From 25cf50425aa4025a38849de2a21ac8bb3d86bd7b Mon Sep 17 00:00:00 2001
From: Morten Hjorth-Jensen
v
This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer).
+This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer). The exercise is also part of the weekly exercises for week 37.
The assumption we have made is that there exists a continuous function \( f(\boldsymbol{x}) \) and a normal distributed error \( \boldsymbol{\varepsilon}\sim N(0, \sigma^2) \) @@ -430,7 +428,7 @@ $$ \mbox{Var}(\boldsymbol{\hat{\beta}}) = \sigma^2 \, (\mathbf{X}^{T} \mathbf{X})^{-1}. $$ -
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). . +
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). A given parameter \( \beta_j \) is given by the diagonal matrix element of the above matrix.
The answer to this exercise should be included in the theory part of the report. +
The answer to this exercise should be included in the theory part of the report. This exercise is also part of the weekly exercises of week 37. Explain what the terms mean and discuss their interpretations.
diff --git a/doc/Projects/2023/Project1/html/Project1-bs.html b/doc/Projects/2023/Project1/html/Project1-bs.html index 3a8cec6ec..96ecf6e87 100644 --- a/doc/Projects/2023/Project1/html/Project1-bs.html +++ b/doc/Projects/2023/Project1/html/Project1-bs.html @@ -381,11 +381,9 @@ Give a critical discussion of the three methods and a judgement of which model fits the data best. - -v
This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer).
+This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer). The exercise is also part of the weekly exercises for week 37.
The assumption we have made is that there exists a continuous function \( f(\boldsymbol{x}) \) and a normal distributed error \( \boldsymbol{\varepsilon}\sim N(0, \sigma^2) \) @@ -430,7 +428,7 @@ $$ \mbox{Var}(\boldsymbol{\hat{\beta}}) = \sigma^2 \, (\mathbf{X}^{T} \mathbf{X})^{-1}. $$ -
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). . +
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). A given parameter \( \beta_j \) is given by the diagonal matrix element of the above matrix.
The answer to this exercise should be included in the theory part of the report. +
The answer to this exercise should be included in the theory part of the report. This exercise is also part of the weekly exercises of week 37. Explain what the terms mean and discuss their interpretations.
diff --git a/doc/Projects/2023/Project1/html/Project1.html b/doc/Projects/2023/Project1/html/Project1.html index ef7099584..5a28b77ea 100644 --- a/doc/Projects/2023/Project1/html/Project1.html +++ b/doc/Projects/2023/Project1/html/Project1.html @@ -417,11 +417,9 @@ Give a critical discussion of the three methods and a judgement of which model fits the data best. - -v
This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer).
+This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer). The exercise is also part of the weekly exercises for week 37.
The assumption we have made is that there exists a continuous function \( f(\boldsymbol{x}) \) and a normal distributed error \( \boldsymbol{\varepsilon}\sim N(0, \sigma^2) \) @@ -466,7 +464,7 @@ $$ \mbox{Var}(\boldsymbol{\hat{\beta}}) = \sigma^2 \, (\mathbf{X}^{T} \mathbf{X})^{-1}. $$ -
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). . +
We can use the last expression when we define a so-called confidence interval for the parameters \( \beta \). A given parameter \( \beta_j \) is given by the diagonal matrix element of the above matrix.
The answer to this exercise should be included in the theory part of the report. +
The answer to this exercise should be included in the theory part of the report. This exercise is also part of the weekly exercises of week 37. Explain what the terms mean and discuss their interpretations.
diff --git a/doc/Projects/2023/Project1/ipynb/Project1.ipynb b/doc/Projects/2023/Project1/ipynb/Project1.ipynb index c867ffa98..c7f114b4a 100644 --- a/doc/Projects/2023/Project1/ipynb/Project1.ipynb +++ b/doc/Projects/2023/Project1/ipynb/Project1.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "markdown", - "id": "1ceb1efb", + "id": "24b1316d", "metadata": { "editable": true }, @@ -14,7 +14,7 @@ }, { "cell_type": "markdown", - "id": "1f73d83b", + "id": "cf021550", "metadata": { "editable": true }, @@ -27,7 +27,7 @@ }, { "cell_type": "markdown", - "id": "df2b7d33", + "id": "30acee10", "metadata": { "editable": true }, @@ -63,7 +63,7 @@ }, { "cell_type": "markdown", - "id": "83787bae", + "id": "589ea9f7", "metadata": { "editable": true }, @@ -85,7 +85,7 @@ }, { "cell_type": "markdown", - "id": "d7ee2bfd", + "id": "6280c2c1", "metadata": { "editable": true }, @@ -100,7 +100,7 @@ }, { "cell_type": "markdown", - "id": "958a455e", + "id": "8ab92f24", "metadata": { "editable": true }, @@ -133,7 +133,7 @@ { "cell_type": "code", "execution_count": 1, - "id": "3deb0a35", + "id": "80b126e0", "metadata": { "collapsed": false, "editable": true @@ -185,7 +185,7 @@ }, { "cell_type": "markdown", - "id": "742ae4b5", + "id": "bbd0b625", "metadata": { "editable": true }, @@ -207,7 +207,7 @@ }, { "cell_type": "markdown", - "id": "e49ae3a8", + "id": "0754efce", "metadata": { "editable": true }, @@ -220,7 +220,7 @@ }, { "cell_type": "markdown", - "id": "091e8f0d", + "id": "b87c82f6", "metadata": { "editable": true }, @@ -232,7 +232,7 @@ }, { "cell_type": "markdown", - "id": "3fad86e6", + "id": "8832d308", "metadata": { "editable": true }, @@ -244,7 +244,7 @@ }, { "cell_type": "markdown", - "id": "d281bfb0", + "id": "55129a1a", "metadata": { "editable": true }, @@ -254,7 +254,7 @@ }, { "cell_type": "markdown", - "id": "af6fb771", + "id": "44243865", "metadata": { "editable": true }, @@ -266,7 +266,7 @@ }, { "cell_type": "markdown", - "id": "0d89eb4e", + "id": "ff13dcbf", "metadata": { "editable": true }, @@ -295,7 +295,7 @@ }, { "cell_type": "markdown", - "id": "e1c30216", + "id": "c2078519", "metadata": { "editable": true }, @@ -313,7 +313,7 @@ }, { "cell_type": "markdown", - "id": "b121745d", + "id": "f718aec9", "metadata": { "editable": true }, @@ -325,21 +325,19 @@ "you can also use the functionalities of **Scikit-Learn** (recommended). Keep in mind that the library **Scikit-Learn** excludes the intercept by default. \n", "Give a\n", "critical discussion of the three methods and a judgement of which\n", - "model fits the data best.\n", - "\n", - "v" + "model fits the data best." ] }, { "cell_type": "markdown", - "id": "56d96043", + "id": "d8b10104", "metadata": { "editable": true }, "source": [ "### Part d): Paper and pencil part\n", "\n", - "This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of [Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer](https://www.springer.com/gp/book/9780387848570)).\n", + "This exercise deals with various mean values and variances in linear regression method (here it may be useful to look up chapter 3, equation (3.8) of [Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer](https://www.springer.com/gp/book/9780387848570)). The exercise is also part of the weekly exercises for week 37.\n", "\n", "The assumption we have made is \n", "that there exists a continuous function $f(\\boldsymbol{x})$ and a normal distributed error $\\boldsymbol{\\varepsilon}\\sim N(0, \\sigma^2)$\n", @@ -348,7 +346,7 @@ }, { "cell_type": "markdown", - "id": "e64c004b", + "id": "c5267aae", "metadata": { "editable": true }, @@ -360,7 +358,7 @@ }, { "cell_type": "markdown", - "id": "6f17d2da", + "id": "fd9640e3", "metadata": { "editable": true }, @@ -371,7 +369,7 @@ }, { "cell_type": "markdown", - "id": "fc4fe2fb", + "id": "630acebb", "metadata": { "editable": true }, @@ -383,7 +381,7 @@ }, { "cell_type": "markdown", - "id": "93babb22", + "id": "254ecf9c", "metadata": { "editable": true }, @@ -395,7 +393,7 @@ }, { "cell_type": "markdown", - "id": "4b4061ba", + "id": "6b4ac35f", "metadata": { "editable": true }, @@ -407,7 +405,7 @@ }, { "cell_type": "markdown", - "id": "1ca5e57a", + "id": "404fb6a5", "metadata": { "editable": true }, @@ -418,7 +416,7 @@ }, { "cell_type": "markdown", - "id": "da25c4ba", + "id": "8d8f00ea", "metadata": { "editable": true }, @@ -430,7 +428,7 @@ }, { "cell_type": "markdown", - "id": "10ce6cfd", + "id": "60f92a7f", "metadata": { "editable": true }, @@ -443,7 +441,7 @@ }, { "cell_type": "markdown", - "id": "8d3f716a", + "id": "6eab356b", "metadata": { "editable": true }, @@ -455,7 +453,7 @@ }, { "cell_type": "markdown", - "id": "e30af070", + "id": "d41efd26", "metadata": { "editable": true }, @@ -465,7 +463,7 @@ }, { "cell_type": "markdown", - "id": "1bc3edc7", + "id": "f2ffda3b", "metadata": { "editable": true }, @@ -477,18 +475,18 @@ }, { "cell_type": "markdown", - "id": "050e1f04", + "id": "78ec374f", "metadata": { "editable": true }, "source": [ - "We can use the last expression when we define a so-called confidence interval for the parameters $\\beta$. .\n", + "We can use the last expression when we define a so-called confidence interval for the parameters $\\beta$. \n", "A given parameter $\\beta_j$ is given by the diagonal matrix element of the above matrix." ] }, { "cell_type": "markdown", - "id": "52486468", + "id": "74bec52c", "metadata": { "editable": true }, @@ -521,7 +519,7 @@ }, { "cell_type": "markdown", - "id": "008e92a6", + "id": "afdd7a0b", "metadata": { "editable": true }, @@ -533,7 +531,7 @@ }, { "cell_type": "markdown", - "id": "123f65c5", + "id": "87974f8e", "metadata": { "editable": true }, @@ -552,7 +550,7 @@ }, { "cell_type": "markdown", - "id": "a4398be8", + "id": "335ac425", "metadata": { "editable": true }, @@ -564,7 +562,7 @@ }, { "cell_type": "markdown", - "id": "3e5bee01", + "id": "16b4fea1", "metadata": { "editable": true }, @@ -578,7 +576,7 @@ }, { "cell_type": "markdown", - "id": "e35fdf55", + "id": "c0f81bfe", "metadata": { "editable": true }, @@ -590,7 +588,7 @@ }, { "cell_type": "markdown", - "id": "ac0994d8", + "id": "ac7c8a94", "metadata": { "editable": true }, @@ -600,7 +598,7 @@ }, { "cell_type": "markdown", - "id": "65ba61ff", + "id": "4cbbc7f8", "metadata": { "editable": true }, @@ -612,7 +610,7 @@ }, { "cell_type": "markdown", - "id": "dd4e4536", + "id": "f52e099f", "metadata": { "editable": true }, @@ -622,7 +620,7 @@ }, { "cell_type": "markdown", - "id": "7730a37d", + "id": "0838f22d", "metadata": { "editable": true }, @@ -634,12 +632,12 @@ }, { "cell_type": "markdown", - "id": "8cefeb53", + "id": "9e5f062d", "metadata": { "editable": true }, "source": [ - "The answer to this exercise should be included in the theory part of the report.\n", + "The answer to this exercise should be included in the theory part of the report. This exercise is also part of the weekly exercises of week 37.\n", "Explain what the terms mean and discuss their interpretations.\n", "\n", "Perform then a bias-variance analysis of the Franke function by\n", @@ -653,7 +651,7 @@ }, { "cell_type": "markdown", - "id": "8ec208c4", + "id": "76ec4768", "metadata": { "editable": true }, @@ -678,7 +676,7 @@ }, { "cell_type": "markdown", - "id": "b142cd65", + "id": "c7b05bee", "metadata": { "editable": true }, @@ -706,7 +704,7 @@ { "cell_type": "code", "execution_count": 2, - "id": "fe878605", + "id": "a562d2f5", "metadata": { "collapsed": false, "editable": true @@ -718,7 +716,7 @@ }, { "cell_type": "markdown", - "id": "10b84fbb", + "id": "9e5a57d7", "metadata": { "editable": true }, @@ -730,7 +728,7 @@ { "cell_type": "code", "execution_count": 3, - "id": "e845fd10", + "id": "b97b3537", "metadata": { "collapsed": false, "editable": true @@ -756,7 +754,7 @@ }, { "cell_type": "markdown", - "id": "0745b3d5", + "id": "db97fd61", "metadata": { "editable": true }, @@ -781,7 +779,7 @@ }, { "cell_type": "markdown", - "id": "2ce7a62f", + "id": "a8a6b74a", "metadata": { "editable": true }, @@ -795,7 +793,7 @@ }, { "cell_type": "markdown", - "id": "34b72e9a", + "id": "4fc8cfed", "metadata": { "editable": true }, @@ -825,7 +823,7 @@ }, { "cell_type": "markdown", - "id": "d0f3a4fc", + "id": "eff5bb90", "metadata": { "editable": true }, @@ -847,7 +845,7 @@ }, { "cell_type": "markdown", - "id": "436bcd57", + "id": "ca7c2f59", "metadata": { "editable": true }, diff --git a/doc/Projects/2023/Project1/ipynb/ipynb-Project1-src.tar.gz b/doc/Projects/2023/Project1/ipynb/ipynb-Project1-src.tar.gz index ca520b06ab53eb3ff89f2f4b68003b8249d86db4..8eef8c769aabba35cf740480e0940b8167abfeab 100644 GIT binary patch literal 193 zcmV;y06za8iwFRQb@gNb1MSaC3c@fD2H>uHia9|^Vw$W4UAPd6c!AWWHq|CINx|OU zK0sHBn<7HK&Cf8yFmu?hH~Vek?><@#LWofcV{(>GiO5_}Fy??Mwryu)aUmZS40r7t4s`eAx@g45h{r*4P zfzbm{#nP;H12zNEb6ca-XvWqF-OYVb#l8CnESpHtC5e`Ix3)4HD1S+bT)O~J!S@YkDy{ytvj-+M!;av@7@a^K?50hyedn?$Cr@L~aI zXvi;j?$MD0Jpwcp8%t{uQ2d^kI`=X{rZXk3Id(UW>kZ-7fup>{KNGos2ykJQ*OgVq zsfhg_7x(uH|0v+ofLdR+Bfm |e^p(5^(FY?*x(Ge*DiUma}RK~x(@(a=L<}dyM4M{nq zlhsNm12Z=^lR;l7f2~-_a^$!Xz3VG@q#`6X7jQfF*wf3zcpTBKhz`fJ4-gAg#Sx3F zA;^*BulGx2qKdV++uMm)l0dF6GZ6=yAP(NPa>&;pWY|Oa*mR7ikeZtb;hJs|tH@ zkOWDVMP*S1MP5c}mIn{J;McI}25WBbv#buMfxcDvdD6FOe--|5n-`((H)iDJwr8?m zhVhy99nmZA>3$_=Yz?mr7>&I4m# Jlkpgo9VhcrqdhU_TRj#%ERdP z?+-sgm@|=79;Fpa^zOFGLSs5QAq&@|F+5`NvT#3Qfa6wkBRj8j`Y=0oRXEV>xtE{J zI0{!;xYCoBfBU02^(?3=Dw8xQz)F^iR^N8`vDZ`k$8B1MrUUC~NMe>n*sk#{O1ibg ztu!gZXP$H`-s{>Y;==IEagOioELD(Iyn3-pT=B9sle?n~*DOEQME=7mMYxjF>~JXy zWKc;^q)C*OEXl*RwV= W`QBWH%w#f)b^X`_FgF!er69* zIAvr*n->ih3>|nw$HkEf#!cXdpWfdDzdFJvz1egoYKEQAavaL;`FZxB@T5NTp1dw2 zRWO%-e|72tqhI4p^}6H9UXK%xRz`Q0htKV_ e;e9%(pZMzYqgUI8`j3lN^*G$g;4&7@%>{_R8ENYUK zun1c(?ateZ1ZOZsJdGe3ti1 k|?jmF50r|K?CG_@OBSq5`Rh9 zf49HS;`tsM4Wo27>q&N=w{1ga-7>*Wu+zgUx#%QU3HE0`TJiTIYt2sV`IL>$9beT^ zk%`h>R+#9Y!W+wtO~1OMdu#7vw48@_%aBAQd%_lkckTbz{AtE^Y8z6zI|%ZXET2<1 zt&x@J`$|7M@1+@zn>&)soz;`(dAw?Ee=}R_<*6^cC@Z6)5-W&GN^ ^wh&@%JPW(4$>5d*xve|e|C zSrXNGo>EzpL`jhcut!u?U~mNne?VeE%sg#ImkW>va{vOk)6+bf6L>zWk&-k@$}+f5 zswnfP@wG=l2VNX;Ft4Jb%mY;-wdBd7>R_2TYh|f-6|rM-3~ZnTDuX-^4^f|pxl*s# zBxHiI)qP{^*#SGFdnUKP&q$)Oe+*L$qrC_<&u@VoQ%jPWEknWuY ZOrLqgmQv?k3Kj+`Gx8|>;r2~+|!GbBMm80|1S}H z4J%FRwJ$C7i~_blvE{+0_b_~!;R9D7PawOemv&Ez#l;zBD$Aq7mlO|Me-U*uFB8KZ zXQ$zb*J0Q1=&TAKC>@T*6^lNZebA^;%pIrWPlJ?Pz6N+`AMLgsb>H4G`A=pXc2WX- zh t`VYXL`luV(>lAuNTVd+ZI z`SaHFK3f=+v4@&vQM6*Fd&`++w9y%?Cn~*y>*!|cq+&~bO_TVWxgffL`7EWfDrJ`6 zOGig7ROPf>)KrJKf4IbOWh^R=y(qGf!n^Q|F$U-gXP1Y0da>`4^zexI<2jq@NKu81 z6BM1$`>d~{V-{0gB~6?QgN5@^A$L|&WXIBqqG2mf0LKiD=5zIJeP6@V|3cp=^H3Rz zi~QN?^YKu8E1w`09HH9`sP48qUOzU3m=OQujK|-Ie;z@je{UQD;Ea`F89OOYNXB~w z*p%2(OTA#SbLYNZv__uyvW&{S@Ff%4H+Lt%apJq1p KjvE^Bwkq zXB+gMvO6Jff3N=ohspmvhq+SMqzc@YqV$h_)6FaD8)|8a(;5*9^k`(<_!fqzuKcg7 zMe|ClP8KR=cdlcN!#_EQ^!>?odXk;Cq}Z(iQQhLv>WuXirC177tW#0+--y&%D@_l* z4bbO27*ihuX~M$PP8gc%iT4+sIA-tL>BIc^2A`kxe `bT>0^d&c1iB;wC{Q9PaYb(cGq$vtq%Hf&k4)N zOWcQye|cdDg)$+j<0!7g5T{yP#HCs!*#-0+Uk5HM_ SI$*lVdkpAKr_}3>)q~lzSN3@%pf&!;Y5MY7LW8801;_2?g(tZu6D)@=+X35=9P$ zbayS5odqbS;|mMWs|EN*ToLPZ8G7!w^tMGiSdIhBW0&b^AW~3mDqDw_4vY_IP`l?5 zcYG%gH-`rhtTuWt#ns8-EC|`%UXAEP6;bM6!G`|CFW}oRl(p=6m!uRr;VQV#3pC`4 z-Jm+5`SIcApFEdDUy}q?q< Ai>^_nC_7ZmuWI zgC}lS-I@B{bkklI5wEO=mKwuemeJg6Ynr+vk(+R~q3MUS)s1mx9DitVe4yX2xNe*+ zo@JNBRkcz&FW}H{cv+IpR=c^ITnx&pD*6F;Nb~5tzrih>Gxwc-OENPw-P~&G)c8|G zKhHHNQ%~j*KN-e`_Apv%!vued>DjE-xMp^&JoG_%R99T B8}DJM45LYS&|dUDv6G0x(ZE)Jus;sHFP5MFble)g;8;wf=9wB> (GZ6`Zl)Q-A)(P(x4HtbcFKWPYCY zyZud`Ws!m7D)2}2+k*WTbtLKHxeEO3D|&}NFZP {pX41qEWolzMyR>;!ISet^P$@2vp7f_;k{+(vtJBKGb>8fCSKzTXCmrOG;AT zjXmsOh9hriT7O-$FF%aDuj%2Y@G2j1p}w5*HHGsk&T=*{L68E+U`3%hoyPfi3Z_B{ zyUsWC1(Nl3%PlzNKrJ_RkDzSWhR1_Z5PAT#{)XTgmKt~B+2gDF0`lew8~BUYK%epG z_zB~v)xbqNP3_tnFJLbS3|U0my7ACNXMllDQ?KKN-hUWrxpE!ad+j=2QV;ZrlSc15 zQ)hNB0$T=Qk!Ni0^<^Nzs~@;V0HoLSFN(NI$}7&X0r3nu9V81Itc3jTI6_uB^=<+7azSu(#?li! zn16YyQKxmN`yB2j-BLGTctBsAF`iAsJ3pOFgNskkfJH$(s~qxe#@1gcp`r&&iJJM6 zPxGUlPK#m}BGfZig3JeI^09ZgRqzo3J7>~^k5qB4%H2&`#APN3eQx&^D&KpDDl$$+ z(9j*^Q46~@)Ng_IKJHUNp@kCv=>l&RVSiJiW+Wi=g$B6{da)ygUIjLwj7scOQpJUk zjBEZD3xq8Fpsn?mE(B`G+9p9;!B-9vuZ)k^iAH>0DG2|d`_qS8>La{#n8~udKok;W za~PqV0CU-KK{We}1fmZg-;-pt&)^NtP~_z ?Ky`RfOx}k2BcdGj4RWpWfqqS zoBnA<`pNnsW&`f5oD^c+d-li$2(M$rnsAKv(;Uk05AVPImMfH7 gNT-IlQ^3W!UHQA{80?_=3%2I~VG_~blkuA=v~ zCpuKeI+FPU7s(DWW|dIJOREjVuF!k@3H}8T;NVVFUJ@U`!LS5ef%JrbP=5*8<1RG+ zidtv+3x!2mRIvaE+NgY*rn7r@I6OT)#j_dXVLbTTy^#`j^K{9B7#T hOyr7Aw?r*k*fEvJfCS3(V! z6y j`j_B)kQXttE+jSfxoHHn!k$_N)+7$EIWp-zVFDx22_`?&wGt>9;{Q-Xd zg7w~d4`+JQ)%G>s0DLqSY2Xb$ud-tJw_J+yMPK&^sw;1?<{vH1qJK&ora4fl(IjjL z5L@oj0Y4{M;^2>1D*0-u^|u@Y*R=c$(gkkVlKc$SrQ Dy|rQFt%8ViFFZ`JU4d2#s)fsB$Mpz`|a4*hCo{YXj6HQ1M4gul)jE1S;$z1 zH?NC($Q5QwekTYUOMgh&9W{s-$&K#PzhJ>+dc< ZyAA}w zY( 5NL&kI6vTfmjiLtC$7D4C0)cI);V(B;*-*Pc0uU_&ZNnj?NQL zup>`cLh(h{N>r>$8I)e2bMrt}v)5?v^376wdBMbm*aOYA^WZiaS4D_U!+GY3z}+Az z<2wL&S-d2OYdSAVd3j{JHG?ne`ToF67pjcU)fhfVvVZ;1DheugV9W|#=iej&_igQq zNMA^&(MO=A#4l)DobZL&WfdLmIG_0j;Z7ZcRLSS9B6*X($caUfTpr!r1TaE{BK`k? z!qO*QYF>U0Q~rLzSRsM~XEDR(BJe#eFzFSfZv%8E@+waAY
@CZ-N4?L5$FRGvvX&=w*oUZlR;l7f9+iB zZyU*x|E|Acfcugl(DeItSOoBMW5jkg_TFYUf^is8BZ+5r@x=9X2(V<9Z z;{*qoH8VZcJyq59>~2|AvNDxXm911zm8*;om5P;@uJTp1rc$X=(#s0XIsnK@Sr;p% zD&?d+Q65*IX|%hDE|u~+RT>=we}>f#qv1hoT)nRhu7#XMP@N61rW4$q;O+zt66D|# z+}$A%+%-6SxVr~;x8QCEcXzko?gY61+^L%FEV`-}@4ox(e!3ZsN44(4n%Jvn@l)xi z$+!EhTn4T*G*;n$tY}$Y%9;9;u~ozRqO{d@%_43!eb!VmS2rzeNQUM^3IuzL5_#<& zC6ccdkK&`kIsVFB*h$> >{Yc>cXN}(i8?MYK1|oxNwCq_K)&>s=a_3L`M4_swSQ*RiW6Ezcj0cxyz-j@$`2l zF9J!Q)nL9Qz=LvTNK2$oP&mwR1l~oVCw5ZZ{XmtK4zZAa7@86du1^|HlYBr87CG;0 zINs!Ta7Dn7#Ymt8M&^}#nN9K)syG0R9X*&yPuUlM0(&fxt^>OeMjkMqAMtrKpPCn5 z5qv~KV~N-%DHjmX)t^=>?vEhLOsF3(&$ u+X8`xzXi;)Kut={y$eStZ%;3?pWPH+_n6|D;xnIFN4@NBDud4 zXQ8K4k=GU!2Al<_*9|@^T(`&Wo-~tm8$3@scZQewMB!wI8z26;!R`-vzt_1qo*r)x z1C*ASr`n(8eViS^&A{{lZx1KIIQ}}Zo@OPR?)GNj=@r=^Gfq*+Ev=+M21nYOp>@dz ztn+Je*00<5m6Ag H-`F)}eqRMVp6O6!(N;K!gGLL>FmIjd?|A)2?c*|us` z70+;<-rb(E&$n!ZHq19Yy-fvLpVu 1ct5DYMJ6&5l+vWi zm7jUr0LekpHX$Dr3K#xXft3P3-$_du`^G4}?Q;bs`%CFZ`Tx>u_906<5LjXCB{M$~ zH;xKmktJ^kh2;Mx?zCltMi5ub(cG0pII`1g!7ga%Br0t1l}gh2R`HnhkiJ4eHJsV- zgFV=WCCO_!6o`=}rp~nVh(%HIWv0M)) M<8hyX^OFfUbEuXo)Hz)RA{YQgB3>p<2YG4&Kyaydx*Hz3o&{IPG%y9RC zbSLU{{vKg>sqBKRX1i<9!rQvq{x%p4sV?(gHWhA<+7t2Ywsf%m&Oc3>xmp+MlJ@B6 zbSb?e#vD)_Mb@>A;MTb#=h|XDKVWy Q2>u;1D7)x*s&{E&Ul zfb1YV3J4hz2`-TrhO^MfmZIvF47ha|3t^iw<3X*Ibp>BEVTsB>{Rr?|^LTALT?>BN zx?Q{87gds{l~|fU`JKOmbl&}P`f%sH&ZHp3aqIYvq_ISR{k^HZgMTy25|u9%)0zA< zEG`#yJ%jB7q?f2%^mfj&VGN8eI=b1l7y%FqpleL<;o6l5u6;}jeh|#DY=q?Gf_B)f zA?v+7_9d{>I8t!#Z@rW#*N9<^E;R%1i9=)j&xD#pYHfV2v(LOp%t0DH>GEwn0T<`H zNad?O;PRu0MyCd9Mr-rME(k&J5y6M$f{Bp*npcH1bECE(!*#=P#@{{ZcJQVmn7~>G zu+eucRUs*SM&F&Lj0_LaIKr*I3d71~4YrT}-oj4y_~4mQS~QuG*VL}s^r~7lp<|&m zMSSGbm-2`h>X{=aIm|I6XEz)@L!w#QqVy`Ji$tmr)+9 9@g33dK6K2XmNQ5~3*K znnk&unWp&k*eqTEGK^IkhHZqFo)7UbMjo8gKe(}(A~8%4DBByO8*Blpn;z~D-g}GX zEyQ52%sAC@G`5Ri>%`t#BO$#EaKRRQDi;go$&(N9j^XdP2a!GHGR;*+VPjrk{CQwp zqi!f!geIAd-YY=n6c 4t+wD z*d95=3Qv(}t5{>iA9vhVuAA{#slMSOKse#ivZmT{$Td^~jl;g#bF (M3G>^xWP*r(bPjm*Teo|J4s1RjVp=MdeIqal6;c9+$7FLMaw^8PvGn%g4VzZYcV>cx zZbP#8+I_b)GZ&hqRR7sfe7128eJeJ@=c4`j=*1c{xZ`U;nuT?&o(8Om!11T>PW!(T z0gE170_0giD@(AOz03#FN{ym~-DK=Is^>hcMG?}M;c~kF9xi65K_<`%jVp;J&^i10 zYv$jbG#Q%Q;GDid^=$*~iS n7p9SR_g1*mP!A9yQPx9_^V5 z6-kS;vASjKI!)Ag*7awteUtcHX@mrg#$nTy+?x_qGufBl#?_|jT~5O6iHmm4F}cKM z1b0-QN$WZueaTAIic{JJ8g31~r;c54L+tIlR*u?MzEFqrLRbJ7_Vh#36?jGrWHA)C zpiCCxuBsYie}a-Q$TSBY*pL$7%fonZxr$c%ybxCW-IBA|z(ekbTKuZ97&2ox@6}T^ zk}@`J6fL=_Qf1taY~>p2=uj6NjCbVD`cVQ|=8!=&LK;>eXO-kAU!7P*3UxtF%sB{l zc7{tLJ|*p%i4ZvA@nPNhbypba--o0ZPvhw4GVz!lv#(8>qt1ze)!9^PhUvz&eHgUN z{~uxE50Zz2&CXpIQ9&lDaxt-iLCun`w49n;REx1n$Sic-!;NObWy7N1kC*2o#d$8A zqt-&^LS 4&I^oR>Exys2{5#JnTpkn z`n2CzpN}fS)nYC0(o8g8sE+dmZH{ev-FiYUM+Q3|?U%V}btSa6JmisvNYSUU)j0G= z#5817XjEi)Wf1MdaFU9xj~}Ia!vFEKP}8XF+axC1AZOo?*M1pHkFl&9#+t&|%ooQh zrfZg!qyrE+O@W~oZ2HwN=-?ph5feqkBK?sq1}>J7`f<{EkKKOl*=_!kLdIt;lmW3> zwDN-^8@j3-OG79 zHT9sg$mvH2Y PU z=~O4ejFq2~=Tw?dG5E;DwFt4Zqyvr)BXpJ6BPA@CvP6%}1?C%*dGxW}v-+Jc9l9@# zjXjG@=)Vmg+-up10=*Km29XL7SesAo@f|JFzMe%HExXI@Uf%?vY?{G^gtRS&=b``o zG_--i4w;>j8r^3iaVWF7A;xlkGXrGao9}20oa_SfxsCZ=_K0wg&}~lEi;rU+!@~D( zh~pLxc$KV4S8jDdUpW5KGhSU=8S=@PT?V1S=>`ZYZCxODO~x%vC6e#}G+7bpPFsdx zn&zZ};3L^T(Do53K>;|a*j$b})I9A6ltH6zNKAU5nRGhqEMtCCt@M;JQlRsuG~Arh z$>5|eL^9!i8tR*#ER7c+Gs8+(x_9%gHa>((ZmH2Dm9&a6%AbB_3WlcRe=)8e4wu>Q zkF|mZaTo5SiM9Uum2|2D0!L)QW{=tU=#le+Tx_IaNNg@YfR6?;X)u}|EW{{C>$Fp( z3ge?l(3fxP-=NRGbpP2IV8niNXsbx^Iaz5v>&%0Go2gvy-m>|{mf9Iw1Z{vPh-eS} z1r58%w5@+Mg^pd8rCMdZZ>q`W@}^a7lG)0gGxT(va>SoM)S>hS@(x{QH P5Ff z^6oU5T*Q`ii(!Un1L5a`ZoOn$fzxmx!{msjnOHY1{M~137wW(VXt!5bKE`zP`xnf{ zGrAZvq#eB4^bGazD}+$cfLIr&Z#)exMTF5h+$B%38p5!TE`6T$4fZ40mm-7AMGdZ% z)-#~G52q_r_x61uEWA^)GFCjEgT`SU+0kax`yN+$_@h2tRD&xRSa40Pl?QyIHBohe z&Eg&N5?;XvJk`U1j`886hSlU_+{U%e;fuT(`PR3>Uyu$Rj?(cPB;SyZS&0M5!sTwe zF*lFK=*8{zLEc4A8c_E);*p!T?@Lc9tg+ldG0si12}{rJ`s|lfVoI5#C;HpJhWnAr z!ZVdzzn;-2?2jjJge7=WB&AhxywokqMfN9QFh^>>K&%x2N*@EgDO^)+Ta!j3yg!Gn z>j&VAIT D_UFDaM=DcU+U(@rrH zQ(U;%C003(9y^&gPGhHY>EiOsnSU8q(Uk5|uL_$fE)Cr!jDv=e^LuY^Shge~Db=^x z(DHGn{4ceDib3iciVw=y@(mqAk4+n$?)r|W8t3U ;VN;YR!)LNL@Q}M zeaCKGJ70VKCb|=&C9QuTS^BMS-rUPD#pjTc^Pe1KQ!jSe&DcM (c7D)xBKj=xXG)Kp*NzlG;c zO}DFxv;4T5FyZ&)=uB_G4FT4gxrJsDp2MG7T|`TJ@s}ntUFiWTiM{8eheM}XU#vHG zr ngPCu!+8nY%hY)`RU| zB^v61f&bidswRzK_*6zz2t3=%SWb?>%ds2FkIrMe^s >B_)n z`?)wdv=RK2kro-OC6_+Xcq9(Un^)Kmz+XG+o>?+HbVDDN2^s$HJSjAW?el*D)f~eH z9kT_)%E88yN{$Ib_J6bQm@p71uK)90h9Q*=3+4sjjgRy6LBxyb4ZzA)i*O0sm%{q7 z6);p<=s*!~az*VQyh9<3{!O~qUqn1E$+I?JicwfG2_4V>=--1b{8raERnXKdH9%xq zAkIHECqC4`0wGu@xPvDHfi_V>mX4X^5Co2k=h#rvLK0Ef`0Cc=b7per 8 zYYb!0m=Xs^?1LlRi`{SD{_yT$HCOQ7(e@5uWLN)HixsARW8xi2{$54!Ah2`k9$^md zWl%^&n7W2C2S@r!AWk~IH-u4Ueq`;^yaVAMgDg+L45g2)JYVEzG$WXf$F|$oOD|B;Rwr!_ z 05l>3u-jJ9nM`O~(&Sds~BxWKRTJa|A0xJ1# zj(BQrM$mR{Ay;>MwVGF!(Hr8$ivDs{Or*s8nC>xD<>^!}YsA+ zeTY=V6ZP&$8R%Ut5B-*4FZ|H$7-#<2FOIxT@A!(_@iv5JirdLkPen2WLP2$> z6&Fj7Z%C#;`HTz6qx{(4_<|5u0hwjn-dI}dvUCP?w%d(?gX8CCnM_wOEOb#YRk0%e z%!?R_j}@PhByo+ZQZcD7vY=zg){gUZsnC^Ifokg4qPBsYE0Xu9%WJ#>oyqvhOv!(> znFA}ZJw-hOr1$nw6PJq9-3-2RsmKKZiY&KuKwNtx{$$?_ zlvz*?KJ^Teu~nyim*B%B1&hxwL4C6)`1e0@x5JxS6_#IwYWQ2G7u VYLe$27 9DnEeZ8an7L}9o zgvwa8#cUEnhwzb3DLef}S62v2Q^rvJWY!LR_6(}lA!iBqK-ABV$n*M)LN`NPwsT(s zfGyiN0lB__>zQryr#GtO&ny(i3;kz2obh`cl(HU2ghDZ8>86D^XDPtOLbtjTUYRx$ z=48!4&Wnh#O5L`xQzDQZ&^6@}nXS MOL+HG_|Z894Yj*+&uC7>CuxgsK>g`I%9T=Aze@ zs$9tyTkBQ*?_+PgIzpwnMQF{Nf);O(nTu;rpw)Mi5VXTnTC~@t#v#ldtX>Ndx%3Rx zm1IJj)>a>qQ^R(403VXDD4)iulQ`w-4fAGJl6$=h3sU-?dpj!$`C4U|V@$cR!lRDo zhd+mXYOWZr$q}hTt#fdOYDM3r>eK6-uAs6Z=DzR@u&X48lNUb0J4ARnv&Vs&MQHB_ z?1hszS