From dd0d8b1dcaa02a3c0117be1127b6e5808d982f34 Mon Sep 17 00:00:00 2001 From: mhjensen Date: Wed, 9 Oct 2019 09:51:41 +0200 Subject: [PATCH] small update to project 2 --- .../2019/Project2/html/Project2-bs.html | 13 ++++++++++--- doc/Projects/2019/Project2/html/Project2.html | 13 ++++++++++--- .../Project2/ipynb/ipynb-Project2-src.tar.gz | Bin 212 -> 192 bytes doc/Projects/2019/Project2/pdf/Project2.p.tex | 14 +++++++++++--- doc/Projects/2019/Project2/pdf/Project2.pdf | Bin 236296 -> 236493 bytes doc/Projects/2019/Project2/pdf/Project2.tex | 14 +++++++++++--- .../Projects/2019/Project2/Project2.do.txt | 12 ++++++++++-- 7 files changed, 52 insertions(+), 14 deletions(-) diff --git a/doc/Projects/2019/Project2/html/Project2-bs.html b/doc/Projects/2019/Project2/html/Project2-bs.html index 6106647d0..996a9801f 100644 --- a/doc/Projects/2019/Project2/html/Project2-bs.html +++ b/doc/Projects/2019/Project2/html/Project2-bs.html @@ -153,7 +153,7 @@ MathJax.Hub.Config({
Department of Physics, University of Oslo, Norway

-

Oct 2, 2019

+

Oct 9, 2019


@@ -265,7 +265,7 @@ Discuss again your choice of cost function.

Train your network and compare the results with those from your Logistic Regression code. -You can test your results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras. +You should test your results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras.

Comment your results and give a critical discussion of the results @@ -290,10 +290,17 @@ project 1. Compare you results from the neural network regression analysis (with a discussion of learning rates and regularization parameters) with those you obtained in project 1. Alternatively, if you opt for other data sets, you would need to run your standard ordinary least squares, Ridge and Lasso calculations using your codes from project 1. +

+Again, we strongly recommend that you compare your own neural Network code and results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras. +

Part e) Critical evaluation of the various algorithms

-After all these glorious calculations, you should now summarize the various algorithms and come with a critical evaluation of their pros and cons. Which algorithm works best for the regression case and which is best for the classification case. These codes can also be part of your final project 3, but now applied to other data sets. +After all these glorious calculations, you should now summarize the +various algorithms and come with a critical evaluation of their pros +and cons. Which algorithm works best for the regression case and which +is best for the classification case. These codes can also be part of +your final project 3, but now applied to other data sets.

Background literature

diff --git a/doc/Projects/2019/Project2/html/Project2.html b/doc/Projects/2019/Project2/html/Project2.html index 3a9330c2b..d83947f4f 100644 --- a/doc/Projects/2019/Project2/html/Project2.html +++ b/doc/Projects/2019/Project2/html/Project2.html @@ -112,7 +112,7 @@ MathJax.Hub.Config({
Department of Physics, University of Oslo, Norway

-

Oct 2, 2019

+

Oct 9, 2019


Classification and Regression, from linear and logistic regression to neural networks

@@ -222,7 +222,7 @@ Discuss again your choice of cost function.

Train your network and compare the results with those from your Logistic Regression code. -You can test your results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras. +You should test your results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras.

Comment your results and give a critical discussion of the results @@ -247,10 +247,17 @@ project 1. Compare you results from the neural network regression analysis (with a discussion of learning rates and regularization parameters) with those you obtained in project 1. Alternatively, if you opt for other data sets, you would need to run your standard ordinary least squares, Ridge and Lasso calculations using your codes from project 1. +

+Again, we strongly recommend that you compare your own neural Network code and results against a similar code using Scikit-Learn (see the examples in the above lecture notes) or tensorflow/keras. +

Part e) Critical evaluation of the various algorithms

-After all these glorious calculations, you should now summarize the various algorithms and come with a critical evaluation of their pros and cons. Which algorithm works best for the regression case and which is best for the classification case. These codes can also be part of your final project 3, but now applied to other data sets. +After all these glorious calculations, you should now summarize the +various algorithms and come with a critical evaluation of their pros +and cons. Which algorithm works best for the regression case and which +is best for the classification case. These codes can also be part of +your final project 3, but now applied to other data sets.

Background literature

diff --git a/doc/Projects/2019/Project2/ipynb/ipynb-Project2-src.tar.gz b/doc/Projects/2019/Project2/ipynb/ipynb-Project2-src.tar.gz index e1b5fa8493b874032aca3eda89e703cc7d0fd1c2..837da8f9490ec409b9f18384f1e24bc99aa68844 100644 GIT binary patch literal 192 zcmV;x06+g9iwFRFk)2%t1MSbv3c@f92k@Qu6nTP?+HE}x?%+Wf;tO=HbDi4GwgdO} z?gR9sco`z}cli?%LbA`+n_U*Uy9*YBh-n@FH7m9R{RpSMxPA7+9>?>JD%rx-j^PheeVze2mk=A$yo~k literal 212 zcmb2|=3tQGoEpo({Pz6$EG9#N63MvSjmXYG$1mYJdM? zzZJZ*~Aj z$Tp;zcBDi~Q0^2^SQ;QsTd#g=6gA+bp5#*;bT!(Syu({g>4@Fo0-`OLtl^_y?3%|^ zA?YHy#r6G?uete0QcD&F*l=dg?4NA-Ov*W#G{0}UkC4;+=Xwt(iL|?5*gB0S08EwH zFHs%NV1n0D1o&ivUj(xzubHJ{XHcay_}cm_lP}6uSCc|dtbqq)t^u~Oe%N$ajQa` zhhyoQRIIo~@Ca_vHpx7}0eg39TeE+xTDIq*@xjUt?n+nN36AJnDy|L(RirxH^Yt`@ zHS*E$jh|EL*$@2iVl08+5myZ+=xDD^lgF@zo<(2^#(CE7xHs>LW2V*BnrH!vPUoZu zAQmnT5aT`Uh|NMMOJ7sIeq(on&(4*DS zkw$x=P5_}*3t5x?6&;JupW_S>ZNtz#(=n-IBD}>9Eet@`3$?5;v*)u`Uf9~B2*Syq zE7bKr-(EL2Hx%6NyUjE|PCQX(iPoq)HX4>Wd~YL|sV$*dwG%nw>N94l9L#!A3|L>C zZ){9xIFv{YZ$p0jym*uNW3}zmqZ66%9~dK#pi^QL@NDGoM}MJ`6t&1IX)K{?aM1i- z-R0nBCIV1Rq;pwv#8C%frL4n4#R723hv1VE0&IkcY$SFZz21KzniO3a9}2gSvkihM z7RzWJVtZ%OFlHoYVSpqN>G|SyZ#+D@wjgoPA>csPy|`trYDk65DvD)54|v%g-Oa3y zYtdZ&yv@4z!n20qu)7MHwHxdV6S;}}5iM?kvI$VaPT@+^D!eBs(B9c*#yl>zoJ|v` z-&}a`YMVNOh{PY0(|~*|^O0TDD@PSCPz|f-f*gWfgf4wdNVu>yv)0qgl}Ru=B}jJf zcLX+|Oa57oSI{GUIz+s(zCrMyNdhmt@^VDCwt_rJ)>t>CQhqNXvYI>-=v?cI-8p8Fhu{4U=bR$)!pG0kGGI+h-jH$C$jAcu z71su-b2Wi>kw;eY*+)<)`_9{$%BcBK&yyB+Ppe$0+YH`_2Z7#CHZ z7?SSf4K?lyK0CP-!Peq*e6E{uHK}IPC4~Z6X@~&La{s`2U#73IWRp~-HmO?hU=d*jzj&2LlNQqy?Q-y zAIFg!-5cb-h}|ovRu){g1CF3J8io5kV1Iha8xSUC;!cU-LV^8Kh?{|S8!$8e;YmVg z4T|xLy~tmBUg5>CBqbqWg$Y@Re&j8Kb+fOW9hr&n;bG|DZuXgH+DFRd<7q6NM{jJh zCjdJ#%g!r-DWoH!v$2;oj_7us=SX9IfG5z*?iQvlY_J9lF?`7HXRrs9n2^?T$;J9o zApn<1NQIpN>LayxGI+ks@!d<+o*{hGvlF1PQzTktEDND!NNz>R9X&q6U~f@oCJhqS z`9YWV<|*>s(x*}W(=Mu#c85?@3!3~93lzczei%G(j2zlVR8cbT30AzM&y+EID#WZ{ zL7$e4LrvliVFhikEFy5mRFA~*0IO>{7SMIfD6C`7j!M9b7%)>xk}H2=Keu;}JdJo_ zv2Ir04nbL#c;It*(SVd5%8JmML3V%Wz-|9GHJ59Nm*sGfdv2DNoFu=q95$Qq%2%J> ze&c0;=bDQ#je%RfSfCB%4JkB$LnJt`Oaar<+Z+}wf}U^BMz!1;eAeI`K<3sp2;9-I zaE;pE!y}o>RX@7F(m)+w6=%c~j|BE7m>N&ssUSer5s#ptZK3flFVbh*ekyZ*OG6<+ z|9#@B|5~ec;Rl+x3>dLFIRq}>3H?c(`Ft<-zt!UmIBxO`p|;_z`W<`hIzztVie${X zT00j!=3Cd=t$mm^$I^gNZ}{=VWL5-^7utqmy#fh8nPE zRM@VpK~66W!vD$X(e0t%Rk!R@m7JOjjRYw(tz06i%18A5?pBQ7E!kW*854=`@8>V# z`g41H2Kkl(==1k_`4=L~*)g8Ny}?hwDJyMbqUg9RX~Ph^Nx_NP*p?mfaI$vhjc=7y z9bfK{BPrImV^dpHv;mlr#}ic-dG@T|#4arKwPnU{{$M&MVBm3!*AtW5v%9sPaX!pM zv)s?>IeMrL_Y8#`A*Yuv#)^79q%FGN{58a}y}Y)f875||)c5r=q}0DUte<6`J*>{<1-MJ`Z! zEuD?ALgE?SmG;k(T@;(UgcB_1)pB2`bu77Ql^Ov*zZDv`bwrsCV=;!LGIG6_(-F`Z zn*G_m1UF5oCTH*GCb2f(2+z0FsEma;Q~$1@yTJzqRSf7$-Mk$7Rl@dB!Byau_n?|v zLu^dL$g79=QSmXWEG19xqe+lEc!n(COsobK4RH8xeNnYKv>Wr!-R9pN@irA;DNb*~t`(R7c&r>`v3u>a*qhVeyY-@mY7nSYiT`icw zf9x*u#{xzJ1Xje*!qHpMRY#>S&6QYKkBYkq@`Itx1uR7Y<+Tvn;HBtIE)}6ku(rR$ zV&~}*uzYs21-O>qHVpXY5}csr6gQF;6z>!!jBAz`$`4^r4H|m=uFvd?LT7Z|!=Tl` z_KE^ToYe&o6SZUmg~U-?`8>OWH`DJOuM}ebdH%`A)3>*+=e$G&lx(2}0;IHX%W`zyF#O^Q66%$460KydXI|Z?Ef>95 ztI1CNLY)G4>JL%-F-|hlTGnPR-|tzD<2DgFME|!Fejoa7E1+n%)GlG)M_U;RCHln6 z9sn&UXNyzc4c#ThRLxNE+zvfCCEj%icpCfE@96ZLG4kmRVc7W=y!U`R*xmOS%{p@e zXQ*?8=Ec`b8hz9MJ~NfoT4Up5?{{iMMz#J!c5De9VeTy&{?F457KJ5<9{x=7`!8bn zKl}3?lX&BBoFSqP-93AplVzgr>H*?FZ-B|sD@#Xvp#xHWJ#OM>l8u7j*V+8#i1Dg1 zgEz*ehJjj4u=4`J&RNOguN?WbX!<{~fd-OTMeHr3GzvX!#ZbPC#^8t%aa^;OCfqpn zLUD2FoxW0nwT`r>cdH-|nEwr;=rjIClo*?Q+s^L&%79Or;xyYY-J~%V7Wez~E&$2; zB~iW`p92fg-_>A#tRZxj(^(ZmlWGApBj*?d+z_Fm)M+kW@_D3x?Ag({Nr6=qR3v+B z!GF#U>JF(GS<~(=OB@`E`&!2WsPv@v#6TGhg*Xc2!`MIV5IYuzA3qmEJ6?jrAk@Jb zgN7dEon(iILr%5lP};INAb{tY0E~&OpeSk{U^f|J*g2szb!g>@*UhCH5qQz(&H<8~ zr5v<+_G!my9NH3vv3=PM!eesM230Erf2tHebR^&8MpOi}aw8FJ=u?~l#1xw&da_aI ziFRikD^VFs+_#w(R(-L(2$EM<5FS+~@dW-rARn{vS)5rSQK1!nHuBAZ2;be$-TmYwuvFd|WI09BHd z9h|pBiiwzw4vR1Klsr!i%y^YRh8&{=GOu_Pl`DLTnfafc7VW)(Tm2= zz)DfF=-cA=3tR^Iq5WeAuu04VO87G*L1M3+9cn@swo|NU2T6-Jq3}XJQa2t)Umema zW(-1VQpUA`aAtBsn5SQQf}*ND=C^MR_Ko$%1AX>QM4dxC&ajcFAHDwt-moWKxI_nb z)&Ic6k1YSyhiOpmb$rEFex_`?wsx{qMb<(ytQoS@lEQwuQyxcv7gMZPEZ%`W+}YlJ zG9Oc{Kwep~wgz1Du%NAytPBg@nw#1@tLZSW) z-uMwzc0rRW`R-v;(t?($@(nnYV7D$@JQ^T)u}6+=O<z6}(cGJ30X3cAZr{&+KZ;IZ3m~ri)fLHq6EroX6z*gWSm?Sl ze)a!G#-n;4t1Nm?5GqLcq)C+gbp^4d@ z$&X^Lnzn({D2g6$*uFTsOAM7-*}wQPO*qwPo%#=Kt|ppDvU}T$HhRIjCX?s)9Zm%a zn@&c6d{78v~ZbjLYhRFB-cV(`W84c`0nW7%FeGE!-^K}^1bPK6~9q=JN9&lem? zq6M}6XK=UeCx7-f-OLIXG05;=SLmsgL`9tFa8&jIUVSIFq(B3)5=be$Jj`_o&KpHQfHb`V7oSIanE=Txb^HAepPr>XRIqk_(ddh>&*_+$vb4WZrpWI1TU zwOY)nC4?^cKJJ0YvbooyT-UyzYM37I@N_A_pk+-Qn^3P^qUB7ypD1Vvc$`^FLjqfK zIt$y60B2b9%Nx7I19>&3weG?*2}}ZMr4ng0tPg6>zoA&iE`(ah68_l#T;8ZtiF;w` z#c<}DwENQPUKA>IC~Xf;xRoRrSN>Z1QGkmSM%~^K6L+H&IqnGF_6nadWVzavVkeCN zR+oiCj$N(F7Y1S-2p?-s{{1BoqJAPjbcFJw|3ZXeJB}A6c!drT(<6w$_fj<+phDrL zqgxk4llzaQEOiRN@6z#6TXISN3hmdPl6qvFu1Kx_7{;rfOXMBbu6vgC5~yfb22 z*^UQ~Wka7P|6YnO1={OsC*x(LHol4VW|p4!5`Z4x$R!0-tV-GQDrVJwD1D%lV`4~u z$QHT%-v}=}bO!&r(~=n8{5)L@emj2gO-vqV3UE(A~cYOnYY_dEi(uf zL<>hToS-GKEkLo+VeY1F-LpYEFPGDd6cBxQb&>5^YV((QW|}H)KZo8#Gr&5(oh&Q0 zG#zVT`Oeo^d#lp*S_J|^Z&ZM3I_}Sa@+>^f;Kbw1J1X{A`Rp^l)qTFlagYTT)MqYs zLkc`7+*+A5FbHE5r;kk2%hn7=`|RPU9se|(s+F)O3u5*ZYe+4D}Fb9(svLi-PfZej^U<&|Omf0JyQeK8T^<+Ty9tp6x` z)c@4GpJexd4{wnXqSZ0Gpk*Xd72eKIiw}dq z0=qG{2!hm1#6p6X%7=L{n|r+XBd$s1&y|~}(XEsT?%N`2g4%u_9+2-f?!Xq&5^OX8 zTt|QVK+uwK6f!Q_L006vLPZF#-ZxRe_uldvk|I$NsWlA-b%P1s%Oz5zD{8J?sTzBo z9Pufa%7DjOKzidUS0}TcKOKCcsDHSz1q9KH7E53sIh$Y^Z}u4<=k`i6E(KaHY9GJ- z4(k>b)PmXnp*sC%RM^@;R#f));y8#nu$ZRg!1Z&I6fYzQEQkzcj+HmD4A)bvY(U@H zQ=EW{1tCZx8265ecEB5+rj7@yA+?N*w^1WHC~P?SQIcL}fJQ4sX@j+S-zt#m8bSL$ zlx$J$Zy1aHe&V~O){f!|E9?t1^o`R6kLVtUmiGnX`T}Uy4ILqdWnD-@Hw9=YfUCpL zd$I40*7w|v{0Fw&E9o%(=M%?Yz2qVChm}x-%*07GR?=Tq116Y+sw`9FtrkK}NI3bB z4M)6`-TZt5cocOau+-NLH#FVfW|+)9iG7_>uzvn@I#@7LlC?|UsW%|&;Tmx5OAWED0*dIF{>A_m9zVYg+fNSD_sj%sVZ4(}Xy*qrCM)LJR61o#r-w>r0X%nuV zjJBxigtD!^-myAPbrIAd0v~pqIOZx;lenou;67M1I%9>4 z`Yab*u-1Yi3STmiCXdJLa$D@(4--z3AW-AGKk{72p^wORF8hA zA`xR*!)KcBx_1_FZ_2g2iOJ!f%0=XJ5S35{^o0ciK?e0S3qZOn$l%P#RX?d+6V@)y z)O!|&iwy`XjdkhtZM;g4g^X4Cvw=S5%DM0<{ny_tGNrn$7Hd7O0gr)Rh@1SCV+Ywf$!!mqbw>%TSHi;s zi2WG@E`R)zCM&5;^0!3PiVS z!d`9ol<1&RhRwJD)P9cT0#{MY^l*_aJ;S^b`UjYl#4r*B;)LC!y))%MZAC+}gp(aS z23vi=T}NxtR9=Ly<-X&ZTZjGkdV#%2#6ePl?7i`@O-on?@a|nDBdEGgFPxg_kvs_C zs#{WE4}jbSaPV>09(p5MzufqyC#y6W2)g&T*0G`jD5vL@ zt&h^W(`4cvzLx|tyfKxBrjF!u4)k=;lNR7?V#4F_luE$oWB*SCCiRR=_8}UK=4q2K zHLxqpAxJdf#W(Qr-04LgcL!fBb=ErPoLO-gLjWtk9Z{ur>luGT3^~e$4_Xy2m(Yv!5Q%UzHUFvMTU77aX_L#fu@Ug zElY4cN}zzRa(88khq(TQcM>T?!Q1T}93@)X^aOkQWi)INm!KYHwOq^!ur}%&wR7%Q zkVB+!86dU^yDr}{{^ji_?q5#)TB4a&8~Inx)b{yyKLmiG(Hl?_B*^jwYh~5}Hhp>$W|CJ&JK~4fMJ(HFW2Q2b-o`u=vI$ zkWZ(1zH+m%A=bdWe^Omxx}s%-iF=#paew1Pd|17AODwp(Q|+e}&6=`tbI>B^M*3I$ zrimR}wX`e3_C@ZIQt|H>K=e;=YG;RfLsxnjZS=?hXp&nLHTUiuHxmB$VEh7|Teg7w zJhJkI@6j~g&hEvjAi=OUKfPsmqQdvgzWLOE0`<=>A>b9pOagV~dP$ceF4hY2KBI_*alRsD|xkd&S`ZuyC{rbtx_5KyPuKm{ww6g~!bz z3<*XVhHSiF*gi*o*T-$S|ydcimA@T;+Ps2TL~*dPF!Z1$Y}iH zI5J^CT;|)F%Bl>^Mt&)Ak_Z{A1yLVp{Zot`E26LK!>1O#ggr9g#=LU&WO6EGpehxfJOJF%f_f$M;17T*0qKA zYlp)?=XC#$h8_(H;Nf=bE`lI{Mix?omkB*B-w)bFpB(6amHw*g!C3?0cdtXqqv+xw zLYTkrs|v`AQMXYAxZBO*S|G8UXBvrsAHq$gy_I&hE!6J@_qce|h+`}eHFnuS5_r2) z^GerHeLLj8m6RB#2Cjy>c{-$;eO=v@i2T0@bm$hla}L9>QeFyF3V-#R3l*4{oR&Q; z7pUo8EZ;P8NgvB5Z|t*UG|vM9e1`-VS$SdFT;_IK*<(Ih?lZ0;; zU6PE)P)E=DEdNHrbOIs`i~Sh}AMbjpwlVw?wwUqKLO-<-FaNj;)eHgKQVOI5lrByK z)X##mKMDKk7}nfBHdKVbU?=HyKnZ<8+-yj;izO;b=!Qh6PkT>zIVq*{;hWYF4IUiOv(qUWoGuC91v*e%TuQ9 zP<{RA?)}i6gc;{wCI({fAdyn%3~P-$JtHv7RMTT z>wVf(?B$_srQ5F-lL2uzx7jGLJA9A3PgHpvbCVkS0FLy!HZ@^hV5tOeyd`#{#N>w} zHwnh4JZcUT$>U>01~y$$lkvm9(jI;L5Z$m8E1zKE7D*x-{C5LdpspV&ee&BaXyU^0TP-OcZ# z*S19dl!N|1+uMOqs%uq)VIKWmZp#IJg>>CEbd-5(&(#*l4 z1Mz$$WCd~H(_f`KUQw_mShpY*A%$2;y<_^Z)V+z~U!t+|alX^P*ocABPnl(qu5tSM z2i2Oo1k%ImICC`GLpp-u!Rz;G>Xxe7%B81O-yHvJR`X=QC|@iDlr4TKrQ`QzW?mH; zV~_<8t(>w5)v%AmfJO{X6cisiUAmR9)jsZakW%RNN|du4kQZGa_8?Y?-OFCv$OT%i zjzBG=Z58O-g?xdQykc)P8BNL`qRIYoI9=P7EE^_!F9*t>d+B&k5W3ZwwVGivbSC_& z3rBdWchUz;VjgG)7&s1CKrz1J8YvMN%3%sa9GS5~KZ>cxrNU=%z#|q7CLjF%b3If2Q!g zUW;M5Bj#v6;7w_+E&KLpQF_<%~8v*yrNQG zSJ#(Zuf+9OZ<=t#`HN)lL}AFhiVfR{e^!1NrXFP&qfnh$HJpg9NnF)7BiABxIq4f} zAW5-ULG#Gl{AS9cEMX^*5QRtloMz}v?N0})jF}W9{-&iCfKBv?==Tw*Dl=$DvOP^J zpgaeKk?f+;v@dsk;QUBVz5g5E*k3wc1pH2v9W|=N7k)@m9I=6sOhOTRgLkOVC#~|a z`Wgn9ap*8#!Puk*_N)XfR9=CvIW=MT#J5qfoXd`RFq^t}C(yR`vB z)>npd{L4h%Li6`eZdF8f)jrV`5o*myD6N&PN6|{r?oTLd4X_p3Y5P7P1!2uLBqkJM zaVrwlLdirXkG|}^BXW=pRr@5a^M@z4T6hm+8^u(bRPrUp+lLEhEM}i~#w>B(Acr?R zFCoK*!UDafikjgDN%CsG=1+Q<4iZ9jb$rCsY4NNMINBreP9R&TM^z)_5ZX9^)i+hM%cYS+! zpWv3a6#NXI88up)WuNX4H(QIPHGQlB5h(%QKeo&z+?Ei>gHc9n&eIMD}&S^32ln9V;HMKJ*XVx@GdsUE<#Jj31c^c3C~LU;H% z^HNj%?NeY!)MK-+zwExNQ8wBzrGqh@B4L(bG8vu*f?PPdQR=*thUesz6N))EMWDU)gnJkN<+wQd3o$C6RQTNyhP}?tQg?IAL)`Y*U7p987#W zAUyOO^!`wnSE;eJ+h~7`Y2SBybt2sp14AK#T;-tPG~1Z_vC|VmRjF-T`9Q%`m6GB? zKFjz1Cc#f#-Kx5uMZ}1Q5tb`1|D2KYgbEzrIvy&OKo}jT7`eFo`S@cC-&21`dzNJt zm1eKv>DGMivOTq?st-Z~k&kkAB`E0F;}g_wCp@FdP$U~*qdrj@-OyFHo>M;_-gV!w zbMHXIa8yP_V%@EPwGeK^(whpUx44gjNAtJAJ(?obYi-n|-7!q{);)m5V{IRj%N77v z#iS!|RhostE>`Rkg$YH`B$dUizv8D6PUt=X-xoZzbtZGOg*1$O2eqbizNiZrBpK!j z1?b|qQdmw`Zh3`uDJ3RoZ$P9c$90zS=b+TgF5?qhAA;p1i=X876iZ?5=0=aU|?a;ubZy77Q6m*~Oa# zbqu5ZX!!h5O)Ro~1GBv?@10B^JS-$^AwSuC4u+o=wco<|W0?JP`je50V6#~X=GWgr zwtYlBu1CZC`&77=0~~$QEMNJw&&`>(?JdgAw#NRvR_(OwqkKox8|M`wB>-arm$4^I zYdo7YjGRAe8n+)(Wu#3`uuOc4|539E>cjf&_2DkfdhFF_II1Y1HVJ~kCJd3@Dz)l~1!B-)kZGjK=~A4OJ}9wf)uJ~jR_7a_$_ zyU6V_gXw!3pSv!h`@j6zI)L0!C8f1X`h$fo6^IFG{_YApyc-Bp@q zKth`sTPB{@1ZRg53=^uE2+S=i4ak)E)cqLD#X_J6g;tsmINCw+c~+uIkpIzZn=0v4*pU=8f~} z`1qg?!!GS_luP%!fy_mjEQ9I;EE_XWCG2u#ZRyC7#|S0+a#b3K^szjD*#bWhj==}3 zXNW0_cgs0Oz;4f_N)?!Mm*HOqBR6+{6>0(Vx`WN(CkpRxkGT|f(#;i$ zifs~h!9Utf$--vXaJ}9y7i?V?V00X$_2rFGSvGEt3g?Z@%FM1Mm+%PlIkabFW4N!b zJyuM6VV|ORnAlWXZx)v>XGp z48|};J(}3o$l&w0BnABYV(raCzP8>pgHUS=+^b#{;aCC?CR3k?uhLHt9-uoQYIxJ2 z{QZ2q91R)FH)bPMp~D*LR?P~Q+ppqzWK%tr)oKmY0ER&B!KV^3zeIQOO*|_5T}G-H zmY!4O&Fxixet-x;Nfyl7bsPC{yuqy?)H&K;<;wNtUN`LI*_VVuB=8RN%ePzf7!QxG z2N_uswBS;C=+sE*n4;tn`L0cL2prd8(tp2{UjHWJwMp3*l54Wmcjr$-$2fU=?x4-L z(8t`F0|V(UGujCF2${*$KR$-ktafgy(D0{!I$HyvM!=wdCS~=USpx_{9sAE}CzLDr zNG0dmQ7u0-)5V}c#^qd!hwhoRgK)eaP(x{g*`!cn1`}s^jvZ;yvd$&MQ}^!lb{LpX zw&V9Y5s6Ij6$@2(ThssuU2OTw#_gY{sP5oP^i9Q;Q!jcgRW?k-c9=&zSRL?}7G!d= z1zwkQO6-d&?2Wy64;S|%eI(AM9qBIkpy(kSyb&C{p#KF}I|ncLyc8Ha^ZzzVUF+F7 zZAmwM-53odbiqsq^f5xg56^lf8=N*OvZg=>3#VD@aWz#$s-m5)?WwJ#it_|Wdq8!? z^oVTRa#mM&8J1Vk%3?M#hpB`ys4Iv4h(!)#%1H=Ym2hi<2~(BC-o%*H)}i_Iv}EF>b6GNz~$lGN5H> z3sRyt8xW;Z;A4ji;S98B2pdV%CGAN>Y zBznMJCLJ4s@eLO72I}Jx+?<;i4Sao{s*&koueAXh*oP|jyCkXca4yU_{xtudxwdyv>lvvuNDUtqs|g#860BCd zRxzT^Jcde0`tV4||3KBN#;&i9MM8qrD+ZOq(K(To4&=Q5#$G zj?!*bh8LyAvAqEyVvp75!Di24Ljgi{HACqDl!OuV#sv7Y2XKcPp&z?L2&JLL9s9C~ z0=iDJuKL(wl|_&{Q*ZYO{38d>pECTPf3OWj_^;-|pR~I%>J#$j_l#yOnB# zTm@^UTo|%8{FrOXig<~O1#V|XdosB@tcM4-(j|GQ!;b)wCx>>t{4}fPhXo_j{vGo` zzEoovisVw#66RicC?XuyO4V(}a46BovajZY^eTeqY}Z}lv=w81WLXkw|jl*Kgez+TyqLZqRg4oT^<3uoU4QJs84Ym6-^l7(%MV5Dp1 z_#Vk>+)Pt^FSSe){HZnbc=<4Kd{snx8sRZ}G6s}nmkWdr|K_dnS-N-jql0ocl0O+c zY!DQvQiOzb-s-brW@2z~ut~~maV;&RI08{-`FHt8r!m=79*0XTI1!Ob1zaTHoaP`I3l`CUcs#aV_ zjE^YG#XUGY|Ww##rZMsqq9pN9f-%J;b4q~(_>^xreYTw=qkr(hP5k)*8PL+mQ zjq2(YQX1jM-!U%O&@{8cn)n#9snufVZzX`~B*tgH!+ClNYIwH&#k1+l?+&I{ADU&f z*FggpswLgRro(O1*so>_AI+Kao`q)Hv{GF|LsD$g>|KJerv#d_WU;)K?1e!cvUq{) z`Kn{N+`6zeMmLWGj`@LSIO)<5*KROFmsK`+uhO42-N+@IUbw2YeMcsgcEgs9_P&6} z1el&DlR1Y>n%lAm)OfU6UvsIt+ccFHhaR!|=#{~d?Q?5hfUVEbKtUB%@hYVpcinF( z1uBb)1kHW)vGVIedHvIy z&)W~uoAUCacU{Oap0y`&T6LhM@HwHz+!wL#Fcv>BSB^DWMlTk#;-opuktQ7_<*W`$ zxQbVp1@?cU>tfoU_v~QhcUj!SRo_Oo8yb%4wXX~n{t9Juus?NQe5Ma>ijM};6|6aA z`=&g4-`5*F__c|^Z>p-Db+mSO7`Qs@o1Oc>_QYY`A+EsQps6z<4SjB9|Fnfai#%Cy zXBno_IBuO#$!cZ)^YF3#=XcMpHW(Y7k$gFw0zo2YAU6cTD`HtdId_KO*Li`+Qo(~@RcJ+zv1Wi|kh(Y=i;ck7! zj+c&%q$vxYLedi4h@d;FHpl-xM`uA5Ht=7ZK?ojibXMuPtZ4Ml-@uzN7MT<|xbw%l zR$!GMbp(V2vhPo*mVB6R=fwNC>Zq}WOzN>vL@8^-!8ice6X&b-f4G3(5^(_DtVf6N zv25Xhi?{k$EMd`l8tO=)_}O1it927;&m}7U-pX?7;`bj;f2pq&`2x5_q?q4>C-JnX z+I#*{VK@713VrBwB^Rm>6d0BcieYMcQ^fO~U8!+g-#;X~W5nB2~O z;Elo4yoj?#{9C+|eLVos~tl8ZZpVabAD&NW@eR0Uecg0Cyn;9~Ex+$k~4oHLwQ z`%}o)Xt9er9t#Nc>U7Y{nl%7p+ha(tlMS(SmrJPrU+C~9Y2X8~IzxVu?i|T@YB(w%7=?u^GMDp;x~7DK}0= zDdA* zyZKpQGwOEopLVLG2?B3_b6>9<5dj1S8KY!_L3G9aedugyNtCiFJkbM@dU({Z;@f}G z;(N8G8k<0nnPFaTWp(K`-NY?OVPR9)Zn@~Qgc}tc$q+&7u{8zI@F*SM3p=2*m9oMh zRi(;62seg-TiU2a8zp7_@pE2R4#_sR$i-OCyrB%cl-^J*(k zPhXdurcSuuPUE}#iQAqj%v;$V6~DhOePgX-ZWr?B)3Z^aC?rSUcPYbJSk7cbSH;SN9M(bEq^;a6^BiWp_HT< z2NpmOmoL*%L-WsvWo*1wu`o%!-ga1#Gjl6&tUq$q`!VUaC;RRgjj|&g zINy;1IqvUh!teCE=zxhD#ayx#ojp2*RN7X&2BBh~!n9Pn^bP{QftwaMcz8sA>K`f4j z3r2%m(MDW3Yomb?6--LXRD%I6+Xe6FqFZdNJGuSLDtm(0NvLxsux9K)94`-?Wt(=+zn49;Gofcc!`zh39 zpF{n+=|Q9UGToGZyIt~o*UYcp&Y}U3UeGQbY2pk^{ z%fn{8vE{?o9BfACDl@evN~M?>>FI=>GT1!$U%KWjB6JlU`s#>v%w0#!hG6ry3sq(P zbK}ljS@MeqR{6$qtv;A`1;0N+hRUU$yq#)}Zwv+S0*#mX9N@BPUtSl^>wvi4_0sxN z%(ooX@LukK^~ZBmxVtC~Um5tt9)n-~XA?DCc+bS~N>Uo6p2Jf*?*I6D zhPugjTT&9k`;Th4N~ysCROFFfFNy$6=tJT|Wq!=LVR#n`!)UET;l{I=0D-Ep z$FgB>B?NIw#o-B*@4f;qG=Tcak4|wUle;S;N5hN-BY#*rC#JDNLZEI~T>k9sHWdpy z#g?}Q+LR7&A=YOmGb3taqorzC-w)MjWW=bM4k>(HuxD}SjqaQiv+v_+qPSftjc%2z zk~=Y)95|YwLn6eK>MAqs+!h~UHK;xmsCX6T*HOC_ZG`{zcerMq12FjKB$bW6iqHPs z+N)|L>rh+|@bB*V`a<{%T+hq~d`Tw!cQ|&OsJQbIKb2`!E9$u_`aJCL(osFi{?zoq zq~@~~BlVBl^wYAFbZ=dqF2whl<*x3XS^dMlHSYcR`m`1Dr8(OjwH3;aC+fOk({XkM zwNH7=%;i1CIAZXl0_fJ9r*@9Kf-%E+NtW5+EF$`MzZKKoS09Dvn5NzuyvXDj5MLl- z^y5r9b}bBE@9nayrw0(c5k89x)*+uznNY$E`)vq+?~W0dU3iAr`<`gVKXm5wS@Cr% z$L-$>vAlNZ8F-OxXTo77{N%lhA9bMzEb~zd|Aw^J0%0Kn102?vJ=Jrk>xB0vtal^l z4j;O0UwQtWdJ9yuLW2K%g3QT2XqQ9j3C?vo6!pX2tG4NJoU6TiU%3QsgaozMAFza4 z7ZQ2S)0y%DGK;T!W^n7ctT;h2I;iq?;S7m|`~zNZtVs>KCHPPE?(I#T++VQg_V#O! ztE)rvdmhTn0j1fOkW+&{?i&N^g-~O~KmPi86s?9a@+I!Vt>^b_3-S6j&pUTY40#*~ zr?wXycsU`KUy^>Q;J?%tE=iZx-3L}djNttsbbrniUqx()1mfyqCP%-?0c-(!q5zYZ?B}ho05FegQxIY z7T%>fy^9{ zQLjD}CT+`n16bG;b>5s^MZeC`kYrYj>lZ{LutA^RRJ1R82!M+Fd|o14P{jhPnspfl{5+U9R!vLi*O|IZtf?u8GE zkRF`^9tF+A%)`OL%9@Us3Qq9<7Xqo^gP?5bDw*K4KyA8_(iSsH|K0=bjf=hMeIIvQ zXreXEkD^j#-Gba_l+6lpTm0pc#n)#(YD6{}`~gkar0qAqmxrAVMOtvMSSa;ko)KQ=_k0ZCauI5w7cKjMUKlcWBH)^A3R4>GivlL5Zds$V8Qli@iys2Ozt?3kKcCl1w*{ z$ts!HPbLpQc6NW!n~bb-US{DUp6(!1A12d>=(AD3HEah6)=O4NQ=9QxF@@X*c8FSD zb%NH|bXk8d%%0o7vuKsZN#jPKH9|!!LL-!bwQV3i4>;(I#uvdBuqbCEU9IIjy|?8= zl}~=2ulw-hW__O_6A-|;FL`#jljbQ^RoO0tHQ zW3`kCy&Ft_gzF}pN2%im$p$I#Xu67LDg4Z5LQx$MdE?^7%O5iDT2T3E^{aGt9WEoB z_vJj99^Yl*Iy=JN&kes^X3_FHJ%S>mrA&@MdA0&Al(ZEL6~3+Ds8B^zw;M0xtP=3e ztVVxT3<#XBEr%F@*^KDl!t&}!_ZMq*7t0wtgH`;ci z|GFu*9NjL`e9RY5wS5*nYrf=*CvDmTy8__xfp+G@zqb8@5U;po?iA+$lwWwo%FHi> zTr`8dWEQ!jH0p1%MZ7X+M*k&Ro#oMhmY)9rM6#bEmvI{b6PKcx171-%Ha9~#HbX{3 zLqb9~GcZLoK|?k*IYL1~I5#slF+?&xAUruXH$yo#LqLmrAqR3H4~igLCGcq<%7+3dqjk7^dJHrr`u;;1p(I z4$fd67GM#UU>R0m72>d_GWca{tYfRZE?P>c=Dk_kP%R&_w5i&1ZD~t&;L_5z>ifN= z9re-@mUh*<`LVR8-bcEnef7)cEFGv{KW*ty{n)#uBlVvjNiG{`iK8WsmN;7CsD+~z k&T0`-|0(|8=a(z_0~3ig5trYZ10M=FI4}w&B}Gq03dx?S7XSbN delta 17376 zcmYhhV{48su=7!~5)}zqk}}*GcEn9VNVm0~=(-h=vdW z=QH&RuTQXl1Y>fI(H8hD2n+~-6!K8=e1D^U2ObqZHM-i8D2OeTF>|l;15#vDq|9NL z8YN%Ck4-+Bt~hka76BXHV8k4iB@%lJk9TT4f$t_uTG)WM1@$_c&NGf|!0m6A<&0+{ zg-#}mS{MsLD|hqS^PTthuUWCFwSpx72aJ(=Nn9GL=Q5t${+R1w0@=Us*NPNLX?1gE zk=$0SbI3_sw}})+(*FGFzd^|`ss3APH&gSl;}^}B@T}LVNlH|h5z)662b7U@q4?;qb`A;enz z-T83_y+`n|8l0>J1$vJ!*5)1q>VcBm#zZmgg~rcO9exi7F?JRmea6a3(`r;2XR*Vd zW26FA>Idvnl?k!PaFjSfCX0sPIc793FE>|r zMci9-TJOHzBOh|n!U$%d{Rs*s63E9g(XtYFX5t4-<2oJ)$81X zAp!s;av5_#`iLmcDa=BO`)`O7Pd~!2-i)Tqu-hMZI%^mX!i6}QC$SdJ=XOWYtxYWr9|8$oN-CYzFm3|A(ZQ2Ifd*K{ z^D9Ok(T?ObAvX{Pi>YU$7^fMLYQ`1Ci4a(4s7SN>Y|3wd#Fnt2!OGPeHU=R%l+z8& zBe65&v_!cVh^gaV;+{7-g9D1TUPBOp@=;GSFAD(9uaIq$wOrhv8Z`}=I7Q5@3BWtY zGK7y1!mRw_W=hJ#I``sMII^vt>i|5WQRujB9qg%`NS?vsUW}`3$elL~m1CBkQIU_= zu?&SE%WUk04*tYUF#R`tg;zTV?)QOh!EN7 zVtk0ik+%#Qi3SRqh``h&wLDu|Uf~o#yn2ij=)?2ICN9@8CXYxw6)=?z1tR*#sRvWf zHRUzW`37>=3c9*HxTYFvwqrq3^B6O`ph(P`Yzf8#IB{_Z??7`VNmXuiQSE zSl^q^P;mPe9U?LraCl{@mw}U$6mJyhO!!9_pLg*dL~P9ef@=xUiWgCc z4?-mk;4_ouVlam zS&kT*b6l+LyjcsvU85<_!!nFuh^K-K}TG&TMflLIIBGx7J+7Ne}fFuaC42wYS78(Xb@G&RBaGO zuHt6F-8xWc??eeWB&FY>X%p#_4zDYp?V2#oCY!r@y~F9s;rfsCE=g2i6$!JMX#=Ti<63{>h@rh;bVlT0{}d>C7>u+7 zFrE}Xb=1QS{Xd4W86ofF899skuR5Hf$GxiIPC1$Kw(Tr+b4TDBQQ^oa5!a$7&EKuU~A2CGlQJKwisbbe3H; z-b+GLk#{hN_A(*DLIEk3qYk&+acO>82Ms=uZ|KDUK>$<-#`#`5bNHsw~f&4BzGY za#5*o>(P@VW_@wN-+Jort#&~c1B&v6V0|mG$^&_9sw8vKtC7u7MfvBUuNT3SHiz*Z zP0dyQJ@iCY&_*Lqt`|1V1yF)65xy8V zFP&)sGbAlj1b-LkHY;H@ElBc+C)Ed9 ztED}*;-28cC&FLtb)h1lPxOv^>YtXH{7a`Zt*y`w8bn`LvAsnh4T#iXVJwu{cRg$_ z&fe^U9MRkOLsxsK3U288%yPa>x}qKW`MP2Nged^HJ8lzR6wJgivM8~xZ$LZ7lp?Tb zp+#ivmmX!wPKcM`F8}_>n>ihgg&n;zW@r^gQUjB16^ew7>a*+C)!9^p#qk=vOe28? z@e{2D;W<>}LM^i4#zs(cL9w2Ahy+(?FFR%I3WxXT`Uq;>QLTWx*WHC&CS-xDi)#?j z3vVH0681{ZgiHvVLazubLKZ{Y53V!5`IKrVJl9p5p2pqRUitd<=FYegBd30*F=}Lr z>_?sFxa-AUGXB_7nW_3Gx#?-bQ~3>jh9;;O-!foZ0s4z5s$!>W2rNwbS1ZOgn;e8e zDQElqg9ul30c?l`yc}k{UdaWLzYG(AS1(Ql+|1|Nv#U3<=FYJLMk!eGhXz=%@}f1A z$@T0nzgHk|Z*(Kr#dE#wnb{f-(y{RjcB&M)@|1JO7#-JW-uJR%mxs?5mSN17@s1DkDaP`bb3M2pplL{_p``+l z%z9EtWx`)ZW$4V|;k9`99(`DVIoxWJM4=uBTnptU`pv+MJhje*ln#yQXkm(87~BD- zwhL#O{+dcKA4Wr9*w84>mRL93=#_%WQK|jOcsUy$8OAQQVehHTtp2z^36GOy*hF1D zt^r=n*=D3yS)*!2MyTscdKbRsYl3VK&I$zF_hDy2ywRUpyW?w$8rfn1AEQ?yD6A(_ ze_A0A7j|V-Vm#jVxapN^7)TI%eDP=CTFgG4IYG_p{z@e}wH#LWnw9?P&IBHK2-jmj z=l%=$o7XKO*F1=GSi+60oLCVF%+cF^zotcISiwVJW7Uh5k5>DzuspNoXeQF1D?f7g zDdLt$a8W(us0+|+*p3u{xWysYH%Ev|mdasl*hZ`sotW(+SGzJc71-9dTb=AU-y_knvSWx6)Jy>U#lqCsbkqnu*;B+>0?;{|gffg_4yQGq zQ*p;G$WgvFwx-$rF8fy1d36G>y4%#iN9sq`ctS_Pz-dEMBZo zqk7P-qC?w|TGALmIQ(A{3515iz{VldK%#RJt}TVpiKZevqcU^krHvAuo&+0gRh(kT z%-t#Jk^--TieYUgrVnS>_y~hmG80nm#g#p_Wtk->T07PbI^(jU>^N+ZZdMenDgW!| z0_dQ*qoN30yN#m(h?8;ixUFy(g4|fjes3v=pqch44Cf%=%hB`<>SmR~nt_)oSKiXa zzBeCxPRHdSRT8$_xi9y-OLi^fQgdC=AstnO5rc3@EYva6UFzvCZXgfb6;P9{KSBZS zY9Gw9?k+I4a1Q}KnY3XJ%v2wNjWn#Ml*`1LudwisLq=tr%_Rzz?Wi$)qNtcDCN<{M zSCZ3>s8bPu7NV_Ycp)#HHS|kY4E1CC@?5d{NC^&8D^{s1XUXk?;eNX6+z4WtI5Ue& zN&VP*b?C51OpCswP>FRnW`)_dic2mqZ7pCjZ4|ng-|4k;Xyn}BK8W{C0aJF~YuM?D z^8tvM;dOk-X6~%)#a>3sS;rI2S-uY1`Vq~^KTtzBkF`xgBafHp)FxL z7%?B-kHbsObpO!0p>uF;X=>QVu1j%VWVIa>n4FzLL-_VIsj9d?*|Fy@@ys;&EYQ2I zAFra)pG{Xl%<{jakUI)-k%Qnru}+DM8fZU0H3(y;shd4CeyMsLmvQ3vi3e9BV<+2Y z4~YOk&l{truddfazaDfN(t7of2`nTwB|oD^uM(iRx-Zd-1_Gkrr8hhr$ zy*4CnMY-J*0gA&kVH$;O@$;7mTT?{A<~CCoqVGWc?=;;)c*mt`vxh_edq6SRJ%k_t9H~SD_hjKU7RG^U=s9_;K0*ESEb^XR z_)h7~-MjsW>P|B{ppn)U6ou`(nCSdGxkbv$p0${ zq~c6x%+d&>976%wJJIbBSu%P17dZ3Sdj)K;8`Q-MN?-2JWdz?UNuoF{{LP*K_fY)1j6pHAYeQWj# zo7iSKRhy_U3b6+>hTpZbDj3HX$_@TfPoM>h-tAp%OKm!9V3`wN5R07Pe&wY|#`OjmeP4KrOL31k{x~Hr%%1fwUlg@`s+OX5mHZDq z7Np{P5P)HJnMI6%d;LdTo%uuv!e+bihi2}duW5!MQUz@j+}~UGH%3j1tiu?{r)$z;CmSo;zC3Qkz8HRzF{?D?k<#G zt-q>X*%%l+t;+%@@vTS0z6+0puJj$q0Q_~U0j{bjO_lW zb1=4sW@jQ~B>bO-mlv8r*390*)sm2zm67Ft>8u`2nfT*Yq~5#w^R21_yjaR4$C)C*quQhmPVYJgZDIkCa-=fT}$=ax?*9z#?z>iL9p z4IqTwDecPa8-Mk8KX1(nf8#lJT2}S2JR(eQEq=M?M)CQUv9qp}Y9|b0G`Cb-CvHhw zl+S}<6oGS{kKJA2yRlZmJYyUghVCa`WnBVXTO*zirBAC~vr{J@kR=4g7$pF)R7|U$+ z#Ig0r6s=j%^!nMG)L(y3Z(0-aXn<1LeV7ulElHuF=QErZp*_ZZobASnr%5(w1ZYs% zd7sjFs?_Ux#wA}kOC!ILso&AyFT%YR+{%F}q1Gu{r9ken=8TL;H3m;;KB;UjywYw= z_WGLl*y@EXJ~GUnS<{huaa&Cq4&^dY$3e1DLTt+4s>-xBq@;s&N<-hz*;&@uu3JCu zi0%z{(Hi4Z`c@O>iV=>~z@vQW~ z&v;&TrwmU|8;{)z-nANUF|xMv;cqy4yk^v2&kc&UQX9@RRij7dXQ4;y0bJAG`5*V} zWBAH4j%%$LFc0x60q8CDScXtvkS?cTOKIill?CSs%ECgx&_QlZy#m10J?U&IqWGkm zFFyDdx{kaUm6bhcg^DdpwZ#Rt_)1_oZxV&E>IpkAlvMbW^pra}yFxSq#*cKfUBF95 zAK~BwYmSVV<9i*ju^(_@0n0C$&^W#@n#f^vg)t>n@tQRHYjEOdI zg>KXmQzSh00g}4!=UIu!apg$J0bz6467RT@LlI3xONR9Ad{3>;08k@D%qL9H-W=o) zL*Nmvv5pX&rt@`mKUiAMG*b;@DF|^c9AELYOo={VdM6XEPH+)Znafai*@6UbrfPQ) zHn?fNr@&7eZ*|1WsOg-ON|=gbV%8lkZc5qWA_Uf9;1JAA5O(W*Z`?k1R)mv3U0{ia zlYc+Z-{mS<A)J6F4Qv zuG`)7ltO?KhI)shJ=XmRLGbdOvcLxe8L#fk|K&M*m(%4N*WD$_Zd2XBDS2OpV?ai zS9q3{)XMfh0nE{%A@Bs%fBOrt(mIXCnz}gN*|c)3(eIPb`tk2CJwtFR)B=n62E*LB zoRhZy<1bNL1N9^2ivhtQF^wUZKh&8dONBl0a$_pev)4S)s~!cju6%A2x+7MJNWWu~ zhwD(5dAZK=KRn*`*F2mt|aS!e}M1-lQPqb|P+$N2GT6Ux2s@<83xQpHA9 zj;0bMWPSlsFLoaQV&)B@xvuMZWYKjLtrr=h!q;UmAv+{+BODVNd=FHI0Brm318l~} z%LFM{z((BBXn(2o$z+*VDJmA@18!@jipR}o;-fo(^mZQl$r}h!5lseD_|4=4Y4#__ zpU>+~E5aa&2Y(1H3jxjy5UtZ%8Yq;j5Vu99G)-k#7feBZV<3abs}5v?GL>>--U}~V zh*3LLDySCzehXKa;*Y1`I zppFYo2ow>(_R^pu_Kkgi`=|~5kGABk*hQ5ZVurPdqLsJb&@y490107uKHEu78gZC9 z)!a+l`(Aq*W!N2O>s5cT59_XuB=)T)Nks@^C!c>ug`hZ^Va7Gh z1QD6k1+y>)$YsMmP?UjR@Z;$SPpR_p<-M7LHC zhIM8wl4BrNaZ(z&ESSgY{G~LuT1CR_WbSBKemZ^vL*OuwCR`YIYk$}K8Rz#1z!wsr ztcCm&K`DnU@)W1=H8C}~#vLXnQq;l@*W(m2@8}c?<}AmwxBqs7!L&gyN`>_X8$oM( z%~dz|pT?>N6-|nkfMXW8D<$|vw5V=Zgw{pgR>}Cqf`m#Q=CZV zP{9fs#xpV<-l2=dCF8pBTj7}n0Qzqph@`MYC=IoWnEa4~??FlK+N$Tjur|O!J88*E zkP7~|O~0!x!-NPIxawHA^!OADbdl z3LQ^=O1RK1Q6|8z9%5KHg z^iYx#gC7?fMy5@a*B+zG0cUY4;E~tUXmdOFHxBQMwE6ct0z$U4pntA{w!i~2K) zbjy_z8nzn=J`xbIRNIq+ZY6~M6w3y$kYKiqu){P{yw+(vczpg_i(k*E%ru!03=6uD z-RW1jus(AIUO#k=^&Y^;x6plQUuF-ENtV8FljBZZI8v?%>2bWVZ{?VKA^A`c zSAh}c#0D@3(WG+7`~{<<-Gd#;io&(=uTx57q03iVt+vXigF3V=~RM)kON z9xUX|L4K2%{EeXR1d@wRV*U1kQ8u5To^^t7k_Ux8VZPMm6Yf3YAODK4(Bd!G&MMs( z*ECf)+(!H@wV|F0Px_U9BDm%Bpqmw!B$U9(u>)Ltg|MDg?<~j2Ux#Lc;ko6((4kcV zcbp~b2WG=nkh&Xr*~RimnUk3$7yH)6>3N^69|KfOhrRT zy1M5$tiy2SS{1su4>Pvl8u42=DYlU!3`;u9znz*(@U#oylL8vRZb+LTBxjnngF=X# zPXQTh=eJQ~P%`b?84K{NqQ_z)-x$*ezUUW*Qk#g5aZj*_0Unwh|Hw4y#|H)5`m%KP z))NvC3%ba`X=kL_D+Qva;26CGR3PRO7}AB=Fv@>@;BPZI5{c)?Cf?I;-J9s7PuS*E zp#(7HWG}Qo4E`WAF-*gnjXxD$dr8`6+ z=`#A|_7PC`&O8@wZmdB7gu%pH_lo-{0eniPwl6%S29T^qA@O1&Xpl+vcp}wwy?_c$ zBwDrUxG=%#ro)0p-IS=QNs%4lMlh)7*l}I%{rZuig3I|uX184gy%fIyO6oHQZ&0x{;5iNjX--D3cZ~b3dD1163DM zLQl36bWM4tC@SBF8xyVJ;xSs**3!locE%XGiYWbY&9%83pz{|A&_X5ZkQ2Gd`u;$HN|GsdD5Jp^g+S&xs^ z@Co8_d+ex7Ao^7<8MQkkDvej-+O;7`O?HTw?(%d82W&QAflZHt`PAA;xRDcdN;(y@;fU=l8Tr}~ZU}FsU0(Fi$lX*srOT@or$RV@sO(1eP zKfE5B)7ez$&H- z0yYM+Wd>H4L#HHGBcV_gP?5dAp-jzn2^u1HV*vqzPqolrB1N%p*j4#u^PM0ql?K9M z)Xehon6z^c834sKh~T!)#wWjN4!aX|l*&^EtS_qad~5@rLbHzxt0HpCBCpV-U`bj8 zB#j*%By5s=#=QvbOv(mpXTUc-W?d)*WF)Gf!;Dx2USra>X2@+v#1{uK;}nmaMlZc^ z;7%-Sz}h>?M62*Fc;=6KLPzBx1AZBS!rV`yFM$)aHXtgyM9sEHSgQj@ViTZM%S4g_ zvZsjDeOD5I_g)MOn+D+@93+JwO-rAy@AwEu8Tu;qX`vp-oE{WxfQV<;DMJx)3tNnH zlE2+rpUC3}XYmIYj7Ybs3JqWmRuEb)Vz-0A-LU6tgy7(1;SSwO^x%U$8rHk@Z*2^r519N4SuWX@ znH-FXlk@*LD6RM{c0~WV`gt8Q!`g`Ox^iN%9)9#DA>qTVS%}N<_(oa*Ywu zr0vzg3PjQBfZcERzc@d{STUTX2jOiDAm}4b=41d`Ln-{yjS(h1+{Y@0>caY|4Wohb z7R2kXnc3#Ty~Z&EeE6=FEQ>;Cl_ukZ1_BzfCGBEE?B{R)#o)chLXtA^h5c_Q>vNJk zouz%yOArRif2D_pka@|wT~6z9W@{JzFxDkr>uKNa2KH#%){>1_;=fkgRV|g1>WKyP z<;VaDEPT*l;+hnu&hVZsoQyLN{t9i4H?Z&aL$7o9G`|ge3cXC6dQ&CZ7F1&7?aVbk z>OXR&1Imu`B?4o@>ZG+{_K+_Z59&Dza$E})%ehXC1GxoVdes~fSvQl=3O@Z9Izo_D zc4}wzPT~DG738A8I@S|fq|C~V>Rs;rb`pRtKA3XLmNey%JEaH=v+`(7$x}=?&lQQ` zbN)p5+1S^Y;m{$EsU>u49lP|ZQ8Cg?mI|&cDYy396P#K4B(R!U%&^XxxT_5P<7I4P zs&uUeb3zfM+rzyq72iVRq~ItoA?jXo9QGmqEl!K^lyJ2Tkx9*LT4%UEWA_PETPZ-a zpK$?{@&$3cqA>|hx}F4@wB7v1kZ$t_VY@QVRFSBp@+WHaKBiYE8goBI!xI3t0E;B5wk5 z+#xTIkYc#?jSrvINfk_D7$%?!RV{B789^uL9F)w+m!27_2ZHysSA~Dnh$CjwI%2v^ z|N1XwXItA|*U^ssm{!%mW`Vj(#J@R1uVJv)=O-z-Pzz8ft&mOhRmcz8A;>P7n{DeI zjbiG+wEyy0DnnjlEq0IZAb5VUe9=Vl9Ij?S(`0f2RFWetCBiQVLNY*<5N;B>LJca} z2N}D;MrX+rH1Sv*z0WB`798+8`e)nT1BTQW3Ew+Qb-2sg?2Agoqc-Mq{CsQ;(o_$_ zAvuAII2}nrH*!BsE|6*h&h(OxbTIK4>L4o20?x@?YQC^Ma_E8g>us#I5Q*}BQ3jx% z${x2-M3Kr(7bIG3#S0jWEBr*xN5hI2y~YJuqhiSU$5JDgHO3+6dju1Xuy1K7k2TyLH|FG4;w6)^F_JMl#(W zRfGZ&PF@!ZP5BZHk2>)1O=hDFgR&@&4x2(z5FEbL{i%^PSSvv554<0t7IsjvtA_l! z(?p|7(moik`Vw`K5n)hhG6u2c!5#6W(lGiK=ZxQKVTJ9C;G&Q4u?jye7YZYdB&<*r zXj1d1_6j@0qDp#VXC8gaf*Blw>HVgYE;^EL2HyJsS7!##WeKE$vLjBm!=ZxDYc~~{ zXtrI@D4g<+;~2nW!M|Ue*7b4;%#7` zFyE2jR3+;G4VxiUksmd~JU+VNwvez^f4EikCnUBOrQ4y{l)rydqE5tnZ&iuH`~iPL-|MTqo~6fhcnL z@s6gvZ+-x`QgK@>oic19B5kR5fvBi9X$!5jt+WiH;RI58L6@WQ#_v?WphpDV(=f3s z#hv(bMIagkRWR0nAwpmx;dlzI*2c}9j$M`(#cFjhg5^~JK|X@PJ1>Mw z|30nL_9a@7AH@5gLbZ*!coBJQ#r3z|WYe^^4SoyVg#wj+n8+clB*mJ;Lu~KPz>v7h zIE|dO_O2Q+K}PEPnGdUAkRsDi7Rm~-w*#wOj@l9Dsa(iqIN2+Z8)Br1s!1**Y~QU& z&p5f;LGPeQ$}`JDuqc*)O3hB(I!_Ykpu`c`~QSM zv;Z8g#_u|V1p*JG_6^@;j9_qM3*L#kM@^p02~crD89Nz{Mr|b-sT>1<0-BEU)PqPP z^Sa(OIBk|3`b3rSN+t11l2+n{?9{aB;?%VH=KHkSYN%`_?K<&eW{S3%w1sjuaq?Re zc>&U|$wDH1viy~7;f_~2l@!VdEb8tAEP%uetoC04BVn-@Q*~5kQn+@}f~V|-ocX+I$t+`OEASlVAsyz}8OgAcYwJJmf>{`MfUI~K!fG%S~ zQJ_UpJi(i`Y`6&70QAn6X#4<#9+ZMtF$fNY!bm@yu&|6YNvuwSFAOBJ&kzs_5)Iiv zX*OO2X0y*>EM2XO{?&H`n_3W)e+6!Q~xD==lfFUrBE=Kq&V`jIru;&W{MCknGZjn zX8Ip6zRcBs5q&&w+RytFAmNScrsiR=n8&y;>t%|HK{Y3;{4kTVAGFy;I_BAY1JX)Z zzn`b3ts@9OcB$h}!|uKd7VNqFsRhe4Rb1GS!r=o$-#zr9lLLFr=%Hvgm%n&^AFO%P z#4pst^VOO!f6YH9vdfKb0owASZrtMT>)FwrOwPWlk%KN2x>~d$fK%UBt418dzkzG# zCk?23My&mNE^;trv<~u;dHQ(l)qN9slXdwpY+={(o`HLrOavb{{yVtGMgfd;l7+k3 zPB*SN- zKEND9SW{Eghw9l7Kw8)J^|FyCJ9e_ipAWUiv_9za&6oCiY_~#ZZRpXcUS`Rk^zR&@ zC!H$&tGJn-37ujCIXbmhlB#q6C1(ITd=fLierj+$v|@t7#~H$ZC#4^xq{OWQxSfXP z5yUWBGli!3H=4~PWrA+NrZ}^7hF6uP=hB%OOLxWeYkt-c;4j|kwHqoEGgaKkgNHDN zQDpEuZ0b)QR~3EqPdEXT;~6j4t7B>O7$Q(eD%(aT04@Ofz<^6Q{CyG;HsV3XY~s?J|1m=YWPW_@sdIdQ*YMDvg>c4YBzrTH>9W!*hwjv_qMQ_nwJ zFWzkJuxfP;l;&SEq?r1O==^Y_^!BB3o9CseJyqVIZz34=(W5YlRLC=N{Ac{5xT(QC zw}_O}?0Y9Hm#C;StqZ{pfa&%SXas4TGFCLwxMdCxkgw{B-}Bkz@nOm=?2_Bs>9x*N z?E!888LFISXIRIXOhuNcL0xgso_TN2%1(0)pDY3BWEDRYno7fUWH6$(phWI@vAM%zr1bVW0;z# zV4A=Uqj4Xfh$23#Mw58w)uY42t%D_oJM@bARqv*5hmY&$lpuD1HOdsA&Jj~h*!twn z1-hC*%}G@`y)h<+x)gi5FWe9Qvx>oEm;yczI7F;ZnW9Sqmdixov&pwP`SHU_6U6;< zrjghW4GUUh{7`Uo(RPsDouwaft3^keF&&MVqCX~a#4$SNwg~Z9GJSDsgQ5+W?X+Mk z4uf8@7Xyig`#B@tJl`l56^d$d=RAbdJ_6fS8xt4L`@u51VHPO<*qp zAcctWl6a-L=PXy3m=#1kYzu4ie>daI(m#>B&+019-?yDTjkAX@KK`W3zr#R>Y-#beG6b4ptj(UP)2$#@2w7eI$XkVYxE z5Nz$wl)>bZY_*dAYF@O4r;%KpKTu8mR#NDAsQC1qlGd@Ko@m%|3nu_);gdrGa4u6k z8s?=2K!lIm1((;qBWA}Q@eaI)Q=vc@Xe>P{J*RqY=W>B+#6LZw!5hJsBY-}=n)^J< z)tRp}bG&Kk9;o64E-+_oFPub5d=L>B2INm#r8~&3nkXjV4Bc{H>|i8uEG7j@v>??c z_Um166s^pjaNDxitCN>AfO8ZA)H2q#YM0vB=l6UEyg1qc4U^`-zZ-8Z(6I+KwvBc& zwNEX`>meJaMv1?L#?ZJu2Tu3Um<#~~W#mee!h^#5`w2qr|=CYGuoVXOd414=@h^Dd7=Bt>%YX zOK;}jXjZTRl62~lDY~PQkIg!`&BYj5lu=?cb>giprspkxytjsI`aOhxUO}xg~%AhRSarxa;5TGf5aC_F6SxwPu`R#u@x+X`&(n}hEXNQpFIEKN4gOfllk--r?6kvo!3@)|-v`QQ_{!%*` zdeM)KF0U$6n<=3JDjAdb@J(>LS&~4l@{qV_yht{J{|r|O`CB^em!-F@t6cPkUL=H$ z^;61qhzhxY?E>7jO~wUdQn9kbg{R9t2V*c(TwNp$YTM!%N!+pbi%GIh+Uq7eKDD=A>h=_&2CMVJBHX zWwB%IO`rnl{@JG%G$l7`Qzvw}v{sj?1#)z7&0;#>+vc9hTW+1lE_6}ift6~t%L}%l zdgbz{SJG?+D=HH-$yZLE%VE=8E6GXjR7<>-PfX0_b<)s6-V29VVM#o} zkj+ax-rHw@dB{v8Fgua*3qK*wl;eftkEmP~x>Kp6O&aYuY|ORkyz_R^MG>BDduLuR z9XL5YJMyI(s>t4jVcNnc{c z6;^#V`m8jlpA7sF`5w%E#y!6mo|#jk%A#aM7H^UqFELUej-U~!0Fxy1<-Mqx*4f!==-J(;3NkJB2uvUweWXf( z!KTt8y)iR!%KSC`WtB9Qf8+`b+woiUYTb#=;tc%kc3rsWbe-yAG^&vh1m;Xd`7|Tu z4~QeZ^QA+!_k0YVDzC}KSFSrdmhEIfYM{F8hfD^IaPNat(+76 zNc7WNz>a*j;!90GEC^di9_&mTc9;ynjFe$QPpud&&4~D;_bYGZ#6(p-RK-wMg#KKi zgD@;R4YAaS>hd7wi>3jPtK!5B@(Ymt{P^KV>-&A)NcNaZqgMB*Tu+Vy?NPqb8tMJd zKCY_D#%EWwN3>OW5>WG_&aEzYDP9Yg@-kAn%ccKrEuMxvL&W;k+ApW4Y@GwH2ow-r z_V7du=DL~&>i5p8Tr?A{@Y69Kl{~zgE>nn(T%U?MOHGAqSuLVk52MmH6D0SF*wak%31d%Zk7=f^WM+7%>g2}#Ry!;)uPLG_#9{nFj86ZhDH z(A%ItU?CD=jbdhcT1(w=l;QzDZ#OGlduyW49x~K8LKhqy0uqY#_Gqq@qgR9Pwb#$z zKR*Kbt%OhV12OdDiWBl=9qrmK4{oS%*@fqLtsfbFyaVSBzAHbkWI4UMAeT4~Jc7?R zwGCN}1z)^(aiZ_F9GCjYgxjF4Gy#|hfvq+fJk)a*90hl#to0)n4_*NMRp0+&aQ&G{ z(Sd$l|AFb}`z^Kz{ef8yC$iq?yOU<^y}8=^kLPRvJ};l{^%Bo~uzg>d&3j0CU)TX5?`s}try>UgYUt!ftqk-b+ zbKswSW2dR1t>Smn0vy1JA4|zZcnwa{4*Z%y9zHI9m!1{(ZlOU76LEyjlJmOez>}v$ z!0PJ<&b2e*U&+LPvR$?1j|r}ZPT zYiz@pm3l21y9LdY-~aXF;23fMI$~ilkY@fy{tF9vxA&pvpqYc(XXV(slfHZx zT5PX-8}Lm-;jq%)$#JSv!t!*atxz9=u~;kGd<99-lu#m^$O6g@f>~x)4;|nz<|_vF_MKa$$%W7RQ60vqq%QWX2RpShx$0+9`!o$ ziJfngAxn!4%i@p1PLX2XV6lG28h{d%lFPW=be*|0g-Y!5(u+$8bLowpBDN8 z2GCT{K@eub|Mikvn{ME+$%xp$_egc?WP5oCaI&qeQnkcVD{MyH9Bd9pEn-JT6b)+j z|I9^M?D!*UCQ_BP{qFnvxU&JD0t|e4KmbY}fI?6KL`3CCKngDj+^gEpW+@(c$fyL{ z4l)S_TD6d)7>Eb#bNPEMjt{hb^9;y{J5b-%}@QzwylEzWqMsnh`B7iQ9rrnvV(K>FDmn zH1NB6fp{R+RCZsCIb)&7c1I##a%+}8VpFt6W)k#kmS0@8X%Gh+yMfRNWOn_)oLO8O zHYIH9(Uk`rX*K%-%Q25^Yyj)7C7VCiJDUOaat z3k8;0kBer;ME6J)OAvN-IKXu(fM?bbC7Ut-&6mXwy#6ZO$&RUU)-N64&AKI2$tPbL zJ8x3zSZA{K+Oc9AE1Lp%3w?OM^T4G`p|073X%ndq6SE2f zYi>XsI`#`$u-dn^26%DVTu{@UUG&gVO#1$F?7cmT-+CB=X zpN=#LcyAu=^wH%Vc`0uJQcdFg_aAGX{xawPgyZqxyOaNayW3x5nl8IOMqtYAiaUpt z-4Fbp|GVDt#%-rtwZ~7VyeZi9@Wr~Tnc^v?y2rINg{#bG{NuTv`$0&p?{=5DN2B}Y zQWNh7>1M*8UNxMM5?47XA9=d@l;7(3*3GjM1Rw3u7ytO>i5>5?Oy+8P<3B}H+q&Ox zVXSoHo@wGxxs6G8$`(-`(S2(ltO%$K*)FZSXp6{)yK9`L*7I}BS+CnJ(Qvcp#r?>V?FSg`R$qZ`qbasnf5$-B(84HzTn59)8T1#iYbdE-yGe1 zK~^=fe)74+OD&CmtuCmX=gJ&a@sIhj|9ny8wOEj)SccOxXEMhHnORsmJG+{ixj4HT zni`rJ8@L!*nj5;9x;R<58k)Jd*eTc$R1yncZ*OdB3C-NHm$rMc1_~VeI5YOJY0P@d z*QU?s?0q(;_}t2miaAFfDR4AJb2+tt zqJKY^PY|*BpvaY^utG^uo73Tr{W&>*4-so?Y5%TPmIddOD?0^vzfHBCBg$SW-jw3e z7;>UXW$Jlp|52DM_O-=tR4b zM!VLEX0NFqx=mL!M!nl=zpAY)qqS=4tKiky4Q?4tX&H@tFIx0ov~ln_&-OaK)bvGT z+>7Qqi3UfBrdo-{WQkU0iDqYsc20{X&lydbGa4gjG*?D6xK8z$UEjNL$)}R!%t+Ix z$1ind-D_PMdi7SgclN5-^V6<{-nY}qT_IkvVav*Uk@J`8e%&`sH+S{?>fB8Czn`?d zcfT?}7I`c9|GT+QCN=05=5!`|%S1nl;65H@5#8TSTmY&$VSfMs diff --git a/doc/Projects/2019/Project2/pdf/Project2.tex b/doc/Projects/2019/Project2/pdf/Project2.tex index 1210054f2..827012c40 100644 --- a/doc/Projects/2019/Project2/pdf/Project2.tex +++ b/doc/Projects/2019/Project2/pdf/Project2.tex @@ -123,7 +123,7 @@ Project 2 on Machine Learning, deadline November 8 % --- begin date --- \begin{center} -Oct 2, 2019 +Oct 9, 2019 \end{center} % --- end date --- @@ -229,7 +229,7 @@ to use the codes in the above lecture slides as starting points. Discuss again your choice of cost function. Train your network and compare the results with those from your Logistic Regression code. -You can test your results against a similar code using \textbf{Scikit-Learn} (see the examples in the above lecture notes) or \textbf{tensorflow/keras}. +You should test your results against a similar code using \textbf{Scikit-Learn} (see the examples in the above lecture notes) or \textbf{tensorflow/keras}. Comment your results and give a critical discussion of the results obtained with the Logistic Regression code and your own Neural Network @@ -250,8 +250,16 @@ project 1. Compare you results from the neural network regression analysis (with a discussion of learning rates and regularization parameters) with those you obtained in project 1. Alternatively, if you opt for other data sets, you would need to run your standard ordinary least squares, Ridge and Lasso calculations using your codes from project 1. +Again, we strongly recommend that you compare your own neural Network code and results against a similar code using \textbf{Scikit-Learn} (see the examples in the above lecture notes) or \textbf{tensorflow/keras}. + + + \paragraph{Part e) Critical evaluation of the various algorithms.} -After all these glorious calculations, you should now summarize the various algorithms and come with a critical evaluation of their pros and cons. Which algorithm works best for the regression case and which is best for the classification case. These codes can also be part of your final project 3, but now applied to other data sets. +After all these glorious calculations, you should now summarize the +various algorithms and come with a critical evaluation of their pros +and cons. Which algorithm works best for the regression case and which +is best for the classification case. These codes can also be part of +your final project 3, but now applied to other data sets. diff --git a/doc/src/Projects/2019/Project2/Project2.do.txt b/doc/src/Projects/2019/Project2/Project2.do.txt index f29d1dd00..fc7ac287b 100644 --- a/doc/src/Projects/2019/Project2/Project2.do.txt +++ b/doc/src/Projects/2019/Project2/Project2.do.txt @@ -93,7 +93,7 @@ to use the codes in the above lecture slides as starting points. Discuss again your choice of cost function. Train your network and compare the results with those from your Logistic Regression code. -You can test your results against a similar code using _Scikit-Learn_ (see the examples in the above lecture notes) or _tensorflow/keras_. +You should test your results against a similar code using _Scikit-Learn_ (see the examples in the above lecture notes) or _tensorflow/keras_. Comment your results and give a critical discussion of the results obtained with the Logistic Regression code and your own Neural Network @@ -115,9 +115,17 @@ project 1. Compare you results from the neural network regression analysis (with a discussion of learning rates and regularization parameters) with those you obtained in project 1. Alternatively, if you opt for other data sets, you would need to run your standard ordinary least squares, Ridge and Lasso calculations using your codes from project 1. +Again, we strongly recommend that you compare your own neural Network code and results against a similar code using _Scikit-Learn_ (see the examples in the above lecture notes) or _tensorflow/keras_. + + + === Part e) Critical evaluation of the various algorithms === -After all these glorious calculations, you should now summarize the various algorithms and come with a critical evaluation of their pros and cons. Which algorithm works best for the regression case and which is best for the classification case. These codes can also be part of your final project 3, but now applied to other data sets. +After all these glorious calculations, you should now summarize the +various algorithms and come with a critical evaluation of their pros +and cons. Which algorithm works best for the regression case and which +is best for the classification case. These codes can also be part of +your final project 3, but now applied to other data sets.