From 81a5432b7808b387cd78f2fada769a0bce5ee1ba Mon Sep 17 00:00:00 2001 From: mhjensen Date: Fri, 14 Sep 2018 05:41:24 +0200 Subject: [PATCH] Some selected typos --- .../Regression/html/._Regression-bs102.html | 2 +- .../Regression/html/._Regression-bs104.html | 13 ++++++++++++- .../Regression/html/Regression-reveal.html | 15 +++++++++++++-- .../Regression/html/Regression-solarized.html | 15 +++++++++++++-- doc/pub/Regression/html/Regression.html | 15 +++++++++++++-- doc/pub/Regression/ipynb/Regression.ipynb | 14 ++++++++++++-- .../ipynb/ipynb-Regression-src.tar.gz | Bin 211 -> 212 bytes .../pdf/Regression-beamer-handouts2x3.pdf | Bin 638699 -> 638704 bytes doc/pub/Regression/pdf/Regression-beamer.pdf | Bin 570094 -> 570097 bytes doc/pub/Regression/pdf/Regression-minted.pdf | Bin 536124 -> 536119 bytes doc/src/Regression/Regression.do.txt | 15 +++++++++++++-- 11 files changed, 77 insertions(+), 12 deletions(-) diff --git a/doc/pub/Regression/html/._Regression-bs102.html b/doc/pub/Regression/html/._Regression-bs102.html index 2fcd8e032..bb82ed2a3 100644 --- a/doc/pub/Regression/html/._Regression-bs102.html +++ b/doc/pub/Regression/html/._Regression-bs102.html @@ -403,7 +403,7 @@ classes).

The goal is to predict the output classes from the design matrix \( X\in\mathbb{R}^{n\times p} \) -made of \( n \) samples, each of which bears \( p \) features. Of cours e the +made of \( n \) samples, each of which bears \( p \) features. The primary goal is to identify the classes to which new unseen samples belong. diff --git a/doc/pub/Regression/html/._Regression-bs104.html b/doc/pub/Regression/html/._Regression-bs104.html index b210240fd..f7dcffe23 100644 --- a/doc/pub/Regression/html/._Regression-bs104.html +++ b/doc/pub/Regression/html/._Regression-bs104.html @@ -396,7 +396,18 @@ MathJax.Hub.Config({

Some selected properties

-This function takes values on the entire real axis. In the case of logistic regression, however, the labels \( y_i \) are discrete variables. One simple way to get a discrete output is to have sign functions that map the output of a linear regressor to \( \{0,1\} \), \( f(s_i)= \) sign$(s_i) = 1$ if \( s_i\ge 0 \) and 0 if otherwise. Indeed, this is commonly known as the "perceptron" in the machine learning literature. This model is extremely simple, and it is favorable in many cases (e.g. noisy data) to have a ``soft" classifier that outputs the probability of a given category. For example, given \( \mathbf{x}_i \), the classifier outputs the probability of being in category \( m \). One such function is the logistic (or sigmoid) function: +This function takes values on the entire real axis. In the case of +logistic regression, however, the labels \( y_i \) are discrete +variables. One simple way to get a discrete output is to have sign +functions that map the output of a linear regressor to \( \{0,1\} \), +\( f(s_i)=sign(s_i)=1 \) if \( s_i\ge 0 \) and 0 if otherwise. Indeed, +this is commonly known as the "perceptron" in the machine learning +literature. This model is extremely simple, and it is favorable in +many cases (e.g. noisy data) to have a ``soft" classifier that outputs +the probability of a given category. For example, given +\( \mathbf{x}_i \), the classifier outputs the probability of being in +category \( m \). One such function is the logistic (or sigmoid) function: + $$ \begin{equation} f(s) = \frac{1}{1+\mathrm e^{-s}}. diff --git a/doc/pub/Regression/html/Regression-reveal.html b/doc/pub/Regression/html/Regression-reveal.html index ef7046053..49c3a7cb8 100644 --- a/doc/pub/Regression/html/Regression-reveal.html +++ b/doc/pub/Regression/html/Regression-reveal.html @@ -3488,7 +3488,7 @@ classes).

The goal is to predict the output classes from the design matrix \( X\in\mathbb{R}^{n\times p} \) -made of \( n \) samples, each of which bears \( p \) features. Of cours e the +made of \( n \) samples, each of which bears \( p \) features. The primary goal is to identify the classes to which new unseen samples belong. @@ -3517,7 +3517,18 @@ where we use the short-hand notation

Some selected properties

-This function takes values on the entire real axis. In the case of logistic regression, however, the labels \( y_i \) are discrete variables. One simple way to get a discrete output is to have sign functions that map the output of a linear regressor to \( \{0,1\} \), \( f(s_i)= \) sign$(s_i) = 1$ if \( s_i\ge 0 \) and 0 if otherwise. Indeed, this is commonly known as the "perceptron" in the machine learning literature. This model is extremely simple, and it is favorable in many cases (e.g. noisy data) to have a ``soft" classifier that outputs the probability of a given category. For example, given \( \mathbf{x}_i \), the classifier outputs the probability of being in category \( m \). One such function is the logistic (or sigmoid) function: +This function takes values on the entire real axis. In the case of +logistic regression, however, the labels \( y_i \) are discrete +variables. One simple way to get a discrete output is to have sign +functions that map the output of a linear regressor to \( \{0,1\} \), +\( f(s_i)=sign(s_i)=1 \) if \( s_i\ge 0 \) and 0 if otherwise. Indeed, +this is commonly known as the "perceptron" in the machine learning +literature. This model is extremely simple, and it is favorable in +many cases (e.g. noisy data) to have a ``soft" classifier that outputs +the probability of a given category. For example, given +\( \mathbf{x}_i \), the classifier outputs the probability of being in +category \( m \). One such function is the logistic (or sigmoid) function: +

 
$$ \begin{equation} diff --git a/doc/pub/Regression/html/Regression-solarized.html b/doc/pub/Regression/html/Regression-solarized.html index 42ef73561..c11e09b23 100644 --- a/doc/pub/Regression/html/Regression-solarized.html +++ b/doc/pub/Regression/html/Regression-solarized.html @@ -3431,7 +3431,7 @@ classes).

The goal is to predict the output classes from the design matrix \( X\in\mathbb{R}^{n\times p} \) -made of \( n \) samples, each of which bears \( p \) features. Of cours e the +made of \( n \) samples, each of which bears \( p \) features. The primary goal is to identify the classes to which new unseen samples belong. @@ -3458,7 +3458,18 @@ where we use the short-hand notation

Some selected properties

-This function takes values on the entire real axis. In the case of logistic regression, however, the labels \( y_i \) are discrete variables. One simple way to get a discrete output is to have sign functions that map the output of a linear regressor to \( \{0,1\} \), \( f(s_i)= \) sign$(s_i) = 1$ if \( s_i\ge 0 \) and 0 if otherwise. Indeed, this is commonly known as the "perceptron" in the machine learning literature. This model is extremely simple, and it is favorable in many cases (e.g. noisy data) to have a ``soft" classifier that outputs the probability of a given category. For example, given \( \mathbf{x}_i \), the classifier outputs the probability of being in category \( m \). One such function is the logistic (or sigmoid) function: +This function takes values on the entire real axis. In the case of +logistic regression, however, the labels \( y_i \) are discrete +variables. One simple way to get a discrete output is to have sign +functions that map the output of a linear regressor to \( \{0,1\} \), +\( f(s_i)=sign(s_i)=1 \) if \( s_i\ge 0 \) and 0 if otherwise. Indeed, +this is commonly known as the "perceptron" in the machine learning +literature. This model is extremely simple, and it is favorable in +many cases (e.g. noisy data) to have a ``soft" classifier that outputs +the probability of a given category. For example, given +\( \mathbf{x}_i \), the classifier outputs the probability of being in +category \( m \). One such function is the logistic (or sigmoid) function: + $$ \begin{equation} f(s) = \frac{1}{1+\mathrm e^{-s}}. diff --git a/doc/pub/Regression/html/Regression.html b/doc/pub/Regression/html/Regression.html index aae956a02..58412808d 100644 --- a/doc/pub/Regression/html/Regression.html +++ b/doc/pub/Regression/html/Regression.html @@ -3436,7 +3436,7 @@ classes).

The goal is to predict the output classes from the design matrix \( X\in\mathbb{R}^{n\times p} \) -made of \( n \) samples, each of which bears \( p \) features. Of cours e the +made of \( n \) samples, each of which bears \( p \) features. The primary goal is to identify the classes to which new unseen samples belong. @@ -3463,7 +3463,18 @@ where we use the short-hand notation

Some selected properties

-This function takes values on the entire real axis. In the case of logistic regression, however, the labels \( y_i \) are discrete variables. One simple way to get a discrete output is to have sign functions that map the output of a linear regressor to \( \{0,1\} \), \( f(s_i)= \) sign$(s_i) = 1$ if \( s_i\ge 0 \) and 0 if otherwise. Indeed, this is commonly known as the "perceptron" in the machine learning literature. This model is extremely simple, and it is favorable in many cases (e.g. noisy data) to have a ``soft" classifier that outputs the probability of a given category. For example, given \( \mathbf{x}_i \), the classifier outputs the probability of being in category \( m \). One such function is the logistic (or sigmoid) function: +This function takes values on the entire real axis. In the case of +logistic regression, however, the labels \( y_i \) are discrete +variables. One simple way to get a discrete output is to have sign +functions that map the output of a linear regressor to \( \{0,1\} \), +\( f(s_i)=sign(s_i)=1 \) if \( s_i\ge 0 \) and 0 if otherwise. Indeed, +this is commonly known as the "perceptron" in the machine learning +literature. This model is extremely simple, and it is favorable in +many cases (e.g. noisy data) to have a ``soft" classifier that outputs +the probability of a given category. For example, given +\( \mathbf{x}_i \), the classifier outputs the probability of being in +category \( m \). One such function is the logistic (or sigmoid) function: + $$ \begin{equation} f(s) = \frac{1}{1+\mathrm e^{-s}}. diff --git a/doc/pub/Regression/ipynb/Regression.ipynb b/doc/pub/Regression/ipynb/Regression.ipynb index 1615fc1bc..bc61ef14c 100644 --- a/doc/pub/Regression/ipynb/Regression.ipynb +++ b/doc/pub/Regression/ipynb/Regression.ipynb @@ -4435,7 +4435,7 @@ "\n", "The goal is to predict the\n", "output classes from the design matrix $X\\in\\mathbb{R}^{n\\times p}$\n", - "made of $n$ samples, each of which bears $p$ features. Of cours e the\n", + "made of $n$ samples, each of which bears $p$ features. The\n", "primary goal is to identify the classes to which new unseen samples\n", "belong.\n", "\n", @@ -4469,7 +4469,17 @@ "\n", "## Some selected properties\n", "\n", - "This function takes values on the entire real axis. In the case of logistic regression, however, the labels $y_i$ are discrete variables. One simple way to get a discrete output is to have sign functions that map the output of a linear regressor to $\\{0,1\\}$, $f(s_i)=$ sign$(s_i) = 1$ if $s_i\\ge 0$ and 0 if otherwise. Indeed, this is commonly known as the \"perceptron\" in the machine learning literature. This model is extremely simple, and it is favorable in many cases (e.g. noisy data) to have a ``soft\" classifier that outputs the probability of a given category. For example, given $\\mathbf{x}_i$, the classifier outputs the probability of being in category $m$. One such function is the logistic (or sigmoid) function:" + "This function takes values on the entire real axis. In the case of\n", + "logistic regression, however, the labels $y_i$ are discrete\n", + "variables. One simple way to get a discrete output is to have sign\n", + "functions that map the output of a linear regressor to $\\{0,1\\}$,\n", + "$f(s_i)=sign(s_i)=1$ if $s_i\\ge 0$ and 0 if otherwise. Indeed,\n", + "this is commonly known as the \"perceptron\" in the machine learning\n", + "literature. This model is extremely simple, and it is favorable in\n", + "many cases (e.g. noisy data) to have a ``soft\" classifier that outputs\n", + "the probability of a given category. For example, given\n", + "$\\mathbf{x}_i$, the classifier outputs the probability of being in\n", + "category $m$. One such function is the logistic (or sigmoid) function:" ] }, { diff --git a/doc/pub/Regression/ipynb/ipynb-Regression-src.tar.gz b/doc/pub/Regression/ipynb/ipynb-Regression-src.tar.gz index db1df18132a57a1d845b80d5269c353b9f15f0ef..883168687b6d3877759e4a86e517575071eb328d 100644 GIT binary patch literal 212 zcmb2|=3qFkJ3E?z`R)0$Sxkl^ZHd?8j_zzyH1;!o-K4imqoaa(VyE99&toc0#ouK` zZ-wrj9Q4a>cE7XTApz`qn#}F6w^|xthFBdv(yMn`^G^y40KH z(%Y=3zG&X|OVhnisr9WcseV2A>-}%XZk7E$Y~LUJ#Q1+l>GMf@bldN+>f5F4=6tB2 zy}R2?RlMfmuY>tZc+Kw4Q@!>1sH^rd{-j&io^DL;yKL7MYqC7Vr~bDB12XubZpU!W LRz;3MgMk46<>zCk literal 211 zcmb2|=3tnuGdr4r`R)0GSxkl^t%=v|j@C9Q8cSX-U8uK9L*yf?lJMLurW-w3p1+k9 zeY5Irr^o;N;vG)}-o>7oAAj#zsO6NCXEm1k+&-l_|IC{?t9G7AFAeryTbk{YEVb6P z_ppz@%iX-C^1owpq{3g?$DGjnb4xn=^;h}-F@6)X?K)E5Z{D+ye`&v?lTM!KY}uNGFl5PCLO*L{ijYdO z%aUEmzD0;+`{OvM`+4)cxQ?0QbAQk8d0ywW%ssPP0*hM$#eo2|l;Ld~M@1sw07aJO z0_wDfKodZe9;4@F0D)`AX6XI_$Wm9t@V@|2dRv5t=0HKhK#--c#V;NzNO8M>NA5K^ z%eJ!CAt>42Rk&%h9eV5*##@?i@Z?ajU#W{l@ZZArN4_5fy`bm9I$9~9TZ@0rb z_JK%HFg-}oMNs)SEwc1tTpKt~Ar{#9yO%TpB)Q!OPz3wP1OTA! zA<0o;02b_nj0W=ljY5zY697kAK!S*7fC!U=p8}LnsHn~+IWhytgHYL>SP}>$W3qu> z7Rr1|Bc$ygt$ROz?yC!NH18^6$If0JD-e%6jL57-oEon6eQ>1w2*g+ z^npz`av;(N0mR)ck02WtKr{~fy-(^&AyZ&Y=WEE;VJZ8X)mq36uqL=35`36zUxR9g zP=cL}zd!`Rn#^Gc2go_+`#9vVGy68Q3lQ|7WbgJSTdhL04{XT0MCfqVdymWMpph&G zUE0@|a6yq^hjm_P?qT`+nomMdDKOzX19dwre_vA}1+@bED4d6Sf};i%p+1LewD;3O z_E&=*JB->Vl{(Oj!>E07-vFu#)-0Jr^#MSJ2+M=;0Bd+dQ=&>yajz+-GGST@W{7F*Jt z(+MF4?fBTH*4K6NW^q{qGdA;hRF3M}x}$xm(U~)a4-(%Lp4xzu4W?8`8 z%3;G2(v_0`C4?^h32B3x&|x1RQ&elzG3HphS^nqM@8Tz>V-nOSC0$KqD<{W1j1!Rr zC2bhN_?6>Pmg_<)<}#vdS~y50!~JiH=J7gdeHD0ryILFP)M$Z+#LHK&4m6l>+k4e8 z_^|a#l#bW>Z^oeX3?>^!TfJZw77x6?N1&POVY2<0pL6KgSlaL5Ud&X6Xcmd^T#wvA zlDE_ajbLVY$B{@sKVO29p6=)5jqeBpcU>t{UhdekGV6TW4}+=#nzJ(CiiVM!h+`=9py# z_LkarDGrK(|NU-WT4|1UrXr|z+Gkg0IbwtHH%fneE94Uv`cY}!-Sg*}rq|_0uBRUx zs}?L(?R|V@W&>utdwHrlxH!js`n8f=mUO09*&9f$_|`g&)U`&;nooia%154mgeP(v z_pF{5t*1@-p>f&J*Wfy3YJ^4e{m+FCVfx=ZDe{L5K}vltfsIyt*s z{EGXJc6ina2sYOC(dfiOxlVy;FPW;$jj|TM?#yIUoMsE7kN_I%unlKn98q5lf2qS* z-+#84A$!bH@pT`P>ZZ-J@hK-^u@q?Aa(4aQ>7NK!LFejw7oIkan$I?QgvZik%VUO0 zu=%G`=gel3t>=uI=v)7@G{RK~_-t+0v1>$-Nk6!=)W%hAt*s?m2AT#^@GXQJggpvGK1Yl8C4&~>hIJ!7zxNH9Vax`odH=pY2PL zn-a4LgYy}f7RNJfQfvCq%+_Nb|5@2+qe_+MF_VkS7p}{m9m!5QtOg9%XPIz4D4hzg zDpK(fIzAC6q*JqfcdLGeJe%aHZExxzh3^%5^H2+ss%prm>87RtN7?pshQUW=SevhU z_zU@=13K;RU?WLbt{s<8HKxJZ-rhu$`dfGwjH#83mbPSE>d&y$2GN=(o<{DXb4tZy zYwKb+^~d~sU&_srR9P2ed6$ii=BwwzUQlp+RGh4q^k*u!?1d(s=$GznsFGZ4#e0l& zIn%6*)BuztX5-%~-+b?5qkcl4FG+W&qLRCcfDo!P(@Y8f5MqF)%Te*p>(FWejG}SO z%cX|6d3RlO47RCys+pNkvE0_U`AmZGc-pB2&e1ntbqm}?Ix)2VW?zi~fyw_m7vbsH zt%`HOOiDj$uAF%}#T_;<`fbfF1-~U`=Kg6R$k6IXvJWF6BkXQs=q2CxGn-zpwq^Nd zwXXWFVn%)Ng$n$MpuV)I6jerxgx2?QHi}*FXO<`mmLtMybS!1z_bIuk?nUMWSTOs> zrHd;+6IpFDyQP`)z_d{~Khe}TV?q`?LltpFB{q>Wu&*!fLM$0#(30UJSBWRsF}}T= z^-al1OHJ_NB~i6e!>0=F(=4Cq_4yrkKGJNb-i;RfgJ$X&q?wtGau7ad(K1^&ST`4H z-?fxHmv-avyA-kF8xl_Fy4c3s?i(3S<@c2DO2?CW&4ucIRL?!Mv!EwbN?-RIxgW|} z|H%3C7taC1K&@)2f^6c30>lT8FT{nhkQ1gb>ekXF{+h~9MVjZ^U$!ufM4jsE@wr98 zLhE7OK*4EgaCRdGLESn2zC@CAHRY=QkSzDtCnDYh^-u2mz-0}%CXbc@SCR__a|4d2 zBq*Eu)%;kJyP}xz=zZqLBF|w;qjDW8pX65Rg_4}rXQww1r7dBD+1Eb2h{0#I87BK( zLizevazX4>NnLVN_f`o+wD6qHs~HMnZNUFbTVR=iOCMp7r_-GjSmKc=^% z=B4~uE%6Uq;-0i9U&z38d--IKp-eFKgs0DEk^TCr&#t-VUCEuc#mJ(HZLh$VP8*>a zWUv1;%}9{#`qoUMo_*MM^Jy}^cv+Y<=~hRp|5jbK9(%;u%=JeK8~sL`evgQOi3cINc`UL9!W&>RGXD z39^<76(@YNveKD?=Cg-Cv^PtnOq)?)9zwbZIe$c;fbbFU}7=JKycTnyQs@*}oJ7{(XEtx-81T1iGxNJ_f?EQ@r!k>zF#&)T))lF{SAXi=`A1Sje*5xixP}~U6k9Ig+H2(Q` zwuhsmS%1G$nIb_&?}2rhLKo+0WO^Vo8o){;IfE6Ep$TGe*3WP|6lVdQqU244o&qcC zSIR8Wsx`GrawkddPAjLKeHVNbmVnN*&1w&)J4uyc5}1bvN_PPKL<01|Wq4cSCnX4p z!%k(mp+51=iLpeA!c?d8BjVW1NR|YfNRi}4ae|%$>|SV#VUh~x`)A5@x*-&V3yQV`Uu(D=8Fe_viG+XToH>VKS?9?K^=2vnAhc z)BZ$|I$dt|nL&G$vOPcdi%lfr9wvJvf4J*$Oe>X#g8B}WSOoogftn%p^LP(4ow)zT z%ac%sm;9*IPAjwE@M~=%&x+sk&Y?}-M{ga^Lp~38idz4`P&!*!K)ns8#{B~BEMzQf z>22ZU^+YE&@WHcLp8qb(F9_ioM)A>pNVG*&st5JpeEyXeI zTw^+wA~E-A&~^yTsd?=Pq`ZoW;kGpDO;I*0v7ltoCO7Jwd-X zDk~qfimOx?^Hm=HQ$zJzgUElA4TfuRbg!ah6(6%4f@}`c4L+gR9I*aS$`wFZf0C16 zqk2!M&a@1gNQm{O5)ELnh&--E5wRv^_Gsa_*!gS9!(j$Zy3RJiugrfs4c8GuvlWh{ z<*j)O6c6E<=CECkPJGkfPcPK=a^rF*W!g7TZsAsPQC=&5?l`~rqhO=V0<4NxQehZ$Q3RvGkoI|85{CG`D*Kk zn^<1wRL&W#H$}b=dlA%%VZ#D7@Se`S~&m7c0~JZ3TLJ-Z|cW!*o^afb7%HcU!ekajjG_)@UgA7&s8UN z`ai$ITatQv+7YX4gl5J$s=^-5pX*uyyqhoApSL$ww@O*NQn^V`Fbq@U(%)|;g#|7Z zC`iU`-4)*UuM9|5*AU%inX0{L+nCZ}@+r8lEPn}Y|qDKEe#^@IP delta 5440 zcmai0cRZEv|MooRm|rq7GmkA~@4ZK4k2oPR%IH`{%HdE}h;U?OCMtl_n!pm7#Jl>41!BR*l|@PUZ(H)1My#&Hxjg=&11#`#0W zfrdUF0M5Y>8gp)xS z{r9z?VU?45pc2WfUML{=bOs7Yl+8l(P^`fcLZa+6=u$uc*KSKFW450U3W$zGLTj0r zY=AP}&-*}AM=s|;0j@0(C=jr}3CR?fNC@18P;g2VKmrumqy@Mkk#Wjr04^ve0s&Y; z(#65}0W6f$AOi418izY22ed#{yD9*1vO&1REqwR)V+3x=7%>0400hqYBG3WlSULdR zu)ms(z@5JWC_+^w9>Ct;X9=;J_95OMMtdu-7t z2P&VOgMScSP4E*`#bpTkL6JHOa1&qT(FK4Tz62)VosW@tSFj7OIz})kP!P)L4+KY` zoIVWrpA*j<4Zeb(?Ax)=dK~x^IwuWD;Awmn$DExk@H5nTrx29JtB$TloaGBp7vJ<_ zq`4gY3gz^_0r&Bh9&@%EK^G_|tQExIV;ysZdO#S|+42*}3+3cZfQ%>3xI+~OzBR`d zB&*H|q!~WriV6!eEb8zc^ukweN`2 z$`6w~DG7JTNx<;1bnMIaa2!uEQ)C~fe^mupU(v1XHE>D#8BFC?#~*=Q}j_NS!H zfD5aVwO9YyV=YBAEJz#buQlFQ#;zg_+Ea{uYO^0EPj{g1yl2%9|KbqNsW{hA#hc8N z-rRLP;?K?H`(k~J?pa9E2RXcYbphN}XhWMG3?}kEr%`OU^#U@6dz$t&dalgh%Ug$@ z@M`=m>kak}y@}vpmSGQR1_rA@&52~jF^a3fT6>CNm{3=&d$_N5DjXrj_l@E*_|#>g z4#<UA&|`-|mhr+WM`1ypRu5Xk02zFmqf$?ggC-NgkC6Jde)nOjtM*Sq^=7 z$9Qyl!6icVHD^3rv)%HDD0dO1We;Z^`yKI$2#bcM0%22Imho2hA7nR+ zg@T7ir7a>0IxgliQAQ2+@RI(f`Wki}JNhl)IWNfrw zgEkry-95%Y%=9oUw~^+@gQ^@tCoPUI=dNVa>bS!BndCIn2KAid(OGd-x|zoGd#*a2 zVt*!lM~%x=5DYKg7IoZ?qr10QIOeAuK+df5u0iv8*xSrlc&+mt?tyaN2r9zFn;|mK zLKaaw)fZpezVix{NLM7?zr8<;v}BaVF74D6Ce^9RMYBfB`K%P>GFhjV3-P1q429zT ziI9R0LM(iOBdz|P?&yIon`TcPQ{%Cljn&V#JegKAlLnPz3xZUtqIJvXWEVr${S-6y zpU6nIZB$`RZCyv-Wqz1-O8%V@^H6>@*>`0_sL6W0+UT7M?eArS_`448B1s~Xz80-L z=_KG*Gm+Bed2Eue_NXqM(l@ZNQCo_XsB&F)D?ce+j)uEgA@3!6)A7Ppxal_V$ve2g zd!u9VwRhRK3Y?$Ha+%k+MnfT)0vBT1(~OnkXmZjB6u-jzmX=mCXLoBt)AjknSaw#^ z5PBEW*!iM$9`Vd4B^_mBUF<4)Pqf;1b34->rF&-a*=Sn~Unc&+uwC>GlVe|)TUm*s z`uOFMNXC`Dvf;{Q=8g5P`x3OPdgIEvchmD?N9!jS0|tMCRzfRStam-^G&aT_W{Ds7 zqxm6rfUQ91kl7jL2s{L)xv{&E02!J}seK`B+SWYUF8W5;b`M7vsKhPj=KPsHw^cjw z=Z{s1+ikamZ~z7rE_Lr42>8|2IxWM6iHppZVJ#Am8b3E|$b%ur#r9A{7rlk9+g7dpBh zD^1~}LgBKFRu;MT^VbItqnfHf|4Xh5L-d%Si0}5Z%6xg_ zmInV?NX;N&Nlnptzc+M-HHv1W(u;AyW9efjP<1+sotv#7;;eVe03GH8$B znv2;^oK0u*fV?%vAx(Krb1r=}O1puPO3bqn(Nd-3>w7k6IHP|K=n44x-U9o@g0yXTT0&&NB*;}j6q>wD`Ak^<|>ou9*EyTwV7dI^E% z7C}OLIYJHhyw?K18%)9~1#37Sn$fzA9aur^n`DqORC)Yq{PdSej4~ z+!*bUwF^jj>v=yWSf}D$kpY%_)A^1SVH)yzcavFm+16|?`5!8ugFBT)&Oa5$GKp2# zKaLzE@)YhXTW`(3y7;BR4SlO4V))0v&iImQVQi|B>0^T6ryVW7hJVkZ(w3du)qK8} z`2}8JqSuW5>KB`H&>a!0+PPBk>%)twOYqf#ixP!!cl$nVI}b9A9ajYJlZb((Mc z2IThZTe7{2`9f`@piREgmF(G;|-FN4Tj@7~LH8elJW@ zYfhkQ=n>UtoJ-~_+?Vwf+`035Gif#F3 zgUs%_?WHn4YR!?T@a|QmTPt4F$B^A_ye1d^>gOz?{wH{4URLqizV%l_l=Cz5Qe3;o z8g-EK_0MIL(tEetF3o5?h&`hu;JE^O^?hskGD#Ah#i_w%uIYDv66onnQuq4es+*^R zg3>2b8YT;b6#FNf0yq_6!>L8& z>@K5l5vSSYxuTC=TZ7g9)(Op7W^?C@9T;zSUf)iu7xxVh2M-e%t}()%G&Ly;;c)CdY>=W z5>}Lq`m7Fm+s)bQXylMdxH+mZO=~vSo0R_2mmcyWc^tUINB9*y-_23T(a%e5f1{=H z>H`B#!f_?<ya*$w!>)kysIF2Tj_CI9@wO&DJP>NP%=Spnt|uNy#1KBk zGWH~t%rLj*v3~b>rkIVTEbP;Z&P=i3&z`Pm6L}ouD;q&vi_TjjdYYRSrZ^I`HRCQ{ zx$iKqY)l*zbGN?kH-=JXTX4jCw)vks_m}&B?p!|)*nhr;+*CtWR$W$0Q$tKnR8mw_ zQ%+V*Q%q7rT~t$DR9i}1Mp~Ie@&Bbxi_Hy-e;}8>*l*0C|!1>WU$w40GTC^=+ z$7)8$UyKrtloa%UFOmrtVK&)Gl@1qe^x3Ve3toGSn10J6i_*Mrn&2j=_Do&v!FxuY z>E0H&6p=kuBK<*+8>WSk%i*;c+`mUPS5?w@Z7=R%e?8OpV`rwE#Lt=*#y6h#y-Bd5 zxMtlt>eB(IBasA84k$X&e~449NQIQIusEs5kc7KnsO3^)G89)`FA$7s`7=z5e$S7l zI@biZke|M6txd=KmVC5`ms)__%6MGw*ZW`9Yx-zQ_AAM^u5&PNV8?M5S7@)7u>|P>B%GScPQCgmleT zKWf%|lv>8x<{3+Hr!@p^WgB{@FchZ@+?(uo9JhRv+KP5`^Jn^eW4ocGu&>T-6)ReJ zewNcSB9&6P#hgzkq}Mf60voc*M%2wVw8$Klnwq0l8>f8lGfidplG!##eydp%%|t2H z^;A!XwmpVx97~b6z9>4j1bjPJLhy$LR&olE@tT?l;@X{XaZFA{vu`g}T1uFY8A*x_ zanrCgM{X8|8f!hNd(W9FJn`RG{th!IO<0jgb?8TCdg>b84~F7M@O>^W7hilMQMoZ<|gWtDEEh-+xFCq@{}>()UZ1^AX`h8zR<(3 zj!YYzarx1zZ*p@QF%_>9`s`t7dX;Y6C{e!ffP4kH)!c1Tp9C+CJt+{RHh%yVoa%s5j;RH_4)Z zt&Os`Z%wGSy;z1WW<$w!Cy*CK&FXWm9z^aDRt@GhEfh)%bq|GUBhf0NbgVwUIG}tf z1e4dWPbO?hiH?bss8iLjQ+g(?^RaRX{bD8seUI8gFMF51#~OJ53Uzawp7n{HjMiA9 z$DI660?L^S)yyveqnY|CFL(HE{57%V9+#+5wxQAtV@fjx^RV2PBUitt{n1?Tw~RGW zO1R+UOR`mg{IqDk>A_5gTTQxwG>ZA-x;T-->v?rd+8#ZI{nLzz>jbo?+*b;SxY$H) zcbOTV8kh%^*rx8DnJQ&qY@Yn+%cQh*L7u>Tn_!uLds`Y%K&4Cw^6LrglQEKjkMlGH zh9;$E1cS+SpA!+aglhb<-woE``#pU*o2N;4itApG3|OxmCh8mJh0x8{o6-=HLCuhE auu}=h`(j(`NZHB6Hdq#Hp*^5_x-=`KOKq`O1v@p-;~ z!1ssO%zI`(JG=9lo!y;1*=9K3W`IW`H=x+lA@hGq)&Rru3^WpRcF!e*8iM@&Dc?iP+ zIpwY&N9R_^fQxx=`&ryD4xx@N?IPPU4v99sf4b7MIL-X;7;(M6a!n6(9DnL;BN-AW z;=*B4i8HxT$)#4*C6ptaq4^BQR%F27)txHFj^2cwFMqnW?#;i$V?AF-7}a6NxE0?_IGV1JHW?-eiDpd-Ml+8lnDIJmGixy0^lH5k z(n5i~JUsxcjL6y4HG-wIRSSx!#t8bTAb9#yC>(bTz|=NXE(Y!pMRRs*4oe}Idaay( zp}m~zgsHyGUyW1ASu6nL&w`=rn zhnrt(#_7ejQ`&F&;!=L=5k1RX`R9v~Mtpz~G0^Ft;%(y(k4h6pCucXN1W`e?9ooi} ztgLXz%T)wo-zwLKrLx^7(}6YjRQA72Kv8rx(S`OmV*c!;T|#HZ+%HyVo*LNutCWOm zms2lZv9cgs-&k^hk*P-rl9w7JW}CuzQc`Ziz3b($Cx3Em?=euoQ`JsHt*c!7Vr28M9+663kc|G3RkC z6>bc$4(TahHdb+a!bzJ6$PXDvYcuaYcj0 zza5AiO)vzxNEUx{|64cnjF>6E5W6;i-C*nu)0|oR`hujDhSl!pl854x5lP=2VPMeC zu3=Rz6U8^#df32gy;LQEzu#%0oh1oD)oKHNwG{#?BKJWIC0>E*TNw7ruJgZ@)fR{~Cg^&)R)~ zQFB!)uL&;({9xjfYaRSbmsuyNkIb@LF>^*(==#()(}5R$@k1%Y)Hm^IWB%a@-|yVpyo z7fKlmmo`6>K>=_Ev&aC#3T)ZvC|amsOvTk*#N!e{e9vX4+S3Ya|reYqjLfm(Yz7q7nciqs1=-Ios6`J{B9JJ9|@m4OgXa-Fh z6k0o6IYqQC4RX&VP&J}0p>iXx2ZlLZ`H&*B?YvsdY`-PI{9IZ%=`Eb%_exBN(vh=+ z+wJ<12kopK7I}V94?D_g8fc|g-K@g$R2-Wds4|lzb}J-p7dSWJ4EKdqo}5VLq)XYw z)cL%ZP*XqBP@>!jpPF>ez8JbwTEO#{}4u}46aBg%sZuVM$vV&e2eir#qzRb@kU`RSqwfT zi?CF3i$^Kf>@iUWMazNU2erMg6Q1|9Jk$C(KH6Jp)IZAfZ{R&9C^vi;Vmy*7YnW9O!?t0<0-`^r&FJ%cR#W1@zidkWc#NH&$8`g1c=yU^() zWpOFKdR!n7nIem$(>B02j@p9c?gDBhs#?2Lhpi#-J^t~0YVVjgnCIDK_Igr<88y(m zW=zUvLWAkI zD-m1*7kqPwLUnH|2g@GhWFPItc$$bB`!86XtMIf9o~HfR6`37pKWX1cxa*}xC$?UO zu+8dZr&u((&8Q&4bnh7Vg!5=!bJZRwK5As7c?#Q%5bsrt4_gE*Z7&)8{3iNPT;?M# zomaq|+gii)cCBgo7r}(~c&Q>EZ85B+xlJAVHAq{sno^tTHHTe&z}TL{ z<_iXQg57s)hzFOO+f{0PcC)CRv( zi(1af@6s5nPwP(mY&Dh?atP~&%D;^V?!pbzSgKQsb@ALN0FoLBPkBG2H@}p0BwG7M z^0J~X2+U62%z`Aj`xfK3AAgbWirW!3X~ol2@l>9y1syyD?+NNq#$REJ5Rz&J2bK@- z|BdYbJiLi}mM{F%IZ!YnuvgPyj`3F|rOzq8`AI}6aCy-zJ{Q_Mvk0TFODbdo^UPvt zIqObAjDvJ9$pl|YIa^6yR!uE~G=E_zq>LM9gAEh3Y6EI%8YnemjteV#ak632uV!kCRSC8yFrx>)Cv8?MJm+;P9z~!X5%es(_FuaYR*>q|oY01SFcm?rmy9?|hnk$N4kYRCs zU!Z$RfEMzO!x%P#DehghEAOGLSQ6b`ZUDPt3;hLqZ$-au6nlqF>%>j^vj9&|uF?U2 zN_jd?C5`mR6ZnH9-XQdQ+ts`no$FmAmJ9-90Yy z%+H?ARcSot^RypLwmcJyLg(v$8Qni{41SZp8dP_8Scau=9C{q9bTRK^&gyZwKr`aJvaH{F>sc;K>L!`E75N!_sUk|h&*hSr zIAg^8iDbgWzrZ*xqUr&lVGI?`kQdnb3IjO2337|Ya_9%by-2xct$U%zNwHGzzcAME zi!=inC1Jo!pqx{5dJDb~HAIx-gBGr)xa0ht^YP#4-}VO5&r^!_uE-*{;pYirg+tH8 z{khq*_mdCC8J<#T)wt8nUa&9lc zp|9eK`0BH67Vh&VY>#PgrLqQAB^P{Gn&`*LqG|Wg2J&8M>=59}?)K3i_;V;h1h8*E zUuD(r3OF&|#>LU@YWIvZs)Jh6*hhqYmG6+dFtgkNAldpz#-thtuJ~4B7tjhD*Q}aU zC9J4S@1tS-IVeNgv%A2nfc~R_2x`_dX};emQpK5hJjXDuO%|d$^?(pCGI&R=@Tb+; zFlJG;Ms^>Lk!5NpvB>)vRdVKtTL^s;Ia!a91Vn_9l&_LocvAOPg{0}b`wov*@M(=v zDABx4Dry9k27NO*ASvb_D?A_%)Q8uh;E-sSpVj2;7(GpqY^mg zGMR!cm`=_sly0?`V0JY6o#52>x*j=t9_0U#lz4fR(yYFUgTW4B_3cYRQin zY3Wr7p0rb$5(-8|41fR4KofK)^{UVCmYD)Oh!dqt!52QM(cSO=`%xR65T-fDQRvTj zrZ#6ABn_n=r+St-Xk3a-TV|6j%#nl(;rkgN{Ur#bfEV$7GqNXa17OElZv}~~=;m{g zlEgC1H|8cQN^E|Q-+W`BkC5M(xYGt#_iE!c!5>(q-89f40B8$-@ab46N{jA>)@yD9 zLiW94>ikUOI)j$|M&yn6x;9aQ#`Op0+DT^7#nkSEQMBRgLIdmR`oynbo3)dXK+d%Z zza}&HgL8ze+Gx>vIQ%^iWyH_gKJt@EkDhWInx4`C7MG?Rl<95` zj!itZ5{%PiZ8*E{8cmoUZ-D*R{zUdPqENHeW}(X8;TVBP<6lTud6;ji(zJ@LBIIv& zS1OhcEWB*86F%_WulPK}rcbGOTe>OpKXlPHU{YV(1AK4I8V4EiYlRw;-#ZxZK{)mf zM9viJd7_%^T!za$0CTe{jwUt}IKXLMr8-Ya!Z0h_q$wi3<=JKm^Dnmd<|@A=;)T_` ze;}!3%%Yh3NMKV3Crs@p)&GpxXs#DTqxst~)kXPnbiD%c&eY2o=2SN}E%a7JHBu&m z6lny0!rw&K$Vn}=W4}m)BE0CUp0}QVXhNzoN#`JvTPlqR_OMO=_4wiIobKza&t zJ2hwc=L26u9?Q;((P;RrfQ8UXr}u?pNoY&RcoVdqr`m_>?ZCpIN&6@Va5wb&uo5x8 z0Kv&mhXnc^SjH2={4Aq(y+^Ohm#mjE#u^v*De@8svk*{=(pW;Hy@Bd4R- zlv|YDajPyRakaXWgnNMy%DFQM?&d3Iu?$Fz*(t>j?0Y>dlQ*K})50S3)+`@5{no71 z@g)hn(YmoFaC*)!k9=Of3vU$O=Ftbg>#u1q`sIH6$rbDeL+y{{fAs9E@y#A*lM+K7 zRWmUqJw~Y#{m`c0hA$wN;ExUboHE|tGk!D@70=7(cL~Ao@7Dz}s#iv4HZC4nyp98u zcbKFwe3I+Qk|x8r?+9SW!I0!WdyM+slum#4yn3nZrd=7Mk+qdtMIzWQ@NSx})Mm_w z*z%a4VkgiO=8^FGBSwYn2St?PoiETPg1Zw;N8O|Mz|LkJpuH7E2IH~2xJ6R%WnNx> z0@Wz!PpIgc#m!lwMy{XN_O-xG&PNTl9H|4YwUH+xdHSOJG>(@nS*L-Di1Z@bf`V%@ z`KY`5HSOviFoJzJ?9;|~j{_ z7&GH?!!2FpQS9gLCNY1b3{Z%F?A3E6P5*ihTN7~))&|_zExX7l&{=e@6r<*6#BYNg zHdUw{?w_j6!O4}v>vZFmkV^O(s4}fr!z`ob8u2m{n3Z>|WPf^imE`Q_s~Rf`UX+}S6w-_CIg zN9P~)PlL}{Ocb&l5p(DvBvD(zMLH7LRa|Ex0=X(?Dshv1YccE%_(1`0;>PaSR^01e z&#+~pAwtp(xYSbtYYA#UnH>9OVwBvBgH#*9J+flulhut4y1zS#I%HIWDHBdRvG4?8 z&cDrr<@!CNlL2QhNH~rtPXAlrqgS<1{(<@rw11%g1LGfn=2b0BBW*MR9zLNANFgTe z|5USiAtnf0M?ppeBFn2NC&MqVD4-}SD+Cb{geVG#@W?6(^UKKbNYaV_zfUZ%xz*pe zJNi1&@j@UOVJQDjEX?UpS85}Qs0p##zm^{7>GPW3`_zu!WxX4+1^JNayvEIcH~ZdzDV9r=xutY{^p!UD8$RN-oXc0SN;8C9Be za&ARK52n4ekc5E^jw>o-Apj>L6p`V@Ob5#Nh9KIK;RXAEufndah3Aol;8^`tNMF2} zV{g{@v^Co(a6FB1*|ykVPI7EABv4Qcq!L}wyODMwDb1S181-Ol?r}Lh5b;_d0;4sZ z8TeX;Xg3G;*%_77n|iIzm(vM7uh63Up`O$i{n)&L$E0=h{NI_?8dpb&D<>xZ(YFq_tMElz4=GnMvk!&m-roz`!^-3H0$U-?Dk zc}LUKT9c6jn&ODOmBcvaQCYe}d6@kVfTZ&{rQI-k9!CEYX5cG$#Aw%a@j7C;Zz+8| zoUw$}6wxk@ytBo{ErqaK7unY=*R!U{MF(w5-j>IU;f}*E*K?PRPD~I;q}|L2rrX&= zQgb9Sy>}%y-)jc+$jp1DCE=f zNMpKftp{@Ep4oP2@b7$5)&lT9~u|- zYv*rKeXk&xFc$%{oMd);Ecmig2)K%ngxhe9qw{Ba1~D^ zbrn!H$2_3)U!|eCwkC#0OlRc?jnhc%H(X^s zq~{|$b`u&WadC6^F^J&lUK6Xg0H@^!I6Hbh^pUS+|N1?a%)FZ7Adb-QU>}h~a249I z0+^>~-GK3&O${lJ8CF|;i+Y~AN?WO_Z z^UrJcM-`gP({sL06gPfw-V`7*AcL`Fh=_J;WT8mpWdcx+!L8Y^#$6D zgw;d1WZ-lBIW>4FTT5UDXAA%yIOAm&+ja zCHr42*wXj~nt@=Mw*Iq{8ar~%a0}@#*7EE+Yi48gIk>2TAgmECZ1erUJxk#lR$oYf zNQ+yNj{V#(WsD$gKoP3;YX3YZM_Oq2J>gqkiES*kKls2Hky_<5y2Ir#Kz+-0(B_0J z+Rtk(Mtw<;03g18>YtXe_K1wK%{@HA3Us2DUiMN~sHWZb zF&d_UFEc6()7Wzw2TY8MTn{g@*i)%#8_RVU&^-*&>Wtc|$_;N8bt;@wJ8qi{*bM~k zJ{wf%Uu~^bl^aZ%IoK9yD$q)~8+1Utb3m1*Rm-x81}>>V2U>=LthdurmCpJnC+&uU zVA_?2a+7;p7V{#7dq603)^vNV<&8nb1#O3I23s8_6D)Xw#*v|`2~;#eWGc3Sp0tI; znLoLS)lq-|YVDmwo@hEfl%Pz`U`gan9;x6K<`FGIiY7jZN`pYY-mFQ^RW++gfC?)E;u X-fwI&#_KT2umpLqnVA)|6tVvU*oB{; delta 7181 zcmai&fxGeYVALyu}b61!_jMr31ZA`KZngGKP)`%Bb_h3xym)CL`^N$*Q7Zd4*3>HquWTI`ptA|Vbs&E z{Si*B618m^#UQ>7Z)Zth9^9Xym3p|sey6qSDBS)kWa*m^wB@b2nYUsl8oeOvK_r=$ z9$_sTQ+%~I@JFy+1CFL%NXs+D(3XhQK( zt1gH^AG=3h7>=#TfbmMlOEK-)Ah6NEiYM#HC+kJ4smW1BSH9aH z>Xl3`qEw5H2xyPk={OcXORCgkF}v!_(GMDE)X>@(oAcNKvoV*bhq z^=kc4`kSYrFBLSZo(~(YcVfm39u6AG9#r8q71EQHaDn1#(*@$U<1&_RtX?=xxyRmZ zK^p=If$NzQ+VsvoEi32MyF{H^+%3mTw(XIQqPTQREA~d|iYi;uzLBX+pm;h^8E#sla_t^ zkC>8fGG4tDM`D-!?Yz`ZBL*3r90Y8-in$D;Btkr2dIwI-Hk>F^Z#r6iP`m1HJ#0H( z9)RjPdr7~FuzUH}S{brybosuobh)8{E9zsa__fjX6g|45mA!LyQbNW^NS0_?TCgzi z=AP;EhfjEgoSOV?oHPz^SvU?Gif&)t)%)Ni>btmY(dGSM?e_e1wC@Z)#tXW9P}NE) z&`kQ|cQO;Ov3D;aL_!_L`y2xUHZSo)^Z)A{ow4oj=bXGMjCgXHO4uh=1bcHQ&G1ny36)ex!`@ZiREP5?3IbKS=q-~L@65J609zq}SUpqPVRWBpuKudyF zY#Y&U^jk39g1XQtN8vkPbcaw{UA`ul4ErhA-L9hiuHvo^I>@fP3i^ED=bVm_kJ&;>SYX7mvb}DmR{!c4^=(i@r&X7TC4TM>SFWzJQOci#_(KE3Y)#5p zZe^d7dbL6MaJOy?v@C*SRevT8?fG@2{&Ea3b?s&GI~ctYN+l^6Rz1ej-l#tsz~#Sy zyJhqR{O6^5m3logEp)GuC(<3Ur|qT+kc|~MU4<67nWky2bz2YV3qN0p?aWe`l@Pf< zG7l_^=vQ?LV1Z+X)@8kq^B8+eEe_FqsXUV`+9o8a>uZp-1>gON{85v+5;jC&_ktM+ zjW^JL`@XGezbQMk>yU=jY-$JhrgD@Wm9{B>(|C;9rGDE8b~fSAGO3w))12(xX>E?~ zz}thRMlJSPUdPN<;Phe29M98b0$G%=F8T@we*LOh80wmMPtP9*Jv$7tZ_gz*Pd@CN z7329hBl|=jM9e@CeN3|XTz2#+p>hx)o$l~~Eewpg@J+lPC5tYO)T2zh-=D)1$KWKy z#RnrrYz0)&8rD{R8CrC&*<$9eV$)5?|5?R#*@>S|*8lDh@T$Ew^h;kaN%PUCA3;$! zkbp-NgOOEjQ~>3FVDy48R4}_14&It8z1VbbTd$YJZo`S9-Xw_Gs&jkZf@c;$XZH&~cMq$)3}@x&{;P&{Nj_e8fMaXR&| z(Jfme5-+unR;Fj~-g3iiaZ<~*vpVW4#I#h}!q?z<6v@3(OQbj{kSh%LkZKrRd7kt!*ehcY+TsE@`G?W-w|!TM`4&VLVA z9ni<17nfx~Pw3T*RL0IQx<#n}Lw>%6`P2178xMT{M*9BL`sy2f>U+Bt;>nJql(2M) zE9y!4OQ%;xUksN@xtWt5=>w4xDmKhOh_agStWSX~y*B&|78*Fy3 z75o}wAE94-sSf^GX^U*oq$qV$-26P*M7gc@5Gu7AS%#<@?MI?<-*Vz_%yP%-nyaPT z3!|$lQ+4(u-50BXKw;-lw6}DX3X$P=?TVPzcyK<;@Z*F9-I>nx?>D@nePpQ-u365G zcl$*78P-Qb&@ktC@4QzKWSL@;0hz(oiNr;yQbK>rCsfu(+wKe_ zV3N{Guq#VO!n%y(;JNd!gbsf!?kkxYboZkg#R-9Rry~NuSV=35ICDRWRj^{0s#>Dr z=0g@RlfD!OcyQH8usXvSlpQ#K5WgHQoBw8nzlg^=;D?x9Q;#k;Gj$P&R8*egls!FB z6!uZA=+iBn66AksVMPkYBYY{kbP0!0>~+D)?Tl7!^Z8R&CkpwLoP_K&*zwm($ElF7 zbB@qdMjs$+uOc<%1CkkWP1dY!Z-jdJe$7pqtp-kD5^d~;I(~T<3bQdz_gl_ZN(GTf?f3N65u8>jghb;vzrb0WjH#eE;D7H}>|T(oL-YNI z7BzPf@waEAP4@7n>^l7T!;z^wd?sS%H#tOjXT?(j39Ca98EZkF!ius1VtfRxj1KbJ zPGb1#41_5R7I^sHzU2paq8O-mSe!f7SgN%Njj}DAS z(>Kehnqg%pQcp?Btqn%>#klYJh#`N*kXc+yiaSZhehKX_E#N1+LeSX&Q!5~LYgc^XODP|w8A5g1Ey7jJJ&B##2Hx_1Rd!JZ%W+wPj-mS#&i;px+ zt)_wIh%8J_=eh|4(HutQ?xEKJYPYTbxokV0F-dd;u}C!2 zrB~y}_}F#Pd6`p=7gx)<4!mI-pSwMCYMn2X5qd{d$cwCQvyuzD-&O zcKpR)6|+&tt`hP7V+qSF*-7DI%DOK#0w;U>mmS}n>UD6m#p(ZL>e?1`RF0-ljVGfV zGJ_}}xjPMnS6Z}}TC^Gwh#OAyyw@_u2>0@15<$uPFg6v! zVNH9gDwhYNTXkCtJ3J)bNXf2J*feDylbzj*_fy^pSSraliVoA=Yo@uIa zE^sWJFNUAWR`B*yR2%zpvyIgBPdyWc^yBxEWZQVEa$sYg0Y;8}(dj1+Vo0twrE3Mo z4_235P0b#}{4FwHn70uDGH}_<1evvYYh3@`HY{z z1}EwCNP<^niZ|+R45y7kdTN$TDSx@hfF)3NbZ|T}3GaT5mQB^kFx>q;o^`vZZRb@V zmj8TzECTES{)uEut9X<0tN12Si9wCv&bu0u&FAlpC>KX$`8m(2NPT>%0*Z{u^9W?7!e>Apja&#WA@g`PnzL|SQA-2lTerb|^1Xk0O-jus5)ZcJvR)L3Y z$kI&Hrb!tQ&=QB?74s=~jB23fD&8Xcz9Oock6M3^GKYXSWq|bS2@eYvZLDSFYLvsf zwd=sk4&ZmTC*p2`KvH~TZduE@?M}Is_$vt57$gsSy<#Y;?vR$Bi?Ut9$@Rf6!XQ`P zoWmR|i+Wfr8eLm?XW7c((yo7qjQr~j6_$-nMapv)1KDvCw@YrkA@J;PPUVl^zpSYh z;K1&Gevu67BM1Oa0*i&%3|L~*A+$ddMh);Hs1>k(3zjJQ`xe^R4w-k6mcLnKX3-r_LG-QRn#&8-p7 zlWf7id%h*TmI!NPnDDH{&G%4`Y`cg73Kr_c+)r&zE!(KvPgz5}VUvDQV>hQls#K5E zYu;XQpBD?zvp?Ho0tWatr|Kl8acfbY($zvg5HBSJsahzr{`631xKgY;}UO+TBx)PR)g z!f}41gUD`dIcPVcoF>}fMF17WL~#w1qT^?e0TPq_gJrg^d%ONYqg0CC4~k zu#CbyhMBECHLW`h6bV&JFfUrMexcTPoHH@0a9N4{UBJ|hH|)O>6_X{^;=I+o;Yl|27-{0RIE+AL#$U_y=a&lqQz578<`GpGbyO0T%85tj)RrixpezwJ@)O zfUJTnpNx>A+$%+S8JXAo3Ig(Sd;*HHg2IY&l62z#?-6rs9yL362OmeeS0X8Ec^m+9 zx|^$k1NrdDQ(&H?rm4zG0avrut*ojlzFVLZ4zY@<3VTWOH)1v(E_4ALeoofr!kwkt+TwWzD3byv)!deW#PMzQlDEE4G1exE^{sWsjwz>fpvCA)tJYdEz?1UTYfh*5atnnh zv}8%R+a1Jl=NujzR)=;ETcfE(1~r)k`1H!a(!$hj_347@vw5f=*?Oc=_*icVTs{a^?pzH745><2xDoFbmxm= zA{~E>*A2q&bBAr>BPtcPdJ1%y=@Hfyf6#GMq}j{S>~zeBBDV^t^up5$ zRGi!GoCJ+P1`!k$dq~osk)qpt>Mf+z>5G+|HJEsoc#aqeIy)QbwP5uEyk@0kO=AT1 zFBpJl$>^BTdOtb+3<2qs$@ux?gkxO~aRo_Cuj1hZ_1d$D$j#zUzY$)t-4jA!V)TX! zT)N=2g6~G$N*5vTdbypTd!N4c63l}YdnI`pewek<2ltY_#u_TwaN(TCXMuxwA+8-s z;p}_oZj=uAzgXOP#`^V!;hdS^Hzk9Llt2ho4hn0pv8T%I2&1YdjNkJ4xN1gvx-}>V zToHuLvJ|)%Yplxihf-ga=bV zD;Y+3XSmrHvy1V7f?svv%Ogw)dv2JXrO&|Q$$-&+DR~uk?2iZFYtDfv(P#YO4^C#N(74x1RpOR^$^VAMtCDQ>ID!Dtq1GmG4Sn=c zlC+1ZX>52YUoB`hj$@eX-%t8)WNBYmaILD7)uFn`bQN~?7$Z>TRfc{PXlwV=avEV~ zp-|4XD`?Nq|7-f~%)Ee*cZEpKvY7f1ikD2JZWW(V{KI3m*A8Y^g zMNYr9^pL$PNA78J0QP@rJoKAyIumRh$gd)DEv2=e1aao|hQ1MP(DRl%dM=`heQ?{T zNiuo}NUe;H^Dq|2xQy_sr&S9FhkHOn2sKyl2GxwR1vQOU zRu?|iHdQKpuC%U#j`}=}Res}3p{(IB*q;7skjCbH#8st|c5{MS$38?l&xZ;na)&n=NDh#g2x?kTDgG@Mk^@hs}M=6p1(kH4%x zYjdd1v)#R@JPpUoVhY3rw%$Hay9`7l78yi90GpXv IK~oX?KSHI%cmMzZ diff --git a/doc/pub/Regression/pdf/Regression-minted.pdf b/doc/pub/Regression/pdf/Regression-minted.pdf index 0bd32443c475c36168d5a5b63822de472b7b2e61..e9cfe6a9714ea9f4285f30e0eef50bfe522c2d48 100644 GIT binary patch delta 9392 zcmajiQ*hp0^e^xjO&UJ2ZQEwksIhIk(J!{1*iPfdYHX*m(Z;rU`u@+HGjp!a{>*pw z&AQmLXU*QTI@U>B)=2{zA%7ht-4xOTI$x4@n$f#{*C;;?(8U`G$~oz`71JwH?YR)= zN>td9v|`BgYmw7XS~eva<~eYRq7%xluC&rCx}d}6UmZ#py_~>3+O%v}Zf!Yzyq&#G zYHmfCsMoWnJsrOrQ{f(9uRh9Gg)~>8xuj$T$(d_BgfF13yqWs7Ysj0m|6H8|x-EKG zehBew&bC%z%p%>NR{z-lMlPhjbHFW>v{d zF=-o#EW0Zj!2?^Aa2h9*@TElpfiBd0jmb3RS`MMe0+6p*ZTfqwtY`r{7s~cxIy&i9 zKCPtqDpu4+`QMlB$LusXu-tJg`sccc&y9rD&D@*XQl##AC(pZvhv6#h%lv`%Fme66 z{?I`>6?WSwtDg*o%JpXJLxo;tl~-*w@E+~Vyfj#YT!MF#1;>VEqR&ylhqB=n((|Wp zIh0r-F=LdS*j#JBq<#+V|AkFafu&kdj1*JqB(|-2w_!$LY`z1p=IK(_oTNq>{E*x& zfjxWqxxvQ59{5QYpF#!Bw<#62q373Jx?>i02vR4K#Bli&m_h#`?}zLVcFakT*K4!K zaQ#RwnCMv`M*YIZRg1t5xJOcV{N-V_esr4UA>)n({`_L`5K+9diHZ6djXbIILQr>K zU>V9(e9O;>T3KwkC%%%9q;kd{+p@%OP{A0Ng^I1Ae8_w0Zc;?9k#s(uMr`f$`{hY9 zXUk+C5GV7^tdoufQ9-=c=>8^YK9(L z!7GZ|nmq?Mi4~P18UG)H&v)tkf~vJ0^Oan5@a6ppwXB(Yu*z(o2ja@EByG5LP)g-$ zEF`X*Fb6w5yg;;Qn- z!iIm6EX5m4_D;A{=3{FqMd$-h*a(-^<3 zvMjWi-BM{9DGgS?VaJPJFF~ZepS^gB7+NFDMB$c%z%c{ zaTsfjZyksLSItMud1#^t!W5JwJLk23$?j=BdA08w`_inp$k#7C7#JmIFyJRlz%{jG zVl0}k&n#^uTLhJ@P*V%*l%~$#;B(v(koX>Fz+LgexiGV}X7= zRM68Pv=%LT5>J>cSvJ^DbY>RL%ac7aB!lbe+W)kLom_!7`pqYGGErip+i0#=RvR>p zC(KuW7Dc30@xF}Q#LG2Amn zgrKf@kQ66{!Y4yc(g~TPc$yZ9G3=Nknoxy0C*i9jiyA?aw7#f(_#EN0O)+pLY^h?_ z9+qu&LQ}BbJ6@fFN3ZA5;CKlfl-`zyw9{eMt%Fd8d!Hv}riO1uBdo5m0`AQLbWh)J zzHW<1ViPIQ`SH@yDbr+=6;G;V%B~oD+2>8PT)oo8*X0bej%Qd1$27(UvTO3AA?#P= zsA+i6f=IPWLw>mCNeH>U_+gfLfnUrGNT-Z(&-q0J{yN4OXBmUHZxV2)?!)MnpX3M# z_xar>7b^wTLV7n#v@~G7>_jVpS#vuXhV}*McLu!{ghWTiurg`w zPVU{$@B1un1!(Zr#*C&Jfi-!qOeGUt#Y#CN`N5^!*f-+Mw56Q@BA|lvE!9LLViH?jav7DN1%X+HDyy+c! zwO-9ctD(-(P-(yu6r1{3pWz=JL60oM~qKl3!mX&w6}et>dK zk3hMGN{_ZM`c5WQ>H2@Ip_vm~;S`{;r+V?LTHM5-&Qe%>x@A)SeZ#H!*dm9e=&x#T zsq~79<~tZc>0*AK?Nj52+X!&t9in9w(0PG4@N=h@q6Eu6Ww$`SB@);Tu+pJPwz=qu z4?A@4(ONFCX(R^~()|eiy_=+mh}CF(SMowvn@Wgm#Cesaitdc>;}Kf|c_T`PkGBmo z_#H#^cl*9k-=H|YJ`foz&XEF;6Fh&l>xpPXlj9yR=jOy=TyjamrrApvTzV1o4AWI+6UUbLN>X+|}YseK(Tzgi7U)a7_p6pZFK;9A3|h*L0$ zv0Me|p;HrxC5=@F|Iz{_)Pj_E-$rq{e~A7FIm03ICjISU+7sVv|$r@AXI>&1%0vl{iKAd`oxqnlN<2G zYl>}atXC(IF=h}bjaJzP0YajlDbj5m_vo5KHVybR8|sAR6cuSH6z*24XdY-1Ezm~v zZKZ|AG5=sxr@KC|%Rm@coBv`Pd(DFhf1*JeJE3BiS~7L>nt?(NC5#hG*o*XVOYZsH z0(VD;>tI}8O)_cqWTQwpaU6kdGhS5Al)QdTi(xPk60JT}&25SL*&IR5Ig)5cOO25Y{nTpvm>YljD);?S`6Z zTlP<}E=>e6G$O97g5J~PJXG_d=JvN~Jzq#NNbx^E3?h`sPm)}5t?z~& zLD@j=!niWr{rdz5OE*hY(kcUn1^7$HF+5Yep&~?VL<>Fs3TywDLxkr0SprC7rYSOu zHAJ0!K8<`y8QR)6$*e`sJi^LlM^z)*dW*|-6)s3ry~K$ZQ^hlIiAhN#gj(3^LPvKw~y}%Uv%H_675}6pg&D3Xi)tFRwW) z=>1tA&J%Nj%JhQlL{E$*IuhK`?;s}bS(x)(ez)Y_LiUQMIj=*~182JSj)XEU?^A|s zBo}C2&0B_0e<4QWeBq`d$u_7wVEnY^2%-?AFn0Mb_m)a?OT8_;TPd3gUtJK86PkpC z7XD|cF9>y|_ab*KlfgVWOO#*lz@J>vYkSTSR!%Za(norb9B)ac9_F$}7)2v6V+a{er{wi?08*yzg!ncn;`AmUo1>m`d{tD<>SUji6}Fc>8)G)2woULNc^@VRWK z3B@U%!eSY3R#Oi?1PYktIUka`v)@I(i+3tsv$%5k$D5Nu;7s(W(#6l%3-p=~-)r9Aj} zB}9cIMu5{b1^LY*x*i#h&h2`_NhEUACLMc6R(|6TaFHV?k)d^Jn`{Z?PRJf-Y7Q~ zprh5x=ekrxDBvVAtDdMjW?#iD1CM+-dzcjIfOq#$CZl3!IpGP{d@8Ppkuiq+W*FRw z3Q(kJU73C!cK&OFS1F>3JEaYF5k^^opNak!?Z{r%(HI|q>tFxdCXW%T1NDliQZcu;q(k@?cS)c| zVtT80TG+D^AiMKj?Qe})ZK5Jgf`4KzEur)J`6qb)#kmkn>Mnolzl23hfJn5{b%E2i zN*}u6A5q?;LoR=K&860AF}oGVLJN|1^~j~&*}yZeW5#6(tfu?&2BVFuhT0+fRE^i_ z^gibivR#$t!kvV{ZX(mhVKM;%8Rs&9F8E`5;3u+w3*9Z|CPOsu67yz*7^zy_QJ22) zCSPs^(-e_aG|C>`D*NYG0N4J=8Ha>Fo_cW@Y-vyTopE^Pt}SjcZ0GRVYWVcVd4xFx zJ@EPW(V@p!<28mi%{`>d1Asl$m z_U=V=S2zvS8`*+Z+RypL-(pf=k3jtt)L-(r+xCoH{JgJs_v6#o0beCT$CHSdwVe@+ zEpNo^BFVKMzjkP4l)3MC%?!sL#h9t}XhQ726Rxaq{)(u|#=EI$Fc!+Ejt^nz^!MOz zV;Ao1Jxam&p5F#=*PI73)wMXRJfqa3d}!*#w>{sZf}{$^AdV{B(cIGio& z0xgsvZ3BLJ#Cigb9}`+N`RtvW6Bvw+M^ezyEyN{nK%(i)3@cWyQDdCGTc#qmsq21Gb!pD{8@;p?foq z7awghiRN)gc7C8h)(A6!7oO3;sWHa$nx&O()M6;xU;ev7ArNU1a=u^pNg5xV!W`?7 z&;)U+M!v~!MD19t-s}(qHEtD-oLh_>&+rQ>+Xl8&K5tc+H=cWUR!2L+Q*@{0;P&vN zt2)jf3@tCmn|LtrKnEFxmvT7~{6KDF<2=AlKEo7kQpWDU61G{GtNOEEV~~7+mpp~Z zqeYrxVGvqq9GDJ?VT|QWRscsrCzM!=$3j|+p~Z4xm8Gc4*6T=-&ys%T_Yu;g#AK&T zGyAG^9luzFN%14ldXIA;1T5*yZJRY`1XVnh@l_hT_>j@5>go?`w9PO^PsydGb0!27 zIc+*BGM}t03|7KwhzfQ&mdX(Qr{qsqqUqxUQN4xw0RW5^Hlc04k18$kY*nvBSDcB{ zw&TsjSRHfY!>oDPp zFi63R0%8>^FLdjLRmVZy#)=!6>aZ&ND`g?_KNp--bR;rtvEqzbAveXj)+&3h7!(5O z9cT|0rAahK=0czqwS=mX-|?mPPB4vr=k-}D^qs=!rpOm~(|&OJYa`YhNqJF8X(&5` zIHqNLv_?Hl$|eRYi+KZ^NC>SIq}Zsz45B{ezy@Ii$8?>z7UQ%J`a_@g*Pf2XOp6Js zE9g3igNJG?TX#Igz8R^YFcO{8aRED|`p@t0AJK0g%zw8^-aq6jE-`9fdhi0F4sAS?+FeDWE6~VeBh0OCEB~6@Mu}I-lwH{jG4tuy`XQO}Rlzda= zCrKTDsu+z$b&2ehGTBz9>7S?i>A?!sj(v;IW-#1jSjbY&-oAr~+&vYMmY)wx4QjFQ ztTM%$OfpWE3HnAJW6b^j4U99CfSrn-&Y0f?pQuQK7$I^R0X_qNW_b>n%xaM^(sCaz zDe{b)2Hz(_5!hzsq%X#b!iHxzjlvUU4>>xVbf%=J*JwD@c zk-Fw4QznP1W!Fjg6jFOp4YhjkW^3nB6&Hp!p!|&>-~xDTJpo9=aF@9BwMEl zg_-TGEgr`gQ;WVh)t>m8k_ z%EzfuEq(lULT@2;zgBNI=bmAJ%km6vMCZ8fxnM}q@A0SUh42vZvnRaXZe-euoc*>K zLZN>WO7wntNxK_PzBQD=_;-&TPzjnN&DY4R4zqiF_oi5z-jsL3>(;gxzLG`+J3JQ= z1eEvzZx+4rjng{7gR9h`K;_xXVMqg6Df-NMY%RpiF^k&${k0IY^GFnJY}a1Fj`8Jd zZt{8$!wjKzS>y}i?@xt}mPAPv)H|F+%e8~Y!5EH+Ryjs+-XvVd3at7Hw&#R&G+ia} z$LaBD_rh$IGBVd)XuG28O7p*77=ha`hhpP>0&Q^zZ*H{E?$x?GrBxU5XC z9?h*_cqwsIAq+>N^xH7VNxA}2db=D|BVC!v0cDuMvhWg_y19|rY@ddFUHs~(d15Co^d1kF zZMeXTNUOFk+%&%QO}yPAn%nHb17xMdIDN5FpuCLn&s%!|TwMK@OgmhOX?~dOmA@yo z1D@XEdNJN2JvZ9%NKCr@iHCG~S%+5bUBe@+nyXyEag>bJS2O9FN_oo7=zs{T{qB;k1lu*l$}q?_AtyzQ)66=Ke*{)9hq7`3)^nrlPS; znSHdAmHj;SoH?AlPUz^9@D}{mn-O`Buuw_~SG{Sb@*%_B<&<#ox%BZAUe_fwX(lKP zQzg29F0M-zX8W$e`P&R-YM(IDEnRKtld?DC)pF;+stMWRAKPTeA59bPcXz=%8C9qG~M8utBstZQNuqv zE@F{~gt*5+gZCju5T$l;59&+>j;_RBxQOFGWRmB6Kfd>!g?UdHF=DKx+IX)I=JF^s z&FYma%lb-0MYAgu9qiZC8#vf0$%5nxlRycI0pW9k+c-91za-i2Lo%%(2f+@qLax+k z`1TsOnd7)}t#< z=i!8>htTyo+iM21kFWatENs z{W?>TWMpylgMXozem^)!?0Nz%(_(NFX)&jERl8;bvPK+s zb*z`o8=^sMGLK0S(jk5AdLwT`8qAIFXgd3y9f?p^({)YTbXF?H^xyaSnP-)&5!}C* z>ITo1Wp-zRP;w!;)MqADr&o{5oyV(cH)1L;oL#Xh7v^wAD?RDSn0O-4Y}#6*8S)HI z9iE&I7Gez`-`uQ@V0&Dxj2#)eYSFj2HJyVx%A(kz0@AW(KL7s z37R{eyjlCEx08bV$L!|m{`q(NO@pV;<2M?D+hEVU6?(I-RY#U7r98u2{nKJ;dV>qk ziw}K|r=-c2>XgwF5Q%k4%=iD54;V5Glv!VG}ddm;-#CT*pg`o9_0k%Da(S&iW5%Q{v9$ za5_1te}VoNn16x&7q}KWIryI|P@Jr3E93}d|Mr$wa)f&r6>&ax30^4B2TM0AQZ`O5jwG-jEkKnjXQ!bK3b=SG zea|^q3ski#U@{LA*A1J6Vv;~IXOy5b#i{3@{x|L8V=5(wM)DzHiF^rn7=cE zu!A7zCB@xl){Sfq{EG1-?uEU(p<5)6h^z+naj3Jg2*bkQc9)I%IkGvJ$O!5b%BQzA zq|G$9gWwbrzNa`Oy8CeevxR>Nvp{%>8~Vu--TC02*NgvJ^S+SGsaT2;tx9H^FvJGaM&ATseX@u*jM|2s%O zYd@=mP!`OH`TzE43EwN{q($kv0zkJR-`T5pFaI|)2X@)) zAJ7peTZEf1W(15`08g*FB%f$jG_0s$w3NjXMhf(*W)-N-dF4VCn!F5K`)RpmA&6fY z6D0v znhMRYS?nqW80h%vLBB#;pE!4p%)fMbKuj<0NY`Q^M`oX1YAZ&J1fuf~GPESgAYkWqNQ;|B66lQ*ed8X~MDjKmg_5U0Wa^;Njy(%}fg``b+T_47< z$7y>5|29c7`PN4!MJGjfM;2t2HnQO;h*y&=PNzu!0RaNqKy4tyM#J6rDw?9?!vEot ze}oSfsqy(_@UI3y0G0e_{zIlVZC8SE#u)52zC@n(pY6QaR8S45CQ$_WnUYx9S@s{= zD|rw&u^ZY>-q$fo^?$_ie@^&PCs#-J6#ox^PD&vM+N&g@_&*0<`qZ8q^b;n6 z(Bt>18axANVd)l6Q4_x(=c0buxZLz1Hm4MLFOtz76&5*DQLJA|^-fMZ-SCM2l-T*v zB4eKTb^sk8l~|d3Tc+pY$n{-~tSwwqup-&_f`nLI?!APASpCa;ZyND}?B4+3v|Qtp zBP+6x8ck>xs$fN0AI;YHdD76e3$J< jY3jjqq7wUC{m&oJ)>3JfX_(9iWC*OEsHmirq(A*HgvYR+f)7qY^m-Mpwh$|81(m%&bJ0OG26+_B1du!5TrDCG4;QUpghj) z?d%2I=~&j}{;E>aeMDm#;_tYLZx~eoPIFS|R!wGVS$l$v-gCFFr?=WPC^vY05S7ONwIie5&t9nyRTek_(=>4cs7w)V-y{1WB6Wv_fIKOd@T(4bPy*9B3{ znilI!j|Yh(puyr6wtn}Y^!o31hO6|>rXs?xY~0=1*-u%b^N9-Qf22V9=k?6?FV_!B zY$jom<+~gWk|8>|L{xbiI`&x{_AprskJ>1k5%pOVaMj*MXU*}ISiIf2(RL2woe#x9 zrhiMz>DFUO-mhMYuOiFRK>B}#Pdk4zc35py@2VS$TkaHXUM07R`5e|mNu)MD1GKDlT19V$hMR<8oAeMcnILLB=KB@vdak6I8a6tD*s5?aT*|c z73yLYUq{JZoTcUxurpi_iOLw~%A?HCe!r2X6Rb3CAB-ns23Y zRjgg*w3}Y{Wrc;=`)!h8Se2zC1dauC{%yAt9n<+~RZLHNMQcNdSgXp# za^>06KaMYv;xM&d_-iASycoX@P8&}h_YX5%w&ph)|90@)vbK_Tc9oIQ7QmL>(ScYE zn5|mTSN(uQY49-CTAwL$9SU?X9+3$RpACpglX}r~XJ#1m`c<^8W)|Er(eC>Vgb(vt zeR&8muJYCg2U;*!Yy5!yN+62%v;N!{>x_k1 z!Y`+|QIO}nB8qzlbBXTnHA}zIN8%%ZP+={IpWO7IK+%}PIWSQoByR$QJGDGbH@p*~ zR9ozfog+zC#Sb%%HBL+EeIV!+Xnxz8!}kyuT9`j%_Mc5P=xXyIIu{Od>eTk4li3hP zdG@b@{kq4*qrnw6oDXoEv}#q9wMrc9DJT8dqLN&=^scX*uacgr*&ad7H!%#X(X4W+cVm^j=S?Qh5D zH^npVrx&Cv&>}=P3RhL<3K`2^e{7~_Dm!vSxgb-rT3Voee(SwlUyeM9Mb(B(;S?Qu z0kxIifZwR8ZAe7=m}b<2KIDd~c2W}qx0X#8GO9`C7%UDysdS{8ElpSY4g4oC?%247%L?5@dW{a|D541~yN{$ty*I^_z#Z$){)#DQS4=eX$e+{)7|<-P}mUDZu~GG486*Y9H*(k&}+saL>w z+EhcYu^NO~jVionkX2Y##_JpVNhDUrhPxf#7V%u{0&rhto0_$`T*$SFLo^HT1kF{D zG*|+VTxh?*rI$f$ge5Of->Vs~1k`H*1!~+B9yQXkgDEtgr?59V{=&grm2l!!w`p*O zh&xSzUG>=kwGda9>yhexQSA6P^CHjZ2DR-_S;$;dIU58PU$X_%AO4EAJXGTGK1LAE zzf2MnHw*HvtZoT&BZu`xPu6k*KrayT6b}PaBwbW#^O!|mmeP}A{MeOJ{&0S#4hA$I(C;f`Z23pSMKSZms>qTQQ7dy7@#XE zOhTK+990{% zVD|I$CxB)LZ?AGgOdHN5t0~>b$i*{>cWj|KQcB^}ph;&B2pu|R(GDMn2VMtd<4>QO znhN3J?4-n`d&(l*wKh7l$#CWOw(FOKOGb*3+Y+HE%*gdt0P}+_pb*!;NR=q&<=DHP z&J?CjiP~4G8pOme3rQtjnarJbaev`ivpIEr;>0KWib8a1jRe0#IZqBb!(a_m@1pBe z&rvd<6vFR@CJ#vRPnZ1g-QK_H^oWXdsB+JmTn8Lg1qumFbsxBUYK`$pmsN@RoF5dZ>)fvq#6dSyFh(2f!|z19JL63vzE_( zL?OuLs%OL$q~s7@N`h=k7lNeZXdoC7LGhJ~b6(^MmFsu)T`fLOaVM$bh(SCayvx5G zFNA=y0vTeYYkm#Yd$A@W<`FHFYiLqF(B<^{v52(xmku>Y*raATdEIS5U{Gb3hZ_+t zcUag|B>ikL)#W1KNmmtQczCMOsmfJYkb)|EDe!(7%lphVcUsN#PZzEc z=U|&C7WQnvqtv<0{hyFo{IAIH$+;pBA|=M*|n zHn3JIor#$1coDZCNeu}dt|ParLDT?r6RYEz%tr%fIH##BU}7zY(y9mX z*mms&#}2#gQelL2mHDKmdJCPOBdixz-)O>{Az&uhBT7nn2!JQi6 zczZ!W2yYc6PY=VT#K;HeIcOU5mMON~HLC6E`z}VR2=EbiS8_*@B@!XVx^`e~C4{_I zCmg#p5-n&6!mzRiY`25)@8{<-%M$)J#j!s=+ehpd-?;^59$7=UC1 z**lZSy1mbua*%mr_&5!ag)|0R%9=o5j3#+u35ALpE%9Mv#8CgZ7|by(5UlXC_VbZH zkNNWz1E2ij@4ePQwW_YJ_T!TU9M_XYBszEkrEDsZtajonzQedXaqfsaq0z8AY45`y zLpZWurc*FSssXbavBJ7y)V(!c^fFI&AmZBFMJ45w9~zpTVQo6q)zaAti`^VyC>-)C zqzO)JxKdnj7U4UnLv7j{1(*?Nlh zp^QKWnPy;)yx(%F2he49o1gwhoQj!A!g{=jnwZ+CBaC2KM@eB*zx44 zQ>!Hz27^;;Ib%~AQvl(iHww&(1*QX>cvmS*$D%6()5V}nB*QRb)EFhr1I^L!WQz@54qF7DiJxPH?#S`h z^@rxU+gIF1ODx%;m0yW?c+RRH_Z9FyBEFjKC20%7Q_NhF$C-EdmIuTW#P9T)o<@9@ zqB2=6M|SC~ag=uxSytV$te~dD3BrVn@EWVy4>@yw8|kTMeruv_d@BwAWAnS`m9dqd z7Z+D&DytCF0WCY%-h3I&oKH_|VpjtJeS5#Xn=O)Dn|*yfyU6lIh@*Ep=SHdSj<-|U(2+F_2G8k8_2V@AD@ik^?#ZXDodD>9x*%#4covGV%>Z|I z&lkTAvfUqY{XPN?AKkCEVl%_f7kdrf_=g+6r6Pj5_Kbf!7geggS>D-s`#&80Jo}yg zlBaKa9mK!a-}z$e69!3R%cJ|KlfAJkr@!DQ{hV)VVj{2Y+-XzD=A8|G{r)qyCWp+6 zuXeh3=gs$S<=EX9odeV_eNnOKSWmt{$dR!i&J!s(l@hxWS8N`B^^mxCT{iX8be8SY z|L$#ri?`7?s9S8Hd%Jh{?P^8$>2sv#{p00t0bfqb-Fpw&0GVpT?sV{;Dd_I^b;n6X zM@~aW#q}(kH)zArnRSkBj`uZZ$LTJ}pGkAz`QRMeLbzw6e*hG&)PCK-_U+{P*Ix<5 zy?)U?fq~t<5B2di@%opew!v)O5VoAPFFQ|M!!40uL_ZN>A^gAd7ZhEgE{VgWAH@Pv zBC2T8srUEfTZ#1Ns?b0Y7~lte@ZUu-6;n|4jV`C2d0=|=XIJD4(?J$Do zK#=VLaaekRG(YHBjsKSZAd1C>JlEr%aA2gyetVU?`;sqYW5P+>f~U))P$*vz=`VWm zs;DP!txyo|YOnHAOdpOWe}cd_;VB&t;+Bj({43WPi|zyqwNDMi&!T8!hBqh8Q@URG zSo}5C-4^jd?WVJSWAj~s{21I5)cImbUqplQ5VElJd~Z-w(H+ntFxOW`aq_D`y-C3Z z(2>91gzPFaP&8p9oPGH5MF3?7Rf&nIh|hwI-6IiViUhnH+g}knxh2WI@-O8a4YeM> z2cLH(Zr};wdQ-R5IdRi`SZtQ&e23>zyiWBP)U!=O#U~Zje`%d$f&YH7^qr0mGzvcU zfAcYv1)bJUWQ`3x9P-QFwqw8uRv**z$i9YTH$AfxTf9XRVMCT>Z2tbGbVP$@N9=2Mb~rfT-+*Q<@xv=ePLAccP^B z1{`dL1rF3j!n2p|_$Byg>V8MfuK+$b!j=#_5Lalan^A!kl#I7(?OBkuz06TiJo(&I z$?uhY=-mgzXr{cU!TipI4-dU-qx#9BVNHmZ%EEc%{d&E1sbH^ zLF5Z@gHEJ1E~OaeXN%{k&A5Yk3ZsVbv*7sDsnvLOAi-7A0-gK_IHo?h!r-84u{<4Z zwglPVwvKsJsPv)Z`fTAdP@m+T623I$UM?SNuL6eAPV0>;u^%s_YK&kyO0I)Brf90O z59k*gpwKmCfD?~?6s6bhgUCZT$x3<&(vP5B`LQS`If@bari5KAY*|~9=A@8xlV823{sf8;augHd>#9g9;mc#&x3mkO&;rhLH>kJNaA!g% z2op)IL%f(F`%*u&iRlWVFtC^5BZx=opkS!^%u}V^q*I#^gEc51O5L-JIP_9MNJ$V& z!(UMv>7oxQei90UFX-Xu;P}w$i0>r)*_5@5J4^WIB|08t@7%^`Xwe)hg7trrHltH0 zM*f;C^am<-A2E<+8Ox>~T0&u2{7hB(nlh`>YE^|@ z4gXYeJFSoe%QVcyz!^5EJ`13Mw4&LHrpS!dm3M`*7vo>%ppnj!s)uj{m3g_7j|0|l z7NKffX%0>JYY;LsbR@@fF@?~PT!Qu#ICL`(sELiK>7R-`yL(64$+z~eKL^%7uQ5N2 z&p!)F5t$%+(Bd%g3zx&<4IE73&rMtaRLYuVf&CR;tIF(d9d@l;h|MQXN`!fYd8NjA zB4z0pA9V<7uUC zE7By{M-FNM%AaXtltwSfbRT;4-zokf_NUgDV_~)+K^FO31NlzEvdTnu5}rDfg$lJg zSzA{RNg+a%ZA)d)31SHNO96AN5MDNE$PSwhN1j}5`#OK1YwIcHwh=Bt7j^>;+O6)(AJm-tREeN(*e6#)`!&cz3N$cJ-Z-op0+f! zh+cKBL3jub$kS9S{=SUZP6=Jgpv{urmC#2Ke?OPG+EDI&NUy(bKlml{l+q4qO5T8J%`_vd zZc8-<-E&>Tm0?3MxJO>p znlOiw+H3ANwRP48r;mr)8lUaro+mv{5(o>@VL&a2zg=WY(UX!x4zrrG)%(t3w^ zg$jF<$02LOgLp^92|}{RGmE2uH!iIB#e{6e(yB8kDRp$c<^M_GeO zn)$8@l4!(I|AZlw3w(w7a$p**Q@2(+StFfLGOR&&q(@4BL``ZpfuJj{u*T2&2f|gu z=nD|}&gH4fc3*rYk1vU1WC(#jd#hIF^!-d)mFhO32NrhjCwlIOhMScMw>}KlSS5bk)YtCfsH5*v%R3d%C4AgeC{%k0c?~g-9&@xJuUSz4(NAg5|+n8Yt{3s*a7r!FH}KLM8%O z;Wq3|;hq#1K_raErNu?i#IpNg=)|}%bA(l*&?f@e_uSy^Fx&< zkt1se%TM`;7_=@?{A&mZbUKc0Rq&ef}^-hV+ZZSBZ~IB~~h`gKERr9ZEsbt22PSbr&jI#$sHmw!0M|*oEa+F>TMciVoV8F7>ihMXU7RGLXS?Txl__yc+@*5Ze(*S`8 zDWA6-h2pzY1YhEz7E?brhI3Hkp;32QIOT+3lj7yCT{094hN}`lfyApI8Db(0X-|e; zPU5kT9IlU76#AZ;8$L1_$>UAz!)m&W_d7@JY>10I)j8 zSUv$#lP{V=&+!5A&S%8h9};URNCwffAr!dG?R5Om{8sM0jdt>ZId`DJ5jmhrZZ8$t z=6nqg6EttDT%~Y{9qMk*pb9DfLc*T^!K{qd`YdEAHn-j^5jH;-EIp1zh$PjR=x?Hj z1W{4P?{Yn-6b7MA!o3v50_m~vJq>~jb_ub-dBi`4SEUPhF`2x8YInZ_`&SNId-JLi zmOP>p7d3p1ImgkP_eJhiO9Ot?RZZ)Fm?Bs=xQpYXu+VEp zzOl`M0jOci==GgeVVGt^_5dL^YCbAY(Wx~9(CI;Ho0Ex|@9x}x+i1-0 zp;h|Nd7gsYQv7hZ0qgI>sS~^&!=N2s&k|k0r{^mvSp(?fYUS?4^keD7bin65kLS;^ zg};{&jFs;YQg1;QcR&Q&RUMvC(SEDn-P>R6HJ>qf{*8eCzK^@#Z5HfW>*>u@4#hPy+9CFy>3Zeeeu7acr^(d%pd) zPybNP&(GgZ^{Z5hOFur)^7Q&{BmX_>{0r!R0rM|l{{>tJg*<%y1{4=h#scLRihsG| zp7P6I7&QTD8D4gNen}}FNohWI0SOsCNiKGFc0L|{eohW<0a0>c_WxgrF(R9im6MGJ zkeq{?Cj$cok^w}MuIQj6jU9aQcHO_XgsV^#o{l?3WRPD5Kx+M^4-C=BS5D32FgM30 zBT`KbkIsihibrxOqEN>h1zH@ACjsM9lAN%MX=zh{%VSBWh@aPu&u+)FS+8?_Cp|VF z=n_6D2*v(`CXH-Z5~6ZGkFMmHYANXF0hy2+ZYJkY|GtWO8*v?AMGA9ziGOezp}-9Z z*5d$@53J{rbb{l};9F1%B_w`5e)Oa%IgnRFtsv`StOOgV!Jms00aNGy#xyuW{Var) znn&yg^n9~=6n4hYLlzvwoJU9xwJJpOP}FN%(q)Ec0-&A3GKGj)4Sf4a=97QDyf*}~ zKI;Ne3kk9x(rGjq{_)yBOy)_o4#OM9Vup_Fnvo|pVwUr5QnP>N`to|t#yhzK^J%=V zwTDJ&;Czxw=^pC}(tikl9c_H5@>>^yswf0%Xo;D`a9Tm21tu|6Qdhwchu zc+Gv7OxPt%lb|6Rn?VK_!d!FRo~YTd7R+7ZKfNFUmwM#816Tx|`%}tJ7}gfAC}2MD zyW9m^DgJPlT<7z5w5IiD2x%5H_K6rwQC~0vjT$5)^e0LH@F!UVGkUD`+)o(BejadXag_Id+9~_2iXsDdZ5#Lgg36< zW>EyR$algAp0}12h)9R{Z+kkg%;yBnn|{Hb5&w1kN|bst!D2eJ#bep#2yuY?&rJB1 zveeJI%gv*;z>O~eE=jHtDI(Y}HeYP;t?{iZZLlD^kpJiN5}zNsT#9q;jW20hZLS>B zgy5eIeRdFDO_cVkeD9qsZsJQ&bzPegDc4md}-Z8Jmq(? z=S}6{cG8aEJTp14QskVQgcNFAC&?uQv5;%Vk2h^q0=S+KKKgbdf&4@BMpbZ^$s7^v zS2HpGfAheV9g=p4!Ua1QR#wX7ZOz#Q`M-03AEA5*nIBdKdQ&6UO0qx0eGoPa%NKNX z_PqceZ&2DLdTAL+jLmK2u@;qB<3H}z|E~qaN#uuTL~1RVg=Wc)oHqj$0JStu9bT8_ z4Z^*8zyfVDo6H*_D#I}l`}y2`U=>k~@wqFJrdtmDyFxTpIYt>F61xn$Y&r5vu`%W2 zz=LJ424We1)f==LSOWys2v`V8hHuG_(}?-6sgUvusi1)~2A}`bP^hJ9{KqHRkVy(! zlT=4cX$6ritMor(Vr480CJZJtApCoI;ljhDf7A}s=qYJH@aw5hyZ;&UACiYJ0M;?u zv2yS=!T);x-xkE-iMpRgnerbdj7XYSb6)zKv5_L#88c6)#n`1