZQaG>^S+5Q&?=eWiG}@(pH&_R@`p)VugX6>|A>C72m*zTSBs0jq?ILB>`JE
z3ee`D)y}dT|D2YC9mV!d=)4u~BNmmp-#3?bwXxvN0d5!A3AI{)1_>35{A0L)kvDHA
z-f*~Qa_6E`0f4*f|2CZ5)e9A(rykk4=sXdI#xLy>xs;2@Vtq{(BbrFoEf&B1v%J%T
zB2^>ai-iDQ9_L0k$FPXkfJ7Mxu4FG-1tXG>O=x;J*Y(3mdv1T>TZ$Ma1o{iO#m|Vl
zWDajQ8VVT@ZK0ix*ZRx?DrAUN!f(btRNwBPYyN{Bd3l7!OH_<+Y@88;lU@(8HUjda6%p+=(1WqOkR_2%>b0F0uz+n0uE3)yG~QlJo@L406rV=
zJ}{wY7}`{;(%P&9C2Y}bj%`7Mc>48K5Rn9sv4=ZqYgq{Yett?{^#<>Pj9&!=2NDZkT5A_#R56&JpPj83Q|>e?ZRXl5T#NYf0sWQFSA~#}umyHR`s5HMl1urU
z^dit@QE$qqs;@WY(k}oSo=o9rz5z1ndD>v5L+}-IU*86d?gz9a)23ti2FB$f@njAH
zSXR=!*9S)pEDC{Fw4Exqg|3yfE3PNAgsjlBGS7;1;1fBYIGT&M0%|;4C5VKho)ZkhT5e3hVBKiIVpDK_DqGs@OR?
zjselO$iBEiACH(`*W5gfHUgn^;3FNMm`I$9@vHJsKYk$L`)%YVxy0mEo|?@Rkc%lE
z9yK({uRJXNnGETI7;clXoz;E);(0{V@)nzmDnV>u-&U|^7g
z;VwEk>Z8_)W@DDksFj)kAX*r(2z(zL$=Jkt+3^kEx#_=RvVPYkFLH}Xt?w_3emUKm
z*?T00C;BKQa_*I(#mOHEVgEB9-<{;yu<#}ERQ28zF&+EyqIed-ZSuNZlYa5jm7cWU5Ji($
zr*=V%uzrN{>8ZNY^{bbSD_Rm)*ACWIo>pNoAKxx`=nem�Oc&Vl
zwBg{-hnOte;p`Qlyp6*>W($FP*1;s*gl&2#F}#QQEHI{ZJMo))_lHuLv9+kFvy>&)
z``BTWm9X!A^2IDX|3{}hUG03d$mcKi>uxrlpeE;}J@Auv+7a`A{CIym+dmiEaRVTr
zcuS&`@HW>?covgDGY65MMg%7@)<-;H!9)j1I{e?%IDv-7R9ToewM1eE-~b1(=??rj
zBG1^+*CI+Wy5_yuU@96z?e#zFLUh8{j=IjA9GCYW0#}BWQJ(w)v|X#cekS0tGd8ej
z?f@H_d04{S#ftEodGS+9e9vjl=a{epdujAj<5Hh?c3RqrXZJ703P6m%?DE%8-Srsw
z81=s3r3)cS@fM){Mgg}viMUL+oTvK_2(Spt9|QW>gHBW)di}R=d13{Ila!>xn90ch
zId$?f%DUUI2cQ@~CID#0k7K8urr*CycK9Ff1I$w>T&oSOUdQ^Z<_;#n{OzN5pl^FH
z;7o}j_D`Vgs(ovhP(4e)XOsqA>_p%LE)zLv-}~PBk_4u&M~L-XK{Od2?+gR974jh6
zL>0f>n`VcB(1s0HS|SD{l!9D}QssVAOQ
zG7f_v6(@)c%mtNEjQ~OWjVI<8YYFB@Rx2>?;tUJwyDa?@E*6o=0YNgBu>@gSxxsw)
zH!0yix!&;CGr+sVs+A!2=gifYG>9;i^a&*i^!6|OxL*w_Sj877h$2Je0k{82vMx|f
zOC2uu#>boBqK>Zk6)GAF#07bIA_{Ko*iPi6ou1jaOZGP}4+27IK#y$5$Pi=Ql5G0E
zLU)4Lh21>{&vye`s4+(rY?DRJe9Xf!?RId}GS^Hyqk0sv>o*6+^LgSf*$+=nQEUYH
zq#9gwG51xoYMd9;jq?7xnX4cAkQIPOjx76v5Uwjf=?`Y^K?>(;*@+Gd)hVwH0ic~&
zk)U}1bhhVJO(vK<`{)Rs)VmS#H*swq_(cXCFEGl)B_MyXK3!v6<6>RB4|i^-gRa*sKWhha)R58h
z)UT}{DC~hxH!gYBI$)A=r$9D@32(pC5#`(G2=oc^GQeJLgl9I8f%=%W{cbLioo?QykSWxiTW5z3ooYz!Tpisjy;e7KhS;W>b
zw4RNxox>yHEp}_lk8CLaca<{Hdkpk?r(ecmXSsjF1=Ad+ub99c=2)i&0(U?!gvk5v
zk_A2O(7-O{9i&H=2yj`;JFO9?eod7xCx;f(ki|{rG3YiLCpWQ;okNiH5wQbt8+0LB
z$qiH4Fc-`~?`HWSbo<)r+LJ?{LuT}nPsI^ihj?{nfB#3(Y}t(sNzpebWgeyO{l~cf
zdkF}j8xF8_NFRS#wfR{l!zugT@$R@qF#&J&6C5>U5muz!-wPv6=d&Oqb7$W3gGnf;
zF*M{o`K}l=xbyk=8*R=TJerDg*>GBYn_>ytm9w^||M?8}lgGmd_o(t&QlKpSTVDCp
z@5Mjz$R!Ri7ty2M2_-TvO{R+b94@c?i}x>#86&0sSD|uczZH-pknMkW-)V+h1!R?S
zggn-#Gpa=28U`SqH8-yAUc-hL{8rB4Xx@u7?kLF0EpMl~$eQM4Mh&bt7a_N3cH?CzrSK4zGlsvP@UIg
z?9o}~Iid?T5*M>{6>}}=RV%CI{o0Sc{5Npx#qhO%fKBs6^LhgZkgJl;I6~w6Cy-Kp
zk;&N5=j}s;8VPXd(VwoACL_J;)v+f0gBhLu{_KA4bF%$NeFc%+>wAqK3&HGV+{b#f8&@+#
z0x5&;aA{5<`(A=c;jGYM2j
z(;`{-VywTxSi8`PT}3;l2{)9|
za0ZF44y+h4jcMRjwH3L>n%6(#=W76le{%bW{3L^eUVDO^0VmPDDKMCu#BxsffMjW#
z>)1JML4jJziX(@o{oD&V5%ZqhRbqDri*H~_{s
zF{j1}HT>IwRf7?wkg>;`84NmL7-MiDMm7BdjLX^Pv#xgk4m2Oj*=|yRWhx70Q_23?
zL6R}P(HYHy9qun+NPlhwtJ(~*1V&Coe8k4CTfkT@acO>ewP`sST{$6!C5gWx2a>N9
z$mWtd>Sft!teFDI-33!?C);$r!M1tT;a>81;RXFTQb4F(Q7keDs{3w&*!9Nl5X|VQ
z--Bp+t?;Fhs0udENc?U_2#7OQ<9eJ!p6n3bn_MPiDdiB*|Ew!=PM^M?(Ct8p6M1(o
zBye8L)#Ms6kChU=8ZJ4N|DEVNQbefT4r*ZqIto+j{oIb;hy)&jQZ9^EpAI?($OfnB
zR>|5Xn6%60Rlsj!IMV=_4eFB1tlFYk%A-XSHLux%%`n2t5s3l9tjs2yoZIj-&;acg
zyW$&WJlg3zFtYClx(!&B#<;0#IBzY2$#+H7LR((GWx<^aGC#0mSU@6=?)cq5E$y
zd5Ru3Zh%&#c`Z>nz$y!2QkcjGCJq1M?pp|737h!+v&YudYbsh-P@@b
zaA`23D62^1A;1bZpF#E1OgN2*9dL-8+}v>>dk5;~Y6D>)wTC#_=^sBi+_7Ai%ZM`f
zMhSzO2aT9=tIin5{-?hSWIIX7Xn~ncV6v^BJ^o7;Z}|vWoXFAu7-w#NyD<+IPO@l0R$aSyg^`%*42Mv9YQE%N1O30}=zOKU_u(E$P(L*K2A2R0
zW)wMQ;zU;jVHgEKz!Wd>3`1`nfq06=ozE1feJk-<|J@b-BdmP4l)(nPLq;hDS?&}^
zw*`}bPO0^LsB0`-62vKSEfFtmJMb}HLAv#her1>!Mop7qLTBF*9A{=OR-W5NV7*0*
zZXdQ10rL?@w?AOS(=T(jT|oUU(0vfN86s*xyP6~%4Hcq|BQ5+LRm9vt0vl`Wx)ylb
z@GXc&Y0dr^cLa?LdzlxDEyiLph(FflI~_)Fgzi__A|vb^CiQbmBlC1<a@RY9o5f
z@ptQNfx3K!1Uejtm89?!onpb#l5JM+?iViNR|GD)8Fo6Fi$IQFtS3a~m0O#{SE35F
zM4xeqz%p!gLanHS%#Yf6!}(m0!)%di5Z6mMvO#x1?}3`dnMMl(+*Ylj)S4&b`wYxW
z)1=bCwtJ6LZ}h%u%-w?>x51U$!kJmhI&I_~0%{6Om7k1TpsQbrh3-Wat7J}E0-k&W
zfGxG)6;}L)Q>DqdsLN?B*4;5wP%a|jR=IWWIwH;ya;=2np^#XqTF*U}-jCWWP7fee
z%a4Z#9MX^9HJ%KlqbKqN;J$FS#XiX0Ie;^-9|N?t#4HvMk~DXkqv?4})~`e!djci&
zTpD$}L-EiODt^j>tdJ21pNKV`w$+80XLKSNpnm(@{;txZ^@?hu`ml$7V=kMyr@=Y&
z&@CzG-jD^bJ6c?TU@A9~);HF|!sC12?6o=D#FNvNj~l%6PovJYxOcpzY82Ub%kp%IGN{r;8(b2-lYjs|Zp}%6Kx*&x83^@eS_sn7YV~DEam*e*=@!fu6
z>McPqMcV+n)`w4<53&TTD!@%LtVxSelTVw;3z(oU1qg|op~s%
z=96wNAT~^hZ{)p9a)3>l3~Ff?Xv6l#IG-u0hN|Ʃwhe7gTt*-I?M$TR`vf>`R#
zM#9c#kw;srRlB?S>hUel_&`PjL*|4$Jk2*ChYh!ZPbRu=o5EK-``iTBhsW{*he`8X
z9tOSK+De5Op~3VV+~0>fBmW>v*Pny5M25n=Ior9>2_x90vhQ
zj|L>e)n@Q&rwfN^qgsGv_M=q^^flu95Kaz=vcuf_Qqu(WPiv#B&&h3%;=aUvdi3fj
zJ1`1N?
z`@`a7jHlrApwh8+fkkS&$K+^(Vs?0!K!Y~A3lxVl3X(ttUoSdY^pHWA{YZ)KGH&T-mp$drw9{dv{+6HG>>XBlj#@rQ{kb9et
zTJP^(TJpZM0+hG?u8jYQY^>5U^5Z?+PRCmyn{*|&4*J&>AUu!~IVL*5Ivu;-gNhgt
zZgH(lmooIkjy1MIQw8*g@FrC79f-N%p2`qkz0zcXKL8bL4@+(HY+sDd+gEbN%spTL
zXhD}DB5#-}aX)t=VIhH~4YLJyaobl=k$Jm6XCW+gVxX;7Ft&Fqsf=Q#_;^)pbsT@W
z?r~VUx2a@SKwH|O$MZc5b0GS7tAyyi8D!hJP3JlihJFfp?A?{ask*HlXhB`z-cu%8>6ntz
z*Q9cYUj8CmS2B*-qM+wZ83TL%Rwp)OsbbtZ7N(MFlf1!x=o6Too$bZ)g@Way951hN
zv4AGt*k>0$->|ehYH0}_xfTmAc(=9ieC0!8O4jzH!tT`ate}!$t0T
z`N`#JMvy6OTTaM7+-QoSvr~r*EB?xP9yXd1+l~kfm)ISv?6ML8fX{LXJj|LhD5X6$
zce-Diad|6U|3!E03x`=cCo8L@V*4ffoR9!Xsgfl+JzBd*SUY$_0*!P5>SyWlN!fIw
zkB(x3NiWj>b2i1fPIPS#HHo`8H{55^O=ZZkzo}5uM6{C!fKA9lL?BBiwQs?+Y62G$
z?XT@QD=|3L=u{X_*O_A0dlD#C>o*uOWn`Wc&*%FQJEE&hXr?xPI+s*Q8Dp5oiBQGa
zypix3q-Ki9Cu%L5Fj?o(2lp+iSaf3Varx8kvy1Uu&!B*R^it@m+2jI0`K2Z_qv*PG
z%^fb8M7U`>xb1@~KoIqh3xj;)O~q!5?wy#zdSx6rKIq~{pFCg}3@&{iiwE#%MmBp=
z@f*&^%PkN#N!b{@0%0-$2dDg!L&QWH6<8^%3<_U?l*z8NT@A2Okzqdv?fe#>+d=0Z
z1QM^SL572}7{|ZznTf0ERrDTBZp^A3GZKU7K@p)~%Dr*5hVL+_bm^q*=cxC{L9)L=
zw~QZiT_LEMqA9m*r&twSPFOvcE*Poxw_3kO2lk%}OWh@9hM=ci!n-fU?xyA?7^G_<
zP59W$PdB=xcZ|jErlqVc5q3={%~v=$h5H>ZM+#X#4v0ucs9->{`TCPAiYqotr0mNn
zG;w8{L{tenYDoQWD_9-KlV_3@_bS=&@fB_D7e!YQ`
z+$O+6jDkmJ54WnrF;8zk?|7tsuF1Meh^5ZfSzTTRUZ#R}%r3$(@_@*LeY(9nNN%h+n45w`?OtEZfg&dAVDbCIe<0^j8K&
zOW&(ScJDo57Yj&LSUr<)(}{(h@o^NBFdo=yeqwalj*V`gxQ9o8x6@H&Z&+djQAmnM{@I+*BfFAibBb_f~t0vZS}XTYA9yE
z76=yWi5^4)=;EoUKdWD>9J3NB)KJJQb)WxoyQ1pLgC?J?({ie>z%Mi)tJMF`XRzAV
zYfkrIyWkXvNluDOXyaD669N-xVA7%gegezyuPojcp+)f(AF+tTcgFWm9~C+kIToYp
zbR)LdSnIqRFf1uC#XU^l>A>*oywXaFP8(|-@PtpUAJMJ++o3FQJOndP;P=oN3n(f
zPc2r!g{*_{Z5$+Z`M)VbusPM3>ceT>Qv7Xlc8?xgyEB-7z6)GK@3f6U
z7<#sw*R9@QdsT!-T!mP2H;CS6o8hR7OwbB)w3A+RQ1(BH#s6EHbsyUB<%^tOlZCsx
zJ9N8~&2rh{qN=G)>lZg|VG3>)*KvKa)wqG}tiQibxIKe*wU0?@@>pwJ32yDt8_No7
z?|8o^+vnK%y_PXMAmP3D(qXDxWps4Z&dzS=XcsB`yf0;9*Ll8;gNsYmseUUc>qTpj
znCnspIC&SsmNoz?&sbG7q;}&#!JLGg#g&yV{u^Sd{hI6Xv$6XpnDn6T{Hjy_7I9?O
zNRi`minw~2I{u4M3+bUTx=}3NQhu&hgeHPxf?rh=SBF&jMt=qpVCu}^uUZH}C0s0v
z6na1&`a1wm2x!HaD12q;>%X1(Iow&c2bNb0OkU}P;EohkA2(T_alv$cBine6bmps9
zb<_s%{5E$k(xU0;wWUt}8@_%sy5T&1qZ;~7dhQOcrO8)8^-}Zq;cwSDT%m#{%V(9K
z#;JBU(hgp#Zf6^1Ih5<-`$yZTH8y6d?NpDjcH`XJ420G0?ZP>oProjo@t#a#4uSqN
z@g6Z7wS>Oh`J~z<&V7|>(QH?$#>PCm@^mA2tEO(W{0tA@U_;%wu8nkY)q*_bGwRrG
zcda{DcpvC3=!kf0W!s?c!h&J)ZucQvJ06qbvC>^T*j5m&jx@)(ZDQB<-Nvbwo?Jh(
zgU-K<&!-&Y!)U&0r2gK}9t6Y4qLn9#2i?2Sq7~33c$P;zP4L|K7k?=SWJsN&LGmG?Xn>+Td
z&dKR_7JqD%{Qw6K=QG|r^DHuE2@Rw4defqnosn8=Gw*}bU|QC)IQOY%Nr9QShloIE
z(5;$n-;!`@XGfUNL#t(C@({!};CxfyPZCf!8z|lv$ZYr0_5++E^i*fEB8H-c^K$&tj6ZvH@myB+*UiGwyK)2*^p
zz3$=Cx8~~=E>;tjg|B`|(|(k|MO1z-)Lh$d$q*aqmukS>9IN~ei67@0As+;WFMhmG
zay0VSCF`03D-U_aKO52ic97SX#|qTa+V=KvXpXI<$Gf&E*um@dGYZf1D;<&-ME4$B
zr&cQ41T>u|X5oN*>I;VWUX|u$>zcCirFaMEVqwEkE5poivH6*1nL2G@nFdvU$fe5K
zt2kC*VFcbJ>3m)@GUyVXXR2td(9iI&(1!)nEJ~g|u2DB*5u8n2yPF;N#x#-lWZ`OW
zs+c*$dhQ^X6KdF^YHd9I%&9nnDz+FjnFE&Fpz=c#6An&NzzU)7Zy7e0u~&N@S#zI_
zXO9}>HqMMAu%#DQqj0
z+4Qz`mieEjl-*B0&b-cyNJ47uC8ftA7jW%-k+^{~-U=D}A{}Dx5vDoZaY@T+klQ-y
z<`Fus&(#NT(Cw(PYBX!`P<%NKF+H2YeKrLQlR*)B+}vcuI;IRd
z+W4sOyh<@+|BN#!C`G3~8X5P&dI_qyu%m5O?+YKG9h$95K=wPKMrT5=S=Vs)ir@Q;
znaa*gf?ESa==T5UEagh8R(}td_FhhLENCZlZa+1V47x!krXgxjq(k1}Msc?;go8r_
zmb)W!&wM%czB+7QLCIgFfiCs(J&q{Mgys)xSfMFEjXvcH>dLuIykW@NhO&buk5iEr
zEZ9Hxac~M-D%*G`fz=W%U1p;E3bJh}>b?WdI9Qv}G|)pao>adt7SGcK)InrfCe}uB
z;LY+cG+!)Xu`8K$%hXD2+$X;B4gH
z_-cIB-pa;C9Fi}2-~NkCu`k-Ae0uT()R!pJg`xKX{L3B{SnYW%4ER;lA4*g>9M>)M
z#2~W>7Ct#ghpqKYm%OY#t}UriD{gd^_4l(+hu|R>T|NPWAFOH8X(Oc_&1^c6iuqR!
ze`K`zX9k-&%Ec6q*q=GP+Q{Vm(W9}AQ2>B0YRNQ&O-41Hj?W&sK9_J?x1D$I%bhQu
zmrVtRRBZ0c_6?k)wy)&f=uV+LC8+xF{k-b6t
z7(u0?6^bt5)1P5~HH9FM*$MX>j?~MyTZ*=!HMH5%UUF{hPMs5(&(R|BXQm8<|EOb%
zz)T>H(TsCf&2RF+d`SLoxZWU$L@i?M+aznU&DdMw-}k;5Hc?NI2kxcF7rnnZJ=+vd
zHy~~Dqi3Wyo%BQEyH4qfrgO(c^m84S&sB_STzadO!CEN8
zHj@h)9qnv~WoEFeI~;Mv-9`Mdubtg7Vxzi;_Ec?AW<2TrKTmS*3J<9wP*?mrZgJ5a
zNtl+%^EgQsZx;0mlx{Cb&ZQvY@Z>|!7&1@y_vQ`X3lhx^AEvk+RG;Py9@4jH=1}bU
z%pm4m>rzFgW@%30f_Fa~w{W#23CFEd#-!q{qPv!SkB-N#*f|o~n>rUQr3mPk3BE*&
z(4Cn+zx59?GOvW7V>ydr)feCO0OZkl#_;RrIu8&rZc+;1pOl8)?LezX83~yuczg*v
zVWJ93`mQeoxH-edHSBKPIzu2fdgf2XkK;mbBOnHC@5_b{6nXA
z&ePW_$vJM>%Gj%>p3{|{derTB!6IhHSD^jNsJ#~Y^%9Ahsg*R|eTekzPK}S!2O__d
z0FbsKE1D^+jGPrcsTVZK_NWIT>|pxLUjhyDN-sH;D!Vhp*dj5ZsNHgaI96|(g32aW
z(hq!Os_K)rPO;ITDXdTFp*qFIVF6#zKzX@k@ErYGmo}$ocQJp^TW)QFNSMpzRWS&M
z$#Q23l}8-|D;;8tlb`KV9yY3H{=9AF#n!>TNr?Qm2_UH!g5Ek>tN9A{Lh}BK!TE89
z-_A1XlUKb4+vB|#Ut}ySKe#Tsu{a7*<3yc-b6{2=^LwZY(uvQOLA^L!6SQBn$6v1$
zoo6=K>O;re5mC^R`9openp`bzk&drrkzU&y=y%e-+lIbJ3nf!}Gxv^6ikg~qcvVUt
z%Ug!J2;Ym!2|1HgyS`xmMkaFb365+rcns8vG6=*QNnQ#rkHv){GmiFJ3%{j-tykd9
z!SQ`hKA~sQiDi-vNR#PF=*ZI8Gn@HZXb%u;zhoPwJ?6Fmm76D(9LWL^>=a$WIM`1P
zaIL~{Cx-WM6WyqDt{*mbicvgog_(BinHjrK=(@4;h
zG@^4Js&trC6GpiP?*CE@H1TNSnH<#Q1hvM3MZQASmPMOasV1coGsLuke2>xTdOp6H
zsCn;@|L2L$OCe8>@8%op1!LKT!8uCc(6!JYLn?#hy_P2Ew9YV3iG3NQ~c%gwJ>OhxQVuO-L(zsY(q1
zf{+3NH!9&q_Z`cT7fG3ls<&&+%TUVYpB{SOtdzLytfrZ91(v0$vn4OBkQ>o?Dj_4ExhQ)dn{i~rnQ--gEf>J};=v%4j9Yim8I3{|pbI1}SEY0Y%e
z72-6(b!1~5#bLYzHl0Ak(~l+8E!5s}iT0Ha-Hy45W;%YJZq=v$4s|mza9!B0+2kQ=
znDw$!*sWvh1=4TLs3Dcu260=L{$b`k8smy(*;G#ZPvyQ&;nbgCwe_=${CXoK?+A$S
z{`}Dzd~F(qWFo!N%+H9vESd+3AebdAy805y*ijI;qr)ocJ1(nn#TMs-W;{FCo`*{6
z5{X1v=Fa<_Q#0~WG5nKHw419yqTa^Vo(VmRjHR-nNiWLWSD)&>4APCbz0?1|^7PcY$zaL0pd-
z^TVmPCHDrInfVx*x%kM#Oyr*GRR#a)QXgi|vrE?7(~7rStqXpp;io^23&EY-UU<#{
zsy+eZJ#?gi-M^yj4WLFDr)4g*_Yf6MXB!JQWUgyGdL_D!7A)O2s}QD-SyLXz<55Z+
zeCAZ7q)=l{x3SehyAdBPVXmkiaHB={8&5&rRKlL-^oTWntAqs6swO;u*|Lg-17^)J
zOGnr|$VhwI;L0D7+U?8UEgQ=JI2iZ}j$-8o)V__<&(dT&n9>L0NKzOf
zp0x{dv!hn2<@4-c>iHh`%kT04KVo2J3DQx@ktjDq8=SyDa8;1rIB6*EACUB0{UmZs&IQ;MO<-98hU;SvKh24V@tGxCMIFZjqXH%waBL>Sr0%mJ!bF
zZlNF>HdWb_*XJ4*Zd%;)^WGq%si-laxYlG5T6^q~KwoyQ-~l;XU!C#SrsMQcNQrRn
zjr^CIX7hK1g#w`|j(1eVj1sn8i6W_&Yf`)1pXb)pRKOY_e0(b?xUX6HG1QJfJ@W-Z
zgisz75(TfeZ);Af9k$9-VHbL$6niIAl<4h$wXp8*qf>tP9v}MZV9o0hcqtd
zGURLFCk;VMEH%3BK&BF840=8;QWV
zSFP=K#|IU>EqRJXIlfla$m=E11}rTIf9M-qYSb{5X^hKp!+A%qtU$bjQhRdd9V66(
z34NAZgac;%9z?^z{R~9Isr|e3jl}7ZGs)@cg;P`66X$iPJ%|EqT$;(TD$>ORe^1GL
z!?SW{U;HEmem2usoWCQLDqUsoEe@ZB?V3+bjC*`XsgFY)>fpJAN|pD-P_j1X!Uuzqejmx{t_Sr(8|vk^zr5iW_qi%0(?u>QJbF(-0wLY9L_2
z8WBn4qQC&TC^Uk%1Um(lo2!6fQGyqgX@ZQK3=$J6>k5B^
z?+%Ro6hoRfbNpS0AH?#gD+$|wC_(J9$`peGF0omkAE^g9uPXjpx+dsM^NA#Ud
zZwg}zh4Rnc@7X4rSUbTTj&9lNf7)L_Id}e{okGd5PtDWBhSeFziqA4k&8ZWNp00z_
zc;TwLAI@|ROq%}|zspEmVPYqT=j@3Jk2R;E85Pc~iD*N$Y0`{68@pm!n`HUP@q^W}
z44$*}w6NR;W9DJ;j!c`uYG_z{UlDC>7f$=k-=Wn~7GmSSN2Q#PRK!udKp
zMObS?ZZe&5adU*CQ-#H1$NSrck1}p|3LAVi>NMZ|A=KW!W3zV~Urwneg(ey`sp)DX
z4^vFPE^gzjp6tnth@zZg%04ZPKSq{H4ct|m#VtkGj8`vyghpH(YmN7M^mJLkHE^Ga
z3hM>8tF{ky{3?5?njTSyg9oi_4X0|DbZ<2)#<`es_&SlEq#oW^sdme|SI89)a=h10
zk7Q)gdyQqg5_LT}!L0>1t^~AbXm|TERkSBD(ne-~Os4+&=_wiYNuRLct0x!VA)S|>
z@Jko%xytT#BV>5)L&K<#DYdjw6YcEae*&f%?(CQKx(55OPqdC3GhNq;$t$%Eck=Sm
zL&HNADjlz_k9y}yh304bHJ6yhRrEC5gPi$BYYk0QXAmDxzbF$TA
zH-vv)iSHs8!=5!9?Z~*$T>`nW3DE%>F(sw5q)tDn?p`n1b1l!UIUow`6g?ds9qMp6
zRdmJMBWyw+q&LtstK1M^Q=39=_tmo(C$yfD;anzE@LRlvZ`R5=VIq;pTUktWA_QGn
z(+Qx~GV|;SSyFW(xGEyrpp#>AR(wF0^K-2ERj1(4*bjb6rBX0(kb=Q2d&5{QqDg~y
zx<7g)4oSVWl&(JAJkvHGsN{rIqwiv5`+xnhtpUBI|Eb~cb@clA5@3HpC{{BF!@S0x)aedz
zd_es#e|LVqf}`av)Br^mB~e)sAh%JSp+R)Z7_YoHKe0ztR?`Am-YA83FM0tm{@*8r
z74d(Rjf;*0nC0C+k7o)}gYo)-rCEQ^bLZ^_^124ZKH85t82@bT@nI|yu@uSk^QU0)
zCjsHNKqo>k4J#E15?Eo?g$AGTGVO5}_%i_FgUG!B)S4qf29mk^1cLN|8Fd~q2!2jLg42a#V{>3iBX8DgIJ(C%=MuqNaCEpVbK-l~Y9~&(%}v19
zLFCA%%?P9R>qdH!d9VT6II$aCuO;_$!^0vDm1K-QUbVm>Zs5IG0ydZ!5+|dopQcZl
zKfbicQ9lcKqPU$l>1h}V<9BIkQ&_M}`anKakP^#)3y6jp%2=Ot!Q8?^nZa2P78n?(
zn^g3DfNyyS;yA!Wc`3LuFfb5EPab@6jJ-D{0CAkl`O^<#H#iU8o#+ZevIYD(hqaa6
zH<2p=gWb98gZGp#XZE#1%M1=C?920z
z)U-O->Zz@T_UPhiI#!f(&DOuE6-I-iAzA7HdN7>ybHQP3rlz?2dCzj*m>IZTkf^G+
zsKRS0#2=&i2rMs-XF@-Y*md0t)wd(PR^U3l->>ed3=
zN~iP6VqT<(77AO-J1ssv0g{UhhjIl
zI2vXpd~xZz7|KJNcqh(k|7=hPfXe6jF@>JVq)vvzhO@3J_7oTF46`t0SxBZR`^Vms
z+q!$8j&iP!
zN0U&badc-6#wH1zV6*V^^#;&CD!szUi=@rO8CtXEAWu|M15T_%Ft+bhx<6pCWbl6a
z8XqK_N`|Q%e<>{|H#c`Ey$O=WR))b9#sDbuhsI0W*y_GU)h*WTE&x++kW-tSr6L08
zq#PbGeHIZEqUi+=MhOyABE>1J-?Z9`MV+)4c~Q@@mq&Jt^s;owt^jL+iCW)iF{n_z
za+cWcB$HWi%lPxUYH)!Sp%Qr=8rhiwE{%eM0(dgH^XF=Bgp@O(uJzzm3}2BoSCgT6
zauX>UA`gF+6Qt;{%oj0$e#;>Zx@$#_nrJ~D5Ptkw*H|4}W)c^nh

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_27_0.png b/doc/LectureNotes/_build/html/_images/chapter1_27_0.png
index bef273ce38fb151ae1508beb546d901e284b9acb..889056124a7c8231a327478a5506b899d1d637d2 100644
GIT binary patch
literal 14221
zcmZvD1yCGOvn~=KxVr@i5Foe(CrAiR(BQJcT^E-SEV#S7yW1jJ2pS-`2U*-*-|*MH
z_q|*73Y%iiIeVo0>#uvh2~$y$!9*iNgM)*^l>IEF3I_)d2Ci>jBLlyQzZA&;e*|2l
zzq+W|o4J6EolM~rja?jU>|JauP2RYfIyqa~+i|l2U+>>oxVSht3$nA@{!arodna@D
z!Q|!)U=UP?&)Uv#aOn6iKk#sA86ZUW0(^XmJH>j++V0fgr
zp}2&Jv%$6+M1%%j+HTX0%^EhoBMjD^}02Fq*Zy$=O+nBMUIJ3JMX(bTw#K0S81zV~0|dAR>)El3oe8?RUn+8?HFYMxem+7|c+B@DXj`s#
zCITxeDvXFFnF@PUCL=>3no0yG)`*zVgaq{N-Mcr8j3HSfUdZhq
zmw8BiRI`8cKYiHT4+sJusAdt5Q`a$$U_HCGVH9jPQz
zfz}C&189NDvz2
zdwa6kewE?)_-fJKe)1|TEUdzyvu&)dP(B$!Ch)#-)?jhd>0ROXoJ*r}s9l-3NyV)1
zjK%0ON&V{&D)y(;0jSUvr+Yq+t1Y?!OW+bT;%0=-@i$}(Gv)RihEGne%lOXx+c$?9
zZSz{A9t5_|i^+v8@UknhT9HDj&*OFXb0)sBaBQw*NS*005gas{db~djAAYXR6djiB
zCrV06ib2Ztqf)oIO`|A@RTqK7Zs`LNyTQm>PhcvKWn6P{S;AQ}x4A_76b-|?r?7L;
zD)z3W+F%z%oA&Iz3bOFGh+EVhjQ8L;C3nIQw2j_fgwE3BMgh{-?;i
ziVBSMQMVCMIeB?fVRsHV5wELnGZos6`wR*x5MT+>^VO=%QuIwBxZwz%pOpi>F#aHa
z!2E!eLPOh`R)+{0`DVl=Zf0gCrMT%i$~*g4bj9!a(RGGW&?R)a9n5X@_Z?(UWiz6_
zAwpHIb$Dtm3(?!8V!qLj0GIp^)*Pmb>s(d1PpTGKneqLwKRn{iPJVPoRd`rfsoG-I
zs@YO)X)d3Tx*Sqqu_BEa(il8WY?nAi4Rs2p9xLrB?IivV;=@j$6Y^3kEKA&%=$~;%ozReJ3X_{Wh2NDP=LF
zoQ{r+xVSimL{)pqQZe|m(e2nXUE
zoHybUX%wp|q|9+VjW@j288!*{12P5)BtKeIE(D8mXQ7b=1dT9Wg9Jp7atFsN36_p|
zp`)Xd@>t-ocV5Izk!JxX!p7eHz!iv$X);%1e6ZMDEo({++@F}3`1bAF4yHz7f|`k`
z=64ou?y0}o6`&1hWQId8>PO};n^WUjr6-KEED@ZkV>6c6K+FFPPzC=#SLDmHj#(KbwRLobHRkvv~H-xH8-%vwCS|I{Vy&A79`g^6*OS{>Q^7Ql+
zaLqc0b&2S!#MD$-v!bc(D!Ed#&(=qZ4(Fr%&i9AS&U>ZBXD5Bwv!qaBBVe<>nVTb?
ztau%By6mVRV^fXTHZOm2bzP2Xqo=3uIE9iwoM%736&0oOzdBsxHX9*X+d1iiqxQML
zdEfy8V^NmdWYEpcB-(;bDSX7kR!_BEu^L&dV-BhtM?Ebi$k&iD)=%^*|detwR
z+`gmMEge|+pGsYRRaMnAy>9_n(UT`1Yj9zLT28P-49#rjeS^wycFM6%REq{
z*Qg{uMN=udxVQu^j->NP6S8PJ3^5hM&CDI_?(T}R^rdiuLVy0u15V*$Z#rBtoo`xO
z41Bb-FrN>n(`MZNuBw6?2ALyvm8j))ZJh)k$2Tvp6eIM
zx#TjXD15!-%y#X(T^qg)7UynMkR~BaY
zZ*Ol$GAtFlr@MT-`0SQKN6pH)(eWW8sob^mHa5#`i9kS_s5a~t+?R!s%}IIo%F18LDI5Xv?41e5eZdp8paH?-{h3N~zl-s_X
zH54=BD`w1c|K8gxfs94*`O_!lbJ+Ox6k|?;RoqNqC2C#IF}~@3O!sJlZt3|EN1Fcqq+@sce?a)$dODrqVrP$5&C3(^IE
z57drWQFOvS%P<~m$T4lnP%Wzv#>PVIKqG0M<${Rq6b7)lWP!zjO4>MjT5VvXvj_kp
zk$|pRMh4=Vnw;#IPgIpF^%UOi|4Y7uT2Twm+C``!h|PuA
z<&^g|=^qRysLCO!#9JRhQ?esDccCd%!1veEWEyCS55#8elUI4_lz$*J1w`&161KBr
z@#1&@91ji-t_2YrluXf0rvfG}db1V?m#D&2nBiQ^v>5{7%E=Dh3>}<9=6gn)L$-lm
zfjc^pWeR-D$r(Fcu2t{*=(#d(cQ2H03?a3sVem5``?`RKRGEKp6A$^BJx)+AnZL
zBN<%dva1RQjU*rV_bdD*ip4W1I<;o1GWY(CAXjtU*=gbYNI-=c(n<%HUwH}m*501g
z+1c4=Wo7fR41u?A-`04Xn^m=LlhxJLrEnM{;4>;!P?~oZ6!h=90aOnk;s{#873dN1
z&#w&Nm0r;YFaO$`E(cIz8(7+`F4}-VrWzpgNR6G
zakg(709Ey7qhvOGi=2Vr=@cO1>n;Mka4D5LdX(zbm(5pTF+<+-^YhlntHsv+s&*V)
zEGps8$!z)nAeij^Ep43Ip=V@t-uaWeaNW9~#6)Op1&u_-jpg1vPUX&B-o=Tx5xH9{
z)7H_^StK}nI3I_7M`v3rEzUw%ixN6A0!9GBNsZ-H(SocnGA5aniwh4xNfgp}es8$}
z=BF&UY0q7lU9Gx4UQXi&SW*x*=FtmTl@M$Rkr;Z1w+j%+P4xM-te6<_j~@e&X{ouP
zDk>^Did>059cUH_>g1EzHa9nOK%Msn5)ueNTt#)`OhEI2-?=?vry*Ygfk3R=>$98>`IGu=h)4oLi8BK`ynh7F(nKTNI
zOa?-R??#A52>txRBK;pls>41vjLVfLs1C6njr2FXkiU7q$kNtfh*L#a4g4pBsR+(3
zXv*Q&sYA?5V-H{}T6>5Eu2-G)_e0x9@R=R5~yP90-plTYr
zb8ue-!G0}Iv*rHdNd6idEipVcVlp{MjzV)3xr!*(9d{S?0^awcD+M?X6D!YZO-jK9
zuSO=I^?%rjIpU$zTNHP+a#sam4jC!9Byx})KeU24mK1_q;WC;DCCStR%*BIjzmsfp
zr7{v33>x88BI=%oT#X})d#b%FOhEI999=wsd1R93i>L$Z!H>K&z;rsEiJ>G^Wh#`=
z0d;BxF12e||HmNy-|XdcV7scE^c%oZN>9|fug%xna;}gurC+k7^>RTFY+*K?d+2o^W`>HzhqZI*>{~{
zGT!zgg26n7uD94{J5gDaRAs=&Y*bvsLZvm=^tl$wYMZGJ(%GJQZ++BUd8YA^&&#ZU
zzS3BJF8|fzg>v#1`1*=L@`XI*UM9=?cJSr$B@vsBc@urKQ~JpoIJ>6M&9Fhi5K~*J
zZ&9H(O=3+=?8LYp+EPuSgF3jR|HgY+0P`HnY#1R}%kvzD!kqUkN%F$;5wG*IN9_vj@(oKS57gap$Kq@I=EsmB;0MNr2Tn96pwo
zHJx6%UOUYbAJ>>|x?-fC;oUIR1QwYBC~?eOZtE?4#mScbgw~zqCqSCru`vt5da?1L
zYUVPV4lh{!!|~tlSRd%^#p51bb(&b4{EMmG%6e$<##6h;k~2n|kH<{ya$Y-!Mq&=u
zzIN`NLfI;?$0QjYM84te$vE0RRfVa>v}U3k?&@t+pIiTn2NZV+O1^`LYc
ztL6{G*SM=~gsjBu{#cY93&q7RPTU*cE(ym&0P*?obXSE)FMM$kqTqh$8>gxhQ{ztK
za(s09)KLh)Fk#jbB|=Uv+)@6=fu>E$e*xVJ=9AsjZwsXnPQ!`Bk5{M;hqeA|K+w?3
zfE+t>ZObSgC)L#Fd%Avk8YURE=rGCEF2$ATY7fFC=}li;ZFbsysZB{-S)AA`+Oxn=
z8{F7kT0B^JFLR+fD>!7<_MN|%z|I`@BAWY-efHTgQ}iFc8656sp%3e?qLl__g#i{c
zN!o%k$OFZAl^0MR&)6pw3h%JgEX_-+Qxv=~HjV9<6Yf!E`Jl(WZ_k
zVo!xZu@i!S648tp@(V7ERu%OvyjOs8fc4>iF~UZGpN9uUp}OS8
z|Vpf
zNE&jyl07LB<2_kF41*Tm$o#lNkJoO5VcDpd&FSR
zdP5*0HMzE9bGhv)Uo3YdJ$pa!a8{~TLVTATd)U%+8$!xdhrDP6CGv~6?T0`k-N~Fb
z&O~a(A|Dk7uxh@^&b@N`13<)C5RP@-P9!+_*y!MIc8l+|a>vcnGwyUX`+J>N%Wbz;
z;=WyPUU0^P{YTn+&mlAJhs*Lv;oDt7s`2Monc@f;B>^O&_x@Lk2OqYu#0+sd4|B^B
zR5kH`Jq}kusQMYg761LYZ+uT5k8+l^UXB^mk=C8E+VRLw!iiuu`kjdN!Py(=)wm5z
zC97GHC@{9Zc{T%m=!doM1Eyg$`&#^xJ+P)7;e+6M}hHU=E*?^t6Z;yZmMo(#8Lhu^8#F($$
zbmCI`7Jox2o5{SNpUmD~-^tC!NpiP;=xX%wG2cU4(ib@-L%0x94(Q2UZ{>sWp~kmC
zyY11o_`kd%>p|x0spj7JbCNlhB`@*%1j!;Lpl-;DDx)?S)zJQ5+gPtx_CtCLB_tyC
z1On<3QhCjTISqu_GB(4A+!b|r60#&3B1>
zeQBQ7u3A=wQ{sBJ&}O@Md)TjSGc-SdVX%IOQvK0}uiQDV9&%)>L;vbi{a@f1%Z4>7
zxnG?PX5Dy?Qhn-fC9~NYS#}nZWb{R}Ei=A#nw7lr^a?-^n0U_%@JXO`NP!WL)`f&d
zV`R5aeHcHHJuIQ0RpEA5R>xZ#ixB}^idQ(pzu7YJO|%*6%Z5+G%4T-$An0zG-8{+P
z+1WRPVGL)`e}@s23B3)+SG^Jmj)R04;PymrF$JJbR@
z-|tz!E0J5@(j0fbt!FXrqq*@>{vbxZknz_&t$&;{0->iIV0M-M`~d;}@V+Py2nr~k
zaPtrnkU}p1z!$k7I4$^U-~F^+jgQt$Ty)mN%HQ-Qbrf(7-!92#6FcX;P7%PbY6JKV
zH6el}p}mK$vmfdWl*BnPv>#p;0)hyGMqc5PqC_B4ot}M=D>ZT^JfX1OyOrOsLRXQ?
z&l`efcYhuC8U{Ok+862QI>wDN6hlCT^hXpSUAo@8~XHKlJRM~N4|bGLo>
zu&&!pC8qA7=6sC0iTI2gi(!Q1!2aij>U52YLdH9(6I9_h-e(`Bgo_2ATBDTQc5j|S
zE~eEr$)8+*q3S(M#-H?{IE?b1%;?6X>@p9S2F)
zy4w8LUKNZj_M`iWaK*1%8^!H}tpx@r08ODjs21x*yK1GK#DzSizz#;CDZim93LphD
z9i6oQ8b8w|gBQ2r_pfNC;9+0aWsdOJCI9H2UHLm&Z)E9M1fO^qYV
zko&77u0?;{nkRCg9t#SfgpjaD=O&yX4S2Wa7d2R%HiQ{jvaPA4+nKPWS*5RMui9=G
zdZh1W&cs>HUTw7uQ2Sd>tu=QCF{!=bZQR{L`(cF9ADr9+l*P#gUF!CP?XSuc&H@E1
zb>Q;e6iFKi?!G6*)R2kl;)7aYirWa2;iLrI{|Gwfrj>6Bc?C)jurr)iMQ^
z6&OMtT!M3CLtUVxzaXhmx}E2XJ>^z>&p-HA8Bu^Qwt%b*i=XVEOTRl>M$|xFIx(`_
zP#|=o-B%)b8RPWyWc9h%!&oI6WAdb%#=BKbj#pcfRTV)*HFpD3UG31DoVcf|Aunyf
z{BgwgxGG0}U#jb7!!qLbMq_h`W&LKu_!rtUlAB?4Bz}RTRl|0b`diC)NH{xemsOME
zp^w2;=bTo2N37df*jlv$#~jUA$G;>Ib@0HW_QvZHEr)#WyY%JTN2uVlRZO4Ytj+O2
z9;L{crNp0aX*dv7D(`q_8o0@9*eoNAhprb%Z}+YV5;f0`=5mG2nsC8qvN4HBYQ??#
z)m~I{J)PSrFGKLaTx^&HO}>cR6@7mh=>1XD63Oc-s|uy3z(FzfF*@q%t#(C^3@>lN
zaYJ8W+iVXs2E{stzbT@WcbDf~3P*YmWOo>nyZ`2218^^m=T`tHY&!%qBiq@}zCODt
z-v6Ly@atyy&ROl=YO~k5a^X%N+;cuIDe--_7`_F3YQ4jaQ5bGc*H1?CriF`}TTxR}
z^Ccg5G8=Q~Zm1LZs`LA1em}L*il5Oc$3VO$N-N95*B1lkts^
zJ(PQaSF=#u{nP{^pdPd2WpcbID-rwu`ii{i2{j&8&G%c@E2%AUsYfe>f*lRR5lob+
zOs4-*`w$Slcc7Z<6p1$b1d!^{0RBdM>c(ip+WeT+j*cJTq(h^FRFT`WD>c%KJE;%n
z3MFOVZ_g`nDM26~XR4}&aulzqS-1%xLIGX*5jth16Sh_BPIMUTK4d*_>MIogbr-$q
zE5%?L4_r6z9ZGZx7uHj6<`vGKxvkLa
z&0djrk@5r>XLqAW21CX&IjoZsqZ}D)B&WvuG+4zV(zrb(fsC0K1-=XStDfd>!|tWQ
z294Q2bavQNULw7vTl>a(JQep?Z0#5;?uZNgwS}^3|Ep(EJI*l8_;So}C^7L%OlWq!
z53s*m8ZH&92$AuxgrwW|nrCYVEN7d7HHNV#{Wt@P4wn+qz`j-?E9yug!yb3%m?H-D
zAsa$(&zo6LV!t%tR&|riQ2mJ4>A7F+MU|ASI#`!mY-J(*{XIjnX&!^S&h2z*HvL^`
z$-k%b`?CS=cKqw1;gikynXer7ouxPQn!`8Kx}>9OuFGE5h@e3fGFcFN1%wY~8P#5qIO;DaQo+$V81cKf53D?)+ui
z9HQFwj&<%Pp4&OOvxmF|jW~+aaLk|#sK~YA{h)A?l_nhbX^ee2=fnGG4o4T{DSN>%
z(Q3sX@ybm)y(1~cjlqtDbAjqt+r8y!?V&F4zKFu;!f$91GSKy`SHCfHP~9ZIgje1-
z5LV-Dn?ciugUbUp=B+)owzEO$fho^<+QAll-$z2bmL9UvTH8e^nCR2;tv*`qshO2~
zdeSY*2}g{L-!joVt|GS|PZT-z_542vD0fc#(c`rGO6H|AAouton|Oxx-N
zr?u$P)3g#Gu_=|W_b`1Tl4CAwt#u~4F7EieH)K-Ysr!UyEt~?&ZrT!FEtx6z;A%y$#?qAn8g>==w*~=rhp{KvA=U(Zs*j^IyFt&
z_G^A16GxLVXsDTw1Bb%87Wz9BJzNwN*McTUBlhOMmk=}P#G9#W_WFP!CmJfXY!;=o
zTa*B>2hm~M2_6Avy49L~Yhj8NRy!f={YA_dtm=jL^vIsV0gr~hzF@sR?$;#vwzQ_G
zN59nJ?~n8-kB}jE+b7MJ9fjl1n8|D(-&0ylrG~gF_hpGV{H8xSsOOlw^i_9%sh^15
z?r-Oiv)cTWW4^i0vsJhiP5)^{p09k=y9~Q1I$vU6|NEuJgByb!ncDYtUx=3>1EDIA
z<8LWH?fEP~6bJgYG&(xJ3E972fUf^!W+VR_c0Vo@m4)6NP+uLA?H3$?XnbYUaxn=S
z_^O5b5kT%I0tl8s$PlxK`8?K&UPokaoZP1i`|&{7<@U|5)n0+wzFkySJHdn)pPGmR1-UudLiB|2zeP8^jj%>P6SxXPa1(Q
z`(Pafe-1&j<|yU-rRZlL>Djo`A!*vt>MjHn3_e<#@+M;(42l#NOdX%^a(Lq%tk^bs
z-frSsrwN&V>SiIeh
zigJLl1Bl?4v?K_UbRqcBArFMHRS+R@F9=+GBqw{=XL2LKKZt|KP!bRt`9d<*ZMlT1
z;>pl6RgL!WTWH(Iy^RICt0y`I{iIPDZ2=1AYTA5xW^9*=wTfpY#=Xw#oPz!|_}!}Z
ziX?VN$FX6u!Hc1cbyw?2-y1#kz6ah-d%}|(PP4+$w}>oK->@9GImSu2KT1L2H)_nx
zRhc{f{)Ig9a0ZQcZs6}`oxVN~MU?qja_ztH`x}UGZ?g5Z
zs)6Z7!^H=K(ACdS7&ZQ7pHpaL(3kuQf8P}k%x2|kc9HsSJzKhcaHT**Uh@&`zrrM#@BxNGi%Cn2e3i
z_RZM?9Ji6z!f!{dRgtZn&TiJT-uI`woA8f)OV2rZ0WML>cUp{Nnj70R5!^5&QREeV
zsJDq>1`91^B&U~j0bLAilr!Nh6>8OdG;^?zGTE!)s!H30Xno_T3(tKeJeJ(*^N#;i
z&({K;-Mm}2yYW73Xda1Gw>O1TQ$_5M=UB$WWz|Pnv4);AS^}PJh;}4gH~S
zI^(ZTtp95KnUbxWA}((4Xm+;IzH*db@yD{{NfFa#D#T>X+-foGltgpI2l)@?T);HJ
znYiTrWJAPRS~(G`BC4Hpu;i0!&VnapJVNrL=Ybc+q!3EE&dF`=`E9`so+j0SLWtz;
zL#+d5vdvR~8%q)$6jEB;dpWDidwFpY9DO|8;9ve|`3KB~A(k9G$z=dr@&VBIEVO*x
z${K-`hxKdHb}!5)cPJ|ae415fKW5rg%!MS}z2}Eh|6c6kW}+TUJ<=x=Jgzo~@cwNf__rY1FbZkiGp%KfeWv
zG}|^e_?Gu>oKA}pRLc*1`rrk7mP;$+ON!1jIhGdw#Of)8sw-mSR@1EvWH6zxl~)v-
zTh|I_3@`6j!B_pv|Grw?vu2O$^(|$Cr@+HHY%8_j2kPoSug2Tjt|F2kVju}7NsOG=
zcwt>!tQp?jUBBWothbu)I$ELr&zMD~>Nk!GEPtIys`#c2tl4yCcYY9GCK4SRPY^3}
zF&1NrZC{$7RFk(Fc@e?v3_niBnkd-3ceApb&?VNPl0O!lw_t93Y_7TYnc2WQIP-X9T%aq%hWka69iwOpuf
zen)>JFz*T4P09aQ*Va$Qi|3uZcOre)ukmJ$=u#v-ZcjMXdhIi~eQqsmsq;h|cv;4!^B!aPmem>6JcX$2rh^g@V
zcq(1iN4$%lxt2E%SP>eVvg)hI&H>rc$Ilw?SIv{_>~xix`TPH{d!ZB`>B$WAA;GD;
z>?<8e6VnnKyoO~UR_?o&*&TGl`xOX!*D4z}L&DSp6f|y6zAg}`22}~(o_kJpn06s+
zgLxk1Rno$3Pc8ViDUAB66tL{A6z`|SoQA3}3n-cD_++C7GW?ii>Ba)l+5A^J$ydx<
zkY#PJcCy5Ml|+S;wwE+cTFMGP*q*BTG(_G$nR37)@@;3&iTf(BW-urDQaK*#1p`({
z`l1ZoY7U2Q6Ih?J_{6A-{Knjn)hkePS~SqM5c5jONTe0~Xiodp*1ktCiD4Tyahx`6
zYo?)xf0F1?o^e1@5)A>haJM9{7-8C+|AC{Fw>}!;*F(yp&yPO56!_=G;-KYWaL&tQ
zUc9&qKT{wSJ!
ztw>{>!j-MSBQBfs@f3nR1&c#jm6-(}shK6G0)%JlI`f`CqeT7k;C^Ne8@Vc
ziR(9B|C*Oguv32PYmQWjO+lMv)fOjX5I$pb{3pzC>`?h%D`;owNUe?DM@EPVVIG;O
zT6~!ED7n4Vki+mn{?PL0z`VJ3U+sHEk>=nbe6&higfGG3eAt+OYJu<%Tf)TQH54|t
zT7i8qK>v!*vyISRnSGWa#f-ffThqzNnMn9LMh+FJQn$9wN9+oFUs++6uHm>T2w%u{
zj!7E<-b}kwCc%}x=BS@x-ZMjMU%+IF#bHn
zHRSOIjDdrPXsZ-l(gF(XtXB_^k{rU>xx(8d=z(g~Y--pnUaPjU2hVjd=h@7xW@6GX
z@@9@6w$&EuYyA{zWT;X@-Cai8Sv3a$bJZ(==UT<(v^k~SlSyEZi@BSa~X
z)LxsOkQ-BEQSrXZW*)!7+Zz`TjYQn5zB+|!OyBG%<
zLy)o3h6cj9DdAN$p}2+(;x^wJ$Bv!97(xyXf&1a=qZYAMljKH)p&8R=KBWDR9Mr(Pa(aDz51B1JMG
zpDnKABn(_NSd%L67ktoR{53iBl#)gwuh5WQKQjBgW!8l(_O#O0gM^c5w7lkI1H*K+
z|K2#{UozK5M2>C{59NjWO|9!=%41zdV^&91c0@DpWwV1yc>H8LQE_@U{}t`DI-K`YNB|jT$IBM{Et;YCABx8OkK_vRf?3H1P8+;
z;_`Jv0$sr;N*ek-4yFm2&tc5@K@<#MhGc(M6$A-LC;>zoNS$JwHe&S+?IqJJvfv7rd
zoUx+oK=A!zp*C8N`WKE#uFL>^tZHX}H6PC(^!c%^!?Hf}pO}#@E_TO!hJ#gb@#Q^i
zDlSPtei4!Ry-ywQX`A(DxfwJ7FdPGZso
z&5~G6N-qKP)1l=DS2*!T*9z-${bni4z0uFg>Buv%VV57p_f{5H`GR>B=u|wuM+*cT
zJU>jzO5i_Q1B$C=gk}U6Ku)?@`YEs!OEo}@JXcTd|6#O{x6(u^6Ij=D7
zYbnF0vd~y7lM;qcHVv^RlUs1IUrgVAN}Q)2mi@_`FRgBQcwSH$
ztw?@#Yst+c6=t@F>LENJmNRWGLQ~B3ZOeg5
zt!mls_sOwkl5@h7#O|wVKx)FvPwoy#L(mvsG#vl^n#uoCano>aR)=mP3X#oMP|o<*
zkA86$1zW7ZRx`#a)2?sZosX6yIW5ODe@HoiLqfl`T8)$Lrk%G+o{K(jdmsJj7++m<
zo_!HT{YSF}sQY@ET9mQe;J=p$WDD^epWBbYutx76It8z5ez(3zo(w-s15=S}m30Ji+@mQ(PmqSSWyAoKm|6MfISVzJP{McOz<
z+I_tWKz%j;G+#j&OJjGuM1s#H_7&7^kK_sO2qDYy>$E+BGiVm;CQG3@a6&m*8i$WN
z7=I>Vf*l>*E;h>}6TuG$j2I;g`FG!oLYQ+JA6-3bhAAP=*%%KKPaR@0lm|}tCl4Wr
zlK!r2GlmG>k-VocO9=MSk4avbk?>PySmG2EL-S>gfa&VC$|?KB-f-5u6&3k2$vDaxg=nYkuul
zt&7hSJGYC^^oM}Rf-43I9S-b!?Gw;Nd)68Ho;WrfK~wPA3AQwVI2-;MYvz?W7^TR9
zCb!fvH)++(MD>Y|b`~)e^8Bj(eE==(7f3i{^dp;GO}|m0*%g5Ai#0dWk
zMLUR%bFQ3NyCnxMWCPMk4Sbd=Wws;m`cEmU|m8MG%%jlXqR){Cu4zB-cb&TyDJsoGp
z>+rE=_@K}E8D|b%-Q*F>N7O>uvb=u%RQ6bZ+sfp3F8JX8;zTO)!+_@U3?3+k0Uij@
m>;HKf?ElIs|L4Y?U!32R
ze{xZCFn4h`b~b}kGcb#Sq@GNE-dbB0(s*mJW1KPDvH=mgzSRvQ8bhe`1K2M?E?NeTzY!~Q`^T-_t{V8PQzePaT8vXquq`U2lAUYa-%
z0}T&^@Ld{SE9tbJc3M+KM@mlDhGf(_O2g2%OukNsNcXC4S+(Rl=)3KA6jSs&V%irH
zXfFd_nPzM~tSNQ9O8Y>c#+!Kd;IqlieYMzh)s#4pW-;VT%n8d)Qe|qM?$U?LPf`sm
z=+tM-|NaHVJwGW@f_y$%l?lVjQA@7q%{rPpJe;7*wvpH6KG~tkn|$?}YPW;gJXmUM
z_N26ws
zae-16hWp3wNnN$WY2vym=KcIp$Snd
zOLnDq37G+hqTm0SLRoU+%C?q@Q&+R0fIx1`)UrX`uP|J0*Io77^POLn#=jbOi%X`f
zz%p$WJHP~F;;74N?$a!?S2j`%FK?#ZTC&E+$9Hwj&FS;r#~{#hb02G3dLSuUmvM3?0A3VaJfGdGb{S?
zW#Gl$)YPYXuM6vC7srmr+dT?>oP71qmKoL)(~8L9zr3j5=|HYbW~$A)ySvBH{z!&m
zt$dGSTa99GH|qL~KGDB5l+_Qs`SfuAEsT0~9x`lpeY~mDRL^ydQ5@GSnvekz&
z>{#1}8PbCf*-!T7{olF*4k6GqyX!Mv#J2L`-d>~WN@K-YcV}m=jJI~H!0v2L>N;!$
zmNUVB8-%1u`l{wE$q7DieA&=irqVA
zJuu~`$A2E;@besTTVe_FhH%@>*6!Xd-EZ>oPG<2t1qGwxa#@YNg6^uUM-INy(
zw!N>6pgsM5??~B*xn_%?PnWsP=VS7WtgL38LC6#RZ|&!MBZ-*#_eyJ~tIaU;^Ybai
z{4(?9QzktQbRF6>e8rsCd&&2nC3CWj;;8M48V;yKNJ7F_Z2RHEhfIE_@8}l!N!E-$
zO}V!q$B;V&@PI2URUNFrbo}->c5Zh;V=tOB#s-8hF7?@c$v0(K^e$S|_m%@#!(rnZ
z&Jh0^R%oVL&1j@pl%bmew
zyw-$axHR_aGeCLtg1trWmzQ4R8jKt9q+s$z*&htYft=K{PFtm
zp*3zN`gdA^YoMrZebR0iV~|Is&Yk4
z;?&kq0xJH2sMglj-4(at=goIJ(0OPDz=yWy4LiF9pd9B1b1B~>!?>K+Wr^7h^p0%I
z2U1t6wrst;1UqggDiBoJfH9_6LTS*#E)V9k;tsg{AANKh><5qhu3C04h=_>zDIeQ>
zfy3sTnVB&)HGM%sBBP{)W9WBdmUx-`_3L|`^q$Rfxgug&o5L6mFbvClIC~J8L=Q^y
zcpOIl>ly)zoF{x>KmpFc_d+8Smqy9jnu%}8hdVhrnfFpdPj3J+EYh>|^yt32x(c^F
z-^2|_4ZITW5oV?3CEsTa7rZ9Z{^4zC?zm6BvcS)jWhu*n`8N
zqb!XpUnhO@Bc$U(6YU%=qNwP&p@a9-W0Ij~YUWv~`XVPCjtK2@%NzC9ZJ)=tF+d0*~-^M88eos6Rq%Pz?FD!y&{X_jQ|$DeJ98oW@B{)wON1A
zInJAWP_nY*BT5F&4Tab3Bh$n^GNjiKF!gx$pPg;g=;-LBh8@|18GJRSefWT;42WFKs~y=q
zcM1wbai?+J9gjf9BtBkGR5poz3E0diw&w6>#d6tJ(X24QhPHa1eX24+E-ES#zMKXd
zG^oiH<#NHsy%DofW8ui``!^9$124+wH0<{d;oRWo0{bauJ2*p)e}$&a*4xggT4=XL
z;!>RLNW)r_O&3UL?C1AHEGLRJvdsGW`ZkYOe*G*g>`h8aA_Yg!RDr{a1jEewzowN`
zv^RTsJIsaX<}DfKt9QeX#iTUu_7{F(M^Hv!pVE0u<=iyBLI!cc*{tpdLeroYps`_6
zwGWkLnOHI%2x`U)6`7Rd0+{{%{b@q(Y5dma>#
z7I|Pv2;S1BBpj62#XHjHA0NG#D}!BgIz*SlrfV#40a~GyCFnFPay8Nfv3P4c9q@Oj
zl-olqxrf?&px^*tXkYI8nudwuV?d{^;p&4ZqeT46^@?9@w5dQ`@={~cuSEFxrs7N*
zYZZ~5e6=E;X#)V(1KA=OO^_`Rh>(zQBwwDy_ANyPATJj@UjH#zx2}uNb2YNIcJwFT
zM*M*=1Eq^u0IWtJC;cF4;$#)Lho1P?;C9G}@hkp>(z{X2heb1>t*ctE(;Cca#w_Jz
zOYDg(Kmf9NJSNZ1Y1;eZSye?MdI0kTC^r@v*XQRY?70C#rXY%avnSW@V-!uMz9fj(
zga{=)3$R&Wfck(H+N{W^PWjHyA2f<@Ft5LKfL2AlyI;}%y14c=Dh}_CkY&$hKUP$W
z{g$L!yhGla%LH0zQ|J-KIQL<$Yt~T?p&x
zNlZy82z&#L=3L*{uu4?O_tBUL`?=b^Bg!)vz@C-b^hv1yIe@s@%v&m2+(kZ*JvUe8FDty3i>9dXahCvpKAXh|B<)uiN}dD5lAi++H#|G_l0K4eGG8rP{u
zA~lUB!`g>CbkuE_BTd|G9>h`LOa_?#`{2$6(i%#BO`Dkln0k!tF!{SJTR@`F@WV6j
z6rm;!TWwP7LU=c-Qf#M}AiT8KpdC;``PmP^I2+v%%}cMxy+b}uxB2+rDTL1dqYyo@
zkAwi@P-j)dIRtSbro!D%lY-%*IrCwu!c%K?Q2Sg%xIaaJRN$o(0^MOO<3j6TsU{5@
zRMOptUh@>&nEU8b;=ojI$~&ztrPi^HutvnCssqHpb+HkMU&GO$A6#tFB=eM!$n9}R;kMra&qJGiOLo|72t^{}=Ot-vAZdMr>tIbG^ID4Ml
zQ_IL20AX~Kb1og10}w$)sDF7{r$o~r*^R8@{`@kJWOCl`yz4kgDBu}Nzbk9tv@xZE
zq3Sw&!uk66f>T(RAK(4uk}KL)6AH6e7|+)y|IX{4uYO)A*P~9DWk8Q($%kcI^)cDa
zDje|W{`q{lHam(+FuFF#Q(~)jJ+l+m=%j2YCf{_0dfe8I;xTgn>|>{NRM*J@M;CIsB;kq2zpgfmaDi(Iay&U@gou|N&M)rnW)Mjb*D
zza_mM9&XymYQq4MmohEbYcI2D3hyNVRm(q1>WC4M0YC;z$!ayO!YZuNNCKIySu;>s
zcwy(rUz@WL_0Z2<7tG<6?+g|4lQ1gHM>H2`M3LRW1J28z%6=0Brs%~PBKR%W-0J_HThrRq3B3@U+Nk4aS&7qQ9x{E>SILFpsSBmlE_~(&I@xnxl
zJv6^Qqie=IgH
zE9}=+j${~)xKPr+CAFJL{!c>l9(2)Em^nTOjEv$Ung5m~Z6?8>Pmk>PyIRM=G*JI1
z%rYJ1hf2Y0KsN-XyN(fH024SKkvvY!fnu@;(D4{+Jv;5l!GF;^u91b$u;!gHg>PO>
zOQf)j0)8Nqv$Qh=cAJ#)>wHJkD2ER>+l%Q(Ak{y#L;J^Di_+n~M8`qiKP(b>~xh_s0JmQH3lL4LVrcsb|kv(VvjqqcN@VwmV6Op*QP
z+4xpVkSZa-l8Sma$^^jt*5}y1-}Yw^Nes|3_?abX^Sl4vMu;500O;
zoQ=l?a}U-&FM9oSGLw^%DrxslLc0Oyee7p&PwCJH0(y`6-_>?9jk_mArj$&}ql}Us
z$6>({aaE}KjkADC5H+>(TG(cIdHwvU;;K^o5R%Z+1=zBo5G(F)zTP$isYkvQL-PI1
zS;F^yZrS?nVUXt9mA<{0mF}Df*j8F!a6SS0a|79
zQ}~W(m~H8ovP|ah7RY;Kk;>qI8hivA#28rSn>cF)z>6X+I(EScu^8g?^|>0HI>C$|
zD=iUWP_-%CjonUt2~E9c=C~16%%1c+Z>6HdI4uwl!9AvsU|OD;$#Xf+uC-Axv2!88
z(|K0Kt5zAy%83`iu%wDuny|0~;Is%a`EbLcNG7IVomt{RmOjbAIwj0w4}TwM`NVya
z+2&q;2K+G^kAAz44$I2#FpeR#Q+4-B-~ky$kys#>^jXBaRFk8{O9Kh0(+u4aZ$eq0P3OgT0rwq
z*uTv+-UWcS4b_~9k}^ChW~Sx8%Ekk!cOJ<6ZDR3=a-o)J6cZo?Ste8y$Fe7;N>oG%azm
zG|=y%SSv+S-T5g@_qwm6ot{xN)O|~@w>0X`sK%38hJO4W*eHBVJxq089$)h9;N_c%<^&mXX{}C7e2UMoKfZ
zBEz#njz(Q{(!Ku@Zi^ou1APs3xEBajWs2l6D_m5GhO>(14ZLV{8mXf5Ia72s;)5Lz
zHM6jJ@5{H(ct>|AQ1+g255;Xmar`C$$&h|`r_=AZs}+%?I$FY{k`ehgu=$%YQjiG3
zdXXB3MJ#W}PTf=&bH%%iw{Yv(Z#IV9`KnCiajY!rHdgv1w`X2nC+eaGU8-ovg(zjk
z6KQGJUEhwec2dHhj=a3ic?zuP2o-xkf4tw?-8mMWPaV`hvin|GnjmK2io+Lk7$PE}
z({(K&69UAmo||W37w87@BCMwvc2ydGiHCRSz`mMCd`od7Z8qTROV#1wOC>V<`gFA7
z{n=TRH(v=9M+c%M7n0yDpZv#!BYE-^7(0*}er)G+$0N~UE5VTHob@KZfFPY#Uqa)Nm(0Q^24#!(A2r*jYll1UAv6;%?&T?j0$e7P@g9
zjk;XyOD^Iy6=f|pkA}(sjv0
zg3WDDY1mp5E$`m2F*6aCmQ)>`|KMP5Jzq1esfg|%(1t!5qWQd)EQ@S#p-mqQdBLSs
zM)ob(QoT2GV%NTV-&hQN0v@vxKTSJ`ZuJzj)W7!Gu$6rt%u$Oe{W?+NdA!uMqPtx{
z`(}oXPE45E#X|k(k7cr_HQskstg&%Vfb}3j!HKJ9YcYb#Z@qe_8$LVo62v==`3xBe*5p6oDZ?F1L#`
zksfpq&=8RQj-KN;Y2wjQRgx+;+Ix)rD&E0jL)@Opz-f4fT9Jwq5))j?%AXx~%ZN(Z
zDKI;&r%>ho#1DH(-BD9GrYcjWPk+cZ5M0GC@WBejfb)IfLtQye`j_O>u2*>yMm@(>
zB$CpTS0PJKq(mYAOe1EM5`*vsAr@)C+1aq?7DEH4$NYk=-2T9tnwmDhx9I}3d>2hq
z#~j1~ySRS6jJbSCOC`^!@fu&1;IMu`<=ga}nAE<&q@jVf
zM!uS!^Ou~O+crWctJp)
zBt|7bpC&W7D%WD!3rCg`q?XZ2ySS=Ui>8Lf%0GVQv$o+RWM^
z;i{FH`$8;|eoX}+II8}YjMTcm5&suDkbX4h3}_6QfAz=-2ym4$Dsn&|E0=RCAtb%>
zN-I}lx4d^lJ6#VJqJtZ4cCv9xq2Jd(_E%sQgT~~f5{ju2Q{c&eYm1b-9sD`m)jD6VT&osPetM!~?8qlWxT{H|n(skl{=zY5hhlh;
z?_ajU^!q2#EzU{~9mW_QbL$lSoGkw=by+%U6$_xljuG06E3n(b=u^#Td}K%2{UzrH
zvPySE!{hw@QPu3zp#y^i%a*Yd;erW00alv;&Q4^0hJ1ccGdGfAVN35msW)`Li;0u(
zvJ#?8Z8UIDlj^T2WD{Hue`rY*u{J0J2T(>{hIkH%JQB08xGy*TOMmaOw5$%ow|;+s
zZ2KQ&&A+X*8N^0h82yuD$o=vDcW$EVn~BV512n4lL3Dl1aRKqT@{p?pzXeM<@5>*<
za}}pMz38P+_Y^-n9$wMe0PugjHKp)FQ3@ZpH3OCu@p%PVPboPm;W4FS^oj;IUehDm
z)#9q){+6`7m4l|q(dZacMPsl0jbPir--@(b>#cY$ybmud2la~_ODZ5LNK9<37Ja_T
zi3@q`c{G_a#u=Kn&2@ZRgG)om58Pk!N+cj`8g=%WlaMO!M_x6<17ZC4qgeRBGDCEK
z{+gl2{XL0Sj1WTB;^)yQ
zr2-MAs7`gAJM_z@YcHp!_5;wHK_W}QZ^z(M?(7j!&J>2umK(`i>bNwv+qz}%4)yHN
zk)56>+0P3lCw^cwRo;_G>94k;(RK9Fy@G36UmdC2JaOH>b&PYO`AIBbqHrxUrzkJa
z{h#~Yz^#7Nxxa4@>nz+?E5}g9i|xq4*KR9u+RZ%Km}2_FztfgJjtj=0E-i}xIkFF_
zC@&^!wMls$nL+%vcdlvbu?ur~3Ak@+9t_ZeWC2tnsP`}GoSI^_8{EA+1+&9}VowwT
zIX`n&=xbf)PmlLPdKW_o?~mr*Iqi9*js0(iN=~9G(Ji_)cyO!3xFjwoIzQ(>v
zRVDobvH=W1(tPk|ZyfScT(<+B=z{c2@7&{8?|7j>3C_fREjg_09fRwyE4ZAW=;dZz
zp-y)P-2+*O5yq+TjN!Zsxztm-FL2cc^??8fa(Y7&hTA(@_uo6xi)@0;Q->wahn|Lf
z_k!*h<5KQa0)5QPZXeHQGv*tuUy2D;hpA-^<(|gNd`R_$cm#PRJL(%
zRdnhXy({4suf?V;gR8PWJ#5u?IUkSAt0G~xsq=CwPaR0HU|p>nZWz1B;=7({I2vuE
z4O6qY3P#Ogc#gHRm5&xsIIfVJbpEqoW9xVzhEnwI?5o?2dq#8J4Xn28vJT~w2H5>U
zg8yS0ROklm>Avtf`es*?`teXwH4>ANh7sffon;y?WUPoZx
z^gN6cl{mM%h+ZwkQ6QK30eCZ=_+$n4Yb*Gv!
zF3slY*9~gD3K{cs2d)A|sNLoryqNts`c~r&{Mh<38)zXgrE2pdFBG^wT?28^Tnl{J
zn>~N_`yPqai_E{ZM>&g^#@i3FtzT4Jo!u0d>7CgLW#1k82@-9t?bAAnO!tslBJ;Z;
zZCrn381usB3l!tM!dPyArMiL=kV=W-UzfYak%6f%J3Qnj~Eza~jEkobRu~iAy`cXn#cj*A$j&OA7>wK8M>P
zOK^D9sJU}cY>S|vdTsR9B7T$sqlmdJD=+Ad~hq$kOvU9NAr3reeW)B
zfV_bFNfp!|a2G7Lr4OFJS#(l<&+VY{gUdR?=mJ&i9Z+c6$Ib4})?H3xU
z!b*Or$P+Cu7K^Le9^a@P59d;eVvbeDRYpU
z;ZVua#mcy-hSF8{xsF$MQ##$GqyqSdZJ14mf|e71yyC?Ir*$hMf+j(DjGyhUAFNpE
zpdA5auvD3aukKD8Yw+~&WWddE(F<#9T^BJOEpG
z7DauqCra-EAtUGF?5IlV>S>~1F2PPR#Kx*LGj_3_f-B{>sTM3cU%=eU4}9j
z&R?r3V*362$OiYapJ(xiq+y6;V$|i{P>gLx#&)XVh3$}IgQ;{@Z{aLuuqq?AZ(oRw
zW5p&2VUY&=u-MFcH*7eJJw$CIxpdYw={H%}4QP$_q29koqorfBzHMoQjU5$S^D@N%
zW9k7tjR$?U;cCz0l}jj29=T}FyPp_cBCh&Bq5`gDf?KT9$j??vichF@QIiXl*xfO|
z%P3{8M%gv7EPVadJcvqK|DapGc*g5L^S~M*&A@STqWQ;cRm!vrX?uFK!TI*~^?Lse
z5|_rvG-fdC0fH@3^ot^H&Z6y!>44RK%2YNbI$JiHfw3X>N?RFG>5O`Ab7jF4GrwOW
zh;vTrN#NNAnerwSdZo`NYs!53wgV?G9mT7LJg1?fbv5cL+dSUIW~am|
zhDD7&f&w$3I-#n0z_1310M$Z~F2vtqa4jhcg^_c0)b!Iw;#*)oWHAk0WZY@RQt@cG-0&#Zn@gF~`uSP^QiL?9vbg5&wHyHPF
zcbJX0$IK@ri~D$zy${HKYpO~jD=#rLZ>1Qc^ILcM^tEFc6I?S*fgCXDe|u)lOGi}}
z+Y>=%yt;+WP($c?H2r>S&+&1YJkh_WoR(58;$5=Z5J0Sq
zYa?W98!wJ;u-R%-fTLo!_WJW$WeSaa)i3bV{^wU)i+a2pBw(~9$-$UysAW1LmyKk#pkW-xmi
zt9qxaILxs!okF*`-qOn$-1&?>!tX%tF)cn^s
zLbcTPq*phOQBc^n@iO+O3z1`r?ZiU(~w)Ke^PN^kk
zjGD9iie374&Pa+};bFFc>z3de0gsRz5?@`^gK2gg+0C91y12$+_+}EbxcH_hE01m3
z#MjcAXG@4`L{W}uCfE@C@RGlhDeE-l@%@}#2C0W7!xYVoyk5QrGrL`)#7200DktO%
zCVO}NlQq6AgFdQjkX^}<*?9pggqAHESlZI~T1Yj8FH&@ms*L236B01ga(UOVTFl^F
z7>vmFlsRGgt3PoVcZ-UiqiB^b*EYmRIE8vf-bb1YF5ihWbAKW`(#GP&)q}jzG{@?Y
zXv-$8p8Sx_Y}nr=9dp_s2{4=L+Y66xSO$0;)bdoomf`t)`}CKK>0bEZ-2HB39TsC0
zcQ&Cu*{-Z4-SafK^bC1BhZdog#j^|yXui(l~I^H8!8t}5H?l8dx2
z3Gd6}orujZkf~Uun4bdfL^FDx{c0gNI0r|vewwI%lwxSzu9RlY5Atmm#aEfi$o2E5
z`NxV7rS;b$cILibPTi+ZN}_f>E=hEv$Wp_3E~(}vDsi0V2L7q@!mu3S3eIFVd=;OX
zf=f5p#Jq%b8?2Uq*w4@$#A>7*K%k;Io+EsRh2d_l)^3M+GN-ov#YUD<^n*inpzR+!
zUHi0K&&Dw
zA99}zeVRh?B3pfud;%>8+g=~nk!4pbT$C6X3gvU`vhu&Iv_Tvi_LE$RMI_wlUK2K<
zyR;PO>ruit)HKMmY${XbA8)4%$eKdHG4ybxymMltbVkx{ieIAl3b$
zbnsWg-`x<}Cho9Vw;z@IAR;PCenG8~+h~&^;j=ewiMV(U%nzA{R~pUY$^|P}nN0smzHmY57eF#%%H@c9a6T
z(AFQOI1H}K$7u(BcLrQ$7A`%tf^rgfAXPykXY;$f^H?z{{~poqzpeVb(`6471zQs)
z?yd&6<>s?p>e+|m>UZ)(Z5KfevsRTlR@Z{WEd|TtDo_lU-CV{a@xl!F^346>q{eB2
z;JgtTJlGt6UfuBdTf@qhG$}gk4N~WoME~
z%1N+P#d>Y#oXccoI^XY{v@g&I;g{X52TuBS5;LO-^eg*^Q$}d2l(1~L#LT{SM{f20
zdC@qAUQp=G3@^7pP13ZW6vT*N5sw`S)!dhAy*DZ$r0^_D5J*ooSrE+sGR2+N+D)iL
z{~J7u7VD;jFPQn=*BO3cHFiXVJ`BFfTed!@lp$cQkQ?9FE>^R^#%!X0c6pJr4}4p=
zpt5A`uy|V6iJ{tw25i)DmNjgBz)%NZq01kA#EkBwF#5mLsM6@5S3Ba7^dl9by5$+s>s43n%Dox$>e^B#>+x+(YE&tHp1QQRyD_2V8Yu^J
zV^0`6U^61x)W3A6Is5$==|$aB9|Z<&XZg;gQHlQ(o(@F`aZw;!D(C|9jl>hNaZ9ds
zv^E?ITuGfmqT86on_BJBV>j$duxxHL(jW50Y37WpOd%Ej_8B+9EjFBD_V3Mn0;SC!
zObCpU(64&TMh@r-x4G8-J>2B&xqE)XS9`di5w^0Er}b~8J6?Ah%XW8YMKKytM
zrMH(rZxmljo4(>$B84>V-B7{7Qc;ENmezlMX*U=--*VCzDI*UY@%7bs(Q+-;-jIhQ
z#i1EPQd~8bjPdPD1B3(*wGxpOlM<;NxttGXds*974_pZp!7x3A+`$SV;RDE`oz@sP
ztqo8hCx_`c=qm&WEh704mOcR2si$nRZQej)N=PE$#{Z9$eL_6R$4FVW%8I~2K{y#n
zl}y=*gki(swhimPt)u>?9JA=1Z7XRhBzMDb+2)4JOjdDuQGb!Eubzh^)x%3mkg4ZX
zYw7Bgj9W0FB@Lk
zma%VUc0$GIQWp<8@+_czTN#b#W@Oc&2eSo~}
zwAI@?2jciCNqv~n>56|xSU3wo{E^<%lNL}vkRCJ~)r*JMcU)~CwT?b@SN%~c^tKfO
zR_LCu3`JeUK1VSL#D)gw%niV2JBegtv=SXUhaJze!{h76IYfMgTwUDj&%K8459EEP
z)7Ved)!}QnERA0@q
zF19w^`=WammdcfDZDQyCPfiI5JXl)F@{wE^$TZ;sN81R#nS@)A&;zEw;ehMNe(Vor
z#NA70HXM4`f~vt%ec-?Y@h?(1Q+)*d5o?`9lu&R&Dt*Mk(b_naB^HSh(0*7oiY#6{
zTl>DySWrtrL|u#59E?;Ao)GRpQnh*cbYp&Du>1CX95Hrw#~yF_;rwZ^i4a1@Ugszo7Vt&6}YU
z;Z?!MA#iu;B{@17<(tEd9!zxEKwISC*svOvmps@WqoilyHnE#pQl?)xf3~M>>DJ1Y
zG|?b6w}>2V>TEhuOzD;Ny)%%NBhhQm6asPKT+=z0asK3yGQ$iB9TGz)Me=zGO8a!v
zbCtLLy_ntqj%C>wtkA@1&OS-`+gDP(-RNxN-U8tXg#+e;+zrMqB|7kl#LSk`aeN})
zW(KuMy&~QV+PSN%_k2#5xu;u(tTmsG4Dfm!f}E+3Su~B}er7WBI6lHsf!uRcu!1u1
zTWW(dMGpRMZ0>6Kc(nAZxQ-4)RXN7vE%E93RPL`3!SGV+l%>nr6l1EPJ$F#-a!qc!
z0HA%nas}4agbBM=n7fg_$WxxpjA5l{uhGLt5`>ljhk(TuHVE4A12-+|r=#XJnnSLz
z(|a4cJLHdJIWSM#<|zujOjT6oGeJ4=7)-lfVyx8R!Ej
z<}D8kInHhD8RTbT75`-A5JuKMt6UM>b573x`OeJ$nxOmb3H4zE-N`NrYXJB@5!?r9
LC8=@=qoDr+QV)|{

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_3_0.png b/doc/LectureNotes/_build/html/_images/chapter1_3_0.png
index 0be2e475d8737b746dd714ffe5682340949646be..656724cd5f5e082dd72dbf223555b261cfb27430 100644
GIT binary patch
literal 10355
zcmd6NXH-)`wC+g=y%(uc#2~0N4ILyPy*H&-#UMTO9)g7;AP}U4u7HU2qI8G~0uqV{
z(jhi}#gZcine?-j#FKnar8J=d3-uelu~#hFbJA+%y0H(CcWcn*spn3F&61
zq9FCS(&1_-$#IflRwhK+uy_Wfk5CRUq4rGFPJ##Co1sR-yi9x
zAR*!T{|1PA`?^T!fEGK92Om=%v
zHhUM8qW;WTR>|`y?M*GSMwgY2HY1~k?vRY({k(kBB1YXphS4`hX`Qcc=;eNsfV?ig
zM=cXTr0OIt&GMK04UO_j%y=z+tF9aGaq{lR
zFOJ?>=(2h)19cu)^(%Jb8o&qH4g<2u#<_74pz@cu?Q|X@a`jHvXfSQ_m1~rM7tNn>
z!FCa*9t!|Xok(CYv^D1gSB+x&oe9vEfI(ytd+%9
z%}M4&I81R4&z?OIhLHJq?=Ie>9^2-O$PcM%Fl!Pv6KoR>TRFqy&m9pzS1*LKCOlv#
zBlx%3R%&Gzb`KG|t-D#Sh&hkFj(Ge!A1wO;q7S${XWy)i0ujW4mio}d3jDe~%*Gk9
zl{H~xb5<3Domnk*|A@^c(b}9{+<~1mVcWot&#?hAbu(>i`&ew72jYayHbZ_kY~O1$
zLb6T*JCSPdiLAlSSRpssiD&Jl*s}?}HCxyTiO4OYzkglUK6FmHPVKbSfrXIv#jv{@5lplLb+
z6gUH>0mt{)fU>`x)?DbrkX2av?UxZ^Xf^G9P&Fkwqjjtif-3}$!JKp!Fwvz0KxS_k#)9(!
z3?J`ZFaxvC!m6qOq}dw!=AXAJ$v)5w&^Rps(x<5I!N1L#zy;<7lJ`D>I*>`y;tHY5
z$+k2EC%{1>`1C@kDT|nj#EzzD*!wA=d9;<6NAz94vqcsRJ~BP(
z1yBaQGQ9zSIC-^RrMZePO|Ox5<<*c&XfME;vc^Sc*2uTbVnJfb1+mYC9-Ob-t(iUh
z0-_TLxz#IeZ*%GlXoXzBS#ZTrxR5-*_*DsivEGo`nn-|f1;ZG!5<;f}L#wrQS5j6N
zV)q1^hQhW0M=>-7V3T62v9t2|X{J?fM&+(H%5dinVDUkL1IX0#z1_mk
zXj1mrJNah5;tn7U$`-{yjz}T@%!$EP1h#ngLiYL1KTG&spY^%qX}U`P3-X<)A`s4;
zd^t>c2fJ)eUgfZES+PYTRsXcMLGG9wbk@Phamb!EA=MXToxhm7yNlS*!sc2M<<3C<
zQDIxYOP^Z|^XSJiiMm^3Kl91rZjpmspAvb1a|j2M+v}<^sT4CtN7a<>elp1vBBH|U
zj?SzJvF5{VMw||L_qys1R~4Laxe*iMAOUlr$#7jdv~M0OOUwar1cwH9qZ$a~5!y?-
z+qCmpZobiMVG~dMnkoo;p((9+#dCiQ<>$1KIdoYBnivC()c3vFh-g~#Sp04opHd}H
z9QhJ}(xVY!a?ZhedLvx06OLP3SCF2AFt^}Lhb$`#2|r$X!lqynmX6>I<8d;Sfbl#+
zL`;UvR3d)1jaF3R{p}bDO0X%=_|TQmX9pe1z)z4Rq@WIK2RNqQcIcq5YC?c)VdJ}i
z@_|Bl7-=oB8;@tJ?HXVH+QpYwVLj;y!o01gY(NZl;?t~qfb5tRd>l`aHNhz%P(@LF
zxw4cB5vab|M!9|67HQ>>yvZ(chqb>vY0Q6rZuXWCIU5zxukmI3GTwvS7RV$glykS8
z#~``z;6{eg)HZ7^9%we`loNNc%Xh(`z!^tIkZxQ`pWb-3&)@$2))pC39^H}2a|EMR
z@v86h2a<3BZwJY%Y0k5rDEJE58Wh58jidaQ%p)UhP9-@DV(?KR1qIbumL;KQ+ecs_IE#o
zR@lBJsKde-E2-a*;}kH7PM0m9pPE1S7mrimT!3>hp;?3#x^hYPM=}Q$w~jc&siUBv
zDZ4a%T&%OLrSy}a6#lG0vR)kxJ{X^uC2#27sjyQhC`n%zKD%wjG_d4Hy;nE@@uK9
z-zc0f{IA04e;#E2C-rhz3rN-}z{0-c`RLX%1qFv|B;6q{Kql|PA9eN47-Mi0mYzbA
z21i;gBX#vt+A>X~G};d+m)HYfJl=?Y#Lw2p-`^x-gaidif`ReS6CPq2Tb)Y;t
zG7OxO0>7Q(Vq=99xIg6q46=0xz$rNng^&><2~#^uS{kGl1I|@w$~Q1DDG{jwGeV$)
zSLHK_#()+mIETpiM7c7y$kL2e5~U;=B(Bn&zu0|~^PcuDl{z%!nm%A&=@B;fEo~EZ
z-DX*mQFDhMIA>nEr2&0D8`;r%Qf&90IaYgTWGo}7qh2pcMIGa^pkClI`0-$GePuJF
z_Bs>kZ2~W?eilSZ3P5Zw7wP=t^e%#wB1yvn%Wt1^n~Iv6mVH)T4R`Abg8BQMkp&*NJE4v);aFl107~kR%P`j&eHT5I7&PHv0yseYnA^q
zdgLYGFYL|LF>D+1I7+)>)e^+iviXcC+MZ$pzl
z4q~9Epp7Se&i~df7Kf|^4jPj2Xe+tiwtsx@DpN}Ew@K$#DtGulp(s5f^>%*XA0-Jg
zuU3jnVqr1~92MJXmVRwf!}i2P;NS3J=hj$vI1PkFf?HgqB9WK2^?j%NC{QV7u^*(DGXbJGbX5(S8yr+EN#ZLOav;+5vPF
z)!En$OX96mZ5fQmWl2XB&R$Go?=VvJkA%CPK|MQZg6)BYM25$KAaoiu(roqCswjI3
zy+VMmbktphmt`sNuCh*Uo10`Xl6C#9b76oS6OKof-$LE=ZiH48ZK3O9CbTeq{SqL7
z+hbPZ{OqGKe6=jM7=8TmF4N)mzXbO%bf}wIr=r}b{-*TA4ccWs%+LwdsWE?)_RXxv
z!pfb^Nh<2^i{b)VB+0=kjb<*TM97Fmk%|CI(5mPqfuXnnlJ5DG#07#O3N<9ff>PJk
zibt{^d2UiyS7}G76q(n`{DiwN^E~s-3Ing-G!kp5wpURS>V4ru?9D_@kh
zn6qbj0-f@2MXNn)FRtUdZKpV~)ii
z+?KK=@n>Jye$D_ROE?bK%XJ?<7GJn6mFc|h^(C6VPFI2IlK@nG4Do_@Ya>@JXH<&?)0^44Mft%?BF!s}XgFB!Xij?$2x=N+CXyf#3#f>B3Wx=L)3Y%ZmK@m+7j1PAZNy`nR*Q
zE?V2fM4^ctI-9fvRUj#I;`!|C7eSV1y1EBG1Au!)p9Mvj$B(f?+;
zqOAAev^}t~^BHynAOx_76kz+W9XWh3e3hQb#C+XIT2J3!F8meIed9Fjo)G0F=9I&C
zZgJEYKeseeT{o^h9I7E*Kw1vLimu=V3*-9VIkDRAxAG3}kF1X39wy!2WBKQiGHMXX
z4F3jXAH0u@bUZ{J>=rzIA8JKXBvHM|+J2|l#ZHOgep$KsXQyF%y}~#=7U$IhGAl9P
z+xbVIUXMtY?0mE^_AL!Z#s~=KLg=`_rwbgo$q%0VYT%0_M&!8DeO_IX*7Yq+)3Kz9
z2&39Al()O#;6hE%*)b@FpJ5uMSbeg&0WZ;=th0pn)l)~*TU(=-^C|hlhi2?Ux@I~y
zK32Yhhe~btdxK}1(@<5%dt{*pnv2s-kFQl=oZo2*oR^y2idYz?!wBhRfP?1f6L=^+
znuMym`Eb87$RbX_7WUNmJo?sM2ISr-4n)}~0I)??HT#2ND*dwHOL(ixK#nYbj1*rY3
zWMQDTkd(y)b*B+d2UV(XmUK97SQA!Cb`6xh+kS1%IZ=*H!$Rxb54LQ7<;)u9+HwLz
zRQIN(NEUite~kO!w3H5J*U9#!g~xM{F~ZkxpKI}f(#6P{1IhPi#vYxL1U2QsYROGf
zZtDkkPvds&s|8hmKUTj)jn7Jx*n^*xw|Mu7MO}S~#9n|XfwnNpjIVL5vXvk132c)J
zb2Z?U3#S4=B;$zEB)8*MK2B+v`vQX~rh$`D042Fgw=ZmsIc4Y?3E+&Qt>&^a)tD<|
z>e5bk19-ot1&SZGfjA+K7F!3(z$|!sbU&|fU?J1AVv9
zC+0VFvE%s9f$2#2b|F8=#`DsHH&2<}TxziNM`p#e7P(-
z43uNV2_XRoP)s6}hmtM+vmguqFHr;L?zq^lpO)5b+`&6Kvb8SX*a(YBGfYEVut^G#
zK)~9q=eON@@Xa{#sB4%=`o&B^7J-h!o`w3}zUBZNY_QRAcHjE4Z=cfNwPh5-@y=VE
zI|P9Myugy#MDM4{d9lxMTf=N7RbI``aW?fEnGbEJ&U;e4R-!j&^x!3u5k9u*&0jx1
z=ALB5)d2gXAk{kc`U)ul4p|saFin`dr0_-DAQiV(MP^|OT5oGuTjy0|Kb)FF-+G7+
z+w@`W@M`s0R58C1VP%^)xF)uIS;G>F(3eCr+w7jL@R;?{zBaFeP2%tGL~7yWJE4b)h!@IgG!Tq|YoA>Ii~J<3k?UoKV!A)p
z#!_J}n2u8+`S9TLNjg6Lyztvcgr~$8mo{sDp~P8?Cl3m<82fE
z;!pR?__=Zf`YB@|*Tt2X#_#YAovCF($mDDNg|b3sDOa9;=3n+rCa$hN4^IZ+b7`9~
z3<0#wZF}sG)!9$J;chO!P*nwf~7;V|}Md0p}paUr1$XuR(3dMTA9Hp4_c*%=^i{AHOOZcGmVU
zcJsZ`x0bTWV)$0G!Won~$2~|P3j28VYOENRaMx9}6spo^D^VEBm@
zs_biX^Wz`a0XC}6B-1`f0?#svQIPZ97l{ydv2Wg^auLl_?`v#i8}2W1xinUA$1o%+
zq?tZbh%bIvtUE6z_NhWOKboPVAR|6DRGx+<_~;Q~STE|%K0v|X|NLo5KE+F=;fk?~
zUSV;$oWQInDmnKjo0y0Oxm~x++@2T?u{L6+=Q&0aMyI8F4WwwpG}phS5$j!v*GrPs
zqKQd&!W*;KN2m2a)uG3^$iAK43_q7)w+ONq|C4a()ZvKv-x*;-{tz{qCX48J9
z!C$S)-YX;~+e?r`^xfI5VvkV>c#m1T+cyGO{MxisaQA~(L(&iK{WO|4buE?-24W6y
z;fDJUD+>>bGp5hfVwx+sw2}h*}uwEU$7@T
zEBb+G!nJHE(ROn${;2%!pb?87Cj6>35B*o}ZdR+IdYAfTt|lpGdAJF3NK~!}r8ufa
zKrlC4t?t+76s#=g#6Kc76|4x0jE!-?HJ1m76E3Mn$~u*HHhZpmQOUMyR|CSjV}x}fi?yL#x6CTKgMgk3|{Z8%?uVLu)n`2&HCHC!WR}|
z1m_(Ov}s8A(zM*XcPgw8`t~)B1h3H6~-j)ZWuAW`_1;yFrFuOa0{h
zCSO8M+r#4x?Ks;A*Xht=+^PQ3!{dc~cu{so^VMfBJ~}t_-Mem(6qn;JFB*st^q1b{
zdeQwW_4SQbEgW|0GBQhpVLeQ;eL<5_vTL$ppZO%2M*fUw6{r}SS>aE0b&lV?c>rn~
zkLb%tJ(x2(|E)zVP@Oiud-hefCyuo@!=Vy2bKXc3M&*+{R|QGn9!5cQUhf_4Vp@eg
z=gVDz-@{$S5&c=G5u`Hkgp>~$$hfxLUqplEOB6j!!e_tC`W1eCKar%8+ohB8%CJQ8
z-feeT@fQKQm3C7yRDgH^sg{%RcBrRnToWfWE-vsu+-7Qcx>oC~cnXGIfK^SQ^1Edp
zpb{o_h6MZe%j4eg0{1b(uo-QW^}uvf<=b={p7&Xf$z5mJSBq4V*-Q{9ld3QW5mzjP
zy7lN9a`gVMN2{ZdBPqHzJd*m3KPgZ`Ld;xzs7(^Mb$x{cUKN&`S6$-|OHf**+tG9Q
z=FoCQZ5qSyeC^DC!@9*iC;fGXDLbj~gIp`U>i>qykv~Fx)|d9EZ7U_>xbb1&nkHYU
zNXSe8ak}ZK{)kKZUTHmD+pF4ptW?9d+(hQv&T7&}m=&B76t2veCpHC~n>i
zJsbS}y)v=GqH6N;U{&>8-Gi&IL-!Mc{OoA%!`wu6h9hfitfZenW$mDR1*`hcimUk%={b^7(;>^JyA2c3rz=LiTKqy574LV5VDt4mX0NZ0*78#QofkL4
zyx$(BIPx6`STWt%Z0Yq`>;aKY`>d|^v>jZlYJ3uYw)ZYehso{f)v9ZMD+x_N#KM(c&mPOl<(L>~q3mPQi49G?iSGfuYGAsSlJcE=%H
zRryphWYX0WwrCe5B+-ZtKM!fB(Hp{8F5b|ZB(vkZcU4;_QDB8M<4*)l#5PDigRPK1EKQ;9#TeN1_SaN(_%q|y9~pG&MIYoa`H0RjM$68#+b3C?6B#PHq`{#lkpVo)yc
z*E4sKr*ud-9mZ#St!kc42l6yM{{D54T2Zbiydl)#?^yl12Y5wa)gD`$c;S!|C8Wa)zo*J@Z8EuUrMf<3gA2Y-K
zjBq+pE3p42UBcy+bT2+>n!}xpvkv=Ao?NY((&i#PmHaE7_7Z>_$pXr8ODYgNa3{__Uz{-
z^a}50f5c-KH|IiJ2-ALcy^)8=g38L{hreXt)DQjM93L+fH+?>PPAlu)>-l?Dp$%hz
zf~#FI{gFH}pv4bjyh9`v3jE~h5|qtBl>2k=IU>wid!)4d<=jl;$TMg$q`uH*KTyQX
z#Z#%=VJnc>eE6V2&7yMAeXU}K!Jbix5cpS@2c8e_mrdD!JYt*{rBWz!N`1q!(o)qh
zSNA3V*6mZf!$*uR?s+rwlm)2>#Z}@ka?}Jr`pPLNdobC?k
zr1DKQn7qJfnfyA)p3tzm8vCA|8(~m1axzrqutRIdWnv&O#KLNlDfxQCnFV)t$La1~
zl>xY_eHAh8)%KxfE7fS4`1a~cV6&y_Z&9TPLgqNvYi`K+)Tel}s<`IijB|c81rC2!
z?z@vDQ9PR@Fk52U=Wpwmm<=?L0Ju`C?<@CQA=EqR$K0wFdZGUu!QbSmg5q}Ni_pYG
zq6P;EW+N?Bf*Cx1j$1Wn_;Pynpgg@nB^QM>y^y}xBx-pu$mTX5DtM}byW`Ar=POSy
zIqhU5%C`OQFXuoX9ly-l-~m%+@+U>5MMY!73o=d}1aQOkur=!PLm=5_c3s`7tLkgSes9~Hgn+Py?BK*77kE_^K}rdW5A@FIsbiu*vbDWzKF1k
ziMpAyDetY@=idweCA!W2aWc6qM%J4OSP}rYhb94V&&fUQ6CHFAbqFd}d)muJzdMtD<7=)^NRINVEihfcX8bA`>T*TN+N1yo7
zwfy-ncGNmWraHf`8t>lsF))yxvA-Vy@%b|!Ls|&in$Mo2!8i~pQc*I#R7k_fNDmmh
ziKF+C6PD6ZDh6t!0#O_e_>5zR;7>E~U`gb9jri_d&|G54X8g$Tc+H=0hqOYkk8Y5V
z=x^UgAUQ#Nl99?oWgJ6Iu;d6CA})0y;VDsip&M
z+HbUk%NtuPD-$wo7c?(od%pU9NRyLwc$9NY^}^K^louvGN8DX4E9TGMV~}8ud;Yio
zYVh8u<5!*r+4PVVQ$l46dhsP01p{P%nzeC(rXVts@V-n}cCXX3HSmMftRmN6WgW5m
z%`zIv4Nj_}EjAvWNt-jh4_{t6n8{e5#`4H4$n9J13@G**!TaHG+Atv(pVDRM@Cj5y
zAi$j-TVwFWSS-syH}Me#x++2p{xPzXYHZ&@gnIEqS=zHdvjzq@QX$kI2sCUCOOF*T
z-`T!Oy_Nb*JYtuMhD=p@F_A+I%!lISl&w6x=nL6sOd6{+pY9Hml?DmtQI%W$*gqGy
zL)&=ijD>yAYwR#$iTg|W#8+40??*R$z~6z)`OUSF(!&h(;&-i$8B$Yaqg_IYPO&E^
z&=hS47aw=jSDwf}&$GXWt(I=~RqqZ|65G9ak>54`VPq8z>~pF2A7XcPH*9z^+(+!Y
z!zUFsrdSUSbTB2u1J4D#zcdGK1g9p3?Dnix32P-?3;T=YMYM&c%pEa={4gMXe$E5e
z;tBQ_9T`N0aW8?{9gSFxutZ?xO$G&)eeA@$CsZ+CldlDft9%iOSR|0Nx-HgdGk@
z0CHKxwc@-VmJWkU8z24pad^lC8|}w*o#|jyD~N2Hp;sPo
zL_~rlDO7q@&>_3%*|sRyAl`d)+^SEo>#k{Fr<)P?fND|n=N`KT3dyR`xm{3#NO&?f
zk&@tP^K4WmY`>g-N*ZAk3B(r~st*|A#+m^;+ARmY%8;(|Tn+4d^SU*rdHE6Da%EQL
zA0K+ADu;!?zRfi7WkKtZX6{cj_Zik4`}{qihxC$k!RlUEtkON?`$P#u*B^&f2j$1c
zcB*hj5y$}tnL1B=`Au{opie*FC@8p~Mj&ktj$tQ^0DVehN&*5?6|h~eqgrxP)nDwg
zU5t`lJn3qEQKca|#?8Wu7h;ugl&aRO>6AU=wiDI>F~@51MgoyG5osrR^6X0
z>Aoei-LaG1@vXs|^=x?cM{CAU7_6yU;M`
zS$@0(SunGT!`7DA8`9sg2P|l_*0~NELfg9^yo9}R9$Y1WNISfE4e(Jo0)dp&km|UDleYl-v&?qmlyDjxCXnwo1ulOD4^(5Z
z^ln?yg*1?^^xX00k%!uqWYugc9mb{{;_$IIANW6OoJg&1TL+I$UxAi9$5vAv9@&PI
z<1Fy&4hWeq`1VWItDk=@TiZeiln5DqoEl)rDC}Bu&nu55U+7xVL&QEAkO*jM=BA5e
zS_uOkeec3oAQs@{aoWrZ>KqBTVXHq6Y;G+rEBeB1Jpsf|
z{>swFhOw@6@%iq5l+kA(9R8eEoJo{MI_=M`OE{m|->+vtqC!p8ZNloqQk^V)hM5yx
z$d+||awMzm-X?G3%ap=5+U`T)_VG7Ce<_d+V-DxayB
z>_9t%Kme>os70#IvQroq#^P0)6rqjc3kX$r(=bpIR3=FdpaWNHs;UfYlJ={}W;YED
zkcx0Iqdydq7!%Sy$A)BG9&k=W5XIEZf9!3QoV*t|l7393PTHuI{%?F-O2WxSv9^g{
Vp2u1$=>Q<0qhY9CdDrpD{{n*maYX#cdBveHyp(_wdq$5=zfbta(fzZ21la5jXNQv+T
z1*Am*>7fY$0vLL~@%P?&=g!=jJNM6b&Yn4Y_H23fJY}E#JRAMMP>1OP_XPj|nBXui
zV*mg}P;L%78cGR@DPfB8xb1u2-1m_O(%1iqw;>O80?TpqPO4;SqUXtD%siypZ_~s9=oW58SC1aNW+{w
zjkT3+AagMFIKehD{ccN!GDB!E9zussm^E0{)xb+VAs`Lo>OyRhb@vxo%%Xeb=Adz4
zfyVyR$N5!P4|y=3ORTH3;S<_;uuEF!{IbRaPbik!lL_qsPN%XH1D1h+tNq^fJeVAL
z{-ex7H`cv1d!Q|i(9q94uVGhB1HjE-t7$vgg=^4hY5*E99B5DMd8l_H4BZrN3SVr3
z+FsecHkx$8*5fto!y+L_V_rlK*Zy!&+uGL@^H*o0ina|7=KVes)Jb1?cq>MJ+dK&mT5wUMJsPS
zvS=`T(H(luwM%1`bTaoY1Z%l-48^Fw?rn}(bb|Xm-tfUJ-(LzzSXAadY?0*QrCYZF
zP9GL+Gh-&@x77aue$nE10Vlv5LU0B?1L;hJ4>8Mayg0d+9DhCn9Em4sUu0n39zYs!
za06WmNJBj3d(z_e;!Koyac#gJ#8c~WXhJpvK^8dUL=%BC2q7ls{4;z0d@dEi8&DT*
zGU;0I0i8f_4nXVf1r2Z1%Tw>L9e-N%ISaw?Wym%MARrb9+mUesrPJG)0kp#FvtQm)
zd79Uq@&PAQIL-ocC>72KZW!NiDg|sN$g$zh0Z*MB?q#Y*W}F}3r@eB;blVL`dQax1
zBY=Uo)d4TUfPKLiqqg9!)?PGf{5M4&Fr(mrO8}^m4kwN=OPb0bXb&;K2f_OQNovoW
zLRx4gcuN8ZW(hi5hPDb4U%vDX(TMErYcYK(>
zaux#8Purn}%V;+p!~(omR6`+X8F-KdzJQMX&~6s91FItyfKH6@G6m!l8Vv#Ty@F+T
z(2cCc0JxD>9v|?GYKX~B2R^9R?6sv0dmG*Rytt4F!%wvp7qd?VNIookgJ70Oin
zOaDwI27Q{}k7Hk2quCN*pR@#{>7;xUPPCF!5uB=O^!Z(uWEgl-%?R)Ca`e>a8ZE{h
zP-YHl&(mxoPk9ZknN6B5g-io8iYvP_5IgO<^VxXTgIvKwhE&?h!Phg}wlHbo!!P2K
z88mj19ZqJSsD?vhTI%9O+}?y%sxV7-^3pv6;9<0$*Y9dLzaqW=sRidme=S$pWCIvP
z45}Ip5`6EcEI0HD18OSHSqmbkx4BuA!1<1C){ir)?|qQdXW-YY{WQ
zvf=H*Ks8snxA=^_8GFXx6qAG)CG+6k${1ykl4E+BEh8ZZ$`WmZl8E|plmp+;@OHnP
zU<+(9-Yuhp&YB&qgH+;Jtr{4`{lgyTzJiKrm)3^GH@Ms;XxQ%uta@&2
zJ6%WPpQWKdC)`hNgixcGht0*Em27oZTzrKfU9Eu1>!H;)uB)dIfI^V-&Myx(nj(}T
zF=dCYgRtww8Zky;QS=tyrpAXx0UcNvA01(F#diPx!FnFZQy%@pHCx$J9PI>B0SMYJ
zzijJiAIwMr>nWh1w+esy8R_H@zS
zZ=SS1y??+1{O;GQwxZC_A1>nbCBmEa+IZv5DGy|+mtf7J%E{KZ4aHh5C029q=4Nfb
z_6(7Za2%%A^XiQ!9k^preY*WDgexzdpjZyj^fYiKRN7)O{Aq1+;AnyfYDUHE
z1bHRNJRZHsg~2AZh?cr#W{huP^qF#K2pbTr6B5Y;gR&+}UBNv8qz#`2Bax$xg+Y*6
zh$jtNsyaNu^z60&%pSl%3Z=G0--X*l}0Ru*>Qj99Vx6zTcLNv
zoB;$DxciUq-&cL&aTHPGrThN(3KzNQ%)
zyR#Q|3*B4$G3;|5K)Rw#tM|Cxl<Q|nhWNhUge2F${B$t)`(zpDCba3z;%zUFAK#7T8oeuu$|ag^b%}_z
z=0LmYY=A8!d@zBRfuN0P$J9Z?-Leq)GLyHF2I+vLRl0uQ{Nr-7AIR36U%`+IxzT65
zJkzUkm%dk2j+8^-?>_UG*Ig3;chv{UyNyo59p3zv5r4p(o}h?vFQw=>^cc2RX;Cj8
z{WJr?>uM4)J&jA1!rfz85~c(4&1C;d;6x2J2_~5EQ^rLGPzL{sUtAX-L_kY*j0bm!
zKa#GcW^`Q@OR>NAyPu&oYh$Bfg9HuY4==2TR?}|z0eVSO4H+VZkLjh&S@Bxd)>rB3
z5KF&m0=Fm5TDvIoH>9n(et~KM#LGx1_wpF~Zk!z1`w@m&VfzkvUg^0>SNcH!-YgnU
z%FIB3pRzzO8&oy^Q{T*JGhDN>y*_Rr&m!
zPKh<$5n(v`=_3c1EYe{=9t{phOl4jnNZ`GhOl`x?zikMBXsK$IhoduL@yvn8580oac<`MqOg`NB-FEg1aHLNs|;`LllEb)`OjNq
zNr(l{6~E!zn0C+>J=|aP0e|humPvDkfgE1=;4<0+bE4LB#f$TsWKzrJI?Au6sHUyR
zsRSAYoMbK{hU@?7kI5gejqJZwm4A~81My@*pVJZ^1N(_8q5}~bj@=Q|{ds)PQlE6v
zP+=570gSkF>Y7XF1%MY~r;dr8m&`_p23&VWY4a4a^AOp-)7B&vMv8t;}6dq@?5HOtb#aPo>**P}QNb2=k6<6&cUTu+ZU0Jj^eU0A<#w>hF@1wl*X^Bh
z%L~~{j2oKo!oQ?9R#NG3>v-wej+kivOv|5DWa;m}U7fRQANFT{ORZlJRUfb
z>%B8JCj2dFX6Adb}SdC}?Bd$~?KEC{3?l+9Y2ISA?y?gxChYy_8jmf)ojHIlu
z1FqN^3$cNpj)~o0pH9RxLsF%Uf~BIAN(dY-FC>dkDO|1>88+Up8(V+9z5Ukbf_vh?
z_G^^-TR+dM16j1OvVx@`rB6o$&hZJ7c?d$llh06s)?xCpiup6Cw4Z=gBF)h!$+*EU
zNjw(4oYmM-Jy)O#Hhmyp;r
zaJ*`*kip?I@JCuI%;$ub+(mCpA1s$+U!jtE%2+0~P<`|7=U3@YTU~1Ayw=MBQHAZ8
zY&wDhu$fxdGI}0R^GkAB*B+8W%%{IJ_3K0Nmyi_V7mDu=J3{Ro(*XX$R>be$V|rBw
zxOIl7?bUo6Fl)4ZH^vg4E29*vdNZ!J6s0KA_2u&Mo9#I%1`^Vg>^R8
zs#0y$giUJg#hbS_*=dC24yT8xS2K9GDJ7SI>|#@a-HVsJwlCi{%I|M(;Vp3Cb@DPG
zy&%I0W#wq)%?h=sN91L7@Y?7r8_i2kjhGN$x~88ifbq%$u0`(Jx4J*MT=M+Nw{&^1
zUokE1j9;09OOZL?Pgnl=*fdki98txlSyPAGtG77QOu|nB0>valI(;=m@=>%Med$w1+JvyH|&)){K}+P5R0%vlY%9
zRr~ch#0zeW22Itn;PD${9sLo0YuAhaWSC7ND=o`|3gYVajFb(U$+|9G7hL*MN1tJ7
zEJjXb#Bh7+h3pm~a_-|t`|#F%{`BAD
zj%izG#y|C4zlFbJgY*7!G3_2$a58J{6^Wd+2FCB?ihmU^)~B!aRB#hlw>f&-x?2gB
zIPIM}dvk^I4B2yWac?(K(PqUKHaEKzJ-{34)h89I7X&qJRP?SkKB(OMlYVn)_Uqe}
zy~%1nclAye!+w(0e>T}l@lh0mBnxvnflTSgqcF|zXi=TY$uom4cTi;>AcG5Py7NseZJj$ba22vE3@9x59&l==Ba_iQr
zcELE9TKgNt^h|mPzi^EpF>sc2ty1$rm_mJ})onSfEw@wochX=6c*Mp?|jb#Kh`FCN;KN_Iqr0kTT1#fD8Mfa
zdyq+cubbuHyPkq?zZM(d-LUZ4!-1ixP^Xf*%{Kg3HBoD?S388rN>x4VOhwa|y!{?A
zr@d_tQC1~+Ib#<;7TFnjdDO2m6YwJyicL4{Z#h}Iw@iu`5U@q8T5}#WEl>Q`>Z`D=
zRkG!7V!dz0`R(&K&0x41MTqK+?qMIe@2sd~yN
zv;8h&YOG|XqY-f(9LZ>FyeY^!>7OkMpMoYH)
zycPP#s@Nl0x7yU-fP|T@nXOb#V6pFzCVtd1tuJq6^vt;)(OI6MLHIInx{Y>6IX-dx
zNg0ad{pQ&W;|a+FTwN4=eoA4epwLF~!5qO=D#qeFj
z&neR%Bz*^CZ)w-Kns`|p--eV#-4fl1ZIQE#cBmIB6P0jDrgZE=J8zIxIO&ll-KcTr
zH@uUbdPtC6pCx~PADJ;+i7
z>!mTbygmxqfAzKd4Smz;LcEMj_g;*dT-v07%G=t%n>NoI&X<@Wq~h}OAcm%LiSaR*
zZmH093zQF3YF*pnSSTWu;n4+BOY3^OVY{raZiFcFzzYFU|KEE7DA%TSs0km=8n
z2w9TA3!Gn5-PUjD%^JQ`O}>yT?;zt+az#^N=yyM!&;N1tV?AtqsN9Rx^8+0}B>5n{
zTzanCR5lz}n&0wAGHdV8&qw!sjt5m99Umgj7*$cu?=Iu3+wNwh+_ud!I6Ym;P$}v?
zp9$`?FPaJ^N=}=UqX$Dr?UhK9uV^w#;w=!${dB#xL6xV(ANO9_W_FDK06%Np#huTF
z24ia`5y}R`+En4_MG=6Rw#FF0-iS2|$o0=_IXj~$xwz$V@CN-00+=$$=Z!lj`niGI
zR)4Uy3aA8;Vd2KhIDgER4Ln7-|j%~IjB1K(HM1}Zm&H$fn6p0MtOgh|E8@x?F>}}
zWI>XAzi)4Yc0em@tE;Z0ma@8ls?g)W)$2sg_5`(K1X0Q(QPfb=udNz^`bpr@ltwZgu1~M8ZPKR72&7NS)bA{P{c<^fHWGKDt8MvvA!?P5IKyf;L*-{Hm+&LhuR-Vd)K0KPc+smnc3Rv1R*H+T>u6jCUc~?}ei^4fn+^Mo4aM
zMpa2(YUnA7SW#PuZ8gZ0D?PcbpJD7y$_p&6>ysviFvs!JUO5oL>MP5=B#VNXqMy=G
z^`wLtul|*%Le>3bwUZ}Ayk1f8MvoyX49fmASA0-~y527_Jj7;7&S&D%Lc%IR@L)n!%|f`n%Eb$)6?;lqhRqE=`%AkLw3eobUPZagk$R
zVXRE5p^|2%eECUOVcl1L{K+AHzg_0^baT9i6tC6lYe>_)R@$SNDZzYORhsbg0cgaP
z%;zK%P&%QGIS>S`xaX6$sI7b
z6%2n%8M@FNF32!i$uKr}x}rnNpRLijIRD59w4D8DfW(Wr+^n&ATSJ$ftl7|A1lSG+pX-e))T{^dM`)wQkF0x4rueQ#w0reU(`
zWPh;-KRGL%EtK%R{3ISu_@ys+?Kjt#qunPnq1UnP4)EJp_0SheFY?0F
zDkdp_=T<&Tml;Q28Zx!jc?3`EBArigRFBeimNz#NsMO)UljKCill5~Ewdnm8F~*fD
z8>Z~A(DaV+X_BTIPZQG68#R)Fcggi4QCeBl*sVmr{|m*{8n?e`kYHP`Ed4pu{Cut#
zv7F0ORVeSM+Z21p1>^heDt_EaZ_Gyd26j`rvcA{(2N1|5{0%3Sh2N-+ntI1Ap5I)Qp)q7*58+W@
z>?c(tF7ao{xvaFD_9XGlI&-8I@1_)2@ExkegiqML5&RrZ7_|&P-r&iclrib8MKW1=
zgzZmfS%%BTMp_{xlQQD0sJ`3`3%4UX>^}F4GEod4nwSkc_bzHbKNCig6oq0dHzoIs
zEAC(;q#XMp5znNbo_8?drkZf~IbYdOwk){%F)F7#CSGsYS|q^QRu{|hYE`=X-hsR3b^Z!7c1;KHL+xIpN#H%05XO7*{(^LqC1lmLeK|lS{>q#;t*_Lcc
zg2`i)y+CYhvP!99doWbRe`dn!%;*}*-^N2-t+EUk(xZ>^iQb`uwdE|TLN~<@iO>cY
zg$mQSFOpVV#=0pl_~=o>e<2EcAe)822OvR8F^d%xw3t%IgkU&_$kG#%uMOk9
z_SKLZoH#P1!OhfCLgwkl<=py@P^vG6fVj-R$JfxOwCd4|FM#3aSN~A;6wxPF^67(X
z?vU;}Xq+5B>AFBzDd=<~M3t=5t;=z6Vd(?@co9C1;4;nojPv57Z*9hY3cEs5MfOeR5F40_y+rLjZeegTvAth;h|Mb~N;*tzy#7VP3mX-^A^n?UV_2e
z{wak0_FJ_RK8{#q8~*t5&EgU~?dEl+7dp)_8zb`GO0c&x_mVzUj~=Y5;&~N&q{sc^
zRcTDMz}|hG;J7}dJ6&Y=IZz{k_5wB_H&8=9YE*>J!<#x(2T!;i`@gjvm;smG&l*XS
z+g|4?vf-G3RA2@EdFnbpXEl|lGMeec1_OLdPrNd%
z1u|##M#Xip5kUKZUa;e~;T+OW;?`p?fwEi;zSxQGNzBxo&4)X0k+g5B^W-cuFrl{+>j|l9=*s9Ac25kO&qy9
z9QTZCcqAu?t?Fb)|E&yQPramJXG+^c4}{VXeBl%*RCK$-SI=?Of?xRWVV?c}uVB_j
bJ7_-kr}-70%|Oa2B>;ZkP^;>$UBv$aKr>P$

diff --git a/doc/LectureNotes/_build/html/_images/chapter1_61_10.png b/doc/LectureNotes/_build/html/_images/chapter1_61_10.png
index 16b5dca787afa007f7e6e74f4863fb236c504308..2c196f8b31df549f7660e282add6c7001c4cae72 100644
GIT binary patch
literal 25394
zcmbTe1zc2J+cgeKr=XOSLAQW(h$u>@w1fzVbT@;5qNEZ6(o!Pb%`hM!jdX{!bPh1U
ze0$LQe$@B*-hcf3jc^WU&e><@y4G6PexsuFfDrF09vT`Nq3pwZYG`N|w86&&2Me6(
z9k(q8KZKp`YdfjinmM@|JD8#=8avrp**aNSJh|p#>fmT$Ys1F{{&8G;=Hz7OD8kKc
z{qF@_whrdpUt=l~z)f)N9_l!vq2XUfebCW9Bv7KEJxP|mC#B(*us-SPcIDA@;I^mW
z%NvgvH1NgP1nz091sSE`ye3n{Si>y*XvCf3$KqG#i>ZoEb_r8eC3mE^b;{rh*^2gE
z<>H!6`m`u)?P9ay>$7Ke3)Lf|qn*NhdOiqobEbFqZoIn_!62LZ=DpKJ2HA-AtN0wU
z5$SJrjASDiUUpuPk6>VZO0EsAxL(o#{>RfXlNG!(7O}rN93C!ljN==}XRcOYAl@Fs
zP2BKFHBA!Fwu?58x;t8anFPKCSu8_W2F`v
zpH!}wPP5BqJpQt3BpV<2o>Pw(Tzys0M!!#yHR@=0sXtX3`vbp4<7^;(cXn2iZl7XA
zdeU;l)}&p
zhE4BvCrc@Dd9p^ZmDZKUdaQKz6h4ieqMR&2KrKMv+Y(Cc*dq?_%~ogDEw^f$3zsCi
z!$Ce*K5jGjDJ|^#;avDq3$<;!d<+-0)%5bWG~um=iz_2VHNJ+fz=SmMz4V12wFJYcNCR@uOS?(}$nE3
zsm~+}zd&!+07o^4%pZEY4Pog+V)9h%r1|$x43q?c;s9R;dWJ1sV+P-dkWQ0u@s?bjl0(o@p7o`-d4y-{RdCg8RI{Vitd~=RXHCv*N_>BP6K%vBKH1jM
zVRE#yPS|KFEnGM^?jnF~Y0-DWFp+B#K1iUS`Tw
zDq+YL628Y0O}bgTO9Y1BLgidonoQ``TJWl0o)oE6e5-YLH1OKK51zup!ZKoXyx*@Z
zxec0z*JVxH0CMV#`b@bm(H=~?8Ya1Y$Gs|G^^k#dcG7hkb0(*t{-fn^KJ{a|@YYd}
z)x4wc!Q5zW>t(|NWA(1n-ZA8yy1iA7=3tWO1Vce{)Bhr|*xMqLR?_fR(5`7(1qL}i
zy=xOyd>65a(q$u=20lMxAMj?9rGDba7)?s?*og%EVIGW2=9@SD>)=!IhCWA(1J*Q<
zlkJneeq|ch344uP?RUrfs|Bt{FU^zOU)>BBi&zI!LE?hA$KFisZueYq2iF{G{u$D<
zHJxdA@zm7R9?TwZ6BFycEr6cMDLC!L_vo|6Y>Ge8N3*UE)K-RI@~QkD#QtnJpB6O7
zTFqvC?eUMCFLU-Dd&^SDv(uH{)Qd&hAp+K8xk@9trO#kx!v^x;OLr#?BT8=>WJ>it
zN7XhMdYRIJM0e~G^Fgou2Id1&8=Y@nJ8jJRu9MESdwue&gz9o?_Js`*rDG>YyDR-!
z#NazjM`G%fz@s`;y*PvN6lxTZz9tn61kPAH6y)C
z#>R%Ls;X+0vp_8a4^+bFjLDSerZ|z%4BF>MIam!~a%*BQsJ$CbCmvLiEgOA&!LoS*
zi*XV|mhqa>4tqn+T9C)f>udu@7X9omStM+iO08Gl+!7Z&q+mhSRZkN5H5L<}JlFBS
zUP*~48FJ=cR93_z=ly~aEMH;yk8AosZ`glt%Taehz)fl)Lc(MOwTxK{L=a-yDCFsu
z9fIjxYucdD2ptSWV?j>F#ayt;WxHKPo>F7eI$Z*@_Y!;>V%{7?hH4}lml5OZn=c1q
zQLm?#2WCmsz)i4kp_*S>a{T^=xp*nb^Zm1Uxl`xDgn2IOxPHEv`zR#Cvpq5i261j-<^zTtWVELrRexTG>crSxRH!8}JPTl;sV~t!VL%WoF
zsON&E7z4Zm3URpSClm_7X*DGW<;35Q4W98k
zGo7#E=iin+%1Zjtb$H+<$*C>b`PpGU@5@DnEXe?I%q5^O<
zmK-dgO;skuKs}P{F}X49Zmbwn@|snpg%OkF>4i7<0$f_ZEKgNgC60W^McyudMNV7%
zsjBNkf$(F|W(!Y^(NS#IsbrXC|3{5%nmzl#by9-b>n*;iRGo)bT^PMDgKt==--N}^
zVloa6JlSpvC$XWAfc2TaC3$Jd9bTlDW@MP)VZ)T8dH-uCflY-ln^J|6>C(tWF~nW0
zv=%cJF6ygR$tgXIM;dt-*tKBmuI4-Y6{_EHM@)u=yt!CYys9E>>?;i*qV4|Srlg^F
zqT6xvS06W&?~`)q;0Y3UJ{jUB`HuU{RUyPC(S&hwLdANmDmj_Bg1Bp@zsDcm><|?H#wN|(8eKxR^Q1FSzH$Q0q
z{=VLH^`cp`T;)uVf=y+yuQc()trk&|{Ee8lN|K{oZFJ}v`E2O+(bMQ}u2JCwJv5BuOJ8mW{-hZj_&1CzxQrX(hS~#QUhNP%KJ3U0kZKKHR
zyS-3}7E^&zaNqIk;{AoYNTsnUBJ=Kqw)wWmI6+&wZ_lC<5@?9T_f)`?o{-NJ@MAkG
z&FV3&k`;aJ+x#`En+I_7AC1j#lZm6t*{E%KsCS^J$W1L(-j8+%jBzK?_zwia9S>Hv
zd3P1LxQXjeuQx`f@m4zXSm_bxSAOypDI0pcxhNHv;NuYT*o956)#vyD$&X=0yJd7K
zD3PR)RGo+v`u0Kv37e*Gk~gnTDdoX5vs91c6H_+W(P9k_V)JsArji}vwe;MC!vpf8
zqe9n6A!keG*JrdF0%NS*;U(=uYr$>=4;%VSVe4e<*F;4RA)85Y}T@`;I@a}tqI;>UxW;L5+3`J`KARd;|i_KU7At;G+a5sm;8^f?eML2T4qw3~Du
zzU6he<-~to1AIX_tUJXu{PwiUO_Hyj0XYa$yC-mtgGu&oqVf0o)WQ9)j_HG8=!ljo
zh!J|zt1DYm4^)%fQ`fw{a_AFu9Vqp#CR~sa*S#}0`CZ>Zk?Bp@l>GG3wZ;=n5)pdC
zgZjENG0{c=y{6EMe9K-oOiPAIRmalKqYlr8_`6f$@hFds(09r2Q6Ae-W#=Nx5&8BK
z$NcRH?y&XkNZzkK;alchG$S0{2jf`E0m8d8D@(k>#qOhWJnWhQIoustp5azu5~(y`
z#PzLC%<7afy&;5Rc4yTe5u`5G1*tWe=%5Or
zi;-fuKs)xu&B%?(@?x;*>!u5f~_>lZo
z=Pp%9d@QUae33+leMGySCcpA}a7ztiWs2u^-9eg#m}E|2&*{m9%A+s>(mPOo1(HuH
zOIc5*TgD_1jNWggBG-Ks$IFA%bzkA|@$s>;Xd|1;9Twz@i;Mfd1whu%W~mfB1>LZP
zPi8PElBnhe1wZPKI!;pbdeX~`6dC&^So_&}=IOS?9c!*A+u@BPV+=xzcOj%}C&?)W
z$_&>^q5>%`;dMTbpSeFvFbOxN@1KKrExezOx)J40A`Q(R2<>bMjSu06ZMU^HSMJs;oSop;A4#+;5zY;8NN;+-
z^xi;5QyZ^7$1R^0cvw7E5yWTojLL0?{fY%^VYejG3q+SfL}FWao)6{XHZ;D(cSoMW
zs{kD8N7YIrpNYQP;)n4n`I(sM8~jJfZ@R&J#Jg`@c~DGwbOc+!c))pTTl#Vxi9Eex
z+#avGQCsDEa>32T3d>sVh$_Gmev0%7>8;aZQlxIp*20zv4=?VAkmfASpTL80Y6RuB
zB(Ga%m%AV-IEj=d4v%YaYHrJ6m@CSp7+4=3LP_X+c-srFV?H%)e0@tb%tog=T)Y9(
zoZJpJeVV*4AK=K{wphiiscOOQHs7WYW5;E?dE7ez$SG
zksBJ?3tv=V606(Tdayyd4AH_u!qo#FG912ll!A}^cvE27+;R+k!Z;zp;Llwj7Ga3?
zfR-+6X|%}s6|r1RjA1D6$qLK1TR_4GT$@^lE(UmOa_G)3HwjpstOyD7tXs@Uuq0!P
zJp6W}#h>s`@=1vixk^UC#{1K*eDOQd_V#EVq81_0s}96T_P$PeLxI7lpv6AFTB!1h
zbYOq6PBH)8ZrvCbi#{QxY5De{t;0`6FjKbT;*{qL7+R0@8z5I1w2Lv0r=y!Yq0
zCu0(e^sac{7Zhe=eTo7@tgJmZ6n-Hr4gk?2?k-+RMd7ITJmIp@o(lh7mtE;
zB3yn#HTSqz1Aiki$MF9^LNZq(&2|*(i@0ls0*?_vajof{;!OX6mF6Uw-M=40f_a=b
zZUHd+56g4?Ebsql`c9{2vr4=j}(<{8?8o
zGvw?0)6=iE18T>Itx4)jbde|yDO(wZQn!QLX>+5hT7N6L=(O-J?D&&yijxjN)g5zH
z)*sJ-tO@^m160ibEPZD2`A@h?lRNOojXnQ<(h(@ld42s0z6;fUmhK!#%e8}n@(XF7
z176heuWJ7t?LHwQ$^5;tH6}Xv)6cQ`Kkfck9chTo(&TEK7n|4m>)Y!SfOpEVU0Pgp
z*r;0St9G_j5?$vf=hDZy1302OPk5&D>X6gg$UDHZU3mMDfs)4v4TLZ#)~FNEBa!2!
zC{|9k`E}+xvx5A4j#)t7m>q7-u5v&Dj2Hz0a~!_~DRZI|DG2kijlR4l0i@dt5Ek3-
z4{CuxhaQbdIjOTVjxSEg{sw%~mEQY!FEu6Q+DbVmXrex|n}UKdqb24A@W}Z1_?^uU
zBF=20YlRmlYg`S@j{rG>3TV+<={z}X>-GUSI#P%i2n9UX#ok;UzO%zQ;>FImrg+<
z{^)Li>fFtHHwvvES7y5=TpjdB5LwVOW-Wej?pceP8A&5$?C*WD(}}Mt1Er*yvtDJa
zoDHJ9{aKm=tAa>p_eAn3QYgmo60=%MX0Fv*9XDNI4Sm?o0Ydc#$wylaVbF!zWrxs(2pM)kwo8>>fY-Ub$EJX1d8-QQfe-B^&=AR_)nYmYk
zTBUrQUtyz@b7bqj6UMj6XcLm5gp
z?~u9!ZFtjMT#@S8AzveYIN7Ce=?ViXg(ULSE)vsKX{?#0QM;v!R9}WYh*$dAi7@a%
zWH9Mr+6T2AWzQEFfOF1{K=x_HB-k49o4#x0W4KPx&n6j@*wo&DDMLYQv(;Ew3{Tw&
zFM#?p%iO2p%k^!K_Iol{KdYJ#73&5;LhQa}z)~Q^CFKSx3e&Bzc#<%Q+BF);$f4)-
z$QzRSGl|Aef%sp$U)9is@GhSb*wl$=-sR+^qxM+mHbgIh_*=~3Yf@;lRsS2IYh|mM
zriG*WXT_w;xJUfxNSb}@hPalC;}@e{_q^<)?sMf0*_$E%wk;;~Fom${6x
zc7?K=QLiH2+pW3UVqAXE($*gED97mR?k7X$%1uzwUtiy?2u?!<6}Oe`3}WuZ$!)n<
zVqBIX&)q=!-6#1mh&yFxvOQDvE{6Dx&^*Pxuiad;Sb-d)p3`B3GsU|mIfz=%G>o1w
zcD?P%76XK#4|VEY%;j&){x5@X(iAEu7%Y8ZHO$AL)>HUyk)W!PAs3zAJn#y|{|@Ep
zQFxq!NC@y-_twYCSXo$%R{AnDax_CxC%N>$0+xmt1#Q~e+EM`tk;B|rP$xzOrWYgF
zlM&lZjJobOw-&!$IGEssH4i`CQ8tc`#&wmvlA*Yy@LWT=5T0bZO3l&0>1t-DRwE=T
zIAJ$Eb#|?>_Gb*x7xL75nl#K?lS#pwS~uy6Y0DCcNICAY_9g#|aK)`HTKgo*t4ovN
zH$uEpt-KrV_9BO;!|VoRfwV0FxR+lyKRG$s?dkk*I~cIX
z%#$*@Nw|)SowCR3x&s9l;l!`M8Hv8Wmom<3?~`~Dns=MPv(4Rp%01h+x-4~ihIfyQ
z&do(XDuh52QfTv4-Q;7!ps;3UA3lrzbjMB+vXvyTv4*>)s!-3*vEVgnc$4fpslK
x+!v>7Mzk}Ggt0^po}i}b6A*3=zmgsrZg&YfPBq(!vY~2
z2w3x26}rFSlNqhDri16C(xC=WBqZE47_YLP;QJ
zrwH|WucP$w*9-#ug_OlE>tnYphd3M@9EPY{L+$|_BC>CY+<4?jOSG_t4nt^Yx0{EJt?kGp(4{<5{uEs3C>^+g)!Fh?s@g5>m88H@7i38>
zX|6Wab-s%`_(nqoh3Xe8ZSSB9k2h>W8jG?_c^9fKuxYZ4x&)>cdLylE+Gr3V54Y>z
zP4jUyQg33~x*$g-q_XBn*GkOo3^}koX{i?0?d_*LB=4rtUgkemwws=};5lw)Y1&|$
zA|BQL=mtTKwL)i!aFtdr873;UJSpr^jZ>i3y8{pATk*6HIsJeI4^6%@TPBz`;pw@O
zAem)^$w8JgVihgo$d@)kmT12`^O}6oxx8X
zoP9VBtaI1DHqTK4dYEW40(3NAc@ozS|^XB?ixEa9Jh6xJ=};3
zp(v(MsBlF~<*+_HQClF~yQhS7PmKL=CU%=@uU@!)H@qd=fY88$voVY?
zMfcI+EVJ8s9*gTV4Kwsw?B4Xj*FMR6r3c%|+wNO1VKDkXYx&1y<*p#}MVVtyWCBh<
zFae^;V)Oo}V@2^G^692G@|3qRW7v#m=9n^Y&F~mO_!6=X-L?q`Ob+4kdKi+c;O3wZ
zWup83Y`DJh-TMnUY8i5SKXjMaYhZ!u!;db`s?8>BJ=W;oG{MkCYXJT`Ar%
z3cS2V$htYFP5tsPNHM$~K_47}_DU=P_BPOee|1pS&K261I78MVyZ*g1>GDF9Bk!B$
z6rFyIrDWCeuv5tsbzxKhWNgp2as94^7+Qk>%)9OX9p3+BV3Y30ioj!06z$Jrdw~D{plEOXed0gg
z_&q>qmyC`SLD9qMZ+~Ll+TLeBeuw)GM4Epb@7zuM2dMs}i~$t>BYto$RQsJ}jsh_3
zk7L|_?}uX4ukZ%qFakyy;N1KH3T>TtYSWmq$h-Vc(6}vjO`UImfP)us$A;+O+Ft?4
zj`Bi!_wRLsg22BYU;eX1rhd_XyrIIM1=485VE=DH0dmCg&ll$<{1dSQB>u-^|1QS=
z!*TY#kLh|U$G_9>WUU|y;QgQcyL;-lf9u%tW~O2wj9A)0g8wt;D6g_Cv
z$;!;xzpox!50SilO!zWG{`v=W!N){3f@8d4rdhQZ!QcV6IN`Qq~`m
z14g=-UEh^8iaX}a=S1(m_XhNsNo-GJ>#(|P<{&54W6r*oT1H@$I;onyq
zn45;+-1}(f+%w~y@>r14Y4R8W2K&Gqq{r^?gu1fAaBE@bl~iFznBCa+Y{hL|1Y!$f
zLF!rO^?0m&f4aOP-CD1i)!KO(w=&^V)^*oahK`C@eeEcE-_9RLl|CHDkiCIU49fhbFYHUMWKwdR-)AZ#dGD
zLfF6pJKG;wY_>qvV(Xy@5-TlkmV7i_x>}~X5z?^JF*@(7&ozi8^wB@uezDtwQ(t8M
zv>{(Ky6%kGa|3#KHKA7kT<+n-F5bp@<@C^S55}AqLVF1S@-=nh}t6UuPwG1nHEReFG9+OE^9BxZXV3TYy&a
zVOuSx3bT3Fy6$b`m13NZ_XRrTMz_|cuW*S{v=I3-)kuZs_%lD~7dL`K{mY>58%~w)
z<}~$>y}ZE++BVPk^uX`BU{h?M-@Uo~!2YRFNCuC0@;L?phkWCopfl
zMgXtqA&qlsDYiYgJBVzs2+Il2kmF)*JHkSXTi>nhD}#(Z}yEczJG1X@Yj-MpGQhUp#@7-G}W;SiSNPib_oVL&@^B{a^qJ8Y%A
z>Xl8FCyLz0X)-?^Ko?*Gl{U~eqB@!lxk~}GuBqbiM
z0$Dv*{eAeUq>M5^g%~|Ak^6(LQb^@h98My>ieBgigTV5nYwttvBN#$aNmSuorZh+;vu6mp)Yo&05M0Q+R2Q16%
z?;c)ygCS}iNb53_8c6pMf;{uuf??;x*ed-P(1q)@p4?X{bMOly8{$CNFoh)7Zp@M5
ze(l6LL&BJ0!E2S8u=sVek*-j-=g4U{3W;N@$%Tvs#Y`UWo@|oZDuGeXO7Ab%rJirE
zCd|0#NFE?#_hfkBi;y}ovhHfJ!&eZ4nD3UB_AEafCC#aY(y0b=NxWEGtq75)Y~t3$
z(yHM|QTj%|`!Hn1XB_Kr--3S8N1q-Bd7L2IU)p3f1cx`N!!eizL9gdj95!R&#MI2u
z%v1F9x1hWzHTspQ&LyI#rW?Q4fJeDqYo&&*BY$fFnF>Gk@N|vxH
zlBuvm&n}iZ&h&GqE;++SrOjRBEc>NZ2hMu*a6o+9SrgPev;D2N`o3<>RsmV&X5a>~
z<02+2Ax86422!t*NzyQJjJ7Y_&^*aD*PYTC_wD6tLI+K
zH^Pi2XlI(WX5VF9r^9dgfp|=o@ZNjT6lU?m^1;&n^QOAf;Ee3%k1VIiB{j_cDg0Oc
z2ex>_2DN72;fwh^-pFM>$~`m=L)=jvn2tJPt!jY=!Cn7?-R4D`asgt*Lg
znNGj>%&~m`@YzSUc1L&UU9f3~rrGH%UdXHurZsCEp)8Y^Lrj&*L7Ojw-h%W9II^CacC8xpBb~
zzAcz(t!L=`a2iM!tj7k|Wg|31q9U?eBN(#Nqc)22BYxMvHFQL3xF7$*Hh*D{KNj+W
zzPEn>h6MjV0K|MkK=iqy{tsmO>+k8L-7vX7uoyW0UHx`o(q{cbWk&%rHMKYQj(r;W
zOy1Epz+uYsV^}V*6*MNfsjsN+;>ng{k3q00i{9|R7*bPOk
zbREFJbsv!9{h5ks-iQM;+VcC
z&!u%MY*1}7tXB`bc3hP+7l0?F1?6?o)FemwvQRccP#5YdDhr?a^tV4pHtBg;<^!+8ixClE_ZrVS6@FY2Ro4
zfRwtuPVU`dZK8ZNZUjTdqtXMo#iW6Mmd*H{-1VcCz53LXqyp&KhcH^v;<+%O
zk%jsBt*Og@ZusTKuF+_;gOs3
z^f!2HefK!u|B>czXBd}g{)3_$ThZg;SjVv%$94lYE(K8l7YA9yzRRwn868ooN-|
zRZI}lL&zswN6>o_7hbzvm?CG5c*r2gj#QYSbxRu3*
zg@zu9(L**TmiO0@BeFa`yaY9TEIguyIOd6XdAb49L(kSXn~)i@{FO`Hz%G`R6q9w1
z-=zae%(|YOogATsQ1QJ29Uiz$zc^V-FmC!_c&?GwHlQi3qj4+ZPVt6yeSDeuSL{U^)4v>Tpc+R#!ExELl7Oe(#2{>aaEU(kSPJzU9_zEN$!TXpP8
z8B`1I$Igv&2DQ62zaX_Bw_*v2EiW}8e!05)3(N=M`42s{;w506Vor$6`X<;HkVB{wy`y9Uko?Mow?f}!N3!frO7G3f_
z6gueyn08#wN&BA1hBf_Msg8w(;k!MhyRDs}6amqg;Okdf`F^_y3=+rT$iC?qNZKc$
zO|MSBd1R|yh;uM{IbTTf^b|*06@K4Bec(*y#@je528zh(fSbfGu*Nf?^u{B7Vd7Wv
zdys&Kck{X;hFtz&iY=(Oi=1ABMfSiHwXRr3NZ%!17_V9iAoP%8S*{E_^FH7=rW7zF?lsV4lgNn5pfb{l4mNE5cmi
zZ^=cpNb@KgwB{>T+pMg0W6-#}_%n6k##^C0i^hlSf|MdGp+VK$a`A=7qFcIdwqhq`
z7#om}ZsYp;znb+Y!oPDWvN{bPbcOYfTF|C8OPyo)G9+A_uCvL*lSaXl%Wc!sO2fXi
z6ocUgh`nwK=okS{C9YPV&Q+!v`N&JImbCg`fAr1(8o;0`S=
zDai)DV&Fn$T4>LU+ol{RXzl(nfcHx9>=uMXBQGK4$;;avQ%5YZT01>Yhq`HJ1CwWZ
zpnkUwH_h^O1mEG*55|z+Ux#!H*v+R(J&#GmTn~HIQ!n>h&)xK*On-kVUGx`DXW-Wx
zD++AfKcbJE!dav{Er`i#)QM>_U21fH98{3k40M-IJ7$nLUcUL^wjSRdBp8P~fSB(C>pf(62TIne6M$x0PoMBPKACuY
z$bY=x)u`i4wWuH8t7|L?Nj~T(i-Kw80qID#6i-{x^dt=bHWW>lmMA7cy;^CIQx0iz
z%gqi0gtD=FRpUx9u@GcplZx&YQ^P2k`zy8^q?OzGGx9PRLf=)EN4=Y*IWl1HUDi_V
zywAQ7)qW@9%Co+`GU%tEKBVLsUU&NGao~(FWM;~4;r`o4
z1UD&HB!UZp?;02$Bn^t+^$WQGH$2zNw7x37
z1O8%axo=h0AOc{B%E0%{lxI!!U;_~C=$uz;i!;jjT|O7S{4ap`$!7k?0DHd9bEyWM
z#cr52{zfkf177dfEq*4hIKBp+48{f1tAFsE3AleQ-GB3_zc1wg{r?RO+5rOjMVtPS
zF(ZsJd;I?_!e5V6Lovd?p`v*}jQ(v|_?@}^`;Go{I~11w$8pmB7cKgC4V(M@89a#r
zx5!UB@3|!d#1w~@KZZ-5#e$BX1f(VVv!kWmVM7S|TM%i0iAx^9Yhf@SYlG|qG!Sm=
zZA``kakIb3!7FFj^L%Pn4S;g@$QeC*B!1
zy+XGGmV^=jo_PUjqa?Dz0HP$~2C(XVQc1lK$)xxlxE(S<4iB0_l2{yYNbrIr5)e2^
z9Ph12OG!!fS?itKLvFm;JDp!#@?
zVzo7Ml{|32w}5007*PFdrgrGWJ$g&bd!l@_Kx}~;M!+<|41%6>pC^cOR7%Vg(Rlgz
zvNf_vp&&Oyqyrte*V@25f(ns9*s;o~+=8kZ#+a589FXWD*BVAUtaqk-TzVb#?jR#U
z&s?YJ6`?yQn#I5;5)gdR^++H%KrRj|__>A!dB9&iD^>T5U+zZH8PUkdF%skjFv!M-
zvOW6Dz?mtK4oVNSXOBcv5E?vbk7A{9pMQsjk`d7_VxfUJtm)(Z;7^Zh1TZl%IUZL>
zfy|b7;3UwXM|}|M=%^M3#RLf!mizY(%VL=4Xh75urOyhw40z9Yo9Z{*&=XP|wR}(^
z)V?5vhRRRSMYKcosnN$VvP}REabUFe-bbk{m>0y8mbW`V;^t_c+d`rV9ru!?78F@4B
zVOz}=mz3{BKAx?XPxY$WgWn{GAEQ9KY3OU-BF%7NeDGzwr*!BK#(o291CVrp4n^Xw
zS2un|7zG~lEK6yZ)7Vs6HLWgD+=)r*ExMi1N9S{l$+e7M5JFIAVNt{ghsbY;Ua3PO
z-S#Xf&p1kLbUg5L3lHvFs0~O=pR#l&5?AcMymujg4SRez(7-&Sj5?L2v;Odh=(4kT
z2sKtTJFEfW?`%{3@o+q5Lv}-vOLm$8-m}3)9XTQYlxPSn^#-KzoTaXke&Q~;1{!Et
zF1|6D7`BwuJInnr!S|3U4awh8HqDcwQhxC(dJf2n`FiPZ2;wam^Hz7Dom}8|TU&cP
z6T>Kh`|FM+Om=5fmqPFU2BPzL5KrJTu$b)jXrsZ;*V=O4kL}-j^ok5|$#}Zndsy!P
z1ij}oRIaN9q6-G-ZSy?*(xK0|@-g?d%f*Te#7m{&c5C5zO!O(jJ0jxjCrw8K&m4c@Q
z&2!WS^{JAg{2mud#V~#h>y)g`yI7tjClUgvO41XPL_Eq_h17GHPJMRT!|mH)<5w?!
zJxgZ0?AGI;V||v;#vi*{)0mE1O27E#m5BCwp`G3t^F0iqB|Lx)UJduZ&lpL*p!L|9
z3?vLupc3%_34t=00U`vII74`LtF=oiQMb@;`@{Gyh16g>E}Nz{WKz}^W0!bHz=zjM
z(6MX%-UVLyZz~x8wAC?kp<-)XpVE%rNeo}64kywe&=Q**QWGq#nv9>JO6l$6&X(XT
ztnb2=Fr0BXraOUG!tIkp$R}awhzJmXyT5l$O4gv|8BFpmhOB|k~%lZc(4+HBt|6}P(Y%>@wE`tH1i-KvV4(kVpR>7Y^HbbGByT4GYXYWxvvM$R?KQooUYY9y6&7U>el03?VZq)_d5M`;53znJSc7Dnx_
z?4-Xj;JMB8cYyadCj3j`{u{ykMgPu|%KTRmSGZ05_Z+FeOOgGs_HQ}(H+JiMky!Xk
zT2BBz)qhi=K+DJqW3&HX9`SptEHJ?TRg+Q6{}S-!{(r%JeTI3=(~j%U6k_q0(5T?&`@j$Fg_F2ZYJ3!`UnxX&YI)J>%xd
zoR8E;7sUAJfSEF%Yhr@>xrQO6Zg|SilivT?mCqV}dieqCKQf?o_aDVBNfmn+7Y*Oh
zdHb+ou%bzc(9Z`Qgn<3T`xjJChE~gHE+SvU>!f^pNjuCCA35=g&|y
zPQq?GkS!j*VWgwaMOg3ZL?_^SS5M(Y2BVRv{dYFoAVqgR&MWnUwJU%
zVvA7_5`bEz*aD=nVRXgXk`ixM>|wcbL>-n~O;)Q^>rn6K
z=IW04+qgl%NClif!-+%Q+#HH{jwdB0eog{eGIpNfGAOyulD677U%@${7n@#WBkK5c|9&Gq#dZt%Q-dBy4uzgo_#QrKPH5E}dgc5YA+fkkelf3c`?%eq?AG>D
zZ&BCrs?{Tq4?&<*heISvmqdpjlN@taPX5R^8Jl{Xxy|5W#-wx1oj{WzLmxZ7XNA^V
zzsR&(fmtP{v#g!l1zOldl01iy8zc&37WQ^3ZtEx!oc28@F
zD^giPw!&s|c_@$i=7X>a4*cA|RX9ZHR6iC-|O8{)BO3{<%!H$xWk~cQ5N|67H
z>^%{p7NY@eyKCCn?j2JVp63j
zOD6i+e^l{1`=2v*?}Zb}G4#RJWzrY&>OJ;?Gp+;~t3DK)zQL
z%^&<3UG0X*JDdrTD`q`}Hj;cqS5L8@FR47=6S3^NH(^dc;*@;-sQze{xzgN?XX)tz
z8>KM0ZUxTA-P9(d?*m44n87pt%KlZ~S?v_Q~Zcgx@q
zDC^}>%FEoH+eki&?q$pev{J(VPaFi=O4=wBr|6p83o^`?yzx`&GJRhdblWh`OSE&D;PVS8Q9ABh20(TQd#>
z?;i35IPFft;*}UKEIGXz;*>8ep;j3%jj-pB=@sIp&1
zLL$N0C*$go%uF!+eo0V?f^}|5#-OHXd+YzP@7KxPg4(&
zaD^A5)JJTgr=K+Q#$+lie`ZL1>Hk9)Kh?5^f)H$FC=X;IWGcqXfQ?SjpAF?wfjuFx
z0r5H8B{-c9=-_*4s#p1+-2r^C1G%
z%3|9XK=j`9=yofk;!KryHG?gSN433j9y(iti=mVVFmxE*kD2A@FMgeocZvUxEo
zZv<3gmkdZPi6B>}Mx_-2;k&N87_jqOz0$}{#yPoW#d{Y6A7iM(M|&*bqraRYTs+yGn)C@PWBdO9Ld1@iyZPQqy-$%vH9C!5t{Rtd
z@kOEw#|l$a<#3<(D8^qA%l|<1s7yQ5!~Pkh|1qGP7oqzv9r|yQykI`*m*GVRG(6ZW
z;yj}-3c|+xe+f6vQxdBxaWy}q0=s{VyN>^hn*CqGzwXt4iT0oOyN8O9{>z>a^R)`n
zf5_y!yo9VJ4afBTnN2f+*SYDh-VX55lFG=#gBoyPUEJ_@mB2fjKE1L$aNVsFI~icz
zyHe;;?AXJ5e&Vi*wn%|Yp3|uj5wK^&xxno|Z1jsF69!skemyPuDbco2!X(ug{bOEl
zg`vR4{Ft@F?hk~+Q$&!(yUloZh{0Yv#5EmyUV_Y8vfPSRF1sdEmDH0D>)r^6oszAa
zr`7hRzecDRdJ+}!I~*%53Y!hW+n)=dQ*gf2X%+42D<7%s`WFg4)AlL^<1kC`-1cFn
zxw=W$NM8<#v#FZ5?oVSuPT7)&>2zeou;NBZ!h|90Kh}VUhig}h$?Yg9qcfT6=DUZz
zAICNMMR=%v>CNYFS$}c(e(UwU)MhF&i%R7K=#(bg;^~#8X6#+v2n1&!B!(r>=s6@KJLvVd^&X%G+;?6Dezrj?<=BE+x;&QS5UOw5SiUjwv;5d
z&_xfpyr*MiyJ6f++^U1{OaNZO?b54(PM^OXO34dUZUnn>WthVAgb+#3Y25C*GZ7f*
zrh_$N&!M>EfzT%Gd4h*L8IxpT)WrQ;nq0wrKD}yVk9FUaY-Y>U$g|yYnNic^K`71KX3tW=eS1wV`K`5v
zM+r{u0(BdC4R2zi=i@?Cn#1i+qsp5wr{ZMxS2Ht(>}r
z->@#}2xTsLGAh?ScB+;ya!l?q=_)WjF#h4W@T0237mig2BjZB01?+CFYuJ00OSw0!
zbF>*zOwvt|-dqQkKA?_V-~r^kP5qHs5@5-ltm{>#PFWsOxO}3YkSM!Y%Wrx*muOYT
zYTLA(cb)I@XU$gvoW_kgZxb>5lL7%09Eumm{`^Sx=h###SNe;lrO%}Z20OFBJlHj*
z%w!4;WLdiBL?PEt4)IHcZ@6_DV`JQJ9hPkP76|)_@LOsCsbPmbw_DFv>HX%HHsbq=
z@zu&bZ@QT>fV{;VP0Z7Y-ra-AaHATJ-V@&$mkxiR+gn)wU^+dtkQiKjT4+*ojdsie
z$q@$qyn{6y+G@PeH&ZR?&VG&0vdeojOt-r*7&-qf39_RP)ql@n06)HC8>-)}cQdA-
zWUXe!ea%_nl8$CqTZ&Y_RlU1txEn}MfoTp*NIY~eALY78T6m-4ut0dPiTPG}{$f6Y0|6oqM)FFQ~~KKQlv}oF<6k6
zASg&Lf=I_8kWlY0tM|O_o_pSN|AaY{Oy+s!naQ_&BAg=RYO^=Lt%wRz-V9N`O^c`x
zJC6FM^-yu)QoWgr`~z{hj4Eu!g_z;k2eF#u7L#RBkH62)oe|=pQLHU``f~DtGd`k!ktddZF8du6lpAWyO|eAJ@Sc-U0Q6J;H(IhNr&U`iA%B
zd@64J;ZkEubX^O*Hk&6BI-HVuAb<1j^*jtjzmr&9e0@)IMXnLBW@HM(#U{%ArR+!f
z8&^0vx8o(KF5ClCBtC+AVX@wEb!~WLYR;3bLOQG!4?5||m$Ni2*0?Q}nn8|I_Ct%Y
z9Z3_BYklMW*1`K??~tFKH7Q%4UJDU^vV**8>|IuexfR>&E0VQPdv}1_9NS^zuH7t9
zAdG(85J_Em!wX#`{2`hyyQL!G!G#o$4y|^y*o=6je(u*aA+;IjXQD<&^KX1Q)yP$o
zSKznsg2l;9LZf=}U
zn;toisXb@AdnR!Yi#{e;P}7|jj4U}kS*A)~Q^;Y7`|7CT0yTNNbblxfX12JRq$vAx
zOCWYQ5W#z%z3a=$3*;`G!|4ZBmbG&oOx!*=GIr_P%MjGmOL=E3&||n2
zvK?cZ#pYIu>YWcB+iMW4K@o;;)Vx4Z-ZnfRow;fMUW|E5uk^38K_;$(uQ`tEg(u{C
zsb9IX4u+|;Wh<4mv^0qZm!lRd2;o2yMiCfy&<8m6h;XUp%%1C1f>ZOs(Uh0{Uup+A
z#saxpwz#oPNB1fz&wSoM$;zF9!M5xtf;r?TmeGIMnDeOU!l(q>S;h4_i<>12Z+oma
z6H)_^8ErmRE9V^dm-}KOuv}p$7p4|0-}GqJ`gyufKgHm_(Ys+;B#CpkwvmUzYR4kl
zR>B1vF&}*TFm-x?(G*W1r}w&^$T*C8x|vMQSm@
zmj>qTM3X0G)bPZ;FKl+>#xA=_QRunTk
zA|zWMa>$a1lY@hU#fC3KySfrG?GRpUP2bH|_j5ZfX2Ttq)6sN=NNwGhtqrj+^c}mK
z()Z2Ev`cQyvWXc|ReM;p>-p|8%g?V&-`LE2`XOfZb*uo&LSRA1a=_2eLGr`yP}Jn-
zoKPD;8f6EP$Bwd$j6GH1tda=aG1_7%_Tm~>R5-7gdYLK_bK-KD!vLdTdD|DvsvAFr
zP_>y1t)pkvhRQZ>I)g`rF8n1@R9!bGm}Z@W`b&@!Ph5BBg-M*BRL!7w_jXSiYDQl`sKH+^AcwLt
z6@U5eTlBljZzgXui#WPRE2HfD+1|1DWua`2=4uk%R9V$h%J{*<
z8ycqt1^%@}X@w`w{qw=4gn$O)l!egNZ`qR=J{^?)`RgU@mhC+B!meCRbRMii1$~m=
zaFGJ^m&O_*FH(rt6)Bq-80mjJ7Evmf*?MPbgPIos3Mxyb-)f@L`z|bh<6o=$jzeP|
zQ}rRm!J1z#_}E}@QAiz!5O}OJm8?5Hyks1E^U|ZMlx=IhVPzC?Ntx=o^GRf+X+wr&
zVv4D=@0uU4pGZ-4GRx-(>Dp($?L6<&-G#HgnxlP-8
zoz-@9_BE6fWK@!q18X1xp5Z(7fRRn|cKE^LBpRzr^Ey&ng=pR2vyv4f=ya{D(qge3
zparzx>V-Y04yErFV=~{DqLmva_^WX;N@y|)O3aGid7vN0lztO3uZNCuk|B1e5W5wA
z8LE$q5%-Mj@Mr37ol;i3spU?sTjC&j_bjn{VuL3nueL7q2iN$J0uUe`6a34^lz3>`
z0%yZ9af7myp$bN1Uj$#12e7)Lqi!q(^}nDr4fb?fBol0+QIa*Rt)r0l3VAFeaaoF7
zWlQvw;KG2qW8XmE&X359ARSq3OFqXa?e&>5qWjl96ge3>Fja@2;*&#(5@$*uZ=x!q
z&2Yy=_x{d?nF_J!2(bt~TbmQXF!TCcjZT%p
zf3`jMbWw+Bi*xM-WPUL&*i%XspNUa^?WiG@pEmUH4U+k
z8~btmh-m{a$DKZ%+lj+Tnmmc?C!)V7Ii58QQIkZMntDqzId7t~vmkrO@FAMo6RwGa~mKtA2U+=AxVVFyGU;R-xL3J-pHuQ@z8oRZ0)Am`$55
zl8-j5(CUaK&U6Nv^R#3%CDBV)P84|7-WVROG^~)v7996fn0Y6Bt@Q9rx)1U@Rz`%E
z!f56sr@O6c&FHzPi^`Z`T=7J@4;~-}8r`Ry16)*RoD*=V9VYP;N!uyXuC~#>%9s=R
z*Q{fY%@FVp4Vrvs^&DKX&s&!a74FU?y&pW`*ML$m{?Vr$vUT#^8_V%Desy4=RTyQEI=SI!KF~I3+%GeU9SKhl-!(_VJ&J`ifs`E8p%b
zQ+jamB&Kkt$hTmVuEl@T1gN#7$5q%E2OC0B$J_@Lm@0g+E9$k^gNyK{f%^)Jbv~5O
z*so?S?XfkMOILD62Fe6
zK>sMeipB7)z#_#Jy}4CahT;nTXo{jLVU0TmNJB_0HuV&Mrc>%=CneB&;2{ax^pmXl
zAJh1E3B`2eMC-NR6aPtoB%5PQBL9IB#&cQqOmhN5kNZ>~`UNO2hmkeT~VQbV+Q
z?&Aw)s2?GKOm3wvCng4&IhoSt4Py{x4SKcI8$rE|^%?VvAkl>%T$BgOw;ynbNt&-l
zgS|b8o{I+wa}5B+db+xQ0^;cASd#*jl0@h7a=W35#jfNn8o?6`ilRWtkq1=JqvC^J
z6`zb}=Qe==WbOLN@Z)SUwv3=2KLO-1iAN1yPt)`CyzV&qe)JDGAMUTnX?07CWrdN&
z?|+eV^?JKC(m&*=dA>d<2N^vO9n~8ljzTK@z-Maj=$Jz66QSqWb$ZS=w;O_bZ6xy#
zKtn%4z*oUGO8{uu4v4ah+{%wZI!8ZGPw*xkUHs*{o>+wl062j)s*~iE*gFp#;vA5^
zKbD%M7Z;1tFmY=__E4k2Xh_r94|V>p3ec<}wOGI9{fos*UvsJ%Qkp1`LZzU|**CwB
z12zjuqX%+EvPD#{gosGg(+d&Ofdn_w_&}FRqe0!?6YN+|fR0Q1ERQ6(XbjB&5DNnR
z9zZOv0KOqD*1)chG)G@KT>4YnPA6BZx?iFBSH`~Nt
zeokflECCO$k=xf)I}ZRJzXVPC5lpL6phQDvFcaEu(f}2PxoAu(h=8-^25*||rLS=j
z28tc(2rMEg3>$^CTPYBd
zd<(Hj|8S#PGVmJiaiZK)9Lx<
zu$t3VK30u~)ct6$p7XR);1ve%@A@E%4KV=K?f_yAvbS50gH#nEA=@Df@PQPNgbT4d
z>s-2HS#vU!7J=FQ7$9FT*Wn5i5#tV6lrs0969A=Or_8+hCg9(J<9-V^IVX!6wL-Ge
zsM>)AxcD|ep*QJj=OTmK_e7@8*a-lKL8lEu2i@=m)4aF4*VdtvIDfbT2HhP7+T$I&
zU0ilH?txI&dq8O$k&X;d{L#R&aR(=&%)B3Z>Clb%c*MzIi{yYpqBRpm2uZMjzUJKom*lhh@o&Nf
znL!|*L;alz|K~3o{MU6&!-y1XUMzo2BFt+voKC1Rs(6*~sDi}3=kor1r#(>!mN2M6
zTHg^kZ9_tQ11H9$i^nhd`88rKoj|S=*gj1_5g?@0@Y}zJ+efYw2riJFjY5U&CfRf*
z!`cSBTfy!EP~bTYfK!Ie5l<$Fw0?y(N@7kBXl!k8Q+oDkh~GIJ)@zmyuu-r8-);p5rrCe6v~k%1
uT$kgA)O|VPK;SRUgn~2@0uCwN
zox{xh*P!?PJmUGC^S*e^v7!=drM`h~@9gs$pTB(FQM*
zb9mr4y>Po?@SljYoVK&Notd+nk)tV=vXQgBjh(ZNrSVl)Q%5IDI|v^a_{VwG!r9s0
zNtB!0_TL6vc8=!UU*js0!9fV@?>~0J!Xmtkd0}HECR1Wz86U{YN^7_$ua0>nhtHy0
zH-{dZwTQOdGReE}jgkDO8Zl#wWXyeejVCr2U*BmTR@+m!R;-?o&Y6B8`+-3z`|B~H
zPeomhu@ek#W}Q8iUA+6rd>x2I-pC-QG`{(I$@%&opZx9md<1>c(DE)L<|noYzZx6!
z<}h4T8}t7ab$mrxS=rCVONJrA!NJ-}betE#PiVG3Gx)19ulZY=t7@-^86=Z$>Xc&@
zdhJ8lHFNK$-wD9duCzD$f##mg&Pp|Ic!sT!qq+JnGJ(hFjFz4Yxh=fB%eAIwaioNG
z!e^gu`Fq}b`>Cc_iGtw!`sV#Eg|d!8cf+Z9jhb+*20xK6WIk>C_Wqt_eDTVO*0t*UATPf8`Ese%;7ecRk;BR`Z>rat$d3v;)%SP97Y4E~
zw?;DHhjov1r-&Wy!R;1%)6XIi`&pnH{C4BXcJ+sN&z_&fzvXk}j0L{%%Ln+cR>XRhj$T3p;+8oVwlnn)^j{AwGPV>?lw
z;=20dTd}F)yZ7%$YCTF8x2OfI?0mVL%{7|87|A;ofmqaf6vp2dh@1fW2)J5
zD4(t?Nr;u5{UzuSw*AqLrM~APbE2TF!uDL3$8uhUu+tn-#kePj*WQY}*BX~zb)i`o
zHHYBq;IDGvLc9s36L){btQgDVx}q(KSmA?IFS13a56%=a9=D+g(EPDk^51KffIB^s9OwEoO0Q
z7d`_6W&VRJuH(x)xloI-Y5_1RKMyviZH5c6pA8IXRJ*OK?5~X-f=_rLcZaBWjd9r6
z*b=z*-NiEt8rNay?j*bVw!RF-!^Nz$;K)exsipu!E@_^x+~9?^J^l&$UHq7-5>Yj#u_q!N`T%a
zi#P@DE)Vr*D%0G#a|g>}qF$oJyr+KBZ75$av~_xFt!BMt(YdeI!^x~GQSRwerW)7P
zCP+p9Tj{e{kjg2%*qb^ns#3`4#VtiPgL-jt4o#fNxg@(x*^o<-Mq=Q>9SFa}4v?Den<1r(nh`h`TMvu=e
z;WUe*ReWHoW;ieQd4p>X?hnPGi&RvYt1K@s-#F`Nn&`VaQW6CEgBkseFBArtSE0l7
z-KFn&9l4K7K0Ya97fcFQP7=i4=vS5cR^>DwikULe?2lf8>wl)ub9X7U+zwnJew*P}
zVD75rXx{vu`#5TkdLTzD5*&1L)BV1@e7aVFzNNW^>&ozJUXzvwHLkYpu{W(3tyvh~
zP%&#WddPyQNyTSI1g?o7Nszw|q$1rzNh8QYxXj1tLoQyw%VX9_>F4Jc&u@9Dv9Xat57o|5u<^sX
z;3743dy3n*yt})5=ISzLaXH$3|0qXOn3e={zVm6&h}P-!dy({?KY#88lWr6u1a5d$
z#bIheo39)NwReI@=s11qWM#ST##vr?%}*l)nE|(0sbxPNPkul7rAv3vDCF7E;*~=b
zVx3CdohQxrXc4zL8e9`;uprAj?JPXRBcc7eIo;||Gd2K*Nz1?|=I57C{A$ZMTd?|i
z?~l2K(Mu%4AvHI9yDq*ybLV6oZ`;PqWU=*H87nLHNGX(Xr1;kT+wk`yn6>{+7M+A=
zw8M1EW*ckT1zK7q(8)^V9$Xpit*!EFTW}2j{8{$W<;&dK^wiV}78Vwnux&8uom%PK
zE?l{ScN<*3$5oEZU?9J?WhS<7)8)R9AwN6z;#rY;A?vc}d3PkP55lW=CD4zZd^-nC
zfip!;a^&KZdVT1?R@~9o!D?SbgjS{Rbkw=0rIsJ5*KR$b-o*(KmHs!8(yS77-cA%E;KXlzAwt
zR1K}$xCd*Jklh?u1d}DaVtb^}$zE$!tv|IIg%k?G_Q#~kp;ON3bF?#6
z)JlI{Kp+O6!TS}BT`lih+xG3D#y+NfKgFm~LRVzsjK00)n;JB06)Qu&=OX@NrVqzy
zzB*beeHYTf_8I(zU-kwV+RSK5gbG3I$#0@hHrDJ8iJF3|>VCg=RWNhG=GvZ8!7Y5@
z!buNgIWuuE8?n+N=bZ_*#!H~C&q$2(dF_nQCN7<6U**YUNLPQEU_FW|0E
z*2$4R8=tiO9E}<%Io-&m(EC&)9qibiQ%8FIqW)BW!*BK~e|7q&*>9v^8TQ$U#S(fSOxLR#r+8P32hTjE5y*GoD}>%^Y&OK5Y0Z=1%ac+}vWMQwOiHO0?nOLGE_iw%KKY!2D+x3A!5j7hps
z|GaF&spA~mm+{sT-t;0)9!f`Vo2F30vUZzap!Ik&D_n>L&ta*QIEYSvDXFPUAS_YL
zh47jPZZewg;XCS@#eqihhVM_-Bc*I>MR?=YvuuY~_3=#f7P~s
z{L$WH#-_U5(p2s_m?@|@&awDg76h}rk5v1op~rA!a?>9ftVepy~Y8^g_9
z3UPk0y^$OGqaDeX^4Jj5Zg~YidZY~Hs8qm_o4YpbTlZwVq2v)yLW%iQ!!?>5lc+vX
z?L%d?OFK&j!Ep#yv=+=io%b5&5l={hlfSUzKzA*bJdTLbfKH<~?feHlXVl@_n_m&I
z5eF$qLeyFORqumFs_x|H=G7EaiA(hPHJV!Cm!w<^UDwCE!0sdLICIXmCIxJOYxM{S
z09msIEKlJDH7f&o(=3dm2botqDnh@R$lO;Bh)Ki$HkBS-%AaE}!K9mgnA5tx_iTtT
zOJxL$+q~z;xkK)BMPW0OW)^(A2!)@ox#Z6L=uY07@@8djt{p{B2o2InjW`TXjYwW&
zV);ZSqj2K({)C@|0A=<6UelUK323_U1oKf)N1|R$~8X$f6$={{kuN<93w=ZPl
z+*fVNu7dyKN{|;#)q!_e)N;NufqhP17EXtoZwU2VcZ%Xfy$|-sM>$t85CqE~pVX`+
zj0$Dp$DE(DXa*-GiJ;|mEuH1?zpq@jY4~@TqWS1Hl49<)5uPSS)tIVLyEbZ=ik{y?mX>V4TLXM
zCh&+RxUn*Au3o_pjh!jDVLst2mZ8|VY~ts>vUs6-pRffvajtif@4>v8(0hGlqwh6s
zv?5j}?Fi{P)%eyGxZQhaHtyzDoc_bb3!Zy~@;T@u>#^L!zP5ytSBdA-3S1mlWKqXF
z?7=6qL2_J0R`!UOn{iaeTT*FbcI#kPZ(}Nm;SvpvBEWupoa=oVgx|h>b6)*1$k6q*
zJju#y;Yj8gsth{>@!tosUZf>zz8A1DzGLXQ|5a+@sVmyj
zq$NnY6WKt?sJl}XB`tF&Uf?Ey4(eFE>h{X5dFZ0tk#w1@qL88Y4?O(l=2czJqt9*4
zq;qLF&9i~Z6F)rm^PWN~yE~4;T|G&Oj7I#J7?-_*bJHZ5`et8i=+jbn5=xvIHVmyA
z-Po-l8TaK#pvmV8Jcl6SlpXae5#K8et;ia2FZR0jl48HF@_CXJsxytD7}o(UB(`Sm
zYNqF#=;bCES@Dc^=H^TsNo@b(=h^CwuGr{k*#{5SP?=)3cX?=Fg2ih(5LYKCT;oYO
zkCli+Q%fo7RerMe=sm|CJ;=)|0=@;7CKL?b@%!eLFsbV|@=!kbEBr5=GnB2MZ}UP!
zz$#%BsmMK2l`m-sv2+)JIkMFKKyw^ex-0cE4uODd*`H
z7)6X8MB6cH)c17|4mRO`NCAO>Ug_B0)O~9#mSzcUbeXGGX
zT=-4;&~e;*r0R@&wKcn{n=-?Bd2vgnx%$n(F#KMr{ozALfB%QgYWH0l?ZJMyzwqgWi$00tJB=E4__U&d5~TkXNxZ3^C^$Yo{l1LY^s%0N#yTD
zAW(;J^Z~)bhWk|M5mdMUHJd|gbJWsE)2uF>gtJW{=E1~}EdBV^K023ve?5PFu7q5A
z0M_3%Qc}1Vdi7|xEiDQufZyRJSTM94C(j3fpebN>pIcj}cRkqvvH-*M-e#nD#B&_r
zqnlPn-@R*@f>n+E!S}V$jP9(aDBTa{B{lk}J^>uHupD<5zIh!}Xf;1v9h&%DapB+d)E0-wrOYFOw1gMpc9XY}^&H6Ze4YH!
z`3jd=CYVd1<4c&Bc9r~9QSGV_%sXZ22^4D0P=0M{okSBdvrh6o)+Yw@k%G}>@R1MR
zqI9Ylz-GU7pdkxDQk$h{Xs_F!sOW!zP+l>u2MPu8{{_t0BDW~h7RfK40Io|I@q;&`
zga6=9^v0GVw;)z$3_u$YOinOdrg)UU1Nh
zp}(+^`jO6y`hE@;4F2N&tkyb^UJSK
z{=$rwB(du|FMsJbzhU5iAv_PsMs_no66ecIx5eGhjNuSm#m7a^l);yYoiCrs48GX|
zb)FSDGU@rZTRy}WK!4OBdiWiuzR2VJL2_$H`fq&sk7NW`BRCs{s8OG+KhGXDTym;k
zTMLmAh+j}O_{%RK$=21lb?OvT#XaeAhljOWBIpzs`!cXCdf#GWzL^7h#Ll63L)CR_*e*$QXC^N7;22mbv1gJQ56qk9RcL
zhN~JIGvbrbR!*|=@_w6lT_3HA0v&u(0a1qeA`bu?mseH>Jo+ZT?GLhN815Ovp>?JZ
zGA8<@%nJ85_`^t{(b??m?6}>;U(#At3t-8W7}4S&LBKk;r-z)LUZp2hA_Vl&`*5pv
z?}zmZT-wwU~MlcX5ioFjV8-)8efS8AU_13L7U`VlIxeZjng&;V8zB+doUJFzZgegw<
zNb@7i)V)BOdRs`E$#B#I(VKFUBx)+z@14aV?6|uAwUp34yjs5+Qf1mx=CZ75Y;1gA
zLE)X_Yw{Z7+C_C$`KkvmY
zZTG_$es}($ga5;%r!SZ@JXH9zRmFd=O}QOAU$g}ABAtsZJ*)y}Ls+Lm##QuvzIE*i
zY|k^8Akk%B;yhCl_y#^E%@BVMjz$jd${XZYhFzbm=2fKRKAyi?T3Mu5P~F$DkiG~<
z&&a#KbK@zfk8gK5bcUbxK8#7ZcHRHL%ER$Tam1LPyh`9(bi|E^%A?L}lOzuaodQSp
z(j;3O&2wlgM|*2Sg0dLgUQeLtH!H!h_13|$9T1;5#KUM4+lND>yeRLrH2tJwhGa8yi1JT2a5h?{Jbp(sl!eLO8UP&+-2GGP}yc?b1yeF%woX
z5oabS!;3w-LcKy{PinwfeB0b`MU`d4*?Sal6b-DX#4N?XYCXfSLS~=BKfAiS+lN?!
z!XvZRa$c3h7%xi=5g6vOyBsr~F6{idMZM0?V>g2gD#YE2>G}3yoNH`>t9^6Xp>>kt
zo<(r)J@cr^SxQ!JXE#0jiQ`DRTGX4+jXrFPcn00vy__8P#qF@9NyF(q7lB$@$wZha
zH5=Ood|gmdci{u_a^nuorO#9E7x#P6!;OV;YTK*j_$CP0rOVEZd+1FqF}d?H>(8ad
z2bklFqEeV#a^mn$4kmZd>@Jdr->!byS4vi;bVJ3?rUy
z(MM9ze7?S_JMb|Uc4;(p6@djsnVNj}W*rkXa5wP%*|=7bcFc&@_{rKl;k;qUK{}YD
z?~dhxAf_lQ=rO>L-PX(9`!VEIwMT98_@}(PU2F==>%-+F-6jW`it(T6^XgnzloZpy
zyU-ui(}wNC;*8sAvJXAsXFu0ceMaVuQrMvR%-^nc(YFvOn(o!G(S}s;F4i5BnW21fBBA-7SNAf*gCP)do^)jHym6pT;E)I2}FJ|N@TbDYg=Ot2A
z6&ttr?QmJ(llFMWhi};c4~KT)MV7faO}@H0K}wKXa^w5+=BW(ZZ9*yW>&sr0C`a}$
zFR2MshVJQlSMfrbh0Wj-N%F8CsXg`8nw;yVO&_C6j~Hy%e@>#|=y~YO2d8@!5yxeV
z%KJzoJDhJ*z7&JE4x3UYD(eJc$QcUX!$gc)UIG+rurY3+VDDddnVEoH1!d7<>fsgW
z?~)WoOizXbU|R2G6u65y3S#>1hmB?uG3aUuuNyB@YN^8#{|@bi&mX=WI=!noXyK;IJ=~t}
zFEY6oNJ6KOBIMWGuApl%F4h*2@IhmM&lF^jmBR
zvOH#?>6digLcJjKOXe{zib7D&B2`>9E1)xa^y5j3GtXlov?SW)ebPB*9?aW3BMlojy!$ida3ZCF=tP~
z_yG-z2NNCXkm~_+UUs_Ej9ac_rV}jM|wL+F{ftIKldod_ePHm{-fSu_OgtUbElQWztJo`0LG_LFaYGIWJ(Mm
z{|93KscH75wQr`JpzePGa&_;=({d#(=f9HZZ-h=5ex&jzjK`qJQwIX5_#dtEH!A-7
zKtR9whdi8sy?+&4e%XZ@@rlOsMYy=t1dNQRtRAt~x;J~=PUK>9Ht=n8za7)x$Nu|F
zFuU8I3?bU^@D#CN);=Y&0t9fX*IVq3TM55U?H^nA7yMvD7MY(OnZLOUX43p~j{o5c
z|L4evobFjWAPj_C^nad_xCKdU!9U0PrPdFzD8)AbhqVvF%EHo@ry~T=r2EY4djK$E
zU1eqtKp_#A$;suMo%w6r*6{$cyydxLktkq&7f|Mra$7j_Q1&m}z2#}^izw=Ai(;kX
zx46J(*7>#2`}kmUAWtU_IC8!DdSU?NVZ}>$2?h{Sm;t;?M9m)zY~C}4R)g#qS#oQ(
zBjJ|m`Lp;WpJyEt@^yaoPXHJiBKDIFxKbe@A(B!w&!4zQ4116b-@SCUBuJX{%HBF>_>;1EpncwPGo*%JC=A^ic83}YDXuboma^^1Um2k1!!BRD4?|m=
zc@Y;_N$OZXrrD-~WMjL)cd`DE{^gZGJhwZMT!uX|2D8Q6C0IoJK*NB4B+b
zid7Y`iAZ{hg^`jh)t*6Mr9C5R_QwJ~&5NL*v%pUL%|pHx(q0u8PriOH0x%Rn&p+dK
zbba)X3Av1J=y-3qM)pycprxMBP9-4Shfi`Xh+(QL-F};xDlScg5JB3QDSWQozrbE|
z8GqOdYuuf6T0pHHfgKCQPE+!KWw>y7zVndD&Y7n~2hma2kA{Lx<;Uf_#O<4AIxIE2
zswroKBAWV0!?-DATAPlw-DiBEk6uZP_^kIJKD$R`@HI~JmZRIb)8$O0$X_Wx&PLl@
z693k-U?V$@Z7p;MJyq!KdSs&Mjol#`L{qTsMG)Jz<|WN#@PCjg&VnJ|*LHrcHFS0?
zH2cPpsx?v_*4>!`*57%&_G|#ri8DF`1~78)1t+`|7xK}vHXDl`ayBi$?O;w`{<#6_
z85QAMTdf6w;R8eL;X^_)HliM)XWt`fCT8BeLF%`db|@eXX)W-Hg0RQd94v;>E&u#vA0d^dsb8mXGpWBYVXO-nq71$VWQBjK7|^(wK^KjynR)8-m9^F
z)b#39o>uq|wET;xtUmSp+N&J9c0bN&Y4VF|VTCVcvF>Bl>^JS;Zs?jr%S6q!JvhIT
z*6O2?F9HMAz|@bX@^<$XqbVhHfUwqoh-64Rue};CwhT`BF=Tf50~)x8%`Q+NYK`N9mKSGYQT~v
ztLFoEs1!-X1d&fo$b?6M_3iOz!SopmP&8fTbl{r-DmbT3Yt*-OSyG<&ZIj`}fab
zmI6R#jYf)1foYC9TISwc8&k}CQl>OC1yJkJ?s9^lEj7j+7_aqEisyMw&Jp%}snfXP
z_O|;i17BZi?QHQ{A-qqz1cEfqt5^t<#>VR{Le;zllqS2^u5IkRZ#nSRe;so)e-k_F
zEWL%p)U+guh=lezaT=vmOnY-Yn@GQA4wxNIW;X+~MCbCVgkB+!(pJt5erjm$kshkI
z?^@dQ`aAovB&xw+0EJ{V;;jzov6uLN8BZNjwpe??=!vzbr$=6rcu$-3$LZYL|o|IZhD=86~w1i>5
zeOzLuQJ|j!LNME-PCXYdUzP!xAOX@>X18YKtL>!(dlP;+UyT1Vb6=^Pv0Bvhha+n+Pnab2z$
z2{S6DNaLfgIVLm_W-Iu*=7mP2Jh(;lN!O@${(Vy8hIsM1%=e|p8@A3|)@UKv(s+&G
z2IcgA`6|z){KuhF6SBe{)YWws&iA*eIl!#&I!3o`f}5chu~tJN&c(#^0;HJ?3P;UK
zcnuq{@Q5g1&AJ>hH(Gtqr2-JyxRqpjAaQ9hhY!S`QlY3r!nvojc)NB1t5P(VDX_0_
z38-ankJ#`SGtvs%jHWUW3gibmUTo>@rLm*26Re;y%*
z7k;wyyeNi;c=Yzgbe>wnv^j3)0~9Zam>f9)A~gS=s=x0$uyfg3-6m
z^1)Q+dPsTeV{{*W6TZCCY7idMp_oZ?@t~t_Y9y{8^|z+x=Sch#2Mss2eJwr9O)Fe^RFn&}x#r=Cmy31K$SxxUckhzJ
zd;f3w%>1^Kk+C}0!4zi|Yx*kRyGKw6pNxIdo26rDy(Oo)M9NJZWjtBtV9(O#9;xar
zv924)oRcej<$j~t_FRkn1IZ2ADwbYlAc1ajcvb27>P_rA6dc*=NtJC*R6DBCS0ASV
z<>&tRtDsE%Q1IsJ%xY17!-GF$TXw0Gv3c%)K-XWu@)Rm`j=?*P;{F0(Dp7^rzY+rf
zYRf<_It3;%3ex`|uKodwofQ6api=^FoaM%!5c`sb7K7
z>G=nm{sQU!k6)Z>)1LYN=g0ok?6c(8P95;?FID$DWfaUgr$792ICxF}8X;a_E&pd+
zml?m?20E-uE3glIJl~9eyXYMiLy6~EB^1_*>$#hF1+a5knwix&UZTt3Rn^{IaMvz3N=Y>in!vjccJ=J)2d_z+$TY~N91pG&~g*7h{Y%>vAfM5*$R=Y!tRuJ
zRe56>#Ed?FZDZp(mO~G|q652FGPyz1Tm8if=&^v^kijeKpp`*V`z4|sua*Nr(hZb&
z{qEoi-kr;E>}dnBfrjh+CV8HXwv1ATUc&2LtXB$SU^kh=BB{Bqrqv
zQhp#|h%*Ti9~l7e%F4+d4$mZOp*4q5d@B-$qxzL()Pp?-@yq$oZ^FY#A3iIi3AfvF
zv8FLsPf5ZVtrw6LXcuEBd{uUYjglZhkWk~tJ3;EN{oBVv9uk%=9u#y1$9spZv{uL*
zo~&N%MEjMLA^)HL;aCrReE1uX7j7ihTUNCqEz1sC>s5M*WnD%zuJMQhdD0-
z+IXKsssxJFq#EkaUeKrZ+ghMa>Lfe4qusm)gmXR(PPfcVtVPP?StIoV1e|Fffcp(a
zIP<=39%a1k$?fX<$_OMIkB^oMjzQ4%x&h?Kwzs10a46r0}sTcfac#0)JA(
zdEwP!wZN~m;<3k-Dd0xNi+gP0c6jY9ycYZ%_Xy;FaScK4o+_3C2%lGOKIyG=Fq!K~
z0|u&(cA+8m|Dpms09ZftU%BPq@TQE64EK>32dE#QnG?$b5izi!^7XgEfYoV
zVFKzR9$QZ{Bg^1?3@bImTTTNfQAj8KqzO%OYHF(=3GGT68f1WyKK9IQoiZppMy}Ai@74AOKu+={t8$Y7(#r2naxv`22eyV-9F*Lm_-M
zSHDhlb)BuXr9}-yi_cSVDSs0O<;kl)f{XP5lNiM^C7a+xj>Wt?(-_Vh?KF9e_(scpI3el((LqyZNN_s<5!|
z>ZX|uur*a-HXsO#gHab5eNDM5ItrF^1U-`{zVEg^ll&htr8}*urv%yLOQ&EJZ^qxi
z@h=ec7aqxUzLNg0=J-3+`H#@p^=th9pV&;jTFL+4&;9~ge;t1g$eyQQDJD8?FsPw6Jh-ozWsl-!GtaTM3w*R3ICeFMe)BDhv=KV
z#9#gY{CGc;TUxfi%f)5E=F0u6%;Z)Y0aIt+YJb~7m}oWSep#$gP4#l-wwzR)3(-cKRuA8B+y_HfXKktZ)`$T8B
zL#ACuqVR&cg2MoNT3-N#KJ<(iZ9PDqH#Hi#dgF0Z8u@()ShwpustuMz9K&aOJbp4*
z3{uUhJa3ONT&b}?A|hH%I-FM5>&<4By(f)ohwcNU7sXRNOj`r{
z3W=vM)PYH%>vbi%)wI1Qc~^(hVt%NrO~y%uajV+a+$m_ZU-;-S(KonpoAA98Ztt{%
z`%fcS?A0oz2-1u7JAH~2oQR*H+?ebKHl>K;Qx%~`^Lszjy=9M9i>K_DO2hljYbic|
z0frhMMN?~kU)&c4|BF$R$f81vY=cCb?3M59j?F19>FAPI(Lns;H{R=BJ?=ZU`Bu(d
z38ybMfXY(E^N(^XcU=TqYVH6%lokS`I&$R;#53o7M(-asWVr78(LunO0S7;_LEw>*
zVPwO=o9kOU+FFJ)?tbJ~uC##l>*M1RmWnY~8q5pqkn#v9WM=be{b*`iPMHN2e>EXH
zazBrG5Z5QDQBmUBJLm2)URKjA-)BKj8awK)2~*w)ZJs7=Oe}EB%+u^I$Dp|3J)kIU
zSNrKgHLVIX=*MZ!KFY^Q>WVk0Tz~Z~-LY!2qagfRqiQDA_P4;=skCy%wdiTOToaRqC#QvyWsmT!oMymy03GhX}jH9Yq7meHzF9t7~Z6ilWH0
zo@Lyu=KJew)#OczG%3XkHx4FLQI<6Z?#P2E)dN!!%{tHBcu^O2jF1KP{z|9A4`D|Y
zO?zoZ>)6wKO&Pr(Q2~*NeWn$$c!%QYE+I6rR?JCGmCV$Y%P}&eVnC@ydU3k}7nw~eetR5D%6dG2Yv!EtehCPRc
zk8tOX8+IQ8XcHuVQF`c*Feb1FH;MAaAP}I;6$n9KS
zjD)*29F$**4ysQ#mvdcSM+ryYfG9aa`f(xk^gqgWtO??j;q{XcuiicGue;_Md)A+*
zWhQ&-pNgAxY4NxQI~X({Z4CQwXxmvFQXa40bI&uFTq;UDVL^95feMJ#nH>8h1SWYN
zt=}R>Cq+wf^}mX&hhlV(h_C$*l?)f`e&=6)E&1)LPVbN9PTmK9`E$if4m{^`O1Qwp
z<$t&RwGxUj|KHTT@2_%%znONi2f5Ly2#pBte-8Z%Oc?;AL_;dRfo1!nG~H>wn^xTY
z!ZT3j0`8W(09t?wnR5UrV1d%PUKkV#ysO}d2s2P+gymkRe>aj|xx}o?V!T!aUT#}6
zQ+*Phx~MK#fcDKD0L_6Rg$bE~D@ep3N&5`c>!?>i^fYs|r9qh%D9OS$+*=tT
z1{RM+?^O^;I@bUKTElf-LV%)~w@0%9>~JBj!?gN;#C|*Cpt=o%>VaQ&<^cfuBn*;7
z4WR25po|g=K>er9fy96h6l0QS6%gWvyo&Mm2`IV?;)BboKmXx_B>#qup6c~c>pq_7J?An_VyN^
znwr8APEmPD&e0Nbne`SY{y9?8`LD09gUBmLQaUs`@||c)_wzyFmpQ1EJSiaZ{5FA!
z*8u?BS(y6^{zXMcN0(l!%L4UI?|O_X%q2gXSHL&T9O()}B~Y{{7wmJi6nfan!s4T)
zIXfQ!et>4_6&~*__KPD{N&v#!>=s+UKt_ghTUHha)D6AYt#Yi?`V
zNc+J@TG+2%y}~pBkXSkC92Diqmv~Y@bf^XhO4ZV*9L${uioPs6u3x_%5+06gc(^r-
zu^QjLy$U!Jh~=F%1OWQQix&Xh;;vQA(*uhz5WYHB07DrA>0;p1qYme>Un-d{eQoao
zC^G8ASc{r@)EFooHnRlcd|Ce&IE)8XZdkp!k3kS^__ZLmBnwO6oq+QR`4e+pHio>U
zo2Ru&+mwkdcy)c!;1Fhi{4LT+%^S{A{5m~j-V>N*{P)e$_m>$SS5dG)_HQ}TDb=!V
z>`xQ96IkGXigN0Ls&92(?cexay!0DC{Uxs{44roBoj~~Xl!^w*K>ru`b_(h-xt)?v
zvp_uH6yW^-0Zc~!C8&x-GIARHt6S!7=fXh*%|BY><3sX5(r3nUKG4xA_k|74&niK0
zjHsad4My%8Z1I9uARiOn&uX)96m%@G6n=S>8lODt>mUky1zVQkuNqn
zX?D|YJigSGV=X2KSN|HwHq^^0>SgrpoNgYs^BVa$CKCbu
z=7tsq6Z8}{msotTf+B_wM7xvH{bQ1e^p2YBYK*=g)P8GhUs~>)dQ90%2ULu!U8PEz
zOUOl`1%b~Cw<);sOvVN2`EnV&>#$-)z8K>;)%PL@
z{M&MH*PAhy#$$kL})FE?8D*$$7G)7AcokN>H+%4clOl;T=!zF#YxrUHa&{mXWJeqVY9Uz{L1(fvR{J;QdT
zX&N7l>F4i1xl0jFs{4$_iZUfET0^^NgH4pjNd6
z3|2AMyLxyZd+7a)Y+eGU|##z|8-CZ`w&{@8Fs1PWg9B6!F8Wi4jgYnSv1BnBn
zZ}D-jwVqa>PC#_HhK2}`CS*!pwabxKy@84i5_u(k+bI}n9fPI(~6K#w=?{ji?`863=`9=Z|*w(niO
zHMmS*a^Jemd78KgX>Rh9g)&6_{SgI91*ucASGi5RWZC&xq#}Rg);7tLtC(4
zI98d?b^!LY`dbywfp6Bsl99OX(Us$#fYu*+-a2XA%(wu4QK}^i3fiRzu#2=7m|sL4
z&Zhz0&#_-wsK^#BsFEr!52VH7-DJ%axpOVg9Y(!Ck)DN^4AfMOh
zVSae+24|Dgnl0(&z}b|uc4oWu!qJ-okB>xZrAWmgCgUv|*dF^dx|&NV420Xp;n7^R
zrM!)0Y98-RHrH&lH*qT^?DbrxSE@r(=2B@)vru4V@vhyxma^n$Mm%|#6{g~@?~r{o
zkz(u2Sa1m4jx?(^^)5{HvaO|`mf-Qe?Kyv)`)m82)cmB8#TfJS;}`w&&8f(iS@|u|
z65EtSD<*gKd(9P64WZ4p8U;VQcsNBpFX%eCbeDEbm|qmM@qArU;n)zDg9l2*MdW|8
zl7Yr{Ie_a3g2m~+-~^&HK*2aDy#B=`*mTBImH&fbPW;GDy&1+*M~DB*qf~!(
z+NS&4*&HGJkCJtY<^WQ23UVZY@|V;8<7pq1Hw){(T)qYj^(3ZEP#?Rgk#6z=ETZq@CR7g2J}|z$YH$>xy9I*Q|+R
z9<>6hX}}{COaS!8JP-oVq|2cCMM2>_H|+S96zbr5z_LJ*)w;`F4zi_FlefC$q@l(vedr~pM<
z*mZ?+wmqhK*3n^UfUT*i3EZStuU&hdlV5!vQ!WacD8WDcdJqNoE~x4Pe|Sl7nFTzE
zf{3!MAH-uGmsycx0WE$YUk_u9T&ALux3S^G0yH@bGFFX)iwmAeA`j*P7WiV|6n=Jg
zzi@I(e9f3?
zm76#8YlVTIgZmbo4-wU^7eK=Y{Yl<)|4nD>oq#cfaam104M0uj<3htm+zz1HxC84A
zP~1zGnijsj_jv+45qtmhNC(x~Q}u*@3m)C?aXPgASfVK<4Am+t~>S`aJ`Uw~8QqVP2$aYUy_$QA%RHB=0&d(&r|k#vRPL
zJw&qDv2H_b+T3sFhM^`iHZ|N@n#*;=)8O$lyAFlhTGMsfh$obz_~!!}3ZMMs#v7H4vtCQplse{GOYgpuO$NzX
z$D4%e&~_rOo{OZh5p;?n)=sQo3G|Aek>em_y}dz-I*$W7b*b
zm~p=KY12oSf<9u*%w(sfKpQy^!AjJE1=3fHku-#qn0w5I3alXK(`@p%Sv%36yWeHp
zU4Y8Tamyg8*!22_*YAb7ZA=#k*6M$B$%~WjsbM3#G>M0n+Lv?`T0JVm<7P^cpIaU0
z8!lVzA*6Rxt?_iG?jxz_=qftP8b6Hm`MFP5v$7jVMv}RX=uumpYad>5%_L549&*s>
z`@Z|NgA{9Hy(mh_-HaU{E$le5y!EI=tRO>OL$6k!Tb_+;Jb9ZLwj$v<(nE1+WhN$n
z-=XVKi?5`|H?3Us_;3%+ge$S$6B@)JjJ4;M;~e!K1G4qyEvC4>3sZ8*2l)E*xR>>V
z$erG*;kr9n%|(vYQB?0dk1;UQtv6=?4=-Tu?}s{C&$H+3CQ7W0jZ8YqtPlk!8F_Ei
z3bj@iAUZY-3AF_;v(%F`kNdrt_GRz>+=vvE^yZG&y34}nYITWJ@=g1&R>^)_U%~
znW!jP?x;=4Jw_KYX4bLi={2Dr+0Vu{D&%z~^5$Xu5cRlXmGPGw>L1QE(+JFkR5J6Y
zi6uRP9yjI?Q=5AEXd(`U;>&uwgQD~}6Bd8Aztqp{lY`)j8=gmJRr*=yz
zEvbeiFD5^`LES`@_Xc~QIb^Y?Ha?Q-$N^1AuT}I!Bo$gg!%G@DpOAzou~*`hxdMBY
z=i=tMd#yWF#-}@0q~K5)r)f#JSb~O-zywEv1iyIXMZH~
zroD{vN8Pqcq&>}nv+^9=SDzEuFC~Fgap;Zf!YjR1LZypm)Xn_hXd}7Ur)Byjtz%DU
zQay-PTLB4xb?4wq-J)fo_@1^z138Kec?{n^WMh-7_Jm
zrXG%vN~MuSo1g!-zIZ2${!&wj36&TPp*e1mz2<{J(+P-UUu6~HvyEbXt09N&T~B=Q
zB$Ppp1P*qZ9>t#lQyK+mj|Aw((Zp~!
zgH)<1i;iJ-EH*FXMk4RrijS9s=Y&3Yb8+V2r=pFl6f9a%k2eK-NV(9ZA^Vk1Du@?a
zk0ebrpxNxbvBeEE6SHZ-S6X{LVr{`8xlLo>8f~`P0G(wmM;)$Mp|VzrtT%Jv4vILk
zOEQzQa%8uY*yUo4I6m8upI5{7n2X5#L5p%t4ZH>eMUZAO4SzY+-dI9!TBQ$ThPrd)wptYH_
z1E|Nx(I_Tb;fOIK)}Np3FGZ%chivi1EID50GZAP-OAt43dx5;M?89V|X_F$w9Qc9#
z@bylQ)raE?gNe6{sP?zBete(KE(dSEiVgQ%#ZW48Q)+ru`
zs0*(Mw#SgnPN-MeNl{oIX84m|zT6DVZp@S2K&*gMr#uYoc}1SPR)M4pYmqDmEBD2t
zBk#lX1Zh=L${j?bN{u$Q7gK_^t48h3_1w|K@e+XJ82O|IAV{%&sNOoo*((cZp9R_F#UodLcR&&bclqIv~=N`6DxGJ
zh|fyll1UbeLx(M7+=JubO^=88_}#_eom9)kTlx{K!vr_6cWE;B3n@B@-gz^Vnsy-8
zY2Z1_^;h?X47d&}R^`+h`>ZL)+>Av+=FXe^PvV~NUcl8|*Q$(q5~_n{auGtYPKyPo^LJLmP>&)?VU
zH`jIjuIu_OpYP}M9-T_+7hH0G6|&uycQVHD5aPC8xsnPyEChAOQYq!
z^2(EY8E0_on$ssi^@I;u3qD;aB=O}o0ZdrJ9LxxlC%*+k&IEY%KB>fb?_GATWbei+3yGY
zO5LanBXSO_ma`z+FASDugW3%Y8iMLUCG`Qn=~#8N*#NAsqLC~m$Vcr9$7m`KVtekD
ztdyl<0#XX!9`5W|*0%0{!hK@W>5tcZ4mSgQ7Eg3N9}=CBxq>+yNFpnm@!R^A46Q{5
zATcHKBjuV*UPFqpagmMD{J&^!aSPCsPQ-@EnyZ9JPEFKXHYEy+%&&b3LLpjxkNhF>
z<`kc@#JXR&-FjHrctG+HcYb_}bNgWOT^dUT?d_t;dr3Vu3Uf^pUNwt}31b?zPll
zx@t2{r|Qg-XNst#SV$ojHA-0UAHC%2dLdl-V_E($&d&Ut_u<3N2q6bqPd8Gf%*7wZ
zPP}8vAxvW3mz&#nh{jrD}|r*CHG&N+1G#(_HOABR8II1
zz8`B
zDpRx7-rQ=kV&d=_;k^H9w6?#FH(SkatH%1tQIQ5B{p$2Fy)>T6(xpuFMy!7BU2$V1
zk?ZZJ8+6i@G5(__rcp#Z-eslwK*c#?lNv%`DVP(vr8V>|xvD2y)`v@Is$4f*i*~%T
zWN^OKtR^TLe`0O**X)3f^ac!V?5s5siBP3r9KT^O;}oq+SjlEGT{&b_JpAQcAxm@M
zflQ;!lfJ9ZU#9w(xF63ZNa#tWv)~sw=P@a^8?vl_$;`-YR{_|Eb-^=Bf=_z;wr}NoRv4}(cE+iuSdfBcxn_q(B@}j;
ziChaQo>1iwPJTjyv%tG`wZ+^0wvw}U0%W2(*{8S_*3THVaxSY+Jw?(-Mg>qT5$!!{1~(?d)%^F3xtz@PW-mC(z4NV5
zd;tt-tBs-iu?9s!6`o&O;Hq?a|Dp58?NHv5{eK|Mlr%n87+8c7TxUyO3{?~;7jirUL&+T*Dl+m3F)X39A&cGiRA#>*u)OM
z9GAOFOb&=L;o$Dt8gs5&x|pnC;}wNyuEaXVR5Y@>l&Cy?sAgzQrLd|*HQ#;5a$B^I
zy^vvedQb#)Yi4t4ya$CHCLnz2teu>ZRMm7Fxmj@aH7xk=6C4+tEa%k^+r|7$yG`2V
z)cm{>4~1&t7S%TQRi_jU!Vee@AK_`=*h>emLP+yAB+L-QUe<3?dfmsn?9P}Ld?oT>
z4+AYu`slly(|dW@kTf=Ms47cL`PkbuwEUKI_2DQfl1nR({i(wJsnnC&1VUu7(Neu|
z@2kT}s`2Dv9t4r;%(u4%$jGTvKD=Q(ij(=K=7)V2s8R}>b-P3e3KF^IhT^2YsSY_U
z9$Bt%JJy%p)%zWPF`{^p4cGz{uon-{ugNqL`XwA|@HqD=^&hm7DvNqlbk5x~wanU-x#2L(h!&
zppX|i_K&32WzIQ@(xK5F&3md7AjgL!(NXF6n
za1P#H8yiWekaz#q7hiorfiF*;MU4Fh0PpmXx18JV6mx-W?Rio4!7%qq~rBv`@%m
zvfj6Kpxk_$OE0x_b$W!gmOY!M)TZKRSG?)(bKH=CPo6bhk;8Un_Bf#{>DnGkgj0l~
z&FScis&Vh*V4t*<-^*kcC5Y5&jpz~Ea;ix^$hF%_W6CGjP-%N*TYH2pv3TWE^NORP$EPl4EqLzfyX$C*WUh3bqg1TDq|_z*3jAJ&MFU
zOVR`4v9{R(@gkJUfcQa5S8cQNpVJl-cV{goFy*(;KL%N9{F1jy`0kLQy9%nHE64b1
z{MEMvTW&fpU
z-uyS|7ATSbFxLEMkVo5pzW-+p=x*Z9pH4E-^H(wYFX#Q6I{d$7jKIZe!2AM~HBsIl
z*s34W^LO{a50U(P#s0s*#{XMmX#wzDFk<5NT^YMU>f}VbZ!N;shmmT4kT*+(6b1n%
z6-Evb#5_Lv->*EwkvzfS{mgrH^40CmhkMj1zjn0k4uJ_v9unXhh`J?VpLsb(1JHMG
z1Alz5)Oim&ZfiFDc+Ci~c(WnL2u>c=jPWK9Cfj;l@18{izDT`n73_74LIYOz~aOx>;y;6*%MDKmUfO
zK#qU{ROwq4XsKh^aHwT5B+j0^8_aMJ(Bkr!6J*kWh5i5)X6{{WM%!A|0T*|0x8bt7
z`q~na06BESDla)$>}5e;Fe2$6K77zXTy;AnehWJ@SBRgCQd#^JWLnX4VI8mR^==lJ
z@K67~(FWyDK4r8dyz21F#Gv3{=n4%8T>x=u3
zGrYY|l+Mn^W(Bf0#9ab_*sc4Ce^ftu#8SH2V8aX3FjPYRe
z^m8?#&cNtU3UtUWxWmvRDGV9<5vRZk+z4xhVeDn8+KMY?h3gM(6>+jmvN0MPVAPob
zlsfZHj%Q%Ae}IZ2!Zn41cgmfzYzez7p6N+SJwypy
zA&&&?DF#x#2BO6oVj>Loa3o^b0&r!FY6Z}(Sr{h<*KZ==tXp1MH+G>K8ygRP7%hJ2
z05ud<8iiD~{Ei*8eAWsj%xl2zV>TuZGJ!$8)o=ZCBhapSM{aI3LJGMbwtisHvV`98
zCD$K>>WgSc+X@bd1;vb)e$cG2RHLo_s9qKaT8>l}d}18YX6EQ7`kJ~kI7
z;|XL;OpGG{`hg)a=5mSwOJ}pWQ1YuoUy5xfHP}#JWi(z$Th%t
z$Ti2SgC$Vo8BVLerBlbtyB3ejN;DlodYjCZKLGN*LfZWJAZDJ?ka}BdSG&Ml9%8h;
z3^ue{2vh)o6fO%

diff --git a/doc/LectureNotes/_build/html/_images/chapter2_245_1.png b/doc/LectureNotes/_build/html/_images/chapter2_245_1.png
new file mode 100644
index 0000000000000000000000000000000000000000..dc3a3ce3ef5be71546dc15552f9becd822286f37
GIT binary patch
literal 9236
zcmZX4c|4TS_y1Ui?Ado}5L%G2Bw5NRRAk9g)?(~ttYcp)dn%Od6w1CYqp^%#8boB9
zu~n!TQ;Tb>prRJw=>1T>U`e7V=oFKOy{LmgKKM&UjrvjXOd|gpqa5?aE{*;TK
zAKLe_yu9cC4v<6nILr6HZpsEin9zDwz7Plt_wk1Yl9??4fk;H@Bd(eSX0MT4)6K@4
zxBnc`!3HJ!zvE8{ygGH_#DxTz%oDKAtGW_#Iwv`~rLPzoKU8A+gW$-e${;QZnLUj|
za3bj6iXkp?3es~%7oTLLuc5iBjm)KXm5RB1_!%|ZP*z5s^B*6-+Ym6`_U32oxO>&T
zPef5pPJOMri|nUhu#r1a{G9rxXDKk4s{zy&A$hw5pBPtI@_3Q`&Yi9pwpQIlKJUJh
zoFz(p=Pj$hl7xB3(@dak;3*+bON%T3gT0E$2s5_Mig7_8Mqe_w#OY<4{70m$GK1aA
z9Cb{s&kut}&lx@pkD9iXm6ej`^h>w#(jHM=Nr}SHD7$7KDgR$%Z-BsbE24K{HZjNl
z_tzRQ=U$_SAVwtl>5IQQN;8d~1y>(4oF%`?&KSwRN_TMPObSBQZk~phs`O^PK+mUW
z|722}UU~QqO`}JY=ZyVt^T{^RQj6Gc&sf&hTj+#U@IyJfO%@(x0^ZsYl@0C_OFu5JR(I|9OFSop
zw1n+#bJ=O3X0Nm_#0U@g4VmqP8^l<(&mMGfkerA7@)SA_l5Y^j!<5<&azltX3}&4;
z)#~p~MGg{i{-03bPIS91C)h1~?@!dMxbDKTKiv*qyLZ^*4Im5K*P3Q#4N&!^;PP!W6g-6kprE1G6MOkBhT
ztS#n0PB?T<$;r7L5D?(Fi!g8+6xvyl*ja^J*3h=?z&6u;BLZUBqxtQKd&;q*4DI_)
z*!J+Sl{ezSGD;}+26vr~c*xOiUUk3)N2eciLaIZk2(sE_4PEKoPwnF+sdH^;hn;&3
zZif}~!(;!pPL5EDm_ucALdeZ6|9+KO`;F*671zqWDcAgDs?M3NI(W-%NONzs<>5?&
zy9If}@8@VVr^NN|8bda1ZydT}h?u#sr=sdxBO^TQ`
zf;a8|ZT>uORr`d5!}#`zVierJo^M3W4f#*l^eInRn?8sE5u5z#RZnL$%g^bCb94->
z_LJ2P9y5(eLdtHGF;k#IW+snN=$dxP!XSA+l7Bj;61nV_mh0{&v#ka
zKfS`A4ydPUpI^9tttMMboKHSxq@e50#{+7Y4oCDuc1?jy<;Mlj9@*fooo+o0#;(9z
z#$m
z#i91{uP!tOc^F4n^t{@zV4ku4q)3ETJpWj{1a~^rCyE@&F
z_)JuNiFfB*RZEI=Zr~$1n-6H9HCY-aV
z{GNBTFMYSn{KwvUyYu$yl!)Kwt1ZAaM4-^E>`UHx6X)+YK2@CyO}(9^a_rXd-A{II
zLpTwv>zDZF
zf`$4AC<_nY4;;F%nWs0Lt6^31018JEj^WM@T#N^h)5{#394vZmI{<8%YzgvB?gPWJ
zSmdygnab|8pPNv25EYL}{JYy(E2Bj$5IRP7$6sS5ke2Of_w%+b>HOU7xn4wUvXRz7
zy&3e-S!+)a*ftnK8jd*h)jZfVB|HVZ$+w$1+Ztlp8th+cQDyk8H_56F$p*~3@W&da
zEs$edf5J8hY8$3(9lHP1s0)4LTQ8bQLXYad$Ptx$u;ljT4AD{iRTX}{s?Vlw>2O9^
z<-Gs%RvADmKfc~aIejg+DhJe6)%b7gHyk~$&A=|Am!}yPxVP4{5jhZWl!%`5H&Q6o08<|wIo`%m;!&m8;B7irwCOEX?(
zF^scoaJvhX6Pl~C4$3Lty4$?$P!5(Q>E+t
zf14i<1Mx(w4S2ZN^ga7}D^(0KtaTJR>mIzl`VmmrMldli^e@x>Mh{uv$|PacB^$GF
z5T&2?gOkHnr+E|9o_KHei9LRId*M&tLKez?NJIv0b93MZw2d`W#q-|x=q@M=-wVK8
zfX0p?)rnzUC`>h^{o6P1i;}qUjJU4mHdfqmp@~>K5p5m`!eX>NB4WA;WmIL``gc~m
z({gXz!tQn~yXa)C6AMrsYTy6jv`EDMAHUVfWRM+w29+mXmRW=W_xmpAV0;$irAsa3
z7?Hot1=JLu=3v&vmY!IYF?@R6MaXc_4D1j*>!W@CR>3Vcz`?eGUC)bdz8?MH|2v)Q
zE_;?L>K*VqgcxyjP;T3r87m%sX~|6Y3l*;8IVTqUlR76Jyh8c_l+|9Jb(qe7nU!$?
zO@%F=vuL{*IMr8H)ixl0v~vT9=rO(_TI9f^uTj;?pi1IFb|i`0ZF*cPL7%0#=0_YI
zZm!QAR{$Y3
zGSjF4zuu$q)1(dW?BrDAMY_xzYcRz9y93;TqTuZ&3A2!zXJJhvoqZD7wGtwh^-}z;
z>vX*=paDr~X=$>IPoxjVFGZ=ri6uK!aer@)Vs`{>nyX|fzIbz&)~@27@)|_aL~yK>
zmv8&>))+cmQ}R8Nzhq>hW1q
z3h?_6km4Crw~cdeB2rd4A|hwJSjw6YcZa>-C&Lit^MmCV(8`|$T{t!`);)ZCIWJCf
zgt9=%n#2Ei-G%jbXr|FuGv1!RPq*H@$3bpm31nH}U)V)R3U04PacG_}J)IXGyiSUj
zEJEz&SkV%lk!+C__-6cmCu7Fc9ebHOmAnGFlJUf@7$NR6Pp;Vo`m*WECWm2&VX_nm
zMM#0pl73Zg5!*-7m?6~(E4%UB72#C7!@~FE=ks$GF}rI*Y5GCQht%KXhs4%Y9EPrq
zPBZ>*D2yw)!F^$W!(q!UEU<(t-j6~c2;ugckYBSE4l8?P5NESALgs5fosyP5FA&hp
z=`EdNlJM_We|NXQzTy03%-qR8yNJTKB#ij}!MelQ@A1;TIoAhHjnb}@-{06hgmKk~
zs@snietz8k?-%xqo}g=ZY8=n{{Nn}jqZ3_9NZ>ieQ4wiQ+3CreYHx|LVopA;RD&C!
z*xXhQyUZ#?|NIIUhD!|sYS7xIB;l^w_C9fOmEJUin-7nl%%EcLmf0^NifuK>NlhGz4+iX=LY
zC(6!;lvr%49@I37|D$7>dB@6#18V%?^Ca~oGljvooF5&ntk5KX^jd?rbn=N}e4=CR
zzpaH1#&3qFIY&+=>Am|mcNDEU8>b$P(1Y4aa#!4gw*TX8*)$>QOI`!^X1ILUn`{;K
zYE>pna-HT4Oukj;%6(;9z&g;uGE;jrjk!hdogn;7`EF&y>=)ZY=q%yz3qez{P70|q
zOQ?N&B`@yfwbo$tM=A8*k3fSs-QxgdLQA|@BYQ@b;v*5w1;
zQ2pzD<@{*xKGDe{kLR_}UYiC8O2*gXBbc%uAe9FR
zk#&(poViAb1bBdlK*&*#{GuUDpXZ0beh5AhzlV>F_tg*^-$giTA%|nW7(Q=F+I8{N~!?BQ&=}%wm<=ulda-2;M>a~
z+fxckU*6pYa4-Nkeh!G{h6{c0i~9B~Q9Y?!8$f1r!4e`KA@Vgh`6-C9l*-)j%O0g+
ze|mto(_OxJCi^Kno^W)I{ShT40UD>mGIr6<+lpzR85$kwPgZa?1Zp6UB_jugu)E0g
zD{25}H~`vjqm!jNMAbSx2)z5hBlZI?EzpNy^+E8B4)ELIwT=W;#?zj@2d_2M+@%*%
z{aSgO)+Mu0FZkyOoq5Hh(!0QNS#1pQup^T+s2{9y$*E#l9!kyR*9TLI0xGP8`N@ahfVyY?inU~jvTj0cWlBjTaSYW1Gwd~1Y&_po!xAaUnmdr%TBR!zJqeqzG8~FV2oc1d{B*oZyxgv;!MV`#pJbVy(|579
zQtYR25JP6yANFc54^?~F0OXhqM6_q{m?6Jj_^ONNd4eE7sQ&xsqn28itf3wi9G
zmuKY#sk0@F5x!A-1mGkFeCPJ_6#P|fbWv}W(XA_$3#vXFUPphxve~o*;e_`;p
zkaoR?DF3S_4)IX~#OZ%{2|O)g4);JNGPQ1rOOaCpL;VUrA{JRLwa&kpF`44B8T_K30?`Bxp9z%&R=hJ(X&Co*9L`!r!&*YZIa$c2J`-y_D;et
z$Pq*=K`}c{QpB4+bnC9=kxk)CgR>NM7NQnQAT6GxnlY8zZxY|%mZ_4Y9%ZHo-m*VH
z@KUH_iF}4b%n~nW$3cWuW1@SOhLQlED@yV1ltrrGPvy7cPs{H3-S!
zr-GEfpQ?SGKlUZ?+!;NeiHfr0`Lmye&oYkYN~V2xiwW+0T&zf%d|Yh;$^Yd+11@AP
zAxQ!jGS$re?l#L!I(P4jvIQw}wE;A3w7qoeZ?>EfDel`Ilhri;O68@TEJg+P7w|UP
z_van*8Nu$zoDa3hAlMgP=8TljoKl3}cSG8+U#Zn&n5v;O)D#Tf1n`?(+q*y^x6P28
zv3tWz{N`G8$X5;*c4DIh$%hZ~HL5owM-v<~eTSmdeDlpKK=P3?8$HQ}G#6cAv!<%^3CS|JDvxY5Qtc+hWKM;`v+`0?O>?y#>jeubGYhG*N81*P5L$PfNZfhm
zq?ABQiZ&gQ#4$DOUKgdNZGnjxCWtY#VW@LF)XddYk36l2aI3$Gg;l9W40Q?!2@_WM
z{dw|~Zo?}j2D={8Z%MCM9_;v(owt!CQ@J0N%Jtk%a{tUJn3PrPZNs_&^L=`hz3AAiWt=9eC!>gqJXrl-G7KX&2ty
z&N?=g&lOOQu6FW69ayb+S8P|NvIDDaVfI^cN1|Sk-`J0{YUt!7@VA=At|TE_5W}ob
z2N!vWDF?=$Vqa$xy_y~|p_gbR7eFrU;(g`%m
zG>+nrpSdX?U3KuUCrfAaICT*X)u7}y;b&@RfQHCp4ZLhVJjnu;3Wu$vq3?3%TU`hC
zmj8Q;nHAJ*{or8*_=&BAQJ(8Hsx8W@Ey7zv4<4)79^5|;NjA}%vECTHvZ^?KVW^=^S@{#GI1FCV48^xtENuI^r<`W
zEEICX*T)C;Od^FgO+J>}h}Cyq7shjpY!$a%JM;`qPU58{Z_IE2b@h6ZkgnhjN#$GV
zCpC$FI}69o>u^O8;!j=5)(+Fz5rUC%gP1OJwZvY_JgPe0=Kn0mA|71L7
z7`kG5fPv}esXXzW4HKOg`?P<5J)ZPVtVNGUNO^P;*KG(=8EkEq?X@*+9TJQN78PE*
zoxvqnS!pAT;o=0sn7^@0T-W}ThVFoRzvmp$HBV;Vh&WO%7Tpn{!Dsb|XbKc}bEY2Ah5>RkG>1~1pR}u-g@b$jKn2cwiw5;^N>C8%P
zJGeT3KP~&kI%m>B?TKO86zhmX_<0*LK|BpB5
zj8Zf|o;9&0;bOD&5=KGaV5PBnk{FQnPS;=nbRM#aQa9%<+8+&Tekm8!4+5wcHH4{K
z2MCUpF+SI-WGPGetHdC0a*s-+6Ay({wQvP^aOr#h>PURqJKaF(mr(GN+5E{39B*RG
zVOyvqr1Ata%OvcYXvPJ`tPaPpDL7oi&b^@N>&6hCp%PWqhJrQ4p%F;jW{^RDlW#y
z!(CDcnz0o3VXe
zT0;P|3<v+IFw4A8(tl$LLx0A0uA+@$PUrHqUxRB?lp8yL>#cFpklxNG)hZgIn4}2;Q=(9t-(&w%OQrgy@YYH)U5;ERg;!z3k*gR^;4K_}sSZm(HE_RHK-hxz5?(a~_fx-FU8SPqwXXEL|PUL?0bc2=I`m
zjE)=P(R)-PcP)<@PKA&oOrG;$xs@lz>`~d9)U|t>03B}43%T_vP2u5ntQOinP@cNK
z9k6Eq_dEJ1+qaWyKb-ryEkrdtK@JJuVc=Z5I%j+U>nx#^mL>9G=y1CniVjB?X|Yyi3;I1u-EAqLSAC?|HVVHtwSE}@r>ub;%yTPb{Sb?6c=ezak&Hj
zABg_etJ`FuZ%wDQc*12SV7>7XKU0>{R8#Bqr2W0Y%pXn5ZRF278#DJCpCk!W!`Hzv
zO~n2;@z`Y_R5kT)Gf2O_-y1C|K2Wp&UVYrzD>z>IV~)A;xnOGE?hBnxYJo%7{|i^keu#vF$E@U2kA8r#u}Ur4TJNvzgJ+epPd9gY)EpVw3Jh3Ar1wZr61Do
z-9;cIJ84nKBrXjMp`76}-RJdb7)+lFw8fnG$?i*`yz?lsrgG;-e4HMQ@z7f<)@*_C
zy~McwL)Q{o&SWJ({IHaFpz!#gGdhC-#vFrB1fNLxf!)6`_-W>+rp7PJJpOx)UYsx{
z&lVTQygW-`>?ADlQ(Y@B$y!Xw{O5AKC&xAt7gs0?c2Sa*UKv6U|6u8ObeadQ|K~bW
z;8Qdx&=ME7dLbtm94nfDlnD+64L$+&o6<9~th!R3Kg-QvQ?2VkHue*2;dK%>!|!eq
zJf37+xapYn64&BOB&<3kx05s0F4kGTug{oLLvG8%_Xq@X9Q@{0aGG?hw2VRJIu5SpaDL1?>;EPhtLUo6tN^fel1bxCXV;rk6v;`R
z9U`eiOrt^`45j||&#%BW(P`33yNFXSf1%aG;tDcL{Q2jHxhLOsIoZK4m)
z*S$hL6->k+$rsa^OQ)0sHR+#!%GKn*`1G~bdArx9i_}2VD8?*qjxs5)nj?`+E7N>c%4?ZftqSIrj2y<$HL`ejPLTo}H%w@w
zm6P%hra;*k-t~#XdX#9|&Qb!UWbc%pWr4e2o#b9@O)r21>6}iCL`6-IGa3YbwA`bg`{xSvg~l&Qx0K3NYT1(1*4aS|e3Xtk7lu(BHp<5Ik&K+*JNtj{o*9
ze%VorP%iWa``nlMdU41!9fwk=Q)QY^SZvWF@vgWdmuao=r;*#`I6HG=X}zq|LYSk^
zBw<_#3kG#{$N$4a#e@)j!s;X~$FPL1inwZ>?}>J?ZGWox;!x_NC7QlWtDA6j0RhOZ
zcr{aly3aTvy7T_jBYsS-HS((rCM`W<91?tTWB3hj?kD*+_m&$(p$?X)UT7R;oY-Xb`Zy>38IXTlklIeX{Nh2DEmI${Dn
zK9|=xUTFW{68`To`&mj#3cEGOCTI7be#ZU=u&&0vM9c;j6$yeqx
zx(6QlMg}7T5O3G6*l;ewhrN^`@%eCG&%#zA@rLq)!QBM+ia?>>fsYi`WN~D)CE9((hGc
z6UehdF+qrvckCV)I$ui^f0v>0g-Mx^H9e@j&CWbh$+C@4@cOImZrenCLn+mZs~LOz
z`j!dt*Ngj1WxKpim)^W+f*uxp_J)S+hcI86?V8Pb-}yU=2dAm`U}}=A{kog1!@`mS
zKbLG#hyW4r3rTmtou#Y<+YDXik}I@L7f94%y;R4ut_;P(E>=b`SC5b^xEs%T7!%xo
zrL-d``96n&K2^fmX_uC-cLqG6;~u8c`Z)KENwM-D*dw*hc!bDfoS6qS2(mMGI+r
z!zfQa-R+xY`UXW#$OclF@8a8U*7@#16Gui*gD^f++&LvjsA64glq~r$!rM7{;3~gW
zkz!UR);>r-~RO?|+ue|9;u!AVNt%G4j^Mbvf`2
P9f-cJA)@k{{iFW_5~=1b

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter2_316_1.png b/doc/LectureNotes/_build/html/_images/chapter2_316_1.png
new file mode 100644
index 0000000000000000000000000000000000000000..07584f6bbceb100d07a9bb9be87c5a061d00ba42
GIT binary patch
literal 10353
zcmZ{K2UOGDw`J(PBuGcm(4>Q*t29GZK)?^_O%1(BB3+u&1*Ax?QdNqAQUyX$Ktl+k
zBGLi^0qKbJGB@AMYc#6kY0faK=$Kpzq$+3U!;V~G~djKa@Ki6pI`x+k+B(|WbX$C%ha{l!(X=SoadGVr|
z;Xkh|%AI;*Zq?eVE?$H)NXjsFeN5tiVbvfiE-s#JMYvz~I3Mjf-BFOluk825K#bL
zR>#rkvjG$L606#VmDCw4f=n)gP)|4c_jzYYO^<;#2a>ZkwpK
zYPhss={4Ty`SY!KfU1Z{H29xF&i(5}A}wmHhhLbBDfvt>V@FIh26J2r
z4Zxu0ou3{CCMKkPQW0rDtVmc<@!5w&?yUWXiNXRT;&+=FQjyb7g(@=mrJrg~2cPR%_dLtaXzG3`T41tlEPWj}QKSKAz=Q@jXA;7U0tTn86k^th2p3
zp7kq5<6Ojg$6@lvDmvG(
zdXXCIca+6G(;YW*72+Gc#xHsPcs)USzc-us_2oZ_LmH>~4-a?lPX&?g4~S%)87n|Y
zN+w=jYkt%;6`VLXch^4r_(2StAU;FfoCkqo=2!A0_6wy53kf|cO;OLa@op)qZKt7xtAC9%4T_VS2?cZm=s%=duO;%%!00BYqG&Qv=`Rrxa@kVLL)sUUJ
z$$)7Z;v+V7$G^Y7;qp~EB5=4H9s3J;aHg(Nhj1fnYd&s=P)zL6okidx+rLuee9z~jS($lZ!*(C+E?gXt-PH%_#}8nHjfq?4?%~dAZ1pSiN>;E`
znafZiX7%SYokXtr6KC1@QTvc2aI5klF4mfVv7&|w?bhDoRX?T(np7GW82-&~z^f&e
zU(Y(%=)E@4B4zzHE?_Ds*0fZ2b3V%<^Kz+v_kGrb-Prb!hv_%X%#Z<z8T$9z?bvFM>5i*D
z9s8moJ1b6L;QK%L%MgKw+rwY0tF2nijaCs;i~GS6f{(HfGg0+MwdL^`%HrKe(8ale03>k`A(={!U-AS?BH4
z$BIFhaCvkoS80>@^+E$Az@A?_U0VaI_MCI*2vr5MRaiBk-g=E^0w;oV@DoST)O0>3
z8poK_!zzm!zJTe6%3!jqbDNpcHecF90&)g|u`O^Bp
zBP-lF_npV6-$z~->Wqs@?ZvwR}B
z-{e8qtmS*y5n7us>z${`yBr`>s+U6EVt(Lk{EY%N#YKI1+VqG?k5~~lL)=N4@`LFG
zeRhz=l_9%pfL>C&L;rG4gvi>4sctWEe6!PA5s1t&l5dbkWm@zfSfh55a~?XOJaE-M
zW%Y4!w_NF1HP@$ZwyWL9r#C(KL_SR{v68YN(NcqbuL)hNN`MA-m0*`+St!QkDs
z!M)92ON0_CwksPDwyWFhCY*olnfP0S0R_jdHi0AG$d)VVKDQZ1?fgS-Z^veD?@pB!
zVo@Pgk{aJ~?nCt{v{~5fkWITRwgfK=aDLpUvjAwA)QpVY)(6WF*O3YqKs`d>69AfN
zU%%cmR^Pu1Kxzocg06d6b{|EuY_(A+QC?nNK!p?${NBxx8kxnT+D_MQbH8{-0^Vc?MN}@gXe_BQYWq@
zr>Z}naJaItafSt(hV)4cQIGaM^c5QxJjlMR&LcfU$!i(?LE8t<<%>X3(
z>*p~g9mhN4sHHs0sXL~els<2b7x%c
zfy6|(M*c&8f*WgzQejF-gk&j{_TMQ1t0l$8~8a7xJJ#sEdu50vq6f7zQ#
zfb+4;?H6Fmq9!pxe;khjT_;o)6?763wQsig-7{8ODl5Jd7Z=E{ge7+Q&P4l;dudgQ
zTi7bF#K+y@QB^(pJ!AOnNO=31#L1|qtpF6DB+svpT>x*d%kB^B+|m3X`wtxc#(CAc
zdDp)A$WH@@d2$of01UBIbVpg?Nqlnm;n_3=RlQ_|4`aRa5lbrS;rGGb@1rTGxP?KU
zhnv|;-tNA}rj>HecTf?mt;k6UP2Kv@#1_08Gj~^G<_5Fb$kRCepeH;OZf#l1;6c&oWMjF)?v#
z59WdexX*%9WPFbm>_!ajsm5KG8w${=1kNzy%gG_s3Fe2XUD?tnxOHnAU4z5RD?!5oYaTp%BB%;yknxANygJ=*(HGaT-N%&Dw
znp1PBK@qiMK<2=VA3xune46o|Jdzt>IG@&c**p>-U8B%Y8tD@0jc;j)(9o4jjMLv_
z71(e%QFO;L$wpWa;-ABvjjrKabWoXQy`onm$M7wuj~l09FQnjbkxRh}fGOW^(gLO=
zxIGu`N1U&ud^hM>k|V9$FhzlarXrjCM4kvRTn)x5j#3whzKptJ6-rHDY-X>bl~Qal
zDL^l8`M9E}A}An9Fk){~91>&!p-ov>?9hW2cH^lYHGkx80lF<8JszQ>`6o&VgEzw9
z-}&R;856;nO?dzyS7lwS6jDRw>L9mN9J!`r(IPtGpK
zW_N6VDITFFp9iIMc9CA-4vIe#s(BAWn)rind52OGv}Bosq!*y0{S`T(3~hF8p|T~q
zNyK=uC}WOneZaA%QC=S@?+889LYC=Yi|r1W?&=e~Mh-lN8=(%$(r~kEHjT$Fg_~&h
zfJJspuI3B&IQ>}wwPDA^?GdN1RbX}FJ1p_~vn|Uq8Z)>rk@D`A=rZ%taW)Pz;sL(c#Ka;xG;4ckS8D@FG!L
zkq^w!IA-$~ND6Z@(rnH|(g6)Q6~gxWD;Rcz3>Uo}BG(;PEH!$!%d=<;L>H>Sx}M|j
zJ}WbnmtlXU&=rV^l2gTboKW&Q>}u$d{MTw;Om21=xcGQG!7%!vD^!nYwPIO8hFVF0
zF@SrxtCgn&
zY=g+QHjUnYCy_R9BM#Tu)QoIz0+{M$B^^;CczHL5^=_6M65?mIQ}w7({&t0jqGaBN;%h3)!la2V*{u#N
zC23+FyMD3baPd|#+LS-r3bt`^Lu=DiN7(7^y?xo!Tt(1CzpS={qQg}o$#jbz=Kv;
zB4JtaElqQIM4tw&x(unqh6;geIzCHmAaBLC_9@Gw`;Krf}rH+<0)ILbhFDQGubn8BOgKDHySYb;)oJ
zy@1GL#ZV{9HzZ-{wMm@$=%Fl;>bTw)0L{{G8Zurv2_(3|jPFMiRfx{TbOerUrW*nH
z8u7L00`yfwzHZgg%ChkUeE~kTuWTZ3d`s3ZNtpgKH{eXNZA5!_px0=R7-oGUqjM8A
zk=+YDY%@77M3eBSomH^orwIimI9xKdOphRE8N`GN4d=g(glX%>2wq=&f*{s~15)gk
zSi8HaP|R?mT_3w;drt~(7WRm?Yhh3>MCo5``1qd*Xx9U*3{i$?eE9d04^=*TO&N+4
zg>SCZG#v>e!uOfpif`qb#lN1d6(VHU!C_%GO$Xl3rLD2YyCBVHI75;_zppy~4+rk<
zU96s<7MY$61IdYLH&WSCZMCsjN+>fRxILcjY~F(z@)q%kvpv+&Mtcsr@#c=V;kt5@n)nSosN-lobGq
zQ=g?%Jv`W}C>A#_PV4ZO;3K2m-K&4kNZB+07f|Xl7Jrd~r;>p(K+yJnv>?h_LYR!s
zPpYL(=N%eE-)~S_##bUzFPa1FKz_`Yf9=VO%V+reo2EcFp%`EmLno)CTPS
zwM7LpW?;=wM{bCKcT^~o5f*FPAu&A6T%zKMMpux9@G-CyhGYbwsy^Jel0j(3GTxu|#
z0f1vOwLGP+)ux^x(*BZ|$O8Eg1FCe#hdn39VISX+<+1?}nFbBLzX)e*2o?s>8mj{J
z1`{PaF3qZQN}0wkRDL4ZgWx9o1SqSrz_BqxC7Zx~>i2ukD+-<<$OJc=#7~#VeVEv1
zta1U^qM(=Ia{nc+7_Fv1<1F&VvQUk^xLiGn#JG%}&Qq4%NxIv%6r&D;7{2%jeI*tn
znuOdOXDULM>gX9z#PW23i087h3QLa6Rdkv~Dl&X{5vRzaO5)0~GK4N?%~>6tapeKp
z%1pOZuc!W=oJVHSBg-Ujiapv_`oC)8#nhRpsq7GaK*BDAGK(EE1WH!8!Act
ze@Q7Um8Koh>d#dus|rb>AzGjPWLG0_O*L@KU~fsxbrjD{Wk8?0qWHBm`|ok&N$Ste
zxD0Hj^u|~7u-Sdw4UN#=_Fe71`G~7nRkw#pwnSz@4zJ%A)RU`LO!^(&VViYY}bhF)^*l
z=HQdrEc9!~$WFDu8SQBnOfS6y_3!U2Y6Z8>gGk#Hz0ALW+th=6&g$9h>)|lLMo`==
z)g>&IQJP&~kLNzz?RiyqfQBr8*ZTGGBsXHYE>)T2fZi}P1SW7`LVctNDGHrs!qN%u#XH8Z{Navr
zgY%FblJ)~MJic`TsBkITmpzuhL(H^MfIblhd|+_H%Bs_tx&f8svlFyqwK$>&*6B9A
z@U*wyC1rg})n6*~v|VHK@b@dQeWC5nN5Z1V^VgmG`!_oGih>y%y`zEDWZ)>i&JC`D
z!6aI9c0Bxax$~$f1aCq;($dgLTWojPj#?B3dsC#1Qu=^v%(k+=usnz>RB8ZC$agtz
zF{}(lEd0dA@zC+pJ1P5=ubpXdd6i>W2kaDZM}#G^S`wySeuK(&f3
zV&L*ZqQ+c#W0254&?G)ZMwKLLnNy@33x9XI$w9p1%iC;f1^PIK`soQ#;^QlyS+1p=
z{iv!utUCHrbyQ7NRaIi9jPhp_o2XILh{^8pNx=HfD0S#=fe)h@lwCaiJX6=W=K9U`
z5_H6k0>P)0t3PI9BS9-r17cfaEr+h?gg5&*kf{!|bfy1ag()^flbdMcs7z2cvl7UR3`Q2-_bDk=GzW`OQUvv(|{T_#|40maP13r{f$i7T{}&ep&gE{iO_~3
z`h{MON5-OQ?6>H*{P3r+tAU^K$&>R58Rb>;?IFjhnl;v7h1xdak2{Y}uCp6lWH6`C
zWPO-EBDj2A#Lv&i;I3f3Edw_OSJi)5+hCwJ$I;)m!YX4kJl+;n=o6Hcaz0j~r6}0A
z!{pfE*`a*Usq{+VZ{>}QAxlCF(N~>oX*6Z@p{{#g;Iu>YrN8TJkLa>i-WduxF=|a_
zifyfDj{d5Ksb)TIFl=ecH$3^g@^t>u@{7kz(xf?cvFSh8TUI3YWk)9-E{r)C*h|KS
z);;r44X|OgQxANlZ(Frv?|{8SOT*^0b;=!jLA=wy8s5q39FFXIn?A1(|GL&KMoaKlH-*q$6tjv9~hv!l)(UP$I)CIp?fhzyA_y=ibHBdd7@b
z4LJ|JP}8|%uP@FWfNqg|g8)ME?XvA#jXE<_;*N_&ACBwsNdt)yjwvwTzH=Iq#%)jSJBC
zL#V>iNhbx^KQx}^be4h#KC6V|pHTb*24CdjNh!88+B1t)faAl^s~@(Su#L06mu}MR
zQK}%iTksHLDlW>J#2+#*i*6e<&}`>irDbNC6tw7_{(+|_UEFD>YC3xq&e6;jc$M+g
zTV&uHT{in*XC)zQoJun$PT4cjW040|SDa;>#
zOeqCX(^DCx!?#eCd@zZONu6*ocqTVGhHb*P{KdB%W0{Ew^d#%${6n?0!;7zAsKf&o
zuHfUWYr77#wNExq7|qjm3ze5|5Vaf9jCaos+eZ50r76Q?oYN&E-0+8IRDR-qXSo#J
zx1)#PJTr2RA={MaE;^?FH#{X~hVA{fXQvBwM13Cg)I5XGnrVqTd7l8ABmI+cUcDh0
zAYIAfDGg)Z&ij`+(uf9=!Ie++a~B6fh9*U~uL*Nh@kdIP57PKPsM^0*kyF4vb3=r4
zr?WD<==~H2Dj;GB
z-|`UUk2tBC-=%O#h|87*Pe>9Co0mMWAp-XFN)%`RNGbO=?dmFI<@}{+rQ~kC+2cS_
z_eOksS{kK97p`JcE88eH_mg_=_}*8!ZOfL7+37b`qx6R)&I(V{QygwZ4Do6`e_X&_
z>4#arhrki;d)}CKF-^zn6EZg5ApbBEoBH4iQwK)
zCH%k{UuuxpOAmoAd71HPh1!?1_qyyT`M#!`V8J@+$&_^N3tihyp}RmVm;hgzEA{`*rNT}F5{*Bu~&RFxoCF^-6*oCo8r&}N!uk+)M3F4`@fWpRZV
zk#@H@27qnc;11z%vn&4>(1G%e{FR`m*(lWI&zlqD*MwABZ)zMoO5s*(EzS9++z@bW
z_wu>-4)&-3zyCT0d5@TE9wuxRa
zI?#9ZYwp$!n0O?}U3s4ESLa3^B#SgIvkpZX7tRZ?9?NHjzt7U{3zu(5-V-Wn#J4P9
zKZPDv^L>&ulj5X|=T3}VLuKAY4eU95IqfcD?@+_k7N3h{RkhCsv)ab`pd|FA!_G1`
zoA6Hebp7EY#qQ+93+lGHEN))k9C>kWVj5iGB+6ec?31?kIB!^C)b^!?r__MXHl10+
zGR9;h!YRW6K67hOuIe6be5#|A+(0-wp#pNpbdO|DzLliEA(Cn=Qv_Miu#$5s_!c{C
z0LtHx;xL?<#|tGX9+JL1Mfg+Hm!iV3!A_XMeJAF{op1z)s{209siFButmL{{#kuj~
z@WD^hRJhx=X?zIEqDQp}f!W%2zy_7TO*2PZ8C~b|+cc^#^z`FCQutgZsWYl2_R2=R{_@GhDBAA<;s6A_LfJ`rb5QM@sBry@cFLm~pnp30}a
zDdFUhVBNj^O+(o)A|f>0P)jRlJz#GL#aHXMbY>2J646krli?W5M#;4ui^PM3NPum)IRDYXycK5;G?vuQlIZyAalf44)
z4di#ADEn?=U-w&=o#d&>$7iO>`loz-M)E%8ea?v|PK=sCrrs~Y@A~S>XBz3Wz=X4Uc5ihtYpR)^k==g-{=&)oUf2)W$?@EsdUENltOgS;Lw
zhLb)RY^{?&R1fqP_X>jTloZ3QbIRNjWC6>$j&9)yaXg?P{HCf%DuWg@5g-xz=`cY9
z%*wD(OQzQfM451PG0==SM-DuVbFT1Tq{ULtV2x;(k3-n}keffHK#`vBO4EHl*wWeV
zhDfE?wX3K%isPz-lMcKBKnf?9-2WNxXFh`W5Svu*4>Y>KmK!!T#a_-0m8WMxa
zFnV6Zuyvj@RsN&d(Tpp)HQDZFSzSL^W)0ob91&k
zN{@UQ-0_{E+Pp7ttL@;{zh~Qej=Zwkab;*4AlmW{!RDpo?|?IE(G`!~UpCQ)ihxg`
zqHTG6mckW4bKuw!oS7?>?%e50U||oQfO@G(K?)py55i&(zR$R%!5UT#$XOIvLp!{T
znxKzd<}DF~+NyRk-HQa}N*5Qje~dDt?$7(d5+@jE$c_hRE6T;qL&hC&ay>j%&W9Iu
zjV`Z00k7#pM`81g8#gxlJp{Y%&Sa^vk8r#g3G%OWDi+^6scL2y#}|G)@q5Hkcq2V-
zB)l1wqdq~g9?p!fB0YkWqdKTYm#u^x?<|Jnu+@JtUHK1F4esm(rVG4(%f;kFkcm(&
zZ~rh+u*}dd8r$tq7zr1a@KR6xz5Lb*tylYj7RSRXc4(&Jh3UEe0Ox9z(W~qA_eaJ4
zd&oMnFJ=8df{tvW0V66kX6)Rc<#OP&YNi7(8{F$RtQ9!+DQK~4ccmM=zv$K_fqtyK
zj0p>-DyOzxNc8UYB6C0R^M}EAjn%v59LrcMa;XaAJ_NEZ5
zX_lAboGiC%I&0Sq8>Ez;RXKMEdV}xlNK?YQhi4fkiq}Ttg6M!<<{2@Y|JK-lz_=e4
zyq%u*T&|$7v;OM8wElmgG)mTk`eigqgTK`8T8}VLQ_kJlvVWELpHy+Nkr(|_;a{z-
zp#`{dke#m@y^9^|GVcve+VLe?pYHmiiTJ8(q<6N|aWV-YXC9FeT{7b|Au%eU2HXa~
zh$=_>?A+N*E2h0ET?yuVbXr+)a)2u~Xv*UBxamPNy*MGGu#Yb~LDnvB#ak6R~yf&_E!R`USQax4lC?+yOQRH)(2UT#=16`6;+!T&OMBx=ol;*v??C{SrfkLUw;X!me
zwQ5-fbyjOT0k(UkP$FL}0AoEoCf*Due`>RkStl}qoAhfI?lBBn^hFTWS3Y3-*-Mm0
z?h%hf_>=V6h9D@?7d}rT&}${Cm8x%bvIOV+;Al<&N+^C247LmWq4dgJq~n@dgGU1A
zC66W4Z+x{2go~Ptl2!f`;o_hCzg^ns+1@j>)-}HZ&|Lw@9bD>tVkl?%S4;mACHdr7
zan8Q&YZ_d(t^r>zulMPH=8%QCGCK`OyfX9_kWh)XRA(JzzX274?n+hifSTRxI9Qk-
zn15Og0GoB2tT@>yi-9;}jHqMuT3q$NC3@P)23GZK+k6|5gl5!&=2vl+R6Eq(Bau<1
z6y-?h7mX_vGL^VvQy$hXo(`mhfDT*i%C|*r3)Q+;fY?{qt?P6!U#CSY-sgT~)8Ve7
ztNi93C(I?FMvPIdWF`b*`3*P6>o58m-p*?{zQKWYDc@n1VxagN+stPA$a%9d<=2&j
ztHdViYOb3&{H^>cy?k+avddXmuv+loF=#d1uY)htgnq$3?c)h2zbIBhn*bleOROY4Qlh4L`{?Nx6CUGdNGt+%FO#+gKd^1H74I#eMweII^XO~1eFjy)
zK^bp=%RuEEc0c+z=1d=JxaiaxW;21h)7H~`z=VE@TJr6soqF784>veP(w_m4e@%qv-D5uLWEcDpwflC_ge{9?YVSEC7wcJ#Y`E<`&b*7f
zNBo1A7iIQo}wNS6Q=EvzjMj33)z_l*hxTqIx#ew}IahnfH%pHX$avMHp>J5zF>Liz}
z>c^VK@1NdmMN#BlL@#3+sak-+QAomJA=WM{WE3v)u52ONk^ru(M;)Swx1y-ulgzLd
z$(Vy1^hzYC0n^LgvODgH9yF{-5Su$+>||Xo{HnugRfnD4LV0FGHOoIdNErlDb+$;3
zb?hOry0$Oseb*eFc=DZT@$6V^+KNJch>zrxHSeTv{RTMCXV_%t^23N{KAKRfHyaR?
zd4yFIcHu*qNArsXzH0(qyoSP-__X<0j63sz`hFHb!%(Lf10tpzzkwN0Zr3}
zEYy4EMwr7-KzmgK5ok#Ey;7`(A)>3jfIPXFGooK{CG`|r*^Sb1A8@YKvu
zYC$jJQoi4m()r@fS(zopI^ifHApsl8zA=$uF+#u(i1q1ItmISkh4bDLB)bUI^R4$*
zeK^XYDhzdjv^QEDm5Jv%4<;MWF3$A14N2*WTrr~QKdUw)ILxgbf)(<@@kpX~rAy*q|wA3iRDTjjZakm4$!zBK23H#cV-W2NGx_TCbIV5|@!J=X}j
zXAM0SgM&A6Hz)au*K<;Qn`LHKF-iIbtZa%=x_?fBn~(;E)ovdizl(iNz3symjYmW%
zQhsgsthet<~tKuFu{t{^DJz!V56d38_os+Qq7#O1}k+}5%
zdeThdl;Damcl@PGwQ=*L^P&v4R-c^@%tyiE8lFBojzX4v=i-3R$0=e~(&~4gOT0d0
z$}ysav_*byrlw=1ROfxw
zJ9xJk^_EEZWY5RpS#|b2+DOetOr$A#C;eHl>!!F6D4yb&`u0Y$BhISKtnUio(HyWwyKk^)9Z(
zrM9~y^gi;CSvt_7{EgE&?S&Z!su=@0hb+i(BvCV$nzYM9f4+n|nKqG*rZRaCIy0|_
zEzIP^qc@=$8d;Ab)9|r7fL1fRGJTS>2>0oBcBKaAB*b5@|LwhG(DL9V?uzypBdPAu
zV|VN9LYY0WSgsO{Xc);tV$MK8Wk{)*qRn(B;-Z(Y(fFa8jd$AYI~BdYC|gVvT-p?Bp)NbIOmdR^HfZbM
zWtW|+-&7;lALHjJi#^RuRRy9@Ut|`^wSG^Qv?7y;4x>9b-*K3Wp+~&-TK)W_rHt!*L>NO(C=vLmj!MyL3I>h$RTy_K%x>dI
zWz-L?6fQpHbogkCj0^X4x0cN$!5SRPak{pdNty}yP34d$Qk
f;a~A)t4hVPhes*%wzK>XP8iU_{&3m%9zXsc;hzEw

literal 5148
zcmbtYc~leU+D`}&WtCC|k)_K3T0~StHX*fEWn@!vKnM_2K(;_+-;+p{+iI~+ydbi*
z6-5REC?p_D2&EO{LX@3^C0tRIH3msQNOFUh(ga<;Jd8r2$%^}hRjP_#<+IOg6c_vWy+`A{VnXASj>U$c+>gbRqGIBs
z!jFHH7!n&79usZ7Me*7EQCNIDDbB&liuiuPmYCR+R`>Gi3Kg5ICb{^=p-}3&D_=C~
zN})aqrTaDbn~nP@7I87hd3eom7-I<23F+Qox=2IG|`&62}^uO&K{+(ziIsT}*E=ht2@y
zR#kQbDlTwt$*2`*w0+sMe9zeyAmN@cuT-lj^|m_7ZfU9}a_;#f^F|HO51L^u&EU!r
zhU-Am-mm1(9~I;m6*&!rxO5`$*sE(ZlhqgA{dc_#?*6!q06PRPxE{{dAcGU%{po
B~0e5zC
zS^puKTC!(8j%h<5=X5=aWN&>L^KwOP$zv)$M`Eph2tElPF^?g@Q$;;
zXf64#+I1MZfZ@C?Ev^xKZWTDcTpr56MGp8K>)^OY$ejn4y5Gg=kJ}i
zpa*vhKZ&TmTrA(qwnZk&nX>L%qlJlSwG0#VPR;4MTMQ8P{>Lw0$+MG>-NOdqs+%oi
zu+i%yTACicu{TMM4oh}{Yn9ILn4*ov3!KbpC@sxtay_Hj1ug+U%#rVU6-t)!||CZ|UxE
z?H8N%Wzal8+Q=#J>pH{6urq_wBKcZ_ow*WBUY##)9K+Sr)K)-GIpisNGtNFqD^%U3
z=HDv#_UU!Nk~~Rv~GW{FIpnQZZ4zKDOyL0CtV7XT9Uq
z*qI?P!dI|t_cSbDe39kW3<^uz~UZ_WxxCjOX;D=F$
z20RHVGVLPZNPB3OUsO{TvbMH9
z$Mb=A6kkl0K?$Wp#wyO*O>2`U`bi#*PQ0Bsr6>bl#)(K8KwqN}54elgG}FIb7$|UF
zVIh%eRDaEK7P}>;CbI)uyurX22I#1(8cR1Z&Lw00;uZ)ZxQP-3zo8|5hZfbsERX?&
zZ(fljC$*-OVsRN&eLUPsaR2Kg=HL}IuJE6wTK`QB{}%xFNw@2P72VGR2$$=mw}$dI
z$C>(pe}?GKa=miK4;K}D*#xrk$9kEW9&jUU=qBQW9+A^k5tCTHw4m%rBM|=ex~{nv
zwPaG5)IgvM&F>Se5859&g5K4|!#!atiu2K@MWY45pORJ!XQ|ret4U~~gm1P)A!Db|
zOhyQ$fJ$(O5rp$k%jR0$qMpyfgA`OYrxKK82)#sQ7aqceO=eRyi76P;s_%#7{(RY9
zc*<>iL2zPpbaZ%+E1Vm=qVd-3CrCpYyaHWC#C0UJw=%$yv&R))xAlHC*O4MIzoVaQ
zBfQEkqu78R#O9@M*%z5oogiJ39jBZPC!Ie=T+%<5E+_;
z6~+;3rrDOn6f(YSNFLI@WEvi_{$#P-%6dU4`6z3fbkc7rf$d9|wy`T;9k>Lz^M0%x
zCqM)=t25U`W!QncldP2pxqk|Irc`jaY}vmZL{l19J+4{J*1WHTymE$SNW)
z_wIYT+D~1DT1Sn-Z@Nc3vjKUhp8sP$e2`j4q}lCbjNbm<>FUs$k9M
zjdFXz&DFh|L4z)rQ36Evg~pXOQ0tU%_A?msf&Ac|oT;&S{BxS@UcFls9AnRA-S7~cOW>h)~%A|)>pUT_o8uU
zkYc$uMN#7s7mP3{(6s<%}M{x*OvWrRfst<
zv4OZKM`dRY52~bAT0t@-<`*m|{h*5`xA`zj}JjS^J2;Oo1;!ik^
z#d>A=$%=_M`r42n`NmiLsjXrQ2lWbV8$j2K<+emAYjq8uYE1g;wwBXa@LoF1{sE-e@
zC>HHEUEv*ji$^W~GqnB%Ck>v7wBXmU$LnXiE=;q6^DUm-_&Y7N@A2yFwo{3Jc6Hg~(ymsmKRQ6xVECRTb4y
z!JuBx8{S~&SDWWCBP)>MGe0=mmurj4KFf3*&B>PTYVm0K!##CysEX?8Km?#^#V4Qa#$>swU01!PDx5
z$IDCJc`Knylif(Ip&6>mEVCZBY~pAMP6=)^7m4g&*t-QOf^Rl;ugHQzVIUKS)&RYZ
z8Y0>!2Y(3&Z`jr5(|LRDbc}gPpZMZCjm*xd-PkelWQV{J%w
zq3PngPhb2pHR_7#(DLj5dBq#qWYuAep(?95SYg*&U$(=g;EPv}jvF}1>y)q+6Ne~(
ztrDLy(?-5UICRo8&@UiYbj8Ux2IyKP*JS_j!+aZ?pEI(N;}LWcFRcMaoLgUYsLFzW
zT)XAM#mNpEugc(4fnu&i?ZnW7$WIbHMK+HAy5(_1;<I_9AgmFnfoy6t
zOx-UfSXoU52KF{LT|Xq*5xQikm=NeOQA7nbjvI{16M|gfIJi(XeN8{^O#4N^^2yKH
z8OQ_{%+y&h(YT(Pc^#QCdHS8YmL~n6Vvn0?DZUcXmE@sv1T^u2Ykx{PTlL#0*Ddsn
zh2SN4uw-naSNt3xxQggNOWxT2@qrjG*b(0KSlk?8)9q9%|1~{?SX_BE;c?rH^9VFP
zAPR!(s^A9sUj8f#umw`)Kb%~4gRg=!8zLO7In$tVblu5X@%#BK_4EnWHE9l(FZO+z
zV4mU*Th+=fB>8P7Ww5PkpLJ1e
zESjTL%?%aHKP5`d1mk5#CmShrcE%o}9NpiOtC%2apPi%1wUq5l>L-=km)jD_Jh2}A
zbpyYp$5UHMSj_464ZsoJqK;G=8Uyadeu#%En^MTKI>IaZeEw0H<6f&`5MD^(JI4^
z&r)qoBtOb_R+)H_@57yR%Q)pT|quDRSjqIELDI-B<%ZPD6ML
zP*{~C*n5jYTc;R(J|bxfO$F!JPHe*V1zlN*>79bs?>Y#DFZ6%66P`_5a)SBddO;X8
z)NIJm+Az21{Lj{s@-$>(mmp0;wxg(FTgj}oe$(zkjAH(F;^6#q8pge{w(yVwd~eW3kq3tIWI-5{?0T-&~jTl;qVYNe-Yswg$maDi~sciVnjJT**1~+f!(Ds
zclygOa$kKg5eW#a2mBmSEtTH(k<
zOL~}mE#CHX_=5Gq%@-3PHxl~oxRPQdbx&Fbi?-YR>Cgi&@zDf5VH*3E-~O}=k4en8
zI=I&hS>9tNXgwygLjU3LqE>Z*o^y8^hJ?pwnrGq8&4g^#AZe3c9jIEWKhabWvdm5z
zY@o(b2gHSPTfuqguw*|qTR%NfxPe`9{ph%+hBJP_Ijbc%5p{B#{%q#uR<~=`eC9FX
zwfz@RBTAuXbK}WcY?bGcXt+dz4P9$>i&nX
zs4a?RW_IZ3>F(3#oFPm>P6F*c(R&ye7&IwKQ6(4{*dM@42pJK02lI5VANa=QB&P19
zY-{4=YTy8dku`9#v$A!vGB>1f0XsOF+uD3*1ilz3%$%I;9J!g8tpD!-Mq39{rorUK
z4B#Rtc9I&7Ffgdt|Gi*g(lQ8OU}0dSM1@q`GEQFHGpOA)=OLX&Tu&X9&1Z=kiK?@F
zsPLhAexQ3tAs-uo-7W9j*R~wwqEKW;5NIA1ITZPQlMX(~UN^3#V;9(EOZG+HuxbaZ-@b=9W(a#R$LX-~N&>b0f8Kd~RnDm#6kIo~o@U!h}h
z$>R;9a}4}+Tcd>!Y6aeNh7!)b8?JCJx76_rcJ{Ba{O{LU20A$>|9g(9`u`nPzz>a<
zkZ5i^=KSwk&d}zsSQl-7p3XcH{@)cp9K(VC_ha~aS_Bt?Bg!o&su&eqfGZy_eJ?}(
z|KqS-{3^C|G*(&9LX3PMB!|%AZHw01H8zc4@%*T$c&s=oSe7`d#-#sZS6_~rpr=nW
zLLv3&8=g05#tMTbJ3Uz^j!H#MCB^=_{%*X&I~Ry^r=N`C(p$1!Gs$30V1coUOM(7CJf4aLi&;?UJRceVbK^bgD!95a6^~i!O`$
zEtgiAzZ&Ip3OIzNs(;iOV3Bh@=C@Cl@S~jlFZpa-TEp`sRp{6<x>W_NO{Kc7%eB+FEE>P627QqKHb}?9uBn8!X
z>0>|Ib!pST68R?oW}kE})3sHoo$#y3LQwYaol3Wz-Oteybofj*L+!eDw$7bO_v>?aD*u;$*OhYhTeN6!<*J>NW@XW
zbC#>2)~C$@foi7*f_cVl3ICfxx=Gu`SffC*TmWBE*Z9UG)1%#Hhv91yt$(cnPOwFN
z3v7l@XGnfN_H}6IvAqj)6|9rvg^|i|{8n7@YXoeoT5VZ$q;>aw-hvw$|GmkEQ!|M=
z1K!aM_pAjptbCNrl&2GgrDg@{f3P%S`roN3gOPR1_q7s>E_1yo$2dz8Rxn*S42v!&jqN+dSVZVI%=NLfs_ozRr+RM&s`0cf^
zBU-|7#J1Q>Jbfi=1ec3z$j-@0yjop63EZa3F=ZAb_!lx=yw3yH5Leu(7+9=tTm`lh
zi*VM^|^hWN64wuD)*D)e|w1b$p}i6?bv1V<#TC;%4YUBm!oAQ{=ofP^t-g=*Ysim
zzw>($58Ycz|L6)9rTMiCAb)i9No}YQ1rH34A)DgA7W2zyd7f^XGQ4(v`lP2m;C7tx
zybE5_v#dlA22CLNZoF95OJm>y
z&n$W01=riCh+R?E-tmrCTeJm%BR`YoMEd7*rZ>DrZU4^a{Z|5SzOd5F86N$}{`kMe
z1b+`Nn(Uf03KZJ&`xb$GPb~-5P_zKV5>EMPM86v3)-HE+gZSWe2p7|4k#c`>V}d`Q
zLU1@>Q@eHN88H#y!}OpQg7)`fsJXh$*R7l5b2>L>%~DuRN2x-le2yZ1$k8Y3
zk8Y(=qZUfYOJI*rn=ZafE3OzO-?#}quWiWDjZM+(I}@GgOKN5f#KvWy0@|7jmxvQ@
ziQQ52DC^&kjme(AVt_$!johAg+_p7MCE5Lvc=i^X4F-FMyg?_t
z(D*B#`bZ|VtR9`G*C@WsPi?iO{mxj1=BboSnjPSjRSZ0=)dOjZrQWI!zhSo!1Ab^;
z`dDeno{lROr>-;an85QSp;F>39^!Fhgn1x%VJ69`n+pFa`eAY2^d9myaCg4wS!;WE
zp&NwZq1rk~Xce&c;2zb}qs)@ZJ7$7iI4wtoMS*jrk6pJ<@WwS>!$3HE5EK8x<^H%JrHjM#F#lkf4`&X5&ZGw#>27&W;
zrpb0(P)<@0W~pbo>`H9I0ZO6xh$Y{bpy&&<^-BzQ)2Iaaq4dvf{XdKyyUboYV7VzuVi9S5-h
zle)GB>oUSiQ>vQYd
zp>TF7_UBDkn@llZy{yjfY6optG8xvib+ra9WAa;|K+COzCh)1^)4P*PE0U|V^vR{-
z%$s6(W?8J2jUsCENP&k9HEXaxl1B^}Xbla3J>7~la6(G5jVEkd?I^LM6s%<-0^?H|
zlKo^$zRTo&b?4w~vKf>)0O^jyhyP*Xp{AvDdd=E1HAU=yMJdHKX~%87VD)2u+baxc
zkhbylvjgD8lg0Ky_Fl-Sg=#E#sFxPU+Pz(b9iLn#mS*Rw#rBXLorYTc5U{H!`7;6@Fs#loUkW1SYxaT_n{*tl
z^C;>f(H+R+KHy8!qQOKpYIFsOB1ur1xr`4$PuvQp?EKy?1$+|)?tXScqJ1wN&r*?I
z^S&e{FHTeKmZ_oYRxVS0nP04M{{l2sXe|IHf(ksN)vVaxwXTLQTNkP`=p8I}K~US5
zRiP%zpR<{CTGoai(FY3mWg@QOs~UZ+p34`~uFy8BoO#8>6gheMjXWF>O8dIIriw2b
zKl{fps`a&#;lltYJdgUmfYx8z1$uASHXXCq)9dk**W6r8XTUT+hJhAcv)Cr3#7Kk&
zl%A|lU*>1GyjlVLY8zYc7@K3zsFm$0+T$!+W}$~TXsxBWnx|dN(pWcM>PjQ?nIbA`
zA_6=`?A%H+W8q9T%E-*d9N3yTKjYcUBk1Xp#Fo_Oc>A#8^>`J1kLn9;eTaVB_9eu!
zu&tD0>Miw-eb_OXF6(pm@&X#XkSgx8&sm}$wuO4R58v6Kb}C9X
z=Cuk%8j0&&^M^_MMWmSiL#2H7e&t2uwOjA)$!l+I6kEI3HM%{N;TM18zqWi6<$Udm
zxG_9Ve*W$I`wWNX@RE}I7hNC~`LmHidTf46;fO_Rw~*~a{sVu4Kq{h#poR3v)oXSJ
zQR%2=Kb_B)~1^
zI)9NendB9i$TcS-e@K+CYIgj4y@Txgdfbubw{AI@r@>G%g{E9QQ#79)t$C9Na|*5z
z&zCAMfCtmWb5QMCvD1@9j}*1IQtxI>w22I&G(e@4=(K!nreW0UT<vn7bz5e*7KXG=P>GA^}LQVuqIRsCY*?(^Hf(xyD=s>^n%QG6%OMlJ8
zF-y>t#DFPKp=%qOA=3r}YjHN%JW9RiyqM7h?%Jw~5l0X1>=P5u>u(r(Z^>`yufm%CC-t
zk0#ODE3Ioj@Mi}YR!5W^srKWuzc+J`*}LcKSXL~(6Oh#E1X`#;)}K8wCexc5QVpoZ&I}OZrDef~@yd9+aIyUjj`_@_c^x13d=L^5`@=km{IV1L#_d_q6t_t$oGvCRL
z$7)O>JaWhjYkC@NMx=Pa53vc`Ig6V@m_n=Pw|0b#)yb#uxv5AM1=S#+j`lro06BQo
zGNCo=lLkE@RRkr0a7Gx)Z8y!NiuugtgXaPCXPMrbFbs|-5Esc2XwJy=rZzM%T0E?}
zSPI|@|HIN)D9(l3Ca0JxX;pmPqSNmn^ATNq`aW6UBKscO_i*{%_3aR!b0SpvGtaGY
z)8brw0ytFxWEa*lF$)z7%04r`dH&*=uZ~mC>FiA3xcq=9)of~Rf0W&xvG%MA^!Z>8DHN{GoCTJklEGe8UQ6kaX-taNd!9J|KPzo!b7I=9hlot_@8
zr`lyL7oytooQDv&ikuay6H`zR@uE!~*6|sQa6Gl!eyGo8l|7D#c1x;W4gwg@q9gA79~yGtIVv8iz}2svb2NTB`7wOoj3cW_F!RCx;*lo(b3p
zX|l=LDCH#po$c`aU`w%?5(d(~4{!m{c6k2h5%T@66Ur&cn98RmfnNM}k7L4jndmx*
z*Xs)ws@0n1ZtXTWV(e4qSyn`Dv1`kaYmlA_(s=FogctNW*}{{ds)P<`+0)!D&U;uee*x&zm$&7#1WL9P*SoyA446M|hy|lGF53se#rDs_|GH&9;COw$48Fkc
zto&T4X3ILX_dWvSj=8g~E0@CU6B%S+7M3qv6>RPdIUK^oOX85}CZaS8pU7B7E~b9p
z_rB1S&_+J8+SQJs2-CdaVarfeXEa(9K{I9mx^>6@7P|iQ{F(oJ7^hNheIwedFFB3w
z#E$t}eRHa{MKXhQznNCyv;hv=T6lL1`bVapxUbv;Gf4^9dfa#rU6RWNhc$ly=t@{s
zwj*gWJ%X1_{($!S&%(&dlep&svssHs@+a4~NZ%_*fr;$<
z&wAtd?Pqvro!_bL5VLLe!aV>Qs(NurH(m}29;J0vyk66Lx^o^<>e1QswC5fzMb$Xs
z5T*bUtX5~=*})MjP%98Al
zE{oIt-CnKHB+wdb*CFU)lm9ka?n~mZ$(h`sRQPuaKML~!VftS1KlSLysD*QU_IoCz9>pXf+lGF%rROLfGEz$AK1Ec
zVSFE>Up{r*mUc$uREqd?hSzH$F27@gi0#`f{YTo(Q|DV(nlRDrcrChRPNHnmCzfEw
z9-5=S{Mo)b{M+2p(r^2^XUoFo49U&Cpg-5IG^uRACGtJTcoo{)+l!b8@1tCQ$XpMj
zl6t;mdSiLp(tAer*`oE?DtN>2UHH$3@MMIxQnXJ^=v%V
zLJ;7{fC^(D2{e8?Q1+hr95HtjDOt7>-A_r{e3^(3DZ=|IlqPMGyf(C^zY6I+GV_qM
z8IHBPw&TC^Kcqf+-ORs{KRZCmi@b>zk;}^YydG^bbeP~;F=V7#76n~1^Mn|6&DE{F
z-fty&1jl{(21Xi5RLXK!CvEU)^&Xi}Jd~G!N*QMUe-qFDxPAQh38RA}>AV?8Slgr_
zB`$T$Y-6keZ92Jny31S1KG4R8F7hW0@sP*+w901|-OMcg946;K>$g&8P${y!to&MC
z^_7oT?c6cESH~4*Of`Ql$H-+(_2kzzYKt+E1Mib;AJnL%_I7q6JnwY#r`#;-wMu8{
z$Y0Mx{)<$uT}
zzP-JDLi-K%Up@{hJ9~Ox3am1NZscUSvSZf7m8I`GFkErauh46@dgxYin6m?9wnqwc
zW8Ivl2<1-@%R=_f#(9F>-+``NbP}@U`0Z)AuHUP1O%2V`e|h?6?mvFNqNf4p*bS}^
zxmu)%i|#a)R
zOSNi|+tI6yEVmUzveSkeOuL0?gb=j6%PM(RLq(XDzU&$D9If~0{0jAb&R(!q^LfMX
z96nGWKWpV1()V5ZxZZ@G6559bBI4q12P;$-J$aGkm%RSg{i_OvzHeufHF7pepFf?b
zX{hcS;D7zX(#$)pZ1lV+a3it4!`yy@ldDwREAS}s7Ha$4Z)?I+0X;v3h@2I(Hb2)7HlfRJ0_GFiSPtRWO6%{;+<&iU
zkfrIhis0%8_Ge4?^=)(jgq6uihURj2`dKEbRY4
zw6JjgH|WWO7F8zu<;3)1P=ddEY6?@}bz5MZ!)Kf0l63uol;P{w>h|(z3r0MMP7Y2v
zT3kuVhQng9ibZfvxATcRuR!JejS8H*rS;kg@eSn<j^FUA1i2LefhqgdQa)E9%TVK~3k|8i+6Q
zZG+8^1rIIWS~g#%Ki{gp#akGboX0cd_%)JFC(0prTEWpAPX@f&+a6$i)NWYRsP(X*g
z^_X=~ggGEwR(2M{gp-m*!iI)Yi)vs86kETVuc5Fl3TOGv;KzUVB1bSSA#DOgJfmEL)t&tnwDe8MMTLcarX{Gu2
zz=o!f8!Z3uvw4+MCAcn}YlR2WjNGyqo!PUUOcI;`WA
z9aDA|ejzF}G}?OFHquA{_rI{j^$i7X{$2E>#j%?OZ09@-5T)+n7Bcb7j4Vg55>%u@
zY(%*)Km_tM?gWdmn^O
z(a>(?mpNCTM76LuoeY*@;crw|L45khYd#Ms5=P07+b=*#GBcRjSE5%9j9h%)P3ol?-e@5xTzK`dhAqG|@!uPA+Twdm
z_8qlN?E711q9|0T+&_D-K4lV+8r#_8`c38X<7t)u33xGAn`?{#hsAdB1C)OZIMs=z
zf31sIvsQ|V#Msj{`$zo`Djeg=8#w9S&;2#8H9X7}(&hNf
z2ptP-&(ooTxR9Q=9?4QmaZ~7lb(;#zfke4_%zuY9^>H(2qKFC}Bfar^fEf~fzAssl-+6?g_p4D8FuM`58q
z%52``3Tj}POKl;kY09to1v1$JU!HxvN|Sr@04pmK9s#
zT2vo4l`02J3By#(D~>hf)CQuv~>KOJ60P%
zXoC7PN4xDocof>|1BucscRs0$6>@N!38gG%>d5&>{{Z-V~i~sEgxOKeBE#
zYGDtfLNKdx1#5Qa!WFiNXPw1&=JG^(-
zEB$k|93}>`7sx9F`Bb{`ZsX%PT=#U^%s%aYXkY&>xHuP?0DeKTbrq@mbD9mFuQW#G
zI9CSnq?x5>+3w{{Ac2Xd0qp_Zo@KfxuzW}EG_X#?!co)nTc2%g$SGirM7;agZ!x$x
zOH%NiQE;!6nq~fI{>S834p?6@V#v$M9@ooyedM#D1?DTz&ytli687Nqpa7rzCJO1e
zunwyCa1gR8W;eJyp$#e8?81fTKWn-29zg!
zhknWf(;&DBu^Eac-C3=;-^uI}4_Pq7a$VwgM?kLvoDXSV73MEMK?-N@-|xu79V_2G
zCsgTPt5<4OXQv*+{OQR>^#K-eYkZ8EHTS#Zbxhi?IO)l=?XGQbf5_9JG_4Fq7(qWo
zH*VUzP@V30SXiX}!)*Z!DNLV&8dWlccjDH$NX;_ghpNK^oU2XLbh;eDF%`CAk0*zF
zW}1mbG|v8gO7C|f=gS~C400w#hJ<2O%Lw#aER=UZ9oHJ5#{&ReP0>bHgzI0{)KDS=
zH!2bqzy5uDM~@2fkRAgLzga`Y$hJHNDVmA)Obi4BCe7~-!Df5x|6XVdG<56QXkqTg
zPhW$MPfnfZ7IDM5q2Tnf@M(kn+cA3&Wt3n0qSV{(8Z@m
z&v*m`zAR0)B3lsO!=YrO+-kNWyN$4k*5IZJ>fXs(#g)2K|LlFs1m*s1?oWe%g
zW1CG+>#c`2KI2n6SwPUnoqNPYkC4EAs?{-f@Zc`QjnAf_u%CeRw6K#y3UshNbwYVZ
zK5kU2X}`ckM*HvGuDU6JX41u>7rbWFvh7wss(y?V$j{gq@=~$g@iF@XnqZ-SdNer@raIZ3z;N%QJug=z
z@hA>Ox?oSxhcaX-qmB_{RBqz1fcwUxGsnR57l|+u%RM332NE?
zL2t839#UR4-~A^y@!iS_3bfMy2x>z`MlVS*&3O5b6Ud0q;{J4*5j6`+{_@m@h+;!$
zR|G1Q>0=f?tBYXMHgnctE)APUdltn6F}jX#
zS}J(xqyQvtw_tVOX2Z(bAN-`CiVJw(fC5e>sHk$flJSisfTx_ya@8{v=6b?bHtL@h
zJHV3x77D}WZg(2#W~EsOPn0HPU*=DSA&Ky+QG{!
zIB)Rlt^>VIW9iL@qaRXgq`5)kE15`)R07nzp9gsF
zL7R4eVTOTBdz`DNUUxCrx@Lfs(S5l6bMB@cyhD`okWcw_SQHPxNA}Y*+M~B0dO|J`
z$0$iEGZ-)noF$wnx6u-!pLWZhNo@Y(4=^c$nLYT)x}33l~~=
z)U*Ld5&vV&X(aavdz7XMd0~%@-|zAFob0#LucUm8$rtVXmCNer@18R<4)?KeTikn3
zv;AM|rebHzQU<_9CBokhdQ57!E;uW{F>d$m(K4kaD;KNW`;*_p8deh=A0K;7b9iIJ
z!{aNbi+UD2CQx!HB~hQa&;w+*{{Y2;PPT)~id&hMS0Oa1HsRS_m3Qc%*=#8eFj{r)A}^)}sXDg+>T
z)HRk(EKReCjYL*jm<_Q!J~|drriABK9i)cqa2j@un|E?@fGmu|Rqd+}sB|Zq)PILb
z1ntVz^Clj6W&P+6oZuOevd@&}m#5)=V#W959{7ms&}vvX-g~es(hOK_;;8I>vUbnU
z6Rd78?Fnsz(%e0DqmTy$o@xc8DPuP(pA$Q~IlOu}&UMyrbbc<)OQR-`u1bEOK5#kR
zrgQO*R{B2Y%Y(8Lp0{zs4OlqElgCuMA4{{=^PABXZf~eQP;>~i22R^>grje{h6@4>
z{6AL_;3rvW=CYt-Nb)aIg8qT7MHU3q#;r>{&
z;Cp<3)sqh`C)SJwfxYX-SejU`<&FA>~v1wh`dx#o{28D1TOT~lAe)Kl(OJG##GG1yBJ?Ivy82uRSi3a`YIB88{p18tcLY4I{e
z&s4S9acu*~3F7Z8ESzy3hk^!0t_O)^?H{(BdslqQUF?M
z7p;R^rCp#kOaX4nBDG`n@}$NDqd(?EYNfNHM?Q0pKMVQWm}Pu;X2zux!;9PHKhzt=
zQani%TG^0)e19(klfVJ1zkdbP`;%@uvPMW)tdEfsd`hBA+pl0S9hrdvIAET992@0H
zs*f@w%M?(k3i4v^6~cK%b=)Q1vK*SbFPHEF+P`?|V%(YH`?-whBjxlO#v=bPY+@L{
zv$51YB_+W|K~h)Olu*S*d{Vyty|sfXW~^b7t9t*VFW>6)sPC&{;=pH>9qTzcRqZ5j
z!xH==t7AyyLeOWI+0bH`>RCmE_3hg68crk0(Wh`yg)N^bImQATXCLgy7d4w{gGPf!
zjb=y3UsjA0k8_NDhJ<2@1caw`_59}Zxrkhua{Sl1Oq&1&kD+b;vJcGD#&K~Y`&WU&
zxrJ7_%hKD5E)KDKEBdWPpMv{;Rv)`1gi|&KlCDK5JRKUmI7dC1A9IngmuuW|e@MGsCZV^&9QVVBGmH^N{4ybRLxwRbe54M%yT6&PgM`oaNnb7Ywe9
zD4CAQl125E=Iv`aYSTRaNopJb$hf}y)5OOnF3)Z9`@PJH{HGU++mOOQ_7OJ*t4SZq
z&73K3Dm>d2VS6!#znmxRJx2A5W&Jg?7R)g77WGxMb%YIC=zWNAKnwWb;Mkhce{%uvja;ZM{1OenQbx-$D}8&OHGOkICN8
z7%K1Xys4)bQBc|lN|p(|YlE+waJ#OlaaBl176QQH%a<`}kftthm|qPAG?s
zJ9r9A*~MI7@&#sPk$|m!lgnDxv~)-nEA0VO3XdmJZy;s-Lad|F$>GD0Yns*TP8;B3
z8`XP;_{OfT>i9el_-6RNoxSm`-+8aQ{_$=H?557*th{>mdxylKo0a^9JKV|J9#jsc
zvw)m6kwN%sBBN-$ViO|N-N}*fc#j?>FV)*46NN^U*OoW0j)gobiB+!|a1qN7SOK~C
zTREMSsa{Ca$8d{ON>&I+_0Kk%{-mauXV#m@m?PBU?>j9$JoD+)-9iNHh@Gt=FFn|C
z7voJR?NtuUIblI0Qb}$L-18VKnT8$?34iKkg@v-GrsBgTBC8RLMMLfmT)4O$-KdJ=
zJrSm>A)8iDLGcs#oO}~3>uFY8Z2U6KO4K;#n|2ET?Jbf7)|TS|7{Qw6U(IfxRYqh-
z@(KzRBCt>{4ZFli<_$D&Jq~1P?YKw-;e6W2@MkU1q6hl+ZW%)c3V+8jpYmnEZ=9B+
z1lWJVa}Z5sW8U{PvPvmjHx@zI(_UX6teHPfF9uWX21E{41&i^$6N5=@m>QOI9D^eN?
zE1T?RKw#^v>q64Yq;%{QEY}@?IuMZZ!I%nP0JB81KyB!>fhiCc`D}g!ik&-qhc#G&
z>d8;TA&Dcn=+W}-H@9dis_cF&nW2OMW7=}9#c*FwM}EJkgIIVve*vzJt0qF?Q&V&2
zo)siAbOOQD2nluVIaq)VA($8b5LB)Pxmb-u;Atg2$4hjJl6R&M;2No_bsaRZTpL&L
zO`A=E^P1f6+IHEpZ>=h#`)}d^y_XAFrHJLpbO9_KhwIkWcAgf%3Z6f;wVjlL4@`o5!zb>{2KvTNx)_3t=(9|pWAr$Xmbmi@U+ejP7`{(Cy0EBNCGAn!1*-_5CiW!?($FLu1<{LEi$Jes07Js+>1?`(Yy&a}#w>qq;-QT^A^{nNhT1DlAZ
zYydO~C%xjOm}$8TFZhuuNzHQE69u$BM*?nD&)=uMiOm2WIru?Qtar|_f6)$tbANKn
zLM%gpMn-dd~SD0k;m>Wu#P%#j!xmgC6_Oz3MW5mkbH7cH45
zsNFS-+F8(55p;BUnZRvbAv_ty(O-lQzZef2T3tm4&)Axh^0;MLNmy^VArIZPGEwtn
zqp?FfKray|lc!36lyDpxtAKVmBfn|F!(RjF8sP0UmRCNJyx(x*Bsh^L4rpDXaoG$t
zBH|V1B>vzeyy@;kGJd2~Y)NMc6e5uO`~eA$Hx$shHvnIi`uf`T4sTC6aSsi8
z4yOkt6^?|Gz0esm9y3S+T>{AtkWGLMsi8LCc;TU4{>dT$P{Xkk2WS3)q8f1bP`P?j
zIp9*4naqsCznh5e=83A5#Xxk}_T-52CF8&0+eZA7rL!T?i>_M9mLod8zNM0RYBST|
z%?kV*fD7U9ST3-c7?$t8DOsC$4qUVG9P$6hn%S4eh+0QE?Kw*H>$KGl2m`2G@(<-0U(3tOYD#wEq
z=S;?jr?!=sufiwIHGou07J|m{J2wvhU^8PIxLe-&)kg4lE*mNy+VWME{5AN1625t;
z5+Xib&<2n?WaD3J>pcf&dXakq_>$@9!V*nl>$32)?hG*lc62H^?pNA1$e7ChnXy-^
zGqf1Ga@*t)gnfX;lm(v9isS-nAp%7h4+?Jgf;E}$I%l-%LkvSUTvsa~(&)8+#raSA)Kr_0V)v4)2-@1m^ZMf~docE7SRpgzCQuY}{rczb!y@dr1qnZ^m=P6ck
z9c{$ZN&(GpNyz}*)a0=ap_B|WSbN<>IbQZ$$(W>1%`*jEzI1QG0M68_5n`26)p%OU
zONIb>Ka(fe%uGE4)3=yqj)AIWYvaHvY|823uXU+m05hnb610nEvX_MN$oHS
zbuU%H8ef)SCv>xhK#j%_6VMBy@qYsAQf6(FU-X}|HHCms-@-6u)1U0~0LX#0PZ+(e
zI%b+*_Cg&4fxRnsV<>#X^0a}g({6F5>od9o51z?Bazde8V~lvY?tgN?nycx1F-3bW
zl9My?i-)K0Kydc28G2kSy@8%GitUl6g9LJBaB~+qYpgpe2fepxd=U<-Ifj{5;Qsq|
zlP%D*sS##-pU)w?IT}x+R!@q)%+?BDaV27o=yu4OkB>hmU=BTUZ_T{YxP9#T-$5O4
z&eLLDPN#>bzGtE|yZ*CPW^;OO<`Xn@0jo@$1LDEMK;p7_kyXUe_53z5D9E&^y%w_r
z_q~<%;I4YLM5a(#k-hN_z!!p`jlQ$1oYOA=)19x0s$lChu1r!-gogAX#=bcNpkkvL
zqyP;|i7{ErFR!}(GkR}Jlcl6ru~hd76!M-6KsMqc+f&=_FafyAof2Mi7X*vhpKv(F0Sm^r|*<<6C3=l{BuB{n}
z9~;>y2gDKWcrsbj(BSx)W;G*saWg3s<@zJ(?(}`r5}kE^RM2MX|LGISBq8sJn=7S~
zn&0}+S!P7B@%0u2*x`xAS?i!JTmu*~f9~towKhZEj`s(5n8<{lj8&f=Kb<9FAZa8;
z>Ry|=E@~%VPHX%u%Mw6|!-uuYpW0cIK>{sfyi5&o0Z%gOwQQq=erg_=zNmvZ$NHr8gThTi=YQfN
zcrQOp6%+tG2Vh60IJmofZ`Xg7}1niv-
zNW?Fx?sz}vw7To@{)G{EA5sA(oF+%dj7@)C@!&_v{`}uc%-bAtv^<>yU@VJ6m9}Yv
z&n}zD*>?s1xyymFYZHOS%*STpv=R~hAtqG(K)Mfb3>X=hM*>s9i-coYnQa=uyS%`(
zlq(#F35P8&j{wS9`Q+2reZ`I<)SIsKFFuFy(!A;``{d8*j<0*!FMyCDV8`tLT+UiJ
z9hEnjFj7xH6;UDm0PEdhpB|b$!l(0Fy>7SuYy3QwoazT8@0Dx~V-Jgj`Ww=C{os^=
z@ZbQzJve&tWu+Z3#Vk^9+Zw*go=}8|{Q@c@6=C#-GuJ!eR*gX+ge;d&TlYI8o%14`
z1QaBfwSd(9rNC^jEk}|O-_gpt@@j`cgbs6e;~=4H4X4^HbTJ5A(+pSaHgjIoo`uc^
zJg&>`dK=*#Rvu-I#g!t#9vW!n_(>vPv%I?9RgiOM--jN};c2g_3qDd@*etDJ;{L)D
zu&n@(a#Ma12xkeU909oXe4hUO8~sL2^SsSF_ZaFh&2~A^#y)GzCp`uL2dC(1Z||^*
zaBggOMVvmYwLAA5kYJTA7HmZ>ePId9%5(GXtHvSsT4#-wJV!Gk&#FtVo_}(Ok{;i5
zR7i0B!=@r~QmkON<0*R%@V!Fezp*7g%8)hykk?_H>3(ukeuYrk8&}$YIKwl~vr<(olX-)tS<&gVxVbDfx
zSZDt|+C$&{5d<(#3w8qD?Na%Y)VFGP$T)?PgSOYnkO9V#y<71MPSKl8H;r#^6?R&iOyHnfLWPG|?yzxdRKB)s&^uTjU0JQ+v
zVsaJe92{Ka{wT4vSM`a*+s{QEe1Tlra0x|M^A+H2h`LFI{*dPzLMSZU2mX;jnHv3O
zRkqzi^uTtvNiEUDiYFNEa~s4DYeU-R4(?G_tDNnAqLX`M;`N`%_{3Fm;GHca2PmyC
zPdR;fH*F}rDNTp(+U71pPb~j3&+@)_g3r+M$xn+T_)wlU@P9};tFWrNuZs)3g0zy-
z5_0I44yBRq6r{VmQM$XkyF|J~8YBoG)u$Bgr>sVwmixl}rVdLRlPDs?J$SE$|$2Qo{DGnj%kR`|`oPXLu+=
z9lr@=;)@I7$RN#j9o6aL8SE9=>!j>nW99@HTrG1uG(1nz>d3UFgiXR!Opv{|o&w@md^o2>iFan)W^PS`
zZm{0Dc$v56$yqADB`HJh#$39;5AUsouehd%{|k#?NShvB{Q1j_L-O5P@19f2Pw`c`
zhtlkO+UIpZVm}Te-g|kT#6(tp0jgw-7CbfJFj&Wv+3X3yt>@T-(yVd%nx~NB?n_2EHufbwc5=N4
z3NwDKRzWcJf0t+~O~o{ZV6AT$}sHJVK9pg2NI*%~V$QPtT_f^{rNy
zcwJmO$ko*n?q@@QP(9-=RlACL9w|T-{JU0Ld+5S5w={36Z;YJ`wb^gjJ&}6*H80n{
zcVUOL&|!LNb}2oIK*~ci4$Y2Hy$qNjS^39Bz{}?D0baIs#qT^n`FTcaJnjDcBbjV)
z>~-bnNIx{FVV6RV|NQCcppOcV4Gd6Q6=E2ZO|%K
zqd=2h)+JvtGX3dxqfz+5^Zf2SF}Uc9Q=9d}VLOjuGuJY5<~63g=IR31z!rsM;fB75
zUabN{ZGAqB3$*j4nN?hwyKa$beKVtQ!^8K}5D>CLYj{A|6^GD@4DsbgbaJb#N
zr3&W`9F;0AF(KiR6Jv|dre_m1UwZ70ah1F8gw0q;Y6f2|lG&
zyc~ZITUkj^q40x!v(&}iERK~eBr<;34BAjTEIFajPvzqbCq;7mR{Xb0uZ_8ky
zgq+`Z_g6Gd#l2|Iig*z&)~+NWx9F7PwFxuufg03aWUk)bg!W^5ZBhP>7rH9$*a;jP
zwr^eb>UZ@*TD`Aa9a?>A^-Xr;m$S7!slPspLb+)Ws4Q0P6NPUsqq}_Xb=lVrIxj_|
z(9q}Caa6Ug8N~`ST|2I1R<#|jeT|Rvm~4{|hwg9lcrrOTj`kK(G7URts?ed1To@u(
z5|3H1$|?LXV{9>z|3aPUhOVQk;GH}zrc@%)kq{0KUHjk<56Nold&+7k8e~z~dW%egjfaKotg(Z%XNy8*VfMvMlXsv}o42rlzsa
zUcneoiXgbA#h?qD--o+evIo}Z>(%E)^_Bj`U-@ms3JX*x!&Tey2u(O*
z;&D_RZoDHq&xRaSJ2Ucc4^R#2X}~Xd)@*Hbj(ez<*!e)&Im-1|yXe*XfRyE)5k`?u
z;}9P1zFc+)k|r=PA=@S22&T_cYo@H
zOL%IR5*cXFsyu5(imqL}{2r-Mi`ONKV;3UM`*A)#tgfRLVV9DItr=b}I1{J~pg-Rf
z^O|o8UU-j9MDpD%kcRFJ%02qmFbNUj>@3A4A&?EHleAGytK8
zlDw>*x2)`IomMFqAx&6B1!yjCy@h_rlmNvAx
zb>eC-pYIh~I>FVxLq3#eUcQJQ}@J^0Vphx9F`^tf#s$Z$`!!@W;xV0r308aunU~e6uCNP&Nw~}6RlWK
z0*V?$1i|ybye{TUP^v%wfiHAO{r8EQpCwZqz;FI*mzCNtEr|urL|xO5ZxsXJ57y46
z&-oK;6H8c_Cipb!7GP@Ee^9Bf$L=W2pQZOH*0?*I6W1z0(4}^%v`$be&5qWKv{
zc8h&`+lupH<(7iaUa}8ucXxLX9}>f4NMlYKCVx={Ab&&iVO1+b2<*k`*c_R;_x1?t3Pb
zQnZIzvRtJck)lz2qp?=3|tsGB=)
zl_s7ZhUwY9^40eyW-*?UKSP_-o?NVU(TL&Cyn7I6#P6>KLGuM1Ab4t=G6?;m^3eA*
z4``R~L5H)xlK;{m_C=^!f_PmnD%&eD^4dko|5$6^VWHj5EiA(h`K1|@jGVpG{l>a)
z2i^hcymuQNxEI#Z4^-`+_jJSi9@J&kS&wjRC-obFLiOn{wo^g-3XWd
zkBzmM-iH2Ic`2Hj4h?y>+=?Q{6-9+$J&@c*b-9JLPxZ*yLQe%}yLo3TB|YA?9fbgK<$UJcPAOc?9e(v4z31Z9O>k4kSX6#v
z7zP%mD&GDF&3>QM`arL5^hyy`j>Ihz9jK({gl1y2cD<5(pGVy?QFmwzKZLJ;!lJN`e@Tb3
zRmq5Ecb}I|tL8ddIz{*&zsu+kTn~f$ZWUY3J^5QSt-hU4h^n#RdH$&+xw!E;_G|41x
zx^TC5)Cc!(8~pg-qXL@C2EKRiuGz}Nh5+pWlt>_3Qp+nQz$+(U?0z8RJ$tyGqInXK
zg7Vwe=8F>>yeC-XES*iaIwECZo|
zW-)-E!>(A|4MnpJvm#h|r+h0VVL!rdlpC5xhiqHhH+epXi#0o}$hDm^1t+TaVOx=G
zN%pr%pH8J&z#`ezb
z>+UB3VYiA@P%Ao}wg}Zp+VbjOCbni`#7oe5$OB63pTkqL=g7qYw(}OW|DqgL0o8H;
zrBh7s(+AD+DvU8!DV0x$2Ee!d9ihCf
zMYgvzNU&r1Xv-T#G3*wI(mV$t!(AX{COUif-AcY?v;Ddev9^BJT+bEi&;k{lPY)xq
z&JYlGo%*P!HD}tA&q)>K6XKdAc5IQZ-~-$FuYE5-vC?*m@Za|qIHsV=n>zhfT4z3u
zb9^(uU$yiXmhC47dkY{Nb)HlSzC7m8-np#^VHj?)`>h|9GdrqT6#=bcV}0FcVD6s$
zKwzYqr&4<1$DR>v&tS7G`#X%&J+@8_oUWidcnR(B9a3;5WGLg?t{6!
zi4HnNrbGZ^b<0W%n1BE9Tm_!UkR^ELD1(?$V=qD_F3R0>`s?5U|nZ0>w_XE9o3Z*JLr?&}XE9oy~
z4S-?@*lcB3sQb8j@ULSR9zF{EIL)sj8SCCzgS8+Mt-C**m^ibhxji~>pmL(tP)0s-
z!)_rGK8j)sO-z+EmVv23B&^`pPr>X>tNY`|kCQm#Toy{a0Z}p6+BxGoE#%ww2#lM}
z8+QQP<~KsV6@38n$DNgYp3%`daSPEcqm{Ha4<9T5^`q7TkG#pD-Up#uqukAFNwnd+FLMUoV#pG1z9GWeQ!VKW^9_
z2u#}c<>P3-$%dS`;~qGn?P1www+hJ@DjA=g7EJ3c@{~Mhjr!4YL>AUt%gM2Wu^^{_
z#;49%l^mYo9WV{tzMguem@B}7bDHM&dLiy#e{Vo4yR&CTwSA-Y)aOuC&Qs|EaC9#IK(xBsxF)YlcqTip)6HIMkuOtC-B@Y)@j*
zYdHJq-C52B`syP?odynw2;FwLYHqQLVE6jGyWHUtD69!VTdaH>Jz3^@t{MQRHW`hn
zaCh&rz|hM2fOHCoc!HPvvY^xivXv3VZ>jRvpe&u6TM>b-Js)CK%gK+g#((v;j2311
zZI4PUV3Ume6GQ+3E?{Cco#UIGE#0VBI1PAhWrui>B#Hr_1^xkeU=)a1KoymxI(1?*
zp$UQ*zH!lBi7KP9LZ~<@{j5}m;TeQ9eD3j!k_k*_bkx(nn
z+WCrzXbb?r#u|loJ^y?Yh<VWP#$
z&e?8v&f%ejqAcI3OG8E+>4CU_NxG!@3N=n`9{K8iW1PI6`Hel5B^ou^2=p)EW*D;w
zL->rrw`y~XC<*dd1IF`1&K?`gEjeT?a+_P
zyL;zt)5SNfvAVDhSn1iL-+Oj%F`G(uu)~AUat@{XH7#RjytATyM^SaNe0oQ+in-evPPc{~W
z)L)JH4Pu3cpJ5q&jXt~WyWZY7J#
zGM%q-`#)*=`?H$@tUsg7M6gN3Av>0@FSeb#ZQM@W6Kth-z1_jMD-Tz`!_7lWj8H6Y
zSN^tSCjIjLV}{9a6R}!RQ2x|DubXwl*1U;&RMP|*HJ;OYO?}(2Nc;OAYPJr(`q{RB
zbE8@790uzC9L*Hw|CrxkFA-7~P`?;h1IDSV4w-LnC1o6
z9u4(BtD)0dKzrR{Wp9SZG{pz$qo7S?N5Q|Z-_?0v+s;Z$vsVyrkMMe#2_IayO*dv)
zP^cpyEvi?t3aTOclg@uodzourUa+y?4sHuJ#{7Pr;3k^*E;hkd=>-;|Q)qXKsBzA)
zAbvPUo`=F^D241|Pew5}qWYt34>q0@ta>GNtHtH1>}J^7wi~B!xO`#OwHX))g-$<{
zJwRR_;w3%@M0?Oj*LfxwyIj3iO+0I=sN8S#VV@-=ES|?Iqo9;hV3^&rj7&1KsBz7J
zk^P;ry?~N6yXMXqYsdVYFl$g5$o5v(*;ZjzXYtr%ecU;rNdfg&msiC!~hGu6kh$NL8L;=Ws~s4v*AHG`Tdi(85nO!XU^MrV%UK&?1JNgfbf
zSq(9#!TyjgE9m-{Rj@jpkE>s%?ISA{C#t#VApYkBdQ~tC2MMeu7G5FWRuV$%Xi=WR
zV1{YWDQv@WjFmR$cMo+&T1SbE)Xuyyin+euH)-WQ;M|27ypj3Ofi;gZ7{wn*6~@L`hIbwp(y{U^sOQ0JR6@gVdmiFF$%&<-m!8P*is56P?Y@l
z^EfDZ#T>uD*Wop&1Oeg$@9w`_h
zhFdPB5_WGprAWbq1M49K(BeBrTRTgGxyzXFI0Fg+lNx3qwlkrYr$;`OQeq4I!AUgh{jfkq(Qb9-ofmtXDyrFIHvl#SM
zQr;uyk5fp;oi#HIBbLyHSFT^wx({Zi$010RbCq?MA#HPPYR7j;)~REC>gsbldkYb|ODE3UZpOI%@=|5-PEr;+V3IAz
zd~cl_l8-NPJvT43GZU$gD5~|E%?;
zG{x$opsd(%n44ePuYh<6$bE9clQIC9!bGlapTvFShW~XyPs%QNI5wHL
zE+(dl6WnSk5#`cbqMp4PY&5sXyr4WMzn6cqR^1ym*HvZrKkbn1f()~2p_F{xYLEa4
zoVsuKPu9I%i@`0Nko!#+;p0fz7761x7`(mkwcnAKZ(;wQ%Qd}szQ@YgAGy8a6zsQ>
zd>k|72_<>6v*MJ;-jSQQ2=ZB0RKA0s<9I3UtKGYikG3ePiXae;KO+^k(%sLokG@&R
z&=9u*a!|yIEI%jl2+f-A{9I&Nwt-BTy>49&>dosg%qP?R+1{P^cJ_v5aS?d7pync~
zxDL%&fPIil51ZOhAkDzp4RxnUphLU)FGZV=9Lvt$OP3nb>?ro%7&)7bU`2uYy{
zyGlLq-GJUiY_Ayo*|h-eWH7x=PKGL))MiE;~9n8XvC7(-K}RDWspGCtn|w+p@CV+jgMh9N0#sE=yR2l7}EY
z>TYtw!&Oh@y*+(ElV=Ca`z*`a#g8Z-!94hwu}$hXAP1ufM2Ov^NgMT7PdSq~01@mrsYgLv7y+m^nKq
zM*Cdm&slK_-ZYP$jjTS7EVVcv0D)24@0`tc|ISaz_ku4I12_ujtzbdV-mVX~H=VbB
zdlNKBke3-QsM<)Hzg3a25%i<9KQ?x_MuCde`X&QNjHC=CYd`mNH?Mt?K3D#)`;#y@
zB!QVXB>t8^7loIL8V%$45c&6V81q&yTW5>;F;?4aS8Xh?IjuaS9IP~0-#w+gip@@j
zeQ>r9CX@?~G$<9O*=6@rQZ&-d)-34c8U#|KWQmA9NHeae*uxxbPZjIfou(B@{sDPo
zd=QE!nkn5pfOIVP?Z5H^J|KrF?=qngE|4jmyBu*Z%<#g8
zHs@78Za%pd0~`q$N5+A~;HO`w#Vcy|2a2+IZV4>t2Bm+NHVAE!!&NB4#8fpk7bi;v
z7@tjjZNmza5c4Gq?OTb(n;3$c`QeI?`VC0WBRoJ+cMRv!8d-=}X#q&NMCW!lJdFEV
z?aHQQCYA7iaZF`p8%}f0j^K@mt(iXzg6%l)HaXU{Tfu-Sm_xREy@r*Qx5QG7jxE
zsj9)XX#`Wib3}ui3=wu#81oDy%)Sr#T9%!<9R+*
zU*4J~8kCNDFgS4Q#h)l*5+7QEQWqJx>|_$)BON{l2LuOQ7`GAP0X1rZB=d@zn^nS
z8y?zh{Htkq5J0TD`w(K^8YCcceeuYdhr`
zl=jVhO7eYn>>Q4MY8+$neEDn~OhzlfPIg!D)OXT`xV!lx1EW9hijf&GqQ~5x*5Ts{
zVm|i-%r9kLi)YVTZ&7{VC4T}{eld&y?4JPqpG@Lyfz$|u+>V-r`YfC?WE)kC)71s&
zo4G6F?N{nW5E)0uUpnMB$k12$u^9NQjCS})_ey=XGwE`K&HXDXzXvL*H{OWkLP*gG
zI0_`GsQe06UkobzGW9c}nS&dY}_iH9p+=K(7D9?!ot?cW2WtnzW!&rRCN199`SpOkQjec{9cPNaJE?pXtfiIYu)VMH$MFuH-cq||^0v|*>Q56_YPg=WciZ6NG_m}^
zj*1f|X0!5EKZncwe)Q`v&-ISU&=RSsK4k)l~q#Y%$43^ebgfUo5m44z^7!JM-eM^C85BI`4YsOa7&KRqJ^;6
zlcPYWm2Rl0Y*vY3JCTaVGOKIuM`fzEjf>sMK6@uHzK+e4v(>*361QDskFl`l(eOXp
z4y@eVA;pPgKHG`vw}+-eo<~n@mmlY;zsw#);?+Lvt$XPG9*fLH1P
zHFr|LbzG=X3dO!)ZFG^L+wU@ORIpTi;JjhCI?Re6OiU#nx1DO(AHQGW;Vl0>IYV0;
z6CV1F1qtF&C=Y#yJ53go*`2W>-stsf?CD(Oc~IW_d*GFv+=#_=zZ_2ZY`mW3B1F)6
z>@tsAa0F2O%|fkzkdBvF;Hn2Y84~GK%07ss$uOwxsycjG_j2~eFCVcOcNMj*?}U^>
zL#aZ6-Rh7e3wS1R#F79>yqh!cBe`Jc7#;QRu<>sz1(t@c&vd-pXnlbuP#N`Zy!$6$
z{s65AoICikPyo7zEt`CZEPmY@E)4G%0#5?cQQ&d}qtgLf4n~I{e=4ze+Zkk80s0=e
zAMYoTaWpxR#f<_i+;!Ku!e-~EGT*NSP;FIPx;
zh5>um`Z_XGa{R=)4>1NGAfmu+&7b1_KR(ymij{57b^S)SH!GQx3hP`iKvrJf(Z|Pz
z1=GdVb?@Y)^V(+xFUM=)%0uhcy3V<;&*kIxd_o3uM5S!2@T3j*iP%ySR(tzcm@7g|C
z$o?gF=%w4kGut`G1r^Zg&EG5A^X|WQ$^cvYZDUK{BqZx|bnf`rV@gVRnWaJXmGEMG
zoMUp32IPPQFAtq-nQob}tgUZxbob2&R?9`#wUCd04+w4o<<0>n{wqWfs6j1LV66C?
zF)`u?Q1+kZ#SCKKrTL*4v}SX2&K}sG0dsU}v$0ZW$zI26+izwio1s@XP9-d0xD8Gd
z=&!BpWh+6F+c$_L3?3g!{(4R1H_eh$*=GAyA@8$rkDc*Cmjx_rg+YzR_h2d%KBOi7^2MT;LWwl^o|u2>l;8c`!r
z^6T&xJi)KdjjcJ)b92+azUCp2yem~Xn5#fZv_NQ-_T_HGK&
zTPh$!K+0QM*#_Ra{(*GA`=tgm4j3%}Q-TB?Y1hd`N{N*99Xrbb!*%iI5&%ZKa-uyk
zVQ}pbfwc>Tu%J01{x^NU1ZiI^!70S9!f@~F^J8ClBHdsAL%-cK)!5nQ`p?UB2Rm2=
zLjtPcjcQA_&c)hDz5htGYHVDzpr5xc9vKxo`vA}4Kp`)Qi$8oJ{uixbxmLv#`_>Hc
zu09aSVzF%?$o>Y?(aB4W0gGY=DZv}*pNNiBtaU2(6?=H&whVRN@%HOqW`^T7WhO?SnrKQPLA;Y*X0=#$Rq69PTLYghN}E
z^T9PpL5iMMHV1Ut-f2dDCJ0%Rqd#~r=5EuuN685=FV@bmYVBgce%1O*}fee>rdWONe)
z2|>JfS-e|J+v`Tw-f)Hm^Y
zcfZh791!_)!5I+sUQy^`1YtXz%==t!fe#~r;LX
zZ7}*%G~@K`l|sKQ`-~w01ENe+KY|
zA6x&zNUKMVCb4hMf`vGQ8FukTe|bfLaR@!x2~{q1@Kv_^6Kgj@i$*}nE5rIShDsa7
z(ixDS)w}tF2-i5gtY$K=dJ9F2uAUbRXlsAA1P<$S!C90h9`JGtz9&Rx`M^1Fe@&YB
zc_d{5-9&N0jeEd<9-JrqcteCG%dv|I4rSMfl+JMROKac>Cm=oS-E^w?(I8f;#9D6<
zCh9t&ZIArr-_S5?CsW%nXKw?q-lIDQzBz;BM3yXcLP{hg@2x5PUxf@M8t~cus*3B~
z=K@xdV4<9sY?iVA$JuYAZce42gp}~=FHN&PkDzV%sW3t
z%)y$6q1D7JUc8O6*_G{@^KXMVf{mG7PJUNU3~xFST2%{$_Fx>!MJiPr7jp!Xdt7O=
z#)Vpzb{Vv6F_*D}s&?~?DweYM*gpH{kyg3iSMx1J?w9Tn_HTA`ulrVb(h#F(qe6mp
zmWV@Z9b>vy;v6jTgV+L5N<&C@4oP9Rg{zud>3{E-FKqw9VD8>F>iG8F8(&yD+w=*c
zgq8xt!CAoiX-M$Oen45w!dbsBOi0t26
z1Ju&z?J&7bb5*Y=!DtpaJT)%fJM~%?jLd*{a8dC&=6=XZd`BG*y~z{2R!~^s;K?hFtV&|
zh^T=;l}GNPUB)H>ff@a92UgdwiXdr(R}81!E!SVaN+Cz1ie)aFIx~cAE}6|wJ&7v5
zc;1Wy+++GKK8fEta6lH8pPcubi%bKzCA__oX(F)95{!M`?n^rw8vR=Pc1>*2i}$C0V%yZGkue4Q!)RRO3C3qpkwbjD+3U>>$1-Lb2>67gj<-K%+eWS!)i=E|rEWMXhhc?gVD
zK}WE!`R+Cr1==_}$Cl5yi9BJqOD?1gq%1sd^mhuq{2Tlr9
zN*&{9*;6LVS8a+DO|T6aj@LC~y-J09PN?3vG>%FkvzV>bjWn5bZ^(hMcqY&
z-AXkW@yd;BQ;7&7Awk@qXHS^U|Nb@Y?74Zmgxqc&SUxU5bT@amK%1C_+0JIX-)BEp
z$t!VshT>WzF4J?OBJ6v6MH?c`kxrO9^O0UrVaBHJM_Wg?|Ge?>0q`J-00
zL*w`M4@cNcQ|M=^FyWp&@_fAqsV+=KtIbP(Fh<9_Asuzh8qCdQoyrD_9tk)I6tO2^
zM>sq^Eb7FTi}CxX*hGrrUrl?>>+W%q$3Pil<|gFIud@$ZgWf0`TMDj)|q3
z6C8-6SEEfjh`9Xk(mV00#>Mc2I9geqDh%<944SCMDo=Obe)0#xsXdN#PZ*B$2NQLS
zGfyiSw1B}aI>e7VpEDJjGK-f4hSpsdn>%a9XAH@#XF!!4zIF`m`c(IcbdUh
zO@?XgtFM+i1J-#w#p%rzRYZ&+ZG=4l(~Iy2cdki6*m>fWb&`4`Sil2f-y|w^H^rsx
z!945jta3s+wkE3Dw;LJg!bt}xxc{tbPhr#A2+l}r%=6PO*}6>_Xt?gw@}2m3U}SmtT4qqXVLp7{8M
zgmdWR-Wx;gP3{E5n?Siti^{&8Q8=Jus~7d(J^J%(+8!Zs>(SKu+6nU#bm=#OIOv148H0*j7luV^L>dw@-uiiYE7#X>tAI=@}a|o)QcOYksyE=Ut&)7Dm9wc@1G1^VuC|?q8FrSw-E->%O=J)`ibJoW4MdfoFL+Zvpf)N*nPe7;cJP|$e#Pm
zTYkiy&+RmA^eE?eFK0e>xSjFlW$n0MZ|H2VpV5Md?(q{#S^p`d=xIxIrBE6O#Pkop
zZk|^8x(g32j&sWwek_Adu{VUfvULksXSZwR&;a3%V!9{(;VR{`8D;a?=di}*6}21{
z)tAZA$yk=y1J#Qs-qTl%c(}#NHyWLfi9V!E#B26ipai6NK6f{i+BFrHn}QQif<(M
z$_o7bDi+LC!kheQe?$_#m!h7q-Mq9TYSl8V?Y*K;q5~rGu}mZgDhn){F>-EK{&Xzc
z-^^9_kqYNsqFq{_q}wuyEm(4_>)QlY7|>V}@f9Hn;i5V&(TxL*8*I|}j<`wrBu;~D
z!fc+yWh}$+*QS_X{6Y1K?^PFWSNI4N-%T3-M={5)k
z9WjA&YjR)Ch4Q^Jo>I}Di7s?|M^SMrs?hj|{1TKh3b1$~plqllY$$V&
zNOIRpp?}~-T)~0WjD<55^09!|(!DjdxS*p^O*AkDZ4xhj$3JuHdGWb({QKR|5pCJ@
zLb}rZ-VV)b!2A)u^L(4V+D-U96Ifd;sF$Z^$+AR5b?92-8VA91vUgMZYAy}?P!=ZB
ztYxadnPAZTnQjA|>R<+G2J1|v$~|(=R(c4r-K3cXIJU@cB-~bb29>WDHzx%eM%vMc
zJPG9L#1NHy+L(yIH@Box?Wdo_jSJ^Z9rqF5&TKTTx&<0F0vZ*
z$)mO;tcr^kjX}ES((W1{-EzSsz?{D{V{p}0EO7gI5|u6*f414(Fh${4KQMkix}s*S
z_`N=cp}rTqxMu(nb^quSKZc-I!B5cIY6+aO$bZ=8pCvZ%Y5;lD-q0%~9lN`ztZoih
z8L!Q?Hobp)pE#d8NBR$3nEP)q!qsXYJsTuQL#(@y1}q{(kLG-#yC2j)IZCDHZa#sD
z^^7X2#@3GMne7Ptr0pz@ZeC&iM+T0b$(73j6=g$@8-$Li^eXhuy1u}kfJv%Mm+a2M
z2NuxKg_XUf=s7ck=n-N?%~M`_haY|0K(ntEJni7)0^~L%Qwr>J)wlzoof@X1pV5c<1L+(P}`;Hru3T9XcowHqbyubeJvu-|-1Nd_=Bz?r6
zSS?+c?NFa(uXtiABv=5T;+qZHA^&B7`X3m{pW^w|RQ25t5{!C8ep#9}PQxnIJH0v5
zYkZY*bV^xcP}t`3$TLw0)2!3RcRmGW<}=W17nhXvR8DZalz%QPSS<;gMy^~}{jr4n
zAK{3{!&a@NEO8dgw=M3_Ty^uN;M0By9{y_RIF8Noc>HBcjCgQ2
z*)wF%jk5-*t9wr@wCne>4`$HsACaO=>IU=s=^4+|xSrK$n1V_Sd|ikxL8k9wYcPNt
z***yV`KmWWRhWdm(^vAln-KXva{F1g9rASch*AvWX>i5dJ&-yF>$B9tUy98j4Ot(P
zu3z>e7eRPuP~L&W+e-dx20b`R;Xu@`KnQPIa*!BxW9;HA(A|Bc6)OQutB
z8e$b2Hhz#f;;f4$!C0-6&we8T=yT4-$W20$V~L*)lVf0kVtzjR&^B9H85Y|8t6&7T
z_NpYGvbj#Mt`GJ@tI~e2;P*lB3O(E$o=R*JfHP1t_dj_M8
zVe=%IBoON6F6W)Q9lZ?~>pJ>LSSEs+4womm?TivDyvDNN=GhQ=pi-hoyKsF55*3b=-mHP^
zgw^g!OrA^Ie-o(6+thgJ*sd&r%nM_R0z>*A{+vdthZG_yR?2O*)Wg}HtoR2StuP45>
zx>0&3d>~l^8$n9d7jj79h=e_ax8&21LY*B#t&0FJ$xU&)b5V#o`=$}MIf24=G6l21
zybsMrDh|zBs=U{L!BfD--xdUlp=ZJ76od@B5tcw-uI}4=!tN|DV4hZ_$fQ1Ka`A6d
z=)aZKwqU^=;FJFI7t~=|R&-~U{Th(`-OlIB*$NlDr6*mUAl7Wf044mHy02Ntb3t@R
zQ#8_W@DfRQt#?YVapS4oUQBxTo1{iLpkRYjg?+hj#%&F3k5k%|m#XK9B>GwET&1X8*l*w(OKZOM^Y!2a
z=9|5Ck+EGQ1p)KJtk-4JgA->2n54{aFm6FP{`C+*H7Osv4HVshFEwap%i9%F+EBvu
zgCRX+-&!6daVnKgm&{mzyQ;90%RDWSk}f429SJTwIdfSAT%2Hq38;gNoNAH*vSh_P
z%w`E+xx2fCR^A=VHz(4uN6vaKcDZ~{S9Z;@zLL)UB$cK?gU`2;#GJ4;Is%TK1abq_
zFo=g2_aILUAw>HQ{+IDcFH+ZH3GEk}<%J^WYO1|2G}}`OeVH6+ubj1BnAX=!EX?fH
z5DoKc>fBU`2%nWg5?NIKJa{T2idllat+|wi*kEcb9*hKd9eoF{fSlFnZhaVVDBqBc
zfEB=AjPf=AjPXg_e8Ub-PM-z!M3am&igbeRy}LHNu%6z(h&Qq)ttk8r2nkpKlk@KF7)MmrndqvR%*ScO!!>#1Id@P
zQ6h{hg_YCQ?G28jU=XF!W9gO!Ay~GgNwU0zcqNoOyF@#vlsos`3kTu1)Vi7k#AkMC
zv6ZPvs8wo^=o(E6*Xw+=oH1|sy=;O?gFF4XJ&()~0ToNuu-TgPo5{vmllc&nSFBIl
zx*%4*-hW-R-}j`-8mRhT(qqR%Z~Ya6?mZ)I$fUm42`5y?vU?*%-fifv+!N%CCH
zuq0}-ch~4IWuD;Dvh?E&^@N*l@OoR&Tj~86=*{91#bplF+Og
zTb2W&D{zH79`w-&lr|~8!g}ci{f@B%)?t-kVPZH*t{}9V`&Usa<$x^(5o1UXe-7Th
zfS`q8U*ys1i;C!sH(oxpX;k8QK&A3>-?`c>ul3!&R!dLR=@|GuNeJXy6Ic?r03VlB
zzxN=1`~4DmUC~};MYGN~;at&D(296(CWtEv9v{_!KpEh{pFKwqIz_Ox5IS+I=F)L(
zXq1DP!6>sc(jCw4GnsFpzII^UdZ$n3M8~`0R4k0vz}`q~aS!;46Js;Z5C3W#WuKWF
zHZZba9}X@s)Jx+z&AS6g+mN&8nvUZtIc7R_qKIom%_aU^p>@NbTmw=?qWp&@
zf4Oa`x^AK_ME#j|W}*m=P+3hPn4vQoa}D$McMj2wM2Dui~DgztW(lZu-VI;mOIuzeCB
zh@MsIFOuFYPsrw*tShW6&$d||+F
z$|pQQN3o9ooR`-G0G;xC^I@xM>F1DnB
zEdn^GAE`x~q=J|8oPmZBik3aV&^l`vsL(mI3Hgy+p=GZdY~Iv=r4VJg&^M-$1+5F~
z+wIi%TccjpHA*MJaf_y
zj2Km0zarA%%YBVsjzd#QuQiy3=jH`Gm6x{hSJ9GjmL9_dnk-6YGCAln6|ChRZ5tZH
zzc!qXl5drGB3BCn)wHI+%$3Q-)TfM_-|inZi8(GAX!e=s=?+;`HyPdrX5(0T=I~P}
zAhIsbg177|kEYyo%YxnMPM2_gs*m$>aKOp(AAtEFL7Tdf0jt)1N9Pm8f}}Lm6D{LJ
z!Fs1EhftWBD8The*$Fs{h<-l;Lf(9o{j-k+%5um`*f1=5o&W>+MKC>ZQ_Yg!!-s`A
zRtJ~RK}*--Qu9Bgt6W9SzbBmrNeAcFe_B`crBd!6ZJq~km7hQBIns?}3JZSi=VIHf
zE#lj~ykD=-#r*Yrb$W3g>rWZl(}yZ?CfWXXq&=g>bKUqM|4YXiW=H30i>HCUeztz=
zN@{aN+cxzMFt*r;eSI!V-;4kK=?l;uG}zYnT3dGac?-WmKJRx$=I1&#>5Gk8
zpcrN%cdPbEviq@5H&lg+y&wIsureMO1NCb4Q1R~`=>3?oQ=tWvQHxT+mJ6fX#;Jg9
zq_mDn`hfOYE-`o%7((%tj~{9wu`3v3uqFXmz;gLZ^>xDK@?y%}#JZVmna!r)Zc4n#
znmFmwsPS?YQWb{zi}u;G%CyTYdgS`cF%eFLg2A@@qvR}{xP=<83mA8E_^>ugV3-Sr
ze8qAV_WxPs(P_3X*(o=!P^TrRjnyx-6;}kt>98Bt#FX!M*i@wS?asuS#fuCu>)us(
zydAZEZJ-+mQoO}k?&H{`w*3N|Yrl4ID7B>Qusz791}VxFN)b}AthS>$l*Ig4AX8(a|9f)KNs4iY6?fHWVt-OA%ZDMBGTYMnSAZRu&M9YSs-BlKK3?Adv%3}n
ze0_#GW&;PcK)nPJ;1_fJ*@cRI|CEvS>vde1Ja$3Ob5>G3Ce`0laCL#o%zQfE6?5Ww
zF^?b5#%0x~*@-ny8tAYYQHj`sHerCwSM;yaEY$jxA#?)Zq4$5pIw^e)7i>U(9q0q=CET+1TL*d?Ghz*APM6TMx6_@4P+{&
z79HemP^fX_g(oZN_}ko*9@33jKbfsj_Gzr4Z0Np8BxtVZ`^Ui;^qQ}lLuznDWl`dX
zzo*Ao)D-*56!t%Utf@I$0USu&L?|jW+8sFGqa6T{)-T~Rm3c~F#2*-7+p<8)bold0K?rDYk90tGI(Ya+9;bma3cnt11!+y-{xHS2C-?DsD9qWe$L~E!&
z|53kKva38l+m!ycYx4~tPjBI8zz~aZ`MD!(mnM4C$|F_Gz;6e~)Dq|SOnojD(zOyp
zfCmWSnz9Yj+SvU}WB>yijqL3+dFI~s?$tX_9L}TyBp100Sclly0~6pf=#h$en9NzI
zCi(=r9RagfBjr7XFXR$AUG>&n6CW3le1J{4g2Vn#bM`+H**ZG@vV{zZ!&%-sTYR0V
zFEe)f_G(14N;Rf(n#UraYS%=9HHq!)a?SHJdgxoKMOu@F$v8aTjmlAlYBDCW(+^=>
z+oI1NcG_M}*n3x%R&}Hm(nC};`*W5i4M(s!6buPLi$wi~_BfylB`pLkJDXtz8sZsm
zSv&g1q70ocx84T*b9GDf>Boc;7((A@4Z-1+p6gkxZ$D_rtPZJR33#V`T6^3(saE5&ccb
zFS^d?F&Or0cuR#czI_G`#HEt3&9VVqzWSVOKB4kJdOhe$E${8@owInSpB>wl|8b>z
zp|CC+aYWyiYxMUo%c3V55+fWW#If?@@+6VR5j&SV{IZFl@Rwt|!fFF;!0)Wq5)7Dp
z%@#!po!lHx(>(4OdE(RNBU24)@FgR
zKNEGF?Rn>Dx{hXO2+De_zAH3Mn3ZB)cn8si|u9jSo4{@Pv98eyWZ;a
z+C3agP4)S(_u$*{>vS&=EFzPsp+c|CCG;~ST>Jjcu8AMQ3>NU?f_ck;?*qJ8ItCOzAw?Uz@dZ?~Ndgo(xf
z97L$w%&y!GIiLL9I|Ok~{~6p~bsyWhWQXB$i}CiuRF9g@o7?Qy5#bI-0#&XLBw0ex
zj0bApAeI>Aal{_s0Ofgvbc_>beyL=c4)c7i$D62D`qOQ5@d1ke)G4SLw==AE$p~ZzA}S
zlu7Kqm-cw&$Nt|horSY)VlIF9Ku8({;4~Z<601PrL6VN=B&3EimX8FqV97=YiTzae
zFX|1I@wy~hIHNg%N}Pukx0bXOnUvBNQN&hXJx3Dw$8lmv9!pQpf4O5Uyfx@O|E=>Y
zObIyiw$m6ern1C1U(&Ng5k!#aC%lsFvT9z{3ObM?`B94}Mc^)1CGMB>JAw4#pq9HF
z(e-)Zo5A&~l-^>jEf!h~BL%YWd{vxp^E+?I^S|>gS?SnnKO{}jvdtiMexY^#TfB0;
z)@DZ`DVlA1d+rx&U8CBv+QrR?ihC6PTuMALj77fMFq3bak+vzZAbMY=oH12S8SF~Mju>Z5MIdDC&EWj(WGzEiPiyUW5w-8QD7hh(!oa_y5
z5A2^FZ)QEvy!yPpc5>CQB;ES2-w(4{HE-SSCyG}47IEbUaiv$E_lI%W4^ZjbJCjG3
zV7!9F-wXN=8D{s!s2>Sky)hJ6y_fAOciL#4MLoWnjBPiW?L>V0)j}iOwMND6b+3~6
zgF5yBC+qWROfsdfd7M=nH7=EK53H^YI&9D&>}wyk^CW@$!6PIy9Ywzb5z3&By)D8t
z>d_k$J>s>op;%veWHZ0JLn#KtGf4b;-{a6~>;%U`eI3rN0>YO9eHI3RDW=cS>18K1
z-8duzNHPXALiG(Azh#0*8MU&p!mIcvEuGO|C5)1p8upbxcIFgv@upqS2+^fza#RQ2
z`76V=6BSWMT)%jxEx==~0+qD1NeDxc*hbLOY@AIfl9_@kz`pBwx|itNwDa%Ci_N7t
zu@bVwcMjL=Cm>sMwQDo39GpIyWE!M&(@bp2JlK*%(MSP?qj};25eIh*aQU*;uCj-e
znB-e~EAO}1hCPjS5>IKRlvx}NYUFPUREm{u)GBDU9D{ArNpL^pM1$bVK32lCP0vfK
zLNR?G&EBO8!Yy_#kg%q+O)tqP?BerTqoL7>MX2F5!0t;(rpjG(x|ti`1Q-
z`1PD`{g#L<#*;;dk19oq-j+nqE`&lyZN`)JIJh;qD!92YN{P*@BNj%_vmW1mYoD{f
zS2&GlLT~7Brm67wJ^W1mTAUh}T4Z!gbwmj25~UkgmAlTYC6X|AYcEhfH5=-X`u`mn
zV>Myqc}3kwMCrA~k;!fS&~H_M?=<=8!{ZE|`magTL<+rI`htz~qL7QMjEQtXDLf1O
zqF3Q_Y`MHD%bcNf5svCnF6+x|!3U|e=tez3$M{hZWlg8<52lLwK3}ECWiqQDy+E3>+E!;N#Q;C+3E0k?MN8=_$oqu`13mDKlu3
z;0FXc^waIH-bAsSiB#nwG
z!rGOk#(7Pc2dY@%>{F%!=+|Q$g
z@wR7jm1+G6?kZ*0AI>>Wig;EU=ZnC(K(FiRIqrQ$S@)BW{nLEix!5rmkt}-p+?C7p
zn)_W$Uyg;%Vr^k8t&%?fCyVJr>Jwq72)j!Cp+si6!KAI6o}xzRgA%pBZAyE{7}jZp
zPr2zFOVO&G-p$#>rWRH-vT>7DQ%v^$r3S`a#SAq}T|&J3!n7?bMdm*G=--<~EMj}n{SaHO?OShkzCd{I^RHbGqxazfI^vrV
zO)-}g^43W3D+j8Wp8dL};cp$+-wzl1OV915rcO1mZ4RBJ>FoqsrGn1OzXrjP#Ie7|
zqZQDUJaek6ZyMY8%6@fVv%V1(;9v-IbCXVE?Kh-*66fl%z~^7Lo=!8}(vHt3d);Pg
zz;h2DdLBUK!5Ug3#*X=;wl|<2%d>=W;LvD4qDQX2<6Ixe0R2Kxh@xcG3yuUOlG2X?
z6Q<%Y!j%}ry{g5v3Gh7R&|SBNxs?%4Y(44Sneu4i!@MhoA4<-`x=9ol@^=>jRDcG;vX8&ws0JDgLd3eX=_9
zUPwwnCK+!)eSBajrt~NhVdNksH$j>I-UFp106|;@C#b1{Sa>EoJ;0(@-FsAzT3Tj-
z=d>S0F<;3#|5^}kWTcia2shLE*6MH>GWE*wvj4ik+T~~$d&kvr6_JzS@^!Jc9n>D3
zEEGM<-ohR=>+s9AjCF))x!(Ryh@alJ5L4NO!pfehKX#Q&()-N-{2lEMK@+QeowmiG
z$b-GKmc>emDLaO#p_nTvG{=|rqbjT+gem6-g3i=U9J`
z+*4TIcW12drTPGdfzbU5QA#f6JabIPa!Y2I8{67Ic1`%-PQS%b~~?)
zC^U^zR<31nO5S1M9jcIQ`431F2okb=O^c117{>BtYw_PNO~ZmQ-oA3}*yO?Qd|hsb
zpzc0*j;EI2n%%p!l6Gi8KL%yRLQL2dM_rDJ|E|R5W$YvImLo(GVR|RJ=`bVegW$LY
zgtB8{xmtJGzz&@26FP+s89kZyv!AS{g($XP()dfEtTGa91cCVIX%s>2z?awz5QY&8
zsQq-C9~O^}+E9(P?9qRKTXylQG}$oE+{Ku`v*KQ#t4zOB=Ywf=>(N)?iS=Y|ogsLd
zr-f?#fXQ{Dm?)ph=?AQ5Tm>b!Qk%KDNrjM!>Fq=-GO$pSgoEwxES6wAuPkM}AUBcCq}Z|tf%TV*
zRno%GPK~2qM9l}}+6X14uu2|a^^4;Ui}e+!MN~|dsphg~6?}um-U6|d!g--W6ewPQ
z@9_BN&A)&;d9h-$C<`=dcw|EODE-QbUajM^U~$aPQ?M9>r4j~spiF&@;eDPvM@z&S
zshGwgP$AAu$S%)~Saq7{J`QjzqrX(cqYvtJvt>bctWZ~am=PZqy-eRia?uJ@`Y@?GJfw>_A;=o@&VU{e`FdW%Ps6c0_x-
zT-L{p%m(MamBO|0DC!;Abf{>Igl&T%QoY0I2&shtfE#+^3etyE7HSJa=i9!#eK~p&
zmuaCQlwCoI0QRqDX*b
zAGY_s4pmDgz66kdAc3Xg$27?4Anou+@|`nj5sA#=z*A6d9`ZO(STaZnd1^w(ogjFj
zFddhj061{HG}V1gD$C&?}2~M
zUO(p+3w^jW7OTQk6dyW18>2J<3X-Fvp;5(SS?)zQE+Op@2km+j35v5sD(Wgj9b4?N
z#7|WMcRV)67z_K?90QD_M9!3-eXx1K2jZ>+46p5RyXTx(-;}m1ayHBg*m+Iqe@`^>6SvA`@jt1~I=$2GxCB1sMb{mi|L0F=&Q+SoCgfw`
zU~83mE6@o+ACU_o8w&#Jls@F+t|i{od7~tDx)q(b#`#2l*XpR|>or((D~v;|$fBa^
z&x}K%t9zSIBI?Co^#Z@HyVGP}HlkiWohGxxlBYl&2rF&zYVjBE`Jes7F?ob&`9xP&
zGdx&|q;h-7nXbZ{x!Ork;>mKMtZ{dt_bm_uFg>~KQ2D8oQWv&f#g5hZqKf)zzbgln
zg6oz>q_3S_*ZFFOR3ii7Xced%_gNv3e}uw*o%@sSPwaKD@`<&h5sG{k`EH4pg>jn&
zG4ptRant>1iVR)I4%mPMrZ7mfqcJ-o8&-HciLRxxTtFBk(?vSPJw)KQ;_?LjSL2Q`
zPYDL1PCK1yAB*^-6lIJL*sZNHg2J``?G>2dB-=M=Q0EZ-Mx?27b92x%%*nSE(eywz
z?%MR##4EAh=YtuvRTilGoE@TSRc-!q7?hTY5=Y0&I4m&~achTGiL3Zr@qtux=L~G&
zL5+$Fl=P~E0lNm;Q*%aAU`*}Op?9j>e?=TCesQZQ21W9U8CPngbs5G8!b>@W?mop?
zSXCP9+`~O~NKL|vO&Y8jAu#9r-v(G(7aI(y|KCGsd0{>@6Vr}8+(K@D&|}2i1AaVH
zk740V+k5DDcwir^g$gMyv^sMhdxWYne;zH}$&e5qoMu+jxD>~s{uU5=OGB?kcDwSz
z?YVx7r;&6vm>24OaW@UkW080FDI&Mg(5+QUruSf#C8M94U+&RqX|Ar7LYv~VurD9o
z*xyu?)DsVe~aY2i=66Xwta;+DZY@J4)R*RRXpOS`nkdtSxrImiQa=`MrZu
zyee4JCvbd>$j0ej7dyk>S(Vs^F=phR3#OSI=v9%kT^zk-NA)t~G4y{$;1!uaG&eWr
z!~=F|Q~2#`V3tHcpEf>Dhg
z3dDhzdb>zX-B?)~?Eu
zjY!>n5Ed0&i=GY8=4rR%ez;E%d?@9am5WtghiO|M41!(!*5cFpn@+ica(BlF~TN)QQq*&*pDN=iYc`DIOjJxDv;x2jF9X^b`Kqb0Go0snZP
z+qEhvRD%p~FCS@7J0*=R&^r2kmuu4<_NN@+LLW@QBb+pG_^b3SsT59E>dbYbu8Zg!
z7zwz#T!!^+GC(hjc74QP{_9pxZ9A@Tv-#vV4(rX;yngE+VOA^!3ssK%dZ%kc?RFkD
zzS+2R8)xy%P4}gl+B}3^vKlHi1{DH%imKl4<|8fw@U0Ro3ml#4t$mWHU+!E#R8GHL
zb!{L2x6*WTY&?2o>h%U@l}#jeGlQpRZ>q)|R^lU21`}*N-B5%tRwry14G4MICKaj2
zBG})SGnZ_2Nf+&pjPh|v+XUr9q#}MK1@NJ}E?3v`l~hswl2Jr}dU>0-oLr9cl29Ns
zT^>hhEwEGTSdH7O{AKRUsbjG)S>Yx)xqzjCN-HHD@~{ZZc7=hG;1&HK9%oe@-ry;
zODNkAbL#BzEXJf?h7!66Rree%30)!Nb2>`MYM`E3$a+f}Q2!-=eXK|qbffZT>#cRa
z3~>>v^WJXe-IkNzI0TRPW7~~r;-X4y%_6phNJYJQWoTh@MsE0X
z|F*odh7jHaeXFTtn>BF5NkK|wr%~1zrGB^(?(hT
zm((Nnw!>7ZX}l%|$Fya3$J=rDChKuEte(z#6%YI6-jeA+N}oeyX0f?=tr@v&D)(-d$zJ>e!EdFS380T?
z;dJlgmPVIceCe&sg;63@Zg~pA;q#n&&jw`m)#?
zEXYw}9I8(b*)>><=fS5Z?1M?&^_=jfChXu79aFy@9F>vYdR2Jay%NU!so(rxCRMGZ
zDe*ov=8XfwB7hvq*~C8~r>2WA2ZXe!7=8_<>TeExd=^L6IDbNVfPK679D4H_8eHx5
zJM!1dimT~1rn@Cqt2t(qmf$O?TY)@@c)#}*=gcL0LBBBR^&I85XR7f%2D3+x6C6u|
zKBz5?3_6?E0&K`)AT)d{NJtT1O7qfwNm3=jS^@gwF-4OV>pNFD-BG$n*7PRN!HD^U$Ep&uXqrb0Iz6parFWl`e6l7DHPLRJY7
z(;BpEBsEA!<`jR=X5AeEvA~PZWw%}C9ia$3nBkA#ovq)zdd4!Sjf&a@Z0Jt3XH;VgQr&+GM(|3#NwK*!7ahdjODA~6@MWJ!l
zR!ixJ!?IA!sLz_KsI;`S*q-qCc42zf_^h7Ud&e_(iG>!=)5e!E1lG7{LqHZL*B3%?
zDaD*Gn-Z&R0gZ5?zMsSegI~t4#*;5;C?9Tb(!Cd>(;;urdq(55q^Z=f>V@7<<;;cy
zlrYO2i1m|OH&IWV0wm4~y%4f1E*;MFesXnlY`r=a@u^8s&a{dEWYDyCVu)w6L_{C{
zYO2@-hr3olQ_n%xw=xL@Y-F?3cx|P?8WuF}fo~qMO_k?GiG=KI&!V3B=;8HaG_1wW*I?0s94mk_Aa<{rR4q{3
zD?jyPs{6`4uW^SOoA^pU?_)o2l2g@+NWC6SL&3J;YKN;fe!LOD62G;l9QsMbs!RkZF#
zN)+cF89@w~LUBI)*k1VCY|j4F)@EGiD%v1++wAO8^a`_1HP6+P$v>Av3t4h-aFCSd
zapUWS)`x2|XUb4@|NCv<_1m`k6$v8+-|I579hi|Lc)8wT?^^7qOoKa|_}8iE&hWN9
zJ>6~%1)yn1{Vk!4T6+}NJ7=g|jyz4vH;6bjv{c$`W|GW+%|Ws(z^R(3G3YK}m-*iU
zqr6+%rwT=iK}4?b!R7dA|
zl_H|G$gLJgLSy#VwmcU#Ely~1;^XIy<>-AZ#!OEe;p;z
z4lnBZLhU-4M~IY4gdf2+W*UbeQAwh{y}=Zwt->e$`MdiaskS@p?4-^ac(-<9I|VN>(3>X|)FC@y`7AI=xQy79c0VbQ=b!=oCw
zD%w!vekpfzygFK^XEp@SG&(pv%IWC%85kPtbo+$MBQUqMe_k|%h|;ynyeN3roH|tH
zgve?VS3kek|B~>yG~eDY*nW;JarOh(>HZ3G(zKU#32j%-mlwXAG&;@kc!>A&yN1ww
zc>ZUjry&v(e6(a~AdnZR@q?co`ftL=v>%9ES{FLIGce3Vo93eJh6KEpXxBFh-d+^XvGY|l@=u6nF3l_VrHXC|H?`0bX^geJKaoeZo6YiFL5
zYAZjHrbimk;iX7flDE77Y9X2fJ+4q%-yocWX+Y}BlzKtryc7B|Lt1$*9z$f6eJGLa
z3|WthFTfV;1`iBR6_R%k%I-Cfe%|vIpW?30fC{(C8v>h)TRdn6Y#SV?PJT43m-iqa
z^*(V6WZUE(6fw-kgJWow6%)5q#Skg24V$DEk3JIWEg62OVRt%ulA%ge@p#Y=!B7Zp
zM-sj!zPc9B7nm3T*d;P52l^m@7#iE{Np4@GjuG>YQo}=7Z5vX1?Q7r#P#FnaboRu)
zs`5K3yx3;1ci0{Cq3K1a)H^N>CaQ}^?}?O~)g&a3pjL52%JtavAtbhhax}_-#GT%W
zGc*I(`5;nTx7RHEcxOP?e<}`9&ieqRV1%45`n>@Nsgx^svA5n6Zu-T8%a>h&xDtc^
zcjzE&#;L!Zq58*pw2T_9U69-yFAJXTwaSFKK@4vRBzI(d-_Iyees_6QQw;&157KprF_vH-_j5U0l?Wvd}8i1N!aptm5TnY>bwZq=-<=
zU|Cm6mG~N4XY
z-*pvxIQmi#AGvst8aoW*EYep!CGhW*Gjb?=7a0S3IZa2
zyHc4s{?%{tvdE9>9dZ|Xm+U&`dX3Gvw_Y5fvm#|%5d*?-v=8ySl}6YzAp>3Ni?_!7
zu7jihpToM2XU3Mhk{7Sn{Oe;Kz?-knmE-t?fzdewOiC~vClj6vyXDOjdIK>hXG%!o
z)>VkF+(nnn2(YlNG3{}2iF_Z<_?}~Vq%N8jGSkm<;ica|p-CbuOCib3AYzJJuQe^+
zn;*Ts4`~`)_58S8ODyHN{8hVm^V>f-#@6HBTN2pL#@vI3E_P4t#c5>dk62f4wQ9EBuECt0IpN
zD?Cw;9L5UZwO2nZ(u+By(SJCkv%GO8LMmOs$07LC-N}a>G9yQmsW6g=H7yr`)dR#=G1Bbr&^Yd%-Ok3YLi7SKfU@A@WZhlKcXB|M~qwGemm6
z)$^#(MaieHeiS)}dg1R{Q$>=2jF|Fql(b2YY#-=Jj}zKbTNfVi_IYd;p<`#9I=LS%duAATLNof
z+gN9cC8g#9B^@w8ltSFc0I1l%?FeyR#BO?v0MLJ3w5L4XHga^YaDEc0;fv?h6?
zp?lEr7Dk?>WKV6i-D_jdX!*3le1l8|#Kyh`5G(|wt1W;ulW3m
z$!U9DI_VGB;=y1A$!jXL?=d#0+>>*@Fum2MzEbI7)Y4!pKxNzZV=MB1T%r7N6*zD!
z#ytOdtdG>I`DhRkm6;AZrvN}~#nFAOWF`2!=0YK8^y&kSIVY|QN6G8zpw6+f5tgwy
zO+wX#gY-a$L%trNdCkHU1puqt2V@5_66Wv3?ZU-jI$rPRxH&0C9O5kT-dF@3YxpXE
zh~1f0S_Oybv~Nu3v&(o=y`C-!)L@9J>q|SH0lZN>+8-%V-Q>pc>H(1?KYdI802O!{
zvGWtlFg-){qA~>8zc$EoEt9`RGeAlZ^ZDBgdmx4Qpkqu`Qq+_jt=SJEQ=~j1?i4}-fpckWNb&mcr>d$|1Bt{t
z(YLhy&9rh9+zM4Ch6D;ojsTDR@&EGSXbx@-_(96eTGF~|dRljh
z-xUb?sXnzEka|x@=@kAX=VH-NyLn*0iifsAp4++)K9zqv2!2cjrant`6S6CDhK#5D
zmAy@S`}gzdO|tKW|7zdCra(wsK79`_b^r?!1G51gFmqFdy8tN#85}^@B~Uw`Z^f3|
z+sXSzgD5`R$zHJ(*?_=YgrDAuKB+Tq!cnl=9#^GohIgq@Mv}EIT!*SaGat!WD1@_4
zZK_s0AUj;^sHe&}p(-3Qui&htITV!Di^+O!yQ+W=Swe%*0&R2BnmGj@>Y!iA-hr)h
zi82SksG{?)WR7bu?2MPj1s&JyoVS>|
zj`Yco98*sR#{oL3sYxX(c_GsT2ty%PU_`fqoq3&qon3YtsOB~%EOrL^BpQxHC{lKl?AcXm!2K2W_C*pv9WFO&mKv
zNxv}sw29LH5cyTx*uFQ{`sI`nx2nNfXXHGDIBjFhduWVIS^iY^`?mg!EmwsuAjr6`
z+-6awF4jL&1MNbYXP3abw|YxlV_E^~l*C=u&$HIHqbiPY3!{9cO%W!^B=b~6MgV4x
ztA75(OD5elG@mm(m#BbOGk@;8Ar6={K?Vu7{bPz)m(qPOskh=9z;Vvy+n^~S88SyQ
zAzkAdJUO^}dEoSfGh%AOnmQA|d+w8jiSEDnMtu!FG)$A+S~@fI+EtYNRSOUN_hncO
zeD`&2C~UFyRg~&2DS;lkrC8sVMmH~R^;PXSMnxvz(tTGH;H2?4+hrZx@jLjYQS5nA7J*=Fv28)ked4xF!-A?_SUr`t{AizC?cCefN?fSDTD;ke>1aR6zUB9BY8
z2^4GjtLO^|h+j#G@4v`jW*C##L#p%zzuQo+6yaf``iB~7(UBfkS^G4?G!{{*)F<9J
zumagX(IMmZv2R_gt=Qo$w2E}r-@efyH?%p@JXBI`)x117j>XjTXNW51RuH1SB1
zq8&=2kFLp3$sz7lfS4=Gk}_NV!|Sg#LGVekI&mr_)7SicB*#t=EA8S|$>&qt`VRX0
zxkaMhhLXF!6$@R$9&ZtmG12
zRt#n&7qElygbLZpnnCDQ7l*2!Qd#th<~`=Qg>(e|=wu(y#j(*7dX&883_eVr58J-|
z0*Q+wEmS3l%8kocl!3_&3Q3X+6z;@vOLXvbQHN;>c7&V
zNoRRokOIOyV6*)HQ|_El1%66HY_;0X<4_NWii!fo^GU1!()zp+^}x$hnW?~z-KQ5^
zD3shygf4_mBf`6lrfl3t_>kqDkcJX#@ocLrG}u)To&xnZLetQV&&xY@9`7I(6lGX*VmjCLQEIwapa|ayDJN}P-nSVdHpe*yE=Me?bk9P81bq=_4h)Cwfu7?1orCb@v
zd+cTU2nz&AI*laDf`I9ZU%^D8`+&lMOtX}h5*@bko-Jk@1sB$9?T8$6OBkUe0j=s!
zp=5=ebw2AXOojr<3G%**Yr9vV&U2{S2py&}VHlvjw(m{-Mb?k$uBFnp(i2m9)0e&K
zL17u}mo?3TJ_^6iogxM6Lv?r}uOEO~}=Qw4hdfCol9D?FcW_
z2bdbS1)k!_t8D0c!aZglWG@BI?E^!_5k!D!gxX||=3
zD(Bf;2w_)H<=9%|t3oDSRx62G#SspRr@Mc*R7Fy-l1H0^
z4IvVMr3b7sMc7Gir6!h_KeF}>DiBU5J^o4m`X}_|iQfwz4n{Lx?&W<0gKX%BWCi6_
zscH>W7fzi0UTKw1e=&@C6yVnX3@CYqp@
zP+SmEULea8hb91PSQrM0&6rBO;^_Q@#CsHnoL4(AHU+_%A`_K{1-bQ)O^}HOT9LW+
zqo5o1u+X6`=Xt-(GLH~A9L(v-SkaUBH8Z
z9=_0CFvIS+ycF%QP~Y@$4p_xU2FfdYg|%aswEBig@M|YV=S${&siAio_RHquC+z#D
zN1Gu>mAn7OR|h60xKVMbz`)f)KPnQCf!{A8GxnJMC6FA=Y~<~@ii{W8#h>7qfKOV7
z410zeYI43XtF4Ez^7pMp-Zx~5I_r?S50+qe+|^o#GO=uW+hAoXANDd`7ypp4>|narO}m!aNf-3
zls>Bb9k4g{6sVt@b16s4uqm;5lE0VJT&Dm`;6`s;7PEZ|ZgkL$XQnB%aXx$v`VK|`
zOyRr}|HR>;2VT$~mU%(8WH5T$RO&0K{(@8|SWL2ikI)bc?`YE1Fj7(+NK
z+U?Oap-RW`1*Cz>dR3nUeh~A!&_CDQJSb96u1udMj;z%RytgRW=pCBg*^AvJ1_+0w
zGeGwF;hR0ZJk8?SC|GXRv&o>Yc_|KGs7?l{E&ybC`=0aNAfcH17Ez3Y^PV{Fe^owv
z9#eeMN<+)*_kMijbr<~lco|GX_xT`UGPiEapCD%4^fzO-Pd=>M0R&dC64vYZ%YS`3
zG!ivozK+DefE+lDK`V=-`m)1vp=srgs;guRRfcrJ&iID@W3PW}ybrZ^#JK+sl3Wee
zfWRd~n^anWM@sIeyBJyKsE-Q1<4j@ZX~)Q{*jOV{3HVGOVP8%0?TUiekY!JA4h5TBGhi
z_o;hOBM)%<2+gmKX?3#mBQBz&vZ2fWp>-r3m&<(>FB*wI)qM`V2wM?RXrnoqyT_Ai
zAdWD~Efs>M*RPwHpFzVTSTD$?r
z#f+wKrp1&E6Bv2g_p*~{qmr}a29(q*brcBL3F0ef$Z*)ut5(Ss7bf3%@N$q5$h_(~
zA307jm$El>Y$uymF6Tw?^;h}Y!c-GNWWe(3H5P#^Lkin?mDK>H%5(h*l`dRFdI_@1
zeGrAcHH%uySM)w=2SUK7EeLd!G+%nRP3BLABf7t%HO!gygY{!#bFSp#O)LJcWpm0e
zFBE8?Cehnw(v8+4g|qTvHN;{?az}siUc=e{&MP?1d}`G6#w@DZwvRV|LiCYckE6)+
zGG3E+Vm-T|$n=JjYMEYD3*{cVSmnW=N!!VXrCoSw6(T2Br*F(E0Bn^>3w67Lj8h(w
zxqRRp>aoJ=4dI=;DP^zRXa{1{dYbDp$SAEZ``vxlDNNnRAOi>$a%=~KJt2{KmI(7L
zq15?@LQ)jsS?jhF!T}!5h>;lXcVxL4Ut?YB=`Vs=u4_0LMrO(t5FaZFFJt88`iI-K
zs!K~wuOt+w7)Q5wLg>8rC~nRX+v#_I%~pF#=$_g*mE-;hIKb%k{{&p~jl@ZP2jj~?
zo*xvev0k&%snBXMg)q`7bBBc`oOlEscv)$hGar67tT(;1i>Bp7U9Lzr;B7aXmN%gb
z?-2+&hyxP@tNLB(h=Vk;nGdY73}h(f-L}4`{X}}T?z&$aGR$4+xS?lNDLS+r^!@k#
zAs=T^Q7g%E#ZQ9|7?KMjk@3u`%`mB9?OS`!!0KItI28C1VaTy&%OaV0^n4!sZy|ez
z$oYs`^vlD#yB1>ZOca~feSWwn^WwJ4Tl>{TkR6)zLP?;b$<-u?sCJnu6aV>6BUi{e
zyCW8rW^ygbZoZuDe{6kaKvZ4SwLYMNG$v|GL+k6Y3GZTM#&bx?~9^N+O$1I=dh%_~2bEY2)D
zYhy4Mm|vV&700CVRSLy1im7pVhip={ap%A6Ya%j|6-pt9WDdB$DwmLJh@Hw7s$m=`
zAh02ae*QcTL0qkBNCRBe?wn<)jJ?Xmncx=MY5B&L4T}8f`tjjFl6%dIyIJmrLq!(K
zX4=mHuDhS8Qj!g
zX&6u1`=h1OF9~#r5ypacnxDONl~9OCA`}zAYl|hX?XQAUmK6&He-k4K9Nv2i
z`hWAxjIMlVUYfKdHCesZg10<%S)8p?F3a53_o)lxGnTVQ>^8YS1ME`-v$UQ6RwX8&
zXO44P4p2ZlAhQNBK9zsGh=%XUMH^}N*3yUA$5v-;P}^6sgg__ZI`BmQ?h&%#0jcuo
z1z>pRXZUA#=PNr{zr)^#4mA1-i&Ou-YNNhk;l|5aeeovh_O8rBu
zHjA?;70AS5)_nDYzI?le^^UW~GTh(c_#985oHso}yaiS3T2YUu%4z4%0|p
z79|Dv*_Nl^f&obldR31y;Ym%Achj39T|Is#+$g^B6Vy^>Al$IVU^>}OEK
zg35>o_0r*AgH&XxDs;T|i2_ys6(5fjyQ&tu}5e{ES`tj0D
z2>^q}PiJ*4V8Xc8?s5X^7&CPaLZvLEoD9O-o?Y&~{FJ`A^67{~!{XE_G*@?bIok|T
zX{f`_fOkuI_7N&<^@-EGP$YfK8fJI$Ldc+b#R*TQR)fL`eSVe`S`h1
zMrG(v$V{UGITLHz_*c=||H^IE)`Zf_p3y^QyWDHPeV~ZSOfSP0Q9Y)1iumm_q~&+&
zvJ^eMD#VBEAu@_m+Zi=sr(<^0XIAGC-0qc?!u(IYgne8J{>3YELg^bWztiM%23;7?
z>sxC~?OgfH!T9=pZc4cdW4M5q7rw|5_BA}24vR?c;yyO_3tFsJ+_=N&BrsA*DP}GF
zQ@c5hv_D+G43X%v$an^{Joh)W2M|1(F{_DcIhf2t0u?90fpoZBwcVLYt`sCzu}$#>
zpzJ7bsy>K3b86fup}ahr4}SX)><*n0{mY-%%p8^&0iZZq^f7V3)VI15lF3t$D9!U?
zOojip4XA~-6pJSd$@tbL=*2GzC9g0haBctdeN
zO*ld8izdM`%2cX^OtBz*LM&XJ_erY77v<{n-7nZ;h`)?!;ha&F*N;5TnD|A?`+1^>
zq&}+p&OCK6nOz<`MPPr+%#HptCT311hk6xm#F~E(M5T1G=uctNx@T9
zRT@nV8UjEFL+?3bELd3M6_@YSE8DEV*nkt~gP?#)jzY4EIS<=8>xG(|=QO^`2aag!
z{r*RK8a>4?0MG?>x@@!~xaf0$`567@$Eq)lbA13%<^O$P4Qj(XH}!jxLk9wT-X57<
zNPJ9%azFGQR739mE?i#UD!cP-ktFTAcxy_e5HPWX1^@1Xf+WhK-5pmaU*;-Nu>gHa
z@kx2O9`u78Lsc#~{nbD>tw`=w@A3^-qBXzRU$_B(pqOXbl!o73FSR)ZRP*+z1JWPV
z4?cGMIh%bDj?F$
zkkx@z^EIN@owx$m3;H#z77nGFe23Y+J#K-MO$UO^cJg4_=lO^K8Dj&g`dV=pKt#|{
z8C7kU4FVeih&sQ06j0Njq4hm>8<)B(VCK7Rjs{$JWO1?p5KCyF2A?ufgRrk{vn-o3GP>YLj2)oULHj
zUhj?kzay!!)GN}=aiQV8EQ6}Snbt;66C!;^XoL1!R%%E-lgVfEOeO5+=YQ6Y6Pn&s
zFVq?eP_f3*fY!T=1M*|%AA*VNg{&oW&irB%>roay)U-F-A^o{7WmUt0!lrDvT8$;I
zS8;c8!7M7#t1r51t8hO#fhSt81v#IEgUM9^eK(YixGf-uU8^}pJeUB&4!K=A&de%#
zZi4w{kLVU=(VW&0R=bNQroVLY0?Nu^I*u1wvLwMvO)JkdtAW?BM8Mwt?QBi3cK2zG
zB6HnGF(MjQ>#T~Oc+@PGnfN6-{qE@NS@?lcb6v2m>+mN_lDeoNq&2y>iLiNL$ASc(
z3qilK4BJo8uWG!|aelFj1$S@?d4tr)jk8I^cIq(<*ea;n%R_k8vin+agU4;-*!J||
ztO{~Rkxp=0eepd9ApO_@!SBp&6Vyz+-rf5>CjIHab~dsp`yX}6qM5g1f^3UgS<$L!
zIsfi%V!!xz-{avoLGZ&9IxxYc068@^qk2C9(-}q4m(o&2fWoSS^>-6UE4M6~>q1)z
z8<$@er(TDkOrytMi83;ph5gxnvz{Uu45%_9Qr!p1MxCRC6qrhH1e@JeC!PE3&I6bI
z@GH=pdON$J(}@z2J+#4Bj}vpKm}1Y)HN>Nj%;LNW(REVaJo*Ynx_L0c-<4CaCX;;=
z!Q#4N9Z_mHorrikwdg`+)Ri)&yux2&bv9jMI!*)HIZm>0y12gIJP_y3+22?bwaF7b
z5a(3Sw?+l?mtEurZ3>?>^k2xmFN`x)%2I3=W($AxGKKA1nPi&dr^ft~5KMnEDj5a!
z@Zy^kJ{KrOvm0AYvfAeBOh2yV)TeSxwS6NyX1=Pg+2O_(!Qnu^QRv!By^Z%qmj7j1
zKjYC=hHpvnsSJ7IcGKa?_rW-w=j}Xg;(1~X+Z>;iYUI%m@c~L~)4{Ncrt3j%vMYFy
zK?bF8z{T$B{VjQiYkA`EVl9MI(kp0Uw$YNh;|hIHs{eXboiq-luBg;+XFAAQTOoZn
zDc!s1gY4BaFhi&dFLvQrToUV7nzt>_oc}0iwRdN*_#ON6Fd43%RQz>aQSmg^G<{pQ
z7x4VGlPlj`6A@x$Ec8&xTliU$XK_7F=6;i1VUPGJCWl4QzR&VX`|U-our++AmWoqo
zq>7U=TvlAW$RXTGHLzcXO@R=j>(}`XxwhkH?^I~Gb*9{foUBQaZH#LKR!yXfNs)+Q
zg}Fg(+tVEHIF1BCQCn)<=}#Oz+Iyy?sYLZ7JIYUO--{Bzo0Z}lDwDQM%Tv+~?Fs*~
zxVvxD@9^9x6vX8-ymn4#IaVTiDv(nx3p6O#M-oM1m&EOz?d(kL^`ecI>ON!q{4wt%
zwEI-f>viBt<@yRnl!)&oBvA84L+XAVa_vv~Pvp*6Xze`L*e-x;)Z(vTgsqR@e6C|L
zQ!V=8kQcZ6UH>0-tYy&|a|DiS^JkC>aCxrI;v;X*7Z+}RF}9vW-D4yuatgoaRh7Mx
z%U6XWjP~41zqw)r9fr}aRNn6lwcek(tskz0-QFLB*mYKH%dJAcu|B@O-ieQ|{W??E
zGNyL@yS-hkySt`r{d9eViY)nw$9(NKDT6|-!s(qEj!R^C@fV}0(dp~LU!pD|HV9t1AwQf2v3Ez`km$ByzZ=VnGfuU*!5m{Z;p5{r~
zY_IAolp`4Q{sR|Dm%ebAD%x!F6rM6H$HBqarOdFoQh_`j6gIFf)SA
z^I+_)R@3K{NOc}(-E^WS>bJ&b7pw`m^lDqu4pUZFzaEv2
zY*FO^b`kG=F_9O^Sg>*-0BIG-R$mpdylWPW7c}`7nfy203q&3l2P^>^+s=VW``H$I
zvWrcOf;04=54)F>0Kq%%To+xCX<~63c#pCVNzq^d;T9hESQ{}rz|`l3;dp__?s|1a
zMVJS4!0h+hxvlVURt#&JHm0luSF8v7Z~|+cRP}c!_-IK2TI{l23SYCXfvcDm>`~wS
zWfkdA6Pjx0LBGbnjP{WnF=+Qo&j0+;o80=9SYE#e?%y2pq0}CSNGEWYNO#?n93Dml
zAMUrDF+Acf*2mh;Lrn5k$y>uTY00ddtbZ(RNFif`#gh8%g6r=xg-GeEC-ijfoe3NA
zS$@mfZdut-Yv*rBSvobsU7sQLi^R9%-)(*-sox@GyYtPeS1zU*v(X}PKvQZ`IC0t#
zxt2ij87_uXjV_SNFF&)B$VRg9=~4jQV*+(Pw!z^_Y%VwHCjGAluwJ)~AJ+oHf_FdW
zx@y(@DZL^i6=NpDbNv&nHFJOutpKFlKf3VUNy(YSeGjkcOlP%=qGG>cgZE3#JQ4MD
z+l}glWGHH}F-e0(2CGvk_zpGO7`#~2zxTcNSx@84s`H^59oUzYsMsa@vfW!eue`RT
zp+O2=D5FZ6H
zcSus9KfeE}H>Wzv8ruK!tI3>x+2F|RE4BIlQs@d-lo>J);U17iFlZ=lQ)cc*Mnfejb8V2qAS3#52iSAn4khwE}m
z`H23?4OhL3*)a%vl(>N7&AJTNRpugXaYKWU6Jxww>9;6p55|F`{*^y|uiReT6;}8P
zGpO@B3IcvVZLE-&%|1-z)O+SPiv+YN?uR2H4LBgD!Pb^ZwIP>FE;7~**RI9bdux*&
zs>W8pgIOB^dy&j0o2nYE3tJWs)gUDD6x;AtdlA2_jFqPOE8_EkJ(|pCY?oMTfd+F@
z;cL7A{%?2TLK>vq59(Dh{CXy3ZLH8_k;KD03R&{c%#YjJV9`)efw|k*Hl__Ye(j39
zBwe0fosd}tJIBw}U_;hFhvzwLw%%eeI{f+XH(J&81Gq^X>O61cJl+s|Y&9k*AFqLn
z_&%FCnAd@t=_v=_2?jL2CW9UTfCGWlEo*{eLe4zkbhkEJ1UAMGiA{i`E8Ac#`$2RM
zJ_d)&j7bw;XD)XvLO%JI3!!Fu*5!`B)?x-Se+HRGzF`ahqfJU{@Q)JJEPFX@Y&AKl
zlPy?H5m}_~@rr`Nnk!qV7b23VV_Y5~iJx8a3AIgTQ~hAp*u;TZ%lAbwovp8r+29$K
zn17`7LsXP#;4TOF`yKbB+R7~Ro=l4OqF@X(oCmoIQNvnO8v?eQEvzVOsVJtq9-mj|
z`OsTTtsgltV;JAPYrgR}MaVF5@|@Yl|0YhZ7ivi+F8ZV*zqiDQxB4Z<0qvMl6vhV+
zbI@rOZxNsFG%izgU>0&?s;0V
zvb!OlRBil1$5W866YAqrl^|GskU;I0z6)Y^
zL1b?{wrjvL+qs_}_pk4jrx5Oce3odMl6Vzz1Et=XjXnnfh!3_Ipxh&jx9Oaw=sWbkolXRk)G2Ka$pHNMg
zxkJlymVYYD2Y`mTBBQ=`rwWEdE2^1JFK3#7RdV|gxNYz8W1z7C+Gen~j+1vcD{>MS
z@e#G%8$*Xx?LsnHYq!DSS?vYQU;@PKiC=TJ-k4VgV9E+?B-=0uXhrgCoP1B&JCc;Y
zW61nr|9+nIthwWNzh%~iVgCd6DUWiB-MA|Icf?%%D}Vn@r9%hUahjXNV8`MGh-FlW
zXlrNuH_44LfVP3;WMIs`6$&LUEPvGUy0w?NU`eX)*BVcu1w%#T3kV|<=!sX=N*{W2
zv^WedsBGI?%72D`k%XeT)Q!tY|BG46WG&{ez-cllqfY)7-fU62h^=K*@K9l-l5S_i
z4gk6ny7S|6^{Oyi;K=j16c|=Q#LP9#)9n27TB5(B`JaBPvk*n4?97sZ
z~dqyMm>yJ9+<#b*MZA-qy*80w`6e-7YLJv$L;oj>d7k9Sx+S^ONt8P^Bp4d|?d@!(
zt&z|M#P@dX3}+g$8@s>=JKi@}&L9n4PNI??vbgVv@;%+6(@K3GbuG{>@Z>*PXi-m=
zoUQSf{U*{9-P%xwYdHdlv%Gj2iza5$8XmY9u%&AJ(czI4_-;yRjHv^{OF$~pGfxpEIIHKS2$|w{@-(EU@O%7)u
zQKsjkF1N%|Ag$Btl7W$4+p@3{Pkf*B1_pI_!31`pBfq~cEjZR214M<@UGp?t#4i#A
z?!!j7wcb;-d&~rh)FXhVXl8`AME?fRF5gp9WeE^5%FtNGVUIKAfE0
z1zzM^!>ODdergbd#2OSSh+{@dVss(~iE+DoYVxP?!8N#(wYBd7nUv0=bqc#IcAUWe
z@+>oPub~HSF=qS0xO3w6r(_39BBx{R=bxXjc2ASK%f7xy(FMBry`_j){@iIk`W&C{
zzUrLh`Ao<%z_3ybO5_Rq%F%gY23SryP>oN{YZ4Y0ful{(+^;o0N4zEf09rgKOfBH|
ze;SVe;3$+07rCXb=&52xeh6$P8eFucWD9a
zz6udfv5aP;t@x|RpZ{a}aq-(jlclbYWMFDY;~JroW;{?q6^JN&(Zsb(wW5H65IjW;
zSk}0YyTGAo+Sdk9RpNj6
z9!AG0JK?q};%t?PKf$AAq(b$?Py?{rq`YHSSa=1f2V4xuB0qLj0r69Yv|w1yH+Q4!
zrVI5`o{-4~9Bn<1o9j>fDz}pdcLYD7%5Ty?ky=_73fgV46Hb!$g9CCsqZ-Gw`ZpvW
zPA|b4d7?uec{PBz-A@{z($-UGuN3N?jiqB~_?U48AtMpHq_H6GC;#9{F8%i(xT?9P
z#8y4W?uK^0FmXl6rrfbhj`n)xC~W(LrYnCUZ`e1){ZXq9Pix>W08@W2m4{$phi*&a
zrD-ofkR*Y775yUVDRxw)JuzJ#AWQga=F>ga-qcuDmBHI%0i7Z+mx{r6HAjwCo*ED{
zE^Ib2PKM|ywZu$!H78Vn8B7n>DR0pQT-`*U%BCublg9#l<`ZUYl}sBW71C#!_3~xY
z%!!Qt{{HXlFl3gS@Y#gqRN`Y(m2I?M$vuVr;B5HB#1FIfN4EX}+_Q1J{$QE6C3SVS
z%up#r*h5?>9ms3L9?d}JLJoNl`q}8prr4H)0IX53IlzW$f(Pi98U?OxMaAsE!;E0)
zqWCEFlIK~nP`yh!(SW7M-Jn$l$i71&|KZPvZ=XRDL%_{(%Ii~~&8BCSGb*-vMX~27
zH^DtIz`S+VoM)mfNG6e1LH@jWdl;QK{YlT~LKvp6?4r=M!!{n0#>Y=uBnT|nuY*yMFKXvWzlyx)*
zBi^!FI%b>;$di*AQKT?`ChaMU|2j#6o{4Gwhv3(&DnTy)%4hsxC?%uWmbf>IdwND3
z30nY$%`UI7GAAv<_})$THU>Mx$_8f)xM3iPFh(oF~(4-wj^q%;z~
zgW;1K6Ept&x_~oUz}z*Z|A8HQ+Y{zTRZg1Kl1p%P-@s&L#l-uY2V?_&mPA@uSO^yp
z{Jm6Gv3wb$^TSj~=0Qe!PeaB{GYlgt_uXiP%k9j?Jt0Prq|U8(X}^tEudEN6OtuF7}D$U&Z!{f@~f6LbvR4Miv~i~$H?4Sr>G4bDo0k|
z?>C=^YL3X#juS|U{_a?~BY&0zUqAhjA~JI9i=&>b1qY7=8FIz{EdCPfPW;IWbt6$?Vt6`shEHc1EWh*v
z5Sfx;^^c^Dp=Uhm7?ZUSjyMyGODc2!p)(viz|Qoy*(J3M;ySwG!Sw@M%D=PW&i=Y~
z;O!}+ON;mw=Q^Lx-4WSj&a)Dsdc&{S*H-uuMX6Jc=HOiqgj^Qhj+X^W)8Nf{5{>8p
zkfe$z<HCsIrV=S4vZIS9{*`*
z&q<&Zemu%o0Cyo?)0NL=*qMejyPfmIgP}I#i+*;mPo;c9bJvSR?>vo#k!F2;r0o-7
z-}#~5l@+esUS@U$eiz1+c+?%@DI|e}eG7MoDGMjRd2_}1yT+8G^u-wv+7qUflznXV
z`3+&Qj=HwbBMRpB=^{R%jKx0S#jw%q#DXyL-Pv@$H+aN8suxhN;snFkp`cDaCRW)O
z{f6mJY(U?B@o*}86+4haITEytLQ!JCf%?xGxNKgf)5nl3ihFT{f%HzN%&z^Newd8+
z=jKrQEw!8dFy)oe)N~owC3*r*Ow^#!<2#DJ-zt1|IpWhkgeNSKu`y%-p=TdZqVbF7
zx7RocC2J-B2y|4d-`zOoa->lsrHYfuZ<&-lYN)0I^dvB^9%GzM8!5EWutjgWTSdNU;jtZ^l3KcNxTP
zKF6#bbbObuRXM|h%j^1@5y1Q2TeQXmc9b&o7AV&UdpeqhMcs!EVNdN
zujr{%Km5kVrxG3bY#jy{6YqOpBTv}(1`=Dr!v9luT&EH+rUe%S38>2xZuSS5$Yo+p
zfaIYJxT>YP-vR4Gav6OOfNOuW
z1z!h9_@lXAXIxYVBOcA{FZJzHQ>u`p7T9NkUR5|*2tYCKV7Q?S?goSk*gnW0vWK;h
zK=^do%<(O%VSy2CWrPp>Omci>LP5^f8q{l+!QR{15=f6S@unj0_1iJa^(XXx(xL-?>W@G^|B~7k)j6=$5G+lIG
zIr|?+^KHPe;HfqRe?>atSws+A+wrgS=FY*OfI3zGpe=Z8Xe73|pf+
znxbN%*=_+gS)DKQ3gAz%SQ$L(UR@rQr@Qx>j~8qN7WEl59!N
z?7;ArRbrkF{KP1IRh&6%LOv&Q@4suWnVB_9@F&#YQvr#c_MzBMPtp2#RStJ5^{$43
zs!Bnzje%{VkW9_el_6XVn=x^NE9JqZ(~y{
z!AH8krtnhOge8CAEp7y9XiqAIrgFWDR@s#@oATs~886aPb=}`_P#f2C2pF0Sf?YD}
z?@d#V$^CHBcvgPoJhtsQTlF&u@AVCH(1Is3tJaH2>^Qa4ee?u`W2ZeuNr)qnE!LGE
zko!VG#T5De7-FKmHO>Ki_M~4Z5kwn-Ls-2J8WrivpMfXBn1&7`2
zw?yS##^l+3IK*YqkU;b$Wb@#L8JCazG|=P{YUT`G=y)Is
zzU_CjMrla4VuyY_xLXoFx?D>de}8&h-_VxBgL~P-5}a*2ln%MQJyrB`Ro4Xk}6y%+E}g}*)6*J?M5I1Zgci)WgkteB3Wxt$BU
zn_Ha6Y_{747;%hA?8>`=0^jxAlsTi&xm=s7(B=yZ<&RtvU@pvDJ)ML9R1m@Mb(RBX
zh*%0lp3BlZ`y2(;3?B)pTg%#tos27Mz`x9Udxc81CyWK9)H74Ja!j3RD*iW(w54BR
z_nwL?JW!K9kxL%->Da{$pC?ub<3BH4jK~;$Z;f+z3rl96aU6NLqBfn$ne=oQQxmx1
zjI^-^dX{}xz={X>0Ukp4nJ%RM5E6Jmv4HrnZj3r9mT+~YpVVkbJ1hGh9Tu;84YW!m
zg=la?9-6T+9>7=d!(UlrcSudxCW}kklAdQy2>fX^VPE6~O(sXr%}3TNg3&DeSwutC
zC#z|T&Rz~eB8tqp@>t_O#Zy+P{v!N2CNp8ADt6!+7STIr)A$x-RIVFGX*=EYu;ASv
z!8Fj>$sf4yk+(UPA9lNcaK9pI(Q&gces$R|t=Lu(#V?Eu!~te{NsFLSA^L8abcZR)z60hT2(!MMFm9yf5ax-sD>-F-$k(h0CS51nr
zk?>Ww{Tr^xd=si(#~r!n`z>jQx5|w-!VNs}9`b0er(lgpE37=OOL>8LNe
zX}_8B1C>u_r4!%niS&as$IvFh{j$gH(m~J)yb>97k@LvZXVGKx>^4JEYsBllj#^2a
zcriY!$*l0$mDF8jL+m|N8A_SQN!H4Fw&XcS!ng$NsnBV7>6)bARe`U^(`|u6V
zBth8KB%atBMW$NGQe`rCo0g)UXuy4AD?{!qF8`TktaLy7w;tbzn~-fdXJ+p&fb|_<
zk-nPOA>Kef+!@T)I^gVi$IY~#pLq~BTGYS!76Wn365JAz*x|DXvvYhGNyQ*7P~Z4wO_
zCGCZ@6GBHzD@CcBL83FA3zuu>hle~D$Pn&-FK1JCDGDcgJKFvI7Kbchl^w#ECSaFG
z`r+guf;;$#+b##s(@3N3?2UN#38aQ6GLv;GqGHs_!lE$HGIC|NeC8d5!Lck96&hZ{
z(d|XeF~lBFiK8|Bg4C@Baa>`w+J0a|aw@|0m%2tZ8BX9?%&zK@7&CTr%eLqJmL{mI
zYd33^I}q7hPnHKN>%ZFfY!)hO-}E$Og<@`@(Q6Uhfk*I4OWrZ0^xNQ8FSHNI<&B%vlr4pZL$`wp<9+N_szf|*fo-zsZ=U{h~6n~;|Y36ZwA>_Yi1qL+Y7jJvRq
zu?HfNC;@om4UPv$Of^2JgAiSfk27b+3LCS<6TRWN&c5Sy1|q9MlPTiG{gz!1eMZ^>gos|
z@u&(3doKAfkDCuVaX-n#S0K$Qbzc|yHOEjTY;AADIJ;T#HBi-LYSknW52Zrxtk)k<<|ti5
zR?7PAx0N>^-@R8Z1(Bu%w;}kX_ncm?F5vAc;}>?337Xq&)!`ihd3Krqk+etbBcaA)wC#{c096{K3|JZ
zO#CqKaSRe;Q1*Tm=H=X18nNP~#{X+Ya*BJ86nP$m^kea{=CkyH^{+(HSSsl%WCk?sxBv?(U)a;=*n+5
zFpHbN$fKswphOc_zsJJJ+yA@|mc)zYXXMtHxSQ%0eq`bLEF9svHm>+RIz2sI6%kTO
zl;}I(6XzO!!u5y?TmYA}Q-Z0m!4O*4H^RdM!=AUz!LGVE_c{*FPAURO>~?@z-`KA6e$D#lYO%*0L*Lg*&!+Rdddqe-FL_T)5K{
z`SF?7jdZeV_YI2n`?JcEe1f~DjEs!8Q<`@i-2UULq_-5P*!+C@zz+TUhJzV$3b6p9
z%J=)ektDV+Nd&EoXetSh$CHAA6mP)owl8Wkh$AOHKK8lbkSFo->VSu*v|Ih_v}xh1
zH+(RaGNWCWQLjS*6VHI$MRo_ro4O{=F+s2psja_xkHw2DlqLDf4b9p0)kRHN=4-G%
zeYg#3WN~+)DQey|K)64NwuR0qrbZTsJbT^^&29}Q8L5iyLXGj!?hGIRE(u)5ke1if
z470e5u*he9WXJd+54$(JknnpyM6PXr^jl~)UUUD|2lMLZYXe_RUdak;cX?6cb{l^1
zi2JYR6|Q>57f}GL5Pq;fDnaEM{BAu})YOw`C%jUPq(`YkQ_es$7^v85r
z<0-$(oejXpQ7tYtnf+IC%xleDa$Dx8t8gRrh!bjDgn%q}q`g+^u%C-Z{a5I^u3N^o
z+^}7C_z?Fh;L+NRUyjCy#83Nj?^Rk}2jTFOv;S5eDy-eewdBa{#EM7a<*B9_
zhG!ybJ9V+H=l7bBxZF>mt-8SA0W#bgM#RorTP~gHf@cOIg_;f2eA#LJC!L;h2ThGw
z>QTM9xMWUgDr`lmI7`PLuTs|EsNK$!^`CGIr7OFC`u7b_Qc^av_)D=uKLU6~t^xNC
zG{&5n7ROUEg6MsR5HfYvBK(*;8j%P^Zw-h0kvgJ)a*@I?W&XLhCG3{DQU!CVE$U-3BoEIJiUbG-L(H0N(
z*Ecsra);g0f%A=(Aseqlx-;5OGiK|Ba{LfZ=KZfNu6A(TUG#t@Fnt6jM9tqGb3aHN
z1Rz6Jin)Vc;qYR?q6<*36b8c0S8fAwSPGi)$s$n?flnN~)l=$_C8=>Jf2q9SC7UM2
z$pJso>@6H9{E{Dweo*$yNSbs*+J}5FAJ%`tTT%Cnz~!1}JugIi`D^qrQn)n+C4uTY
z=C!yQL{_;W&wd!yoQIlD>iX(e@R&O`o{StEqC4&_E1K*qHfr=BH@D!LSsVkEG^u$1
z2h&PzN2FXH>(2G-k(Y72W{R$y705
zAaxHQj%k#7aE<%$H@9@?%-4&0?-Kscj7gxb8b%0rBm=M0qADT#-_-`71gX=jfRST2
zwNIcmmDYc^8y5GX7#KrX-5&w#w43ei?KU1V-^bN&Qfq!C-OVOx9?u<_$ez$6_TMrU
zPg8J1a~oB@6-(%~nP1>ZQIm=jPcofXC1>zhN+XntkxbPm60E-
z9ZMID6}kl%zVy^;k8|l3nj2hxgBEmg(9a$>090DYDsO*X;qvEen*!=UKMuzwX3)=-
z$zA%{IyAfYUNZ~`T?CpEM-0EOJr%6)4@ZUQ0U{Ynv{?p)d
z`;b@x^*pUevYJj}EiMteC28rvOG4SPm1cBBUKEE->%g1*N%;H7mFAIxeHWhw%aQU(
zoFWI67Re=pn%@msh#jPr6F8sJMu?BkSRd6}ZBQAMi^Mcj;dO(;f3P$;Om2!=bO%=3
zkDC9>0$5|KT(Y0hG$skRuBZd|;XCrDfR5r{XgM75xo1(ET%WBj{9)d<`I8ZBNU>=u
z^Zw^8aMXG$ZwI-LkoWVIhj7$F)fZ$w9m4AGyz8@Y%f@A5>P5axi|AKx&n-$P9gWi!
z3acAA!(5Z+rid7I9O&%j7hoVt97QM@2J&vd;Sv>cuXKHV&8wb5d%rS&IaL~*tyq}R
z(6+aCdzJTqt&~V*hk(^lNd-m=w{0VE+P(NGIn>*@8E@C5gVT`LTi+To4BEfkRD$&yZ#V~%v#18Z)gQ`vQgTL{7oB-|
z_r3VaH?Q=2ab%$+yjt-D>*>CGqC~gE;1SvkJgwl{HQHW?X&@3cmj4x--Qm$Ak~;7k
zSzRT+gMg4&03OIYY3z8osXPP_P99)dP956s4vlPlH*+{FPgL_+Sb!|FV@k;N|Hg>TlAW>%NV8eFXUm7YB$v)A
zd`B8*zx`#37(@&+ZbIrS~&b|5eC^F=F4+T&(Vx-+BWHblh#bO6UICW8+2Zv7#J
z)%Hv3Z=`N42}|y#+r#-#_{uJB5Q`j(q&GPQ1wwx8#hHJ?{r!@=JmB8O$L<9IyI+Ik
zgBag?obO^G>~O~PJVQbV+7B#YlbXgYu9|VGP^}lhToDJMv%32E9Cymn%F1U*pV6%6
z7V!z$$Xmoi(nb44(CzP}*0}B{0(Vpn^3_EqP{B!CP6NsG9+2Ki-$peYb`sq0JZY*|
zeMuJ2>ejM%er#R*A{Ui+l6eq_!uQ@^(EGc0Fy|o%kl@1HQ%vTotV_x;0Rn2VE+9~~
z{N^KlV^CTw@wmIYFh0?dKhA%#w6wJS{RGw6QwooSfE>I|17019wW|VjjGt?Pjn*~+
zUINQyXXj+#CV$@QQosQ{n2&2667W4CoBbbL3fL5reRP?x_t+YfH7@VHx|1ZS`XV-X
z(RLsSmke+v?Gr~{ak(~aC_Dm!-%-v%V4TIn83C{XhUiV!S*+pP
zg4@}CDc-%AYy>my&1}?VQ{GK4_(JyP<_5SG-d`bsDjq0OIo6rXNy(|G*2zMS44PIV
zQ5u&j^>jsZi(71=v$Zb`^Fw~@tB{$
zz}kBD8aguSoMw*dD(@J-T)1h%MIQf35MJu}gmE(@JOVgB{yNF^If-cQvB&-QZ`I_W
zJV(A#)up18zUDq(`Lv{Tck`0s1pvhzCx6(}r`N!G{BC!Lds=5VNxqcM8>K^lO)4+X
zxKVe0aRKyWjfAe<4YNE%l#!^F9(zW#Gn-ItM`TYe+6(OHd^Ek~B?we;9&V?!>s=sk~Vy7lei{w|5G*rq!&0j1UkR9BS`*Oixx$UADO`y6|8gM{E7Q^YcqbKXU!0*E~~`q?&2zF{Lj&q!Cflx5!e^Ao8Yd
z^pH{dxB@58YZMc%3NtFXoXEI`&a~}&WtMoD>n(Y{S-)8dbs}f65`)5VJ*FK=s~bmc
zpsiHZ>q?@J4NR+@^pgU1tK|ca;_`qYeig09?5^!s=?d*X?VtB$d-p(wm86a-sicGB
z4LYvt$osC+{KQn-#CO@0zv`nEzPS#gt;oHG(?Bbx7V7i$TYD~Q?0wR-&?(`P0>Phe
zprZn&&^Myl+gED7Lngt|i!B$xLq6~uOCDL-K;f-0U5!kKXU9PB{v;Um-%)hO_uI9s70;lww
z8!{gO5=-vDWSgDCp(sW!MCTylmoqb}2K*(!+uFUfaC}^AzwLtS
zez7B9+RrvGtlrCmigLdGvvw-95ClHXAKZ=hZO!C`|c25jF6j#2hWkU;8m93cuKcGnmzAq
zV7P!?Mf)`@?n50a3YZ(l@J=Y@WR7`EG63RQEGps~2}Jw01%GroP*;A5QU@SGjX^Js
zA_wE=C1e;LdjX_G)Z77cLw-vqt}BV+`^!LaLXmep)!(#N?{m_IyeUy=8rgoD
z=Z~jlu-R^5?d+Ti0`nk_+-`i@(2KiEZBcYsmn*cvI6ypcs}8=KP&9p+!KsydupTIM
zQ>DsEClYimAbmt4-3Ppse4Hxr@oh~!*0B(AEFw|Y|a<`*!@D$dGXO7Fi#V&;Av^TSoPcI`Tn%shvF&&?i^lTnudt9w6*ZT1ZtAh_eGK*=Z%
z{AXIOPD>x2DJ20T$TX=@iwF0kkAQr$%9L{dfOh#JUBA_QI8E{eHjALE`+cI%?W2%$dy6gz9DHFp8dSlA-FN#odbpGYafB~fa}Y3M
z`#{B7x|@LmmYT1sMF+*KG16Y)RkZ9!YKI<;KHN@&naEXJP?MJK6To3Co};q7Ij8oh
z=gp30*dG6>HX*=S2ZYy=Fi?iOE7wa?AuYm;*FZ9X9Hd%n3#eQHlA#7Ltqzqns&ZRHdcslSjRr9-s$Se@d>sEKa7JGDsLL%S>EESNXDwt$-?_cisQvO2LivHBWpks-oT?#-$`20k9eIPG
z@wmEvJY#pEG=omyk3?|?t`H1jrT$NKaec}Luz*OPeD0IbcMdy2SypT&e!Oqrzry2nDnVi*YJYot1B;i5NQ1qB$*tff^$&d>~N~
zn4Z8dQQ|#fozC-8z`-tSSvcf`QQ~}-$hG5+PsRl;J3C7}_fj>r~SVh{$i5g-f6oXgDQ*U9tTC;9Go&-$MCe#dIt
z9lM?#k2A<9xwCAwdO}Ggr$+vc)p<*85Ip>LO_JOxUtp-S2rq8t=x0JKA-+?Uk6hI?
zpI)-UWX~z*2hTOh!l&R&&E*>OXB`}sd14kt=yT2n=JLEIB@g5I9e~s*%#-y7O4ICKmsnH
zbb82^R{%zhJ})MBzgs?^Fo~u`CXz?tkcs|w%aV_lP&A${S(hp=Jx0^^BZeo!UbiFBcIAB+a?;b+^z8Wg*{ta|B{L6@
z_aI8;R#EYWSZL)STj;JS9d$)kRJz!uk+I}C7$C6hwn#92ef4(V
zH*<41q41f5p|jJEU>I<@2mLB%3?blFi^}bNjN*!X4cU8p&kh)&fq}NY0`LRBva=e$
zuJ0x%N2rz}&6Lt&-YuP_+u2K1fJIjx-0w4$~_S*p7KGo<`Ss|8mhG>13*eY-i~5qXVX5ovN^=_TX2o;vu-$&ByQ&b
zEkflXP+#QI=+Xr29ZxKU2zARpZrjW*bFW`dsV${N0p{1{a(c;ztf&(#9KHIzvX1K&BvIyBD!4_RC2N_v>|Xo!~;IyHr{Y%89E
zJ>ody`xr9gmsrw66JNBlB4eKk!QgW#pZ@DESqdS#7eYQ#{x%ZaQb21~RJAP|1s2Er
z{P<;)g3W^)zQaT@MO0(rS+7>w7iw^dfe;rJT=g}vYF`(G0$f<&lZumhtfjF`jk<(=
zKAXIE(->3hW9oDXqx+rL^Sff=XZ7f4t
zArle~e)0X5+;B&!R!Wy-cBI+w&3toU`c*k&F{&SkRQ+Bz+Jgb!MIr+v0sHN9Mp3TA
z@cjGrIrDR0KSJKjXAl&K55p@3vL
zGKTCi%xcN$m~Npc{7_4Q)4AC-D<~7Jo-yVz^ZIpcdV0a0VjOa70Jb6-43fvzE39*F
zv((&}UB1Q~dEZy7y++W8b>W{yYb&4o=)7?ov4MOXWgtb0&s1l!m)QEnCgn?AJ$o~i
zr6uV#JWDIT+A#sHzh@ISbI-YPqnvbSH)K3EzVDckfgcUj!
zNj>4hF2^$#FPG^59UsJ9n}=B$z{opK+n`vrAKN5iGaG&B4m{Vg?8`{j;p58jM_!?M
zZLA_5oFmC5A^Pu9vMolbe)wLj_-_wd88iKHKLDIX
zej+Ay$DTv&(%%mK%Vij_(Pd9iSre}j`k}eoqf({N+tEFjz&9IMy#3IF#A7XuCc}#o^m1^PuPYL^jpwGZwS_K
zs6*ow<&5efmg0bvTf=d%|ChjGQPKZ+(KcQKD%zn=t_(y;F1k=uT)(RDAnve1Zxfu0
hTimKuq2k-i7VD9U!$egq>6{gOLJtvx?;bdw_aC(Oc9sAD

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter4_116_0.png b/doc/LectureNotes/_build/html/_images/chapter4_116_0.png
new file mode 100644
index 0000000000000000000000000000000000000000..dfd44cb864c330522f1b312d0828240e042880fa
GIT binary patch
literal 28657
zcmZsD1yodR6z!-;cXy7oATYwvodVJVQW8UVw{(MwN(_yFfV4<=sg!gM2na}nbia@P
z#d>e8xBicT|IVE|U!Jq~KIa&tp{|IBO^pqKK=726`1uiUI4YOg$89GpEIY|R?jpRrj{oNd9%nae
z-tQ@O8F$ZeQ8I9cKyXOze(piiGbkYtXA@;PX&s-8ofh9O+_t`pw`FCer-8yK>$c05
z_JJq#xQ8`ZECh*|OxWF6%I)k#lp3F?xJhuH$ud{Uzv;!dQR2oC?fk*sS*_X5*bg=9
zrzRILGTV*cNo&|@INg4>cWI*ACb>|uSw0r{(f@h2yPvOwR&mLcc2VQyb?fk9>)zoi
zOl?ex5KWq}Tca!c1ncfEG}kB4>AT;2_urV2+`WY<6{>dk4hxtNa9uhc^1pA0RtyZA
zn9vnFop-NyTgCoXSGRO}c`&a1+#qYr2DZOlGt5jL_sZ9ois5p)ud=@0XI%2XCy{i$
zeJGkeyn@;b5?VMP^1csQxZd?%iDAoXr&&BxI20b1ycJtMfJ$g`7(BvP1ZBc?qyaONF7`e{*=g*&8AQ0$dV`I^MD}+5Ef_>eT{FTn%
zlY*}-j9UE2hQ$t5hb?k7Fh8jDG#m4z5Mf8)2MimPi+moVdC5#fpjA9-o89}?rq*zJ
zVWG3PR|NFr==$!ps--1HtN8YoO_|udisNd2xaDEhG%5H~j#*PKUp0s2Eqk*G13dA;
z6Z)ymPdWCU@P*^@(`8wP3qq}8V#6}>(G60qV(7acbf2_1wznOa(&jr(P6WNYzExHK
zGhaLo5!uPY+I`o>;D$X
zl0Rl}P>q|1M}}0s9_1u<&`siVzM66{ZIQGD7P8)BV>D#sDW3_98qdMui$!3d+M?dz
zN@k|x5tprV-5gO8+kUT*M7)n!qK{YzXvM|w!oubK%Ll)G4gX5vV!VTxdJ4L>4-z@mZQB8@aOTc4-4#4B9kw4j51+uO8j>+8oW
zz5~OJ7d&c4C6;%NK47Pfsc?uKu%`;h<2kf7m{*`?pAB2)mP84`+3y
z39KJo8`Pux;~u(R06PSk*61A_b)0W-r}5sM@5~gftNQ9eGg0EXoISSDS@0~yb#1_a
zn>&Nge^*=EmHN5FD;-_=CX_-GYV>rcd=8tm08d^sr-YMKo-avPH=?lk7felZ(l3%r
zj9yfwK)t)yRm>-AYu5d@@4e7=Swgvev5G|m5*kA&uPphIF*b7!wjko;lO?q7)x7%H
zxmHZCOs}f(XSG?c*Xb_Z6>701{)I^c!q-VGtG&UU!ew{fH$&9>-uiI1A~TFZEHL%%GMl&ZTaIb0QQF~G0GRxfWEQCrq$|+y3
zG*7QI*RTw{=8b&LH>m(uaz~z=;K8Up!2yCH@%mdQ>~d!u@=#9TH5cWS`SaF|6rLD1
zhzNtuFe(#6l|u(vWj{Jzm6Pq=m{3C*0Q_K4=IiojvsRKWQ7bOL%uV1gBxy+Jjp)Ry)-{+g_s39}kqPH%^rakEGwKj2d9gj
z9LGiwsiAuuQclq4TKKxjhzFj@Q*Q*CQ-9Um&WMoLwhNq9#O>=G9WYGTJjRXY4U);q
z$mPIG!MU3yh8z0eQ}2q@VIp}VFacdXaVf4FqgFXNh`Lmbhe;ay0GMG4;ofyT2#PkO(g)+YQ0=L3PD(!h>fiJS+cJg?|71AY|@(
z5N|+;ATA$c%rRKD7+Aa-tUKt$k!;@r105Xn?6+8QK}+E=p7if{;Q}TnTxLmIWGq=x
zR~L-#kDI}tL*cJ;Prp<+vt$o%(2#6|$)&FQw(y;{uN-vlcy+CA%GvcUYn;YTf(&AxgWkCdB1V~5%>wI*K^Og-c5hQKlH1n;M3Cm%OdrIc*O_A+k9(v
zL0eM?VGLIg`Xq`M<<0j}cxmRy!n;zF8C++&u94
zeR#ffeBpOuQbA!-K>$9gUgZ9%=Tw@~sw0RH>^hEx0Nx~1hLTmAan`Y%%ly@f){
zE2}&s8UBU@OE<;E_!Y}Vb3(CU7D7yBQf%awmTYci!+Agurc8zHz-QCag
zyIN|u23-EMuW8rHMWTHZb_IL>=hydI$b)bJZpjGF%)b$Q&j
zZceG-9RNPvLqme!>I@#QW@eVPjzkY4WMR~xyH8n&zR3fB#fW({@hFl)&_46S^p0Y!
zAzy-wmvfzSO^h+hg*FOIKR^=ZRrL*Pes-MrNqwSTGf(p*$2W-{6inEe`}@VSd+Kt8{!1Q>PlVPx=jUaHEd~+V-dCHJ
znOOe`Xym
zJ{vg&1>e&Tm2wzME}xt?(-~Q*kb{WyrR_ps6Zv%_`=?s`pXnPmI{KD(Ep79U@o|e<
zf4=&QDI2OrtJGpLmVlmmI5|6u4d8@7r=xJDdWN_^JpgIT2exB#-t*SI#NEoEua)&a
z&e73L#M+d49g-XCK6rS
zH#hkKD10`~9ZO6cgH~8%t;RJV7pmog=4z<)Zd{1dxz@1In5{pwYNvMU=3Yh13BlT-
z7l
zLHXLF%QeF{TeH){n?pNsg$Z-gt>_C{NluC)S>w-7*a$4${787y5ZkA5`4Bp-qbeIm6%8mX&rp&!5fGU%gv8$Sq7kPELfcSgais`^M}!
zW=TmpN45WHs+6-+`Ah6LIITJ`<^`oGEw7crgL0w@=6-jvr`W!q_%dantY+a%89<3(
zh~=+8E{@cvUjA_?2K|9{n6ANv|2&`9r^ld^p
zNxC@h2hsMDb~_2-N{d*TSsAca?S=KP?iRBw*uXsQ?O9HF#ntz(jv8MKnYkX|l2iP6
zcL)XNP&>9Kk7wvvy}vhBf4}!*`}9Z3qyB>q$ZS(N7&VLQuFMz=V1=3ts{*Y_JaJCr
zHTvy(y;4IJe)@2z@*68EuO>Q)Y*Py6nRxDmz`7q%L>~-KTM!IkXSf(u2)!|#xIt?+
z7PCR(%CZ=R{A2^O%g6$4!0Z6Cdi&#(KmUFISCnCujk`Q!w4dDP%2Lv;kea7{#BZ%;
zV7Giv
z-&PL7&+`q9#KPpv{qll!zK}jDePql@rpc2sF!yB=M)Eg|*;JbFw};5mDaP^Va*InF
zIsfat4(0k2+CJHLXCV|<0M>b5Yz`W254lhlGOoHc^kEh;fKfWtVeq8PePP)1^4|RJ
zFJ@A2R2~yEdl}=?VAT1KpTwDkz25Iw@$Ltr|FjN(VQ`>K7Kb`k-Lfa!p6F>5)DW!?
z-)#F{o%Kn6BV}+E$@Fth@zv+$&IlRX5b3EAGk~edwwTD~zL|QaR89RWLE&Ms4mxlL
zwX+Kd0!Gbyj(iy3nqfRbadNWpyPgw8mS3-=F-IRQ-YK9
zRT4bp^@E1_C9!lxJmANQ-#^ATnsJ!@+P|Gr{t;lyCx{Vs>%G1-0Ri@uISVQxf^^sT
zN{oa&cJ`EKzw24`^aKYT603MW#0HwKyTH{HD~8Aw7zt_5{g(W5OE>{d=1;qMIv#@$
zzKL(;iL!z&8NeQ&#%wM52WannBy#X5;l0&Nq74k8z>Ok`AKRGKev_%r#z{uZWc~y*
z6p}rhq}2VOLTL8TrwKLrP|O|?8t>@uU(?q{4{0`86tyGl!9U&k-!asZn9B^v3^@yT3Y}p0Qp~4zTuSu%5gt{9CsasRE;sC0j
z5@WY_!-JW2_uH~%AAZ!~ryDwROvVn@#0b0EP)j$A;_*aVvo}9J{c@^%JS+JxysN#rT_EC4Ed7)GpBW?@ZEX=N5y3yu&$>s0
z-9KK{iKMfMW{drwD<^gqT@D
zYi8ni*J|z%>wh$5-brXEyUp_C*UACfwpSlVCf`Vc<{~mJxU#Hip{~SEL$@cur0zVb
z%2S(DSc987jikH}@@Wq9lfOkIkEmq8}R*
zi0(@ZEbLnm?-8Kqg{cYH+j=*-^SvnFHJtl$A~@%l*^v4pl3(NSjWqCkwhHFsVnQdq
zuFjbrM0xBTh-5V(US)F{;T~~H+9{#~q1J;}kby@$A5_vOkx1m0xWzI6>332{S{jf-
zXlMRrG3&ssx`iD=;tb1Tfh+*}WEL1mHaGW@T(jz544}m}&
zGc0f${5kF*T(+wDPD_dCUBe*S)8A+g2O3rQtMqP%bl4023r1FJ=$PVI
z=xQWirng}rw1C&=S~*7t7(o+rqxyU6ni-~kZfsKd7_As%7g_?%KS?lc16|!`==l~s
zGc)n=ie-ATRr=GX*V|rQ&yWVrUfjZI8~xB5Cm)k1)F)F}>3P8^*d0TDNE$crHTM!Q
z_Lq?l8{b+{In~Y08}Xzxc#CAZH1tjt+h=H!KAhm}|M;Y+Ja*UHw9DX}KQ-4U)1nGc`GT97yhN*5)XJ&BgQ$d~NAT)YiByAJwu==;_1ad=nD
zQ;(QSHsY9gW9YcuQ%&|ipm~NME>UC?5OR*FF3lPu7&A>uKT1Hr=Uu*nxr+ym2v?Q;
zwaHUzy!ZX`;{DoB#J}Q1(7H7Y0GV2i==A*RLIP4ZCoZ@(C7d`_TB5dAInPM-o$Ttz
z^Po`Yz^#``1zKlel4lPB&bkDeyGpqYbzdJ^8p;&)IhnC$
z>erd_8XbY*t-4{I2ard|;GYN3gVzbW3hQ#9BaZ9bH{rD(#|(DBH
zBioa?y17X!;c5pV8jk#qe?9>ZsFE2z*gwVnRVyN<*rJzSn5$R1NmbCCW&AT&pZiEP
z$h}Ft@zh`GG>jl2K6(rg;EW4OK|bH{t*4eX1I$m5YAh-J`mwx_*L^GUj8DR(7%{VX
zb9j^T_@lcmJ6vfq)CFhw==TP;GumxE=AV*F%zTY+8X)2N}Faa
z4sWYset>fH4PMn$B}@UYj21?hop(8Gr57H
zKg>vp%S4I|w1E+Q0cl)F;>wXO0sUC!voRP@y)0{NEo-R6^FPQ6B0K2x`q8|a9^?{|
z?7P5Kyy{r6u>`Al9kv^yJw#$WL?S#yY^$dK!105ErJQk@$B}H&BzIezd9Ab40K&cX
zq2CI?ron-!-d;{YLA`sGn^vmLK!DE2j9?})=y4neAd4HwAm)5YI(B{Y)CrO2^-$&1
z^mM`Czadg@0Kw>f!-RRibBP@^$VpSV9v4E7J5xQq%^7{mM-jwSn6;zj^4^ot$S74}
z>6DWF&yY#y(9pL6kyJozx3@#jS5D6ZENUb?be^d4r&iRO3-6v8my+MwcO9nFou0NM
z_R2yUTn%p5P)%a9j)DWP;J%(G89{AJH$6LUH`NywG)}JuxaM34EfUQmyQY=X(p5QW
z`_evK#~rF?!sr@0q>ox^ukjTfhZptCz3{&r#W40@#4j;LXx&la>U88*$A9<3
z(URZ6voC?B0yuU#X1;Fo0krOwgHcb(hXD3K1UU*`mzAkGI6E_fsbprvybn~0ZgW@A
ze8eO!m=@SdN>V42^y0o3+R_xWmj!0Fd0Ak7WLF`tuxCX~ve)_xQsq!Fq1-ny$Z@52FI~KpS=wDD%cFiDSnL=f|IK9HXd-uB#hH01tp?23zL+
z<}8NVSff}P&6?{eqW;s)!(kjkc`WrJ)UqRr*Uyq@%|aB_+QZp8SXaND|o1D+^7&3WVWYEt+1nTGo
zf*FPO%M{ZmD+)UE+vQ32pBCmOSC}{4X?KbXDS!U6d^_^(&R>u6l8GP%K3r)+j|0Tb
zsvy#MvF4IBa5oazyX9qJpwNdE76V>5Jv-YzIT(o;yO-h%Zk|(v4gza66_51dvkLJcR>lHr+eFr@By@Ggy6J-
zg_E4M1M`HB0`ZEWg4P%e2%a!xWVF!HO5}p3z0BA}(F)$dM=Mdw8GK~I4oM8*=thg^
z{bZ>YlP)kl-@Z)(!V;lZiuJl+#-a*N97EN-p_Zb-!w{rzqcu|jB?lqSN7O9Rv1@7ygDDE
zxJ4QJ`Ju{L-_ehssFs*9gy*p%yn3fe3u06XMg<~yFz;`B6}1Y3MU>9|nrK$6)4jGP
zvheStq6YVgR|Xar*b)O!*q7e!3d6Ec!!nkkrp#8Ktf_qDf)Yk*y!})5zlL;KO5I_W
z)t_;E1_w6Bw{&611moc~v$`6a`6ZXM*4bnY*7qQ$7&hM+{Weq$zvS=zK{-RsFU3mA
z)g3DbT}yxEcuDz4iXQg4A}~hQx8^HKus4s>XMR}vv;I8~=ncr?p7XqCU2g2v%2u5hcf@XQxetN=^dVE=0MMwrKt6X@JuJr4kgsae%Xv4I}ly&
z)+}W-ax8y&@Ei7Ue#4^+k&nOeNw$pM)ts_;$F4RWG?jvQKa^b;M=TH~f?s@+URI5#
zK#lk?65jvycsX;3EABhjcf`yVnl!8j=~y14R6s)a`4Ken1EMCxtsz>a05r>(EaAi&
zm_)B1B9=K?xr=l86~y72XN!q^t>K7h1KE3(h;pq?TA;xKj|V7ohPn{eSt_83xOW5%
zJOVt(tsz{mRA57aP|m029z1g({B`dJrn$3E)}BgHA0&rO)=KVA@%YD?dtMD%;VK2;mTn|s!&}Qc`HPFGAm?J90$>C{@$VZ;0Sq~n?M7Qvv_6ff6D8jqBkCB1HON<2mq?{Qk@MQE`6Dve-1~zdWAXtx&E7W(J-f`X}9E%zmL!
zQL_iRtQXDpT|eCOv9AfNTRd56sf$StQB$$#f0RFhHoNnUV`0^;Bfd2u_i=>~q(Vrl
zpEC;c0?JCsI!Z>0Eli4SFvb|R;^WieU^K9GtW6pi`lL!1$V4zTl;@Y7+0sHdO{f^V
zUF~xFLMhAdW}amS=d+-iIR}f=1LNUG9g;j?h87t@7a92HLe2S{5tCpR1O-1XdHF9L
zK2&awf;3)`pd(>=^0{w76R@W{Y=*Q+qJh37^*!4X#!yn!A^Ti{t*e_;qLM;Mb6Q$2
zIkF>H75U3^af;~Ykb8ns!%NJ|n7pYisjBI;1LuVHCKhjtisNl>Z`SEs-{8#H#Auwg
z+wX^BJ|@3*)XuBE35#L@<#`m<9MH(6|u&)4O1A7c+yF;48CBZ~#
zMR0ijC2e=jmtMkDgCY53B)kfvJKBd#KBVNEn)I#%1`c1}0U=uM^Y=4>NYSdlV=D(a
zCX#2qr*XQ!3?g$xk-vDG+OK||o@xq)`|l;R&Ygw<7#GtG+K7YT5q=Q)D*v-a?QyFe=Q%A7Ce-RUn5
zg0eXGIns@wC(<81?mfv9u6p*q_<$OXwvR3~bSM7yr2wr$-%r9L;|2c0RDpXGOqikP
zX`}1b@*j7i6-YoUT+x@vo;am)CyQBRi0qPn)`LDMg(mtzWvO93s|(~8?H^YJx2(JB
z>K^p)1`+Y3;Ly;}oTc3UGhKeGNKXZ%hhJ%89|sHH3|r^gazffSJUaFWI@WPUyQ_>V
zqZ49WL2kmsMk|vhr&N8VkLalB8dnQ*O0zDs&a91@p2CWnLd!BzNu8f9BO`ZMUblc}6@dLcxbZZ->_pDsGS4dI9})j6F(>F0{Q%bK2ab-t-idR(fZNcHd#G
zhxnfc32I{isZ#*k_JXVq^n-Oa^2C+Onwl`b7W@jK*Bb2i
zQ{sBx=;%r}p~fQ`3SJ<_G`0%s~`Fe
zw7Ws~*k>6rN9Att6CMGp$^bxLgOS4Um6b;6BZvl}E2%p4`P)Inl9xB`JjzL^I`YSs
z1loP|XlD7P%Ow}1aF9{~G8uNXg~{}Fzo7%I0JP0xky%H3Im?$Pe*radJX2<3p;syOvy1=b(iVD+i3Q9om2AVgV9{F>X5Qe`rW?
z(Ls=7aBkHNSk(RYsa3a$6|u4^M&bj;51jXGXwT+drN(psfD#XNN*}4Y*LUy+N)>80
zmQVMC)*2sex|TEL)RbAyBe){^`YoRTi2}dDxPa(mcI-oYU0+48!vxoC%9G0)n5BrXb
zV)#O_JV?sON&ccp!hB4|X{09!M+Z!nVU>mBbP#>$P^4x0+N8teFNevasr@zY7w=fr
z+G8+YF@|0zXJ4T$UQ7Z?Qm~H^ypM6VA0-(td3z`c5=Vp%MB5HU&)9DNWdA;%GVKG<
zbI-!vo#akRQiHQxn8V!4^d6MycDMh6aQN?Y^ZdJ|8L-gX0~N
z2iN{?>{B!2OEH{&>q69wIol2}b+DqxGK~(hB2_J-VSTReH(8@GKKb8Ac8ySJ*$Hl~
zQTmkWg)=Hu&TLoC&_Wk1FhOT5;};~K&3MLnu*ne0(-bsfnM+kxISXxj3F+waiGEkD
zfBAVrrO`*!R10W)c^+}@z~Tn4KYcd-xSxl5i(U(mcoIF`Vsmn92n0F!N?ngh
z;h}x)Qqaz>w4-bO2xNHAxYoy8dS&5$-}p3v=)ltfjHs@BMHHF9=9(`NU<#mP;r`J;
zRruv3qzU%HS0QA5XV4dzpg8W6r?A<Ts6
zSY5VeawtKhEW=iL&de8J(DU@TWIq~L&Skzgpt%p2Gv&AKe9Osyf-20T(D%s-HPJkM
zwt?N}uf3-q#nZWNw+8PC&CoGj2H=xoXR7v_1TX%D9r@E=nqIVg6LPZ-Hl6d_o#)Xk
z+WX^b{&=X9HBFG3Y<%kxjM~cmMA&orIRWuDR@Y}ay^0jITyl9Dup<&)(s3E4y!K>X
zxd|<7%X{XbI{{zwqu+W`Kdk-^YGc^4ZsaFCu(;h^e|fu;!c=J|z3$ocFthv0XUX>-
zeiVjR6V7oX)4bLA7TNTU`uA#N_$4W0M;0M&Zux{tVc%EwpDyrun1r~-m(J#Fb{g~i
z53^ZQzvlB&$D4-olw^J$ERb}Mom7KB*ch2TUZf
zTnLUfuZu1%L04NDXt;kQyqlz3p8)fv0xT5FkRUQ_t&8>nfK1^acMHb(dEzGoh8c<(
z)rF({?)l}o!t0iYvxFf|A%*CHI)5qTU&?2!-;*{weQ=GIA+gnB*0~eOEvhzBu7?9A
z#YH6W{U8m<*0zu-od6UeyyP3`GA%wYnK+zqQr|lcCUsx@Mql1YGgad-Ao!m3vf5(A
zwOR6**c&Uhk_l$RG7*)6uM&KNvMX~BLec!YL9Web9XA`n8Dd)GAXZdZ5aTt7r#ruS
z+Tyw!gsVYn_nwX9#DHeRqwY0LrZDSrx-fcq32ESI*!shI)IIUn?5U9~V36r+5<6ow
zzDkvflxky@)z$aqH5oC9Gt_w#v`>F2d#0#7QquCCB^lo*4*ZK*U*a|`Ze)%r?4KA3
zTzul#=lrP%=#PCrmNi1-T0<#8uKPJB8|*-)c|zOAM2VZHWjhKbz`52&LF7le!`e5f
zQimvgPESimY^2_aSS%i0n^UgL2_hfV8e&Og6Eqm@{H7rGm+>7VVk5+q`e^$lU0#|R
zO*#TB>fI1
zj=eZ1Wu*ncDNhQ+vS#cU3530qRe^qx)^s}yHtBR)*jTJBi%XCH4l1qb!>+>SIdu4%
z-K)2mEF#Zo`7hu>ESUC`cyHI>=`aP{2btpyc^;!{y{(6IR>O~?FC);9luUm4wz^h$
z(5Io@E*)vflZ`J}HOAFw9}+v5|D9m
z3hEkrE}1}K8bFjJc~igYYkIXz!3SbmaS554}$
zRl)dHsU2ZK=0G2Kq}jaRi6b6*Jy8)vw&X%tYlzo&QQC&xwipa*1LCnzGI*7}rKrNl
z*i`5VxfAsYD{dn9^#7Bo&pqth|8gOeEz9SAhZ*62!KAIWn=PN4^L2uNJ*jpqP0~ca
zSc~V-W#JudA4cE|T1*s6C+C!JDp
zdO%-!)GXj?M=;6@neUd1HbQ|P+o_VniTy04X0(a(^Y+Fc*FTtZOD?}_vah)%P8)`S
zO#FMfC(h^Upzolm?XqeA^7?GLPa+azBn}6i+(m{v%3jv(`?sC?hrzwgKm9G+QzNm%
zT&{HS3&~OUQh8?+;6M5srbaXCL|yP4A3-{*Dprv;@k8BRW`^CU^wobuJh!(hI}_qe
zhHZ#?!cU&XXfKg2Q;rPGlS^VeFw$c5maz`AIa_3m3$?0b$Oq0*AQ7COBmTs=-yP
znC9@m(chMu)bhk<@2+(_RTWYdqw&SRu}LBU!wBp;gifOIiFhF1B-J(#QhS
zzM&+>=#lRI9HBf1;vLPq%Y=eXj%WH~BKDz_f%=~u-}5~hp~nMg=Yd(VOBa1|cRFM5
z1KomyZOt$=mH0pQdx=PyWPBv9K*tXBRSMFbvC;Xp+ZSN{Cc2s0F}9Ir0&
zftyYpfG(U~RZZ4O(pMzi&|F)GeAn%za5QA4*_B4Z0e2`PD^~flN{*lOf_HZ+aK?`=
zq4A|rhN^+@l@3+&$ki0dU}U9bX;N63?!v$ld}V%AQjE(Xt;z#S6u;{}YcIk3NfYqw
zuAp1!bXOESTIl@$XAJ>0v(64AuhPfu@9+QbF!c2FNMG54dZEx%%tYE(Q+7`S)&#OU
z1ct@Jin4>Kg`5|<<9J(be1{H)C6hpj3Mk-NIXeY~KPzL|H%FSc+;@o$s-%jEm8QO{
zWKd$PX$L__dw!Bc>|JfroxME&uSzNV>JQuhU8S^iS9|pTST3sXmmmTTVyHf1UWbRY
zLg=1ltoIuAqO9l3b&GXE3vD0Vw)O6SRD4D9ZB^C>z(Or}HKD#+@<;g%Nrz_k>n8d6
z-C00sPF?En{s+zInrJEoB09BvZ%%1hK}AY0-QB}zD?UK0(u%^a+IkN}=rArl?GiJH
z2_j@Z>u|2^fY7}LyT{@%&_zH2l{2uyNn-ynLBc5zf$l-3%8k27o{NP9NF;%bUWm_f
z4BK7d%>#(jLURnrAE%=wunSsZfQApR4b~D#@{W4%(bWxg-@HJG~XS29y
z|Npi{043io$Xg{eP+oEd677OM8!X>(xQ@a!*t7$+4w+3|pK}jG!5@4Ak8sH0QE?&(hR#DxEG3M@c=*2_Lj|3J*A9_0w$QM&c0`@+Qz1?
zA;b?m{=ZBTPh2D=5a>5C!FI9mc}3vNF4O?F5P3YOHK=jpOWO}9664@A`tOp+vH~xX
zF6EnK!WdaG@j`Q?$G%rN
z>*unj{gu{CwZRERkFCZs=l^}~C*T^VB?-)B
zM>4WvSa7h(Pjvy?QH4QjMbd34HFEq}N{z3I?e721x(*%k|N5PQlXsF!Vim_6BPbnD
zL#sG)%nr`Uc3j?*5kbEtP_Bpb1p)87w{vplixJNwKDl0o8j@@onzuz
zY65*&=A^l!X;blxWf!FxAgZJ?V(B
zB!XqzYzQBZOJfrS*H6Y+YEt>JQg+QSX~MpQxA%nt%4{=_&%j*alImJ*xN43wVnT-P
zZR(2k_@rqWz9c)O9#YEcfDU2+7h#c;%q*mmR!u5g@B_=d(gv|k`Io+ewTfO5puj{d
z_3X4+Ds2l}krPU3LPcqTXJ|BuFJV`(8;x)^TCVg-j%xGFHwq(y(qfps`ktrV{KQ*7
zUNbu^qy7Hn!`a<_
z)$n&_!3S}sAd6oz(Yx&ub%ty^rxCl{t~qO2yllx@Z`ljny90KC?RKJr4-}Y9vown>P4LS!2Kz@%XGW49bp0>0OUxEs*$lC+S8;mUf
zU16gpuO0T_ca${nuXD{ka&qqgzrm(nWM!o(v(7No%(>P3tS$P3mhx5b9~K5_k<8tO
zTaxTEH2ySPBOVG!`;J$>273<5NtU$C`I6ywCAeFbKs+4erGsw{JnNFYk_9Dl3_fll
zuSxRQX-|UF^qGlSy{(|8FoOJ9*M_VLA2>V`*SIr%^f<6aNxTqCTzKf4Z8*{CA
z=dfc#q}~A?#%uq|XMmt^Z+rxuU3bJRu-V~C2tnZ&9L15^o)X?GATp%$`S7^I`a+(ueIKXed`qvM6Z9i4MM7QUI
z0g=+x1yJxl#3bNAKH&CpQOLBWx;k`bW+wm2z}(>#N1I;Un01?%)@t78&=nhDNQnY1HgV5~u`{GzpJ>s(BvtFZt|F=`{!CxAyk-XJv~Qk0I@Ml3VffKx|M1
z8yo_%couK37yCdhPa-JX48BUb<)ioCt_sNz<16<-yr)SXjc;+4RBL8ZAWIw@*!#+t
zrmCrLU~2eSdK;t2cJ@~!_nmqEp(IfCerZ=B5$kk`rr}4uUf6$VaK3hwJM(!v;pUID
z?)EV@@k3VQq=y62lR
zAXV7aM(dU0LlBHh2R<4VW7XyPQlaALO7Czhgv<-Ga1XphS_tLmY_k^#JJUu+cAqXp
zzIpJZuk~pHT0lS*_SdR=O;Z=i-j~uIMQv%`b#jisUW2*=5}@&;b|VMZp7_@lXrTkT
z?bSpRD#hR#+^D_YzDxRsZ>?{J2M%qZcp$c(s!^1c<%0hD3u1JkSJ@B=}J@V$XN2FUQj0pV%kSgS^r3j}k+?
zH_^3ZJsUl6+$!J?puZFgSjMeB+tEN6`+0!AO3eyqRjTYGRer}5`>JMr&OvX)LGSng
zO1#_ELA{5%-zpk0%e73$GzBC6w<2GIS#{6=@sF!H7RVi533?@i0_g{#c~5c+RTq9p
zz3`4-g5!hHy|YxTF#!4f3i+BqSBw~33rFqdF-!+`PEh+qFr2Nac^?yDqVH
zNzycPNlkP5GD~eafSZyCgr0Ya?L@b|2x?i_(2Q5HK*k5yAcDv)QYV(>!mI<$na9m^
zOnc;%5Pxjn^<$L9YBKZROnZxMZ_|q1zQ+~w+NmK2&3Z+$ZR2&&nI)HcCd{wjE}*%3
zexZMP=X)vZWRvPQFyNFYvD+7lai~*I$qJ;Cblt9S{|>5;jm2d0aEE+)=yr;0s43gt
zx3BzmjL>RDP=25ljkvXJ{UhZ_?k2`q!Sp3x?rk=a
zOGRYnrB~ei4SqutodL|UxDgWTa6)GvooB5JwQXw@q
z?fp9351vyh+2N7oTB8iL0&*Q_`^}gJUNRr)e#6rbeEIcRR$KCwN~66QXzsUNx^B7q
z?airyqTRNb*6Z$8p=F_y`K)0kCVGk49de-Er4ObD5hy0>MBx%U$WS}r|qYKbjp?jZ-2Ykx*XUTgh6ks6
zZkYAd62J>q+CFJgM<9&T;`d}|Iyr-c@(pQpkPSCs5Kg&kxlv}ncCDV+>q!zyz1?6=nO~te*(5_&5QBD
z-p05$&?bqP1T8x>*K~FjP!L6zpfF+duE$l(a^@j&P%~ghy8RalJ_+F?zHsgDB@<+SGdjKZA1*Z%gR3M7yeFcL$QphjlC0Jrq
zt`^40WCf}9zs=ejL=4N)4R}b+IGinhu;S?^5P;7~nF$O$iq1n*JCxEi^pn-vi)~Kh
zh#~SFn^h03myGALWqBt!ih>|l9WB2ped+@s&-}LsFM-7}N_m1R!T=&g_^BLSA;*JG
z`bhdN4+!t@)Ix{hL8E9pbTk?Ks}inXJQNr36&0IbmZ8t?u-esR145c+p
z4b*@02jfI?kwf;IZH68Yp~VE@kN{=D)ecm5PjPu~+O`$Wm{Gkf>&?T~{3`valM-^CuUeWgYK?cV{9=kN51MFTD{OR;*ts~Lj-I@7zr;u+W
zZi>qqw91vv^YtN73_+CmQ5M7NfF%CMQcZ7GO*;`aOB+Smy?~`g_b~)tu?MH8GQtz*
zWZOmY@~Wuzo6TZ{=ClXhmwiCpuwT`?!kbKoj?K_sY0}kl^+#k+nCRnKuueX3y}%C`
z;;Lv4m172H3LWBeG@=NKetF-R5kwy!dy>ZekG9S`9Lx9r|F;rKM)oEoBs*kg&#Y21
zvuCo`osr6_n@iE}!4$kMHmJ9>;lfboBP-zAoo=ozL_2
zdOjcL4cs&7VFI%y7sOV0^78EM_$jk;M>LC)it~qD`-G<`oYk_(cVFglYg5cgwC~g;
z+xrmrzFhc5AX@@Bk+7pw;1Ph?$5G^5+QwhIw4N_+?;iedH3(k#uV+tgaYREKS@1C=
z{ab-^@%*5hm3BY}+f=-2~gb~0Jks9P+g@uPU3Z^863xDNN^`^srV
zo{ng+^b2?vWn-;6IS7PviBSKMi@D8>S8rp{7LMU%5*61^J6lyrNi5n)L-rLV4GkE8
zSls2NMBNfhXayrNSK62@@)_r^ncf^=hLY+6J{}6nbLrYk+Qr@TdCCXl(Vq;b{!&fU
zl*47c05N4LcpjWp(C~-hG>xu?ODiIwi|>eLUC4x%MxU_W$*TlXA)R~g^1`$1<8kO+
zqqc|*ZM4W~(O|odQ3KmG-V3>&)uquj;=LoYPHn~`>4=^v+&XQ}pZTUo+-g~5jXM>^
zGg2EqBur2ig?#dppJ$mrnVxK$Li_yxuV)JT`(1=9j_ZUH2!sZS+sv8Z2W0798XtEU
zoiagu2w>ysrG8hDB0f1cGwJ@kzTw)LJJ7D{bJ|7onNV-})00HZ<@VF@w=cd&{6YOP
zLTwZa{!wK(N%5S9f}D=^3DP;x1_1*hC4B4sC->4U@k2A7Kjb&J-dUm}X3`uKUi+Do
zLX0TkXS#7${4&O7z`LiCDX
z4j)ZU0-23Y=_Oivxq`%~vQp9YaYB)gm$$mZFS>eLkOh;(tBh!KTIz6G_B)oj=0YQO
zk?<2ttNFjH=-XFk#EF9LsD_cZJmpj)c##3aR)qe`;uuNMUG8^xxtC_wo=r19@0iOu
zFBB3;=`@40)$?sbvw6oUuZDqW>2}q|!L?d%x7t8#os;1%Du>z4H}h*s;u3ujVTnWk
zrJw8LlW4ujw|dx}=8~0`;R!Es8I89@ve^_nI&f1kN(ryM*K#W~M8B{{3_7Dg)j|
zdU`0ozZ<(Q(qOQisFZe^5{#t7oJqXRX4t3t*@MiLZq%$w%rNbcW$~Lh$D0JZ(G(y4
z)_q|K_PJM4>(;C(gK7np4cw3D=_Bb+vY&rEe&s(BfV1Pl|2!I>
z@MA68(bShFQ_pv98q>vtxAZ9Ac3yd_1_TEmH;o^pcHpcZtGWz@+Q7;6z)5M0z9Svj
zPr;l7Qxc{EjPdc6*X;CgZzh~z@Ev(;k;?yk+H8t#$S!%_;`DxSJ{8NN#2ov(DAUL~
z1q)6_e-d0-#D+FP50?Kd72hw2e;qukQ7Q~D);v33K_>W2*%$>35aff)BIlB=pGWYb
z{hZppZwfGhXn>#tWC5~tn`cWORZ0BE-v>s&WjAJX*k(yxQau0k7R}RJ`g{WVvbk@S
zYr1^r;UWDb4nKH&ZRp&D#O6xF`Pq*GYC*A`iV)FodqMEY+OwP>F--W#L>cieXqR|Q
zIRQ}vC1hJ=X#Kdsh&PdZ6h)O)XJk@6$y9A*l31@q+VQG%HBXl-11@ZN{YtQAPA5Z8
z2B)ukdhANtIt&LOWAcW|Zmt{W>%HzAdfl0Y;_qM!X(7rPJO?AB<%Yn$blg(<%!jHK1`y|k
zu#aqm(+$efF-ve}`o(`!PnkN8RPQ~(r9Hm(VtnmfA6L&Oh5A=uc&5PQR|8`RYTFRHE**BVhrj#&en0N$DaO!P
zr12&iq%B%*`l2}}Y>I#;!!tg!8-j9wS>2B+Q^~4JwqL9uc4JL;DX?1~Jn#w?0DCaS
zOT}NVw0GXoW4H~j=hwX4Oc558SHh2OO{ivG8W)`X2VNir?ij^W=QbT=VLxguk1kQU
zu-HybJ5{-F7J(6E+1z9!;e*?u2uady2=a?gcT`^wXH}e6Foo70Xz$E^kl)IgIwO$|
zQlBrD6h!La^pV;;sMJc%j{#@9RLI#!9o%Z74|H;8Y_Y;#R-C9=s@F-m@dL3ww9ay
zcQ3<^?#G&lK_OmZH;nAYVz5eq5oyOSg&n~+U3>-uFK!x3!6w`hw5}#v7
ze3AsdO>YNB)ip6Cc!Z6roe_S`FuFPk$q4N1;KjzpK|2yRVvh8Jq6W_H;aQ6-{2_=X
z%MxQ{ji>a!4Y%G-$w(=)XJ9=>YTA$j**d7c(YUnXnBk+*PN&?lSMXykw$UTPurPbx
zyI7Ov(?ss7XGnso*+}A@;AJW6pK;
z@PgKXsf7)R(YzX^d3E!5e(^1EZnn+Bh)+}lMVGvjn`;t6Y?vv^p!3y)
zS7gfWp#&nZ6+$xzWJ&8H&K-$hJ4GTe<8$9pbW@Yz8MpMtwJF*rULh2U@BxwC$qxL>
z1h8#;u%R$^;PkQblFeVGJ2zh=QC4QRr
zm}?|3gDuc#atR5;(Xp}5zkf>=6iuB)?LEPP07|IR9OQ{UHr9|#_V1h6CgSPoY2Z{j
z4TqM1A;@G};$VD*A^tVtPah`9$E!Zh6YEm2J}(H2yDgw`7)F6Op{-o1KX`H#)#v_n
zjo~ca-JxpJ8|U=(^lTHsUxkj)(|d}D@CuuwO*4LR5l5m%kl@z*?5bs!f}w}X5WwoZ
z!X7=Y!ched%gGxNy!E34e(*8@CwIt~f#tVW
z^PpQSe0_;a^%?q?C?F)@=P3_e3Y@?!I*h=YXhYa-9@U&XfS`K~lMg~?omMV~J47hQ
z3`0%|0nxNx<=oOc@1kcH1is9Ss;2HsI)NJyf^OyHsYt5dYXJr7x2vazjUfW?fuC2l
zqi>O%mK0!edq1Ju+!IDs3UkDDLS700KdjNJ=2r8(5E4D}Q0V=$=GVMG9Ve10#V!-f
zzK7gzn|Zo=@?b1?APTJeq+v8;Wdum1VJj58d<++4NxR(lveerTI)lb6dMyVI*kc}H
zZ}7&^H$VK6ytlB6e>r?BhAmX~ogMgDxy!~rHd?*zw&oF9Qq)lp79WZsA{?aIB9Je<
zx_PTaz*z<;Og+grNdmK`0OFH{^8gTdKYoO7a(g#`S5_AH*=5`_mtks?xde%!B>m~W%Zo&Jb@qsHS>p~j%8OcJ=STlw{dKUZd0|HPibTcUE2M8
zDsGoj0#Ojr$XX5!Q_!J+_thyc3yl3fzp!IWE}r7X@u?BZv^>K&9BHeY(0&pV0R=R^
z_H=x$d18VUCDMM6JEn`^a$G0aSTpo_9zJ?B_M{MUQl|(fWEI4gNbISzX@o2=-nc>V({IY}ui?u`323?m>1wfR6
z$>Z`da?EzCj@z6KO0
zs4z80Q{FfLK^@k%l7KELw_MH8aVeY-vz
z0VjvuTSUEB#3d8?*Cvd&$fP0MT|Dx|3n7RET5MZTVUoW6rqx{B*6EajoI_K^!H<0HF_vA|NC;EMtwPWHua>Z*IpwG&DH`_tkqwo~l84N+@G
zEt#V&onwkpe3^eUB&9Vq=L2sL6B62Vb+2qQ{`uVxaTg)IQB41bGL;ZTQKj#9c;j?<
z77iQJK!v#1!skNbEAF*h;9F}^om(Ie%;5Ym1Rk6`5PQhD`)$Fc!r*DfRfM)IAUVk7v2kW^9E>WmZiOvP?h6HT|2H
zVp~Ow0FQkKllp*fnl0LJUJDSH=~Gpc_gKQCw_h}wUU+fFX}S>ufWMiYwQ)|WmkS8l
zbFn7{kcKwS3thCh&zk+F?CQnin>r)Lc
zv$C#C)OnF~a|m4kspjV9hQ#ZU&^y2y4L{AB*i)jxgm
zBsdRcB08AR7XdFJ_4Xn8E}G3gkJ7vtF)_aS(-66+#$Fx0bI_vP_Psq_&=IU6Ot+ot`uY$u$LwFT
z39D~cg?;>?sP<+^CTUF7Q^M$c_L_B-*N$q=U!w9$`E&Y<@ef`ZOn!yJj`Nw}2*bjC%}uvUIi
z(Z_gaoOY+o=um3T2ZE>n2M`rpsi5hisF
z#aB|(<~}d>TXXR4MNd6bX7a%ubEIP4`-J~zB;aVo
zxo&a~pb8;SvjfXZZZ6lq@_?OQJdfj6zfUfbtWlgWpb{m3x&^4Hg9ADd5q)0=G#WmF
zA>6SX6e?#h>~#O|(WH$>R%SpRKuEN0lO!6rF~(6GFK=ON-MyaYapqk~T=TZsHE^QK
z`0=P|>`~3w%>d%OZ|mg-cCHJ4Ja0DENdHJqi_TI*Fe^B4txo8C2IuneNVJfV+W_2j
z&^bV0HGVG&_#DFp7eMQREPv4ZsoUJ1Tk{&XX2MeqaS2|atvtk&@*j4pg7A0NZrKRD
zN39dS;Gz5jP{5_7ZRjxHjQUV3tZIQ#et|e*b{w)0IuHYbr>&?6E4~#(;TxI6^qeU%gf=c**Rm?l>*)G;KeVq1sfaaSt9|CHfSZ=
zH}cY4OqgYSdHXUQhLPJtgS7Q0bo<5}@xJ~dM$p0lElx?f4v%jqRxX_%!r($~RAH^Q
zc)J9*(nwuc0M!N9J%GeH5@luEA<5*asqc4|3Us`WwgZ4WdGc&uJgyUCD;%uCp1lFt
z{?F*1+A9RJ&NTB+2A2_-F@RT2wONj@bvynUS%^4SShlWu{>@`ngrdhGAOj{p^F6JX@SF{v$KDtLhIys|%)UP*_C79a&a@f;PF7;fqaePhv^-U(lmqEcLOel$`
z$Z@yPB|C~dgQdZ}Ar*>pw*ag3AKoHotK-?+=F9DjdU`rMRPbU2{!*78cq0s7N|F*F
zh@dEyR*L0YEDGe??F>+K(Jvimj8Cyz5hjAyX^oNKx#Sz#1unI2iyZ3{HFiEEIfGW7
zc(1OL5lwQc09ZYF)Lfs<_
z!3zZeHgLGUN#e*xfmZ36HO1pm%=j93wji}T3IY(;_ri`VniRlW1*eAg7obl9+%w@!
zO0Nwd!+cHHIt5vyXynpK&pdt2^-WkqVKq3`10$CEo$l{G_rfBzSvlr$v7ye7FN};>jsvE!Y;SafKRcNYK@g)!aj&r5I|}mv+w+8yH}OW%(pT&FVzA6!-cHox|~#
z)(J&%;{i9W>?c6)fQSt#x9OdmsKH7(-OGH%5_(YW`FN}k5?gYYBaL|Tl|@pPvf6#_6g}-
ztg`eCYVI3-S@tx8#+n5|72~U`SHVsJ2i5-}ZK5`UKyj3$!p&QXne=@rFmdK=yTUezi^B6$*o0AuqMtimP}X
zW1Tq1B_&4JoWYw^b1yTQ%<}WXAv7uAY}U?8IhmVpk^X=;&Uct=qn$JIwQye(MS*>jrJ(^aLWmQ&~E-C)w9~0L1cHE#{0H4C7
zN9i%%<~q^{LFXs$4#__i36VR8JV^cSRAtz8HJh5@r8${RDnN+dRR5ZLOI+zOH#~;T^j(F6$5jVf1A9DmKqPAY!MtYBauR7NXgX@DLgC$=bMUv15(Hn{#F5PBdz4@nI^Q}0>lh5X2q$1>?
ztpySW2%T0ueA6v|{~ALP$H2Y^Ey)Df?4GXK?q0stI$Sq~;z1Yic(hM!L_!@1;8A+J
zUDh(K09o&JgUJhzAiqpgXid$zuI_F&iVKnTp=nCb(mBJ|`zPoxc`9SNt_ekm#-yukz+L^PsfN)cxGMC1tL?|C)=6cz3^F0Gbm*
zCR<&VTv%A}*iCI!h!~$F33B#U?}1L=quwtc0dGBduXdtNyyJ-Vp>@KTjKO-lQyc`v
zuc!p8Q`3ttE6h*!qrGr@v2t1;`NGfO%C3H;ph3%$Mz|(~$?I}c-gcNA0_b>Q;X^dK
zbKvq@cGZm%z9rTMsQmew2loWi+CMrEnj2WbN=7vUx!U)5;{fm7ihlHn(!t?cwclZO
zmN;!MVQAaMlU3Tm`Kt^OeUa!us7R=q&n}3iXFR4L!ia495+xTPn1oviIUMfp!*1wVWH
z3%qfM?@t(mXwCzVk|VmkrzdtjqN5TP=(lPO4@jG20z2cT0~O5a0D6`Eg5#z;Mj9~^
zEP1ZOCWuL70~i@&j!L_|)(rufQD9U1?ZX_``NHTVa}B6>-O=`sCW1B0foGLFM)Bf_LLJe$@cA*O`8%A
zl{^pCNckNXh0Oj^&OZ4GW@Lr??7{#9Fc>{)Rk2)mDL|OeNk~W-_n&j<6A>APiCMgO
z1k6JgQUy*5B{7C&m#qtU&tSSoAMA#nSJcykT{uEbcBO?s0bOJ%Xm{s*{*c=ANbMvO
zyfordbl6%bFzvP%kQ#)>_QsVvOY=YiFB#p8FK!_cA*H|6WE
zmOzL+Yy|p?MxwQxNFq7n@|9!)?T9fOF5b_>!7$S8~d}LgQMGQ{;@J
zh(+y_M~(#>`^UkT2;6qe&p7D^meuw%VJpYivOoAehgS8F{gCh$f2Db;TNjTY7D_6I
z_wwh&h@Yac1;n^=1bEL5q@Qbtu1PU%e4%C(Cs5RZY}l(L(c+aeN8GSywyN>elHSLi
zIvDm!o_!D@jUYoslERsf|6B#-cfXriO+2Mi(#D=U0^yU-Dpnd-JD-QN^xK}jE$2E#
zp6i84Ol=UGX)`y8Ow~VY%t@hL%xn~5@W!!rFiv^ei~d;&S$v9Myk_~`Y@SZG(pT0L
ztgD`Nh4g6mZ_R=o*BzEe*M0xAZ_0Kv(bsM+k5RJ`&OPg0RA%YhG~R`kJ@GHeQ3V}4
z6oui`f@+TyJcNG_AAfGRke=!+r7Wq-`ovA-mn;Ubx|6Y}{g7QgOSL23pU8nb
zrWHY^!`CNCO+Ase@c}PDGk8J(QGUgZIr3_|IGI=h0c~nWU1pJasjbLCFs(|B(T@>V
z7tw;L3VG_sbtVSobzRxUd~IAw=($nrXckqaG0nNr57&9+-YER3=YQSnOd8akS5TVg
zP3f
z|M~&tbT{lfz>UG44$?pcwgZKRqKJ0swqBIz#Y$39_r}voQBfE@=P`_n!eBk<*pY?Q
z%pdq4aB@iLR`BD=T2RJTKB=Aa?KN(TT@}Mn_yDBqV!QRLwQ9Oo*5!7Ifp@v*jXM~3
zKlBLHJ^D`3WE%bQ=%-kZIj;Fz!W$9iWcfS8J>wWA*GM^3&lp@6aTwz?{$V`uHb#8t
z9p_rn-<*Q2+zkGw>!K_BoTHAFyo==#r6Ey(U}2~ozvmU(^2qvT#IiT<#lgYSA-GcrkNu@NbvljGMZ=ZtX-NP
z#{|v^{0E
zVC?f}^ImTufn-=M@!_q_8btsDUJo7JhZvK8-!7~)2nDR6kuy=W0(>cDj+XU04?{Dc
zoQYRn-t`2Ie;F2L1-M%fr?h?^hNV3LssKPSo}?PAwF2HfI70;?(FbRr*dF{SKl+#6
zo)|U*DNOH&@vqB&YeY)z-0QS}yq<$YipdFxVq|Rq6V1@jeCqD34^!#rTPOWbj^gBD
zxPak=b)A#|x7`;{LZAkMJFuaI_Tn9{Pva%}7VAugc=5dlu$%+*;IUA-K5|Y
z$PF@e{IJ7Cm7t^j-T)?WnM2}K)M?@q1dax)28OE|x32>blu@$R!pVvFWMTSW5@G_!
zg}HvSvzmG(8qP#s(bs#x4TT4Gsm^i`;}f069a)%zgj)`+q8{8i)u9
zA&@ZGN-(#5;5;8r0_Q8FFP180LjUtNoIkS^MEkrX7{JQ(ot+KatvFQaA#!F4nZVW)
z@M9igY|nmJ~23Gmt_M
zN~we+f~$)-k5`N;4ZzRVuZ)SqC{_~kCqrv2`{v^hh6BLx%0nG=9A0EtG&nX
zH!PSU@9+Ah$?;Hah+DVAro#PKtP)vH=hNxf$S1$pq&iWkpGcqq!MUOwji*Th_T>K|
zAu)7nT(|Xg|8@YyBxr0Xg!!+EHl!@^tB!DyZ9X(J`wdwF*7)}uL(l%!^d5;oXmc83
z)HJk95+1_D6su5?ymlyU!X5JmQdHx6{}yFPnx?0*-uyJJ-@b+W)amBJAu{X7y>B7C
z#=;TLM)Vrw$-kY)CnD0@xmrkxZyFBTLY2^c1|%@*lzs~J%8@SfL%ybeg2(+b2g=Hl
zk5031JP~)7SN%57%ur~geE_GXJlXFwZQq6yGya<#zzOUo3)mjI4~0uADk^yXk>K0J
zRe9sY(;rWOS4fL|jL=JgxJk%glu%?~ttAMe!Lc1Wrg0u6E5$Wz{!>xxM`U_15clUF
zws-FH3&I>fHTkwk|1GvByLHowoD^p|T#$^I=Hc7m*t^^>7E{s0$kQ>>bi6$%osw?l
z;X&CewaZXsR1>rYq1+??0~0A0Pz^XrVWUViulmWpVB^1ezO)@KqU*^2*+L>C`7w1e
zN44-?i;Xmv_LQ@Gu$%*@*W9k(n|+z%XCU9H&e!l9%Hd?fkG%o_(Q&;h79;@e*PP^%y5J@xJ%Wpck_HUN&m=)N#K+O?+
z1E~N4~#Rfo;%fNAz@HDChA3dNk?FV
zS}R%W(Vs7Dkq-f`(rrzSa$fs!138>42d&$x4~cBPpCK~;E=wORYq0^~XgW6_bnv(lr`WbOhbt(4W&fiNr!3NACe?;uoFG_j;ic;1*{1_YzQ?jYe3$pCEXK+;B
zD3tr`&Rmt*Ki|0D{dg961drF6j;avw%SA|F|H!%FUVrdAf3Gh@`Z-8Z3@Y&dox!z7
zeKKpB{5Ju^^xqTawAGUT_XL7aD3V9Sh4$J1J;VO{1$!qrvhOV9;#2+S;rKn2s-mVs
Jk=(n7KENm>Sx6-dQy}xZ;`BZrL=}{3%*v;FG
zET^~b&u89_Dom8up!%H`qC>4_l1e2|ME3gi*n1g;m~ec(T&nvHQSX0Cztp00|N2xb
znDUNHjfsLTElVL$)RQOaiRP=#radxHR-<KY15bMeYB-;MpTWv`o&zRF&0PB_o=*X*XcEv%C9NboZl{
zU37~|*J9p#3o;*k|L@0oDK+sJ*UmWjAHEFYZ&{>k
z!N%GfLR}7>U6kMWUbdi4(Om(72hW1;q0#8G?A>z)|C6em?hiMcK@TOb5RzAU+|B><
z<|ca@S&Y)|1>K-g+-iH%n{x`c(q&@vZRQU#Xa_YnAC5+?X`oJxD$p09(?IU*8?BUpHmaz|MP78ckKJmT?F$He13|K
zN*%SXKJgGFhC>&GNDPo+%R#2ck*YhZ}-&q{$EN=4LfIWH7;=_rBH#oHk
zr>!?v-j`>**X@#L1VZBC6Fb_)ntS&@?B72dD8E$jU%+K-Lk4am^Ylu~u7+zX#bQ$}
zlrHRAPwdXLZ#F(GmS5Gh;v>c>PMvN}oO($w+ejBk!D|s>owX521P7s=ZnpK>T3U9R
zV{1K2mi1d!Bzwv)2g4RjX29JPk_e2BiMhMY3$8dRV@tR~!JS?g
zURq)o7ngXek?co=INL3Lettq?O`%e`gflQ++Bntn56fp~Huf%+4s>2SvofWyPqa)3CU!D_qTz7v@#Mb!DMUggmryn7xC4k`>xL9;OxEV(Ky_
zHx|q^;iEATqR#_QpV@xGRR-@6P%`Jk<7wILJXFUq7OtQTzl{*Ua$fj>Bdg@JABq$H$zqB}3AfOu9jXYk;W4IuMfPc_
zikr%9&=Nc}6d@Y1y+`^)5utQWfrPVeTeb3?hhmd3;!Ym&aAq_YgqkGc-*JO=tv;xc
z=Q)T|M3GiW-`d^1D*^h5Y$x#h3ee`wY%g3~(3WZ)^o&C}z=hFdfB)B2SSUV*hlrOU
z;rDtqDBJ8a9fUw-ZEZqUmZ9gcE+`G_nhbD0URO;;iWh3KFaP_;btg=d%hqJ9W
zWMNX+Ce(Qowu|eC-bzC^uGG#eG%2Xsg$4$eB;JiQ%nlhdYYk1nHIL>TocF6O=Q;L`X=)wwY(fR5tmOp%*Ct`F-{RTeWWJeTXIP&W#oXTRtkXQaEOgfQqxF|)k$+2U2
zABEV+80XxF9YSZqW469!HK-LnQXPwf#7>Z|%^Oe+?}$L!Dv&_+_`h
zgiO^GXwq`EbXKX}KM*$H!~k$`)|n+;{sRE;KPTffiGKj&4O;Qx=nAl
z#^ys8Q@c?Dm&vol*C>feIZ$-_fU
z|oo`s4|2t3Ni3I^?Bw$Y2R-hkoooIXzWVFE*Q!Dtg`y~A0J
zLPA2)T&Z$oc%a0*FOF{eii^kWiIJY}O1>&k58Z`S=-~bH6a|GMTQU*doa$O1(%Jc#
z^3JWgvWRX|u2<&=2ajBse`?u*z9*Bb8oX9uOpRZNh|+0Gv3OEwldD&nXHdqJ3TFmw
zMt~Eq9S0nO&~~hC&%32sZYz2AGlJ3G_YTf+pt=KP=OHRg)y6b)X%H
z+*lIeI}gQ#;!9Nju$UDPVSmDl{`f_0B7t*=8;wai^L14t62hM&lVO
zI7}(b=6m9}UmM>x4K9ioZHP}z6rMKn(n?Idm=Zcg51smEEOsUv0lg4rp_sNlMvGVL
zaUP*dkooq4CG-h4Lh+$IL^3IJpm78QG#~jGATh02y*R{Ogm?zVlf%tJPx9MZ0NmT?
zeQ$cdZ@Xx46{RZW={|ynLrwJCdK@$Z4j6e22OXGwd|0?Z6e}
zuPcZ7?fn7%E$#c5TP)Vu($LTWIs;E;e-grl9@CQi{Lp(%l%NN2aSdZylK$-Z+xpg^
zV0*zB9z5JtbA|1N1f%`X2Dky;VYk&tr8W^2r=fF0Cl|>gDJ$bjwOlVyPgg@}Uf&K0=2i0emYr?N
z7nS2TVH=lKBCJC-iYnEj7|9mzH2!`!$YG{ubkuO%_Qel%Hh^2J^_0W{FNI*~Ynncl
zIw2-lS;_<|ITGy?69t~F;0mKp*`m$>h$3(~-SPVqh}=B^bb&J-6zy)8oWub+7+Ejp
zy>Y?q`4v9iyT0yX-}|lKd~a`Y^H<<@<_BBIVVZ!xK=DT53e(BA0j>2o#NnX{BT1(XKYz9zrSBnGXGvmoJJ
z!M&dow4nIZ+43HA^0c`t)buJt;w-BwU3p;+bYW8D-Fd$ZMdR%!wf+jnF!PuYDl#B7k9c!Hf0%|m8!I`|@(;s2-o23T1d8d#)
z>MdL9^GOs+PB1fKS+&0yvA%S9iIg=OKXn(G!*Y=^Q(#
zAn+4@OjPEG`~7?v)YT_0U8)u8APSI&_t8HG1A|2C#-zt+m$!R+t%QZ
z9P0EK^d>z^FtZ(V7YPOUqQbB)frse#?x~B9Pt@TdV$yWUCwsC6f}oj%g-zLqJJv?P
z;cnxb$f)lc94VN&Y+A|6_nAq@Sb+9+LleS=`^xA}*dz92cJOh`Chu7Qs@qy%0#7DO
z*)N}}(>{fdLpk@Ldkz{opjEZmZyvq=)#}fa*lFp277qKF@Xdk2&sJ0H1@z^*?{3xo
z#L=$$XYtb+OTK_Uwzq%l7G(SVogAw<_16r9w3!-cTXB8MDGjx?p!tG60j0A*+I4r@
zbj5Io*t1=VA?O2uBhnOq1m+@a%7>twaM?;7>0DK5!OV}F_rgf9U=H7B9`bB7MfqO&
zt{)1=hbn|A2&XHi@XC={eB0c#Gcb5k(ydWU3C5%OA;GJL00#$UGT8O
zPc!WWA1>0oaTxYp(cY|`Jkt$-NY
z0Z)@TGs6V+j6A#fozvLmmHwrCLG3s>a82?}O$=&>WwwKv@ejCtt
zX}eskqQIkRJG|jS8w?@nPwK5}`SV&V6~0)5oCjDYY^^>`vCvweZ2xaRMQpE&@78=X
z8juP30{MzDPhQh_w|aVty*(nokabPiboI=34U;`JK2&#O{XM@u2^U`{OX7PQ
zy@84dZq1VSRkpf~@s^V*)(Lz{kN-Sk(dVpZpA9zoq?)`4x1GFvSrB)t8kL4DtB?MM
ztd43}J+?oe+ar#gh79pct&U4);a}%*e!#oUkP(We_u`f%Pl*CFx`mj8vSSk|y@6wq6?l?ma&j
z26)EwnQyLY*XZa&fI7Uh37EQ!d87eVJ3Kt>vRA#IU!eXszIW3D0NglP@y{$C^aK;O
zO1byNWtb%_uaKP8Go{I#n(2yE-KL*!_mAOx@*z(8w|!gV>m3rx{(E#cLW*S@Ws!
z8P3-!ck?uP)MoNDgO{s$M_6zFwS4XeNAEJ1fOZ>X@P=MB>dS;*N
zZg3N&r)S^4-c!V&wsCy>_f~Z_BUZ|a)}TnB2aLG!0+>myTD$=&=_DKUgPr;281-VD
zkwczLC0+PhQrd5U$}49g9yfF>YeMKWbgxl0Eyp=`7T>zpnu5g0X1_pRO$@sdto0CTK}W6CO2BLcpLehP|%lxO=)I-Znqa0F*&%
z0cAXwk$0f~81;g3kT}Y{0RrqyuF58mf@_7YH$Gei`XA*p%5bIj`k~NILHY40QEYqC
z0VEN>o#N%`>2h<83#e%tOO!%@F;7@V%$3-}UfzX1Nc?3?E>=d&S}_QvPV8WOTl`g&kW&hBXGCR@IV
z!?|O5(p^NhW}+)N<3ITbzS_KQeral>3&S%AjzNdK!jU{SOOQTaWZ@vJ&t6GqQ&3<%
z6K*?Pd-e_?D3UciZ~!>_y)vIBN;7JFs+h186EBhNF4TyJ;i9+f(VO!lTbwcml
zJ>iyaQlv`_*+L(YwG=*oC@i-y0dn0g>R+wT%U-GqCzng0y-4-|
z*>86NBdQ3-fHPd)dL3S5a_ajKe4BK&?jDd5yw4Ifm&!IxM6xPqJq?~_!&h{P8(9<|
z$UntM!$R+~l{LZ*OLqt@Hthx90BLAtrCZ$!0!jjqJIa`+h4)^j6WR|#(?q7XJ(s*p
zGgts-qqw0=iwcW#pYAl~`Lp$?UcrCD#yBIz`X-mwQ_p4mi(u23oZkR+^Ru(iyFA{P
zx4Fim@6@n%dM0z|Z#Sms3I>tX3oiuOmE*jy!)KyAzPw?>g|v=y=%1Lw^#q+%mP^$P
zILly#-BnxtZdQ{i7FD#c5xhW~(tvo|3w8OT5a(7Ex=N32GE>qj&W
z^MFG{X1zS{F#v<>tHOQrDqSIx;;!Ow@mKjSad3D{0X|z}BRaM;&&@_83qU3w;{K-{
zs**SE$#p;-_z1Kn1R_bAcl!Do8js-)eeA`rwyVju0DSX)nx#yv=2dUS_{eDd(mgptxz!Vi#9-3DfHt$
zRw-5eH<=;kac4u)Zd3Ae|BsvMeaHmrsr#$^dHY;%dH!VaYt}tu9R9#5sLE
z+`e;Z>hWkFbBA-g=8`LJTa_J&^W}-?)FKCM6M?8+X9^80-~5kQW4uUqShTl8kMG|I
zQ#S#662S`z`Jxo6OrGI!XB8JF$_wm`cF0HOIY+DY)nOf2RNV>D&yj8GJr(WJwK1_TK}!$dfR;(XgclP-7-Xd
zpotw)a1*{sE-85jhKA>_p-N5mm@hO|E8eTW0s$ZmFXe2k;JDjtMBvnXTu9b=ebst%
zT1#_+5Bx2C`kR9f*Ef5`Dpw3#+#~y|nK=`ltJ$d?bWTkKXf%O*fqI%E3+87bAPFyi
zGhd)vtQT>lj(iR286{(cALyopg@!&haZpgU`oS(KE*^#b!-N|1n}Ra~sHs6JUo?;l
zWb@Zg5>OL+7unO{}a&F69pJRE|ad
zc_OWJQY>P+#WRPX*rc;to%c3u-WI6bO55{2@fP!Cep*f9X&dffXUZv@95esmloS#W
zIS5Ol9|p1@S9m9z-*yonePT0m|E`zN<4_8=?03TzAAX4sxfqNDpf?8GY7q5~>N
zR94$;U?S0bbsdgX>)IJ`ALA!_&X-=jG78=l(cz#9xd0MWba6)g6>AjyP6)#j?)~Ah
z?z`W&>eP$DZDLAd-c&~jq(i-;0f|+H)OZ#}>_VmpW#`K}w`G;ECB8{95=BZqB)Xup
zntI~-?(VD4A*?k|-;w-X*j9U8K^FSrN9o!(iwS<-`smEh&Z_>9tYLrLT;-i5ZS;L&#g+P|
zSW{L)<&Mk2y$DnBX{tN>7M+eofrcMaDv(dfQJ0E@ba!E40jnja$6rFOYD%77pc>F~!08{Q
z!c8PTd^c+Cu#!a6j0T>YNuHSLC!hUK%>X}ud&w*4lppwiKw;o4jf3+L(KF1q1FP)w
z$Xa)G8So#vl`jDi0Ynq%?42(5%pAht>SG&K>fH0-y;n!ZlcX+I{6HWsG!$1*`U&xj
zpe|aA6rhxM&m30{Zvn0gP_jb<`L4J_O~!Nm+OhRmK0*Fae~HkkSlg0A;oIY9JPD4q
znUmOks?ut^!{VkMk$Xt(he!KiGg8xOyDptwZq!=P_wAcb5z|8NIhn_0D
zCONkd)iR7C1ismZhzhc4Z{mo~WTp|nJ)cyp`c%WAw+qNe>g?aqs;l0aFQ)
z%6&3>+QHLh&Ie4EH~DIE&4F30wvXQ=8-9hqKWiB&I-v1wCgI}Zat?T?9f=Oqi!EDW
zxHb=Tk4GTt$Vv^$cruz;e?{7C?IcCk4Vp6r_Th&NI(v9TVW>(z!eOQzFc@cVZ#?Aj
zo+0Y^;oj94i{y5r2q>1AWv9!y))1AzgCwk?2N!WK=VSCrdt%iaye|YCzp9u1Nqgx`
zlVV}QL%pyS>svXIl#M_|L`3vix&aK=xk#3px+Wwl8k4m*OZQC$`;p{+6xrC5iKMD2
z4%-?-X47SltwrZPMb;D~+O?j3V1=D7C*9-`(}qvM=dg!@@Vcrm(S@U#2LVXvIFFI7<~`uCbsbf^i{w|Z&%TkMyh4gK
zuF{;aYjq#0vr@h5NS*f(?>YNjiP-l1gf<#GOe>5st$ecOUhHnp1B;w2eBu;<5SZ!`
z%^Ltr&|g`qIws~&dUszcHxlCz#5@&>hf5a$`3O@SWX-d#lpWI%R(_H4w|!2kx-?gcCY0=em~
zGB87cT%mK~NR1Eh1!%HrTyAFfBiTNHnu(IVmtq0r;O}~+cgRVGn2mt?7=z8Y?fINM
zk+_p=M-ym<6SlWwZlxWYM^4#ujdwH~MHGi=*Wv*U%Uz_C>XO6l&(dSzw*^D(DY&dT
z)7=p1{aCb2oQlxD_3zd;HTH9622M#!58f*czuE3o{;A=HOVN@mpE2Cczg?7Y7UTHD
z0FdsQS*!xeQQDV3cq!=+#TDF7?;?(;YPMR}raZLv6H@6q_q~HRH%0VH|BS`GVNaH-
zH?Tk~yZ*v=jIQ4kn7I-Wgx>pZR;D%cx>Po)fxg%BY8=05UKqTcWh1yp&hvmH$#&-66Jf%iU#sq<*jf%h3wD_nxzqCY(X(qO
z_PWhN4i>v-Np#p{dns;ZE3#6hGf6zdn%POBEzCV!eQE{O`)V&GGcEsMee$qjNxIMF
zKHDm`!6-?cn5$4bqbq##&EENL-^!iX_5M3Ng
zK;-$sb+PNkMs+dL)3=`n0Mqq45
z_32HH;C1tS>3;0o^jJ
z&Ad{lLKxZ08>7+d;^OQa3CzP33w?WnAfU5Ub#XKrlo3K$M)FfIUXE4?GoIyW*yrbV
z{j3E8gRJ&i=w(eaTwv`L*A0R$LK$5&!7g6_&0}z+E_8mj+%k3vRo1lB6$G;57l)E`
z!QHEsjlaQch-V?gPZvHj9pxPxywFtrYfPK<_i3@!9C=*C=5(6!gwJGCc?NFtV8w*e
z$7rwdv3`pS`B9eP#8*2~~8O5vjORXLaw52EGsaezHdS2msC3O%PrcXRPq`
zZ5>l7YPVA2nrnI+h{Xb~RlkxQ>x7_)_3u+ZrajbmJwPo1hfkrP$5L;`o?1R3ZbmGj
zQVI|yL2l^PQxBo6%(qiDn+?AXm$BB}sQ`ZjI%#h&CJt0s`VQp3$n!LDL^X&W?6;F%Tk#MDt(>;R0o8Kj>@(we(1J$J#n-bL*h5e`WS~y0LK4}A
zhllt6_CMyNbD<~j+MVwNzS0i_4F?sr9_LRp85IB&Lh~Qz!TnJWbWhIDX@FB@YOKLZ
z88#WVLmJ!r7`OudT5Ui>7`2zbJ^N7Y!>VR2{SGggTiVO4d)=);x24$drL!xTpZ>083Nve{DNPIQNdQ##)_e(8Dx?0B=sgi^524tYkcQu%5LS!O8~AhbX(J_Fk}u1w?29J^~^
zl>N*csf+0@TOx#1d7Gdb3`Du&IxcraSGY1T}s?y@@EURj-kH?-AJgyfWA_WY)
zfjLC#z;T-%L_#?%KQDYoEXy;NnVqE2ip755jBXnWDJMse|7b$T6YgQJj9d*OzVvoy|zXK
zDSiM@?D2xmxo_IEp0>RHPm{bkXuW&_a#?{&q9$YO0zi9=%i?(=ASx`})8EhUU;=n#
zERLi!`-)km4oLglfk1kBD~2qqgoNrso14WwQCM@1bYT1dE)$EbFY%#mXNLhGtse*p
zEZk+@<0K=H<_M`@^C{A_9`?w66)^SX!ZCm9$Di~kua-T3__x&9qWd<#AlAu{Na7>k
ztmGv=0B%CE?%#b0X=Jne)Q4B>U{^%*WA+Iy9^uhb>iyS@WohJ`oPx_CmDWX5RYo`Q
zp*ufmVeBIBa89jb(ypCcC*|E22!j&XYs!@8mq;9EB@O2WXP?;5ST^sD(L
zH?3#0Yx7;55#?r}hCmS*wf2C>g5CK2-K-{n;#3`AdP>X!G1;GhH(k-H@=79SWYjt_
zQ@uH`bPnla0BDn$lhcRlC?yDgwH`63*xnt|MAg`TyARb;^*L5ee_1vM){I}p7VX{v
zX-fESPMCcB6wZ_^1fkc^CK~_vm%I<%gBoqMtE8gQDC@Z;OSa8bUAK+ol9c}VuLiZO
zNww13X)(RtaO%^sY(*AYonqwf0^+Cuug1TAz%ndh{;12COu0(KN9y4__A1+n_wgT`
z)PF8;8jYKAXKmGO1XZSi*jp|Nl6_@VUNs&%&qE}rQ~1fW64=fV@9>e)T`!p!?R)PX
zsaXbA_zszhn34}nsL6a2N=lvsXRPvANI&|H^tH(Db6U5#ERhw!88O1_K1R{Hk9YEQIAR#o9NWDE0%&lxxcDBk
zCnBpj!MMrnn9#9F>xdBeZK`YDTuh>f2;LPfLeAus!;qV%JUZfMou`OtsbO@rOrO5Kq|fR_x2BG$i7l$NIxat`BsMhfi~eHT7n
z-X}yNQ^DI#7)o3VnD+Dg!L}HH)YcYwCMx3pEcpO5a6=G}03kM0iDw#LAu5c$1NIXe
zB$zyap2LH1Y~a8r2!-)8SOV6%9oQ*>y!T|3<
z8u{TyM95V*5P*mX5b=Aa(v_7aYz2(mK+ynxf!bY~s)F#YJqF0i+xrucS^z@gAqu@c
zqzqD_RJhY1Zr9w53q}fWZu-Z-R(OwH_OmnkOpGswC_HZoH^I25bB?*A-+evR3%7jO
zzQQdEn>ag9i(4MxeY9T5driDuppleEqBUh7Xw&?Kz7l`QmqK5?*QJ_aQIoxUd`Q9o
z8CUJ}8jGIMkv!{x5(^uDQK`XSs*`Sk%Qqs>{UQAF<9AN(^}=fJu!1ih-pcEaul1Ix
z+Ehw;F~2HfTFs$x;m1SdTDt5mtaQ9_hj|*R#RrE{D;Y^kjJ%8y#LDRDo;%ly3tZU=
zd?E#J=ixB<)bX-@)Gcbd@u&S4ZKX+^T0PFn^mr&yfjSKs&Zg$8?jjcXLOI>i79ei|
zaF3M{2?-s?5fRXyMg{M$yW?a06o7^5%3_SlL*)|YdqCEbCud|oRProTNXWbH#LK7l
zKmDGbu{CNWVCQpMb^Zl<8a8xM%E*+%Z7h*`bg<*lM5r>Bw
zz(2A6-mj!umrt{MWiPM+jtbBi?(0X&Tf-uN{tuuXxafQ|RScWY1v2BL37(Z@bEZC*
zhg5;^R4YN~$1nX)tfpO!aX*N;v8EnU$63bW$djHFV1dDMtSt~k4Co=Z=g*Sq8c74-L
z_|L0+#`w(>`}f+pg)PhB6AO2-_zjHyq!X4lYchUThTKw84t2K17t8r2v&wcXT@F%h
zakd^26Q#xP?F2LP<{Wvkr)7w~BGv@}M+QA@1nL51DLA|k(9-8leJ&*SgTY}2Hq|#s
z0c6REu5W*Xy(2}vc5+o&O#u%9Mp*)*qAL;S~OV5@c`C
zWos{ptYMR79RmTf&hL~9B&QdhgO``BKxDd*Q(?+JC2A-h#KwSb4DLtw`g)Wgei94b
zgtNE&f};EM*HI@cklmg313+0+RP)XAnaf5gJ0@?sJH0I*Q12gQ1H#TOYuZLp=bY`=JOuU1Qvx0(-KeSNt%`#
zWJcu{#l=`nu`zQYHmk%o@y+A3U?wsF_3!o7cUTK?3Q2x6*bKl4CHzF!^Bft*hdv3Y
zFV-GkJ2TkM(4erCAQZeuZ=8;4orKg8lk1J5LD}(prcxVlMpK@DQA%?^$D-DJ
z`YfZ^N7n8Fgj&Hthu6b@Zn(<`WCMAjtnt7NAf9Ww)79;0&aaoLyWBClbM}|MZ38yB-3B
z(DIfvqKANrUkC_>Jf0Qmw6a!in+bpiEKI;PK&kP{+VB2x+#l7_FSn)HWjYD#GIgtT
zRSX6>;8b^MyuK`H)c3{x4E4<9%d|M|Rb+vGa=b4TKQcQW`SVG^sXXD}nT4S!ShrBu
zs~-!wW@*q5DZ`_0hNOj(mLpq;^fH6TsG5Hp{OjR&AgO40SX^w57M9h*Lg7)fg>Eene~=ja@W;fq0XhN+ZILX&>!h2%l0Ej8CUhnLGe<5UHy^35vE^wmudgCllNHj2
zR)r)b+y9G+4R>~mVqz&k5#_zqa#!8tuw0i65
zfMM>%6_;#2Hqe&_QXb|tRCPT6hRau5{+ESrRw%5^m?>P9BXgXRxH$Smz(U|b+UJVF
z`?}WE$?sm!e_nou%ZX6NLDacKPvf(6dT=U2+=-lDscfK>hbt_!O5@>w!=L{=zuUU{
zYGm^S8ZJ2@DCq77-De5j#P_xwsshoUWE-Q8)H{!k@8A;QHy5-yW)%K@rCDMzDcFpU
zQjC{rw9QcbvzMcSUnU5
zW~3>9wK#VeY2cn8%)yWjtZD&sI3{Snb?NMk2&{6%12eb(Q{i9W3CmMJz=G3k<0%;!
zU2j;DI=#iZ*4ZW?noKX1?t&O6=B~}n8S12jg5d>N5O5iLpuwhOCjs%IQ9;@rgr
zpJ&iaiuex@fC1yqgpdedyzd0{FxE@oJCwH16eqd8z_IaLu
zrrFI0U>Ya(LCh`wsedsStIk3?&rSK1G%~$PV}|N9GDJSIk5i7)e>y`M$PC>d-(Ao+
zDhRjG{p>R}8W#(P+Q`|Po|*I8_GU)285vZLaBp7JIm}QwTCcn75qU0BHPDZr3uy)F
z{xREb$nGmDu6aHpKn?N*A6buqbJexhXAZT;)wk$v=;e4*A1$_H0WxJ2t~+^pb0*g)
z{uJGu?%9#ZgL#Yj8$U(WKb`&kydbb=w%Gyr2?)qX=#}EPwXLj#k--j#Zy`HDanO2t
zo>b@zg3vN00Kj3bdk-Y=MxNVy@-+Pq)B>@0FHGEP2Bg9q;=+g5?$!m}U|x#=rIL@Z
z0h=>yAS7Yqfdonk7)Y3?5-<=gt7z!xo$t}X-~I`gn?$XR)BLu50rC;?B_%OwlPm!F
z0c-{Yw-OtKcZ+2e*M_TFE|Jnd_Nso_`+pCG(+cTnCLLdD#}6tN_*beP1cGyFLj8)}
zVsR-i-l+U}mu3FwL6lOV&F%Y@4QYPa+y*f3ELV|axAONZ2TB1Q#GSvtj?p)1e%Af6
z5DV3_p;H*SC7+QQ9CsaO1>1NYp~|*zz!hfX^lOg0zv!uBwA6?U($)B_>s3JEWa0qv
zoe`+hG>KeB@K>_zr)-Dk;lO8q)$n~iv&zC`21b@$j8kpO3p%S#ieqX6h4+l{m{X2I
z!KbrdIvkI)hFJstD8X!$3j074?e8b01=!RZkn2&;z|_Eg63{|~gnd}D5j;dda9L;%
zMeg-pPa->ERWC1;P;vjMjbNwn1}o=$&xLi2Gk!SqmT^(`BE6<2|9u8AjlT|tct!?lrNb?Mtc(wR_eEb2b?Q78t1`=^
z=&cOP5fP90F?_DLwXS&)le8bJx$Ja+mbP(4L9n^L9pntH{?gW`{fFsRvK(cbk5Zo#
zo`lPqe9=dN_%#GUrEp;09Ns;9F<^gTz#zb66{}U!uAs1gZQtvnT09PH`tJs1z{LiW
z1W-tDwIh`-@{|)E0Qmr>IIhy}g105f=
z|6|r%fMxNo3E7~^+Hq<#HsURI@3`HuHYqI%>#GCYr^jb&`#o>^r(eCE%(f}6(TEKO
zJH5a4ALFbOJ?4RUFu~nC>_D=_moImHPytF)n|zR)JipzJdW!*{A8h<5m&OoU*;4}v>#}#Fs0c<<+d}@Uc
zS^ax!tE!4(y9xN@V0$kH3}^7!FK#p&SZVx~g|0U++Js+&C0N@d-||1tZUnoY6KsKO
z^W|j16}7}+`pCjs;Le)q%}r%O01;-P(Bbc)4^|uFb!QMY>S_+v+!ubB$NX
zlMj$ARa;0`R8~}^jl);S@gVJ7G`3;c5Gs2lkhI|>k(p*a&PK=1JGPO(yzl*ZkC*5%
zj=)c1hyd~DYy##>vD2&I4qw)poi
zH4@p7;0+8RFcv}RJ+a>$tVMvBQugGu%tF(%LZ|dXkjZTRieKxXPiO;t;+ZxYDQDS~
z{2^){{giIXg#)RpY3A_G+`j6@Igq?awlrM>!E>6ctoz%BO-gbe{L7SQKSkI_+UnXk
zY?s>CJ#d~nMzLYbvFAZD7VIT*S~YWN2sDHV^kG??bWcomS*Y&TSLzjVXzMo`S%S6D
z-pb@SqEjmEFl&LknNGHKSH+H6kB{;4N3Ba&*cd67&|_Dzq6=!s4!qFUD*1h>4tM}W
zAQtPgL%{tD&Hmmuv#n*i9>K3@VBMcQ*EBq5J#Kaa`Npu8|5AL!Y&%686jvlLgX;z*
z(tf===Kxx<1{g-R7JN{a7Ld3Q1y(e5TsW9H;mgaA4A)r?%!(8vMz_pWHOGC;u&m#D
z3>FY)l}JB&Irtl~?k5CnnVl^%UN05|UjZ}uhR(=u`^6DDz#m@gfA{3{6e+O=7D^nP
z&?j2m+*f1Viy(*Lg$adMS_3hTfRteYnEgORSCfGk@%_I)FW0wj0WiB|R{?3+xfZ)C52q2u(Q0$IxOi71nwYSF$o)7nIK#c^H5ewJp^tPA|wCkUZ25JNWsdgkT8|^E}L|
z2O;D2UW5#%gRVXQ*75$`);<2SK`V*QqP++J)d@PWK!VEI^*}ST%R*NW1PsA=O`WLz
z2-4L+DOv!wQvR%Kebc{YuGHtIgl!WOw}1^NF#SzY&GXYEvCJV8zVr$XCZ9y?^N-qi
zS*h?9GCbynIhb?UhWnBwrNHgtV5yH*Tr#}r)?y4cclO@5jAW;(ngz+ISRU?eM6R
z*R+{7&wej&OG^eTvc#BHa_Bt>kc^wM?`b037)*G@{rq!q6^EtOgF8?0UIznE?Ecg?
zhKflh4}?8IIS#(NH2B&*&H>&qZvdPefNbC9vhO|b1Kxpj^e8#;J@(`Q%r@!4Uvf}C
zVAU@Yv*0FETrHvISKIqQtTVm`=|6;A!0bu`9klKs5YsSU^K?~N;pG9{53sT*EK1C_
zwp?=(8hvE=0qlnWPf1(*fgboj6O2z5iRX2oud=D>$qHcobpDK(g?FlVAfYDJu2=q{
z`$B;~{gdOx+#I
zk)!|i|FIwhI9Fe*T9(EOu$9kjN1n(ZENgF0uqYHjzxTKao7UA)n2h_)!W_VGxovhI
z@Y#TXr~bY9ZzQa3QJ8%RcM*rzaV3-if|+1F>F?7WG>w0^b`xpop95IEgR1{)EI%Q7
zQ!$2o1Fq#~K-AvjL>+LlQWaA=q_V+QqD^*1KWW
z5sh_WSpQ;T_=je{Yt3{oG7ohei>VOp0{IT1th`RD9PrJYf08mOG*9%@UF5M|&A>`@
zQLamGp;iewJR<<54UOi5PH}YbNS7lK7kWv)=HlE1oFP;8M)cZacm)yd#=-mEb>l08
zPd@q?9Z|`;e%<9B?d#S+%rFM@hv+j?(e4?U^C>b7U%^H=l-miz^1<^RCQtVAfUuS
zy1TnUI;B&P?m9F`9U7$L2uO=`NFyyJc(>o*|K7PXXT}-F(c$&Hd$0AZ=lQH;Dd8N8
zmjYszz0)M@;p!oKgAk^0ljzMfX)E+%RdG8fd)}XddA~{mHP4>ce&){Ea?20#DpHxJ
zHd}<){oNRKNywly==U4){tJPwy`0o(SOCNf&>O#N#PtNZsL7}0gV_$gfW$zC(BJa!
zE(8+CYoKlB6>*RIJY5fc1`5mlf*jQ
zn$qp52W{Yv-gVbC4VHhl25bwG!8_;XuWT2IL7M*7UoP|sMrZ#V+vs<<6Hy`6kAuv&
zRXd&x>Z{qe+Zo?0*uK@XAtNRIPVJoU*v64?6HLCiQf#mOt=k>FVoT+9***JBV?+3}
zlj>E#$E(ucwv{k$IS%WNDv6MWj56Sy27a{ulQR1H+QAsk
zUTV$kPwC&iyp`y2NTkDakn63iw7TA63?!Mz-P^uPktcfd=09~7v{@K)uYlvHTy}v7
ziynscO1fNq!1!C=PkNR0iZcSF=m8#J;W#|}rW_1P0o$3&+Z5TKAkE>|MMJm_1}E{a
zd0l7G8RaFDGuxRudj56yo}@hF+3!~O2arN}s5zb?u=rb!?|@CUhGOGS2ZvA1ypWz8
zmcB;FofPtTn3eFoSk$bBK#4z)C+h8HwWXEd(eHYaGO7lC^b~f^s$(dJS@R$;l6VeH
zx#K*2%Q^*?v6t`{>d4Rd1u633>3c5)ikff2WjtOl&SX)op-OqZf>)3wl00O|p|=(f
z-B>Khj?}Y3PUwsFs6F2+l+CyI24)j(x#ys5*Bo#=1%-U~6ONDTF{WYj6v;6ccqKF&
z^Jz18r)d6w3Ema=-CLW9P^B73N!4Jg0DvTq!w+w7RTqa_Y7nxn)Zm1X#78THQ!Otg
zzwa}kq^1UrT+`g&dwX@i!M}Fpa=e22r{L9vhBjMvT;RleyZ@%A|Hujd3i`0)Qk~97
zm@6rLAly_@UpUDPngQe(!T<3e!J?;T4V+
zVQ`$*WhF5rI4ZR;#m(%vrotP!XMGzj+Lzi^>=E@%IzRkg5B`O-W$BFJ5k>36Z|NtD
zn$yh;Q*RQd2RsZ0Q7*OPvN#ahbTy0lWWT&kxQg5U%Ghl!*i)xOf32Vzk|~{jMzLqv
zP}8}i&H{d1E#*zPLkzqY9O2b?M}m0ZF=ng^dW(ExS}_CFc+~L^F2#S*N8%?)|HS8o
z1`r#2B?GL1zn%VienMbUP)i=K{AbpjEDLi?^K8(+F8HnKn>*ygC`Tqvjw_R3QADd(
z$A);;v=1hn;Dq2K|L0eVoUQQouhd=sKfc1Ad+)xbx=0L=y%svs`(+%VLO2orCec%S
zjg!X*VkMx#PadheZ9KF<|MI1!RApTbn^`NgbXw=|@Y7Q?xN>M^Onq^06rNEkksD3?
zZ@o$mJ@Yq@HGOAG-^p~=3q2A$wkeAlyJ)L3)Y`tE_xWADnmUoC>aJ=*Eoa0X#Mc#BMoJL+>b{?KjduYw2isQPLic(wv4JmU
z0By{YqR-?^dro~+YrA@S;M`A<;pO`CFAGK1vPyV1GO`J2>47Z;NM5L8map&FK-|r$
zk^G>S$&4ohq%IB)6psFEH90u4AKux3WImRyzStGepTTqD<}w$hemu~!f4}C<3}(iPATOLMh{nWO`!
zj`V>hprmvL(4#R`oc(}1V2rne;#`IyS*CrN(A}%OI^l0oYITM6BDk4J@i2GYGYbt^
zD-PifdNYTSSe(7*qBs$#1zjb&9)9^DCh1k0>1w)7^?)=1n~oZe?R5^DfUX;WeY0|w2NsMnhZ8V6|0&y&W`QnEDf9YbG8#rXfuNR)bE8A
zN)2!#&fLoUzss3HX1fl(fMe;2dBP~VyRcp2Rx-7NOVQXEL%vgH>NX8yxvN})9x-l;
zip?pLSZ<>zd3XUA57~?@JsR@lPb5Cii0Mtaqp&66H0>8{+a9*#T=Y4tTSA7sE>D00nPZp=S0`m
zXxU9&I=ub68g<)$C^1>{hS{d><#w_E%rUDEi-A#_n*^W`dIHXOkXi?S1^4K_w3b_byzC8
zg0B9188wYfQT0h3by}sru*GDK(t)B+Er##YJg&25m!Z>ySq(
zFzFwn%ao7YmSWSI;n$r_$x$WBB7D-8|C9HR3e#-J3LsF~Hi?0ZsJk$`zTsIa=i{|4
zI6P+AV4p#UjTqc$CNI4?2YlaCl3-#+{9}#f?LjXQ?bEd5{dM4w5`@TfE%)ziXc;OsN3znb7RR{T^6
zwD8vee{H>u`ND^9D06?dn)Hk8!3_9WKzgUf0*p>WC0!xU4A9CoSmOpvYaWNm78O`u
z%3-N~a3c~Fko|}R<)3sZ#PSnfwts|f5jSP|iAVO{a|u0GT8rg`9(&B)R`kdF>gi1t
zmh8#_i~l^QS4XJNz7@>~u0AT8@%n@%pnaDzf@82{8R{h3Vo)~Dw#(PrQSB5k$8G|%
zv27H5pXEmBSieDw^Lntj>|s;YIbA(#+wd$*iW;_u*6`BX5BLL4EBDyrN7gEU~zYk_^cfNCEWyHm*g+`5-M9
z=f|N1LXlxWDFDJAu&BrEXb=EHaX3gFOshTXs8IZp_a`~fRm9Nke{p!=R7tkx`vFvk
ze?F5j%g`pL9BYwZV6=GeU)Tpa01!mBTNQp&%P2z+Fi5*6z>kz_@i@qlucmRQ7+pdM
zn}tw5U1^B*{kFAu!oQ8F%k``AHBqv@d3
zr(~p;HLKWPqj{ta=T{vx9X+rpeN0kNF7LeP8FpV*+~Xh_`erfBo!@KbvHj^3rrpUE
zd+p_6=HGTS4`U7N?+7d#SmE@h%cWrV7-k?^bonS-uj^Zfo(d?@!U(LN0m`rKLYYSO
z+$T@GbU2xmdj7KCGHKVP&5&(d-apGp*7q@)hiqfO=`#=*CxOE{E*1P8#w}ftO3z&i
zXW*B(a2s7kD=B*&fTXINC3LiolQo#>VMnOe@1ts`##L6jF&K+Ir7*D>m6i)NL1JVa
z0>WVQs~g6@FTYwYF5v$!Sv_S#08o)u=~)Zqg27Znz!FJ|hZp}ZQ+MMR4yLfcP!HV1
zs`U=#c&p3HZ&-fDkhNdlmUxEfrLs%2reNi){YS!hS@-k9_H}BCOz{V$5!dP457)%c9Id5QbiW`T^-F@m+0>)mkF{_
z!>dZ=-*D`qi|Yy9u#&r+)Jt2EkEw2M)R@crEtuC7m(9A`O}yfF(?WNcRD2aho-nmh
z;jssWFo&DiZivn5%Q)l|tum|k2Me=zDHYKmXRB>CuDel^447guGHQL@@T4K!lA|Yf
z^kC5`psaLCXoS}mLs>I0iO~6TmE3eYMmX4UFm9bv$F^mxWXsiqP13C)wPD8U%T;=*
zVXhgmiDgSEpfmu1Uhv)Bw{=;4#~WJm8od0#%ZsI}&yE0bGD^yCA8M~Uk1yH(U&Aiv
z!v<&m0-Zt+sDm*OY4R4^ma;U}EFv)j_4$1YC0N2asm?rh%_U;AIbSs9^g2WJIU)>e
z5jf=gpilhxpkx$j2>Kj~iG?@Ki_a?Nt=~1G?S$-BYY>!-A&-%=}1tKa%F%Kv8IJfIv=oCdCu_s<$+b5&gP<4}Mx
zjhTwZgLP_b@1*5%TpQnRMvHt*RC5hUoyXv*4%k>08C{Sz{#(&hi#k^&r5U(d
zIUj&+p`OIIlnrItWd7?J8bWlt6sHMsNOUsysVM7fpy9GG8*Y^!8j8yDO5q~*pC_o1
z9-VY`h?3+G11l{svvXL-ZvsjYVQJ|EfQm6d`c_YHSA%kL+F1jN9FYEDYFxZ;2*)f*
zz%;=`Csnl}wNLD=y6iSUJ<>gE?drm(86!a!zl@+@A;hBJE>wR<_{oKH)AZwQB@pBN
zXS3VXNU(=oK>nIPMkO<%S}#U+b~)98^+;C;W88*~IC$9l`qE2O5lwmiGI3(ZI1{g^
z$A7Yd2n~C(vU09mD1{uIm%0N|ft8)aBhn-eEM;;mCuQ8D2tod29{btUPs^;S9wPB!
z3$eiNOf_Dc7PsqndR2wevN8ozy50E8QEe1z{M1B1ozqn=nG>6wrBF#z!G_<-5Tdf!
z^|ngTP%Z`${~=a6_eq3U^`#wOG+sCN!0EVeGY};w-{$;XvFv*@QwQ(j5$P(9l4|YC
zZ!?KR5ghC5dRD0B5t=h%u2QM8^R`uA27pU)@hX_ubH*Qe*6P<37|K>oCk2+3N2t_^mHnTBe)D;3>=wCLkhg+hM}4=^0rR3HoD|*cLUbWjt5K)elv#FE6#@V$%hWrrdc%n
z5m~QeYfu{f%$mYRj(g+FFyFB$QG4ZoK=#*_mnergp&ly>t&Wr>^r;2kZZjG<%NTUK
z%i9wcE_Hd6SFqqG?Blb;$~UI;tin!mXN)me0c4;b3T;-=sl3q
zatPoK4eKwKsFa_}RbJVreH_~cbcAoVvtcRax75>nv8FAmabwSvcL+et683%}VzZ?R
z1}*Xt$^xqRzYg2k`NJw8VWHhWY+DrkUxN+!Widef5C1P`ZstGth6}3T1w!<;m0>Rx
zm4PK)FrIaP7DySI?~5c+_-R>qaYB!#Mr_s51Ez~j(3ggTttL8izhd0p*T9inVpLQI
zr1>?ZWV6uL6pJ^X(#L${v~bKy;-OOsLl5N7C(V>#iF(lF
zeItkIM#4?hl4Q{5#ZVJcSlp9bev(FY`Z4Eri^HQri-)Qv9zA6hHJ`ldErux>R0+m%
zV3~rL8Isj^W}?(l@v)xmB^Jjpn-_VU-19vw05y1fpKClhJg6QZ`|Gum
z+plu1teB}8IM=9{tsl!M0^DiDqflW&q5I@iF))K3Fz(rnII$8
z)B(H^aIL;C6qsBBl=5Nb;t@=H_#g3YpBzjD0ih814Gh=XBnF5#xv<0Pw|x~u>OJW;
zVZVtq6bE;ZpI{giWJ2pnQ&36jU+~sBWI?~HHe}MORqjXdWtd$v5uZOoQNuZLrhLcL
zy!o$&Kp+K8udCeZbxKo>6w(kln5f<*<>l!@8UqgKdUQk;V0QeX6NBb2NPb!LNhLwh
zZD0L94=)}l$|9dt4|57Bv%m>b^@+Uda?|xo(G}f~g<;{~_7NWW+vfPzrJHh*_Cuso
zEx27W4Hp5*IC&&}w%I#QWaTVH{$>V=bf1F?7?^t5h_UDJe8Pd5MdoDR_
z5zTYO|MPNv|4o_qaIphU+QM={))aDzB^M-MA)2*a+us-6eyPv&NgAGmAmXBavhXF|
z4lw-&MrS0%jX;~sq4mYI4*z%5a*a5#+V~B7G_~Dy<9Fxynql(7OWU)D@hAo19E@hv
z6f!KQ(n%7*X}+*>8NTVqefX`}!2T6JoY^8Q`1liUpK;84cQ4G=G#%pS8D!E-C#_Op
zh~L#vr3P~|!T^y4PSFaSRcYjLvkZ1-B=t(2nFAJjJZI(A3;vCZYggVBK`LsRN7klzA=`Xj)
zc8)-&&8K(GC*T$fqzP}Nc&CowO>U$0Y+J5V5u(W>CXg%^@vY1BV~{Q-Aov|ZV@|zA
zLXK($5R4_!`Tancklh5P2)?(GX>DcLVHdo|zN-|w?vN~szyHVOdjI#A5fDiq-+j5i
ze|h+enyZ>^Mwmn2c*Zfg&k%JaryLPRHNIWJIz|6_#_%ynG1d=|5PG&Ne~@&8wzDll)-wF|BwtYi1{R&O(B>QEn6eNSh`B
z7_)o~N+jo%CLGbQ_6AFypTRwTTQ`LPV4KF};
z2dxNcI7;@NKhFo{C!9R5J}yYIq{xzDpk(g}E8z;qGu~D_&X0QbBqU6=HZ^*Ax6px<
zK{P>qL(qkEwP^)TZ&z0^WyLn8VlFF`V4lVadBLkgR`YIW
z<4Ud8yQL;>&wYGFysCVXiDm5R+JYRYe{%J$FCltKo8TT<2dgeabIiQ
zpU3`i(LdGWAa*N(LkES=Z|Ei^C*V{)7bkXxjEwSrOpubzQ)LlHx$M-r4>lb|!SB^h
z+Tgd|ltUV9{@4Iv45^3H@Bgrs1*XIkc_u>hQq0)~k0PQop8O=8I&I>RM)3agWzBEg
z`x;`2h-@7wv!!yAUVwN-^s8&k=1E5^#_^Lo$H%5FL?M
zw1W8#QK$%3*t6~P4AcH^Gf1q-BhfkD99n2d+jG)|_3O$`+;XM7!nNwWd!=
zJa%oVd@z%R;=DvDGbXyp*CHSG#<$Rj4#KUZLo`Y4Y(PlDuNn0_o%6G{v*hxvupn}W
zIG41;%|>i^mD`)e?h8=n&-3A^9V}?tEsBVs>l?z&?ic04Md_5!Q<8AUj4)~T!7xjh
z`ia>K=$?>77b=J{@R3JCU2ErOS+eR&Lb8VJA!Y_I
zZ}lC1C20Z;UbVa~;1??-{z}Te?I%rvzu!ojsB{C_rP_B{zW)R;t-#CG{6%^qhS^WW
zW@KcA-yZk#B4wuN5sUm(A02WQlxXkLp;iBLsul+kpedP_Qfv4pmsLfAV7Fqp
zt`fi@BjC)NnhjO}Rs&<~%6qyd&$kgY85x>BnibA-t&6eg&G*2AgF2wA4tJ&xX27(_
zl{=kdU-Xo*`_$N#E&HQ+&Q~Wn5YCdF)LfKdri$yMgPUIYB~4{r5`Q9yo~o;@#O}sQ
z8@~5uDj{@?laq)
zI=)L8+3`5-LN?_K-;|#@IIPFYs{&BWn))lNtbM#md+f0x4kPZSPe5w`yR;Rw^A()U4sZ6uJ>Oj2chYjHybftr;KqaXxEw
zKl}fG^EE3u2Uk3V+3OHLQJyY(elUCbUSR)HxX`hBdA{9zQ=B%9(a|-hpGq_@Fz2f`qypsa
zk7Jhj;hM!BF&%NO_;O=#FAO-&BQL0W$(Pwo{`c%^FIx>*Tm56^D~#w0joTEdx-wj%
zi+gX5(qux_mSZtKegKc|I+dKYrUoqC6n!Fd~a@&`XV$NQ%=KS%
zQSUo$rQPcSWz(PaR%=FlE!bKl)d4OYdsqy@~>c;QbM3Oj+&l+pX?-@RB@7-7B3vuEhi=ld7JFnuoMBE@zjBSC8jf3IO
zP3+7wY4vNA`ao2alb)Drww^oFr3{6VgCQ9^=d1d#BcE^&!kpiB)|3x@nop(jC%9S%
z{5Y%_5N($@av2Pm#&Tf_^4T#JUS-vKK9zI&s9R9eRw4O%9#LWOWUI;vg}7Io^wbW=
z(T0pFn{7V&KdNi64G)yaNnZ#kx;)j(rq95_ptlj?gDN09zvJTRP|MdH19Q9bCSPc&
z3#+~`M2(!+e37!BBt&B)txQTkdTM5jqZM{qgWc3m7pIA}@dq^sAP1Eq+?j2S1a1n|
zvMj}oKA4ayEC}$ZQ~(I~V+xH5t312=h3w19Sx*;d<%OEP|2;Opv*hl0*-)PvIpnSw
zaM^s4N1G}}>Yl=RN;yO%@>8k+t&fY)MMq7o@HQZPtd4^=s{}O)gXmklLABmFl?X|y
zOG#xWk#cvr{_yo@ye3Nr9by1%v4^NwY7RBv>pY3I;+A$8Z?b2?*&&T{x8i^3nnlmB
z(x&|6{Z=BEStt8PTa$bN+d-AwLg8)m*YM3AUR@!3vsSB?+#A^jWB+Gbd>1%+e$gMQ
z^qYjBuz|hR!4T~c7{FQN$r+DRIkD`h5g?mUV@
z_*g;)(;YAMkU)Az<4e7#(?3pYJr&BnJ}sT}7F)dHrQy5~QcHPn=8zn}`RHW)CsxQ)
zHFO)lIGRwR8U{}6*DrsCyk#$TUoz*>(KKBQK6X3l`Znkc;N7(k`?cZ6X_^n3i(u+i
z4RABfla5!PfhE(m6*w>3mjRbDN>d?PT7RCv!>(o*CNzuNxQMH=v(YH{O2vlwLcgrr
zSIoanlO=^18>v})>m_Ft{J;hYnh`=c=B&JPDeO&VgU#KVG{P*W;tUJ;QHS!)rK!ax
z6DPK0wyXjPW2S4r+?r@3#jYUF{%gK%n-mCbL}Aa3$AT5?#J0`_Dj|}>A`PHPSn2@cC2I&w(VkO55%lD+RtmrcQineO@!N8%z9-jFYr)G{T!lJI1Hpa+}w7>oO-gjDg
zA<{gVOa$XF+eKyeZfc1^M{xMyM};8t2p{RuaDxs-S{k&vhUy;3LD(v~g#A0`t)@j3
z$I~!2eK>!`P9kCPz=X(Z8i!4d0yTQkq_WwN*LV0BS`oT=1G;u<*r`{4n+$ocAt8+$
z5&5-R@bPpk7*N#+7ESt2nAa=HbcL#q2Ck2XzE=Q3eX1;JcuSklzjXpp25pTb%QBFx
z*QCI8kqP9yNyp~?k*&7e^u__GlJzrXf*d8iO4hpo*~9+UL>Z01H#RQ1O!b9xgLdw=
zwM`s2ylCvhocm}D2I`P#_L`1?aNQPBm4@xrNmI!U?jdC19O`qV1S@@49GETVzBRk#
zAE7Afyn*_b*#*kGM%hLSg*>Mm_0ko(q+o)L)j}9E!Vy)1BTcGy2>neE)~@jPH+!co
zwtmxB90yA*2z;lTAJW+QX*3&Ohw4{(br%ii&PuVSZJVB9gnBDxGa!q$Dgh2rOs2iBm5oB}Wf#8j;D46A=h>^n*<=
zF*{8KsVslk8kRQR=Z@7(E6T~t=Z^!-VTE%zvtg>yBO40>Bu~qmY?Xp53OF_hcbJlT#}xGK=}$VYcX2Hj;!y$s&+vf94r%DNOCAC=EUk*Ij@VV&6Qv+K!y
z)%U}sS(|tI>RBclv4a!Glh!obNYZ=GrmON_34(b&0D9JS1(J6$ps|$=sh#HpWSSFx
zqjP?}FDw6r{a>cJecMw6n}z_Eks#dwTY^5AteKcD5oWj#n}ZFMI{ZAR-*WZ-wmXc}
zXOa-tci8_!vVH;VHOl6JJIAkoW|Yt!V-lSpY4ifBI!**qW;1+Ot;uUycT9>+A*6yZ
zW#os)GyZYA*;(7Cp*>Q*ct}-5)-aV~>0|~n>#|GRH;5mp!F6aFVhW#(+4Nc+G1^={
zYF(@ricQNo=*t}o4H2oA4zr4uHw=;3=%Et)
z@QiA6DZ0o5MTPnw`WB81+{(pTSA7}{{lqww}|X;*5Z9(k%|S8e?@Fn4ps
z244K7+s~_|F6mS78O#0Sko*}h;xSC3vKJs{^HFK%Lx={^vT}9&ZFLb+HK#H&@}WLI
z>kt)#TC8Z_I!$$SCrb(r2ZyfivQNg}MjP27Q{kvf+lQ5wRsLquSah
zJR2Ib6dn@qDr|-p!p6E^*$r4^FeCKMRIXv{HaashHZ{f;o}OndUB9AsrlDFJm-c%}
zgJMGtr-AjkIUT{5qK=>`y1RP%lVE57H
zNZlMLVt@&TKt(UQ|vsIT&Uw^3Hzx2Y-Olh_KOq`yZ1I@nSt|diS{+f
z7xl`#y)7FzmTK@B;aV)8Ge~0gQN>@S=DnM(BkOb+va7;@Jx$gNf9?#8-l=
zTMd;}AIzEE=S+1@GU0~xp9hZdNpLZ=+1#z#>1YIS2=*EixcC;h36F%ol{*{4R6?<*
z1S8K^>?Kko&Dr&iS`s%!Y__UUgW~;1JEE|(8ayY?Gw|6tJ1%Y_^~j?ZW3vCsY`Mw|
zMw9iZCtK8iVn%8{C-p*GX#D`ub%^-?;)gg9Q1|xNpAaH|Zp(L8AEsFu+#|sv
zVk4k$qMkayv&R8k#u!sI2alVWO4qi&?o-RYWacXRL6$e0MWlXi4;kneraOh0VolFb`u_$z!Vg?
zEl;f;zUK9$1>U4No%{=IkSulD@sV@Ux7b&LJa(Z$47LJBD91T+8lGit+A|TGU*Y{k
z#rFYkL`Dqe2^fW5xzN+UnelSJ4mwVz#d>dr@`bt&sv@hqjYLXrQC87~jPjmeZ}NF1
zd~;Z4y53&W;B0=!xGH)yb+)<)Te#a0wl>GxyLB-3Vj)Z9_V|>|Q)jzf$6h|FQs=JC
z$Fp&nRX)z?HfMNZ?BK`bfw^*83AV;$x6f;s9Z_~a)UY*sa6m@wY01Em{RpZA9-c~r
z^~y-T&uV1?BRl5R*LN|Bne&rTozc3t`-3dmSUcm^jmXC_$4h+d08RqFa&ge2b#aG_
zJF2>Jng&wU#P*s9u(^g({*`M}bt9OYXjk>Zj~w*o!%s(%???hqN!Hsq0{#pIQoj0o
z{_6SN-@u27ze4{}V6XK851a~?9si{;^iJ2G1PU@dL=FBPGN3~9F%b3jWZPvP`go+f
zPe`L=cA6YzHRU&iBe7_ZAX`7zKc@25cspty)`RxRqmtMChT~OAQlBW&XURveAxI|0
z>>M+xKUvK!cC63tpf
zOH8XQGz!QOeyn`6dM19milod2n{=#9PT6o5^e29O*G%Hw{ijnlBiuSJD{!b%DwSr2
z=R@bSuSGQD0xohBV?BV$b@(gQRIO|#38&WP;CKO_>WRKvK^G_arqDB*nyx1ZS2V8!
z;RK0qJH>QbB0_#J#*t_@9-JuQog(2IZyBGIkiq(=UIZ^o9e+>nRcRVx_aOlj#F3{#
z0g@J
z!{FcI!s{f?o20bUGhoBRK?SQ!NE3!Yf8>yS4@uM{PE>9Yf->}JZ$q);IXF-S#l+vD
zS=Uf%g;`9pr75bwlr>f*B5SJft;v#YgP!4bB{D{hqz3a~1{e&m)jG{(RaxdC7@X*h
zy5H&>bVdpee+%*xLd9?9ME{~&py{-IV>P4k+XqUaZTv9^ScAui=lzn2{i#q~4L(7#
zp61{p-3O=zwPZj1eB>D!L0AlZw!iS~i#Ju4bXKF+v}Ky!MZ9+S61p*0;92Y(iw?_d
zx>cajBPqt@r@Ga4Ej5g}SwNkoP4$pZ{X*jzkmYAR5>~2wrxOELpJE{F5_gIcX2k5)CO-m+ymX<+U9D
zKDoaF4;`etxNyKm5uh6e1Hun-8DKrPx3`0J5rs9g;J5w#<3NFgL*Tdirb$-@d{|@{
ziJ=GkU?UrN8Nllo9;2?pl9C8^(|r7jvg}9#KAnc|mhwI@djpmYNYk{`eC?1MhDdj>
z4jFTO@v0JmSTyK>JP-H)c9V3&4#4XH@bP5ipUj(`b@2Xy)m5Ss1oLuh9olT)wUxvv
z&BOQjylFl-oagGjX-_nE;HOE>x2ZdOF*})si76${;yN3)Y6QM}Jg9R9<4GJH9Pem+
z8YVz#$10NE>npZb|I0s_^O_P6Nj;-dRgfOCgg5*)(u!-4RNdYI_UiyBHIcw3u&(16
z?Bj%ONr%MOI2a(AkE+OkwiYUh$=4DZe`629o*(Pw8b#W=>7C#<)ZBLu0)A4df*?Qf
zC##%&T-LBJfP&t>3_-+F&ka8&0q8yUc`1X^$`s;&B%SU})
zxV+3}nzQ|KruAHCFg0W0l2{H7h8Uwct8Tx5M%=0@jV$Gc1rH?R@b9FePP$HfIKLu^
z{h=vI|Ee~-1xb9c@sv|(3W52@RFa)}__GNqguo@tl$SOmSqSx7#a<#dGx*$YWZz3r
zxME|!g7x;Mwat$w!7e4i3(J|jzV0|vG
zOlfZNuT^k{G(KPV}f5mgvE-OrGqe-qnCdxEuI$o4i7bTd2;h7
zUMlPgydbC7rv(4*lyeyd;cAcu=K%(^a
zINBcSs1-l8DfX#V|08Z^S}*s0`LF1&=^zas%=cxsB2>FNmQ`cKP~DPo2F9;WnMILY5&>hiR>2
zB~=}hsME32CQD6KdJzup>*U4Oi|#vr)R?ehI|6YzDX5-C9_KmRs%TQmnOxK1&wZyG
z0|k>zVSO~`7q#WTrx8lXC}tkI#GJn0N#pF13Z%;KHtyBrELk{a(7B&Y6hyy-P7_lR
zA_-t5Rmah>6`iFn&D<&;8AvdwAw8Ew$XNH8kbqCPN&OCkm87c>t1(!2%QNpK3Kn!N
z@HGWK^>SkYn=a~*CDJYB=5w}yoI>@gECPZJ%-Skujfv@}7%+VOOJ?=I^J
z&WZlycN9diMIrTBK`yEonfj#xR*Olh{dWyQWsD8z6FZXQ0*bds7TpAYlrJORmjx>HrR-wV+WUwXGX#EOeBRLf*T<((pJPyN@
zipIp6yelszXLZR;(qh!_Ri!ug$<9v*?2oq^^YfbyK9i(+}u=^74Eq;5cOEuQj|LRo^}EU2_^6h
zd7}SNs#Q>Stb;j*!B>X=e7jZ>o3xoEB_~GX(g*l3l6Zud2&i&PVuJE_^uia86;p1b4S{p>5
z8n5sD@9n2Vu#9iP_Y9W2EZ8qn{)}{2mjnxzR;P&_n{vNlu_&CU=$FX8K*KY0oV)aR
zmqvL<&3Y>M(ao}la0_kikIPISbgi1l)MR)E@kB=pGtbw7yFiy??s#m+wo+_Pls8RE
z-S%FUOcYcXptKNS7F(tC^&P05Uicr#+^>V79755n0f;qVpDgbV?cU-T)tJ}zzTnJ?
zjAxOnGV*aHbpocTq$gzpF6pp!dtip#lb07QjKm3Lx2c2;o7&I$Yb8aLSX7W}tiH+J
z*HqlPrQ+^z
z+S6AUw+!E4aM&-HV2H$qOy7D9;dNYu#C7rrOhr06-AZ)!UP+pFo
z@A!wROvSL6VFyuGEZyk&&)!tA&6=lh_8{UL#c;o#Spz{8Q#w91UmlxSBc;PsCn>N>~I7bms
zm20FX`j5feTSm!vd1p$!YMXIa5eQGCRJM&Wg`lk$$%J=6o1O
zbY(psi>YX6s{)sEwzKu6F=fY|UkZmc&}YY@e7Z1tq`ppVFNk^gm;;fe3Ag=gTWfV)
zPxh32?`LM+!aJ~@N);GMA%8XQ9;e4z;z+bLZyz3>t0w3_L6qAhF
zN%1F
zS0)nOKmKjDxfjOTt_)C@)8rpAh
zOCN+cT=^n1m}5ddIL!zVka0Okr}E=$QQ#(^Fl`KhMd@5b(O5h`sj+qXH6u{V&pCEU
zaDRf4*V#%AgPO^I=~hURw@vK3XZFVpl*R=`tpozEJDiS7InQFDsKvCTYsKpKn51!$
z!D>TiuxW2|9`BC@Zm^p9)JWD
zkv=!+gR9|mB-?ur)y;7Z$#FvzUoqPFF$N1v%*Zt-Dd>6kajI5fP^@4N>VX~
zsCc6}Q_qIT22$j?(AFrX^xD76sObHhim?uhk5ED#NVWE8uZy|FA2e!>rn#xEh>^?>
zz~1pFl{NTSL_~j27G&mi3Fe}C;Wu~FyesGLedavoS3g7jTtC)peYtHad5@Y=NBENW
zXv8hMa`1Sm6?Iy%_D@aozo$S|LyrJiZQh(s^+@?f?(lCo$*9k
z$nk+GZ@}t^d0FcxomAX;=n#`YK%gy@j1`6IH5(YH>B9SWu`}&w<%u(0qV|U2mesSL?3v%Jfs>o1z@rI8NfaD#+hcf=$Abm2Q~Gt?c4n&L3o7vrIOB
zqy!4=K*3zt_Bk2v>yHn8<~r%qN`4%IMUeQ!qhgy2LwyP2F&M68P+`p2?!y$4%os}K
ze!SY&AAax)+@36Yy`SYAONTcWh0loVMti`5pU>aLC`D*hR7A~CrZqm~)Q3h1Md0?z
z<3&Sp_x#YBQF*>eG}G>QhgKkwoFc5HXmm9uO$H9wx>r_m<#
z{M-~Pe-I;T!@1EWsJGz8dtmV`ag6mQk=1vg=ITo+5M0%?SWX^^jH#=w_F9>sUry{WQ*Z!1Zmo4)Ka|+B_#|SRlL)(EIwC;mwO*1RMw7i~pz(yd)O6
zsgI0S9lr7{vzX3JGZC>jSRY%2s@MM1P|w*{B1U^wTvGR;UHVhe)RG)_k^QW0isVqk
zPVw*OVw*oGv+3y-)`~5^${ove_#$_?NT}AXme(yX&?gRPPi*TJX!jG`H?2~Vn+w8SD
z4@k`e9f9}($B#>{wIkxjZ}fX7^^YUF3m^76Vu#HEvn-Rqr*BBZN%67(Mjk2YeUo3B
zD5YoEqXCzW^aL&>Hey}~l7z16sQr&Ha!Gt+2!)Jpv*Ki1%Hc-;GH`z$@g&BFR<&I}
zZCQ==*_uFROO)3kO;|KB
z%8qzb4c}3QcnOr;7{k7%U(=QD(mL9vjeQ|Q3MpBuQky&A?aT_pdwV8)vs%>Ljo;J#
z#KYs`U(3ku%h?@^$kREeLczpUw0;5BEc^!4sufZBR3;AC5?Y5GKSxg%njONOc8T4A
za&PNxpUXzW&+<9;Wo+h`6a7riwpZlFf`_&tb4<>h#C-bJT)P`!O<}?HSiA&b1HZyX3OzNaJ?7
zw8*5J&03GkFS@bOpXbDz>!x;g)yThF;j^#zRz@~N_oSN1+qsKF@&0r5sn%&__i9VW
zi^YP0OCa0FXt{Z!!3o95z)RKuA93W#e2{CqLrR7-;M;`k2rpC`=heoBJ#T+SlA*+a
zN~?d_q4?D}40{42G=6y8|5a%#^VB)|_hokb`3qcAy}qsi6P;)MmJN?@{cU72(PpMv
zt2l!)t7ICY{HN8i)l$~+D55gxXp*ZL{Z3!7PPVve_~T&OEQ=1bBENOZpX(w!iH&%*
z;zCI(tNOgGE1eP9opi;X`<)LiikTSicG}hiV{L{_PHIlYJRv7%wI?N}2~&6Fm|k0w
zR7sE)FPd<{%jZLOOO#&(Fx_(2AHB92q4+z7dTIz6>Gc*9&6_Y0{Uuc|EQ*@ffE~Ep
z0++JLDqQQ-W&?z=6x_-x?@EV{{iaau2}n+$x%ie%&^xsQSLbCMRKO#ia9E@kskN!j
zXEvOP5u+w>R?G4Il08{p#o+j8=-*10p8P&P%G;BBN^tXR9iBRLSpUoAp~dyuVPEmg
zLur#I$8ZFd6Tnha`bGHAr*$9yOj})dDL0gGrQPcpcAFC}6>DMC)Re=~VsVB?u#Z(C
z?Gn|$ETFJ90fhDU28juH;TK~r31yN4=xfYvjr^}Y43FILRsSC6a6%U8W}s_r6U573
zcLNa~Mt{d5ZM&Eg&sWoP--QTTj9=Wb_uEyWI18`f;Ag#TaPqD?!^S)o^q|Ic+S`7k
z(?|=c-}jc8Yq$9hpXj&kw}+OxB@UDi{dubB8L0MXcKe(gHvJLF=SN~GdM99~`@M<*
zTSS*8TcF%zEEXR~P?kAECnJ3tBl-Y$RGea=UDR%AaJ#OgzdJINXAPWT5Twyht=_@p
zz;RL=HSQ%_jUk{*Gs@P_G8(Jqdfv(++>riM-t~@{T}s8`+IByIx*eyu)pWkU-Us4~
zfkewghS^8!56?1Jt;|i6Fc;%m73KA7t*hyxx^MoEyXX;~ll`-M|1HOJI{zy~_tjXS
zQMulu#vLu@3s<2COShIkI(#Sbb6q#w9Y2tmbXf}bTz^fShhN!p61VNfLMBh40y@}F
zQj$o#+34yV(Uw%*VU|m0m#Cm@T1Pj@B#w{A<6a@psg85N1&ddmzxP}7SxQt_b`|m4
zqU`Her9vgd2%FedeeddsIfABFmVPJVQP~d2)3?Q+`amqEr=mYm6eJ3~7o7{e5VitJ
z%05402Xpsw?au~Mo%qcBT(H8*ThpT8tPw{=}Op-HsahKK|bS
zIF!R<-p`)^7Rf4deFiX>?tTFwj|E_$PnU_w_v`4KlQd0GcK#%qZmr0vlIii6?S~P#(&%Y&yx~S47i~k)ohKwS
z5QnnOU
zQIqg;ZCOT9LO<0?#;LnbVfr`pRhl5K5>1db*dT74E7Xta&YVQ`>9C@0lS^$zB$S*qO
zXH>a?A8Gg-j&iefNvh(J|Gf?A`HjOfB4KPh=O@5g(UPn?zdpp=UkSb|tu%IE=n~}e
zC3a%f4_t~bgBhlc!a6BEr70O|n+dOf6a6d{4H810ekyrlw!=485HEU6?<09PGc{lG
z(eFx@252;+vXJGLCmt3@pE#ggIB36AXYx#A;yTZC0{=!fBIrKwd6^T3iku;5pf7Yd(B<1ad1-Tago|9&u=e@RI
zAUFT4SgH%G|G$I*p4gHH7tJ5I3_3(98(jFbagZn5p>ARirhHnyMeh$uI5df598SLc
zLhq?`ukdYsNyRh!IZ5;5hX}JfdM7#vK~
z!DTIi?-FgPv;~2n=E!;XXOlPiyO6ulH8%~(YrJ|TjnoYrlnuB@G{j~MRDs9+zI@w>vmjXa=*Z{AQ=gW&L@WN0Z8@pE?A36^%{Dv_%`cLvP~LF4b|o%j
zNCW;UUyU?gY>}YBW2M?ck)P3Pa1L|98ECcxL#gn(3Qg7U)tU`T+rmubC;7dI2I|rB
zM@f&T6J4u=U-&=m@_2Qa#9XnEZJ_@}`Z}6GwmcroBFmU^gjuYknyrH?P%jmm_gQ@M
zr;Cngv#)W+A%qx;i5z(zuAb}@6%skOsg$Ba>5GtDI}7g1>o47b5XoNT6S?Q^WfV_w
zfn*IIzK`t5Fe8r=oHVd0DBBbb{}iNp3aa4n+C&h&)EnIc@FZRcGI%8Ep@7+(#Pg3c
z(dPrtYWWG`5+k8K2mCwcI=q_e4~rcprHQsf@M_=qlXUXcZY4lFzI=h3jWL;*2)#mm
zmsRGc$Ohk`?%np%X$?@kW8HB)vesrMI5}({)c&{o=b)RHE*;T3>IKh051O}laPR}r
zMqe0ofT5ZzVeU-sa6R?J>b|{4kWjpKlk50uqe1M5)=w|QI=*Ft0l@bR}G%eis
zRf<#XpN(@XF?AXZOGCL6$ameeP6b3PupNqKoLU$P?+AZx|LFXrqU0`LA|9-oHnJOR
zBCYt{gB8VT+;PKwCsR+0j{sC$wH=b-1{Y6Go*f|yVHS*0S$5>vZ=0Io&t88lvUicJ
zy|#PrDT4~CvS|ECCWNkVNkgWNJkM40Ep2E{Ku!ojCBVK1fbsZUWtHfzY|J5yW{&9Il@@)k;`mOAs?vjk8X=%j?Y2B+i3P
z&{qcZ4O>aNw0|3k-EP8dYDf$D7#7n)7&_M6TXuzyT{YW
zUkVmkMEy2wfJ^Y`25b&bCPDoo%}(r)g+B}N(4jIcf3}0M1n=ZvHuxllzO86K9v69f
z2)Qe5_F|q#h#|+WAYf@G;k)Yfxd6Smog^`dwmeCZTNEf>WS)
zL=bNogvGh-m8D_cmg|bV&8vD=5y5K_AFPOgyXXU&I+
zu$*`x*}3NYpihbdg>7kbwN~W+pb}&(-bX{b~m~;*
zHtZgYUA*0OTilGAgZ}6L!ReG%c@p@Hq^Smr>X>p
zH^%szafw*LH*X_`sGh%hPF%=c_4@Uk$VHizCRc{-U8=G#C7KXBgk)q=)(8T=A
z0-ci&j^_v-=yY5hhs0L-+~AVgZ;3Xi5VxJiJmoZ0v6!laIQ6NG!b-(seUOlUI
z=2Tg>P~;3n%>oiaJN569RT4lJ*f5J1?JJtP@72k4B9yaP=fC*f+}t{dAAh`H(u6Nn4Og3ODRXsz}uK#KeKlg=0MALe^!YuO(h-A``tE&6{4UXX3KjUJ@H@R
zqfL>z?6Z>SzR6KTLO*ofYN=_}!l4zHCPKxmj#TZ`87=2K;a-E=aqDDh+X@_8~OZv8_RRvikmyAEQI7vcZ?cz4_e<(%3tPx{=@x#39Eco_pl|T3||BRwf
z(bq_4*MgRkf65UYJPttD0-Oy1GheE2i9X8mk;?!yKKDu2-PlOh-pL*XSuz2hIv_WU
zasey<=iLzGDx9msO$JCTW<1j>NDOG6>H|VRhGFnki~HYR^0sB!0XAh|lN#QD&yt>W
z%V~;dpj3Y#NSO$LKR=ScYIH&a84Zsyw|t%E;pSg9Z3kr_ok@p7jB@(g9_4aTLZr0^
zv5s7m%NWGPVr#!6u}Vi=1d+*1$j-QctwEhFN&l|)gUfwPNza>@NJ&>3kDGX#a(?l;rd6~)GP?c!QX52KFxXk
z+;>U-;#x5c$(KK&_^F2*h>mZ-F#b(xqB`mdF)ex{%Je+KrXBO_htAl($HoY|`=!$P
zog0Fyjayvs$W{5L&$h$0^&ylsO4#~GX6J#
z$r68Yv02Ao`>{M10g~imEJC^?gE8}>Hb@saDp8iwI$=r54)=!vc$2%lgs~$gU`1UY
zqwh{HrzN1U5sA4C#=b)*WR-?cKvGwYvQ6~CWC<=~`{j=$>P;ptWs=Ehc`G%u<7VEb
z1!X2-8p40#Q-)}o>eQR&(#tX)3LX`Mx_IZmiVr6^V%3=_)iqF+_n4q3ISliH9Lb&3w?
z&P!h9_Wgod(0PCP;Psq*;dx%opFl>g3Sw}P9^EyPy>NOA;~on6^1x8%X{dkoMK%Wz
zMwqC#IrNNbwHx3g4L?#qf|y$Z0&qa;wBXDd3f>(Fr+@AS~#ysABS
z9PN4U{Av&mHoncXeBLVv7&?Iua^i*dZoqS6?TX?x!kC$Xb?pDjTJynPso8tBG-Z2_
zFw5v^*6ux2bLn&!w^B$&UQ{?Mg>u(_ez|678W72~X{o4H4$vDa<;*_4y!t1tJ!-!_
zxP1Zy`myx^$+eiXm4lnbc7Qh=1!Sv)pATeRz4)hYl#!F`+E6@`h4=ZuN!oyQgYRKo
ze5{KfP!ebqC=jYNd^Z!)TZUgPRe4jlmiwEQ`qK{lOg>nid)!g?Bil)WS1^E|TDsp?
zF}qH%!(Bt@aTcp&f39HZVJ<00TME9jWHNv3@*Pq*mTCRUD!x8#XTPR
zQSV>47^LTL;6+R1)~d_(%ZZ@{2;M>(IT^cLu=T0bKWQ9`%3IQu3t
zRp`xmzp4Hr&qWB5&{}DoKvTMU*f-3@2K8JkPc3&#VViTLggq`0sMGS@WsL#fzVL0KC$;r
zhB{lQ=1obru4uk>4@J3qyGA59ug*}!^epi}9+&1BaQ{RA3{6;FSO
z!Ffr%nQzJpStKc%s&9$`Fx%Dze-vSX(Nu|p;vF*?V>FLV>oon$i}3$Iwfuljt7Ur+
z@?C!@XVI+HGO5()NYG3!p7|9sd(j288FK;&o@fSfiYDHKNd;AM%cy!n&=8^xJR|@0
zrDE?Y*%Nir8tI(DE!oZAGLq$s*iC~hq`tz-M}xACYceeoR&$MN^It?IncPq_um4cz
zP8NXjAHVtY^fR#Zgm}nB&6kvdj;*6EtULe0mPGsMQV0Wv3Lk|`n8O=diQXyA=b1Vx
z7_bs+FH0kT7HLAFiJ^jyoZSBrIfO7wZXM&4xjz2od+8O5yPC3zVp}e(9iH4?ktela
zAeZ@IE=q3Hlw2CL2T1>sB(9h*bF5P^0=(byL&41VJD<>$@5vC$qr!I5P#Iq6y4W#0
z9&wi$sTCP%0+;WJXb$c@Dod$2wU<)RX>3Vj)x?sFNzD$uEcB)Dw
zG~AS{j_)l~+|nh12Nq`0J1ZEdsnV*@wK3xFi|W+pz2W{h%w&)=C`gd}ND=jEXj(Q&
zyEQE{*u#>!q7nBoO^;5hp)qV;l(f(<10m9a+Az6?%D0s;`eP1$_&3Np=%aB)WD1J$
z(>4zKtg(GKArN(ieAUGBQN_)ag6-E2Jk5@a74s3;$0ZbgA{pl{Evl(t0>Q2<4{$@n
zYv0qlX%!&0c*8&`8hN5fEO@SFdk)O_qjaPWGD=@AyMv}qcIg;!>LDn-N8yc;p?IeL
zm^nW=6(|bGMiq&UXct;!A9HSZHwB-bo=52DwbO;qXX8`-
zl8r!$WZHIPvqib08Hrs7-2g!@9$Vl0owVQe2qZqhb7zH~rQh8^48HlaYsQ~=av(t|
zfad#&0OuoY(0MF(43mUO76nQ6>`Vig?O2zu=f~NF%A)U&o{D9~yXhaS+~TwpKkU0j
zDqY4NbY@-6e=xp(MDdTxsPcM;;f&gxX!e3*OgjC)WcuM_m{Z
zjHugsMDh#L2G=^V6iu=Q1+AY`{E5ga=S*G??KI2LUT?kkIrshB2KKp4$mShP0~G8$
zrd_1$%xekorZJ~|9Lw=R%Wl^n7JKJ`!5O0;A)f0Zx7Yf+igkPm3yBJ97s6=}$qsV9
zwqNBFLl$I9eIl=9Uh?BjriKf3MI@eYCQ)Aw>Q+8=;_sO)>EZH;olc>MD55oeW|o+U
zd5MrIzHvXK&1bO1d3XblFA+zPEE%k~F3dPM|yo79&6~
zmunO>eSs^ms?p1mICn22e(QNTU&<;!(TvZUB6%8q$M(GaW-Q>mJ2<fC?nHN@fWO}jrSZ*Q&X1*^ui^5!RZ4uMq}Mi~<@uN`NmfDrhy}@R^bu5IGpqV-dm47WIr>okClJHaAoi_L
z{VTA*VT>a!n}OQuebrlYPMuX-A}uNnPCtJ4KB_8x$?j8RwiBPMX{)?2r@?u^+z!uq
z@Nq8%$-or9AyT6C+K|Pif5HXB7Be|M0*P~MO|SW=nnoY)k}vYv^giM6kjdC_?OKc2
zSjRte5FEXZb&hKguG6%wm-*kY!i|7Lf+F&BdZ_N9UsJJ`jLY~v*
zLR110PeQ7FY)|grG-u%?o5E9xr{%*eLNhYDrnj8JXXP>$e;Y{s)$+$jaJe2rD`gEl
zM64}5Do$Gvf`arv$*=q#x=j$mF$@+;93roa@@OJaT0^)ts+rpN;(_v+mU7ou)A=v#
zpq=g|298WzonI!7j)q`}OrUyk@tLcUBA=pEGyBuq>9smU@wQK^!1Ee6`zIm9X)uL>
zHBQwNXGLovnWj|V1g|6&V;a_egae#=&kLM23Z!!f5+D9O^1Z%XQ9zfyqHnZx`p!iH
znmesc36mgyowTNb>BcA>tW#FPyWQ7SSr!)mT10l`wyYP0ObGFFSqq$MunlN
z{&=&^!@G8Y=mnEC?7f{$*Y1!cTw)bEY(w@i1;=rsDV>*@;L6oM-5t8rIvl2!KYe#lm;N5JJIP8iUy9M3*ceFA+ee$z)dZ(qcF?`Oy
zV!hND5msxj18Wu7e8P}=)S-K2x6xdh
zG-Sms?#(9$^i~a!_!*Rou@k>f%XpHCV6Bgb@F_pKS5=v{6JZxk1eA#ns;^A=(BldA
z#xlmwXc<%XLV`AA^(Fk#zW>17r7_IUNv+CXzbE)}heQ0()JrMqbI`bgz1QK~6r63H
z&l13=sU290H_94d{iOhJxO%;M_4bO}>EL3Yukjc1aV+k?kl80H!zbeZ>3I>*2pN_+
zcAt42W@?x%o^cS~zql5VNB#E*+WyrkK%OtyWu`Vk6D_SsO;)1{t9Gntm-b`(mT8jg
z;yV6!P3Fy#(H^3Ms8AxGjh{i7d^&jdalFBI9kr~RummDl7gxL#Zh}c8GHjtBSv#Td##^
z3SXFXYR<4&YD*f%Qv5#jc4)8;2b;iZOWP
zbv}}POn;_&fHrn}&LiDS>)*YE-^#&lZeNOEs`YJU0NOVbSWF@EB61RattFkJ!;=T%
z&QA!>h%uf~B+_bgJTeBXk)89#F2p)*t{-to=^}RZ?kdW>32MP1_?ltuj?wcUY%C|0wBcxbnn7O}IEniN@4oR%KbG^6g
zfMr9hWmUmqPS-d_!e1fI`lx(lx1V)L`(?aR?HejhRbvLO${l>;D8I(aYp|={s3g0?
ziqSePH4rcJi?3VW#i*Xx?SbysH=o|W$YZc_vgBc3t$%!7UZ4rEGYr&we9rZEVxPt1>4=lBySW
z=Nq{>(e?t|n(jVDfZm6CS7#UL?IvQn{KXcpXmyA2r#~3+9L>s!Lx7^_Wyic|h|{#H
zaB`R>6H{!yU2Wv%Nw0B*8Qm|%;ZcX!(Y-h^gb*O8BPTLHo^9Wq**VEi(PZor+w%+m
zx^;Y+ucJSl4KqzR6t`)eM<_&&D7n07+Fv3@$Ao)@@MML*slFj>3TM7q;0%s(JrBPe
z${G{B@6owec`g30WZ<(a6N1hZkBoyNJ3IZQ?g`r(l^1jsMUV`OT37pMSktdv?I7Mm
z$6gb=7#y`&8WIDc1=HpKWjLYv$cZ>JZLPt^*^3;OGU=_wMF$%|(oU$0PfSoSemCe{
zBYlY<{?usBjO%T((3ib+}
zV?DPv_X9O92~tqcVnw11N1S7UZ8vS%a0T{XwU<(BnyKr55$L+%GHI$QrnUUfFUPvh
zxKnHGS-1mOD9s?Pav9#*56?v*1l=q2i~GJ5lOp%|5@qN2XxFhsKo8h?-{w@@S|;5R
zY~0~4cz%O@OEa({Vi74W9gro-orvmP+7!X2bSS2`|7^KW$%QE
zEi7vUjoMz9WGv&mW+R4t`|Q`xJI~mGJbFMdpq|UR3#+RHhjBeEkmz-jiLhts-`Khg
zniFG6uu7j9%aT5?U>1Md$jqx;*Rm2@?F5n@tb*vm=3bN<)AM!F471cXX8N$m?1f5ALkZLskO}_#G-7o2dk&RA4_|7k(a*c}=EM);tSc=6>N-2m
z|LNBExpf}ZY`J00xZ$x~9hBPX_}h}yvw8I`?6@-a&obuHPSW~6v4bCq!88esjh0Lz
zPh+t<)F%d*ucd-_Y^Jv
zro)W=D$l6=lWg_R%eDkMEtQbO?N|EkTjo8(fgOmL4-%iQIBMT;C;_!jT
zoHmuQYhrAgz$Wk>xib`lOD8UyJw@(}It;_|uJ|a%k>L~%T>r|H(HcIB25#Z!D_J*$
za^cKvt6Dr~hC7{@?w@#~>0_5&G5W2}z1&Ww-KLr-Zev&rl5EE|IuF;v`JkY#&KAx*ElR*}bv(LNDAJY+~WvP{VJ0r-?wMUg%EHzDqNY
z8NhM7R%3grhV?F+f(dxudf8g|TFbG`E=^1DfL?rP?{669B25v$UW0XP)eK#f
z)WaSgJpbcio5EuJ3_$uSc;g-t@(>OWt)4Y4~Due9DADN
zPQl$lhW{3Kd%3@Ugm4XZvU95x5X|vlF*B1W3^k{eGSVaToPi;b~+E4KQ#
z^sj!|U(+_;v*FpZwR9P5-z9qF?yI0wOlbhN_i==H`xUoly1~c+t`&9Jf@%&Dk#c64
z7J{Va6GNJ1O9L4_Wy}}s9(Gf3@~t$;naqT3AjLRb#iM+TD2a9L(jq-CT!`!@1Y*ZL
z)6pqKElffq5N#LyRq)`$13$kaNB;35(Vx
z$EI84_Kf_4c7elp6|E3&Bx6tWNSUmV9E)?|p^w!e9=Fe)tR_QCs6~+3F4K7It{Mm8
ztC&c$Qbn8qIaksCRNptp9*33GVBM8LJ}Y@PlWc6WaKsC{FdE|G-8`%w9Fq!}QF_!HJ
zx0n}9gl_l&I*M04_7hw5fJ3-3{jnlSP36;OD3eEnrbdCTZ{`Obfd}g<<#cTa)?mba
zgqKB`^<2#|6Bd_F^}C_@CC$_!0X_Lvfb?0t5i+n*Yemy)a$y+P{6%9UEZ68Ta9O2YM{G$8c
z#K#V_L=mym!P$4?thv5bjcPfqfDs{-!bo1T$AOX%yh7ze2jJ3A+eGd})*-cdWD
zpSCWwPprbklSE;DV`q;N#mqO4;&MAB0rz!y?bELdE;SPc!x!!6%3fO@M$>R4_HWLP
zGLSJ|;fo~24@Ry!Tg~OKPluQaCj%j+zuGUa6$jV;#vW9EgtRoT20>>K4NA0NH9w`c
z8&y`8t3{Ujbz-ICoI5MKyOa|GMKb0xJCa&p1XW32v`6WjJJrj0F!$mbAp6N@>gx*9CM
zA_pcCfN;mFTtHD?C+JX#6ivSTEz$yTFT}mH;@#h9AYc-&4BVbzRqw8kknueg=4MvH|f;cif&GD=W$&5@aG-eIuguE^5m&1Bb1lK3816ESBehjn;s$
zjS*^sbSlG4NzhL0exBHtS)cgJD^1WPv?tI&)C_h=fj63MAFXEF!Pc~IsVH6H@SL>V
z6bUQtb4%lk9Zc7l=Dhi=xkHL?A=xyCzt%)7_{#PHwd?U~+8;XPDl~vaW?}F>FVF+w
zMtF&4_-~%*&)i_B+&MvLQAv7o0iZKBzsESCC`fMBzJJ0uJ3rVq8o=8=u@npTk5q}z
z(A=QlUs5x!zZr^W>kyTUQ7_K=P$P&ZuVpNz%N3G8#`lnvfuHhCdnS;qRNC4YEYc_5
zvBXr+5*_YD1u4mVUtP8#B5h+3A7o}*s4NN3D*T10vKHAU|3UnAVyQRdkpLf^*05!p
zV|W~-#kYUU-pA1)*Zx@HxCK`V4$3dGRJ`LdJlKyXw-b6IeZxte!h7hC>bBpXDl(~9
zkHI$sHyLZ4omn}7E=@W9avM<3?)crj-1zQ=7cloaG|!io(;DxkDUS@q6rW=NK0IQzkL3{V}`)V
zqLrBlWY)mzK_?Cz2*P*#XiG9vJQ7R%OvT7h{WGzQsT)Jx*#6!lHFTl9PIBPZE`?ZE@$Gnm1Biq+%K~kOyV+G&
zmS*7j4O`JyuZ>SfW6w1|%!iyN^B218Rc+Z%JbC+}K!?xs+Ivn)ds{^P12)#>E44XH
zdho+>NuuQt8GOeEiT&20*)>g9&jh%{hIT-U?@ei2ukV
zbAU=@3Af;Z1~1F-!Gnd{R7LIx%(>(p}Dba`^%
z0FKVJYNt8s-~T_{yEL7_ye`Tvq2~OrY3l)<9c%A)nzV8@1iztHletZ#dKXb;lS>;b
z~@ePzN!x6Qp3CFR;4ExY_3
z-Ql}yDR%;Z2Ek7^6%|I+xYY3sT1;0={39p-bV#YI=^
zX=F}kofq(O_O2o_a_t>NDeo2??fTp^K=7qhIA_=<%+(&0m%VT_>KJs|v6)$qq;Agz
z9-abe6O)XukTIzzVQB9!?jhOoBL95ernj
z<6F%ndq=*Jn$L9pq?axJS|n^O5C&GnPExf%qG{sBKV;Kpq>GZc+L%4dRn(Py8iWHV
z+gy|%rkOw6r`3EmBdPFvyusHTWZMTjX^`&x{O;R%jBuvUe@O}mZOD9;i?A|leBkiL
z*Bp$xJeX;C+0SFP4sWLQvW^Z@P~s|?2ApH`6tQLh1`!@IxSe9s)*AOuy0
z2(npNJqyb!&|N5AZ1F+YNo_Vo&c%o?dv#u{+04|0nrE0|W80`9=#F92p7o3teigbqG&1lF;pPCi*stlZqEf%Kdd&5cpXy_W}vb3ynXOzO-w5d6)3V>
zplPvkb4vJ|j)+=OP!`bo`SUJrlt_bbDMQPA6?k^)kY?o@rEB(Vk=FY)qA-nIz`e3|{0_>H;T-q0C
zfWZ%@4Z}M7GGb@UhLEK+T{aTt!4ye6)7t0&us>8MLOkbNDw!$bxj+}!Mg27%vKtZc
zwSq7S&b`wpvhRg&8w@lmkK$S9MFl)fRUbif3^eX%0kPJ6*la^xy%^9iB$S-&!
zZZgf4X~tB2Viss(xDvq80X{8p>=v0pJ)D)2_bBAuemuXVa
zX*~4@2kHn4(`4uP)gw0kw>#mr`Sv9XHQR>$5+)PVsq8e$6Y%vAm~e{-FFCxx!uc1>
zhYfgJeJn!>ylI#ycAgRLdem93H?QPn%cVdEGR_}8{*P?lpt^uAMdM+(G~|1qt9WZa
z_F=$_AEY(>z|crHWmU}3Fn@x7y_q+)(JUDLP
zPb<+>vs|P_7#HRtuV6jV}D!MAsBC(Uc7oe9e}K&{P=5K
zTK~PH5mQNd+G=gbykybI(AmsEBbaht15*;~C`Ov--AhtR#cJorT=Hpz=qX&2TM|8OfyU!=)XZRu2b6pnenJ>0ms2>0h0Jln%K^bX
zH6yz}jV*z5;P|ac#=u|Nhr|tJkti2zvihC&7%rjrcthV2a{X#yHMe!G`OU*W@$iGE
zdUo^iC8(zt*m}-FS9cVNbLVs_ro9|4xr4D@na`@AnLf#HWH16wg@?)afNwy3XVDRJ
zn1#C|5nez?p*!c0Ki`ypo#o-h@o$R6CuymCp{H`uWXjnYFSne2PN$Ep?GQ~xDdg$A
zQJVBJk~gQf$({2S-sr#F?lG6@{M6S$KP3e1sPhi95eQ)cBYGq+{mW!Km=)Zw6EJr{_XFvqEd<~G%^&*2;EhQ%0x5D
zc?ShnkbXFL%%cSii5`?w7H84UcTSV@%$Qu>i6hDDyDW(-d+<%%;Lc8TI6kUQV1h3#
zz3RW)k&N0eVFauo7}#1=%Zt6Kj6Y5|FZ9llF^!v07+mWdGuihyNdhvFQmKAbP*z;u
zqi{4i{uGfeLQ-44llI!_6)nPNNBo1OU5aLb8=I<3%AzN5mW<0kCvxP723o#c@>es>
zW%vdRp|u-Frz1UQycN_26f;%+H6$-0T;++)V0N9R@zssp0KW2H8OP2E$oD!eh)uPY
z0*FFM=C2(<>AY{I(rf+Pw;kt*9*Mv)8&VZ(^eNELseH1U86xu`?kCUo|9WDFBLp@4
z5YxW9n4bknoDxmvSf_I}=f|$Bl41~8jb!8-y6lT8ftPKx%gk+QTfwqw9{EMxK$~tO
z=q*BUaFR-l&qMCSwIns$del}b3*)5w(@bl}{%My*E1cymQ_05&<#$>HjFj2)mb&<
zN-niF{{_nxFu0qmSF_ktt#cEyIs{>|I?nh3@hleCs~fB*NU%%Adm~H@k$C-(ess%o
z*-teE2E$`QB*2Iz2s^p8qvE(k=ZJPnLC!^N(v(K@`eb8;f$xj^ir#v%^YdZ&KP|YM
zrE_N?*B{c4Z=O8^4#F)-gyl}O${NJ4)kZrXt(O_|unbKkQ`BoN!J9Bg;
zWm2t7b`vE_CzGp*WESWJRd_3y9>1^AVAtfWt*Y|0c@%Kaf)NrgjXyv?w__pqTE|k5
z>p(Ec{iX#kNJCkX&7w$~?CjHbqlAL3?STA}**Y)n
zXr9s5j3$3Y{U|@y
zc|lxqzzq%wyc5=Il)vIfFuYN70_-s+vYLv&hqm0qP5U|MG6lu2_(#DpAyY*J>yO)7
z@)c4?0~^B@uoVnyidl4OE@(-nPA@87vksV3d64`a>d_Xyfod1I>1d(h7kx95OG!T@
zpwiH#7TzK3`tQ1rZktVflZPg6`ohDHoxJN0*h7oyAANbf*YnF`a=ehtox7H)?9AXB
zzPWjcKrL?z%0E!*T+RYXbx%Vu?9O1x2QFzT#-=vyIE)_fjmRVn^(M}gw#jDi5Czf;
zO$TZMY3xMqerL|c&2e}x+IHotebSDFoN?0^C|c|7Z$94oq1k>MBDr
zflr;T7~cDO!A!D{I5qxoDL9sSa`f!CTIPo;4+tI_7u)tszBq`MJRMbQ=KM*VF>tzg
zYWZi!uu+O3ok1c9C2GRdt8U|+T#2_EKrcSQVWoIRLkPT>{$vV@L}X*WS*PDM(D;Ta
zBPyVFUq;2OiT5JH>Juj+$U!sH8NMCewAVHJ+)`P(&S+7tjwCxAf>8~imyUBRbooVVj4lK@G1N6|j292dAqJ$17ZR9+i
zB(2Qi_h}WQG$CFgdsQed&r@v!}W#e9C=dUh0br+uMm!nmVtmtLR
z--@_mc-&L_Tso?zg@pZye3AaD;q5X1&QwbLJ$iWXuy7t8fwO|QlI(GJIN@F%yJV<0
z&AH=D?Tb{FsGR#xCoDU!1sb?wZP9f?BrGM>=VK35@!>8_R)+FGv1g_61}yYvr39@^
zQ@9*h-``a&FGafBmEBjgs@AvMI^pgk5p~9W(G1`FIgz?Mv3}WBh&CKQh#qkSV^y6x
z(U!S`0tbb!_rPW|cBhNJBS8k9mMR}4|45fnqy(xOc-TGP<6f6Kn~lp41w
zMI6NAC9uFhf1KcR;bdL?>f<*HM(#;L_)miuEbSDt@JDBRzG~P;IKJU|Xb7$NueN-R
zQSpiYLkcdgzE6E2&;}Yq3ctj1k_CAgH>_G3!LBnQl&wM5;N=lhL7Dm0taPkes-Taw
zF>$63`!@0Y0qT>zvOc<3xe>}nu=4r|lF^NFGE0_k7A~iAbeBA8(4DFmaNoX*nZ&$~
znc}y_FAsZQuXf5rR78Q&`e~`YO0$C5H9lsw(p|2(fR+$wwJ8j50A&x=&?jBYpXXB0
zbt0Od$UhXXTVE$1{OoS_QhaLT&Mufn`}~6X6^Clh^oa|>!dK>r@P`x4jZ;M|>Eav5
zQg<>HE)<=q+ZNuYY}+xMki_4NJor{6#xh16m({*4T?X4G?04;ZY{>EEzfptY>>6v{
zPie(R=2uUPoV6*mx7zLi9lKacmG8=Xmje&e19Q
zwSPthWI@hiFeIQ!2dS%b;CGxez3BucA$=nw)FUX7l}la1&^zdnNaZB?e0IAuJQS-~
z+Jdn~Mc2JQ4yI^)$nrU=@1vm2QiQ^N|H!Jn
zmdV)RWh#jX%vYZ=5Rt}!7D|#?Np;ur+>&Z-%`Wq?8=#w0+vO{HoH}%|L>vKG^k4Wh
z6bA(X0^pYlSKbJ5>hVY4MpDzbK!0&3{xP}A5JVJk$f+nN)+M=m174^m*u$%##+N2w
zd}rd}U$ulFdDpPHm@}(pfE@=|B8$y-`I3H>f-d=ke)In`qYwsPXx-XheYid3JRHyp
zx_kR4>nd^fy6wcc{oX(CiVOA)!hj~Mp~
zug6SNq@6AO&u;HyVqYpBb`?n2{~9d{#X1V7`8uF|F=zUZtPC9y8?l{KK-()@i3^VYjn(=02%W93sj6%{`Gc+(
zZG~RjUTRSIYKvy6T$-LpQ(G?m`CtDNPq_~GzRHB>the{-JMRybcPrX=n%gK&{1r~m
z4z8CD)PP7sz|W|hIW76VE4nrtetQibT{szq*$
z5!VP}KkPG3BAW3^&LnA9f0j$z-_Q>%O(>LiCffetAYrT79Tl1mAgm9IijA
z!o!?b%OrRpAQ=6hC-CEpS%9-phf6Iu(NtfpHZA%iR`7B3mS+gy2U>9>M33;)88NA|
zsEGF3hlx3omTyjqT0!MWMC`A0{T*9h6JLoJI!P7(NzES)v3aj%zTMt^%dci7{UjdI
z_Lj$q00g5Ze%BE(
zecZ45J3B{V884_AMJ|bqGEHIT5$jbj{fP*TAnFY5L#-tpJ#@mQpg)?SMoOJOAXR0;
zR6NI=`Q2V*RJ`jzhFQOO*mONx@c!jIGoi?4@0r`o`Rq?F-PD7Vl@_$QRBOg7AkYTH=C`m*_sGXp%qh7R
z+o~4sxj345z^gdOBKWm0jOqt6e)D+@YbGgLQ%|c6PTo@NE%Wdg&WtC!AmMQ3vH;nrTjBY9(%j%BHAD;whq7%_B}9G>U)9Y>-YZQ*ZVG(e^)F!
zGI~!qZ->B8=>xUz-x$DR6Qy1g-S@Tf-)((#=Ta0CW=^Ko#VemS59&;$(qP9|z(div
ztGK6zolS&<+zMlYG{E{Dj7}q@#Vbx;to?zDk+hajtqk_qY!!ada_`USkA<<8IZdjs
zzRdbRv9r_>VFft;g5F$dyFJ89O>2%ea_$x0OMF%9j5%V@_J5;-Bk1r5Gks$D;+{l0
zmi4Ph^Iu=gj|obvnF0$#&Z~K$f&zr?FJqNRZ$P9KEp$e2+nP9+!+LJ-1Lbn>*Wq5m
z#$V#KmX-|QK5k$HmMmThGlNGA$UXZuvMQ>51(?qn;#lzZI$64Qe##TvWS4GP7^8<$MHWN9jQze1imHiBQX&UJ|LepPij2$t}4Jfh8WV|yxOsm^V5
z(JMfteUDrExKA#Rr|$(vP2rN@%^8tH7e)l_{4bjZj*bZK?p(S1>8yHXa(j6$chSK!
zfa5d3nPsDxVV=K-+NY5*jv%j1G|XIhN7Va*H=mUjUHMHXD~|~nN3@|dg#@hi%-ja3
zVG#T35pGr-jYqV$M>5?G(5IkDIlp!dO*
zq+~=*zf%~xKLvrA0G+T8%*ETcnb#%*Uy2bVrvFL8n+d1p7Lf$32xN}t^$_#h9c$j#
zY9@dy>Cdawf@tL{CA^nScIjRG5NOANFk`>VrH6(-*g$0A+J87z4r$T__UvOV
zzD^@N_{<|9sMS_uW3THp;AmyUPs`{7lNiU}9~E@O_gDNpghyDX<;#W@x@eDV$`^a5
zI({f6O;fPon4oFI?_YEEaS<*roDdBSd?B}MZi+?UxH5(uV`U1Adb;aYY+Yv=HS5_D
zx2vIKHVWMB6@I@fkYSIZAg<202Wa(ZF{A#%Exfj0L*(0DJ{NyuUq06aw5H4xLx*Rp
zrMdUnX1mYuT5dglQ=`sck#QvhV7LN9c-hO`3hH!3!2mvTIZ}?%0T%>UzLAB?E
zFroH(^`pHil*F$TNRA)Uf<7~JZ)gznW2$5o%+K@Dex9567(^d^86QD@Y2RxsiaG>q
z5a8!MTD5p^lfQE)@OmkkN|qwN%{h+puh>V%4R>glu0g-_I;(Z@x7?K4R%v;Et2beq
zAhP0``U&=&tYA;57eK#ux1NPY6c4)rHu<45_RJv`uy!0UgW3Fdj742&ba?bvG#=jY
zfOiPsGy72pDW!xYb5$I&e9{FB*N>c+-wFpHXx6!17?NaILlo~8!@YA;L@hf;s=&Ct+QNd2}qEJDv
zfPJQl{rolP;kxL(mU119Rx3e({q&>sZWCVF+NaVFJmN2ckr7k8k{c%;hphk=LSMrn
z|K78E(4HF|B$8zyq<817)@NKdl&_99TUUK8DhzJOG1elBdRz?w{`aq7xiE)Xyowlz
zZ1Bk|tGcR@LyGLvg+R-J?98tnV;trd
zzT6=b@)1s@FjxPYy}pn@#*OUHN-pq65%8=haqAn7*Y*NNV%3vKTQ`i^
zyUb??{`?U_3?NL{?+h`_#px^*Aq2mF6)N^1|BSv6RoeR)#FcriJETmdljv2=Tw!$?
zsRi5E=l3%Xb$N_{bT
zQx2R|aG@w5Nd%jBt|soFI{s;kkABL}s3<
z(b6xZb9&`|eQZ{^E5Th9hqsJ%0RpqBwm|H)Ri*~?sD=DP-cmZdmJPC+
zvR)$%4U0>0&4p8<+WyR*Ynq%)l1ZEqLy*}CBJ3=Bk6}JHhim6n6~obzK-}Q2BT)>G
z)S`_c+~WNQ#t{)AJ^Wc*q;qF}%$d9frkpaxDn_K4!*@Wy0$TwH(CBz%oh;>zaeW~-
zHlQ`hoB+BLHCP@6JKWf(S)*AfIv91v;-WGgK13xselV-Bc;X~oT0ML0c(zn_a`K5V9bZ#iPJe+cW
zG4N9zL1zE5kZ|uy-tq-G4+clABB#v*IZO{%Ac;d^Rsy_?Lowo69PgJ3Z{kqr34
zcxQ$KT#%Hx=+z(78=za=Qme`T0FbVZ&%oOSijft$ur-^?1wWg@%@#+(G7J9W2xqYZ
zK@t}OA_66d3`!)UTy4q(aKi4qvkxhDEXmZII4w+V*C_bdM~KuG|pl#e0uXm`=^`h_G8;0(^8Ibx^G@WE4F=-#zHU7v>1
zu2JM~Ex;I2SoV{xb+x%)L@KdUD)>eF-|%O$AeVTl#`q}2s)`v4BzgaP{P`EvG~%0*
zv{ECIrQvRH{w^@J<#GTGo)QzX(V>aYOY1
zd9eF1x!O}s#o~veFB|u)kQ*^U=>_njD@g(5dzjQF;=3309)o+ZCYcvd0e)y%4L4c4
zVw1Y%*Pq+YCxc02-e1BcC$uLc-KoaZxYqrw1^e4F<;tsv%ie=)%_lqD{~@BNs%Jk{
zJ4EgYYC@bc_Of`aG68<`xxNl%M{=9^SmZ~BWX2yuILBI~;p{%V{It!6pSS-m`ID;I
z%=e$Aq5{q)N@;mKms`(94(Tj#mL#Y$yIfTjD3&uQa8>X}9p=xDX=vI!r1ZG(f5>tb
z-iL&=TFpN3RAZ;F&AjsmmCy5$GC%+|t#?yDrqWruj-{l!1bdwM@xmQ*`esm>59~*f
zgmt?Mn$}=_Xv7q+wvpwh(t>pFjO^)?)!1+Yi>)YXNYL>^{WfbH4mMsrUhGdKxXTtd+zGLzG
zd{1tedJ6tCYw?C$cFK1g9xpzChFgFL;dOJ&>8TC82ygY`cL4xvkD-odvv+g^NDYA>
zDxe}XN})QuKw$1T=z$t-3b$Gu8mTt!iR@+SY3DlDhjVe(76gxUM>^(*pIWYEuRfVhh=Dq6O;u0
zvACy&GVu`1FLf)J9!<62#a}IE#Z3wnZw$hx{i~xzNrJullLq%Qv2y$N&UMMW{Z6=B
zi$%TX4uEqEUIdqHp%EIQP2>O^u<{fE#~g$M{?j({5LzaT)m8OREgalvxW&pDrwtO_
zXU^=z-8I^auF}T)j*VyfLid{XjjfL_da9}Lg3>$KG=j*S)a80Q)##VZ>wYOAGBa<0>JlSIV-i=0$GT$T@C-@gbLF>X){|
z%@{SS-E(W<&eqs-^nb+OowNTi1=*EXd=zK`zWUvNzdjFq5?B>?OMmn2XjbO1;7SVM
z6!v2{o_=}Zgqa?)w5!4VWJJveiO|cGg8L<4H*AV}f$Tk0ripWFqXGE-R1xETLaf8^
zps*QKDrN-A0<<3eg$){8Tp(KkP)3@+O>2s5@WK5GQPdMr6ryBdtz_Y~jNi||L+4&U
zEx9$yfgDD%&w#&5
zsDlpoc4%vHTrLTc9aluQEDhQ^;BwWXhXj
zXNCA$X8TI``u_A2^6D5$@FGjgy+g?W)-8gBB1J{A7UQj$fs(s-MCZr`N_kcf9qtUj)XTx
zU--yRpq@V_T&}CY-LS|
z@TZDZ&Z}Qs$z!>*#m;la!`yCHHza_6P%qVu)f~Fh$WXHllnH^)(REsj1BH(GYPl#v
zuQio(fjWaStKk%n9Rkb9P?k?QwKOj7+#qgkxp>d%|3Ns{A^gx81$YF>4BZWDlm7rE
zSj1H)EOQeeM!tQ~qoYDD?E~tofiP4#bH@qEce9`x!Foo3e4UwqV1H2%B~qg=KE|(n
z@o9P07;A=S|WHnb~_OpIjV{H8Qp^~Srj~D*n?uPAq11vRPHT;od&=ESqdYFEv
zxoFlnT};II*_pTGldRw0<%Sn+7U$yjL4CZ
zw=W6Ip$p*2c6@TNczpm{v=bfrSHJ(c9LDf0he|~^mja{J|Ea+y8Kz#AZ10*cxQF8D
zm3z`O+2=EE_1;QAg;^iL_Pe+g=3O4
zN@tNHvN<860VBGaCeCBE?)Yopv3U8klLnTzDK`+;7%y
zh>D6eoUh!J&`Cb=s|8E*yWd}9v7Cab!2~_COU2OUVdlSX
zBNopE7*8+@y0D=DS+!HhV7*m80=G*yb6<^8VF;nyQlj6-9juCMmnkenBA8kGV
zt#Br0YSrNG^N?k-udjEUROyDE4v%pB#qY!|_-WixQ_D;)x*jHm_3p$3aQsh=@Mbb`
zEs>#~U$r&>kT~YONYr;-aB%;H2;=9r4@4FQA}i$F>84t``Gto&;XyF^vMK71ex#QV
z!^azdg+PRsL#U9U^Zjc3-aBSW+gpU|iyTU^GV}LjxYd`;8k=HPt{_xJU>*dKp)RXv
z$=`ziBx$u2yqTVt!X>)d-sYBAlg8?ek+ei)4WlEQT{;k9rXNAziM+{z_&3{3)rZC;
z>u3VHkh5_n32X55Ooi3)i27Sb122Klrp*>*+_~&=m2ji_!0ngWq;pg`$s}8ZjeIdp
z`=Vw-a<=3s-5>9`;O|Mz%KPj&gXEvrtpqbC00_#nYo(oW1n8#BJPABD+#U6)piIHP
z13tWCCA*tE<-UBo@5F4+Snai=;sa`)AJp$)we{W`aZ<%e(!`4jb>ib>V1wrjI90li
zr1(_b6cf{&A_xAnu%P_NhK5?KOi)2ep`DrLB6?#X@hc+}3CRv3W{E?BgD($p>uB`t=v8M)JEGAW!f3h
z;~*q5NDHZ
zs?BOo`1$|T!l|7(lzkXNYe;<0pI!m$*8=q)6awxb;R-NRG8zbnog(EL_q!9cbU6To
zD{dhF9OTsj7@06Ie^9h`ojP};>aGQ{h^)E9iN_3!()Ar>*0Xh_)(`@NItp1@3|V
z-<8SIT27?4fL1rt(|s%K-R;h40m~;WxFNr`BX4-!gxyV2R^-uZc>98#>#Oio5s3|k
z#9hkRXol{?pO&b;fd`MV0Uy?K7T(R}69(KklQ2tYB%uz3O(0!({~%(Ly2b;*A%Y4h
zvdy%hUGP>_#8%Y+znppckZAhqYmyj0vgZjWrtS1;?AxyqBER;
z{|n%51C)%;txUj<(AJC5C!%rW!|}Fnp^vAlB6o~|6huxPa;ndRTP;Q$YPrafjSwv{r}Q;2;OvhU_J|RpE^d0ed+EA|j6seHwHkKC
zV}WM;vNJH7ftJ2)U~!a%Su_E(O=mJ-C~1VGsd9DCI!?LP0SseLK4q=V95}?RJhZgg
zBAgdKmE#x9|52t}@0K!*Y-D58s=1n@y}RS2H5|
zq^bHVMKFhXGvu$c8+-iUo@>(+{*;m!f?aByv_6qkQc{OWgwU(Kq;7@$d*g3{^9EoP
z4Sb1Knf@J0#DIz-BdgT$qZvN7M!%XfO3(HNMgN(&2|LPU2HR1kGuHoQR8AXzU=T@`
zhJ_iBbMe=${A6w~EJ_oP>%HvAL$<-Cl|+rVdlV!D1+!6Le#0aE<$?$V*vQbaXuiB0S#p63$fs=gb)U
ze5KxqjJ=AfM{@Gfl@kXTE+ZMqn1|&
zca4!oD{OqrgCkBVq@E=Bf>vEQeh=bY7d-|kKd37kG7h|SFS=2msh2F%;CC#K|KP%#
z&_b||m@1v$_fzy#O9!yY_;aR_6s28(hm+>94ulPEWycd%d6L%~_~Sq@0QdlyoSfWx
zmofV`$
zmkBjU+1`#xttnF0X+kukLye4UV#W6RfnEL#DZoDx)3I&aMlMl8Z>}N<|LxXH#ft_w
zF$rVX|8vR+llb>ag^lFK_}F}MpXXbNE481lT&jD%AJK-nCF=n=M9u9}euZ<1PxQi|
zf2QIIa41=IAbSupw3d5Nce>Nbpj@u$t0!|5C9zw{aUF6fi~Pvbr4YN7dIo)Aof)TQ2g
z&;KS^fQA#wb@_M535Y04`jK5}T7nTn@djy)DDl~np5mLLFRAB0t=*U;_NRogEFZ&B
z=4j&CNHM6~#tSUilIz@8lItYXgk7X_g}`P~#|zt$flsYAVr}tgZCY9u+Io$m5O>b*
z54MbBy}Qi_*dVn(Mh=n}oB9=R+CPYHYqm$`{B!-Q;SF_-Fp^-|g9wbi%-(w}U6*Xv
zh;5ene03N6GQ=AMS~l8PpcSq(jXdCoyekiRdZCxWw_71$NGEeXYq4dIl^f_kcg&_!
z^Mvy|E&vR{HhB+25eo*|DEd#hh{`5f-pwR@D)8wH08G&4;a6N07?Um!rvCJ&I!FD-
z_`&F?dg_R=Z48nt%$6M<%95w0jvop#zW_eq(+TJDU6)(Ng|4HuFv8b=t7=3b=+Md#
z$QtJd80qGXRWhloQ!B3Fbbor7K9aRrDE!tF!&pt@|IzRzJv{Q$kxxJ2_a0V+@)Sml
z)R3aFk653p}9bgCLf1MU)R{JKzQhNxT
z<|OEg#w;DjbFtDUu|t33q)p)a`h)1{O)#+9<+x5{Y2%YN$txgJ4?>o&HK)8m62T0Y
z2ob20j55*3k){ONb=2sz-TIORm~^A2@(F((>2?OECkPH_
z!ptMxnNV7odq|HgSN~pa`F-eoxR>Od2@J!m1`gh{Ks?kb4CME^1oW=>zFGIGvN6=m
z#iEi6f;bK2(3>?P7qSm|ooLvh`rJFgBJ9cybj$vIx0LZivaL5tgvt_IC2P9w=Rw*j)ZPe!@f1^541^z+7XVtKzw?
zQu}{>j^F;oKS28BU6M(zI%BT_s&`ytK%r`nSQsk2xvv*Qprj9>>@KBvk7bBKT-xYQEQt<8@NqH
zB*Q@g@}K_;l4}$*q+Ytp^?zVr{z5Olqwk}nKqBbNOC4oaEdMs{E3w^q&Yc5=x!suB
zAgqStF@oFnBkN=mSius_dCO??P#XD6mX|9-Gxeusdzf$oO_Al6aTYrpv2ftAfHN#?
zdqG~^-1JX_QqD`Bu3H{rA*t^VXB|SQb^ZE@2SsoX^D)I%xP8rsF=Iz6=-%eWasAY>
zi{Y^}vjlgZamutr;SAv6EP@5^zVW<{)HBHVYRz+r1M^cTbLT%7@=GN^o%Uh2P$=lW
z^WSUO{qs%X*C@T$IX$sv?cf57hlhc$9rKqwe(eJOe!;%~dAeCHf?`MT0blHdvY*j^
zvO;Xt&H`0--D#-P!wP1?Kz1%wf@5>zVqGhQ-f5P5VA#4k_EVX&8$mmI0*d5uWX}c)fh+^uJcsQ=+bS`Z=0+m(skL<#vLD
zEa=-y8*6TV!46O(21NkXr&lCUl
z$z=_V9YQv3!IvLFdyWAd;e$uzm1IRTYz7HXn=l@*=zkp%vac8uOK1rgtZBWI!@xyn
z;6ZyT6Ye^h=1Z&|yRjR(?Q=0mMxB-DfBJ
zI2dI2WbvoLhGstdIgq@_D`+pRdY<^UUA0e~-n*n$Yz%~TtAbeyd5j9+f7Nq*<=mQy
zmhVlyYBaw?>8({mR=!-nbiXiQ8lh^|MgO=^G}q~b6b;97c055Cch+U2=0I;G;OtSB?7B{fS%`CC7@gVn2)Twz${x#$9#+-E06f
z+{P!VmDG-tNB+4_nEx~Ph&G|9p$a|JA~o2WI|e8o=zDITb&|ktF!0tkRF@2u26n-|
zt3tL2i+Z;zdi7UAng3q)s~a1UR%l$dL%zNkygUI0y#UrhD<>xh=-0^8Cn8K
z?VV6%C`o1|=DnWpdBDPe-4DakDX~VfVUGV<;=Mumgmv{jb^^S>K$to-wqwoRXD?~Q
z*4z!<$Rwq4NWAwFlIZ+F`nFZ|d3?lWCL9^pWNqf~dl~7G_KBebYEU-;S`Xf=d#y**
zw=D)WFHfjCJ_F2iYsYb`Jm5L*`)~dHRo82W$7g5wGob%GXl%Sj1s%(oGB0y+%vU5S
zAm$-ogcrkte;76MA>6qnN(KgBRZIUPwQOy1M@PE?ye5ET#>~Ppb?VASc4v@EL!XFZ
zSx-_jM~{JFnKH2tc&ZgG9Gi1P0sHbPOC9QTxm?BKEN$iS;FT@EI>T?+0!}Fvjv=7w9Xj~
zHbCj~0a)pBwX(_`dt3IQKAkbTl_6X5;SI7gfRwbbP^JVkp{g5Z2vASj0E^8Ely%#v
zOrNbxA9ft2_b*BuZp>7vtsn4@m(KT>&WBz2Gz|FjW>1-ias_oF0T-}~OG_f$G8*^R
z+&z7L2!L5NE77FsD?lhhLP8o5+6I?}e!))n?&0x*^Vu`=p{Wjjg-m!Me!b@S8t}vW
zND1VZ`P`iO-$#7_hLZiR0}t29oC0Hu`%YA$Id^5>83c6_h&SzI_Q@gRNV&qH(FbI0
zckc8FLNS}v)rg`EN7jP?Gv$U;)~7F@kER9UX?lWuLVq{dsD)0>@;8)xC}i_>gvlGy
zVF;}XHi4dQ@Ge~O5RNHQE`j|lU9sOCkWKTeE@Qt2MP@?b30
zDt!M%yC-^~t_SdEs7xB&F=^cVI9H|;Eb>~Mgq4I)sUk7AuGst@bNK?uMp@w_zgOCn
zUHdtU_w~FhjepK7I|~*iWrNHj@^tLfE7=B|`LXFQXa(gKBuFYq3kI5r65z8%cVD>j
z6fM{|g!5ZNdMH-xWh6Le@*LX!P6$KwHvlttx5?86Hlmx}$GjO0JQU-vCCB^#Ity62
zFQ2wxr8-!8S(S!|VBX`nU_;*xB=^a@JUkGEgoHdyh6n}L)^Y&48-Rc@KqN%>y#(l*
z$+vaq&6AwyyI0)9yEbSc-gTNT&ir6)z094bC$b8t>1KP7rVZxYQ3Y1k?K62CI6hzh
z*OaaYaFPJDO8_4A=?s}FVmY%SqG5-m$;Tu~E8o+|^azz2#SrbT2NoQnwas)7b0!Y6
zVr5i6N3%jA`-*-D+rT+3ZJX9Uxly#W2U+^okTY7dV&_Sd~PVp(SC~zXvWKv
zdP?}$5sEGidt(m%sX*?sv|ph(c-g{@bvl7j+Rzt=6Z4fM>KiHR?i?+ihoU!sg?Y?;
z=|ZK($U+JIN&dEO5+wn^R7bZenXPQDVua8D^C$y$!9M*11B#f=xol#MD>Bv-Im_ie
zwyvYUlTmp-IpC#h0hS
zPeagLqA#La&wu9l2ogzJrp<1O0ZpNL61pYzY4)0J){I41l>c!7P}TrvhC!-oBxJES
zqOY9&1dwjmdkE6&`k%qO(fenjj$N$|g#Y{wCat+It7E*;I`6DH5eFPDNINe{9}@K*
z@e`zhUtF&cwqcN#*7nP_HNdes*s&RLr68Mp%@&G0DG3i78wV3MDb0$eWIWTZ3k6?>
zpVgp#Q~x%rt#@NK9OpgT3TD!u?v`h!tvlZ5ViqnL%zwr?KmG{X-z2v&*ow%6xBvY8
z>-fGCrO*zs9A|4>xMMN>KIAtE&S1MFVnbQ7nyK66if2xddlK;&Z!XvA%AI*5*)
zQD+`!{_d;ymVy0Fh8ho}Ur|1ZtZoT~gH~IefT>a*N=eA)*t%L2(nPaJ-XRrP`<(-3
zG!GPWr?=8w*yXSq8#iT4YCLRUZ?M^y{3#5hR|SwEbH(M#`8)dI){yM=AZ
z_pEiLvttyR@dh1WBHfw&w4qGTDXw(*O%x6pFK7vUp>#@!!qz
z{{9bpA0KAdu4ZlQL}_58j5yeUb;nyV<~oVF;^yP|We(^ZII-~dyx6FE0{yoG3nEaH
zSm4{d;XWv6&`<}NJGy#Qn#!)4K0BNa+mZ?NQm%zXw^@aFtW$bwQ%8WBY<+-XyaB1u9#2%O*4(>4zCZ8DH
zX`i>e`)5c{der|N*1`QHAO6rpaG~S_>u5M5h1Fr0V;CMQ4I9%ms=igJ-_~TzFAf(G
z$y+x?;?JWW__jS2@HcnZ9*K%76J;Z&b22sC$KAuE$=9TuH{8q!GvUPBGebmS?YP32?mlgaJJE1waVb!
zbsN}=wx)weDmEf4m+?+T_U9y9>9#vTC>Rff$s+z&EW_{v1-HspMAoPxO|=+Q990_S
zcK~lCmm|N9PD1=k#-ml1XFhSRRJ^Qew_y9!ikC$&g@>tFO8#_L#L=?B)B;$9ouskY
zAcXFs2*j+LyVp5Fad*3sS@bPa*rQpAc*FRzOe0Ao$@|Tb+Du~$hCET4AUWAa>`~Cf
z)_ZFVYN3fD`u8IdT2!Ah6YRl<8+Bc|oJ03u_f>OFteN^*#^oAo#7Pad>kPXFy7(j`
z4wxGD@m+H#+ZBwFBZ&iNO`wzhMlEVt0=
z%GrOS+5T_7i>y`ybM1^MUy-<_!bpH3Y>jCHL5V>A&%CP2iGTCEs4gwJgKY@UST$E4
zf?Vsx1;Sa2g>kJ1uqYXKlzH`pXHQmGW%tw=i@*4p;o|l|=+HL*Mx^;4E0V|13P8eQ
zRd&8Jrf`ydJQ6pdgpTP8C36=#M9eh*R=JcNkr}Nzc%2Sn1-Du3yE(zQ%4&@(IGgCPD4dzef=dE2KXN9Uu$a=vgDsahowO94
z73kBtzL2xK#}Yq;(hW;g2PNGX4@1de9b96&sjv!%B)D_9PV+i6nl~rrVL?7#Ts!fw
z_3k!5=l0W-CzAXq>~_MT$(!STiR~?+DNHZtVi4b&C+~L6diu#79*xSRf4Vozzq!?4
zuA?VRGZ}7=G!E{~7rDW!zqf0u8vMH;$u2ab#-EFxGQ{DB}LT!E^0VYc)GFHHQYMap3ZEIr4YU!mQ+!X}@DLEOO
zzf03BB-PESB%ftz^M!6lk#4@qYS|HlJ<;zM0oUX#FEh+-^7Y!r!p`SSD$W?}Nmy=t
zi7v7D3VtOfh%YY~%D17liBuNRP(%)RYT^43Vzn@2qq9{3N8h}-xXv7|h#EHBU11_+
zqtkHu&TlWk{$WR@JjBu&cG4}4L%9tcp|lVHGgeCxg_wOG(dgoU&ns+`QPLIZi;33r
zomox4W78v!(&?8*=Y~o&)F8nPKY9m6vrV`2d*cehJp-O}k#fW3FHC`TW5E9e=ms5{
zm?w|N^ivDujejqS(({b$(hu*_Up?Z0ssOX9PuHo0#F_Fs;WVdmevfGQFRyNL_4zH`Z%^oKY4
zZXT=F^J1M}#hZc{7?mR;R=fSH^ET{KmqLP!N+emvRloD#!4E1(rv+RaBJfSL#q%Q@
zZ`867k@dWrJiSZFpn7`cFtUzy#=l00=#>$nZvt?EIx;WnQdnS^Mbfz9!4$%w8*I>h
zFL?7@*L-%%pUutc$QFMVC!-DmnO#YJ7g5%s1gV!2B#w$xTg?9++XFyZ1^Pr{$sd#E
z+PPCrD8HPU)r9kFq8yu1Mrx+c?=LR4&Vb{G*@l)LDQq
z(S-@~(n1Tu%FF2+8UO@C22joo5Xn5vAZ6oEt2t-NQ~ayULx%AkHN_L#DK!ct!1BaV
zoU}#=s_>b6$mzDSr4N$3c@M)2e)t_T<0AN-yd;Mc_R@m36(YB@{T@UqLedE-AS3DK
z=xQ=d!Nkx=yObK9lV=x@3xmeGEZQ?FN`!Dseau#EnmbO%8zK#_@?(ZbvG#iB{LyRV
zFADgQRlM0I+g4!PpezEhknXhan94by#j~97Rn%jvHxAOEkg}OzgGAydnd42nT{tPO
zkn1c26&!8?-WTHmi>4{PCTU~Ce`?G*1jz1yDxVmRW(h}!r}s8>1Yci++iOF#a7MLE
z+#_8}LP!rw7bD38I8U;d%5tY-oq2>mfKT%i954~RwRO^Wj*iHBwm|P+4`3E3OLy{)v@mv))?H`?50FWQi6}
zdT`%@J;oxn%&eB7*WA+dgqN8eB2c4obj=A?3Z0UiT8^1=t;t9i;b&T#0&dQSjHwc8
zcO2XYVSJeq!U`9E@Z?GEFU9Vdo7cDa*T_B{#s5xL%Gg4aLBNEx4@ELbN3fuBImwn-
zN%&Tr=aY|n=DHl)3zDiU&y4RY9k@Os+B$q$$dUC?0rPu}myB7axk#3%v7jot4)pm6
zI7OT8zz_e{*aCsu=N%@K%L6fTdmAUuC*UWaJ0|cpc7iqmGAP-ANmm-mBer{nGm}h-
zmXl_hoqKc6PXP)x0}YnHi6GFSvv@+SJ3%?nsa3eG^W_}#U&B=bY0wlS6z|asd)Ea<`9SMwOy$?DMLHcV=Me!e!uE1SyY1{wq}|xSZqCi;
zT?{Mq`nQtJBA}`)h@n{3JAUE)`rj1ka?Sbj_{bm1!QYvFj+zeLT7s%I>z(Q2i65-F
zgN`kmI&6_+qmx{C>&JK!Dxiwo2~K+#pT$fR?A_8576(f_0r^Bm&W`~iEY7cs4b;6R
zON`*nD*?=QU8Hn5)HPd%8v$buX#jbc0llAY=ls?%kXn!}zX-&*NU?H|%?c3)a$aR6
zphT-R!};eaafFKqQKQMWOzMC5yF=UC(>Z_z8T|Z7Sij6XeTeq-N>NcU?%?fBS!AWN
z0HhW%ouLK`9@*J>m%$}qRJE2ran%5b2j|v0Qnhod)B=04k$qIzSsZ;{GFtySR&qHd
z|C(<7pvKHm(;0bFZ2VQpVfpE;Hi|J3ZBymIhwR=c!jQp
zQ;Ku;x}Q(5QyJ$<^3--4Aj~ve1aS^lt5KyU@pMWcit}oCr
z1KtxLVE{m;N$Bvf>9uU6s;H1&7f|Dusz}r>2?Iq`*T+6{$NRC`bKB%dmDzNDbi5%B
ztO?W!d)u<^6^mB=dOmN`Lj$`~@^^ld4$>3b>?gceyMLlF)f-uSt0p5~z-@Qa!=pAot5jI<1yQ8Byb%6YIIo`;SoN7QX@pTz_`M@R4696~A}4E>~3dTIzneb6`L
zdQBhEu*(0`_y#WXSzftQ$1q&o?yGBSk!Cm*D)qLmuE=hm^4X1sTpqqzpmkwc8|U;wrG~y4td1XDY9cEf`A>OaH9eK
z(iSQqlJ*=uvEM;;5f`y%_=%;`Mg}TraVvTAS#NM|g*_e*(kE5#zzaof^peVVQqTnqHgNr#3b5{&{?)djo7k<4L8x{&znfy`cNdDfi%-J}=T&>j|=^LlgNCOWzAElvT*gg#p;5ROxCSg?;`Ew39Kyunby)wwA
zO*v9wd!L0Raz$isrQ-vP&uJfVB{*Upw#APe>Ew1$KYbuCa3>G~Mox&(aDOF>dr0@-
zFXd2wW{!JS;S_};v+B*aj){(e=8zTGMbQG3elPQ|l*S5X^AqlW%7)BS<2KXy1IgB^kd_#>V@%@|2uTP;7Nv
zz43{hLM2+_XO>!N-`shk=PMKJ(}O)`eKf@xc}y*W1GmB2JEcw4P>kdw?_3pz9KHE4+v-TGW<{OLVsLH(dYsB)rMr=)ZNr
z{rMAVacwQ6;ur`H;M#%i$VFqql?*5|qUG%oMLxKWMD?o-(q^mYH2}>@xON2~s2Hgl
z!F;7!iVeHq%uju^1gIjBp!ZTupY~fVw>9)ztu)xwM&=A8>o
z%dlxF=;ZcSTZ=Ea#Nj)%{Qd`R{V`@b6S#bL^~9~ADz4SOozR4G;wT(cX&hk?mna<{
z&$0Z1DE#m2UB^)9%({4}OK1&tiNBU@uri4k!3r-EN`*}5d`a^q@`8Xpf0{ZJxnM>C
zP5tq5))uj5$pzwS^l_2OHV|iQlL>GenoC=(vjp@74_x}hd5TTPc={i`V1f;R>n(N;
zJj32uE1vivwp%v@AUOhaE8+nM7veko+xAIDp!NW-jzL+4qK%fNvWRmN`b1NyZkUFK
z1^}KB%TC&!iR8mtT%uZDU
z+6_w=Bp|f_Tii(;6?zNjAT0nL<9KSLLf0hP&~@0%)rv`v*AFXuUR>V6I}f=j~r
zgv-LKSfx$FB4{0iZe`{>c4(-l7**46>WkDo6BSo$&;UN~BPIU{oMgeD-ASnN;f=r|_6LJ`yQg&3X0a?|Wd=3yijX3^LKEI>nRD7ua(W1#uSE
z6(>-S@F~FI@dq6_VyF7pY6T9iKp}NB0gE$;l}T`&}r-HZoZ@)
z%^$6v*gX~4eb26Hzqwxi4~uX~kftMhuNS6sa(kPvj;GDj=i2&QH0O#(m=h5zt)1UQ
zpGd+uV!DmHY!RZeUGOa|8No#y!=f%ijgU7rT}%FGnbSDX4Ed!1MH_^?t-LTEzCH~U
z^Su^cY}$D^ZfglecCEH3&(A8({J8lIA&-kC|7o9
z=y^2h$h@I(kgiVBXlF)m{1)i)n6$pT5GG#de+M{>6Gg*m>vD@;uf;2Gry=g&T2{<$
zk}6`1D%jIx)40)(R3CaU2loJv6$9>2)tWkOY)_unI$NDF4B6+@l^Q$CC7r`cg*%uh
z!nHNb^A%-bVGAeK<_pzIg{oDGY(MYU#*YaBJ-9}$}gu7M7BeXj9vSm4>MBZ#;%=_Dm-V25WZJz6;
zz$SNIr*i)DPCRcN&at!5pvO&kiJp-QHU$@e|Y-Juq?YS
zSX4?%q&o~+K)OLvknUExJ47T@T1uq5q`Mmgq+1%KL0U?r;VjZuczJyq9W_cyOGob#cgd}ceK_qh@ojxDO>1?3S_)sf^{-Xll#_y9Rt!VnEHjKG;JzNhIUc9^0Xuzp|&+Xz%=OUCiz)T
z02^dx_MoBR$FdxaQBn%U%HO|4$LDGr9&i%37-ZtSe*B*oH2Nn5+TUX3zkhk<6~d8n
z!4(5EdQ;QAon3MWbq5Ts`%Cp0Dv&mKRNJG)hCF0!N;CKg1McP@Ax!#iW8DZE%APu-
zz!U|Q$kTo1pgG~XtSY20Cl_;?3miinx$3b-PPK9!84jh2we6{~lhx7~Dy_z%@%B?6
zB*@3UVEh)*ectaUtz2l#jy4g%vGJ1Ng|t~^qE?6_F8wj7;v#Y-g5ph4sozJ(Qu*#ad_Gqt72MqEWc+@
z_*a_*mbEd9E<<#Lycw9?hRw&OsnS?Q%-FP488}rLIJ)8SqA5+Ut@$~6T0l{FyWF?1
zdL6yCvGM!#^z*?X1u?EHpoT6kq^1^xeGQ1*_TQMX>pT4qWZ(Qvy%u)6zT|a%OJ6WQ
z%l{qhGn<>w2M6UvZS<8}>jJcGP*S)7e-Iy?H(^w{y+pGn)zLA*KptCuA<1Zfppfs2Xp}{>HJYxe2u!vpvx5Yah~d
zHM!{zoPk>7qW8QVmOsSha9*~hwFRX8xGGUwiVnzi)XnO*Lpyyvs7OAWK~TsLnZ|};
z8JaLVa2Oh1Pp|AvyOKDq{WRnWjo;0;`%WHWwBC-LG@HkD(CC0|!tJx2Jym$x++CfA
zn?ZST$-@&6Pb2(8(S>h!>R6nHPRG!waNB=M9G7$g&E^YRd?Bmo^i2Nu6G7eAuYrQA
z6_1fOL&Vj{8yns`v48h2?C8jz_OlKDC#;3(X`nkhVS(i3Q!>Xb7x|l4_!oKn2(Paj
zHU5X*PCJHHU?4pD#__ig-!z#F?83h{X#RrAx<}k1C*6-Gm4#K?;{gQSH0dks
zrIrqUdZ+I1<8`eN1EP%VA`1$VP%a(&l{Ttl&0@wJj}bX6j5{!XGyAW@s|AZcU#r6m
z$*4Yr|6EbB^oVfRKJ&*Ms6RWqaHHO})vC@_$x*hjwu^n9Y+1?_ESE+qZtL)7
z!bLCYW1z$%J>U}_s`8BIHWT2B(Rmy?xP?qlQT*e`wI203Xja4z)G(kSxcqeuVEIbGP+2P_pz}0
zQr1y})4!tl?_d1-t*i^9b0?!#nwLLpY0~@`5BN|~++v^VC5aHuJW;NCeoZnn!Q7a5
z^dbZ1V3Fm!>;WX7EeO;8l`pZE#hkmhp>W&yXwnd~F6YbjXj9YirIfnjazZw=v;_~C
zmrRD{wQo9gLx&K$j6h_M0UvpAScq#w)uQ%n+l0DSdi(4rtehqCLp~=!O&S^uYT(R-
zI{b4)ylyQTLY;0(mHlo8yru;bo)czfwCUfdWBbtg0Q)XoUgEvJsG^iCiM3X&xKI2a
z%K?th8{C|n97tS&DuqYD|l
z3KY;7%;1xR7AO`Ca6GfNi6#!z_%cKoztp>|*J3l00y(E&@;Kmtb$>8fpZ<&ry8912*^D{+^{I$Ojm?Q;{
zuZ-|HOjZBg{rsd=i)S{b>ZHRGO7Wu@j!krm?6z@zH{NceGqVf}?QasV*yuRWMv>)$
zA}#H=gces=Be@;JR;Z$T%lWGCgz~=^y1o$S&i<)L$D*07$}pf!GiyzxK$a9a#Wa+d
zK5ZR%7`;`1pja)3pw~X3P8``Ijug-j9O_Z(Je0c845Z&n}fP)}E{gP$k9}>1+t|`Z=T#CVB;nHQ}
z;MT8lOzVf9a#ep+nfjK|n@pzBatnmUR7MCo@0iT>g?wj{38z(Au_T{jT^5`XmryB)
z*dnCZ1+zne5<}LAWzP@|gF3p|8j105a&A|p-Ir=;x>-LWRTop%jpkm^>OLIQsTdP_
zCm%4n-XT>SCec7(rf~|RmmYOGzwC~aF)3|m@EaYq%1V4O6G8*#A=!A`kO;oq%vn%Q
z)n`jj-vZ@{mYuIiht^7qxsyg=b)2wo+m=uE|3s#Yc65&SmM-u&F$d)
z7q!Fi{wVb7?M2g?ax5alV28ylGb;j9WbsJXED`mouz?%J;Ut2To$4%2`iuFDI
zoDLfuPW9)HT{PLn(U1B$>DkqEzAZ!2II@GU)rjT9Ks&4RX){^J%XoLUY}cP;wiaR}
z#d)}lX^bh02L2!PbNB?F?qD)+-=v88u%wz4c`3C0t@-x-Z0`m31pEE~shJ^x&5Uv$
z2N-;1M<0gB1Dio{_osf8%Chvr5_E}jrT?HaVdL#L#f`D48Q!eadhK#r21?;yH9MQ=
zrF#02`VRU7a~3Wa?=i)qKVYX#2dCe_>G82zuCpqG&HcU+VBbtD!h(ZUU_nt6NYkv=
z7e8!`12)`s#qH`EviG=TTK)3R>}DsmV}5=mD|MU((0NHahhXC|LEU}9_y3+4t1t50
zR!a`{_JO;*e}9p8DvU=wbw
zh|Qf{wdX{NzZpWGeOqu+HZ5+poepNS6OV2vWkPS9@AZ?{{qV!sKT%=Np=fA8^V-i}w42A+ucFNE%2=8p-UXvimPr9_
zVtofC<3)bWA`ku2=8zt$RHo9-^YV(!5GvGL`Uy@^OP2~v1RWjxot>Qw^;rqNwaqcx;U?3!S7~Nu0C-1Juv)9bcNTd4RfH}`gvzKe};tByEBg$bXpd@u|5ZGWW@TzVwL_D
zv(=^SS^DE-y4*+$slg+@mG=K|XDtxqH5_r#UF=`V3>Rz}7NdPp4@x-2eyAo`JfWP|
z+`B8C_wNF0}U3hwkwsEtnWpIM*Kp2IX8nj%TeL5j~1n
z3E2=o8iZ+EFdo=ct;+DM@{z|5ge*NQ(it$V?Q3@+E8Ap_Ef{veJU-`SW*&5S_m1GM
z_ok_&=hWz?s`Izp{^Vj(IGH+HTOU@}Kv+=-EFQ*Cs<1@C_>b2YR^CvztkJxVJ~94p
z!R>DT-4M1M@wWuK>v6u64uT-0N`Q=vJ9o$Rs#a?IJO3iU&S~Cze$jfQc?AF+WVDWd
zTtP|@xdG2P{~-6SKw`9Ows@$@uKdJ8LYeGnH962`2G@XkK%$e(Oz#%T?VOXs>vqx8
z@DAEZka^O^1bA|X
zwOZoL?slze>)Edw4ju>rGmahJlCu@+#|BcftzR#S$LaN)Uwg?5d^-&hH}Sy5as3qO
z%yWW7+c$X}!(b7Zk>6@!Cs;U
ztSX`Q6URo5??abs(Tz__&~){f+M5j24wv1QZvn~Ne`JTz=Zy*^M9t{7qz6LMpveDpQ9l1V+f
z`8&pa{S)dBW1>a57<0TW5;_TXsd4cRu4h)4ums#(395^nspQmi2@?|&6z5!UdQsJG
z9C%9&1>2+5TDu%^I4LOJ(pwFWq*(U^%~pThRL!qcLlbDGM@_5#daz(e`V--)km}Py
zgc?W-<1gVgWoOBL{zbOOvFQ#ame^<|8#UiA)(}OTgzd-Xc&3MV4D>Pm?Cfb|KstYC
zvKpJS<^MOGm`%@B@~Z)X>fWP4ea^(t;cX?CC-A<)>oH>9fT`F!I}9kBP-$WA3OJNMM@Fs#sfM3cVnG6`CV(SsV8meF3UCJW;19nZ
zlYH7N2~X>~jj$O(Zgijf=J&#nq09auCd8f8>&?_oTA_m=QcCZ8Qw1vvmx*_R~Yn~!#3#uR6pH))vlf$^R?-#
z#@pI8`uR-qqKAFw4z<~2L#2}%?zH1gY9sM{t6~!)m?O3{7bU#A_SM)9_x00leyBg3qjUDpkFmDW1I{hGh-B2;S2RgW}dgpMF>
zB;-Nu+SK%z=HMZ|7s27(oRdBrA3?Zwu@UbK4h>D2-f;X)nS^a7Aka2HPwlK{mXJ_m
z@9zF6<>r>F*k5c#QTLMHJ9gWQI_36g=Vn6MQGoHRr&}f0ma;Us9>*HqoXywUEnY<{
z?kNvOIc(p%pYXvicJS~B;huN9SJHCLSFabJ9@A&ic)nHO_IKkUv?Y;*BEtrJCh+d?
ziE3u=`jOlI;vLA8_yD6YH=RF^j;aWFUz3@cnWg(N#P%&ME+X7!^~}%vTO>eB6(I8V
z>1pDG0n?33<-FU$u%Ml{YqvPUTLzs`P*SyxkK@CV?6^eBJXK}DkLojl4|F?05&PF&
zkb?xXQX5ce5rn3_Kz~C4uXx9!`xChnyN^)s0zz?a(1Q&9c%VVzls+nc*(i-Yr6a3m%kbH;YJw5T9Ac#zb~hTWqX+eqkDmk$
zWJo^S3gsd;h|to!+I{Z-qPl)AN{@;9G5#=6tboD_piKCP5Q|HYhv2{v7!~@r<~^tpQ55?Q8I}A71J2IuF$=84?dHV
zlVv1zCR)%=mOE|FL%?sc`T&ip6CI!H+ba{Z4c{
zTn)-Cu}bjI%zWIq=egHDGV)O0^H{C5(iReApT6;)eCtWLb?m;gw{X$45NvnZX@^)!
zQtG|_)(hn}5gd|OuYbNie|37jcjM(TFzGSS8#!xyyYsIaZcwiGAN&Vh4i;OD@PftP
z3BUh`2(VKWA!3M-h*RG64vC-9&gdTK_P|=V``3cl@)hStYRmsG+n&5t%xc
z2{bm7A=(iwXIV3k?|yix
zO*$Q>BdEdsUukUPKLste3y+m0dEgLQpk#}xP(?4U_-Q?Vp!^X!OPEDhcVw`KX{@0U
zPbstUsVCN0b~LW#zK2k_D#K$LmPWaV(Ix3byh8>$_Vw~dnZvZA59h5Xj3(1|=5xUs
z-R3~{`;F^U?dlA#n`jyQ*;(h%+3H&sJs5q)4P@qFt7Rc9;B<#M77xF=mkt3nTLM3}
z*yFa+OI-!&oWt2YK_5XD@A2yT1
zaP-JGwgKLg$iE-H|n_vln7REx_h6tXZV3n1upE3Ko6yZ|o
zDPy_cF&DM$Q^wKq^ucZv*EdO)IfC6R{Gw?d5B*O@M^fLNB`f)~&1^SdbW4BJLKGt+Xjq9^bL1>P4Al?m
zqN^1PcGI+rsH0`rW42{x=izfpU)n9NH+*)zfn{Rj*iNyRxxyC$s+}umu9}
zrURj{?B#E;xVVT)c`o=FFC?bDPxjjd@
zm!J`As&~Z};l0ZxjLN>TcTke5(r)+ov$E=YUN+t=-Q@sfHNtIKJw*6CQ4a2Gf4H;L
zy{wfzZPA6_^p%6x-T~a7!NEtG3#Mh7%0C$1A-?$F+CVR>=HG5VH|;=X5TVH&yTk)?
zc!k`X{`^GhEie3XN)wyeF>vCk{&>|j%)tqtVysKscjcSkc9hL*p`>6JR2f=hBB()~
zD+kJ7pu20hhO3
z4BWOq3>l=vXAN1b1Y8BNRy)5usQH2|F>BA?kDJLn4G&TL1l;oTEcgd-wbv8YKXoDN
zeqOF@v{cWG@P;j(&T?PfFj&Lwo(_&b)10kTAph)csS|NfcW=Mvm+#X@0vseR>4Hin
z4|^-Erl!+-bh9m>Q#OznF+XmQLk>jMmBmGDfIh%82M4pm<%Fl95!?G5b7zl|^%`bU@0
zzvq|G=0yV2sl8UEyby2aBJ}tQQ`~loD(*PO!VE5ru9nN#8*Vt85z>O~K-LYXX!@ib
z+PR#Znz}q)k&`ZDMS>?9W$pC55@g9qqlrKSr4PD-@`B$8JW@>&j*yB*EuJcWnYNaHxQ!PflchBqjhEum
zKpvsSa$XP4oXym%jv1HH{&QvAv0OfkPnKiOI_qqYK2N@=UF#0}@U_F4Gy#EHE|ZIG
zx`I48m7M$Oeg0Ix@!4(TsYVw}RI@3|7;?%#)ONxBdKr(!gMLfqa)#O=cH63+R~At0
ze!JGUoPT`n=`x2#P@H=Kqx`650V&NTS+mqQ8@Jpn_({?1)9kF
zxAc;$>*s=k;BN8eLZZ>c0n_Ism=5mlI3yb{_KN&`e8ffozsv7mp0NWuG3`cK=7*Z@
z;PgPF=(2EryKJ+)xoc!se`$a_HB-N(Gm_SBd1psl*BNxvw{IyQPzr(I$ov5Tij-S6
zu&*z8SX3iVRZDnHNO!Wso(B{*`oY04qc-M*nkI-;6GNKLk?`L#?31_ZMsNlej)v8GwAe=*Nyonr-SlHkb*0(CmF!Wl-YAf
zn<}$N$YX@I#U8cyEnjl^j_sqyLf-F%?3-3fzAuK*@`}_x$l@gXN!e~ar&4F%`LC~n
zpoF$(O*VhHJ@Fn^wbDrZFYE4v%U~wqpO&h&UbV6Ol!I-_>n~|V?g_Uc(;c3eBp_t8
zI2qzfwHyC5#O0(=$y-}q)~0KXEVpTAjl*et!a?NW7V?+YJY(AWM=|9ef^JVt;PepS
zQKsqB+S_sz8!hPFT_-mcOZd3{4o*lA+&>2gSA01`t0Vsd<{Ip(87}xALw)@}$t*zJ
zK2P$&S17%Y?5>}JJ(wj;
zb*62lt7Lh9A2r(odNn#L%oIcsiY-Vit8Cy0Z>!2ggE_y(H
z+R!6TpHn7bVpYJDG*Y`}m^1qDQ|yza=?QfSCedl@nEOsN=2ax3Zz5H1UV)}VP!rX>
zQ<&D>V=RHC9{AbG%{~)9AtuU1Xq9-X`kuVOkf}jRfuR6Pl1KPs221hz)%rPEn|nrqA*ri}w~p9fX(`Eyp^{T|>J$)BT}
z#C;N~q9!)#oP9V%C4fQGQGSh(!rkT8AHBMoCbgm;*%6R!@hFxtO;HhL>o(wK?d5
zIs?+#%Ia#|bg+yjTw*__^k=I=
zQn0>Ve^>_vzIga6(SoTFpN-Q&1gMM0EY)1NRC40-NTU*cEfT2TD|tn3c`$RKM|o{i
z`?oir*7Yms&#mTNP5gSBOnUa^F{KOR6C0R$VS0UT0^AOdJX$>#{BNB3#^$5HA)>Tl
zeLC|<;k|RndmIbNpzJn>s0*g-D`v1&Zfom;g3`=%6D>e-Fc9Vrn%^T}MZjTo`$%GA
z*QvQOG?|ueQQq8~UoojU7b$>udJvwaw7x{r}I$4z`;2R)L{A75Nv
zGO0%1McucsMS35c8B<2?1gTh}X>si?b1SoIRhfQ14(6y=;M7(6o9wn~swn=#4o;j$UlHJ1S)HK1h(yBI0
z5#yf6NBA;j&6iAvX3w^&Rhef$7u8*PboNMh66tIV7k$k3=wFBWBeAc2YaMTse~mc9
z{qEY(G4r&<;jCrEq{+r
zcg%!IwRu3m7Vx@6%O5AN;9iz}=nzgr-?i6|`he2VS`jNJoHg+u7ZPun)AM
zmF;atUHy~*?OufcejEsxH&(%t*5}t4bGEU^e)`V}M;#`!Wog!Y80B4x0MBpuF<(wy;Xs
z&i;{IqRuEbn!6iDX~()zroa8K&C9BM*!3b-#K6;wCxRM!y@0Kum}XNw;3IpO%w^~>CxplSiV5JGrhL%l1w68-w{z0Zt1v|?D{Z*C!*Z!t;}<6g`jkAcAU#)6Af
z)SAFkaEx?t_E`n=+XBVjB-d9RgRK1FQp+_b(Og-X!{BL{Er6;g$eLJD6LInaOmz<
za@EiM5#{6+iNBlEJ1(L9Pq|MiSapxA(4;)w-++jkurl{&yxHhdkcOoo(I=U8++t`K
zpdTm_X>aNJsA@WOq^Q7Yx3t_Ek9aO9l+H-spw5kK(5HIVz~{+i9o(A}_VCsBp=cRo
z+!%xK4h1#yYWZAK9>+)tcSKw9_Et9zlA!A?h_h);t0{f|LHFWww{dP0E3oGSzI*R6
zqI3-ZyjRZaav+P6F>dYi{Bh@v0L>ekGA+UC^v7t8pS9)Kv{JYML>U(Q8a(BpYJCT$
zi?Eg3d-?nvEhmQ-YR(6GM!E~-piVpUg~ALwFdhhalKgwO*hvRc5izFFt*2|=ZRPmO
z)ZZ1#8y=zkkHXax>&N>4<-G#3C5N$FcHkCR*&M9_tUQV;sNbF`RZ_aCgA6x$im*4q
z?89}(cL1|)Zz{mv0z*d273=04ZOawTd;`ApI6qK)?m)_sy@LZuAD1d$7y$t~47j@T
zQYd2y^_~Ddd(9CEK3U4rd0IoFcPk}0c#pds6B@yEiAEDy*BzBTdOuV6eFjaPwSRfG
zg>|K&s@6^$E0bS#c42bTaIM~N)SvxEjjlKQ8Apq@=H)rlA#Op5_dQ3Jg7gFmA;CiDcU~
zuk(MaI%YWwZoBqsaww&j>`P0Tve@3hd#xGM^TTwKM^G^6SXU<|w5iDxN=hoQ*~)p`
z1Cra`orL0V1dMr}vZf3(GJ^Gnvd``OZ&4inVL?eVB~a_oBL@Y4YODF*Fu%mNvj6QN
zg`2;^t+$g70V}~H_|YB^{$=2|uSM_eF*deRy}JlZ1J;8&hF|UA(Y}3JeO|#GI9LzD
zfrS8FO!D39m7N_91Fn1-YrWL{qr!;fCo5F3;+Zex<)z0s
zcw?1JA-HIfhq-UqBc#iSEAZot8No+ZUL_MT_DYq!sBXtPC4lS?tK{;XuqHD~&QJ#>
z7BfuaFV7Hsv#lirJ9rllc4m%D)9Lm1I8qPx$?P6KVNz*of7c`=pp!WI_v@$i#RAjm
ztsxg-!W4r7ZKP2S-#;&GH&(}L^zF)o1w|6|n_qyh({LY^_YiYx^%>h;nVHhsVnz-!
zRkm{Rx@`RtsMbF7Cf<3&UMG!|B9nCo0m7s-G>qBFO3ZM1_Wu02PY~k&DG8J(o*#Ds
zGf<+SNI_dtfAc9C9Q1%mB{pioNnn`L>LJv%w#GZ}afDeGf0|Bto<2}n{J&0&TZ{0i
zkMJoDtf>F`P-=kdI#?fXp%n-z{sw{RR7wZLyK$7|)CgDtcAwp^NZ=70#k15=Dm5`i
zXKZRT>4#6p>i@uaq09Z0AQB|}LJ>h{KAR=aB-Mr$qkA>g)tYldBG`(O?Q=r}AM3IY
zEcBlKCX*=An=bV6^6r;aK?YmPr*mD?xVG5_fagCgaK7<|k#1)BBWP|dVAZ%4u?Ie)
zZ(~oF%N6WZ@^*muz@w5Cf^W;Y!!a(N)JX=T?_PcuP`vEjT-ZvorhkIoP|}g(ocBy`
zJFHt4%{FeCrKaV`h-^2H+Qw5Ixz_~fSxa?3699LG>0pwbk
zk4E*a?VlbLKyS?HRM#~ISx)I?`>)%k0;kIg?|&<&&_5~%-r&V(4gH5rkwA#w?jl92
zSc`;ym89}*(BRF#d=ik!JLu6LYp`fP^596s1lPk4UpjvSL$iQ#D@S`6s4_TZkFEVO
z<2C!Fi$u69*Q-pl^&?(V?4K_=0ujuW=wWdw$%okxnt@97
z*BaoRd&JIc`_26lZ_ZYul7(i#O#{;pY6Ye~QEbFE#f_~koB`9v$o-%my-(!o0+SSgo?_t4)wLypml2?3t17gvD
z%Ybd*f`NSRawGhD<5t-F@A&aPeQ;xU`P=TA07RZpY&|Ld*%O1eZwbNN4L0nRD0>3f
zfI#Yt0h|Sse3PH&Zb{#1024gzVsTSVX#|{)$SGkq+cm!vL0TFamnpbJJ
zty0(TLuKJu2eEDb$`-FG%^-Ybsgy1^xaGM(4aKFL)ya^HFMs*Pkl%e}gu#3-W
zMGo`N^8BPZ`^}h~9_&y-*@x+
zG=kHhNc!RHyJHwS1N|#|5tlmHzTx<4QAb`D;~ai=@@-yAPuLphn}+giNxB~u8IgCH
zlH5yv=a2H~etP_rV_=u#L(*tPx9vAvuXeqrtLM?SUW5<(>JH<_Bpo%xN-j?|guJ0d+jy
z;^cjxR5%3;bp52;JLC2*mGU(`@!=GtKwT#h)Seu7@E
zncd%_X;CSI922a`^1(6W2!*DC1n6gPhNVg-b^0!Jc2Cx1jH45>`bnrH%2~{fQeK32
zMNXBXUmuZs#Lr6;`BV2RjeVrEWF#zkUf|<1CimT$FWAlR0V`S9M8*s_O{bzv*tDQQ
zw2b~ypXLKly5Quk!mcK)c7#R(UFIP+8N
zf0z$Ab)acTB
zvAqtLv)&gKrwTWl!hU0U#oNo{8Vlioc7Xzg+xFH`)f34rCSp3vMIrR$Lb?8pCr6?j
zm_mc^u-hHL3nN7n+3G<7aBb-_$(=vbFx8qX820)Ey3AGP=q6r~Iw*wez!yBnE@yfb
zINV(xL6AQkGKBj3xXFz`T&N!@y)Al$opb3E7y}$!)TC-=C0knD9AmRBUgogN21?dI
zDgQwuKU~m2I=}ke$jPLk34{fg=Omk9zD6Q
zoC!`o18;B2JiI&aGL>#iK>tqjyT;9bPY)RBqI)w~E<(AlvCYhKdacGCoSl8cy)W&S
z7Z=A}gxgN6|2qhWZZ2EeNbMT^0M|`b`z)1)(<*d#i$MS7r>^kVFY0tZ8V~_0
zcTO(AMLW<~EiG
zkJR@0qFl#Bo<_?6*hAj``8Ddj<&nO{yivbS%j(dk4C6pU=xHT0#TN47E`~nAVj@vp
zJhgjb@XlB(6~pP#^IVU6oMc$)B-vm3^k_Kn{8JRjTYBVFRx$obB(e()xm#HU+fb2L
zfBgHL3~R*lKt9ko{G(`kx3y%=V*S|9my;Ur(|IFo_@`zcpYuodh>SXe>aMFda{6ln
zo1)7>f;MowK3(p8(!0LJB{7^b?HE4@_=FV!>Q{Y(5Aa7xXopM9IU&37vybuKe#y)zF8k$fStGx+?O>4=hRqIjbX#FwhbvbNgyNiQ?V74
zy&#MmImaPl{CUuSQOqd^I+bXXe%INVg{*;!$$2{QhSX#OV4hn!=P%8oz8@HBQybX>K9Fn+c
zYyaS2Cx|tn^p84(lWFg&sU@J~^z(yfSB+e^WD0r9-&exnES{T~z<0vXJn-86S6II3
zV8sw{+YiiZXNP&rP{yYSdE(&OMfcs3`r|1UbCqIS~sBG
zS(mx%Jnh+w&_Pp*Y?XYa!zHZkzkh!=8e1MOM>d{17wvz4pl|gwp!^AIf@~*CRd|Ar
zH=dgbdR!w5cdU}daqhd%mI;UDFqZvg%&xz<+a+hu^yD3tnIlYmy9JWMtiv3~mwc;k
z>pWrD#D{-A!W#?8%B~GC+@R*mN2()AoY;eMGz1zq(zKB@?SGFt+#nRHN*jM#h}+5n
zc5tJBLZklY4jflk*Z9zq0Va`=7+^XkgNa6qxwO;FH4MbVi0l9nPiGCCclD#3joA>z00dGw1EIaF)FA@TQsx
z9rlorQp<1GPe9w__sTQ%wLz6w)}l~E(yYK)^yFSN^ety*qU5UIj{}SgU8%6>h8(;i
zKKH>p;@8WMimh!tMqGc$%Lb-YMdnxue}Pn|`UW3p%qdGj!N$~Nj+}G@eST%XmGW!b
z&M{xcQ!_1y8-Cu$lRU`!hRQP0EJL7y0!!+1e6Qu61E=ihr)@rJBmDWLmu9^HJ)Gw?#LHY!oi8nu)^e3ZEDz(Daq|*(X)fl_B2o?Vx8F
zXDXklVxs@{M6nth5r%HOsF^ircPuutF$P9YxA@Kb_W@PP+J--Wt*s$JbDYuopL+V8
zy}eBBr$hD&shN}8U^PcvUw2Tq@xD=v>NNLEJDcEO(dmR1k8sMC&!Xw5a`&{02;ilE
zAa-mEj<@jZ5#crz;nUZJS9MJq|Di6gvsmx@EjJgh@s`V1R?L7fE{faTxUKMKXJN
z@@VV$n}mfja+JJToPjto?4%fzdkDB5nBX}cHs>HnP}``7v-yKeR)}+gdvPAJ?FWH2
z`i9@=E;X}SeVN+3O9RGivSZ)mCo?V_L>67Fgw|dmWe{52XcHu4(>vEkyeyHj_|wrY
z{B7Os=i&pb(yh)#W)N8-OpeYb#%BL~XJyK!Shx1?FJD9W@(}qR$qpM4hI)Qy&mER7
z7%~VVfo9BQ^UHMY*BB{U*+lF21dJ2Y?=0jDB3n?YD(vk3W?$>v$AhQoyMx49-A-tE~>7uyG8
zxv#t^Z}Alnkz!Wz5Dj!Oe{v_P4~tvJA%*Wd*y%$fh%(#fZ2FQvcRY2ist63XSL^^%8eMp=i{!jz-lP
z!y^5sBa%qnOgXdBxJ@x8t$C^m1FX#DvU3jwCx2y+oj}>5!ZlaGLc=-tq=6<_nU^Io
zG2eH5G2Z}HA}$PzNaV1U;)VHX!1_CCgtc}N58(uZz8!(f=3XJ2lqD(loqc~1)0v;6
zzc6fY#wMRu*j|T_oQmKaHwF!SxhH&DxIJRtPj-oQ4WzRnmryY;DPa>Q_P+W1r&o@Mh0LNh5Me+S&zc9Rr7C!!vEBHxNpc>F8f-1
zhrF*(haai}!4=GSK=tu4-&SXiHB-;N4MGE#p=4$lY_+oBX%`i^$_!&ND}yOR
zB@4YgN%|CB^rVWapV0}BykYG!=MV)wdVJg@L&+Bg*ZQ;J4P|;+Se|GL44_lhH)%4Z
zYZsNi8^y|Geu~dW5gvC~Q{kLX_y8Z&p28`*mnO9|BF_`_oTQk0OD9uPo8+H-%tBPw
ztK{hDFj^wvI{BoLJT=C|UoZXTZ-ZsX@?2hh(ol^nnbhCfNSgxs5G9^D3LC=AR!YwZ
z6Bb7Xl?AEs`{CZrjz6+#v)v&un1dwLkHNXGTq#fWHc6{mJ`qvgAks&xTBE2
z7pVajGySiVWMtRZ&4Oq|fsuma0%m`FXAQjCf{qskji_gta=db@k4vH15~#?|gM%Yf
zbhG89b}6`B{kF5A2!pQHk6`B%R1M&k!@a#a;CHl^Gu@W{b1fHQi5Tzy*B-og{io&A
zt=A9trdT!KZ7o^@0ux1k(Mz<jZRE?Ws^XYNT{ohD5@H}DzDBl
zT~PIf4P})Io8*^>n8<4FD>=GCUIdH)G6Z*SW#;XAtVy}&D9o`Y?(*MUpIJ2v5ZrHG
z=CtHx6~9VMun3^F?OBLz;_hK^a1o>0p3X0+BBY5;KQ=tD^SJ*Gr_x*LjIB#Ahe+@7n@R9XO_HCV2OlnlCxboL3fYv{Ete
z<${N;RPRMyNl6LLD_M{ZmGjD#s*R&WkJcJ$YhO;pt{x!)--}uu!q^vC$q4XiV}tsL
zdR^uK{XY#>Y%qGcq2c0X7Rd1`rum2N8%-BIsh83oyOUmXy+fN#|7G_5PfGl??nMe_
z_1WF!q8fdv%!kdjZ(Usv_Lm-aEENU=8QDqjQ8%pfw;F>Q#HLSMe>Sc72w&|gm)}|t
zFuqE$KtNoLGU!@2-ma&}0Gk3b_8Xoj27^GvWhkqUIs^KPFmMw29cIkInqFrA95HZ<
zCi;wksSm%4s_VI}5-(0nR^Nx@0F&u9;#bw^(%j7Q6^Bew3X@-(Cm)A&MUtZDsv~{O
zAVbagM3+-g7hcC_s**LodE8T7Z0=w5b3>_KFu`B}5GGkc&+M*wg^_<}&|O8krCoq&fnsFU@l5fCbENQ&fE~zzhHUYGvkIfV><;J8UPn`{gt$M@~Q(2*Zgwpv~~iV|8~BB@4n
zmoF>$R1J@bp?k%tuDy9(#I2HUsI+c@hj1f8cq|a^hH>oGP#%^tOYngMmq(Jx3%3QmzOU`vP%
zxuLlmZVhpNL5gq?fc{LV`uMyc`UYD_-e0>Z$8
zFz?>^rbe&vLD}CEr$OYbyAe38r7z;k<*$H+2_HXU>g!xuqPFBDNhBU9jkI*gwkRL@
zzzLQ)IKrsXB0l{hf7bJ!>?J|gNbSIfXMvAn47dqRc?4>Co~5!tLMOWuzRGYR$Dl)}
zwjXjEzSW#km_T>0YS{d0Ori%`Y@xxchtXRPihqujdkH=;84(x_Q=W)22-u#yT+dKs
zx5D38iE5~S&MKCOSd6ufA$n+BR@>z^)uv`TnK`mA8P%}$hSq9|pupg0XcD(N8_eh-
z7JC0+oOFgVJv1Jd_suChc6j1m)WWWI<{&z^7{7Jnz@(vTe?rO0(L#YaajDz)4%A2m#Rx{xXUd@=N?vFCt
zp5rHR$^lK#ad1#ECC?90DI=L(Z5rUnp?(?k(}aU+5qox)zuVMl*yWJPmdj!?RTXYe
z##`38OcT#XFn2e&#|j`~y<9%;yYF3hUZMce|v6T%RGx
zn*YFWQEk^p+Ce};;GOGnQ3;ueA?BibUrIll*!e?cfNqm6E`O16)3&Mu
z&G3`+LWpK9Qny+<8FC{Dg
ze-1euYG(JH``5wu@>%Fvab;#
zYf=iyG8p?lwnBwL){u}rBx|^K*_#n#B*G{=V;du~Os;(!@8iDj-|w6`XU^|;X3lxe
z?>zH7-{*t_-yl1U>=lndgelg1`_T77emt+wU
z6gEJ)5)-U8$ewB8I^g}?tjb&*C_kCse?uofPp>Uy;(q_F*|-d>GF_9--41R(wXSM1
zOu!~q1kWmT=gHK$#b@2h;5424nzLtbVzD}VXcCsQu}?8tqf;9;7239)3bC7daFDzvF{t)VUR5nkwSkw
zZ!kbxIJB>;Iu03HnMS{EpDl6gly^JilCr!MB&%ILrqxE!XVokR3~LZjWL)-2w)~Mp
zK~Lkiv9
z3uFN8b%!+q^3U9y;kS}qkdo&G1qe&5AJ~V^=6VC%!GWqpqw5lY%6!H7ZryikfE&>E
z?HA~!w2JhVjX}8|La38Q+-c6-;0VEL3S``pv&ectkVLUk-UjO*=WaA>@e)Oj!c~*q
z9;#;7t!2}WAz(m05?;=J$jKJE370tiHe#S@9DO32!E8M?u_S~;>QwfO8x3&XLR#JQ
zj$ZK-GC2U$sPz?()+&Z5rBkbet>2?kh|3=iRx6{pHN4j98yawomOOZl`?5Q3O;Zf2
z+f=~Mqu$tvxoR~G8S-U0V5%8Znc^}NjS(j&D5U43Oa?t%
z>`JDwv*(1w8jG_7#a@_yj^Wq!l^?>jtAN$But_vC6k(k^dLqv?wm*?~VwU~{~oiSo)whh^+
z04y^N|21!tZ#181Jmc*QnN_YTcN1km{133(V+pGDw}v_>ZaqhQiLQ~X9IL(z6~2NK
zFi`C!(HE$dxqX5LbS?2d1wD+6aE8IUt%LCQt-N2@{ai0m>szO}OsH-Lo^TH$OWq;P
zGQ)mf7wkQ5COJ1t1sQo?Z%FFTFR#EpYzxNmN*r@MXXG@J@;$uv;c=9PP3}0`E>ofx
z?Z-$rYda>t$>HA?#OdK^h8;v#r$X-in64KvKH?;`|3^``%YkSLe4>G<1n9q@`!$}b
z5c8sS&~={*w0Z!n1$eC0Hqdy%rN_G2;qytuBtQs{9gDZ)04FLtla}E)zc$$1y-8^x
zZR|*$Xp}f+21Xw^ljZ4>X851kBIun_%i09r%EtcDY#v~lUTEZU{^jBcsB-|MFz9Wv
z;L&yW{Ms4L%5Qw_EzL1bDuh{mTT49>1ckNDwjVt`y$A?8q2Rowu{29rpZ#x6p)!Rs
zd#dp-FbnVlMl3Kz?mjD^+8`bcXQZrGU_Ivx^HGTpf4|xa=+(@Caad$5xpl#tc(C^h9p1kK$DPjOnCs;Y6eu8IrZ;vw}!}+B2AwU=bXo0)Va1(It0L%T-5}2?J
z{?0?W3u4q(j>T+_bP$vwECf(Z5DDq@kU+E^eE&m8=EwCO+Bl(Iv6n
zKqZ)*KG{VTzCw?<7mT<|Ji`9)NpdeGB{B525+t|q?e~K#0O&vm=*ZE~
zM|@9bvQ(oNFHX`~MSN1dpj^xvvzEjD&gQ!lNzPOtqd4P0inlggJFRU?2>j-fS_e!K&9dSZhg
z9$ri#^^zVhF0MuD7S6kbO(9L&kH!J9+QMQ2@5;iTtXI?ABbjC?&ffkRgs9az-2MYd
zD+>yWX5#egIe2*y{t2$Z!8eVKzwf`czA`A>OsLT7T3+rcf1x0%AGf&33WLF(x%9=M
z+49);!>*Jra4K5n&^Q+FQR+D$D*Uz@zEngaF*C~obyaMy*hCclXpr@|*u3`H^OjL}
zof}I_OX1E=FA#}rHp!a_cbY}5#HZM+v)ttR0{&r97hR$vkG}
zWLRTe*j0isR!LDux#?u1C!B}>pf=O$f3yT`Sk+j=52e~S-qH}F6oMXJU}0vaiV}#H
z^1}MB#Jys5Xfb*j)9yC_cdq;AV+ZB6Rf`e(v;!YY?7`Zz7URi^CcfTZ_1G5iEET#m
zJ=pJOYC(55C|@jtX4N150Yv^$MMaKel=n3?)FE|7ymc%WC+DuE%7C@t4_A4R4{Do$
zvhq<<`|Mm__5Ja~h>!9T_F%JlIrq5V&M`7g)cTG`b$bG@f5#7XoCWvdX;`0)*QohwbQ+!
z5ad64&tf_rsG~_&r#mcQEcxxT;W(Vm_U`Uy{Y&#Z@S23?-2k*>&bjXG9
zl9Et<1%*F}M5G0fwc&t<^pt~>GaC2L40-R*L#2S%hBH(QbM|4AN-@QuaDUDPN(vaN7}pI`IL%D%BTfhaTSbe)dR;Zd!t2YJ3vHLFj76+T;B96f$y=gq8XvGM$BHG{46
zVux){R7X%g7(i2|
z)!(C^JlGF*&2svYk&%!NE?XBT)4$U~)zsCQDou|_e+foj(rm2p>`*k4%!T$?v=dldovf7Pk}QI1`mmPQs_rLC`=nVwpcG6*RpHi;HQ%Mk$a?
z`*R@oa-bf+hrRtd1qB7*DgwGR$#^WC#r5nz!Hu5&{wSapw6fwkX+KO8!Ol=S%5wJh
z_8P+B^bs9lhlhdrF|2i?*S}5htvR#y5Z;j%Moz$0t_q@ES@LIGM
z-ZP&x^!4dsETiU5vlx`CrzehbaujjAIO?~I=l2T;pjUmB)Z{XFCGY2z#F+b;p@Q_|
zcJ^pNdlhA6Lo>5IOD3tJil1l!N=Zq{C0^csbu$sx%(wK@-&2#McD7d@va+)BiHY4^
z&X4@}hlPUnO>b<;Y!zaF=|H~pK2hk4>20uvky{5MQYOlVb8YjpO&IByBs
z(;cuJ92{_Pa}zD|Fd-*gQBhHy7vqrhm6est%geXGWD*h*!V?G*+ew`)kCdmV&!xQ-
zq&T~{_>yN-r!gq*+pl>#=CLu!d3jRXHJU)6HHe>{ezm{Y+DiVvSOeB4X&^pr1SQ}j@S0ae=yy*i6Acnf3K%0^@NBxk
n7!5xQf!Wy!`QPpTe|gOGOMQvAEuwA{0$z7??rT?R+J^rJp_%(Q

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter4_57_2.png b/doc/LectureNotes/_build/html/_images/chapter4_57_2.png
new file mode 100644
index 0000000000000000000000000000000000000000..fa199fba32c334c8dd5da01f91f4738dd8d5359d
GIT binary patch
literal 11509
zcmaia2RK(@`~Tr%GK4;MN&}E
z?w>aZI5=AhHa;myf;-&0{2s#le|Gj*wJt_>N0_#l8nrf_h)%`GwJDbYCL&_imKiC
z#aUMzEcV3(Qc8CA2f;WBSnLBKx$tmRY62oEDz$2ajh$VMftA2L@7C>#y{+E_c?G>MFDABC`uW-RmumN)bskskuka{pi7mAF`Zz3h
ziO1M`3IC*K5|ON|?8erX+1hN+#z+{;%!|;xJpPJ}(Wu2ge?sEoDAd)}Ib&!}oQVCF
zDBh)Z(eIttx;-hgs^a5Eu6^l7
zBBGm=mMuwCAKd16qfeaI=Mfa_@~~STZz6a#W4chW5=%)(cTz@>ho@~@c+Yls>rb_Y
z;787*w6sS8S&8D~Uv76xwrR&QQVVf&mtEW9q|YVs=s_8M$)e43cE~P5Oo-=(2RAO
zY+;s=l*C2qqmKpE$1rIysyuJ6YHrTcwDx>-pF@RRQj+oRd_Pr5{G&&#A|f;qyxF+A*lcgTl6itcoEQL7BewQF5rYgBkREIzAL8PA`anwbT@x9c-&
z&rm9O`&Qx7B}(!W(&0@HxjDGF@MZTlIFw@;LrCZahWpE0^xob)VrFJ`@A2bMTE45p
z-!nDBV`D>EiMrofkl5PU;UMzz@}<7JlCNI98iu#RhDYod?LGgQdu{CZo67p?5R#Fx
zv7l?lKC~7V7Ks1yva>;{BhK%xHw}v;HTPi+($ZKT#mA?=dPVFq(HsyKM(96tDM8jx
zhL?}8{gaPZf2Ch#eQJCAbCveD78>syhKwxB)ibM~82g@*(Nb4eP*o+ullO-!6&@?Y
z6BD6h4RPjky~PD3C1K^|Qln#I=~r_LJyvdY_w<8BCA0J<@
z&dn5hlDdCXz@QW!W35>0wNY?)FO{hG+ut@`YBe=At={_cyKd4i#ps!|$6HD&DmJ(c
zqCT>(EB^-fg(L~>?7Y0AU%q|~NmmRb_F9{z!jq>&(~P^PXXNEe<_)p5@9hV8;i3Ld
zJ~yU*WNF1**3)>*Pj&J_!eV!-g{f)!@bvVlq;oFHlz8&tadGspgLTtS!o!I^4-6Qp
zUc5*P144c}d17CCl7WgUeMM?x>`^O3XozV|=#eW|uAm({Spm)7&#
zdwKT#rEIG+KQ&4{R$j6a{jA)#ANX)&X=!P+)fqNOEX5p)4Msm_cA)&4`@~br;P2mc
zqX<5WC^9WXUBUIGwYAWQh$Fov4slLn^$~sBe|{4pQyn=pXqcv)_rgUa@6JcRdnf87
zDt01h(ZUt5cv?JXv;>C6kz&#$kGqQ4IZAB|F^EmMOYveR*h
z>(V>F=^i|IAY|RagUGE+e5-50zM|pQtb#n|v%Q9oNbj!xu0))kPgKf}#EJnZKMQ98TI|h;_5h-cy_Cn=;8?9!E
z!w_B8Go-4slMX@vrue!vU+v+^xZ_>lzf&V7W@hYr#}Gy_>tM)WNEIAlG)V*p2ME2nR~$GW>qpt&LdLufYe
z;ZEG~-z~{f1@GT0KUc17rS9wPoyjv+%~5;aZnCqnv@u)MS^!BSnptWs-6t|KawBLdWExnS50Al(buhQO
z1_)%~Tg?m-5)$E)OLD%xzAGthuMcL5qi)?KDe&h7(OS~l0
zlSdIa>>Ma$xy?Omc`s~UKiua~NxM&Q?AWmg?dQs$YinO<#-cI!;^j+rK|u<{ai$}O
z+jn7o$S*ECn;=6b6zO(J2Zf|c#UhNaEO(Z&u0xnqnd_@a$EAZR*
zMSl8SmW5D=X>b>^d+R!_bCjMQlWkOB3@9V%HcJGjkd#rBsIIOKvKAqdB9&jZHn-kFS>s`tk!G5X9Eg&5ROzeJln+CQ!O($n~VSyseGtzwe
zd**%)2GMu
z-&q9!>x*b8DlS%XbQH?V&oA`ebcW@Xy6fT+1UV13BQjItiCIG|bM~uOfj}icxK86D
zL}X-jQVSJ9Xm0B1IR*$?^T}s>cxo#9N0YRWc|B=&Z?74wk&uXp;`QstfJj0>kOK0{
z_mwy_JQK6Ibma<%R$ZFp2v4bUdX^?AeKkdSmpEs~SQ1_t~oW3W#s)Ea&C
z97)~Z;YL~8Ys2yktyGzfwLS-6!E|sQ>0q@Iq&?2?^75KEIK<(T&<%&N_)^9d!MC8W
z4dd0lwfI_K*?L0v>QytCE$pnkx_V^G!p+(4!ci|p1qC68LFwd`xajCwmWZl9Lu1_
z8@v=59=<23X-Y+PpYy-#2lm9T?%zI6Fnz$YHnDHW6LLM$fryE{No+OQed;0T*R*iD
ztRRP+&%^oc5zrscU+N47-&^KA$ir?ZWf21nwk@NiD;;iGhz;BZy)|GYCw8iWo!0%O9E{}esh
z@g)SP+IfIMf8vb5eN%1bPf67jfGb7$>&gWz_H7k8I-Ep^5=bN-cnN4FPQt?(96WMF
z3YAdU0Os(2{eq-wc9`x_iIBR3!4Fl{;d^1WRB9PjiKBF@1_M;~(
z=qjJPWJZ`#Vd3|W&p4A-OO6URPR#%A!FL`(*@I<6Ck=~X=D&Ex2>n)cg-EB$G3R`?
zL{SxBX-OEbj_>E^m)!N^`*%ue>cA7{J%^WneRbd4oD{y
zfry)%8=gY^f!v#!o8MV0EGbd3w--QI#l`6X(Q@w?pu0<&t;^dK*9+vR3KE{Ty@f?c
zQqu8cA0VHBQS^dlKVMzxuJ91?_V%`BesWn4i#?OpdWv64N($0yRBN&R>Dk#?*R`i0
z9UyOp!CX!Nc5QT3?u7t{Out@qODy&)x{U;@Jprq;Jv5N=dHDD+2q!0}36QA9*4D7A
zx%ddgd-dnfIEaaz9a*t{4CtDMgXgjXM8w5aKr00X1(}+g2LaI4eEmwo#>QsVnRo1^
zR%$GqXcNdlrH&)G-+%lN_FA)r{37FK5YI}aa6Y-l{HfPBaX183{jt|gr>L`WtgY=c
zo;WcZ@2W3*70ECfhs8^uL!${cRVzaT!uzKGe`@mQ6+6R|Xy7v*-1x6Se4yMJfBho-
z8~JTRXR_n?j27oZXDiHz0{!DzSh_q0GDFDd_P6&(mfzfMEB0{-_VFctqI0IMi`kan
zX`)i3&`pK^y)W~%p*|l!h;MV4E({}(12>}{PgVHt@A>c>2~nI*PI5fWg>5k>Z9Y!d
z47fP~q$1(`%EhE}sd#{#8yAmlT#OPkZ9%w5~jokKM}rTEFZne$3O4hr3Y09Uk?+qSO%Qa9{;A`l*Hr>{H;&#aj4$|fDt2*CYqg$*PK`Q1ZWA=SlWXI{_J8T9x
zjh8IP+Is8ofwdw2pNw`X|IC>t`(#v}kCyRC#X($u4`O7E)D=uk*vuADw>K#Y
z?prxCerF49Nf!p0FF98pUy{xGY5bYftEZ=@;H@>gOo_c&OnFaF+(4EZjZOcz76tv8
zi%D^^VO>V+4gH%@l%FhUG<{ia7-uqZ5i~lajPI4d;N#OG;nNUVDl5)B_IBl+dE1_MJp7I~DEuiY}z{Kd5{;JiFFU-D?JxCzpGg
zI+dP|wtBBbs^#Z9Ep2VKS?1&IFCxVH`@O0=I*wWwGp6{Ye$H8_%q{7Sc>sxl6f0=M
zs^z?I`2EhB4zGc}Xi$9zp2e^DMQ59+#g9$}g=bZilxQR8<1;cpik+R^I>D9sYii0Y
zZfx>(Mnwj-i}>-Q4bvUk@13U2~^cnA;(5
zU9}2I-<9MTah9~rJ1r#TTG_%Ncz$2>L-~7uyUQdy&r*V>M4AXBnI2$BNJ#R${AdO}
zyy#}<=GybUDBLL+1G%CG17c!M-)i+B6d6p4s3Vy$otvTE_|s+6UHVHAk;`f^6fzyX
z$$x#7`~uU{438)}2C2cWD$9<7Q^InqZW?_ca&WY@wF?Rh19Eb1=U+2jdi-9u%U$Ef
zcS4_m@11vAsG1fodN}AYUtJzwbvx#;zqi{CzMT)578XlTPeDyh)#Oi|Re<8hnY5}M
z6BGt<{({>k3Plev#rhsc4$u;Z%Z=4@lb42_-FgE>+9rg)Jp$ZA{E2_$AQYz$fC(G$
z_hERXA{w+grzwi@7-CzwLxitAM0ipjp?||J6JIo~jmt$a6A6xS{l)$!e_STY13tjx
zbO^;5;%1723x@>ZXu?0ivihq#I;Fp9^)FWb_q$Qn0q8Cy>wZiSz`5oCD>^*w}6ln)Qr_+)w_P3jhkD=GBrAy
z)T&s&V)eTQgUBsXuslM)eEHJQ#Xo%^B0O7jx(HKPRHW$SRJdV~`?I>GMbnankB<+;
zHU
zjDn?O7ZamJorb4Rd(0cIfDha8+Sd3J4aSwkzsH*%Ze<6BH)iKt;ccQ&%qhey
zt*m%uXHw(gJbn69!fiHTkc5N%%t`fGIcri%adB}t-_z-H<#!i4ZREiZClw=T1n09(
z_S%&zY_hT}XV0FM?1L4w7j0Bpy_}_K3Q7s}Rg|L{0!By8fKYd;*kQ5CiUFJ2+w5Br
zG~ABkja2UL?k3}n_Zz=_c@_5|JUm)GNFVE$=CLvXYJ;XPm(Z9Hh~*uX05a7%9Otom
zGTXlQWR-hcd@|d!BJlPsnHhHWGrZ~sn4CMrHY3#`ZPvw&U%x7XxSD8wdYR=871c3X
zpFs+&vGtrBI0%a~xo^<*6c7;5a(%*!o=I!6dyS2%73+gukPhUHj84FV9D^co`es7m
zjKj|f2@JKhwQ|qf)84&1>QK3R7Pv!p>?t8xYl?fo1;eiy!Gfi0JQ45|j0|t=7R@H-
zJadK`LCs(g#4NCd%KG|@U|s#rJqr{*y>>gG!=1JhMAp5O6mk_6m4t)@Rt^r#$&)9+
zQL*>wRKeans>>Tc9+uwG9e`TwQ{9DQihU0sKK!E($wFAi1f%0ZxZJP^`kWEWTPhZo
zIA$rg4h>V_rP8CB~`?`BWx4Ew5r(zxYONHOt_5@C*
zc#)^4r!$;$zUkAUguSV&%R3FRP>eYxC`gDp3>g_2xU##e4+o=QozJRQM3;1uzW@s$
zJR(9Ke4LSy5uD4He{5w(K_H1Xl8?*oZ^ms*?(J@)%xqOI!iH6hK-JrOhofX+sFLOL
z=g*#e;o*%8=;D#zgM^R{p-*8~>HI6why-MSMYgaU-D!Qa!XW!@cV~0O^AzX8xb??q
zjry-I<0{iN(g?E0t51H6&aG8iWyFkYk1UnW!ABeBUlVP*lpi$(f$2x~3Iac4>v^ri
zm`?^05fKq^HCG8t*^${28Z{X%Yn{heQE5QGH2mi4&Qp0H`jAS&#qo)WnG%>e$BWuidHR?a;Pf!a
zcuRCxi^0+fnN|lMt_{y3dtNW*=4fri;_7NGxT@_rm$|{G|8y9@WHaRjd)bf+r`pmn
z$S`pW7ISC&|Uja_edAVE72?0xE4H)FinKRZqk5T=i
z8F5*!*zF=kH3msaO1gLdetH+^S=D0@|KM4$?v<66GJu)udHP@*3UEB$Dy@Q^hXUK_
z)9qlvf26}=Z$>kYPs)ihl0h172ghJ5J1o4B=aduKi%vvMOUrLO`3M~Jc>n=xW9-7u
zT>8kUsJh8i->n>(-)!t>)UD%JA#{5KuvqLcRC9u%G~hoIvU<(PLuu&XMAAO$3hZEH
z1-r5eY~R{R@FUy*nN$MtNx-Vkp*evF+4oC)q=((IW$d`3_wAV&jo=W#+pdn8e@AJ!
z(GSDdYx4uSC1#K};;9!@^~?#E!7pRuGk|PsP4(&CRQaXD?Ux
z0ZXWkIpJF{TekAh1S*y-2BPzo?3O*}e}drt{cUCmM+AagNkIV@mY7BAB&s|JIy`rH
zA40_pEdJ59bj3C+Q6eHDIRk^^6%`e(YvpjR!5Ues9phb+na5(uV%52J|10!qf#gJs
z8i0T+5jM9D=)#r(rm3l^{b}r5C|d#cwaB1_BqoOnw$SE>C|41KfOeBC`31>xR6GwO
zck9+kViFQ3NXJfnP^!RnaBx_KQ&I19?1133eszKkiU8b%5)=y7Y-~8Fm+b2+gIc|h
zPv0;D=W28h5=p`t3w1b!1}O62_`D8CNKEA5;UR)^uUF}N9-MVjK|w(@hXahzt$+RY
zt&Whz8$r51`AE(8m#u6(!%87|f15V+wLc?cqDeU!qF=tL}iS(H%M^>IV02
zp^OhXWPBb=>CLYr;MM*I)&Y+r!J+YCjjw4Yj7|x?4mO4v%A9l#x!0pof`8>)|Ef?I
zc~Q#+n_uy7v=S2OzcRu@&!jY5(@&#$$-KMLLGbu-=5(E$0xPGqmfui-ZP@W}X?t~r
z?}fp~$a5|pHJt#iF{I*a1qyqcFdCp0vQDbl$x5FX6{fDm%ou0qLFy^j~6bdl+Qpfo0tf_
z#V=aKzqHm^*jDatvX5k`1pA!Z31rxF%dqs0bEu(+#)}uXB1T56`tV}JV(vYAdaaV%
zfYq9_hr^?HHzcdWO)on6dq9)GeAc*1S9GwVLjc5bJ>`Lu&8
zBPbnLGMGYWX=%AjfAQ@EjM(e7j%s19LoK`-VzW5gPL3?PuFq7uiWb`D6}472kX5a{
z(V8yP*Fm!wL2i$U%1P$f*w{8Y+K5)rS|h_#*+p?dK^&3qDiEh=>s(rnCp
zZC#q9YPrHxp4ZQ=cYWssLFeO%#yz`>V*#7L*Q(z?#m6-|j`!s8`_ja$x_qh4(cg?e
zlo#E<;AheJ<-d8WCeOfnrhWaG;ER`oSZGE-SOw+@NttAD!F%H2=f^`xnPqS-TXkC%
z_uZ$dJxj}KdV3kUEBDElCR=b2z)}N)-r=n`I~0T6$rnG6!+Y8it_t4q{9_x(!R;3u
zQykK5@y^vMluRnteC#!iM4~HmL+ei9a-_Ob^pc{NxmJfqP=|xOYM
zet7P>2-r6MC(ic^K{)a!$x@ts|2{#=L1qWD7lgkuVe;pTiGhaN=T18o9HhCK(R}wq
z@l;O{J^1`k7cla)hzw{AC(R%CJ$LK%4t4rU{iuBk-IE6nY-e%
z6PceAT4ibF7@5@V7A>xt{b@WMJ+fje80}}17|>GH$m2+5$Tek}xebIi987CyF%Whd
z#iDfqu*`jDz2>iV+@ZZWYrD09+Qk0q0=D13mRi1{RrPfkBER^%-~o5m6Y|<0KWI`?
zQnnbLg=fp_j9OG2il)CZlkR8&YsG=>dnhS)1P+W{lrVvlyAp)Dqa?%^4g@2qBiVr^
zI?DLpUQ#=r334FYANMCFgizP
z=RoD>I0(oG*0X1?TBd?=DgWmNlP;7ZNJyge8!QfB?%cU^1;E);Zyad9$@|-nl`0vvx<}b14^Z&00RSi%_^;l#Nd0vR-a%N>V!`*#ig1}TtY
z)L}Vx>Ch^+9mKYA+nYh=kq*479V*
z14m}~a9oGMjg?qiM<=kVN{-*4loAIAr_}iu@#yF%sUh8|{>>G%(m=U5X2OY&!Lm`)
z(2R_ax7j%Lfjw8j
zNU0m{elpzbndbg&LlYAd1X@~J(40a7%ZWBI#PeNqTpm|zGXDhKbds$c?Cj97u-v^4
zzR{Q1x2kpEZn^a8R{Jc*%ckG?_5
zcSi!f0HPf&xk5RlCg*Y%%GWqgo;2Ir*_43d6HxL(r!i%4o}L?5`l5wl=p+MI2^1MU
z6f+s2Q%tYakqvxNv^NXXrhY5hS^-6T<_&4>R*ukN
z0JW7PG$&c%7T65LU+~igE_9HKir#$`wHDgq?zqh-A`0MBUjvwtzeVUkf&_f5Vk-9o@is#I!rDEC6>Cw@^
zM~@yc#(hK5Lql;CJhOCiTvP5&f5%4)ficm~#2v_s^~sJ1mtNWuifk8A_
zpXfzsUV^5Z63uV$D>QgCpuQmVb)pThJ#Q_#-`g=k&zwK>CZHXh+?q*p^71%m?RtYt
z20clhE5|56?~t|h8i?gfy1J-MGP5bJpews!5&4g7{(raZ{zuNWw>
zf8UGo#(US{@LuoXoPE|_YtCQI9ik{NaqkYv9Rvb#PfAiu34uU4f!`+Rx8ZNzi3Q8T
zA3RPkHJp@fO`Kfy9gGoj`c8IMwoX>&1`l0~9URSVZP=LL7sEp{Cnq~cUS?+N|MLY*
zwhpGu9r0Bu@Fo~`lA4YP1SUT6Hwq#tg$RLgmXH#Aq2iXZIpwNBpicO6Zbm7VPEvWJ
z{&?D8n@DD(9yPpan$HEbDM0B$ILdiAQ$yRGwrz6eL;oK
z-^538b@>LU2UTsY>sY%b`8=#}Nr;R=}%+zlpjh^_~#%?f(6~eJ_@h-apUcMV&v!RhR#$Ne2_>fM+*c>7
z*$&r+WY{A5PQ!4hF<;%GDov4zXL=$N$B-8m8mjy>R@zunGPqPemG}3*u6X*_Ira-}
z(h3R+L#2l85l_}fii^Dt*J3Tkq94h8@;*OO4Gai)5-Xj>N=5ae+G$nExWZy+VDfs|
zOB_9_q64QMN9&<_k+w>)ZfdRT_KlAT2_0d?Y(%_Hk6(TucvU!(@e(s%wV(~&46$=`
zWYUww(VfDB5C0&XoScYQ3~p6cR_1N3I||bHcX+_Mq(7QpSWv|{y1G2)_dE)#abC~X
zD1PC0JB8c+eM^hj)!91p-cpZDf5yw0!c4`-Czr=F%2NvBf^7P~Zar1W{Rv+Q9zRi{O+SLtD1ZDV7J@IHBQ
z#aa~~l(MArbc?mCqv_<5;4PKaigkP7(nCIc$as}0{)t&n*LhE|(Z}24VC7v!1;@?X
zcfUCN`yEEc6U}8i{ijHWPdc6{JRkto#>OT^!1KY)n>RVF$6E?tS3Oqyrs2^;*P<@x
zu_I!ne|dhqT)W+n>AXJtw?JKCJ+LF0)B28Zp;m?2-&|aBUZ*$LSCA{iWL^BSd@ZcMMX@A^W$y$r%ylV)_dsN+7?GG;EE(mz9*!{
zocLi+<9W1^?z%nA&BupNM1=mOM&_y~nd^z*{@I~rcX#*q@87>DriUwSH8G=qeMKVp$du1>OTToC?Ok6ygMgptSX6kqB
zGj+4^@+WL2-4FTsFG5@B#a!s%Ffq=qtOS>ou)r}ue9^9UAb6@0j7iA!RpWTF+Ogbf
z43C_gJc-M;7k2gT-kx-&yqg=h-|c(UbaZsB%WP-F-rn8@hK4BU_wM!lv0qtVAFQ-~
zeY82z9z`QzHc`olaNk=*d2_n=&uXkRfbn(Zchf!!GBUE@(9q_Af#}vyLdT6!mgxrX
zu8fzVtmQ&f4JOgE0e^1rI4*s)ooTR|X-NI?gB$jF0-GsTfqL=MSea@5p?nH=6Wk~w
z4hw>he+MZAy|~LvdZ?+XZ@h&UKVZ-h4#9i$oSj{^q&Uon&iCrV&CLyAcXqIPmr^DO
zi=5kG5j`X%Wbt<_reuxn7q}Z_+;+kh)7^<|)n2Ep#d?h<@LJ)4VBthoIOmcbJ6HZE
zT2DSn=bP1Ulnv(s{N#s%IZWp!FnbaEOHY**N38!6H;T>F~`uZ)MK
zV>2J1eX5$@0`#ckb@Jdx?CmOfD`idik9J9dZA=Y~JnvpYloT233@r
zH2M}Z6kQ`xSbM0wR+9g#_L>iAAMEa8+(MJE;0?3t`SnY3oEu+~uOO|wX5>?#^u-WGR{J2WYPBTAo~9(-v9rf{=X}Kg{=5n;=oq`IN;2;oeJ*+Bk~#mu
zqUAOC9<03j)!at>JOr2kyL
z#PPkpl&!Fus;Oi*SGu*|i`L&r`BlruR~6p^oBxZ4wWVcu0OrjVh<3;oT*&919
zsLTZ3e^!(qCzm+z_J!ZmEDzaO?LJ(bo$ZV1aR*r%iD9Fb;3Unpgl}bYEKvawQt66c
zRGyX%vDC-JMueoj-sAZjJ^Q-Cq7~}opx4aQRI10}8p<0Rn}~PsP$(!UJ~8QNk0
zR~RgB)Gi1M4DFRP(754hQT_^3>H0mxj8gpTRVnotou}aHZG8ci;^W3EzsoAmoqAiH
z;l1rW#&@Kr3l&)Yu_}z~S6?wT1eUb^y5+x8tf)&irT;ZOVUq9y!CBRlyQcYeT~?8?
zk=!}vR8XDjTa?AvBHCbr*SS-E`{I2i!e6rGTM0`ZefmXYG00ye5l=85ogCakMb$Uh
zv-L|(NWdW=IEYfXSa&Q~^J3YWe%?#|IfzbgYg^LIje%Y_1+~nV(APJE$#R6=yUEkQ
zU{q@~<}!Ah{!w2XSNfZ`7E_in?aDm`4~y~eSMf;N!o8S7{z}~{leD`$wkpdvMJZB{
zI{TM^huir+H`mvddzSWqx4@&?#M+fi{?X;H*J!tKzo=Z*+7$RkKXNlZ{;GF!$x`BU
z7xUazVDPx|x{0O2o%Hvs`73WWrqklTwT62#lC+df=0(R>EIcp8SVNDGh>xS9$o>4w
zXYfVP3khy*Ul(4Zv&Uyv{ns)T)5#ywRk}5a5s&Pc
z1V48y56>EhR_Ik5t*1V=7q93c`>?pl)@C4G>U^$oq+V~DE7`CYZV(w=m^{o$N?Ivf
zIep&#X~EC>D%$mRe!J006K4=O6tn0cOs06
zUnf<_+cO}%8O{eM8BWLchJW$JkazL%eV}b@q>c2MCTPE*$YN^THZCh&*`1fGoYZbq
zXqp+@&OYt(;lEow=*yc8%l$pth3ZgQAX_0_@oD2vbCIV^dJWyshDK{$nGsP@QDPDj
zusk-c+dDX%En4neUFokD?N=+^EpBA-+r0O)>$WzhpU>X(e78>Fld{4mQ_WWb
zn(0S03AH+hmKd!_`oz%Lr1wJn@a)o3snzejds0HK$M!%L3IU@|s;8_C$++px^5}V77^<_YNUwC&*G65Ma2QT66lRsy8Gt
zkqYwxyVPOAR!h6%5GNXjVS(YBFEYFH4?Z!NOIcW1MPKa-y3fZr)zRHvS$VAVjgT;2
zSmkdc{mq&hULw-(m)VbNJ$g99c|7#S>V{v9*Nh&Zq50#>OJT
z!futk*{-bT36oO_07WG~3+#)fD?T@9(a4rhMe(}osL%>r0lfF?-A%Ne6)u4;vN@wZ2}Z`ULw5PhxDMHSG&N=U-1DB^)mVW1zy)?%nfms!x}C=^0a{LKC%`|TZ3?PHDCxR0Ej
zPbL@$E8W8|c&9j03oW>KX!^(ode`W!zxfQbrO~mlM9uh2bXU!|JXB|LL|k0xexz5>
z1^CsleCg=9qj#uMu^sJc;jr}Av?qzPe89Ku4;gmJ;LwW^p~K`1Bp;1*KUksK9Ir6i
z7{%>QT==SGV;aAm*?HM`l8;|kWXsJGzK36+LYM5lLLMN}lx9!LqDFhKmVS{-sw4tX
zNmLx{mj{eV5;1g#KRgauIXNrYtwy<(k1kc|PeM~a46v;*I?psbs(*GL<5UjaF+WBs
z3WLi3k4rq0&R3}zj~_Km)J#kfg5J07t}avoW~MzkzkX^{>Upi@xx?g{A*1=Hcn0O_
ztV^b%w_n?wtM%g!xf6c_C!5H9B#Cx&1-x<^xXQw5}a(7=o;<5qdt=QqVPV
zo1fyUbD9@M!T{cfY?y=smqe9bqzHBBg+!9iRKtHl@t9C)dRG*_>Y4h?(D
z^z5KB-?N6`!zbl=lWZ$4F-kJB*5bx#gUd5t^t)lyoSg3{$qMvmKeMZ!p1c0C9Y3&6
z8mzbj$d+Kd(m=zJ;U(t^*HDKwdLrLXRJZxWq^ANW=6)_N=ixq>H9yD
z(FRSS_4n_j)BRNQnT8mx=yP=*RqNl2Sa$bwvFo1Qesf#pt4DB`>9#~R@?Px9OG=_!
z4qu2_SuqHB9_cT2^Jr)oeOq)l8Cbh@-5XTDo~L|mGF9XIoP*=d?9bm`*Oxd}R$+rv
zy|o5xdNQ^;AcC|(_trN{xCzwD%F4>l@wHTucD31`FX`E$GRl=Tokcn7rdBq|xJ6ZA
z0pj8(E2PRt|4e%OD8hTJ%bczO3{VL=U7WlkVb5*NSAC}1wXxaQGtfz1s`=hmy>E(E
zogF{iAzDTPElBy$jyra@6Llc}ACo`M{5r@6q-11ggSD<)x^+AMT0#tI>3qH!ml#{G
zc^<2Rg40CA#D?bPd2=1nUR#bkpEcgUVqL(Q(}15q%1wG;(~nYStz_@VPPRWA_B|!_a_=b
zK?-m0%LAcrkIJ7m<@k`(CJYr!?CcqHmu)fK*bkPClz~&}f{u><=IlUfW2_AS?VbBi
z=awo{q+|2VxJ#K83h=xvk~j(!a^$m65}eNy0yA;p*sH!27a#axucO=O^D=^rC*isN
zQ_-NS;Lo3jbAG9qPrd47FGcQ!pr|NZIK6XgLj?@Fb)TqTez4%Btr&onHe6mXD*k8B
z4c{*>Br?);b6k~xK?B|M+!lq1%Z9b4rbg?0Wg_QK%`MwCCzdhG|7f0|({eAPcqB#d
z<^)$sWo5ekLN@R;(}}s0zg_|oPl0YugocH+v>KxTpQi;i?&1`4cE0qM;ljbQ4<=k%
z=VTXrf`aKSEiIP3cj1#sT+R+=*sU~@(|oi7mGX=_KI*6NI3{MxHTF^)XpQ(qt|^He
zI%-(m7|Jh1`#Hc$uYnzSf1Trg66}Y2A)%p$zx@08eXqh;jenOIb>7KWD=b?5&Mv)<
z=XLU}CO<7r5aYdG(G46TBGj|CkC)rY@MABOhN?hGxO?~RpOF#M-Ff55ANJw;&EiG{
z7?y(Z!;emm>%C52%n{3Mrr~SChCTE0s!M#X&wjd32;U0>agT)k>kGd?k_d+NKWmmd
z{r$V6qePh=<1diB+QJ>(&R_UaJ#Td^5f?l)l$opk?8bk$qeJULH4mI6H$OEo)?AM6lFV{ED#>?fR&j~Y}1x{Rb>pw^y@!IMlw|e?LCY_p&GSkma
z+tZkDKMS1iY*vv&-5#_v8Ven!J6RjrXLVfSs`I$tTcAGo_nUb;mK@q6H~@0#>9&b%
z?ICi>K2#DBWPkHjgW%LQ4uUWw`r!3Cc^E?ZcEkvZFGj$;J_=*jxut_R*MnOI{=GF1J`EwH3
zNR&!XV;^5#yN-97Qv?R`B=EUXftWX9XQW5S=qcM9=Rid@85TL>^blD|a7al*3)ZR6
zj<*Ap@@{=G>Qu5C{jM5ad)rJCyMZeb56}FB^OLAaHai?ECt>)Bw*~6VY%Uw<6oOue
zpNH$}dAjx4fthDDd5leZKkVKuv`0ei4G~HI+27AHb`E+b$fL@jMN{y*y{)eo0EtJ=
zv>+Sd|1`E79dkiz=>2H!6Cb$hRZMJb!|6KBi?em!PtuP-M{gZ+uBJU^5N5;>1O*nF
zgTu*=J{)i&5Dy8Nb-#j^pYIG3w*qMMD>IEk$J;Zee{)|EOYU5+vE{N3JCf#6dB8gp
z+r!V1kp<2!AKwrV5b(LMr!{PoeQ4U7l7uDqCEIcb#O0RWUM!H45m0JA)Yf{II)D52
z4gHSv(@PQp0*>CCVw6o+Nh+jF5{JUSMx&N?_?#BOHJ@p5;wRf^h)0@`#p}HD%csRSgb@%0jjyVv-eNtTVYIg5(gWbsDwli_!Z9nG;ki-T3cI7tP-Ey
zBORW*I4z^4qnn-E)fm(dxvMCHEeOlc&BMdux~0v)z<~5EoY(#~jg1jIIy!>boWSqS
zn#^Upl;(TYG%^z3pZOoJ|6lKWvES#6>{gfOt|%+x6=vN705@AP3w3Ai`^|XX;FbTa$@!SxVpE=s#fAmCh8G
zi!P)AJyg&&ShtI==3PoHe{{W%5Yo3gVFjNhSR{S6ol=6;=fWr~ER3Bz`b11C9;v^2
zxr2?8!tb84HRXx&=exckpu3mvZlH3IC7x(&aT3-e?J=o56)>A@{}$_|;?nrc!m{Nb
z(^gw@6|Of0@eE1%K0=;?y}do7iD!3a<%zuIg=cn_ll(PAFc|dkEt_3yM7zVg?%UeDl8oWJm2no^Nf(O$cSu5kv
zw}cd-#W?R15Cmr~;MBNm{!H_wFj*g#%RgS!@;stPc8RDS=p{#6Q?2FZ_!xKZ(n6IC
zBjpO2u0J>DWseFeXKH#466;3AwfAYjeUV^SXk)}eJ_I@PsfaH=m$n*pZcko@K7cYs
zT?UN-+3YIWOecAq*KSoiB(hl!^MfnI*u1OvI0GtRnY^Nnre3P_W%G?ot)UA9+1Da^t3h=%qbgi$!X9h%quO{okrW|@PSx-&Qv!RiSPvT^{poQCa#cp+=!0682lH>G
z;r7U8HIx9cWVf081lVmLTaHgWoFphc9cT`omi#*21xSzo0gtOPXn>_={ro1qDSl9M
zcpMgITSCg9QCh-gku&>m16TtCM{-IqlIZmH(}_q(O6}60y`^>CeTg(0f@P6Kt1W`O
z>0cKiirG;9JsR<_emfQF+W))}iwC5nU3>p*7<6h9fL{RZ()|7t)${yFsX!+;Tup$^
z)q2mZn~(CZ6A_#UsUp3`4`usaV2r_5qFnCPN!O{xik=M6%9`+Nbls*H%$7siUoM;D
z`c_u5P;yvlDhcx4@85dQUTO>VfdJ@bKJ{~`Vusxb80=<$K6)IIm~pf2R!CN%U7v3^
zmO1>Bwix~QnA+MKGchETAbzs>{46Q%2~;?nD?KGHF%w>
zh)O{XUyG1Jrz#%+JCB$>m)ZN=rF~>1D4cY}l9zV(zKm<|)$T$&nV=Vuudnab)wyk5
zeLWjZk&0^ac6SJ#DSM?amA_;-*tJzNoN$bs;NSq4OL(H3&1UuoM=pr2!C;l{gvGFJ
zc4MPEQz8=m&K-mKmXJhT^?VzGL#l?9thOVlD3p7Lhx#Ww=_x|K58j&eXni+}B1OGJ
z@Yl|0-ZI|ieI)RMAr5oA+sw?&01bjGDk{cr0&lf`W0v3tCaQt4aqDDt=*nnG*(5o0
zslQWk6PSP+hmP5vAuNb`ps*7|Uv$?s3ZU>QFU&Tj}|c+zuAFI$!v
z?AXyNTS{_1ms`TnqVOJlrKkmR3|ayUoh*@GaWOU!Be9n+&3S351fTZYN@La&-22xh
z5&Kj%`o8ea?k*L${fT&!LHQmJgcZ0HZOZXQj7VU;v*nV1LbXKXYLu|b#C`iB5+F?y
zAsZYib4*tYn+|r%OGU+jyADXZaG@=HE#jH#NRBK5YtFYZO&pf>q{hudFdgjvTWDx8
zY^L&H1qb>yW3Q}A5o1RkW}UxYTU
zr(Xe~W^n1O3{$uCN{yrd%_)Fbp;PPf2uyDLUw9(H-Gr2seMuZb;PvtWmVlap1^u=C
z?_b%i2tqCTsrJRXx;k`hYv^B}YHZQGuh{{x#N0eW3iAhzO
z795P8lDdrS*2t#&Mfio!iT)2-=k&t?xt;C!z$v&xrhT7PEdgdmX~IsXg->alI49=i
z<-JS1M}to<-wNPjFdbtz&D$yXYPj$Ge}JUX;RvJ0kok8DQXFN*UW8RhB^Fu
zYIm?AwbZTouv5o~)ck1Hv^xRmLI;P183Cr=n5vDZSNOcsj4fnj;sr&ZwXU|PNbzw;
z3_UYGoh%B$#g3n0>8YT92bLhOnD6XQ6i8{T_GhBOXZ-5vkt>qHpem29I(BS}qJ0V
zE9=pnfEyAL?MB9BQ5XTcE6=A_lA}5|_c7Gv&bj6wmgmpk
zob1eL*SSUUIDJa(C*B2kzvh3w
zpuonYd-(7zFf}jv3)l=t&dtq15@!D2zt-6%KWH)NaLQM0X2!Ly((_}=ZI?c}T$R3m|NhU=(4}vM
zj535t{6*5gz@bgTVtgGlv+zxZ@lg~AQz5~@iuG9kL2dUCb%^miRh52PrSVk~7s`(N
z>E5H${bdEmS}eb+g(P2|{ww`8^w%E`B8T+G)i=MUHj-s2b-ugl3DJZbg|86N50!WfVBR+7=e*uJ=CAm4FZ@C
ze!Q_vrXTYdXag4@h37A)*fqqE93cw;v`o6zWz%hcsmF9v5>SXg=t)ZDW-NP)on39=
z6F~VHpG4Q~N%B2j0cFT2E|jnjwFW2MG7&(6+9C#puA
zG(^l|oq|@d!JJHBQL>jpkK)m7I53=VDY-aZ)<*!m3;>y5I`(PdT!;T3CTPk+S0{eZ
z?Uagj1(G>-<7xmpdzG(@%_cVaVL*9lz(
znVt#zp_DbsqW#&AkeU=?H~OazT(YvVr*lDHqX(vHk25@vby>9@xA=atd7t)N{|0
#ES~P*8@ha=a5{hzD?(U
zLjCKPNFv`zNI0qQi#!$NJ=NBUH#9cRTK(hSp*GMn4$M|MZW4kQt`)awp$kP&2*@H=
z7IY;z&S4Y+SgyyRe+x#0k(Fli>zLm*4`7V6KxdCIE*>Mc&89t}}^!lCw*l$E9B
z=fBHVru~@U4qV^<@eDl?_a77;9UZwKda=qlllNRKEGUvLIvY?lfhXq)<+FQ1(13?R
zFfM6F#Oxit0zVO;4=~&|CaWm}L|*L9KYFZN=btT#KVR^1i{k`}UcL{MJa(|L*TR-F
zhl{q~~xmg)0ri7OC6V(uj7
ztMz;Am}EZYsb}q1@nwdQT#bmap7i4OtT4&g5D#?(=Z%5E!O+Ys!5GTJ
zJPk$zuwLN-S(k19z(C;fZif<&{5zug1=8R@%z8gNqkVamNe3|kOIBJcF_MA$+uMta
zEf5Dgg=UOw;h?Pi4*TJcqbaGaP1Y69oc=14H$`AwervM&bTy~q;_6%wl-ofkH<4hK
zugSBU8ymUW)vRb(WV1VS&G{I1pz}tii=u(uc>MC)Cm=LxT?Z00vM#AE1s*vTWH(hY1jO^lDTf`HO+edUP+
z{?J}yiS4u2#us-JT;r;D0u-wnuLU~+E2|c-k6Z_a5*Eui5Q>C;Jx!Plf*4}S_wS0X
z{FJ8kc!=xsOLkHTPP)i?qljzl6kEGaKilx`^2>!
zGwL4PiTb)aZl4Qope#QpCXxYX8jEP;)xP>0bmE{CD;<3Ol!*xo+&Kgw`TzjO05+T3
z+Qg9`-0;Z>{~{QdMx)#eH>uaP%~iFa#ez4%YL%ypM^A302Skvu@4u4Uu4mjc#Jsws6wUGE-fsD7~qCG?Dcf
zQQ=wcg!M5o_<@0eNbL{etHvg>(2X907!gGi)L*?{#qYWF5`Y9OE7HU&t!A(N7jWmk
z0bn%5j~_pbw1na1%R6&^VU$82-XeS16HLu}k>4TfVnNTQY>eex;V2?<17fk+N?A{Z2X1k~y7n5IvSJKLg+2mvjq)48CG`84d*`CZ*Y)r>X
z)!<+rF_slKDOAbNrp-Jrr>)tusTcfh=~sXwyM;bACD_T$YL(ujU~z0K(1O+FBB^J#2nv$V2G)>PXakaA^FnW2$w%
zF7i-7A|A$7IyW{pHe`hM>f-dt>&kB`$#~TG0v-n+T{gAAW6%JsVm3AY`RuT3N+5V`hvYtpcTn
zPt`h*BIcReHzMc@8D{+%WU%?mTxk6zC;Bsk%%|(9nwq5If5J{5t8)*6L(yUinn;6G
z3|+o1^dVYm>h}Qba|NDLHQa{nX0W%oOUC<2WkG+TO^-;xqb?AvDkC&NCF!C;0~MC)
z5H-1gyoW^~=l_drsf?U2L<$iJ`S5euO#U1lJ;|JKBB~=Fx#(gm_oXT=1$N45JN*+7
zCHg(@(2^1i<(%F(Ki_U6*O59|lB=mvW(xtqUH}MLwf+BJEQ|K__Ud2`zxNq?a}yHh
zyWf{Far34s&B##@SebNZ2Rw!H#eF}HL2FU`>H?J=+Br3!6ju|Z`ZW(7#l&s_WLvd4
zSn5e09UIes^9CV2l&@;<>Z%Yoo0*ZJ(dZ)p6x;yDCDQO~akM&tazLMXV{F`WaoP*Q
zzmzkvEFkPpcjZ*iiDcsGwq@1$OcpqP-d2*$9
zaGNhMZ$s3ZfJvL^t$gYR;Io?$o+}hDVDq`R=&FfK7Z*9<*MYqV1QKGayv?K)<>hdD
zL=Jg@*0}*{06ySwTshgRJb|tUY=t5FNrF~NTb|%{-ZtPfS@22-zzIl
zD~-Me?;urUZP1n?8+{YXD@Z!GOOX+0B%PAxYC1nVQv%4ZFm47p0HBn`@os|J7z1Gf
z)qEAHVbYl&_J%d2J%D>4R^FY;7yn2mPBCBQ2}E?7;oOc@S|?l7fBEvz=jzlHfgHjB
z9454Y&vjD+U0FpxV
zHaE#Y%>$O|+rysZje#|95Lnc$o4!hv82-@qB_tul#ceuRje)5X3}A!lZ<>CoA1~}0qDl)e^h>D5=^MGo9amsdf
zv>B0Os?;h9iLv2AK6@v!C!IPVTxOlBuT{{Tw)loQUryF82p1`otc|I9%JZw
z*#j+rsK-au0=17HKSHTTwR3UF)GUibHV!I5nVMD;xXrL-yT&VoAd3|YDk5-XO=uA!
z1?H;Qy1~p(*bf-9vrPbRfOOxP@zMqkVIUIOAsg#SUwJ-SV`Zm%M$K0*Cxu{zbO`^{
zx)MS`)bIQ3yX?F&R8S1XRje{y6~iaE8S9tn&=4r+p{FNqgJq_sFvcJzER29y=dhC`
z2$T=BVrK(RwEv&EiH_JVGOnuo-(FX}LV`hmtqu;isLj*UdI&R2Ltumat!?9GN1c1`
z==it=H!GFDBBj(0^+Z?dL^J^WXs%-N&&|~Nt$q+Js&XG+lO-O$I
z5lB8%dU-fwEa2!PTY1!zMUuJ>H4%~bfP%O;C#p<_zpdu~gX0ia3TP3HiO)S
z#HM1y8Q?~VzoMM)(uA*&P6gpZ$UDwKdTyl9d%xP{0Pp%%Jf-;ma4{vtvpI
z(LLDK80BmFbNQc9VM-zwJgp>T&`~sq6BJyPif@MPy?062#gG^5MGdnkAeR5k=%NA`
z*62>(^)r4Ug{?
z1LEw=Lrl!y(7+%cbRZBPVkqhqLjjk?F*;`HH}fJ57-*Jig<5?_o1V1QV9e5eq^LWY
zN6s%*+4c%uPa?ipj!?jm5v7bw-mS*~J72=BpLcS02DEhl-Mbga%$Rt164EjN3Bha#
z=Mo9!65U_G&Q?j`JtPFoE_UeA63mh!==}r^0IYhH`7V9ytV9
zX3C8Gno!P=y%T3%K#N1Hv1kl6qc#pKJ-sZ~vOWk)E%oY1=E&=CyKa$8)f|3;Q;(zh
z5tjVr5;D;?<}L_y`^OIklrJ_@XMYP{>md)7{|OP?w9szn|D9s<)$JWFBy~UPmKmhN
zp!)j|hC&CajB1MK5{8ET8tvw80Agq#7{Ch#U;o;%Qde3-V8ga#M
z8Z6EMG=eV_6(7PX{5zhhLV+{G6UU%Q#nm4YS>t=nBy?Su0=X}c0bgop5TT%40p%ct
z-h;Th*ndtz5y`^JN)0&}&;|A&v!ojD4zmv(;Cw=QiH4s)3^^fLpsojAbwQ_S*LOf1
zdn0y#)m}KQ?C9jLlk@+8
zT~ojy@U+T$!UP8N&@nNyOnS!u7SB9y>W13q(9OA0X;YIZcLMUINS!CJ~JO=I-YAebcRvhP}5peWDSwl0`GN_kdIPeO*DE#R;%zWL83-Dw_bxZ|k)ET?*JI!3xZqs-p&e&I
zPK)5N;`1>Nq)-K;va(6+m!a)DS(CM{@uXZuC2sZuorc@fIw1Lw08q|%h9e2{%ugFH
zE0eXWyT|H1(HgI6;vk=0qTlQRxo-I?n|d&9vh*6aHbzS#VaQpjNE;QZ6C)N@GrODr
zT4vlW-~A;yQ01W(A-;eI^rN?;c)BdLge>SeCHEv%yqTz+N;^n|Zw*rqq;H`cet*+6Z?p}65duXCxkRHS3tYTT
zCro2<$q>Geg2G2IQSW_D2W>q=_}zPcH@h?n0gr{{#G?y+i6+x$j@)o7aMyYwDvc>DFJy*o-?t}2%o6tm*+Ef#TWks
zpsPd+^7^O*^iR}lBuCK4q
z-Me8>;NOgX$A!{Er$8YA1RrK8?%jW;29rp`PsVM*A%Q_$zSccxEa~81cgE@bu#4OK
z_w2@(oE!;|t*YJkJxX=G%*-CMu#_zRHpY6Js^f!2Hj4SSxR?t(=i5KKyTg9`(5S2X
ztW@KqX)#h1nmj>`1p5*4N3{lpaknUmh!U(Guz?kIpuTbE~tF|`Xr-Q!x`s)u*)He3i<-@IAxWC*^&
zVsEPNXtg67sAE~CeOu)=sh@}NKirperGz#L
zIhj(|ZC&6AZWkvku!>+EB|5MBw2hB{3cR;&cVI@`-4p&;je#;ELTGk={#TuQ8H@wb
zgQQWQ{`w@*PcJYt}rz}A4_76eF;zsAQ2935#K7XP%vzwz*^?b1>w_$ovyfEG_0)HZESqVH|S2CfCi9j)NWvH
zE&Z9l(%-)>Lo5W3-*@W;jJ<&U`uEr1KCVuU6H>s0Lkp$#NDHlz{PUJ1sSy`LZLGkE^Na)nc!8>tK%H==g)P6+zINF)%SD^4MEugzjR7
z)@f+aw#I$@n848X@uTF$WivWP2(4bj`)gljG+y2}w%6nZYK%EzA$M9@pM-?TE`q@a
z2zgL!YtEBXI@%U1OWfg?gRW~63iG6+H7(^#GjzdnKGroKd_cA
znj1GoRn9Td(dqOWZakU)@Zkd;1A}2BW+gd&}N#<$sIY%p~g#70F&$_(?6HqiTV8p*|VBpT5_TRs6`(8h9xU75zrO|M#
zG>wJzx76IPmmN_w@oW3Myfr7Ms2PfjvbYLryVa>E6|^Vt^niusWeLcE0FeQ5<@Z^@
z6Z+Bn&z~uvH$m71Ie&K>i%b$G@3~zzh+xuRMS^?&CY)awT~eCDF*c2lT<%MI1+)eb
zR%sLrxa%7jfZO9c_+1|#(s(_Tz8f+w_pW}vs|Tq7?Ja~K9asMNfC`c$tp@!N#L&7<
z8#OjAuJ)D||Jhlvm>u>PL(pEmib7Iz8X6jE>R%9QZtCk}*5s#*kOT+(b3%fdT9LL$
zn&3?Y_+zx7zN9OpT?r4!IiBv_mprOAX#KFi_pfo{=B$z;gn=
z7DP>jhrRCzK0BM?#DImTW9
z0TNs$``vk-BHjA#y?;{2%9~^Rp81=*TNL48xym_$q+E32Uce6aK|Fv3yh9s)M^jTX
zv!Fo6$mo`65MB5d1^0LLp#~TyC^z3{gxv{SeI7EH7pE}|$}sD$qC%^tHcT8Lk(~Re
zX}LQQpPrt6sW&wg5HO4lW`JA@0c-GGBmDes0+4QRZoa9fr}uC*n@?(I4I~v}u(998
z$J2v|{=$zef!7&s{u(LM1HGUi>}|*OVIfeN?V+dx?(F&b4hyXP7$91_PAlEWh$qAc
zVVVY5lN?O*e6QL@2yc!%qi9q+V6?T-`3D5(>+45R<31(0lajJj0%?}LWoF4YUxa3S
zuWuq!d7X_r@d$s%7Q6*Rr9FZ?-KcYbjL*d%bWBKGCT=7-0x}1V_G4|_3GfLh{!B1m
z(gK4<@R$bG0k<5G{=ANLxPbUT@aGn|Gvk$)ad2e!4)W08{hCHc7ybNSCxiHeOwvG>
zqyxIVy?sgk5AX+GcEuHrfeZv5J4`{2@O!T{H3uETOa>N+Psj{E!4su0=-6{xKCG>!
z$77?T1JL>7m|Y>nj0Q2mCq<1nk-G?zF`|cS98f(Uf<10B_lpMxYKUM8f=uuN4d&C9
zh6{an4!lJAGg=?()G&Y>19N-wzhe0xHpw1wExk`#0_?B6&wCv<5d2X7WRB7*g=+gW5LAJ}e~awlX}F92oE?3{BxVZVB
zERhmL>ldv8QB2|XP4tq_(
z>TANW{Z$_nNGCNw4N7&{j1Fvv@+bu=|6`Th?2qfO1o>5#>V0R%%C;aJKUh<7Nf0(kFzNr}``&mkYJAwR-vIFs09?wkMur5)`^<&M$f(Ko
zbSfg>!l5}RQXgCi(01+ugIYM+OeW#aSJ*E~N`coUah#vD=xpbA=5<}3yTYS`rX@ze
zR9lJ^Ix=2eyeeY-3@?J4(g`j$1S923rXkzEI*`Q&p}5Rfnb{0FNlr>-zSlMfhxd?S
zVjytF5Ga7Y_r}0Lq(IFY+|f^CL7&V)Z)Bc@;TRxM5i}CfcXS{mlcQSjRR|B#EOmo}
z1#I-c8tgcD1|L{GqaIp(xUL>^6WpJQwQVsi$bfvZ5)5-*SAz#Y1vNAX!oIsWS&&2y
zb%8t*7pDe5-r?#(6Kt_C2!d7F>NWrRl?3#a1)0yjcTc%c@;Ds^&+V4NJU{#chd_ZP
z+*M{1vi>d(r6bZRq$JGH;=;Sm*
z1tYhj<>h1Uyt{bjY9j;oJv`ZjpIP_hBcJPDhRU{udYuAD9*#={KBvZ*)sC;{VOAqI
z*DQ&{vJ;*g^zLiFzw7;3$hNLPVi_P4OuyKJIRj6lk_1enqRQL@0WF6`Zlq8E99W}t
zksd)tPM)h-#`#G4;&sj(MB^c_-pG_!tt}H_C3duOh3G0;;D=M&r2;Vvo(fr?nUpJU04WQb+~0rW#yl})
zxy~%r1d#o|*OKWctB*gj0LysNDQMHH#MdMC*l_C$=8Jea!SBamza^cG6k|PKn*uuD
z@5ZRO?L_ZQjD}LpFHQlhHEz#kL@(}}nJI=#-HIta3D6FK$K*U(S$3^2bUy>bMeyhw
zS)VaA4zu$rwzjrewQRd~F|o5hF?xqd+8$#!DC=5`>G%EpIVB~d@gMCL9O@6K0ZoyV
ztPN(C&CbkxUtX@atsr@;-8boSUn@}#`Q*uQXUMTi5F^A%LJMTpBcl!FZ%hquY4fnU
zYpa1;m~j7Km!}ZMGDEE6S66KUWa9Rsv%#I_T9k!9NCBG-ynH3Ct=_S*ja5oThq<`!
zLhIU8A%#6bWV?2pj2uaWcNeU!k6=hpPuEpJwGJoL>fF@dEVs5Z7Msix=81qQ}{A`!`l`Eg%fezLT{>#>Zy&9ktvBRO^I)mm&
z{nE^d{m$6_28V|WXT_sqJoUXP7J^I5%fTG3XcbpH3+8W*ckhn94A$hXPdNh%i&9Ob
zJP8*+YLhcLJ
z5K7*HbQ+)^+=2-hZ8623S
zH68#PIwt85{w9q!`mI}4o{ncTDUR!CTp84Znvn+46+5?W>)j>Z$Oye3Q1R2liQ#{L
zrSQ-yI+cUOUk}u4X{Ie%V~m>sD}jz~a$Tf*_Uzf0Tsskz2Wf2AfC(QW>`67(mJuzZ
zELI=jz&MxbaEQ^47=65gdDDktaTRlQg%?d1VBR20xwpl4d^~=SHJsd&>i5Pi?;znu
zg#!anulJ?=*Xp^p7tks161przg3l=JVvl@(rRD>Om`A=T4i0fA6#|5VJHkUNj>Bdz
zwxeed_WY=E?EMl)7IU4s?rRk^oLt=$_v*nPGrT!uU7HE7Bt)L8-)k-H*?3X2Odf2u
zY5fKt=m`~{$qQ;Kf^2h>wfh0x2E)IF*X-lH%>#t5l`{gWAK~@uTe5dAl%IyMF5jt0
zAf|nF^RD>XHMpqiCwzS;oNdV@Y%QhLHr)M{?hCte0OwNy5M2?!d)D02(wAZ}^P(Y6
z^VA&jc8ojc`)VC6iOC-O{Z8+F^Dn&h9zMVHL|oULk*Qd(=I|Af(2#{E5~C9Azlor)_?DrxKAMG
zV7~D4#17XaG70Hgp=p0*65;Dc*MX!rxg1MT5)leWpem3Dd@L&yc6h!YEkAafhUXX5
zeVMDir9Cw8)#9FGlXJg+I5e;J-l{h5Id(X$kKnbAO--%Bmw@xE0%d^kYgOy~W7b-J
z!W2dNZwgj3Qm_~#Q(v_9X1%!CN{QR5HRI>;=l%h!DPO7$5APB%url}`@nv!8N#_K58b
z4J+Gz5N16rqL{$_~hW__+#K6b#9@i^If^vE&c*)PL2Z2R{9@R|0P3-q0-|Dt+
z>}+n1z>WJ$@t$9AuhGPB7hiOMNN=G|qm{E?USMWDW_=dn3Ypne;VoIPdF#F?FAvv`QWzcIW{jkJUUd)~&(T>QO4wpBDk$4X1qDxEq7)fv6x@_SXM#|&Ht5-L-Q;!lI6YTx{2@?W+(YQ;5$kWZ
z!%d6c5I#=)PC~|jpK*z|O~uSufB%D!5EW`ygFlC!W>*6}dCP#$p7lpivYr4K{*qC+*W
zEdXH72vm^DZ>6uVU*Xa}Htb(?@-@Wsc9TEIDC3t+o`_T*%YJk1JX%R&w1E#T9whm<
zRvyi+Mb*of_kR8Q`l6m5)AlDX%e_U^ss%;I3hy{2Y=D@tA50xSqCU&?=jye^jVvsA
z$T?+>{rG|NIBrhGFOh{z%Z|ma2Gk%vwZ!Mc?~v*Ng2D#?;kwPNDJ}lc*>w-Zga#iB
zJ%MlcL?A5d8=A5=P>`wk_7f6P#v~^|dQ7|KukC92^l5{DfLgTF^kAJ2{?SVqXZ`%H
ztDy_BwwVOR&jBU4X+nRdqu+p=nwGe`fwv-K|xTZA1mvRSi3$m0iE7>R(*GE#&pCb0llN}$G@9g2|ffN
zNgosh0tSVwUZ-3=XoI$OFD9K`UebRae;x@1gB
zu!cV$-dO0l6$(rd!dwOTu!-8#4UsSGU5q5Wv9Ll1-JJYeNJK>WYYjy++f}=s8Pl)!
zP+2oYUHeCEj2uuEjV^WkCKSH)(?5@&rf9%48P&^U#Kxuvf%pr{T*SBdm}N>%asB(T!*w&Xz>xl75^gVae~3a)qd2;ogM%Q>
z{|z-NIsP^=GfQUG3>PsM2bprghAbd(w1axbARPSyq10&D`dRIGD`Cz=k%JfQkk=|b
zt=R~DDnm;JCU|EoE8uB~9?Ra;N=u-X?Qf^)qodEm*yWP7$R#9DIgPN8iZXvyY9DIiE?yl&t(Md;|3
z5eT}?i^|!WW%>e;2{x&lu4L^}J*+42-pUve2)O|suwPEjRTdIaB!uDK!UnuXQPF6)QHL6QpHJN^%YBihsow_%Cd4AqR|K6*)UQ5=J$M_pQoIHa
zweq*OkIy7n_#j9_nr6+7jZ>8_!{c58gcDd@>g&-#F<(GfeegVh3+%7bR14W(F2cgHl@3-h4;^sp&H3Wza}P{+!sn|*$kFvRextodRN-BG7@UD2r4Y0l1%
zAu2bu6W*fJB<>3K49eB*7Y?AU&b&xyUhWGhBBTip*1IEDLn(_>Galz=e?{#)HsXpT
zEj**@u(z;QGxXYjVsK{#vjX0cdGn4MZeihV2(iHacOITl1OX6XA%cqI6r$Dx2yZW#
zrzatUdSTT*<=Bv$8j+p_KAbCJ>4(+hmU4lOlFV?^04mx85K8Q1KIKOwi_8W`pqYKR
zJ(J~oT!&h+4?|cd;j#31WtkQp8M(a_*0#=f`9h*9E@h|S1M*m2Z+P`8ofJeUnCHtH
z8^I=v6%km(wshJdJQ2;6?yJ^q!x*q%j#o
zyKbE@62S#RyWQOi=KBKNA?;~!IAGy|tV7%g(-QA4P?lSEw}(Ras2VF(Z1S@@h8%%c
zSvAYV0-GCT@HGN|50V=_e`T5&9trR^T=PAoC`N6stD%b~h>t7viR%z6ucMib
zP5!0RnBg(&i)XO?A^HLQ#`-ODl5!HrZRj6u+kl4g!J|k02pqu{wh^#F#b6y16kfHr
z*JVHO_c!{Qwl2+}C>*d6!nOe?#cOAHP>n6oXnS}RI=hm~D=Z}OOXAQk5hioj*-j{5
zeKCEq2u9w)KET7194Rrs0|m!yZX8OlWhY;%9CwjGg%Z||bR~8Na5a9HTNYE=#g$V%>uC5#p
zG8bJtGHBMWEhw(9_XX|%`0oUezC*Ct;C%@y@mYXRx5o?Y{
zMsvDelFQopj@j>heNmQj(qwXxMw?w-d>$|-Cdg}(l9=IjIA_-mA$ZT{U9;P(szMHC
zh*MlZ{XKlbuc<3~&=4C#uEUUw@7l4SKOd#r?A~iZZ9_e!>F(=G2)9v%PI2?Yc5pPo
z&=_5yYFIQmJw2%X`}KT10qyFvaWGYFHb~T(&cB%fsVRUZG8k3r8o27uht^qMx6q`~
z%e;IHEeke@-*`F32kT^qo2(wkQ(K_VA|SV#8f6;7{zr6g5K~q3@);CLgz$572R&dq
z(>FB#1cCT=m=iTmXhJcG#^s!*jZH;R49ON`C4PWv%&+&F9b-vPPst+jjufrV;@ZM7
zmtR0yzJUrfq^o|kmF)Rwm7+BAx
zB8~6dv;C)@uI}&Nd-Q}uST{NH!^s>AJ@wRX?dPh7P5X;r?5v=Ep2pMz5aL3r)}^;j
zW7Qzhz&aT>zZeDX1x&xvsmn3Rc_Ij9#M|-bF*vhI0{7Fl6wL3C_uCm^y^49^pXVCg#0GOO|G%1e!doBY|1b2i<87`V`q4X9wE`=URe(ncpjhE_5{fxUJmvGO`R
zPy@wxT;a%Lgz*&wiI}^%Ma>cjLN~YQ)E`Tpv+scEDc|;tA*|@-p6ABF1^{aIMwHol
zDX>hoj+a)N&VAVXG4zSPXx=NT*E@rW&-L~EIzzRyso^$luTSVSy=Amih60Q=)$bOAfEhdCW0gZNEXuJcm2^07Ruk>`AET$Ew
zkba^KfyNKzvaddA67X1owoZ*Ya}nx3Xf&E}-0D54y0em?kP%om!ga9-Iuq1XYUHRO;6xqRGJrWO^*~VC>m3vxvaI
zI-KS~gpUB56M=p)^YuBG)^emzke6Ti$vq4Tl~YN*-G$#3q*TN+>%m4s20|EG4v&m->S8HBBc1^*0MFJl+gQMvVtm&omtwZOIPSf42aUfHVjxm`u3A(-C5b
zh3SJh7y(a!7P`P{SkEBj;X!dcu-zB+>fK>n52TBbk(c0Tioup_dX?~_=J0D>eCNWA
zhio)6vW|#j#A5PFtj~lV{*1Qv4(QmS!p}y;7Q!o;a4-yBL`0FEn>dxERZJfXJP_13
zvJj%bd{NabCF8Y*Gme-x-nq`6Ie(bXhCp$=w_3ru#P-->jgZw>H^`iBMO~64G;jko
ze4DGF2QY^6>>j>2n$s)oxjMdM6l`1+t?g&iIdHEI{GBU=9+2?1;I{Sv2HPp&JZQcu
z*!FLE0ktaxuQlQ>_Ekr`>BZcLjw=cs7qSv+U`V6^lxtqP={Wih;&=!UA!<2#U@5Q)
zG!sq@Cu%l*XXj}y5ncU#o_YzmUNvH2K~sU#QEX`UzPr}=>dH;-b=fB=*Pjx{I~>k#
z6T?1x{F;8McgW(@P$dL*$y9?t4)?2Uuu*!!H*w*$^-a;!A&%Vtj|@oVXE6IXpcRa4
zrwlPlw@T@kEg7ER?TE;wVBX!({6v^md1x6swA|r}{=9EA%_FR%3^+9q^9Z1E5Hcx%
zJiGEW%`|z`fN4dUjeN6rqN!tcI}3|EwkqBLcICWpvjy&KAK;iHk`hS&gcwV*`YUY+
zjMBMyF)_FEQ2k5}x0Ax#md3`TlRva7Zfslut$}
z+tR{uy&IZzC}V&0mQGSlK|+`yW|JEldzeKmG+8wMNqIofDdJDPcvOb%K;ET#D08PPokpq1gbM9Env9ml8*lIN|lb)>(j<;aOy&
z-KVIhWMLZ%xY~TgjhB2TMs{9H5nvHAh6Ak&Epg2N4nVk*a2+0d_lipvy
zqOCccoSa~hTO&nLOG3`fK8yr=I)05ES&e;o8yaN@WBzFWHw3QFnd(N3)9l~5&q99?
zG~AYS@K!RqOH8IeGCq&zW%E2ZzaNkPpWqHhN)qig3yZL{85?J3XJ6gYT{>q#(J;iE
zIkfTrA0tRLoI_Z{L@jxhgx+aCHqqs}PE24M_6eV}DK!I)RtsYo4%CstFns3B8Jru1
zVp|UQ9UCxVt42^=XJ_Y@UBX5mDTOdZ4@IF0{vl8Wk6fny0T
zbjCjx)m!xeC)>urKqTFvV4Hrt&$#`d)>+v)%ap0sxhNuk1KC>ku;iHp`1|@PW<-T3
zaRb4n=>{l{O&IvV8kM`$oO&x`COV~rqh?3jQrh2K
ziy58<=O!$2Iz2S-So9HN@9VHz;-@GDPGix~F#HJPNd9`#W$HQj=-^6>)yWGh@W?)a
zTNK|;$<)?1GYgYJ<=os{z*Ui|t#zqZED{qLey=8;Ur5j}b%r<@G;rmJP>!yAs~B|p
zV9_SwpS9fkVPxO_QMd7KX&D(AK<3$0S}tM1{WjL7rbx?TAW2J0Kf&OVUDD`+d23;C
zEIB_vpIVO7WBzmK$({~)NXA!9StJ6F#MFCWQz4ENz!6fWY7Q;3@~zJQF{e)ZsJp%O
zdzJl96wuUYTPUsq8A6vatqY&jNDm1-wp-{j2|8jU#3n6GHJrzm#+P04c?PLKLDkGy
zknslQw-$W4#PzX0UliO5X7w$A?}?%Tanv*5cU*|EL7dsh9&F4FDO8p
zB`m6C^E9x_^iN(8buIDIBVcO)2f2K&jUlTWH)-v%=z~qU4r=s~(NUN_!Jx4~Qnkk3
ze)6pUvBgTj?^%JvFHfgf{XrlH-kQwsJ_m^hyC$bJKJrM`Fl^f2vX*3l(=Z=0_-tnkawWh>mgTfnn8?TUvKYiFa;&hX9%Ko-a7C>7l&yR3phcX#<>f+
z7)%>)<5_m>D2JYWVto8^$6H6@3*bmOkGURN5MJYT%L|K(#)gJ8?;${jtL6+vgUWMw
zVrojmW)eYdSQ=aKuYqQi<9}H5y$IPMwZC93A9Cuj695_Y+sMR|=&HtmBfY|I*hC|><+)NeLhBRA*Lr1A-Kgzoq~mS
z_wHSxD_`~!ABV;PDXT|d!J_R(j1O_siD(kcL<_{Jd-&y(ySkn-w5lNUJPbB2UChyp
z$S%t^uk}jYJ)dtu?kcz$Y6EQ?c@xdK|$7l^`{Ozbma9`7U
zMMvw0_eZ6tr9tks7osWvbba035NBhX3PSkKBrch_nfPmJ49*N=gmNx#Jb;_F6&xEL9v>2z&?==M
z+<4r%&9t<%=vhipvc-^zhocG=W_0kqk$!_tgpcam
z4+a53LF8&~!AH|eVM_wzs8LjGQL}*B(C2lZ#~q@|hbajNnS@kz7VPtD<6oQ4Nq9
zand7h!QH=T>2Ok54IulQFy^3Gz?|Yv^o(emFsN@u9NHwJ(+v1Do-?4Mj05e$HWp!M;uLtRPIa!
z-lI$rg_J0VSg3fvc^@{Q_r)^nh7$#e)-*V;aVHe_=Ga5wIl?7&3IU=wv0J}x9pn%+
zkgNgn-vO&j9Ig@PGq55&3_6z@bb(0W2^>}{!Jps_;(o*`Y6KyKRQ*;jIAY40nwX7^
zjft}fVfJgBdABJrq~a2hV2{%OPZ!wo_Y?Nclw3Q
H<=g)U)ZJ*y

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_images/chapter4_57_4.png b/doc/LectureNotes/_build/html/_images/chapter4_57_4.png
new file mode 100644
index 0000000000000000000000000000000000000000..e242a4058461cc1324ae0ed72eb50b2a4ab1acbb
GIT binary patch
literal 30569
zcmZs@1yEgEv@E)b<3WQH+}%C6LvVL@cTI2)4#9$l;1Jw`LvRo7?(XpCKKI}E>R0_b
zmEsgdt-aQobN1-b-D8J+mJ>&W!-0c9Ac&F@B8m_Q%nkTw`tA+*%@>hS8SsP4SybIw
z$UKy7|6^e*a@;=duO0D1?ONDKZ7LQ8XkZ
z!n@0DZOlG*&Mxz(r>B3v5eX;6l8Bq`wg&du7>DX^Sdqtvjqu+`)-8EPFq{1SIiU61
zT=4$tkS+4ymWzkS?3UeZI9;;E^U{2#ETeI-yF18Xy~n`HDy_(dlZz{RVz-L5%w>NX
z1s%Ps-eG;xhKGymbWEJks3k^;ovI)5tUnlCo8@wgs9
z^gDwQe_Kw4BMUsd_g@Cfl5HdKccVYvV?Bxki%}nfMlSP_hzQ0Xo5j@rbWIeKUc1j9
z=H&8aQnpwW4jz|d;O0QGQlT6^x9dUQSgv?jSXj1r3<3125g!=hQd0W5!q9eS
z$_dL`@2LI2b|AU7V@~~7Y1B&+qX6z3i^Fz7$w;T!{Zu&1>1?APdjHB-(e+TOShW;#ZW8kB=X1F?g+RcYvNu)CV%+zhTIE+4*jrP6b*4k#nJgzM
zAz%~s?w7a8JwH7{pTFMs@y_V^F>o|;TKJuixy3|*Y->oLSEPTXIloH#*6Ih
z;qPqwl@8#Kep}Bfa(UlW+zh(^Ehfy}enFNxWY61OIE5!KgR-al1MsB!os+S9k9|b3?L{e-8@@mtBS_RYpo(Jw>s`Qe_kD=+
zM1OyXr~>-tT6Y98e)(IMol(J5E~hOtBRDuX#pdSb$*a}Qki;Z9t*7lvHmhlHQEaCR
zvmFokKpZhyu16Tq?w^*1Yinz(T&9ifFO^8`usxiyy;N3ON>|=;g#*r?8F-!P0)G4x
z$A{CUxLUQ=L~Lvr{v_hjqdyf$WXRF>r%Q9qMl$_9)SFBOl1lX2o+{@R2~ui1%vqWB
z|3Ex1b_CmeA2WHT3uM!BCUz;c>lx|k>0iBl`|@*zz92aL;vyoi5^2;-$MYmpp9V*Y
z!NGzc;IM`oc88yy4KP%@9%@3$v>SxF!!gG8FAtVlaNoXti^FF5%0DIs83hF;o!@6}
zjS23Qc7xtCwmgwu2gkpnf>A1o_Eu=#(b-w~d3kP5(?$!RR%Ygt_B^lS#yfx*h`8+5
zKaBh1zzO^g-lp;0qslMkZ1qaRgM|h(a9IB)(uC-iQxFq>{r#J@Kt3xvBLg4e;NYNK
zq_`1kCyITq7F{nd^j=i?=O
zaLFZ~>qYS%)dHE+kTp+~*Kp|WKNm~2IF@!xKsL0?4-0UxB^>zcl@vJwG{@srNX
z&`?xLN(yb5aA*lJEC4Q*V%2=>*(y9vd;eCSd;W_#aYCOi=vk<-oT{j-B%-5}^`WDu
zKWezX{b`d(sjwEipULI)MqXb2a)X3~WIgqI27KgniH7jz-^AUy8XA2$i8ldXo*$3>
zCQCHX{4Xw?6%`eq&V5^S<3Tt0WdJ?ri!41F-cfhObP!nU%q0v8GxGxt6)qXAC>>X9M?k%+kX`IMU8SiOT0
zWMOf!vaW6$Y$3?OViU)NgY)5BqFw8)7=H9y^Vur0v9YneHZ4U()R%9Ol-m
z!Ux|nbe*pEg1Xmf&T7bk5Nz3a~#E%CxxvQg(+B`u6Qk6k3eu!2+B>qoSgU#AXTZ>yxmY{3+<|
z&08p+WxCN98yFM>D!-q;mDNCBEXnR{HO2G&@-wIB1!W?&N=JM97o8^8u|j#mCBJ7M
z==#+y`$~U#@!B0Azec6u>-k!nTwWZkZ;X+dSxCRJ;mV&c$b4Hj;D}T??x<70>EKY$)!Ia~wYA;pK+bh8K*NnwnV^z`)5@Nmz)dvwuQ>>R)YZ9{sP
zTzhO~8V*wl_LZ%R2?Xtu^IS-Ykzx5=Y4o%ectJ-?jt69lql6Ox|RQd1{Y!Fz*=5g`wZTW
zqCcf?aq^=~%F|_d$Lx)hh*?T}69Eklx^-AfVki7DOPzfM$G3#^s|yU#0@AMT)0yUD
zuG)D*|8TZnEE@sWSzcbVNh|SjaT1b}?ce|ao*0eKBL>cJXjm8&<4P;Y0{iram&4Z%
zq9h>B-0m%VbTnVGs3dIkilMhH}j%aqpIb@Eq!l656k|BcSrmX3H%$Hmu&DVqP?Bh
zkX;u~OAqB-Resr{Gp|#_xVSibdq%x+Sa^7&&R{lBPY(p}b&+FU#CYB-^;hu
z?Z+)V=LR;3}>kh@S`#dX4PZC~&f{WWpk>%YtRh&Fq@4z!X
zJ&jJK7?>@*{CE6O;_CI$b9Hr1=RyMql0m~*4|Fqqru$dRq+C7j0h#{0P&r|e2Z36@E|+Mgq!1=zRhUm`&x)(B^*^p?NGfy4!A@rZ&+oQW(}*4-J%LsytYO8x~1dOLn1}
z`GQG?YlW5ACk={_kc0%I%id)7$gx**GquTpFIhW$tStRvH2I3=D}CRkCiww4#Z(2O
z@SKXZzueJ58Oq5ubce$C%yev#QX8)UL(8>a?SLR)Grus~{(5mR`_^jzK)Q*Uol=CI
zkqU0WZJF@NMxzehB%FUamw|ZSy1p{8r9EWQbz%*1N3%6I-92nO>D&7ep5tcOTrrn=
zCyI9vP{fy8%T25t--4#lID+IuN4?L?Lvuw`P39B+FLS)uf4=isrFCvahsojY6LpGi
zl1FO|hl~625tvcxwbrmyRQ-4YzRDBsr0wtq2D{i|s>cmwwewRnK_SI0!6;!q?I9Xo
zKGkz&N4h~@RPJ(S`sH)nU14Eio1dTDJuf|je*E}kIelSzf7Rl=Kdm%Tzk$>I9#X>3GAw+EV
zMt!I@7D^$uMPe;i;f1dy&FoFwv7+U_rXlb4&QV>cq|9x8Yh-3q=wM7-*Wi7N3aaFr
zA7W9OCE5*{o|iZqp}O}08(-t!$7EB2MbjCG+bEma-C3T7x1za?pXey9|I{jQ&VpTl
z0D+rrVCwOrlw*&^5hr(y@9GlazuV#a&Q?6q6SY6J8HLL+EhG%$^GyNLzlN-`c$p!ria=M|1XE#(K3AO^56+^A}%X6Jn
z=KCcYCB)aDd-({!Yh$`JOfHj`x{#_Uc
z7KH?l(<^x;R0Lv;sOoI9SL7hF=J4cW^XPC;dWfxLMJb9OVXOpR9q2}=L3gnV9u)`*
zr&sf6DU7Qx5#9t$XPbnB;`&5JMi;>5W%u(?4pB;JRPx^AZ@1rKq!K>Hf?&MNbp)p@F^sFfg4f*hT*2
z)H2=wsp&4CEn0$C%3g@_T>kkc%WD5w;NfUh!QbAeIf*D-d#5YR2sEazDJUpF6$R8`
zDuGe~PHTR_Esmk?0`Am5B~ZB*uc5TIiGv}eGz$Gwzr_)mC?khHlO%S(zqZx~26{5G
zAfPW0GPEruF|^(&>*$tg2Wh7+2zs6f@fshtrsd_Qm2W0p*yl-J8sdnPKGAWL1#`hf
zgi5nq42I0~ewIF<>WJDT|Q
z!qeyI>E(#k+5~C}T1CoF^z-!yc26s@X&e!Y?KtSp%AK541|3YtLc_8FbLvm^F?fNt
z*OR4M^82Gzfu!!S<*&7)K5}+wOI=M)y-V=>HK%;-oZDE~6dW7oCp|1-s{6;VBkl-B
zKfl^gJ?e|HeJllDS9L^0M84Cut9GL^N`_qT&Y0c#(>1KKBF&Z(f4itU(&e
zfeJjm50nd!+~a)TljAj(96!bzu3bwPb7J_a>P%N!<7y__mKi7$*8$cC(OZtyZgwZg
zkP|=OqGQB~(hNzYgI>W5x(y1_p8rU@bN*^^i*=$5@7S}#@fKC-}?>wlNA5I2t!X)U~nBp;@yA{P}(_>@UR)`y*k>^^lFJkt%OB=e|M=Qy$2=iZAMJ<
zAXwZ_?cR@h2}%xPQ#gU8Dn}`9TITC(-S9Vet=*|CFp%b_yH9W*=zrOK<_ZO;VH=8&aR(pFvm^lI}JuA?Edod{$?u;`{w5NICnap^ij?S
zkGaN|9FmeO$d8y#Za+Nes4U~r06S%(&}V|(4TEs1&#&+EMRp(&KoKK{MvWeHvsu{>
z=aE+VQi)vlEA^z;h2P2|`>KUV&}NgRZGI6y5de;pf<-#?UMw$H~a^B{l5t#
zl9H0y=3|C~>AT`-?1wc-bThH1QnA}9=s|6+%M6wi6`Cl2F{E!k*6!|@
zcKDomzd7^S`mJ6As*{e+G@zcB5z}LUdMdSmeaCLM)EA0E0;Klb>D3KOAj^43c?QR-
zJY7f3{Vr9D6MCJmAFe7B9;Qc6zir(m>Y_2@h;N}=6$>ybeFM}^1l%j~%1S24zu8&P
zYmTAVy55hM&I=29{6YE9z1BhfypJWjmBWr9g_#$foWMKPG1f$(;cHSlcjZ|6=bL4@
zQGW6-qZfNqR{rJ3?v^uU(%Zx3=J;}%6or$-Gfl1(N$bg-Q?7bcmbqaQtJ@l`rwlbZ
z(+#X7mIz%vf5kd~1}1&|`U1cJE)ek)l7XI0G7pNlN5)!(4qUL`i(JaTl
zVsNZMOYUqSzThm~$F7n{alqDLlR*paWy8NDYkyiXC8a`)SxDxnzE8{7Q)d&Yd
zU;7#xV8Peg*R0Vp#vGywH-~k4Hy0Bz5)z@kK#&9rGhSnb3>K%3HfaT!fVZ#KX5QY-
zO}ub|$#gI!L>Czj&T#Ix#pF+g0N^Z)H#nJT*4f$=*^3bTzBgvPHjIO@uc#?-TCbXW
zlQ6O%4QD3zhd7R9@c9j(!Z)7Cor_Dd>wSTkz~8x9i;$O;2rX19qz5%U0+3ECeAQOd
z*jHJ=W6G0^|M%Mh?M(oTo}M14Zz8#3*8@uQy3Hw#F8gO&X?BjQ9k0RGq=3~m)>M?@
z-bY?gp4a)F(A}?VyY-iTjEdnbE$`B3p~XA%oyeZd}C!WKIIdSm&c04yS6w$t|T
zU@D6=*ih#*7dvBGab8FUv9AcRvBPq6bLU(g9f_Hl&$3C-y1LqJskhk2hg%+iKzHe8(<q6E#L3oOn04>yW2*G*nc3H@|w#F`yqx951!(E;O(xC@B24
zngJ4w>&n{NS(lBMd*2XGQqn}s>4+9XyaDTnnj+=7M7g2A+kXj!`>s8M5i+EQE)JHMluhwVwM8>rb!Kc+!b1@VP`?0pMzkVeG5p
z!$VnMHdSG|Z&{Z`8AeAin=%(k3LNSBu#A
z6d9-+Tz6Q@2vJ7IMfYSDO&sT>#KZujGUflfCELF%AuIw;TzgN-&E;x)q8?4*TvtvT
zwMOajRN)=r7=C9!2YnA{WBXTn@7qm|NCLj}(9l<0!N1
z%7nqIHb34>3o-X^`hsJm6Sb&gp=+3H|GqaD<$lS^4=(|217RV$gV6;D2|QV-^7y8Va#P%Su4<
zZ>pxV*^?5)kmi2x7&tY<{_F79QKEbGUh$60IALEgyS4hj{c#KhGczhh;S2ApS+$Gf
z=iv0wO-F#1(U=&?9koWFB43#65(Gpg2BU1+P;|VwSs%Bo(P{lb(fJGck0yfvhQ};0
zSVJQt^E}RzfHA5KEU0`J31av@z0$rh)#DaH&)^1rp;Y46=ATPLdr&rYwu$M_&VvLL
z7zNaoetm)!oAyKtNYqB^kLc()Qp^i&_%t+c*Cu>>hOsj(OW(Oc#=J%DSTgUn&gFbm
z_CKxP?P+Cb@fe4?w=Fi1Czw?xvl)#n5xbW(3zmnhh)GDW`5$Dxy>Uk8V&w6>;svd5?5*(&j3!3*_wcjsiX$Up;^LoTIWze+Ngsd9dd;jGT
zwXL?8cmvfX=4y&1<>XcmW+ya5NNFX4cSNI|uf#|{*onXV0Yr|)s;}U&WV6EzpcIU%
zM;Vfi`1<6MAUeWE5S(reDr)Mu7=F)~sbaMkqxB9-t@pN>yfc}-&|8WgGhQoS1TUHO
znqKZyXgqDu5~W{ZkgF!N%h|~XM%)T2LaeT?JSDh=FBo4RWdJkd>&b~1Cnx7g+p|y4
z55dmSeG>>rIpE8V>5pbYuwjfNV&+yAIaW%$CdM@>lj>XPniAHUc#vAvH)3AM+4b>*
za7i=&=I7B+P~rT9|8frGDtL4U!EZwwSP1CLOb_RVSj|TzJP&UpAFLwqHjhcZbddo$
zX3hAA&f_?!1a+ild5jDz;zDSh5GY42T_R+d0o9zmN=QbA)u<=(Pj4ZL*VTb)GKb})
zNSqfxDW7|+kw|ah+4b0`n#C3mF6;cCvPqIG1?QcqS+J39F9A8VQ{I)15>f%}v~dtV
zPc5@Shx*YHV)yWH^tW&NrNXE<(DxEqb8q`C$<`OCDK7*~_Sc6A&YWhF&V=@u?IZ`b
zt>DIkDXB$|B6AzWTvQDrcJ|=sM>}G_r(0YI(8~Ud*zs=;s7-P*XR|?HPrVbvmq#<<
zDc5CHjM4r49Whl#Dstb3igc&gzFB&o+MXYX5rs&(YDaSk0yFX&kYx+@vz`AuujueQelFY_1mO{Td
z2P9f|g||ot>-ouhMpjl4#l^V$Gj`cB*z&tI5zpV|se=k+$b&YU%8CvP*uoX6D&oNr
z$JequHUywkqtj-N@9EV-eA(kvbaTVG!-j*|yxXY^C}T=P%=6DQiHo8-AGXZp?x
zoR^hj16iK7Xqc|aav5{)(a}Srq7aW;IAWQMdJJsno|A~3HqwY$QPNU+wmR#kT_p)T2^KYRJYI4BXF@AXA#t_
zmA22aiW89#bcPLy>hFq^TcQi0E!`a~(*2(Ydw*B>5cRktcenQluJ?6SbGq9Lh=!wQ
zd*6nGnu1BI9+{hq2GT>CA>@ukV7tu+iI7%td=#;{w$LD;DT;rTp%>LLneA>1m5z8j
z*lqokRa^=WMJ#K|A1dx^*kr&`+)R$YIqeN
z`o*sDw_Y(feNfo_>tN)>*-9vJy!nO`OqcA1ttU2
zDKg|97qZQa;6hqDjrg*EYTiMd!8`ABnoa$HPT27&5
zlexLuA5s6k@%^E8Xnv~g|JZPiP_ER=@Pn1eYIXISuhL$QF
zFgb}~37El_%Tt%7j?UKi`k8ksa!HOMKn~(|wI1q@*siRqT05*;Hs}bvX2L=Mk9hVT
z`iOW9?|TlLwyb*Cr#(GclXKWpF%nUS(v>14B;;THS(=MxTI!_SQx^dXa1MPizz5V@
zA2=Q!Fh3V34en7?WiHH;R*8a!bjs`KLyd)6IlAqzD~?!#S66;F%l0I3j*=0CamNNr
z#fmW>4t<85@l+<7fUjd~2zTl#6eagxtxXK6n*kGOf@TeSPkBUo@?U)Z(;h5`IAD&a
zaP%!9vunY>dqJ9gxlX^d+R%N7gHo-cr1ndgjI^8_;P0&)_s83cKm1v0tY$zoV+{z!
z{ZY;xK!Agfh>({uTF=(9yiX+0{5Sts?bt$*Suxe`j^Znt5X?E-mEe?KVR)Jh>~}5T
zU?fx;i8~(7sa$yS1-wD*9+J((L75fyH3{l26|j%kPXQRS
z+n(3I2JY03qUyw5034I8~w$NkoYb7L)hh-?20$cruTt@P5B-srTU>1o6iIT1Q
zQLEcrpHjutBre>Yl8VL0l5P8KJ6=3eAbEPZs$%+4u=R}YdCkvPuFOA-^
zgLac^5b!Wg#-%T7ki<8X#uHI7oyaJI(bPJofF6Re$#qBpSR97hwA00rq>IM>_>hHO
zK_U<7-}o@jYIfxcWqQc}-g>O=~k
z*Bh~16!{Uyn_C}wu*fw?vY{b;4559FP4Y`b#M9agi(|vzlX0aL(t5dkuhIQ27-jZQ{4|;oyL}N!MiZOwPGk
zK;V&F+1Fl3Gti5-KO_jIzR1T?IG06Gb!>L5LLm`@!Uf3s2;;ED72=8vlY#1;&rZ-R
zYAE#v5~Oorp;8zKXp+qKv5vuMgB}wVIV7AL(;(qm|DbeJV>f+kp`jW|Yk?F3fkU@9
zo%)3sw;&n~?P62CnEDU6J;ZIU<8s(zoN4I472R*e8!%_R!>DrVocUibi69{}b2t(k
zn&5nm%TWf>oyUOHgqjfxUl0$JEIDN$kKhRZOP#_?(@0Awh^9k28?&9|Vx*rM{g_?~bN0<<@^t74Y!r211)3
z@(K&jdvJySRd0VIb&Iu9`Eki>fSZhR
zZ4XG(T+)tW%;$2?eT$^$oA51iBcppSe`?m}0MW*OAerIxFAYAY-&1-(zMV__vg@m9w5sf_F_al+yEcLI4fjE|Xp@B$}i6D%Ufc8jwF`X6BRg
zG*O_YaXD_h2MM*HX(HIzO%RCUt;>zdy@>3va>cF2E0qw8xw^9;}xyhlUp1j0cFP8gw
zv)sD(i(lkGsyf^AtkEep$5^uBMo24^x&OW#lST$AOH_Nn+iJTk57KUyc#)Sb{rm(u$^r=>T$1!oOE7Ss-4&<@}x>QazXzn1Vg
zxX*M>@2&>A@WaJ~Q0s#uGmz}v-O)|Qa)MZG7vOT4{QSh}BwKrD*FEHOXqkO?{*Pz`
zN$2%I1$iMuLqniZKz=CXioJUM8k}|r*d!q#gZxSHhdlUF=6N1D=sW=!I_Dk=Y4EYh
zK#d`!Ofeg!*;P)DWqWI2WirZH_0S87+jdr4d_saq3_;7ky5n~s3^&D1OT?P5gf3dy
z6^x9>`3GdT99&$wdwPNafo8OsW7XEy#^U$p1j%!mEdC6T&qKq)k^t#)i?Oxcc<#R0
z9235yNM0_sfG`0(V)}k>x@z%rM`v4Y_O#pqE~wRH`^Ruxc57Xst3BYJQ4p8WKJ!tEF#Nazih3rB7W7eqd}(0Wph6T#oNw
zU+diL7MjF(D+)$eZuT<`6L49A%sk!LX^}M#&d*veR)%D|`1{2!uR-`51=*w52y&67
zd6o%f^7|2#+nga{;>6tw0`{rViCi^p=e0mC$mWl>aq*-yP@0VaH
ztRwQmCyjyauzt2F4OID)vw`voa5s`iLdIV&R=SNcno#<0(-a^(&ZF~%%{?e4Z3&U+
zKuwSR*3|^raq)bkb}VmF7mg#IR0jmU#is1Nll!i)=Uju%9_PJD4XJ0iSV>THVvL!o
zp!}cIruus~)oN^m_@A?I5PcFze+-&;5ZtSRQMvm(ko!GYVbh0*sJ;2l5}Dh@QrO!Y
z4}ye*)cjC2dHG>Y12hsb&MVfG?WOP=sTM}kW|34+3${v0zH^`1_T>V35zU1Pui4S%fm_;6%k(=jAFtN
zZ)ga*E>2trL7gk91|(L7S@q7JXSJiF@1cz`kE`{|AHpGWGFeafqTw4O6zqY0aB(Z$
ze%5?A64qZE2JHnYUgGa1PjluR%*@ObfRgCCS8O&_D$W32A(u8dwc1fRJbHNiLu>hjTVT0aefmXy>J=gXjf%X4{
zb|k)!J(RQoS+U0xIrm3tbx-UjB1jbdF=rJBQ}}JpRBO8k^5s@!4e+r;%u@jZeowqs
z(@uXPW~c^goyLCG4Q`(fvJuH3%oqztL)w*Qo1CtkHiRpca;;?{?6&hNY$5
zira(vC`?2tra7w?>^g&8
zyieNQiHV8r?d`j|yG8fE1@>JtC$7){w;==xf6}$V(c>A%#<#*3Xe;A`!bRN+_~yDa
z$Cx!xq6_GEj$#f>#n?P!)$~WH58rqi^hA`ENzo`;2Y`MHF`xU*1>)-YGO56NBMt6YfR2IhoF=;*p3(!6dBv&Eo
zyCqnvMjtV>&t?!jG!A>Btekhpzx-b@cKC;&@Ywd)aqRmBS@2qbJY+smfY{#N4(aOY
z(PX7G@PGy@6$eLr)#wj}O5u*cHo-3HVZ}_>=*W0=-X96%Ox1mHwreQ>nfHlQ>tTgv
zT-2%;YfZ;n3biHajIgos#?4m&$JibSYp;q&q}W=-}%$A
zd}oIe@X4M(U&@&czLww3uk7uS_v?h>?S=8(EVY1huqkzSeJl^A7p4UE05c+&N%=+x
z6JFwSu*PQ_-1Ac3CcjYpyxFZYPm<~MK40ZkV3ta<%bB%wvEmSGB=t4Oqr)FJpR*en
zxPiffLU4pZ7K}pNC$LXCGi61sOdIwO2MxLRrssjR({y~GHXju}l`9@JJU;I&
z*5BU`X37#k^z-20AcgZ(2?SEY5)%h2C`4>yi(dKdj1mEtSP(ci;W^(@QY63x0Udr?
zpAc;St~IZ_v`yQ19xalqhgHw-j4xiZj#eRaWjgibU=FZJg7Yz81m^yzX%;ozor!`z
z#EFj+>HqfnR)OTfOxG8NijmKQMs}2cKo+9@-&Rz7@&Thk$2}l7j9ynSekm9Ia^Vzu
z2k*X?p_-YQxu@X+Cf;Dc6iNbB{B>s5uoZJ;K}h)A?=r2p4*CYa9d2SkQ429+Q!-!q
z2h{Mj0?G8?vjn*DS`c3a`9K7ahS7k3dh)6xbt5{fptxEgq_tE4Q3q6g7gJq7tqeLZJKM+YUa1siEeNks_a2Ug=0$2I&`3g(W$fswmvxQ{Iiyb#%X
zLrYFf1_!ZfVY8d^j;|U9Sy$R^ZXmCN@Gmjr2`2E-Wl~P~mzG?9BY}M)3s&D(AqKH&*jrV|Jb}>RO
zxcP*OuX_hnEOP{;3=N6(eArUEBZmA6r;2BwQAO}$l#`iA#MRUA>-EXW3RuP7<>f3e
z^a&;)K1E>CF=B0CucuT0e^LzFOWFTejkuSyZd2z^w?^`WdvR6n6m!36^Z2GyDe;?~
zt`P#k+H|4b4V&Xg=Qa(4n9y_2i3v=yxgO4$h|?0omn?Y&0O4Rqx?yd2vRGSvJv1{5
zpm(oRnAh!cb&j!R6cZfL~SgUW!%K1Wjo-tzAvb~
z-&o6G2T9R!ct1z_vz*2fty?)(nz7FS`wMc#V;GJ5KDi``yScfkmT6M}*zlG^^(HW&
zw{()VnK?cCQ=Aat3ey6HJ8#aEx&XAPrAMN^KJbW`?*P75yb}7B2qgi`oF!{P1!J?;TeTy_Ieblg{6o8OC4f+t8Ri
zK2O~KOt}H5X#XnLu{rH4rVU_;9Dxns@hNz$D;(;q-nb|K_?Z?EjKbFjGXK>i_Azs1
zYI#U4ExP?R$v9P-V2nzs(6{^RYbVB384!TRhQ@@;bku-I_ns>fVQd@5OFqP&pAFtD@_*+62HO+f+O~34-DD?hmGxHShS(0*{w_+p51#|Ya(zacX0o=>$s%~h{UvVkf{qUZ
zO**&BS0sE*$Mb?)1wp`VCW}}A9VE(>LiMKqzmC2&aP(!5_D*%f8~?^RwHYYNKoab!
zdwyyLR94plnb0P`px_K7A;*f8aOE<%`%5$?YLFz~#SZkn_o1&fJw_a784RnGiG9KP
zQ+0#txK9I^NLx@7DH{vsq~d?ruC?Pf;#Uo_k1M3m{QY=-1aP>-qAxbV0UHBaf(Shc
zEPryN4@YLe_R+6H{!;26Bcg8Km&1?^o_4b&SW+1n>gO6)_WAJ4gkhpcCge0Vaahdc
zOs)=6!Q3T)-jRb@cyr+N-M=Hs%mB+I1k|{hBP+`1-&Tokg;LmRvPPbmfVT}Is>?VA
zdvI$};EDq?ea$w@i$8#Dq0uECB(#kFlf79%%LEL~KzA@9JQ6g#6uP!{Xbz4zev8&t
zrp6Q?m#I#cD5txykbhe9V0Zyb6wk?S=-~rUrlMH6e3E0(DF&>l
zz`jk|=>P%ulaKKf@}^*{eK3RPx^ll%uPqZ~LSs`>%!@vsq38B@=`|fE=h^5MV|BxG
zfhKON8+c5BhL067L9n2CDXtX7*Wp&UDPqH!e8Nqx{3ROIiaOJ6&$ytgdau#(e{??R
zB>w9D{s0JLa5?Xc`b{_aK61~L>-lwIcC8->ph-yt-WvB!>x<45WP;ktCAljJ-ppW2
z-aAN%iz-6->6=ZRzm~WuZ^kDkZV#$Q%$Ju!KtFQUiG2PUeRJ0i9qmo9*u*
zcP%)jn;YJg+!Jfk2ChGrElfi$pD80I_C`1)TK{rSRxw=dUF;^UBO*V@rHJ?5clP|R
z+&2uK9gEE}!4vhH1_8nCHU0i_oh~u^7%$n+v}*ivmtWx-MElRM)n{0RxD%<3cCZ}l
zw#5Tt(}eypW=}Z5a)9)Mf(jjf1R}apt+fXDYJ#sZF)`EGn-;gX*ThYIO#dtL5jGZN
ze*5O%)YLSGgop^{%A@JDYJUzuzsmq%H~5Bv9<(}q0)nhdnC%y!l`5G7K}RmX-no13|tzxLtn*eCa{+)tnU=_|>@L
z$xE}rl=dHj0R(%O5h11XegVIoofOjI2;^9cbY%ec4+0bIa9m0*BugpT<;prnzK9%9
z-xeUcOUTN0kBy;td3k|GR00&Rq3<<+`uoE+jBgFlh18=(%i8DX=RqFZxb@*o8VuHo
z0wQES`4a_%i8nwoQ=5qh4>x>zxP{PyuE4;)UNiZKDE{;*>KRSX3bDv4}Rc?Gch)1^n1c9
zki8EFBkLf1)Lo`?@8IB39~y&TqnU|kG!So;&X?&Hs!Kh)X}@hJR@ARk7{Jff{BtP!
z4<9nwHSp%f${l08V;9P`8)UYJZEnv7B=hp~ne_jpf!O|zK)=pAqN)Jps)6?Z;u26x
zSTPq1%w1Ywjm6`?_q=2V*8$59xjN#b1-Fyk(EPlgCY@h^t5JO%Ja_cDjkWBxqp3og6Uqk-&sF&Z
zzk)U@ogbujQ+R>xf5q1OXn~~bc&v-n9iBRH;CtGQ4!(JUC`{*FtdRsVf5@CGZRpMo
zh%MnGI-7siTn{BUw7hV^%+dlvh(L$hWaYd6C^A$Cq20c_EN(uNdJ7$M+b}+lbO+2E
zT3{T}B~Q3MTN~n}AupFdOqZq>ic(Lv`~(6XEoXJ0v(g*Lk*h68xpwz|
zf=U(Wo|B%)_3FQ=$il17iOgk_v{}`0rXcz6TZoC0W=d~sU}iD^wd#(Pm7y;i#E&*M
zj6E;+23mc@Nuh7&HQ)gslWoP(?1y;u8j-*uO;`R%PDB?AI5tYEA$_wEaOq0F+bQ-m
zBmiWj111QlPCqp!aQ!p+?mrjEXn=_J4;y`Z?l9!Xn(Wm80=C4)v;UvM&O4mz{*U)66)h@JRyWCx(6Duj
zL?J38GqXWLBC}FSvUgTCSs{BRm1JdP@0}gl>pVaD{mwbpIoEa0KX=!ye7~Rhe!s@^
z`FLH8KciV}PkvHjm_6J8U=4N)O34Hzj_`mD|WCW8EHz{c{Q%;W2cyL
z=LRr8wUP(y_r2e=NkT&$I&M1B^7i%D{lO4-UgBM?c>b+&G-L9RcATthp@8PvLU+g!
zAq#d_SE3UmVA{r=|KL{~`b1>EKcRc5|6BGpLikHql}BBj!{&#zHk)-n(Ro3$F4E*~
zn_rE_Y_Q#<>z@qx>md}!tA*5;%-i@@gzQY--ktZnL+|k1x%#>K33b<((;nYKJ5+Vr
zzm@)o9?hUW9`d6#Gwg+u;a+K_mh1^LhzTAm!~+O0v$%5fQ-i>8$1F}q9lkf(U2bxE
z3yHe_Nt)X>d{-UrvK2g?Pn_5xJyxHj7f3(%t!tKSUvXJd+Pv<^$Hhe6Y?X(kN=iG%
z;Y>xj%Cb)REqrr&Q;oZ;qvHYX8Hv(ZBBJk`|j1}!Bj$CU}1H)qYh
zO?HY9eHo&N_Wl{=rzl*DW2#f1@At{&65`3IA+dq;k>yKseY>)>TCW_>*Z-f)B{K!j7;-4#5>7&GfP7ZPsYT3=|JHQgyBo65C)E_a$H7wL3@F8%Rz*+O{jdOYy|1H{PH)O55u{k^UY{6|ntr(@-pa1q|IBRSCY8ng2A1&QSoU39Dyy=Y
zMsujz?8YJrE6U#Q5?3d8F-bWdKmVuo7P?q0aRww;TE2vUHM-L=+Wg-XBMTfQ`z;#3ZJf=NYBvKjdAXsHA(-cD`zBOd0t1xzF~@(=YxHqkQ8`
zLU@X~Xc~bFS=8N%)=KPyU;Nn156j5N?0M0#7FtXxsRnh4+0FRW3-{+dmTYt2E#ZwA
zKXFGNK^vY$O^;$HpSr(}`RXRY@a7kL@80ez6vt;FY$At(2@2d>knKng*wj1cyxkM+
zdZ=KEttr4t?#k!WD?Yyew`1N|V^1v?Kfe!!guM&UP}_c>euB8_(y#iwDaS2L&wfBU
zvAgXcj_d(H$$WNGVf|-;_W2Y%C#lK(YAw()wbzS(;BDY<*R#PR{Ug(u1~L^FXXj%P
zj>|%3#X_V-+B-r)=g1tOX^`|$=2y$vK$WB;caH*;-G3WKbej#MuE%H?4ZWhP2i}5E
zK!O00Zd}WRrasEI$fmsi^qer5?QLga3Ayw!U!=V*?%urE*Q|;8d)b{`DcL%lee&IM
zot6(Qx(gp2d-5d5W-Nt|q;S=OM%ZdJG+Lip@?|LGP}`-x#=W0V^J<%HjiNu()G83a
zT-gb5;N$a1!e}5HlqWwyuvLOlHDYD;NSxPlN
zjuxH$cJd7gyT%8Sxg_d}fHlT~P1pQ?nnrNVJIse_iYDRluPT@-t}e2>iE`;|1)Fc8
zTHPgL5oA{MB8~WRuiJh2$5H``fJwVXG9%e+p=B!3)J(1Ur5$&U<3gkTHfR4czOAC@
z@1KGjVGCYQ@^fp;dwh}ZEC@y@icV=
zVQXSM%|5B-r_@EH(V2_I5#<|f)*>PzFt}fRv6s%(|9TC*A>ef#$i}IKEya8QyFjN@
z*f!^by;4X>`V)Jnd1L;|L+Ydd%Op?Ty)e5x60PX^`gKoITAXsRJuItS&nWX`YYJyM
zY53nm|Q?C(;?HPS+2-Qpg*)D!?r*)y1pZg0L%bUr9?x?Va(
z;afocKGw8InYvvSh9#_i2mKH1eeB-#!aHk+QvSJ`APO;YaSvHv@u6m&z^zx}6|`1Y
zi_0sMd>~I^l3EAVWV>{Ee8$|!+HEy7@L^jwoFr5HuerO*vQp!}J)MlSuWJq5I_n}O
z>-kj9r5{(OoIWQg>8EE0qyO^GUNv?Ii9AE0Kt_t2g3`R<=Ybju_MbS!*3ca!A@ethyb
zOWHLur>@h;3N3n6W@Rp;-6gI08-v=79GU`^3vdHBKP1r}sPTi!O@^xmxWO$Dl&bFL=mC!Bn9
z#v~+mEHIHIwWYHeU7`OXJ-!;EwE--jol-b3Gef=6zfM~+KneOwa(VDp|1os+bjV|w>Q{PAvgH=_z5<0
z1}-ixXu)frksmd&ssLT@GYgKF%()7CoHXY^j>KSWQCnNkEE(4;LXWlOi=-JK6
zfUwJUN$_z>KQQM5g+;AY>T{@Z|m@N0z=>HPqLufl>ra0R7GyPJ=hJYNt6i4NUy}M(+7`~&b)IAV%EQ9SG#(4L@jRBIvXBQ-4JZyM*UfBlI-E9)#E)21GFpV<
z?*3JI)rv^*`j3a}uKfPzbe;$l5$XR1)3j;JUDyTHYXYk^B(JbZYl!C2xnGD=SbGIU
z@6A)RVEJuzbljbwn)XC`0lTrJh;qdwU!PBX%Y%{Zw#l}}gLdu15=Z6E5@w~?4v*bi
z&$)Pj6{Oq{-L|`Y`4r9*>RCc`4F$MX!M=+YG+^2)*iL(<-`qF>i9>BH4S)seDSfp?
zC!#C9c@8@b%2ioesUz@8Q2j8!yL?FyUQwtZ;UOh3Wysdbv0s6H8TnN+CLPC9S3`dzm->xBfh{E(*#tR_q@WIRM5G1e8y(
z%tO2o2QCwJdpuUgQ0W&VM{mY%SiNSQEph?)>`Kxnr-{Qj)o`7%eFQePJI6m
zqZJ&WzJJJxsGXU%HGeSc_SsESv;W{h7+a?sy4i71wcex~ysA(21g%a>!qalSH;=0+
zrJ(rvTB*y=hqDLVZDKf0{e_MzW9EokGsmdZrb>A*#!0b)k3g=+y0c
z?#hK@IFd-_u;#vEnb6FA&3(&tR2~(G5Dq2N
z!=g5a*VlIv)vjmw-Hl7NC)4lLt`&6r)fz-NoXLr{R?Jxaej{sQ@WzwUpW0<|5mvKb
zjSUQNDTP1_+M$cv19(~>_8+5GaRQxp&!=Z29myH9UbSzUU~BudMYJ*R9ZTU=`bwVl
z*z8K-KqD#lv+vV0Dl!ClwF^HI4>cGucBi6iiTGf{iRI%qbRzFdO_*`^{>+3(EPn!
zc8vXBqyFCi-KcNXbUE*hw|6O8T-Z)_;py|^R)6*K<$bw8$MjKK?8AY$^S^st4#Mv9
zi;72+^l?JMMIb)6p;qWWKOvHCSoswPf%qJh-863X89Kb0k>__R<3Qy6zjs_t4ETqf
zvBhSZS~d})sfStdh7$*B|ExpGvZ)4xx))>?0Si-66g#FI2*@}s=+1_y1xp-n1NpRCIcPleE(6q6uu>fN!}30>qc
zLh23UP8iWM^uA2lJi0zK`je}0E`V3sjoPp_T#J-6lt3_o_u~yt!;p6+h_N9d=-aOJs7Ie3Dy)iNvOaEBTx+r4(^uG0m*#vyo
z%^TdAScyNs!S@zs$)
zjU5h?%VB;45*ZJANhLoS9y>2c(`98{?}%`4>{nII8D4y*&Ze`_leWFwVzpi@t24)N
z$gNE({&SBt+HALn^7YXZU%kFARu39ZEb~>l)Rzskn-_4fI^S}8g_&58Ds7A
z_U8&bNK5+(uZw1hqd%~k_t#g|Ya%4v@DVWV5;DT|b&-wC@qe*cM*={MR+3a&b3>k`
zs(f0>wh=!kPl18^gbqC;tUqv)&uK!4B>9hUF>ROECmuzoNipCOTH_si6eM3R6$x{7
zy<1|v-pM6q%?-0Dp8n~%<_MrDCF`QTz-3g-TK1cfaeK!2VYhPMev?Ok!=w4>X`5oZ
z{24O4APJCE-7t(xyA
z4D>S`p6+w~6i+t0VNd>c!vGg>$*6UlZ3mbbgdl-~tcx?P#u4u-H}{wK&+PqsRAHWf
zQK!k(l^G;3$0@NCJO{5t`f$ybtj=6DG23e+tCubFLoTxtNtL*`yAaea&v!lQ>2DZa
z5`Ro^R%MrrUalM6qIcnamN-$I60X^3B-6tPo
z?xvb5x8w2oec`8S-B@_!1&3*d8y{{->FI5Swgw1db!a8$^7_E)_ZGTQ&AOA=5kW`)
zl&^ZZ_zcoD;*eTcr>r;=g1hhc+MVQy<+mTg!4BMr@kseHFm*#Sb-AI`+fkAzcw$N
zK<9l6(Jh+EQi%=G4{_aBpGlm_y9cx7#i@TkeX^7--st^2WXoZ+0i~_IB|6z?vDrHf
z8N)g;+}}CTF`Xi|ht0X)smZ|$=Cb*AUSO$ZSXe&y4-I{NMn0u)8WvBHgq27xr2YhB
zdqXx=iji-n>&yezHj9BXzh#HIK7`=P>2S_PSiz|uWW6k-{5p|@>J*i($pnOD;*X5-
zkaG|%oXO51kqHsR7$0tDvTH7WiE}%IP-OMP8~J^@-QrIB=vywK+2gok$^nq^g>0ww
zorA5d15tg-LAlc>e;pZOTDTdORG(m5=h8%U<%V3BpbpFVb!*~duCSR>N8*T6_QG3`
zxoJ;L55YLDsin1qzyh3@m0-rm%0z>Q3FA!Vtuy~gl9)0p)+*l=ooyV@j=Q*p_8QmB
zk1@W)bf6S2%1{1fA3VaI4_bIHQ9CfKWQ|kY@L-mHB|0aDpi=7LcVBYxlidGNqnxOW
zc6w_a<`wWQ^zj#|7^y9Ui!SAjGkgDPHM29GWTw$Q{S({?R*FecDqibkB!0Ra&GkmJSDZN_xB-%VKX0su^rcGR4md9v%|0
zz6Y)6RNb!!*AlLH$6j)(ov%elukXwClpA3P-hH56^I|$iGCjvo3cJtkwm)Au$56KT
zH_VLmt>tj4P_~I__{C{H@b7Fl05>1JCilAN>J?zpIT;zB%c59OR_3f-q&L;l0=q~;F)leH1A<@jXDAG$lE^_c)_R5EM
zAM4&9OwP?$&{9Eb)cIY}3;!-;I0ST_p(#yg+qq05@lC{yKWErP`fgMx0$aluK0a}7
z?6OyZzn6B~jzaQ}h=qyx^YmR?0jG77T~zEBahDMp`*Ep;10x>@B)(5$BUP7;J{cb$
zAH)xdidVmsmq#^gh*?E|CIQ|Puql0#%F6W;nT@|dVwKovRiU4s_YBV#it9dSyAE?(f5X8t4o_;o
zJjmve?znabP7!CM`3N_+|06EWNJFD)R5gvd{un4g<*Tc+NAYm|0PQR^EDSGKFTB}`
zowd<3vaSbWPaZZQY^^9rlkV(hVrCYtl;|%R93^#H2NcsRXn5`C7T%u?=>ur!ZZxKI
z^kN|Gb3%qd-fSRmHw}%zj`>h
zbUSK%57cp#!?TeG34=z?y*LwfMU~{mFk}~icSKrGITD4tmGQgaEuP0JLAJZ#5?oU#
zA=2U#`>-4|CeGHM&uoAx#rN4WQmDrJ#x=d(zb_zO2++Ls##??9gi+`{D5GhIzj_
zax<^0_2Cml6GN&vFzLXmkG%2w)cZbrI>kI3
zN^*$#>aN=>|)D_
zKTWpzd!{P8Qe22Q5AE!m-i-DJ1}qqoaHk#WyKmpVT~3+Z$eb3HyQ68*w3BPS|4}9L
zE8}DTGtLREWEQi5i~+XJpARJ^w@yw@))smYrx)%ieV{NWhZrh?I?R^)GXt{r_V$<@
zpybl)U~eyIHA+PSoPV6f;CIFGBI4VbXSvUWbO5{B0^>+(YAU=4NQe*P+PrXc+g;{D
z!3Ll+E_7dS@4vdb{~WvU+GbSi55DTzw@}TqTFIN=%Ab+J9{SQPK%{O${FASFTW^Jpl9^Ex5won1$4kZ*|u_syggbSsAzr
zQ@Ia*9l#s8TswxeELgy}59u{#0brWdr}p-h3cBkSL5K!oLA7(
zP^1u=e|G2^uTe=Wfy@{4XRMjGy}zoO-|{uQB(d;#5W|G+W|r+Ye#L&2e1yg0iYPe#_A2Wc6qWRWF$86x#5KoF{(v8k0z_8o~#K;YHtp1@;Fy
z(#Sd}uM5Q=KTPxBgkZ*LV2tkUy=!E0
zs7SyxyoR}C#IXt^Q2IO-hlw9;`qt*=rMf5Y&yD_MyC+b7e32&cDI5Q=x`z%_*RT-Y
zL<^{uN~_B&SRal;Zrp^}_4}`k1B3hN4;+xhV~CM1F<|8FVGyQ@t_$jn9J&!K^c6gk
zK5VnZ{E|nrLzungIlfPz7P?W}1eMpmzP`4~bAm#Xw1wA7GYhL2OVFHy61WNgK@P3_
z2M=Eajo~42dGP|C)^c2^M#D=1+>xkE0I+#XL%+J3e^v{K$S4d-BeAiy_3`pL2iMQ>
zO953)YL-d3QWzQrGqoKGBE}H4V5S+0K;sbp91n@EE`TSAkv5U&=~P!wxK34=i`gVT
zHGYJn-jF!f14(Uyc9AgloS*yyn!~rtb1@P2;bL-
zi94|uy_soc!V5$v1k20IiP32&=_=qYg+D1C7wLveSd!?(;iD3cp*oIB&);27vKX#m
zz#kV%9M?YV6Hyhys2g~{4B;EY^X^Gd5AOWr=Ul8(s&)BI{q;`k7!5<_|L0HSIwzP_mcXL(#A
zqAOj0Cj#fD6<>wbKUJ!G1`UeM}r=^)?Pil~*0Kme$`)9Jc
z@ABLj_yjSme(P&1@R*6qhZhkah4E59xJ##6(w*^I`9PlU?k=A_Rf4+*`Ik>mPmf7=
z4&@C%;ht@WaF~*T!5x)JOj|O>Wk5L_4-qA+Lg+VC)Ph7G@x!H7rhD=|)H;ZlzT+9p
zeKgC8g_4w%lsF@j;^k2;{J0XY@9m%O4#7{Zzx?NyNk^_X=9>}Ye*n@C0<{#mL@$5{
zB$(Ot22T&PDG^t&d4XzxzOe3ePbmqs6aj!ND8auf62i3tIY#KcQ%U)bzDqS0*rm~=G#m8&
zFlT8Hy=!ci>&ErI@87?VcVuLmd7T_OsX!LVj*)(HII=?$5)y-WL{$+dr=c&&F8}6~
z5(Y{Av~pZuGM?MOx4dJ2^@y_pg-i}0K{E1-pr~iZXtD2ziIXrx>>S3)$tozQVAJwL
z){iAO77AK9P0dg&*7YdzALrzhBSy~kI8*pxb4qG&D|RSY-bgE)c?I-E8cu+N_|o`o
zVM3O1eZ#}Vlslk+wbe@;i|`Q#sHkFqK$w$=SilKhr{>j8#y`NEMo!$fs{m(U_4mU)
z7N0p>gej5VF*=d>$w4*+V&)u5*-8>+iV|xv5NM{2I^sRXN|yj{Bq!#mUBgry;(r0t
zk?b_W04{$lh{Ugfiz0{VX2b_!MwM8@@P6hrK;S8FrPU?jqQIImUE1wBAJb$g0cdg*>$#5
zOzwfv#1(lDWHq=2$d7@9NfJ)dLTW;`cklIuUqeF|utvU$>VZN5^QJ}|pP~y%%w{5i
zEOPs9>H#=R;S!!rFCT`Z`zDMY;19dQAFYf5iuCmK$ZHt!P+~CNdZchW)>;&t`?1Ku
zCt-VHqkG}BALsR5k0GbE#OAdzepVnt=4GKIR*BZP*aZj^^^dvTci6&b=^Ujb`U5Cok}0Jh^FVx_&!BGmqb$_H-1+Se5s
zBoNZ$o+)RWw6ZaJ!jnvRaxvQjt3@21O#7J%HuTEM8;DQ#XX3(zR0@kkqajLT+K9+2N0Fx
z#s_wisgezGysZS|CMf7ISl??AwW9Wm_luW?=!oyclK~|mM@>x)%7$jJ@gjY-N4);T
z-`{^Z92eUhA`}v0NEo6+#8rg*1IW+E&=eh{qpQTmqh4ft8h&?raeIh~DT%`_b`T)}
z5hhtJ{VDEuQuzjUca$hz5MxMj^&d`k$)kkRoT^pXxcO>S7(JuJ-(O%*ju5ftf%Z2F
zdlaqAoXL(ohd*M2FbKDX9993SBpKm*MulQ&ePs$=uA#v|ohIQJ^epJJ}I&zOphZl_Lf9eT0tsRQupcq0B#ox8UO$Q

literal 0
HcmV?d00001

diff --git a/doc/LectureNotes/_build/html/_sources/chapter1.ipynb b/doc/LectureNotes/_build/html/_sources/chapter1.ipynb
index c930e0a6b..9dab26cdc 100644
--- a/doc/LectureNotes/_build/html/_sources/chapter1.ipynb
+++ b/doc/LectureNotes/_build/html/_sources/chapter1.ipynb
@@ -13,38 +13,37 @@
     "\n",
     "\n",
     "\n",
-    "Our emphasis throughout this series of lectures  \n",
-    "is on understanding the mathematical aspects of\n",
-    "different algorithms used in the fields of data analysis and machine learning. \n",
+    "Our emphasis throughout this series of lectures is on understanding\n",
+    "the mathematical aspects of different algorithms used in the fields of\n",
+    "data analysis and machine learning.\n",
     "\n",
-    "However, where possible we will emphasize the\n",
-    "importance of using available software. We start thus with a hands-on\n",
-    "and top-down approach to machine learning. The aim is thus to start with\n",
-    "relevant data or data we have produced \n",
-    "and use these to introduce statistical data analysis\n",
-    "concepts and machine learning algorithms before we delve into the\n",
-    "algorithms themselves. The examples we will use in the beginning, start with simple\n",
-    "polynomials with random noise added. We will use the Python\n",
-    "software package [Scikit-Learn](http://scikit-learn.org/stable/) and\n",
-    "introduce various machine learning algorithms to make fits of\n",
-    "the data and predictions. We move thereafter to more interesting\n",
-    "cases such as data from say experiments (below we will look at experimental nuclear binding energies as an example).\n",
-    "These are examples where we can easily set up the data and\n",
-    "then use machine learning algorithms included in for example\n",
-    "**Scikit-Learn**. \n",
+    "However, where possible we will emphasize the importance of using\n",
+    "available software. We start thus with a hands-on and top-down\n",
+    "approach to machine learning. The aim is thus to start with relevant\n",
+    "data or data we have produced and use these to introduce statistical\n",
+    "data analysis concepts and machine learning algorithms before we delve\n",
+    "into the algorithms themselves. The examples we will use in the\n",
+    "beginning, start with simple polynomials with random noise added. We\n",
+    "will use the Python software package\n",
+    "[Scikit-Learn](http://scikit-learn.org/stable/) and introduce various\n",
+    "machine learning algorithms to make fits of the data and\n",
+    "predictions. We move thereafter to more interesting cases such as data\n",
+    "from say experiments (below we will look at experimental nuclear\n",
+    "binding energies as an example).  These are examples where we can\n",
+    "easily set up the data and then use machine learning algorithms\n",
+    "included in for example **Scikit-Learn**.\n",
     "\n",
     "These examples will serve us the purpose of getting\n",
     "started. Furthermore, they allow us to catch more than two birds with\n",
     "a stone. They will allow us to bring in some programming specific\n",
-    "topics and tools as well as showing the power of various Python \n",
-    "libraries for machine learning and statistical data analysis.  \n",
+    "topics and tools as well as showing the power of various Python\n",
+    "libraries for machine learning and statistical data analysis.\n",
     "\n",
-    "Here, we will mainly focus on two\n",
-    "specific Python packages for Machine Learning, Scikit-Learn and\n",
-    "Tensorflow (see below for links etc).  Moreover, the examples we\n",
-    "introduce will serve as inputs to many of our discussions later, as\n",
-    "well as allowing you to set up models and produce your own data and\n",
-    "get started with programming.\n",
+    "Here, we will mainly focus on two specific Python packages for Machine\n",
+    "Learning, Scikit-Learn and Tensorflow (see below for links etc).\n",
+    "Moreover, the examples we introduce will serve as inputs to many of\n",
+    "our discussions later, as well as allowing you to set up models and\n",
+    "produce your own data and get started with programming.\n",
     "\n",
     "\n",
     "\n",
@@ -286,7 +285,17 @@
     "[gallery](https://matplotlib.org/gallery/index.html) of examples. In\n",
     "this example we plot our original values of $x$ and $y$ as well as the\n",
     "prediction **ypredict** ($\\tilde{y}$), which attempts at fitting our\n",
-    "data with a straight line.\n",
+    "data with a straight line.  Note also that **Scikit-Learn** requires a\n",
+    "matrix as input for the input values $x$ and $y$. In the above code we\n",
+    "have solved this by declaring $x$ and $y$ as arrays of dimension\n",
+    "$n\\times 1$.\n",
+    "\n",
+    "In the code here we have also made a new array for $x\\in [0,1]$. Our\n",
+    "prediction is computed for these values, meaning that they were not\n",
+    "included in the data set used to *train* (or fit) the model.\n",
+    "This is a recurrring theme in machine learning and data analysis. We would like to train a model on a specific given data set.\n",
+    "Thereafter we wish to apply it to data which were not included in the training. Below we will encounter this again in the so-called *train-validate-test* spliting. We will typically split our data into different sets, oen for training, one for validation and finally, our data from the untouched test vault!\n",
+    "\n",
     "\n",
     "The Python code follows here."
    ]
@@ -311,6 +320,7 @@
     "y = 2*x+np.random.randn(100,1)\n",
     "linreg = LinearRegression()\n",
     "linreg.fit(x,y)\n",
+    "# This is our new x-array to which we test our model\n",
     "xnew = np.array([[0],[1]])\n",
     "ypredict = linreg.predict(xnew)\n",
     "\n",
@@ -460,8 +470,7 @@
     "Depending on the parameter in front of the normal distribution, we may\n",
     "have a small or larger relative error. Try to play around with\n",
     "different training data sets and study (graphically) the value of the\n",
-    "relative error. Note also that **Scikit-Learn** requires a matrix as input for the input values $x$ and $y$. In the above code we have\n",
-    "solved this by declaring $x$ and $y$ as arrays of dimension $n\\times 1$.\n",
+    "relative error.\n",
     "\n",
     "As mentioned above, **Scikit-Learn** has an impressive functionality.\n",
     "We can for example extract the values of $\\alpha$ and $\\beta$ and\n",
@@ -1039,7 +1048,7 @@
    "metadata": {},
    "source": [
     "The next step, and we will define this mathematically later, is to set up the so-called **design matrix**. We will throughout call this matrix $\\boldsymbol{X}$.\n",
-    "It has dimensionality $p\\times n$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit."
+    "It has dimensionality $n\\times p$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit."
    ]
   },
   {
@@ -1064,9 +1073,12 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Note well that we have made life simple here. We perform a fit in terms of the number of nucleons only. A more sophisticated fit can be done by including an explicit dependence on the number of protons and neutrons in the asymmetry and Coulomb terms.\n",
+    "Note well that we have made life simple here. We perform a fit in\n",
+    "terms of the number of nucleons only.  A more sophisticated fit can be\n",
+    "done by including an explicit dependence on the number of protons and\n",
+    "neutrons in the asymmetry and Coulomb terms. We leave this as an exercise to you the reader.\n",
     "\n",
-    "With **scikitlearn** we are now ready to use linear regression and fit our data."
+    "With **Scikit-Learn** we are now ready to use linear regression and fit our data."
    ]
   },
   {
@@ -1124,7 +1136,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "As a teaser, let us now see how we can do this with decision trees using **scikit-learn**. Later we will switch to so-called **random forests**!"
+    "As a teaser, let us now see how we can do this with decision trees using **Scikit-Learn**. Later we will switch to so-called **random forests**!"
    ]
   },
   {
@@ -1172,6 +1184,17 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "With a deeper and deeper tree level, we can almost reproduce every\n",
+    "single data point by increasing the max depth of the tree.\n",
+    "We can actually decide to make a decision tree which fits every single point.\n",
+    "As we will\n",
+    "see later, this has the benefit that we can really train a model which\n",
+    "traverses every single data point. However, the price we pay is that\n",
+    "we will easily overfit. That is, if we apply our model to unseen data,\n",
+    "we will most likely fail miserably in our attempt at making\n",
+    "predictions. As an exercise, try to make the tree level larger by adjusting the maximum depth variable. When printing out the predicition, you will note that the binding energy of every nucleus is accurately reproduced.\n",
+    "\n",
+    "\n",
     "The **seaborn** package allows us to visualize data in an efficient way. Note that we use **scikit-learn**'s multi-layer perceptron (or feed forward neural network) \n",
     "functionality."
    ]
@@ -1977,6 +2000,47 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "We can then compute the second derivative of the cost function, which in our case is the second derivative\n",
+    "of the means squared error. This leads to"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\frac{\\partial^2 C(\\boldsymbol{\\beta})}{\\partial \\boldsymbol{\\beta}^T\\partial \\boldsymbol{\\beta}} =\\frac{2}{n}\\boldsymbol{X}^T\\boldsymbol{X}.\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This quantity defines was what is called the Hessian matrix (the second derivative of a function we want to optimize).\n",
+    "\n",
+    "The Hessian matrix plays an important role and is defined for the mean squared error  as"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\boldsymbol{H}=\\boldsymbol{X}^T\\boldsymbol{X}.\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The Hessian matrix for ordinary least squares is also proportional to\n",
+    "the covariance matrix. As we will see in the chapter on Ridge and Lasso regression, This means that we can use the Singular Value Decomposition of a matrix  to find\n",
+    "the eigenvalues of the covariance matrix and the Hessian matrix in\n",
+    "terms of the singular values.\n",
+    "\n",
+    "\n",
     "The residuals $\\boldsymbol{\\epsilon}$ are in turn given by"
    ]
   },
@@ -2529,9 +2593,7 @@
     "**pandas** again, rather extensively in order to organize our data.\n",
     "\n",
     "The difference now is that we use **Scikit-Learn's** regression tools\n",
-    "instead of our own matrix inversion implementation. Furthermore, we\n",
-    "sneak in **Ridge** regression (to be discussed below) which includes a\n",
-    "hyperparameter $\\lambda$, also to be explained below."
+    "instead of our own matrix inversion implementation."
    ]
   },
   {
@@ -2603,18 +2665,6 @@
     "print('Mean absolute error: %.2f' % mean_absolute_error(Energies, ytilde))\n",
     "print(clf.coef_, clf.intercept_)\n",
     "\n",
-    "# The Ridge regression with a hyperparameter lambda = 0.1\n",
-    "_lambda = 0.1\n",
-    "clf_ridge = skl.Ridge(alpha=_lambda).fit(X, Energies)\n",
-    "yridge = clf_ridge.predict(X)\n",
-    "EoS['Eridge']  = yridge\n",
-    "# The mean squared error                               \n",
-    "print(\"Mean squared error: %.2f\" % mean_squared_error(Energies, yridge))\n",
-    "# Explained variance score: 1 is perfect prediction                                 \n",
-    "print('Variance score: %.2f' % r2_score(Energies, yridge))\n",
-    "# Mean absolute error                                                           \n",
-    "print('Mean absolute error: %.2f' % mean_absolute_error(Energies, yridge))\n",
-    "print(clf_ridge.coef_, clf_ridge.intercept_)\n",
     "\n",
     "fig, ax = plt.subplots()\n",
     "ax.set_xlabel(r'$\\rho[\\mathrm{fm}^{-3}]$')\n",
@@ -2623,8 +2673,6 @@
     "            label='Theoretical data')\n",
     "ax.plot(EoS['Density'], EoS['Eols'], alpha=0.7, lw=2, c='m',\n",
     "            label='OLS')\n",
-    "ax.plot(EoS['Density'], EoS['Eridge'], alpha=0.7, lw=2, c='g',\n",
-    "            label='Ridge $\\lambda = 0.1$')\n",
     "ax.legend()\n",
     "save_fig(\"EoSfitting\")\n",
     "plt.show()"
@@ -2637,13 +2685,11 @@
     "The above simple polynomial in density $\\rho$ gives an excellent fit\n",
     "to the data. \n",
     "\n",
-    "We note also that there is a small deviation between the\n",
-    "standard OLS and the Ridge regression at higher densities. We discuss this in more detail\n",
-    "below.\n",
     "\n",
     "\n",
     "## Splitting our Data in Training and Test data\n",
     "\n",
+    "\n",
     "It is normal in essentially all Machine Learning studies to split the\n",
     "data in a training set and a test set (sometimes also an additional\n",
     "validation set).  **Scikit-Learn** has an own function for this. There\n",
@@ -2653,6 +2699,107 @@
     "postpone a discussion of this splitting to the end of these notes and\n",
     "our discussion of the so-called **bias-variance** tradeoff. Here we\n",
     "limit ourselves to repeat the above equation of state fitting example\n",
+    "but now splitting the data into a training set and a test set.\n",
+    "\n",
+    "Let us study some examples. The first code here takes a simple\n",
+    "one-dimensional second-order polynomial and we fit it to a\n",
+    "second-order polynomial. Depending on the strength of the added noise,\n",
+    "the various measures like the $R2$ score or the mean-squared error,\n",
+    "the fit becomes better or worse."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "\n",
+    "\n",
+    "def R2(y_data, y_model):\n",
+    "    return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)\n",
+    "def MSE(y_data,y_model):\n",
+    "    n = np.size(y_model)\n",
+    "    return np.sum((y_data-y_model)**2)/n\n",
+    "\n",
+    "x = np.random.rand(100)\n",
+    "y = 2.0+5*x*x+0.1*np.random.randn(100)\n",
+    "\n",
+    "\n",
+    "#  The design matrix now as function of a given polynomial\n",
+    "X = np.zeros((len(x),3))\n",
+    "X[:,0] = 1.0\n",
+    "X[:,1] = x\n",
+    "X[:,2] = x**2\n",
+    "# We split the data in test and training data\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n",
+    "# matrix inversion to find beta\n",
+    "beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train\n",
+    "print(beta)\n",
+    "# and then make the prediction\n",
+    "ytilde = X_train @ beta\n",
+    "print(\"Training R2\")\n",
+    "print(R2(y_train,ytilde))\n",
+    "print(\"Training MSE\")\n",
+    "print(MSE(y_train,ytilde))\n",
+    "ypredict = X_test @ beta\n",
+    "print(\"Test R2\")\n",
+    "print(R2(y_test,ypredict))\n",
+    "print(\"Test MSE\")\n",
+    "print(MSE(y_test,ypredict))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Alternatively, you could write your own test-train splitting function as shown here."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "# equivalently in numpy\n",
+    "def train_test_split_numpy(inputs, labels, train_size, test_size):\n",
+    "    n_inputs = len(inputs)\n",
+    "    inputs_shuffled = inputs.copy()\n",
+    "    labels_shuffled = labels.copy()\n",
+    "\n",
+    "    np.random.shuffle(inputs_shuffled)\n",
+    "    np.random.shuffle(labels_shuffled)\n",
+    "\n",
+    "    train_end = int(n_inputs*train_size)\n",
+    "    X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]\n",
+    "    Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]\n",
+    "\n",
+    "    return X_train, X_test, Y_train, Y_test"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "But since **scikit-learn** has its own function for doing this and since\n",
+    "it interfaces easily with **tensorflow** and other libraries, we\n",
+    "normally recommend using the latter functionality.\n",
+    "\n",
+    "\n",
+    "As another example, we apply the training and testing split to \n",
+    "to the above equation of state fitting example\n",
     "but now splitting the data into a training set and a test set."
    ]
   },
@@ -3050,119 +3197,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Splitting our Data in Training and Test data\n",
-    "\n",
-    "\n",
-    "It is normal in essentially all Machine Learning studies to split the\n",
-    "data in a training set and a test set (sometimes also an additional\n",
-    "validation set).  **Scikit-Learn** has an own function for this. There\n",
-    "is no explicit recipe for how much data should be included as training\n",
-    "data and say test data.  An accepted rule of thumb is to use\n",
-    "approximately $2/3$ to $4/5$ of the data as training data. We will\n",
-    "postpone a discussion of this splitting to the end of these notes and\n",
-    "our discussion of the so-called **bias-variance** tradeoff. Here we\n",
-    "limit ourselves to repeat the above equation of state fitting example\n",
-    "but now splitting the data into a training set and a test set.\n",
-    "\n",
-    "Let us study some examples. The first code here takes a simple\n",
-    "one-dimensional second-order polynomial and we fit it to a\n",
-    "second-order polynomial. Depending on the strength of the added noise,\n",
-    "the various measures like the $R2$ score or the mean-squared error,\n",
-    "the fit becomes better or worse."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "collapsed": false,
-    "editable": true
-   },
-   "outputs": [],
-   "source": [
-    "import os\n",
-    "import numpy as np\n",
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "from sklearn.model_selection import train_test_split\n",
-    "\n",
-    "\n",
-    "def R2(y_data, y_model):\n",
-    "    return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)\n",
-    "def MSE(y_data,y_model):\n",
-    "    n = np.size(y_model)\n",
-    "    return np.sum((y_data-y_model)**2)/n\n",
-    "\n",
-    "x = np.random.rand(100)\n",
-    "y = 2.0+5*x*x+0.1*np.random.randn(100)\n",
-    "\n",
-    "\n",
-    "#  The design matrix now as function of a given polynomial\n",
-    "X = np.zeros((len(x),3))\n",
-    "X[:,0] = 1.0\n",
-    "X[:,1] = x\n",
-    "X[:,2] = x**2\n",
-    "# We split the data in test and training data\n",
-    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n",
-    "# matrix inversion to find beta\n",
-    "beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train\n",
-    "print(beta)\n",
-    "# and then make the prediction\n",
-    "ytilde = X_train @ beta\n",
-    "print(\"Training R2\")\n",
-    "print(R2(y_train,ytilde))\n",
-    "print(\"Training MSE\")\n",
-    "print(MSE(y_train,ytilde))\n",
-    "ypredict = X_test @ beta\n",
-    "print(\"Test R2\")\n",
-    "print(R2(y_test,ypredict))\n",
-    "print(\"Test MSE\")\n",
-    "print(MSE(y_test,ypredict))"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Alternatively, you could write your own test-train splitting function as shown here."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "collapsed": false,
-    "editable": true
-   },
-   "outputs": [],
-   "source": [
-    "# equivalently in numpy\n",
-    "def train_test_split_numpy(inputs, labels, train_size, test_size):\n",
-    "    n_inputs = len(inputs)\n",
-    "    inputs_shuffled = inputs.copy()\n",
-    "    labels_shuffled = labels.copy()\n",
-    "\n",
-    "    np.random.shuffle(inputs_shuffled)\n",
-    "    np.random.shuffle(labels_shuffled)\n",
-    "\n",
-    "    train_end = int(n_inputs*train_size)\n",
-    "    X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]\n",
-    "    Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]\n",
-    "\n",
-    "    return X_train, X_test, Y_train, Y_test"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "But since **scikit-learn** has its own function for doing this and since\n",
-    "it interfaces easily with **tensorflow** and other libraries, we\n",
-    "normally recommend using the latter functionality.\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
     "## Reducing the number of degrees of freedom, overarching view\n",
     "\n",
     "Many Machine Learning problems involve thousands or even millions of\n",
diff --git a/doc/LectureNotes/_build/html/_sources/chapter2.ipynb b/doc/LectureNotes/_build/html/_sources/chapter2.ipynb
index 651eadf4a..3049b1030 100644
--- a/doc/LectureNotes/_build/html/_sources/chapter2.ipynb
+++ b/doc/LectureNotes/_build/html/_sources/chapter2.ipynb
@@ -13,7 +13,7 @@
     "What is presented here is a mathematical analysis of various regression algorithms (ordinary least  squares, Ridge and Lasso Regression). The analysis is based on an important algorithm in linear algebra, the so-called Singular Value Decomposition (SVD). \n",
     "\n",
     "\n",
-    "We have shown that in ordinary least squares the optimal parameters $\\beta$ are given by"
+    "We have shown that in ordinary least squares (OLS) the optimal parameters $\\beta$ are given by"
    ]
   },
   {
@@ -21,7 +21,7 @@
    "metadata": {},
    "source": [
     "$$\n",
-    "\\hat{\\boldsymbol{\\beta}} = \\left(\\boldsymbol{X}^T\\boldsymbol{X}\\right)^{-1}\\boldsymbol{X}^T\\boldsymbol{y}.\n",
+    "\\hat{\\boldsymbol{\\beta}}_{\\mathrm{OLS}} = \\left(\\boldsymbol{X}^T\\boldsymbol{X}\\right)^{-1}\\boldsymbol{X}^T\\boldsymbol{y}.\n",
     "$$"
    ]
   },
@@ -79,7 +79,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "The matrix $\\boldsymbol{A}$ has the important property that $\\boldsymbol{A}^2=\\boldsymbol{A}$. This is the definition of a projection matrix.\n",
+    "The matrix $\\boldsymbol{A}$ has the important property that $\\boldsymbol{A}^2=\\boldsymbol{A}$. This is the definition of a [projection matrix](https://en.wikipedia.org/wiki/Projection_matrix).\n",
     "We can then interpret our optimal model $\\tilde{\\boldsymbol{y}}$ as being represented  by an orthogonal  projection of $\\boldsymbol{y}$ onto a space defined by the column vectors of $\\boldsymbol{X}$.  In our case here the matrix $\\boldsymbol{A}$ is a square matrix. If it is a general rectangular matrix we have an oblique projection matrix.\n",
     "\n",
     "\n",
@@ -545,7 +545,7 @@
     "\n",
     "## Code for SVD and Inversion of Matrices\n",
     "\n",
-    "How do we use the SVD to invert a matrix $\\boldsymbol{X}^\\boldsymbol{X}$ which is singular or near singular?\n",
+    "How do we use the SVD to invert a matrix $\\boldsymbol{X}^T\\boldsymbol{X}$ which is singular or near singular?\n",
     "The simple answer is to use the linear algebra function for the pseudoinverse, that is"
    ]
   },
@@ -3258,6 +3258,8 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "**More text will be added to this example.**\n",
+    "\n",
     "## Linking the regression analysis with a statistical interpretation\n",
     "\n",
     "We will now couple the discussions of ordinary least squares, Ridge\n",
@@ -3624,7 +3626,9 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "It is a conditional probability (see below) and reads as the likelihood of a domain of events $\\boldsymbol{D}$ given a set of parameters $\\boldsymbol{\\beta}$.\n",
+    "It is a conditional probability (see below) and reads as the\n",
+    "likelihood of a domain of events $\\boldsymbol{D}$ given a set of parameters\n",
+    "$\\boldsymbol{\\beta}$.\n",
     "\n",
     "\n",
     "In statistics, maximum likelihood estimation (MLE) is a method of\n",
@@ -3728,7 +3732,7 @@
     "We define also the likelihood for $X$ and $Y$ as $p(X)$ and $p(Y)$ respectively.\n",
     "The likelihood of a specific event $x_i$ (or $y_i$) is then written as $p(X=x_i)$ or just $p(x_i)=p_i$. \n",
     "\n",
-    "**Union of events is given by.**"
+    "The union of events is given by"
    ]
   },
   {
@@ -3744,7 +3748,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "**The product rule (aka joint probability) is given by.**"
+    "The product rule (aka joint probability) is given by"
    ]
   },
   {
@@ -3762,8 +3766,6 @@
    "source": [
     "where we read $p(X\\vert Y)$ as the likelihood of obtaining $X$ given $Y$.\n",
     "\n",
-    "\n",
-    "\n",
     "If we have independent events then $p(X,Y)=p(X)p(Y)$.\n",
     "\n",
     "\n",
@@ -3840,21 +3842,27 @@
     "the parameter space represented by $X$. This function is not\n",
     "necesseraly normalized and is normally called the likelihood function.\n",
     "\n",
-    "The function $p(X)$ on the right hand side is called the prior while the function on the left hand side is the called the posterior probability. The denominator on the right hand side serves as a normalization factor for the posterior distribution.\n",
+    "The function $p(X)$ on the right hand side is called the prior while\n",
+    "the function on the left hand side is the called the posterior\n",
+    "probability. The denominator on the right hand side serves as a\n",
+    "normalization factor for the posterior distribution.\n",
     "\n",
     "Let us try to illustrate Bayes' theorem through an example.\n",
     "\n",
     "\n",
-    "Let us suppose that you are undergoing a series of mammography scans in\n",
-    "order to rule out possible breast cancer cases.  We define the\n",
+    "Let us suppose that you are undergoing a series of mammography scans\n",
+    "in order to rule out possible breast cancer cases.  We define the\n",
     "sensitivity for a positive event by the variable $X$. It takes binary\n",
     "values with $X=1$ representing a positive event and $X=0$ being a\n",
     "negative event. We reserve $Y$ as a classification parameter for\n",
-    "either a negative or a positive breast cancer confirmation. (Short note on wordings: positive here means having breast cancer, although none of us would consider this being a  positive thing).\n",
+    "either a negative or a positive breast cancer confirmation. (Short\n",
+    "note on wordings: positive here means having breast cancer, although\n",
+    "none of us would consider this being a positive thing).\n",
     "\n",
     "We let $Y=1$ represent the the case of having breast cancer and $Y=0$ as not.\n",
     "\n",
-    "Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$, that is we have"
+    "Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$ (the numbers here are all made up),\n",
+    "that is we have"
    ]
   },
   {
@@ -3870,12 +3878,32 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "This obviously sounds  scary since many would conclude that if the test is positive, there is a likelihood of $80\\%$ for having cancer.\n",
-    "It is however not correct, as the following Bayesian analysis shows.\n",
+    "This obviously sounds scary since many would conclude that if the test\n",
+    "is positive, there is a likelihood of $80\\%$ for having cancer.  It is\n",
+    "however not correct, as the following Bayesian analysis shows. The correct question to pose is *what is the probability of having breast cancer in case of a positive test?*\n",
+    "We are thus interested in"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "p(Y=1\\vert X=1),\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "instead  of $p(X=1\\vert Y=1)$.\n",
     "\n",
     "\n",
-    "If we look at various national surveys on breast cancer, the general likelihood of developing breast cancer is a very small number.\n",
-    "Let us assume that the prior probability in the population as a whole is"
+    "\n",
+    "If we look at various national surveys on breast cancer, the general\n",
+    "likelihood of developing breast cancer is a very small number.  Let us\n",
+    "assume that the prior probability in the population as a whole is"
    ]
   },
   {
@@ -3891,7 +3919,8 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We need also to account for the fact that the test may produce a false positive result (false alarm). Let us here assume that we have"
+    "We need also to account for the fact that the test may produce a false\n",
+    "positive result (false alarm). Let us here assume that we have"
    ]
   },
   {
@@ -3907,15 +3936,38 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Using Bayes' theorem we can then find the posterior probability that the person has breast cancer in case of a positive test, that is we can compute"
+    "Using Bayes' theorem we can then find the posterior probability that\n",
+    "the person has breast cancer in case of a positive test, that is we\n",
+    "can compute"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "\n",
+    "
\n",
+    "\n",
     "$$\n",
-    "p(Y=1\\vert X=1)=\\frac{p(X=1\\vert Y=1)p(Y=1)}{p(X=1\\vert Y=1)p(Y=1)+p(X=1\\vert Y=0)p(Y=0)}=\\frac{0.8\\times 0.004}{0.8\\times 0.004+0.1\\times 0.996}=0.031.\n",
+    "\\begin{equation}\n",
+    "p(Y=1\\vert X=1)=\\frac{p(X=1\\vert Y=1)p(Y=1)}{p(X=1\\vert Y=1)p(Y=1)+p(X=1\\vert Y=0)p(Y=0)}= \n",
+    "\\label{_auto2} \\tag{2}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation} \n",
+    " \\frac{0.8\\times 0.004}{0.8\\times 0.004+0.1\\times 0.996}=0.031.\n",
+    "\\label{_auto3} \\tag{3}\n",
+    "\\end{equation}\n",
     "$$"
    ]
   },
@@ -3941,9 +3993,12 @@
     "parameters $\\boldsymbol{\\beta}$ using OLS, Ridge and Lasso regression.  For\n",
     "OLS, you will notice as function of the noise and polynomial degree,\n",
     "that the parameters $\\beta$ will fluctuate from order to order in the\n",
-    "polynomial fit and that for larger and larger polynomial degrees of freedom, the parameters will tend to increase in value for OLS.\n",
+    "polynomial fit and that for larger and larger polynomial degrees of\n",
+    "freedom, the parameters will tend to increase in value for OLS.\n",
     "\n",
-    "For Ridge and Lasso regression, the higher order parameters will typically be reduced, providing thereby less fluctuations from one order to another one."
+    "For Ridge and Lasso regression, the higher order parameters will\n",
+    "typically be reduced, providing thereby less fluctuations from one\n",
+    "order to another one."
    ]
   },
   {
@@ -4008,8 +4063,8 @@
     "    # Compute the MSE and print it\n",
     "    MSERidgePredict[i] = MSE(y_test,ypredictRidge)\n",
     "    MSELassoPredict[i] = MSE(y_test,ypredictLasso)\n",
-    "    print(lmb,RegRidge.coef_)\n",
-    "    print(lmb,RegLasso.coef_)\n",
+    "#    print(lmb,RegRidge.coef_)\n",
+    "#    print(lmb,RegLasso.coef_)\n",
     "# Now plot the results\n",
     "plt.figure()\n",
     "plt.plot(np.log10(lambdas), MSERidgePredict, 'b', label = 'MSE Ridge Test')\n",
@@ -4024,10 +4079,13 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "How can we understand this?  \n",
+    "How can we understand this?  **More text to be added**.\n",
     "\n",
     "\n",
     "\n",
+    "## Linking Bayes' Theorem with Ridge and Lasso Regression\n",
+    "\n",
+    "\n",
     "Using Bayes' theorem we can gain a better intuition about Ridge and Lasso regression. \n",
     "\n",
     "For ordinary least squares we postulated that the maximum likelihood for the doamin of events $\\boldsymbol{D}$ (one-dimensional case)"
diff --git a/doc/LectureNotes/_build/html/_sources/chapter3.ipynb b/doc/LectureNotes/_build/html/_sources/chapter3.ipynb
index d7f8222d5..b2933e098 100644
--- a/doc/LectureNotes/_build/html/_sources/chapter3.ipynb
+++ b/doc/LectureNotes/_build/html/_sources/chapter3.ipynb
@@ -575,196 +575,6 @@
     "plt.show()"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Various steps in cross-validation\n",
-    "\n",
-    "When the repetitive splitting of the data set is done randomly,\n",
-    "samples may accidently end up in a fast majority of the splits in\n",
-    "either training or test set. Such samples may have an unbalanced\n",
-    "influence on either model building or prediction evaluation. To avoid\n",
-    "this $k$-fold cross-validation structures the data splitting. The\n",
-    "samples are divided into $k$ more or less equally sized exhaustive and\n",
-    "mutually exclusive subsets. In turn (at each split) one of these\n",
-    "subsets plays the role of the test set while the union of the\n",
-    "remaining subsets constitutes the training set. Such a splitting\n",
-    "warrants a balanced representation of each sample in both training and\n",
-    "test set over the splits. Still the division into the $k$ subsets\n",
-    "involves a degree of randomness. This may be fully excluded when\n",
-    "choosing $k=n$. This particular case is referred to as leave-one-out\n",
-    "cross-validation (LOOCV). \n",
-    "\n",
-    "\n",
-    "* Define a range of interest for the penalty parameter.\n",
-    "\n",
-    "* Divide the data set into training and test set comprising samples $\\{1, \\ldots, n\\} \\setminus i$ and $\\{ i \\}$, respectively.\n",
-    "\n",
-    "* Fit the linear regression model by means of ridge estimation  for each $\\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\\boldsymbol{\\sigma}_{-i}^2(\\lambda)$, as"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "$$\n",
-    "\\begin{align*}\n",
-    "\\boldsymbol{\\beta}_{-i}(\\lambda) & =  ( \\boldsymbol{X}_{-i, \\ast}^{T}\n",
-    "\\boldsymbol{X}_{-i, \\ast} + \\lambda \\boldsymbol{I}_{pp})^{-1}\n",
-    "\\boldsymbol{X}_{-i, \\ast}^{T} \\boldsymbol{y}_{-i}\n",
-    "\\end{align*}\n",
-    "$$"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "* Evaluate the prediction performance of these models on the test set by $\\log\\{L[y_i, \\boldsymbol{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}$. Or, by the prediction error $|y_i - \\boldsymbol{X}_{i, \\ast} \\boldsymbol{\\beta}_{-i}(\\lambda)|$, the relative error, the error squared or the R2 score function.\n",
-    "\n",
-    "* Repeat the first three steps  such that each sample plays the role of the test set once.\n",
-    "\n",
-    "* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "$$\n",
-    "\\begin{align*}\n",
-    "\\frac{1}{n} \\sum_{i = 1}^n \\log\\{L[y_i, \\mathbf{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}.\n",
-    "\\end{align*}\n",
-    "$$"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "For the various values of $k$\n",
-    "\n",
-    "1. shuffle the dataset randomly.\n",
-    "\n",
-    "2. Split the dataset into $k$ groups.\n",
-    "\n",
-    "3. For each unique group:\n",
-    "\n",
-    "a. Decide which group to use as set for test data\n",
-    "\n",
-    "b. Take the remaining groups as a training data set\n",
-    "\n",
-    "c. Fit a model on the training set and evaluate it on the test set\n",
-    "\n",
-    "d. Retain the evaluation score and discard the model\n",
-    "\n",
-    "\n",
-    "5. Summarize the model using the sample of model evaluation scores\n",
-    "\n",
-    "The code here uses Ridge regression with cross-validation (CV)  resampling and $k$-fold CV in order to fit a specific polynomial."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {
-    "collapsed": false,
-    "editable": true
-   },
-   "outputs": [],
-   "source": [
-    "import numpy as np\n",
-    "import matplotlib.pyplot as plt\n",
-    "from sklearn.model_selection import KFold\n",
-    "from sklearn.linear_model import Ridge\n",
-    "from sklearn.model_selection import cross_val_score\n",
-    "from sklearn.preprocessing import PolynomialFeatures\n",
-    "\n",
-    "# A seed just to ensure that the random numbers are the same for every run.\n",
-    "# Useful for eventual debugging.\n",
-    "np.random.seed(3155)\n",
-    "\n",
-    "# Generate the data.\n",
-    "nsamples = 100\n",
-    "x = np.random.randn(nsamples)\n",
-    "y = 3*x**2 + np.random.randn(nsamples)\n",
-    "\n",
-    "## Cross-validation on Ridge regression using KFold only\n",
-    "\n",
-    "# Decide degree on polynomial to fit\n",
-    "poly = PolynomialFeatures(degree = 6)\n",
-    "\n",
-    "# Decide which values of lambda to use\n",
-    "nlambdas = 500\n",
-    "lambdas = np.logspace(-3, 5, nlambdas)\n",
-    "\n",
-    "# Initialize a KFold instance\n",
-    "k = 5\n",
-    "kfold = KFold(n_splits = k)\n",
-    "\n",
-    "# Perform the cross-validation to estimate MSE\n",
-    "scores_KFold = np.zeros((nlambdas, k))\n",
-    "\n",
-    "i = 0\n",
-    "for lmb in lambdas:\n",
-    "    ridge = Ridge(alpha = lmb)\n",
-    "    j = 0\n",
-    "    for train_inds, test_inds in kfold.split(x):\n",
-    "        xtrain = x[train_inds]\n",
-    "        ytrain = y[train_inds]\n",
-    "\n",
-    "        xtest = x[test_inds]\n",
-    "        ytest = y[test_inds]\n",
-    "\n",
-    "        Xtrain = poly.fit_transform(xtrain[:, np.newaxis])\n",
-    "        ridge.fit(Xtrain, ytrain[:, np.newaxis])\n",
-    "\n",
-    "        Xtest = poly.fit_transform(xtest[:, np.newaxis])\n",
-    "        ypred = ridge.predict(Xtest)\n",
-    "\n",
-    "        scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)\n",
-    "\n",
-    "        j += 1\n",
-    "    i += 1\n",
-    "\n",
-    "\n",
-    "estimated_mse_KFold = np.mean(scores_KFold, axis = 1)\n",
-    "\n",
-    "## Cross-validation using cross_val_score from sklearn along with KFold\n",
-    "\n",
-    "# kfold is an instance initialized above as:\n",
-    "# kfold = KFold(n_splits = k)\n",
-    "\n",
-    "estimated_mse_sklearn = np.zeros(nlambdas)\n",
-    "i = 0\n",
-    "for lmb in lambdas:\n",
-    "    ridge = Ridge(alpha = lmb)\n",
-    "\n",
-    "    X = poly.fit_transform(x[:, np.newaxis])\n",
-    "    estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)\n",
-    "\n",
-    "    # cross_val_score return an array containing the estimated negative mse for every fold.\n",
-    "    # we have to the the mean of every array in order to get an estimate of the mse of the model\n",
-    "    estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)\n",
-    "\n",
-    "    i += 1\n",
-    "\n",
-    "## Plot and compare the slightly different ways to perform cross-validation\n",
-    "\n",
-    "plt.figure()\n",
-    "\n",
-    "plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')\n",
-    "plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')\n",
-    "\n",
-    "plt.xlabel('log10(lambda)')\n",
-    "plt.ylabel('mse')\n",
-    "\n",
-    "plt.legend()\n",
-    "\n",
-    "plt.show()"
-   ]
-  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -1221,6 +1031,203 @@
     "plt.show()"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Cross-validation\n",
+    "\n",
+    "When the repetitive splitting of the data set is done randomly,\n",
+    "samples may accidently end up in a fast majority of the splits in\n",
+    "either training or test set. Such samples may have an unbalanced\n",
+    "influence on either model building or prediction evaluation. To avoid\n",
+    "this $k$-fold cross-validation structures the data splitting. The\n",
+    "samples are divided into $k$ more or less equally sized exhaustive and\n",
+    "mutually exclusive subsets. In turn (at each split) one of these\n",
+    "subsets plays the role of the test set while the union of the\n",
+    "remaining subsets constitutes the training set. Such a splitting\n",
+    "warrants a balanced representation of each sample in both training and\n",
+    "test set over the splits. Still the division into the $k$ subsets\n",
+    "involves a degree of randomness. This may be fully excluded when\n",
+    "choosing $k=n$. This particular case is referred to as leave-one-out\n",
+    "cross-validation (LOOCV). \n",
+    "\n",
+    "\n",
+    "* Define a range of interest for the penalty parameter.\n",
+    "\n",
+    "* Divide the data set into training and test set comprising samples $\\{1, \\ldots, n\\} \\setminus i$ and $\\{ i \\}$, respectively.\n",
+    "\n",
+    "* Fit the linear regression model by means of ridge estimation  for each $\\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\\boldsymbol{\\sigma}_{-i}^2(\\lambda)$, as"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\begin{align*}\n",
+    "\\boldsymbol{\\beta}_{-i}(\\lambda) & =  ( \\boldsymbol{X}_{-i, \\ast}^{T}\n",
+    "\\boldsymbol{X}_{-i, \\ast} + \\lambda \\boldsymbol{I}_{pp})^{-1}\n",
+    "\\boldsymbol{X}_{-i, \\ast}^{T} \\boldsymbol{y}_{-i}\n",
+    "\\end{align*}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "* Evaluate the prediction performance of these models on the test set by $\\log\\{L[y_i, \\boldsymbol{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}$. Or, by the prediction error $|y_i - \\boldsymbol{X}_{i, \\ast} \\boldsymbol{\\beta}_{-i}(\\lambda)|$, the relative error, the error squared or the R2 score function.\n",
+    "\n",
+    "* Repeat the first three steps  such that each sample plays the role of the test set once.\n",
+    "\n",
+    "* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\begin{align*}\n",
+    "\\frac{1}{n} \\sum_{i = 1}^n \\log\\{L[y_i, \\mathbf{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}.\n",
+    "\\end{align*}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "For the various values of $k$\n",
+    "\n",
+    "1. shuffle the dataset randomly.\n",
+    "\n",
+    "2. Split the dataset into $k$ groups.\n",
+    "\n",
+    "3. For each unique group:\n",
+    "\n",
+    "a. Decide which group to use as set for test data\n",
+    "\n",
+    "b. Take the remaining groups as a training data set\n",
+    "\n",
+    "c. Fit a model on the training set and evaluate it on the test set\n",
+    "\n",
+    "d. Retain the evaluation score and discard the model\n",
+    "\n",
+    "\n",
+    "5. Summarize the model using the sample of model evaluation scores\n",
+    "\n",
+    "The code here uses Ridge regression with cross-validation (CV)  resampling and $k$-fold CV in order to fit a specific polynomial."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "from sklearn.model_selection import KFold\n",
+    "from sklearn.linear_model import Ridge\n",
+    "from sklearn.model_selection import cross_val_score\n",
+    "from sklearn.preprocessing import PolynomialFeatures\n",
+    "\n",
+    "# A seed just to ensure that the random numbers are the same for every run.\n",
+    "# Useful for eventual debugging.\n",
+    "np.random.seed(3155)\n",
+    "\n",
+    "# Generate the data.\n",
+    "nsamples = 100\n",
+    "x = np.random.randn(nsamples)\n",
+    "y = 3*x**2 + np.random.randn(nsamples)\n",
+    "\n",
+    "## Cross-validation on Ridge regression using KFold only\n",
+    "\n",
+    "# Decide degree on polynomial to fit\n",
+    "poly = PolynomialFeatures(degree = 6)\n",
+    "\n",
+    "# Decide which values of lambda to use\n",
+    "nlambdas = 500\n",
+    "lambdas = np.logspace(-3, 5, nlambdas)\n",
+    "\n",
+    "# Initialize a KFold instance\n",
+    "k = 5\n",
+    "kfold = KFold(n_splits = k)\n",
+    "\n",
+    "# Perform the cross-validation to estimate MSE\n",
+    "scores_KFold = np.zeros((nlambdas, k))\n",
+    "\n",
+    "i = 0\n",
+    "for lmb in lambdas:\n",
+    "    ridge = Ridge(alpha = lmb)\n",
+    "    j = 0\n",
+    "    for train_inds, test_inds in kfold.split(x):\n",
+    "        xtrain = x[train_inds]\n",
+    "        ytrain = y[train_inds]\n",
+    "\n",
+    "        xtest = x[test_inds]\n",
+    "        ytest = y[test_inds]\n",
+    "\n",
+    "        Xtrain = poly.fit_transform(xtrain[:, np.newaxis])\n",
+    "        ridge.fit(Xtrain, ytrain[:, np.newaxis])\n",
+    "\n",
+    "        Xtest = poly.fit_transform(xtest[:, np.newaxis])\n",
+    "        ypred = ridge.predict(Xtest)\n",
+    "\n",
+    "        scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)\n",
+    "\n",
+    "        j += 1\n",
+    "    i += 1\n",
+    "\n",
+    "\n",
+    "estimated_mse_KFold = np.mean(scores_KFold, axis = 1)\n",
+    "\n",
+    "## Cross-validation using cross_val_score from sklearn along with KFold\n",
+    "\n",
+    "# kfold is an instance initialized above as:\n",
+    "# kfold = KFold(n_splits = k)\n",
+    "\n",
+    "estimated_mse_sklearn = np.zeros(nlambdas)\n",
+    "i = 0\n",
+    "for lmb in lambdas:\n",
+    "    ridge = Ridge(alpha = lmb)\n",
+    "\n",
+    "    X = poly.fit_transform(x[:, np.newaxis])\n",
+    "    estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)\n",
+    "\n",
+    "    # cross_val_score return an array containing the estimated negative mse for every fold.\n",
+    "    # we have to the the mean of every array in order to get an estimate of the mse of the model\n",
+    "    estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)\n",
+    "\n",
+    "    i += 1\n",
+    "\n",
+    "## Plot and compare the slightly different ways to perform cross-validation\n",
+    "\n",
+    "plt.figure()\n",
+    "\n",
+    "plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')\n",
+    "plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')\n",
+    "\n",
+    "plt.xlabel('log10(lambda)')\n",
+    "plt.ylabel('mse')\n",
+    "\n",
+    "plt.legend()\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "More examples of the application of cross-validation follow here."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
diff --git a/doc/LectureNotes/_build/html/_sources/chapter4.ipynb b/doc/LectureNotes/_build/html/_sources/chapter4.ipynb
index 7cd60a191..81b9d76c8 100644
--- a/doc/LectureNotes/_build/html/_sources/chapter4.ipynb
+++ b/doc/LectureNotes/_build/html/_sources/chapter4.ipynb
@@ -7,8 +7,6 @@
     "# Logistic Regression\n",
     "\n",
     "\n",
-    "[Video of Lecture](https://www.uio.no/studier/emner/matnat/fys/FYS-STK3155/h20/forelesningsvideoer/LectureSeptember18.mp4?vrtx=view-as-webpage)\n",
-    "\n",
     "\n",
     "## Logistic Regression\n",
     "\n",
@@ -911,7 +909,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Optimization, the central part of any Machine Learning algortithm\n",
+    "# Optimization, the central part of any Machine Learning algortithm\n",
     "\n",
     "Almost every problem in machine learning and data science starts with\n",
     "a dataset $X$, a model $g(\\beta)$, which is a function of the\n",
@@ -924,7 +922,6 @@
     "\n",
     "\n",
     "\n",
-    "## Revisiting our Logistic Regression case\n",
     "\n",
     "In our discussion on Logistic Regression we studied the \n",
     "case of\n",
@@ -953,7 +950,6 @@
     "where $\\boldsymbol{\\beta}$ are the weights we wish to extract from data, in our case $\\beta_0$ and $\\beta_1$. \n",
     "\n",
     "\n",
-    "## The equations to solve\n",
     "\n",
     "Our compact equations used a definition of a vector $\\boldsymbol{y}$ with $n$\n",
     "elements $y_i$, an $n\\times p$ matrix $\\boldsymbol{X}$ which contains the\n",
@@ -995,7 +991,6 @@
     "This defines what is called  the Hessian matrix.\n",
     "\n",
     "\n",
-    "## Solving using Newton-Raphson's method\n",
     "\n",
     "If we can set up these equations, Newton-Raphson's iterative method is normally the method of choice. It requires however that we can compute in an efficient way the  matrices that define the first and second derivatives. \n",
     "\n",
@@ -1036,9 +1031,6 @@
     "If we can compute these matrices, in particular the Hessian, the above is often the easiest method to implement. \n",
     "\n",
     "\n",
-    "\n",
-    "## Brief reminder on Newton-Raphson's method\n",
-    "\n",
     "Let us quickly remind ourselves how we derive the above method.\n",
     "\n",
     "Perhaps the most celebrated of all one-dimensional root-finding\n",
@@ -1050,7 +1042,6 @@
     "normally discourage the use of this method.\n",
     "\n",
     "\n",
-    "## The equations\n",
     "\n",
     "The Newton-Raphson formula consists geometrically of extending the\n",
     "tangent line at a current point until it crosses zero, then setting\n",
@@ -1125,8 +1116,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Simple geometric interpretation\n",
-    "\n",
     "The above is Newton-Raphson's method. It has a simple geometric\n",
     "interpretation, namely $x_{n+1}$ is the point where the tangent from\n",
     "$(x_n,f(x_n))$ crosses the $x$-axis.  Close to the solution,\n",
@@ -1141,7 +1130,6 @@
     "\n",
     "\n",
     "\n",
-    "## Extending to more than one variable\n",
     "\n",
     "Newton's method can be generalized to systems of several non-linear equations\n",
     "and variables. Consider the case with two equations"
@@ -1278,8 +1266,6 @@
     "we are always moving towards smaller function values, i.e a minimum.\n",
     "\n",
     "\n",
-    "## More on Steepest descent\n",
-    "\n",
     "The previous observation is the basis of the method of steepest\n",
     "descent, which is also referred to as just gradient descent (GD). One\n",
     "starts with an initial guess $\\mathbf{x}_0$ for a minimum of $F$ and\n",
@@ -1303,8 +1289,6 @@
     "the learning rate within the context of Machine Learning.\n",
     "\n",
     "\n",
-    "## The ideal\n",
-    "\n",
     "Ideally the sequence $\\{\\mathbf{x}_k \\}_{k=0}$ converges to a global\n",
     "minimum of the function $F$. In general we do not know if we are in a\n",
     "global or local minimum. In the special case when $F$ is a convex\n",
@@ -1325,7 +1309,6 @@
     "\n",
     "\n",
     "\n",
-    "## The sensitiveness of the gradient descent\n",
     "\n",
     "The gradient descent method \n",
     "is sensitive to the choice of learning rate $\\gamma_k$. This is due\n",
@@ -1356,13 +1339,10 @@
     "regular polygons (triangles, rectangles, pentagons, etc...).\n",
     "\n",
     "\n",
-    "## Convex function\n",
     "\n",
     "**Convex function**: Let $X \\subset \\mathbb{R}^n$ be a convex set. Assume that the function $f: X \\rightarrow \\mathbb{R}$ is continuous, then $f$ is said to be convex if $$f(tx_1 + (1-t)x_2) \\leq tf(x_1) + (1-t)f(x_2) $$ for all $x_1, x_2 \\in X$ and for all $t \\in [0,1]$. If $\\leq$ is replaced with a strict inequaltiy in the definition, we demand $x_1 \\neq x_2$ and $t\\in(0,1)$ then $f$ is said to be strictly convex. For a single variable function, convexity means that if you draw a straight line connecting $f(x_1)$ and $f(x_2)$, the value of the function on the interval $[x_1,x_2]$ is always below the line as illustrated below.\n",
     "\n",
     "\n",
-    "## Conditions on convex functions\n",
-    "\n",
     "In the following we state first and second-order conditions which\n",
     "ensures convexity of a function $f$. We write $D_f$ to denote the\n",
     "domain of $f$, i.e the subset of $R^n$ where $f$ is defined. For more\n",
@@ -1395,7 +1375,6 @@
     "This condition is particularly useful since it gives us an procedure for determining if the function under consideration is convex, apart from using the definition.\n",
     "\n",
     "\n",
-    "## More on convex functions\n",
     "\n",
     "The next result is of great importance to us and the reason why we are\n",
     "going on about convex functions. In machine learning we frequently\n",
@@ -1418,7 +1397,7 @@
     "This result means that if we know that the cost/loss function is convex and we are able to find a minimum, we are guaranteed that it is a global minimum.\n",
     "\n",
     "\n",
-    "## Some simple problems\n",
+    "### Some simple problems\n",
     "\n",
     "1. Show that $f(x)=x^2$ is convex for $x \\in \\mathbb{R}$ using the definition of convexity. Hint: If you re-write the definition, $f$ is convex if the following holds for all $x,y \\in D_f$ and any $\\lambda \\in [0,1]$ $\\lambda f(x)+(1-\\lambda)f(y)-f(\\lambda x + (1-\\lambda) y ) \\geq 0$.\n",
     "\n",
@@ -1443,13 +1422,6 @@
     "Using the definition of convexity, try to show that a function satisfying the properties above is convex (the third condition is not needed to show this).\n",
     "\n",
     "\n",
-    "\n",
-    "## Friday September 25\n",
-    "\n",
-    "[Video of Lecture](https://www.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/forelesningsvideoer/LectureSeptember25.mp4?vrtx=view-as-webpage) and [link to handwritten notes](https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/NotesSeptember25.pdf).\n",
-    "\n",
-    "\n",
-    "\n",
     "## Standard steepest descent\n",
     "\n",
     "\n",
@@ -1498,7 +1470,6 @@
     "When we have found the exact solution, $\\boldsymbol{r}=0$.\n",
     "\n",
     "\n",
-    "## Gradient method\n",
     "\n",
     "The residual is zero when we reach the minimum of the quadratic equation"
    ]
@@ -1520,9 +1491,6 @@
     "symmetric.  This defines also the Hessian and we want it to be  positive definite.  \n",
     "\n",
     "\n",
-    "\n",
-    "## Steepest descent  method\n",
-    "\n",
     "We denote the initial guess for $\\boldsymbol{x}$ as $\\boldsymbol{x}_0$. \n",
     "We can assume without loss of generality that"
    ]
@@ -1559,8 +1527,6 @@
     "instead.\n",
     "\n",
     "\n",
-    "\n",
-    "## Steepest descent  method\n",
     "One can show that the solution $\\boldsymbol{x}$ is also the unique minimizer of the quadratic form"
    ]
   },
@@ -1599,9 +1565,6 @@
     "$\\boldsymbol{x}_0=0$ it is equal $-\\boldsymbol{b}$.\n",
     "\n",
     "\n",
-    "\n",
-    "\n",
-    "## Final expressions\n",
     "We can compute the residual iteratively as"
    ]
   },
@@ -1678,13 +1641,6 @@
     "$$"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Steepest descent example"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -1848,10 +1804,6 @@
     "Two vectors are conjugate if they are orthogonal with respect to \n",
     "this inner product. Being conjugate is a symmetric relation: if $\\boldsymbol{s}$ is conjugate to $\\boldsymbol{t}$, then $\\boldsymbol{t}$ is conjugate to $\\boldsymbol{s}$.\n",
     "\n",
-    "\n",
-    "\n",
-    "\n",
-    "## Conjugate gradient method\n",
     "An example is given by the eigenvectors of the matrix"
    ]
   },
@@ -1868,13 +1820,8 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "which is zero unless $i=j$.\n",
+    "which is zero unless $i=j$. \n",
     "\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
-    "## Conjugate gradient method\n",
     "Assume now that we have a symmetric positive-definite matrix $\\boldsymbol{A}$ of size\n",
     "$n\\times n$. At each iteration $i+1$ we obtain the conjugate direction of a vector"
    ]
@@ -1910,7 +1857,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Conjugate gradient method\n",
     "The coefficients are given by"
    ]
   },
@@ -1959,8 +1905,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Conjugate gradient method and iterations\n",
-    "\n",
     "If we choose the conjugate vectors $\\boldsymbol{p}_k$ carefully, \n",
     "then we may not need all of them to obtain a good approximation to the solution \n",
     "$\\boldsymbol{x}$. \n",
@@ -2003,11 +1947,6 @@
    "source": [
     "instead.\n",
     "\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
-    "## Conjugate gradient method\n",
     "One can show that the solution $\\boldsymbol{x}$ is also the unique minimizer of the quadratic form"
    ]
   },
@@ -2047,11 +1986,6 @@
     "The other vectors in the basis will be conjugate to the gradient, \n",
     "hence the name conjugate gradient method.\n",
     "\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
-    "## Conjugate gradient method\n",
     "Let  $\\boldsymbol{r}_k$ be the residual at the $k$-th step:"
    ]
   },
@@ -2090,7 +2024,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Conjugate gradient method\n",
     "We can also  compute the residual iteratively as"
    ]
   },
@@ -2155,14 +2088,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Revisiting our first homework\n",
+    "## Revisiting our Linear Regression Solvers\n",
     "\n",
     "We will use linear regression as a case study for the gradient descent\n",
     "methods. Linear regression is a great test case for the gradient\n",
     "descent methods discussed in the lectures since it has several\n",
     "desirable properties such as:\n",
     "\n",
-    "1. An analytical solution (recall homework set 1).\n",
+    "1. An analytical solution.\n",
     "\n",
     "2. The gradient can be computed analytically.\n",
     "\n",
@@ -2221,8 +2154,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Gradient descent example\n",
-    "\n",
     "Let $\\mathbf{y} = (y_1,\\cdots,y_n)^T$, $\\mathbf{\\boldsymbol{y}} = (\\boldsymbol{y}_1,\\cdots,\\boldsymbol{y}_n)^T$ and $\\beta = (\\beta_0, \\beta_1)^T$\n",
     "\n",
     "It is convenient to write $\\mathbf{\\boldsymbol{y}} = X\\beta$ where $X \\in \\mathbb{R}^{100 \\times 2} $ is the design matrix given by (we keep the intercept here)"
@@ -2264,8 +2195,6 @@
     "and we want to find $\\beta$ such that $C(\\beta)$ is minimized.\n",
     "\n",
     "\n",
-    "## The derivative of the cost/loss function\n",
-    "\n",
     "Computing $\\partial C(\\beta) / \\partial \\beta_0$ and $\\partial C(\\beta) / \\partial \\beta_1$ we can show  that the gradient can be written as"
    ]
   },
@@ -2287,7 +2216,6 @@
     "where $X$ is the design matrix defined above.\n",
     "\n",
     "\n",
-    "## The Hessian matrix\n",
     "The Hessian matrix of $C(\\beta)$ is given by"
    ]
   },
@@ -2311,10 +2239,6 @@
     "\n",
     "\n",
     "\n",
-    "\n",
-    "\n",
-    "## Simple program\n",
-    "\n",
     "We can now write a program that minimizes $C(\\beta)$ using the gradient descent method with a constant learning rate $\\gamma$ according to"
    ]
   },
@@ -2339,9 +2263,8 @@
     "$\\beta= (X^TX)^{-1} X^T \\mathbf{y}$.\n",
     "\n",
     "\n",
-    "## Gradient Descent Example\n",
     "\n",
-    "Here our simple example"
+    "Here is our simple example"
    ]
   },
   {
@@ -2405,7 +2328,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## And a corresponding example using **scikit-learn**"
+    "Alternatively, we can use **Scikit-Learn** as done here"
    ]
   },
   {
@@ -2439,8 +2362,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Gradient descent and Ridge\n",
-    "\n",
     "We have also discussed Ridge regression where the loss function contains a regularized term given by the $L_2$ norm of $\\beta$,"
    ]
   },
@@ -2487,13 +2408,6 @@
     "$$"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Program example for gradient descent with Ridge Regression"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -2590,8 +2504,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Computation of gradients\n",
-    "\n",
     "This in turn means that the gradient can be\n",
     "computed as a sum over $i$-gradients"
    ]
@@ -2617,7 +2529,7 @@
     "$k=1,\\cdots,n/M$.\n",
     "\n",
     "\n",
-    "## SGD example\n",
+    "\n",
     "As an example, suppose we have $10$ data points $(\\mathbf{x}_1,\\cdots, \\mathbf{x}_{10})$ \n",
     "and we choose to have $M=5$ minibathces,\n",
     "then each minibatch contains two data points. In particular we have\n",
@@ -2648,8 +2560,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## The gradient step\n",
-    "\n",
     "Thus a gradient descent step now looks like"
    ]
   },
@@ -2671,10 +2581,7 @@
     "probability from $[1,n/M]$. An iteration over the number of\n",
     "minibathces (n/M) is commonly referred to as an epoch. Thus it is\n",
     "typical to choose a number of epochs and for each epoch iterate over\n",
-    "the number of minibatches, as exemplified in the code below.\n",
-    "\n",
-    "\n",
-    "## Simple example code"
+    "the number of minibatches, as exemplified in the code below."
    ]
   },
   {
@@ -2715,7 +2622,6 @@
     "all $n$ datapoints.\n",
     "\n",
     "\n",
-    "## When do we stop?\n",
     "\n",
     "A natural question is when do we stop the search for a new minimum?\n",
     "One possibility is to compute the full gradient after a given number\n",
@@ -2729,7 +2635,6 @@
     "gave the lowest value.\n",
     "\n",
     "\n",
-    "## Slightly different approach\n",
     "\n",
     "Another approach is to let the step length $\\gamma_j$ depend on the\n",
     "number of epochs in such a way that it becomes very small after a\n",
@@ -2783,7 +2688,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Program for stochastic gradient"
+    "### Program for stochastic gradient"
    ]
   },
   {
@@ -2865,7 +2770,922 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "**Challenge**: try to write a similar code for a Logistic Regression case."
+    "## Momentum based GD\n",
+    "\n",
+    "The stochastic gradient descent (SGD) is almost always used with a\n",
+    "*momentum* or inertia term that serves as a memory of the direction we\n",
+    "are moving in parameter space.  This is typically implemented as\n",
+    "follows"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\mathbf{v}_{t}=\\gamma \\mathbf{v}_{t-1}+\\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t) \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation} \n",
+    "\\boldsymbol{\\theta}_{t+1}= \\boldsymbol{\\theta}_t -\\mathbf{v}_{t},\n",
+    "\\label{_auto2} \\tag{3}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "where we have introduced a momentum parameter $\\gamma$, with\n",
+    "$0\\le\\gamma\\le 1$, and for brevity we dropped the explicit notation to\n",
+    "indicate the gradient is to be taken over a different mini-batch at\n",
+    "each step. We call this algorithm gradient descent with momentum\n",
+    "(GDM). From these equations, it is clear that $\\mathbf{v}_t$ is a\n",
+    "running average of recently encountered gradients and\n",
+    "$(1-\\gamma)^{-1}$ sets the characteristic time scale for the memory\n",
+    "used in the averaging procedure. Consistent with this, when\n",
+    "$\\gamma=0$, this just reduces down to ordinary SGD as discussed\n",
+    "earlier. An equivalent way of writing the updates is"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\Delta \\boldsymbol{\\theta}_{t+1} = \\gamma \\Delta \\boldsymbol{\\theta}_t -\\ \\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t),\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "where we have defined $\\Delta \\boldsymbol{\\theta}_{t}= \\boldsymbol{\\theta}_t-\\boldsymbol{\\theta}_{t-1}$.\n",
+    "\n",
+    "\n",
+    "\n",
+    "Let us try to get more intuition from these equations. It is helpful\n",
+    "to consider a simple physical analogy with a particle of mass $m$\n",
+    "moving in a viscous medium with drag coefficient $\\mu$ and potential\n",
+    "$E(\\mathbf{w})$. If we denote the particle's position by $\\mathbf{w}$,\n",
+    "then its motion is described by"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "m {d^2 \\mathbf{w} \\over dt^2} + \\mu {d \\mathbf{w} \\over dt }= -\\nabla_w E(\\mathbf{w}).\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can discretize this equation in the usual way to get"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "m { \\mathbf{w}_{t+\\Delta t}-2 \\mathbf{w}_{t} +\\mathbf{w}_{t-\\Delta t} \\over (\\Delta t)^2}+\\mu {\\mathbf{w}_{t+\\Delta t}- \\mathbf{w}_{t} \\over \\Delta t} = -\\nabla_w E(\\mathbf{w}).\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Rearranging this equation, we can rewrite this as"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\Delta \\mathbf{w}_{t +\\Delta t}= - { (\\Delta t)^2 \\over m +\\mu \\Delta t} \\nabla_w E(\\mathbf{w})+ {m \\over m +\\mu \\Delta t} \\Delta \\mathbf{w}_t.\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Notice that this equation is identical to previous one if we identify\n",
+    "the position of the particle, $\\mathbf{w}$, with the parameters\n",
+    "$\\boldsymbol{\\theta}$. This allows us to identify the momentum\n",
+    "parameter and learning rate with the mass of the particle and the\n",
+    "viscous drag as:"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\gamma= {m \\over m +\\mu \\Delta t }, \\qquad \\eta = {(\\Delta t)^2 \\over m +\\mu \\Delta t}.\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Thus, as the name suggests, the momentum parameter is proportional to\n",
+    "the mass of the particle and effectively provides inertia.\n",
+    "Furthermore, in the large viscosity/small learning rate limit, our\n",
+    "memory time scales as $(1-\\gamma)^{-1} \\approx m/(\\mu \\Delta t)$.\n",
+    "\n",
+    "Why is momentum useful? SGD momentum helps the gradient descent\n",
+    "algorithm gain speed in directions with persistent but small gradients\n",
+    "even in the presence of stochasticity, while suppressing oscillations\n",
+    "in high-curvature directions. This becomes especially important in\n",
+    "situations where the landscape is shallow and flat in some directions\n",
+    "and narrow and steep in others. It has been argued that first-order\n",
+    "methods (with appropriate initial conditions) can perform comparable\n",
+    "to more expensive second order methods, especially in the context of\n",
+    "complex deep learning models.\n",
+    "\n",
+    "These beneficial properties of momentum can sometimes become even more\n",
+    "pronounced by using a slight modification of the classical momentum\n",
+    "algorithm called Nesterov Accelerated Gradient (NAG).\n",
+    "\n",
+    "In the NAG algorithm, rather than calculating the gradient at the\n",
+    "current parameters, $\\nabla_\\theta E(\\boldsymbol{\\theta}_t)$, one\n",
+    "calculates the gradient at the expected value of the parameters given\n",
+    "our current momentum, $\\nabla_\\theta E(\\boldsymbol{\\theta}_t +\\gamma\n",
+    "\\mathbf{v}_{t-1})$. This yields the NAG update rule"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\mathbf{v}_{t}=\\gamma \\mathbf{v}_{t-1}+\\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t +\\gamma \\mathbf{v}_{t-1}) \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation} \n",
+    "\\boldsymbol{\\theta}_{t+1}= \\boldsymbol{\\theta}_t -\\mathbf{v}_{t}.\n",
+    "\\label{_auto3} \\tag{4}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "One of the major advantages of NAG is that it allows for the use of a larger learning rate than GDM for the same choice of $\\gamma$.\n",
+    "\n",
+    "\n",
+    "\n",
+    "In stochastic gradient descent, with and without momentum, we still\n",
+    "have to specify a schedule for tuning the learning rates $\\eta_t$\n",
+    "as a function of time.  As discussed in the context of Newton's\n",
+    "method, this presents a number of dilemmas. The learning rate is\n",
+    "limited by the steepest direction which can change depending on the\n",
+    "current position in the landscape. To circumvent this problem, ideally\n",
+    "our algorithm would keep track of curvature and take large steps in\n",
+    "shallow, flat directions and small steps in steep, narrow directions.\n",
+    "Second-order methods accomplish this by calculating or approximating\n",
+    "the Hessian and normalizing the learning rate by the\n",
+    "curvature. However, this is very computationally expensive for\n",
+    "extremely large models. Ideally, we would like to be able to\n",
+    "adaptively change the step size to match the landscape without paying\n",
+    "the steep computational price of calculating or approximating\n",
+    "Hessians.\n",
+    "\n",
+    "Recently, a number of methods have been introduced that accomplish\n",
+    "this by tracking not only the gradient, but also the second moment of\n",
+    "the gradient. These methods include AdaGrad, AdaDelta, RMS-Prop, and\n",
+    "ADAM.\n",
+    "\n",
+    "\n",
+    "### RMS prop\n",
+    "\n",
+    "In RMS prop, in addition to keeping a running average of the first\n",
+    "moment of the gradient, we also keep track of the second moment\n",
+    "denoted by $\\mathbf{s}_t=\\mathbb{E}[\\mathbf{g}_t^2]$. The update rule\n",
+    "for RMS prop is given by"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation}\n",
+    "\\mathbf{g}_t = \\nabla_\\theta E(\\boldsymbol{\\theta}) \n",
+    "\\label{_auto4} \\tag{5}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\mathbf{s}_t =\\beta \\mathbf{s}_{t-1} +(1-\\beta)\\mathbf{g}_t^2 \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_t - \\eta_t { \\mathbf{g}_t \\over \\sqrt{\\mathbf{s}_t +\\epsilon}}, \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "where $\\beta$ controls the averaging time of the second moment and is\n",
+    "typically taken to be about $\\beta=0.9$, $\\eta_t$ is a learning rate\n",
+    "typically chosen to be $10^{-3}$, and $\\epsilon\\sim 10^{-8} $ is a\n",
+    "small regularization constant to prevent divergences. Multiplication\n",
+    "and division by vectors is understood as an element-wise operation. It\n",
+    "is clear from this formula that the learning rate is reduced in\n",
+    "directions where the norm of the gradient is consistently large. This\n",
+    "greatly speeds up the convergence by allowing us to use a larger\n",
+    "learning rate for flat directions.\n",
+    "\n",
+    "### ADAM optimizer\n",
+    "\n",
+    "A related algorithm is the ADAM optimizer. In ADAM, we keep a running\n",
+    "average of both the first and second moment of the gradient and use\n",
+    "this information to adaptively change the learning rate for different\n",
+    "parameters. In addition to keeping a running average of the first and\n",
+    "second moments of the gradient\n",
+    "(i.e. $\\mathbf{m}_t=\\mathbb{E}[\\mathbf{g}_t]$ and\n",
+    "$\\mathbf{s}_t=\\mathbb{E}[\\mathbf{g}^2_t]$, respectively), ADAM\n",
+    "performs an additional bias correction to account for the fact that we\n",
+    "are estimating the first two moments of the gradient using a running\n",
+    "average (denoted by the hats in the update rule below). The update\n",
+    "rule for ADAM is given by (where multiplication and division are once\n",
+    "again understood to be element-wise operations below)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation}\n",
+    "\\mathbf{g}_t = \\nabla_\\theta E(\\boldsymbol{\\theta}) \n",
+    "\\label{_auto5} \\tag{6}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\mathbf{m}_t = \\beta_1 \\mathbf{m}_{t-1} + (1-\\beta_1) \\mathbf{g}_t \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\mathbf{s}_t =\\beta_2 \\mathbf{s}_{t-1} +(1-\\beta_2)\\mathbf{g}_t^2 \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\boldsymbol{\\mathbf{m}}_t={\\mathbf{m}_t \\over 1-\\beta_1^t} \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\boldsymbol{\\mathbf{s}}_t ={\\mathbf{s}_t \\over1-\\beta_2^t} \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_t - \\eta_t { \\boldsymbol{\\mathbf{m}}_t \\over \\sqrt{\\boldsymbol{\\mathbf{s}}_t} +\\epsilon}, \\nonumber\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "
\n",
+    "\n",
+    "$$\n",
+    "\\begin{equation} \n",
+    "\\label{_auto6} \\tag{7}\n",
+    "\\end{equation}\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "where $\\beta_1$ and $\\beta_2$ set the memory lifetime of the first and\n",
+    "second moment and are typically taken to be $0.9$ and $0.99$\n",
+    "respectively, and $\\eta$ and $\\epsilon$ are identical to RMSprop.\n",
+    "\n",
+    "Like in RMSprop, the effective step size of a parameter depends on the\n",
+    "magnitude of its gradient squared.  To understand this better, let us\n",
+    "rewrite this expression in terms of the variance\n",
+    "$\\boldsymbol{\\sigma}_t^2 = \\boldsymbol{\\mathbf{s}}_t -\n",
+    "(\\boldsymbol{\\mathbf{m}}_t)^2$. Consider a single parameter $\\theta_t$. The\n",
+    "update rule for this parameter is given by"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "\\Delta \\theta_{t+1}= -\\eta_t { \\boldsymbol{m}_t \\over \\sqrt{\\sigma_t^2 +  m_t^2 }+\\epsilon}.\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Practical tips\n",
+    "\n",
+    "* **Randomize the data when making mini-batches**. It is always important to randomly shuffle the data when forming mini-batches. Otherwise, the gradient descent method can fit spurious correlations resulting from the order in which data is presented.\n",
+    "\n",
+    "* **Transform your inputs**. Learning becomes difficult when our landscape has a mixture of steep and flat directions. One simple trick for minimizing these situations is to standardize the data by subtracting the mean and normalizing the variance of input variables. Whenever possible, also decorrelate the inputs. To understand why this is helpful, consider the case of linear regression. It is easy to show that for the squared error cost function, the Hessian of the cost function is just the correlation matrix between the inputs. Thus, by standardizing the inputs, we are ensuring that the landscape looks homogeneous in all directions in parameter space. Since most deep networks can be viewed as linear transformations followed by a non-linearity at each layer, we expect this intuition to hold beyond the linear case.\n",
+    "\n",
+    "* **Monitor the out-of-sample performance.** Always monitor the performance of your model on a validation set (a small portion of the training data that is held out of the training process to serve as a proxy for the test set. If the validation error starts increasing, then the model is beginning to overfit. Terminate the learning process. This *early stopping* significantly improves performance in many settings.\n",
+    "\n",
+    "* **Adaptive optimization methods don't always have good generalization.** Recent studies have shown that adaptive methods such as ADAM, RMSPorp, and AdaGrad tend to have poor generalization compared to SGD or SGD with momentum, particularly in the high-dimensional limit (i.e. the number of parameters exceeds the number of data points). Although it is not clear at this stage why these methods perform so well in training deep neural networks, simpler procedures like properly-tuned SGD may work as well or better in these applications.\n",
+    "\n",
+    "## Automatic differentiation\n",
+    "\n",
+    "[Automatic differentiation (AD)](https://en.wikipedia.org/wiki/Automatic_differentiation), \n",
+    "also called algorithmic\n",
+    "differentiation or computational differentiation,is a set of\n",
+    "techniques to numerically evaluate the derivative of a function\n",
+    "specified by a computer program. AD exploits the fact that every\n",
+    "computer program, no matter how complicated, executes a sequence of\n",
+    "elementary arithmetic operations (addition, subtraction,\n",
+    "multiplication, division, etc.) and elementary functions (exp, log,\n",
+    "sin, cos, etc.). By applying the chain rule repeatedly to these\n",
+    "operations, derivatives of arbitrary order can be computed\n",
+    "automatically, accurately to working precision, and using at most a\n",
+    "small constant factor more arithmetic operations than the original\n",
+    "program.\n",
+    "\n",
+    "Automatic differentiation is neither:\n",
+    "\n",
+    "* Symbolic differentiation, nor\n",
+    "\n",
+    "* Numerical differentiation (the method of finite differences).\n",
+    "\n",
+    "Symbolic differentiation can lead to inefficient code and faces the\n",
+    "difficulty of converting a computer program into a single expression,\n",
+    "while numerical differentiation can introduce round-off errors in the\n",
+    "discretization process and cancellation\n",
+    "\n",
+    "\n",
+    "\n",
+    "Python has tools for so-called **automatic differentiation**.\n",
+    "Consider the following example"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "f(x) = \\sin\\left(2\\pi x + x^2\\right)\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "which has the following derivative"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "$$\n",
+    "f'(x) = \\cos\\left(2\\pi x + x^2\\right)\\left(2\\pi + 2x\\right)\n",
+    "$$"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Using **autograd** we have"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "\n",
+    "# To do elementwise differentiation:\n",
+    "from autograd import elementwise_grad as egrad \n",
+    "\n",
+    "# To plot:\n",
+    "import matplotlib.pyplot as plt \n",
+    "\n",
+    "\n",
+    "def f(x):\n",
+    "    return np.sin(2*np.pi*x + x**2)\n",
+    "\n",
+    "def f_grad_analytic(x):\n",
+    "    return np.cos(2*np.pi*x + x**2)*(2*np.pi + 2*x)\n",
+    "\n",
+    "# Do the comparison:\n",
+    "x = np.linspace(0,1,1000)\n",
+    "\n",
+    "f_grad = egrad(f)\n",
+    "\n",
+    "computed = f_grad(x)\n",
+    "analytic = f_grad_analytic(x)\n",
+    "\n",
+    "plt.title('Derivative computed from Autograd compared with the analytical derivative')\n",
+    "plt.plot(x,computed,label='autograd')\n",
+    "plt.plot(x,analytic,label='analytic')\n",
+    "\n",
+    "plt.xlabel('x')\n",
+    "plt.ylabel('y')\n",
+    "plt.legend()\n",
+    "\n",
+    "plt.show()\n",
+    "\n",
+    "print(\"The max absolute difference is: %g\"%(np.max(np.abs(computed - analytic))))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here we\n",
+    "experiment with what kind of functions Autograd is capable\n",
+    "of finding the gradient of. The following Python functions are just\n",
+    "meant to illustrate what Autograd can do, but please feel free to\n",
+    "experiment with other, possibly more complicated, functions as well."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "\n",
+    "def f1(x):\n",
+    "    return x**3 + 1\n",
+    "\n",
+    "f1_grad = grad(f1)\n",
+    "\n",
+    "# Remember to send in float as argument to the computed gradient from Autograd!\n",
+    "a = 1.0\n",
+    "\n",
+    "# See the evaluated gradient at a using autograd:\n",
+    "print(\"The gradient of f1 evaluated at a = %g using autograd is: %g\"%(a,f1_grad(a)))\n",
+    "\n",
+    "# Compare with the analytical derivative, that is f1'(x) = 3*x**2 \n",
+    "grad_analytical = 3*a**2\n",
+    "print(\"The gradient of f1 evaluated at a = %g by finding the analytic expression is: %g\"%(a,grad_analytical))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "To differentiate with respect to two (or more) arguments of a Python\n",
+    "function, Autograd need to know at which variable the function if\n",
+    "being differentiated with respect to."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f2(x1,x2):\n",
+    "    return 3*x1**3 + x2*(x1 - 5) + 1\n",
+    "\n",
+    "# By sending the argument 0, Autograd will compute the derivative w.r.t the first variable, in this case x1\n",
+    "f2_grad_x1 = grad(f2,0)\n",
+    "\n",
+    "# ... and differentiate w.r.t x2 by sending 1 as an additional arugment to grad\n",
+    "f2_grad_x2 = grad(f2,1)\n",
+    "\n",
+    "x1 = 1.0\n",
+    "x2 = 3.0 \n",
+    "\n",
+    "print(\"Evaluating at x1 = %g, x2 = %g\"%(x1,x2))\n",
+    "print(\"-\"*30)\n",
+    "\n",
+    "# Compare with the analytical derivatives:\n",
+    "\n",
+    "# Derivative of f2 w.r.t x1 is: 9*x1**2 + x2:\n",
+    "f2_grad_x1_analytical = 9*x1**2 + x2\n",
+    "\n",
+    "# Derivative of f2 w.r.t x2 is: x1 - 5:\n",
+    "f2_grad_x2_analytical = x1 - 5\n",
+    "\n",
+    "# See the evaluated derivations:\n",
+    "print(\"The derivative of f2 w.r.t x1: %g\"%( f2_grad_x1(x1,x2) ))\n",
+    "print(\"The analytical derivative of f2 w.r.t x1: %g\"%( f2_grad_x1(x1,x2) ))\n",
+    "\n",
+    "print()\n",
+    "\n",
+    "print(\"The derivative of f2 w.r.t x2: %g\"%( f2_grad_x2(x1,x2) ))\n",
+    "print(\"The analytical derivative of f2 w.r.t x2: %g\"%( f2_grad_x2(x1,x2) ))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Note that the grad function will not produce the true gradient of the function. The true gradient of a function with two or more variables will produce a vector, where each element is the function differentiated w.r.t a variable."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f3(x): # Assumes x is an array of length 5 or higher\n",
+    "    return 2*x[0] + 3*x[1] + 5*x[2] + 7*x[3] + 11*x[4]**2\n",
+    "\n",
+    "f3_grad = grad(f3)\n",
+    "\n",
+    "x = np.linspace(0,4,5)\n",
+    "\n",
+    "# Print the computed gradient:\n",
+    "print(\"The computed gradient of f3 is: \", f3_grad(x))\n",
+    "\n",
+    "# The analytical gradient is: (2, 3, 5, 7, 22*x[4])\n",
+    "f3_grad_analytical = np.array([2, 3, 5, 7, 22*x[4]])\n",
+    "\n",
+    "# Print the analytical gradient:\n",
+    "print(\"The analytical gradient of f3 is: \", f3_grad_analytical)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Note that in this case, when sending an array as input argument, the\n",
+    "output from Autograd is another array. This is the true gradient of\n",
+    "the function, as opposed to the function in the previous example. By\n",
+    "using arrays to represent the variables, the output from Autograd\n",
+    "might be easier to work with, as the output is closer to what one\n",
+    "could expect form a gradient-evaluting function."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f4(x):\n",
+    "    return np.sqrt(1+x**2) + np.exp(x) + np.sin(2*np.pi*x)\n",
+    "\n",
+    "f4_grad = grad(f4)\n",
+    "\n",
+    "x = 2.7\n",
+    "\n",
+    "# Print the computed derivative:\n",
+    "print(\"The computed derivative of f4 at x = %g is: %g\"%(x,f4_grad(x)))\n",
+    "\n",
+    "# The analytical derivative is: x/sqrt(1 + x**2) + exp(x) + cos(2*pi*x)*2*pi\n",
+    "f4_grad_analytical = x/np.sqrt(1 + x**2) + np.exp(x) + np.cos(2*np.pi*x)*2*np.pi\n",
+    "\n",
+    "# Print the analytical gradient:\n",
+    "print(\"The analytical gradient of f4 at x = %g is: %g\"%(x,f4_grad_analytical))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f5(x):\n",
+    "    if x >= 0:\n",
+    "        return x**2\n",
+    "    else:\n",
+    "        return -3*x + 1\n",
+    "\n",
+    "f5_grad = grad(f5)\n",
+    "\n",
+    "x = 2.7\n",
+    "\n",
+    "# Print the computed derivative:\n",
+    "print(\"The computed derivative of f5 at x = %g is: %g\"%(x,f5_grad(x)))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "2\n",
+    "6\n",
+    " \n",
+    "<\n",
+    "<\n",
+    "<\n",
+    "!\n",
+    "!\n",
+    "C\n",
+    "O\n",
+    "D\n",
+    "E\n",
+    "_\n",
+    "B\n",
+    "L\n",
+    "O\n",
+    "C\n",
+    "K\n",
+    " \n",
+    " \n",
+    "p\n",
+    "y\n",
+    "c\n",
+    "o\n",
+    "d"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "# Both of the functions are implementation of the sum: sum(x**i) for i = 0, ..., 9\n",
+    "# The analytical derivative is: sum(i*x**(i-1)) \n",
+    "f6_grad_analytical = 0\n",
+    "for i in range(10):\n",
+    "    f6_grad_analytical += i*x**(i-1)\n",
+    "\n",
+    "print(\"The analytical derivative of f6 at x = %g is: %g\"%(x,f6_grad_analytical))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "\n",
+    "def f7(n): # Assume that n is an integer\n",
+    "    if n == 1 or n == 0:\n",
+    "        return 1\n",
+    "    else:\n",
+    "        return n*f7(n-1)\n",
+    "\n",
+    "f7_grad = grad(f7)\n",
+    "\n",
+    "n = 2.0\n",
+    "\n",
+    "print(\"The computed derivative of f7 at n = %d is: %g\"%(n,f7_grad(n)))\n",
+    "\n",
+    "# The function f7 is an implementation of the factorial of n.\n",
+    "# By using the product rule, one can find that the derivative is:\n",
+    "\n",
+    "f7_grad_analytical = 0\n",
+    "for i in range(int(n)-1):\n",
+    "    tmp = 1\n",
+    "    for k in range(int(n)-1):\n",
+    "        if k != i:\n",
+    "            tmp *= (n - k)\n",
+    "    f7_grad_analytical += tmp\n",
+    "\n",
+    "print(\"The analytical derivative of f7 at n = %d is: %g\"%(n,f7_grad_analytical))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Note that if n is equal to zero or one, Autograd will give an error message. This message appears when the output is independent on input.\n",
+    "\n",
+    "\n",
+    "Autograd supports many features. However, there are some functions that is not supported (yet) by Autograd.\n",
+    "\n",
+    "Assigning a value to the variable being differentiated with respect to"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f8(x): # Assume x is an array\n",
+    "    x[2] = 3\n",
+    "    return x*2\n",
+    "\n",
+    "f8_grad = grad(f8)\n",
+    "\n",
+    "x = 8.4\n",
+    "\n",
+    "print(\"The derivative of f8 is:\",f8_grad(x))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here, Autograd tells us that an 'ArrayBox' does not support item assignment. The item assignment is done when the program tries to assign x[2] to the value 3. However, Autograd has implemented the computation of the derivative such that this assignment is not possible."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f9(a): # Assume a is an array with 2 elements\n",
+    "    b = np.array([1.0,2.0])\n",
+    "    return a.dot(b)\n",
+    "\n",
+    "f9_grad = grad(f9)\n",
+    "\n",
+    "x = np.array([1.0,0.0])\n",
+    "\n",
+    "print(\"The derivative of f9 is:\",f9_grad(x))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here we are told that the 'dot' function does not belong to Autograd's\n",
+    "version of a Numpy array.  To overcome this, an alternative syntax\n",
+    "which also computed the dot product can be used:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "import autograd.numpy as np\n",
+    "from autograd import grad\n",
+    "def f9_alternative(x): # Assume a is an array with 2 elements\n",
+    "    b = np.array([1.0,2.0])\n",
+    "    return np.dot(x,b) # The same as x_1*b_1 + x_2*b_2\n",
+    "\n",
+    "f9_alternative_grad = grad(f9_alternative)\n",
+    "\n",
+    "x = np.array([3.0,0.0])\n",
+    "\n",
+    "print(\"The gradient of f9 is:\",f9_alternative_grad(x))\n",
+    "\n",
+    "# The analytical gradient of the dot product of vectors x and b with two elements (x_1,x_2) and (b_1, b_2) respectively\n",
+    "# w.r.t x is (b_1, b_2)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The documentation recommends to avoid inplace operations such as"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "a += b\n",
+    "a -= b\n",
+    "a*= b\n",
+    "a /=b"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "More examples will be added, in particular how to compare autograd with own codes for the gradients."
    ]
   }
  ],
diff --git a/doc/LectureNotes/_build/html/chapter1.html b/doc/LectureNotes/_build/html/chapter1.html
index 3e61ffcbd..2e7c0a670 100644
--- a/doc/LectureNotes/_build/html/chapter1.html
+++ b/doc/LectureNotes/_build/html/chapter1.html
@@ -159,7 +159,7 @@
  
  
  • 
   
-   7. Support Vector Machines, overarching aims
+   8. Support Vector Machines, overarching aims
   
  
    • 
 
@@ -171,12 +171,12 @@
 
@@ -188,12 +188,12 @@
 
@@ -205,12 +205,12 @@
 
@@ -357,39 +357,34 @@
    3.7. Housing data, the code
   
  
- 
  • 
-  
-   3.8. Splitting our Data in Training and Test data
-  
- 
    • 
  
  • 
   
-   3.9. Reducing the number of degrees of freedom, overarching view
+   3.8. Reducing the number of degrees of freedom, overarching view
   
  
    • 
  
  • 
   
-   3.10. Testing the Means Squared Error as function of Complexity
+   3.9. Testing the Means Squared Error as function of Complexity
   
  
    • 
  
  • 
   
-   3.11. Exercises
+   3.10. Exercises
   
   
@@ -409,35 +404,34 @@
 
    3. Linear Regression¶
    
 
    
 
    3.1. Introduction¶
    
-
    Our emphasis throughout this series of lectures
    
-is on understanding the mathematical aspects of
-different algorithms used in the fields of data analysis and machine learning.
    
-
    However, where possible we will emphasize the
-importance of using available software. We start thus with a hands-on
-and top-down approach to machine learning. The aim is thus to start with
-relevant data or data we have produced
-and use these to introduce statistical data analysis
-concepts and machine learning algorithms before we delve into the
-algorithms themselves. The examples we will use in the beginning, start with simple
-polynomials with random noise added. We will use the Python
-software package Scikit-Learn and
-introduce various machine learning algorithms to make fits of
-the data and predictions. We move thereafter to more interesting
-cases such as data from say experiments (below we will look at experimental nuclear binding energies as an example).
-These are examples where we can easily set up the data and
-then use machine learning algorithms included in for example
-Scikit-Learn.
    
+
    Our emphasis throughout this series of lectures is on understanding
+the mathematical aspects of different algorithms used in the fields of
+data analysis and machine learning.
    
+
    However, where possible we will emphasize the importance of using
+available software. We start thus with a hands-on and top-down
+approach to machine learning. The aim is thus to start with relevant
+data or data we have produced and use these to introduce statistical
+data analysis concepts and machine learning algorithms before we delve
+into the algorithms themselves. The examples we will use in the
+beginning, start with simple polynomials with random noise added. We
+will use the Python software package
+Scikit-Learn and introduce various
+machine learning algorithms to make fits of the data and
+predictions. We move thereafter to more interesting cases such as data
+from say experiments (below we will look at experimental nuclear
+binding energies as an example).  These are examples where we can
+easily set up the data and then use machine learning algorithms
+included in for example Scikit-Learn.
    
 
    These examples will serve us the purpose of getting
 started. Furthermore, they allow us to catch more than two birds with
 a stone. They will allow us to bring in some programming specific
 topics and tools as well as showing the power of various Python
 libraries for machine learning and statistical data analysis.
    
-
    Here, we will mainly focus on two
-specific Python packages for Machine Learning, Scikit-Learn and
-Tensorflow (see below for links etc).  Moreover, the examples we
-introduce will serve as inputs to many of our discussions later, as
-well as allowing you to set up models and produce your own data and
-get started with programming.
    
+
    Here, we will mainly focus on two specific Python packages for Machine
+Learning, Scikit-Learn and Tensorflow (see below for links etc).
+Moreover, the examples we introduce will serve as inputs to many of
+our discussions later, as well as allowing you to set up models and
+produce your own data and get started with programming.
    
 
    
 
    
 
    3.2. What is Machine Learning?¶
    
@@ -625,7 +619,15 @@ quality figures. Feel free to explore the extensive
 gallery of examples. In
 this example we plot our original values of \(x\) and \(y\) as well as the
 prediction ypredict (\(\tilde{y}\)), which attempts at fitting our
-data with a straight line.
    
+data with a straight line.  Note also that Scikit-Learn requires a
+matrix as input for the input values \(x\) and \(y\). In the above code we
+have solved this by declaring \(x\) and \(y\) as arrays of dimension
+\(n\times 1\).
    
+
    In the code here we have also made a new array for \(x\in [0,1]\). Our
+prediction is computed for these values, meaning that they were not
+included in the data set used to train (or fit) the model.
+This is a recurrring theme in machine learning and data analysis. We would like to train a model on a specific given data set.
+Thereafter we wish to apply it to data which were not included in the training. Below we will encounter this again in the so-called train-validate-test spliting. We will typically split our data into different sets, oen for training, one for validation and finally, our data from the untouched test vault!
    
 
    The Python code follows here.
    
 
    
 
    
@@ -640,6 +642,7 @@ data with a straight line.
    
 y = 2*x+np.random.randn(100,1)
 linreg = LinearRegression()
 linreg.fit(x,y)
+# This is our new x-array to which we test our model
 xnew = np.array([[0],[1]])
 ypredict = linreg.predict(xnew)
 
@@ -745,8 +748,7 @@ to be dominated by outliers.
    
 
    Depending on the parameter in front of the normal distribution, we may
 have a small or larger relative error. Try to play around with
 different training data sets and study (graphically) the value of the
-relative error. Note also that Scikit-Learn requires a matrix as input for the input values \(x\) and \(y\). In the above code we have
-solved this by declaring \(x\) and \(y\) as arrays of dimension \(n\times 1\).
    
+relative error.
    
 
    As mentioned above, Scikit-Learn has an impressive functionality.
 We can for example extract the values of \(\alpha\) and \(\beta\) and
 their error estimates, or the variance and standard deviation and many
@@ -787,13 +789,13 @@ example of the functionality of Scikit-Learn.
    
 
    
 
    
 
    The intercept alpha: 
- [1.84860939]
+ [1.89992521]
 Coefficient beta : 
- [[5.2571699]]
-Mean squared error: 0.25
-Variance score: 0.90
+ [[5.12908963]]
+Mean squared error: 0.20
+Variance score: 0.92
 Mean squared log error: 0.01
-Mean absolute error: 0.40
+Mean absolute error: 0.36
 
    
 
    
 _images/chapter1_13_1.png
@@ -893,7 +895,7 @@ a linear \(x\)-dependence we s
 
    
 
    
 _images/chapter1_27_0.png
-
    0.0050000000000000044
+
    0.005000000000000007
 
    
 
    
 
    
@@ -1117,7 +1119,7 @@ A
 
    
 
    
 
    The next step, and we will define this mathematically later, is to set up the so-called design matrix. We will throughout call this matrix \(\boldsymbol{X}\).
-It has dimensionality \(p\times n\), where \(n\) is the number of data points and \(p\) are the so-called predictors. In our case here they are given by the number of polynomials in \(A\) we wish to include in the fit.
    
+It has dimensionality \(n\times p\), where \(n\) is the number of data points and \(p\) are the so-called predictors. In our case here they are given by the number of polynomials in \(A\) we wish to include in the fit.
    
 
    
 
    
 
    # Now we set up the design matrix X
@@ -1131,8 +1133,11 @@ It has dimensionality \(p\times n\)
 
    
 
    
-
    Note well that we have made life simple here. We perform a fit in terms of the number of nucleons only. A more sophisticated fit can be done by including an explicit dependence on the number of protons and neutrons in the asymmetry and Coulomb terms.
    
-
    With scikitlearn we are now ready to use linear regression and fit our data.
    
+
    Note well that we have made life simple here. We perform a fit in
+terms of the number of nucleons only.  A more sophisticated fit can be
+done by including an explicit dependence on the number of protons and
+neutrons in the asymmetry and Coulomb terms. We leave this as an exercise to you the reader.
    
+
    With Scikit-Learn we are now ready to use linear regression and fit our data.
    
 
    
 
    
 
    clf = skl.LinearRegression().fit(X, Energies)
@@ -1176,7 +1181,7 @@ Mean absolute error: 0.05
 _images/chapter1_57_1.png
 
    
 
    
-
    As a teaser, let us now see how we can do this with decision trees using scikit-learn. Later we will switch to so-called random forests!
    
+
    As a teaser, let us now see how we can do this with decision trees using Scikit-Learn. Later we will switch to so-called random forests!
    
 
    
 
    
 
    #Decision Tree Regression
@@ -1228,11 +1233,20 @@ A
 270 3344  160  110  270      Ds  7.253775  7.253775
 
 [267 rows x 6 columns]
-0.009883615646716186
+0.009883615646716188
 
    
 
    
 
    
 
    
+
    With a deeper and deeper tree level, we can almost reproduce every
+single data point by increasing the max depth of the tree.
+We can actually decide to make a decision tree which fits every single point.
+As we will
+see later, this has the benefit that we can really train a model which
+traverses every single data point. However, the price we pay is that
+we will easily overfit. That is, if we apply our model to unseen data,
+we will most likely fail miserably in our attempt at making
+predictions. As an exercise, try to make the tree level larger by adjusting the maximum depth variable. When printing out the predicition, you will note that the binding energy of every nucleus is accurately reproduced.
    
 
    The seaborn package allows us to visualize data in an efficient way. Note that we use scikit-learn’s multi-layer perceptron (or feed forward neural network)
 functionality.
    
 
    
@@ -1270,14 +1284,6 @@ functionality.
    
 
    
 
    
 
    
-
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
-  warnings.warn(
-/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
-  warnings.warn(
-/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
-  warnings.warn(
-
    
-
    
 
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
   warnings.warn(
 /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
@@ -1334,16 +1340,12 @@ functionality.
    
   warnings.warn(
 /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
   warnings.warn(
+
    
+
    
+
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
+  warnings.warn(
 /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
   warnings.warn(
-
    
-
    
-
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
-  warnings.warn(
-
    
-
    
-
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
-  warnings.warn(
 /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
   warnings.warn(
 
    
@@ -1360,7 +1362,11 @@ functionality.
    
   warnings.warn(
    • -_images/chapter1_61_11.png +
    /Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.
+  warnings.warn(
+
    +
    +_images/chapter1_61_10.png
    @@ -1878,6 +1884,22 @@ K

    \[ \frac{\partial\log{\vert\boldsymbol{A}\vert}}{\partial \boldsymbol{A}}=(\boldsymbol{A}^{-1})^T. \]
    +

    We can then compute the second derivative of the cost function, which in our case is the second derivative +of the means squared error. This leads to

    +
    +\[ +\frac{\partial^2 C(\boldsymbol{\beta})}{\partial \boldsymbol{\beta}^T\partial \boldsymbol{\beta}} =\frac{2}{n}\boldsymbol{X}^T\boldsymbol{X}. +\]
    +

    This quantity defines was what is called the Hessian matrix (the second derivative of a function we want to optimize).

    +

    The Hessian matrix plays an important role and is defined for the mean squared error as

    +
    +\[ +\boldsymbol{H}=\boldsymbol{X}^T\boldsymbol{X}. +\]
    +

    The Hessian matrix for ordinary least squares is also proportional to +the covariance matrix. As we will see in the chapter on Ridge and Lasso regression, This means that we can use the Singular Value Decomposition of a matrix to find +the eigenvalues of the covariance matrix and the Hessian matrix in +terms of the singular values.

    The residuals \(\boldsymbol{\epsilon}\) are in turn given by

    \[ @@ -1935,7 +1957,7 @@ write

    -_images/chapter1_132_0.png +_images/chapter1_136_0.png

    We can easily test our fit by computing the \(R2\) score that we discussed in connection with the functionality of Scikit-Learn in the introductory slides. @@ -2138,9 +2160,7 @@ time the file is presented as a standard csv file.

    before, with the same initializations and declarations. We use also pandas again, rather extensively in order to organize our data.

    The difference now is that we use Scikit-Learn’s regression tools -instead of our own matrix inversion implementation. Furthermore, we -sneak in Ridge regression (to be discussed below) which includes a -hyperparameter \(\lambda\), also to be explained below.

    +instead of our own matrix inversion implementation.

    # Common imports
@@ -2203,18 +2223,6 @@ hyperparameter \(\lambda\), al
 print('Mean absolute error: %.2f' % mean_absolute_error(Energies, ytilde))
 print(clf.coef_, clf.intercept_)
 
-# The Ridge regression with a hyperparameter lambda = 0.1
-_lambda = 0.1
-clf_ridge = skl.Ridge(alpha=_lambda).fit(X, Energies)
-yridge = clf_ridge.predict(X)
-EoS['Eridge']  = yridge
-# The mean squared error                               
-print("Mean squared error: %.2f" % mean_squared_error(Energies, yridge))
-# Explained variance score: 1 is perfect prediction                                 
-print('Variance score: %.2f' % r2_score(Energies, yridge))
-# Mean absolute error                                                           
-print('Mean absolute error: %.2f' % mean_absolute_error(Energies, yridge))
-print(clf_ridge.coef_, clf_ridge.intercept_)
 
 fig, ax = plt.subplots()
 ax.set_xlabel(r'$\rho[\mathrm{fm}^{-3}]$')
@@ -2223,8 +2231,6 @@ hyperparameter \(\lambda\), al
             label='Theoretical data')
 ax.plot(EoS['Density'], EoS['Eols'], alpha=0.7, lw=2, c='m',
             label='OLS')
-ax.plot(EoS['Density'], EoS['Eridge'], alpha=0.7, lw=2, c='g',
-            label='Ridge $\lambda = 0.1$')
 ax.legend()
 save_fig("EoSfitting")
 plt.show()
@@ -2236,20 +2242,13 @@ hyperparameter \(\lambda\), al
 Variance score: 1.00
 Mean absolute error: 2.83
 [   0.          618.32047562 -861.13519106 1404.91549644] -11.057088709963637
-Mean squared error: 197.93
-Variance score: 1.00
-Mean absolute error: 11.69
-[  0.          28.18220995 282.79902342 842.30879705] 12.946893955211749
    -_images/chapter1_179_1.png +_images/chapter1_183_1.png

    The above simple polynomial in density \(\rho\) gives an excellent fit to the data.

    -

    We note also that there is a small deviation between the -standard OLS and the Ridge regression at higher densities. We discuss this in more detail -below.

    @@ -2264,6 +2263,95 @@ postpone a discussion of this splitting to the end of these notes and our discussion of the so-called bias-variance tradeoff. Here we limit ourselves to repeat the above equation of state fitting example but now splitting the data into a training set and a test set.

    +

    Let us study some examples. The first code here takes a simple +one-dimensional second-order polynomial and we fit it to a +second-order polynomial. Depending on the strength of the added noise, +the various measures like the \(R2\) score or the mean-squared error, +the fit becomes better or worse.

    +
    +
    +
    import os
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+
+
+def R2(y_data, y_model):
+    return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)
+def MSE(y_data,y_model):
+    n = np.size(y_model)
+    return np.sum((y_data-y_model)**2)/n
+
+x = np.random.rand(100)
+y = 2.0+5*x*x+0.1*np.random.randn(100)
+
+
+#  The design matrix now as function of a given polynomial
+X = np.zeros((len(x),3))
+X[:,0] = 1.0
+X[:,1] = x
+X[:,2] = x**2
+# We split the data in test and training data
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
+# matrix inversion to find beta
+beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train
+print(beta)
+# and then make the prediction
+ytilde = X_train @ beta
+print("Training R2")
+print(R2(y_train,ytilde))
+print("Training MSE")
+print(MSE(y_train,ytilde))
+ypredict = X_test @ beta
+print("Test R2")
+print(R2(y_test,ypredict))
+print("Test MSE")
+print(MSE(y_test,ypredict))
+
    +
    +
    +
    +
    [ 2.00247083e+00 -3.54710530e-03  5.02096350e+00]
+Training R2
+0.99620088364924
+Training MSE
+0.008171076530362356
+Test R2
+0.9968966158779216
+Test MSE
+0.006530396683779109
+
    +
    +
    +
    +

    Alternatively, you could write your own test-train splitting function as shown here.

    +
    +
    +
    # equivalently in numpy
+def train_test_split_numpy(inputs, labels, train_size, test_size):
+    n_inputs = len(inputs)
+    inputs_shuffled = inputs.copy()
+    labels_shuffled = labels.copy()
+
+    np.random.shuffle(inputs_shuffled)
+    np.random.shuffle(labels_shuffled)
+
+    train_end = int(n_inputs*train_size)
+    X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]
+    Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]
+
+    return X_train, X_test, Y_train, Y_test
+
    +
    +
    +
    +

    But since scikit-learn has its own function for doing this and since +it interfaces easily with tensorflow and other libraries, we +normally recommend using the latter functionality.

    +

    As another example, we apply the training and testing split to +to the above equation of state fitting example +but now splitting the data into a training set and a test set.

    import os
@@ -2335,13 +2423,13 @@ but now splitting the data into a training set and a test set.
    
 
    Training R2
-0.9999869956119286
+0.9999850282256434
 Training MSE
-5.745136489050356
+6.653725417896576
 Test R2
-0.9999787681537219
+0.9999887274726137
 Test MSE
-9.712199063818309
+5.071447584032141
    @@ -2460,7 +2548,7 @@ dtype: int64 warnings.warn(msg, FutureWarning)
    -_images/chapter1_191_1.png +_images/chapter1_199_1.png

    It is now useful to look at the correlation matrix

    @@ -2478,7 +2566,7 @@ dtype: int64 
    <AxesSubplot:>
    -_images/chapter1_193_1.png +_images/chapter1_201_1.png

    From the above coorelation plot we can see that MEDV is strongly correlated to LSTAT and RM. We see also that RAD and TAX are stronly correlated, but we don’t include this in our features together to avoid multi-colinearity

    @@ -2501,7 +2589,7 @@ dtype: int64
    -_images/chapter1_195_0.png +_images/chapter1_203_0.png

    Now we start training our model

    @@ -2599,111 +2687,12 @@ R2 score is 0.6628996975186952
    -_images/chapter1_202_0.png +_images/chapter1_210_0.png
    -
    -

    3.8. Splitting our Data in Training and Test data¶

    -

    It is normal in essentially all Machine Learning studies to split the -data in a training set and a test set (sometimes also an additional -validation set). Scikit-Learn has an own function for this. There -is no explicit recipe for how much data should be included as training -data and say test data. An accepted rule of thumb is to use -approximately \(2/3\) to \(4/5\) of the data as training data. We will -postpone a discussion of this splitting to the end of these notes and -our discussion of the so-called bias-variance tradeoff. Here we -limit ourselves to repeat the above equation of state fitting example -but now splitting the data into a training set and a test set.

    -

    Let us study some examples. The first code here takes a simple -one-dimensional second-order polynomial and we fit it to a -second-order polynomial. Depending on the strength of the added noise, -the various measures like the \(R2\) score or the mean-squared error, -the fit becomes better or worse.

    -
    -
    -
    import os
-import numpy as np
-import pandas as pd
-import matplotlib.pyplot as plt
-from sklearn.model_selection import train_test_split
-
-
-def R2(y_data, y_model):
-    return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)
-def MSE(y_data,y_model):
-    n = np.size(y_model)
-    return np.sum((y_data-y_model)**2)/n
-
-x = np.random.rand(100)
-y = 2.0+5*x*x+0.1*np.random.randn(100)
-
-
-#  The design matrix now as function of a given polynomial
-X = np.zeros((len(x),3))
-X[:,0] = 1.0
-X[:,1] = x
-X[:,2] = x**2
-# We split the data in test and training data
-X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
-# matrix inversion to find beta
-beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train
-print(beta)
-# and then make the prediction
-ytilde = X_train @ beta
-print("Training R2")
-print(R2(y_train,ytilde))
-print("Training MSE")
-print(MSE(y_train,ytilde))
-ypredict = X_test @ beta
-print("Test R2")
-print(R2(y_test,ypredict))
-print("Test MSE")
-print(MSE(y_test,ypredict))
-
    -
    -
    -
    -
    [ 2.05802766 -0.30626706  5.27262964]
-Training R2
-0.9967292151090247
-Training MSE
-0.007931713723267314
-Test R2
-0.9966744029509663
-Test MSE
-0.006968283076248407
-
    -
    -
    -
    -

    Alternatively, you could write your own test-train splitting function as shown here.

    -
    -
    -
    # equivalently in numpy
-def train_test_split_numpy(inputs, labels, train_size, test_size):
-    n_inputs = len(inputs)
-    inputs_shuffled = inputs.copy()
-    labels_shuffled = labels.copy()
-
-    np.random.shuffle(inputs_shuffled)
-    np.random.shuffle(labels_shuffled)
-
-    train_end = int(n_inputs*train_size)
-    X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]
-    Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]
-
-    return X_train, X_test, Y_train, Y_test
-
    -
    -
    -
    -

    But since scikit-learn has its own function for doing this and since -it interfaces easily with tensorflow and other libraries, we -normally recommend using the latter functionality.

    -
    -

    3.9. Reducing the number of degrees of freedom, overarching view¶

    +

    3.8. Reducing the number of degrees of freedom, overarching view¶

    Many Machine Learning problems involve thousands or even millions of features for each training instance. Not only does this make training extremely slow, it can also make it much harder to find a good @@ -3156,7 +3145,7 @@ x_j^{(i)} \rightarrow (b-a)\frac{x_j^{(i)} - \min(x_j)}{\max(x_j) - \min(x_j)} -

    where \(\min(x_j)\) and \(\max(x_j)\) return the minimum and maximum value of \(x_j\) over the data set, respectively.

    -

    3.10. Testing the Means Squared Error as function of Complexity¶

    +

    3.9. Testing the Means Squared Error as function of Complexity¶

    Before we proceed with a more detailed analysis of the so-called Bias-Variance tradeoff, we present here an example of the relation between model complexity and the mean squared error for the triaining @@ -3208,14 +3197,14 @@ ourmodel (here in terms of the polynomial degree of the model).

    -_images/chapter1_214_0.png +_images/chapter1_218_0.png
    -

    3.11. Exercises¶

    +

    3.10. Exercises¶

    -

    3.11.1. Exercise: Setting up various Python environments¶

    +

    3.10.1. Exercise: Setting up various Python environments¶

    The first exercise here is of a mere technical art. We want you to have

    • git as a version control software and to establish a user account on a provider like GitHub. Other providers like GitLab etc are equally fine. You can also use the University of Oslo GitHub facilities.

      • @@ -3271,7 +3260,7 @@ license.

      We recommend using Anaconda if you are not too familiar with setting paths in a terminal environment.

    -

    3.11.2. Exercise: making your own data and exploring scikit-learn¶

    +

    3.10.2. Exercise: making your own data and exploring scikit-learn¶

    We will generate our own dataset for a function \(y(x)\) where \(x \in [0,1]\) and defined by random numbers computed with the uniform distribution. The function \(y\) is a quadratic polynomial in \(x\) with added stochastic noise according to the normal distribution \(\cal {N}(0,1)\). The following simple Python instructions define our \(x\) and \(y\) values (with 100 data points).

    @@ -3307,7 +3296,7 @@ R^2(\boldsymbol{y}, \tilde{\boldsymbol{y}}) = 1 - \frac{\sum_{i=0}^{n - 1} (y_i Discuss the meaning of these results. Try also to vary the coefficient in front of the added stochastic noise term and discuss the quality of the fits.

    -

    3.11.3. Exercise: Normalizing our data¶

    +

    3.10.3. Exercise: Normalizing our data¶

    A much used approach before starting to train the data is to preprocess our data. Normally the data may need a rescaling and/or may be sensitive to extreme values. Scaling the data renders our inputs much more diff --git a/doc/LectureNotes/_build/html/chapter10.html b/doc/LectureNotes/_build/html/chapter10.html index f570beb9f..75098e166 100644 --- a/doc/LectureNotes/_build/html/chapter10.html +++ b/doc/LectureNotes/_build/html/chapter10.html @@ -5,7 +5,7 @@ - 13. Building a Feed Forward Neural Network — Applied Data Analysis and Machine Learning + 14. Building a Feed Forward Neural Network — Applied Data Analysis and Machine Learning @@ -54,7 +54,7 @@ - + @@ -138,17 +138,17 @@

    @@ -170,12 +170,12 @@ @@ -187,12 +187,12 @@ @@ -204,12 +204,12 @@ @@ -285,151 +285,151 @@ @@ -444,7 +444,7 @@
    -

    13. Building a Feed Forward Neural Network¶

    +

    14. Building a Feed Forward Neural Network¶

    We are now gong to develop an example based on the MNIST data base. This is a classification problem and we need to use our cross-entropy function we discussed in connection with logistic @@ -472,7 +472,7 @@ P(y = 1 \mid \hat{x}, \hat{\theta}) = 1 - P(y = 0 \mid \hat{x}, \hat{\theta}) ,

    where \(y \in \{0, 1\}\) and \(\hat{\theta}\) represents the weights and biases of our network.

    -

    13.1. Defining the cost function¶

    +

    14.1. Defining the cost function¶

    Our cost function is given as (see the Logistic regression lectures)

    \[ @@ -510,7 +510,7 @@ P(\mathcal{D} \mid \hat{\theta}) = \prod_{i=1}^n \prod_{c=0}^{C-1} [P(y_{ic} = 1

    See the logistic regression lectures for a full definition of the cost function.

    The back propagation equations need now only a small change, namely the definition of a new cost function. We are thus ready to use the same equations as before!

    -

    13.1.1. Example: binary classification problem¶

    +

    14.1.1. Example: binary classification problem¶

    As an example of the above, relevant for project 2 as well, let us consider a binary class. As discussed in our logistic regression lectures, we defined a cost function in terms of the parameters \(\beta\) as

    \[ @@ -552,7 +552,7 @@ Our cost function at the final layer

    In case we use another activation function than the logistic one, we need to evaluate other derivatives.

    -

    13.1.2. The Softmax function¶

    +

    14.1.2. The Softmax function¶

    In case we employ the more general case given by the Softmax equation, we need to evaluate the derivative of the activation function with respect to the activation \(z_i^l\), that is we need

    \[ @@ -573,7 +573,7 @@ f(z_i^l) = \frac{\exp{(z_i^l)}}{\sum_{m=1}^K\exp{(z_m^l)}}.
    -

    13.2. Developing a code for doing neural networks with back propagation¶

    +

    14.2. Developing a code for doing neural networks with back propagation¶

    One can identify a set of key steps when using neural networks to solve supervised learning problems:

    1. Collect and pre-process data

      2. @@ -584,7 +584,7 @@ f(z_i^l) = \frac{\exp{(z_i^l)}}{\sum_{m=1}^K\exp{(z_m^l)}}.

    2. Adjust hyperparameters (if necessary, network architecture)

    -

    13.2.1. Collect and pre-process data¶

    +

    14.2.1. Collect and pre-process data¶

    Here we will be using the MNIST dataset, which is readily available through the scikit-learn package. You may also find it for example here.
    The MNIST (Modified National Institute of Standards and Technology) database is a large database @@ -679,7 +679,7 @@ X = (n_inputs, n_features) = (1797, 64)

    -

    13.2.2. Train and test datasets¶

    +

    14.2.2. Train and test datasets¶

    Performing analysis before partitioning the dataset is a major error, that can lead to incorrect conclusions.

    We will reserve \(80 \%\) of our dataset for training and \(20 \%\) for testing.

    It is important that the train and test datasets are drawn randomly from our dataset, to ensure @@ -728,7 +728,7 @@ Number of test images: 360

    -

    13.2.3. Define model and architecture¶

    +

    14.2.3. Define model and architecture¶

    Our simple feed-forward neural network will consist of an input layer, a single hidden layer and an output layer. The activation \(y\) of each neuron is a weighted sum of inputs, passed through an activation function. In case of the simple perceptron model we have

    \[ z = \sum_{i=1}^n w_i a_i ,\]
    @@ -758,7 +758,7 @@ We will be using the sigmoid function which is inspired by probability theory (see logistic regression) and was most commonly used until about 2011. See the discussion below concerning other activation functions.

    -

    13.2.4. Layers¶

    +

    14.2.4. Layers¶

    • Input

    @@ -815,7 +815,7 @@ of values. Without it, any input with the value 0 will be mapped to zero (before
    -

    13.2.5. Feed-forward pass¶

    +

    14.2.5. Feed-forward pass¶

    Denote \(F\) the number of features, \(H\) the number of hidden neurons and \(C\) the number of categories.
    For each input image we calculate a weighted sum of input features (pixel values) to each neuron \(j\) in the hidden layer \(l\):

    @@ -908,7 +908,7 @@ correct label for image 0: 6
    -

    13.2.6. Choose cost function and optimizer¶

    +

    14.2.6. Choose cost function and optimizer¶

    To measure how well our neural network is doing we need to introduce a cost function.
    We will call the function that gives the error of a single sample output the loss function, and the function that gives the total error of our network across all samples the cost function. @@ -927,7 +927,7 @@ probability of the correct category \ you got the correct label. The probability of category \(c\) is given by the softmax function. The vector \(\hat{\theta}\) represents the parameters of our network, i.e. all the weights and biases.

    -

    13.2.7. Optimizing the cost function¶

    +

    14.2.7. Optimizing the cost function¶

    The network is trained by finding the weights and biases that minimize the cost function. One of the most widely used classes of methods is gradient descent and its generalizations. The idea behind gradient descent is simply to adjust the weights in the direction where the gradient of the cost function is large and negative. This ensures we flow toward a local minimum of the cost function.
    Each parameter \(\theta\) is iteratively adjusted according to the rule

    @@ -953,7 +953,7 @@ We denote each minibatch \(B_k\)The various optmization methods, with codes and algorithms, are discussed in our lectures on Gradient descent approaches.

    -

    13.2.8. Regularization¶

    +

    14.2.8. Regularization¶

    It is common to add an extra term to the cost function, proportional to the size of the weights. This is equivalent to constraining the size of the weights, so that they do not grow out of control. @@ -976,7 +976,7 @@ above. This is a clever use of the chain rule that allows us to calculate the gradient efficently.

    -

    13.2.9. Matrix multiplication¶

    +

    14.2.9. Matrix multiplication¶

    To more efficently train our network these equations are implemented using matrix operations.
    The error in the output layer is calculated simply as, with \(\hat{t}\) being our targets,

    @@ -1089,7 +1089,7 @@ the Hadamard product, meaning element-wise multiplication.

    -

    13.3. Improving performance¶

    +

    14.3. Improving performance¶

    As we can see the network does not seem to be learning at all. It seems to be just guessing the label for each image.
    In order to obtain a network that does something useful, we will have to do a bit more work.

    The choice of hyperparameters such as learning rate and regularization parameter is hugely influential for the performance of the network. Typically a grid-search is performed, wherein we test different hyperparameters separated by orders of magnitude. For example we could test the learning rates \(\eta = 10^{-6}, 10^{-5},...,10^{-1}\) with different regularization parameters \(\lambda = 10^{-6},...,10^{-0}\).

    @@ -1206,7 +1206,7 @@ being realizations of this object with different hyperparameters. An implementat
    -

    13.4. Evaluate model performance on test data¶

    +

    14.4. Evaluate model performance on test data¶

    To measure the performance of our network we evaluate how well it does it data it has never seen before, i.e. the test data.
    We measure the performance of the network using the accuracy score.
    The accuracy is as you would expect just the number of images correctly labeled divided by the total number of images. A perfect classifier will have an accuracy score of \(1\).

    @@ -1242,7 +1242,7 @@ The accuracy is as you would expect just the number of images correctly labeled
    -

    13.5. Adjust hyperparameters¶

    +

    14.5. Adjust hyperparameters¶

    We now perform a grid search to find the optimal hyperparameters for the network.
    Note that we are only using 1 layer with 50 neurons, and human performance is estimated to be around \(98\%\) (\(2\%\) error rate).

    @@ -1530,7 +1530,7 @@ Accuracy score on test set: 0.07777777777777778
    -

    13.6. Visualization¶

    +

    14.6. Visualization¶

    # visual representation of grid search
@@ -1580,7 +1580,7 @@ Accuracy score on test set:  0.07777777777777778
    -

    13.7. scikit-learn implementation¶

    +

    14.7. scikit-learn implementation¶

    scikit-learn focuses more on traditional machine learning methods, such as regression, clustering, decision trees, etc. As such, it has only two types of @@ -1939,7 +1939,7 @@ Accuracy score on test set: 0.1527777777777778

    -

    13.8. Visualization¶

    +

    14.8. Visualization¶

    # optional
@@ -1986,7 +1986,7 @@ Accuracy score on test set:  0.1527777777777778
    -

    13.9. Building neural networks in Tensorflow and Keras¶

    +

    14.9. Building neural networks in Tensorflow and Keras¶

    Now we want to build on the experience gained from our neural network implementation in NumPy and scikit-learn and use it to construct a neural network in Tensorflow. Once we have constructed a neural network in NumPy and Tensorflow, building one in Keras is really quite trivial, though the performance may suffer.

    @@ -2214,7 +2214,7 @@ If you have Anaconda installed you may run the following command

    -

    13.10. The Breast Cancer Data, now with Keras¶

    +

    14.10. The Breast Cancer Data, now with Keras¶

    import tensorflow as tf
@@ -2388,7 +2388,7 @@ If you have Anaconda installed you may run the following command
    -

    13.11. Fine-tuning neural network hyperparameters¶

    +

    14.11. Fine-tuning neural network hyperparameters¶

    The flexibility of neural networks is also one of their main drawbacks: there are many hyperparameters to tweak. Not only can you use any imaginable network topology (how neurons/nodes are interconnected), @@ -2419,7 +2419,7 @@ of training data. However, you will rarely have to train such networks from scra common to reuse parts of a pretrained state-of-the-art network that performs a similar task.

    -

    13.12. Which activation function should I use?¶

    +

    14.12. Which activation function should I use?¶

    The Back propagation algorithm we derived above works by going from the output layer to the input layer, propagating the error gradient on the way. Once the algorithm has computed the gradient of the cost @@ -2484,7 +2484,7 @@ it does not saturate for positive values (and also because it is quite fast to compute).

    -

    13.13. The RELU function family¶

    +

    14.13. The RELU function family¶

    The ReLU activation function suffers from a problem known as the dying ReLUs: during training, some neurons effectively die, meaning they stop outputting anything other than 0.

    @@ -2521,7 +2521,7 @@ bootstrap to evaluate other activation functions.

    -

    13.14. Batch Normalization¶

    +

    14.14. Batch Normalization¶

    Batch Normalization aims to address the vanishing/exploding gradients problems, and more generally the problem that the distribution of each layer’s inputs changes during training, as the parameters of the previous layers change.

    @@ -2534,7 +2534,7 @@ standard deviation. It does so by evaluating the mean and standard deviation of mini-batch, from this the name batch normalization.

    -

    13.15. Dropout¶

    +

    14.15. Dropout¶

    It is a fairly simple algorithm: at every training step, every neuron (including the input neurons but excluding the output neurons) has a probability \(p\) of being temporarily dropped out, meaning it will be entirely ignored during this training step, but it may be active during the next step.

    @@ -2543,7 +2543,7 @@ hyperparameter \(p\) is called It is viewed as one of the most popular regularization techniques.

    -

    13.16. Gradient Clipping¶

    +

    14.16. Gradient Clipping¶

    A popular technique to lessen the exploding gradients problem is to simply clip the gradients during backpropagation so that they never exceed some threshold (this is mostly useful for recurrent neural networks).

    @@ -2552,7 +2552,7 @@ networks).

    Normalization is preferred.

    -

    13.17. A top-down perspective on Neural networks¶

    +

    14.17. A top-down perspective on Neural networks¶

    The first thing we would like to do is divide the data into two or three parts. A training set, a validation or dev (development) set, and a test set. The test set is the data on which we want to make @@ -2586,7 +2586,7 @@ can serve as another important diagnostic when using DNNs for supervised learning.

    -

    13.18. Limitations of supervised learning with deep networks¶

    +

    14.18. Limitations of supervised learning with deep networks¶

    Like all statistical methods, supervised learning using neural networks has important limitations. This is especially important when one seeks to apply these methods, especially to physics problems. Like @@ -2630,7 +2630,7 @@ features).

    - 12. Neural networks + 13. Neural networks
    diff --git a/doc/LectureNotes/_build/html/chapter2.html b/doc/LectureNotes/_build/html/chapter2.html index e41ee49d1..846cae837 100644 --- a/doc/LectureNotes/_build/html/chapter2.html +++ b/doc/LectureNotes/_build/html/chapter2.html @@ -159,7 +159,7 @@
  • - 7. Support Vector Machines, overarching aims + 8. Support Vector Machines, overarching aims
    • @@ -171,12 +171,12 @@ @@ -188,12 +188,12 @@ @@ -205,12 +205,12 @@ @@ -349,6 +349,11 @@ 4.13. Bayes’ Theorem and Ridge and Lasso Regression +
  • + + 4.14. Linking Bayes’ Theorem with Ridge and Lasso Regression + +
    • @@ -365,10 +370,10 @@

    4.1. Mathematical Interpretation of Ordinary Least Squares¶

    What is presented here is a mathematical analysis of various regression algorithms (ordinary least squares, Ridge and Lasso Regression). The analysis is based on an important algorithm in linear algebra, the so-called Singular Value Decomposition (SVD).

    -

    We have shown that in ordinary least squares the optimal parameters \(\beta\) are given by

    +

    We have shown that in ordinary least squares (OLS) the optimal parameters \(\beta\) are given by

    \[ -\hat{\boldsymbol{\beta}} = \left(\boldsymbol{X}^T\boldsymbol{X}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}. +\hat{\boldsymbol{\beta}}_{\mathrm{OLS}} = \left(\boldsymbol{X}^T\boldsymbol{X}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}. \]

    The hat over \(\boldsymbol{\beta}\) means we have the optimal parameters after minimization of the cost function.

    This means that our best model is defined as

    @@ -386,7 +391,7 @@ \[ \tilde{\boldsymbol{y}}=\boldsymbol{X}\hat{\boldsymbol{\beta}} = \boldsymbol{A}\boldsymbol{y}. \]
    -

    The matrix \(\boldsymbol{A}\) has the important property that \(\boldsymbol{A}^2=\boldsymbol{A}\). This is the definition of a projection matrix. +

    The matrix \(\boldsymbol{A}\) has the important property that \(\boldsymbol{A}^2=\boldsymbol{A}\). This is the definition of a projection matrix. We can then interpret our optimal model \(\tilde{\boldsymbol{y}}\) as being represented by an orthogonal projection of \(\boldsymbol{y}\) onto a space defined by the column vectors of \(\boldsymbol{X}\). In our case here the matrix \(\boldsymbol{A}\) is a square matrix. If it is a general rectangular matrix we have an oblique projection matrix.

    We have defined the residual error as

    @@ -659,7 +664,7 @@ example

    4.5. Code for SVD and Inversion of Matrices¶

    -

    How do we use the SVD to invert a matrix \(\boldsymbol{X}^\boldsymbol{X}\) which is singular or near singular? +

    How do we use the SVD to invert a matrix \(\boldsymbol{X}^T\boldsymbol{X}\) which is singular or near singular? The simple answer is to use the linear algebra function for the pseudoinverse, that is

    @@ -1096,10 +1101,10 @@ covariance matrix through the np.linalg.eig() function.

    -
    0.059680303635344434
-4.260227742627244
-[[0.89704131 2.55795935]
- [2.55795935 8.24340751]]
+
    -0.1831277634928002
+3.3041320306136366
+[[1.03929932 3.07670437]
+ [3.07670437 9.7701384 ]]
    @@ -1136,10 +1141,10 @@ a more brute force way. Here we scale the mean values for each column of the des
    -
    0.08362077210702115
-1.8429949116841184
-[[1.        0.6996584]
- [0.6996584 1.       ]]
+
    0.09456011349477329
+1.6600855222624895
+[[1.         0.70183798]
+ [0.70183798 1.        ]]
    @@ -1169,30 +1174,30 @@ this matrix we easily see that it is a positive definite matrix.

    -
    [[-1.25981533 -3.84976606]
- [-0.52570079 -0.62262506]
- [ 0.51707172  3.46973015]
- [-0.47265243 -1.87030083]
- [ 0.73734906  2.46183815]
- [ 0.4918372   2.05531364]
- [ 0.3893239   0.47862383]
- [-0.41237437 -0.85058354]
- [ 0.77622336  0.54697204]
- [-0.24126232 -1.81920231]]
+
    [[ 0.54316925  0.40968888]
+ [-1.8393131  -5.8788431 ]
+ [ 0.24613822  0.9957744 ]
+ [ 1.38892672  3.93679587]
+ [ 0.06540809 -0.84008474]
+ [ 0.49841285  2.31081134]
+ [-0.49323032 -1.25263916]
+ [ 0.08380813 -0.7925146 ]
+ [ 0.64078247  2.92814088]
+ [-1.13410232 -1.81712976]]
           0         1
-0 -1.259815 -3.849766
-1 -0.525701 -0.622625
-2  0.517072  3.469730
-3 -0.472652 -1.870301
-4  0.737349  2.461838
-5  0.491837  2.055314
-6  0.389324  0.478624
-7 -0.412374 -0.850584
-8  0.776223  0.546972
-9 -0.241262 -1.819202
+0  0.543169  0.409689
+1 -1.839313 -5.878843
+2  0.246138  0.995774
+3  1.388927  3.936796
+4  0.065408 -0.840085
+5  0.498413  2.310811
+6 -0.493230 -1.252639
+7  0.083808 -0.792515
+8  0.640782  2.928141
+9 -1.134102 -1.817130
           0         1
-0  1.000000  0.881323
-1  0.881323  1.000000
+0  1.000000  0.942726
+1  0.942726  1.000000
    @@ -1249,37 +1254,37 @@ this matrix we easily see that it is a positive definite matrix.

         0         1         2         3         4         5         6         7   \
 0   0.0  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000   
-1   0.0  0.082272  0.078892  0.081102  0.080738  0.080182  0.072012  0.072246   
-2   0.0  0.078892  0.076822  0.075975  0.076166  0.076251  0.066705  0.067202   
-3   0.0  0.081102  0.075975  0.085259  0.084103  0.082616  0.078750  0.078649   
-4   0.0  0.080738  0.076166  0.084103  0.083246  0.082091  0.077338  0.077407   
-5   0.0  0.080182  0.076251  0.082616  0.082091  0.081309  0.075567  0.075818   
-6   0.0  0.072012  0.066705  0.078750  0.077338  0.075567  0.074693  0.074434   
-7   0.0  0.072246  0.067202  0.078649  0.077407  0.075818  0.074434  0.074285   
-8   0.0  0.072468  0.067718  0.078474  0.077416  0.076026  0.074077  0.074044   
-9   0.0  0.072661  0.068244  0.078197  0.077339  0.076169  0.073591  0.073681   
-10  0.0  0.063190  0.058247  0.070949  0.069542  0.067794  0.068593  0.068293   
-11  0.0  0.063523  0.058719  0.071156  0.069853  0.068213  0.068717  0.068491   
-12  0.0  0.063880  0.059225  0.071359  0.070166  0.068640  0.068821  0.068673   
-13  0.0  0.064257  0.059765  0.071549  0.070473  0.069071  0.068896  0.068828   
-14  0.0  0.064647  0.060335  0.071714  0.070764  0.069496  0.068928  0.068945   
+1   0.0  0.080633  0.085811  0.085748  0.084826  0.083221  0.080341  0.078342   
+2   0.0  0.085811  0.092659  0.093248  0.092932  0.091659  0.088414  0.086561   
+3   0.0  0.085748  0.093248  0.096639  0.096802  0.095941  0.093928  0.092288   
+4   0.0  0.084826  0.092932  0.096802  0.097346  0.096763  0.094774  0.093328   
+5   0.0  0.083221  0.091659  0.095941  0.096763  0.096406  0.094513  0.093236   
+6   0.0  0.080341  0.088414  0.093928  0.094774  0.094513  0.093573  0.092375   
+7   0.0  0.078342  0.086561  0.092288  0.093328  0.093236  0.092375  0.091318   
+8   0.0  0.076213  0.084461  0.090355  0.091535  0.091578  0.090817  0.089881   
+9   0.0  0.074037  0.082239  0.088258  0.089540  0.089693  0.089039  0.088209   
+10  0.0  0.073546  0.081474  0.088137  0.089326  0.089437  0.089352  0.088484   
+11  0.0  0.071453  0.079337  0.086052  0.087334  0.087544  0.087526  0.086757   
+12  0.0  0.069371  0.077163  0.083905  0.085253  0.085544  0.085598  0.084916   
+13  0.0  0.067326  0.074996  0.081744  0.083138  0.083495  0.083622  0.083015   
+14  0.0  0.065334  0.072862  0.079597  0.081024  0.081435  0.081632  0.081092   
 
           8         9         10        11        12        13        14  
 0   0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  
-1   0.072468  0.072661  0.063190  0.063523  0.063880  0.064257  0.064647  
-2   0.067718  0.068244  0.058247  0.058719  0.059225  0.059765  0.060335  
-3   0.078474  0.078197  0.070949  0.071156  0.071359  0.071549  0.071714  
-4   0.077416  0.077339  0.069542  0.069853  0.070166  0.070473  0.070764  
-5   0.076026  0.076169  0.067794  0.068213  0.068640  0.069071  0.069496  
-6   0.074077  0.073591  0.068593  0.068717  0.068821  0.068896  0.068928  
-7   0.074044  0.073681  0.068293  0.068491  0.068673  0.068828  0.068945  
-8   0.073926  0.073695  0.067889  0.068165  0.068427  0.068667  0.068873  
-9   0.073695  0.073607  0.067355  0.067711  0.068056  0.068385  0.068686  
-10  0.067889  0.067355  0.063925  0.064013  0.064075  0.064101  0.064078  
-11  0.068165  0.067711  0.064013  0.064156  0.064273  0.064357  0.064393  
-12  0.068427  0.068056  0.064075  0.064273  0.064449  0.064591  0.064689  
-13  0.068667  0.068385  0.064101  0.064357  0.064591  0.064795  0.064958  
-14  0.068873  0.068686  0.064078  0.064393  0.064689  0.064958  0.065188  
+1   0.076213  0.074037  0.073546  0.071453  0.069371  0.067326  0.065334  
+2   0.084461  0.082239  0.081474  0.079337  0.077163  0.074996  0.072862  
+3   0.090355  0.088258  0.088137  0.086052  0.083905  0.081744  0.079597  
+4   0.091535  0.089540  0.089326  0.087334  0.085253  0.083138  0.081024  
+5   0.091578  0.089693  0.089437  0.087544  0.085544  0.083495  0.081435  
+6   0.090817  0.089039  0.089352  0.087526  0.085598  0.083622  0.081632  
+7   0.089881  0.088209  0.088484  0.086757  0.084916  0.083015  0.081092  
+8   0.088556  0.086985  0.087242  0.085610  0.083855  0.082032  0.080182  
+9   0.086985  0.085510  0.085763  0.084221  0.082549  0.080806  0.079029  
+10  0.087242  0.085763  0.086430  0.084864  0.083177  0.081425  0.079643  
+11  0.085610  0.084221  0.084864  0.083384  0.081778  0.080101  0.078390  
+12  0.083855  0.082549  0.083177  0.081778  0.080249  0.078646  0.077005  
+13  0.082032  0.080806  0.081425  0.080101  0.078646  0.077115  0.075542  
+14  0.080182  0.079029  0.079643  0.078390  0.077005  0.075542  0.074036
    @@ -1758,13 +1763,11 @@ We select values of the hyperparameter 
    [2. 2.]
-
    -
    -
    Training MSE for OLS
+Training MSE for OLS
 3.0
    -_images/chapter2_245_2.png +_images/chapter2_245_1.png

    We see here that we reach a plateau for the Ridge results. Writing out the coefficients \(\boldsymbol{\beta}\), we that they are getting smaller and smaller and our error stabilizes since the predicted values of \(\tilde{\boldsymbol{y}}\) approach zero.

    @@ -2438,6 +2441,7 @@ Optimal solution found.
    +

    More text will be added to this example.

    4.11. Linking the regression analysis with a statistical interpretation¶

    @@ -2619,7 +2623,9 @@ We can now rewrite the above probability as

    \[ p(\boldsymbol{D}\vert\boldsymbol{\beta})=\prod_{i=0}^{n-1}\frac{1}{\sqrt{2\pi\sigma^2}}\exp{\left[-\frac{(y_i-\boldsymbol{X}_{i,*}\boldsymbol{\beta})^2}{2\sigma^2}\right]}. \]
    -

    It is a conditional probability (see below) and reads as the likelihood of a domain of events \(\boldsymbol{D}\) given a set of parameters \(\boldsymbol{\beta}\).

    +

    It is a conditional probability (see below) and reads as the +likelihood of a domain of events \(\boldsymbol{D}\) given a set of parameters +\(\boldsymbol{\beta}\).

    In statistics, maximum likelihood estimation (MLE) is a method of estimating the parameters of an assumed probability distribution, given some observed data. This is achieved by maximizing a likelihood @@ -2663,12 +2669,12 @@ Bayes’ theorem is extremely simple to derive. But to do so we need some basic

    Assume we have two domains of events \(X=[x_0,x_1,\dots,x_{n-1}]\) and \(Y=[y_0,y_1,\dots,y_{n-1}]\).

    We define also the likelihood for \(X\) and \(Y\) as \(p(X)\) and \(p(Y)\) respectively. The likelihood of a specific event \(x_i\) (or \(y_i\)) is then written as \(p(X=x_i)\) or just \(p(x_i)=p_i\).

    -

    Union of events is given by.

    +

    The union of events is given by

    \[ p(X \cup Y)= p(X)+p(Y)-p(X \cap Y). \]
    -

    The product rule (aka joint probability) is given by.

    +

    The product rule (aka joint probability) is given by

    \[ p(X \cup Y)= p(X,Y)= p(X\vert Y)p(Y)=p(Y\vert X)p(X), @@ -2700,37 +2706,68 @@ p(X\vert Y)= \frac{p(X,Y)}{\sum_{i=0}^{n-1}p(Y\vert X=x_i)p(x_i)}=\frac{p(Y\vert evaluated for the observed data \(Y\) and can be viewed as a function of the parameter space represented by \(X\). This function is not necesseraly normalized and is normally called the likelihood function.

    -

    The function \(p(X)\) on the right hand side is called the prior while the function on the left hand side is the called the posterior probability. The denominator on the right hand side serves as a normalization factor for the posterior distribution.

    +

    The function \(p(X)\) on the right hand side is called the prior while +the function on the left hand side is the called the posterior +probability. The denominator on the right hand side serves as a +normalization factor for the posterior distribution.

    Let us try to illustrate Bayes’ theorem through an example.

    -

    Let us suppose that you are undergoing a series of mammography scans in -order to rule out possible breast cancer cases. We define the +

    Let us suppose that you are undergoing a series of mammography scans +in order to rule out possible breast cancer cases. We define the sensitivity for a positive event by the variable \(X\). It takes binary values with \(X=1\) representing a positive event and \(X=0\) being a negative event. We reserve \(Y\) as a classification parameter for -either a negative or a positive breast cancer confirmation. (Short note on wordings: positive here means having breast cancer, although none of us would consider this being a positive thing).

    +either a negative or a positive breast cancer confirmation. (Short +note on wordings: positive here means having breast cancer, although +none of us would consider this being a positive thing).

    We let \(Y=1\) represent the the case of having breast cancer and \(Y=0\) as not.

    -

    Let us assume that if you have breast cancer, the test will be positive with a probability of \(0.8\), that is we have

    +

    Let us assume that if you have breast cancer, the test will be positive with a probability of \(0.8\) (the numbers here are all made up), +that is we have

    \[ p(X=1\vert Y=1) =0.8. \]
    -

    This obviously sounds scary since many would conclude that if the test is positive, there is a likelihood of \(80\%\) for having cancer. -It is however not correct, as the following Bayesian analysis shows.

    -

    If we look at various national surveys on breast cancer, the general likelihood of developing breast cancer is a very small number. -Let us assume that the prior probability in the population as a whole is

    +

    This obviously sounds scary since many would conclude that if the test +is positive, there is a likelihood of \(80\%\) for having cancer. It is +however not correct, as the following Bayesian analysis shows. The correct question to pose is what is the probability of having breast cancer in case of a positive test? +We are thus interested in

    +
    +\[ +p(Y=1\vert X=1), +\]
    +

    instead of \(p(X=1\vert Y=1)\).

    +

    If we look at various national surveys on breast cancer, the general +likelihood of developing breast cancer is a very small number. Let us +assume that the prior probability in the population as a whole is

    \[ p(Y=1) =0.004. \]
    -

    We need also to account for the fact that the test may produce a false positive result (false alarm). Let us here assume that we have

    +

    We need also to account for the fact that the test may produce a false +positive result (false alarm). Let us here assume that we have

    \[ p(X=1\vert Y=0) =0.1. \]
    -

    Using Bayes’ theorem we can then find the posterior probability that the person has breast cancer in case of a positive test, that is we can compute

    +

    Using Bayes’ theorem we can then find the posterior probability that +the person has breast cancer in case of a positive test, that is we +can compute

    + +
    \[ -p(Y=1\vert X=1)=\frac{p(X=1\vert Y=1)p(Y=1)}{p(X=1\vert Y=1)p(Y=1)+p(X=1\vert Y=0)p(Y=0)}=\frac{0.8\times 0.004}{0.8\times 0.004+0.1\times 0.996}=0.031. +\begin{equation} +p(Y=1\vert X=1)=\frac{p(X=1\vert Y=1)p(Y=1)}{p(X=1\vert Y=1)p(Y=1)+p(X=1\vert Y=0)p(Y=0)}= +\label{_auto2} \tag{2} +\end{equation} +\]
    + +
    +
    +\[ +\begin{equation} + \frac{0.8\times 0.004}{0.8\times 0.004+0.1\times 0.996}=0.031. +\label{_auto3} \tag{3} +\end{equation} \]

    That is, in case of a positive test, there is only a \(3\%\) chance of having breast cancer!

    @@ -2745,8 +2782,11 @@ more intuitive way of understanding what Ridge and Lasso express.

    parameters \(\boldsymbol{\beta}\) using OLS, Ridge and Lasso regression. For OLS, you will notice as function of the noise and polynomial degree, that the parameters \(\beta\) will fluctuate from order to order in the -polynomial fit and that for larger and larger polynomial degrees of freedom, the parameters will tend to increase in value for OLS.

    -

    For Ridge and Lasso regression, the higher order parameters will typically be reduced, providing thereby less fluctuations from one order to another one.

    +polynomial fit and that for larger and larger polynomial degrees of +freedom, the parameters will tend to increase in value for OLS.

    +

    For Ridge and Lasso regression, the higher order parameters will +typically be reduced, providing thereby less fluctuations from one +order to another one.

    import numpy as np
@@ -2802,8 +2842,8 @@ polynomial fit and that for larger and larger polynomial degrees of freedom, the
     # Compute the MSE and print it
     MSERidgePredict[i] = MSE(y_test,ypredictRidge)
     MSELassoPredict[i] = MSE(y_test,ypredictLasso)
-    print(lmb,RegRidge.coef_)
-    print(lmb,RegLasso.coef_)
+#    print(lmb,RegRidge.coef_)
+#    print(lmb,RegLasso.coef_)
 # Now plot the results
 plt.figure()
 plt.plot(np.log10(lambdas), MSERidgePredict, 'b', label = 'MSE Ridge Test')
@@ -2819,22 +2859,15 @@ polynomial fit and that for larger and larger polynomial degrees of freedom, the
 
    [ 1.0169643   0.27924636 -1.4087793   1.03308408  0.        ]
 Test MSE OLS
 0.958228616652075
-0.001 [ 1.0170259   0.27852549 -1.40702     1.03193199  0.        ]
-0.001 [ 1.034342   -0.18063928 -0.          0.          0.        ]
 
    
 
    
-
    0.021544346900318832 [ 1.01825571  0.26412372 -1.37186301  1.00890601  0.        ]
-0.021544346900318832 [ 0.92280994 -0.         -0.         -0.          0.        ]
-0.46415888336127775 [ 1.0344707   0.07160764 -0.89928965  0.69843037  0.        ]
-0.46415888336127775 [0.48019541 0.         0.         0.         0.        ]
-10.0 [ 1.04529095 -0.18224665 -0.15751596  0.16831012  0.        ]
-10.0 [0. 0. 0. 0. 0.]
-
    
-
    
-_images/chapter2_313_2.png
+_images/chapter2_316_1.png
    -

    How can we understand this?

    +

    How can we understand this? More text to be added.

    +
    +
    +

    4.14. Linking Bayes’ Theorem with Ridge and Lasso Regression¶

    Using Bayes’ theorem we can gain a better intuition about Ridge and Lasso regression.

    For ordinary least squares we postulated that the maximum likelihood for the doamin of events \(\boldsymbol{D}\) (one-dimensional case)

    diff --git a/doc/LectureNotes/_build/html/chapter3.html b/doc/LectureNotes/_build/html/chapter3.html index 4735f8747..7a900ba95 100644 --- a/doc/LectureNotes/_build/html/chapter3.html +++ b/doc/LectureNotes/_build/html/chapter3.html @@ -159,7 +159,7 @@
  • - 7. Support Vector Machines, overarching aims + 8. Support Vector Machines, overarching aims
    • @@ -171,12 +171,12 @@ @@ -188,12 +188,12 @@ @@ -205,12 +205,12 @@ @@ -307,13 +307,13 @@
  • - - 5.4. Various steps in cross-validation + + 5.4. The bias-variance tradeoff
  • - - 5.5. The bias-variance tradeoff + + 5.5. Cross-validation
    • @@ -587,10 +587,10 @@ number \(i\) is left out. Usin
    -
    Runtime: 0.136687 sec
+
    Runtime: 0.138472 sec
 Jackknife Statistics :
 original           bias      std. error
- 99.9196        99.9096        0.149076
+ 100.094        100.084        0.150306
    @@ -718,10 +718,10 @@ theorem.

    -
    Runtime: 1.78082 sec
+
    Runtime: 2.06834 sec
 Bootstrap Statistics :
 original           bias      std. error
- 99.9769  14.9969        99.9765        0.149832
+ 99.9933  15.0354        99.9941        0.149739
 
    
 
    
 
    ---------------------------------------------------------------------------
@@ -769,158 +769,8 @@ original           bias      std. error
    -
    -

    5.4. Various steps in cross-validation¶

    -

    When the repetitive splitting of the data set is done randomly, -samples may accidently end up in a fast majority of the splits in -either training or test set. Such samples may have an unbalanced -influence on either model building or prediction evaluation. To avoid -this \(k\)-fold cross-validation structures the data splitting. The -samples are divided into \(k\) more or less equally sized exhaustive and -mutually exclusive subsets. In turn (at each split) one of these -subsets plays the role of the test set while the union of the -remaining subsets constitutes the training set. Such a splitting -warrants a balanced representation of each sample in both training and -test set over the splits. Still the division into the \(k\) subsets -involves a degree of randomness. This may be fully excluded when -choosing \(k=n\). This particular case is referred to as leave-one-out -cross-validation (LOOCV).

    -
      -

    • Define a range of interest for the penalty parameter.

      • -

    • Divide the data set into training and test set comprising samples \(\{1, \ldots, n\} \setminus i\) and \(\{ i \}\), respectively.

      • -

    • Fit the linear regression model by means of ridge estimation for each \(\lambda\) in the grid using the training set, and the corresponding estimate of the error variance \(\boldsymbol{\sigma}_{-i}^2(\lambda)\), as

      • -
    -
    -\[ -\begin{align*} -\boldsymbol{\beta}_{-i}(\lambda) & = ( \boldsymbol{X}_{-i, \ast}^{T} -\boldsymbol{X}_{-i, \ast} + \lambda \boldsymbol{I}_{pp})^{-1} -\boldsymbol{X}_{-i, \ast}^{T} \boldsymbol{y}_{-i} -\end{align*} -\]
    -
      -

    • Evaluate the prediction performance of these models on the test set by \(\log\{L[y_i, \boldsymbol{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}\). Or, by the prediction error \(|y_i - \boldsymbol{X}_{i, \ast} \boldsymbol{\beta}_{-i}(\lambda)|\), the relative error, the error squared or the R2 score function.

      • -

    • Repeat the first three steps such that each sample plays the role of the test set once.

      • -

    • Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as

      • -
    -
    -\[ -\begin{align*} -\frac{1}{n} \sum_{i = 1}^n \log\{L[y_i, \mathbf{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}. -\end{align*} -\]
    -

    For the various values of \(k\)

    -
      -

    1. shuffle the dataset randomly.

      2. -

    2. Split the dataset into \(k\) groups.

      3. -

    3. For each unique group:

      4. -
    -

    a. Decide which group to use as set for test data

    -

    b. Take the remaining groups as a training data set

    -

    c. Fit a model on the training set and evaluate it on the test set

    -

    d. Retain the evaluation score and discard the model

    -
      -

    1. Summarize the model using the sample of model evaluation scores

      2. -
    -

    The code here uses Ridge regression with cross-validation (CV) resampling and \(k\)-fold CV in order to fit a specific polynomial.

    -
    -
    -
    import numpy as np
-import matplotlib.pyplot as plt
-from sklearn.model_selection import KFold
-from sklearn.linear_model import Ridge
-from sklearn.model_selection import cross_val_score
-from sklearn.preprocessing import PolynomialFeatures
-
-# A seed just to ensure that the random numbers are the same for every run.
-# Useful for eventual debugging.
-np.random.seed(3155)
-
-# Generate the data.
-nsamples = 100
-x = np.random.randn(nsamples)
-y = 3*x**2 + np.random.randn(nsamples)
-
-## Cross-validation on Ridge regression using KFold only
-
-# Decide degree on polynomial to fit
-poly = PolynomialFeatures(degree = 6)
-
-# Decide which values of lambda to use
-nlambdas = 500
-lambdas = np.logspace(-3, 5, nlambdas)
-
-# Initialize a KFold instance
-k = 5
-kfold = KFold(n_splits = k)
-
-# Perform the cross-validation to estimate MSE
-scores_KFold = np.zeros((nlambdas, k))
-
-i = 0
-for lmb in lambdas:
-    ridge = Ridge(alpha = lmb)
-    j = 0
-    for train_inds, test_inds in kfold.split(x):
-        xtrain = x[train_inds]
-        ytrain = y[train_inds]
-
-        xtest = x[test_inds]
-        ytest = y[test_inds]
-
-        Xtrain = poly.fit_transform(xtrain[:, np.newaxis])
-        ridge.fit(Xtrain, ytrain[:, np.newaxis])
-
-        Xtest = poly.fit_transform(xtest[:, np.newaxis])
-        ypred = ridge.predict(Xtest)
-
-        scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)
-
-        j += 1
-    i += 1
-
-
-estimated_mse_KFold = np.mean(scores_KFold, axis = 1)
-
-## Cross-validation using cross_val_score from sklearn along with KFold
-
-# kfold is an instance initialized above as:
-# kfold = KFold(n_splits = k)
-
-estimated_mse_sklearn = np.zeros(nlambdas)
-i = 0
-for lmb in lambdas:
-    ridge = Ridge(alpha = lmb)
-
-    X = poly.fit_transform(x[:, np.newaxis])
-    estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)
-
-    # cross_val_score return an array containing the estimated negative mse for every fold.
-    # we have to the the mean of every array in order to get an estimate of the mse of the model
-    estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)
-
-    i += 1
-
-## Plot and compare the slightly different ways to perform cross-validation
-
-plt.figure()
-
-plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')
-plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')
-
-plt.xlabel('log10(lambda)')
-plt.ylabel('mse')
-
-plt.legend()
-
-plt.show()
-
    -
    -
    -
    -
    -

    5.5. The bias-variance tradeoff¶

    +

    5.4. The bias-variance tradeoff¶

    We will discuss the bias-variance tradeoff in the context of continuous predictions such as regression. However, many of the intuitions and ideas discussed here also carry over to classification @@ -1265,6 +1115,157 @@ flexible statistical methods have higher variance.

    +
    +
    +

    5.5. Cross-validation¶

    +

    When the repetitive splitting of the data set is done randomly, +samples may accidently end up in a fast majority of the splits in +either training or test set. Such samples may have an unbalanced +influence on either model building or prediction evaluation. To avoid +this \(k\)-fold cross-validation structures the data splitting. The +samples are divided into \(k\) more or less equally sized exhaustive and +mutually exclusive subsets. In turn (at each split) one of these +subsets plays the role of the test set while the union of the +remaining subsets constitutes the training set. Such a splitting +warrants a balanced representation of each sample in both training and +test set over the splits. Still the division into the \(k\) subsets +involves a degree of randomness. This may be fully excluded when +choosing \(k=n\). This particular case is referred to as leave-one-out +cross-validation (LOOCV).

    +
      +

    • Define a range of interest for the penalty parameter.

      • +

    • Divide the data set into training and test set comprising samples \(\{1, \ldots, n\} \setminus i\) and \(\{ i \}\), respectively.

      • +

    • Fit the linear regression model by means of ridge estimation for each \(\lambda\) in the grid using the training set, and the corresponding estimate of the error variance \(\boldsymbol{\sigma}_{-i}^2(\lambda)\), as

      • +
    +
    +\[ +\begin{align*} +\boldsymbol{\beta}_{-i}(\lambda) & = ( \boldsymbol{X}_{-i, \ast}^{T} +\boldsymbol{X}_{-i, \ast} + \lambda \boldsymbol{I}_{pp})^{-1} +\boldsymbol{X}_{-i, \ast}^{T} \boldsymbol{y}_{-i} +\end{align*} +\]
    +
      +

    • Evaluate the prediction performance of these models on the test set by \(\log\{L[y_i, \boldsymbol{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}\). Or, by the prediction error \(|y_i - \boldsymbol{X}_{i, \ast} \boldsymbol{\beta}_{-i}(\lambda)|\), the relative error, the error squared or the R2 score function.

      • +

    • Repeat the first three steps such that each sample plays the role of the test set once.

      • +

    • Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as

      • +
    +
    +\[ +\begin{align*} +\frac{1}{n} \sum_{i = 1}^n \log\{L[y_i, \mathbf{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}. +\end{align*} +\]
    +

    For the various values of \(k\)

    +
      +

    1. shuffle the dataset randomly.

      2. +

    2. Split the dataset into \(k\) groups.

      3. +

    3. For each unique group:

      4. +
    +

    a. Decide which group to use as set for test data

    +

    b. Take the remaining groups as a training data set

    +

    c. Fit a model on the training set and evaluate it on the test set

    +

    d. Retain the evaluation score and discard the model

    +
      +

    1. Summarize the model using the sample of model evaluation scores

      2. +
    +

    The code here uses Ridge regression with cross-validation (CV) resampling and \(k\)-fold CV in order to fit a specific polynomial.

    +
    +
    +
    import numpy as np
+import matplotlib.pyplot as plt
+from sklearn.model_selection import KFold
+from sklearn.linear_model import Ridge
+from sklearn.model_selection import cross_val_score
+from sklearn.preprocessing import PolynomialFeatures
+
+# A seed just to ensure that the random numbers are the same for every run.
+# Useful for eventual debugging.
+np.random.seed(3155)
+
+# Generate the data.
+nsamples = 100
+x = np.random.randn(nsamples)
+y = 3*x**2 + np.random.randn(nsamples)
+
+## Cross-validation on Ridge regression using KFold only
+
+# Decide degree on polynomial to fit
+poly = PolynomialFeatures(degree = 6)
+
+# Decide which values of lambda to use
+nlambdas = 500
+lambdas = np.logspace(-3, 5, nlambdas)
+
+# Initialize a KFold instance
+k = 5
+kfold = KFold(n_splits = k)
+
+# Perform the cross-validation to estimate MSE
+scores_KFold = np.zeros((nlambdas, k))
+
+i = 0
+for lmb in lambdas:
+    ridge = Ridge(alpha = lmb)
+    j = 0
+    for train_inds, test_inds in kfold.split(x):
+        xtrain = x[train_inds]
+        ytrain = y[train_inds]
+
+        xtest = x[test_inds]
+        ytest = y[test_inds]
+
+        Xtrain = poly.fit_transform(xtrain[:, np.newaxis])
+        ridge.fit(Xtrain, ytrain[:, np.newaxis])
+
+        Xtest = poly.fit_transform(xtest[:, np.newaxis])
+        ypred = ridge.predict(Xtest)
+
+        scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)
+
+        j += 1
+    i += 1
+
+
+estimated_mse_KFold = np.mean(scores_KFold, axis = 1)
+
+## Cross-validation using cross_val_score from sklearn along with KFold
+
+# kfold is an instance initialized above as:
+# kfold = KFold(n_splits = k)
+
+estimated_mse_sklearn = np.zeros(nlambdas)
+i = 0
+for lmb in lambdas:
+    ridge = Ridge(alpha = lmb)
+
+    X = poly.fit_transform(x[:, np.newaxis])
+    estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)
+
+    # cross_val_score return an array containing the estimated negative mse for every fold.
+    # we have to the the mean of every array in order to get an estimate of the mse of the model
+    estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)
+
+    i += 1
+
+## Plot and compare the slightly different ways to perform cross-validation
+
+plt.figure()
+
+plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')
+plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')
+
+plt.xlabel('log10(lambda)')
+plt.ylabel('mse')
+
+plt.legend()
+
+plt.show()
+
    +
    +
    +
    +

    More examples of the application of cross-validation follow here.

    # Common imports
diff --git a/doc/LectureNotes/_build/html/chapter4.html b/doc/LectureNotes/_build/html/chapter4.html
index ee803c07c..bb0ddc790 100644
--- a/doc/LectureNotes/_build/html/chapter4.html
+++ b/doc/LectureNotes/_build/html/chapter4.html
@@ -54,7 +54,7 @@
     
     
     
-    
+    
     
     
     
@@ -159,7 +159,7 @@
  
  
  • 
   
-   7. Support Vector Machines, overarching aims
+   8. Support Vector Machines, overarching aims
   
  
    • 
 
@@ -171,12 +171,12 @@
 
@@ -188,12 +188,12 @@
 
@@ -205,12 +205,12 @@
 
@@ -284,283 +284,120 @@
    -

    6.6. Optimization, the central part of any Machine Learning algortithm¶

    +

    7. Optimization, the central part of any Machine Learning algortithm¶

    Almost every problem in machine learning and data science starts with a dataset \(X\), a model \(g(\beta)\), which is a function of the parameters \(\beta\) and a cost function \(C(X, g(\beta))\) that allows @@ -1307,9 +1135,6 @@ us to judge how well the model \(g(\b the cost function. Ideally we would be able to solve for \(\beta\) analytically, however this is not possible in general and we must use some approximative/numerical method to compute the minimum.

    -
    -
    -

    6.7. Revisiting our Logistic Regression case¶

    In our discussion on Logistic Regression we studied the case of two classes, with \(y_i\) either @@ -1324,9 +1149,6 @@ p(y_i=0|x_i,\boldsymbol{\beta}) &= 1 - p(y_i=1|x_i,\boldsymbol{\beta}), \end{align*} \end{split}\]

    where \(\boldsymbol{\beta}\) are the weights we wish to extract from data, in our case \(\beta_0\) and \(\beta_1\).

    -
    -
    -

    6.8. The equations to solve¶

    Our compact equations used a definition of a vector \(\boldsymbol{y}\) with \(n\) elements \(y_i\), an \(n\times p\) matrix \(\boldsymbol{X}\) which contains the \(x_i\) values and a vector \(\boldsymbol{p}\) of fitted probabilities @@ -1343,9 +1165,6 @@ the first derivative of the cost function as

    \frac{\partial^2 \mathcal{C}(\boldsymbol{\beta})}{\partial \boldsymbol{\beta}\partial \boldsymbol{\beta}^T} = \boldsymbol{X}^T\boldsymbol{W}\boldsymbol{X}. \]

    This defines what is called the Hessian matrix.

    -
    -
    -

    6.9. Solving using Newton-Raphson’s method¶

    If we can set up these equations, Newton-Raphson’s iterative method is normally the method of choice. It requires however that we can compute in an efficient way the matrices that define the first and second derivatives.

    Our iterative scheme is then given by

    @@ -1359,9 +1178,6 @@ the first derivative of the cost function as

    \]

    The right-hand side is computed with the old values of \(\beta\).

    If we can compute these matrices, in particular the Hessian, the above is often the easiest method to implement.

    -
    -
    -

    6.10. Brief reminder on Newton-Raphson’s method¶

    Let us quickly remind ourselves how we derive the above method.

    Perhaps the most celebrated of all one-dimensional root-finding routines is Newton’s method, also called the Newton-Raphson @@ -1370,9 +1186,6 @@ function \(f\) and its derivat If you can only calculate the derivative numerically and/or your function is not of the smooth type, we normally discourage the use of this method.

    -
    -
    -

    6.11. The equations¶

    The Newton-Raphson formula consists geometrically of extending the tangent line at a current point until it crosses zero, then setting the next guess to the abscissa of that zero-crossing. The mathematics @@ -1401,9 +1214,6 @@ s\approx x-\frac{f(x)}{f'(x)}. \[ x_{n+1}=x_n-\frac{f(x_n)}{f'(x_n)}. \]

    -
    -
    -

    6.12. Simple geometric interpretation¶

    The above is Newton-Raphson’s method. It has a simple geometric interpretation, namely \(x_{n+1}\) is the point where the tangent from \((x_n,f(x_n))\) crosses the \(x\)-axis. Close to the solution, @@ -1415,9 +1225,6 @@ from the true root as to let the search interval include a local maximum or minimum of the function. If an iteration places a trial guess near such a local extremum, so that the first derivative nearly vanishes, then Newton-Raphson may fail totally

    -
    -
    -

    6.13. Extending to more than one variable¶

    Newton’s method can be generalized to systems of several non-linear equations and variables. Consider the case with two equations

    @@ -1463,9 +1270,8 @@ is to understand that difficulties may arise in case \(\boldsymbol{J}\) is nearly singular.

    It is rather straightforward to extend the above scheme to systems of more than two non-linear equations. In our case, the Jacobian matrix is given by the Hessian that represents the second derivative of cost function.

    -
    -

    6.14. Steepest descent¶

    +

    7.1. Steepest descent¶

    The basic idea of gradient descent is that a function \(F(\mathbf{x})\), \(\mathbf{x} \equiv (x_1,\cdots,x_n)\), decreases fastest if one goes from \(\bf {x}\) in the @@ -1479,9 +1285,6 @@ direction of the negative gradient \(

    For \(\gamma_k\) small enough, then \(F(\mathbf{x}_{k+1}) \leq F(\mathbf{x}_k)\). This means that for a sufficiently small \(\gamma_k\) we are always moving towards smaller function values, i.e a minimum.

    -
    -
    -

    6.15. More on Steepest descent¶

    The previous observation is the basis of the method of steepest descent, which is also referred to as just gradient descent (GD). One starts with an initial guess \(\mathbf{x}_0\) for a minimum of \(F\) and @@ -1492,9 +1295,6 @@ computes new approximations according to

    \]

    The parameter \(\gamma_k\) is often referred to as the step length or the learning rate within the context of Machine Learning.

    -
    -
    -

    6.16. The ideal¶

    Ideally the sequence \(\{\mathbf{x}_k \}_{k=0}\) converges to a global minimum of the function \(F\). In general we do not know if we are in a global or local minimum. In the special case when \(F\) is a convex @@ -1510,9 +1310,6 @@ have a very good intial guess. This also implies that the scheme is sensitive to the chosen initial condition.

    Note that the gradient is a function of \(\mathbf{x} = (x_1,\cdots,x_n)\) which makes it expensive to compute numerically.

    -
    -
    -

    6.17. The sensitiveness of the gradient descent¶

    The gradient descent method is sensitive to the choice of learning rate \(\gamma_k\). This is due to the fact that we are only guaranteed that \(F(\mathbf{x}_{k+1}) \leq @@ -1525,7 +1322,7 @@ randomness. One such method is that of Stochastic Gradient Descent (SGD), see below.

    -

    6.18. Convex functions¶

    +

    7.2. Convex functions¶

    Ideally we want our cost/loss function to be convex(concave).

    First we give the definition of a convex set: A set \(C\) in \(\mathbb{R}^n\) is said to be convex if, for all \(x\) and \(y\) in \(C\) and @@ -1535,13 +1332,7 @@ connecting \(x\) and The convex subsets of \(\mathbb{R}\) are the intervals of \(\mathbb{R}\). Examples of convex sets of \(\mathbb{R}^2\) are the regular polygons (triangles, rectangles, pentagons, etc…).

    -
    -
    -

    6.19. Convex function¶

    Convex function: Let \(X \subset \mathbb{R}^n\) be a convex set. Assume that the function \(f: X \rightarrow \mathbb{R}\) is continuous, then \(f\) is said to be convex if $\(f(tx_1 + (1-t)x_2) \leq tf(x_1) + (1-t)f(x_2) \)\( for all \)x_1, x_2 \in X\( and for all \)t \in [0,1]\(. If \)\leq\( is replaced with a strict inequaltiy in the definition, we demand \)x_1 \neq x_2\( and \)t\in(0,1)\( then \)f\( is said to be strictly convex. For a single variable function, convexity means that if you draw a straight line connecting \)f(x_1)\( and \)f(x_2)\(, the value of the function on the interval \)[x_1,x_2]$ is always below the line as illustrated below.

    -
    -
    -

    6.20. Conditions on convex functions¶

    In the following we state first and second-order conditions which ensures convexity of a function \(f\). We write \(D_f\) to denote the domain of \(f\), i.e the subset of \(R^n\) where \(f\) is defined. For more @@ -1563,9 +1354,6 @@ Hessian is positive semi-definite for all \(f''(x) \geq 0\). Geometrically this means that \(f\) has nonnegative curvature everywhere.

    This condition is particularly useful since it gives us an procedure for determining if the function under consideration is convex, apart from using the definition.

    -
    -
    -

    6.21. More on convex functions¶

    The next result is of great importance to us and the reason why we are going on about convex functions. In machine learning we frequently have to minimize a loss/cost function in order to find the best @@ -1579,9 +1367,8 @@ is convex the following result provides invaluable information:

    is minimal, where \(f\) is convex and differentiable. Then, any point \(x^*\) that satisfies \(\nabla f(x^*) = 0\) is a global minimum.

    This result means that if we know that the cost/loss function is convex and we are able to find a minimum, we are guaranteed that it is a global minimum.

    -
    -

    6.22. Some simple problems¶

    +

    7.2.1. Some simple problems¶

    1. Show that \(f(x)=x^2\) is convex for \(x \in \mathbb{R}\) using the definition of convexity. Hint: If you re-write the definition, \(f\) is convex if the following holds for all \(x,y \in D_f\) and any \(\lambda \in [0,1]\) \(\lambda f(x)+(1-\lambda)f(y)-f(\lambda x + (1-\lambda) y ) \geq 0\).

    2. Using the second order condition show that the following functions are convex on the specified domain.

      3. @@ -1601,12 +1388,9 @@ is minimal, where \(f\) is con

      Using the definition of convexity, try to show that a function satisfying the properties above is convex (the third condition is not needed to show this).

    -
    -

    6.23. Friday September 25¶

    -

    Video of Lecture and link to handwritten notes.

    -

    6.24. Standard steepest descent¶

    +

    7.3. Standard steepest descent¶

    Before we proceed, we would like to discuss the approach called the standard Steepest descent (different from the above steepest descent discussion), which again leads to us having to be able to compute a matrix. It belongs to the class of Conjugate Gradient methods (CG).

    @@ -1626,9 +1410,6 @@ algebra of the type

    \]

    where \(\boldsymbol{r}\) is the so-called residual or error in the iterative process.

    When we have found the exact solution, \(\boldsymbol{r}=0\).

    -
    -
    -

    6.25. Gradient method¶

    The residual is zero when we reach the minimum of the quadratic equation

    \[ @@ -1636,9 +1417,6 @@ P(\boldsymbol{x})=\frac{1}{2}\boldsymbol{x}^T\boldsymbol{A}\boldsymbol{x} - \bol \]

    with the constraint that the matrix \(\boldsymbol{A}\) is positive definite and symmetric. This defines also the Hessian and we want it to be positive definite.

    -
    -
    -

    6.26. Steepest descent method¶

    We denote the initial guess for \(\boldsymbol{x}\) as \(\boldsymbol{x}_0\). We can assume without loss of generality that

    @@ -1651,9 +1429,6 @@ We can assume without loss of generality that

    \boldsymbol{A}\boldsymbol{z} = \boldsymbol{b}-\boldsymbol{A}\boldsymbol{x}_0, \]

    instead.

    -
    -
    -

    6.27. Steepest descent method¶

    One can show that the solution \(\boldsymbol{x}\) is also the unique minimizer of the quadratic form

    \[ @@ -1668,9 +1443,6 @@ which equals

    \]

    and \(\boldsymbol{x}_0=0\) it is equal \(-\boldsymbol{b}\).

    -
    -
    -

    6.28. Final expressions¶

    We can compute the residual iteratively as

    \[ @@ -1696,9 +1468,6 @@ which equals

    \[ \boldsymbol{x}_{k+1}=\boldsymbol{x}_k-\alpha_k\boldsymbol{r}_{k}, \]
    -
    -
    -

    6.29. Steepest descent example¶

    import numpy as np
@@ -1724,6 +1493,12 @@ which equals
    +
    +
    <mpl_toolkits.mplot3d.art3d.Poly3DCollection at 0x7febdb109490>
+
    +
    +_images/chapter4_114_1.png +

    And then as countor plot

    @@ -1734,6 +1509,9 @@ which equals

    +
    +_images/chapter4_116_0.png +

    Find guesses

    @@ -1757,6 +1535,11 @@ which equals

    +
    +
    [ 1.33333333 -0.26666667]
+
    +
    +

    What happened?

    @@ -1768,10 +1551,16 @@ which equals

    +
    +
    [<matplotlib.lines.Line2D at 0x7febdb6780d0>]
+
    +
    +_images/chapter4_122_1.png +
    -

    6.30. Conjugate gradient method¶

    +

    7.4. Conjugate gradient method¶

    In the CG method we define so-called conjugate directions and two vectors \(\boldsymbol{s}\) and \(\boldsymbol{t}\) are said to be @@ -1788,18 +1577,12 @@ of our vectors \(\boldsymbol{x}_i\)

    Two vectors are conjugate if they are orthogonal with respect to this inner product. Being conjugate is a symmetric relation: if \(\boldsymbol{s}\) is conjugate to \(\boldsymbol{t}\), then \(\boldsymbol{t}\) is conjugate to \(\boldsymbol{s}\).

    -
    -
    -

    6.31. Conjugate gradient method¶

    An example is given by the eigenvectors of the matrix

    \[ \boldsymbol{v}_i^T\boldsymbol{A}\boldsymbol{v}_j= \lambda\boldsymbol{v}_i^T\boldsymbol{v}_j, \]

    which is zero unless \(i=j\).

    -
    -
    -

    6.32. Conjugate gradient method¶

    Assume now that we have a symmetric positive-definite matrix \(\boldsymbol{A}\) of size \(n\times n\). At each iteration \(i+1\) we obtain the conjugate direction of a vector

    @@ -1813,9 +1596,6 @@ Then the \(\boldsymbol{p}_{i}\) -
    -
    -

    6.33. Conjugate gradient method¶

    The coefficients are given by

    \[ @@ -1831,9 +1611,6 @@ Then the \(\boldsymbol{p}_{i}\) -
    -
    -

    6.34. Conjugate gradient method and iterations¶

    If we choose the conjugate vectors \(\boldsymbol{p}_k\) carefully, then we may not need all of them to obtain a good approximation to the solution \(\boldsymbol{x}\). @@ -1852,9 +1629,6 @@ We can assume without loss of generality that

    \boldsymbol{A}\boldsymbol{z} = \boldsymbol{b}-\boldsymbol{A}\boldsymbol{x}_0, \]

    instead.

    -
    -
    -

    6.35. Conjugate gradient method¶

    One can show that the solution \(\boldsymbol{x}\) is also the unique minimizer of the quadratic form

    \[ @@ -1871,9 +1645,6 @@ which equals

    \(\boldsymbol{x}_0=0\) it is equal \(-\boldsymbol{b}\). The other vectors in the basis will be conjugate to the gradient, hence the name conjugate gradient method.

    -
    -
    -

    6.36. Conjugate gradient method¶

    Let \(\boldsymbol{r}_k\) be the residual at the \(k\)-th step:

    \[ @@ -1890,9 +1661,6 @@ This gives the following expression

    \[ \boldsymbol{p}_{k+1}=\boldsymbol{r}_k-\frac{\boldsymbol{p}_k^T \boldsymbol{A}\boldsymbol{r}_k}{\boldsymbol{p}_k^T\boldsymbol{A}\boldsymbol{p}_k} \boldsymbol{p}_k. \]
    -
    -
    -

    6.37. Conjugate gradient method¶

    We can also compute the residual iteratively as

    \[ @@ -1914,14 +1682,14 @@ This gives the following expression

    \boldsymbol{r}_{k+1}=\boldsymbol{r}_k-\boldsymbol{A}\boldsymbol{p}_{k}, \]
    -
    -

    6.38. Revisiting our first homework¶

    +
    +

    7.5. Revisiting our Linear Regression Solvers¶

    We will use linear regression as a case study for the gradient descent methods. Linear regression is a great test case for the gradient descent methods discussed in the lectures since it has several desirable properties such as:

      -

    1. An analytical solution (recall homework set 1).

      2. +

    2. An analytical solution.

    3. The gradient can be computed analytically.

    4. The cost function is convex which guarantees that gradient descent converges for small enough learning rates

    @@ -1933,6 +1701,17 @@ desirable properties such as:

    +
    +
    ---------------------------------------------------------------------------
+NameError                                 Traceback (most recent call last)
+<ipython-input-14-be0d226abb41> in <module>
+----> 1 x = 2*np.random.rand(m,1)
+      2 y = 4+3*x+np.random.randn(m,1)
+
+NameError: name 'm' is not defined
+
    +
    +

    with \(x_i \in [0,1] \) is chosen randomly using a uniform distribution. Additionally we have a stochastic noise chosen according to a normal distribution \(\cal {N}(0,1)\). The linear regression model is given by

    @@ -1945,9 +1724,6 @@ h_\beta(x) = \boldsymbol{y} = \beta_0 + \beta_1 x, \[ \boldsymbol{y}_i = \beta_0 + \beta_1 x_i. \] - -
    -

    6.39. Gradient descent example¶

    Let \(\mathbf{y} = (y_1,\cdots,y_n)^T\), \(\mathbf{\boldsymbol{y}} = (\boldsymbol{y}_1,\cdots,\boldsymbol{y}_n)^T\) and \(\beta = (\beta_0, \beta_1)^T\)

    It is convenient to write \(\mathbf{\boldsymbol{y}} = X\beta\) where \(X \in \mathbb{R}^{100 \times 2} \) is the design matrix given by (we keep the intercept here)

    @@ -1964,9 +1740,6 @@ X \equiv \begin{bmatrix} C(\beta) = \frac{1}{n}||X\beta-\mathbf{y}||_{2}^{2} = \frac{1}{n}\sum_{i=1}^{100}\left[ (\beta_0 + \beta_1 x_i)^2 - 2 y_i (\beta_0 + \beta_1 x_i) + y_i^2\right] \]

    and we want to find \(\beta\) such that \(C(\beta)\) is minimized.

    -
    -
    -

    6.40. The derivative of the cost/loss function¶

    Computing \(\partial C(\beta) / \partial \beta_0\) and \(\partial C(\beta) / \partial \beta_1\) we can show that the gradient can be written as

    \[\begin{split} @@ -1975,9 +1748,6 @@ C(\beta) = \frac{1}{n}||X\beta-\mathbf{y}||_{2}^{2} = \frac{1}{n}\sum_{i=1}^{100 \end{bmatrix} = \frac{2}{n}X^T(X\beta - \mathbf{y}), \end{split}\]

    where \(X\) is the design matrix defined above.

    -
    -
    -

    6.41. The Hessian matrix¶

    The Hessian matrix of \(C(\beta)\) is given by

    \[\begin{split} @@ -1987,9 +1757,6 @@ C(\beta) = \frac{1}{n}||X\beta-\mathbf{y}||_{2}^{2} = \frac{1}{n}\sum_{i=1}^{100 \end{bmatrix} = \frac{2}{n}X^T X. \end{split}\]

    This result implies that \(C(\beta)\) is a convex function since the matrix \(X^T X\) always is positive semi-definite.

    -
    -
    -

    6.42. Simple program¶

    We can now write a program that minimizes \(C(\beta)\) using the gradient descent method with a constant learning rate \(\gamma\) according to

    \[ @@ -2000,10 +1767,7 @@ C(\beta) = \frac{1}{n}||X\beta-\mathbf{y}||_{2}^{2} = \frac{1}{n}\sum_{i=1}^{100 when \(||\nabla_\beta C(\beta_k) || \leq \epsilon = 10^{-8}\). Note that the code below does not include the latter stop criterion.

    And finally we can compare our solution for \(\beta\) with the analytic result given by \(\beta= (X^TX)^{-1} X^T \mathbf{y}\).

    -
    -
    -

    6.43. Gradient Descent Example¶

    -

    Here our simple example

    +

    Here is our simple example

    # Importing various packages
@@ -2055,9 +1819,7 @@ when \(||\nabla_\beta C(\beta_k) || \
    -
    -
    -

    6.44. And a corresponding example using scikit-learn¶

    +

    Alternatively, we can use Scikit-Learn as done here

    # Importing various packages
@@ -2080,9 +1842,6 @@ when \(||\nabla_\beta C(\beta_k) || \
    -
    -
    -

    6.45. Gradient descent and Ridge¶

    We have also discussed Ridge regression where the loss function contains a regularized term given by the \(L_2\) norm of \(\beta\),

    \[ @@ -2100,9 +1859,6 @@ C_{\text{ridge}}(\beta) = \frac{1}{n}||X\beta -\mathbf{y}||^2 + \lambda ||\beta| \[ \beta_{\text{ridge}} = \left(X^T X + \lambda I_{2 \times 2} \right)^{-1} X^T \mathbf{y}. \]
    -
    -
    -

    6.46. Program example for gradient descent with Ridge Regression¶

    from random import random, seed
@@ -2154,7 +1910,7 @@ C_{\text{ridge}}(\beta) = \frac{1}{n}||X\beta -\mathbf{y}||^2 + \lambda ||\beta|
    -

    6.47. Using gradient descent methods, limitations¶

    +

    7.6. Using gradient descent methods, limitations¶

    • Gradient descent (GD) finds local minima of our function. Since the GD algorithm is deterministic, if it converges, it will converge to a local minimum of our cost/loss/risk function. Because in ML we are often dealing with extremely rugged landscapes with many local minima, this can lead to poor performance.

    • GD is sensitive to initial conditions. One consequence of the local nature of GD is that initial conditions matter. Depending on where one starts, one will end up at a different local minima. Therefore, it is very important to think about how one initializes the training process. This is true for GD as well as more complicated variants of GD.

      • @@ -2165,7 +1921,7 @@ C_{\text{ridge}}(\beta) = \frac{1}{n}||X\beta -\mathbf{y}||^2 + \lambda ||\beta|
    -

    6.48. Stochastic Gradient Descent¶

    +

    7.7. Stochastic Gradient Descent¶

    Stochastic gradient descent (SGD) and variants thereof address some of the shortcomings of the Gradient descent method discussed above.

    The underlying idea of SGD comes from the observation that the cost @@ -2176,9 +1932,6 @@ sum over \(n\) data points -

    -
    -

    6.49. Computation of gradients¶

    This in turn means that the gradient can be computed as a sum over \(i\)-gradients

    @@ -2191,9 +1944,6 @@ gradient on a subset of the data called minibatches. If there are \(M\), there will be \(n/M\) minibatches. We denote these minibatches by \(B_k\) where \(k=1,\cdots,n/M\).

    -
    -
    -

    6.50. SGD example¶

    As an example, suppose we have \(10\) data points \((\mathbf{x}_1,\cdots, \mathbf{x}_{10})\) and we choose to have \(M=5\) minibathces, then each minibatch contains two data points. In particular we have @@ -2212,9 +1962,6 @@ C(\mathbf{\beta}) = \sum_{i=1}^n \nabla_\beta c_i(\mathbf{x}_i, \mathbf{\beta}) \rightarrow \sum_{i \in B_k}^n \nabla_\beta c_i(\mathbf{x}_i, \mathbf{\beta}). \]

    -
    -
    -

    6.51. The gradient step¶

    Thus a gradient descent step now looks like

    \[ @@ -2226,9 +1973,6 @@ probability from \([1,n/M]\). minibathces (n/M) is commonly referred to as an epoch. Thus it is typical to choose a number of epochs and for each epoch iterate over the number of minibatches, as exemplified in the code below.

    -
    -
    -

    6.52. Simple example code¶

    import numpy as np 
@@ -2256,9 +2000,6 @@ the size of the minibatches are small relative to the number of
 datapoints (\(M <  n\)), the computation of the gradient is much
 cheaper since we sum over the datapoints in the \(k-th\) minibatch and not
 all \(n\) datapoints.
    
-
    -
    -

    6.53. When do we stop?¶

    A natural question is when do we stop the search for a new minimum? One possibility is to compute the full gradient after a given number of epochs and check if the norm of the gradient is smaller than some @@ -2269,9 +2010,6 @@ evaluate the cost function at this point, store the result and continue the search. If the test kicks in at a later stage we can compare the values of the cost function and keep the \(\beta\) that gave the lowest value.

    -
    -
    -

    6.54. Slightly different approach¶

    Another approach is to let the step length \(\gamma_j\) depend on the number of epochs in such a way that it becomes very small after a reasonable time such that we do not move at all.

    @@ -2311,9 +2049,8 @@ function.

    -
    -

    6.55. Program for stochastic gradient¶

    +

    7.7.1. Program for stochastic gradient¶

    # Importing various packages
@@ -2384,7 +2121,594 @@ function.
    -

    Challenge: try to write a similar code for a Logistic Regression case.

    +
    +
    +
    +

    7.8. Momentum based GD¶

    +

    The stochastic gradient descent (SGD) is almost always used with a +momentum or inertia term that serves as a memory of the direction we +are moving in parameter space. This is typically implemented as +follows

    +
    +\[ +\mathbf{v}_{t}=\gamma \mathbf{v}_{t-1}+\eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t) \nonumber +\]
    + +
    +
    +\[ +\begin{equation} +\boldsymbol{\theta}_{t+1}= \boldsymbol{\theta}_t -\mathbf{v}_{t}, +\label{_auto2} \tag{3} +\end{equation} +\]
    +

    where we have introduced a momentum parameter \(\gamma\), with +\(0\le\gamma\le 1\), and for brevity we dropped the explicit notation to +indicate the gradient is to be taken over a different mini-batch at +each step. We call this algorithm gradient descent with momentum +(GDM). From these equations, it is clear that \(\mathbf{v}_t\) is a +running average of recently encountered gradients and +\((1-\gamma)^{-1}\) sets the characteristic time scale for the memory +used in the averaging procedure. Consistent with this, when +\(\gamma=0\), this just reduces down to ordinary SGD as discussed +earlier. An equivalent way of writing the updates is

    +
    +\[ +\Delta \boldsymbol{\theta}_{t+1} = \gamma \Delta \boldsymbol{\theta}_t -\ \eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t), +\]
    +

    where we have defined \(\Delta \boldsymbol{\theta}_{t}= \boldsymbol{\theta}_t-\boldsymbol{\theta}_{t-1}\).

    +

    Let us try to get more intuition from these equations. It is helpful +to consider a simple physical analogy with a particle of mass \(m\) +moving in a viscous medium with drag coefficient \(\mu\) and potential +\(E(\mathbf{w})\). If we denote the particle’s position by \(\mathbf{w}\), +then its motion is described by

    +
    +\[ +m {d^2 \mathbf{w} \over dt^2} + \mu {d \mathbf{w} \over dt }= -\nabla_w E(\mathbf{w}). +\]
    +

    We can discretize this equation in the usual way to get

    +
    +\[ +m { \mathbf{w}_{t+\Delta t}-2 \mathbf{w}_{t} +\mathbf{w}_{t-\Delta t} \over (\Delta t)^2}+\mu {\mathbf{w}_{t+\Delta t}- \mathbf{w}_{t} \over \Delta t} = -\nabla_w E(\mathbf{w}). +\]
    +

    Rearranging this equation, we can rewrite this as

    +
    +\[ +\Delta \mathbf{w}_{t +\Delta t}= - { (\Delta t)^2 \over m +\mu \Delta t} \nabla_w E(\mathbf{w})+ {m \over m +\mu \Delta t} \Delta \mathbf{w}_t. +\]
    +

    Notice that this equation is identical to previous one if we identify +the position of the particle, \(\mathbf{w}\), with the parameters +\(\boldsymbol{\theta}\). This allows us to identify the momentum +parameter and learning rate with the mass of the particle and the +viscous drag as:

    +
    +\[ +\gamma= {m \over m +\mu \Delta t }, \qquad \eta = {(\Delta t)^2 \over m +\mu \Delta t}. +\]
    +

    Thus, as the name suggests, the momentum parameter is proportional to +the mass of the particle and effectively provides inertia. +Furthermore, in the large viscosity/small learning rate limit, our +memory time scales as \((1-\gamma)^{-1} \approx m/(\mu \Delta t)\).

    +

    Why is momentum useful? SGD momentum helps the gradient descent +algorithm gain speed in directions with persistent but small gradients +even in the presence of stochasticity, while suppressing oscillations +in high-curvature directions. This becomes especially important in +situations where the landscape is shallow and flat in some directions +and narrow and steep in others. It has been argued that first-order +methods (with appropriate initial conditions) can perform comparable +to more expensive second order methods, especially in the context of +complex deep learning models.

    +

    These beneficial properties of momentum can sometimes become even more +pronounced by using a slight modification of the classical momentum +algorithm called Nesterov Accelerated Gradient (NAG).

    +

    In the NAG algorithm, rather than calculating the gradient at the +current parameters, \(\nabla_\theta E(\boldsymbol{\theta}_t)\), one +calculates the gradient at the expected value of the parameters given +our current momentum, \(\nabla_\theta E(\boldsymbol{\theta}_t +\gamma +\mathbf{v}_{t-1})\). This yields the NAG update rule

    +
    +\[ +\mathbf{v}_{t}=\gamma \mathbf{v}_{t-1}+\eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t +\gamma \mathbf{v}_{t-1}) \nonumber +\]
    + +
    +
    +\[ +\begin{equation} +\boldsymbol{\theta}_{t+1}= \boldsymbol{\theta}_t -\mathbf{v}_{t}. +\label{_auto3} \tag{4} +\end{equation} +\]
    +

    One of the major advantages of NAG is that it allows for the use of a larger learning rate than GDM for the same choice of \(\gamma\).

    +

    In stochastic gradient descent, with and without momentum, we still +have to specify a schedule for tuning the learning rates \(\eta_t\) +as a function of time. As discussed in the context of Newton’s +method, this presents a number of dilemmas. The learning rate is +limited by the steepest direction which can change depending on the +current position in the landscape. To circumvent this problem, ideally +our algorithm would keep track of curvature and take large steps in +shallow, flat directions and small steps in steep, narrow directions. +Second-order methods accomplish this by calculating or approximating +the Hessian and normalizing the learning rate by the +curvature. However, this is very computationally expensive for +extremely large models. Ideally, we would like to be able to +adaptively change the step size to match the landscape without paying +the steep computational price of calculating or approximating +Hessians.

    +

    Recently, a number of methods have been introduced that accomplish +this by tracking not only the gradient, but also the second moment of +the gradient. These methods include AdaGrad, AdaDelta, RMS-Prop, and +ADAM.

    +
    +

    7.8.1. RMS prop¶

    +

    In RMS prop, in addition to keeping a running average of the first +moment of the gradient, we also keep track of the second moment +denoted by \(\mathbf{s}_t=\mathbb{E}[\mathbf{g}_t^2]\). The update rule +for RMS prop is given by

    + +
    +
    +\[ +\begin{equation} +\mathbf{g}_t = \nabla_\theta E(\boldsymbol{\theta}) +\label{_auto4} \tag{5} +\end{equation} +\]
    +
    +\[ +\mathbf{s}_t =\beta \mathbf{s}_{t-1} +(1-\beta)\mathbf{g}_t^2 \nonumber +\]
    +
    +\[ +\boldsymbol{\theta}_{t+1}=\boldsymbol{\theta}_t - \eta_t { \mathbf{g}_t \over \sqrt{\mathbf{s}_t +\epsilon}}, \nonumber +\]
    +

    where \(\beta\) controls the averaging time of the second moment and is +typically taken to be about \(\beta=0.9\), \(\eta_t\) is a learning rate +typically chosen to be \(10^{-3}\), and \(\epsilon\sim 10^{-8} \) is a +small regularization constant to prevent divergences. Multiplication +and division by vectors is understood as an element-wise operation. It +is clear from this formula that the learning rate is reduced in +directions where the norm of the gradient is consistently large. This +greatly speeds up the convergence by allowing us to use a larger +learning rate for flat directions.

    +
    +
    +

    7.8.2. ADAM optimizer¶

    +

    A related algorithm is the ADAM optimizer. In ADAM, we keep a running +average of both the first and second moment of the gradient and use +this information to adaptively change the learning rate for different +parameters. In addition to keeping a running average of the first and +second moments of the gradient +(i.e. \(\mathbf{m}_t=\mathbb{E}[\mathbf{g}_t]\) and +\(\mathbf{s}_t=\mathbb{E}[\mathbf{g}^2_t]\), respectively), ADAM +performs an additional bias correction to account for the fact that we +are estimating the first two moments of the gradient using a running +average (denoted by the hats in the update rule below). The update +rule for ADAM is given by (where multiplication and division are once +again understood to be element-wise operations below)

    + +
    +
    +\[ +\begin{equation} +\mathbf{g}_t = \nabla_\theta E(\boldsymbol{\theta}) +\label{_auto5} \tag{6} +\end{equation} +\]
    +
    +\[ +\mathbf{m}_t = \beta_1 \mathbf{m}_{t-1} + (1-\beta_1) \mathbf{g}_t \nonumber +\]
    +
    +\[ +\mathbf{s}_t =\beta_2 \mathbf{s}_{t-1} +(1-\beta_2)\mathbf{g}_t^2 \nonumber +\]
    +
    +\[ +\boldsymbol{\mathbf{m}}_t={\mathbf{m}_t \over 1-\beta_1^t} \nonumber +\]
    +
    +\[ +\boldsymbol{\mathbf{s}}_t ={\mathbf{s}_t \over1-\beta_2^t} \nonumber +\]
    +
    +\[ +\boldsymbol{\theta}_{t+1}=\boldsymbol{\theta}_t - \eta_t { \boldsymbol{\mathbf{m}}_t \over \sqrt{\boldsymbol{\mathbf{s}}_t} +\epsilon}, \nonumber +\]
    + +
    +
    +\[ +\begin{equation} +\label{_auto6} \tag{7} +\end{equation} +\]
    +

    where \(\beta_1\) and \(\beta_2\) set the memory lifetime of the first and +second moment and are typically taken to be \(0.9\) and \(0.99\) +respectively, and \(\eta\) and \(\epsilon\) are identical to RMSprop.

    +

    Like in RMSprop, the effective step size of a parameter depends on the +magnitude of its gradient squared. To understand this better, let us +rewrite this expression in terms of the variance +\(\boldsymbol{\sigma}_t^2 = \boldsymbol{\mathbf{s}}_t - +(\boldsymbol{\mathbf{m}}_t)^2\). Consider a single parameter \(\theta_t\). The +update rule for this parameter is given by

    +
    +\[ +\Delta \theta_{t+1}= -\eta_t { \boldsymbol{m}_t \over \sqrt{\sigma_t^2 + m_t^2 }+\epsilon}. +\]
    +
    +
    +
    +

    7.9. Practical tips¶

    +
      +

    • Randomize the data when making mini-batches. It is always important to randomly shuffle the data when forming mini-batches. Otherwise, the gradient descent method can fit spurious correlations resulting from the order in which data is presented.

      • +

    • Transform your inputs. Learning becomes difficult when our landscape has a mixture of steep and flat directions. One simple trick for minimizing these situations is to standardize the data by subtracting the mean and normalizing the variance of input variables. Whenever possible, also decorrelate the inputs. To understand why this is helpful, consider the case of linear regression. It is easy to show that for the squared error cost function, the Hessian of the cost function is just the correlation matrix between the inputs. Thus, by standardizing the inputs, we are ensuring that the landscape looks homogeneous in all directions in parameter space. Since most deep networks can be viewed as linear transformations followed by a non-linearity at each layer, we expect this intuition to hold beyond the linear case.

      • +

    • Monitor the out-of-sample performance. Always monitor the performance of your model on a validation set (a small portion of the training data that is held out of the training process to serve as a proxy for the test set. If the validation error starts increasing, then the model is beginning to overfit. Terminate the learning process. This early stopping significantly improves performance in many settings.

      • +

    • Adaptive optimization methods don’t always have good generalization. Recent studies have shown that adaptive methods such as ADAM, RMSPorp, and AdaGrad tend to have poor generalization compared to SGD or SGD with momentum, particularly in the high-dimensional limit (i.e. the number of parameters exceeds the number of data points). Although it is not clear at this stage why these methods perform so well in training deep neural networks, simpler procedures like properly-tuned SGD may work as well or better in these applications.

      • +
    +
    +
    +

    7.10. Automatic differentiation¶

    +

    Automatic differentiation (AD), +also called algorithmic +differentiation or computational differentiation,is a set of +techniques to numerically evaluate the derivative of a function +specified by a computer program. AD exploits the fact that every +computer program, no matter how complicated, executes a sequence of +elementary arithmetic operations (addition, subtraction, +multiplication, division, etc.) and elementary functions (exp, log, +sin, cos, etc.). By applying the chain rule repeatedly to these +operations, derivatives of arbitrary order can be computed +automatically, accurately to working precision, and using at most a +small constant factor more arithmetic operations than the original +program.

    +

    Automatic differentiation is neither:

    +
      +

    • Symbolic differentiation, nor

      • +

    • Numerical differentiation (the method of finite differences).

      • +
    +

    Symbolic differentiation can lead to inefficient code and faces the +difficulty of converting a computer program into a single expression, +while numerical differentiation can introduce round-off errors in the +discretization process and cancellation

    +

    Python has tools for so-called automatic differentiation. +Consider the following example

    +
    +\[ +f(x) = \sin\left(2\pi x + x^2\right) +\]
    +

    which has the following derivative

    +
    +\[ +f'(x) = \cos\left(2\pi x + x^2\right)\left(2\pi + 2x\right) +\]
    +

    Using autograd we have

    +
    +
    +
    import autograd.numpy as np
+
+# To do elementwise differentiation:
+from autograd import elementwise_grad as egrad 
+
+# To plot:
+import matplotlib.pyplot as plt 
+
+
+def f(x):
+    return np.sin(2*np.pi*x + x**2)
+
+def f_grad_analytic(x):
+    return np.cos(2*np.pi*x + x**2)*(2*np.pi + 2*x)
+
+# Do the comparison:
+x = np.linspace(0,1,1000)
+
+f_grad = egrad(f)
+
+computed = f_grad(x)
+analytic = f_grad_analytic(x)
+
+plt.title('Derivative computed from Autograd compared with the analytical derivative')
+plt.plot(x,computed,label='autograd')
+plt.plot(x,analytic,label='analytic')
+
+plt.xlabel('x')
+plt.ylabel('y')
+plt.legend()
+
+plt.show()
+
+print("The max absolute difference is: %g"%(np.max(np.abs(computed - analytic))))
+
    +
    +
    +
    +

    Here we +experiment with what kind of functions Autograd is capable +of finding the gradient of. The following Python functions are just +meant to illustrate what Autograd can do, but please feel free to +experiment with other, possibly more complicated, functions as well.

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+
+def f1(x):
+    return x**3 + 1
+
+f1_grad = grad(f1)
+
+# Remember to send in float as argument to the computed gradient from Autograd!
+a = 1.0
+
+# See the evaluated gradient at a using autograd:
+print("The gradient of f1 evaluated at a = %g using autograd is: %g"%(a,f1_grad(a)))
+
+# Compare with the analytical derivative, that is f1'(x) = 3*x**2 
+grad_analytical = 3*a**2
+print("The gradient of f1 evaluated at a = %g by finding the analytic expression is: %g"%(a,grad_analytical))
+
    +
    +
    +
    +

    To differentiate with respect to two (or more) arguments of a Python +function, Autograd need to know at which variable the function if +being differentiated with respect to.

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f2(x1,x2):
+    return 3*x1**3 + x2*(x1 - 5) + 1
+
+# By sending the argument 0, Autograd will compute the derivative w.r.t the first variable, in this case x1
+f2_grad_x1 = grad(f2,0)
+
+# ... and differentiate w.r.t x2 by sending 1 as an additional arugment to grad
+f2_grad_x2 = grad(f2,1)
+
+x1 = 1.0
+x2 = 3.0 
+
+print("Evaluating at x1 = %g, x2 = %g"%(x1,x2))
+print("-"*30)
+
+# Compare with the analytical derivatives:
+
+# Derivative of f2 w.r.t x1 is: 9*x1**2 + x2:
+f2_grad_x1_analytical = 9*x1**2 + x2
+
+# Derivative of f2 w.r.t x2 is: x1 - 5:
+f2_grad_x2_analytical = x1 - 5
+
+# See the evaluated derivations:
+print("The derivative of f2 w.r.t x1: %g"%( f2_grad_x1(x1,x2) ))
+print("The analytical derivative of f2 w.r.t x1: %g"%( f2_grad_x1(x1,x2) ))
+
+print()
+
+print("The derivative of f2 w.r.t x2: %g"%( f2_grad_x2(x1,x2) ))
+print("The analytical derivative of f2 w.r.t x2: %g"%( f2_grad_x2(x1,x2) ))
+
    +
    +
    +
    +

    Note that the grad function will not produce the true gradient of the function. The true gradient of a function with two or more variables will produce a vector, where each element is the function differentiated w.r.t a variable.

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f3(x): # Assumes x is an array of length 5 or higher
+    return 2*x[0] + 3*x[1] + 5*x[2] + 7*x[3] + 11*x[4]**2
+
+f3_grad = grad(f3)
+
+x = np.linspace(0,4,5)
+
+# Print the computed gradient:
+print("The computed gradient of f3 is: ", f3_grad(x))
+
+# The analytical gradient is: (2, 3, 5, 7, 22*x[4])
+f3_grad_analytical = np.array([2, 3, 5, 7, 22*x[4]])
+
+# Print the analytical gradient:
+print("The analytical gradient of f3 is: ", f3_grad_analytical)
+
    +
    +
    +
    +

    Note that in this case, when sending an array as input argument, the +output from Autograd is another array. This is the true gradient of +the function, as opposed to the function in the previous example. By +using arrays to represent the variables, the output from Autograd +might be easier to work with, as the output is closer to what one +could expect form a gradient-evaluting function.

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f4(x):
+    return np.sqrt(1+x**2) + np.exp(x) + np.sin(2*np.pi*x)
+
+f4_grad = grad(f4)
+
+x = 2.7
+
+# Print the computed derivative:
+print("The computed derivative of f4 at x = %g is: %g"%(x,f4_grad(x)))
+
+# The analytical derivative is: x/sqrt(1 + x**2) + exp(x) + cos(2*pi*x)*2*pi
+f4_grad_analytical = x/np.sqrt(1 + x**2) + np.exp(x) + np.cos(2*np.pi*x)*2*np.pi
+
+# Print the analytical gradient:
+print("The analytical gradient of f4 at x = %g is: %g"%(x,f4_grad_analytical))
+
    +
    +
    +
    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f5(x):
+    if x >= 0:
+        return x**2
+    else:
+        return -3*x + 1
+
+f5_grad = grad(f5)
+
+x = 2.7
+
+# Print the computed derivative:
+print("The computed derivative of f5 at x = %g is: %g"%(x,f5_grad(x)))
+
    +
    +
    +
    +

    2 +6

    +

    < +< +< +! +! +C +O +D +E +_ +B +L +O +C +K

    +

    p +y +c +o +d

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+# Both of the functions are implementation of the sum: sum(x**i) for i = 0, ..., 9
+# The analytical derivative is: sum(i*x**(i-1)) 
+f6_grad_analytical = 0
+for i in range(10):
+    f6_grad_analytical += i*x**(i-1)
+
+print("The analytical derivative of f6 at x = %g is: %g"%(x,f6_grad_analytical))
+
    +
    +
    +
    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+
+def f7(n): # Assume that n is an integer
+    if n == 1 or n == 0:
+        return 1
+    else:
+        return n*f7(n-1)
+
+f7_grad = grad(f7)
+
+n = 2.0
+
+print("The computed derivative of f7 at n = %d is: %g"%(n,f7_grad(n)))
+
+# The function f7 is an implementation of the factorial of n.
+# By using the product rule, one can find that the derivative is:
+
+f7_grad_analytical = 0
+for i in range(int(n)-1):
+    tmp = 1
+    for k in range(int(n)-1):
+        if k != i:
+            tmp *= (n - k)
+    f7_grad_analytical += tmp
+
+print("The analytical derivative of f7 at n = %d is: %g"%(n,f7_grad_analytical))
+
    +
    +
    +
    +

    Note that if n is equal to zero or one, Autograd will give an error message. This message appears when the output is independent on input.

    +

    Autograd supports many features. However, there are some functions that is not supported (yet) by Autograd.

    +

    Assigning a value to the variable being differentiated with respect to

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f8(x): # Assume x is an array
+    x[2] = 3
+    return x*2
+
+f8_grad = grad(f8)
+
+x = 8.4
+
+print("The derivative of f8 is:",f8_grad(x))
+
    +
    +
    +
    +

    Here, Autograd tells us that an ‘ArrayBox’ does not support item assignment. The item assignment is done when the program tries to assign x[2] to the value 3. However, Autograd has implemented the computation of the derivative such that this assignment is not possible.

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f9(a): # Assume a is an array with 2 elements
+    b = np.array([1.0,2.0])
+    return a.dot(b)
+
+f9_grad = grad(f9)
+
+x = np.array([1.0,0.0])
+
+print("The derivative of f9 is:",f9_grad(x))
+
    +
    +
    +
    +

    Here we are told that the ‘dot’ function does not belong to Autograd’s +version of a Numpy array. To overcome this, an alternative syntax +which also computed the dot product can be used:

    +
    +
    +
    import autograd.numpy as np
+from autograd import grad
+def f9_alternative(x): # Assume a is an array with 2 elements
+    b = np.array([1.0,2.0])
+    return np.dot(x,b) # The same as x_1*b_1 + x_2*b_2
+
+f9_alternative_grad = grad(f9_alternative)
+
+x = np.array([3.0,0.0])
+
+print("The gradient of f9 is:",f9_alternative_grad(x))
+
+# The analytical gradient of the dot product of vectors x and b with two elements (x_1,x_2) and (b_1, b_2) respectively
+# w.r.t x is (b_1, b_2).
+
    +
    +
    +
    +

    The documentation recommends to avoid inplace operations such as

    +
    +
    +
    a += b
+a -= b
+a*= b
+a /=b
+
    +
    +
    +
    +

    More examples will be added, in particular how to compare autograd with own codes for the gradients.

    @@ -2414,7 +2738,7 @@ function.

    5. Resampling Methods - 7. Support Vector Machines, overarching aims + 8. Support Vector Machines, overarching aims
    diff --git a/doc/LectureNotes/_build/html/chapter5.html b/doc/LectureNotes/_build/html/chapter5.html index f4d68f250..288731c0c 100644 --- a/doc/LectureNotes/_build/html/chapter5.html +++ b/doc/LectureNotes/_build/html/chapter5.html @@ -5,7 +5,7 @@ - 7. Support Vector Machines, overarching aims — Applied Data Analysis and Machine Learning + 8. Support Vector Machines, overarching aims — Applied Data Analysis and Machine Learning @@ -54,7 +54,7 @@ - + @@ -139,17 +139,17 @@ @@ -171,12 +171,12 @@ @@ -188,12 +188,12 @@ @@ -205,12 +205,12 @@ @@ -286,49 +286,49 @@ @@ -343,7 +343,7 @@
    -

    7. Support Vector Machines, overarching aims¶

    +

    8. Support Vector Machines, overarching aims¶

    A Support Vector Machine (SVM) is a very powerful and versatile Machine Learning method, capable of performing linear or nonlinear classification, regression, and even outlier detection. It is one of @@ -365,7 +365,7 @@ below. We distinguish also between linear and non-linear approaches. The latter are the most frequent ones since it is rather unlikely that we can separate classes easily by say straight lines.

    -

    7.1. Hyperplanes and all that¶

    +

    8.1. Hyperplanes and all that¶

    The theory behind support vector machines (SVM hereafter) is based on the mathematical description of so-called hyperplanes. Let us start with a two-dimensional case. This will also allow us to introduce our @@ -509,7 +509,7 @@ y_i\left(b+w_1x_{i1}+w_2x_{i2}+\dots +w_px_{ip}\right) > 0. \]

    When we try to separate hyperplanes, if it exists, we can use it to construct a natural classifier: a test observation is assigned a given class depending on which side of the hyperplane it is located.

    -

    7.1.1. The two-dimensional case¶

    +

    8.1.1. The two-dimensional case¶

    Let us try to develop our intuition about SVMs by limiting ourselves to a two-dimensional plane. To separate the two classes of data points, there are many possible lines (hyperplanes if you prefer a more strict naming)
    @@ -578,7 +578,7 @@ properly into two distinct classes, we may not experience a converge at all.

    -

    7.1.2. A better approach¶

    +

    8.1.2. A better approach¶

    A better approach is rather to try to define a large margin between the two classes (if they are well separated from the beginning).

    Thus, we wish to find a margin \(M\) with \(\boldsymbol{w}\) normalized to @@ -611,7 +611,7 @@ about Lagrangian multipliers.

    -

    7.2. A quick Reminder on Lagrangian Multipliers¶

    +

    8.2. A quick Reminder on Lagrangian Multipliers¶

    Consider a function of three independent variables \(f(x,y,z)\) . For the function \(f\) to be an extreme we have

    @@ -767,7 +767,7 @@ y_i = \mathrm{sign}(\boldsymbol{w}^T\boldsymbol{x}_i+b).

    Below we discuss how to find the optimal values of \(\lambda_i\). Before we proceed however, we discuss now the so-called soft classifier.

    -

    7.3. A soft classifier¶

    +

    8.3. A soft classifier¶

    Till now, the margin is strictly defined by the support vectors. This defines what is called a hard classifier, that is the margins are well defined.

    Suppose now that classes overlap in feature space, as shown in the figure here. One way to deal with this problem before we define the @@ -851,7 +851,7 @@ y_i(\boldsymbol{w}^T\boldsymbol{x}_i+b) -(1-\xi_) \geq 0 \hspace{0.1cm}\forall i \]

    -

    7.4. Kernels and non-linearity¶

    +

    8.4. Kernels and non-linearity¶

    The cases we have studied till now, were all characterized by two classes with a close to linear separability. The classifiers we have described so far find linear boundaries in our input feature space. It is @@ -984,7 +984,7 @@ Given a kernel \(K\) and the t \(0\leq \lambda_i\) and \(\lambda_i \leq C\). These two inequalities define then the matrix \(\boldsymbol{G}\) and the vector \(\boldsymbol{h}\).

    -

    7.5. Different kernels and Mercer’s theorem¶

    +

    8.5. Different kernels and Mercer’s theorem¶

    There are several popular kernels being used. These are

    1. Linear: \(K(\boldsymbol{x},\boldsymbol{y})=\boldsymbol{x}^T\boldsymbol{y}\),

      2. @@ -1010,7 +1010,7 @@ don’t respect all of Mercer’s conditions, yet they generally work well in practice.

    -

    7.6. The moons example¶

    +

    8.6. The moons example¶

    from __future__ import division, print_function, unicode_literals
@@ -1222,7 +1222,7 @@ Phi(-1.0, 1) = [0.30119421]
    -

    7.7. Mathematical optimization of convex functions¶

    +

    8.7. Mathematical optimization of convex functions¶

    A mathematical (quadratic) optimization problem, or just optimization problem, has the form

    \[\begin{split} @@ -1367,7 +1367,7 @@ With the slack constants this leads to the additional constraint 6. Logistic Regression - 8. Decision trees, overarching aims + 9. Decision trees, overarching aims
    diff --git a/doc/LectureNotes/_build/html/chapter6.html b/doc/LectureNotes/_build/html/chapter6.html index 5b28a4d7f..a7a382fbf 100644 --- a/doc/LectureNotes/_build/html/chapter6.html +++ b/doc/LectureNotes/_build/html/chapter6.html @@ -5,7 +5,7 @@ - 8. Decision trees, overarching aims — Applied Data Analysis and Machine Learning + 9. Decision trees, overarching aims — Applied Data Analysis and Machine Learning @@ -54,8 +54,8 @@ - - + + @@ -139,17 +139,17 @@ @@ -171,12 +171,12 @@ @@ -188,12 +188,12 @@ @@ -205,12 +205,12 @@ @@ -286,100 +286,100 @@

    A good read is for example Vidal, Ma and Sastry.

    -

    10.1. Introducing the Covariance and Correlation functions¶

    +

    11.1. Introducing the Covariance and Correlation functions¶

    Before we discuss the PCA theorem, we need to remind ourselves about the definition of the covariance and the correlation function. These are quantities

    Suppose we have defined two vectors @@ -532,7 +532,7 @@ covariance matrix through the np.linalg.eig() function.

    -

    10.2. Correlation Matrix¶

    +

    11.2. Correlation Matrix¶

    The previous example can be converted into the correlation matrix by simply scaling the matrix elements with the variances. We should also subtract the mean values for each column. This leads to the following @@ -752,7 +752,7 @@ x_{10}x_{00}+x_{11}x_{01} & x_{10}^2+x_{11}^2\\

    It is easy to generalize this to a matrix \(\boldsymbol{X}\in {\mathbb{R}}^{n\times p}\).

    -

    10.3. Towards the PCA theorem¶

    +

    11.3. Towards the PCA theorem¶

    We have that the covariance matrix (the correlation matrix involves a simple rescaling) is given as

    \[ @@ -790,7 +790,7 @@ dimensions where the eigenvalues are very small. If \(p\) to \(l << p\) and handle only \(l\) features/predictors.

    -

    10.3.1. The Algorithm before theorem¶

    +

    11.3.1. The Algorithm before theorem¶

    Here’s how we would proceed in setting up the algorithm for the PCA, see also discussion below here.

    • Set up the datapoints for the design/feature matrix \(\boldsymbol{X}\) with \(\boldsymbol{X}\in {\mathbb{R}}^{n\times p}\), with the predictors/features \(p\) referring to the column numbers and the entries \(n\) being the row elements.

      • @@ -815,7 +815,7 @@ x_{n-1,0} & x_{n-1,1} & x_{n-1,2}& \dots & \dots x_{n-1,p-1}\\
    -

    10.3.2. Writing our own PCA code¶

    +

    11.3.2. Writing our own PCA code¶

    We will use a simple example first with two-dimensional data drawn from a multivariate normal distribution with the following mean and covariance matrix (we have fixed these quantities but will play around with them below):

    @@ -937,7 +937,7 @@ Our own code here is not very elegant and asks for obvious improvements. It is t The plot shows how the data are clustered around a line with slope close to one. Is this expected? Try to change the covariance and the mean values. For example, try to make the variance of the first element much larger than that of the second diagonal element. Try also to shrink the covariance (the non-diagonal elements) and see how the data points are distributed.

    -

    10.3.3. Diagonalize the sample covariance matrix to obtain the principal components¶

    +

    11.3.3. Diagonalize the sample covariance matrix to obtain the principal components¶

    Now we are ready to solve for the principal components! To do so we diagonalize the sample covariance matrix \(\Sigma\). We can use the function np.linalg.eig to do so. It will return the eigenvalues and @@ -1005,7 +1005,7 @@ Second eigenvector

    -

    10.4. Classical PCA Theorem¶

    +

    11.4. Classical PCA Theorem¶

    We assume now that we have a design matrix \(\boldsymbol{X}\) which has been centered as discussed above. For the sake of simplicity we skip the overline symbol. The matrix is defined in terms of the various column @@ -1065,7 +1065,7 @@ chapter 12.4 and discussion therein.

    For more details, see for example Vidal, Ma and Sastry, chapter 2.

    -

    10.6. PCA and scikit-learn¶

    +

    11.6. PCA and scikit-learn¶

    Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The following code applies PCA to reduce the dimensionality of the dataset down to two dimensions (note that it automatically takes care of centering the data):

    @@ -1299,7 +1299,7 @@ available via the \(explained\_varian variance that lies along the axis of each principal component.

    -

    10.7. Back to the Cancer Data¶

    +

    11.7. Back to the Cancer Data¶

    We can now repeat the above but applied to real data, in this case our breast cancer data. Here we compute performance scores on the training data using logistic regression.

    @@ -1382,7 +1382,7 @@ a float between 0.0 and 1.0, indicating the ratio of variance you wish to preser
    -

    10.7.1. Incremental PCA¶

    +

    11.7.1. Incremental PCA¶

    One problem with the preceding implementation of PCA is that it requires the whole training set to fit in memory in order for the SVD algorithm to run. Fortunately, Incremental PCA (IPCA) algorithms have been developed: you can split the training set into mini-batches and feed an IPCA algorithm one minibatch @@ -1390,14 +1390,14 @@ at a time. This is useful for large training sets, and also to apply PCA online instances arrive).

    -

    10.7.2. Randomized PCA¶

    +

    11.7.2. Randomized PCA¶

    Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic algorithm that quickly finds an approximation of the first d principal components. Its computational complexity is \(O(m \times d^2)+O(d^3)\), instead of \(O(m \times n^2) + O(n^3)\), so it is dramatically faster than the previous algorithms when \(d\) is much smaller than \(n\).

    -

    10.7.3. Kernel PCA¶

    +

    11.7.3. Kernel PCA¶

    The kernel trick is a mathematical technique that implicitly maps instances into a very high-dimensional space (called the feature space), enabling nonlinear classification and regression with Support Vector Machines. Recall that a linear decision boundary in the high-dimensional feature @@ -1419,7 +1419,7 @@ For example, the following code uses Scikit-Learn’s KernelPCA class to perform

    -

    10.8. Other techniques¶

    +

    11.8. Other techniques¶

    There are many other dimensionality reduction techniques, several of which are available in Scikit-Learn.

    Here are some of the most popular:

    @@ -170,12 +170,12 @@ @@ -187,12 +187,12 @@ @@ -204,12 +204,12 @@ diff --git a/doc/LectureNotes/_build/html/reports/chapter4.log b/doc/LectureNotes/_build/html/reports/chapter4.log index 64735e6fa..ab0fdb8ff 100644 --- a/doc/LectureNotes/_build/html/reports/chapter4.log +++ b/doc/LectureNotes/_build/html/reports/chapter4.log @@ -17,61 +17,16 @@ Traceback (most recent call last): raise CellExecutionError.from_cell_and_msg(cell, exec_reply['content']) nbclient.exceptions.CellExecutionError: An error occurred while executing the following cell: ------------------ -import matplotlib.pyplot as plt -import numpy as np -from sklearn.model_selection import train_test_split -from sklearn.datasets import load_breast_cancer -from sklearn.linear_model import LogisticRegression - -# Load the data -cancer = load_breast_cancer() - -X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0) -print(X_train.shape) -print(X_test.shape) -# Logistic Regression -logreg = LogisticRegression(solver='lbfgs') -logreg.fit(X_train, y_train) -print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test))) -#now scale the data -from sklearn.preprocessing import StandardScaler -scaler = StandardScaler() -scaler.fit(X_train) -X_train_scaled = scaler.transform(X_train) -X_test_scaled = scaler.transform(X_test) -# Logistic Regression -logreg.fit(X_train_scaled, y_train) -print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test))) - - -from sklearn.preprocessing import LabelEncoder -from sklearn.model_selection import cross_validate -#Cross validation -accuracy = cross_validate(logreg,X_test_scaled,y_test,cv=10)['test_score'] -print(accuracy) -print("Test set accuracy with Logistic Regression and scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test))) - - -import scikitplot as skplt -y_pred = logreg.predict(X_test_scaled) -skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True) -plt.show() -y_probas = logreg.predict_proba(X_test_scaled) -skplt.metrics.plot_roc(y_test, y_probas) -plt.show() -skplt.metrics.plot_cumulative_gain(y_test, y_probas) -plt.show() +x = 2*np.random.rand(m,1) +y = 4+3*x+np.random.randn(m,1) ------------------ --------------------------------------------------------------------------- -ModuleNotFoundError Traceback (most recent call last) - in  - 34  - 35  ----> 36 import scikitplot as skplt - 37 y_pred = logreg.predict(X_test_scaled) - 38 skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True) +NameError Traceback (most recent call last) + in  +----> 1 x = 2*np.random.rand(m,1) + 2 y = 4+3*x+np.random.randn(m,1) -ModuleNotFoundError: No module named 'scikitplot' -ModuleNotFoundError: No module named 'scikitplot' +NameError: name 'm' is not defined +NameError: name 'm' is not defined diff --git a/doc/LectureNotes/_build/html/search.html b/doc/LectureNotes/_build/html/search.html index 1ff418ad7..f5d056516 100644 --- a/doc/LectureNotes/_build/html/search.html +++ b/doc/LectureNotes/_build/html/search.html @@ -162,7 +162,7 @@
  • - 7. Support Vector Machines, overarching aims + 8. Support Vector Machines, overarching aims
    • @@ -174,12 +174,12 @@ @@ -191,12 +191,12 @@ @@ -208,12 +208,12 @@ diff --git a/doc/LectureNotes/_build/html/searchindex.js b/doc/LectureNotes/_build/html/searchindex.js index 08e6b3d20..18ce0c3c5 100644 --- a/doc/LectureNotes/_build/html/searchindex.js +++ b/doc/LectureNotes/_build/html/searchindex.js @@ -1 +1 @@ -Search.setIndex({docnames:["Clustering","chapter1","chapter10","chapter2","chapter3","chapter4","chapter5","chapter6","chapter7","chapter8","chapter9","content","intro","linalg","schedule","statistics","teachers","textbooks"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":3,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":2,"sphinx.domains.rst":2,"sphinx.domains.std":1,"sphinx.ext.intersphinx":1,sphinx:56},filenames:["Clustering.ipynb","chapter1.ipynb","chapter10.ipynb","chapter2.ipynb","chapter3.ipynb","chapter4.ipynb","chapter5.ipynb","chapter6.ipynb","chapter7.ipynb","chapter8.ipynb","chapter9.ipynb","content.md","intro.md","linalg.ipynb","schedule.md","statistics.ipynb","teachers.md","textbooks.md"],objects:{},objnames:{},objtypes:{},terms:{"000":2,"000000":[1,3,9],"00000000e":3,"0001":2,"00015921913736794912":7,"00019998":3,"00024087":3,"00029012":3,"00034944":3,"00042089":3,"00050694":3,"00061058":3,"00073541":3,"00079968":3,"00088573":3,"00096314":3,"001":[2,3,5,6],"00106677":3,"00109273":[],"00115999":3,"00128479":3,"00139705":3,"00154733":3,"00168251":3,"00186347":3,"00200":6,"00202624":3,"00224413":3,"0024401":3,"00270244":3,"00293838":3,"0032542":3,"0032873138755776365":15,"00353823":3,"003704":1,"003717":1,"003759":1,"003774":1,"003788":1,"00391839":3,"004":3,"00426027":3,"00433417":9,"004456043243408203":0,"00445655":9,"00471782":3,"0049999999999999845":[],"004999999999999996":[],"004999999999999997":[],"0050000000000000044":1,"005000000000000007":[],"00512927":3,"0056799":3,"00617499":3,"00683748":3,"006968283076248407":1,"00727646693":1,"00728323e":[],"007315":1,"0074331":3,"007789":[],"007824":1,"007931713723267314":1,"00823002":3,"0086649156":1,"008675369724976777":3,"008885578722629236":[],"00890601":3,"00894639":3,"009154":1,"009163470508352228":3,"009735":9,"00973536":9,"009790":1,"009883615646716182":[],"009883615646716184":[],"009883615646716186":1,"00990475":3,"009911":1,"01057384067458835":[],"010679893512872646":[],"01076611":3,"0110":15,"011347":1,"01191824":3,"01210814993019007":[],"01295356":3,"0140617":[],"01433809":3,"01503674":[],"015037":[],"01514394":[],"015144":[],"01558197":3,"0169643":3,"0170259":3,"01724499":3,"018232":1,"01825571":3,"01867234e":15,"01873344":9,"01873869":3,"02073509":3,"021544346900318832":3,"02161783e":15,"021901":9,"02190139":9,"02252765":3,"02492265":3,"02493044e":[],"02568378":[],"025709":[1,9],"02625193":6,"02707227":3,"02730126581656065":15,"02881304":[],"029688":[],"02968834":[],"029733":1,"02994311":3,"03019554":[],"030196":[],"030670":[],"03067028":[],"03099776":3,"031":3,"03193199":3,"03251863":3,"03256632e":2,"03303359":[],"033037772005753835":[],"03308408":3,"034342":3,"0344707":3,"0359565":3,"03607832":[],"03652792":[],"036528":[],"03697069":[],"03707133":9,"03750367":13,"03787596148305236":1,"03791824e":15,"038300":[1,9],"03871832":[],"039039":3,"04205220e":15,"04260073e":[],"04315108":3,"04330858":[],"04346721":3,"04421672e":[],"044334":1,"044402":1,"04455272":[],"044553":[],"04529095":3,"0458":7,"04648335":3,"04683565":3,"04757618e":[],"04829457939163212":15,"051115":[],"051649":[1,9],"0517473":3,"052645":[],"053053":[],"0530533":[],"05318162":[],"053182":[],"05364854":6,"05396545e":[],"054187":[],"054276":[],"054674":[],"055":[],"055085":[],"05531364":3,"055314":3,"055676":1,"055744":[],"055774":[],"055831":[],"055887":[],"055916":[],"05614483":3,"056326":[],"056329":9,"056484":[],"057035":[],"057088709963637":1,"057320":[],"057332":[],"05755374":[],"057571":[],"057654":[],"057919":[],"05802766":1,"058178":[],"058247":3,"058282":[],"058719":3,"058805":[],"058892":[],"059225":3,"059245":[],"059344":9,"059406":[],"059455":[],"059531":[],"059571":[],"059680303635344434":3,"059765":3,"059806":[],"059865":[],"059908":[],"059994":[],"060064":9,"060070":9,"060308":[],"060335":3,"06041458e":[],"060520":[],"060919":[],"061164":[],"061176":[],"061255":[],"06126507e":15,"061330":[],"061601":9,"061652":[],"061774":[],"061842":[],"061855":[],"061896":[],"06200174":3,"06209126":[],"062188":[],"062327":[],"062376":[],"062639":[],"062659":[],"062660":9,"062731":[],"062822":[],"062829":[],"062879":[],"062885":[],"062923":9,"062979":[],"063055":[],"063097":[],"063122":[],"063178":9,"063190":3,"063324":9,"063395":[],"063415":[],"063523":3,"063526":[],"063615":[],"063670":[],"063727":[],"063759":[],"063763":[],"063880":3,"063882":[],"063925":3,"063982":[],"064000":[],"064013":3,"064075":3,"064078":3,"064101":3,"06412177e":15,"064156":3,"064257":3,"064273":3,"064304":[],"064354":[],"064357":3,"064393":3,"064410":[],"064449":3,"064472":[],"064591":3,"064647":3,"064648":9,"064686":[],"064689":3,"064783":9,"064795":3,"06492086e":[],"064929":[],"064958":3,"064964":[],"064982":[],"065":[],"065003":[],"065086":[],"065110":9,"065144":[],"065151":[],"065188":3,"065200":[],"065202":[],"065204":[],"065355":[],"065591":[],"065618":[],"065660":[],"065680":[],"065762":[],"065860":9,"065994":[],"066002":9,"06600226":9,"066030":9,"066047":[],"066074":9,"066092":[],"066116":[],"06619182206626131":[],"066249":[],"066272":[],"066294":[],"066312":9,"066389":[],"066403":9,"066487":[],"066629":[],"066683":9,"066705":3,"066761":[],"066788":[],"066835":[],"066837":[1,9],"066951":9,"066963":[],"067043":[],"067061":[],"067123":[],"067132":9,"067175":9,"067202":3,"06724062":3,"067355":3,"067440":9,"067504":[],"067531":[],"067577":[],"067601":[],"067610":[],"067667":9,"067685":9,"067711":3,"067718":3,"067769":[],"067777":[],"067794":3,"067844":[],"067865":[],"067889":3,"067892":[],"067985":[],"067990":[],"068056":3,"068073":[],"068165":3,"068185":[],"068213":3,"068244":3,"068276":[],"068293":3,"068351":9,"068385":3,"068427":3,"068460":[],"068491":3,"068514":[],"068593":3,"068640":3,"068654":[],"068667":3,"068673":3,"068686":3,"068697":9,"068717":3,"068818":9,"068821":3,"068828":3,"068838":9,"068840":[],"068873":3,"068896":3,"068920":[],"068928":3,"068945":3,"069005":[],"069028":[],"069071":3,"069081":[],"069098":[],"069195":9,"069243":9,"069282":[],"069369":[],"069413":9,"069496":3,"069503":[],"069510":[],"069516":[],"069528":[],"069542":3,"069584":1,"069635":[],"069641":9,"069775":9,"069814":[],"069848":[],"069853":3,"069859":[],"069872":[],"069890":[],"069902":[],"069988":9,"070043":1,"070103":[],"070129":[],"070131":[],"070138":[],"070166":3,"070182":[],"070188":[],"070254":[],"070337":9,"070456":[],"070473":3,"070517":[],"070548":9,"070653":[],"070762":9,"070764":3,"070791":9,"070812":[],"070815":[],"070867":9,"070889":[1,9],"070949":3,"070967":[],"071016":[],"071049":9,"07106781e":[],"071096":[],"07110274":13,"071156":3,"071248":[],"071268":[],"0713":1,"071313":[],"071325":[],"071359":3,"071435":9,"07145103":9,"071481":[],"071502":[],"071547":9,"071549":3,"07155335":[],"071576":[],"071579":[],"07159175":[],"07160764":3,"071660":9,"071667":[],"071681":[],"071684":[],"071714":3,"071761":[],"071855":[],"071887":[],"071917":[],"072012":3,"072094":[],"072198":[],"072246":3,"072279":[],"072424":[],"072468":3,"072471":[],"072555":9,"072589":[],"072598":[],"072620":[],"072654":9,"072661":3,"072710":9,"072826":[],"072879":9,"072910":[],"073":[],"073020":[],"073035":[],"073096":9,"073310":9,"073333":[],"073352":[],"073371":9,"073372":[],"073483":[],"073567":[],"073591":3,"073598":[1,9],"073602":[],"073607":3,"073656":9,"073681":3,"073695":3,"073765":9,"073851":[],"073902":[],"073915":1,"073926":3,"073966":[],"074":[],"074027":[],"074044":3,"074067":[1,9],"074077":3,"074096":[],"074099":[],"074134":[],"074170":[],"074191":[],"074201":9,"074207":[],"07421084":3,"074230":[],"07423848370736122":[],"074248":[],"074285":3,"074286":[],"074331":9,"074432":[],"074434":3,"074453":[],"074488":[],"07456491":3,"074568":[],"074618":9,"074665":[],"074693":3,"074696":[],"074848":[],"074979":9,"075084":[],"075119":[],"075163":[],"075194":9,"075212":[],"07521771":[],"075218":[],"075261":[],"075266":[],"075289":[],"075296":9,"075309":[],"075526":9,"075545":9,"0755452":9,"075562":[],"075567":3,"075615":9,"075754":[],"075772":[],"07581375":[],"075814":[],"075818":3,"075821":[],"075865":[],"075975":3,"075993":9,"076026":3,"076062":[],"076166":3,"076169":3,"076189":[],"076251":3,"076283":[],"076320":[],"076367":9,"076398":[],"076408":[],"076410":9,"076466":9,"076469":[],"076730":[],"076822":3,"076938":[1,9],"077010":9,"077046":9,"077143":9,"077160":[],"077212":[],"07729012413236423":9,"077303":[],"077305":9,"077338":3,"077339":3,"077386":[],"077407":3,"077416":3,"07744472306026946":7,"077469":[],"077533":[],"077623":[],"077771":9,"07777777777777778":2,"077891":[],"077927":9,"078049":[],"078076":[],"078099":9,"078103":9,"078197":3,"078244":9,"078279":[],"078311":[],"078388":[],"078399":[],"07839924":[],"078451":[],"078474":3,"078649":3,"078674":[],"078693":9,"078710":9,"078746":9,"078750":3,"078892":3,"078927":[],"078930":[],"078937":[],"078966":[],"079001":[],"079037":[],"079137":[],"079139":[],"079333":[],"079365":[],"07944154":13,"079624":[],"079785":9,"079848":9,"079956":[],"080024":9,"080084":1,"080115":[],"080182":3,"080223":[],"080264":9,"080325":9,"080336":[],"080381":[],"080387":[],"080398":[],"08043851":3,"080570":9,"080626":9,"080675":9,"080738":3,"080845":[],"080871":[],"081077":9,"081102":3,"081103":[],"081129":9,"081210":9,"081227":[],"081260":[],"081309":3,"081402":[],"081480":[],"081489":9,"081514":[],"08170444":[],"081816":[],"081886":[],"082091":3,"082211":9,"082225":9,"082247":9,"082272":3,"08248290e":[],"08251898":15,"082616":3,"082687":[],"08271336":[],"082866":[],"082990":[],"083128":[],"08318298e":2,"0832":[],"083246":3,"083295":9,"083376":[],"083527":1,"083577":[],"083618":[],"08362077210702115":3,"083657":[],"083658":9,"083728":[],"083775":[],"084092":[],"084103":3,"08415761":[],"084167":[],"084181":[],"084249":9,"084251":[],"084604":9,"084813":9,"084873":[],"084931":[],"08501147":[],"085121":9,"085259":3,"085285":[],"085289":[],"085597":[],"085646":9,"085676":[],"085928":[],"085951":[],"08611111111111111":2,"086116":[],"086325":[],"086465":[],"086592":[],"08665060086846632":[],"086652":9,"086974":9,"087202":9,"087482":[],"087563":9,"0881981":3,"08839750e":[],"088660":[],"08871404":3,"088871":[],"08888888888888889":2,"089300":[],"089425":9,"089504":[],"089916":[],"089931":[],"090274":9,"09103481e":15,"091159":[],"09166666666666666":2,"0917":7,"091979":[],"093755":9,"09487315108590391":9,"095173":[],"095472":[],"09609807":3,"096726":[],"09672604":[],"09678277e":15,"09741585e":[],"097452":[],"09903804":6,"099763":[],"09976319":[],"100":[0,1,2,3,4,5,6,7,8,9,13,15,16],"1000":[0,1,2,5,6,9,12,15],"10000":[0,3,4,8,9,15],"100000":6,"10001":8,"1001":15,"1002":15,"1003":15,"1005":15,"1007125":[],"1009":15,"10094646e":[],"1011":15,"1013":15,"10131725":[],"1013904243":15,"1015":15,"102":1,"1022964509394572":2,"1023":15,"1026":15,"1027":15,"103":2,"1030":15,"10307631":[],"10327559":13,"1037":15,"10378326e":2,"1038":15,"1040":15,"10405456":9,"104411":1,"1047":15,"105137868830763":15,"10555555555555556":2,"10589577":3,"106095":[1,9],"108":1,"10806972":[],"108070":[],"10896672e":15,"10898112e":[],"10th":7,"10x":1,"110":1,"1100":15,"1101":15,"11022302e":3,"111":[2,5,10],"1111111111111111":2,"112283":1,"112383":[1,9],"11304709e":15,"11388888888888889":2,"11456076e":[],"11462415":3,"11507992e":2,"11666666666666667":2,"117":6,"118318":1,"11944444444444445":2,"12002944":[],"1203284":6,"121":[6,7,8],"122":[6,7,8],"12222222222222222":2,"12283463":[],"123190":1,"123459876":15,"12366979e":15,"124":1,"12497018e":[],"12618549":3,"127773":15,"12777777777777777":2,"1298":7,"12adb44b1c20":5,"13055555555555556":2,"13060772":[],"13162821e":[],"13209041":[],"13266452e":[],"133":5,"13328820e":[],"13457922":[],"13519106":1,"13579199e":[],"1361111111111111":2,"13646574":3,"136687":4,"137268":[],"137400784702912":1,"137652":[1,9],"13803928":[],"13865173":3,"138775":[1,9],"13891144e":[],"139462":[],"139475":[],"1404":1,"143":5,"1437":2,"1440501043841336":2,"1445":4,"14459063":[],"1446":4,"1447":4,"1448":4,"1449":4,"14662074":[],"14722222222222223":2,"147400":[1,9],"147420":[1,9],"147722":[],"148":[],"149076":4,"149366":[],"149667":[],"149832":4,"149894":[],"149903":[],"14g":4,"150":6,"15005476":3,"15024669":[],"151":[],"152636":[],"1527777777777778":2,"153036":1,"153106":1,"15324631":[],"15332528e":[],"154720":1,"15475232":[],"154911":1,"155491":1,"155687":1,"155883":1,"156":1,"157":1,"15751596":3,"158":1,"159":1,"15979239e":15,"15g":4,"160":1,"16111111111111112":2,"16211139":3,"16342407":3,"16496581e":[],"16553696":[],"16637855e":15,"16666666666666666":2,"16807":15,"16831012":3,"16861772":[],"168618":[],"16b8e3cda33a":2,"17117385":[],"17174962e":2,"17234827e":[],"17248441e":[],"17385778e":[],"174497":[],"17449733":[],"175300":1,"1758506":[],"17654307":[],"17707436":[],"17777777777777778":2,"17801022":3,"17917768":3,"17949575":3,"17953942":9,"1797":2,"180092462880674":[],"18029127":3,"18063928":3,"18220995":1,"18224665":3,"18333333333333332":2,"18404906e":[],"184519":1,"1856411":[],"18611111111111112":2,"18620601e":[],"18673098":9,"189496":1,"18954529":[],"189622":1,"18968431e":15,"1914224774238273":[],"19207979":3,"1940":1,"1943":10,"197":1,"1970":13,"1973":7,"197370":[1,9],"19742904e":[],"1979":4,"1989":15,"19937":15,"19955871":[],"199559":[],"19972087e":15,"1_1":10,"1_2":10,"1_3":10,"1cm":[1,6,8,15],"1e10":0,"200":[1,6,7,8,15],"2000":1,"200000":1,"2004":5,"2006":17,"2010":2,"2011":2,"2015":2,"2016":1,"2018":[1,4],"201841":[],"20184113":[],"2021":0,"20277777777777778":2,"203757":[],"20375701":[],"205466494327873":[],"20632410e":[],"207545":[1,9],"20772452":[],"20819609e":[],"20833333333333334":2,"20843563e":[],"20906175e":15,"210340":[1,9],"21058097":3,"212":[],"21208310e":[],"212327334149492":[],"2125":[],"2126":[],"2127":[],"2128":[],"2129":[],"213":[],"213103":[1,9],"21347282":[],"213743":[1,9],"214":[],"2147483647":15,"215":[],"216":[],"21623344":[],"216290":[1,9],"216683":[1,9],"22044605e":3,"22076362":[],"221":6,"221180":1,"22209371e":[],"222400":1,"22443803":[],"22527008e":15,"22690428":3,"22717936e":15,"22847924":3,"22999784":[],"229998":[],"23117916e":15,"23167717":3,"232435":1,"23333333333333334":2,"2335879":[],"23382086e":[],"23438810e":[],"241262":3,"24126232":3,"2433e":3,"24340751":3,"24444444444444444":2,"24829908":3,"250":[5,7],"25000":1,"250000":1,"250154":1,"251559":[],"2515593":[],"251879":1,"25190225e":[],"25226753e":15,"25240108":[],"252436":1,"253":[],"25303483":[],"253775":1,"254":[],"254509":[],"25450941":[],"25457052":[],"255":[],"255001":1,"2551":1,"256":[],"256962":1,"257":[],"2571699":1,"25726439e":[],"2572e3a4b38d":2,"25803281":[],"259107":[],"25910749":[],"259153":[1,9],"259815":3,"25981533":3,"260227742627244":3,"2627588":13,"26301436":3,"26381865":[],"263819":[],"264":1,"26412372":3,"265":1,"2653678":[],"266":1,"267":1,"26710969":3,"26780278":3,"2683":4,"2684":4,"2685":4,"2686":4,"2687":4,"26890510e":[],"269":1,"26974938e":[],"270":1,"27262964":1,"27296891e":[],"276263":[1,9],"27753165e":15,"27852549":3,"27924636":3,"27n_":15,"280647":[1,9],"28166741":[],"282":1,"282727":[1,9],"28310983":[],"283110":[],"2836":15,"28475098":6,"2861":15,"2873":7,"2882":15,"2886":15,"2890":1,"2892":15,"28971976":9,"289720":9,"290":[],"291":[],"2915":15,"291614":[],"29167186":3,"292":[],"293":[],"2931":[],"294":[],"295656121491569":[],"296247":1,"2968":[],"297219777724628":9,"297260":1,"2980":[],"298273":1,"2983233":13,"298375":1,"2990":[],"299444":9,"29944428":9,"299748":1,"2_1":10,"2_2":10,"2_3":10,"2_i":10,"2_m":15,"2_x":15,"2cm":6,"2ff97f4bf03b":15,"2nd":7,"2x_ix_jy_iy_j":6,"2x_j":6,"2y_i":8,"2y_j":6,"30000":1,"300162456113691":[],"30119421":6,"30626706":1,"306854":[],"30685416":[],"30879705":1,"30968181":[],"31109204e":[],"31318084":3,"31457796":3,"3155":[3,4],"3156929654100207":7,"31608475e":[],"31718909":9,"31730641":[],"317367":9,"31853484":[],"31896852":6,"3200":2,"32047562":1,"323291478597321":15,"3250":2,"32521615e":15,"326238":1,"327631":1,"32938847":13,"32945844e":15,"3304":1,"33066907e":3,"3310":1,"3317":1,"331939":1,"333":5,"333333":1,"3338":1,"3344":1,"33443859e":[],"33544681":[],"33569998e":[],"33861512":[],"338869":[],"33886902":[],"339535706819584":15,"33953571":15,"340782":[1,9],"34114547":3,"34114641e":[],"34172919":[],"342680":1,"3436":1,"3437":1,"344172":1,"34569596":3,"346433":9,"34643337":9,"34902789e":15,"350387":[],"35038714":[],"351636":[1,9],"35176067":[],"35182854":3,"35216172e":[],"35367281e":[],"356399":1,"357508":1,"358869339268145":15,"35886934":15,"359":3,"359640894899012":[],"360":2,"360688":1,"361556":1,"3621311":3,"36436520e":[],"36468301":[],"3655222":3,"369139":[1,9],"36941772":[],"369418":[],"37186301":3,"372889":[],"3728892":[],"37416969":9,"374170":9,"37738324":[],"38207279e":[],"38216436":[],"38629436":13,"38777878e":3,"3893239":3,"389324":3,"38937995e":15,"38986237":[],"39579407":3,"396740":9,"39674043":9,"39706038":3,"397700":[1,9],"39792608e":[],"399836":1,"3cd19a0768e1":[],"4000":17,"401842":[1,9],"40212127":[],"404":1,"40425078e":15,"405890":[1,9],"40702":3,"40708470e":[],"40859":[],"4087793":3,"40902095":[],"412374":3,"41237437":3,"4140e":3,"41433969":3,"41511965e":2,"415634483874318":[],"416694683938511":9,"4171578884124756":0,"41754964":[],"41770932":[],"418506":[1,9],"41876428e":[],"4200e":3,"4201e":3,"4203e":3,"42441033":3,"426":5,"42847770e":15,"42937310e":15,"43054282":3,"4332e":3,"435163":1,"435490":[],"43549028":[],"43766686":9,"438136":1,"44089210e":3,"442600":[1,9],"443217":1,"44395541":[],"44625466e":15,"44970586e":2,"45013332e":[],"45019484":[],"450257":[1,9],"4557763":9,"455947":1,"458027":1,"458078":[1,9],"45937170e":[],"45960079":3,"46016532e":[],"461":15,"461838":3,"46183815":3,"462":5,"46323168e":15,"4632e":3,"46415888336127775":3,"46423858e":[],"465564":[],"46556436":[],"466":15,"46675058":[],"46914544e":[],"46929603e":[],"469730":3,"46973015":3,"47042744":3,"470714":1,"47079457e":15,"47125748":3,"47132891":3,"472652":3,"47265243":3,"47441766":[],"47566390e":15,"47654764e":[],"47815203":9,"47862383":3,"478624":3,"48019541":3,"48154187202453613":0,"48257387":16,"48471852e":[],"48476997":9,"48608063e":[],"48994188":3,"491837":3,"4918372":3,"4940954":1,"497221":[],"49722108":[],"49865673":[],"4990":15,"4992":15,"4997":15,"4c4c7f":[7,8],"4y_i":8,"500":[2,4,5,7,8,15],"500000":1,"50000455":3,"50000553":3,"50000718":3,"50000855":3,"50000969":3,"50001063":3,"50001142":3,"50001207":3,"50001261":3,"50001306":3,"50001343":3,"50001374":3,"500014":3,"50001414":3,"50001422":3,"50001439":3,"50001454":3,"50001466":3,"50001476":3,"50001485":3,"50001492":3,"50001498":3,"50001502":3,"50001506":3,"5000151":3,"50001512":3,"50001515":3,"50001517":3,"50001518":3,"50001519":3,"50001521":3,"50001522":3,"50001523":3,"50001524":3,"50001525":3,"5018":15,"50321091":3,"50394742":[],"506":1,"507d50":[7,8],"5098779":[],"509878":[],"50j":5,"50x10":2,"510":2,"5120":0,"512132":1,"51257863e":[],"51345668e":[],"51363731e":[],"514219":1,"51523276e":[],"515768":[],"51576807":[],"51707172":3,"517072":3,"51893804e":[],"51943726":9,"519842":1,"52158335":[],"5222222222222223":2,"52570079":3,"525701":3,"526744":[1,9],"52687171":[],"526872":[],"52722156":[],"52874252":3,"52d2c51caad1":[],"5303329":9,"5305555555555556":2,"531280":1,"535457":[],"53545715":[],"5378811":9,"539261":[1,9],"54039921":3,"54096582":[],"540966":[],"54121682":[],"541217":[],"54152940e":[],"541605":1,"54237024":15,"544439":1,"546972":3,"54697204":3,"54702088e":[],"55111512e":3,"55138385":9,"551384":9,"55280484":[],"5555555555555556":2,"55707065":[],"557071":[],"557795":[1,9],"55795935":3,"55854694":9,"56033697":3,"56198284":3,"56216797e":[],"564374":[1,9],"56536":1,"569":2,"56939714":3,"57051369":[],"571":3,"574465":[1,9],"5755035":[],"57781668":[],"57871326":[],"581766":9,"58176612":9,"582":[1,2],"58228342e":15,"58239999":[],"582400":[],"583595":1,"584804":1,"58521266":9,"585213":9,"58596975":[],"585970":[],"587401":1,"58836420e":15,"5888888888888889":2,"59007674e":15,"59304755e":[],"5944444444444444":2,"59480085":[],"5cm":15,"60122668e":15,"60293962":3,"60394236e":[],"60420593":3,"60673226":9,"60949193":[],"609492":[],"60999846":[],"61069091e":15,"6111111111111112":2,"61124978":[],"61234223":[],"612939":1,"613579":1,"614808":[1,9],"61505887e":[],"61745046e":[],"618":1,"6183694":[],"61869821":[],"618982":1,"622539":9,"62253933":9,"622625":3,"62262506":3,"6226921":[],"62316154e":15,"62359224e":15,"62373464":9,"625":5,"62783293":[],"62856593":[],"62894215":3,"629961":1,"6300745149331701":1,"63315151":[],"63339159":[],"63374631":[],"63437572":[],"63442451e":[],"63498144":3,"636323":9,"63632311":9,"63680118":[],"637129335071195":1,"63993205e":[],"64012627":3,"64166831e":[],"64447921":[],"64580686":[],"646283":[1,9],"64669382":[],"646694":[],"647473":[1,9],"64857826e":[],"649382":[1,9],"64x50":2,"650024":[],"65002433":[],"6510573774179256":15,"65105738":15,"65238878":13,"65245958":[],"653095390463358":[],"653702":13,"6544e":3,"65482578":[],"65572035":[],"65599927":[],"65766387":[],"65885453":3,"65933852":[],"65939208e":[],"66020213e":[],"6614":[],"66183486":9,"661835":9,"6628996975186952":1,"66302359":[],"66383151":[],"66677842":[],"66800261":[],"668172":1,"66880047":[],"67006792":[],"67060602":[],"671089":1,"6714":[],"67171347e":15,"672721":1,"67279536":[],"67303655":9,"673037":9,"67407338e":15,"67450955":[],"67708423":[],"677149":[],"67714918":[],"6796265324852733":[],"68002363":[],"680024":[],"68034946e":[],"6813":4,"6814":4,"6815":4,"6816":4,"6817":4,"6818813252071303":15,"68188133":15,"68192193":3,"68316185":[],"68342382":[],"68542204":3,"68616263":13,"68729414":[],"68937695":[],"689519":[1,9],"690617":1,"69069n_":15,"693361":1,"69347005":[],"693850":9,"69385025":9,"69519693":[],"695197":[],"69573183":[],"69695259":3,"69843037":3,"6996584":3,"69981195e":15,"6999536":9,"6ea927cc6e88":2,"6n_":15,"7022283":[],"70234019":[],"70415861":[],"70523024e":[],"70589906":[],"7070e":3,"70710678":3,"70790937":[],"70832814":3,"70885528e":[],"70886748":[],"70946493e":[],"712018":[1,9],"712199063818309":1,"71281409":[],"71351486":[],"71442781":13,"71606852":[],"7162":[],"71669651":[],"7172":[],"72108703":[],"72174172":9,"72218808":[],"72312577":[],"7236674":3,"7240496":[],"725394195434945":[],"72780613e":15,"72859758":3,"72879865e":[],"72981762":6,"73091052e":[],"731000":1,"733096":1,"73453972":[],"7371165871823337":[],"737349":3,"73734906":3,"73921714":[],"74081822":6,"74107697":9,"74143127e":[],"7432283":[],"74382593":[],"74495014":[],"745136489050356":1,"74840212":3,"74845978":[],"74921867":[],"749765":1,"750445":1,"751699":[1,9],"75170092":3,"75195757":[],"7522047280566193":[],"75382481":[],"75524378":[],"75629493":[],"756352":1,"75841112":[],"75932862":[],"76172241e":[],"762":[5,9],"7621419":[],"76290332":[],"763880":[],"76388013":[],"76497666":[],"765":5,"76504618":[],"76570177":[],"76648901e":[],"76936315":3,"7693978131030923":9,"77152076":3,"77156117e":3,"7718":7,"772b904ae9cb":4,"77317984":[],"7733":[],"77350269e":[],"774300":1,"776223":3,"77622336":3,"77661393e":15,"77714169":6,"78080633":[],"78082":4,"78195":[],"78857629":[],"78941903":3,"78944806":[],"7899453":[],"79111643":3,"79295029e":15,"793167":1,"79326583e":15,"79328828":[],"793701":1,"794282":[1,9],"79459035":[],"79648291":[],"79902342":1,"7c394b1e8b71":7,"7d7d58":[7,8],"800":5,"80004454e":[],"800266":[],"80026635":[],"80121":[],"80469739":3,"8055555555555556":2,"81160425":3,"81620806":[],"81633628":9,"816454":1,"816847":1,"81781888":9,"819202":3,"81920231":3,"82198978":3,"8265786":3,"827265":1,"82889306e":15,"8305555555555556":2,"83425361":9,"834254":9,"83512277":3,"83614019":[],"8388888888888889":2,"83935285":[],"84087101":[],"842":1,"842436":1,"84251485":[],"842515":[],"8429949116841184":3,"84355903e":2,"84372853":[],"84443254e":2,"84658093e":15,"8479552268981934":0,"84860939":1,"849766":3,"84976606":3,"84994524":3,"85058354":3,"850584":3,"8520127":[],"85278450e":[],"85396354":[],"85450859":9,"85463934e":[],"85497163e":[],"85546305e":[],"85601992":3,"861":1,"86117291":3,"86134827":3,"86145244":9,"8638888888888889":2,"86436607":[],"86574276":[],"8666666666666667":2,"86692943":[],"87030083":3,"870301":3,"8718475896381779":15,"87184759":15,"8722222222222222":2,"87381451":3,"875":2,"87761937":[],"8777777777777778":2,"87972591":[],"8802":[],"88046261":3,"88046462":[],"8805555555555555":2,"881323":3,"88297395":[],"88336879":3,"88559559":[],"8867467323038865":[],"88693966e":15,"88712946":[],"8888888888888888":2,"8901":[],"8914984":[],"8921171964770647":9,"8923":[],"89288636":9,"89383322":[],"89410423":3,"8944444444444445":2,"89704131":3,"89742056":[],"897421":[],"898500":9,"89850037":9,"89928965":3,"89942598":[],"89975818":[],"8x8":2,"90075537":3,"90220243":3,"90233874":[],"90266948":3,"90316476":[],"9040":7,"9055555555555556":2,"90618734e":15,"907307":9,"90730735":9,"90825063":[],"9096":4,"91012519e":[],"9111111111111111":2,"91128596":3,"91145266e":15,"91417278":[],"91549644":1,"9166666666666666":2,"916978":[],"91697817":[],"9171356":[],"91760278":3,"91812702":3,"918992":1,"9196":4,"9222222222222223":2,"922312":[],"92231203":[],"92280994":3,"924018":1,"925":2,"92507116e":2,"92578916":3,"92605247":[],"92645039":9,"92646965":9,"926470":9,"92732185e":[],"9277777777777778":2,"9279671770201344":15,"92857143":5,"92919670e":15,"9305555555555556":2,"931":1,"93100040e":15,"93155188":3,"93158979":3,"932734":[],"93273404":[],"933":3,"9354":[],"9361111111111111":2,"937":15,"937082":1,"93799826":3,"938":15,"9388888888888889":2,"939":[1,15],"94019247e":[],"94034531":[],"94284104":3,"94320205":3,"9444444444444444":2,"94536341":15,"94591015":13,"94639099":9,"946893955211749":1,"947543":9,"9482527":3,"948729":[],"95014575":[],"95231424":3,"9527777777777777":2,"95284275":3,"95351665":3,"954":15,"95443703e":[],"9547578478889096":1,"95517094":[],"955171":[],"9555555555555556":2,"95628168":[],"956282":[],"956563":[1,9],"95684892":3,"958228616652075":3,"9583333333333334":2,"959247":[],"960":15,"96024953":3,"96084663":3,"961":15,"96104648":[],"9611111111111111":2,"962":15,"963499":9,"96349948":9,"9640435":3,"965548":1,"965885569080809":[],"96688672":3,"9674916":3,"967809":[1,9],"96793117e":[],"97005689":3,"97101567e":15,"9722222222222222":2,"97243128":3,"97300836":3,"975":2,"97507735":3,"9756404":[],"976":6,"97622676":[],"976227":[],"9765":4,"9769":4,"97705827":3,"97723801":[],"97758848":3,"9777777777777777":2,"9780387310732":17,"9780387848570":17,"9781492032632":17,"97926491":3,"9804422":[],"98046438":[],"9805555555555555":2,"98091621":3,"981321":1,"98139097":3,"98275501":3,"98314755":[],"983148":[],"983310":1,"98413059":3,"98452685":[],"98454786":3,"985":15,"98526763":[],"98528992":[],"98531221":[],"98566191":3,"986":15,"9860803":[],"98609175":15,"986091753050161":15,"9861111111111112":2,"98661465":[],"98680716":3,"98716878":3,"9877742":[],"987902":[],"98808176":3,"988663":[],"9888888888888889":2,"98892195e":15,"98893512":[],"989":15,"9890348":3,"98927731":[],"9893447":3,"9898ff":[7,8],"99009525":3,"99009739":[],"9901168":[],"99013921":[],"99016161":[],"99018401":[],"99043999":[],"99088801":3,"9909252":[],"991":15,"991072":9,"99107239":9,"99115119":3,"99126104":[],"99133007":[],"99160404":[],"99176998":3,"99190487":[],"99194716":[],"992":15,"99218987":[],"99219378":[],"99242605":[],"99242921":3,"99248001":[],"99265097":3,"99273355":[],"99276945":[],"993":15,"99305549":[],"99305802":[],"99311297":[],"99316252":3,"99346398":[],"99363129":[],"99363383":[],"99371056":3,"99389612":3,"99393624":[],"994":4,"99400444":[],"99418903":9,"99420743":[],"99420997":[],"99428016":[],"9943201":3,"9945452":[],"99462421":[],"99473581":[],"9947756":3,"99478645":[],"99478898":[],"99492986":3,"99498985":[],"995":4,"99501236":[],"99503487":[],"99505739":[],"9950799":[],"99527696":[],"99528218":3,"9953048353087299":[],"99536326":[],"9953658":[],"99539415":3,"99544872":[],"9954538761021741":[],"99566069":3,"99578809":3,"99579317":[],"99581841":[],"99594294":[],"996":[3,4],"99600927":[],"99608161":3,"99636015":[],"9963961":3,"99650061":3,"99652042":[],"99652296":[],"99655111":[],"9966744029509663":1,"9967292151090247":1,"9967458":3,"9969":4,"99696351":[],"997":4,"99700706":3,"99709215":3,"99709325":[],"99710078":[],"99723611":[],"99728435":[],"99729756":3,"99730848":[],"99751458":3,"99758326":3,"99763569":[],"99767893":[],"99775587":3,"99782689":[],"99793613":3,"99799099":3,"998":4,"99813653":3,"99817842":[],"99825997":[],"99828624":3,"9983295":3,"99836973":[],"99845267":3,"9984806":[],"998577":3,"99861053":3,"99871521":3,"99881845":3,"99883879":[],"99884384":3,"99891285":[],"99893323":3,"999":[7,15],"99901896":3,"99903755":3,"99911427":3,"99918546":3,"99919837":3,"99926459":3,"9993237":3,"99933188":3,"99938942":3,"9994385":3,"99944272":3,"99945628":[],"99949306":3,"99953381":3,"99953475":3,"99957911":3,"99961294":3,"99965056":3,"99967865":3,"99970988":3,"9997332":3,"99975913":3,"99977849":3,"99980002":3,"9998161":3,"99984732":3,"99987324":3,"99988687":[],"99989476":3,"9999095":[],"99991263":3,"99992746":3,"99993212":[],"99993978":3,"99995":3,"99995475":[],"99995818594196":[],"999974281880048":[],"99997737":[],"9999787681537219":1,"9999794306626945":[],"9999822527140678":[],"9999864543345858":[],"9999868619217517":[],"9999869956119286":1,"9999878260589065":[],"9999910208315801":[],"9b9cf4fa1a95":[],"\u00f8yvind":16,"abstract":2,"break":[0,1,9],"byte":13,"case":[1,2,3,4,9,10,12,13,14],"catch":1,"char":15,"class":[1,2,4,5,6,7,9,10,15],"const":15,"default":[1,2,5,13],"ekstr\u00f8m":16,"export":7,"f\u00f8470":16,"final":[0,1,2,3,4,6,7,8,9,14,15,16],"float":[0,1,3,7,9,13,15],"function":[0,3,4,7,12,13],"import":[0,1,2,4,5,6,7,8,9,10],"int":[0,1,2,3,5,9,13,15],"long":[1,2,5,10,15],"new":[0,1,2,3,4,5,6,7,8,9,13,15],"null":15,"public":[1,12],"return":[0,1,2,3,4,5,6,7,9,13,15],"s\u00f8rli":16,"sch\u00f8yen":16,"short":[0,3,11],"steinsv\u00e5g":16,"super":3,"switch":[0,1],"throw":15,"true":[0,1,2,3,4,5,6,7,8,10,15],"try":[0,1,2,3,5,6,7,8,9,12,13,15],"var":[3,4,8,9,15],"while":[0,1,2,3,4,5,6,7,9,10,15],AGE:1,Adding:2,Age:5,And:[0,1,4,7,12,15],Are:9,Being:5,But:[0,1,2,3,4,7,8,15],CAS:[],DIS:1,Doing:[3,5,8],EoS:[1,4],FYS:14,For:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],Going:2,Ising:[3,10],Its:[2,9],MDS:9,NNs:10,N_s:6,Not:[1,2,3,4],OLS:[1,4],One:[1,2,3,4,5,6,9,10,15],PCs:[9,12],RMS:15,Such:[4,10,15],That:[0,1,3,5,8,9,10,15],The:[8,13,14,16,17],Then:[0,1,2,3,4,5,6,7,8,9,10,13,15],There:[0,1,3,6,7,9,10,11,13,14,15,16],These:[0,1,3,6,7,8,9,10,13,15],Use:[1,7],Useful:[3,4,13],Using:[1,3,4,6,8,10,13],With:[1,3,4,6,7,8,9,10,13,15],__class__:8,__doc__:4,__future__:[6,7],__getattr__:[],__init__:[2,4],__mosek:3,__name__:[4,8],_auto10:10,_auto1:[3,5,10,13,15],_auto2:[10,13,15],_auto3:[10,13],_auto4:[10,13],_auto5:[10,13],_auto6:[10,13],_auto7:[10,13],_auto8:10,_auto9:10,_ax:4,_base:6,_build:[12,17],_check_optimize_result:[5,9],_compon:9,_datafram:[],_depth:7,_fraction:7,_lambda:1,_leaf:7,_logist:[5,9],_make_index:[],_multilayer_perceptron:[1,2],_node:7,_num_sampl:[],_ratio:9,_sampl:7,_split:7,_varianc:9,_weight:7,a0faa0:[7,8],a77d5ac269b2:[],a_0:1,a_1a:1,a_2a:1,a_3:1,a_3a:1,a_4:1,a_4a:1,a_h:2,a_i:[1,2,10],a_j:[2,10],a_k:[2,10],aaron:17,ab_channel:12,abandon:2,abbrevi:14,abid:15,abil:[1,8],abl:[2,3,5,8,10,15],abort:15,about:[0,1,2,3,4,5,6,7,8,9,10,12,13,17],abov:[0,1,2,3,4,5,6,7,8,9,10,13,15],abovement:4,abs:[0,1,3],abscissa:5,absolut:[1,3,4],acccess:[],accept:[1,7],access:[1,9,15],accid:4,accmod:3,accompani:1,accomplish:[6,7],accord:[0,1,2,3,4,5,7,10,15],accordingli:9,account:[1,3,15],accumul:[10,15],accur:[4,8,15],accuraci:[1,2,3,5,7,8,9,10],accuracy_scor:[1,2,8],accuracy_score_numpi:2,achiev:[1,2,3,4,6,10,13],aco:15,acquaint:12,acquir:[2,12],acr:1,across:[2,4,7,12],act:[2,13],action:15,activ:[0,1,7,14],actual:[1,2,3,4,6,9,13],ada_clf:8,adaboostclassifi:8,adam:2,adapt:[1,4,5,17],add:[1,2,3,4,6,8,9,10,15],add_subplot:[0,2,5,10],added:[1,2,3,5,6,13],addendum:3,adding:[0,2,4,13],addit:[0,1,3,4,5,6,7,8,10,12,13,15,16,17],addition:[5,10],address:[2,5,7,9,17],adjac:10,adjoint:3,adjust:[5,10],admir:1,advanc:[4,10,17],advantag:[2,3,4,5,8,13],afecionado:[],affect:[],affin:[1,6,9],aficionado:[],aforement:0,african:1,after:[0,1,2,3,4,5,7,9,10,12,13,15],afterward:1,again:[0,1,2,3,4,5,6,8,9,10,15],against:[2,5,8],age:[1,5],agegroup:5,agegroupmean:5,aggreg:[7,8],agorithm:8,agre:[3,15],ahead:7,aid:9,aim:[0,1,2,4,5,9,12,13],ainv:3,aka:3,alarm:3,albeit:0,algebra:[1,3,5,12,14],algo:15,algorithm:[1,2,3,4,5,6,12,13,14,15,17],align:[1,3,4,5,6,15],all:[0,1,2,3,4,5,7,8,9,10,12,13,14,15,16,17],allevi:[2,5],alloc:13,allow:[1,2,3,4,5,6,8,12,13],almost:[1,2,4,5,6,9,15],alon:7,along:[0,3,4,7,8,9,12,13],alpha:[0,1,2,3,4,5,6,7,8,15],alpha_:8,alpha_i:5,alpha_k:5,alpha_m:8,alpha_opt:5,alreadi:[3,8,10,12,13,15],also:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],alter:[2,15],altern:[1,2,3,5,6,7,9,13],although:[2,3,4,6,8],alwai:[0,1,3,4,5,10,15],ame2016:1,american:1,among:[1,3,7,8,10],amongst:3,amount:[0,1,2,4,6,8,12],an_:15,anaconda3:[1,2,4,5,6,9],anaconda:[1,2,12],analys:[4,15],analysi:[2,4,5,13,14,17],analyt:[1,3,4,5,10,12],analyz:[1,2,3,15],andrew:2,angl:[1,7,15],ani:[0,1,2,3,4,6,7,8,10,15],anim:10,ann:10,annot:[1,2,5,6],anoth:[1,2,3,4,5,6,8,9,10,13,15],ans:15,ansatz:1,answer:[1,2,3,4,13],anymor:[2,6],anyon:6,anyth:[2,15],anytim:16,apach:2,apart:[5,9],api:[2,12],appear:[1,2,13,15],append:[2,4,5,6,7,15],appendcon:3,appendvar:3,appli:[1,2,4,5,6,7,8,9,10,15,17],applic:[1,2,3,4,5,7,10,15,17],approach:[2,3,4,7,8,9,10,12,15,17],appropri:[4,7,10,12,15],approx:[1,5,8,9,15],approxim:[1,2,3,4,5,8,9,15],apt:[1,12],aptli:0,aragorn:[],arang:[2,4,5,7,8,10],arbitrari:[2,5,6,10,15],arbitrarili:[1,2,9],architectur:[10,17],area:[1,17],arg:4,argc:15,argmax:[2,9],argmin:[0,8],argsort:9,argu:2,arguabl:0,argument:[1,3,9,10,15],argv:15,aris:[1,4,5,10,15],arithmet:[1,13],arma:15,armadillo:[13,15],around:[1,2,3,4,9,15],arrai:[0,1,2,3,4,5,6,7,9,10,12,15],arriv:[1,7,9,13,15],arrow:10,arrowprop:6,art:[1,2,12],articl:[0,1,4,8,15],artifici:[1,5,10,17],artificialneuron:10,artist:4,asarrai:[1,7],asc:3,ascii:15,ask:[3,4,9,10],aspect:[1,12],assembl:1,assess:[1,4],assign:[0,1,5,6,7,10,14,17],assign_points_to_clust:0,associ:[0,1,4,7,10,15],assum:[0,1,2,3,4,5,6,7,8,9,10,13,15],assumpt:[1,3,4,7,9,15],ast:[1,3,4],astyp:[7,8],asymmetri:1,asymptot:4,atoi:15,atom:1,attempt:[1,5,6,8],attend:14,attent:[1,13],attr:[],attract:[1,8],attribut:[1,7],attributeerror:4,audi:1,aurelien:[1,14,17],author:[1,2,8,15],authour:1,auto:[7,8,15],autocor:15,autocorrelation_tim:15,autocorrelform:15,autocovari:15,autoencod:12,autoencond:12,autograd:12,autom:[1,12],automac:13,automag:[],automat:[1,2,9,12,13],autonom:17,avail:[1,2,4,8,9,12,13,14,17],averag:[0,1,2,4,7,8,15,16],avg:15,avoid:[0,1,3,4,7,9,13],awai:15,awar:8,award:16,axes3d:5,axes:[1,4,5,6,7,8,9],axessubplot:1,axhlin:6,axi:[0,1,2,4,5,6,7,8,9,10,15],axiom:3,axlabel:1,axvlin:6,b_1:[5,10],b_5:5,b_group:7,b_i:[1,2,10],b_ia_:1,b_index:7,b_j:[2,10],b_k:[2,5,10],b_m:10,b_score:7,b_valu:7,bachelor:14,back:[1,3,6,7,8,13,15],backbon:13,backend:2,background:17,backpropag:2,backtrack:7,backup:13,backward:[2,10,13],bad:15,badli:15,bag:[7,12,14],bag_clf:8,baggin:[],baggingboot:8,baggingclassifi:8,baggingtre:8,balanc:4,band:13,bandwidth:13,bar:[1,9],barber:17,bare:8,barebon:0,base:[0,1,2,3,5,6,7,8,12,15,16,17],basi:[3,5,6,8,9,10,13],basic:[6,10,12,14,15],batch:[5,9,10],batch_siz:2,bay:5,bayesian:[3,12,17],becaus:[0,1,2,3,4,5,6,7,10],becom:[0,1,2,3,4,5,7,10,15],been:[1,2,4,9,10,12,13],befor:[0,1,2,3,4,5,6,10,13,15],beforehand:[0,1,15],begin:[0,1,2,3,4,5,6,7,9,10,13,15],behav:[2,4,5],behavior:[1,2,5],behaviour:10,behind:[1,2,5,6],being:[0,1,2,3,5,6,8,9,10,15],believ:[7,13],belong:[0,3,5,6,7],below:[1,2,3,4,5,6,7,8,9,10,13,15],benchmark:8,bendik:16,benefici:2,benefit:[1,2,5,9,12],bengio:[2,14,17],benign:[2,5],besid:3,bessel:3,best:[0,1,2,3,4,5,6,7,8,10,15,16],beta:[1,2,3,4,5,8,9],beta_0:[1,2,3,5],beta_0x_:1,beta_1:[1,2,3,5,8],beta_1x_0:1,beta_1x_1:[1,5],beta_1x_2:1,beta_1x_:1,beta_1x_i:5,beta_2:1,beta_2x_0:1,beta_2x_1:1,beta_2x_2:[1,5],beta_2x_:1,beta_:[1,5],beta_i:[1,3],beta_j:[1,3,5],beta_k:5,beta_linreg:5,beta_m:8,beta_mg_m:8,beta_p:5,beta_px_p:5,better:[0,1,2,3,7,8,9,10],between:[0,1,2,3,4,5,6,7,9,10,15],beyond:[1,2,3,5,6],bia:[1,2,3,6,7,8,10,14,15],bias:[2,3,4,7,10],big:[0,1,2,3,4],bigger:2,bigr:10,bike:7,bilbo:[],bilek:16,billion:[10,12],bin:[1,4,5,15],binari:[1,3,5,7,8,10,14,15],bind:[1,4],binomi:12,binsboot:4,bioinformat:1,biolog:[2,10,17],bios1100:12,bird:1,birth:[],bishop:[14,17],bit:[0,2,13,15],bitwis:15,bla:[3,13],black:[0,6,7],block:[0,4,8,12,13],blockingavg:15,blockingstd:15,blockingvar:15,blocksiz:15,blocksizemax:15,blocksizemin:15,blue:1,bmatrix:[1,2,3,5,6,9,13],bmi:2,bodi:[1,2,10],bold:2,boldfac:[1,3],boldsymbol:[0,1,2,3,4,5,6,8,9],boltzmann:[10,12],book:17,bool:0,boost:[2,7,12,14],boostrap:8,bootavg:15,bootstd:15,bootstrap:[2,12,14],bootvar:15,bootvec:15,borrow:[],boston_dataset:1,bot:6,both:[0,1,2,3,4,5,6,7,8,12,13,15,16],bottl:5,bottom:4,bound:[1,3,4,6,10],boundari:[6,9,10],boundkei:3,box:7,boyd:[5,6],bracket:15,brain:[2,5,10],branch:7,breast:[3,5,9],brew:[1,12],brg:6,briefli:1,bring:[1,3,8],broad:1,broadcast:0,browser:[],brute:[3,9],bsol:[],bsubex:[],build:[1,3,4,8,13,15],built:[1,2,4],bunch:9,busi:1,bzl:3,c46dd114b2af:6,c_0:15,c_1:10,c_2:10,c_3:10,c_4:10,c_i:[5,10],c_k:15,cach:8,cal:[1,5,6,8,10],calcul:[0,1,2,3,4,5,6,7,8,9,10,13],call:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],callabl:4,calor:1,cambridg:[5,17],came:0,can:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],cancel:1,cancer:[3,8],cancerpd:5,candid:[6,7,8],cannot:[1,2,3,5,6,7,14,15],canopi:[1,12],cap:3,capabl:[1,2,6,12],capita:1,captur:[9,10],card:[1,5],cardin:2,care:[0,9],carefulli:5,carlo:[1,4,12,15,17],carri:[4,5],cart:8,casella:17,cast:2,categor:[1,2,7,9],categori:[0,1,2,5,8,10,14],categorical_crossentropi:2,caus:[1,3,4,15],causal:1,causat:1,cax:2,cbar:2,ccc:[3,10],cdf:15,cdot:[0,1,4,5,10,13,15],celebr:5,center:[0,1,2,4,5,6,7,9,15],centr:17,central:[1,3,4,6,13],centroid:[0,15],centroid_differ:0,centroid_list:0,certain:[0,1,4,5,7,15],cha:1,chain:[2,12,15],challeng:[0,5],chanc:[2,3,5,15],chang:[0,1,2,3,4,5,6,7,9,10,13,15],chapter:[0,4,8,9,13,14,17],charact:[1,3,6,15],character:[6,7,8,10,15],characterist:[1,2,8],charg:1,charl:1,chd:5,chddata:5,cheap:3,cheaper:[2,5],chebychev:0,check:[1,2,3,5,9,13],check_consistent_length:[],chemic:15,chen:8,choic:[0,1,2,4,5,7,10,13,15],choleski:[3,13],choos:[0,4,5,7,8,9,15],chosen:[1,2,4,5,6,7,8,15],chosen_datapoint:2,christian:17,christoph:[14,17],cin:15,circ:[2,10],circl:[1,6,10],circumfer:7,circumv:[2,3],clariti:[0,15],class_nam:7,class_val:7,class_valu:7,classic:[5,7],classif:[1,3,4,5,6,9,10,12,14,17],classifi:[1,2,5,7,8,9],classificaton:2,classifii:8,clean:2,clear:[2,3,8,10],clearer:0,clearli:[1,3,4,5,6,15],clever:[0,2,8],clf3:1,clf:[1,6,7,8],clf_ridg:1,clip:15,close:[0,1,2,4,5,6,7,9,10,15,17],closer:3,closest:[0,5,6,9],closur:12,cloud:12,clust:0,cluster:[1,2,4,9,12,14],cluster_label:0,cmap:[1,2,6,7,8],cmath:15,cmb:14,cmd:7,cn_:15,cnn:10,cntk:12,code:[4,6,12,13,14,17],coef0:6,coef:1,coef_:[1,3,5,6,7],coeffici:[1,3,4,5,6,7,13],coerc:[1,4],coin:[8,15],coin_toss:8,col:[1,9],colab:12,cold:7,colinear:1,collaps:6,collect:[1,4,8,9,12,15,17],collinear:3,color:[1,4,6,7,8,15],colorbar:2,colsample_bytre:8,colsaobject:8,column:[1,2,3,4,5,6,7,9,10,13],columntransform:7,com:[12,16,17],combin:[2,3,4,5,8,15],come:[0,1,2,3,5,10],comma:[],command:[1,2,15],comment:1,commerci:[1,12],commod:1,common:[0,1,2,3,4,5,7,9,15],commonli:[0,1,2,4,5,7],commun:[1,10],compact:[0,1,2,3,4,5,7,9,10],compar:[0,1,3,4,5,9,13],compat:5,compet:1,competit:8,compil:[1,2,12,13,15],complet:[1,7,10],completenn:10,complex:[2,3,4,6,7,9,10],complic:[0,1,2,4,5,7],compon:[0,1,2,3,4,5,7,12,14],components_:9,compos:[0,7,10],compphys:[12,14,17],compress:1,compris:4,compromis:3,compulsori:12,comput:[0,1,2,3,4,6,8,9,10,12,13,14,17],computation:[1,4,5,7,15],con:3,concaten:0,concav:[2,5],concentr:[1,8],concept:[0,1,12],conceptu:[5,10],concern:[0,1,2,5],concic:[],conclud:[1,3],conclus:2,conda:[1,2,12],condit:[1,3,4,6,7,9,15],conduct:12,coneqp:3,confid:[1,3,4,5,6],confirm:[3,10],confus:[3,4,8,13],confusion_matrix:7,congruenti:15,conjug:6,conjugaci:5,connect:[1,2,5,7,9,10,13],consequ:[3,4,5,6,8,10],conserv:[0,3],consid:[0,1,2,3,4,5,6,7,8,10,13,15],consider:[1,2,3,5],consist:[1,2,4,5,10,15],constant:[1,3,5,6,10,15],constitu:1,constitut:4,constrain:[2,3,5,9],constraint:[3,5,6],construct:[1,2,3,4,5,6,7,8,9,13,15,17],contact:1,contain:[0,1,3,4,5,6,7,9,10,13,15,17],contemporari:17,content:[2,12,13],context:[4,5,8],continu:[1,2,3,4,5,6,7,8,10,13],contour:[5,7,8],contourf:[6,7,8],contrast:[2,7,8,10],contribut:[1,3,15],contributor:1,control:[1,2,7,12],conveni:[1,3,4,5,10,13],convent:10,converg:[0,1,2,3,5,6,9],convergencewarn:[1,2,5,6,9],convert:[1,2,3,7,9,13],convex:3,convinc:5,convolut:[2,12,14],cool:7,coordin:[0,3,10],coorel:1,copi:[0,1,2],core:8,corel:1,coronari:5,corr:[1,3,5,9],correalt:[9,12],correct:[0,1,2,3,13,15],correctli:[2,4,8],correl:[1,2,3,5,8,10,12],correlation_matrix:[1,3,5,9],correspond:[0,1,3,4,6,7,9,10,12,13,15],cortex:10,cos:[1,4,7],cosin:[0,4],cost:[1,3,4,6,7,10,15],could:[1,2,3,4,5,6,7,8,9,10,13,15],coulomb:1,count:[1,7,14,15,16],countor:5,coupl:3,cours:[1,2,3,9,14,15],courvil:[14,17],cout:15,cov:[3,4,9,13,15],cov_xi:[3,9],cov_xx:[3,9],cov_yi:[3,9],covari:[5,12,13],covariance_matrix:[0,3,9],cover:[1,3,11,12,17],covert:1,covxi:15,covxx:15,covxz:15,covyi:15,covyz:15,covzz:15,cpu:2,creat:[2,3,7,8,9,10,12],create_biases_and_weight:2,create_neural_network_kera:2,create_x:[3,9],credit:[1,5],crim:1,crime:1,criteria:[0,1,7,8,15],criterion:[5,7,8],cross:[1,2,5,7,8,12,14,15],cross_val_scor:4,cross_valid:[5,8],crossvalid:4,crucial:[2,15],csr_matrix:13,cstdlib:15,csv:[1,4,5,7],ctnk:2,cubic:1,cumbersom:3,cumsum:[8,9],cumul:[4,8],cumulative_heads_ratio:8,cup:3,current:[0,2,5],curs:1,curv:[5,8,10],curvatur:5,custom:[0,3],custom_cmap2:[7,8],custom_cmap:[7,8],cutpoint:7,cvxbook:5,cvxopt:[3,6],cyber:17,cycl:[0,2,10,15],d_f:5,dagger:[3,13],dai:[2,7,12],dalen:16,darget:7,darkr:15,dat:1,dat_id:[1,4,5,7],data1:0,data2:0,data3:0,data4:0,data:[0,3,4,6,8,10,13,17],data_id:[1,4,5,7],data_indic:2,data_panda:[],data_path:[1,4,5,7],databas:2,datafil:[1,4,5,7],datafram:[1,3,5,7,9],datapoint:[2,3,4,5,9],dataset:[0,1,4,5,6,7,8,9],date:1,daughter:8,david:17,dbh:2,dbo:2,dcomposit:13,dcost:3,dead:2,deal:[0,1,2,3,5,6,9,13,15],debt:5,debug:[3,4],decad:1,decai:[1,5,15],decent:8,decid:[3,4,7],decim:1,decis:[1,2,6,9,12,14,17],decision_funct:6,decision_tre:7,decisiontreeclassifi:[7,8],decisiontreeregressor:[1,7,8],declar:[1,13],decompos:[3,13],decomposit:[1,10],decompost:3,decorrel:8,decreas:[2,3,4,5,8,9],deduc:1,deep:[5,10,12,14,17],deep_tree_clf1:7,deep_tree_clf2:7,deep_tree_clf:[7,8],deepen:12,deeper:0,deeplearningbook:17,def:[0,1,2,3,4,5,6,7,8,9,15],def_covari:15,defect:3,defici:3,defin:[0,1,3,4,5,6,7,8,9,10,13,15],definit:[0,2,3,4,5,6,8,9,10,13],defint:15,degre:[3,4,6,7,8,9,15],del:2,delet:[4,15],deliv:14,delta:[0,1,6,10,15],delta_:[2,13],delta_h:[1,2],delta_j:10,delta_k:10,delta_l:2,delta_n:1,delug:12,delv:1,demand:5,demonstr:[1,3,4,5,9,10,12],denomin:[2,3],denot:[2,4,5,15],dens:[0,2],densiti:[0,1,4,15],depart:16,depend:[0,1,2,3,4,5,6,9,10,12,15],depict:15,deploy:[1,12],deprec:1,depth:[7,8,13],deriv:[1,2,4,6,8,9,12],descend:[3,7,9],descent:[1,2,6,10],descr:1,describ:[0,1,3,4,6,8,9,10,13],descript:[0,1,6,7],design:[1,2,3,4,5,8,9,10],designmatrix:1,desir:[0,1,3,5],despit:[2,10],destroi:13,det:[3,13],detail:[0,1,5,9,13],detect:[6,10],determin:[1,3,5,6,7,8,9,10,13,15],determinist:[5,15],dev:[2,15],develop:[0,1,3,6,8,9,10,12,13],deviat:[1,2,3,4],devis:10,df1:[],diag:[3,6],diagnost:[2,8],diagon:[1,3,5,13,15],diagonaliz:3,diagram:8,dict:6,dict_kei:1,dictionari:1,did:[0,1,2,3,5,8,9],die:2,diffeent:6,differ:[0,1,2,3,4,7,8,9,10,12,13,15,17],differenti:[5,12,13],difficult:[0,1,2,4,8,15],difficulti:[1,2,5],digit:[1,2,14,16],dilut:2,dim:[0,9],dimens:[0,1,2,3,6,9,13],dimension:[0,1,3,4,5,7,9,12,13],dimensionless:1,diment:13,dimes:0,direct:[0,1,2,5,9,10],directli:[0,2,3,15],directori:[],disadvantag:1,discard:[4,9],disciplin:[1,10],disclaim:15,discourag:5,discov:1,discover:3,discret:[2,3,5],discrimin:[5,8,9],discuss:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],diseas:5,disk:[],disord:[2,5],displai:[0,1,2,4,5,6,7,8,9,10,15],displaystyl:[1,3],displot:1,disregard:1,dissimilar:[0,9],dist:0,distanc:[0,1,6,7,9,15],distance_list:7,distinct:[0,5,6,7,8],distinctli:6,distinguish:[1,5,6,15],distplot:1,distribut:[0,1,2,4,5,8,9,12,13],distrubut:[1,12],div:3,dive:[1,6],diverg:[2,5],divid:[1,2,3,4,6,7,9,10,15],divis:[4,6,7,13,15],dna:5,dnn:[1,2,10],dnn_kera:2,dnn_model:2,dnn_numpi:2,dnn_scikit:[1,2],doamin:3,doc:[12,14,17],doconc:[],document:[5,9],doe:[0,1,2,3,4,5,6,8,9,10,13,15],doesn:[7,10],dog:2,doi:0,doing:[1,3,4,9],domain:[3,5,6],domin:1,don:[0,1,2,6,9,12,15],done:[0,1,3,4,7,8,9,13,15],dot:[1,3,5,6,7,8,9,10,13,15],doubl:[13,15],doubli:2,down:[1,5,7,9,10],download:[1,2,3,13,17],dozen:2,dramat:9,draw:[4,5,8],drawback:[1,2,5],drawn:[2,4,5,9,15],dre:3,drop:[1,2,3,4,9,15],dropna:[1,4],dtype:[0,1,2,13],dub:1,due:[0,2,3,4,5,6,8,10],dummi:1,dure:[1,2,6,7,9,12],dwell:1,dwh:2,dwo:2,dx_1:15,dx_n:15,dying:2,each:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,16],eapprox:1,earli:2,earlier:[1,3,5,6,7,9,10],eas:[0,7],easi:[1,3,4,5,6,7,8,9,10,12,13],easier:[3,4,6,7,15],easiest:5,easili:[1,2,3,5,6,7,8,9,10,13],eastern:16,ebind:1,eblock:7,econometr:[],economi:3,ecosystem:12,ect:14,edgecolor:4,edit:[],edu:5,educ:1,eface79dac2c:8,eff:15,effect:[2,8,15],effic:2,effici:[1,5,8,12,13,15],efron:[4,15],eig:[3,5,9,13,15],eigen:15,eigenpair:[3,9],eigenvalu:[3,5,6,9,13],eigenvector:[3,5,9],eight:13,eigval:[13,15],eigvalu:[5,9],eigvec:[13,15],eigvector:[5,9],eispack:13,either:[1,2,3,4,5,6,7,8,9,15],ekstrom:16,elabor:15,electr:[1,10],electur:17,eleg:9,element:[2,3,4,5,6,9,10,12,13,14,17],elementari:[8,13],elessar:[],elif:0,elim:13,elimin:[3,6],els:[2,5,7,10,15],elu:2,elus:1,email:[14,16],embed:[1,9],embodi:4,emit:15,emner:17,emphas:[1,8,12],emphasi:[1,12,17],empir:[2,9,15],emploi:[1,2,3,4,5,9,15],employ:1,empti:[4,8],emul:10,enabl:9,encapsul:0,encod:[0,1,3,7,9],encompass:[1,15],encount:[1,2,3,5,15],end:[0,1,2,3,4,5,6,7,8,9,10,13,15],endl:15,endpoint:15,energi:[1,4],enforc:10,eng:17,engin:[1,2,12,15],english:17,enough:[1,4,5],ensembl:[2,7,14,15],ensur:[1,2,3,4,5,9,15],enter:[3,15],enthought:[1,12],entir:[2,5,7,12,15],entiti:[7,10,13],entri:[1,3,6,9,10,13],entropi:[2,5,8],enumer:[1,2,6],env:[3,15],environ:[12,17],eol:1,eosfit:1,epoch:[1,2,5,10],epsilon:[1,3,4,5],epsilon_0:1,epsilon_1:1,epsilon_2:1,epsilon_:1,epsilon_i:1,eqnarrai:[3,4],equal:[0,1,2,3,4,5,6,7,9,10,13,15],equat:[0,2,3,4,6,7,8,9,13,15],equilibrium:10,equiv:[5,13,15],equival:[1,2,3,6,9,12,13],erf:15,eriador:[],eridg:1,err:[1,8],err_:4,errat:5,errno:[],error:[2,3,4,5,7,9,10,12,15],error_estimate_corr_tim:15,error_hidden:2,error_output:2,escap:5,esl:0,esol:[],especi:[2,7,10],essenti:[0,1,3,7,8,10,15],establish:[1,8,9],estim:[1,2,3,4,5,8,9,12,15],estimated_mse_fold:4,estimated_mse_kfold:4,estimated_mse_sklearn:4,esubex:[],eta0:[5,6],eta:[1,2,5,6,10],eta_v:[1,2],etc:[0,1,2,3,5,6,7,9,10,12,13,15],ethic:12,etsim:4,euclidean:[0,1],evalu:[1,3,4,5,7,15],even:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],event:[3,5,8],eventu:[3,4,5,9,10,16],everi:[0,1,2,3,4,5,7,8,9,10,12,15],everyth:10,everywher:5,evolv:1,exact:[1,3,5,9,10,13,15],exactli:[1,10,12],examin:4,exampl:[3,4,9,10,12,13,14,17],exce:[2,10],excel:[0,1,2,3,8,17],except:[3,4,6,7,15],excess:1,excit:1,exclud:[2,4,10],exclus:[1,2,4,15],execut:3,exemplifi:5,exercis:[3,12,14],exhaust:4,exhibit:[1,6],exist:[0,1,2,3,4,5,6,7,13,17],exit:[3,13,15],exp:[1,2,3,4,5,6,8,9,10,15],exp_term:2,expand:[3,5,9],expans:[1,3,5,6,8,10],expect:[1,2,3,4,5,9,10,12],expectation_value_of_h_wrt_p:15,expens:[4,5,8,15],experi:[1,2,4,5,6,12],experiment:[1,4,7,15],expert:[2,7],explain:[0,1,5,7,8,9,15],explained_variance_ratio_:9,explanatori:1,explicit:[1,13],explicitli:[0,1],explod:2,exploit:[1,10],explor:[2,5,6,12],expon:2,exponenti:[1,2,3,5,8],export_graphviz:7,export_text:7,exporttext:7,expos:12,express:[1,3,4,8,10,13,15],exptmean:15,exptvari:15,extend:[3,9,12],extens:[1,10,12],extent:[1,2,4,17],extern:7,extra:[2,3],extract:[1,3,5,6,9,13],extrapol:1,extrem:[0,1,2,3,5,6,7,13],extremum:5,extrins:9,eye:[0,1,3,5,13],f11:1,f12:1,f13:1,f1d:5,f6d7a289d493:13,f_0:8,f_1:[5,8],f_2:[5,10],f_3:10,f_d:15,f_i:[4,10],f_m:8,face:5,facecolor:[4,6,15],facil:[1,12],facilit:10,fact:[1,2,3,5,7,9,10],factor:[1,2,3,7,8,9,13,15],fafab0:[7,8],fail:[4,5,6,9,16],failur:5,fairli:[0,2,15],fall:[6,7,14],fals:[0,1,2,3,4,5,7,8],famili:[1,5,6,15],familiar:[1,3,6,12,13,15],famou:[4,10,13],far:[0,1,3,5,6,9,10,15],fashion:[1,7,8],fast:[2,4,5,8,10,12,15],faster:[2,9],fastest:5,favor:5,favorit:15,featur:[1,2,3,4,5,6,8,9,10,12,15],feature_nam:[1,2,5,7],feautur:7,fed:2,feed:[1,9,12,14],feed_forward:2,feed_forward_out:2,feed_forward_train:2,feedforward:[2,10],feel:[0,1,3,9,12,16],feet:1,few:[0,2,7,11,15],fewer:[1,7,9],ffnn:[2,10],field:[1,10,12],fifth:1,fig:[0,1,2,5,10],fig_id:[1,4,5,7],figaxi:15,figsiz:[1,2,4,5,6,7,8],figur:[0,1,2,3,4,5,6,7,8,10,12],figure_id:[1,4,5,7],figurefil:[1,4,5,7],file:[1,2,3,4,5,6,7,13,15],filenam:[1,15],filenotfounderror:[],fileout:15,fill:[3,7],financ:1,find:[0,1,2,3,4,5,6,7,8,9,10,12,15],fine:[0,1],finit:[3,4,10,15],first:[0,1,2,3,4,6,7,8,9,13,15,17],firsteigvector:9,fit:[2,3,4,5,6,7,9,10,15],fit_intercept:[1,3,4],fit_mod:7,fit_transform:[1,4,6,7,9],fiti:1,five:[1,7],fix:[1,4,5,8,9,10],fkkt:3,flag:0,flat:[5,10],flatten:[2,3],flexibl:[1,2,4,6,8,10],float32:7,float64:[1,13],flop:[3,13],flow:[2,10],fluctuat:3,fly:9,flyvbjerg:15,fmesh:5,focu:[1,3,4,12,17],focus:[2,5,13],fold:[4,7],folder:1,follow:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,16,17],font:[1,5,15],fontdict:15,fontsiz:[2,6,7,8,15],fontweight:2,foral:6,forc:[1,3,8,9],forecast:10,forest:[1,2,7,12,14],forget:9,form:[1,3,4,5,6,7,9,10,12,13,15],formal:[0,15],format:[1,2,4,5,6,7,8,9,12,15,17],formatstrformatt:5,formul:[0,9],formula:[5,15],forth:10,fortran2003:12,fortran90:15,fortran:[1,12,13],fortun:[1,9],forward:[1,4,12,13,14],found:[2,3,4,5,10],foundat:12,four:[3,6,10,13,14],fourier:1,fourth:10,frac:[0,1,2,3,4,5,6,7,8,9,10,13,15],fraction:7,frame:5,framework:[2,6,8,15],frank:[3,9],frankefunct:[3,9],free:[1,9,12,13,15,16,17],freecodecamp:12,freedom:3,freeli:1,frequenc:[4,5,15],frequent:[1,5,6,7],frequentist:12,fresh:8,fret:0,fridai:14,friedman:[14,17],frodo:[],from:[0,1,2,4,5,6,7,9,12,13,14,15,16,17],from_cod:7,front:[1,3],fruit:0,fstream:15,fulfil:[3,10],full:[1,2,3,5,7,8,15],full_matric:3,fulli:[4,10,14,15],fun:12,func:4,functionali:9,fundament:[1,4,12],further:7,furthermor:[1,3,4,5,9,10,12],futur:[1,6,7],futurewarn:1,fys:16,g_1:8,g_2:8,g_m:8,gain:[0,2,3,7,8],galleri:1,gamge:[],gamma1:6,gamma2:6,gamma:[1,5,6,7,8,9],gamma_0:8,gamma_1:8,gamma_1x:8,gamma_:1,gamma_i:[1,6,15],gamma_j:5,gamma_k:5,gamma_m:8,gamma_x:1,gap:[3,6],gate:10,gather:[1,2,10],gaug:10,gaussbacksub:13,gaussian:[0,3,4,6],gaussian_point:0,gaussian_rbf:6,gave:5,gbc:14,gca:[4,5,6],gd_clf:8,gdclassiffiercgain:8,gdclassiffierconfus:8,gdclassiffierroc:8,gdregress:8,gemv:3,gen:15,gender:1,gener:[0,1,2,3,4,5,6,8,9,10,13,17],generallay:10,generate_simple_clustering_dataset:0,genom:12,geodes:9,geometr:1,geometri:3,georg:17,geq:[3,5,6,7],geron:[1,14,17],get:[0,1,2,3,4,5,7,8,9,12,13,15],get_distances_to_clust:0,get_dummi:7,get_split:7,get_yaxi:6,getattr:4,getsolutionslic:3,gibb:12,gini:8,gini_index:7,git:[1,12],github:[1,12,14,17],gitlab:[1,12],give:[0,1,2,3,5,6,7,8,10,12,15,17],given:[0,1,2,3,4,5,6,7,8,9,10,13,15],glare:0,global:[5,15],glorot:2,gmail:16,goal:[1,5,7],goe:[0,1,2,3,4,5,13],going:[1,2,3,4,5,6,7,9,10],golden:5,gone:3,gong:2,good:[0,2,3,5,7,8,9,12,15,17],goodfellow:[14,17],googl:[0,2,12],got:2,gpu:2,grade:14,gradient:[1,6,7,10,12,14],gradientboostingclassifi:8,gradientboostingregressor:8,gradual:[0,2],graph:[2,5,7,9,10],graph_from_dot_data:7,graphic:[1,2,7],grasp:1,gray_r:2,great:5,greater:[2,5,15],greedi:7,green:[1,7,15],grid:[2,4,5,6,10,15],grossli:5,ground:1,group:[0,1,4,5,7,12,14],groupbi:1,grow:[2,7,8],growth:1,guarante:[1,5,15],guess:[0,2,5,8],guestrin:8,guid:2,h_1:5,h_2:5,h_m:8,habit:1,had:[1,2,4,5],hadamard:[2,10],half:[2,6,7],halv:8,hand:[1,2,3,5,9,10,12,13,14,15,17],handl:[1,2,3,7,9,12],handle_unknown:7,handsid:10,handwrit:10,handwritten:[2,3,5],happen:[0,2,3,5,8,15],hard:[2,5,6,8],hardcopi:12,harder:[1,2],has:[0,1,2,3,4,5,6,7,8,9,10,13,15],hasn:[1,2],hassl:[1,12],hast:12,hasti:[0,1,14,17],hat:[2,3,5,7,8,9,10,13,15],have:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],haven:2,hdf5:[],head:[1,8,15],header:1,heads_proba:8,health:1,hear:1,heart:[1,5],heatmap:[1,2,5],heavili:1,heavisid:2,height:2,help:[0,1,2,10],helper:0,henc:[1,3,4,5,6,7,8,10],her:5,here:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],hereaft:[1,6,10],hermitian:13,hessenberg:13,hessian:3,heterogen:[7,8],hidden:[2,10],hidden_bia:2,hidden_bias_gradi:2,hidden_layer_s:[1,2],hidden_weight:2,hidden_weights_gradi:2,hierarch:[0,3],high:[0,1,2,3,4,5,7,8,9,12,13],higher:[1,2,3,4,5,6],highest:2,highli:[1,8,12,13,15,17],highwai:1,hing:6,hint:5,hip:12,hire:1,his:5,hist:[4,5,15],histogram:[1,4,5,15],histor:[5,9],histori:10,histplot:1,histtyp:4,hitherto:3,hjorth:16,hobbi:15,hoc:3,hoff:17,hold:[0,2,4,5,15],holder:1,home:1,homogen:[2,7,8],hopefulli:[1,9,15],horizont:9,hors:5,hot:[2,7],hour:[2,12,14,15,16],how:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,17],howev:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],hspace:[1,6,8,15],hstack:2,htf:14,html:[5,9,12,14,17],http:[5,9,12,13,14,17],huang:1,huber:1,huge:[2,12],human:[1,2,7,10],humid:7,hundr:2,hungri:2,hybrid:14,hydrogen:1,hyperbol:[2,10],hyperparam:6,hyperparamet:[0,1,3,7],hyperplan:9,i_1:[3,4],i_2:[3,4],ian:17,idea:[1,2,4,5,7,8,10,13,15],ideal:[0,1,4,6,15],idem:4,ident:[3,4,10,13],identical:3,identifi:[0,1,2,5,6,7,9,10],idum:15,ieor:15,ifi:17,ifs:12,ignor:[1,2,7],iii:13,ijca2016907841:0,illustr:[0,3,5,8,10,12],imag:[2,7,9,10,17],image_path:[1,4,5,7],imagin:2,immedi:[1,12],implement:[0,1,3,5,6,7,8,9,10],impli:[3,4,5,13],implicitli:[9,15],impos:[1,9,10],imposs:[1,3],impress:[1,10],improv:[0,1,3,7,8,9],impur:7,imshow:2,in3050:17,in4080:17,in4300:17,in5400:17,inaccur:5,inact:10,inadequ:1,includ:[1,2,3,5,9,10,12,15,16,17],include_bia:[4,7],incom:10,inconsist:[],incorrect:2,incoveni:6,increas:[1,2,3,4,5,6,7,9,10,15],increasingli:15,ind:4,inde:[1,3],indent:15,indentationerror:15,independ:[1,3,4,5,6,10,15],index:[0,1,2,8,12,15,17],index_col:1,indic:[1,2,3,7,8,9],indispens:4,individu:[2,5,8,10,15],indu:1,indx:13,ineffici:0,inequ:6,inequaltii:5,inf1000:12,inf1100:12,inf1100l:12,inf1110:12,inf3000:17,inf4490:17,inf5860:17,inf:1,infeas:7,infer:[1,2,4,17],inferenc:2,infil:[1,4,5,7],infin:[3,4,5,9],infinitesim:15,influenc:[4,8],influenti:2,info:[],inform:[0,1,2,4,5,7,9,10,13,17],infti:[5,15],ingeni:5,ingredi:[1,7],inher:4,inherit:13,initi:[0,1,2,4,5,8,13,15],initialis:15,inject:0,inlin:[0,1,2,3,4,5,6,7,8,9,10,13,15],inner:[4,5],innov:17,input:[0,1,2,3,4,5,6,7,8,10,13,15],input_dim:2,inputs:2,inputs_shuffl:[1,2],insert:[3,6,8,15],insid:[1,5],insight:[1,2,3,12,17],insist:5,inspir:[1,2,10,17],instal:[1,2,3,7,14],instanc:[1,2,4,5,7,9],instanti:8,instead:[0,1,2,3,4,5,6,7,9,13,15],institut:2,instruct:[1,2],int32:8,int64:1,int_0:15,int_:15,int_a:15,intak:1,integ:[0,2,13,15],integer_vector:2,integr:15,intellig:[0,1,17],intend:8,intens:2,intention:0,interact:[1,7,10,12],intercept:[1,5,6,9],intercept_:[1,5,6,7],interchang:[3,10,13],interconnect:2,interest:[1,2,3,4,5,6,7,10,12,15],interfac:[1,2,13],interior:[1,7],intermedi:13,intern:[2,8,10],interpol:[2,10],interpr:3,interpret:[1,2,7,8,10,13,14,15],interv:[1,3,4,5,15],intial:5,intimid:[],intract:1,intrins:[9,13,15],intro:[12,17],introduc:[1,2,3,5,6,8,10,13,15],introduct:[2,5,14,17],introductori:[1,13,17],intuit:[1,3,4,6,10],inv:[1,3,5],invalid:[2,6,13],invalu:[1,5,12],invari:2,invd:3,inver:6,invers:[1,5],invers_period:15,inverse_transform:6,invert:[1,3,5,8],invok:[1,6],involv:[1,4,5,9,10],iomanip:15,ios:15,iostream:15,ipca:9,ipynb:12,ipython:[0,1,2,4,5,6,7,8,9,12,13,15],irani:0,iri:[6,7],irreduc:4,irrelev:3,irrespect:1,isn:3,isnul:1,isomap:9,issu:[2,7,13],it_arrai:5,item:1,items:13,iter:[0,1,2,4,6,9,15],itr:3,its:[0,1,2,3,4,5,6,7,8,9,10,12,13,17],itself:[3,4,10,15],jackknavg:15,jackknif:[4,12],jackknstd:15,jackknvar:15,jackknvec:15,jacobian:5,jargon:15,jensen:16,jerom:17,job:[6,8],join:[1,4,5,7],joint:3,journal:15,judg:5,julia:[12,13],jump:[0,15],jupyt:[0,1,12,17],just:[0,1,2,3,4,5,6,7,8,9,10,15],justif:1,justifi:8,k_mean:0,kappa_d:15,karim:16,karlsen:16,karush:6,keep:[0,1,2,3,5,9,13],keepdim:[2,4,8],kei:[1,2,10,17],kept:0,kera:[1,12,14],kernel:[1,2,12],kernel_regular:2,kernelpca:9,kev:1,kevin:17,keyword:13,kfold:4,kick:[2,5],kind:[0,1,6,10],kjm:12,kkt:[3,6],kktsolver:3,kmeanspoint:0,kn_k:0,know:[0,1,2,3,5,6,12,15],knowledg:[1,12],known:[2,3,4,5,6,7,10,13,15,17],kondev:1,kpca:9,kroneck:0,kuhn:6,kwarg:4,kwown:1,l1_l2:2,l1regl:3,l1regls_mosek2:3,l1regls_mosek:3,l_1:5,l_2:5,l_j:10,la_i:10,la_k:10,lab:[12,14],label:[0,1,2,3,4,5,6,7,8,10,12,13,15],labelencod:[5,8],labels:[6,7],labels_shuffl:[1,2],labor:0,laboratori:14,lack:[0,1],lagrang:[6,9],lambda:[1,2,3,4,5,6,8,10,15],lambda_0:9,lambda_1:[3,6,9],lambda_2:[6,9],lambda_:9,lambda_i:[6,9],lambda_iy_i:6,lambda_jy_iy_j:6,lambda_k:6,lambda_n:[3,6],lamda:2,land:[1,6],landmark:6,landscap:5,langl:[1,9,15],languag:[1,2,6,12,13,17],lapack:[3,13],laplac:3,laptop:12,larg:[1,2,3,4,5,6,7,8,9,12,13,15,17],larger:[1,3,4,5,6,8,9,15],largest:[6,9],lasso:[1,4,5,12,14],last:[0,1,2,3,4,5,6,7,8,10,13,15],later:[0,1,2,5,6,10,12,15],latex:[],latter:[1,3,5,6,9,13,15],lattic:10,law:1,layer:1,lbfg:[5,7,8,9],lbl:4,lcc:[3,4],lda:9,ldot:[1,4,9,15],lead:[1,2,3,4,5,6,7,8,9,10,13,15],leaf:7,leaki:2,lear:5,learn:[3,4,6,7,8,10,13,14,17],learner:8,learning_r:[6,8],learning_rate_init:[1,2],learning_schedul:5,least:[0,1,4,5,6,8,9,12,13,15],leav:[1,2,3,4,7,9],lectur:[1,2,3,5,8,9,10,12,13,14,17],lecturenot:[12,17],left:[1,2,3,4,5,6,7,8,9,10,13,15],leftarrow:[6,10],legend:[1,3,4,5,6,7,8],len:[0,1,2,3,4,6,7,8,9,10,13,15],length:[1,2,5,6,7,12,15],leq:[0,1,3,5,6,15],less:[1,2,3,4,6,7,15],lessen:2,let:[0,1,2,3,4,5,6,7,8,9,10,13,15],letter:[1,13,15],level:[1,2,3,4,7,12,13,14],lib:[1,2,4,5,6,9],liblinear:[6,8],librari:[1,2,3,7,8,9,13,15,17],licens:[1,2,12],lie:[1,4,9,15],lies:[6,9],life:[1,2,6,10],lift:0,light:[],like:[0,1,2,3,4,5,7,8,9,10,12,13,15],likelihood:[1,2,3,7],lim_:15,limit:[1,3,4,6,9,10,13],lin_clf:6,lin_model:1,lin_reg:7,linalg:[1,3,5,6,9,13,15],line1:6,line2:6,line3:6,line:[1,2,4,5,6,9,13,15],linear:[2,3,4,5,7,8,9,10,12,14,15],linear_model:[1,3,4,5,6,7,8,9],linear_regress:4,linearli:3,linearloc:5,linearregress:[1,4,5,7],linearsvc:6,liner:2,linerar:8,linewidth:[1,4,6,7,8],link:[1,5,7,10,12],linlag:3,linpack:13,linreg:1,linspac:[1,4,6,7,8,13,15],linu:16,linuek:16,linux:[1,2,12],liquid:1,list:[0,1,2,3,7,12],listedcolormap:[7,8],literatur:[0,2,5,17],littl:[2,7,10],live:6,lle:1,lloyd:0,lmb:[3,4],lmbd:[1,2],lmbd_val:[1,2],lmbda:5,load:[1,2,5,7,8],load_boston:1,load_breast_canc:[2,5,7,8,9],load_digit:2,load_iri:[6,7],loc:[1,4,5,6,7,8],local:[1,2,5,10],locat:6,log10:[3,4],log:[1,2,3,4,5,7,8,9,13],log_:1,log_clf:8,logarithm:[1,3,5,13],logic:[1,2,7],logist:[1,2,6,7,8,9,10,12,14],logisticregress:[5,7,8,9],logit:5,logreg:[5,7,8,9],logspac:[1,2,3,4],longer:[0,6,8,13,15],longest:0,loocv:4,look:[0,1,2,3,4,5,6,7,8,9,13,15],loop:[0,2,4,8,10,12,13,15],lose:2,loss:[1,2,3,4,6,8,9,13],lot:[0,1,2,4],low:[1,4,7,8,9,15],lower:[1,2,7,8,13],lowercas:13,lowest:[5,7,15],lstat:1,lstsq:1,lubksb:13,ludcmp:13,lux:13,lvert:2,m_1:0,m_h:1,m_k:0,m_l:10,m_n:1,m_p:1,machin:[2,3,4,7,8,9,10,13,14,17],machinelearn:[12,14,17],mackai:17,made:[1,2,3,4,5,7,9,10],mae:1,magic:0,magnitud:[2,5],mai:[1,2,3,4,5,6,7,9,10,12,13,15],mail:14,main:[1,2,3,5,7,13,17],mainli:[1,3,4,5,7],maintain:4,major:[2,4,5,7,8,13],make:[0,2,3,4,5,6,9,10,12,13,15,17],make_moon:[6,7,8],make_pipelin:[1,4,8],makedir:[1,4,5,7],makeplot:1,malcondit:13,malign:[2,5,7],mammographi:3,manag:[1,12],manhattan:0,mani:[0,1,2,3,4,5,6,7,9,11,12,13,15,17],manifold:9,manual:0,map:[0,1,2,4,5,6,9,10,15],margin:[1,3,6],marit:1,mark:[],marker:[1,5,13],markov:12,marsaglia:15,mask:15,mass:[1,2,3],massag:1,masses2016:1,masses2016ol:1,masses2016tre:1,masseval2016:1,master:14,mat1100:12,mat1110:12,mat1120:12,mat3155:14,mat4155:14,mat:12,match:[0,2,3,5],materi:[3,5,13],math:[0,5,10,13,15,17],mathbb:[0,1,3,4,5,6,9,10,13,15],mathbf:[1,3,4,5,6,13,15],mathcal:[2,3,4,5],mathemat:[0,1,5,9,10,12,13,15,17],mathemati:[],mathemt:[],mathrm:[0,1,2,3,4,5,6,7,8,9,10,15],matmul:[2,3],matnat:[16,17],matplotlib:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],matric:[1,2,5,6,9,12,13],matrix:[1,4,6,8],matshow:2,matter:5,max:[1,2,5,7,8,10],max_depth:[1,7,8],max_it:[1,2,5,6,9],max_iter:0,max_leaf_nod:8,max_sampl:8,maxdegre:[1,4,8],maxdepth:8,maxim:[2,3,5,6,9],maximum:[0,1,2,3,5,6,7,8],maxpolydegre:[3,4],mbox:[3,4],mcculloch:10,mcint:15,mcintsqr2:15,mean:[2,3,4,5,7,8,9,10,12,13,14],mean_absolute_error:1,mean_divisor:0,mean_i:15,mean_matrix:0,mean_squared_error:[1,4,5,8],mean_squared_log_error:1,mean_vector:0,mean_x:15,meaning:[1,5],meansquarederror:1,meant:[5,8],meantempvec:15,meanvec:15,measur:[0,1,2,3,4,7,9,10,15],mechan:[1,15],median:1,medicin:10,medium:6,medv:1,meet:[1,16],mehta:1,memori:[9,10,13],mention:[0,1,5,10,15],mere:1,mersienn:15,meshgrid:[3,6,7,8,9],messag:3,met:[1,6],meteorolog:7,method:[0,1,2,3,6,9,10,12,13,14,17],metric:[0,1,2,4,5,7,8],metropoli:12,mev:[1,15],mglearn:12,mgrid:5,mhjensen:[1,2,5,6,9],microsoft:17,mid:2,midpoint:7,might:[1,2,5,7],mild:7,miller:15,million:1,mimic:10,min:[1,3,6,7],min_:[0,1,3],min_samples_leaf:7,mind:[0,1,5],mine:12,mini:[2,5,9,10],minibatch:[2,5,9],minibathc:5,minim:[0,1,2,3,4,5,6,7,8,9,10,15],minima:[1,2,5],minimum:[1,2,4,5,6,7,9],minkowski:0,minmaxscal:1,minor:15,minst:2,minu:5,mirror:7,misclassif:[6,7,8],misclassifi:[6,8],mismatch:2,miss:[1,8],mit:17,mix:2,mkdir:[1,4,5,7],mlab:[4,15],mle:[3,5],mlp:2,mlpclassifi:2,mlpregressor:1,mnist:[2,9],mod:15,mode:[14,15],model:[0,3,4,5,6,7,8,9,12,15,17],model_select:[1,2,3,4,5,7,8,9],moder:8,modern:[1,4,5,12],modif:10,modifi:[1,2,3,5,6,8,10],modul:[1,4,5,7,8,9,13],modular:15,modulenotfounderror:[5,7],modulo:15,moe:[3,9],moment:[3,4],monoton:[3,10,15],mont:[1,4,12,15,17],montecarlocycl:15,moor:3,more:[1,2,3,4,6,7,8,9,10,12,14,15],moreov:1,morten:16,mosek:3,most:[0,1,2,3,4,5,6,7,8,9,10,12,15],mostli:[2,9],motion:1,motiv:2,move:[0,1,3,5,7,10,15],mpl:[1,5],mpl_toolkit:5,mplot3d:5,mplregressor:2,mse:[1,3,4,7,8],mse_simpletre:8,mselassopredict:3,mselassotrain:3,msepredict:3,mseridgepredict:3,msetrain:3,msg:1,msle:1,mt19937_64:15,mu0:15,mu1:15,mu2:15,mu_:15,mu_n:9,mu_x:15,much:[1,2,3,4,5,6,7,8,9,10,13,15],mul:3,multi:[1,2,5,12],multiclass:[2,5],multidimension:[9,10],multilay:2,multinomi:5,multipl:[3,4,5,10,15],multipli:[3,5,9,13,15],multiplum:6,multitud:[],multivari:[1,8,9,12,15],multivariate_norm:[0,9],murphi:[9,17],must:[0,2,3,4,5,6,8,10,15],mutat:5,mutual:[2,4,5],mx_:15,myriad:[1,12],mz1:15,mz2:15,n_0:[10,15],n_b:15,n_boostrap:[4,8],n_bootstrap:4,n_categori:2,n_cluster:0,n_compon:9,n_epoch:5,n_estim:8,n_featur:2,n_hidden_neuron:[1,2],n_i:15,n_input:[1,2],n_instanc:7,n_iter_i:[5,9],n_job:8,n_k:0,n_l:[10,15],n_layer:2,n_m:7,n_neuron:2,n_neurons_layer1:2,n_neurons_layer2:2,n_point:0,n_sampl:[0,4,6,7,8],n_split:4,nabla:[2,5],nabla_:5,naimi:1,naiv:[0,5],nall:1,name:[0,1,2,3,4,5,6,7,8,10,12,13,16],nameerror:8,namespac:15,nation:[2,3],nativ:12,natur:[1,2,5,6,7,10,15,17],navier:10,nb_:13,nbconvert:[],nboot:15,nearest:[2,9],nearli:5,neat:[],neccesari:4,necessari:[0,1,2,6],necessarili:[1,9,15],necesserali:3,neck:5,need:[0,1,2,3,4,5,6,7,8,9,10,13,15],neg:[1,2,3,4,5,8,15],neg_mean_squared_error:4,neglect:15,neglig:15,neighbor:9,neq:[0,5,15],nervou:10,nest:[7,10],net:10,netlib:13,network:[1,7,12,14,17],neural:[1,5,12,14,17],neural_network:[1,2],neuralnetwork:2,neuron:[2,10],neutral:1,neutron:1,never:[2,4,7,15],new_hobbit:[],new_sig:4,newaxi:[1,4,7],newli:1,newton:[2,6,15],next:[0,1,2,3,5,6,7,15],next_guess:5,nian:16,nice:[0,1,2,3,9],nichola:16,nicholaskarlsen1102:16,niter:5,nitric:1,nlambda:[3,4],nm_n:1,nmse:4,nn_model:2,node:[2,7,8,10],nois:[1,3,4,5,6,7,8],noisi:[2,4],non:[0,1,2,3,4,5,7,8,9,10,13,15],none:[0,1,2,3,4,5,7,8,15],nonetheless:0,nonlinear:[4,6,7,9,10],nonneg:[4,5,7],nonparametr:4,nonsens:15,nonsingular:13,nonumb:[5,6,13],nor:2,norm:[1,2,3,4,5,6,9],normal:[3,4,5,6,7,8,9,10,12,13],normali:13,normpdf:4,notat:[0,1,3,4,15],note:[0,1,2,3,4,5,6,9,10,12,13,14,15,17],notebook:[0,1,2,7,12],noth:[0,2,3,6,10,15],notic:[3,10,13,15],novel:[4,8],novemb:2,now:[0,1,3,4,5,6,8,9,10,13,15],nowadai:[1,2,7,12],nox:1,np_assign_points_to_clust:0,np_get_distances_to_clust:0,np_k_mean:0,nsampl:4,nspin:15,nthi:1,nuclear:3,nuclei:[1,15],nucleon:1,nucleu:1,num_tre:8,number:[0,2,3,4,5,6,7,8,9,10,13,14,16],numberid:5,numer:[1,3,4,5,7,8,9,10,12,13,17],numpi:[0,1,2,3,4,5,6,7,8,9,10,12,15],nunmpi:3,obei:[5,9],object:[1,2,3,4,6,8,13],objsens:3,obliqu:3,observ:[0,2,3,4,5,6,7,8,9,10],obtain:[0,1,2,3,4,5,6,7,8,10,13,15],obviou:[3,9,15],obviouli:1,obvious:[1,3,13],occupi:1,occur:[1,6,7,15],odd:[1,5],off:[2,3,4,7,15],offend:[],offer:[4,9,12,13,14],offic:16,offici:14,ofil:15,ofstream:15,often:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],ofter:13,old:[2,3,5,8],olsbeta:3,omit:[1,3],onc:[0,2,4,7,9,15],one:[0,1,2,3,4,6,7,8,9,12,13,15],onehot:2,onehot_vector:2,onehotencod:7,ones:[1,3,5,6,7,8,9,13],onli:[0,1,2,3,4,5,6,7,8,9,10,13,15],onlin:[9,14],onto:[3,9],open:[1,2,4,5,7,12,14,15],oper:[1,2,3,4,8,9,10,12,13,15],operation:15,opinion:0,oplu:15,opmiz:5,opportun:1,opposit:[2,3,6],opt:[1,2,3,4,5,6,9],optim:[0,1,3,4,7,8,9,14],optimis:2,option:[1,2,3,5,6,9],optmiz:[2,6],orang:1,order:[1,2,3,4,5,6,7,8,9,10,13,15],ordinari:[1,4,5,9,12,14],oreilli:17,org:[5,9,12,13,17],organ:[0,4,5,8,13],orient:[2,3,4,15],origin:[1,3,4,6,9,10,13,15],orthogn:3,orthogon:[1,3,5,6,9,13],orthonorm:3,oscar:2,oslo:[1,14,16],osx:[1,12],other:[0,1,2,3,4,5,6,8,12,13,14,17],otherwis:[1,2,5],ouput:[3,5,10],our:[2,4,6,7,8,10,12,13],ourmodel:1,ourselv:[0,1,3,5,6,9],out:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],out_fil:7,outcom:[1,5,7,8,10,15],outdoor:7,outer:10,outfilenam:15,outlier:[1,6],outlin:[4,8,9],outlook:7,outperform:8,output:[1,2,3,5,6,7,8,10,13],output_bia:2,output_bias_gradi:2,output_weight:2,output_weights_gradi:2,outputlayer1:10,outputlayer2:10,over:[0,1,2,3,4,5,7,8,10,15],overal:[2,8],overcast:7,overcom:10,overdetermin:1,overfit:[2,4,7,8],overflow:[2,3],overhead:10,overlap:[5,6,7],overlin:[0,1,3,4,7,8,9,13],overst:1,overview:[0,17],own:[3,5,6,10,12,13],owner:1,oxid:1,oyvinssc:16,p_i:[3,15],p_j:15,p_n:15,p_x:15,pack:1,packag:[1,2,3,4,5,6,9,12,15],page:[1,12],pai:[2,7],painless:[],pair:[1,7,12,15],panda:[1,3,4,5,7,9,12],panel:[],paper:2,paradigm:1,parallel:[8,13],paramet:[1,2,3,4,5,6,7,8,10,15],parameter:[1,8],parametr:[1,4],paramt:3,park:15,part:[0,1,2,3,4,8,13,14,15,17],partial:[1,2,3,5,6,8,9,10,15],particip:[12,14],particl:[1,15],particular:[1,2,3,4,5,7,8,9,10,15,17],particularli:[3,4,5,6,9,15],partit:[2,7],pass:[0,10,15],past:[8,15],patch:[4,15],path:[1,4,5,7,12],patient:5,pattern:[1,10,14,17],pauli:1,pca:[1,5,12,14],pcost:3,pdf:[1,3,4,7,17],pedagog:1,penalti:[4,5],penros:3,pentagon:5,peopl:[1,2,7,12],per:[1,2,4,14],percentag:[1,8,9],perceptron:[1,2,5],peregrin:[],perfect:[1,2],perfectli:4,perform:[0,1,3,4,5,6,8,9,10,12,13,15],perhap:[1,3,5],perimet:2,period:2,permut:9,person:[3,5,14,16],perspect:17,pertin:10,petal:[6,7],peter:17,petersen:15,phantom:15,phase:10,phatak:0,phenomena:15,phi:6,phi_k:6,philip:16,philosophi:5,phone:16,phrase:1,physic:[1,2,5,10,15,16,17],pick:[0,2,5,7,8,9,15],pickl:2,pictur:1,pie:12,piec:[0,9],pillow:[1,12],pinv:3,pip3:[1,2],pip:[1,2,12],pipelin:[1,4,6,8],pippin:[],pise:0,pitt:10,pixel:2,pixel_height:2,pixel_width:2,place:[1,4,5,6,13],plai:[1,3,4,6,9,12],plain:[5,6,8,10],plan:[4,7,16,17],plane:[6,7],plateau:[3,15],platform:12,plausibl:10,pleas:[1,5,9],plenti:2,plethora:10,plot:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],plot_confusion_matrix:[5,8],plot_cumulative_gain:[5,8],plot_data:2,plot_dataset:6,plot_decision_boundari:[7,8],plot_import:8,plot_predict:6,plot_regression_predict:7,plot_roc:[5,8],plot_surfac:5,plot_train:7,plot_tre:[7,8],plt:[0,1,2,3,4,5,6,7,8,9,10,13,15],plu:[1,3,5],png:[1,4,5,7],point:[0,1,2,3,4,5,6,7,8,9,13,15,16],points_in_clust:0,poisson:12,poli:[4,6],poly100_kernel_svm_clf:6,poly3:1,poly3_plot:1,poly_featur:[6,7],poly_features10:7,poly_fit10:7,poly_fit:7,poly_kernel_svm_clf:6,polydegre:[1,3,4,8],polygon:5,polym:10,polynomi:[1,3,4,5,6,7,8,9],polynomial_featur:4,polynomial_svm_clf:6,polynomialfeatur:[1,4,6,7],polytrop:[1,4],poor:[2,5],poorli:1,popul:[1,3],popular:[1,2,4,5,6,7,9,10,12,13,15],popularli:1,portabl:8,portion:9,pose:[1,9,15],posit:[0,1,2,3,5,6,8,9,13,15],possibl:[1,2,3,4,5,6,7,8,9,10,12,13,15,16],posterior:3,postpon:1,postul:3,potenti:[1,3,10],potr:3,potrf:3,pott:10,power:[1,2,3,4,6,7,10],practic:[1,3,4,5,6,15],practition:[1,2],pre:3,preced:[2,9,10,15],preceq:6,precis:[1,3,9,13,15],pred:4,predict:[1,2,3,4,5,6,7,8,12,17],predict_prob:2,predict_proba:[5,8],predictor:[1,3,5,7,8,9],prefer:[1,2,6,7,9,12],prepar:1,preprocess:[1,4,5,6,7,8,9],prerequisit:1,present:[1,3,7,10,15],preserv:9,press:[5,17],pretrain:2,pretti:[1,6,7,12],prev_centroid:0,prevent:15,previou:[1,2,3,5,6,8,9,10,13,15],previous:[0,7,8,15],price:[1,7],primal:6,primari:[1,5],primarili:0,prime:15,princip:[1,3,5,12,14],principl:[0,1,4,5,6],print:[0,1,2,3,4,5,6,7,8,9,13,15],print_funct:[6,7],printout:1,prior:[1,3,4],privat:1,prob:[2,15],probabilist:[1,17],probabl:[1,2,4,5,8,12],problem:[1,3,4,6,7,8,9,10,12,13,14,15],proce:[1,3,5,6,7,8,9,13],procedur:[3,4,5,6,8,9],proceed:13,process:[0,1,4,5,7,8,10,12,13,15,17],prod:17,prod_:[2,3,5],produc:[0,1,3,7,8,9,10,12,13,15],product:[1,2,3,4,5,6,10,12,13],profess:1,profil:0,program:[0,1,2,3,6,10,12,13,14,15],programm:13,progress:[0,2],progression_plot:0,prohibit:4,project:[1,2,3,5,9,12,14,15,16],project_root_dir:[1,4,5,7],promin:10,promis:6,prone:7,pronounc:12,proof:[1,5,9,10],prop:4,proper:[1,4],properli:[0,2,6,8],properti:[1,2,4,5,10,13],proport:[1,2,3,7,9,15],propos:[2,8],propto:[3,5],proton:1,prove:5,provid:[1,2,3,4,5,6,7,8,10,12,13,15,17],proxi:2,prun:0,prune:7,pseudoinv:3,pseudoinvers:3,pseudorandom:15,psycholog:1,ptratio:1,punish:[1,2],pure:[7,15],purest:7,puriti:7,purpos:[0,1,8,10],put:2,putarow:3,putboundslic:3,putclist:3,putobjsens:3,putqobj:3,pycod:[],pydata:12,pydot:7,pyhton2:[],pylab:[1,5],pypi:12,pyplot:[0,1,2,3,4,5,6,7,8,9,10,13,15],pythagora:[0,3],python2:1,python3:[1,2,4,5,6,9,12],python:[2,3,6,9,10,14],pytorch:[1,12],qquad:[9,13],quad:[2,5,13],quadrat:[1,3,5,6,7],qualit:[7,15],qualiti:[1,7,12],quantifi:2,quantil:8,quantit:[1,4,7],quantiti:[0,1,3,4,5,7,8,9,10,13,15],quantum:10,quartil:1,queri:7,question:[1,3,4,5,7,9,10],quick:15,quickli:[2,5,7,9],quirk:0,quit:[2,3,4,7,8,10],quot:15,r2_score:1,r2score:1,r_1:7,r_2:7,r_j:7,r_m:7,rad:1,radial:[1,6,10],radioact:15,radiu:[1,2],rain:7,rais:4,ramp:2,ran1:15,ran2:15,ran3:15,rand:[1,3,4,5,7,8,13,15],rand_max:15,randint:[4,5,7],randn:[1,2,3,4,5,7,9],random:[0,1,2,3,4,5,6,7,12,13,14],random_devic:15,random_forest_model:8,random_index:5,random_indic:2,random_st:[1,5,6,7,8,9],randomforestclassifi:8,randomli:[0,2,4,5,7,15],randomnumbergener:15,rang:[0,1,2,3,4,5,7,8,9,10,13,15],rangl:[1,9,15],rangle_x:15,rank:3,raphson:[2,6],rapidli:1,rare:2,rate:[1,2,5,6,7,8,10],rather:[1,2,3,4,5,6,7,8,9,10,13,15],ratio:[5,7,8,9],rational:1,ravel:[3,4,5,6,7,8,9],raw:15,rbf:[6,9,10],rbf_kernel_svm_clf:6,rbf_pca:9,rcond:1,rcparam:[1,2,5,6,7,8,15],reach:[0,1,2,3,4,5,7,8,9,10,15],read:[0,1,3,4,5,6,9,10,13,14,15,17],read_csv:[1,4,5,7],read_fwf:1,readabl:0,reader:[1,13,15],readi:[0,1,2,3,6,8,9,10,13],readili:2,real:[1,2,4,5,8,9,10,13],realist:6,realiti:15,realiz:[2,10],realli:2,reason:[0,1,2,5,8,17],reassign:2,recal:[3,4,5,7,8,9,10,13,15],recalcul:15,receiv:[2,8,10,15],recent:[1,4,5,7,8],recept:10,recip:[1,5,13],reciproc:3,recogn:[1,3,8],recognit:[1,2,10,14,17],recommend:[1,6,12,13,14,15,17],reconsid:7,reconstruct:9,record:[8,14],recreat:15,rectangl:[4,5,7],rectangular:3,rectifi:[2,10],recur:[1,12],recurr:[2,12,14],recurs:[7,12,13],recycl:15,red:[1,4,6,7],redefin:[1,8],reduc:[2,3,5,7,8,9],reduct:[1,8,9,12,15],refer:[0,1,2,3,4,5,9,10,13,17],refin:10,refit:4,reflect:[1,2,15],refresh:[12,14],reg:[8,9],regard:[2,5,7],regardless:10,region:[7,10],regist:15,reglasso:3,regr_1:[1,7],regr_2:[1,7],regr_3:[1,7],regress:[2,4,6,9,10,12,14],regressor:[1,5,8],regridg:3,regular:[1,3,5,7],reilli:[1,17],reinforc:[1,6,12],reiter:2,rel:[1,4,5,7,10,15],relat:[0,1,2,3,5,9,13,15],relationship:[1,7],relativeerror:1,releas:[2,12],relev:[1,2,3,5,9,12,15],reli:[1,6],reliabl:[5,15],remain:[2,4,10,13,15],remaind:15,remark:2,rememb:[1,6,13],remind:[1,3,9,13,15],remov:[1,3],render:1,reorder:[3,5],reorgan:1,repeat:[0,1,2,3,4,5,7,8,9,13,15],repeated:1,repeatedli:[4,8,15],repetit:[4,14],rephras:5,replac:[0,1,2,3,4,5,8,10],replica:4,repositori:1,repres:[1,2,3,4,5,6,7,8,10,15],represent:[1,2,4,15],reproduc:[1,3,7,10,12,15],repuls:1,request:1,requir:[1,2,3,4,5,6,7,9,10,13],resampl:[1,5,8,12,14,15],rescal:[1,9,10],rescu:3,research:[1,12,17],resembl:[4,15],reserv:[2,3,4,15],reset:15,reshap:[0,1,2,4,6,7,8],residenti:1,residu:[1,3,5],resiz:3,respect:[0,1,2,3,4,5,6,8,9,10,15],respond:10,respons:[1,5,7,10],rest:[1,3],restat:[1,10],restrict:[1,7,10],result:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],ret:4,retail:1,retain:[3,4],return_data:0,return_x_i:7,reus:2,reveal:[1,10],revers:2,review:[12,13],revisit:0,reward:1,rewrit:[1,3,4,5,6,8,9,10,13,15],rewritten:[4,6,8,15],rewrot:5,rgoj5yh7evk:12,rho:[1,8],rho_1:8,rho_2:8,rho_m:8,rhs:3,rich:1,rid:0,ride:7,rideclass:7,ridedata:7,ridg:[1,4,9,12,14],ridgebeta:3,right:[1,2,3,4,5,6,7,8,10,13,15],rightarrow:[1,2,3,5,6,9,10,15],rigor:1,rise:1,risk:[1,5],river:1,rmse:1,rmsprop:2,rnd_clf:8,rnn:10,rntrick1:15,rntrick2:15,rntrick3:15,rntrick4:15,robert:17,robust:1,robustscal:1,roc:8,role:[1,3,4,6,12],room:[1,16],root:[1,3,5,7,15],rot:[],rotat:[2,6,7,8],rotation_matrix:7,roughli:2,round:[1,5,7],routin:[5,13],row:[1,2,3,4,5,7,9,13],rrr:3,rthe:3,rug:5,rule:[1,2,3],run:[0,1,2,3,4,5,6,7,9,12],runtim:[0,2,4],runtimewarn:2,rust:[1,12,13],rvert:2,rvert_2:2,rwidth:4,s_i:5,saddl:5,safe:15,sai:[0,1,2,3,4,5,6,7,8,9,10,13,15],said:[4,5,7],sake:[1,3,5,9],sale:1,sam:[],same:[0,1,2,3,4,5,6,7,9,10,13,15],samm:8,sampl:[0,1,2,3,4,5,6,7,8,12,13],sample_vari:0,sampleexptvari:15,samwis:[],sanitize_sequ:4,sastri:9,satisfactori:1,satisfi:[2,4,5,6,13,15],satur:[2,4],save:[1,4,5,7],save_fig:[1,4,5,7,8],savefig:[1,4,5,7,15],saw:3,scalabl:8,scalar:[3,4,8],scale:[1,2,3,5,6,7,8,9,10,12,16],scaler:[1,5,6,7,8,9],scan:[3,5],scari:3,scatter:[0,1,2,4,5,6,7],scenario:5,scheme:[2,5],schrage:15,scienc:[1,2,5,8,10,12,14,15,17],scientif:[1,12],scientist:[0,1],scikit:[3,6,7,8,12,13,14,17],scikitlearn:1,scikitplot:[5,8],scipi:[1,3,5,12,13],score:[1,2,4,5,7,8,9,16],scores_kfold:4,scratch:2,sdg:5,seaborn:[1,2,5],seamless:[1,12],search:[1,2,3,5,7],sec:4,second:[0,1,3,4,5,6,7,9,10,12,13,15],secondeigvector:9,secondli:10,section:[0,9,11,13,14,15],sector:1,see:[0,1,2,3,4,5,6,8,9,10,12,13,14,15],seed:[0,1,2,3,4,5,6,7,9,15],seek:[2,6],seem:2,seemingli:1,seen:[1,2,3,8,10,15],segment:5,seldomli:1,select:[2,3,4,6,7,8,9,14,17],self:[2,3,4,15,17],semest:[5,14],semi:[5,6],send:[3,10,16],senior:14,sens:[1,4,6],sensit:[1,3,4,7],sentenc:10,separ:[0,1,2,4,6,7,10,12,15],sequenc:[0,5,7,8,10,12,13,15],sequenti:[2,8,10,15],seri:[1,2,3,5,8,9,10,13,14],serif:[1,5,15],serv:[1,2,3,5,17],session:[2,14],set:[0,2,3,4,5,6,8,9,12,13,15],set_:4,set_label:4,set_stream:3,set_tick:[2,6],set_ticklabel:2,set_titl:[0,1,2,5,10],set_xlabel:[1,2,5,10],set_xlim:[5,10],set_xticklabel:2,set_ylabel:[1,2,5],set_ylim:[5,10],set_ytick:5,set_yticklabel:2,setiosflag:15,setminu:4,setosa:[6,7],setosa_or_versicolor:6,setp:4,setprecis:15,setse:3,setup:[2,3,6,12],setw:15,sever:[1,3,4,5,6,7,9,10,12,13,14,15],sgd:2,sgd_clf:6,sgdclassifi:6,sgdreg:5,sgdregressor:5,sgn:3,shape:[0,1,2,3,4,5,6,7,8,9,13],share:2,she:5,shift:[2,10,15],shire:[],shortcom:5,shorter:15,shorthand:[],shortli:13,should:[0,1,3,4,6,7,9,10,13],show:[0,1,2,3,4,5,6,7,8,9,10,13,15],shown:[1,3,5,6,9,10,15],showpoint:15,shrink:[3,6,9],shrinkag:3,shrunk:9,shuffl:[1,2,4],side:[1,3,5,6,10,13],sigh:12,sigma0:15,sigma1:15,sigma2:15,sigma:[1,2,3,4,5,8,9,10,13,15],sigma_0:3,sigma_1:3,sigma_2:3,sigma_:[3,13,15],sigma_fn:[5,10],sigma_i:[1,3],sigma_j:3,sigma_m:15,sigma_n:[9,15],sigma_x:15,sigmoid:[2,5,6,8,10],sigmundson:16,sign:[2,5,6,8,15],signal:[2,8,10],signific:2,significantli:[2,5,15],sim:[3,4,15],similar:[0,1,2,3,4,5,6,7,8,9,12,13],similarli:[1,2,3,6,8,15],simpl:[0,2,3,4,6,8,9,10,12,13],simplepredict:8,simpler:[0,1,2,3,12],simplest:[0,1,2,7,8,10],simpletre:8,simpli:[1,2,3,4,6,7,8,9,10,12,13,15],simplic:[0,3,5,6,7,8,9,10],simplicti:3,simplifi:[1,4,7,12],simplist:15,simul:[4,15],simultan:4,sin:[1,2,7,10,13],sinc:[1,2,3,4,5,6,7,8,9,13,15,17],sine:10,singl:[2,3,5,6,7,10,13,15],singular:[1,4,5,13],site:[1,2,4,5,6,9,14],situat:[1,3,5],six:15,size:[1,2,3,4,5,6,7,8,9,13,15],sketch:8,ski:7,skill:1,skip:9,skl:1,sklearn:[1,2,3,4,5,6,7,8,9],skplt:[5,8],slack:6,slice:13,slide:[1,15],slight:4,slightli:[2,3,4,8,15],slope:[6,9,10],slow:[1,5,6],slower:[3,13],slowli:10,slp:2,small:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],smaller:[1,2,3,4,5,6,7,9,15],smallest:[0,1],smallest_row_index:0,smart:4,smooth:[1,5],sne:9,sneak:1,sns:[1,2,5],soar:4,social:1,soft:[2,5,8,10],soften:6,softmax:5,softwar:[1,6,12,13,14],sol:6,sole:1,solid:[1,5],solitem:3,soltyp:3,solut:[1,2,3,4,5,6,8,9,13,15],solutionsummari:3,solv:[1,2,3,6,8,9,10,13],solver:[3,5,6,7,8,9,13],some:[0,1,2,3,4,6,7,8,9,10,13,15],someth:[1,2,5,7,9,15],sometim:[0,1,2,9,10],soon:13,sophist:1,sopt:5,sort:[3,4,7,9,15],sound:3,sourc:[0,1,2,4,12,13,15],space:[0,1,2,3,5,6,7,9,10,15],span:[1,3,7,9,13],spare:2,spars:[3,13],sparse_mtx:13,sparsiti:8,spatial:[2,10],spdiag:3,speak:15,special:[4,5,8,10,13,15],specif:[1,2,3,4,5,6,7,9,10,12,13,15],specifi:[0,1,3,4,5,7,9,15],specifici:[1,8],spectral:2,speech:[1,2,10],speed:2,spend:15,sphere:1,spite:1,spline:6,split:[0,2,3,4,6,7,8,9,15],splitter:[2,8],spmatrix:3,spontan:15,spread:[1,9,15],springer:17,sqquar:3,sqrt:[1,3,4,5,6,8,9,15],squar:[0,2,4,5,6,7,9,12,13,15],squarederror:8,squaredeuclidean:0,squash:10,srand:15,stabil:3,stabl:[1,3,5,7,9,12],stack:4,stage:[3,5],stai:[1,3,9],stand:[1,3,7,10],standard:[1,2,3,4,6,8,10,13],standardscal:[1,5,6,7,8,9],stanford:5,start:[0,1,2,3,4,5,6,7,8,9,10,13,15],start_tim:0,startpoint:15,stat:4,state:[0,2,3,5,6,8,9,10,12,15],statement:[1,5,13],statis:[],statist:[0,1,2,5,7,8,9,10,13,14,17],statu:[1,5,9],std:[1,4,15],stdev:15,stdout:3,steep:5,step:[0,1,2,7,8,9,10,13,15],step_fn:[5,10],step_length:5,steps_list:7,stian:16,still:[3,4,5,9,15],stimuli:10,stk2100:17,stk4021:17,stk4051:17,stk5000:17,stk:17,stochast:[1,2,3,4,6,9,10],stoke:10,stone:[1,5],stop:[0,2,7,9,15],storag:3,store:[1,2,5,9,15],str:2,straight:[1,4,5,6],straightforward:[1,3,4,5,6,7,8,13],strategi:[1,2,7],stratifi:4,streamtyp:3,strength:[0,1,3],stretch:9,strict:[5,6],strictli:[5,6],string:[2,15],stroke:5,strong:[7,8,10,15],strongli:[1,6,12,13],stronli:1,structur:[0,1,2,4,7,8,10,12],stuck:[2,5],student:[1,14,16,17],studi:[0,1,3,5,6,9,10,12,17],studier:17,style:[1,5,7],sub:[7,10],subdivid:[1,13],subfield:1,subject:[3,6,15],subplot:[0,1,2,4,5,6,7,8],subplots_adjust:[6,15],subprogram:13,subract:1,subroutin:1,subscript:2,subsequ:[2,3,4,10,13,15],subset:[2,4,5,7,10,12],subspac:[1,6,9],substanti:[7,8],substep:9,substitut:[4,10,13],subsubset:7,subtl:2,subtract:[1,3,4,9,13,15],subtre:7,succeed:1,success:[5,7,15],successfulli:7,sucess:15,sudo:[1,12],suffer:[1,2,3,8],suffici:[2,4,5,6,9],suggest:[2,5,17],suit:[6,10],suitabl:[1,15],sum:[0,1,2,3,4,5,6,7,8,9,10,15],sum_:[0,1,2,3,4,5,6,7,8,9,10,13,15],sum_i:[3,4,5,6],sum_k:[6,10,13],summar:[0,3,4,7],summari:[0,2,8,14],summat:[],sunni:7,superscript:[2,10],supervis:[1,3,4,5,7,10,12],supplement:5,support:[1,2,7,8,9,12],suppos:[1,3,4,5,6,8,9,10,13],suppress:3,sure:2,surfac:1,surpris:1,surround:12,survei:[1,3],svc:[6,7,8],svd:[1,4,9],svdinv:3,svm:[6,7,8,9],svm_clf:[6,8],swath:3,symbol:[2,3,9,12,15],symmeteri:2,symmetr:[1,3,5,6,9,10,13],sympi:[1,12],synonim:15,syntax:[2,13],syntaxerror:[2,6,13],syrk:3,sys:[3,5],system:[1,2,3,5,7,8,10,12,13,17],systemat:[4,15],t_0:[5,7],t_1:5,t_b:8,t_i:[2,10],t_j:[3,10],t_k:7,tabl:[7,15,16],tabul:1,tabular:[],tackl:0,tag:[0,3,5,10,13,15],taht:1,tail:15,tailor:[6,9],taiwan:1,take:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],taken:[1,2,4,8,13],tangent:[2,5,10],tanh:[2,5,6,10],target:[1,2,3,5,6,7,8,9,10],target_nam:7,task:[0,1,2,3,4,7,9,10],tau:[3,15],tax:1,taylor:5,taylornr:5,team:2,teaser:1,technic:[0,1,3],techniqu:[1,2,6,8,12,15,17],technolog:[1,2],tek5040:17,tell:[1,4,5,8,9,15],temp1:2,temp2:2,temp:2,temperatur:[1,7],temporarili:2,ten:[],tend:[0,3,4,6,7,8,10],tendenc:1,tension:4,tensorflow:[0,1,6,12,13,14,17],term1:[3,9],term2:[3,9],term3:[3,9],term4:[3,9],term:[0,1,2,3,4,5,6,7,8,9,10,15],termin:[1,3,7,8],test:[3,4,5,6,7,8,15],test_accuraci:2,test_data:0,test_ind:4,test_pr:2,test_predict:2,test_scor:[5,8],test_siz:[1,2,3,4,8],test_split:7,tester:[],testerror:[1,4],text:[1,2,3,5,6,7,9,13,15,17],textual:7,textur:2,than:[1,2,3,4,7,8,9,10,12,15],thats:0,theano:[2,12],thei:[0,1,2,3,4,5,6,7,9,10,13,15],them:[1,2,5,6,7,8,9,10,13],theme:1,themselv:[1,15],thenc:4,theorem:[4,5],theoret:[1,8],theori:[1,2,5,6,7,10,12,17],thereaft:[1,4,9,10,13],therebi:[1,3,5,9,15],therefor:[1,2,4,5,6,9,15],therein:9,thereof:[1,4,5],theta:[2,4,5,15],theta_:2,theta_i:2,theta_k:15,theta_linreg:5,thi:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,17],thing:[0,1,2,3,5,7,15],think:[0,1,2,4,5,7,10,15],third:[1,5],thirti:5,thorough:0,those:[0,3,6,7,8,9,13,14],though:[2,13,15],thought:[0,4,15],thousand:[1,2],three:[1,2,3,4,6,7,10,13,14,16],threshold:[2,5,7,8,9,10],through:[0,1,2,3,5,6,9,10,12,13,15],throughout:[0,1,3,12,13,15],thu:[1,2,3,4,5,6,8,9,10,15,16],thumb:1,thursdai:14,tibshirani:[14,17],ticker:[5,15],tight_layout:[2,5],tightli:9,tild:[1,3,4,9,15],till:[1,5,6,7,8,10,13],time:[0,1,2,3,4,5,6,7,8,9,10,13,14,15],timefunct:15,tini:2,tip:11,titl:[1,2,4,5,6,7,8,15],to_categor:2,to_categorical_numpi:2,to_numer:[1,4],to_str:15,togeth:[1,6,9],toi:0,toler:0,tomographi:10,too:[1,3,4,5,7,9,15,17],took:6,tool:[0,1,2,4,12,15],toolbox:6,top:[1,3,4,7,8,12],topic:[0,1,3,5,6,12,17],topolog:[2,10],toss:8,total:[0,1,2,4,5,6,8,9,10,15,16],totalclustervari:0,totalscatt:0,totalvari:15,toward:[2,5,10],town:1,tpng:7,traceback:[4,5,7,8],track:[0,5,13],tract:1,tractabl:1,trade:[3,4,7],tradeoff:[1,3,14],tradit:[1,2,4],train:[3,4,5,6,7,8,9,10],train_accuraci:[1,2],train_end:[1,2],train_ind:4,train_pr:2,train_siz:[1,2],train_test_split:[1,2,3,4,5,7,8,9],train_test_split_numpi:[1,2],trainerror:1,trainingerror:4,trait:1,tran:3,transfer:7,transform:[1,3,4,5,6,7,8,9,10,12,13],transit:10,translat:[2,8],transpos:[2,3,9],treat:[1,2,4,5,10,15],tree:[1,2,12,14],tree_clf:[7,8],tree_clf_:7,tree_clf_sr:7,tree_reg1:7,tree_reg2:7,tree_reg:7,trend:15,trevor:17,tri:7,triain:1,trial:[1,4,5,15],triangl:5,triangular:13,trick:[6,9,15],trickier:15,tridiagon:13,trillion:12,trivial:[1,2,3,9,15],troubl:[1,6,10],true_divid:2,true_fun:4,tucker:6,tumor:[5,7],tumour:5,tunabl:2,tune:[7,13],turn:[1,2,3,4,5,6,7,8,9,10,13,15],tutori:2,tweak:[2,8,15],twice:5,twist:9,twister:15,two:[0,1,2,3,4,5,7,8,9,10,13,14,15,17],tx_1:5,type:[1,2,4,5,6,8,13,15],typeerror:[],typic:[1,2,3,5,7,8,10,15],u_i:10,u_m:8,ubuntu:[1,12],uci:1,uio:[16,17],unari:13,unbalanc:[4,7],unbias:[1,3,4,15],uncertainti:[1,3],uncertitud:15,unchang:2,uncorrel:8,undefin:3,under:[1,2,3,4,5,8,12],underdetermin:1,underfit:[2,4],underflowproblem:3,undergo:3,undergradu:14,underli:[1,2,5,7,15],underset:0,understand:[0,1,2,3,5,8,12],understood:[0,6],undesir:6,undetermin:[3,6],unexpect:[4,15],unexpected:15,unfortun:[2,6,7,8],unicode_liter:[6,7],uniform:[1,2,3,5,9],uniform_real_distribut:15,uniformli:[5,15],unifrompdf:15,unimport:5,union:[3,4],uniqu:[0,1,4,5,13],unique_cluster_label:0,unit:[1,2,3,8,10,15],unitari:[3,13],unitarili:13,uniti:15,univari:15,univers:[1,2,5,14,16],unix:2,unknow:[1,13],unknown:[1,2,3,4,6,8,13],unknowwn:10,unlabel:2,unless:[1,4,5,9],unlik:[2,5,6,15],unnecessarili:7,unravel:2,unrol:9,unseen:[5,7],unstabl:2,unsupervis:[1,2,10,12,14],unsymmetr:13,until:[0,2,5,7,10,15],unusu:10,updat:[0,2,4,8,10],upload:[12,17],upon:[2,9,13],upper:[1,6,7,13],uppercas:[13,15],ups:[],usag:[1,6,12,15],usd10000:1,usd:1,use:[0,1,3,4,5,6,7,8,9,10,12,13,14],usecol:1,used:[0,1,2,3,4,6,7,8,9,10,12,13,15,17],useful:[1,2,3,4,5,7,9,10,12,13,15,17],useless:2,user:[1,2,5,6,9,12,13],uses:[1,2,3,4,7,9,10,13,15],usetex:15,using:[0,2,3,4,6,7,8,9,10,13],usr:15,usual:[0,1,5,10],util:[0,2,4,5,8],v_0:9,valid:[1,2,5,7,8,12,14,15],valu:[0,1,2,4,5,6,7,8,10,12,13],valuat:7,valueerror:[],van:1,vandenbergh:[5,6],vandermond:1,vanilla:[1,9],vanish:[2,5,15],var_x:15,varabl:6,varepsilon:[3,4],varepsilon_:[3,4],varepsilon_i:[3,4],vari:[1,2,4,8],variabl:[0,1,2,3,4,6,8,9,10,13],varianc:[0,1,2,3,5,7,8,9,12,13,14],variance_i:[3,9],variance_x:[3,9],variant:[1,2,4,5,6,10],variat:9,varieti:[1,10,12],variou:[2,3,5,6,7,9,10,12,13,15],vartempvec:15,varvec:15,vaue:2,vdot:5,vec:[4,15],vector:[0,1,2,3,4,5,7,8,9,12],vector_mean:0,ventur:[1,6,12],verbos:2,veri:[0,1,2,3,4,5,6,7,8,9,10,15,17],verifi:[9,13],versatil:6,versicolor:[6,7],version:[0,1,8,12,13,15],versu:2,vert:[1,2,3,5,6,7,9],vert_1:3,vert_2:[3,9],vertic:4,via:[1,3,4,5,6,7,8,9,10,12,13,14,15],vidal:9,video:[1,2,5,10,12,14],view:[2,3,10,15,17],violat:6,virginica:7,viridi:[1,2],virtual:2,vision:1,visual:[1,9,10,12],visualis:2,viz:[6,15],vmax:2,vmc:15,vmin:2,volum:1,vote:8,voting_clf:8,votingclassifi:8,votingsimpl:8,vstack:[3,9,13,15],w_1:[6,13],w_1x_1:6,w_1x_:6,w_2:[6,13],w_2x_2:6,w_2x_:6,w_3:13,w_4:13,w_i:[2,8],w_ix_i:10,w_j:13,w_m:13,w_px_:6,w_px_p:6,wai:[0,1,2,3,4,5,6,8,9,10,11,13,15],walk:7,walker:15,wang:1,want:[0,1,2,3,4,5,6,7,8,9,10,12,15],warn:[1,2,6],warrant:4,watch:12,wavelet:6,weak:[0,7,8],weather:[2,10],web:[12,14],websit:[13,14],wedg:[6,15],wednesdai:14,wee:9,week:[3,5],weekli:[12,17],weight:[1,2,4,5,7,8,10,15],welcom:[6,12],well:[1,2,3,4,5,6,7,8,10,12,13,15,17],went:6,were:[0,1,2,3,4,5,6,8,9,10,15],wessel:1,what:[0,2,3,4,5,6,7,8,9,10,12,13,14],when:[0,1,2,3,4,6,7,8,9,10,13,15],whenev:15,where:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,16],wherea:[4,15],wherein:[2,10],whether:[1,3,5,7,15],which:[0,1,3,4,5,6,7,8,9,10,12,13,14,16],whichev:2,white:7,who:[1,14],whole:[0,2,3,7,9],whose:[1,4,8,15],whow:[3,9],why:[1,2,5],wide:[1,2,4,5,10,12,13],widehat:4,width:[1,6,7],wieringen:1,win:8,wind:7,wing:16,wiscons:5,wisconsin:8,wise:[1,2,3,10],wish:[0,1,3,5,6,9,13],with_std:1,within:[0,1,5,7,10,15,17],withinclust:0,without:[1,2,3,5,6,7,9,10],won:1,wonder:6,word:[0,1,2,3,15],work:[0,1,2,4,5,6,7,12,14,15],world:[1,6],worldwid:1,wors:[1,2,4],worth:7,would:[1,2,3,4,5,6,7,8,9,10,13,15],wrap:13,write:[1,2,3,5,6,10,11,13,14,15],written:[1,3,5,9,10,12,13,15],wrong:[2,6],wrongli:[8,15],wrote:[3,9],wrt:8,wth:8,www:[12,13,17],wx_1:6,x0s:6,x1_exampl:6,x1d:6,x1s:[6,7,8],x2d:[6,9],x2d_train:9,x2dsl:9,x2s:[6,7,8],x3s:6,x_0:[1,3,9,13],x_1:[1,3,4,5,6,7,8,9,13,15],x_2:[1,3,4,5,6,7,8,9,13,15],x_3:[6,13,15],x_4:13,x_center:9,x_data:2,x_data_ful:2,x_i:[0,1,2,3,4,5,6,7,8,9,10,13,15],x_ix_:1,x_iy_i:6,x_j:[1,6,7,10,15],x_jy_j:6,x_k:[0,10,13,15],x_l:15,x_m:[10,13,15],x_n:[1,4,5,6,9,10,13,15],x_new:[7,8],x_p:[5,7],x_poli:7,x_poly10:7,x_reduc:9,x_scale:6,x_test:[1,2,3,4,5,7,8,9],x_test_scal:[1,5,7,8,9],x_train:[1,2,3,4,5,7,8,9],x_train_scal:[1,5,7,8,9],x_val:2,xarrai:12,xavier:2,xbnew:5,xcode:[1,12],xdclassiffierconfus:8,xdclassiffierroc:8,xg_clf:8,xgb:8,xgbclassifi:8,xgboost:7,xgboot:8,xgbregressor:8,xgparam:8,xgtree:8,xi_1:6,xi_:6,xi_i:6,xlabel:[1,2,3,4,5,6,7,8,15],xlim:[4,8],xmesh:5,xnew:[1,5],xpanda:1,xpd:[3,9],xplot:1,xscale:1,xsr:7,xt_x:5,xtest:4,xtick:[4,6,7],xtrain:4,xytext:6,y_0:[1,3,9,13],y_1:[1,3,5,6,7,9,13],y_1y_1:6,y_1y_1k:6,y_1y_2:6,y_1y_2k:6,y_1y_n:6,y_1y_nk:6,y_2:[1,3,6,7,9,13],y_2y_1:6,y_2y_1k:6,y_2y_2:6,y_2y_2k:6,y_3:[1,7,13],y_4:13,y_data:[1,2,3],y_data_ful:2,y_decis:6,y_fit:1,y_i:[1,2,3,4,5,6,7,8,9,10,13],y_if_:8,y_ix_:1,y_ix_i:[5,6],y_iy_jk:6,y_j:[4,6,10],y_k:10,y_m:13,y_model:[1,3],y_n:[5,6],y_ny_1:6,y_ny_1k:6,y_ny_2:6,y_ny_2k:6,y_ny_n:6,y_ny_nk:6,y_plot:7,y_pred1:7,y_pred2:7,y_pred:[1,2,4,5,6,7,8],y_pred_rf:8,y_pred_tre:8,y_proba:[5,8],y_test:[1,2,3,4,5,7,8,9],y_test_onehot:2,y_test_predict:1,y_train:[1,2,3,4,5,7,8,9],y_train_onehot:2,y_train_predict:1,y_val:2,year:[1,12],yes:[4,5],yet:[1,2,4,6,9],yield:[0,1,3,4,5,6,8,10,13,15],ylabel:[1,2,3,4,5,6,7,8,15],ylim:4,ymesh:5,yoshua:[2,17],you:[0,1,2,3,4,5,6,7,8,9,12,13,15,17],young:1,your:[0,2,4,5,6,9,12,13,15],yourself:[5,9],youtub:12,ypred:4,ypredict2:5,ypredict:[1,5],ypredictlasso:3,ypredictol:3,ypredictridg:3,yridg:1,ytest:4,ytick:[4,6,7],ytild:[1,4],ytildelasso:3,ytildenp:1,ytildeol:3,ytilderidg:3,ytrain:4,z_0:13,z_1:13,z_2:13,z_c:2,z_h:2,z_i:[2,10],z_j:[2,10],z_k:10,z_m:2,z_mod:7,z_o:2,zaman:15,zero:[0,1,2,3,4,5,6,7,8,9,10,13,15],zm_h:1,zone:1},titles:["11. Clustering Analysis","3. Linear Regression","13. Building a Feed Forward Neural Network","4. Ridge and Lasso Regression","5. Resampling Methods","6. Logistic Regression","7. Support Vector Machines, overarching aims","8. Decision trees, overarching aims","9. Ensemble Methods: From a Single Tree to Many Trees and Extreme Boosting, Meet the Jungle of Methods","10. Basic ideas of the Principal Component Analysis (PCA)","12. Neural networks","Content in Jupyter Book","Applied Data Analysis and Machine Learning, FYS-STK3155/4155 at the University of Oslo, Norway","2. Linear Algebra, Handling of Arrays and more Python Features","Teaching schedule with links to material","1. Elements of Probability Theory and Statistical Data Analysis","Teachers and Grading","Textbooks"],titleterms:{"2021":16,"4155":12,"case":[5,6,8,15],"final":[5,10],"function":[1,2,5,6,8,9,10,15],"import":[3,13,15],And:5,Eye:8,FYS:12,OLS:3,The:[0,1,2,3,4,5,6,7,9,10,12,15],Useful:12,Using:5,activ:[2,10],actual:15,adaboost:8,adapt:8,adjust:2,again:7,aim:[6,7],algebra:13,algorithm:[0,7,8,9,10],algortithm:5,all:6,analys:3,analysi:[0,1,3,9,12,15],ani:5,anoth:7,appli:12,approach:[1,5,6],approxim:10,architectur:2,arrai:13,assist:16,august:14,autocorrel:15,back:[2,9,10],background:12,bag:8,basic:[0,1,3,5,7,8,9,13],batch:2,bay:3,befor:9,better:[6,15],bia:4,binari:2,binomi:15,bird:8,block:15,book:11,boost:8,bootstrap:[4,8,15],boston:1,breast:2,brief:5,bring:10,build:[2,7],calcul:15,cancer:[2,5,7,9],cart:7,central:[5,12,15],chain:10,chang:8,chi:1,choos:2,classic:9,classif:[2,7,8],classifi:6,clip:2,cluster:0,code:[0,1,2,3,5,7,9,10,15],collect:2,compar:8,complex:1,compon:9,comput:[5,7,15],con:7,concept:15,condit:5,conjug:5,content:11,continu:15,convex:[5,6],convolut:10,correl:[9,15],correspond:5,cost:[2,5,8],cours:[12,17],covari:[3,9,15],cross:4,cumul:15,cython:[],data:[1,2,5,7,9,12,15],dataset:2,decemb:14,decis:[7,8],decomposit:[3,9,13],deep:2,defin:2,definit:15,degre:1,demonstr:15,dens:1,deriv:[3,5,10],descent:[5,8],develop:2,deviat:15,diagon:9,dice:15,differ:[5,6],dimension:6,disadvantag:7,discret:15,disguis:15,distribut:[3,15],doing:2,domain:15,down:2,dropout:2,economi:[],element:[1,15],elimin:13,ensembl:8,entropi:7,environ:1,equat:[1,5,10],error:[1,8],etc:[],evalu:2,event:15,exampl:[1,2,5,6,7,8,15],exercis:1,expect:15,experi:15,explor:1,exponenti:15,express:5,extend:5,extrem:8,fall:16,famili:2,famou:15,fantast:[],featur:[7,13],feed:[2,10],fine:2,first:[5,10],fit:[1,8],forest:8,forward:[2,10],frank:[],freedom:1,frequentist:1,fridai:5,from:[3,8,10],further:3,gaussian:[13,15],gener:[7,15],geometr:[5,9],gini:7,good:1,grade:16,gradient:[2,5,8],handl:13,has:12,hessian:5,homework:5,hous:1,how:15,hyperparamet:2,hyperplan:6,id3:7,idea:[0,9],ideal:5,implement:[2,15],implic:3,improv:2,increment:9,index:7,inform:16,instal:12,instructor:16,interpret:[3,5,9],introduc:9,introduct:[1,4,12,13],invers:[3,13],iter:[5,8],its:15,jackknif:15,julia:[],jungl:8,jupyt:11,kera:2,kernel:[6,9],lagrangian:6,lasso:3,later:3,layer:[2,10],learn:[1,2,5,9,12],least:3,level:8,librari:12,likelihood:5,limit:[2,5,15],linear:[1,6,13],link:[3,9,14,17],logist:5,loss:5,machin:[1,5,6,12],main:15,make:[1,7,8],mani:[8,10],materi:14,math:3,mathemat:[3,6],matric:3,matrix:[2,3,5,9,10,13,15],matter:1,mean:[0,1,15],meet:[3,8,15],mercer:6,mersenn:15,method:[4,5,7,8,15],mlp:10,model:[1,2,10],moment:15,moon:[6,7],more:[0,5,13],multilay:10,multipl:2,multipli:6,name:15,need:[],network:[2,5,10],neural:[2,10],newton:5,non:6,normal:[1,2,15],norwai:12,notat:10,novemb:14,now:[2,7],nuclear:1,nueral:5,numba:[],number:[1,15],numer:15,numpi:13,numpython:0,observ:15,obtain:9,octob:14,one:[5,10],optim:[2,5,6,12],ordinari:3,organ:1,oslo:[12,17],other:[7,9,10,15],our:[0,1,3,5,9,15],outcom:12,output:15,overarch:[1,6,7],overview:8,own:[0,1,8,9],packag:13,panda:[],part:[5,12],pass:2,pca:9,pdf:15,perceptron:10,perform:[2,7],period:15,perspect:2,poisson:15,pre:2,preprocess:[],prerequisit:12,princip:9,pro:7,probabl:[3,15],problem:[2,5],procedur:7,process:2,program:5,propag:[2,10],properti:[3,15],pseudo:15,python:[0,1,7,12,13,15],quick:6,ran0:15,random:[8,9,15],raphson:5,read:7,recip:15,recurr:10,reduc:1,regress:[1,3,5,7,8],regular:2,relev:17,relu:2,remind:[4,5,6],requir:12,resampl:4,revisit:5,ridg:[3,5],rng:15,rule:10,sampl:[9,15],schedul:14,schemat:7,scikit:[1,2,5,9],select:15,semest:16,sensit:5,septemb:[5,14],set:[1,7,10],sgd:5,should:[2,15],simpl:[1,5,7,15],singl:8,singular:[3,9],situat:15,size:[],slightli:5,soft:6,softmax:2,softwar:[],solv:5,some:5,split:1,squar:[1,3,8],standard:[5,15],state:1,statist:[3,4,12,15],steepest:[5,8],step:[4,5],stk3155:12,stochast:[5,15],stop:5,supervis:2,support:6,svd:3,teach:[14,16],teacher:16,techniqu:9,technolog:12,tensorflow:2,test:[1,2],textbook:17,than:5,theorem:[3,6,9,10,15],theori:15,three:15,togeth:10,top:2,toss:15,toward:[0,9],tradeoff:4,train:[1,2],tree:[7,8],tune:2,two:[6,12],type:10,uncorrel:15,understand:[],uniform:15,univers:[10,12,17],use:[2,15],used:5,using:[1,5,15],valid:4,valu:[3,9,15],variabl:[5,15],varianc:[4,15],variou:[1,4],vector:[6,10,13],view:[1,8],visual:[2,7],wai:7,week:14,weekli:14,what:[1,15],when:5,which:[2,15],why:15,wisconsin:5,write:[0,9],xgboost:8,your:[1,8]}}) \ No newline at end of file +Search.setIndex({docnames:["Clustering","chapter1","chapter10","chapter2","chapter3","chapter4","chapter5","chapter6","chapter7","chapter8","chapter9","content","intro","linalg","schedule","statistics","teachers","textbooks"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":3,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":2,"sphinx.domains.rst":2,"sphinx.domains.std":1,"sphinx.ext.intersphinx":1,sphinx:56},filenames:["Clustering.ipynb","chapter1.ipynb","chapter10.ipynb","chapter2.ipynb","chapter3.ipynb","chapter4.ipynb","chapter5.ipynb","chapter6.ipynb","chapter7.ipynb","chapter8.ipynb","chapter9.ipynb","content.md","intro.md","linalg.ipynb","schedule.md","statistics.ipynb","teachers.md","textbooks.md"],objects:{},objnames:{},objtypes:{},terms:{"000":2,"000000":[1,3,9],"00000000e":3,"0001":2,"00015921913736794912":7,"00019998":3,"00024087":3,"00029012":3,"00034944":3,"00042089":3,"00050694":3,"00061058":3,"00073541":3,"00079968":3,"00088573":3,"00096314":3,"001":[2,5,6],"00106677":3,"00109273":[],"00115999":3,"00128479":3,"00139705":3,"00154733":3,"00168251":3,"00186347":3,"00200":6,"00202624":3,"00224413":3,"0024401":3,"00247083e":1,"00270244":3,"00293838":3,"0032542":3,"0032873138755776365":15,"00353823":3,"003704":1,"003717":1,"003759":1,"003774":1,"003788":1,"00391839":3,"004":3,"00426027":3,"00433417":9,"004456043243408203":0,"00445655":9,"00471782":3,"0049999999999999845":[],"004999999999999996":[],"004999999999999997":[],"0050000000000000044":[],"005000000000000007":1,"00512927":3,"0056799":3,"00617499":3,"006530396683779109":1,"00683748":3,"006968283076248407":[],"00727646693":1,"00728323e":[],"007315":1,"0074331":3,"007789":[],"007824":1,"007931713723267314":[],"008171076530362356":1,"00823002":3,"0086649156":1,"008675369724976777":3,"008885578722629236":[],"00890601":[],"00894639":3,"009154":1,"009163470508352228":3,"009735":9,"00973536":9,"009790":1,"009883615646716182":[],"009883615646716184":[],"009883615646716186":[],"009883615646716188":1,"00990475":3,"009911":1,"01057384067458835":[],"010679893512872646":[],"01076611":3,"0110":15,"011347":1,"01191824":3,"01210814993019007":[],"01295356":3,"0140617":[],"01433809":3,"01503674":[],"015037":[],"01514394":[],"015144":[],"01558197":3,"0169643":3,"0170259":[],"01724499":3,"018232":1,"01825571":[],"01867234e":15,"01873344":9,"01873869":3,"02073509":3,"02096350e":1,"021544346900318832":[],"02161783e":15,"021901":9,"02190139":9,"02252765":3,"02492265":3,"02493044e":[],"02568378":[],"025709":[1,9],"02625193":6,"02707227":3,"02730126581656065":15,"02881304":[],"029688":[],"02968834":[],"029733":1,"02994311":3,"03019554":[],"030196":[],"030670":[],"03067028":[],"03099776":3,"031":3,"03193199":[],"03251863":3,"03256632e":2,"03303359":[],"033037772005753835":[],"03308408":3,"034342":[],"0344707":[],"0354":4,"0359565":3,"03607832":[],"03652792":[],"036528":[],"03697069":[],"03707133":9,"03750367":13,"03787596148305236":1,"03791824e":15,"038300":[1,9],"03871832":[],"039039":3,"03929932":3,"04205220e":15,"04260073e":[],"04315108":3,"04330858":[],"04346721":3,"04421672e":[],"044334":1,"044402":1,"04455272":[],"044553":[],"04529095":[],"0458":7,"04648335":3,"04683565":3,"04757618e":[],"04829457939163212":15,"051115":[],"051649":[1,9],"0517473":3,"052645":[],"053053":[],"0530533":[],"05318162":[],"053182":[],"05364854":6,"05396545e":[],"054187":[],"054276":[],"054674":[],"055":[],"055085":[],"05531364":[],"055314":[],"055676":1,"055744":[],"055774":[],"055831":[],"055887":[],"055916":[],"05614483":3,"056326":[],"056329":9,"056484":[],"057035":[],"057088709963637":1,"057320":[],"057332":[],"05755374":[],"057571":[],"057654":[],"057919":[],"05802766":[],"058178":[],"058247":[],"058282":[],"058719":[],"058805":[],"058892":[],"059225":[],"059245":[],"059344":9,"059406":[],"059455":[],"059531":[],"059571":[],"059680303635344434":[],"059765":[],"059806":[],"059865":[],"059908":[],"059994":[],"060064":9,"060070":9,"060308":[],"060335":[],"06041458e":[],"060520":[],"060919":[],"061164":[],"061176":[],"061255":[],"06126507e":15,"061330":[],"061601":9,"061652":[],"061774":[],"061842":[],"061855":[],"061896":[],"06200174":3,"06209126":[],"062188":[],"062327":[],"062376":[],"062639":[],"062659":[],"062660":9,"062731":[],"062822":[],"062829":[],"062879":[],"062885":[],"062923":9,"062979":[],"063055":[],"063097":[],"063122":[],"063178":9,"063190":[],"063324":9,"063395":[],"063415":[],"063523":[],"063526":[],"063615":[],"063670":[],"063727":[],"063759":[],"063763":[],"063880":[],"063882":[],"063925":[],"063982":[],"064000":[],"064013":[],"064075":[],"064078":[],"064101":[],"06412177e":15,"064156":[],"064257":[],"064273":[],"064304":[],"064354":[],"064357":[],"064393":[],"064410":[],"064449":[],"064472":[],"064591":[],"064647":[],"064648":9,"064686":[],"064689":[],"064783":9,"064795":[],"06492086e":[],"064929":[],"064958":[],"064964":[],"064982":[],"065":[],"065003":[],"065086":[],"065110":9,"065144":[],"065151":[],"065188":[],"065200":[],"065202":[],"065204":[],"065334":3,"065355":[],"065408":3,"06540809":3,"065591":[],"065618":[],"065660":[],"065680":[],"065762":[],"065860":9,"065994":[],"066002":9,"06600226":9,"066030":9,"066047":[],"066074":9,"066092":[],"066116":[],"06619182206626131":[],"066249":[],"066272":[],"066294":[],"066312":9,"066389":[],"066403":9,"066487":[],"066629":[],"066683":9,"066705":[],"066761":[],"066788":[],"066835":[],"066837":[1,9],"066951":9,"066963":[],"067043":[],"067061":[],"067123":[],"067132":9,"067175":9,"067202":[],"06724062":3,"067326":3,"067355":[],"067440":9,"067504":[],"067531":[],"067577":[],"067601":[],"067610":[],"067667":9,"067685":9,"067711":[],"067718":[],"067769":[],"067777":[],"067794":[],"067844":[],"067865":[],"067889":[],"067892":[],"067985":[],"067990":[],"068056":[],"068073":[],"068165":[],"068185":[],"068213":[],"068244":[],"068276":[],"068293":[],"06834":4,"068351":9,"068385":[],"068427":[],"068460":[],"068491":[],"068514":[],"068593":[],"068640":[],"068654":[],"068667":[],"068673":[],"068686":[],"068697":9,"068717":[],"068818":9,"068821":[],"068828":[],"068838":9,"068840":[],"068873":[],"068896":[],"068920":[],"068928":[],"068945":[],"069005":[],"069028":[],"069071":[],"069081":[],"069098":[],"069195":9,"069243":9,"069282":[],"069369":[],"069371":3,"069413":9,"069496":[],"069503":[],"069510":[],"069516":[],"069528":[],"069542":[],"069584":1,"069635":[],"069641":9,"069775":9,"069814":[],"069848":[],"069853":[],"069859":[],"069872":[],"069890":[],"069902":[],"069988":9,"070043":1,"070103":[],"070129":[],"070131":[],"070138":[],"070166":[],"070182":[],"070188":[],"070254":[],"070337":9,"070456":[],"070473":[],"070517":[],"070548":9,"070653":[],"070762":9,"070764":[],"070791":9,"070812":[],"070815":[],"070867":9,"070889":[1,9],"070949":[],"070967":[],"071016":[],"071049":9,"07106781e":[],"071096":[],"07110274":13,"071156":[],"071248":[],"071268":[],"0713":1,"071313":[],"071325":[],"071359":[],"071435":9,"071447584032141":1,"07145103":9,"071453":3,"071481":[],"071502":[],"071547":9,"071549":[],"07155335":[],"071576":[],"071579":[],"07159175":[],"07160764":[],"071660":9,"071667":[],"071681":[],"071684":[],"071714":[],"071761":[],"071855":[],"071887":[],"071917":[],"072012":[],"072094":[],"072198":[],"072246":[],"072279":[],"072424":[],"072468":[],"072471":[],"072555":9,"072589":[],"072598":[],"072620":[],"072654":9,"072661":[],"072710":9,"072826":[],"072862":3,"072879":9,"072910":[],"073":[],"073020":[],"073035":[],"073096":9,"073310":9,"073333":[],"073352":[],"073371":9,"073372":[],"073483":[],"073546":3,"073567":[],"073591":[],"073598":[1,9],"073602":[],"073607":[],"073656":9,"073681":[],"073695":[],"073765":9,"073851":[],"073902":[],"073915":1,"073926":[],"073966":[],"074":[],"074027":[],"074036":3,"074037":3,"074044":[],"074067":[1,9],"074077":[],"074096":[],"074099":[],"074134":[],"074170":[],"074191":[],"074201":9,"074207":[],"07421084":3,"074230":[],"07423848370736122":[],"074248":[],"074285":[],"074286":[],"074331":9,"074432":[],"074434":[],"074453":[],"074488":[],"07456491":3,"074568":[],"074618":9,"074665":[],"074693":[],"074696":[],"074848":[],"074979":9,"074996":3,"075084":[],"075119":[],"075163":[],"075194":9,"075212":[],"07521771":[],"075218":[],"075261":[],"075266":[],"075289":[],"075296":9,"075309":[],"075526":9,"075542":3,"075545":9,"0755452":9,"075562":[],"075567":[],"075615":9,"075754":[],"075772":[],"07581375":[],"075814":[],"075818":[],"075821":[],"075865":[],"075975":[],"075993":9,"076026":[],"076062":[],"076166":[],"076169":[],"076189":[],"076213":3,"076251":[],"076283":[],"076320":[],"076367":9,"076398":[],"076408":[],"076410":9,"076466":9,"076469":[],"07670437":3,"076730":[],"076822":[],"076938":[1,9],"077005":3,"077010":9,"077046":9,"077115":3,"077143":9,"077160":[],"077163":3,"077212":[],"07729012413236423":9,"077303":[],"077305":9,"077338":[],"077339":[],"077386":[],"077407":[],"077416":[],"07744472306026946":7,"077469":[],"077533":[],"077623":[],"077771":9,"07777777777777778":2,"077891":[],"077927":9,"078049":[],"078076":[],"078099":9,"078103":9,"078197":[],"078244":9,"078279":[],"078311":[],"078342":3,"078388":[],"078390":3,"078399":[],"07839924":[],"078451":[],"078474":[],"078646":3,"078649":[],"078674":[],"078693":9,"078710":9,"078746":9,"078750":[],"078892":[],"078927":[],"078930":[],"078937":[],"078966":[],"079001":[],"079029":3,"079037":[],"079137":[],"079139":[],"079333":[],"079337":3,"079365":[],"07944154":13,"079597":3,"079624":[],"079643":3,"079785":9,"079848":9,"079956":[],"080024":9,"080084":1,"080101":3,"080115":[],"080182":3,"080223":[],"080249":3,"080264":9,"080325":9,"080336":[],"080341":3,"080381":[],"080387":[],"080398":[],"08043851":3,"080570":9,"080626":9,"080633":3,"080675":9,"080738":[],"080806":3,"080845":[],"080871":[],"081024":3,"081077":9,"081092":3,"081102":[],"081103":[],"081129":9,"081210":9,"081227":[],"081260":[],"081309":[],"081402":[],"081425":3,"081435":3,"081474":3,"081480":[],"081489":9,"081514":[],"081632":3,"08170444":[],"081744":3,"081778":3,"081816":[],"081886":[],"082032":3,"082091":[],"082211":9,"082225":9,"082239":3,"082247":9,"082272":[],"08248290e":[],"08251898":15,"082549":3,"082616":[],"082687":[],"08271336":[],"082866":[],"082990":[],"083015":3,"083128":[],"083138":3,"083177":3,"08318298e":2,"0832":[],"083221":3,"083246":[],"083295":9,"083376":[],"083384":3,"083495":3,"083527":1,"083577":[],"083618":[],"08362077210702115":[],"083622":3,"083657":[],"083658":9,"083728":[],"083775":[],"083808":3,"08380813":3,"083855":3,"083905":3,"084":4,"084092":[],"084103":[],"08415761":[],"084167":[],"084181":[],"084221":3,"084249":9,"084251":[],"084461":3,"084604":9,"084813":9,"084826":3,"084864":3,"084873":[],"084916":3,"084931":[],"08501147":[],"085121":9,"085253":3,"085259":[],"085285":[],"085289":[],"085510":3,"085544":3,"085597":[],"085598":3,"085610":3,"085646":9,"085676":[],"085748":3,"085763":3,"085811":3,"085928":[],"085951":[],"086052":3,"08611111111111111":2,"086116":[],"086325":[],"086430":3,"086465":[],"086561":3,"086592":[],"08665060086846632":[],"086652":9,"086757":3,"086974":9,"086985":3,"087202":9,"087242":3,"087334":3,"087482":[],"087526":3,"087544":3,"087563":9,"088137":3,"0881981":3,"088209":3,"088258":3,"08839750e":[],"088414":3,"088484":3,"088556":3,"088660":[],"08871404":3,"088871":[],"08888888888888889":2,"089039":3,"089300":[],"089326":3,"089352":3,"089425":9,"089437":3,"089504":[],"089540":3,"089693":3,"089881":3,"089916":[],"089931":[],"090274":9,"090355":3,"090817":3,"09103481e":15,"091159":[],"091318":3,"091535":3,"091578":3,"091659":3,"09166666666666666":2,"0917":7,"091979":[],"092288":3,"092375":3,"092659":3,"092932":3,"093236":3,"093248":3,"093328":3,"093573":3,"093755":9,"093928":3,"094":4,"094513":3,"09456011349477329":3,"094774":3,"09487315108590391":9,"095173":[],"095472":[],"095941":3,"09609807":3,"096406":3,"096639":3,"096726":[],"09672604":[],"096763":3,"09678277e":15,"096802":3,"097346":3,"09741585e":[],"097452":[],"09903804":6,"099763":[],"09976319":[],"0x7febdb109490":5,"0x7febdb6780d0":5,"100":[0,1,2,3,4,5,6,7,8,9,13,15,16],"1000":[0,1,2,5,6,9,12,15],"10000":[0,3,4,8,9,15],"100000":6,"10001":8,"1001":15,"1002":15,"1003":15,"1005":15,"1007125":[],"1009":15,"10094646e":[],"1011":15,"1013":15,"10131725":[],"1013904243":15,"1015":15,"102":1,"1022964509394572":2,"1023":15,"1026":15,"1027":15,"103":2,"1030":15,"10307631":[],"10327559":13,"1037":15,"10378326e":2,"1038":15,"1040":15,"10405456":9,"104411":1,"1047":15,"105137868830763":15,"10555555555555556":2,"10589577":3,"106095":[1,9],"108":1,"10806972":[],"108070":[],"10896672e":15,"10898112e":[],"10th":7,"10x":1,"110":1,"1100":15,"1101":15,"11022302e":3,"111":[2,5,10],"1111111111111111":2,"112283":1,"112383":[1,9],"11304709e":15,"11388888888888889":2,"11456076e":[],"11462415":3,"11507992e":2,"11666666666666667":2,"117":6,"118318":1,"11944444444444445":2,"12002944":[],"1203284":6,"121":[6,7,8],"122":[6,7,8],"12222222222222222":2,"12283463":[],"123190":1,"123459876":15,"12366979e":15,"124":1,"12497018e":[],"12618549":3,"127773":15,"12777777777777777":2,"12908963":1,"1298":7,"12adb44b1c20":[],"13055555555555556":2,"13060772":[],"13162821e":[],"13209041":[],"13266452e":[],"133":5,"13328820e":[],"134102":3,"13410232":3,"13457922":[],"13519106":1,"13579199e":[],"1361111111111111":2,"13646574":3,"136687":[],"137268":[],"137400784702912":1,"137652":[1,9],"13803928":[],"138472":4,"13865173":3,"138775":[1,9],"13891144e":[],"139462":[],"139475":[],"1404":1,"143":5,"1437":2,"1440501043841336":2,"1445":4,"14459063":[],"1446":4,"1447":4,"1448":4,"1449":4,"14662074":[],"14722222222222223":2,"147400":[1,9],"147420":[1,9],"147722":[],"148":[],"149076":[],"149366":[],"149667":[],"149739":4,"149832":[],"149894":[],"149903":[],"14g":4,"150":6,"15005476":3,"15024669":[],"150306":4,"151":[],"152636":[],"1527777777777778":2,"153036":1,"153106":1,"15324631":[],"15332528e":[],"154720":1,"15475232":[],"154911":1,"155491":1,"155687":1,"155883":1,"156":1,"157":1,"15751596":[],"158":1,"159":1,"15979239e":15,"15g":4,"160":1,"16111111111111112":2,"16211139":3,"16342407":3,"16496581e":[],"16553696":[],"16637855e":15,"16666666666666666":2,"16807":15,"16831012":[],"16861772":[],"168618":[],"16b8e3cda33a":2,"17117385":[],"17174962e":2,"17234827e":[],"17248441e":[],"17385778e":[],"174497":[],"17449733":[],"175300":1,"1758506":[],"17654307":[],"17707436":[],"17777777777777778":2,"17801022":3,"17917768":3,"17949575":3,"17953942":9,"1797":2,"180092462880674":[],"18029127":3,"18063928":[],"18220995":[],"18224665":[],"1831277634928002":3,"18333333333333332":2,"18404906e":[],"184519":1,"1856411":[],"18611111111111112":2,"18620601e":[],"18673098":9,"189496":1,"18954529":[],"189622":1,"18968431e":15,"1914224774238273":[],"19207979":3,"1940":1,"1943":10,"197":[],"1970":13,"1973":7,"197370":[1,9],"19742904e":[],"1979":4,"1989":15,"19937":15,"19955871":[],"199559":[],"19972087e":15,"1_1":10,"1_2":10,"1_3":10,"1cm":[1,6,8,15],"1e10":0,"200":[1,6,7,8,15],"2000":1,"200000":1,"2004":5,"2006":17,"2010":2,"2011":2,"2015":2,"2016":1,"2018":[1,4],"201841":[],"20184113":[],"2021":0,"20277777777777778":2,"203757":[],"20375701":[],"205466494327873":[],"20632410e":[],"207545":[1,9],"20772452":[],"20819609e":[],"20833333333333334":2,"20843563e":[],"20906175e":15,"210340":[1,9],"21058097":3,"212":[],"21208310e":[],"212327334149492":[],"2125":[],"2126":[],"2127":[],"2128":[],"2129":[],"213":[],"213103":[1,9],"21347282":[],"213743":[1,9],"214":[],"2147483647":15,"215":[],"216":[],"21623344":[],"216290":[1,9],"216683":[1,9],"22044605e":3,"22076362":[],"221":6,"221180":1,"22209371e":[],"222400":1,"22443803":[],"22527008e":15,"22690428":3,"22717936e":15,"22847924":3,"22999784":[],"229998":[],"23117916e":15,"23167717":3,"232435":1,"23333333333333334":2,"2335879":[],"23382086e":[],"23438810e":[],"241262":[],"24126232":[],"2433e":3,"24340751":[],"24444444444444444":2,"246138":3,"24613822":3,"24829908":3,"250":[5,7],"25000":1,"250000":1,"250154":1,"251559":[],"2515593":[],"251879":1,"25190225e":[],"25226753e":15,"25240108":[],"252436":1,"252639":3,"25263916":3,"253":[],"25303483":[],"253775":1,"254":[],"254509":[],"25450941":[],"25457052":[],"255":[],"255001":1,"2551":1,"256":[],"256962":1,"257":[],"2571699":[],"25726439e":[],"2572e3a4b38d":2,"25803281":[],"259107":[],"25910749":[],"259153":[1,9],"259815":[],"25981533":[],"260227742627244":[],"2627588":13,"26301436":3,"26381865":[],"263819":[],"264":1,"26412372":[],"265":1,"2653678":[],"266":1,"26666667":5,"267":1,"26710969":3,"26780278":3,"2683":4,"2684":4,"2685":4,"2686":4,"2687":4,"26890510e":[],"269":1,"26974938e":[],"270":1,"27262964":[],"27296891e":[],"276263":[1,9],"27753165e":15,"27852549":[],"27924636":3,"27n_":15,"280647":[1,9],"28166741":[],"282":[],"282727":[1,9],"28310983":[],"283110":[],"2836":15,"28475098":6,"2861":15,"2873":7,"2882":15,"2886":15,"2890":1,"2892":15,"28971976":9,"289720":9,"290":[],"291":[],"2915":15,"291614":[],"29167186":3,"292":[],"293":[],"2931":[],"294":[],"295656121491569":[],"296247":1,"2968":[],"297219777724628":9,"297260":1,"2980":[],"298273":1,"2983233":13,"298375":1,"2990":[],"299444":9,"29944428":9,"299748":1,"2_1":10,"2_2":10,"2_3":10,"2_i":10,"2_m":15,"2_t":5,"2_x":15,"2cm":6,"2ff97f4bf03b":15,"2nd":7,"2x_ix_jy_iy_j":6,"2x_j":6,"2y_i":8,"2y_j":6,"30000":1,"300162456113691":[],"30119421":6,"3041320306136366":3,"30626706":[],"306854":[],"30685416":[],"30879705":[],"30968181":[],"310811":3,"31081134":3,"31109204e":[],"31318084":3,"31457796":3,"3155":[3,4],"3156929654100207":7,"31608475e":[],"31718909":9,"31730641":[],"317367":9,"31853484":[],"31896852":6,"3200":2,"32047562":1,"323291478597321":15,"3250":2,"32521615e":15,"326238":1,"327631":1,"32938847":13,"32945844e":15,"3304":1,"33066907e":3,"3310":1,"3317":1,"331939":1,"333":5,"333333":1,"33333333":5,"3338":1,"3344":1,"33443859e":[],"33544681":[],"33569998e":[],"33861512":[],"338869":[],"33886902":[],"339535706819584":15,"33953571":15,"340782":[1,9],"34114547":3,"34114641e":[],"34172919":[],"342680":1,"3436":1,"3437":1,"344172":1,"34569596":3,"346433":9,"34643337":9,"34902789e":15,"350387":[],"35038714":[],"351636":[1,9],"35176067":[],"35182854":3,"35216172e":[],"35367281e":[],"356399":1,"357508":1,"358869339268145":15,"35886934":15,"359":3,"359640894899012":[],"360":2,"360688":1,"361556":1,"3621311":3,"36436520e":[],"36468301":[],"3655222":3,"369139":[1,9],"36941772":[],"369418":[],"37186301":[],"372889":[],"3728892":[],"37416969":9,"374170":9,"37738324":[],"38207279e":[],"38216436":[],"38629436":13,"38777878e":3,"38892672":3,"388927":3,"3893239":[],"389324":[],"38937995e":15,"38986237":[],"39579407":3,"396740":9,"39674043":9,"39706038":3,"397700":[1,9],"39792608e":[],"399836":1,"3cd19a0768e1":[],"4000":17,"401842":[1,9],"40212127":[],"404":1,"40425078e":15,"405890":[1,9],"40702":[],"40708470e":[],"40859":[],"4087793":3,"40902095":[],"40968888":3,"409689":3,"412374":[],"41237437":[],"4140e":3,"41433969":3,"41511965e":2,"415634483874318":[],"416694683938511":9,"4171578884124756":0,"41754964":[],"41770932":[],"418506":[1,9],"41876428e":[],"4200e":3,"4201e":3,"4203e":3,"42441033":3,"426":5,"42847770e":15,"42937310e":15,"43054282":3,"4332e":3,"435163":1,"435490":[],"43549028":[],"43766686":9,"438136":1,"44089210e":3,"442600":[1,9],"443217":1,"44395541":[],"44625466e":15,"44970586e":2,"45013332e":[],"45019484":[],"450257":[1,9],"4557763":9,"455947":1,"458027":1,"458078":[1,9],"45937170e":[],"45960079":3,"46016532e":[],"461":15,"461838":[],"46183815":[],"462":5,"46323168e":15,"4632e":3,"46415888336127775":[],"46423858e":[],"465564":[],"46556436":[],"466":15,"46675058":[],"46914544e":[],"46929603e":[],"469730":[],"46973015":[],"47042744":3,"470714":1,"47079457e":15,"47125748":3,"47132891":3,"472652":[],"47265243":[],"47441766":[],"47566390e":15,"47654764e":[],"47815203":9,"47862383":[],"478624":[],"48019541":[],"48154187202453613":0,"48257387":16,"48471852e":[],"48476997":9,"48608063e":[],"48994188":3,"491837":[],"4918372":[],"493230":3,"49323032":3,"4940954":1,"497221":[],"49722108":[],"49841285":3,"498413":3,"49865673":[],"4990":15,"4992":15,"4997":15,"4c4c7f":[7,8],"4y_i":8,"500":[2,4,5,7,8,15],"500000":1,"50000455":3,"50000553":3,"50000718":3,"50000855":3,"50000969":3,"50001063":3,"50001142":3,"50001207":3,"50001261":3,"50001306":3,"50001343":3,"50001374":3,"500014":3,"50001414":3,"50001422":3,"50001439":3,"50001454":3,"50001466":3,"50001476":3,"50001485":3,"50001492":3,"50001498":3,"50001502":3,"50001506":3,"5000151":3,"50001512":3,"50001515":3,"50001517":3,"50001518":3,"50001519":3,"50001521":3,"50001522":3,"50001523":3,"50001524":3,"50001525":3,"5018":15,"50321091":3,"50394742":[],"506":1,"507d50":[7,8],"5098779":[],"509878":[],"50j":5,"50x10":2,"510":2,"5120":0,"512132":1,"51257863e":[],"51345668e":[],"51363731e":[],"514219":1,"51523276e":[],"515768":[],"51576807":[],"51707172":[],"517072":[],"51893804e":[],"51943726":9,"519842":1,"52158335":[],"5222222222222223":2,"52570079":[],"525701":[],"526744":[1,9],"52687171":[],"526872":[],"52722156":[],"52874252":3,"52d2c51caad1":[],"5303329":9,"5305555555555556":2,"531280":1,"535457":[],"53545715":[],"5378811":9,"539261":[1,9],"54039921":3,"54096582":[],"540966":[],"54121682":[],"541217":[],"54152940e":[],"541605":1,"54237024":15,"543169":3,"54316925":3,"544439":1,"546972":[],"54697204":[],"54702088e":[],"54710530e":1,"55111512e":3,"55138385":9,"551384":9,"55280484":[],"5555555555555556":2,"55707065":[],"557071":[],"557795":[1,9],"55795935":[],"55854694":9,"56033697":3,"56198284":3,"56216797e":[],"564374":[1,9],"56536":1,"569":2,"56939714":3,"57051369":[],"571":3,"574465":[1,9],"5755035":[],"57781668":[],"57871326":[],"581766":9,"58176612":9,"582":[1,2],"58228342e":15,"58239999":[],"582400":[],"583595":1,"584804":1,"58521266":9,"585213":9,"58596975":[],"585970":[],"587401":1,"58836420e":15,"5888888888888889":2,"59007674e":15,"59304755e":[],"5944444444444444":2,"59480085":[],"5cm":15,"60122668e":15,"60293962":3,"60394236e":[],"60420593":3,"60673226":9,"60949193":[],"609492":[],"60999846":[],"61069091e":15,"6111111111111112":2,"61124978":[],"61234223":[],"612939":1,"613579":1,"614808":[1,9],"61505887e":[],"61745046e":[],"618":1,"6183694":[],"61869821":[],"618982":1,"622539":9,"62253933":9,"622625":[],"62262506":[],"6226921":[],"62316154e":15,"62359224e":15,"62373464":9,"625":5,"62783293":[],"62856593":[],"62894215":3,"629961":1,"6300745149331701":1,"63315151":[],"63339159":[],"63374631":[],"63437572":[],"63442451e":[],"63498144":3,"636323":9,"63632311":9,"63680118":[],"637129335071195":1,"63993205e":[],"64012627":3,"640782":3,"64078247":3,"64166831e":[],"64447921":[],"64580686":[],"646283":[1,9],"64669382":[],"646694":[],"647473":[1,9],"64857826e":[],"649382":[1,9],"64x50":2,"650024":[],"65002433":[],"6510573774179256":15,"65105738":15,"65238878":13,"65245958":[],"653095390463358":[],"653702":13,"653725417896576":1,"6544e":3,"65482578":[],"65572035":[],"65599927":[],"65766387":[],"65885453":3,"65933852":[],"65939208e":[],"6600855222624895":3,"66020213e":[],"6614":[],"66183486":9,"661835":9,"6628996975186952":1,"66302359":[],"66383151":[],"66677842":[],"66800261":[],"668172":1,"66880047":[],"67006792":[],"67060602":[],"671089":1,"6714":[],"67171347e":15,"672721":1,"67279536":[],"67303655":9,"673037":9,"67407338e":15,"67450955":[],"67708423":[],"677149":[],"67714918":[],"6796265324852733":[],"68002363":[],"680024":[],"68034946e":[],"6813":4,"6814":4,"6815":4,"6816":4,"6817":4,"6818813252071303":15,"68188133":15,"68192193":3,"68316185":[],"68342382":[],"68542204":3,"68616263":13,"68729414":[],"68937695":[],"689519":[1,9],"690617":1,"69069n_":15,"693361":1,"69347005":[],"693850":9,"69385025":9,"69519693":[],"695197":[],"69573183":[],"69695259":3,"69843037":[],"6996584":[],"69981195e":15,"6999536":9,"6ea927cc6e88":2,"6n_":15,"70183798":3,"7022283":[],"70234019":[],"70415861":[],"70523024e":[],"70589906":[],"7070e":3,"70710678":3,"70790937":[],"70832814":3,"70885528e":[],"70886748":[],"70946493e":[],"712018":[1,9],"712199063818309":[],"71281409":[],"71351486":[],"71442781":13,"71606852":[],"7162":[],"71669651":[],"7172":[],"72108703":[],"72174172":9,"72218808":[],"72312577":[],"7236674":3,"7240496":[],"725394195434945":[],"72780613e":15,"72859758":3,"72879865e":[],"72981762":6,"73091052e":[],"731000":1,"733096":1,"73453972":[],"7371165871823337":[],"737349":[],"73734906":[],"73921714":[],"74081822":6,"74107697":9,"74143127e":[],"7432283":[],"74382593":[],"74495014":[],"745136489050356":[],"74840212":3,"74845978":[],"74921867":[],"749765":1,"750445":1,"751699":[1,9],"75170092":3,"75195757":[],"7522047280566193":[],"75382481":[],"75524378":[],"75629493":[],"756352":1,"75841112":[],"75932862":[],"76172241e":[],"762":[5,9],"7621419":[],"76290332":[],"763880":[],"76388013":[],"76497666":[],"765":5,"76504618":[],"76570177":[],"76648901e":[],"76936315":3,"7693978131030923":9,"7701384":3,"77152076":3,"77156117e":3,"7718":7,"772b904ae9cb":4,"77317984":[],"7733":[],"77350269e":[],"774300":1,"776223":[],"77622336":[],"77661393e":15,"77714169":6,"78080633":[],"78082":[],"78195":[],"78857629":[],"78941903":3,"78944806":[],"7899453":[],"79111643":3,"7925146":3,"792515":3,"79295029e":15,"793167":1,"79326583e":15,"79328828":[],"793701":1,"794282":[1,9],"79459035":[],"79648291":[],"79902342":[],"7c394b1e8b71":7,"7d7d58":[7,8],"800":5,"80004454e":[],"800266":[],"80026635":[],"80121":[],"80469739":3,"8055555555555556":2,"81160425":3,"81620806":[],"81633628":9,"816454":1,"816847":1,"81712976":3,"817130":3,"81781888":9,"819202":[],"81920231":[],"82198978":3,"8265786":3,"827265":1,"82889306e":15,"8305555555555556":2,"83425361":9,"834254":9,"83512277":3,"83614019":[],"8388888888888889":2,"839313":3,"8393131":3,"83935285":[],"84008474":3,"840085":3,"84087101":[],"842":[],"842436":1,"84251485":[],"842515":[],"8429949116841184":[],"84355903e":2,"84372853":[],"84443254e":2,"84658093e":15,"8479552268981934":0,"84860939":[],"849766":[],"84976606":[],"84994524":3,"85058354":[],"850584":[],"8520127":[],"85278450e":[],"85396354":[],"85450859":9,"85463934e":[],"85497163e":[],"85546305e":[],"85601992":3,"861":1,"86117291":3,"86134827":3,"86145244":9,"8638888888888889":2,"86436607":[],"86574276":[],"8666666666666667":2,"86692943":[],"87030083":[],"870301":[],"8718475896381779":15,"87184759":15,"8722222222222222":2,"87381451":3,"875":2,"87761937":[],"8777777777777778":2,"878843":3,"8788431":3,"87972591":[],"8802":[],"88046261":3,"88046462":[],"8805555555555555":2,"881323":[],"88297395":[],"88336879":3,"88559559":[],"8867467323038865":[],"88693966e":15,"88712946":[],"8888888888888888":2,"8901":[],"8914984":[],"8921171964770647":9,"8923":[],"89288636":9,"89383322":[],"89410423":3,"8944444444444445":2,"89704131":[],"89742056":[],"897421":[],"898500":9,"89850037":9,"89928965":[],"89942598":[],"89975818":[],"89992521":1,"8x8":2,"90075537":3,"90220243":3,"90233874":[],"90266948":3,"90316476":[],"9040":7,"9055555555555556":2,"90618734e":15,"907307":9,"90730735":9,"90825063":[],"9096":[],"91012519e":[],"9111111111111111":2,"91128596":3,"91145266e":15,"91417278":[],"91549644":1,"9166666666666666":2,"916978":[],"91697817":[],"9171356":[],"91760278":3,"91812702":3,"918992":1,"9196":[],"9222222222222223":2,"922312":[],"92231203":[],"92280994":[],"924018":1,"925":2,"92507116e":2,"92578916":3,"92605247":[],"92645039":9,"92646965":9,"926470":9,"92732185e":[],"9277777777777778":2,"9279671770201344":15,"92814088":3,"928141":3,"92857143":5,"92919670e":15,"9305555555555556":2,"931":1,"93100040e":15,"93155188":3,"93158979":3,"932734":[],"93273404":[],"933":3,"9354":[],"9361111111111111":2,"93679587":3,"936796":3,"937":15,"937082":1,"93799826":3,"938":15,"9388888888888889":2,"939":[1,15],"94019247e":[],"94034531":[],"942726":3,"94284104":3,"94320205":3,"9444444444444444":2,"94536341":15,"94591015":13,"94639099":9,"946893955211749":[],"947543":9,"9482527":3,"948729":[],"95014575":[],"95231424":3,"9527777777777777":2,"95284275":3,"95351665":3,"954":15,"95443703e":[],"9547578478889096":1,"95517094":[],"955171":[],"9555555555555556":2,"95628168":[],"956282":[],"956563":[1,9],"95684892":3,"958228616652075":3,"9583333333333334":2,"959247":[],"960":15,"96024953":3,"96084663":3,"961":15,"96104648":[],"9611111111111111":2,"962":15,"963499":9,"96349948":9,"9640435":3,"965548":1,"965885569080809":[],"96688672":3,"9674916":3,"967809":[1,9],"96793117e":[],"97005689":3,"97101567e":15,"9722222222222222":2,"97243128":3,"97300836":3,"975":2,"97507735":3,"9756404":[],"976":6,"97622676":[],"976227":[],"9765":[],"9769":[],"97705827":3,"97723801":[],"97758848":3,"9777777777777777":2,"9780387310732":17,"9780387848570":17,"9781492032632":17,"97926491":3,"9804422":[],"98046438":[],"9805555555555555":2,"98091621":3,"981321":1,"98139097":3,"98275501":3,"98314755":[],"983148":[],"983310":1,"98413059":3,"98452685":[],"98454786":3,"985":15,"98526763":[],"98528992":[],"98531221":[],"98566191":3,"986":15,"9860803":[],"98609175":15,"986091753050161":15,"9861111111111112":2,"98661465":[],"98680716":3,"98716878":3,"9877742":[],"987902":[],"98808176":3,"988663":[],"9888888888888889":2,"98892195e":15,"98893512":[],"989":15,"9890348":3,"98927731":[],"9893447":3,"9898ff":[7,8],"99009525":3,"99009739":[],"9901168":[],"99013921":[],"99016161":[],"99018401":[],"99043999":[],"99088801":3,"9909252":[],"991":15,"991072":9,"99107239":9,"99115119":3,"99126104":[],"99133007":[],"99160404":[],"99176998":3,"99190487":[],"99194716":[],"992":15,"99218987":[],"99219378":[],"99242605":[],"99242921":3,"99248001":[],"99265097":3,"99273355":[],"99276945":[],"993":15,"99305549":[],"99305802":[],"99311297":[],"99316252":3,"9933":4,"99346398":[],"99363129":[],"99363383":[],"99371056":3,"99389612":3,"99393624":[],"994":4,"99400444":[],"9941":4,"99418903":9,"99420743":[],"99420997":[],"99428016":[],"9943201":3,"9945452":[],"99462421":[],"99473581":[],"9947756":3,"99478645":[],"99478898":[],"99492986":3,"99498985":[],"995":4,"99501236":[],"99503487":[],"99505739":[],"9950799":[],"99527696":[],"99528218":3,"9953048353087299":[],"99536326":[],"9953658":[],"99539415":3,"99544872":[],"9954538761021741":[],"99566069":3,"995774":3,"9957744":3,"99578809":3,"99579317":[],"99581841":[],"99594294":[],"996":[3,4],"99600927":[],"99608161":3,"99620088364924":1,"99636015":[],"9963961":3,"99650061":3,"99652042":[],"99652296":[],"99655111":[],"9966744029509663":[],"9967292151090247":[],"9967458":3,"9968966158779216":1,"9969":[],"99696351":[],"997":4,"99700706":3,"99709215":3,"99709325":[],"99710078":[],"99723611":[],"99728435":[],"99729756":3,"99730848":[],"99751458":3,"99758326":3,"99763569":[],"99767893":[],"99775587":3,"99782689":[],"99793613":3,"99799099":3,"998":4,"99813653":3,"99817842":[],"99825997":[],"99828624":3,"9983295":3,"99836973":[],"99845267":3,"9984806":[],"998577":3,"99861053":3,"99871521":3,"99881845":3,"99883879":[],"99884384":3,"99891285":[],"99893323":3,"999":[7,15],"99901896":3,"99903755":3,"99911427":3,"99918546":3,"99919837":3,"99926459":3,"9993237":3,"99933188":3,"99938942":3,"9994385":3,"99944272":3,"99945628":[],"99949306":3,"99953381":3,"99953475":3,"99957911":3,"99961294":3,"99965056":3,"99967865":3,"99970988":3,"9997332":3,"99975913":3,"99977849":3,"99980002":3,"9998161":3,"99984732":3,"99987324":3,"99988687":[],"99989476":3,"9999095":[],"99991263":3,"99992746":3,"99993212":[],"99993978":3,"99995":3,"99995475":[],"99995818594196":[],"999974281880048":[],"99997737":[],"9999787681537219":[],"9999794306626945":[],"9999822527140678":[],"9999850282256434":1,"9999864543345858":[],"9999868619217517":[],"9999869956119286":[],"9999878260589065":[],"9999887274726137":1,"9999910208315801":[],"9b9cf4fa1a95":[],"\u00f8yvind":16,"abstract":2,"break":[0,1,9],"byte":13,"case":[1,2,3,4,5,9,10,12,13,14],"catch":1,"char":15,"class":[1,2,4,5,6,7,9,10,15],"const":15,"default":[1,2,5,13],"ekstr\u00f8m":16,"export":7,"f\u00f8470":16,"final":[0,1,2,3,4,5,6,7,8,9,14,15,16],"float":[0,1,3,5,7,9,13,15],"function":[0,3,4,7,12,13],"import":[0,1,2,4,5,6,7,8,9,10],"int":[0,1,2,3,5,9,13,15],"long":[1,2,5,10,15],"new":[0,1,2,3,4,5,6,7,8,9,13,15],"null":15,"public":[1,12],"return":[0,1,2,3,4,5,6,7,9,13,15],"s\u00f8rli":16,"sch\u00f8yen":16,"short":[0,3,11],"steinsv\u00e5g":16,"super":3,"switch":[0,1],"throw":15,"true":[0,1,2,3,4,5,6,7,8,10,15],"try":[0,1,2,3,5,6,7,8,9,12,13,15],"var":[3,4,8,9,15],"while":[0,1,2,3,4,5,6,7,9,10,15],AGE:1,Adding:2,Age:5,And:[0,1,4,5,7,12,15],Are:9,Being:5,But:[0,1,2,3,4,7,8,15],CAS:[],DIS:1,Doing:[3,5,8],EoS:[1,4],FYS:14,For:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],Going:2,Ising:[3,10],Its:[2,9],MDS:9,NNs:10,N_s:6,Not:[1,2,3,4],OLS:[1,4],One:[1,2,3,4,5,6,9,10,15],PCs:[9,12],RMS:15,Such:[4,10,15],That:[0,1,3,5,8,9,10,15],The:[8,13,14,16,17],Then:[0,1,2,3,4,5,6,7,8,9,10,13,15],There:[0,1,3,6,7,9,10,11,13,14,15,16],These:[0,1,3,5,6,7,8,9,10,13,15],Use:[1,7],Useful:[3,4,13],Using:[1,3,4,6,8,10,13],With:[1,3,4,6,7,8,9,10,13,15],__class__:8,__doc__:4,__future__:[6,7],__getattr__:[],__init__:[2,4],__mosek:3,__name__:[4,8],_auto10:10,_auto1:[3,5,10,13,15],_auto2:[3,5,10,13,15],_auto3:[3,5,10,13],_auto4:[5,10,13],_auto5:[5,10,13],_auto6:[5,10,13],_auto7:[10,13],_auto8:10,_auto9:10,_ax:4,_base:6,_build:[12,17],_check_optimize_result:[5,9],_compon:9,_datafram:[],_depth:7,_fraction:7,_lambda:[],_leaf:7,_logist:[5,9],_make_index:[],_multilayer_perceptron:[1,2],_node:7,_num_sampl:[],_ratio:9,_sampl:7,_split:7,_varianc:9,_weight:7,a0faa0:[7,8],a77d5ac269b2:[],a_0:1,a_1a:1,a_2a:1,a_3:1,a_3a:1,a_4:1,a_4a:1,a_h:2,a_i:[1,2,10],a_j:[2,10],a_k:[2,10],aaron:17,ab_channel:12,abandon:2,abbrevi:14,abid:15,abil:[1,8],abl:[2,3,5,8,10,15],abort:15,about:[0,1,2,3,4,5,6,7,8,9,10,12,13,17],abov:[0,1,2,3,4,5,6,7,8,9,10,13,15],abovement:4,abs:[0,1,3,5],abscissa:5,absolut:[1,3,4,5],acccess:[],acceler:5,accept:[1,7],access:[1,9,15],accid:4,accmod:3,accompani:1,accomplish:[5,6,7],accord:[0,1,2,3,4,5,7,10,15],accordingli:9,account:[1,3,5,15],accumul:[10,15],accur:[1,4,5,8,15],accuraci:[1,2,3,5,7,8,9,10],accuracy_scor:[1,2,8],accuracy_score_numpi:2,achiev:[1,2,3,4,6,10,13],aco:15,acquaint:12,acquir:[2,12],acr:1,across:[2,4,7,12],act:[2,13],action:15,activ:[0,1,7,14],actual:[1,2,3,4,6,9,13],ada_clf:8,adaboostclassifi:8,adadelta:5,adagrad:5,adam:2,adapt:[1,4,5,17],add:[1,2,3,4,6,8,9,10,15],add_subplot:[0,2,5,10],added:[1,2,3,5,6,13],addendum:3,adding:[0,2,4,13],addit:[0,1,3,4,5,6,7,8,10,12,13,15,16,17],addition:[5,10],address:[2,5,7,9,17],adjac:10,adjoint:3,adjust:[1,5,10],admir:1,advanc:[4,10,17],advantag:[2,3,4,5,8,13],afecionado:[],affect:[],affin:[1,6,9],aficionado:[],aforement:0,african:1,after:[0,1,2,3,4,5,7,9,10,12,13,15],afterward:1,again:[0,1,2,3,4,5,6,8,9,10,15],against:[2,5,8],age:[1,5],agegroup:5,agegroupmean:5,aggreg:[7,8],agorithm:8,agre:[3,15],ahead:7,aid:9,aim:[0,1,2,4,5,9,12,13],ainv:3,aka:3,alarm:3,albeit:0,algebra:[1,3,5,12,14],algo:15,algorithm:[1,2,3,4,5,6,12,13,14,15,17],align:[1,3,4,5,6,15],all:[0,1,2,3,4,5,7,8,9,10,12,13,14,15,16,17],allevi:[2,5],alloc:13,allow:[1,2,3,4,5,6,8,12,13],almost:[1,2,4,5,6,9,15],alon:7,along:[0,3,4,7,8,9,12,13],alpha:[0,1,2,3,4,5,6,7,8,15],alpha_:8,alpha_i:5,alpha_k:5,alpha_m:8,alpha_opt:5,alreadi:[3,8,10,12,13,15],also:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],alter:[2,15],altern:[1,2,3,5,6,7,9,13],although:[2,3,4,5,6,8],alwai:[0,1,3,4,5,10,15],ame2016:1,american:1,among:[1,3,7,8,10],amongst:3,amount:[0,1,2,4,6,8,12],an_:15,anaconda3:[1,2,4,5,6,9],anaconda:[1,2,12],analog:5,analys:[4,15],analysi:[2,4,5,13,14,17],analyt:[1,3,4,5,10,12],analyz:[1,2,3,15],andrew:2,angl:[1,7,15],ani:[0,1,2,3,4,6,7,8,10,15],anim:10,ann:10,annot:[1,2,5,6],anoth:[1,2,3,4,5,6,8,9,10,13,15],ans:15,ansatz:1,answer:[1,2,3,4,13],anymor:[2,6],anyon:6,anyth:[2,15],anytim:16,apach:2,apart:[5,9],api:[2,12],appear:[1,2,5,13,15],append:[2,4,5,6,7,15],appendcon:3,appendvar:3,appli:[1,2,4,5,6,7,8,9,10,15,17],applic:[1,2,3,4,5,7,10,15,17],approach:[2,3,4,5,7,8,9,10,12,15,17],appropri:[4,5,7,10,12,15],approx:[1,5,8,9,15],approxim:[1,2,3,4,5,8,9,15],apt:[1,12],aptli:0,aragorn:[],arang:[2,4,5,7,8,10],arbitrari:[2,5,6,10,15],arbitrarili:[1,2,9],architectur:[10,17],area:[1,17],arg:4,argc:15,argmax:[2,9],argmin:[0,8],argsort:9,argu:[2,5],arguabl:0,argument:[1,3,5,9,10,15],argv:15,aris:[1,4,5,10,15],arithmet:[1,5,13],arma:15,armadillo:[13,15],around:[1,2,3,4,9,15],arrai:[0,1,2,3,4,5,6,7,9,10,12,15],arraybox:5,arriv:[1,7,9,13,15],arrow:10,arrowprop:6,art3d:5,art:[1,2,12],articl:[0,1,4,8,15],artifici:[1,5,10,17],artificialneuron:10,artist:4,arug:5,asarrai:[1,7],asc:3,ascii:15,ask:[3,4,9,10],aspect:[1,12],assembl:1,assess:[1,4],assign:[0,1,5,6,7,10,14,17],assign_points_to_clust:0,associ:[0,1,4,7,10,15],assum:[0,1,2,3,4,5,6,7,8,9,10,13,15],assumpt:[1,3,4,7,9,15],ast:[1,3,4],astyp:[7,8],asymmetri:1,asymptot:4,atoi:15,atom:1,attempt:[1,5,6,8],attend:14,attent:[1,13],attr:[],attract:[1,8],attribut:[1,7],attributeerror:4,audi:1,aurelien:[1,14,17],author:[1,2,8,15],authour:1,auto:[7,8,15],autocor:15,autocorrelation_tim:15,autocorrelform:15,autocovari:15,autoencod:12,autoencond:12,autograd:[5,12],autom:[1,12],automac:13,automag:[],automat:[1,2,9,12,13],autonom:17,avail:[1,2,4,8,9,12,13,14,17],averag:[0,1,2,4,5,7,8,15,16],avg:15,avoid:[0,1,3,4,5,7,9,13],awai:15,awar:8,award:16,axes3d:5,axes:[1,4,5,6,7,8,9],axessubplot:1,axhlin:6,axi:[0,1,2,4,5,6,7,8,9,10,15],axiom:3,axlabel:1,axvlin:6,b_1:[5,10],b_2:5,b_5:5,b_group:7,b_i:[1,2,10],b_ia_:1,b_index:7,b_j:[2,10],b_k:[2,5,10],b_m:10,b_score:7,b_valu:7,bachelor:14,back:[1,3,6,7,8,13,15],backbon:13,backend:2,background:17,backpropag:2,backtrack:7,backup:13,backward:[2,10,13],bad:15,badli:15,bag:[7,12,14],bag_clf:8,baggin:[],baggingboot:8,baggingclassifi:8,baggingtre:8,balanc:4,band:13,bandwidth:13,bar:[1,9],barber:17,bare:8,barebon:0,base:[0,1,2,3,6,7,8,12,15,16,17],basi:[3,5,6,8,9,10,13],basic:[6,10,12,14,15],batch:[5,9,10],batch_siz:2,bay:5,bayesian:[3,12,17],be0d226abb41:5,becaus:[0,1,2,3,4,5,6,7,10],becom:[0,1,2,3,4,5,7,10,15],been:[1,2,4,5,9,10,12,13],befor:[0,1,2,3,4,5,6,10,13,15],beforehand:[0,1,15],begin:[0,1,2,3,4,5,6,7,9,10,13,15],behav:[2,4,5],behavior:[1,2,5],behaviour:10,behind:[1,2,5,6],being:[0,1,2,3,5,6,8,9,10,15],believ:[7,13],belong:[0,3,5,6,7],below:[1,2,3,4,5,6,7,8,9,10,13,15],benchmark:8,bendik:16,benefici:[2,5],benefit:[1,2,5,9,12],bengio:[2,14,17],benign:[2,5],besid:3,bessel:3,best:[0,1,2,3,4,5,6,7,8,10,15,16],beta:[1,2,3,4,5,8,9],beta_0:[1,2,3,5],beta_0x_:1,beta_1:[1,2,3,5,8],beta_1x_0:1,beta_1x_1:[1,5],beta_1x_2:1,beta_1x_:1,beta_1x_i:5,beta_2:[1,5],beta_2x_0:1,beta_2x_1:1,beta_2x_2:[1,5],beta_2x_:1,beta_:[1,5],beta_i:[1,3],beta_j:[1,3,5],beta_k:5,beta_linreg:5,beta_m:8,beta_mg_m:8,beta_p:5,beta_px_p:5,better:[0,1,2,3,5,7,8,9,10],between:[0,1,2,3,4,5,6,7,9,10,15],beyond:[1,2,3,5,6],bia:[1,2,3,5,6,7,8,10,14,15],bias:[2,3,4,7,10],big:[0,1,2,3,4],bigger:2,bigr:10,bike:7,bilbo:[],bilek:16,billion:[10,12],bin:[1,4,5,15],binari:[1,3,5,7,8,10,14,15],bind:[1,4],binomi:12,binsboot:4,bioinformat:1,biolog:[2,10,17],bios1100:12,bird:1,birth:[],bishop:[14,17],bit:[0,2,13,15],bitwis:15,bla:[3,13],black:[0,6,7],block:[0,4,8,12,13],blockingavg:15,blockingstd:15,blockingvar:15,blocksiz:15,blocksizemax:15,blocksizemin:15,blue:1,bmatrix:[1,2,3,5,6,9,13],bmi:2,bodi:[1,2,10],bold:2,boldfac:[1,3],boldsymbol:[0,1,2,3,4,5,6,8,9],boltzmann:[10,12],book:17,bool:0,boost:[2,7,12,14],boostrap:8,bootavg:15,bootstd:15,bootstrap:[2,12,14],bootvar:15,bootvec:15,borrow:[],boston_dataset:1,bot:6,both:[0,1,2,3,4,5,6,7,8,12,13,15,16],bottl:5,bottom:4,bound:[1,3,4,6,10],boundari:[6,9,10],boundkei:3,box:7,boyd:[5,6],bracket:15,brain:[2,5,10],branch:7,breast:[3,5,9],breviti:5,brew:[1,12],brg:6,briefli:1,bring:[1,3,8],broad:1,broadcast:0,browser:[],brute:[3,9],bsol:[],bsubex:[],build:[1,3,4,8,13,15],built:[1,2,4],bunch:9,busi:1,bzl:3,c46dd114b2af:6,c_0:15,c_1:10,c_2:10,c_3:10,c_4:10,c_i:[5,10],c_k:15,cach:8,cal:[1,5,6,8,10],calcul:[0,1,2,3,4,5,6,7,8,9,10,13],call:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],callabl:4,calor:1,cambridg:[5,17],came:0,can:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],cancel:[1,5],cancer:[3,8],cancerpd:5,candid:[6,7,8],cannot:[1,2,3,5,6,7,14,15],canopi:[1,12],cap:3,capabl:[1,2,5,6,12],capita:1,captur:[9,10],card:[1,5],cardin:2,care:[0,9],carefulli:5,carlo:[1,4,12,15,17],carri:[4,5],cart:8,casella:17,cast:2,categor:[1,2,7,9],categori:[0,1,2,5,8,10,14],categorical_crossentropi:2,caus:[1,3,4,15],causal:1,causat:1,cax:2,cbar:2,ccc:[3,10],cdf:15,cdot:[0,1,4,5,10,13,15],celebr:5,center:[0,1,2,4,5,6,7,9,15],centr:17,central:[1,3,4,6,13],centroid:[0,15],centroid_differ:0,centroid_list:0,certain:[0,1,4,5,7,15],cha:1,chain:[2,5,12,15],challeng:0,chanc:[2,3,5,15],chang:[0,1,2,3,4,5,6,7,9,10,13,15],chapter:[0,1,4,8,9,13,14,17],charact:[1,3,6,15],character:[6,7,8,10,15],characterist:[1,2,5,8],charg:1,charl:1,chd:5,chddata:5,cheap:3,cheaper:[2,5],chebychev:0,check:[1,2,3,5,9,13],check_consistent_length:[],chemic:15,chen:8,choic:[0,1,2,4,5,7,10,13,15],choleski:[3,13],choos:[0,4,5,7,8,9,15],chosen:[1,2,4,5,6,7,8,15],chosen_datapoint:2,christian:17,christoph:[14,17],cin:15,circ:[2,10],circl:[1,6,10],circumfer:7,circumv:[2,3,5],clariti:[0,15],class_nam:7,class_val:7,class_valu:7,classic:[5,7],classif:[1,3,4,5,6,9,10,12,14,17],classifi:[1,2,5,7,8,9],classificaton:2,classifii:8,clean:2,clear:[2,3,5,8,10],clearer:0,clearli:[1,3,4,5,6,15],clever:[0,2,8],clf3:1,clf:[1,6,7,8],clf_ridg:[],clip:15,close:[0,1,2,4,5,6,7,9,10,15,17],closer:[3,5],closest:[0,5,6,9],closur:12,cloud:12,clust:0,cluster:[1,2,4,9,12,14],cluster_label:0,cmap:[1,2,6,7,8],cmath:15,cmb:14,cmd:7,cn_:15,cnn:10,cntk:12,code:[4,5,6,12,13,14,17],coef0:6,coef:1,coef_:[1,3,5,6,7],coeffici:[1,3,4,5,6,7,13],coerc:[1,4],coin:[8,15],coin_toss:8,col:[1,9],colab:12,cold:7,colinear:1,collaps:6,collect:[1,4,8,9,12,15,17],collinear:3,color:[1,4,6,7,8,15],colorbar:2,colsample_bytre:8,colsaobject:8,column:[1,2,3,4,5,6,7,9,10,13],columntransform:7,com:[12,16,17],combin:[2,3,4,5,8,15],come:[0,1,2,3,5,10],comma:[],command:[1,2,15],comment:1,commerci:[1,12],commod:1,common:[0,1,2,3,4,5,7,9,15],commonli:[0,1,2,4,5,7],commun:[1,10],compact:[0,1,2,3,4,5,7,9,10],compar:[0,1,3,4,5,9,13],comparison:5,compat:5,compet:1,competit:8,compil:[1,2,12,13,15],complet:[1,7,10],completenn:10,complex:[2,3,4,5,6,7,9,10],complic:[0,1,2,4,5,7],compon:[0,1,2,3,4,5,7,12,14],components_:9,compos:[0,7,10],compphys:[12,14,17],compress:1,compris:4,compromis:3,compulsori:12,comput:[0,1,2,3,4,5,6,8,9,10,12,13,14,17],computation:[1,4,5,7,15],con:3,concaten:0,concav:[2,5],concentr:[1,8],concept:[0,1,12],conceptu:[5,10],concern:[0,1,2,5],concic:[],conclud:[1,3],conclus:2,conda:[1,2,12],condit:[1,3,4,5,6,7,9,15],conduct:12,coneqp:3,confid:[1,3,4,5,6],confirm:[3,10],confus:[3,4,8,13],confusion_matrix:7,congruenti:15,conjug:6,conjugaci:5,connect:[1,2,5,7,9,10,13],consequ:[3,4,5,6,8,10],conserv:[0,3],consid:[0,1,2,3,4,5,6,7,8,10,13,15],consider:[1,2,3,5],consist:[1,2,4,5,10,15],constant:[1,3,5,6,10,15],constitu:1,constitut:4,constrain:[2,3,5,9],constraint:[3,5,6],construct:[1,2,3,4,5,6,7,8,9,13,15,17],contact:1,contain:[0,1,3,4,5,6,7,9,10,13,15,17],contemporari:17,content:[2,12,13],context:[4,5,8],continu:[1,2,3,4,5,6,7,8,10,13],contour:[5,7,8],contourf:[6,7,8],contrast:[2,7,8,10],contribut:[1,3,15],contributor:1,control:[1,2,5,7,12],conveni:[1,3,4,5,10,13],convent:10,converg:[0,1,2,3,5,6,9],convergencewarn:[1,2,5,6,9],convert:[1,2,3,5,7,9,13],convex:3,convinc:5,convolut:[2,12,14],cool:7,coordin:[0,3,10],coorel:1,copi:[0,1,2],core:8,corel:1,coronari:5,corr:[1,3,5,9],correalt:[9,12],correct:[0,1,2,3,5,13,15],correctli:[2,4,8],correl:[1,2,3,5,8,10,12],correlation_matrix:[1,3,5,9],correspond:[0,1,3,4,6,7,9,10,12,13,15],cortex:10,cos:[1,4,5,7],cosin:[0,4],cost:[1,3,4,5,6,7,10,15],could:[1,2,3,4,5,6,7,8,9,10,13,15],coulomb:1,count:[1,7,14,15,16],countor:5,coupl:3,cours:[1,2,3,9,14,15],courvil:[14,17],cout:15,cov:[3,4,9,13,15],cov_xi:[3,9],cov_xx:[3,9],cov_yi:[3,9],covari:[1,5,12,13],covariance_matrix:[0,3,9],cover:[1,3,11,12,17],covert:1,covxi:15,covxx:15,covxz:15,covyi:15,covyz:15,covzz:15,cpu:2,creat:[2,3,7,8,9,10,12],create_biases_and_weight:2,create_neural_network_kera:2,create_x:[3,9],credit:[1,5],crim:1,crime:1,criteria:[0,1,7,8,15],criterion:[5,7,8],cross:[1,2,5,7,8,12,14,15],cross_val_scor:4,cross_valid:[5,8],crossvalid:4,crucial:[2,15],csr_matrix:13,cstdlib:15,csv:[1,4,5,7],ctnk:2,cubic:1,cumbersom:3,cumsum:[8,9],cumul:[4,8],cumulative_heads_ratio:8,cup:3,current:[0,2,5],curs:1,curv:[5,8,10],curvatur:5,custom:[0,3],custom_cmap2:[7,8],custom_cmap:[7,8],cutpoint:7,cvxbook:5,cvxopt:[3,6],cyber:17,cycl:[0,2,10,15],d_f:5,dagger:[3,13],dai:[2,7,12],dalen:16,darget:7,darkr:15,dat:1,dat_id:[1,4,5,7],data1:0,data2:0,data3:0,data4:0,data:[0,3,4,6,8,10,13,17],data_id:[1,4,5,7],data_indic:2,data_panda:[],data_path:[1,4,5,7],databas:2,datafil:[1,4,5,7],datafram:[1,3,5,7,9],datapoint:[2,3,4,5,9],dataset:[0,1,4,5,6,7,8,9],date:1,daughter:8,david:17,dbh:2,dbo:2,dcomposit:13,dcost:3,dead:2,deal:[0,1,2,3,5,6,9,13,15],debt:5,debug:[3,4],decad:1,decai:[1,5,15],decent:8,decid:[1,3,4,7],decim:1,decis:[1,2,6,9,12,14,17],decision_funct:6,decision_tre:7,decisiontreeclassifi:[7,8],decisiontreeregressor:[1,7,8],declar:[1,13],decompos:[3,13],decomposit:[1,10],decompost:3,decorrel:[5,8],decreas:[2,3,4,5,8,9],deduc:1,deep:[5,10,12,14,17],deep_tree_clf1:7,deep_tree_clf2:7,deep_tree_clf:[7,8],deepen:12,deeper:[0,1],deeplearningbook:17,def:[0,1,2,3,4,5,6,7,8,9,15],def_covari:15,defect:3,defici:3,defin:[0,1,3,4,5,6,7,8,9,10,13,15],definit:[0,2,3,4,5,6,8,9,10,13],defint:15,degre:[3,4,6,7,8,9,15],del:2,delet:[4,15],deliv:14,delta:[0,1,5,6,10,15],delta_:[2,13],delta_h:[1,2],delta_j:10,delta_k:10,delta_l:2,delta_n:1,delug:12,delv:1,demand:5,demonstr:[1,3,4,5,9,10,12],denomin:[2,3],denot:[2,4,5,15],dens:[0,2],densiti:[0,1,4,15],depart:16,depend:[0,1,2,3,4,5,6,9,10,12,15],depict:15,deploy:[1,12],deprec:1,depth:[1,7,8,13],deriv:[1,2,4,5,6,8,9,12],descend:[3,7,9],descent:[1,2,6,10],descr:1,describ:[0,1,3,4,5,6,8,9,10,13],descript:[0,1,6,7],design:[1,2,3,4,5,8,9,10],designmatrix:1,desir:[0,1,3,5],despit:[2,10],destroi:13,det:[3,13],detail:[0,1,5,9,13],detect:[6,10],determin:[1,3,5,6,7,8,9,10,13,15],determinist:[5,15],dev:[2,15],develop:[0,1,3,6,8,9,10,12,13],deviat:[1,2,3,4],devis:10,df1:[],diag:[3,6],diagnost:[2,8],diagon:[1,3,5,13,15],diagonaliz:3,diagram:8,dict:6,dict_kei:1,dictionari:1,did:[0,1,2,3,5,8,9],die:2,diffeent:6,differ:[0,1,2,3,4,5,7,8,9,10,12,13,15,17],differenti:[12,13],difficult:[0,1,2,4,5,8,15],difficulti:[1,2,5],digit:[1,2,14,16],dilemma:5,dilut:2,dim:[0,9],dimens:[0,1,2,3,6,9,13],dimension:[0,1,3,4,5,7,9,12,13],dimensionless:1,diment:13,dimes:0,direct:[0,1,2,5,9,10],directli:[0,2,3,15],directori:[],disadvantag:1,discard:[4,9],disciplin:[1,10],disclaim:15,discourag:5,discov:1,discover:3,discret:[2,3,5],discrimin:[5,8,9],discuss:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,17],diseas:5,disk:[],disord:[2,5],displai:[0,1,2,4,5,6,7,8,9,10,15],displaystyl:[1,3],displot:1,disregard:1,dissimilar:[0,9],dist:0,distanc:[0,1,6,7,9,15],distance_list:7,distinct:[0,5,6,7,8],distinctli:6,distinguish:[1,5,6,15],distplot:1,distribut:[0,1,2,4,5,8,9,12,13],distrubut:[1,12],div:3,dive:[1,6],diverg:[2,5],divid:[1,2,3,4,6,7,9,10,15],divis:[4,5,6,7,13,15],dna:5,dnn:[1,2,10],dnn_kera:2,dnn_model:2,dnn_numpi:2,dnn_scikit:[1,2],doamin:3,doc:[12,14,17],doconc:[],document:[5,9],doe:[0,1,2,3,4,5,6,8,9,10,13,15],doesn:[7,10],dog:2,doi:0,doing:[1,3,4,9],domain:[3,5,6],domin:1,don:[0,1,2,5,6,9,12,15],done:[0,1,3,4,5,7,8,9,13,15],dot:[1,3,5,6,7,8,9,10,13,15],doubl:[13,15],doubli:2,down:[1,5,7,9,10],download:[1,2,3,13,17],dozen:2,drag:5,dramat:9,draw:[4,5,8],drawback:[1,2,5],drawn:[2,4,5,9,15],dre:3,drop:[1,2,3,4,5,9,15],dropna:[1,4],dtype:[0,1,2,13],dub:1,due:[0,2,3,4,5,6,8,10],dummi:1,dure:[1,2,6,7,9,12],dwell:1,dwh:2,dwo:2,dx_1:15,dx_n:15,dying:2,each:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,16],eapprox:1,earli:[2,5],earlier:[1,3,5,6,7,9,10],eas:[0,7],easi:[1,3,4,5,6,7,8,9,10,12,13],easier:[3,4,5,6,7,15],easiest:5,easili:[1,2,3,5,6,7,8,9,10,13],eastern:16,ebind:1,eblock:7,econometr:[],economi:3,ecosystem:12,ect:14,edgecolor:4,edit:[],edu:5,educ:1,eface79dac2c:8,eff:15,effect:[2,5,8,15],effic:2,effici:[1,5,8,12,13,15],efron:[4,15],egrad:5,eig:[3,5,9,13,15],eigen:15,eigenpair:[3,9],eigenvalu:[1,3,5,6,9,13],eigenvector:[3,5,9],eight:13,eigval:[13,15],eigvalu:[5,9],eigvec:[13,15],eigvector:[5,9],eispack:13,either:[1,2,3,4,5,6,7,8,9,15],ekstrom:16,elabor:15,electr:[1,10],electur:17,eleg:9,element:[2,3,4,5,6,9,10,12,13,14,17],elementari:[5,8,13],elementwis:5,elementwise_grad:5,elessar:[],elif:0,elim:13,elimin:[3,6],els:[2,5,7,10,15],elu:2,elus:1,email:[14,16],embed:[1,9],embodi:4,emit:15,emner:17,emphas:[1,8,12],emphasi:[1,12,17],empir:[2,9,15],emploi:[1,2,3,4,5,9,15],employ:1,empti:[4,8],emul:10,enabl:9,encapsul:0,encod:[0,1,3,7,9],encompass:[1,15],encount:[1,2,3,5,15],end:[0,1,2,3,4,5,6,7,8,9,10,13,15],endl:15,endpoint:15,energi:[1,4],enforc:10,eng:17,engin:[1,2,12,15],english:17,enough:[1,4,5],ensembl:[2,7,14,15],ensur:[1,2,3,4,5,9,15],enter:[3,15],enthought:[1,12],entir:[2,5,7,12,15],entiti:[7,10,13],entri:[1,3,6,9,10,13],entropi:[2,5,8],enumer:[1,2,6],env:[3,15],environ:[12,17],eol:1,eosfit:1,epoch:[1,2,5,10],epsilon:[1,3,4,5],epsilon_0:1,epsilon_1:1,epsilon_2:1,epsilon_:1,epsilon_i:1,eqnarrai:[3,4],equal:[0,1,2,3,4,5,6,7,9,10,13,15],equat:[0,2,3,4,5,6,7,8,9,13,15],equilibrium:10,equiv:[5,13,15],equival:[1,2,3,5,6,9,12,13],erf:15,eriador:[],eridg:[],err:[1,8],err_:4,errat:5,errno:[],error:[2,3,4,5,7,9,10,12,15],error_estimate_corr_tim:15,error_hidden:2,error_output:2,escap:5,esl:0,esol:[],especi:[2,5,7,10],essenti:[0,1,3,7,8,10,15],establish:[1,8,9],estim:[1,2,3,4,5,8,9,12,15],estimated_mse_fold:4,estimated_mse_kfold:4,estimated_mse_sklearn:4,esubex:[],eta0:[5,6],eta:[1,2,5,6,10],eta_:5,eta_t:5,eta_v:[1,2],etc:[0,1,2,3,5,6,7,9,10,12,13,15],ethic:12,etsim:4,euclidean:[0,1],evalu:[1,3,4,5,7,15],evalut:5,even:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],event:[3,5,8],eventu:[3,4,5,9,10,16],everi:[0,1,2,3,4,5,7,8,9,10,12,15],everyth:10,everywher:5,evolv:1,exact:[1,3,5,9,10,13,15],exactli:[1,10,12],examin:4,exampl:[3,4,9,10,12,13,14,17],exce:[2,5,10],excel:[0,1,2,3,8,17],except:[3,4,6,7,15],excess:1,excit:1,exclud:[2,4,10],exclus:[1,2,4,15],execut:[3,5],exemplifi:5,exercis:[3,12,14],exhaust:4,exhibit:[1,6],exist:[0,1,2,3,4,5,6,7,13,17],exit:[3,13,15],exp:[1,2,3,4,5,6,8,9,10,15],exp_term:2,expand:[3,5,9],expans:[1,3,5,6,8,10],expect:[1,2,3,4,5,9,10,12],expectation_value_of_h_wrt_p:15,expens:[4,5,8,15],experi:[1,2,4,5,6,12],experiment:[1,4,7,15],expert:[2,7],explain:[0,1,5,7,8,9,15],explained_variance_ratio_:9,explanatori:1,explicit:[1,5,13],explicitli:[0,1],explod:2,exploit:[1,5,10],explor:[2,5,6,12],expon:2,exponenti:[1,2,3,5,8],export_graphviz:7,export_text:7,exporttext:7,expos:12,express:[1,3,4,5,8,10,13,15],exptmean:15,exptvari:15,extend:[3,5,9,12],extens:[1,10,12],extent:[1,2,4,17],extern:7,extra:[2,3],extract:[1,3,5,6,9,13],extrapol:1,extrem:[0,1,2,3,5,6,7,13],extremum:5,extrins:9,eye:[0,1,3,5,13],f11:1,f12:1,f13:1,f1_grad:5,f1d:5,f2_grad_x1:5,f2_grad_x1_analyt:5,f2_grad_x2:5,f2_grad_x2_analyt:5,f3_grad:5,f3_grad_analyt:5,f4_grad:5,f4_grad_analyt:5,f5_grad:5,f6_grad_analyt:5,f6d7a289d493:13,f7_grad:5,f7_grad_analyt:5,f8_grad:5,f9_altern:5,f9_alternative_grad:5,f9_grad:5,f_0:8,f_1:[5,8],f_2:[5,10],f_3:10,f_d:15,f_grad:5,f_grad_analyt:5,f_i:[4,10],f_m:8,face:5,facecolor:[4,6,15],facil:[1,12],facilit:10,fact:[1,2,3,5,7,9,10],factor:[1,2,3,5,7,8,9,13,15],factori:5,fafab0:[7,8],fail:[1,4,5,6,9,16],failur:5,fairli:[0,2,15],fall:[6,7,14],fals:[0,1,2,3,4,5,7,8],famili:[1,5,6,15],familiar:[1,3,6,12,13,15],famou:[4,10,13],far:[0,1,3,5,6,9,10,15],fashion:[1,7,8],fast:[2,4,5,8,10,12,15],faster:[2,9],fastest:5,favor:5,favorit:15,featur:[1,2,3,4,5,6,8,9,10,12,15],feature_nam:[1,2,5,7],feautur:7,fed:2,feed:[1,9,12,14],feed_forward:2,feed_forward_out:2,feed_forward_train:2,feedforward:[2,10],feel:[0,1,3,5,9,12,16],feet:1,few:[0,2,7,11,15],fewer:[1,7,9],ffnn:[2,10],field:[1,10,12],fifth:1,fig:[0,1,2,5,10],fig_id:[1,4,5,7],figaxi:15,figsiz:[1,2,4,5,6,7,8],figur:[0,1,2,3,4,5,6,7,8,10,12],figure_id:[1,4,5,7],figurefil:[1,4,5,7],file:[1,2,3,4,5,6,7,13,15],filenam:[1,15],filenotfounderror:[],fileout:15,fill:[3,7],financ:1,find:[0,1,2,3,4,5,6,7,8,9,10,12,15],fine:[0,1],finit:[3,4,5,10,15],first:[0,1,2,3,4,5,6,7,8,9,13,15,17],firsteigvector:9,fit:[2,3,4,5,6,7,9,10,15],fit_intercept:[1,3,4],fit_mod:7,fit_transform:[1,4,6,7,9],fiti:1,five:[1,7],fix:[1,4,5,8,9,10],fkkt:3,flag:0,flat:[5,10],flatten:[2,3],flexibl:[1,2,4,6,8,10],float32:7,float64:[1,13],flop:[3,13],flow:[2,10],fluctuat:3,fly:9,flyvbjerg:15,fmesh:5,focu:[1,3,4,12,17],focus:[2,5,13],fold:[4,7],folder:1,follow:[0,1,2,3,4,5,6,7,8,9,10,12,13,14,15,16,17],font:[1,5,15],fontdict:15,fontsiz:[2,6,7,8,15],fontweight:2,foral:6,forc:[1,3,8,9],forecast:10,forest:[1,2,7,12,14],forget:9,form:[1,3,4,5,6,7,9,10,12,13,15],formal:[0,15],format:[1,2,4,5,6,7,8,9,12,15,17],formatstrformatt:5,formul:[0,9],formula:[5,15],forth:10,fortran2003:12,fortran90:15,fortran:[1,12,13],fortun:[1,9],forward:[1,4,12,13,14],found:[2,3,4,5,10],foundat:12,four:[3,6,10,13,14],fourier:1,fourth:10,frac:[0,1,2,3,4,5,6,7,8,9,10,13,15],fraction:7,frame:5,framework:[2,6,8,15],frank:[3,9],frankefunct:[3,9],free:[1,5,9,12,13,15,16,17],freecodecamp:12,freedom:3,freeli:1,frequenc:[4,5,15],frequent:[1,5,6,7],frequentist:12,fresh:8,fret:0,fridai:14,friedman:[14,17],frodo:[],from:[0,1,2,4,5,6,7,9,12,13,14,15,16,17],from_cod:7,front:[1,3],fruit:0,fstream:15,fulfil:[3,10],full:[1,2,3,5,7,8,15],full_matric:3,fulli:[4,10,14,15],fun:12,func:4,functionali:9,fundament:[1,4,12],further:7,furthermor:[1,3,4,5,9,10,12],futur:[1,6,7],futurewarn:1,fys:16,g_1:8,g_2:8,g_m:8,gain:[0,2,3,5,7,8],galleri:1,gamge:[],gamma1:6,gamma2:6,gamma:[1,5,6,7,8,9],gamma_0:8,gamma_1:8,gamma_1x:8,gamma_:1,gamma_i:[1,6,15],gamma_j:5,gamma_k:5,gamma_m:8,gamma_x:1,gap:[3,6],gate:10,gather:[1,2,10],gaug:10,gaussbacksub:13,gaussian:[0,3,4,6],gaussian_point:0,gaussian_rbf:6,gave:5,gbc:14,gca:[4,5,6],gd_clf:8,gdclassiffiercgain:8,gdclassiffierconfus:8,gdclassiffierroc:8,gdm:5,gdregress:8,gemv:3,gen:15,gender:1,gener:[0,1,2,3,4,5,6,8,9,10,13,17],generallay:10,generate_simple_clustering_dataset:0,genom:12,geodes:9,geometr:[1,5],geometri:3,georg:17,geq:[3,5,6,7],geron:[1,14,17],get:[0,1,2,3,4,5,7,8,9,12,13,15],get_distances_to_clust:0,get_dummi:7,get_split:7,get_yaxi:6,getattr:4,getsolutionslic:3,gibb:12,gini:8,gini_index:7,git:[1,12],github:[1,12,14,17],gitlab:[1,12],give:[0,1,2,3,5,6,7,8,10,12,15,17],given:[0,1,2,3,4,5,6,7,8,9,10,13,15],glare:0,global:[5,15],glorot:2,gmail:16,goal:[1,5,7],goe:[0,1,2,3,4,5,13],going:[1,2,3,4,5,6,7,9,10],golden:5,gone:3,gong:2,good:[0,2,3,5,7,8,9,12,15,17],goodfellow:[14,17],googl:[0,2,12],got:2,gpu:2,grad:5,grad_analyt:5,grade:14,gradient:[1,6,7,10,12,14],gradientboostingclassifi:8,gradientboostingregressor:8,gradual:[0,2],graph:[2,5,7,9,10],graph_from_dot_data:7,graphic:[1,2,7],grasp:1,gray_r:2,great:5,greater:[2,5,15],greatli:5,greedi:7,green:[1,7,15],grid:[2,4,5,6,10,15],grossli:5,ground:1,group:[0,1,4,5,7,12,14],groupbi:1,grow:[2,7,8],growth:1,guarante:[1,5,15],guess:[0,2,5,8],guestrin:8,guid:2,h_1:5,h_2:5,h_m:8,habit:1,had:[1,2,4,5],hadamard:[2,10],half:[2,6,7],halv:8,hand:[1,2,3,5,9,10,12,13,14,15,17],handl:[1,2,3,7,9,12],handle_unknown:7,handsid:10,handwrit:10,handwritten:[2,3],happen:[0,2,3,5,8,15],hard:[2,5,6,8],hardcopi:12,harder:[1,2],has:[0,1,2,3,4,5,6,7,8,9,10,13,15],hasn:[1,2],hassl:[1,12],hast:12,hasti:[0,1,14,17],hat:[2,3,5,7,8,9,10,13,15],have:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],haven:2,hdf5:[],head:[1,8,15],header:1,heads_proba:8,health:1,hear:1,heart:[1,5],heatmap:[1,2,5],heavili:1,heavisid:2,height:2,held:5,help:[0,1,2,5,10],helper:0,henc:[1,3,4,5,6,7,8,10],her:5,here:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,17],hereaft:[1,6,10],hermitian:13,hessenberg:13,hessian:[1,3,5],heterogen:[7,8],hidden:[2,10],hidden_bia:2,hidden_bias_gradi:2,hidden_layer_s:[1,2],hidden_weight:2,hidden_weights_gradi:2,hierarch:[0,3],high:[0,1,2,3,4,5,7,8,9,12,13],higher:[1,2,3,4,5,6],highest:2,highli:[1,8,12,13,15,17],highwai:1,hing:6,hint:5,hip:12,hire:1,his:5,hist:[4,5,15],histogram:[1,4,5,15],histor:[5,9],histori:10,histplot:1,histtyp:4,hitherto:3,hjorth:16,hobbi:15,hoc:3,hoff:17,hold:[0,2,4,5,15],holder:1,home:1,homework:5,homogen:[2,5,7,8],hopefulli:[1,9,15],horizont:9,hors:5,hot:[2,7],hour:[2,12,14,15,16],how:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,17],howev:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],hspace:[1,6,8,15],hstack:2,htf:14,html:[5,9,12,14,17],http:[5,9,12,13,14,17],huang:1,huber:1,huge:[2,12],human:[1,2,7,10],humid:7,hundr:2,hungri:2,hybrid:14,hydrogen:1,hyperbol:[2,10],hyperparam:6,hyperparamet:[0,3,7],hyperplan:9,i_1:[3,4],i_2:[3,4],ian:17,idea:[1,2,4,5,7,8,10,13,15],ideal:[0,1,4,5,6,15],idem:4,ident:[3,4,5,10,13],identical:3,identifi:[0,1,2,5,6,7,9,10],idum:15,ieor:15,ifi:17,ifs:12,ignor:[1,2,7],iii:13,ijca2016907841:0,illustr:[0,3,5,8,10,12],imag:[2,7,9,10,17],image_path:[1,4,5,7],imagin:2,immedi:[1,12],implement:[0,1,3,5,6,7,8,9,10],impli:[3,4,5,13],implicitli:[9,15],impos:[1,9,10],imposs:[1,3],impress:[1,10],improv:[0,1,3,5,7,8,9],impur:7,imshow:2,in3050:17,in4080:17,in4300:17,in5400:17,inaccur:5,inact:10,inadequ:1,includ:[1,2,3,5,9,10,12,15,16,17],include_bia:[4,7],incom:10,inconsist:[],incorrect:2,incoveni:6,increas:[1,2,3,4,5,6,7,9,10,15],increasingli:15,ind:4,inde:[1,3],indent:15,indentationerror:15,independ:[1,3,4,5,6,10,15],index:[0,1,2,8,12,15,17],index_col:1,indic:[1,2,3,5,7,8,9],indispens:4,individu:[2,5,8,10,15],indu:1,indx:13,ineffici:[0,5],inequ:6,inequaltii:5,inertia:5,inf1000:12,inf1100:12,inf1100l:12,inf1110:12,inf3000:17,inf4490:17,inf5860:17,inf:1,infeas:7,infer:[1,2,4,17],inferenc:2,infil:[1,4,5,7],infin:[3,4,5,9],infinitesim:15,influenc:[4,8],influenti:2,info:[],inform:[0,1,2,4,5,7,9,10,13,17],infti:[5,15],ingeni:5,ingredi:[1,7],inher:4,inherit:13,initi:[0,1,2,4,5,8,13,15],initialis:15,inject:0,inlin:[0,1,2,3,4,5,6,7,8,9,10,13,15],inner:[4,5],innov:17,inplac:5,input:[0,1,2,3,4,5,6,7,8,10,13,15],input_dim:2,inputs:2,inputs_shuffl:[1,2],insert:[3,6,8,15],insid:[1,5],insight:[1,2,3,12,17],insist:5,inspir:[1,2,10,17],instal:[1,2,3,7,14],instanc:[1,2,4,5,7,9],instanti:8,instead:[0,1,2,3,4,5,6,7,9,13,15],institut:2,instruct:[1,2],int32:8,int64:1,int_0:15,int_:15,int_a:15,intak:1,integ:[0,2,5,13,15],integer_vector:2,integr:15,intellig:[0,1,17],intend:8,intens:2,intention:0,interact:[1,7,10,12],intercept:[1,5,6,9],intercept_:[1,5,6,7],interchang:[3,10,13],interconnect:2,interest:[1,2,3,4,5,6,7,10,12,15],interfac:[1,2,13],interior:[1,7],intermedi:13,intern:[2,8,10],interpol:[2,10],interpr:3,interpret:[1,2,5,7,8,10,13,14,15],interv:[1,3,4,5,15],intial:5,intimid:[],intract:1,intrins:[9,13,15],intro:[12,17],introduc:[1,2,3,5,6,8,10,13,15],introduct:[2,5,14,17],introductori:[1,13,17],intuit:[1,3,4,5,6,10],inv:[1,3,5],invalid:[2,6,13],invalu:[1,5,12],invari:2,invd:3,inver:6,invers:[1,5],invers_period:15,inverse_transform:6,invert:[1,3,5,8],invok:[1,6],involv:[1,4,5,9,10],iomanip:15,ios:15,iostream:15,ipca:9,ipynb:12,ipython:[0,1,2,4,5,6,7,8,9,12,13,15],irani:0,iri:[6,7],irreduc:4,irrelev:3,irrespect:1,isn:3,isnul:1,isomap:9,issu:[2,7,13],it_arrai:5,item:[1,5],items:13,iter:[0,1,2,4,5,6,9,15],itr:3,its:[0,1,2,3,4,5,6,7,8,9,10,12,13,17],itself:[3,4,10,15],jackknavg:15,jackknif:[4,12],jackknstd:15,jackknvar:15,jackknvec:15,jacobian:5,jargon:15,jensen:16,jerom:17,job:[6,8],join:[1,4,5,7],joint:3,journal:15,judg:5,julia:[12,13],jump:[0,15],jupyt:[0,1,12,17],just:[0,1,2,3,4,5,6,7,8,9,10,15],justif:1,justifi:8,k_mean:0,kappa_d:15,karim:16,karlsen:16,karush:6,keep:[0,1,2,3,5,9,13],keepdim:[2,4,8],kei:[1,2,10,17],kept:0,kera:[1,12,14],kernel:[1,2,12],kernel_regular:2,kernelpca:9,kev:1,kevin:17,keyword:13,kfold:4,kick:[2,5],kind:[0,1,5,6,10],kjm:12,kkt:[3,6],kktsolver:3,kmeanspoint:0,kn_k:0,know:[0,1,2,3,5,6,12,15],knowledg:[1,12],known:[2,3,4,5,6,7,10,13,15,17],kondev:1,kpca:9,kroneck:0,kuhn:6,kwarg:4,kwown:1,l1_l2:2,l1regl:3,l1regls_mosek2:3,l1regls_mosek:3,l_1:5,l_2:5,l_j:10,la_i:10,la_k:10,lab:[12,14],label:[0,1,2,3,4,5,6,7,8,10,12,13,15],labelencod:[5,8],labels:[6,7],labels_shuffl:[1,2],labor:0,laboratori:14,lack:[0,1],lagrang:[6,9],lambda:[1,2,3,4,5,6,8,10,15],lambda_0:9,lambda_1:[3,6,9],lambda_2:[6,9],lambda_:9,lambda_i:[6,9],lambda_iy_i:6,lambda_jy_iy_j:6,lambda_k:6,lambda_n:[3,6],lamda:2,land:[1,6],landmark:6,landscap:5,langl:[1,9,15],languag:[1,2,6,12,13,17],lapack:[3,13],laplac:3,laptop:12,larg:[1,2,3,4,5,6,7,8,9,12,13,15,17],larger:[1,3,4,5,6,8,9,15],largest:[6,9],lasso:[1,4,5,12,14],last:[0,1,2,3,4,5,6,7,8,10,13,15],later:[0,1,2,5,6,10,12,15],latex:[],latter:[1,3,5,6,9,13,15],lattic:10,law:1,layer:[1,5],lbfg:[5,7,8,9],lbl:4,lcc:[3,4],lda:9,ldot:[1,4,9,15],lead:[1,2,3,4,5,6,7,8,9,10,13,15],leaf:7,leaki:2,lear:5,learn:[3,4,6,7,8,10,13,14,17],learner:8,learning_r:[6,8],learning_rate_init:[1,2],learning_schedul:5,least:[0,1,4,5,6,8,9,12,13,15],leav:[1,2,3,4,7,9],lectur:[1,2,3,5,8,9,10,12,13,14,17],lecturenot:[12,17],left:[1,2,3,4,5,6,7,8,9,10,13,15],leftarrow:[6,10],legend:[1,3,4,5,6,7,8],len:[0,1,2,3,4,6,7,8,9,10,13,15],length:[1,2,5,6,7,12,15],leq:[0,1,3,5,6,15],less:[1,2,3,4,6,7,15],lessen:2,let:[0,1,2,3,4,5,6,7,8,9,10,13,15],letter:[1,13,15],level:[1,2,3,4,7,12,13,14],lib:[1,2,4,5,6,9],liblinear:[6,8],librari:[1,2,3,7,8,9,13,15,17],licens:[1,2,12],lie:[1,4,9,15],lies:[6,9],life:[1,2,6,10],lifetim:5,lift:0,light:[],like:[0,1,2,3,4,5,7,8,9,10,12,13,15],likelihood:[1,2,3,7],lim_:15,limit:[1,3,4,6,9,10,13],lin_clf:6,lin_model:1,lin_reg:7,linalg:[1,3,5,6,9,13,15],line1:6,line2:6,line2d:5,line3:6,line:[1,2,4,5,6,9,13,15],linear:[2,3,4,7,8,9,10,12,14,15],linear_model:[1,3,4,5,6,7,8,9],linear_regress:4,linearli:3,linearloc:5,linearregress:[1,4,5,7],linearsvc:6,liner:2,linerar:8,linewidth:[1,4,6,7,8],link:[1,7,10,12],linlag:3,linpack:13,linreg:1,linspac:[1,4,5,6,7,8,13,15],linu:16,linuek:16,linux:[1,2,12],liquid:1,list:[0,1,2,3,7,12],listedcolormap:[7,8],literatur:[0,2,5,17],littl:[2,7,10],live:6,lle:1,lloyd:0,lmb:[3,4],lmbd:[1,2],lmbd_val:[1,2],lmbda:5,load:[1,2,5,7,8],load_boston:1,load_breast_canc:[2,5,7,8,9],load_digit:2,load_iri:[6,7],loc:[1,4,5,6,7,8],local:[1,2,5,10],locat:6,log10:[3,4],log:[1,2,3,4,5,7,8,9,13],log_:1,log_clf:8,logarithm:[1,3,5,13],logic:[1,2,7],logist:[1,2,6,7,8,9,10,12,14],logisticregress:[5,7,8,9],logit:5,logreg:[5,7,8,9],logspac:[1,2,3,4],longer:[0,6,8,13,15],longest:0,loocv:4,look:[0,1,2,3,4,5,6,7,8,9,13,15],loop:[0,2,4,8,10,12,13,15],lose:2,loss:[1,2,3,4,5,6,8,9,13],lot:[0,1,2,4],low:[1,4,7,8,9,15],lower:[1,2,7,8,13],lowercas:13,lowest:[5,7,15],lstat:1,lstsq:1,lubksb:13,ludcmp:13,lux:13,lvert:2,m_1:0,m_h:1,m_k:0,m_l:10,m_n:1,m_p:1,m_t:5,machin:[2,3,4,7,8,9,10,13,14,17],machinelearn:[12,14,17],mackai:17,made:[1,2,3,4,5,7,9,10],mae:1,magic:0,magnitud:[2,5],mai:[1,2,3,4,5,6,7,9,10,12,13,15],mail:14,main:[1,2,3,5,7,13,17],mainli:[1,3,4,5,7],maintain:4,major:[2,4,5,7,8,13],make:[0,2,3,4,5,6,9,10,12,13,15,17],make_moon:[6,7,8],make_pipelin:[1,4,8],makedir:[1,4,5,7],makeplot:1,malcondit:13,malign:[2,5,7],mammographi:3,manag:[1,12],manhattan:0,mani:[0,1,2,3,4,5,6,7,9,11,12,13,15,17],manifold:9,manual:0,map:[0,1,2,4,5,6,9,10,15],margin:[1,3,6],marit:1,mark:[],marker:[1,5,13],markov:12,marsaglia:15,mask:15,mass:[1,2,3,5],massag:1,masses2016:1,masses2016ol:1,masses2016tre:1,masseval2016:1,master:14,mat1100:12,mat1110:12,mat1120:12,mat3155:14,mat4155:14,mat:12,match:[0,2,3,5],materi:[3,5,13],math:[0,5,10,13,15,17],mathbb:[0,1,3,4,5,6,9,10,13,15],mathbf:[1,3,4,5,6,13,15],mathcal:[2,3,4,5],mathemat:[0,1,5,9,10,12,13,15,17],mathemati:[],mathemt:[],mathrm:[0,1,2,3,4,5,6,7,8,9,10,15],matmul:[2,3],matnat:[16,17],matplotlib:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],matric:[1,2,5,6,9,12,13],matrix:[1,4,5,6,8],matshow:2,matter:5,max:[1,2,5,7,8,10],max_depth:[1,7,8],max_it:[1,2,5,6,9],max_iter:0,max_leaf_nod:8,max_sampl:8,maxdegre:[1,4,8],maxdepth:8,maxim:[2,3,5,6,9],maximum:[0,1,2,3,5,6,7,8],maxpolydegre:[3,4],mbox:[3,4],mcculloch:10,mcint:15,mcintsqr2:15,mean:[2,3,4,5,7,8,9,10,12,13,14],mean_absolute_error:1,mean_divisor:0,mean_i:15,mean_matrix:0,mean_squared_error:[1,4,5,8],mean_squared_log_error:1,mean_vector:0,mean_x:15,meaning:[1,5],meansquarederror:1,meant:[5,8],meantempvec:15,meanvec:15,measur:[0,1,2,3,4,7,9,10,15],mechan:[1,15],median:1,medicin:10,medium:[5,6],medv:1,meet:[1,16],mehta:1,memori:[5,9,10,13],mention:[0,1,5,10,15],mere:1,mersienn:15,meshgrid:[3,6,7,8,9],messag:[3,5],met:[1,6],meteorolog:7,method:[0,1,2,3,6,9,10,12,13,14,17],metric:[0,1,2,4,5,7,8],metropoli:12,mev:[1,15],mglearn:12,mgrid:5,mhjensen:[1,2,5,6,9],microsoft:17,mid:2,midpoint:7,might:[1,2,5,7],mild:7,miller:15,million:1,mimic:10,min:[1,3,6,7],min_:[0,1,3],min_samples_leaf:7,mind:[0,1,5],mine:12,mini:[2,5,9,10],minibatch:[2,5,9],minibathc:5,minim:[0,1,2,3,4,5,6,7,8,9,10,15],minima:[1,2,5],minimum:[1,2,4,5,6,7,9],minkowski:0,minmaxscal:1,minor:15,minst:2,minu:5,mirror:7,misclassif:[6,7,8],misclassifi:[6,8],miser:1,mismatch:2,miss:[1,8],mit:17,mix:2,mixtur:5,mkdir:[1,4,5,7],mlab:[4,15],mle:[3,5],mlp:2,mlpclassifi:2,mlpregressor:1,mnist:[2,9],mod:15,mode:[14,15],model:[0,3,4,5,6,7,8,9,12,15,17],model_select:[1,2,3,4,5,7,8,9],moder:8,modern:[1,4,5,12],modif:[5,10],modifi:[1,2,3,5,6,8,10],modul:[1,4,5,7,8,9,13],modular:15,modulenotfounderror:7,modulo:15,moe:[3,9],moment:[3,4,5],monitor:5,monoton:[3,10,15],mont:[1,4,12,15,17],montecarlocycl:15,moor:3,more:[1,2,3,4,5,6,7,8,9,10,12,14,15],moreov:1,morten:16,mosek:3,most:[0,1,2,3,4,5,6,7,8,9,10,12,15],mostli:[2,9],motion:[1,5],motiv:2,move:[0,1,3,5,7,10,15],mpl:[1,5],mpl_toolkit:5,mplot3d:5,mplregressor:2,mse:[1,3,4,7,8],mse_simpletre:8,mselassopredict:3,mselassotrain:3,msepredict:3,mseridgepredict:3,msetrain:3,msg:1,msle:1,mt19937_64:15,mu0:15,mu1:15,mu2:15,mu_:15,mu_n:9,mu_x:15,much:[1,2,3,4,5,6,7,8,9,10,13,15],mul:3,multi:[1,2,5,12],multiclass:[2,5],multidimension:[9,10],multilay:2,multinomi:5,multipl:[3,4,5,10,15],multipli:[3,5,9,13,15],multiplum:6,multitud:[],multivari:[1,8,9,12,15],multivariate_norm:[0,9],murphi:[9,17],must:[0,2,3,4,5,6,8,10,15],mutat:5,mutual:[2,4,5],mx_:15,myriad:[1,12],mz1:15,mz2:15,n_0:[10,15],n_b:15,n_boostrap:[4,8],n_bootstrap:4,n_categori:2,n_cluster:0,n_compon:9,n_epoch:5,n_estim:8,n_featur:2,n_hidden_neuron:[1,2],n_i:15,n_input:[1,2],n_instanc:7,n_iter_i:[5,9],n_job:8,n_k:0,n_l:[10,15],n_layer:2,n_m:7,n_neuron:2,n_neurons_layer1:2,n_neurons_layer2:2,n_point:0,n_sampl:[0,4,6,7,8],n_split:4,nabla:[2,5],nabla_:5,nabla_w:5,nag:5,naimi:1,naiv:[0,5],nall:1,name:[0,1,2,3,4,5,6,7,8,10,12,13,16],nameerror:[5,8],namespac:15,narrow:5,nation:[2,3],nativ:12,natur:[1,2,5,6,7,10,15,17],navier:10,nb_:13,nbconvert:[],nboot:15,nearest:[2,9],nearli:5,neat:[],neccesari:4,necessari:[0,1,2,6],necessarili:[1,9,15],necesserali:3,neck:5,need:[0,1,2,3,4,5,6,7,8,9,10,13,15],neg:[1,2,3,4,5,8,15],neg_mean_squared_error:4,neglect:15,neglig:15,neighbor:9,neither:5,neq:[0,5,15],nervou:10,nest:[7,10],nesterov:5,net:10,netlib:13,network:[1,7,12,14,17],neural:[1,5,12,14,17],neural_network:[1,2],neuralnetwork:2,neuron:[2,10],neutral:1,neutron:1,never:[2,4,7,15],new_hobbit:[],new_sig:4,newaxi:[1,4,7],newli:1,newton:[2,5,6,15],next:[0,1,2,3,5,6,7,15],next_guess:5,nian:16,nice:[0,1,2,3,9],nichola:16,nicholaskarlsen1102:16,niter:5,nitric:1,nlambda:[3,4],nm_n:1,nmse:4,nn_model:2,node:[2,7,8,10],nois:[1,3,4,5,6,7,8],noisi:[2,4],non:[0,1,2,3,4,5,7,8,9,10,13,15],none:[0,1,2,3,4,5,7,8,15],nonetheless:0,nonlinear:[4,6,7,9,10],nonneg:[4,5,7],nonparametr:4,nonsens:15,nonsingular:13,nonumb:[5,6,13],nor:[2,5],norm:[1,2,3,4,5,6,9],normal:[3,4,5,6,7,8,9,10,12,13],normali:13,normpdf:4,notat:[0,1,3,4,5,15],note:[0,1,2,3,4,5,6,9,10,12,13,14,15,17],notebook:[0,1,2,7,12],noth:[0,2,3,6,10,15],notic:[3,5,10,13,15],novel:[4,8],novemb:2,now:[0,1,3,4,5,6,8,9,10,13,15],nowadai:[1,2,7,12],nox:1,np_assign_points_to_clust:0,np_get_distances_to_clust:0,np_k_mean:0,nsampl:4,nspin:15,nthi:1,nuclear:3,nuclei:[1,15],nucleon:1,nucleu:1,num_tre:8,number:[0,2,3,4,5,6,7,8,9,10,13,14,16],numberid:5,numer:[1,3,4,5,7,8,9,10,12,13,17],numpi:[0,1,2,3,4,5,6,7,8,9,10,12,15],nunmpi:3,obei:[5,9],object:[1,2,3,4,6,8,13],objsens:3,obliqu:3,observ:[0,2,3,4,5,6,7,8,9,10],obtain:[0,1,2,3,4,5,6,7,8,10,13,15],obviou:[3,9,15],obviouli:1,obvious:[1,3,13],occupi:1,occur:[1,6,7,15],odd:[1,5],oen:1,off:[2,3,4,5,7,15],offend:[],offer:[4,9,12,13,14],offic:16,offici:14,ofil:15,ofstream:15,often:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],ofter:13,old:[2,3,5,8],olsbeta:3,omit:[1,3],onc:[0,2,4,5,7,9,15],one:[0,1,2,3,4,5,6,7,8,9,12,13,15],onehot:2,onehot_vector:2,onehotencod:7,ones:[1,3,5,6,7,8,9,13],onli:[0,1,2,3,4,5,6,7,8,9,10,13,15],onlin:[9,14],onto:[3,9],open:[1,2,4,5,7,12,14,15],oper:[1,2,3,4,5,8,9,10,12,13,15],operation:15,opinion:0,oplu:15,opmiz:5,opportun:1,oppos:5,opposit:[2,3,6],opt:[1,2,3,4,5,6,9],optim:[0,1,3,4,7,8,9,14],optimis:2,option:[1,2,3,5,6,9],optmiz:[2,6],orang:1,order:[1,2,3,4,5,6,7,8,9,10,13,15],ordinari:[1,4,5,9,12,14],oreilli:17,org:[5,9,12,13,17],organ:[0,4,5,8,13],orient:[2,3,4,15],origin:[1,3,4,5,6,9,10,13,15],orthogn:3,orthogon:[1,3,5,6,9,13],orthonorm:3,oscar:2,oscil:5,oslo:[1,14,16],osx:[1,12],other:[0,1,2,3,4,5,6,8,12,13,14,17],otherwis:[1,2,5],ouput:[3,5,10],our:[2,4,6,7,8,10,12,13],ourmodel:1,ourselv:[0,1,3,5,6,9],out:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],out_fil:7,outcom:[1,5,7,8,10,15],outdoor:7,outer:10,outfilenam:15,outlier:[1,6],outlin:[4,8,9],outlook:7,outperform:8,output:[1,2,3,5,6,7,8,10,13],output_bia:2,output_bias_gradi:2,output_weight:2,output_weights_gradi:2,outputlayer1:10,outputlayer2:10,over1:5,over:[0,1,2,3,4,5,7,8,10,15],overal:[2,8],overcast:7,overcom:[5,10],overdetermin:1,overfit:[1,2,4,5,7,8],overflow:[2,3],overhead:10,overlap:[5,6,7],overlin:[0,1,3,4,7,8,9,13],overst:1,overview:[0,17],own:[3,5,6,10,12,13],owner:1,oxid:1,oyvinssc:16,p_i:[3,15],p_j:15,p_n:15,p_x:15,pack:1,packag:[1,2,3,4,5,6,9,12,15],page:[1,12],pai:[1,2,5,7],painless:[],pair:[1,7,12,15],panda:[1,3,4,5,7,9,12],panel:[],paper:2,paradigm:1,parallel:[8,13],paramet:[1,2,3,4,5,6,7,8,10,15],parameter:[1,8],parametr:[1,4],paramt:3,park:15,part:[0,1,2,3,4,8,13,14,15,17],partial:[1,2,3,5,6,8,9,10,15],particip:[12,14],particl:[1,5,15],particular:[1,2,3,4,5,7,8,9,10,15,17],particularli:[3,4,5,6,9,15],partit:[2,7],pass:[0,10,15],past:[8,15],patch:[4,15],path:[1,4,5,7,12],patient:5,pattern:[1,10,14,17],pauli:1,pca:[1,5,12,14],pcost:3,pdf:[1,3,4,7,17],pedagog:1,penalti:[4,5],penros:3,pentagon:5,peopl:[1,2,7,12],per:[1,2,4,14],percentag:[1,8,9],perceptron:[1,2,5],peregrin:[],perfect:[1,2],perfectli:4,perform:[0,1,3,4,5,6,8,9,10,12,13,15],perhap:[1,3,5],perimet:2,period:2,permut:9,persist:5,person:[3,5,14,16],perspect:17,pertin:10,petal:[6,7],peter:17,petersen:15,phantom:15,phase:10,phatak:0,phenomena:15,phi:6,phi_k:6,philip:16,philosophi:5,phone:16,phrase:1,physic:[1,2,5,10,15,16,17],pick:[0,2,5,7,8,9,15],pickl:2,pictur:1,pie:12,piec:[0,9],pillow:[1,12],pinv:3,pip3:[1,2],pip:[1,2,12],pipelin:[1,4,6,8],pippin:[],pise:0,pitt:10,pixel:2,pixel_height:2,pixel_width:2,place:[1,4,5,6,13],plai:[1,3,4,6,9,12],plain:[5,6,8,10],plan:[4,7,16,17],plane:[6,7],plateau:[3,15],platform:12,plausibl:10,pleas:[1,5,9],plenti:2,plethora:10,plot:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],plot_confusion_matrix:[5,8],plot_cumulative_gain:[5,8],plot_data:2,plot_dataset:6,plot_decision_boundari:[7,8],plot_import:8,plot_predict:6,plot_regression_predict:7,plot_roc:[5,8],plot_surfac:5,plot_train:7,plot_tre:[7,8],plt:[0,1,2,3,4,5,6,7,8,9,10,13,15],plu:[1,3,5],png:[1,4,5,7],point:[0,1,2,3,4,5,6,7,8,9,13,15,16],points_in_clust:0,poisson:12,poli:[4,6],poly100_kernel_svm_clf:6,poly3:1,poly3_plot:1,poly3dcollect:5,poly_featur:[6,7],poly_features10:7,poly_fit10:7,poly_fit:7,poly_kernel_svm_clf:6,polydegre:[1,3,4,8],polygon:5,polym:10,polynomi:[1,3,4,5,6,7,8,9],polynomial_featur:4,polynomial_svm_clf:6,polynomialfeatur:[1,4,6,7],polytrop:[1,4],poor:[2,5],poorli:1,popul:[1,3],popular:[1,2,4,5,6,7,9,10,12,13,15],popularli:1,portabl:8,portion:[5,9],pose:[1,3,9,15],posit:[0,1,2,3,5,6,8,9,13,15],possibl:[1,2,3,4,5,6,7,8,9,10,12,13,15,16],posterior:3,postpon:1,postul:3,potenti:[1,3,5,10],potr:3,potrf:3,pott:10,power:[1,2,3,4,6,7,10],practic:[1,3,4,6,15],practition:[1,2],pre:3,preced:[2,9,10,15],preceq:6,precis:[1,3,5,9,13,15],pred:4,predicit:1,predict:[1,2,3,4,5,6,7,8,12,17],predict_prob:2,predict_proba:[5,8],predictor:[1,3,5,7,8,9],prefer:[1,2,6,7,9,12],prepar:1,preprocess:[1,4,5,6,7,8,9],prerequisit:1,presenc:5,present:[1,3,5,7,10,15],preserv:9,press:[5,17],pretrain:2,pretti:[1,6,7,12],prev_centroid:0,prevent:[5,15],previou:[1,2,3,5,6,8,9,10,13,15],previous:[0,7,8,15],price:[1,5,7],primal:6,primari:[1,5],primarili:0,prime:15,princip:[1,3,5,12,14],principl:[0,1,4,5,6],print:[0,1,2,3,4,5,6,7,8,9,13,15],print_funct:[6,7],printout:1,prior:[1,3,4],privat:1,prob:[2,15],probabilist:[1,17],probabl:[1,2,4,5,8,12],problem:[1,3,4,6,7,8,9,10,12,13,14,15],proce:[1,3,5,6,7,8,9,13],procedur:[3,4,5,6,8,9],proceed:13,process:[0,1,4,5,7,8,10,12,13,15,17],prod:17,prod_:[2,3,5],produc:[0,1,3,5,7,8,9,10,12,13,15],product:[1,2,3,4,5,6,10,12,13],profess:1,profil:0,program:[0,1,2,3,6,10,12,13,14,15],programm:13,progress:[0,2],progression_plot:0,prohibit:4,project:[1,2,3,5,9,12,14,15,16],project_root_dir:[1,4,5,7],promin:10,promis:6,prone:7,pronounc:[5,12],proof:[1,5,9,10],prop:4,proper:[1,4],properli:[0,2,5,6,8],properti:[1,2,4,5,10,13],proport:[1,2,3,5,7,9,15],propos:[2,8],propto:[3,5],proton:1,prove:5,provid:[1,2,3,4,5,6,7,8,10,12,13,15,17],proxi:[2,5],prun:0,prune:7,pseudoinv:3,pseudoinvers:3,pseudorandom:15,psycholog:1,ptratio:1,punish:[1,2],pure:[7,15],purest:7,puriti:7,purpos:[0,1,8,10],put:2,putarow:3,putboundslic:3,putclist:3,putobjsens:3,putqobj:3,pycod:[],pydata:12,pydot:7,pyhton2:[],pylab:[1,5],pypi:12,pyplot:[0,1,2,3,4,5,6,7,8,9,10,13,15],pythagora:[0,3],python2:1,python3:[1,2,4,5,6,9,12],python:[2,3,5,6,9,10,14],pytorch:[1,12],qquad:[5,9,13],quad:[2,5,13],quadrat:[1,3,5,6,7],qualit:[7,15],qualiti:[1,7,12],quantifi:2,quantil:8,quantit:[1,4,7],quantiti:[0,1,3,4,5,7,8,9,10,13,15],quantum:10,quartil:1,queri:7,question:[1,3,4,5,7,9,10],quick:15,quickli:[2,5,7,9],quirk:0,quit:[2,3,4,7,8,10],quot:15,r2_score:1,r2score:1,r_1:7,r_2:7,r_j:7,r_m:7,rad:1,radial:[1,6,10],radioact:15,radiu:[1,2],rain:7,rais:4,ramp:2,ran1:15,ran2:15,ran3:15,rand:[1,3,4,5,7,8,13,15],rand_max:15,randint:[4,5,7],randn:[1,2,3,4,5,7,9],random:[0,1,2,3,4,5,6,7,12,13,14],random_devic:15,random_forest_model:8,random_index:5,random_indic:2,random_st:[1,5,6,7,8,9],randomforestclassifi:8,randomli:[0,2,4,5,7,15],randomnumbergener:15,rang:[0,1,2,3,4,5,7,8,9,10,13,15],rangl:[1,9,15],rangle_x:15,rank:3,raphson:[2,5,6],rapidli:1,rare:2,rate:[1,2,5,6,7,8,10],rather:[1,2,3,4,5,6,7,8,9,10,13,15],ratio:[5,7,8,9],rational:1,ravel:[3,4,5,6,7,8,9],raw:15,rbf:[6,9,10],rbf_kernel_svm_clf:6,rbf_pca:9,rcond:1,rcparam:[1,2,5,6,7,8,15],reach:[0,1,2,3,4,5,7,8,9,10,15],read:[0,1,3,4,5,6,9,10,13,14,15,17],read_csv:[1,4,5,7],read_fwf:1,readabl:0,reader:[1,13,15],readi:[0,1,2,3,6,8,9,10,13],readili:2,real:[1,2,4,5,8,9,10,13],realist:6,realiti:15,realiz:[2,10],realli:[1,2],rearrang:5,reason:[0,1,2,5,8,17],reassign:2,recal:[3,4,7,8,9,10,13,15],recalcul:15,receiv:[2,8,10,15],recent:[1,4,5,7,8],recept:10,recip:[1,5,13],reciproc:3,recogn:[1,3,8],recognit:[1,2,10,14,17],recommend:[1,5,6,12,13,14,15,17],reconsid:7,reconstruct:9,record:[8,14],recreat:15,rectangl:[4,5,7],rectangular:3,rectifi:[2,10],recur:[1,12],recurr:[1,2,12,14],recurs:[7,12,13],recycl:15,red:[1,4,6,7],redefin:[1,8],reduc:[2,3,5,7,8,9],reduct:[1,8,9,12,15],refer:[0,1,2,3,4,5,9,10,13,17],refin:10,refit:4,reflect:[1,2,15],refresh:[12,14],reg:[8,9],regard:[2,5,7],regardless:10,region:[7,10],regist:15,reglasso:3,regr_1:[1,7],regr_2:[1,7],regr_3:[1,7],regress:[2,4,6,9,10,12,14],regressor:[1,5,8],regridg:3,regular:[1,3,5,7],reilli:[1,17],reinforc:[1,6,12],reiter:2,rel:[1,4,5,7,10,15],relat:[0,1,2,3,5,9,13,15],relationship:[1,7],relativeerror:1,releas:[2,12],relev:[1,2,3,5,9,12,15],reli:[1,6],reliabl:[5,15],remain:[2,4,10,13,15],remaind:15,remark:2,rememb:[1,5,6,13],remind:[1,3,5,9,13,15],remov:[1,3],render:1,reorder:[3,5],reorgan:1,repeat:[0,1,2,3,4,5,7,8,9,13,15],repeated:1,repeatedli:[4,5,8,15],repetit:[4,14],rephras:5,replac:[0,1,2,3,4,5,8,10],replica:4,repositori:1,repres:[1,2,3,4,5,6,7,8,10,15],represent:[1,2,4,15],reproduc:[1,3,7,10,12,15],repuls:1,request:1,requir:[1,2,3,4,5,6,7,9,10,13],resampl:[1,5,8,12,14,15],rescal:[1,9,10],rescu:3,research:[1,12,17],resembl:[4,15],reserv:[2,3,4,15],reset:15,reshap:[0,1,2,4,6,7,8],residenti:1,residu:[1,3,5],resiz:3,respect:[0,1,2,3,4,5,6,8,9,10,15],respond:10,respons:[1,5,7,10],rest:[1,3],restat:[1,10],restrict:[1,7,10],result:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],ret:4,retail:1,retain:[3,4],return_data:0,return_x_i:7,reus:2,reveal:[1,10],revers:2,review:[12,13],revisit:0,reward:1,rewrit:[1,3,4,5,6,8,9,10,13,15],rewritten:[4,6,8,15],rewrot:5,rgoj5yh7evk:12,rho:[1,8],rho_1:8,rho_2:8,rho_m:8,rhs:3,rich:1,rid:0,ride:7,rideclass:7,ridedata:7,ridg:[1,4,5,9,12,14],ridgebeta:3,right:[1,2,3,4,5,6,7,8,10,13,15],rightarrow:[1,2,3,5,6,9,10,15],rigor:1,rise:1,risk:[1,5],river:1,rmse:1,rmsporp:5,rmsprop:[2,5],rnd_clf:8,rnn:10,rntrick1:15,rntrick2:15,rntrick3:15,rntrick4:15,robert:17,robust:1,robustscal:1,roc:8,role:[1,3,4,6,12],room:[1,16],root:[1,3,5,7,15],rot:[],rotat:[2,6,7,8],rotation_matrix:7,roughli:2,round:[1,5,7],routin:[5,13],row:[1,2,3,4,5,7,9,13],rrr:3,rthe:3,rug:5,rule:[1,2,3,5],run:[0,1,2,3,4,5,6,7,9,12],runtim:[0,2,4],runtimewarn:2,rust:[1,12,13],rvert:2,rvert_2:2,rwidth:4,s_i:5,saddl:5,safe:15,sai:[0,1,2,3,4,5,6,7,8,9,10,13,15],said:[4,5,7],sake:[1,3,5,9],sale:1,sam:[],same:[0,1,2,3,4,5,6,7,9,10,13,15],samm:8,sampl:[0,1,2,3,4,5,6,7,8,12,13],sample_vari:0,sampleexptvari:15,samwis:[],sanitize_sequ:4,sastri:9,satisfactori:1,satisfi:[2,4,5,6,13,15],satur:[2,4],save:[1,4,5,7],save_fig:[1,4,5,7,8],savefig:[1,4,5,7,15],saw:3,scalabl:8,scalar:[3,4,8],scale:[1,2,3,5,6,7,8,9,10,12,16],scaler:[1,5,6,7,8,9],scan:[3,5],scari:3,scatter:[0,1,2,4,5,6,7],scenario:5,schedul:5,scheme:[2,5],schrage:15,scienc:[1,2,5,8,10,12,14,15,17],scientif:[1,12],scientist:[0,1],scikit:[3,5,6,7,8,12,13,14,17],scikitlearn:[],scikitplot:[5,8],scipi:[1,3,5,12,13],score:[1,2,4,5,7,8,9,16],scores_kfold:4,scratch:2,sdg:5,seaborn:[1,2,5],seamless:[1,12],search:[1,2,3,5,7],sec:4,second:[0,1,3,4,5,6,7,9,10,12,13,15],secondeigvector:9,secondli:10,section:[0,9,11,13,14,15],sector:1,see:[0,1,2,3,4,5,6,8,9,10,12,13,14,15],seed:[0,1,2,3,4,5,6,7,9,15],seek:[2,6],seem:2,seemingli:1,seen:[1,2,3,8,10,15],segment:5,seldomli:1,select:[2,3,4,6,7,8,9,14,17],self:[2,3,4,15,17],semest:[5,14],semi:[5,6],send:[3,5,10,16],senior:14,sens:[1,4,6],sensit:[1,3,4,5,7],sentenc:10,separ:[0,1,2,4,6,7,10,12,15],sequenc:[0,5,7,8,10,12,13,15],sequenti:[2,8,10,15],seri:[1,2,3,5,8,9,10,13,14],serif:[1,5,15],serv:[1,2,3,5,17],session:[2,14],set:[0,2,3,4,5,6,8,9,12,13,15],set_:4,set_label:4,set_stream:3,set_tick:[2,6],set_ticklabel:2,set_titl:[0,1,2,5,10],set_xlabel:[1,2,5,10],set_xlim:[5,10],set_xticklabel:2,set_ylabel:[1,2,5],set_ylim:[5,10],set_ytick:5,set_yticklabel:2,setiosflag:15,setminu:4,setosa:[6,7],setosa_or_versicolor:6,setp:4,setprecis:15,setse:3,setup:[2,3,6,12],setw:15,sever:[1,3,4,5,6,7,9,10,12,13,14,15],sgd:[2,5],sgd_clf:6,sgdclassifi:6,sgdreg:5,sgdregressor:5,sgn:3,shallow:5,shape:[0,1,2,3,4,5,6,7,8,9,13],share:2,she:5,shift:[2,10,15],shire:[],shortcom:5,shorter:15,shorthand:[],shortli:13,should:[0,1,3,4,6,7,9,10,13],show:[0,1,2,3,4,5,6,7,8,9,10,13,15],shown:[1,3,5,6,9,10,15],showpoint:15,shrink:[3,6,9],shrinkag:3,shrunk:9,shuffl:[1,2,4,5],side:[1,3,5,6,10,13],sigh:12,sigma0:15,sigma1:15,sigma2:15,sigma:[1,2,3,4,5,8,9,10,13,15],sigma_0:3,sigma_1:3,sigma_2:3,sigma_:[3,13,15],sigma_fn:[5,10],sigma_i:[1,3],sigma_j:3,sigma_m:15,sigma_n:[9,15],sigma_t:5,sigma_x:15,sigmoid:[2,5,6,8,10],sigmundson:16,sign:[2,5,6,8,15],signal:[2,8,10],signific:2,significantli:[2,5,15],sim:[3,4,5,15],similar:[0,1,2,3,4,5,6,7,8,9,12,13],similarli:[1,2,3,6,8,15],simpl:[0,2,3,4,6,8,9,10,12,13],simplepredict:8,simpler:[0,1,2,3,5,12],simplest:[0,1,2,7,8,10],simpletre:8,simpli:[1,2,3,4,6,7,8,9,10,12,13,15],simplic:[0,3,5,6,7,8,9,10],simplicti:3,simplifi:[1,4,7,12],simplist:15,simul:[4,15],simultan:4,sin:[1,2,5,7,10,13],sinc:[1,2,3,4,5,6,7,8,9,13,15,17],sine:10,singl:[1,2,3,5,6,7,10,13,15],singular:[1,4,5,13],site:[1,2,4,5,6,9,14],situat:[1,3,5],six:15,size:[1,2,3,4,5,6,7,8,9,13,15],sketch:8,ski:7,skill:1,skip:9,skl:1,sklearn:[1,2,3,4,5,6,7,8,9],skplt:[5,8],slack:6,slice:13,slide:[1,15],slight:[4,5],slightli:[2,3,4,5,8,15],slope:[6,9,10],slow:[1,5,6],slower:[3,13],slowli:10,slp:2,small:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],smaller:[1,2,3,4,5,6,7,9,15],smallest:[0,1],smallest_row_index:0,smart:4,smooth:[1,5],sne:9,sneak:[],sns:[1,2,5],soar:4,social:1,soft:[2,5,8,10],soften:6,softmax:5,softwar:[1,6,12,13,14],sol:6,sole:1,solid:[1,5],solitem:3,soltyp:3,solut:[1,2,3,4,5,6,8,9,13,15],solutionsummari:3,solv:[1,2,3,5,6,8,9,10,13],solver:[3,6,7,8,9,13],some:[0,1,2,3,4,6,7,8,9,10,13,15],someth:[1,2,5,7,9,15],sometim:[0,1,2,5,9,10],soon:13,sophist:1,sopt:5,sort:[3,4,7,9,15],sound:3,sourc:[0,1,2,4,12,13,15],space:[0,1,2,3,5,6,7,9,10,15],span:[1,3,7,9,13],spare:2,spars:[3,13],sparse_mtx:13,sparsiti:8,spatial:[2,10],spdiag:3,speak:15,special:[4,5,8,10,13,15],specif:[1,2,3,4,5,6,7,9,10,12,13,15],specifi:[0,1,3,4,5,7,9,15],specifici:[1,8],spectral:2,speech:[1,2,10],speed:[2,5],spend:15,sphere:1,spite:1,spline:6,split:[0,2,3,4,6,7,8,9,15],splite:1,splitter:[2,8],spmatrix:3,spontan:15,spread:[1,9,15],springer:17,spuriou:5,sqquar:3,sqrt:[1,3,4,5,6,8,9,15],squar:[0,2,4,5,6,7,9,12,13,15],squarederror:8,squaredeuclidean:0,squash:10,srand:15,stabil:3,stabl:[1,3,5,7,9,12],stack:4,stage:[3,5],stai:[1,3,9],stand:[1,3,7,10],standard:[1,2,3,4,6,8,10,13],standardscal:[1,5,6,7,8,9],stanford:5,start:[0,1,2,3,4,5,6,7,8,9,10,13,15],start_tim:0,startpoint:15,stat:4,state:[0,2,3,5,6,8,9,10,12,15],statement:[1,5,13],statis:[],statist:[0,1,2,5,7,8,9,10,13,14,17],statu:[1,5,9],std:[1,4,15],stdev:15,stdout:3,steep:5,step:[0,1,2,4,5,7,8,9,10,13,15],step_fn:[5,10],step_length:5,steps_list:7,stian:16,still:[3,4,5,9,15],stimuli:10,stk2100:17,stk4021:17,stk4051:17,stk5000:17,stk:17,stochast:[1,2,3,4,6,9,10],stoke:10,stone:[1,5],stop:[0,2,5,7,9,15],storag:3,store:[1,2,5,9,15],str:2,straight:[1,4,5,6],straightforward:[1,3,4,5,6,7,8,13],strategi:[1,2,7],stratifi:4,streamtyp:3,strength:[0,1,3],stretch:9,strict:[5,6],strictli:[5,6],string:[2,15],stroke:5,strong:[7,8,10,15],strongli:[1,6,12,13],stronli:1,structur:[0,1,2,4,7,8,10,12],stuck:[2,5],student:[1,14,16,17],studi:[0,1,3,5,6,9,10,12,17],studier:17,style:[1,5,7],sub:[7,10],subdivid:[1,13],subfield:1,subject:[3,6,15],subplot:[0,1,2,4,5,6,7,8],subplots_adjust:[6,15],subprogram:13,subract:1,subroutin:1,subscript:2,subsequ:[2,3,4,10,13,15],subset:[2,4,5,7,10,12],subspac:[1,6,9],substanti:[7,8],substep:9,substitut:[4,10,13],subsubset:7,subtl:2,subtract:[1,3,4,5,9,13,15],subtre:7,succeed:1,success:[5,7,15],successfulli:7,sucess:15,sudo:[1,12],suffer:[1,2,3,8],suffici:[2,4,5,6,9],suggest:[2,5,17],suit:[6,10],suitabl:[1,15],sum:[0,1,2,3,4,5,6,7,8,9,10,15],sum_:[0,1,2,3,4,5,6,7,8,9,10,13,15],sum_i:[3,4,5,6],sum_k:[6,10,13],summar:[0,3,4,7],summari:[0,2,8,14],summat:[],sunni:7,superscript:[2,10],supervis:[1,3,4,5,7,10,12],supplement:5,support:[1,2,5,7,8,9,12],suppos:[1,3,4,5,6,8,9,10,13],suppress:[3,5],sure:2,surfac:1,surpris:1,surround:12,survei:[1,3],svc:[6,7,8],svd:[1,4,9],svdinv:3,svm:[6,7,8,9],svm_clf:[6,8],swath:3,symbol:[2,3,5,9,12,15],symmeteri:2,symmetr:[1,3,5,6,9,10,13],sympi:[1,12],synonim:15,syntax:[2,5,13],syntaxerror:[2,6,13],syrk:3,sys:[3,5],system:[1,2,3,5,7,8,10,12,13,17],systemat:[4,15],t_0:[5,7],t_1:5,t_b:8,t_i:[2,10],t_j:[3,10],t_k:7,tabl:[7,15,16],tabul:1,tabular:[],tackl:0,tag:[0,3,5,10,13,15],taht:1,tail:15,tailor:[6,9],taiwan:1,take:[0,1,2,3,4,5,6,7,8,9,10,12,13,15],taken:[1,2,4,5,8,13],tangent:[2,5,10],tanh:[2,5,6,10],target:[1,2,3,5,6,7,8,9,10],target_nam:7,task:[0,1,2,3,4,7,9,10],tau:[3,15],tax:1,taylor:5,taylornr:5,team:2,teaser:1,technic:[0,1,3],techniqu:[1,2,5,6,8,12,15,17],technolog:[1,2],tek5040:17,tell:[1,4,5,8,9,15],temp1:2,temp2:2,temp:2,temperatur:[1,7],temporarili:2,ten:[],tend:[0,3,4,5,6,7,8,10],tendenc:1,tension:4,tensorflow:[0,1,6,12,13,14,17],term1:[3,9],term2:[3,9],term3:[3,9],term4:[3,9],term:[0,1,2,3,4,5,6,7,8,9,10,15],termin:[1,3,5,7,8],test:[3,4,5,6,7,8,15],test_accuraci:2,test_data:0,test_ind:4,test_pr:2,test_predict:2,test_scor:[5,8],test_siz:[1,2,3,4,8],test_split:7,tester:[],testerror:[1,4],text:[1,2,3,5,6,7,9,13,15,17],textual:7,textur:2,than:[1,2,3,4,5,7,8,9,10,12,15],thats:0,theano:[2,12],thei:[0,1,2,3,4,5,6,7,9,10,13,15],them:[1,2,5,6,7,8,9,10,13],theme:1,themselv:[1,15],thenc:4,theorem:[4,5],theoret:[1,8],theori:[1,2,5,6,7,10,12,17],thereaft:[1,4,9,10,13],therebi:[1,3,5,9,15],therefor:[1,2,4,5,6,9,15],therein:9,thereof:[1,4,5],theta:[2,4,5,15],theta_:[2,5],theta_i:2,theta_k:15,theta_linreg:5,theta_t:5,thi:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,17],thing:[0,1,2,3,5,7,15],think:[0,1,2,4,5,7,10,15],third:[1,5],thirti:5,thorough:0,those:[0,3,6,7,8,9,13,14],though:[2,13,15],thought:[0,4,15],thousand:[1,2],three:[1,2,3,4,6,7,10,13,14,16],threshold:[2,5,7,8,9,10],through:[0,1,2,3,5,6,9,10,12,13,15],throughout:[0,1,3,12,13,15],thu:[1,2,3,4,5,6,8,9,10,15,16],thumb:1,thursdai:14,tibshirani:[14,17],ticker:[5,15],tight_layout:[2,5],tightli:9,tild:[1,3,4,9,15],till:[1,5,6,7,8,10,13],time:[0,1,2,3,4,5,6,7,8,9,10,13,14,15],timefunct:15,tini:2,tip:11,titl:[1,2,4,5,6,7,8,15],tmp:5,to_categor:2,to_categorical_numpi:2,to_numer:[1,4],to_str:15,togeth:[1,6,9],toi:0,told:5,toler:0,tomographi:10,too:[1,3,4,5,7,9,15,17],took:6,tool:[0,1,2,4,5,12,15],toolbox:6,top:[1,3,4,7,8,12],topic:[0,1,3,5,6,12,17],topolog:[2,10],toss:8,total:[0,1,2,4,5,6,8,9,10,15,16],totalclustervari:0,totalscatt:0,totalvari:15,toward:[2,5,10],town:1,tpng:7,traceback:[4,5,7,8],track:[0,5,13],tract:1,tractabl:1,trade:[3,4,7],tradeoff:[1,3,14],tradit:[1,2,4],train:[3,4,5,6,7,8,9,10],train_accuraci:[1,2],train_end:[1,2],train_ind:4,train_pr:2,train_siz:[1,2],train_test_split:[1,2,3,4,5,7,8,9],train_test_split_numpi:[1,2],trainerror:1,trainingerror:4,trait:1,tran:3,transfer:7,transform:[1,3,4,5,6,7,8,9,10,12,13],transit:10,translat:[2,8],transpos:[2,3,9],travers:1,treat:[1,2,4,5,10,15],tree:[1,2,12,14],tree_clf:[7,8],tree_clf_:7,tree_clf_sr:7,tree_reg1:7,tree_reg2:7,tree_reg:7,trend:15,trevor:17,tri:[5,7],triain:1,trial:[1,4,5,15],triangl:5,triangular:13,trick:[5,6,9,15],trickier:15,tridiagon:13,trillion:12,trivial:[1,2,3,9,15],troubl:[1,6,10],true_divid:2,true_fun:4,tucker:6,tumor:[5,7],tumour:5,tunabl:2,tune:[5,7,13],turn:[1,2,3,4,5,6,7,8,9,10,13,15],tutori:2,tweak:[2,8,15],twice:5,twist:9,twister:15,two:[0,1,2,3,4,5,7,8,9,10,13,14,15,17],tx_1:5,type:[1,2,4,5,6,8,13,15],typeerror:[],typic:[1,2,3,5,7,8,10,15],u_i:10,u_m:8,ubuntu:[1,12],uci:1,uio:[16,17],unari:13,unbalanc:[4,7],unbias:[1,3,4,15],uncertainti:[1,3],uncertitud:15,unchang:2,uncorrel:8,undefin:3,under:[1,2,3,4,5,8,12],underdetermin:1,underfit:[2,4],underflowproblem:3,undergo:3,undergradu:14,underli:[1,2,5,7,15],underset:0,understand:[0,1,2,3,5,8,12],understood:[0,5,6],undesir:6,undetermin:[3,6],unexpect:[4,15],unexpected:15,unfortun:[2,6,7,8],unicode_liter:[6,7],uniform:[1,2,3,5,9],uniform_real_distribut:15,uniformli:[5,15],unifrompdf:15,unimport:5,union:[3,4],uniqu:[0,1,4,5,13],unique_cluster_label:0,unit:[1,2,3,8,10,15],unitari:[3,13],unitarili:13,uniti:15,univari:15,univers:[1,2,5,14,16],unix:2,unknow:[1,13],unknown:[1,2,3,4,6,8,13],unknowwn:10,unlabel:2,unless:[1,4,5,9],unlik:[2,5,6,15],unnecessarili:7,unravel:2,unrol:9,unseen:[1,5,7],unstabl:2,unsupervis:[1,2,10,12,14],unsymmetr:13,until:[0,2,5,7,10,15],untouch:1,unusu:10,updat:[0,2,4,5,8,10],upload:[12,17],upon:[2,9,13],upper:[1,6,7,13],uppercas:[13,15],ups:[],usag:[1,6,12,15],usd10000:1,usd:1,use:[0,1,3,4,5,6,7,8,9,10,12,13,14],usecol:1,used:[0,1,2,3,4,6,7,8,9,10,12,13,15,17],useful:[1,2,3,4,5,7,9,10,12,13,15,17],useless:2,user:[1,2,5,6,9,12,13],uses:[1,2,3,4,7,9,10,13,15],usetex:15,using:[0,2,3,4,5,6,7,8,9,10,13],usr:15,usual:[0,1,5,10],util:[0,2,4,5,8],v_0:9,valid:[1,2,5,7,8,12,14,15],valu:[0,1,2,4,5,6,7,8,10,12,13],valuat:7,valueerror:[],van:1,vandenbergh:[5,6],vandermond:1,vanilla:[1,9],vanish:[2,5,15],var_x:15,varabl:6,varepsilon:[3,4],varepsilon_:[3,4],varepsilon_i:[3,4],vari:[1,2,4,8],variabl:[0,1,2,3,4,5,6,8,9,10,13],varianc:[0,1,2,3,5,7,8,9,12,13,14],variance_i:[3,9],variance_x:[3,9],variant:[1,2,4,5,6,10],variat:9,varieti:[1,10,12],variou:[2,3,4,5,6,7,9,10,12,13,15],vartempvec:15,varvec:15,vaue:2,vault:1,vdot:5,vec:[4,15],vector:[0,1,2,3,4,5,7,8,9,12],vector_mean:0,ventur:[1,6,12],verbos:2,veri:[0,1,2,3,4,5,6,7,8,9,10,15,17],verifi:[9,13],versatil:6,versicolor:[6,7],version:[0,1,5,8,12,13,15],versu:2,vert:[1,2,3,5,6,7,9],vert_1:3,vert_2:[3,9],vertic:4,via:[1,3,4,5,6,7,8,9,10,12,13,14,15],vidal:9,video:[1,2,10,12,14],view:[2,3,5,10,15,17],violat:6,virginica:7,viridi:[1,2],virtual:2,viscos:5,viscou:5,vision:1,visual:[1,9,10,12],visualis:2,viz:[6,15],vmax:2,vmc:15,vmin:2,volum:1,vote:8,voting_clf:8,votingclassifi:8,votingsimpl:8,vstack:[3,9,13,15],w_1:[6,13],w_1x_1:6,w_1x_:6,w_2:[6,13],w_2x_2:6,w_2x_:6,w_3:13,w_4:13,w_i:[2,8],w_ix_i:10,w_j:13,w_m:13,w_px_:6,w_px_p:6,wai:[0,1,2,3,4,5,6,8,9,10,11,13,15],walk:7,walker:15,wang:1,want:[0,1,2,3,4,5,6,7,8,9,10,12,15],warn:[1,2,6],warrant:4,watch:12,wavelet:6,weak:[0,7,8],weather:[2,10],web:[12,14],websit:[13,14],wedg:[6,15],wednesdai:14,wee:9,week:[3,5],weekli:[12,17],weight:[1,2,4,5,7,8,10,15],welcom:[6,12],well:[1,2,3,4,5,6,7,8,10,12,13,15,17],went:6,were:[0,1,2,3,4,5,6,8,9,10,15],wessel:1,what:[0,2,3,4,5,6,7,8,9,10,12,13,14],when:[0,1,2,3,4,5,6,7,8,9,10,13,15],whenev:[5,15],where:[0,1,2,3,4,5,6,7,8,9,10,12,13,15,16],wherea:[4,15],wherein:[2,10],whether:[1,3,5,7,15],which:[0,1,3,4,5,6,7,8,9,10,12,13,14,16],whichev:2,white:7,who:[1,14],whole:[0,2,3,7,9],whose:[1,4,8,15],whow:[3,9],why:[1,2,5],wide:[1,2,4,5,10,12,13],widehat:4,width:[1,6,7],wieringen:1,win:8,wind:7,wing:16,wiscons:5,wisconsin:8,wise:[1,2,3,5,10],wish:[0,1,3,5,6,9,13],with_std:1,within:[0,1,5,7,10,15,17],withinclust:0,without:[1,2,3,5,6,7,9,10],won:1,wonder:6,word:[0,1,2,3,15],work:[0,1,2,4,5,6,7,12,14,15],world:[1,6],worldwid:1,wors:[1,2,4],worth:7,would:[1,2,3,4,5,6,7,8,9,10,13,15],wrap:13,write:[1,2,3,5,6,10,11,13,14,15],written:[1,3,5,9,10,12,13,15],wrong:[2,6],wrongli:[8,15],wrote:[3,9],wrt:8,wth:8,www:[12,13,17],wx_1:6,x0s:6,x1_exampl:6,x1d:6,x1s:[6,7,8],x2d:[6,9],x2d_train:9,x2dsl:9,x2s:[6,7,8],x3s:6,x_0:[1,3,9,13],x_1:[1,3,4,5,6,7,8,9,13,15],x_2:[1,3,4,5,6,7,8,9,13,15],x_3:[6,13,15],x_4:13,x_center:9,x_data:2,x_data_ful:2,x_i:[0,1,2,3,4,5,6,7,8,9,10,13,15],x_ix_:1,x_iy_i:6,x_j:[1,6,7,10,15],x_jy_j:6,x_k:[0,10,13,15],x_l:15,x_m:[10,13,15],x_n:[1,4,5,6,9,10,13,15],x_new:[7,8],x_p:[5,7],x_poli:7,x_poly10:7,x_reduc:9,x_scale:6,x_test:[1,2,3,4,5,7,8,9],x_test_scal:[1,5,7,8,9],x_train:[1,2,3,4,5,7,8,9],x_train_scal:[1,5,7,8,9],x_val:2,xarrai:12,xavier:2,xbnew:5,xcode:[1,12],xdclassiffierconfus:8,xdclassiffierroc:8,xg_clf:8,xgb:8,xgbclassifi:8,xgboost:7,xgboot:8,xgbregressor:8,xgparam:8,xgtree:8,xi_1:6,xi_:6,xi_i:6,xlabel:[1,2,3,4,5,6,7,8,15],xlim:[4,8],xmesh:5,xnew:[1,5],xpanda:1,xpd:[3,9],xplot:1,xscale:1,xsr:7,xt_x:5,xtest:4,xtick:[4,6,7],xtrain:4,xytext:6,y_0:[1,3,9,13],y_1:[1,3,5,6,7,9,13],y_1y_1:6,y_1y_1k:6,y_1y_2:6,y_1y_2k:6,y_1y_n:6,y_1y_nk:6,y_2:[1,3,6,7,9,13],y_2y_1:6,y_2y_1k:6,y_2y_2:6,y_2y_2k:6,y_3:[1,7,13],y_4:13,y_data:[1,2,3],y_data_ful:2,y_decis:6,y_fit:1,y_i:[1,2,3,4,5,6,7,8,9,10,13],y_if_:8,y_ix_:1,y_ix_i:[5,6],y_iy_jk:6,y_j:[4,6,10],y_k:10,y_m:13,y_model:[1,3],y_n:[5,6],y_ny_1:6,y_ny_1k:6,y_ny_2:6,y_ny_2k:6,y_ny_n:6,y_ny_nk:6,y_plot:7,y_pred1:7,y_pred2:7,y_pred:[1,2,4,5,6,7,8],y_pred_rf:8,y_pred_tre:8,y_proba:[5,8],y_test:[1,2,3,4,5,7,8,9],y_test_onehot:2,y_test_predict:1,y_train:[1,2,3,4,5,7,8,9],y_train_onehot:2,y_train_predict:1,y_val:2,year:[1,12],yes:[4,5],yet:[1,2,4,5,6,9],yield:[0,1,3,4,5,6,8,10,13,15],ylabel:[1,2,3,4,5,6,7,8,15],ylim:4,ymesh:5,yoshua:[2,17],you:[0,1,2,3,4,5,6,7,8,9,12,13,15,17],young:1,your:[0,2,4,5,6,9,12,13,15],yourself:[5,9],youtub:12,ypred:4,ypredict2:5,ypredict:[1,5],ypredictlasso:3,ypredictol:3,ypredictridg:3,yridg:[],ytest:4,ytick:[4,6,7],ytild:[1,4],ytildelasso:3,ytildenp:1,ytildeol:3,ytilderidg:3,ytrain:4,z_0:13,z_1:13,z_2:13,z_c:2,z_h:2,z_i:[2,10],z_j:[2,10],z_k:10,z_m:2,z_mod:7,z_o:2,zaman:15,zero:[0,1,2,3,4,5,6,7,8,9,10,13,15],zm_h:1,zone:1},titles:["12. Clustering Analysis","3. Linear Regression","14. Building a Feed Forward Neural Network","4. Ridge and Lasso Regression","5. Resampling Methods","6. Logistic Regression","8. Support Vector Machines, overarching aims","9. Decision trees, overarching aims","10. Ensemble Methods: From a Single Tree to Many Trees and Extreme Boosting, Meet the Jungle of Methods","11. Basic ideas of the Principal Component Analysis (PCA)","13. Neural networks","Content in Jupyter Book","Applied Data Analysis and Machine Learning, FYS-STK3155/4155 at the University of Oslo, Norway","2. Linear Algebra, Handling of Arrays and more Python Features","Teaching schedule with links to material","1. Elements of Probability Theory and Statistical Data Analysis","Teachers and Grading","Textbooks"],titleterms:{"2021":16,"4155":12,"case":[6,8,15],"final":10,"function":[1,2,5,6,8,9,10,15],"import":[3,13,15],And:[],Eye:8,FYS:12,OLS:3,RMS:5,The:[0,1,2,3,4,5,6,7,9,10,12,15],Useful:12,Using:5,activ:[2,10],actual:15,adaboost:8,adam:5,adapt:8,adjust:2,again:7,aim:[6,7],algebra:13,algorithm:[0,7,8,9,10],algortithm:5,all:6,analys:3,analysi:[0,1,3,9,12,15],ani:5,anoth:7,appli:12,approach:[1,6],approxim:10,architectur:2,arrai:13,assist:16,august:14,autocorrel:15,automat:5,back:[2,9,10],background:12,bag:8,base:5,basic:[0,1,3,5,7,8,9,13],batch:2,bay:3,befor:9,better:[6,15],bia:4,binari:2,binomi:15,bird:8,block:15,book:11,boost:8,bootstrap:[4,8,15],boston:1,breast:2,brief:[],bring:10,build:[2,7],calcul:15,cancer:[2,5,7,9],cart:7,central:[5,12,15],chain:10,chang:8,chi:1,choos:2,classic:9,classif:[2,7,8],classifi:6,clip:2,cluster:0,code:[0,1,2,3,7,9,10,15],collect:2,compar:8,complex:1,compon:9,comput:[7,15],con:7,concept:15,condit:[],conjug:5,content:11,continu:15,convex:[5,6],convolut:10,correl:[9,15],correspond:[],cost:[2,8],cours:[12,17],covari:[3,9,15],cross:4,cumul:15,cython:[],data:[1,2,5,7,9,12,15],dataset:2,decemb:14,decis:[7,8],decomposit:[3,9,13],deep:2,defin:2,definit:15,degre:1,demonstr:15,dens:1,deriv:[3,10],descent:[5,8],develop:2,deviat:15,diagon:9,dice:15,differ:6,differenti:5,dimension:6,disadvantag:7,discret:15,disguis:15,distribut:[3,15],doing:2,domain:15,down:2,dropout:2,economi:[],element:[1,15],elimin:13,ensembl:8,entropi:7,environ:1,equat:[1,10],error:[1,8],etc:[],evalu:2,event:15,exampl:[1,2,5,6,7,8,15],exercis:1,expect:15,experi:15,explor:1,exponenti:15,express:[],extend:[],extrem:8,fall:16,famili:2,famou:15,fantast:[],featur:[7,13],feed:[2,10],fine:2,first:10,fit:[1,8],forest:8,forward:[2,10],frank:[],freedom:1,frequentist:1,fridai:[],from:[3,8,10],further:3,gaussian:[13,15],gener:[7,15],geometr:9,gini:7,good:1,grade:16,gradient:[2,5,8],handl:13,has:12,hessian:[],homework:[],hous:1,how:15,hyperparamet:2,hyperplan:6,id3:7,idea:[0,9],ideal:[],implement:[2,15],implic:3,improv:2,increment:9,index:7,inform:16,instal:12,instructor:16,interpret:[3,9],introduc:9,introduct:[1,4,12,13],invers:[3,13],iter:8,its:15,jackknif:15,julia:[],jungl:8,jupyt:11,kera:2,kernel:[6,9],lagrangian:6,lasso:3,later:3,layer:[2,10],learn:[1,2,5,9,12],least:3,level:8,librari:12,likelihood:5,limit:[2,5,15],linear:[1,5,6,13],link:[3,9,14,17],logist:5,loss:[],machin:[1,5,6,12],main:15,make:[1,7,8],mani:[8,10],materi:14,math:3,mathemat:[3,6],matric:3,matrix:[2,3,9,10,13,15],matter:1,mean:[0,1,15],meet:[3,8,15],mercer:6,mersenn:15,method:[4,5,7,8,15],mlp:10,model:[1,2,10],moment:15,momentum:5,moon:[6,7],more:[0,13],multilay:10,multipl:2,multipli:6,name:15,need:[],network:[2,5,10],neural:[2,10],newton:[],non:6,normal:[1,2,15],norwai:12,notat:10,novemb:14,now:[2,7],nuclear:1,nueral:5,numba:[],number:[1,15],numer:15,numpi:13,numpython:0,observ:15,obtain:9,octob:14,one:10,optim:[2,5,6,12],ordinari:3,organ:1,oslo:[12,17],other:[7,9,10,15],our:[0,1,3,5,9,15],outcom:12,output:15,overarch:[1,6,7],overview:8,own:[0,1,8,9],packag:13,panda:[],part:[5,12],pass:2,pca:9,pdf:15,perceptron:10,perform:[2,7],period:15,perspect:2,poisson:15,practic:5,pre:2,preprocess:[],prerequisit:12,princip:9,pro:7,probabl:[3,15],problem:[2,5],procedur:7,process:2,program:5,prop:5,propag:[2,10],properti:[3,15],pseudo:15,python:[0,1,7,12,13,15],quick:6,ran0:15,random:[8,9,15],raphson:[],read:7,recip:15,recurr:10,reduc:1,regress:[1,3,5,7,8],regular:2,relev:17,relu:2,remind:[4,6],requir:12,resampl:4,revisit:5,ridg:3,rng:15,rule:10,sampl:[9,15],schedul:14,schemat:7,scikit:[1,2,9],select:15,semest:16,sensit:[],septemb:14,set:[1,7,10],sgd:[],should:[2,15],simpl:[1,5,7,15],singl:8,singular:[3,9],situat:15,size:[],slightli:[],soft:6,softmax:2,softwar:[],solv:[],solver:5,some:5,split:1,squar:[1,3,8],standard:[5,15],state:1,statist:[3,4,12,15],steepest:[5,8],step:[],stk3155:12,stochast:[5,15],stop:[],supervis:2,support:6,svd:3,teach:[14,16],teacher:16,techniqu:9,technolog:12,tensorflow:2,test:[1,2],textbook:17,than:[],theorem:[3,6,9,10,15],theori:15,three:15,tip:5,togeth:10,top:2,toss:15,toward:[0,9],tradeoff:4,train:[1,2],tree:[7,8],tune:2,two:[6,12],type:10,uncorrel:15,understand:[],uniform:15,univers:[10,12,17],use:[2,15],used:5,using:[1,15],valid:4,valu:[3,9,15],variabl:15,varianc:[4,15],variou:1,vector:[6,10,13],view:[1,8],visual:[2,7],wai:7,week:14,weekli:14,what:[1,15],when:[],which:[2,15],why:15,wisconsin:5,write:[0,9],xgboost:8,your:[1,8]}}) \ No newline at end of file diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb index 0dbf33b5b..8d73f71ab 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter1.ipynb @@ -13,38 +13,37 @@ "\n", "\n", "\n", - "Our emphasis throughout this series of lectures \n", - "is on understanding the mathematical aspects of\n", - "different algorithms used in the fields of data analysis and machine learning. \n", + "Our emphasis throughout this series of lectures is on understanding\n", + "the mathematical aspects of different algorithms used in the fields of\n", + "data analysis and machine learning.\n", "\n", - "However, where possible we will emphasize the\n", - "importance of using available software. We start thus with a hands-on\n", - "and top-down approach to machine learning. The aim is thus to start with\n", - "relevant data or data we have produced \n", - "and use these to introduce statistical data analysis\n", - "concepts and machine learning algorithms before we delve into the\n", - "algorithms themselves. The examples we will use in the beginning, start with simple\n", - "polynomials with random noise added. We will use the Python\n", - "software package [Scikit-Learn](http://scikit-learn.org/stable/) and\n", - "introduce various machine learning algorithms to make fits of\n", - "the data and predictions. We move thereafter to more interesting\n", - "cases such as data from say experiments (below we will look at experimental nuclear binding energies as an example).\n", - "These are examples where we can easily set up the data and\n", - "then use machine learning algorithms included in for example\n", - "**Scikit-Learn**. \n", + "However, where possible we will emphasize the importance of using\n", + "available software. We start thus with a hands-on and top-down\n", + "approach to machine learning. The aim is thus to start with relevant\n", + "data or data we have produced and use these to introduce statistical\n", + "data analysis concepts and machine learning algorithms before we delve\n", + "into the algorithms themselves. The examples we will use in the\n", + "beginning, start with simple polynomials with random noise added. We\n", + "will use the Python software package\n", + "[Scikit-Learn](http://scikit-learn.org/stable/) and introduce various\n", + "machine learning algorithms to make fits of the data and\n", + "predictions. We move thereafter to more interesting cases such as data\n", + "from say experiments (below we will look at experimental nuclear\n", + "binding energies as an example). These are examples where we can\n", + "easily set up the data and then use machine learning algorithms\n", + "included in for example **Scikit-Learn**.\n", "\n", "These examples will serve us the purpose of getting\n", "started. Furthermore, they allow us to catch more than two birds with\n", "a stone. They will allow us to bring in some programming specific\n", - "topics and tools as well as showing the power of various Python \n", - "libraries for machine learning and statistical data analysis. \n", + "topics and tools as well as showing the power of various Python\n", + "libraries for machine learning and statistical data analysis.\n", "\n", - "Here, we will mainly focus on two\n", - "specific Python packages for Machine Learning, Scikit-Learn and\n", - "Tensorflow (see below for links etc). Moreover, the examples we\n", - "introduce will serve as inputs to many of our discussions later, as\n", - "well as allowing you to set up models and produce your own data and\n", - "get started with programming.\n", + "Here, we will mainly focus on two specific Python packages for Machine\n", + "Learning, Scikit-Learn and Tensorflow (see below for links etc).\n", + "Moreover, the examples we introduce will serve as inputs to many of\n", + "our discussions later, as well as allowing you to set up models and\n", + "produce your own data and get started with programming.\n", "\n", "\n", "\n", @@ -286,7 +285,17 @@ "[gallery](https://matplotlib.org/gallery/index.html) of examples. In\n", "this example we plot our original values of $x$ and $y$ as well as the\n", "prediction **ypredict** ($\\tilde{y}$), which attempts at fitting our\n", - "data with a straight line.\n", + "data with a straight line. Note also that **Scikit-Learn** requires a\n", + "matrix as input for the input values $x$ and $y$. In the above code we\n", + "have solved this by declaring $x$ and $y$ as arrays of dimension\n", + "$n\\times 1$.\n", + "\n", + "In the code here we have also made a new array for $x\\in [0,1]$. Our\n", + "prediction is computed for these values, meaning that they were not\n", + "included in the data set used to *train* (or fit) the model.\n", + "This is a recurrring theme in machine learning and data analysis. We would like to train a model on a specific given data set.\n", + "Thereafter we wish to apply it to data which were not included in the training. Below we will encounter this again in the so-called *train-validate-test* spliting. We will typically split our data into different sets, oen for training, one for validation and finally, our data from the untouched test vault!\n", + "\n", "\n", "The Python code follows here." ] @@ -301,7 +310,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
    " ] @@ -327,6 +336,7 @@ "y = 2*x+np.random.randn(100,1)\n", "linreg = LinearRegression()\n", "linreg.fit(x,y)\n", + "# This is our new x-array to which we test our model\n", "xnew = np.array([[0],[1]])\n", "ypredict = linreg.predict(xnew)\n", "\n", @@ -451,7 +461,7 @@ "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
    " ] @@ -492,8 +502,7 @@ "Depending on the parameter in front of the normal distribution, we may\n", "have a small or larger relative error. Try to play around with\n", "different training data sets and study (graphically) the value of the\n", - "relative error. Note also that **Scikit-Learn** requires a matrix as input for the input values $x$ and $y$. In the above code we have\n", - "solved this by declaring $x$ and $y$ as arrays of dimension $n\\times 1$.\n", + "relative error.\n", "\n", "As mentioned above, **Scikit-Learn** has an impressive functionality.\n", "We can for example extract the values of $\\alpha$ and $\\beta$ and\n", @@ -517,18 +526,18 @@ "output_type": "stream", "text": [ "The intercept alpha: \n", - " [1.84860939]\n", + " [1.89992521]\n", "Coefficient beta : \n", - " [[5.2571699]]\n", - "Mean squared error: 0.25\n", - "Variance score: 0.90\n", + " [[5.12908963]]\n", + "Mean squared error: 0.20\n", + "Variance score: 0.92\n", "Mean squared log error: 0.01\n", - "Mean absolute error: 0.40\n" + "Mean absolute error: 0.36\n" ] }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
    " ] @@ -721,7 +730,7 @@ "outputs": [ { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAA0L0lEQVR4nO3deXxU5dXA8d+ZmawQEnaRRRaBCggRIgKiRQFXlPK2INhWUFpESltrF9Fqi1Vffa1LtVoVW/e14FKroiKibKIGZAn7FiAQQgghgSRkm+f9407CJJlJJpk1k/P9fOYzM/feufe5k8yZZ859FjHGoJRSKrrYwl0ApZRSgafBXSmlopAGd6WUikIa3JVSKgppcFdKqSjkCHcBADp06GB69uwZ7mIopVSzsnbt2qPGmI6e1kVEcO/Zsyfp6enhLoZSSjUrIrLP2zpNyyilVBTS4K6UUlFIg7tSSkWhiMi5e1JeXk5WVhanTp0Kd1GiTnx8PN26dSMmJibcRVFKBUnEBvesrCySkpLo2bMnIhLu4kQNYwx5eXlkZWXRq1evcBdHKRUkEZuWOXXqFO3bt9fAHmAiQvv27fUXkVJRLmKDO6CBPUj0fVUqApTkwJKL4d9trPuSnIDuPqKDu1JKRa2Vk+HoV1BxwrpfOTmgu9fgXo/Dhw8zdepU+vTpw7Bhw7jqqqvYsWNHva/p2bMnR48erbP8mWee4eWXX/b52Ha7ndTU1OpbZmYmo0aNAiAzM5PXX3+9cSejlIos+evBVFiPTYX1PIAi9oJquBljmDRpEtOnT+fNN98EYMOGDeTk5NCvX79G72/27NmN2j4hIYH169fXWLZ69WrgdHC//vrrG10OpVSEaJtq1dhNBYjDeh5AWnP3YtmyZcTExNQIykOGDOGiiy7iiy++YMKECdXL586dy4svvlj9/KGHHuLcc89l+PDh7Nq1C4D58+fz8MMPA7Br1y7GjRvHkCFDGDp0KLt37/apTK1btwZg3rx5rFixgtTUVB577DF/T1UpFQ6jF0KHkeBIsu5HLwzo7ptFzf2e/25my6HCgO5zwJlt+PM1A72uz8jIYNiwYU3ad3JyMps2beLll1/m1ltv5YMPPqix/sc//jHz5s1j0qRJnDp1CqfTWWcfJSUlpKamAtCrVy/efffd6nUPPvggDz/8cJ39KqWakYTOMH550HbfLIJ7czNt2rTq+9/85jc11p04cYKDBw8yadIkwOpQ5ImntIxSSvmqWQT3+mrYwTJw4EAWLVrkcZ3D4ahR267dZty9qaE2O1RKhUODOXcReV5EjohIhtuyt0RkveuWKSLrXct7ikiJ27pnglj2oLr00kspLS1lwYIF1cs2btzIihUrOOuss9iyZQulpaUcP36cpUuX1njtW2+9VX0/cuTIGuuSkpLo1q0b7733HgClpaUUFxc3qmxJSUmcOHGiCWellGopfLmg+iJwhfsCY8x1xphUY0wq8Dbwjtvq3VXrjDGNayISQUSEd999l88++4w+ffowcOBA7rjjDs444wy6d+/OlClTGDRoEFOmTOG8886r8dr8/HwGDx7M448/7vGC5yuvvMITTzzB4MGDGTVqFIcPH25U2QYPHozdbmfIkCF6QVUp5ZEYYxreSKQn8IExZlCt5QLsBy41xuz0tl1D0tLSTO3JOrZu3co555zTmN2oRtD3V6nmT0TWGmPSPK3ztynkRUCOMWan27JeIvKdiHwpIhfVU6hZIpIuIum5ubl+FkMppSJYkIca8MTf4D4NeMPteTbQwxhzHnAb8LqItPH0QmPMAmNMmjEmrWNHj1MAKqVUdAjyUAOeNDm4i4gD+B/graplxphSY0ye6/FaYDfQ+O6cSikVTYI81IAn/tTcxwHbjDFZVQtEpKOI2F2PewN9gT3+FVEppZq5tqnWEAMQlKEGPPGlKeQbwFdAfxHJEpGZrlVTqZmSAbgY2OhqGrkImG2MORbA8iqlVPMT5KEGPGmwE5MxZpqX5TM8LHsbq2mkUkqpKkEeasATHTjMi7y8vOrhds844wy6du1a/bysrCwgxygvL2fevHn07duXoUOHMnLkSBYvXlzva2bMmOGx52x6ejq/+tWvAlIupVTz1yyGHwiH9u3bV4/tMn/+fFq3bs3vfve76vUVFRU4HP69fXfffTfZ2dlkZGQQFxdHTk4OX375ZZP2lZaWRlqax+auSqkWSGvujTBjxgxmz57NBRdcwB/+8Icaw/gCDBo0iMzMTABeffVVhg8fTmpqKjfffDOVlZU19lVcXMxzzz3H3//+d+Li4gDo3LkzU6ZMAU4P7wuwaNEiZsyYUf38s88+Iy0tjX79+lWPDOk+DPHJkye58cYbOffccxk8eDBvv62ZMqVamuZRc7/1Vgj0CImpqfC3vzX6ZVlZWaxevRq73c78+fM9brN161beeustVq1aRUxMDHPmzOG1117jhhtuqN5m165d9OjRgzZtPHYDqFdmZibffPMNu3fv5pJLLqkeM77KvffeWz3sMFjDISilWpbmEdwjyOTJk7Hb7fVus3TpUtauXcv5558PWGOzd+rUKWBlmDJlCjabjb59+9K7d2+2bdtWY/1nn31WPXsUQNu2bQN2bKVU89A8gnsTatjB0qpVq+rH3ob+NcYwffp0HnjgAa/7Ofvss9m/fz+FhYUea+/uQwXXN6Swp+dKKaU5dz/07NmTdevWAbBu3Tr27t0LwNixY1m0aBFHjhwB4NixY+zbt6/GaxMTE5k5cya//vWvq1vf5ObmsnCh1f61c+fObN26FafTWWMWJoCFCxfidDrZvXs3e/bsoX///jXWjx8/nqeeeqr6uaZllGp5NLj74Yc//CHHjh1j4MCBPPnkk9UTZw8YMID77ruPyy67jMGDBzN+/Hiys7PrvP6+++6jY8eODBgwgEGDBjFhwoTqWvyDDz7IhAkTGDVqFF26dKnxuh49ejB8+HCuvPJKnnnmmTqzOd11113k5+czaNAghgwZwrJly4L0DiilIpVPQ/4Gmw75G3r6/irV/AVzyF+llFIRSIO7UkpFIQ3uSikVhTS4K6VUFNLgrpRSUUiDu1JKRSEN7vWw2+2kpqYycOBAhgwZwiOPPFKjR6onmZmZvP766yEqoVJKeabBvR4JCQmsX7+ezZs3s2TJEhYvXsw999xT72s0uCulIoEGdx916tSJBQsW8OSTT2KMITMzk4suuoihQ4cydOhQVq9eDcC8efNYsWIFqampPPbYY163U0qpoDLG1HsDngeOABluy+YDB4H1rttVbuvuAHYB24HLG9q/MYZhw4aZ2rZs2VJnWb2KDxvz6UXGvJVk3RcfbtzrPWjVqlWdZcnJyebw4cOmqKjIlJSUGGOM2bFjh6k6h2XLlpmrr766entv24Vbo99fpVTEAdKNl7jqy6iQLwJPAi/XWv6YMeZh9wUiMgBr4uyBwJnAZyLSzxhTSbCtnAxHvwJTYd2vnBzUOQvLy8uZO3cu69evx263s2PHDr+2U0qpQPJlguzlItLTx/1NBN40xpQCe0VkFzAc+KrpRfRR/norsIN1n78+4IfYs2cPdrudTp06cc8999C5c2c2bNiA0+msM3hXlccee8yn7ZRSKpD8ybnPFZGNIvK8iFTNBtEVOOC2TZZrWR0iMktE0kUkPTc3149iuLRNBXF9V4nDeh5Aubm5zJ49m7lz5yIiFBQU0KVLF2w2G6+88kr1NHpJSUmcOHGi+nXetlNKqWBqanB/GugDpALZwCON3YExZoExJs0Yk9axY8cmFsPN6IXQYSQ4kqz70Qv93mVJSUl1U8hx48Zx2WWX8ec//xmAOXPm8NJLLzFkyBC2bdtWPYnH4MGDsdvtDBkyhMcee8zrdkopFUw+DfnrSst8YIwZVN86EbkDwBjzgGvdJ8B8Y0y9aRkd8jf09P1VqvkL+JC/IuI+e8QkIMP1+H1gqojEiUgvoC/wTVOOoZRSqukavKAqIm8AY4AOIpIF/BkYIyKpgAEygZsBjDGbReTfwBagAvhFSFrKKKWUqsGX1jLTPCz+Vz3b3w/c70+h3Palkz8HgS+pOKWUS0mO1bQ6f73VUGP0QkjoHO5SNShie6jGx8eTl5engSjAjDHk5eVpk0ylfFXVh6bixOk+NM2AL52YwqJbt25kZWURkGaSqob4+Hi6desW7mIo1Tw0pg9NBNXyIza4x8TE0KtXr3AXQynV0rVNPd37vaE+NCHuKV+fiE3LKKVURGhMH5oQ9JT3VcTW3JVSKiIkdPa99t2YWn6Qac1dKaUCJQg95ZtKa+5KKRUojanlB5nW3JVSKgppcFdKqSikwV0ppaKQBnellIpCGtyVUioKaXBXSqkopMFdKaWikAZ3pZSKQhrclVIqCmlwV0qpKKTBXSmlolCDwV1EnheRIyKS4bbsryKyTUQ2isi7IpLiWt5TREpEZL3r9kwQy66UUsoLX2ruLwJX1Fq2BBhkjBkM7ADucFu32xiT6rrNDkwxlVJKNUaDwd0Ysxw4VmvZp8ZUjUjPGkDnbFNKqQgSiJz7TcBit+e9ROQ7EflSRC7y9iIRmSUi6SKSrvOkKqVUYPkV3EXkj0AF8JprUTbQwxhzHnAb8LqItPH0WmPMAmNMmjEmrWPHjv4UQymlVC1NDu4iMgOYAPzYGGMAjDGlxpg81+O1wG6gXwDKqZRSqhGaFNxF5ArgD8C1xphit+UdRcTuetwb6AvsCURBlVJK+a7BafZE5A1gDNBBRLKAP2O1jokDlogIwBpXy5iLgb+ISDngBGYbY4553LFSSqmgaTC4G2OmeVj8Ly/bvg287W+hlFJK+Ud7qCqlVEkOLLkY/t3Gui/JCXeJ/KbBXSmlVk6Go19BxQnrfuXkcJfIbxrclVIqfz1U9cs0FdbzZk6Du1Iq+jWUdmmbCuK6BCkO63kz1+AFVaWUanZKcqzUSv56K1A7y+DYWqtWXpV2Gb/89PajF9bcfvTC8JQ7gDS4K6WiT1UOvSqYY8BUWus8pV0SOtcM9lFA0zJKqehTO4duTNSlXRqiNXelVPSoSsdUFJ1eJg5oNwxssVGVdmmIBnelVPSoSsfgtJ6LHTqMtIJ5QuewFi3UNC2jlIoe7ukYAHuilUsPdGBvBp2eNLgrpaJHqJo0NoNOTxrclVLRY/RCKw3jSDqdjgmGZtDpSXPuSqno0VCTxtrt35uai2+berqpZYS2vtGau1Kq5QhUOiVUvxD8oDV3pVTLEah0SjPo9KQ1d6VUyxGFY8h4o8FdKdVyNIN0SqBoWkYp1XK4p1MCdXE1QvlUcxeR50XkiIhkuC1rJyJLRGSn676ta7mIyBMisktENorI0GAVXimlmqwZtFX3h69pmReBK2otmwcsNcb0BZa6ngNcCfR13WYBT/tfTKWUCrBm0FbdHz4Fd2PMcuBYrcUTgZdcj18CfuC2/GVjWQOkiEiXAJRVKaUCJ8ovrvpzQbWzMSbb9fgwUJWs6goccNsuy7WsBhGZJSLpIpKem5vrRzGUUlEnFGO3RPnF1YBcUDXGGBExjXzNAmABQFpaWqNeq5SKcrUn26g9c1IgNIO26v7wp+aeU5Vucd0fcS0/CHR3266ba5lSSvkmyvPhoeBPcH8fmO56PB34j9vyG1ytZkYABW7pG6WUaliU58NDwdemkG8AXwH9RSRLRGYCDwLjRWQnMM71HOAjYA+wC3gOmBPwUiuloluU58NDQYwJf7o7LS3NpKenh7sYSikVWitXwplnQu/eTXq5iKw1xqR5WqfDDyilVDgsXQqXXw5zgpPc0OCulIpOkTwV3scfw4QJVo39pZca3r4JNLgrpaJTpA4vsGgRXHstnHMOLFsGnYMzno0Gd6VUdIrE5pQvvgjXXQfDh8Pnn0OHDkE7lAZ3pVR0irTmlH//O9x4I4wdC598AikpQT2cBnelVHSKlOaUxsD//i/86lfwgx/Af/8LrVoF/bA6nrtSKjpFwvACxsC8efDQQ/CTn8ALL4AjNGFXg7tSSgWD0wm/+AU88wzccgs8+STYQpcs0bSMUipyRXJzxvqUl8P06VZgnzcPnnoqpIEdNLgrpSJZpDZnrM/Jk3DNNfDqq1au/YEHQMTjpp9vy2Hd/vygFEODu1IqcnlrzhipNfojR+CSS+Czz+Cf/4Q77vC4WUWlk4c+3sZNL6bz1Oe7glIUDe5KqcjlrTljOGr0DX2h7NkDF14ImzfDe+/BzJked5NdUML1z33NP77YzdTzu/PUj4MzzbQGd6VU5PLWnDEcHZTq+0JZtw5GjoRjx6zOSRMmeNzFsm1HuOrxFWQcKuDRKUN48IeDiY+xB6W42lpGKRW5vDVnbJt6eqamUHVQqv2FkrvCqsGXzoVpM6F9e2vMmO99r85LyyudPPzJdp5dvodzurThyevPo0/H1kEtrtbclVLNTzg6KLmniKq8twp+MNUaAGz1ao+B/cCxYqY8+xXPLt/DT0ecxbtzRgU9sIPW3JVSzVE4OiiNXmilYnJXgAE+BN5wwgA7LF8Oycl1XvJxxmH+sGgDxsBT1w/l6sFdQlZcDe5KKeWLqi+UxaPhr6thmYERAneNqBPYSysqeeCjbby4OpPB3ZJ5ctpQerRPDGlxNbgrpZSvjh+Hh+1WYJ8UCzcPh4sX1dgk82gRc99YR8bBQmaO7sXtV3yPWEfoM+BNDu4i0h94y21Rb+BPQArwcyDXtfxOY8xHTT2OUkpFhD17rFYwu3ZZQ/dOn15jtTGGhelZzP/vZmLsNp67IY3xA4IzVrsvmhzcjTHbgVQAEbEDB4F3gRuBx4wxDweigEopFVAlOVbuPH+9dZF09EIr5VKf1autER0rKuDTT2HMmBqr84vKuOOdTXy8+TCj+rTnkSlD6JKcEKQT8E2gfiuMBXYbY/YFaH9KKRUcje0A9eabcOmlVl59zZo6gX3lzqNc8fhylm7L4c6rvserMy8Ie2CHwAX3qcAbbs/nishGEXleRNp6eoGIzBKRdBFJz83N9bSJUkoFnq8doIyB++6DadOsmZPWrIF+/apXl1ZUcv+HW/jJv74mKT6Gd+dcyKyL+2CzeR5HJtT8Du4iEgtcC1Q1NH0a6IOVsskGHvH0OmPMAmNMmjEmrWPHjv4WQymlfOPLDE3FxXD99XD33XDDDbBkidVJyWVHzgl+8NRqnluxl5+OOIv/zh3NoK51m0KGUyBay1wJrDPG5ABU3QOIyHPABwE4hlJKBUZVe3X3nLu7zEyYNAk2bLBGdLz99upRHY0xvPzVPv73o620jnPwr+lpjD0nfBdN6xOI4D4Nt5SMiHQxxmS7nk4CMgJwDKVUS9WUC6D1qa8D1BdfwOTJ1njsH34IV15ZvepwwSluf3sjX+7I5ZL+HXnoR0PomBTX9HIEmV9pGRFpBYwH3nFb/JCIbBKRjcAlwG/8OYZSKoo0ZajeUIwAaYw1gfW4cdCxI3zzTXVgN8bwzrosLnvsS77Ze4y/TBzI8zPOj+jADn7W3I0xRUD7Wst+6leJlFLRqypQm4rTgbqhYQSCPQLkqVMwZ441v+m118Irr0CbNgDknijlznc3sWRLDmlnteXhyUPo2SH4k1sHgvZQVUqFTp2RFVdZtfj60i3BHAHy0CH4n/+Br7+GP/0J/vzn6unwPtyYzV3vbaKorJI/XnUON43uhT1CWsL4QkeFVEqFTp2RFZ0Np1uCNQLkl1/CsGGQkQFvvw333AM2G/lFZcx9fR2/eH0dPdol8tGvRvPzi3s3q8AOWnNXSoWSe0uVymIwldby+tItgR4B0umEv/4V7rwTzj7bauY4aBAAS7bkcMc7mygoKeN3l/Vj9vf74LA3zzqwBnelVOi4B+olF4dhwo18q936Bx/AlCnWPKdJSRwvLuMvH2zhnXUHOadLG16+aTgDzmwT/PIEkQZ3pVR4NNTePNDS061mjgcPwhNPwNy51rDsGw8x//3NHC8u55eXns0vL+0bllEcA02Du1IquLy1Uw/VhBvGwLPPwq9/DZ07WxNrjBhBdkEJd7+XwWdbjzC4WzIv33RBs6+tu2v+X09KqchWXzv1prR7b4yTJ+GnP4VbboFLLoF163AOv4BX1+zjskeXs3LXUf541Tm8c8uowAf2YJ9bAzS4K6WCq7526sHsoPTdd5CWBq+/Dn/5C3z0EXtMPFOfW8Nd72VwbrdkPrn1Yn5+ce/gXDQNReeremhaRikVXPW1Uw9GByWnEx5/HObNswb7WrKE8jGXsODLPTy+dCfxDhsP/XAwk9O6IRLE5o3B7nzVAK25K6WCq7526r6M0NgYOTlw9dVw221w+eWwcSPf9R3KtU+u4q+fbGfs9zrx2W3fZ8r53YMb2CHw59ZIWnNXSgVXfRdOA9li5pNPrKnvjh+HJ5/k+Iyf8X+f7ODNb/fTKSmOZ34yjCsGndH0/TdWqFsD1aLBXSllCfToi74IRIuZ0lKrQ9Kjj8LAgZhPP+Xt8nY88OhyjpeUc9OFvfjN+H60jgtxuPN2biF6nzUto5SyhPkCYJNs3QqjRlmB/ZZb2PnB51y3uojfLdzAWe0T+e/c0dw9YUDoA3t9QvQ+R9AZK6XCytsFwHDU6BtSddH0zjshMZHSfy/i0bie/OvZNbS2FfPg91YyZcpd2BIjsN16iC60as1dKWXxdgEw0mr0e/dabdZvuw0zbhxfvLOMS3e35dnVh5mUsoyl/W9matyz2FZNCW85vQnRhVatuSulLN4uAIa5SV81Y6yxYG67DUTI+ds/+H3roSxffID+nZNY2G8+58enu7YlfOVsSIgutGpwVyqa+JNC8XYBMJjjqfvq0CH42c9g8WIqvj+Gf9zwR57YVUbCsePcdfU5TB/Vk5jPE+CoI7zl9EWIhl3Q4K5UNGnKTEcNCWeTPmPgjTesQb5OnWLdb+dzc9II8naWMvX87vz2sv50aB0X/nJGIL+Du4hkAieASqDCGJMmIu2At4CeQCYwxRiT7++xlFINCEYKpbE1zapfD8fWgdgBJ7Q9r/EXYg8csKa/++ADTp6Xxm1X/4ZPy5NJ65TEi9cOZFDXZP/KGeUCdUH1EmNMqjEmzfV8HrDUGNMXWOp6rpQKtjD3igRO/3qoLIKKQqg42bgLsU4nPPUUDBiA8/PPeecntzF43N1sSuzM41NTWTh7ZN3AruoIVlpmIjDG9fgl4Avg9iAdSylVJRJSE+6/Hqo09Cuiqra/aS38ywFbCskcNpqfjZjJgZTOzL24N7PH9CExVjPJvgrEO2WAT0XEAM8aYxYAnY0x2a71h4E6v8VEZBYwC6BHjx4BKIZSKiJSE+4XYKs09Cti2Q/hha8w7zkpi43h/olzeLn/lVw9+ExeuPJ7dG+XGOxSR51ABPfRxpiDItIJWCIi29xXGmOMK/BTa/kCYAFAWlpanfVKqWaq6teDp5y7J6tXY27+CslysmxQGr8fcyu9Oh3mnekXMrRH25AWPZr4HdyNMQdd90dE5F1gOJAjIl2MMdki0gU44u9xlFIRrClNMHNz4fbb4YUXyEtO4fc/+jX7zunC/Wc+zeW9bUiP34ai5FHLrwuqItJKRJKqHgOXARnA+8B012bTgf/4cxylVASob2ahxvRirayEp5/G2bcflS+/wjMX/JAf/OIFLr0qn08GzuOKPjbkopbdjDEQ/K25dwbedY2L7ABeN8Z8LCLfAv8WkZnAPiBC+wErpXxWXxt6X5tgfvst5bN+Rsz6jXzTYxD3TpnFJRMvZPHYc0mK/wHwePDPo4XwK7gbY/YAQzwszwPG+rNvpVSEqS+AN9SLNS+P0tvvIPb5f5LfKoX7r/ktcaPK+GeX++kS2w/itX16oGm7IqWak3CO0FhfAPfWBLOigtJ/PI3z7j8Rc6KQ54ddy9YJZ/Or7m/SJ/6gtU1+cWjK38JocFeqOQnG8AK+qq8NvYcmmGUffEjRL39D28ydrOk+iI9/fwdTbprAzM0T4GiO1Yg6kseAaeY0uCvVnIRzhEYf29CXZ2wmd9ZczvzqCw6ldOGFWx5gzO2zmH9WO2uDtgHuaBWJ481HAA3uSjUnkTBCoxdlh4+Q+cvf0/udV2ntiOfFSXPof++d3Dawa80NA93Ryp9fM1H8xaCTdSjVnIxeCB1GgiPJuo+AkQ9Lc/eSft1oSnv2pM/br/LpyAms//wbpr/9JCNrB/Zg8OfXTKRNRBJAWnNXqjmJhOEFXE6VlJL+l7/R9x8PkVZ4jHVn96f8+hiuvDAfuXBg8AtQVeuuKDq9zNOvmfpq55EyEUkQaM1dqXCrr3NQBCouLWfJfU+T3aMvox+cR2GbVmy6tTfnzd/OBf0zkKOrQnMuVbVunNZzsXv+NVNf7TwSRtEMEg3uSoVbpKUGvHzZFJSU85+HFrCn9zmMv3sOMbZytj/xFH3/dRbnDt+P1ZcRwBmac6k9+qQ90fpVUztnXl/tPALTXIGiaRmlwi3SUgO1LlAe/uxGFu+aTv8nH2LinnXkJSezd0YXeo07Ap3ehNGLTqc9KovBVFr7Cfa5+Hpxub7tIijNFWga3JUKN0/Bxz1PnDzA2q5gS2hadLi+bHae6s7H6SMZ8vFObtw7lZNJKWRP60KXy7JpH1twelv3ALnk4tC15vF17Ppwj3EfphY5Ykz4R9tNS0sz6enp4S6GUoHnywfb0zbutWd34rDSB0GqbRpj+Hbhj1m6NJ4LP13PxZnfUdQ6kfLf3kHK726Fr66qGbxrlyWKmxY2yNu51/7CC+DfT0TWus2AV3OdBnelgsjTB9tTTbJ2APx3Gytv7U3HiwIaOCudhiVbclj+/Dtc9c4CRu/bQHHreJjSg8T/WwwdelsbhiJ4h+sLwt/jegvitf+WjiSYUhiQImtwVyqU3IOEew4arA927TSMp5qce6DwJEA1wIKSchZ+u58tL7/N5CWvMnL/JkradcBx++3EzJ0DiWGYASmINd2gHtdbEA9TzV1z7koFmreUCnZAIHfF6UXeLjq61+6rcu55Xzf8Onf11ER3HTnJKyt2UfTam9y4aiE/O7KHU53OoPKRR0mYfXN4gnqVcF1g9ve43i7chinnr8FdqUCrM0G0DRytALFq8u68XXT01IqjsRcra7V6ca6YzJdnLeK1z7fS9Z03+Nm379G9IIdTZ/eDB58n/vrrIS6u0acbcOEaYsHf43oL4mFqkaPBXalA85Z2+XeburX5xrSt9rUGWFVjd/1COFGZwDv5Y3l33SWM+epWHv7uQ1KKCykfMRLuWED8hAlgi6AuL+Fq3eLvcSOsWaUGd6UCqSQHnGVgnFaPyXbDTgcJX3Lt9fE1eLhq7JuK+/DasavYvr0H16Uv4a0t84irKMd5zTVw++3EXHhhk04x6MIVJN2PGwWtfjS4KxVIKyfDsbVYXeIdYIs9HRRCUCMtKq3gvzvb8sbhB+myKY+Za//D+Qe2UBkr2L8PTBuGbeb7AT9u1AnnuPkB0uTgLiLdgZex5lE1wAJjzOMiMh/4OZDr2vROY8xH/hZUqWahvotyQayRbs0u5PWv9/P5qq1c8207nvvufjoV5uPsIHA92McYaAU4dgTl+FEn0noNN4E/NfcK4LfGmHUikgSsFZElrnWPGWMe9r94SoVRU36ah/Bi4MnSCj7amM2b3+yDNWv48cZP+NPWFcSUl2JSk2FcPLZhMWBKInL894gWrL9jCNM9TQ7uxphsINv1+ISIbAVCMHizUiHSlJ/mQU69OJ2GNXvzWJSexapvd3LFhqU8nPEpvQ/vxbRujdw0A+bORQYNsl7gKZi4i4LcclAE6+8YwnRPQDoxiUhPYDkwCLgNmAEUAulYtft8D6+ZBcwC6NGjx7B9+/b5XQ6l/NZQB6QA9SxsrP15xSxal8Xb6Qc4Y/M6bsj4lKu2riCmrBRz/vnIrFkwdSq0bt24HYerw1BLFeDeqkHtxCQirYG3gVuNMYUi8jRwL1Ye/l7gEeCm2q8zxiwAFoDVQ9XfcigVEPWN6VLfT/Mg1ICLSiv4aFM2i9ZmkblxB5O2fMFbO76k26G9mKQkZOZN8POfI+ed1/SDeMsta40+OEKYtvMruItIDFZgf80Y8w6AMSbHbf1zwAd+lVCpUPLWAamhn+YB+rldWlHJl9tz+c+GQ6zesI+Lt6zid9u/JG3XWsQYGDkS7r0Lue46aNXKepE/gdhbsImC1iIRKYRt+P1pLSPAv4CtxphH3ZZ3ceXjASYBGf4VUakQampbdD9aV1Q6DWv25PGf9Qf5ZNMhBuxYz7TtX/Do1pXEnSrG9OyJ3H03/PSncPbZdXfgTyD2FmwaOh+t2dfP2/sTwjb8/tTcLwR+CmwSkfWuZXcC00QkFSstkwnc7McxlAoObx++ptasGvlz2xjDdweO8/76Q3y48RBdd2xi0o6V3LFrNW3zcqy0y4+nwQ03IKNH19+DtHYgzl1lnZ8vwdZbsGnofLRmX78IeH/8aS2zEhAPq7RNu4p83j58DdWs/PhScDoN6/bn83HGYT7OyCZlWwYTt6/g412raX80GxMbi1x+OUybhkycaA3eVZIDS8fUX0Num1pzMDKc/geThs4nCtqBB1UEvD/aQ1W1TE398DXyS6G80sk3e4+xOCObTzIO02nXFibs/Iq3d62m85EDGIcDGT8ernvACugpKb4dz93ohfBe15otezydT2NSKQ19yYVrcK/mIgLeHw3uqmVq6ofPhy+FU+WVrNp1lMUZh/ki4yD9tn/HVbu/5ra939AuLwdjsyGXXgrX/QmZNAnat/freCR0hg6jGj6fQKYKwj11XaSLgPdHg7tqmRrz4auq8R5bB5WnTi93C6JHTpzii225LN2Ww7rNBxi2/Vuu3vM1f9n9LYlFJzDx8VbKZeJEZMIE6NjRt3L6+iXky/kEMlXQ2AuDLe0CbASMEKkzMano1diA4sscmC5OI2wuH8zSzi+wbEc+RZu2M2ZPOpdlfsPQA9txVJRj2rdHrrkGJk6E8eNPN11szPEhcEGxznnYwNEacELb84IbcLWzVFDoNHuqZfIWUBo7kbGrV+GJygRWnxzCshPn89XRQfTbfYAxe9IZn/ktnY7nAWC6gozsCnNfgwsvBIfrx7EvXzTBDoDV47yvwhq10k2wA24Q5xFtyXSaPdX8NRQcPa331kTQW+651vYVxzayYd8xluf/gjUHkzF7YMS+DP5n/1LuO/gUjkonJsGODKiEIcBgkI6AoxC+//2a5fcl3x3sFhZVqQJPk28Hu0VHBFxgbGk0uKvmwVNwdM8zV01h577eWxNBL0HUpKSy99A+Vhaey8rCVAp3t2HIM79l1L6N3JKVQXx5GUaAnjZkUhdIS0LO3GlNjVrFW+DyJXCHKgC6H6dKsANuBFxgbGk0LaOaB08/6z0FKff11+ys20TQ7XXGWcH+8m58LdeSHnMjx1Z+xVnbNzJi/yZGHNhEUqk132nlOQOwj70ULr3UqpG3a+e5TAAdL2p6yiVQ1wga4n6BWOyEJOfeXDXlF2MI30PNuavmp/aHxllmzXDkHhzz19cNrlAzeLoFVYOD3YlX83X7+9j05SIqNhXQe38Ww7K2MuTwDuIrygEo79kLx7ixyNixMGYMnHGG5zI2Jkde+3zS/gHpc/wLCv4cv7FfBC2llUttDb3HYb5QrMFdNT+1PzTthllT1rnXNrGdTsVgtwb4wlQHobKYjmzeu5f1nz9O9uYSKvfY6Ld/L2kHt9LnWBYATruN0h4xxPcvRfrbYEQaTP3atzL6E/gCERQac5GyqccLZfCKxC+Sht7jMF8o1guqqvmpnaMu2GJ9aGo05zsd0E1KKtmDX2P9ITiwKp3yl/5G222bGHhoJ9cf2UtcpVUrL22TQuWIkZgxc5ALL8S2/2oSbCddB3WCbavvZfSnLXMgLp42Jkff1OOFsht9BIzHUkdD73EEXyjW4K7Cz1ONzduHxhVs8iuS2JZ/Foczu1BccD6xWzLoc2A8lx7ZS3xFGQClcTGc7NWeimtnEPf9S2DYMOLOPrvmIFxLzmvah9PfWmYggkJjLlI29XihDF4RMB5LHQ29xxF8oVjTMir8PP30d/vQ5MaNYI/z1+Sv3YlzzQe02Z/HWUey6V5QPXUAp+ITKeg/ENv559O23ec4Ou2EzpVgD1Iu2t90RX0XNSHw6YnmkHPXjk6Npjl3FdlcectThQ4O7e9IflYHyk6NIHb3Ttpm7aVr3kHiKq0aXaXYONKpA0Vd44np25b21/ya1kMHwv7fQsEGSB4Aed9So5NOMPKggcq1egpo0DKDnC9fJJGYlw8jzbkr3wT6g1OSA8snuoItYE+AE1BW1J88+wwKd2ZTvm0L8dtT6JBrSCk+SW+ygWzKbVs43P5MCjq3ZvugXrQ6q5gz+hylVfcKunTpX7fFwrE1VjDM83AxNBiphEClK7ylIiItPeFJoP9ffLmGEYq8fJR8gWjNvSXwtRmer931q15fO6Uw6i04IbBrA87PfkPhjgMU5cTjPGIj9mg5ycdOEl9WXqNoua1SONS+I4WdWuHsIiR2d9LpnLac2e47YhzGrTWMG3sraDf0dHnyv4OKk3hmg0mHAt9ePFC1zEipuTflPQhHGiUUrVOaUXpI0zLRyv0DmTzAWlawpW4Ad++9KQ6wJ0JFEVDVuccOjgSoLKnb4aeqhUrWaiiohAI7HI+Fo6WUHrVx6mgczmOCI7+SxIJS7M6aY5YUxcRzILkzh9p2pLBdIhUdHDg6V9D6jGLOPPMovZMPEW8rc3uFDcTmuWMSnC5/7fPx9CXg6wczWB/mpnZcgtDXHJvyHoS6GWBJDvy3H1S4jhGswNuMxsHRtExz0lDvQfdg4B603dMRuStg8bme928qoKwQSoCiqlslFJ6EQqAATAFUFDqoPGmHua1xFJTgKKsK2pVYLwanI4a8pGQOJ3Xg8JntyfteMmVndMe0OkpC+yJSOhVyZvuj9I4/RH/HPsTTvF012K39G6fn1WJ3dV76rmbawlSe7tRU+0vOvfWCt9ppsFpp+DoWu6fg1FDACnTqoCnvQaibAa6c7KqUuNgTg9M6pbHnFaFpnKAFdxG5Angc6xP7T2PMg4E+xqHjJTyxdCcpibG0TYwhJTHG9dh6npwYQ0pCLLGOeuaf9EednHL86fG+3R87Eq375IHWfcEWKwg5y09/iByJVlCrLOV0jdoldwW8e6b1uNIJp6h7KwFK3R4Xud1OgikCZ5ENisBW7ES8/GCrFBt5rZI5mphi3TqlkNczmbzEFE4mxWPaCI52ThI6nKJ9cgHd4o7QPSaHAbF7SLafrCeA26gzEmFtjlana2XVy9rg3jHJY/qo3VDfam/e8rXBClLBDH6Bzj03payhbgaYv56anw0TnCDa2POKxPb5BCm4i4gdeAoYD2QB34rI+8aYLYE8Tt7RQ3zx3SYKyuKoMA7EgGDAGART/by1FJNiKyTFfpJW9lKSKCLRVkIrexlJnCDRdopW9nJay0kSpYRWjgoSnEXEShmxdjtxlBIrlcRWniDOWU4MNmIqihGnsf7Xqm7OYqjAimGVxW7rTlrLKr6Gclw31+Oyqucn3R6DKQdnmWDKBMpAToGcMti8ZCtqq7DZOBmfSGF8K47HJ3Esrg0F7ZM43q01BXGtKYhvTWG8dV+eaMeZZMOW7KR1Ugmd4vI5IyaPTo48usYcYWjMNjrHHCPBVtq4P5DYrRmC6hkX3drObTiBGmxwzY66H+CmBJWSHDi6+nTayb126ml/gaiNBTP4BfrXRlPKGuoJKUL1S6Gx5xWJ7fMJXs19OLDLGLMHQETeBCYCAQ3u5y7+EWvuXx/IXYaME6E0JoZSR6x1s1uPTzliKbXHUuKIo9QRQ2mStawoNpHi2HhOxiZQHJNg3cfGUxSbQFFsAsUxcZh4sMcbbIlOYuPLSbKX0MZ+kiR7MSn2E7RzFNDOUcjZ9izaOU7QPr6CtvbjxDj9ySfWkyO3J9b8kLgHEE/pE/caUFXA9xRQmxJUVk6ueT0BTgcHT/tz/yJqam0smMEv0IEuAmYOalCkdhiK0F6qwQruXYEDbs+zgAvcNxCRWcAsgB49ejTtKDE74UdVO3TdvD2uvcxtXSVCmYnhlImllFhKTSzlxkE5dipwUI6DCuyUSQxlthjKJIZSWwylEkupxFBhd1AhdirsdiptdirERoXdToXNQYXNTrnNToU4qLDbMLF2nA5B7AaHzYkNJw6pxC6VOKjEJtbzGKkgwVZKgpQSZyujra2QM225xEspCbZS4m2ueymjtb2Y1rYSbGKzUhum8nS+vnYqCKBgT4MzDdVhbw0pA2u2IXcf88XTBBC1/8kbCiDBrunWYKt//xFaG6sWqYEumCL1CyhC/xZhu6BqjFkALACrtUyTdtJnKLRZ0fB2DbBjSKCMBMoa3tgrm/ecO1B/vtl2OucuDqovoro3Oay6eCoCKUPAllwrYG8JTPqgan/HMxq+oOt+PPfrDyLQLq3x/+ShrOl6+1XgbfsIqY1Vi9RA1xJF6N8iKE0hRWQkMN8Yc7nr+R0AxpgHPG3f5KaQvl7Q9OdxVdCtGoHQfX1VELv4P763ow5EMFaNF6qx0pUKoZC3cxcRB7ADGAscBL4FrjfGbPa0vbZzV0qpxgt5O3djTIWIzAU+wWoK+by3wK6UUirwgpZzN8Z8BHwUrP0rpZTyLki9e5RSSoWTBnellIpCGtyVUioKaXBXSqkoFBFD/opILrCvgc06AEdDUJxI1ZLPX8+95WrJ5+/LuZ9ljOnoaUVEBHdfiEi6t/acLUFLPn8995Z57tCyz9/fc9e0jFJKRSEN7kopFYWaU3BfEO4ChFlLPn8995arJZ+/X+febHLuSimlfNecau5KKaV8pMFdKaWiUMQFdxG5QkS2i8guEZnnYX2ciLzlWv+1iPQMQzGDwodzv01EtojIRhFZKiJnhaOcwdLQ+btt90MRMSISNU3kfDl3EZni+vtvFpHXQ13GYPLhf7+HiCwTke9c//9XhaOcgSYiz4vIERHJ8LJeROQJ1/uyUUSG+rxzY0zE3LCGB94N9AZigQ3AgFrbzAGecT2eCrwV7nKH8NwvARJdj2+JlnP39fxd2yUBy4E1QFq4yx3Cv31f4Dugret5p3CXO8TnvwC4xfV4AJAZ7nIH6NwvBoYCGV7WXwUsxpoUdATwta/7jrSae/XE2saYMqBqYm13E4GXXI8XAWNFRGj+Gjx3Y8wyY4xrOijWAN1CXMZg8uVvD3Av8H/AKQ/rmitfzv3nwFPGmHwAY8yREJcxmHw5fwO0cT1OBg6FsHxBY4xZDhyrZ5OJwMvGsgZIEZEuvuw70oK7p4m1u3rbxhhTARQA7UNSuuDy5dzdzcT6Ro8WDZ6/6ydpd2PMh6EsWAj48rfvB/QTkVUiskZErghZ6YLPl/OfD/xERLKw5on4ZWiKFnaNjQvVwjZBtmo6EfkJkAZ8P9xlCRURsQGPAjPCXJRwcWClZsZg/WJbLiLnGmOOh7NQITQNeNEY84hrjuZXRGSQMaa+medbtEiruR8Eurs97+Za5nEb11ytyUBeSEoXXL6cOyIyDvgjcK0xpjREZQuFhs4/CRgEfCEimVj5x/ej5KKqL3/7LOB9Y0y5MWYv1hzFfUNUvmDz5fxnAv8GMMZ8BcRjDawV7XyKC55EWnD/FugrIr1EJBbrgun7tbZ5H5juevwj4HPjuvLQzDV47iJyHvAsVmCPppwrNHD+xpgCY0wHY0xPY0xPrGsO1xpjomFmdV/+79/DqrUjIh2w0jR7QljGYPLl/PcDYwFE5Bys4J4b0lKGx/vADa5WMyOAAmNMtk+vDPfVYi9Xh3dgXT3/o2vZX7A+yGD9URcCu4BvgN7hLnMIz/0zIAdY77q9H+4yh/L8a237BVHSWsbHv71gpaW2AJuAqeEuc4jPfwCwCqslzXrgsnCXOUDn/QaQDZRj/TqbCcwGZrv93Z9yvS+bGvM/r8MPKKVUFIq0tIxSSqkA0OCulFJRSIO7UkpFIQ3uSikVhTS4K6VUFNLgrpRSUUiDu1JKRaH/B/3lCXFtiwzMAAAAAElFTkSuQmCC\n", + "image/png": "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\n", "text/plain": [ "
    " ] @@ -738,7 +747,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.0050000000000000044\n" + "0.005000000000000007\n" ] } ], @@ -1157,7 +1166,7 @@ "metadata": {}, "source": [ "The next step, and we will define this mathematically later, is to set up the so-called **design matrix**. We will throughout call this matrix $\\boldsymbol{X}$.\n", - "It has dimensionality $p\\times n$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit." + "It has dimensionality $n\\times p$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit." ] }, { @@ -1182,9 +1191,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Note well that we have made life simple here. We perform a fit in terms of the number of nucleons only. A more sophisticated fit can be done by including an explicit dependence on the number of protons and neutrons in the asymmetry and Coulomb terms.\n", + "Note well that we have made life simple here. We perform a fit in\n", + "terms of the number of nucleons only. A more sophisticated fit can be\n", + "done by including an explicit dependence on the number of protons and\n", + "neutrons in the asymmetry and Coulomb terms. We leave this as an exercise to you the reader.\n", "\n", - "With **scikitlearn** we are now ready to use linear regression and fit our data." + "With **Scikit-Learn** we are now ready to use linear regression and fit our data." ] }, { @@ -1266,7 +1278,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "As a teaser, let us now see how we can do this with decision trees using **scikit-learn**. Later we will switch to so-called **random forests**!" + "As a teaser, let us now see how we can do this with decision trees using **Scikit-Learn**. Later we will switch to so-called **random forests**!" ] }, { @@ -1310,7 +1322,7 @@ "270 3344 160 110 270 Ds 7.253775 7.253775\n", "\n", "[267 rows x 6 columns]\n", - "0.009883615646716186\n" + "0.009883615646716188\n" ] } ], @@ -1351,6 +1363,17 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "With a deeper and deeper tree level, we can almost reproduce every\n", + "single data point by increasing the max depth of the tree.\n", + "We can actually decide to make a decision tree which fits every single point.\n", + "As we will\n", + "see later, this has the benefit that we can really train a model which\n", + "traverses every single data point. However, the price we pay is that\n", + "we will easily overfit. That is, if we apply our model to unseen data,\n", + "we will most likely fail miserably in our attempt at making\n", + "predictions. As an exercise, try to make the tree level larger by adjusting the maximum depth variable. When printing out the predicition, you will note that the binding energy of every nucleus is accurately reproduced.\n", + "\n", + "\n", "The **seaborn** package allows us to visualize data in an efficient way. Note that we use **scikit-learn**'s multi-layer perceptron (or feed forward neural network) \n", "functionality." ] @@ -1363,18 +1386,6 @@ "editable": true }, "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n" - ] - }, { "name": "stderr", "output_type": "stream", @@ -1454,8 +1465,6 @@ "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n" ] }, @@ -1463,6 +1472,10 @@ "name": "stderr", "output_type": "stream", "text": [ + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", + "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", + " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n" ] @@ -1491,22 +1504,20 @@ "name": "stderr", "output_type": "stream", "text": [ - "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", - " warnings.warn(\n", "/Users/mhjensen/opt/anaconda3/lib/python3.8/site-packages/sklearn/neural_network/_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n", " warnings.warn(\n" ] }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAJjCAYAAADQ22+yAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAABfx0lEQVR4nO3dd3hUZd6H8XsmCU1AFAlFbKyKrmJZUVddRbARQIqKIhYQFRuIDYkoYAEFZAUEdO1gF1YBcUFUFGxYUF+xrGVRVKQkiBipKTPvH2dpC4wZcjIh4f5wzXUxZ87kPL85U575Ps85E4nH43EkSZIUmmhZN0CSJKmisYMlSZIUMjtYkiRJIbODJUmSFDI7WJIkSSGzgyVJkhSy9LJugLQjGzhwIB9++CEA8+bNY/fdd6dKlSoAPPfcc+v//0cuvfRS+vTpw7777rvVdUaOHMlee+1F+/btS9zudXr27MkHH3zAzJkzqVq1amh/V5LKu4jnwZK2Dy1atGDkyJE0adKkrJtSLEuWLKFdu3YceuihnHjiiZx77rll3SRJ2m44RChtp0aNGsXFF1/M6aefzg033MDSpUu58sorOeecc2jRogUXXHABv/zyCxB0zj777DPef/99OnXqRO/evWnfvj2tWrXivffeAyA7O5tHHnkEgCZNmjBq1Cg6depEixYtGDt2LABFRUXcddddnHLKKZxxxhnceuutXHDBBVts3/jx4znmmGPo0KEDjz/+OBt/V/v000/p2LEjbdq0oUOHDsyePTvh8saNG7Ns2bL19193/f3336dt27Z06tSJtm3bkp+fz8CBA+nYsSOtWrUiKyuLjz76CICVK1dy0003cdppp9GqVSvuueceVq1axZFHHsn333+//m9fdNFFvPbaa2HsIknaKjtY0nbs559/ZuLEiQwbNox//etfHHbYYTz33HPMmDGDKlWqMHny5M3uM3fuXLp168akSZM466yzGD169Gbr5Ofns8suu/Dss89y77338ve//521a9cyYcIEvvjiC1566SWeffZZfvrppy22q7CwkPHjx9O2bVtatGjBL7/8wptvvglAQUEBV111FVdddRUvvfQSd9xxB3feeSf5+flbXB6LxRI+Bt9++y1///vfefHFF/niiy/IycnhueeeY+rUqXTo0IGHHnoIgHvvvZe1a9cydepUJk2axMcff8xnn31G+/btmTBhAgA//vgj33//Pc2bN09qP0hSspyDJW3HDjvsMNLTg5dply5dmDNnDo899hjz58/n22+/5dBDD93sPg0aNODAAw8E4M9//jMTJ07c4t8+6aSTADjooIPIz89n1apVzJo1i3bt2lG5cmUAzjnnHJ544onN7jtjxgxisRjHH3886enptGrVinHjxtGsWTO++eYbotEoJ554IgAHH3wwU6ZM4Ysvvtji8j9Sv359dt99dwAOP/xwdt555/Wdv/fff5+ddtoJgHfffZebbrqJtLQ00tLSePLJJwHIzMzk/PPP59prr+W5557jrLPOIi0t7Q+3K0klYQdL2o5Vq1Zt/f/vvvtu5s6dy5lnnsnRRx9NYWEhW5pCufHE+EgkssV1gPWdqEgkAkA8Hl/fmVsnGt1yyP3MM8+wZs0aTj31VCBIxHJzc/n2229JS0tb/zfX+eabb7a6vFGjRpssy8/P3+T6xo/BzJkzGTRoEBdddBEnnXQSjRo14sUXXwQgPT19k7+/aNEiqlSpwj777EPjxo2ZMWMGU6ZMWZ9mSVJpcohQKifefvttunTpQvv27alduzbvvvsuRUVFoW6jWbNmvPjii+Tn51NYWLjF9Ov777/ngw8+YOLEibz++uu8/vrrvP322zRt2pRx48bRqFEjIpEI77zzDgBffPEFXbp02eryWCzGrrvuymeffQbAq6++utX2vfPOOzRv3pzOnTvTpEkTXnvttfWPwTHHHMPEiROJxWLk5+dz9dVXrz9Cs3PnzgwdOpRDDz2UunXrhvqYSdKWmGBJ5cRVV13F0KFDue+++0hLS+Mvf/kLP/74Y6jbOOOMM/j+++9p37491apVo2HDhpudfuGZZ57h5JNPZs8999xkeY8ePbjsssu47rrrGDVqFHfeeSdDhw4lIyODUaNGUalSpa0uv+WWW7j99tupWbMmxx57LHXq1Nli+zp16sQNN9zA6aefTlpaGk2bNuWVV14hFovRo0cPBg0aRLt27SgqKqJVq1brE7bmzZtzyy230KlTp1AfL0naGk/TIGm9t99+m19++YV27doBwXm6KleuTO/evcu4ZSXz8ccf069fP1566aXNhiklqTTYwZK03pIlS8jOzuaXX36hqKiIAw44gFtvvZUaNWqUddO2WZ8+ffjggw8YMmQIRx11VFk3R9IOwg6WJEkq96ZMmcL9999PQUEBXbt25bzzzivT9jgHS5IklWtLlixh+PDhvPDCC1SqVIlOnTpx9NFHJ/z5sNJmB0uSJG038vLyyMvL22x5zZo1qVmz5hbv8+677/LXv/6VWrVqAXDaaafx8ssv06NHj9JsakLlvoO1etE+Zd0ESZJSqmr97/94pRDFFu+fsm2Nm9Bzi79A0aNHD3r27LnF++Tk5Gxy9HFmZiZz584ttTYWR7nvYEmSpIqjS5cudOjQYbPlW0uvgC2eULmsjxi2gyVJkrYbiYYCt6Zu3brMmTNn/fWcnBwyMzPDblpSPJO7JElKKJbCf9vi2GOPZfbs2SxbtozVq1fzyiuvcMIJJ4T8KCTHBEuSJJVrdevW5dprr+XCCy+koKCAs846i0MOOaRM21Tuz4PlJHdJ0o4m1ZPc1y5q9McrhaRy/e9Stq3S5BChJElSyBwilCRJCcUo14NdZcIES5IkKWQmWJIkKaFtPbpvR2aCJUmSFDITLEmSlFBR+T7hQJkwwZIkSQqZCZYkSUrIowiTZ4IlSZIUMjtYkiRJIXOIUJIkJVTkEGHSTLAkSZJCZoIlSZIScpJ78kywJEmSQmaCJUmSEvJEo8kzwZIkSQqZCZYkSUrIn3pOngmWJElSyEywJElSQp4HK3kmWJIkSSEzwZIkSQkVGWAlzQRLkiQpZCZYkiQpIY8iTJ4JliRJUshMsCRJUkJFRMq6CeWOCZYkSVLI7GBJkiSFzCFCSZKUUMzTNCTNBEuSJClkJlhJev2tCPc/lkYkAjVrwIDeheyxe3Dbc5OiTPxXlLVrIxy4f4xbbyyiUqWybW/YtlZ/UREMHpnGR58Gffa/HR3j2iuKiFSQeZFTpkd5YvyG7yMrVkbIyYXpEwpIT4dBw9P4+j9RqlaJ0y4rxrlnVKyDmhPVX3tXaN4ugzq7bfiK26VTjNanVJzH4I/qX+e6funUqR3npmuKyqCVpWtrr/0b+qfz488b1lu4OMIRh8YZeWdh2TU2RDv6c38dJ7knLxKPx8t18Ld60T4p29aatXBiuwzGP1zAng3hiQlR3v8oyujBhcx4M8Loh9MZO7qAGtWh963pHNQ4RrfzKs4LLVH9k6dFeemVKP8YVkgsDl2uSqfruUWcemK5fnptUUEhXHx1Om1bxjirbYxb7kojPQ36XV9ELAbX3JLOOe2KOOHYilc7bF7//B/h6r4ZvPhkQVk3LSX+t/51HnsmyuPPpnFq81iF62Aleu1v7POvIvQekM5jowqol1lGjS1F29Nzv2r971O6vS9/2j1l2/rzHj//8UrlQMoSrHnz5jF9+nQWL15MNBolMzOT448/niZNmqSqCSUWKwLiwTcYiLN6NVT+b0I1ZXoaF5xdxM41g+s3X1dIQQX7vElUfywGq9dAfgHEY1BYuOG2imbs01F23YX1H67//jpCdq8i0tIgLQ2O/2uMV2dFOeHYivUhu87/1v9/X0RJi8Il16TzWx6c3CzGJefHSEsr44aWkv+tH+DDTyK8+0GUs9rGyPu9DBtXShK99tcpKID+d6XTu0dhhexcwY793DfBSl5KOlhPPfUU48eP57TTTlvfocrNzaVfv360bduWbt26paIZJVatGtx8XRFdeqRTqyYUxWDsqKAX9eOCCMuWw5W908n9BQ4/JM61l1WsD9hE9bdtGePVmVFOPSuDoiI45sg4zSpggvPrcnh8fBrPPrSh99zkz3H+9WqUw5oUUZAPM96Mkl5BB9+3VH9REfy1aYxrLy9izVroeVM6O1WD8ztWnPR2nS3Vn7MUho5K4767C/nnixXwk5XEr/11Jk6NUqd2nBbHV7zXPfjcV/JS8jHw+OOPM2nSJKpWrbrJ8osuuogOHTqUmw7Wt99FeHBcGi+MLWCP3eHp56PcMCCd5x4upKAQ3psTZcSgQipXgn53pTHq4TRu7FlxOlmJ6n9gXBq71Irz+sRC1qyFa29J5/Hnolx4TsV6o3n+pSgnHhdj9/obll13RRHD70+j0yXp7FY7eMP99POKefzIluo/s82GfVypElzQsYhnXkirkB8y/1t/QSFk355O7x5F1Kldtm0rTYle++vmWT45IY1+11eMeVdbsqM/92NxE6xkpaSDlZ6eTmHh5i+8NWvWkJGRkYombLP7Hk1j5jvBE+vHBRFOPjG2flL7Oe1jDBuTxvLfIHO3OC2Oj1F9p+C2VqfEePDx8v9ttrj1z3gzGCbLyICMDDj9tBivzSrfHayNaz/xuDhXdivilTeim3WaV66Cay7fMDz82NNR9ti9/H+LL279L70SZf8/xdn/T0HN8Tikl/+nfrHq//LrCD8vijBsTFDwL8sixGKQnw8DbizfX66K+9rfpRZ89W2EoiJoelj5f96Dz32FIyUdrMsvv5z27dtzzDHHUKdOHSAYInzvvfe49tprU9GEbXZltyKu/G/A9sHHEQYMSeeXZUXU3hXeeDvC7vWCN5iTm8V45Y0oZ7SJUbkSvPF2lIMal/83m+LWf+D+cV55I8qRhxdRUAiz3o1yyJ/Lb+cKNq0dIO93+PHnCIcevOl+nfBiGitXwk3XFPHLMnjhpTTu6l/+v8kXt/7/fB/htVlR/n57kOQ+OzGNVhXgKKri1H/oQXGmT9gwZHT/Y0GnoyJMci/uax9gzv9FOOrwWIU5anhHf+5viXOwkpeSDtbpp5/OUUcdxezZs8nJySEej9O0aVN69uxJ3bp1U9GEUBz1lzhdOhVxyTUZZGTEqVkDhg8KPkjPbhfjtzzo3D2DohgcuF+M668s/2+yG0tU/w1XFTH43jTaX5BBNC3O0X+J07VzxXqj+fHnCHV2hYz/edVcfF4RNw9K58yu6cSBy7sWcfAB5b9z/b+2Vv9lXYoYPDKNjt0yKCiEU06McUbrirXvYev17wgSvfYheGwa1Kt4z/l1dvTnvraNp2mQJKmcSfVpGj78ce+UbevIPeenbFulqWLOxJUkSSpDO2DYLUmSkuFRhMkzwZIkSQqZCZYkSUrIowiTZ4IlSZIUMjtYkiRJIXOIUJIkJVQUN49Jlo+YJElSyEywJElSQjHzmKT5iEmSJIXMBEuSJCXkaRqSZ4IlSZIUMhMsSZKUkEcRJs9HTJIkKWQmWJIkKaGYc7CSZoIlSZIUMhMsSZKUUJF5TNJ8xCRJkkJmgiVJkhLyKMLk+YhJkiSFzARLkiQl5G8RJs9HTJIkKWR2sCRJkkLmEKEkSUqoKO6JRpNlgiVJkhQyEyxJkpSQJxpNno+YJElSyEywJElSQjFPNJo0HzFJkqSQmWBJkqSEnIOVPB8xSZKkkJlgSZKkhDwPVvJMsCRJkkJW7hOsjEhaWTdBUhkoisfKugllKi3i92Oljj/2nDwfMUmSpJCV+wRLkiSVriLPg5U0HzFJkqSQmWBJkqSEYngUYbJMsCRJkkJmB0uSJClkdrAkSVJCRfFoyi6lYeTIkYwaNWr99by8PLp3705WVhbnnXceubm5AOTn59O7d2+ysrLo0KED8+bNAyAejzNkyBBatmxJq1at+Oijj/5wm3awJElShfT777/Tt29fHn300U2WjxgxgqZNmzJt2jQ6duzIoEGDAHjiiSeoWrUq06ZNo2/fvmRnZwMwffp05s2bx9SpUxkzZgzZ2dkUFhYm3LYdLEmSlFAR0ZRd8vLyWLBgwWaXvLy8pNs9Y8YM9t57by666KJNls+cOZPTTz8dgDZt2vDmm29SUFDAzJkzadu2LQBHHnkkv/76KwsXLmTWrFm0atWKaDTKPvvsQ4MGDfjkk08SbtujCCVJ0nZj3LhxjB49erPlPXr0oGfPnkn9rfbt2wNsMjwIkJOTQ506dQBIT0+nevXqLFu2bJPlAHXq1GHx4sXk5OSQmZm52fJE7GBJkqSEYin8secuXbrQoUOHzZbXrFlzq/eZNm0ad9111ybLGjVqxNixY4u93Wh0y4N60WiUeDxe7PXXsYMlSZK2GzVr1kzYmdqSrKwssrKyir1+ZmYmS5cupV69ehQWFrJixQpq1apFZmYmubm57LXXXgDk5uaSmZlJ3bp110+E33h5Is7BkiRJCaVyDlYqNGvWjEmTJgEwdepUmjZtSkZGBs2aNWPy5MkAzJkzh8qVK9OgQQNOOOEEpkyZQlFRET/88APz58+nSZMmCbdhgiVJknYovXr1Ijs7m9atW1OjRg2GDRsGwAUXXED//v1p3bo1lSpVYujQoQC0bNmSuXPnrp8AP2jQIKpUqZJwG5H4lgYWy5HCxfuWdRMklYGieKysm1Cm0iIOQOzI0uv9J6XbG/7vU1O2rWsPfCVl2ypNvkIlSZJC5hChJElKqMgfe06aCZYkSVLITLAkSVJCsVL6jcCKzEdMkiQpZCZYkiQpIedgJc8ES5IkKWQmWJIkKSHnYCXPR0ySJClkdrAkSZJC5hChJElKqMghwqT5iEmSJIXMBEuSJCUU8zQNSTPBkiRJCpkJliRJSsg5WMnzEZMkSQqZCZYkSUooFncOVrJMsCRJkkJmgiVJkhIqMo9Jmo+YJElSyEywttGMt+CmO6N8MC0GwJ0jI8yZu2GMOicX6tSGiY/FyqqJpeKpFyI8NzlCJAJ7NIDbeseovQv8vgL6DY3y/Y8Qi0G7lnEu6Rwv6+aGamu1b6zXLVHq7BbnlmsqVu0b+9/nPkDHS6OsWQsZGcH1NifH6XZuxXkMXpwe4Ynxaeuv/74yeI2/OqGQ2rsGyxbnwPlXpjPh4UJ2qVU27UyF/93/+QXB+99H/33/O/7oONdfHictLdFfKZ925Nqdg5U8O1jb4IcFcPf9UWIbfX707RUHggU/L4ILe0a5s2/F6lx98TWMfS7CC4/EqFEd7r4vwqhHItx6Q5xRj0SoWyfOiNvjrFoN7bpGaXpInMMOLutWhyNR7es88nSEj+ZCyxZl2NBStqXn/qrV8NNCeGtyjIwK+o7S9rQ4bU8rBKCgEC66Oo2LO8fXd65enB7hvkfTyFlasT+EtrT/n34hwq/LI0weGyMWC977Xn4DWp9ccTrYsGPXrm3jEGGSVq+BPgOj3HjV1jtPA+6OcuHZcQ7cL4UNS4GDGsPUp4IOxtq1sCQXatUMbrvp6ji9rwjeVHJ/gfx8qF69DBsbskS1A7z/Mbz9QYSz21XcN9atPfc/+zdUqwpX9onSvmuUwaMjrFlbRo1MgceejrLrLtCxbfA45CyFN96OMmZIYRm3rHRtbf93PSfOsFtjRKOwPC9Is3euWbFeBzty7evEiKbsUlFUnEpS5LZhEc4+PU7jRlu+/a33YHEunH9mxXyRZaQHMXmLjlE+mhuhQ6ugzkgE0tOhz8AI7S+KcuRhcfbZo4wbG7Kt1Z6zFAaPijK0X4y0CvyK2tpzf+VqOOqwOMNvj/HcAzEWLYkw/MGKmeT8uhzGjY9yY4+i9csyd4PhdxTxp73LrFkpkei9LyMd7nkgQsvOUWrvCkcckvr2laYduXZtu5R8HCxcuDDhpbx4ZmKEtDQ4o/XWO0+PT4hyceeKOQa/zknHwzsvxriya5zuN0SJbfSlbsgtcd6eHOO33yPcP67ifcj+b+35+XDDbVH69IxRp3ZZt670JHrutzgOBt8Sp/pOULkydD8/xoy3Kt6+B/jnS1GaHxenYf2ybklqFee977rL4sx+KUaDenFuv6fi7P8dufaNFcUjKbtUFCmZMXHZZZcxf/58MjMzicc3fZJGIhFmzJiRimZsk1GPRHjj3WCHZ6TDmrVwxsURCgqCoaIzLo7yjyExMneDZcth7r9h5MCKk15tXP+f9orTqX18/Te0M1oFbyZ5v8PnX8P+jYJv8ztVg1YnxXl1VoR189LKo+LU/vnXwZy7oWOC7ypLl0GsKBgivf3G8ls7FP+5/8XXUKM6ND00uF88HqSZ5d2YR6PMfCfYryceF+OqbjGmvxElu2fRH9yzYiju/l+wCHatBXvvEazXvmWcO0dGqSiv/R2tdoUnEv/fHk8pWLFiBZ07d2bAgAEcccQRof7twsX7hvr3iuvnRdDuoihzXt4Q38x4C56eGOWReyrW5PZ1PvoUet8R5fmHY+xSCya/HGHs+AgTH43Rb0jwLW/A9XEKCuDqW6Ic0zROl7MrxhtNoto3NuaxCL/+RoU+ivB/n/vPToowcVqEcfcGk9yzB0VoWB96XVq6j0FRPLWvs7zf4eSO6bzzUuFWJ/MfcmIGsyYVpOQowrRI2YxH/+/+v39chLlfRhg1KJiLdOuwCFWrBPMyK5rtqfb0ev8p9W1s7Nr/65SybQ0/7NmUbas0peR7ZvXq1Rk4cCATJkwIvYO1PflhQYTd61W8N5V1jjgUup8fp+s1UdLSILM2jBoUvNH0vjJIdNpfFCUCtDg+zgVnVZzHIlHtO7qz28ZZsAg6XhKlsAiOOjzOFV0qzr5f58efI9TZlQp7pOS2urhznMGjglQnGoG/NIlzTfeKt/+3ZEeq3dM0JC8lCVZpKqsES1LZSnWCtb0pqwRL24dUJ1i9Pjk3ZdsaefgzKdtWafK7mCRJSigWt0OfLB8xSZKkkJlgSZKkhIpwDlayTLAkSZJCZoIlSZIS8ijC5JlgSZIkhcwES5IkJeRRhMnzEZMkSQqZCZYkSUoo5lGESTPBkiRJCpkJliRJSqjIowiTZoIlSZIUMhMsSZKUkEcRJs9HTJIkKWR2sCRJkkLmEKEkSUrIn8pJngmWJElSyEywJElSQp5oNHkmWJIkSSEzwZIkSQk5Byt5JliSJEkhM8GSJEkJeaLR5PmISZIkhcwES5IkJeQcrOSZYEmSJIXMBEuSJCXkebCSZ4IlSZIUMhMsSZKUkHOwkmeCJUmSFDITLEmSlJAJVvJMsCRJkkJmB0uSJClkDhFKkqSEHCJMXrnvYEUN4aQdUjTia1/S9qvcd7AkSVLpMsFKnl8BJUmSQmaCJUmSEvKncpJngiVJkhQyEyxJkpSQc7CSZ4IlSZIUMhMsSZKUkAlW8kywJEmSQmaCJUmSEjLBSp4JliRJUshMsCRJUkImWMkzwZIkSQqZCZYkSUooboKVNBMsSZKkkNnBkiRJFdJHH33EmWeeSbt27ejSpQs///wzAHl5eXTv3p2srCzOO+88cnNzAcjPz6d3795kZWXRoUMH5s2bB0A8HmfIkCG0bNmSVq1a8dFHH/3htu1gSZKkhGJEUnYJU+/evRk0aBCTJ0/m9NNPZ+DAgQCMGDGCpk2bMm3aNDp27MigQYMAeOKJJ6hatSrTpk2jb9++ZGdnAzB9+nTmzZvH1KlTGTNmDNnZ2RQWFibcth0sSZK03cjLy2PBggWbXfLy8pL6O/n5+fTq1YsDDjgAgMaNG7No0SIAZs6cyemnnw5AmzZtePPNNykoKGDmzJm0bdsWgCOPPJJff/2VhQsXMmvWLFq1akU0GmWfffahQYMGfPLJJwm37yR3SZKUUCpP0zBu3DhGjx692fIePXrQs2fPYv+dSpUq0a5dOwBisRijR4/m5JNPBiAnJ4c6deoAkJ6eTvXq1Vm2bNkmywHq1KnD4sWLycnJITMzc7PlidjBkiRJ240uXbrQoUOHzZbXrFlzq/eZNm0ad9111ybLGjVqxNixY8nPz18/pHfZZZdt9W9Eo1se1ItGo8Tj8WKvv44dLEmSlFAqT9NQs2bNhJ2pLcnKyiIrK2uz5StXruSKK66gVq1a3H///WRkZACQmZnJ0qVLqVevHoWFhaxYsYJatWqRmZlJbm4ue+21FwC5ublkZmZSt27d9RPhN16eiHOwJElShdS7d2/22msvRo4cSaVKldYvb9asGZMmTQJg6tSpNG3alIyMDJo1a8bkyZMBmDNnDpUrV6ZBgwaccMIJTJkyhaKiIn744Qfmz59PkyZNEm47Et9S7lWOxBbvX9ZNkCQppaL1vknp9o55JTtl25p96uBQ/s6XX35Jhw4d2HfffUlPDwbsMjMzeeihh1i+fDnZ2dn89NNP1KhRg2HDhtGwYUPWrl1L//79+fzzz6lUqRIDBw7koIMOIh6PM3ToUN58800AbrrpJv72t78l3L4dLEmSyhk7WNs/52BJkqSE/Kmc5DkHS5IkKWQmWJIkKaFUngerojDBkiRJCpkJliRJSqh8Hw5XNkywJEmSQmaCJUmSEorhHKxkmWBJkiSFzA6WJElSyBwilCRJCXmi0eSZYEmSJIXMBGsrZs6G4Q9CfgE0bgQD+0D1nba87mtvQfadMGdacH3QSJgzd8PtS3KhTm2Y/Fjptzssxan/m3kw8F5YsQKiaXDb9XBQ4+C2V2bBA08G929QFwb3hV12Tn0d26I4tQ8ZA9Nnws41g+t77wHDb4X8fBh0L7z3MexUFU48FnpcBNFy9FWmJPt+R3nub3X/FwSPwUf/fQyOPxpuuBzS0lJZQcmUpP5YDP7+AMx6D6IR2Ksh3HYD7ForxUWEpCSPRUXjiUaT5489b8Gy5XB6F3hqDOzdEIb9A1auggHXbb7u/AVw2Y2wdBl89PLmt/+8CM7vCffdBQfuF3pTS0Vx6l+9Bk49N3jDafZXmPF28MY69Qn4/Cu44iZ49j7YvT7cNRrWroVbry+zkoqtuPu+0xXQ5yo4/OBNl496FOb+G0YPhIwMGDAMDtgXzjsjZSWUSEn3/cYq6nMftr7/xz4HH30GI28POhvn94QLzoTWJ6eshBIpaf0TXoKpM+CBIVCpEtx9f/DeOOTmlJUQmpI+FqUt1T/2fNi/+qVsW//X+o6Ubas0pex79WuvvcYTTzzBjz/+uMny5557LlVNKLZ3PoSDDwheVADntoOXXtv8RGur10CfgcGLa2v63Q1dzi4/HzBQvPrf+RD23D34gAVocdyGb20vvgpntg46VwA9usLF56aq9SVTnNrz8+Hf/4FHn4X23eDqfrBwSXDbF99AqxZQuXKQWp10PEyflfo6tlVJ9/3GKupzP9H+73oO3HNrsO+X58HvKzYkG+VBSevfd2+44YqgcwVwcOMNt5U3JX0sKpp4PHWXiiIlHaxhw4bx5JNPMn/+fDp16sTkyZPX3/bss8+moglJWZwD9TM3XK9bB1asjLBy1abrDRgGZ58eRMdb8uZ7sDg3+AZbnhSn/vk/wW67ws1D4Kzu0O16KCzacFtREVzVN3jTuWME7FQtpSVss+LUnvMLHH04XNcdJj4Ch/4ZevQN3hgOORCmvR58080vgH+9Brm/pL6ObVXSfb9ORX7uJ9r/ABnpQaJ3WmeovSsccUhqayiJktZ/+MFw0H8HFX77He57HFqemNISQhPGc0E7tpR0sGbNmsXDDz9Mv379ePrppxk5ciTTpgUTlrbHEcpYbMvLN55H8/RESE8LkpqtGTcBLu1cvuZfQPHqLywKPkTPPh3++SCcfwZc3if4RldYCG+8GwwJvvBw8GHc/+7UtL2kilN7w/rw4FDYZ0+IRKBbJ/hxIfy8GC7pDPvuA+deCd2ug8MODj5wy4uS7vt1KvJzP9H+X+f6y+C9l2D3enDbPaXb5jCFVf+PP8MFPeGIJtC5Q+m2ubSE9VhUFPF4JGWXiiIlHax4PE4kEjxoe++9Nw888ACDBg3i/fffX798e1K/7qapw5KlsHONONWqblg26WX47GvocDFc1gfWrA3+n7M0uH3Z8mAuzmknprLl4ShO/Zm1gzeVQ/8cXD/pb0Fq9dNCyNwN/nZkMLk5GoUOWfB/X6S2hm1VnNq/ngeTp296v3g86HD/lhcME704Fp4cFUzs33P3lDQ9FCXd91Dxn/uJ9v/Hn8H3PwXLMtKhQ0v4MrVTZUqkpPUDvP9x8AWjfcvgS9Z2+BZfLGE8FtqxpaSD1bJlSy644ALmzg0Ordlvv/0YOXIk11xzzWZzsrYHxx0Jn34ZTGAHeO7FYJ7JxsY/AFPGBrHwA0OgSuXg/5m7Bbd//Fkwfr/xi7G8KE79xx8NCxfDF18H1z/8NHgjbVgfTmsWHEX062/Bba++GTwW5UFxao9E4M57YcGi4Pozk6Dxn6BeJrzxDtw6LHiTXbkKxo2H009JaQklUtJ9DxX/uZ9o/7/3MQweHaS4sRhMeRX++peUllAiJa3/k8+hZ7/gqOFunVLa9NCV9LGoaEywkpeyowhnz55NZmYmf/rTn9YvW7RoEY8++ig337zth5iUxlGEEHQQhj8IBQWwx+7BG8aChcHE3YmPbLruz4ug7UWbHkX4yDPBN9mBN5ZK80pdcer/8FMYdj+sWgOVMqBvzw3zTZ6ZBE9PgngsOE3DwD4bOp/bu+LU/uIr8NDTECsK5mYM7BPUWVgIA/4evDHHYtCxDVx0TtnWk6yS7vsd4bm/tf2fXwB3jQoen2gE/tIkOAimapWyrSkZJam/23Xw2VcbOtsQDJOOHlQ2tZRUSR6L0pbqowibvDggZdv6rO1tKdtWafI0DZIklTOp7mAdNPnWlG3ri3ap21ZpKkenP5QkSSofytHxTZIkqSyU77GusmGCJUmSFDITLEmSlFBFOrovVUywJEmSQmYHS5IkKWQOEUqSpIQcIkyeCZYkSVLITLAkSVJCnqUheSZYkiRJITPBkiRJCTkHK3kmWJIkSSEzwZIkSYk5CStpJliSJEkhM8GSJEkJOQcreSZYkiRJITPBkiRJCcWdg5U0EyxJkqSQmWBJkqSEnIOVPBMsSZKkkJlgSZKkxEywkmaCJUmSFDI7WJIkSSFziFCSJCXkaRqSZ4IlSZIUMhMsSZKUmAlW0kywJEmSQmaCJUmSEvJEo8kzwZIkSQqZCZYkSUrMOVhJM8GSJEkKmQmWJElKyDlYyTPBkiRJCpkJliRJSsw5WEkzwZIkSQqZCZYkSfoDzsFKlgmWJElSyEywJElSYs7BSpoJliRJUsjsYEmSJIXMIUJJkpSYQ4RJM8GSJEkKmQmWJElKzJ/KSZoJliRJUshMsCRJUkJx52AlzQRLkiQpZCZYkiQpMROspJlgSZIkhcwES5IkJeZRhEkzwZIkSQqZCZYkSUoo4hyspJlgSZIkhcwES5IkJWaClTQTLEmSpJCZYEmSpMQ8ijBpJliSJEkhs4MlSZIUMjtYkiQpsXgKLyGaM2cOZ5xxBqeffjqXX345v/32GwB5eXl0796drKwszjvvPHJzcwHIz8+nd+/eZGVl0aFDB+bNmxeUH48zZMgQWrZsSatWrfjoo4/+cNt2sCRJUoV00003MXToUKZMmcK+++7LI488AsCIESNo2rQp06ZNo2PHjgwaNAiAJ554gqpVqzJt2jT69u1LdnY2ANOnT2fevHlMnTqVMWPGkJ2dTWFhYcJt28GSJEmJpTDBysvLY8GCBZtd8vLykm721KlT2XfffSkoKGDJkiXUrFkTgJkzZ3L66acD0KZNG958800KCgqYOXMmbdu2BeDII4/k119/ZeHChcyaNYtWrVoRjUbZZ599aNCgAZ988knCbXsUoSRJ2m6MGzeO0aNHb7a8R48e9OzZM6m/lZGRwddff81FF11Eeno61113HQA5OTnUqVMHgPT0dKpXr86yZcs2WQ5Qp04dFi9eTE5ODpmZmZstT8QOliRJSiyFJxrt0qULHTp02Gz5uvRpS6ZNm8Zdd921ybJGjRoxduxYGjduzLvvvsuzzz7Ltddey7PPPrvFvxGNbnlQLxqNEo9v/gBsbf117GBJkqTtRs2aNRN2prYkKyuLrKysTZatXbuW1157jZNPPhmAtm3bMmTIEAAyMzNZunQp9erVo7CwkBUrVlCrVi0yMzPJzc1lr732AiA3N5fMzEzq1q27fiL8xssTcQ6WJElKLB5J3SUk6enp3HbbbXz++edAkHL95S9/AaBZs2ZMmjQJCOZpNW3alIyMDJo1a8bkyZOB4AjEypUr06BBA0444QSmTJlCUVERP/zwA/Pnz6dJkyaJtx9aJZIkSduJtLQ0hg8fTv/+/SkqKqJu3brrjxbs1asX2dnZtG7dmho1ajBs2DAALrjgAvr370/r1q2pVKkSQ4cOBaBly5bMnTt3/QT4QYMGUaVKlYTbj8S3NLBYjsQW71/WTZAkKaWi9b5J6fb2Gf33lG3r+x7Xp2xbpckhQkmSpJA5RChJkhIr12NdZcMES5IkKWQmWFsxczYMfxDyC6BxIxjYB6rvtOk6r74Jox+DaBRqVoc7boQ9d4c1a+GO4fDZVxCLw6EHQr9roUrlsqllW5SkfoBj20LdDedqo1snOP2U1LW/JP6o9kkvw7gJG67/vgKW5MIb/4Tddi3ftUPJ6r9jBPz484bbFiyCIw+F+zY9Pc12zf2/7fVXrgS3DIXvfoR4DNq1hEs7p76GkihJ/enpcNs98NV/oGoVOCMLzj8z9TVo++Ak9y1YthxO7wJPjYG9G8Kwf8DKVTDgug3rrFkbvJFOfAT2aghjx8Psj+CBITDiYVi0BO66CeJxuHFgsM7VF4fe1FJR0vq//xGuuAlefqrMSthmxal9YwWFcEFP6JAF57Qt37VDyevf2Gf/hl4D4KnRUD/x6WK2G+7/ktU/aCREotC3J6xaDad3hWH94PCDU1nFtitp/dl3Qloa3H4DFMWgx81wbntofmz4bU35JPdRKZzk3tNJ7kmZP38+S5YsAWDChAkMHDiQqVOnpmrzSXnnQzj4gOAFBnBuO3jptaCztE5RUXB9xcrg+qrVwbc3CL6xX35hkOykpcGB+8HCJamtoSRKWv8nn0NaFLr0gnYXwZixwfrlQXFq39jDT0PtXTZ0Lspz7VDy+tfJL4Cb7oKbepSfzhW4/0taf9+r4cYrgv/n/gL5+VCjeum3Oywlrf+Lb6DdqcH7fqUMaHYMvDIrNW0vbZF46i4VxR8OET766KOMHj2aoqIiGjRoQOPGjddf9t9/fxo2bPiHGxk7dixPPPEEsViMv/71ryxatIhTTjmF559/nu+//56rrroqlGLCsjhn0w+FunVgxcoIK1fF10fFO1ULvtWcexXUqgmxWPBNHeC4Izfc9+fF8Pg/4bYbUtf+kipp/YVFcGxT6H1FkHRdnh1E7F06pr6WZBWn9nV+XQ5jn4PnH96wrDzXDiWvf53n/wV1doNTTijV5obO/V+y+iORYJjsxoEwfRac/DfYZ4+UND0UJa3/kANh8itweJOgc/nqrODx0I7pD3f9Aw88wNChQznkkEP46aef+Oabb/j666958803+fbbbwHYb7/9eOaZZ7b6N55//nmmTp3K0qVLadOmDe+99x6VK1emY8eOnHXWWdtdBysW2/LyjX926Jt5cP/j8NK4YN7RE/+EXv2DIbPIf09E+8XX0PMWOK9D6UTEpaWk9Z99+ob1KlWCrmfDE8+Xjw+Z4tS+zvgp0OJv0LD+hmXluXYoef3rjJsQDJOUN+7/LS9Pdv8PvSX4AtarP9w3Dnp2C7edpaWk9fe5EobeD2dcAnV2DTrbn3xeOm1NuRDPsL6j+MMOVvXq1TnxxBNJT08nMzOTI444YpPbFyxYsL6jtTWxWIxKlSqx++67061bNypX3jDbu2g7zM/r14W5/95wfclS2LlGnGpVNyx7+8NgXsG6Sd2dO8DgMbD8N9ilFvxrRjDR/ZZe0KYcTXCFktf/5vtwwL7Q+E/BbfF4+fkWV5za15n2RjAksrHJ08tv7VDy+gG+/CYYFjvysFJrZqlx/5es/rc/gP0bQeZuQcrd+qTyNURW0vpXrIIbLg9SfYCHnoY9/3iQRxXUH87B6t69OxMmTNjq7Q0bNqR58+YJ/8app57K+eefT1FRET179gTgq6++onPnzpv9OOP24Lgj4dMvYf6C4PpzL0KL4zZd58/7w4efwtJlwfUZbwffZHapBdNnwp33wsPDyl/nCkpe/7ffw6hHgw/ZNWvhqYmQlfgpst0oTu0Av/0eHC33v5N3y3PtUPL6IXheHP2XDUlueeL+L1n9094I5p3F48EQ2bQ3gudCeVHS+p+bHOx/CN4b//kStDmpdNus7dcfHkV4+OGHU1BQwAknnMDxxx/PgQceSOPGjaladQtd+gQ+/PBDjjxyw+Sk7777jp9++olmzZptW8v/q7R+KmfWe8GhugUFsMfuMLgvLFgI/e4OhsEgePN8+gXIyICda8At18B++8BpnYNDdzc+VPvwg6H/taXS1FJRkvpXr4GBI4I3qoJCaHkiXHNp+fnALU7tn/0bbrgDpj+96X3Le+1QsvoBbh8OdWrDFRemtt1hcf9ve/15v8Ot9wQdzQhw0vHQ86ItD7Ftr0pS/8pV0GcQ/PBz0Mnsfh60PbV02pnqowgbjbgnZdv67pqtHLZZzvxhB+unn37iq6++4uuvv+brr7/mq6++YuHChTRs2JDp06enqp1b5W8RSpJ2NHawtn9/ODtgjz32YI899uCUUzaMda1atYpvvkntzpUkSWWkAp0+IVW2KbitVq0ahx12WMhNkSRJqhjK0fEtkiSpLFSkE4CmSjmaeihJklQ+mGBJkqTETLCSZoIlSZIUMhMsSZKUmAlW0kywJEmSQmaCJUmSEvIowuSZYEmSJIXMBEuSJCUWL0c/qLmdMMGSJEkKmQmWJElKzDlYSTPBkiRJCpkdLEmSpJA5RChJkhLyNA3JM8GSJEkKmQmWJElKzAQraSZYkiRJITPBkiRJCTkHK3kmWJIkSSEzwZIkSYmZYCXNBEuSJClkJliSJCkxE6ykmWBJkiSFzARLkiQl5FGEyTPBkiRJCpkdLEmSpJDZwZIkSQqZc7AkSVJizsFKmgmWJElSyOxgSZIkhcwhQkmSlJCnaUieCZYkSVLITLAkSVJiJlhJM8GSJEkKmQmWJElKzAQraSZYkiRJITPBkiRJCXkUYfJMsCRJkkJmgiVJkhIzwUqaCZYkSVLITLAkSVJCzsFKngmWJElSyEywJElSYiZYSTPBkiRJCpkJliRJSswEK2kmWJIkSSGzgyVJkhQyhwglSVJCnqYheSZYkiRJITPBkiRJiZlgJc0ES5IkKWQmWJIkKTETrKSZYEmSJIXMBEuSJCXkUYTJM8GSJEkKmQmWJElKzAQraSZYkiRJITPBkiRJCTkHK3kmWJIkSSEzwZIkSYmZYCXNBEuSJClkJliSJCkxE6ykmWBJkiSFzA6WJElSyOxgSZKkhCIpvJSGL7/8koMPPnj99fz8fHr37k1WVhYdOnRg3rx5AMTjcYYMGULLli1p1aoVH3300fr7PProo7Rs2ZLTTjuNV1555Q+36RwsSZJUYa1evZrbb7+dgoKC9cueeOIJqlatyrRp0/jwww/Jzs5mwoQJTJ8+nXnz5jF16lR++OEHunfvzrRp0/jyyy958cUXmTx5MitWrOCcc87hqKOOolatWlvdrgmWJElKLJ66S15eHgsWLNjskpeXt01NHzx4MF27dt1k2cyZM2nbti0ARx55JL/++isLFy5k1qxZtGrVimg0yj777EODBg345JNPePPNNznllFOoXLkytWvX5qijjmLmzJkJt2uCtRUzZ8PwByG/ABo3goF9oPpOG26f9DKMm7Dh+u8rYEkuvPFP2G3XDct73gKZu0G/a1LW9FDsyPXvyLVDyes/ti3UrbPh9m6d4PRTUtf+knL/u/8T1b+x196C7DthzrTgen4+DLoX3vsYdqoKJx4LPS6CqFFGUsaNG8fo0aM3W96jRw969uyZ1N+aMWMGa9asoWXLlpssz8nJoU6dDU/UOnXqsHjxYnJycsjMzNzi8iZNmmy2PBE7WFuwbDncPBieGgN7N4Rh/4C/PwADrtuwTvuWwQWgoBAu6AmXnrfpG+zDT8NHcyGrRUqbX2I7cv07cu1Q8vq//xFq1oCJj5RJ80vM/e/+/6P615m/AO6+H+Ibnb7ggSdh4RJ48THIyIABw+CZSXDeGamqoPSk8qdyunTpQocOHTZbXrNmza3eZ9q0adx1112bLGvUqBErVqxg7NixxdpuNBolHt+80ETLE/69Ym01ZIMHDy6LzRbbOx/CwQcELzCAc9vBS69t+kLa2MNPQ+1d4Jy2G5a9/zG8/QGc06702xu2Hbn+Hbl2KHn9n3wOaVHo0gvaXQRjxkJRUUqaHgr3v/u/OPWvXgN9BkKfqzZd/sU30KoFVK4cpFYnHQ/TZ6Wm7RVJzZo1adiw4WaXRB2srKws3nzzzU0urVu3Zvny5Zx33nm0axe8INu1a8eKFSvIzMwkNzd3/f1zc3PJzMykbt26SS1PpNQTrJtuummzZa+//jq//fYbwGY9zu3B4hyov9HjVrcOrFgZYeWq+GZR8a/LYexz8PzDG5blLIU7R8FDw2D8iylpcqh25Pp35Nqh5PUXFsGxTaH3FbBmLVyeHQyvdOmYkuaXmPvf/V+c+gcMg7NPD4YQN3bIgTDtdTi1WZBg/es1yP0lNW0vdeXwRKMdO3akY8cNT77GjRszefJkAJo1a8bkyZNp2rQpc+bMoXLlyjRo0IATTjiB559/njZt2rBgwQLmz59PkyZNqFy5Mv379+eiiy5i9erVvPfee/Tq1Svh9ku9g1WrVi0mTZrE5Zdfvr73+d5773HUUUeV9qa3WSy25eVbSgPHT4EWf4OG9YPrBYVw3W1wU0/IrF16bSxNO3L9O3LtULL6IfjQWadSJeh6NjzxfPn5gHX/b3m5+3/D/5+eCOlpcGZr+HnRputd0hlGPATnXhkMlWa1gK/nlV57te0uuOAC+vfvT+vWralUqRJDhw4FoGXLlsydO3f9BPhBgwZRpUoVDjnkENq2bctZZ51FYWEhV199NXXr1k24jVLvYPXp04cTTjiBESNGcN1113H00Uczbty4LY6vbi/q14W5/95wfclS2LlGnGpVN1932hvQ9+oN1z//KnjRDRkTXF+6LIjI1+bDwBtLt91h2ZHr35Frh5LVDzB5OhywLzT+U3A9Hof0cjTT0/3v/v+j+ie9DKvXQoeLoaAgSOo6XAwPDIFIBLqeAzdeGaw79XXYc/fU1lBqymGC9b++/vrr9f+vXLkyQ4YM2WydSCRCnz596NOnz2a3devWjW7duhV7eyl56h9zzDEceOCBDBgwgJkzZ1K0nQ/KH3ckDL0vmMS4d0N47kVocdzm6/32O/z4Mxy+4dxlHH5wcDTNOqMfg19/K19HEu3I9e/ItUPJ6gf49nt49U0YeXuQ6Dw1EdqcnJq2h8H97/7/o/rHP7Dh/z8vgrYXbZjUP/7F4CjEMXfCqtUwbnxwFKV2TCmb5F6rVi1GjhxJo0aNNjk0cntUexcYlA3X9IfWF8A338GNVwXfUDtcvGG9HxdAndqQUY6+oRXHjlz/jlw7lLz+q7rCzjWCCc7tLoLDD4KObVJaQom4/93/xal/a85oBbvUgtO7QsfLoGVzOO3EUm50ikTiqbtUFJH4lo49LEdii/cv6yZIkpRS0XrfpHR7h/UcnrJt/d+oa1O2rdJUwb5/SZKk0JXrKKZseH5ZSZKkkJlgSZKkhCrS3KhUMcGSJEkKmR0sSZKkkDlEKEmSEnOIMGkmWJIkSSEzwZIkSQk5yT15JliSJEkhM8GSJEmJmWAlzQRLkiQpZCZYkiQpMROspJlgSZIkhcwES5IkJeRRhMkzwZIkSQqZCZYkSUrMBCtpJliSJEkhM8GSJEkJReJGWMkywZIkSQqZCZYkSUrMACtpJliSJEkhs4MlSZIUMocIJUlSQp5oNHkmWJIkSSEzwZIkSYmZYCXNBEuSJClkJliSJCkh52AlzwRLkiQpZCZYkiQpMROspJlgSZIkhcwES5IkJeQcrOSZYEmSJIXMBEuSJCVmgpU0EyxJkqSQmWBJkqSEnIOVPBMsSZKkkJlgSZKkxOJGWMkywZIkSQqZHSxJkqSQOUQoSZIScpJ78kywJEmSQmaCJUmSEjPBSpoJliRJUshMsCRJUkKRWFm3oPwxwZIkSQqZCZYkSUrMOVhJM8GSJEkKmQmWJElKyPNgJc8ES5IkKWQmWJIkKTF/7DlpJliSJEkhM8GSJEkJOQcreSZYkiRJITPBkiRJiZlgJc0ES5IkKWR2sCRJkkLmEKEkSUrISe7JM8GSJEkKmQmWJElKzBONJs0ES5IkKWQmWJIkKSHnYCXPBEuSJClkJliSJCkxE6ykmWBJkiSFzARLkiQl5Bys5JlgSZIkhcwES5IkJRYzwkqWCZYkSVLIdvgEa+ZsGP4g5BdA40YwsA9U36n46xUVwZAx8PaHwf8vOgc6tQvu8/7HMOS+YHmtmnBTTzhg3+C2x56D56dCehrsUgtuux723D1lZf+h4jwu38yDgffCihUQTQtqOKjxhtvzfocLroZBfeDgA1Lb/m1V0ufDOotyoNMVMOmRYP+u886HMOwfMPGRUi9lm+2o+36d4tQ/ZAxMnwk71wyu770HDL81+P8rs+CBJ4P7N6gLg/vCLjunsoKSKe5rAOC1tyD7TpgzLbgei8HfH4BZ70E0Ans1hNtugF1rpaz5oXv1TRj9GESjULM63HHj5u/Vk16GcRM2XP99BSzJhTf+Cbvtmtr2lhoDrKTt0AnWsuVw82AYeQdMexIaNgjeHJJZ77kXYf4CePExGP8APP5PmPvv4AV2dT/ofQVMfgwGXAfX3gr5+fDuHHj+X/DsfTDpUTjleOg7OIWF/4HiPC6r18DFN8DF58ILj8AVF0LvgRtun/UenH05fP9jSpteImE8HyB4sz2/J+QsjaxftmYtjHg4eA4UFpV2JdtuR9336xT3OfDJ5/D3AUFHeeIjGzpXn38Fd4yAe2+HKWODjtfIh1PX/pIqbv0QvO/dff+mv6Dy/FT48ht44SF4cWzQERkyJhUtLx1r1kKfQXDvHcF+bn4cDLp38/Xat9zwXBj/QNCpuuWaCtS50jZJSQdr7ty56/8/e/ZsBg8ezLBhw/j0009TsfmteufD4Nv13g2D6+e2g5de2/wnlxKt99pbcEYWpKfDzjWgVQuY8gr8sACqV4djjgju02gvqF4N/u+L4EXX/7oN3woPPgAWLklNzcVRnMflnQ+DN89mfw2utzhuw4cMwJPPw103QZ3dUtbsEgvj+ZCzFGa8DQ8M2fQ+b38QdEwG9in9OkpiR9336xSn/vx8+Pd/4NFnoX234IvUutfvi6/Cma1h9/rB9R5dg45oeVHc18DqNdBnIPS5atPl++4NN1wBlSoF1w9uvH29tyWrqCiofcXK4Pqq1VC5UuL7PPw01N4Fzmlb+u1LpUg8dZeKIiUdrAEDBgDw1FNPceedd1KvXj122203+vfvz5NPPpmKJmzR4hyon7nhet06sGJlhJWrir/e4lyo9z+3Lc4NvrmuWh28YQF89m/4z3zI/QX2bwRHHRYsz8+Hex6AlieWQoHbqDiPy/yfgo7izUPgrO7Q7fpNk5mH7obDD05dm8MQxvMhczcYNTD4oNnYycfDTT2gVo1Sa34odtR9v05x6s/5BY4+HK7rHiQWh/4ZevQNPojn/xR8KF/VN+h83TECdqqW8jK2WXFfAwOGwdmnB0OIGzv8YDho/+D/v/0O9z2+fb23JWunasHow7lXwQlnwNMT4frLtr7+r8th7HPBdBAppUOE48eP5/HHH6dr16507dqVp556qkw7WLHYlpdHo8Vfb0u3pUWDdGrMoGAuRvtuMHk6HP0XyMjYsN6y5cFQS7WqcM2l21RCqSjO41JYBG++F7zJ/vNBOP8MuLxP0GEsr8J4PpR3O+q+X6c49TesDw8OhX32hEgEunWCHxfCz4uhsBDeeBduvR5eePi/afXdqWl7GIpT/9MTg7mjZ7be+t/58We4oCcc0QQ6dwi3jan0zTy4/3F4aRy8+QJcdj706r95orfO+CnQ4m/Bc0RKyST3wsJCYrEYtWvXplq1DV/nKlWqRDTFn0r3PhK8AUIQ++6/0TewJUth5xpxqlXd9D716wbzqra0Xv26QSq1Ts5SqJsZvFFVqwqPj9xwW+sLNkyO/HoeXNk3SDZuvALS0sKtsyQS1btOZu3gA+bQPwfXT/ob9BsKPy2EP+2d0uaWSNjPh/JuR9r3W1Kc+r+eB1/9B9qdtmFZPB50OjJ3C55DdWoHyztkwUXXpqbtYShO/ZNehtVrocPFUFAQzFPqcHEwLJ65W3Bwz3W3BUOj3TqlvoaS2vg9Yf5PcNqJG963O3eAwWNg+W+bHryyzrQ3oO/VqWppim2tV6mtSknvZpdddqFZs2b85z//WT9cOHv2bDp16kTLli1T0YT1rr54w2TEZ++HT78MJmtCMGG9xXGb3+e4I7e+3knHwQtTg2+ueb/D1BnBB04kApf1CSa9Arz8RjBPq/GfgvlZXa6BKy8Mho22p84VJK53neOPhoWL4Yuvg+sffhrUXN6+uYX9fCjvdqR9vyXFqT8SgTvvhQWLguvPTApe1/Uy4bRmwST/X38Lbnv1zfJ1FGVx6h//QDCBf+IjQaeqSuXg/5m7BZP/e/YLjpwsj50r2PQ94f7BwfN76bLgthlvB8/zLXWufvs9SO7K6/B4RTVp0iT+9re/0a5dO9q1a8fw4cMBWLhwIeeddx4tW7bkiiuuYOXKYKJdXl4e3bt3Jysri/POO4/c3FwA8vPz6d27N1lZWXTo0IF58+b94bZTkmA9/vjjAHz33Xfk5eUBQXp19dVXc+KJJ6aiCVtUexcYlA3X9A++ie2xe/DGAEHHqN/dwYss0Xqd2gXDA+3/+23unLYb5lfd3S8YHsgvDL7Rjh4UvDk//DSsWQNPvhBcACplwHP/SPlDsEVbq3fjx6RObRg1CG4fDqvWBO2/9w6oXLmsW7/twng+lHc76r5fpzj1798Ibu4FV9wEsaJgntKw/sH9mx8XzMG8sBfEY8FpGrb3Axs2Vpz6Exn1aBB03PNgcAHYvV7w3lce/fUvQUexS69gesfONTbU8r+PyY8LgtdGRgU9+VF5nXz+2WefkZ2dTZs2bTZZftttt9G5c2dat27NmDFjuO++++jduzcjRoygadOmPPjgg0yaNIlBgwYxYsQInnjiCapWrcq0adP48MMPyc7OZsKECVvZaiASj5fv3C+2eP+yboIkSSkVrfdNSrfX/LQhf7xSSCZPuGJ9GLOxmjVrUrNmzaT+1tlnn03NmjXJyclh//33p1+/flSrVo2jjz6aDz74gPT0dBYtWsT555/PjBkzaNGiBU899RT169ensLCQo446ivfff59u3brRq1cvmjZtCsDJJ5/M448/ToMGDba67QowLVeSJJWqeOou48aN46STTtrsMm7cuKSbXadOHXr27MnkyZOpX78+t99+O7/++ivVq1cnPT19/TpLlgTnE8nJyaFOnToApKenU716dZYtW7bJ8nX3Wbx4ccJtV9AwU5IklUddunShQ4fNDz9NlF5NmzaNu+66a5NljRo1YuzYseuvX3LJJZx88snceOONm90/EolstmydrR2M90cH6dnBkiRJCUVSOJtoW4YCs7KyyMrK2mTZ77//ztixY+natSsA8Xic9PR0dt11V1asWEFRURFpaWnk5uaSmRmcAC4zM5OlS5dSr149CgsLWbFiBbVq1SIzM5Pc3Fz22msvgE3uszUOEUqSpAqnWrVqPPzww+t/NebJJ5/klFNOISMjg6ZNmzJ16lQgONLwhBNOAKBZs2ZMmjQJgKlTp9K0aVMyMjJo1qwZkydPBmDOnDlUrlw54fwrcJK7JEnlTqonubc4KXU/mPv6jOzQ/tacOXMYNGgQa9asYe+992bo0KHUqFGDn3/+mezsbH755Rfq16/PPffcw84778zy5cvJzs7mp59+okaNGgwbNoyGDRuydu1a+vfvz+eff06lSpUYOHAgBx10UMJt28GSJKmcsYO1/XMOliRJSiiVc7AqCudgSZIkhcwES5IkJWaAlTQTLEmSpJCZYEmSpMScg5U0EyxJkqSQmWBJkqSEIgZYSTPBkiRJCpkdLEmSpJA5RChJkhJzknvSTLAkSZJCZoIlSZISisTKugXljwmWJElSyEywJElSYs7BSpoJliRJUshMsCRJUmIGWEkzwZIkSQqZCZYkSUoo4hyspJlgSZIkhcwES5IkJWaClTQTLEmSpJCZYEmSpMQ8k3vSTLAkSZJCZoIlSZIS8ijC5JlgSZIkhcwOliRJUsgcIpQkSYk5RJg0EyxJkqSQmWBJ5VRsBz9uuii+Y9ffZvcjyroJZWZ5l2PKugll7sPHUrxBE6ykmWBJkiSFzARLkiQltmMHxtvEBEuSJClkJliSJCkhTzSaPBMsSZKkkJlgSZKkxEywkmaCJUmSFDITLEmSlJgJVtJMsCRJkkJmgiVJkhIzwUqaCZYkSVLITLAkSVJinsk9aSZYkiRJIbODJUmSFDKHCCVJUkL+VE7yTLAkSZJCZoIlSZISM8FKmgmWJElSyEywJElSYjETrGSZYEmSJIXMBEuSJCXmHKykmWBJkiSFzARLkiQlZoKVNBMsSZKkkJlgSZKkxEywkmaCJUmSFDITLEmSlJjnwUqaCZYkSVLITLAkSVJi8VhZt6DcMcGSJEkKmR0sSZKkkDlEKEmSEvM0DUkzwZIkSQqZCZYkSUrM0zQkzQ7W/5g5G4Y/CPkF0LgRDOwD1XdKfr1FOdDpCpj0COxSa8Pydz6EYf+AiY+UeilJK2nta9bCHcPhs6+C1+KhB0K/ayE/H7pcs+nf+OY76H05dD0nJaUVS0nrX54Ht90DX/0HqlaBM7Lg/DOD+3z2b7hrNKxeA0VFcElnaHtqausrqSmvRHj02QiRCFStDDddHePgA+DOkRHmzI2sXy8nF+rUhomPVayjjr75DgaPTGPFygjRaJz+1xfx58bBbed0T2ftWsj47ztqq1NiXNRp+6p/74P3pMe93dhp52rEimKMuPxBvv34u83WO6HjMZx3c/DE/W3p74y8/AF+/s/ibd5u5aqVuO6hK9j38L2JRKM8nP0k707+sFS2VRzNDv8Tt17akuZXjkm43p/3qcdDN51Nq+se5LcVa7Z5e9FIhGvPbcZfD96btGiUJ1+ewwsz5wJwxAF7cPXZx5OelsbagkKGPfUGX35fuvUrdRwi3Miy5XDzYBh5B0x7Eho2gL8/kPx6k16G83tCztINHzpr1sKIh+HaW6GwqLQrSV4Ytf/jiaC2SY/C5EeDmh98EmrWCDqU6y5ntoK/NIHzzkxlhYmFUf/g0VCtKrw0Dp69H958H954N5i60Ks/9LgoqP/BoTBkDMxfkMoKS+b7H2HY/REevDvGC4/EuOzCGL36BW8ffXvFeeGRYPmogTEqV4I7+25fnYuSWr0GLr8hnYvOjTH+4UIuuzBG9sCgN7VqNSz4GSY8Urj+sr11ripXrcTg6bcw/u7JXHHEjTw58J9kP3n1ZuvVytyZXvd355Y2d3HZYTfwzsT36THq4hJt+4Jbz2b1yjVcfNC19Dn1Dq4ecwm77b5rqWzrj+xRtxa9zmlGNBJJuN7O1auQfeFJVMooeQZxxomHsEfdXeh0yzi63P4U5576F/68Tz3S06LceUVrBo19lfMGPMGjU97j9ktblnh7pSYeT92lgkhZB+utt94iLy8PgEmTJnH77bfz/PPPp2rzxfLOh3DwAbB3w+D6ue3gpdc239+J1stZCjPehgeGbHqftz8I3qQH9in9OrZFGLUfeShcfiFEo5CWBgfuBwuXbHr/HxYEHbEhN2/4tr89CKP+L76BdqcGtVfKgGbHwCuzggTvyq5wbNPgPvUyYZedYUlOysorsUoZcPuNMerUDq4f1BiWLgtSvI0NuDvKhWfHOXC/1LexNM3+MMIeDeIc/9fgCXHicXGG3VoIwOdfRahaFXpkp3HGRekMHR1lzdqybO3mjjj1UBbNW8IH0z4BYPaLcxh4zj2brbc85zfOrncJuQt+IZoWJXOvOuT9smL97Z37nsF9c4bwj4/v5tYXelO7/i6b3P+CAR25YEDHTZb9rf1RTH3oNQByf1rKR6/MpdnZx/7htsJWuVI6t1+axYhnZyZcLxKBO7q34r7n397strbHH8zjA87jyVvPZ8wNZ7JXvU3rb3Pcnxlw8WmbLDvxiH2Z8tbnFMXi/L5qLa+8/zVZxxxIYVGMVtc9yDc/5gKwe51a/LZy25MybX9S0sEaNGgQDzzwAGvXrmXEiBFMmTKFfffdl1dffZWBAwemognFsjgH6mduuF63DqxYGWHlquKvl7kbjBoI++696X1OPh5u6gG1apRa80skjNqPOxL22SNY/vNiePyfcNqJm95/xMNw3hnQoG6plLHNwqj/kANh8itQUAgrV8GrsyD3F6hcGc5qveE+418MUo9DDyrdmsK0e/2gwwhBZ3LomAjNjws6Xuu89R4szoXzz6w430DXmb8gQu1dYcDQNDp1T6P79Wnrk+iVq+Cow+P8/bYinvlHIYtyIox8cPsaHGi4f32WLV7OdQ9fwZgPBjPklX6kpadtcd2iwiL2P6IRz/z0AK0vPZlJo6cBcPIFJwTDjEffxOV/6c0H0z7huocu/8Nt19mjNrk//bL+eu7Pv7Bbw9oJt1Ua+nY5mRdmzeXbn5YmXO+yDsfyxXeLee/zHzZZ/pfGDWl93J+59K7nOP/WJ3l82hyG9mj7h9utu2sNlizb0HHM+fV36u5aHYCiohi71qzGv+7pztVnH8/jUz/chspSxAQraSnJEN555x2mTJlCWloaM2fOZPz48VSqVIlzzjmHNm3apKIJxRLbSqofjW7beuVJmLV/8TX0vAXO6wDNj92wfFFOkAANvLFkbS0NYdTf50oYej+ccQnU2TVIrD75fNP1Hnoq6Hg+dDdUqVzydqfaqtVw811RFufCA0M3fTAenxDl4s5x0rb8uV2uFRbC2+9HeHh4EYf8Oc4bb0e4qk86058rpPlxcZoft2Hc/9Lziri2fzp9em4/w4TpGekc1epwere4la8++A/HtG3KoH/15fy9r6Agv3Cz9b/56DvOaXApTU87jIEv3cSFf7qKv7Y+gsZH7ct9HwbxfDQtSuVqlQAY8Hxv6u+TyS71agFwXLujWPR9DredeTeRLbwxxopiCbe18rdVm92nJM5qfihFRTGmvPUF9WvX3Op6xx2yDwc3qk/Pv28+unLcIfuwR2YtHr250/plO1evQs2dqnB5h2M5dL8G1NypClUrV+Kp284nv6CIiwY+Q2QLw5FFG00YX5a3itbXPUjjvTK5r/dZXHTH0/y4ZHnJCtZ2ISUdrCpVqvDLL7+QmZlJ7dq1WbVqFZUqVWL16tWkp5ftONG9jwTzZABWrIT9G224bclS2LlGnGpVN71P/bow999/vN72rjRq/9eMYKL7Lb2gzSmb3nf6TDj5b7BTtdBL2SZh179wCdxwOdT67/v3Q0/Dnv8dSszPh5vugnk/wLP3BYnQ9m7UIxHeeDf4cGh+bJwz28S56qYof9orzmMj4pt0EJctDx6XkQMrzrfPMY9GmflO0DlYngd77wmH/Dmor/nf4tx6NyxYBD/+HKH6TtD00OC2eDzCVsKhlOpy2zkcc3owLl2jdnV++upnvvrgP0AwRHjdQ1dQv1Fdfvzq5/X3qV1/F/ZpsidzXvkUgDnT/49Veato8Kd6RNOiPDd0Mi/94xUAMiqlU32XIIm57cy7AdYPDz5x24T1fzPnx6XsWr8Wv/6307Bbg12Z9+n8hNva0uT7ZF3W/lhOODx4URcUxqhSKZ2nbjuf9LQ0Kv/3/72GT2Tp8pXr79P2+IPJ3KU6T956/vpl/+hzNrc/Mp20aISps//N6AlvAcFQYp1a1clbuYahT74OBEOERxywB7c9Mn39/Zf8ksdutTYcLVNnl+rkLPudnapW4sgD92Tmx8E++fqHHL79KZc/NayzfXawKlCylCop6d306NGDs846i9atW9OoUSMuuOACjjnmGN5++20uueSSVDRhq66+OLgA/PIrtLsomHy8d0N47kVocdzm9znuSBh63x+vt70Lu/bpM+HOe+HhYcE8pf8151M4tVmplZO0sOt/bjKsWAX9rgnmJ/3zJRjWL7jtmgFQFIOnx1BuOuI9L47T8+LgTXV5Hpx9aZT2WXGu7Lr5G+0nnwX7vLzUVhxXdYtxVbcgaVn6C7Tvms6XX8OfG8OcTyMQgd3rwYf/F+HBxyM8OrKIjPQgyTutRdmnV+MGPMe4Ac8BsEvdWjz8xXD2+0sjvv34O5ocfyDxeJxF3286ETCjSgY3P3stVx2ZzcJ5izn0xINIS0/jx38vYM4rn9K6+ym8/tRbrPp9NV1uP4d9D29E9ml3JGzH7Bc/pHX3U7j3yofYbfddadryMJ4a9HzCbYXhgUnv8sCkdzdbXr92TZ4deCHnDXhys9v6jJmyyfUPH7uOy4eM57cVa9i5ehVu7noKz7zyMb/8tpIzTzyUTqcczll9xyZsx6xP5tH2+IN46//mUbVyJU49qjGDH59BLBanX7dTWZa3irn/WUijBrXZu/6ufPHdohLVre1HSjpYLVq0YL/99uO1117jhx9+4LDDDmOnnXZi8ODBHHLIIaloQrHU3gUGZcM1/aGgAPbYHQb3DW77/Cvod3dwFFii9cqrMGq/58HgS06/uzf83cMPhv7XBv+fvyD4QNoehVF/9/OhzyA4vWvwOFzVFZocCB9/Bm+8G2HvPeKc12PDNq+/DP52VKor3TbPTY6wKAdeeyvCa29tGPJ49J4YtXaGHxZE2L1exf2Gu1ttGDGwiIEj0li9OkKlSjD89iIqV4aOp8dYsDDKOZemU1QERx4e5/ILy76DtbFflyzn1g5D6TnmEqrsVJmCtYXcduYwCtYWULv+Lgz6V19ubn0ni7/P4e+X3E//f14PcVixfCX92g5m7ep8pj08g91235V7Z99JPB4n58el3H3Rpqc62Di5WmfcgPH0uu9SHvrsHqJpUR668QkWfRcc/bK1baXSbrV2YuS1HTZLs/7Xe5//wONTP2TMDWcSi8dZuTqfG0dv2iF76Z0veemdLzdZ9vwbn9IwsxZP334B6elpTJw5l4+/DjqRvUe9yPWdTyQ9LUp+QRG3PDCVnF9Lb6J/iWxtfoS2KhKPl+/cL7Z4/7JuglQmYuzYb3hF8R27/ja7H1HWTSgzy7scU9ZNKHMfPnZdSreXVf+qlG1r2qLE5ygrL7ajA+UlSdJ2qXxnMWWiHB/3JkmStH0ywZIkSYmZYCXNBEuSJClkdrAkSZJCZgdLkiQlFoun7hKinJwcunfvTvv27enUqRMLFgSnyMjLy6N79+5kZWVx3nnnkZsb/CZkfn4+vXv3Jisriw4dOjBv3jwA4vE4Q4YMoWXLlrRq1YqPPvroD7dtB0uSJFVIN954I82bN2fSpEm0a9eOYcOGATBixAiaNm3KtGnT6NixI4MGDQLgiSeeoGrVqkybNo2+ffuSnZ0NwPTp05k3bx5Tp05lzJgxZGdnU1i4+c9MbcxJ7pIkKaF4Cs87l5eXR15e3mbLa9asSc2aW/8tyf+1bNkyvvrqKx577DEAzjzzTI45JjiH2syZM3nqqacAaNOmDbfffjsFBQXMnDmTXr16AXDkkUfy66+/snDhQmbNmkWrVq2IRqPss88+NGjQgE8++YQjjzxyq9u3gyVJkrYb48aNY/To0Zst79GjBz179iz23/npp59o0KABd955J++//z4NGjSgX7/g98tycnKoU6cOAOnp6VSvXp1ly5ZtshygTp06LF68mJycHDIzMzdbnogdLEmSlFjIc6MS6dKlCx06dNhseaL0atq0adx1112bLNtrr7348ssv6dmzJzfffDMTJkwgOzubJ554Yot/Ixrd8qypaDTKln70Zmvrr2MHS5IkbTeSHQoEyMrKIisra5NlP/74Ix06dKB58+ZAMBQ4cOBAADIzM1m6dCn16tWjsLCQFStWUKtWLTIzM8nNzWWvvfYCIDc3l8zMTOrWrbt+IvzGyxNxkrskSUosHk/dJSR77rkndevWZdasWQC88cYbHHTQQQA0a9aMSZMmATB16lSaNm1KRkYGzZo1Y/LkyQDMmTOHypUr06BBA0444QSmTJlCUVERP/zwA/Pnz6dJkyYJt2+CJUmSKqTRo0czYMAA7r77bqpXr87gwYMB6NWrF9nZ2bRu3ZoaNWqsP7rwggsuoH///rRu3ZpKlSoxdOhQAFq2bMncuXNp27YtAIMGDaJKlSoJtx2Jb2lgsRyJLd6/rJsglYkYqTuqZ3tUlMKjmrZHbXY/oqybUGaWdzmmrJtQ5j587LqUbq/lzt1Stq2Xf3s0ZdsqTQ4RSpIkhcwhQkmSlFj5HuwqEyZYkiRJITPBkiRJCcVjO/acx21hgiVJkhQyEyxJkpSYc7CSZoIlSZIUMjtYkiRJIXOIUJIkJZbCH3uuKEywJEmSQmaCJUmSEtvBf5pqW5hgSZIkhcwES5IkJRR3DlbSTLAkSZJCZoIlSZIScw5W0kywJEmSQmaCJUmSEnIOVvJMsCRJkkJmgiVJkhJzDlbSTLAkSZJCFonH4w6sSpIkhcgES5IkKWR2sCRJkkJmB0uSJClkdrAkSZJCZgdLkiQpZHawJEmSQmYHS5IkKWR2sCRJkkJmB0uSJClkdrAkSZJCZgdLkiQpZHawJEmSQmYHS5IkKWR2sEpgypQptGrVilNOOYWnnnqqrJuTcitWrKBNmzYsWLCgrJuScqNHj6Z169a0bt2aoUOHlnVzUm7kyJG0atWK1q1b89hjj5V1c8rEkCFDyM7OLutmpNyFF15I69atadeuHe3atePTTz8t6yal1Ouvv84ZZ5xBy5YtGThwYFk3R9ux9LJuQHm1ZMkShg8fzgsvvEClSpXo1KkTRx99NPvuu29ZNy0lPv30U2655Rbmz59f1k1JuXfffZe3336biRMnEolEuOSSS3j11Vc55ZRTyrppKfHBBx/w3nvv8eKLL1JYWEirVq1o1qwZjRo1Kuumpczs2bOZOHEiJ554Ylk3JaXi8TjfffcdM2fOJD19x/v4+OmnnxgwYAATJkygdu3adOnShVmzZtGsWbOybpq2QyZY2+jdd9/lr3/9K7Vq1aJatWqcdtppvPzyy2XdrJQZP348AwYMIDMzs6ybknJ16tQhOzubSpUqkZGRwZ/+9CcWLlxY1s1KmaOOOorHH3+c9PR0fvnlF4qKiqhWrVpZNytlli9fzvDhw7n88svLuikp99133xGJRLj00ktp27YtTz75ZFk3KaVeffVVWrVqRb169cjIyGD48OEceuihZd0sbad2vK8gIcnJyaFOnTrrr2dmZjJ37twybFFqDRo0qKybUGb222+/9f+fP38+U6dO5dlnny3DFqVeRkYG9957L48++igtW7akbt26Zd2klOnfvz/XXnstixYtKuumpFxeXh7HHHMMt956K2vWrOHCCy9kn3324bjjjivrpqXEDz/8QEZGBhdffDG5ubk0b96ca665pqybpe2UCdY2isfjmy2LRCJl0BKVlW+//ZZu3brRp08f9t5777JuTspdffXVzJ49m0WLFjF+/Piybk5KTJgwgfr163PMMceUdVPKxOGHH87QoUOpVq0au+66K2eddRazZs0q62alTFFREbNnz+buu+9m/PjxfPbZZ0ycOLGsm6XtlB2sbVS3bl2WLl26/npOTs4OOVy2o/roo4/o2rUr119/PR06dCjr5qTUvHnz+Pe//w1A1apVOfXUU/n666/LuFWpMXXqVN555x3atWvHvffey+uvv86dd95Z1s1KmTlz5jB79uz11+Px+A41F2u33XbjmGOOYdddd6VKlSqcdNJJO9TIhZJjB2sbHXvsscyePZtly5axevVqXnnlFU444YSybpZSYNGiRVx11VUMGzaM1q1bl3VzUm7BggXccsst5Ofnk5+fz4wZMzjiiCPKulkp8dhjj/HSSy8xefJkrr76alq0aEHfvn3Lulkp8/vvvzN06FDWrl3LihUrmDhx4g5zcAdA8+bNefvtt8nLy6OoqIi33nqLgw46qKybpe3UjvPVI2R169bl2muv5cILL6SgoICzzjqLQw45pKybpRR45JFHWLt2LYMHD16/rFOnTpx77rll2KrUadasGZ9++int27cnLS2NU089dYfsaO6Imjdvvn7fx2IxOnfuzOGHH17WzUqZQw89lEsuuYTOnTtTUFDAcccdx5lnnlnWzdJ2KhLf0mQiSZIkbTOHCCVJkkJmB0uSJClkdrAkSZJCZgdLkiQpZHawJEmSQmYHS5IkKWR2sCRJkkJmB0tS0rKysjjhhBP49ttvy7opkrRdsoMlKWkvvfQSe++9N9OnTy/rpkjSdskOlqSkpaWlccQRR+wwP/IsScnytwglJW3NmjX861//wl/akqQtM8GSlLThw4dTt25dfvrpJ1auXFnWzZGk7Y4dLElJ+eSTT3j55ZcZNWoUNWrU4JtvvinrJknSdscOlqRiW7t2LTfddBO33XYbtWrV4oADDnAeliRtgR0sScU2cuRIDj/8cE488UQADjjgAL766quybZQkbYfsYEkqlrlz5/Lyyy/Tt2/f9csOPPBAEyxJ2oJI3MOAJEmSQmWCJUmSFDI7WJIkSSGzgyVJkhQyO1iSJEkhs4MlSZIUMjtYkiRJIbODJUmSFLL/B9J07RddsyZ8AAAAAElFTkSuQmCC\n", + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAJjCAYAAADQ22+yAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAABin0lEQVR4nO3dd3gUVd+H8Xs3hSJVJFQLPCo2RFTsimCjCIgVUQTsBbCiYC+ggiiCvYAFUQEVEAVRUKzY0MfeXhQV6VhCT7K77x+jlAdYsmayIeH+cOW62NnZzDkzW377PWcmkUQikUCSJEmhiZZ0AyRJksoaCyxJkqSQWWBJkiSFzAJLkiQpZBZYkiRJIbPAkiRJCllmSTdA2pL169ePjz76CICZM2dSr149ypcvD8CoUaNW/39TzjnnHK666ip23HHHja4zZMgQtt9+e4477rgit/sfPXv25MMPP2TatGlUqFAhtN8rSaVdxOtgSZuHli1bMmTIEBo3blzSTSmU+fPn06FDB5o0acLhhx/OqaeeWtJNkqTNhkOE0mbqnnvu4ayzzqJdu3ZcccUVLFq0iAsvvJBTTjmFli1b0qVLFxYvXgwExdkXX3zBBx98QKdOnejduzfHHXccbdq04f333wegT58+DBs2DIDGjRtzzz330KlTJ1q2bMnjjz8OQCwW47bbbuOoo47i+OOP58Ybb6RLly4bbN/o0aM58MAD6dixI08++SRrf1f77LPPOOmkkzj22GPp2LEj06dPT7q8UaNG/P7776sf/8/tDz74gPbt29OpUyfat29PXl4e/fr146STTqJNmza0bt2aGTNmALBs2TL69u3LMcccQ5s2bbjrrrtYvnw5zZo146efflr9u7t3786UKVPCOESStFEWWNJm7LfffmPs2LEMGjSIl19+mb322otRo0YxdepUypcvz/jx49d7zOeff86ZZ57JuHHjOPHEE7n33nvXWycvL4/q1avz7LPPMnToUO68805WrVrFmDFj+Oqrr3jppZd49tln+fXXXzfYroKCAkaPHk379u1p2bIlixcv5q233gIgPz+fiy66iIsuuoiXXnqJW265hVtvvZW8vLwNLo/H40n3wQ8//MCdd97Jiy++yFdffcWCBQsYNWoUEydOpGPHjjzyyCMADB06lFWrVjFx4kTGjRvHJ598whdffMFxxx3HmDFjAPjll1/46aefaNGiRUrHQZJS5RwsaTO21157kZkZvEy7du3Kxx9/zGOPPcasWbP44YcfaNKkyXqPqVu3LrvuuisAu+22G2PHjt3g7z7iiCMA2H333cnLy2P58uW8+eabdOjQgXLlygFwyimnMGLEiPUeO3XqVOLxOIceeiiZmZm0adOGJ554gubNm/P9998TjUY5/PDDAdhjjz2YMGECX3311QaXb0qdOnWoV68eAE2bNqVq1aqri78PPviArbbaCoD33nuPvn37kpGRQUZGBk899RQAOTk5nH766Vx66aWMGjWKE088kYyMjE1uV5KKwgJL2oxVrFhx9f/vuOMOPv/8c0444QT2339/CgoK2NAUyrUnxkcikQ2uA6wuoiKRCACJRGJ1MfePaHTDIfczzzzDypUrOfroo4EgEVu4cCE//PADGRkZq3/nP77//vuNLm/YsOE6y/Ly8ta5vfY+mDZtGv3796d79+4cccQRNGzYkBdffBGAzMzMdX7/3LlzKV++PA0aNKBRo0ZMnTqVCRMmrE6zJKk4OUQolRLvvPMOXbt25bjjjqNGjRq89957xGKxULfRvHlzXnzxRfLy8igoKNhg+vXTTz/x4YcfMnbsWF5//XVef/113nnnHfbdd1+eeOIJGjZsSCQS4d133wXgq6++omvXrhtdHo/H2Xrrrfniiy8AeO211zbavnfffZcWLVrQuXNnGjduzJQpU1bvgwMPPJCxY8cSj8fJy8ujV69eq8/Q7Ny5MwMHDqRJkybUqlUr1H0mSRtigiWVEhdddBEDBw7k/vvvJyMjg7333ptffvkl1G0cf/zx/PTTTxx33HFUrFiR+vXrr3f5hWeeeYYjjzyS7bbbbp3lPXr04LzzzuOyyy7jnnvu4dZbb2XgwIFkZWVxzz33kJ2dvdHl1157LTfffDNVqlThoIMOombNmhtsX6dOnbjiiito164dGRkZ7Lvvvrz66qvE43F69OhB//796dChA7FYjDZt2qxO2Fq0aMG1115Lp06dQt1fkrQxXqZB0mrvvPMOixcvpkOHDkBwna5y5crRu3fvEm5Z0XzyySdcd911vPTSS+sNU0pScbDAkrTa/Pnz6dOnD4sXLyYWi7HLLrtw4403Urly5ZJu2r921VVX8eGHHzJgwAD222+/km6OpC2EBZYkSSr1JkyYwAMPPEB+fj7dunXjtNNOK9H2OAdLkiSVavPnz2fw4MG88MILZGdn06lTJ/bff/+kfz6suFlgSZKkzUZubi65ubnrLa9SpQpVqlTZ4GPee+89DjjgAKpVqwbAMcccwyuvvEKPHj2Ks6lJlfoCa8XcBiXdBEklIE7yK8Cr7Ip6hSEq1Plp0yuFKD5v57Rt64kxPTf4Fyh69OhBz549N/iYBQsWrHP2cU5ODp9//nmxtbEwSn2BJUmSyo6uXbvSsWPH9ZZvLL0CNnhB5ZI+Y9gCS5IkbTaSDQVuTK1atfj4449X316wYAE5OTlhNy0l5qySJCmpeBr//RsHHXQQ06dP5/fff2fFihW8+uqrHHbYYSHvhdSYYEmSpFKtVq1aXHrppZxxxhnk5+dz4oknsueee5Zom0r9dbCc5C5tmZzkvuVyknv6J7mvmttw0yuFpFydH9O2reLks1SSJClkDhFKkqSk4pTqwa4SYYIlSZIUMhMsSZKUlHMeU2eCJUmSFDITLEmSlFSsdF9woESYYEmSJIXMBEuSJCXlWYSpM8GSJEkKmQWWJElSyBwilCRJScUcIkyZCZYkSVLITLAkSVJSTnJPnQmWJElSyEywJElSUl5oNHUmWJIkSSEzwZIkSUn5p55TZ4IlSZIUMhMsSZKUlNfBSp0JliRJUshMsCRJUlIxA6yUmWBJkiSFzARLkiQl5VmEqTPBkiRJCpkJliRJSipGpKSbUOqYYEmSJIXMAkuSJClkDhFKkqSk4l6mIWUmWJIkSSEzwUrBhMlRRoxeU5MuXRZhwUKYPCaf7Gy4aWAmP/0CiUSEdsfE6N65bJ3Ymqz/NbaGFh2yqLnNmq85XTvFaXtU2dkHd96fwWvTolSpHPRxh+0SDLwhxoChGcz4bM0E0IWLImxTI8GY4QUl1dRi8cOPEW4fksHSZZARhWsvj7Fbo2BfnHpuJqtWRcjMDG63OSpOt05l59gDPPtCBqPGZlAuGxpsn6DPJflUrQIrV8Htd2fy9bdR4gnYY9c4fS4poHy5km5xuF5/O8qDj2USjUCVygmu613AtvUSLFkKNw/MYtYvEeIJaHdMjG6dYyXd3FBt7Lmfnw+3D83g08+D98WD949zyXkxMjJKuMHFwEnuqbPASkG7Y+K0Oyb40MgvgLN6ZXJm5zg1toYBQzPIqZlg0M0xVqyA47tlsXeTBE12Lzu5arL+z/oFKleG0cPKVlGxts++jHD79QXstce6x/SqXms+TH6bC2f2yuKWvmXrA2bFSrjgikxuuLKAQw9I8MY7Ea7ul8m4EfmsWAG//hbhjfH5ZJXRd5SPPo3y+NOZPHH/KmrlwEuvRuk3KIs7bs5n2IhMYrEIzw7LI5GAa/tn8djITC44s+y8FlauCvr17KN5bFc/wVNjMrjjnkyG3p7PA8MzyamZ4I6bg+fCid3K0bRJvMy89yV77j87Nsoff0Z47rF84nE4s1cmr06L0vqIsvXlQv9O2t4OZ86cyeTJk5k3bx7RaJScnBwOPfRQGjdunK4mhOrxp6NsXR1ObB+8kK7sGSP292fqwsWQnw+VtiobbzAb8r/9/+9XUTKicPYlmfyVC0c2j3P26fEy800uLw++/SHCk6MyuHUwbFsPrriogDq11l3v5kGZnH5yjF12KlvHfvpHEerXTXDoAUG/Dj84Qb06QQHx5bcRKlaAnn0yWbQY9t8nQc9zYmUqwfnmuwj77xOnVk5w+4hD49xyRxb5+bB3kzh1ayeI/h3uNtopzo+zytbsi3gMSMDSZcHtFSsgOzv4f++eBWu990XIz4fKW5VIM4tFsud+l5PjdDo+TjQKf/wJS5ZGqFq5bL32/2GClbq0vAuMHDmSyy67DIDGjRuz++67A3DdddcxfPjwdDQhVH/8CU+OzqB3jzXfUCMRyMyEq/tlcGL3LPbdK84O25ZcG4vThvofi8EB+8a5f2ABw4YU8N5HUZ55oex8yCxcDM32TtDrnAJGPVpA493iXHJNJom13kvf+SDC/AUROh9f9r69/jw7wjZbw40DM+h8bibnX565+kN12XJo1jTOoJsKGPlgAfMWRBj6cBmprP+2+65xPvo0ypx5we3xkzLIz4/wZy4c2CzO9tsGT4Q58+Dp5zI5snnZSjArVoSrL8une49sjj6hHKPGZtLr3OD1/8973zX9sji5ezb77LVmf5QFyZ77AFmZMOShDI7tnMXW1RM03bPs9F1FE0kkEsX+bDjmmGMYN24cFSpUWGf5ihUr6NixI6+88sq//t0r5jYoavNS9uhTUX7+NbLRYaDly+HyGzJpvFuCC7uXrTda2HT/Aaa8GeGZFzIYNqTsDJOsLZGAQ9pmMXpYPvXqBMvOvyKT1kfE6dC67BVYj4yIMuypDB4ZXEDj3YJhkn53ZjJpVP7qJOMfX38X4fLrg/uKUzzNf7xj3MsZjB6XQSQCHdrEeGB4JmNHrKJa1eD+r7+LcMV12ZzQvoCzTi9br/sffoxw+bVZ3HdHPtvWS/DM8xmMm5jBs4/mEVkr2Fi+HK64IYvGuyW4oHvxvfajaTw/q7DP/fwCuPmODBJAvzRMEahQ56di38baPvslfYlBk+1+Tdu2ilNahggzMzMpKFj/xbZy5UqysrLS0YR/7f7hGUx7N3gHOfzgBBeeGePVN6Jc2XPdF9B7H0bYsWGCnG2Cb3utWsaZ+lbpT3AK2/+XXo2y838S7PyfoF5PJCCzlIcYa/e9fl1oeWicY49e86G+dh9//xO+/CbC4FvKTnG1dv//yo2ww3YJGu8WHN8WhyS4+Q6YPTeYf1VpK9inSdk59gAPDM/kzXeD1/C+TeOcfFyM49oGz/vFvwf3V60SrDt5apTb7s7iqovzaX1k2XgOrN3/X2ZHOPLwONvWC47xycfFuPO+TP78C775PspODePULEPvfYV97v+VG2Hragm23zZIstq3ijNgaBl48isUaSmwzj//fI477jgOPPBAatasCcDChQt5//33ufTSS9PRhH/twjNjXHjmmtu5S+CX3yI0+Z+Jzq9OizL1bbj2shj5+cHtA/Yt/W+0he3///0UYcqbUe68uYD8Anh2bAZtSvkZhGv3/f9+jNC9VyZNG8epVwdGj4+y038Sq+fk/PeLCLs3SvA/IW2ptnb/Fy0OTtz4+rsIuzVKBGdNRqBebfj4vxEefjLKsCEFZGXCiDFRjm5Zuo89wAVnFnDB3/2f9UuEcy/J5rknVlFpK3jkyUyOaRkjEoEp06IMvCeL++/IY7ddys7w0Nr9//CTKDcNyGLx71Bja5j2TpS6tRNUrwavTYvy+ttRrrmsgPz84Pb+pfy9r7DP/SlvRvj86wzu7ldANAoTp0RptnfZeQ6szTlYqUvLECHA/PnzmT59OgsWLCCRSFCrVi0OPPBAatWqtekHJ5HuIcIvv43Q9+ZMJjy97vBH7hLof1cG//dThEgk+JZzQffY6omvZcXG+r9iJdw+JIMvvo6SXwBHHR6n59mxdYYPSruXX40y/Oko8XiEnJoJbrxyzST3x5+N8vMvEW64smwNDa1txmcRBj+YsXqC85U9YjTdM0E8Dnc/lMHb06MUxIL5WH16xdYbOgxbuocIn30hGCJMJGCvxnGuuji4FEOH07JZsjRCzlqXKGnSOE7fS8rW8PioscFlKrKyoGrlBFddXMB/GiRYsgT635XFzL/f+w4/JM753QuK9b0vnUOEsPHnfn4+DLw3uExLNAJNGye47MIYFcoXf5vSPUT4yS/bpW1be2/3S9q2VZzSVmAVl5KYgyWp5KW7wNLmI90F1uYo3QXWR7/skLZtNdtuVtq2VZx8lkqSJIWsjF4WUJIkhSWeKEPzPdLEBEuSJClkJliSJCkpzyJMnQmWJElSyCywJEmSQuYQoSRJSiqWMI9JlXtMkiQpZCZYkiQpqbh5TMrcY5IkSSEzwZIkSUl5mYbUmWBJkiSFzARLkiQl5VmEqXOPSZIkhcwES5IkJRV3DlbKTLAkSZJCZoIlSZKSipnHpMw9JkmSFDITLEmSlJRnEabOPSZJkhQyEyxJkpSUf4swde4xSZKkkFlgSZIkhcwhQkmSlFQs4YVGU2WCJUmSFDITLEmSlJQXGk2de0ySJClkJliSJCmpuBcaTZl7TJIkKWQmWJIkKSnnYKXOPSZJkhQyEyxJkpSU18FKnQmWJElSyEp9ghXFqlpbpozIlv39yG/UUvr4x55T5x6TJEkKWalPsCRJUvGKeR2slLnHJEmSQmaCJUmSkoo73zllJliSJEkhs8CSJEkKmUOEkiQpKSe5p849JkmSFDITLEmSlJR/7Dl17jFJkqSQmWBJkqSk4v5pqpSZYEmSJIXMBEuSJCXlHKzUucckSZJCZoIlSZKSinsdrJS5xyRJkkJmgiVJkpKK+ceeU2aCJUmSFDITLEmSlJRzsFLnHpMkSQqZCZYkSUrKOVipM8GSJEkKmQmWJElKyjlYqXOPSZIkhcwCS5IkKWQOEUqSpKRiDhGmzD0mSZIUMhMsSZKUVNzLNKTMBEuSJClkJliSJCkp52Clzj0mSZIUMhMsSZKUVDzhHKxUmWBJkqQybciQIdxzzz2rb+fm5nLuuefSunVrTjvtNBYuXAhAXl4evXv3pnXr1nTs2JGZM2cCkEgkGDBgAK1ataJNmzbMmDFjk9u0wJIkSUnFiKbtJ0xLlizh6quvZvjw4essv/vuu9l3332ZNGkSJ510Ev379wdgxIgRVKhQgUmTJnH11VfTp08fACZPnszMmTOZOHEi9913H3369KGgoCDpti2wJEnSZiM3N5fZs2ev95Obm5vy75o6dSo77LAD3bt3X2f5tGnTaNeuHQDHHnssb731Fvn5+UybNo327dsD0KxZM/744w/mzJnDm2++SZs2bYhGozRo0IC6devy6aefJt22c7BSNOj+KK9Oi1K1cnB7h+0S3HFDjPx8uG1olE8+D2rWQ/aPc+l5cTIySrCxxej1tyNcc1sG0ycGFfztQ6PM+GxNvb5gEWxTA54fnrzCL43+t+9byrEf+UKEUeMjRCKwbV24qXecGtWD+w5pHyWn5pp1z+yU4NijEiXT0GLyzAtRRr8YHONt6yW44YoYNapDLAa3Dlnz/D9k/ziXXxAnUsamrGzsvS8eh7sfjvL2+1EiEdi+foLrLo+xdbUSbW6x+N/XPsAp52ayahVk/f1p2uaoON07xUuohcUnnXOwnnjiCe699971lvfo0YOePXum9LuOO+44gHWGBwEWLFhAzZrBm1ZmZiaVKlXi999/X2c5QM2aNZk3bx4LFiwgJydnveXJWGCl6L9fRhh4fYy99lj3w+OZsVH++DPCC48VEI9Dt14ZTJ4Woc0RZetDBuDn2XDnAxnE13oP6dMrDgQLfpsL3Xpl0r9v2SuuNtT3LeHYf/UdPD4qwgvD4lSuBHfcH+GeYRFuvCLBT79AlcrwwrCy96Hyj6+/gydGRRkzrIDKlYJi477hUa6/PM5Lr0aY9WuE54cXEE/AGRdl8NqbEY4+vOwcf9j4e9/YiRG+/j7CqIcLyM6Gux6MMuj+DG69OlZCLS0eG3rtL18Bs3+DaeMLVhdYKrquXbvSsWPH9ZZXqVJlo4+ZNGkSt9122zrLGjZsyOOPP17o7UajGx7Ui0ajJBLrv543tv4/fEqkIC8Pvv0hwhOjovQbHGHbegmuvChGnVpwxslxTj0eolH4409YsjSy+pteWbJiJfTtn8EVF8Xoc8uGI5qbBmVwxslxdtkpzY0rZhvr+5Zw7HdvBBNHxsnKhFWrYP5CqF8nuO+/X0bIiEK3i6P8lQtHNU9wXpdEmUrwdmsEE0YWrO7/gkUR6tUJ3nBj8eC5kZcPiTjkF0TIzi7hBocs2Xvfjg3gsvNjq/u8e6MEz44rW/Hdxl77X34boUIF6NEng4WLIxywT5xe58QpX64EG1tM4mmcUVSlSpWkxdSGtG7dmtatWxd6/ZycHBYtWkTt2rUpKChg6dKlVKtWjZycHBYuXMj2228PwMKFC8nJyaFWrVqrJ8KvvTwZ52ClYMFi2G/vBL3OiTHm0QL23C3Bxddk8k9hm5UJdz8UpU3nTGpUT7D3nmXrGyzALXdmcFK7ODs33HDf3v4gwrwFETofX/bSjGR93xKOfVYmTH0bWp4UZcbnETq2CfpYEIMD903w8B1xnhga592PIox8oWx9wELQ/9ffjnDUSZl88nmE41oHz/EOrRJUqQRHnphJyxMy2bZegsMPKlvHP9l7X5PdE+y2c7Be7hJ48IkMjm5etvq/sdf+suWwX9MEd94U45kHC5i7IMKQh/1YLQ2aN2/OuHHjAJg4cSL77rsvWVlZNG/enPHjxwPw8ccfU65cOerWrcthhx3GhAkTiMVi/Pzzz8yaNYvGjRsn3UZanglz5sxJ+lNa1K8D9w+I0WA7iESg2ylxfp0Dv601DHvJeXHeeamAurWh3+Ay9BUeeHZclIwMVn+wbshTY6Kc1TlWptILKFzfy/Kx/8cRh8K7L8a5sFuCc6+IEo/DSe0SXH1xguzsYKiw68lxpr5d9gosgJaHJnjrxQLO7xbj/N6ZxOPw4BNRtq4G08YW8NqYAnJzg+HEsqQw732//hZMDdi7cZxOHcvOF6xkr/0WBye49ZoYlbaCcuXgnNNivP5O2Tr2/4glImn7SYeLL76Y//73v7Rt25ann36a66+/HoAuXbqQl5dH27Zt6d+/PwMHDgSgVatW7LTTTrRv354LL7yQ/v37U758+aTbiCQ2NLAYsnbt2jFr1ixycnLWG8eMRCJMnTr1X//uVXMbFrV5Sd03PMq0d4MXTP26CVoeGqfd0UEfEgk4sE0m454oYO78CNWrJdhh2+BxH30a4bahGbzwWOmeh7R2/7OyEqxYGSEzA/ILYNavsFMDuG9AATnbwO9/wrGnZTLluQIqVijZdoehsH3/bW7JHPuMSPG+kd8zLMIb7wVvdv/ZPkGn4xLss2dwXywGex0Z5e1xcd56P0KjHRM0+k9w3+RpMPrFKMPuKt4P2ViieH//2se/4Q4JTukQX51MxmKwz1GZTBtbQPeLM+l7cYz9mgb3jZ8U4bU3o9x7e+meg1TY977aOfDhpxF635RB905xupWBCd6Ffe1//X2ESlvBvk2C/fLVtxGuvDmDl58u/vf9cnV+LPZtrO3S/3ZK27YG7/Vs2rZVnNIyB+uZZ56hc+fO3HDDDeyzzz7p2GRoLjozzkVnBm8YP/wYfENr2riA+nVg1PgoO/8nQe0cGP9KhM+/jjKkX4xoFF6eEmW/vUv/G83a/V/bb3Ph+O6ZjBm25o3kv19E2L1RokwUV1D4vo+dWDaPfc+zEvQ8K/jgmPEZ9L45yvOPxqleDV56LcKODaBaVfjhJ3jtrSh33xwnvyCY9N/2yNI/RLT28Z/xefDBOebRAqpXg5enrOn/rjsnmPxGhP2aJsgvgGnvRdlzt7LV/2Tvff/9MsKl12Yw4PoYh+xf+vsNhX/tv/FuhIefjDB8SIysTHhyTJRjWpb+177CkZYCq1KlSvTr148xY8aUugJrbTs1hL69YvTsGwwN1KqZYMB1wbfUM0+NM+DeKCednUkkAk0bx7n4nC3rhfbzbxHq1i7pVqTflnDs92kC556eoNslwXBJTg24p3/Qxwu7Jeh/NxzXPUpBARxzeIITjy0bH7T/2GfPBOd0iXPmJZlkZkDNbRLc3S/4kL3yohi3Dc2gfZdMMjJg/73jdO9cto5/sve++x+LkgCGPJzBkIeD9evVSXB3v9Kd4BXGSe3izJ4T5ZRzMonFoFnTBOefUbaO/T/8UzmpS8sQYXEq7iFCaXNV3EOEm7viHiKUNmfpHiK8+NNT07atIU2fSdu2ipOXaZAkSUnFE1v2F7p/wz0mSZIUMhMsSZKUVAznYKXKBEuSJClkJliSJCkpzyJMnQmWJElSyEywJElSUp5FmDr3mCRJUshMsCRJUlJxzyJMmQmWJElSyEywJElSUjHPIkyZCZYkSVLITLAkSVJSnkWYOveYJElSyCywJEmSQuYQoSRJSso/lZM6EyxJkqSQmWBJkqSkvNBo6kywJEmSQmaCJUmSknIOVupMsCRJkkJmgiVJkpLyQqOpc49JkiSFzARLkiQl5Rys1JlgSZIkhcwES5IkJeV1sFJngiVJkhQyEyxJkpSUc7BSZ4IlSZIUMhMsSZKUlAlW6kywJEmSQmaBJUmSFDKHCCVJUlIOEaau1BdYWZFS3wVJ/0I0YgAvafNldSJJkpIywUqdXwElSZJCZoIlSZKS8k/lpM4ES5IkKWQmWJIkKSnnYKXOBEuSJClkJliSJCkpE6zUmWBJkiSFzARLkiQlZYKVOhMsSZKkkJlgSZKkpEywUmeCJUmSFDITLEmSlFTCBCtlJliSJEkhs8CSJEkKmUOEkiQpKf/Yc+pMsCRJkkJmgiVJkpLyMg2pM8GSJEkKmQmWJElKyss0pM4ES5IkKWQmWJIkKSnnYKXOBEuSJClkJliSJCkp52ClzgRLkiQpZCZYkiQpKedgpc4ES5IkKWQmWJIkKalEoqRbUPqYYEmSJIXMBEuSJCUVxzlYqTLBkiRJCpkFliRJUsgcIpQkSUl5odHUmWBJkiSFzARrI6ZNh8EPQ14+NGoI/a6CSlutu86A+2DyNKhaJbi9w7Yw+EaIx+HOh+DN9yEage3rw01XwNbV0tyJIihK/wFOOAdWrYKsrOD2sUfCWaemq/VFU5i+fz8T+g2FpUshmgE3XQ67N9oyjv24V+CJMWtuL1kK8xfCG88F690yGL74FuIJaLIrXHcplC+X/n78W0V97r/6Jjz0VPD4urXg9quhetV09qBoinL8y2XDtQPhx18gEYcOreCczunvQ1EU5fjn5UH/ofD+J7BVBTj8IOjRHaJlIMrwQqOpiyQSpfvqFvF5O4f+O3//E9p1hZH3wQ71YdCDsGw53HDZuut1ugCuugia7rHu8jEvwcSp8NAAyM6GOx6ARb/DgGtCb2qxKGr/l6+Aw0+Ed8dDVikr4QvT9xUr4ehTgzfe5gfA1HeComriiC3n2P8jvwC69ISOreGU9nD3ozB3PtzWN7huzpX9giKz11lp7ca/VtTn/pffwgV94dn7oV4duO3e4IvGjZenrQtFUtTj338IRKJwdc/gfaBdNxh03fr7aXNV1ON/z3D4/Bu4t1/w5fKGQbDLjnDa8eG3NVr7+/B/aRJ7vXxd2rb137a3pG1bxSltdfWUKVMYMWIEv/zyyzrLR40ala4mFNq7H8EeuwQvMIBTO8BLU9a90FpeHnzzfzD8WTjuTOh1HcyZH9y34w5wxQXBByzAHo3W3FcaFLX/X3wDFSvA+VdB+27Bh8zKVWnvxr9SmL6/+xFsVy8orgBaHrwmvdgSjv3aHn0aalQPPlwBmjWB888IvrFnZMCuO5W9/id77r/4GpzQNiiuAHp0Kz3JLRT9+F/dC668IPj/wsXBvqpcqfjbHZaiHv+vvoc2LaFcueA1cMShMPnN9PejOCQS6fspK9JSYA0aNIinnnqKWbNm0alTJ8aPH7/6vmeffTYdTUjJvAVQJ2fN7Vo1YemyCMuWr1m2YDHs3xQuOxfGDoMmu0GPq4MnR9M9YPe/g7W/lsD9T0Krw9PahSIpav+XrYD99oIhN8OYh4JE466H096Nf6UwfZ/1K2yzNVwzAE48F868HApiwX1bwrH/xx9/wuOjoG/PNcsObgYNtg3+/9s8ePI5OObw4mxxuIr63J/1K8RicNHVwYfvLXfDVhXT3o1/rajHPxKBzMwguWzfPXgf+Of5UBoU9fjvuStMej1IvfLy4eUpQaGpLVNaCqw333yTRx99lOuuu46nn36aIUOGMGnSJAA2xxHKeHzDy9ceR69fBx4eCA22C95UzuwEv8wJPlT+8ctvQXy+T2Po3LF42xymova/5cEw8Npg3kK5cnDe6TDl7fS0vagK0/eCGLz1PpzcDp57GE4/Pkjr8vLWrFOWj/0/Rk+AlocEz4X/9dV3Qf9P6wgtDgq3jcWpqM/9ggJ4471gSPCFR4NC/Po70tP2MIR1/AdeC++N//tLxhPhtrE4FfX4n90ZdmwAp14IZ14Ge+1R+qZJbEwiEUnbT1mRlgIrkUgQiQQ7bYcdduChhx6if//+fPDBB6uXb07q1Fr3W8f8RVC1coKKFdYs+24mjJ+87uMSCcjMCP7/wSfBi+y4VsGb7WbYzY0qav/feBc++mzd5aXlTaYwfc+pEby5NtktuH3EIUFq8euc4HZZP/b/mPRGMPfmf708Fc66PPiGf16X4mtrcSjqcz9nGzikGdSsEXwod2wN//0qPW0PQ1GP/zsfwoJFwf+3qghtj4Cv0ztVqEiKevz/yoVup8CLj8NT9wQnN2xXLy1N12YoLQVWq1at6NKlC59//jkAO+20E0OGDOGSSy5Zb07W5uDgZvDZ1zBrdnB71ItBKrO2SARuHQqz5wa3nxkHjf4DtXPg0y+h53XB2UNndkpr00NR1P7PWwh33B/Mu4rF4PHR0LpFWrvwrxWm74fuD3PmBSkNBMVkJBJ8s90Sjj0EycQvv60/yXfytOB58eggOPaoYm9u6Ir63D+meXAG6R9/Bfe99lYwp6e0KOrxn/QG3Pd4UHDk5QW399+72JsdmqIe/zfehRsH/T1VYjk8MRralcLXwYaYYKUubWcRTp8+nZycHP7zn/+sXjZ37lyGDx/ONdf8+1OsiuMsQgjeJAc/DPn5sG294ANz9hy47o5g3B3gxVfhkachHgvG6vtdFZyWfeZlwWnqa0fn9WrDvf2LpanFoij9/+dSBdPeC4bT9m8K1168ZuL35q4wff/oMxj0ACxfCdlZwVlT++y55Rz7L76BK26ByU+v+9hjOgen7dequWZZ0z3g+kvT1/6iKspzH4IP3KfHBZcpqFsruC9nm5LqTeqKcvxzl8CNd8EPP0GEYJJ3z1J2mYKiHP+CArjhzqBIi8fhpGOh+ynF0850n0XY+MUb0ratL9rflLZtFScv0yBJUimT7gJr9/E3pm1bX3VI37aKUyn6XiFJklQ6lJKpx5IkqaSU7rGukmGCJUmSFDITLEmSlFRZOrsvXUywJEmSQmaBJUmSFDKHCCVJUlIOEabOBEuSJClkJliSJCkpr9KQOhMsSZKkkJlgSZKkpJyDlToTLEmSpJCZYEmSpOSchJUyEyxJkqSQmWBJkqSknIOVOhMsSZKkkJlgSZKkpBLOwUqZCZYkSVLITLAkSVJSzsFKnQmWJElSyEywJElSciZYKTPBkiRJCpkFliRJUsgssCRJUlKJRPp+wjRjxgxOOOEEOnToQNeuXfntt98AyM3N5dxzz6V169acdtppLFy4EIC8vDx69+5N69at6dixIzNnzvy7/wkGDBhAq1ataNOmDTNmzNjkti2wJElSmdS7d2/69+/P+PHjadeuHf369QPg7rvvZt9992XSpEmcdNJJ9O/fH4ARI0ZQoUIFJk2axNVXX02fPn0AmDx5MjNnzmTixIncd9999OnTh4KCgqTbtsCSJEnJJdL3k5uby+zZs9f7yc3NTanJeXl5XHzxxeyyyy4ANGrUiLlz5wIwbdo02rVrB8Cxxx7LW2+9RX5+PtOmTaN9+/YANGvWjD/++IM5c+bw5ptv0qZNG6LRKA0aNKBu3bp8+umnSbfvWYSSJGmz8cQTT3Dvvfeut7xHjx707Nmz0L8nOzubDh06ABCPx7n33ns58sgjAViwYAE1a9YEIDMzk0qVKvH777+vsxygZs2azJs3jwULFpCTk7Pe8mQssCRJUlLpvNBo165d6dix43rLq1SpstHHTJo0idtuu22dZQ0bNuTxxx8nLy9v9ZDeeeedt9HfEY1ueFAvGo2S2MDksI2t/w8LLEmStNmoUqVK0mJqQ1q3bk3r1q3XW75s2TIuuOACqlWrxgMPPEBWVhYAOTk5LFq0iNq1a1NQUMDSpUupVq0aOTk5LFy4kO233x6AhQsXkpOTQ61atVZPhF97eTLOwZIkScmlcQ5WmHr37s3222/PkCFDyM7OXr28efPmjBs3DoCJEyey7777kpWVRfPmzRk/fjwAH3/8MeXKlaNu3bocdthhTJgwgVgsxs8//8ysWbNo3Lhx0m1HEhvKvUqR+LydS7oJkiSlVbT292ndXoOnbtv0SiH56fS+ofyer7/+mo4dO7LjjjuSmRkM2OXk5PDII4/w559/0qdPH3799VcqV67MoEGDqF+/PqtWreL666/nyy+/JDs7m379+rH77ruTSCQYOHAgb731FgB9+/blkEMOSbp9CyxJkkqZdBdYO4y4PW3bmtWlT9q2VZwcIpQkSQqZk9wlSVJypXqsq2SYYEmSJIXMBEuSJG1C+q6DVVaYYEmSJIXMBEuSJCXnHKyUmWBJkiSFzAJLkiQpZA4RSpKk5BwiTJkJliRJUshMsCRJUnIJL9OQKhMsSZKkkJlgSZKkpBLOwUqZCZYkSVLITLAkSVJyJlgpM8GSJEkKmQmWJElKzrMIU2aCJUmSFDITLEmSlFTEOVgpM8GSJEkKmQmWJElKzgQrZSZYkiRJITPBkiRJyXkWYcpMsCRJkkJmgSVJkhQyhwglSVJyTnJPmQmWJElSyEywJElSciZYKTPBkiRJCpkJliRJSs4EK2UmWJIkSSEzwZIkScl5odGUmWBJkiSFzARLkiQlFXEOVspMsCRJkkJmgiVJkpIzwUqZCZYkSVLITLA2Ytp0GPww5OVDo4bQ7yqotNW667z2Ftz7GESjUKUS3HIlbFcvuO+g9lCr5pp1z+wE7Y5KX/uLqqj9f3osPPcyrFoFuzeCfldCdnb6+xGGou6L0qYo/V2yFK4dCD/+Aok4dGgF53QumX6EYUs79rDpPo97BZ4Ys+b2kqUwfyG88Vyw3i2D4YtvIZ6AJrvCdZdC+XLp78e/VZT+V68Kt9wNH38W3HfYAdD7Aoh4At4WKZJIJEp18Beft3Pov/P3P6FdVxh5H+xQHwY9CMuWww2XrVln5aqgiBo7DLavD4+Phukz4KEB8NMvcEFfeGVk6E1Li6L2/9W3YMgjweOrVIJLboDGu8A5p5VYl/61ou6L0qao/e0/BCJRuLonLF8B7brBoOug6R4l1aN/b0s79lC4Pq8tvwC69ISOreGU9nD3ozB3PtzWFxIJuLJfsF96nZXWbvxrRe3/2EkwfjIMuzMoMDtfCGedCq1ahN/WaO3vw/+lSTS45860beunnpenbVvFKW1DhLNmzWL+/PkAjBkzhn79+jFx4sR0bT4l734Ee+wSvMAATu0AL00J3jD+EYsFt5cuC24vXwHl/k5oPv0SMqLQ9WLo0B3uezxYv7Qoav9fnAzdToFqVYJv9TdeDu2PTm8fwlLUfVHaFLW/V/eCKy8I/r9wMeTlQeVK6Wt/mLa0Yw+F6/PaHn0aalQPiguAZk3g/DOC131GBuy6E8yZn562h6Go/Y/FYcXKIP3KywsKsNL8fFhbJJG+n7Jik0OEw4cP59577yUWi1G3bl0aNWq0+mfnnXemfv36m9zI448/zogRI4jH4xxwwAHMnTuXo446iueff56ffvqJiy66KJTOhGXeAqiTs+Z2rZqwdFmEZcsTq6PirSoG32pOvSgoJOJxGHlvcF9BDA7aN4iGV66C8/sEEXPXk9Lfl3+jqP2f9evfiVVvWLAI9tkTrjg//f0IQ1H3RWlT1P5GIpCZGSQXk9+EIw+BBtumvx9h2NKOPRSuz//44094fBQ8/+iaZQc3W/P/3+bBk8/BTVcUa5NDVdT+d2wFk6fB4ScEnwMHN4MWB6ej5docbTLBeuihhxg4cCCvvfYa/fr1Y//992f+/Pk88sgjHHfccey7776ceuqpSX/H888/z8SJE3nqqad45ZVXeOihhzjttNN44IEHmDx5cmidCUs8vuHl0bX21vcz4YEn4aUn4K0X4LzT4eLrg286J7eDay4O5hxVqQzdToYpb6en7WEoav/zC+C9j2HwjTDmYfgrNxg6KI2Kui9Km7D6O/BaeG88/LUE7n+ieNtcXLa0Yw+F6/M/Rk+AlodA/Trr3/fVd8HQ2WkdocVB4baxOBW1//c9HszDenscTHsueO97bFRxtLQEJCLp+ykjNllgVapUicMPP5ycnBz22WcfTj31VG688UaeeeYZPv74Y8aNG8e5556b9HfE43Gys7OpV68eZ555JuXKrZnxGNsMx87q1AqGN/4xfxFUrZygYoU1y975KJhX8s9k1s4d4Yef4M+/gjH472auWTeRCL7VlxZF7X/ONnDkoUFql50F7Y6Gz75Kbx/CUtR9UdoUtb/vfBiklhCkO22PgK/TO1UkNFvasYfC9fkfk94I5h79r5enwlmXw2Xnwnldiq+txaGo/X/tbTihTfC+V7kSHNcKPvi0eNuszdcmC6xzzz2XMWPGbPT++vXr06JF8hl8Rx99NKeffjqxWIyePXsC8O2339K5c2dat97AK7SEHdwMPvsaZs0Obo96EVr+T8y7287w0Wew6Pfg9tR3gm8y1asFb7D3DA/mZ6xcBSPHQutimORYXIra/2OaBzH5ylVBcTn17WBeQ2lU1H1R2hS1v5PeCL7FJxLBHJRJb8D+e6ezB+HZ0o49FK7PECSTv/y2/skLk6fBrUPh0UFwbCk6a/ofRe3/bjsFz3kIkvzX34UmuxVvm7X52uRZhE2bNiU/P5/DDjuMQw89lF133ZVGjRpRocIGSvokPvroI5o1WzNA/+OPP/Lrr7/SvHnzf9fyvxXHWYQAb74fnKqbnw/b1oPbr4bZc+C6O4IzhiAonJ5+AbKyoGpluPYS2KlBMMmx393BCzW/AFodDpecU7pO1S1K/2MxeHAETHo9mPS5207BPIz/ncNQWhRlX5RGRelv7hK48a7gS0YEOOJQ6Nl9w0MspcGWduyhcH3+4hu44haY/PS6jz2mc3DZgrUvUdN0D7j+0vS1v6iK0v8//grOpP36h+A5f+DecOVFkFUMIxjpPouw4d13pW1bP16ykdM2S5lNFli//vor3377Ld999x3fffcd3377LXPmzKF+/fqbxfyp4iqwJEnaXFlgbf42WVdvu+22bLvtthx11Jq8d/ny5Xz/fSmdWCFJklJTSk/cKEn/KrivWLEie+21V8hNkSRJKhtK0bltkiSpJJSlC4CmSymdeipJkrT5MsGSJEnJmWClzARLkiQpZCZYkiQpOROslJlgSZIkhcwES5IkJeVZhKkzwZIkSQqZCZYkSUouUYr+mO5mwgRLkiQpZCZYkiQpOedgpcwES5IkKWQWWJIkSSFziFCSJCXlZRpSZ4IlSZIUMhMsSZKUnAlWykywJEmSQmaCJUmSknIOVupMsCRJkkJmgiVJkpIzwUqZCZYkSVLITLAkSVJyJlgpM8GSJEkKmQmWJElKyrMIU2eCJUmSFDILLEmSpJBZYEmSJIXMOViSJCk552ClzARLkiQpZBZYkiRJIXOIUJIkJeVlGlJngiVJkhQyEyxJkpScCVbKTLAkSZJCZoIlSZKSM8FKmQmWJElSyEywJElSUp5FmDoTLEmSpJCZYEmSpORMsFJmgiVJkhQyEyxJkpSUc7BSZ4IlSZIUMhMsSZKUnAlWykywJEmSQmaCJUmSkjPBSpkJliRJUsgssCRJkkLmEKEkSUrKyzSkzgRLkiQpZCZYkiQpOROslJlgSZIkhcwES5IkJWeClTITLEmSpJCZYEmSpKQ8izB1JliSJEkhs8CSJEnJJdL4E6KPP/6Y448/nnbt2nH++efz119/AZCbm8u5555L69atOe2001i4cCEAeXl59O7dm9atW9OxY0dmzpwZdD+RYMCAAbRq1Yo2bdowY8aMTW7bAkuSJJVJffv2ZeDAgUyYMIEdd9yRYcOGAXD33Xez7777MmnSJE466ST69+8PwIgRI6hQoQKTJk3i6quvpk+fPgBMnjyZmTNnMnHiRO677z769OlDQUFB0m1bYEmSpKQiifT95ObmMnv27PV+cnNzU273xIkT2XHHHcnPz2f+/PlUqVIFgGnTptGuXTsAjj32WN566y3y8/OZNm0a7du3B6BZs2b88ccfzJkzhzfffJM2bdoQjUZp0KABdevW5dNPP026bSe5S5KkzcYTTzzBvffeu97yHj160LNnz5R+V1ZWFt999x3du3cnMzOTyy67DIAFCxZQs2ZNADIzM6lUqRK///77OssBatasybx581iwYAE5OTnrLU/GAkuSJCWXxrMIu3btSseOHddb/k/6tCGTJk3itttuW2dZw4YNefzxx2nUqBHvvfcezz77LJdeeinPPvvsBn9HNLrhQb1oNEoisf4O2Nj6/7DAkiRJm40qVaokLaY2pHXr1rRu3XqdZatWrWLKlCkceeSRALRv354BAwYAkJOTw6JFi6hduzYFBQUsXbqUatWqkZOTw8KFC9l+++0BWLhwITk5OdSqVWv1RPi1lyfjHCxJkpRcKTyLMDMzk5tuuokvv/wSCFKuvffeG4DmzZszbtw4IJinte+++5KVlUXz5s0ZP348EJyBWK5cOerWrcthhx3GhAkTiMVi/Pzzz8yaNYvGjRsn3354XZEkSdo8ZGRkMHjwYK6//npisRi1atVafbbgxRdfTJ8+fWjbti2VK1dm0KBBAHTp0oXrr7+etm3bkp2dzcCBAwFo1aoVn3/++eoJ8P3796d8+fJJtx9JbGhgsRSJz9u5pJsgSVJaRWt/n9btNb5scNq29cVdl6ZtW8XJBEuSJCUVKekGlELOwZIkSQqZCZYkSUquVE8mKhkWWBsxbToMfhjy8qFRQ+h3FVTaas39416BJ8asub1kKcxfCG88B9tsDQe1h1prrlXGmZ2g3VHpa39Rbcn935L7DkXv/z96Xgs528B1l6St6aELa1+UJkXpc6Wt4JbB8MW3EE9Ak13hukuhfLn09yMMm9oXACNfgGfGQyQC29WFm3tDjeol015tXpzkvgG//wntusLI+2CH+jDoQVi2HG64bMPr5xdAl57QsTWc0h5++gUu6AuvjAy9aWmxJfd/S+47FL3//3j0aRj+LLRuWXoLrLD2RWlS1D7f/SjMnQ+39YVEAq7sB9vXh15npbUboSjMvvjqO+h1PYwbBpUrwcD7g3VuuqL425fuSe5NLknfJPfP7i4bk9xLZA7W7bffXhKbLbR3P4I9dgleVACndoCXpgRvGBvy6NPBN5Z/3lQ//RIyotD1YujQHe57HGKxtDQ9FFty/7fkvkPR+w/wwSfwzodwSofib29xCmNflDZF7XOzJnD+GRCNQkYG7LoTzJmfnraHrTD7YvdGwZepypVg1aogyauW2vUxVYYV+xBh375911v2+uuv89dffwGsd2n7zcG8BVBnrQu01qoJS5dFWLY8sV48/Mef8PgoeP7RNcsKYnDQvtD7Ali5Cs7vE8TKXU9KS/OLbEvu/5bcdyh6/xcsglvvgUcGwegX09LkYlPUfVEaFbXPBzdb8//f5sGTz6UnzSkOhd0XWZkw5W247g7IzoKepTCtK5RSPdZVMoq9wKpWrRrjxo3j/PPPX33p+/fff5/99tuvuDf9r8XjG16+oT87NHoCtDwE6tdZs+zkdmv+n50N3U6GEc+Xng/ZLbn/W3LfoWj9zy+Ay26Cvj0hp0bxtTFdivpcKI3C6vNX3wVz8E7rCC0OCreN6ZLKvjjy0OBn9AQ45wqY/PSG19OWpdifAldddRV33XUXEydOpG7dunTs2JGqVavSsWPHDf4xx81BnVqwcPGa2/MXQdXKCSpWWH/dSW8E8w/WNn4yfDdzze1EAjJL0ekEW3L/t+S+Q9H6/+W38NtcGHAfdDwLRr0Ik16HawcWf7uLQ1GfC6VRGH1+eSqcdTlcdi6c16X42lrcCrMvfp4NMz5fc/uENsGQ6F9L0tfOtCmFfyqnpKWlxj7wwAN56KGHePrppxkwYACxzXxSysHN4LOvYdbs4PaoF6Hlweuv99cS+OU3aLrHust/+AnuGR7MvVm5CkaOhdYtir/dYdmS+78l9x2K1v+mewRnko0dFvyc0j6Y5N7vyvS0PWxFfS6URkXt8+RpcOtQeHQQHFuKzpzdkMLsi4WL4fKbg+FSgAmvwU4NoHrVtDZVm6m0hZjVqlVjyJAhNGzYkJo1a276ASWoRnXo3wcuuR7adoHvf4QrLwq+oXdca3z9l9lQs0YwBr+2i7pB1crBJOcO3aHp7nDSsWntQpFsyf3fkvsORe9/WbIl7oui9vmuh4PU9ro7gvU7ngU3p+/ks1AVZl/s2wTOOx3OuCRYNvF1uLd/iTa72EQS6fspK7xMgyRJpUy6L9OwV8/0Vcr/vadsXKahDHznkiRJxapURzElw/McJEmSQmaCJUmSkipLc6PSxQRLkiQpZBZYkiRJIXOIUJIkJecQYcpMsCRJkkJmgiVJkpJyknvqTLAkSZJCZoIlSZKSM8FKmQmWJElSyEywJElSciZYKTPBkiRJCpkJliRJSsqzCFNngiVJkhQyEyxJkpScCVbKTLAkSZJCZoIlSZKSiiSMsFJlgiVJkhQyEyxJkpScAVbKTLAkSZJCZoElSZIUMocIJUlSUl5oNHUmWJIkSSEzwZIkScmZYKXMBEuSJClkJliSJCkp52ClzgRLkiQpZCZYkiQpOROslJlgSZIkhcwES5IkJeUcrNSZYEmSJIXMBEuSJCVngpUyEyxJkqSQmWBJkqSknIOVOhMsSZKkkJlgSZKk5BJGWKkywZIkSQqZBZYkSVLIHCKUJElJOck9dSZYkiRJITPBkiRJyZlgpcwES5IkKWQmWJIkKalIvKRbUPqYYEmSJIXMBEuSJCXnHKyUmWBJkiSFzARLkiQl5XWwUmeCJUmSFDITLEmSlJx/7DllJliSJEkhM8GSJElJOQcrdSZYkiRJITPBkiRJyZlgpcwES5IkKWQWWJIkSSFziFCSJCXlJPfUmWBJkiSFzARLkiQl54VGU2aCJUmSFDITLEmSlJRzsFJngiVJkhQyEyxJkpScCVbKTLAkSZJCZoIlSZKScg5W6kywJEmSQmaCJUmSkosbYaXKBEuSJClkJlgbMW06DH4Y8vKhUUPodxVU2mrddQbcB5OnQdUqwe0dtoXBN665P3cJdOkF/a+CPXZJV8v/vcL0eVPrHdQeatVcs+6ZnaDdUfDdTLjpLlixEiLAJefAYQekpVuFFkb/AeYugE4XwLhhUL0a/N8s6H3LmvtjMfjhpwhDbklw9GHF3auiG/kCPDMeIhHYri7c3BtqVF9/vdfegnsfg2gUqlSCW66E7eqlv71FVZjnQbJ9srHXQFmQSMDVt8NODYJ+AaxcBbcMhi++DUKOJrvCdZdC+XIl29Z/qzDH/8VXYfizwfEvXw6u6RW8x+flQf+h8P4nsFUFOPwg6NE9eE2UegZYKSsLhz10v/8J19wOQ26BSU9B/bpw50Prr/fpl3DnDTB2WPCzdnH15vtw8vnw0y/panXRFLbPydb76ReoUnnN/hg7bM0Hy5X9gjfkscPg9mvg0huDN7DNRRj9Bxj3CpzeExYsiqxetuMO6+6Tg5tB2yNKR3H11XcwfBQ8cx9MeBy2rw9Dh62/3spVcFV/GHpL0McWBwcfNKVNYZ4HyfZJstdAaTdzFnS/FF55Y93lD46AghiMGw7jhwfPhYefKpEmFllhjv9Pv8AdD8DDdwTH9/wzoNd1wX0PPQVz5sOLj8Fzj8DCxfDMuHT3QpuLtBRYn3/++er/T58+ndtvv51Bgwbx2WefpWPzKXv3o+DbyA71g9undoCXpqz7p5jy8uCb/wu+xRx3ZvACmzN/zf1PPQ+39YWa26S37f9WYfq8qfU+/RIyotD1YujQHe57PEhrAJ5/BI44JPj/r3OChCNjMyrvw+j/gkUw9R14aMDGt/PxZzD5Tbjx8uLpR9h2bwSvjITKlWDVKpi/EKpVWX+9WCzYB0uXBbeXr4By2eltaxgK8zxItk+SvQZKu6fHQcfW0KrFusubNQmKjGgUMjJg153WfS8sTQpz/LOzgnQ2p0Zwe49GsOj34AvjV99Dm5ZQrlywP444NHi9lwWRRPp+yoq0fMTdcMMNAIwcOZJbb72V2rVrs80223D99dfz1FOb31edeQugTs6a27VqwtJlEZYtX7NswWLYvylcdm7wLabJbtDj6jUvxEfugKZ7pLfdRVGYPm9qvYIYHLRv0PcRQ4M3q6deCNbL/Hsw+uhTg2L07M7Bm/HmIoz+52wD9/QLEquNueMBuOTsDQ89bq6yMmHK23D4SfDx59CxzfrrbFURbrgMTr0IDjsenh4Ll5+X/rYWVWGfBxvbJ8leA6XddZdAh2PWX35wM2iwbfD/3+bBk8/BMYens2XhKczxr1cHDj8w+H8iEUwVaXFwUHjtuStMeh2WLQ8KrpenBCmWtkxpzRBGjx7Nk08+Sbdu3ejWrRsjR47cLAuseHzDy9ceR69fBx4eCA22C8bhz+wEv8wJ3mBKo8L0eVPrndwOrrkYsrODYZJuJwcfQv+IRODVZ2Dy0/DI08E8hc1FGP3flE+/hD/+gmOPTK1tm4MjD4XpL8JF3eCcK9bfD9/PhAeehJeegLdegPNOh4uvXz8B3Nylcnw3tE829Rooy776Drr0hNM6QouDSro1/04qx3/5Crj0Bvj5N7ild7Ds7M6wYwM49UI48zLYa4+gGNeWKS0FVkFBAfF4nBo1alCxYsXVy7Ozs4luhrP/6tRa91vH/EVQtXKCihXWLPtuJoyfvO7jEgnI3IxSmU0ZOgw6nhX8PPfypvsMyffN+MnBfvlHIhEkV3n58PLUNW9e9evAgfvANz8UX98KI+z+b8qk14MEYDN8yq9j7f1yxc0wY80IPye0CYZ//lqy7mPe+ShIbP+Z1N65I/zwE/z5V/raHYbCHN+fZ298n2zsNVAarf082NC8u7W9PBXOujxI9M/rkp72FYfCvr7nzIfOF0E0A564OyimAf7KhW6nwIuPw1P3QPWqpfNEjw1KJNL3U0ak5a2+evXqNG/enP/7v/9bPVw4ffp0OnXqRKtWrdLRhJQc3Aw++xpmzQ5uj3oRWh687jqRCNw6FGbPDW4/Mw4a/Qdq51Bq9DprzUTcZx/YdJ8h+b754Se4Z3gw52TlKhg5Flq3CKLzocNg4tRgvQWL4INPg7kbJSns/m/KR5/BAXuH0/bitPZ+6dQBLr8Z/vgzuG/Ca8EZZNWrrvuY3XYO+rfo9+D21HeCQrp6tXS2vOgKc3wXLt74PtnYa6A0Wvt50Ousja83eVrwXvjoIDi2lE/oL8zx/zMXzugFRx0Gd92w7tmSb7wLNw4KaoRly+GJ0WXnJAelLpJIpK9c/PHHH8nNzWWvvfZixowZLFmyhMMPP7xIvzM+b+dwGvc/3nw/OFU3Px+2rQe3Xw2z58B1f585AsGpuo88DfFYMFbf7yqoW2vd33PEKTDkptJxmYYN9blaFfjy23X7vbH1VqyEfncHb1D5BdDq8OByDJFIMIR0893Bm04kAud0htYtS7CzG1DU/q9t1+YR3hufWKfA2PsYmDiidBXhEHx5eHpckM7WrAHXXxoUT/+7X0aOhadfgKwsqFoZrr0kKDxKm8K89je2T5K9BsqKvrete5mGYzrDkqXrXpqi6R7BPimNNnX8HxwRFNE7N1z3ccPvgspbwQ13Bsc/HoeTjoXupxRPO6O1vy+eX7wRLY5JcvZOyN6YfFXatlWc0lpgFYfiKrAkSdpcWWBt/krp7ABJkpQ2pTqKKRmb+XRbSZKk0scES5IkJRUp3bOJSoQJliRJUshMsCRJUnIbuQirNs4ES5IkKWQmWJIkKSnnYKXOBEuSJClkJliSJCk5A6yUmWBJkiSFzARLkiQl5xyslJlgSZIkhcwES5IkJRUxwEqZCZYkSVLILLAkSZJCZoElSZKSSyTS91MMvv76a/bYY4/Vt/Py8ujduzetW7emY8eOzJw58+9uJhgwYACtWrWiTZs2zJgxY/Vjhg8fTqtWrTjmmGN49dVXN7lN52BJkqQya8WKFdx8883k5+evXjZixAgqVKjApEmT+Oijj+jTpw9jxoxh8uTJzJw5k4kTJ/Lzzz9z7rnnMmnSJL7++mtefPFFxo8fz9KlSznllFPYb7/9qFat2ka3a4IlSZKSisTT95Obm8vs2bPX+8nNzf1Xbb/99tvp1q3bOsumTZtG+/btAWjWrBl//PEHc+bM4c0336RNmzZEo1EaNGhA3bp1+fTTT3nrrbc46qijKFeuHDVq1GC//fZj2rRpSbdrgiVJkjYbTzzxBPfee+96y3v06EHPnj1T+l1Tp05l5cqVtGrVap3lCxYsoGbNmqtv16xZk3nz5rFgwQJycnI2uLxx48brLU/GAkuSJCWXxguNdu3alY4dO663vEqVKht9zKRJk7jtttvWWdawYUOWLl3K448/XqjtRqNREhvoZ7LlyVhgSZKkzUaVKlWSFlMb0rp1a1q3br3OsjFjxvDQQw9x2mmnrV7WoUMHRo4cSU5ODgsXLmT77bcHYOHCheTk5FCrVi0WLly4ev1kyxs0aJC0Tc7BkiRJySXS+BOSk046iSlTpjB+/HjGjx8PwPjx46lUqRLNmzdfvezjjz+mXLly1K1bl8MOO4wJEyYQi8X4+eefmTVrFo0bN+awww7j1VdfZcWKFfz++++8//77HHjggUm3b4IlSZK2KF26dOH666+nbdu2ZGdnM3DgQABatWrF559/vnoCfP/+/Slfvjx77rkn7du358QTT6SgoIBevXpRq1atpNuIJDY0sFiKxOftXNJNkCQpraK1v0/r9o4+8Ja0bevV6delbVvFySFCSZKkkDlEKEmSkivdg10lwgRLkiQpZCZYkiQpuXhJN6D0McGSJEkKmQmWJElKKuIcrJSZYEmSJIXMAkuSJClkDhFKkqTkHCJMmQmWJElSyEywJJVKI5ZsU9JNKFGVoytKugkl5oGddizpJpS419J92QQTrJSZYEmSJIXMBEuSJCXnhUZTZoIlSZIUMhMsSZKUlBcaTZ0JliRJUshMsCRJUnImWCkzwZIkSQqZCZYkSUrOBCtlJliSJEkhM8GSJEnJmWClzARLkiQpZCZYkiQpOa/knjITLEmSpJBZYEmSJIXMIUJJkpSUfyondSZYkiRJITPBkiRJyZlgpcwES5IkKWQmWJIkKbm4CVaqTLAkSZJCZoIlSZKScw5WykywJEmSQmaCJUmSkjPBSpkJliRJUshMsCRJUnImWCkzwZIkSQqZCZYkSUrO62ClzARLkiQpZCZYkiQpuUS8pFtQ6phgSZIkhcwCS5IkKWQOEUqSpOS8TEPKTLAkSZJCZoIlSZKS8zINKbPA+h/TpsPghyEvHxo1hH5XQaWtCr9eLAYD7oN3Pgr+3/0U6NQheMwHn8AdD0JBAZQrB9f0gj13TW//NqYw/d7YOitXwS2D4Ytvg9dgk13hukuhfDlYsAiuvh0W/Q7xOJzdGdofXTJ9TNaHwq6X7PjOmg3X3g5/5kLFCnD71dBw+yBZHzIMprwVrLfHLnDDZVCh/Jrtfft/cE5veHts8e+DVI18AZ4ZD5EIbFcXbu4NNaqvv973M6HfUFi6FKIZcNPlsHuj9Lc3mQWzCnj1wWWsWp4gEoXWPSpRZ8f13wZ//Sqf1x5ZRjwGmdkRjjl/K+rs9O/fLvNXJnj5nqXMnxkjkUjQottWNDowG4AfZ+QxbcRy4jGIRKHFGRVpuE/2v95WKub9FOfFB/NYuSxBNBqhY89s6u307wc28lYmeGFIHnN+jJOIQ6vuWex+ULDfvp8R49Un8onHE0QicEy3bHbeJyOsrhTaQR2aceUTPTiuWtcN3r/7wbtw4eBuZGRmkLcyj3t7DuP7GT/+6+2Vq5DNZY9cwI5NdyASjfJon6d4b/xHABx20oGcds0JAPy1aAlDzn+I3/5v3r/eljYvDhGu5fc/4ZrbYcgtMOkpqF8X7nwotfVGvRh80L74GIx+CJ58Dj7/JvigvuwmuPkKGDcczu8CV/VPZ+82rjD9TrbOgyOgIBb0a/zwoOB6+KngvsGPBEXkuOHwyB1w012wcHE6e1e4PhR2vY0dX4ArbwmKrZeehB7dodf1QXH12tvw3kfwwjCY8ASsXAkjngseU1AAj4+Gs6+A5SvSsRdS89V3MHwUPHMfTHgctq8PQ4etv96KlXDWFXDWqUE/LzgDevdLe3OTyl+Z4JnrcjnghAqcNbQah3SqyPhBSza47ot3LqVl9604+55qHHhCeSYMXlqkbb/99HKyy0c478FqnHpLFSY/sJTcRTFWLoszftBS2l1aibPvqUa7SyoxduBSVi0v/rQgb2WC4deu5LATs+h1bwVanprJqDtWFel3Th2ZT3YFuOyhCpzVvxzj78/nr0VxVi5L8OzAVZx0eTa97q3AiZeV45nbV6Wln2urt2Ntzr3jDKLRjX/09RnRk0eueorz9+7NqIHj6f14jyJts8uNJ7Ni2UrO2v1Srjr6Fnrddzbb1NuaajlVufiBc7n22Ns4b68reHfsB/S456wibatYJRLp+ykj0lZgvf322+Tm5gIwbtw4br75Zp5//vl0bb5Q3v0oSBd2qB/cPrUDvDRl/eOdbL0pb8PxrSEzE6pWhjYtYcKrkJ0F056H3XYO1ps9B6pVSW//NqYw/U62TrMmcP4ZEI1CRgbsuhPMmR+sF4vB0mXBeitWQWZGsF5JKM7jO38h/PgLtDkieMxhBwRFx9c/wNGHwcj7gufAsuWw+M81x/7rH4Lk5+6b0rILUrZ7I3hlJFSuBKtWBf3c0PP23Y9gu3rQ/IDgdsuDYfCNaW3qJv34aT7V62SwY7MgHdpp/yyOv6ryBteNx2Hl0uC6P3krIDNrzX3vjlrOsIv/5NGef/Jcv1yWLF73+kBvjVzOWyOXr7Psu+l57HVMEFlWzcmgQdNsvnk7j3gBHHPBVtTcPkh5ttkuAxKwIrf4rzn0wycxtq4TZZdmQYq06wEZnNq33Or733g2n3t6rmBojxWMuHkVuf/TzylP5THlqbx1ln01PUazVkFfquVE2WnvKJ+/FSNWAMddmE2t7YMXf852ERIJWJabvg/TchWyuWpELx66/Imk60UzolSqHsTaFSqXJ29l/ur7Ol99PPd/PIAHP7mDG1/oTY0660a5XW44iS43nLTOskOO24+Jj0wBYOGvi5jx6uc0P/kg/lzwFyfXPpuFsxcTzYiSs31NchcXrZDX5iUtQ4T9+/fnm2++YfDgwdx999188cUXHHHEEbz22mt88803XHvtteloxibNWwB1ctbcrlUTli6LsGx5Yp1hpGTrzVsItf/nvu9mBv/PygyGyk44B/74C+66oXj7U1iF6XeydQ5utmb5b/OCVOemK4Lbl50Lp/eEV6bBH3/ClRdueHgpHYrz+M5dADnbrFs81q4ZFCS77xwc+5EvBEOFOdvAkYcF6+y5a/Dz29zi6XMYsjKDwvK6O4IisecGvmTP+hW22RquGRDsj8qV4Irz09/WZH7/LcZW1aK8PGQp838qoPxWUVp2r7jBdY+9eCue67eE1x5ezsplcU69Jagqv5i6igWzYnS/qyrRjAifvrKSiUOXcspNyb8t5S6KU6XmmidH5W2iLFkUp2LVKLsdtqaoeeupFWxdL0q12sU/dLbotwSVq0d4/u5VzP0xQflK0PrMoPj8ZGoB82bFufDu8mRkRPhwUgHPD8mj+83lk/7OvxYmqLbNmn5WrRHhr0UJtqoaYc/maz5upozIZ5t6Ubaunb5vW5c8eB4vP/waP37+c9L17jzrAW4c25sL7+5OpWpbcdXRtwBwZJfD2GGP7eixf1/isThtzjmSyx45n2uOvS3p76u5bQ0W/romtl/422K2qV8DgFhBjJ33acgtE/pSrkI2fVtvJsMaG1KGkqV0SUuB9e677zJhwgQyMjKYNm0ao0ePJjs7m1NOOYVjjz02HU0olPhGvjT+b+KSbL0N3Zex1uO32RrefB6++h7OvBT+swM02PZfNTc0hel3Ydb56jvoeS2c1hFaHBQs690vGDY69bhgaK3rxdBk95KZe1acx3dj7z1r/+7TjofOHYMi6+LrYcTQTbd5c3HkocHP6AlwzhUw+el1+1YQg7feh8fvhia7wdR34PyrYOooyE7PdKJNiscSzJyRx2m3VqFeo0p8/34eo27M5aLHqpOZFVm93tI/4ky8Zxmn316VOjtl8t30PF64bQnnP1ydHz7KY+73BQy/5C8guLh1/qrg4D/XL5c/58dZ9kfwJPn+/Tyq1Ypy4rVVNvj8iKxVQ8VjCaY8upyZH+fRuX96ou14DL77OMbZt5Vju10y+Hp6AY/fsJKrHq/Atx/E+PX7OPddvHL1uvl/jx6OuGUVf8yPs+SPoFNfvx+jeq0oXa4rt8F+RtfqZyyW4OVH8vn+4xhn31pu/ZWLSbsLjiZWEGPyY29Qa/uaG12vWk5VLn34PK44/Aa+n/EjB3VoxvVjLqd7o14c0HYfGu23I/d/NAAIkq5yFYMn9w3P96ZOgxyq164GwMEd9mPuTwu46YQ7iGwgso/H1ryRfD/jR06pew77HrMX/V7qyxn/uYhlfy1f7zEqfdJSYJUvX57FixeTk5NDjRo1WL58OdnZ2axYsYLMzJKdZz90GLzxXvD/pctg54Zr7pu/CKpWTlCxwrqPqVNrzbyb/12vTq115xgtWAS1cmDJUnj/Ezjq7+Ri952h0Y7ww48lX2Al609h13l5ajDR/dqL4dijgmV//AmffAGP3RXc3qE+HLQvfPxZ+gqsdB3fOjlBOplIBJPBIUivatcMJrDH48HwcCQCJ7VdMwdrc7T2PvvP9kFxvM+ewe0T2gTz6P5aAtWrrnlMTg1osF1QXAEccQhcNxB+nRN8iSgpbz61nB8+CIaxVixJUKN+BvUaBeN9Ox+QzctD4c95MbbZds370K9f5VMlJ7p6UnujA7OZ8kiExb/GSMThgBMrsE+bIMkpyE+wcmlQVZx4bVAY/TM8eNhpa9KxqjWjLP09TqXqwYft0sVxchoElceKpXFeuDWYC9b1zqpUrFJ8qc5rI/L45oMYAMtzoWb9CNvtErRjtwMzeWFIHr/PTRCPQ/OTMjmgbdbqfq74e7pal+uCwuif4cEjT19TQVetGWHJHwkqbx28CHIXJ6jTMOjPiiUJRt66ikQCLrirPFtVWVPUFoeuN53Cge32BSA/r4ByFbN58JM7yMzOJLtC8P9r2t7K4rl/rH5M40N3ZcHPi1ZPan9v/EdcMLgb2+1aj2hGlFEDx/PSg68CkJWdSaXqlQC46YQ7AFYPD464aczq37ngl0VsXacaf8z/E4Bt6m7NzM9mUaNOdRo03o6PX/0MgI8n/5flucup+5/a/PDJv59UX2xMsFKWlny2R48enHjiiQwYMICGDRvSpUsXbr31Vk4++WS6d++ejiZsVK+zYOyw4OfZB+Czr4OkBYIJzS0PXv8xBzfb+HpHHAwvTAwmL+cugYlTgw+baBSuHRAUHAA//AQ//bJ5nEWYrD+FWWfyNLh1KDw6aE1xBVCtajCENvnN4PYff6a3uIL0Hd/aObBtXZj4erDeOx8Gx3znhsGQ2TUDgjlZAOMmwwF7F2+/i2LtfdapA1x+c3DsACa8Bjs1WLe4Ajh0f5gzL0gxAT76LCgm69dJa9PX0/z0ipx9TzXOvqca3QdX5a/5ceb+XwEAv3yZTyQC1WqtOxyX0yCTRT/HWPxbUIj89l0++asSbF0vSsO9s/hs8kpWLQ8SiLeeWs6Ld2563sxO+2fz6SvBEyB3UYyZM/LYab9sCvITPHtdLtVqZ3DqLVWKtbgCOKpLMMm8170VuOju8vwxP8FvPwR9+emLGESgeu0IO++TwUeTC1j59yT010bkM3rQpifA73ZABh9OCvbvX4vifD8jxi77Z1CQn2DYtauoXivCmf3LFXtxBfDEDaM4f+/enL93b3oe0Jdz97yc8/fuzTVtbyVvRR7n7917neIK4MfPf2aHPbal3k7BE3eX/XakXMVyzP5+Lh+/+hmtzzqCipWDb2Rdbz6Fq57sucl2TH/xI9qeG7wxblNva/ZttRfvvzSDrPJZXPPspdT9T20Amhy+OxmZGfzyzewwd4NKUCSRSE9Z+uuvvzJlyhR+/vlnYrEY22yzDS1atGDPPfcs0u+Nz9s5pBYG3nw/OD0/Px+2rRecal+tCnz5bTAHZeyw5OsVFMDAB+C9j4P7TmkPZ3YKHvPhf+GO+4PhlOwsuPTczeeDdkP9mT2ncH0+pnOQ0NVaK3lvugdcf2mQ3vQbEly6IBqB00+Ak9uVTB+T9SGM4ztrNlx/RzC/rlx2MA9t97+fnvcMDwrRjAzYcYcg6atebU27fpsL7bvDjFfSuDMK6Zlx8PS44ASFmjWC41q/zvr77KPPYNADsHxl8Py+uuea5Ks4jFiyTcqP+eXLfF4fvpy8lQkys+Coc7di292zWLI4zqgbcznlxipUrhHlm3dW8c4zKyACWeUiHHFmRbbdPYtEPMHbz6zgm3fyiESgSs0obXtuReVtks+ZyluR4JX7lzJvZox4PMEhp1Rkjxbl+PKNVbx451Jqbp9BZK3aqv3llcjZIXm6Xzla9NNOf/oixsRhQQGZkQXtzstmh90ziMcTTH06ny/fDoquajUjHH9xNlW3SV4ArlqRYNy9ecyZGVymoUWnLJq2zOTTNwoYfUcetbaPrDNkePLl5ajdIPWi8oGddkz5Mf+otX1NHvniLtpX6QJAjTrV6f/y1avTrMNOPIDTrzuJRCLBquWreKj3CL5691sikQhdbjiJw048kEQiwYJfFnHXOQ+yeM7vSbdXfqvyXHz/Oey4dwOiGVGe7v88U0e+DcAhx+/P6dedCAlY+ucyHrjscWb+d1ah+vFafMymVwpR6zoXpW1bk+bel7ZtFae0FVjFJewCS1Lp8G8KrLIkjAKrtCpKgVVWWGBt/rzQqCRJSq50ZzElwguNSpIkhcwES5IkJWeClTITLEmSpJBZYEmSJIXMIUJJkpRc3CHCVJlgSZIkhcwES5IkJZVIbOSPtGqjTLAkSZJCZoIlSZKScw5WykywJEmSQmaCJUmSkvNCoykzwZIkSQqZCZYkSUou7lmEqTLBkiRJCpkJliRJSs45WCkzwZIkSQqZCZYkSUoq4RyslJlgSZIkhcwES5IkJeccrJSZYEmSJIXMAkuSJClkDhFKkqTk/GPPKTPBkiRJCpkJliRJSi7hZRpSZYIlSZIUMhMsSZKUVMI5WCkzwZIkSQqZCZYkSUrOOVgpM8GSJEkKmQmWJElKyjlYqTPBkiRJCpkJliRJSs45WCkzwZIkSQpZJJFIOLAqSZIUIhMsSZKkkFlgSZIkhcwCS5IkKWQWWJIkSSGzwJIkSQqZBZYkSVLILLAkSZJCZoElSZIUMgssSZKkkFlgSZIkhcwCS5IkKWQWWJIkSSGzwJIkSQqZBVYRTJgwgTZt2nDUUUcxcuTIkm5O2i1dupRjjz2W2bNnl3RT0u7ee++lbdu2tG3bloEDB5Z0c9JuyJAhtGnThrZt2/LYY4+VdHNKxIABA+jTp09JNyPtzjjjDNq2bUuHDh3o0KEDn332WUk3Ka1ef/11jj/+eFq1akW/fv1KujnajGWWdANKq/nz5zN48GBeeOEFsrOz6dSpE/vvvz877rhjSTctLT777DOuvfZaZs2aVdJNSbv33nuPd955h7FjxxKJRDj77LN57bXXOOqoo0q6aWnx4Ycf8v777/Piiy9SUFBAmzZtaN68OQ0bNizppqXN9OnTGTt2LIcffnhJNyWtEokEP/74I9OmTSMzc8v7+Pj111+54YYbGDNmDDVq1KBr1668+eabNG/evKSbps2QCda/9N5773HAAQdQrVo1KlasyDHHHMMrr7xS0s1Km9GjR3PDDTeQk5NT0k1Ju5o1a9KnTx+ys7PJysriP//5D3PmzCnpZqXNfvvtx5NPPklmZiaLFy8mFotRsWLFkm5W2vz5558MHjyY888/v6SbknY//vgjkUiEc845h/bt2/PUU0+VdJPS6rXXXqNNmzbUrl2brKwsBg8eTJMmTUq6WdpMbXlfQUKyYMECatasufp2Tk4On3/+eQm2KL369+9f0k0oMTvttNPq/8+aNYuJEyfy7LPPlmCL0i8rK4uhQ4cyfPhwWrVqRa1atUq6SWlz/fXXc+mllzJ37tySbkra5ebmcuCBB3LjjTeycuVKzjjjDBo0aMDBBx9c0k1Li59//pmsrCzOOussFi5cSIsWLbjkkktKulnaTJlg/UuJRGK9ZZFIpARaopLyww8/cOaZZ3LVVVexww47lHRz0q5Xr15Mnz6duXPnMnr06JJuTlqMGTOGOnXqcOCBB5Z0U0pE06ZNGThwIBUrVmTrrbfmxBNP5M033yzpZqVNLBZj+vTp3HHHHYwePZovvviCsWPHlnSztJmywPqXatWqxaJFi1bfXrBgwRY5XLalmjFjBt26dePyyy+nY8eOJd2ctJo5cybffPMNABUqVODoo4/mu+++K+FWpcfEiRN599136dChA0OHDuX111/n1ltvLelmpc3HH3/M9OnTV99OJBJb1FysbbbZhgMPPJCtt96a8uXLc8QRR2xRIxdKjQXWv3TQQQcxffp0fv/9d1asWMGrr77KYYcdVtLNUhrMnTuXiy66iEGDBtG2bduSbk7azZ49m2uvvZa8vDzy8vKYOnUq++yzT0k3Ky0ee+wxXnrpJcaPH0+vXr1o2bIlV199dUk3K22WLFnCwIEDWbVqFUuXLmXs2LFbzMkdAC1atOCdd94hNzeXWCzG22+/ze67717SzdJmasv56hGyWrVqcemll3LGGWeQn5/PiSeeyJ577lnSzVIaDBs2jFWrVnH77bevXtapUydOPfXUEmxV+jRv3pzPPvuM4447joyMDI4++ugtstDcErVo0WL1sY/H43Tu3JmmTZuWdLPSpkmTJpx99tl07tyZ/Px8Dj74YE444YSSbpY2U5HEhiYTSZIk6V9ziFCSJClkFliSJEkhs8CSJEkKmQWWJElSyCywJEmSQmaBJUmSFDILLEmSpJBZYElKWevWrTnssMP44YcfSropkrRZssCSlLKXXnqJHXbYgcmTJ5d0UyRps2SBJSllGRkZ7LPPPlvMH3mWpFT5twglpWzlypW8/PLL+Je2JGnDTLAkpWzw4MHUqlWLX3/9lWXLlpV0cyRps2OBJSkln376Ka+88gr33HMPlStX5vvvvy/pJknSZscCS1KhrVq1ir59+3LTTTdRrVo1dtllF+dhSdIGWGBJKrQhQ4bQtGlTDj/8cAB22WUXvv3225JtlCRthiywJBXK559/ziuvvMLVV1+9etmuu+5qgiVJGxBJeBqQJElSqEywJEmSQmaBJUmSFDILLEmSpJBZYEmSJIXMAkuSJClkFliSJEkhs8CSJEkK2f8DRQL6GXqKYn0AAAAASUVORK5CYII=\n", "text/plain": [ "
    " ] }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_61_11.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png" } }, "output_type": "display_data" @@ -2457,6 +2468,47 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "We can then compute the second derivative of the cost function, which in our case is the second derivative\n", + "of the means squared error. This leads to" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\frac{\\partial^2 C(\\boldsymbol{\\beta})}{\\partial \\boldsymbol{\\beta}^T\\partial \\boldsymbol{\\beta}} =\\frac{2}{n}\\boldsymbol{X}^T\\boldsymbol{X}.\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This quantity defines was what is called the Hessian matrix (the second derivative of a function we want to optimize).\n", + "\n", + "The Hessian matrix plays an important role and is defined for the mean squared error as" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\boldsymbol{H}=\\boldsymbol{X}^T\\boldsymbol{X}.\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The Hessian matrix for ordinary least squares is also proportional to\n", + "the covariance matrix. As we will see in the chapter on Ridge and Lasso regression, This means that we can use the Singular Value Decomposition of a matrix to find\n", + "the eigenvalues of the covariance matrix and the Hessian matrix in\n", + "terms of the singular values.\n", + "\n", + "\n", "The residuals $\\boldsymbol{\\epsilon}$ are in turn given by" ] }, @@ -2574,7 +2626,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_132_0.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_136_0.png" } }, "output_type": "display_data" @@ -3060,9 +3112,7 @@ "**pandas** again, rather extensively in order to organize our data.\n", "\n", "The difference now is that we use **Scikit-Learn's** regression tools\n", - "instead of our own matrix inversion implementation. Furthermore, we\n", - "sneak in **Ridge** regression (to be discussed below) which includes a\n", - "hyperparameter $\\lambda$, also to be explained below." + "instead of our own matrix inversion implementation." ] }, { @@ -3080,23 +3130,19 @@ "Mean squared error: 12.36\n", "Variance score: 1.00\n", "Mean absolute error: 2.83\n", - "[ 0. 618.32047562 -861.13519106 1404.91549644] -11.057088709963637\n", - "Mean squared error: 197.93\n", - "Variance score: 1.00\n", - "Mean absolute error: 11.69\n", - "[ 0. 28.18220995 282.79902342 842.30879705] 12.946893955211749\n" + "[ 0. 618.32047562 -861.13519106 1404.91549644] -11.057088709963637\n" ] }, { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
    " ] }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_179_1.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_183_1.png" } }, "output_type": "display_data" @@ -3163,18 +3209,6 @@ "print('Mean absolute error: %.2f' % mean_absolute_error(Energies, ytilde))\n", "print(clf.coef_, clf.intercept_)\n", "\n", - "# The Ridge regression with a hyperparameter lambda = 0.1\n", - "_lambda = 0.1\n", - "clf_ridge = skl.Ridge(alpha=_lambda).fit(X, Energies)\n", - "yridge = clf_ridge.predict(X)\n", - "EoS['Eridge'] = yridge\n", - "# The mean squared error \n", - "print(\"Mean squared error: %.2f\" % mean_squared_error(Energies, yridge))\n", - "# Explained variance score: 1 is perfect prediction \n", - "print('Variance score: %.2f' % r2_score(Energies, yridge))\n", - "# Mean absolute error \n", - "print('Mean absolute error: %.2f' % mean_absolute_error(Energies, yridge))\n", - "print(clf_ridge.coef_, clf_ridge.intercept_)\n", "\n", "fig, ax = plt.subplots()\n", "ax.set_xlabel(r'$\\rho[\\mathrm{fm}^{-3}]$')\n", @@ -3183,8 +3217,6 @@ " label='Theoretical data')\n", "ax.plot(EoS['Density'], EoS['Eols'], alpha=0.7, lw=2, c='m',\n", " label='OLS')\n", - "ax.plot(EoS['Density'], EoS['Eridge'], alpha=0.7, lw=2, c='g',\n", - " label='Ridge $\\lambda = 0.1$')\n", "ax.legend()\n", "save_fig(\"EoSfitting\")\n", "plt.show()" @@ -3197,13 +3229,11 @@ "The above simple polynomial in density $\\rho$ gives an excellent fit\n", "to the data. \n", "\n", - "We note also that there is a small deviation between the\n", - "standard OLS and the Ridge regression at higher densities. We discuss this in more detail\n", - "below.\n", "\n", "\n", "## Splitting our Data in Training and Test data\n", "\n", + "\n", "It is normal in essentially all Machine Learning studies to split the\n", "data in a training set and a test set (sometimes also an additional\n", "validation set). **Scikit-Learn** has an own function for this. There\n", @@ -3213,7 +3243,13 @@ "postpone a discussion of this splitting to the end of these notes and\n", "our discussion of the so-called **bias-variance** tradeoff. Here we\n", "limit ourselves to repeat the above equation of state fitting example\n", - "but now splitting the data into a training set and a test set." + "but now splitting the data into a training set and a test set.\n", + "\n", + "Let us study some examples. The first code here takes a simple\n", + "one-dimensional second-order polynomial and we fit it to a\n", + "second-order polynomial. Depending on the strength of the added noise,\n", + "the various measures like the $R2$ score or the mean-squared error,\n", + "the fit becomes better or worse." ] }, { @@ -3228,14 +3264,125 @@ "name": "stdout", "output_type": "stream", "text": [ + "[ 2.00247083e+00 -3.54710530e-03 5.02096350e+00]\n", "Training R2\n", - "0.9999869956119286\n", + "0.99620088364924\n", "Training MSE\n", - "5.745136489050356\n", + "0.008171076530362356\n", "Test R2\n", - "0.9999787681537219\n", + "0.9968966158779216\n", "Test MSE\n", - "9.712199063818309\n" + "0.006530396683779109\n" + ] + } + ], + "source": [ + "import os\n", + "import numpy as np\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "\n", + "def R2(y_data, y_model):\n", + " return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)\n", + "def MSE(y_data,y_model):\n", + " n = np.size(y_model)\n", + " return np.sum((y_data-y_model)**2)/n\n", + "\n", + "x = np.random.rand(100)\n", + "y = 2.0+5*x*x+0.1*np.random.randn(100)\n", + "\n", + "\n", + "# The design matrix now as function of a given polynomial\n", + "X = np.zeros((len(x),3))\n", + "X[:,0] = 1.0\n", + "X[:,1] = x\n", + "X[:,2] = x**2\n", + "# We split the data in test and training data\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n", + "# matrix inversion to find beta\n", + "beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train\n", + "print(beta)\n", + "# and then make the prediction\n", + "ytilde = X_train @ beta\n", + "print(\"Training R2\")\n", + "print(R2(y_train,ytilde))\n", + "print(\"Training MSE\")\n", + "print(MSE(y_train,ytilde))\n", + "ypredict = X_test @ beta\n", + "print(\"Test R2\")\n", + "print(R2(y_test,ypredict))\n", + "print(\"Test MSE\")\n", + "print(MSE(y_test,ypredict))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Alternatively, you could write your own test-train splitting function as shown here." + ] + }, + { + "cell_type": "code", + "execution_count": 25, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "# equivalently in numpy\n", + "def train_test_split_numpy(inputs, labels, train_size, test_size):\n", + " n_inputs = len(inputs)\n", + " inputs_shuffled = inputs.copy()\n", + " labels_shuffled = labels.copy()\n", + "\n", + " np.random.shuffle(inputs_shuffled)\n", + " np.random.shuffle(labels_shuffled)\n", + "\n", + " train_end = int(n_inputs*train_size)\n", + " X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]\n", + " Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]\n", + "\n", + " return X_train, X_test, Y_train, Y_test" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "But since **scikit-learn** has its own function for doing this and since\n", + "it interfaces easily with **tensorflow** and other libraries, we\n", + "normally recommend using the latter functionality.\n", + "\n", + "\n", + "As another example, we apply the training and testing split to \n", + "to the above equation of state fitting example\n", + "but now splitting the data into a training set and a test set." + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Training R2\n", + "0.9999850282256434\n", + "Training MSE\n", + "6.653725417896576\n", + "Test R2\n", + "0.9999887274726137\n", + "Test MSE\n", + "5.071447584032141\n" ] } ], @@ -3352,7 +3499,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 27, "metadata": { "collapsed": false, "editable": true @@ -3375,7 +3522,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 28, "metadata": { "collapsed": false, "editable": true @@ -3387,7 +3534,7 @@ "dict_keys(['data', 'target', 'feature_names', 'DESCR', 'filename'])" ] }, - "execution_count": 26, + "execution_count": 28, "metadata": {}, "output_type": "execute_result" } @@ -3411,7 +3558,7 @@ }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 29, "metadata": { "collapsed": false, "editable": true @@ -3432,7 +3579,7 @@ }, { "cell_type": "code", - "execution_count": 28, + "execution_count": 30, "metadata": { "collapsed": false, "editable": true @@ -3458,7 +3605,7 @@ "dtype: int64" ] }, - "execution_count": 28, + "execution_count": 30, "metadata": {}, "output_type": "execute_result" } @@ -3477,7 +3624,7 @@ }, { "cell_type": "code", - "execution_count": 29, + "execution_count": 31, "metadata": { "collapsed": false, "editable": true @@ -3500,7 +3647,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_191_1.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_199_1.png" } }, "output_type": "display_data" @@ -3524,7 +3671,7 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": 32, "metadata": { "collapsed": false, "editable": true @@ -3536,7 +3683,7 @@ "" ] }, - "execution_count": 30, + "execution_count": 32, "metadata": {}, "output_type": "execute_result" }, @@ -3549,7 +3696,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_193_1.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_201_1.png" } }, "output_type": "display_data" @@ -3572,7 +3719,7 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": 33, "metadata": { "collapsed": false, "editable": true @@ -3587,7 +3734,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_195_0.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_203_0.png" } }, "output_type": "display_data" @@ -3618,7 +3765,7 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 34, "metadata": { "collapsed": false, "editable": true @@ -3638,7 +3785,7 @@ }, { "cell_type": "code", - "execution_count": 33, + "execution_count": 35, "metadata": { "collapsed": false, "editable": true @@ -3676,7 +3823,7 @@ }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 36, "metadata": { "collapsed": false, "editable": true @@ -3735,7 +3882,7 @@ }, { "cell_type": "code", - "execution_count": 35, + "execution_count": 37, "metadata": { "collapsed": false, "editable": true @@ -3750,7 +3897,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_202_0.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_210_0.png" } }, "output_type": "display_data" @@ -3767,135 +3914,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Splitting our Data in Training and Test data\n", - "\n", - "\n", - "It is normal in essentially all Machine Learning studies to split the\n", - "data in a training set and a test set (sometimes also an additional\n", - "validation set). **Scikit-Learn** has an own function for this. There\n", - "is no explicit recipe for how much data should be included as training\n", - "data and say test data. An accepted rule of thumb is to use\n", - "approximately $2/3$ to $4/5$ of the data as training data. We will\n", - "postpone a discussion of this splitting to the end of these notes and\n", - "our discussion of the so-called **bias-variance** tradeoff. Here we\n", - "limit ourselves to repeat the above equation of state fitting example\n", - "but now splitting the data into a training set and a test set.\n", - "\n", - "Let us study some examples. The first code here takes a simple\n", - "one-dimensional second-order polynomial and we fit it to a\n", - "second-order polynomial. Depending on the strength of the added noise,\n", - "the various measures like the $R2$ score or the mean-squared error,\n", - "the fit becomes better or worse." - ] - }, - { - "cell_type": "code", - "execution_count": 36, - "metadata": { - "collapsed": false, - "editable": true - }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[ 2.05802766 -0.30626706 5.27262964]\n", - "Training R2\n", - "0.9967292151090247\n", - "Training MSE\n", - "0.007931713723267314\n", - "Test R2\n", - "0.9966744029509663\n", - "Test MSE\n", - "0.006968283076248407\n" - ] - } - ], - "source": [ - "import os\n", - "import numpy as np\n", - "import pandas as pd\n", - "import matplotlib.pyplot as plt\n", - "from sklearn.model_selection import train_test_split\n", - "\n", - "\n", - "def R2(y_data, y_model):\n", - " return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2)\n", - "def MSE(y_data,y_model):\n", - " n = np.size(y_model)\n", - " return np.sum((y_data-y_model)**2)/n\n", - "\n", - "x = np.random.rand(100)\n", - "y = 2.0+5*x*x+0.1*np.random.randn(100)\n", - "\n", - "\n", - "# The design matrix now as function of a given polynomial\n", - "X = np.zeros((len(x),3))\n", - "X[:,0] = 1.0\n", - "X[:,1] = x\n", - "X[:,2] = x**2\n", - "# We split the data in test and training data\n", - "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)\n", - "# matrix inversion to find beta\n", - "beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train\n", - "print(beta)\n", - "# and then make the prediction\n", - "ytilde = X_train @ beta\n", - "print(\"Training R2\")\n", - "print(R2(y_train,ytilde))\n", - "print(\"Training MSE\")\n", - "print(MSE(y_train,ytilde))\n", - "ypredict = X_test @ beta\n", - "print(\"Test R2\")\n", - "print(R2(y_test,ypredict))\n", - "print(\"Test MSE\")\n", - "print(MSE(y_test,ypredict))" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Alternatively, you could write your own test-train splitting function as shown here." - ] - }, - { - "cell_type": "code", - "execution_count": 37, - "metadata": { - "collapsed": false, - "editable": true - }, - "outputs": [], - "source": [ - "# equivalently in numpy\n", - "def train_test_split_numpy(inputs, labels, train_size, test_size):\n", - " n_inputs = len(inputs)\n", - " inputs_shuffled = inputs.copy()\n", - " labels_shuffled = labels.copy()\n", - "\n", - " np.random.shuffle(inputs_shuffled)\n", - " np.random.shuffle(labels_shuffled)\n", - "\n", - " train_end = int(n_inputs*train_size)\n", - " X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:]\n", - " Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:]\n", - "\n", - " return X_train, X_test, Y_train, Y_test" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "But since **scikit-learn** has its own function for doing this and since\n", - "it interfaces easily with **tensorflow** and other libraries, we\n", - "normally recommend using the latter functionality.\n", - "\n", - "\n", - "\n", - "\n", "## Reducing the number of degrees of freedom, overarching view\n", "\n", "Many Machine Learning problems involve thousands or even millions of\n", @@ -4509,7 +4527,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_214_0.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter1_218_0.png" } }, "output_type": "display_data" diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1.py b/doc/LectureNotes/_build/jupyter_execute/chapter1.py index f7147cccd..be4c4230e 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter1.py +++ b/doc/LectureNotes/_build/jupyter_execute/chapter1.py @@ -7,38 +7,37 @@ -Our emphasis throughout this series of lectures -is on understanding the mathematical aspects of -different algorithms used in the fields of data analysis and machine learning. +Our emphasis throughout this series of lectures is on understanding +the mathematical aspects of different algorithms used in the fields of +data analysis and machine learning. -However, where possible we will emphasize the -importance of using available software. We start thus with a hands-on -and top-down approach to machine learning. The aim is thus to start with -relevant data or data we have produced -and use these to introduce statistical data analysis -concepts and machine learning algorithms before we delve into the -algorithms themselves. The examples we will use in the beginning, start with simple -polynomials with random noise added. We will use the Python -software package [Scikit-Learn](http://scikit-learn.org/stable/) and -introduce various machine learning algorithms to make fits of -the data and predictions. We move thereafter to more interesting -cases such as data from say experiments (below we will look at experimental nuclear binding energies as an example). -These are examples where we can easily set up the data and -then use machine learning algorithms included in for example -**Scikit-Learn**. +However, where possible we will emphasize the importance of using +available software. We start thus with a hands-on and top-down +approach to machine learning. The aim is thus to start with relevant +data or data we have produced and use these to introduce statistical +data analysis concepts and machine learning algorithms before we delve +into the algorithms themselves. The examples we will use in the +beginning, start with simple polynomials with random noise added. We +will use the Python software package +[Scikit-Learn](http://scikit-learn.org/stable/) and introduce various +machine learning algorithms to make fits of the data and +predictions. We move thereafter to more interesting cases such as data +from say experiments (below we will look at experimental nuclear +binding energies as an example). These are examples where we can +easily set up the data and then use machine learning algorithms +included in for example **Scikit-Learn**. These examples will serve us the purpose of getting started. Furthermore, they allow us to catch more than two birds with a stone. They will allow us to bring in some programming specific -topics and tools as well as showing the power of various Python -libraries for machine learning and statistical data analysis. +topics and tools as well as showing the power of various Python +libraries for machine learning and statistical data analysis. -Here, we will mainly focus on two -specific Python packages for Machine Learning, Scikit-Learn and -Tensorflow (see below for links etc). Moreover, the examples we -introduce will serve as inputs to many of our discussions later, as -well as allowing you to set up models and produce your own data and -get started with programming. +Here, we will mainly focus on two specific Python packages for Machine +Learning, Scikit-Learn and Tensorflow (see below for links etc). +Moreover, the examples we introduce will serve as inputs to many of +our discussions later, as well as allowing you to set up models and +produce your own data and get started with programming. @@ -270,7 +269,17 @@ quality figures. Feel free to explore the extensive [gallery](https://matplotlib.org/gallery/index.html) of examples. In this example we plot our original values of $x$ and $y$ as well as the prediction **ypredict** ($\tilde{y}$), which attempts at fitting our -data with a straight line. +data with a straight line. Note also that **Scikit-Learn** requires a +matrix as input for the input values $x$ and $y$. In the above code we +have solved this by declaring $x$ and $y$ as arrays of dimension +$n\times 1$. + +In the code here we have also made a new array for $x\in [0,1]$. Our +prediction is computed for these values, meaning that they were not +included in the data set used to *train* (or fit) the model. +This is a recurrring theme in machine learning and data analysis. We would like to train a model on a specific given data set. +Thereafter we wish to apply it to data which were not included in the training. Below we will encounter this again in the so-called *train-validate-test* spliting. We will typically split our data into different sets, oen for training, one for validation and finally, our data from the untouched test vault! + The Python code follows here. @@ -285,6 +294,7 @@ x = np.random.rand(100,1) y = 2*x+np.random.randn(100,1) linreg = LinearRegression() linreg.fit(x,y) +# This is our new x-array to which we test our model xnew = np.array([[0],[1]]) ypredict = linreg.predict(xnew) @@ -384,8 +394,7 @@ plt.show() Depending on the parameter in front of the normal distribution, we may have a small or larger relative error. Try to play around with different training data sets and study (graphically) the value of the -relative error. Note also that **Scikit-Learn** requires a matrix as input for the input values $x$ and $y$. In the above code we have -solved this by declaring $x$ and $y$ as arrays of dimension $n\times 1$. +relative error. As mentioned above, **Scikit-Learn** has an impressive functionality. We can for example extract the values of $\alpha$ and $\beta$ and @@ -728,7 +737,7 @@ Energies = Masses['Ebinding'] print(Masses) The next step, and we will define this mathematically later, is to set up the so-called **design matrix**. We will throughout call this matrix $\boldsymbol{X}$. -It has dimensionality $p\times n$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit. +It has dimensionality $n\times p$, where $n$ is the number of data points and $p$ are the so-called predictors. In our case here they are given by the number of polynomials in $A$ we wish to include in the fit. # Now we set up the design matrix X X = np.zeros((len(A),5)) @@ -738,9 +747,12 @@ X[:,2] = A**(2.0/3.0) X[:,3] = A**(-1.0/3.0) X[:,4] = A**(-1.0) -Note well that we have made life simple here. We perform a fit in terms of the number of nucleons only. A more sophisticated fit can be done by including an explicit dependence on the number of protons and neutrons in the asymmetry and Coulomb terms. +Note well that we have made life simple here. We perform a fit in +terms of the number of nucleons only. A more sophisticated fit can be +done by including an explicit dependence on the number of protons and +neutrons in the asymmetry and Coulomb terms. We leave this as an exercise to you the reader. -With **scikitlearn** we are now ready to use linear regression and fit our data. +With **Scikit-Learn** we are now ready to use linear regression and fit our data. clf = skl.LinearRegression().fit(X, Energies) fity = clf.predict(X) @@ -768,7 +780,7 @@ ax.legend() save_fig("Masses2016") plt.show() -As a teaser, let us now see how we can do this with decision trees using **scikit-learn**. Later we will switch to so-called **random forests**! +As a teaser, let us now see how we can do this with decision trees using **Scikit-Learn**. Later we will switch to so-called **random forests**! #Decision Tree Regression @@ -801,6 +813,17 @@ plt.show() print(Masses) print(np.mean( (Energies-y_1)**2)) +With a deeper and deeper tree level, we can almost reproduce every +single data point by increasing the max depth of the tree. +We can actually decide to make a decision tree which fits every single point. +As we will +see later, this has the benefit that we can really train a model which +traverses every single data point. However, the price we pay is that +we will easily overfit. That is, if we apply our model to unseen data, +we will most likely fail miserably in our attempt at making +predictions. As an exercise, try to make the tree level larger by adjusting the maximum depth variable. When printing out the predicition, you will note that the binding energy of every nucleus is accurately reproduced. + + The **seaborn** package allows us to visualize data in an efficient way. Note that we use **scikit-learn**'s multi-layer perceptron (or feed forward neural network) functionality. @@ -1291,6 +1314,27 @@ $$ \frac{\partial\log{\vert\boldsymbol{A}\vert}}{\partial \boldsymbol{A}}=(\boldsymbol{A}^{-1})^T. $$ +We can then compute the second derivative of the cost function, which in our case is the second derivative +of the means squared error. This leads to + +$$ +\frac{\partial^2 C(\boldsymbol{\beta})}{\partial \boldsymbol{\beta}^T\partial \boldsymbol{\beta}} =\frac{2}{n}\boldsymbol{X}^T\boldsymbol{X}. +$$ + +This quantity defines was what is called the Hessian matrix (the second derivative of a function we want to optimize). + +The Hessian matrix plays an important role and is defined for the mean squared error as + +$$ +\boldsymbol{H}=\boldsymbol{X}^T\boldsymbol{X}. +$$ + +The Hessian matrix for ordinary least squares is also proportional to +the covariance matrix. As we will see in the chapter on Ridge and Lasso regression, This means that we can use the Singular Value Decomposition of a matrix to find +the eigenvalues of the covariance matrix and the Hessian matrix in +terms of the singular values. + + The residuals $\boldsymbol{\epsilon}$ are in turn given by $$ @@ -1523,9 +1567,7 @@ before, with the same initializations and declarations. We use also **pandas** again, rather extensively in order to organize our data. The difference now is that we use **Scikit-Learn's** regression tools -instead of our own matrix inversion implementation. Furthermore, we -sneak in **Ridge** regression (to be discussed below) which includes a -hyperparameter $\lambda$, also to be explained below. +instead of our own matrix inversion implementation. # Common imports import os @@ -1587,18 +1629,6 @@ print('Variance score: %.2f' % r2_score(Energies, ytilde)) print('Mean absolute error: %.2f' % mean_absolute_error(Energies, ytilde)) print(clf.coef_, clf.intercept_) -# The Ridge regression with a hyperparameter lambda = 0.1 -_lambda = 0.1 -clf_ridge = skl.Ridge(alpha=_lambda).fit(X, Energies) -yridge = clf_ridge.predict(X) -EoS['Eridge'] = yridge -# The mean squared error -print("Mean squared error: %.2f" % mean_squared_error(Energies, yridge)) -# Explained variance score: 1 is perfect prediction -print('Variance score: %.2f' % r2_score(Energies, yridge)) -# Mean absolute error -print('Mean absolute error: %.2f' % mean_absolute_error(Energies, yridge)) -print(clf_ridge.coef_, clf_ridge.intercept_) fig, ax = plt.subplots() ax.set_xlabel(r'$\rho[\mathrm{fm}^{-3}]$') @@ -1607,8 +1637,6 @@ ax.plot(EoS['Density'], EoS['Energy'], alpha=0.7, lw=2, label='Theoretical data') ax.plot(EoS['Density'], EoS['Eols'], alpha=0.7, lw=2, c='m', label='OLS') -ax.plot(EoS['Density'], EoS['Eridge'], alpha=0.7, lw=2, c='g', - label='Ridge $\lambda = 0.1$') ax.legend() save_fig("EoSfitting") plt.show() @@ -1616,13 +1644,11 @@ plt.show() The above simple polynomial in density $\rho$ gives an excellent fit to the data. -We note also that there is a small deviation between the -standard OLS and the Ridge regression at higher densities. We discuss this in more detail -below. ## Splitting our Data in Training and Test data + It is normal in essentially all Machine Learning studies to split the data in a training set and a test set (sometimes also an additional validation set). **Scikit-Learn** has an own function for this. There @@ -1634,6 +1660,77 @@ our discussion of the so-called **bias-variance** tradeoff. Here we limit ourselves to repeat the above equation of state fitting example but now splitting the data into a training set and a test set. +Let us study some examples. The first code here takes a simple +one-dimensional second-order polynomial and we fit it to a +second-order polynomial. Depending on the strength of the added noise, +the various measures like the $R2$ score or the mean-squared error, +the fit becomes better or worse. + +import os +import numpy as np +import pandas as pd +import matplotlib.pyplot as plt +from sklearn.model_selection import train_test_split + + +def R2(y_data, y_model): + return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2) +def MSE(y_data,y_model): + n = np.size(y_model) + return np.sum((y_data-y_model)**2)/n + +x = np.random.rand(100) +y = 2.0+5*x*x+0.1*np.random.randn(100) + + +# The design matrix now as function of a given polynomial +X = np.zeros((len(x),3)) +X[:,0] = 1.0 +X[:,1] = x +X[:,2] = x**2 +# We split the data in test and training data +X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2) +# matrix inversion to find beta +beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train +print(beta) +# and then make the prediction +ytilde = X_train @ beta +print("Training R2") +print(R2(y_train,ytilde)) +print("Training MSE") +print(MSE(y_train,ytilde)) +ypredict = X_test @ beta +print("Test R2") +print(R2(y_test,ypredict)) +print("Test MSE") +print(MSE(y_test,ypredict)) + +Alternatively, you could write your own test-train splitting function as shown here. + +# equivalently in numpy +def train_test_split_numpy(inputs, labels, train_size, test_size): + n_inputs = len(inputs) + inputs_shuffled = inputs.copy() + labels_shuffled = labels.copy() + + np.random.shuffle(inputs_shuffled) + np.random.shuffle(labels_shuffled) + + train_end = int(n_inputs*train_size) + X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:] + Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:] + + return X_train, X_test, Y_train, Y_test + +But since **scikit-learn** has its own function for doing this and since +it interfaces easily with **tensorflow** and other libraries, we +normally recommend using the latter functionality. + + +As another example, we apply the training and testing split to +to the above equation of state fitting example +but now splitting the data into a training set and a test set. + import os import numpy as np import pandas as pd @@ -1854,89 +1951,6 @@ print('R2 score is {}'.format(r2)) plt.scatter(Y_test, y_test_predict) plt.show() -## Splitting our Data in Training and Test data - - -It is normal in essentially all Machine Learning studies to split the -data in a training set and a test set (sometimes also an additional -validation set). **Scikit-Learn** has an own function for this. There -is no explicit recipe for how much data should be included as training -data and say test data. An accepted rule of thumb is to use -approximately $2/3$ to $4/5$ of the data as training data. We will -postpone a discussion of this splitting to the end of these notes and -our discussion of the so-called **bias-variance** tradeoff. Here we -limit ourselves to repeat the above equation of state fitting example -but now splitting the data into a training set and a test set. - -Let us study some examples. The first code here takes a simple -one-dimensional second-order polynomial and we fit it to a -second-order polynomial. Depending on the strength of the added noise, -the various measures like the $R2$ score or the mean-squared error, -the fit becomes better or worse. - -import os -import numpy as np -import pandas as pd -import matplotlib.pyplot as plt -from sklearn.model_selection import train_test_split - - -def R2(y_data, y_model): - return 1 - np.sum((y_data - y_model) ** 2) / np.sum((y_data - np.mean(y_data)) ** 2) -def MSE(y_data,y_model): - n = np.size(y_model) - return np.sum((y_data-y_model)**2)/n - -x = np.random.rand(100) -y = 2.0+5*x*x+0.1*np.random.randn(100) - - -# The design matrix now as function of a given polynomial -X = np.zeros((len(x),3)) -X[:,0] = 1.0 -X[:,1] = x -X[:,2] = x**2 -# We split the data in test and training data -X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2) -# matrix inversion to find beta -beta = np.linalg.inv(X_train.T @ X_train) @ X_train.T @ y_train -print(beta) -# and then make the prediction -ytilde = X_train @ beta -print("Training R2") -print(R2(y_train,ytilde)) -print("Training MSE") -print(MSE(y_train,ytilde)) -ypredict = X_test @ beta -print("Test R2") -print(R2(y_test,ypredict)) -print("Test MSE") -print(MSE(y_test,ypredict)) - -Alternatively, you could write your own test-train splitting function as shown here. - -# equivalently in numpy -def train_test_split_numpy(inputs, labels, train_size, test_size): - n_inputs = len(inputs) - inputs_shuffled = inputs.copy() - labels_shuffled = labels.copy() - - np.random.shuffle(inputs_shuffled) - np.random.shuffle(labels_shuffled) - - train_end = int(n_inputs*train_size) - X_train, X_test = inputs_shuffled[:train_end], inputs_shuffled[train_end:] - Y_train, Y_test = labels_shuffled[:train_end], labels_shuffled[train_end:] - - return X_train, X_test, Y_train, Y_test - -But since **scikit-learn** has its own function for doing this and since -it interfaces easily with **tensorflow** and other libraries, we -normally recommend using the latter functionality. - - - - ## Reducing the number of degrees of freedom, overarching view Many Machine Learning problems involve thousands or even millions of diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_11_0.png index 8585b7491cb359fdc358093418f34f4a361e7869..65d331dc5e178e89457d6a5492e5d80b639a355f 100644 GIT binary patch literal 6183 zcmd5=cT^K;yH7yKA_BY1iqb8ESSY%HNC_;W!hkek=v9Omx4m zS_GtpGK!!P2}(y=gh(erT9B3-w|sZ+y=T8azjM#boO$2(JoBD+p7MKs&vVDz)Ieyr z#BLM{B?KCtyM#iaUFcP_Pg%CdWdCGgAwG$vQ(0g20?6P^KMQ(+jhOf}x5?52i(p~T}DgJ)2 zqD#4@VC#80C#nL)NFh^Gh_wUXbZ zSwmx}vI0>F6mTO9ZZ-`h!kzLOzV9kPnG}XHd{c|+3o4l{Ei%JddT6{0G1w`?8H{vrG?{U*CCdPIF$oNZSx4$LsrR!sk5kL#zZ_re z1q$T`QyINr6M8F=Injuv_ETDhQna5?wsLY>r$<82nymRJ|3pP|>FNwYRl=`(Atq*=xA_K0sn9pRL)f<{#BW8x+|>Q$G( z(`^0(w35t{6b&6AxQ;_S>E%)E)s8hsIv(j`uIoY6cC4W2&NZ&8u@myavMe%l z-IPAgTu7;3^~n=G%Ml5^*jZ7?<083KD~opSQfU&d6&>i3R;EA1va|Vz`Lt_sAI5k( zXT&2<2XvFaNK3Uz&)Z!s4fBJfL7e`2sH7$27?-W)*ZjE#}^Ta%3dh?=+>)oK=-vGQoh=|l|KXCE|y^FT9^sxLmXi?TIMm7|@Fl>0sf1a=@|e zW@8-hO(HzDe`(PMg+s}fkceRg&1p0}ws$?HR%f@ipW!w~FjOVLBV<(hy~b7D1a%Z8 zS{dcCYo|cVB<&ZgL9Jf_#m=nTKoS0kTQcJ^+;No_2~7GE}{sgMgIRV&R_E3AHMLvVvv&# zxLjPS132>3KD)2J+|ko=h#E$}L|1S<5ilBcvi314%N8=@ALVKA{XpPqL_IF6^%#;x zs((si@|5kO&1B|~cA9W2-#qd(=UV+-YzO^R0@`mzd<>&q8d(4|3pel#ptOVD{S$UX zIido#KEE<+FbBI_gMy+cB&AZ)D^xkZv-*`RY3NxgsoHn6;YKx|QP)8qF}G_c0vtQn zpN`;L90jyP()d8QO(l+8D|Y4>^$vI%HN`^-!x;u%>YP^QamoBW)V3d-|1H$$Uar_h zCaGSDDu*yhANU*}F!s>>=oBQbz=?ax;L0KvEx^205)*7ew|a7423QZj&N-jx!A2zIdKbVg(2ccDiXi@T3sgjIFhu$|-i!4YdKtHGwU z6hNAXdsKfOk%CIe+2j;u?%5ODu*X<|bS1iabfRgK6xgLLF*Rh~DF4wZe#*|U>!5o^ zR;#D*u5%|9fSIIIH1J01dyQM0vLXKd5A7j1uWr1PWCW*e*+E>=rdxa*^!oZaI?Dmt zXYy1A?2?k03Y7hoF_n^m8GfI6y|?Fh3eXXM&FszQQF8QVWV|(8MNi{e$PywP-&()? zvjLB;6O^49M&NFJ@(r4aK?64=e}ckA*2}vUj9^ul3j%S#4PfyFQ_n9d_TtkcU{{9t zrw6UxQO!JMsGXBkoBf;(J*-+}eIT+|je%d4iLD@VyYWenD zGkG1Mq`(;u88HA6)~4q*0p*5Ep8yxbrKf;%o;f`E`mb^143~yihKmhYLuNcX`$OBS zxEsd%{f~qukkPHB4}~|7nDYtFyi!`X>H+X@+Id^4mgx30-uXFG$w@iHfPW;J@iRb4 znICPA5L9^U@Cq+E; zR6y%*RD?wq*KYd_1HU^!{T0zbQd2#jkkaefBkH%p0MI?H)$fUWltWk>n@||9_}@wE z&JXgSmxCpyw%E%6@+~I$rmOoqF9d!K4lb>^bZhWZvZq_6-V@QF~+g)0eIok zqOaK2>-HzM{bm;XEpqKv;F*uRY-`Gv*-Pykmr)`6Da7*r-0{t0aKZopBsr< z6R4@|<}d>HXq5xlxc;WKpUf5NmT9zLuW}o&(oIit!1}Y!UGk2Fl|g)OS@JPbhn@P*VmKZzXtYW6ZE?~X%5)^95>6`gHE4`l%+!{fLAZef&*h@zhX{0p&i+(9s{R$pB~ z`mQ4(%l|l?J5}h^2lsKz?Q^3> zp8f30B>At{Xlfr&sP++Wcsg9+(HdWWDEA+IWcyg(MS~l{RJw<#`b}u!*>mmi#iS!| z$+mv(8pgj12<2+W_6oP(5XjrntoCQElx+-o)WvhWYVuXj$FzDiKHpR~&GlJKP4^!t z%d?oB=BVCW|Cs5KIuvDg`EcaP?3eQ5Ki-p43a!DCvJEX+PoDBdmL~Dy-UHmwdu|q#ncU zv-e8OpVE9X>gHSRxVcBLIX6*;Gtki$nA<}R-`swMSv_z;JZ$LGFNg1Ev!`=q2OI{; zIo+f0@XHHBvAWIq?v|3LtLDVlwo+wqMOixp-(lNlgt^64T;B z(P|lkSKm8boxxG-cSC0g45Q}JN)iE_TT$!~7DqMUtaql!H*p>kQ;#IWAGk?UElwki z{#zC?)*|>GlteH>S+y-)6DQPsc=F1RzwxQEX)ot9C8LSIxAgnd}adotQjbh>AlDq5kBV~9+MDp~-vXR(EPO1azQT4)DwPKCB z$3yn~a10+=X5E2a7$t;EMVE#zZ#mGVM~0?&SpKCUuC^sTMj>vBHOW$HRToCyRbP+3 zOY!#3c;)r1(@5VxBgqb5?)EGuKYknNQ|s((W^**C{q*Nmo4}-9m|sem>Glw8(MPFz zOBUjgKiwaaYw|Ge2(r!C?Kq;cBSBXSUv?DA_|O=>KR`0+Q=is*vhB?C&!CjN_-bqD z$>#*J^1Sut-Q}4%*MyOhXKwGaT{xT|xcvEz{V)|oLy>lYxn&* zHz8JWMx*X2y(yi|93eU^Wm?D(-)U~dlW8VqKN|>z32)EUi7*|S-{#L?uzg1tDg_e4 zRHcH>_|!@=?J;xwSX0iArBgbBxmE2(^CJ|t_bq2xz7mHd(aS_mc8jyTckc)kg1$Je zvFdMp0biW&VH%Typ#`uSe;8K3x>>p6ELCMnzjoq=u=gxmf#iiD^B$mbNoe&OxsKZa zMc0M!!mIz3$W13HSBI~6(iOJR{q)wtbcg7Pu(renQ^_-Ca7$DydMr!Nms>ynrB(?R zc4f_`r`X{ZXZin2)%ByRS>*mqwY;5XdQ&@x3O-o?Bo_hx)@QK7Z-3wOGA=rV@fNF<&J==SQzwyuTp zFIv%!^_h{vi{BkUx}(i%`ZQR7whIU}?S}8q6#8ccMYY%0aQ2_Uf}XY$C8|S@NjlC@ z$aRDAoFLQoLlU0|=tV6gJzADlQ7bNyFTJA{fQQ2OJt8%!O-SH8?o}V9%X8ay{5Qj6 zRy(YqaG0iU9IJs@&z-I14-JJ46l-WVMB?;PcWZcl^;Z#ca8c5E6(fh0Fo8@n;kYM* zVk3U5*Pzvpc4DX9a~qDx2x_0Ea!dkEU=^J8h);Z%ShG;WNvaP0IDM?6psG%lxpu|Z ziK(9pE}eCiTof#JYWFg%5Dt}G9EX%VWTY9+uHkpeBam((p)D0bJ72l>PoR zA@O~K%o(Xiq*1-82OUAvTbeVONV8X2klqltB+y%KlH|TH|DHLfB{BWww2O(GLf}rj zS3h9##xeW3V(F=Jj1su?4gDDMA`@G#pNb1G7QM$%o^cHPR7Kht_3A%t@=r=mo&dOX zj&6ZC@!uAUcROhE6WgXusc>2F%9smXR=n%r8*-BHB{&|*MwOS4=4Q4s=H_f_j^it| pJ-Kb{3NDA?_Okz;Jn!hPl+S+q`nH@Mw`q?8&zqhrJL~%Me*qAD1nU3* literal 6370 zcmd5>cT`j9woihB^l=zbx-%YNR0NesQOZ~ljz|*^z4y=(dI{A6;~Q2sK1N5CSBK^pY3veRt-)JMaDT{&?%Ith2JR&tCcVclK}pzTe*Gmbs~+ zsF1V}0)Y?(jV@e4Ado4%OL7lC&q6p*C&~NL4mGd|wZwXbMz{uhB1~LEasJp)e~jC~ zaL?cn3^q_*mG@CO=o1=>3puT(7Vw9IDmK_#?Jc1?oA)PzI3wE-1VUKi+l54AW**Gqs1#q}(}TSP{a>JD#3!C2)Qb!r(P&eC0VayiP$XUMw+!kMlzcz zGKtB2O=4SJfPGS`*p?W_hv}UlkvN7bChkX7e;B_0j94vld-F$T5NNs}vl7E3GA71$ z%Nt9$>5IyW*T?pZ^SA> zoEIm=w2j^F+7&28V{`_Nbm|J1T6L8X-7y+B?h4#tiRSU;($DM)vxe&u;&jL3(^3uGn8Z1}GAGB~bN3EV#V${BQGNQ`sV zut4jeGo(Mqibv!8oVle9u1|)H-A|~zB0@B01REJwuK77UQ#U1fIn2HbLhN z<=H^`$od4YL)#P7>i0Wc3CF)Xh-9@);?x`?s~ z8ednu#M(0p^->}}Q5=PRs;$8BZuq#$$42cOq#k;#O@JpK9a35rV^)H3RrHm7Fqi)6 z=4zAzrxvh>Wer2(Cf}VX`Z3kR#)noy+@>h>Wb2dVY_weK^QJ5}OQaghxQ5Kblu@*D z3lB@{WBG%KvGhs0;rpE~4)Dwu2Ut=+jV{8lFHCTc@<0BQk63x7118AS2nt&@xHIZ+Z)J$LMQ@!z{@WSXJ zF!S~;=asR#nW0Y<^@Lc;G8k@6+b@vW*hEy#4Aq4T7ERo{#?dZVXw(5(MOgP{Cp-X3 zzwagI4=lp>Z3;W@Y80veeYG^>JTPd=w%To|5c~Dn-T18uW%Dwi)KPN; z*PH{LnniuM@AYXRO##KQhV(CeDWUA6^a{v8G*cI4%GOrQ{9uJQ=@x+53kC*Ymw)Ni z5(yLN5FL`t?v`0`hnDAw8y4lOO~@kCd{yv@Veg(4R$^LO_5+VYI6YEb(^+_Db0u3i z1|t*(!lyIWT#JddYE%Ps#Eye_eH}6Ia`0hN5iy*Pt4FP&#G*AV@QVd37znHOEoZJ| z-vm-_ISO>Xw0k%&lLe}Y0 zkAW5L(yl?exL`O9_TSYh5U!?U3q@Gdf}qM$Vil5(oVKFn9_~{DXI;i!EudB8ek2{` zjM6iQG?i-5)S9&1CSZmW@UZnVaaNj{ndn-s@L}aPrVjvT16#~+#;{l3zhR>O0MMy) zl2LFtoCVLB1TWs3<_5rB%Ip3xH*N9(rcV)^^%{4*@1cv+#);vCaIfzjCliUeHjo*L zfvOMa;eom~t2MU;LtzkWYlsWTGX72$)*Onb)A49?%VJ}=??ckz_Y;xKmByV|ilu;z z;`%9OI^`mBrdjEM#~G5G3#-4h-KiMBkRNHkc7zDlH1`BLqFm;Kmp9a5P20*iW?yLk z7?68ixK}a)p}ex3*Da&*jz+0aw!H^nbpBdzX`9jUf@|-7_y#)kVu&PopfwyJ=irp>;k3IP}Kndhh#_3Mj73z^?4HNn44MOEB89b+95*i<5>P zOG3qOT!rr($XQh(&Yq{_Kw$=uBXZ=_3nDL`r%Sm8uro^42-U%8S&~}YdkE|qLU+0r zPb4UO-C>A3Xx#se=T;PAbCtFa4E((&8ZW^KWm+pi4*^X0ZG29PF(z*1t0;Q^Gzi~8 z@Iuv06zG*rik|#J;Kgn*;q;(dm@Kd+%6!u9du!hF+jsGIrJI{$cZqqM00{Wintj_hY7pc)&AH55kP?_F#anaL!sSqMT&flPs^ageEeS0bQQVPmwwR81B=rW5p*!xqXcUE@Je1z4d^_GV^F5NB2509? z76SM*9p>+iK>6h_EkK^}mlpt|z#R5^`?uzwsL{+|rY!RZ<~5*51idM8oTn@Vw)S!| znloq097zE(&+X}3qm?X@k}_ymiyLMUAI*FO(#PwtLRU&zB1FBOs9pFA+9xn>u+?=_B z4X(*X!IC&ZTp+aV!&|#Hto|b}Re!BMWj=O+;BQ#D{t?wW5siS;p#y~`9XPNDzO2K; zqaO1orXACmB0&6BiAvhl^tn>X35q~kj0SaKZ<=4k$D5=O<@X(Dd$vYvDx_3FJNRm8 z=b6;5t}c46F5aqE1hj+T5p!s1;*O@JGFcI306OBB9qZP331F6{NAEtt&N!7$aexu-1IC)OXO@1iMm|e43WPTUGT>d6Q zbZ~fQE-%$D{>w?2!8`2ct!RP^#35T~2|Cr}*`ATEYRDr5Uarn0NNjGHeHKmork4Jj zP{9i=r31&(ZY3K)qXG`Q9E4*8*M5OUA5QiF(jbvXAp7^^s(M-jjw}9vf zx`+rqc5H%$5i{eJSU2%BPovmL{c*DVU(^b1NI1o@&ShA`rn+ps&X0b{8URZeSL3?e ze8cCj%@2R1Bo=b}?X%(%gw-5sY5`2Ty**x%;1GQLXGvMDZV>`-Y;Z^@WjIgTp?!~E25WfOVL9#(SW`FIju7n&rC4ZosFjKHA}dAA%Qf1Dc;3<?_muwQ`+vZ__AV*VvD#0SS4LmEX~6?krGwq#nOS}-aT$0W zQ2NLJRECC~xdwFX1(WgC>rWbM@vPDmw&EJ(0;-R{6-;p*dm#M@r_rj0m+~Lz&UP6i z<|@n)6s}jX;0jx$+2Hy-Ai>fI8-S``o*uZRLL7m%H0I!gff! zui)^=zOm10mG)O`MC=w)2R4?Q&%11Icx48-A|XSCZHlyVrGXQEXyBEQ`% z(?6b3inYz@InR+W|C$(?74fp7<`IEa>;ZRM z)h)By=hIJGo;493)j zE$>qy6KTO-3`x{42Ne|3fG~~4X7j%bbRScx!~`nQpvm!8T$HJF3WG|Q8pZ!*|fpia(xPX`K!Mt%?G^-^Q zEXxc}?d|!v{o)k0EiI;SRX(c!QlGXc$jF(rZ}1yP(#-?7HPz4c2$P3PB^=+S$MnuV z>@GcL-LiDCWG~)16lMFB*Wku{PZoRVJX34!lcOcICc;T;*fD$_I(n3>5{^0( zeb)>bXggD@OxB6p8!5!OxwlJScx}&jObn%HxKNw*FD>!ojws5$hPL=!*^dCr=RtOM z1Oq6xo+$P;y8KB^q35yLmrjT+4UQKxxmP9dw_26tcs>UwXl!>e5o=P@ikW$-BD8jk zF_DaN&U1^kg$Z_(1Y2xxoW?a=$LuEm)nJY3}5X&?W)%*9WyVM~Z1 z*3Ul+&X865>+iIGv}5KeuK&bYrIeKTcFadbI7pVG80nMy(8_hr^%wPR;K^9D!uNMr zIZvYaa6kM4HC&&{?x^sI%lV4nl-n28TqN)sC#jAN*Pp>25nf;Q*Rl7XCOf5uu}hH` z*^4r1TZHs7&`wo7c6arwclAgzntCWfna^)wHcGCh{+hkrMR4cZRK`*7tF%?CT6s)O z9Bet4H1IeBu)4HvQHNlR|HtR4HhF;32_jqm&S5jnVqk z8%gP{4bjJYF z3^nAphWGr=`RlvB^If{HxtQ5|uV<}wKXtF?34NiiOhQOUh=YSe^6aUiCJqiR8eAr~ zZh&vhGZgT^hlGpLYZoo3rHi|XvjvWtiHoBh)Wy!m^uC*gGt36+AjAv)ao@LgadCu6 z^6}aK_YJ&IXDhz$`05n!5(3AkdN3RuA`0wgM%YW^i1)eR~{*w)1D~?+Nqb9 zqm8bRZC~V#r?;-Rv1ES1xicQCbVHFuT<{yQ_y;t;>#awG&57i+zxW<(MTT?ah3AEY zhlqa*=N95Nwjg0m3M@*UsfUz$Z)471{)ydEXLo`BZJOR*StQqTt^a$trjY3FKOl3& z&_`o-o18qlrGmX4mxYC;!io|FK5%pSsmaL+!~<(XLY}s8B9s&ruc@2xvasB7B!q$6 z-@(}GalJE3`69Nr$L;s}`$hKmzaMIb<;We3cI213MVvque(O?*8J_gQdFbe4q|Us) zwND-OGX`UJJk1w=8&l*qO^Ez04b?JMn^IlnBB=}u3AvuqW}v9ZI`_YkAW=uB^g@U| zWk<*F@>Kb-7bBQRoPIjxde%S7y|@r2*W!OZ3VmdJx*N5P$_$(7tFTx>kC{b9dKpWm zE+{a0$t8r@Q-{mqw%?C3rdC#7b&z_;&F(;$VZF6%tIDE}# zcbHDcEQ*XTy_TidTs|TF4w-5WCy>rZ({i|Ta+m0f#-Yx~r8w+bAJ;Ks@W5?)9Vi_u z6~Uk;(+Xemji;S$4K_MF`F7FP;sLluXDVbOyZpf$qwk=^a0?T?5%WtL7D1AllD>Ab z%cN5O^V#G2=2QOC<_pPPj|bESj->4AK6Dzg81al3AzURjU%a5BxXFu0s&$vE_{T@B zQ(uejfn6>_Ue|j%rl*{~({oMnZcAp{pU6$~N{NglJ)DI4&+JCJJ0oR|9M`h_Z$4I0 z;c}lnr*NoSyIkC=-(EVLZ2lR>CyU&y_D|GOmxw6!SyoAL9q|^Q56E_LcSn~;`36XB zpOHhoCl`7#OrA?7pMJ7Hy@uaUxz9E-c&%pkjwU7Rul}7NY;hg+b*P8ikCml)NDt)7 zopXxXPqaiUGxlri8aCdPsf=hdfSZLgCv$#LU)HP5A!}Z(EQf_`r3o{0pUHEUg*~P! zE8h_~emq4mPYBq;H&55bs+&BSAeSK|Lc`-Ze;1Yi0PjCcj+8+{Z%a>Zm{!QRQSRaMSD` z1z!esJ2UR*xO5e*6(Nf9uSF>h(iIUIt8^9k@j*$qN+7F9$~=uHG^%%3H~QYWYr$e5-mrODpYg)F{>VEG3&5DVIxn2M`g&_S zyK8f~a*e@E=c~LI+n48(=oO~xk0B5)8JX$WqoaKnG)k=5B6t(x;9xnJ*`6rlO^O5b z{*m12xW?g5F7L|(B!8mI`>g>=-)Wl@@5=ANdo*wP4e#F}HPQPemC*J2v)GM0CUkVB z0d34PUazIXNH_1N_H0+0*qafjK^^5qKdxQ#9~XYyK@iBCX;Jqo%EMcEJl-jZo4>aRQD$JU z6Qrd^qMc;YNm6afxK+Cs{1rr_6BGn#O}2{OmyizaCdyAI0ej`gt`f_pxo*=b#Oq!aiZw=j^3uR|q z`K&-Xea_@@JrElCo5+P) z+1d}vk}H~s`hBvc)!84LCv!1zN*w0%iVSa(DLN=jl9FPC|0w6@N8C{fc*go5`y~0mri-RMZiblWK){Y$)SlrRZ1H|+)5Oom zk2HiAei=|d)Y{2|sJ|#PQShf~%Vyqn6c50s{KTltG}YU7P9Lf)|C@~^s|o3{R%)ot z?X^^?6cj|bPi8(o6q{4Y72)LTJkGM_f(t=d4Z@3Gj{df#Fj?5;2Z&Np`SMy?RT%Ya z;>-Q$@Y@=Sw9QufkJosHXs@5_7&6U@*iB0YEYt1~16yT5+IB*MH#x8wP-aiW|9eF7 z*@$9DN1Mx!hoS5&0gI&e{R9ioNEXAok10{Oi(ayYBS%8j#X%E_*O6$?3rxTCQS-Xf zCq{|;%A74~Q_M5B7Xz|Zb~;p&pILn3o@$PkAAF#@Bfgwa`}^U8?8KGmA!UjXt^RhVze1M0Y-@i+PqPhSNl_3lhL?YV3b8A>$c+N-pI5@9%Whe$TCfha9%nSQU4 zw$57!r|8YNwuC>hdE!)Y6qy;!PlOpe2`3S@rJR92W0sePl9|z)><1y;`?>zLc&Ne4 z>gc-=@6aFxh8GkTeRY0AYb;bjK_@gig7TtvP7@2~iIX}XmDVKplP&*558Ul4u5(@g z(?k&!cfmI@aeL)^){4WCEAnXkw_#FGI#*2t93Evl@GzcdXgH?zki-KBrz~KF6R&DJ zY41~LJYKhH&P*?Lr!9bc;>{OfkxqgxhQoAyWqqwLFS=PaI5w1W9t3*g(3p8N3jG|p z;N}|He`{ed^(v{E*f@(McntcF(6-mg`+(Hjoe7&kW?9jn3*ej_=937>q}#oXZ?z<` zUG2kqK4xRTQ!`%gcYh};niUR-ci$ysZ1mL&Utgs5-SZ@q(|@y4E2RAmX3vDfhtpZn zQvN(<_Y1jdx9LEUB8R^5^h}RPFzuvx&Grw+| zb=zs|a?UbRx~Q?jk#Lo1an1C&*{r+b3I4_WO{iS^ERBM(f1GMT%dZz{RqHpqEeR)G-6JK!7(4Ps`aOIQt(f4RD|nvtZ5u+}Zez!Z9arTojV8^eNU6@;|^YuFh*w&sndQaFVc_7(=5({ zJXG=reDv3Z?8yj~8F65PiY>o}r1o2D{w`E8r=S8pA<%IHn_Y%Q1zGKmglkm4szuaF z4GE8{nCTb4GO|Om7Of$id))uzo+`D zI^x+~y*MRla<;TFs{Qsy$-BhMJ>LnxYk6vZ%+}w!aTB&g#5mrdEE>-|nVJDLl;bbY_#LXHKpQqH!e=K;0{|H}p3nUUP}oy&O_l8DYM0b8r99 zf|E+7C{tP2J!zmgf3HwQe`Z1y;7&7-y-ir;p*6HHqeS+Jf*FI=4L*Oa#r3VNg)b_+ z1=4cc4KS{#(xwxp<28*-vk80&4n%JFsL@PYjd;Cb>d~t+$ee_5Z;kkcc!9nTk=(~p>m9ZyS1JOOw8CVe)4OI;A z5R~^-2+zq87pZzVF=?Au8kL_^S|O94a+I?hG!6!ji5biM`yxRD*Wai-min@Wc2d)RxCDp>|lcW9KA1eWNNqNhbrcNPFry)YJ&*=*)gpv#nVdG8fgd3zDp9$@GjA zZCEa=q<*jDHBty~Cxy;B!lOOT<{gNCmfD7h78RuQAG8u*$&Sj^7gz{@qn(Pco_Dp7G+=sH;=T z+FgI`OPj6cOTQ*`>Nbw!LMVHM4`$C+%%t~qb549baJe(B40z;36z<|4M(yTeD(1bF zJYkV2)_eUun@uP$*=b+elT(=8mg?YSOg^IKXphlo@WG$-hahG;f6~X2bW2N%{-F^? zMy*7vU*)Q zJ8eX@s3eH^R?VW@y3ao|$c92#h}A_|WZHWMU*ul(KBBb#EV<1f(U$- z6`^{I(?<%5hopTKAxaFU#UeieXmkL<}urxY9G^kKN?QnNZ1ClT%dAb&?3DH~KLM zTeUkwb>tQC+O7%be{_wP>()yaBc-;!!QpWlzJIkCJeQ&k+{P{b*vf0m?6$EndZJth zq_@->4t}|X3!EX%3OzS=MJU7HN79w?T%Zu`W5!1@Vb|3!AIDXd9u5CV&4EBZHO+9> zs=ik(c(>cd1h2NE=eolJ;5qf>dk*h&Ev=Mul*o&ikQd1z;S=~T%~=ZGB+6YJFLLWu z1fWnu(597ie_6Sc!6-K=MDB?Q`mX<4&S!Ve`6~58=(92>L#EQEsb?i~^Hr?kf8RZ^ zn`B{R;U9I$={v~1!o|14(UeF!l*@0NNw)%+VLFe#*8U-B>4$jcWZutecR8sH9A$z8 z@q_9#vj!~WMa8MqrNb!*$kl>uAkV%$ReajP@b?ROTV(%6q?hX8!;P`m){~2=1b(?^E(>)tr`Jk3pbEsb4>2u8FVt(Nl2kTBjH#k*twy@Xe+7 zts>{KGAqIExv+)#p;e=9kX2ikgpr2@ggjxOvt?n;BrOk75Wd60NboW_|ID;k z8rtn=0Tk+UvARa{XJ{ll11YS7T-gz`9IsnRpdHg+|Gnbp(4t-R%7wd(!{7r=>YwHr z#R9TDJe1KrpTVqIS$-f3?zv7KktFT~iO4k44k_TDU^9Iyr!d*H?Rd8WuHyuGMy(zG z8&jWuQn8anYL6EW1lX0mcmC-xNp;iZ_g>DpE5g1wLzecXh+O}xb5u_sU$f7|jpL`& zAVf0bS}FHjydQBOFqK1N_vqKmcYf+<8*aa@(89OlT`T%jkkk8E2#;#O^wi@s{ZWc=(%=e z)kVp!?i)KY(K4D=<3ji~c&Tz0TiDVjf8BhcH~n1E-0HdafRv&cX@7Bcy-`ppo_Ct@ zo7+|#ax5Dv-pG2TQfE@6*msyxg@TeZepFGa#d0UJP7wF4;Jm`9!5N))kPjzJYGyHG za5NLP9~dt8HayrU`y)nN#~4*Ib`6^Zav_w=qXOkcs{muOZk81pU*H-4`>n*17XHw3 z*SP9?`9bz%cNKp@Pt9`)&Gi!u8b;q*9{P~w2b|5?=Gm`_l%;o`B#Lc>$N8!zO;IjK zR;>pFcMa_(b6!wqu#uZM=#vl~L4IE>u~Z=_zPX4WP>@$ru%IoA{Y>hAo*F3eh5wLiyv!}r3O2lUXn_Ep> zZqs*L0Z$_61{x%T&TH3dUEZhhC=}teuq=#$MGj56)z3rlCMo8JFnb;(B>ER?eF0cE zd@{Mpw1ZRyyS0ADL#LZ%`BD+-?$2s&bP+f$>7#CISd-r(PpsqaW1%9W2n)8^%v=?((70klPRaVA@4y5Qi7jbhr>r255{=GNFf{xM;vo_eC&>xh zJZLzDMu2wzHd`RZ=$8RgR$-~I_cYpfq-ec;pq6#g6k$9j*;7nGK7N|=xql;$&a(P< zOY7uCDpwt2s;}P3O_<&0b6`2yk{^9hqjn_XHEMkj?@r5rXbpP5SI;faOvG*;Qxx$+ zmno}>SskYJVPu4rghYTrt!0uEjCi}E`kW0DAMugcEys~N4F6{J?TGwG)X|9#HUbhR z-?Re-Nto4?a0t(k(|*gJnp-c{7dJqe-^!5q%7(|gdUgJ;!JZ6nqfFw<72-fU!2BrM ze@&X;;Nc+vbF;Qlzdc&z`PigMcZq0nYsKd;Mr}+l-a{JkJ~;0YHTnA?5lrUF?Z%|{3s!6&-lGstZ(iy=PEjpC2c)Mo%(4mk&8X8X@7zJiVH)sHH&Hm zo4kTqN?aT(2#+KuAeO&3U9f-@Gj$4T9iSBDw`#s|r>A>Kz#O2X4}ecYjS&0-jvqFlaLf6-Ft)a1^G;b)Ne7ikWL0V zy630;$hEv|IIrYpfoC9wQLxjl5E0SZTPOVciRZe&grbt7xyE>$l7fAB&;$*G@ChU^ zQ+fMwTuBM>v<$76_t1B(z@Y<7gE9!dKGSy`k(BDrSj-35XYTU59tgZiq5xj@BuJc@ zJpuBL199|X@L<>X7gG<%p?anIc+>eg$+dg*GS6TPN044ViR!W+iq3;_=)(^G^zm*t1NyHn#Lnum&Omg{W6)>5>+B>a69TenA zy{9B$Ax|A8n3M5g^pBA7+<)gbgWoj zk1CKH@_7S7h?e=TR9Z!2DQCUsGX&g+-8)61BK` z53#$Xl3ZOsGmnwgKdq!OdpD$6lGwz6!&7NnIHOKZPW~bxYP~krP`c8T3nEGuDS3hT zRD5<^&8253o5t9yg{Y^Q%*?hwUA_(uCJR6gNji$g<*2#HTl}~-WPh{qqbE)s{QFKs ztD-^&OOoGx7M2(V6WPc;4#}+qe)sbs>IzTlcYqUbl4C##S`)xyXxt*_q2J!i+~V`xBLXmH`QB6tU3AV&PmIf+bnq5V<&hG+cD zmT<|6StheB1@=%iQ99{^Oyqp;6uo{TnZ~<5D%$STnzYsqK!FxCPeg%b<<~|n@^8!~ zd17jNygdNDPV(M9UyzzS`dKzVlXA8olT>me14^fr%a`X(Qfo;sIL2g)!O8?dO=_dE z7j*{P(ZoMr*xaapgdIBup2LfJTr=x_*QMo5S%BbZBPy`@pDkrDU#W}scT0IL}(v;H&(SZifY;mG;B zbDthOLfen_^&KU}H|O*~!S0V^7kOJ2T^@^h>sgAP_|DcG8q&R_nGn^^v?2?0_t?|- zyem$@>uP9k5DSi;Re8ZE;mzkIHd{MOVl4r79MGbnSJcIRcwpb30FbnoP9=$E3cC9i zWon3<|Kqv6*G^`K`g~5umJtY!u_ss43Y_ffMf%@{w^nc)lv1KHKKBbKyYNH(C}8>! zw%TlPQvsq&L>hO1MoW)m0If3G$20d0kM8ag5Z8hO(IsGz*bdhp0`ZE#dv_i=2Bw59)5}Mysd6=e$xvqUGj;7y<&_l+YUC2)1elS=S*Sg)S5#E& z%;IMXrqxS~@7qVbV0@S7_LPD=n)3<3W%MW_INqf{MyP>#wY@J&UM|uyg+b9~t^1 z^2&1yoSHV&Z+h?N^w`sOxcYu-GO^|t1F!BMyZCZX%XpLSs(9hQP6woT-WEz)P7MvU zeR-&24M-rWFeh+q4p;-eSCs@%>!kx{jzDj>XM;dO#UgY zt^Y_P7s>pO@xrKh3M?sRK5s|)8F71Hk9OYF+0U;gVeu`Xea}+vZ(df-MnzxEYGxP* zRTzW+*>1y8<90Fj8Mz!mCL>J0xajMv1@|gwieJH}HfRCYi%qDMiCIO8PYEUJYw)ND z0t<$SUW5Q5A$fRje8^n9*&zB~Bs}R6?i}lFNZ<;r)#dHhP6XcW6+oETh6qedP89LS z0&GteM`9<9equd786M4K8=Qw*|MCL(czsnj3ok%44w2gG<`jcLjI{5E@C!%a3 z+qvV5g4+*%4Zrq+W4mccvlMIPa)33WX-Cg;6aCFx&R$m#JrH@xGXHByL{(AI{3a#J zj2LNI%^g-lagDo-$B++b2d4^nm7|4&J{5?2>EHs)DnjER7V9lIp~};mhF6V~)EY#e zByr52&=8?SRV^!l%x+IjWS>0jJ0C(Ef@6(sDfF*r_A$T~3|J@lhWGlRjBIXr3+K$b zgE4Kj5~5xv_HV^9DrP3NVs|FF%_n>wg39rMdI}d@$t;SUXG2;cr!Hzf^&_CzXj9AYGBpoj}7UA-&zvjqGe!MYQ4#{baA@9 zGNKP%#-{NN=z+zh7!5WCQr*I0ZIdUYl~T+sb(UD^rvlho{`~n zYA2IR;q%)KOL1%?IIZ*57W085ZQCXQNBF@c-m;RK5Z}mWSINj@9v>WJOB~OAJkqXb z>vy!+f?P`h#XcPYy#rS1zHz79&FR66ULiR-Y%}PzBgb|UF@CFo6yNEl0yN26;?8@d z|B+reAMogUEN&<)OvDU(Gibd5Pnc>X7)qBJL1S@B*&B64xU-jN`D0|wRZu80A#>PgU zb^Y0epVY>zvCqrRd%pP2blVDczQxq>sDUTUehK38g`$MiqGMsd7@}>0G$RZ zf^-jU7TG>RgD$&<>OiP-=N1JadfWf<)Mwj!c5$zduM4gMmDp*))2p=Xm$jcV=kK(m zx75Yja9sWjevRB=M3M**PTwQ7TQEcS2_71}?5go5?#@E!{89|>Y0-;i7^fX^S!MJD^AXK8-$+6{M zs=i@m$E(TOd)-vU7_2jAz4&%yvd`aq*kYB;oe< zjW~Q}ELiX0fKiRl7O0C1YQLOr)khKWHGIaPS72MUE6yitxfEeP3;uK#2*&m@w3%F? z`=HTe&2y*qW~C3tnbCLuC2#H09Z(xK3u_}_-vP_7o8_y$a+-aL;Awf800L?pRSyQw z&50^$G(zZXO~ZfTmjYRKNr{6o95O__ZRQZjy_vwq!t%as*yIqv&F`Sl*lE`HC|W&N z;%Ew13uMZjCO(XoE*0)zBwX_XD|NjcO#a@s$4%JE!15Ffxty03dF}0Vohzt@DcMU} zgF}+AE1>2F{QkEwn}ljO{0njg)1%?nrs202v)SaJM-?T#*DVW~GJk;;@z;hhg2{J( z(6-8~mj>uNsL8u}kk7F0D}{!1bhb*o^Vx0}MO0VepOR;b^~(@cq-6zokbmRHSpz_T z#9sFUP%?2UvGzTtW`9Vf7Cj)P*XZS1HKFU;9$OpdeD~^w7BeFwou@NLvNK0bNAq>O zppuLD@_hCZTdt&3{<`SZRv0xO+nvx?h5e^r#iS98I#}uWRXy4rx}08Dm=mbPahqWF zmhHyM-ghy2hu3d6C2JSz>_$iVm>a_@ZGCjHzT|peK)X*-Eo^14mCWxmncvRG`m-5- zOuTa~somw`pKYZn8uPbS{H#9e= zPdd~`MEUJ!{W*izo7V`A7K1eR4rRSC;sWfZSFrKv0(J9o(DvooiX2j8)Sw;JpX|Oo zx%C1D>o=0nz{VO*-SBezu$-Kwyv2%$6AMtWSpVc)#OZHgXflCsm5Ry5JefbuJE`r4 zmIX`Ty2Z!uyf9VRO zkt?5$o0wyG!Xh~zju($_xVpMdxlP(L z_#MwTobC1f9%&Ie}L?FkDOqbm)EjWd`r6~92(`n!fNTe>$ z&=TNsU(w&zYJz?fOrB0!8VLdB0^tyd$(23elm);Z8XPRIo)pTT$_&O0w10_c|*tj&SZ6HBL zU(>Z<$AF2Lrmt<}x;U_0A>fukBI;|ue*J2(ubzIQt4nQfZ;y>(Pc!YcS3E5hE&emw zbe-h|Fs^X`*onYyReBxR#_<^sG#<^=S5NA=rs<~P1P2xigDreFRxDMPl$@L>=P!$M zlhNaTe}BJs)ytTlM`uG1@$|2Bp^H}=)lM-B3J#gh=59*t|5DHcb~Qls4Z$_!q}%4K zybW+8t!2afmq52#fb%l&^c#A{j=MA9-c;$lu|KRaE5?(29VDmINx-P#R)9N4L5b{3 zlM8Sj6uI0NubY;u$!Tlr)}ZJ0>o~2IhM(%n^S>`2hM>%#5~pA;qJJ?%*(yhVVQhol zy#}X6pfvc!hHynEZ;GJk6jxumI>lt0L0BjF+HnI=&7yNT`g8X6NP|)AzKXkf^-IU} zlzHr$l1i@LMPCmskUJOV(kaS2OB91L zV*O}2pc#p%UT{zr{23_VAgN5&Owe#M192Ifb*_n#u5MR(7h@l|^d)pSC|_6E-o@rG z;Gp%dEcC62_Lfk!-IZgZsc%Suu--8l%v9oj^TXQshm~e+$<24!iCCX~|5Gu!boHx5 zNS83!?mK~)TkX|3df3>T;5PUpmY)n;3(YD{iFyEtygxkP5r*f*8ivWp8Au=>X^LNQttT z3>vSp4hNPptgXRr@-TO9bp$KM?QEUI70hvtFoJO?QvfR^AzE-&2y7@Q|o^W2;#(I{>EhN=m!J0JCGjW?fJP3rhtna`eY5H`Qe8o7q4nT|MUf9M) zG>E8OVGO*-&f^+GTs8#je>n)z6`Be0BbH==zyE8RV9d!!Fl?#zQ$RNM1#!W+g>luv zM3nNiQQpg^h<{G8#^oFM?fFq6#NG|a&KT%9-mPTDp!N^$u{N*=0+lK|8(S=1h+Y;N792127(iXT@Hg!zC^^eAey z#IO$(!*l%bAl`}K-K6@dBIUk$!~Y>hecP|sP zdYz0Z=73D!C{e_GbY{Jyh(8a38gM`7eNxmPd8Iub;HT1lbpB*Zta66NcW zZ45i-dlMT};#by!CQ!mm=iTrXn8Uw>MMEkTUSFBU7LHn)M#!77m+`<(mVuc7!yeYj zaV0%i4lY=8gLODI8^nrlb1lW#NQ5mD8~D9@>HM;e#~uVaS1*nD5(WUAfz2d8DEIsf zIY;miDOcx!&>oC{Gm6Ifc(0<2yBKSAS~K}?8jRk&|GfbF+NW#ylzk>!V_1fo6-)`( zv3u;oJv_iij=K`VT3l7M#;eQ#$8|lrEycclgLyg*>oLCXUz0ACMwAdZf>n9^oUer^ zxK_s>t7)2Evlb;v7eO-O65tbEH? zqjxJ1QLMhl4-pv(fl0pA@_IReWVoMsWrH3Q!k(zU)4H;x-}g<;ElRd-1EclQE5ih9 zTSNmTCflTou!{rEM)CYGNwrt|6jG38gypZS(mdI6Q{Cv~&gRJq&o`?VPjAf9w6JXC zVFRZ*rvgmBq?A-GpsbEDNPl(U62>bm7`!*i8C-t-^T)1}LkW2kFAhKBvxvKUuJZ}s2)>nhTKwHPVZx3RY`mWDzg zS(TNQ*jom9bW!kZthkR0{_@ugSk{|)-}wf5XS|0PaHMBJX~ONOAZtkJ&p}tT*zbuq zQsZj3i2p~WE#-~1MG%vdlh`U}E@ZXC^mBdvNr&{(Tz2i+Vmo(WPlj}u+G!LtH0VHt z(Iiw{T;<|QqUI)a?Vta#F`x$JD)21+%YO>gz6&JAN>C`@xLsT@lV1YJD$0SVkpek_ zrr%y-7df*)7d2`CBcV3-5%_5uO~hOC?m?7v1@@-h;E(_M_5?u0;jE;C zM82ifn!$So;*V?3+Cl8_|L&Rk-$~-|Bp8E7Cf$-~mDXdEo7xE|#P%MCfv>RLpVY7s zJu7TJ;0&%kP;pU+?0lN7nc{mm4pJ}NccAe93v56lcvwR&6=X-ndv$lYbap^e7mr0M zXg{i~DUs4sOM7K@x;~tKA)K4;=d~Ztj$QJ^!SazL{0Y&XhZxRVAA^0tYy zonX0?jyR3T@^g!D0YDJJ*ETj74(d`57FBlu4VFOWGl$2Pj!c%%c+1;#Y>IBdBU`Pk z!dpBT1%#CJqO86UT+n~7hh4A1Lx6;Yg!K-=fcPRYDEi9iaTxhp4&rd`z6DTz2=zpI z&$Lgw&lbzlFS5&F2*jow@jx_>(*P*mjpoY>M6wuJmyLT&l{{I^ss#{u%6dmwM8v5TpW5+svl@CNc`z<2b)vdfDGI_0)b2s4e<})P3Cfa~ zPJRV@Y?OEXHnu3#DPCx2y5I+4kX7^f;pEo(iCu(fMKdXA7^6KJPbxQ=nVC^{cXwaL zT1l>U|7%FuQ*I5JgICM70zncmR36gJR5?ArysN&h?l-grQrg%F^+PB3!pw*x7gsNn zTVP0;E>Pds2Da_lfYi3MFMOtD>>?=?%z_@(bRgm?3AmW|teh`(iVW~#Tp6YT3DR(LDf1v9hxP5BiGN=Sv zUIhMDriMJp0?knUK7em!Pe)dQ%0(>YiMx}{=Nr0S+MMPh*>4hjuM@mex~DkR2V~1j zfaY2O=Mx>ShM;f5WkB}S?j074j$lbXBUfrMsAR6LI3E2R0K|>FX3GhyKEXvp#jF0$ zgz_sRHXQiB?ZJ$;0mERC2?KQ8b`^}2nAowKIe6zina|tbyI4wO%4Jy1+vftbCayC8 z7Axz5!Ks6s!MU6H?5}R~#3DKjab3s_Up#tk6}07KR9DA?GZWCp8LDU+g{(FGN%%%z zAA?T6RMpkhwHVC@rp$3riF$L#7c^$-;vOG&$u$k5jXjrR4Zsz0TNzFTYl%^S**J)bt3aa@m($Bt~Y4F6F+$)~AYSd^E zpl4#s6$JiD7G{ydDzQUTIM(}KeL}BwhPQu8DP(KN{u46PZ8;I{-ssT zFE||kXzRY@rotTgU%F=wt{O>Ice{Yr+r%3qrN&4@_vxRYf)%&rpPTnOfD(vqLArtt z=t1U>pPS-L2sk-ZCp{EPQ-j5#;(lE=`p9b371O~k`;UeOncQ(J*8`L<~k46=ppfS=o zvzHe~?{^FRaAuEc0rw3Wu1MUcU8$?8s=TKhq7$E(@ZNF6YB1nVpn0Io6fEn_=c8%% z-e;Tw@rG{W>cQiHHQLrmD+OHcYwpelPc)7=ID-I1pUSISTc{ty>XTdN58AylDdrq% z=5Jt+47Qq19g>og+)mz*)26o>u(pH?V3mPeA0A#nf( z@?T36%;0=0&H~pFxBwV1S!R=m^g;WIJ|F>H!k)%qgFze6#oscR&C_FE;oOj&9cV7t1J)aEn6Ui`z%D$g9LrY$bI^XZ3;d@guWOwj0`8i9-Z7yP zDeL`v6kC`H>PbnJsTIwIK~FVXJvj&oDo*zyrT>=>>i=srP;<%jfY<(@s^3cw^orv= NQ&Lwfk$?N){{YXv@uL6$ literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_13_1.png index b145ce82a3c0168dd9cd94435a2e21c69993403e..d70ab26eed2c169cb7994caa15171b77cf728c3f 100644 GIT binary patch literal 9562 zcmZ{KcUV(R({~aebR;66pnz17CcPIGkbr=S^eSrTA@p7Xw~7>jAYF<#0hA7**Qit} z(EyIu^bro-p5lXWlOXhR=Kv?w-ExE{=TuFTH(SJmIpEl#e){v#&40 zM@34?HaYja`0JU+kSDNW#PePNp3f4iKR z|Co9YYW;%{%fVjqsrRufmQit=PZ$y-{MyS{K`@Xw2 zfpOAQ?K|I?J0!Q<;=dz{A=Md=ia5YvquzXXWV5|ls9`XZAtZLd{kt-2My8=MEl3t4 zO4M=?5FlBi;AJOj!btS0V2%aWi}yCE1yyfw@(JCMx|jJl;#~q~&~@a#0UkUbq1_o_ z7A9#am_uI%-k|cQ8l!U3?mR42t1N*AD{j4*Xa9MSXOBz23lY?d~xs6xT_KVEeh^(92AZ2NZT`k20)v& z@aGbh9eVggp&3!qg?_MOzuRWgUYst&VlY(7GiY5NRgttO4V{XVZOJiTB>CS!{DXd( z@LbITuZ&VKXg0;u3Q4%O)_+P?I4(!q>j$Txox#_kb7{4@cy_GO{A1L+G~dtYDWREL zq>;2;^}=mcch69Qy<$DDFsU?Q?!q6Mt{APqIH_K_eHM<=kR;W2kM#(xj|C43<=k{J zSUe2!=$^v`VO_9OLhDI;YEUQrEYG!_neMy*1-l5{hE>-TfmP^t_oqfQM&J#a3V!odgM?vMXzVY)XFsu zp8h!=%2P!X$b{hAGMzWa7Dsk~?C8|!k8Y48-kQrFG*xe(yCDzd>TfXG+$K%NH1a6X zD#`Oprti^Wc@^Gyff0;VW!t;M$Xb#8(S)(wk-)QlY)fQ-qFuz*Er)r!`C}}n&{>G~ z^v~d=n7{X5Dw~|A(+P)O zqA>!_azQ%cVkEFVk>(0>v>(_Z4RY`zUVY_*Ac&Xk;_D!ps+;^n9v^Utkr6`l4uC_d z#B^VT_A~@y-r%lzJtHB8EU>T>|9fP>nx3nbKAMoI%4I74q zh1#svz9F=OE%4DI*>XN!*J^2~o0*)9{RR|lYZmhG6+tjTD>_K#${OUaF-rfSM-dW; z(0(TzAlv%gh3NdvgfZS(|I=EW>x=@UTKRfFra;gW+CUoU{KPGis)T5IB0v$CuGegd zSCt7*d_%=;9X>lgPN*$kr*Gd^$=+WYNizLyj?Id(7(S0eTh0$-eUO_EI;l|MOrKsh zOm_{%9F?O@vBA)bk)t(oELFCo+Qvw*YIHgsw- z{Pij5H$zFjOj{CA&G;D4`Wp)hCACa7+O&y(}-U=55z z3kco-of;;7i(n7@fHsF_xrCl6LL&rSs?o~OOYdUH$QGn}!rZD3il;!(D}r-NYqw6} zf~nO2F9Us(Ji@bQ3j8BA;6)D6@{|OA7b3DRvC45eFgFxq_Y-^eE7Vq2Q|* zTr`A4&>U^}{U@Pqj?~0ch(jv+P12P_&XjnmMyexzXgd#z&yRk~Wd4C*$0{7f?Kinu zb0ZK4P57mN@*RJJJmYF!=`6qj_LgMwYvL%JYr2 z=PTP5hiylt1v;g@AE%og%UMfp${v7l?vHllLQx6%k&MmX3iI@k?QA%?!lIg=P-$=f zJt#AC6t6+n;yom?V^4rocQZeK^v*ToC3ECk~Hy!F@FCQHqc<@*jBO`8~JIn7}N%N_Ew#EIL z`^1y!pkJcG0|Vca({}A^1SeP!1JXHn`TpJF15iAe!q5rnmF)LJ7DqS>p~Cjj{L6JK zrH#fV;%dcI%jWNN&NQ&TN&mzHTX6+UZdMz@e`C7EqYt&TtfH8H*8XfxRexOqa8(Jw zWFqd*cL%dArziE{acjU8yha2rcMqFnvL3>mnVDdt*@n>FpM5qARh-S3UbBwI7Hj6z zf?m^szIj*Pl;t)K3i%bk(zZ@QFGt?#O*i6t#^pRR&3^}T@`w1g#a!wG92~T{GRo-2 zkaXG_?01MmsYGMX4#$M)BB_v>qdFZXcV6s&?c4R$=nWwoig?UUUjx|fpq2ur8tnS6 za>fmo@w}Fpudq&6%ki9ij}9c1UGun{ESGvj%-*r*6 z>@`1KZ;+RcQ6r&i<$lZraB@;xg%ht?sQvkPoo=8^rCbps9u(UVy}U{ODUnkf7UC~9 zpC7Foz*3G6Cjyn_U%*C5gv*hq(XBNcUdgEQW1W%`#Ya<{YLbk@8JQ9z66?dp+lr;5 zKZpyGVL^KZs?pfj%zcAR@gy>c}eRM~mMEim<1lohTtlpM>nO6-Cjx#QlTkgG0J{eR(cqrqX@2 zEcJatM+wi=l7~y@X2-Ds_HNZxf`Ob19u2(v~Y*9m5yOCW;5WA z`#v8Y&v6f%>11a@t(aT<*Jl*w^%myLwP*RV5YoMoF{^z!Sn3XvW*y5M=GGn9Q|FCSfywvP!PU-J|5fHYe1d`3Z z9>kgHf@`|(#HwelE^gpfA}Y52)VpD*=KW`;8in{P+)qz+P*?{TVMeqV?HRGiP0r?a-d|ah{LGeh zxcT50l8--}`BeW2=HFc#lw5MAW&LrG5_xTCy)r^Qbb;;chFIy8Q)lO2mIS#GPcQ*$ z#Wc;-TK}IJ8Y#cH=;Az#Xe}cZx3DJ{>xs@h5LoC*IK5yKSa=EXdT=ez4W`U4T{*;u zUk1A>_U2cJuCh!@C5kI#`v-{xiCN7)#ztQWC@xcV?ob;`ny4)ew#%snuu7+~b(ntM zn0YW*HdK;j?le&|#JAPhv%WsFSbT&J9CXCEq?_tV)I@}tiRIh+w1-R}9>9 zwUSoa%&&SjQh3$QgtBUeE8?#Vl-i|su-|d@6Ma3F5!VGXD=PtMnT&RAq5fL@1nfX; zG6&#bjpSLK@i?PK2L}E%k~AUs$%N%JC2s&u^Zo|4!wN^{wSTCP6OE-NEH?VfAXige z=|)>Ohv{ko%c}O&2kcVwS`G>uW|bQ@r2)jD7a^ATV5;@;$Kim;A#9_E^$rnV$dj1# z1&MW^PR*q9G``!m88+dPPj%_~8IhRGadoqUmuD{@ty}#pI$DdH7KBmhPa#W*(dMQ_ zsq7BCB0o(AyAl@=a}=|^a@%vGqYU`W#dy;2!S&=?rV`1g(Q)6X5PWw^OIw>RYL=Hs z2ijq)<)fS>*=oOi=hX5C$|`C&;FXTm`{6BNX6hDhGtGhy@X7bJ)P%CkqvlB3_nC(O z-WHB_Q*`dQDB0c4Evt|+V$K{jO%;7d-Y~cX+TCR|d1R1Y)N%3Sew?}{wYPq^PMlc zG{x+7K%T&SjpvMIVu=EODL5_qxFUWGRvjVBAS45H$VAvpSAC47jBU_Q;y@H-b#tjtpj%eQTb`ug9*uTWJdKen*A0MH$9Lr9Ka?xVclavng6C*S!rOpTan zC8}>BFZseNPcDKmI-^_<)5W*-p9w5ER$Kn;zPTyk2S(Od0oztY{G1^(JkUmw(=MaU4l${X z(COQ))nt(<5nXP2J+M!Y?(qy@Jj1y#hT+5y@=qPu}{FmMCjF~uOZAjor72}orrf6 zdUcw9^%cveOD0aUvs(S>#SL^tA61v-Xum`SzNog$s>xFs4yMWp^Y{16SGj}`%jyw~ z|3Fc8R!y$T;aVS@=0kY!0lBE@zeAK*GYUl)Kbu79#Bsn^o#6R0&Cdzd0YiV;{iko| zv#bQO@5<*<9P4BSi`$(S+Zq2PQ&y~j$%u)Q8P}gE#@DKF$h4}HYoIy^BQxRx&u{I1 z$qc4i(R_z>4=yZ3@t8DTOi7mTj3FMQYL(goL4*hHi7#cyi_PURpcmQ7* zW+Je_Di3So`AZ{#YPjtaw((j~7y2^G#joy$%qd)dcS$On@jsiP#XcdG7tbws!7shu z%|2<=bug3vEK9d!NVU&Kja3fjR<3*mQ;4sXdqi-4rkD3kPyfWJhk>S2Vjn1` zHu`Am4aN;DHw)Rzm#@>6*%V~WDtm4R z%&3f47x;us4z>holuw3GF?*lqwhpdH^1B}Too@HMu`D`1juA4*YWTC{)`xl#QhCPb zNW#A&*LSc$^2j#FL^lO$hcKoik9ckrc%M5Rr{_}eOOqp ziGT&obqEBP7!&ZI(_CPoC?-&Lt1JqK6HivGLn+8>ujW_56%YOuj^6EcENxPgk6Ptd zjo2#|(BRGHI9nRn9pDL`_B!TBu*ZNoq&B8KVZtQ!r=9$Fcr4xh6#wYe@^`4!1b*FD z#ryj!gzasvx$gu|y0vLUknn5C-380ZU1p%}XD_PJI@sWOLQLS}hA><4T*R3XV~+eC zaYI|$S3xhT+1+i5K879ygjgmR&rG}C|Ql+TFEDlPW+ zWL&26Kl25U)+b3zZ3z|6O>ybpGDU=DZ8F7WS>GHwgzA-MSY= z6VzK@(m4KOInX;aOzZ34NeW9^P|LF z-kv%R%tvL`r_UxCQ!if+-TLbB@)Cm&K$@&h8+H4k+eaA*=DkeZd(B$|UBMU{8}a%D z9YNz+mnqsh_>HoGKd+|UOr<$jJwVJR;*!s`G7T~O!Bp~je=vI`b&VFD^^qp`{qDRy zUZEtX_NQIW!BmBX|77iqSlN^*T4N=mO6sKCX9vu`9TAvfdDwoe_o*9~zZ}fk>sl%< zwD2T!Qz#OK#WUKhB-;p#JgH?OhZ7T_Uwo|M_F>nq)0HSZY;PV*2zoDC3|IHy*-07# zAbM|hF*9)M99m0Vd0s!3?`FesEN|+@acGsOXi({gc+h49Cj=d|*lp!66H|PWP@47^ zC$BU^&g`?j@g){-{zyMmFXOf7BnQfbJ4$CRCtPeZraI-BM?EI+PXqpN6w)troQ zel-8$`+oNtG^ZdiUCNk}hkd?wNLCbks!sK#!Zw$xDBVu9@Ud^Q+Trw?G>}^{Ej=&q z0D3aOV1{$7zQU;AMIj0Gv%UqJX+{fIxtQQDTVT3hIbt}zU5V=vVBO2o9?%t$08IOt z6Kj9M%*0aW+ZF zGF^sYW-vJFd16kxvIDs3&*V5Xcd;F*@>%;83o`0g_8&l^$Ub1a0 zkGXbl(A`~$b>vIF$*n!v1qF&P{HenNfGul1YOcNVgVMTm%`4>Mj1mrWb|~SlWl#L$ z!*|QXjN#9Z`@@HuqaO8RG1j7@tqtB>y#tdN&qco4H6%00xGh4be;qP`n6FN&r=jk;Y;mNV* z1Cv}`BR|UVAnd7{xYfvYE3~@m*Hs(j1a&Uop?PN-L5#9TwGC>uW&P zSTD&M!BNB;)fJToLE{9^u8xCjxFF_5YZ!feFm$y6K4aDGlE6vdzMb%DYO;rVFWc<1 z_mhst#eDt(K_7IXYDki$xXt&nm?ig=a`7x%Et&vHo zX#jV_fcNCl#VR$r(7Go0iZZ7sjQQmrmleU8o{t2e90=fI&!?+}zo^bNA*^#h-^Al3 zit{#XS~LAjbt^x9L+%Pm@YAbt>UNqN98GVh9PEY9h2`wh_@xlUNSV03ubZ&6K7qK^ z`Kh_rDMl;PcHcUF(Z4R=f8O0f&^eKUD$9Y|JWM=9RGT*&nCT)hn z+5R4Wpp4pxIknG)Pqy@C1KszeG;s65Fy`1q~f4$)<* zCufX>vJ>Et_^7`_m0o3^H~#7|18VASDk2R^gm-gLWx|in|+`@ zla*WWM?vq_rGYRrfo}uya_vYyH%Quq<#^SzMz%lxYqjh0t*>DUw%xF2S;@C+L<|Vh zlhN2x`*l7C*mKX&5N7s?6rt7e^6P*B_j~c4aeU<>{R2#EgN~#-uUw4B z9ANS??a2H#k$!w(U*1%`kn!rE3oE#(Z)0{w8CCBx{#=pqC$}O?gt#mQa&^wptK~6! zA|;@G$^MpB^dYC^k^<)QZ}#sFM+;t7fN_jSSd~Rb&(JQLRRw&llO zB+@&>dH2T8T6>1&Gc0w4MOZ(0>qku8Z2GCW{g$zz(2!Yr?FwUg*RLC@4%-V~5?{g; z98=inB-@oP4r;I6+iX*8G5B-GM7);sO?KBB%xmf|n=avEi&mTlNzq%hZ=cG20@W^w zNFKv>J(Cst;`>eLy^$H-lx%6iI4{ z+Y}=SeziVXhv@i=knarj`y6Lm4jLNqMG9VZA5WTIn=+9ymAJClbozIQ7;BmBR3_LJ z{Q9;$)kv;-UmgpHqNx<=nn2VTqk^p1_)!BLi;_dl|2$Af9emE#-1{iV?7P_(a(GRS zdCBGDpnEL)U*zWr{w1m+-8To_uosq8`#F2=c<2>*VCd^%``872u$43AaP?K9c$%*g zB2e2#g2Xp}d!B-bgb_kibUB<7hCCE|7xsCM-C%6_1c5jKM;~pQnWJ!31QkNtp0-bw zA%JLP(bn9-w%2P<;Z#VVKw0!^*BWJh*O{US238$@tAsWwn_y9KtM?c8Puh?nOO=_1 zO?XO*7?>g9O zpIYM-PnQ)Ddq1~QE3vd1NlRNeWs$xq5!#0lbNPx}R4n7A>w5wQN%MbJS|(n}9Yk7iPH!#G#?00Elhvr(N)DskJkv=SRIqNW+YeaYRRc=0t{A z!PpOa>~E~R*7Ik8a|M1GGZd2DI7q*Yau~43xm%_R2=Uxe4cM3=%$D}5Q;rqxF%7#p zpK=uSh3pPt>}T^GVP-yc8yYI!L2)fX&99n!k}@Z*f}>i8RftV}E&qK2^(vO{iWd7G zNV<0Pg_HwiG4z?Qn`fH+5a-8?yBC)~_*;3W*f&UCL$*>pcI!aU2L@yO`zVa^`l%dsUqIu_ONG6y!VO2E;3L~<43)@T%l7)?*5~)#a~BdV--=Q-<96*uuz3t??Xo- zE5F&bbA%jQt@%CGzAo3xy_&_J{V#cnPY zgU%??_E$$-4@eVqp1-|&zeHn`p|Z{$27B zH^`0P1n>ehn5M0l`%=Ymbw^_J5f3_V03g+c|LJyvV82ja$~0}ZaGZLZznyV<_GNlH zC@-$_QsjNX8?);frZiCkk@j_`p$5;bS2Q5i+>MMDR^KO5#c*}b`zk1B&9}jJak{dH z$&hW{m$?=+JEtvy*SD0Q=6WYcbc%s$i}&v)4kwV9&eR{+qqO*(M#JQF*+0$fv14wGx8pgx|~qE6-&YGWWMj>?Qm>TL{JU!|dDznOy; z*aM&xOCSX=f`0zP5x;xdLIIj@@oyx?*n>1h=$TT4adm+=5I427LwU#ZUfgSTdB2*A zFDcR5x<6lyI)8+BL^*Ehc#v(VKM;=2agm-n1(#)>1=}$~jlmi^m#Ky2FEOcC=x9%~ zL*D3gCPTagp>0b-5b3QK*79k@i{6fuT5yu@RcOhd<-ckQs^}7)ZvvpsV+2{q=*7bw0K%d`c*o8G%*U?*eIlSt(?~x~-&#+U7TL4P%(OJ3q_O{d2U2jP zOv*O@|5e{$u5Ym_DSS5Ryya;XbX*wY?3t z^CQk(ov$p)o~|a3lh^0x4%>XuoLo7pt8xcm1ft53nH~n?t1*&**@XOGZfj5P_rfa_ z4|F?0eOKU?gQWJuY(nlLKdvz6h`G;y)F3O5MKjCj8#0s)V_7oQ&6}Wf73I$ADTmPl z)&UAVW@~W09M<^E2d+XN%!cEgD>ZNRz}Pe zkjO@o5h^W#4{Vct03uK3d84QG=2W_jAEh=WcCpYE(;fbdodF}Z?|`lYHDD< z>4l@FE{acdMS}Wjj(I=uz^VST{${AT%bU}1Eug*qOT>P~O%u?%N$CBgdtE4ZvrQ?S zO7@X%pF6e1-7xS)`C>EN&~nrdl)4y%V$UUr5bXiR9R1S8M$}FCKD=*itzplO8uWFc zVol}fceCl-D)m^zbnXg^{7F|_H+AbKy!+xNlDeh`k9yX<3?Gr~7^cP<2yM;PaN7*Km?2%25aN9=jffufHVZZ*c9JUVo&^d%g~U5i9Q_oYA0#0Fq=(Hd08UXW}&CAEBEAtGZhm7df{v^+GuxBRTjqEXU5lCcK!GMU7z?SKT zFS=xKy{Gf{Ay`ng`8PV!0o9M@Bub%MhX+4^jx-5Iw>p2gE{Oj)3p9eY6&K6XG^{1E z>yBGDmg|9CN|G25TwcK+GnM2i4v>@pqs_X`m!AiK`?m_$8e^0{bT|}EnMc>(=>;i1 z(c7~FRKA>|$^&7?lk=m&N0nh;fiXa(48YP7i z8P~jk4G={yx~ws(CU~e2EC|yPa9{pMBYKPFEU2By{k0e+d0Yjg&rCWUowUXt&9Pe9 z?ZtxOAg*1H=5p6vCj;wCgMj2jHL5lg}@ z1+wsKtZx!y)l0?**I&-qTukgL6aa}{MR)tRkl);93H;fg-{D%Ame%~50wHY<$OKKd zut;D(TYIk9c>Z>->=qCE@S*&amYD`4d}RC+BS>G6j~;=zCUOe0G(;C0OCUdbg!Hi1 z_2W1^kf8ijp+obW_w<1ZifpxJWrqMT8Ps=$`i?-TWNgYPopRmnPiEo&@Q^mY{nXHB-QKx8H!-;!7&YtP7y}|r zdCRoAg_;HnL!lNaAfuqhs8TGbu8e?2y1W#_rsdhzUu?e#mAaEUOP0MqCNv|fINRc1 zxCik(*KuY=suoEzs;T`e;>pIhp9xVq&?8Ff(0R|}gI^NBjdBLt3?l^G-_#kjb0<}X zBOhXOIpuE)ZWCOe{|&-0Nj4>g_`+*W7S?SE3=~vPfNdl=b9`oXQ?wc|c!HoXb2rD2 zA44F|@R!GcmC9RO|SNk1zJFy2) zL{%IsM`}{295tM6l z5oY6levLmXi>~5WF?c$#Pq)p5FD-6*y8pB9oOA?#1}dgn^D65v!UdlmjR)q4P7kZ0 zN{tDnxfj(s(7_6FBBxKgcr&&uSL{;lX|50tD`p=b=-0R)&$oA$j(dlQi*JQLtYI@vIsOjnXV-l@QSTMg z7yH;ChQU(&da0Ll*6Gs_aE?X$*_ zJ$DA$V|IV6#^YX0e;4N7t$uyZ>=_o~E08M-fNmWcbRRw-!0=&GL38!V2cG;&#RN;* zhryX;kc@yIK!eOaXP-aHB4{Rkk+Ux!?r=XVYdP2nb}IBg4g3mJ8~~K{63bY&E0p98 zkky(GQHcUXwE8R6^>fZe#)wa2JV#A3U5U*2cRROz%NZ#I{OmF{kp~mC80^j^_pH+N zs;2LDkDb?vi8-DTtbVA3Sw@L11?e&od7LU75SM(84@d__|LXYCbYe1vbp?pBeI_7L zHj;==nFFFfnSN;yPYApLKQ<_mAEKto6|O!)@2X-D9SU44q307oQL~WJ;N-oqZ)M~w zB~-2yG%cn>-Y~!Wl5E63T1@?nJZ7a;1zgvMRJ1*0VF7jh#H4UXU6wNq)tW3;oVi)b zll|c+OL@)vW4zdT?qU;(QRLDPi-*T#&$6W=`*28b0<8$zIXQCrXqt~lxY>JOq6PAc zvDK>{qp8xEhG*)uh~cizFiu^Rexly)UYh!NV6SU>)f|rx^I+juROOPncduw?=gWuY zXpe-1xk%sL`k>3N9WYbYCNX>dp2U0GuT5qRXZKn`j=oG49jl7`zEX9M!aRvCrATS| z765195k5ctc=%dQoe#g?%Sw$;t$}5k(9wjEd|F5;xSiF7UwIXIyAAx<;`jR3PI3pW zg8Ehu7^ha<32+xy8^ZVeS0+M}=^QLA=9ajC1l)^G-ZkB9Crdq~S-xqnK(_K=f6%Bh zby0I@*weI3Hv{=%X|rh8e)LJY?G3DU-)&lFd1?;D>o>NCIwSlbcDMc`FxNe}L%%!a zy?M3;+*$CiFv~Y6YWuR4@UH?FUZAt|zcTOb%Z|doB*ptqbg`SeZ^J>fgp&0VA2ZPY zmGAYnW#Dj5klytdt8$p3xqy`fXZmRU^>5_%w`Id5`(xGfN{|*Yo9Y;Pn3Kq<9WfvO z$=H>u@81_v(iA|I(gE*Pi2ey*7}INQyVmAg12+w3IZO2_Y1$2;Ns7-A1r7W7iy z9=OEDmAEiZw^_1-jbe;LbZCPKH&bRi$L0?a4YxX(9F?sTM=uZm>6Lvet8602nRcV^ z_JbPba0U~mwNxR&w4{4#_ZC9)Xs=Y`9O25Zv%MX3@>mBLEc$K_JQ2|7lzURG8X>P5 z_-!%JOZfdyiN7@iu>+z%_YoD;t!VNpo9OUbEv7`MDJJafGRxS?uky-*H#?JblWvm< zkJ_25tS`dK(weKT1&)9ETC_)cO4Tad^(m}O#7ufHMouf0{b-XM>aXFh&9%Nri|R6{ z$Sx{Y?tZLO!`oN>4qF$_7(MrusMjl1U=!z)yQG%@_=bxzRy)6NS5lM{(CwEha0P?< zG~wcs`>Li1GK8!fIWc*wPr8SAuC7_j5#?ufm}adW)7; zXJh)vi<=jClYBuKJKQ2)i+iT;On(Uc(ZXc`f1&_t0NQZmJHbYirnyq(HTu-ZwE#O@ z{jzvbdCSNXJyJAfeKN^~e*FA8j?KZi7ICgA#YVMlUaZT}oTiumvY7XRlr&;=!f%`V z(~)y_Q&OQ*edOa%&T5A{v?2cT`KcGTrd+_z>O>fLVc5y1P{CJI?p@wP6{9KzHg*QX zDwk^Ky8&gB4PRPX@}&f5Lmk5$_qDB=+rj=#a$)$n>N|+d;FJCDZw>R4bCUmtX(qjL zm&Q@c`;4v%O}rU1ZDdgP3xmvO>Fn!(o9$O0@^qZVvsgVf4t>CnMSU|i{#GJHxSApz zX8M2^iWo(JGx+}ng^<)pZ@^!za1lXE3^qn_(UbJ<-4%m?x|2Gm9`e{FcO1_SS6aBd zg$MDg;~&QaiD>B;#@wbBE2s`K(jh z0KYx)Q2y4PQ@Qj7Dp^{h8o$B8^0E$0^13oOgkE)F%BYcp>t|QBYth>Jj_*#f7e|`< zv0`vVsgA>i7ECgi;Tv`anz#3e^i zM?1&dOS22l-n~RvNo21wyKEmXl{`8GJFjf_lXD*9=7p*Zs%ZWNl1;uB}qEf z!=niavz(EYn!9ynBY|SbH7}=1{3-;c+J$~M{$?3-56`m&8|xe;Fs>UG zV<#pc>GtM*$eQ)?#+rYl_@gufPJu(WDus8ZY{yahTCOic zZgG=Z(sY^bSTE=G4}^<>`d)zu0BIW7rK3?f>A-!d4#hrGq?%?@Vi>3PMteG=Z%ys9 zCkEMy2LI`4_{oFQfTd&D*mqalGxc`qoW+H_J5U!n#f_Ss@82gB?s}c^lV2@`ok|-s zX_A+Kj>pUXT+2}l@#Qsp1zFI=?~8n4#J#VHDm-wjT6;2)4bMf2>8D>VNj5|B?o(S7 z_KfALd0XzFc5(4(i}f*OwrY!SWcgp{K;OynlxZf!7UKGv-@Ow6o>)F`3CA&#HL7{@ z7&?zO7wC-6@s;Ir_es%UmAHG-8_k$f%TUe&^2UnqF@_v|&Cb!A(y{#EbvPGZv5~(g z3Ml1}2q&udoyIAcHi(LgkF30aKd(Gp1T=+t|Dj6^=XE4#J6)=s9ri5LeSs7$jFFJo zzY44ku4C;F@l$H*E~^ha3d&{Ehl{{M^o36Xf6id`azy*S6L1Ahj0&nbj2tDi&T&M>2<1tZOPV>v=e=ix zvgA3JFPOBV(r<`C2iF@11CrEB7)h)^TgeojhQ(}nmwSXYBVS*U3H|ZzNu`F>j}8G9 zcFV(dM+TR7S7~7H;ro~-$I zAz^a1ekh!&UPQ<@-=6bC);}M#M)e#V6&}teN5a5H*vfCCqS%YzxDA~cmjwJTjgc=; zU6Fc+8BhQMUI~|~uhaoedWh@GNYr0>^~PE4;>FccK2syC5ssyVr|CxeUYNaP^c!aZ z2UqKI9R?V?wfFOvDi*%bd{1F%ZaO9eSdce zPz^)Jfhygdw-^v`TN+Zed{x#18`T_#J?W(3(gG!BFpTdZ|M@HU81=-I)(lkj6C|lv zpWps;Ns7_`sPaNzWN%*;Ygr1aG1%)^()d(kPz^}fggn*EYqjuOztQobUN;5V=^WwD z*74FH@Uo0%F@+Ge5Y0!C6~E(In!YItdmi& z4$TW}C=)%pb!_lHLz5xqx~Z5@)5!lRFuh*+W{$!>Cr3AWS^9s{5$~{-4fV=i@3*L0 z2o={vZ2AMO#Zy%1bj`Sw^k~5qL!+-py?K4+i#{D!&Ej*hp4Y%li76HpN}X!6OPF?0%+E8zKGE$lgru%I z-6)Y|p>ebW-6ipKUU-)E#jN6~Gw9t^bYvCY4Y#9Xrtm~PdTVJn*=WXcy2X_SKZY8; zraGR*Fu;c8g3{A(@=P!nosm2_^G)fUmNvA`gPi~1` z=MB|H=icfJM6e>N6rMC|E0o=fFy_ea5Y9dc`?r;>e4E*RS^| zVLn@vv9BZR_YNq*TQUb0#tgLqDX5BK6#$_s}Nu+m?slt6}*y-yg*= zt+HCY@QKgG3;U1?N)k%uYGZFJ;2h0kpUO5hu zB$}4Xho;*CGd5&CQEa$YjP}7jzho(Wma!RilL23W_*L67NJ`i+=&|`fVt;HOM|acB zkxj(U#dxTb4%Tuj6b3(J_DnxoQu2weO$)D#$NuGK4|kV-pRXwW!CtGsEJ~xe7$Y~f z6X!d|4!fT7Q4Z&8x^LX*var#YUzCP1<*RVMJP3ECI_q7OFI37fXmkoYkP5u?PZ|o; zF93B;&}~SnFALdkDKN@@shje{>aGcb*{a;g%500C?KwThcT74dlzm#oC(iO?;IH3- zYuvUEc(c+L>qN1Y217hICKO4XYtn`qZ^afer(H5z#D)E+&drZ)xM8>#a!3y&#q+*m zl{$hx_4XqdQy(gtM%P1WK*$MIyaL zHY1JQH~O5f=dVNkhB&LHFIby*EA2+lVCkWAlP6YFNa&1+psHjSSyFQ7kRC4x=>Bsh z$WtP^0*j6ACBOR@iBt9u4~nTDDvwikd2;$eO{3ffVFRB{vJo@P50Lg70h*%x(Bb1= zIbS7M{@8H>{qN|VFG~`XYL(lT%nE9=-1*ZSv2L`ygL}>LK~bI`?)@e929#NnEOWe0 zJxtCFrNVG28&LZBBXpR8XS}|A*zP(C5t_SyOAW^e{5g$D00c2!=-pXf z%TI***~F}#o@+LWEn&)Tt2=z-(l6OqW%pEI9et0GO?1Q<^jUFY2w-3mn%HY{{HM3s zMM=@JrL3Zs(TSpRCpY;-!IOwj{%y?YU1~{^LcE4kcKaNlVj<#!frW3axZ9)Jp zE|bM?CPtxRnCrld0Oi|%X)s1{5>EvXIp1}1d2SLS(5@%qhrNPC;__wgW0u;d%!M@bvK#w!c)6TEgvQ9o8zTTT7hK*bn3!0x7 zcfZ!@qs%DBl>_|CyJ-QSlw--6ScRruF=E`Ddd%UkCl1CBa};=*T-j}4Fwjhy)6oMC z2qX5vX&SZLH|;$X>YcL(GqX=igOhGbK)G9Q{1l68=c314`&VYwpf-*yxW~^G;uz6y z<$qh5D`z}kfm4>M48wJXLcl-gkqDc@Uj5{7zAI|Zh$^%Ezw|7u)>eX(t5%yvTj$&j zQ|Anrd7?H9gXx4cH%A8qTuXMWzo%Ylag8g7wS5ZaQz}*W_bEO^os-^L$eSz2T-(Xt z=dACRaBCoA>p5n9`*%CM0cSBgNRL}|;o{Q2+|56A6$KaJhftfwGKbK!07sprqkipl zwk!T+ww>1Jgrds0Gw1iNf%`C*%Kpid(S*~i)0ifjM-Fs7FCMB8Pj?E4&89MS3h;!h zNW)(XSVZz?q@faB1G&=DP!_b6GD5zgiMEf$i?hVV-(7@&8p+0+#?+ld)UX0cEsX1j zdwlvL4_sVozY)%Ow|!V2F`LnOY+yW!iv0blM|&@Mapi<6@(guyac;7xP>^bTV0u!# zc=4n=*EWNuolUsEsEf-=C0O7J5BFBBapW}pC-|QykV*mK7{CrzWg|)gOIqt;>==dP ziV-0AI@jdt$F67lm=OK0h{QL>Ckdvf;>;VqYA3Ew@2ht-7yLU|p z(|la9Oy_939pN2qILAgh1&DFP9NPhgbVfigs9j~X2-yO?5257Y&s|4P1GN2h6OrA0 zS&Yb0MG!cc+jJ>9GbWuNfZp|2pI)WLs$h=8UVk9G+oyto6Fp2f9VND0ps-ncZ$TY6 zslf+=0d<$cEzKpV;VxAm?!b@?A-xxERB=E@%3v~;B58)jiS$TF;Q=V?f_w z?{Mp+m6$LXz=&}G2K#@mZt^X=j~|9be|{~3oMY+6$6@ftVdVn$7;UPDC^N|l7nMsW zT$Jl@b9=oolaT=`HQkH>+CbV`6=q>JuSR&a%*;`$IB&2clyna1VaA2p1^)13q#z!5 z6&@4KILe@n`Oz|cv~|z`(%tU-2``aja9s`8OXp~@iBg~D+cB1FKvKXMy4A$_`FZ|! zjy^d`ahynKM`!%|4grb%M8FJGibe-5h6l!EbPA{gqD&+$_#|;4X>PRo)Q6Ch?r1Th z61sJEupB~3`VP5slV{oeNC(3Qc%t`!XdXgx_!ZtYyiMa1qhLc!8%{raQ6svnr%hu$ zj0N-lWJH`X_|^Wo({9q6?%Dbb#4EIV-?oIsro~-FHx^9=sF!%7w@5pMG#An{0(BIJ zyK|C~CxwB&DT)npWbyBKg{e${wV43aYc_WY!I2&L*$@^?^U>#vy`bp#gYJGicP`jm zM7O3aE`yH1b7`2h!2rHKGoVhR*#&TAX^h$9CEU+ZzuP`h_nsh72>wb(v^+Z5Y(Yml zUll0JDcXC;eQhqo1NZ~<$pGv06fgC)PW!OIxDYYSGt(2T=u^FBo!cprs1xWM?PyD4 z#YZIY7>vQYF*Rl_+5wiJugs*}oZj`(I4KzHj4%3m8tOfb&`PtT@y7qJ+jpeNPQxkc z{u-S6`j`V@U&b>B^XDGnz^Q0t>pGKWs zlh}_04#8g^OGwNb(|E4r?#?k;LVkt&?*)iw)k9jEJ4;M4lF-01e#$R;;D)>1`jzx5 mvxI*S3DLHa{y!W%PNK3TLhaU?KhjP?0j5X`M8gHoL5+sTWC`bm$pf*VfB7z_}=bSTw5(GhV1_cxqB z7sEo@D}L4(0oOpm=y;b1^k!yH`Oj*P#6-UY*ub_ZNiVP8(-i47*Q61o07SqjS0;km83P zOjL507KYYfj&{XfzpC<2rT2k>*8GVPk&#`cwuU;oy1B1^!idAe!X8^#uyAv8*Ecm` zve_@ceDUJNw~P$kv6T}j_s2(A+NTN zRy^nS2F{0#3SYi_xmiy{fcgC8%jVl5f_5scrSowYLpDn1qV$5PTFbw^Mk3QZxE3Y7 zJUl!qPPSVswi+pFVo* zgr8sS9@69u{aQ9x=@dC(95vx}9uEtUG5{Y(K5c-gvya{{o+i->aR(x1FR> zyyS;;|u&}2IvZU|c&8-_pMmlu&pg7Eje-e|Buo=JLK#g+ZruVc=H+as* zTXtLVt4qgmy-PlsbwG@}tX%ESI=z@b)b{7|?SlqZ>Xc8k$$m6RGAsS&n<#?G22xGF zg@t5~qhgqt9g%a+KtJlE8~KKv)2Aqr>4Pm@LAZDVul!>zE!q&FePSV(`mKTLv)F=y zHT3iTDnmiS?DWRp-vMEGLW$7V|I(IxPKBQIoDD`~FG`tjq+<0q<#xh%$NJgS2#$FFXN6jh`8285fN8MS&ARNhMX zve}!QyL?;i%}228ekrciX{`0Zk&=_cp=D(Bx1MR->U*Ur|MKOmhuot_7@?t|NHMWQ z_R5@`kIl`Jzv}9~fBROmWS7J~_9`+WVspFIa|)JJgTMy!=N~`9LWE!A*HBniw3}w1 zhO<_WHfO0%CPt zAKu}TCd${rDi)NI8pzdad9hw~EPQs@Il$+#d&!T$a3{LDqPBd4d6k%*^j7QG#_6C#E$Cw!ZY;Xh?bs*& zoTujYp!^9_H&LjjiHYR?pCiwelp2GnMh2@rX^f2p5Z|)0rYCDLXXoZP4BDJ7T)2P{ zBY21{d^oEs{-Mmb$hr8<9UOMzaGKnuzIoT*n;3$^cl_2kV-^U)C+ZW zv6cRxW65iS?ERHlDJi6eJHy)WVB+LgZLN`QJuF1U8(WpL2!sCbD>J1^P)oEyMSM0DYa9xOw8t zVSenoFNW;G_xbl;UI^6gj(b&F+N>b=3F=iN8(v<1e$L1m++el#(JpxqwYv~W;9)}h z!-?#%Ih({i+sgB^a<{WXY>Sd91B+69G8Gk^&eyWy;$^Pt5aOFG9M!DU>F8ua$JVvx ze%oCMj(pRC>^v}PRO{J$A79EU=ChCCwD{Q%{y1GjLMZvdBAH$#?TwFDy6&7EZzOXI z(bN0heHO)~{Gom98k7FI@`X4;jRm6WnI zS`=R;j@>?YOWuI_9^2h^hi$7;Z1LdXIBz7CK-KhzhZ?u<8FTsI5KMiIxA>fr;w=`K zw6L7vX*V86blo@CVU3}_VegKRJ8NmF1jb<1DU(Hs_WhYgnaMw&eO4!`38GuyHRxCm z!ry zozabC&V_-Kb;r&;x5MLf0gR5jTwD=an-!Y;AQEzI+%$!>VN8)fCKl2$$ zqs+2cvF!NLlIN|zsAcbtyCz|B_DD$U6dfCv7lp-g;Dd@Omm|%~XdD&llB0i2ST*i) z5zj|y%{$h`*7LFVNrig9^F0LDBSl5Uilb#FfF_T*62ep-ZbVo z;n@GpV{UG);$V7%*U3r*u{JEmVc3yc-{`R27(z=o&8fyhdT#~)W2F2gQNJwnQtcO( z%pZ1@CtANC{?wEC*6N0|-(tyPI6oS2JvcnfwbaB85k7i0-|o6!k$k+Od5$c$7#rI- z-0dNTfAfAg?o^)pl$3Cb&!OB06}9n&u^5l7i=j$r;I6}tg9@n_#w1RWAi{J0A4#g**?NX!Tf9G-O^t8@!(bc9o z8*QHW(z7G$i&)(44h)2$QJ)G%JF3Tr#EiAE=<{Pk1Jub@-?5*vzUrwGp7P<~s7iI* z5H=iRxJe@tv%9;|aa@YNsPLK7-G`%W!Rg5!-XS6yEq^}OQIx3x6pL`JNN}7O6LYq{ ztK^AGl)h`B#NheohF8ARk?)!NRZjfkr3AUDjEwTlNZ;|w;~cvsKeyvG7Khcr#`2A` zKmrl9M$NNrY^4l`e1A#zLkrDY6a!Q>b5?}?^!`%eu3ap2^&|S~9O@WGI!Ajea^j7S zyp)DXo2b6tgUraNV z&z)|uW7@;=Hu>C5FJJJ`=YnWBr_mp+GKaMeG|Y9p1POcLXR*dYx=7_hLNa@-w4R~i zR)@edt5P)t@r@hPcKiVWOM^y*f&K+AS?VTdfS3Hj9k%8WSRq7XpCgM~O^DHvVcHh2 zZnAYv;=o}<-io~7GQqqr#nH9?`+r8haNk77hOK+}>4A#L>kN*bTi>M2WH z_jTB9Njf7f8b7BpsO+=|tM%C*9IKXENIRCHj271N;_jd)$C%Jzm9L|pHII!? z?X531oAFatvruiA`j)@rH)x;ZP%gh0h*=(@YePP0_E}nLUB=4D3x2)(CKuvlNzQXR zTuvB+_TIe!O^APjTlGX~U%n6$+bxp>;n28!%++lQi;MgH>m4qDAn63&P(Isv#K)wh z4m1e*_Ko(|tv7~A&e$1>rD90rr{Q%iv8VKR`50g;+O3gvlxzv*9;sBQzYZVJGVJc{ zeaT}!T)#cn3SXaeo}Umv5pmCTe{~9enlv=@+`U&vM&>HI&U_Zde2bqyP9@(U*Lp@g zEG*1vw@bo#f5%?DQu>V@7-JXWaN%!dN^f;E=lmyBxvavGznNH8MeAn?r$MKf;H<)7Bw1252 zbH+-O`C4}@y-F!5xe&dwI;Vjj{~+ksuWc8VC^dd4K+$!m=ExcBwFp-H(8^*&cq=W( z{fL%u*O>*&{b2{A-^Vp(JnEazEPe0;0AM$&JB~Hf5s0(_E|oSgpf~N$wC;R}hndm$ z^yN!pF)^|AB`ho~q_}tzH`G>}K~_SDreo|Lx#LcN*qQ7&Ak8wyurpr zGzC-5yuqQ?>PiulP7);Nwg2V0Ub?{e;K7%ytYFXcIG-IZjP-n#OYU$-jWwAUkNMV3X^oaxOUH6xc`YP+qVXUcCiV2G z%|l$~*9HEiwhJ#*RHP=V+++D{4W2{-X(A(SpmcdfWGqD9cA32@HG4%C#rt9P6^(Rb z>CItPwml3ZzJKvQuebLf;`y{|F67y+5K5=h%m`T9unrF`7vDxjn!p4r&>%j1yA4-- zU$>;w=Zi_%&iG4D)Xt8HUUpsVc}_L#Tlv3HqlGd1hyGy`c?1nplNm=>M)3#J?3G}# zwbMOWbl8b|t1EURQ!VmMYFluhj=|!lGB!a#*QD^(zeujaqVo-1m9poxx^-H%KOyY8 zu765ROCTV3{f*|P5RiAw{f5eK6NLAO#q&Ipj_VFAnz}X4^0Pq!oouqd{j|-q8S~gC zh|uDorT@xJM89{3O)~NHrVlz>aFr3waWkLu`Yd}tZ8IQpKN8 z*RW*PzXxEW>KG0SO-X;H-K)H8dSh{|+b!zv?|yr~{8}lqBaN#VFTVv@yiTYl!hvNwlVh)$-I9*W^gHA-M9;U)qdCPwX3a?1!>ucA z^h81~MKfjW*2;b36PajW(CkbltklSm!D;=nU#BpJoF|WH3+9FJ@30aeuQk5X)bp8n z>3UvrS!juDq;W)FO6czb;^a+Wc_A!<>~O^Tw0}nO4E-cVe`lDeY*}OLH_47oObBIB z4TLDVdOrKh&mt)KG#Z!tVtMbIePdr8sPd&*G`3> zz!>Ur+|SQE(6MGJSB)FFc3OqrZuNdpV(eRn-q%0ferK-R@`WsITeioaVz^jUhChmL5A4l=wIZXJ_Y8 zkK;8o&3U|0o&5XvZ(n8=WLg8C)4`pMs{Fv;DNi$%a7D_BD0CYemu5F+SWGqsdV4>{ z#s+AU_4f9v^?EBPC=3-C6SYRO8aBMYu{u^B4M7;{Ww|`3Egf{yOA>M&EU~_gfC`FH zPH2x9(n2j8UKES^l^V`Y!MY%FPDv@L&z|RJ%o-I39G));2`6;Our)0N9{>Jrm%xqx z_%UXQv55#ap3%lyciQDmGK9nWXiE&cZVco%U}S-a+1eY>bXCh~H<;OKq5;y0MnpuU z%yC23!GR}Hz>%JZhXitvjSa`Wd-rDN=V4x+xVZ@;6w+nyLRK#;<1ezD=ztQeGlm7$ zx033@q>tj#>bMJ;8q07~fzDCTBV-}Lj)QSG(rFcQ^A|4b=6t;6Hn7Nls@z@CQ%%CG zTm#t1sDgF z=yPe@WdbhTMJ)OJR=f;O_(S7kS9HaYXnIUa{Azps*!3H8)-N4yPCF-t#*meFEt#q( zHy_&yVGCk=#l_up*Ucz^64RG2U)G5>K0oneahU@?aMXT4UGQYHZq#k39hg%z zk(#9W{JE%}-mFRwE-r3BP|zJg!G!nkuO!GWkoSloXQh3$b1W1{Rz!yUGWw;IS41ri zw-7G-15 z9D#lHCQ?a>gpZGpH&IsU=W8(?9XfV)_KO!UevFF3t^u$EV4%LUGvw{ti?A5o8||Pn z$_4~Ju3qHVmc7G9er@E-gW`6720_8WkuOq21{@vsiKjR=Ha3G5&YUdKfrPik6T0qL z33mz~Gt;^FWrT*{}3RpJHZQ{xtXJA4CVj!GVDsb+TK+BXh9_kQQ#$ z>5%gmr26if-;p^a+w8q)HI4wba~zkYM6=rCl~QapBcg!vU*#gX-_ya5VF6QCgx6-9 z2DgNkG9?apetdPqgHmx}?IJG#>K%_}J&s(D(<7z~X(W>6Sz=wwvOP|-z;o{@8$xRG zr`r+ZP%Jo=a~X?wP#}2B1}D!v#l)~afBp>jcmMu#@c482i9vHM)>Q#4FhEz26nyDAaAs#&XOBSwX`1aC=6J@Y%$ejJ_1Nk_fJX z#d1^3itFMmYjlGy;-FRY?3b!#C6?$wmbyxj897=i@garsHXUN*5i3Ei$DVM41J37k z%Pk5@ni+@yTJYYKfj&z@{MULUMv2@iu1GM0@Hykb`v5f7bQX>b7$8QBPann@vz&y zG_;UhPYti`#SM9dNOGPhr}j9|2FxA;=@MoiFyFVo=c$ZTwdhMMYvV0^S3PNzTa_|v z;u&Xo_WBwuVUe{n*e1%%B#)o0ptSkv=(aWO@fm2cha}c(uOsME9+yt}3SY!IbS>pD zHijwJa9!hGuS98_p*+t~3AdlNCkhll!iB+ci``Xp+m+{6{KBop(Pb#=nkBR9J+zuz zs5Bdv_z>~-U7|~c_DGRA#ZNLqLd`vM7vXr}&R>cP^~;u1!z(U-GSZJ!rRw-B5izi4 zt$q+qLcr~aK^0e@{?yL6PPj!V8y#4P;N6!rmm4+se$lzKbHyfxEOWv zDAdJf9AUNcv)Bgy@+`*q8-vJuVz>kR3zT6qc2>91yQ8ex@Vr&A*zT4R6!~9+feJUy z*nf@IaQV)@e2vg}>wbpGA22mj5vDl@le-C&SJpAPzI`+_?ix z>}t|@wq`Z<18HR1-4;H#FmBU*`B&T=tw70;tL-hdZv_y>sqRs&bL4U~h0{pq2N!UC? zm-(xTg)wdP*z|U@afFE#>QVmjO^8j{Heoe2qWF}&HQ(e@C1qsvn?K(AARq59nXgNR zty;zZtAk2{T@CxJ)*~nx7z~$>AzbX#8yA{eTW2gik0}BI0!FQys45^uQ3|`2pqI0; zl8MQ(ljjh1)sVx*I{iN^UvD^$3M(^s8ixsLU>FUo_bT_IV;@Jzg(+EEW3fax(Hy1>nrRc0 zL=Cj&&32CxO6KJ?23$=LICx4K6aOGLE9v*M+Ti%b`7G{f=K_cHKIaW2% zdr=8Euo-qxGLzu4MB@UoAYs*z&eo`0E__ZHIuC20$U@*($C`w7Dc*_q3E9KX4C*=~ z5J^LyO|v5ek5}`Vm2yeYaHKIH7?K416n{PrWjSKq$#YuL@o%7uq;i)d|QMNA+ zZ#_EG$daDzR_}e1er~&sQ?g#R0&_E7{c+LHuIKo9fP`iG3NIjK4+fS8U4% z`$ySwA3l0>vsE)rB&50lXc^-mei&Q?CeHgqvr`O{HT;|yazVkU&?h>(KQ&+Dg<#Oq zi+5&_5Kc_ll?|4#2L@FhlKDOoRYnjo?9+*dY($cX59A_|vP=(t&d(^4CopDR_~c$} z71q==wUI6o*}c73{2F;tQjT@9^@HEH411ltIC2+z!|T}MD`C|I%CM*p_Od<4$n5?7 zoy61=!y_W3g;gEB%3d?PF~1PIAN~f9s`nAB;x1Ul&CSj5z^E?Kbx#w|zWTT=ML(yH z4y(C_0`wXan4Tg~84?*=QDb_w8|ZG&w&305^8fhp8fq76e2v~KkOM#YA_@vz(SP9e z?qZh@;8OT`j2M3V`>fR`c~D8KMI|Hvhmj5Tj zZq;BCXcfM8MvXa^g7d<8ukTgq@{bHb7Pne&EW-*G78Zkgf5Nt8k>trqEra&B1}Iq2 zBpOeA`)L)>{@Xh{_ZS&@b<5ffZi|SVzwoHaS6l4z2ejK$Sv#HvnpwLxO-TYve`fm~ z^AUNV_}=`yP`WjLw!75TG4e${cRuG-wQRnO0ERPhy6#wUI3JG=I|Y^#Nu(N1Z?JGl zHer&~wts^;&PcIrJ5oMBRYvK^%Rk z`ZW?CS!>@#5a#}rO8OJ{k+zy%vz(z_`?%N+c>S36uecsddro5+@EoXxTuL+4?ZLfr#Xl%W(9}g+dy3p02Z`)xV#DjMZe5tz zSmn3hE{-F2vp4jxQ4sknoF(}WIOCx+a(DL&)WeM@e)#J9zac=BE+nNE6huH3bauMo z$q8%t|Bk6UoVarRn@f9#soe7T!#n$F(u(>Yi2{tT`XZ9*YuEdroi5$k>_Q2EEJ(fr zjLgilXc!;=(9+=S6d+V@w_{BpdFp4*dhEN82xe zXsWY5TKZB+sk_)x)1V{4S!D$R$;W8b-|40(1$hgc@PHVNwHV$q0gSUUcjejRzXY2n zEdzsJd;G&0n^H1%oqBSa3+*3{{(;K_5&Lgi^>gf>;g>s7*Dcv;;}_Vp7mktr>6t&Q zYk#RQd{*@X>{xQ{2$uWT2m1S8sHz4Inf22O(sj>9L@I@Ew+LUsy^k=uK_$NZR-@oD z?xu}ColJC-Fk+)(2cOfh??_)`a1S#4L#jMvYjK ze+sYm$Dyw7sD?F`shsN#PZR;Wt7N0v(lF)rtrr z`rSK4?o5Dpx|yBu5CF~B!$m~>@TtIPVTlI7EO$^w_f~|3`no#VViyN9sEEWJ8J<&} zB3a^^QttvU555HJK;LI0kS?*v${(AJ`v$z64717O!fSbl%Cs1h41lo zf5WUOs#Z7aZ;_$Ph(GACoDi-}wmR+{WUfk}pt^7|Blhev`w8PMl6)1PBY!m(?wA0{ za79Ix*0RE5K9>UGI<`wj6!rC6LIDAp6MilM2XsKN7>D6gG0yLF;)9LDPn)Ofize|5 zQTeoCAv)8Npoyor2nhKUA|mx??(a&}u%oVdW>r>ERF)(E9?>u(DhiGydT#DQb`zj$ z98Z+V&piwoV#%>e@cb9MeFY5gx?42UFkotD8b9c6O*eqQENMW~GZ8!*JdaiArk{a# zN|ncOXDY8UBC20&yO_<)?v+Q_X!IZ24+3=E*93Vm(2li)gf@uYhma6Tz}beDCprW+ z^d@WGe2`D|FVMQ%5b0LtAR_io(d&_<9-{*NZkl>L8g6%|sK2B^M7N@;?4h9{UP^qh z_8qKQdbT!)-gXAHgQ85F(D)6w>J|o^R|g&54)Zhk_DN4<=t$lKYsu=?beg}~EqGX8 zqBV4e^7KKv;DA+lcF-g;4MK;a;*C^^5GBO5&}T{-8g*sd$1MU%`017#@oX%?!SZO- zT5kw950ps7{;=up`q**MQf+PR$rOR7QO+-9+8jHcPtfj#7Z3Fn6lQAJ;um>73F&4p zrHmXSZw`s=VKB0?8iFei0ixCgOic9Z8c?&9@gZfTd+@|w@c~6Y%^im0i@w@gS8*rH zR?3M6zW;<|yCp@9S+mnAA_sm#As46}a&s&vgw1#;`uh6Nz=>VAkxM+0t~*GdidlPR z0SA0G4Q%oL>x8vwD=ycFKYf-&Ry53X4SarrL+!fC?n^*}U$Z_|?!43f@DN}@JLoIf zz^g4NnmXoe?L46z)O_2juZ6!qvs|Br=WpuQl4gA(R?(u1;I*0cKb|FEvzinIw~aBN zTptp8y7rN&1R}SpDcZR?2e#Nt8>7Bbx-NhCk#ZY<@hu{w@Ap@k_|dh~cpmd#Kr@7e zU8lsCK;LxxGxeY0GMWaVvGV>sOTUPX^auMA$-0XxtM&r9Zoy4gIPai6H^3T@`M#;H z7J*#JffhAlh233rd*K2|4U2me{ZyX{3b3W=Ujq@fUBWx;i{(thlUSCs+D`+zcS%hHA&#{FXH3MEL$xZITExos4%AB1gN?b9jfGHF(Wg6d>Q@xUs5(ktC9!EXNHrumONghYWz`9VR@(>z7 zh6wGnF*fFw7zI3s*z@vh9{VD7WI`$VvS)PdADdD^HB7$C%p3&f3UHr&`}S>eDLvUD zXc(Dx8GMgsGVR)j#}hBq)B?1*yIhSV49@gubJtXz%LNHbrKvGRU0nEW=i6#c3cApg0ye_zp!FK-XVRp}wplO_c z0kf=MV0IDp{K#076|WkZ5~n+M(?>38^HaJ9@0t*7c8kwFU?BjrS{JS9(K2FL?cSJ= zflSkH5m9RN-NTU^8z8ti@;HXA2&Zf9aivj#xgZ#nelNQm$6S3E{z;T?MoH$Ls!9oAsOr<8= z@~#+tA*|~8F`OX-<5IvC?dcfa6uQSir1*0Y!{Y_Ep#n?~Ao=N#`MMrS(75-0dKVyB zXDd?5T6rzK73+)y`|4Wrw_5g?_M3H@X?J(X5I~=F78v&|judB8&)w_xrtA=|T}Y-% zs~E+nC;G|%mfbq2ag%iG25H~L&HUukObF7U;^2eQomo(!6nu`}v{!OC^pFg`_-5G* zsDOiOKsU`yPdB%69M zZt3AuH-(@DwHGEe@iGy;)%hvP^y1@CCM`1TW7Lf;&-39HbpF6Iq8N!Au?!i|bro~_ z*o3f}Z6-utrNnwhQdZXCd`Az~_^88(IrvL3pZ^PugSA1loM3t!btMc`>(F`uY!OJR>^Nfq+m{4CHGLk=HG= z9g#UTqPjV`wvDXEh1}8>%ejz8Qen6BV!l0op-SHhq-(qN(b&O3WkgtbxD=4+U^)g; z{u&JpK@G5Q&WCf+rI3P#s@w%U&vuf>dNxXC@S#HJYW*QmPfG4vJ1*38Xws@}VnRtq zPt-M5NA9Wa_}J&9Qa#DvsmmEtR6~QBir)@x(l%^Ao(^OcWO4cjj59LU8042~f@KI6cZhYvY@XddOdg z41nY@9N>%y1UC^NCg5lR5Ja!Cb!XPf-RrHuNOguiIp$}gQOV|89J$*$jD?i$u z&#Z{;3U2GlKc4zdus00mo?a5D?nuq*2mqi^81i8yd+Di`SD{96?7x0VVI0RGf5}_Z zrIP|t<#|EaX13XRK|S+dz&iWifb}il?twT|ke7ecvsAB>Taq5wd^oodov%HH@$|ji z)ftLG9)I=$)F)dUp+Y%ihuyfW*Ti|zX`SVG^p1?%pHGkU^z}o-!+m;|^!k1%-l3-l zHTqvZS_B&U0Wko(Pf}7+3taSu(xjpJeSLg~tNA|6o}QiG!>jwGli!kg@ZZ*8@Fe6* z4=lUzoCac+6KK@2LasbeXF|~j^#`H`0#OXqnURr^yK5uSv7Ck%*r0Ml6VDp%dx|Nk zsm@ysq$`E=p;5F<8E-!%Q>Enn!WUIOeoi0!p{}22ur)G3Qo&&yF=??wH`fdY2ZD2N1X3~@H%#*G*`J0xpc0zm6;YTOGLvy-d#=K|kUsCYzf_Q!Tj1E45yMVVAD={D7Ojgw5z{YKWK zoljieL<|{+c4HVF>pG44R-CGHl&p2b`^cV*;ClVQja^;Y5jBtlc9~~b^k_smjafA&g7-xRY^dS@rgT z=t7V#8oUE|xD zO27|Dlvw6xJGN+n)&pS0hBATCYa5fLPV2ZsuxiMI*zg0UmzC`V4NFQyj+_)uy zPeQ5U4!=yGXo6pHNk`BM|I8~3?&Q%A&Ce!FV1be3j{PEcH-Gr}31$Uk0ZAWF*OOrI%zC$Z{7P&LB;IV{NXzv+dXrTjFM`PYmrv=ug!CfU;Ah@rik70u)w1j0g;qtcZ zeGk>i;}a<|l4s{<^2?J9blXB#5Z$Yg7~j+}3>Dmq|1o`k(hjHk%^EpBGf9Rb!77M{ zx3VvKL39Op7}V!tvbVF)xv)7^hZYrj`}(X0)E&_!Hs;6wX_0eE)iX}9!To)B(rPAq zW_VE~Fm_#N#JZh3@p3n&D{wa%U6@|lWR(1N{?I=(N`(ME3w{Q_HR?efXoyQa>6~r{ zR{RdD&Ho$km)6Ll8W%I&eN*YDfT)8w27UCKA+r?Sg`zw34qxoi z^x*-n8+L`P@td{@cvy$cTBH6k;ugx?q_~xML90&NeQZoGE~VHg3ap%YIxc(wgiLEjAMyvWHxT=#9*POV>u`~pp3Q*-y-qk}Ovz++eSwSu6{ zpa-jZm&XM$n5&CpZ*LEGcAdA6ffgDeQc~zpu2=q&Z+GtBUry-xGG|++=2nT~)wJwv zkeA051y>9vj-%G*BtDpm+6$)$D*y9@iqXp~q5E zm!aUzR4bGep0*kN9JwA@i(OtHFlT@3k}3kzXv$Jg|GFe0ulKzA-i4VxrCAZCz9 z&0U6+Zg}aQJBZ4__}pv{J+Y~hZvT9zxXpdqlBOpvBYPdG=X!PeHeGiN?Dbl-3AtgO z@_ler68~1bG*aA;-A2zaMfbJwH5x)fRtPKDwwXEJtS}Too8@{-p4c3*cN{uM27U*V zRHABsf6{vh$w~nw((7^?Q@v*8M!n;)vx^%(0U z>{1jA6A#3Yao4dBr8*M0zlS1R2FDuV(nf~3`h%Ow{Ep^-&m{sMQuJE$mQB^@_Jqdi zy5&ioK-c%cs;xkOF8wPA8!5axT88YPc2$D%jhS<~g{rCBy^XAN-D-@8W-$OyJ`zud zw)?c7?l6s_wnkFw$b>71sB~3V>fx9mI0lQ{_mM&W<*H$anD;%pd#Y{pDLSt?ZIBGZ zj+c4@G?8iKFAyJXZOE}*QKY%L$MQd9MG`yBc)D&CUA-+~N0;XwQ~Zn=QW4yup;ejM z>PeQDWmQLmM;)|{L#FT8P?l+gGVG2)^ZzhC%Td2p`b0Q>TN6O5eK?RM)V z9sY!gMm>OFZoG%CfES2OYwkk)qV?vxx|YIVT8FapfnA7clmAri+qc+GrH{e9eCY~8 z_(w%vRIj-oUxdorjQ-1_qMsr)+*8PSxO}}tMOa@k($QU&!l-E?PR=ygiJ}aq=6^F~ zhjEQ~S`OlL>HR89Sgn7Lf+bQTF-y&4*xFCIZQEL-kGR&)|M4J0XK{k{O1?_*c7XrE zbu2FumK{-c0SC%Cy+;j8_vn9_f5xM_`2@92SmvjRte3clC=i7mlsG|ci0aUN8I@EP z*{R1t*q3|bwr|t*p3jBROvHqQmI?@?N2~k%MQhD>cgioUrn~OO|1{7kFVgSz?|56> z_tOo$Z|^UN;#zGZxmAm)ZY?~&99*U#bo3KLsSK6J4uG@v(Q+3n9i8C|gwa=c^bd<8 z)~nb_cFD**NM1L+qt=2s^> zBRQdaLN+M!{*jQ`MH~}H=m?nkL!OSoN6BJiE1Sbw={0j08S$!52a!d~q%HqbYLQfI zqr*J+b=x?LB6NHQT@c008Q^W^QW|hEZb3 zdQ)P=a%kn(U`+CNdTy_30yqTrU*q^DW8D_~NG(K1*4-+uH=<=3eUFYV1@W9_c{ySA z9QL3i&5T|tm0#UErkeMK?w3woC9Y4TQfb$o{k^UFpDGuF9>W?3T{TX*4!|<%gY~>aU{`|{fzg_<7SBjh`34gG-%x6y!0hn8N4`W zXJ^vM!iiILem$uY1lj$HYOmY3eiWbKx{d0B!oDKRXIlIn9 z^Wg#vt&@W-@W`W;QJ2Cj0+EgyjTcu%Ulv=5fAIhEr2Khy{~g#Axp8g?#{Z%d@<5Sj z)zCm9PZQE^W1fCD!{zCMtzUdgP$lGsi%zjoa%>A&~_8DRu?*!v5ZO5L;mak1-b z{J8Z*;Clc4kAM0xK=Rkz+MewmGSeR3*ZBIcR{97OfTOa>=y7_H3o`$5rVsq%On1pm zQAp8M?J4xDIt=ue-07XgMEffS^Yn3n7k~NcRldX8YpCF%Sp?Qae*bJdA|)jy25=fk zKWX5NQMD+AW)d7|7l9TwgP$!p2ruX<0D7iU-51?k(gkP0djZnU{(oK>fVLk0XM=yc z4yD(T9?!sruFHz6vyJh;y*A+cI;Na`Y>=&xIXpYan%o$0-v9m-q;hVVq>MOK%-c$GrZN^>s^Gm2|Yc(U9MKU90^szC5&K&5M6ucwV;5GHg z7sAFX{RAkK0;CEa*#ur~kp6B0o7V_zS*KIyY)kkDa3VoDUWYc^1r<+tjNnEVNt5QE z8n(yjyPfV0X#B{T8!5F-X9qek6_l;U(AyFnaoH2+z6-|aU*!c6lE-5^JedsaOrUdf zgQ{o@xNAQM#WkQS`U)Y4Hq(I83}mG*D9zxOm3#X1y{pH6dkaA?=X`7AR>3I-kW_W1 z*_Gm;enTbDWKrz;eK^V(-HQ_cFa#Kk|GZ(Z;5WW{`9n@8TTNTg7!6VZs1)JR(F;iv z2_aw%CjdYRohj=B|CmLS9jn?61azy^$mK?=^YX%gt=}Bf4as#ov;mW@a)ndw!>8uC z;5!w4{P-8&!oSU`wYBwnxGOkcGhSuW-@Df$RLw_&c?Cg@-Y4K*vg_KJ`|UJbb?eqG zw9EqpA~|Rn;7Mt3{6ImGk4*kL0i7T0e|YydCOXmotQ#UE9?haY+7rLS{s}Sp=-)Ps zp_=dU)XA@3zjidfhCf>{K3A~4O?MsVuTP)2kGvu6LDuG5!0>Qsjn4t5v%Fcfc83=oAfWM=5Ju=Y8f!?Av z=ocxh6;E8qG3uftVNq+Y4lM>{rlTH z>cl_*h4)j``w$5G_n9=9@#GtIr5JGjij%bcWQ{@vOnzVY; zL51#tFT50FHU+4|z=H#)zpwTv0A=5L=EFk{r%f$*bHh&;)#wJDS~KB2h99twN=kT9 z6;M}lXw@Jr%GZiqRFnO)<%z(e5k??#@)67&<(lU($%Y4;Q*H-SetT6}z@MYl6dQuND5+=9bb)C-*$VMQTWjG*(X?d8 zO%qd7*lRenAoYM)=6ezQ6Xu|)XcC(94Zw~CzhDiP6Hfi#5&7e&epO3rFm>3} zU44l5AeN;l*U^}&CGw!x5Q-nxC;2W=XOX*WQ# zvoO}>(*@!!!0~i8zTh0j^EzHHC-8!CU%zofOi8Iz}|{h(LQ0si|c_aKs=J zaA1bruvHQS3v_#b#8T7bkj^I5juZwBMfp`V&(pp20f3|1+cprxSzO4^`3^C`?3}R0 z&&Kxg$B!RJ_mq$<@#ve>27v-%SmUV+yz~Gl?4F;h4>xYyrTg0sG`hUJ+}zOtHX3eN z!04@xw&FwY*6La}GA-#lftP;?m7vo{*iqL>Nt+-ALTZ_Wt!?=0?PWCcx*Y3!frPLa zy=Gv>w|+BaC)l2OVPJw60WEjWMpvKc&yb*mpv}H>u+)|T69<1b`TQh(VELMH2f8tR zl!JqVWw8Gob|8-q_W~-!^Q3O%u0NOSGA1pZ1w48MLlHun&-DLnA*TPwl=W}BYw;YT Zx?F5p?8?VVc$W}DN=z1+^;paM{{R&Lek}k1 literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_199_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_199_1.png new file mode 100644 index 0000000000000000000000000000000000000000..f19e88b164f90fb08abd355a23c2b47a742e10e3 GIT binary patch literal 25198 zcmd43by!vFw>CT#MO4ZlB@9p+B&4KKq(K_#W+5Q0w4#8dAf3_;($Xpd(#;YO5b0D} z`Ww^zJLmo0^S$vq|9#iK_OeGakF*kSB zt>?$bJKs55r#)kJ&PF5BA0~>J0!WoF6u&Ddud%4f7(29CVmoB3slZgOR9In4+D8iZ zmb4qJtn4DSU-dDiJ}r1B{y_q7NA_a4+(moEn-2U9BH1~YPhM-C?vT^#Zt?Lm=}L&L zY)I0ftaab3L>2VRlMfD(LoarF5=yfoT4dq01&s+0HP(|po3b|E1# zCnr|37V=F#!r3YEmuEtJcKL88pHCUoCsE<#4Wr9H1Nfx_S**J=yvxIv~m95UV`5!P*>Vb-51&#P+8w~uRcE7-)ds_ zTzj#Sk&)s0XVP!5uP=3>KWoWtVn2!U%9X7dcCV@>;m*!Z0hiST=amtXwrse^$9v+= zW>wYI+EzpPy&K&kMFdyw%r}wg5A>!pUpIax`4DvCmbb1VxBXFWWF9{g~=6-RyIC*Mn)|GJlWT3UC4 z0;5UFG+BhkkB;_#S$*-pz{JF~=nz$19hI$*>042VXz73K93&R{|2c}*+wZkm$C7ji zhB<=r^H~OmhMYZgRaDYCIy!7B-R>rxuV~nvjjvv*7!@ZWf|Y+reG@rTcqe{Xah8*tv$q zL!(gvi8r_iho*AWiI?<6-9;-IgBo=cTm8Om_A@mnke(a(jc_fbT1*SI)c@nlrvW=8 zH9&+|HEU;a&OD0FZ;UPZcyY%`V7bI$@dT23(Ud||b#3qDNN?b7EWPE!lt*raZiTK5 z6vibcF6;K^kY0a>Yj6`$UIqw7|Cg7TB|K#Bt5JD&d9;HdmlPNBk{6t4GYm^MKX}k_ z(SaLB^uv_97b`WR4#>u6V*1~S26PLw$7{4aUPtZ)`VD%r zQ7pZSgLx7p5ZjS?^{;K~Lj{j`+_oQQq^EzKvA3N3{A#4kQU$vm5OB`P(a~akx-~pK zLpB=IpD7PcgH4g}bLJX|{u6W_ET_|9DG(t5=Gak(XB^^zQnDg8Xb zEv_4>tla0P+SXec9X+ILow09AoLwBOpA1Ppyowy9Yn+HRg6-LmkKXCZnpK z&J7bqK4*v)uT1>$MYr&958E*0LnNF%m*iH6)mmKCzKpMT;NJV?{}oSqyhdt0c`6+@ zF%#pfa~oT${BAYLbmisGX3m~+Ew|;HGXLaG<@5Wa9f!VR$n!zW!uB-gtFNJnscB9{ zMa9xmTpXpTnVECV*wh*W7nh1n<)gK=wKx&4db)+}QQL+eLENx097^KqQFC31dDf#9 zIra5=D=RA_wa-dtJL9qCLhS5{3Rx=j^z`BBTT580@!DsU0*=`wB_*4yHNRxS$36G0 z!B*J*BXIxsV^DZ{^3i72-fp*7A*@KZ;b4Y+tLS*IC?(*m!JD;RAB@FzZd$)r>!bmK zxNCoHW|8gt*%Vj~!T!9B+ulM}&5@ptPJXnW>q6?`xpU|CxbIUSM>@u`S#4bj{lx({ zcMrs*!v~F&wZm1oT1=R3(05J0M7*C|?t?I9cAnC=B6=u$+0iGT)Kb$qb^ZI=x-;(J z?AF3l(^9>|-(>-Q8QB(3bBFCtAcnpr_*XHfPm0ww?BpwGRAxW52)G#}r;)JnIlAub z>4g!ebag)k9*fTNHBM6lcbl4j_u(OY?CI^gINn+U9w+Lm(sZ)&9&y)wD~L5NWn*d! zzdZHca_Ae&*xeiY1f6uE#lD;s%=#FR_s@#Q<3 zx_FmPBG*r&agzU8g>(&}8#fEH*sI-Ejk@DMFSvuhN)L>uISF8w3B<$^QMwcs9DmBh zdheveF|n!y=T3@b*1#tF!BE3uP4os-gn7bj1!jD>2)=!UDWaT~-jU`zOV}A7kw@-V zfDOHL#kfDBmm;^&7Q3(*zj7MCv&)W8Q9@e|dC?oHPX-T2n z#pwt+f%uTX3cqmSONKa8(h-?PMf1sW*Jr7EP~2s6o=ZE&cpx16Q6Ebp8mz+&ipOa!{&4~9W z^ze;N`}l6t7DWrvt4n-eyXKZ;*QheMy0xULrt&oj2O&;^1ap0;HTPz@^BMhkl8diNj{#r5mm4kW2y94Cb&VXK^?MD*A~8Lv2Egbe|mxL zWJD51DX1}5`B^K@Uaj=n?Y0LYgT(>&(GQ5BA4n~8`r}kxt(CSOKJ%XFK=#|$S~{LG z$3>k|mygL;5RTvIW4HAN+^6tWjY&3Y+aT&z2p_yoooB!RHiC;j%ngeKeP;o|v5fRT z%!Tnz-o3_>OKOvr zVIsubJm-qU6EnU(eQMQ<@z^$;PcsDW#4?D`0(jpZBizb+a5yhP8; zdqYn>@7{hWHWt0h^~tR+9K>tZ(Ww-ccf;XrM~M9ezp>l|$LMY9ytjrNmvNBP3+QzO z)rZqx^7(P9Oug0$nWqGY@P)ZDtEV>MT?0x{C$3VW4^!~gJbp&uwOT53 z^D;v;kC|p}7mpj5Jh~x0-RK1#s_DkV-+1l{x_2{qrCMm_(&dy?3{KM_-beQ#ibct^ zVd$jcG~^ZRO{-R)HZ^}!1<>c6Edg&jLx}5Yq%qB^nhq$Br=APQCk?dk(-n9$+I-m_ zAvQ&C(6z8gI+$RE{hjbz1T5pOOKokv_hMX(I3bdH7CpAz4`tf0ffA+HvsmWRTV##C zb0U&g+Bm)68f_+Pk?ju=R^;4M)Tg=#kkry_4zx3ZEEfE zq0w0>x`{{L!^#M9_!4$7-`zc1+c{qC(*4YM>Y3abvoGx&U5ajJ5Pw4S`!_7enHV=L z#Kq!%CQOQ)Z%%f|YHqbD{d+`s=+gLgv!Qu1DBy-s?4LBTmrHe6M$dl()7 zbt}V36SdOCIDzZVc+IySUP=tG(`1Be)1%k^93nC2XCTKI^!*xnxJj{QMC9T_Cy$!{ zxb#OyUqcHit>9*SMK$+)uy&8A#My_kHHxg(L62~~A4qB88dUcWvaWMW9uWsAnq+a= zP#~$0LcAAb2(e!=#TISzUuj;fy^eTE&QERNJFi`#--rzttl_)cR{jHhic^zsRbnP8|h$;q8x2Yl{jAhtlJg}s zf`*m%UNspht7|9ZE{>Xz`wrV7hzD^=iH@#r0gu(t{O(*b)|cSQ=FdR(#rfVevhU`D zxoWP{VN#oWOGWJ*_yvzD6~fDGCxnIm{O}ohZ|HO9@wa!8%o@H-H<7oA1y*XN)_Iwv z)tw=IMJ4?@n3dmouXf=HQvdD3(-OPwvA>PhVbxP1f}2Fgzlku9AL9eiygB;p+YRYn zji{PtRxV~-M6%>f=YhV8%8kl>MO)AR0y}Gtjs}Z7)h^vq81Oo1@tWVCIIcfBoH(w- zYW4MBa;>wav#Yt(8G_hR6%N&3+M-NMPE8FXi!%Ke26gQKRQTkZTlueyZCQU8rY=hj z-M1Pl>>Dv`(JHGVY$g}2|7da0wp_jYgsKA@Z45i;o@3`q8kiw9n*8+V>%!6c^~j=m zuH;?0Yb+K@ z^R|wTW#^ifDP=RW%((dYt>w~g=P?^{-{tMEx_hLeJE=}!(h%CnEAf#FY$m49R=-z` zR9MF^E-ej?j~8~v^P*5xcXVTYecDk#?cu}g3kUe*8?&8n+h_W^y^b7KhD)WwH}mul zWzLhb;GyR-^6Xn0L{a;`sm_p#7h`km!s)vCn^|#0)6WevjzV=_pi}j*J(@kbd#+#C zWh~?MrCXS$020?9&(Clhem;ek&qN{z@3kL2;&EQqWYw?Nxf3scyqkQS^UHImW^sPL z=l5je);HtOSE!MdmF=rKSBq74(O{e--bl+oX)`R%%}mj~VGj`<2bq8MCd5{=_GvLG zyPo?eYekVmN9;skVBl7tEW2*m#O6Z(RvUBCOHNAU?Fzqk-aA9d6=M8_1M&|+5t^j8 zPI%)rpYufIiZ|X%SxPGqbcd|weJOo7HPF+?Sj4`^_7hSDgy&jfK$jG%JAhCPP< z7Nu!sQZi{*|HlsyKqMi6;Mh#83G5zx%0@jIuB^rKkbrFWgs5aH3l8}c!O*7>#B)l9 zo+ZP|;$1eSOhSxN0GclVv1K5-iz5aL%W6&IG$|zviGJ5Hz&CiunSmHT4pPIDNZb}d z67TnR^8@_lZihZb{~YFJe9By_<*~LG2(kew20uH|Psm>2F@(3lWd@wtw@uRI0sp`| zgO8^OBx*tiFc4d*oa2Y>aOEBc>9(%jhDG~}qAQ<$dhadeBSD^Wi{JR4{=5{VS4i?D zr0_L#zKncS157~9LGNc}QmV~ee68ddqF>;7;LlQb z+q&iE<1QGz4+t)1?wv6}6CE)mx;L1|Ym1uCrrx0mGH6=!8Rdvp!YSnXDab(&Rsr-x1rN-mP9U2I`d(zjAOjbV0$><3sk7%Pl35f!E8!#lc(B3O`ukSpyNLK-M&B>pe^78zOrk%K6Wr-ezItHaTVt(l%75 z28tVX7i;h0_-2yaM)T^ion5##>Z{)X(&D`i=L`i*&5d~gmiak^dFYx(zAg=VFz+F6 z_51>WuuWamKOkP2kN@cdor`@w23^@_8hGZ0QLkEs#QLie9Bhq=7qgf3&)myWzxj5eK!GjONS82Mo3{b-^j}!P4h7=uK&w_^m0q zSUQ&(47(Gl^7_>}SrkqphCy)si9g^~0zZ4xl_~7MJ$qIa9u&dnwC9uiyA1WJ9)j^s zWHY|eoN{1f5?l6Jt~5>Zr6TDqe)V1p`-_OE3+&sTymYM8ZNj`}GLBPew=Vvql5ymB zEPG)Gu%Hlq`(6)683c_>=lf?h8{GiAvRdWfg2XuWFT_E4 zx9EB_ERwO zl(B_;S!2t57fdaNc1zrA!vhbHt5g6fH?@WraCypixgF%9hLx|w;T0}(lbJ*El|$bh zZpq-WZ~jL9qF45Gm9&7~Fhe6p2b|6yF1WYHPoh(557pBc)3`1-vXIwtUcV(RZ!J8? z;}iqS`)!UcU>e1i~DaQ zBwGvr(>&Q~1Rwnuel#_roWwf~pfDJCdUKQ|!od|u;6(3F(=D7r_cR~EtWJ3aVKV9u zK|kfRSV>E#B!e&Tzs}t(%N-l7S=a`s8(M<@BX{Ml4zt%s1JWQ*^*8Y>5`8(je~tRc zAAQq9N4o$53ly3FWL0;k_gn9IBE-V5PlYlnkFp4p?q+Ejmzo&j-EWX_Y+~S|SGe`z zD-rVf{)6`Q=HM$%qt=y{pI)5Su5n(uH!3SDTmE>MlU3(~5|Pm6n}ehCEeq6lpDInY zgc6xbA{v&!2%IL6;0oY^$BBf^v@@eXK0o-{X7A)wK9sMOHka(BcjeCGo@=xR zy(h_PdFl&cQZ&6wMGb6?9A!3R)rHnTW9=R7G_WUkljZs&u7*oTOaykfEG$SRIK)$~ z-jbe+s{_gdjX0v$SlS;|+6*oamqvz%Uyp9jcy*o(WfOHrnVFaddV5n^-`+tMNf?7o5a?~|4siPiq3ktZyF*M4C2ouvM@0li@g30{i#3 z<5AizJZk%2h{i@o6|=zEDPnGX>)k&Q%^?p|J}$j&ahSRG@ISDgyRiPR9qReHW!a&M zJ4JV&e>`1@?&GntqU7d;HGinv(Y=*x`R7e{h<=^&F+Ig4*qw%ScQPd~!D4cId65a1MCNVyt z(cc+&E28l_u;Zs9EH5v&JUZCbF2|^4JDempcy6FXOOrnYNms_-A%@yqiRr6!Ux)+` z2D762&^-uhPXi@bd?B8Ji-Q9rt*<<8E@~%C zKYn{Nw2gzG0Kr!fqV&ChcM7Q9i9gM}bkbNoH|yi9lLpm0+X-sn?J3e}eVrkk76k1w z5ramCfaYT6OU$Ftl0s{6aFB_SQQ5=81H0ncpx+WoOZ1EnnYDPlG5|?Jn?p$7`AUCr zd=rJSN|Im_q<4eV_NbYf)oNL(5GrLYtyt)oesiknH0zIGR1v2k^yaTv70DCr>2Z2% z-x0z&PcnVe5)%0f-%qI=h`|f^?%7K>Ek3_ITRj_R6%pAq`1_-Q0lDw^goApTFCAB< zJvjqxN=*$jN<9x-vXxFDAKy6)mmIiw?YOuethH=K7P-%akMi1lCl6ObIvmn#l>s^I ztuNm;lX`ws#@D*mb{{%vF+WO7pFvd>o-Rf6%tZ7vf=3-Vn5tfM9I)f&rt!)Mn88PH zGS!Rp75sFaeOcpg@(Ss7k;jK<+_7AJ>f3~T6s>uzMI+)_OJ_OT5i8&@8??T$QT}Yh zj9S=Tb7#zD!u!;z9uX7(Qf=#O3j>P{{1bSAv|H$qOt@K4Yxvl3n%2z=CxJ1#QF1Bv z0uCbAm$q?lVzbF*J@>07Zenl*}`wF7ZXbnC_ql}}gY7{}^n?cOr8Cv&u!=>i3;UlA?5|sEKKMG*Kr&Ko+ z2GfLJKd%z;D)L3`2H7rk-eYe5vk1Ph5L3H^AK_pC3A3u6USTYk`Qq`>#Bp9~YO0bs z4u^M01aBLr*s-(T`*+xMfU=oPLWmj%FgUo{p@k#QHhZPEny=VeKo`xPk19^sqk6S= zql;JyhePS)^LOZ8133yuVb=1A9-g-E)7v={a$NJH69&lEd`AgX*ifGi4h*E@oFTM; zh{)nXAQ{%r!@9V6nkV4={^y~uwI)0j&bfs-P}uIJeEh(y)`&_Sc4Z*HxRw;>qv%2}M^HinRRAY(Mo5BX#J#=; zz6LLO;sc_cr>)~a)Q(Wu2IQHMFh%X_y;-PI2Q5smJ|F$5lgS2q3PsA>op=@$slgmUpQBA%U7t^)4F)S(y_d4B)8xV?K2 zXSlIeD-nXL1Pdba{Eb6@^Crj}6rfqh%F4Re$m%Noqn$SW)329tgH;41L|(D?+Na!LuMQfbA+H}4Zj zD;ry7-3>-d6`T|)(_u{AZ=2_$KT<~#$q<#nIgMpzsyHL(%Q_jo$w80UY&m%G5d4cm zm|8Bgjvn;vp;0@3xHEypHXJ#1#_=$t_xU=>MF90lL}GJ*N7zj)brD~du2ee(;9B$+ zyR=7>wA1!g5iv2@#|PcVs&;n8M{Bglx6!YPN|b6qV*}{I&jB76P(IQ`FFZsQ(wxCn z9lQ5Guh`b6m?u5EZ)$295FD(gtQ?l9kjRwP&riIIY?eX^Reg@^djEW_N!|L^2QNM%S=CrQrgMlp!|wzysquSp8> zvKig1kWgjd+lG~U&OlJ~HKTtLMvLHkw?~=AcMh%Gx%uVHb=ha~ftV|TAKtUe1Ofj5 zJ(g*rQ)qpw9NZfIT)gHPLhxzlNjTmq+4xTfJBv%_JaTBy$p`c5@z%L!(RAQuGhLQ~Q*Y2BUb)&^> zDyb=jzflWw{AD~!oo{3rP&EbkW-fxe%+{arCI;F?xs3y#TDbq+X1ZpEgQ5NaIG?hF zY!-%c%RFtTp)D{3MglPU!3HxEDpCEG<@wT`J4=^yv1&yEr%B02@B)u%R z#xkH;(}q?>;?Pu+yk=`IdMhtZ?M*zLUn1>>g;}wtYDDtqAaZ#8+Q5^_^O~kYm1CKw&y!lLjFgjEvqK0dF>NnCU z@N{ZHXN!ft3>1+b?G+ts2L}hI1f7+P1Qz(RaU}-2HjL(_CQs^CY}2~%fxa#*WV`3 zkGuUE%qyyA@@u@7+?}^<@v~x{OeOE`vz<8Zo&Dg4Qjhe15t2882CT8aT==x3i z;T30(jrfUWm8i|lm%=NwjNfg9or%t*+$Fs45r6R#n@4YwA-RBhZ+1e^(;0L9yXSoy z4SzXLwA)oB*k@H0x{55UKPyVFd`3_3$6=e^N`L!sv$*x^10_tVN=sf``7+n!n)XMl z@c*FM|3}5wX;i%V)D7Yt!1!D;y{BSm0dt%BKK5s{$wm7f?E4A-Q2OJF)$ZfW3)XtlI7bN{ts~JM z#|Pgo&tzHb7=~gnacob8cyNlO|3k6pYLB0MZK$9CZtLn}2L?cZQaZXq z6DkEgkagEO*Nslz8k+{_o~Wv;ljg>V4)ED$iYfTXH5N;kBsj>P>_5sG%bB)F z?Rxhrcj4sK>LoV$1U_YE_52GLl(ulSCk2@;`?r+$V+;@1nz~vN|I8B9jENPIIkyLIXhgnu)1aII#cuO@U2KJtYWNlAw{p5#ThSZu+K{80s@OuW;Q}L-{0P z{5G>06oo`%+FCeWGZHrSI+;P`Z-+qZ-3xXwa$`Glw!-a@P?lH&45&IfQ^ zTlcxn0m)>RxZW$u!e`|*nP{9nzpsB%nd9aR)qjV)D}oSn%4+F1REMd>jr}I2zGk`a z>sjJ%=k$2Y2&U-MR~d10bBUp~ z9@i1%&m{=QDK9>4kZsST*F?dn}sn|F< z*|qEx;D80ejJ6SU!8D&=d3-1X!A|%WAiL*pz~T&F zh)KXvMa+e0{B8Z?si69gc7JUBloJZ_{h^b6U9KwY-pV*awmlhe^K$wUNJaz%VxbG| z8VKP$rlg<1zG);f`hqQFf~vKDL?!}*27vPYHR0wj2JVzCJ=TrCPb5A2Sy-oc{u(2+ zK1SNnCGfx!#41*c!kb`e^AynSU%Sy)w(l<`miY$9TMu&4h0nBji;XyHls^^TFh`x= z@8ME8#|7P<92K?BzY7Gx&~>!uz*D)a8q=+1y7~P}7)6OT3pBlq(Iq^9C5YG7)I5ZC z(aaM=+Cp()xj_Lm;*matW|c7X$g^{8RUp>I30{*@KM&8jC`YlP4*r+hU5=*A>x8h) z$gkDHH=UT?1iN7Xv8UoV1th}DmN?9#Py7J+!*hY3ze5>r6yQ95&@rnx3A%Wd&9v^c$`dn{y za~7b@g)XCFi`f`CYCh#6a1f~HuNpzJYZGjIO{{SVpaCYvJX9Yc z3H<=#%nI6E<#=p2yvqv;i-^eCzLpHYX!DHdo8!Z3Da+i^kO$!|%AesbLvR-XCj|ml zJt&YxC82L)I4>t37@~Fu9=56kHh!-+=(ZLo-3s3q=;F_xh0FMk59?m5>CCOV^x-|x zAv608J&EsU(H;~Hpd7=@wEiON{XFh8K9#d}P?B6d*GUFSi2dD_%=bpLxXtC495($t%*)gA?_)XSul2D0i+&&3 z*|U^rdRP6m83GPI$RjFU)G!uiFM#omqdw#F1#0v*8Ha_#I&&1H8G>uRd_EN^54;lW zJ#IvkQ7>re93hzGE{&pnSVMquFQ2>^a&yP5jm3i50zD_7|9cTaufbT)hB3I$$73qHT&L*bgl?O-5 zi-lbf7A%3+q8|G91E6FWAJfeN`kgYW6a5SXmXyY_Uh2lh2YqU+v zG^Wj{c%G|&T}^p3w@+p6?{Z(2;Kx|VN>}|fUWo(3iDNjwI2KB%&Djw9zo!7P;wC?K zd9bUoyxQ3)B!N&}(WeSJs5UVPSwd1)Y7QZm)5woY@KAK9BO8*(bq9FhT>`^-JBaDM zGRj|rk0gWH-h>Lr)x{+P6*!vZwlDK}(=kGep=zphYc)xRr|c=3!wjFpY7FKKL!lAt zG^N_bbfcCfj!W*-yA*laM6fmF*cxgf(hdg${5~Tt+XY z^}T~J8%3EX7@4p~k7U}M4yW%nweVM0@KZYv^LL&!Kw&&ic0ps7^_g65tH)6nh1Ep8 zpHFY?me-Z-mF!T2UnqiRtI&>XY$i!njZO0Ho3hTgY^f1kqAnr?XWqv{4JZw#{()<7 zf^5w6y*VJILUZ^(QRE}Y8V$UmQEj`D+w=819YMPSw;@$9B3>LK(KhGat5Li&{|pCyEDcYT-xT34{@#m9qkX`3ou{vc>96F5*cC|b-h+Ngcwu$e@216KUHa@0>U z@L$9-B$qCw(aT0X2!HYEo#$SE^5Q?bmH!lIa}nO5(pluopKyH$qEQ!ihKH{9Iz_aj z*|sW1;*X2Ao%ojjs!o95(MDRYpI z(b86ryMN@D7wvO>wRY2 z28L!MZxMgkJ9J(9g=Ubt?Ow~evYGc$>@3wk5BE2C7GyTW^}7s`$l`Wh9rZ_J^`xoJ zyxp2M|AZ-bbhGC%G35u$Dd@aBzgL}^c^`I->~T>+!OGsq)S9EUt+n;w#6jl-13mpc zMWL+-D#F>cd(*C7wi&gFCqQ~UvCJS^RHZVd%FTW6*J^H0`6YYRH}TvC-D7W`rPMUm z_8@yqCJna#9`^n6zYs!>ZKl9?RgSI%wZ%eP#_9JDF|?m1=(0MH=rW!I=3xQsG8o&-0SJM)R*OIKIa9^K&T3{u`+&8r z+n!`ZxodB{1#^`^GWH?RlCytUS;$IW-E#wi31s9M=PN>?mjldEO3-Qin9fnfc93s? z&;|O?n}O14oY13()__Jo2)#CDl4Ev!nKGPWf` zo16-CB0T2q6_~DatI_97sAPu@ndgl#eMjr@K1&=_Jgf=)1Ir8_y_vqs_k9UND~a?m zDGZu$4t={Dy*+_F(sNz%2cc-3_VgV=LXICdgJm>outgp%OOfH1^DdyGlat=UQ*e7l zwDJAA?dF((RZKo9>9tQkj6uZ?sleCd=KSGUq<~&4J_T}})|6uYR>P>X%<2GNCk={k zK5|2cVB_WtN>b4-3*}uhXY4?(UxCk^@$2TEw|JH&>~_d7Y<-|<_@s0+U)ym751eSz zug=`5BzJiE0n%etn}<_3O_XbO^o@WvK?(5o+w)^&oviITrQ|_I%Xh}Ot^GkXw0V>y z)5U?~uc`Cn%qC1vVYmA*plHb6oaRRr{5UO3Y;ljayKiGzaL+?!&@D%s%1S@u>IG!8 z*%zxxe6xTP#9AOj*z7%;+5*1mAMWSsvirlBjqhu&cJ0cCo}bWqHr$aCU4;d%xN6v=58U$*Y{dwK#gt>?;nmo0VI1{8oj` zVAzQ%x7aOcl2Hs8oLw=WhS3DjmnmMH0=}-Vg_QEx=nTIDtzgw`{__!_%m!|5%x|@2q`D|Iv=*P zyUd%AZ~+;r4~9WT`KWf!U_uMz2meU3MZZw!zbUjmp5Y!YP(Pg45G>z#-vWF7L>m3H z=kX*XjA_VI#d;3zYv)HXcQt(hrXXg?2*Y= z&!ax7<=QUx2-Yz1Sm{)R|1>Ui^#V95RChPQTRge&=9lIann`TyT>NgB$ZT+Ivcw^Ur$U!OSnq6Hl4+A2W zM-g{3ZVG_wNEU2PYkDjNsyKDNe0;d=^VZe1@dVn!YB|aA0G2~D^83@-;T6)cJLWON z%HG*zQ|}krD}#IWzw2cCWOXgLU?;~8<>0ut(h^rsk4`~c1Uk8>cvw~~sLa?#;R7~CUPcNH|R)Y@>` zp)X<{N_kbfzSaAQaz#+D2n2S8@iA%SeypdYPyA3M_dTKWX>UjI>rC0wEX6J$I9#+3 z#6MMnx<8ZEBZ?M~F+v^K&LHTaP!l?|I5aFV6EKs{X>cd7DK#~PvHF>rCzfxve-Ms< zSO6^RidbV$b!8|M0O8a9{CW5<-{z}ElXg;PLOB4nU=BnnQw~}#Z*(SM><`G&Z*|?V z8s0h`6nN^hl6$_ydGlE-RlSmJn!Nr|LFSlHP@{nxME}@0vl;IaVg*e%v(R=3Xx=v8 z9z&UxoDplxiGqKP%EX5-&@^0(NL~O-IBOdfkyn@_k(?|Nu#n@;VGh*sT^2Ro?ynl4 z5W))I%_wy$fSpH$;N$}bVmX6L4L=l}Bx*b>x6+$KAN2Pmqz6`bt+(V?VNU31tH;S! zcGBDg37zdK19q~X0D>p+=@@egpSZu#Xz{$wK0FKcxHmMz*haQ3%{tZ2zjHgw=4TTt zr26X@e;yH$I`fbvR-c6-VzBL&B;xAF)&4PyyWBY!2CMGR-D0NyWC>Wcj@Y=;@@U|0 zFhx|Zi?iDNTJ&~q9U1ePlipymZ0|vqb|>%N+xw=F5JhPrbaaxx&2h7#EbzItv{lp{ znO#vlwycR<2mAYzw&-9jpgWcdL{oqRr}$ZaT&!G2FRuu%^j0T~ZB<<)fZYFqfOc;>WMB{_z8Px{ zJxy%A{{5`$JbPAqB30J$!4{3j`(OFPVTdH2_6b}e2o6v?m}WME+@TdD7~*tb7H1mx z)L!i}y3X!ojC|OllQqHlaJ`Kaw+<`=WgCEm(9-li<&P=UtWo#037-vW?_>T&jb+zo zSGh{gElBFk(X1v%KF}t_D&Sr%_DVt}BI>dgY#q)=o6<$*%1||oQ9e9hg!YL0-XitX z^!ww5gD{b&o;2IzPtbW-rgQ$35vKUszJ~KyTWYnS#?02VSSCVpIq=TaYQuDhYnz{= zZnz#~2<`3lLGz0LxZ-Kf=|ERZa}3LHtTF_lgVjJWC|Uc6u#O!+sXu;G6UJQCV;LudU-C{Z+Zy}>Sg7{ zZ`oS8z^D)w!t-Ysity?46-iK1vC0y5nPOmD#CezV%BG;!EgTasusL5OGi_;5E`rP{ zN8O%ReX5O?wV-g3g7FcEVOQmXD3UVOcI1>H*B`svg;HwsHR5}h1c?C|fWjsq2w7`9 z(*ZrbjD<=~=nq*xUEt=Pl)37}0E#khrm+|+?HcGu5+>U8^ess-^^zFF<{B~p=i=qE zLpS`G)Oy_WzEkjJGXswEmRjS`ov^7{xx=dYUepR(SZCp8X_`1#hq8rW7oJrn@k8cEHp&;-; zN#i81qi5(4JbV40)DRjW*H0em1=<;)jgTyAII4#6m};jbI%ycU-`lQiC^+6ZKGM(r zX-Mp9*nR<8&W;nTO(piqp!|8@*km!8p8~?*h2Rx)Rgc0>*l#(|`~RC|o9<}kUnZ5B^ns9I|Hr?YInitvr$W<>ff5 z!^)Fy>{Y)aX^LtN_tq0qzVr62%V-GW9v&uyZ3|Im&mC;+$A)454%}8R9tt0hyon*s zGu@D~Y;>z5t2NdzzP-&a@f<3q-sqikucwF!_w4bZdqdFpVZK7K+uSW< zzKgWvmV0$tqe19#-* z{Of*B>0%n{@0w@t1eeu@|Dk#H$w6ZDEX{UySeva7NY zG1{4>TIbPq)Vwc#ywL9UiY~;b)pfnppj^9JsEQJPu0TsmD;KTV{#C(^+3Y;MOZNE8 zUL)u*QKJMXnrmSJgQ$q)`)}kmHc7EXz^~TTm5N7}mXG9uqPC;xLww zv6UpPyVG0&q_t#MjIl2$BqA=;rd2wtXMdJ3#7)(G(9Kfx&rk#wt*+4dTy#_gR43?c zjdUMOX`@n?#i2*GYVcD9QSACOxqRd=$G_(zRg>dqkO zP5e5z*z!}5dtKF35Xy4ys;o!MnPpO6rG!x=9Gs@54J)^W-C@-izA0D{0oK5KOf*rs z)P9)iG24eJHKW$Z6Mo=1SYwikc%2h^At!TtVG7s3`xtN28*ms_SxF}P-0E)a?@9TG z&)gA1Ql<`!4Hcw=&=0GvTHXc+@Vu{eU)9oL4YL!bJ!5{;LXvPp`y^hSBH65qvpVK@ zO#ra>yi|m#WUxUG;OX4y_gXw|I_h6%S8%;W<_@OFX9vEJ3vJZ+jCho7w+Ws=<@b`o zpea+seoN6Z<|3Ih`E%O)Iq0=FV<~pqI{mSi*ng#Q;3&T+3CdQ|8qEb>j|V1S8y&)- z*fV&8lCq;2mm-s6pVV{Erf}fEFE#}E&k>#4`o3fuVE2vNiX~6Nei}tO5?eO!%UA{ahGcGUkz!5Ov93~2eveq6F{ARYcrsod zbEkxM+l1j?PmkU0+i`rHQvv$|DDf72`Xx!+>FXoC>-XV8jA$B9(dzI7GzSwHZiM=W zv19cepNk%id0a&WET&asOIF3hGc#PVT0%PD2|M*{o?9w z8d}Mcy|#m0Sdf6UnCIf6u?SO=uxK6nHqPtehU9xQOMd7OWYO5|kYRS@Wr5;32Z2`< zX!Aody*(CFVEf$> zyX^bEy?#}T$ji$b_pTX5B;{i>i4?7%LXc-STf2aSK_#|WEzOTc`5&zgFsaTBg*3oI z9hhgiFbTDn$mmlgElW0!$;@7qRS_=LW&vHK3K%&I>)Zm!!uZOjiK~W)Se@MW`FqqR z2KEjY=Qo#ZMFdfu9c<=1eN7g0BoQ?|u<$`euTn*MYanXHi70%pd0ov`@bXUKRu0nF zzpy=2=cEb>rmb?pX2$MB1IQ8cE-*Sl0nwuEBA{WE7l%_lwCRH8% zmXbN70Oid<->S+Fm8H~2Ea+Tw@onb}dV$EKi_9+e2`_xFDXdA}Y2*An;&&!*(@eUG zRIVGFU#x{b(;FB$$X+n!+}fC+8E6pJNNv#LRJbH%7DH{N1Y{DE;MhA|sk|jB~n-w8j0Sw zZA2Z!FDT1m6BC}b%&qY?c>-E=?qb!=!4FhmOs2teF8gSzNeK0j9X1BlaORuS6i>Do z+VQR&y@Oa0spq-#0U855=HeCFbw@NSS4^5|5M_*sO-cQJM(o}MZ|FSK#g@lwu47(S z#V|$_M=gK!ZhC8L*=esofv{#7=i82=2Jcd$F3E`VJ8g)5jir~iGDZVSK1Dc=D`{OI~_H>B)!Nprg$j z)F0PL=OfKJdxh%k%U3YZe7NI#h$a#*3=Qq=lOO5j*t=9=$hvN`OWAsFzN{iz9je|N zJPwaZxy8Czt4iGsUSo=VMY)FeUA#HrWThqYAqcE}wZSO}il?Dt8~)Ve@uy0t^^Wld z!72vCb}x9sbgy&7Ai(utCd?y<@|f8e7F`iW4VVbJuqITj&00MbcY%o}E*HKldc3Fe zdR12`8^A+H$esCK83QJWa<}w{v)1F(J)Bzkz^0(qK`x{K{;GF4#`k#oPNw16&Iiwf z&o&?==}R4M@EbsG}GO=^Xnm;p};SghM@y$Q4mQWmuvSJx>q zwB-b}#!}$c~=EM)EZGMQs2xcLJEt8|=q^dU^h$3DNbmCKK11GodSx`WfN z;#St=mDr`-~C%bgb=k3Vmn&lWDk78vv1<;g;|2Y`Aa%h>eSp7p){^!NW?`>sk z5!C_!bD#!MyUTsAsqvv2v@l>MOa9oR0lMakd(LXfRw`;9@wrRg$2)TJc?Zwm(J{gh zmY71gS#aya-k!^Fe0PL)L?Hsmg1qK%bH>AUQ$x^cg7sdtnRv)>G4B7<$eG7Oxwdh9 zI+Z%b!KsKWov5fp2W4+{%DzP?Mhn?8maH{Hs3^-}>=Mb^49b!s|P zEMuMJy>93H^DdwFbN;an<9Y7qzOVbbuj}{wf|!7>Cok66jU#P5n)YKCIB+3^FI1|D56dR^tlu z^{*e2(+2ybH$G_-J)5t%v*c)&G1e6N<_svs`##`Jo!8vo{^^UQPxozlVa^^6Xq~o! zXkP_39XJLjCJnv%j(_uOz`dEhkhy(}TghYcx;>9k*_5*|_PzKzxuiy^olpZPsPs#6 z>+L?R3d#X&l`p@Aa9S0DbTtL z#f5~eQ>NMbzlJK;gQ=on{|?_*GjAHUN2sUOnPu<)t)I=7t$8|JV9#5O0$R-+qz&)4 zX)aPk*G5k%@qK{PF)B`^Y3)0?UnF+;$R&)@aS%F#1JCt0uRK{OW6?L<*x%7JXP1TahR)vP=WTrcv!1ibs+`pNk_K^#wTQyK3yjZ%AJAWA0x2u6$|srMq0} z)xFLio}Mkctg9ft`2=W5QOOuW+k%W%fVrSIznB0ncHN%{r&-$)B{8dRjoQ(|;Z<@nVr{dw)D|y~&BFz$wf;$02 zc?EI|3U4@7TvN-^tzV7Y!K*d^zqeDvJ-G)q^_JZ?$X3)Ef;G!X5*c}5tlx>%S<68e z!OZ+-6nC;XGAMY;re?8I7(bF4TkhU>rdLfNmL(Vjr7kJW^Yka8&m}rQnbv6RGxoYR zC!2b>G*9Hpipmnj}TYZD8-|ntmhhD0U7Vkos|&g>Zb@ zSAVes-4|5xFLt?NFuM$G&BSlZ^G=E1?#Q9j2F3eLVWD=>fUc)#1E&YY=JxQ=>l|(!F^Ds$%pep%`C`Ns zw1TV2;~}zXl2;tj(unbj69wMYcS<=gx0F@{OnMRq*M$f%Cqj4pRBfljyS*ZYyvMIE zua71Xsv_FEr;4`L^vhh0GBOy6ZfDD%EcasS)l^oSqZ+i^o-3%#Pfq?0N9{8jN6*bP z z_=F+pfa9S?K&kx(lXj-HBN4LZ!bsy@rQIg2B{ zNc2{P`~*Y|k_T}ukOJ+~Y|DaXNKuv-ci)-hw&K6oAJyFSRPlVE*1T1p5VpxVX_Z7G zkt&=%iX*k{y~s!!0vXzrIk~hKnAt7ZRJbN0kqlxCu2x`unVn-uN4`K=(C)o4RbyN3 z9QQEb%a7ViL&Q}>VfK8rL(A_#tK8`<=e&K>x~SP;LGex7@BDD<^HOc_w^+L9@0H3e ztp1qh5HKkCE;jdf4uAmrV}9QA)l+d#r+o0|H%&g5$E{vkd>Mkx|I>gA$UaR8PNCk) zUjV?>cMD=v77rj=X{NsyBOPlqRx9wy6C054Sz7kzLGx!djDPZ<59NSUADJnka|zYL z-Bp*s{=SV_E`RW=2~QHvx$iHk;O+hmuy`i^Z`SG<6U(cTE&LC8&+>Z{LN;>Pc30x) zTTHZQzAISvTkKrnf%8Rc7tR}=zl)94t}Q#=#+nQ7^{|ZU-IL$@WL@&fIwCPa=<2x> z?onR`CGY%ElU@m>B;weqi1!mOj-;c=6Z@OLN&)H=Wl?e4B200hI5 z-ag&11wUdkugaVL6B~Gq^h&4pCgt<`6$p{V!+!AoI22H*kLB8OznuR{1 zEI=`h9%|Wu=v+|uFXH4cEACu#Np{;+?donqE(c(XA-8`1OXKVh4M zF7HhJP&Iu2kHD&c3i!9*G;~8yMX8sO9)2CMO%jBmV+O z|^FfUrC8YI9pm; z*0YypqkTr2#fplG>OnFTEps{s6`Vc+g4YXlN;}|Yl$#%@UoNk#+(E|%*LW?@xzvr! zieb4k$V^BNDAwvhxcmdXjXuv3oXJlm5i&DO?AA~yo*ep}{YLZOP|DviA z;mmUz1-VSHAt$D$%Ig6owFFRGd2IGgn5PQ%uTm}xp8$|38eR~14SA}mcn_V)&621w0i=esunwiOf4+7G+Yc(RZ(#U*WWU<$=vkJb#-rRXlSSe&Ro%JZOt_R zOsXGy(F+m>#47gh@w$sV;?q*vfAj?m3iSGQUu$e`PJ+n|(_pKpC^V&I8ok3wg@tk5 z6M%CDt|{y#d-jNH5svMpw6S!d2_x=EK16mLurS;@D{OqYVN6ciOLeKtz zx(~3iUi(vKB@y^ZPfsuYAVC|D(Gp=ouq@H*bAN6z8ST27*#Z8Utj0!tf(Tzrp+)}I zjqB#l1+&hrlE2uq)P*Ldu%r4Lu{&NgFt)x8m;G#=G6;6R<(cGKGJ0yQu*eh{&4;#c z7u5P5^88wN9z2q7L+^>A*GAOBn@fT@Zo@-E&M<%MDyM@~zULXHW+Q*u1@U~az-#vC_* zUx2N}Zec(zyc2n*AU=V%uHe}-Gaz`CS|1Q)9|+Kria8!J@#)z~oy-zuvPhj$LUMA0 zTPvpV-MbbTy@aHsJ$hARdFji;tO(^MVUPcb@N#MpPuOFWmIHmI>v;aT>V8fP%&)7| zhaLT}ie15NgI2q2%q_9TrY05&RAprraIuV1$NLa`rDEzZ%P?1^IpDu;Lr6F+I%7%rq8{Qw`K$h!|P<|<)lYl;AU)ke* zzw23cr!rMb27ZKIwyW&*p=wKnt?}cq=x)@9+DP5#Sk|>V&=Afb>})qxlF5EW#dGfNWWaW` z_@0MYFPBOLatQY- z``)$U8-#xvjxf6e0;NGMX95TNyjE_0{sNP?0TT|ER_Od{PQEyaUP??#%7gVdw=@mP zSpVY#_Rrl@JzfrrRFs!H_g48X_bpK9BO`x-%0f0iN1Af{La)A_9!)WLsYyq>6X14p zLqrp#V)$02psH#K^NemnnAy_hei(9?m$8v+c9w#hdS@ua50{~Lar%aaXZ7_Pb-Sk^ z`@5ePv-0&wZDE|u=_ijLKW4g@cep0DjkN=b->BEKPC&GWhmzzGodtEy_q#{=J9vDU waE1HRV@(gxyu5z#Fh6|C+5w~e4?ppaI<4t86)M;#4S#{r(=|R-e8M60U$ntLOaK4? literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_201_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_201_1.png new file mode 100644 index 0000000000000000000000000000000000000000..13fd31349a36b9b40f6d70753bd35a54c52c5129 GIT binary patch literal 97061 zcmbTe1ymf*+U`390>J|W*8ss?g9i!j?(XjHKDfI(!GgO5*TLQ0-StfVdw<{l&ROTK zb?;patDEkcma6Kix1Znp0F;#$L4?DB0{{SsVxmIw0KkVz0083X6Et{B-`eOVc){WL zQ`J$y#@NwC&)x_irRQjCY2#>VrcdN-Wba^RW6ewl{-Gf%l!Jl<-Pd|zNwx6N0)g3 zzkkBk83xN|f5B=2yW$A}v+zV6zdYKyFJTHR-vmNim(O^8Xb@r3Wfw?E-Q0EdckrC| zxmpSP@6Iy0YEdJF^Z7qR{quq2Fyn{$_j=&{SDZ?JqEFzDKrGS!uhq)lfr05#1@aFc zJ~$Tfcs|4s@VHM@YH=BlrTuhwZ&^Dh`LDzJuUW5mNM-YSUEJP|IkyxSlRNB^1~Dn*vjP98n2Jj4+>iMD z16Mj$@SOfprL#Kdj)9huEk2fvZ zb{!cO&5!>aDGV->)93AKArytA$>W|5K&?_~csNz)crf8tP(TWegjaHZ9G3gx-wj`x z$>YJcy}sON1A1`j3q$8{z0l2S3o&hi`=`Jc^rJs+n4!@>5zznLYr>!=x2gX28Gl(J z(A5d6^xO>&QOR2-->PZ*Q`IBKkbM?yVXBod9xwx zz|BC=i>^GPlad?jXw;(glNf*In&UZ!B0-ZhsU$Po7gy|%xgu4$oMSE7z2%UH{3wU~ z+RSomF;%0+cK*wOU?$k!OPz>W@w3}Na`<#ZzO>XlSNuvL0rYA}hqf{=>8X0+b4x9YP>+GfU}+$l`Y7P@PV^xEeZg$VO`{M_ZUkXeV1DSX`AHAP~CC{$6heMGw1vi0}N{?TU@N zk()+^a{VEf7ZF_Hhfa|W%dC#90?Q<&XMXFZyK`UR{I9+ADYV+Sz{pU3szfHsO;IZL z61cAQPM!VMzV)Nb{g-grxojWFU*b}3T+DmTExP78gYwrbG)cQAgrBqHq7>n3ms?~P zyZu=b({@eHuU&q80YpkHGnT6TG?ah!$wT>}zVy*3{){L-9|@W0_J&SWYFs|wbPi1| zXbF3{lanfC#$dT_8pl~!IhW3KrbO^X^!1cW_ONsPRd6u}qKHJ8NRC3(=hAKHZs~`l5JC|~QG_~ouzaFKd|bkxMaWLO zmt_O@{aB^S0RDY`(H%HNzYfy(CX6_zYrvouj_!9|xXWoipKHxqVG1oinA8taiWFk& z75+l6Q6E#$8+J)|I-@!JS$TQ0KEh*6fQP-6)m3Z;UBLbMIyn)M?=PjYt+TVJ)>f`e zE|(&uGUalOMmhs+6B82vczm>Vb~Zj+X;#yI-s}na&c)?$Fp(Sms55oBGM}wSz|7%v z=jqqhSzV8BCypOxuwd~;w8lE@%KVlgvlXRun%pf8wySF?Lby!$ zVqmZy&w5X+O^2es%_(`T^@#UW7H0pe)nom=d{o_yfN@FARVb3XNnDD$?sr!lv2@Hh z1{18K3nGM8Yv-rgW^f#dXwK#F(L%;aYDQINp8olkqs=ETdJ`7 zkPYU37u~YtX$MbO3)^IJE;C6xUYw0o`i(4Z?2Qo~`xKog4FP7}+Bo{O#A~bL*^ie! zP&RS>m&@PWcVkmDbhkEtzSMB^M&SEWegC}emtQk8IIP_D+k$?}ID%x(;M1=zK*W=I za$(&Ax#`w+XeKxJHhmRN#opXrgw|&F@Z?C!f;r9Cx9*$ksrF{))L%x2+B|JRP$Bhkd4CbyFf`vy|p2(@&ozs>=3r%z#zus zalUc+_=F<;6PbA|#YO}V7&bkjfU8kXXL#Rsb}yi~l1e^$~ZDk7;p}!2? zBx{x?8f1^*Rd|MAFg&x3Lyb>eFy_|Q;wfFsW2@JgourQLox$tzju%~OP6+JFmlJ_9 zm~gYbt9R~WIN9i8}zEokF9c`q$R7dP0!@Zj2(}8IICg;N@{&bV?>xZ;;u~hK=r7}<54}Ky-WbkWFUcBnPRL5@e>b)yr=!8~RdLZVv=sR& zqhn`3gj}M1FWlMk{9y#O;qQD=F7Z?=QWBtTqd<r#B5QBdmPklBKHpS3-y zj{vj-QXqU$)g-xt7EJuMph;Bb6KXo(gut=Xl6@`&d#{`jMax24e z$?#dKNV{vKLoeYsLy;Nw>3L8ig$3wpoD~8~wiscAb5@8d_H%rKu%|c|7`&LE_v8HxnBRDB$sqm|4QvvHvnYPMXaFUs{?o z8DqcsA$ZJb{x#&+WH}k=jQiWcEt``)2#D8MYDX(<(khkcL^;wd*qN_fCMF}X-@kaw z;yAq;vK~^dPdoD_GZ|3m0vO~&?u|tJLy|W`hI>~E;DZ>wzNBKN?d=;x7&+c zkTvu=b^yO@kIEYr&8E09oJbuQ9K5x;>34GrBOCyYsM_L4d1MKV6A~1ZJ88nZ4bC4Re=NmVq+AM)5jNdLCFeDhG}cl_&TZBq7MCij z{Vk)9PqoUqq&zhji=2(L&VF4v!m|wNEy$qcLk-fdY`g2jo?kU3bmmjM`x$2Y^-E7) z3M(VOO;Y!>wTN}?^YuX^ZpoBl+spG|`pX_!-POVP>^bNlQN-LLn)79htq=)W&|Du* zxL+*sw)FlI~&4V^Ns7EH$ImyqB5eo^OC^Nssr zj4AXJ+`z&L6o~>!rb5y$M|3BxV`le9*C~bRg~lToz}t7*q+sjFMfY1=LniD-(E3h) zs)hS5Y?M@g_b9`5b$ja9%hE&Af+7^;f6GA~mRoh{H z&)?(jX#`%cP|V+9un={S%+Zeqb*IiVMwG|^rR_YwezQMkbJ8G`mQtb`Y)5QgnDnowS4T+%9K)Soa!D42H)nHv=HM18d#kQO<#MMyhUU`e%>iNN} z)a*p-mIokX&6DePpTynr29ALEdypbUWu~jD>Feu$;@^@*lX2n!V_~uWwy*Aj2VN?<;AY)pb13X%TW0`!}ol2FB zcStd$?TMLRC+vA<4GQg*%xY~4AQQ2E-YJ{m_Rqr%!5FE>(;Yts`&*cKAvJOQ%y>Wm+ z_^1vVhDtajjE&#Sj#=Uwqy^D%qw*g27d!W0c8B$JKxea#W0WlZb z(FM;>`t^1s$~s;@mtLPuuX4yAPC-LHJ9Sc8JYGTFtk1PJ;5&GLw)*wb^Gx&#>rHt0d#f=~I*y1v zQ1Gfpmnl+>+5bASdPhK)Et{p$XowhyCN#MzGA=e6Nh(*Xg|1g01EP_GBU(uC4p8(# zkoE{AkFB)=E&!65M1Y`uG{tz&R7&2F+f0m-^IGjoZ7wLPG>{K|XL(I0rgFb@sIinp zP%fb&;ZjxKD4dg*Z6pXQjI2^L0Xl_lA&=){iK2dsE<9|R$?oBW@6ojT1<-=_)FTZC zCDUvXi%UHg6x~GORH&TwOqth%Mzp_|B|k^m;vquD4+W0+MK%gYi8}))b!P8>M)DxTM$E z)qSN~wiY|xxMfQcXMR*!AmiXy?h{02!AJpj1I@Qj=EEi#k*Qt4er|h$@Fcz zvFFK@rxN_4w%Z*5!x=dbUbEXBPGt8hkS~(SmKG9%;&M3~y}Xr_k}6fLCT6!@*|^#r z3DQsb!T;zrD`RPQZ5*D?uH%v^0!-WeO;?D&uUf2QL6F&z%N!jW1RHP<+S?pD#>@Eq z6@*g(IJ~zeqE>({6|@5+&~n9lU+X8O1^(OyW;6RgxaEtfo+6f(_JVLTizAhNm#KZW zFz3}?wm8GlGWQKM%J7T}>qmC0!1qK<2YO>?XD~PFn7hX1{)xh4az7}12 zG^zw&oQX1_VL&=N?n>L@loeg}3`MjQXo(#r#QQ{^s-E#l{Znos?>}5!X|l zvfGnY=4f|my2GsLN6}U%#AXMh0_i2I;T$i_smiy+LSb{OfWx|G(NP+JESpXE@H&i? zCQXfugu#qUiL+*zJTO;-T^{sa>5}DD=G`@I#gt_oe|g54dBvJ(N3#p%Yteb9$>pVF zLLlK`L1vW_oSKLYkZNHYc(3NWoV0sb@!-#oTC#IuYL4KvI=ulA|^-IUTR;}U9)@rDQv*0AqdVoh}FRkehPO5;k2a1_o@4=DT;e+PH z#Ht`p(f8LgLdyBd+!9S}HM>)1qm@EXEE)haEC;m7M<8qJhR(~k_MGoP+Y z73!UYg;Q=$#I9_fv~C(*W|1YG(t8yf#YRt##DdbszZ+R9;5Ruxg^hD7-ziJ3ZlqRA zu^umFBv&<-I=^yGJl}p4#b`*XJDiTX4!OnE&UJT0Qxh{Gbza&|$OGaYy3m`HioKY_ z0Y~~{9|~c>{a;Z1=y+{bbjtcYsLdXqVQw!$#U^m+{czdlTKn&Fpvr>d)bWCjPFOKe zFP+J`u(0r3uLrt63O`(uZ2RT$8jsU4G!mEH81Y*&v)7X&nRJHe{n_eYo%4>B%kiK~ z1v~@I{!)q^>C0>fpc3nIoHQqb*ky!@63})rhu_I8JRrlG6N*NzP{7Yi(YP<0>X+I= z`BqLv^ZPxxVJ)xZm@oL*We90arg6@dU9KmZF(Yyz(6KkS_J)O2x-5opm)FAkN4K|$ zcqk)%l*^_aW00i+(qQodlgykRp|(^xEtopwI{diYWsYa<7FS$Cfivsj15i?YM@y%Y z{)PWs^6#NL)qs-+981f3K_BDl3_ec9?K0x91c4ukHI7dHWItc19)g~nkUKbcCszis zbc};c*t+FOnJkUNS~p^*pbXK@&q&gf)*@h~V%qBQ6I+v;LZ=TdP57dGvlc#;$x@W_ zjh4XiQ1(}TfmZBx<}=aHw@eY|A{B2N+-WJMsB^SEj{O3sr>_1wvA>OBYwX8UBCbiD zuWY9XSH2#05sYF~K_tRC8A4r37(>L=-D^k+Oa0MbWhaV{f0`#Ocu9M%?gCV!Fj`(| z4$O=($58JEK76dzyTSnsHzgR1UfH>}NbOD5!{9C1`aV8m;R1kwVyyT|&NC7(nhdmId!mc8aN#QtFwF=H1OU6XK2inRZ!Z|4rFI z23>AnH8!<=7MoBcKpB6fl&p?5XoOIzl7!yLmot9`FZxY3zljwsjr08(w^KqMkFvP4 zJLlRoW_YqUFxK84Mm{8tVi}3p4J5#V;#ctYMgpxh5Fivr#v3-Xy`)EOrV^EJ`o<9) zfay<#2!60iO>WlD-*SObWX_NFG#E4!YsRz;F)809Vz{^r&JM+PE;g}RTS@902Xf35 z5YjrZwNY7(;KazKb4nCp(``B+QX>jhp5gK^f-CAmGko8een*@GtOk~+m#^9k zw~L)s?sDjEDsvUQYa}yiXpyokA3d* z{NJ?Xo7*v_Yff*x(LaTSYa9>d&sJMQH0cJ0hvDey=|{)Lu)o_9nSip!^^_3vVHdNDeP&Qxnv;Jy8TMB~lk%pG*qGjsHmsXb@&m2!-EjoCqcu zEn5!vM(mxv_>)yC^n8(*jPwVeIGvj@$JZM8L*Un|rUH%X}Z`d{E6ER6W(y8MyfpAMr185qm`^uMk}K<+Eu|MgmaE0>S(;*-=79(ufOCYCH{9a_f)P7R0keN^N z(s?uATK+zX%3x}TD=O=e3e3jV;z+qh9kvsl^OAmrg{+q`eF3z69%0+g@XTp}No~hk zapll_yeJ%StG`^!eg9Ded=@M&2U2m!k064!jquWuS}kR?{is5xeT-h=_u6u8CX1U7 zq^PjlCXsn|$IvCSINDE8za5{%^GFPChV#qk0r7ey{JB4Nul<{jR2_UV-QOH4c* z_FT$?O8tl>#-=B*k}^94!73LMFmMsyTy}8?&Uc-lU4NjDyr8JewuWIZe7wGzh3&8I zr>Q=d_k7bam>BorzSHXI zg_!k<^u2bDK*_c$;Xg?ch6+#)oW~O47FUr_rfcFuI~8cXBjrqfKEWI{u0re;>eV&Z z8!qWHpipXNp%kOc;JLsZ=;SGKA&AFts!nIZ0?P9b8xZ=p&*w-_j6rVHSu#<|yo9c} zi9QD^hV`s4V$C~15WTIi{z*EY;F2%~E5Ujj5&RShaGr#ZHERx54I_tyw_3OiX18CS z6d$>d-$5&Z-(Zre*BofULK|4|aI1jEx+U}zs{FBg?;j|oBOpI#kjNimja7I20Y~;@ zC$ZhNI7{{9lG*8sT*G0gklBx?YjRTSC)BBF`$VxpognX$uG`~zI~Nzwl}f4nH#Vy! zA+Y`ci;N5^LHzV72OOmn(x(4^03|n*_PqDECMg4w(wRK(lWWsc{c&y<#!OxhmhLR4 z1Js>A`aw9ER#x?#N|yvY+1^$mQ&p$3Z6Y9;ocRd?)g{%bvc4ZrbvTmX0%psFg=?t@ zoQWY&+_IdPCLMt&TFOQK;94HRCuEC_VWi%9Y(XY@9kG$0KkyG_w(cr zhY%u|fKYsn*HgP598u|dsMQvd!4sJo{92;@QJ2Y7wcjy-?ew>1Jwa+krZ#qsfsw%w zriBT!*?}Y{>stn(Q$;eBhi!QY$El3bYR1yNj^Ic19ApZcJsUv6q+;0B`p?E;K|#S5 zkGJ1iDlwXT%KZqU@?8mNbJ7YN82njUKL>}D;u0pK(LRh#{BC!M7JZ>)ff5I#qVM878S^8ZbLF z)Rt0>ka>Zz)fVJIolq#TF>OY}tJSZk3Hn%jCv`I% zm)cyv_Xt^Pt^Wd(I56MAt50cBy1%UO=zOD+cOZXyG zE;9yH$Zyb@9($ZB9n&S(`&FS7H;D6`PDi#67n>2C)~hX>Cnph!iRiGfu%qehRb4fS zR(pT`LW3Jbh>fSLbz3`|a)n}XsqMjtDj;$sG|@#Aa*=b=;Fp@{IEQnY&WU;hjiq|? zr}I9y0{-h3?xx$&WL^jW-pVaAeoNzQsR9N1cNJQhcgFU|%|@GttK*;Fzcj*ZJXbuY zaF`R|L_)(;JB-pu0)ca#E`8O_@FcP zl17hAXv$OoFLlQZr>%{Xe~+(Nq8{~@?{WQF#iiQdrPp8Mez}nBq+>}&tF)|eadBsW zC-EJu4Xd@%naL33x)+YcJKiurSJdu_`8)*Qq+fsu#*G(5lyoS~ zejlhsEKs%Yb=4O%0IWW2_9ak?a0@TVq(N%ad3&{sOo&+IMNan+iMJ!Yt?A8R=vezs@Ud-&ZDN= zEW#}<5`&>7)wmKf!#n0u=gdS+=A_l63`GqqnJItG|GYgxIU+i(rrMoBK}a*^Q7+Zy z{_5<>_pq;OWDx#zTP?nrhd(58mk*4Fj12CrC<8+V&iUr6O!Z7YpPxdxHciIPuTM^F zHfy6TPR9%k3^`h@&JT}|b~gu;ABGGz4FxSX(lhDVC{P6*3=kwEYi54J1YX~qR74D% z?%ix(+W6#)KYzd1PiJ+a(OPPJqd31s4$;30gbFs~MnZ(^nOXUQrm#)pkyRhZPyuIa zO`z@8CFqpYAj%#x9(`Sc7N>Vlyk!$wl=zwOFOUi%H1@}HCc60^s>vi<2)Ouzo^V2}5 zY$+@Dv?2~`Jd)q)MTHf9-Lop z8T{_C(q0v0C*)@|oW8wDYFKxSm}BlBRy5zpc-l=r8~dM7K?eHP{&GmoN6m94&$~7y znw@Ya$S=@c@ftFwYwh2dnh^vI1aX{A0~6a5wcU24oePZArOmoG2S+7i(hSGYYHXRS zLqJEPgboMnii~y%T%GCmn0GOIm_je-D!1T~){0T%5A!aZ&wxEnt0f+JCm--!P$$z2 z9(sKHay{BU6Xu6Lm>O}@Y*%v2!OTp6>ypC|!vLH>mj$>6uKDH=g&-lRqu zzZb!Z;STtXNfjL=fmMdN;?)7P^Es?UA69|Q4mDRO=-9hr`bnd1dBfjRM z_L14G3*kkFzKkdRUcSwc5qyjkIR)@a1A#i|w9JL$V(c9}X$|SE8qLefw7dF+f>y*< zUTHBTRLn8ud57$LJdTZ5s$K%HrfrxyK(7!F#?u76luG_v*&T~T&{0P{QMauQtlBs~ zDu&+N)QWavD(7rB-9GV{Cv1h}(jICRBW&Mmk+N$2Y!M%8F76&ZlZ|ihQP;CD@A`wr z?vM!kBdy1M@NmKL_=dG;2B&|L^*Oc9!#V1?>%CDj=DM1ySW`ttN6&Wb?L_01#Gzw4 zivImBJTqtVV)l}CbYXAo^tw@0Mgrvg<`X~d{MMn9U^Q_66CHhITKLs<&5}qSUZFh1 zd74@xN%Oqgx;2;_+azS>rYy&qzI2QbQU zkI?gbDgE-=9!5iP($2KzQFACN)q+F-!^*GDamMVh_w5|P_%!`<^L0<>;9cfabc@w7 zq^_lL#5&1(gc%2Iu`jr+!)RHq0aC`&;v5y9Arz$f*O)1p7D^*yxcqzr$@9+Ww0;=uq ztRYDF0;P&mtu8i4^Hnm<_BOmCnDp8aAt5kh>Fg$8+2v%RI$$WC{P@=UXWoDlSU4&u zEM#;tziYSJ?vLyNfueQ1AHS{FzH}UsH#E7`_Kl@iwT|ZtjoGUvPeo;yO;@o-ORrC) z4h;SY?0z=6zSwa>l3uuYubG7V?bH_)0bd^Mi1%}LN$H`W&Kckx?We`zB7P64maoig zHQ5_MFb9Hm+FcCT4}=bqD;A?IqDk?UBoic#o#+Umo_~9K4q#fuFHtkd&~PS9$cSa7!F}g5z{yUv3G-1$~zvAlbgn@^XQGCrRdO0-xThZ~X7+opA(b z)e|2J9PI!t$R~fYi{GbGMJnbPoMJ_Uku9S6wcPGxTgx?%6gNlk!&AunV(NF*tLkYibJB zz`$Us-eLgUHf4&LN(<^{B} zd4gtw2wrUWa9PY*8!Ok0dg3bFO~iqryH`6EEQ@R?rSj5kHWbO^=JUx;^;0v?=1|}V zrn8ogPw1ymt?3n0l9fiNX6Z)aIrW~*Oq5oI;#jJhj@wk{sjW8mo!;(Ay=T1zrF?3{ zL=J;-wAQDtK{G*}byN$|Z4|2mPzI(F#hA z_-z=CCTloB1SHohiG8psm+4e!fL>XbvO2=w3tl`fz&s+aFR)vlB{9%_L&L6_wrU`yujq z+|9I03aw^++1I$=xG)qoZ#iXT+M4}03v|19(XMbIev9H@6PJL^zS}sx<7g*nJ0oz9K zif<36$w#*TPL@2z0{;RA6C=eQm2Vo5tmI(xg=VKW&$gYZDFxVdf*qaB?dI?22LZ;M zkU1(<+T~iU%I6tD*B1X4=w%PUew0w~$by26EhaCIS}c=Y1WuR2o|Z_W5wo>rpw~8~ z7vOc+L!hIhbMMpv4^02Eg?!egHsr(D*x2aW8H)dlgv`v$Jnt4w6w8z&QAmV0z=?$6 z;qZS)tf>Mqus#e9Zoy3iTstTKkE@2$ypu_)$QvOfqz?`~5Rq!92(&Xr(0yf`V zsw856#Q&S`20_ zQPr5Vf1NpmJehEu0u4*_9-81UzNk$={J3eta0L8=kf8Wo3I8VuSy%e;A6)W(2S)$c zPX0|;pmIh38=U+M!uq#~h3HdI_QAhvhJxyC+d%VY&wml5&w?NSMVp{rUQnQny^5db*0#K|74Magnphr#D&LM^!>pjIK{SzK1(R_X%?c%^^WNn!%8?a zdi{h22eB5jDo-;t4&m@wt)m^pSH3)xH(K5ynh3W=_;BfZWvqB9cCMvh`I}n$Rh@)Y z5!Gk^V>FqKaC}fKS$C2e2t_(`j(iEj#ok|yJHoLT~D>&z`=nz4AZ4RKd3)2C>T<KsmZ!FltEV&5$MvQr#pi&F$n#gn+pl^9A$v)RIrREcTiF4;Bq{Z#S_Q&%h|S*SdfZlJ|QY+rD2 zm64hG2o9HOByDDYp@ZFbU=A%?Z>|KkxPVDaF?bSjb#?trK#;}gQ>I*D@9E`x-uX@d zP7iHyJoE=!UijW$9K2tW0Wm3Rx-6DBKwQkv=R8nOs-J^-LQFKMohYL8s^o}CupKBQ$D@SkjaN4%+6IZgP`_IS@k>_*na;g|=qF#6K^$h1)<7M&qqO$Fk z^SCogNbS2PF7V|eKbzcaKAUA5SXf7>fc6$ra2@vUnD5@)LMpafuBJ!53knKr_oLIB z$BnG5#vjlL%CXuLU@Q7jw+Kf8Rs5VzM!%B!yo6NRT4hY)j-mWLArG0AT1SIN@ulVU znb4La2Fl9FSv6*gz;>qpSWZsV(|dX}WdBU1)cf!b|6@5B^(#ZDR7vs*NMp*=<7kUE z(!wYv_-OyqP2JA4)7w9nCZ&C$}*yJ=AQ)dnRwG^~vK} zk3Ns6NJY^;L;TIiymegLC|5drlD&`!8NVVXCS$UH?qZ)Qp>OPWS#n5Po+Yv}!(Sp8 zz;ML;mzg+zMBut5wmx&+DnkB=j_0cf0yC5JhM(#uXy7d*;}+k+x}V&P4R~}j2Ah1; z8U0xEI}O##E(0H6rk5>zrL_)!86AAjqrPqd83rE<`FlYga?~ROoR-;hRyU$(4!N=q zt+-`zs16y*!^bTS@_$>ZV1PND=I&5N?DgTgB?G@Vf5g@J8#8u}(BW1@CNmuHW^CU+ zGI-W(cP&ouRD`+d);2;a@{qScBWv%=m2w9%x-7+)(NyM;hldBt;^pP#zg89`Lf%nu zXhEq~LrOqEaDEHUpkCkW@`F5CZj2@oiTdrZFFjYT)&rKZ-n@)?2Ft+yk!vWY(7OQo z-4nS4MD=3~DntgH&7W7=t2S%HY^$PA?i^_cq1+?0Ic{16EAG7#*#H32*;6Xg>hjTF z>XA{FRQ#T@_Hwbz?)i92j2E^HdGR{uHHycI94QQPe9mu&dVJc=~#X+2U*| z(W)%<_j#!`my2vl2Z>$5LKE4}o8h^$c%9=6fYQ-^?@}C2UV9KR)_5q zC!7}6Icm_ib=zOx)n%uKp?tGg#qGEFb(3}-hu9-flloa`^H+;B5*#*0#_QJ%BSQ`5 zXE7rz6{V{XILuaCD1FwOMy~fe04EC_AzAG!>l~mR%KT68;b0KK_hI9}RAxZB(sRzs zG|Ra4gis9K!$Djf2Xdm;+)kPscD!bVt}~(_FbG9TV)4U7ULEbovdY*rUUF7Po;a+A z_>#Ftr^j5?ATS-|ur~hQ6i^TRnms#jJyI{u#gwPyR3Hz_RQhFKpAAab(1;^ zp8F-1)9<^KtfynWgveTgX(e8n(99LWIMg4xKg_Md$+SKC;lR!pKNN~DL@GqF)Ds1T zU_w}L%<5{lPuX2{-?@I5QBdu{oXMH(L7y_EPa~Fk*ZDC2tTTN^`@P|n0haL+94I=w z+HC&G<+IO3gabJ%Rk`64ta?lOMmxX0wmpBcYBjn}MZm(sVenX$WTRWi*mWzF44e7^ z^vi{?55g<-iy?RP!H#FBEMBiC?MTq? z-wfa|MNDjInw*V|4eWPBqNXm>+#ipcSt4*UGzHrx6tPeZ&=^?8564}wnYB3i9&bXVdGgElFKL%lErPn`>6Rq*#opwbt z8llUPk9B@Cd@~#jR!n;}S1hfG3V`**?(Kb>zVvuQ^*v{M7rm74qr3wu2#*DS(=bz; zjOH2%<>6;nHKDu1c?h2b}ByG5e@OSc91 zjR4&1kidWCWn`+|(I>#mi%CCcQ!Vd@ARP8w8gy>$W5(>OttBat4=<0L(szDxSEL+A zZa~^G`yCS?D-ujxU*{G-J&fVx zNmD>(!d+~-6;@uyTm;{s%j5)e1y6GkCGjzN=k*~kEpxt+5{;v1AesY#> z=oRXV%SgyD%Cb-VUzDA7R8?=k?NJ0p1QbvjM7pILlx~pj?(UFM5D=uhk?!u6ZjkQo zZZ^H|!~2W(J?H$+9rymV8H~*y?6uc^p0&Pn&d=;!xlpO6{ttmjy3d&Rnfx}T$Fx&o z0PX7&CcBSf2+xGKkFu(zxAi4e<+*m+BO!UBNWt%JFBQNIpYg~D0=qN*DMzgqkCdX} zx2}e5ID{bouDL0g^lDm;$82{Hxw&1X!IQ+sT7<&9n2s?Ke?Ho_!Xic|H}CV_5mzy- zCXOcG;^5*`hr=o`A~upJe~4Ug1V?5>QlGL#+o$O!iD88d9YB%lR~%!NV|C3<+Yd+< z8G0NZBVG>)rYjmHUf;?{8Tx)BYj8RIU8CQR^cvG(hr5!IIU}-?g0`b#x5U6jcnli} z;-y!E^|B{q5T&>wPodhT(EvTSVl!8yxzJ(6asBAovxWm zCvuc3{|AYStMg@j6EWw#1VlqVJv4e&S>23&L(b-XZ`Y&~j-ix>JFj6*spa8RmNqrT zp$PSz@1X@<_GX;>^xW$MrE)tmFXB!UGT{cP_s7$L^sZA$8GO@t4v2ap5F6BCc(2AtbpE?LSIyYojNC!yeOrC zPu2}Jufm@8)-Cv?GO-5{u2usni6s^tDXWgo7GFb4$*c3GR0>q7kHr z%=&McnO_qS5P%S>Uu`YBjg3l|ijaVSwodB&E06mtBU+6*fu3-Z?a5*SwJLL+-#+VPBdPlC%}3iQ8EJ}BmOC@9MWACkIgkoiB=!y6XFSfs&stK zL!t5b3qEYU?ZFsak_#AUkG_%Y-_p4=mr+XQc9t@h!E}fp)9v*_=+J&Z5WqScwKB^;Sl`Z(*DBR)5eQ;=DD9g}L{5cOaWMJ1c$I`0{QLG0)AkSSh+O z-$OfZmN=_UQrA#PnsUs(GZ$zoHv5xb&9aNP`t=N~I>wNaV;k`geixY8Y4Ti<@o)tH zk+ASl{X1bXv>jEbL&16QL}(c!(_)RFVPE=(t;$dV6NQDawmB+_SEZ8tqBGVb)SMOL zmSVlx5>IQn3lz)VSkl)yDy{KBDdVG5h`(0oCiK;f$LkDGb|&&gr^_;7FgY3sx`OYb z_uZnSH-JeYoGONQuxb1V4d(#9YYC&!f(gF?67VK-zif<&)+AlSGIho#fmKtGFzkbk z9e>i~D5vn1?wkvjuU3uYJZ{6u_cu|z zoKh=YC+&9kD)H^)aA)CEY)j0X7Z|~lM6<+7#QtcPaAj^t6U7;$b02Q+N;`20Nc)Ir zkt*Ta;7Bx*!u^}7uSJ1(WuFDks+yvIe@E~t{t-<6|FkXajU+GB z@H~L!uvCEtWVYCOhfbzEFrhHLGNT3cJol8O}bRlw+WJYQ*ZRJ!tMWjKkrEx1LRsT{pP zKLEF1zDPKaXXI?O+|OSF3qjk=O#IoIy_=hx@eGqE?d5R0U_^x-T&LcqmwGS<)Ax+x z7$Pc#N9lxe-&2&{XJe$jOH*Mpq3Elo<4cW1Tf*t)3Za9^k)My7rZV~sJfq3H7djn# z%moXLFLEWsjI}Wr^fqBj%-SFl8e1(WM!;*LXt%qg!9qQ)`WBjU8~O&nFWo;Or%aC$ z)hAKE`6Yhv;un>32DW1Cw)2T(_6eAV>kTGbaXHfz&0qmil7j=p^tSNKut_O&nO411qEp+GU+rfq2|W^NUsj zxiU+ut3x@jW^`DX&FV@#MrGrf#umP&NyN#1d*TFEBT!xb|Io-a;cG`S9$!ETpE@WSwhs81qjKm9+O49 zIE+n~4OkG^R+Xbst-X|-#@GSgcW*fDh|tRl+$ndZG@mk2Uv5jLmkEiwHq^89E6_@s zGx@7Ewd;>O2KxkYXuG6%SXAzp<7d(+#DvnY4-cD zHcG$Z!Q-I#Ggr?D+qEurgU8>4zU$;~ekQfN5Gay=xq{KZ(PQ?Ad|x*DRbSWgK8sJ+ zB)+Y4+egF0o5u=u0b9RC6ob^^e-+nDKV0{$=uw4Vy!A%R`I{SIEQXQ+`G+2a}D%+_p3M~va&80OK|x{g?;6L zS1Y{teJZGuQZWu2UghPCAq2dn0AK^O9@~q(=|B?LZ3eaf!eEg-JvL)UnCfond|P2Tz%}lhbRErtnmV`vfE+1y82~tDuOk#X~$&4d21na>D^_cepN{AArB z=$dXYW^%ynYGn2LveZ$J>QkCPv=EiC--}yDyrqRk*F(3$kxcP&Go?YKOu*z}baNhr zSfKp@?D9;1mNYShvO=cy2UH87t!~%;N_Kqu=0W|I_pju}f6)~>Jlz8G|1GCLkA3qo zDgRPb5dKrWfWmzJOLutrABv2C@z{!^@*i+6FPSY-rg-ROk~!nFMuR7KgQ?`-^1DQ% z9%Qxg%lnfD!dd(eqmRNbZI^@R9&#-u(o}A%v$7fC58BEzmE}GK3g$xa`)AP<55kIR zg3v<{=yeJ%IHJUsaj|s8W?Y{wni?!M(fHxMK{1}|rTY(@J0E*Jf`xxDVSxo!NJ1iT;gIBSjY{)R{|fKz5v(5yf#G74JD2MPwO*Z2 zx=w#AeIO>i7nq940IUClNC&y`r~fnFA}Gjnw`K-laDW5HF_e&ypl4~Byy^x_FBu=b z;gwZYA3w0>q>{OM03YU2jVF|+v$K=QY~sb2;dEi68=gb)zx1K|Rmc6A4)DL(Cll~` zaG{Y(blWrmo50@Q9s~Zv)DB2!fnE3a_fs=740@d}@VT5n_&k3d%V-q5!2WMRs5={f z@*_X|aG@G&(&pm$?DLsa7^?`8g}DG)ZtkRfrQ$N9A$)apb!EUx=!>S&1D>4E>*#+I z-rl7g9Nt^;Kek-~k{^gTFyj09`s!L)r5sNU4hDm*YGq~R@dv7*Uo0}4hKP0`gE;P;BS$vdp!gW-63uH`fEUS(MOu%p)hRn@1M7Z zGIO;e-lUv;bQtuWktvc!)c1Y@1H*H{Jo%T*;Lh8F(q&r$FTIhWl&3uL6fS@O_|jD(FLeZp$=6Q4)jPp4#0 zbTbWaY(`#T!=G<=zVgOodnB!1_vIFK%}Wx)3mc5_%-O$Kgldd&Mu`5Xo07x|6v%zM zUW(GPU?xMR-+Z4KElIlWPY}k5Vv-UnW)sT^zxsLRhwsXe29fy{a%1^3f-d1&Hmd)P zZKhbLx;@6p$!X+rdvn;ly0_Qg=n5(0f^H2H0ivJ?xLe+uDkUm69C$}hPrrCi06$^K zq{{5YD^#>@%?J`Qe7WoMyWamy8Uq7j&vTK%5aqVgtmO6^KUpF_b*Q3}i8Kjy-3@=B287CA;tT`RZ;s;OytwJ)V*7c-|XD z3)3GG1cY*qFA4{Yug0HAg{b!}yh(1q_8Za8I%s;Ptx|-U1R2bb$i8+-<~Pw3Dr8Qt z;W9^RW{DH>(t4&~1i4?TlXhjvoL+XpWLMebkkjFm`s(Ojc@I21;5giX*Sho@A@@5s zP$Ek8+Rm%}h;L~CbYuUVJ1(-tKyYqCVR*0Xg( z2pG(-mPuuFwcAym2L?7b7*}8`Ck$JkKLbauZ+LPy(-3)=E}NqqJV%JkI&Hrouz8M2 z)j)BBjV7r-AK}Nb32%5sOv{ucPMTac^$kMGo@#)KUzSR8wY8N`nLUCVM^Hz0A6wvV zt@cEd|AGq-_5(*U;JceuP2RYvP$1@4A0Id%^D#cG;d3=M59{{U))wS*5WgStFj$B$ z4(1cquO3z#mlOS9T=rJ53WxQr1GfYV!99j*M|}NDn^`}$l$S90hdlh-mO(r@Mg2yS z&L!tha)%Lq4qH#};v{dL{9Ks=^er{I@1A!aL;nh!1o74kHhnTkRFz+yK47fv)f-R4 zwqZ?)SVMd9VX0`8?acUkfjKyZl}x+u(@6U2$Mof%v;8LomyAkr3zG{e9;s$sRTXCA zXNx~hzZjhEv@d=ZDwBQ*(^4;`GuycBIvJ}-B|g<+_fs)lRGe(58fVJhl$^o+8(0!D z;t*4--M=4?ALyYpaq`y|)pb4llIJ~hWDHGdxLg-Y-n2H>4BAqo6LD_zyxO}Ck}hK> zZO9+|-GZ8ZQCT}B4gEpRK=rU7NQ3YJ1iXX~?nF%APZq8@DA~}0s@7K}d>i+9o!`d4 zbC|!)pLR6V6fVRSsL7MDjFujc7irn-4X-ttQSWf1e=^B&Fs42CfQFCAef}cPXUWd& zM7T1MFjSdcKPZz|e^B;WDM=&g1P~@w3647Vt4ZXO=3a}^mc=pU3ZBf9Wn95MetEY_ z-(O!bmRr6v=^xAAjOw&fh>Q@2o?A+Yn#p&Q9hxpGI?g#HoIdgA=zkKxLtkQJ2LO+$ zZLOm|-(O5CNDTum*m{69g&r^gzz!@5?x8v>j*hIuBO)G!ie&~8@FpYU*-pA%(UNwU zi#!YLc22}Z_aB+V^vNA|os5gIY5hHsIN_Q#dNoNRpE8qF5j%H|ii1&2NPje7lIpo2 zeXm&LG>*ltEbF=qLmRyKjWkAy8v9K-3CVk|RU0wwC$@Ha`f(=>`JFiu^ypP@qCc9iQXtnNN7ApTm~Uw?`Ib#9Kh4;&@Qxkc&Hn*w&!+85xB zv*6+Uv@Su7 zTI+Yh`O|6{%#m!QHP=P6S{)M~zrXu(e!iJF&oyZ1-gtY8KU|CxYS>|sL*KwX6ub>J zXfKIH>WO18@VxD>6?zwgdbNpXd7;_5cyL;3sgfk^yLoT-sb15bcd|^&b9z!#U)Jls zq3iIjX>IaAOQt}Onev4L2pY`w1XuiIy>U~uW@BJP%OF)@^r4=U*jp~=IPJo{MpBT# zT^~Hmn)BfXANMQhjY8w;GXmizR&@d{=P$rZBx^EJZMx9FJ~1)zKq@;pNM5Ha2Ul|= zH8y%;k>nqN!-?JP(z>%Ej?sw0upd)hLj(MT3;>tez+Ha@N$FWmNEOQTIONFX6Dl=S zxUxI@99%ol@@pg#Ta(-FHPtDdd;^t|lmjXP(+t`pV#yMYgDgP=$BT_?ygO`_#6)x@RXF4z|gH7(0U&N{NOUGKc1NiWLi@uyrtF6 z9{qgJ{O)?VFVLhjSTRGOBO*C6+#vaz(yOU)>I4f7U2F7vvLd4w_ntOeba|3H*DbZ` zT&^;6m5x=le?$c{~sii4QqBdu;5sStRdmM#J;2f{lcA?InGFRe9VUiQNMYW-&9h+onq? z&Vt9}JVjJJO&{Nl3U{A{aLBpJ zwA0_bdhw|d%(sbt-UtM#VI+hSt^`ICNioL6PW zh#8sbQsQJ6o<{G@7Qb~TK}R_wLZK&(U6sHz$mM+WA^W(_beytC-37883D9UqYdvUh z=rkpO`DLochDbVvXMJNJ;qv;r%w|;x0D??rt7rf=P494_5f?c5q)!($uh5!n1|Kj! ziCbI*v#RUwsrF$$_&h3fak;iia`76RA1ED!PAP~v=7sdjJo=TrZ;EDf)A4k(u6~6~ zs_J}0U(HI&Dc;Yb&^EToqxl3@?R4iX2-mJr1H)z&OL)gwzhR;LaB#_UtxOZx+`H|n z>5Db+(?^0~&~etpE_=)lHP9~lNcK2b1x$0E{Ho&DSh3`pq!QP`IPId5FjM+fQh^48 z466b1{rin(xKHa&{If4~v3&~RF79StdXPI|MEgq+Y+Nhx+LYW;BUhz#5JIll#;E-S zDfW33Ydy`}%Grh1{QR`=*UWLO2CI%YLW1o|i!|^fCU2DDz7`^IWV%W`<932H?wl!L zhR7*3U;Df78&Soxy+<2HG@^af;ndEs!8Ua1I%2?ss_e zT3iO48>qQY;!+=P5*A(JH5DkaeEEjxsVATQE+D)EretSTXf6)o*|_#}Xi>d{Bt~bZ zjz_c_e>-V2{R%K61|awkIb3Z$J&>Tl!Q^zLW{3*}vkn04B?PximxC@ScpUAUDV~E~ zaEJ`arza=6;Eoo|Mao?vmf++OZ1bJ@K@FltLYZPwC;QOw{|Th zXbO%Jf@z@4$2jlvHQ_fE^lBDn7`p&WbTLuL(nM^3v@fhXqkZ1JCBkL)gv&@L z^v$cja#0+c1s)`SS*bFs_d{G0da{kw>x#k4jTX!`xMXN;InmN60T^Rh$>m= zuD{4@Thv>=8lI2!#cGOll7+=W)>T0e&fOC(IYLgUN5q2dwuU<_9OmQr`FeMJSLf%~ zWyOl|(OSqxi_HbazM72ypIvcl>)dJgaWds-)BMhIGzv^kz5Nqxrs?4T3@BXdJ*3O3 zz2GwzWg&LQ_ynVE!K0(AowtMHej`gEQ9yp>3(+LC4mr3SM9I{kn>DT*Aku48ce|@K z)F@;+Z^j<TEhs2BIyq^bn@jxHw9R}XsdsuhuHNZz1c=dK*GMwh`UN)Rt?_&e&-*K~ zrlzI>RP!v1;>7dZ?LC>oa{W%(@!k4^`*u>D z3wUWm5uMiRFt*W|ZYo+?-_Vet6sZoOyeeyr$(PN$hd2$;wJ-o)ez&5@fE#J9lgFC5~ zrnn8zv`>r<%aZ+iQ& z#)Lb%&es{o4uwmb3FGCi)6#faL z{|6lZpIpk{g#LouJmXs?`w!HMbI5-T8BJ|CssA)clb;EM>^=>xA8Y%0BNIv`^Yk*b z!t4)De)=mRes2HMtt@xGcP&3}-UBz%S5y^D5lmh)^dL3X4}}v)BV=DPN0m_Vaj`)+v(>6v8#+2Qj!;o_YK)vAq~EM@BOaiS{Q6C69ei=^f@C@uF} z5}G8jlqtq$X|u@ZaUdB{ZmRJ^y77icp;y#-tLu~^L8gPOQ7LF{B$A;|NLMx!9c=S_ zkYp0)O@DQx&jkO2U}LDDHjy^_FUX1}_lX`PaDceEat=oIg}L$oTPTPLx#sUiVLqaY$DcaeeO26&E1 z+OE*(@l#@qFKT`hZpS_t@bK_+jW?mMh_X>JFj6wJJfTCeOg8$RdWd$6Vv*Kq`r8}@ zTGijm48#p2s5Exa?zueUEa&~UqBa|(F?Z>Bh#vgWg%qc%N-bk=cbD%?oN@bxxJiLO zIv3rrI#0sD^)ztXhQj0pr*o2v1*Kd^3tusA+}#z^_j98|lMn>tH~oj|Pvu|*CnN~z zsW}}YF4ye!Wf96bC(_CyUiV2atoV8=3u!2;qgG`J;bM-nOx8-oPj*=Lz9bfdqX=<- zmRt$%z<1H*(3L4C;<6(=Hkbkvsv5iU-ipgNsy9taf^@^P8MSd!q=@m+n5jZS)9BMQ z8OkHqwZ4t*UC~~{6f@p#q{(|c(C+X6ChIUG0$VFRBJpH z5mykLuEFVE%W2aJyPmcDqcz|ftL&tbkmM1JgZ#7Hu4-?I;50!xeUd~%8o2QvoEaMT zaNTzlhjoQJ49=a}=wfg@A#Td6wWPD_XCb~tNO5(aDN_`N@rBo)v!`POXPI*E+t%IR zI5^Ty``5jPnBIf#33THSvkxC&1_uXk&nLB1lK(V06B84cXtz9`sj>how8qwKH63_( zTtTn~Zgh14bO!5GP*`|%XQvlTySmL#1pC{@D(^87bp0i9l3X8xU8P_qG83fHyjR?*yiB)8KM0A7Q}8AqS-GyTB}KN zF|^fokL7;pF|#=pkFz>n&JO(iiM4lujh+iSyER3(OyvX=Cfj>gt6AXN6)_ttXF1@E zkD}EwkcFRd=_6yS2!#|;RBFG^`=)QnlF~jO%vfrwfSAuxX_vIS$n`fkC*|-k3Qx|= z4!#hxZ-SmGZu0ZHWjg*Xmh=><`D}&0-;Vl5(U91IR!2F6bA+_Mu?3fKLIxgxmz$w- z+Ye^Q($GLu7Q+H=8yaA*o>`yft;M+^b6RypaWWqWJldIghQmn2lGOCI(FcK7a=!(F zuQBC$^{V=g?}Rv^u#j2HN~>eDQAjq^l+n%YR=I3SE!%P^*?TN^S1hTnh{w8tAw6}I#9*w*d#hWgXy)N z=Y7J*O7XI4D_Eg1nzjNtC&sK42vRzS2zm^%>|1#F^yT*+udar|kiiWs0p0FE3f(B{?NQ)EuS3do~nzkiwxo>%K+NHMnoGP|JLt`!r(@9)%c6(5%MmSBNi|#{H;SvpPwouIBmW^)=DpD%?#gQU zg%t3n3WeYY@uewX2X;@ECWthz z8y+uZYxtQgHU>#0@#X&H8@D|pm`^zLoNp=HN3p{mk~oD5ruEvd@!sd@famRy_q z<~6IU=e17}-t0(`LJO`06>FjIV$$ZfYj{{|%OR7U`d!20W{4L%4VOr;xPax)Uk`Ek zoEokk>+K$@JU3l&;Lse>8iJ2vvpIDcoE;;muAG&J-FIYc6o<{SA7deJ)r}AlS2EZx z90;niNU*!Rwn*kaeWcwtF@Lt7ev_I!Z9io5&DW>=ZkJmnS&1BdZ?4vtVszux%a=Jy z$$j^){e8#qk!dZE4nrfY+zAAT@6QXk6D=v^3Wh* zT2Q!8eUrgb{l{9pw8-fw?e11%v-U!5rc{zavs63&byQ7+N;4O>BCYUV7~>4RpUF0H zVts&0Jt>e43W2YD(M6~PSpV!l4~fRt?XsL{&bbB)|DgX*rjEN5WrVJ_{&#f>;Z*sz zO5LL3qN7wHd!xOemFBmA-{Vk^P27_Jt3sSO+}wGxZnJvbDJz!V;B=5>F)5$&6uXWv z`ziL6DTBA!r@)Qo0j-k{&^mJ-d_RjF@9=I8Flp*{ig;mQ`C+*xKbN@=Ur-N$wg70U z`5Xffr8X2dfM56}MiOgRq1zvm)lW+1m7Qc6#PUw_Dn@bP{p)d!-Ll(}H@%bj z{T;B)o1(sk_RhN`LAE1W4120 zk)02s;id$7ZztS!ER21*9J5E2Rlpj$5+u{rE6~5Uf^64A7}7oOaTKypih?xEKF#;Y z9=@KmZIVNg z-89EEYVtN4`YZI{(pm(e={>tYS`o_EPoyX(i(3z$4#FIDeJv+$Z!bUJVKuM6plR{= zILBruX><=Eq%Y^g!-zXpM#J!F|Me`=*UD6aTBg}eU`7i{1a<{b)uH{{rGQ1R1m)zp z!v`H0S*(+Q^H9u94`?}+p|Eq7o}9A`8!0<=HbNh%_3h2-CD++-To!;ZO_{yt96Wr> z!4-d_bhycp5bF87VfpzI$Qb(|VwAnW`W_gV0p=tk#NpJH4!aZU2lK4iv0!+bqh8DK zU`+&NWmg!ols)%X606$(TLZwUA$6pD_#dw9Vc^Q{Uwg!O;@;`X4ZAFWzRn|(7nd4$cIZfR#+z_#m#yvGeJCa8I?@w9cB?uaz zTB0P=bZEeqO|M-ot)*vLsB>Fp$ctZZw7b`KefiU3rtVxj4{bX{JLT(#O3b0 zsF=>KRkdvr;s?qB6R4cQ&F%N)J9gVMrlux3hb+lNHZKzK$qhoxZb7C76K%-{qXO;8th@;?L#BA#d_P_Q$YF6YaaB{}|o|rOS3d0KAA7 z`Rrfr@;&u0oKC|;h|dzo5~jK_g~ssPx8V8dE1A+qpEt{QUz+*eP~@T>)?Q?^Hjt=uO~M z-7e z->!r?%*4$+Q4px)W=~_RyBl12Nbygb!wF@5RnlnCG$7Cw5pPoaXe!D+!#s(VpW}W8 z%E{8P|Gw&dj{`oLxLoV%g8F+j?1V41VW>m~EVE-41arIk`(gE$Y+rs)&SsQ;^}1m0 zJ=#*J8EiN-yKgX2Z+48|os?p<7a>SY++)moO?LnWROtTx(^_vw%8f99PbT}+EeKm7 zv`;_rhS5oD1b5&Cg0#rMA4LLRanmI*SZCreWqNy`Uk|!6DZ-r{}*3Wt`hQa*Ghd+W;60CtanvEO)qdsbqvry-Nh!Mlex!CH1 zXfl$4K&4oq4NQ0f0s?E|O=nM8L!)xPjWuDy^ll`H)JL7elIr+BMK~1>FR^~%ziY^E zN{N8j1u6ms6ss73Z#AsqR9se?px+_cg38fQU7NXL6fq(j4^eJ~sM+(%^9r;Dq}X=RVK)CZhC$Y>5k`z* z;BYAQ@458rl4%Uoi{6|aBkw9^O$kqo4;D&ahH%p5oa8nR-V#u5gs8<%6n)V#*~jGY z`&uNBRCfX7d036Bb~)G5u-1vkHy#$xc;(+?NUT?<=R=Nt63Db_T*?~$M&EEV?{4<) zYdJ`>1BxPp?j5m&CJI{aVkJUx@@2$JykfbVRBT1^K|H9;ZEJ;7Et~VzE1JB>u|NFJ zT+ut3Wx+!{?p=O`&GQ23!=dp-D`yr2JJB>*PRUn@iiYYE<25EmJCpf>wYHg$Fqv&( zgoBd@9eKpX5qqO(Iy%HB9kQ6sWt>yg^;ll>_$nhLtx_cu!y5)_5siUD{s@v0o{a& zTnG<;>5|88=WD0$0U-9X>w zcVQyN(}g6}bQ@Y^v7xR$JC4elnws_pb8#R_h>M3up;Uwms2MG@vp7C{Enw<0)!@ST z;MRr#TfheJp5byizl4Q_1(yp)7Z--R6NS(dDFP`p49OE#Dtc^gVOP$Ak`LjI1s_z) zxP-j3LQAasIX79i7+!{IPw8Qv{ay(AfQK(Jk_A^g_~!MifwZ1JU%tnUp>!Q%AHkacmzo_!blkuWV5d#K3@77puM?x5)Y$SJg@q)0xl+4E2k3T)% zkNOP&%>Vz@k{!qcbLrQiZDi@i7k$p6l)~yjNJ<~qM0JriVw@x28l0aK0%0+vMUE#IYKFFcG`U8K&h6MDDhjh=F|5Voe{mvWFn;X+w7Jk*|z zS5zm&-j%tp5I?iD*$t_T$Ww|bC!7i%Zp?XNb$WSYO!2C?(nopaABMsfALdGn^_=#|`5wG)n5rWeN09VVCkJD5r{1rWHmAj0^m zJ}vtM&L=dv$d#j3y>{gC3T}-fq_Q;p%7SXu-3AeJC3fI z?yo#4X65S}{_qibas%0>9c8=+$2fb81t}W;8~Y;f&6d6yJ+Ddxt(wVI4{c&ZkJ^^| zFblwv(?8z~QntN443xC zRPnn6(T;an?vqZ1&1XYKPfQ{?%CLx0-?8iwUM)nZZw3IfLW;PTe*41qFIHnZ)CF_{({|t({!Oz+jkgY>aLGo$(3mN+6SS^v zdBE=sF%n(4x48cO78|O*Ez_`0M&dPmuDfWY#naH+I}GD@k45kF2;en3zsd)Y(guv7 zX~F)igJLjr$zmeMK1HS2h76{EzHlmvkps=9`Pqe?u0L1-_edM6tth36@ZrjI{aE>+ zo|Psj8|f&}@w7y}yKwlkf$OFOXnoGM{?M%M0fCI~~=F9^xGVGouj{ z6cqMAt@60L9tP=lltDDk*7FBrH%r&TVdQw1{Xs(VLoLBTi;LYW8O?lGRx4~v7ZbN6 zz=!$R6<=A~T=~i4{cc?San==~<9Yp|vdl~3H!gOmYMQ6ecaWg;1{O{Vs{_NZ7)MFR zTS(gDqiM={nPJrfrQ=oBY({{1 z>;bWShYcDK!jGXd;wI77*8Wwh*L6Zx@P}6QX!K8d#aRM1 zl-kR!J{%U!D4%kCpZ}}u`0X=6VG)sMzyZ7QJCz@#Nk2mlB*BE3LWy=)&~kZsnWLI# zTsXD%FW2-X#+AL&*QqJxhidG5UA@J-^7(%omvzm}CF&jb*<55Hnq%cgLx#7PP9Zp+ zf8%YIc`YZ4G|Z<;K5kxtP6$RFzCnyyA7B) z_{(oCD0qJ>^$ZNK)6`h!fA9_r3@oy1IZj?0-P?R8lTNn2J~L^$yHNA;w)2#B{Hvjw z{N2g!3N@nLsi#GTVHYqB?@lhxC2)WSIM*9iNHCAf39*ij4gmRj1G_)yx!wk&a^UCo z)@1L%{6pV+@{ukGn|TE|>IH;^!XqQww@0&00Op|69t8yj6w7n~i~V$!1r7izeP&ht zchy5M-$g_*ATCKHbHxEkbfnTu38?P^z=nZIulocrrv}=fHPL^Y=(h@ikp}D+?0t5i zvwAYO@8JH9Km=n@MhfmsD!{`}>9+ z-`(2gNA{Wt3Py@R2%T7cuJC{$BB6Ts@F2nmx*w|2n0>+d>Q|k0RLP;!k;}BkPpVP7r+dzq|B@=W{+g&Z@yp?} z6~NoIglL{LW6g^Et6+Pfl3r5S;c<}fScZJ?z+y*!J&y=6s=CoxXRU%Hi1NK^YPsEO z^ltBFYhv(Bk4CxX+MDaY%vzgpvCNqhTF``jeF z=v@{xW>d`gDa^)leZ_z1EHLxO1KAU%qUP>afjUwaxpVFUE`P@G$e?t*yK|*bN9r@%+S;YL;9v7b$%60qV z?PC{kYGtK<6L})9l|aY$X1;9zz5>yy(<^I;V08ST9lW+$ZLeSI1UM=qVKFYHht#B3S; z<;xGiwFnqRUGwuvppJ4hPl0NpP)+y`wXZ*kleLcvg@C6WxGebnP7KN1AF3nfl>%4% z=*^TzKVyHQ<2Pnaua-)(z#`*th8Rr#TE8W5b;85zI4`letxuRYp9>;JcY-X%!+c#r z52)uPuBorbP|a{Qp{Rfh7$dLvRosy&g333ip}4>sZK=$XY7oJ+wttrxVHX?6$%*1* zL<$&tapuQK9gu|PjGHEs>(D&M6Mz`9MkyL6>Xwx|&$Qx-@ zQCj;vXFbr0+4uXB_4N>@u}qQuID>8ES$y)C|9^Sesy&agMClSR&cJgbvBG?O{-&dWT;U+(HXuu~p$qzoHs+hQ8G z)}khthV{nc&>`KVy*n$m7#(SeM^Oq{`?YmmV96a7ixF0lav%1QeRHH^>wP|oD|~^r z_n`!>#l6>kad309{Na*dA+6*|qNu6WkG_>$0SVQ9#AY3RpKq&dF5fk=TtuaDvyq9+ z*Ub4o2NGIt5~{Q6r_@u&PqeKbkxGca9gL_!6!k%cea)o& zswSJF$K;D?r0GrSfpS<7E~7Hrk}T*B{T_BBfTWVWx_cK+J9>i|X-AD<(61X3Xocn9W>=^`95>q(oc)NWJYn?ckp-!?siD|SYEs}J zv0cfsH#3?uk1ZUEYA_)q8!iM3_C+s6?`PSIBi%jDox`qUh5eH0L=RN~yaU zolo^%!!GwLOU7M!l8Bg`&e#i@^Ukl%P@yiXNWF1!38jbaijGzb!r6-MQ)P~&9lpdAVmZex*h5gf}VAcB7AF-f(ml+V=Fd2MT6K$j9^ z>K9ST&d4!Ar5dETd}mSfxTdDY8YFEWc6zY$0~hgo09qs=B7y~}5s&L_GI`3VqFdHQ zH*arm{;%s4lNB6vR=h-{?{U9qp=kaT+U$g1kBn1sqjzT&{d5e|&~W;K%7{~9ja1=zLOFhu0#k#i09-o8z~=XcW6%*7v12&6?kiw4k}4soUCk38+8jOhOmCQ%8zo7^#z9=OC7Tb@TS@7##hdg?gKM*p= zGT+H{m)hCAB^kWbym4|^J7u)rUXq+u^qqb#c9k{mVsnEvT)eX}({h*-)pzW)Qn`BA zkz}OY%yAE+-di)On?ZoVvVQk?GVPPa$RYPTDmL^l?de7(JHN$>14$Ww5K|iiV;TDP~zKhmz~y@a;Ag1{-1NF^+Nzr^unY#DLY&X&q}hsa%p!yTSQa@fTK!` zCXhXNt{R8q3KGWOoll@KXc`3YSikY3_1%=(eALvA+bl8{mvX+*oYq-n{0z3E69VKuZyc&Qu%#_giVh67@3B~xUANavN}cXa1N1*3j>@w5{qDOHWf z`cn8ZBuQG|Sw5Mu@n9$7eV#L$Z0j)mm*n%laMH2ciZ^L(S~O^O0-&bw>V5Y zURM;90b?>>b8^VRo3yd90a-)YQeD9!t>)mEn2UCY_1@Qjh$OE)TkYWVmXP`;MqF2! z+K}Z9c|rfVWlM*j02%>mVZueyXow$%Wm>Q7@rvhk^`ilEY7^#EoMv~&;VlVSBaE?6 zVRa@B^60YSRK#!IT(`l=Q%c}4w|oxZjW3`^z1Sa)4)G_nSyoT4PQmHYl@jnXoct)E z%oNQM9jg|HGv_Nup)dw9`4tDdey3# zVlJuMnMO`X`2_=xxI{s}LmS+JBI&lL=gM3IcP!Md zH#R!-v9rqBh6a|_;tGZGpW3qVVHE@4QOWgeI|!dL_s#YvbHIPmvbsG6%lX9or^8F} zXFoc){8s5`iQ{V~gL#|T?`oktmyDG%zg_Q?gQaGJjm}40z^Y~^gq7S#V<3P|IuR{G zO=I5%7n%BsoS(?v02fv$-1Z8ayz<6&Dreed@R(F@NMT=#Q|ZwjK@Gzq!*o3-DnvEw zQvG6oDwJH8zYK?&d*j90g%3_nnf#Oc<8K^zIL9}vlQM=BxwEwhRf88oMz-V|0usiK zSyLsL)C;!BdAo8c%T!(AQE?sq4t|fUsrG~SY1Fa~6Mi)2!=e=PE*+J*3mGvuIKA7g zwEJMF%tXo(EmGKj?u+=Xe-`eG!|IMNL7J?bZkzs|8%~(@0rs}E7IxAmUB{dTf@@qO zfuEm{u7zGuR8;1I^;cN~@t3c$+D?#8bL13)aK-9Q zIs)fE_u=wb>s*%BKACrJue{p~8X_KGj}Wqmr8tdA@!uYf8@o%P5kMDE=fPS2-HX$gKmX6tj3-Ya8G}pc^QklEn4&h=}58JqfjyEK4(-=I) z@OX}m#N4FD-_iJ}9qNoIiQieej!VC}zo}?dJ*Z`5x}TajZ5@bkv-y_d*I`A;`h}KRjSDJhJwq22B^VswPD_UMZF#*M5q(~~K|#uj75O0c<|_;z63 zd5C5-=(L%>Y~lltmS-f^(+xtP5%);s%v8W9bQbAUrSK|NDY|W0u{CTG^zCB~lE_(l zkrr-B4cy00sjRZ(#vHq7_UtDXeb&>8>Cvw+9*rfsksisEv;2obMPRCZ4vH(j#l&E|dNq^Z6iuU6 z1~AQFV6^CQLj%mUo1oURXYxfM8RTcUlLZXA;u`_3=MyyLHg<2BdEyExUU;9I&*N=3 z8)1Kv=9?aMMF30ke3>i~)u`*Kv&W2!Z;I3Az7COs{{NxuEugCE*Kl7@1VutxS^;UK zK^g(+?vn2AE~UFWmF@;<=?3Wr$wedGtn+8@{e9oL_w0M_Ib-Np1J+!uG3Q$IAMfvd zo(HZfB;H{{o`PaBI5;XJgJ*QQ-i~VDkzalf zsCU{gOOEHD*Iww|UnBqyZ1BLm&NE&f6`EZg^=!}0_d^sUVO zlWU&|^1jeH`mH3qsSxlj(tGW%c*5n7La| zxY?mnNygr8 z4i;wD!9r9~_KM4JmvNKStiBMz$G?`N9krU7;|E7fTBqwktslPA$OX5XW5&(H?5BYD z_`$iQShi@kiyv-_%K|ST^xjzueyA%{6MIa8*%bv!TI_Vr_CmPj-WhKAdq%aQ?wEc| zB-lW@WSj5RH8S(#tyHX0*mp;b1BPcb<`P6aF5fk_)Q9}9e!z{(Pn=H&KaX77iRHJt zKYE`4otRpET3wd6rVL-W}A4dNymoDS5#AYuIY@c^gcVO-hmjLp?JlI=2V z_aR~O;%fS6Aq8P&iAA+{aXLm<1Y!r>cZ`TwDY*>&?K{7@-DJ;1k!Lk(pN;kAbjxug zo?3rx_1ltDue5dN8FZb=>Zu3iVY%nePlpDf5-V&%nM1doV)-+~ zhQZ~05+3(zbB@@#<_VpfQ@qn9<*4@poEN-eD0<9>;ezEEveb=Z+| zzIOPPWQ#xP{Rw5KS6N0JX79ZB9L*QevE@8vNW)q<=#WbN&!V#v?wY08Q zB5qu<5>dShq;T4E!S3|V8h&rKVV2oy5Oc}3NG#E)@pQ6$zSn_h-&+N%eJ-}V*#1Y6I#7bG16>fh#G=hoerccYI#4bu#$b=bw%EuznJ%ICq->O5F%;^a{+27n{H%gv9C}2HcV_OJ z-h>gqFr90+uwUHBrD|V>7eCF<==!&k zfy6_DO&T4Z?#v%@oqdh&sgr^Tl^WHslhlfgjAXQZP+P}xi6AkZP^j=-{V2Z?H7Pe` znpA8pw3lgWfXQg{L*9+_TKoBJ+N@n#w@BhQ`J$lkFvIR3Ll}?N12~Ks8KchY;WyDX z4xPFg8>=`1%9129c_Dq*)_E^OjgZEEYdV&rVqWcYno_+e$A4dlhSD>8oKqFG!HIqp(Lwqtk z9V^LG1yd_7J<3|S5NljN>Wj04aZ#p`ij41^`^^;_aNk^M z@BH0tp;6KE;A&yYn5yu4u|lPr{1vM)UD+OlBzmyV4NZ7sXo|8ZuG4kZ0Bqme^PEKt zy3;BxmO!KN9IB{4N_qt6Hyxq^iD0pf9H_Kgn^W1%-BiQC{ZxZl@e%FrrMa&q3Z$sT zGa))*)G?kQwhtD1fV9H3jD=Vm4iAAaE%{d+zJHI$*Rzjva<$7d$ z$X;<5Hx+VxXlYO|$;tXP7Yte<@wdhKNm z*vF$|5cZzEcsX0;%xb5=(f;MFB!0Fh?|1cKaas*N)V2g-G(z0Y0$2sLA7=2Nlg+@$ zh|%pA3)9Q(uLtxf%0{&NTmI11%s%@+YzDjJJz5VJiJM`Biu{ZUSqvp@J%DWTj{)Q=8@OF^ZX`Y zFTXr;Pa`*MD9)I-p&2xvy+DFgO&r^^RqJ+6Oc)L68sQflAHUk-fC%F zs$8}5`_JkZ|FY7-47mI=579nfX4Cw)pAVcT-nqAX!OvS~Vuk=ypkH^r+JW!>TkReVdj3CYm4t$3bpQR<&jrxHAJ%5B?YX^#D5Yvi(CA)b>4Ev0>4E5)20S4GXb=s0>ejGVfF=LA|L<{+AhTZCKGQq=>Iws zS)ri6taZN1!u{o6)z< zJ|qI^feSa*kA6ChP6uWTD?L41rF*7!E_{hXdP$5_RF7vbTM3LT>rZ8Z1NOLLW{x{N zem*C&<0$%ROlGb1!nDxM5VQ5OT+I;@9`f!VS@}9Z4ql=9EIlX`=zn)GoM__~*h0lQ zNh?-M?U793!2>;UL@VAMi~SN}V$zJ1kT#tsmiOwhugVW@ydPMCOi8yG>~}DC-KNIl z?QQMJshLu&=x%Z2A-(x%77xd0$<(ZyX{Bnv$~ct~8?=)7veDISkBRzm#U;tmx2_bk zt5Qp~lBW_6F$fYnwhqt329C*Zr)#ful=*V-$z^Yd&(LYt zTOkBSdgv+=#SS>iT_FpR#AYKRlimLj7G=LXgfWn}w%ru?d1awfNhi zGHn1T;gNjpm}AtZ31)opsQG%2_tCJt2{Jq0tSOM&>CE#R(%q8+JXzb+K%<96$^eg*h9{?v|HTbA6pFSEH#>2 zT9O%bpRBIV$GQU$4^qH$!Fc1w$H6m}KupF~rz)NF@g$^yENmx|nj13|mh^leDUK>q zEm4b1idbUJuf0G(kl~0K6Q)qamZe{eOqwX~wm(w1xFwrVf^9^1JSy42;_mI+kMzk* z#*z_+=F`O^HPQCl%Z+?X8kffNK^!o&nJv=n)^}MjOiWDR)B7gps;9u5@UZohu`+)Deo4C6DGY`pHH9<`qTBfqK*eToE039zt;)v#$mv!p= zK!mI@d?ZB?g@*Ytc<%neHaX@c!Z&d-$mLe-j|pA!i&8WFl|1PTrf%|lM(C_e$<9iO z0xZuThZ>{x`kC{At0}79%KpzBYz;J-QrlsV1^GsnjUn(F|Z&W_G-V6lju&Y7EXeTMi@sOY>5v@a{oen-zCX)2KCsd7cTGkGGL_eUsU z?D@do)MY2r1@HCHuNTZ5&O7UxNNnl1rxu8W^x_4knWG@_wvu2G0nytG`obOFB;%nqS_`p(XU^Bq#7+ z$x0jKY|7qdg45a~jK-!9{oYGCenCs>*sVGlO)uv_preRQ_|hEnb8nQF3!N!G#kYvi zNjANeH*{KG!u0V)SFrhS>R?)gi^Uq$lOi<(y8212)wEVJ(&;Yjp9G~h_ynOMe0ReQ zIY?S{Gc+K^>~*>m{Z1iGQ3ZV-!^;u0E`HD;r&Y=AG(rA~S)UipGvGTCO1sfn6gQ;G zp(1Z|S8kuMCBv{Z|J5l;j5G&<_xs$?F;iu2CnVJovlZO56_1#2TzN1ycB{O~w<9{W zM@Sc8suIe=*a-PngZ6Abvo(@i+RdZ87gJYN+JfAc!*geY7%fbuYScX*=Uh+C6Mu^y z!6TKcgv>#XHU}@Ddh0r!NmCfA%x2FpG{WDKd`N3$M!K9DR~r1y-u%+OQOdYz6t3fm z;0}*c6O?st`pG2|k>q*u&l)F;!o@6Vt=8}pR%0*s6`~Qd$pSPGS@jMi=z-3SkxV`r zVAACqgJmtQ7-d+r4@No4m&pID3vuMfyBOhRk%3-$=WVIZQ&;I8UeDiBe)3}@? zj#_Uc;F0kdoc2{e;{xM2z=sCX9{mN;Juc@V%HF#eN3Y; zdePcKT;$BPYNmsk;e^dXSK2~tE;k*oKCXP&ccbfXbRv?nAcMhOAtE_l zdy!qW%@kkyWp`77rE>mW?0Uh~fOYvQLefAqG+t=A3$?nkSoSO>bpl7@?0qKPj%1!SsnN=eLo0ZmLslEg1(-Xj~YrFT9DSIOnZ* z^I&xFXNOauv|z$`QH|m~>d*D1rbzy4%*ahAU6}6Pp{Rgrm=)(+T5Nm#mzc_1ZuKNxTCtu8S2)UmWQWYC29IX0Fd7AySgXK{;Fh9voVM|WltqyCGnEu*v>JK|t5F^lD9 zNcyShL7Hr@p*Ka!l?mBaf{sxcyXMc5OI0Rsj!;e(S#M8cRK;(jjQG3=g}yYqzxbJ2 zXYWTi&Uh!L(WC?}9&*IOKI$34BHAz6zy{ay6hWN(?+#z3Y%^?x-^-pMzwu1Hn0D%U zFmxS=!bmQDSroKy-$>1bSQw9mF7Xn)MoHDLOczmy%a+BI@BWGSyDC)f{+mLCF0?Zqts^-iNK64-fK%pQQE|39naU zn^DbgBwXw`%an~8O>tnP3if^2=d zZ_+wSy*x&Jb#;Bc3W4OaTVGloxMr)@nsI|>fc&Yj6Roo!UeNnX&_AP4q(}uOmh3&p z`}Ih^0+mr4i*?Xe$;dkOARieP#`2o_&-PgM*B8|Nxs8(nkynQIH_XTPT(M493MDut zz&|tWXe6r99fPv^R}NXM*Y4{)E!QL57*YG>;3>0v;ix_+wuOes?~U!Gu;z zdV3#Ec8NsW>O4y^u{=_nj-L=c_#8dU0cp1K>MJd?n;m^;Wb}bHHL~Off~~}i2_!j5 zXSwFB?cfaBoHyL=?1z10so_9Xj+sRA#Pv~t+p;ZYV;JLwOHA0gFVY)2EI{{(03vPDQoC_36;zigu z>uc<)VHsb$2Jl=7=@myavk=|RWxohLpPHa~ch+zx1vz-|^7+q;W!70r}&bhX0@8{SATnlm2;bnoSn#*SAq z#_B)J|MWDbVHW39l^b3n@=4fSRCPE+aqh&p;vq};Ig5$1>>{n_Mh`rVWW^>fH)|95%T5my;b=(EV%p75(GGWWeGxW1{x>{hZ2A{G zGh{EZ0L(31hJ3HK@g_9AiPZPv3gdd?9hW=3aB{SNY3kG5U8SxD2gVc>%gBaOETd^` zq-!#t7GD+I_#;my?`0NFr!)AU^F1N%&eiT1+}w*n;o%gltdPKOZWsGq8&s!fTjvK8 z=?nEo_K?G;{O-$x@_5co@`O14B?ixyYP!H^DUNPnWwlSBuj&)^MvH+VGV z!p;&4`b3EBM@~iLQ=(0;!)sM2r{<81{yo~&=(0|%gz#4mxuCKWdGH=W(vNt4*S*@~ ze)Jawr%o#CmWpW6Ny@{#`KcecvZY0ZV#~Y_S!^BKZ4cz*zI+r4Pkw?s&*fFL{kkfH zSA-BBfq!79`h&HGzGA{fth@aA)UmOVq4cbIZzi|`RLA`WTQ+x~?SRQc1_n(=t19=B zq-9|-0cAR_KrMlXbOtY;irs1&@lePBx|!X#fnd>}~isN`kO;(@Y#Mh^jyM#Z+JS=1TYC4*~+aNX&XHD0^fqivQ4sSd4b!&#&*D zRECDBvf)W+(CpTq>K)xJfnFr=@6Kio9jxy^RNt@HZG_iAs6hNg}BP5CNY^;$_ySH<_c&v zln=&vB&T|vDW~=nH+|*1VYgW;@DoSyc+u=US=I~w?h4b7!FFS)X$6U zFC%Z^k(ZsPDsq_X1=pX4OrR)yu%$`DHazbgXQSGbG||Jm?5w3|OQqN2f1`x&S$Zapk7=H(Hx&tgAz zVyk5kZ*qO04_?CMZzmhFk8<;FA8Zfa0+kp8;6o`NZ5&8g#dl}cuNOw=TvwezlTWfZ z7Px)#wQ!xYtJ1W+zaqCXKLQLo;NbNUP|bLUr$39W@ee(nJ9P-{H6(-y(|ebyXE<2A zmg?WinI>sY3RbxuAJ5w(8uhJ{ zY#^sBU0^q)GN&wse?H<_RQz?3GiySLhDwxD*-PyC3{SOi;R4@q{G+nLvkh$W?!mXV zzYb-pZk8#DTWz_Nv)gMrbv~mOm$agSN)=1+HGAH<)K1&-&l?JOG(oDb*?N(%&3cA-hDWltbx%`#nl|*b zc)i>fFwWuCtLlncgS)ZoMc!Nd+a7DC*QsiaR?og#^9@(Fr{W-oa@kREf5~z=unEBc z;?nPNUZXDS=n?bw9yuUrIhB4rO7v%E&(QDF<;RuApUd>Hwj%>0#}`?erHDV`!Ft~F zrf&V_dIf2^(%x2QJi%0T!TfTE{BsJ%-M^4;#)_wKHp=|*Ejbg$uGm6dk)s<9fO0Rk$fXw{5rt0k$}f7E*OJ?$#4J@z-`fRym^&7uwsbIP2Iz(DId11wlJI8-bg2<*w<8+B%E_H8g}!L>}bfCTmM5L$Ki z)Wg!-(~VW;gkkl;>8Iu(?Lo-z(;A;y*`d97+>K8n{mHcN^WQp!I=Z)bSDB4s3F6`T z>F*>w%x&noGyIK@Y%F9I(n^L@S8`DX-!HzSlL#&hj&n@k%ZkEUyiCE0?TZ!DD6DYD zLs!7v`LfELT!A}?XGqy(#8Tg!zo`!Y#$ZBhz%p5McyJF$%ev?F$yVHM*fl)45NPbN zFDwF}?GgrsiY}b96*o)n6y?S;(=S+4*g_UgB+sPx(DlD!LQ4&m`iC-J39)c?f|YBi zoT4I9r-PT{qFK2@IjKoYBk_IN zv^JRM&|}nnCi>1qk5r7WroVuelPcICZ82gkn@UvDpg^Cr@5e(j5OA>T7hih07clMwC^6i`}yR^_%G*E$(l*%cw)aKD1 zJq86_i_?{^$%6?{j1PK_Xgs)q3=>zToG?byyLQBt95{1iZ#QU*P4GEHy^w^BDPfAV zMn8}5b4}p4(17aIHRVUOTiN2)NAIQSbV z-KKH7##~*w)LG04I5<3ds(~iq8?ZSB8{`)+UVH;22!tXDA2IJu754)7lVuJG3lter z?7F;}lj&QvS5~yF9*xZv_q^=h841JIwN5OLTx?6`?*8cI{b0g!-$Lp3SG}FF{l_2m zgr{l2vpIFx{)BHuY$3qg=~tVDqu<~P&P^g+=B_NjZ7eoY4=pH+Z3L*)cW>dGC7fxa zMoVvKj|fen0bp|F@tptTl$4!K`Y*8PebP_CQMLzciUsewY9i&gcdQ{3lKX?Hl&rIpeGM|G4};KeqW#zWf_Z z*uRrQg09N>(*C~yWIbX~z@EAAgRTl7f{Ut3g@7H}5ZXVIMAgk-1r?+l1!|+=kEk*G zMCm2hS6GQiqsNOvMeqH1iPWcFYk?Q$6k;H92}Cqwb%R5`sJ#<>Jn{%D`teASMd$)D z%Uk2WbxuNeGb_|@{)UJDQXu^e5dR;dCt`2>gnzV2@c)p-{|+Suno<4}KthQ==}(vy z75ta`qd@dOr1m*7D7^eZF{>{U%=CfnliGyiK{9avX*3=)fi&>HBNX%pE1}SpO?(2? z8`BJ?`zx}v;l!ipLV1OeZXo$G)6b3Ci@#=O`T|x3v1!r%!9pEr7p#SvLqt9=QMIce ziMFk6`?Q##w*Cuv$ZgS|i0GOPEH$CxvKzsRiS@e~^6pN=SFQ){=2=>l z9RIP&*1PTz`Xv1H_o`f2=!&;AX*o()PuC{L%pR|aQY<_(Z&#P1^pLMk>m@X)grIe_ ztwxjm>#WGz#rbQg)~!F5D=F_JjIv}Yf7nA{!MMm;Q0#T@#$dYke5s~1oHZk(n^Lam zeJVR$wq$ZlqhYQ%Hl#PRY^-)cC;LQ|r8u%Pr(JZzj)|C$!fx4(g$kg^63)5~D;t@A z1E&@beB8^2NxDRJjkQ*5WI6kpo#D~Y{{^Zc?G*x`)~YPSRyXb9hMBFn;|4T;r%W5z z-@dIjZthJV{QeV7fAUV-3M{Pb6UwqJcF>UyC$jnL&l3Ym@fqxT7&Z_&WQe)-yC+$J zN_}6u=;&^Qy&fzWQm*2)-L$UjNO7WSjrU5eB zyu1EQ6`Ei2=3Ss<>{euMxV8lP8anKAH@u+|xtCG>hH$j$Lc{Q^Fjt%7tZa145$cwA z!xM;f=HqNx?4a1UimLtR`s}m8h7C%=TrT6Vu+yVaKzu^~&a+z=R|p5W9mX9t%6Een zL>TUe2ke#ie15`~9z4lzuDYxqAJ7{JHmi-v%GCYImD)0fnp-Njr(VKHxp9i6vS)jo zJ>Re^>5c5Iak6AFH8r?9c}bP&h3I;K^0@LN27>!n6{;OtdD0@OIaM@1FQ2x1P^UMip$$>!Qwo zGs2YJ)|1Ln{-Ugg5xu`45krXFlwkYzyE#$t8K`^}sMk`1V&!9e%OfFwfkFO?u+*|o zu|W2t3hd2>^HSb1JN4+Un(MlGJennW=I7F&n=OH<@WM|n& zA-nmwrrKbA)BO>M#f9;D_d*)tRcOGhK)ev-mPpzq>)BWJfu3RHWvcpD&h% z*AHuljJ6M)92J%j@taFr!E|-Y=SUvy9iLzUT_V2uN{-Amo}(+G&T=4NAcQ%QNBfIa zoe$h%)hmSEOG|X4$;$>?I#PqRog936q$289v9Ae&XNI&m5^8DoB37c4;|^!vtk?qw z-9N{Njvlk-a7smN>({4m66n$mWeg}|Hm?(tz5xU=VkNQ4GTlkHw+`Y!*Q8s zjp;LU9J5UJ>3sIWv{}~GAN}C{8O^3EGzZ-iPIQJAqeXM(+R$$zk9<=DU;3(}>Z>-5 zeQ5mf%Tb47ota~v(YsR;c&mJKpJ;7HLNr|5m*Yq%)vBki7N*N8zkGG$hN?SV@C@b* zZ=8@zgv+h=^-qyfc)1oUrDm>L7S7mQUsfLF$c^eTeAO8@yh?f!6ZV$+j~w~q-6nj+ zsF_Hr{YEQln#*_Z?wuszvsQiTR5?|v$Ba2s=ZCX|(oYnRwhbM~gG=e^Ch}t$tce(( z!e=2Ocbj!RZ($rQ7+T$2hQJDyT#=n z>^7dSc$^}GZ3JzUk*jnz^dbg8*>BL54_{+wr2Mhj<~ zV|u@WFgg}iCxtueQy?S|2#nx8+h*G+-NP+*sWuZ@c9Ry+frGh3)y%O-Ors|$Lb2%V z)aN_dA1c4j=v-*x03?5{KNbR_)f+N-i5)%nIAI_3gTq3ME_)Hqw+1)uKpxD`IMY+G zpkt+u{FFVPv*IqKT;X*8zm+Pj&HgP_3Ip}=BevlNTjr(=r^ESCk(L!LnTLsYd% z;%DM*TSt@m9>E~>5a!SlODx^X`n}3T0~v?u9S6e)-F&o*(OJ@@T?bSwQ5BBGMv-v)iYa;mtK@ z(6!DtaI-ZhcKsH;ge9};5FX~+Urz4-2L#6c8wB$A0ZA9ff-VXBDuidn#%0k6%$C>} z%1LvoQFxhp4&jAfVE1)g^GR1u-(;1~OhO!6~7#wLH-kEt!tuQ`vq{>?}pG4IAyDv#!9 zhBtzrun?D=bkJVSf1tP=!%7z z8YZ4?G=5y{adrGDA+yQf-*j0d%Ebknl#EQ0{r<%8LtWjps)oj<>T>&7hAl<_Bk7x4 zGj49~7%xxZ)9uw*Zb*M{ZWC$$w>jkEjS5vvHtat8D_-x6CP2@%sG`b)G!Uyu#qSXz^qz&uJ{V`0i$AYoPV;!oh;eoAIdj2 zR!&^1?%WIRI{jWe_O!Lk6dp&U^!d$26I#mmha*CM!aI4{9p5XeDmyv`I`+Bi6Z0lc{BCE7Q%#+^QuWZIR!js4xOa2lxo zURC;;DRUAR`%(2_>}KO8b#91lweZPiEoql<_wj9cJrG~+9r6pPUU0l&q{U_ic+NWQ z>6+koSBGoVC5J$cdl#F3(K#5i+>p8ncjkjYWHJ$oS62&yVecGmKVrq-FBN8gk5X5A zTiI7e;pp-qd8TY9ue3Mf-Fl)ne<9i4gXfTXosgat3KOnTsfIaz?R8)Gl0UFkS#q$t zO^hujg*KgVaVdKYy*XN4yj^*!>=DDqQ$}7kQb1C*Cx&%hl%d5XQ`kWw+7`1kr^D_F zAez>*TyX6QoW6Dxbg*3IdhbpmS9Cyb>8})`b(i!DXXPh@|gV+@!EjQOyOTGMq(()rea5pxwy_} zH~=)oQM2;h;f|zT)P7*9!ID-+#P+;x0^PqP@f*$Ng^EQ45LBW&v=xtg!jB){x-MVv z#V=eI_HW+OQZ}4lZ7QpD2A;1MPe5KEdFWPHtaYu@YP2$zf3eFbXB#4ARQcWL#CEcF zC^W2{S%2_gS(4UPzPM{T`>Qq0D?8)jkf$q->a=fYi0tj#=x}`Ji3Ww+C;b{pb_@B- z+$xT^voGT1yV*sM=^c0%-25PsPv;({jHZdzhL`X!`v+Ja4rLE-8@%#<;pj^45#EFjkLp^pGm~tLvqu4j#H9#n{;5wiHOL`w2=7+Ry!%5KCN77#Ja6D z?{9QQ4VO3lVMVmzMBA!ntxMmL%%(%$Ufwz=$Q7|B*O9K*eX6^y@_vPo!^0WO=md$Q zf?GS%`!TFAIZ&7Q*=w2_eLd!C(YuXTTXZQKJTq65#2Q)qnwa?m7eH{h~774^6WJpNq1g#0{ z(XD1V7Eb>J)b;NC|5ozjJQta7Tl-UWB<08V72~KC+q|!@=D0fUgv7#O*IS-y3 z(%#%S3#=*F)Ujo!g(6EU&$}Hcw!o3{-hyI$HAM2eC&B?8ow+&9;fi~X#@6AXxTXbn zVIc?K)nus%!jn8Y7PF?K?}w9zac5=7K85idjFCV+nP7F1(@T-MU%_Yb#;|%%S-HgRF|Gb#-q}b8B8+PS}VLuz}1o0`Oqv< zDp*J$Vf?CR_SRs#<|E66>2sgl_-8W2yfbw6cJ?HU@1xIZ0EU~tvV~ypz7W9Tfo4$_^0QI=bZ-{Ke9AF1S}#oif5VsJY3J7t>9-weP@obBfOX{g#jb zO7eg*wcs_i;q_J*#}aq2&{z4w9N`wTuI?_3yZvWi$YU01YxmyCaUPg70y3}`PJ|>i zwO0?6s`W!X!(S$^465utAj~#$G1-i=kYxH*HC>{|l?S-Q7bs-J%?f?b49r$8`r*1J z!b$h~N;+99KO#&>t1!k&eRso;lM!wFd{1n5?dwvNVZm3BVZtKD#R3yGKO5zCM%lkm zE69gQr?DGAxY1o<>T9x=O(<@tYuL`*azDt?dtWS>J04f68DO93ou7I+5V9g+{#NGfNd$f9`_24N7?7*kaL*Kx zrFsxMMe%T%hP@fO6K^Wt6aULk@i5Rdv#$-802%#HMRixxcLYA3;5WQV?%!L_DC4LP zorDe58V!68KNe5ZH7S&P(JjRu3I?2S?v>MT{K!a8M+U(tzh}&$%-|co)-ylKF=oX^ zF+aRz0D1OfnmW`{Oa+Ge(p>8I_196D3Y}HFrB220)SIy^FUx@hx@SN8(C)8y&#nn( zs;jGWEs|sI?9Kz)tBj=Z?XGW|QH@1!H_1cru%f+xqxeXbqvU9gTO&zWme@iakg0II z$j7~1zp$voV!v0$mPKz~{NSnR7PiqD-m>pet4lou!KrLH+x@4RO<=TQ zse=DnE;sBpZ`Mn`H0EZ}>4GUq4Q{y^;QOe9ldYxKvN$$~kB3g@1H?Fl^fkDYt*=`p|%`?+EDn!d6PmYo5fJD*B_?-;%OM z1B<_cxduj=ay+PydC&P?9}ST73EnQLIqVAcVBN2)>sf|(_JkYUongEP#ohs+EJQfG zBDv_mKOreqf3BKh2gi>0vNbev#-j+Fh~LVGqRFzHg|^0v%k_MM3;J61;hazF4m>5F z5uDcgeT{|uKjFuDs_e{I>8RH-Rjthn-nS185g`as%Z4z@ACH)ClmP$pSfFRNgCo_7#BRKzN%1==|d5 z;Tbbw0j+p`;^m)9xAMD^;Oefe^>#0Iz(!37&cE951Q_nA&mOgH4|VQcax`py zcp35jQ>h{vKJMTy+91r8!x$0hjvfgJ)d0>Ciw$HY2U?Hv)d$9&f`YSgS5 z7YVe-#e*#sAF#ddxq1U?JAm)(IZv~y(;>ELvbdxndCA@~jQ^0|)@BIc+&#a1`M!m5T9F>*?~vIwR&AX+FmU{ps^) zK=j7h0kStfN1SCY3LPLzdX3qi+s_4KQ>GQNRUBaZJ>6Djt1Yyk;B^3UX2%$569a4f z>AvYJ-vCLtEa`*Y{l!AdjVEz*3NaNGK?Hk67IX5~Op^jwsr-{^wLi)-LxGZR@6+pO zW|iy$c@&6$?28>n_|#>>Xe|4lDo1@<4P(9`bWR7(UAT4P1lQEZ_lZmzi0#&>6|F4jTI;z>9ikCcNQ$e}XWyZt64i~tb(N9#=I-vUYLjt0YhP;#5?H3|^A z`EYfBqLFaG!l@uGNs@Vclyh_uoYZI71^>y$BBU;VJX@Wrv6Ns0R7F^}XW+S6t#JU;chK27>_fC$XDzOWP7TWz9cGn=lhi)l8XFZjQV?%M|&~Q zNNWCwvsbqmTo$jf)452&ntH5;DkN_X^}bswSzx|?g``~ZMXwAmEC$B-NZQ-=K;kf9 zM{rEnUfCQumyvbAqWFG!u*5jWxo))|bc(L6=;=9W+~*l~t5*xH*7FBp6Z7$0)d zxP6DRUzZ^``2@dsTyL+gKfR8TqEtnf($X4G;Rkg|lbJF?!+|*18sr)Acxpy-0B831 zAN4wsM5gXkZC7hAL3L74IpG5Fhv`Hf92}gE+0CI0!Ue7Fo$AEbDsQ=VmX6a7$D!NH zfhZ-Eo;LZCH{qc_Gy%8Zrq-?nq@&;r($Yq+_*`BKZ$*}dveazjZvI^F|Lr}xd*Ze92H_|9r~Ztcyr+n>{%Sv?Fjy#53;Jbc|-In%oM z;pXOM2h4gk6QNTk&$j1lX+Af#-p7CVm$K}+z}ThM->NdfPo~^vA7A~g76QsD8&5uj zy8l=RcHsYG0w)*upvwR870-XN&cBX#}p(bcm4Ftx#vyAK7#ebUv zLT!VCgZ=#cUf+?CleapyXnS2hqo=0_A3;S$MV944T^f(Oa{V3X*U!n%SEx3Uf@yTz z%d5Zp9Q>DZ?2})kPtT9Th5ATfK=FErnB9tCy*F~yv8At1G%Y>dapL{^_lvEbTr-;> zYzMd)zfM>9+nq1y>y3RraGe}FM&+EFoo(ikLGkucB8k3R-#q%;!|!3;!ouRAE;B6Q$II?4?KwhYN{*3@T&FuXYgR- zWa^s zdgNX9_seaPFusumh~S5He?p(jBh*(cW|Kic^^Sp$AKK8s0ai}N$7|hCFOTL*yyO2ta&Z43xq@qJ z=Z28~8cF6tnD9s1%MH7KE0gG7ylgJzns=%~zVCeC4mZAUK;lwtM3yAJgW$+a(W&;G zM6S_huKOK58H!I&Zd&qW2?_alv-7Qrxmv0ftY=j6zzf&cU9CF{VlvXqlrwhxTn3J@Zzw>{ZuxWc1l`{V&U3w_HG_oPo&-{y-dydx91aoA*q11;u!u@g@umM1Gm{i4SfJef9KPb}iGeGSY9a z$UgmmTcKe6(8N@ziQ~V*PsYSi^oxHS)!NUYlN+YWOe^U;Q^4+L;M}PhOUEQckRdud zIF#(*C#UFerf0M69(-6DBT|FF$OOWuDx)KEVSiymg7thIV>P>U{^xFK7*H;j)Z~ts z4>ylVi0PqU`K*65?`gvF?2<-2dyK(&hjSp|efw`-9z+W+@9X0pOSVJk*CE<>n-Xpn zHh26bv_t0u1f2H4o!Z(xl5g|lj2Vq0_kBxibFExm6lw3DDBF0fYwlZ?oW7g6ZPcE! zR*GzAh9_l@KfPCL-;}A5#lc*V$NI0S~?Cyu_2BOmNPiP zk`bJ=L1!2mZ%H9;;3^PTYhj2cuN+Mw;f9QKGEgaCNeOcnA+Yx~+hCsPLd%wlP&}W& z100BP#t?M?n<_TA$~|S3MbCRb50rn8{RzEcrDQ#1Z{o#q8%^NiTZ*3iX&Yq!P;oZF z4WSlYTP+d5Mn@L_?tqWZIe>-rjE+WvAp1(w3G&a$y^+KUKxqm_$^3dN=)?W-#l<5l zbJ(AawpwaD?nLFoRGprkMB2)18Qztyy|xpz=eDGqJ)0o?k8Z~+{mZ$yy#YT)fP1|D z#f&$M&Bi7FnmjWHV*yn!Jlm?W2?2!If0H9I-WSISt%1=AglH(uc&#nT}#~TCzWVuwSJNk_H1~a&vfWIKw4SVUUEb>e(|2daEEe| zah388ep7RLj__0B+duP}pW^;k%@@mqwPD-HNP9x^-d!@Y9;|07G6@T1m|S(hruuSq z9D@ELv@BIbw3IaPS4ty<9PlZ#qt%GgW-cRZ47%3S;ZO#OH}}OAYyIb(zquXj z?X*MfV^U1c#KyPm&|G>OQ|v>hG%c%t{DTb1k}>({kDdPjezK+OW#mC{JTsCxO3K?o zF1BoyX;oK5BXgghTO4DkMdVI{pC`&+Q_fuum*;hqV+W!@pYQfosko3j4`r34Nza*h zld@g(G|z3&p?~~OnJ-8yU^reOAO23Z8Bf?{{t6h5ZgTt30+lv&b_{E6K#`TUWRK`W zE?C`|ryQR3{iXy=C{TarvpVyYaydM2SO7WQ1lU4(JZ~_*fCa)B2vdgQa~hN#M3PB8 z<*-eMlJx=Z7%a7(N{)RTu2(eWS%CSHO22gf*|x?Sbc^S!f@vBDy7vfjT8N3U&d<_C zaj;ZxCZDH!FOC&$t0U%%2nlIo_Tp&FB=#e-pM@5%6wChE&lQ37hd#P*J7w8km`K+? zS~MLnXZ#9?r=To@~gPsargyjhOgSZ?NzrXEdNjDhuG2`B0`y zEC2U^&)~B^b4^^w-aI55^);!fhm8WqHX+OT$q9u7opX~baV>=>WaHI;unX$u6n`H~ z@~I3)*%CD?@V7C}c@|_bgP;(1Wre5H;t7d&)5i#xH_3Ewt|NiLaD2C^={`7K?RLTz zjzK=NQeTIPxf0yiZ6Y&Qp8;0@j_Qzee2)y&5$;0laPIzFDkge4S83`gF@iga#5{SZ z3aQmq;d)9Mx@hSo9bnDP$+EO>}doAJXlRGjf zy@N-D!;)f0+QRor?_I{FYiJir*jt6pPqgoL-I*Cs?WZDwbvL)31+jF3@Gfdld{)!Rwq92%w*0M+EZC*EF?Nm>aOqC%b9Z+4Y#JJp z#D=T&cMu}06mC8}M{qW)?3+}(eu6M&e4W!aMM=(3iHqfoRf(sHZ|O}-33A4vrO~b7 z;1$38Q;GcjBwAdxgnSoEZluGETeG4hMO2mMyQjEKpvEC^$XzSj)=Xp~^=a|@Fd?$w zMqhdfs}}30GCfnaRSn9mO?yR{H|X?#E_pWJEK;yW{Cv+YY(gz;LoH4o7ZDQ?vt6;k$x$6cyt-tY1gHhtDj->041v>cvL3W zuZy_0T;M{Cx4Br`O@yZvWl>;kE$%u}LW_SES=RcVQlfVt!;5|)b2!2>_~yjh?Yy#D z%Gv^>vOkW;yZkTMU?U1zy8&ZmxxlfQ4j>JlUY!s%>|8kB93TCbQ7&y+zK2#jS@pi* z?db}i{fjO;h(?MNRiAo3M+J zvlaHB8aCV4vYO8NHM$Ow#jNuH+HK!8?rzxWbb_KgWiC6u;n)M`2L}y){&hCrUXc^G zl>B-$4B<;^Li}Bow57=)=7#|~CdXdeqG{W~U1R0K5b8K0hhn0gq;qtc%Ysz=-Tmz1 zv{Ek$B@Mgj9yEy|r|@|`vVX{cQVk~?!HY1jDYvdmJt;U z)AFHinoGuzvA8#YZmLoZ=Ph|jYqo?eDaC^MV0m?o2tCa&amJ`Q zm=HuArHp~ySmK>(ygz94dQkJ9+8r(jtU||+&wvd%HOegtJL%UE7c|ZW>kArgOD>ed z`zqnYBLU0t8e=|3mnc7E||dO8k_AQ;WZNTp?D9Pem> zQjn0K5!!#!H#nEft$z%}QxNzVSdZ_oEF7Jc6qJCYDQMn(W(lHIqFaZrBE!W%zUN{y zC)%z|$c8vu>5ZJKE|U$dX>MwT3aQyCS*lJSV$~gm zhx+X5ddB#Mj^L|bF&W6Ah;3YP{}nVL-3AFG<}!tz_ufu4IKO6v1SQ<13hK>?s4X<+ z5=7$&pI+D;PReK411sjaoBTj78TH6d6FMnKFK^y0kql<4&Dy)d@G(WN8&F!HzE3GE z_Ra*NT^)vo#$Ww3E*AZxCqC_}(#f$Ol0$WLP*j|cP&I$AJej|((nF4)BVss&g;H17 zcy&c7Upvq{81NVHkmJF_%`Nl2TG{>{rjY|@j!BLcccb66Fs5R)mkN)R`x|g|wo*wiGw^lffb?>dQc>4Buul z3LBWwpTR-830>LR>X+@p2jZmivy-t&@5(ocPsWCnp{O6hV`{(Oz z1?W8L*#wR`b`=@t8!_s!4kR0{9NBPHKKlYvIqxAn zzQK;-?^xmg&gBsOE;Aj>-2-Y5lyj)_UgH&GYBtYH*M~^@`}?cwI|f)t@E(Wir7gRh z+J(+s`>V7FsH^jRflM~t0el`TM?_{fN+!z-LJCTjZ)gW4`24rf zbdc!@8gv}I;Uibbl|#P}*wN3ro+0fgP0@^S3muMoReSG0yC|LH3pRn6^$#7^tm%%W zjQo=4Iy>Fiy)J}>NFOC--PRW7B62hAeUxsrhdRrjm#ZCUG{sB2VPm{0eqPXB`+ooZ z%K$@07DoM^W)JzLX@Ci-F>W|mMsqi%CAWQRy`~Ihblx1HQXqBHvhmTvzbT(6fggAJ zJx{~IEAeUb?!7sVxr&Pg@&!pu5}x-+X)XJ8UJvX(PJ7z;@l~$W5Bn4m7;z#|^n>tT z7}5?NXv2qmnF`;#78>2)&=cUDnjxoGm*TK#$9&hNttx+-E$W{47%_o{m3R(&Vz5L6 zlT5w`4tIgO&A}v2w==U)VgDc04h*?|M z)_n9C6O_@9K9jMpP%bNRB#70klhTYuP(r7&Y*Sl{a)no(It{ZJDL{ z&)w{(M{UFr8BNQ5$Gz6&FZ~x$&o=GH-@%+TxzVYw?$wH)bU2d%T{3UE6uIU=R@p`? zuFo2>JGDG-ierBLSujgN_U-+M*ueLJ7w9-3*gl3*BCOd&YeFP5!b<9p*>H=O;2f$h zt^Q0AK5^#57^<}!gzHh+Lz<^Qx5(hQFX=4Q)2|9+cOvLF$wC}GcT8Y_$YyO|$aqs)oFn|&ZN zIu}p4u{N{x(P6>@xtb=~BJ^L~SfQ@b&>UW+Lr$;xQvNoJU-!i2XXjD4H zx+>f{n4tLnLr&&fFbDpXEp@YI0IGha<>g;FZVrIV z%$}MnfS-+kb_>~5p5AXTh%Zr5yMOz%RU=h(u@;=9{e zPg#$Id4!AK3)s9Z^7oz@-m_btC!q4k3&&&qzOZmj(2w8uW36dxD}#~m$~!p3b;nb! zjLRpf1Q)2`eAnwy`2E~mTn>;aIw%x;j&MVqzv^myAHyCXyd{U7S}=d-VqUBqVz~V? z^8R)-Z1%8*Y8vpL2G1Z_5?p9jhwwMg4@7AE3;(MW=-&x_loN4>t#JQwIxa+~`{-^v9Qx^Qw+eXux|BFW01AZ5|95ZHw||Y;DTzBEW&gbGjFr&990aRpErp8B?Lhs~5<*ohhy>p9m))3~G?wf`V!F3V=lL?eBk_lVdlbTaGS^`8y zISxXEu;sQl&SOzeK^pv#P|SM|FpLm^5ij%ZLIF2Er~dsQPNvf!V_FOTko7v!%+tH8 zBbcP&B~qpQpW2LYn4fHeV-oUhu6L5;IF3$v7e6_;kw?H&B`3`yN-=Zxw>EV(hf_}A z=BX|uBFz%^uhtZ9Ci!3%R>4=8{QI%rubyXe);#)~_Fo9F#M@W)UsD-0{EV{8E&N zWpbYzj(?^M+ied=A%R{zTp&L>vBX)g$@)l}p<&rmaKIfj2 zxA{!Xy9Y?_Xt9Hfk1@$s*sshi?o(KX_grBTvx)IFck=FcqA#NCHw}9{)~+na;i4qZ zN3*h_(gWa0)`IA6jGJzW?d?d`wly;=n}!NiJpBjp0~eHsD&dATc&+!fAo=$5Vgn1W z1iB@4ps-ubkS6hzr|DcA97)NUV)d6uPRaBHU!j^dQ()3-f7lvIIX*wn2K~}crIKK} ze|o)@=6HQ#{KS+4c|E5MLqo%-o>c(FnjE|Z^bs>4uu4d{@|&;sEFgdy7Ng`P*_&L) zQBn~S+p)p)_=#qu1ieG@`#;2EsPtp3n%SuvjQR3mXTz9LE#A*42O=gdVz$&(QP_|) zdHwe@r{C9}i~Q6R8L@WL^Kl?jrVAN<$*=k;t^He#Qov_Iv90`Or+z$VOR-r`@94qT^Uj{}Kbbeyck+^Hz4Ry^)XrNMeB75GUr35Mw0r?VXV$wh++qqB(OmX3Ta zpAcg-2g?I=IQa6JaU{vy8J35@c*DYV7Lc?LkV0^DE}2y_EHsEWy=-vmz{kJ2PXCl#PUbM{tjfdr)2*`)@;eM=vC59jPn_{a%qU5_pE%3^1 z!!b(*y2_EWY!TmjYJJhooXpau8xW@#2Kj1jf9;w7pK%pxx&N=T|D`;nu779WStp^< z;0@f!5?dcSD8_~g5uk_uwv`_;PFYm=g!4RdFV0LdC{A?L5X0O(Hlo3RcDE;vk_)BB zkzNZ?s3|m9`b61Wu0WlLN9v97a=Y#rzdSz9c6r$-)xXJ6=Qmkl(f( zI}~h@HH$<}9;7+NNW&?L!wyCsaRz84qmkrxfz&4~6J3YfSgH~pPdn!(A z@bUDWFU)sFgUt~T1ByQRc((BRL-wpp+TdvoW79sRe=R|0a>i-zK#$RNhst&AaVsqg zOLU9p?O2Hp_Y zQ|KU61RE#hCLF0n3i_a3hiLG&>HgB5{OkTLRnl)VE@VwQR)(|T8-EAPSM zaX8knFce4mq12oU^WJgn|HBl^YDgng>wgj%(i;A6M24ljeLP6dmg)9bTT)Fi(q-j0 z1I2Q--9P-JB7V=q&rjKVAT{r4Db+qMc~kG`c@*(C>Pi7&K^(pCJ|+Bk{~Z%UYL8>d z+@Qtj_LQLzbanj@K8J82J*Qp|IK10O@+=7uU zv(>6Ck&VG1aA~+@h9l)`!LRGK#3Snk{=9ub$~N1b)-g)p+KyuMa3-2s(Nv;)%Ei-g z9v?TX_uS6n(OlETI>k`KFX7v4Qck7y%8)j{?OZ`G{_;7e+#&y9Z7_@=m>OPN1GVgI zXZ$;`*@AnA=YG}nsyyk#z^UYp6+%0%IU~+;Cpe0$AqT5dV15naH=B!)x5^SEKUkV z3~Kw`^&AWseJX&T8C(zw z8{VuSXA2w^7UC(Q_t(Y?Y%F3+U7k832Gllem<{G?c#e_G(OE_&dLu77$=8g`yS|OH zZ=W8Eig{k3Q&&HjZ*EcOi;(7w=6cohm_7&nL!|VyV(bJ&m_Dv3x5`!@FkBcOOOg&4 zP2bisEC|G86CZ_9Ozh0O^z(a;UJviY&E7UOyOhVHs3pz+1xkDm7h9RM9p4_oPr{pvC=N~j2}ie9iyc|=#l6K4r1CGrM=vEj3jHHIU; zgt*X%7I(JYI5`Cm4vj|n5`j@y44hmS9HioYL{lIDo$pd16l(99_w`96^IfsZ5v}ux zvO-M#keqHK6#ZVosOpsW*yG?LzE(HKVY)s@Ux%Hb4#J2(5=#PK^{q2zj z+t=%e67-Yjn$TiV*fL#MNfo8iw#niKkl>Ck_$ca-h#8KE8b@4n?%WKCV6mmU_GJE3 zf)8YzkN51=)LZ_OL|diI*%$J%wj5nq_m1J~6)oj|xAEBFK~9$mM}MBcedfTvh+IYu zFzE@k-JZ-t8>9vmfh7wMX^HH+RJ9Ih@GNtDsgHyF0&Yegsi(iRc6Y0hBx3+2&Ud`o znFtLap98O}=0Wu%~p}YR=`l^ptF+C|LV~lBYsF zM8s<=@7$Ltk$khxA{h4VB$;B`ZLG=ujzGhQ{9^LhW#4k~ICq-^e^ADnaUIkqj_Y{N8C0A_;rv)o%;hmaPbZ$p{GXyl`=-9 zLsb8G1^+r}rc~z@s!fLpsZ|QjYF!4GZ!0GNCLRYE4WU|ET0jAOc#CIcw=FCTOhimf zCpg!B{P^)}aIpASg;}IQc!TlfokY>zNLKC_CT924C^*Hf)jUG&eq9CPXVnNA^>dU& z4+t7KP?WZhKg;Xzv`n@(1tdL~63lVjARKXa?NhU% z6zqi0k50?`&$PInBWg)0o~)?e=>^Ve*k4w|`4ginl{2z;E`S{T#a;2%i%(2Vo*oxd z%(0%XF*qsBN5vXl+q9E+jcY4Pa8UIzoX4no+p!5?&GI@fcsImN(i6wWaK8d}!>Z5y z=f;^!_>MSD(7HGCI0G{JFM{ntx%D#`O687Q@WlRQxbO z_9`mQYP&hSNR1Z~>#V9h;g=>IlV}8RcWlR&3;Sp2e;YfTNmn%Xn zn6Ipusmr=*unO;W@f@5U37N}(T0cc9%>Ef4(I#pC0&)2HvS!_yz zdWD|WGbNmRMmTiC0JaM6XcBl;bqpXs$jRWDDlchVec0ZreSBR8Z^RUyWbnEypo#!- zgtX%)3ZhUH2CBeRDb{1u$=Y^{pb5Sl}M>< z;bKe3&#ZZQz2lopxf?knux&r@Ro{=mF*Cz74(|()$LU*`fZ|l{c*c)xjISBN!d@|7 zRdf?s>uAUo!H3KHHt`cI3jPPs4UW7y4-B+U&9<%gu^av8-*Lc7Vk~`Oo~xJd{!Hn~ z3k+&I9@)0s7www8ANj(F_yM9jn^vy{KSom2$cVJ{@qrsyCy43jLa3B--kT2L69T_6 z=*UU)x-lS)742lI~Ax_JSt6u8(V3y>_OoA)DnQ}(FBQ_yNS`bV)t`PV8sl;=c zky}(JIxl+Gd2M)$8C(9p5Rv*X8yOC{C)=07J~yyqmOw_C5;z4Y54Fc>q8gk7fdmAH zMQdt9<`i?m6Yc~-FiH6yZht_TMgFdE^rcgoeY?U%Detd`>bW2uEpaVpuUZ0Nob!eC zAInH%P#ebMt72-ijXYQ`Nu}RwD)V97uOi^}+`Sm5J3W~7&uNhf_A30cJ4=9Jee<^5 zMHC(DYmu{MblCC?KP#@ZCv3XgvAX-+q`42*2@KfxjIo(q3iTJTcV;PiW-e17^(K0H zREW6`RXbK04*ThRjz0ObaT<4Ze%KS^=zq?VV|a0)!>e<>zG~xNJLvVL!lIcPYzlhP z+ivy1u;{clWFq;=q7B+XAcOsVeLq2kR3516l$QPx82G!hldSt(tyr@YmH+OSX`1In z7#Gnp#2*44S=UGHs2X*488%D7ILsU%-v=%@gWgE8i2@ZwA|fKq8Y>Cl%bRUMPQ&<362*jky(<6(m{_%?E8Qc zygt{lKQ}NbrQp49H{>?#W?6r?`O&;fsI}oLtuCjS22LvWQ^|1ih&pC!eQ){eB@T^* zaxiS7z1{TAGR}X`8c|VVDvZH|4w{OhfIS9(JO-C^2*3$W_JO-hct zY*2!#vyq}SRFnhR2{WxUTx$qqYI(x+wOW`?j`EhW7X=`N7)-^(K<&++ z;3#3q{2A$#8_AozJDw+~q3yqa7g=jxRjGgMD=4i~A4QuTxoR^+o5JGj?hRY<5#kkU z8!t{J$L{ZJE3WsrkvEBJp+?_7@S12tEd~Q8as35Q)H*zRq~Awm1fY=Y|KZB&>^Ha= zeW3%*%3jde8ezWmqto{dz{v%Kklv5v@N7wfhOzyL^MOT9Jb(aDm&Y!ouym*BS7GR0 zS3!3|p70j6AO<^tjBQ3i_q$HfNP7#oF@tO%_6CD-Q)U3dyR`+c0l@X-eLyqpy)v-; z+7%cUU^+R$O1r>`m7pzaCQsQ>h2d=9%Wfp2J9PHU8= zk%wy^#k~^hs9HciWTvV`7KmL(8#COXKI+vc4V8+g)Cd>FBawVx;p9Ccx8P8sH@+&! ziD=h1Wr4p7=i2Z@aufWZcXr`#s=s|K`hcp_TNp(`@Yo{%Zo6*n-cni@rB}b`wAD&h zqFwFT=6ABMu7@Ao7yIKFo+_~z85uEXRD%IPK;P1`f2I9g@<8mXRnw*>*2#37*RoaSs;nE>%38Zp0)MMa{{z;xd|>0Esg>c3p`XLK1RUNUjK@H4<3!LsX?;@KbPb;l zeaBzrR-7lQ7*7RROtr=IdoM-(XPac6vN@GXj!)T~0j+mW3Z41md0Vnj=zqVHtJ2!7 z&UTG|0$Kl#ha2?vh%Wq>{`M;H-y$w|)7^D}|0Sax>v@O$XFc)C_OkZBlQ2L2O|`>b zqyFb8i+e97ay+Alt$L?ck^Wn8_tlizkmedNc-CwSSw;W&cSr;+ph#>^QF9I zEWnOf>Dh&eW`I^!QEjk3HX2=|9T;4NL{HHAi#^TD`V*lB9m4>C`6S@095E}p6#KK> zyD^>beEZ$V3|z!|?w16ZoyS-9cb+l2FGTlvWAg&UI+yU&<;V6CXUb?0ei|Ga1Q`S* ziev`h#&qcZ>{4%;BE7&wOjpcy32nT;w|Dyn7YmKpIzSUP95tV^cw5HKJuo;#-|$0T zaf)6<@&|a~pi;4)MK`aZ5`#=7Zq2KEH7cmbPYiLdyuJz*6W@rv6mOc&=w0F9owPW9 zYhc5>o@RUbEd9OOsFf;zpqDbC5fd@}(uQ2?x!E`ixoL%kY8&{acDO~uc1#d>OvtB7d_lElnX!G zF~_s=J>#6pukq9|Uo7N}eDS(%rRY>gLg*6@RIz()u4wI4-r!)J_XfSC$Mc?n9u}%F zR}w6%???hUs+skUVKcK2ABG7i`Y(#h56;Oxpebr{*$2gD*1O_C1!2mR?My3qAAnjW zZfn8@-xpC8Jf#jnVdh)o$l00_SA%eS*scC&_@R^CI=G{s?8HeWgrXAL4$SdBf7;+f zi1{WveydPpV+>_O)Y$P*&rh-=ZHb35xeM@U3{2Sw)@HBFlUl!4B}3{f#5@Pw-6Pgd zJ1dQeSVa~u5asj20jYGc4qFjTrtgH$)J-LlbcNlW@lAi+N+DrdbIfJY%!y&*tw(@j zG3m)B^u<*X*&2+sO1Ga+K*Rl)O3O3r7~tVu4EBmH(R=fy(OY}HXD>}=D6>e1D2VW8 z*>moXJ87xr$BCSz>cc>d`JuO8*;Zdu6`or+OwR1|q)1`+Q6(pjYPX7<+)0pcxfbXb z-XNvCe-)Du$US|?=)U({F*105I6}#Pq(aFNZS~BFeBI`yS1L0Sq}aMxgWw9evI*1f z^rV<)%Pr~w64vgv^8*6#LHWAPT)<8HRJ(yfz;y%);f}!rV)fu4w6XDN$MRDqi48Oe zFk8*MhoV+1u~J{IcQnNFi`u>qh4{Mi}+HNk>NRbQ?*xGRDZ3 z)ZgEcNDe~N?XDFF{RmIkoMLI>MFMC0gPhm0W~`|TbhuH1)= zMfx*De0|5~YfBaCPbO*p@RcE=^Lk(}I=|2t%=kn!kN%}J2k9*snUY1)N=NS-&q*QLZJs`i&K{emow@54%_K)8hL)r=?2#OM?nK_4**idSaE{NI`hhc&(Ig@{c z+cjKYuz??un=2R6-f!hF_+VS^z@v0|KB9SoFAe*U@e)0_&9iuaIY>m_;AGte`@VYb z{yydUmhgbDxroOo)_$h^oF0^h#O|oX$KPc*hSE^J6!gK(gyBh4 z9V!#Mol!8azpD{$VWa9lEyB0D8t{J4kF3$Rs=D^W^_Af5@nUe_8+tvY=>XoVjA4)1 zMw#`njn+}ib8INwGxP5tp|xB^FkErb5`Dc6ZQ)#Gt4NqoO}EJ_9(8lC9ck+RGYF-- zNU1fsACoFnBd%1CoS8N%uHeV4X=;`i`}b(q_7PFO-$jN*Q58+Y3A&Su>920Wl7F>j zQY`h|K2ztK85#9{#my5}(U8qv{w|{GY_PuAp-4SBHHkuTzrLoxUmP8u2E))AG?`fP ze;uP&$y+tioPS$c-EOh{ITB0uoxU>Br`H+kXC5Uo43E!DQd&scY&qZ!TZOPLUIy9R zIMTX~Dx)>Sg>UnIiJpAej*R-CQ9F@Et1-(xvr=osgAOaZ(ogRB5*gljrb7^L{uFY# zqP`EL%OyPnw5PXPxz&PFSS`zqflP0twm8S<-!^AFA_S@>QvX?B?*JP5^?vR04A`as ziDC<6SN9g`qd=VQsS*bWeVyPRXp|fH={M#=T=wTW4;C9az3w!VA!dW!y?x)6C&jK=qH^oG-ebZO;$K2WBJN;hwoVMHI(+^XJ1sXH!8OZX(_Hh z?zcbK<4m%4-JVU-RL_-y;QNy%IJi8B0~OW=KaE9RouatV-J=7?<^ z=rw;=JI%;4$ybt)+f_yn+uYM>i~@bwq!O!*0fn46p9ivdOz53@>gXpS-o$ahlDbWq+k{n{&Q6gl+0&GM}53SK{J{1R2=DPO`Sg0K943Is3%kLkZ(rg!G3}qT;x^5QCrL8k#2me6O6pVj=?x57b?c1KeCTn>lM+wP9q{_>mvBMB|kgD0KW zQ2(84*bou< zQPWgfV*=i~qmnB-#KYmhm{WGH?B$p&?=9_~81lPw--iXISbBH_n9v7i`?)Gi_rnM= zB@bz1wS})&Wv1QK?`UB`>%%}Qx2ioVks~B5$bXvP&dq-9NY9M*rck6onr^?##c&}_ zO2Q3aZlpemd?R8WOAPC_h&OzdkfgHg`RShPKqE5Anm62tAFT0m%Ln*yn)gK8-#4x{ z;=TD^j5)T7=*TlS7f2X>AG@w+kg!d(W)w#gGDIp(o1$_;*;;O+!mGQ1NMvrr)e=z9 z@g+oTJ;!$Pf+=r%edoST6m`;rz@6cqyy2pEhB$ZD5K^wyjBH#Deo*E7^A=80?4KA% z&4zHcSpAgXDTsyAHUg1?e$P$eNMn zcU}s)j@+>xY0mJ5>9lmwX2nGYF@jv@{W%6Y$iac^cNBs=y;eS;%YfFW<5BS&*C2M= zB~ohYKrW}%O#2GRp*Dehm&INYxP~RqSt>~B~HLauk3ZmSLMW~81z#jMcNgb3q@AH zriV&EI+eR6ChLg!vm~Fg=d;@=;Z0Q2szgy5|W-UDt(G%KT?|ITctC8{gi+< z*rF$|xP1GGu^c0jW2%Z)Y*hE~PDvwv)UcBhzaNuq#KpSC32w_)ze^lCz!O_EZkAlu zxMyG8nDbFNWKp1y>~!F;YXbuzdJ+=RLiQyZh9m;bfDyv>yNdSEG3QdPpDsVg*A#5()NKd4nwuHzvoXS*-f9Cln)M21jNK(SuLlG_GaSP zZ@66!UVOa2Ip3Qp_W`DSys3Tg0T`ra%2V1oIvQ^D#qfCD#sBuxyRWd6J=I{v9k-2i z#nCY$EhtbyF5U1$gxDT@km^^JJy`d>Pv$fe7WzTJtI5v_6+S-~S>Z$@QKO&Dbzt7E zh25*45%~=C!KyTqXO-x|&Pjh56mh(JOC4vDb+~vjr?DGyAMId$rdKgBg(S>QjOUoO^vGeb9EwFY6GOTHKZ0R zF59UdHYm`16l=XzKGWw3YAzPFC5T_LLtx&q3nM}BlL;jHGGyP-tp$5V2ralN=7!!o z_&zwo8XB9*Sc%YLp(L{|j5wXt@*iVg!{m&JiTo@9_gAVw3_+~4FJ}w&xsd= zU}4Z%ClJTZw^A?cSrQvBM>{*7hd13qiK^1GP$3FH7}FHaC7MA~5$WAjrJ*roPe4)| zEfJqjxN$Y~?A``H!bM19-WoGOj(#x~ES)0_(LQ_d2kp0cdD?v@H+;rUhU zRdSJ4I*Lfj`Uac2D$^GW8vADtGs#m6W_OL5;^Y_RzU%BxM?@5p42BkPym*hoIbJ4mH7sDYED>r8>*O9iAwa(kHWV;l#;uF8&FZZYNQ>#EM_=O|4t9 zPUB&9m`e3L_AiGm z(l6Cm%nQ@l^o^~17n3fLAb}iMxi<<5N;JJrBB14Z&61Fl`@^9U$~U{6s|{UnjSZ)Xs}$n~qTMu-&~dwv-z#7A z_cGz(;kR7-pP^IyNMmdtF^73?IrD0sg1{@m4@Nc3TXt@xo!oe~;>xtEA;hF;-i-Xy z@urVv43%mPHk7q{gTq7k@diXl)(;!+Pl(JrJg7ux^DZB)cch(UVV8F#%eRE zQ;YNV23u+5#{Eqm&qT8e{2MU>{Av~717HDH|n0I zNKb0jK4J#NO=J?|whxMGa?mneQ4s^!EOD|KMXGSL=V|#Tnf}{C53E%Ex5+!6jvmJ- zQ}r@NDf1#zE(wyyns2DivWg29E8oj`HWEcqdA2 zDX12hn2k?bb7yjfUuaA(x?oIDlSwKFpVWLLLSk1( z-sXtjtu9n3NajX@bPdp_QBu+VHDmsi#_T={0HrS<|e-RS8x zIPXh>=le@|__nsTub#K5o~MXNNJgE3Z;)_VC@=fbG=d3v69HGU%w!N3P`^YS9GJYk zyz1)f6`DRDv!5`3u(u~2NDEs$+qn_{=If8fZ;H~_7gK(IL2mL3igs}^f}2x+Dhu93 zCRJuL?EHk->CsDBrEvAC9W`A(yulQ8$2)O~bAXzUD;4U{_DFZ5&>LUQj-$T|P~3z< z0+CYqlgn)~>+K)7=AmmC9U=Ht37Q7PhVkrXC{Wz0Hl`DijI_K}_(n`M^os)o=-fDp zLikjLSRYy4xInLvxK43$(9b(Kb(Ax5``-dpwI$&f^W;So3^2u*I(vqZSl7?O8|+~) z@5Kx%?Vvw!F_k{Z+>{Ku^2z3p!sFvK!cN===;P#1nHA@k({ z*PY+Ic zesA~ip1i@;9&(553~SYG3!w!^V3pWVoyUw2s`c?RlqI*N$)U~wH0SSDlDlw>(EavzT&2akrqd1}p6ts!#T|ciY@g@jS@xvxit+2suwcecN7iDFA9s`(- zpi+u2LIGzS!4on;5nvGGdKr;Li+%ljzglfLLjUl`Mvv>;xObf*kr$^-f9)QZOsPiU z_l_T8<(|F`RWv;O-V(+eE|nrNPObr>9 zoC>`(TZXL84G;DvuQRe<3sZjNNXLvvGwc7;Z9~4t`R$c!TOAl)xzY$XzTs*e9*ME|bOvUFQPf3%ZXjw`KJv=@0 z)k*=nj1cNcG6PFBV9)!iT9^$ua-ZGYekvzU5ry?I#jJQTuD9$rhJIYX^|#%c4Jb7n z??q!^I76u_cY8cgOi8tPC?Ubt_=EpgLb7fq+HuPcXJfE+=X$Z{{q;KfvedEhPZay* zI-U}8@ui84*$|D|THCnGf%JB{3&GbVW1giRFLqeCmZlN+KiRxZfJM_%1{@Cdna4(u zE$(6{P*dLLL}XFTHJ|yvNPFwBs@A{jciD;{QqmwDA}L*pbVzqeH%K>#lt_1Xw{(Ma zcPvV}yPLJn&EDJnd!KXO=Q{6m&L3XtqSl(+YtDO)@g1L0T56{RCi5&bDxteGRoS4l zoUZTR)}_}ltG~ljeD@CL8<~6C*8dn@Ei>eTYLEVkogqB^?e5YiAMocxkG^49V*KTz zQND(4yMKLe^XC(R!DX%1E_d&*FMF8!{w2vjfWAjw)$IQi61_x#Ff9G;{Ybt49$)s* zljz)R-u5Ahbsyv1UmxKH;Fw@-CZ{?=QLI)d6q~l63X+TN&yTsmpJv+w2Mev&rx85H zb3MMKE<{Nmx0i4OzToiEWDOke3@C-hJ_ldW*~AZ5D2f-?Q=ml8%*1bO81_WI=>$aT zlG4(i5>qf=uU>1T1!#%Y)z$T1nzZx2{^k1d-Ai@;2tS$uYi1L)2|js&mjYbV0x++_oi9OB^M@MYnb2VO4# z{viYF63DqXqSvov|4Hw`mRU?E!PL;#;Uln0nY7!!L#Gu@O=%`3Ccuf)TtAdp01W5R zQ)A-b^!Om*PX5$-0PueZ!II6D z14F64V3iH$%2K#p@3O|Rn8g)}gR9x)@gOf!F%T;IfePCG=OB1hnQV>a-T?d`xS&}s zwmbqs3Z!4)Zs(4xEIwEKpQVP-L-?9fI_)7YtNA;XIq;nDde`MxDAN8h5~beP=f;;I zATY@Pv1DJiegAWuh(zGKD2l_Dt`f=R1s^4od2+@?zgavkYXRG5kzB!5wQIW+bvnfm z(iYEF!5S4RtNq_S~>G*g{c#rzT#lw);P$LWFohe72a5%8NZOQV~NLrbz}fZselbRvlwa>S{UWh%&mZ_k6lE z$LgKHGf{8inRxJoST?t--agpt(en@)nbJrGi_youbWIQ6i;9t0URp6){JtNF5RXD4 z1uY7Uh!een!-S5mPD7BJVzw#@dKDWrQ)Tk2-Isa9>Z8NLBEn2AAMfHUH0a_HjPo$7 zusUqL7L9m25K)|U+tm9|s${kE-T5Ud@EBKs$9SkLCP;gVpR(N2%>Uc%*A(+L1XN&^ z#}5aOoSxcl?-63MGhVL)pAAJM>>$1&=ErL9fqd}b+=er^eEuqh&G}}aXd&^6{>yx8 zqtTgD+jNQIGoN{Plce%Z9fivUwXB%b1X(JIqT~i2_*Zsv1DWx{V7>VdSI2Z41l?lm zB$+(+Fx#CUvGJV3sI>=DVNsVP?VUtrC4z=cy&S62qc98!i#fjN& z)iDE;dZamOf#xjmlP zqQ-7X!L6}-&%OZg$2J!CyU#yH(Pu|u4kHt9a?Lq}HOXFT+ayhfkGrld?XUHwu6mqk z?ml|IaDsGEcX{`Sz2~b`5$I1N(>-_IYdu{%yU9R8$(sl&{_{;wmUrmeA3RGVr6Z`l zE3zppd3f3ovECnb1Wpddq{3sJ&iEI1#804_XiSDLt%QVVYHb*RVQBZlEr7dz^u&C`8=)UT5&|6m7dtJw$eJA$_yP`6SeiUW*-?nf zzZ|L<0_<^Hapo~dDM*iRSD48S!ltIRxTmAr*u(vBXK~dXD5Ery1{4IP4tHHICAWJXuj7kpu)jq%lS;U1mhJ!{7Ut=Od@uj{XS>sHOW22c@Ugcv^rcs z2$ABz;4{zSi;lU@na<+h zXOwb3Ojm3B@w-a)&yF^nes;(9ZA}PRACpfD1w(|VDxM!t6qZGXzQaMixgBob5ZY$0 zGAFYC0?9;`U+u}j=SINBel3!b%;o3Ew5)SmYs^~MUno}f3~8k5xiiaNsvhRD-MwFZ zvjK;_^_O-&mQT{;+GSjLR+!vW;?9R#&l%9(*JLq2t#b*Ue0`goe&al<#bYyABT@9n z#Nsw6dmBxqn?utll_O6%CW=l?rh+!uq*$WYNlA&pN%0$J)KBh{2 z22-(uXZuP8^uZsK6LtUR7E^43h;qTxI2wvZ@*4&O+&3!4GzzPY48Y zlZD&Scx4V($lo<uT;db&!00aicHeU@SUit7`pONeNKQl6~)?x9&D4vx9NBG z;?NR_C{h_YY1~ z%(IW6bx$2PMC#3Bfn!}S1>Ib>!npLpIN4JJ ze=ych2nY~WRSN1GU79L8uwfevU}?-C;tGNd#FN-q!y?VjdV%MvP8AIorS;O&Ah6OE|A_e0MjA0vv2AdA zh^jJ4nM2Vhql%m@)~q2PvmBlRk*h`5!HLasrQmq4+aAIpwVLl|!=7AvPPiGYCz>Cm z7dVT+lqYC;Oa!%+!=-v+;GAi2V$go%ua9Gwd)u2UttZyIs4Ux#-~QDewi9HI33yw#PVumA|K%a*!qKM##X zwuTh(jjZ(9q;di(lJGt+bUuas)bA$hqorIFRfF_ZI<~MO`92olRT17+RI>2rJ0!_%al4l4=Zeu0WwMRZ$fm7J)I5GlQRok9P47m#h7o_k0pD^ zqNiH|+};SvRu?d-x-|7y7F|Nk@iZ!)i$r&Ja761wE?*k_zZz*(Pb60-6pSjSO^Z3t zHotz#UmcqGD5%%)DoB4Fx$aq`-qrV$>!fEILq3gyt`>~bo)IzpQErT0tEWPr%`F0Z z-&fuIxX=h`Sle%U?mNKvQ-QGZ?Nm!!plsP~qEi(t@u^zl+x=K86ErI}_&bA7s57mJ z5y?;;PHU}7YgVI0Q|;P1lCsL?w7SC+-o;y)1&e_^jfUFA)QaxzZXiLF0+kT3Z*Icg zYD%yc6O5S^Cg4)KIJ3(oc9$uP0Dq#)rR84Jl6CCQcmH zjr$we1Gc%qw@nN#`|>#}RJS>;qgY;f%_1@sSbwo2rSjF0FArck6F{)s?k?dt>qWU5 z#14o|WG+D|QDgfuL4OHfv2b5vIP!-MJPLswFxXYR!O@7vuzrsg??p>?8`;C=KH-(I z-At+u1tl)Ah*V71bhFUIOhx1^E_Q`MX#Q+a)oW~P1C|7ODeYtQbsmEl(K#ygb}(8@gJ_Kn*7l&#KgzL(W4*mZ9_kptYn zyec01#66z{zdM`*hiyFraT(S8J#I=byk(kCnG}x`Byll3O0A-f2TSN{+a3zL+rhel zbf?kf(YLJ#5r(@GDzpN_XeOmVF19blS(srsBucd7kR%6p%{6Zz{!G(7$i!p6T3=63 zPXG)!UCzT{(H7UUz=)}&nHbD01!$YR&R)w#qhy^#bSjEo<>o+ zaDYUHQM5i+H3pxI8u8wPy3blV-L+Kn+JCz-CAwZs-IrmKu)y$rh@9S!^k*+JrR-Tz zgA*JteIzr~oBRV!&}0TQ9%tU)R-nz$s5kn{DJgR;fdz@H3URIxs{OE@(gtJ0!+8dr zFb-X=N6mA&FI%IX8HUIG+a~#O?2U}s6b#Pc!B_Y=mNv)xh9gWlqZ8iF(Gu}Zc(?0% z)DruF0_+Ad`@XVi7Px6I)ErhB734)1$A-OoHR5YxpXaGK{Za-N8U0Y5?M~GZwKFo zmy!`Q2|sHpW&p?PB!wQ~+*s1(g`<^hjt35#=+wsSX&~ zRVcQJkg5wZO^UFxLE5(#qp>n(Mfax9<$;-r7$lXSR%g$6*$?_(q=TL)Ir5dG-*}w$ z#cytJ91DCAKL8B|Z1R5Y%*LL`W|u~WPE9Tx^s{CVu)RO>L7JL~Wj6h;?D@D&BF?L? z<$nGkFNNc_LU`#-wIK;F?TEWym38^@%)Fv>y{RRiC*S5!Z{Vo!CtNe8(ow@!IOf~U zwklsmrgE?ib{gB%Q-ZD?PrS1#d2DX8Z*6o1~^tk(^Yngc-r@vHo6#Lp#L+OtA#gPAnwY=C+${; zfSZ=CZjaF32CRJ!AMh)U(={5smscaVJ`FEg5##BfUoY9>MRzr0GZI>mOcE6z)a0(1N#CT$hD5kZByjQj=n~9guQ{LoW#>KEGA8(A)?^zLsaF z%HJJNMcD=kL_)5=W115(7$6ByAg9iHSbWGQxjYd_gRb48egd=G^G`sV6h0_|vM z+}Y)PB`X=4>6q~AR60NBFcH%#GHh(&OA(T$N59Z&Zmk3Tx=P?+1>&z;ftg&cbCWGX z)$ih&udU9p`B4~Y0q_H<}ic{m;fc*_#o!4T)1>iQ#a&=74F7D!KUf+fTU&q_@*{#DaInj}iEzbxV z(EZ~usJPv+s?<8`zg!EIxuuLH5XoJ-Vhf`lJ{d$ky)_Rn)r~|?{53e2e`!391vdJ_ z%_ycvBknwz#u3>wHy^-F`$&!=E&>Lfc%Yqa?(p7}O%MjRekf`*e`x7I8&^Qju9=2E zI;K#|&E=#-a*%i?A1;pQO#(Y5sOy@TiI2JW;HQnf5E#2q4kwRccC=%XD2T?)y*^LR zQ^DvYbc1+*8{>8sh93`dt^WuPG8mpD6NlNMibjV0T%f<{qu@ zDGvGH`7bC3+RQz}TCn%P+jLpvc`-Q2KZEUx?b1dll1;KP#{8+c$2rkGnUue?tjC#M zBi{wF39i6pG};Tz`L%G_iVe?5ykh>$DO)wb7oSVQ!uS~e!5Zn>l2$Pw$2YX5q2DBQ zX<1l>13CT~)B8*?EBie|z*Oj!){F;71eWX4a54|2?aerWx3po1| zIl+z=bYU)y4wXCUJ~=aP_Yg2-Oi6QbF+YlGH|kzz=;2=l?E3>zc0>PVZ$lFiGu65& zgw|}E$;l7%H}4JcjMW>}H*2QxomTr6)z+7v8g)5QPGVNG_rp@hrMrjYatrgc3kNh# zVXg^nH_KJRJze~?;fv5Tgz*H^bWf)o}`0E2uv6y(2s0PcKEUEc-fed*C#!q zGA%ktdKLw9FW^tH!A3AA&c!Efgd!vf@hdIMw*Xg@ z52^g)H)7~yZ#>EIcN=;Q2X;JME(GW5C3Na<1yfjLEsv+0#Icw?Gct730TZpTYs`^n za#AUMbn^h0Gd@4s9BX1`Ql(7q-2Ngct|QW)HK3Ap{93NK)Oe`pk>vcnoc^)d8q44s zom&wRQ3h&Ommd^wIao*-KFvP{(`dgUh?tJaJkxq8a%Rr?v*c2rKPNZ4Mi3sX;g*ui z_q~iJ5lNcf?!Hf^zxO~&E+;4=$>*%&5#tLBZ?3uSf}9J6tyh{~WxQ#Ly-v8dUz0eK z+inw3=1qJQ9fvYeaRi)bJ}3nsaI{!YrcG7MRjNVRp6Sz)4ThJoqTVbQ1l{?s-W#6{ z_7=D*_D$P8epZeWEUet{l8Wd>e#vMkXSBl^69;>Gjho-fWjJ6!DfCKi?2Y0^st9!T zzgD<@xH6elwY@DEud(|f>?!gKgTQ@J)U;*nvYsdFjO~c438L(lw>!)?Px`v8D`$nn zyykb>pY&E8#Ioj$9ol`djRN)L59#2w#ENo1l)VmL6}mDqzEa$po8+ZPB(6LJ8P_AV7u zWsVP__>}PL_iucAdwW3C6NqOq`(kfz0Bk&nuDBci83not9tkXnl82 zJ_(yBRK0{9cNLg1?6b#>?XQP%&7i5j#n#IS^zj^%p4PEv z^x^u@9--*tU$2O<02e@z7O%g)*&2JgCHx@d%hknN0#b_7)RN2UIWuDo&CHRc%2=~I zR9jSzGM26eqAkM|kH`$!o0d zlth7tx2ycq6sNTjM=1V^fa|H7*Lehr;A-F7zNw$Ck4&O1QmlIk9@41vE{&HVZpT-d zw-q;8WVmFOIu1nqwe(Bvf9i~A6Otpv37%)UK0|t;*VX26m)%(ZnDru2psCVx18OJL zXsc3E+P-tVm|=E8p||+dh5;p}2B)eW!q}-DW4jSwSSV90t?KSL5EWndf#`&g-NREzvcBZpL1VGyMS*fzi23iwpseos-aN9_1dV38 z74L2N158Xs-~$_t<)AGtN`nACnQHcUZ*RC@;8sJ$jw8`u5tk|s`Pan6-!6~WkAK_G zmzY3lL8n6t;Nh`Ztvr>8d#44qy?1FBu&kLtQ=#47Y&VD)LqKyY8?*7)XHXlJk+s!i zYd%vM2m&#TRx83_m$mD5r&MM!^Jy7cQ3;V6-K9LYfgaLe`WfFWiMSh8n5)8u{BeDBX;jO6BN5} z`$C?g@lb_NoL{_gk_7DO$XX8^2_9>Px5>6Bq=D;tklxW*Mz{?BX)INE%U`r{g z_z@E&06n`HGd3E|uiU;`!za2TAewrU0U}YiQubH@eOWIdR@o!Eb3Dk?;_r!%_XN-* zjql|>Xx}p zsY4f~|19`CXu%6s+TTlyw?dGct7Y10J2O4GPninRo2{ZJST~Qh-8Fl6m`5p9W~nq8 zDc`sT>^0+OY|umA8N5g-7;mdi5e;;#lFryyPBts4=YPaVm*^#qB#ash9Y`%0ek~b` zzG{+!)DVoG&zsE%=lE87Wxbxn=Qmq}n0I9!gPagJl}!>l#et*Hi|3ziJme98*0cts z+q^j2;kgDsS~a+1CgHTVa`?nlkKz7raxGm+;lDKVk6Jc<&!+4%d#ga@e2r) zng5pb(F0L>P~E6spjx$wl!bBBtGZj=(b17MY$z-?s`M;TKYl!t zC4p-^QP2(M59V#;NUW|~HnfuyPh@6wbYCfaHV-XKaIDhOz1fG<4yEkOU{c+)dbIZ7 z_9tEw-tnOCiqqzvEe5SdOs6tVM(2N`B|Afot!xLQ_r94w#4e1VGoY@5HOA7#CLu_Q zrsq`DELMb9V2fa@gfXkMmM&DImg=NBqPZ ztBj(&8<(c0knMWL=@!ZBM7v-gtc-4<5sutwZem3qD?w!AH`qui>{rqSSZ}vGG2H<9 z#9i&THddT*`|i2=g)2C_J$jGpM!5RK_PyR?t4BT5ry-yp1FTW)c7N0}%9aBTu6L-o zs<;J(Almgtu>?^%bhuBQ!+ZXP#Zo7JK2UmnPBmTG6WnRl2CS3T8zfJu-()X+0`0Fj z8baOzT4pAI3Zz$QQy!e>mLLSNMy@ znVa?afZzJ;W$i(VX>SEGOyE+Q#1&tTQrd0yVXRpy5d3F2h{?D(vopM~bl0L%+@eJE zWuKZ2*)j&f_@<{tHVA&V@AxxZMN%{M&2o#(H*^@{NND^O`?x7c&-D-D(>FqTHN{I; zt|DiPt5fBK+hfe!eWBxCIldKTzM-8qoWKdLmU#Xcu-NDKn~6ZK!}Vq&m>s4tMOPzb09dSG7Kg39t$shZ%p%^-DcUJ0&Y zg>KU2@9G3Tk5VV@k*bFwrJqS<_-OQh2hcUdy?VQDTD$&<4?1VQ!B~NJN#^TfA#A?> zjaY0=JTXYNdD!Jj>zNy@l;F2Ib2(qt_F}u4fw~MalYAS+#qEJ#37A`x8RV7VG+ZrSjz+2M&n$WgB`#lX^F9E#=h@Tzx{qTm|rE6_IHrN`MY9{Y|qv1hisDL7U^{DIGv_6e}}K_Lp4f>F+XT8 z$Yu2P37A$#+(ebP9M#Y_RbLUebKz6X`PWNH>z)bFZ?lE;J#Qir;Ok*-8%0ZsnD zMLLqZ|BiITxYt;*!;^}IB-Pn$qC6_DewuZ^A~#Qn+g-QXp7;B;I&<`P=9;wXSR%i_ z1__K~J#jw6Bc>E~pQDNtWidzmHRH^&xERWAPEzkach#>vnk{LtF_fCXI&3~nmDrSXoj-+a+gGckjSz?Y(?TFq+*irkrHNE?Nld_ zP;MIuDD^JLSu7x=GL>$8$`mxWpyfinieC_LcFAGvZJxKN_-4M@2v8zO_MiLjrncJ%xSMrSHfqK>^z5o@&;~Hz_R+HEYZW$FuU41T9 zh3bb#Eo8%UeT>?~;hUT}YYK+2EBt;xdw z?B9nHQI&b0p5K;dN&AOT0XRT=jzESh?{m^;{fFwhJHH`cOZ7=IHV(M-M%snaX<$1E zVNAo~xGf1}8-kHqxfhN1i zgac2J!5yXx`SI%QH@?(gv$Fw2A3K_wlZh?<&U+LFRTFU91%Syw>2gEqJI)$_u6+YM zo0+O6ha!W|pI-wl1{u^M-}QubtsYhJY~?jvBL85F48@>&u_2_$RGiSQK&}^xNpA<6SZXZxi#qUHE=8xH(*14Pz+=4A(At5_CdcGju)9+E{J&vpbO^a-;g#@NI5rdu^X~gnK+ymQ-7vm@jhu%jI%N7xr^^`JIV+! zM{lWJ&qs0@I4AeTluUj}mmkgA8_QSpN5JU0PygjJ-PsO%#IoJlnVnpYRMcoToRNmp z`OJb^p}=BtzSfqv#PxdbymnM2#LLHLq|88ay2@fOgy^I2Cl)tq36SFhL)d9z(X`$Z@Q;WFZN_iTrvy-($L!V}!eYVQ)0Z>zr+mU$K6vUoO~zuxuJ zzRhZdE2K9>n^@%9QMpmqAE?Vy)fQMyPWOSjd=Nq(6{LaqzC8{Dn?-ZJ*%cFyW?bAv z9%=tpv1@FT(#WK_kA=|cEwKl-`|M{ER!jud*pkUYa%*_KK%St#3Xwsmi9}y%Ut}Nh z=J2xw>@pv66G39dHzEi>R@~DSwyh^IYk;NV}a|%BjrODWkVhl2CFF2+T=?w+6tjXf0 znorKV=m_hjG3QALjO?+Bk|_;3v)JzH!@)Y(yjD2(=C8L2JJC);EI*#6qcZap^(2tV zZh&`3FiHC14pY?TbC;6x^4BCETSMHIyMzZFx;a<+@3gTW{y{oH(Y$37;vX|*jwsn8 zV$;{|NoB4*xuC@p0`fq$`VEI4;LFnX@|>=;7L&xlg~p8YrI2 zkBR?jdP1*DOzziCS!#wG)0y{-fd<>RG#fIS`Y7V?Zm??%IP2N0_ZAg-?Rt*W$Ajx9 zp30AjD{?h34w4QZe``U2RzM`c4JdWNG46|*#p}-)GT^nu7W>zz|4Vp+!T0|>JP})k zvbj$7uj((A7wPrHw@kZ_6%tV_?a_Sf%QR<($U3OG4*j|-{+r{uSys?w7uL;_e&(Id z8a0MA0vNpEApr(Pf69mD$$Cd-O;jk&av zUsiuZc`ThRnfwXFn7~fv0cZpK48o}n&dvsJz)cp1wvNu_DOB^WwO!NwmKn5-4!YtX zgMbFB^}{>;tjX;bF;U_x*u+ZbE4*#I-s1#A^bnvRG=^o#?3@R#JbwTeliIeL?Z(+9Q$f&3R?$OiJ)7mE2PJ>Mo zhVMYK;0tKa0==tRT3RPt z_}A-Lq5Ja;q~6~^*QpQK4}W#&QD{ z7Eh*6tsaRGbFn4JH{NR#>eMka=wF!nNU2(g2cqV9zwKgNx6jw8!b|O3X`GPm{0O8* zp`)HhA#q`a0kI~)C9nl9!6I-8Du7Gysi`QDb87-WB)7k--Z?%Z^7(7o@N99&scH@V zq&i%6V2IVb$r@Rs?nR*x4g8iEJ64679aCE+`bsH2A933rZXFy5uQPH;jEMT6%ZKo!p>E00 zEOh;RFwYKHAd90y@2VK`5}>H)XQ@MhO$@0WCq3SQALF%o2QRn6J=(i)zL%x17{%Ux0c}M zgDHbzU9AU3LTY|RCK8JwU1UqMIm$kgC?yg=zVjIMm;M6i2Uy>r89gG;v{DJJ1W5q&5F1GKZr^U$R=fj z%~*F|>@owBto3S-1Q&M$pzil5WEl)fVzXO)U93nAP$bcsZ-Znj8mLB$?kHcN0rEQTk;1WT7l7yz#9qDUN<}-bB z0#XS=v6czm>8o0AG>zi7)bJ-H;k@}_ zgsd0eqxRlz%+H0p6+ZrKWaihYs4g*(L8&WNQlsPWPMhU?2AS{pbpr8nBO`Tok<V8FQHQPR zQMjM1i)Ej>aw&kw#sv*Q=hz3PjhP*j7QPm&cd710t6~vvnnSuhLkx`2q9m2Of7vA= zny&k4!gF;q4Cw90?uRq|e;|>?C#+J%7UEB&yNV<*#a1^r414=(U&@~a)ub&wiE(xj zX{pu%nF<86jgK_W+rgOv`>+iHAr#)7H%mgu9kc!TdXdP-3He{No=g-d3s;$Y<|>yx zI^XxDH;DpFzGSy)4}bqRw6sNl5a@SsK-m!x8&){FwYn;jE_iyBuqi)_gpdmFdWOg4 zv{YC>Kl55-1ou)=s{W9@=LKnC}@v2o%s4kwSt=7Hh^ z8CN;9)10is-rVfAnviarz} zUY%*zy>aXU@0@WH1>g8+^TUHi9;gNHUYZ0{fjx?uiM7>Y>Y;+TW~5n-8j+A$R!83~ zVy39yw77&Ih>64&;6TZDd(7T7r$XAy=KRgpUkR=vobCv_EbNo+xfi9%c7>|Ok;vws zz~w|UxKl%>=YNfX+ z09Fwnc6Ui>Bs8LR_>$pKEgL@nwSG!n0+*303Y@({gMRtRxL#Hpd2};3m_G#$AFx_( zx-abwM?KUAn?n12R~%s77*;?91fyBtTpvMA7RME^5DzW>*m0_=%QFN$&cLtxX`0-S zB!-Qnt^bw}6aT^cj=&j8JbVC@yVb(5p~^*9=I~96hidccKV@X90Ld69!eIH|`pI9m zN6ZMgjnM31)$s~+hG^t(aR+-g=hteRET4D$H0yt{r_lsFifDb?8p}g+raZXR0+Cov>)cPvi}ZfnB~qkqr{2^MkKh(jiLWfd?G49DQy3j zr?z(MUw`1{u#B$=625r~pnCpz9jW(DkZ<)VhW9tf<;vUz(39O^*sc@M%A?6te} zkSe;ltXQGvK%WCagm+)ego(2GK$;4}j_;hU=RgG$-@C5|hlVEkd4_;|F@v2+Wom&+ zcmVI&gA_^y)S5)@3Ag7zm2dFZG*5b?1qX5+P>4SQm=(g`Fsr+;!}6;DEoc@upj6P! z95qS8TD_OuOLk}V-L4IA!SvtGba|)YwPGfX!b6G4vO!5S7?yV+hE@JfvSwR=Jn0UX zak;C@f>!yVpB&dA35KLO9~z$E2;sW;z_DEDl7K@38FU$QEFa}q@aKNE>ui(dykR21 zs`1Xdx!5mpE)jHE?#@fVeM)YAzcy*M0`sQ=B=eSE!^ z@#V#O^6K+}rdr^>xUMA<7`&MT(8>DxeXDZxaW*W}^+;Qz95p?D(v@ z=`G1O1b5~M2bq172Y;Y9!#i{3&>RVeJPrR^HVitxh4ZG^geSjpz0gpU83ooxK4H*q zb!le9^jBvqPRyCU=DTYeG$WTGZFioWEN*E+fhvdA$s8I<4>$B@f-jo_El-=|vjsx=lBl~F06=yU*=A9j7;>`NZqui7c zqJ~faAyakC`wF7o0Gjh~0g%AK1_nEFYYMLhK>79JsekHp2C1E&Ha{W!vY*t*usvPz z2HX#__AOrngyr0%;hT`%FnNVqwNJmSeFsL073eByr}|a`M?<@H3zAQ{g;D*_-s;`N z`KfVOhKdJTQY^@K>y{=cuFzLWTknw`YSx5b^*R+$TziT?7A5Nm-U=x<#PQ?t44s+O z9wymuV$&;jD5yQz4PcS&Z7h5e&{OL=m86}SxT&&jT{{8$_A`XuN8+&WV^%_N55p_zV)}zi8l0=^s0~rfXWSEj~S-I2H(` z`|g0%{FA-)DB*2YJOrH`uNNP4Z>Zi_L=?T}5>r`Y_3U!h4wlGk69ubqI_?eYzc5Si z5hq_h9{w%--Opfw!u7BK&OC9(VuQDl6fMP0H%JuTH;dOYp3g{TXjOs7Z={z_*?1t|A zu6L5Es&JzX@(v^;V)aDY#!JK(2@oUYX6Bw@8PNNorP-Dgk7*A1auVH&n8frDmM^D0{PW59p=oM9EkABj#iw{^uAH95rCE5*5iY; zgp2)#svI?j{+e+k=gC{128xlf!e>oS2#8L*`(^9gr+YV)NyK{|r#uU6sn^$QHPIVx zFs6eC-?%rr>i&FY{;;ZUt|qJ2fCv*xN}BpX+OKb*x5M4B9<@L?>J>X z$NRyWLG&Blmx{%%u8qmHQQ{KhcN>{mw}>lqq^UByY>~@yl+YOYVr_>)N*OFOm-{j* zV2et*#lP5I)`f^bWZb3HijH#DShq-jw|V^HESBGf{~g%P$`#6cQ$5IPm^gI& z0|HCZ`wIj%C}`_{2Lki8%hMh|k}FW~c)(feVieKc)z?=u8ihHX^E4t-b0iD%jYoaR zgIMV4Df#WWs}j;NpT3!Asf-?@=N=%?d{m*v?hH;bH{_C7kr$6$67YRt2KiZ0vA8vE zz^WBRxd>`R@TjQR=lQRu3=*+8?eZ2+DU!!BOs8b}l6kSerV03`NKw+Wvg&k&9Fni1 z!Q$scJIAM|gWKHW@Gae$D+q+9}Ah8^A#`bKY4qUxX-O=n^wn|p_# zM}NQmMXNNaRBGN+HHlG~ZF_D}O{rY;5P6@srNN%!6TwyTFpDQA4=#oct#ZVvP*b|Q>TTfDs^S|l*(!3p=|fJH z!l108z#7;&Qh4b-(&QUl3NR$vgAO>#qN#i4qy+58Ju}(=&xp+B5t8*!qeF6*w-bYh zcLAN%)JSyDA~}>IHS1JkxfWts#E{sD1)I;!K+AX2Y=5fgD%AKpdN)TcpONWt-}@1f z@{Q?lV(>a$naTnjIhu~N2kvNiiSMz6R4bFOqu1mEnsmuCvd{owQ7F@Y<&wK>x)vOF z9(K4X{;rbE7wiE&e*h2=oJrbcckgn@^JxfwkRBb}K4be1WVqk$jRb^Htk2)(pBql5 zqCh`sen^#*%9h|ZhwEJ7FBlone&+2#)5}HG<#F&j%}x~NOlnB#H8cBiWyjF%{37h9 zxlW}mLUxPN>At|&0)udzxT%wi;MP9KW`py0KH)3#1z&I6ll@i0L3>}QRGQi|g5U1^ z*>m{pwg=_IEFU>}uB}wcb*0_dN~1h>lhloJHO@bptJms{)wsx;+`4}*gd;=Ht;*N4 z)8}RLz*o<@^Z$hi+lUxu;Yjp?_Jyi(*n9tcvgk4mO{`{RzkRy^<frS`~g58h<$oGQ4pzAK>?ou!U3hK;aN8K#NB%{F*W~JaaS5o z<=U^QT}r8hWQatPlqsZb*kO|rQf9HslsRLTNE1S)WtOQZ37Hd-iVTr?j7Y{>q)b^W z!?~V!@BO~}oKNR>z8s(Y{FY}u_kI7T`#)Tl`RG=B`Dn~|E7G-X{ii<)O~~jT`rM zSo4i_9vG6lB5>@hD9^i)xI+K?kpbFYUzVhHMtIZqwAz&rol=Vvo-ucfUsrdsoPHd2 zE8te)rJ@xbmj?;aJ7s;-@7+we>yY(S{Nk70dfX=0(<7gGtZ-7sbDCYH8SGpiUg@a4 zGQ07x(XPF_vg7A%l%L24S<#%Vsg`kEbbPJqtoixBJ@9lb3`4WC;pd7w`B^n-XAcB! z*s|}BMjg7h-H+(4Cd!1MSq@5hfaF}IU%=F^H;5DtqN9j>|mrqG7& zv~q>|mj0UsD&j$oCQOX$KKq^%SSfAF**{Qg&c?p0_D24u_?0P(C}FmHukJPO%JJRv zl0i!GW#8ZdV+JXiy$#X3m(Jg+mn{ei2{j*!qc$2!?n36YDplL+PTT5|uC1Qcr#()z zB9508#y<=vOYZn_josI-3)~!hVWuUXN**HVcyp_t96v949$ZoruX!ShE^Y}5iq7X5 zJ+*Y3-oo^`#+R;BJx%8rj~* z&6<4DtZj!a+mEk)EtRVm=@94CKh~Xfc6MHus^lkeD?p8G^Qm*~F;C|Bx%K;`<-8Ze z6?|xy<|a;)BO`5k^7yj!a~~prUZl}BKIB$*3P(Y_oUuNAAWeC$fhXmnsI&pF7bxx=iYz zIymM0a4O2KW*)vPKgjTBgQmMEj8NE<_akFucEP~M2VsHy4kzo>3ns?Gi?9T50mt|o(Thx7g_j_~FF zXH4XDcycLC@E;Wew$w_OdBJ}dUzD!@x4OYu`oD`9|BES^PnGfmj1r$K6~;vVtU_S0 z<}U=V8```7{?7l;UxfzUuieK7*X`cf|0nU7OesiMft7 z+BY~3zIb<=nV&!g1FqJCOGB^e{vn=`gVWTM>7Ub;8K=`!{pTTzBelG~ztnkTx3#qq zUv}o9#y$K`sTHZr_Zp}R13@jdPMsWHZ{a=PsSgs7Q9r8bTyMqXGTqAJ@K!l_`QChU z26zA&kxJ66AYOqFB&0cd4R~xL0LJo)ioA1%e-sSsJ<{geamb_&R0R2Xc`p%gG4rdo zq#j!Ac;O8JO4jn(?QD)wbRf&VPm`1ObCPoN=$UIlzl1k%maoW})fRT;K&c40 zOcXI91U6y-WBulx-&9>jIu4C`Y{pid^7K3s6BBc>*r>x+Dor)yJ}QHxCMh49+sevM zCsd#L`1t(2eQ0xl&1g&F7_YL61^LlX!R55)adc4d~|r_obzQ zS&os^SW{APZmy_xQ{tQ}3_Djr7xrmb0V zRm5j;D4?z7o-z{!tek&Dt-4z-Qx%rJDQ7|zbuL`Ef8Hl}$H8>uKz3HdYGy=2P=(y5 z2bTQ|lfSM)6v@w@KaWaC5XQI5UtKalt4_Sza7~CwVZ~SL*!ksTP6h90hc8rbuxU$I z?~N&O8Z@Or$C#!iE#I=*G$qQGTGhV`4mCc0{5E=tL~Q^usN>~ThSj=Fy_d>_Dz?EL z5R1>t6U#(a8v~KC0!ffBCD4P4#VN9il^`p(#aC z6oV7TY2zurn;WDZ-V2aCJx6GgBPFq&-!t3~$jO=54wRR&vak?4w|DPeb$}2?Z{bx~ zQgxB@Y0jR~(X7=b3jn{;)W2*m-iJD43Bn++Hqe>g743Cti0vaSK$VXZo4kGd_H>A5 zCJu~vK~uXySe?eMukx#sC&=GPDRFT^0Ke2Y>MNnsDlZMKL6`LfAWRWp zp)gA~Zr((l;s+N=5z5agZC|%Uoi#Jk6O0sxfUGS4FmoIzr?$;#h1uEK3e$y#hKd+J zy`$vE2;?NnRgY%al_4aD7Z(?S6G@K+Zl>@kUez!%;-qsaO@tp9{_=&^<})TxGVB2# zqNJ=GFLr*7Z82v{824rliCfUnH6JG@8_^j2ePLk?VkmRx^$Qy<#ju{9o)V*6NlD3G z7!C89J43Kaw{G2{%X1k$;lG*VHICU{s1HqR{r&xQ`1!$|yQkzR4;NRkYMOWl!Lx@7 zrW#Ni2MuB%IF7Yq`yhHRJ3FI6Q(lp8FM70@Fd*#@PsK~w?SLa00$g7@C<=0mS3zE| zjg@uIsIR)FhMbTPR8diJ-Vptr@>dpAa`fOe;=A9QfwKl6#(?tXr28jn$3&~$6oq7C zVW=q+_@J((wGX@^oKm*i zwr#ssSm+x1)79O5^y90WxcH|4^Yu)ri=}}2+aLX{t*xFUlK7Pc`xCK;-ob`yXP#|L z`ZX3T!CZ$bMrDBFj73*x0uJFEttrqb=b=WtME~>4(+?`|z__2l3ee3oG#sW_^XAPu zxFa;lCQzhvX}zOXKfZn-ShfJ+N>Bn}NXq6}w%qyGzu=T^L`Je4I&_E#25=gc->f|& z5<`ta2SHrI1S2gEpd;tpj)O)Po>OCEHPe-=O&GP3_57Av&tW=ZIltu4NkK*zDACTq zK2?M3AfV9{6ckL#J(=;!X)^#{m!WYKc(q#AFNGI}D(XHs!G=b20+1uqs8|@^O}Jo8 z8I1J(d@(V5)O+z#Xswt`w6N}{o_up^%oGW7@SCrr+hh;3we`4t%6L}G#lr*w>HA{4 z%A#zLj>X5vtEj1+xu>*lX=!P3A}x$=xg+(_BOx^H+(ahY_^CZAhKQ)0(|RV(D2Km6w0d@s+nPqsWri{wu;;M6+KZz zZztw8t&R0Tnx?S2nQfg?Q4I0FuUkj*|h%hWWh>hipP)$K3s zWn^F#V>96>4?7Jc5-iqz&EfNdcPEP!=8HnDiTDTP*C!y78=06!gGFC_gd3$UB#Y9@ zP2Dz|N%>y&Y9+1T#?5_4HEjqhP*G7Sck|Wyb&BAe``XW;PDT!H@?=RLwh&PkG42ne zV{fK%d9vpi_Vlz6>ocId!m3x5cj?aykI!Uzy?-@`O7H5=!Bx+>31+*lY;ha2+^Q(A z=cw_1>CO#9ku1lk_!GiGlgg{|TC%2H0~lHq>?5rdG>x~@&LO@6Pf7!9ILz2bh^2qo z{&Gc?+WEbY^>dr%DcvDQe-NJu;b8#vnmz3Q&6nCj|kdrDeb zT7F^S+K(lJSBGK9!J_96?x6y&d0X1TlwC(};}>1U=ZSdnq`nd-N!SkpR2oB&i!fBk zp7?HHWV`?)i=I7LQZ5J&LHRP{Ejv#gUT9V0b)NsCn`Y+7E^3+%7C!%ONvpe4un+~V%Syl{Nzh%qyDOQuhM z`0&%{s2}?EKa{yw)z*$bX!kC-daSq7S8;Z{x!I$B)e?p_02XUyW#!rn*V2E|H&Alv zeXJLBj~qFoUFK#})jW3=y1SJ}u5R{Jl%Aemk>iK=+N*=C(CPm<2H=8`QgVg}wRb?y ztn#nSY;EW+q8k94HoiYAe3dwA!kVO|@zc}O`#Uva?0UDDSj$DnFr7?mLG0R?@%0*(fIlaV0Yg%i4+YqXXz`MBFf%FyyZAe0<&d^~5;O zrO6&~f>Rz*b|AD5yZQz||Gd6S5eR>|-VT3VkUyWZY-2#5qY<(0=N zc%5vc;-GuV0c&Wd?^C*Sho`1aq3zQd^e<@WBsG2CV@M%&oz9Rk%pK~jdxy|S)@$z3 zP~cX4izOmQ%#`ivyFH7Dj`Z#0E4#tsB(vn)VXgSPcMB12^C)<4n8+0w(O<4t?`pQg zW|Z<&C7QhOjz_#bMs0I+B#yo@K`Iztq4lYQvHzA`CtO@+HZiQ9A+7+|uV3#kCq*i* zEUsl>V9-pKs|SucUO)(fDke8HqWc$f0ls?eZ@$sh&COe7WMqQaMUykKv2VCiRpcFP z-1@Mi4+fu^hsSZz8!g0dp-l}S!ZZ6E)E_*ec7=tNwHk4~2Fwyu0}N&@Oe-?<&0JYO zlm>3gJ`v>ue^IH+UOesr{2i<$Bb#tF+DlO>kP;p&vN@x$vS=Oy))2~rVYo{%n>GQc zQ}i^ca(TfR@I1IIHa;!Kj`f;$KRSIA6U;C2xr&obX-y8HL+!O}mPuKkr>C0{N`mZi z9cM_*HzB@jT8hh22$0JzPdmox!Au6QON-?G?3Is>hBkLU`sAu2UTPfMjaG-ul*J10RJ?I;TiEYGn-w+nh%F2rF z4wuXpU<+Jbp;~**O+<{#mDU?Ir+7uqbIv6tiksJ17#%Gi~Mu}Rh-6CDMjx<55QIGWMd*Z0NJ z%t_cvdr$AqyV{vLdx@k40%CDV$%d|c`iY)=e!>t}2eKS?ct7{xCk3tIjD9y@p(6f? zp`Q>n{h3?8CF}&wqUzDR<}6%1w9yw{MZqE}JSvzrZ?3MY;_o}l!p_c4fVyM1LqaJO z&(l|Cm$>yeZQ4{NG>w`ikre?{oS)Jw+|>);U6+=T$w13Ak#fU@AR$r9)77Hw?z5Xp tf2Osz&BXr)z-7bVsN%>){U3i_WuRDgj_S_4$I;M6 z^UcgNb9_|v+_TS)wbxqjyWV~MKS>CpBI6-LAP`hxAwFpcgqK%BXo{hbxl`ceF)5gNY+{VOE>$RP(m9?R{8N)m9i~6;Jjg5shI}MHL|N6i? zb1QwC_HPvlppnNGLQ2*U$P-M&ZzM=u0uBV?sUghwLDnH*Ys%S0_9&6(YBV->k~m!h zB@@f$Qs7NkKbWX~VySno zD&I$>C;T$OXUFrX%pZQ_u)H9m<82~|)zRzS2|Yy7?@fRwY$S{x5RzEK6Rh=m9o?r_ z4jkR@Nn!CnL;rWJ`JGKcp#LspZqjhSe^>jnNFly|myVw>@qeG8Zzjh0?_pdm4FCT= zk3w=`20`aj|2gnqs(|nMoy+f{r#u??PTVt-iTfSE#=Osex}- z<8atf3xvKxg*eQ9erPeGr0-bNf#_ult>eCqSRCWyB1II>*BTlckm%^>(_Z^JuY*1w zLgjLk29~-#)E|mPcPggrl#gvw$+fm8e`RN9|1LM3aGAUV-A4xFUCblku+uo%o^seM zYWt+77NSG%uWVlX9F6<*_0=BXHM%6%5f)_Dz8+3+b)L?P@Dw`@`6S!h{knIG={}IX=OqH^b+d@K8soNT2^w><;nTc#@^J98+Ity zF%CrI;&jiz)U-z!dbt5a;E~dyuH;=(+fr z3)1D@MC0|Ncoy^4$zRr%OVJ8i3$4Ls%+X*$$T6gIu%_LwU0pV{3A?UL2hxREw@L?h z_q%BHua6faB~vppT0%)N!X;z)pip&H*Q4R=+}y_QSO&|jvQe$0^jN8?@Uv*;lC<}fV}-AB8b;Qlm0?Is`8zeU;)3w|x*9UdLcWyYH|#xWVA zs8>54=yk_2J<)CrYA7`vq}6QvvKeNKusnKmFgS~dzZKvtrfKzR^)tJC${VcT2)rb!L>@zUh{oyLx4aB&$=bQac=dtVkn2bj>rmAfJ zmKsKGZ%8C@`U4xC&-kW>c#>AX_vPX7@%#$UHJgvGFP9P5%@||ZZ7J*ag3V_Os<#_~ zeOtfE%gJdN%8}_G8Sz0pzMfZjoD zJZV4UTD4OKrqkuDOT8DDS5;2VyCo1?8_Xj0K4M?XdYHQ%#2XkIHh}$+d~ov1dS!lM z=89RqKxIv)%dcJ1lRO}GCiwNgZJ(;KV8i_X=>q8d5nJ)*@Ekx7JFOuEOZE&xE#X9DVI6*M zEKM>}@XNE$@V}~)_1p7TwuC1|)r)Wing>(?#L`2aIKRlv|M!Mwk7a0e=(ULJ1EnVQ z3M~n%1L32Hrx&>GzQta4t(pAS*Z*Jc?f-3Tk6C71IwD9`H#R;(>GtWa_i3&@!R%x6 zT#(oiY z%*lw(w`jj-w_ayU>s+XuEADVOn3XE6b$W!0?)Uipp!Y)2X@FCYT@YpH$7lCm5(+Jq zF8%s8-6UCIn_9?Ym>Vn+R+0CokI{*Sj3*Gd0g*-=FR9Y(7RR?G0`RFHe|D)Po_=)M zR=~epjaOPas_Wkvl*r3#o!+i9QEm;+|NYrMbp1GdYWhL#$?<}-E(^;X^Js)nD*1e$ z`xQeH*x4qKgNBEM8cUtge~WdWh=__>PT7vmF(jX%^m3i3RK#ft3+kgWZ#xit%n9$7 zTrNEO4#tZ3H$)aq9^7&(&VP~`1Gjg)^B5w5l37$xMzvy&i|%k7{Sgv&8BRW7sYRRU zS~v8S1?^Q!;PnwwM3bS6sf}l`s<%sKzcbsJ5?C`Sj8>eWRDP=7JUrD9pOV(oiOOgH zMz&JtsI!733D=W%tThihFO?=#JhxYOr-z%DansXAWLq2KA0DGY`sx+ztTELrOxY)r zuP^i;d3r2$eM`&8xX%M0>)V|Vq?s>CzNBsq!u1-I;6PPSvAl1$J-H^2Nj)(zlE@__ z@-YXcywujXP9pYwaqiZz3=g!{PRP{I`RLOzv2JFt#3$o#Q~&!a7_>NsZvNeo32baB zWMxCy`;`oD-KdL|Q`^B$k)ToC<_<5yu36pF=@81 zJ0L*BO<-vTW@ZB7;^N0_U@tdK*E&DB`Vohx?^EYS+~sD(RdB2%P{=?;lr;X7Vdq=F z5X`Ww|1oT<_#jqbNB6j7a(sl=?|*Ez+2cq$8m*t}>mp=vcSXXcW_x+TlPt?TeaMc4xHxQr2Cnich+ZD5E!fUXZ1pi8Bo&Z^XG$iA0EcuJ zBEhl6sAAp2RJoGmyjxc2`gX<9^EK9mQf+M;vKtzNwKkVpA185q!6;exkexxRF=vbR zz-=Hws06IDeDx+`ony~_#KrX&m>U1#cbq^n173lrbK)tVIb56WO>sv6uLHW*(Gh{5 zz($CCc)<8hpE%5N8dsaDGF1F0ekArefn71p?2G#Gc317IgCAih;h*c1h-5K@B}}}g zuaIWX_R8Yp$-i8@Ou_d0#hkbm>V84Q&dy$Z{emAo`LvmAcg~kk@7nHkS2GBor8l2c zu{=BHX~yFtRp~fCO(~hgZzfgNX<3f+az<6wSGpvYgP`-U>P@%*yOAmnaJFQ< z@D5bBw8ELU<~a!-b!Et_nPw4ig)uC;>Od61{W8-(6FhD%g`3NyHM1~`H4r9Fr%&p9 zLH4oG7P`2 zTa6b3*K*N>P1DZZmJLj}JTL73!GufRC9PU_H8%X;O?Yc_ZNYk+erc1g3XCeoc8_ng z{Db-R>0WJdeEhEG>j3YsUq5(D&Ew!xd`87zbKF`f*BFsjy z&c(4I88hmQry2tXF%ycY>(oLl3pKz1VFX`LD2^t)0N1bd#!~^V; zLb&J8(7-Zt)R^SnWh0CZDmVA=3$10rbQQd3IJj5K_QwN{Kd}0rk^ob(fTz`5f+k-( zP4p2=3 z!AFQ<`1U{3dyS1IH&>0kGnu}8P>{^g(NFVTp8tfGYU%dF=&>0iXw#Jowvb)U$J%7E zEp?)M%dQl@)Syt3QI+!nW|u^!%lbB-_ND(KCWRO7GRf0_fS?<> zQRtO~C>08L-;>7B+eiJ#DgUBXgbEWUTmAOLcbB_bhTfsZ=iPGz>=@fU_zxIR(z+Le0(AVcfaE{w2!qxx!l1S)V z$3&l{49n@}@-Q2AeEgS*@N~c{>5$s}&iXV7nb}z}Iz9|se4?Nu?m^pH-IQ{@O3h@C zBDF&MU!)FyXe3lZ7TBbOzlH4;cuMn+iXwCH` z;@``6hR!P?&oBk9y0}daE!g9|HoMVGO|qBJLT;P7BbY}MkNz3ezI(5d3G1F1l&#$) zn25?Y^7gjh#Vj9z`muZ9ldKZ9cClchy#MpgCxe?iAdP-XFSF2YH1!`Q2Q zZ(1{X%8+ZBuE1hAJZ|@NkLYI9slUn7LkbhC8r0@xIX1wUwVkX5*_!_?Nr1Qf^{9wA zNrgpjB9|D{QOr~#2k(D9N=sFm84=Gq-1`ccR}3Sg1h>w+)sGkd`uan9 znAU&sPMTwMCtHT}TjGq@DF40N0-3ibnb(N&m$32shw!;`FnuwMh0=I0$a?akvM)Wb znTD5Y9z5!O#&f$u$b_|ZIp2g6UlSJ({UkI)jCC+h;g{nHJ`P^*>XnDrY18coQ+NCG z&KBg<8OZ&qnKpeVnusHsN|z;Hft=`4{Oa!0_5aag91lpmk`8&i!rBL#+$wEV5_>e| zS0sxKux}TQO}x;2eOUUKT20jj%#UVZ-1^UgvaZ8}Ch1bD>;uJr`zRUKL)ZZf@Mxk6 zaeMfRAW#9DFa(S!novzKz)tu;F@vz4F?2Yv8-b0$r8IiGgW^KYK?hBA*V5*%sBWLWF9R~e94Q{Nqn&iVuR}hF zbkVc`jgO}?t>j_eNV)ddI+tQud7>Aw*t`uhb~Y-$So8gK={1vb7NR>_a3_oHFLa+z zk=Euo>R&nKQX8N4d%W?Ct_W{1$4AgQn`nz zkO$9@8;B$L1wO(PsCj!($ZwV){5o}rCGfmaL@K6N;Sp&#H9bRVUyh-1x5w?!&uENX zOUS=z`FGlIS8KVkyf%5kG@Qa1UOh31H{SNY8u`l-_>=~Iq1hZR^um#?G!*yKtqorn znZF>AO77MF0!CMZH;#SesdwA$55Mb_O8FewlhH@BN~WB5YAY8)I3N)C$xa(~-R^w^ z(mq{rztT^arl{cI9>T*nDk}ZO9Y0l2K-n~~?Bo^;Ot$@*HZ2so^3z;^v2i$Dh=72= zVZVcn&Ul0x@;g^9Zx_Km+#t=G0}0NnQvf%7@Fz`pS8wM=p6&q=)an&nkMCUwAG^~b zx!Y+I;k>ar91k*Ee+{pvV`cSwLR0e$^3$Lnx;go4Fh7*+o2|{U2Jk%8%C!jlOiL?i z+aX0vB#sw4dbs;vNoqHcDApb2n`;6D0T_onkeA#pI801TV{OJpM!Nu7R37auriS#s zGMU_SD*vgX?v8L@iC=g4)FnE^f;O+0YCHPJsY5eRV5!D*NJl_x;$ywC2r zoX+`Exy75$_?Ko72EY^40RE)eBopv5W(ps20nfH^olJTyFYBS8 zplARrq2|WPrXwXoAcFZzJ#l~!1$aIeBgBfWX`%qW%1oPw4)30nX{t**R##X#VqFY3 z1i+$zxC)Yptm-f6$xUZJ%7;;_Op0rx~O?f?iI>vsT%(x+65U%o4JsO4KI zTCoO$4qv3)GOXTT)T9iKw@SHA2AtVkw-)YFM3HJCvhb65UqmdFigu{}_(F*_=*F3m zU;M2oSqUeA$GSX$Ae^W@FS$;;7D(%soeNmW;3i0MJ^x-v@;68ke^1T9Pm%W=B8=*iyYatLboA@SvAC~)X6L*sRDTrogON1YLW@u&emY=z5f$`bgVZvWiMOK{ zO#|5YNSR#@_X|vc@r6;euAC)Gwz!+P-cA_QGQ2j`O+)1$kN5fy7$o2x_$Tsw8AX#I zDz-tMfx^TOW5g3!qnV;5IQ#0X$aji@9tjqY$iI8JzhYEG&l`{ydoz@a21S+l5%N?3 zs9a3@?`JCL8_$fU#lMd@JR)OKnAb}Uy=ZJ4f4Zw}G&WDQz9-3|@+BthIwo3fW=IJ8 z>6y`8;OYusrOIb2Z{>}6)<~9aT5abFYAK_@=Ew!E9mKu8-`6IMKLW|(oY2y$L_~~S z>8LK5pjjX~-!1iFs;V8D1vlp(#;eJ@{s_DGHJZXuPr_5)nCE`k^vyPrD$=GdDEu=< zVt1TYOtUQe7#+$1avmha&Q~`2L?tCBbf>G+yQ%w8I>_$7O6ZxH+-@&iZR+iJ)o`il z9W5{r5Xzof6dROBxD*)c^4|dzQL_df@CucdZeHs(lf0OU@&fW}T=qQyxaHh4!KSwx zS39UBIA@s1D3)n_$hd5s5{bw)d(ncuS|naYmmj*~=+^z^%^k2Ug78IADFeqlBuRnw zPM!oR{M^i?0Qt81r#qsK7zy$U!N;y#neSejspw3Kb>8gr@bc`IdHE-@#?N0B;;EYD z{w(jm0y%v{)5Hl;bZ>qQ7i~;uLF1NvYAAngTCgxNf!j4}&^l@c1C)7--Q(evzRdBN z>t%;YKL>-U9LMAGosEW|1LLBK>OPB!sjBwM@Tl8tT3gD$W$OoDmR#uSjQbg$Ks1@y zk$)7t;x;AISI;YVkcRPrtQPM{@K`0Z#Ik0iJlkgx9tu=KT;2*pi*toR0qd*x- z>7c2u)GZYo0OfFfQ)27eecZ;p<3?n@9``kdcEJ4*6CMgc(St&_mrjqr2l-nI>++|5 zFJDi*cgg+X=8fpTT6#rntjV4a z$=61HA51H_u|_{n;71YRLRBm6&M-hh035O$U3ni94C`f9(s8sRbHIZFo zI;vRf1tecGR)ZEFJ~1}l&7H5E-VYKm&CbCMo}SjeUr|<-D|W<(_2rGfTVa^sjX!GX zlK5;?b~UPxXS7${`>N~mEB?FqAk>Mf@6>^ZYZUr=3SIL1)yb{i(WwY4P`h9a!g#MY z@kvPe2t&zjen0%C6)e8aD+Tc$1C>FW?h$y6<1TrU>(Zy*nTeAljn=a=gPBpxN;f=y z{750BS$opaJShcv%QK-HBd4@;u7>YlF-LW3nA;NqCgx#5vH$|%dLhR;Esd7GU0Qjk zPRm3~M)HnO_~vCFNRP6(*Cpp|;xAci-9i~>-=8Pg=BomxI3LbN{g~HG3!~$Ww^*&T z^A;c)5S@9UE|q-1dDb!8dWLGiY8;foY&LxFH7EpiF) zno92i0U(J1wa3g|q;Mu}vkBK#0tA6q({(&qp(ne?mtQCN(!GU=eX;0|M7GKxyPH(L zh~J+8$~UwfW;@V?pGP|E`j7t9nbwgZySvn(lkR%9boC_h-LlRIM|K&adI9PetbYkm z{`8MUDzrWNv?JL87XqXcs7|y!lpec^kgwN=soGA@9C1H{APhFE?_m%M-rpKH@YU_; z=NlLihG3ZVP7)x)xN)?L;V&h$MV0N3X8?c$P=}yPey1U~Y-@FOvxB=lBcEeON2`Jh zmX7)J&}r=+wYl$e74F-xGMUbT^Ly16ws=XQOSsmo!OoE@l=M^j*jX<@J+NHi0PX=x zCJGN#q(88S<^lhQnBM`XeB6PJ(#+<3Q%dKnJaW3qh)(Nl}@ zNW?r>LHpxdz}uOhF?5N{k74>w@dZP^7eiCV^{4=E@(go`P}?WrW(78jIhMpE@_3l@ z3-B`hC(Vjt@i%~P-&cDGDai`$Ta%0xo+gKj3bs|zAWcX70NTm#`GxcPR^YWlqv7($Z_X`L<)rr;OiMKHScGY=ZE%$ytW4?8y>0#$382ML<0{6ymH>zWvCjOD{;w6jYj96+} zW%z4I>Z2aFB>5by5?88`Xc88HKVR4O3&6RTZ7Tj6Z^X@xF} zE_Hm&FqVjR6DgBgOnOPR8)J}t#Bih^yxKDfkOg9!$X=_pLqEWh$pU`+a4Z}2-lMQ# z3|HoF_eW%L87$)lXP!B^MBX82>9XG(WUtzXCGN?kg)Y~u_%|?ud8JmTeeQ83EXcMm zl(%0W`zBRl68E<9HXsed=sH%7(q-$WWw8Dz{wDC=+EHYXp66d&*?)WUtDr2p62Pur zhY}bAXm$tNEV6>g!b8PVFtV89fy(7}vNMcuqi)%YZ{udo{v3C22EZrMS@#TrlEt{~ zu@QDIuu)-aIQ?EhrajoQ)-6J+M}=YN+}|uH8yc~&Zt_$*?+=Y+CdklJe>7LJ*K?u| z_m3f15LrpuHi@VBb<^GY26XX{=wY$9Bv<=WGI_m7@cDVFgRIaT4 z(-TB{4%_4Tfvn`+E7{pR`Y-RVkH5)Y7b$D627x~*H+#gwJftP6T#-YJp}|S;14MIW z_0{(~;YfolIa^=vNySg=f5{MN7f$G1rYoZ`k3hhzfe1T>4f(IGLGxIQWWG&K{8gWX zX8b|H3jrD?nv8CoZzFHvHSyFl=72yBc4tI(`!#>sOU`(rzu&O7KI6!RLVkyfUL!}` z4z4XA^B1wllDN68E;;v~3PK&b4LT=4_r5Bigtb3$h&L4-k|32p^F8t$_lB}1=vi_i zriaq0^C~hTR*Kh%0iBWxBlpQG;U%u9(oC=C(yu2Z+{FrK8l?#q?<+(K@tpUE?shZu zy?=HO!L*<5f2$9|R(C-^nY2cW=pc@Zk1qo943_w3LQZ5)syl*cQKV3K-=^{|zk-tl z`O`q5``Ca|pd^*PHU!jDZshQI(AAo7$`HxaSPi z^@QPa0P^3U*6XX8>%gpJH@qBKWCb-x5Sd+KR)Q^`%hkItn?<9PuDB1^_)fL zfkfP=`du5?=o%os4IvGSIt95Og5U8RfNklFg@VSF%o#`I=A9 z(4W(R9`j`hF$I@m@ZzH9UZU5=s8QZ8)@wmQ?EfYTI*!^ckVS)736iQ&vDa&5!m&e^ z#WYyM#yA=T~DSekH?MoT4>zR!E9D#u+UIYG+ds{gredzYfRTTvbvno z8W|ZChlb*_JSBA5{apLiH!QZFW1xQ2d_y`h@Cn4fau#;yL*A!}lERmbzc5>M6zG4l z@308T6h{p{PJBhObXdw&#tu_S4ED>eMDRd7_ft^3Yu@U~6Ul6Zu96rAMR5KWq8A>c z0fD$7Q}rREf}PBSy2Ochs^Y{~&yOtrhlhib3(*Q{4GW<>A%JdRWMZ1zxxU&#NJ<6? zEjIEiw9fXIPRB>G_sEbg{ID&p}CzWQVKgTxf-_ z{2%Spv*I$8JjD58V2Gov9^2|7f%O<{@K53a!V|{c(F-R0T|fH2PDpPy?pE8&+?E?b zjR?9*r)X=Nu{G+0>+b$fH?SekX*^2fc2KkC?HdKwur8Rhe8f4aM*aIFiwH2-pK}FW za@X4e>f`~H^^>}K*LWj*aon!#i25j^n1|5MRMZ{4ZCn)JZPHeNEW|nfyXm7lEa?5 zXp=ZD1pIJX#R(2zo)*?tuA(AifmiW(_{#k`K+&eUSx4LzC+!d?=TJlLmBU>69_=EF zWwu;vd+c7$w&T6L9New;A_a~VQTK}TEBVk8RS8pPMN4>E^}iY!Vc_F%ad}8XNW=Mg z2vHi7PL&Y<*2&7lZEgG5sd}}R5X?8^Hy(97W6y zy;J>G9@%%C6sn=d;hGH|5R)|lv5t{uNn=;%vsS=Jp*&rpVW9}eHngr93F~M4 z1$--AegS)-K&TX5#p}MQVaiXYSr)conOg|Ck_Mo))b^>8qK!~GjH6&{gV{%dLq>sd zu1N_)G_>OnVVAzIPrUtp(ATn05MJ#*yZ}CuPgxmXuP-UM?Ul3G$r&iUI{=Z?Pm4KT zu!JgZXG@jqxp@{qnzZm$+3W*-ET4=F=5kNG!mhxV&tuQaySvZCpJss?pVwEVG}Tyj z_;N!ULhlf_Gf`LgkfFD*1;{UI90!~Zvb@;sM?0QBt&G4!M`sJB!$EiK2z8Lp6UZ4e zG=b1ATwK)o3{?P?!87{#c1kl?-(8QM>Y4TzqhH(M_noK=MW4(V9Qk^VV?&EYp4ooA zLO6mXmwn+2z!Z}~X2*>STM*=N5H3S&-f=93&l^*tjuha!Y3}A%tb3?{nwRob@{>J6 z$?YDFex&yR)`QZ-M9c0X84nYJ@GH?0m^SVf}?&i9(27yMu%xrqh7m0U%6qbUY*O z3uU@{YK1vsbG2UEXXz z*X%t2YC&4V0b*+9@}L$R@t+Lzg8AH`GUw-!ai$TY?<10_hsQCo1=i``D%ZBzHFk7A zhMiZBlmm|PITa#~c6Wc>2^+On`0Xv>BC&YDAGk&rgqKK)%I4S(f&D0Oj| z#nW=Sw@~7qmzUS%jjiU#u~QR?6ogbwg`cfE;a-1z7v=JZQwf+IU^Y^WY}mFZF09;M zs6jC3kSNCPO4AD#DwanaaymK#EYRlPFga>Wl5i+-<@A_2?X2HUL87E7$8!Q?(DI-d zfy~kxmX+P}Ua0wg^Zms`?H^b><>*6h?(Q)rM5J_1NUXDsMoBd!G9LZ62NmPjq^&|I7ec@G<#^3UVPUl_z zV)x!>tnK<9wcrZSl!F8ODj^oz3PwVAIG+A3CPIKhiOWsJ!Z#q?48tE;Vb#CMU@

    &eKlAWslZhA5>-EV zufh_feUC!MziTz&(0`l4AaqK%ZUT(Ra9alcpNh%ROsBgpvW{-4ZXoyX)%s|~NrRr4 z8qddWF$;NDwVGN}ADHTwlJaI0k3S11(Izp5{$V!RSFp|s1WuIgS8s>)i>>M9;N9ZY zPqVodu=x9S)+;R=(p;SElgmQRh&YZHrsMH14c?)Llci#LSq?pT9ab>>gu5TalPHV&5XC|hyCY~mJ^Ai9|Z?`qRB*!7-)i12Tma2b~ooc8WV*?Et zHsec`4Ksfioow0U4hk3*Gyz8py9VYdCqc3*6yHCkpRWxE{C?j!&?ftmw`ja9`_Pi9 zs3nxf+GFCG6NW)3U`jha3d`Ubx-wf%Xga_M|9~O@k}fwoJ`@zQ#}m|8eBXs%_$}AP$tX{83sz zweZS5+{2PY$-&M-I@6^&1kx%|7o3&=0r+_5S&P_AR#k2g(4pV15`rlW|) zWmQfCLX?M};=v&MsITeJgK!nP<}fGAI4RGLK<1bfJ2&Cx`!6}zz98NOkW1x}U_C#; zqna~WjOb03z|2R-qc2@39GCfc{K%qbol(qHy#Q7(6 zM%@K<(Ccn6Tm!QvU+8pp5m+>)i5{3R7D<(LbikH`LQHJOaeSg1se58{^`l!M zOrf*oACAxCxW~CiN-dwr8uB!Mf5^O$kh>AbxMts>hC*2o9~z-MwNIykEUHop`C3w zLF6?76s>~+gOhPSxn-gB(4TW5Y>w`u8$^aA>7_U%o`-s_SRZG;v{Qo+2lTq1``nI8 z#=b8BU8~q}k=9EQ0v)4{sNx8R>BpjusS5F)q>_yW!}tT|)oRW`5gc~%>J-cDk5z(w z{QOF9Mw#90U^e-vv})(@oOlCPE#__peTPHpImG=00$4nCELTiu2?*k#BtLeIeeaFF z0n=R^^BfWxvxFHBAZS%|OPtcB$*(EIt4<3DuuT`C3oQmTNF$yZXW*4@NGKipnOMFGcEP^|b2+`pR zEx*0Pa|{n>Vfs9{L|)$~u@27E*XVp+OFR!AJlNe(zs6IpFl*_EBD0sW45m;lr;k+m#9{caYrQRq#=@PX2G-58N(0o4`6noN}?f-98!fF1DjVe8rh z`la3MH2ZDbxs{jwQ~xT_p)L1h6%@{=P-gD2J=oab2nmuxYJp;w%_I|6hE8QvxDU@e z=C5fOrC)hT5x&1-lXmCJzt$*sPm#Np%X@1!oi4vULq7TJ1P|n<5j*O&INh}V#G`|Q zURg;_$Q*W?zh^FbW@dq2-#Wd>8@UU+V+~Unkd>U5lgW&d3Gx9QIl|4}HH!nc^(2~V z)WP-K{<^}pL;r~&0AQNRzmaOiz;pahRrMu_CA+0~K3n|{=ygOKq4GzI7BJ*M>DVi^ zVQ((^u=lc43qcZLrz09vluJ!peHs4R$-%o1; zkuxZbeNy!LoNzDVux=mIUEN*!h?f==HULq!g}L~(+^lH@C4>I&^qz!-GV)lm+U(RL zD6}{bAeXi9w%F+Bk5N!EUwiLMC#W!0sn|*-fx@`sNV@d%S4c3BP8w# z9GXF9|1{x1{zoqSD0fTh_~VegNF+}|dRbzpTOj5ob4fBkmVl3N`pJwtp(M{GW0Kps zVY3!pX=&-~UK?Q>(3P(7!vlOG$uP>g%Ogt1J=D9Ox<GGqbUY5eANL+5?*-SiFj=r z(0-*{EE{txbY<^5OwW{_<960_^qVulmH0H%-G4A|7L}*_TlxeFV8Gc4m%)} zF(>ZVCzY$IffoJz*qY~uYlxp~jI^cxuxtKh3y)#@oRh%9RhFI8duW{~7C>>bf?#nX zS)7s!T_jRARf1{|+PM3eh*|0qvHBQ5G@&*u`5NgxC2GvuWaWt6Uk54Qzn;QwC0i+8 zv)%DY8Blxu+S z9rLPl2oxADnak>W(3opcAZ&eT11CQ16kp}_(;EMBg|kZ8zzn;yF(*()g`A79{5TFitYnAID%>n&0*sXYrL#bMb^+! zN0hiK(BE}ti*@{gO`(3`DG<11uC;pk;iC5$n{Wjw#+hVQ7+duFf2srY&}#ctP`1{+ zRyV32Neo%IoAs5DscW;2mO(2EAY}l~fHbj>yB-u~MB_JXvl^9jS3l`%LcJxP|PM!va5oUB*+1EuD6y|@16+{2S+dF=5>Oiqt`#z)!T`_ ze=m4k=vCRtouM6OgWxN`D2N%~i=Cc&T&6Ranpw3ulO7m8lUFiEuuO0`H@C<$a)$_7 zu?ORGLjLtM_RNfJ?{bjR5ds0A!Uxl9ZPWQ%bn`)GJZc7g`S~4BT4^!_|6QKdG^TlE zy*cDpEV6@xbw17ap^aGjH=FH#th_?v+jlGS%<51Yoh*=JeiUwmy{QAaD)9W)5g+$Q zPk_w1pzI00{idLy@XL=mLCfC6=-4({J~<>Q7?wz0A1Q!_PVl6$@FjA7I`dlz)6+vB zWcb+aH`Wi6zLRTY^J`m{n+4H*pZ~hg0=h4-spi}5Qju_e_0}kNQ0$JdWzpfZyxDTd zZR8Zuj`JW-f*o~>Xpz19z2AMe`nIMgZ%dhsSH0&Q9Y4(1;xjmQ2h=|V8RifQZ}+OX z)S3*UVKFvm6i`!BgatPg@VTUiB}(0)AccUXwrtXkWU5F3>y9Wd#`l&1#l&tZZ4`#} zJ1hT6$Wj3o19Y0x0qdV$^>n?~`KQUzE0Fmoss0k@PW({}MWQ2GR!ta|?`B}QBl;_J zNG(z%&=GqI5If+&Od)*QteB&l1Er~{!(9dAo8M%!F*%n|MVFHQ5R=4vC(Ks1~ApnSi5aOo)Y{P7Z%8QO6<$gs;-?*M>0y zu5HQJt}E7bgK~v*1P_4z$U^J&eW|_4lc$9CE+VmUpeV>E#rRT*cU=QIFbwB_n&;0jL%7rY=^RtH-3F85pPW(}!P?eMPre1_hS*+dA!B9dv#J zH=D`Qt!zvomo|znTuJ;6{g?kJlwNCveW-s-JNIj5^iJ}4DGTK9&-|`|>{lf$p$yR7 z$dPxImW4_gy=A7L+W9AO&%))kcQA)q8<2QT_47(JWAD2)3grY>{BhRtD0=doi!}U> zh%s>>UIEmB*_NB3>Gf{gkPga-YahEO!dswjPoXM6fGnUj2s!APkdJ zp@+S+vlsg@Zo^Ouk%{%l{UNlvZC_T6skbaI+KZ#Cx$cNTw!yPnCUagrzH1^O11-53 zhc_5U)4HvYxlY(cweB%I>-pUYO_I7pgVtD6>F`~^0##9UbqLxUl7uamX(b{QR)Dro zsW^#A+Y!a$OXz*^2}b6OewJpGj+rrI-wz7fj9#>7v;#afZhwb9spa#5SKKKW-6N-^sn@&)OutFm`k8I6Dzylu0AJej+caMv?ZIynVFCYh#I$r zGQY|1*j`=r%6j$g5^u$bYAYs=K9QKYNkHGtp8twOIF4i2Ryq@P_Xr|1f0f2?Atc?U z0wj;kN0?cX0e}J}o~M<&a1vE!0pi}dxh?nshU$JeJ8VdvZV>wWvYMnbJv{b~Qv||M ztkRaGxr4C12SR|sEm7-z8evG-2C6KJwSB;g$JGs{uv|~~8laVFPV2oJ_#d5o{se?2 ztGb(hHXrPt*jdZz`O6mYntusX{dXMD037K-6aZ~vEe{66Y0+l6;QRuiRlgq?gZvnYUwcgErEzIu^uciH^$ z#)?Fzb3t)PFUSLd!mtytRl5F8BxN};i7A7xgjD(3t2!7RBEjNpXRGhZ7K%Z#F*tna zH#xiF7$uzUF;FFgw3v*78mf27{>Se&=-!qq6 z_bv*6*ij2{xM=li$E~aBbHei$-CSmf?xm38HjJ zQo=s_jjcKANqmAe$b@A*#8QR;xhK{GCxQdO#b27lnc4+fQk^v2xK^CD6(__3gDPGh z?ilyk4?0#S>R6dDj(5MMKW_s^0AeYECrEVtsn-&ez!x)^_cC{ik4fZb&X#l8LMn5r zFs?lZS6p&Cg&P@bEFUs$Er%}HczaUr7`#aKl-)HFB8)M(-r5-i2u;ivnfr}0hUhtm zLi6q21U^BJE|x7q1sD?c8NanBO$Mht%Om!gt^Y$5uLg z2_A;ekf{MInLV%(Q_;=dmnHwaSAx!L$?YdcocgwRt?$q4#I!MYeZIvDWoYv~*{`kQ zG{1@5lVn$w9X}2S!ah0`U4X*p;-&aTWejdkUz`Y*?Co&htm!DqFyKA8TBDfQ(TnTZqcC7(GNl<8%Z!;+M=V$H9DCi0zE+hBi;W&IQ(??PMsrM47R595O&`d{GC zm*y&e3-JoX?%qm`lAIz0sMtsd$f(lHx)Z(+9dB4r{@rjsb)6@HLIWu6x&{A^O@m_* zKsIM%9Gy!NG?YF7eGoDIW7fnM>l+nYSz~JK$(&IL=wq_rFLvd2!^CD#A|Go*)XAw_ zz!^TZ16S=kvI@@-Y6UG&&HZc9_EPjQv7v5p?UZZC4|dJNkKu*9|Bl%GfwA&pwV2=s za~Fb~2S^ZOEZyD;M7N2T$vHN#SR+x2H@^n#UtI)-fs?6hryr5M0Z4Q-*`>6W?`3=o zdrVm`Sp!^>RL1jbzI<&nUFW>K&SJXBtgM7JMf?G8-}y`sW`i=eqUE(@7KV@m^`?FE z*rY$5xxUiHpja%GFdOYlj1u3UM7VHfccWub5)vyQ1q&S;Jirvr@W-j6ch6np^NOX} z0B7+^ug`W#D3wEHr$*_-uNqT046uOx%eN#tDj#HC=G$^FA z4bFj03d;H_M_h|k#s39h^LKHG2z8@Fs?F6FY;A6aex8UKpZ+h#BmP~@k(E7eKG3axov<%*!q?eM34wBC+ znPp5Aoy3gG=yf$$d4+6U!Ae`w$1;n(Mmp<2$$#>%1i#sTfZeV^EH1OSh4+PUv690iAo|C3d&SSPzKJ#)M0QYZ<{>t}x;?Mlgz zCm+)i-79gQV_+}LGm-b?Eqb_IP`zLG8W4E=)^OwHQyP^rPp?;k(H)mot=F6)@*t)K z4u-P#Q#vaTQlg1Nu+j0EQ0 zO|rUxlRjolZ>*4e>XhFyn=o1l@9;+YJ-K1ME!*5=nyf@I>7u?iQ_o!f8FxFI+~ftO zOitN3|5aTG@-)v=Px`j#HHm>cxIb%%CqjJo_$wsTuww`ca>O~>G4-Kb* z?JvvLP9b3;fgo;sgA_L|Z}rXy1|W&xy7P9ZuAA;a>A0|7z-6?1XdStq#1f$_L;^(@L!ELnq)1GA7D7=x& z1W;!?mAz!;ydHluwUSdSVGNfXkBQGGC4wcdB>FwU5y}jIr_nA?N|Xv%++{j);pVWH z^7W6XydnWa+Hfx;1U-4{sz9t)Z{JsYsP99$TjJ=R!o; zXW2Y!<7fUWwW|T=$B}GiZR*3r(f4b)f1JXtlNI2K8{==Yuos4d^8c9T!6pv|&HK6> zLolE8efb<)X}s}-z=s4HbnN)-Uaf132RJOHcU+h!oc>MC>2bxh6V*02OQvEwEO!d_ zS|}RN1ugt^uF03qY_igmC#&K_v0(>YK-yCG7a+N3K{d_#YI)?!kkuH4`=o>6K%Q;l z()VQ7pFu#gqoyMcyG@(Sw|~MJrA=a;EDS*WpFp$XV^6vWm=>TEji`N+Qhu3b95lP< z8OLjvD1fH>9zPVC^u+t-IPh=@{sPA7>(-}}8#n0o#9dww;GzzoOI|54jl zM^*KG`(hvpDqV_zl!UaDAfbmY5dU(2HQPx)KR|&v@_-zN8u3X&`#Ub zA*$x?Q}-mL0@nWLrq}6Hb1gl&{Xj0EckEbG!S{J_3V~jezPo#C*oDzU`*(jZFr2&A z*NYxQ2?B7|-ny2h<~xYW2jns<0ZP^% zi5}>iweB{YxFh3+^%9Sx63eP?hYVKFwY2_c&&SFy!Qg(-R;0y$nrFkLIK!R&&PYu_ zS}SoK(&Mrlp`Ab#rKNoL(%bXNOX<21($jpxiCb? z=IYhm##Aij*IdrgC}@HHo;S!6WMnXP#tSGf=raEdRG;U`ewX^Y60P1vLoZeKG!iaI zir^(uACzxZY2Zu5UX2*$wK2OoyGB337~zd>_eUwkM{cY4t-E3i)UBx`^Ciy~@Cg_D z@6vON%_{cnfexogUGO}@RmO%^U+AdWfb{ptLjvUIQHD_UVwA0wSbZ z;OU^$)1?)BrchsRmXR%I=5@#nuIFcUQIgwm5+QB062o)+q<>bKQUAl;SYF^=Zt z#eLO^-wn0{4r2&%RD0E&mrqj0Lb$ZaE)jwuGqac4Kadk4Oxw@khx(p<1EI??5`+); z&lf=_Tik-}!81Yt)63XjiPkej<$bJv$5tsTUvIW?`#o>mJi4$2AG#3ZJk za}8Y&athS40J;8rWXPB`>`BH1By@Z2haa-ZrSm@ClZO@Zt-p1lHouhN#65_J3L5>- z!%?^jkX5Ou(?RbyMFF!wO>DF9OFK4`lg1)VQmFm8n)CQpjl7AOopZC^f+H~U75eO$AhFsIFxvfjYSXZWO_8}d| z;}^_P8b%Sm-k$~szh&)NJX`A4wzwHZhMdDenZ}+cXl#E=<|fnQ&885_V0F2GD!%Xm z_2vg-ml+pNXY*nAC`?PnS#i2^V3n>3+Z#mczfAApzH%S}fFrVzX!S#i*lK$-*Eer? zt5ajs@G5JA!$=|U0(o7H75{eu1~2j$?;P5;3w2=#bYG&8?PYhweB-S6wgomWsy z133QYJ$t|g2ecZPx|XDzOJYl&;q>03KO zQJmew;)e&knS?soDmZQ!;D4wQwU|F*UTT=Hj=nN`M9Cn5VZUz;s&%-Ax8%-odG-xo zWu)s$jmG%3;9im2ygTJ0OJVFSX&x70XEVeD)P~Kz*B>G(db5q&@w{RUTD4|Xchh#| zFJ~$SWx&yd#<7dQFDO@9p-0s2TnO9nalCoos#~iv`rG;0-s4JrutwmA?#Gh*%tk zs-NZ9+o~K>)m^Pgr$4Q<3Nl3@@_jtKgBF1-R?aJ3Q|(Xvl}tyHqbO;eBA# z90(Vavi&r3n)bZ+;&Ou)k&>Rm{d_1{V`ynPJ{Dr7gx4lU)ZURAqyOTY8C<4y%5Kh= zH}yxiC5cMrLKk(f?otx&r&Dx`NsPWhS~H6VQWq%FVUAwRL=8G+18ibJG*vyCU1d8c_{)#jc{d`lFiL~M)mZL(Y8L5KUVqA60C$iL}Yz0=Nx(2cog?GE?d0p<3&bkIpWhi(;uSRI-i_G$qE{COQZ)f(@_&+shqIt z{Ast0bHy^k(SsppSOKl632vea*Yt4ctRdC3xoQ0PIKD*AE7EiSF*S^@23mG9zU_Aw zIGtX62?P-~4f}zra`C%bATy`Pom*)BJAJbCSc>xnkEomj%}mt2C^t9peH0dWU5zvp zxm%po&(2kdH-q$mP2jL(fayP+)GdyDoAzbbWQ&1y)Zn+FCiEe$Q$RKxl2{HkZnzWp1I|Ey+iqzm7~Y=7>TON9w28ozc{|o z&L6{ao4x65AgkYxe`u62?StR|La4uTfXp^5EAONEJ2ytY4@XjSmphgC*ciAOgNFLD zWwX;MW<+wlH?`kSlW*0WQq%d1w}BQ*@akrwrCXx9rKB@Y*%$(p)wU=NKjPe00+kp< zAV_<14VrWJoLth9BLfP#!FAwv_Lee_AHtSQIju zzBbqPJkyBGx$5-d)_@d25NpWHn+~Q_xQYcO3L`UpHjU*n9-e3aRHeXZ=Xbvuz};QK z$_&fPLxCO#pU;Z->Ye(?aacASNi&+f>=6!68T_DjFpWPOvr{l>%*3-zj64gppHmPcT zsf}Q*<=Il$GDQwLm%cp=Et9$(^^VCVWBAd;unF0;3Vo22Cp5>GS62I1#7K#s+zh7niHZZ0!PsH_P0VJ$B|Z!LS7#erJS*izpwcY&A@bgCD#KZ zq&McZ7fe8k2bcw3y_>?%_%Qk8FPUlLA(mITyh$sjGx!iwcIAfDXpC{>*NC-Okf%J+ zuqvrZIKcj5+i=E3Hux@4ShHZ$a5uk?h1iQKExMHHG_qG^omVUl$|P>4woyx~k}=^QSz7FQ2lrmo$H{0~0C^V|_5CE?7TTG8JVz2o(RAGM@xZ*XJYFtOO>*F=Rem@NFT z8X1ZzAz6W@3XmpgUY6k%_LhI4>6f>GR+irVyU7%9_eQU&Q%4^Wly02^T&)t@xxe~! z6!oGv1t?0T{GjW%F{Xh(ccfc?xdLGkBFY_uLDPCj`!@RKl!TdUDVzw4fMPmbcfwgpg&`|-N(t0NX!j=SHj&3L`uwOf0FrPsNu}MOUpBCGlNw40P4Hzxz5DM{do3> z_jZO{L34N+SE3`hh;g_~7;Ny73&yRg`Ejm}@UjC9Tcn=#^RnF@UvW%5^&fnK=VEu{ zklS{kAWx7o>)z9CJ@x-g!)FLFOD}UkHcfEqaDcUpP|<-H_upBm(G@px2UeBjT5CfW zswJH#vnaKHEd9MHBQ$q;2z)KkBiZqF=qQmojNCZO|9yn`|+B)`~Y9iQ`$G>HGNo{szi3P;jKmq zi`XG(vq7ViVpvM*=LFv)C?#M^>Tw-!KeXg!bC+4GKqL~F!p9-s+*~E0Bz17xjvcl= z&B}6tDG%2W<_sa_8CK-ks0Awh1JF4KZ z(Ojy6+~?s;2?z_k!CDAvgZDnQJNoXN1;vv~?G(^f z_VW5f0`=Fq$E)+f8Ujp6qX_Q_TdeF)M0O3z1EfY@Wwy)fDcnN;10$snEda@yLad4_ zY<_q07RZQ_K1|R)pBnz$+kxVEOyqa>!#})0=BKxC-_yK=xZHt>x4Rxe?#kbx;{hjlOp6eSTU2 z$L(qIvZdr$&6Miwn>D&p8_}`m`9nKe*+C4=QlPf?)QPzJ*Y97>Ti<)psAOh;q-9$! zEmSR&qaGgXZl*Re)^iZ7v{j?WNEjhQa$8SP(uS`0ab}*2U*6q!dJoA|)gU90AUJz5 zy_0-zh;B;51(O$=_Bz3xYHrq-_;c4I<(Iy`Xp-4@VHb=0HPlzw4U&BEz75AxmfgzB zA1w^h8aB^U5pAGL>w|YCtd-evyO)u(Wa%Xh7+%~rM>O`-8l^Vh4&z@P z+9|TBQ( z%*oYj8{lZy!hM!A5ht#d6icNv(|R%xxh24Io@JmY;@-E@07hr{xYzAxGxeLAvTf47 zXv=+_UWkZ6{g;f>ql;4?YRnB3P(DRnD3l3h!W4*@A?KmCmJVN}mz^fmdF4Cx#T>(b zEe*mIPt*5pXT$A%R+0bMcAokfNywyq@A8fo+c(?4-F(0WZk3Z`rO~Gjo)fk<0=d$f zr-61WCEAW^Y?i}v&#?_6o%_EVaM_1O6wHQnqJm)10BL#i%lxfR|%_+UFtliJ; zqd1)dlO2OAh^@X++S4_w((K$wuFyyy{66bPRTegDX2BIfph{mSWx>e*VFGH$=fk9f z{`67ShMJc1XPpPAC2{^tJtfKF@TXj^KkRHrBlX_X`TsI_H98eVZdjp7`7Isc@j)4OAQ4Ux?6v?s|orLCwnV&P;8VmaX-JRC< zvi#5kw8^A4^$7>?SJbYeD^IaB{1$qoF~F>JvA+w={O-y7@5lnp46PQo?5pPG`loc8 zkv&~D`%3S9N*U{z-zPhV_SjmBKA#b@bpzC>vQpr3c!4ERBq}f}?S70m8KXeBzS6iZ z-VB#j9Y5RK{ff>jE7nWf+!y2`LmC2fAn3%|6Q*yu-tV4pK@zD=d#~7 z*O4f~!eRcFhd0ZVCwvA<|8E<_7+cP|%s98N@w*ncIyKq6l4~^~g$1t6mk-(mR?Xe+VNg?}xI^bNohw-J0r z{EEInh3yqBq2w&%P-nTmsIJl%CBxpIj;6OtXqjadN;Jd04qF(A!tT}9@~%T(k^Kdy zPT1Y#4JAInLm|%2Cs46mw>4ZJpmHu@s}_p7h!v`va7lbp=2G4F=9R~KY%EYRdkpsA zJ@{(K_-$T&v%_!dKE+$-_=MAdH~kKZNY-mzLDXdm`kE2%QI^wdbnk|kV`5&S=N4Ys ztm**_5fnbutO6;oNwd*cbXmp^E~Ol5aZGIYr>OHXvp0_Gern)3(7RfuzZMg(c+=qz z(2hgZKEBj2;XMPlUn8h%(&0GUn|}XHe^vd}#`hvIO6$`h^tX^O1Q}8)eRJvWrTcPC03;0tklVtcyP9+D zcY|L_g>lbqdG{awv|@&-KRB9O##!cp@i_m0D3X~O9!z@K1f`1mfGvR>>3QyzAA>Hf zz00NCIb#|8)zT~H;x@JN1+K_ZUNvLv4vg8?GuIss${#i%3UHyHh+I((u!po%-*#5> ztCcutgv&Cd5ZpF;j_vaBc0_xPT>JONP7&91q+d{B|D$yCb*7_6kT)K$qic|N{6A-Q zQCrSFUzZNWc7G+KqjWO{ALdM655!v+d0J$%Pzy(#TkhUAtsHRSQ5Mm+&&!zx=B*b$ zs&-n4W~T8ZugYD<7mz27xn@xVsvM@(_2HA&xBP`7RS#AaDu!zojnaRIxsk+Zso0j} zx)lVOk$uQ(VoLgLQXU+_?+n1Utl9=CCg*LIfF-G*lA* zxp+4~E6-!rgA;5^S3hz*OPn^!vB5SzxS$fc?ij{)9`m{Vsdin=r>qGV?zgHrUW*gk z+`^-Z^~`?T-yAgay3)ocVQv*d`&dvk^zL7ji~{QllBHC`CP~G~hG|`ub3>^>bXGav zRqaG~5ORr3?Xy&MKU0B4B$d(O1Q{qrLE8T7*b&+p%`O;!$7OUXACc8iGx3>{_u>?pmx-!^APz~4(5Op|nd#z3 z@u3HvHd6uMM~;XanR`$IP7K-&UkwZ%s##2}n2rTV;UPlWFEy9r#ebRo(}$-MfOWx_c|o)!#awScd_k{f?hI$N}1=6 ztEzccemU~nY{+QBt$p8=Fj^&XsZt@1@w?jSCv!^NnPyHmF7|IWewCst*l-N%SUWN> zF3r=sBb91#Up@v3H7QJe;`8vPr+gEESq->oJ-Q}lud{)H{_1^fO!-ox4_cQaMQZ#g zQ->c3>6uvF}*!X-qONHp6w)Ha-jFOU~s=68tEj2kxY10SOElLTm?`l@5 zQoa6g0nR&39_4kMjj5rIi+$4oT$1!byY$vFsqZn0ub;}AHr4J%AM5Kqu7<=?c`)Q2 zlnh-#)WC`}PZ6{A@>kb?pnxGNU-#tMzFJrS<)ZzXGj3?|@c#v1R; zzl~i={&v=XGVA)BS+$K;K?zUeX#4q_c^0AoZ}cV;3>%U4h*kWlNx7I&-gj+=l5Po> zkv~?6#M^DnWWUqA1)O%4)d?w)K9@fTjG4?O@Umf4m-6Qz?iUvn{3FJ)!Txo_)$BB& zydJfizfcg2S`HPn^g(`g$>(jb1G2% zg*_;~Uo)9>eeyA)S61)G{wrVq3q8-OybHzBdThPUy~za5G~8Xt z(5^Z&Pm-4stDc#*cx2KU>|i5oFi)B&=9?9>j*e!N!F>J_uk4cJ4K!J4`(N+0Za!>l zvl3oAihZNCeEiw9Jmjqi2h~GrX*!PCfw)@{SiG!Q)H_obRVcXNXmiU5B{P5s{b6Gp z2NYxtCoEicBHknA5t#USIY#>e7cYHjX-1lC!-cx3z^Lj|&4m})duUK<3QkHLq-jS6 z4s)6{0SZ`|L=5dQvR2x7l=~Jtbyg=Km&aP~QIaQ8DA02RlKc=A4duCJ9$WQ@Z0_HG zx7f(@lSHh__aD;VO&LS`uC}b~CMefUy;cu8y=Yw~{PxQayK;(39qya=I<{~KNr@Bt zZLptt##HS;;dkItow?!b7YgzTS;rHd8}|e7`ls7SsxVeH19n$CCvOJ)vUq?mSg6xN4lhv$G64ZUq?^GU65$~FTX%BNKovurI3>m zrgPtr=?loTf6XDf7birm)a+yC)15~~euEC0tSIb~YCBd+Ubz<}L}z12(>^K(47D+) zhKWO9M94B_VRA~G#yJ-&5A_ndWJNp)RJeZ#v1^3>eyeiny09!!SyO~gs38*~r;d#; za-2Ok1C;yJ{g?!c$W|*8%Sh~GL1Ds#Mg}6%+^=)rS>DC|7;t!n#pX@2_sIAbk<24C zsuvC;Ro~K|=M4~s;^^@4r9$gD_q)TnQ~&XdN{^BSEGE0Vca zUvs=}-9mo?QNMdd%3Z}JQ%=XUDTYKtwsA>#56*fe#GAV0l(JjoJ3g8FdRXmz0;>!syX}_X@Ex;j1X@qZA7VhS2bY={TuvA}xvY z-st69>z3CERRksH4)YwlN^G|j*fwVEPRp*f#E=)gp3n+WI&T)FRJQg2R50qx{o2^_ zZ-fQYhy^1}-;;%tLmGt`u2Y=#dvvQ+3)IjOz9k@F1PZQ}0$$8&xOtLIX&=w?(tFMR zCvv^4Ep>uc(dGCLJNM|KH(IpbW0L(z-PshKz=k$lf$k{@Vb1uf`A2mU3(Z*;snzl? zF-1*qlH4W8#1y!!T@h?SU~XsT{w`bUUFwfq?)tKTHa2d7sl&2iz0W(t`XQhiautgU zG-!d~O>91#-wYU&4gHfHeNd3GExCY7EO>gQ;oRqIPh2%!sWK;y!xWt?XaGU7f3CrO0%w+= zFs?9+gGvr?Z)g4zQ3VqfxNR?v2O}pD+;H|b*vdnvbEn*=soApkfa4DMIViK!ZAJq^ zqLs#zYU=oR1z;fru|9%rgZ1xkridwx3J5XXU35G51A--sW(U z7vV$4I&PBg{6E>){OI$-1jE@FWbh0LX16zX*i_Uneo>;k!F^1ZgBJC~*{G!&ifQ>t z;Y2Ebvp$8ogIqyUq0O_i%Sc;BXPGQ-t@AA_+YMq;dX3DIqp^k~?d}GA8qq*NK`l45 z23>s%K1UzBm@%@o5q@zCgdDk}+ zGkwY=xlhy^XtUBv*t0+QSrV4~W8R`M_8YJ;KOa`%J>V^G9ICGdBrBq&1X>8+8UQP0 zJ#+X5z3aM~zPsMmedgN&h{bQfU$Vdk1Xg0{VU>xX^K#j5?VC*EoxZ?h#*V(F%;Kc2 zWfCOSk-JrV;sKw0f$_9MlKg0~dsRmc!Cv2H%++DmBi|wIuv}55Xu>i!zr8E^)89J2 zP1myNU+5VBo!y1){jpW&Z(UA}i{^>-Q-HH_VeCLl18n~Uy`pOY9riso>qM+uxjEp<_nwhRIB4pSgO#!0C!OH+N zne@oB&f^P&09$YsfViwE0(ReIaFerPASrNAnx+?{VhHwaZt4U|tTuDORw zuGEtMQFamh61107uqMc(g+1%9R&;WV7V#Xr!zqUNk!-xaIxJG4v+KAZWRqY!E&P08 zr(1X@4QY>-DOt>5xfQ&)d8n4|IUdMgXY#J>KCG~{K4LzJOgE)DyS8nP+4W9a)@ z_5uzcFVwodX;J%-rPHx}cJa1p8M_p>@_-RVO5p~BU#V$N;_`_AKmRlsbN!w+@@u2_ zAXQVxwXwEh6bsOL`%EnMVc-}_jk!g?nVxl_@^5K!K&|{bfq07iMuRPU+5Kr50Mzt{ z$)MEGwR+CECeN|Sx$Zr>jg(POBqO`6 z;h4|&*yxp4{d}8{ySp9he(H|o3dB9@e8rnbj!?XU$Qy7$ zqK>rJ!n|r9w=7<{XEORwrZAO^3zF=Xe*pQ#M~PBB71c?NU!6A&Mue~b^b?}P4dzOm z!Tm8;ns>LmSxKK}r9R2+DuQdZ;ksn}#dQ&5=)fqD*mo z>Jo(3?4C2Z?)u|<-TMg)eV>kmAxKkZqFkBiM@DdP;)}r z!%i>lvC_dsmY5~Pk)i%w>*NK!Yvy#Um*MM$x_5W@~dq*eey+x6QvWVFo>~ zj??hfwy1wtT6oR0JWsuWQw7t#QLXUg5)3wpri)huPs}p_#{jT-*&IQj0U|%jpv{s& zy;g1rx!AMe>aT=2L+XfA9Nxb3#U(DcMfZ>Y`cCe`;c#&9bm6G8l#l7-<}Rbzl#t## zzgz>*RlvPjHYkGic>5V+nOKe=&RaIJ!VkSK)bH!5O*?f>nv~Y^mb#OHOq$gp*kHlZ?H}YeV}!w zHneEaaW)|T427P%}z0JDwM*F2ziXQV(-DUa4!2MZwd zUf>V_AV#leA&+W=x@i`Z)<)xZzf&clzSE<(g(2w<@Yd-HV9KMcDRkhqI))zXg zMt_cLKP_%u=9Au!<60J`jTt8X=+uCrZ2jQA(1+-sv;sr%*Dq6I(lf}+i);+N**k6W zo#f4Qs7Y|b#QVz7HA#x!3EGEB?L=WEjob13wyW$6Duz4N1R3t+Pr&A?N06g^*p==^ zVIe=E&x4pVSOV|DFWs}25eU(}zIN@6`vMtj7l6kC;T_0I4 zc5GHVFXK9|!KGHb7GjY{?f*|dH!9@tw>`vuvP}Zr7-{IK8m+Nk_*`!*9}`iGisV#i zc(z}q@p5ZHFK5{2{hIKiOC>ZdCzg_@0)N*eS?fd=?4pz&BXcI5tKh}lZ=y8`_F|SI zm*w1nOy+G@!fm*hiiQZ(Y+m6}M!M}^g-qMSx>H-jKD28!T!ZF3qiY_m)+%+Ape71O zT)26K=de6*lq4R{jXYwnryN@IN!m+bVC6VeTqw5DJ-=CxHfX8U z5%jmHyfDC*=-c?#r_9fnP9eL1R^P@Eh@NL2X7P4P4gkdfZMfr-Rok*lO?iI&&;Nw0 zOgZ15Q$RpsSulPNq3ft?km7y*^B!0s`r?voFFtl@e4zj7>0BZg%^5m6T>i`gZlBtu zC>KS8sup6EO5_Quz2suA$jPAib;;+5f`Cxvg^+vD#%A2Zi^EQIxZq|&Psb`g^46dH zUR`qGpD*EnFVWL0@&9o}8TIKi#Bmh%dCYz5#eCD4Oz(gDj`Z3VTYG*wo6`^^P}t0G zlBd;wv7J@nrb4cT3HciRlG$OWylQXI33_q+Dxu0LLA7yDk#hk1B7?VUO$b#+5Ca^Q zQ~n0GHAgn!`ogr)^)2ghk|={#;Lvf%l{+8y8mXzcav#v;_MX4dCcqS z9|OMcAmRo7NGsx{;L2hcpaHT|GfUM-A4ZhU!VMyDNIXS8?nZmF7Dv?4CW2C4e>KojPlXnvvcwr z!R67}?RYDx)x4l?-Z4VQ@o>+&6YLjp@Tt#Y)dGnC$;&w!4s8yBS-)Lnzp83^f|!UR zns49M5cw@P%Vc4M8-YM~5f!-gx;1xv@zUVB(cjW@BWFH_cx=-nY5;%UTmOuI#}c4V$W7TCP*pab`2oHl!)x;-32&)g&Lbvuh!rNy$zE#c#5>yK4W6D1 z!40G3j*gDf^CQIAT>+f+%pUW+Nrb?5aA=5sV3$HM?6NrFI7JF(wk-80$C&A_UqhXa zc81C4?0K0don|g=z@sI`92^{4%Jm95Z>H^iP|1VoJyd1U8(sn2Quxl z^715QWiR}3Zh@S99Y$eEcxwMLczpT+V0vOLgCxlAb`a?7czK-yD;f9$=Cb=R!FmS= zePVe{Q;Lg;Hk_#5>DXKVtI|A7+2l@JcHx zFcUX5kTUb?@o%dWe#}~yJ5J=-)bQ|IL|~^9+bt@WXXu1`x#i_FpQsIf$)OnUoeKksSn zsRJ_P=CZR{Ve&-&nLG)OlM=TRS~Ti6NkJ&&aes#5-bkzHykm)fAIvu2PoE^U6tE?@*952Zl}q0?ZT$ISg%~VbO~V;z1aS#0>>ZA-%4<3Job`T z&F{vdh$Ok!th{OWz7lyVB#B;|!c6Y|nLwa`+R8Vq^Qyd@zaf6^LzX8FA}0&79G>7? z%LjUWg-#pODaV%&7nMC5i;X*80NrFJx1_t>wRR+=&JA(2xhqixj%w{=0X7dH^e`@1?s#- z;Ov7)lnHv?GcGMa^&{J(Q{d39Cw-t+ROAjxS<1f!(qa|+gC#Riq=+FTaJ$~ zr?5y#Nh?8UXrpH8h=ZtA-(sx7YKl76NSTBL{wD3}DlCcwLFCCN-f?}gJK+d^rYVh; zL%ODD>vM>dFbHs_*49#a?mBwDt#jQyfVX*X-Fhu{_yZr*i-KWo+~%xZKMns?MOQZn zcvUyGO3yc(9Ie5y^a^?$3Ubse^OTmCBYMP=l9Jf)w6gw8c+uUeo@5w*L@noEibV(| zW2$Sjv&g9RPt5=Fgs3%yrZqxY)USi1rgP22`PbhfIO@(QveAb)xq-0LG<5ArUp~F&TDB6>UWD`V6%PJ1LU8~sR8WUq2h-Al1YCc=&($Y=rn!dGfQl;!d4Q@1kEZN#^{@my#M+)#_(IjAwvU;WCJ8e4SqEG zAl~0#F2&`3V23P);$j})lw7G8x4}bHr{m&~Ef3y^JveX~x9eA|glQQjm;PdZ?w!%` z;XZ6w)=R?$(mb(EzXo$kJZn~~mmnmRm5brFQe(l#ix;Y$M1nzkk_s6=C9$LJ6tQEo zv19y`b$qZ3f6fqmhfop+hC=tnC+t0N68FDqS$PVv8Rs@Shp1X1Tk6)N>soSo`OumP zNjehwa&qtj$k^{f_(TQ)xhxNXoTmfB!!w}HD=jaN*SC1fvUWpC$8Dpzx_}T>HqG;| zbO*(gjP3^aq<>zM|J5V?f0DI;K{9R1Q$eS7T4cjOl(N@~?3acN;2DDyTCBo3Bpf4A z4nh~QJh5JE&dMT`I>7}rxXmYF#ut!SVg!cRC8j+svtJ^LU`=CvyoZ11HL{ekULy9y zMy0to1FZJiw!dg;=3*z!E~K006`61y!amWfT{g%#}~)@3q~E$ z?0YQ&^fg)9EIaCzNF+NGgvnv`j+9%(hiHqV^bDYEa4R!vy}UeGcNH-=Z|~>` zQ(&@&Fg6W2LzCF?j2Pk~ej4j=)v@7o!|_ss$qD>M3af$%6WVMlCP zy~u?1ak@F+k6bV~U2nOKAqit6;&m=~wB28FqSxSY=Z^wuQ7iQstUwwu>b$%>A=xxB z2#MgAj*^u0Sy*AD^F$Ky{p^Yga!?Tlhi zSN3*gAPz&4cUhwvVapB)abGt2)}RRX@|g(^SM(qOT%JOtbzw)@>qP-O6*-*A$tkG6 z^6=rq7$IAx{JP&-$RtOCc7|fS$=)dB8^i_m2ds$7I%o%r>a{8?ufcsABlJ`K!)fv& zlddbd+7*}M`7Qd~CqTUo=BRot?3>6Z=R0FxO-||^LgLg4`JQy%jX(FOnyJ)`B+!V) zGtBSD-9}*j7IWL4b#xR?nU~a&jX|_5-1iXLG=%$R;PNg8M(VKX_AXB$!deOmw;%4> z!{dcjeKXPX*csV~k+gYrqQ)8)0Z^1bC88>9vMTMILdc$`ScP1qf2^ugPJ{CfuE zv0`pL3&+m`O>=9g;Wj(r%DLSSw$Tk1leLr}8V`EYuV~Lh#M%SdV?Rhyly~d+VK|XJ zQ|Mk+PIflz+lDYj!X&8?fq&ffE-MdW$P1ldeb^q0I%8H6v^}d5*q%0)I~=pXWkDE&!q1^$AXuvq!KFs?ql!&178tr7krWXG zh7KJNq`T|eqrP>%_nh;cway>ka=F-Z-+Ny>?kj%R@7lrYs`8ZNjN||SN<{@Z4FCu+ z01!-{B!N$!L|mwWepI`Jw{UiQ=x7d99y&j^wRg6)GGTKycXYC{ef}Rrq6g2q%1Ry?<9)QbRigJHwx+gA<+1n;< zdI+s_ylm6bC@rvXo+b! z_xN*&j{Y<49$8{y0Y-=)P1nGgpoPKv$_9 zv7cvr$m_Z^%CJ4*=O-t3uolzs^0f30oAhqG+8w#3?rwGF>yJ|kLr(zz=4jiI%{N8f zoaK+k80{MlUFGCnhK1dBbKCyZT2oUqIn^GeC6t~wzm_W_UplvB6t{J@jbQ)Tr3ISE zcWMgk%KK(;y1Hc_A?dB+@Y{ZVehWjT-0bY+fxA+z!^6Y5Mr}Jgad$U`J3^M0-Z;$l zd07cPeOaoe&?tS842of_Q$q{b=&!9fwIgcmC+e@e%xR8#%>21JZ+v{icRgW3PG4U? zQQF6IdM5FP3$LfAr*(x1)S(ZN0b8?K$w|A*QeIn49YPJv;gV|#qrUr#s~`D_nkdCb z&j{77$c{Sqs<;?m27o{P?UxEQ$yGVYgOv}fNBhgIE(3;i31W7+o7M1hye@B)yD%isgH9{DwZc~!bX;AVx)Jsp+XYHqdV~T1 zWKf2`Y*{*#H=R3hh6k#47ypBUUE_3_5*1v%Ro8n{;fLiY_;x|Uoj>FJAsRse5Q>vD zQRa2y3U~>eE-tRoo1PqG&@iJ#M7o(eMl9u9zb|%}pi;!~1MrK8)~wyh=pX=~3OiB3 zySJ}x{z$GQlIZX2jb;Ymz^_?R7YhFJCYA9XnyZ@-$dz+HEEeG;1>p6fqAq_YnE*)t zWV&SAwswghJ^hRpXlQzWGCZZy+&cP!BaNphf78t4EUhvPih%7li<6~1`uFe%_VB9G zNt~kYN13(TDa8Kg;F9_2jijbk ze_kl|IE>Gfkl5WikdA2tzaI8zF0ibAZZb27OWoGOzgHDLO|77e zu93$ZZb~}8Tlg6i0l3W5F~)&+6-pl+w@<=V`bUW-0^ zW;hVpLBxBho3J}_I%#mu5$tlpc8Q>LG4Wal{s4b_U#PtiJWvDYkO3tpF~F>4^fOWO ztReNqKEhB9M}V+^wqvW!RB*Dl#=i%y>6&RVk71@8fij%6?PYVObKE#FKU7RM;pZft zrXKqX4e@uE36&gpPWgYjQb&>f4DK#7cAPj``JX-vR4V`YtFvJua7Khm?8Y4=-iSmW z_ApsF$Wj3F6$Kv$uDaK>Q0*erS6_QxO*zk+b2|KGW^IY>mDw;192OBFKc_4V+fZSy zwQ?AieB{{7h=Y<&AldGZMsjsj-?&M8s)!}$7_DfZ{zp4O1P{w>q&Z8@Eis8y;%l6F z;m+$uBv;|L6gn`wu3G0M7Bk;P9^Q8`0WY9rM?rkq{EUaOU%z`g)lk|LXp82T?&8v%ZpN@wmjr z&24(Pw^{5m-|ty;9!33Kx~IU+$A-3p?c?k1O`>RxlBuuXV~k(<9nJW?TJ}5MUkyLr z311ys_7#7{BE>jTVIOW-ahN?a&p=xZ;x5V_dDI{rvSh3b|*S^ z+=t#NOR57e=fVSpQST3PNPee#Wx9Saugq3&gr_6QVyr(cDJoppPmi9g^R%5^pbzOJjXY8w8>PZHeF7Bb8~5Er^QJo zG4ooc3m9anN{;+%u<69 zQo_{K^h#Fqg90lF2B3_1EK zuBaJtiBnQ;eD*xW`g*K)%G&BYiHL@Ns(PE;w_L{Iz#MaSvi(akJwF`Pqb(@ynqGX9 z%FCKLK+8bvVIzv+3M7boW3iJLQ=7=8!xu_3;Xe(1Nu0>`v!d%)F0#i z@bfF;4>n1Hg9HT-fI8-3=OmdJ#uL6hxzK$BE~x^U^SN|*Mf#uI!9Ds%XbM+^%^6fc ziTU@detcLDZJKdD$`4S6$Hi_!^Y$Loqx?P(G4`)IG~@cSmzYp0)y%+LkC#~#NtKN`267k<|Cs0R=u&xVj zp=3H+r^Cp1um(Ln#Q^1sAE4BDPr&+asb-B=@g0%_r4L@K1GOtrlfgC2KldJ;F&fo~ zB>s|bWc#sW*3xex7jJZveL34bM&yW2c*a8?C3y0+&U9)lE;Vo15Gy1jDEhPto}jdm z&H|SfZy5P&XLS&$>y%YLOb(23uK0uh56bhYR`-GbJ?1T(bLP)wpdwElB&?I!qj z%th*OJbh4b=Zlzix&Xk^paPoEb0Y7$TSFKbPS{|%`15@131abLbR$)XCOlVXD2|5j z5Wh!m=5;zw=mQHlq9hv5oiAXL`kZr<{E3s$BcHbwd{l=cD6C!M+O-~95I~##8poL% z*l1h#wj3Ap8r>8#(sCn&v#db{vNlOo)}&@l8iYH}6#V0Haj1UL z1*Y)U26t0z@`QrEzc#d{*>r2~p1;>HtGlV*)wb?%ApN4yZdK?5l#mncB7dQPcgXf*^*&>#B2>Gkf;dQ1siga z#lM9HmHG#A)v<5cdn&TTh}2&xJ2j9bXGWwj-`d88Z{7vqp$H`eZ#Veb#w)&6XHu2i zw)~tV^z%(5QFbX#<=A8lRcxrkw~=DkT>TI*c99A6=|#F4I;lNQ$T;^9o|n-`p+ylB zNy=(sx9p#9s%IxYaaaBWZO0U$!j(rxa1is=nsa3ksQam+Qa}F$B4y1HIvq5l$fS+_ zIj8faYM8Zj5ir+*pCHH8uR5*|Ism*3&X7C+4SN(J7W zMO`F~Bzf;hXe!m16>1F9|A-+%-P^3*(fT|O)G1Wq>$$6p6)c9N1_y*}lu$-r^TjQb zviNsp>O1_tM|L$-TIb@GvE%(5u|)m|gLZBt-%3bEQD6<8$Uq64*y_ZiE-kfNZ(|v;nH@%CDr3?x?|1$5P%F z8f)%FQpNgljZ_bJGm-Me8|Su{+$$kFdg2t86G+Ef^t@_ys%82GNeuoAP!Sce^O`j* zG%X;7kx)RS*fY_(0x2a7hG6n!Zq@8*0ZLR-0?)3AIab7Pavhe<#Syff4WLD84~+;?+8z+=%1V47D@lppoL<}}2e(yG56Sdw?rG&-f{!flj}*x;0+;PYQ>{K~ zgyv=_5iV7FeE(FT4i!w%S2J_ADJNq z3Ua5&?A+YkBsaQm6?^Zl1qJ=gn=T5xtKpPjvS_@Xj94|u(K(ubDeHbOT13KzqJ_JuoK!4 zw6wGoH{j*ziNiBG0zVlYUmxN5B_qAkE|=C$zBH%r^8XjW#UtQ&pRlTxn`*4Ur zyY}26_oGF@fDh`SW1dzI$Hw^Z$iaF~V!yDSo#jyWIDf$UP;}b2PG$aR|5MIS2OBFX zeqs4=dh~_y{#fa1*o-eN&9YC#oFdlHZ;zBB$!KCT%TIk^J1e?s*?3j62*PZOeywmH z+3$r0IL$+xU#y(>&gAmKh2Z+UqU`j8+Nn?aN4FYmWBjCwEpSSEm!~Tb7THmRrZ`ye5}w zmJ9bytf=$~w8FBDA3IOhmGN(dr{Th9J|d+g5aZFOFlthpns-%ITV7qbX<;-{>0r7r zSj_Ps0JkBX(t2A5XK|qB*}Lm~FJgxE7FgxO>Z8fJ$%i3?;v(0SnxmhSV0iqU)$3zFHAvUkE_=nVI+5_CyURJ+*?Kcd*vC^X|&x+(Mgo-4%!~ z8Ic-MZ-#r)4e4@o4z~M-t&XX>byN2zymjj&9kZ zNNHbvM>`*wia{vfwiZ29iZ}ssGp77Q%rb5f02mT(m-M6pfM_QOHZ%~ffvBwd%JEFB zuNJG~{F}T28xnuXvnWNgN1Y3@5BoRr_7sps26@%fO8wk&L=jw^q?bz6J8k8fO3F_k z>*VF_DI$v!?-)91x6l!F55thqW>Qw%8yI9SvoS&4Asl|sVT=!CcH)F!>A9NBK4sWO9K%hZpJTS+AMK=<$oq;;WYD?y2n{nN_;Wk)FBl4>M6 z5&7~giGThXR8B~1PbW_v-9C75Q84gcG0t?>q>V=!&BDT<;$g1Y(JT$tlL|+nsY=!{zb4C`1Fg(QvSwNUD zCC9#mKkmq#ECYz!(ahcIUQ{dzWTm(7&t&0wMxKI`pzS7pRB|jogroi%Cyc8fBH88ZfRqF6}iUU~fzdWA?8BF^#BfT(giNebbhO+)NKs;j{4*7ceyMYK!LObFT zXWrr8%*T4!Z|FP6o1l)~x&X|Ap@dVAz{oX!sj#hy!NO^C^A~S|Hu8XS!r|MmNuX3( zu5RtNoXb}4mcavQQOM{|#nT`a{z4FeHo*W|rM4b6O8A14RQ*W+35bAN75yNE0GK5p zC^+d+01R<;%aXNF5FyH|&XN562_Xq1K)CY!qncFkU5V(jw{AFqYsGQO?F5xSy=BTr^Y_jgJ2`D8T70K}`kMeL6{ zbv5fW>|uY*&IShs4H$U|6jn{g>(+U?U@9tFpHnj7{t6&+F+>T#YDox$ZJ!d`TwPq7 zAl@=K3D2Wg&XB5ptE{MVZ*_1`*V)aj>!30#i(55GQV%OzT`gENH^7aDE0eDhitO43 za^JwRl9f;ZGb~2^E-67Oj@Nob>+-qjy8ixtZLDldYgg08#zEJ}Ty>33OV0h}jJi|T zKL4se!SIgSkk%{Dw`>2<6;jP?;DXh1*xQ(mV+{`r>x|}raezbX918Aq@4D}Sd+qLY zg2T7Qz`+_fdwgxF%!mnJg}K-MhrGP}zX`Yn1F)jX`Q|*z*mE``?s#c=IUG-*iW;AR z!_qP!eFeI`A$9#X%KLxPgcV89j;}L8THeIl+Q(gDQx6T>g@Pg1V^D)jsp zKJLGDg8xN_vMmf<1UK|2#WP=v@%|9*n4a-R*ipTfl@OdggA7CV@2h#wU?J)pkAwp1 zKUVVyz?qBPTw=eEQRF!yK^>-edPPK_o1V~4_IGMCQdaAu2R%n7h8+NQ19Ihj_f_Q> zSmlCMlaB6*3MGGdwDkBg-BdijI#`G=1^)lmM0yQV_y-03u7$!v=N-6HxT!l3uVG?{ z!4$+(b9m6QeUx_9#rFm~dkZ%g{s%R+tN>7tgz&5ao_pLCeG&K;DkQom1o%r*qZG$> zn#}M=W^gx!(*@SL%&K?*ye0A*?yo|b@-It$-WgK^ny?-L=|)YMAxS#749EFXCk+U2w4w7O8*)o-mL z!r)^;;P07RIrSB0O2tkyp9f1V)%cC;CB+_%(kCBnWKEtvUO8=3B^{5@v21>&#_fpc3-zC)sFEQWkG(MBwz3|pf@qAS;D;3V={bwz| z{9Q2rVpIr`U8;KQHWDUF&A*bR-gOzvJeb)rWU1E0Y+cLkciTL_`h#xIvabJBLs32o z?IY_m^b5kLZMT=0+bOg7Y8yB{#%k#_&-P|n`bvt4(Mm~4z45&PK(<0NH^y&`+wbhe zPK!A1Pq*>l^L*M(r~P(5?2I%Vjvxyu9h+B90tNZzfUVrT!jJ;}ik3{31l(wqlS7x( zb`!VsYB0Xj2Ye?rt*AyZ+rcwTVz${>NN%GUSs-DJ@QFJ|BKpAn1U_^JGV0nYJo`os z)cptr`ut%dA}QAz6>$3eyqSvx_(`YTfnJX69f%Sa?EdsVDIq{ zjRu-|8G?c5g?win0qB;5qwjdrTNCR7|LHFp_)Uwj{%`t2yc1N%z{@Mc%5D71l=#~$ zzwfMsjdEwyYr}%fWf2stS*tt!uQ_8e)6f4E5Q$rxYkOi3uYg>=ex3g5)2FM0rs21r z`80a}EN0@37N2|X&1P{lN0Ly-x{*)H8PP4__Q^~t1}sY{B^_;5}q4FV{3b|bIFvdZar8 zC>xtnwb+@z)q+V#Nz)n{CX%`NfAAUaD%p-zi{7|-lf&d_;wBWdr5KtJ8yj?VbhJ~` zuwS$D*{z-iI`ZQ)@T`yDRLQ8F7(T^!j(pw7#>PhZA*Wdwqf6eAAU5Y87sh_1HaMx0t=Z(2$9giaAJGu*z(DW{`Vfq(Vsb&l~66 z#yu<#_68bW;W^JQ{rzI|T~YV5REM|)mW+?S8{>Vl@bL=IIr3GnhHbC7eXmwByXS^} zhno%uJFD*No#K;?ffU7V%j1=^nF;V(36JOB2r9|dCK|jYQg4y1!>^^ky7J+4YhRz{ zT@{t)aNmuz!AgfTpGJN)nM3D=u^REC+2kW0JV};(xL`T?TAn5N)vH(T@D$kDae;J3 z@6?c^UdwfK`18S&R&m0XIC!ruyp`2Y@;SWYhL1AP;#%-g1}X;4@TvJ9KY}jYbZsI! zEqNv?Dhki!&i5B6d+DYB$jM3yX>V^IEH-WZWCP3B7`0gQ2M^ecVD?fuUs!{SR2B>_ zvl(cC$7$g4{mm8`rxe`G&kgJ^H?R&Ex*6IIV|W%hU%!R+WIo*$1r2RyUFLYb^4oJJ z6F5z=Rc|KV=3#ClwHmDE&5m^`KTpdW3+bXnb?Bh~AF$b>Ox{U(()q93eDJ0rP`s@w KS19}7$^QX4#N{9W literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_218_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_218_0.png new file mode 100644 index 0000000000000000000000000000000000000000..cbac4fd70f9f2d17f9f7159143d4097c1198bfa4 GIT binary patch literal 22611 zcma%j1zeO{_caFM1p!gIC8R?_Is_?^?p6r}1f&KSN>o6QPLUEpke2RHq`PAn5UGJ- z$f4^yGu*en|NF)D@`IUqp7Wg8XYIAu-e*EI)s+bGsqt}ea0peDA8O;^TyDU@xwLfs z8hFPq@V+znCFZ7J=%(Xjt! z#rgOg{&@kflgo3yj>M`I(8yn}l#N_*a0qT=elFpBN}MzB}Da0?8C-E1dH>-@Jtom6mFYm zkN-F^sSAb3@n5}X`Am+n&H5d=fhGj8iVL;K%U|D!T;;7@DTL;SkMC@CxXo6mr_Kd9 zMK1e9JcU*AfL<}aP3m+6zj647u7h#oba=iq2fwT&<@mra0s0II@QW&PfB*;Qb=a~p zXYswmq1&%k(yfszB2i11Hrt=^E;i*& z)Gsvt8cYvWR!~rg$#U%zQ|eCOp*eE`?Z5sQ%y1Ub)pb9=kS9yZ8)Cn`z!8(x@3ZvC z%x8<$xgoQ$(R_c_zlz@T_nhN+pSXSYBh^Tx%X*k#--wzBTman`8RaMMmOBH4TmXM#KK<;c~ixvvWxTk4g4&y1)53TH`#cv{V4@ zwVY~P{c7&LKM9zLhrTF#59OG>joBPN|HHddhYQ13{FKbh3Z+jr-wPLX$n>?z$9S)r_%MYX=2XmTpVvdvBrR zX;O8&1M9@Eo+$IV2fO zRz|hVd?*+B%CFlwD=oyc{S(dE^RpJ{>F&4{B^Uz%x~AVO zM`YH=46Nxs8Ih8Z^`nr+-9qncyQfD(P6)zlZ?_M9N-UdZi+M~c-D=ub+*aca%m>Gt z&fE`A{Zie>jAB^Q`*UcV>a_%IdIp#yCBJxYG#C)eO#8l_07b-R!}S2V!Pp-y4c|>hSSP=hu%D5=@N}sOrwUEgB8Z&o<7BZfi(t!73ewd8($r zP+0hA$($ehuS`Bq*VEF9#;km8Ek4|P!-dDJR#!uEE0%uN_jETM>EZ)(s9G6LKP%_% z9+|FLUTK#(%h+3+%rvWmtoontFH_6-)$13T=E?`>l|xVjb`B1s731dlIvS}(Kjo^x zm4z~=D;iz>3`sHur$Oeh%DPm+8=0`0vVLik;>M#qdanf%9*ag(Vwscge&|ew)*=i4 z)7{$NZwP`$Ci3#23QTaAptet=kDl%b>)K8P1nq&@fbte>>KR!&=+Vtjo3${uXvy;*Uw(?F#|ov{6w(b3xTbCvWnw3ztu+K#TN zr1|mw#s?`_EL)mq9;_xbR6MYSMD3$y9{5YoK%`@Si<8Sfj+1j0o!Inp04%ev8Jt~3 ze1Do?ce}e7(IdG0;^ZL2&N1OrGT!-emr%Glp2~6I%+2dpb&ZUuB#6MOAD@=Y%LfzL zwgeHD%qx5#Upd{MH83+vP4n6AobZBGF24Fz;YbAbP7C*~kdDQ!p;!%Ry5rgTNSWL> zZ?tN6$6#oblfMBjE^h5cV;%a>2m4^O5%^E)Y{qwUNMDq89Wtwo5JC5)Ntq_I0!^LcrBfmM~`9%$T~nsx8y5#gjILUwWS={$X!QPi-P?mnnMfqY9mlW1*J zA}ekxy%myl)WYCb{7ST7X~_n&vvShJT@7V2MNWwtu3f$a%>A)yBCj4h=@6NLX$F;` z4I+rfaloK)n+%I^aF2Ay+LgOhK^a3s+Hlu?sZws$Na>?*349hsph2}XNu$}*b&uOy zsi1oUZ|gSBDjWA3&Ig7y&Z}sNa@HI6WfoqEq67J+vrL>Cz{E^;b$5?fHCC9?M;Zh) zzwkZm7FofQK}Z+x*CLD7T_pWZJbjO0zRw$4Kk$vO!|MS08sruhs_iWC41gVx4>qx{ zVbgPs8+ADmvGe|OpSOL|$I}7Ae(!FA$5o0N z#E?o(Jj8f7+j)D7I{BD#|8r@~HGAbuZb`Hc`=slJ`&w8|yeQ^5|GK`)J-awYL zrwx?#pGD4i&I^x@jpcp+uDL&ZZbn8%hAb$vv1L)ooZrL>@Ibi z53hPLRxo)q+3)wI&}3Gh-&0mK!ABqb*bK}XftARu=2s%>){MdjJCyNl&nID%{h8y} z7+?bsh-f-PE(0I358vM~Zt_bls$F$8Tc52&$xH1`+|Da59)2$4otwaYSg^gbQ#ZYH zcD%8QwZV0n^PPJtLq%p1yGV<3S9oF5f;m)rDmqNXZzOsW!Wr>nQ%>oTbFBp312ZO?j6c6Lp9dEcr5 z7z`Z*fYVFTJe*4KK4iUmKD%q~ecJ%o5^5gPTRbX5)z0(om}TmdaPqKSe)SnDFLRl` zs*3~D1p#G!{REbkhW(k*QtNY0SOj9)wSPQCY`J4S!!EviG4#}tQZaJ>*N&GIof4B$ zV>br8wrVmM(R(c$om#lo-N5{>L6LtQvawok>J7G~%MI{%o5s_~m zyFxkpCp&y2xD=)n%dS;3(7MKB_;CdUMGc@wV*MZ?^uq_d(V#}kJ=yl0*l&4xjlp*| zqvv-Pk^RM}k(MT?YL$h1p8DrEE$^vA#8eSf7zXJ6Y;YFijfi~dWQK)wb>qf&Dy~+s zqOMAc#r%T&0z1$)Yyd{=edluFHL+UGZFXX8tQ*6U$!=G6Cw{ zTD#LHF*-HHseiu< zC8q+qYbUkSBREy*V6p}UJQ(;mbFllH1~&Z9y(`Mg5rB&)vS55=?a^v}1z^u=$@kSF z0qAu2#$FX5ls}WZnD`HV@*8d8G=TVUlaZ5mp&<4%^nd5g-`&fpSn(#ki2ujHJp?46ACbWa9dl3MkbIW?Aqe&K$4*zgeuf=jP|_?d^qE z4`qHX@8s6k`zVL=aCXrQ?Yy1aS={&Z+pd;tICZWF4c)f-r1-`pz6OePx*_wQ0T>i< zMA!f!u6RM1>D?i5yjHfmnqRyctwcY#z1Y3GTkJ1Yw0!MLUY>ch@*lhTBcH>(Z{Ltx z#(tGw!QwIa=;nUWn*7UN=Pok8>0bT+?yi%v2a$}nO~W*o16~8hZ`uCiiKbtZ48#X9 zaK~Zd`yEI?D~2!KTi z?BkmU4Ptf#N^ajPStr#1w8%1CZ;-#a`S2geR6zm_X8gVKZL}M^Wpou0bZ;U`tsh`< zNSivt`?dG9VH5 zH&UyBwiY(nP4rbU4)C=u08GV_QtRLgGkpX)$NoD!=x?+q2hCH*FJH_WK6o_q#f04j z?-OWZ`;2*L(%+8>9+2p{Y%YRDTj!w;Ia%X5_8@`N8mKE z8CQm(-h(EzuVi>!C0(jnaQTEUfb})Ez{Sn{Hx3El7xPK-ZieL6=EB7Q6mCM5u3vOn z$uz^G_mePYe=PN-=ErjxKq!ZBHT`T8pbWqm_`T0c~H zZaphBEP3W>lKxY?D?#tVAXaEk;hN^=MPyw1h??cJ*zUz45WnXRYsi>={rWW~>*xrI z31pj2QBGvnucFh>57Uo+5c_?0@Js8uhpL>ozB%~%!jLm(qi3G*v#0k{NbDV-PL*1#u1(BuA>I)1MzIKsh#)KcYxrui2SEm*O# z*S|=Qlp7|fUvIqG%^&8tyt=yTw$m$OzdDjzyH?uGq9SuN82N0xxasW3P?Wv^Be(J1 z*g4z=nuEx}veM#<<(W@+8s=XJjGQT+{QseJQ@0dAYXb}4ifr}NyyJ}~3vHpRGKUMX zCEhP#2`z(8ZGG&Qoc`YEuYvbln=Gra9lon)V341itAZRa27J`@*?7y`_r`-(KD)D% z!=ojUCMp_SNF5ydCxorsYsW@qNJm5JYmMu2#K5?++TVmj`$3a@F<1tmybTY1RmG@w ztz-r-+QO)6=YnX^@x?w0cto@V5*ziIAQHw%EHOe3Q$le6^HV5#P^rS_&;=|2kmov& z);*de{#dZb19S~!9xS!KK(n;!B^qwya^(-?JgYA2IWniYEVf7J6)4c&pgxXEJxS2B z!{y+KU+7V-%7T+8%ouT#F=Qgz;i`Ka4}k!dTjhd%@r}QqHE~$iOBj1W5_qQryHhT9 zqZQaG>^S+5Q&?=eWiG}@(pH&_R@`p)VugX6>|A>C72m*zTSBs0jq?ILB>`JE z3ee`D)y}dT|D2YC9mV!d=)4u~BNmmp-#3?bwXxvN0d5!A3AI{)1_>35{A0L)kvDHA z-f*~Qa_6E`0f4*f|2CZ5)e9A(rykk4=sXdI#xLy>xs;2@Vtq{(BbrFoEf&B1v%J%T zB2^>ai-iDQ9_L0k$FPXkfJ7Mxu4FG-1tXG>O=x;J*Y(3mdv1T>TZ$Ma1o{iO#m|Vl zWDajQ8VVT@ZK0ix*ZRx?DrAUN!f(btRNwBPYyN{Bd3l7!OH_<+Y@88;lU@(8HUjda6%p+=(1WqOkR_2%>b0F0uz+n0uE3)yG~QlJo@L406rV= zJ}{wY7}`{;(%P&9C2Y}bj%`7Mc>48K5Rn9sv4=ZqYgq{Yett?{^#<>Pj9&!=2NDZkT5A_#R56&JpPj83Q|>e?ZRXl5T#NYf0sWQFSA~#}umyHR`s5HMl1urU z^dit@QE$qqs;@WY(k}oSo=o9rz5z1ndD>v5L+}-IU*86d?gz9a)23ti2FB$f@njAH zSXR=!*9S)pEDC{Fw4Exqg|3yfE3PNAgsjlBGS7;1;1fBYIGT&M0%|;4C5VKho)ZkhT5e3hVBKiIVpDK_DqGs@OR? zjselO$iBEiACH(`*W5gfHUgn^;3FNMm`I$9@vHJsKYk$L`)%YVxy0mEo|?@Rkc%lE z9yK({uRJXNnGETI7;clXoz;E);(0{V@)nzmDnV>u-&U|^7g z;VwEk>Z8_)W@DDksFj)kAX*r(2z(zL$=Jkt+3^kEx#_=RvVPYkFLH}Xt?w_3emUKm z*?T00C;BKQa_*I(#mOHEVgEB9-<{;yu<#}ERQ28zF&+EyqIed-ZSuNZlYa5jm7cWU5Ji($ zr*=V%uzrN{>8ZNY^{bbSD_Rm)*ACWIo>pNoAKxx`=nem�Oc&Vl zwBg{-hnOte;p`Qlyp6*>W($FP*1;s*gl&2#F}#QQEHI{ZJMo))_lHuLv9+kFvy>&) z``BTWm9X!A^2IDX|3{}hUG03d$mcKi>uxrlpeE;}J@Auv+7a`A{CIym+dmiEaRVTr zcuS&`@HW>?covgDGY65MMg%7@)<-;H!9)j1I{e?%IDv-7R9ToewM1eE-~b1(=??rj zBG1^+*CI+Wy5_yuU@96z?e#zFLUh8{j=IjA9GCYW0#}BWQJ(w)v|X#cekS0tGd8ej z?f@H_d04{S#ftEodGS+9e9vjl=a{epdujAj<5Hh?c3RqrXZJ703P6m%?DE%8-Srsw z81=s3r3)cS@fM){Mgg}viMUL+oTvK_2(Spt9|QW>gHBW)di}R=d13{Ila!>xn90ch zId$?f%DUUI2cQ@~CID#0k7K8urr*CycK9Ff1I$w>T&oSOUdQ^Z<_;#n{OzN5pl^FH z;7o}j_D`Vgs(ovhP(4e)XOsqA>_p%LE)zLv-}~PBk_4u&M~L-XK{Od2?+gR974jh6 zL>0f>n`VcB(1s0HS|SD{l!9D}QssVAOQ zG7f_v6(@)c%mtNEjQ~OWjVI<8YYFB@Rx2>?;tUJwyDa?@E*6o=0YNgBu>@gSxxsw) zH!0yix!&;CGr+sVs+A!2=gifYG>9;i^a&*i^!6|OxL*w_Sj877h$2Je0k{82vMx|f zOC2uu#>boBqK>Zk6)GAF#07bIA_{Ko*iPi6ou1jaOZGP}4+27IK#y$5$Pi=Ql5G0E zLU)4Lh21>{&vye`s4+(rY?DRJe9Xf!?RId}GS^Hyqk0sv>o*6+^LgSf*$+=nQEUYH zq#9gwG51xoYMd9;jq?7xnX4cAkQIPOjx76v5Uwjf=?`Y^K?>(;*@+Gd)hVwH0ic~& zk)U}1bhhVJO(vK<`{)Rs)VmS#H*swq_(cXCFEGl)B_MyXK3!v6<6>RB4|i^-gRa*sKWhha)R58h z)UT}{DC~hxH!gYBI$)A=r$9D@32(pC5#`(G2=oc^GQeJLgl9I8f%=%W{cbLioo?QykSWxiTW5z3ooYz!Tpisjy;e7KhS;W>b zw4RNxox>yHEp}_lk8CLaca<{Hdkpk?r(ecmXSsjF1=Ad+ub99c=2)i&0(U?!gvk5v zk_A2O(7-O{9i&H=2yj`;JFO9?eod7xCx;f(ki|{rG3YiLCpWQ;okNiH5wQbt8+0LB z$qiH4Fc-`~?`HWSbo<)r+LJ?{LuT}nPsI^ihj?{nfB#3(Y}t(sNzpebWgeyO{l~cf zdkF}j8xF8_NFRS#wfR{l!zugT@$R@qF#&J&6C5>U5muz!-wPv6=d&Oqb7$W3gGnf; zF*M{o`K}l=xbyk=8*R=TJerDg*>GBYn_>ytm9w^||M?8}lgGmd_o(t&QlKpSTVDCp z@5Mjz$R!Ri7ty2M2_-TvO{R+b94@c?i}x>#86&0sSD|uczZH-pknMkW-)V+h1!R?S zggn-#Gpa=28U`SqH8-yAUc-hL{8rB4Xx@u7?kLF0EpMl~$eQM4Mh&bt7a_N3cH?CzrSK4zGlsvP@UIg z?9o}~Iid?T5*M>{6>}}=RV%CI{o0Sc{5Npx#qhO%fKBs6^LhgZkgJl;I6~w6Cy-Kp zk;&N5=j}s;8VPXd(VwoACL_J;)v+f0gBhLu{_KA4bF%$NeFc%+>wAqK3&HGV+{b#f8&@+# z0x5&;aA{5<`(A=c;jGYM2j z(;`{-VywTxSi8`PT}3;l2{)9| za0ZF44y+h4jcMRjwH3L>n%6(#=W76le{%bW{3L^eUVDO^0VmPDDKMCu#BxsffMjW# z>)1JML4jJziX(@o{oD&V5%ZqhRbqDri*H~_{s zF{j1}HT>IwRf7?wkg>;`84NmL7-MiDMm7BdjLX^Pv#xgk4m2Oj*=|yRWhx70Q_23? zL6R}P(HYHy9qun+NPlhwtJ(~*1V&Coe8k4CTfkT@acO>ewP`sST{$6!C5gWx2a>N9 z$mWtd>Sft!teFDI-33!?C);$r!M1tT;a>81;RXFTQb4F(Q7keDs{3w&*!9Nl5X|VQ z--Bp+t?;Fhs0udENc?U_2#7OQ<9eJ!p6n3bn_MPiDdiB*|Ew!=PM^M?(Ct8p6M1(o zBye8L)#Ms6kChU=8ZJ4N|DEVNQbefT4r*ZqIto+j{oIb;hy)&jQZ9^EpAI?($OfnB zR>|5Xn6%60Rlsj!IMV=_4eFB1tlFYk%A-XSHLux%%`n2t5s3l9tjs2yoZIj-&;acg zyW$&WJlg3zFtYClx(!&B#<;0#IBzY2$#+H7LR((GWx<^aGC#0mSU@6=?)cq5E$y zd5Ru3Zh%&#c`Z>nz$y!2QkcjGCJq1M?pp|737h!+v&YudYbsh-P@@b zaA`23D62^1A;1bZpF#E1OgN2*9dL-8+}v>>dk5;~Y6D>)wTC#_=^sBi+_7Ai%ZM`f zMhSzO2aT9=tIin5{-?hSWIIX7Xn~ncV6v^BJ^o7;Z}|vWoXFAu7-w#NyD<+IPO@l0R$aSyg^`%*42Mv9YQE%N1O30}=zOKU_u(E$P(L*K2A2R0 zW)wMQ;zU;jVHgEKz!Wd>3`1`nfq06=ozE1feJk-<|J@b-BdmP4l)(nPLq;hDS?&}^ zw*`}bPO0^LsB0`-62vKSEfFtmJMb}HLAv#her1>!Mop7qLTBF*9A{=OR-W5NV7*0* zZXdQ10rL?@w?AOS(=T(jT|oUU(0vfN86s*xyP6~%4Hcq|BQ5+LRm9vt0vl`Wx)ylb z@GXc&Y0dr^cLa?LdzlxDEyiLph(FflI~_)Fgzi__A|vb^CiQbmBlC1<a@RY9o5f z@ptQNfx3K!1Uejtm89?!onpb#l5JM+?iViNR|GD)8Fo6Fi$IQFtS3a~m0O#{SE35F zM4xeqz%p!gLanHS%#Yf6!}(m0!)%di5Z6mMvO#x1?}3`dnMMl(+*Ylj)S4&b`wYxW z)1=bCwtJ6LZ}h%u%-w?>x51U$!kJmhI&I_~0%{6Om7k1TpsQbrh3-Wat7J}E0-k&W zfGxG)6;}L)Q>DqdsLN?B*4;5wP%a|jR=IWWIwH;ya;=2np^#XqTF*U}-jCWWP7fee z%a4Z#9MX^9HJ%KlqbKqN;J$FS#XiX0Ie;^-9|N?t#4HvMk~DXkqv?4})~`e!djci& zTpD$}L-EiODt^j>tdJ21pNKV`w$+80XLKSNpnm(@{;txZ^@?hu`ml$7V=kMyr@=Y& z&@CzG-jD^bJ6c?TU@A9~);HF|!sC12?6o=D#FNvNj~l%6PovJYxOcpzY82Ub%kp%IGN{r;8(b2-lYjs|Zp}%6Kx*&x83^@eS_sn7YV~DEam*e*=@!fu6 z>McPqMcV+n)`w4<53&TTD!@%LtVxSelTVw;3z(oU1qg|op~s% z=96wNAT~^hZ{)p9a)3>l3~Ff?Xv6l#IG-u0hN|Ʃwhe7gTt*-I?M$TR`vf>`R# zM#9c#kw;srRlB?S>hUel_&`PjL*|4$Jk2*ChYh!ZPbRu=o5EK-``iTBhsW{*he`8X z9tOSK+De5Op~3VV+~0>fBmW>v*Pny5M25n=Ior9>2_x90vhQ zj|L>e)n@Q&rwfN^qgsGv_M=q^^flu95Kaz=vcuf_Qqu(WPiv#B&&h3%;=aUvdi3fj zJ1`1N? z`@`a7jHlrApwh8+fkkS&$K+^(Vs?0!K!Y~A3lxVl3X(ttUoSdY^pHWA{YZ)KGH&T-mp$drw9{dv{+6HG>>XBlj#@rQ{kb9et zTJP^(TJpZM0+hG?u8jYQY^>5U^5Z?+PRCmyn{*|&4*J&>AUu!~IVL*5Ivu;-gNhgt zZgH(lmooIkjy1MIQw8*g@FrC79f-N%p2`qkz0zcXKL8bL4@+(HY+sDd+gEbN%spTL zXhD}DB5#-}aX)t=VIhH~4YLJyaobl=k$Jm6XCW+gVxX;7Ft&Fqsf=Q#_;^)pbsT@W z?r~VUx2a@SKwH|O$MZc5b0GS7tAyyi8D!hJP3JlihJFfp?A?{ask*HlXhB`z-cu%8>6ntz z*Q9cYUj8CmS2B*-qM+wZ83TL%Rwp)OsbbtZ7N(MFlf1!x=o6Too$bZ)g@Way951hN zv4AGt*k>0$->|ehYH0}_xfTmAc(=9ieC0!8O4jzH!tT`ate}!$t0T z`N`#JMvy6OTTaM7+-QoSvr~r*EB?xP9yXd1+l~kfm)ISv?6ML8fX{LXJj|LhD5X6$ zce-Diad|6U|3!E03x`=cCo8L@V*4ffoR9!Xsgfl+JzBd*SUY$_0*!P5>SyWlN!fIw zkB(x3NiWj>b2i1fPIPS#HHo`8H{55^O=ZZkzo}5uM6{C!fKA9lL?BBiwQs?+Y62G$ z?XT@QD=|3L=u{X_*O_A0dlD#C>o*uOWn`Wc&*%FQJEE&hXr?xPI+s*Q8Dp5oiBQGa zypix3q-Ki9Cu%L5Fj?o(2lp+iSaf3Varx8kvy1Uu&!B*R^it@m+2jI0`K2Z_qv*PG z%^fb8M7U`>xb1@~KoIqh3xj;)O~q!5?wy#zdSx6rKIq~{pFCg}3@&{iiwE#%MmBp= z@f*&^%PkN#N!b{@0%0-$2dDg!L&QWH6<8^%3<_U?l*z8NT@A2Okzqdv?fe#>+d=0Z z1QM^SL572}7{|ZznTf0ERrDTBZp^A3GZKU7K@p)~%Dr*5hVL+_bm^q*=cxC{L9)L= zw~QZiT_LEMqA9m*r&twSPFOvcE*Poxw_3kO2lk%}OWh@9hM=ci!n-fU?xyA?7^G_< zP59W$PdB=xcZ|jErlqVc5q3={%~v=$h5H>ZM+#X#4v0ucs9->{`TCPAiYqotr0mNn zG;w8{L{tenYDoQWD_9-KlV_3@_bS=&@fB_D7e!YQ` z+$O+6jDkmJ54WnrF;8zk?|7tsuF1Meh^5ZfSzTTRUZ#R}%r3$(@_@*LeY(9nNN%h+n45w`?OtEZfg&dAVDbCIe<0^j8K& zOW&(ScJDo57Yj&LSUr<)(}{(h@o^NBFdo=yeqwalj*V`gxQ9o8x6@H&Z&+djQAmnM{@I+*BfFAibBb_f~t0vZS}XTYA9yE z76=yWi5^4)=;EoUKdWD>9J3NB)KJJQb)WxoyQ1pLgC?J?({ie>z%Mi)tJMF`XRzAV zYfkrIyWkXvNluDOXyaD669N-xVA7%gegezyuPojcp+)f(AF+tTcgFWm9~C+kIToYp zbR)LdSnIqRFf1uC#XU^l>A>*oywXaFP8(|-@PtpUAJMJ++o3FQJOndP;P=oN3n(f zPc2r!g{*_{Z5$+Z`M)VbusPM3>ceT>Qv7Xlc8?xgyEB-7z6)GK@3f6U z7<#sw*R9@QdsT!-T!mP2H;CS6o8hR7OwbB)w3A+RQ1(BH#s6EHbsyUB<%^tOlZCsx zJ9N8~&2rh{qN=G)>lZg|VG3>)*KvKa)wqG}tiQibxIKe*wU0?@@>pwJ32yDt8_No7 z?|8o^+vnK%y_PXMAmP3D(qXDxWps4Z&dzS=XcsB`yf0;9*Ll8;gNsYmseUUc>qTpj znCnspIC&SsmNoz?&sbG7q;}&#!JLGg#g&yV{u^Sd{hI6Xv$6XpnDn6T{Hjy_7I9?O zNRi`minw~2I{u4M3+bUTx=}3NQhu&hgeHPxf?rh=SBF&jMt=qpVCu}^uUZH}C0s0v z6na1&`a1wm2x!HaD12q;>%X1(Iow&c2bNb0OkU}P;EohkA2(T_alv$cBine6bmps9 zb<_s%{5E$k(xU0;wWUt}8@_%sy5T&1qZ;~7dhQOcrO8)8^-}Zq;cwSDT%m#{%V(9K z#;JBU(hgp#Zf6^1Ih5<-`$yZTH8y6d?NpDjcH`XJ420G0?ZP>oProjo@t#a#4uSqN z@g6Z7wS>Oh`J~z<&V7|>(QH?$#>PCm@^mA2tEO(W{0tA@U_;%wu8nkY)q*_bGwRrG zcda{DcpvC3=!kf0W!s?c!h&J)ZucQvJ06qbvC>^T*j5m&jx@)(ZDQB<-Nvbwo?Jh( zgU-K<&!-&Y!)U&0r2gK}9t6Y4qLn9#2i?2Sq7~33c$P;zP4L|K7k?=SWJsN&LGmG?Xn>+Td z&dKR_7JqD%{Qw6K=QG|r^DHuE2@Rw4defqnosn8=Gw*}bU|QC)IQOY%Nr9QShloIE z(5;$n-;!`@XGfUNL#t(C@({!};CxfyPZCf!8z|lv$ZYr0_5++E^i*fEB8H-c^K$&tj6ZvH@myB+*UiGwyK)2*^p zz3$=Cx8~~=E>;tjg|B`|(|(k|MO1z-)Lh$d$q*aqmukS>9IN~ei67@0As+;WFMhmG zay0VSCF`03D-U_aKO52ic97SX#|qTa+V=KvXpXI<$Gf&E*um@dGYZf1D;<&-ME4$B zr&cQ41T>u|X5oN*>I;VWUX|u$>zcCirFaMEVqwEkE5poivH6*1nL2G@nFdvU$fe5K zt2kC*VFcbJ>3m)@GUyVXXR2td(9iI&(1!)nEJ~g|u2DB*5u8n2yPF;N#x#-lWZ`OW zs+c*$dhQ^X6KdF^YHd9I%&9nnDz+FjnFE&Fpz=c#6An&NzzU)7Zy7e0u~&N@S#zI_ zXO9}>HqMMAu%#DQqj0 z+4Qz`mieEjl-*B0&b-cyNJ47uC8ftA7jW%-k+^{~-U=D}A{}Dx5vDoZaY@T+klQ-y z<`Fus&(#NT(Cw(PYBX!`P<%NKF+H2YeKrLQlR*)B+}vcuI;IRd z+W4sOyh<@+|BN#!C`G3~8X5P&dI_qyu%m5O?+YKG9h$95K=wPKMrT5=S=Vs)ir@Q; znaa*gf?ESa==T5UEagh8R(}td_FhhLENCZlZa+1V47x!krXgxjq(k1}Msc?;go8r_ zmb)W!&wM%czB+7QLCIgFfiCs(J&q{Mgys)xSfMFEjXvcH>dLuIykW@NhO&buk5iEr zEZ9Hxac~M-D%*G`fz=W%U1p;E3bJh}>b?WdI9Qv}G|)pao>adt7SGcK)InrfCe}uB z;LY+cG+!)Xu`8K$%hXD2+$X;B4gH z_-cIB-pa;C9Fi}2-~NkCu`k-Ae0uT()R!pJg`xKX{L3B{SnYW%4ER;lA4*g>9M>)M z#2~W>7Ct#ghpqKYm%OY#t}UriD{gd^_4l(+hu|R>T|NPWAFOH8X(Oc_&1^c6iuqR! ze`K`zX9k-&%Ec6q*q=GP+Q{Vm(W9}AQ2>B0YRNQ&O-41Hj?W&sK9_J?x1D$I%bhQu zmrVtRRBZ0c_6?k)wy)&f=uV+LC8+xF{k-b6t z7(u0?6^bt5)1P5~HH9FM*$MX>j?~MyTZ*=!HMH5%UUF{hPMs5(&(R|BXQm8<|EOb% zz)T>H(TsCf&2RF+d`SLoxZWU$L@i?M+aznU&DdMw-}k;5Hc?NI2kxcF7rnnZJ=+vd zHy~~Dqi3Wyo%BQEyH4qfrgO(c^m84S&sB_STzadO!CEN8 zHj@h)9qnv~WoEFeI~;Mv-9`Mdubtg7Vxzi;_Ec?AW<2TrKTmS*3J<9wP*?mrZgJ5a zNtl+%^EgQsZx;0mlx{Cb&ZQvY@Z>|!7&1@y_vQ`X3lhx^AEvk+RG;Py9@4jH=1}bU z%pm4m>rzFgW@%30f_Fa~w{W#23CFEd#-!q{qPv!SkB-N#*f|o~n>rUQr3mPk3BE*& z(4Cn+zx59?GOvW7V>ydr)feCO0OZkl#_;RrIu8&rZc+;1pOl8)?LezX83~yuczg*v zVWJ93`mQeoxH-edHSBKPIzu2fdgf2XkK;mbBOnHC@5_b{6nXA z&ePW_$vJM>%Gj%>p3{|{derTB!6IhHSD^jNsJ#~Y^%9Ahsg*R|eTekzPK}S!2O__d z0FbsKE1D^+jGPrcsTVZK_NWIT>|pxLUjhyDN-sH;D!Vhp*dj5ZsNHgaI96|(g32aW z(hq!Os_K)rPO;ITDXdTFp*qFIVF6#zKzX@k@ErYGmo}$ocQJp^TW)QFNSMpzRWS&M z$#Q23l}8-|D;;8tlb`KV9yY3H{=9AF#n!>TNr?Qm2_UH!g5Ek>tN9A{Lh}BK!TE89 z-_A1XlUKb4+vB|#Ut}ySKe#Tsu{a7*<3yc-b6{2=^LwZY(uvQOLA^L!6SQBn$6v1$ zoo6=K>O;re5mC^R`9openp`bzk&drrkzU&y=y%e-+lIbJ3nf!}Gxv^6ikg~qcvVUt z%Ug!J2;Ym!2|1HgyS`xmMkaFb365+rcns8vG6=*QNnQ#rkHv){GmiFJ3%{j-tykd9 z!SQ`hKA~sQiDi-vNR#PF=*ZI8Gn@HZXb%u;zhoPwJ?6Fmm76D(9LWL^>=a$WIM`1P zaIL~{Cx-WM6WyqDt{*mbicvgog_(BinHjrK=(@4;h zG@^4Js&trC6GpiP?*CE@H1TNSnH<#Q1hvM3MZQASmPMOasV1coGsLuke2>xTdOp6H zsCn;@|L2L$OCe8>@8%op1!LKT!8uCc(6!JYLn?#hy_P2Ew9YV3iG3NQ~c%gwJ>OhxQVuO-L(zsY(q1 zf{+3NH!9&q_Z`cT7fG3ls<&&+%TUVYpB{SOtdzLytfrZ91(v0$vn4OBkQ>o?Dj_4ExhQ)dn{i~rnQ--gEf>J};=v%4j9Yim8I3{|pbI1}SEY0Y%e z72-6(b!1~5#bLYzHl0Ak(~l+8E!5s}iT0Ha-Hy45W;%YJZq=v$4s|mza9!B0+2kQ= znDw$!*sWvh1=4TLs3Dcu260=L{$b`k8smy(*;G#ZPvyQ&;nbgCwe_=${CXoK?+A$S z{`}Dzd~F(qWFo!N%+H9vESd+3AebdAy805y*ijI;qr)ocJ1(nn#TMs-W;{FCo`*{6 z5{X1v=Fa<_Q#0~WG5nKHw419yqTa^Vo(VmRjHR-nNiWLWSD)&>4APCbz0?1|^7PcY$zaL0pd- z^TVmPCHDrInfVx*x%kM#Oyr*GRR#a)QXgi|vrE?7(~7rStqXpp;io^23&EY-UU<#{ zsy+eZJ#?gi-M^yj4WLFDr)4g*_Yf6MXB!JQWUgyGdL_D!7A)O2s}QD-SyLXz<55Z+ zeCAZ7q)=l{x3SehyAdBPVXmkiaHB={8&5&rRKlL-^oTWntAqs6swO;u*|Lg-17^)J zOGnr|$VhwI;L0D7+U?8UEgQ=JI2iZ}j$-8o)V__<&(dT&n9>L0NKzOf zp0x{dv!hn2<@4-c>iHh`%kT04KVo2J3DQx@ktjDq8=SyDa8;1rIB6*EACUB0{UmZs&IQ;MO<-98hU;SvKh24V@tGxCMIFZjqXH%waBL>Sr0%mJ!bF zZlNF>HdWb_*XJ4*Zd%;)^WGq%si-laxYlG5T6^q~KwoyQ-~l;XU!C#SrsMQcNQrRn zjr^CIX7hK1g#w`|j(1eVj1sn8i6W_&Yf`)1pXb)pRKOY_e0(b?xUX6HG1QJfJ@W-Z zgisz75(TfeZ);Af9k$9-VHbL$6niIAl<4h$wXp8*qf>tP9v}MZV9o0hcqtd zGURLFCk;VMEH%3BK&BF840=8;QWV zSFP=K#|IU>EqRJXIlfla$m=E11}rTIf9M-qYSb{5X^hKp!+A%qtU$bjQhRdd9V66( z34NAZgac;%9z?^z{R~9Isr|e3jl}7ZGs)@cg;P`66X$iPJ%|EqT$;(TD$>ORe^1GL z!?SW{U;HEmem2usoWCQLDqUsoEe@ZB?V3+bjC*`XsgFY)>fpJAN|pD-P_j1X!Uuzqejmx{t_Sr(8|vk^zr5iW_qi%0(?u>QJbF(-0wLY9L_2 z8WBn4qQC&TC^Uk%1Um(lo2!6fQGyqgX@ZQK3=$J6>k5B^ z?+%Ro6hoRfbNpS0AH?#gD+$|wC_(J9$`peGF0omkAE^g9uPXjpx+dsM^NA#Ud zZwg}zh4Rnc@7X4rSUbTTj&9lNf7)L_Id}e{okGd5PtDWBhSeFziqA4k&8ZWNp00z_ zc;TwLAI@|ROq%}|zspEmVPYqT=j@3Jk2R;E85Pc~iD*N$Y0`{68@pm!n`HUP@q^W} z44$*}w6NR;W9DJ;j!c`uYG_z{UlDC>7f$=k-=Wn~7GmSSN2Q#PRK!udKp zMObS?ZZe&5adU*CQ-#H1$NSrck1}p|3LAVi>NMZ|A=KW!W3zV~Urwneg(ey`sp)DX z4^vFPE^gzjp6tnth@zZg%04ZPKSq{H4ct|m#VtkGj8`vyghpH(YmN7M^mJLkHE^Ga z3hM>8tF{ky{3?5?njTSyg9oi_4X0|DbZ<2)#<`es_&SlEq#oW^sdme|SI89)a=h10 zk7Q)gdyQqg5_LT}!L0>1t^~AbXm|TERkSBD(ne-~Os4+&=_wiYNuRLct0x!VA)S|> z@Jko%xytT#BV>5)L&K<#DYdjw6YcEae*&f%?(CQKx(55OPqdC3GhNq;$t$%Eck=Sm zL&HNADjlz_k9y}yh304bHJ6yhRrEC5gPi$BYYk0QXAmDxzbF$TA zH-vv)iSHs8!=5!9?Z~*$T>`nW3DE%>F(sw5q)tDn?p`n1b1l!UIUow`6g?ds9qMp6 zRdmJMBWyw+q&LtstK1M^Q=39=_tmo(C$yfD;anzE@LRlvZ`R5=VIq;pTUktWA_QGn z(+Qx~GV|;SSyFW(xGEyrpp#>AR(wF0^K-2ERj1(4*bjb6rBX0(kb=Q2d&5{QqDg~y zx<7g)4oSVWl&(JAJkvHGsN{rIqwiv5`+xnhtpUBI|Eb~cb@clA5@3HpC{{BF!@S0x)aedz zd_es#e|LVqf}`av)Br^mB~e)sAh%JSp+R)Z7_YoHKe0ztR?`Am-YA83FM0tm{@*8r z74d(Rjf;*0nC0C+k7o)}gYo)-rCEQ^bLZ^_^124ZKH85t82@bT@nI|yu@uSk^QU0) zCjsHNKqo>k4J#E15?Eo?g$AGTGVO5}_%i_FgUG!B)S4qf29mk^1cLN|8Fd~q2!2jLg42a#V{>3iBX8DgIJ(C%=MuqNaCEpVbK-l~Y9~&(%}v19 zLFCA%%?P9R>qdH!d9VT6II$aCuO;_$!^0vDm1K-QUbVm>Zs5IG0ydZ!5+|dopQcZl zKfbicQ9lcKqPU$l>1h}V<9BIkQ&_M}`anKakP^#)3y6jp%2=Ot!Q8?^nZa2P78n?( zn^g3DfNyyS;yA!Wc`3LuFfb5EPab@6jJ-D{0CAkl`O^<#H#iU8o#+ZevIYD(hqaa6 zH<2p=gWb98gZGp#XZE#1%M1=C?920z z)U-O->Zz@T_UPhiI#!f(&DOuE6-I-iAzA7HdN7>ybHQP3rlz?2dCzj*m>IZTkf^G+ zsKRS0#2=&i2rMs-XF@-Y*md0t)wd(PR^U3l->>ed3= zN~iP6VqT<(77AO-J1ssv0g{UhhjIl zI2vXpd~xZz7|KJNcqh(k|7=hPfXe6jF@>JVq)vvzhO@3J_7oTF46`t0SxBZR`^Vms z+q!$8j&iP! zN0U&badc-6#wH1zV6*V^^#;&CD!szUi=@rO8CtXEAWu|M15T_%Ft+bhx<6pCWbl6a z8XqK_N`|Q%e<>{|H#c`Ey$O=WR))b9#sDbuhsI0W*y_GU)h*WTE&x++kW-tSr6L08 zq#PbGeHIZEqUi+=MhOyABE>1J-?Z9`MV+)4c~Q@@mq&Jt^s;owt^jL+iCW)iF{n_z za+cWcB$HWi%lPxUYH)!Sp%Qr=8rhiwE{%eM0(dgH^XF=Bgp@O(uJzzm3}2BoSCgT6 zauX>UA`gF+6Qt;{%oj0$e#;>Zx@$#_nrJ~D5Ptkw*H|4}W)c^nh literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_27_0.png index bef273ce38fb151ae1508beb546d901e284b9acb..889056124a7c8231a327478a5506b899d1d637d2 100644 GIT binary patch literal 14221 zcmZvD1yCGOvn~=KxVr@i5Foe(CrAiR(BQJcT^E-SEV#S7yW1jJ2pS-`2U*-*-|*MH z_q|*73Y%iiIeVo0>#uvh2~$y$!9*iNgM)*^l>IEF3I_)d2Ci>jBLlyQzZA&;e*|2l zzq+W|o4J6EolM~rja?jU>|JauP2RYfIyqa~+i|l2U+>>oxVSht3$nA@{!arodna@D z!Q|!)U=UP?&)Uv#aOn6iKk#sA86ZUW0(^XmJH>j++V0fgr zp}2&Jv%$6+M1%%j+HTX0%^EhoBMjD^}02Fq*Zy$=O+nBMUIJ3JMX(bTw#K0S81zV~0|dAR>)El3oe8?RUn+8?HFYMxem+7|c+B@DXj`s# zCITxeDvXFFnF@PUCL=>3no0yG)`*zVgaq{N-Mcr8j3HSfUdZhq zmw8BiRI`8cKYiHT4+sJusAdt5Q`a$$U_HCGVH9jPQz zfz}C&189NDvz2 zdwa6kewE?)_-fJKe)1|TEUdzyvu&)dP(B$!Ch)#-)?jhd>0ROXoJ*r}s9l-3NyV)1 zjK%0ON&V{&D)y(;0jSUvr+Yq+t1Y?!OW+bT;%0=-@i$}(Gv)RihEGne%lOXx+c$?9 zZSz{A9t5_|i^+v8@UknhT9HDj&*OFXb0)sBaBQw*NS*005gas{db~djAAYXR6djiB zCrV06ib2Ztqf)oIO`|A@RTqK7Zs`LNyTQm>PhcvKWn6P{S;AQ}x4A_76b-|?r?7L; zD)z3W+F%z%oA&Iz3bOFGh+EVhjQ8L;C3nIQw2j_fgwE3BMgh{-?;i ziVBSMQMVCMIeB?fVRsHV5wELnGZos6`wR*x5MT+>^VO=%QuIwBxZwz%pOpi>F#aHa z!2E!eLPOh`R)+{0`DVl=Zf0gCrMT%i$~*g4bj9!a(RGGW&?R)a9n5X@_Z?(UWiz6_ zAwpHIb$Dtm3(?!8V!qLj0GIp^)*Pmb>s(d1PpTGKneqLwKRn{iPJVPoRd`rfsoG-I zs@YO)X)d3Tx*Sqqu_BEa(il8WY?nAi4Rs2p9xLrB?IivV;=@j$6Y^3kEKA&%=$~;%ozReJ3X_{Wh2NDP=LF zoQ{r+xVSimL{)pqQZe|m(e2nXUE zoHybUX%wp|q|9+VjW@j288!*{12P5)BtKeIE(D8mXQ7b=1dT9Wg9Jp7atFsN36_p| zp`)Xd@>t-ocV5Izk!JxX!p7eHz!iv$X);%1e6ZMDEo({++@F}3`1bAF4yHz7f|`k` z=64ou?y0}o6`&1hWQId8>PO};n^WUjr6-KEED@ZkV>6c6K+FFPPzC=#SLDmHj#(KbwRLobHRkvv~H-xH8-%vwCS|I{Vy&A79`g^6*OS{>Q^7Ql+ zaLqc0b&2S!#MD$-v!bc(D!Ed#&(=qZ4(Fr%&i9AS&U>ZBXD5Bwv!qaBBVe<>nVTb? ztau%By6mVRV^fXTHZOm2bzP2Xqo=3uIE9iwoM%736&0oOzdBsxHX9*X+d1iiqxQML zdEfy8V^NmdWYEpcB-(;bDSX7kR!_BEu^L&dV-BhtM?Ebi$k&iD)=%^*|detwR z+`gmMEge|+pGsYRRaMnAy>9_n(UT`1Yj9zLT28P-49#rjeS^wycFM6%REq{ z*Qg{uMN=udxVQu^j->NP6S8PJ3^5hM&CDI_?(T}R^rdiuLVy0u15V*$Z#rBtoo`xO z41Bb-FrN>n(`MZNuBw6?2ALyvm8j))ZJh)k$2Tvp6eIM zx#TjXD15!-%y#X(T^qg)7UynMkR~BaY zZ*Ol$GAtFlr@MT-`0SQKN6pH)(eWW8sob^mHa5#`i9kS_s5a~t+?R!s%}IIo%F18LDI5Xv?41e5eZdp8paH?-{h3N~zl-s_X zH54=BD`w1c|K8gxfs94*`O_!lbJ+Ox6k|?;RoqNqC2C#IF}~@3O!sJlZt3|EN1Fcqq+@sce?a)$dODrqVrP$5&C3(^IE z57drWQFOvS%P<~m$T4lnP%Wzv#>PVIKqG0M<${Rq6b7)lWP!zjO4>MjT5VvXvj_kp zk$|pRMh4=Vnw;#IPgIpF^%UOi|4Y7uT2Twm+C``!h|PuA z<&^g|=^qRysLCO!#9JRhQ?esDccCd%!1veEWEyCS55#8elUI4_lz$*J1w`&161KBr z@#1&@91ji-t_2YrluXf0rvfG}db1V?m#D&2nBiQ^v>5{7%E=Dh3>}<9=6gn)L$-lm zfjc^pWeR-D$r(Fcu2t{*=(#d(cQ2H03?a3sVem5``?`RKRGEKp6A$^BJx)+AnZL zBN<%dva1RQjU*rV_bdD*ip4W1I<;o1GWY(CAXjtU*=gbYNI-=c(n<%HUwH}m*501g z+1c4=Wo7fR41u?A-`04Xn^m=LlhxJLrEnM{;4>;!P?~oZ6!h=90aOnk;s{#873dN1 z&#w&Nm0r;YFaO$`E(cIz8(7+`F4}-VrWzpgNR6G zakg(709Ey7qhvOGi=2Vr=@cO1>n;Mka4D5LdX(zbm(5pTF+<+-^YhlntHsv+s&*V) zEGps8$!z)nAeij^Ep43Ip=V@t-uaWeaNW9~#6)Op1&u_-jpg1vPUX&B-o=Tx5xH9{ z)7H_^StK}nI3I_7M`v3rEzUw%ixN6A0!9GBNsZ-H(SocnGA5aniwh4xNfgp}es8$} z=BF&UY0q7lU9Gx4UQXi&SW*x*=FtmTl@M$Rkr;Z1w+j%+P4xM-te6<_j~@e&X{ouP zDk>^Did>059cUH_>g1EzHa9nOK%Msn5)ueNTt#)`OhEI2-?=?vry*Ygfk3R=>$98>`IGu=h)4oLi8BK`ynh7F(nKTNI zOa?-R??#A52>txRBK;pls>41vjLVfLs1C6njr2FXkiU7q$kNtfh*L#a4g4pBsR+(3 zXv*Q&sYA?5V-H{}T6>5Eu2-G)_e0x9@R=R5~yP90-plTYr zb8ue-!G0}Iv*rHdNd6idEipVcVlp{MjzV)3xr!*(9d{S?0^awcD+M?X6D!YZO-jK9 zuSO=I^?%rjIpU$zTNHP+a#sam4jC!9Byx})KeU24mK1_q;WC;DCCStR%*BIjzmsfp zr7{v33>x88BI=%oT#X})d#b%FOhEI999=wsd1R93i>L$Z!H>K&z;rsEiJ>G^Wh#`= z0d;BxF12e||HmNy-|XdcV7scE^c%oZN>9|fug%xna;}gurC+k7^>RTFY+*K?d+2o^W`>HzhqZI*>{~{ zGT!zgg26n7uD94{J5gDaRAs=&Y*bvsLZvm=^tl$wYMZGJ(%GJQZ++BUd8YA^&&#ZU zzS3BJF8|fzg>v#1`1*=L@`XI*UM9=?cJSr$B@vsBc@urKQ~JpoIJ>6M&9Fhi5K~*J zZ&9H(O=3+=?8LYp+EPuSgF3jR|HgY+0P`HnY#1R}%kvzD!kqUkN%F$;5wG*IN9_vj@(oKS57gap$Kq@I=EsmB;0MNr2Tn96pwo zHJx6%UOUYbAJ>>|x?-fC;oUIR1QwYBC~?eOZtE?4#mScbgw~zqCqSCru`vt5da?1L zYUVPV4lh{!!|~tlSRd%^#p51bb(&b4{EMmG%6e$<##6h;k~2n|kH<{ya$Y-!Mq&=u zzIN`NLfI;?$0QjYM84te$vE0RRfVa>v}U3k?&@t+pIiTn2NZV+O1^`LYc ztL6{G*SM=~gsjBu{#cY93&q7RPTU*cE(ym&0P*?obXSE)FMM$kqTqh$8>gxhQ{ztK za(s09)KLh)Fk#jbB|=Uv+)@6=fu>E$e*xVJ=9AsjZwsXnPQ!`Bk5{M;hqeA|K+w?3 zfE+t>ZObSgC)L#Fd%Avk8YURE=rGCEF2$ATY7fFC=}li;ZFbsysZB{-S)AA`+Oxn= z8{F7kT0B^JFLR+fD>!7<_MN|%z|I`@BAWY-efHTgQ}iFc8656sp%3e?qLl__g#i{c zN!o%k$OFZAl^0MR&)6pw3h%JgEX_-+Qxv=~HjV9<6Yf!E`Jl(WZ_k zVo!xZu@i!S648tp@(V7ERu%OvyjOs8fc4>iF~UZGpN9uUp}OS8 z|Vpf zNE&jyl07LB<2_kF41*Tm$o#lNkJoO5VcDpd&FSR zdP5*0HMzE9bGhv)Uo3YdJ$pa!a8{~TLVTATd)U%+8$!xdhrDP6CGv~6?T0`k-N~Fb z&O~a(A|Dk7uxh@^&b@N`13<)C5RP@-P9!+_*y!MIc8l+|a>vcnGwyUX`+J>N%Wbz; z;=WyPUU0^P{YTn+&mlAJhs*Lv;oDt7s`2Monc@f;B>^O&_x@Lk2OqYu#0+sd4|B^B zR5kH`Jq}kusQMYg761LYZ+uT5k8+l^UXB^mk=C8E+VRLw!iiuu`kjdN!Py(=)wm5z zC97GHC@{9Zc{T%m=!doM1Eyg$`&#^xJ+P)7;e+6M}hHU=E*?^t6Z;yZmMo(#8Lhu^8#F($$ zbmCI`7Jox2o5{SNpUmD~-^tC!NpiP;=xX%wG2cU4(ib@-L%0x94(Q2UZ{>sWp~kmC zyY11o_`kd%>p|x0spj7JbCNlhB`@*%1j!;Lpl-;DDx)?S)zJQ5+gPtx_CtCLB_tyC z1On<3QhCjTISqu_GB(4A+!b|r60#&3B1> zeQBQ7u3A=wQ{sBJ&}O@Md)TjSGc-SdVX%IOQvK0}uiQDV9&%)>L;vbi{a@f1%Z4>7 zxnG?PX5Dy?Qhn-fC9~NYS#}nZWb{R}Ei=A#nw7lr^a?-^n0U_%@JXO`NP!WL)`f&d zV`R5aeHcHHJuIQ0RpEA5R>xZ#ixB}^idQ(pzu7YJO|%*6%Z5+G%4T-$An0zG-8{+P z+1WRPVGL)`e}@s23B3)+SG^Jmj)R04;PymrF$JJbR@ z-|tz!E0J5@(j0fbt!FXrqq*@>{vbxZknz_&t$&;{0->iIV0M-M`~d;}@V+Py2nr~k zaPtrnkU}p1z!$k7I4$^U-~F^+jgQt$Ty)mN%HQ-Qbrf(7-!92#6FcX;P7%PbY6JKV zH6el}p}mK$vmfdWl*BnPv>#p;0)hyGMqc5PqC_B4ot}M=D>ZT^JfX1OyOrOsLRXQ? z&l`efcYhuC8U{Ok+862QI>wDN6hlCT^hXpSUAo@8~XHKlJRM~N4|bGLo> zu&&!pC8qA7=6sC0iTI2gi(!Q1!2aij>U52YLdH9(6I9_h-e(`Bgo_2ATBDTQc5j|S zE~eEr$)8+*q3S(M#-H?{IE?b1%;?6X>@p9S2F) zy4w8LUKNZj_M`iWaK*1%8^!H}tpx@r08ODjs21x*yK1GK#DzSizz#;CDZim93LphD z9i6oQ8b8w|gBQ2r_pfNC;9+0aWsdOJCI9H2UHLm&Z)E9M1fO^qYV zko&77u0?;{nkRCg9t#SfgpjaD=O&yX4S2Wa7d2R%HiQ{jvaPA4+nKPWS*5RMui9=G zdZh1W&cs>HUTw7uQ2Sd>tu=QCF{!=bZQR{L`(cF9ADr9+l*P#gUF!CP?XSuc&H@E1 zb>Q;e6iFKi?!G6*)R2kl;)7aYirWa2;iLrI{|Gwfrj>6Bc?C)jurr)iMQ^ z6&OMtT!M3CLtUVxzaXhmx}E2XJ>^z>&p-HA8Bu^Qwt%b*i=XVEOTRl>M$|xFIx(`_ zP#|=o-B%)b8RPWyWc9h%!&oI6WAdb%#=BKbj#pcfRTV)*HFpD3UG31DoVcf|Aunyf z{BgwgxGG0}U#jb7!!qLbMq_h`W&LKu_!rtUlAB?4Bz}RTRl|0b`diC)NH{xemsOME zp^w2;=bTo2N37df*jlv$#~jUA$G;>Ib@0HW_QvZHEr)#WyY%JTN2uVlRZO4Ytj+O2 z9;L{crNp0aX*dv7D(`q_8o0@9*eoNAhprb%Z}+YV5;f0`=5mG2nsC8qvN4HBYQ??# z)m~I{J)PSrFGKLaTx^&HO}>cR6@7mh=>1XD63Oc-s|uy3z(FzfF*@q%t#(C^3@>lN zaYJ8W+iVXs2E{stzbT@WcbDf~3P*YmWOo>nyZ`2218^^m=T`tHY&!%qBiq@}zCODt z-v6Ly@atyy&ROl=YO~k5a^X%N+;cuIDe--_7`_F3YQ4jaQ5bGc*H1?CriF`}TTxR} z^Ccg5G8=Q~Zm1LZs`LA1em}L*il5Oc$3VO$N-N95*B1lkts^ zJ(PQaSF=#u{nP{^pdPd2WpcbID-rwu`ii{i2{j&8&G%c@E2%AUsYfe>f*lRR5lob+ zOs4-*`w$Slcc7Z<6p1$b1d!^{0RBdM>c(ip+WeT+j*cJTq(h^FRFT`WD>c%KJE;%n z3MFOVZ_g`nDM26~XR4}&aulzqS-1%xLIGX*5jth16Sh_BPIMUTK4d*_>MIogbr-$q zE5%?L4_r6z9ZGZx7uHj6<`vGKxvkLa z&0djrk@5r>XLqAW21CX&IjoZsqZ}D)B&WvuG+4zV(zrb(fsC0K1-=XStDfd>!|tWQ z294Q2bavQNULw7vTl>a(JQep?Z0#5;?uZNgwS}^3|Ep(EJI*l8_;So}C^7L%OlWq! z53s*m8ZH&92$AuxgrwW|nrCYVEN7d7HHNV#{Wt@P4wn+qz`j-?E9yug!yb3%m?H-D zAsa$(&zo6LV!t%tR&|riQ2mJ4>A7F+MU|ASI#`!mY-J(*{XIjnX&!^S&h2z*HvL^` z$-k%b`?CS=cKqw1;gikynXer7ouxPQn!`8Kx}>9OuFGE5h@e3fGFcFN1%wY~8P#5qIO;DaQo+$V81cKf53D?)+ui z9HQFwj&<%Pp4&OOvxmF|jW~+aaLk|#sK~YA{h)A?l_nhbX^ee2=fnGG4o4T{DSN>% z(Q3sX@ybm)y(1~cjlqtDbAjqt+r8y!?V&F4zKFu;!f$91GSKy`SHCfHP~9ZIgje1- z5LV-Dn?ciugUbUp=B+)owzEO$fho^<+QAll-$z2bmL9UvTH8e^nCR2;tv*`qshO2~ zdeSY*2}g{L-!joVt|GS|PZT-z_542vD0fc#(c`rGO6H|AAouton|Oxx-N zr?u$P)3g#Gu_=|W_b`1Tl4CAwt#u~4F7EieH)K-Ysr!UyEt~?&ZrT!FEtx6z;A%y$#?qAn8g>==w*~=rhp{KvA=U(Zs*j^IyFt& z_G^A16GxLVXsDTw1Bb%87Wz9BJzNwN*McTUBlhOMmk=}P#G9#W_WFP!CmJfXY!;=o zTa*B>2hm~M2_6Avy49L~Yhj8NRy!f={YA_dtm=jL^vIsV0gr~hzF@sR?$;#vwzQ_G zN59nJ?~n8-kB}jE+b7MJ9fjl1n8|D(-&0ylrG~gF_hpGV{H8xSsOOlw^i_9%sh^15 z?r-Oiv)cTWW4^i0vsJhiP5)^{p09k=y9~Q1I$vU6|NEuJgByb!ncDYtUx=3>1EDIA z<8LWH?fEP~6bJgYG&(xJ3E972fUf^!W+VR_c0Vo@m4)6NP+uLA?H3$?XnbYUaxn=S z_^O5b5kT%I0tl8s$PlxK`8?K&UPokaoZP1i`|&{7<@U|5)n0+wzFkySJHdn)pPGmR1-UudLiB|2zeP8^jj%>P6SxXPa1(Q z`(Pafe-1&j<|yU-rRZlL>Djo`A!*vt>MjHn3_e<#@+M;(42l#NOdX%^a(Lq%tk^bs z-frSsrwN&V>SiIeh zigJLl1Bl?4v?K_UbRqcBArFMHRS+R@F9=+GBqw{=XL2LKKZt|KP!bRt`9d<*ZMlT1 z;>pl6RgL!WTWH(Iy^RICt0y`I{iIPDZ2=1AYTA5xW^9*=wTfpY#=Xw#oPz!|_}!}Z ziX?VN$FX6u!Hc1cbyw?2-y1#kz6ah-d%}|(PP4+$w}>oK->@9GImSu2KT1L2H)_nx zRhc{f{)Ig9a0ZQcZs6}`oxVN~MU?qja_ztH`x}UGZ?g5Z zs)6Z7!^H=K(ACdS7&ZQ7pHpaL(3kuQf8P}k%x2|kc9HsSJzKhcaHT**Uh@&`zrrM#@BxNGi%Cn2e3i z_RZM?9Ji6z!f!{dRgtZn&TiJT-uI`woA8f)OV2rZ0WML>cUp{Nnj70R5!^5&QREeV zsJDq>1`91^B&U~j0bLAilr!Nh6>8OdG;^?zGTE!)s!H30Xno_T3(tKeJeJ(*^N#;i z&({K;-Mm}2yYW73Xda1Gw>O1TQ$_5M=UB$WWz|Pnv4);AS^}PJh;}4gH~S zI^(ZTtp95KnUbxWA}((4Xm+;IzH*db@yD{{NfFa#D#T>X+-foGltgpI2l)@?T);HJ znYiTrWJAPRS~(G`BC4Hpu;i0!&VnapJVNrL=Ybc+q!3EE&dF`=`E9`so+j0SLWtz; zL#+d5vdvR~8%q)$6jEB;dpWDidwFpY9DO|8;9ve|`3KB~A(k9G$z=dr@&VBIEVO*x z${K-`hxKdHb}!5)cPJ|ae415fKW5rg%!MS}z2}Eh|6c6kW}+TUJ<=x=Jgzo~@cwNf__rY1FbZkiGp%KfeWv zG}|^e_?Gu>oKA}pRLc*1`rrk7mP;$+ON!1jIhGdw#Of)8sw-mSR@1EvWH6zxl~)v- zTh|I_3@`6j!B_pv|Grw?vu2O$^(|$Cr@+HHY%8_j2kPoSug2Tjt|F2kVju}7NsOG= zcwt>!tQp?jUBBWothbu)I$ELr&zMD~>Nk!GEPtIys`#c2tl4yCcYY9GCK4SRPY^3} zF&1NrZC{$7RFk(Fc@e?v3_niBnkd-3ceApb&?VNPl0O!lw_t93Y_7TYnc2WQIP-X9T%aq%hWka69iwOpuf zen)>JFz*T4P09aQ*Va$Qi|3uZcOre)ukmJ$=u#v-ZcjMXdhIi~eQqsmsq;h|cv;4!^B!aPmem>6JcX$2rh^g@V zcq(1iN4$%lxt2E%SP>eVvg)hI&H>rc$Ilw?SIv{_>~xix`TPH{d!ZB`>B$WAA;GD; z>?<8e6VnnKyoO~UR_?o&*&TGl`xOX!*D4z}L&DSp6f|y6zAg}`22}~(o_kJpn06s+ zgLxk1Rno$3Pc8ViDUAB66tL{A6z`|SoQA3}3n-cD_++C7GW?ii>Ba)l+5A^J$ydx< zkY#PJcCy5Ml|+S;wwE+cTFMGP*q*BTG(_G$nR37)@@;3&iTf(BW-urDQaK*#1p`({ z`l1ZoY7U2Q6Ih?J_{6A-{Knjn)hkePS~SqM5c5jONTe0~Xiodp*1ktCiD4Tyahx`6 zYo?)xf0F1?o^e1@5)A>haJM9{7-8C+|AC{Fw>}!;*F(yp&yPO56!_=G;-KYWaL&tQ zUc9&qKT{wSJ! ztw>{>!j-MSBQBfs@f3nR1&c#jm6-(}shK6G0)%JlI`f`CqeT7k;C^Ne8@Vc ziR(9B|C*Oguv32PYmQWjO+lMv)fOjX5I$pb{3pzC>`?h%D`;owNUe?DM@EPVVIG;O zT6~!ED7n4Vki+mn{?PL0z`VJ3U+sHEk>=nbe6&higfGG3eAt+OYJu<%Tf)TQH54|t zT7i8qK>v!*vyISRnSGWa#f-ffThqzNnMn9LMh+FJQn$9wN9+oFUs++6uHm>T2w%u{ zj!7E<-b}kwCc%}x=BS@x-ZMjMU%+IF#bHn zHRSOIjDdrPXsZ-l(gF(XtXB_^k{rU>xx(8d=z(g~Y--pnUaPjU2hVjd=h@7xW@6GX z@@9@6w$&EuYyA{zWT;X@-Cai8Sv3a$bJZ(==UT<(v^k~SlSyEZi@BSa~X z)LxsOkQ-BEQSrXZW*)!7+Zz`TjYQn5zB+|!OyBG%< zLy)o3h6cj9DdAN$p}2+(;x^wJ$Bv!97(xyXf&1a=qZYAMljKH)p&8R=KBWDR9Mr(Pa(aDz51B1JMG zpDnKABn(_NSd%L67ktoR{53iBl#)gwuh5WQKQjBgW!8l(_O#O0gM^c5w7lkI1H*K+ z|K2#{UozK5M2>C{59NjWO|9!=%41zdV^&91c0@DpWwV1yc>H8LQE_@U{}t`DI-K`YNB|jT$IBM{Et;YCABx8OkK_vRf?3H1P8+; z;_`Jv0$sr;N*ek-4yFm2&tc5@K@<#MhGc(M6$A-LC;>zoNS$JwHe&S+?IqJJvfv7rd zoUx+oK=A!zp*C8N`WKE#uFL>^tZHX}H6PC(^!c%^!?Hf}pO}#@E_TO!hJ#gb@#Q^i zDlSPtei4!Ry-ywQX`A(DxfwJ7FdPGZso z&5~G6N-qKP)1l=DS2*!T*9z-${bni4z0uFg>Buv%VV57p_f{5H`GR>B=u|wuM+*cT zJU>jzO5i_Q1B$C=gk}U6Ku)?@`YEs!OEo}@JXcTd|6#O{x6(u^6Ij=D7 zYbnF0vd~y7lM;qcHVv^RlUs1IUrgVAN}Q)2mi@_`FRgBQcwSH$ ztw?@#Yst+c6=t@F>LENJmNRWGLQ~B3ZOeg5 zt!mls_sOwkl5@h7#O|wVKx)FvPwoy#L(mvsG#vl^n#uoCano>aR)=mP3X#oMP|o<* zkA86$1zW7ZRx`#a)2?sZosX6yIW5ODe@HoiLqfl`T8)$Lrk%G+o{K(jdmsJj7++m< zo_!HT{YSF}sQY@ET9mQe;J=p$WDD^epWBbYutx76It8z5ez(3zo(w-s15=S}m30Ji+@mQ(PmqSSWyAoKm|6MfISVzJP{McOz< z+I_tWKz%j;G+#j&OJjGuM1s#H_7&7^kK_sO2qDYy>$E+BGiVm;CQG3@a6&m*8i$WN z7=I>Vf*l>*E;h>}6TuG$j2I;g`FG!oLYQ+JA6-3bhAAP=*%%KKPaR@0lm|}tCl4Wr zlK!r2GlmG>k-VocO9=MSk4avbk?>PySmG2EL-S>gfa&VC$|?KB-f-5u6&3k2$vDaxg=nYkuul zt&7hSJGYC^^oM}Rf-43I9S-b!?Gw;Nd)68Ho;WrfK~wPA3AQwVI2-;MYvz?W7^TR9 zCb!fvH)++(MD>Y|b`~)e^8Bj(eE==(7f3i{^dp;GO}|m0*%g5Ai#0dWk zMLUR%bFQ3NyCnxMWCPMk4Sbd=Wws;m`cEmU|m8MG%%jlXqR){Cu4zB-cb&TyDJsoGp z>+rE=_@K}E8D|b%-Q*F>N7O>uvb=u%RQ6bZ+sfp3F8JX8;zTO)!+_@U3?3+k0Uij@ m>;HKf?ElIs|L4Y?U!32R ze{xZCFn4h`b~b}kGcb#Sq@GNE-dbB0(s*mJW1KPDvH=mgzSRvQ8bhe`1K2M?E?NeTzY!~Q`^T-_t{V8PQzePaT8vXquq`U2lAUYa-% z0}T&^@Ld{SE9tbJc3M+KM@mlDhGf(_O2g2%OukNsNcXC4S+(Rl=)3KA6jSs&V%irH zXfFd_nPzM~tSNQ9O8Y>c#+!Kd;IqlieYMzh)s#4pW-;VT%n8d)Qe|qM?$U?LPf`sm z=+tM-|NaHVJwGW@f_y$%l?lVjQA@7q%{rPpJe;7*wvpH6KG~tkn|$?}YPW;gJXmUM z_N26ws zae-16hWp3wNnN$WY2vym=KcIp$Snd zOLnDq37G+hqTm0SLRoU+%C?q@Q&+R0fIx1`)UrX`uP|J0*Io77^POLn#=jbOi%X`f zz%p$WJHP~F;;74N?$a!?S2j`%FK?#ZTC&E+$9Hwj&FS;r#~{#hb02G3dLSuUmvM3?0A3VaJfGdGb{S? zW#Gl$)YPYXuM6vC7srmr+dT?>oP71qmKoL)(~8L9zr3j5=|HYbW~$A)ySvBH{z!&m zt$dGSTa99GH|qL~KGDB5l+_Qs`SfuAEsT0~9x`lpeY~mDRL^ydQ5@GSnvekz& z>{#1}8PbCf*-!T7{olF*4k6GqyX!Mv#J2L`-d>~WN@K-YcV}m=jJI~H!0v2L>N;!$ zmNUVB8-%1u`l{wE$q7DieA&=irqVA zJuu~`$A2E;@besTTVe_FhH%@>*6!Xd-EZ>oPG<2t1qGwxa#@YNg6^uUM-INy( zw!N>6pgsM5??~B*xn_%?PnWsP=VS7WtgL38LC6#RZ|&!MBZ-*#_eyJ~tIaU;^Ybai z{4(?9QzktQbRF6>e8rsCd&&2nC3CWj;;8M48V;yKNJ7F_Z2RHEhfIE_@8}l!N!E-$ zO}V!q$B;V&@PI2URUNFrbo}->c5Zh;V=tOB#s-8hF7?@c$v0(K^e$S|_m%@#!(rnZ z&Jh0^R%oVL&1j@pl%bmew zyw-$axHR_aGeCLtg1trWmzQ4R8jKt9q+s$z*&htYft=K{PFtm zp*3zN`gdA^YoMrZebR0iV~|Is&Yk4 z;?&kq0xJH2sMglj-4(at=goIJ(0OPDz=yWy4LiF9pd9B1b1B~>!?>K+Wr^7h^p0%I z2U1t6wrst;1UqggDiBoJfH9_6LTS*#E)V9k;tsg{AANKh><5qhu3C04h=_>zDIeQ> zfy3sTnVB&)HGM%sBBP{)W9WBdmUx-`_3L|`^q$Rfxgug&o5L6mFbvClIC~J8L=Q^y zcpOIl>ly)zoF{x>KmpFc_d+8Smqy9jnu%}8hdVhrnfFpdPj3J+EYh>|^yt32x(c^F z-^2|_4ZITW5oV?3CEsTa7rZ9Z{^4zC?zm6BvcS)jWhu*n`8N zqb!XpUnhO@Bc$U(6YU%=qNwP&p@a9-W0Ij~YUWv~`XVPCjtK2@%NzC9ZJ)=tF+d0*~-^M88eos6Rq%Pz?FD!y&{X_jQ|$DeJ98oW@B{)wON1A zInJAWP_nY*BT5F&4Tab3Bh$n^GNjiKF!gx$pPg;g=;-LBh8@|18GJRSefWT;42WFKs~y=q zcM1wbai?+J9gjf9BtBkGR5poz3E0diw&w6>#d6tJ(X24QhPHa1eX24+E-ES#zMKXd zG^oiH<#NHsy%DofW8ui``!^9$124+wH0<{d;oRWo0{bauJ2*p)e}$&a*4xggT4=XL z;!>RLNW)r_O&3UL?C1AHEGLRJvdsGW`ZkYOe*G*g>`h8aA_Yg!RDr{a1jEewzowN` zv^RTsJIsaX<}DfKt9QeX#iTUu_7{F(M^Hv!pVE0u<=iyBLI!cc*{tpdLeroYps`_6 zwGWkLnOHI%2x`U)6`7Rd0+{{%{b@q(Y5dma># z7I|Pv2;S1BBpj62#XHjHA0NG#D}!BgIz*SlrfV#40a~GyCFnFPay8Nfv3P4c9q@Oj zl-olqxrf?&px^*tXkYI8nudwuV?d{^;p&4ZqeT46^@?9@w5dQ`@={~cuSEFxrs7N* zYZZ~5e6=E;X#)V(1KA=OO^_`Rh>(zQBwwDy_ANyPATJj@UjH#zx2}uNb2YNIcJwFT zM*M*=1Eq^u0IWtJC;cF4;$#)Lho1P?;C9G}@hkp>(z{X2heb1>t*ctE(;Cca#w_Jz zOYDg(Kmf9NJSNZ1Y1;eZSye?MdI0kTC^r@v*XQRY?70C#rXY%avnSW@V-!uMz9fj( zga{=)3$R&Wfck(H+N{W^PWjHyA2f<@Ft5LKfL2AlyI;}%y14c=Dh}_CkY&$hKUP$W z{g$L!yhGla%LH0zQ|J-KIQL<$Yt~T?p&x zNlZy82z&#L=3L*{uu4?O_tBUL`?=b^Bg!)vz@C-b^hv1yIe@s@%v&m2+(kZ*JvUe8FDty3i>9dXahCvpKAXh|B<)uiN}dD5lAi++H#|G_l0K4eGG8rP{u zA~lUB!`g>CbkuE_BTd|G9>h`LOa_?#`{2$6(i%#BO`Dkln0k!tF!{SJTR@`F@WV6j z6rm;!TWwP7LU=c-Qf#M}AiT8KpdC;``PmP^I2+v%%}cMxy+b}uxB2+rDTL1dqYyo@ zkAwi@P-j)dIRtSbro!D%lY-%*IrCwu!c%K?Q2Sg%xIaaJRN$o(0^MOO<3j6TsU{5@ zRMOptUh@>&nEU8b;=ojI$~&ztrPi^HutvnCssqHpb+HkMU&GO$A6#tFB=eM!$n9}R;kMra&qJGiOLo|72t^{}=Ot-vAZdMr>tIbG^ID4Ml zQ_IL20AX~Kb1og10}w$)sDF7{r$o~r*^R8@{`@kJWOCl`yz4kgDBu}Nzbk9tv@xZE zq3Sw&!uk66f>T(RAK(4uk}KL)6AH6e7|+)y|IX{4uYO)A*P~9DWk8Q($%kcI^)cDa zDje|W{`q{lHam(+FuFF#Q(~)jJ+l+m=%j2YCf{_0dfe8I;xTgn>|>{NRM*J@M;CIsB;kq2zpgfmaDi(Iay&U@gou|N&M)rnW)Mjb*D zza_mM9&XymYQq4MmohEbYcI2D3hyNVRm(q1>WC4M0YC;z$!ayO!YZuNNCKIySu;>s zcwy(rUz@WL_0Z2<7tG<6?+g|4lQ1gHM>H2`M3LRW1J28z%6=0Brs%~PBKR%W-0J_HThrRq3B3@U+Nk4aS&7qQ9x{E>SILFpsSBmlE_~(&I@xnxl zJv6^Qqie=IgH zE9}=+j${~)xKPr+CAFJL{!c>l9(2)Em^nTOjEv$Ung5m~Z6?8>Pmk>PyIRM=G*JI1 z%rYJ1hf2Y0KsN-XyN(fH024SKkvvY!fnu@;(D4{+Jv;5l!GF;^u91b$u;!gHg>PO> zOQf)j0)8Nqv$Qh=cAJ#)>wHJkD2ER>+l%Q(Ak{y#L;J^Di_+n~M8`qiKP(b>~xh_s0JmQH3lL4LVrcsb|kv(VvjqqcN@VwmV6Op*QP z+4xpVkSZa-l8Sma$^^jt*5}y1-}Yw^Nes|3_?abX^Sl4vMu;500O; zoQ=l?a}U-&FM9oSGLw^%DrxslLc0Oyee7p&PwCJH0(y`6-_>?9jk_mArj$&}ql}Us z$6>({aaE}KjkADC5H+>(TG(cIdHwvU;;K^o5R%Z+1=zBo5G(F)zTP$isYkvQL-PI1 zS;F^yZrS?nVUXt9mA<{0mF}Df*j8F!a6SS0a|79 zQ}~W(m~H8ovP|ah7RY;Kk;>qI8hivA#28rSn>cF)z>6X+I(EScu^8g?^|>0HI>C$| zD=iUWP_-%CjonUt2~E9c=C~16%%1c+Z>6HdI4uwl!9AvsU|OD;$#Xf+uC-Axv2!88 z(|K0Kt5zAy%83`iu%wDuny|0~;Is%a`EbLcNG7IVomt{RmOjbAIwj0w4}TwM`NVya z+2&q;2K+G^kAAz44$I2#FpeR#Q+4-B-~ky$kys#>^jXBaRFk8{O9Kh0(+u4aZ$eq0P3OgT0rwq z*uTv+-UWcS4b_~9k}^ChW~Sx8%Ekk!cOJ<6ZDR3=a-o)J6cZo?Ste8y$Fe7;N>oG%azm zG|=y%SSv+S-T5g@_qwm6ot{xN)O|~@w>0X`sK%38hJO4W*eHBVJxq089$)h9;N_c%<^&mXX{}C7e2UMoKfZ zBEz#njz(Q{(!Ku@Zi^ou1APs3xEBajWs2l6D_m5GhO>(14ZLV{8mXf5Ia72s;)5Lz zHM6jJ@5{H(ct>|AQ1+g255;Xmar`C$$&h|`r_=AZs}+%?I$FY{k`ehgu=$%YQjiG3 zdXXB3MJ#W}PTf=&bH%%iw{Yv(Z#IV9`KnCiajY!rHdgv1w`X2nC+eaGU8-ovg(zjk z6KQGJUEhwec2dHhj=a3ic?zuP2o-xkf4tw?-8mMWPaV`hvin|GnjmK2io+Lk7$PE} z({(K&69UAmo||W37w87@BCMwvc2ydGiHCRSz`mMCd`od7Z8qTROV#1wOC>V<`gFA7 z{n=TRH(v=9M+c%M7n0yDpZv#!BYE-^7(0*}er)G+$0N~UE5VTHob@KZfFPY#Uqa)Nm(0Q^24#!(A2r*jYll1UAv6;%?&T?j0$e7P@g9 zjk;XyOD^Iy6=f|pkA}(sjv0 zg3WDDY1mp5E$`m2F*6aCmQ)>`|KMP5Jzq1esfg|%(1t!5qWQd)EQ@S#p-mqQdBLSs zM)ob(QoT2GV%NTV-&hQN0v@vxKTSJ`ZuJzj)W7!Gu$6rt%u$Oe{W?+NdA!uMqPtx{ z`(}oXPE45E#X|k(k7cr_HQskstg&%Vfb}3j!HKJ9YcYb#Z@qe_8$LVo62v==`3xBe*5p6oDZ?F1L#` zksfpq&=8RQj-KN;Y2wjQRgx+;+Ix)rD&E0jL)@Opz-f4fT9Jwq5))j?%AXx~%ZN(Z zDKI;&r%>ho#1DH(-BD9GrYcjWPk+cZ5M0GC@WBejfb)IfLtQye`j_O>u2*>yMm@(> zB$CpTS0PJKq(mYAOe1EM5`*vsAr@)C+1aq?7DEH4$NYk=-2T9tnwmDhx9I}3d>2hq z#~j1~ySRS6jJbSCOC`^!@fu&1;IMu`<=ga}nAE<&q@jVf zM!uS!^Ou~O+crWctJp) zBt|7bpC&W7D%WD!3rCg`q?XZ2ySS=Ui>8Lf%0GVQv$o+RWM^ z;i{FH`$8;|eoX}+II8}YjMTcm5&suDkbX4h3}_6QfAz=-2ym4$Dsn&|E0=RCAtb%> zN-I}lx4d^lJ6#VJqJtZ4cCv9xq2Jd(_E%sQgT~~f5{ju2Q{c&eYm1b-9sD`m)jD6VT&osPetM!~?8qlWxT{H|n(skl{=zY5hhlh; z?_ajU^!q2#EzU{~9mW_QbL$lSoGkw=by+%U6$_xljuG06E3n(b=u^#Td}K%2{UzrH zvPySE!{hw@QPu3zp#y^i%a*Yd;erW00alv;&Q4^0hJ1ccGdGfAVN35msW)`Li;0u( zvJ#?8Z8UIDlj^T2WD{Hue`rY*u{J0J2T(>{hIkH%JQB08xGy*TOMmaOw5$%ow|;+s zZ2KQ&&A+X*8N^0h82yuD$o=vDcW$EVn~BV512n4lL3Dl1aRKqT@{p?pzXeM<@5>*< za}}pMz38P+_Y^-n9$wMe0PugjHKp)FQ3@ZpH3OCu@p%PVPboPm;W4FS^oj;IUehDm z)#9q){+6`7m4l|q(dZacMPsl0jbPir--@(b>#cY$ybmud2la~_ODZ5LNK9<37Ja_T zi3@q`c{G_a#u=Kn&2@ZRgG)om58Pk!N+cj`8g=%WlaMO!M_x6<17ZC4qgeRBGDCEK z{+gl2{XL0Sj1WTB;^)yQ zr2-MAs7`gAJM_z@YcHp!_5;wHK_W}QZ^z(M?(7j!&J>2umK(`i>bNwv+qz}%4)yHN zk)56>+0P3lCw^cwRo;_G>94k;(RK9Fy@G36UmdC2JaOH>b&PYO`AIBbqHrxUrzkJa z{h#~Yz^#7Nxxa4@>nz+?E5}g9i|xq4*KR9u+RZ%Km}2_FztfgJjtj=0E-i}xIkFF_ zC@&^!wMls$nL+%vcdlvbu?ur~3Ak@+9t_ZeWC2tnsP`}GoSI^_8{EA+1+&9}VowwT zIX`n&=xbf)PmlLPdKW_o?~mr*Iqi9*js0(iN=~9G(Ji_)cyO!3xFjwoIzQ(>v zRVDobvH=W1(tPk|ZyfScT(<+B=z{c2@7&{8?|7j>3C_fREjg_09fRwyE4ZAW=;dZz zp-y)P-2+*O5yq+TjN!Zsxztm-FL2cc^??8fa(Y7&hTA(@_uo6xi)@0;Q->wahn|Lf z_k!*h<5KQa0)5QPZXeHQGv*tuUy2D;hpA-^<(|gNd`R_$cm#PRJL(% zRdnhXy({4suf?V;gR8PWJ#5u?IUkSAt0G~xsq=CwPaR0HU|p>nZWz1B;=7({I2vuE z4O6qY3P#Ogc#gHRm5&xsIIfVJbpEqoW9xVzhEnwI?5o?2dq#8J4Xn28vJT~w2H5>U zg8yS0ROklm>Avtf`es*?`teXwH4>ANh7sffon;y?WUPoZx z^gN6cl{mM%h+ZwkQ6QK30eCZ=_+$n4Yb*Gv! zF3slY*9~gD3K{cs2d)A|sNLoryqNts`c~r&{Mh<38)zXgrE2pdFBG^wT?28^Tnl{J zn>~N_`yPqai_E{ZM>&g^#@i3FtzT4Jo!u0d>7CgLW#1k82@-9t?bAAnO!tslBJ;Z; zZCrn381usB3l!tM!dPyArMiL=kV=W-UzfYak%6f%J3Qnj~Eza~jEkobRu~iAy`cXn#cj*A$j&OA7>wK8M>P zOK^D9sJU}cY>S|vdTsR9B7T$sqlmdJD=+Ad~hq$kOvU9NAr3reeW)B zfV_bFNfp!|a2G7Lr4OFJS#(l<&+VY{gUdR?=mJ&i9Z+c6$Ib4})?H3xU z!b*Or$P+Cu7K^Le9^a@P59d;eVvbeDRYpU z;ZVua#mcy-hSF8{xsF$MQ##$GqyqSdZJ14mf|e71yyC?Ir*$hMf+j(DjGyhUAFNpE zpdA5auvD3aukKD8Yw+~&WWddE(F<#9T^BJOEpG z7DauqCra-EAtUGF?5IlV>S>~1F2PPR#Kx*LGj_3_f-B{>sTM3cU%=eU4}9j z&R?r3V*362$OiYapJ(xiq+y6;V$|i{P>gLx#&)XVh3$}IgQ;{@Z{aLuuqq?AZ(oRw zW5p&2VUY&=u-MFcH*7eJJw$CIxpdYw={H%}4QP$_q29koqorfBzHMoQjU5$S^D@N% zW9k7tjR$?U;cCz0l}jj29=T}FyPp_cBCh&Bq5`gDf?KT9$j??vichF@QIiXl*xfO| z%P3{8M%gv7EPVadJcvqK|DapGc*g5L^S~M*&A@STqWQ;cRm!vrX?uFK!TI*~^?Lse z5|_rvG-fdC0fH@3^ot^H&Z6y!>44RK%2YNbI$JiHfw3X>N?RFG>5O`Ab7jF4GrwOW zh;vTrN#NNAnerwSdZo`NYs!53wgV?G9mT7LJg1?fbv5cL+dSUIW~am| zhDD7&f&w$3I-#n0z_1310M$Z~F2vtqa4jhcg^_c0)b!Iw;#*)oWHAk0WZY@RQt@cG-0&#Zn@gF~`uSP^QiL?9vbg5&wHyHPF zcbJX0$IK@ri~D$zy${HKYpO~jD=#rLZ>1Qc^ILcM^tEFc6I?S*fgCXDe|u)lOGi}} z+Y>=%yt;+WP($c?H2r>S&+&1YJkh_WoR(58;$5=Z5J0Sq zYa?W98!wJ;u-R%-fTLo!_WJW$WeSaa)i3bV{^wU)i+a2pBw(~9$-$UysAW1LmyKk#pkW-xmi zt9qxaILxs!okF*`-qOn$-1&?>!tX%tF)cn^s zLbcTPq*phOQBc^n@iO+O3z1`r?ZiU(~w)Ke^PN^kk zjGD9iie374&Pa+};bFFc>z3de0gsRz5?@`^gK2gg+0C91y12$+_+}EbxcH_hE01m3 z#MjcAXG@4`L{W}uCfE@C@RGlhDeE-l@%@}#2C0W7!xYVoyk5QrGrL`)#7200DktO% zCVO}NlQq6AgFdQjkX^}<*?9pggqAHESlZI~T1Yj8FH&@ms*L236B01ga(UOVTFl^F z7>vmFlsRGgt3PoVcZ-UiqiB^b*EYmRIE8vf-bb1YF5ihWbAKW`(#GP&)q}jzG{@?Y zXv-$8p8Sx_Y}nr=9dp_s2{4=L+Y66xSO$0;)bdoomf`t)`}CKK>0bEZ-2HB39TsC0 zcQ&Cu*{-Z4-SafK^bC1BhZdog#j^|yXui(l~I^H8!8t}5H?l8dx2 z3Gd6}orujZkf~Uun4bdfL^FDx{c0gNI0r|vewwI%lwxSzu9RlY5Atmm#aEfi$o2E5 z`NxV7rS;b$cILibPTi+ZN}_f>E=hEv$Wp_3E~(}vDsi0V2L7q@!mu3S3eIFVd=;OX zf=f5p#Jq%b8?2Uq*w4@$#A>7*K%k;Io+EsRh2d_l)^3M+GN-ov#YUD<^n*inpzR+! zUHi0K&&Dw zA99}zeVRh?B3pfud;%>8+g=~nk!4pbT$C6X3gvU`vhu&Iv_Tvi_LE$RMI_wlUK2K< zyR;PO>ruit)HKMmY${XbA8)4%$eKdHG4ybxymMltbVkx{ieIAl3b$ zbnsWg-`x<}Cho9Vw;z@IAR;PCenG8~+h~&^;j=ewiMV(U%nzA{R~pUY$^|P}nN0smzHmY57eF#%%H@c9a6T z(AFQOI1H}K$7u(BcLrQ$7A`%tf^rgfAXPykXY;$f^H?z{{~poqzpeVb(`6471zQs) z?yd&6<>s?p>e+|m>UZ)(Z5KfevsRTlR@Z{WEd|TtDo_lU-CV{a@xl!F^346>q{eB2 z;JgtTJlGt6UfuBdTf@qhG$}gk4N~WoME~ z%1N+P#d>Y#oXccoI^XY{v@g&I;g{X52TuBS5;LO-^eg*^Q$}d2l(1~L#LT{SM{f20 zdC@qAUQp=G3@^7pP13ZW6vT*N5sw`S)!dhAy*DZ$r0^_D5J*ooSrE+sGR2+N+D)iL z{~J7u7VD;jFPQn=*BO3cHFiXVJ`BFfTed!@lp$cQkQ?9FE>^R^#%!X0c6pJr4}4p= zpt5A`uy|V6iJ{tw25i)DmNjgBz)%NZq01kA#EkBwF#5mLsM6@5S3Ba7^dl9by5$+s>s43n%Dox$>e^B#>+x+(YE&tHp1QQRyD_2V8Yu^J zV^0`6U^61x)W3A6Is5$==|$aB9|Z<&XZg;gQHlQ(o(@F`aZw;!D(C|9jl>hNaZ9ds zv^E?ITuGfmqT86on_BJBV>j$duxxHL(jW50Y37WpOd%Ej_8B+9EjFBD_V3Mn0;SC! zObCpU(64&TMh@r-x4G8-J>2B&xqE)XS9`di5w^0Er}b~8J6?Ah%XW8YMKKytM zrMH(rZxmljo4(>$B84>V-B7{7Qc;ENmezlMX*U=--*VCzDI*UY@%7bs(Q+-;-jIhQ z#i1EPQd~8bjPdPD1B3(*wGxpOlM<;NxttGXds*974_pZp!7x3A+`$SV;RDE`oz@sP ztqo8hCx_`c=qm&WEh704mOcR2si$nRZQej)N=PE$#{Z9$eL_6R$4FVW%8I~2K{y#n zl}y=*gki(swhimPt)u>?9JA=1Z7XRhBzMDb+2)4JOjdDuQGb!Eubzh^)x%3mkg4ZX zYw7Bgj9W0FB@Lk zma%VUc0$GIQWp<8@+_czTN#b#W@Oc&2eSo~} zwAI@?2jciCNqv~n>56|xSU3wo{E^<%lNL}vkRCJ~)r*JMcU)~CwT?b@SN%~c^tKfO zR_LCu3`JeUK1VSL#D)gw%niV2JBegtv=SXUhaJze!{h76IYfMgTwUDj&%K8459EEP z)7Ved)!}QnERA0@q zF19w^`=WammdcfDZDQyCPfiI5JXl)F@{wE^$TZ;sN81R#nS@)A&;zEw;ehMNe(Vor z#NA70HXM4`f~vt%ec-?Y@h?(1Q+)*d5o?`9lu&R&Dt*Mk(b_naB^HSh(0*7oiY#6{ zTl>DySWrtrL|u#59E?;Ao)GRpQnh*cbYp&Du>1CX95Hrw#~yF_;rwZ^i4a1@Ugszo7Vt&6}YU z;Z?!MA#iu;B{@17<(tEd9!zxEKwISC*svOvmps@WqoilyHnE#pQl?)xf3~M>>DJ1Y zG|?b6w}>2V>TEhuOzD;Ny)%%NBhhQm6asPKT+=z0asK3yGQ$iB9TGz)Me=zGO8a!v zbCtLLy_ntqj%C>wtkA@1&OS-`+gDP(-RNxN-U8tXg#+e;+zrMqB|7kl#LSk`aeN}) zW(KuMy&~QV+PSN%_k2#5xu;u(tTmsG4Dfm!f}E+3Su~B}er7WBI6lHsf!uRcu!1u1 zTWW(dMGpRMZ0>6Kc(nAZxQ-4)RXN7vE%E93RPL`3!SGV+l%>nr6l1EPJ$F#-a!qc! z0HA%nas}4agbBM=n7fg_$WxxpjA5l{uhGLt5`>ljhk(TuHVE4A12-+|r=#XJnnSLz z(|a4cJLHdJIWSM#<|zujOjT6oGeJ4=7)-lfVyx8R!Ej z<}D8kInHhD8RTbT75`-A5JuKMt6UM>b573x`OeJ$nxOmb3H4zE-N`NrYXJB@5!?r9 LC8=@=qoDr+QV)|{ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_3_0.png index 0be2e475d8737b746dd714ffe5682340949646be..656724cd5f5e082dd72dbf223555b261cfb27430 100644 GIT binary patch literal 10355 zcmd6NXH-)`wC+g=y%(uc#2~0N4ILyPy*H&-#UMTO9)g7;AP}U4u7HU2qI8G~0uqV{ z(jhi}#gZcine?-j#FKnar8J=d3-uelu~#hFbJA+%y0H(CcWcn*spn3F&61 zq9FCS(&1_-$#IflRwhK+uy_Wfk5CRUq4rGFPJ##Co1sR-yi9x zAR*!T{|1PA`?^T!fEGK92Om=%v zHhUM8qW;WTR>|`y?M*GSMwgY2HY1~k?vRY({k(kBB1YXphS4`hX`Qcc=;eNsfV?ig zM=cXTr0OIt&GMK04UO_j%y=z+tF9aGaq{lR zFOJ?>=(2h)19cu)^(%Jb8o&qH4g<2u#<_74pz@cu?Q|X@a`jHvXfSQ_m1~rM7tNn> z!FCa*9t!|Xok(CYv^D1gSB+x&oe9vEfI(ytd+%9 z%}M4&I81R4&z?OIhLHJq?=Ie>9^2-O$PcM%Fl!Pv6KoR>TRFqy&m9pzS1*LKCOlv# zBlx%3R%&Gzb`KG|t-D#Sh&hkFj(Ge!A1wO;q7S${XWy)i0ujW4mio}d3jDe~%*Gk9 zl{H~xb5<3Domnk*|A@^c(b}9{+<~1mVcWot&#?hAbu(>i`&ew72jYayHbZ_kY~O1$ zLb6T*JCSPdiLAlSSRpssiD&Jl*s}?}HCxyTiO4OYzkglUK6FmHPVKbSfrXIv#jv{@5lplLb+ z6gUH>0mt{)fU>`x)?DbrkX2av?UxZ^Xf^G9P&Fkwqjjtif-3}$!JKp!Fwvz0KxS_k#)9(! z3?J`ZFaxvC!m6qOq}dw!=AXAJ$v)5w&^Rps(x<5I!N1L#zy;<7lJ`D>I*>`y;tHY5 z$+k2EC%{1>`1C@kDT|nj#EzzD*!wA=d9;<6NAz94vqcsRJ~BP( z1yBaQGQ9zSIC-^RrMZePO|Ox5<<*c&XfME;vc^Sc*2uTbVnJfb1+mYC9-Ob-t(iUh z0-_TLxz#IeZ*%GlXoXzBS#ZTrxR5-*_*DsivEGo`nn-|f1;ZG!5<;f}L#wrQS5j6N zV)q1^hQhW0M=>-7V3T62v9t2|X{J?fM&+(H%5dinVDUkL1IX0#z1_mk zXj1mrJNah5;tn7U$`-{yjz}T@%!$EP1h#ngLiYL1KTG&spY^%qX}U`P3-X<)A`s4; zd^t>c2fJ)eUgfZES+PYTRsXcMLGG9wbk@Phamb!EA=MXToxhm7yNlS*!sc2M<<3C< zQDIxYOP^Z|^XSJiiMm^3Kl91rZjpmspAvb1a|j2M+v}<^sT4CtN7a<>elp1vBBH|U zj?SzJvF5{VMw||L_qys1R~4Laxe*iMAOUlr$#7jdv~M0OOUwar1cwH9qZ$a~5!y?- z+qCmpZobiMVG~dMnkoo;p((9+#dCiQ<>$1KIdoYBnivC()c3vFh-g~#Sp04opHd}H z9QhJ}(xVY!a?ZhedLvx06OLP3SCF2AFt^}Lhb$`#2|r$X!lqynmX6>I<8d;Sfbl#+ zL`;UvR3d)1jaF3R{p}bDO0X%=_|TQmX9pe1z)z4Rq@WIK2RNqQcIcq5YC?c)VdJ}i z@_|Bl7-=oB8;@tJ?HXVH+QpYwVLj;y!o01gY(NZl;?t~qfb5tRd>l`aHNhz%P(@LF zxw4cB5vab|M!9|67HQ>>yvZ(chqb>vY0Q6rZuXWCIU5zxukmI3GTwvS7RV$glykS8 z#~``z;6{eg)HZ7^9%we`loNNc%Xh(`z!^tIkZxQ`pWb-3&)@$2))pC39^H}2a|EMR z@v86h2a<3BZwJY%Y0k5rDEJE58Wh58jidaQ%p)UhP9-@DV(?KR1qIbumL;KQ+ecs_IE#o zR@lBJsKde-E2-a*;}kH7PM0m9pPE1S7mrimT!3>hp;?3#x^hYPM=}Q$w~jc&siUBv zDZ4a%T&%OLrSy}a6#lG0vR)kxJ{X^uC2#27sjyQhC`n%zKD%wjG_d4Hy;nE@@uK9 z-zc0f{IA04e;#E2C-rhz3rN-}z{0-c`RLX%1qFv|B;6q{Kql|PA9eN47-Mi0mYzbA z21i;gBX#vt+A>X~G};d+m)HYfJl=?Y#Lw2p-`^x-gaidif`ReS6CPq2Tb)Y;t zG7OxO0>7Q(Vq=99xIg6q46=0xz$rNng^&><2~#^uS{kGl1I|@w$~Q1DDG{jwGeV$) zSLHK_#()+mIETpiM7c7y$kL2e5~U;=B(Bn&zu0|~^PcuDl{z%!nm%A&=@B;fEo~EZ z-DX*mQFDhMIA>nEr2&0D8`;r%Qf&90IaYgTWGo}7qh2pcMIGa^pkClI`0-$GePuJF z_Bs>kZ2~W?eilSZ3P5Zw7wP=t^e%#wB1yvn%Wt1^n~Iv6mVH)T4R`Abg8BQMkp&*NJE4v);aFl107~kR%P`j&eHT5I7&PHv0yseYnA^q zdgLYGFYL|LF>D+1I7+)>)e^+iviXcC+MZ$pzl z4q~9Epp7Se&i~df7Kf|^4jPj2Xe+tiwtsx@DpN}Ew@K$#DtGulp(s5f^>%*XA0-Jg zuU3jnVqr1~92MJXmVRwf!}i2P;NS3J=hj$vI1PkFf?HgqB9WK2^?j%NC{QV7u^*(DGXbJGbX5(S8yr+EN#ZLOav;+5vPF z)!En$OX96mZ5fQmWl2XB&R$Go?=VvJkA%CPK|MQZg6)BYM25$KAaoiu(roqCswjI3 zy+VMmbktphmt`sNuCh*Uo10`Xl6C#9b76oS6OKof-$LE=ZiH48ZK3O9CbTeq{SqL7 z+hbPZ{OqGKe6=jM7=8TmF4N)mzXbO%bf}wIr=r}b{-*TA4ccWs%+LwdsWE?)_RXxv z!pfb^Nh<2^i{b)VB+0=kjb<*TM97Fmk%|CI(5mPqfuXnnlJ5DG#07#O3N<9ff>PJk zibt{^d2UiyS7}G76q(n`{DiwN^E~s-3Ing-G!kp5wpURS>V4ru?9D_@kh zn6qbj0-f@2MXNn)FRtUdZKpV~)ii z+?KK=@n>Jye$D_ROE?bK%XJ?<7GJn6mFc|h^(C6VPFI2IlK@nG4Do_@Ya>@JXH<&?)0^44Mft%?BF!s}XgFB!Xij?$2x=N+CXyf#3#f>B3Wx=L)3Y%ZmK@m+7j1PAZNy`nR*Q zE?V2fM4^ctI-9fvRUj#I;`!|C7eSV1y1EBG1Au!)p9Mvj$B(f?+; zqOAAev^}t~^BHynAOx_76kz+W9XWh3e3hQb#C+XIT2J3!F8meIed9Fjo)G0F=9I&C zZgJEYKeseeT{o^h9I7E*Kw1vLimu=V3*-9VIkDRAxAG3}kF1X39wy!2WBKQiGHMXX z4F3jXAH0u@bUZ{J>=rzIA8JKXBvHM|+J2|l#ZHOgep$KsXQyF%y}~#=7U$IhGAl9P z+xbVIUXMtY?0mE^_AL!Z#s~=KLg=`_rwbgo$q%0VYT%0_M&!8DeO_IX*7Yq+)3Kz9 z2&39Al()O#;6hE%*)b@FpJ5uMSbeg&0WZ;=th0pn)l)~*TU(=-^C|hlhi2?Ux@I~y zK32Yhhe~btdxK}1(@<5%dt{*pnv2s-kFQl=oZo2*oR^y2idYz?!wBhRfP?1f6L=^+ znuMym`Eb87$RbX_7WUNmJo?sM2ISr-4n)}~0I)??HT#2ND*dwHOL(ixK#nYbj1*rY3 zWMQDTkd(y)b*B+d2UV(XmUK97SQA!Cb`6xh+kS1%IZ=*H!$Rxb54LQ7<;)u9+HwLz zRQIN(NEUite~kO!w3H5J*U9#!g~xM{F~ZkxpKI}f(#6P{1IhPi#vYxL1U2QsYROGf zZtDkkPvds&s|8hmKUTj)jn7Jx*n^*xw|Mu7MO}S~#9n|XfwnNpjIVL5vXvk132c)J zb2Z?U3#S4=B;$zEB)8*MK2B+v`vQX~rh$`D042Fgw=ZmsIc4Y?3E+&Qt>&^a)tD<| z>e5bk19-ot1&SZGfjA+K7F!3(z$|!sbU&|fU?J1AVv9 zC+0VFvE%s9f$2#2b|F8=#`DsHH&2<}TxziNM`p#e7P(- z43uNV2_XRoP)s6}hmtM+vmguqFHr;L?zq^lpO)5b+`&6Kvb8SX*a(YBGfYEVut^G# zK)~9q=eON@@Xa{#sB4%=`o&B^7J-h!o`w3}zUBZNY_QRAcHjE4Z=cfNwPh5-@y=VE zI|P9Myugy#MDM4{d9lxMTf=N7RbI``aW?fEnGbEJ&U;e4R-!j&^x!3u5k9u*&0jx1 z=ALB5)d2gXAk{kc`U)ul4p|saFin`dr0_-DAQiV(MP^|OT5oGuTjy0|Kb)FF-+G7+ z+w@`W@M`s0R58C1VP%^)xF)uIS;G>F(3eCr+w7jL@R;?{zBaFeP2%tGL~7yWJE4b)h!@IgG!Tq|YoA>Ii~J<3k?UoKV!A)p z#!_J}n2u8+`S9TLNjg6Lyztvcgr~$8mo{sDp~P8?Cl3m<82fE z;!pR?__=Zf`YB@|*Tt2X#_#YAovCF($mDDNg|b3sDOa9;=3n+rCa$hN4^IZ+b7`9~ z3<0#wZF}sG)!9$J;chO!P*nwf~7;V|}Md0p}paUr1$XuR(3dMTA9Hp4_c*%=^i{AHOOZcGmVU zcJsZ`x0bTWV)$0G!Won~$2~|P3j28VYOENRaMx9}6spo^D^VEBm@ zs_biX^Wz`a0XC}6B-1`f0?#svQIPZ97l{ydv2Wg^auLl_?`v#i8}2W1xinUA$1o%+ zq?tZbh%bIvtUE6z_NhWOKboPVAR|6DRGx+<_~;Q~STE|%K0v|X|NLo5KE+F=;fk?~ zUSV;$oWQInDmnKjo0y0Oxm~x++@2T?u{L6+=Q&0aMyI8F4WwwpG}phS5$j!v*GrPs zqKQd&!W*;KN2m2a)uG3^$iAK43_q7)w+ONq|C4a()ZvKv-x*;-{tz{qCX48J9 z!C$S)-YX;~+e?r`^xfI5VvkV>c#m1T+cyGO{MxisaQA~(L(&iK{WO|4buE?-24W6y z;fDJUD+>>bGp5hfVwx+sw2}h*}uwEU$7@T zEBb+G!nJHE(ROn${;2%!pb?87Cj6>35B*o}ZdR+IdYAfTt|lpGdAJF3NK~!}r8ufa zKrlC4t?t+76s#=g#6Kc76|4x0jE!-?HJ1m76E3Mn$~u*HHhZpmQOUMyR|CSjV}x}fi?yL#x6CTKgMgk3|{Z8%?uVLu)n`2&HCHC!WR}| z1m_(Ov}s8A(zM*XcPgw8`t~)B1h3H6~-j)ZWuAW`_1;yFrFuOa0{h zCSO8M+r#4x?Ks;A*Xht=+^PQ3!{dc~cu{so^VMfBJ~}t_-Mem(6qn;JFB*st^q1b{ zdeQwW_4SQbEgW|0GBQhpVLeQ;eL<5_vTL$ppZO%2M*fUw6{r}SS>aE0b&lV?c>rn~ zkLb%tJ(x2(|E)zVP@Oiud-hefCyuo@!=Vy2bKXc3M&*+{R|QGn9!5cQUhf_4Vp@eg z=gVDz-@{$S5&c=G5u`Hkgp>~$$hfxLUqplEOB6j!!e_tC`W1eCKar%8+ohB8%CJQ8 z-feeT@fQKQm3C7yRDgH^sg{%RcBrRnToWfWE-vsu+-7Qcx>oC~cnXGIfK^SQ^1Edp zpb{o_h6MZe%j4eg0{1b(uo-QW^}uvf<=b={p7&Xf$z5mJSBq4V*-Q{9ld3QW5mzjP zy7lN9a`gVMN2{ZdBPqHzJd*m3KPgZ`Ld;xzs7(^Mb$x{cUKN&`S6$-|OHf**+tG9Q z=FoCQZ5qSyeC^DC!@9*iC;fGXDLbj~gIp`U>i>qykv~Fx)|d9EZ7U_>xbb1&nkHYU zNXSe8ak}ZK{)kKZUTHmD+pF4ptW?9d+(hQv&T7&}m=&B76t2veCpHC~n>i zJsbS}y)v=GqH6N;U{&>8-Gi&IL-!Mc{OoA%!`wu6h9hfitfZenW$mDR1*`hcimUk%={b^7(;>^JyA2c3rz=LiTKqy574LV5VDt4mX0NZ0*78#QofkL4 zyx$(BIPx6`STWt%Z0Yq`>;aKY`>d|^v>jZlYJ3uYw)ZYehso{f)v9ZMD+x_N#KM(c&mPOl<(L>~q3mPQi49G?iSGfuYGAsSlJcE=%H zRryphWYX0WwrCe5B+-ZtKM!fB(Hp{8F5b|ZB(vkZcU4;_QDB8M<4*)l#5PDigRPK1EKQ;9#TeN1_SaN(_%q|y9~pG&MIYoa`H0RjM$68#+b3C?6B#PHq`{#lkpVo)yc z*E4sKr*ud-9mZ#St!kc42l6yM{{D54T2Zbiydl)#?^yl12Y5wa)gD`$c;S!|C8Wa)zo*J@Z8EuUrMf<3gA2Y-K zjBq+pE3p42UBcy+bT2+>n!}xpvkv=Ao?NY((&i#PmHaE7_7Z>_$pXr8ODYgNa3{__Uz{- z^a}50f5c-KH|IiJ2-ALcy^)8=g38L{hreXt)DQjM93L+fH+?>PPAlu)>-l?Dp$%hz zf~#FI{gFH}pv4bjyh9`v3jE~h5|qtBl>2k=IU>wid!)4d<=jl;$TMg$q`uH*KTyQX z#Z#%=VJnc>eE6V2&7yMAeXU}K!Jbix5cpS@2c8e_mrdD!JYt*{rBWz!N`1q!(o)qh zSNA3V*6mZf!$*uR?s+rwlm)2>#Z}@ka?}Jr`pPLNdobC?k zr1DKQn7qJfnfyA)p3tzm8vCA|8(~m1axzrqutRIdWnv&O#KLNlDfxQCnFV)t$La1~ zl>xY_eHAh8)%KxfE7fS4`1a~cV6&y_Z&9TPLgqNvYi`K+)Tel}s<`IijB|c81rC2! z?z@vDQ9PR@Fk52U=Wpwmm<=?L0Ju`C?<@CQA=EqR$K0wFdZGUu!QbSmg5q}Ni_pYG zq6P;EW+N?Bf*Cx1j$1Wn_;Pynpgg@nB^QM>y^y}xBx-pu$mTX5DtM}byW`Ar=POSy zIqhU5%C`OQFXuoX9ly-l-~m%+@+U>5MMY!73o=d}1aQOkur=!PLm=5_c3s`7tLkgSes9~Hgn+Py?BK*77kE_^K}rdW5A@FIsbiu*vbDWzKF1k ziMpAyDetY@=idweCA!W2aWc6qM%J4OSP}rYhb94V&&fUQ6CHFAbqFd}d)muJzdMtD<7=)^NRINVEihfcX8bA`>T*TN+N1yo7 zwfy-ncGNmWraHf`8t>lsF))yxvA-Vy@%b|!Ls|&in$Mo2!8i~pQc*I#R7k_fNDmmh ziKF+C6PD6ZDh6t!0#O_e_>5zR;7>E~U`gb9jri_d&|G54X8g$Tc+H=0hqOYkk8Y5V z=x^UgAUQ#Nl99?oWgJ6Iu;d6CA})0y;VDsip&M z+HbUk%NtuPD-$wo7c?(od%pU9NRyLwc$9NY^}^K^louvGN8DX4E9TGMV~}8ud;Yio zYVh8u<5!*r+4PVVQ$l46dhsP01p{P%nzeC(rXVts@V-n}cCXX3HSmMftRmN6WgW5m z%`zIv4Nj_}EjAvWNt-jh4_{t6n8{e5#`4H4$n9J13@G**!TaHG+Atv(pVDRM@Cj5y zAi$j-TVwFWSS-syH}Me#x++2p{xPzXYHZ&@gnIEqS=zHdvjzq@QX$kI2sCUCOOF*T z-`T!Oy_Nb*JYtuMhD=p@F_A+I%!lISl&w6x=nL6sOd6{+pY9Hml?DmtQI%W$*gqGy zL)&=ijD>yAYwR#$iTg|W#8+40??*R$z~6z)`OUSF(!&h(;&-i$8B$Yaqg_IYPO&E^ z&=hS47aw=jSDwf}&$GXWt(I=~RqqZ|65G9ak>54`VPq8z>~pF2A7XcPH*9z^+(+!Y z!zUFsrdSUSbTB2u1J4D#zcdGK1g9p3?Dnix32P-?3;T=YMYM&c%pEa={4gMXe$E5e z;tBQ_9T`N0aW8?{9gSFxutZ?xO$G&)eeA@$CsZ+CldlDft9%iOSR|0Nx-HgdGk@ z0CHKxwc@-VmJWkU8z24pad^lC8|}w*o#|jyD~N2Hp;sPo zL_~rlDO7q@&>_3%*|sRyAl`d)+^SEo>#k{Fr<)P?fND|n=N`KT3dyR`xm{3#NO&?f zk&@tP^K4WmY`>g-N*ZAk3B(r~st*|A#+m^;+ARmY%8;(|Tn+4d^SU*rdHE6Da%EQL zA0K+ADu;!?zRfi7WkKtZX6{cj_Zik4`}{qihxC$k!RlUEtkON?`$P#u*B^&f2j$1c zcB*hj5y$}tnL1B=`Au{opie*FC@8p~Mj&ktj$tQ^0DVehN&*5?6|h~eqgrxP)nDwg zU5t`lJn3qEQKca|#?8Wu7h;ugl&aRO>6AU=wiDI>F~@51MgoyG5osrR^6X0 z>Aoei-LaG1@vXs|^=x?cM{CAU7_6yU;M` zS$@0(SunGT!`7DA8`9sg2P|l_*0~NELfg9^yo9}R9$Y1WNISfE4e(Jo0)dp&km|UDleYl-v&?qmlyDjxCXnwo1ulOD4^(5Z z^ln?yg*1?^^xX00k%!uqWYugc9mb{{;_$IIANW6OoJg&1TL+I$UxAi9$5vAv9@&PI z<1Fy&4hWeq`1VWItDk=@TiZeiln5DqoEl)rDC}Bu&nu55U+7xVL&QEAkO*jM=BA5e zS_uOkeec3oAQs@{aoWrZ>KqBTVXHq6Y;G+rEBeB1Jpsf| z{>swFhOw@6@%iq5l+kA(9R8eEoJo{MI_=M`OE{m|->+vtqC!p8ZNloqQk^V)hM5yx z$d+||awMzm-X?G3%ap=5+U`T)_VG7Ce<_d+V-DxayB z>_9t%Kme>os70#IvQroq#^P0)6rqjc3kX$r(=bpIR3=FdpaWNHs;UfYlJ={}W;YED zkcx0Iqdydq7!%Sy$A)BG9&k=W5XIEZf9!3QoV*t|l7393PTHuI{%?F-O2WxSv9^g{ Vp2u1$=>Q<0qhY9CdDrpD{{n*maYX#cdBveHyp(_wdq$5=zfbta(fzZ21la5jXNQv+T z1*Am*>7fY$0vLL~@%P?&=g!=jJNM6b&Yn4Y_H23fJY}E#JRAMMP>1OP_XPj|nBXui zV*mg}P;L%78cGR@DPfB8xb1u2-1m_O(%1iqw;>O80?TpqPO4;SqUXtD%siypZ_~s9=oW58SC1aNW+{w zjkT3+AagMFIKehD{ccN!GDB!E9zussm^E0{)xb+VAs`Lo>OyRhb@vxo%%Xeb=Adz4 zfyVyR$N5!P4|y=3ORTH3;S<_;uuEF!{IbRaPbik!lL_qsPN%XH1D1h+tNq^fJeVAL z{-ex7H`cv1d!Q|i(9q94uVGhB1HjE-t7$vgg=^4hY5*E99B5DMd8l_H4BZrN3SVr3 z+FsecHkx$8*5fto!y+L_V_rlK*Zy!&+uGL@^H*o0ina|7=KVes)Jb1?cq>MJ+dK&mT5wUMJsPS zvS=`T(H(luwM%1`bTaoY1Z%l-48^Fw?rn}(bb|Xm-tfUJ-(LzzSXAadY?0*QrCYZF zP9GL+Gh-&@x77aue$nE10Vlv5LU0B?1L;hJ4>8Mayg0d+9DhCn9Em4sUu0n39zYs! za06WmNJBj3d(z_e;!Koyac#gJ#8c~WXhJpvK^8dUL=%BC2q7ls{4;z0d@dEi8&DT* zGU;0I0i8f_4nXVf1r2Z1%Tw>L9e-N%ISaw?Wym%MARrb9+mUesrPJG)0kp#FvtQm) zd79Uq@&PAQIL-ocC>72KZW!NiDg|sN$g$zh0Z*MB?q#Y*W}F}3r@eB;blVL`dQax1 zBY=Uo)d4TUfPKLiqqg9!)?PGf{5M4&Fr(mrO8}^m4kwN=OPb0bXb&;K2f_OQNovoW zLRx4gcuN8ZW(hi5hPDb4U%vDX(TMErYcYK(> zaux#8Purn}%V;+p!~(omR6`+X8F-KdzJQMX&~6s91FItyfKH6@G6m!l8Vv#Ty@F+T z(2cCc0JxD>9v|?GYKX~B2R^9R?6sv0dmG*Rytt4F!%wvp7qd?VNIookgJ70Oin zOaDwI27Q{}k7Hk2quCN*pR@#{>7;xUPPCF!5uB=O^!Z(uWEgl-%?R)Ca`e>a8ZE{h zP-YHl&(mxoPk9ZknN6B5g-io8iYvP_5IgO<^VxXTgIvKwhE&?h!Phg}wlHbo!!P2K z88mj19ZqJSsD?vhTI%9O+}?y%sxV7-^3pv6;9<0$*Y9dLzaqW=sRidme=S$pWCIvP z45}Ip5`6EcEI0HD18OSHSqmbkx4BuA!1<1C){ir)?|qQdXW-YY{WQ zvf=H*Ks8snxA=^_8GFXx6qAG)CG+6k${1ykl4E+BEh8ZZ$`WmZl8E|plmp+;@OHnP zU<+(9-Yuhp&YB&qgH+;Jtr{4`{lgyTzJiKrm)3^GH@Ms;XxQ%uta@&2 zJ6%WPpQWKdC)`hNgixcGht0*Em27oZTzrKfU9Eu1>!H;)uB)dIfI^V-&Myx(nj(}T zF=dCYgRtww8Zky;QS=tyrpAXx0UcNvA01(F#diPx!FnFZQy%@pHCx$J9PI>B0SMYJ zzijJiAIwMr>nWh1w+esy8R_H@zS zZ=SS1y??+1{O;GQwxZC_A1>nbCBmEa+IZv5DGy|+mtf7J%E{KZ4aHh5C029q=4Nfb z_6(7Za2%%A^XiQ!9k^preY*WDgexzdpjZyj^fYiKRN7)O{Aq1+;AnyfYDUHE z1bHRNJRZHsg~2AZh?cr#W{huP^qF#K2pbTr6B5Y;gR&+}UBNv8qz#`2Bax$xg+Y*6 zh$jtNsyaNu^z60&%pSl%3Z=G0--X*l}0Ru*>Qj99Vx6zTcLNv zoB;$DxciUq-&cL&aTHPGrThN(3KzNQ%) zyR#Q|3*B4$G3;|5K)Rw#tM|Cxl<Q|nhWNhUge2F${B$t)`(zpDCba3z;%zUFAK#7T8oeuu$|ag^b%}_z z=0LmYY=A8!d@zBRfuN0P$J9Z?-Leq)GLyHF2I+vLRl0uQ{Nr-7AIR36U%`+IxzT65 zJkzUkm%dk2j+8^-?>_UG*Ig3;chv{UyNyo59p3zv5r4p(o}h?vFQw=>^cc2RX;Cj8 z{WJr?>uM4)J&jA1!rfz85~c(4&1C;d;6x2J2_~5EQ^rLGPzL{sUtAX-L_kY*j0bm! zKa#GcW^`Q@OR>NAyPu&oYh$Bfg9HuY4==2TR?}|z0eVSO4H+VZkLjh&S@Bxd)>rB3 z5KF&m0=Fm5TDvIoH>9n(et~KM#LGx1_wpF~Zk!z1`w@m&VfzkvUg^0>SNcH!-YgnU z%FIB3pRzzO8&oy^Q{T*JGhDN>y*_Rr&m! zPKh<$5n(v`=_3c1EYe{=9t{phOl4jnNZ`GhOl`x?zikMBXsK$IhoduL@yvn8580oac<`MqOg`NB-FEg1aHLNs|;`LllEb)`OjNq zNr(l{6~E!zn0C+>J=|aP0e|humPvDkfgE1=;4<0+bE4LB#f$TsWKzrJI?Au6sHUyR zsRSAYoMbK{hU@?7kI5gejqJZwm4A~81My@*pVJZ^1N(_8q5}~bj@=Q|{ds)PQlE6v zP+=570gSkF>Y7XF1%MY~r;dr8m&`_p23&VWY4a4a^AOp-)7B&vMv8t;}6dq@?5HOtb#aPo>**P}QNb2=k6<6&cUTu+ZU0Jj^eU0A<#w>hF@1wl*X^Bh z%L~~{j2oKo!oQ?9R#NG3>v-wej+kivOv|5DWa;m}U7fRQANFT{ORZlJRUfb z>%B8JCj2dFX6Adb}SdC}?Bd$~?KEC{3?l+9Y2ISA?y?gxChYy_8jmf)ojHIlu z1FqN^3$cNpj)~o0pH9RxLsF%Uf~BIAN(dY-FC>dkDO|1>88+Up8(V+9z5Ukbf_vh? z_G^^-TR+dM16j1OvVx@`rB6o$&hZJ7c?d$llh06s)?xCpiup6Cw4Z=gBF)h!$+*EU zNjw(4oYmM-Jy)O#Hhmyp;r zaJ*`*kip?I@JCuI%;$ub+(mCpA1s$+U!jtE%2+0~P<`|7=U3@YTU~1Ayw=MBQHAZ8 zY&wDhu$fxdGI}0R^GkAB*B+8W%%{IJ_3K0Nmyi_V7mDu=J3{Ro(*XX$R>be$V|rBw zxOIl7?bUo6Fl)4ZH^vg4E29*vdNZ!J6s0KA_2u&Mo9#I%1`^Vg>^R8 zs#0y$giUJg#hbS_*=dC24yT8xS2K9GDJ7SI>|#@a-HVsJwlCi{%I|M(;Vp3Cb@DPG zy&%I0W#wq)%?h=sN91L7@Y?7r8_i2kjhGN$x~88ifbq%$u0`(Jx4J*MT=M+Nw{&^1 zUokE1j9;09OOZL?Pgnl=*fdki98txlSyPAGtG77QOu|nB0>valI(;=m@=>%Med$w1+JvyH|&)){K}+P5R0%vlY%9 zRr~ch#0zeW22Itn;PD${9sLo0YuAhaWSC7ND=o`|3gYVajFb(U$+|9G7hL*MN1tJ7 zEJjXb#Bh7+h3pm~a_-|t`|#F%{`BAD zj%izG#y|C4zlFbJgY*7!G3_2$a58J{6^Wd+2FCB?ihmU^)~B!aRB#hlw>f&-x?2gB zIPIM}dvk^I4B2yWac?(K(PqUKHaEKzJ-{34)h89I7X&qJRP?SkKB(OMlYVn)_Uqe} zy~%1nclAye!+w(0e>T}l@lh0mBnxvnflTSgqcF|zXi=TY$uom4cTi;>AcG5Py7NseZJj$ba22vE3@9x59&l==Ba_iQr zcELE9TKgNt^h|mPzi^EpF>sc2ty1$rm_mJ})onSfEw@wochX=6c*Mp?|jb#Kh`FCN;KN_Iqr0kTT1#fD8Mfa zdyq+cubbuHyPkq?zZM(d-LUZ4!-1ixP^Xf*%{Kg3HBoD?S388rN>x4VOhwa|y!{?A zr@d_tQC1~+Ib#<;7TFnjdDO2m6YwJyicL4{Z#h}Iw@iu`5U@q8T5}#WEl>Q`>Z`D= zRkG!7V!dz0`R(&K&0x41MTqK+?qMIe@2sd~yN zv;8h&YOG|XqY-f(9LZ>FyeY^!>7OkMpMoYH) zycPP#s@Nl0x7yU-fP|T@nXOb#V6pFzCVtd1tuJq6^vt;)(OI6MLHIInx{Y>6IX-dx zNg0ad{pQ&W;|a+FTwN4=eoA4epwLF~!5qO=D#qeFj z&neR%Bz*^CZ)w-Kns`|p--eV#-4fl1ZIQE#cBmIB6P0jDrgZE=J8zIxIO&ll-KcTr zH@uUbdPtC6pCx~PADJ;+i7 z>!mTbygmxqfAzKd4Smz;LcEMj_g;*dT-v07%G=t%n>NoI&X<@Wq~h}OAcm%LiSaR* zZmH093zQF3YF*pnSSTWu;n4+BOY3^OVY{raZiFcFzzYFU|KEE7DA%TSs0km=8n z2w9TA3!Gn5-PUjD%^JQ`O}>yT?;zt+az#^N=yyM!&;N1tV?AtqsN9Rx^8+0}B>5n{ zTzanCR5lz}n&0wAGHdV8&qw!sjt5m99Umgj7*$cu?=Iu3+wNwh+_ud!I6Ym;P$}v? zp9$`?FPaJ^N=}=UqX$Dr?UhK9uV^w#;w=!${dB#xL6xV(ANO9_W_FDK06%Np#huTF z24ia`5y}R`+En4_MG=6Rw#FF0-iS2|$o0=_IXj~$xwz$V@CN-00+=$$=Z!lj`niGI zR)4Uy3aA8;Vd2KhIDgER4Ln7-|j%~IjB1K(HM1}Zm&H$fn6p0MtOgh|E8@x?F>}} zWI>XAzi)4Yc0em@tE;Z0ma@8ls?g)W)$2sg_5`(K1X0Q(QPfb=udNz^`bpr@ltwZgu1~M8ZPKR72&7NS)bA{P{c<^fHWGKDt8MvvA!?P5IKyf;L*-{Hm+&LhuR-Vd)K0KPc+smnc3Rv1R*H+T>u6jCUc~?}ei^4fn+^Mo4aM zMpa2(YUnA7SW#PuZ8gZ0D?PcbpJD7y$_p&6>ysviFvs!JUO5oL>MP5=B#VNXqMy=G z^`wLtul|*%Le>3bwUZ}Ayk1f8MvoyX49fmASA0-~y527_Jj7;7&S&D%Lc%IR@L)n!%|f`n%Eb$)6?;lqhRqE=`%AkLw3eobUPZagk$R zVXRE5p^|2%eECUOVcl1L{K+AHzg_0^baT9i6tC6lYe>_)R@$SNDZzYORhsbg0cgaP z%;zK%P&%QGIS>S`xaX6$sI7b z6%2n%8M@FNF32!i$uKr}x}rnNpRLijIRD59w4D8DfW(Wr+^n&ATSJ$ftl7|A1lSG+pX-e))T{^dM`)wQkF0x4rueQ#w0reU(` zWPh;-KRGL%EtK%R{3ISu_@ys+?Kjt#qunPnq1UnP4)EJp_0SheFY?0F zDkdp_=T<&Tml;Q28Zx!jc?3`EBArigRFBeimNz#NsMO)UljKCill5~Ewdnm8F~*fD z8>Z~A(DaV+X_BTIPZQG68#R)Fcggi4QCeBl*sVmr{|m*{8n?e`kYHP`Ed4pu{Cut# zv7F0ORVeSM+Z21p1>^heDt_EaZ_Gyd26j`rvcA{(2N1|5{0%3Sh2N-+ntI1Ap5I)Qp)q7*58+W@ z>?c(tF7ao{xvaFD_9XGlI&-8I@1_)2@ExkegiqML5&RrZ7_|&P-r&iclrib8MKW1= zgzZmfS%%BTMp_{xlQQD0sJ`3`3%4UX>^}F4GEod4nwSkc_bzHbKNCig6oq0dHzoIs zEAC(;q#XMp5znNbo_8?drkZf~IbYdOwk){%F)F7#CSGsYS|q^QRu{|hYE`=X-hsR3b^Z!7c1;KHL+xIpN#H%05XO7*{(^LqC1lmLeK|lS{>q#;t*_Lcc zg2`i)y+CYhvP!99doWbRe`dn!%;*}*-^N2-t+EUk(xZ>^iQb`uwdE|TLN~<@iO>cY zg$mQSFOpVV#=0pl_~=o>e<2EcAe)822OvR8F^d%xw3t%IgkU&_$kG#%uMOk9 z_SKLZoH#P1!OhfCLgwkl<=py@P^vG6fVj-R$JfxOwCd4|FM#3aSN~A;6wxPF^67(X z?vU;}Xq+5B>AFBzDd=<~M3t=5t;=z6Vd(?@co9C1;4;nojPv57Z*9hY3cEs5MfOeR5F40_y+rLjZeegTvAth;h|Mb~N;*tzy#7VP3mX-^A^n?UV_2e z{wak0_FJ_RK8{#q8~*t5&EgU~?dEl+7dp)_8zb`GO0c&x_mVzUj~=Y5;&~N&q{sc^ zRcTDMz}|hG;J7}dJ6&Y=IZz{k_5wB_H&8=9YE*>J!<#x(2T!;i`@gjvm;smG&l*XS z+g|4?vf-G3RA2@EdFnbpXEl|lGMeec1_OLdPrNd% z1u|##M#Xip5kUKZUa;e~;T+OW;?`p?fwEi;zSxQGNzBxo&4)X0k+g5B^W-cuFrl{+>j|l9=*s9Ac25kO&qy9 z9QTZCcqAu?t?Fb)|E&yQPramJXG+^c4}{VXeBl%*RCK$-SI=?Of?xRWVV?c}uVB_j bJ7_-kr}-70%|Oa2B>;ZkP^;>$UBv$aKr>P$ diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png b/doc/LectureNotes/_build/jupyter_execute/chapter1_61_10.png index 16b5dca787afa007f7e6e74f4863fb236c504308..2c196f8b31df549f7660e282add6c7001c4cae72 100644 GIT binary patch literal 25394 zcmbTe1zc2J+cgeKr=XOSLAQW(h$u>@w1fzVbT@;5qNEZ6(o!Pb%`hM!jdX{!bPh1U ze0$LQe$@B*-hcf3jc^WU&e><@y4G6PexsuFfDrF09vT`Nq3pwZYG`N|w86&&2Me6( z9k(q8KZKp`YdfjinmM@|JD8#=8avrp**aNSJh|p#>fmT$Ys1F{{&8G;=Hz7OD8kKc z{qF@_whrdpUt=l~z)f)N9_l!vq2XUfebCW9Bv7KEJxP|mC#B(*us-SPcIDA@;I^mW z%NvgvH1NgP1nz091sSE`ye3n{Si>y*XvCf3$KqG#i>ZoEb_r8eC3mE^b;{rh*^2gE z<>H!6`m`u)?P9ay>$7Ke3)Lf|qn*NhdOiqobEbFqZoIn_!62LZ=DpKJ2HA-AtN0wU z5$SJrjASDiUUpuPk6>VZO0EsAxL(o#{>RfXlNG!(7O}rN93C!ljN==}XRcOYAl@Fs zP2BKFHBA!Fwu?58x;t8anFPKCSu8_W2F`v zpH!}wPP5BqJpQt3BpV<2o>Pw(Tzys0M!!#yHR@=0sXtX3`vbp4<7^;(cXn2iZl7XA zdeU;l)}&p zhE4BvCrc@Dd9p^ZmDZKUdaQKz6h4ieqMR&2KrKMv+Y(Cc*dq?_%~ogDEw^f$3zsCi z!$Ce*K5jGjDJ|^#;avDq3$<;!d<+-0)%5bWG~um=iz_2VHNJ+fz=SmMz4V12wFJYcNCR@uOS?(}$nE3 zsm~+}zd&!+07o^4%pZEY4Pog+V)9h%r1|$x43q?c;s9R;dWJ1sV+P-dkWQ0u@s?bjl0(o@p7o`-d4y-{RdCg8RI{Vitd~=RXHCv*N_>BP6K%vBKH1jM zVRE#yPS|KFEnGM^?jnF~Y0-DWFp+B#K1iUS`Tw zDq+YL628Y0O}bgTO9Y1BLgidonoQ``TJWl0o)oE6e5-YLH1OKK51zup!ZKoXyx*@Z zxec0z*JVxH0CMV#`b@bm(H=~?8Ya1Y$Gs|G^^k#dcG7hkb0(*t{-fn^KJ{a|@YYd} z)x4wc!Q5zW>t(|NWA(1n-ZA8yy1iA7=3tWO1Vce{)Bhr|*xMqLR?_fR(5`7(1qL}i zy=xOyd>65a(q$u=20lMxAMj?9rGDba7)?s?*og%EVIGW2=9@SD>)=!IhCWA(1J*Q< zlkJneeq|ch344uP?RUrfs|Bt{FU^zOU)>BBi&zI!LE?hA$KFisZueYq2iF{G{u$D< zHJxdA@zm7R9?TwZ6BFycEr6cMDLC!L_vo|6Y>Ge8N3*UE)K-RI@~QkD#QtnJpB6O7 zTFqvC?eUMCFLU-Dd&^SDv(uH{)Qd&hAp+K8xk@9trO#kx!v^x;OLr#?BT8=>WJ>it zN7XhMdYRIJM0e~G^Fgou2Id1&8=Y@nJ8jJRu9MESdwue&gz9o?_Js`*rDG>YyDR-! z#NazjM`G%fz@s`;y*PvN6lxTZz9tn61kPAH6y)C z#>R%Ls;X+0vp_8a4^+bFjLDSerZ|z%4BF>MIam!~a%*BQsJ$CbCmvLiEgOA&!LoS* zi*XV|mhqa>4tqn+T9C)f>udu@7X9omStM+iO08Gl+!7Z&q+mhSRZkN5H5L<}JlFBS zUP*~48FJ=cR93_z=ly~aEMH;yk8AosZ`glt%Taehz)fl)Lc(MOwTxK{L=a-yDCFsu z9fIjxYucdD2ptSWV?j>F#ayt;WxHKPo>F7eI$Z*@_Y!;>V%{7?hH4}lml5OZn=c1q zQLm?#2WCmsz)i4kp_*S>a{T^=xp*nb^Zm1Uxl`xDgn2IOxPHEv`zR#Cvpq5i261j-<^zTtWVELrRexTG>crSxRH!8}JPTl;sV~t!VL%WoF zsON&E7z4Zm3URpSClm_7X*DGW<;35Q4W98k zGo7#E=iin+%1Zjtb$H+<$*C>b`PpGU@5@DnEXe?I%q5^O< zmK-dgO;skuKs}P{F}X49Zmbwn@|snpg%OkF>4i7<0$f_ZEKgNgC60W^McyudMNV7% zsjBNkf$(F|W(!Y^(NS#IsbrXC|3{5%nmzl#by9-b>n*;iRGo)bT^PMDgKt==--N}^ zVloa6JlSpvC$XWAfc2TaC3$Jd9bTlDW@MP)VZ)T8dH-uCflY-ln^J|6>C(tWF~nW0 zv=%cJF6ygR$tgXIM;dt-*tKBmuI4-Y6{_EHM@)u=yt!CYys9E>>?;i*qV4|Srlg^F zqT6xvS06W&?~`)q;0Y3UJ{jUB`HuU{RUyPC(S&hwLdANmDmj_Bg1Bp@zsDcm><|?H#wN|(8eKxR^Q1FSzH$Q0q z{=VLH^`cp`T;)uVf=y+yuQc()trk&|{Ee8lN|K{oZFJ}v`E2O+(bMQ}u2JCwJv5BuOJ8mW{-hZj_&1CzxQrX(hS~#QUhNP%KJ3U0kZKKHR zyS-3}7E^&zaNqIk;{AoYNTsnUBJ=Kqw)wWmI6+&wZ_lC<5@?9T_f)`?o{-NJ@MAkG z&FV3&k`;aJ+x#`En+I_7AC1j#lZm6t*{E%KsCS^J$W1L(-j8+%jBzK?_zwia9S>Hv zd3P1LxQXjeuQx`f@m4zXSm_bxSAOypDI0pcxhNHv;NuYT*o956)#vyD$&X=0yJd7K zD3PR)RGo+v`u0Kv37e*Gk~gnTDdoX5vs91c6H_+W(P9k_V)JsArji}vwe;MC!vpf8 zqe9n6A!keG*JrdF0%NS*;U(=uYr$>=4;%VSVe4e<*F;4RA)85Y}T@`;I@a}tqI;>UxW;L5+3`J`KARd;|i_KU7At;G+a5sm;8^f?eML2T4qw3~Du zzU6he<-~to1AIX_tUJXu{PwiUO_Hyj0XYa$yC-mtgGu&oqVf0o)WQ9)j_HG8=!ljo zh!J|zt1DYm4^)%fQ`fw{a_AFu9Vqp#CR~sa*S#}0`CZ>Zk?Bp@l>GG3wZ;=n5)pdC zgZjENG0{c=y{6EMe9K-oOiPAIRmalKqYlr8_`6f$@hFds(09r2Q6Ae-W#=Nx5&8BK z$NcRH?y&XkNZzkK;alchG$S0{2jf`E0m8d8D@(k>#qOhWJnWhQIoustp5azu5~(y` z#PzLC%<7afy&;5Rc4yTe5u`5G1*tWe=%5Or zi;-fuKs)xu&B%?(@?x;*>!u5f~_>lZo z=Pp%9d@QUae33+leMGySCcpA}a7ztiWs2u^-9eg#m}E|2&*{m9%A+s>(mPOo1(HuH zOIc5*TgD_1jNWggBG-Ks$IFA%bzkA|@$s>;Xd|1;9Twz@i;Mfd1whu%W~mfB1>LZP zPi8PElBnhe1wZPKI!;pbdeX~`6dC&^So_&}=IOS?9c!*A+u@BPV+=xzcOj%}C&?)W z$_&>^q5>%`;dMTbpSeFvFbOxN@1KKrExezOx)J40A`Q(R2<>bMjSu06ZMU^HSMJs;oSop;A4#+;5zY;8NN;+- z^xi;5QyZ^7$1R^0cvw7E5yWTojLL0?{fY%^VYejG3q+SfL}FWao)6{XHZ;D(cSoMW zs{kD8N7YIrpNYQP;)n4n`I(sM8~jJfZ@R&J#Jg`@c~DGwbOc+!c))pTTl#Vxi9Eex z+#avGQCsDEa>32T3d>sVh$_Gmev0%7>8;aZQlxIp*20zv4=?VAkmfASpTL80Y6RuB zB(Ga%m%AV-IEj=d4v%YaYHrJ6m@CSp7+4=3LP_X+c-srFV?H%)e0@tb%tog=T)Y9( zoZJpJeVV*4AK=K{wphiiscOOQHs7WYW5;E?dE7ez$SG zksBJ?3tv=V606(Tdayyd4AH_u!qo#FG912ll!A}^cvE27+;R+k!Z;zp;Llwj7Ga3? zfR-+6X|%}s6|r1RjA1D6$qLK1TR_4GT$@^lE(UmOa_G)3HwjpstOyD7tXs@Uuq0!P zJp6W}#h>s`@=1vixk^UC#{1K*eDOQd_V#EVq81_0s}96T_P$PeLxI7lpv6AFTB!1h zbYOq6PBH)8ZrvCbi#{QxY5De{t;0`6FjKbT;*{qL7+R0@8z5I1w2Lv0r=y!Yq0 zCu0(e^sac{7Zhe=eTo7@tgJmZ6n-Hr4gk?2?k-+RMd7ITJmIp@o(lh7mtE; zB3yn#HTSqz1Aiki$MF9^LNZq(&2|*(i@0ls0*?_vajof{;!OX6mF6Uw-M=40f_a=b zZUHd+56g4?Ebsql`c9{2vr4=j}(<{8?8o zGvw?0)6=iE18T>Itx4)jbde|yDO(wZQn!QLX>+5hT7N6L=(O-J?D&&yijxjN)g5zH z)*sJ-tO@^m160ibEPZD2`A@h?lRNOojXnQ<(h(@ld42s0z6;fUmhK!#%e8}n@(XF7 z176heuWJ7t?LHwQ$^5;tH6}Xv)6cQ`Kkfck9chTo(&TEK7n|4m>)Y!SfOpEVU0Pgp z*r;0St9G_j5?$vf=hDZy1302OPk5&D>X6gg$UDHZU3mMDfs)4v4TLZ#)~FNEBa!2! zC{|9k`E}+xvx5A4j#)t7m>q7-u5v&Dj2Hz0a~!_~DRZI|DG2kijlR4l0i@dt5Ek3- z4{CuxhaQbdIjOTVjxSEg{sw%~mEQY!FEu6Q+DbVmXrex|n}UKdqb24A@W}Z1_?^uU zBF=20YlRmlYg`S@j{rG>3TV+<={z}X>-GUSI#P%i2n9UX#ok;UzO%zQ;>FImrg+< z{^)Li>fFtHHwvvES7y5=TpjdB5LwVOW-Wej?pceP8A&5$?C*WD(}}Mt1Er*yvtDJa zoDHJ9{aKm=tAa>p_eAn3QYgmo60=%MX0Fv*9XDNI4Sm?o0Ydc#$wylaVbF!zWrxs(2pM)kwo8>>fY-Ub$EJX1d8-QQfe-B^&=AR_)nYmYk zTBUrQUtyz@b7bqj6UMj6XcLm5gp z?~u9!ZFtjMT#@S8AzveYIN7Ce=?ViXg(ULSE)vsKX{?#0QM;v!R9}WYh*$dAi7@a% zWH9Mr+6T2AWzQEFfOF1{K=x_HB-k49o4#x0W4KPx&n6j@*wo&DDMLYQv(;Ew3{Tw& zFM#?p%iO2p%k^!K_Iol{KdYJ#73&5;LhQa}z)~Q^CFKSx3e&Bzc#<%Q+BF);$f4)- z$QzRSGl|Aef%sp$U)9is@GhSb*wl$=-sR+^qxM+mHbgIh_*=~3Yf@;lRsS2IYh|mM zriG*WXT_w;xJUfxNSb}@hPalC;}@e{_q^<)?sMf0*_$E%wk;;~Fom${6x zc7?K=QLiH2+pW3UVqAXE($*gED97mR?k7X$%1uzwUtiy?2u?!<6}Oe`3}WuZ$!)n< zVqBIX&)q=!-6#1mh&yFxvOQDvE{6Dx&^*Pxuiad;Sb-d)p3`B3GsU|mIfz=%G>o1w zcD?P%76XK#4|VEY%;j&){x5@X(iAEu7%Y8ZHO$AL)>HUyk)W!PAs3zAJn#y|{|@Ep zQFxq!NC@y-_twYCSXo$%R{AnDax_CxC%N>$0+xmt1#Q~e+EM`tk;B|rP$xzOrWYgF zlM&lZjJobOw-&!$IGEssH4i`CQ8tc`#&wmvlA*Yy@LWT=5T0bZO3l&0>1t-DRwE=T zIAJ$Eb#|?>_Gb*x7xL75nl#K?lS#pwS~uy6Y0DCcNICAY_9g#|aK)`HTKgo*t4ovN zH$uEpt-KrV_9BO;!|VoRfwV0FxR+lyKRG$s?dkk*I~cIX z%#$*@Nw|)SowCR3x&s9l;l!`M8Hv8Wmom<3?~`~Dns=MPv(4Rp%01h+x-4~ihIfyQ z&do(XDuh52QfTv4-Q;7!ps;3UA3lrzbjMB+vXvyTv4*>)s!-3*vEVgnc$4fpslK

    x+!v>7Mzk}Ggt0^po}i}b6A*3=zmgsrZg&YfPBq(!vY~2 z2w3x26}rFSlNqhDri16C(xC=WBqZE47_YLP;QJ zrwH|WucP$w*9-#ug_OlE>tnYphd3M@9EPY{L+$|_BC>CY+<4?jOSG_t4nt^Yx0{EJt?kGp(4{<5{uEs3C>^+g)!Fh?s@g5>m88H@7i38> zX|6Wab-s%`_(nqoh3Xe8ZSSB9k2h>W8jG?_c^9fKuxYZ4x&)>cdLylE+Gr3V54Y>z zP4jUyQg33~x*$g-q_XBn*GkOo3^}koX{i?0?d_*LB=4rtUgkemwws=};5lw)Y1&|$ zA|BQL=mtTKwL)i!aFtdr873;UJSpr^jZ>i3y8{pATk*6HIsJeI4^6%@TPBz`;pw@O zAem)^$w8JgVihgo$d@)kmT12`^O}6oxx8X zoP9VBtaI1DHqTK4dYEW40(3NAc@ozS|^XB?ixEa9Jh6xJ=};3 zp(v(MsBlF~<*+_HQClF~yQhS7PmKL=CU%=@uU@!)H@qd=fY88$voVY? zMfcI+EVJ8s9*gTV4Kwsw?B4Xj*FMR6r3c%|+wNO1VKDkXYx&1y<*p#}MVVtyWCBh< zFae^;V)Oo}V@2^G^692G@|3qRW7v#m=9n^Y&F~mO_!6=X-L?q`Ob+4kdKi+c;O3wZ zWup83Y`DJh-TMnUY8i5SKXjMaYhZ!u!;db`s?8>BJ=W;oG{MkCYXJT`Ar% z3cS2V$htYFP5tsPNHM$~K_47}_DU=P_BPOee|1pS&K261I78MVyZ*g1>GDF9Bk!B$ z6rFyIrDWCeuv5tsbzxKhWNgp2as94^7+Qk>%)9OX9p3+BV3Y30ioj!06z$Jrdw~D{plEOXed0gg z_&q>qmyC`SLD9qMZ+~Ll+TLeBeuw)GM4Epb@7zuM2dMs}i~$t>BYto$RQsJ}jsh_3 zk7L|_?}uX4ukZ%qFakyy;N1KH3T>TtYSWmq$h-Vc(6}vjO`UImfP)us$A;+O+Ft?4 zj`Bi!_wRLsg22BYU;eX1rhd_XyrIIM1=485VE=DH0dmCg&ll$<{1dSQB>u-^|1QS= z!*TY#kLh|U$G_9>WUU|y;QgQcyL;-lf9u%tW~O2wj9A)0g8wt;D6g_Cv z$;!;xzpox!50SilO!zWG{`v=W!N){3f@8d4rdhQZ!QcV6IN`Qq~`m z14g=-UEh^8iaX}a=S1(m_XhNsNo-GJ>#(|P<{&54W6r*oT1H@$I;onyq zn45;+-1}(f+%w~y@>r14Y4R8W2K&Gqq{r^?gu1fAaBE@bl~iFznBCa+Y{hL|1Y!$f zLF!rO^?0m&f4aOP-CD1i)!KO(w=&^V)^*oahK`C@eeEcE-_9RLl|CHDkiCIU49fhbFYHUMWKwdR-)AZ#dGD zLfF6pJKG;wY_>qvV(Xy@5-TlkmV7i_x>}~X5z?^JF*@(7&ozi8^wB@uezDtwQ(t8M zv>{(Ky6%kGa|3#KHKA7kT<+n-F5bp@<@C^S55}AqLVF1S@-=nh}t6UuPwG1nHEReFG9+OE^9BxZXV3TYy&a zVOuSx3bT3Fy6$b`m13NZ_XRrTMz_|cuW*S{v=I3-)kuZs_%lD~7dL`K{mY>58%~w) z<}~$>y}ZE++BVPk^uX`BU{h?M-@Uo~!2YRFNCuC0@;L?phkWCopfl zMgXtqA&qlsDYiYgJBVzs2+Il2kmF)*JHkSXTi>nhD}#(Z}yEczJG1X@Yj-MpGQhUp#@7-G}W;SiSNPib_oVL&@^B{a^qJ8Y%A z>Xl8FCyLz0X)-?^Ko?*Gl{U~eqB@!lxk~}GuBqbiM z0$Dv*{eAeUq>M5^g%~|Ak^6(LQb^@h98My>ieBgigTV5nYwttvBN#$aNmSuorZh+;vu6mp)Yo&05M0Q+R2Q16% z?;c)ygCS}iNb53_8c6pMf;{uuf??;x*ed-P(1q)@p4?X{bMOly8{$CNFoh)7Zp@M5 ze(l6LL&BJ0!E2S8u=sVek*-j-=g4U{3W;N@$%Tvs#Y`UWo@|oZDuGeXO7Ab%rJirE zCd|0#NFE?#_hfkBi;y}ovhHfJ!&eZ4nD3UB_AEafCC#aY(y0b=NxWEGtq75)Y~t3$ z(yHM|QTj%|`!Hn1XB_Kr--3S8N1q-Bd7L2IU)p3f1cx`N!!eizL9gdj95!R&#MI2u z%v1F9x1hWzHTspQ&LyI#rW?Q4fJeDqYo&&*BY$fFnF>Gk@N|vxH zlBuvm&n}iZ&h&GqE;++SrOjRBEc>NZ2hMu*a6o+9SrgPev;D2N`o3<>RsmV&X5a>~ z<02+2Ax86422!t*NzyQJjJ7Y_&^*aD*PYTC_wD6tLI+K zH^Pi2XlI(WX5VF9r^9dgfp|=o@ZNjT6lU?m^1;&n^QOAf;Ee3%k1VIiB{j_cDg0Oc z2ex>_2DN72;fwh^-pFM>$~`m=L)=jvn2tJPt!jY=!Cn7?-R4D`asgt*Lg znNGj>%&~m`@YzSUc1L&UU9f3~rrGH%UdXHurZsCEp)8Y^Lrj&*L7Ojw-h%W9II^CacC8xpBb~ zzAcz(t!L=`a2iM!tj7k|Wg|31q9U?eBN(#Nqc)22BYxMvHFQL3xF7$*Hh*D{KNj+W zzPEn>h6MjV0K|MkK=iqy{tsmO>+k8L-7vX7uoyW0UHx`o(q{cbWk&%rHMKYQj(r;W zOy1Epz+uYsV^}V*6*MNfsjsN+;>ng{k3q00i{9|R7*bPOk zbREFJbsv!9{h5ks-iQM;+VcC z&!u%MY*1}7tXB`bc3hP+7l0?F1?6?o)FemwvQRccP#5YdDhr?a^tV4pHtBg;<^!+8ixClE_ZrVS6@FY2Ro4 zfRwtuPVU`dZK8ZNZUjTdqtXMo#iW6Mmd*H{-1VcCz53LXqyp&KhcH^v;<+%O zk%jsBt*Og@ZusTKuF+_;gOs3 z^f!2HefK!u|B>czXBd}g{)3_$ThZg;SjVv%$94lYE(K8l7YA9yzRRwn868ooN-| zRZI}lL&zswN6>o_7hbzvm?CG5c*r2gj#QYSbxRu3* zg@zu9(L**TmiO0@BeFa`yaY9TEIguyIOd6XdAb49L(kSXn~)i@{FO`Hz%G`R6q9w1 z-=zae%(|YOogATsQ1QJ29Uiz$zc^V-FmC!_c&?GwHlQi3qj4+ZPVt6yeSDeuSL{U^)4v>Tpc+R#!ExELl7Oe(#2{>aaEU(kSPJzU9_zEN$!TXpP8 z8B`1I$Igv&2DQ62zaX_Bw_*v2EiW}8e!05)3(N=M`42s{;w506Vor$6`X<;HkVB{wy`y9Uko?Mow?f}!N3!frO7G3f_ z6gueyn08#wN&BA1hBf_Msg8w(;k!MhyRDs}6amqg;Okdf`F^_y3=+rT$iC?qNZKc$ zO|MSBd1R|yh;uM{IbTTf^b|*06@K4Bec(*y#@je528zh(fSbfGu*Nf?^u{B7Vd7Wv zdys&Kck{X;hFtz&iY=(Oi=1ABMfSiHwXRr3NZ%!17_V9iAoP%8S*{E_^FH7=rW7zF?lsV4lgNn5pfb{l4mNE5cmi zZ^=cpNb@KgwB{>T+pMg0W6-#}_%n6k##^C0i^hlSf|MdGp+VK$a`A=7qFcIdwqhq` z7#om}ZsYp;znb+Y!oPDWvN{bPbcOYfTF|C8OPyo)G9+A_uCvL*lSaXl%Wc!sO2fXi z6ocUgh`nwK=okS{C9YPV&Q+!v`N&JImbCg`fAr1(8o;0`S= zDai)DV&Fn$T4>LU+ol{RXzl(nfcHx9>=uMXBQGK4$;;avQ%5YZT01>Yhq`HJ1CwWZ zpnkUwH_h^O1mEG*55|z+Ux#!H*v+R(J&#GmTn~HIQ!n>h&)xK*On-kVUGx`DXW-Wx zD++AfKcbJE!dav{Er`i#)QM>_U21fH98{3k40M-IJ7$nLUcUL^wjSRdBp8P~fSB(C>pf(62TIne6M$x0PoMBPKACuY z$bY=x)u`i4wWuH8t7|L?Nj~T(i-Kw80qID#6i-{x^dt=bHWW>lmMA7cy;^CIQx0iz z%gqi0gtD=FRpUx9u@GcplZx&YQ^P2k`zy8^q?OzGGx9PRLf=)EN4=Y*IWl1HUDi_V zywAQ7)qW@9%Co+`GU%tEKBVLsUU&NGao~(FWM;~4;r`o4 z1UD&HB!UZp?;02$Bn^t+^$WQGH$2zNw7x37 z1O8%axo=h0AOc{B%E0%{lxI!!U;_~C=$uz;i!;jjT|O7S{4ap`$!7k?0DHd9bEyWM z#cr52{zfkf177dfEq*4hIKBp+48{f1tAFsE3AleQ-GB3_zc1wg{r?RO+5rOjMVtPS zF(ZsJd;I?_!e5V6Lovd?p`v*}jQ(v|_?@}^`;Go{I~11w$8pmB7cKgC4V(M@89a#r zx5!UB@3|!d#1w~@KZZ-5#e$BX1f(VVv!kWmVM7S|TM%i0iAx^9Yhf@SYlG|qG!Sm= zZA``kakIb3!7FFj^L%Pn4S;g@$QeC*B!1 zy+XGGmV^=jo_PUjqa?Dz0HP$~2C(XVQc1lK$)xxlxE(S<4iB0_l2{yYNbrIr5)e2^ z9Ph12OG!!fS?itKLvFm;JDp!#@? zVzo7Ml{|32w}5007*PFdrgrGWJ$g&bd!l@_Kx}~;M!+<|41%6>pC^cOR7%Vg(Rlgz zvNf_vp&&Oyqyrte*V@25f(ns9*s;o~+=8kZ#+a589FXWD*BVAUtaqk-TzVb#?jR#U z&s?YJ6`?yQn#I5;5)gdR^++H%KrRj|__>A!dB9&iD^>T5U+zZH8PUkdF%skjFv!M- zvOW6Dz?mtK4oVNSXOBcv5E?vbk7A{9pMQsjk`d7_VxfUJtm)(Z;7^Zh1TZl%IUZL> zfy|b7;3UwXM|}|M=%^M3#RLf!mizY(%VL=4Xh75urOyhw40z9Yo9Z{*&=XP|wR}(^ z)V?5vhRRRSMYKcosnN$VvP}REabUFe-bbk{m>0y8mbW`V;^t_c+d`rV9ru!?78F@4B zVOz}=mz3{BKAx?XPxY$WgWn{GAEQ9KY3OU-BF%7NeDGzwr*!BK#(o291CVrp4n^Xw zS2un|7zG~lEK6yZ)7Vs6HLWgD+=)r*ExMi1N9S{l$+e7M5JFIAVNt{ghsbY;Ua3PO z-S#Xf&p1kLbUg5L3lHvFs0~O=pR#l&5?AcMymujg4SRez(7-&Sj5?L2v;Odh=(4kT z2sKtTJFEfW?`%{3@o+q5Lv}-vOLm$8-m}3)9XTQYlxPSn^#-KzoTaXke&Q~;1{!Et zF1|6D7`BwuJInnr!S|3U4awh8HqDcwQhxC(dJf2n`FiPZ2;wam^Hz7Dom}8|TU&cP z6T>Kh`|FM+Om=5fmqPFU2BPzL5KrJTu$b)jXrsZ;*V=O4kL}-j^ok5|$#}Zndsy!P z1ij}oRIaN9q6-G-ZSy?*(xK0|@-g?d%f*Te#7m{&c5C5zO!O(jJ0jxjCrw8K&m4c@Q z&2!WS^{JAg{2mud#V~#h>y)g`yI7tjClUgvO41XPL_Eq_h17GHPJMRT!|mH)<5w?! zJxgZ0?AGI;V||v;#vi*{)0mE1O27E#m5BCwp`G3t^F0iqB|Lx)UJduZ&lpL*p!L|9 z3?vLupc3%_34t=00U`vII74`LtF=oiQMb@;`@{Gyh16g>E}Nz{WKz}^W0!bHz=zjM z(6MX%-UVLyZz~x8wAC?kp<-)XpVE%rNeo}64kywe&=Q**QWGq#nv9>JO6l$6&X(XT ztnb2=Fr0BXraOUG!tIkp$R}awhzJmXyT5l$O4gv|8BFpmhOB|k~%lZc(4+HBt|6}P(Y%>@wE`tH1i-KvV4(kVpR>7Y^HbbGByT4GYXYWxvvM$R?KQooUYY9y6&7U>el03?VZq)_d5M`;53znJSc7Dnx_ z?4-Xj;JMB8cYyadCj3j`{u{ykMgPu|%KTRmSGZ05_Z+FeOOgGs_HQ}(H+JiMky!Xk zT2BBz)qhi=K+DJqW3&HX9`SptEHJ?TRg+Q6{}S-!{(r%JeTI3=(~j%U6k_q0(5T?&`@j$Fg_F2ZYJ3!`UnxX&YI)J>%xd zoR8E;7sUAJfSEF%Yhr@>xrQO6Zg|SilivT?mCqV}dieqCKQf?o_aDVBNfmn+7Y*Oh zdHb+ou%bzc(9Z`Qgn<3T`xjJChE~gHE+SvU>!f^pNjuCCA35=g&|y zPQq?GkS!j*VWgwaMOg3ZL?_^SS5M(Y2BVRv{dYFoAVqgR&MWnUwJU% zVvA7_5`bEz*aD=nVRXgXk`ixM>|wcbL>-n~O;)Q^>rn6K z=IW04+qgl%NClif!-+%Q+#HH{jwdB0eog{eGIpNfGAOyulD677U%@${7n@#WBkK5c|9&Gq#dZt%Q-dBy4uzgo_#QrKPH5E}dgc5YA+fkkelf3c`?%eq?AG>D zZ&BCrs?{Tq4?&<*heISvmqdpjlN@taPX5R^8Jl{Xxy|5W#-wx1oj{WzLmxZ7XNA^V zzsR&(fmtP{v#g!l1zOldl01iy8zc&37WQ^3ZtEx!oc28@F zD^giPw!&s|c_@$i=7X>a4*cA|RX9ZHR6iC-|O8{)BO3{<%!H$xWk~cQ5N|67H z>^%{p7NY@eyKCCn?j2JVp63j zOD6i+e^l{1`=2v*?}Zb}G4#RJWzrY&>OJ;?Gp+;~t3DK)zQL z%^&<3UG0X*JDdrTD`q`}Hj;cqS5L8@FR47=6S3^NH(^dc;*@;-sQze{xzgN?XX)tz z8>KM0ZUxTA-P9(d?*m44n87pt%KlZ~S?v_Q~Zcgx@q zDC^}>%FEoH+eki&?q$pev{J(VPaFi=O4=wBr|6p83o^`?yzx`&GJRhdblWh`OSE&D;PVS8Q9ABh20(TQd#> z?;i35IPFft;*}UKEIGXz;*>8ep;j3%jj-pB=@sIp&1 zLL$N0C*$go%uF!+eo0V?f^}|5#-OHXd+YzP@7KxPg4(& zaD^A5)JJTgr=K+Q#$+lie`ZL1>Hk9)Kh?5^f)H$FC=X;IWGcqXfQ?SjpAF?wfjuFx z0r5H8B{-c9=-_*4s#p1+-2r^C1G% z%3|9XK=j`9=yofk;!KryHG?gSN433j9y(iti=mVVFmxE*kD2A@FMgeocZvUxEo zZv<3gmkdZPi6B>}Mx_-2;k&N87_jqOz0$}{#yPoW#d{Y6A7iM(M|&*bqraRYTs+yGn)C@PWBdO9Ld1@iyZPQqy-$%vH9C!5t{Rtd z@kOEw#|l$a<#3<(D8^qA%l|<1s7yQ5!~Pkh|1qGP7oqzv9r|yQykI`*m*GVRG(6ZW z;yj}-3c|+xe+f6vQxdBxaWy}q0=s{VyN>^hn*CqGzwXt4iT0oOyN8O9{>z>a^R)`n zf5_y!yo9VJ4afBTnN2f+*SYDh-VX55lFG=#gBoyPUEJ_@mB2fjKE1L$aNVsFI~icz zyHe;;?AXJ5e&Vi*wn%|Yp3|uj5wK^&xxno|Z1jsF69!skemyPuDbco2!X(ug{bOEl zg`vR4{Ft@F?hk~+Q$&!(yUloZh{0Yv#5EmyUV_Y8vfPSRF1sdEmDH0D>)r^6oszAa zr`7hRzecDRdJ+}!I~*%53Y!hW+n)=dQ*gf2X%+42D<7%s`WFg4)AlL^<1kC`-1cFn zxw=W$NM8<#v#FZ5?oVSuPT7)&>2zeou;NBZ!h|90Kh}VUhig}h$?Yg9qcfT6=DUZz zAICNMMR=%v>CNYFS$}c(e(UwU)MhF&i%R7K=#(bg;^~#8X6#+v2n1&!B!(r>=s6@KJLvVd^&X%G+;?6Dezrj?<=BE+x;&QS5UOw5SiUjwv;5d z&_xfpyr*MiyJ6f++^U1{OaNZO?b54(PM^OXO34dUZUnn>WthVAgb+#3Y25C*GZ7f* zrh_$N&!M>EfzT%Gd4h*L8IxpT)WrQ;nq0wrKD}yVk9FUaY-Y>U$g|yYnNic^K`71KX3tW=eS1wV`K`5v zM+r{u0(BdC4R2zi=i@?Cn#1i+qsp5wr{ZMxS2Ht(>}r z->@#}2xTsLGAh?ScB+;ya!l?q=_)WjF#h4W@T0237mig2BjZB01?+CFYuJ00OSw0! zbF>*zOwvt|-dqQkKA?_V-~r^kP5qHs5@5-ltm{>#PFWsOxO}3YkSM!Y%Wrx*muOYT zYTLA(cb)I@XU$gvoW_kgZxb>5lL7%09Eumm{`^Sx=h###SNe;lrO%}Z20OFBJlHj* z%w!4;WLdiBL?PEt4)IHcZ@6_DV`JQJ9hPkP76|)_@LOsCsbPmbw_DFv>HX%HHsbq= z@zu&bZ@QT>fV{;VP0Z7Y-ra-AaHATJ-V@&$mkxiR+gn)wU^+dtkQiKjT4+*ojdsie z$q@$qyn{6y+G@PeH&ZR?&VG&0vdeojOt-r*7&-qf39_RP)ql@n06)HC8>-)}cQdA- zWUXe!ea%_nl8$CqTZ&Y_RlU1txEn}MfoTp*NIY~eALY78T6m-4ut0dPiTPG}{$f6Y0|6oqM)FFQ~~KKQlv}oF<6k6 zASg&Lf=I_8kWlY0tM|O_o_pSN|AaY{Oy+s!naQ_&BAg=RYO^=Lt%wRz-V9N`O^c`x zJC6FM^-yu)QoWgr`~z{hj4Eu!g_z;k2eF#u7L#RBkH62)oe|=pQLHU``f~DtGd`k!ktddZF8du6lpAWyO|eAJ@Sc-U0Q6J;H(IhNr&U`iA%B zd@64J;ZkEubX^O*Hk&6BI-HVuAb<1j^*jtjzmr&9e0@)IMXnLBW@HM(#U{%ArR+!f z8&^0vx8o(KF5ClCBtC+AVX@wEb!~WLYR;3bLOQG!4?5||m$Ni2*0?Q}nn8|I_Ct%Y z9Z3_BYklMW*1`K??~tFKH7Q%4UJDU^vV**8>|IuexfR>&E0VQPdv}1_9NS^zuH7t9 zAdG(85J_Em!wX#`{2`hyyQL!G!G#o$4y|^y*o=6je(u*aA+;IjXQD<&^KX1Q)yP$o zSKznsg2l;9LZf=}U zn;toisXb@AdnR!Yi#{e;P}7|jj4U}kS*A)~Q^;Y7`|7CT0yTNNbblxfX12JRq$vAx zOCWYQ5W#z%z3a=$3*;`G!|4ZBmbG&oOx!*=GIr_P%MjGmOL=E3&||n2 zvK?cZ#pYIu>YWcB+iMW4K@o;;)Vx4Z-ZnfRow;fMUW|E5uk^38K_;$(uQ`tEg(u{C zsb9IX4u+|;Wh<4mv^0qZm!lRd2;o2yMiCfy&<8m6h;XUp%%1C1f>ZOs(Uh0{Uup+A z#saxpwz#oPNB1fz&wSoM$;zF9!M5xtf;r?TmeGIMnDeOU!l(q>S;h4_i<>12Z+oma z6H)_^8ErmRE9V^dm-}KOuv}p$7p4|0-}GqJ`gyufKgHm_(Ys+;B#CpkwvmUzYR4kl zR>B1vF&}*TFm-x?(G*W1r}w&^$T*C8x|vMQSm@ zmj>qTM3X0G)bPZ;FKl+>#xA=_QRunTk zA|zWMa>$a1lY@hU#fC3KySfrG?GRpUP2bH|_j5ZfX2Ttq)6sN=NNwGhtqrj+^c}mK z()Z2Ev`cQyvWXc|ReM;p>-p|8%g?V&-`LE2`XOfZb*uo&LSRA1a=_2eLGr`yP}Jn- zoKPD;8f6EP$Bwd$j6GH1tda=aG1_7%_Tm~>R5-7gdYLK_bK-KD!vLdTdD|DvsvAFr zP_>y1t)pkvhRQZ>I)g`rF8n1@R9!bGm}Z@W`b&@!Ph5BBg-M*BRL!7w_jXSiYDQl`sKH+^AcwLt z6@U5eTlBljZzgXui#WPRE2HfD+1|1DWua`2=4uk%R9V$h%J{*< z8ycqt1^%@}X@w`w{qw=4gn$O)l!egNZ`qR=J{^?)`RgU@mhC+B!meCRbRMii1$~m= zaFGJ^m&O_*FH(rt6)Bq-80mjJ7Evmf*?MPbgPIos3Mxyb-)f@L`z|bh<6o=$jzeP| zQ}rRm!J1z#_}E}@QAiz!5O}OJm8?5Hyks1E^U|ZMlx=IhVPzC?Ntx=o^GRf+X+wr& zVv4D=@0uU4pGZ-4GRx-(>Dp($?L6<&-G#HgnxlP-8 zoz-@9_BE6fWK@!q18X1xp5Z(7fRRn|cKE^LBpRzr^Ey&ng=pR2vyv4f=ya{D(qge3 zparzx>V-Y04yErFV=~{DqLmva_^WX;N@y|)O3aGid7vN0lztO3uZNCuk|B1e5W5wA z8LE$q5%-Mj@Mr37ol;i3spU?sTjC&j_bjn{VuL3nueL7q2iN$J0uUe`6a34^lz3>` z0%yZ9af7myp$bN1Uj$#12e7)Lqi!q(^}nDr4fb?fBol0+QIa*Rt)r0l3VAFeaaoF7 zWlQvw;KG2qW8XmE&X359ARSq3OFqXa?e&>5qWjl96ge3>Fja@2;*&#(5@$*uZ=x!q z&2Yy=_x{d?nF_J!2(bt~TbmQXF!TCcjZT%p zf3`jMbWw+Bi*xM-WPUL&*i%XspNUa^?WiG@pEmUH4U+k z8~btmh-m{a$DKZ%+lj+Tnmmc?C!)V7Ii58QQIkZMntDqzId7t~vmkrO@FAMo6RwGa~mKtA2U+=AxVVFyGU;R-xL3J-pHuQ@z8oRZ0)Am`$55 zl8-j5(CUaK&U6Nv^R#3%CDBV)P84|7-WVROG^~)v7996fn0Y6Bt@Q9rx)1U@Rz`%E z!f56sr@O6c&FHzPi^`Z`T=7J@4;~-}8r`Ry16)*RoD*=V9VYP;N!uyXuC~#>%9s=R z*Q{fY%@FVp4Vrvs^&DKX&s&!a74FU?y&pW`*ML$m{?Vr$vUT#^8_V%Desy4=RTyQEI=SI!KF~I3+%GeU9SKhl-!(_VJ&J`ifs`E8p%b zQ+jamB&Kkt$hTmVuEl@T1gN#7$5q%E2OC0B$J_@Lm@0g+E9$k^gNyK{f%^)Jbv~5O z*so?S?XfkMOILD62Fe6 zK>sMeipB7)z#_#Jy}4CahT;nTXo{jLVU0TmNJB_0HuV&Mrc>%=CneB&;2{ax^pmXl zAJh1E3B`2eMC-NR6aPtoB%5PQBL9IB#&cQqOmhN5kNZ>~`UNO2hmkeT~VQbV+Q z?&Aw)s2?GKOm3wvCng4&IhoSt4Py{x4SKcI8$rE|^%?VvAkl>%T$BgOw;ynbNt&-l zgS|b8o{I+wa}5B+db+xQ0^;cASd#*jl0@h7a=W35#jfNn8o?6`ilRWtkq1=JqvC^J z6`zb}=Qe==WbOLN@Z)SUwv3=2KLO-1iAN1yPt)`CyzV&qe)JDGAMUTnX?07CWrdN& z?|+eV^?JKC(m&*=dA>d<2N^vO9n~8ljzTK@z-Maj=$Jz66QSqWb$ZS=w;O_bZ6xy# zKtn%4z*oUGO8{uu4v4ah+{%wZI!8ZGPw*xkUHs*{o>+wl062j)s*~iE*gFp#;vA5^ zKbD%M7Z;1tFmY=__E4k2Xh_r94|V>p3ec<}wOGI9{fos*UvsJ%Qkp1`LZzU|**CwB z12zjuqX%+EvPD#{gosGg(+d&Ofdn_w_&}FRqe0!?6YN+|fR0Q1ERQ6(XbjB&5DNnR z9zZOv0KOqD*1)chG)G@KT>4YnPA6BZx?iFBSH`~Nt zeokflECCO$k=xf)I}ZRJzXVPC5lpL6phQDvFcaEu(f}2PxoAu(h=8-^25*||rLS=j z28tc(2rMEg3>$^CTPYBd zd<(Hj|8S#PGVmJiaiZK)9Lx< zu$t3VK30u~)ct6$p7XR);1ve%@A@E%4KV=K?f_yAvbS50gH#nEA=@Df@PQPNgbT4d z>s-2HS#vU!7J=FQ7$9FT*Wn5i5#tV6lrs0969A=Or_8+hCg9(J<9-V^IVX!6wL-Ge zsM>)AxcD|ep*QJj=OTmK_e7@8*a-lKL8lEu2i@=m)4aF4*VdtvIDfbT2HhP7+T$I& zU0ilH?txI&dq8O$k&X;d{L#R&aR(=&%)B3Z>Clb%c*MzIi{yYpqBRpm2uZMjzUJKom*lhh@o&Nf znL!|*L;alz|K~3o{MU6&!-y1XUMzo2BFt+voKC1Rs(6*~sDi}3=kor1r#(>!mN2M6 zTHg^kZ9_tQ11H9$i^nhd`88rKoj|S=*gj1_5g?@0@Y}zJ+efYw2riJFjY5U&CfRf* z!`cSBTfy!EP~bTYfK!Ie5l<$Fw0?y(N@7kBXl!k8Q+oDkh~GIJ)@zmyuu-r8-);p5rrCe6v~k%1 uT$kgA)O|VPK;SRUgn~2@0uCwN zox{xh*P!?PJmUGC^S*e^v7!=drM`h~@9gs$pTB(FQM* zb9mr4y>Po?@SljYoVK&Notd+nk)tV=vXQgBjh(ZNrSVl)Q%5IDI|v^a_{VwG!r9s0 zNtB!0_TL6vc8=!UU*js0!9fV@?>~0J!Xmtkd0}HECR1Wz86U{YN^7_$ua0>nhtHy0 zH-{dZwTQOdGReE}jgkDO8Zl#wWXyeejVCr2U*BmTR@+m!R;-?o&Y6B8`+-3z`|B~H zPeomhu@ek#W}Q8iUA+6rd>x2I-pC-QG`{(I$@%&opZx9md<1>c(DE)L<|noYzZx6! z<}h4T8}t7ab$mrxS=rCVONJrA!NJ-}betE#PiVG3Gx)19ulZY=t7@-^86=Z$>Xc&@ zdhJ8lHFNK$-wD9duCzD$f##mg&Pp|Ic!sT!qq+JnGJ(hFjFz4Yxh=fB%eAIwaioNG z!e^gu`Fq}b`>Cc_iGtw!`sV#Eg|d!8cf+Z9jhb+*20xK6WIk>C_Wqt_eDTVO*0t*UATPf8`Ese%;7ecRk;BR`Z>rat$d3v;)%SP97Y4E~ zw?;DHhjov1r-&Wy!R;1%)6XIi`&pnH{C4BXcJ+sN&z_&fzvXk}j0L{%%Ln+cR>XRhj$T3p;+8oVwlnn)^j{AwGPV>?lw z;=20dTd}F)yZ7%$YCTF8x2OfI?0mVL%{7|87|A;ofmqaf6vp2dh@1fW2)J5 zD4(t?Nr;u5{UzuSw*AqLrM~APbE2TF!uDL3$8uhUu+tn-#kePj*WQY}*BX~zb)i`o zHHYBq;IDGvLc9s36L){btQgDVx}q(KSmA?IFS13a56%=a9=D+g(EPDk^51KffIB^s9OwEoO0Q z7d`_6W&VRJuH(x)xloI-Y5_1RKMyviZH5c6pA8IXRJ*OK?5~X-f=_rLcZaBWjd9r6 z*b=z*-NiEt8rNay?j*bVw!RF-!^Nz$;K)exsipu!E@_^x+~9?^J^l&$UHq7-5>Yj#u_q!N`T%a zi#P@DE)Vr*D%0G#a|g>}qF$oJyr+KBZ75$av~_xFt!BMt(YdeI!^x~GQSRwerW)7P zCP+p9Tj{e{kjg2%*qb^ns#3`4#VtiPgL-jt4o#fNxg@(x*^o<-Mq=Q>9SFa}4v?Den<1r(nh`h`TMvu=e z;WUe*ReWHoW;ieQd4p>X?hnPGi&RvYt1K@s-#F`Nn&`VaQW6CEgBkseFBArtSE0l7 z-KFn&9l4K7K0Ya97fcFQP7=i4=vS5cR^>DwikULe?2lf8>wl)ub9X7U+zwnJew*P} zVD75rXx{vu`#5TkdLTzD5*&1L)BV1@e7aVFzNNW^>&ozJUXzvwHLkYpu{W(3tyvh~ zP%&#WddPyQNyTSI1g?o7Nszw|q$1rzNh8QYxXj1tLoQyw%VX9_>F4Jc&u@9Dv9Xat57o|5u<^sX z;3743dy3n*yt})5=ISzLaXH$3|0qXOn3e={zVm6&h}P-!dy({?KY#88lWr6u1a5d$ z#bIheo39)NwReI@=s11qWM#ST##vr?%}*l)nE|(0sbxPNPkul7rAv3vDCF7E;*~=b zVx3CdohQxrXc4zL8e9`;uprAj?JPXRBcc7eIo;||Gd2K*Nz1?|=I57C{A$ZMTd?|i z?~l2K(Mu%4AvHI9yDq*ybLV6oZ`;PqWU=*H87nLHNGX(Xr1;kT+wk`yn6>{+7M+A= zw8M1EW*ckT1zK7q(8)^V9$Xpit*!EFTW}2j{8{$W<;&dK^wiV}78Vwnux&8uom%PK zE?l{ScN<*3$5oEZU?9J?WhS<7)8)R9AwN6z;#rY;A?vc}d3PkP55lW=CD4zZd^-nC zfip!;a^&KZdVT1?R@~9o!D?SbgjS{Rbkw=0rIsJ5*KR$b-o*(KmHs!8(yS77-cA%E;KXlzAwt zR1K}$xCd*Jklh?u1d}DaVtb^}$zE$!tv|IIg%k?G_Q#~kp;ON3bF?#6 z)JlI{Kp+O6!TS}BT`lih+xG3D#y+NfKgFm~LRVzsjK00)n;JB06)Qu&=OX@NrVqzy zzB*beeHYTf_8I(zU-kwV+RSK5gbG3I$#0@hHrDJ8iJF3|>VCg=RWNhG=GvZ8!7Y5@ z!buNgIWuuE8?n+N=bZ_*#!H~C&q$2(dF_nQCN7<6U**YUNLPQEU_FW|0E z*2$4R8=tiO9E}<%Io-&m(EC&)9qibiQ%8FIqW)BW!*BK~e|7q&*>9v^8TQ$U#S(fSOxLR#r+8P32hTjE5y*GoD}>%^Y&OK5Y0Z=1%ac+}vWMQwOiHO0?nOLGE_iw%KKY!2D+x3A!5j7hps z|GaF&spA~mm+{sT-t;0)9!f`Vo2F30vUZzap!Ik&D_n>L&ta*QIEYSvDXFPUAS_YL zh47jPZZewg;XCS@#eqihhVM_-Bc*I>MR?=YvuuY~_3=#f7P~s z{L$WH#-_U5(p2s_m?@|@&awDg76h}rk5v1op~rA!a?>9ftVepy~Y8^g_9 z3UPk0y^$OGqaDeX^4Jj5Zg~YidZY~Hs8qm_o4YpbTlZwVq2v)yLW%iQ!!?>5lc+vX z?L%d?OFK&j!Ep#yv=+=io%b5&5l={hlfSUzKzA*bJdTLbfKH<~?feHlXVl@_n_m&I z5eF$qLeyFORqumFs_x|H=G7EaiA(hPHJV!Cm!w<^UDwCE!0sdLICIXmCIxJOYxM{S z09msIEKlJDH7f&o(=3dm2botqDnh@R$lO;Bh)Ki$HkBS-%AaE}!K9mgnA5tx_iTtT zOJxL$+q~z;xkK)BMPW0OW)^(A2!)@ox#Z6L=uY07@@8djt{p{B2o2InjW`TXjYwW& zV);ZSqj2K({)C@|0A=<6UelUK323_U1oKf)N1|R$~8X$f6$={{kuN<93w=ZPl z+*fVNu7dyKN{|;#)q!_e)N;NufqhP17EXtoZwU2VcZ%Xfy$|-sM>$t85CqE~pVX`+ zj0$Dp$DE(DXa*-GiJ;|mEuH1?zpq@jY4~@TqWS1Hl49<)5uPSS)tIVLyEbZ=ik{y?mX>V4TLXM zCh&+RxUn*Au3o_pjh!jDVLst2mZ8|VY~ts>vUs6-pRffvajtif@4>v8(0hGlqwh6s zv?5j}?Fi{P)%eyGxZQhaHtyzDoc_bb3!Zy~@;T@u>#^L!zP5ytSBdA-3S1mlWKqXF z?7=6qL2_J0R`!UOn{iaeTT*FbcI#kPZ(}Nm;SvpvBEWupoa=oVgx|h>b6)*1$k6q* zJju#y;Yj8gsth{>@!tosUZf>zz8A1DzGLXQ|5a+@sVmyj zq$NnY6WKt?sJl}XB`tF&Uf?Ey4(eFE>h{X5dFZ0tk#w1@qL88Y4?O(l=2czJqt9*4 zq;qLF&9i~Z6F)rm^PWN~yE~4;T|G&Oj7I#J7?-_*bJHZ5`et8i=+jbn5=xvIHVmyA z-Po-l8TaK#pvmV8Jcl6SlpXae5#K8et;ia2FZR0jl48HF@_CXJsxytD7}o(UB(`Sm zYNqF#=;bCES@Dc^=H^TsNo@b(=h^CwuGr{k*#{5SP?=)3cX?=Fg2ih(5LYKCT;oYO zkCli+Q%fo7RerMe=sm|CJ;=)|0=@;7CKL?b@%!eLFsbV|@=!kbEBr5=GnB2MZ}UP! zz$#%BsmMK2l`m-sv2+)JIkMFKKyw^ex-0cE4uODd*`H z7)6X8MB6cH)c17|4mRO`NCAO>Ug_B0)O~9#mSzcUbeXGGX zT=-4;&~e;*r0R@&wKcn{n=-?Bd2vgnx%$n(F#KMr{ozALfB%QgYWH0l?ZJMyzwqgWi$00tJB=E4__U&d5~TkXNxZ3^C^$Yo{l1LY^s%0N#yTD zAW(;J^Z~)bhWk|M5mdMUHJd|gbJWsE)2uF>gtJW{=E1~}EdBV^K023ve?5PFu7q5A z0M_3%Qc}1Vdi7|xEiDQufZyRJSTM94C(j3fpebN>pIcj}cRkqvvH-*M-e#nD#B&_r zqnlPn-@R*@f>n+E!S}V$jP9(aDBTa{B{lk}J^>uHupD<5zIh!}Xf;1v9h&%DapB+d)E0-wrOYFOw1gMpc9XY}^&H6Ze4YH! z`3jd=CYVd1<4c&Bc9r~9QSGV_%sXZ22^4D0P=0M{okSBdvrh6o)+Yw@k%G}>@R1MR zqI9Ylz-GU7pdkxDQk$h{Xs_F!sOW!zP+l>u2MPu8{{_t0BDW~h7RfK40Io|I@q;&` zga6=9^v0GVw;)z$3_u$YOinOdrg)UU1Nh zp}(+^`jO6y`hE@;4F2N&tkyb^UJSK z{=$rwB(du|FMsJbzhU5iAv_PsMs_no66ecIx5eGhjNuSm#m7a^l);yYoiCrs48GX| zb)FSDGU@rZTRy}WK!4OBdiWiuzR2VJL2_$H`fq&sk7NW`BRCs{s8OG+KhGXDTym;k zTMLmAh+j}O_{%RK$=21lb?OvT#XaeAhljOWBIpzs`!cXCdf#GWzL^7h#Ll63L)CR_*e*$QXC^N7;22mbv1gJQ56qk9RcL zhN~JIGvbrbR!*|=@_w6lT_3HA0v&u(0a1qeA`bu?mseH>Jo+ZT?GLhN815Ovp>?JZ zGA8<@%nJ85_`^t{(b??m?6}>;U(#At3t-8W7}4S&LBKk;r-z)LUZp2hA_Vl&`*5pv z?}zmZT-wwU~MlcX5ioFjV8-)8efS8AU_13L7U`VlIxeZjng&;V8zB+doUJFzZgegw< zNb@7i)V)BOdRs`E$#B#I(VKFUBx)+z@14aV?6|uAwUp34yjs5+Qf1mx=CZ75Y;1gA zLE)X_Yw{Z7+C_C$`KkvmY zZTG_$es}($ga5;%r!SZ@JXH9zRmFd=O}QOAU$g}ABAtsZJ*)y}Ls+Lm##QuvzIE*i zY|k^8Akk%B;yhCl_y#^E%@BVMjz$jd${XZYhFzbm=2fKRKAyi?T3Mu5P~F$DkiG~< z&&a#KbK@zfk8gK5bcUbxK8#7ZcHRHL%ER$Tam1LPyh`9(bi|E^%A?L}lOzuaodQSp z(j;3O&2wlgM|*2Sg0dLgUQeLtH!H!h_13|$9T1;5#KUM4+lND>yeRLrH2tJwhGa8yi1JT2a5h?{Jbp(sl!eLO8UP&+-2GGP}yc?b1yeF%woX z5oabS!;3w-LcKy{PinwfeB0b`MU`d4*?Sal6b-DX#4N?XYCXfSLS~=BKfAiS+lN?! z!XvZRa$c3h7%xi=5g6vOyBsr~F6{idMZM0?V>g2gD#YE2>G}3yoNH`>t9^6Xp>>kt zo<(r)J@cr^SxQ!JXE#0jiQ`DRTGX4+jXrFPcn00vy__8P#qF@9NyF(q7lB$@$wZha zH5=Ood|gmdci{u_a^nuorO#9E7x#P6!;OV;YTK*j_$CP0rOVEZd+1FqF}d?H>(8ad z2bklFqEeV#a^mn$4kmZd>@Jdr->!byS4vi;bVJ3?rUy z(MM9ze7?S_JMb|Uc4;(p6@djsnVNj}W*rkXa5wP%*|=7bcFc&@_{rKl;k;qUK{}YD z?~dhxAf_lQ=rO>L-PX(9`!VEIwMT98_@}(PU2F==>%-+F-6jW`it(T6^XgnzloZpy zyU-ui(}wNC;*8sAvJXAsXFu0ceMaVuQrMvR%-^nc(YFvOn(o!G(S}s;F4i5BnW21fBBA-7SNAf*gCP)do^)jHym6pT;E)I2}FJ|N@TbDYg=Ot2A z6&ttr?QmJ(llFMWhi};c4~KT)MV7faO}@H0K}wKXa^w5+=BW(ZZ9*yW>&sr0C`a}$ zFR2MshVJQlSMfrbh0Wj-N%F8CsXg`8nw;yVO&_C6j~Hy%e@>#|=y~YO2d8@!5yxeV z%KJzoJDhJ*z7&JE4x3UYD(eJc$QcUX!$gc)UIG+rurY3+VDDddnVEoH1!d7<>fsgW z?~)WoOizXbU|R2G6u65y3S#>1hmB?uG3aUuuNyB@YN^8#{|@bi&mX=WI=!noXyK;IJ=~t} zFEY6oNJ6KOBIMWGuApl%F4h*2@IhmM&lF^jmBR zvOH#?>6digLcJjKOXe{zib7D&B2`>9E1)xa^y5j3GtXlov?SW)ebPB*9?aW3BMlojy!$ida3ZCF=tP~ z_yG-z2NNCXkm~_+UUs_Ej9ac_rV}jM|wL+F{ftIKldod_ePHm{-fSu_OgtUbElQWztJo`0LG_LFaYGIWJ(Mm z{|93KscH75wQr`JpzePGa&_;=({d#(=f9HZZ-h=5ex&jzjK`qJQwIX5_#dtEH!A-7 zKtR9whdi8sy?+&4e%XZ@@rlOsMYy=t1dNQRtRAt~x;J~=PUK>9Ht=n8za7)x$Nu|F zFuU8I3?bU^@D#CN);=Y&0t9fX*IVq3TM55U?H^nA7yMvD7MY(OnZLOUX43p~j{o5c z|L4evobFjWAPj_C^nad_xCKdU!9U0PrPdFzD8)AbhqVvF%EHo@ry~T=r2EY4djK$E zU1eqtKp_#A$;suMo%w6r*6{$cyydxLktkq&7f|Mra$7j_Q1&m}z2#}^izw=Ai(;kX zx46J(*7>#2`}kmUAWtU_IC8!DdSU?NVZ}>$2?h{Sm;t;?M9m)zY~C}4R)g#qS#oQ( zBjJ|m`Lp;WpJyEt@^yaoPXHJiBKDIFxKbe@A(B!w&!4zQ4116b-@SCUBuJX{%HBF>_>;1EpncwPGo*%JC=A^ic83}YDXuboma^^1Um2k1!!BRD4?|m= zc@Y;_N$OZXrrD-~WMjL)cd`DE{^gZGJhwZMT!uX|2D8Q6C0IoJK*NB4B+b zid7Y`iAZ{hg^`jh)t*6Mr9C5R_QwJ~&5NL*v%pUL%|pHx(q0u8PriOH0x%Rn&p+dK zbba)X3Av1J=y-3qM)pycprxMBP9-4Shfi`Xh+(QL-F};xDlScg5JB3QDSWQozrbE| z8GqOdYuuf6T0pHHfgKCQPE+!KWw>y7zVndD&Y7n~2hma2kA{Lx<;Uf_#O<4AIxIE2 zswroKBAWV0!?-DATAPlw-DiBEk6uZP_^kIJKD$R`@HI~JmZRIb)8$O0$X_Wx&PLl@ z693k-U?V$@Z7p;MJyq!KdSs&Mjol#`L{qTsMG)Jz<|WN#@PCjg&VnJ|*LHrcHFS0? zH2cPpsx?v_*4>!`*57%&_G|#ri8DF`1~78)1t+`|7xK}vHXDl`ayBi$?O;w`{<#6_ z85QAMTdf6w;R8eL;X^_)HliM)XWt`fCT8BeLF%`db|@eXX)W-Hg0RQd94v;>E&u#vA0d^dsb8mXGpWBYVXO-nq71$VWQBjK7|^(wK^KjynR)8-m9^F z)b#39o>uq|wET;xtUmSp+N&J9c0bN&Y4VF|VTCVcvF>Bl>^JS;Zs?jr%S6q!JvhIT z*6O2?F9HMAz|@bX@^<$XqbVhHfUwqoh-64Rue};CwhT`BF=Tf50~)x8%`Q+NYK`N9mKSGYQT~v ztLFoEs1!-X1d&fo$b?6M_3iOz!SopmP&8fTbl{r-DmbT3Yt*-OSyG<&ZIj`}fab zmI6R#jYf)1foYC9TISwc8&k}CQl>OC1yJkJ?s9^lEj7j+7_aqEisyMw&Jp%}snfXP z_O|;i17BZi?QHQ{A-qqz1cEfqt5^t<#>VR{Le;zllqS2^u5IkRZ#nSRe;so)e-k_F zEWL%p)U+guh=lezaT=vmOnY-Yn@GQA4wxNIW;X+~MCbCVgkB+!(pJt5erjm$kshkI z?^@dQ`aAovB&xw+0EJ{V;;jzov6uLN8BZNjwpe??=!vzbr$=6rcu$-3$LZYL|o|IZhD=86~w1i>5 zeOzLuQJ|j!LNME-PCXYdUzP!xAOX@>X18YKtL>!(dlP;+UyT1Vb6=^Pv0Bvhha+n+Pnab2z$ z2{S6DNaLfgIVLm_W-Iu*=7mP2Jh(;lN!O@${(Vy8hIsM1%=e|p8@A3|)@UKv(s+&G z2IcgA`6|z){KuhF6SBe{)YWws&iA*eIl!#&I!3o`f}5chu~tJN&c(#^0;HJ?3P;UK zcnuq{@Q5g1&AJ>hH(Gtqr2-JyxRqpjAaQ9hhY!S`QlY3r!nvojc)NB1t5P(VDX_0_ z38-ankJ#`SGtvs%jHWUW3gibmUTo>@rLm*26Re;y%* z7k;wyyeNi;c=Yzgbe>wnv^j3)0~9Zam>f9)A~gS=s=x0$uyfg3-6m z^1)Q+dPsTeV{{*W6TZCCY7idMp_oZ?@t~t_Y9y{8^|z+x=Sch#2Mss2eJwr9O)Fe^RFn&}x#r=Cmy31K$SxxUckhzJ zd;f3w%>1^Kk+C}0!4zi|Yx*kRyGKw6pNxIdo26rDy(Oo)M9NJZWjtBtV9(O#9;xar zv924)oRcej<$j~t_FRkn1IZ2ADwbYlAc1ajcvb27>P_rA6dc*=NtJC*R6DBCS0ASV z<>&tRtDsE%Q1IsJ%xY17!-GF$TXw0Gv3c%)K-XWu@)Rm`j=?*P;{F0(Dp7^rzY+rf zYRf<_It3;%3ex`|uKodwofQ6api=^FoaM%!5c`sb7K7 z>G=nm{sQU!k6)Z>)1LYN=g0ok?6c(8P95;?FID$DWfaUgr$792ICxF}8X;a_E&pd+ zml?m?20E-uE3glIJl~9eyXYMiLy6~EB^1_*>$#hF1+a5knwix&UZTt3Rn^{IaMvz3N=Y>in!vjccJ=J)2d_z+$TY~N91pG&~g*7h{Y%>vAfM5*$R=Y!tRuJ zRe56>#Ed?FZDZp(mO~G|q652FGPyz1Tm8if=&^v^kijeKpp`*V`z4|sua*Nr(hZb& z{qEoi-kr;E>}dnBfrjh+CV8HXwv1ATUc&2LtXB$SU^kh=BB{Bqrqv zQhp#|h%*Ti9~l7e%F4+d4$mZOp*4q5d@B-$qxzL()Pp?-@yq$oZ^FY#A3iIi3AfvF zv8FLsPf5ZVtrw6LXcuEBd{uUYjglZhkWk~tJ3;EN{oBVv9uk%=9u#y1$9spZv{uL* zo~&N%MEjMLA^)HL;aCrReE1uX7j7ihTUNCqEz1sC>s5M*WnD%zuJMQhdD0- z+IXKsssxJFq#EkaUeKrZ+ghMa>Lfe4qusm)gmXR(PPfcVtVPP?StIoV1e|Fffcp(a zIP<=39%a1k$?fX<$_OMIkB^oMjzQ4%x&h?Kwzs10a46r0}sTcfac#0)JA( zdEwP!wZN~m;<3k-Dd0xNi+gP0c6jY9ycYZ%_Xy;FaScK4o+_3C2%lGOKIyG=Fq!K~ z0|u&(cA+8m|Dpms09ZftU%BPq@TQE64EK>32dE#QnG?$b5izi!^7XgEfYoV zVFKzR9$QZ{Bg^1?3@bImTTTNfQAj8KqzO%OYHF(=3GGT68f1WyKK9IQoiZppMy}Ai@74AOKu+={t8$Y7(#r2naxv`22eyV-9F*Lm_-M zSHDhlb)BuXr9}-yi_cSVDSs0O<;kl)f{XP5lNiM^C7a+xj>Wt?(-_Vh?KF9e_(scpI3el((LqyZNN_s<5!| z>ZX|uur*a-HXsO#gHab5eNDM5ItrF^1U-`{zVEg^ll&htr8}*urv%yLOQ&EJZ^qxi z@h=ec7aqxUzLNg0=J-3+`H#@p^=th9pV&;jTFL+4&;9~ge;t1g$eyQQDJD8?FsPw6Jh-ozWsl-!GtaTM3w*R3ICeFMe)BDhv=KV z#9#gY{CGc;TUxfi%f)5E=F0u6%;Z)Y0aIt+YJb~7m}oWSep#$gP4#l-wwzR)3(-cKRuA8B+y_HfXKktZ)`$T8B zL#ACuqVR&cg2MoNT3-N#KJ<(iZ9PDqH#Hi#dgF0Z8u@()ShwpustuMz9K&aOJbp4* z3{uUhJa3ONT&b}?A|hH%I-FM5>&<4By(f)ohwcNU7sXRNOj`r{ z3W=vM)PYH%>vbi%)wI1Qc~^(hVt%NrO~y%uajV+a+$m_ZU-;-S(KonpoAA98Ztt{% z`%fcS?A0oz2-1u7JAH~2oQR*H+?ebKHl>K;Qx%~`^Lszjy=9M9i>K_DO2hljYbic| z0frhMMN?~kU)&c4|BF$R$f81vY=cCb?3M59j?F19>FAPI(Lns;H{R=BJ?=ZU`Bu(d z38ybMfXY(E^N(^XcU=TqYVH6%lokS`I&$R;#53o7M(-asWVr78(LunO0S7;_LEw>* zVPwO=o9kOU+FFJ)?tbJ~uC##l>*M1RmWnY~8q5pqkn#v9WM=be{b*`iPMHN2e>EXH zazBrG5Z5QDQBmUBJLm2)URKjA-)BKj8awK)2~*w)ZJs7=Oe}EB%+u^I$Dp|3J)kIU zSNrKgHLVIX=*MZ!KFY^Q>WVk0Tz~Z~-LY!2qagfRqiQDA_P4;=skCy%wdiTOToaRqC#QvyWsmT!oMymy03GhX}jH9Yq7meHzF9t7~Z6ilWH0 zo@Lyu=KJew)#OczG%3XkHx4FLQI<6Z?#P2E)dN!!%{tHBcu^O2jF1KP{z|9A4`D|Y zO?zoZ>)6wKO&Pr(Q2~*NeWn$$c!%QYE+I6rR?JCGmCV$Y%P}&eVnC@ydU3k}7nw~eetR5D%6dG2Yv!EtehCPRc zk8tOX8+IQ8XcHuVQF`c*Feb1FH;MAaAP}I;6$n9KS zjD)*29F$**4ysQ#mvdcSM+ryYfG9aa`f(xk^gqgWtO??j;q{XcuiicGue;_Md)A+* zWhQ&-pNgAxY4NxQI~X({Z4CQwXxmvFQXa40bI&uFTq;UDVL^95feMJ#nH>8h1SWYN zt=}R>Cq+wf^}mX&hhlV(h_C$*l?)f`e&=6)E&1)LPVbN9PTmK9`E$if4m{^`O1Qwp z<$t&RwGxUj|KHTT@2_%%znONi2f5Ly2#pBte-8Z%Oc?;AL_;dRfo1!nG~H>wn^xTY z!ZT3j0`8W(09t?wnR5UrV1d%PUKkV#ysO}d2s2P+gymkRe>aj|xx}o?V!T!aUT#}6 zQ+*Phx~MK#fcDKD0L_6Rg$bE~D@ep3N&5`c>!?>i^fYs|r9qh%D9OS$+*=tT z1{RM+?^O^;I@bUKTElf-LV%)~w@0%9>~JBj!?gN;#C|*Cpt=o%>VaQ&<^cfuBn*;7 z4WR25po|g=K>er9fy96h6l0QS6%gWvyo&Mm2`IV?;)BboKmXx_B>#qup6c~c>pq_7J?An_VyN^ znwr8APEmPD&e0Nbne`SY{y9?8`LD09gUBmLQaUs`@||c)_wzyFmpQ1EJSiaZ{5FA! z*8u?BS(y6^{zXMcN0(l!%L4UI?|O_X%q2gXSHL&T9O()}B~Y{{7wmJi6nfan!s4T) zIXfQ!et>4_6&~*__KPD{N&v#!>=s+UKt_ghTUHha)D6AYt#Yi?`V zNc+J@TG+2%y}~pBkXSkC92Diqmv~Y@bf^XhO4ZV*9L${uioPs6u3x_%5+06gc(^r- zu^QjLy$U!Jh~=F%1OWQQix&Xh;;vQA(*uhz5WYHB07DrA>0;p1qYme>Un-d{eQoao zC^G8ASc{r@)EFooHnRlcd|Ce&IE)8XZdkp!k3kS^__ZLmBnwO6oq+QR`4e+pHio>U zo2Ru&+mwkdcy)c!;1Fhi{4LT+%^S{A{5m~j-V>N*{P)e$_m>$SS5dG)_HQ}TDb=!V z>`xQ96IkGXigN0Ls&92(?cexay!0DC{Uxs{44roBoj~~Xl!^w*K>ru`b_(h-xt)?v zvp_uH6yW^-0Zc~!C8&x-GIARHt6S!7=fXh*%|BY><3sX5(r3nUKG4xA_k|74&niK0 zjHsad4My%8Z1I9uARiOn&uX)96m%@G6n=S>8lODt>mUky1zVQkuNqn zX?D|YJigSGV=X2KSN|HwHq^^0>SgrpoNgYs^BVa$CKCbu z=7tsq6Z8}{msotTf+B_wM7xvH{bQ1e^p2YBYK*=g)P8GhUs~>)dQ90%2ULu!U8PEz zOUOl`1%b~Cw<);sOvVN2`EnV&>#$-)z8K>;)%PL@ z{M&MH*PAhy#$$kL})FE?8D*$$7G)7AcokN>H+%4clOl;T=!zF#YxrUHa&{mXWJeqVY9Uz{L1(fvR{J;QdT zX&N7l>F4i1xl0jFs{4$_iZUfET0^^NgH4pjNd6 z3|2AMyLxyZd+7a)Y+eGU|##z|8-CZ`w&{@8Fs1PWg9B6!F8Wi4jgYnSv1BnBn zZ}D-jwVqa>PC#_HhK2}`CS*!pwabxKy@84i5_u(k+bI}n9fPI(~6K#w=?{ji?`863=`9=Z|*w(niO zHMmS*a^Jemd78KgX>Rh9g)&6_{SgI91*ucASGi5RWZC&xq#}Rg);7tLtC(4 zI98d?b^!LY`dbywfp6Bsl99OX(Us$#fYu*+-a2XA%(wu4QK}^i3fiRzu#2=7m|sL4 z&Zhz0&#_-wsK^#BsFEr!52VH7-DJ%axpOVg9Y(!Ck)DN^4AfMOh zVSae+24|Dgnl0(&z}b|uc4oWu!qJ-okB>xZrAWmgCgUv|*dF^dx|&NV420Xp;n7^R zrM!)0Y98-RHrH&lH*qT^?DbrxSE@r(=2B@)vru4V@vhyxma^n$Mm%|#6{g~@?~r{o zkz(u2Sa1m4jx?(^^)5{HvaO|`mf-Qe?Kyv)`)m82)cmB8#TfJS;}`w&&8f(iS@|u| z65EtSD<*gKd(9P64WZ4p8U;VQcsNBpFX%eCbeDEbm|qmM@qArU;n)zDg9l2*MdW|8 zl7Yr{Ie_a3g2m~+-~^&HK*2aDy#B=`*mTBImH&fbPW;GDy&1+*M~DB*qf~!( z+NS&4*&HGJkCJtY<^WQ23UVZY@|V;8<7pq1Hw){(T)qYj^(3ZEP#?Rgk#6z=ETZq@CR7g2J}|z$YH$>xy9I*Q|+R z9<>6hX}}{COaS!8JP-oVq|2cCMM2>_H|+S96zbr5z_LJ*)w;`F4zi_FlefC$q@l(vedr~pM< z*mZ?+wmqhK*3n^UfUT*i3EZStuU&hdlV5!vQ!WacD8WDcdJqNoE~x4Pe|Sl7nFTzE zf{3!MAH-uGmsycx0WE$YUk_u9T&ALux3S^G0yH@bGFFX)iwmAeA`j*P7WiV|6n=Jg zzi@I(e9f3? zm76#8YlVTIgZmbo4-wU^7eK=Y{Yl<)|4nD>oq#cfaam104M0uj<3htm+zz1HxC84A zP~1zGnijsj_jv+45qtmhNC(x~Q}u*@3m)C?aXPgASfVK<4Am+t~>S`aJ`Uw~8QqVP2$aYUy_$QA%RHB=0&d(&r|k#vRPL zJw&qDv2H_b+T3sFhM^`iHZ|N@n#*;=)8O$lyAFlhTGMsfh$obz_~!!}3ZMMs#v7H4vtCQplse{GOYgpuO$NzX z$D4%e&~_rOo{OZh5p;?n)=sQo3G|Aek>em_y}dz-I*$W7b*b zm~p=KY12oSf<9u*%w(sfKpQy^!AjJE1=3fHku-#qn0w5I3alXK(`@p%Sv%36yWeHp zU4Y8Tamyg8*!22_*YAb7ZA=#k*6M$B$%~WjsbM3#G>M0n+Lv?`T0JVm<7P^cpIaU0 z8!lVzA*6Rxt?_iG?jxz_=qftP8b6Hm`MFP5v$7jVMv}RX=uumpYad>5%_L549&*s> z`@Z|NgA{9Hy(mh_-HaU{E$le5y!EI=tRO>OL$6k!Tb_+;Jb9ZLwj$v<(nE1+WhN$n z-=XVKi?5`|H?3Us_;3%+ge$S$6B@)JjJ4;M;~e!K1G4qyEvC4>3sZ8*2l)E*xR>>V z$erG*;kr9n%|(vYQB?0dk1;UQtv6=?4=-Tu?}s{C&$H+3CQ7W0jZ8YqtPlk!8F_Ei z3bj@iAUZY-3AF_;v(%F`kNdrt_GRz>+=vvE^yZG&y34}nYITWJ@=g1&R>^)_U%~ znW!jP?x;=4Jw_KYX4bLi={2Dr+0Vu{D&%z~^5$Xu5cRlXmGPGw>L1QE(+JFkR5J6Y zi6uRP9yjI?Q=5AEXd(`U;>&uwgQD~}6Bd8Aztqp{lY`)j8=gmJRr*=yz zEvbeiFD5^`LES`@_Xc~QIb^Y?Ha?Q-$N^1AuT}I!Bo$gg!%G@DpOAzou~*`hxdMBY z=i=tMd#yWF#-}@0q~K5)r)f#JSb~O-zywEv1iyIXMZH~ zroD{vN8Pqcq&>}nv+^9=SDzEuFC~Fgap;Zf!YjR1LZypm)Xn_hXd}7Ur)Byjtz%DU zQay-PTLB4xb?4wq-J)fo_@1^z138Kec?{n^WMh-7_Jm zrXG%vN~MuSo1g!-zIZ2${!&wj36&TPp*e1mz2<{J(+P-UUu6~HvyEbXt09N&T~B=Q zB$Ppp1P*qZ9>t#lQyK+mj|Aw((Zp~! zgH)<1i;iJ-EH*FXMk4RrijS9s=Y&3Yb8+V2r=pFl6f9a%k2eK-NV(9ZA^Vk1Du@?a zk0ebrpxNxbvBeEE6SHZ-S6X{LVr{`8xlLo>8f~`P0G(wmM;)$Mp|VzrtT%Jv4vILk zOEQzQa%8uY*yUo4I6m8upI5{7n2X5#L5p%t4ZH>eMUZAO4SzY+-dI9!TBQ$ThPrd)wptYH_ z1E|Nx(I_Tb;fOIK)}Np3FGZ%chivi1EID50GZAP-OAt43dx5;M?89V|X_F$w9Qc9# z@bylQ)raE?gNe6{sP?zBete(KE(dSEiVgQ%#ZW48Q)+ru` zs0*(Mw#SgnPN-MeNl{oIX84m|zT6DVZp@S2K&*gMr#uYoc}1SPR)M4pYmqDmEBD2t zBk#lX1Zh=L${j?bN{u$Q7gK_^t48h3_1w|K@e+XJ82O|IAV{%&sNOoo*((cZp9R_F#UodLcR&&bclqIv~=N`6DxGJ zh|fyll1UbeLx(M7+=JubO^=88_}#_eom9)kTlx{K!vr_6cWE;B3n@B@-gz^Vnsy-8 zY2Z1_^;h?X47d&}R^`+h`>ZL)+>Av+=FXe^PvV~NUcl8|*Q$(q5~_n{auGtYPKyPo^LJLmP>&)?VU zH`jIjuIu_OpYP}M9-T_+7hH0G6|&uycQVHD5aPC8xsnPyEChAOQYq! z^2(EY8E0_on$ssi^@I;u3qD;aB=O}o0ZdrJ9LxxlC%*+k&IEY%KB>fb?_GATWbei+3yGY zO5LanBXSO_ma`z+FASDugW3%Y8iMLUCG`Qn=~#8N*#NAsqLC~m$Vcr9$7m`KVtekD ztdyl<0#XX!9`5W|*0%0{!hK@W>5tcZ4mSgQ7Eg3N9}=CBxq>+yNFpnm@!R^A46Q{5 zATcHKBjuV*UPFqpagmMD{J&^!aSPCsPQ-@EnyZ9JPEFKXHYEy+%&&b3LLpjxkNhF> z<`kc@#JXR&-FjHrctG+HcYb_}bNgWOT^dUT?d_t;dr3Vu3Uf^pUNwt}31b?zPll zx@t2{r|Qg-XNst#SV$ojHA-0UAHC%2dLdl-V_E($&d&Ut_u<3N2q6bqPd8Gf%*7wZ zPP}8vAxvW3mz&#nh{jrD}|r*CHG&N+1G#(_HOABR8II1 zz8`B zDpRx7-rQ=kV&d=_;k^H9w6?#FH(SkatH%1tQIQ5B{p$2Fy)>T6(xpuFMy!7BU2$V1 zk?ZZJ8+6i@G5(__rcp#Z-eslwK*c#?lNv%`DVP(vr8V>|xvD2y)`v@Is$4f*i*~%T zWN^OKtR^TLe`0O**X)3f^ac!V?5s5siBP3r9KT^O;}oq+SjlEGT{&b_JpAQcAxm@M zflQ;!lfJ9ZU#9w(xF63ZNa#tWv)~sw=P@a^8?vl_$;`-YR{_|Eb-^=Bf=_z;wr}NoRv4}(cE+iuSdfBcxn_q(B@}j; ziChaQo>1iwPJTjyv%tG`wZ+^0wvw}U0%W2(*{8S_*3THVaxSY+Jw?(-Mg>qT5$!!{1~(?d)%^F3xtz@PW-mC(z4NV5 zd;tt-tBs-iu?9s!6`o&O;Hq?a|Dp58?NHv5{eK|Mlr%n87+8c7TxUyO3{?~;7jirUL&+T*Dl+m3F)X39A&cGiRA#>*u)OM z9GAOFOb&=L;o$Dt8gs5&x|pnC;}wNyuEaXVR5Y@>l&Cy?sAgzQrLd|*HQ#;5a$B^I zy^vvedQb#)Yi4t4ya$CHCLnz2teu>ZRMm7Fxmj@aH7xk=6C4+tEa%k^+r|7$yG`2V z)cm{>4~1&t7S%TQRi_jU!Vee@AK_`=*h>emLP+yAB+L-QUe<3?dfmsn?9P}Ld?oT> z4+AYu`slly(|dW@kTf=Ms47cL`PkbuwEUKI_2DQfl1nR({i(wJsnnC&1VUu7(Neu| z@2kT}s`2Dv9t4r;%(u4%$jGTvKD=Q(ij(=K=7)V2s8R}>b-P3e3KF^IhT^2YsSY_U z9$Bt%JJy%p)%zWPF`{^p4cGz{uon-{ugNqL`XwA|@HqD=^&hm7DvNqlbk5x~wanU-x#2L(h!& zppX|i_K&32WzIQ@(xK5F&3md7AjgL!(NXF6n za1P#H8yiWekaz#q7hiorfiF*;MU4Fh0PpmXx18JV6mx-W?Rio4!7%qq~rBv`@%m zvfj6Kpxk_$OE0x_b$W!gmOY!M)TZKRSG?)(bKH=CPo6bhk;8Un_Bf#{>DnGkgj0l~ z&FScis&Vh*V4t*<-^*kcC5Y5&jpz~Ea;ix^$hF%_W6CGjP-%N*TYH2pv3TWE^NORP$EPl4EqLzfyX$C*WUh3bqg1TDq|_z*3jAJ&MFU zOVR`4v9{R(@gkJUfcQa5S8cQNpVJl-cV{goFy*(;KL%N9{F1jy`0kLQy9%nHE64b1 z{MEMvTW&fpU z-uyS|7ATSbFxLEMkVo5pzW-+p=x*Z9pH4E-^H(wYFX#Q6I{d$7jKIZe!2AM~HBsIl z*s34W^LO{a50U(P#s0s*#{XMmX#wzDFk<5NT^YMU>f}VbZ!N;shmmT4kT*+(6b1n% z6-Evb#5_Lv->*EwkvzfS{mgrH^40CmhkMj1zjn0k4uJ_v9unXhh`J?VpLsb(1JHMG z1Alz5)Oim&ZfiFDc+Ci~c(WnL2u>c=jPWK9Cfj;l@18{izDT`n73_74LIYOz~aOx>;y;6*%MDKmUfO zK#qU{ROwq4XsKh^aHwT5B+j0^8_aMJ(Bkr!6J*kWh5i5)X6{{WM%!A|0T*|0x8bt7 z`q~na06BESDla)$>}5e;Fe2$6K77zXTy;AnehWJ@SBRgCQd#^JWLnX4VI8mR^==lJ z@K67~(FWyDK4r8dyz21F#Gv3{=n4%8T>x=u3 zGrYY|l+Mn^W(Bf0#9ab_*sc4Ce^ftu#8SH2V8aX3FjPYRe z^m8?#&cNtU3UtUWxWmvRDGV9<5vRZk+z4xhVeDn8+KMY?h3gM(6>+jmvN0MPVAPob zlsfZHj%Q%Ae}IZ2!Zn41cgmfzYzez7p6N+SJwypy zA&&&?DF#x#2BO6oVj>Loa3o^b0&r!FY6Z}(Sr{h<*KZ==tXp1MH+G>K8ygRP7%hJ2 z05ud<8iiD~{Ei*8eAWsj%xl2zV>TuZGJ!$8)o=ZCBhapSM{aI3LJGMbwtisHvV`98 zCD$K>>WgSc+X@bd1;vb)e$cG2RHLo_s9qKaT8>l}d}18YX6EQ7`kJ~kI7 z;|XL;OpGG{`hg)a=5mSwOJ}pWQ1YuoUy5xfHP}#JWi(z$Th%t z$Ti2SgC$Vo8BVLerBlbtyB3ejN;DlodYjCZKLGN*LfZWJAZDJ?ka}BdSG&Ml9%8h; z3^ue{2vh)o6fO% diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb b/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb index 1ced6d808..e167c4f90 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb +++ b/doc/LectureNotes/_build/jupyter_execute/chapter2.ipynb @@ -13,7 +13,7 @@ "What is presented here is a mathematical analysis of various regression algorithms (ordinary least squares, Ridge and Lasso Regression). The analysis is based on an important algorithm in linear algebra, the so-called Singular Value Decomposition (SVD). \n", "\n", "\n", - "We have shown that in ordinary least squares the optimal parameters $\\beta$ are given by" + "We have shown that in ordinary least squares (OLS) the optimal parameters $\\beta$ are given by" ] }, { @@ -21,7 +21,7 @@ "metadata": {}, "source": [ "$$\n", - "\\hat{\\boldsymbol{\\beta}} = \\left(\\boldsymbol{X}^T\\boldsymbol{X}\\right)^{-1}\\boldsymbol{X}^T\\boldsymbol{y}.\n", + "\\hat{\\boldsymbol{\\beta}}_{\\mathrm{OLS}} = \\left(\\boldsymbol{X}^T\\boldsymbol{X}\\right)^{-1}\\boldsymbol{X}^T\\boldsymbol{y}.\n", "$$" ] }, @@ -79,7 +79,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "The matrix $\\boldsymbol{A}$ has the important property that $\\boldsymbol{A}^2=\\boldsymbol{A}$. This is the definition of a projection matrix.\n", + "The matrix $\\boldsymbol{A}$ has the important property that $\\boldsymbol{A}^2=\\boldsymbol{A}$. This is the definition of a [projection matrix](https://en.wikipedia.org/wiki/Projection_matrix).\n", "We can then interpret our optimal model $\\tilde{\\boldsymbol{y}}$ as being represented by an orthogonal projection of $\\boldsymbol{y}$ onto a space defined by the column vectors of $\\boldsymbol{X}$. In our case here the matrix $\\boldsymbol{A}$ is a square matrix. If it is a general rectangular matrix we have an oblique projection matrix.\n", "\n", "\n", @@ -567,7 +567,7 @@ "\n", "## Code for SVD and Inversion of Matrices\n", "\n", - "How do we use the SVD to invert a matrix $\\boldsymbol{X}^\\boldsymbol{X}$ which is singular or near singular?\n", + "How do we use the SVD to invert a matrix $\\boldsymbol{X}^T\\boldsymbol{X}$ which is singular or near singular?\n", "The simple answer is to use the linear algebra function for the pseudoinverse, that is" ] }, @@ -1470,10 +1470,10 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.059680303635344434\n", - "4.260227742627244\n", - "[[0.89704131 2.55795935]\n", - " [2.55795935 8.24340751]]\n" + "-0.1831277634928002\n", + "3.3041320306136366\n", + "[[1.03929932 3.07670437]\n", + " [3.07670437 9.7701384 ]]\n" ] } ], @@ -1513,10 +1513,10 @@ "name": "stdout", "output_type": "stream", "text": [ - "0.08362077210702115\n", - "1.8429949116841184\n", - "[[1. 0.6996584]\n", - " [0.6996584 1. ]]\n" + "0.09456011349477329\n", + "1.6600855222624895\n", + "[[1. 0.70183798]\n", + " [0.70183798 1. ]]\n" ] } ], @@ -1571,30 +1571,30 @@ "name": "stdout", "output_type": "stream", "text": [ - "[[-1.25981533 -3.84976606]\n", - " [-0.52570079 -0.62262506]\n", - " [ 0.51707172 3.46973015]\n", - " [-0.47265243 -1.87030083]\n", - " [ 0.73734906 2.46183815]\n", - " [ 0.4918372 2.05531364]\n", - " [ 0.3893239 0.47862383]\n", - " [-0.41237437 -0.85058354]\n", - " [ 0.77622336 0.54697204]\n", - " [-0.24126232 -1.81920231]]\n", + "[[ 0.54316925 0.40968888]\n", + " [-1.8393131 -5.8788431 ]\n", + " [ 0.24613822 0.9957744 ]\n", + " [ 1.38892672 3.93679587]\n", + " [ 0.06540809 -0.84008474]\n", + " [ 0.49841285 2.31081134]\n", + " [-0.49323032 -1.25263916]\n", + " [ 0.08380813 -0.7925146 ]\n", + " [ 0.64078247 2.92814088]\n", + " [-1.13410232 -1.81712976]]\n", " 0 1\n", - "0 -1.259815 -3.849766\n", - "1 -0.525701 -0.622625\n", - "2 0.517072 3.469730\n", - "3 -0.472652 -1.870301\n", - "4 0.737349 2.461838\n", - "5 0.491837 2.055314\n", - "6 0.389324 0.478624\n", - "7 -0.412374 -0.850584\n", - "8 0.776223 0.546972\n", - "9 -0.241262 -1.819202\n", + "0 0.543169 0.409689\n", + "1 -1.839313 -5.878843\n", + "2 0.246138 0.995774\n", + "3 1.388927 3.936796\n", + "4 0.065408 -0.840085\n", + "5 0.498413 2.310811\n", + "6 -0.493230 -1.252639\n", + "7 0.083808 -0.792515\n", + "8 0.640782 2.928141\n", + "9 -1.134102 -1.817130\n", " 0 1\n", - "0 1.000000 0.881323\n", - "1 0.881323 1.000000\n" + "0 1.000000 0.942726\n", + "1 0.942726 1.000000\n" ] } ], @@ -1636,37 +1636,37 @@ "text": [ " 0 1 2 3 4 5 6 7 \\\n", "0 0.0 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 \n", - "1 0.0 0.082272 0.078892 0.081102 0.080738 0.080182 0.072012 0.072246 \n", - "2 0.0 0.078892 0.076822 0.075975 0.076166 0.076251 0.066705 0.067202 \n", - "3 0.0 0.081102 0.075975 0.085259 0.084103 0.082616 0.078750 0.078649 \n", - "4 0.0 0.080738 0.076166 0.084103 0.083246 0.082091 0.077338 0.077407 \n", - "5 0.0 0.080182 0.076251 0.082616 0.082091 0.081309 0.075567 0.075818 \n", - "6 0.0 0.072012 0.066705 0.078750 0.077338 0.075567 0.074693 0.074434 \n", - "7 0.0 0.072246 0.067202 0.078649 0.077407 0.075818 0.074434 0.074285 \n", - "8 0.0 0.072468 0.067718 0.078474 0.077416 0.076026 0.074077 0.074044 \n", - "9 0.0 0.072661 0.068244 0.078197 0.077339 0.076169 0.073591 0.073681 \n", - "10 0.0 0.063190 0.058247 0.070949 0.069542 0.067794 0.068593 0.068293 \n", - "11 0.0 0.063523 0.058719 0.071156 0.069853 0.068213 0.068717 0.068491 \n", - "12 0.0 0.063880 0.059225 0.071359 0.070166 0.068640 0.068821 0.068673 \n", - "13 0.0 0.064257 0.059765 0.071549 0.070473 0.069071 0.068896 0.068828 \n", - "14 0.0 0.064647 0.060335 0.071714 0.070764 0.069496 0.068928 0.068945 \n", + "1 0.0 0.080633 0.085811 0.085748 0.084826 0.083221 0.080341 0.078342 \n", + "2 0.0 0.085811 0.092659 0.093248 0.092932 0.091659 0.088414 0.086561 \n", + "3 0.0 0.085748 0.093248 0.096639 0.096802 0.095941 0.093928 0.092288 \n", + "4 0.0 0.084826 0.092932 0.096802 0.097346 0.096763 0.094774 0.093328 \n", + "5 0.0 0.083221 0.091659 0.095941 0.096763 0.096406 0.094513 0.093236 \n", + "6 0.0 0.080341 0.088414 0.093928 0.094774 0.094513 0.093573 0.092375 \n", + "7 0.0 0.078342 0.086561 0.092288 0.093328 0.093236 0.092375 0.091318 \n", + "8 0.0 0.076213 0.084461 0.090355 0.091535 0.091578 0.090817 0.089881 \n", + "9 0.0 0.074037 0.082239 0.088258 0.089540 0.089693 0.089039 0.088209 \n", + "10 0.0 0.073546 0.081474 0.088137 0.089326 0.089437 0.089352 0.088484 \n", + "11 0.0 0.071453 0.079337 0.086052 0.087334 0.087544 0.087526 0.086757 \n", + "12 0.0 0.069371 0.077163 0.083905 0.085253 0.085544 0.085598 0.084916 \n", + "13 0.0 0.067326 0.074996 0.081744 0.083138 0.083495 0.083622 0.083015 \n", + "14 0.0 0.065334 0.072862 0.079597 0.081024 0.081435 0.081632 0.081092 \n", "\n", " 8 9 10 11 12 13 14 \n", "0 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 \n", - "1 0.072468 0.072661 0.063190 0.063523 0.063880 0.064257 0.064647 \n", - "2 0.067718 0.068244 0.058247 0.058719 0.059225 0.059765 0.060335 \n", - "3 0.078474 0.078197 0.070949 0.071156 0.071359 0.071549 0.071714 \n", - "4 0.077416 0.077339 0.069542 0.069853 0.070166 0.070473 0.070764 \n", - "5 0.076026 0.076169 0.067794 0.068213 0.068640 0.069071 0.069496 \n", - "6 0.074077 0.073591 0.068593 0.068717 0.068821 0.068896 0.068928 \n", - "7 0.074044 0.073681 0.068293 0.068491 0.068673 0.068828 0.068945 \n", - "8 0.073926 0.073695 0.067889 0.068165 0.068427 0.068667 0.068873 \n", - "9 0.073695 0.073607 0.067355 0.067711 0.068056 0.068385 0.068686 \n", - "10 0.067889 0.067355 0.063925 0.064013 0.064075 0.064101 0.064078 \n", - "11 0.068165 0.067711 0.064013 0.064156 0.064273 0.064357 0.064393 \n", - "12 0.068427 0.068056 0.064075 0.064273 0.064449 0.064591 0.064689 \n", - "13 0.068667 0.068385 0.064101 0.064357 0.064591 0.064795 0.064958 \n", - "14 0.068873 0.068686 0.064078 0.064393 0.064689 0.064958 0.065188 \n" + "1 0.076213 0.074037 0.073546 0.071453 0.069371 0.067326 0.065334 \n", + "2 0.084461 0.082239 0.081474 0.079337 0.077163 0.074996 0.072862 \n", + "3 0.090355 0.088258 0.088137 0.086052 0.083905 0.081744 0.079597 \n", + "4 0.091535 0.089540 0.089326 0.087334 0.085253 0.083138 0.081024 \n", + "5 0.091578 0.089693 0.089437 0.087544 0.085544 0.083495 0.081435 \n", + "6 0.090817 0.089039 0.089352 0.087526 0.085598 0.083622 0.081632 \n", + "7 0.089881 0.088209 0.088484 0.086757 0.084916 0.083015 0.081092 \n", + "8 0.088556 0.086985 0.087242 0.085610 0.083855 0.082032 0.080182 \n", + "9 0.086985 0.085510 0.085763 0.084221 0.082549 0.080806 0.079029 \n", + "10 0.087242 0.085763 0.086430 0.084864 0.083177 0.081425 0.079643 \n", + "11 0.085610 0.084221 0.084864 0.083384 0.081778 0.080101 0.078390 \n", + "12 0.083855 0.082549 0.083177 0.081778 0.080249 0.078646 0.077005 \n", + "13 0.082032 0.080806 0.081425 0.080101 0.078646 0.077115 0.075542 \n", + "14 0.080182 0.079029 0.079643 0.078390 0.077005 0.075542 0.074036 \n" ] } ], @@ -2866,14 +2866,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "[2. 2.]" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\n", + "[2. 2.]\n", "Training MSE for OLS\n", "3.0\n" ] @@ -2887,7 +2880,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter2_245_2.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter2_245_1.png" }, "needs_background": "light" }, @@ -3707,6 +3700,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ + "**More text will be added to this example.**\n", + "\n", "## Linking the regression analysis with a statistical interpretation\n", "\n", "We will now couple the discussions of ordinary least squares, Ridge\n", @@ -4073,7 +4068,9 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "It is a conditional probability (see below) and reads as the likelihood of a domain of events $\\boldsymbol{D}$ given a set of parameters $\\boldsymbol{\\beta}$.\n", + "It is a conditional probability (see below) and reads as the\n", + "likelihood of a domain of events $\\boldsymbol{D}$ given a set of parameters\n", + "$\\boldsymbol{\\beta}$.\n", "\n", "\n", "In statistics, maximum likelihood estimation (MLE) is a method of\n", @@ -4177,7 +4174,7 @@ "We define also the likelihood for $X$ and $Y$ as $p(X)$ and $p(Y)$ respectively.\n", "The likelihood of a specific event $x_i$ (or $y_i$) is then written as $p(X=x_i)$ or just $p(x_i)=p_i$. \n", "\n", - "**Union of events is given by.**" + "The union of events is given by" ] }, { @@ -4193,7 +4190,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "**The product rule (aka joint probability) is given by.**" + "The product rule (aka joint probability) is given by" ] }, { @@ -4211,8 +4208,6 @@ "source": [ "where we read $p(X\\vert Y)$ as the likelihood of obtaining $X$ given $Y$.\n", "\n", - "\n", - "\n", "If we have independent events then $p(X,Y)=p(X)p(Y)$.\n", "\n", "\n", @@ -4289,21 +4284,27 @@ "the parameter space represented by $X$. This function is not\n", "necesseraly normalized and is normally called the likelihood function.\n", "\n", - "The function $p(X)$ on the right hand side is called the prior while the function on the left hand side is the called the posterior probability. The denominator on the right hand side serves as a normalization factor for the posterior distribution.\n", + "The function $p(X)$ on the right hand side is called the prior while\n", + "the function on the left hand side is the called the posterior\n", + "probability. The denominator on the right hand side serves as a\n", + "normalization factor for the posterior distribution.\n", "\n", "Let us try to illustrate Bayes' theorem through an example.\n", "\n", "\n", - "Let us suppose that you are undergoing a series of mammography scans in\n", - "order to rule out possible breast cancer cases. We define the\n", + "Let us suppose that you are undergoing a series of mammography scans\n", + "in order to rule out possible breast cancer cases. We define the\n", "sensitivity for a positive event by the variable $X$. It takes binary\n", "values with $X=1$ representing a positive event and $X=0$ being a\n", "negative event. We reserve $Y$ as a classification parameter for\n", - "either a negative or a positive breast cancer confirmation. (Short note on wordings: positive here means having breast cancer, although none of us would consider this being a positive thing).\n", + "either a negative or a positive breast cancer confirmation. (Short\n", + "note on wordings: positive here means having breast cancer, although\n", + "none of us would consider this being a positive thing).\n", "\n", "We let $Y=1$ represent the the case of having breast cancer and $Y=0$ as not.\n", "\n", - "Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$, that is we have" + "Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$ (the numbers here are all made up),\n", + "that is we have" ] }, { @@ -4319,12 +4320,32 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "This obviously sounds scary since many would conclude that if the test is positive, there is a likelihood of $80\\%$ for having cancer.\n", - "It is however not correct, as the following Bayesian analysis shows.\n", + "This obviously sounds scary since many would conclude that if the test\n", + "is positive, there is a likelihood of $80\\%$ for having cancer. It is\n", + "however not correct, as the following Bayesian analysis shows. The correct question to pose is *what is the probability of having breast cancer in case of a positive test?*\n", + "We are thus interested in" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "p(Y=1\\vert X=1),\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "instead of $p(X=1\\vert Y=1)$.\n", "\n", "\n", - "If we look at various national surveys on breast cancer, the general likelihood of developing breast cancer is a very small number.\n", - "Let us assume that the prior probability in the population as a whole is" + "\n", + "If we look at various national surveys on breast cancer, the general\n", + "likelihood of developing breast cancer is a very small number. Let us\n", + "assume that the prior probability in the population as a whole is" ] }, { @@ -4340,7 +4361,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "We need also to account for the fact that the test may produce a false positive result (false alarm). Let us here assume that we have" + "We need also to account for the fact that the test may produce a false\n", + "positive result (false alarm). Let us here assume that we have" ] }, { @@ -4356,15 +4378,38 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Using Bayes' theorem we can then find the posterior probability that the person has breast cancer in case of a positive test, that is we can compute" + "Using Bayes' theorem we can then find the posterior probability that\n", + "the person has breast cancer in case of a positive test, that is we\n", + "can compute" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ + "\n", + "

    \n", + "\n", "$$\n", - "p(Y=1\\vert X=1)=\\frac{p(X=1\\vert Y=1)p(Y=1)}{p(X=1\\vert Y=1)p(Y=1)+p(X=1\\vert Y=0)p(Y=0)}=\\frac{0.8\\times 0.004}{0.8\\times 0.004+0.1\\times 0.996}=0.031.\n", + "\\begin{equation}\n", + "p(Y=1\\vert X=1)=\\frac{p(X=1\\vert Y=1)p(Y=1)}{p(X=1\\vert Y=1)p(Y=1)+p(X=1\\vert Y=0)p(Y=0)}= \n", + "\\label{_auto2} \\tag{2}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation} \n", + " \\frac{0.8\\times 0.004}{0.8\\times 0.004+0.1\\times 0.996}=0.031.\n", + "\\label{_auto3} \\tag{3}\n", + "\\end{equation}\n", "$$" ] }, @@ -4390,9 +4435,12 @@ "parameters $\\boldsymbol{\\beta}$ using OLS, Ridge and Lasso regression. For\n", "OLS, you will notice as function of the noise and polynomial degree,\n", "that the parameters $\\beta$ will fluctuate from order to order in the\n", - "polynomial fit and that for larger and larger polynomial degrees of freedom, the parameters will tend to increase in value for OLS.\n", + "polynomial fit and that for larger and larger polynomial degrees of\n", + "freedom, the parameters will tend to increase in value for OLS.\n", "\n", - "For Ridge and Lasso regression, the higher order parameters will typically be reduced, providing thereby less fluctuations from one order to another one." + "For Ridge and Lasso regression, the higher order parameters will\n", + "typically be reduced, providing thereby less fluctuations from one\n", + "order to another one." ] }, { @@ -4409,21 +4457,7 @@ "text": [ "[ 1.0169643 0.27924636 -1.4087793 1.03308408 0. ]\n", "Test MSE OLS\n", - "0.958228616652075\n", - "0.001 [ 1.0170259 0.27852549 -1.40702 1.03193199 0. ]\n", - "0.001 [ 1.034342 -0.18063928 -0. 0. 0. ]\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "0.021544346900318832 [ 1.01825571 0.26412372 -1.37186301 1.00890601 0. ]\n", - "0.021544346900318832 [ 0.92280994 -0. -0. -0. 0. ]\n", - "0.46415888336127775 [ 1.0344707 0.07160764 -0.89928965 0.69843037 0. ]\n", - "0.46415888336127775 [0.48019541 0. 0. 0. 0. ]\n", - "10.0 [ 1.04529095 -0.18224665 -0.15751596 0.16831012 0. ]\n", - "10.0 [0. 0. 0. 0. 0.]\n" + "0.958228616652075\n" ] }, { @@ -4435,7 +4469,7 @@ }, "metadata": { "filenames": { - "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter2_313_2.png" + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter2_316_1.png" }, "needs_background": "light" }, @@ -4496,8 +4530,8 @@ " # Compute the MSE and print it\n", " MSERidgePredict[i] = MSE(y_test,ypredictRidge)\n", " MSELassoPredict[i] = MSE(y_test,ypredictLasso)\n", - " print(lmb,RegRidge.coef_)\n", - " print(lmb,RegLasso.coef_)\n", + "# print(lmb,RegRidge.coef_)\n", + "# print(lmb,RegLasso.coef_)\n", "# Now plot the results\n", "plt.figure()\n", "plt.plot(np.log10(lambdas), MSERidgePredict, 'b', label = 'MSE Ridge Test')\n", @@ -4512,10 +4546,13 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "How can we understand this? \n", + "How can we understand this? **More text to be added**.\n", "\n", "\n", "\n", + "## Linking Bayes' Theorem with Ridge and Lasso Regression\n", + "\n", + "\n", "Using Bayes' theorem we can gain a better intuition about Ridge and Lasso regression. \n", "\n", "For ordinary least squares we postulated that the maximum likelihood for the doamin of events $\\boldsymbol{D}$ (one-dimensional case)" diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter2.py b/doc/LectureNotes/_build/jupyter_execute/chapter2.py index 99539ec90..73becd4f1 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter2.py +++ b/doc/LectureNotes/_build/jupyter_execute/chapter2.py @@ -7,10 +7,10 @@ What is presented here is a mathematical analysis of various regression algorithms (ordinary least squares, Ridge and Lasso Regression). The analysis is based on an important algorithm in linear algebra, the so-called Singular Value Decomposition (SVD). -We have shown that in ordinary least squares the optimal parameters $\beta$ are given by +We have shown that in ordinary least squares (OLS) the optimal parameters $\beta$ are given by $$ -\hat{\boldsymbol{\beta}} = \left(\boldsymbol{X}^T\boldsymbol{X}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}. +\hat{\boldsymbol{\beta}}_{\mathrm{OLS}} = \left(\boldsymbol{X}^T\boldsymbol{X}\right)^{-1}\boldsymbol{X}^T\boldsymbol{y}. $$ The **hat** over $\boldsymbol{\beta}$ means we have the optimal parameters after minimization of the cost function. @@ -33,7 +33,7 @@ $$ \tilde{\boldsymbol{y}}=\boldsymbol{X}\hat{\boldsymbol{\beta}} = \boldsymbol{A}\boldsymbol{y}. $$ -The matrix $\boldsymbol{A}$ has the important property that $\boldsymbol{A}^2=\boldsymbol{A}$. This is the definition of a projection matrix. +The matrix $\boldsymbol{A}$ has the important property that $\boldsymbol{A}^2=\boldsymbol{A}$. This is the definition of a [projection matrix](https://en.wikipedia.org/wiki/Projection_matrix). We can then interpret our optimal model $\tilde{\boldsymbol{y}}$ as being represented by an orthogonal projection of $\boldsymbol{y}$ onto a space defined by the column vectors of $\boldsymbol{X}$. In our case here the matrix $\boldsymbol{A}$ is a square matrix. If it is a general rectangular matrix we have an oblique projection matrix. @@ -344,7 +344,7 @@ example ## Code for SVD and Inversion of Matrices -How do we use the SVD to invert a matrix $\boldsymbol{X}^\boldsymbol{X}$ which is singular or near singular? +How do we use the SVD to invert a matrix $\boldsymbol{X}^T\boldsymbol{X}$ which is singular or near singular? The simple answer is to use the linear algebra function for the pseudoinverse, that is #Ainv = np.linlag.pinv(A) @@ -1917,6 +1917,8 @@ X = matrix( [ [ 2, 0, 1], [0, 1, 3]]) y = matrix( [4, 2, 3]) x = l1regls(X,y) +**More text will be added to this example.** + ## Linking the regression analysis with a statistical interpretation We will now couple the discussions of ordinary least squares, Ridge @@ -2133,7 +2135,9 @@ $$ p(\boldsymbol{D}\vert\boldsymbol{\beta})=\prod_{i=0}^{n-1}\frac{1}{\sqrt{2\pi\sigma^2}}\exp{\left[-\frac{(y_i-\boldsymbol{X}_{i,*}\boldsymbol{\beta})^2}{2\sigma^2}\right]}. $$ -It is a conditional probability (see below) and reads as the likelihood of a domain of events $\boldsymbol{D}$ given a set of parameters $\boldsymbol{\beta}$. +It is a conditional probability (see below) and reads as the +likelihood of a domain of events $\boldsymbol{D}$ given a set of parameters +$\boldsymbol{\beta}$. In statistics, maximum likelihood estimation (MLE) is a method of @@ -2197,13 +2201,13 @@ Assume we have two domains of events $X=[x_0,x_1,\dots,x_{n-1}]$ and $Y=[y_0,y_1 We define also the likelihood for $X$ and $Y$ as $p(X)$ and $p(Y)$ respectively. The likelihood of a specific event $x_i$ (or $y_i$) is then written as $p(X=x_i)$ or just $p(x_i)=p_i$. -**Union of events is given by.** +The union of events is given by $$ p(X \cup Y)= p(X)+p(Y)-p(X \cap Y). $$ -**The product rule (aka joint probability) is given by.** +The product rule (aka joint probability) is given by $$ p(X \cup Y)= p(X,Y)= p(X\vert Y)p(Y)=p(Y\vert X)p(X), @@ -2211,8 +2215,6 @@ $$ where we read $p(X\vert Y)$ as the likelihood of obtaining $X$ given $Y$. - - If we have independent events then $p(X,Y)=p(X)p(Y)$. @@ -2249,47 +2251,82 @@ evaluated for the observed data $Y$ and can be viewed as a function of the parameter space represented by $X$. This function is not necesseraly normalized and is normally called the likelihood function. -The function $p(X)$ on the right hand side is called the prior while the function on the left hand side is the called the posterior probability. The denominator on the right hand side serves as a normalization factor for the posterior distribution. +The function $p(X)$ on the right hand side is called the prior while +the function on the left hand side is the called the posterior +probability. The denominator on the right hand side serves as a +normalization factor for the posterior distribution. Let us try to illustrate Bayes' theorem through an example. -Let us suppose that you are undergoing a series of mammography scans in -order to rule out possible breast cancer cases. We define the +Let us suppose that you are undergoing a series of mammography scans +in order to rule out possible breast cancer cases. We define the sensitivity for a positive event by the variable $X$. It takes binary values with $X=1$ representing a positive event and $X=0$ being a negative event. We reserve $Y$ as a classification parameter for -either a negative or a positive breast cancer confirmation. (Short note on wordings: positive here means having breast cancer, although none of us would consider this being a positive thing). +either a negative or a positive breast cancer confirmation. (Short +note on wordings: positive here means having breast cancer, although +none of us would consider this being a positive thing). We let $Y=1$ represent the the case of having breast cancer and $Y=0$ as not. -Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$, that is we have +Let us assume that if you have breast cancer, the test will be positive with a probability of $0.8$ (the numbers here are all made up), +that is we have $$ p(X=1\vert Y=1) =0.8. $$ -This obviously sounds scary since many would conclude that if the test is positive, there is a likelihood of $80\%$ for having cancer. -It is however not correct, as the following Bayesian analysis shows. +This obviously sounds scary since many would conclude that if the test +is positive, there is a likelihood of $80\%$ for having cancer. It is +however not correct, as the following Bayesian analysis shows. The correct question to pose is *what is the probability of having breast cancer in case of a positive test?* +We are thus interested in + +$$ +p(Y=1\vert X=1), +$$ + +instead of $p(X=1\vert Y=1)$. -If we look at various national surveys on breast cancer, the general likelihood of developing breast cancer is a very small number. -Let us assume that the prior probability in the population as a whole is + +If we look at various national surveys on breast cancer, the general +likelihood of developing breast cancer is a very small number. Let us +assume that the prior probability in the population as a whole is $$ p(Y=1) =0.004. $$ -We need also to account for the fact that the test may produce a false positive result (false alarm). Let us here assume that we have +We need also to account for the fact that the test may produce a false +positive result (false alarm). Let us here assume that we have $$ p(X=1\vert Y=0) =0.1. $$ -Using Bayes' theorem we can then find the posterior probability that the person has breast cancer in case of a positive test, that is we can compute +Using Bayes' theorem we can then find the posterior probability that +the person has breast cancer in case of a positive test, that is we +can compute + + +
    $$ -p(Y=1\vert X=1)=\frac{p(X=1\vert Y=1)p(Y=1)}{p(X=1\vert Y=1)p(Y=1)+p(X=1\vert Y=0)p(Y=0)}=\frac{0.8\times 0.004}{0.8\times 0.004+0.1\times 0.996}=0.031. +\begin{equation} +p(Y=1\vert X=1)=\frac{p(X=1\vert Y=1)p(Y=1)}{p(X=1\vert Y=1)p(Y=1)+p(X=1\vert Y=0)p(Y=0)}= +\label{_auto2} \tag{2} +\end{equation} +$$ + + +
    + +$$ +\begin{equation} + \frac{0.8\times 0.004}{0.8\times 0.004+0.1\times 0.996}=0.031. +\label{_auto3} \tag{3} +\end{equation} $$ That is, in case of a positive test, there is only a $3\%$ chance of having breast cancer! @@ -2310,9 +2347,12 @@ We will play around with a study of the values for the optimal parameters $\boldsymbol{\beta}$ using OLS, Ridge and Lasso regression. For OLS, you will notice as function of the noise and polynomial degree, that the parameters $\beta$ will fluctuate from order to order in the -polynomial fit and that for larger and larger polynomial degrees of freedom, the parameters will tend to increase in value for OLS. +polynomial fit and that for larger and larger polynomial degrees of +freedom, the parameters will tend to increase in value for OLS. -For Ridge and Lasso regression, the higher order parameters will typically be reduced, providing thereby less fluctuations from one order to another one. +For Ridge and Lasso regression, the higher order parameters will +typically be reduced, providing thereby less fluctuations from one +order to another one. import numpy as np import matplotlib.pyplot as plt @@ -2367,8 +2407,8 @@ for i in range(nlambdas): # Compute the MSE and print it MSERidgePredict[i] = MSE(y_test,ypredictRidge) MSELassoPredict[i] = MSE(y_test,ypredictLasso) - print(lmb,RegRidge.coef_) - print(lmb,RegLasso.coef_) +# print(lmb,RegRidge.coef_) +# print(lmb,RegLasso.coef_) # Now plot the results plt.figure() plt.plot(np.log10(lambdas), MSERidgePredict, 'b', label = 'MSE Ridge Test') @@ -2378,10 +2418,13 @@ plt.ylabel('MSE') plt.legend() plt.show() -How can we understand this? +How can we understand this? **More text to be added**. +## Linking Bayes' Theorem with Ridge and Lasso Regression + + Using Bayes' theorem we can gain a better intuition about Ridge and Lasso regression. For ordinary least squares we postulated that the maximum likelihood for the doamin of events $\boldsymbol{D}$ (one-dimensional case) diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter2_245_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter2_245_1.png new file mode 100644 index 0000000000000000000000000000000000000000..dc3a3ce3ef5be71546dc15552f9becd822286f37 GIT binary patch literal 9236 zcmZX4c|4TS_y1Ui?Ado}5L%G2Bw5NRRAk9g)?(~ttYcp)dn%Od6w1CYqp^%#8boB9 zu~n!TQ;Tb>prRJw=>1T>U`e7V=oFKOy{LmgKKM&UjrvjXOd|gpqa5?aE{*;TK zAKLe_yu9cC4v<6nILr6HZpsEin9zDwz7Plt_wk1Yl9??4fk;H@Bd(eSX0MT4)6K@4 zxBnc`!3HJ!zvE8{ygGH_#DxTz%oDKAtGW_#Iwv`~rLPzoKU8A+gW$-e${;QZnLUj| za3bj6iXkp?3es~%7oTLLuc5iBjm)KXm5RB1_!%|ZP*z5s^B*6-+Ym6`_U32oxO>&T zPef5pPJOMri|nUhu#r1a{G9rxXDKk4s{zy&A$hw5pBPtI@_3Q`&Yi9pwpQIlKJUJh zoFz(p=Pj$hl7xB3(@dak;3*+bON%T3gT0E$2s5_Mig7_8Mqe_w#OY<4{70m$GK1aA z9Cb{s&kut}&lx@pkD9iXm6ej`^h>w#(jHM=Nr}SHD7$7KDgR$%Z-BsbE24K{HZjNl z_tzRQ=U$_SAVwtl>5IQQN;8d~1y>(4oF%`?&KSwRN_TMPObSBQZk~phs`O^PK+mUW z|722}UU~QqO`}JY=ZyVt^T{^RQj6Gc&sf&hTj+#U@IyJfO%@(x0^ZsYl@0C_OFu5JR(I|9OFSop zw1n+#bJ=O3X0Nm_#0U@g4VmqP8^l<(&mMGfkerA7@)SA_l5Y^j!<5<&azltX3}&4; z)#~p~MGg{i{-03bPIS91C)h1~?@!dMxbDKTKiv*qyLZ^*4Im5K*P3Q#4N&!^;PP!W6g-6kprE1G6MOkBhT ztS#n0PB?T<$;r7L5D?(Fi!g8+6xvyl*ja^J*3h=?z&6u;BLZUBqxtQKd&;q*4DI_) z*!J+Sl{ezSGD;}+26vr~c*xOiUUk3)N2eciLaIZk2(sE_4PEKoPwnF+sdH^;hn;&3 zZif}~!(;!pPL5EDm_ucALdeZ6|9+KO`;F*671zqWDcAgDs?M3NI(W-%NONzs<>5?& zy9If}@8@VVr^NN|8bda1ZydT}h?u#sr=sdxBO^TQ` zf;a8|ZT>uORr`d5!}#`zVierJo^M3W4f#*l^eInRn?8sE5u5z#RZnL$%g^bCb94-> z_LJ2P9y5(eLdtHGF;k#IW+snN=$dxP!XSA+l7Bj;61nV_mh0{&v#ka zKfS`A4ydPUpI^9tttMMboKHSxq@e50#{+7Y4oCDuc1?jy<;Mlj9@*fooo+o0#;(9z z#$m z#i91{uP!tOc^F4n^t{@zV4ku4q)3ETJpWj{1a~^rCyE@&F z_)JuNiFfB*RZEI=Zr~$1n-6H9HCY-aV z{GNBTFMYSn{KwvUyYu$yl!)Kwt1ZAaM4-^E>`UHx6X)+YK2@CyO}(9^a_rXd-A{II zLpTwv>zDZF zf`$4AC<_nY4;;F%nWs0Lt6^31018JEj^WM@T#N^h)5{#394vZmI{<8%YzgvB?gPWJ zSmdygnab|8pPNv25EYL}{JYy(E2Bj$5IRP7$6sS5ke2Of_w%+b>HOU7xn4wUvXRz7 zy&3e-S!+)a*ftnK8jd*h)jZfVB|HVZ$+w$1+Ztlp8th+cQDyk8H_56F$p*~3@W&da zEs$edf5J8hY8$3(9lHP1s0)4LTQ8bQLXYad$Ptx$u;ljT4AD{iRTX}{s?Vlw>2O9^ z<-Gs%RvADmKfc~aIejg+DhJe6)%b7gHyk~$&A=|Am!}yPxVP4{5jhZWl!%`5H&Q6o08<|wIo`%m;!&m8;B7irwCOEX?( zF^scoaJvhX6Pl~C4$3Lty4$?$P!5(Q>E+t zf14i<1Mx(w4S2ZN^ga7}D^(0KtaTJR>mIzl`VmmrMldli^e@x>Mh{uv$|PacB^$GF z5T&2?gOkHnr+E|9o_KHei9LRId*M&tLKez?NJIv0b93MZw2d`W#q-|x=q@M=-wVK8 zfX0p?)rnzUC`>h^{o6P1i;}qUjJU4mHdfqmp@~>K5p5m`!eX>NB4WA;WmIL``gc~m z({gXz!tQn~yXa)C6AMrsYTy6jv`EDMAHUVfWRM+w29+mXmRW=W_xmpAV0;$irAsa3 z7?Hot1=JLu=3v&vmY!IYF?@R6MaXc_4D1j*>!W@CR>3Vcz`?eGUC)bdz8?MH|2v)Q zE_;?L>K*VqgcxyjP;T3r87m%sX~|6Y3l*;8IVTqUlR76Jyh8c_l+|9Jb(qe7nU!$? zO@%F=vuL{*IMr8H)ixl0v~vT9=rO(_TI9f^uTj;?pi1IFb|i`0ZF*cPL7%0#=0_YI zZm!QAR{$Y3 zGSjF4zuu$q)1(dW?BrDAMY_xzYcRz9y93;TqTuZ&3A2!zXJJhvoqZD7wGtwh^-}z; z>vX*=paDr~X=$>IPoxjVFGZ=ri6uK!aer@)Vs`{>nyX|fzIbz&)~@27@)|_aL~yK> zmv8&>))+cmQ}R8Nzhq>hW1q z3h?_6km4Crw~cdeB2rd4A|hwJSjw6YcZa>-C&Lit^MmCV(8`|$T{t!`);)ZCIWJCf zgt9=%n#2Ei-G%jbXr|FuGv1!RPq*H@$3bpm31nH}U)V)R3U04PacG_}J)IXGyiSUj zEJEz&SkV%lk!+C__-6cmCu7Fc9ebHOmAnGFlJUf@7$NR6Pp;Vo`m*WECWm2&VX_nm zMM#0pl73Zg5!*-7m?6~(E4%UB72#C7!@~FE=ks$GF}rI*Y5GCQht%KXhs4%Y9EPrq zPBZ>*D2yw)!F^$W!(q!UEU<(t-j6~c2;ugckYBSE4l8?P5NESALgs5fosyP5FA&hp z=`EdNlJM_We|NXQzTy03%-qR8yNJTKB#ij}!MelQ@A1;TIoAhHjnb}@-{06hgmKk~ zs@snietz8k?-%xqo}g=ZY8=n{{Nn}jqZ3_9NZ>ieQ4wiQ+3CreYHx|LVopA;RD&C! z*xXhQyUZ#?|NIIUhD!|sYS7xIB;l^w_C9fOmEJUin-7nl%%EcLmf0^NifuK>NlhGz4+iX=LY zC(6!;lvr%49@I37|D$7>dB@6#18V%?^Ca~oGljvooF5&ntk5KX^jd?rbn=N}e4=CR zzpaH1#&3qFIY&+=>Am|mcNDEU8>b$P(1Y4aa#!4gw*TX8*)$>QOI`!^X1ILUn`{;K zYE>pna-HT4Oukj;%6(;9z&g;uGE;jrjk!hdogn;7`EF&y>=)ZY=q%yz3qez{P70|q zOQ?N&B`@yfwbo$tM=A8*k3fSs-QxgdLQA|@BYQ@b;v*5w1; zQ2pzD<@{*xKGDe{kLR_}UYiC8O2*gXBbc%uAe9FR zk#&(poViAb1bBdlK*&*#{GuUDpXZ0beh5AhzlV>F_tg*^-$giTA%|nW7(Q=F+I8{N~!?BQ&=}%wm<=ulda-2;M>a~ z+fxckU*6pYa4-Nkeh!G{h6{c0i~9B~Q9Y?!8$f1r!4e`KA@Vgh`6-C9l*-)j%O0g+ ze|mto(_OxJCi^Kno^W)I{ShT40UD>mGIr6<+lpzR85$kwPgZa?1Zp6UB_jugu)E0g zD{25}H~`vjqm!jNMAbSx2)z5hBlZI?EzpNy^+E8B4)ELIwT=W;#?zj@2d_2M+@%*% z{aSgO)+Mu0FZkyOoq5Hh(!0QNS#1pQup^T+s2{9y$*E#l9!kyR*9TLI0xGP8`N@ahfVyY?inU~jvTj0cWlBjTaSYW1Gwd~1Y&_po!xAaUnmdr%TBR!zJqeqzG8~FV2oc1d{B*oZyxgv;!MV`#pJbVy(|579 zQtYR25JP6yANFc54^?~F0OXhqM6_q{m?6Jj_^ONNd4eE7sQ&xsqn28itf3wi9G zmuKY#sk0@F5x!A-1mGkFeCPJ_6#P|fbWv}W(XA_$3#vXFUPphxve~o*;e_`;p zkaoR?DF3S_4)IX~#OZ%{2|O)g4);JNGPQ1rOOaCpL;VUrA{JRLwa&kpF`44B8T_K30?`Bxp9z%&R=hJ(X&Co*9L`!r!&*YZIa$c2J`-y_D;et z$Pq*=K`}c{QpB4+bnC9=kxk)CgR>NM7NQnQAT6GxnlY8zZxY|%mZ_4Y9%ZHo-m*VH z@KUH_iF}4b%n~nW$3cWuW1@SOhLQlED@yV1ltrrGPvy7cPs{H3-S! zr-GEfpQ?SGKlUZ?+!;NeiHfr0`Lmye&oYkYN~V2xiwW+0T&zf%d|Yh;$^Yd+11@AP zAxQ!jGS$re?l#L!I(P4jvIQw}wE;A3w7qoeZ?>EfDel`Ilhri;O68@TEJg+P7w|UP z_van*8Nu$zoDa3hAlMgP=8TljoKl3}cSG8+U#Zn&n5v;O)D#Tf1n`?(+q*y^x6P28 zv3tWz{N`G8$X5;*c4DIh$%hZ~HL5owM-v<~eTSmdeDlpKK=P3?8$HQ}G#6cAv!<%^3CS|JDvxY5Qtc+hWKM;`v+`0?O>?y#>jeubGYhG*N81*P5L$PfNZfhm zq?ABQiZ&gQ#4$DOUKgdNZGnjxCWtY#VW@LF)XddYk36l2aI3$Gg;l9W40Q?!2@_WM z{dw|~Zo?}j2D={8Z%MCM9_;v(owt!CQ@J0N%Jtk%a{tUJn3PrPZNs_&^L=`hz3AAiWt=9eC!>gqJXrl-G7KX&2ty z&N?=g&lOOQu6FW69ayb+S8P|NvIDDaVfI^cN1|Sk-`J0{YUt!7@VA=At|TE_5W}ob z2N!vWDF?=$Vqa$xy_y~|p_gbR7eFrU;(g`%m zG>+nrpSdX?U3KuUCrfAaICT*X)u7}y;b&@RfQHCp4ZLhVJjnu;3Wu$vq3?3%TU`hC zmj8Q;nHAJ*{or8*_=&BAQJ(8Hsx8W@Ey7zv4<4)79^5|;NjA}%vECTHvZ^?KVW^=^S@{#GI1FCV48^xtENuI^r<`W zEEICX*T)C;Od^FgO+J>}h}Cyq7shjpY!$a%JM;`qPU58{Z_IE2b@h6ZkgnhjN#$GV zCpC$FI}69o>u^O8;!j=5)(+Fz5rUC%gP1OJwZvY_JgPe0=Kn0mA|71L7 z7`kG5fPv}esXXzW4HKOg`?P<5J)ZPVtVNGUNO^P;*KG(=8EkEq?X@*+9TJQN78PE* zoxvqnS!pAT;o=0sn7^@0T-W}ThVFoRzvmp$HBV;Vh&WO%7Tpn{!Dsb|XbKc}bEY2Ah5>RkG>1~1pR}u-g@b$jKn2cwiw5;^N>C8%P zJGeT3KP~&kI%m>B?TKO86zhmX_<0*LK|BpB5 zj8Zf|o;9&0;bOD&5=KGaV5PBnk{FQnPS;=nbRM#aQa9%<+8+&Tekm8!4+5wcHH4{K z2MCUpF+SI-WGPGetHdC0a*s-+6Ay({wQvP^aOr#h>PURqJKaF(mr(GN+5E{39B*RG zVOyvqr1Ata%OvcYXvPJ`tPaPpDL7oi&b^@N>&6hCp%PWqhJrQ4p%F;jW{^RDlW#y z!(CDcnz0o3VXe zT0;P|3<v+IFw4A8(tl$LLx0A0uA+@$PUrHqUxRB?lp8yL>#cFpklxNG)hZgIn4}2;Q=(9t-(&w%OQrgy@YYH)U5;ERg;!z3k*gR^;4K_}sSZm(HE_RHK-hxz5?(a~_fx-FU8SPqwXXEL|PUL?0bc2=I`m zjE)=P(R)-PcP)<@PKA&oOrG;$xs@lz>`~d9)U|t>03B}43%T_vP2u5ntQOinP@cNK z9k6Eq_dEJ1+qaWyKb-ryEkrdtK@JJuVc=Z5I%j+U>nx#^mL>9G=y1CniVjB?X|Yyi3;I1u-EAqLSAC?|HVVHtwSE}@r>ub;%yTPb{Sb?6c=ezak&Hj zABg_etJ`FuZ%wDQc*12SV7>7XKU0>{R8#Bqr2W0Y%pXn5ZRF278#DJCpCk!W!`Hzv zO~n2;@z`Y_R5kT)Gf2O_-y1C|K2Wp&UVYrzD>z>IV~)A;xnOGE?hBnxYJo%7{|i^keu#vF$E@U2kA8r#u}Ur4TJNvzgJ+epPd9gY)EpVw3Jh3Ar1wZr61Do z-9;cIJ84nKBrXjMp`76}-RJdb7)+lFw8fnG$?i*`yz?lsrgG;-e4HMQ@z7f<)@*_C zy~McwL)Q{o&SWJ({IHaFpz!#gGdhC-#vFrB1fNLxf!)6`_-W>+rp7PJJpOx)UYsx{ z&lVTQygW-`>?ADlQ(Y@B$y!Xw{O5AKC&xAt7gs0?c2Sa*UKv6U|6u8ObeadQ|K~bW z;8Qdx&=ME7dLbtm94nfDlnD+64L$+&o6<9~th!R3Kg-QvQ?2VkHue*2;dK%>!|!eq zJf37+xapYn64&BOB&<3kx05s0F4kGTug{oLLvG8%_Xq@X9Q@{0aGG?hw2VRJIu5SpaDL1?>;EPhtLUo6tN^fel1bxCXV;rk6v;`R z9U`eiOrt^`45j||&#%BW(P`33yNFXSf1%aG;tDcL{Q2jHxhLOsIoZK4m) z*S$hL6->k+$rsa^OQ)0sHR+#!%GKn*`1G~bdArx9i_}2VD8?*qjxs5)nj?`+E7N>c%4?ZftqSIrj2y<$HL`ejPLTo}H%w@w zm6P%hra;*k-t~#XdX#9|&Qb!UWbc%pWr4e2o#b9@O)r21>6}iCL`6-IGa3YbwA`bg`{xSvg~l&Qx0K3NYT1(1*4aS|e3Xtk7lu(BHp<5Ik&K+*JNtj{o*9 ze%VorP%iWa``nlMdU41!9fwk=Q)QY^SZvWF@vgWdmuao=r;*#`I6HG=X}zq|LYSk^ zBw<_#3kG#{$N$4a#e@)j!s;X~$FPL1inwZ>?}>J?ZGWox;!x_NC7QlWtDA6j0RhOZ zcr{aly3aTvy7T_jBYsS-HS((rCM`W<91?tTWB3hj?kD*+_m&$(p$?X)UT7R;oY-Xb`Zy>38IXTlklIeX{Nh2DEmI${Dn zK9|=xUTFW{68`To`&mj#3cEGOCTI7be#ZU=u&&0vM9c;j6$yeqx zx(6QlMg}7T5O3G6*l;ewhrN^`@%eCG&%#zA@rLq)!QBM+ia?>>fsYi`WN~D)CE9((hGc z6UehdF+qrvckCV)I$ui^f0v>0g-Mx^H9e@j&CWbh$+C@4@cOImZrenCLn+mZs~LOz z`j!dt*Ngj1WxKpim)^W+f*uxp_J)S+hcI86?V8Pb-}yU=2dAm`U}}=A{kog1!@`mS zKbLG#hyW4r3rTmtou#Y<+YDXik}I@L7f94%y;R4ut_;P(E>=b`SC5b^xEs%T7!%xo zrL-d``96n&K2^fmX_uC-cLqG6;~u8c`Z)KENwM-D*dw*hc!bDfoS6qS2(mMGI+r z!zfQa-R+xY`UXW#$OclF@8a8U*7@#16Gui*gD^f++&LvjsA64glq~r$!rM7{;3~gW zkz!UR);>r-~RO?|+ue|9;u!AVNt%G4j^Mbvf`2 P9f-cJA)@k{{iFW_5~=1b literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter2_316_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter2_316_1.png new file mode 100644 index 0000000000000000000000000000000000000000..07584f6bbceb100d07a9bb9be87c5a061d00ba42 GIT binary patch literal 10353 zcmZ{K2UOGDw`J(PBuGcm(4>Q*t29GZK)?^_O%1(BB3+u&1*Ax?QdNqAQUyX$Ktl+k zBGLi^0qKbJGB@AMYc#6kY0faK=$Kpzq$+3U!;V~G~djKa@Ki6pI`x+k+B(|WbX$C%ha{l!(X=SoadGVr| z;Xkh|%AI;*Zq?eVE?$H)NXjsFeN5tiVbvfiE-s#JMYvz~I3Mjf-BFOluk825K#bL zR>#rkvjG$L606#VmDCw4f=n)gP)|4c_jzYYO^<;#2a>ZkwpK zYPhss={4Ty`SY!KfU1Z{H29xF&i(5}A}wmHhhLbBDfvt>V@FIh26J2r z4Zxu0ou3{CCMKkPQW0rDtVmc<@!5w&?yUWXiNXRT;&+=FQjyb7g(@=mrJrg~2cPR%_dLtaXzG3`T41tlEPWj}QKSKAz=Q@jXA;7U0tTn86k^th2p3 zp7kq5<6Ojg$6@lvDmvG( zdXXCIca+6G(;YW*72+Gc#xHsPcs)USzc-us_2oZ_LmH>~4-a?lPX&?g4~S%)87n|Y zN+w=jYkt%;6`VLXch^4r_(2StAU;FfoCkqo=2!A0_6wy53kf|cO;OLa@op)qZKt7xtAC9%4T_VS2?cZm=s%=duO;%%!00BYqG&Qv=`Rrxa@kVLL)sUUJ z$$)7Z;v+V7$G^Y7;qp~EB5=4H9s3J;aHg(Nhj1fnYd&s=P)zL6okidx+rLuee9z~jS($lZ!*(C+E?gXt-PH%_#}8nHjfq?4?%~dAZ1pSiN>;E` znafZiX7%SYokXtr6KC1@QTvc2aI5klF4mfVv7&|w?bhDoRX?T(np7GW82-&~z^f&e zU(Y(%=)E@4B4zzHE?_Ds*0fZ2b3V%<^Kz+v_kGrb-Prb!hv_%X%#Z<z8T$9z?bvFM>5i*D z9s8moJ1b6L;QK%L%MgKw+rwY0tF2nijaCs;i~GS6f{(HfGg0+MwdL^`%HrKe(8ale03>k`A(={!U-AS?BH4 z$BIFhaCvkoS80>@^+E$Az@A?_U0VaI_MCI*2vr5MRaiBk-g=E^0w;oV@DoST)O0>3 z8poK_!zzm!zJTe6%3!jqbDNpcHecF90&)g|u`O^Bp zBP-lF_npV6-$z~->Wqs@?ZvwR}B z-{e8qtmS*y5n7us>z${`yBr`>s+U6EVt(Lk{EY%N#YKI1+VqG?k5~~lL)=N4@`LFG zeRhz=l_9%pfL>C&L;rG4gvi>4sctWEe6!PA5s1t&l5dbkWm@zfSfh55a~?XOJaE-M zW%Y4!w_NF1HP@$ZwyWL9r#C(KL_SR{v68YN(NcqbuL)hNN`MA-m0*`+St!QkDs z!M)92ON0_CwksPDwyWFhCY*olnfP0S0R_jdHi0AG$d)VVKDQZ1?fgS-Z^veD?@pB! zVo@Pgk{aJ~?nCt{v{~5fkWITRwgfK=aDLpUvjAwA)QpVY)(6WF*O3YqKs`d>69AfN zU%%cmR^Pu1Kxzocg06d6b{|EuY_(A+QC?nNK!p?${NBxx8kxnT+D_MQbH8{-0^Vc?MN}@gXe_BQYWq@ zr>Z}naJaItafSt(hV)4cQIGaM^c5QxJjlMR&LcfU$!i(?LE8t<<%>X3( z>*p~g9mhN4sHHs0sXL~els<2b7x%c zfy6|(M*c&8f*WgzQejF-gk&j{_TMQ1t0l$8~8a7xJJ#sEdu50vq6f7zQ# zfb+4;?H6Fmq9!pxe;khjT_;o)6?763wQsig-7{8ODl5Jd7Z=E{ge7+Q&P4l;dudgQ zTi7bF#K+y@QB^(pJ!AOnNO=31#L1|qtpF6DB+svpT>x*d%kB^B+|m3X`wtxc#(CAc zdDp)A$WH@@d2$of01UBIbVpg?Nqlnm;n_3=RlQ_|4`aRa5lbrS;rGGb@1rTGxP?KU zhnv|;-tNA}rj>HecTf?mt;k6UP2Kv@#1_08Gj~^G<_5Fb$kRCepeH;OZf#l1;6c&oWMjF)?v# z59WdexX*%9WPFbm>_!ajsm5KG8w${=1kNzy%gG_s3Fe2XUD?tnxOHnAU4z5RD?!5oYaTp%BB%;yknxANygJ=*(HGaT-N%&Dw znp1PBK@qiMK<2=VA3xune46o|Jdzt>IG@&c**p>-U8B%Y8tD@0jc;j)(9o4jjMLv_ z71(e%QFO;L$wpWa;-ABvjjrKabWoXQy`onm$M7wuj~l09FQnjbkxRh}fGOW^(gLO= zxIGu`N1U&ud^hM>k|V9$FhzlarXrjCM4kvRTn)x5j#3whzKptJ6-rHDY-X>bl~Qal zDL^l8`M9E}A}An9Fk){~91>&!p-ov>?9hW2cH^lYHGkx80lF<8JszQ>`6o&VgEzw9 z-}&R;856;nO?dzyS7lwS6jDRw>L9mN9J!`r(IPtGpK zW_N6VDITFFp9iIMc9CA-4vIe#s(BAWn)rind52OGv}Bosq!*y0{S`T(3~hF8p|T~q zNyK=uC}WOneZaA%QC=S@?+889LYC=Yi|r1W?&=e~Mh-lN8=(%$(r~kEHjT$Fg_~&h zfJJspuI3B&IQ>}wwPDA^?GdN1RbX}FJ1p_~vn|Uq8Z)>rk@D`A=rZ%taW)Pz;sL(c#Ka;xG;4ckS8D@FG!L zkq^w!IA-$~ND6Z@(rnH|(g6)Q6~gxWD;Rcz3>Uo}BG(;PEH!$!%d=<;L>H>Sx}M|j zJ}WbnmtlXU&=rV^l2gTboKW&Q>}u$d{MTw;Om21=xcGQG!7%!vD^!nYwPIO8hFVF0 zF@SrxtCgn& zY=g+QHjUnYCy_R9BM#Tu)QoIz0+{M$B^^;CczHL5^=_6M65?mIQ}w7({&t0jqGaBN;%h3)!la2V*{u#N zC23+FyMD3baPd|#+LS-r3bt`^Lu=DiN7(7^y?xo!Tt(1CzpS={qQg}o$#jbz=Kv; zB4JtaElqQIM4tw&x(unqh6;geIzCHmAaBLC_9@Gw`;Krf}rH+<0)ILbhFDQGubn8BOgKDHySYb;)oJ zy@1GL#ZV{9HzZ-{wMm@$=%Fl;>bTw)0L{{G8Zurv2_(3|jPFMiRfx{TbOerUrW*nH z8u7L00`yfwzHZgg%ChkUeE~kTuWTZ3d`s3ZNtpgKH{eXNZA5!_px0=R7-oGUqjM8A zk=+YDY%@77M3eBSomH^orwIimI9xKdOphRE8N`GN4d=g(glX%>2wq=&f*{s~15)gk zSi8HaP|R?mT_3w;drt~(7WRm?Yhh3>MCo5``1qd*Xx9U*3{i$?eE9d04^=*TO&N+4 zg>SCZG#v>e!uOfpif`qb#lN1d6(VHU!C_%GO$Xl3rLD2YyCBVHI75;_zppy~4+rk< zU96s<7MY$61IdYLH&WSCZMCsjN+>fRxILcjY~F(z@)q%kvpv+&Mtcsr@#c=V;kt5@n)nSosN-lobGq zQ=g?%Jv`W}C>A#_PV4ZO;3K2m-K&4kNZB+07f|Xl7Jrd~r;>p(K+yJnv>?h_LYR!s zPpYL(=N%eE-)~S_##bUzFPa1FKz_`Yf9=VO%V+reo2EcFp%`EmLno)CTPS zwM7LpW?;=wM{bCKcT^~o5f*FPAu&A6T%zKMMpux9@G-CyhGYbwsy^Jel0j(3GTxu|# z0f1vOwLGP+)ux^x(*BZ|$O8Eg1FCe#hdn39VISX+<+1?}nFbBLzX)e*2o?s>8mj{J z1`{PaF3qZQN}0wkRDL4ZgWx9o1SqSrz_BqxC7Zx~>i2ukD+-<<$OJc=#7~#VeVEv1 zta1U^qM(=Ia{nc+7_Fv1<1F&VvQUk^xLiGn#JG%}&Qq4%NxIv%6r&D;7{2%jeI*tn znuOdOXDULM>gX9z#PW23i087h3QLa6Rdkv~Dl&X{5vRzaO5)0~GK4N?%~>6tapeKp z%1pOZuc!W=oJVHSBg-Ujiapv_`oC)8#nhRpsq7GaK*BDAGK(EE1WH!8!Act ze@Q7Um8Koh>d#dus|rb>AzGjPWLG0_O*L@KU~fsxbrjD{Wk8?0qWHBm`|ok&N$Ste zxD0Hj^u|~7u-Sdw4UN#=_Fe71`G~7nRkw#pwnSz@4zJ%A)RU`LO!^(&VViYY}bhF)^*l z=HQdrEc9!~$WFDu8SQBnOfS6y_3!U2Y6Z8>gGk#Hz0ALW+th=6&g$9h>)|lLMo`== z)g>&IQJP&~kLNzz?RiyqfQBr8*ZTGGBsXHYE>)T2fZi}P1SW7`LVctNDGHrs!qN%u#XH8Z{Navr zgY%FblJ)~MJic`TsBkITmpzuhL(H^MfIblhd|+_H%Bs_tx&f8svlFyqwK$>&*6B9A z@U*wyC1rg})n6*~v|VHK@b@dQeWC5nN5Z1V^VgmG`!_oGih>y%y`zEDWZ)>i&JC`D z!6aI9c0Bxax$~$f1aCq;($dgLTWojPj#?B3dsC#1Qu=^v%(k+=usnz>RB8ZC$agtz zF{}(lEd0dA@zC+pJ1P5=ubpXdd6i>W2kaDZM}#G^S`wySeuK(&f3 zV&L*ZqQ+c#W0254&?G)ZMwKLLnNy@33x9XI$w9p1%iC;f1^PIK`soQ#;^QlyS+1p= z{iv!utUCHrbyQ7NRaIi9jPhp_o2XILh{^8pNx=HfD0S#=fe)h@lwCaiJX6=W=K9U` z5_H6k0>P)0t3PI9BS9-r17cfaEr+h?gg5&*kf{!|bfy1ag()^flbdMcs7z2cvl7UR3`Q2-_bDk=GzW`OQUvv(|{T_#|40maP13r{f$i7T{}&ep&gE{iO_~3 z`h{MON5-OQ?6>H*{P3r+tAU^K$&>R58Rb>;?IFjhnl;v7h1xdak2{Y}uCp6lWH6`C zWPO-EBDj2A#Lv&i;I3f3Edw_OSJi)5+hCwJ$I;)m!YX4kJl+;n=o6Hcaz0j~r6}0A z!{pfE*`a*Usq{+VZ{>}QAxlCF(N~>oX*6Z@p{{#g;Iu>YrN8TJkLa>i-WduxF=|a_ zifyfDj{d5Ksb)TIFl=ecH$3^g@^t>u@{7kz(xf?cvFSh8TUI3YWk)9-E{r)C*h|KS z);;r44X|OgQxANlZ(Frv?|{8SOT*^0b;=!jLA=wy8s5q39FFXIn?A1(|GL&KMoaKlH-*q$6tjv9~hv!l)(UP$I)CIp?fhzyA_y=ibHBdd7@b z4LJ|JP}8|%uP@FWfNqg|g8)ME?XvA#jXE<_;*N_&ACBwsNdt)yjwvwTzH=Iq#%)jSJBC zL#V>iNhbx^KQx}^be4h#KC6V|pHTb*24CdjNh!88+B1t)faAl^s~@(Su#L06mu}MR zQK}%iTksHLDlW>J#2+#*i*6e<&}`>irDbNC6tw7_{(+|_UEFD>YC3xq&e6;jc$M+g zTV&uHT{in*XC)zQoJun$PT4cjW040|SDa;># zOeqCX(^DCx!?#eCd@zZONu6*ocqTVGhHb*P{KdB%W0{Ew^d#%${6n?0!;7zAsKf&o zuHfUWYr77#wNExq7|qjm3ze5|5Vaf9jCaos+eZ50r76Q?oYN&E-0+8IRDR-qXSo#J zx1)#PJTr2RA={MaE;^?FH#{X~hVA{fXQvBwM13Cg)I5XGnrVqTd7l8ABmI+cUcDh0 zAYIAfDGg)Z&ij`+(uf9=!Ie++a~B6fh9*U~uL*Nh@kdIP57PKPsM^0*kyF4vb3=r4 zr?WD<==~H2Dj;GB z-|`UUk2tBC-=%O#h|87*Pe>9Co0mMWAp-XFN)%`RNGbO=?dmFI<@}{+rQ~kC+2cS_ z_eOksS{kK97p`JcE88eH_mg_=_}*8!ZOfL7+37b`qx6R)&I(V{QygwZ4Do6`e_X&_ z>4#arhrki;d)}CKF-^zn6EZg5ApbBEoBH4iQwK) zCH%k{UuuxpOAmoAd71HPh1!?1_qyyT`M#!`V8J@+$&_^N3tihyp}RmVm;hgzEA{`*rNT}F5{*Bu~&RFxoCF^-6*oCo8r&}N!uk+)M3F4`@fWpRZV zk#@H@27qnc;11z%vn&4>(1G%e{FR`m*(lWI&zlqD*MwABZ)zMoO5s*(EzS9++z@bW z_wu>-4)&-3zyCT0d5@TE9wuxRa zI?#9ZYwp$!n0O?}U3s4ESLa3^B#SgIvkpZX7tRZ?9?NHjzt7U{3zu(5-V-Wn#J4P9 zKZPDv^L>&ulj5X|=T3}VLuKAY4eU95IqfcD?@+_k7N3h{RkhCsv)ab`pd|FA!_G1` zoA6Hebp7EY#qQ+93+lGHEN))k9C>kWVj5iGB+6ec?31?kIB!^C)b^!?r__MXHl10+ zGR9;h!YRW6K67hOuIe6be5#|A+(0-wp#pNpbdO|DzLliEA(Cn=Qv_Miu#$5s_!c{C z0LtHx;xL?<#|tGX9+JL1Mfg+Hm!iV3!A_XMeJAF{op1z)s{209siFButmL{{#kuj~ z@WD^hRJhx=X?zIEqDQp}f!W%2zy_7TO*2PZ8C~b|+cc^#^z`FCQutgZsWYl2_R" ] @@ -627,196 +627,6 @@ "plt.show()" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Various steps in cross-validation\n", - "\n", - "When the repetitive splitting of the data set is done randomly,\n", - "samples may accidently end up in a fast majority of the splits in\n", - "either training or test set. Such samples may have an unbalanced\n", - "influence on either model building or prediction evaluation. To avoid\n", - "this $k$-fold cross-validation structures the data splitting. The\n", - "samples are divided into $k$ more or less equally sized exhaustive and\n", - "mutually exclusive subsets. In turn (at each split) one of these\n", - "subsets plays the role of the test set while the union of the\n", - "remaining subsets constitutes the training set. Such a splitting\n", - "warrants a balanced representation of each sample in both training and\n", - "test set over the splits. Still the division into the $k$ subsets\n", - "involves a degree of randomness. This may be fully excluded when\n", - "choosing $k=n$. This particular case is referred to as leave-one-out\n", - "cross-validation (LOOCV). \n", - "\n", - "\n", - "* Define a range of interest for the penalty parameter.\n", - "\n", - "* Divide the data set into training and test set comprising samples $\\{1, \\ldots, n\\} \\setminus i$ and $\\{ i \\}$, respectively.\n", - "\n", - "* Fit the linear regression model by means of ridge estimation for each $\\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\\boldsymbol{\\sigma}_{-i}^2(\\lambda)$, as" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "$$\n", - "\\begin{align*}\n", - "\\boldsymbol{\\beta}_{-i}(\\lambda) & = ( \\boldsymbol{X}_{-i, \\ast}^{T}\n", - "\\boldsymbol{X}_{-i, \\ast} + \\lambda \\boldsymbol{I}_{pp})^{-1}\n", - "\\boldsymbol{X}_{-i, \\ast}^{T} \\boldsymbol{y}_{-i}\n", - "\\end{align*}\n", - "$$" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "* Evaluate the prediction performance of these models on the test set by $\\log\\{L[y_i, \\boldsymbol{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}$. Or, by the prediction error $|y_i - \\boldsymbol{X}_{i, \\ast} \\boldsymbol{\\beta}_{-i}(\\lambda)|$, the relative error, the error squared or the R2 score function.\n", - "\n", - "* Repeat the first three steps such that each sample plays the role of the test set once.\n", - "\n", - "* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "$$\n", - "\\begin{align*}\n", - "\\frac{1}{n} \\sum_{i = 1}^n \\log\\{L[y_i, \\mathbf{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}.\n", - "\\end{align*}\n", - "$$" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "For the various values of $k$\n", - "\n", - "1. shuffle the dataset randomly.\n", - "\n", - "2. Split the dataset into $k$ groups.\n", - "\n", - "3. For each unique group:\n", - "\n", - "a. Decide which group to use as set for test data\n", - "\n", - "b. Take the remaining groups as a training data set\n", - "\n", - "c. Fit a model on the training set and evaluate it on the test set\n", - "\n", - "d. Retain the evaluation score and discard the model\n", - "\n", - "\n", - "5. Summarize the model using the sample of model evaluation scores\n", - "\n", - "The code here uses Ridge regression with cross-validation (CV) resampling and $k$-fold CV in order to fit a specific polynomial." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "collapsed": false, - "editable": true - }, - "outputs": [], - "source": [ - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "from sklearn.model_selection import KFold\n", - "from sklearn.linear_model import Ridge\n", - "from sklearn.model_selection import cross_val_score\n", - "from sklearn.preprocessing import PolynomialFeatures\n", - "\n", - "# A seed just to ensure that the random numbers are the same for every run.\n", - "# Useful for eventual debugging.\n", - "np.random.seed(3155)\n", - "\n", - "# Generate the data.\n", - "nsamples = 100\n", - "x = np.random.randn(nsamples)\n", - "y = 3*x**2 + np.random.randn(nsamples)\n", - "\n", - "## Cross-validation on Ridge regression using KFold only\n", - "\n", - "# Decide degree on polynomial to fit\n", - "poly = PolynomialFeatures(degree = 6)\n", - "\n", - "# Decide which values of lambda to use\n", - "nlambdas = 500\n", - "lambdas = np.logspace(-3, 5, nlambdas)\n", - "\n", - "# Initialize a KFold instance\n", - "k = 5\n", - "kfold = KFold(n_splits = k)\n", - "\n", - "# Perform the cross-validation to estimate MSE\n", - "scores_KFold = np.zeros((nlambdas, k))\n", - "\n", - "i = 0\n", - "for lmb in lambdas:\n", - " ridge = Ridge(alpha = lmb)\n", - " j = 0\n", - " for train_inds, test_inds in kfold.split(x):\n", - " xtrain = x[train_inds]\n", - " ytrain = y[train_inds]\n", - "\n", - " xtest = x[test_inds]\n", - " ytest = y[test_inds]\n", - "\n", - " Xtrain = poly.fit_transform(xtrain[:, np.newaxis])\n", - " ridge.fit(Xtrain, ytrain[:, np.newaxis])\n", - "\n", - " Xtest = poly.fit_transform(xtest[:, np.newaxis])\n", - " ypred = ridge.predict(Xtest)\n", - "\n", - " scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)\n", - "\n", - " j += 1\n", - " i += 1\n", - "\n", - "\n", - "estimated_mse_KFold = np.mean(scores_KFold, axis = 1)\n", - "\n", - "## Cross-validation using cross_val_score from sklearn along with KFold\n", - "\n", - "# kfold is an instance initialized above as:\n", - "# kfold = KFold(n_splits = k)\n", - "\n", - "estimated_mse_sklearn = np.zeros(nlambdas)\n", - "i = 0\n", - "for lmb in lambdas:\n", - " ridge = Ridge(alpha = lmb)\n", - "\n", - " X = poly.fit_transform(x[:, np.newaxis])\n", - " estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)\n", - "\n", - " # cross_val_score return an array containing the estimated negative mse for every fold.\n", - " # we have to the the mean of every array in order to get an estimate of the mse of the model\n", - " estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)\n", - "\n", - " i += 1\n", - "\n", - "## Plot and compare the slightly different ways to perform cross-validation\n", - "\n", - "plt.figure()\n", - "\n", - "plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')\n", - "plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')\n", - "\n", - "plt.xlabel('log10(lambda)')\n", - "plt.ylabel('mse')\n", - "\n", - "plt.legend()\n", - "\n", - "plt.show()" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -1273,6 +1083,203 @@ "plt.show()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Cross-validation\n", + "\n", + "When the repetitive splitting of the data set is done randomly,\n", + "samples may accidently end up in a fast majority of the splits in\n", + "either training or test set. Such samples may have an unbalanced\n", + "influence on either model building or prediction evaluation. To avoid\n", + "this $k$-fold cross-validation structures the data splitting. The\n", + "samples are divided into $k$ more or less equally sized exhaustive and\n", + "mutually exclusive subsets. In turn (at each split) one of these\n", + "subsets plays the role of the test set while the union of the\n", + "remaining subsets constitutes the training set. Such a splitting\n", + "warrants a balanced representation of each sample in both training and\n", + "test set over the splits. Still the division into the $k$ subsets\n", + "involves a degree of randomness. This may be fully excluded when\n", + "choosing $k=n$. This particular case is referred to as leave-one-out\n", + "cross-validation (LOOCV). \n", + "\n", + "\n", + "* Define a range of interest for the penalty parameter.\n", + "\n", + "* Divide the data set into training and test set comprising samples $\\{1, \\ldots, n\\} \\setminus i$ and $\\{ i \\}$, respectively.\n", + "\n", + "* Fit the linear regression model by means of ridge estimation for each $\\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\\boldsymbol{\\sigma}_{-i}^2(\\lambda)$, as" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\begin{align*}\n", + "\\boldsymbol{\\beta}_{-i}(\\lambda) & = ( \\boldsymbol{X}_{-i, \\ast}^{T}\n", + "\\boldsymbol{X}_{-i, \\ast} + \\lambda \\boldsymbol{I}_{pp})^{-1}\n", + "\\boldsymbol{X}_{-i, \\ast}^{T} \\boldsymbol{y}_{-i}\n", + "\\end{align*}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "* Evaluate the prediction performance of these models on the test set by $\\log\\{L[y_i, \\boldsymbol{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}$. Or, by the prediction error $|y_i - \\boldsymbol{X}_{i, \\ast} \\boldsymbol{\\beta}_{-i}(\\lambda)|$, the relative error, the error squared or the R2 score function.\n", + "\n", + "* Repeat the first three steps such that each sample plays the role of the test set once.\n", + "\n", + "* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\begin{align*}\n", + "\\frac{1}{n} \\sum_{i = 1}^n \\log\\{L[y_i, \\mathbf{X}_{i, \\ast}; \\boldsymbol{\\beta}_{-i}(\\lambda), \\boldsymbol{\\sigma}_{-i}^2(\\lambda)]\\}.\n", + "\\end{align*}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "For the various values of $k$\n", + "\n", + "1. shuffle the dataset randomly.\n", + "\n", + "2. Split the dataset into $k$ groups.\n", + "\n", + "3. For each unique group:\n", + "\n", + "a. Decide which group to use as set for test data\n", + "\n", + "b. Take the remaining groups as a training data set\n", + "\n", + "c. Fit a model on the training set and evaluate it on the test set\n", + "\n", + "d. Retain the evaluation score and discard the model\n", + "\n", + "\n", + "5. Summarize the model using the sample of model evaluation scores\n", + "\n", + "The code here uses Ridge regression with cross-validation (CV) resampling and $k$-fold CV in order to fit a specific polynomial." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.model_selection import KFold\n", + "from sklearn.linear_model import Ridge\n", + "from sklearn.model_selection import cross_val_score\n", + "from sklearn.preprocessing import PolynomialFeatures\n", + "\n", + "# A seed just to ensure that the random numbers are the same for every run.\n", + "# Useful for eventual debugging.\n", + "np.random.seed(3155)\n", + "\n", + "# Generate the data.\n", + "nsamples = 100\n", + "x = np.random.randn(nsamples)\n", + "y = 3*x**2 + np.random.randn(nsamples)\n", + "\n", + "## Cross-validation on Ridge regression using KFold only\n", + "\n", + "# Decide degree on polynomial to fit\n", + "poly = PolynomialFeatures(degree = 6)\n", + "\n", + "# Decide which values of lambda to use\n", + "nlambdas = 500\n", + "lambdas = np.logspace(-3, 5, nlambdas)\n", + "\n", + "# Initialize a KFold instance\n", + "k = 5\n", + "kfold = KFold(n_splits = k)\n", + "\n", + "# Perform the cross-validation to estimate MSE\n", + "scores_KFold = np.zeros((nlambdas, k))\n", + "\n", + "i = 0\n", + "for lmb in lambdas:\n", + " ridge = Ridge(alpha = lmb)\n", + " j = 0\n", + " for train_inds, test_inds in kfold.split(x):\n", + " xtrain = x[train_inds]\n", + " ytrain = y[train_inds]\n", + "\n", + " xtest = x[test_inds]\n", + " ytest = y[test_inds]\n", + "\n", + " Xtrain = poly.fit_transform(xtrain[:, np.newaxis])\n", + " ridge.fit(Xtrain, ytrain[:, np.newaxis])\n", + "\n", + " Xtest = poly.fit_transform(xtest[:, np.newaxis])\n", + " ypred = ridge.predict(Xtest)\n", + "\n", + " scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred)\n", + "\n", + " j += 1\n", + " i += 1\n", + "\n", + "\n", + "estimated_mse_KFold = np.mean(scores_KFold, axis = 1)\n", + "\n", + "## Cross-validation using cross_val_score from sklearn along with KFold\n", + "\n", + "# kfold is an instance initialized above as:\n", + "# kfold = KFold(n_splits = k)\n", + "\n", + "estimated_mse_sklearn = np.zeros(nlambdas)\n", + "i = 0\n", + "for lmb in lambdas:\n", + " ridge = Ridge(alpha = lmb)\n", + "\n", + " X = poly.fit_transform(x[:, np.newaxis])\n", + " estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold)\n", + "\n", + " # cross_val_score return an array containing the estimated negative mse for every fold.\n", + " # we have to the the mean of every array in order to get an estimate of the mse of the model\n", + " estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds)\n", + "\n", + " i += 1\n", + "\n", + "## Plot and compare the slightly different ways to perform cross-validation\n", + "\n", + "plt.figure()\n", + "\n", + "plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score')\n", + "plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold')\n", + "\n", + "plt.xlabel('log10(lambda)')\n", + "plt.ylabel('mse')\n", + "\n", + "plt.legend()\n", + "\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "More examples of the application of cross-validation follow here." + ] + }, { "cell_type": "code", "execution_count": null, diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter3.py b/doc/LectureNotes/_build/jupyter_execute/chapter3.py index 6aaa20907..6c9083dee 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter3.py +++ b/doc/LectureNotes/_build/jupyter_execute/chapter3.py @@ -433,161 +433,6 @@ plt.grid(True) plt.show() -## Various steps in cross-validation - -When the repetitive splitting of the data set is done randomly, -samples may accidently end up in a fast majority of the splits in -either training or test set. Such samples may have an unbalanced -influence on either model building or prediction evaluation. To avoid -this $k$-fold cross-validation structures the data splitting. The -samples are divided into $k$ more or less equally sized exhaustive and -mutually exclusive subsets. In turn (at each split) one of these -subsets plays the role of the test set while the union of the -remaining subsets constitutes the training set. Such a splitting -warrants a balanced representation of each sample in both training and -test set over the splits. Still the division into the $k$ subsets -involves a degree of randomness. This may be fully excluded when -choosing $k=n$. This particular case is referred to as leave-one-out -cross-validation (LOOCV). - - -* Define a range of interest for the penalty parameter. - -* Divide the data set into training and test set comprising samples $\{1, \ldots, n\} \setminus i$ and $\{ i \}$, respectively. - -* Fit the linear regression model by means of ridge estimation for each $\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\boldsymbol{\sigma}_{-i}^2(\lambda)$, as - -$$ -\begin{align*} -\boldsymbol{\beta}_{-i}(\lambda) & = ( \boldsymbol{X}_{-i, \ast}^{T} -\boldsymbol{X}_{-i, \ast} + \lambda \boldsymbol{I}_{pp})^{-1} -\boldsymbol{X}_{-i, \ast}^{T} \boldsymbol{y}_{-i} -\end{align*} -$$ - -* Evaluate the prediction performance of these models on the test set by $\log\{L[y_i, \boldsymbol{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}$. Or, by the prediction error $|y_i - \boldsymbol{X}_{i, \ast} \boldsymbol{\beta}_{-i}(\lambda)|$, the relative error, the error squared or the R2 score function. - -* Repeat the first three steps such that each sample plays the role of the test set once. - -* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as - -$$ -\begin{align*} -\frac{1}{n} \sum_{i = 1}^n \log\{L[y_i, \mathbf{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}. -\end{align*} -$$ - -For the various values of $k$ - -1. shuffle the dataset randomly. - -2. Split the dataset into $k$ groups. - -3. For each unique group: - -a. Decide which group to use as set for test data - -b. Take the remaining groups as a training data set - -c. Fit a model on the training set and evaluate it on the test set - -d. Retain the evaluation score and discard the model - - -5. Summarize the model using the sample of model evaluation scores - -The code here uses Ridge regression with cross-validation (CV) resampling and $k$-fold CV in order to fit a specific polynomial. - -import numpy as np -import matplotlib.pyplot as plt -from sklearn.model_selection import KFold -from sklearn.linear_model import Ridge -from sklearn.model_selection import cross_val_score -from sklearn.preprocessing import PolynomialFeatures - -# A seed just to ensure that the random numbers are the same for every run. -# Useful for eventual debugging. -np.random.seed(3155) - -# Generate the data. -nsamples = 100 -x = np.random.randn(nsamples) -y = 3*x**2 + np.random.randn(nsamples) - -## Cross-validation on Ridge regression using KFold only - -# Decide degree on polynomial to fit -poly = PolynomialFeatures(degree = 6) - -# Decide which values of lambda to use -nlambdas = 500 -lambdas = np.logspace(-3, 5, nlambdas) - -# Initialize a KFold instance -k = 5 -kfold = KFold(n_splits = k) - -# Perform the cross-validation to estimate MSE -scores_KFold = np.zeros((nlambdas, k)) - -i = 0 -for lmb in lambdas: - ridge = Ridge(alpha = lmb) - j = 0 - for train_inds, test_inds in kfold.split(x): - xtrain = x[train_inds] - ytrain = y[train_inds] - - xtest = x[test_inds] - ytest = y[test_inds] - - Xtrain = poly.fit_transform(xtrain[:, np.newaxis]) - ridge.fit(Xtrain, ytrain[:, np.newaxis]) - - Xtest = poly.fit_transform(xtest[:, np.newaxis]) - ypred = ridge.predict(Xtest) - - scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred) - - j += 1 - i += 1 - - -estimated_mse_KFold = np.mean(scores_KFold, axis = 1) - -## Cross-validation using cross_val_score from sklearn along with KFold - -# kfold is an instance initialized above as: -# kfold = KFold(n_splits = k) - -estimated_mse_sklearn = np.zeros(nlambdas) -i = 0 -for lmb in lambdas: - ridge = Ridge(alpha = lmb) - - X = poly.fit_transform(x[:, np.newaxis]) - estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold) - - # cross_val_score return an array containing the estimated negative mse for every fold. - # we have to the the mean of every array in order to get an estimate of the mse of the model - estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds) - - i += 1 - -## Plot and compare the slightly different ways to perform cross-validation - -plt.figure() - -plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score') -plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold') - -plt.xlabel('log10(lambda)') -plt.ylabel('mse') - -plt.legend() - -plt.show() - ## The bias-variance tradeoff @@ -934,6 +779,163 @@ plt.ylabel('log10[MSE]') plt.legend() plt.show() +## Cross-validation + +When the repetitive splitting of the data set is done randomly, +samples may accidently end up in a fast majority of the splits in +either training or test set. Such samples may have an unbalanced +influence on either model building or prediction evaluation. To avoid +this $k$-fold cross-validation structures the data splitting. The +samples are divided into $k$ more or less equally sized exhaustive and +mutually exclusive subsets. In turn (at each split) one of these +subsets plays the role of the test set while the union of the +remaining subsets constitutes the training set. Such a splitting +warrants a balanced representation of each sample in both training and +test set over the splits. Still the division into the $k$ subsets +involves a degree of randomness. This may be fully excluded when +choosing $k=n$. This particular case is referred to as leave-one-out +cross-validation (LOOCV). + + +* Define a range of interest for the penalty parameter. + +* Divide the data set into training and test set comprising samples $\{1, \ldots, n\} \setminus i$ and $\{ i \}$, respectively. + +* Fit the linear regression model by means of ridge estimation for each $\lambda$ in the grid using the training set, and the corresponding estimate of the error variance $\boldsymbol{\sigma}_{-i}^2(\lambda)$, as + +$$ +\begin{align*} +\boldsymbol{\beta}_{-i}(\lambda) & = ( \boldsymbol{X}_{-i, \ast}^{T} +\boldsymbol{X}_{-i, \ast} + \lambda \boldsymbol{I}_{pp})^{-1} +\boldsymbol{X}_{-i, \ast}^{T} \boldsymbol{y}_{-i} +\end{align*} +$$ + +* Evaluate the prediction performance of these models on the test set by $\log\{L[y_i, \boldsymbol{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}$. Or, by the prediction error $|y_i - \boldsymbol{X}_{i, \ast} \boldsymbol{\beta}_{-i}(\lambda)|$, the relative error, the error squared or the R2 score function. + +* Repeat the first three steps such that each sample plays the role of the test set once. + +* Average the prediction performances of the test sets at each grid point of the penalty bias/parameter. It is an estimate of the prediction performance of the model corresponding to this value of the penalty parameter on novel data. It is defined as + +$$ +\begin{align*} +\frac{1}{n} \sum_{i = 1}^n \log\{L[y_i, \mathbf{X}_{i, \ast}; \boldsymbol{\beta}_{-i}(\lambda), \boldsymbol{\sigma}_{-i}^2(\lambda)]\}. +\end{align*} +$$ + +For the various values of $k$ + +1. shuffle the dataset randomly. + +2. Split the dataset into $k$ groups. + +3. For each unique group: + +a. Decide which group to use as set for test data + +b. Take the remaining groups as a training data set + +c. Fit a model on the training set and evaluate it on the test set + +d. Retain the evaluation score and discard the model + + +5. Summarize the model using the sample of model evaluation scores + +The code here uses Ridge regression with cross-validation (CV) resampling and $k$-fold CV in order to fit a specific polynomial. + +import numpy as np +import matplotlib.pyplot as plt +from sklearn.model_selection import KFold +from sklearn.linear_model import Ridge +from sklearn.model_selection import cross_val_score +from sklearn.preprocessing import PolynomialFeatures + +# A seed just to ensure that the random numbers are the same for every run. +# Useful for eventual debugging. +np.random.seed(3155) + +# Generate the data. +nsamples = 100 +x = np.random.randn(nsamples) +y = 3*x**2 + np.random.randn(nsamples) + +## Cross-validation on Ridge regression using KFold only + +# Decide degree on polynomial to fit +poly = PolynomialFeatures(degree = 6) + +# Decide which values of lambda to use +nlambdas = 500 +lambdas = np.logspace(-3, 5, nlambdas) + +# Initialize a KFold instance +k = 5 +kfold = KFold(n_splits = k) + +# Perform the cross-validation to estimate MSE +scores_KFold = np.zeros((nlambdas, k)) + +i = 0 +for lmb in lambdas: + ridge = Ridge(alpha = lmb) + j = 0 + for train_inds, test_inds in kfold.split(x): + xtrain = x[train_inds] + ytrain = y[train_inds] + + xtest = x[test_inds] + ytest = y[test_inds] + + Xtrain = poly.fit_transform(xtrain[:, np.newaxis]) + ridge.fit(Xtrain, ytrain[:, np.newaxis]) + + Xtest = poly.fit_transform(xtest[:, np.newaxis]) + ypred = ridge.predict(Xtest) + + scores_KFold[i,j] = np.sum((ypred - ytest[:, np.newaxis])**2)/np.size(ypred) + + j += 1 + i += 1 + + +estimated_mse_KFold = np.mean(scores_KFold, axis = 1) + +## Cross-validation using cross_val_score from sklearn along with KFold + +# kfold is an instance initialized above as: +# kfold = KFold(n_splits = k) + +estimated_mse_sklearn = np.zeros(nlambdas) +i = 0 +for lmb in lambdas: + ridge = Ridge(alpha = lmb) + + X = poly.fit_transform(x[:, np.newaxis]) + estimated_mse_folds = cross_val_score(ridge, X, y[:, np.newaxis], scoring='neg_mean_squared_error', cv=kfold) + + # cross_val_score return an array containing the estimated negative mse for every fold. + # we have to the the mean of every array in order to get an estimate of the mse of the model + estimated_mse_sklearn[i] = np.mean(-estimated_mse_folds) + + i += 1 + +## Plot and compare the slightly different ways to perform cross-validation + +plt.figure() + +plt.plot(np.log10(lambdas), estimated_mse_sklearn, label = 'cross_val_score') +plt.plot(np.log10(lambdas), estimated_mse_KFold, 'r--', label = 'KFold') + +plt.xlabel('log10(lambda)') +plt.ylabel('mse') + +plt.legend() + +plt.show() + +More examples of the application of cross-validation follow here. + # Common imports import os import numpy as np diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter3_25_2.png b/doc/LectureNotes/_build/jupyter_execute/chapter3_25_2.png index b77a1febbbca6d54d73c1e26de0b254d433ede55..dada82b441f4a8ba20ebcab5f87860a422153e41 100644 GIT binary patch literal 4835 zcma)=3piBy|HtRzTBNe&erY?XkdWlwuEcR|)1eGT#%2=R{_@GhDBAA<;s6A_LfJ`rb5QM@sBry@cFLm~pnp30}a zDdFUhVBNj^O+(o)A|f>0P)jRlJz#GL#aHXMbY>2J646krli?W5M#;4ui^PM3NPum)IRDYXycK5;G?vuQlIZyAalf44) z4di#ADEn?=U-w&=o#d&>$7iO>`loz-M)E%8ea?v|PK=sCrrs~Y@A~S>XBz3Wz=X4Uc5ihtYpR)^k==g-{=&)oUf2)W$?@EsdUENltOgS;Lw zhLb)RY^{?&R1fqP_X>jTloZ3QbIRNjWC6>$j&9)yaXg?P{HCf%DuWg@5g-xz=`cY9 z%*wD(OQzQfM451PG0==SM-DuVbFT1Tq{ULtV2x;(k3-n}keffHK#`vBO4EHl*wWeV zhDfE?wX3K%isPz-lMcKBKnf?9-2WNxXFh`W5Svu*4>Y>KmK!!T#a_-0m8WMxa zFnV6Zuyvj@RsN&d(Tpp)HQDZFSzSL^W)0ob91&k zN{@UQ-0_{E+Pp7ttL@;{zh~Qej=Zwkab;*4AlmW{!RDpo?|?IE(G`!~UpCQ)ihxg` zqHTG6mckW4bKuw!oS7?>?%e50U||oQfO@G(K?)py55i&(zR$R%!5UT#$XOIvLp!{T znxKzd<}DF~+NyRk-HQa}N*5Qje~dDt?$7(d5+@jE$c_hRE6T;qL&hC&ay>j%&W9Iu zjV`Z00k7#pM`81g8#gxlJp{Y%&Sa^vk8r#g3G%OWDi+^6scL2y#}|G)@q5Hkcq2V- zB)l1wqdq~g9?p!fB0YkWqdKTYm#u^x?<|Jnu+@JtUHK1F4esm(rVG4(%f;kFkcm(& zZ~rh+u*}dd8r$tq7zr1a@KR6xz5Lb*tylYj7RSRXc4(&Jh3UEe0Ox9z(W~qA_eaJ4 zd&oMnFJ=8df{tvW0V66kX6)Rc<#OP&YNi7(8{F$RtQ9!+DQK~4ccmM=zv$K_fqtyK zj0p>-DyOzxNc8UYB6C0R^M}EAjn%v59LrcMa;XaAJ_NEZ5 zX_lAboGiC%I&0Sq8>Ez;RXKMEdV}xlNK?YQhi4fkiq}Ttg6M!<<{2@Y|JK-lz_=e4 zyq%u*T&|$7v;OM8wElmgG)mTk`eigqgTK`8T8}VLQ_kJlvVWELpHy+Nkr(|_;a{z- zp#`{dke#m@y^9^|GVcve+VLe?pYHmiiTJ8(q<6N|aWV-YXC9FeT{7b|Au%eU2HXa~ zh$=_>?A+N*E2h0ET?yuVbXr+)a)2u~Xv*UBxamPNy*MGGu#Yb~LDnvB#ak6R~yf&_E!R`USQax4lC?+yOQRH)(2UT#=16`6;+!T&OMBx=ol;*v??C{SrfkLUw;X!me zwQ5-fbyjOT0k(UkP$FL}0AoEoCf*Due`>RkStl}qoAhfI?lBBn^hFTWS3Y3-*-Mm0 z?h%hf_>=V6h9D@?7d}rT&}${Cm8x%bvIOV+;Al<&N+^C247LmWq4dgJq~n@dgGU1A zC66W4Z+x{2go~Ptl2!f`;o_hCzg^ns+1@j>)-}HZ&|Lw@9bD>tVkl?%S4;mACHdr7 zan8Q&YZ_d(t^r>zulMPH=8%QCGCK`OyfX9_kWh)XRA(JzzX274?n+hifSTRxI9Qk- zn15Og0GoB2tT@>yi-9;}jHqMuT3q$NC3@P)23GZK+k6|5gl5!&=2vl+R6Eq(Bau<1 z6y-?h7mX_vGL^VvQy$hXo(`mhfDT*i%C|*r3)Q+;fY?{qt?P6!U#CSY-sgT~)8Ve7 ztNi93C(I?FMvPIdWF`b*`3*P6>o58m-p*?{zQKWYDc@n1VxagN+stPA$a%9d<=2&j ztHdViYOb3&{H^>cy?k+avddXmuv+loF=#d1uY)htgnq$3?c)h2zbIBhn*bleOROY4Qlh4L`{?Nx6CUGdNGt+%FO#+gKd^1H74I#eMweII^XO~1eFjy) zK^bp=%RuEEc0c+z=1d=JxaiaxW;21h)7H~`z=VE@TJr6soqF784>veP(w_m4e@%qv-D5uLWEcDpwflC_ge{9?YVSEC7wcJ#Y`E<`&b*7f zNBo1A7iIQo}wNS6Q=EvzjMj33)z_l*hxTqIx#ew}IahnfH%pHX$avMHp>J5zF>Liz} z>c^VK@1NdmMN#BlL@#3+sak-+QAomJA=WM{WE3v)u52ONk^ru(M;)Swx1y-ulgzLd z$(Vy1^hzYC0n^LgvODgH9yF{-5Su$+>||Xo{HnugRfnD4LV0FGHOoIdNErlDb+$;3 zb?hOry0$Oseb*eFc=DZT@$6V^+KNJch>zrxHSeTv{RTMCXV_%t^23N{KAKRfHyaR? zd4yFIcHu*qNArsXzH0(qyoSP-__X<0j63sz`hFHb!%(Lf10tpzzkwN0Zr3} zEYy4EMwr7-KzmgK5ok#Ey;7`(A)>3jfIPXFGooK{CG`|r*^Sb1A8@YKvu zYC$jJQoi4m()r@fS(zopI^ifHApsl8zA=$uF+#u(i1q1ItmISkh4bDLB)bUI^R4$* zeK^XYDhzdjv^QEDm5Jv%4<;MWF3$A14N2*WTrr~QKdUw)ILxgbf)(<@@kpX~rAy*q|wA3iRDTjjZakm4$!zBK23H#cV-W2NGx_TCbIV5|@!J=X}j zXAM0SgM&A6Hz)au*K<;Qn`LHKF-iIbtZa%=x_?fBn~(;E)ovdizl(iNz3symjYmW% zQhsgsthet<~tKuFu{t{^DJz!V56d38_os+Qq7#O1}k+}5% zdeThdl;Damcl@PGwQ=*L^P&v4R-c^@%tyiE8lFBojzX4v=i-3R$0=e~(&~4gOT0d0 z$}ysav_*byrlw=1ROfxw zJ9xJk^_EEZWY5RpS#|b2+DOetOr$A#C;eHl>!!F6D4yb&`u0Y$BhISKtnUio(HyWwyKk^)9Z( zrM9~y^gi;CSvt_7{EgE&?S&Z!su=@0hb+i(BvCV$nzYM9f4+n|nKqG*rZRaCIy0|_ zEzIP^qc@=$8d;Ab)9|r7fL1fRGJTS>2>0oBcBKaAB*b5@|LwhG(DL9V?uzypBdPAu zV|VN9LYY0WSgsO{Xc);tV$MK8Wk{)*qRn(B;-Z(Y(fFa8jd$AYI~BdYC|gVvT-p?Bp)NbIOmdR^HfZbM zWtW|+-&7;lALHjJi#^RuRRy9@Ut|`^wSG^Qv?7y;4x>9b-*K3Wp+~&-TK)W_rHt!*L>NO(C=vLmj!MyL3I>h$RTy_K%x>dI zWz-L?6fQpHbogkCj0^X4x0cN$!5SRPak{pdNty}yP34d$Qk f;a~A)t4hVPhes*%wzK>XP8iU_{&3m%9zXsc;hzEw literal 5148 zcmbtYc~leU+D`}&WtCC|k)_K3T0~StHX*fEWn@!vKnM_2K(;_+-;+p{+iI~+ydbi* z6-5REC?p_D2&EO{LX@3^C0tRIH3msQNOFUh(ga<;Jd8r2$%^}hRjP_#<+IOg6c_vWy+`A{VnXASj>U$c+>gbRqGIBs z!jFHH7!n&79usZ7Me*7EQCNIDDbB&liuiuPmYCR+R`>Gi3Kg5ICb{^=p-}3&D_=C~ zN})aqrTaDbn~nP@7I87hd3eom7-I<23F+Qox=2IG|`&62}^uO&K{+(ziIsT}*E=ht2@y zR#kQbDlTwt$*2`*w0+sMe9zeyAmN@cuT-lj^|m_7ZfU9}a_;#f^F|HO51L^u&EU!r zhU-Am-mm1(9~I;m6*&!rxO5`$*sE(ZlhqgA{dc_#?*6!q06PRPxE{{dAcGU%{po

    B~0e5zC zS^puKTC!(8j%h<5=X5=aWN&>L^KwOP$zv)$M`Eph2tElPF^?g@Q$;; zXf64#+I1MZfZ@C?Ev^xKZWTDcTpr56MGp8K>)^OY$ejn4y5Gg=kJ}i zpa*vhKZ&TmTrA(qwnZk&nX>L%qlJlSwG0#VPR;4MTMQ8P{>Lw0$+MG>-NOdqs+%oi zu+i%yTACicu{TMM4oh}{Yn9ILn4*ov3!KbpC@sxtay_Hj1ug+U%#rVU6-t)!||CZ|UxE z?H8N%Wzal8+Q=#J>pH{6urq_wBKcZ_ow*WBUY##)9K+Sr)K)-GIpisNGtNFqD^%U3 z=HDv#_UU!Nk~~Rv~GW{FIpnQZZ4zKDOyL0CtV7XT9Uq z*qI?P!dI|t_cSbDe39kW3<^uz~UZ_WxxCjOX;D=F$ z20RHVGVLPZNPB3OUsO{TvbMH9 z$Mb=A6kkl0K?$Wp#wyO*O>2`U`bi#*PQ0Bsr6>bl#)(K8KwqN}54elgG}FIb7$|UF zVIh%eRDaEK7P}>;CbI)uyurX22I#1(8cR1Z&Lw00;uZ)ZxQP-3zo8|5hZfbsERX?& zZ(fljC$*-OVsRN&eLUPsaR2Kg=HL}IuJE6wTK`QB{}%xFNw@2P72VGR2$$=mw}$dI z$C>(pe}?GKa=miK4;K}D*#xrk$9kEW9&jUU=qBQW9+A^k5tCTHw4m%rBM|=ex~{nv zwPaG5)IgvM&F>Se5859&g5K4|!#!atiu2K@MWY45pORJ!XQ|ret4U~~gm1P)A!Db| zOhyQ$fJ$(O5rp$k%jR0$qMpyfgA`OYrxKK82)#sQ7aqceO=eRyi76P;s_%#7{(RY9 zc*<>iL2zPpbaZ%+E1Vm=qVd-3CrCpYyaHWC#C0UJw=%$yv&R))xAlHC*O4MIzoVaQ zBfQEkqu78R#O9@M*%z5oogiJ39jBZPC!Ie=T+%<5E+_; z6~+;3rrDOn6f(YSNFLI@WEvi_{$#P-%6dU4`6z3fbkc7rf$d9|wy`T;9k>Lz^M0%x zCqM)=t25U`W!QncldP2pxqk|Irc`jaY}vmZL{l19J+4{J*1WHTymE$SNW) z_wIYT+D~1DT1Sn-Z@Nc3vjKUhp8sP$e2`j4q}lCbjNbm<>FUs$k9M zjdFXz&DFh|L4z)rQ36Evg~pXOQ0tU%_A?msf&Ac|oT;&S{BxS@UcFls9AnRA-S7~cOW>h)~%A|)>pUT_o8uU zkYc$uMN#7s7mP3{(6s<%}M{x*OvWrRfst< zv4OZKM`dRY52~bAT0t@-<`*m|{h*5`xA`zj}JjS^J2;Oo1;!ik^ z#d>A=$%=_M`r42n`NmiLsjXrQ2lWbV8$j2K<+emAYjq8uYE1g;wwBXa@LoF1{sE-e@ zC>HHEUEv*ji$^W~GqnB%Ck>v7wBXmU$LnXiE=;q6^DUm-_&Y7N@A2yFwo{3Jc6Hg~(ymsmKRQ6xVECRTb4y z!JuBx8{S~&SDWWCBP)>MGe0=mmurj4KFf3*&B>PTYVm0K!##CysEX?8Km?#^#V4Qa#$>swU01!PDx5 z$IDCJc`Knylif(Ip&6>mEVCZBY~pAMP6=)^7m4g&*t-QOf^Rl;ugHQzVIUKS)&RYZ z8Y0>!2Y(3&Z`jr5(|LRDbc}gPpZMZCjm*xd-PkelWQV{J%w zq3PngPhb2pHR_7#(DLj5dBq#qWYuAep(?95SYg*&U$(=g;EPv}jvF}1>y)q+6Ne~( ztrDLy(?-5UICRo8&@UiYbj8Ux2IyKP*JS_j!+aZ?pEI(N;}LWcFRcMaoLgUYsLFzW zT)XAM#mNpEugc(4fnu&i?ZnW7$WIbHMK+HAy5(_1;<I_9AgmFnfoy6t zOx-UfSXoU52KF{LT|Xq*5xQikm=NeOQA7nbjvI{16M|gfIJi(XeN8{^O#4N^^2yKH z8OQ_{%+y&h(YT(Pc^#QCdHS8YmL~n6Vvn0?DZUcXmE@sv1T^u2Ykx{PTlL#0*Ddsn zh2SN4uw-naSNt3xxQggNOWxT2@qrjG*b(0KSlk?8)9q9%|1~{?SX_BE;c?rH^9VFP zAPR!(s^A9sUj8f#umw`)Kb%~4gRg=!8zLO7In$tVblu5X@%#BK_4EnWHE9l(FZO+z zV4mU*Th+=fB>8P7Ww5PkpLJ1e zESjTL%?%aHKP5`d1mk5#CmShrcE%o}9NpiOtC%2apPi%1wUq5l>L-=km)jD_Jh2}A zbpyYp$5UHMSj_464ZsoJqK;G=8Uyadeu#%En^MTKI>IaZeEw0H<6f&`5MD^(JI4^ z&r)qoBtOb_R+)H_@57yR%Q)pT|quDRSjqIELDI-B<%ZPD6ML zP*{~C*n5jYTc;R(J|bxfO$F!JPHe*V1zlN*>79bs?>Y#DFZ6%66P`_5a)SBddO;X8 z)NIJm+Az21{Lj{s@-$>(mmp0;wxg(FTgj}oe$(zkjAH(F;^6#q8pge{w(yVwd~eW3kq3tIWI-5{?0T-&~jTl;qVYNe-Yswg$maDi~sciVnjJT**1~+f!(Ds zclygOa$kKg5eW#a2mBmSEtTH(k< zOL~}mE#CHX_=5Gq%@-3PHxl~oxRPQdbx&Fbi?-YR>Cgi&@zDf5VH*3E-~O}=k4en8 zI=I&hS>9tNXgwygLjU3LqE>Z*o^y8^hJ?pwnrGq8&4g^#AZe3c9jIEWKhabWvdm5z zY@o(b2gHSPTfuqguw*|qTR%NfxPe`9{ph%+hBJP_Ijbc%5p{B#{%q#uR<~=`eC9FX zwfz@RBTAuXbK\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 34\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 35\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 36\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mscikitplot\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mskplt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 37\u001b[0m \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlogreg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test_scaled\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 38\u001b[0m \u001b[0mskplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmetrics\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot_confusion_matrix\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0my_test\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnormalize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", - "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'scikitplot'" - ] + "data": { + "image/png": "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\n", + "text/plain": [ + "

    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_57_2.png" + } + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_57_3.png" + } + }, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_57_4.png" + } + }, + "output_type": "display_data" } ], "source": [ @@ -1207,7 +1236,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Optimization, the central part of any Machine Learning algortithm\n", + "# Optimization, the central part of any Machine Learning algortithm\n", "\n", "Almost every problem in machine learning and data science starts with\n", "a dataset $X$, a model $g(\\beta)$, which is a function of the\n", @@ -1220,7 +1249,6 @@ "\n", "\n", "\n", - "## Revisiting our Logistic Regression case\n", "\n", "In our discussion on Logistic Regression we studied the \n", "case of\n", @@ -1249,7 +1277,6 @@ "where $\\boldsymbol{\\beta}$ are the weights we wish to extract from data, in our case $\\beta_0$ and $\\beta_1$. \n", "\n", "\n", - "## The equations to solve\n", "\n", "Our compact equations used a definition of a vector $\\boldsymbol{y}$ with $n$\n", "elements $y_i$, an $n\\times p$ matrix $\\boldsymbol{X}$ which contains the\n", @@ -1291,7 +1318,6 @@ "This defines what is called the Hessian matrix.\n", "\n", "\n", - "## Solving using Newton-Raphson's method\n", "\n", "If we can set up these equations, Newton-Raphson's iterative method is normally the method of choice. It requires however that we can compute in an efficient way the matrices that define the first and second derivatives. \n", "\n", @@ -1332,9 +1358,6 @@ "If we can compute these matrices, in particular the Hessian, the above is often the easiest method to implement. \n", "\n", "\n", - "\n", - "## Brief reminder on Newton-Raphson's method\n", - "\n", "Let us quickly remind ourselves how we derive the above method.\n", "\n", "Perhaps the most celebrated of all one-dimensional root-finding\n", @@ -1346,7 +1369,6 @@ "normally discourage the use of this method.\n", "\n", "\n", - "## The equations\n", "\n", "The Newton-Raphson formula consists geometrically of extending the\n", "tangent line at a current point until it crosses zero, then setting\n", @@ -1421,8 +1443,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Simple geometric interpretation\n", - "\n", "The above is Newton-Raphson's method. It has a simple geometric\n", "interpretation, namely $x_{n+1}$ is the point where the tangent from\n", "$(x_n,f(x_n))$ crosses the $x$-axis. Close to the solution,\n", @@ -1437,7 +1457,6 @@ "\n", "\n", "\n", - "## Extending to more than one variable\n", "\n", "Newton's method can be generalized to systems of several non-linear equations\n", "and variables. Consider the case with two equations" @@ -1574,8 +1593,6 @@ "we are always moving towards smaller function values, i.e a minimum.\n", "\n", "\n", - "## More on Steepest descent\n", - "\n", "The previous observation is the basis of the method of steepest\n", "descent, which is also referred to as just gradient descent (GD). One\n", "starts with an initial guess $\\mathbf{x}_0$ for a minimum of $F$ and\n", @@ -1599,8 +1616,6 @@ "the learning rate within the context of Machine Learning.\n", "\n", "\n", - "## The ideal\n", - "\n", "Ideally the sequence $\\{\\mathbf{x}_k \\}_{k=0}$ converges to a global\n", "minimum of the function $F$. In general we do not know if we are in a\n", "global or local minimum. In the special case when $F$ is a convex\n", @@ -1621,7 +1636,6 @@ "\n", "\n", "\n", - "## The sensitiveness of the gradient descent\n", "\n", "The gradient descent method \n", "is sensitive to the choice of learning rate $\\gamma_k$. This is due\n", @@ -1652,13 +1666,10 @@ "regular polygons (triangles, rectangles, pentagons, etc...).\n", "\n", "\n", - "## Convex function\n", "\n", "**Convex function**: Let $X \\subset \\mathbb{R}^n$ be a convex set. Assume that the function $f: X \\rightarrow \\mathbb{R}$ is continuous, then $f$ is said to be convex if $$f(tx_1 + (1-t)x_2) \\leq tf(x_1) + (1-t)f(x_2) $$ for all $x_1, x_2 \\in X$ and for all $t \\in [0,1]$. If $\\leq$ is replaced with a strict inequaltiy in the definition, we demand $x_1 \\neq x_2$ and $t\\in(0,1)$ then $f$ is said to be strictly convex. For a single variable function, convexity means that if you draw a straight line connecting $f(x_1)$ and $f(x_2)$, the value of the function on the interval $[x_1,x_2]$ is always below the line as illustrated below.\n", "\n", "\n", - "## Conditions on convex functions\n", - "\n", "In the following we state first and second-order conditions which\n", "ensures convexity of a function $f$. We write $D_f$ to denote the\n", "domain of $f$, i.e the subset of $R^n$ where $f$ is defined. For more\n", @@ -1691,7 +1702,6 @@ "This condition is particularly useful since it gives us an procedure for determining if the function under consideration is convex, apart from using the definition.\n", "\n", "\n", - "## More on convex functions\n", "\n", "The next result is of great importance to us and the reason why we are\n", "going on about convex functions. In machine learning we frequently\n", @@ -1714,7 +1724,7 @@ "This result means that if we know that the cost/loss function is convex and we are able to find a minimum, we are guaranteed that it is a global minimum.\n", "\n", "\n", - "## Some simple problems\n", + "### Some simple problems\n", "\n", "1. Show that $f(x)=x^2$ is convex for $x \\in \\mathbb{R}$ using the definition of convexity. Hint: If you re-write the definition, $f$ is convex if the following holds for all $x,y \\in D_f$ and any $\\lambda \\in [0,1]$ $\\lambda f(x)+(1-\\lambda)f(y)-f(\\lambda x + (1-\\lambda) y ) \\geq 0$.\n", "\n", @@ -1739,13 +1749,6 @@ "Using the definition of convexity, try to show that a function satisfying the properties above is convex (the third condition is not needed to show this).\n", "\n", "\n", - "\n", - "## Friday September 25\n", - "\n", - "[Video of Lecture](https://www.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/forelesningsvideoer/LectureSeptember25.mp4?vrtx=view-as-webpage) and [link to handwritten notes](https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/NotesSeptember25.pdf).\n", - "\n", - "\n", - "\n", "## Standard steepest descent\n", "\n", "\n", @@ -1794,7 +1797,6 @@ "When we have found the exact solution, $\\boldsymbol{r}=0$.\n", "\n", "\n", - "## Gradient method\n", "\n", "The residual is zero when we reach the minimum of the quadratic equation" ] @@ -1816,9 +1818,6 @@ "symmetric. This defines also the Hessian and we want it to be positive definite. \n", "\n", "\n", - "\n", - "## Steepest descent method\n", - "\n", "We denote the initial guess for $\\boldsymbol{x}$ as $\\boldsymbol{x}_0$. \n", "We can assume without loss of generality that" ] @@ -1855,8 +1854,6 @@ "instead.\n", "\n", "\n", - "\n", - "## Steepest descent method\n", "One can show that the solution $\\boldsymbol{x}$ is also the unique minimizer of the quadratic form" ] }, @@ -1895,9 +1892,6 @@ "$\\boldsymbol{x}_0=0$ it is equal $-\\boldsymbol{b}$.\n", "\n", "\n", - "\n", - "\n", - "## Final expressions\n", "We can compute the residual iteratively as" ] }, @@ -1974,21 +1968,39 @@ "$$" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Steepest descent example" - ] - }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "metadata": { "collapsed": false, "editable": true }, - "outputs": [], + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_114_1.png" + } + }, + "output_type": "display_data" + } + ], "source": [ "import numpy as np\n", "import numpy.linalg as la\n", @@ -2021,12 +2033,27 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "metadata": { "collapsed": false, "editable": true }, - "outputs": [], + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_116_0.png" + } + }, + "output_type": "display_data" + } + ], "source": [ "pt.axis(\"equal\")\n", "pt.contour(xmesh, ymesh, fmesh)\n", @@ -2042,7 +2069,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "metadata": { "collapsed": false, "editable": true @@ -2062,12 +2089,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": { "collapsed": false, "editable": true }, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[ 1.33333333 -0.26666667]\n" + ] + } + ], "source": [ "def f1d(alpha):\n", " return f(x + alpha*s)\n", @@ -2087,12 +2122,37 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, "metadata": { "collapsed": false, "editable": true }, - "outputs": [], + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
    " + ] + }, + "metadata": { + "filenames": { + "image/png": "/Users/mhjensen/Teaching/MachineLearning/doc/LectureNotes/_build/jupyter_execute/chapter4_122_1.png" + } + }, + "output_type": "display_data" + } + ], "source": [ "pt.axis(\"equal\")\n", "pt.contour(xmesh, ymesh, fmesh, 50)\n", @@ -2144,10 +2204,6 @@ "Two vectors are conjugate if they are orthogonal with respect to \n", "this inner product. Being conjugate is a symmetric relation: if $\\boldsymbol{s}$ is conjugate to $\\boldsymbol{t}$, then $\\boldsymbol{t}$ is conjugate to $\\boldsymbol{s}$.\n", "\n", - "\n", - "\n", - "\n", - "## Conjugate gradient method\n", "An example is given by the eigenvectors of the matrix" ] }, @@ -2164,13 +2220,8 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "which is zero unless $i=j$.\n", + "which is zero unless $i=j$. \n", "\n", - "\n", - "\n", - "\n", - "\n", - "## Conjugate gradient method\n", "Assume now that we have a symmetric positive-definite matrix $\\boldsymbol{A}$ of size\n", "$n\\times n$. At each iteration $i+1$ we obtain the conjugate direction of a vector" ] @@ -2206,7 +2257,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Conjugate gradient method\n", "The coefficients are given by" ] }, @@ -2255,8 +2305,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Conjugate gradient method and iterations\n", - "\n", "If we choose the conjugate vectors $\\boldsymbol{p}_k$ carefully, \n", "then we may not need all of them to obtain a good approximation to the solution \n", "$\\boldsymbol{x}$. \n", @@ -2299,11 +2347,6 @@ "source": [ "instead.\n", "\n", - "\n", - "\n", - "\n", - "\n", - "## Conjugate gradient method\n", "One can show that the solution $\\boldsymbol{x}$ is also the unique minimizer of the quadratic form" ] }, @@ -2343,11 +2386,6 @@ "The other vectors in the basis will be conjugate to the gradient, \n", "hence the name conjugate gradient method.\n", "\n", - "\n", - "\n", - "\n", - "\n", - "## Conjugate gradient method\n", "Let $\\boldsymbol{r}_k$ be the residual at the $k$-th step:" ] }, @@ -2386,7 +2424,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Conjugate gradient method\n", "We can also compute the residual iteratively as" ] }, @@ -2451,14 +2488,14 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Revisiting our first homework\n", + "## Revisiting our Linear Regression Solvers\n", "\n", "We will use linear regression as a case study for the gradient descent\n", "methods. Linear regression is a great test case for the gradient\n", "descent methods discussed in the lectures since it has several\n", "desirable properties such as:\n", "\n", - "1. An analytical solution (recall homework set 1).\n", + "1. An analytical solution.\n", "\n", "2. The gradient can be computed analytically.\n", "\n", @@ -2469,12 +2506,24 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 14, "metadata": { "collapsed": false, "editable": true }, - "outputs": [], + "outputs": [ + { + "ename": "NameError", + "evalue": "name 'm' is not defined", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrand\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m4\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m3\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mNameError\u001b[0m: name 'm' is not defined" + ] + } + ], "source": [ "x = 2*np.random.rand(m,1)\n", "y = 4+3*x+np.random.randn(m,1)" @@ -2517,8 +2566,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Gradient descent example\n", - "\n", "Let $\\mathbf{y} = (y_1,\\cdots,y_n)^T$, $\\mathbf{\\boldsymbol{y}} = (\\boldsymbol{y}_1,\\cdots,\\boldsymbol{y}_n)^T$ and $\\beta = (\\beta_0, \\beta_1)^T$\n", "\n", "It is convenient to write $\\mathbf{\\boldsymbol{y}} = X\\beta$ where $X \\in \\mathbb{R}^{100 \\times 2} $ is the design matrix given by (we keep the intercept here)" @@ -2560,8 +2607,6 @@ "and we want to find $\\beta$ such that $C(\\beta)$ is minimized.\n", "\n", "\n", - "## The derivative of the cost/loss function\n", - "\n", "Computing $\\partial C(\\beta) / \\partial \\beta_0$ and $\\partial C(\\beta) / \\partial \\beta_1$ we can show that the gradient can be written as" ] }, @@ -2583,7 +2628,6 @@ "where $X$ is the design matrix defined above.\n", "\n", "\n", - "## The Hessian matrix\n", "The Hessian matrix of $C(\\beta)$ is given by" ] }, @@ -2607,10 +2651,6 @@ "\n", "\n", "\n", - "\n", - "\n", - "## Simple program\n", - "\n", "We can now write a program that minimizes $C(\\beta)$ using the gradient descent method with a constant learning rate $\\gamma$ according to" ] }, @@ -2635,9 +2675,8 @@ "$\\beta= (X^TX)^{-1} X^T \\mathbf{y}$.\n", "\n", "\n", - "## Gradient Descent Example\n", "\n", - "Here our simple example" + "Here is our simple example" ] }, { @@ -2701,7 +2740,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## And a corresponding example using **scikit-learn**" + "Alternatively, we can use **Scikit-Learn** as done here" ] }, { @@ -2735,8 +2774,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Gradient descent and Ridge\n", - "\n", "We have also discussed Ridge regression where the loss function contains a regularized term given by the $L_2$ norm of $\\beta$," ] }, @@ -2783,13 +2820,6 @@ "$$" ] }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Program example for gradient descent with Ridge Regression" - ] - }, { "cell_type": "code", "execution_count": null, @@ -2886,8 +2916,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Computation of gradients\n", - "\n", "This in turn means that the gradient can be\n", "computed as a sum over $i$-gradients" ] @@ -2913,7 +2941,7 @@ "$k=1,\\cdots,n/M$.\n", "\n", "\n", - "## SGD example\n", + "\n", "As an example, suppose we have $10$ data points $(\\mathbf{x}_1,\\cdots, \\mathbf{x}_{10})$ \n", "and we choose to have $M=5$ minibathces,\n", "then each minibatch contains two data points. In particular we have\n", @@ -2944,8 +2972,6 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## The gradient step\n", - "\n", "Thus a gradient descent step now looks like" ] }, @@ -2967,10 +2993,7 @@ "probability from $[1,n/M]$. An iteration over the number of\n", "minibathces (n/M) is commonly referred to as an epoch. Thus it is\n", "typical to choose a number of epochs and for each epoch iterate over\n", - "the number of minibatches, as exemplified in the code below.\n", - "\n", - "\n", - "## Simple example code" + "the number of minibatches, as exemplified in the code below." ] }, { @@ -3011,7 +3034,6 @@ "all $n$ datapoints.\n", "\n", "\n", - "## When do we stop?\n", "\n", "A natural question is when do we stop the search for a new minimum?\n", "One possibility is to compute the full gradient after a given number\n", @@ -3025,7 +3047,6 @@ "gave the lowest value.\n", "\n", "\n", - "## Slightly different approach\n", "\n", "Another approach is to let the step length $\\gamma_j$ depend on the\n", "number of epochs in such a way that it becomes very small after a\n", @@ -3079,7 +3100,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "## Program for stochastic gradient" + "### Program for stochastic gradient" ] }, { @@ -3161,7 +3182,922 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "**Challenge**: try to write a similar code for a Logistic Regression case." + "## Momentum based GD\n", + "\n", + "The stochastic gradient descent (SGD) is almost always used with a\n", + "*momentum* or inertia term that serves as a memory of the direction we\n", + "are moving in parameter space. This is typically implemented as\n", + "follows" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\mathbf{v}_{t}=\\gamma \\mathbf{v}_{t-1}+\\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t) \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation} \n", + "\\boldsymbol{\\theta}_{t+1}= \\boldsymbol{\\theta}_t -\\mathbf{v}_{t},\n", + "\\label{_auto2} \\tag{3}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "where we have introduced a momentum parameter $\\gamma$, with\n", + "$0\\le\\gamma\\le 1$, and for brevity we dropped the explicit notation to\n", + "indicate the gradient is to be taken over a different mini-batch at\n", + "each step. We call this algorithm gradient descent with momentum\n", + "(GDM). From these equations, it is clear that $\\mathbf{v}_t$ is a\n", + "running average of recently encountered gradients and\n", + "$(1-\\gamma)^{-1}$ sets the characteristic time scale for the memory\n", + "used in the averaging procedure. Consistent with this, when\n", + "$\\gamma=0$, this just reduces down to ordinary SGD as discussed\n", + "earlier. An equivalent way of writing the updates is" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\Delta \\boldsymbol{\\theta}_{t+1} = \\gamma \\Delta \\boldsymbol{\\theta}_t -\\ \\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t),\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "where we have defined $\\Delta \\boldsymbol{\\theta}_{t}= \\boldsymbol{\\theta}_t-\\boldsymbol{\\theta}_{t-1}$.\n", + "\n", + "\n", + "\n", + "Let us try to get more intuition from these equations. It is helpful\n", + "to consider a simple physical analogy with a particle of mass $m$\n", + "moving in a viscous medium with drag coefficient $\\mu$ and potential\n", + "$E(\\mathbf{w})$. If we denote the particle's position by $\\mathbf{w}$,\n", + "then its motion is described by" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "m {d^2 \\mathbf{w} \\over dt^2} + \\mu {d \\mathbf{w} \\over dt }= -\\nabla_w E(\\mathbf{w}).\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can discretize this equation in the usual way to get" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "m { \\mathbf{w}_{t+\\Delta t}-2 \\mathbf{w}_{t} +\\mathbf{w}_{t-\\Delta t} \\over (\\Delta t)^2}+\\mu {\\mathbf{w}_{t+\\Delta t}- \\mathbf{w}_{t} \\over \\Delta t} = -\\nabla_w E(\\mathbf{w}).\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Rearranging this equation, we can rewrite this as" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\Delta \\mathbf{w}_{t +\\Delta t}= - { (\\Delta t)^2 \\over m +\\mu \\Delta t} \\nabla_w E(\\mathbf{w})+ {m \\over m +\\mu \\Delta t} \\Delta \\mathbf{w}_t.\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Notice that this equation is identical to previous one if we identify\n", + "the position of the particle, $\\mathbf{w}$, with the parameters\n", + "$\\boldsymbol{\\theta}$. This allows us to identify the momentum\n", + "parameter and learning rate with the mass of the particle and the\n", + "viscous drag as:" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\gamma= {m \\over m +\\mu \\Delta t }, \\qquad \\eta = {(\\Delta t)^2 \\over m +\\mu \\Delta t}.\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Thus, as the name suggests, the momentum parameter is proportional to\n", + "the mass of the particle and effectively provides inertia.\n", + "Furthermore, in the large viscosity/small learning rate limit, our\n", + "memory time scales as $(1-\\gamma)^{-1} \\approx m/(\\mu \\Delta t)$.\n", + "\n", + "Why is momentum useful? SGD momentum helps the gradient descent\n", + "algorithm gain speed in directions with persistent but small gradients\n", + "even in the presence of stochasticity, while suppressing oscillations\n", + "in high-curvature directions. This becomes especially important in\n", + "situations where the landscape is shallow and flat in some directions\n", + "and narrow and steep in others. It has been argued that first-order\n", + "methods (with appropriate initial conditions) can perform comparable\n", + "to more expensive second order methods, especially in the context of\n", + "complex deep learning models.\n", + "\n", + "These beneficial properties of momentum can sometimes become even more\n", + "pronounced by using a slight modification of the classical momentum\n", + "algorithm called Nesterov Accelerated Gradient (NAG).\n", + "\n", + "In the NAG algorithm, rather than calculating the gradient at the\n", + "current parameters, $\\nabla_\\theta E(\\boldsymbol{\\theta}_t)$, one\n", + "calculates the gradient at the expected value of the parameters given\n", + "our current momentum, $\\nabla_\\theta E(\\boldsymbol{\\theta}_t +\\gamma\n", + "\\mathbf{v}_{t-1})$. This yields the NAG update rule" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\mathbf{v}_{t}=\\gamma \\mathbf{v}_{t-1}+\\eta_{t}\\nabla_\\theta E(\\boldsymbol{\\theta}_t +\\gamma \\mathbf{v}_{t-1}) \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation} \n", + "\\boldsymbol{\\theta}_{t+1}= \\boldsymbol{\\theta}_t -\\mathbf{v}_{t}.\n", + "\\label{_auto3} \\tag{4}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "One of the major advantages of NAG is that it allows for the use of a larger learning rate than GDM for the same choice of $\\gamma$.\n", + "\n", + "\n", + "\n", + "In stochastic gradient descent, with and without momentum, we still\n", + "have to specify a schedule for tuning the learning rates $\\eta_t$\n", + "as a function of time. As discussed in the context of Newton's\n", + "method, this presents a number of dilemmas. The learning rate is\n", + "limited by the steepest direction which can change depending on the\n", + "current position in the landscape. To circumvent this problem, ideally\n", + "our algorithm would keep track of curvature and take large steps in\n", + "shallow, flat directions and small steps in steep, narrow directions.\n", + "Second-order methods accomplish this by calculating or approximating\n", + "the Hessian and normalizing the learning rate by the\n", + "curvature. However, this is very computationally expensive for\n", + "extremely large models. Ideally, we would like to be able to\n", + "adaptively change the step size to match the landscape without paying\n", + "the steep computational price of calculating or approximating\n", + "Hessians.\n", + "\n", + "Recently, a number of methods have been introduced that accomplish\n", + "this by tracking not only the gradient, but also the second moment of\n", + "the gradient. These methods include AdaGrad, AdaDelta, RMS-Prop, and\n", + "ADAM.\n", + "\n", + "\n", + "### RMS prop\n", + "\n", + "In RMS prop, in addition to keeping a running average of the first\n", + "moment of the gradient, we also keep track of the second moment\n", + "denoted by $\\mathbf{s}_t=\\mathbb{E}[\\mathbf{g}_t^2]$. The update rule\n", + "for RMS prop is given by" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation}\n", + "\\mathbf{g}_t = \\nabla_\\theta E(\\boldsymbol{\\theta}) \n", + "\\label{_auto4} \\tag{5}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\mathbf{s}_t =\\beta \\mathbf{s}_{t-1} +(1-\\beta)\\mathbf{g}_t^2 \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_t - \\eta_t { \\mathbf{g}_t \\over \\sqrt{\\mathbf{s}_t +\\epsilon}}, \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "where $\\beta$ controls the averaging time of the second moment and is\n", + "typically taken to be about $\\beta=0.9$, $\\eta_t$ is a learning rate\n", + "typically chosen to be $10^{-3}$, and $\\epsilon\\sim 10^{-8} $ is a\n", + "small regularization constant to prevent divergences. Multiplication\n", + "and division by vectors is understood as an element-wise operation. It\n", + "is clear from this formula that the learning rate is reduced in\n", + "directions where the norm of the gradient is consistently large. This\n", + "greatly speeds up the convergence by allowing us to use a larger\n", + "learning rate for flat directions.\n", + "\n", + "### ADAM optimizer\n", + "\n", + "A related algorithm is the ADAM optimizer. In ADAM, we keep a running\n", + "average of both the first and second moment of the gradient and use\n", + "this information to adaptively change the learning rate for different\n", + "parameters. In addition to keeping a running average of the first and\n", + "second moments of the gradient\n", + "(i.e. $\\mathbf{m}_t=\\mathbb{E}[\\mathbf{g}_t]$ and\n", + "$\\mathbf{s}_t=\\mathbb{E}[\\mathbf{g}^2_t]$, respectively), ADAM\n", + "performs an additional bias correction to account for the fact that we\n", + "are estimating the first two moments of the gradient using a running\n", + "average (denoted by the hats in the update rule below). The update\n", + "rule for ADAM is given by (where multiplication and division are once\n", + "again understood to be element-wise operations below)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation}\n", + "\\mathbf{g}_t = \\nabla_\\theta E(\\boldsymbol{\\theta}) \n", + "\\label{_auto5} \\tag{6}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\mathbf{m}_t = \\beta_1 \\mathbf{m}_{t-1} + (1-\\beta_1) \\mathbf{g}_t \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\mathbf{s}_t =\\beta_2 \\mathbf{s}_{t-1} +(1-\\beta_2)\\mathbf{g}_t^2 \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\boldsymbol{\\mathbf{m}}_t={\\mathbf{m}_t \\over 1-\\beta_1^t} \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\boldsymbol{\\mathbf{s}}_t ={\\mathbf{s}_t \\over1-\\beta_2^t} \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_t - \\eta_t { \\boldsymbol{\\mathbf{m}}_t \\over \\sqrt{\\boldsymbol{\\mathbf{s}}_t} +\\epsilon}, \\nonumber\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "\n", + "
    \n", + "\n", + "$$\n", + "\\begin{equation} \n", + "\\label{_auto6} \\tag{7}\n", + "\\end{equation}\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "where $\\beta_1$ and $\\beta_2$ set the memory lifetime of the first and\n", + "second moment and are typically taken to be $0.9$ and $0.99$\n", + "respectively, and $\\eta$ and $\\epsilon$ are identical to RMSprop.\n", + "\n", + "Like in RMSprop, the effective step size of a parameter depends on the\n", + "magnitude of its gradient squared. To understand this better, let us\n", + "rewrite this expression in terms of the variance\n", + "$\\boldsymbol{\\sigma}_t^2 = \\boldsymbol{\\mathbf{s}}_t -\n", + "(\\boldsymbol{\\mathbf{m}}_t)^2$. Consider a single parameter $\\theta_t$. The\n", + "update rule for this parameter is given by" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "\\Delta \\theta_{t+1}= -\\eta_t { \\boldsymbol{m}_t \\over \\sqrt{\\sigma_t^2 + m_t^2 }+\\epsilon}.\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Practical tips\n", + "\n", + "* **Randomize the data when making mini-batches**. It is always important to randomly shuffle the data when forming mini-batches. Otherwise, the gradient descent method can fit spurious correlations resulting from the order in which data is presented.\n", + "\n", + "* **Transform your inputs**. Learning becomes difficult when our landscape has a mixture of steep and flat directions. One simple trick for minimizing these situations is to standardize the data by subtracting the mean and normalizing the variance of input variables. Whenever possible, also decorrelate the inputs. To understand why this is helpful, consider the case of linear regression. It is easy to show that for the squared error cost function, the Hessian of the cost function is just the correlation matrix between the inputs. Thus, by standardizing the inputs, we are ensuring that the landscape looks homogeneous in all directions in parameter space. Since most deep networks can be viewed as linear transformations followed by a non-linearity at each layer, we expect this intuition to hold beyond the linear case.\n", + "\n", + "* **Monitor the out-of-sample performance.** Always monitor the performance of your model on a validation set (a small portion of the training data that is held out of the training process to serve as a proxy for the test set. If the validation error starts increasing, then the model is beginning to overfit. Terminate the learning process. This *early stopping* significantly improves performance in many settings.\n", + "\n", + "* **Adaptive optimization methods don't always have good generalization.** Recent studies have shown that adaptive methods such as ADAM, RMSPorp, and AdaGrad tend to have poor generalization compared to SGD or SGD with momentum, particularly in the high-dimensional limit (i.e. the number of parameters exceeds the number of data points). Although it is not clear at this stage why these methods perform so well in training deep neural networks, simpler procedures like properly-tuned SGD may work as well or better in these applications.\n", + "\n", + "## Automatic differentiation\n", + "\n", + "[Automatic differentiation (AD)](https://en.wikipedia.org/wiki/Automatic_differentiation), \n", + "also called algorithmic\n", + "differentiation or computational differentiation,is a set of\n", + "techniques to numerically evaluate the derivative of a function\n", + "specified by a computer program. AD exploits the fact that every\n", + "computer program, no matter how complicated, executes a sequence of\n", + "elementary arithmetic operations (addition, subtraction,\n", + "multiplication, division, etc.) and elementary functions (exp, log,\n", + "sin, cos, etc.). By applying the chain rule repeatedly to these\n", + "operations, derivatives of arbitrary order can be computed\n", + "automatically, accurately to working precision, and using at most a\n", + "small constant factor more arithmetic operations than the original\n", + "program.\n", + "\n", + "Automatic differentiation is neither:\n", + "\n", + "* Symbolic differentiation, nor\n", + "\n", + "* Numerical differentiation (the method of finite differences).\n", + "\n", + "Symbolic differentiation can lead to inefficient code and faces the\n", + "difficulty of converting a computer program into a single expression,\n", + "while numerical differentiation can introduce round-off errors in the\n", + "discretization process and cancellation\n", + "\n", + "\n", + "\n", + "Python has tools for so-called **automatic differentiation**.\n", + "Consider the following example" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "f(x) = \\sin\\left(2\\pi x + x^2\\right)\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "which has the following derivative" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "$$\n", + "f'(x) = \\cos\\left(2\\pi x + x^2\\right)\\left(2\\pi + 2x\\right)\n", + "$$" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Using **autograd** we have" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "\n", + "# To do elementwise differentiation:\n", + "from autograd import elementwise_grad as egrad \n", + "\n", + "# To plot:\n", + "import matplotlib.pyplot as plt \n", + "\n", + "\n", + "def f(x):\n", + " return np.sin(2*np.pi*x + x**2)\n", + "\n", + "def f_grad_analytic(x):\n", + " return np.cos(2*np.pi*x + x**2)*(2*np.pi + 2*x)\n", + "\n", + "# Do the comparison:\n", + "x = np.linspace(0,1,1000)\n", + "\n", + "f_grad = egrad(f)\n", + "\n", + "computed = f_grad(x)\n", + "analytic = f_grad_analytic(x)\n", + "\n", + "plt.title('Derivative computed from Autograd compared with the analytical derivative')\n", + "plt.plot(x,computed,label='autograd')\n", + "plt.plot(x,analytic,label='analytic')\n", + "\n", + "plt.xlabel('x')\n", + "plt.ylabel('y')\n", + "plt.legend()\n", + "\n", + "plt.show()\n", + "\n", + "print(\"The max absolute difference is: %g\"%(np.max(np.abs(computed - analytic))))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here we\n", + "experiment with what kind of functions Autograd is capable\n", + "of finding the gradient of. The following Python functions are just\n", + "meant to illustrate what Autograd can do, but please feel free to\n", + "experiment with other, possibly more complicated, functions as well." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "\n", + "def f1(x):\n", + " return x**3 + 1\n", + "\n", + "f1_grad = grad(f1)\n", + "\n", + "# Remember to send in float as argument to the computed gradient from Autograd!\n", + "a = 1.0\n", + "\n", + "# See the evaluated gradient at a using autograd:\n", + "print(\"The gradient of f1 evaluated at a = %g using autograd is: %g\"%(a,f1_grad(a)))\n", + "\n", + "# Compare with the analytical derivative, that is f1'(x) = 3*x**2 \n", + "grad_analytical = 3*a**2\n", + "print(\"The gradient of f1 evaluated at a = %g by finding the analytic expression is: %g\"%(a,grad_analytical))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "To differentiate with respect to two (or more) arguments of a Python\n", + "function, Autograd need to know at which variable the function if\n", + "being differentiated with respect to." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f2(x1,x2):\n", + " return 3*x1**3 + x2*(x1 - 5) + 1\n", + "\n", + "# By sending the argument 0, Autograd will compute the derivative w.r.t the first variable, in this case x1\n", + "f2_grad_x1 = grad(f2,0)\n", + "\n", + "# ... and differentiate w.r.t x2 by sending 1 as an additional arugment to grad\n", + "f2_grad_x2 = grad(f2,1)\n", + "\n", + "x1 = 1.0\n", + "x2 = 3.0 \n", + "\n", + "print(\"Evaluating at x1 = %g, x2 = %g\"%(x1,x2))\n", + "print(\"-\"*30)\n", + "\n", + "# Compare with the analytical derivatives:\n", + "\n", + "# Derivative of f2 w.r.t x1 is: 9*x1**2 + x2:\n", + "f2_grad_x1_analytical = 9*x1**2 + x2\n", + "\n", + "# Derivative of f2 w.r.t x2 is: x1 - 5:\n", + "f2_grad_x2_analytical = x1 - 5\n", + "\n", + "# See the evaluated derivations:\n", + "print(\"The derivative of f2 w.r.t x1: %g\"%( f2_grad_x1(x1,x2) ))\n", + "print(\"The analytical derivative of f2 w.r.t x1: %g\"%( f2_grad_x1(x1,x2) ))\n", + "\n", + "print()\n", + "\n", + "print(\"The derivative of f2 w.r.t x2: %g\"%( f2_grad_x2(x1,x2) ))\n", + "print(\"The analytical derivative of f2 w.r.t x2: %g\"%( f2_grad_x2(x1,x2) ))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Note that the grad function will not produce the true gradient of the function. The true gradient of a function with two or more variables will produce a vector, where each element is the function differentiated w.r.t a variable." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f3(x): # Assumes x is an array of length 5 or higher\n", + " return 2*x[0] + 3*x[1] + 5*x[2] + 7*x[3] + 11*x[4]**2\n", + "\n", + "f3_grad = grad(f3)\n", + "\n", + "x = np.linspace(0,4,5)\n", + "\n", + "# Print the computed gradient:\n", + "print(\"The computed gradient of f3 is: \", f3_grad(x))\n", + "\n", + "# The analytical gradient is: (2, 3, 5, 7, 22*x[4])\n", + "f3_grad_analytical = np.array([2, 3, 5, 7, 22*x[4]])\n", + "\n", + "# Print the analytical gradient:\n", + "print(\"The analytical gradient of f3 is: \", f3_grad_analytical)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Note that in this case, when sending an array as input argument, the\n", + "output from Autograd is another array. This is the true gradient of\n", + "the function, as opposed to the function in the previous example. By\n", + "using arrays to represent the variables, the output from Autograd\n", + "might be easier to work with, as the output is closer to what one\n", + "could expect form a gradient-evaluting function." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f4(x):\n", + " return np.sqrt(1+x**2) + np.exp(x) + np.sin(2*np.pi*x)\n", + "\n", + "f4_grad = grad(f4)\n", + "\n", + "x = 2.7\n", + "\n", + "# Print the computed derivative:\n", + "print(\"The computed derivative of f4 at x = %g is: %g\"%(x,f4_grad(x)))\n", + "\n", + "# The analytical derivative is: x/sqrt(1 + x**2) + exp(x) + cos(2*pi*x)*2*pi\n", + "f4_grad_analytical = x/np.sqrt(1 + x**2) + np.exp(x) + np.cos(2*np.pi*x)*2*np.pi\n", + "\n", + "# Print the analytical gradient:\n", + "print(\"The analytical gradient of f4 at x = %g is: %g\"%(x,f4_grad_analytical))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f5(x):\n", + " if x >= 0:\n", + " return x**2\n", + " else:\n", + " return -3*x + 1\n", + "\n", + "f5_grad = grad(f5)\n", + "\n", + "x = 2.7\n", + "\n", + "# Print the computed derivative:\n", + "print(\"The computed derivative of f5 at x = %g is: %g\"%(x,f5_grad(x)))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "2\n", + "6\n", + " \n", + "<\n", + "<\n", + "<\n", + "!\n", + "!\n", + "C\n", + "O\n", + "D\n", + "E\n", + "_\n", + "B\n", + "L\n", + "O\n", + "C\n", + "K\n", + " \n", + " \n", + "p\n", + "y\n", + "c\n", + "o\n", + "d" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "# Both of the functions are implementation of the sum: sum(x**i) for i = 0, ..., 9\n", + "# The analytical derivative is: sum(i*x**(i-1)) \n", + "f6_grad_analytical = 0\n", + "for i in range(10):\n", + " f6_grad_analytical += i*x**(i-1)\n", + "\n", + "print(\"The analytical derivative of f6 at x = %g is: %g\"%(x,f6_grad_analytical))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "\n", + "def f7(n): # Assume that n is an integer\n", + " if n == 1 or n == 0:\n", + " return 1\n", + " else:\n", + " return n*f7(n-1)\n", + "\n", + "f7_grad = grad(f7)\n", + "\n", + "n = 2.0\n", + "\n", + "print(\"The computed derivative of f7 at n = %d is: %g\"%(n,f7_grad(n)))\n", + "\n", + "# The function f7 is an implementation of the factorial of n.\n", + "# By using the product rule, one can find that the derivative is:\n", + "\n", + "f7_grad_analytical = 0\n", + "for i in range(int(n)-1):\n", + " tmp = 1\n", + " for k in range(int(n)-1):\n", + " if k != i:\n", + " tmp *= (n - k)\n", + " f7_grad_analytical += tmp\n", + "\n", + "print(\"The analytical derivative of f7 at n = %d is: %g\"%(n,f7_grad_analytical))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Note that if n is equal to zero or one, Autograd will give an error message. This message appears when the output is independent on input.\n", + "\n", + "\n", + "Autograd supports many features. However, there are some functions that is not supported (yet) by Autograd.\n", + "\n", + "Assigning a value to the variable being differentiated with respect to" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f8(x): # Assume x is an array\n", + " x[2] = 3\n", + " return x*2\n", + "\n", + "f8_grad = grad(f8)\n", + "\n", + "x = 8.4\n", + "\n", + "print(\"The derivative of f8 is:\",f8_grad(x))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here, Autograd tells us that an 'ArrayBox' does not support item assignment. The item assignment is done when the program tries to assign x[2] to the value 3. However, Autograd has implemented the computation of the derivative such that this assignment is not possible." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f9(a): # Assume a is an array with 2 elements\n", + " b = np.array([1.0,2.0])\n", + " return a.dot(b)\n", + "\n", + "f9_grad = grad(f9)\n", + "\n", + "x = np.array([1.0,0.0])\n", + "\n", + "print(\"The derivative of f9 is:\",f9_grad(x))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Here we are told that the 'dot' function does not belong to Autograd's\n", + "version of a Numpy array. To overcome this, an alternative syntax\n", + "which also computed the dot product can be used:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "import autograd.numpy as np\n", + "from autograd import grad\n", + "def f9_alternative(x): # Assume a is an array with 2 elements\n", + " b = np.array([1.0,2.0])\n", + " return np.dot(x,b) # The same as x_1*b_1 + x_2*b_2\n", + "\n", + "f9_alternative_grad = grad(f9_alternative)\n", + "\n", + "x = np.array([3.0,0.0])\n", + "\n", + "print(\"The gradient of f9 is:\",f9_alternative_grad(x))\n", + "\n", + "# The analytical gradient of the dot product of vectors x and b with two elements (x_1,x_2) and (b_1, b_2) respectively\n", + "# w.r.t x is (b_1, b_2)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "The documentation recommends to avoid inplace operations such as" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": false, + "editable": true + }, + "outputs": [], + "source": [ + "a += b\n", + "a -= b\n", + "a*= b\n", + "a /=b" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "More examples will be added, in particular how to compare autograd with own codes for the gradients." ] } ], diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter4.py b/doc/LectureNotes/_build/jupyter_execute/chapter4.py index 8e216da7c..8b1ee3013 100644 --- a/doc/LectureNotes/_build/jupyter_execute/chapter4.py +++ b/doc/LectureNotes/_build/jupyter_execute/chapter4.py @@ -1,8 +1,6 @@ # Logistic Regression -[Video of Lecture](https://www.uio.no/studier/emner/matnat/fys/FYS-STK3155/h20/forelesningsvideoer/LectureSeptember18.mp4?vrtx=view-as-webpage) - ## Logistic Regression @@ -575,7 +573,7 @@ plt.show() skplt.metrics.plot_cumulative_gain(y_test, y_probas) plt.show() -## Optimization, the central part of any Machine Learning algortithm +# Optimization, the central part of any Machine Learning algortithm Almost every problem in machine learning and data science starts with a dataset $X$, a model $g(\beta)$, which is a function of the @@ -588,7 +586,6 @@ some approximative/numerical method to compute the minimum. -## Revisiting our Logistic Regression case In our discussion on Logistic Regression we studied the case of @@ -607,7 +604,6 @@ $$ where $\boldsymbol{\beta}$ are the weights we wish to extract from data, in our case $\beta_0$ and $\beta_1$. -## The equations to solve Our compact equations used a definition of a vector $\boldsymbol{y}$ with $n$ elements $y_i$, an $n\times p$ matrix $\boldsymbol{X}$ which contains the @@ -629,7 +625,6 @@ $$ This defines what is called the Hessian matrix. -## Solving using Newton-Raphson's method If we can set up these equations, Newton-Raphson's iterative method is normally the method of choice. It requires however that we can compute in an efficient way the matrices that define the first and second derivatives. @@ -650,9 +645,6 @@ The right-hand side is computed with the old values of $\beta$. If we can compute these matrices, in particular the Hessian, the above is often the easiest method to implement. - -## Brief reminder on Newton-Raphson's method - Let us quickly remind ourselves how we derive the above method. Perhaps the most celebrated of all one-dimensional root-finding @@ -664,7 +656,6 @@ numerically and/or your function is not of the smooth type, we normally discourage the use of this method. -## The equations The Newton-Raphson formula consists geometrically of extending the tangent line at a current point until it crosses zero, then setting @@ -699,8 +690,6 @@ $$ x_{n+1}=x_n-\frac{f(x_n)}{f'(x_n)}. $$ -## Simple geometric interpretation - The above is Newton-Raphson's method. It has a simple geometric interpretation, namely $x_{n+1}$ is the point where the tangent from $(x_n,f(x_n))$ crosses the $x$-axis. Close to the solution, @@ -715,7 +704,6 @@ vanishes, then Newton-Raphson may fail totally -## Extending to more than one variable Newton's method can be generalized to systems of several non-linear equations and variables. Consider the case with two equations @@ -792,8 +780,6 @@ F(\mathbf{x}_k)$. This means that for a sufficiently small $\gamma_k$ we are always moving towards smaller function values, i.e a minimum. -## More on Steepest descent - The previous observation is the basis of the method of steepest descent, which is also referred to as just gradient descent (GD). One starts with an initial guess $\mathbf{x}_0$ for a minimum of $F$ and @@ -807,8 +793,6 @@ The parameter $\gamma_k$ is often referred to as the step length or the learning rate within the context of Machine Learning. -## The ideal - Ideally the sequence $\{\mathbf{x}_k \}_{k=0}$ converges to a global minimum of the function $F$. In general we do not know if we are in a global or local minimum. In the special case when $F$ is a convex @@ -829,7 +813,6 @@ Note that the gradient is a function of $\mathbf{x} = -## The sensitiveness of the gradient descent The gradient descent method is sensitive to the choice of learning rate $\gamma_k$. This is due @@ -860,13 +843,10 @@ $\mathbb{R}$. Examples of convex sets of $\mathbb{R}^2$ are the regular polygons (triangles, rectangles, pentagons, etc...). -## Convex function **Convex function**: Let $X \subset \mathbb{R}^n$ be a convex set. Assume that the function $f: X \rightarrow \mathbb{R}$ is continuous, then $f$ is said to be convex if $$f(tx_1 + (1-t)x_2) \leq tf(x_1) + (1-t)f(x_2) $$ for all $x_1, x_2 \in X$ and for all $t \in [0,1]$. If $\leq$ is replaced with a strict inequaltiy in the definition, we demand $x_1 \neq x_2$ and $t\in(0,1)$ then $f$ is said to be strictly convex. For a single variable function, convexity means that if you draw a straight line connecting $f(x_1)$ and $f(x_2)$, the value of the function on the interval $[x_1,x_2]$ is always below the line as illustrated below. -## Conditions on convex functions - In the following we state first and second-order conditions which ensures convexity of a function $f$. We write $D_f$ to denote the domain of $f$, i.e the subset of $R^n$ where $f$ is defined. For more @@ -899,7 +879,6 @@ everywhere. This condition is particularly useful since it gives us an procedure for determining if the function under consideration is convex, apart from using the definition. -## More on convex functions The next result is of great importance to us and the reason why we are going on about convex functions. In machine learning we frequently @@ -922,7 +901,7 @@ $x^*$ that satisfies $\nabla f(x^*) = 0$ is a global minimum. This result means that if we know that the cost/loss function is convex and we are able to find a minimum, we are guaranteed that it is a global minimum. -## Some simple problems +### Some simple problems 1. Show that $f(x)=x^2$ is convex for $x \in \mathbb{R}$ using the definition of convexity. Hint: If you re-write the definition, $f$ is convex if the following holds for all $x,y \in D_f$ and any $\lambda \in [0,1]$ $\lambda f(x)+(1-\lambda)f(y)-f(\lambda x + (1-\lambda) y ) \geq 0$. @@ -947,13 +926,6 @@ This result means that if we know that the cost/loss function is convex and we a Using the definition of convexity, try to show that a function satisfying the properties above is convex (the third condition is not needed to show this). - -## Friday September 25 - -[Video of Lecture](https://www.uio.no/studier/emner/matnat/fys/FYS-STK4155/h20/forelesningsvideoer/LectureSeptember25.mp4?vrtx=view-as-webpage) and [link to handwritten notes](https://github.com/CompPhysics/MachineLearning/blob/master/doc/HandWrittenNotes/NotesSeptember25.pdf). - - - ## Standard steepest descent @@ -982,7 +954,6 @@ where $\boldsymbol{r}$ is the so-called residual or error in the iterative proce When we have found the exact solution, $\boldsymbol{r}=0$. -## Gradient method The residual is zero when we reach the minimum of the quadratic equation @@ -994,9 +965,6 @@ with the constraint that the matrix $\boldsymbol{A}$ is positive definite and symmetric. This defines also the Hessian and we want it to be positive definite. - -## Steepest descent method - We denote the initial guess for $\boldsymbol{x}$ as $\boldsymbol{x}_0$. We can assume without loss of generality that @@ -1013,8 +981,6 @@ $$ instead. - -## Steepest descent method One can show that the solution $\boldsymbol{x}$ is also the unique minimizer of the quadratic form $$ @@ -1033,9 +999,6 @@ and $\boldsymbol{x}_0=0$ it is equal $-\boldsymbol{b}$. - - -## Final expressions We can compute the residual iteratively as $$ @@ -1066,8 +1029,6 @@ $$ \boldsymbol{x}_{k+1}=\boldsymbol{x}_k-\alpha_k\boldsymbol{r}_{k}, $$ -## Steepest descent example - import numpy as np import numpy.linalg as la @@ -1137,23 +1098,14 @@ $$ Two vectors are conjugate if they are orthogonal with respect to this inner product. Being conjugate is a symmetric relation: if $\boldsymbol{s}$ is conjugate to $\boldsymbol{t}$, then $\boldsymbol{t}$ is conjugate to $\boldsymbol{s}$. - - - -## Conjugate gradient method An example is given by the eigenvectors of the matrix $$ \boldsymbol{v}_i^T\boldsymbol{A}\boldsymbol{v}_j= \lambda\boldsymbol{v}_i^T\boldsymbol{v}_j, $$ -which is zero unless $i=j$. +which is zero unless $i=j$. - - - - -## Conjugate gradient method Assume now that we have a symmetric positive-definite matrix $\boldsymbol{A}$ of size $n\times n$. At each iteration $i+1$ we obtain the conjugate direction of a vector @@ -1169,7 +1121,6 @@ $$ \boldsymbol{x} = \sum^{n}_{i=1} \alpha_i \boldsymbol{p}_i. $$ -## Conjugate gradient method The coefficients are given by $$ @@ -1188,8 +1139,6 @@ $$ \alpha_k = \frac{\boldsymbol{p}_k^T \boldsymbol{b}}{\boldsymbol{p}_k^T \boldsymbol{A} \boldsymbol{p}_k} $$ -## Conjugate gradient method and iterations - If we choose the conjugate vectors $\boldsymbol{p}_k$ carefully, then we may not need all of them to obtain a good approximation to the solution $\boldsymbol{x}$. @@ -1212,11 +1161,6 @@ $$ instead. - - - - -## Conjugate gradient method One can show that the solution $\boldsymbol{x}$ is also the unique minimizer of the quadratic form $$ @@ -1236,11 +1180,6 @@ $\boldsymbol{x}_0=0$ it is equal $-\boldsymbol{b}$. The other vectors in the basis will be conjugate to the gradient, hence the name conjugate gradient method. - - - - -## Conjugate gradient method Let $\boldsymbol{r}_k$ be the residual at the $k$-th step: $$ @@ -1259,7 +1198,6 @@ $$ \boldsymbol{p}_{k+1}=\boldsymbol{r}_k-\frac{\boldsymbol{p}_k^T \boldsymbol{A}\boldsymbol{r}_k}{\boldsymbol{p}_k^T\boldsymbol{A}\boldsymbol{p}_k} \boldsymbol{p}_k. $$ -## Conjugate gradient method We can also compute the residual iteratively as $$ @@ -1284,14 +1222,14 @@ $$ \boldsymbol{r}_{k+1}=\boldsymbol{r}_k-\boldsymbol{A}\boldsymbol{p}_{k}, $$ -## Revisiting our first homework +## Revisiting our Linear Regression Solvers We will use linear regression as a case study for the gradient descent methods. Linear regression is a great test case for the gradient descent methods discussed in the lectures since it has several desirable properties such as: -1. An analytical solution (recall homework set 1). +1. An analytical solution. 2. The gradient can be computed analytically. @@ -1315,8 +1253,6 @@ $$ \boldsymbol{y}_i = \beta_0 + \beta_1 x_i. $$ -## Gradient descent example - Let $\mathbf{y} = (y_1,\cdots,y_n)^T$, $\mathbf{\boldsymbol{y}} = (\boldsymbol{y}_1,\cdots,\boldsymbol{y}_n)^T$ and $\beta = (\beta_0, \beta_1)^T$ It is convenient to write $\mathbf{\boldsymbol{y}} = X\beta$ where $X \in \mathbb{R}^{100 \times 2} $ is the design matrix given by (we keep the intercept here) @@ -1338,8 +1274,6 @@ $$ and we want to find $\beta$ such that $C(\beta)$ is minimized. -## The derivative of the cost/loss function - Computing $\partial C(\beta) / \partial \beta_0$ and $\partial C(\beta) / \partial \beta_1$ we can show that the gradient can be written as $$ @@ -1351,7 +1285,6 @@ $$ where $X$ is the design matrix defined above. -## The Hessian matrix The Hessian matrix of $C(\beta)$ is given by $$ @@ -1365,10 +1298,6 @@ This result implies that $C(\beta)$ is a convex function since the matrix $X^T X - - -## Simple program - We can now write a program that minimizes $C(\beta)$ using the gradient descent method with a constant learning rate $\gamma$ according to $$ @@ -1383,9 +1312,8 @@ And finally we can compare our solution for $\beta$ with the analytic result giv $\beta= (X^TX)^{-1} X^T \mathbf{y}$. -## Gradient Descent Example -Here our simple example +Here is our simple example # Importing various packages @@ -1434,7 +1362,7 @@ plt.ylabel(r'$y$') plt.title(r'Gradient descent example') plt.show() -## And a corresponding example using **scikit-learn** +Alternatively, we can use **Scikit-Learn** as done here # Importing various packages from random import random, seed @@ -1453,8 +1381,6 @@ sgdreg = SGDRegressor(max_iter = 50, penalty=None, eta0=0.1) sgdreg.fit(x,y.ravel()) print(sgdreg.intercept_, sgdreg.coef_) -## Gradient descent and Ridge - We have also discussed Ridge regression where the loss function contains a regularized term given by the $L_2$ norm of $\beta$, $$ @@ -1475,8 +1401,6 @@ $$ \beta_{\text{ridge}} = \left(X^T X + \lambda I_{2 \times 2} \right)^{-1} X^T \mathbf{y}. $$ -## Program example for gradient descent with Ridge Regression - from random import random, seed import numpy as np import matplotlib.pyplot as plt @@ -1549,8 +1473,6 @@ C(\mathbf{\beta}) = \sum_{i=1}^n c_i(\mathbf{x}_i, \mathbf{\beta}). $$ -## Computation of gradients - This in turn means that the gradient can be computed as a sum over $i$-gradients @@ -1566,7 +1488,7 @@ minibatches. We denote these minibatches by $B_k$ where $k=1,\cdots,n/M$. -## SGD example + As an example, suppose we have $10$ data points $(\mathbf{x}_1,\cdots, \mathbf{x}_{10})$ and we choose to have $M=5$ minibathces, then each minibatch contains two data points. In particular we have @@ -1587,8 +1509,6 @@ C(\mathbf{\beta}) = \sum_{i=1}^n \nabla_\beta c_i(\mathbf{x}_i, c_i(\mathbf{x}_i, \mathbf{\beta}). $$ -## The gradient step - Thus a gradient descent step now looks like $$ @@ -1602,9 +1522,6 @@ minibathces (n/M) is commonly referred to as an epoch. Thus it is typical to choose a number of epochs and for each epoch iterate over the number of minibatches, as exemplified in the code below. - -## Simple example code - import numpy as np n = 100 #100 datapoints @@ -1629,7 +1546,6 @@ cheaper since we sum over the datapoints in the $k-th$ minibatch and not all $n$ datapoints. -## When do we stop? A natural question is when do we stop the search for a new minimum? One possibility is to compute the full gradient after a given number @@ -1643,7 +1559,6 @@ compare the values of the cost function and keep the $\beta$ that gave the lowest value. -## Slightly different approach Another approach is to let the step length $\gamma_j$ depend on the number of epochs in such a way that it becomes very small after a @@ -1682,7 +1597,7 @@ for epoch in range(1,n_epochs+1): print("gamma_j after %d epochs: %g" % (n_epochs,gamma_j)) -## Program for stochastic gradient +### Program for stochastic gradient # Importing various packages from math import exp, sqrt @@ -1749,4 +1664,579 @@ plt.ylabel(r'$y$') plt.title(r'Random numbers ') plt.show() -**Challenge**: try to write a similar code for a Logistic Regression case. \ No newline at end of file +## Momentum based GD + +The stochastic gradient descent (SGD) is almost always used with a +*momentum* or inertia term that serves as a memory of the direction we +are moving in parameter space. This is typically implemented as +follows + +$$ +\mathbf{v}_{t}=\gamma \mathbf{v}_{t-1}+\eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t) \nonumber +$$ + + +
    + +$$ +\begin{equation} +\boldsymbol{\theta}_{t+1}= \boldsymbol{\theta}_t -\mathbf{v}_{t}, +\label{_auto2} \tag{3} +\end{equation} +$$ + +where we have introduced a momentum parameter $\gamma$, with +$0\le\gamma\le 1$, and for brevity we dropped the explicit notation to +indicate the gradient is to be taken over a different mini-batch at +each step. We call this algorithm gradient descent with momentum +(GDM). From these equations, it is clear that $\mathbf{v}_t$ is a +running average of recently encountered gradients and +$(1-\gamma)^{-1}$ sets the characteristic time scale for the memory +used in the averaging procedure. Consistent with this, when +$\gamma=0$, this just reduces down to ordinary SGD as discussed +earlier. An equivalent way of writing the updates is + +$$ +\Delta \boldsymbol{\theta}_{t+1} = \gamma \Delta \boldsymbol{\theta}_t -\ \eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t), +$$ + +where we have defined $\Delta \boldsymbol{\theta}_{t}= \boldsymbol{\theta}_t-\boldsymbol{\theta}_{t-1}$. + + + +Let us try to get more intuition from these equations. It is helpful +to consider a simple physical analogy with a particle of mass $m$ +moving in a viscous medium with drag coefficient $\mu$ and potential +$E(\mathbf{w})$. If we denote the particle's position by $\mathbf{w}$, +then its motion is described by + +$$ +m {d^2 \mathbf{w} \over dt^2} + \mu {d \mathbf{w} \over dt }= -\nabla_w E(\mathbf{w}). +$$ + +We can discretize this equation in the usual way to get + +$$ +m { \mathbf{w}_{t+\Delta t}-2 \mathbf{w}_{t} +\mathbf{w}_{t-\Delta t} \over (\Delta t)^2}+\mu {\mathbf{w}_{t+\Delta t}- \mathbf{w}_{t} \over \Delta t} = -\nabla_w E(\mathbf{w}). +$$ + +Rearranging this equation, we can rewrite this as + +$$ +\Delta \mathbf{w}_{t +\Delta t}= - { (\Delta t)^2 \over m +\mu \Delta t} \nabla_w E(\mathbf{w})+ {m \over m +\mu \Delta t} \Delta \mathbf{w}_t. +$$ + +Notice that this equation is identical to previous one if we identify +the position of the particle, $\mathbf{w}$, with the parameters +$\boldsymbol{\theta}$. This allows us to identify the momentum +parameter and learning rate with the mass of the particle and the +viscous drag as: + +$$ +\gamma= {m \over m +\mu \Delta t }, \qquad \eta = {(\Delta t)^2 \over m +\mu \Delta t}. +$$ + +Thus, as the name suggests, the momentum parameter is proportional to +the mass of the particle and effectively provides inertia. +Furthermore, in the large viscosity/small learning rate limit, our +memory time scales as $(1-\gamma)^{-1} \approx m/(\mu \Delta t)$. + +Why is momentum useful? SGD momentum helps the gradient descent +algorithm gain speed in directions with persistent but small gradients +even in the presence of stochasticity, while suppressing oscillations +in high-curvature directions. This becomes especially important in +situations where the landscape is shallow and flat in some directions +and narrow and steep in others. It has been argued that first-order +methods (with appropriate initial conditions) can perform comparable +to more expensive second order methods, especially in the context of +complex deep learning models. + +These beneficial properties of momentum can sometimes become even more +pronounced by using a slight modification of the classical momentum +algorithm called Nesterov Accelerated Gradient (NAG). + +In the NAG algorithm, rather than calculating the gradient at the +current parameters, $\nabla_\theta E(\boldsymbol{\theta}_t)$, one +calculates the gradient at the expected value of the parameters given +our current momentum, $\nabla_\theta E(\boldsymbol{\theta}_t +\gamma +\mathbf{v}_{t-1})$. This yields the NAG update rule + +$$ +\mathbf{v}_{t}=\gamma \mathbf{v}_{t-1}+\eta_{t}\nabla_\theta E(\boldsymbol{\theta}_t +\gamma \mathbf{v}_{t-1}) \nonumber +$$ + + +
    + +$$ +\begin{equation} +\boldsymbol{\theta}_{t+1}= \boldsymbol{\theta}_t -\mathbf{v}_{t}. +\label{_auto3} \tag{4} +\end{equation} +$$ + +One of the major advantages of NAG is that it allows for the use of a larger learning rate than GDM for the same choice of $\gamma$. + + + +In stochastic gradient descent, with and without momentum, we still +have to specify a schedule for tuning the learning rates $\eta_t$ +as a function of time. As discussed in the context of Newton's +method, this presents a number of dilemmas. The learning rate is +limited by the steepest direction which can change depending on the +current position in the landscape. To circumvent this problem, ideally +our algorithm would keep track of curvature and take large steps in +shallow, flat directions and small steps in steep, narrow directions. +Second-order methods accomplish this by calculating or approximating +the Hessian and normalizing the learning rate by the +curvature. However, this is very computationally expensive for +extremely large models. Ideally, we would like to be able to +adaptively change the step size to match the landscape without paying +the steep computational price of calculating or approximating +Hessians. + +Recently, a number of methods have been introduced that accomplish +this by tracking not only the gradient, but also the second moment of +the gradient. These methods include AdaGrad, AdaDelta, RMS-Prop, and +ADAM. + + +### RMS prop + +In RMS prop, in addition to keeping a running average of the first +moment of the gradient, we also keep track of the second moment +denoted by $\mathbf{s}_t=\mathbb{E}[\mathbf{g}_t^2]$. The update rule +for RMS prop is given by + + +
    + +$$ +\begin{equation} +\mathbf{g}_t = \nabla_\theta E(\boldsymbol{\theta}) +\label{_auto4} \tag{5} +\end{equation} +$$ + +$$ +\mathbf{s}_t =\beta \mathbf{s}_{t-1} +(1-\beta)\mathbf{g}_t^2 \nonumber +$$ + +$$ +\boldsymbol{\theta}_{t+1}=\boldsymbol{\theta}_t - \eta_t { \mathbf{g}_t \over \sqrt{\mathbf{s}_t +\epsilon}}, \nonumber +$$ + +where $\beta$ controls the averaging time of the second moment and is +typically taken to be about $\beta=0.9$, $\eta_t$ is a learning rate +typically chosen to be $10^{-3}$, and $\epsilon\sim 10^{-8} $ is a +small regularization constant to prevent divergences. Multiplication +and division by vectors is understood as an element-wise operation. It +is clear from this formula that the learning rate is reduced in +directions where the norm of the gradient is consistently large. This +greatly speeds up the convergence by allowing us to use a larger +learning rate for flat directions. + +### ADAM optimizer + +A related algorithm is the ADAM optimizer. In ADAM, we keep a running +average of both the first and second moment of the gradient and use +this information to adaptively change the learning rate for different +parameters. In addition to keeping a running average of the first and +second moments of the gradient +(i.e. $\mathbf{m}_t=\mathbb{E}[\mathbf{g}_t]$ and +$\mathbf{s}_t=\mathbb{E}[\mathbf{g}^2_t]$, respectively), ADAM +performs an additional bias correction to account for the fact that we +are estimating the first two moments of the gradient using a running +average (denoted by the hats in the update rule below). The update +rule for ADAM is given by (where multiplication and division are once +again understood to be element-wise operations below) + + +
    + +$$ +\begin{equation} +\mathbf{g}_t = \nabla_\theta E(\boldsymbol{\theta}) +\label{_auto5} \tag{6} +\end{equation} +$$ + +$$ +\mathbf{m}_t = \beta_1 \mathbf{m}_{t-1} + (1-\beta_1) \mathbf{g}_t \nonumber +$$ + +$$ +\mathbf{s}_t =\beta_2 \mathbf{s}_{t-1} +(1-\beta_2)\mathbf{g}_t^2 \nonumber +$$ + +$$ +\boldsymbol{\mathbf{m}}_t={\mathbf{m}_t \over 1-\beta_1^t} \nonumber +$$ + +$$ +\boldsymbol{\mathbf{s}}_t ={\mathbf{s}_t \over1-\beta_2^t} \nonumber +$$ + +$$ +\boldsymbol{\theta}_{t+1}=\boldsymbol{\theta}_t - \eta_t { \boldsymbol{\mathbf{m}}_t \over \sqrt{\boldsymbol{\mathbf{s}}_t} +\epsilon}, \nonumber +$$ + + +
    + +$$ +\begin{equation} +\label{_auto6} \tag{7} +\end{equation} +$$ + +where $\beta_1$ and $\beta_2$ set the memory lifetime of the first and +second moment and are typically taken to be $0.9$ and $0.99$ +respectively, and $\eta$ and $\epsilon$ are identical to RMSprop. + +Like in RMSprop, the effective step size of a parameter depends on the +magnitude of its gradient squared. To understand this better, let us +rewrite this expression in terms of the variance +$\boldsymbol{\sigma}_t^2 = \boldsymbol{\mathbf{s}}_t - +(\boldsymbol{\mathbf{m}}_t)^2$. Consider a single parameter $\theta_t$. The +update rule for this parameter is given by + +$$ +\Delta \theta_{t+1}= -\eta_t { \boldsymbol{m}_t \over \sqrt{\sigma_t^2 + m_t^2 }+\epsilon}. +$$ + +## Practical tips + +* **Randomize the data when making mini-batches**. It is always important to randomly shuffle the data when forming mini-batches. Otherwise, the gradient descent method can fit spurious correlations resulting from the order in which data is presented. + +* **Transform your inputs**. Learning becomes difficult when our landscape has a mixture of steep and flat directions. One simple trick for minimizing these situations is to standardize the data by subtracting the mean and normalizing the variance of input variables. Whenever possible, also decorrelate the inputs. To understand why this is helpful, consider the case of linear regression. It is easy to show that for the squared error cost function, the Hessian of the cost function is just the correlation matrix between the inputs. Thus, by standardizing the inputs, we are ensuring that the landscape looks homogeneous in all directions in parameter space. Since most deep networks can be viewed as linear transformations followed by a non-linearity at each layer, we expect this intuition to hold beyond the linear case. + +* **Monitor the out-of-sample performance.** Always monitor the performance of your model on a validation set (a small portion of the training data that is held out of the training process to serve as a proxy for the test set. If the validation error starts increasing, then the model is beginning to overfit. Terminate the learning process. This *early stopping* significantly improves performance in many settings. + +* **Adaptive optimization methods don't always have good generalization.** Recent studies have shown that adaptive methods such as ADAM, RMSPorp, and AdaGrad tend to have poor generalization compared to SGD or SGD with momentum, particularly in the high-dimensional limit (i.e. the number of parameters exceeds the number of data points). Although it is not clear at this stage why these methods perform so well in training deep neural networks, simpler procedures like properly-tuned SGD may work as well or better in these applications. + +## Automatic differentiation + +[Automatic differentiation (AD)](https://en.wikipedia.org/wiki/Automatic_differentiation), +also called algorithmic +differentiation or computational differentiation,is a set of +techniques to numerically evaluate the derivative of a function +specified by a computer program. AD exploits the fact that every +computer program, no matter how complicated, executes a sequence of +elementary arithmetic operations (addition, subtraction, +multiplication, division, etc.) and elementary functions (exp, log, +sin, cos, etc.). By applying the chain rule repeatedly to these +operations, derivatives of arbitrary order can be computed +automatically, accurately to working precision, and using at most a +small constant factor more arithmetic operations than the original +program. + +Automatic differentiation is neither: + +* Symbolic differentiation, nor + +* Numerical differentiation (the method of finite differences). + +Symbolic differentiation can lead to inefficient code and faces the +difficulty of converting a computer program into a single expression, +while numerical differentiation can introduce round-off errors in the +discretization process and cancellation + + + +Python has tools for so-called **automatic differentiation**. +Consider the following example + +$$ +f(x) = \sin\left(2\pi x + x^2\right) +$$ + +which has the following derivative + +$$ +f'(x) = \cos\left(2\pi x + x^2\right)\left(2\pi + 2x\right) +$$ + +Using **autograd** we have + +import autograd.numpy as np + +# To do elementwise differentiation: +from autograd import elementwise_grad as egrad + +# To plot: +import matplotlib.pyplot as plt + + +def f(x): + return np.sin(2*np.pi*x + x**2) + +def f_grad_analytic(x): + return np.cos(2*np.pi*x + x**2)*(2*np.pi + 2*x) + +# Do the comparison: +x = np.linspace(0,1,1000) + +f_grad = egrad(f) + +computed = f_grad(x) +analytic = f_grad_analytic(x) + +plt.title('Derivative computed from Autograd compared with the analytical derivative') +plt.plot(x,computed,label='autograd') +plt.plot(x,analytic,label='analytic') + +plt.xlabel('x') +plt.ylabel('y') +plt.legend() + +plt.show() + +print("The max absolute difference is: %g"%(np.max(np.abs(computed - analytic)))) + +Here we +experiment with what kind of functions Autograd is capable +of finding the gradient of. The following Python functions are just +meant to illustrate what Autograd can do, but please feel free to +experiment with other, possibly more complicated, functions as well. + +import autograd.numpy as np +from autograd import grad + +def f1(x): + return x**3 + 1 + +f1_grad = grad(f1) + +# Remember to send in float as argument to the computed gradient from Autograd! +a = 1.0 + +# See the evaluated gradient at a using autograd: +print("The gradient of f1 evaluated at a = %g using autograd is: %g"%(a,f1_grad(a))) + +# Compare with the analytical derivative, that is f1'(x) = 3*x**2 +grad_analytical = 3*a**2 +print("The gradient of f1 evaluated at a = %g by finding the analytic expression is: %g"%(a,grad_analytical)) + +To differentiate with respect to two (or more) arguments of a Python +function, Autograd need to know at which variable the function if +being differentiated with respect to. + +import autograd.numpy as np +from autograd import grad +def f2(x1,x2): + return 3*x1**3 + x2*(x1 - 5) + 1 + +# By sending the argument 0, Autograd will compute the derivative w.r.t the first variable, in this case x1 +f2_grad_x1 = grad(f2,0) + +# ... and differentiate w.r.t x2 by sending 1 as an additional arugment to grad +f2_grad_x2 = grad(f2,1) + +x1 = 1.0 +x2 = 3.0 + +print("Evaluating at x1 = %g, x2 = %g"%(x1,x2)) +print("-"*30) + +# Compare with the analytical derivatives: + +# Derivative of f2 w.r.t x1 is: 9*x1**2 + x2: +f2_grad_x1_analytical = 9*x1**2 + x2 + +# Derivative of f2 w.r.t x2 is: x1 - 5: +f2_grad_x2_analytical = x1 - 5 + +# See the evaluated derivations: +print("The derivative of f2 w.r.t x1: %g"%( f2_grad_x1(x1,x2) )) +print("The analytical derivative of f2 w.r.t x1: %g"%( f2_grad_x1(x1,x2) )) + +print() + +print("The derivative of f2 w.r.t x2: %g"%( f2_grad_x2(x1,x2) )) +print("The analytical derivative of f2 w.r.t x2: %g"%( f2_grad_x2(x1,x2) )) + +Note that the grad function will not produce the true gradient of the function. The true gradient of a function with two or more variables will produce a vector, where each element is the function differentiated w.r.t a variable. + +import autograd.numpy as np +from autograd import grad +def f3(x): # Assumes x is an array of length 5 or higher + return 2*x[0] + 3*x[1] + 5*x[2] + 7*x[3] + 11*x[4]**2 + +f3_grad = grad(f3) + +x = np.linspace(0,4,5) + +# Print the computed gradient: +print("The computed gradient of f3 is: ", f3_grad(x)) + +# The analytical gradient is: (2, 3, 5, 7, 22*x[4]) +f3_grad_analytical = np.array([2, 3, 5, 7, 22*x[4]]) + +# Print the analytical gradient: +print("The analytical gradient of f3 is: ", f3_grad_analytical) + +Note that in this case, when sending an array as input argument, the +output from Autograd is another array. This is the true gradient of +the function, as opposed to the function in the previous example. By +using arrays to represent the variables, the output from Autograd +might be easier to work with, as the output is closer to what one +could expect form a gradient-evaluting function. + +import autograd.numpy as np +from autograd import grad +def f4(x): + return np.sqrt(1+x**2) + np.exp(x) + np.sin(2*np.pi*x) + +f4_grad = grad(f4) + +x = 2.7 + +# Print the computed derivative: +print("The computed derivative of f4 at x = %g is: %g"%(x,f4_grad(x))) + +# The analytical derivative is: x/sqrt(1 + x**2) + exp(x) + cos(2*pi*x)*2*pi +f4_grad_analytical = x/np.sqrt(1 + x**2) + np.exp(x) + np.cos(2*np.pi*x)*2*np.pi + +# Print the analytical gradient: +print("The analytical gradient of f4 at x = %g is: %g"%(x,f4_grad_analytical)) + +import autograd.numpy as np +from autograd import grad +def f5(x): + if x >= 0: + return x**2 + else: + return -3*x + 1 + +f5_grad = grad(f5) + +x = 2.7 + +# Print the computed derivative: +print("The computed derivative of f5 at x = %g is: %g"%(x,f5_grad(x))) + +2 +6 + +< +< +< +! +! +C +O +D +E +_ +B +L +O +C +K + + +p +y +c +o +d + +import autograd.numpy as np +from autograd import grad +# Both of the functions are implementation of the sum: sum(x**i) for i = 0, ..., 9 +# The analytical derivative is: sum(i*x**(i-1)) +f6_grad_analytical = 0 +for i in range(10): + f6_grad_analytical += i*x**(i-1) + +print("The analytical derivative of f6 at x = %g is: %g"%(x,f6_grad_analytical)) + +import autograd.numpy as np +from autograd import grad + +def f7(n): # Assume that n is an integer + if n == 1 or n == 0: + return 1 + else: + return n*f7(n-1) + +f7_grad = grad(f7) + +n = 2.0 + +print("The computed derivative of f7 at n = %d is: %g"%(n,f7_grad(n))) + +# The function f7 is an implementation of the factorial of n. +# By using the product rule, one can find that the derivative is: + +f7_grad_analytical = 0 +for i in range(int(n)-1): + tmp = 1 + for k in range(int(n)-1): + if k != i: + tmp *= (n - k) + f7_grad_analytical += tmp + +print("The analytical derivative of f7 at n = %d is: %g"%(n,f7_grad_analytical)) + +Note that if n is equal to zero or one, Autograd will give an error message. This message appears when the output is independent on input. + + +Autograd supports many features. However, there are some functions that is not supported (yet) by Autograd. + +Assigning a value to the variable being differentiated with respect to + +import autograd.numpy as np +from autograd import grad +def f8(x): # Assume x is an array + x[2] = 3 + return x*2 + +f8_grad = grad(f8) + +x = 8.4 + +print("The derivative of f8 is:",f8_grad(x)) + +Here, Autograd tells us that an 'ArrayBox' does not support item assignment. The item assignment is done when the program tries to assign x[2] to the value 3. However, Autograd has implemented the computation of the derivative such that this assignment is not possible. + +import autograd.numpy as np +from autograd import grad +def f9(a): # Assume a is an array with 2 elements + b = np.array([1.0,2.0]) + return a.dot(b) + +f9_grad = grad(f9) + +x = np.array([1.0,0.0]) + +print("The derivative of f9 is:",f9_grad(x)) + +Here we are told that the 'dot' function does not belong to Autograd's +version of a Numpy array. To overcome this, an alternative syntax +which also computed the dot product can be used: + +import autograd.numpy as np +from autograd import grad +def f9_alternative(x): # Assume a is an array with 2 elements + b = np.array([1.0,2.0]) + return np.dot(x,b) # The same as x_1*b_1 + x_2*b_2 + +f9_alternative_grad = grad(f9_alternative) + +x = np.array([3.0,0.0]) + +print("The gradient of f9 is:",f9_alternative_grad(x)) + +# The analytical gradient of the dot product of vectors x and b with two elements (x_1,x_2) and (b_1, b_2) respectively +# w.r.t x is (b_1, b_2). + +The documentation recommends to avoid inplace operations such as + +a += b +a -= b +a*= b +a /=b + +More examples will be added, in particular how to compare autograd with own codes for the gradients. \ No newline at end of file diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter4_114_1.png b/doc/LectureNotes/_build/jupyter_execute/chapter4_114_1.png new file mode 100644 index 0000000000000000000000000000000000000000..6c642a6a1e422f9fbe644baf63a261788f73b91b GIT binary patch literal 70740 zcmdRW1ydYtur(GW1PJaDf^2|5Ah^2(*WkXmyGw8uch>}WcY?bGcXt+dz4P9$>i&nX zs4a?RW_IZ3>F(3#oFPm>P6F*c(R&ye7&IwKQ6(4{*dM@42pJK02lI5VANa=QB&P19 zY-{4=YTy8dku`9#v$A!vGB>1f0XsOF+uD3*1ilz3%$%I;9J!g8tpD!-Mq39{rorUK z4B#Rtc9I&7Ffgdt|Gi*g(lQ8OU}0dSM1@q`GEQFHGpOA)=OLX&Tu&X9&1Z=kiK?@F zsPLhAexQ3tAs-uo-7W9j*R~wwqEKW;5NIA1ITZPQlMX(~UN^3#V;9(EOZG+HuxbaZ-@b=9W(a#R$LX-~N&>b0f8Kd~RnDm#6kIo~o@U!h}h z$>R;9a}4}+Tcd>!Y6aeNh7!)b8?JCJx76_rcJ{Ba{O{LU20A$>|9g(9`u`nPzz>a< zkZ5i^=KSwk&d}zsSQl-7p3XcH{@)cp9K(VC_ha~aS_Bt?Bg!o&su&eqfGZy_eJ?}( z|KqS-{3^C|G*(&9LX3PMB!|%AZHw01H8zc4@%*T$c&s=oSe7`d#-#sZS6_~rpr=nW zLLv3&8=g05#tMTbJ3Uz^j!H#MCB^=_{%*X&I~Ry^r=N`C(p$1!Gs$30V1coUOM(7CJf4aLi&;?UJRceVbK^bgD!95a6^~i!O`$ zEtgiAzZ&Ip3OIzNs(;iOV3Bh@=C@Cl@S~jlFZpa-TEp`sRp{6<x>W_NO{Kc7%eB+FEE>P627QqKHb}?9uBn8!X z>0>|Ib!pST68R?oW}kE})3sHoo$#y3LQwYaol3Wz-Oteybofj*L+!eDw$7bO_v>?aD*u;$*OhYhTeN6!<*J>NW@XW zbC#>2)~C$@foi7*f_cVl3ICfxx=Gu`SffC*TmWBE*Z9UG)1%#Hhv91yt$(cnPOwFN z3v7l@XGnfN_H}6IvAqj)6|9rvg^|i|{8n7@YXoeoT5VZ$q;>aw-hvw$|GmkEQ!|M= z1K!aM_pAjptbCNrl&2GgrDg@{f3P%S`roN3gOPR1_q7s>E_1yo$2dz8Rxn*S42v!&jqN+dSVZVI%=NLfs_ozRr+RM&s`0cf^ zBU-|7#J1Q>Jbfi=1ec3z$j-@0yjop63EZa3F=ZAb_!lx=yw3yH5Leu(7+9=tTm`lh zi*VM^|^hWN64wuD)*D)e|w1b$p}i6?bv1V<#TC;%4YUBm!oAQ{=ofP^t-g=*Ysim zzw>($58Ycz|L6)9rTMiCAb)i9No}YQ1rH34A)DgA7W2zyd7f^XGQ4(v`lP2m;C7tx zybE5_v#dlA22CLNZoF95OJm>y z&n$W01=riCh+R?E-tmrCTeJm%BR`YoMEd7*rZ>DrZU4^a{Z|5SzOd5F86N$}{`kMe z1b+`Nn(Uf03KZJ&`xb$GPb~-5P_zKV5>EMPM86v3)-HE+gZSWe2p7|4k#c`>V}d`Q zLU1@>Q@eHN88H#y!}OpQg7)`fsJXh$*R7l5b2>L>%~DuRN2x-le2yZ1$k8Y3 zk8Y(=qZUfYOJI*rn=ZafE3OzO-?#}quWiWDjZM+(I}@GgOKN5f#KvWy0@|7jmxvQ@ ziQQ52DC^&kjme(AVt_$!johAg+_p7MCE5Lvc=i^X4F-FMyg?_t z(D*B#`bZ|VtR9`G*C@WsPi?iO{mxj1=BboSnjPSjRSZ0=)dOjZrQWI!zhSo!1Ab^; z`dDeno{lROr>-;an85QSp;F>39^!Fhgn1x%VJ69`n+pFa`eAY2^d9myaCg4wS!;WE zp&NwZq1rk~Xce&c;2zb}qs)@ZJ7$7iI4wtoMS*jrk6pJ<@WwS>!$3HE5EK8x<^H%JrHjM#F#lkf4`&X5&ZGw#>27&W; zrpb0(P)<@0W~pbo>`H9I0ZO6xh$Y{bpy&&<^-BzQ)2Iaaq4dvf{XdKyyUboYV7VzuVi9S5-h zle)GB>oUSiQ>vQYd zp>TF7_UBDkn@llZy{yjfY6optG8xvib+ra9WAa;|K+COzCh)1^)4P*PE0U|V^vR{- z%$s6(W?8J2jUsCENP&k9HEXaxl1B^}Xbla3J>7~la6(G5jVEkd?I^LM6s%<-0^?H| zlKo^$zRTo&b?4w~vKf>)0O^jyhyP*Xp{AvDdd=E1HAU=yMJdHKX~%87VD)2u+baxc zkhbylvjgD8lg0Ky_Fl-Sg=#E#sFxPU+Pz(b9iLn#mS*Rw#rBXLorYTc5U{H!`7;6@Fs#loUkW1SYxaT_n{*tl z^C;>f(H+R+KHy8!qQOKpYIFsOB1ur1xr`4$PuvQp?EKy?1$+|)?tXScqJ1wN&r*?I z^S&e{FHTeKmZ_oYRxVS0nP04M{{l2sXe|IHf(ksN)vVaxwXTLQTNkP`=p8I}K~US5 zRiP%zpR<{CTGoai(FY3mWg@QOs~UZ+p34`~uFy8BoO#8>6gheMjXWF>O8dIIriw2b zKl{fps`a&#;lltYJdgUmfYx8z1$uASHXXCq)9dk**W6r8XTUT+hJhAcv)Cr3#7Kk& zl%A|lU*>1GyjlVLY8zYc7@K3zsFm$0+T$!+W}$~TXsxBWnx|dN(pWcM>PjQ?nIbA` zA_6=`?A%H+W8q9T%E-*d9N3yTKjYcUBk1Xp#Fo_Oc>A#8^>`J1kLn9;eTaVB_9eu! zu&tD0>Miw-eb_OXF6(pm@&X#XkSgx8&sm}$wuO4R58v6Kb}C9X z=Cuk%8j0&&^M^_MMWmSiL#2H7e&t2uwOjA)$!l+I6kEI3HM%{N;TM18zqWi6<$Udm zxG_9Ve*W$I`wWNX@RE}I7hNC~`LmHidTf46;fO_Rw~*~a{sVu4Kq{h#poR3v)oXSJ zQR%2=Kb_B)~1^ zI)9NendB9i$TcS-e@K+CYIgj4y@Txgdfbubw{AI@r@>G%g{E9QQ#79)t$C9Na|*5z z&zCAMfCtmWb5QMCvD1@9j}*1IQtxI>w22I&G(e@4=(K!nreW0UT<vn7bz5e*7KXG=P>GA^}LQVuqIRsCY*?(^Hf(xyD=s>^n%QG6%OMlJ8 zF-y>t#DFPKp=%qOA=3r}YjHN%JW9RiyqM7h?%Jw~5l0X1>=P5u>u(r(Z^>`yufm%CC-t zk0#ODE3Ioj@Mi}YR!5W^srKWuzc+J`*}LcKSXL~(6Oh#E1X`#;)}K8wCexc5QVpoZ&I}OZrDef~@yd9+aIyUjj`_@_c^x13d=L^5`@=km{IV1L#_d_q6t_t$oGvCRL z$7)O>JaWhjYkC@NMx=Pa53vc`Ig6V@m_n=Pw|0b#)yb#uxv5AM1=S#+j`lro06BQo zGNCo=lLkE@RRkr0a7Gx)Z8y!NiuugtgXaPCXPMrbFbs|-5Esc2XwJy=rZzM%T0E?} zSPI|@|HIN)D9(l3Ca0JxX;pmPqSNmn^ATNq`aW6UBKscO_i*{%_3aR!b0SpvGtaGY z)8brw0ytFxWEa*lF$)z7%04r`dH&*=uZ~mC>FiA3xcq=9)of~Rf0W&xvG%MA^!Z>8DHN{GoCTJklEGe8UQ6kaX-taNd!9J|KPzo!b7I=9hlot_@8 zr`lyL7oytooQDv&ikuay6H`zR@uE!~*6|sQa6Gl!eyGo8l|7D#c1x;W4gwg@q9gA79~yGtIVv8iz}2svb2NTB`7wOoj3cW_F!RCx;*lo(b3p zX|l=LDCH#po$c`aU`w%?5(d(~4{!m{c6k2h5%T@66Ur&cn98RmfnNM}k7L4jndmx* z*Xs)ws@0n1ZtXTWV(e4qSyn`Dv1`kaYmlA_(s=FogctNW*}{{ds)P<`+0)!D&U;uee*x&zm$&7#1WL9P*SoyA446M|hy|lGF53se#rDs_|GH&9;COw$48Fkc zto&T4X3ILX_dWvSj=8g~E0@CU6B%S+7M3qv6>RPdIUK^oOX85}CZaS8pU7B7E~b9p z_rB1S&_+J8+SQJs2-CdaVarfeXEa(9K{I9mx^>6@7P|iQ{F(oJ7^hNheIwedFFB3w z#E$t}eRHa{MKXhQznNCyv;hv=T6lL1`bVapxUbv;Gf4^9dfa#rU6RWNhc$ly=t@{s zwj*gWJ%X1_{($!S&%(&dlep&svssHs@+a4~NZ%_*fr;$< z&wAtd?Pqvro!_bL5VLLe!aV>Qs(NurH(m}29;J0vyk66Lx^o^<>e1QswC5fzMb$Xs z5T*bUtX5~=*})MjP%98Al zE{oIt-CnKHB+wdb*CFU)lm9ka?n~mZ$(h`sRQPuaKML~!VftS1KlSLysD*QU_IoCz9>pXf+lGF%rROLfGEz$AK1Ec zVSFE>Up{r*mUc$uREqd?hSzH$F27@gi0#`f{YTo(Q|DV(nlRDrcrChRPNHnmCzfEw z9-5=S{Mo)b{M+2p(r^2^XUoFo49U&Cpg-5IG^uRACGtJTcoo{)+l!b8@1tCQ$XpMj zl6t;mdSiLp(tAer*`oE?DtN>2UHH$3@MMIxQnXJ^=v%V zLJ;7{fC^(D2{e8?Q1+hr95HtjDOt7>-A_r{e3^(3DZ=|IlqPMGyf(C^zY6I+GV_qM z8IHBPw&TC^Kcqf+-ORs{KRZCmi@b>zk;}^YydG^bbeP~;F=V7#76n~1^Mn|6&DE{F z-fty&1jl{(21Xi5RLXK!CvEU)^&Xi}Jd~G!N*QMUe-qFDxPAQh38RA}>AV?8Slgr_ zB`$T$Y-6keZ92Jny31S1KG4R8F7hW0@sP*+w901|-OMcg946;K>$g&8P${y!to&MC z^_7oT?c6cESH~4*Of`Ql$H-+(_2kzzYKt+E1Mib;AJnL%_I7q6JnwY#r`#;-wMu8{ z$Y0Mx{)<$uT} zzP-JDLi-K%Up@{hJ9~Ox3am1NZscUSvSZf7m8I`GFkErauh46@dgxYin6m?9wnqwc zW8Ivl2<1-@%R=_f#(9F>-+``NbP}@U`0Z)AuHUP1O%2V`e|h?6?mvFNqNf4p*bS}^ zxmu)%i|#a)R zOSNi|+tI6yEVmUzveSkeOuL0?gb=j6%PM(RLq(XDzU&$D9If~0{0jAb&R(!q^LfMX z96nGWKWpV1()V5ZxZZ@G6559bBI4q12P;$-J$aGkm%RSg{i_OvzHeufHF7pepFf?b zX{hcS;D7zX(#$)pZ1lV+a3it4!`yy@ldDwREAS}s7Ha$4Z)?I+0X;v3h@2I(Hb2)7HlfRJ0_GFiSPtRWO6%{;+<&iU zkfrIhis0%8_Ge4?^=)(jgq6uihURj2`dKEbRY4 zw6JjgH|WWO7F8zu<;3)1P=ddEY6?@}bz5MZ!)Kf0l63uol;P{w>h|(z3r0MMP7Y2v zT3kuVhQng9ibZfvxATcRuR!JejS8H*rS;kg@eSn<j^FUA1i2LefhqgdQa)E9%TVK~3k|8i+6Q zZG+8^1rIIWS~g#%Ki{gp#akGboX0cd_%)JFC(0prTEWpAPX@f&+a6$i)NWYRsP(X*g z^_X=~ggGEwR(2M{gp-m*!iI)Yi)vs86kETVuc5Fl3TOGv;KzUVB1bSSA#DOgJfmEL)t&tnwDe8MMTLcarX{Gu2 zz=o!f8!Z3uvw4+MCAcn}YlR2WjNGyqo!PUUOcI;`WA z9aDA|ejzF}G}?OFHquA{_rI{j^$i7X{$2E>#j%?OZ09@-5T)+n7Bcb7j4Vg55>%u@ zY(%*)Km_tM?gWdmn^O z(a>(?mpNCTM76LuoeY*@;crw|L45khYd#Ms5=P07+b=*#GBcRjSE5%9j9h%)P3ol?-e@5xTzK`dhAqG|@!uPA+Twdm z_8qlN?E711q9|0T+&_D-K4lV+8r#_8`c38X<7t)u33xGAn`?{#hsAdB1C)OZIMs=z zf31sIvsQ|V#Msj{`$zo`Djeg=8#w9S&;2#8H9X7}(&hNf z2ptP-&(ooTxR9Q=9?4QmaZ~7lb(;#zfke4_%zuY9^>H(2qKFC}Bfar^fEf~fzAssl-+6?g_p4D8FuM`58q z%52``3Tj}POKl;kY09to1v1$JU!HxvN|Sr@04pmK9s# zT2vo4l`02J3By#(D~>hf)CQuv~>KOJ60P% zXoC7PN4xDocof>|1BucscRs0$6>@N!38gG%>d5&>{{Z-V~i~sEgxOKeBE# zYGDtfLNKdx1#5Qa!WFiNXPw1&=JG^(- zEB$k|93}>`7sx9F`Bb{`ZsX%PT=#U^%s%aYXkY&>xHuP?0DeKTbrq@mbD9mFuQW#G zI9CSnq?x5>+3w{{Ac2Xd0qp_Zo@KfxuzW}EG_X#?!co)nTc2%g$SGirM7;agZ!x$x zOH%NiQE;!6nq~fI{>S834p?6@V#v$M9@ooyedM#D1?DTz&ytli687Nqpa7rzCJO1e zunwyCa1gR8W;eJyp$#e8?81fTKWn-29zg! zhknWf(;&DBu^Eac-C3=;-^uI}4_Pq7a$VwgM?kLvoDXSV73MEMK?-N@-|xu79V_2G zCsgTPt5<4OXQv*+{OQR>^#K-eYkZ8EHTS#Zbxhi?IO)l=?XGQbf5_9JG_4Fq7(qWo zH*VUzP@V30SXiX}!)*Z!DNLV&8dWlccjDH$NX;_ghpNK^oU2XLbh;eDF%`CAk0*zF zW}1mbG|v8gO7C|f=gS~C400w#hJ<2O%Lw#aER=UZ9oHJ5#{&ReP0>bHgzI0{)KDS= zH!2bqzy5uDM~@2fkRAgLzga`Y$hJHNDVmA)Obi4BCe7~-!Df5x|6XVdG<56QXkqTg zPhW$MPfnfZ7IDM5q2Tnf@M(kn+cA3&Wt3n0qSV{(8Z@m z&v*m`zAR0)B3lsO!=YrO+-kNWyN$4k*5IZJ>fXs(#g)2K|LlFs1m*s1?oWe%g zW1CG+>#c`2KI2n6SwPUnoqNPYkC4EAs?{-f@Zc`QjnAf_u%CeRw6K#y3UshNbwYVZ zK5kU2X}`ckM*HvGuDU6JX41u>7rbWFvh7wss(y?V$j{gq@=~$g@iF@XnqZ-SdNer@raIZ3z;N%QJug=z z@hA>Ox?oSxhcaX-qmB_{RBqz1fcwUxGsnR57l|+u%RM332NE? zL2t839#UR4-~A^y@!iS_3bfMy2x>z`MlVS*&3O5b6Ud0q;{J4*5j6`+{_@m@h+;!$ zR|G1Q>0=f?tBYXMHgnctE)APUdltn6F}jX# zS}J(xqyQvtw_tVOX2Z(bAN-`CiVJw(fC5e>sHk$flJSisfTx_ya@8{v=6b?bHtL@h zJHV3x77D}WZg(2#W~EsOPn0HPU*=DSA&Ky+QG{! zIB)Rlt^>VIW9iL@qaRXgq`5)kE15`)R07nzp9gsF zL7R4eVTOTBdz`DNUUxCrx@Lfs(S5l6bMB@cyhD`okWcw_SQHPxNA}Y*+M~B0dO|J` z$0$iEGZ-)noF$wnx6u-!pLWZhNo@Y(4=^c$nLYT)x}33l~~= z)U*Ld5&vV&X(aavdz7XMd0~%@-|zAFob0#LucUm8$rtVXmCNer@18R<4)?KeTikn3 zv;AM|rebHzQU<_9CBokhdQ57!E;uW{F>d$m(K4kaD;KNW`;*_p8deh=A0K;7b9iIJ z!{aNbi+UD2CQx!HB~hQa&;w+*{{Y2;PPT)~id&hMS0Oa1HsRS_m3Qc%*=#8eFj{r)A}^)}sXDg+>T z)HRk(EKReCjYL*jm<_Q!J~|drriABK9i)cqa2j@un|E?@fGmu|Rqd+}sB|Zq)PILb z1ntVz^Clj6W&P+6oZuOevd@&}m#5)=V#W959{7ms&}vvX-g~es(hOK_;;8I>vUbnU z6Rd78?Fnsz(%e0DqmTy$o@xc8DPuP(pA$Q~IlOu}&UMyrbbc<)OQR-`u1bEOK5#kR zrgQO*R{B2Y%Y(8Lp0{zs4OlqElgCuMA4{{=^PABXZf~eQP;>~i22R^>grje{h6@4> z{6AL_;3rvW=CYt-Nb)aIg8qT7MHU3q#;r>{& z;Cp<3)sqh`C)SJwfxYX-SejU`<&FA>~v1wh`dx#o{28D1TOT~lAe)Kl(OJG##GG1yBJ?Ivy82uRSi3a`YIB88{p18tcLY4I{e z&s4S9acu*~3F7Z8ESzy3hk^!0t_O)^?H{(BdslqQUF?M z7p;R^rCp#kOaX4nBDG`n@}$NDqd(?EYNfNHM?Q0pKMVQWm}Pu;X2zux!;9PHKhzt= zQani%TG^0)e19(klfVJ1zkdbP`;%@uvPMW)tdEfsd`hBA+pl0S9hrdvIAET992@0H zs*f@w%M?(k3i4v^6~cK%b=)Q1vK*SbFPHEF+P`?|V%(YH`?-whBjxlO#v=bPY+@L{ zv$51YB_+W|K~h)Olu*S*d{Vyty|sfXW~^b7t9t*VFW>6)sPC&{;=pH>9qTzcRqZ5j z!xH==t7AyyLeOWI+0bH`>RCmE_3hg68crk0(Wh`yg)N^bImQATXCLgy7d4w{gGPf! zjb=y3UsjA0k8_NDhJ<2@1caw`_59}Zxrkhua{Sl1Oq&1&kD+b;vJcGD#&K~Y`&WU& zxrJ7_%hKD5E)KDKEBdWPpMv{;Rv)`1gi|&KlCDK5JRKUmI7dC1A9IngmuuW|e@MGsCZV^&9QVVBGmH^N{4ybRLxwRbe54M%yT6&PgM`oaNnb7Ywe9 zD4CAQl125E=Iv`aYSTRaNopJb$hf}y)5OOnF3)Z9`@PJH{HGU++mOOQ_7OJ*t4SZq z&73K3Dm>d2VS6!#znmxRJx2A5W&Jg?7R)g77WGxMb%YIC=zWNAKnwWb;Mkhce{%uvja;ZM{1OenQbx-$D}8&OHGOkICN8 z7%K1Xys4)bQBc|lN|p(|YlE+waJ#OlaaBl176QQH%a<`}kftthm|qPAG?s zJ9r9A*~MI7@&#sPk$|m!lgnDxv~)-nEA0VO3XdmJZy;s-Lad|F$>GD0Yns*TP8;B3 z8`XP;_{OfT>i9el_-6RNoxSm`-+8aQ{_$=H?557*th{>mdxylKo0a^9JKV|J9#jsc zvw)m6kwN%sBBN-$ViO|N-N}*fc#j?>FV)*46NN^U*OoW0j)gobiB+!|a1qN7SOK~C zTREMSsa{Ca$8d{ON>&I+_0Kk%{-mauXV#m@m?PBU?>j9$JoD+)-9iNHh@Gt=FFn|C z7voJR?NtuUIblI0Qb}$L-18VKnT8$?34iKkg@v-GrsBgTBC8RLMMLfmT)4O$-KdJ= zJrSm>A)8iDLGcs#oO}~3>uFY8Z2U6KO4K;#n|2ET?Jbf7)|TS|7{Qw6U(IfxRYqh- z@(KzRBCt>{4ZFli<_$D&Jq~1P?YKw-;e6W2@MkU1q6hl+ZW%)c3V+8jpYmnEZ=9B+ z1lWJVa}Z5sW8U{PvPvmjHx@zI(_UX6teHPfF9uWX21E{41&i^$6N5=@m>QOI9D^eN? zE1T?RKw#^v>q64Yq;%{QEY}@?IuMZZ!I%nP0JB81KyB!>fhiCc`D}g!ik&-qhc#G& z>d8;TA&Dcn=+W}-H@9dis_cF&nW2OMW7=}9#c*FwM}EJkgIIVve*vzJt0qF?Q&V&2 zo)siAbOOQD2nluVIaq)VA($8b5LB)Pxmb-u;Atg2$4hjJl6R&M;2No_bsaRZTpL&L zO`A=E^P1f6+IHEpZ>=h#`)}d^y_XAFrHJLpbO9_KhwIkWcAgf%3Z6f;wVjlL4@`o5!zb>{2KvTNx)_3t=(9|pWAr$Xmbmi@U+ejP7`{(Cy0EBNCGAn!1*-_5CiW!?($FLu1<{LEi$Jes07Js+>1?`(Yy&a}#w>qq;-QT^A^{nNhT1DlAZ zYydO~C%xjOm}$8TFZhuuNzHQE69u$BM*?nD&)=uMiOm2WIru?Qtar|_f6)$tbANKn zLM%gpMn-dd~SD0k;m>Wu#P%#j!xmgC6_Oz3MW5mkbH7cH45 zsNFS-+F8(55p;BUnZRvbAv_ty(O-lQzZef2T3tm4&)Axh^0;MLNmy^VArIZPGEwtn zqp?FfKray|lc!36lyDpxtAKVmBfn|F!(RjF8sP0UmRCNJyx(x*Bsh^L4rpDXaoG$t zBH|V1B>vzeyy@;kGJd2~Y)NMc6e5uO`~eA$Hx$shHvnIi`uf`T4sTC6aSsi8 z4yOkt6^?|Gz0esm9y3S+T>{AtkWGLMsi8LCc;TU4{>dT$P{Xkk2WS3)q8f1bP`P?j zIp9*4naqsCznh5e=83A5#Xxk}_T-52CF8&0+eZA7rL!T?i>_M9mLod8zNM0RYBST| z%?kV*fD7U9ST3-c7?$t8DOsC$4qUVG9P$6hn%S4eh+0QE?Kw*H>$KGl2m`2G@(<-0U(3tOYD#wEq z=S;?jr?!=sufiwIHGou07J|m{J2wvhU^8PIxLe-&)kg4lE*mNy+VWME{5AN1625t; z5+Xib&<2n?WaD3J>pcf&dXakq_>$@9!V*nl>$32)?hG*lc62H^?pNA1$e7ChnXy-^ zGqf1Ga@*t)gnfX;lm(v9isS-nAp%7h4+?Jgf;E}$I%l-%LkvSUTvsa~(&)8+#raSA)Kr_0V)v4)2-@1m^ZMf~docE7SRpgzCQuY}{rczb!y@dr1qnZ^m=P6ck z9c{$ZN&(GpNyz}*)a0=ap_B|WSbN<>IbQZ$$(W>1%`*jEzI1QG0M68_5n`26)p%OU zONIb>Ka(fe%uGE4)3=yqj)AIWYvaHvY|823uXU+m05hnb610nEvX_MN$oHS zbuU%H8ef)SCv>xhK#j%_6VMBy@qYsAQf6(FU-X}|HHCms-@-6u)1U0~0LX#0PZ+(e zI%b+*_Cg&4fxRnsV<>#X^0a}g({6F5>od9o51z?Bazde8V~lvY?tgN?nycx1F-3bW zl9My?i-)K0Kydc28G2kSy@8%GitUl6g9LJBaB~+qYpgpe2fepxd=U<-Ifj{5;Qsq| zlP%D*sS##-pU)w?IT}x+R!@q)%+?BDaV27o=yu4OkB>hmU=BTUZ_T{YxP9#T-$5O4 z&eLLDPN#>bzGtE|yZ*CPW^;OO<`Xn@0jo@$1LDEMK;p7_kyXUe_53z5D9E&^y%w_r z_q~<%;I4YLM5a(#k-hN_z!!p`jlQ$1oYOA=)19x0s$lChu1r!-gogAX#=bcNpkkvL zqyP;|i7{ErFR!}(GkR}Jlcl6ru~hd76!M-6KsMqc+f&=_FafyAof2Mi7X*vhpKv(F0Sm^r|*<<6C3=l{BuB{n} z9~;>y2gDKWcrsbj(BSx)W;G*saWg3s<@zJ(?(}`r5}kE^RM2MX|LGISBq8sJn=7S~ zn&0}+S!P7B@%0u2*x`xAS?i!JTmu*~f9~towKhZEj`s(5n8<{lj8&f=Kb<9FAZa8; z>Ry|=E@~%VPHX%u%Mw6|!-uuYpW0cIK>{sfyi5&o0Z%gOwQQq=erg_=zNmvZ$NHr8gThTi=YQfN zcrQOp6%+tG2Vh60IJmofZ`Xg7}1niv- zNW?Fx?sz}vw7To@{)G{EA5sA(oF+%dj7@)C@!&_v{`}uc%-bAtv^<>yU@VJ6m9}Yv z&n}zD*>?s1xyymFYZHOS%*STpv=R~hAtqG(K)Mfb3>X=hM*>s9i-coYnQa=uyS%`( zlq(#F35P8&j{wS9`Q+2reZ`I<)SIsKFFuFy(!A;``{d8*j<0*!FMyCDV8`tLT+UiJ z9hEnjFj7xH6;UDm0PEdhpB|b$!l(0Fy>7SuYy3QwoazT8@0Dx~V-Jgj`Ww=C{os^= z@ZbQzJve&tWu+Z3#Vk^9+Zw*go=}8|{Q@c@6=C#-GuJ!eR*gX+ge;d&TlYI8o%14` z1QaBfwSd(9rNC^jEk}|O-_gpt@@j`cgbs6e;~=4H4X4^HbTJ5A(+pSaHgjIoo`uc^ zJg&>`dK=*#Rvu-I#g!t#9vW!n_(>vPv%I?9RgiOM--jN};c2g_3qDd@*etDJ;{L)D zu&n@(a#Ma12xkeU909oXe4hUO8~sL2^SsSF_ZaFh&2~A^#y)GzCp`uL2dC(1Z||^* zaBggOMVvmYwLAA5kYJTA7HmZ>ePId9%5(GXtHvSsT4#-wJV!Gk&#FtVo_}(Ok{;i5 zR7i0B!=@r~QmkON<0*R%@V!Fezp*7g%8)hykk?_H>3(ukeuYrk8&}$YIKwl~vr<(olX-)tS<&gVxVbDfx zSZDt|+C$&{5d<(#3w8qD?Na%Y)VFGP$T)?PgSOYnkO9V#y<71MPSKl8H;r#^6?R&iOyHnfLWPG|?yzxdRKB)s&^uTjU0JQ+v zVsaJe92{Ka{wT4vSM`a*+s{QEe1Tlra0x|M^A+H2h`LFI{*dPzLMSZU2mX;jnHv3O zRkqzi^uTtvNiEUDiYFNEa~s4DYeU-R4(?G_tDNnAqLX`M;`N`%_{3Fm;GHca2PmyC zPdR;fH*F}rDNTp(+U71pPb~j3&+@)_g3r+M$xn+T_)wlU@P9};tFWrNuZs)3g0zy- z5_0I44yBRq6r{VmQM$XkyF|J~8YBoG)u$Bgr>sVwmixl}rVdLRlPDs?J$SE$|$2Qo{DGnj%kR`|`oPXLu+= z9lr@=;)@I7$RN#j9o6aL8SE9=>!j>nW99@HTrG1uG(1nz>d3UFgiXR!Opv{|o&w@md^o2>iFan)W^PS` zZm{0Dc$v56$yqADB`HJh#$39;5AUsouehd%{|k#?NShvB{Q1j_L-O5P@19f2Pw`c` zhtlkO+UIpZVm}Te-g|kT#6(tp0jgw-7CbfJFj&Wv+3X3yt>@T-(yVd%nx~NB?n_2EHufbwc5=N4 z3NwDKRzWcJf0t+~O~o{ZV6AT$}sHJVK9pg2NI*%~V$QPtT_f^{rNy zcwJmO$ko*n?q@@QP(9-=RlACL9w|T-{JU0Ld+5S5w={36Z;YJ`wb^gjJ&}6*H80n{ zcVUOL&|!LNb}2oIK*~ci4$Y2Hy$qNjS^39Bz{}?D0baIs#qT^n`FTcaJnjDcBbjV) z>~-bnNIx{FVV6RV|NQCcppOcV4Gd6Q6=E2ZO|%K zqd=2h)+JvtGX3dxqfz+5^Zf2SF}Uc9Q=9d}VLOjuGuJY5<~63g=IR31z!rsM;fB75 zUabN{ZGAqB3$*j4nN?hwyKa$beKVtQ!^8K}5D>CLYj{A|6^GD@4DsbgbaJb#N zr3&W`9F;0AF(KiR6Jv|dre_m1UwZ70ah1F8gw0q;Y6f2|lG& zyc~ZITUkj^q40x!v(&}iERK~eBr<;34BAjTEIFajPvzqbCq;7mR{Xb0uZ_8ky zgq+`Z_g6Gd#l2|Iig*z&)~+NWx9F7PwFxuufg03aWUk)bg!W^5ZBhP>7rH9$*a;jP zwr^eb>UZ@*TD`Aa9a?>A^-Xr;m$S7!slPspLb+)Ws4Q0P6NPUsqq}_Xb=lVrIxj_| z(9q}Caa6Ug8N~`ST|2I1R<#|jeT|Rvm~4{|hwg9lcrrOTj`kK(G7URts?ed1To@u( z5|3H1$|?LXV{9>z|3aPUhOVQk;GH}zrc@%)kq{0KUHjk<56Nold&+7k8e~z~dW%egjfaKotg(Z%XNy8*VfMvMlXsv}o42rlzsa zUcneoiXgbA#h?qD--o+evIo}Z>(%E)^_Bj`U-@ms3JX*x!&Tey2u(O* z;&D_RZoDHq&xRaSJ2Ucc4^R#2X}~Xd)@*Hbj(ez<*!e)&Im-1|yXe*XfRyE)5k`?u z;}9P1zFc+)k|r=PA=@S22&T_cYo@H zOL%IR5*cXFsyu5(imqL}{2r-Mi`ONKV;3UM`*A)#tgfRLVV9DItr=b}I1{J~pg-Rf z^O|o8UU-j9MDpD%kcRFJ%02qmFbNUj>@3A4A&?EHleAGytK8 zlDw>*x2)`IomMFqAx&6B1!yjCy@h_rlmNvAx zb>eC-pYIh~I>FVxLq3#eUcQJQ}@J^0Vphx9F`^tf#s$Z$`!!@W;xV0r308aunU~e6uCNP&Nw~}6RlWK z0*V?$1i|ybye{TUP^v%wfiHAO{r8EQpCwZqz;FI*mzCNtEr|urL|xO5ZxsXJ57y46 z&-oK;6H8c_Cipb!7GP@Ee^9Bf$L=W2pQZOH*0?*I6W1z0(4}^%v`$be&5qWKv{ zc8h&`+lupH<(7iaUa}8ucXxLX9}>f4NMlYKCVx={Ab&&iVO1+b2<*k`*c_R;_x1?t3Pb zQnZIzvRtJck)lz2qp?=3|tsGB=) zl_s7ZhUwY9^40eyW-*?UKSP_-o?NVU(TL&Cyn7I6#P6>KLGuM1Ab4t=G6?;m^3eA* z4``R~L5H)xlK;{m_C=^!f_PmnD%&eD^4dko|5$6^VWHj5EiA(h`K1|@jGVpG{l>a) z2i^hcymuQNxEI#Z4^-`+_jJSi9@J&kS&wjRC-obFLiOn{wo^g-3XWd zkBzmM-iH2Ic`2Hj4h?y>+=?Q{6-9+$J&@c*b-9JLPxZ*yLQe%}yLo3TB|YA?9fbgK<$UJcPAOc?9e(v4z31Z9O>k4kSX6#v z7zP%mD&GDF&3>QM`arL5^hyy`j>Ihz9jK({gl1y2cD<5(pGVy?QFmwzKZLJ;!lJN`e@Tb3 zRmq5Ecb}I|tL8ddIz{*&zsu+kTn~f$ZWUY3J^5QSt-hU4h^n#RdH$&+xw!E;_G|41x zx^TC5)Cc!(8~pg-qXL@C2EKRiuGz}Nh5+pWlt>_3Qp+nQz$+(U?0z8RJ$tyGqInXK zg7Vwe=8F>>yeC-XES*iaIwECZo| zW-)-E!>(A|4MnpJvm#h|r+h0VVL!rdlpC5xhiqHhH+epXi#0o}$hDm^1t+TaVOx=G zN%pr%pH8J&z#`ezb z>+UB3VYiA@P%Ao}wg}Zp+VbjOCbni`#7oe5$OB63pTkqL=g7qYw(}OW|DqgL0o8H; zrBh7s(+AD+DvU8!DV0x$2Ee!d9ihCf zMYgvzNU&r1Xv-T#G3*wI(mV$t!(AX{COUif-AcY?v;Ddev9^BJT+bEi&;k{lPY)xq z&JYlGo%*P!HD}tA&q)>K6XKdAc5IQZ-~-$FuYE5-vC?*m@Za|qIHsV=n>zhfT4z3u zb9^(uU$yiXmhC47dkY{Nb)HlSzC7m8-np#^VHj?)`>h|9GdrqT6#=bcV}0FcVD6s$ zKwzYqr&4<1$DR>v&tS7G`#X%&J+@8_oUWidcnR(B9a3;5WGLg?t{6! zi4HnNrbGZ^b<0W%n1BE9Tm_!UkR^ELD1(?$V=qD_F3R0>`s?5U|nZ0>w_XE9o3Z*JLr?&}XE9oy~ z4S-?@*lcB3sQb8j@ULSR9zF{EIL)sj8SCCzgS8+Mt-C**m^ibhxji~>pmL(tP)0s- z!)_rGK8j)sO-z+EmVv23B&^`pPr>X>tNY`|kCQm#Toy{a0Z}p6+BxGoE#%ww2#lM} z8+QQP<~KsV6@38n$DNgYp3%`daSPEcqm{Ha4<9T5^`q7TkG#pD-Up#uqukAFNwnd+FLMUoV#pG1z9GWeQ!VKW^9_ z2u#}c<>P3-$%dS`;~qGn?P1www+hJ@DjA=g7EJ3c@{~Mhjr!4YL>AUt%gM2Wu^^{_ z#;49%l^mYo9WV{tzMguem@B}7bDHM&dLiy#e{Vo4yR&CTwSA-Y)aOuC&Qs|EaC9#IK(xBsxF)YlcqTip)6HIMkuOtC-B@Y)@j* zYdHJq-C52B`syP?odynw2;FwLYHqQLVE6jGyWHUtD69!VTdaH>Jz3^@t{MQRHW`hn zaCh&rz|hM2fOHCoc!HPvvY^xivXv3VZ>jRvpe&u6TM>b-Js)CK%gK+g#((v;j2311 zZI4PUV3Ume6GQ+3E?{Cco#UIGE#0VBI1PAhWrui>B#Hr_1^xkeU=)a1KoymxI(1?* zp$UQ*zH!lBi7KP9LZ~<@{j5}m;TeQ9eD3j!k_k*_bkx(nn z+WCrzXbb?r#u|loJ^y?Yh<VWP#$ z&e?8v&f%ejqAcI3OG8E+>4CU_NxG!@3N=n`9{K8iW1PI6`Hel5B^ou^2=p)EW*D;w zL->rrw`y~XC<*dd1IF`1&K?`gEjeT?a+_P zyL;zt)5SNfvAVDhSn1iL-+Oj%F`G(uu)~AUat@{XH7#RjytATyM^SaNe0oQ+in-evPPc{~W z)L)JH4Pu3cpJ5q&jXt~WyWZY7J# zGM%q-`#)*=`?H$@tUsg7M6gN3Av>0@FSeb#ZQM@W6Kth-z1_jMD-Tz`!_7lWj8H6Y zSN^tSCjIjLV}{9a6R}!RQ2x|DubXwl*1U;&RMP|*HJ;OYO?}(2Nc;OAYPJr(`q{RB zbE8@790uzC9L*Hw|CrxkFA-7~P`?;h1IDSV4w-LnC1o6 z9u4(BtD)0dKzrR{Wp9SZG{pz$qo7S?N5Q|Z-_?0v+s;Z$vsVyrkMMe#2_IayO*dv) zP^cpyEvi?t3aTOclg@uodzourUa+y?4sHuJ#{7Pr;3k^*E;hkd=>-;|Q)qXKsBzA) zAbvPUo`=F^D241|Pew5}qWYt34>q0@ta>GNtHtH1>}J^7wi~B!xO`#OwHX))g-$<{ zJwRR_;w3%@M0?Oj*LfxwyIj3iO+0I=sN8S#VV@-=ES|?Iqo9;hV3^&rj7&1KsBz7J zk^P;ry?~N6yXMXqYsdVYFl$g5$o5v(*;ZjzXYtr%ecU;rNdfg&msiC!~hGu6kh$NL8L;=Ws~s4v*AHG`Tdi(85nO!XU^MrV%UK&?1JNgfbf zSq(9#!TyjgE9m-{Rj@jpkE>s%?ISA{C#t#VApYkBdQ~tC2MMeu7G5FWRuV$%Xi=WR zV1{YWDQv@WjFmR$cMo+&T1SbE)Xuyyin+euH)-WQ;M|27ypj3Ofi;gZ7{wn*6~@L`hIbwp(y{U^sOQ0JR6@gVdmiFF$%&<-m!8P*is56P?Y@l z^EfDZ#T>uD*Wop&1Oeg$@9w`_h zhFdPB5_WGprAWbq1M49K(BeBrTRTgGxyzXFI0Fg+lNx3qwlkrYr$;`OQeq4I!AUgh{jfkq(Qb9-ofmtXDyrFIHvl#SM zQr;uyk5fp;oi#HIBbLyHSFT^wx({Zi$010RbCq?MA#HPPYR7j;)~REC>gsbldkYb|ODE3UZpOI%@=|5-PEr;+V3IAz zd~cl_l8-NPJvT43GZU$gD5~|E%?; zG{x$opsd(%n44ePuYh<6$bE9clQIC9!bGlapTvFShW~XyPs%QNI5wHL zE+(dl6WnSk5#`cbqMp4PY&5sXyr4WMzn6cqR^1ym*HvZrKkbn1f()~2p_F{xYLEa4 zoVsuKPu9I%i@`0Nko!#+;p0fz7761x7`(mkwcnAKZ(;wQ%Qd}szQ@YgAGy8a6zsQ> zd>k|72_<>6v*MJ;-jSQQ2=ZB0RKA0s<9I3UtKGYikG3ePiXae;KO+^k(%sLokG@&R z&=9u*a!|yIEI%jl2+f-A{9I&Nwt-BTy>49&>dosg%qP?R+1{P^cJ_v5aS?d7pync~ zxDL%&fPIil51ZOhAkDzp4RxnUphLU)FGZV=9Lvt$OP3nb>?ro%7&)7bU`2uYy{ zyGlLq-GJUiY_Ayo*|h-eWH7x=PKGL))MiE;~9n8XvC7(-K}RDWspGCtn|w+p@CV+jgMh9N0#sE=yR2l7}EY z>TYtw!&Oh@y*+(ElV=Ca`z*`a#g8Z-!94hwu}$hXAP1ufM2Ov^NgMT7PdSq~01@mrsYgLv7y+m^nKq zM*Cdm&slK_-ZYP$jjTS7EVVcv0D)24@0`tc|ISaz_ku4I12_ujtzbdV-mVX~H=VbB zdlNKBke3-QsM<)Hzg3a25%i<9KQ?x_MuCde`X&QNjHC=CYd`mNH?Mt?K3D#)`;#y@ zB!QVXB>t8^7loIL8V%$45c&6V81q&yTW5>;F;?4aS8Xh?IjuaS9IP~0-#w+gip@@j zeQ>r9CX@?~G$<9O*=6@rQZ&-d)-34c8U#|KWQmA9NHeae*uxxbPZjIfou(B@{sDPo zd=QE!nkn5pfOIVP?Z5H^J|KrF?=qngE|4jmyBu*Z%<#g8 zHs@78Za%pd0~`q$N5+A~;HO`w#Vcy|2a2+IZV4>t2Bm+NHVAE!!&NB4#8fpk7bi;v z7@tjjZNmza5c4Gq?OTb(n;3$c`QeI?`VC0WBRoJ+cMRv!8d-=}X#q&NMCW!lJdFEV z?aHQQCYA7iaZF`p8%}f0j^K@mt(iXzg6%l)HaXU{Tfu-Sm_xREy@r*Qx5QG7jxE zsj9)XX#`Wib3}ui3=wu#81oDy%)Sr#T9%!<9R+* zU*4J~8kCNDFgS4Q#h)l*5+7QEQWqJx>|_$)BON{l2LuOQ7`GAP0X1rZB=d@zn^nS z8y?zh{Htkq5J0TD`w(K^8YCcceeuYdhr` zl=jVhO7eYn>>Q4MY8+$neEDn~OhzlfPIg!D)OXT`xV!lx1EW9hijf&GqQ~5x*5Ts{ zVm|i-%r9kLi)YVTZ&7{VC4T}{eld&y?4JPqpG@Lyfz$|u+>V-r`YfC?WE)kC)71s& zo4G6F?N{nW5E)0uUpnMB$k12$u^9NQjCS})_ey=XGwE`K&HXDXzXvL*H{OWkLP*gG zI0_`GsQe06UkobzGW9c}nS&dY}_iH9p+=K(7D9?!ot?cW2WtnzW!&rRCN199`SpOkQjec{9cPNaJE?pXtfiIYu)VMH$MFuH-cq||^0v|*>Q56_YPg=WciZ6NG_m}^ zj*1f|X0!5EKZncwe)Q`v&-ISU&=RSsK4k)l~q#Y%$43^ebgfUo5m44z^7!JM-eM^C85BI`4YsOa7&KRqJ^;6 zlcPYWm2Rl0Y*vY3JCTaVGOKIuM`fzEjf>sMK6@uHzK+e4v(>*361QDskFl`l(eOXp z4y@eVA;pPgKHG`vw}+-eo<~n@mmlY;zsw#);?+Lvt$XPG9*fLH1P zHFr|LbzG=X3dO!)ZFG^L+wU@ORIpTi;JjhCI?Re6OiU#nx1DO(AHQGW;Vl0>IYV0; z6CV1F1qtF&C=Y#yJ53go*`2W>-stsf?CD(Oc~IW_d*GFv+=#_=zZ_2ZY`mW3B1F)6 z>@tsAa0F2O%|fkzkdBvF;Hn2Y84~GK%07ss$uOwxsycjG_j2~eFCVcOcNMj*?}U^> zL#aZ6-Rh7e3wS1R#F79>yqh!cBe`Jc7#;QRu<>sz1(t@c&vd-pXnlbuP#N`Zy!$6$ z{s65AoICikPyo7zEt`CZEPmY@E)4G%0#5?cQQ&d}qtgLf4n~I{e=4ze+Zkk80s0=e zAMYoTaWpxR#f<_i+;!Ku!e-~EGT*NSP;FIPx; zh5>um`Z_XGa{R=)4>1NGAfmu+&7b1_KR(ymij{57b^S)SH!GQx3hP`iKvrJf(Z|Pz z1=GdVb?@Y)^V(+xFUM=)%0uhcy3V<;&*kIxd_o3uM5S!2@T3j*iP%ySR(tzcm@7g|C z$o?gF=%w4kGut`G1r^Zg&EG5A^X|WQ$^cvYZDUK{BqZx|bnf`rV@gVRnWaJXmGEMG zoMUp32IPPQFAtq-nQob}tgUZxbob2&R?9`#wUCd04+w4o<<0>n{wqWfs6j1LV66C? zF)`u?Q1+kZ#SCKKrTL*4v}SX2&K}sG0dsU}v$0ZW$zI26+izwio1s@XP9-d0xD8Gd z=&!BpWh+6F+c$_L3?3g!{(4R1H_eh$*=GAyA@8$rkDc*Cmjx_rg+YzR_h2d%KBOi7^2MT;LWwl^o|u2>l;8c`!r z^6T&xJi)KdjjcJ)b92+azUCp2yem~Xn5#fZv_NQ-_T_HGK& zTPh$!K+0QM*#_Ra{(*GA`=tgm4j3%}Q-TB?Y1hd`N{N*99Xrbb!*%iI5&%ZKa-uyk zVQ}pbfwc>Tu%J01{x^NU1ZiI^!70S9!f@~F^J8ClBHdsAL%-cK)!5nQ`p?UB2Rm2= zLjtPcjcQA_&c)hDz5htGYHVDzpr5xc9vKxo`vA}4Kp`)Qi$8oJ{uixbxmLv#`_>Hc zu09aSVzF%?$o>Y?(aB4W0gGY=DZv}*pNNiBtaU2(6?=H&whVRN@%HOqW`^T7WhO?SnrKQPLA;Y*X0=#$Rq69PTLYghN}E z^T9PpL5iMMHV1Ut-f2dDCJ0%Rqd#~r=5EuuN685=FV@bmYVBgce%1O*}fee>rdWONe) z2|>JfS-e|J+v`Tw-f)Hm^Y zcfZh791!_)!5I+sUQy^`1YtXz%==t!fe#~r;LX zZ7}*%G~@K`l|sKQ`-~w01ENe+KY| zA6x&zNUKMVCb4hMf`vGQ8FukTe|bfLaR@!x2~{q1@Kv_^6Kgj@i$*}nE5rIShDsa7 z(ixDS)w}tF2-i5gtY$K=dJ9F2uAUbRXlsAA1P<$S!C90h9`JGtz9&Rx`M^1Fe@&YB zc_d{5-9&N0jeEd<9-JrqcteCG%dv|I4rSMfl+JMROKac>Cm=oS-E^w?(I8f;#9D6< zCh9t&ZIArr-_S5?CsW%nXKw?q-lIDQzBz;BM3yXcLP{hg@2x5PUxf@M8t~cus*3B~ z=K@xdV4<9sY?iVA$JuYAZce42gp}~=FHN&PkDzV%sW3t z%)y$6q1D7JUc8O6*_G{@^KXMVf{mG7PJUNU3~xFST2%{$_Fx>!MJiPr7jp!Xdt7O= z#)Vpzb{Vv6F_*D}s&?~?DweYM*gpH{kyg3iSMx1J?w9Tn_HTA`ulrVb(h#F(qe6mp zmWV@Z9b>vy;v6jTgV+L5N<&C@4oP9Rg{zud>3{E-FKqw9VD8>F>iG8F8(&yD+w=*c zgq8xt!CAoiX-M$Oen45w!dbsBOi0t26 z1Ju&z?J&7bb5*Y=!DtpaJT)%fJM~%?jLd*{a8dC&=6=XZd`BG*y~z{2R!~^s;K?hFtV&| zh^T=;l}GNPUB)H>ff@a92UgdwiXdr(R}81!E!SVaN+Cz1ie)aFIx~cAE}6|wJ&7v5 zc;1Wy+++GKK8fEta6lH8pPcubi%bKzCA__oX(F)95{!M`?n^rw8vR=Pc1>*2i}$C0V%yZGkue4Q!)RRO3C3qpkwbjD+3U>>$1-Lb2>67gj<-K%+eWS!)i=E|rEWMXhhc?gVD zK}WE!`R+Cr1==_}$Cl5yi9BJqOD?1gq%1sd^mhuq{2Tlr9 zN*&{9*;6LVS8a+DO|T6aj@LC~y-J09PN?3vG>%FkvzV>bjWn5bZ^(hMcqY& z-AXkW@yd;BQ;7&7Awk@qXHS^U|Nb@Y?74Zmgxqc&SUxU5bT@amK%1C_+0JIX-)BEp z$t!VshT>WzF4J?OBJ6v6MH?c`kxrO9^O0UrVaBHJM_Wg?|Ge?>0q`J-00 zL*w`M4@cNcQ|M=^FyWp&@_fAqsV+=KtIbP(Fh<9_Asuzh8qCdQoyrD_9tk)I6tO2^ zM>sq^Eb7FTi}CxX*hGrrUrl?>>+W%q$3Pil<|gFIud@$ZgWf0`TMDj)|q3 z6C8-6SEEfjh`9Xk(mV00#>Mc2I9geqDh%<944SCMDo=Obe)0#xsXdN#PZ*B$2NQLS zGfyiSw1B}aI>e7VpEDJjGK-f4hSpsdn>%a9XAH@#XF!!4zIF`m`c(IcbdUh zO@?XgtFM+i1J-#w#p%rzRYZ&+ZG=4l(~Iy2cdki6*m>fWb&`4`Sil2f-y|w^H^rsx z!945jta3s+wkE3Dw;LJg!bt}xxc{tbPhr#A2+l}r%=6PO*}6>_Xt?gw@}2m3U}SmtT4qqXVLp7{8M zgmdWR-Wx;gP3{E5n?Siti^{&8Q8=Jus~7d(J^J%(+8!Zs>(SKu+6nU#bm=#OIOv148H0*j7luV^L>dw@-uiiYE7#X>tAI=@}a|o)QcOYksyE=Ut&)7Dm9wc@1G1^VuC|?q8FrSw-E->%O=J)`ibJoW4MdfoFL+Zvpf)N*nPe7;cJP|$e#Pm zTYkiy&+RmA^eE?eFK0e>xSjFlW$n0MZ|H2VpV5Md?(q{#S^p`d=xIxIrBE6O#Pkop zZk|^8x(g32j&sWwek_Adu{VUfvULksXSZwR&;a3%V!9{(;VR{`8D;a?=di}*6}21{ z)tAZA$yk=y1J#Qs-qTl%c(}#NHyWLfi9V!E#B26ipai6NK6f{i+BFrHn}QQif<(M z$_o7bDi+LC!kheQe?$_#m!h7q-Mq9TYSl8V?Y*K;q5~rGu}mZgDhn){F>-EK{&Xzc z-^^9_kqYNsqFq{_q}wuyEm(4_>)QlY7|>V}@f9Hn;i5V&(TxL*8*I|}j<`wrBu;~D z!fc+yWh}$+*QS_X{6Y1K?^PFWSNI4N-%T3-M={5)k z9WjA&YjR)Ch4Q^Jo>I}Di7s?|M^SMrs?hj|{1TKh3b1$~plqllY$$V& zNOIRpp?}~-T)~0WjD<55^09!|(!DjdxS*p^O*AkDZ4xhj$3JuHdGWb({QKR|5pCJ@ zLb}rZ-VV)b!2A)u^L(4V+D-U96Ifd;sF$Z^$+AR5b?92-8VA91vUgMZYAy}?P!=ZB ztYxadnPAZTnQjA|>R<+G2J1|v$~|(=R(c4r-K3cXIJU@cB-~bb29>WDHzx%eM%vMc zJPG9L#1NHy+L(yIH@Box?Wdo_jSJ^Z9rqF5&TKTTx&<0F0vZ* z$)mO;tcr^kjX}ES((W1{-EzSsz?{D{V{p}0EO7gI5|u6*f414(Fh${4KQMkix}s*S z_`N=cp}rTqxMu(nb^quSKZc-I!B5cIY6+aO$bZ=8pCvZ%Y5;lD-q0%~9lN`ztZoih z8L!Q?Hobp)pE#d8NBR$3nEP)q!qsXYJsTuQL#(@y1}q{(kLG-#yC2j)IZCDHZa#sD z^^7X2#@3GMne7Ptr0pz@ZeC&iM+T0b$(73j6=g$@8-$Li^eXhuy1u}kfJv%Mm+a2M z2NuxKg_XUf=s7ck=n-N?%~M`_haY|0K(ntEJni7)0^~L%Qwr>J)wlzoof@X1pV5c<1L+(P}`;Hru3T9XcowHqbyubeJvu-|-1Nd_=Bz?r6 zSS?+c?NFa(uXtiABv=5T;+qZHA^&B7`X3m{pW^w|RQ25t5{!C8ep#9}PQxnIJH0v5 zYkZY*bV^xcP}t`3$TLw0)2!3RcRmGW<}=W17nhXvR8DZalz%QPSS<;gMy^~}{jr4n zAK{3{!&a@NEO8dgw=M3_Ty^uN;M0By9{y_RIF8Noc>HBcjCgQ2 z*)wF%jk5-*t9wr@wCne>4`$HsACaO=>IU=s=^4+|xSrK$n1V_Sd|ikxL8k9wYcPNt z***yV`KmWWRhWdm(^vAln-KXva{F1g9rASch*AvWX>i5dJ&-yF>$B9tUy98j4Ot(P zu3z>e7eRPuP~L&W+e-dx20b`R;Xu@`KnQPIa*!BxW9;HA(A|Bc6)OQutB z8e$b2Hhz#f;;f4$!C0-6&we8T=yT4-$W20$V~L*)lVf0kVtzjR&^B9H85Y|8t6&7T z_NpYGvbj#Mt`GJ@tI~e2;P*lB3O(E$o=R*JfHP1t_dj_M8 zVe=%IBoON6F6W)Q9lZ?~>pJ>LSSEs+4womm?TivDyvDNN=GhQ=pi-hoyKsF55*3b=-mHP^ zgw^g!OrA^Ie-o(6+thgJ*sd&r%nM_R0z>*A{+vdthZG_yR?2O*)Wg}HtoR2StuP45> zx>0&3d>~l^8$n9d7jj79h=e_ax8&21LY*B#t&0FJ$xU&)b5V#o`=$}MIf24=G6l21 zybsMrDh|zBs=U{L!BfD--xdUlp=ZJ76od@B5tcw-uI}4=!tN|DV4hZ_$fQ1Ka`A6d z=)aZKwqU^=;FJFI7t~=|R&-~U{Th(`-OlIB*$NlDr6*mUAl7Wf044mHy02Ntb3t@R zQ#8_W@DfRQt#?YVapS4oUQBxTo1{iLpkRYjg?+hj#%&F3k5k%|m#XK9B>GwET&1X8*l*w(OKZOM^Y!2a z=9|5Ck+EGQ1p)KJtk-4JgA->2n54{aFm6FP{`C+*H7Osv4HVshFEwap%i9%F+EBvu zgCRX+-&!6daVnKgm&{mzyQ;90%RDWSk}f429SJTwIdfSAT%2Hq38;gNoNAH*vSh_P z%w`E+xx2fCR^A=VHz(4uN6vaKcDZ~{S9Z;@zLL)UB$cK?gU`2;#GJ4;Is%TK1abq_ zFo=g2_aILUAw>HQ{+IDcFH+ZH3GEk}<%J^WYO1|2G}}`OeVH6+ubj1BnAX=!EX?fH z5DoKc>fBU`2%nWg5?NIKJa{T2idllat+|wi*kEcb9*hKd9eoF{fSlFnZhaVVDBqBc zfEB=AjPf=AjPXg_e8Ub-PM-z!M3am&igbeRy}LHNu%6z(h&Qq)ttk8r2nkpKlk@KF7)MmrndqvR%*ScO!!>#1Id@P zQ6h{hg_YCQ?G28jU=XF!W9gO!Ay~GgNwU0zcqNoOyF@#vlsos`3kTu1)Vi7k#AkMC zv6ZPvs8wo^=o(E6*Xw+=oH1|sy=;O?gFF4XJ&()~0ToNuu-TgPo5{vmllc&nSFBIl zx*%4*-hW-R-}j`-8mRhT(qqR%Z~Ya6?mZ)I$fUm42`5y?vU?*%-fifv+!N%CCH zuq0}-ch~4IWuD;Dvh?E&^@N*l@OoR&Tj~86=*{91#bplF+Og zTb2W&D{zH79`w-&lr|~8!g}ci{f@B%)?t-kVPZH*t{}9V`&Usa<$x^(5o1UXe-7Th zfS`q8U*ys1i;C!sH(oxpX;k8QK&A3>-?`c>ul3!&R!dLR=@|GuNeJXy6Ic?r03VlB zzxN=1`~4DmUC~};MYGN~;at&D(296(CWtEv9v{_!KpEh{pFKwqIz_Ox5IS+I=F)L( zXq1DP!6>sc(jCw4GnsFpzII^UdZ$n3M8~`0R4k0vz}`q~aS!;46Js;Z5C3W#WuKWF zHZZba9}X@s)Jx+z&AS6g+mN&8nvUZtIc7R_qKIom%_aU^p>@NbTmw=?qWp&@ zf4Oa`x^AK_ME#j|W}*m=P+3hPn4vQoa}D$McMj2wM2Dui~DgztW(lZu-VI;mOIuzeCB zh@MsIFOuFYPsrw*tShW6&$d||+F z$|pQQN3o9ooR`-G0G;xC^I@xM>F1DnB zEdn^GAE`x~q=J|8oPmZBik3aV&^l`vsL(mI3Hgy+p=GZdY~Iv=r4VJg&^M-$1+5F~ z+wIi%TccjpHA*MJaf_y zj2Km0zarA%%YBVsjzd#QuQiy3=jH`Gm6x{hSJ9GjmL9_dnk-6YGCAln6|ChRZ5tZH zzc!qXl5drGB3BCn)wHI+%$3Q-)TfM_-|inZi8(GAX!e=s=?+;`HyPdrX5(0T=I~P} zAhIsbg177|kEYyo%YxnMPM2_gs*m$>aKOp(AAtEFL7Tdf0jt)1N9Pm8f}}Lm6D{LJ z!Fs1EhftWBD8The*$Fs{h<-l;Lf(9o{j-k+%5um`*f1=5o&W>+MKC>ZQ_Yg!!-s`A zRtJ~RK}*--Qu9Bgt6W9SzbBmrNeAcFe_B`crBd!6ZJq~km7hQBIns?}3JZSi=VIHf zE#lj~ykD=-#r*Yrb$W3g>rWZl(}yZ?CfWXXq&=g>bKUqM|4YXiW=H30i>HCUeztz= zN@{aN+cxzMFt*r;eSI!V-;4kK=?l;uG}zYnT3dGac?-WmKJRx$=I1&#>5Gk8 zpcrN%cdPbEviq@5H&lg+y&wIsureMO1NCb4Q1R~`=>3?oQ=tWvQHxT+mJ6fX#;Jg9 zq_mDn`hfOYE-`o%7((%tj~{9wu`3v3uqFXmz;gLZ^>xDK@?y%}#JZVmna!r)Zc4n# znmFmwsPS?YQWb{zi}u;G%CyTYdgS`cF%eFLg2A@@qvR}{xP=<83mA8E_^>ugV3-Sr ze8qAV_WxPs(P_3X*(o=!P^TrRjnyx-6;}kt>98Bt#FX!M*i@wS?asuS#fuCu>)us( zydAZEZJ-+mQoO}k?&H{`w*3N|Yrl4ID7B>Qusz791}VxFN)b}AthS>$l*Ig4AX8(a|9f)KNs4iY6?fHWVt-OA%ZDMBGTYMnSAZRu&M9YSs-BlKK3?Adv%3}n ze0_#GW&;PcK)nPJ;1_fJ*@cRI|CEvS>vde1Ja$3Ob5>G3Ce`0laCL#o%zQfE6?5Ww zF^?b5#%0x~*@-ny8tAYYQHj`sHerCwSM;yaEY$jxA#?)Zq4$5pIw^e)7i>U(9q0q=CET+1TL*d?Ghz*APM6TMx6_@4P+{& z79HemP^fX_g(oZN_}ko*9@33jKbfsj_Gzr4Z0Np8BxtVZ`^Ui;^qQ}lLuznDWl`dX zzo*Ao)D-*56!t%Utf@I$0USu&L?|jW+8sFGqa6T{)-T~Rm3c~F#2*-7+p<8)bold0K?rDYk90tGI(Ya+9;bma3cnt11!+y-{xHS2C-?DsD9qWe$L~E!& z|53kKva38l+m!ycYx4~tPjBI8zz~aZ`MD!(mnM4C$|F_Gz;6e~)Dq|SOnojD(zOyp zfCmWSnz9Yj+SvU}WB>yijqL3+dFI~s?$tX_9L}TyBp100Sclly0~6pf=#h$en9NzI zCi(=r9RagfBjr7XFXR$AUG>&n6CW3le1J{4g2Vn#bM`+H**ZG@vV{zZ!&%-sTYR0V zFEe)f_G(14N;Rf(n#UraYS%=9HHq!)a?SHJdgxoKMOu@F$v8aTjmlAlYBDCW(+^=> z+oI1NcG_M}*n3x%R&}Hm(nC};`*W5i4M(s!6buPLi$wi~_BfylB`pLkJDXtz8sZsm zSv&g1q70ocx84T*b9GDf>Boc;7((A@4Z-1+p6gkxZ$D_rtPZJR33#V`T6^3(saE5&ccb zFS^d?F&Or0cuR#czI_G`#HEt3&9VVqzWSVOKB4kJdOhe$E${8@owInSpB>wl|8b>z zp|CC+aYWyiYxMUo%c3V55+fWW#If?@@+6VR5j&SV{IZFl@Rwt|!fFF;!0)Wq5)7Dp z%@#!po!lHx(>(4OdE(RNBU24)@FgR zKNEGF?Rn>Dx{hXO2+De_zAH3Mn3ZB)cn8si|u9jSo4{@Pv98eyWZ;a z+C3agP4)S(_u$*{>vS&=EFzPsp+c|CCG;~ST>Jjcu8AMQ3>NU?f_ck;?*qJ8ItCOzAw?Uz@dZ?~Ndgo(xf z97L$w%&y!GIiLL9I|Ok~{~6p~bsyWhWQXB$i}CiuRF9g@o7?Qy5#bI-0#&XLBw0ex zj0bApAeI>Aal{_s0Ofgvbc_>beyL=c4)c7i$D62D`qOQ5@d1ke)G4SLw==AE$p~ZzA}S zlu7Kqm-cw&$Nt|horSY)VlIF9Ku8({;4~Z<601PrL6VN=B&3EimX8FqV97=YiTzae zFX|1I@wy~hIHNg%N}Pukx0bXOnUvBNQN&hXJx3Dw$8lmv9!pQpf4O5Uyfx@O|E=>Y zObIyiw$m6ern1C1U(&Ng5k!#aC%lsFvT9z{3ObM?`B94}Mc^)1CGMB>JAw4#pq9HF z(e-)Zo5A&~l-^>jEf!h~BL%YWd{vxp^E+?I^S|>gS?SnnKO{}jvdtiMexY^#TfB0; z)@DZ`DVlA1d+rx&U8CBv+QrR?ihC6PTuMALj77fMFq3bak+vzZAbMY=oH12S8SF~Mju>Z5MIdDC&EWj(WGzEiPiyUW5w-8QD7hh(!oa_y5 z5A2^FZ)QEvy!yPpc5>CQB;ES2-w(4{HE-SSCyG}47IEbUaiv$E_lI%W4^ZjbJCjG3 zV7!9F-wXN=8D{s!s2>Sky)hJ6y_fAOciL#4MLoWnjBPiW?L>V0)j}iOwMND6b+3~6 zgF5yBC+qWROfsdfd7M=nH7=EK53H^YI&9D&>}wyk^CW@$!6PIy9Ywzb5z3&By)D8t z>d_k$J>s>op;%veWHZ0JLn#KtGf4b;-{a6~>;%U`eI3rN0>YO9eHI3RDW=cS>18K1 z-8duzNHPXALiG(Azh#0*8MU&p!mIcvEuGO|C5)1p8upbxcIFgv@upqS2+^fza#RQ2 z`76V=6BSWMT)%jxEx==~0+qD1NeDxc*hbLOY@AIfl9_@kz`pBwx|itNwDa%Ci_N7t zu@bVwcMjL=Cm>sMwQDo39GpIyWE!M&(@bp2JlK*%(MSP?qj};25eIh*aQU*;uCj-e znB-e~EAO}1hCPjS5>IKRlvx}NYUFPUREm{u)GBDU9D{ArNpL^pM1$bVK32lCP0vfK zLNR?G&EBO8!Yy_#kg%q+O)tqP?BerTqoL7>MX2F5!0t;(rpjG(x|ti`1Q- z`1PD`{g#L<#*;;dk19oq-j+nqE`&lyZN`)JIJh;qD!92YN{P*@BNj%_vmW1mYoD{f zS2&GlLT~7Brm67wJ^W1mTAUh}T4Z!gbwmj25~UkgmAlTYC6X|AYcEhfH5=-X`u`mn zV>Myqc}3kwMCrA~k;!fS&~H_M?=<=8!{ZE|`magTL<+rI`htz~qL7QMjEQtXDLf1O zqF3Q_Y`MHD%bcNf5svCnF6+x|!3U|e=tez3$M{hZWlg8<52lLwK3}ECWiqQDy+E3>+E!;N#Q;C+3E0k?MN8=_$oqu`13mDKlu3 z;0FXc^waIH-bAsSiB#nwG z!rGOk#(7Pc2dY@%>{F%!=+|Q$g z@wR7jm1+G6?kZ*0AI>>Wig;EU=ZnC(K(FiRIqrQ$S@)BW{nLEix!5rmkt}-p+?C7p zn)_W$Uyg;%Vr^k8t&%?fCyVJr>Jwq72)j!Cp+si6!KAI6o}xzRgA%pBZAyE{7}jZp zPr2zFOVO&G-p$#>rWRH-vT>7DQ%v^$r3S`a#SAq}T|&J3!n7?bMdm*G=--<~EMj}n{SaHO?OShkzCd{I^RHbGqxazfI^vrV zO)-}g^43W3D+j8Wp8dL};cp$+-wzl1OV915rcO1mZ4RBJ>FoqsrGn1OzXrjP#Ie7| zqZQDUJaek6ZyMY8%6@fVv%V1(;9v-IbCXVE?Kh-*66fl%z~^7Lo=!8}(vHt3d);Pg zz;h2DdLBUK!5Ug3#*X=;wl|<2%d>=W;LvD4qDQX2<6Ixe0R2Kxh@xcG3yuUOlG2X? z6Q<%Y!j%}ry{g5v3Gh7R&|SBNxs?%4Y(44Sneu4i!@MhoA4<-`x=9ol@^=>jRDcG;vX8&ws0JDgLd3eX=_9 zUPwwnCK+!)eSBajrt~NhVdNksH$j>I-UFp106|;@C#b1{Sa>EoJ;0(@-FsAzT3Tj- z=d>S0F<;3#|5^}kWTcia2shLE*6MH>GWE*wvj4ik+T~~$d&kvr6_JzS@^!Jc9n>D3 zEEGM<-ohR=>+s9AjCF))x!(Ryh@alJ5L4NO!pfehKX#Q&()-N-{2lEMK@+QeowmiG z$b-GKmc>emDLaO#p_nTvG{=|rqbjT+gem6-g3i=U9J` z+*4TIcW12drTPGdfzbU5QA#f6JabIPa!Y2I8{67Ic1`%-PQS%b~~?) zC^U^zR<31nO5S1M9jcIQ`431F2okb=O^c117{>BtYw_PNO~ZmQ-oA3}*yO?Qd|hsb zpzc0*j;EI2n%%p!l6Gi8KL%yRLQL2dM_rDJ|E|R5W$YvImLo(GVR|RJ=`bVegW$LY zgtB8{xmtJGzz&@26FP+s89kZyv!AS{g($XP()dfEtTGa91cCVIX%s>2z?awz5QY&8 zsQq-C9~O^}+E9(P?9qRKTXylQG}$oE+{Ku`v*KQ#t4zOB=Ywf=>(N)?iS=Y|ogsLd zr-f?#fXQ{Dm?)ph=?AQ5Tm>b!Qk%KDNrjM!>Fq=-GO$pSgoEwxES6wAuPkM}AUBcCq}Z|tf%TV* zRno%GPK~2qM9l}}+6X14uu2|a^^4;Ui}e+!MN~|dsphg~6?}um-U6|d!g--W6ewPQ z@9_BN&A)&;d9h-$C<`=dcw|EODE-QbUajM^U~$aPQ?M9>r4j~spiF&@;eDPvM@z&S zshGwgP$AAu$S%)~Saq7{J`QjzqrX(cqYvtJvt>bctWZ~am=PZqy-eRia?uJ@`Y@?GJfw>_A;=o@&VU{e`FdW%Ps6c0_x- zT-L{p%m(MamBO|0DC!;Abf{>Igl&T%QoY0I2&shtfE#+^3etyE7HSJa=i9!#eK~p& zmuaCQlwCoI0QRqDX*b zAGY_s4pmDgz66kdAc3Xg$27?4Anou+@|`nj5sA#=z*A6d9`ZO(STaZnd1^w(ogjFj zFddhj061{HG}V1gD$C&?}2~M zUO(p+3w^jW7OTQk6dyW18>2J<3X-Fvp;5(SS?)zQE+Op@2km+j35v5sD(Wgj9b4?N z#7|WMcRV)67z_K?90QD_M9!3-eXx1K2jZ>+46p5RyXTx(-;}m1ayHBg*m+Iqe@`^>6SvA`@jt1~I=$2GxCB1sMb{mi|L0F=&Q+SoCgfw` zU~83mE6@o+ACU_o8w&#Jls@F+t|i{od7~tDx)q(b#`#2l*XpR|>or((D~v;|$fBa^ z&x}K%t9zSIBI?Co^#Z@HyVGP}HlkiWohGxxlBYl&2rF&zYVjBE`Jes7F?ob&`9xP& zGdx&|q;h-7nXbZ{x!Ork;>mKMtZ{dt_bm_uFg>~KQ2D8oQWv&f#g5hZqKf)zzbgln zg6oz>q_3S_*ZFFOR3ii7Xced%_gNv3e}uw*o%@sSPwaKD@`<&h5sG{k`EH4pg>jn& zG4ptRant>1iVR)I4%mPMrZ7mfqcJ-o8&-HciLRxxTtFBk(?vSPJw)KQ;_?LjSL2Q` zPYDL1PCK1yAB*^-6lIJL*sZNHg2J``?G>2dB-=M=Q0EZ-Mx?27b92x%%*nSE(eywz z?%MR##4EAh=YtuvRTilGoE@TSRc-!q7?hTY5=Y0&I4m&~achTGiL3Zr@qtux=L~G& zL5+$Fl=P~E0lNm;Q*%aAU`*}Op?9j>e?=TCesQZQ21W9U8CPngbs5G8!b>@W?mop? zSXCP9+`~O~NKL|vO&Y8jAu#9r-v(G(7aI(y|KCGsd0{>@6Vr}8+(K@D&|}2i1AaVH zk740V+k5DDcwir^g$gMyv^sMhdxWYne;zH}$&e5qoMu+jxD>~s{uU5=OGB?kcDwSz z?YVx7r;&6vm>24OaW@UkW080FDI&Mg(5+QUruSf#C8M94U+&RqX|Ar7LYv~VurD9o z*xyu?)DsVe~aY2i=66Xwta;+DZY@J4)R*RRXpOS`nkdtSxrImiQa=`MrZu zyee4JCvbd>$j0ej7dyk>S(Vs^F=phR3#OSI=v9%kT^zk-NA)t~G4y{$;1!uaG&eWr z!~=F|Q~2#`V3tHcpEf>Dhg z3dDhzdb>zX-B?)~?Eu zjY!>n5Ed0&i=GY8=4rR%ez;E%d?@9am5WtghiO|M41!(!*5cFpn@+ica(BlF~TN)QQq*&*pDN=iYc`DIOjJxDv;x2jF9X^b`Kqb0Go0snZP z+qEhvRD%p~FCS@7J0*=R&^r2kmuu4<_NN@+LLW@QBb+pG_^b3SsT59E>dbYbu8Zg! z7zwz#T!!^+GC(hjc74QP{_9pxZ9A@Tv-#vV4(rX;yngE+VOA^!3ssK%dZ%kc?RFkD zzS+2R8)xy%P4}gl+B}3^vKlHi1{DH%imKl4<|8fw@U0Ro3ml#4t$mWHU+!E#R8GHL zb!{L2x6*WTY&?2o>h%U@l}#jeGlQpRZ>q)|R^lU21`}*N-B5%tRwry14G4MICKaj2 zBG})SGnZ_2Nf+&pjPh|v+XUr9q#}MK1@NJ}E?3v`l~hswl2Jr}dU>0-oLr9cl29Ns zT^>hhEwEGTSdH7O{AKRUsbjG)S>Yx)xqzjCN-HHD@~{ZZc7=hG;1&HK9%oe@-ry; zODNkAbL#BzEXJf?h7!66Rree%30)!Nb2>`MYM`E3$a+f}Q2!-=eXK|qbffZT>#cRa z3~>>v^WJXe-IkNzI0TRPW7~~r;-X4y%_6phNJYJQWoTh@MsE0X z|F*odh7jHaeXFTtn>BF5NkK|wr%~1zrGB^(?(hT zm((Nnw!>7ZX}l%|$Fya3$J=rDChKuEte(z#6%YI6-jeA+N}oeyX0f?=tr@v&D)(-d$zJ>e!EdFS380T? z;dJlgmPVIceCe&sg;63@Zg~pA;q#n&&jw`m)#? zEXYw}9I8(b*)>><=fS5Z?1M?&^_=jfChXu79aFy@9F>vYdR2Jay%NU!so(rxCRMGZ zDe*ov=8XfwB7hvq*~C8~r>2WA2ZXe!7=8_<>TeExd=^L6IDbNVfPK679D4H_8eHx5 zJM!1dimT~1rn@Cqt2t(qmf$O?TY)@@c)#}*=gcL0LBBBR^&I85XR7f%2D3+x6C6u| zKBz5?3_6?E0&K`)AT)d{NJtT1O7qfwNm3=jS^@gwF-4OV>pNFD-BG$n*7PRN!HD^U$Ep&uXqrb0Iz6parFWl`e6l7DHPLRJY7 z(;BpEBsEA!<`jR=X5AeEvA~PZWw%}C9ia$3nBkA#ovq)zdd4!Sjf&a@Z0Jt3XH;VgQr&+GM(|3#NwK*!7ahdjODA~6@MWJ!l zR!ixJ!?IA!sLz_KsI;`S*q-qCc42zf_^h7Ud&e_(iG>!=)5e!E1lG7{LqHZL*B3%? zDaD*Gn-Z&R0gZ5?zMsSegI~t4#*;5;C?9Tb(!Cd>(;;urdq(55q^Z=f>V@7<<;;cy zlrYO2i1m|OH&IWV0wm4~y%4f1E*;MFesXnlY`r=a@u^8s&a{dEWYDyCVu)w6L_{C{ zYO2@-hr3olQ_n%xw=xL@Y-F?3cx|P?8WuF}fo~qMO_k?GiG=KI&!V3B=;8HaG_1wW*I?0s94mk_Aa<{rR4q{3 zD?jyPs{6`4uW^SOoA^pU?_)o2l2g@+NWC6SL&3J;YKN;fe!LOD62G;l9QsMbs!RkZF# zN)+cF89@w~LUBI)*k1VCY|j4F)@EGiD%v1++wAO8^a`_1HP6+P$v>Av3t4h-aFCSd zapUWS)`x2|XUb4@|NCv<_1m`k6$v8+-|I579hi|Lc)8wT?^^7qOoKa|_}8iE&hWN9 zJ>6~%1)yn1{Vk!4T6+}NJ7=g|jyz4vH;6bjv{c$`W|GW+%|Ws(z^R(3G3YK}m-*iU zqr6+%rwT=iK}4?b!R7dA| zl_H|G$gLJgLSy#VwmcU#Ely~1;^XIy<>-AZ#!OEe;p;z z4lnBZLhU-4M~IY4gdf2+W*UbeQAwh{y}=Zwt->e$`MdiaskS@p?4-^ac(-<9I|VN>(3>X|)FC@y`7AI=xQy79c0VbQ=b!=oCw zD%w!vekpfzygFK^XEp@SG&(pv%IWC%85kPtbo+$MBQUqMe_k|%h|;ynyeN3roH|tH zgve?VS3kek|B~>yG~eDY*nW;JarOh(>HZ3G(zKU#32j%-mlwXAG&;@kc!>A&yN1ww zc>ZUjry&v(e6(a~AdnZR@q?co`ftL=v>%9ES{FLIGce3Vo93eJh6KEpXxBFh-d+^XvGY|l@=u6nF3l_VrHXC|H?`0bX^geJKaoeZo6YiFL5 zYAZjHrbimk;iX7flDE77Y9X2fJ+4q%-yocWX+Y}BlzKtryc7B|Lt1$*9z$f6eJGLa z3|WthFTfV;1`iBR6_R%k%I-Cfe%|vIpW?30fC{(C8v>h)TRdn6Y#SV?PJT43m-iqa z^*(V6WZUE(6fw-kgJWow6%)5q#Skg24V$DEk3JIWEg62OVRt%ulA%ge@p#Y=!B7Zp zM-sj!zPc9B7nm3T*d;P52l^m@7#iE{Np4@GjuG>YQo}=7Z5vX1?Q7r#P#FnaboRu) zs`5K3yx3;1ci0{Cq3K1a)H^N>CaQ}^?}?O~)g&a3pjL52%JtavAtbhhax}_-#GT%W zGc*I(`5;nTx7RHEcxOP?e<}`9&ieqRV1%45`n>@Nsgx^svA5n6Zu-T8%a>h&xDtc^ zcjzE&#;L!Zq58*pw2T_9U69-yFAJXTwaSFKK@4vRBzI(d-_Iyees_6QQw;&157KprF_vH-_j5U0l?Wvd}8i1N!aptm5TnY>bwZq=-<= zU|Cm6mG~N4XY z-*pvxIQmi#AGvst8aoW*EYep!CGhW*Gjb?=7a0S3IZa2 zyHc4s{?%{tvdE9>9dZ|Xm+U&`dX3Gvw_Y5fvm#|%5d*?-v=8ySl}6YzAp>3Ni?_!7 zu7jihpToM2XU3Mhk{7Sn{Oe;Kz?-knmE-t?fzdewOiC~vClj6vyXDOjdIK>hXG%!o z)>VkF+(nnn2(YlNG3{}2iF_Z<_?}~Vq%N8jGSkm<;ica|p-CbuOCib3AYzJJuQe^+ zn;*Ts4`~`)_58S8ODyHN{8hVm^V>f-#@6HBTN2pL#@vI3E_P4t#c5>dk62f4wQ9EBuECt0IpN zD?Cw;9L5UZwO2nZ(u+By(SJCkv%GO8LMmOs$07LC-N}a>G9yQmsW6g=H7yr`)dR#=G1Bbr&^Yd%-Ok3YLi7SKfU@A@WZhlKcXB|M~qwGemm6 z)$^#(MaieHeiS)}dg1R{Q$>=2jF|Fql(b2YY#-=Jj}zKbTNfVi_IYd;p<`#9I=LS%duAATLNof z+gN9cC8g#9B^@w8ltSFc0I1l%?FeyR#BO?v0MLJ3w5L4XHga^YaDEc0;fv?h6? zp?lEr7Dk?>WKV6i-D_jdX!*3le1l8|#Kyh`5G(|wt1W;ulW3m z$!U9DI_VGB;=y1A$!jXL?=d#0+>>*@Fum2MzEbI7)Y4!pKxNzZV=MB1T%r7N6*zD! z#ytOdtdG>I`DhRkm6;AZrvN}~#nFAOWF`2!=0YK8^y&kSIVY|QN6G8zpw6+f5tgwy zO+wX#gY-a$L%trNdCkHU1puqt2V@5_66Wv3?ZU-jI$rPRxH&0C9O5kT-dF@3YxpXE zh~1f0S_Oybv~Nu3v&(o=y`C-!)L@9J>q|SH0lZN>+8-%V-Q>pc>H(1?KYdI802O!{ zvGWtlFg-){qA~>8zc$EoEt9`RGeAlZ^ZDBgdmx4Qpkqu`Qq+_jt=SJEQ=~j1?i4}-fpckWNb&mcr>d$|1Bt{t z(YLhy&9rh9+zM4Ch6D;ojsTDR@&EGSXbx@-_(96eTGF~|dRljh z-xUb?sXnzEka|x@=@kAX=VH-NyLn*0iifsAp4++)K9zqv2!2cjrant`6S6CDhK#5D zmAy@S`}gzdO|tKW|7zdCra(wsK79`_b^r?!1G51gFmqFdy8tN#85}^@B~Uw`Z^f3| z+sXSzgD5`R$zHJ(*?_=YgrDAuKB+Tq!cnl=9#^GohIgq@Mv}EIT!*SaGat!WD1@_4 zZK_s0AUj;^sHe&}p(-3Qui&htITV!Di^+O!yQ+W=Swe%*0&R2BnmGj@>Y!iA-hr)h zi82SksG{?)WR7bu?2MPj1s&JyoVS>| zj`Yco98*sR#{oL3sYxX(c_GsT2ty%PU_`fqoq3&qon3YtsOB~%EOrL^BpQxHC{lKl?AcXm!2K2W_C*pv9WFO&mKv zNxv}sw29LH5cyTx*uFQ{`sI`nx2nNfXXHGDIBjFhduWVIS^iY^`?mg!EmwsuAjr6` z+-6awF4jL&1MNbYXP3abw|YxlV_E^~l*C=u&$HIHqbiPY3!{9cO%W!^B=b~6MgV4x ztA75(OD5elG@mm(m#BbOGk@;8Ar6={K?Vu7{bPz)m(qPOskh=9z;Vvy+n^~S88SyQ zAzkAdJUO^}dEoSfGh%AOnmQA|d+w8jiSEDnMtu!FG)$A+S~@fI+EtYNRSOUN_hncO zeD`&2C~UFyRg~&2DS;lkrC8sVMmH~R^;PXSMnxvz(tTGH;H2?4+hrZx@jLjYQS5nA7J*=Fv28)ked4xF!-A?_SUr`t{AizC?cCefN?fSDTD;ke>1aR6zUB9BY8 z2^4GjtLO^|h+j#G@4v`jW*C##L#p%zzuQo+6yaf``iB~7(UBfkS^G4?G!{{*)F<9J zumagX(IMmZv2R_gt=Qo$w2E}r-@efyH?%p@JXBI`)x117j>XjTXNW51RuH1SB1 zq8&=2kFLp3$sz7lfS4=Gk}_NV!|Sg#LGVekI&mr_)7SicB*#t=EA8S|$>&qt`VRX0 zxkaMhhLXF!6$@R$9&ZtmG12 zRt#n&7qElygbLZpnnCDQ7l*2!Qd#th<~`=Qg>(e|=wu(y#j(*7dX&883_eVr58J-| z0*Q+wEmS3l%8kocl!3_&3Q3X+6z;@vOLXvbQHN;>c7&V zNoRRokOIOyV6*)HQ|_El1%66HY_;0X<4_NWii!fo^GU1!()zp+^}x$hnW?~z-KQ5^ zD3shygf4_mBf`6lrfl3t_>kqDkcJX#@ocLrG}u)To&xnZLetQV&&xY@9`7I(6lGX*VmjCLQEIwapa|ayDJN}P-nSVdHpe*yE=Me?bk9P81bq=_4h)Cwfu7?1orCb@v zd+cTU2nz&AI*laDf`I9ZU%^D8`+&lMOtX}h5*@bko-Jk@1sB$9?T8$6OBkUe0j=s! zp=5=ebw2AXOojr<3G%**Yr9vV&U2{S2py&}VHlvjw(m{-Mb?k$uBFnp(i2m9)0e&K zL17u}mo?3TJ_^6iogxM6Lv?r}uOEO~}=Qw4hdfCol9D?FcW_ z2bdbS1)k!_t8D0c!aZglWG@BI?E^!_5k!D!gxX||=3 zD(Bf;2w_)H<=9%|t3oDSRx62G#SspRr@Mc*R7Fy-l1H0^ z4IvVMr3b7sMc7Gir6!h_KeF}>DiBU5J^o4m`X}_|iQfwz4n{Lx?&W<0gKX%BWCi6_ zscH>W7fzi0UTKw1e=&@C6yVnX3@CYqp@ zP+SmEULea8hb91PSQrM0&6rBO;^_Q@#CsHnoL4(AHU+_%A`_K{1-bQ)O^}HOT9LW+ zqo5o1u+X6`=Xt-(GLH~A9L(v-SkaUBH8Z z9=_0CFvIS+ycF%QP~Y@$4p_xU2FfdYg|%aswEBig@M|YV=S${&siAio_RHquC+z#D zN1Gu>mAn7OR|h60xKVMbz`)f)KPnQCf!{A8GxnJMC6FA=Y~<~@ii{W8#h>7qfKOV7 z410zeYI43XtF4Ez^7pMp-Zx~5I_r?S50+qe+|^o#GO=uW+hAoXANDd`7ypp4>|narO}m!aNf-3 zls>Bb9k4g{6sVt@b16s4uqm;5lE0VJT&Dm`;6`s;7PEZ|ZgkL$XQnB%aXx$v`VK|` zOyRr}|HR>;2VT$~mU%(8WH5T$RO&0K{(@8|SWL2ikI)bc?`YE1Fj7(+NK z+U?Oap-RW`1*Cz>dR3nUeh~A!&_CDQJSb96u1udMj;z%RytgRW=pCBg*^AvJ1_+0w zGeGwF;hR0ZJk8?SC|GXRv&o>Yc_|KGs7?l{E&ybC`=0aNAfcH17Ez3Y^PV{Fe^owv z9#eeMN<+)*_kMijbr<~lco|GX_xT`UGPiEapCD%4^fzO-Pd=>M0R&dC64vYZ%YS`3 zG!ivozK+DefE+lDK`V=-`m)1vp=srgs;guRRfcrJ&iID@W3PW}ybrZ^#JK+sl3Wee zfWRd~n^anWM@sIeyBJyKsE-Q1<4j@ZX~)Q{*jOV{3HVGOVP8%0?TUiekY!JA4h5TBGhi z_o;hOBM)%<2+gmKX?3#mBQBz&vZ2fWp>-r3m&<(>FB*wI)qM`V2wM?RXrnoqyT_Ai zAdWD~Efs>M*RPwHpFzVTSTD$?r z#f+wKrp1&E6Bv2g_p*~{qmr}a29(q*brcBL3F0ef$Z*)ut5(Ss7bf3%@N$q5$h_(~ zA307jm$El>Y$uymF6Tw?^;h}Y!c-GNWWe(3H5P#^Lkin?mDK>H%5(h*l`dRFdI_@1 zeGrAcHH%uySM)w=2SUK7EeLd!G+%nRP3BLABf7t%HO!gygY{!#bFSp#O)LJcWpm0e zFBE8?Cehnw(v8+4g|qTvHN;{?az}siUc=e{&MP?1d}`G6#w@DZwvRV|LiCYckE6)+ zGG3E+Vm-T|$n=JjYMEYD3*{cVSmnW=N!!VXrCoSw6(T2Br*F(E0Bn^>3w67Lj8h(w zxqRRp>aoJ=4dI=;DP^zRXa{1{dYbDp$SAEZ``vxlDNNnRAOi>$a%=~KJt2{KmI(7L zq15?@LQ)jsS?jhF!T}!5h>;lXcVxL4Ut?YB=`Vs=u4_0LMrO(t5FaZFFJt88`iI-K zs!K~wuOt+w7)Q5wLg>8rC~nRX+v#_I%~pF#=$_g*mE-;hIKb%k{{&p~jl@ZP2jj~? zo*xvev0k&%snBXMg)q`7bBBc`oOlEscv)$hGar67tT(;1i>Bp7U9Lzr;B7aXmN%gb z?-2+&hyxP@tNLB(h=Vk;nGdY73}h(f-L}4`{X}}T?z&$aGR$4+xS?lNDLS+r^!@k# zAs=T^Q7g%E#ZQ9|7?KMjk@3u`%`mB9?OS`!!0KItI28C1VaTy&%OaV0^n4!sZy|ez z$oYs`^vlD#yB1>ZOca~feSWwn^WwJ4Tl>{TkR6)zLP?;b$<-u?sCJnu6aV>6BUi{e zyCW8rW^ygbZoZuDe{6kaKvZ4SwLYMNG$v|GL+k6Y3GZTM#&bx?~9^N+O$1I=dh%_~2bEY2)D zYhy4Mm|vV&700CVRSLy1im7pVhip={ap%A6Ya%j|6-pt9WDdB$DwmLJh@Hw7s$m=` zAh02ae*QcTL0qkBNCRBe?wn<)jJ?Xmncx=MY5B&L4T}8f`tjjFl6%dIyIJmrLq!(K zX4=mHuDhS8Qj!g zX&6u1`=h1OF9~#r5ypacnxDONl~9OCA`}zAYl|hX?XQAUmK6&He-k4K9Nv2i z`hWAxjIMlVUYfKdHCesZg10<%S)8p?F3a53_o)lxGnTVQ>^8YS1ME`-v$UQ6RwX8& zXO44P4p2ZlAhQNBK9zsGh=%XUMH^}N*3yUA$5v-;P}^6sgg__ZI`BmQ?h&%#0jcuo z1z>pRXZUA#=PNr{zr)^#4mA1-i&Ou-YNNhk;l|5aeeovh_O8rBu zHjA?;70AS5)_nDYzI?le^^UW~GTh(c_#985oHso}yaiS3T2YUu%4z4%0|p z79|Dv*_Nl^f&obldR31y;Ym%Achj39T|Is#+$g^B6Vy^>Al$IVU^>}OEK zg35>o_0r*AgH&XxDs;T|i2_ys6(5fjyQ&tu}5e{ES`tj0D z2>^q}PiJ*4V8Xc8?s5X^7&CPaLZvLEoD9O-o?Y&~{FJ`A^67{~!{XE_G*@?bIok|T zX{f`_fOkuI_7N&<^@-EGP$YfK8fJI$Ldc+b#R*TQR)fL`eSVe`S`h1 zMrG(v$V{UGITLHz_*c=||H^IE)`Zf_p3y^QyWDHPeV~ZSOfSP0Q9Y)1iumm_q~&+& zvJ^eMD#VBEAu@_m+Zi=sr(<^0XIAGC-0qc?!u(IYgne8J{>3YELg^bWztiM%23;7? z>sxC~?OgfH!T9=pZc4cdW4M5q7rw|5_BA}24vR?c;yyO_3tFsJ+_=N&BrsA*DP}GF zQ@c5hv_D+G43X%v$an^{Joh)W2M|1(F{_DcIhf2t0u?90fpoZBwcVLYt`sCzu}$#> zpzJ7bsy>K3b86fup}ahr4}SX)><*n0{mY-%%p8^&0iZZq^f7V3)VI15lF3t$D9!U? zOojip4XA~-6pJSd$@tbL=*2GzC9g0haBctdeN zO*ld8izdM`%2cX^OtBz*LM&XJ_erY77v<{n-7nZ;h`)?!;ha&F*N;5TnD|A?`+1^> zq&}+p&OCK6nOz<`MPPr+%#HptCT311hk6xm#F~E(M5T1G=uctNx@T9 zRT@nV8UjEFL+?3bELd3M6_@YSE8DEV*nkt~gP?#)jzY4EIS<=8>xG(|=QO^`2aag! z{r*RK8a>4?0MG?>x@@!~xaf0$`567@$Eq)lbA13%<^O$P4Qj(XH}!jxLk9wT-X57< zNPJ9%azFGQR739mE?i#UD!cP-ktFTAcxy_e5HPWX1^@1Xf+WhK-5pmaU*;-Nu>gHa z@kx2O9`u78Lsc#~{nbD>tw`=w@A3^-qBXzRU$_B(pqOXbl!o73FSR)ZRP*+z1JWPV z4?cGMIh%bDj?F$ zkkx@z^EIN@owx$m3;H#z77nGFe23Y+J#K-MO$UO^cJg4_=lO^K8Dj&g`dV=pKt#|{ z8C7kU4FVeih&sQ06j0Njq4hm>8<)B(VCK7Rjs{$JWO1?p5KCyF2A?ufgRrk{vn-o3GP>YLj2)oULHj zUhj?kzay!!)GN}=aiQV8EQ6}Snbt;66C!;^XoL1!R%%E-lgVfEOeO5+=YQ6Y6Pn&s zFVq?eP_f3*fY!T=1M*|%AA*VNg{&oW&irB%>roay)U-F-A^o{7WmUt0!lrDvT8$;I zS8;c8!7M7#t1r51t8hO#fhSt81v#IEgUM9^eK(YixGf-uU8^}pJeUB&4!K=A&de%# zZi4w{kLVU=(VW&0R=bNQroVLY0?Nu^I*u1wvLwMvO)JkdtAW?BM8Mwt?QBi3cK2zG zB6HnGF(MjQ>#T~Oc+@PGnfN6-{qE@NS@?lcb6v2m>+mN_lDeoNq&2y>iLiNL$ASc( z3qilK4BJo8uWG!|aelFj1$S@?d4tr)jk8I^cIq(<*ea;n%R_k8vin+agU4;-*!J|| ztO{~Rkxp=0eepd9ApO_@!SBp&6Vyz+-rf5>CjIHab~dsp`yX}6qM5g1f^3UgS<$L! zIsfi%V!!xz-{avoLGZ&9IxxYc068@^qk2C9(-}q4m(o&2fWoSS^>-6UE4M6~>q1)z z8<$@er(TDkOrytMi83;ph5gxnvz{Uu45%_9Qr!p1MxCRC6qrhH1e@JeC!PE3&I6bI z@GH=pdON$J(}@z2J+#4Bj}vpKm}1Y)HN>Nj%;LNW(REVaJo*Ynx_L0c-<4CaCX;;= z!Q#4N9Z_mHorrikwdg`+)Ri)&yux2&bv9jMI!*)HIZm>0y12gIJP_y3+22?bwaF7b z5a(3Sw?+l?mtEurZ3>?>^k2xmFN`x)%2I3=W($AxGKKA1nPi&dr^ft~5KMnEDj5a! z@Zy^kJ{KrOvm0AYvfAeBOh2yV)TeSxwS6NyX1=Pg+2O_(!Qnu^QRv!By^Z%qmj7j1 zKjYC=hHpvnsSJ7IcGKa?_rW-w=j}Xg;(1~X+Z>;iYUI%m@c~L~)4{Ncrt3j%vMYFy zK?bF8z{T$B{VjQiYkA`EVl9MI(kp0Uw$YNh;|hIHs{eXboiq-luBg;+XFAAQTOoZn zDc!s1gY4BaFhi&dFLvQrToUV7nzt>_oc}0iwRdN*_#ON6Fd43%RQz>aQSmg^G<{pQ z7x4VGlPlj`6A@x$Ec8&xTliU$XK_7F=6;i1VUPGJCWl4QzR&VX`|U-our++AmWoqo zq>7U=TvlAW$RXTGHLzcXO@R=j>(}`XxwhkH?^I~Gb*9{foUBQaZH#LKR!yXfNs)+Q zg}Fg(+tVEHIF1BCQCn)<=}#Oz+Iyy?sYLZ7JIYUO--{Bzo0Z}lDwDQM%Tv+~?Fs*~ zxVvxD@9^9x6vX8-ymn4#IaVTiDv(nx3p6O#M-oM1m&EOz?d(kL^`ecI>ON!q{4wt% zwEI-f>viBt<@yRnl!)&oBvA84L+XAVa_vv~Pvp*6Xze`L*e-x;)Z(vTgsqR@e6C|L zQ!V=8kQcZ6UH>0-tYy&|a|DiS^JkC>aCxrI;v;X*7Z+}RF}9vW-D4yuatgoaRh7Mx z%U6XWjP~41zqw)r9fr}aRNn6lwcek(tskz0-QFLB*mYKH%dJAcu|B@O-ieQ|{W??E zGNyL@yS-hkySt`r{d9eViY)nw$9(NKDT6|-!s(qEj!R^C@fV}0(dp~LU!pD|HV9t1AwQf2v3Ez`km$ByzZ=VnGfuU*!5m{Z;p5{r~ zY_IAolp`4Q{sR|Dm%ebAD%x!F6rM6H$HBqarOdFoQh_`j6gIFf)SA z^I+_)R@3K{NOc}(-E^WS>bJ&b7pw`m^lDqu4pUZFzaEv2 zY*FO^b`kG=F_9O^Sg>*-0BIG-R$mpdylWPW7c}`7nfy203q&3l2P^>^+s=VW``H$I zvWrcOf;04=54)F>0Kq%%To+xCX<~63c#pCVNzq^d;T9hESQ{}rz|`l3;dp__?s|1a zMVJS4!0h+hxvlVURt#&JHm0luSF8v7Z~|+cRP}c!_-IK2TI{l23SYCXfvcDm>`~wS zWfkdA6Pjx0LBGbnjP{WnF=+Qo&j0+;o80=9SYE#e?%y2pq0}CSNGEWYNO#?n93Dml zAMUrDF+Acf*2mh;Lrn5k$y>uTY00ddtbZ(RNFif`#gh8%g6r=xg-GeEC-ijfoe3NA zS$@mfZdut-Yv*rBSvobsU7sQLi^R9%-)(*-sox@GyYtPeS1zU*v(X}PKvQZ`IC0t# zxt2ij87_uXjV_SNFF&)B$VRg9=~4jQV*+(Pw!z^_Y%VwHCjGAluwJ)~AJ+oHf_FdW zx@y(@DZL^i6=NpDbNv&nHFJOutpKFlKf3VUNy(YSeGjkcOlP%=qGG>cgZE3#JQ4MD z+l}glWGHH}F-e0(2CGvk_zpGO7`#~2zxTcNSx@84s`H^59oUzYsMsa@vfW!eue`RT zp+O2=D5FZ6H zcSus9KfeE}H>Wzv8ruK!tI3>x+2F|RE4BIlQs@d-lo>J);U17iFlZ=lQ)cc*Mnfejb8V2qAS3#52iSAn4khwE}m z`H23?4OhL3*)a%vl(>N7&AJTNRpugXaYKWU6Jxww>9;6p55|F`{*^y|uiReT6;}8P zGpO@B3IcvVZLE-&%|1-z)O+SPiv+YN?uR2H4LBgD!Pb^ZwIP>FE;7~**RI9bdux*& zs>W8pgIOB^dy&j0o2nYE3tJWs)gUDD6x;AtdlA2_jFqPOE8_EkJ(|pCY?oMTfd+F@ z;cL7A{%?2TLK>vq59(Dh{CXy3ZLH8_k;KD03R&{c%#YjJV9`)efw|k*Hl__Ye(j39 zBwe0fosd}tJIBw}U_;hFhvzwLw%%eeI{f+XH(J&81Gq^X>O61cJl+s|Y&9k*AFqLn z_&%FCnAd@t=_v=_2?jL2CW9UTfCGWlEo*{eLe4zkbhkEJ1UAMGiA{i`E8Ac#`$2RM zJ_d)&j7bw;XD)XvLO%JI3!!Fu*5!`B)?x-Se+HRGzF`ahqfJU{@Q)JJEPFX@Y&AKl zlPy?H5m}_~@rr`Nnk!qV7b23VV_Y5~iJx8a3AIgTQ~hAp*u;TZ%lAbwovp8r+29$K zn17`7LsXP#;4TOF`yKbB+R7~Ro=l4OqF@X(oCmoIQNvnO8v?eQEvzVOsVJtq9-mj| z`OsTTtsgltV;JAPYrgR}MaVF5@|@Yl|0YhZ7ivi+F8ZV*zqiDQxB4Z<0qvMl6vhV+ zbI@rOZxNsFG%izgU>0&?s;0V zvb!OlRBil1$5W866YAqrl^|GskU;I0z6)Y^ zL1b?{wrjvL+qs_}_pk4jrx5Oce3odMl6Vzz1Et=XjXnnfh!3_Ipxh&jx9Oaw=sWbkolXRk)G2Ka$pHNMg zxkJlymVYYD2Y`mTBBQ=`rwWEdE2^1JFK3#7RdV|gxNYz8W1z7C+Gen~j+1vcD{>MS z@e#G%8$*Xx?LsnHYq!DSS?vYQU;@PKiC=TJ-k4VgV9E+?B-=0uXhrgCoP1B&JCc;Y zW61nr|9+nIthwWNzh%~iVgCd6DUWiB-MA|Icf?%%D}Vn@r9%hUahjXNV8`MGh-FlW zXlrNuH_44LfVP3;WMIs`6$&LUEPvGUy0w?NU`eX)*BVcu1w%#T3kV|<=!sX=N*{W2 zv^WedsBGI?%72D`k%XeT)Q!tY|BG46WG&{ez-cllqfY)7-fU62h^=K*@K9l-l5S_i z4gk6ny7S|6^{Oyi;K=j16c|=Q#LP9#)9n27TB5(B`JaBPvk*n4?97sZ z~dqyMm>yJ9+<#b*MZA-qy*80w`6e-7YLJv$L;oj>d7k9Sx+S^ONt8P^Bp4d|?d@!( zt&z|M#P@dX3}+g$8@s>=JKi@}&L9n4PNI??vbgVv@;%+6(@K3GbuG{>@Z>*PXi-m= zoUQSf{U*{9-P%xwYdHdlv%Gj2iza5$8XmY9u%&AJ(czI4_-;yRjHv^{OF$~pGfxpEIIHKS2$|w{@-(EU@O%7)u zQKsjkF1N%|Ag$Btl7W$4+p@3{Pkf*B1_pI_!31`pBfq~cEjZR214M<@UGp?t#4i#A z?!!j7wcb;-d&~rh)FXhVXl8`AME?fRF5gp9WeE^5%FtNGVUIKAfE0 z1zzM^!>ODdergbd#2OSSh+{@dVss(~iE+DoYVxP?!8N#(wYBd7nUv0=bqc#IcAUWe z@+>oPub~HSF=qS0xO3w6r(_39BBx{R=bxXjc2ASK%f7xy(FMBry`_j){@iIk`W&C{ zzUrLh`Ao<%z_3ybO5_Rq%F%gY23SryP>oN{YZ4Y0ful{(+^;o0N4zEf09rgKOfBH| ze;SVe;3$+07rCXb=&52xeh6$P8eFucWD9a zz6udfv5aP;t@x|RpZ{a}aq-(jlclbYWMFDY;~JroW;{?q6^JN&(Zsb(wW5H65IjW; zSk}0YyTGAo+Sdk9RpNj6 z9!AG0JK?q};%t?PKf$AAq(b$?Py?{rq`YHSSa=1f2V4xuB0qLj0r69Yv|w1yH+Q4! zrVI5`o{-4~9Bn<1o9j>fDz}pdcLYD7%5Ty?ky=_73fgV46Hb!$g9CCsqZ-Gw`ZpvW zPA|b4d7?uec{PBz-A@{z($-UGuN3N?jiqB~_?U48AtMpHq_H6GC;#9{F8%i(xT?9P z#8y4W?uK^0FmXl6rrfbhj`n)xC~W(LrYnCUZ`e1){ZXq9Pix>W08@W2m4{$phi*&a zrD-ofkR*Y775yUVDRxw)JuzJ#AWQga=F>ga-qcuDmBHI%0i7Z+mx{r6HAjwCo*ED{ zE^Ib2PKM|ywZu$!H78Vn8B7n>DR0pQT-`*U%BCublg9#l<`ZUYl}sBW71C#!_3~xY z%!!Qt{{HXlFl3gS@Y#gqRN`Y(m2I?M$vuVr;B5HB#1FIfN4EX}+_Q1J{$QE6C3SVS z%up#r*h5?>9ms3L9?d}JLJoNl`q}8prr4H)0IX53IlzW$f(Pi98U?OxMaAsE!;E0) zqWCEFlIK~nP`yh!(SW7M-Jn$l$i71&|KZPvZ=XRDL%_{(%Ii~~&8BCSGb*-vMX~27 zH^DtIz`S+VoM)mfNG6e1LH@jWdl;QK{YlT~LKvp6?4r=M!!{n0#>Y=uBnT|nuY*yMFKXvWzlyx)* zBi^!FI%b>;$di*AQKT?`ChaMU|2j#6o{4Gwhv3(&DnTy)%4hsxC?%uWmbf>IdwND3 z30nY$%`UI7GAAv<_})$THU>Mx$_8f)xM3iPFh(oF~(4-wj^q%;z~ zgW;1K6Ept&x_~oUz}z*Z|A8HQ+Y{zTRZg1Kl1p%P-@s&L#l-uY2V?_&mPA@uSO^yp z{Jm6Gv3wb$^TSj~=0Qe!PeaB{GYlgt_uXiP%k9j?Jt0Prq|U8(X}^tEudEN6OtuF7}D$U&Z!{f@~f6LbvR4Miv~i~$H?4Sr>G4bDo0k| z?>C=^YL3X#juS|U{_a?~BY&0zUqAhjA~JI9i=&>b1qY7=8FIz{EdCPfPW;IWbt6$?Vt6`shEHc1EWh*v z5Sfx;^^c^Dp=Uhm7?ZUSjyMyGODc2!p)(viz|Qoy*(J3M;ySwG!Sw@M%D=PW&i=Y~ z;O!}+ON;mw=Q^Lx-4WSj&a)Dsdc&{S*H-uuMX6Jc=HOiqgj^Qhj+X^W)8Nf{5{>8p zkfe$z<HCsIrV=S4vZIS9{*`* z&q<&Zemu%o0Cyo?)0NL=*qMejyPfmIgP}I#i+*;mPo;c9bJvSR?>vo#k!F2;r0o-7 z-}#~5l@+esUS@U$eiz1+c+?%@DI|e}eG7MoDGMjRd2_}1yT+8G^u-wv+7qUflznXV z`3+&Qj=HwbBMRpB=^{R%jKx0S#jw%q#DXyL-Pv@$H+aN8suxhN;snFkp`cDaCRW)O z{f6mJY(U?B@o*}86+4haITEytLQ!JCf%?xGxNKgf)5nl3ihFT{f%HzN%&z^Newd8+ z=jKrQEw!8dFy)oe)N~owC3*r*Ow^#!<2#DJ-zt1|IpWhkgeNSKu`y%-p=TdZqVbF7 zx7RocC2J-B2y|4d-`zOoa->lsrHYfuZ<&-lYN)0I^dvB^9%GzM8!5EWutjgWTSdNU;jtZ^l3KcNxTP zKF6#bbbObuRXM|h%j^1@5y1Q2TeQXmc9b&o7AV&UdpeqhMcs!EVNdN zujr{%Km5kVrxG3bY#jy{6YqOpBTv}(1`=Dr!v9luT&EH+rUe%S38>2xZuSS5$Yo+p zfaIYJxT>YP-vR4Gav6OOfNOuW z1z!h9_@lXAXIxYVBOcA{FZJzHQ>u`p7T9NkUR5|*2tYCKV7Q?S?goSk*gnW0vWK;h zK=^do%<(O%VSy2CWrPp>Omci>LP5^f8q{l+!QR{15=f6S@unj0_1iJa^(XXx(xL-?>W@G^|B~7k)j6=$5G+lIG zIr|?+^KHPe;HfqRe?>atSws+A+wrgS=FY*OfI3zGpe=Z8Xe73|pf+ znxbN%*=_+gS)DKQ3gAz%SQ$L(UR@rQr@Qx>j~8qN7WEl59!N z?7;ArRbrkF{KP1IRh&6%LOv&Q@4suWnVB_9@F&#YQvr#c_MzBMPtp2#RStJ5^{$43 zs!Bnzje%{VkW9_el_6XVn=x^NE9JqZ(~y{ z!AH8krtnhOge8CAEp7y9XiqAIrgFWDR@s#@oATs~886aPb=}`_P#f2C2pF0Sf?YD} z?@d#V$^CHBcvgPoJhtsQTlF&u@AVCH(1Is3tJaH2>^Qa4ee?u`W2ZeuNr)qnE!LGE zko!VG#T5De7-FKmHO>Ki_M~4Z5kwn-Ls-2J8WrivpMfXBn1&7`2 zw?yS##^l+3IK*YqkU;b$Wb@#L8JCazG|=P{YUT`G=y)Is zzU_CjMrla4VuyY_xLXoFx?D>de}8&h-_VxBgL~P-5}a*2ln%MQJyrB`Ro4Xk}6y%+E}g}*)6*J?M5I1Zgci)WgkteB3Wxt$BU zn_Ha6Y_{747;%hA?8>`=0^jxAlsTi&xm=s7(B=yZ<&RtvU@pvDJ)ML9R1m@Mb(RBX zh*%0lp3BlZ`y2(;3?B)pTg%#tos27Mz`x9Udxc81CyWK9)H74Ja!j3RD*iW(w54BR z_nwL?JW!K9kxL%->Da{$pC?ub<3BH4jK~;$Z;f+z3rl96aU6NLqBfn$ne=oQQxmx1 zjI^-^dX{}xz={X>0Ukp4nJ%RM5E6Jmv4HrnZj3r9mT+~YpVVkbJ1hGh9Tu;84YW!m zg=la?9-6T+9>7=d!(UlrcSudxCW}kklAdQy2>fX^VPE6~O(sXr%}3TNg3&DeSwutC zC#z|T&Rz~eB8tqp@>t_O#Zy+P{v!N2CNp8ADt6!+7STIr)A$x-RIVFGX*=EYu;ASv z!8Fj>$sf4yk+(UPA9lNcaK9pI(Q&gces$R|t=Lu(#V?Eu!~te{NsFLSA^L8abcZR)z60hT2(!MMFm9yf5ax-sD>-F-$k(h0CS51nr zk?>Ww{Tr^xd=si(#~r!n`z>jQx5|w-!VNs}9`b0er(lgpE37=OOL>8LNe zX}_8B1C>u_r4!%niS&as$IvFh{j$gH(m~J)yb>97k@LvZXVGKx>^4JEYsBllj#^2a zcriY!$*l0$mDF8jL+m|N8A_SQN!H4Fw&XcS!ng$NsnBV7>6)bARe`U^(`|u6V zBth8KB%atBMW$NGQe`rCo0g)UXuy4AD?{!qF8`TktaLy7w;tbzn~-fdXJ+p&fb|_< zk-nPOA>Kef+!@T)I^gVi$IY~#pLq~BTGYS!76Wn365JAz*x|DXvvYhGNyQ*7P~Z4wO_ zCGCZ@6GBHzD@CcBL83FA3zuu>hle~D$Pn&-FK1JCDGDcgJKFvI7Kbchl^w#ECSaFG z`r+guf;;$#+b##s(@3N3?2UN#38aQ6GLv;GqGHs_!lE$HGIC|NeC8d5!Lck96&hZ{ z(d|XeF~lBFiK8|Bg4C@Baa>`w+J0a|aw@|0m%2tZ8BX9?%&zK@7&CTr%eLqJmL{mI zYd33^I}q7hPnHKN>%ZFfY!)hO-}E$Og<@`@(Q6Uhfk*I4OWrZ0^xNQ8FSHNI<&B%vlr4pZL$`wp<9+N_szf|*fo-zsZ=U{h~6n~;|Y36ZwA>_Yi1qL+Y7jJvRq zu?HfNC;@om4UPv$Of^2JgAiSfk27b+3LCS<6TRWN&c5Sy1|q9MlPTiG{gz!1eMZ^>gos| z@u&(3doKAfkDCuVaX-n#S0K$Qbzc|yHOEjTY;AADIJ;T#HBi-LYSknW52Zrxtk)k<<|ti5 zR?7PAx0N>^-@R8Z1(Bu%w;}kX_ncm?F5vAc;}>?337Xq&)!`ihd3Krqk+etbBcaA)wC#{c096{K3|JZ zO#CqKaSRe;Q1*Tm=H=X18nNP~#{X+Ya*BJ86nP$m^kea{=CkyH^{+(HSSsl%WCk?sxBv?(U)a;=*n+5 zFpHbN$fKswphOc_zsJJJ+yA@|mc)zYXXMtHxSQ%0eq`bLEF9svHm>+RIz2sI6%kTO zl;}I(6XzO!!u5y?TmYA}Q-Z0m!4O*4H^RdM!=AUz!LGVE_c{*FPAURO>~?@z-`KA6e$D#lYO%*0L*Lg*&!+Rdddqe-FL_T)5K{ z`SF?7jdZeV_YI2n`?JcEe1f~DjEs!8Q<`@i-2UULq_-5P*!+C@zz+TUhJzV$3b6p9 z%J=)ektDV+Nd&EoXetSh$CHAA6mP)owl8Wkh$AOHKK8lbkSFo->VSu*v|Ih_v}xh1 zH+(RaGNWCWQLjS*6VHI$MRo_ro4O{=F+s2psja_xkHw2DlqLDf4b9p0)kRHN=4-G% zeYg#3WN~+)DQey|K)64NwuR0qrbZTsJbT^^&29}Q8L5iyLXGj!?hGIRE(u)5ke1if z470e5u*he9WXJd+54$(JknnpyM6PXr^jl~)UUUD|2lMLZYXe_RUdak;cX?6cb{l^1 zi2JYR6|Q>57f}GL5Pq;fDnaEM{BAu})YOw`C%jUPq(`YkQ_es$7^v85r z<0-$(oejXpQ7tYtnf+IC%xleDa$Dx8t8gRrh!bjDgn%q}q`g+^u%C-Z{a5I^u3N^o z+^}7C_z?Fh;L+NRUyjCy#83Nj?^Rk}2jTFOv;S5eDy-eewdBa{#EM7a<*B9_ zhG!ybJ9V+H=l7bBxZF>mt-8SA0W#bgM#RorTP~gHf@cOIg_;f2eA#LJC!L;h2ThGw z>QTM9xMWUgDr`lmI7`PLuTs|EsNK$!^`CGIr7OFC`u7b_Qc^av_)D=uKLU6~t^xNC zG{&5n7ROUEg6MsR5HfYvBK(*;8j%P^Zw-h0kvgJ)a*@I?W&XLhCG3{DQU!CVE$U-3BoEIJiUbG-L(H0N( z*Ecsra);g0f%A=(Aseqlx-;5OGiK|Ba{LfZ=KZfNu6A(TUG#t@Fnt6jM9tqGb3aHN z1Rz6Jin)Vc;qYR?q6<*36b8c0S8fAwSPGi)$s$n?flnN~)l=$_C8=>Jf2q9SC7UM2 z$pJso>@6H9{E{Dweo*$yNSbs*+J}5FAJ%`tTT%Cnz~!1}JugIi`D^qrQn)n+C4uTY z=C!yQL{_;W&wd!yoQIlD>iX(e@R&O`o{StEqC4&_E1K*qHfr=BH@D!LSsVkEG^u$1 z2h&PzN2FXH>(2G-k(Y72W{R$y705 zAaxHQj%k#7aE<%$H@9@?%-4&0?-Kscj7gxb8b%0rBm=M0qADT#-_-`71gX=jfRST2 zwNIcmmDYc^8y5GX7#KrX-5&w#w43ei?KU1V-^bN&Qfq!C-OVOx9?u<_$ez$6_TMrU zPg8J1a~oB@6-(%~nP1>ZQIm=jPcofXC1>zhN+XntkxbPm60E- z9ZMID6}kl%zVy^;k8|l3nj2hxgBEmg(9a$>090DYDsO*X;qvEen*!=UKMuzwX3)=- z$zA%{IyAfYUNZ~`T?CpEM-0EOJr%6)4@ZUQ0U{Ynv{?p)d z`;b@x^*pUevYJj}EiMteC28rvOG4SPm1cBBUKEE->%g1*N%;H7mFAIxeHWhw%aQU( zoFWI67Re=pn%@msh#jPr6F8sJMu?BkSRd6}ZBQAMi^Mcj;dO(;f3P$;Om2!=bO%=3 zkDC9>0$5|KT(Y0hG$skRuBZd|;XCrDfR5r{XgM75xo1(ET%WBj{9)d<`I8ZBNU>=u z^Zw^8aMXG$ZwI-LkoWVIhj7$F)fZ$w9m4AGyz8@Y%f@A5>P5axi|AKx&n-$P9gWi! z3acAA!(5Z+rid7I9O&%j7hoVt97QM@2J&vd;Sv>cuXKHV&8wb5d%rS&IaL~*tyq}R z(6+aCdzJTqt&~V*hk(^lNd-m=w{0VE+P(NGIn>*@8E@C5gVT`LTi+To4BEfkRD$&yZ#V~%v#18Z)gQ`vQgTL{7oB-| z_r3VaH?Q=2ab%$+yjt-D>*>CGqC~gE;1SvkJgwl{HQHW?X&@3cmj4x--Qm$Ak~;7k zSzRT+gMg4&03OIYY3z8osXPP_P99)dP956s4vlPlH*+{FPgL_+Sb!|FV@k;N|Hg>TlAW>%NV8eFXUm7YB$v)A zd`B8*zx`#37(@&+ZbIrS~&b|5eC^F=F4+T&(Vx-+BWHblh#bO6UICW8+2Zv7#J z)%Hv3Z=`N42}|y#+r#-#_{uJB5Q`j(q&GPQ1wwx8#hHJ?{r!@=JmB8O$L<9IyI+Ik zgBag?obO^G>~O~PJVQbV+7B#YlbXgYu9|VGP^}lhToDJMv%32E9Cymn%F1U*pV6%6 z7V!z$$Xmoi(nb44(CzP}*0}B{0(Vpn^3_EqP{B!CP6NsG9+2Ki-$peYb`sq0JZY*| zeMuJ2>ejM%er#R*A{Ui+l6eq_!uQ@^(EGc0Fy|o%kl@1HQ%vTotV_x;0Rn2VE+9~~ z{N^KlV^CTw@wmIYFh0?dKhA%#w6wJS{RGw6QwooSfE>I|17019wW|VjjGt?Pjn*~+ zUINQyXXj+#CV$@QQosQ{n2&2667W4CoBbbL3fL5reRP?x_t+YfH7@VHx|1ZS`XV-X z(RLsSmke+v?Gr~{ak(~aC_Dm!-%-v%V4TIn83C{XhUiV!S*+pP zg4@}CDc-%AYy>my&1}?VQ{GK4_(JyP<_5SG-d`bsDjq0OIo6rXNy(|G*2zMS44PIV zQ5u&j^>jsZi(71=v$Zb`^Fw~@tB{$ zz}kBD8aguSoMw*dD(@J-T)1h%MIQf35MJu}gmE(@JOVgB{yNF^If-cQvB&-QZ`I_W zJV(A#)up18zUDq(`Lv{Tck`0s1pvhzCx6(}r`N!G{BC!Lds=5VNxqcM8>K^lO)4+X zxKVe0aRKyWjfAe<4YNE%l#!^F9(zW#Gn-ItM`TYe+6(OHd^Ek~B?we;9&V?!>s=sk~Vy7lei{w|5G*rq!&0j1UkR9BS`*Oixx$UADO`y6|8gM{E7Q^YcqbKXU!0*E~~`q?&2zF{Lj&q!Cflx5!e^Ao8Yd z^pH{dxB@58YZMc%3NtFXoXEI`&a~}&WtMoD>n(Y{S-)8dbs}f65`)5VJ*FK=s~bmc zpsiHZ>q?@J4NR+@^pgU1tK|ca;_`qYeig09?5^!s=?d*X?VtB$d-p(wm86a-sicGB z4LYvt$osC+{KQn-#CO@0zv`nEzPS#gt;oHG(?Bbx7V7i$TYD~Q?0wR-&?(`P0>Phe zprZn&&^Myl+gED7Lngt|i!B$xLq6~uOCDL-K;f-0U5!kKXU9PB{v;Um-%)hO_uI9s70;lww z8!{gO5=-vDWSgDCp(sW!MCTylmoqb}2K*(!+uFUfaC}^AzwLtS zez7B9+RrvGtlrCmigLdGvvw-95ClHXAKZ=hZO!C`|c25jF6j#2hWkU;8m93cuKcGnmzAq zV7P!?Mf)`@?n50a3YZ(l@J=Y@WR7`EG63RQEGps~2}Jw01%GroP*;A5QU@SGjX^Js zA_wE=C1e;LdjX_G)Z77cLw-vqt}BV+`^!LaLXmep)!(#N?{m_IyeUy=8rgoD z=Z~jlu-R^5?d+Ti0`nk_+-`i@(2KiEZBcYsmn*cvI6ypcs}8=KP&9p+!KsydupTIM zQ>DsEClYimAbmt4-3Ppse4Hxr@oh~!*0B(AEFw|Y|a<`*!@D$dGXO7Fi#V&;Av^TSoPcI`Tn%shvF&&?i^lTnudt9w6*ZT1ZtAh_eGK*=Z% z{AXIOPD>x2DJ20T$TX=@iwF0kkAQr$%9L{dfOh#JUBA_QI8E{eHjALE`+cI%?W2%$dy6gz9DHFp8dSlA-FN#odbpGYafB~fa}Y3M z`#{B7x|@LmmYT1sMF+*KG16Y)RkZ9!YKI<;KHN@&naEXJP?MJK6To3Co};q7Ij8oh z=gp30*dG6>HX*=S2ZYy=Fi?iOE7wa?AuYm;*FZ9X9Hd%n3#eQHlA#7Ltqzqns&ZRHdcslSjRr9-s$Se@d>sEKa7JGDsLL%S>EESNXDwt$-?_cisQvO2LivHBWpks-oT?#-$`20k9eIPG z@wmEvJY#pEG=omyk3?|?t`H1jrT$NKaec}Luz*OPeD0IbcMdy2SypT&e!Oqrzry2nDnVi*YJYot1B;i5NQ1qB$*tff^$&d>~N~ zn4Z8dQQ|#fozC-8z`-tSSvcf`QQ~}-$hG5+PsRl;J3C7}_fj>r~SVh{$i5g-f6oXgDQ*U9tTC;9Go&-$MCe#dIt z9lM?#k2A<9xwCAwdO}Ggr$+vc)p<*85Ip>LO_JOxUtp-S2rq8t=x0JKA-+?Uk6hI? zpI)-UWX~z*2hTOh!l&R&&E*>OXB`}sd14kt=yT2n=JLEIB@g5I9e~s*%#-y7O4ICKmsnH zbb82^R{%zhJ})MBzgs?^Fo~u`CXz?tkcs|w%aV_lP&A${S(hp=Jx0^^BZeo!UbiFBcIAB+a?;b+^z8Wg*{ta|B{L6@ z_aI8;R#EYWSZL)STj;JS9d$)kRJz!uk+I}C7$C6hwn#92ef4(V zH*<41q41f5p|jJEU>I<@2mLB%3?blFi^}bNjN*!X4cU8p&kh)&fq}NY0`LRBva=e$ zuJ0x%N2rz}&6Lt&-YuP_+u2K1fJIjx-0w4$~_S*p7KGo<`Ss|8mhG>13*eY-i~5qXVX5ovN^=_TX2o;vu-$&ByQ&b zEkflXP+#QI=+Xr29ZxKU2zARpZrjW*bFW`dsV${N0p{1{a(c;ztf&(#9KHIzvX1K&BvIyBD!4_RC2N_v>|Xo!~;IyHr{Y%89E zJ>ody`xr9gmsrw66JNBlB4eKk!QgW#pZ@DESqdS#7eYQ#{x%ZaQb21~RJAP|1s2Er z{P<;)g3W^)zQaT@MO0(rS+7>w7iw^dfe;rJT=g}vYF`(G0$f<&lZumhtfjF`jk<(= zKAXIE(->3hW9oDXqx+rL^Sff=XZ7f4t zArle~e)0X5+;B&!R!Wy-cBI+w&3toU`c*k&F{&SkRQ+Bz+Jgb!MIr+v0sHN9Mp3TA z@cjGrIrDR0KSJKjXAl&K55p@3vL zGKTCi%xcN$m~Npc{7_4Q)4AC-D<~7Jo-yVz^ZIpcdV0a0VjOa70Jb6-43fvzE39*F zv((&}UB1Q~dEZy7y++W8b>W{yYb&4o=)7?ov4MOXWgtb0&s1l!m)QEnCgn?AJ$o~i zr6uV#JWDIT+A#sHzh@ISbI-YPqnvbSH)K3EzVDckfgcUj! zNj>4hF2^$#FPG^59UsJ9n}=B$z{opK+n`vrAKN5iGaG&B4m{Vg?8`{j;p58jM_!?M zZLA_5oFmC5A^Pu9vMolbe)wLj_-_wd88iKHKLDIX zej+Ay$DTv&(%%mK%Vij_(Pd9iSre}j`k}eoqf({N+tEFjz&9IMy#3IF#A7XuCc}#o^m1^PuPYL^jpwGZwS_K zs6*ow<&5efmg0bvTf=d%|ChjGQPKZ+(KcQKD%zn=t_(y;F1k=uT)(RDAnve1Zxfu0 hTimKuq2k-i7VD9U!$egq>6{gOLJtvx?;bdw_aC(Oc9sAD literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter4_116_0.png b/doc/LectureNotes/_build/jupyter_execute/chapter4_116_0.png new file mode 100644 index 0000000000000000000000000000000000000000..dfd44cb864c330522f1b312d0828240e042880fa GIT binary patch literal 28657 zcmZsD1yodR6z!-;cXy7oATYwvodVJVQW8UVw{(MwN(_yFfV4<=sg!gM2na}nbia@P z#d>e8xBicT|IVE|U!Jq~KIa&tp{|IBO^pqKK=726`1uiUI4YOg$89GpEIY|R?jpRrj{oNd9%nae z-tQ@O8F$ZeQ8I9cKyXOze(piiGbkYtXA@;PX&s-8ofh9O+_t`pw`FCer-8yK>$c05 z_JJq#xQ8`ZECh*|OxWF6%I)k#lp3F?xJhuH$ud{Uzv;!dQR2oC?fk*sS*_X5*bg=9 zrzRILGTV*cNo&|@INg4>cWI*ACb>|uSw0r{(f@h2yPvOwR&mLcc2VQyb?fk9>)zoi zOl?ex5KWq}Tca!c1ncfEG}kB4>AT;2_urV2+`WY<6{>dk4hxtNa9uhc^1pA0RtyZA zn9vnFop-NyTgCoXSGRO}c`&a1+#qYr2DZOlGt5jL_sZ9ois5p)ud=@0XI%2XCy{i$ zeJGkeyn@;b5?VMP^1csQxZd?%iDAoXr&&BxI20b1ycJtMfJ$g`7(BvP1ZBc?qyaONF7`e{*=g*&8AQ0$dV`I^MD}+5Ef_>eT{FTn% zlY*}-j9UE2hQ$t5hb?k7Fh8jDG#m4z5Mf8)2MimPi+moVdC5#fpjA9-o89}?rq*zJ zVWG3PR|NFr==$!ps--1HtN8YoO_|udisNd2xaDEhG%5H~j#*PKUp0s2Eqk*G13dA; z6Z)ymPdWCU@P*^@(`8wP3qq}8V#6}>(G60qV(7acbf2_1wznOa(&jr(P6WNYzExHK zGhaLo5!uPY+I`o>;D$X zl0Rl}P>q|1M}}0s9_1u<&`siVzM66{ZIQGD7P8)BV>D#sDW3_98qdMui$!3d+M?dz zN@k|x5tprV-5gO8+kUT*M7)n!qK{YzXvM|w!oubK%Ll)G4gX5vV!VTxdJ4L>4-z@mZQB8@aOTc4-4#4B9kw4j51+uO8j>+8oW zz5~OJ7d&c4C6;%NK47Pfsc?uKu%`;h<2kf7m{*`?pAB2)mP84`+3y z39KJo8`Pux;~u(R06PSk*61A_b)0W-r}5sM@5~gftNQ9eGg0EXoISSDS@0~yb#1_a zn>&Nge^*=EmHN5FD;-_=CX_-GYV>rcd=8tm08d^sr-YMKo-avPH=?lk7felZ(l3%r zj9yfwK)t)yRm>-AYu5d@@4e7=Swgvev5G|m5*kA&uPphIF*b7!wjko;lO?q7)x7%H zxmHZCOs}f(XSG?c*Xb_Z6>701{)I^c!q-VGtG&UU!ew{fH$&9>-uiI1A~TFZEHL%%GMl&ZTaIb0QQF~G0GRxfWEQCrq$|+y3 zG*7QI*RTw{=8b&LH>m(uaz~z=;K8Up!2yCH@%mdQ>~d!u@=#9TH5cWS`SaF|6rLD1 zhzNtuFe(#6l|u(vWj{Jzm6Pq=m{3C*0Q_K4=IiojvsRKWQ7bOL%uV1gBxy+Jjp)Ry)-{+g_s39}kqPH%^rakEGwKj2d9gj z9LGiwsiAuuQclq4TKKxjhzFj@Q*Q*CQ-9Um&WMoLwhNq9#O>=G9WYGTJjRXY4U);q z$mPIG!MU3yh8z0eQ}2q@VIp}VFacdXaVf4FqgFXNh`Lmbhe;ay0GMG4;ofyT2#PkO(g)+YQ0=L3PD(!h>fiJS+cJg?|71AY|@( z5N|+;ATA$c%rRKD7+Aa-tUKt$k!;@r105Xn?6+8QK}+E=p7if{;Q}TnTxLmIWGq=x zR~L-#kDI}tL*cJ;Prp<+vt$o%(2#6|$)&FQw(y;{uN-vlcy+CA%GvcUYn;YTf(&AxgWkCdB1V~5%>wI*K^Og-c5hQKlH1n;M3Cm%OdrIc*O_A+k9(v zL0eM?VGLIg`Xq`M<<0j}cxmRy!n;zF8C++&u94 zeR#ffeBpOuQbA!-K>$9gUgZ9%=Tw@~sw0RH>^hEx0Nx~1hLTmAan`Y%%ly@f){ zE2}&s8UBU@OE<;E_!Y}Vb3(CU7D7yBQf%awmTYci!+Agurc8zHz-QCag zyIN|u23-EMuW8rHMWTHZb_IL>=hydI$b)bJZpjGF%)b$Q&j zZceG-9RNPvLqme!>I@#QW@eVPjzkY4WMR~xyH8n&zR3fB#fW({@hFl)&_46S^p0Y! zAzy-wmvfzSO^h+hg*FOIKR^=ZRrL*Pes-MrNqwSTGf(p*$2W-{6inEe`}@VSd+Kt8{!1Q>PlVPx=jUaHEd~+V-dCHJ znOOe`Xym zJ{vg&1>e&Tm2wzME}xt?(-~Q*kb{WyrR_ps6Zv%_`=?s`pXnPmI{KD(Ep79U@o|e< zf4=&QDI2OrtJGpLmVlmmI5|6u4d8@7r=xJDdWN_^JpgIT2exB#-t*SI#NEoEua)&a z&e73L#M+d49g-XCK6rS zH#hkKD10`~9ZO6cgH~8%t;RJV7pmog=4z<)Zd{1dxz@1In5{pwYNvMU=3Yh13BlT- z7l zLHXLF%QeF{TeH){n?pNsg$Z-gt>_C{NluC)S>w-7*a$4${787y5ZkA5`4Bp-qbeIm6%8mX&rp&!5fGU%gv8$Sq7kPELfcSgais`^M}! zW=TmpN45WHs+6-+`Ah6LIITJ`<^`oGEw7crgL0w@=6-jvr`W!q_%dantY+a%89<3( zh~=+8E{@cvUjA_?2K|9{n6ANv|2&`9r^ld^p zNxC@h2hsMDb~_2-N{d*TSsAca?S=KP?iRBw*uXsQ?O9HF#ntz(jv8MKnYkX|l2iP6 zcL)XNP&>9Kk7wvvy}vhBf4}!*`}9Z3qyB>q$ZS(N7&VLQuFMz=V1=3ts{*Y_JaJCr zHTvy(y;4IJe)@2z@*68EuO>Q)Y*Py6nRxDmz`7q%L>~-KTM!IkXSf(u2)!|#xIt?+ z7PCR(%CZ=R{A2^O%g6$4!0Z6Cdi&#(KmUFISCnCujk`Q!w4dDP%2Lv;kea7{#BZ%; zV7Giv z-&PL7&+`q9#KPpv{qll!zK}jDePql@rpc2sF!yB=M)Eg|*;JbFw};5mDaP^Va*InF zIsfat4(0k2+CJHLXCV|<0M>b5Yz`W254lhlGOoHc^kEh;fKfWtVeq8PePP)1^4|RJ zFJ@A2R2~yEdl}=?VAT1KpTwDkz25Iw@$Ltr|FjN(VQ`>K7Kb`k-Lfa!p6F>5)DW!? z-)#F{o%Kn6BV}+E$@Fth@zv+$&IlRX5b3EAGk~edwwTD~zL|QaR89RWLE&Ms4mxlL zwX+Kd0!Gbyj(iy3nqfRbadNWpyPgw8mS3-=F-IRQ-YK9 zRT4bp^@E1_C9!lxJmANQ-#^ATnsJ!@+P|Gr{t;lyCx{Vs>%G1-0Ri@uISVQxf^^sT zN{oa&cJ`EKzw24`^aKYT603MW#0HwKyTH{HD~8Aw7zt_5{g(W5OE>{d=1;qMIv#@$ zzKL(;iL!z&8NeQ&#%wM52WannBy#X5;l0&Nq74k8z>Ok`AKRGKev_%r#z{uZWc~y* z6p}rhq}2VOLTL8TrwKLrP|O|?8t>@uU(?q{4{0`86tyGl!9U&k-!asZn9B^v3^@yT3Y}p0Qp~4zTuSu%5gt{9CsasRE;sC0j z5@WY_!-JW2_uH~%AAZ!~ryDwROvVn@#0b0EP)j$A;_*aVvo}9J{c@^%JS+JxysN#rT_EC4Ed7)GpBW?@ZEX=N5y3yu&$>s0 z-9KK{iKMfMW{drwD<^gqT@D zYi8ni*J|z%>wh$5-brXEyUp_C*UACfwpSlVCf`Vc<{~mJxU#Hip{~SEL$@cur0zVb z%2S(DSc987jikH}@@Wq9lfOkIkEmq8}R* zi0(@ZEbLnm?-8Kqg{cYH+j=*-^SvnFHJtl$A~@%l*^v4pl3(NSjWqCkwhHFsVnQdq zuFjbrM0xBTh-5V(US)F{;T~~H+9{#~q1J;}kby@$A5_vOkx1m0xWzI6>332{S{jf- zXlMRrG3&ssx`iD=;tb1Tfh+*}WEL1mHaGW@T(jz544}m}& zGc0f${5kF*T(+wDPD_dCUBe*S)8A+g2O3rQtMqP%bl4023r1FJ=$PVI z=xQWirng}rw1C&=S~*7t7(o+rqxyU6ni-~kZfsKd7_As%7g_?%KS?lc16|!`==l~s zGc)n=ie-ATRr=GX*V|rQ&yWVrUfjZI8~xB5Cm)k1)F)F}>3P8^*d0TDNE$crHTM!Q z_Lq?l8{b+{In~Y08}Xzxc#CAZH1tjt+h=H!KAhm}|M;Y+Ja*UHw9DX}KQ-4U)1nGc`GT97yhN*5)XJ&BgQ$d~NAT)YiByAJwu==;_1ad=nD zQ;(QSHsY9gW9YcuQ%&|ipm~NME>UC?5OR*FF3lPu7&A>uKT1Hr=Uu*nxr+ym2v?Q; zwaHUzy!ZX`;{DoB#J}Q1(7H7Y0GV2i==A*RLIP4ZCoZ@(C7d`_TB5dAInPM-o$Ttz z^Po`Yz^#``1zKlel4lPB&bkDeyGpqYbzdJ^8p;&)IhnC$ z>erd_8XbY*t-4{I2ard|;GYN3gVzbW3hQ#9BaZ9bH{rD(#|(DBH zBioa?y17X!;c5pV8jk#qe?9>ZsFE2z*gwVnRVyN<*rJzSn5$R1NmbCCW&AT&pZiEP z$h}Ft@zh`GG>jl2K6(rg;EW4OK|bH{t*4eX1I$m5YAh-J`mwx_*L^GUj8DR(7%{VX zb9j^T_@lcmJ6vfq)CFhw==TP;GumxE=AV*F%zTY+8X)2N}Faa z4sWYset>fH4PMn$B}@UYj21?hop(8Gr57H zKg>vp%S4I|w1E+Q0cl)F;>wXO0sUC!voRP@y)0{NEo-R6^FPQ6B0K2x`q8|a9^?{| z?7P5Kyy{r6u>`Al9kv^yJw#$WL?S#yY^$dK!105ErJQk@$B}H&BzIezd9Ab40K&cX zq2CI?ron-!-d;{YLA`sGn^vmLK!DE2j9?})=y4neAd4HwAm)5YI(B{Y)CrO2^-$&1 z^mM`Czadg@0Kw>f!-RRibBP@^$VpSV9v4E7J5xQq%^7{mM-jwSn6;zj^4^ot$S74} z>6DWF&yY#y(9pL6kyJozx3@#jS5D6ZENUb?be^d4r&iRO3-6v8my+MwcO9nFou0NM z_R2yUTn%p5P)%a9j)DWP;J%(G89{AJH$6LUH`NywG)}JuxaM34EfUQmyQY=X(p5QW z`_evK#~rF?!sr@0q>ox^ukjTfhZptCz3{&r#W40@#4j;LXx&la>U88*$A9<3 z(URZ6voC?B0yuU#X1;Fo0krOwgHcb(hXD3K1UU*`mzAkGI6E_fsbprvybn~0ZgW@A ze8eO!m=@SdN>V42^y0o3+R_xWmj!0Fd0Ak7WLF`tuxCX~ve)_xQsq!Fq1-ny$Z@52FI~KpS=wDD%cFiDSnL=f|IK9HXd-uB#hH01tp?23zL+ z<}8NVSff}P&6?{eqW;s)!(kjkc`WrJ)UqRr*Uyq@%|aB_+QZp8SXaND|o1D+^7&3WVWYEt+1nTGo zf*FPO%M{ZmD+)UE+vQ32pBCmOSC}{4X?KbXDS!U6d^_^(&R>u6l8GP%K3r)+j|0Tb zsvy#MvF4IBa5oazyX9qJpwNdE76V>5Jv-YzIT(o;yO-h%Zk|(v4gza66_51dvkLJcR>lHr+eFr@By@Ggy6J- zg_E4M1M`HB0`ZEWg4P%e2%a!xWVF!HO5}p3z0BA}(F)$dM=Mdw8GK~I4oM8*=thg^ z{bZ>YlP)kl-@Z)(!V;lZiuJl+#-a*N97EN-p_Zb-!w{rzqcu|jB?lqSN7O9Rv1@7ygDDE zxJ4QJ`Ju{L-_ehssFs*9gy*p%yn3fe3u06XMg<~yFz;`B6}1Y3MU>9|nrK$6)4jGP zvheStq6YVgR|Xar*b)O!*q7e!3d6Ec!!nkkrp#8Ktf_qDf)Yk*y!})5zlL;KO5I_W z)t_;E1_w6Bw{&611moc~v$`6a`6ZXM*4bnY*7qQ$7&hM+{Weq$zvS=zK{-RsFU3mA z)g3DbT}yxEcuDz4iXQg4A}~hQx8^HKus4s>XMR}vv;I8~=ncr?p7XqCU2g2v%2u5hcf@XQxetN=^dVE=0MMwrKt6X@JuJr4kgsae%Xv4I}ly& z)+}W-ax8y&@Ei7Ue#4^+k&nOeNw$pM)ts_;$F4RWG?jvQKa^b;M=TH~f?s@+URI5# zK#lk?65jvycsX;3EABhjcf`yVnl!8j=~y14R6s)a`4Ken1EMCxtsz>a05r>(EaAi& zm_)B1B9=K?xr=l86~y72XN!q^t>K7h1KE3(h;pq?TA;xKj|V7ohPn{eSt_83xOW5% zJOVt(tsz{mRA57aP|m029z1g({B`dJrn$3E)}BgHA0&rO)=KVA@%YD?dtMD%;VK2;mTn|s!&}Qc`HPFGAm?J90$>C{@$VZ;0Sq~n?M7Qvv_6ff6D8jqBkCB1HON<2mq?{Qk@MQE`6Dve-1~zdWAXtx&E7W(J-f`X}9E%zmL! zQL_iRtQXDpT|eCOv9AfNTRd56sf$StQB$$#f0RFhHoNnUV`0^;Bfd2u_i=>~q(Vrl zpEC;c0?JCsI!Z>0Eli4SFvb|R;^WieU^K9GtW6pi`lL!1$V4zTl;@Y7+0sHdO{f^V zUF~xFLMhAdW}amS=d+-iIR}f=1LNUG9g;j?h87t@7a92HLe2S{5tCpR1O-1XdHF9L zK2&awf;3)`pd(>=^0{w76R@W{Y=*Q+qJh37^*!4X#!yn!A^Ti{t*e_;qLM;Mb6Q$2 zIkF>H75U3^af;~Ykb8ns!%NJ|n7pYisjBI;1LuVHCKhjtisNl>Z`SEs-{8#H#Auwg z+wX^BJ|@3*)XuBE35#L@<#`m<9MH(6|u&)4O1A7c+yF;48CBZ~# zMR0ijC2e=jmtMkDgCY53B)kfvJKBd#KBVNEn)I#%1`c1}0U=uM^Y=4>NYSdlV=D(a zCX#2qr*XQ!3?g$xk-vDG+OK||o@xq)`|l;R&Ygw<7#GtG+K7YT5q=Q)D*v-a?QyFe=Q%A7Ce-RUn5 zg0eXGIns@wC(<81?mfv9u6p*q_<$OXwvR3~bSM7yr2wr$-%r9L;|2c0RDpXGOqikP zX`}1b@*j7i6-YoUT+x@vo;am)CyQBRi0qPn)`LDMg(mtzWvO93s|(~8?H^YJx2(JB z>K^p)1`+Y3;Ly;}oTc3UGhKeGNKXZ%hhJ%89|sHH3|r^gazffSJUaFWI@WPUyQ_>V zqZ49WL2kmsMk|vhr&N8VkLalB8dnQ*O0zDs&a91@p2CWnLd!BzNu8f9BO`ZMUblc}6@dLcxbZZ->_pDsGS4dI9})j6F(>F0{Q%bK2ab-t-idR(fZNcHd#G zhxnfc32I{isZ#*k_JXVq^n-Oa^2C+Onwl`b7W@jK*Bb2i zQ{sBx=;%r}p~fQ`3SJ<_G`0%s~`Fe zw7Ws~*k>6rN9Att6CMGp$^bxLgOS4Um6b;6BZvl}E2%p4`P)Inl9xB`JjzL^I`YSs z1loP|XlD7P%Ow}1aF9{~G8uNXg~{}Fzo7%I0JP0xky%H3Im?$Pe*radJX2<3p;syOvy1=b(iVD+i3Q9om2AVgV9{F>X5Qe`rW? z(Ls=7aBkHNSk(RYsa3a$6|u4^M&bj;51jXGXwT+drN(psfD#XNN*}4Y*LUy+N)>80 zmQVMC)*2sex|TEL)RbAyBe){^`YoRTi2}dDxPa(mcI-oYU0+48!vxoC%9G0)n5BrXb zV)#O_JV?sON&ccp!hB4|X{09!M+Z!nVU>mBbP#>$P^4x0+N8teFNevasr@zY7w=fr z+G8+YF@|0zXJ4T$UQ7Z?Qm~H^ypM6VA0-(td3z`c5=Vp%MB5HU&)9DNWdA;%GVKG< zbI-!vo#akRQiHQxn8V!4^d6MycDMh6aQN?Y^ZdJ|8L-gX0~N z2iN{?>{B!2OEH{&>q69wIol2}b+DqxGK~(hB2_J-VSTReH(8@GKKb8Ac8ySJ*$Hl~ zQTmkWg)=Hu&TLoC&_Wk1FhOT5;};~K&3MLnu*ne0(-bsfnM+kxISXxj3F+waiGEkD zfBAVrrO`*!R10W)c^+}@z~Tn4KYcd-xSxl5i(U(mcoIF`Vsmn92n0F!N?ngh z;h}x)Qqaz>w4-bO2xNHAxYoy8dS&5$-}p3v=)ltfjHs@BMHHF9=9(`NU<#mP;r`J; zRruv3qzU%HS0QA5XV4dzpg8W6r?A<Ts6 zSY5VeawtKhEW=iL&de8J(DU@TWIq~L&Skzgpt%p2Gv&AKe9Osyf-20T(D%s-HPJkM zwt?N}uf3-q#nZWNw+8PC&CoGj2H=xoXR7v_1TX%D9r@E=nqIVg6LPZ-Hl6d_o#)Xk z+WX^b{&=X9HBFG3Y<%kxjM~cmMA&orIRWuDR@Y}ay^0jITyl9Dup<&)(s3E4y!K>X zxd|<7%X{XbI{{zwqu+W`Kdk-^YGc^4ZsaFCu(;h^e|fu;!c=J|z3$ocFthv0XUX>- zeiVjR6V7oX)4bLA7TNTU`uA#N_$4W0M;0M&Zux{tVc%EwpDyrun1r~-m(J#Fb{g~i z53^ZQzvlB&$D4-olw^J$ERb}Mom7KB*ch2TUZf zTnLUfuZu1%L04NDXt;kQyqlz3p8)fv0xT5FkRUQ_t&8>nfK1^acMHb(dEzGoh8c<( z)rF({?)l}o!t0iYvxFf|A%*CHI)5qTU&?2!-;*{weQ=GIA+gnB*0~eOEvhzBu7?9A z#YH6W{U8m<*0zu-od6UeyyP3`GA%wYnK+zqQr|lcCUsx@Mql1YGgad-Ao!m3vf5(A zwOR6**c&Uhk_l$RG7*)6uM&KNvMX~BLec!YL9Web9XA`n8Dd)GAXZdZ5aTt7r#ruS z+Tyw!gsVYn_nwX9#DHeRqwY0LrZDSrx-fcq32ESI*!shI)IIUn?5U9~V36r+5<6ow zzDkvflxky@)z$aqH5oC9Gt_w#v`>F2d#0#7QquCCB^lo*4*ZK*U*a|`Ze)%r?4KA3 zTzul#=lrP%=#PCrmNi1-T0<#8uKPJB8|*-)c|zOAM2VZHWjhKbz`52&LF7le!`e5f zQimvgPESimY^2_aSS%i0n^UgL2_hfV8e&Og6Eqm@{H7rGm+>7VVk5+q`e^$lU0#|R zO*#TB>fI1 zj=eZ1Wu*ncDNhQ+vS#cU3530qRe^qx)^s}yHtBR)*jTJBi%XCH4l1qb!>+>SIdu4% z-K)2mEF#Zo`7hu>ESUC`cyHI>=`aP{2btpyc^;!{y{(6IR>O~?FC);9luUm4wz^h$ z(5Io@E*)vflZ`J}HOAFw9}+v5|D9m z3hEkrE}1}K8bFjJc~igYYkIXz!3SbmaS554}$ zRl)dHsU2ZK=0G2Kq}jaRi6b6*Jy8)vw&X%tYlzo&QQC&xwipa*1LCnzGI*7}rKrNl z*i`5VxfAsYD{dn9^#7Bo&pqth|8gOeEz9SAhZ*62!KAIWn=PN4^L2uNJ*jpqP0~ca zSc~V-W#JudA4cE|T1*s6C+C!JDp zdO%-!)GXj?M=;6@neUd1HbQ|P+o_VniTy04X0(a(^Y+Fc*FTtZOD?}_vah)%P8)`S zO#FMfC(h^Upzolm?XqeA^7?GLPa+azBn}6i+(m{v%3jv(`?sC?hrzwgKm9G+QzNm% zT&{HS3&~OUQh8?+;6M5srbaXCL|yP4A3-{*Dprv;@k8BRW`^CU^wobuJh!(hI}_qe zhHZ#?!cU&XXfKg2Q;rPGlS^VeFw$c5maz`AIa_3m3$?0b$Oq0*AQ7COBmTs=-yP znC9@m(chMu)bhk<@2+(_RTWYdqw&SRu}LBU!wBp;gifOIiFhF1B-J(#QhS zzM&+>=#lRI9HBf1;vLPq%Y=eXj%WH~BKDz_f%=~u-}5~hp~nMg=Yd(VOBa1|cRFM5 z1KomyZOt$=mH0pQdx=PyWPBv9K*tXBRSMFbvC;Xp+ZSN{Cc2s0F}9Ir0& zftyYpfG(U~RZZ4O(pMzi&|F)GeAn%za5QA4*_B4Z0e2`PD^~flN{*lOf_HZ+aK?`= zq4A|rhN^+@l@3+&$ki0dU}U9bX;N63?!v$ld}V%AQjE(Xt;z#S6u;{}YcIk3NfYqw zuAp1!bXOESTIl@$XAJ>0v(64AuhPfu@9+QbF!c2FNMG54dZEx%%tYE(Q+7`S)&#OU z1ct@Jin4>Kg`5|<<9J(be1{H)C6hpj3Mk-NIXeY~KPzL|H%FSc+;@o$s-%jEm8QO{ zWKd$PX$L__dw!Bc>|JfroxME&uSzNV>JQuhU8S^iS9|pTST3sXmmmTTVyHf1UWbRY zLg=1ltoIuAqO9l3b&GXE3vD0Vw)O6SRD4D9ZB^C>z(Or}HKD#+@<;g%Nrz_k>n8d6 z-C00sPF?En{s+zInrJEoB09BvZ%%1hK}AY0-QB}zD?UK0(u%^a+IkN}=rArl?GiJH z2_j@Z>u|2^fY7}LyT{@%&_zH2l{2uyNn-ynLBc5zf$l-3%8k27o{NP9NF;%bUWm_f z4BK7d%>#(jLURnrAE%=wunSsZfQApR4b~D#@{W4%(bWxg-@HJG~XS29y z|Npi{043io$Xg{eP+oEd677OM8!X>(xQ@a!*t7$+4w+3|pK}jG!5@4Ak8sH0QE?&(hR#DxEG3M@c=*2_Lj|3J*A9_0w$QM&c0`@+Qz1? zA;b?m{=ZBTPh2D=5a>5C!FI9mc}3vNF4O?F5P3YOHK=jpOWO}9664@A`tOp+vH~xX zF6EnK!WdaG@j`Q?$G%rN z>*unj{gu{CwZRERkFCZs=l^}~C*T^VB?-)B zM>4WvSa7h(Pjvy?QH4QjMbd34HFEq}N{z3I?e721x(*%k|N5PQlXsF!Vim_6BPbnD zL#sG)%nr`Uc3j?*5kbEtP_Bpb1p)87w{vplixJNwKDl0o8j@@onzuz zY65*&=A^l!X;blxWf!FxAgZJ?V(B zB!XqzYzQBZOJfrS*H6Y+YEt>JQg+QSX~MpQxA%nt%4{=_&%j*alImJ*xN43wVnT-P zZR(2k_@rqWz9c)O9#YEcfDU2+7h#c;%q*mmR!u5g@B_=d(gv|k`Io+ewTfO5puj{d z_3X4+Ds2l}krPU3LPcqTXJ|BuFJV`(8;x)^TCVg-j%xGFHwq(y(qfps`ktrV{KQ*7 zUNbu^qy7Hn!`a<_ z)$n&_!3S}sAd6oz(Yx&ub%ty^rxCl{t~qO2yllx@Z`ljny90KC?RKJr4-}Y9vown>P4LS!2Kz@%XGW49bp0>0OUxEs*$lC+S8;mUf zU16gpuO0T_ca${nuXD{ka&qqgzrm(nWM!o(v(7No%(>P3tS$P3mhx5b9~K5_k<8tO zTaxTEH2ySPBOVG!`;J$>273<5NtU$C`I6ywCAeFbKs+4erGsw{JnNFYk_9Dl3_fll zuSxRQX-|UF^qGlSy{(|8FoOJ9*M_VLA2>V`*SIr%^f<6aNxTqCTzKf4Z8*{CA z=dfc#q}~A?#%uq|XMmt^Z+rxuU3bJRu-V~C2tnZ&9L15^o)X?GATp%$`S7^I`a+(ueIKXed`qvM6Z9i4MM7QUI z0g=+x1yJxl#3bNAKH&CpQOLBWx;k`bW+wm2z}(>#N1I;Un01?%)@t78&=nhDNQnY1HgV5~u`{GzpJ>s(BvtFZt|F=`{!CxAyk-XJv~Qk0I@Ml3VffKx|M1 z8yo_%couK37yCdhPa-JX48BUb<)ioCt_sNz<16<-yr)SXjc;+4RBL8ZAWIw@*!#+t zrmCrLU~2eSdK;t2cJ@~!_nmqEp(IfCerZ=B5$kk`rr}4uUf6$VaK3hwJM(!v;pUID z?)EV@@k3VQq=y62lR zAXV7aM(dU0LlBHh2R<4VW7XyPQlaALO7Czhgv<-Ga1XphS_tLmY_k^#JJUu+cAqXp zzIpJZuk~pHT0lS*_SdR=O;Z=i-j~uIMQv%`b#jisUW2*=5}@&;b|VMZp7_@lXrTkT z?bSpRD#hR#+^D_YzDxRsZ>?{J2M%qZcp$c(s!^1c<%0hD3u1JkSJ@B=}J@V$XN2FUQj0pV%kSgS^r3j}k+? zH_^3ZJsUl6+$!J?puZFgSjMeB+tEN6`+0!AO3eyqRjTYGRer}5`>JMr&OvX)LGSng zO1#_ELA{5%-zpk0%e73$GzBC6w<2GIS#{6=@sF!H7RVi533?@i0_g{#c~5c+RTq9p zz3`4-g5!hHy|YxTF#!4f3i+BqSBw~33rFqdF-!+`PEh+qFr2Nac^?yDqVH zNzycPNlkP5GD~eafSZyCgr0Ya?L@b|2x?i_(2Q5HK*k5yAcDv)QYV(>!mI<$na9m^ zOnc;%5Pxjn^<$L9YBKZROnZxMZ_|q1zQ+~w+NmK2&3Z+$ZR2&&nI)HcCd{wjE}*%3 zexZMP=X)vZWRvPQFyNFYvD+7lai~*I$qJ;Cblt9S{|>5;jm2d0aEE+)=yr;0s43gt zx3BzmjL>RDP=25ljkvXJ{UhZ_?k2`q!Sp3x?rk=a zOGRYnrB~ei4SqutodL|UxDgWTa6)GvooB5JwQXw@q z?fp9351vyh+2N7oTB8iL0&*Q_`^}gJUNRr)e#6rbeEIcRR$KCwN~66QXzsUNx^B7q z?airyqTRNb*6Z$8p=F_y`K)0kCVGk49de-Er4ObD5hy0>MBx%U$WS}r|qYKbjp?jZ-2Ykx*XUTgh6ks6 zZkYAd62J>q+CFJgM<9&T;`d}|Iyr-c@(pQpkPSCs5Kg&kxlv}ncCDV+>q!zyz1?6=nO~te*(5_&5QBD z-p05$&?bqP1T8x>*K~FjP!L6zpfF+duE$l(a^@j&P%~ghy8RalJ_+F?zHsgDB@<+SGdjKZA1*Z%gR3M7yeFcL$QphjlC0Jrq zt`^40WCf}9zs=ejL=4N)4R}b+IGinhu;S?^5P;7~nF$O$iq1n*JCxEi^pn-vi)~Kh zh#~SFn^h03myGALWqBt!ih>|l9WB2ped+@s&-}LsFM-7}N_m1R!T=&g_^BLSA;*JG z`bhdN4+!t@)Ix{hL8E9pbTk?Ks}inXJQNr36&0IbmZ8t?u-esR145c+p z4b*@02jfI?kwf;IZH68Yp~VE@kN{=D)ecm5PjPu~+O`$Wm{Gkf>&?T~{3`valM-^CuUeWgYK?cV{9=kN51MFTD{OR;*ts~Lj-I@7zr;u+W zZi>qqw91vv^YtN73_+CmQ5M7NfF%CMQcZ7GO*;`aOB+Smy?~`g_b~)tu?MH8GQtz* zWZOmY@~Wuzo6TZ{=ClXhmwiCpuwT`?!kbKoj?K_sY0}kl^+#k+nCRnKuueX3y}%C` z;;Lv4m172H3LWBeG@=NKetF-R5kwy!dy>ZekG9S`9Lx9r|F;rKM)oEoBs*kg&#Y21 zvuCo`osr6_n@iE}!4$kMHmJ9>;lfboBP-zAoo=ozL_2 zdOjcL4cs&7VFI%y7sOV0^78EM_$jk;M>LC)it~qD`-G<`oYk_(cVFglYg5cgwC~g; z+xrmrzFhc5AX@@Bk+7pw;1Ph?$5G^5+QwhIw4N_+?;iedH3(k#uV+tgaYREKS@1C= z{ab-^@%*5hm3BY}+f=-2~gb~0Jks9P+g@uPU3Z^863xDNN^`^srV zo{ng+^b2?vWn-;6IS7PviBSKMi@D8>S8rp{7LMU%5*61^J6lyrNi5n)L-rLV4GkE8 zSls2NMBNfhXayrNSK62@@)_r^ncf^=hLY+6J{}6nbLrYk+Qr@TdCCXl(Vq;b{!&fU zl*47c05N4LcpjWp(C~-hG>xu?ODiIwi|>eLUC4x%MxU_W$*TlXA)R~g^1`$1<8kO+ zqqc|*ZM4W~(O|odQ3KmG-V3>&)uquj;=LoYPHn~`>4=^v+&XQ}pZTUo+-g~5jXM>^ zGg2EqBur2ig?#dppJ$mrnVxK$Li_yxuV)JT`(1=9j_ZUH2!sZS+sv8Z2W0798XtEU zoiagu2w>ysrG8hDB0f1cGwJ@kzTw)LJJ7D{bJ|7onNV-})00HZ<@VF@w=cd&{6YOP zLTwZa{!wK(N%5S9f}D=^3DP;x1_1*hC4B4sC->4U@k2A7Kjb&J-dUm}X3`uKUi+Do zLX0TkXS#7${4&O7z`LiCDX z4j)ZU0-23Y=_Oivxq`%~vQp9YaYB)gm$$mZFS>eLkOh;(tBh!KTIz6G_B)oj=0YQO zk?<2ttNFjH=-XFk#EF9LsD_cZJmpj)c##3aR)qe`;uuNMUG8^xxtC_wo=r19@0iOu zFBB3;=`@40)$?sbvw6oUuZDqW>2}q|!L?d%x7t8#os;1%Du>z4H}h*s;u3ujVTnWk zrJw8LlW4ujw|dx}=8~0`;R!Es8I89@ve^_nI&f1kN(ryM*K#W~M8B{{3_7Dg)j| zdU`0ozZ<(Q(qOQisFZe^5{#t7oJqXRX4t3t*@MiLZq%$w%rNbcW$~Lh$D0JZ(G(y4 z)_q|K_PJM4>(;C(gK7np4cw3D=_Bb+vY&rEe&s(BfV1Pl|2!I> z@MA68(bShFQ_pv98q>vtxAZ9Ac3yd_1_TEmH;o^pcHpcZtGWz@+Q7;6z)5M0z9Svj zPr;l7Qxc{EjPdc6*X;CgZzh~z@Ev(;k;?yk+H8t#$S!%_;`DxSJ{8NN#2ov(DAUL~ z1q)6_e-d0-#D+FP50?Kd72hw2e;qukQ7Q~D);v33K_>W2*%$>35aff)BIlB=pGWYb z{hZppZwfGhXn>#tWC5~tn`cWORZ0BE-v>s&WjAJX*k(yxQau0k7R}RJ`g{WVvbk@S zYr1^r;UWDb4nKH&ZRp&D#O6xF`Pq*GYC*A`iV)FodqMEY+OwP>F--W#L>cieXqR|Q zIRQ}vC1hJ=X#Kdsh&PdZ6h)O)XJk@6$y9A*l31@q+VQG%HBXl-11@ZN{YtQAPA5Z8 z2B)ukdhANtIt&LOWAcW|Zmt{W>%HzAdfl0Y;_qM!X(7rPJO?AB<%Yn$blg(<%!jHK1`y|k zu#aqm(+$efF-ve}`o(`!PnkN8RPQ~(r9Hm(VtnmfA6L&Oh5A=uc&5PQR|8`RYTFRHE**BVhrj#&en0N$DaO!P zr12&iq%B%*`l2}}Y>I#;!!tg!8-j9wS>2B+Q^~4JwqL9uc4JL;DX?1~Jn#w?0DCaS zOT}NVw0GXoW4H~j=hwX4Oc558SHh2OO{ivG8W)`X2VNir?ij^W=QbT=VLxguk1kQU zu-HybJ5{-F7J(6E+1z9!;e*?u2uady2=a?gcT`^wXH}e6Foo70Xz$E^kl)IgIwO$| zQlBrD6h!La^pV;;sMJc%j{#@9RLI#!9o%Z74|H;8Y_Y;#R-C9=s@F-m@dL3ww9ay zcQ3<^?#G&lK_OmZH;nAYVz5eq5oyOSg&n~+U3>-uFK!x3!6w`hw5}#v7 ze3AsdO>YNB)ip6Cc!Z6roe_S`FuFPk$q4N1;KjzpK|2yRVvh8Jq6W_H;aQ6-{2_=X z%MxQ{ji>a!4Y%G-$w(=)XJ9=>YTA$j**d7c(YUnXnBk+*PN&?lSMXykw$UTPurPbx zyI7Ov(?ss7XGnso*+}A@;AJW6pK; z@PgKXsf7)R(YzX^d3E!5e(^1EZnn+Bh)+}lMVGvjn`;t6Y?vv^p!3y) zS7gfWp#&nZ6+$xzWJ&8H&K-$hJ4GTe<8$9pbW@Yz8MpMtwJF*rULh2U@BxwC$qxL> z1h8#;u%R$^;PkQblFeVGJ2zh=QC4QRr zm}?|3gDuc#atR5;(Xp}5zkf>=6iuB)?LEPP07|IR9OQ{UHr9|#_V1h6CgSPoY2Z{j z4TqM1A;@G};$VD*A^tVtPah`9$E!Zh6YEm2J}(H2yDgw`7)F6Op{-o1KX`H#)#v_n zjo~ca-JxpJ8|U=(^lTHsUxkj)(|d}D@CuuwO*4LR5l5m%kl@z*?5bs!f}w}X5WwoZ z!X7=Y!ched%gGxNy!E34e(*8@CwIt~f#tVW z^PpQSe0_;a^%?q?C?F)@=P3_e3Y@?!I*h=YXhYa-9@U&XfS`K~lMg~?omMV~J47hQ z3`0%|0nxNx<=oOc@1kcH1is9Ss;2HsI)NJyf^OyHsYt5dYXJr7x2vazjUfW?fuC2l zqi>O%mK0!edq1Ju+!IDs3UkDDLS700KdjNJ=2r8(5E4D}Q0V=$=GVMG9Ve10#V!-f zzK7gzn|Zo=@?b1?APTJeq+v8;Wdum1VJj58d<++4NxR(lveerTI)lb6dMyVI*kc}H zZ}7&^H$VK6ytlB6e>r?BhAmX~ogMgDxy!~rHd?*zw&oF9Qq)lp79WZsA{?aIB9Je< zx_PTaz*z<;Og+grNdmK`0OFH{^8gTdKYoO7a(g#`S5_AH*=5`_mtks?xde%!B>m~W%Zo&Jb@qsHS>p~j%8OcJ=STlw{dKUZd0|HPibTcUE2M8 zDsGoj0#Ojr$XX5!Q_!J+_thyc3yl3fzp!IWE}r7X@u?BZv^>K&9BHeY(0&pV0R=R^ z_H=x$d18VUCDMM6JEn`^a$G0aSTpo_9zJ?B_M{MUQl|(fWEI4gNbISzX@o2=-nc>V({IY}ui?u`323?m>1wfR6 z$>Z`da?EzCj@z6KO0 zs4z80Q{FfLK^@k%l7KELw_MH8aVeY-vz z0VjvuTSUEB#3d8?*Cvd&$fP0MT|Dx|3n7RET5MZTVUoW6rqx{B*6EajoI_K^!H<0HF_vA|NC;EMtwPWHua>Z*IpwG&DH`_tkqwo~l84N+@G zEt#V&onwkpe3^eUB&9Vq=L2sL6B62Vb+2qQ{`uVxaTg)IQB41bGL;ZTQKj#9c;j?< z77iQJK!v#1!skNbEAF*h;9F}^om(Ie%;5Ym1Rk6`5PQhD`)$Fc!r*DfRfM)IAUVk7v2kW^9E>WmZiOvP?h6HT|2H zVp~Ow0FQkKllp*fnl0LJUJDSH=~Gpc_gKQCw_h}wUU+fFX}S>ufWMiYwQ)|WmkS8l zbFn7{kcKwS3thCh&zk+F?CQnin>r)Lc zv$C#C)OnF~a|m4kspjV9hQ#ZU&^y2y4L{AB*i)jxgm zBsdRcB08AR7XdFJ_4Xn8E}G3gkJ7vtF)_aS(-66+#$Fx0bI_vP_Psq_&=IU6Ot+ot`uY$u$LwFT z39D~cg?;>?sP<+^CTUF7Q^M$c_L_B-*N$q=U!w9$`E&Y<@ef`ZOn!yJj`Nw}2*bjC%}uvUIi z(Z_gaoOY+o=um3T2ZE>n2M`rpsi5hisF z#aB|(<~}d>TXXR4MNd6bX7a%ubEIP4`-J~zB;aVo zxo&a~pb8;SvjfXZZZ6lq@_?OQJdfj6zfUfbtWlgWpb{m3x&^4Hg9ADd5q)0=G#WmF zA>6SX6e?#h>~#O|(WH$>R%SpRKuEN0lO!6rF~(6GFK=ON-MyaYapqk~T=TZsHE^QK z`0=P|>`~3w%>d%OZ|mg-cCHJ4Ja0DENdHJqi_TI*Fe^B4txo8C2IuneNVJfV+W_2j z&^bV0HGVG&_#DFp7eMQREPv4ZsoUJ1Tk{&XX2MeqaS2|atvtk&@*j4pg7A0NZrKRD zN39dS;Gz5jP{5_7ZRjxHjQUV3tZIQ#et|e*b{w)0IuHYbr>&?6E4~#(;TxI6^qeU%gf=c**Rm?l>*)G;KeVq1sfaaSt9|CHfSZ= zH}cY4OqgYSdHXUQhLPJtgS7Q0bo<5}@xJ~dM$p0lElx?f4v%jqRxX_%!r($~RAH^Q zc)J9*(nwuc0M!N9J%GeH5@luEA<5*asqc4|3Us`WwgZ4WdGc&uJgyUCD;%uCp1lFt z{?F*1+A9RJ&NTB+2A2_-F@RT2wONj@bvynUS%^4SShlWu{>@`ngrdhGAOj{p^F6JX@SF{v$KDtLhIys|%)UP*_C79a&a@f;PF7;fqaePhv^-U(lmqEcLOel$` z$Z@yPB|C~dgQdZ}Ar*>pw*ag3AKoHotK-?+=F9DjdU`rMRPbU2{!*78cq0s7N|F*F zh@dEyR*L0YEDGe??F>+K(Jvimj8Cyz5hjAyX^oNKx#Sz#1unI2iyZ3{HFiEEIfGW7 zc(1OL5lwQc09ZYF)Lfs<_ z!3zZeHgLGUN#e*xfmZ36HO1pm%=j93wji}T3IY(;_ri`VniRlW1*eAg7obl9+%w@! zO0Nwd!+cHHIt5vyXynpK&pdt2^-WkqVKq3`10$CEo$l{G_rfBzSvlr$v7ye7FN};>jsvE!Y;SafKRcNYK@g)!aj&r5I|}mv+w+8yH}OW%(pT&FVzA6!-cHox|~# z)(J&%;{i9W>?c6)fQSt#x9OdmsKH7(-OGH%5_(YW`FN}k5?gYYBaL|Tl|@pPvf6#_6g}- ztg`eCYVI3-S@tx8#+n5|72~U`SHVsJ2i5-}ZK5`UKyj3$!p&QXne=@rFmdK=yTUezi^B6$*o0AuqMtimP}X zW1Tq1B_&4JoWYw^b1yTQ%<}WXAv7uAY}U?8IhmVpk^X=;&Uct=qn$JIwQye(MS*>jrJ(^aLWmQ&~E-C)w9~0L1cHE#{0H4C7 zN9i%%<~q^{LFXs$4#__i36VR8JV^cSRAtz8HJh5@r8${RDnN+dRR5ZLOI+zOH#~;T^j(F6$5jVf1A9DmKqPAY!MtYBauR7NXgX@DLgC$=bMUv15(Hn{#F5PBdz4@nI^Q}0>lh5X2q$1>? ztpySW2%T0ueA6v|{~ALP$H2Y^Ey)Df?4GXK?q0stI$Sq~;z1Yic(hM!L_!@1;8A+J zUDh(K09o&JgUJhzAiqpgXid$zuI_F&iVKnTp=nCb(mBJ|`zPoxc`9SNt_ekm#-yukz+L^PsfN)cxGMC1tL?|C)=6cz3^F0Gbm* zCR<&VTv%A}*iCI!h!~$F33B#U?}1L=quwtc0dGBduXdtNyyJ-Vp>@KTjKO-lQyc`v zuc!p8Q`3ttE6h*!qrGr@v2t1;`NGfO%C3H;ph3%$Mz|(~$?I}c-gcNA0_b>Q;X^dK zbKvq@cGZm%z9rTMsQmew2loWi+CMrEnj2WbN=7vUx!U)5;{fm7ihlHn(!t?cwclZO zmN;!MVQAaMlU3Tm`Kt^OeUa!us7R=q&n}3iXFR4L!ia495+xTPn1oviIUMfp!*1wVWH z3%qfM?@t(mXwCzVk|VmkrzdtjqN5TP=(lPO4@jG20z2cT0~O5a0D6`Eg5#z;Mj9~^ zEP1ZOCWuL70~i@&j!L_|)(rufQD9U1?ZX_``NHTVa}B6>-O=`sCW1B0foGLFM)Bf_LLJe$@cA*O`8%A zl{^pCNckNXh0Oj^&OZ4GW@Lr??7{#9Fc>{)Rk2)mDL|OeNk~W-_n&j<6A>APiCMgO z1k6JgQUy*5B{7C&m#qtU&tSSoAMA#nSJcykT{uEbcBO?s0bOJ%Xm{s*{*c=ANbMvO zyfordbl6%bFzvP%kQ#)>_QsVvOY=YiFB#p8FK!_cA*H|6WE zmOzL+Yy|p?MxwQxNFq7n@|9!)?T9fOF5b_>!7$S8~d}LgQMGQ{;@J zh(+y_M~(#>`^UkT2;6qe&p7D^meuw%VJpYivOoAehgS8F{gCh$f2Db;TNjTY7D_6I z_wwh&h@Yac1;n^=1bEL5q@Qbtu1PU%e4%C(Cs5RZY}l(L(c+aeN8GSywyN>elHSLi zIvDm!o_!D@jUYoslERsf|6B#-cfXriO+2Mi(#D=U0^yU-Dpnd-JD-QN^xK}jE$2E# zp6i84Ol=UGX)`y8Ow~VY%t@hL%xn~5@W!!rFiv^ei~d;&S$v9Myk_~`Y@SZG(pT0L ztgD`Nh4g6mZ_R=o*BzEe*M0xAZ_0Kv(bsM+k5RJ`&OPg0RA%YhG~R`kJ@GHeQ3V}4 z6oui`f@+TyJcNG_AAfGRke=!+r7Wq-`ovA-mn;Ubx|6Y}{g7QgOSL23pU8nb zrWHY^!`CNCO+Ase@c}PDGk8J(QGUgZIr3_|IGI=h0c~nWU1pJasjbLCFs(|B(T@>V z7tw;L3VG_sbtVSobzRxUd~IAw=($nrXckqaG0nNr57&9+-YER3=YQSnOd8akS5TVg zP3f z|M~&tbT{lfz>UG44$?pcwgZKRqKJ0swqBIz#Y$39_r}voQBfE@=P`_n!eBk<*pY?Q z%pdq4aB@iLR`BD=T2RJTKB=Aa?KN(TT@}Mn_yDBqV!QRLwQ9Oo*5!7Ifp@v*jXM~3 zKlBLHJ^D`3WE%bQ=%-kZIj;Fz!W$9iWcfS8J>wWA*GM^3&lp@6aTwz?{$V`uHb#8t z9p_rn-<*Q2+zkGw>!K_BoTHAFyo==#r6Ey(U}2~ozvmU(^2qvT#IiT<#lgYSA-GcrkNu@NbvljGMZ=ZtX-NP z#{|v^{0E zVC?f}^ImTufn-=M@!_q_8btsDUJo7JhZvK8-!7~)2nDR6kuy=W0(>cDj+XU04?{Dc zoQYRn-t`2Ie;F2L1-M%fr?h?^hNV3LssKPSo}?PAwF2HfI70;?(FbRr*dF{SKl+#6 zo)|U*DNOH&@vqB&YeY)z-0QS}yq<$YipdFxVq|Rq6V1@jeCqD34^!#rTPOWbj^gBD zxPak=b)A#|x7`;{LZAkMJFuaI_Tn9{Pva%}7VAugc=5dlu$%+*;IUA-K5|Y z$PF@e{IJ7Cm7t^j-T)?WnM2}K)M?@q1dax)28OE|x32>blu@$R!pVvFWMTSW5@G_! zg}HvSvzmG(8qP#s(bs#x4TT4Gsm^i`;}f069a)%zgj)`+q8{8i)u9 zA&@ZGN-(#5;5;8r0_Q8FFP180LjUtNoIkS^MEkrX7{JQ(ot+KatvFQaA#!F4nZVW) z@M9igY|nmJ~23Gmt_M zN~we+f~$)-k5`N;4ZzRVuZ)SqC{_~kCqrv2`{v^hh6BLx%0nG=9A0EtG&nX zH!PSU@9+Ah$?;Hah+DVAro#PKtP)vH=hNxf$S1$pq&iWkpGcqq!MUOwji*Th_T>K| zAu)7nT(|Xg|8@YyBxr0Xg!!+EHl!@^tB!DyZ9X(J`wdwF*7)}uL(l%!^d5;oXmc83 z)HJk95+1_D6su5?ymlyU!X5JmQdHx6{}yFPnx?0*-uyJJ-@b+W)amBJAu{X7y>B7C z#=;TLM)Vrw$-kY)CnD0@xmrkxZyFBTLY2^c1|%@*lzs~J%8@SfL%ybeg2(+b2g=Hl zk5031JP~)7SN%57%ur~geE_GXJlXFwZQq6yGya<#zzOUo3)mjI4~0uADk^yXk>K0J zRe9sY(;rWOS4fL|jL=JgxJk%glu%?~ttAMe!Lc1Wrg0u6E5$Wz{!>xxM`U_15clUF zws-FH3&I>fHTkwk|1GvByLHowoD^p|T#$^I=Hc7m*t^^>7E{s0$kQ>>bi6$%osw?l z;X&CewaZXsR1>rYq1+??0~0A0Pz^XrVWUViulmWpVB^1ezO)@KqU*^2*+L>C`7w1e zN44-?i;Xmv_LQ@Gu$%*@*W9k(n|+z%XCU9H&e!l9%Hd?fkG%o_(Q&;h79;@e*PP^%y5J@xJ%Wpck_HUN&m=)N#K+O?+ z1E~N4~#Rfo;%fNAz@HDChA3dNk?FV zS}R%W(Vs7Dkq-f`(rrzSa$fs!138>42d&$x4~cBPpCK~;E=wORYq0^~XgW6_bnv(lr`WbOhbt(4W&fiNr!3NACe?;uoFG_j;ic;1*{1_YzQ?jYe3$pCEXK+;B zD3tr`&Rmt*Ki|0D{dg961drF6j;avw%SA|F|H!%FUVrdAf3Gh@`Z-8Z3@Y&dox!z7 zeKKpB{5Ju^^xqTawAGUT_XL7aD3V9Sh4$J1J;VO{1$!qrvhOV9;#2+S;rKn2s-mVs Jk=(n7KENm>Sx6-dQy}xZ;`BZrL=}{3%*v;FG zET^~b&u89_Dom8up!%H`qC>4_l1e2|ME3gi*n1g;m~ec(T&nvHQSX0Cztp00|N2xb znDUNHjfsLTElVL$)RQOaiRP=#radxHR-<KY15bMeYB-;MpTWv`o&zRF&0PB_o=*X*XcEv%C9NboZl{ zU37~|*J9p#3o;*k|L@0oDK+sJ*UmWjAHEFYZ&{>k z!N%GfLR}7>U6kMWUbdi4(Om(72hW1;q0#8G?A>z)|C6em?hiMcK@TOb5RzAU+|B>< z<|ca@S&Y)|1>K-g+-iH%n{x`c(q&@vZRQU#Xa_YnAC5+?X`oJxD$p09(?IU*8?BUpHmaz|MP78ckKJmT?F$He13|K zN*%SXKJgGFhC>&GNDPo+%R#2ck*YhZ}-&q{$EN=4LfIWH7;=_rBH#oHk zr>!?v-j`>**X@#L1VZBC6Fb_)ntS&@?B72dD8E$jU%+K-Lk4am^Ylu~u7+zX#bQ$} zlrHRAPwdXLZ#F(GmS5Gh;v>c>PMvN}oO($w+ejBk!D|s>owX521P7s=ZnpK>T3U9R zV{1K2mi1d!Bzwv)2g4RjX29JPk_e2BiMhMY3$8dRV@tR~!JS?g zURq)o7ngXek?co=INL3Lettq?O`%e`gflQ++Bntn56fp~Huf%+4s>2SvofWyPqa)3CU!D_qTz7v@#Mb!DMUggmryn7xC4k`>xL9;OxEV(Ky_ zHx|q^;iEATqR#_QpV@xGRR-@6P%`Jk<7wILJXFUq7OtQTzl{*Ua$fj>Bdg@JABq$H$zqB}3AfOu9jXYk;W4IuMfPc_ zikr%9&=Nc}6d@Y1y+`^)5utQWfrPVeTeb3?hhmd3;!Ym&aAq_YgqkGc-*JO=tv;xc z=Q)T|M3GiW-`d^1D*^h5Y$x#h3ee`wY%g3~(3WZ)^o&C}z=hFdfB)B2SSUV*hlrOU z;rDtqDBJ8a9fUw-ZEZqUmZ9gcE+`G_nhbD0URO;;iWh3KFaP_;btg=d%hqJ9W zWMNX+Ce(Qowu|eC-bzC^uGG#eG%2Xsg$4$eB;JiQ%nlhdYYk1nHIL>TocF6O=Q;L`X=)wwY(fR5tmOp%*Ct`F-{RTeWWJeTXIP&W#oXTRtkXQaEOgfQqxF|)k$+2U2 zABEV+80XxF9YSZqW469!HK-LnQXPwf#7>Z|%^Oe+?}$L!Dv&_+_`h zgiO^GXwq`EbXKX}KM*$H!~k$`)|n+;{sRE;KPTffiGKj&4O;Qx=nAl z#^ys8Q@c?Dm&vol*C>feIZ$-_fU z|oo`s4|2t3Ni3I^?Bw$Y2R-hkoooIXzWVFE*Q!Dtg`y~A0J zLPA2)T&Z$oc%a0*FOF{eii^kWiIJY}O1>&k58Z`S=-~bH6a|GMTQU*doa$O1(%Jc# z^3JWgvWRX|u2<&=2ajBse`?u*z9*Bb8oX9uOpRZNh|+0Gv3OEwldD&nXHdqJ3TFmw zMt~Eq9S0nO&~~hC&%32sZYz2AGlJ3G_YTf+pt=KP=OHRg)y6b)X%H z+*lIeI}gQ#;!9Nju$UDPVSmDl{`f_0B7t*=8;wai^L14t62hM&lVO zI7}(b=6m9}UmM>x4K9ioZHP}z6rMKn(n?Idm=Zcg51smEEOsUv0lg4rp_sNlMvGVL zaUP*dkooq4CG-h4Lh+$IL^3IJpm78QG#~jGATh02y*R{Ogm?zVlf%tJPx9MZ0NmT? zeQ$cdZ@Xx46{RZW={|ynLrwJCdK@$Z4j6e22OXGwd|0?Z6e} zuPcZ7?fn7%E$#c5TP)Vu($LTWIs;E;e-grl9@CQi{Lp(%l%NN2aSdZylK$-Z+xpg^ zV0*zB9z5JtbA|1N1f%`X2Dky;VYk&tr8W^2r=fF0Cl|>gDJ$bjwOlVyPgg@}Uf&K0=2i0emYr?N z7nS2TVH=lKBCJC-iYnEj7|9mzH2!`!$YG{ubkuO%_Qel%Hh^2J^_0W{FNI*~Ynncl zIw2-lS;_<|ITGy?69t~F;0mKp*`m$>h$3(~-SPVqh}=B^bb&J-6zy)8oWub+7+Ejp zy>Y?q`4v9iyT0yX-}|lKd~a`Y^H<<@<_BBIVVZ!xK=DT53e(BA0j>2o#NnX{BT1(XKYz9zrSBnGXGvmoJJ z!M&dow4nIZ+43HA^0c`t)buJt;w-BwU3p;+bYW8D-Fd$ZMdR%!wf+jnF!PuYDl#B7k9c!Hf0%|m8!I`|@(;s2-o23T1d8d#) z>MdL9^GOs+PB1fKS+&0yvA%S9iIg=OKXn(G!*Y=^Q(# zAn+4@OjPEG`~7?v)YT_0U8)u8APSI&_t8HG1A|2C#-zt+m$!R+t%QZ z9P0EK^d>z^FtZ(V7YPOUqQbB)frse#?x~B9Pt@TdV$yWUCwsC6f}oj%g-zLqJJv?P z;cnxb$f)lc94VN&Y+A|6_nAq@Sb+9+LleS=`^xA}*dz92cJOh`Chu7Qs@qy%0#7DO z*)N}}(>{fdLpk@Ldkz{opjEZmZyvq=)#}fa*lFp277qKF@Xdk2&sJ0H1@z^*?{3xo z#L=$$XYtb+OTK_Uwzq%l7G(SVogAw<_16r9w3!-cTXB8MDGjx?p!tG60j0A*+I4r@ zbj5Io*t1=VA?O2uBhnOq1m+@a%7>twaM?;7>0DK5!OV}F_rgf9U=H7B9`bB7MfqO& zt{)1=hbn|A2&XHi@XC={eB0c#Gcb5k(ydWU3C5%OA;GJL00#$UGT8O zPc!WWA1>0oaTxYp(cY|`Jkt$-NY z0Z)@TGs6V+j6A#fozvLmmHwrCLG3s>a82?}O$=&>WwwKv@ejCtt zX}eskqQIkRJG|jS8w?@nPwK5}`SV&V6~0)5oCjDYY^^>`vCvweZ2xaRMQpE&@78=X z8juP30{MzDPhQh_w|aVty*(nokabPiboI=34U;`JK2&#O{XM@u2^U`{OX7PQ zy@84dZq1VSRkpf~@s^V*)(Lz{kN-Sk(dVpZpA9zoq?)`4x1GFvSrB)t8kL4DtB?MM ztd43}J+?oe+ar#gh79pct&U4);a}%*e!#oUkP(We_u`f%Pl*CFx`mj8vSSk|y@6wq6?l?ma&j z26)EwnQyLY*XZa&fI7Uh37EQ!d87eVJ3Kt>vRA#IU!eXszIW3D0NglP@y{$C^aK;O zO1byNWtb%_uaKP8Go{I#n(2yE-KL*!_mAOx@*z(8w|!gV>m3rx{(E#cLW*S@Ws! z8P3-!ck?uP)MoNDgO{s$M_6zFwS4XeNAEJ1fOZ>X@P=MB>dS;*N zZg3N&r)S^4-c!V&wsCy>_f~Z_BUZ|a)}TnB2aLG!0+>myTD$=&=_DKUgPr;281-VD zkwczLC0+PhQrd5U$}49g9yfF>YeMKWbgxl0Eyp=`7T>zpnu5g0X1_pRO$@sdto0CTK}W6CO2BLcpLehP|%lxO=)I-Znqa0F*&% z0cAXwk$0f~81;g3kT}Y{0RrqyuF58mf@_7YH$Gei`XA*p%5bIj`k~NILHY40QEYqC z0VEN>o#N%`>2h<83#e%tOO!%@F;7@V%$3-}UfzX1Nc?3?E>=d&S}_QvPV8WOTl`g&kW&hBXGCR@IV z!?|O5(p^NhW}+)N<3ITbzS_KQeral>3&S%AjzNdK!jU{SOOQTaWZ@vJ&t6GqQ&3<% z6K*?Pd-e_?D3UciZ~!>_y)vIBN;7JFs+h186EBhNF4TyJ;i9+f(VO!lTbwcml zJ>iyaQlv`_*+L(YwG=*oC@i-y0dn0g>R+wT%U-GqCzng0y-4-| z*>86NBdQ3-fHPd)dL3S5a_ajKe4BK&?jDd5yw4Ifm&!IxM6xPqJq?~_!&h{P8(9<| z$UntM!$R+~l{LZ*OLqt@Hthx90BLAtrCZ$!0!jjqJIa`+h4)^j6WR|#(?q7XJ(s*p zGgts-qqw0=iwcW#pYAl~`Lp$?UcrCD#yBIz`X-mwQ_p4mi(u23oZkR+^Ru(iyFA{P zx4Fim@6@n%dM0z|Z#Sms3I>tX3oiuOmE*jy!)KyAzPw?>g|v=y=%1Lw^#q+%mP^$P zILly#-BnxtZdQ{i7FD#c5xhW~(tvo|3w8OT5a(7Ex=N32GE>qj&W z^MFG{X1zS{F#v<>tHOQrDqSIx;;!Ow@mKjSad3D{0X|z}BRaM;&&@_83qU3w;{K-{ zs**SE$#p;-_z1Kn1R_bAcl!Do8js-)eeA`rwyVju0DSX)nx#yv=2dUS_{eDd(mgptxz!Vi#9-3DfHt$ zRw-5eH<=;kac4u)Zd3Ae|BsvMeaHmrsr#$^dHY;%dH!VaYt}tu9R9#5sLE z+`e;Z>hWkFbBA-g=8`LJTa_J&^W}-?)FKCM6M?8+X9^80-~5kQW4uUqShTl8kMG|I zQ#S#662S`z`Jxo6OrGI!XB8JF$_wm`cF0HOIY+DY)nOf2RNV>D&yj8GJr(WJwK1_TK}!$dfR;(XgclP-7-Xd zpotw)a1*{sE-85jhKA>_p-N5mm@hO|E8eTW0s$ZmFXe2k;JDjtMBvnXTu9b=ebst% zT1#_+5Bx2C`kR9f*Ef5`Dpw3#+#~y|nK=`ltJ$d?bWTkKXf%O*fqI%E3+87bAPFyi zGhd)vtQT>lj(iR286{(cALyopg@!&haZpgU`oS(KE*^#b!-N|1n}Ra~sHs6JUo?;l zWb@Zg5>OL+7unO{}a&F69pJRE|ad zc_OWJQY>P+#WRPX*rc;to%c3u-WI6bO55{2@fP!Cep*f9X&dffXUZv@95esmloS#W zIS5Ol9|p1@S9m9z-*yonePT0m|E`zN<4_8=?03TzAAX4sxfqNDpf?8GY7q5~>N zR94$;U?S0bbsdgX>)IJ`ALA!_&X-=jG78=l(cz#9xd0MWba6)g6>AjyP6)#j?)~Ah z?z`W&>eP$DZDLAd-c&~jq(i-;0f|+H)OZ#}>_VmpW#`K}w`G;ECB8{95=BZqB)Xup zntI~-?(VD4A*?k|-;w-X*j9U8K^FSrN9o!(iwS<-`smEh&Z_>9tYLrLT;-i5ZS;L&#g+P| zSW{L)<&Mk2y$DnBX{tN>7M+eofrcMaDv(dfQJ0E@ba!E40jnja$6rFOYD%77pc>F~!08{Q z!c8PTd^c+Cu#!a6j0T>YNuHSLC!hUK%>X}ud&w*4lppwiKw;o4jf3+L(KF1q1FP)w z$Xa)G8So#vl`jDi0Ynq%?42(5%pAht>SG&K>fH0-y;n!ZlcX+I{6HWsG!$1*`U&xj zpe|aA6rhxM&m30{Zvn0gP_jb<`L4J_O~!Nm+OhRmK0*Fae~HkkSlg0A;oIY9JPD4q znUmOks?ut^!{VkMk$Xt(he!KiGg8xOyDptwZq!=P_wAcb5z|8NIhn_0D zCONkd)iR7C1ismZhzhc4Z{mo~WTp|nJ)cyp`c%WAw+qNe>g?aqs;l0aFQ) z%6&3>+QHLh&Ie4EH~DIE&4F30wvXQ=8-9hqKWiB&I-v1wCgI}Zat?T?9f=Oqi!EDW zxHb=Tk4GTt$Vv^$cruz;e?{7C?IcCk4Vp6r_Th&NI(v9TVW>(z!eOQzFc@cVZ#?Aj zo+0Y^;oj94i{y5r2q>1AWv9!y))1AzgCwk?2N!WK=VSCrdt%iaye|YCzp9u1Nqgx` zlVV}QL%pyS>svXIl#M_|L`3vix&aK=xk#3px+Wwl8k4m*OZQC$`;p{+6xrC5iKMD2 z4%-?-X47SltwrZPMb;D~+O?j3V1=D7C*9-`(}qvM=dg!@@Vcrm(S@U#2LVXvIFFI7<~`uCbsbf^i{w|Z&%TkMyh4gK zuF{;aYjq#0vr@h5NS*f(?>YNjiP-l1gf<#GOe>5st$ecOUhHnp1B;w2eBu;<5SZ!` z%^Ltr&|g`qIws~&dUszcHxlCz#5@&>hf5a$`3O@SWX-d#lpWI%R(_H4w|!2kx-?gcCY0=em~ zGB87cT%mK~NR1Eh1!%HrTyAFfBiTNHnu(IVmtq0r;O}~+cgRVGn2mt?7=z8Y?fINM zk+_p=M-ym<6SlWwZlxWYM^4#ujdwH~MHGi=*Wv*U%Uz_C>XO6l&(dSzw*^D(DY&dT z)7=p1{aCb2oQlxD_3zd;HTH9622M#!58f*czuE3o{;A=HOVN@mpE2Cczg?7Y7UTHD z0FdsQS*!xeQQDV3cq!=+#TDF7?;?(;YPMR}raZLv6H@6q_q~HRH%0VH|BS`GVNaH- zH?Tk~yZ*v=jIQ4kn7I-Wgx>pZR;D%cx>Po)fxg%BY8=05UKqTcWh1yp&hvmH$#&-66Jf%iU#sq<*jf%h3wD_nxzqCY(X(qO z_PWhN4i>v-Np#p{dns;ZE3#6hGf6zdn%POBEzCV!eQE{O`)V&GGcEsMee$qjNxIMF zKHDm`!6-?cn5$4bqbq##&EENL-^!iX_5M3Ng zK;-$sb+PNkMs+dL)3=`n0Mqq45 z_32HH;C1tS>3;0o^jJ z&Ad{lLKxZ08>7+d;^OQa3CzP33w?WnAfU5Ub#XKrlo3K$M)FfIUXE4?GoIyW*yrbV z{j3E8gRJ&i=w(eaTwv`L*A0R$LK$5&!7g6_&0}z+E_8mj+%k3vRo1lB6$G;57l)E` z!QHEsjlaQch-V?gPZvHj9pxPxywFtrYfPK<_i3@!9C=*C=5(6!gwJGCc?NFtV8w*e z$7rwdv3`pS`B9eP#8*2~~8O5vjORXLaw52EGsaezHdS2msC3O%PrcXRPq` zZ5>l7YPVA2nrnI+h{Xb~RlkxQ>x7_)_3u+ZrajbmJwPo1hfkrP$5L;`o?1R3ZbmGj zQVI|yL2l^PQxBo6%(qiDn+?AXm$BB}sQ`ZjI%#h&CJt0s`VQp3$n!LDL^X&W?6;F%Tk#MDt(>;R0o8Kj>@(we(1J$J#n-bL*h5e`WS~y0LK4}A zhllt6_CMyNbD<~j+MVwNzS0i_4F?sr9_LRp85IB&Lh~Qz!TnJWbWhIDX@FB@YOKLZ z88#WVLmJ!r7`OudT5Ui>7`2zbJ^N7Y!>VR2{SGggTiVO4d)=);x24$drL!xTpZ>083Nve{DNPIQNdQ##)_e(8Dx?0B=sgi^524tYkcQu%5LS!O8~AhbX(J_Fk}u1w?29J^~^ zl>N*csf+0@TOx#1d7Gdb3`Du&IxcraSGY1T}s?y@@EURj-kH?-AJgyfWA_WY) zfjLC#z;T-%L_#?%KQDYoEXy;NnVqE2ip755jBXnWDJMse|7b$T6YgQJj9d*OzVvoy|zXK zDSiM@?D2xmxo_IEp0>RHPm{bkXuW&_a#?{&q9$YO0zi9=%i?(=ASx`})8EhUU;=n# zERLi!`-)km4oLglfk1kBD~2qqgoNrso14WwQCM@1bYT1dE)$EbFY%#mXNLhGtse*p zEZk+@<0K=H<_M`@^C{A_9`?w66)^SX!ZCm9$Di~kua-T3__x&9qWd<#AlAu{Na7>k ztmGv=0B%CE?%#b0X=Jne)Q4B>U{^%*WA+Iy9^uhb>iyS@WohJ`oPx_CmDWX5RYo`Q zp*ufmVeBIBa89jb(ypCcC*|E22!j&XYs!@8mq;9EB@O2WXP?;5ST^sD(L zH?3#0Yx7;55#?r}hCmS*wf2C>g5CK2-K-{n;#3`AdP>X!G1;GhH(k-H@=79SWYjt_ zQ@uH`bPnla0BDn$lhcRlC?yDgwH`63*xnt|MAg`TyARb;^*L5ee_1vM){I}p7VX{v zX-fESPMCcB6wZ_^1fkc^CK~_vm%I<%gBoqMtE8gQDC@Z;OSa8bUAK+ol9c}VuLiZO zNww13X)(RtaO%^sY(*AYonqwf0^+Cuug1TAz%ndh{;12COu0(KN9y4__A1+n_wgT` z)PF8;8jYKAXKmGO1XZSi*jp|Nl6_@VUNs&%&qE}rQ~1fW64=fV@9>e)T`!p!?R)PX zsaXbA_zszhn34}nsL6a2N=lvsXRPvANI&|H^tH(Db6U5#ERhw!88O1_K1R{Hk9YEQIAR#o9NWDE0%&lxxcDBk zCnBpj!MMrnn9#9F>xdBeZK`YDTuh>f2;LPfLeAus!;qV%JUZfMou`OtsbO@rOrO5Kq|fR_x2BG$i7l$NIxat`BsMhfi~eHT7n z-X}yNQ^DI#7)o3VnD+Dg!L}HH)YcYwCMx3pEcpO5a6=G}03kM0iDw#LAu5c$1NIXe zB$zyap2LH1Y~a8r2!-)8SOV6%9oQ*>y!T|3< z8u{TyM95V*5P*mX5b=Aa(v_7aYz2(mK+ynxf!bY~s)F#YJqF0i+xrucS^z@gAqu@c zqzqD_RJhY1Zr9w53q}fWZu-Z-R(OwH_OmnkOpGswC_HZoH^I25bB?*A-+evR3%7jO zzQQdEn>ag9i(4MxeY9T5driDuppleEqBUh7Xw&?Kz7l`QmqK5?*QJ_aQIoxUd`Q9o z8CUJ}8jGIMkv!{x5(^uDQK`XSs*`Sk%Qqs>{UQAF<9AN(^}=fJu!1ih-pcEaul1Ix z+Ehw;F~2HfTFs$x;m1SdTDt5mtaQ9_hj|*R#RrE{D;Y^kjJ%8y#LDRDo;%ly3tZU= zd?E#J=ixB<)bX-@)Gcbd@u&S4ZKX+^T0PFn^mr&yfjSKs&Zg$8?jjcXLOI>i79ei| zaF3M{2?-s?5fRXyMg{M$yW?a06o7^5%3_SlL*)|YdqCEbCud|oRProTNXWbH#LK7l zKmDGbu{CNWVCQpMb^Zl<8a8xM%E*+%Z7h*`bg<*lM5r>Bw zz(2A6-mj!umrt{MWiPM+jtbBi?(0X&Tf-uN{tuuXxafQ|RScWY1v2BL37(Z@bEZC* zhg5;^R4YN~$1nX)tfpO!aX*N;v8EnU$63bW$djHFV1dDMtSt~k4Co=Z=g*Sq8c74-L z_|L0+#`w(>`}f+pg)PhB6AO2-_zjHyq!X4lYchUThTKw84t2K17t8r2v&wcXT@F%h zakd^26Q#xP?F2LP<{Wvkr)7w~BGv@}M+QA@1nL51DLA|k(9-8leJ&*SgTY}2Hq|#s z0c6REu5W*Xy(2}vc5+o&O#u%9Mp*)*qAL;S~OV5@c`C zWos{ptYMR79RmTf&hL~9B&QdhgO``BKxDd*Q(?+JC2A-h#KwSb4DLtw`g)Wgei94b zgtNE&f};EM*HI@cklmg313+0+RP)XAnaf5gJ0@?sJH0I*Q12gQ1H#TOYuZLp=bY`=JOuU1Qvx0(-KeSNt%`# zWJcu{#l=`nu`zQYHmk%o@y+A3U?wsF_3!o7cUTK?3Q2x6*bKl4CHzF!^Bft*hdv3Y zFV-GkJ2TkM(4erCAQZeuZ=8;4orKg8lk1J5LD}(prcxVlMpK@DQA%?^$D-DJ z`YfZ^N7n8Fgj&Hthu6b@Zn(<`WCMAjtnt7NAf9Ww)79;0&aaoLyWBClbM}|MZ38yB-3B z(DIfvqKANrUkC_>Jf0Qmw6a!in+bpiEKI;PK&kP{+VB2x+#l7_FSn)HWjYD#GIgtT zRSX6>;8b^MyuK`H)c3{x4E4<9%d|M|Rb+vGa=b4TKQcQW`SVG^sXXD}nT4S!ShrBu zs~-!wW@*q5DZ`_0hNOj(mLpq;^fH6TsG5Hp{OjR&AgO40SX^w57M9h*Lg7)fg>Eene~=ja@W;fq0XhN+ZILX&>!h2%l0Ej8CUhnLGe<5UHy^35vE^wmudgCllNHj2 zR)r)b+y9G+4R>~mVqz&k5#_zqa#!8tuw0i65 zfMM>%6_;#2Hqe&_QXb|tRCPT6hRau5{+ESrRw%5^m?>P9BXgXRxH$Smz(U|b+UJVF z`?}WE$?sm!e_nou%ZX6NLDacKPvf(6dT=U2+=-lDscfK>hbt_!O5@>w!=L{=zuUU{ zYGm^S8ZJ2@DCq77-De5j#P_xwsshoUWE-Q8)H{!k@8A;QHy5-yW)%K@rCDMzDcFpU zQjC{rw9QcbvzMcSUnU5 zW~3>9wK#VeY2cn8%)yWjtZD&sI3{Snb?NMk2&{6%12eb(Q{i9W3CmMJz=G3k<0%;! zU2j;DI=#iZ*4ZW?noKX1?t&O6=B~}n8S12jg5d>N5O5iLpuwhOCjs%IQ9;@rgr zpJ&iaiuex@fC1yqgpdedyzd0{FxE@oJCwH16eqd8z_IaLu zrrFI0U>Ya(LCh`wsedsStIk3?&rSK1G%~$PV}|N9GDJSIk5i7)e>y`M$PC>d-(Ao+ zDhRjG{p>R}8W#(P+Q`|Po|*I8_GU)285vZLaBp7JIm}QwTCcn75qU0BHPDZr3uy)F z{xREb$nGmDu6aHpKn?N*A6buqbJexhXAZT;)wk$v=;e4*A1$_H0WxJ2t~+^pb0*g) z{uJGu?%9#ZgL#Yj8$U(WKb`&kydbb=w%Gyr2?)qX=#}EPwXLj#k--j#Zy`HDanO2t zo>b@zg3vN00Kj3bdk-Y=MxNVy@-+Pq)B>@0FHGEP2Bg9q;=+g5?$!m}U|x#=rIL@Z z0h=>yAS7Yqfdonk7)Y3?5-<=gt7z!xo$t}X-~I`gn?$XR)BLu50rC;?B_%OwlPm!F z0c-{Yw-OtKcZ+2e*M_TFE|Jnd_Nso_`+pCG(+cTnCLLdD#}6tN_*beP1cGyFLj8)} zVsR-i-l+U}mu3FwL6lOV&F%Y@4QYPa+y*f3ELV|axAONZ2TB1Q#GSvtj?p)1e%Af6 z5DV3_p;H*SC7+QQ9CsaO1>1NYp~|*zz!hfX^lOg0zv!uBwA6?U($)B_>s3JEWa0qv zoe`+hG>KeB@K>_zr)-Dk;lO8q)$n~iv&zC`21b@$j8kpO3p%S#ieqX6h4+l{m{X2I z!KbrdIvkI)hFJstD8X!$3j074?e8b01=!RZkn2&;z|_Eg63{|~gnd}D5j;dda9L;% zMeg-pPa->ERWC1;P;vjMjbNwn1}o=$&xLi2Gk!SqmT^(`BE6<2|9u8AjlT|tct!?lrNb?Mtc(wR_eEb2b?Q78t1`=^ z=&cOP5fP90F?_DLwXS&)le8bJx$Ja+mbP(4L9n^L9pntH{?gW`{fFsRvK(cbk5Zo# zo`lPqe9=dN_%#GUrEp;09Ns;9F<^gTz#zb66{}U!uAs1gZQtvnT09PH`tJs1z{LiW z1W-tDwIh`-@{|)E0Qmr>IIhy}g105f= z|6|r%fMxNo3E7~^+Hq<#HsURI@3`HuHYqI%>#GCYr^jb&`#o>^r(eCE%(f}6(TEKO zJH5a4ALFbOJ?4RUFu~nC>_D=_moImHPytF)n|zR)JipzJdW!*{A8h<5m&OoU*;4}v>#}#Fs0c<<+d}@Uc zS^ax!tE!4(y9xN@V0$kH3}^7!FK#p&SZVx~g|0U++Js+&C0N@d-||1tZUnoY6KsKO z^W|j16}7}+`pCjs;Le)q%}r%O01;-P(Bbc)4^|uFb!QMY>S_+v+!ubB$NX zlMj$ARa;0`R8~}^jl);S@gVJ7G`3;c5Gs2lkhI|>k(p*a&PK=1JGPO(yzl*ZkC*5% zj=)c1hyd~DYy##>vD2&I4qw)poi zH4@p7;0+8RFcv}RJ+a>$tVMvBQugGu%tF(%LZ|dXkjZTRieKxXPiO;t;+ZxYDQDS~ z{2^){{giIXg#)RpY3A_G+`j6@Igq?awlrM>!E>6ctoz%BO-gbe{L7SQKSkI_+UnXk zY?s>CJ#d~nMzLYbvFAZD7VIT*S~YWN2sDHV^kG??bWcomS*Y&TSLzjVXzMo`S%S6D z-pb@SqEjmEFl&LknNGHKSH+H6kB{;4N3Ba&*cd67&|_Dzq6=!s4!qFUD*1h>4tM}W zAQtPgL%{tD&Hmmuv#n*i9>K3@VBMcQ*EBq5J#Kaa`Npu8|5AL!Y&%686jvlLgX;z* z(tf===Kxx<1{g-R7JN{a7Ld3Q1y(e5TsW9H;mgaA4A)r?%!(8vMz_pWHOGC;u&m#D z3>FY)l}JB&Irtl~?k5CnnVl^%UN05|UjZ}uhR(=u`^6DDz#m@gfA{3{6e+O=7D^nP z&?j2m+*f1Viy(*Lg$adMS_3hTfRteYnEgORSCfGk@%_I)FW0wj0WiB|R{?3+xfZ)C52q2u(Q0$IxOi71nwYSF$o)7nIK#c^H5ewJp^tPA|wCkUZ25JNWsdgkT8|^E}L| z2O;D2UW5#%gRVXQ*75$`);<2SK`V*QqP++J)d@PWK!VEI^*}ST%R*NW1PsA=O`WLz z2-4L+DOv!wQvR%Kebc{YuGHtIgl!WOw}1^NF#SzY&GXYEvCJV8zVr$XCZ9y?^N-qi zS*h?9GCbynIhb?UhWnBwrNHgtV5yH*Tr#}r)?y4cclO@5jAW;(ngz+ISRU?eM6R z*R+{7&wej&OG^eTvc#BHa_Bt>kc^wM?`b037)*G@{rq!q6^EtOgF8?0UIznE?Ecg? zhKflh4}?8IIS#(NH2B&*&H>&qZvdPefNbC9vhO|b1Kxpj^e8#;J@(`Q%r@!4Uvf}C zVAU@Yv*0FETrHvISKIqQtTVm`=|6;A!0bu`9klKs5YsSU^K?~N;pG9{53sT*EK1C_ zwp?=(8hvE=0qlnWPf1(*fgboj6O2z5iRX2oud=D>$qHcobpDK(g?FlVAfYDJu2=q{ z`$B;~{gdOx+#I zk)!|i|FIwhI9Fe*T9(EOu$9kjN1n(ZENgF0uqYHjzxTKao7UA)n2h_)!W_VGxovhI z@Y#TXr~bY9ZzQa3QJ8%RcM*rzaV3-if|+1F>F?7WG>w0^b`xpop95IEgR1{)EI%Q7 zQ!$2o1Fq#~K-AvjL>+LlQWaA=q_V+QqD^*1KWW z5sh_WSpQ;T_=je{Yt3{oG7ohei>VOp0{IT1th`RD9PrJYf08mOG*9%@UF5M|&A>`@ zQLamGp;iewJR<<54UOi5PH}YbNS7lK7kWv)=HlE1oFP;8M)cZacm)yd#=-mEb>l08 zPd@q?9Z|`;e%<9B?d#S+%rFM@hv+j?(e4?U^C>b7U%^H=l-miz^1<^RCQtVAfUuS zy1TnUI;B&P?m9F`9U7$L2uO=`NFyyJc(>o*|K7PXXT}-F(c$&Hd$0AZ=lQH;Dd8N8 zmjYszz0)M@;p!oKgAk^0ljzMfX)E+%RdG8fd)}XddA~{mHP4>ce&){Ea?20#DpHxJ zHd}<){oNRKNywly==U4){tJPwy`0o(SOCNf&>O#N#PtNZsL7}0gV_$gfW$zC(BJa! zE(8+CYoKlB6>*RIJY5fc1`5mlf*jQ zn$qp52W{Yv-gVbC4VHhl25bwG!8_;XuWT2IL7M*7UoP|sMrZ#V+vs<<6Hy`6kAuv& zRXd&x>Z{qe+Zo?0*uK@XAtNRIPVJoU*v64?6HLCiQf#mOt=k>FVoT+9***JBV?+3} zlj>E#$E(ucwv{k$IS%WNDv6MWj56Sy27a{ulQR1H+QAsk zUTV$kPwC&iyp`y2NTkDakn63iw7TA63?!Mz-P^uPktcfd=09~7v{@K)uYlvHTy}v7 ziynscO1fNq!1!C=PkNR0iZcSF=m8#J;W#|}rW_1P0o$3&+Z5TKAkE>|MMJm_1}E{a zd0l7G8RaFDGuxRudj56yo}@hF+3!~O2arN}s5zb?u=rb!?|@CUhGOGS2ZvA1ypWz8 zmcB;FofPtTn3eFoSk$bBK#4z)C+h8HwWXEd(eHYaGO7lC^b~f^s$(dJS@R$;l6VeH zx#K*2%Q^*?v6t`{>d4Rd1u633>3c5)ikff2WjtOl&SX)op-OqZf>)3wl00O|p|=(f z-B>Khj?}Y3PUwsFs6F2+l+CyI24)j(x#ys5*Bo#=1%-U~6ONDTF{WYj6v;6ccqKF& z^Jz18r)d6w3Ema=-CLW9P^B73N!4Jg0DvTq!w+w7RTqa_Y7nxn)Zm1X#78THQ!Otg zzwa}kq^1UrT+`g&dwX@i!M}Fpa=e22r{L9vhBjMvT;RleyZ@%A|Hujd3i`0)Qk~97 zm@6rLAly_@UpUDPngQe(!T<3e!J?;T4V+ zVQ`$*WhF5rI4ZR;#m(%vrotP!XMGzj+Lzi^>=E@%IzRkg5B`O-W$BFJ5k>36Z|NtD zn$yh;Q*RQd2RsZ0Q7*OPvN#ahbTy0lWWT&kxQg5U%Ghl!*i)xOf32Vzk|~{jMzLqv zP}8}i&H{d1E#*zPLkzqY9O2b?M}m0ZF=ng^dW(ExS}_CFc+~L^F2#S*N8%?)|HS8o z1`r#2B?GL1zn%VienMbUP)i=K{AbpjEDLi?^K8(+F8HnKn>*ygC`Tqvjw_R3QADd( z$A);;v=1hn;Dq2K|L0eVoUQQouhd=sKfc1Ad+)xbx=0L=y%svs`(+%VLO2orCec%S zjg!X*VkMx#PadheZ9KF<|MI1!RApTbn^`NgbXw=|@Y7Q?xN>M^Onq^06rNEkksD3? zZ@o$mJ@Yq@HGOAG-^p~=3q2A$wkeAlyJ)L3)Y`tE_xWADnmUoC>aJ=*Eoa0X#Mc#BMoJL+>b{?KjduYw2isQPLic(wv4JmU z0By{YqR-?^dro~+YrA@S;M`A<;pO`CFAGK1vPyV1GO`J2>47Z;NM5L8map&FK-|r$ zk^G>S$&4ohq%IB)6psFEH90u4AKux3WImRyzStGepTTqD<}w$hemu~!f4}C<3}(iPATOLMh{nWO`! zj`V>hprmvL(4#R`oc(}1V2rne;#`IyS*CrN(A}%OI^l0oYITM6BDk4J@i2GYGYbt^ zD-PifdNYTSSe(7*qBs$#1zjb&9)9^DCh1k0>1w)7^?)=1n~oZe?R5^DfUX;WeY0|w2NsMnhZ8V6|0&y&W`QnEDf9YbG8#rXfuNR)bE8A zN)2!#&fLoUzss3HX1fl(fMe;2dBP~VyRcp2Rx-7NOVQXEL%vgH>NX8yxvN})9x-l; zip?pLSZ<>zd3XUA57~?@JsR@lPb5Cii0Mtaqp&66H0>8{+a9*#T=Y4tTSA7sE>D00nPZp=S0`m zXxU9&I=ub68g<)$C^1>{hS{d><#w_E%rUDEi-A#_n*^W`dIHXOkXi?S1^4K_w3b_byzC8 zg0B9188wYfQT0h3by}sru*GDK(t)B+Er##YJg&25m!Z>ySq( zFzFwn%ao7YmSWSI;n$r_$x$WBB7D-8|C9HR3e#-J3LsF~Hi?0ZsJk$`zTsIa=i{|4 zI6P+AV4p#UjTqc$CNI4?2YlaCl3-#+{9}#f?LjXQ?bEd5{dM4w5`@TfE%)ziXc;OsN3znb7RR{T^6 zwD8vee{H>u`ND^9D06?dn)Hk8!3_9WKzgUf0*p>WC0!xU4A9CoSmOpvYaWNm78O`u z%3-N~a3c~Fko|}R<)3sZ#PSnfwts|f5jSP|iAVO{a|u0GT8rg`9(&B)R`kdF>gi1t zmh8#_i~l^QS4XJNz7@>~u0AT8@%n@%pnaDzf@82{8R{h3Vo)~Dw#(PrQSB5k$8G|% zv27H5pXEmBSieDw^Lntj>|s;YIbA(#+wd$*iW;_u*6`BX5BLL4EBDyrN7gEU~zYk_^cfNCEWyHm*g+`5-M9 z=f|N1LXlxWDFDJAu&BrEXb=EHaX3gFOshTXs8IZp_a`~fRm9Nke{p!=R7tkx`vFvk ze?F5j%g`pL9BYwZV6=GeU)Tpa01!mBTNQp&%P2z+Fi5*6z>kz_@i@qlucmRQ7+pdM zn}tw5U1^B*{kFAu!oQ8F%k``AHBqv@d3 zr(~p;HLKWPqj{ta=T{vx9X+rpeN0kNF7LeP8FpV*+~Xh_`erfBo!@KbvHj^3rrpUE zd+p_6=HGTS4`U7N?+7d#SmE@h%cWrV7-k?^bonS-uj^Zfo(d?@!U(LN0m`rKLYYSO z+$T@GbU2xmdj7KCGHKVP&5&(d-apGp*7q@)hiqfO=`#=*CxOE{E*1P8#w}ftO3z&i zXW*B(a2s7kD=B*&fTXINC3LiolQo#>VMnOe@1ts`##L6jF&K+Ir7*D>m6i)NL1JVa z0>WVQs~g6@FTYwYF5v$!Sv_S#08o)u=~)Zqg27Znz!FJ|hZp}ZQ+MMR4yLfcP!HV1 zs`U=#c&p3HZ&-fDkhNdlmUxEfrLs%2reNi){YS!hS@-k9_H}BCOz{V$5!dP457)%c9Id5QbiW`T^-F@m+0>)mkF{_ z!>dZ=-*D`qi|Yy9u#&r+)Jt2EkEw2M)R@crEtuC7m(9A`O}yfF(?WNcRD2aho-nmh z;jssWFo&DiZivn5%Q)l|tum|k2Me=zDHYKmXRB>CuDel^447guGHQL@@T4K!lA|Yf z^kC5`psaLCXoS}mLs>I0iO~6TmE3eYMmX4UFm9bv$F^mxWXsiqP13C)wPD8U%T;=* zVXhgmiDgSEpfmu1Uhv)Bw{=;4#~WJm8od0#%ZsI}&yE0bGD^yCA8M~Uk1yH(U&Aiv z!v<&m0-Zt+sDm*OY4R4^ma;U}EFv)j_4$1YC0N2asm?rh%_U;AIbSs9^g2WJIU)>e z5jf=gpilhxpkx$j2>Kj~iG?@Ki_a?Nt=~1G?S$-BYY>!-A&-%=}1tKa%F%Kv8IJfIv=oCdCu_s<$+b5&gP<4}Mx zjhTwZgLP_b@1*5%TpQnRMvHt*RC5hUoyXv*4%k>08C{Sz{#(&hi#k^&r5U(d zIUj&+p`OIIlnrItWd7?J8bWlt6sHMsNOUsysVM7fpy9GG8*Y^!8j8yDO5q~*pC_o1 z9-VY`h?3+G11l{svvXL-ZvsjYVQJ|EfQm6d`c_YHSA%kL+F1jN9FYEDYFxZ;2*)f* zz%;=`Csnl}wNLD=y6iSUJ<>gE?drm(86!a!zl@+@A;hBJE>wR<_{oKH)AZwQB@pBN zXS3VXNU(=oK>nIPMkO<%S}#U+b~)98^+;C;W88*~IC$9l`qE2O5lwmiGI3(ZI1{g^ z$A7Yd2n~C(vU09mD1{uIm%0N|ft8)aBhn-eEM;;mCuQ8D2tod29{btUPs^;S9wPB! z3$eiNOf_Dc7PsqndR2wevN8ozy50E8QEe1z{M1B1ozqn=nG>6wrBF#z!G_<-5Tdf! z^|ngTP%Z`${~=a6_eq3U^`#wOG+sCN!0EVeGY};w-{$;XvFv*@QwQ(j5$P(9l4|YC zZ!?KR5ghC5dRD0B5t=h%u2QM8^R`uA27pU)@hX_ubH*Qe*6P<37|K>oCk2+3N2t_^mHnTBe)D;3>=wCLkhg+hM}4=^0rR3HoD|*cLUbWjt5K)elv#FE6#@V$%hWrrdc%n z5m~QeYfu{f%$mYRj(g+FFyFB$QG4ZoK=#*_mnergp&ly>t&Wr>^r;2kZZjG<%NTUK z%i9wcE_Hd6SFqqG?Blb;$~UI;tin!mXN)me0c4;b3T;-=sl3q zatPoK4eKwKsFa_}RbJVreH_~cbcAoVvtcRax75>nv8FAmabwSvcL+et683%}VzZ?R z1}*Xt$^xqRzYg2k`NJw8VWHhWY+DrkUxN+!Widef5C1P`ZstGth6}3T1w!<;m0>Rx zm4PK)FrIaP7DySI?~5c+_-R>qaYB!#Mr_s51Ez~j(3ggTttL8izhd0p*T9inVpLQI zr1>?ZWV6uL6pJ^X(#L${v~bKy;-OOsLl5N7C(V>#iF(lF zeItkIM#4?hl4Q{5#ZVJcSlp9bev(FY`Z4Eri^HQri-)Qv9zA6hHJ`ldErux>R0+m% zV3~rL8Isj^W}?(l@v)xmB^Jjpn-_VU-19vw05y1fpKClhJg6QZ`|Gum z+plu1teB}8IM=9{tsl!M0^DiDqflW&q5I@iF))K3Fz(rnII$8 z)B(H^aIL;C6qsBBl=5Nb;t@=H_#g3YpBzjD0ih814Gh=XBnF5#xv<0Pw|x~u>OJW; zVZVtq6bE;ZpI{giWJ2pnQ&36jU+~sBWI?~HHe}MORqjXdWtd$v5uZOoQNuZLrhLcL zy!o$&Kp+K8udCeZbxKo>6w(kln5f<*<>l!@8UqgKdUQk;V0QeX6NBb2NPb!LNhLwh zZD0L94=)}l$|9dt4|57Bv%m>b^@+Uda?|xo(G}f~g<;{~_7NWW+vfPzrJHh*_Cuso zEx27W4Hp5*IC&&}w%I#QWaTVH{$>V=bf1F?7?^t5h_UDJe8Pd5MdoDR_ z5zTYO|MPNv|4o_qaIphU+QM={))aDzB^M-MA)2*a+us-6eyPv&NgAGmAmXBavhXF| z4lw-&MrS0%jX;~sq4mYI4*z%5a*a5#+V~B7G_~Dy<9Fxynql(7OWU)D@hAo19E@hv z6f!KQ(n%7*X}+*>8NTVqefX`}!2T6JoY^8Q`1liUpK;84cQ4G=G#%pS8D!E-C#_Op zh~L#vr3P~|!T^y4PSFaSRcYjLvkZ1-B=t(2nFAJjJZI(A3;vCZYggVBK`LsRN7klzA=`Xj) zc8)-&&8K(GC*T$fqzP}Nc&CowO>U$0Y+J5V5u(W>CXg%^@vY1BV~{Q-Aov|ZV@|zA zLXK($5R4_!`Tancklh5P2)?(GX>DcLVHdo|zN-|w?vN~szyHVOdjI#A5fDiq-+j5i ze|h+enyZ>^Mwmn2c*Zfg&k%JaryLPRHNIWJIz|6_#_%ynG1d=|5PG&Ne~@&8wzDll)-wF|BwtYi1{R&O(B>QEn6eNSh`B z7_)o~N+jo%CLGbQ_6AFypTRwTTQ`LPV4KF}; z2dxNcI7;@NKhFo{C!9R5J}yYIq{xzDpk(g}E8z;qGu~D_&X0QbBqU6=HZ^*Ax6px< zK{P>qL(qkEwP^)TZ&z0^WyLn8VlFF`V4lVadBLkgR`YIW z<4Ud8yQL;>&wYGFysCVXiDm5R+JYRYe{%J$FCltKo8TT<2dgeabIiQ zpU3`i(LdGWAa*N(LkES=Z|Ei^C*V{)7bkXxjEwSrOpubzQ)LlHx$M-r4>lb|!SB^h z+Tgd|ltUV9{@4Iv45^3H@Bgrs1*XIkc_u>hQq0)~k0PQop8O=8I&I>RM)3agWzBEg z`x;`2h-@7wv!!yAUVwN-^s8&k=1E5^#_^Lo$H%5FL?M zw1W8#QK$%3*t6~P4AcH^Gf1q-BhfkD99n2d+jG)|_3O$`+;XM7!nNwWd!= zJa%oVd@z%R;=DvDGbXyp*CHSG#<$Rj4#KUZLo`Y4Y(PlDuNn0_o%6G{v*hxvupn}W zIG41;%|>i^mD`)e?h8=n&-3A^9V}?tEsBVs>l?z&?ic04Md_5!Q<8AUj4)~T!7xjh z`ia>K=$?>77b=J{@R3JCU2ErOS+eR&Lb8VJA!Y_I zZ}lC1C20Z;UbVa~;1??-{z}Te?I%rvzu!ojsB{C_rP_B{zW)R;t-#CG{6%^qhS^WW zW@KcA-yZk#B4wuN5sUm(A02WQlxXkLp;iBLsul+kpedP_Qfv4pmsLfAV7Fqp zt`fi@BjC)NnhjO}Rs&<~%6qyd&$kgY85x>BnibA-t&6eg&G*2AgF2wA4tJ&xX27(_ zl{=kdU-Xo*`_$N#E&HQ+&Q~Wn5YCdF)LfKdri$yMgPUIYB~4{r5`Q9yo~o;@#O}sQ z8@~5uDj{@?laq) zI=)L8+3`5-LN?_K-;|#@IIPFYs{&BWn))lNtbM#md+f0x4kPZSPe5w`yR;Rw^A()U4sZ6uJ>Oj2chYjHybftr;KqaXxEw zKl}fG^EE3u2Uk3V+3OHLQJyY(elUCbUSR)HxX`hBdA{9zQ=B%9(a|-hpGq_@Fz2f`qypsa zk7Jhj;hM!BF&%NO_;O=#FAO-&BQL0W$(Pwo{`c%^FIx>*Tm56^D~#w0joTEdx-wj% zi+gX5(qux_mSZtKegKc|I+dKYrUoqC6n!Fd~a@&`XV$NQ%=KS% zQSUo$rQPcSWz(PaR%=FlE!bKl)d4OYdsqy@~>c;QbM3Oj+&l+pX?-@RB@7-7B3vuEhi=ld7JFnuoMBE@zjBSC8jf3IO zP3+7wY4vNA`ao2alb)Drww^oFr3{6VgCQ9^=d1d#BcE^&!kpiB)|3x@nop(jC%9S% z{5Y%_5N($@av2Pm#&Tf_^4T#JUS-vKK9zI&s9R9eRw4O%9#LWOWUI;vg}7Io^wbW= z(T0pFn{7V&KdNi64G)yaNnZ#kx;)j(rq95_ptlj?gDN09zvJTRP|MdH19Q9bCSPc& z3#+~`M2(!+e37!BBt&B)txQTkdTM5jqZM{qgWc3m7pIA}@dq^sAP1Eq+?j2S1a1n| zvMj}oKA4ayEC}$ZQ~(I~V+xH5t312=h3w19Sx*;d<%OEP|2;Opv*hl0*-)PvIpnSw zaM^s4N1G}}>Yl=RN;yO%@>8k+t&fY)MMq7o@HQZPtd4^=s{}O)gXmklLABmFl?X|y zOG#xWk#cvr{_yo@ye3Nr9by1%v4^NwY7RBv>pY3I;+A$8Z?b2?*&&T{x8i^3nnlmB z(x&|6{Z=BEStt8PTa$bN+d-AwLg8)m*YM3AUR@!3vsSB?+#A^jWB+Gbd>1%+e$gMQ z^qYjBuz|hR!4T~c7{FQN$r+DRIkD`h5g?mUV@ z_*g;)(;YAMkU)Az<4e7#(?3pYJr&BnJ}sT}7F)dHrQy5~QcHPn=8zn}`RHW)CsxQ) zHFO)lIGRwR8U{}6*DrsCyk#$TUoz*>(KKBQK6X3l`Znkc;N7(k`?cZ6X_^n3i(u+i z4RABfla5!PfhE(m6*w>3mjRbDN>d?PT7RCv!>(o*CNzuNxQMH=v(YH{O2vlwLcgrr zSIoanlO=^18>v})>m_Ft{J;hYnh`=c=B&JPDeO&VgU#KVG{P*W;tUJ;QHS!)rK!ax z6DPK0wyXjPW2S4r+?r@3#jYUF{%gK%n-mCbL}Aa3$AT5?#J0`_Dj|}>A`PHPSn2@cC2I&w(VkO55%lD+RtmrcQineO@!N8%z9-jFYr)G{T!lJI1Hpa+}w7>oO-gjDg zA<{gVOa$XF+eKyeZfc1^M{xMyM};8t2p{RuaDxs-S{k&vhUy;3LD(v~g#A0`t)@j3 z$I~!2eK>!`P9kCPz=X(Z8i!4d0yTQkq_WwN*LV0BS`oT=1G;u<*r`{4n+$ocAt8+$ z5&5-R@bPpk7*N#+7ESt2nAa=HbcL#q2Ck2XzE=Q3eX1;JcuSklzjXpp25pTb%QBFx z*QCI8kqP9yNyp~?k*&7e^u__GlJzrXf*d8iO4hpo*~9+UL>Z01H#RQ1O!b9xgLdw= zwM`s2ylCvhocm}D2I`P#_L`1?aNQPBm4@xrNmI!U?jdC19O`qV1S@@49GETVzBRk# zAE7Afyn*_b*#*kGM%hLSg*>Mm_0ko(q+o)L)j}9E!Vy)1BTcGy2>neE)~@jPH+!co zwtmxB90yA*2z;lTAJW+QX*3&Ohw4{(br%ii&PuVSZJVB9gnBDxGa!q$Dgh2rOs2iBm5oB}Wf#8j;D46A=h>^n*<= zF*{8KsVslk8kRQR=Z@7(E6T~t=Z^!-VTE%zvtg>yBO40>Bu~qmY?Xp53OF_hcbJlT#}xGK=}$VYcX2Hj;!y$s&+vf94r%DNOCAC=EUk*Ij@VV&6Qv+K!y z)%U}sS(|tI>RBclv4a!Glh!obNYZ=GrmON_34(b&0D9JS1(J6$ps|$=sh#HpWSSFx zqjP?}FDw6r{a>cJecMw6n}z_Eks#dwTY^5AteKcD5oWj#n}ZFMI{ZAR-*WZ-wmXc} zXOa-tci8_!vVH;VHOl6JJIAkoW|Yt!V-lSpY4ifBI!**qW;1+Ot;uUycT9>+A*6yZ zW#os)GyZYA*;(7Cp*>Q*ct}-5)-aV~>0|~n>#|GRH;5mp!F6aFVhW#(+4Nc+G1^={ zYF(@ricQNo=*t}o4H2oA4zr4uHw=;3=%Et) z@QiA6DZ0o5MTPnw`WB81+{(pTSA7}{{lqww}|X;*5Z9(k%|S8e?@Fn4ps z244K7+s~_|F6mS78O#0Sko*}h;xSC3vKJs{^HFK%Lx={^vT}9&ZFLb+HK#H&@}WLI z>kt)#TC8Z_I!$$SCrb(r2ZyfivQNg}MjP27Q{kvf+lQ5wRsLquSah zJR2Ib6dn@qDr|-p!p6E^*$r4^FeCKMRIXv{HaashHZ{f;o}OndUB9AsrlDFJm-c%} zgJMGtr-AjkIUT{5qK=>`y1RP%lVE57H zNZlMLVt@&TKt(UQ|vsIT&Uw^3Hzx2Y-Olh_KOq`yZ1I@nSt|diS{+f z7xl`#y)7FzmTK@B;aV)8Ge~0gQN>@S=DnM(BkOb+va7;@Jx$gNf9?#8-l= zTMd;}AIzEE=S+1@GU0~xp9hZdNpLZ=+1#z#>1YIS2=*EixcC;h36F%ol{*{4R6?<* z1S8K^>?Kko&Dr&iS`s%!Y__UUgW~;1JEE|(8ayY?Gw|6tJ1%Y_^~j?ZW3vCsY`Mw| zMw9iZCtK8iVn%8{C-p*GX#D`ub%^-?;)gg9Q1|xNpAaH|Zp(L8AEsFu+#|sv zVk4k$qMkayv&R8k#u!sI2alVWO4qi&?o-RYWacXRL6$e0MWlXi4;kneraOh0VolFb`u_$z!Vg? zEl;f;zUK9$1>U4No%{=IkSulD@sV@Ux7b&LJa(Z$47LJBD91T+8lGit+A|TGU*Y{k z#rFYkL`Dqe2^fW5xzN+UnelSJ4mwVz#d>dr@`bt&sv@hqjYLXrQC87~jPjmeZ}NF1 zd~;Z4y53&W;B0=!xGH)yb+)<)Te#a0wl>GxyLB-3Vj)Z9_V|>|Q)jzf$6h|FQs=JC z$Fp&nRX)z?HfMNZ?BK`bfw^*83AV;$x6f;s9Z_~a)UY*sa6m@wY01Em{RpZA9-c~r z^~y-T&uV1?BRl5R*LN|Bne&rTozc3t`-3dmSUcm^jmXC_$4h+d08RqFa&ge2b#aG_ zJF2>Jng&wU#P*s9u(^g({*`M}bt9OYXjk>Zj~w*o!%s(%???hqN!Hsq0{#pIQoj0o z{_6SN-@u27ze4{}V6XK851a~?9si{;^iJ2G1PU@dL=FBPGN3~9F%b3jWZPvP`go+f zPe`L=cA6YzHRU&iBe7_ZAX`7zKc@25cspty)`RxRqmtMChT~OAQlBW&XURveAxI|0 z>>M+xKUvK!cC63tpf zOH8XQGz!QOeyn`6dM19milod2n{=#9PT6o5^e29O*G%Hw{ijnlBiuSJD{!b%DwSr2 z=R@bSuSGQD0xohBV?BV$b@(gQRIO|#38&WP;CKO_>WRKvK^G_arqDB*nyx1ZS2V8! z;RK0qJH>QbB0_#J#*t_@9-JuQog(2IZyBGIkiq(=UIZ^o9e+>nRcRVx_aOlj#F3{# z0g@J z!{FcI!s{f?o20bUGhoBRK?SQ!NE3!Yf8>yS4@uM{PE>9Yf->}JZ$q);IXF-S#l+vD zS=Uf%g;`9pr75bwlr>f*B5SJft;v#YgP!4bB{D{hqz3a~1{e&m)jG{(RaxdC7@X*h zy5H&>bVdpee+%*xLd9?9ME{~&py{-IV>P4k+XqUaZTv9^ScAui=lzn2{i#q~4L(7# zp61{p-3O=zwPZj1eB>D!L0AlZw!iS~i#Ju4bXKF+v}Ky!MZ9+S61p*0;92Y(iw?_d zx>cajBPqt@r@Ga4Ej5g}SwNkoP4$pZ{X*jzkmYAR5>~2wrxOELpJE{F5_gIcX2k5)CO-m+ymX<+U9D zKDoaF4;`etxNyKm5uh6e1Hun-8DKrPx3`0J5rs9g;J5w#<3NFgL*Tdirb$-@d{|@{ ziJ=GkU?UrN8Nllo9;2?pl9C8^(|r7jvg}9#KAnc|mhwI@djpmYNYk{`eC?1MhDdj> z4jFTO@v0JmSTyK>JP-H)c9V3&4#4XH@bP5ipUj(`b@2Xy)m5Ss1oLuh9olT)wUxvv z&BOQjylFl-oagGjX-_nE;HOE>x2ZdOF*})si76${;yN3)Y6QM}Jg9R9<4GJH9Pem+ z8YVz#$10NE>npZb|I0s_^O_P6Nj;-dRgfOCgg5*)(u!-4RNdYI_UiyBHIcw3u&(16 z?Bj%ONr%MOI2a(AkE+OkwiYUh$=4DZe`629o*(Pw8b#W=>7C#<)ZBLu0)A4df*?Qf zC##%&T-LBJfP&t>3_-+F&ka8&0q8yUc`1X^$`s;&B%SU}) zxV+3}nzQ|KruAHCFg0W0l2{H7h8Uwct8Tx5M%=0@jV$Gc1rH?R@b9FePP$HfIKLu^ z{h=vI|Ee~-1xb9c@sv|(3W52@RFa)}__GNqguo@tl$SOmSqSx7#a<#dGx*$YWZz3r zxME|!g7x;Mwat$w!7e4i3(J|jzV0|vG zOlfZNuT^k{G(KPV}f5mgvE-OrGqe-qnCdxEuI$o4i7bTd2;h7 zUMlPgydbC7rv(4*lyeyd;cAcu=K%(^a zINBcSs1-l8DfX#V|08Z^S}*s0`LF1&=^zas%=cxsB2>FNmQ`cKP~DPo2F9;WnMILY5&>hiR>2 zB~=}hsME32CQD6KdJzup>*U4Oi|#vr)R?ehI|6YzDX5-C9_KmRs%TQmnOxK1&wZyG z0|k>zVSO~`7q#WTrx8lXC}tkI#GJn0N#pF13Z%;KHtyBrELk{a(7B&Y6hyy-P7_lR zA_-t5Rmah>6`iFn&D<&;8AvdwAw8Ew$XNH8kbqCPN&OCkm87c>t1(!2%QNpK3Kn!N z@HGWK^>SkYn=a~*CDJYB=5w}yoI>@gECPZJ%-Skujfv@}7%+VOOJ?=I^J z&WZlycN9diMIrTBK`yEonfj#xR*Olh{dWyQWsD8z6FZXQ0*bds7TpAYlrJORmjx>HrR-wV+WUwXGX#EOeBRLf*T<((pJPyN@ zipIp6yelszXLZR;(qh!_Ri!ug$<9v*?2oq^^YfbyK9i(+}u=^74Eq;5cOEuQj|LRo^}EU2_^6h zd7}SNs#Q>Stb;j*!B>X=e7jZ>o3xoEB_~GX(g*l3l6Zud2&i&PVuJE_^uia86;p1b4S{p>5 z8n5sD@9n2Vu#9iP_Y9W2EZ8qn{)}{2mjnxzR;P&_n{vNlu_&CU=$FX8K*KY0oV)aR zmqvL<&3Y>M(ao}la0_kikIPISbgi1l)MR)E@kB=pGtbw7yFiy??s#m+wo+_Pls8RE z-S%FUOcYcXptKNS7F(tC^&P05Uicr#+^>V79755n0f;qVpDgbV?cU-T)tJ}zzTnJ? zjAxOnGV*aHbpocTq$gzpF6pp!dtip#lb07QjKm3Lx2c2;o7&I$Yb8aLSX7W}tiH+J z*HqlPrQ+^z z+S6AUw+!E4aM&-HV2H$qOy7D9;dNYu#C7rrOhr06-AZ)!UP+pFo z@A!wROvSL6VFyuGEZyk&&)!tA&6=lh_8{UL#c;o#Spz{8Q#w91UmlxSBc;PsCn>N>~I7bms zm20FX`j5feTSm!vd1p$!YMXIa5eQGCRJM&Wg`lk$$%J=6o1O zbY(psi>YX6s{)sEwzKu6F=fY|UkZmc&}YY@e7Z1tq`ppVFNk^gm;;fe3Ag=gTWfV) zPxh32?`LM+!aJ~@N);GMA%8XQ9;e4z;z+bLZyz3>t0w3_L6qAhF zN%1F zS0)nOKmKjDxfjOTt_)C@)8rpAh zOCN+cT=^n1m}5ddIL!zVka0Okr}E=$QQ#(^Fl`KhMd@5b(O5h`sj+qXH6u{V&pCEU zaDRf4*V#%AgPO^I=~hURw@vK3XZFVpl*R=`tpozEJDiS7InQFDsKvCTYsKpKn51!$ z!D>TiuxW2|9`BC@Zm^p9)JWD zkv=!+gR9|mB-?ur)y;7Z$#FvzUoqPFF$N1v%*Zt-Dd>6kajI5fP^@4N>VX~ zsCc6}Q_qIT22$j?(AFrX^xD76sObHhim?uhk5ED#NVWE8uZy|FA2e!>rn#xEh>^?> zz~1pFl{NTSL_~j27G&mi3Fe}C;Wu~FyesGLedavoS3g7jTtC)peYtHad5@Y=NBENW zXv8hMa`1Sm6?Iy%_D@aozo$S|LyrJiZQh(s^+@?f?(lCo$*9k z$nk+GZ@}t^d0FcxomAX;=n#`YK%gy@j1`6IH5(YH>B9SWu`}&w<%u(0qV|U2mesSL?3v%Jfs>o1z@rI8NfaD#+hcf=$Abm2Q~Gt?c4n&L3o7vrIOB zqy!4=K*3zt_Bk2v>yHn8<~r%qN`4%IMUeQ!qhgy2LwyP2F&M68P+`p2?!y$4%os}K ze!SY&AAax)+@36Yy`SYAONTcWh0loVMti`5pU>aLC`D*hR7A~CrZqm~)Q3h1Md0?z z<3&Sp_x#YBQF*>eG}G>QhgKkwoFc5HXmm9uO$H9wx>r_m<# z{M-~Pe-I;T!@1EWsJGz8dtmV`ag6mQk=1vg=ITo+5M0%?SWX^^jH#=w_F9>sUry{WQ*Z!1Zmo4)Ka|+B_#|SRlL)(EIwC;mwO*1RMw7i~pz(yd)O6 zsgI0S9lr7{vzX3JGZC>jSRY%2s@MM1P|w*{B1U^wTvGR;UHVhe)RG)_k^QW0isVqk zPVw*OVw*oGv+3y-)`~5^${ove_#$_?NT}AXme(yX&?gRPPi*TJX!jG`H?2~Vn+w8SD z4@k`e9f9}($B#>{wIkxjZ}fX7^^YUF3m^76Vu#HEvn-Rqr*BBZN%67(Mjk2YeUo3B zD5YoEqXCzW^aL&>Hey}~l7z16sQr&Ha!Gt+2!)Jpv*Ki1%Hc-;GH`z$@g&BFR<&I} zZCQ==*_uFROO)3kO;|KB z%8qzb4c}3QcnOr;7{k7%U(=QD(mL9vjeQ|Q3MpBuQky&A?aT_pdwV8)vs%>Ljo;J# z#KYs`U(3ku%h?@^$kREeLczpUw0;5BEc^!4sufZBR3;AC5?Y5GKSxg%njONOc8T4A za&PNxpUXzW&+<9;Wo+h`6a7riwpZlFf`_&tb4<>h#C-bJT)P`!O<}?HSiA&b1HZyX3OzNaJ?7 zw8*5J&03GkFS@bOpXbDz>!x;g)yThF;j^#zRz@~N_oSN1+qsKF@&0r5sn%&__i9VW zi^YP0OCa0FXt{Z!!3o95z)RKuA93W#e2{CqLrR7-;M;`k2rpC`=heoBJ#T+SlA*+a zN~?d_q4?D}40{42G=6y8|5a%#^VB)|_hokb`3qcAy}qsi6P;)MmJN?@{cU72(PpMv zt2l!)t7ICY{HN8i)l$~+D55gxXp*ZL{Z3!7PPVve_~T&OEQ=1bBENOZpX(w!iH&%* z;zCI(tNOgGE1eP9opi;X`<)LiikTSicG}hiV{L{_PHIlYJRv7%wI?N}2~&6Fm|k0w zR7sE)FPd<{%jZLOOO#&(Fx_(2AHB92q4+z7dTIz6>Gc*9&6_Y0{Uuc|EQ*@ffE~Ep z0++JLDqQQ-W&?z=6x_-x?@EV{{iaau2}n+$x%ie%&^xsQSLbCMRKO#ia9E@kskN!j zXEvOP5u+w>R?G4Il08{p#o+j8=-*10p8P&P%G;BBN^tXR9iBRLSpUoAp~dyuVPEmg zLur#I$8ZFd6Tnha`bGHAr*$9yOj})dDL0gGrQPcpcAFC}6>DMC)Re=~VsVB?u#Z(C z?Gn|$ETFJ90fhDU28juH;TK~r31yN4=xfYvjr^}Y43FILRsSC6a6%U8W}s_r6U573 zcLNa~Mt{d5ZM&Eg&sWoP--QTTj9=Wb_uEyWI18`f;Ag#TaPqD?!^S)o^q|Ic+S`7k z(?|=c-}jc8Yq$9hpXj&kw}+OxB@UDi{dubB8L0MXcKe(gHvJLF=SN~GdM99~`@M<* zTSS*8TcF%zEEXR~P?kAECnJ3tBl-Y$RGea=UDR%AaJ#OgzdJINXAPWT5Twyht=_@p zz;RL=HSQ%_jUk{*Gs@P_G8(Jqdfv(++>riM-t~@{T}s8`+IByIx*eyu)pWkU-Us4~ zfkewghS^8!56?1Jt;|i6Fc;%m73KA7t*hyxx^MoEyXX;~ll`-M|1HOJI{zy~_tjXS zQMulu#vLu@3s<2COShIkI(#Sbb6q#w9Y2tmbXf}bTz^fShhN!p61VNfLMBh40y@}F zQj$o#+34yV(Uw%*VU|m0m#Cm@T1Pj@B#w{A<6a@psg85N1&ddmzxP}7SxQt_b`|m4 zqU`Her9vgd2%FedeeddsIfABFmVPJVQP~d2)3?Q+`amqEr=mYm6eJ3~7o7{e5VitJ z%05402Xpsw?au~Mo%qcBT(H8*ThpT8tPw{=}Op-HsahKK|bS zIF!R<-p`)^7Rf4deFiX>?tTFwj|E_$PnU_w_v`4KlQd0GcK#%qZmr0vlIii6?S~P#(&%Y&yx~S47i~k)ohKwS z5QnnOU zQIqg;ZCOT9LO<0?#;LnbVfr`pRhl5K5>1db*dT74E7Xta&YVQ`>9C@0lS^$zB$S*qO zXH>a?A8Gg-j&iefNvh(J|Gf?A`HjOfB4KPh=O@5g(UPn?zdpp=UkSb|tu%IE=n~}e zC3a%f4_t~bgBhlc!a6BEr70O|n+dOf6a6d{4H810ekyrlw!=485HEU6?<09PGc{lG z(eFx@252;+vXJGLCmt3@pE#ggIB36AXYx#A;yTZC0{=!fBIrKwd6^T3iku;5pf7Yd(B<1ad1-Tago|9&u=e@RI zAUFT4SgH%G|G$I*p4gHH7tJ5I3_3(98(jFbagZn5p>ARirhHnyMeh$uI5df598SLc zLhq?`ukdYsNyRh!IZ5;5hX}JfdM7#vK~ z!DTIi?-FgPv;~2n=E!;XXOlPiyO6ulH8%~(YrJ|TjnoYrlnuB@G{j~MRDs9+zI@w>vmjXa=*Z{AQ=gW&L@WN0Z8@pE?A36^%{Dv_%`cLvP~LF4b|o%j zNCW;UUyU?gY>}YBW2M?ck)P3Pa1L|98ECcxL#gn(3Qg7U)tU`T+rmubC;7dI2I|rB zM@f&T6J4u=U-&=m@_2Qa#9XnEZJ_@}`Z}6GwmcroBFmU^gjuYknyrH?P%jmm_gQ@M zr;Cngv#)W+A%qx;i5z(zuAb}@6%skOsg$Ba>5GtDI}7g1>o47b5XoNT6S?Q^WfV_w zfn*IIzK`t5Fe8r=oHVd0DBBbb{}iNp3aa4n+C&h&)EnIc@FZRcGI%8Ep@7+(#Pg3c z(dPrtYWWG`5+k8K2mCwcI=q_e4~rcprHQsf@M_=qlXUXcZY4lFzI=h3jWL;*2)#mm zmsRGc$Ohk`?%np%X$?@kW8HB)vesrMI5}({)c&{o=b)RHE*;T3>IKh051O}laPR}r zMqe0ofT5ZzVeU-sa6R?J>b|{4kWjpKlk50uqe1M5)=w|QI=*Ft0l@bR}G%eis zRf<#XpN(@XF?AXZOGCL6$ameeP6b3PupNqKoLU$P?+AZx|LFXrqU0`LA|9-oHnJOR zBCYt{gB8VT+;PKwCsR+0j{sC$wH=b-1{Y6Go*f|yVHS*0S$5>vZ=0Io&t88lvUicJ zy|#PrDT4~CvS|ECCWNkVNkgWNJkM40Ep2E{Ku!ojCBVK1fbsZUWtHfzY|J5yW{&9Il@@)k;`mOAs?vjk8X=%j?Y2B+i3P z&{qcZ4O>aNw0|3k-EP8dYDf$D7#7n)7&_M6TXuzyT{YW zUkVmkMEy2wfJ^Y`25b&bCPDoo%}(r)g+B}N(4jIcf3}0M1n=ZvHuxllzO86K9v69f z2)Qe5_F|q#h#|+WAYf@G;k)Yfxd6Smog^`dwmeCZTNEf>WS) zL=bNogvGh-m8D_cmg|bV&8vD=5y5K_AFPOgyXXU&I+ zu$*`x*}3NYpihbdg>7kbwN~W+pb}&(-bX{b~m~;* zHtZgYUA*0OTilGAgZ}6L!ReG%c@p@Hq^Smr>X>p zH^%szafw*LH*X_`sGh%hPF%=c_4@Uk$VHizCRc{-U8=G#C7KXBgk)q=)(8T=A z0-ci&j^_v-=yY5hhs0L-+~AVgZ;3 Xi5VxJiJmoZ0v6!laIQ6NG!b-(seUOlUI z=2Tg>P~;3n%>oiaJN569RT4lJ*f5J1?JJtP@72k4B9yaP=fC*f+}t{dAAh`H(u6Nn4Og3ODRXsz}uK#KeKlg=0MALe^!YuO(h-A``tE&6{4UXX3KjUJ@H@R zqfL>z?6Z>SzR6KTLO*ofYN=_}!l4zHCPKxmj#TZ`87=2K;a-E=aqDDh+X@_8~OZv8_RRvikmyAEQI7vcZ?cz4_e<(%3tPx{=@x#39Eco_pl|T3||BRwf z(bq_4*MgRkf65UYJPttD0-Oy1GheE2i9X8mk;?!yKKDu2-PlOh-pL*XSuz2hIv_WU zasey<=iLzGDx9msO$JCTW<1j>NDOG6>H|VRhGFnki~HYR^0sB!0XAh|lN#QD&yt>W z%V~;dpj3Y#NSO$LKR=ScYIH&a84Zsyw|t%E;pSg9Z3kr_ok@p7jB@(g9_4aTLZr0^ zv5s7m%NWGPVr#!6u}Vi=1d+*1$j-QctwEhFN&l|)gUfwPNza>@NJ&>3kDGX#a(?l;rd6~)GP?c!QX52KFxXk z+;>U-;#x5c$(KK&_^F2*h>mZ-F#b(xqB`mdF)ex{%Je+KrXBO_htAl($HoY|`=!$P zog0Fyjayvs$W{5L&$h$0^&ylsO4#~GX6J# z$r68Yv02Ao`>{M10g~imEJC^?gE8}>Hb@saDp8iwI$=r54)=!vc$2%lgs~$gU`1UY zqwh{HrzN1U5sA4C#=b)*WR-?cKvGwYvQ6~CWC<=~`{j=$>P;ptWs=Ehc`G%u<7VEb z1!X2-8p40#Q-)}o>eQR&(#tX)3LX`Mx_IZmiVr6^V%3=_)iqF+_n4q3ISliH9Lb&3w? z&P!h9_Wgod(0PCP;Psq*;dx%opFl>g3Sw}P9^EyPy>NOA;~on6^1x8%X{dkoMK%Wz zMwqC#IrNNbwHx3g4L?#qf|y$Z0&qa;wBXDd3f>(Fr+@AS~#ysABS z9PN4U{Av&mHoncXeBLVv7&?Iua^i*dZoqS6?TX?x!kC$Xb?pDjTJynPso8tBG-Z2_ zFw5v^*6ux2bLn&!w^B$&UQ{?Mg>u(_ez|678W72~X{o4H4$vDa<;*_4y!t1tJ!-!_ zxP1Zy`myx^$+eiXm4lnbc7Qh=1!Sv)pATeRz4)hYl#!F`+E6@`h4=ZuN!oyQgYRKo ze5{KfP!ebqC=jYNd^Z!)TZUgPRe4jlmiwEQ`qK{lOg>nid)!g?Bil)WS1^E|TDsp? zF}qH%!(Bt@aTcp&f39HZVJ<00TME9jWHNv3@*Pq*mTCRUD!x8#XTPR zQSV>47^LTL;6+R1)~d_(%ZZ@{2;M>(IT^cLu=T0bKWQ9`%3IQu3t zRp`xmzp4Hr&qWB5&{}DoKvTMU*f-3@2K8JkPc3&#VViTLggq`0sMGS@WsL#fzVL0KC$;r zhB{lQ=1obru4uk>4@J3qyGA59ug*}!^epi}9+&1BaQ{RA3{6;FSO z!Ffr%nQzJpStKc%s&9$`Fx%Dze-vSX(Nu|p;vF*?V>FLV>oon$i}3$Iwfuljt7Ur+ z@?C!@XVI+HGO5()NYG3!p7|9sd(j288FK;&o@fSfiYDHKNd;AM%cy!n&=8^xJR|@0 zrDE?Y*%Nir8tI(DE!oZAGLq$s*iC~hq`tz-M}xACYceeoR&$MN^It?IncPq_um4cz zP8NXjAHVtY^fR#Zgm}nB&6kvdj;*6EtULe0mPGsMQV0Wv3Lk|`n8O=diQXyA=b1Vx z7_bs+FH0kT7HLAFiJ^jyoZSBrIfO7wZXM&4xjz2od+8O5yPC3zVp}e(9iH4?ktela zAeZ@IE=q3Hlw2CL2T1>sB(9h*bF5P^0=(byL&41VJD<>$@5vC$qr!I5P#Iq6y4W#0 z9&wi$sTCP%0+;WJXb$c@Dod$2wU<)RX>3Vj)x?sFNzD$uEcB)Dw zG~AS{j_)l~+|nh12Nq`0J1ZEdsnV*@wK3xFi|W+pz2W{h%w&)=C`gd}ND=jEXj(Q& zyEQE{*u#>!q7nBoO^;5hp)qV;l(f(<10m9a+Az6?%D0s;`eP1$_&3Np=%aB)WD1J$ z(>4zKtg(GKArN(ieAUGBQN_)ag6-E2Jk5@a74s3;$0ZbgA{pl{Evl(t0>Q2<4{$@n zYv0qlX%!&0c*8&`8hN5fEO@SFdk)O_qjaPWGD=@AyMv}qcIg;!>LDn-N8yc;p?IeL zm^nW=6(|bGMiq&UXct;!A9HSZHwB-bo=52DwbO;qXX8`- zl8r!$WZHIPvqib08Hrs7-2g!@9$Vl0owVQe2qZqhb7zH~rQh8^48HlaYsQ~=av(t| zfad#&0OuoY(0MF(43mUO76nQ6>`Vig?O2zu=f~NF%A)U&o{D9~yXhaS+~TwpKkU0j zDqY4NbY@-6e=xp(MDdTxsPcM;;f&gxX!e3*OgjC)WcuM_m{Z zjHugsMDh#L2G=^V6iu=Q1+AY`{E5ga=S*G??KI2LUT?kkIrshB2KKp4$mShP0~G8$ zrd_1$%xekorZJ~|9Lw=R%Wl^n7JKJ`!5O0;A)f0Zx7Yf+igkPm3yBJ97s6=}$qsV9 zwqNBFLl$I9eIl=9Uh?BjriKf3MI@eYCQ)Aw>Q+8=;_sO)>EZH;olc>MD55oeW|o+U zd5MrIzHvXK&1bO1d3XblFA+zPEE%k~F3dPM|yo79&6~ zmunO>eSs^ms?p1mICn22e(QNTU&<;!(TvZUB6%8q$M(GaW-Q>mJ2<fC?nHN@fWO}jrSZ*Q&X1*^ui^5!RZ4uMq}Mi~<@uN`NmfDrhy}@R^bu5IGpqV-dm47WIr>okClJHaAoi_L z{VTA*VT>a!n}OQuebrlYPMuX-A}uNnPCtJ4KB_8x$?j8RwiBPMX{)?2r@?u^+z!uq z@Nq8%$-or9AyT6C+K|Pif5HXB7Be|M0*P~MO|SW=nnoY)k}vYv^giM6kjdC_?OKc2 zSjRte5FEXZb&hKguG6%wm-*kY!i|7Lf+F&BdZ_N9UsJJ`jLY~v* zLR110PeQ7FY)|grG-u%?o5E9xr{%*eLNhYDrnj8JXXP>$e;Y{s)$+$jaJe2rD`gEl zM64}5Do$Gvf`arv$*=q#x=j$mF$@+;93roa@@OJaT0^)ts+rpN;(_v+mU7ou)A=v# zpq=g|298WzonI!7j)q`}OrUyk@tLcUBA=pEGyBuq>9smU@wQK^!1Ee6`zIm9X)uL> zHBQwNXGLovnWj|V1g|6&V;a_egae#=&kLM23Z!!f5+D9O^1Z%XQ9zfyqHnZx`p!iH znmesc36mgyowTNb>BcA>tW#FPyWQ7SSr!)mT10l`wyYP0ObGFFSqq$MunlN z{&=&^!@G8Y=mnEC?7f{$*Y1!cTw)bEY(w@i1;=rsDV>*@;L6oM-5t8rIvl2!KYe#lm;N5JJIP8iUy9M3*ceFA+ee$z)dZ(qcF?`Oy zV!hND5msxj18Wu7e8P}=)S-K2x6xdh zG-Sms?#(9$^i~a!_!*Rou@k>f%XpHCV6Bgb@F_pKS5=v{6JZxk1eA#ns;^A=(BldA z#xlmwXc<%XLV`AA^(Fk#zW>17r7_IUNv+CXzbE)}heQ0()JrMqbI`bgz1QK~6r63H z&l13=sU290H_94d{iOhJxO%;M_4bO}>EL3Yukjc1aV+k?kl80H!zbeZ>3I>*2pN_+ zcAt42W@?x%o^cS~zql5VNB#E*+WyrkK%OtyWu`Vk6D_SsO;)1{t9Gntm-b`(mT8jg z;yV6!P3Fy#(H^3Ms8AxGjh{i7d^&jdalFBI9kr~RummDl7gxL#Zh}c8GHjtBSv#Td##^ z3SXFXYR<4&YD*f%Qv5#jc4)8;2b;iZOWP zbv}}POn;_&fHrn}&LiDS>)*YE-^#&lZeNOEs`YJU0NOVbSWF@EB61RattFkJ!;=T% z&QA!>h%uf~B+_bgJTeBXk)89#F2p)*t{-to=^}RZ?kdW>32MP1_?ltuj?wcUY%C|0wBcxbnn7O}IEniN@4oR%KbG^6g zfMr9hWmUmqPS-d_!e1fI`lx(lx1V)L`(?aR?HejhRbvLO${l>;D8I(aYp|={s3g0? ziqSePH4rcJi?3VW#i*Xx?SbysH=o|W$YZc_vgBc3t$%!7UZ4rEGYr&we9rZEVxPt1>4=lBySW z=Nq{>(e?t|n(jVDfZm6CS7#UL?IvQn{KXcpXmyA2r#~3+9L>s!Lx7^_Wyic|h|{#H zaB`R>6H{!yU2Wv%Nw0B*8Qm|%;ZcX!(Y-h^gb*O8BPTLHo^9Wq**VEi(PZor+w%+m zx^;Y+ucJSl4KqzR6t`)eM<_&&D7n07+Fv3@$Ao)@@MML*slFj>3TM7q;0%s(JrBPe z${G{B@6owec`g30WZ<(a6N1hZkBoyNJ3IZQ?g`r(l^1jsMUV`OT37pMSktdv?I7Mm z$6gb=7#y`&8WIDc1=HpKWjLYv$cZ>JZLPt^*^3;OGU=_wMF$%|(oU$0PfSoSemCe{ zBYlY<{?usBjO%T((3ib+} zV?DPv_X9O92~tqcVnw11N1S7UZ8vS%a0T{XwU<(BnyKr55$L+%GHI$QrnUUfFUPvh zxKnHGS-1mOD9s?Pav9#*56?v*1l=q2i~GJ5lOp%|5@qN2XxFhsKo8h?-{w@@S|;5R zY~0~4cz%O@OEa({Vi74W9gro-orvmP+7!X2bSS2`|7^KW$%QE zEi7vUjoMz9WGv&mW+R4t`|Q`xJI~mGJbFMdpq|UR3#+RHhjBeEkmz-jiLhts-`Khg zniFG6uu7j9%aT5?U>1Md$jqx;*Rm2@?F5n@tb*vm=3bN<)AM!F471cXX8N$m?1f5ALkZLskO}_#G-7o2dk&RA4_|7k(a*c}=EM);tSc=6>N-2m z|LNBExpf}ZY`J00xZ$x~9hBPX_}h}yvw8I`?6@-a&obuHPSW~6v4bCq!88esjh0Lz zPh+t<)F%d*ucd-_Y^Jv zro)W=D$l6=lWg_R%eDkMEtQbO?N|EkTjo8(fgOmL4-%iQIBMT;C;_!jT zoHmuQYhrAgz$Wk>xib`lOD8UyJw@(}It;_|uJ|a%k>L~%T>r|H(HcIB25#Z!D_J*$ za^cKvt6Dr~hC7{@?w@#~>0_5&G5W2}z1&Ww-KLr-Zev&rl5EE|IuF;v`JkY#&KAx*ElR*}bv(LNDAJY+~WvP{VJ0r-?wMUg%EHzDqNY z8NhM7R%3grhV?F+f(dxudf8g|TFbG`E=^1DfL?rP?{669B25v$UW0XP)eK#f z)WaSgJpbcio5EuJ3_$uSc;g-t@(>OWt)4Y4~Due9DADN zPQl$lhW{3Kd%3@Ugm4XZvU95x5X|vlF*B1W3^k{eGSVaToPi;b~+E4KQ# z^sj!|U(+_;v*FpZwR9P5-z9qF?yI0wOlbhN_i==H`xUoly1~c+t`&9Jf@%&Dk#c64 z7J{Va6GNJ1O9L4_Wy}}s9(Gf3@~t$;naqT3AjLRb#iM+TD2a9L(jq-CT!`!@1Y*ZL z)6pqKElffq5N#LyRq)`$13$kaNB;35(Vx z$EI84_Kf_4c7elp6|E3&Bx6tWNSUmV9E)?|p^w!e9=Fe)tR_QCs6~+3F4K7It{Mm8 ztC&c$Qbn8qIaksCRNptp9*33GVBM8LJ}Y@PlWc6WaKsC{FdE|G-8`%w9Fq!}QF_!HJ zx0n}9gl_l&I*M04_7hw5fJ3-3{jnlSP36;OD3eEnrbdCTZ{`Obfd}g<<#cTa)?mba zgqKB`^<2#|6Bd_F^}C_@CC$_!0X_Lvfb?0t5i+n*Yemy)a$y+P{6%9UEZ68Ta9O2YM{G$8c z#K#V_L=mym!P$4?thv5bjcPfqfDs{-!bo1T$AOX%yh7ze2jJ3A+eGd})*-cdWD zpSCWwPprbklSE;DV`q;N#mqO4;&MAB0rz!y?bELdE;SPc!x!!6%3fO@M$>R4_HWLP zGLSJ|;fo~24@Ry!Tg~OKPluQaCj%j+zuGUa6$jV;#vW9EgtRoT20>>K4NA0NH9w`c z8&y`8t3{Ujbz-ICoI5MKyOa|GMKb0xJCa&p1XW32v`6WjJJrj0F!$mbAp6N@>gx*9CM zA_pcCfN;mFTtHD?C+JX#6ivSTEz$yTFT}mH;@#h9AYc-&4BVbzRqw8kknueg=4MvH|f;cif&GD=W$&5@aG-eIuguE^5m&1Bb1lK3816ESBehjn;s$ zjS*^sbSlG4NzhL0exBHtS)cgJD^1WPv?tI&)C_h=fj63MAFXEF!Pc~IsVH6H@SL>V z6bUQtb4%lk9Zc7l=Dhi=xkHL?A=xyCzt%)7_{#PHwd?U~+8;XPDl~vaW?}F>FVF+w zMtF&4_-~%*&)i_B+&MvLQAv7o0iZKBzsESCC`fMBzJJ0uJ3rVq8o=8=u@npTk5q}z z(A=QlUs5x!zZr^W>kyTUQ7_K=P$P&ZuVpNz%N3G8#`lnvfuHhCdnS;qRNC4YEYc_5 zvBXr+5*_YD1u4mVUtP8#B5h+3A7o}*s4NN3D*T10vKHAU|3UnAVyQRdkpLf^*05!p zV|W~-#kYUU-pA1)*Zx@HxCK`V4$3dGRJ`LdJlKyXw-b6IeZxte!h7hC>bBpXDl(~9 zkHI$sHyLZ4omn}7E=@W9avM<3?)crj-1zQ=7cloaG|!io(;DxkDUS@q6rW=NK0IQzkL3{V}`)V zqLrBlWY)mzK_?Cz2*P*#XiG9vJQ7R%OvT7h{WGzQsT)Jx*#6!lHFTl9PIBPZE`?ZE@$Gnm1Biq+%K~kOyV+G& zmS*7j4O`JyuZ>SfW6w1|%!iyN^B218Rc+Z%JbC+}K!?xs+Ivn)ds{^P12)#>E44XH zdho+>NuuQt8GOeEiT&20*)>g9&jh%{hIT-U?@ei2ukV zbAU=@3Af;Z1~1F-!Gnd{R7LIx%(>(p}Dba`^% z0FKVJYNt8s-~T_{yEL7_ye`Tvq2~OrY3l)<9c%A)nzV8@1iztHletZ#dKXb;lS>;b z~@ePzN!x6Qp3CFR;4ExY_3 z-Ql}yDR%;Z2Ek7^6%|I+xYY3sT1;0={39p-bV#YI=^ zX=F}kofq(O_O2o_a_t>NDeo2??fTp^K=7qhIA_=<%+(&0m%VT_>KJs|v6)$qq;Agz z9-abe6O)XukTIzzVQB9!?jhOoBL95ernj z<6F%ndq=*Jn$L9pq?axJS|n^O5C&GnPExf%qG{sBKV;Kpq>GZc+L%4dRn(Py8iWHV z+gy|%rkOw6r`3EmBdPFvyusHTWZMTjX^`&x{O;R%jBuvUe@O}mZOD9;i?A|leBkiL z*Bp$xJeX;C+0SFP4sWLQvW^Z@P~s|?2ApH`6tQLh1`!@IxSe9s)*AOuy0 z2(npNJqyb!&|N5AZ1F+YNo_Vo&c%o?dv#u{+04|0nrE0|W80`9=#F92p7o3teigbqG&1lF;pPCi*stlZqEf%Kdd&5cpXy_W}vb3ynXOzO-w5d6)3V> zplPvkb4vJ|j)+=OP!`bo`SUJrlt_bbDMQPA6?k^)kY?o@rEB(Vk=FY)qA-nIz`e3|{0_>H;T-q0C zfWZ%@4Z}M7GGb@UhLEK+T{aTt!4ye6)7t0&us>8MLOkbNDw!$bxj+}!Mg27%vKtZc zwSq7S&b`wpvhRg&8w@lmkK$S9MFl)fRUbif3^eX%0kPJ6*la^xy%^9iB$S-&! zZZgf4X~tB2Viss(xDvq80X{8p>=v0pJ)D)2_bBAuemuXVa zX*~4@2kHn4(`4uP)gw0kw>#mr`Sv9XHQR>$5+)PVsq8e$6Y%vAm~e{-FFCxx!uc1> zhYfgJeJn!>ylI#ycAgRLdem93H?QPn%cVdEGR_}8{*P?lpt^uAMdM+(G~|1qt9WZa z_F=$_AEY(>z|crHWmU}3Fn@x7y_q+)(JUDLP zPb<+>vs|P_7#HRtuV6jV}D!MAsBC(Uc7oe9e}K&{P=5K zTK~PH5mQNd+G=gbykybI(AmsEBbaht15*;~C`Ov--AhtR#cJorT=Hpz=qX&2TM|8OfyU!=)XZRu2b6pnenJ>0ms2>0h0Jln%K^bX zH6yz}jV*z5;P|ac#=u|Nhr|tJkti2zvihC&7%rjrcthV2a{X#yHMe!G`OU*W@$iGE zdUo^iC8(zt*m}-FS9cVNbLVs_ro9|4xr4D@na`@AnLf#HWH16wg@?)afNwy3XVDRJ zn1#C|5nez?p*!c0Ki`ypo#o-h@o$R6CuymCp{H`uWXjnYFSne2PN$Ep?GQ~xDdg$A zQJVBJk~gQf$({2S-sr#F?lG6@{M6S$KP3e1sPhi95eQ)cBYGq+{mW!Km=)Zw6EJr{_XFvqEd<~G%^&*2;EhQ%0x5D zc?ShnkbXFL%%cSii5`?w7H84UcTSV@%$Qu>i6hDDyDW(-d+<%%;Lc8TI6kUQV1h3# zz3RW)k&N0eVFauo7}#1=%Zt6Kj6Y5|FZ9llF^!v07+mWdGuihyNdhvFQmKAbP*z;u zqi{4i{uGfeLQ-44llI!_6)nPNNBo1OU5aLb8=I<3%AzN5mW<0kCvxP723o#c@>es> zW%vdRp|u-Frz1UQycN_26f;%+H6$-0T;++)V0N9R@zssp0KW2H8OP2E$oD!eh)uPY z0*FFM=C2(<>AY{I(rf+Pw;kt*9*Mv)8&VZ(^eNELseH1U86xu`?kCUo|9WDFBLp@4 z5YxW9n4bknoDxmvSf_I}=f|$Bl41~8jb!8-y6lT8ftPKx%gk+QTfwqw9{EMxK$~tO z=q*BUaFR-l&qMCSwIns$del}b3*)5w(@bl}{%My*E1cymQ_05&<#$>HjFj2)mb&< zN-niF{{_nxFu0qmSF_ktt#cEyIs{>|I?nh3@hleCs~fB*NU%%Adm~H@k$C-(ess%o z*-teE2E$`QB*2Iz2s^p8qvE(k=ZJPnLC!^N(v(K@`eb8;f$xj^ir#v%^YdZ&KP|YM zrE_N?*B{c4Z=O8^4#F)-gyl}O${NJ4)kZrXt(O_|unbKkQ`BoN!J9Bg; zWm2t7b`vE_CzGp*WESWJRd_3y9>1^AVAtfWt*Y|0c@%Kaf)NrgjXyv?w__pqTE|k5 z>p(Ec{iX#kNJCkX&7w$~?CjHbqlAL3?STA}**Y)n zXr9s5j3$3Y{U|@y zc|lxqzzq%wyc5=Il)vIfFuYN70_-s+vYLv&hqm0qP5U|MG6lu2_(#DpAyY*J>yO)7 z@)c4?0~^B@uoVnyidl4OE@(-nPA@87vksV3d64`a>d_Xyfod1I>1d(h7kx95OG!T@ zpwiH#7TzK3`tQ1rZktVflZPg6`ohDHoxJN0*h7oyAANbf*YnF`a=ehtox7H)?9AXB zzPWjcKrL?z%0E!*T+RYXbx%Vu?9O1x2QFzT#-=vyIE)_fjmRVn^(M}gw#jDi5Czf; zO$TZMY3xMqerL|c&2e}x+IHotebSDFoN?0^C|c|7Z$94oq1k>MBDr zflr;T7~cDO!A!D{I5qxoDL9sSa`f!CTIPo;4+tI_7u)tszBq`MJRMbQ=KM*VF>tzg zYWZi!uu+O3ok1c9C2GRdt8U|+T#2_EKrcSQVWoIRLkPT>{$vV@L}X*WS*PDM(D;Ta zBPyVFUq;2OiT5JH>Juj+$U!sH8NMCewAVHJ+)`P(&S+7tjwCxAf>8~imyUBRbooVVj4lK@G1N6|j292dAqJ$17ZR9+i zB(2Qi_h}WQG$CFgdsQed&r@v!}W#e9C=dUh0br+uMm!nmVtmtLR z--@_mc-&L_Tso?zg@pZye3AaD;q5X1&QwbLJ$iWXuy7t8fwO|QlI(GJIN@F%yJV<0 z&AH=D?Tb{FsGR#xCoDU!1sb?wZP9f?BrGM>=VK35@!>8_R)+FGv1g_61}yYvr39@^ zQ@9*h-``a&FGafBmEBjgs@AvMI^pgk5p~9W(G1`FIgz?Mv3}WBh&CKQh#qkSV^y6x z(U!S`0tbb!_rPW|cBhNJBS8k9mMR}4|45fnqy(xOc-TGP<6f6Kn~lp41w zMI6NAC9uFhf1KcR;bdL?>f<*HM(#;L_)miuEbSDt@JDBRzG~P;IKJU|Xb7$NueN-R zQSpiYLkcdgzE6E2&;}Yq3ctj1k_CAgH>_G3!LBnQl&wM5;N=lhL7Dm0taPkes-Taw zF>$63`!@0Y0qT>zvOc<3xe>}nu=4r|lF^NFGE0_k7A~iAbeBA8(4DFmaNoX*nZ&$~ znc}y_FAsZQuXf5rR78Q&`e~`YO0$C5H9lsw(p|2(fR+$wwJ8j50A&x=&?jBYpXXB0 zbt0Od$UhXXTVE$1{OoS_QhaLT&Mufn`}~6X6^Clh^oa|>!dK>r@P`x4jZ;M|>Eav5 zQg<>HE)<=q+ZNuYY}+xMki_4NJor{6#xh16m({*4T?X4G?04;ZY{>EEzfptY>>6v{ zPie(R=2uUPoV6*mx7zLi9lKacmG8=Xmje&e19Q zwSPthWI@hiFeIQ!2dS%b;CGxez3BucA$=nw)FUX7l}la1&^zdnNaZB?e0IAuJQS-~ z+Jdn~Mc2JQ4yI^)$nrU=@1vm2QiQ^N|H!Jn zmdV)RWh#jX%vYZ=5Rt}!7D|#?Np;ur+>&Z-%`Wq?8=#w0+vO{HoH}%|L>vKG^k4Wh z6bA(X0^pYlSKbJ5>hVY4MpDzbK!0&3{xP}A5JVJk$f+nN)+M=m174^m*u$%##+N2w zd}rd}U$ulFdDpPHm@}(pfE@=|B8$y-`I3H>f-d=ke)In`qYwsPXx-XheYid3JRHyp zx_kR4>nd^fy6wcc{oX(CiVOA)!hj~Mp~ zug6SNq@6AO&u;HyVqYpBb`?n2{~9d{#X1V7`8uF|F=zUZtPC9y8?l{KK-()@i3^VYjn(=02%W93sj6%{`Gc+( zZG~RjUTRSIYKvy6T$-LpQ(G?m`CtDNPq_~GzRHB>the{-JMRybcPrX=n%gK&{1r~m z4z8CD)PP7sz|W|hIW76VE4nrtetQibT{szq*$ z5!VP}KkPG3BAW3^&LnA9f0j$z-_Q>%O(>LiCffetAYrT79Tl1mAgm9IijA z!o!?b%OrRpAQ=6hC-CEpS%9-phf6Iu(NtfpHZA%iR`7B3mS+gy2U>9>M33;)88NA| zsEGF3hlx3omTyjqT0!MWMC`A0{T*9h6JLoJI!P7(NzES)v3aj%zTMt^%dci7{UjdI z_Lj$q00g5Ze%BE( zecZ45J3B{V884_AMJ|bqGEHIT5$jbj{fP*TAnFY5L#-tpJ#@mQpg)?SMoOJOAXR0; zR6NI=`Q2V*RJ`jzhFQOO*mONx@c!jIGoi?4@0r`o`Rq?F-PD7Vl@_$QRBOg7AkYTH=C`m*_sGXp%qh7R z+o~4sxj345z^gdOBKWm0jOqt6e)D+@YbGgLQ%|c6PTo@NE%Wdg&WtC!AmMQ3vH;nrTjBY9(%j%BHAD;whq7%_B}9G>U)9Y>-YZQ*ZVG(e^)F! zGI~!qZ->B8=>xUz-x$DR6Qy1g-S@Tf-)((#=Ta0CW=^Ko#VemS59&;$(qP9|z(div ztGK6zolS&<+zMlYG{E{Dj7}q@#Vbx;to?zDk+hajtqk_qY!!ada_`USkA<<8IZdjs zzRdbRv9r_>VFft;g5F$dyFJ89O>2%ea_$x0OMF%9j5%V@_J5;-Bk1r5Gks$D;+{l0 zmi4Ph^Iu=gj|obvnF0$#&Z~K$f&zr?FJqNRZ$P9KEp$e2+nP9+!+LJ-1Lbn>*Wq5m z#$V#KmX-|QK5k$HmMmThGlNGA$UXZuvMQ>51(?qn;#lzZI$64Qe##TvWS4GP7^8<$MHWN9jQze1imHiBQX&UJ|LepPij2$t}4Jfh8WV|yxOsm^V5 z(JMfteUDrExKA#Rr|$(vP2rN@%^8tH7e)l_{4bjZj*bZK?p(S1>8yHXa(j6$chSK! zfa5d3nPsDxVV=K-+NY5*jv%j1G|XIhN7Va*H=mUjUHMHXD~|~nN3@|dg#@hi%-ja3 zVG#T35pGr-jYqV$M>5?G(5IkDIlp!dO* zq+~=*zf%~xKLvrA0G+T8%*ETcnb#%*Uy2bVrvFL8n+d1p7Lf$32xN}t^$_#h9c$j# zY9@dy>Cdawf@tL{CA^nScIjRG5NOANFk`>VrH6(-*g$0A+J87z4r$T__UvOV zzD^@N_{<|9sMS_uW3THp;AmyUPs`{7lNiU}9~E@O_gDNpghyDX<;#W@x@eDV$`^a5 zI({f6O;fPon4oFI?_YEEaS<*roDdBSd?B}MZi+?UxH5(uV`U1Adb;aYY+Yv=HS5_D zx2vIKHVWMB6@I@fkYSIZAg<202Wa(ZF{A#%Exfj0L*(0DJ{NyuUq06aw5H4xLx*Rp zrMdUnX1mYuT5dglQ=`sck#QvhV7LN9c-hO`3hH!3!2mvTIZ}?%0T%>UzLAB?E zFroH(^`pHil*F$TNRA)Uf<7~JZ)gznW2$5o%+K@Dex9567(^d^86QD@Y2RxsiaG>q z5a8!MTD5p^lfQE)@OmkkN|qwN%{h+puh>V%4R>glu0g-_I;(Z@x7?K4R%v;Et2beq zAhP0``U&=&tYA;57eK#ux1NPY6c4)rHu<45_RJv`uy!0UgW3Fdj742&ba?bvG#=jY zfOiPsGy72pDW!xYb5$I&e9{FB*N>c+-wFpHXx6!17?NaILlo~8!@YA;L@hf;s=&Ct+QNd2}qEJDv zfPJQl{rolP;kxL(mU119Rx3e({q&>sZWCVF+NaVFJmN2ckr7k8k{c%;hphk=LSMrn z|K78E(4HF|B$8zyq<817)@NKdl&_99TUUK8DhzJOG1elBdRz?w{`aq7xiE)Xyowlz zZ1Bk|tGcR@LyGLvg+R-J?98tnV;trd zzT6=b@)1s@FjxPYy}pn@#*OUHN-pq65%8=haqAn7*Y*NNV%3vKTQ`i^ zyUb??{`?U_3?NL{?+h`_#px^*Aq2mF6)N^1|BSv6RoeR)#FcriJETmdljv2=Tw!$? zsRi5E=l3%Xb$N_{bT zQx2R|aG@w5Nd%jBt|soFI{s;kkABL}s3< z(b6xZb9&`|eQZ{^E5Th9hqsJ%0RpqBwm|H)Ri*~?sD=DP-cmZdmJPC+ zvR)$%4U0>0&4p8<+WyR*Ynq%)l1ZEqLy*}CBJ3=Bk6}JHhim6n6~obzK-}Q2BT)>G z)S`_c+~WNQ#t{)AJ^Wc*q;qF}%$d9frkpaxDn_K4!*@Wy0$TwH(CBz%oh;>zaeW~- zHlQ`hoB+BLHCP@6JKWf(S)*AfIv91v;-WGgK13xselV-Bc;X~oT0ML0c(zn_a`K5V9bZ#iPJe+cW zG4N9zL1zE5kZ|uy-tq-G4+clABB#v*IZO{%Ac;d^Rsy_?Lowo69PgJ3Z{kqr34 zcxQ$KT#%Hx=+z(78=za=Qme`T0FbVZ&%oOSijft$ur-^?1wWg@%@#+(G7J9W2xqYZ zK@t}OA_66d3`!)UTy4q(aKi4qvkxhDEXmZII4w+V*C_bdM~KuG|pl#e0uXm`=^`h_G8;0(^8Ibx^G@WE4F=-#zHU7v>1 zu2JM~Ex;I2SoV{xb+x%)L@KdUD)>eF-|%O$AeVTl#`q}2s)`v4BzgaP{P`EvG~%0* zv{ECIrQvRH{w^@J<#GTGo)QzX(V>aYOY1 zd9eF1x!O}s#o~veFB|u)kQ*^U=>_njD@g(5dzjQF;=3309)o+ZCYcvd0e)y%4L4c4 zVw1Y%*Pq+YCxc02-e1BcC$uLc-KoaZxYqrw1^e4F<;tsv%ie=)%_lqD{~@BNs%Jk{ zJ4EgYYC@bc_Of`aG68<`xxNl%M{=9^SmZ~BWX2yuILBI~;p{%V{It!6pSS-m`ID;I z%=e$Aq5{q)N@;mKms`(94(Tj#mL#Y$yIfTjD3&uQa8>X}9p=xDX=vI!r1ZG(f5>tb z-iL&=TFpN3RAZ;F&AjsmmCy5$GC%+|t#?yDrqWruj-{l!1bdwM@xmQ*`esm>59~*f zgmt?Mn$}=_Xv7q+wvpwh(t>pFjO^)?)!1+Yi>)YXNYL>^{WfbH4mMsrUhGdKxXTtd+zGLzG zd{1tedJ6tCYw?C$cFK1g9xpzChFgFL;dOJ&>8TC82ygY`cL4xvkD-odvv+g^NDYA> zDxe}XN})QuKw$1T=z$t-3b$Gu8mTt!iR@+SY3DlDhjVe(76gxUM>^(*pIWYEuRfVhh=Dq6O;u0 zvACy&GVu`1FLf)J9!<62#a}IE#Z3wnZw$hx{i~xzNrJullLq%Qv2y$N&UMMW{Z6=B zi$%TX4uEqEUIdqHp%EIQP2>O^u<{fE#~g$M{?j({5LzaT)m8OREgalvxW&pDrwtO_ zXU^=z-8I^auF}T)j*VyfLid{XjjfL_da9}Lg3>$KG=j*S)a80Q)##VZ>wYOAGBa<0>JlSIV-i=0$GT$T@C-@gbLF>X){| z%@{SS-E(W<&eqs-^nb+OowNTi1=*EXd=zK`zWUvNzdjFq5?B>?OMmn2XjbO1;7SVM z6!v2{o_=}Zgqa?)w5!4VWJJveiO|cGg8L<4H*AV}f$Tk0ripWFqXGE-R1xETLaf8^ zps*QKDrN-A0<<3eg$){8Tp(KkP)3@+O>2s5@WK5GQPdMr6ryBdtz_Y~jNi||L+4&U zEx9$yfgDD%&w#&5 zsDlpoc4%vHTrLTc9aluQEDhQ^;BwWXhXj zXNCA$X8TI``u_A2^6D5$@FGjgy+g?W)-8gBB1J{A7UQj$fs(s-MCZr`N_kcf9qtUj)XTx zU--yRpq@V_T&}CY-LS| z@TZDZ&Z}Qs$z!>*#m;la!`yCHHza_6P%qVu)f~Fh$WXHllnH^)(REsj1BH(GYPl#v zuQio(fjWaStKk%n9Rkb9P?k?QwKOj7+#qgkxp>d%|3Ns{A^gx81$YF>4BZWDlm7rE zSj1H)EOQeeM!tQ~qoYDD?E~tofiP4#bH@qEce9`x!Foo3e4UwqV1H2%B~qg=KE|(n z@o9P07;A=S|WHnb~_OpIjV{H8Qp^~Srj~D*n?uPAq11vRPHT;od&=ESqdYFEv zxoFlnT};II*_pTGldRw0<%Sn+7U$yjL4CZ zw=W6Ip$p*2c6@TNczpm{v=bfrSHJ(c9LDf0he|~^mja{J|Ea+y8Kz#AZ10*cxQF8D zm3z`O+2=EE_1;QAg;^iL_Pe+g=3O4 zN@tNHvN<860VBGaCeCBE?)Yopv3U8klLnTzDK`+;7%y zh>D6eoUh!J&`Cb=s|8E*yWd}9v7Cab!2~_COU2OUVdlSX zBNopE7*8+@y0D=DS+!HhV7*m80=G*yb6<^8VF;nyQlj6-9juCMmnkenBA8kGV zt#Br0YSrNG^N?k-udjEUROyDE4v%pB#qY!|_-WixQ_D;)x*jHm_3p$3aQsh=@Mbb` zEs>#~U$r&>kT~YONYr;-aB%;H2;=9r4@4FQA}i$F>84t``Gto&;XyF^vMK71ex#QV z!^azdg+PRsL#U9U^Zjc3-aBSW+gpU|iyTU^GV}LjxYd`;8k=HPt{_xJU>*dKp)RXv z$=`ziBx$u2yqTVt!X>)d-sYBAlg8?ek+ei)4WlEQT{;k9rXNAziM+{z_&3{3)rZC; z>u3VHkh5_n32X55Ooi3)i27Sb122Klrp*>*+_~&=m2ji_!0ngWq;pg`$s}8ZjeIdp z`=Vw-a<=3s-5>9`;O|Mz%KPj&gXEvrtpqbC00_#nYo(oW1n8#BJPABD+#U6)piIHP z13tWCCA*tE<-UBo@5F4+Snai=;sa`)AJp$)we{W`aZ<%e(!`4jb>ib>V1wrjI90li zr1(_b6cf{&A_xAnu%P_NhK5?KOi)2ep`DrLB6?#X@hc+}3CRv3W{E?BgD($p>uB`t=v8M)JEGAW!f3h z;~*q5NDHZ zs?BOo`1$|T!l|7(lzkXNYe;<0pI!m$*8=q)6awxb;R-NRG8zbnog(EL_q!9cbU6To zD{dhF9OTsj7@06Ie^9h`ojP};>aGQ{h^)E9iN_3!()Ar>*0Xh_)(`@NItp1@3|V z-<8SIT27?4fL1rt(|s%K-R;h40m~;WxFNr`BX4-!gxyV2R^-uZc>98#>#Oio5s3|k z#9hkRXol{?pO&b;fd`MV0Uy?K7T(R}69(KklQ2tYB%uz3O(0!({~%(Ly2b;*A%Y4h zvdy%hUGP>_#8%Y+znppckZAhqYmyj0vgZjWrtS1;?AxyqBER; z{|n%51C)%;txUj<(AJC5C!%rW!|}Fnp^vAlB6o~|6huxPa;ndRTP;Q$YPrafjSwv{r}Q;2;OvhU_J|RpE^d0ed+EA|j6seHwHkKC zV}WM;vNJH7ftJ2)U~!a%Su_E(O=mJ-C~1VGsd9DCI!?LP0SseLK4q=V95}?RJhZgg zBAgdKmE#x9|52t}@0K!*Y-D58s=1n@y}RS2H5| zq^bHVMKFhXGvu$c8+-iUo@>(+{*;m!f?aByv_6qkQc{OWgwU(Kq;7@$d*g3{^9EoP z4Sb1Knf@J0#DIz-BdgT$qZvN7M!%XfO3(HNMgN(&2|LPU2HR1kGuHoQR8AXzU=T@` zhJ_iBbMe=${A6w~EJ_oP>%HvAL$<-Cl|+rVdlV!D1+!6Le#0aE<$?$V*vQbaXuiB0S#p63$fs=gb)U ze5KxqjJ=AfM{@Gfl@kXTE+ZMqn1|& zca4!oD{OqrgCkBVq@E=Bf>vEQeh=bY7d-|kKd37kG7h|SFS=2msh2F%;CC#K|KP%# z&_b||m@1v$_fzy#O9!yY_;aR_6s28(hm+>94ulPEWycd%d6L%~_~Sq@0QdlyoSfWx zmofV`$ zmkBjU+1`#xttnF0X+kukLye4UV#W6RfnEL#DZoDx)3I&aMlMl8Z>}N<|LxXH#ft_w zF$rVX|8vR+llb>ag^lFK_}F}MpXXbNE481lT&jD%AJK-nCF=n=M9u9}euZ<1PxQi| zf2QIIa41=IAbSupw3d5Nce>Nbpj@u$t0!|5C9zw{aUF6fi~Pvbr4YN7dIo)Aof)TQ2g z&;KS^fQA#wb@_M535Y04`jK5}T7nTn@djy)DDl~np5mLLFRAB0t=*U;_NRogEFZ&B z=4j&CNHM6~#tSUilIz@8lItYXgk7X_g}`P~#|zt$flsYAVr}tgZCY9u+Io$m5O>b* z54MbBy}Qi_*dVn(Mh=n}oB9=R+CPYHYqm$`{B!-Q;SF_-Fp^-|g9wbi%-(w}U6*Xv zh;5ene03N6GQ=AMS~l8PpcSq(jXdCoyekiRdZCxWw_71$NGEeXYq4dIl^f_kcg&_! z^Mvy|E&vR{HhB+25eo*|DEd#hh{`5f-pwR@D)8wH08G&4;a6N07?Um!rvCJ&I!FD- z_`&F?dg_R=Z48nt%$6M<%95w0jvop#zW_eq(+TJDU6)(Ng|4HuFv8b=t7=3b=+Md# z$QtJd80qGXRWhloQ!B3Fbbor7K9aRrDE!tF!&pt@|IzRzJv{Q$kxxJ2_a0V+@)Sml z)R3aFk653p}9bgCLf1MU)R{JKzQhNxT z<|OEg#w;DjbFtDUu|t33q)p)a`h)1{O)#+9<+x5{Y2%YN$txgJ4?>o&HK)8m62T0Y z2ob20j55*3k){ONb=2sz-TIORm~^A2@(F((>2?OECkPH_ z!ptMxnNV7odq|HgSN~pa`F-eoxR>Od2@J!m1`gh{Ks?kb4CME^1oW=>zFGIGvN6=m z#iEi6f;bK2(3>?P7qSm|ooLvh`rJFgBJ9cybj$vIx0LZivaL5tgvt_IC2P9w=Rw*j)ZPe!@f1^541^z+7XVtKzw? zQu}{>j^F;oKS28BU6M(zI%BT_s&`ytK%r`nSQsk2xvv*Qprj9>>@KBvk7bBKT-xYQEQt<8@NqH zB*Q@g@}K_;l4}$*q+Ytp^?zVr{z5Olqwk}nKqBbNOC4oaEdMs{E3w^q&Yc5=x!suB zAgqStF@oFnBkN=mSius_dCO??P#XD6mX|9-Gxeusdzf$oO_Al6aTYrpv2ftAfHN#? zdqG~^-1JX_QqD`Bu3H{rA*t^VXB|SQb^ZE@2SsoX^D)I%xP8rsF=Iz6=-%eWasAY> zi{Y^}vjlgZamutr;SAv6EP@5^zVW<{)HBHVYRz+r1M^cTbLT%7@=GN^o%Uh2P$=lW z^WSUO{qs%X*C@T$IX$sv?cf57hlhc$9rKqwe(eJOe!;%~dAeCHf?`MT0blHdvY*j^ zvO;Xt&H`0--D#-P!wP1?Kz1%wf@5>zVqGhQ-f5P5VA#4k_EVX&8$mmI0*d5uWX}c)fh+^uJcsQ=+bS`Z=0+m(skL<#vLD zEa=-y8*6TV!46O(21NkXr&lCUl z$z=_V9YQv3!IvLFdyWAd;e$uzm1IRTYz7HXn=l@*=zkp%vac8uOK1rgtZBWI!@xyn z;6ZyT6Ye^h=1Z&|yRjR(?Q=0mMxB-DfBJ zI2dI2WbvoLhGstdIgq@_D`+pRdY<^UUA0e~-n*n$Yz%~TtAbeyd5j9+f7Nq*<=mQy zmhVlyYBaw?>8({mR=!-nbiXiQ8lh^|MgO=^G}q~b6b;97c055Cch+U2=0I;G;OtSB?7B{fS%`CC7@gVn2)Twz${x#$9#+-E06f z+{P!VmDG-tNB+4_nEx~Ph&G|9p$a|JA~o2WI|e8o=zDITb&|ktF!0tkRF@2u26n-| zt3tL2i+Z;zdi7UAng3q)s~a1UR%l$dL%zNkygUI0y#UrhD<>xh=-0^8Cn8K z?VV6%C`o1|=DnWpdBDPe-4DakDX~VfVUGV<;=Mumgmv{jb^^S>K$to-wqwoRXD?~Q z*4z!<$Rwq4NWAwFlIZ+F`nFZ|d3?lWCL9^pWNqf~dl~7G_KBebYEU-;S`Xf=d#y** zw=D)WFHfjCJ_F2iYsYb`Jm5L*`)~dHRo82W$7g5wGob%GXl%Sj1s%(oGB0y+%vU5S zAm$-ogcrkte;76MA>6qnN(KgBRZIUPwQOy1M@PE?ye5ET#>~Ppb?VASc4v@EL!XFZ zSx-_jM~{JFnKH2tc&ZgG9Gi1P0sHbPOC9QTxm?BKEN$iS;FT@EI>T?+0!}Fvjv=7w9Xj~ zHbCj~0a)pBwX(_`dt3IQKAkbTl_6X5;SI7gfRwbbP^JVkp{g5Z2vASj0E^8Ely%#v zOrNbxA9ft2_b*BuZp>7vtsn4@m(KT>&WBz2Gz|FjW>1-ias_oF0T-}~OG_f$G8*^R z+&z7L2!L5NE77FsD?lhhLP8o5+6I?}e!))n?&0x*^Vu`=p{Wjjg-m!Me!b@S8t}vW zND1VZ`P`iO-$#7_hLZiR0}t29oC0Hu`%YA$Id^5>83c6_h&SzI_Q@gRNV&qH(FbI0 zckc8FLNS}v)rg`EN7jP?Gv$U;)~7F@kER9UX?lWuLVq{dsD)0>@;8)xC}i_>gvlGy zVF;}XHi4dQ@Ge~O5RNHQE`j|lU9sOCkWKTeE@Qt2MP@?b30 zDt!M%yC-^~t_SdEs7xB&F=^cVI9H|;Eb>~Mgq4I)sUk7AuGst@bNK?uMp@w_zgOCn zUHdtU_w~FhjepK7I|~*iWrNHj@^tLfE7=B|`LXFQXa(gKBuFYq3kI5r65z8%cVD>j z6fM{|g!5ZNdMH-xWh6Le@*LX!P6$KwHvlttx5?86Hlmx}$GjO0JQU-vCCB^#Ity62 zFQ2wxr8-!8S(S!|VBX`nU_;*xB=^a@JUkGEgoHdyh6n}L)^Y&48-Rc@KqN%>y#(l* z$+vaq&6AwyyI0)9yEbSc-gTNT&ir6)z094bC$b8t>1KP7rVZxYQ3Y1k?K62CI6hzh z*OaaYaFPJDO8_4A=?s}FVmY%SqG5-m$;Tu~E8o+|^azz2#SrbT2NoQnwas)7b0!Y6 zVr5i6N3%jA`-*-D+rT+3ZJX9Uxly#W2U+^okTY7dV&_Sd~PVp(SC~zXvWKv zdP?}$5sEGidt(m%sX*?sv|ph(c-g{@bvl7j+Rzt=6Z4fM>KiHR?i?+ihoU!sg?Y?; z=|ZK($U+JIN&dEO5+wn^R7bZenXPQDVua8D^C$y$!9M*11B#f=xol#MD>Bv-Im_ie zwyvYUlTmp-IpC#h0hS zPeagLqA#La&wu9l2ogzJrp<1O0ZpNL61pYzY4)0J){I41l>c!7P}TrvhC!-oBxJES zqOY9&1dwjmdkE6&`k%qO(fenjj$N$|g#Y{wCat+It7E*;I`6DH5eFPDNINe{9}@K* z@e`zhUtF&cwqcN#*7nP_HNdes*s&RLr68Mp%@&G0DG3i78wV3MDb0$eWIWTZ3k6?> zpVgp#Q~x%rt#@NK9OpgT3TD!u?v`h!tvlZ5ViqnL%zwr?KmG{X-z2v&*ow%6xBvY8 z>-fGCrO*zs9A|4>xMMN>KIAtE&S1MFVnbQ7nyK66if2xddlK;&Z!XvA%AI*5*) zQD+`!{_d;ymVy0Fh8ho}Ur|1ZtZoT~gH~IefT>a*N=eA)*t%L2(nPaJ-XRrP`<(-3 zG!GPWr?=8w*yXSq8#iT4YCLRUZ?M^y{3#5hR|SwEbH(M#`8)dI){yM=AZ z_pEiLvttyR@dh1WBHfw&w4qGTDXw(*O%x6pFK7vUp>#@!!qz z{{9bpA0KAdu4ZlQL}_58j5yeUb;nyV<~oVF;^yP|We(^ZII-~dyx6FE0{yoG3nEaH zSm4{d;XWv6&`<}NJGy#Qn#!)4K0BNa+mZ?NQm%zXw^@aFtW$bwQ%8WBY<+-XyaB1u9#2%O*4(>4zCZ8DH zX`i>e`)5c{der|N*1`QHAO6rpaG~S_>u5M5h1Fr0V;CMQ4I9%ms=igJ-_~TzFAf(G z$y+x?;?JWW__jS2@HcnZ9*K%76J;Z&b22sC$KAuE$=9TuH{8q!GvUPBGebmS?YP32?mlgaJJE1waVb! zbsN}=wx)weDmEf4m+?+T_U9y9>9#vTC>Rff$s+z&EW_{v1-HspMAoPxO|=+Q990_S zcK~lCmm|N9PD1=k#-ml1XFhSRRJ^Qew_y9!ikC$&g@>tFO8#_L#L=?B)B;$9ouskY zAcXFs2*j+LyVp5Fad*3sS@bPa*rQpAc*FRzOe0Ao$@|Tb+Du~$hCET4AUWAa>`~Cf z)_ZFVYN3fD`u8IdT2!Ah6YRl<8+Bc|oJ03u_f>OFteN^*#^oAo#7Pad>kPXFy7(j` z4wxGD@m+H#+ZBwFBZ&iNO`wzhMlEVt0= z%GrOS+5T_7i>y`ybM1^MUy-<_!bpH3Y>jCHL5V>A&%CP2iGTCEs4gwJgKY@UST$E4 zf?Vsx1;Sa2g>kJ1uqYXKlzH`pXHQmGW%tw=i@*4p;o|l|=+HL*Mx^;4E0V|13P8eQ zRd&8Jrf`ydJQ6pdgpTP8C36=#M9eh*R=JcNkr}Nzc%2Sn1-Du3yE(zQ%4&@(IGgCPD4dzef=dE2KXN9Uu$a=vgDsahowO94 z73kBtzL2xK#}Yq;(hW;g2PNGX4@1de9b96&sjv!%B)D_9PV+i6nl~rrVL?7#Ts!fw z_3k!5=l0W-CzAXq>~_MT$(!STiR~?+DNHZtVi4b&C+~L6diu#79*xSRf4Vozzq!?4 zuA?VRGZ}7=G!E{~7rDW!zqf0u8vMH;$u2ab#-EFxGQ{DB}LT!E^0VYc)GFHHQYMap3ZEIr4YU!mQ+!X}@DLEOO zzf03BB-PESB%ftz^M!6lk#4@qYS|HlJ<;zM0oUX#FEh+-^7Y!r!p`SSD$W?}Nmy=t zi7v7D3VtOfh%YY~%D17liBuNRP(%)RYT^43Vzn@2qq9{3N8h}-xXv7|h#EHBU11_+ zqtkHu&TlWk{$WR@JjBu&cG4}4L%9tcp|lVHGgeCxg_wOG(dgoU&ns+`QPLIZi;33r zomox4W78v!(&?8*=Y~o&)F8nPKY9m6vrV`2d*cehJp-O}k#fW3FHC`TW5E9e=ms5{ zm?w|N^ivDujejqS(({b$(hu*_Up?Z0ssOX9PuHo0#F_Fs;WVdmevfGQFRyNL_4zH`Z%^oKY4 zZXT=F^J1M}#hZc{7?mR;R=fSH^ET{KmqLP!N+emvRloD#!4E1(rv+RaBJfSL#q%Q@ zZ`867k@dWrJiSZFpn7`cFtUzy#=l00=#>$nZvt?EIx;WnQdnS^Mbfz9!4$%w8*I>h zFL?7@*L-%%pUutc$QFMVC!-DmnO#YJ7g5%s1gV!2B#w$xTg?9++XFyZ1^Pr{$sd#E z+PPCrD8HPU)r9kFq8yu1Mrx+c?=LR4&Vb{G*@l)LDQq z(S-@~(n1Tu%FF2+8UO@C22joo5Xn5vAZ6oEt2t-NQ~ayULx%AkHN_L#DK!ct!1BaV zoU}#=s_>b6$mzDSr4N$3c@M)2e)t_T<0AN-yd;Mc_R@m36(YB@{T@UqLedE-AS3DK z=xQ=d!Nkx=yObK9lV=x@3xmeGEZQ?FN`!Dseau#EnmbO%8zK#_@?(ZbvG#iB{LyRV zFADgQRlM0I+g4!PpezEhknXhan94by#j~97Rn%jvHxAOEkg}OzgGAydnd42nT{tPO zkn1c26&!8?-WTHmi>4{PCTU~Ce`?G*1jz1yDxVmRW(h}!r}s8>1Yci++iOF#a7MLE z+#_8}LP!rw7bD38I8U;d%5tY-oq2>mfKT%i954~RwRO^Wj*iHBwm|P+4`3E3OLy{)v@mv))?H`?50FWQi6} zdT`%@J;oxn%&eB7*WA+dgqN8eB2c4obj=A?3Z0UiT8^1=t;t9i;b&T#0&dQSjHwc8 zcO2XYVSJeq!U`9E@Z?GEFU9Vdo7cDa*T_B{#s5xL%Gg4aLBNEx4@ELbN3fuBImwn- zN%&Tr=aY|n=DHl)3zDiU&y4RY9k@Os+B$q$$dUC?0rPu}myB7axk#3%v7jot4)pm6 zI7OT8zz_e{*aCsu=N%@K%L6fTdmAUuC*UWaJ0|cpc7iqmGAP-ANmm-mBer{nGm}h- zmXl_hoqKc6PXP)x0}YnHi6GFSvv@+SJ3%?nsa3eG^W_}#U&B=bY0wlS6z|asd)Ea<`9SMwOy$?DMLHcV=Me!e!uE1SyY1{wq}|xSZqCi; zT?{Mq`nQtJBA}`)h@n{3JAUE)`rj1ka?Sbj_{bm1!QYvFj+zeLT7s%I>z(Q2i65-F zgN`kmI&6_+qmx{C>&JK!Dxiwo2~K+#pT$fR?A_8576(f_0r^Bm&W`~iEY7cs4b;6R zON`*nD*?=QU8Hn5)HPd%8v$buX#jbc0llAY=ls?%kXn!}zX-&*NU?H|%?c3)a$aR6 zphT-R!};eaafFKqQKQMWOzMC5yF=UC(>Z_z8T|Z7Sij6XeTeq-N>NcU?%?fBS!AWN z0HhW%ouLK`9@*J>m%$}qRJE2ran%5b2j|v0Qnhod)B=04k$qIzSsZ;{GFtySR&qHd z|C(<7pvKHm(;0bFZ2VQpVfpE;Hi|J3ZBymIhwR=c!jQp zQ;Ku;x}Q(5QyJ$<^3--4Aj~ve1aS^lt5KyU@pMWcit}oCr z1KtxLVE{m;N$Bvf>9uU6s;H1&7f|Dusz}r>2?Iq`*T+6{$NRC`bKB%dmDzNDbi5%B ztO?W!d)u<^6^mB=dOmN`Lj$`~@^^ld4$>3b>?gceyMLlF)f-uSt0p5~z-@Qa!=pAot5jI<1yQ8Byb%6YIIo`;SoN7QX@pTz_`M@R4696~A}4E>~3dTIzneb6`L zdQBhEu*(0`_y#WXSzftQ$1q&o?yGBSk!Cm*D)qLmuE=hm^4X1sTpqqzpmkwc8|U;wrG~y4td1XDY9cEf`A>OaH9eK z(iSQqlJ*=uvEM;;5f`y%_=%;`Mg}TraVvTAS#NM|g*_e*(kE5#zzaof^peVVQqTnqHgNr#3b5{&{?)djo7k<4L8x{&znfy`cNdDfi%-J}=T&>j|=^LlgNCOWzAElvT*gg#p;5ROxCSg?;`Ew39Kyunby)wwA zO*v9wd!L0Raz$isrQ-vP&uJfVB{*Upw#APe>Ew1$KYbuCa3>G~Mox&(aDOF>dr0@- zFXd2wW{!JS;S_};v+B*aj){(e=8zTGMbQG3elPQ|l*S5X^AqlW%7)BS<2KXy1IgB^kd_#>V@%@|2uTP;7Nv zz43{hLM2+_XO>!N-`shk=PMKJ(}O)`eKf@xc}y*W1GmB2JEcw4P>kdw?_3pz9KHE4+v-TGW<{OLVsLH(dYsB)rMr=)ZNr z{rMAVacwQ6;ur`H;M#%i$VFqql?*5|qUG%oMLxKWMD?o-(q^mYH2}>@xON2~s2Hgl z!F;7!iVeHq%uju^1gIjBp!ZTupY~fVw>9)ztu)xwM&=A8>o z%dlxF=;ZcSTZ=Ea#Nj)%{Qd`R{V`@b6S#bL^~9~ADz4SOozR4G;wT(cX&hk?mna<{ z&$0Z1DE#m2UB^)9%({4}OK1&tiNBU@uri4k!3r-EN`*}5d`a^q@`8Xpf0{ZJxnM>C zP5tq5))uj5$pzwS^l_2OHV|iQlL>GenoC=(vjp@74_x}hd5TTPc={i`V1f;R>n(N; zJj32uE1vivwp%v@AUOhaE8+nM7veko+xAIDp!NW-jzL+4qK%fNvWRmN`b1NyZkUFK z1^}KB%TC&!iR8mtT%uZDU z+6_w=Bp|f_Tii(;6?zNjAT0nL<9KSLLf0hP&~@0%)rv`v*AFXuUR>V6I}f=j~r zgv-LKSfx$FB4{0iZe`{>c4(-l7**46>WkDo6BSo$&;UN~BPIU{oMgeD-ASnN;f=r|_6LJ`yQg&3X0a?|Wd=3yijX3^LKEI>nRD7ua(W1#uSE z6(>-S@F~FI@dq6_VyF7pY6T9iKp}NB0gE$;l}T`&}r-HZoZ@) z%^$6v*gX~4eb26Hzqwxi4~uX~kftMhuNS6sa(kPvj;GDj=i2&QH0O#(m=h5zt)1UQ zpGd+uV!DmHY!RZeUGOa|8No#y!=f%ijgU7rT}%FGnbSDX4Ed!1MH_^?t-LTEzCH~U z^Su^cY}$D^ZfglecCEH3&(A8({J8lIA&-kC|7o9 z=y^2h$h@I(kgiVBXlF)m{1)i)n6$pT5GG#de+M{>6Gg*m>vD@;uf;2Gry=g&T2{<$ zk}6`1D%jIx)40)(R3CaU2loJv6$9>2)tWkOY)_unI$NDF4B6+@l^Q$CC7r`cg*%uh z!nHNb^A%-bVGAeK<_pzIg{oDGY(MYU#*YaBJ-9}$}gu7M7BeXj9vSm4>MBZ#;%=_Dm-V25WZJz6; zz$SNIr*i)DPCRcN&at!5pvO&kiJp-QHU$@e|Y-Juq?YS zSX4?%q&o~+K)OLvknUExJ47T@T1uq5q`Mmgq+1%KL0U?r;VjZuczJyq9W_cyOGob#cgd}ceK_qh@ojxDO>1?3S_)sf^{-Xll#_y9Rt!VnEHjKG;JzNhIUc9^0Xuzp|&+Xz%=OUCiz)T z02^dx_MoBR$FdxaQBn%U%HO|4$LDGr9&i%37-ZtSe*B*oH2Nn5+TUX3zkhk<6~d8n z!4(5EdQ;QAon3MWbq5Ts`%Cp0Dv&mKRNJG)hCF0!N;CKg1McP@Ax!#iW8DZE%APu- zz!U|Q$kTo1pgG~XtSY20Cl_;?3miinx$3b-PPK9!84jh2we6{~lhx7~Dy_z%@%B?6 zB*@3UVEh)*ectaUtz2l#jy4g%vGJ1Ng|t~^qE?6_F8wj7;v#Y-g5ph4sozJ(Qu*#ad_Gqt72MqEWc+@ z_*a_*mbEd9E<<#Lycw9?hRw&OsnS?Q%-FP488}rLIJ)8SqA5+Ut@$~6T0l{FyWF?1 zdL6yCvGM!#^z*?X1u?EHpoT6kq^1^xeGQ1*_TQMX>pT4qWZ(Qvy%u)6zT|a%OJ6WQ z%l{qhGn<>w2M6UvZS<8}>jJcGP*S)7e-Iy?H(^w{y+pGn)zLA*KptCuA<1Zfppfs2Xp}{>HJYxe2u!vpvx5Yah~d zHM!{zoPk>7qW8QVmOsSha9*~hwFRX8xGGUwiVnzi)XnO*Lpyyvs7OAWK~TsLnZ|}; z8JaLVa2Oh1Pp|AvyOKDq{WRnWjo;0;`%WHWwBC-LG@HkD(CC0|!tJx2Jym$x++CfA zn?ZST$-@&6Pb2(8(S>h!>R6nHPRG!waNB=M9G7$g&E^YRd?Bmo^i2Nu6G7eAuYrQA z6_1fOL&Vj{8yns`v48h2?C8jz_OlKDC#;3(X`nkhVS(i3Q!>Xb7x|l4_!oKn2(Paj zHU5X*PCJHHU?4pD#__ig-!z#F?83h{X#RrAx<}k1C*6-Gm4#K?;{gQSH0dks zrIrqUdZ+I1<8`eN1EP%VA`1$VP%a(&l{Ttl&0@wJj}bX6j5{!XGyAW@s|AZcU#r6m z$*4Yr|6EbB^oVfRKJ&*Ms6RWqaHHO})vC@_$x*hjwu^n9Y+1?_ESE+qZtL)7 z!bLCYW1z$%J>U}_s`8BIHWT2B(Rmy?xP?qlQT*e`wI203Xja4z)G(kSxcqeuVEIbGP+2P_pz}0 zQr1y})4!tl?_d1-t*i^9b0?!#nwLLpY0~@`5BN|~++v^VC5aHuJW;NCeoZnn!Q7a5 z^dbZ1V3Fm!>;WX7EeO;8l`pZE#hkmhp>W&yXwnd~F6YbjXj9YirIfnjazZw=v;_~C zmrRD{wQo9gLx&K$j6h_M0UvpAScq#w)uQ%n+l0DSdi(4rtehqCLp~=!O&S^uYT(R- zI{b4)ylyQTLY;0(mHlo8yru;bo)czfwCUfdWBbtg0Q)XoUgEvJsG^iCiM3X&xKI2a z%K?th8{C|n97tS&DuqYD|l z3KY;7%;1xR7AO`Ca6GfNi6#!z_%cKoztp>|*J3l00y(E&@;Kmtb$>8fpZ<&ry8912*^D{+^{I$Ojm?Q;{ zuZ-|HOjZBg{rsd=i)S{b>ZHRGO7Wu@j!krm?6z@zH{NceGqVf}?QasV*yuRWMv>)$ zA}#H=gces=Be@;JR;Z$T%lWGCgz~=^y1o$S&i<)L$D*07$}pf!GiyzxK$a9a#Wa+d zK5ZR%7`;`1pja)3pw~X3P8``Ijug-j9O_Z(Je0c845Z&n}fP)}E{gP$k9}>1+t|`Z=T#CVB;nHQ} z;MT8lOzVf9a#ep+nfjK|n@pzBatnmUR7MCo@0iT>g?wj{38z(Au_T{jT^5`XmryB) z*dnCZ1+zne5<}LAWzP@|gF3p|8j105a&A|p-Ir=;x>-LWRTop%jpkm^>OLIQsTdP_ zCm%4n-XT>SCec7(rf~|RmmYOGzwC~aF)3|m@EaYq%1V4O6G8*#A=!A`kO;oq%vn%Q z)n`jj-vZ@{mYuIiht^7qxsyg=b)2wo+m=uE|3s#Yc65&SmM-u&F$d) z7q!Fi{wVb7?M2g?ax5alV28ylGb;j9WbsJXED`mouz?%J;Ut2To$4%2`iuFDI zoDLfuPW9)HT{PLn(U1B$>DkqEzAZ!2II@GU)rjT9Ks&4RX){^J%XoLUY}cP;wiaR} z#d)}lX^bh02L2!PbNB?F?qD)+-=v88u%wz4c`3C0t@-x-Z0`m31pEE~shJ^x&5Uv$ z2N-;1M<0gB1Dio{_osf8%Chvr5_E}jrT?HaVdL#L#f`D48Q!eadhK#r21?;yH9MQ= zrF#02`VRU7a~3Wa?=i)qKVYX#2dCe_>G82zuCpqG&HcU+VBbtD!h(ZUU_nt6NYkv= z7e8!`12)`s#qH`EviG=TTK)3R>}DsmV}5=mD|MU((0NHahhXC|LEU}9_y3+4t1t50 zR!a`{_JO;*e}9p8DvU=wbw zh|Qf{wdX{NzZpWGeOqu+HZ5+poepNS6OV2vWkPS9@AZ?{{qV!sKT%=Np=fA8^V-i}w42A+ucFNE%2=8p-UXvimPr9_ zVtofC<3)bWA`ku2=8zt$RHo9-^YV(!5GvGL`Uy@^OP2~v1RWjxot>Qw^;rqNwaqcx;U?3!S7~Nu0C-1Juv)9bcNTd4RfH}`gvzKe};tByEBg$bXpd@u|5ZGWW@TzVwL_D zv(=^SS^DE-y4*+$slg+@mG=K|XDtxqH5_r#UF=`V3>Rz}7NdPp4@x-2eyAo`JfWP| z+`B8C_wNF0}U3hwkwsEtnWpIM*Kp2IX8nj%TeL5j~1n z3E2=o8iZ+EFdo=ct;+DM@{z|5ge*NQ(it$V?Q3@+E8Ap_Ef{veJU-`SW*&5S_m1GM z_ok_&=hWz?s`Izp{^Vj(IGH+HTOU@}Kv+=-EFQ*Cs<1@C_>b2YR^CvztkJxVJ~94p z!R>DT-4M1M@wWuK>v6u64uT-0N`Q=vJ9o$Rs#a?IJO3iU&S~Cze$jfQc?AF+WVDWd zTtP|@xdG2P{~-6SKw`9Ows@$@uKdJ8LYeGnH962`2G@XkK%$e(Oz#%T?VOXs>vqx8 z@DAEZka^O^1bA|X zwOZoL?slze>)Edw4ju>rGmahJlCu@+#|BcftzR#S$LaN)Uwg?5d^-&hH}Sy5as3qO z%yWW7+c$X}!(b7Zk>6@!Cs;U ztSX`Q6URo5??abs(Tz__&~){f+M5j24wv1QZvn~Ne`JTz=Zy*^M9t{7qz6LMpveDpQ9l1V+f z`8&pa{S)dBW1>a57<0TW5;_TXsd4cRu4h)4ums#(395^nspQmi2@?|&6z5!UdQsJG z9C%9&1>2+5TDu%^I4LOJ(pwFWq*(U^%~pThRL!qcLlbDGM@_5#daz(e`V--)km}Py zgc?W-<1gVgWoOBL{zbOOvFQ#ame^<|8#UiA)(}OTgzd-Xc&3MV4D>Pm?Cfb|KstYC zvKpJS<^MOGm`%@B@~Z)X>fWP4ea^(t;cX?CC-A<)>oH>9fT`F!I}9kBP-$WA3OJNMM@Fs#sfM3cVnG6`CV(SsV8meF3UCJW;19nZ zlYH7N2~X>~jj$O(Zgijf=J&#nq09auCd8f8>&?_oTA_m=QcCZ8Qw1vvmx*_R~Yn~!#3#uR6pH))vlf$^R?-# z#@pI8`uR-qqKAFw4z<~2L#2}%?zH1gY9sM{t6~!)m?O3{7bU#A_SM)9_x00leyBg3qjUDpkFmDW1I{hGh-B2;S2RgW}dgpMF> zB;-Nu+SK%z=HMZ|7s27(oRdBrA3?Zwu@UbK4h>D2-f;X)nS^a7Aka2HPwlK{mXJ_m z@9zF6<>r>F*k5c#QTLMHJ9gWQI_36g=Vn6MQGoHRr&}f0ma;Us9>*HqoXywUEnY<{ z?kNvOIc(p%pYXvicJS~B;huN9SJHCLSFabJ9@A&ic)nHO_IKkUv?Y;*BEtrJCh+d? ziE3u=`jOlI;vLA8_yD6YH=RF^j;aWFUz3@cnWg(N#P%&ME+X7!^~}%vTO>eB6(I8V z>1pDG0n?33<-FU$u%Ml{YqvPUTLzs`P*SyxkK@CV?6^eBJXK}DkLojl4|F?05&PF& zkb?xXQX5ce5rn3_Kz~C4uXx9!`xChnyN^)s0zz?a(1Q&9c%VVzls+nc*(i-Yr6a3m%kbH;YJw5T9Ac#zb~hTWqX+eqkDmk$ zWJo^S3gsd;h|to!+I{Z-qPl)AN{@;9G5#=6tboD_piKCP5Q|HYhv2{v7!~@r<~^tpQ55?Q8I}A71J2IuF$=84?dHV zlVv1zCR)%=mOE|FL%?sc`T&ip6CI!H+ba{Z4c{ zTn)-Cu}bjI%zWIq=egHDGV)O0^H{C5(iReApT6;)eCtWLb?m;gw{X$45NvnZX@^)! zQtG|_)(hn}5gd|OuYbNie|37jcjM(TFzGSS8#!xyyYsIaZcwiGAN&Vh4i;OD@PftP z3BUh`2(VKWA!3M-h*RG64vC-9&gdTK_P|=V``3cl@)hStYRmsG+n&5t%xc z2{bm7A=(iwXIV3k?|yix zO*$Q>BdEdsUukUPKLste3y+m0dEgLQpk#}xP(?4U_-Q?Vp!^X!OPEDhcVw`KX{@0U zPbstUsVCN0b~LW#zK2k_D#K$LmPWaV(Ix3byh8>$_Vw~dnZvZA59h5Xj3(1|=5xUs z-R3~{`;F^U?dlA#n`jyQ*;(h%+3H&sJs5q)4P@qFt7Rc9;B<#M77xF=mkt3nTLM3} z*yFa+OI-!&oWt2YK_5XD@A2yT1 zaP-JGwgKLg$iE-H|n_vln7REx_h6tXZV3n1upE3Ko6yZ|o zDPy_cF&DM$Q^wKq^ucZv*EdO)IfC6R{Gw?d5B*O@M^fLNB`f)~&1^SdbW4BJLKGt+Xjq9^bL1>P4Al?m zqN^1PcGI+rsH0`rW42{x=izfpU)n9NH+*)zfn{Rj*iNyRxxyC$s+}umu9} zrURj{?B#E;xVVT)c`o=FFC?bDPxjjd@ zm!J`As&~Z};l0ZxjLN>TcTke5(r)+ov$E=YUN+t=-Q@sfHNtIKJw*6CQ4a2Gf4H;L zy{wfzZPA6_^p%6x-T~a7!NEtG3#Mh7%0C$1A-?$F+CVR>=HG5VH|;=X5TVH&yTk)? zc!k`X{`^GhEie3XN)wyeF>vCk{&>|j%)tqtVysKscjcSkc9hL*p`>6JR2f=hBB()~ zD+kJ7pu20hhO3 z4BWOq3>l=vXAN1b1Y8BNRy)5usQH2|F>BA?kDJLn4G&TL1l;oTEcgd-wbv8YKXoDN zeqOF@v{cWG@P;j(&T?PfFj&Lwo(_&b)10kTAph)csS|NfcW=Mvm+#X@0vseR>4Hin z4|^-Erl!+-bh9m>Q#OznF+XmQLk>jMmBmGDfIh%82M4pm<%Fl95!?G5b7zl|^%`bU@0 zzvq|G=0yV2sl8UEyby2aBJ}tQQ`~loD(*PO!VE5ru9nN#8*Vt85z>O~K-LYXX!@ib z+PR#Znz}q)k&`ZDMS>?9W$pC55@g9qqlrKSr4PD-@`B$8JW@>&j*yB*EuJcWnYNaHxQ!PflchBqjhEum zKpvsSa$XP4oXym%jv1HH{&QvAv0OfkPnKiOI_qqYK2N@=UF#0}@U_F4Gy#EHE|ZIG zx`I48m7M$Oeg0Ix@!4(TsYVw}RI@3|7;?%#)ONxBdKr(!gMLfqa)#O=cH63+R~At0 ze!JGUoPT`n=`x2#P@H=Kqx`650V&NTS+mqQ8@Jpn_({?1)9kF zxAc;$>*s=k;BN8eLZZ>c0n_Ism=5mlI3yb{_KN&`e8ffozsv7mp0NWuG3`cK=7*Z@ z;PgPF=(2EryKJ+)xoc!se`$a_HB-N(Gm_SBd1psl*BNxvw{IyQPzr(I$ov5Tij-S6 zu&*z8SX3iVRZDnHNO!Wso(B{*`oY04qc-M*nkI-;6GNKLk?`L#?31_ZMsNlej)v8GwAe=*Nyonr-SlHkb*0(CmF!Wl-YAf zn<}$N$YX@I#U8cyEnjl^j_sqyLf-F%?3-3fzAuK*@`}_x$l@gXN!e~ar&4F%`LC~n zpoF$(O*VhHJ@Fn^wbDrZFYE4v%U~wqpO&h&UbV6Ol!I-_>n~|V?g_Uc(;c3eBp_t8 zI2qzfwHyC5#O0(=$y-}q)~0KXEVpTAjl*et!a?NW7V?+YJY(AWM=|9ef^JVt;PepS zQKsqB+S_sz8!hPFT_-mcOZd3{4o*lA+&>2gSA01`t0Vsd<{Ip(87}xALw)@}$t*zJ zK2P$&S17%Y?5>}JJ(wj; zb*62lt7Lh9A2r(odNn#L%oIcsiY-Vit8Cy0Z>!2ggE_y(H z+R!6TpHn7bVpYJDG*Y`}m^1qDQ|yza=?QfSCedl@nEOsN=2ax3Zz5H1UV)}VP!rX> zQ<&D>V=RHC9{AbG%{~)9AtuU1Xq9-X`kuVOkf}jRfuR6Pl1KPs221hz)%rPEn|nrqA*ri}w~p9fX(`Eyp^{T|>J$)BT} z#C;N~q9!)#oP9V%C4fQGQGSh(!rkT8AHBMoCbgm;*%6R!@hFxtO;HhL>o(wK?d5 zIs?+#%Ia#|bg+yjTw*__^k=I= zQn0>Ve^>_vzIga6(SoTFpN-Q&1gMM0EY)1NRC40-NTU*cEfT2TD|tn3c`$RKM|o{i z`?oir*7Yms&#mTNP5gSBOnUa^F{KOR6C0R$VS0UT0^AOdJX$>#{BNB3#^$5HA)>Tl zeLC|<;k|RndmIbNpzJn>s0*g-D`v1&Zfom;g3`=%6D>e-Fc9Vrn%^T}MZjTo`$%GA z*QvQOG?|ueQQq8~UoojU7b$>udJvwaw7x{r}I$4z`;2R)L{A75Nv zGO0%1McucsMS35c8B<2?1gTh}X>si?b1SoIRhfQ14(6y=;M7(6o9wn~swn=#4o;j$UlHJ1S)HK1h(yBI0 z5#yf6NBA;j&6iAvX3w^&Rhef$7u8*PboNMh66tIV7k$k3=wFBWBeAc2YaMTse~mc9 z{qEY(G4r&<;jCrEq{+r zcg%!IwRu3m7Vx@6%O5AN;9iz}=nzgr-?i6|`he2VS`jNJoHg+u7ZPun)AM zmF;atUHy~*?OufcejEsxH&(%t*5}t4bGEU^e)`V}M;#`!Wog!Y80B4x0MBpuF<(wy;Xs z&i;{IqRuEbn!6iDX~()zroa8K&C9BM*!3b-#K6;wCxRM!y@0Kum}XNw;3IpO%w^~>CxplSiV5JGrhL%l1w68-w{z0Zt1v|?D{Z*C!*Z!t;}<6g`jkAcAU#)6Af z)SAFkaEx?t_E`n=+XBVjB-d9RgRK1FQp+_b(Og-X!{BL{Er6;g$eLJD6LInaOmz< za@EiM5#{6+iNBlEJ1(L9Pq|MiSapxA(4;)w-++jkurl{&yxHhdkcOoo(I=U8++t`K zpdTm_X>aNJsA@WOq^Q7Yx3t_Ek9aO9l+H-spw5kK(5HIVz~{+i9o(A}_VCsBp=cRo z+!%xK4h1#yYWZAK9>+)tcSKw9_Et9zlA!A?h_h);t0{f|LHFWww{dP0E3oGSzI*R6 zqI3-ZyjRZaav+P6F>dYi{Bh@v0L>ekGA+UC^v7t8pS9)Kv{JYML>U(Q8a(BpYJCT$ zi?Eg3d-?nvEhmQ-YR(6GM!E~-piVpUg~ALwFdhhalKgwO*hvRc5izFFt*2|=ZRPmO z)ZZ1#8y=zkkHXax>&N>4<-G#3C5N$FcHkCR*&M9_tUQV;sNbF`RZ_aCgA6x$im*4q z?89}(cL1|)Zz{mv0z*d273=04ZOawTd;`ApI6qK)?m)_sy@LZuAD1d$7y$t~47j@T zQYd2y^_~Ddd(9CEK3U4rd0IoFcPk}0c#pds6B@yEiAEDy*BzBTdOuV6eFjaPwSRfG zg>|K&s@6^$E0bS#c42bTaIM~N)SvxEjjlKQ8Apq@=H)rlA#Op5_dQ3Jg7gFmA;CiDcU~ zuk(MaI%YWwZoBqsaww&j>`P0Tve@3hd#xGM^TTwKM^G^6SXU<|w5iDxN=hoQ*~)p` z1Cra`orL0V1dMr}vZf3(GJ^Gnvd``OZ&4inVL?eVB~a_oBL@Y4YODF*Fu%mNvj6QN zg`2;^t+$g70V}~H_|YB^{$=2|uSM_eF*deRy}JlZ1J;8&hF|UA(Y}3JeO|#GI9LzD zfrS8FO!D39m7N_91Fn1-YrWL{qr!;fCo5F3;+Zex<)z0s zcw?1JA-HIfhq-UqBc#iSEAZot8No+ZUL_MT_DYq!sBXtPC4lS?tK{;XuqHD~&QJ#> z7BfuaFV7Hsv#lirJ9rllc4m%D)9Lm1I8qPx$?P6KVNz*of7c`=pp!WI_v@$i#RAjm ztsxg-!W4r7ZKP2S-#;&GH&(}L^zF)o1w|6|n_qyh({LY^_YiYx^%>h;nVHhsVnz-! zRkm{Rx@`RtsMbF7Cf<3&UMG!|B9nCo0m7s-G>qBFO3ZM1_Wu02PY~k&DG8J(o*#Ds zGf<+SNI_dtfAc9C9Q1%mB{pioNnn`L>LJv%w#GZ}afDeGf0|Bto<2}n{J&0&TZ{0i zkMJoDtf>F`P-=kdI#?fXp%n-z{sw{RR7wZLyK$7|)CgDtcAwp^NZ=70#k15=Dm5`i zXKZRT>4#6p>i@uaq09Z0AQB|}LJ>h{KAR=aB-Mr$qkA>g)tYldBG`(O?Q=r}AM3IY zEcBlKCX*=An=bV6^6r;aK?YmPr*mD?xVG5_fagCgaK7<|k#1)BBWP|dVAZ%4u?Ie) zZ(~oF%N6WZ@^*muz@w5Cf^W;Y!!a(N)JX=T?_PcuP`vEjT-ZvorhkIoP|}g(ocBy` zJFHt4%{FeCrKaV`h-^2H+Qw5Ixz_~fSxa?3699LG>0pwbk zk4E*a?VlbLKyS?HRM#~ISx)I?`>)%k0;kIg?|&<&&_5~%-r&V(4gH5rkwA#w?jl92 zSc`;ym89}*(BRF#d=ik!JLu6LYp`fP^596s1lPk4UpjvSL$iQ#D@S`6s4_TZkFEVO z<2C!Fi$u69*Q-pl^&?(V?4K_=0ujuW=wWdw$%okxnt@97 z*BaoRd&JIc`_26lZ_ZYul7(i#O#{;pY6Ye~QEbFE#f_~koB`9v$o-%my-(!o0+SSgo?_t4)wLypml2?3t17gvD z%Ybd*f`NSRawGhD<5t-F@A&aPeQ;xU`P=TA07RZpY&|Ld*%O1eZwbNN4L0nRD0>3f zfI#Yt0h|Sse3PH&Zb{#1024gzVsTSVX#|{)$SGkq+cm!vL0TFamnpbJJ zty0(TLuKJu2eEDb$`-FG%^-Ybsgy1^xaGM(4aKFL)ya^HFMs*Pkl%e}gu#3-W zMGo`N^8BPZ`^}h~9_&y-*@x+ zG=kHhNc!RHyJHwS1N|#|5tlmHzTx<4QAb`D;~ai=@@-yAPuLphn}+giNxB~u8IgCH zlH5yv=a2H~etP_rV_=u#L(*tPx9vAvuXeqrtLM?SUW5<(>JH<_Bpo%xN-j?|guJ0d+jy z;^cjxR5%3;bp52;JLC2*mGU(`@!=GtKwT#h)Seu7@E zncd%_X;CSI922a`^1(6W2!*DC1n6gPhNVg-b^0!Jc2Cx1jH45>`bnrH%2~{fQeK32 zMNXBXUmuZs#Lr6;`BV2RjeVrEWF#zkUf|<1CimT$FWAlR0V`S9M8*s_O{bzv*tDQQ zw2b~ypXLKly5Quk!mcK)c7#R(UFIP+8N zf0z$Ab)acTB zvAqtLv)&gKrwTWl!hU0U#oNo{8Vlioc7Xzg+xFH`)f34rCSp3vMIrR$Lb?8pCr6?j zm_mc^u-hHL3nN7n+3G<7aBb-_$(=vbFx8qX820)Ey3AGP=q6r~Iw*wez!yBnE@yfb zINV(xL6AQkGKBj3xXFz`T&N!@y)Al$opb3E7y}$!)TC-=C0knD9AmRBUgogN21?dI zDgQwuKU~m2I=}ke$jPLk34{fg=Omk9zD6Q zoC!`o18;B2JiI&aGL>#iK>tqjyT;9bPY)RBqI)w~E<(AlvCYhKdacGCoSl8cy)W&S z7Z=A}gxgN6|2qhWZZ2EeNbMT^0M|`b`z)1)(<*d#i$MS7r>^kVFY0tZ8V~_0 zcTO(AMLW<~EiG zkJR@0qFl#Bo<_?6*hAj``8Ddj<&nO{yivbS%j(dk4C6pU=xHT0#TN47E`~nAVj@vp zJhgjb@XlB(6~pP#^IVU6oMc$)B-vm3^k_Kn{8JRjTYBVFRx$obB(e()xm#HU+fb2L zfBgHL3~R*lKt9ko{G(`kx3y%=V*S|9my;Ur(|IFo_@`zcpYuodh>SXe>aMFda{6ln zo1)7>f;MowK3(p8(!0LJB{7^b?HE4@_=FV!>Q{Y(5Aa7xXopM9IU&37vybuKe#y)zF8k$fStGx+?O>4=hRqIjbX#FwhbvbNgyNiQ?V74 zy&#MmImaPl{CUuSQOqd^I+bXXe%INVg{*;!$$2{QhSX#OV4hn!=P%8oz8@HBQybX>K9Fn+c zYyaS2Cx|tn^p84(lWFg&sU@J~^z(yfSB+e^WD0r9-&exnES{T~z<0vXJn-86S6II3 zV8sw{+YiiZXNP&rP{yYSdE(&OMfcs3`r|1UbCqIS~sBG zS(mx%Jnh+w&_Pp*Y?XYa!zHZkzkh!=8e1MOM>d{17wvz4pl|gwp!^AIf@~*CRd|Ar zH=dgbdR!w5cdU}daqhd%mI;UDFqZvg%&xz<+a+hu^yD3tnIlYmy9JWMtiv3~mwc;k z>pWrD#D{-A!W#?8%B~GC+@R*mN2()AoY;eMGz1zq(zKB@?SGFt+#nRHN*jM#h}+5n zc5tJBLZklY4jflk*Z9zq0Va`=7+^XkgNa6qxwO;FH4MbVi0l9nPiGCCclD#3joA>z00dGw1EIaF)FA@TQsx z9rlorQp<1GPe9w__sTQ%wLz6w)}l~E(yYK)^yFSN^ety*qU5UIj{}SgU8%6>h8(;i zKKH>p;@8WMimh!tMqGc$%Lb-YMdnxue}Pn|`UW3p%qdGj!N$~Nj+}G@eST%XmGW!b z&M{xcQ!_1y8-Cu$lRU`!hRQP0EJL7y0!!+1e6Qu61E=ihr)@rJBmDWLmu9^HJ)Gw?#LHY!oi8nu)^e3ZEDz(Daq|*(X)fl_B2o?Vx8F zXDXklVxs@{M6nth5r%HOsF^ircPuutF$P9YxA@Kb_W@PP+J--Wt*s$JbDYuopL+V8 zy}eBBr$hD&shN}8U^PcvUw2Tq@xD=v>NNLEJDcEO(dmR1k8sMC&!Xw5a`&{02;ilE zAa-mEj<@jZ5#crz;nUZJS9MJq|Di6gvsmx@EjJgh@s`V1R?L7fE{faTxUKMKXJN z@@VV$n}mfja+JJToPjto?4%fzdkDB5nBX}cHs>HnP}``7v-yKeR)}+gdvPAJ?FWH2 z`i9@=E;X}SeVN+3O9RGivSZ)mCo?V_L>67Fgw|dmWe{52XcHu4(>vEkyeyHj_|wrY z{B7Os=i&pb(yh)#W)N8-OpeYb#%BL~XJyK!Shx1?FJD9W@(}qR$qpM4hI)Qy&mER7 z7%~VVfo9BQ^UHMY*BB{U*+lF21dJ2Y?=0jDB3n?YD(vk3W?$>v$AhQoyMx49-A-tE~>7uyG8 zxv#t^Z}Alnkz!Wz5Dj!Oe{v_P4~tvJA%*Wd*y%$fh%(#fZ2FQvcRY2ist63XSL^^%8eMp=i{!jz-lP z!y^5sBa%qnOgXdBxJ@x8t$C^m1FX#DvU3jwCx2y+oj}>5!ZlaGLc=-tq=6<_nU^Io zG2eH5G2Z}HA}$PzNaV1U;)VHX!1_CCgtc}N58(uZz8!(f=3XJ2lqD(loqc~1)0v;6 zzc6fY#wMRu*j|T_oQmKaHwF!SxhH&DxIJRtPj-oQ4WzRnmryY;DPa>Q_P+W1r&o@Mh0LNh5Me+S&zc9Rr7C!!vEBHxNpc>F8f-1 zhrF*(haai}!4=GSK=tu4-&SXiHB-;N4MGE#p=4$lY_+oBX%`i^$_!&ND}yOR zB@4YgN%|CB^rVWapV0}BykYG!=MV)wdVJg@L&+Bg*ZQ;J4P|;+Se|GL44_lhH)%4Z zYZsNi8^y|Geu~dW5gvC~Q{kLX_y8Z&p28`*mnO9|BF_`_oTQk0OD9uPo8+H-%tBPw ztK{hDFj^wvI{BoLJT=C|UoZXTZ-ZsX@?2hh(ol^nnbhCfNSgxs5G9^D3LC=AR!YwZ z6Bb7Xl?AEs`{CZrjz6+#v)v&un1dwLkHNXGTq#fWHc6{mJ`qvgAks&xTBE2 z7pVajGySiVWMtRZ&4Oq|fsuma0%m`FXAQjCf{qskji_gta=db@k4vH15~#?|gM%Yf zbhG89b}6`B{kF5A2!pQHk6`B%R1M&k!@a#a;CHl^Gu@W{b1fHQi5Tzy*B-og{io&A zt=A9trdT!KZ7o^@0ux1k(Mz<jZRE?Ws^XYNT{ohD5@H}DzDBl zT~PIf4P})Io8*^>n8<4FD>=GCUIdH)G6Z*SW#;XAtVy}&D9o`Y?(*MUpIJ2v5ZrHG z=CtHx6~9VMun3^F?OBLz;_hK^a1o>0p3X0+BBY5;KQ=tD^SJ*Gr_x*LjIB#Ahe+@7n@R9XO_HCV2OlnlCxboL3fYv{Ete z<${N;RPRMyNl6LLD_M{ZmGjD#s*R&WkJcJ$YhO;pt{x!)--}uu!q^vC$q4XiV}tsL zdR^uK{XY#>Y%qGcq2c0X7Rd1`rum2N8%-BIsh83oyOUmXy+fN#|7G_5PfGl??nMe_ z_1WF!q8fdv%!kdjZ(Usv_Lm-aEENU=8QDqjQ8%pfw;F>Q#HLSMe>Sc72w&|gm)}|t zFuqE$KtNoLGU!@2-ma&}0Gk3b_8Xoj27^GvWhkqUIs^KPFmMw29cIkInqFrA95HZ< zCi;wksSm%4s_VI}5-(0nR^Nx@0F&u9;#bw^(%j7Q6^Bew3X@-(Cm)A&MUtZDsv~{O zAVbagM3+-g7hcC_s**LodE8T7Z0=w5b3>_KFu`B}5GGkc&+M*wg^_<}&|O8krCoq&fnsFU@l5fCbENQ&fE~zzhHUYGvkIfV><;J8UPn`{gt$M@~Q(2*Zgwpv~~iV|8~BB@4n zmoF>$R1J@bp?k%tuDy9(#I2HUsI+c@hj1f8cq|a^hH>oGP#%^tOYngMmq(Jx3%3QmzOU`vP% zxuLlmZVhpNL5gq?fc{LV`uMyc`UYD_-e0>Z$8 zFz?>^rbe&vLD}CEr$OYbyAe38r7z;k<*$H+2_HXU>g!xuqPFBDNhBU9jkI*gwkRL@ zzzLQ)IKrsXB0l{hf7bJ!>?J|gNbSIfXMvAn47dqRc?4>Co~5!tLMOWuzRGYR$Dl)} zwjXjEzSW#km_T>0YS{d0Ori%`Y@xxchtXRPihqujdkH=;84(x_Q=W)22-u#yT+dKs zx5D38iE5~S&MKCOSd6ufA$n+BR@>z^)uv`TnK`mA8P%}$hSq9|pupg0XcD(N8_eh- z7JC0+oOFgVJv1Jd_suChc6j1m)WWWI<{&z^7{7Jnz@(vTe?rO0(L#YaajDz)4%A2m#Rx{xXUd@=N?vFCt zp5rHR$^lK#ad1#ECC?90DI=L(Z5rUnp?(?k(}aU+5qox)zuVMl*yWJPmdj!?RTXYe z##`38OcT#XFn2e&#|j`~y<9%;yYF3hUZMce|v6T%RGx zn*YFWQEk^p+Ce};;GOGnQ3;ueA?BibUrIll*!e?cfNqm6E`O16)3&Mu z&G3`+LWpK9Qny+<8FC{Dg ze-1euYG(JH``5wu@>%Fvab;# zYf=iyG8p?lwnBwL){u}rBx|^K*_#n#B*G{=V;du~Os;(!@8iDj-|w6`XU^|;X3lxe z?>zH7-{*t_-yl1U>=lndgelg1`_T77emt+wU z6gEJ)5)-U8$ewB8I^g}?tjb&*C_kCse?uofPp>Uy;(q_F*|-d>GF_9--41R(wXSM1 zOu!~q1kWmT=gHK$#b@2h;5424nzLtbVzD}VXcCsQu}?8tqf;9;7239)3bC7daFDzvF{t)VUR5nkwSkw zZ!kbxIJB>;Iu03HnMS{EpDl6gly^JilCr!MB&%ILrqxE!XVokR3~LZjWL)-2w)~Mp zK~Lkiv9 z3uFN8b%!+q^3U9y;kS}qkdo&G1qe&5AJ~V^=6VC%!GWqpqw5lY%6!H7ZryikfE&>E z?HA~!w2JhVjX}8|La38Q+-c6-;0VEL3S``pv&ectkVLUk-UjO*=WaA>@e)Oj!c~*q z9;#;7t!2}WAz(m05?;=J$jKJE370tiHe#S@9DO32!E8M?u_S~;>QwfO8x3&XLR#JQ zj$ZK-GC2U$sPz?()+&Z5rBkbet>2?kh|3=iRx6{pHN4j98yawomOOZl`?5Q3O;Zf2 z+f=~Mqu$tvxoR~G8S-U0V5%8Znc^}NjS(j&D5U43Oa?t% z>`JDwv*(1w8jG_7#a@_yj^Wq!l^?>jtAN$But_vC6k(k^dLqv?wm*?~VwU~{~oiSo)whh^+ z04y^N|21!tZ#181Jmc*QnN_YTcN1km{133(V+pGDw}v_>ZaqhQiLQ~X9IL(z6~2NK zFi`C!(HE$dxqX5LbS?2d1wD+6aE8IUt%LCQt-N2@{ai0m>szO}OsH-Lo^TH$OWq;P zGQ)mf7wkQ5COJ1t1sQo?Z%FFTFR#EpYzxNmN*r@MXXG@J@;$uv;c=9PP3}0`E>ofx z?Z-$rYda>t$>HA?#OdK^h8;v#r$X-in64KvKH?;`|3^``%YkSLe4>G<1n9q@`!$}b z5c8sS&~={*w0Z!n1$eC0Hqdy%rN_G2;qytuBtQs{9gDZ)04FLtla}E)zc$$1y-8^x zZR|*$Xp}f+21Xw^ljZ4>X851kBIun_%i09r%EtcDY#v~lUTEZU{^jBcsB-|MFz9Wv z;L&yW{Ms4L%5Qw_EzL1bDuh{mTT49>1ckNDwjVt`y$A?8q2Rowu{29rpZ#x6p)!Rs zd#dp-FbnVlMl3Kz?mjD^+8`bcXQZrGU_Ivx^HGTpf4|xa=+(@Caad$5xpl#tc(C^h9p1kK$DPjOnCs;Y6eu8IrZ;vw}!}+B2AwU=bXo0)Va1(It0L%T-5}2?J z{?0?W3u4q(j>T+_bP$vwECf(Z5DDq@kU+E^eE&m8=EwCO+Bl(Iv6n zKqZ)*KG{VTzCw?<7mT<|Ji`9)NpdeGB{B525+t|q?e~K#0O&vm=*ZE~ zM|@9bvQ(oNFHX`~MSN1dpj^xvvzEjD&gQ!lNzPOtqd4P0inlggJFRU?2>j-fS_e!K&9dSZhg z9$ri#^^zVhF0MuD7S6kbO(9L&kH!J9+QMQ2@5;iTtXI?ABbjC?&ffkRgs9az-2MYd zD+>yWX5#egIe2*y{t2$Z!8eVKzwf`czA`A>OsLT7T3+rcf1x0%AGf&33WLF(x%9=M z+49);!>*Jra4K5n&^Q+FQR+D$D*Uz@zEngaF*C~obyaMy*hCclXpr@|*u3`H^OjL} zof}I_OX1E=FA#}rHp!a_cbY}5#HZM+v)ttR0{&r97hR$vkG} zWLRTe*j0isR!LDux#?u1C!B}>pf=O$f3yT`Sk+j=52e~S-qH}F6oMXJU}0vaiV}#H z^1}MB#Jys5Xfb*j)9yC_cdq;AV+ZB6Rf`e(v;!YY?7`Zz7URi^CcfTZ_1G5iEET#m zJ=pJOYC(55C|@jtX4N150Yv^$MMaKel=n3?)FE|7ymc%WC+DuE%7C@t4_A4R4{Do$ zvhq<<`|Mm__5Ja~h>!9T_F%JlIrq5V&M`7g)cTG`b$bG@f5#7XoCWvdX;`0)*QohwbQ+! z5ad64&tf_rsG~_&r#mcQEcxxT;W(Vm_U`Uy{Y&#Z@S23?-2k*>&bjXG9 zl9Et<1%*F}M5G0fwc&t<^pt~>GaC2L40-R*L#2S%hBH(QbM|4AN-@QuaDUDPN(vaN7}pI`IL%D%BTfhaTSbe)dR;Zd!t2YJ3vHLFj76+T;B96f$y=gq8XvGM$BHG{46 zVux){R7X%g7(i2| z)!(C^JlGF*&2svYk&%!NE?XBT)4$U~)zsCQDou|_e+foj(rm2p>`*k4%!T$?v=dldovf7Pk}QI1`mmPQs_rLC`=nVwpcG6*RpHi;HQ%Mk$a? z`*R@oa-bf+hrRtd1qB7*DgwGR$#^WC#r5nz!Hu5&{wSapw6fwkX+KO8!Ol=S%5wJh z_8P+B^bs9lhlhdrF|2i?*S}5htvR#y5Z;j%Moz$0t_q@ES@LIGM z-ZP&x^!4dsETiU5vlx`CrzehbaujjAIO?~I=l2T;pjUmB)Z{XFCGY2z#F+b;p@Q_| zcJ^pNdlhA6Lo>5IOD3tJil1l!N=Zq{C0^csbu$sx%(wK@-&2#McD7d@va+)BiHY4^ z&X4@}hlPUnO>b<;Y!zaF=|H~pK2hk4>20uvky{5MQYOlVb8YjpO&IByBs z(;cuJ92{_Pa}zD|Fd-*gQBhHy7vqrhm6est%geXGWD*h*!V?G*+ew`)kCdmV&!xQ- zq&T~{_>yN-r!gq*+pl>#=CLu!d3jRXHJU)6HHe>{ezm{Y+DiVvSOeB4X&^pr1SQ}j@S0ae=yy*i6Acnf3K%0^@NBxk n7!5xQf!Wy!`QPpTe|gOGOMQvAEuwA{0$z7??rT?R+J^rJp_%(Q literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter4_57_2.png b/doc/LectureNotes/_build/jupyter_execute/chapter4_57_2.png new file mode 100644 index 0000000000000000000000000000000000000000..fa199fba32c334c8dd5da01f91f4738dd8d5359d GIT binary patch literal 11509 zcmaia2RK(@`~Tr%GK4;MN&}E z?w>aZI5=AhHa;myf;-&0{2s#le|Gj*wJt_>N0_#l8nrf_h)%`GwJDbYCL&_imKiC z#aUMzEcV3(Qc8CA2f;WBSnLBKx$tmRY62oEDz$2ajh$VMftA2L@7C>#y{+E_c?G>MFDABC`uW-RmumN)bskskuka{pi7mAF`Zz3h ziO1M`3IC*K5|ON|?8erX+1hN+#z+{;%!|;xJpPJ}(Wu2ge?sEoDAd)}Ib&!}oQVCF zDBh)Z(eIttx;-hgs^a5Eu6^l7 zBBGm=mMuwCAKd16qfeaI=Mfa_@~~STZz6a#W4chW5=%)(cTz@>ho@~@c+Yls>rb_Y z;787*w6sS8S&8D~Uv76xwrR&QQVVf&mtEW9q|YVs=s_8M$)e43cE~P5Oo-=(2RAO zY+;s=l*C2qqmKpE$1rIysyuJ6YHrTcwDx>-pF@RRQj+oRd_Pr5{G&&#A|f;qyxF+A*lcgTl6itcoEQL7BewQF5rYgBkREIzAL8PA`anwbT@x9c-& z&rm9O`&Qx7B}(!W(&0@HxjDGF@MZTlIFw@;LrCZahWpE0^xob)VrFJ`@A2bMTE45p z-!nDBV`D>EiMrofkl5PU;UMzz@}<7JlCNI98iu#RhDYod?LGgQdu{CZo67p?5R#Fx zv7l?lKC~7V7Ks1yva>;{BhK%xHw}v;HTPi+($ZKT#mA?=dPVFq(HsyKM(96tDM8jx zhL?}8{gaPZf2Ch#eQJCAbCveD78>syhKwxB)ibM~82g@*(Nb4eP*o+ullO-!6&@?Y z6BD6h4RPjky~PD3C1K^|Qln#I=~r_LJyvdY_w<8BCA0J<@ z&dn5hlDdCXz@QW!W35>0wNY?)FO{hG+ut@`YBe=At={_cyKd4i#ps!|$6HD&DmJ(c zqCT>(EB^-fg(L~>?7Y0AU%q|~NmmRb_F9{z!jq>&(~P^PXXNEe<_)p5@9hV8;i3Ld zJ~yU*WNF1**3)>*Pj&J_!eV!-g{f)!@bvVlq;oFHlz8&tadGspgLTtS!o!I^4-6Qp zUc5*P144c}d17CCl7WgUeMM?x>`^O3XozV|=#eW|uAm({Spm)7&# zdwKT#rEIG+KQ&4{R$j6a{jA)#ANX)&X=!P+)fqNOEX5p)4Msm_cA)&4`@~br;P2mc zqX<5WC^9WXUBUIGwYAWQh$Fov4slLn^$~sBe|{4pQyn=pXqcv)_rgUa@6JcRdnf87 zDt01h(ZUt5cv?JXv;>C6kz&#$kGqQ4IZAB|F^EmMOYveR*h z>(V>F=^i|IAY|RagUGE+e5-50zM|pQtb#n|v%Q9oNbj!xu0))kPgKf}#EJnZKMQ98TI|h;_5h-cy_Cn=;8?9!E z!w_B8Go-4slMX@vrue!vU+v+^xZ_>lzf&V7W@hYr#}Gy_>tM)WNEIAlG)V*p2ME2nR~$GW>qpt&LdLufYe z;ZEG~-z~{f1@GT0KUc17rS9wPoyjv+%~5;aZnCqnv@u)MS^!BSnptWs-6t|KawBLdWExnS50Al(buhQO z1_)%~Tg?m-5)$E)OLD%xzAGthuMcL5qi)?KDe&h7(OS~l0 zlSdIa>>Ma$xy?Omc`s~UKiua~NxM&Q?AWmg?dQs$YinO<#-cI!;^j+rK|u<{ai$}O z+jn7o$S*ECn;=6b6zO(J2Zf|c#UhNaEO(Z&u0xnqnd_@a$EAZR* zMSl8SmW5D=X>b>^d+R!_bCjMQlWkOB3@9V%HcJGjkd#rBsIIOKvKAqdB9&jZHn-kFS>s`tk!G5X9Eg&5ROzeJln+CQ!O($n~VSyseGtzwe zd**%)2GMu z-&q9!>x*b8DlS%XbQH?V&oA`ebcW@Xy6fT+1UV13BQjItiCIG|bM~uOfj}icxK86D zL}X-jQVSJ9Xm0B1IR*$?^T}s>cxo#9N0YRWc|B=&Z?74wk&uXp;`QstfJj0>kOK0{ z_mwy_JQK6Ibma<%R$ZFp2v4bUdX^?AeKkdSmpEs~SQ1_t~oW3W#s)Ea&C z97)~Z;YL~8Ys2yktyGzfwLS-6!E|sQ>0q@Iq&?2?^75KEIK<(T&<%&N_)^9d!MC8W z4dd0lwfI_K*?L0v>QytCE$pnkx_V^G!p+(4!ci|p1qC68LFwd`xajCwmWZl9Lu1_ z8@v=59=<23X-Y+PpYy-#2lm9T?%zI6Fnz$YHnDHW6LLM$fryE{No+OQed;0T*R*iD ztRRP+&%^oc5zrscU+N47-&^KA$ir?ZWf21nwk@NiD;;iGhz;BZy)|GYCw8iWo!0%O9E{}esh z@g)SP+IfIMf8vb5eN%1bPf67jfGb7$>&gWz_H7k8I-Ep^5=bN-cnN4FPQt?(96WMF z3YAdU0Os(2{eq-wc9`x_iIBR3!4Fl{;d^1WRB9PjiKBF@1_M;~( z=qjJPWJZ`#Vd3|W&p4A-OO6URPR#%A!FL`(*@I<6Ck=~X=D&Ex2>n)cg-EB$G3R`? zL{SxBX-OEbj_>E^m)!N^`*%ue>cA7{J%^WneRbd4oD{y zfry)%8=gY^f!v#!o8MV0EGbd3w--QI#l`6X(Q@w?pu0<&t;^dK*9+vR3KE{Ty@f?c zQqu8cA0VHBQS^dlKVMzxuJ91?_V%`BesWn4i#?OpdWv64N($0yRBN&R>Dk#?*R`i0 z9UyOp!CX!Nc5QT3?u7t{Out@qODy&)x{U;@Jprq;Jv5N=dHDD+2q!0}36QA9*4D7A zx%ddgd-dnfIEaaz9a*t{4CtDMgXgjXM8w5aKr00X1(}+g2LaI4eEmwo#>QsVnRo1^ zR%$GqXcNdlrH&)G-+%lN_FA)r{37FK5YI}aa6Y-l{HfPBaX183{jt|gr>L`WtgY=c zo;WcZ@2W3*70ECfhs8^uL!${cRVzaT!uzKGe`@mQ6+6R|Xy7v*-1x6Se4yMJfBho- z8~JTRXR_n?j27oZXDiHz0{!DzSh_q0GDFDd_P6&(mfzfMEB0{-_VFctqI0IMi`kan zX`)i3&`pK^y)W~%p*|l!h;MV4E({}(12>}{PgVHt@A>c>2~nI*PI5fWg>5k>Z9Y!d z47fP~q$1(`%EhE}sd#{#8yAmlT#OPkZ9%w5~jokKM}rTEFZne$3O4hr3Y09Uk?+qSO%Qa9{;A`l*Hr>{H;&#aj4$|fDt2*CYqg$*PK`Q1ZWA=SlWXI{_J8T9x zjh8IP+Is8ofwdw2pNw`X|IC>t`(#v}kCyRC#X($u4`O7E)D=uk*vuADw>K#Y z?prxCerF49Nf!p0FF98pUy{xGY5bYftEZ=@;H@>gOo_c&OnFaF+(4EZjZOcz76tv8 zi%D^^VO>V+4gH%@l%FhUG<{ia7-uqZ5i~lajPI4d;N#OG;nNUVDl5)B_IBl+dE1_MJp7I~DEuiY}z{Kd5{;JiFFU-D?JxCzpGg zI+dP|wtBBbs^#Z9Ep2VKS?1&IFCxVH`@O0=I*wWwGp6{Ye$H8_%q{7Sc>sxl6f0=M zs^z?I`2EhB4zGc}Xi$9zp2e^DMQ59+#g9$}g=bZilxQR8<1;cpik+R^I>D9sYii0Y zZfx>(Mnwj-i}>-Q4bvUk@13U2~^cnA;(5 zU9}2I-<9MTah9~rJ1r#TTG_%Ncz$2>L-~7uyUQdy&r*V>M4AXBnI2$BNJ#R${AdO} zyy#}<=GybUDBLL+1G%CG17c!M-)i+B6d6p4s3Vy$otvTE_|s+6UHVHAk;`f^6fzyX z$$x#7`~uU{438)}2C2cWD$9<7Q^InqZW?_ca&WY@wF?Rh19Eb1=U+2jdi-9u%U$Ef zcS4_m@11vAsG1fodN}AYUtJzwbvx#;zqi{CzMT)578XlTPeDyh)#Oi|Re<8hnY5}M z6BGt<{({>k3Plev#rhsc4$u;Z%Z=4@lb42_-FgE>+9rg)Jp$ZA{E2_$AQYz$fC(G$ z_hERXA{w+grzwi@7-CzwLxitAM0ipjp?||J6JIo~jmt$a6A6xS{l)$!e_STY13tjx zbO^;5;%1723x@>ZXu?0ivihq#I;Fp9^)FWb_q$Qn0q8Cy>wZiSz`5oCD>^*w}6ln)Qr_+)w_P3jhkD=GBrAy z)T&s&V)eTQgUBsXuslM)eEHJQ#Xo%^B0O7jx(HKPRHW$SRJdV~`?I>GMbnankB<+; zHU zjDn?O7ZamJorb4Rd(0cIfDha8+Sd3J4aSwkzsH*%Ze<6BH)iKt;ccQ&%qhey zt*m%uXHw(gJbn69!fiHTkc5N%%t`fGIcri%adB}t-_z-H<#!i4ZREiZClw=T1n09( z_S%&zY_hT}XV0FM?1L4w7j0Bpy_}_K3Q7s}Rg|L{0!By8fKYd;*kQ5CiUFJ2+w5Br zG~ABkja2UL?k3}n_Zz=_c@_5|JUm)GNFVE$=CLvXYJ;XPm(Z9Hh~*uX05a7%9Otom zGTXlQWR-hcd@|d!BJlPsnHhHWGrZ~sn4CMrHY3#`ZPvw&U%x7XxSD8wdYR=871c3X zpFs+&vGtrBI0%a~xo^<*6c7;5a(%*!o=I!6dyS2%73+gukPhUHj84FV9D^co`es7m zjKj|f2@JKhwQ|qf)84&1>QK3R7Pv!p>?t8xYl?fo1;eiy!Gfi0JQ45|j0|t=7R@H- zJadK`LCs(g#4NCd%KG|@U|s#rJqr{*y>>gG!=1JhMAp5O6mk_6m4t)@Rt^r#$&)9+ zQL*>wRKeans>>Tc9+uwG9e`TwQ{9DQihU0sKK!E($wFAi1f%0ZxZJP^`kWEWTPhZo zIA$rg4h>V_rP8CB~`?`BWx4Ew5r(zxYONHOt_5@C* zc#)^4r!$;$zUkAUguSV&%R3FRP>eYxC`gDp3>g_2xU##e4+o=QozJRQM3;1uzW@s$ zJR(9Ke4LSy5uD4He{5w(K_H1Xl8?*oZ^ms*?(J@)%xqOI!iH6hK-JrOhofX+sFLOL z=g*#e;o*%8=;D#zgM^R{p-*8~>HI6why-MSMYgaU-D!Qa!XW!@cV~0O^AzX8xb??q zjry-I<0{iN(g?E0t51H6&aG8iWyFkYk1UnW!ABeBUlVP*lpi$(f$2x~3Iac4>v^ri zm`?^05fKq^HCG8t*^${28Z{X%Yn{heQE5QGH2mi4&Qp0H`jAS&#qo)WnG%>e$BWuidHR?a;Pf!a zcuRCxi^0+fnN|lMt_{y3dtNW*=4fri;_7NGxT@_rm$|{G|8y9@WHaRjd)bf+r`pmn z$S`pW7ISC&|Uja_edAVE72?0xE4H)FinKRZqk5T=i z8F5*!*zF=kH3msaO1gLdetH+^S=D0@|KM4$?v<66GJu)udHP@*3UEB$Dy@Q^hXUK_ z)9qlvf26}=Z$>kYPs)ihl0h172ghJ5J1o4B=aduKi%vvMOUrLO`3M~Jc>n=xW9-7u zT>8kUsJh8i->n>(-)!t>)UD%JA#{5KuvqLcRC9u%G~hoIvU<(PLuu&XMAAO$3hZEH z1-r5eY~R{R@FUy*nN$MtNx-Vkp*evF+4oC)q=((IW$d`3_wAV&jo=W#+pdn8e@AJ! z(GSDdYx4uSC1#K};;9!@^~?#E!7pRuGk|PsP4(&CRQaXD?Ux z0ZXWkIpJF{TekAh1S*y-2BPzo?3O*}e}drt{cUCmM+AagNkIV@mY7BAB&s|JIy`rH zA40_pEdJ59bj3C+Q6eHDIRk^^6%`e(YvpjR!5Ues9phb+na5(uV%52J|10!qf#gJs z8i0T+5jM9D=)#r(rm3l^{b}r5C|d#cwaB1_BqoOnw$SE>C|41KfOeBC`31>xR6GwO zck9+kViFQ3NXJfnP^!RnaBx_KQ&I19?1133eszKkiU8b%5)=y7Y-~8Fm+b2+gIc|h zPv0;D=W28h5=p`t3w1b!1}O62_`D8CNKEA5;UR)^uUF}N9-MVjK|w(@hXahzt$+RY zt&Whz8$r51`AE(8m#u6(!%87|f15V+wLc?cqDeU!qF=tL}iS(H%M^>IV02 zp^OhXWPBb=>CLYr;MM*I)&Y+r!J+YCjjw4Yj7|x?4mO4v%A9l#x!0pof`8>)|Ef?I zc~Q#+n_uy7v=S2OzcRu@&!jY5(@&#$$-KMLLGbu-=5(E$0xPGqmfui-ZP@W}X?t~r z?}fp~$a5|pHJt#iF{I*a1qyqcFdCp0vQDbl$x5FX6{fDm%ou0qLFy^j~6bdl+Qpfo0tf_ z#V=aKzqHm^*jDatvX5k`1pA!Z31rxF%dqs0bEu(+#)}uXB1T56`tV}JV(vYAdaaV% zfYq9_hr^?HHzcdWO)on6dq9)GeAc*1S9GwVLjc5bJ>`Lu&8 zBPbnLGMGYWX=%AjfAQ@EjM(e7j%s19LoK`-VzW5gPL3?PuFq7uiWb`D6}472kX5a{ z(V8yP*Fm!wL2i$U%1P$f*w{8Y+K5)rS|h_#*+p?dK^&3qDiEh=>s(rnCp zZC#q9YPrHxp4ZQ=cYWssLFeO%#yz`>V*#7L*Q(z?#m6-|j`!s8`_ja$x_qh4(cg?e zlo#E<;AheJ<-d8WCeOfnrhWaG;ER`oSZGE-SOw+@NttAD!F%H2=f^`xnPqS-TXkC% z_uZ$dJxj}KdV3kUEBDElCR=b2z)}N)-r=n`I~0T6$rnG6!+Y8it_t4q{9_x(!R;3u zQykK5@y^vMluRnteC#!iM4~HmL+ei9a-_Ob^pc{NxmJfqP=|xOYM zet7P>2-r6MC(ic^K{)a!$x@ts|2{#=L1qWD7lgkuVe;pTiGhaN=T18o9HhCK(R}wq z@l;O{J^1`k7cla)hzw{AC(R%CJ$LK%4t4rU{iuBk-IE6nY-e% z6PceAT4ibF7@5@V7A>xt{b@WMJ+fje80}}17|>GH$m2+5$Tek}xebIi987CyF%Whd z#iDfqu*`jDz2>iV+@ZZWYrD09+Qk0q0=D13mRi1{RrPfkBER^%-~o5m6Y|<0KWI`? zQnnbLg=fp_j9OG2il)CZlkR8&YsG=>dnhS)1P+W{lrVvlyAp)Dqa?%^4g@2qBiVr^ zI?DLpUQ#=r334FYANMCFgizP z=RoD>I0(oG*0X1?TBd?=DgWmNlP;7ZNJyge8!QfB?%cU^1;E);Zyad9$@|-nl`0vvx<}b14^Z&00RSi%_^;l#Nd0vR-a%N>V!`*#ig1}TtY z)L}Vx>Ch^+9mKYA+nYh=kq*479V* z14m}~a9oGMjg?qiM<=kVN{-*4loAIAr_}iu@#yF%sUh8|{>>G%(m=U5X2OY&!Lm`) z(2R_ax7j%Lfjw8j zNU0m{elpzbndbg&LlYAd1X@~J(40a7%ZWBI#PeNqTpm|zGXDhKbds$c?Cj97u-v^4 zzR{Q1x2kpEZn^a8R{Jc*%ckG?_5 zcSi!f0HPf&xk5RlCg*Y%%GWqgo;2Ir*_43d6HxL(r!i%4o}L?5`l5wl=p+MI2^1MU z6f+s2Q%tYakqvxNv^NXXrhY5hS^-6T<_&4>R*ukN z0JW7PG$&c%7T65LU+~igE_9HKir#$`wHDgq?zqh-A`0MBUjvwtzeVUkf&_f5Vk-9o@is#I!rDEC6>Cw@^ zM~@yc#(hK5Lql;CJhOCiTvP5&f5%4)ficm~#2v_s^~sJ1mtNWuifk8A_ zpXfzsUV^5Z63uV$D>QgCpuQmVb)pThJ#Q_#-`g=k&zwK>CZHXh+?q*p^71%m?RtYt z20clhE5|56?~t|h8i?gfy1J-MGP5bJpews!5&4g7{(raZ{zuNWw> zf8UGo#(US{@LuoXoPE|_YtCQI9ik{NaqkYv9Rvb#PfAiu34uU4f!`+Rx8ZNzi3Q8T zA3RPkHJp@fO`Kfy9gGoj`c8IMwoX>&1`l0~9URSVZP=LL7sEp{Cnq~cUS?+N|MLY* zwhpGu9r0Bu@Fo~`lA4YP1SUT6Hwq#tg$RLgmXH#Aq2iXZIpwNBpicO6Zbm7VPEvWJ z{&?D8n@DD(9yPpan$HEbDM0B$ILdiAQ$yRGwrz6eL;oK z-^538b@>LU2UTsY>sY%b`8=#}Nr;R=}%+zlpjh^_~#%?f(6~eJ_@h-apUcMV&v!RhR#$Ne2_>fM+*c>7 z*$&r+WY{A5PQ!4hF<;%GDov4zXL=$N$B-8m8mjy>R@zunGPqPemG}3*u6X*_Ira-} z(h3R+L#2l85l_}fii^Dt*J3Tkq94h8@;*OO4Gai)5-Xj>N=5ae+G$nExWZy+VDfs| zOB_9_q64QMN9&<_k+w>)ZfdRT_KlAT2_0d?Y(%_Hk6(TucvU!(@e(s%wV(~&46$=` zWYUww(VfDB5C0&XoScYQ3~p6cR_1N3I||bHcX+_Mq(7QpSWv|{y1G2)_dE)#abC~X zD1PC0JB8c+eM^hj)!91p-cpZDf5yw0!c4`-Czr=F%2NvBf^7P~Zar1W{Rv+Q9zRi{O+SLtD1ZDV7J@IHBQ z#aa~~l(MArbc?mCqv_<5;4PKaigkP7(nCIc$as}0{)t&n*LhE|(Z}24VC7v!1;@?X zcfUCN`yEEc6U}8i{ijHWPdc6{JRkto#>OT^!1KY)n>RVF$6E?tS3Oqyrs2^;*P<@x zu_I!ne|dhqT)W+n>AXJtw?JKCJ+LF0)B28Zp;m?2-&|aBUZ*$LSCA{iWL^BSd@ZcMMX@A^W$y$r%ylV)_dsN+7?GG;EE(mz9*!{ zocLi+<9W1^?z%nA&BupNM1=mOM&_y~nd^z*{@I~rcX#*q@87>DriUwSH8G=qeMKVp$du1>OTToC?Ok6ygMgptSX6kqB zGj+4^@+WL2-4FTsFG5@B#a!s%Ffq=qtOS>ou)r}ue9^9UAb6@0j7iA!RpWTF+Ogbf z43C_gJc-M;7k2gT-kx-&yqg=h-|c(UbaZsB%WP-F-rn8@hK4BU_wM!lv0qtVAFQ-~ zeY82z9z`QzHc`olaNk=*d2_n=&uXkRfbn(Zchf!!GBUE@(9q_Af#}vyLdT6!mgxrX zu8fzVtmQ&f4JOgE0e^1rI4*s)ooTR|X-NI?gB$jF0-GsTfqL=MSea@5p?nH=6Wk~w z4hw>he+MZAy|~LvdZ?+XZ@h&UKVZ-h4#9i$oSj{^q&Uon&iCrV&CLyAcXqIPmr^DO zi=5kG5j`X%Wbt<_reuxn7q}Z_+;+kh)7^<|)n2Ep#d?h<@LJ)4VBthoIOmcbJ6HZE zT2DSn=bP1Ulnv(s{N#s%IZWp!FnbaEOHY**N38!6H;T>F~`uZ)MK zV>2J1eX5$@0`#ckb@Jdx?CmOfD`idik9J9dZA=Y~JnvpYloT233@r zH2M}Z6kQ`xSbM0wR+9g#_L>iAAMEa8+(MJE;0?3t`SnY3oEu+~uOO|wX5>?#^u-WGR{J2WYPBTAo~9(-v9rf{=X}Kg{=5n;=oq`IN;2;oeJ*+Bk~#mu zqUAOC9<03j)!at>JOr2kyL z#PPkpl&!Fus;Oi*SGu*|i`L&r`BlruR~6p^oBxZ4wWVcu0OrjVh<3;oT*&919 zsLTZ3e^!(qCzm+z_J!ZmEDzaO?LJ(bo$ZV1aR*r%iD9Fb;3Unpgl}bYEKvawQt66c zRGyX%vDC-JMueoj-sAZjJ^Q-Cq7~}opx4aQRI10}8p<0Rn}~PsP$(!UJ~8QNk0 zR~RgB)Gi1M4DFRP(754hQT_^3>H0mxj8gpTRVnotou}aHZG8ci;^W3EzsoAmoqAiH z;l1rW#&@Kr3l&)Yu_}z~S6?wT1eUb^y5+x8tf)&irT;ZOVUq9y!CBRlyQcYeT~?8? zk=!}vR8XDjTa?AvBHCbr*SS-E`{I2i!e6rGTM0`ZefmXYG00ye5l=85ogCakMb$Uh zv-L|(NWdW=IEYfXSa&Q~^J3YWe%?#|IfzbgYg^LIje%Y_1+~nV(APJE$#R6=yUEkQ zU{q@~<}!Ah{!w2XSNfZ`7E_in?aDm`4~y~eSMf;N!o8S7{z}~{leD`$wkpdvMJZB{ zI{TM^huir+H`mvddzSWqx4@&?#M+fi{?X;H*J!tKzo=Z*+7$RkKXNlZ{;GF!$x`BU z7xUazVDPx|x{0O2o%Hvs`73WWrqklTwT62#lC+df=0(R>EIcp8SVNDGh>xS9$o>4w zXYfVP3khy*Ul(4Zv&Uyv{ns)T)5#ywRk}5a5s&Pc z1V48y56>EhR_Ik5t*1V=7q93c`>?pl)@C4G>U^$oq+V~DE7`CYZV(w=m^{o$N?Ivf zIep&#X~EC>D%$mRe!J006K4=O6tn0cOs06 zUnf<_+cO}%8O{eM8BWLchJW$JkazL%eV}b@q>c2MCTPE*$YN^THZCh&*`1fGoYZbq zXqp+@&OYt(;lEow=*yc8%l$pth3ZgQAX_0_@oD2vbCIV^dJWyshDK{$nGsP@QDPDj zusk-c+dDX%En4neUFokD?N=+^EpBA-+r0O)>$WzhpU>X(e78>Fld{4mQ_WWb zn(0S03AH+hmKd!_`oz%Lr1wJn@a)o3snzejds0HK$M!%L3IU@|s;8_C$++px^5}V77^<_YNUwC&*G65Ma2QT66lRsy8Gt zkqYwxyVPOAR!h6%5GNXjVS(YBFEYFH4?Z!NOIcW1MPKa-y3fZr)zRHvS$VAVjgT;2 zSmkdc{mq&hULw-(m)VbNJ$g99c|7#S>V{v9*Nh&Zq50#>OJT z!futk*{-bT36oO_07WG~3+#)fD?T@9(a4rhMe(}osL%>r0lfF?-A%Ne6)u4;vN@wZ2}Z`ULw5PhxDMHSG&N=U-1DB^)mVW1zy)?%nfms!x}C=^0a{LKC%`|TZ3?PHDCxR0Ej zPbL@$E8W8|c&9j03oW>KX!^(ode`W!zxfQbrO~mlM9uh2bXU!|JXB|LL|k0xexz5> z1^CsleCg=9qj#uMu^sJc;jr}Av?qzPe89Ku4;gmJ;LwW^p~K`1Bp;1*KUksK9Ir6i z7{%>QT==SGV;aAm*?HM`l8;|kWXsJGzK36+LYM5lLLMN}lx9!LqDFhKmVS{-sw4tX zNmLx{mj{eV5;1g#KRgauIXNrYtwy<(k1kc|PeM~a46v;*I?psbs(*GL<5UjaF+WBs z3WLi3k4rq0&R3}zj~_Km)J#kfg5J07t}avoW~MzkzkX^{>Upi@xx?g{A*1=Hcn0O_ ztV^b%w_n?wtM%g!xf6c_C!5H9B#Cx&1-x<^xXQw5}a(7=o;<5qdt=QqVPV zo1fyUbD9@M!T{cfY?y=smqe9bqzHBBg+!9iRKtHl@t9C)dRG*_>Y4h?(D z^z5KB-?N6`!zbl=lWZ$4F-kJB*5bx#gUd5t^t)lyoSg3{$qMvmKeMZ!p1c0C9Y3&6 z8mzbj$d+Kd(m=zJ;U(t^*HDKwdLrLXRJZxWq^ANW=6)_N=ixq>H9yD z(FRSS_4n_j)BRNQnT8mx=yP=*RqNl2Sa$bwvFo1Qesf#pt4DB`>9#~R@?Px9OG=_! z4qu2_SuqHB9_cT2^Jr)oeOq)l8Cbh@-5XTDo~L|mGF9XIoP*=d?9bm`*Oxd}R$+rv zy|o5xdNQ^;AcC|(_trN{xCzwD%F4>l@wHTucD31`FX`E$GRl=Tokcn7rdBq|xJ6ZA z0pj8(E2PRt|4e%OD8hTJ%bczO3{VL=U7WlkVb5*NSAC}1wXxaQGtfz1s`=hmy>E(E zogF{iAzDTPElBy$jyra@6Llc}ACo`M{5r@6q-11ggSD<)x^+AMT0#tI>3qH!ml#{G zc^<2Rg40CA#D?bPd2=1nUR#bkpEcgUVqL(Q(}15q%1wG;(~nYStz_@VPPRWA_B|!_a_=b zK?-m0%LAcrkIJ7m<@k`(CJYr!?CcqHmu)fK*bkPClz~&}f{u><=IlUfW2_AS?VbBi z=awo{q+|2VxJ#K83h=xvk~j(!a^$m65}eNy0yA;p*sH!27a#axucO=O^D=^rC*isN zQ_-NS;Lo3jbAG9qPrd47FGcQ!pr|NZIK6XgLj?@Fb)TqTez4%Btr&onHe6mXD*k8B z4c{*>Br?);b6k~xK?B|M+!lq1%Z9b4rbg?0Wg_QK%`MwCCzdhG|7f0|({eAPcqB#d z<^)$sWo5ekLN@R;(}}s0zg_|oPl0YugocH+v>KxTpQi;i?&1`4cE0qM;ljbQ4<=k% z=VTXrf`aKSEiIP3cj1#sT+R+=*sU~@(|oi7mGX=_KI*6NI3{MxHTF^)XpQ(qt|^He zI%-(m7|Jh1`#Hc$uYnzSf1Trg66}Y2A)%p$zx@08eXqh;jenOIb>7KWD=b?5&Mv)< z=XLU}CO<7r5aYdG(G46TBGj|CkC)rY@MABOhN?hGxO?~RpOF#M-Ff55ANJw;&EiG{ z7?y(Z!;emm>%C52%n{3Mrr~SChCTE0s!M#X&wjd32;U0>agT)k>kGd?k_d+NKWmmd z{r$V6qePh=<1diB+QJ>(&R_UaJ#Td^5f?l)l$opk?8bk$qeJULH4mI6H$OEo)?AM6lFV{ED#>?fR&j~Y}1x{Rb>pw^y@!IMlw|e?LCY_p&GSkma z+tZkDKMS1iY*vv&-5#_v8Ven!J6RjrXLVfSs`I$tTcAGo_nUb;mK@q6H~@0#>9&b% z?ICi>K2#DBWPkHjgW%LQ4uUWw`r!3Cc^E?ZcEkvZFGj$;J_=*jxut_R*MnOI{=GF1J`EwH3 zNR&!XV;^5#yN-97Qv?R`B=EUXftWX9XQW5S=qcM9=Rid@85TL>^blD|a7al*3)ZR6 zj<*Ap@@{=G>Qu5C{jM5ad)rJCyMZeb56}FB^OLAaHai?ECt>)Bw*~6VY%Uw<6oOue zpNH$}dAjx4fthDDd5leZKkVKuv`0ei4G~HI+27AHb`E+b$fL@jMN{y*y{)eo0EtJ= zv>+Sd|1`E79dkiz=>2H!6Cb$hRZMJb!|6KBi?em!PtuP-M{gZ+uBJU^5N5;>1O*nF zgTu*=J{)i&5Dy8Nb-#j^pYIG3w*qMMD>IEk$J;Zee{)|EOYU5+vE{N3JCf#6dB8gp z+r!V1kp<2!AKwrV5b(LMr!{PoeQ4U7l7uDqCEIcb#O0RWUM!H45m0JA)Yf{II)D52 z4gHSv(@PQp0*>CCVw6o+Nh+jF5{JUSMx&N?_?#BOHJ@p5;wRf^h)0@`#p}HD%csRSgb@%0jjyVv-eNtTVYIg5(gWbsDwli_!Z9nG;ki-T3cI7tP-Ey zBORW*I4z^4qnn-E)fm(dxvMCHEeOlc&BMdux~0v)z<~5EoY(#~jg1jIIy!>boWSqS zn#^Upl;(TYG%^z3pZOoJ|6lKWvES#6>{gfOt|%+x6=vN705@AP3w3Ai`^|XX;FbTa$@!SxVpE=s#fAmCh8G zi!P)AJyg&&ShtI==3PoHe{{W%5Yo3gVFjNhSR{S6ol=6;=fWr~ER3Bz`b11C9;v^2 zxr2?8!tb84HRXx&=exckpu3mvZlH3IC7x(&aT3-e?J=o56)>A@{}$_|;?nrc!m{Nb z(^gw@6|Of0@eE1%K0=;?y}do7iD!3a<%zuIg=cn_ll(PAFc|dkEt_3yM7zVg?%UeDl8oWJm2no^Nf(O$cSu5kv zw}cd-#W?R15Cmr~;MBNm{!H_wFj*g#%RgS!@;stPc8RDS=p{#6Q?2FZ_!xKZ(n6IC zBjpO2u0J>DWseFeXKH#466;3AwfAYjeUV^SXk)}eJ_I@PsfaH=m$n*pZcko@K7cYs zT?UN-+3YIWOecAq*KSoiB(hl!^MfnI*u1OvI0GtRnY^Nnre3P_W%G?ot)UA9+1Da^t3h=%qbgi$!X9h%quO{okrW|@PSx-&Qv!RiSPvT^{poQCa#cp+=!0682lH>G z;r7U8HIx9cWVf081lVmLTaHgWoFphc9cT`omi#*21xSzo0gtOPXn>_={ro1qDSl9M zcpMgITSCg9QCh-gku&>m16TtCM{-IqlIZmH(}_q(O6}60y`^>CeTg(0f@P6Kt1W`O z>0cKiirG;9JsR<_emfQF+W))}iwC5nU3>p*7<6h9fL{RZ()|7t)${yFsX!+;Tup$^ z)q2mZn~(CZ6A_#UsUp3`4`usaV2r_5qFnCPN!O{xik=M6%9`+Nbls*H%$7siUoM;D z`c_u5P;yvlDhcx4@85dQUTO>VfdJ@bKJ{~`Vusxb80=<$K6)IIm~pf2R!CN%U7v3^ zmO1>Bwix~QnA+MKGchETAbzs>{46Q%2~;?nD?KGHF%w> zh)O{XUyG1Jrz#%+JCB$>m)ZN=rF~>1D4cY}l9zV(zKm<|)$T$&nV=Vuudnab)wyk5 zeLWjZk&0^ac6SJ#DSM?amA_;-*tJzNoN$bs;NSq4OL(H3&1UuoM=pr2!C;l{gvGFJ zc4MPEQz8=m&K-mKmXJhT^?VzGL#l?9thOVlD3p7Lhx#Ww=_x|K58j&eXni+}B1OGJ z@Yl|0-ZI|ieI)RMAr5oA+sw?&01bjGDk{cr0&lf`W0v3tCaQt4aqDDt=*nnG*(5o0 zslQWk6PSP+hmP5vAuNb`ps*7|Uv$?s3ZU>QFU&Tj}|c+zuAFI$!v z?AXyNTS{_1ms`TnqVOJlrKkmR3|ayUoh*@GaWOU!Be9n+&3S351fTZYN@La&-22xh z5&Kj%`o8ea?k*L${fT&!LHQmJgcZ0HZOZXQj7VU;v*nV1LbXKXYLu|b#C`iB5+F?y zAsZYib4*tYn+|r%OGU+jyADXZaG@=HE#jH#NRBK5YtFYZO&pf>q{hudFdgjvTWDx8 zY^L&H1qb>yW3Q}A5o1RkW}UxYTU zr(Xe~W^n1O3{$uCN{yrd%_)Fbp;PPf2uyDLUw9(H-Gr2seMuZb;PvtWmVlap1^u=C z?_b%i2tqCTsrJRXx;k`hYv^B}YHZQGuh{{x#N0eW3iAhzO z795P8lDdrS*2t#&Mfio!iT)2-=k&t?xt;C!z$v&xrhT7PEdgdmX~IsXg->alI49=i z<-JS1M}to<-wNPjFdbtz&D$yXYPj$Ge}JUX;RvJ0kok8DQXFN*UW8RhB^Fu zYIm?AwbZTouv5o~)ck1Hv^xRmLI;P183Cr=n5vDZSNOcsj4fnj;sr&ZwXU|PNbzw; z3_UYGoh%B$#g3n0>8YT92bLhOnD6XQ6i8{T_GhBOXZ-5vkt>qHpem29I(BS}qJ0V zE9=pnfEyAL?MB9BQ5XTcE6=A_lA}5|_c7Gv&bj6wmgmpk zob1eL*SSUUIDJa(C*B2kzvh3w zpuonYd-(7zFf}jv3)l=t&dtq15@!D2zt-6%KWH)NaLQM0X2!Ly((_}=ZI?c}T$R3m|NhU=(4}vM zj535t{6*5gz@bgTVtgGlv+zxZ@lg~AQz5~@iuG9kL2dUCb%^miRh52PrSVk~7s`(N z>E5H${bdEmS}eb+g(P2|{ww`8^w%E`B8T+G)i=MUHj-s2b-ugl3DJZbg|86N50!WfVBR+7=e*uJ=CAm4FZ@C ze!Q_vrXTYdXag4@h37A)*fqqE93cw;v`o6zWz%hcsmF9v5>SXg=t)ZDW-NP)on39= z6F~VHpG4Q~N%B2j0cFT2E|jnjwFW2MG7&(6+9C#puA zG(^l|oq|@d!JJHBQL>jpkK)m7I53=VDY-aZ)<*!m3;>y5I`(PdT!;T3CTPk+S0{eZ z?Uagj1(G>-<7xmpdzG(@%_cVaVL*9lz( znVt#zp_DbsqW#&AkeU=?H~OazT(YvVr*lDHqX(vHk25@vby>9@xA=atd7t)N{|0

    #ES~P*8@ha=a5{hzD?(U zLjCKPNFv`zNI0qQi#!$NJ=NBUH#9cRTK(hSp*GMn4$M|MZW4kQt`)awp$kP&2*@H= z7IY;z&S4Y+SgyyRe+x#0k(Fli>zLm*4`7V6KxdCIE*>Mc&89t}}^!lCw*l$E9B z=fBHVru~@U4qV^<@eDl?_a77;9UZwKda=qlllNRKEGUvLIvY?lfhXq)<+FQ1(13?R zFfM6F#Oxit0zVO;4=~&|CaWm}L|*L9KYFZN=btT#KVR^1i{k`}UcL{MJa(|L*TR-F zhl{q~~xmg)0ri7OC6V(uj7 ztMz;Am}EZYsb}q1@nwdQT#bmap7i4OtT4&g5D#?(=Z%5E!O+Ys!5GTJ zJPk$zuwLN-S(k19z(C;fZif<&{5zug1=8R@%z8gNqkVamNe3|kOIBJcF_MA$+uMta zEf5Dgg=UOw;h?Pi4*TJcqbaGaP1Y69oc=14H$`AwervM&bTy~q;_6%wl-ofkH<4hK zugSBU8ymUW)vRb(WV1VS&G{I1pz}tii=u(uc>MC)Cm=LxT?Z00vM#AE1s*vTWH(hY1jO^lDTf`HO+edUP+ z{?J}yiS4u2#us-JT;r;D0u-wnuLU~+E2|c-k6Z_a5*Eui5Q>C;Jx!Plf*4}S_wS0X z{FJ8kc!=xsOLkHTPP)i?qljzl6kEGaKilx`^2>! zGwL4PiTb)aZl4Qope#QpCXxYX8jEP;)xP>0bmE{CD;<3Ol!*xo+&Kgw`TzjO05+T3 z+Qg9`-0;Z>{~{QdMx)#eH>uaP%~iFa#ez4%YL%ypM^A302Skvu@4u4Uu4mjc#Jsws6wUGE-fsD7~qCG?Dcf zQQ=wcg!M5o_<@0eNbL{etHvg>(2X907!gGi)L*?{#qYWF5`Y9OE7HU&t!A(N7jWmk z0bn%5j~_pbw1na1%R6&^VU$82-XeS16HLu}k>4TfVnNTQY>eex;V2?<17fk+N?A{Z2X1k~y7n5IvSJKLg+2mvjq)48CG`84d*`CZ*Y)r>X z)!<+rF_slKDOAbNrp-Jrr>)tusTcfh=~sXwyM;bACD_T$YL(ujU~z0K(1O+FBB^J#2nv$V2G)>PXakaA^FnW2$w% zF7i-7A|A$7IyW{pHe`hM>f-dt>&kB`$#~TG0v-n+T{gAAW6%JsVm3AY`RuT3N+5V`hvYtpcTn zPt`h*BIcReHzMc@8D{+%WU%?mTxk6zC;Bsk%%|(9nwq5If5J{5t8)*6L(yUinn;6G z3|+o1^dVYm>h}Qba|NDLHQa{nX0W%oOUC<2WkG+TO^-;xqb?AvDkC&NCF!C;0~MC) z5H-1gyoW^~=l_drsf?U2L<$iJ`S5euO#U1lJ;|JKBB~=Fx#(gm_oXT=1$N45JN*+7 zCHg(@(2^1i<(%F(Ki_U6*O59|lB=mvW(xtqUH}MLwf+BJEQ|K__Ud2`zxNq?a}yHh zyWf{Far34s&B##@SebNZ2Rw!H#eF}HL2FU`>H?J=+Br3!6ju|Z`ZW(7#l&s_WLvd4 zSn5e09UIes^9CV2l&@;<>Z%Yoo0*ZJ(dZ)p6x;yDCDQO~akM&tazLMXV{F`WaoP*Q zzmzkvEFkPpcjZ*iiDcsGwq@1$OcpqP-d2*$9 zaGNhMZ$s3ZfJvL^t$gYR;Io?$o+}hDVDq`R=&FfK7Z*9<*MYqV1QKGayv?K)<>hdD zL=Jg@*0}*{06ySwTshgRJb|tUY=t5FNrF~NTb|%{-ZtPfS@22-zzIl zD~-Me?;urUZP1n?8+{YXD@Z!GOOX+0B%PAxYC1nVQv%4ZFm47p0HBn`@os|J7z1Gf z)qEAHVbYl&_J%d2J%D>4R^FY;7yn2mPBCBQ2}E?7;oOc@S|?l7fBEvz=jzlHfgHjB z9454Y&vjD+U0FpxV zHaE#Y%>$O|+rysZje#|95Lnc$o4!hv82-@qB_tul#ceuRje)5X3}A!lZ<>CoA1~}0qDl)e^h>D5=^MGo9amsdf zv>B0Os?;h9iLv2AK6@v!C!IPVTxOlBuT{{Tw)loQUryF82p1`otc|I9%JZw z*#j+rsK-au0=17HKSHTTwR3UF)GUibHV!I5nVMD;xXrL-yT&VoAd3|YDk5-XO=uA! z1?H;Qy1~p(*bf-9vrPbRfOOxP@zMqkVIUIOAsg#SUwJ-SV`Zm%M$K0*Cxu{zbO`^{ zx)MS`)bIQ3yX?F&R8S1XRje{y6~iaE8S9tn&=4r+p{FNqgJq_sFvcJzER29y=dhC` z2$T=BVrK(RwEv&EiH_JVGOnuo-(FX}LV`hmtqu;isLj*UdI&R2Ltumat!?9GN1c1` z==it=H!GFDBBj(0^+Z?dL^J^WXs%-N&&|~Nt$q+Js&XG+lO-O$I z5lB8%dU-fwEa2!PTY1!zMUuJ>H4%~bfP%O;C#p<_zpdu~gX0ia3TP3HiO)S z#HM1y8Q?~VzoMM)(uA*&P6gpZ$UDwKdTyl9d%xP{0Pp%%Jf-;ma4{vtvpI z(LLDK80BmFbNQc9VM-zwJgp>T&`~sq6BJyPif@MPy?062#gG^5MGdnkAeR5k=%NA` z*62>(^)r4Ug{? z1LEw=Lrl!y(7+%cbRZBPVkqhqLjjk?F*;`HH}fJ57-*Jig<5?_o1V1QV9e5eq^LWY zN6s%*+4c%uPa?ipj!?jm5v7bw-mS*~J72=BpLcS02DEhl-Mbga%$Rt164EjN3Bha# z=Mo9!65U_G&Q?j`JtPFoE_UeA63mh!==}r^0IYhH`7V9ytV9 zX3C8Gno!P=y%T3%K#N1Hv1kl6qc#pKJ-sZ~vOWk)E%oY1=E&=CyKa$8)f|3;Q;(zh z5tjVr5;D;?<}L_y`^OIklrJ_@XMYP{>md)7{|OP?w9szn|D9s<)$JWFBy~UPmKmhN zp!)j|hC&CajB1MK5{8ET8tvw80Agq#7{Ch#U;o;%Qde3-V8ga#M z8Z6EMG=eV_6(7PX{5zhhLV+{G6UU%Q#nm4YS>t=nBy?Su0=X}c0bgop5TT%40p%ct z-h;Th*ndtz5y`^JN)0&}&;|A&v!ojD4zmv(;Cw=QiH4s)3^^fLpsojAbwQ_S*LOf1 zdn0y#)m}KQ?C9jLlk@+8 zT~ojy@U+T$!UP8N&@nNyOnS!u7SB9y>W13q(9OA0X;YIZcLMUINS!CJ~JO=I-YAebcRvhP}5peWDSwl0`GN_kdIPeO*DE#R;%zWL83-Dw_bxZ|k)ET?*JI!3xZqs-p&e&I zPK)5N;`1>Nq)-K;va(6+m!a)DS(CM{@uXZuC2sZuorc@fIw1Lw08q|%h9e2{%ugFH zE0eXWyT|H1(HgI6;vk=0qTlQRxo-I?n|d&9vh*6aHbzS#VaQpjNE;QZ6C)N@GrODr zT4vlW-~A;yQ01W(A-;eI^rN?;c)BdLge>SeCHEv%yqTz+N;^n|Zw*rqq;H`cet*+6Z?p}65duXCxkRHS3tYTT zCro2<$q>Geg2G2IQSW_D2W>q=_}zPcH@h?n0gr{{#G?y+i6+x$j@)o7aMyYwDvc>DFJy*o-?t}2%o6tm*+Ef#TWks zpsPd+^7^O*^iR}lBuCK4q z-Me8>;NOgX$A!{Er$8YA1RrK8?%jW;29rp`PsVM*A%Q_$zSccxEa~81cgE@bu#4OK z_w2@(oE!;|t*YJkJxX=G%*-CMu#_zRHpY6Js^f!2Hj4SSxR?t(=i5KKyTg9`(5S2X ztW@KqX)#h1nmj>`1p5*4N3{lpaknUmh!U(Guz?kIpuTbE~tF|`Xr-Q!x`s)u*)He3i<-@IAxWC*^& zVsEPNXtg67sAE~CeOu)=sh@}NKirperGz#L zIhj(|ZC&6AZWkvku!>+EB|5MBw2hB{3cR;&cVI@`-4p&;je#;ELTGk={#TuQ8H@wb zgQQWQ{`w@*PcJYt}rz}A4_76eF;zsAQ2935#K7XP%vzwz*^?b1>w_$ovyfEG_0)HZESqVH|S2CfCi9j)NWvH zE&Z9l(%-)>Lo5W3-*@W;jJ<&U`uEr1KCVuU6H>s0Lkp$#NDHlz{PUJ1sSy`LZLGkE^Na)nc!8>tK%H==g)P6+zINF)%SD^4MEugzjR7 z)@f+aw#I$@n848X@uTF$WivWP2(4bj`)gljG+y2}w%6nZYK%EzA$M9@pM-?TE`q@a z2zgL!YtEBXI@%U1OWfg?gRW~63iG6+H7(^#GjzdnKGroKd_cA znj1GoRn9Td(dqOWZakU)@Zkd;1A}2BW+gd&}N#<$sIY%p~g#70F&$_(?6HqiTV8p*|VBpT5_TRs6`(8h9xU75zrO|M# zG>wJzx76IPmmN_w@oW3Myfr7Ms2PfjvbYLryVa>E6|^Vt^niusWeLcE0FeQ5<@Z^@ z6Z+Bn&z~uvH$m71Ie&K>i%b$G@3~zzh+xuRMS^?&CY)awT~eCDF*c2lT<%MI1+)eb zR%sLrxa%7jfZO9c_+1|#(s(_Tz8f+w_pW}vs|Tq7?Ja~K9asMNfC`c$tp@!N#L&7< z8#OjAuJ)D||Jhlvm>u>PL(pEmib7Iz8X6jE>R%9QZtCk}*5s#*kOT+(b3%fdT9LL$ zn&3?Y_+zx7zN9OpT?r4!IiBv_mprOAX#KFi_pfo{=B$z;gn= z7DP>jhrRCzK0BM?#DImTW9 z0TNs$``vk-BHjA#y?;{2%9~^Rp81=*TNL48xym_$q+E32Uce6aK|Fv3yh9s)M^jTX zv!Fo6$mo`65MB5d1^0LLp#~TyC^z3{gxv{SeI7EH7pE}|$}sD$qC%^tHcT8Lk(~Re zX}LQQpPrt6sW&wg5HO4lW`JA@0c-GGBmDes0+4QRZoa9fr}uC*n@?(I4I~v}u(998 z$J2v|{=$zef!7&s{u(LM1HGUi>}|*OVIfeN?V+dx?(F&b4hyXP7$91_PAlEWh$qAc zVVVY5lN?O*e6QL@2yc!%qi9q+V6?T-`3D5(>+45R<31(0lajJj0%?}LWoF4YUxa3S zuWuq!d7X_r@d$s%7Q6*Rr9FZ?-KcYbjL*d%bWBKGCT=7-0x}1V_G4|_3GfLh{!B1m z(gK4<@R$bG0k<5G{=ANLxPbUT@aGn|Gvk$)ad2e!4)W08{hCHc7ybNSCxiHeOwvG> zqyxIVy?sgk5AX+GcEuHrfeZv5J4`{2@O!T{H3uETOa>N+Psj{E!4su0=-6{xKCG>! z$77?T1JL>7m|Y>nj0Q2mCq<1nk-G?zF`|cS98f(Uf<10B_lpMxYKUM8f=uuN4d&C9 zh6{an4!lJAGg=?()G&Y>19N-wzhe0xHpw1wExk`#0_?B6&wCv<5d2X7WRB7*g=+gW5LAJ}e~awlX}F92oE?3{BxVZVB zERhmL>ldv8QB2|XP4tq_( z>TANW{Z$_nNGCNw4N7&{j1Fvv@+bu=|6`Th?2qfO1o>5#>V0R%%C;aJKUh<7Nf0(kFzNr}``&mkYJAwR-vIFs09?wkMur5)`^<&M$f(Ko zbSfg>!l5}RQXgCi(01+ugIYM+OeW#aSJ*E~N`coUah#vD=xpbA=5<}3yTYS`rX@ze zR9lJ^Ix=2eyeeY-3@?J4(g`j$1S923rXkzEI*`Q&p}5Rfnb{0FNlr>-zSlMfhxd?S zVjytF5Ga7Y_r}0Lq(IFY+|f^CL7&V)Z)Bc@;TRxM5i}CfcXS{mlcQSjRR|B#EOmo} z1#I-c8tgcD1|L{GqaIp(xUL>^6WpJQwQVsi$bfvZ5)5-*SAz#Y1vNAX!oIsWS&&2y zb%8t*7pDe5-r?#(6Kt_C2!d7F>NWrRl?3#a1)0yjcTc%c@;Ds^&+V4NJU{#chd_ZP z+*M{1vi>d(r6bZRq$JGH;=;Sm* z1tYhj<>h1Uyt{bjY9j;oJv`ZjpIP_hBcJPDhRU{udYuAD9*#={KBvZ*)sC;{VOAqI z*DQ&{vJ;*g^zLiFzw7;3$hNLPVi_P4OuyKJIRj6lk_1enqRQL@0WF6`Zlq8E99W}t zksd)tPM)h-#`#G4;&sj(MB^c_-pG_!tt}H_C3duOh3G0;;D=M&r2;Vvo(fr?nUpJU04WQb+~0rW#yl}) zxy~%r1d#o|*OKWctB*gj0LysNDQMHH#MdMC*l_C$=8Jea!SBamza^cG6k|PKn*uuD z@5ZRO?L_ZQjD}LpFHQlhHEz#kL@(}}nJI=#-HIta3D6FK$K*U(S$3^2bUy>bMeyhw zS)VaA4zu$rwzjrewQRd~F|o5hF?xqd+8$#!DC=5`>G%EpIVB~d@gMCL9O@6K0ZoyV ztPN(C&CbkxUtX@atsr@;-8boSUn@}#`Q*uQXUMTi5F^A%LJMTpBcl!FZ%hquY4fnU zYpa1;m~j7Km!}ZMGDEE6S66KUWa9Rsv%#I_T9k!9NCBG-ynH3Ct=_S*ja5oThq<`! zLhIU8A%#6bWV?2pj2uaWcNeU!k6=hpPuEpJwGJoL>fF@dEVs5Z7Msix=81qQ}{A`!`l`Eg%fezLT{>#>Zy&9ktvBRO^I)mm& z{nE^d{m$6_28V|WXT_sqJoUXP7J^I5%fTG3XcbpH3+8W*ckhn94A$hXPdNh%i&9Ob zJP8*+YLhcLJ z5K7*HbQ+)^+=2-hZ8623S zH68#PIwt85{w9q!`mI}4o{ncTDUR!CTp84Znvn+46+5?W>)j>Z$Oye3Q1R2liQ#{L zrSQ-yI+cUOUk}u4X{Ie%V~m>sD}jz~a$Tf*_Uzf0Tsskz2Wf2AfC(QW>`67(mJuzZ zELI=jz&MxbaEQ^47=65gdDDktaTRlQg%?d1VBR20xwpl4d^~=SHJsd&>i5Pi?;znu zg#!anulJ?=*Xp^p7tks161przg3l=JVvl@(rRD>Om`A=T4i0fA6#|5VJHkUNj>Bdz zwxeed_WY=E?EMl)7IU4s?rRk^oLt=$_v*nPGrT!uU7HE7Bt)L8-)k-H*?3X2Odf2u zY5fKt=m`~{$qQ;Kf^2h>wfh0x2E)IF*X-lH%>#t5l`{gWAK~@uTe5dAl%IyMF5jt0 zAf|nF^RD>XHMpqiCwzS;oNdV@Y%QhLHr)M{?hCte0OwNy5M2?!d)D02(wAZ}^P(Y6 z^VA&jc8ojc`)VC6iOC-O{Z8+F^Dn&h9zMVHL|oULk*Qd(=I|Af(2#{E5~C9Azlor)_?DrxKAMG zV7~D4#17XaG70Hgp=p0*65;Dc*MX!rxg1MT5)leWpem3Dd@L&yc6h!YEkAafhUXX5 zeVMDir9Cw8)#9FGlXJg+I5e;J-l{h5Id(X$kKnbAO--%Bmw@xE0%d^kYgOy~W7b-J z!W2dNZwgj3Qm_~#Q(v_9X1%!CN{QR5HRI>;=l%h!DPO7$5APB%url}`@nv!8N#_K58b z4J+Gz5N16rqL{$_~hW__+#K6b#9@i^If^vE&c*)PL2Z2R{9@R|0P3-q0-|Dt+ z>}+n1z>WJ$@t$9AuhGPB7hiOMNN=G|qm{E?USMWDW_=dn3Ypne;VoIPdF#F?FAvv`QWzcIW{jkJUUd)~&(T>QO4wpBDk$4X1qDxEq7)fv6x@_SXM#|&Ht5-L-Q;!lI6YTx{2@?W+(YQ;5$kWZ z!%d6c5I#=)PC~|jpK*z|O~uSufB%D!5EW`ygFlC!W>*6}dCP#$p7lpivYr4K{*qC+*W zEdXH72vm^DZ>6uVU*Xa}Htb(?@-@Wsc9TEIDC3t+o`_T*%YJk1JX%R&w1E#T9whm< zRvyi+Mb*of_kR8Q`l6m5)AlDX%e_U^ss%;I3hy{2Y=D@tA50xSqCU&?=jye^jVvsA z$T?+>{rG|NIBrhGFOh{z%Z|ma2Gk%vwZ!Mc?~v*Ng2D#?;kwPNDJ}lc*>w-Zga#iB zJ%MlcL?A5d8=A5=P>`wk_7f6P#v~^|dQ7|KukC92^l5{DfLgTF^kAJ2{?SVqXZ`%H ztDy_BwwVOR&jBU4X+nRdqu+p=nwGe`fwv-K|xTZA1mvRSi3$m0iE7>R(*GE#&pCb0llN}$G@9g2|ffN zNgosh0tSVwUZ-3=XoI$OFD9K`UebRae;x@1gB zu!cV$-dO0l6$(rd!dwOTu!-8#4UsSGU5q5Wv9Ll1-JJYeNJK>WYYjy++f}=s8Pl)! zP+2oYUHeCEj2uuEjV^WkCKSH)(?5@&rf9%48P&^U#Kxuvf%pr{T*SBdm}N>%asB(T!*w&Xz>xl75^gVae~3a)qd2;ogM%Q> z{|z-NIsP^=GfQUG3>PsM2bprghAbd(w1axbARPSyq10&D`dRIGD`Cz=k%JfQkk=|b zt=R~DDnm;JCU|EoE8uB~9?Ra;N=u-X?Qf^)qodEm*yWP7$R#9DIgPN8iZXvyY9DIiE?yl&t(Md;|3 z5eT}?i^|!WW%>e;2{x&lu4L^}J*+42-pUve2)O|suwPEjRTdIaB!uDK!UnuXQPF6)QHL6QpHJN^%YBihsow_%Cd4AqR|K6*)UQ5=J$M_pQoIHa zweq*OkIy7n_#j9_nr6+7jZ>8_!{c58gcDd@>g&-#F<(GfeegVh3+%7bR14W(F2cgHl@3-h4;^sp&H3Wza}P{+!sn|*$kFvRextodRN-BG7@UD2r4Y0l1% zAu2bu6W*fJB<>3K49eB*7Y?AU&b&xyUhWGhBBTip*1IEDLn(_>Galz=e?{#)HsXpT zEj**@u(z;QGxXYjVsK{#vjX0cdGn4MZeihV2(iHacOITl1OX6XA%cqI6r$Dx2yZW# zrzatUdSTT*<=Bv$8j+p_KAbCJ>4(+hmU4lOlFV?^04mx85K8Q1KIKOwi_8W`pqYKR zJ(J~oT!&h+4?|cd;j#31WtkQp8M(a_*0#=f`9h*9E@h|S1M*m2Z+P`8ofJeUnCHtH z8^I=v6%km(wshJdJQ2;6?yJ^q!x*q%j#o zyKbE@62S#RyWQOi=KBKNA?;~!IAGy|tV7%g(-QA4P?lSEw}(Ras2VF(Z1S@@h8%%c zSvAYV0-GCT@HGN|50V=_e`T5&9trR^T=PAoC`N6stD%b~h>t7viR%z6ucMib zP5!0RnBg(&i)XO?A^HLQ#`-ODl5!HrZRj6u+kl4g!J|k02pqu{wh^#F#b6y16kfHr z*JVHO_c!{Qwl2+}C>*d6!nOe?#cOAHP>n6oXnS}RI=hm~D=Z}OOXAQk5hioj*-j{5 zeKCEq2u9w)KET7194Rrs0|m!yZX8OlWhY;%9CwjGg%Z||bR~8Na5a9HTNYE=#g$V%>uC5#p zG8bJtGHBMWEhw(9_XX|%`0oUezC*Ct;C%@y@mYXRx5o?Y{ zMsvDelFQopj@j>heNmQj(qwXxMw?w-d>$|-Cdg}(l9=IjIA_-mA$ZT{U9;P(szMHC zh*MlZ{XKlbuc<3~&=4C#uEUUw@7l4SKOd#r?A~iZZ9_e!>F(=G2)9v%PI2?Yc5pPo z&=_5yYFIQmJw2%X`}KT10qyFvaWGYFHb~T(&cB%fsVRUZG8k3r8o27uht^qMx6q`~ z%e;IHEeke@-*`F32kT^qo2(wkQ(K_VA|SV#8f6;7{zr6g5K~q3@);CLgz$572R&dq z(>FB#1cCT=m=iTmXhJcG#^s!*jZH;R49ON`C4PWv%&+&F9b-vPPst+jjufrV;@ZM7 zmtR0yzJUrfq^o|kmF)Rwm7+BAx zB8~6dv;C)@uI}&Nd-Q}uST{NH!^s>AJ@wRX?dPh7P5X;r?5v=Ep2pMz5aL3r)}^;j zW7Qzhz&aT>zZeDX1x&xvsmn3Rc_Ij9#M|-bF*vhI0{7Fl6wL3C_uCm^y^49^pXVCg#0GOO|G%1e!doBY|1b2i<87`V`q4X9wE`=URe(ncpjhE_5{fxUJmvGO`R zPy@wxT;a%Lgz*&wiI}^%Ma>cjLN~YQ)E`Tpv+scEDc|;tA*|@-p6ABF1^{aIMwHol zDX>hoj+a)N&VAVXG4zSPXx=NT*E@rW&-L~EIzzRyso^$luTSVSy=Amih60Q=)$bOAfEhdCW0gZNEXuJcm2^07Ruk>`AET$Ew zkba^KfyNKzvaddA67X1owoZ*Ya}nx3Xf&E}-0D54y0em?kP%om!ga9-Iuq1XYUHRO;6xqRGJrWO^*~VC>m3vxvaI zI-KS~gpUB56M=p)^YuBG)^emzke6Ti$vq4Tl~YN*-G$#3q*TN+>%m4s20|EG4v&m->S8HBBc1^*0MFJl+gQMvVtm&omtwZOIPSf42aUfHVjxm`u3A(-C5b zh3SJh7y(a!7P`P{SkEBj;X!dcu-zB+>fK>n52TBbk(c0Tioup_dX?~_=J0D>eCNWA zhio)6vW|#j#A5PFtj~lV{*1Qv4(QmS!p}y;7Q!o;a4-yBL`0FEn>dxERZJfXJP_13 zvJj%bd{NabCF8Y*Gme-x-nq`6Ie(bXhCp$=w_3ru#P-->jgZw>H^`iBMO~64G;jko ze4DGF2QY^6>>j>2n$s)oxjMdM6l`1+t?g&iIdHEI{GBU=9+2?1;I{Sv2HPp&JZQcu z*!FLE0ktaxuQlQ>_Ekr`>BZcLjw=cs7qSv+U`V6^lxtqP={Wih;&=!UA!<2#U@5Q) zG!sq@Cu%l*XXj}y5ncU#o_YzmUNvH2K~sU#QEX`UzPr}=>dH;-b=fB=*Pjx{I~>k# z6T?1x{F;8McgW(@P$dL*$y9?t4)?2Uuu*!!H*w*$^-a;!A&%Vtj|@oVXE6IXpcRa4 zrwlPlw@T@kEg7ER?TE;wVBX!({6v^md1x6swA|r}{=9EA%_FR%3^+9q^9Z1E5Hcx% zJiGEW%`|z`fN4dUjeN6rqN!tcI}3|EwkqBLcICWpvjy&KAK;iHk`hS&gcwV*`YUY+ zjMBMyF)_FEQ2k5}x0Ax#md3`TlRva7Zfslut$} z+tR{uy&IZzC}V&0mQGSlK|+`yW|JEldzeKmG+8wMNqIofDdJDPcvOb%K;ET#D08PPokpq1gbM9Env9ml8*lIN|lb)>(j<;aOy& z-KVIhWMLZ%xY~TgjhB2TMs{9H5nvHAh6Ak&Epg2N4nVk*a2+0d_lipvy zqOCccoSa~hTO&nLOG3`fK8yr=I)05ES&e;o8yaN@WBzFWHw3QFnd(N3)9l~5&q99? zG~AYS@K!RqOH8IeGCq&zW%E2ZzaNkPpWqHhN)qig3yZL{85?J3XJ6gYT{>q#(J;iE zIkfTrA0tRLoI_Z{L@jxhgx+aCHqqs}PE24M_6eV}DK!I)RtsYo4%CstFns3B8Jru1 zVp|UQ9UCxVt42^=XJ_Y@UBX5mDTOdZ4@IF0{vl8Wk6fny0T zbjCjx)m!xeC)>urKqTFvV4Hrt&$#`d)>+v)%ap0sxhNuk1KC>ku;iHp`1|@PW<-T3 zaRb4n=>{l{O&IvV8kM`$oO&x`COV~rqh?3jQrh2K ziy58<=O!$2Iz2S-So9HN@9VHz;-@GDPGix~F#HJPNd9`#W$HQj=-^6>)yWGh@W?)a zTNK|;$<)?1GYgYJ<=os{z*Ui|t#zqZED{qLey=8;Ur5j}b%r<@G;rmJP>!yAs~B|p zV9_SwpS9fkVPxO_QMd7KX&D(AK<3$0S}tM1{WjL7rbx?TAW2J0Kf&OVUDD`+d23;C zEIB_vpIVO7WBzmK$({~)NXA!9StJ6F#MFCWQz4ENz!6fWY7Q;3@~zJQF{e)ZsJp%O zdzJl96wuUYTPUsq8A6vatqY&jNDm1-wp-{j2|8jU#3n6GHJrzm#+P04c?PLKLDkGy zknslQw-$W4#PzX0UliO5X7w$A?}?%Tanv*5cU*|EL7dsh9&F4FDO8p zB`m6C^E9x_^iN(8buIDIBVcO)2f2K&jUlTWH)-v%=z~qU4r=s~(NUN_!Jx4~Qnkk3 ze)6pUvBgTj?^%JvFHfgf{XrlH-kQwsJ_m^hyC$bJKJrM`Fl^f2vX*3l(=Z=0_-tnkawWh>mgTfnn8?TUvKYiFa;&hX9%Ko-a7C>7l&yR3phcX#<>f+ z7)%>)<5_m>D2JYWVto8^$6H6@3*bmOkGURN5MJYT%L|K(#)gJ8?;${jtL6+vgUWMw zVrojmW)eYdSQ=aKuYqQi<9}H5y$IPMwZC93A9Cuj695_Y+sMR|=&HtmBfY|I*hC|><+)NeLhBRA*Lr1A-Kgzoq~mS z_wHSxD_`~!ABV;PDXT|d!J_R(j1O_siD(kcL<_{Jd-&y(ySkn-w5lNUJPbB2UChyp z$S%t^uk}jYJ)dtu?kcz$Y6EQ?c@xdK|$7l^`{Ozbma9`7U zMMvw0_eZ6tr9tks7osWvbba035NBhX3PSkKBrch_nfPmJ49*N=gmNx#Jb;_F6&xEL9v>2z&?==M z+<4r%&9t<%=vhipvc-^zhocG=W_0kqk$!_tgpcam z4+a53LF8&~!AH|eVM_wzs8LjGQL}*B(C2lZ#~q@|hbajNnS@kz7VPtD<6oQ4Nq9 zand7h!QH=T>2Ok54IulQFy^3Gz?|Yv^o(emFsN@u9NHwJ(+v1Do-?4Mj05e$HWp!M;uLtRPIa! z-lI$rg_J0VSg3fvc^@{Q_r)^nh7$#e)-*V;aVHe_=Ga5wIl?7&3IU=wv0J}x9pn%+ zkgNgn-vO&j9Ig@PGq55&3_6z@bb(0W2^>}{!Jps_;(o*`Y6KyKRQ*;jIAY40nwX7^ zjft}fVfJgBdABJrq~a2hV2{%OPZ!wo_Y?Nclw3Q H<=g)U)ZJ*y literal 0 HcmV?d00001 diff --git a/doc/LectureNotes/_build/jupyter_execute/chapter4_57_4.png b/doc/LectureNotes/_build/jupyter_execute/chapter4_57_4.png new file mode 100644 index 0000000000000000000000000000000000000000..e242a4058461cc1324ae0ed72eb50b2a4ab1acbb GIT binary patch literal 30569 zcmZs@1yEgEv@E)b<3WQH+}%C6LvVL@cTI2)4#9$l;1Jw`LvRo7?(XpCKKI}E>R0_b zmEsgdt-aQobN1-b-D8J+mJ>&W!-0c9Ac&F@B8m_Q%nkTw`tA+*%@>hS8SsP4SybIw z$UKy7|6^e*a@;=duO0D1?ONDKZ7LQ8XkZ z!n@0DZOlG*&Mxz(r>B3v5eX;6l8Bq`wg&du7>DX^Sdqtvjqu+`)-8EPFq{1SIiU61 zT=4$tkS+4ymWzkS?3UeZI9;;E^U{2#ETeI-yF18Xy~n`HDy_(dlZz{RVz-L5%w>NX z1s%Ps-eG;xhKGymbWEJks3k^;ovI)5tUnlCo8@wgs9 z^gDwQe_Kw4BMUsd_g@Cfl5HdKccVYvV?Bxki%}nfMlSP_hzQ0Xo5j@rbWIeKUc1j9 z=H&8aQnpwW4jz|d;O0QGQlT6^x9dUQSgv?jSXj1r3<3125g!=hQd0W5!q9eS z$_dL`@2LI2b|AU7V@~~7Y1B&+qX6z3i^Fz7$w;T!{Zu&1>1?APdjHB-(e+TOShW;#ZW8kB=X1F?g+RcYvNu)CV%+zhTIE+4*jrP6b*4k#nJgzM zAz%~s?w7a8JwH7{pTFMs@y_V^F>o|;TKJuixy3|*Y->oLSEPTXIloH#*6Ih z;qPqwl@8#Kep}Bfa(UlW+zh(^Ehfy}enFNxWY61OIE5!KgR-al1MsB!os+S9k9|b3?L{e-8@@mtBS_RYpo(Jw>s`Qe_kD=+ zM1OyXr~>-tT6Y98e)(IMol(J5E~hOtBRDuX#pdSb$*a}Qki;Z9t*7lvHmhlHQEaCR zvmFokKpZhyu16Tq?w^*1Yinz(T&9ifFO^8`usxiyy;N3ON>|=;g#*r?8F-!P0)G4x z$A{CUxLUQ=L~Lvr{v_hjqdyf$WXRF>r%Q9qMl$_9)SFBOl1lX2o+{@R2~ui1%vqWB z|3Ex1b_CmeA2WHT3uM!BCUz;c>lx|k>0iBl`|@*zz92aL;vyoi5^2;-$MYmpp9V*Y z!NGzc;IM`oc88yy4KP%@9%@3$v>SxF!!gG8FAtVlaNoXti^FF5%0DIs83hF;o!@6} zjS23Qc7xtCwmgwu2gkpnf>A1o_Eu=#(b-w~d3kP5(?$!RR%Ygt_B^lS#yfx*h`8+5 zKaBh1zzO^g-lp;0qslMkZ1qaRgM|h(a9IB)(uC-iQxFq>{r#J@Kt3xvBLg4e;NYNK zq_`1kCyITq7F{nd^j=i?=O zaLFZ~>qYS%)dHE+kTp+~*Kp|WKNm~2IF@!xKsL0?4-0UxB^>zcl@vJwG{@srNX z&`?xLN(yb5aA*lJEC4Q*V%2=>*(y9vd;eCSd;W_#aYCOi=vk<-oT{j-B%-5}^`WDu zKWezX{b`d(sjwEipULI)MqXb2a)X3~WIgqI27KgniH7jz-^AUy8XA2$i8ldXo*$3> zCQCHX{4Xw?6%`eq&V5^S<3Tt0WdJ?ri!41F-cfhObP!nU%q0v8GxGxt6)qXAC>>X9M?k%+kX`IMU8SiOT0 zWMOf!vaW6$Y$3?OViU)NgY)5BqFw8)7=H9y^Vur0v9YneHZ4U()R%9Ol-m z!Ux|nbe*pEg1Xmf&T7bk5Nz3a~#E%CxxvQg(+B`u6Qk6k3eu!2+B>qoSgU#AXTZ>yxmY{3+<| z&08p+WxCN98yFM>D!-q;mDNCBEXnR{HO2G&@-wIB1!W?&N=JM97o8^8u|j#mCBJ7M z==#+y`$~U#@!B0Azec6u>-k!nTwWZkZ;X+dSxCRJ;mV&c$b4Hj;D}T??x<70>EKY$)!Ia~wYA;pK+bh8K*NnwnV^z`)5@Nmz)dvwuQ>>R)YZ9{sP zTzhO~8V*wl_LZ%R2?Xtu^IS-Ykzx5=Y4o%ectJ-?jt69lql6Ox|RQd1{Y!Fz*=5g`wZTW zqCcf?aq^=~%F|_d$Lx)hh*?T}69Eklx^-AfVki7DOPzfM$G3#^s|yU#0@AMT)0yUD zuG)D*|8TZnEE@sWSzcbVNh|SjaT1b}?ce|ao*0eKBL>cJXjm8&<4P;Y0{iram&4Z% zq9h>B-0m%VbTnVGs3dIkilMhH}j%aqpIb@Eq!l656k|BcSrmX3H%$Hmu&DVqP?Bh zkX;u~OAqB-Resr{Gp|#_xVSibdq%x+Sa^7&&R{lBPY(p}b&+FU#CYB-^;hu z?Z+)V=LR;3}>kh@S`#dX4PZC~&f{WWpk>%YtRh&Fq@4z!X zJ&jJK7?>@*{CE6O;_CI$b9Hr1=RyMql0m~*4|Fqqru$dRq+C7j0h#{0P&r|e2Z36@E|+Mgq!1=zRhUm`&x)(B^*^p?NGfy4!A@rZ&+oQW(}*4-J%LsytYO8x~1dOLn1} z`GQG?YlW5ACk={_kc0%I%id)7$gx**GquTpFIhW$tStRvH2I3=D}CRkCiww4#Z(2O z@SKXZzueJ58Oq5ubce$C%yev#QX8)UL(8>a?SLR)Grus~{(5mR`_^jzK)Q*Uol=CI zkqU0WZJF@NMxzehB%FUamw|ZSy1p{8r9EWQbz%*1N3%6I-92nO>D&7ep5tcOTrrn= zCyI9vP{fy8%T25t--4#lID+IuN4?L?Lvuw`P39B+FLS)uf4=isrFCvahsojY6LpGi zl1FO|hl~625tvcxwbrmyRQ-4YzRDBsr0wtq2D{i|s>cmwwewRnK_SI0!6;!q?I9Xo zKGkz&N4h~@RPJ(S`sH)nU14Eio1dTDJuf|je*E}kIelSzf7Rl=Kdm%Tzk$>I9#X>3GAw+EV zMt!I@7D^$uMPe;i;f1dy&FoFwv7+U_rXlb4&QV>cq|9x8Yh-3q=wM7-*Wi7N3aaFr zA7W9OCE5*{o|iZqp}O}08(-t!$7EB2MbjCG+bEma-C3T7x1za?pXey9|I{jQ&VpTl z0D+rrVCwOrlw*&^5hr(y@9GlazuV#a&Q?6q6SY6J8HLL+EhG%$^GyNLzlN-`c$p!ria=M|1XE#(K3AO^56+^A}%X6Jn z=KCcYCB)aDd-({!Yh$`JOfHj`x{#_Uc z7KH?l(<^x;R0Lv;sOoI9SL7hF=J4cW^XPC;dWfxLMJb9OVXOpR9q2}=L3gnV9u)`* zr&sf6DU7Qx5#9t$XPbnB;`&5JMi;>5W%u(?4pB;JRPx^AZ@1rKq!K>Hf?&MNbp)p@F^sFfg4f*hT*2 z)H2=wsp&4CEn0$C%3g@_T>kkc%WD5w;NfUh!QbAeIf*D-d#5YR2sEazDJUpF6$R8` zDuGe~PHTR_Esmk?0`Am5B~ZB*uc5TIiGv}eGz$Gwzr_)mC?khHlO%S(zqZx~26{5G zAfPW0GPEruF|^(&>*$tg2Wh7+2zs6f@fshtrsd_Qm2W0p*yl-J8sdnPKGAWL1#`hf zgi5nq42I0~ewIF<>WJDT|Q z!qeyI>E(#k+5~C}T1CoF^z-!yc26s@X&e!Y?KtSp%AK541|3YtLc_8FbLvm^F?fNt z*OR4M^82Gzfu!!S<*&7)K5}+wOI=M)y-V=>HK%;-oZDE~6dW7oCp|1-s{6;VBkl-B zKfl^gJ?e|HeJllDS9L^0M84Cut9GL^N`_qT&Y0c#(>1KKBF&Z(f4itU(&e zfeJjm50nd!+~a)TljAj(96!bzu3bwPb7J_a>P%N!<7y__mKi7$*8$cC(OZtyZgwZg zkP|=OqGQB~(hNzYgI>W5x(y1_p8rU@bN*^^i*=$5@7S}#@fKC-}?>wlNA5I2t!X)U~nBp;@yA{P}(_>@UR)`y*k>^^lFJkt%OB=e|M=Qy$2=iZAMJ< zAXwZ_?cR@h2}%xPQ#gU8Dn}`9TITC(-S9Vet=*|CFp%b_yH9W*=zrOK<_ZO;VH=8&aR(pFvm^lI}JuA?Edod{$?u;`{w5NICnap^ij?S zkGaN|9FmeO$d8y#Za+Nes4U~r06S%(&}V|(4TEs1&#&+EMRp(&KoKK{MvWeHvsu{> z=aE+VQi)vlEA^z;h2P2|`>KUV&}NgRZGI6y5de;pf<-#?UMw$H~a^B{l5t# zl9H0y=3|C~>AT`-?1wc-bThH1QnA}9=s|6+%M6wi6`Cl2F{E!k*6!|@ zcKDomzd7^S`mJ6As*{e+G@zcB5z}LUdMdSmeaCLM)EA0E0;Klb>D3KOAj^43c?QR- zJY7f3{Vr9D6MCJmAFe7B9;Qc6zir(m>Y_2@h;N}=6$>ybeFM}^1l%j~%1S24zu8&P zYmTAVy55hM&I=29{6YE9z1BhfypJWjmBWr9g_#$foWMKPG1f$(;cHSlcjZ|6=bL4@ zQGW6-qZfNqR{rJ3?v^uU(%Zx3=J;}%6or$-Gfl1(N$bg-Q?7bcmbqaQtJ@l`rwlbZ z(+#X7mIz%vf5kd~1}1&|`U1cJE)ek)l7XI0G7pNlN5)!(4qUL`i(JaTl zVsNZMOYUqSzThm~$F7n{alqDLlR*paWy8NDYkyiXC8a`)SxDxnzE8{7Q)d&Yd zU;7#xV8Peg*R0Vp#vGywH-~k4Hy0Bz5)z@kK#&9rGhSnb3>K%3HfaT!fVZ#KX5QY- zO}ub|$#gI!L>Czj&T#Ix#pF+g0N^Z)H#nJT*4f$=*^3bTzBgvPHjIO@uc#?-TCbXW zlQ6O%4QD3zhd7R9@c9j(!Z)7Cor_Dd>wSTkz~8x9i;$O;2rX19qz5%U0+3ECeAQOd z*jHJ=W6G0^|M%Mh?M(oTo}M14Zz8#3*8@uQy3Hw#F8gO&X?BjQ9k0RGq=3~m)>M?@ z-bY?gp4a)F(A}?VyY-iTjEdnbE$`B3p~XA%oyeZd}C!WKIIdSm&c04yS6w$t|T zU@D6=*ih#*7dvBGab8FUv9AcRvBPq6bLU(g9f_Hl&$3C-y1LqJskhk2hg%+iKzHe8(<q6E#L3oOn04>yW2*G*nc3H@|w#F`yqx951!(E;O(xC@B24 zngJ4w>&n{NS(lBMd*2XGQqn}s>4+9XyaDTnnj+=7M7g2A+kXj!`>s8M5i+EQE)JHMluhwVwM8>rb!Kc+!b1@VP`?0pMzkVeG5p z!$VnMHdSG|Z&{Z`8AeAin=%(k3LNSBu#A z6d9-+Tz6Q@2vJ7IMfYSDO&sT>#KZujGUflfCELF%AuIw;TzgN-&E;x)q8?4*TvtvT zwMOajRN)=r7=C9!2YnA{WBXTn@7qm|NCLj}(9l<0!N1 z%7nqIHb34>3o-X^`hsJm6Sb&gp=+3H|GqaD<$lS^4=(|217RV$gV6;D2|QV-^7y8Va#P%Su4< zZ>pxV*^?5)kmi2x7&tY<{_F79QKEbGUh$60IALEgyS4hj{c#KhGczhh;S2ApS+$Gf z=iv0wO-F#1(U=&?9koWFB43#65(Gpg2BU1+P;|VwSs%Bo(P{lb(fJGck0yfvhQ};0 zSVJQt^E}RzfHA5KEU0`J31av@z0$rh)#DaH&)^1rp;Y46=ATPLdr&rYwu$M_&VvLL z7zNaoetm)!oAyKtNYqB^kLc()Qp^i&_%t+c*Cu>>hOsj(OW(Oc#=J%DSTgUn&gFbm z_CKxP?P+Cb@fe4?w=Fi1Czw?xvl)#n5xbW(3zmnhh)GDW`5$Dxy>Uk8V&w6>;svd5?5*(&j3!3*_wcjsiX$Up;^LoTIWze+Ngsd9dd;jGT zwXL?8cmvfX=4y&1<>XcmW+ya5NNFX4cSNI|uf#|{*onXV0Yr|)s;}U&WV6EzpcIU% zM;Vfi`1<6MAUeWE5S(reDr)Mu7=F)~sbaMkqxB9-t@pN>yfc}-&|8WgGhQoS1TUHO znqKZyXgqDu5~W{ZkgF!N%h|~XM%)T2LaeT?JSDh=FBo4RWdJkd>&b~1Cnx7g+p|y4 z55dmSeG>>rIpE8V>5pbYuwjfNV&+yAIaW%$CdM@>lj>XPniAHUc#vAvH)3AM+4b>* za7i=&=I7B+P~rT9|8frGDtL4U!EZwwSP1CLOb_RVSj|TzJP&UpAFLwqHjhcZbddo$ zX3hAA&f_?!1a+ild5jDz;zDSh5GY42T_R+d0o9zmN=QbA)u<=(Pj4ZL*VTb)GKb}) zNSqfxDW7|+kw|ah+4b0`n#C3mF6;cCvPqIG1?QcqS+J39F9A8VQ{I)15>f%}v~dtV zPc5@Shx*YHV)yWH^tW&NrNXE<(DxEqb8q`C$<`OCDK7*~_Sc6A&YWhF&V=@u?IZ`b zt>DIkDXB$|B6AzWTvQDrcJ|=sM>}G_r(0YI(8~Ud*zs=;s7-P*XR|?HPrVbvmq#<< zDc5CHjM4r49Whl#Dstb3igc&gzFB&o+MXYX5rs&(YDaSk0yFX&kYx+@vz`AuujueQelFY_1mO{Td z2P9f|g||ot>-ouhMpjl4#l^V$Gj`cB*z&tI5zpV|se=k+$b&YU%8CvP*uoX6D&oNr z$JequHUywkqtj-N@9EV-eA(kvbaTVG!-j*|yxXY^C}T=P%=6DQiHo8-AGXZp?x zoR^hj16iK7Xqc|aav5{)(a}Srq7aW;IAWQMdJJsno|A~3HqwY$QPNU+wmR#kT_p)T2^KYRJYI4BXF@AXA#t_ zmA22aiW89#bcPLy>hFq^TcQi0E!`a~(*2(Ydw*B>5cRktcenQluJ?6SbGq9Lh=!wQ zd*6nGnu1BI9+{hq2GT>CA>@ukV7tu+iI7%td=#;{w$LD;DT;rTp%>LLneA>1m5z8j z*lqokRa^=WMJ#K|A1dx^*kr&`+)R$YIqeN z`o*sDw_Y(feNfo_>tN)>*-9vJy!nO`OqcA1ttU2 zDKg|97qZQa;6hqDjrg*EYTiMd!8`ABnoa$HPT27&5 zlexLuA5s6k@%^E8Xnv~g|JZPiP_ER=@Pn1eYIXISuhL$QF zFgb}~37El_%Tt%7j?UKi`k8ksa!HOMKn~(|wI1q@*siRqT05*;Hs}bvX2L=Mk9hVT z`iOW9?|TlLwyb*Cr#(GclXKWpF%nUS(v>14B;;THS(=MxTI!_SQx^dXa1MPizz5V@ zA2=Q!Fh3V34en7?WiHH;R*8a!bjs`KLyd)6IlAqzD~?!#S66;F%l0I3j*=0CamNNr z#fmW>4t<85@l+<7fUjd~2zTl#6eagxtxXK6n*kGOf@TeSPkBUo@?U)Z(;h5`IAD&a zaP%!9vunY>dqJ9gxlX^d+R%N7gHo-cr1ndgjI^8_;P0&)_s83cKm1v0tY$zoV+{z! z{ZY;xK!Agfh>({uTF=(9yiX+0{5Sts?bt$*Suxe`j^Znt5X?E-mEe?KVR)Jh>~}5T zU?fx;i8~(7sa$yS1-wD*9+J((L75fyH3{l26|j%kPXQRS z+n(3I2JY03qUyw5034I8~w$NkoYb7L)hh-?20$cruTt@P5B-srTU>1o6iIT1Q zQLEcrpHjutBre>Yl8VL0l5P8KJ6=3eAbEPZs$%+4u=R}YdCkvPuFOA-^ zgLac^5b!Wg#-%T7ki<8X#uHI7oyaJI(bPJofF6Re$#qBpSR97hwA00rq>IM>_>hHO zK_U<7-}o@jYIfxcWqQc}-g>O=~k z*Bh~16!{Uyn_C}wu*fw?vY{b;4559FP4Y`b#M9agi(|vzlX0aL(t5dkuhIQ27-jZQ{4|;oyL}N!MiZOwPGk zK;V&F+1Fl3Gti5-KO_jIzR1T?IG06Gb!>L5LLm`@!Uf3s2;;ED72=8vlY#1;&rZ-R zYAE#v5~Oorp;8zKXp+qKv5vuMgB}wVIV7AL(;(qm|DbeJV>f+kp`jW|Yk?F3fkU@9 zo%)3sw;&n~?P62CnEDU6J;ZIU<8s(zoN4I472R*e8!%_R!>DrVocUibi69{}b2t(k zn&5nm%TWf>oyUOHgqjfxUl0$JEIDN$kKhRZOP#_?(@0Awh^9k28?&9|Vx*rM{g_?~bN0<<@^t74Y!r211)3 z@(K&jdvJySRd0VIb&Iu9`Eki>fSZhR zZ4XG(T+)tW%;$2?eT$^$oA51iBcppSe`?m}0MW*OAerIxFAYAY-&1-(zMV__vg@m9w5sf_F_al+yEcLI4fjE|Xp@B$}i6D%Ufc8jwF`X6BRg zG*O_YaXD_h2MM*HX(HIzO%RCUt;>zdy@>3va>cF2E0qw8xw^9;}xyhlUp1j0cFP8gw zv)sD(i(lkGsyf^AtkEep$5^uBMo24^x&OW#lST$AOH_Nn+iJTk57KUyc#)Sb{rm(u$^r=>T$1!oOE7Ss-4&<@}x>QazXzn1Vg zxX*M>@2&>A@WaJ~Q0s#uGmz}v-O)|Qa)MZG7vOT4{QSh}BwKrD*FEHOXqkO?{*Pz` zN$2%I1$iMuLqniZKz=CXioJUM8k}|r*d!q#gZxSHhdlUF=6N1D=sW=!I_Dk=Y4EYh zK#d`!Ofeg!*;P)DWqWI2WirZH_0S87+jdr4d_saq3_;7ky5n~s3^&D1OT?P5gf3dy z6^x9>`3GdT99&$wdwPNafo8OsW7XEy#^U$p1j%!mEdC6T&qKq)k^t#)i?Oxcc<#R0 z9235yNM0_sfG`0(V)}k>x@z%rM`v4Y_O#pqE~wRH`^Ruxc57Xst3BYJQ4p8WKJ!tEF#Nazih3rB7W7eqd}(0Wph6T#oNw zU+diL7MjF(D+)$eZuT<`6L49A%sk!LX^}M#&d*veR)%D|`1{2!uR-`51=*w52y&67 zd6o%f^7|2#+nga{;>6tw0`{rViCi^p=e0mC$mWl>aq*-yP@0VaH ztRwQmCyjyauzt2F4OID)vw`voa5s`iLdIV&R=SNcno#<0(-a^(&ZF~%%{?e4Z3&U+ zKuwSR*3|^raq)bkb}VmF7mg#IR0jmU#is1Nll!i)=Uju%9_PJD4XJ0iSV>THVvL!o zp!}cIruus~)oN^m_@A?I5PcFze+-&;5ZtSRQMvm(ko!GYVbh0*sJ;2l5}Dh@QrO!Y z4}ye*)cjC2dHG>Y12hsb&MVfG?WOP=sTM}kW|34+3${v0zH^`1_T>V35zU1Pui4S%fm_;6%k(=jAFtN zZ)ga*E>2trL7gk91|(L7S@q7JXSJiF@1cz`kE`{|AHpGWGFeafqTw4O6zqY0aB(Z$ ze%5?A64qZE2JHnYUgGa1PjluR%*@ObfRgCCS8O&_D$W32A(u8dwc1fRJbHNiLu>hjTVT0aefmXy>J=gXjf%X4{ zb|k)!J(RQoS+U0xIrm3tbx-UjB1jbdF=rJBQ}}JpRBO8k^5s@!4e+r;%u@jZeowqs z(@uXPW~c^goyLCG4Q`(fvJuH3%oqztL)w*Qo1CtkHiRpca;;?{?6&hNY$5 zira(vC`?2tra7w?>^g&8 zyieNQiHV8r?d`j|yG8fE1@>JtC$7){w;==xf6}$V(c>A%#<#*3Xe;A`!bRN+_~yDa z$Cx!xq6_GEj$#f>#n?P!)$~WH58rqi^hA`ENzo`;2Y`MHF`xU*1>)-YGO56NBMt6YfR2IhoF=;*p3(!6dBv&Eo zyCqnvMjtV>&t?!jG!A>Btekhpzx-b@cKC;&@Ywd)aqRmBS@2qbJY+smfY{#N4(aOY z(PX7G@PGy@6$eLr)#wj}O5u*cHo-3HVZ}_>=*W0=-X96%Ox1mHwreQ>nfHlQ>tTgv zT-2%;YfZ;n3biHajIgos#?4m&$JibSYp;q&q}W=-}%$A zd}oIe@X4M(U&@&czLww3uk7uS_v?h>?S=8(EVY1huqkzSeJl^A7p4UE05c+&N%=+x z6JFwSu*PQ_-1Ac3CcjYpyxFZYPm<~MK40ZkV3ta<%bB%wvEmSGB=t4Oqr)FJpR*en zxPiffLU4pZ7K}pNC$LXCGi61sOdIwO2MxLRrssjR({y~GHXju}l`9@JJU;I& z*5BU`X37#k^z-20AcgZ(2?SEY5)%h2C`4>yi(dKdj1mEtSP(ci;W^(@QY63x0Udr? zpAc;St~IZ_v`yQ19xalqhgHw-j4xiZj#eRaWjgibU=FZJg7Yz81m^yzX%;ozor!`z z#EFj+>HqfnR)OTfOxG8NijmKQMs}2cKo+9@-&Rz7@&Thk$2}l7j9ynSekm9Ia^Vzu z2k*X?p_-YQxu@X+Cf;Dc6iNbB{B>s5uoZJ;K}h)A?=r2p4*CYa9d2SkQ429+Q!-!q z2h{Mj0?G8?vjn*DS`c3a`9K7ahS7k3dh)6xbt5{fptxEgq_tE4Q3q6g7gJq7tqeLZJKM+YUa1siEeNks_a2Ug=0$2I&`3g(W$fswmvxQ{Iiyb#%X zLrYFf1_!ZfVY8d^j;|U9Sy$R^ZXmCN@Gmjr2`2E-Wl~P~mzG?9BY}M)3s&D(AqKH&*jrV|Jb}>RO zxcP*OuX_hnEOP{;3=N6(eArUEBZmA6r;2BwQAO}$l#`iA#MRUA>-EXW3RuP7<>f3e z^a&;)K1E>CF=B0CucuT0e^LzFOWFTejkuSyZd2z^w?^`WdvR6n6m!36^Z2GyDe;?~ zt`P#k+H|4b4V&Xg=Qa(4n9y_2i3v=yxgO4$h|?0omn?Y&0O4Rqx?yd2vRGSvJv1{5 zpm(oRnAh!cb&j!R6cZfL~SgUW!%K1Wjo-tzAvb~ z-&o6G2T9R!ct1z_vz*2fty?)(nz7FS`wMc#V;GJ5KDi``yScfkmT6M}*zlG^^(HW& zw{()VnK?cCQ=Aat3ey6HJ8#aEx&XAPrAMN^KJbW`?*P75yb}7B2qgi`oF!{P1!J?;TeTy_Ieblg{6o8OC4f+t8Ri zK2O~KOt}H5X#XnLu{rH4rVU_;9Dxns@hNz$D;(;q-nb|K_?Z?EjKbFjGXK>i_Azs1 zYI#U4ExP?R$v9P-V2nzs(6{^RYbVB384!TRhQ@@;bku-I_ns>fVQd@5OFqP&pAFtD@_*+62HO+f+O~34-DD?hmGxHShS(0*{w_+p51#|Ya(zacX0o=>$s%~h{UvVkf{qUZ zO**&BS0sE*$Mb?)1wp`VCW}}A9VE(>LiMKqzmC2&aP(!5_D*%f8~?^RwHYYNKoab! zdwyyLR94plnb0P`px_K7A;*f8aOE<%`%5$?YLFz~#SZkn_o1&fJw_a784RnGiG9KP zQ+0#txK9I^NLx@7DH{vsq~d?ruC?Pf;#Uo_k1M3m{QY=-1aP>-qAxbV0UHBaf(Shc zEPryN4@YLe_R+6H{!;26Bcg8Km&1?^o_4b&SW+1n>gO6)_WAJ4gkhpcCge0Vaahdc zOs)=6!Q3T)-jRb@cyr+N-M=Hs%mB+I1k|{hBP+`1-&Tokg;LmRvPPbmfVT}Is>?VA zdvI$};EDq?ea$w@i$8#Dq0uECB(#kFlf79%%LEL~KzA@9JQ6g#6uP!{Xbz4zev8&t zrp6Q?m#I#cD5txykbhe9V0Zyb6wk?S=-~rUrlMH6e3E0(DF&>l zz`jk|=>P%ulaKKf@}^*{eK3RPx^ll%uPqZ~LSs`>%!@vsq38B@=`|fE=h^5MV|BxG zfhKON8+c5BhL067L9n2CDXtX7*Wp&UDPqH!e8Nqx{3ROIiaOJ6&$ytgdau#(e{??R zB>w9D{s0JLa5?Xc`b{_aK61~L>-lwIcC8->ph-yt-WvB!>x<45WP;ktCAljJ-ppW2 z-aAN%iz-6->6=ZRzm~WuZ^kDkZV#$Q%$Ju!KtFQUiG2PUeRJ0i9qmo9*u* zcP%)jn;YJg+!Jfk2ChGrElfi$pD80I_C`1)TK{rSRxw=dUF;^UBO*V@rHJ?5clP|R z+&2uK9gEE}!4vhH1_8nCHU0i_oh~u^7%$n+v}*ivmtWx-MElRM)n{0RxD%<3cCZ}l zw#5Tt(}eypW=}Z5a)9)Mf(jjf1R}apt+fXDYJ#sZF)`EGn-;gX*ThYIO#dtL5jGZN ze*5O%)YLSGgop^{%A@JDYJUzuzsmq%H~5Bv9<(}q0)nhdnC%y!l`5G7K}RmX-no13|tzxLtn*eCa{+)tnU=_|>@L z$xE}rl=dHj0R(%O5h11XegVIoofOjI2;^9cbY%ec4+0bIa9m0*BugpT<;prnzK9%9 z-xeUcOUTN0kBy;td3k|GR00&Rq3<<+`uoE+jBgFlh18=(%i8DX=RqFZxb@*o8VuHo z0wQES`4a_%i8nwoQ=5qh4>x>zxP{PyuE4;)UNiZKDE{;*>KRSX3bDv4}Rc?Gch)1^n1c9 zki8EFBkLf1)Lo`?@8IB39~y&TqnU|kG!So;&X?&Hs!Kh)X}@hJR@ARk7{Jff{BtP! z4<9nwHSp%f${l08V;9P`8)UYJZEnv7B=hp~ne_jpf!O|zK)=pAqN)Jps)6?Z;u26x zSTPq1%w1Ywjm6`?_q=2V*8$59xjN#b1-Fyk(EPlgCY@h^t5JO%Ja_cDjkWBxqp3og6Uqk-&sF&Z zzk)U@ogbujQ+R>xf5q1OXn~~bc&v-n9iBRH;CtGQ4!(JUC`{*FtdRsVf5@CGZRpMo zh%MnGI-7siTn{BUw7hV^%+dlvh(L$hWaYd6C^A$Cq20c_EN(uNdJ7$M+b}+lbO+2E zT3{T}B~Q3MTN~n}AupFdOqZq>ic(Lv`~(6XEoXJ0v(g*Lk*h68xpwz| zf=U(Wo|B%)_3FQ=$il17iOgk_v{}`0rXcz6TZoC0W=d~sU}iD^wd#(Pm7y;i#E&*M zj6E;+23mc@Nuh7&HQ)gslWoP(?1y;u8j-*uO;`R%PDB?AI5tYEA$_wEaOq0F+bQ-m zBmiWj111QlPCqp!aQ!p+?mrjEXn=_J4;y`Z?l9!Xn(Wm80=C4)v;UvM&O4mz{*U)66)h@JRyWCx(6Duj zL?J38GqXWLBC}FSvUgTCSs{BRm1JdP@0}gl>pVaD{mwbpIoEa0KX=!ye7~Rhe!s@^ z`FLH8KciV}PkvHjm_6J8U=4N)O34Hzj_`mD|WCW8EHz{c{Q%;W2cyL z=LRr8wUP(y_r2e=NkT&$I&M1B^7i%D{lO4-UgBM?c>b+&G-L9RcATthp@8PvLU+g! zAq#d_SE3UmVA{r=|KL{~`b1>EKcRc5|6BGpLikHql}BBj!{&#zHk)-n(Ro3$F4E*~ zn_rE_Y_Q#<>z@qx>md}!tA*5;%-i@@gzQY--ktZnL+|k1x%#>K33b<((;nYKJ5+Vr zzm@)o9?hUW9`d6#Gwg+u;a+K_mh1^LhzTAm!~+O0v$%5fQ-i>8$1F}q9lkf(U2bxE z3yHe_Nt)X>d{-UrvK2g?Pn_5xJyxHj7f3(%t!tKSUvXJd+Pv<^$Hhe6Y?X(kN=iG% z;Y>xj%Cb)REqrr&Q;oZ;qvHYX8Hv(ZBBJk`|j1}!Bj$CU}1H)qYh zO?HY9eHo&N_Wl{=rzl*DW2#f1@At{&65`3IA+dq;k>yKseY>)>TCW_>*Z-f)B{K!j7;-4#5>7&GfP7ZPsYT3=|JHQgyBo65C)E_a$H7wL3@F8%Rz*+O{jdOYy|1H{PH)O55u{k^UY{6|ntr(@-pa1q|IBRSCY8ng2A1&QSoU39Dyy=Y zMsujz?8YJrE6U#Q5?3d8F-bWdKmVuo7P?q0aRww;TE2vUHM-L=+Wg-XBMTfQ`z;#3ZJf=NYBvKjdAXsHA(-cD`zBOd0t1xzF~@(=YxHqkQ8` zLU@X~Xc~bFS=8N%)=KPyU;Nn156j5N?0M0#7FtXxsRnh4+0FRW3-{+dmTYt2E#ZwA zKXFGNK^vY$O^;$HpSr(}`RXRY@a7kL@80ez6vt;FY$At(2@2d>knKng*wj1cyxkM+ zdZ=KEttr4t?#k!WD?Yyew`1N|V^1v?Kfe!!guM&UP}_c>euB8_(y#iwDaS2L&wfBU zvAgXcj_d(H$$WNGVf|-;_W2Y%C#lK(YAw()wbzS(;BDY<*R#PR{Ug(u1~L^FXXj%P zj>|%3#X_V-+B-r)=g1tOX^`|$=2y$vK$WB;caH*;-G3WKbej#MuE%H?4ZWhP2i}5E zK!O00Zd}WRrasEI$fmsi^qer5?QLga3Ayw!U!=V*?%urE*Q|;8d)b{`DcL%lee&IM zot6(Qx(gp2d-5d5W-Nt|q;S=OM%ZdJG+Lip@?|LGP}`-x#=W0V^J<%HjiNu()G83a zT-gb5;N$a1!e}5HlqWwyuvLOlHDYD;NSxPlN zjuxH$cJd7gyT%8Sxg_d}fHlT~P1pQ?nnrNVJIse_iYDRluPT@-t}e2>iE`;|1)Fc8 zTHPgL5oA{MB8~WRuiJh2$5H``fJwVXG9%e+p=B!3)J(1Ur5$&U<3gkTHfR4czOAC@ z@1KGjVGCYQ@^fp;dwh}ZEC@y@icV= zVQXSM%|5B-r_@EH(V2_I5#<|f)*>PzFt}fRv6s%(|9TC*A>ef#$i}IKEya8QyFjN@ z*f!^by;4X>`V)Jnd1L;|L+Ydd%Op?Ty)e5x60PX^`gKoITAXsRJuItS&nWX`YYJyM zY53nm|Q?C(;?HPS+2-Qpg*)D!?r*)y1pZg0L%bUr9?x?Va( z;afocKGw8InYvvSh9#_i2mKH1eeB-#!aHk+QvSJ`APO;YaSvHv@u6m&z^zx}6|`1Y zi_0sMd>~I^l3EAVWV>{Ee8$|!+HEy7@L^jwoFr5HuerO*vQp!}J)MlSuWJq5I_n}O z>-kj9r5{(OoIWQg>8EE0qyO^GUNv?Ii9AE0Kt_t2g3`R<=Ybju_MbS!*3ca!A@ethyb zOWHLur>@h;3N3n6W@Rp;-6gI08-v=79GU`^3vdHBKP1r}sPTi!O@^xmxWO$Dl&bFL=mC!Bn9 z#v~+mEHIHIwWYHeU7`OXJ-!;EwE--jol-b3Gef=6zfM~+KneOwa(VDp|1os+bjV|w>Q{PAvgH=_z5<0 z1}-ixXu)frksmd&ssLT@GYgKF%()7CoHXY^j>KSWQCnNkEE(4;LXWlOi=-JK6 zfUwJUN$_z>KQQM5g+;AY>T{@Z|m@N0z=>HPqLufl>ra0R7GyPJ=hJYNt6i4NUy}M(+7`~&b)IAV%EQ9SG#(4L@jRBIvXBQ-4JZyM*UfBlI-E9)#E)21GFpV< z?*3JI)rv^*`j3a}uKfPzbe;$l5$XR1)3j;JUDyTHYXYk^B(JbZYl!C2xnGD=SbGIU z@6A)RVEJuzbljbwn)XC`0lTrJh;qdwU!PBX%Y%{Zw#l}}gLdu15=Z6E5@w~?4v*bi z&$)Pj6{Oq{-L|`Y`4r9*>RCc`4F$MX!M=+YG+^2)*iL(<-`qF>i9>BH4S)seDSfp? zC!#C9c@8@b%2ioesUz@8Q2j8!yL?FyUQwtZ;UOh3Wysdbv0s6H8TnN+CLPC9S3`dzm->xBfh{E(*#tR_q@WIRM5G1e8y( z%tO2o2QCwJdpuUgQ0W&VM{mY%SiNSQEph?)>`Kxnr-{Qj)o`7%eFQePJI6m zqZJ&WzJJJxsGXU%HGeSc_SsESv;W{h7+a?sy4i71wcex~ysA(21g%a>!qalSH;=0+ zrJ(rvTB*y=hqDLVZDKf0{e_MzW9EokGsmdZrb>A*#!0b)k3g=+y0c z?#hK@IFd-_u;#vEnb6FA&3(&tR2~(G5Dq2N z!=g5a*VlIv)vjmw-Hl7NC)4lLt`&6r)fz-NoXLr{R?Jxaej{sQ@WzwUpW0<|5mvKb zjSUQNDTP1_+M$cv19(~>_8+5GaRQxp&!=Z29myH9UbSzUU~BudMYJ*R9ZTU=`bwVl z*z8K-KqD#lv+vV0Dl!ClwF^HI4>cGucBi6iiTGf{iRI%qbRzFdO_*`^{>+3(EPn! zc8vXBqyFCi-KcNXbUE*hw|6O8T-Z)_;py|^R)6*K<$bw8$MjKK?8AY$^S^st4#Mv9 zi;72+^l?JMMIb)6p;qWWKOvHCSoswPf%qJh-863X89Kb0k>__R<3Qy6zjs_t4ETqf zvBhSZS~d})sfStdh7$*B|ExpGvZ)4xx))>?0Si-66g#FI2*@}s=+1_y1xp-n1NpRCIcPleE(6q6uu>fN!}30>qc zLh23UP8iWM^uA2lJi0zK`je}0E`V3sjoPp_T#J-6lt3_o_u~yt!;p6+h_N9d=-aOJs7Ie3Dy)iNvOaEBTx+r4(^uG0m*#vyo z%^TdAScyNs!S@zs$) zjU5h?%VB;45*ZJANhLoS9y>2c(`98{?}%`4>{nII8D4y*&Ze`_leWFwVzpi@t24)N z$gNE({&SBt+HALn^7YXZU%kFARu39ZEb~>l)Rzskn-_4fI^S}8g_&58Ds7A z_U8&bNK5+(uZw1hqd%~k_t#g|Ya%4v@DVWV5;DT|b&-wC@qe*cM*={MR+3a&b3>k` zs(f0>wh=!kPl18^gbqC;tUqv)&uK!4B>9hUF>ROECmuzoNipCOTH_si6eM3R6$x{7 zy<1|v-pM6q%?-0Dp8n~%<_MrDCF`QTz-3g-TK1cfaeK!2VYhPMev?Ok!=w4>X`5oZ z{24O4APJCE-7t(xyA z4D>S`p6+w~6i+t0VNd>c!vGg>$*6UlZ3mbbgdl-~tcx?P#u4u-H}{wK&+PqsRAHWf zQK!k(l^G;3$0@NCJO{5t`f$ybtj=6DG23e+tCubFLoTxtNtL*`yAaea&v!lQ>2DZa z5`Ro^R%MrrUalM6qIcnamN-$I60X^3B-6tPo z?xvb5x8w2oec`8S-B@_!1&3*d8y{{->FI5Swgw1db!a8$^7_E)_ZGTQ&AOA=5kW`) zl&^ZZ_zcoD;*eTcr>r;=g1hhc+MVQy<+mTg!4BMr@kseHFm*#Sb-AI`+fkAzcw$N zK<9l6(Jh+EQi%=G4{_aBpGlm_y9cx7#i@TkeX^7--st^2WXoZ+0i~_IB|6z?vDrHf z8N)g;+}}CTF`Xi|ht0X)smZ|$=Cb*AUSO$ZSXe&y4-I{NMn0u)8WvBHgq27xr2YhB zdqXx=iji-n>&yezHj9BXzh#HIK7`=P>2S_PSiz|uWW6k-{5p|@>J*i($pnOD;*X5- zkaG|%oXO51kqHsR7$0tDvTH7WiE}%IP-OMP8~J^@-QrIB=vywK+2gok$^nq^g>0ww zorA5d15tg-LAlc>e;pZOTDTdORG(m5=h8%U<%V3BpbpFVb!*~duCSR>N8*T6_QG3` zxoJ;L55YLDsin1qzyh3@m0-rm%0z>Q3FA!Vtuy~gl9)0p)+*l=ooyV@j=Q*p_8QmB zk1@W)bf6S2%1{1fA3VaI4_bIHQ9CfKWQ|kY@L-mHB|0aDpi=7LcVBYxlidGNqnxOW zc6w_a<`wWQ^zj#|7^y9Ui!SAjGkgDPHM29GWTw$Q{S({?R*FecDqibkB!0Ra&GkmJSDZN_xB-%VKX0su^rcGR4md9v%|0 zz6Y)6RNb!!*AlLH$6j)(ov%elukXwClpA3P-hH56^I|$iGCjvo3cJtkwm)Au$56KT zH_VLmt>tj4P_~I__{C{H@b7Fl05>1JCilAN>J?zpIT;zB%c59OR_3f-q&L;l0=q~;F)leH1A<@jXDAG$lE^_c)_R5EM zAM4&9OwP?$&{9Eb)cIY}3;!-;I0ST_p(#yg+qq05@lC{yKWErP`fgMx0$aluK0a}7 z?6OyZzn6B~jzaQ}h=qyx^YmR?0jG77T~zEBahDMp`*Ep;10x>@B)(5$BUP7;J{cb$ zAH)xdidVmsmq#^gh*?E|CIQ|Puql0#%F6W;nT@|dVwKovRiU4s_YBV#it9dSyAE?(f5X8t4o_;o zJjmve?znabP7!CM`3N_+|06EWNJFD)R5gvd{un4g<*Tc+NAYm|0PQR^EDSGKFTB}` zowd<3vaSbWPaZZQY^^9rlkV(hVrCYtl;|%R93^#H2NcsRXn5`C7T%u?=>ur!ZZxKI z^kN|Gb3%qd-fSRmHw}%zj`>h zbUSK%57cp#!?TeG34=z?y*LwfMU~{mFk}~icSKrGITD4tmGQgaEuP0JLAJZ#5?oU# zA=2U#`>-4|CeGHM&uoAx#rN4WQmDrJ#x=d(zb_zO2++Ls##??9gi+`{D5GhIzj_ zax<^0_2Cml6GN&vFzLXmkG%2w)cZbrI>kI3 zN^*$#>aN=>|)D_ zKTWpzd!{P8Qe22Q5AE!m-i-DJ1}qqoaHk#WyKmpVT~3+Z$eb3HyQ68*w3BPS|4}9L zE8}DTGtLREWEQi5i~+XJpARJ^w@yw@))smYrx)%ieV{NWhZrh?I?R^)GXt{r_V$<@ zpybl)U~eyIHA+PSoPV6f;CIFGBI4VbXSvUWbO5{B0^>+(YAU=4NQe*P+PrXc+g;{D z!3Ll+E_7dS@4vdb{~WvU+GbSi55DTzw@}TqTFIN=%Ab+J9{SQPK%{O${FASFTW^Jpl9^Ex5won1$4kZ*|u_syggbSsAzr zQ@Ia*9l#s8TswxeELgy}59u{#0brWdr}p-h3cBkSL5K!oLA7( zP^1u=e|G2^uTe=Wfy@{4XRMjGy}zoO-|{uQB(d;#5W|G+W|r+Ye#L&2e1yg0iYPe#_A2Wc6qWRWF$86x#5KoF{(v8k0z_8o~#K;YHtp1@;Fy z(#Sd}uM5Q=KTPxBgkZ*LV2tkUy=!E0 zs7SyxyoR}C#IXt^Q2IO-hlw9;`qt*=rMf5Y&yD_MyC+b7e32&cDI5Q=x`z%_*RT-Y zL<^{uN~_B&SRal;Zrp^}_4}`k1B3hN4;+xhV~CM1F<|8FVGyQ@t_$jn9J&!K^c6gk zK5VnZ{E|nrLzungIlfPz7P?W}1eMpmzP`4~bAm#Xw1wA7GYhL2OVFHy61WNgK@P3_ z2M=Eajo~42dGP|C)^c2^M#D=1+>xkE0I+#XL%+J3e^v{K$S4d-BeAiy_3`pL2iMQ> zO953)YL-d3QWzQrGqoKGBE}H4V5S+0K;sbp91n@EE`TSAkv5U&=~P!wxK34=i`gVT zHGYJn-jF!f14(Uyc9AgloS*yyn!~rtb1@P2;bL- zi94|uy_soc!V5$v1k20IiP32&=_=qYg+D1C7wLveSd!?(;iD3cp*oIB&);27vKX#m zz#kV%9M?YV6Hyhys2g~{4B;EY^X^Gd5AOWr=Ul8(s&)BI{q;`k7!5<_|L0HSIwzP_mcXL(#A zqAOj0Cj#fD6<>wbKUJ!G1`UeM}r=^)?Pil~*0Kme$`)9Jc z@ABLj_yjSme(P&1@R*6qhZhkah4E59xJ##6(w*^I`9PlU?k=A_Rf4+*`Ik>mPmf7= z4&@C%;ht@WaF~*T!5x)JOj|O>Wk5L_4-qA+Lg+VC)Ph7G@x!H7rhD=|)H;ZlzT+9p zeKgC8g_4w%lsF@j;^k2;{J0XY@9m%O4#7{Zzx?NyNk^_X=9>}Ye*n@C0<{#mL@$5{ zB$(Ot22T&PDG^t&d4XzxzOe3ePbmqs6aj!ND8auf62i3tIY#KcQ%U)bzDqS0*rm~=G#m8& zFlT8Hy=!ci>&ErI@87?VcVuLmd7T_OsX!LVj*)(HII=?$5)y-WL{$+dr=c&&F8}6~ z5(Y{Av~pZuGM?MOx4dJ2^@y_pg-i}0K{E1-pr~iZXtD2ziIXrx>>S3)$tozQVAJwL z){iAO77AK9P0dg&*7YdzALrzhBSy~kI8*pxb4qG&D|RSY-bgE)c?I-E8cu+N_|o`o zVM3O1eZ#}Vlslk+wbe@;i|`Q#sHkFqK$w$=SilKhr{>j8#y`NEMo!$fs{m(U_4mU) z7N0p>gej5VF*=d>$w4*+V&)u5*-8>+iV|xv5NM{2I^sRXN|yj{Bq!#mUBgry;(r0t zk?b_W04{$lh{Ugfiz0{VX2b_!MwM8@@P6hrK;S8FrPU?jqQIImUE1wBAJb$g0cdg*>$#5 zOzwfv#1(lDWHq=2$d7@9NfJ)dLTW;`cklIuUqeF|utvU$>VZN5^QJ}|pP~y%%w{5i zEOPs9>H#=R;S!!rFCT`Z`zDMY;19dQAFYf5iuCmK$ZHt!P+~CNdZchW)>;&t`?1Ku zCt-VHqkG}BALsR5k0GbE#OAdzepVnt=4GKIR*BZP*aZj^^^dvTci6&b=^Ujb`U5Cok}0Jh^FVx_&!BGmqb$_H-1+Se5s zBoNZ$o+)RWw6ZaJ!jnvRaxvQjt3@21O#7J%HuTEM8;DQ#XX3(zR0@kkqajLT+K9+2N0Fx z#s_wisgezGysZS|CMf7ISl??AwW9Wm_luW?=!oyclK~|mM@>x)%7$jJ@gjY-N4);T z-`{^Z92eUhA`}v0NEo6+#8rg*1IW+E&=eh{qpQTmqh4ft8h&?raeIh~DT%`_b`T)} z5hhtJ{VDEuQuzjUca$hz5MxMj^&d`k$)kkRoT^pXxcO>S7(JuJ-(O%*ju5ftf%Z2F zdlaqAoXL(ohd*M2FbKDX9993SBpKm*MulQ&ePs$=uA#v|ohIQJ^epJJ}I&zOphZl_Lf9eT0tsRQupcq0B#ox8UO$Q literal 0 HcmV?d00001