From 052f6f6798452d1900996ae5d8249a2a2256e5cd Mon Sep 17 00:00:00 2001 From: mhjensen Date: Wed, 2 Oct 2019 22:32:31 +0200 Subject: [PATCH] Project 2 --- .../2019/Project2/html/Project2-bs.html | 429 +++++++++++++++++ doc/Projects/2019/Project2/html/Project2.html | 370 +++++++++++++++ .../Project2/ipynb/ipynb-Project2-src.tar.gz | Bin 0 -> 212 bytes doc/Projects/2019/Project2/pdf/Project2.p.tex | 431 ++++++++++++++++++ doc/Projects/2019/Project2/pdf/Project2.pdf | Bin 0 -> 236296 bytes doc/Projects/2019/Project2/pdf/Project2.tex | 403 ++++++++++++++++ .../Projects/2019/Project2/Project2.do.txt | 239 ++++++++++ doc/src/Projects/2019/Project2/clean.sh | 3 + doc/src/Projects/2019/Project2/make.sh | 79 ++++ 9 files changed, 1954 insertions(+) create mode 100644 doc/Projects/2019/Project2/html/Project2-bs.html create mode 100644 doc/Projects/2019/Project2/html/Project2.html create mode 100644 doc/Projects/2019/Project2/ipynb/ipynb-Project2-src.tar.gz create mode 100644 doc/Projects/2019/Project2/pdf/Project2.p.tex create mode 100644 doc/Projects/2019/Project2/pdf/Project2.pdf create mode 100644 doc/Projects/2019/Project2/pdf/Project2.tex create mode 100644 doc/src/Projects/2019/Project2/Project2.do.txt create mode 100755 doc/src/Projects/2019/Project2/clean.sh create mode 100755 doc/src/Projects/2019/Project2/make.sh diff --git a/doc/Projects/2019/Project2/html/Project2-bs.html b/doc/Projects/2019/Project2/html/Project2-bs.html new file mode 100644 index 000000000..6106647d0 --- /dev/null +++ b/doc/Projects/2019/Project2/html/Project2-bs.html @@ -0,0 +1,429 @@ + + + + + + + + +Project 2 on Machine Learning, deadline November 8 + + + + + + + + + + + + + + + + + + + + + + + +
+
+ + Project 2 on Machine Learning, deadline November 8 +
+ +
+ +
+
+ + +
+ +

 

 

 

+ + + + + + +
+

Project 2 on Machine Learning, deadline November 8

+ +

+ + +

+Data Analysis and Machine Learning FYS-STK3155/FYS4155 +
+ +

+ + +

Department of Physics, University of Oslo, Norway
+
+

+

Oct 2, 2019

+
+

+

+ +

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

+ +

+The main aim of this project is to study both classification and +regression problems, where we can reuse the regression algorithms studied +in project 1. We will include logistic regresion for classification +problems and write our own multilayer perceptron code for studying +both regression and classification problems. The codes developed in +project 1, including bootstrap and/or cross-validation as well as the +computation of the mean-squared error and/or the \( R2 \) or the accuracy score (classification problems) functions can +also be utilized in the present analysis. + +

+The data sets that we propose here are (the default sets) + +

+ +However, if you would like to study other data sets, feel free to +propose other sets. What we listed here are mere suggestions from our +side. If you opt for another data set, consider using a set which +has been studied in the scientific literature. This makes it easier +for you to compare and analyze your results. It is also an essential +elements of the scientific discussion. + +

Part a): Write your Logistic Regression code, first step

+ +

+If you opt for the credit card data, your first task is to familiarize yourself with the data set and the scientific article. +We recommend also that you study the code example in the Logistic Regression. + +

+Write the part of the code which reads in the data and sets up the relevant data sets. + +

Part b): Write your Logistic Regression code, second step

+ +

+Now you should write your Logistic Regression code with the aim to +reproduce the Logistic Regression analysis of the scientific +article. + +

+Define your cost function and the design matrix before you start writing your code. + +

+In order to find the optimal parameters of your logistic regressor you +should include a gradient descent solver, as discussed in the +gradient descent +lectures. +Since we don't have so many data points, you may just code the +standard gradient descent with a given learning rate, or even attempt +to use the Newton-Raphson method. Alternatively, it may be useful for +the next part on neural networks to implement a stochastic gradient +descent with and without mini-batches. Stochastic gradient with +mini-batches may give the best results. You could finally compare your +code with the output from scikit-learn's toolbox for optimization +methods applied to logistic regression. + +

+To measure the performance of our classification problem we use the +so-called accuracy score. The accuracy is as you would expect just +the number of correctly guessed targets \( t_i \) divided by the total +number of targets. A perfect classifier will have an accuracy score of +\( 1 \). + +$$ +\text{Accuracy} = \frac{\sum_{i=1}^n I(t_i = y_i)}{n} , +$$ + +

+where \( I \) is the indicator function, \( 1 \) if \( t_i = y_i \) and \( 0 \) +otherwise if we have a binary classifcation problem. Here \( t_i \) +represents the target and \( y_i \) the outputs of your Logistic +Regression code. + +

+You can compare your own results with those obtained using +scikit-learn. + +

Part c): Writing your own Neural Network code

+ +

+Your aim now, and this is the central part of this project, is to +write to your own Feed Forward Neural Network code implementing the back +propagation algorithm discussed in the lecture +slides. We +start with the Logistic Regression case and the data set discussed in parts a) and b) but train +now the network to find the optimal weights and biases. You are free +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. + +

+Comment your results and give a critical discussion of the results +obtained with the Logistic Regression code and your own Neural Network +code. Make an analysis of the regularization parameters and the learning rates employed to find the optimal accurary score. + +

+A useful reference on the back progagation algorithm is Nielsen's +book. It is an excellent +read. + +

Part d): Regression analysis using neural networks

+ +

+Here we will change the cost function for our neural network code +developed in part c) in order to perform a regression (fitting a +function or some data set) analysis. As stated above, our default data +sets could be either the Franke function or the terrain data from +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. + +

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. + +

Background literature

+ +
    +
  1. The text of Michael Nielsen is highly recommended, see Nielsen's book. It is an excellent read.
    2. +
  2. The textbook of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer, chapters 3 and 7 are the most relevant ones for the analysis here.
    3. +
  3. Mehta et al, arXiv 1803.08823, A high-bias, low-variance introduction to Machine Learning for physicists, ArXiv:1803.08823.
    4. +
+ +

Introduction to numerical projects

+ +

+Here follows a brief recipe and recommendation on how to write a report for each +project. + +

+ +

Format for electronic delivery of report and programs

+ +

+The preferred format for the report is a PDF file. You can also use DOC or postscript formats or as an ipython notebook file. As programming language we prefer that you choose between C/C++, Fortran2008 or Python. The following prescription should be followed when preparing the report: + +

+ +Finally, +we encourage you to collaborate. Optimal working groups consist of +2-3 students. You can then hand in a common report. + +

Software and needed installations

+ +

+If you have Python installed (we recommend Python3) and you feel pretty familiar with installing different packages, +we recommend that you install the following Python packages via pip as + +

    +
  1. pip install numpy scipy matplotlib ipython scikit-learn tensorflow sympy pandas pillow
    2. +
+ +For Python3, replace pip with pip3. + +

+See below for a discussion of tensorflow and scikit-learn. + +

+For OSX users we recommend also, after having installed Xcode, to install brew. Brew allows +for a seamless installation of additional software via for example + +

    +
  1. brew install python3
    2. +
+ +For Linux users, with its variety of distributions like for example the widely popular Ubuntu distribution +you can use pip as well and simply install Python as + +
    +
  1. sudo apt-get install python3 (or python for python2.7)
    2. +
+ +etc etc. + +

+If you don't want to install various Python packages with their dependencies separately, we recommend two widely used distrubutions which set up all relevant dependencies for Python, namely + +

    +
  1. Anaconda Anaconda is an open source distribution of the Python and R programming languages for large-scale data processing, predictive analytics, and scientific computing, that aims to simplify package management and deployment. Package versions are managed by the package management system conda
    2. +
  2. Enthought canopy is a Python distribution for scientific and analytic computing distribution and analysis environment, available for free and under a commercial license.
    3. +
+ +Popular software packages written in Python for ML are + + + +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. + +
    +
  1. For project 3, you should feel free to use your own codes from projects 1 and 2, eventually write your own for SVMs and/or Decision trees and random forests' or use the available functionality of scikit-learn, tensorflow, etc.
    2. +
  2. The estimates you used and tested in projects 1 and 2 should also be included, that is the \( R2 \)-score, MSE, cross-validation and/or bootstrap if these are relevant.
    3. +
  3. If possible, you should link the data sets with exisiting research and analyses thereof. Scientific articles which have used Machine Learning algorithms to analyze the data are highly welcome. Perhaps you can improve previous analyses and even publish a new article?
    4. +
  4. A critical assessment of the methods with ditto perspectives and recommendations is also something you need to include.
    5. +
+ + +

+ +

+ + +
+ + + + + + + +
+ © 1999-2019, "Data Analysis and Machine Learning FYS-STK3155/FYS4155":"http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html". Released under CC Attribution-NonCommercial 4.0 license +
+ + + + + + diff --git a/doc/Projects/2019/Project2/html/Project2.html b/doc/Projects/2019/Project2/html/Project2.html new file mode 100644 index 000000000..3a9330c2b --- /dev/null +++ b/doc/Projects/2019/Project2/html/Project2.html @@ -0,0 +1,370 @@ + + + + + + + + +Project 2 on Machine Learning, deadline November 8 + + + + + + + + + + + + + + + + + + + + + + + +

Project 2 on Machine Learning, deadline November 8

+ +

+ + +

+Data Analysis and Machine Learning FYS-STK3155/FYS4155 +
+ +

+ + +

Department of Physics, University of Oslo, Norway
+
+

+

Oct 2, 2019

+
+ +

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

+ +

+The main aim of this project is to study both classification and +regression problems, where we can reuse the regression algorithms studied +in project 1. We will include logistic regresion for classification +problems and write our own multilayer perceptron code for studying +both regression and classification problems. The codes developed in +project 1, including bootstrap and/or cross-validation as well as the +computation of the mean-squared error and/or the \( R2 \) or the accuracy score (classification problems) functions can +also be utilized in the present analysis. + +

+The data sets that we propose here are (the default sets) + +

+ +However, if you would like to study other data sets, feel free to +propose other sets. What we listed here are mere suggestions from our +side. If you opt for another data set, consider using a set which +has been studied in the scientific literature. This makes it easier +for you to compare and analyze your results. It is also an essential +elements of the scientific discussion. + +

Part a): Write your Logistic Regression code, first step

+ +

+If you opt for the credit card data, your first task is to familiarize yourself with the data set and the scientific article. +We recommend also that you study the code example in the Logistic Regression. + +

+Write the part of the code which reads in the data and sets up the relevant data sets. + +

Part b): Write your Logistic Regression code, second step

+ +

+Now you should write your Logistic Regression code with the aim to +reproduce the Logistic Regression analysis of the scientific +article. + +

+Define your cost function and the design matrix before you start writing your code. + +

+In order to find the optimal parameters of your logistic regressor you +should include a gradient descent solver, as discussed in the +gradient descent +lectures. +Since we don't have so many data points, you may just code the +standard gradient descent with a given learning rate, or even attempt +to use the Newton-Raphson method. Alternatively, it may be useful for +the next part on neural networks to implement a stochastic gradient +descent with and without mini-batches. Stochastic gradient with +mini-batches may give the best results. You could finally compare your +code with the output from scikit-learn's toolbox for optimization +methods applied to logistic regression. + +

+To measure the performance of our classification problem we use the +so-called accuracy score. The accuracy is as you would expect just +the number of correctly guessed targets \( t_i \) divided by the total +number of targets. A perfect classifier will have an accuracy score of +\( 1 \). + +$$ +\text{Accuracy} = \frac{\sum_{i=1}^n I(t_i = y_i)}{n} , +$$ + +

+where \( I \) is the indicator function, \( 1 \) if \( t_i = y_i \) and \( 0 \) +otherwise if we have a binary classifcation problem. Here \( t_i \) +represents the target and \( y_i \) the outputs of your Logistic +Regression code. + +

+You can compare your own results with those obtained using +scikit-learn. + +

Part c): Writing your own Neural Network code

+ +

+Your aim now, and this is the central part of this project, is to +write to your own Feed Forward Neural Network code implementing the back +propagation algorithm discussed in the lecture +slides. We +start with the Logistic Regression case and the data set discussed in parts a) and b) but train +now the network to find the optimal weights and biases. You are free +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. + +

+Comment your results and give a critical discussion of the results +obtained with the Logistic Regression code and your own Neural Network +code. Make an analysis of the regularization parameters and the learning rates employed to find the optimal accurary score. + +

+A useful reference on the back progagation algorithm is Nielsen's +book. It is an excellent +read. + +

Part d): Regression analysis using neural networks

+ +

+Here we will change the cost function for our neural network code +developed in part c) in order to perform a regression (fitting a +function or some data set) analysis. As stated above, our default data +sets could be either the Franke function or the terrain data from +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. + +

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. + +

Background literature

+ +
    +
  1. The text of Michael Nielsen is highly recommended, see Nielsen's book. It is an excellent read.
    2. +
  2. The textbook of Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer, chapters 3 and 7 are the most relevant ones for the analysis here.
    3. +
  3. Mehta et al, arXiv 1803.08823, A high-bias, low-variance introduction to Machine Learning for physicists, ArXiv:1803.08823.
    4. +
+ +

Introduction to numerical projects

+ +

+Here follows a brief recipe and recommendation on how to write a report for each +project. + +

+ +

Format for electronic delivery of report and programs

+ +

+The preferred format for the report is a PDF file. You can also use DOC or postscript formats or as an ipython notebook file. As programming language we prefer that you choose between C/C++, Fortran2008 or Python. The following prescription should be followed when preparing the report: + +

+ +Finally, +we encourage you to collaborate. Optimal working groups consist of +2-3 students. You can then hand in a common report. + +

Software and needed installations

+ +

+If you have Python installed (we recommend Python3) and you feel pretty familiar with installing different packages, +we recommend that you install the following Python packages via pip as + +

    +
  1. pip install numpy scipy matplotlib ipython scikit-learn tensorflow sympy pandas pillow
    2. +
+ +For Python3, replace pip with pip3. + +

+See below for a discussion of tensorflow and scikit-learn. + +

+For OSX users we recommend also, after having installed Xcode, to install brew. Brew allows +for a seamless installation of additional software via for example + +

    +
  1. brew install python3
    2. +
+ +For Linux users, with its variety of distributions like for example the widely popular Ubuntu distribution +you can use pip as well and simply install Python as + +
    +
  1. sudo apt-get install python3 (or python for python2.7)
    2. +
+ +etc etc. + +

+If you don't want to install various Python packages with their dependencies separately, we recommend two widely used distrubutions which set up all relevant dependencies for Python, namely + +

    +
  1. Anaconda Anaconda is an open source distribution of the Python and R programming languages for large-scale data processing, predictive analytics, and scientific computing, that aims to simplify package management and deployment. Package versions are managed by the package management system conda
    2. +
  2. Enthought canopy is a Python distribution for scientific and analytic computing distribution and analysis environment, available for free and under a commercial license.
    3. +
+ +Popular software packages written in Python for ML are + + + +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. + +
    +
  1. For project 3, you should feel free to use your own codes from projects 1 and 2, eventually write your own for SVMs and/or Decision trees and random forests' or use the available functionality of scikit-learn, tensorflow, etc.
    2. +
  2. The estimates you used and tested in projects 1 and 2 should also be included, that is the \( R2 \)-score, MSE, cross-validation and/or bootstrap if these are relevant.
    3. +
  3. If possible, you should link the data sets with exisiting research and analyses thereof. Scientific articles which have used Machine Learning algorithms to analyze the data are highly welcome. Perhaps you can improve previous analyses and even publish a new article?
    4. +
  4. A critical assessment of the methods with ditto perspectives and recommendations is also something you need to include.
    5. +
+ + + + + +
+ © 1999-2019, "Data Analysis and Machine Learning FYS-STK3155/FYS4155":"http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html". Released under CC Attribution-NonCommercial 4.0 license +
+ + + + + + diff --git a/doc/Projects/2019/Project2/ipynb/ipynb-Project2-src.tar.gz b/doc/Projects/2019/Project2/ipynb/ipynb-Project2-src.tar.gz new file mode 100644 index 0000000000000000000000000000000000000000..e1b5fa8493b874032aca3eda89e703cc7d0fd1c2 GIT binary patch literal 212 zcmb2|=3tQGoEpo({Pz6$EG9#N63MvSjmXYG$1mYJdM? zzZJZ!}kUQ0W5YaBUX4Ip#@U566sdU&7P(u7BuaGi24 z(^bZ2G_*lZf@A~{bUq7Ee0_$47K+O^L0{yzBs3%hQY=6%@cWJV9ePsy)a>a~wT9qaxqhY}!n%|Cn4*AElz4O`YJw26X<0|t0`|1*0>?Cv($E#6Jv-biWrLf3 zJp8a^D0SZy`}$9xzXt}!Gk%@}m|~$`UYkJya>#dcBu0i3N;$Y*DBen8hCc<2Tb!i5eV7d&8H?O zp{hoBDE^^gDEJ`?O7t@Bfb0oLflGvy4DWG-3*R8h?^*WqL5lX$gLDaC2ZgE*i3(YQ zpN$ORgd^XKA~+u-20PWtQU(xxgT$H6$dXUh%t*h62#YmSdEU zotqIlX>h0vB|n8gI+hX$0clk1BQ5!XN$0|`7r`qBCt@g4;OZ;XL(9yheX1!jl7M)6 zu^NRffQ8~UR#UiHZ$(#O+#O6foimCA;YRvXAh{OF?|w(wt3xA$5C#cTP9~GnIB5#L z(Z!oYiN1pGUy3rO6U}EzVJHF?2Vke1q#cuP#S_Dg5?p7j%CP@zDQtwqk+h`6&etC` z0ULllMb)yS}@2m&swl53H(TH2o;Hw&6N zMar)W!av70LWmN^s{Z0(PAR}X_vTSNva6l%dP1kvb>BMJQ$3M7gTuR+RNat2Zyc$_ zDnFy9m~3Dj2}6m>R`J?9R2Ikm z9hqQJy?rpXzPFI2=>9D>LTozXWZ4(%tV&4Qr8UJM3okDv(Ja`V^hlIgaPb~OVj}Q@ zX9dxYA5}~MLM_8@F2~PAa*oVmdJxL@Mw`)2?35YtXal_3goD|E@?HvejQ(X*aBILgZfTfz6Ih%_48N(0R0(#*&UgXNv(6T^B&3`+7jsmtQ_8s>5P zh;f&>Ez&$s9Ni@$-tJzg(-lDosF12?3Ij-FbU#?;pUKpA3%F&4S~S@C=>0c>-L-5n z8gW{%R}Xbjd3E3{Qd?xmeJ-KvGuUC@zF4v zvYGlYtbEnYqK8ee@ZO0sa9G-40Au5UcqDpw(0xRO`d8qhKcfg(T09?A+ii>J;twnZ zE286^^7c}wVupE?L>Uz(skn1=S|rZIX8%w}EAF5ZKYu{4htA}zF5S%YB#^$EAkHu=36TBXCUg&O42>a`sydk*N%(X(H2TXf;MfpPu^K9(Z6fbU7 zAZC$M;1)r@KP|K=}Hx7O3=n6132 zZiS&POUClsKWjkBj7f)UPb0hEx8rnwPtFur;b+^Q=bo8mr=-dqE=A5Hzw+0nbv=6- z;IpM&G*0!DtQX=$4Tcn%%ApF4Y*4^-_BKSOilP1V}hwiB$=@~1JFdJ>|oE8VqG@Wjv|}iC0U%fk^lqV zHvLNY@z4lPVgh_h5g#6*_t^SRT*ayFcOSRKg86l|xP}}?)LfxS z^TpWd-kof%Jocd0wptarcQ1e+SVI+eQy6mxv;RdU|Ed08qG4tIZ?s`$;^O|FqD{KC zZsJuta=@NpamS>P1~IuxCWg1vA_ccNa_ba`aPqOR?XT&A%hMp z%w{B16V(k9&I-y*8ZMlQKyoZ6 z%fma443A)J5n#YP;g##Y@H!+%`c2J-4=-kI5KYrZCh}dk%E+9*1t#iGo9?Wb8~!TG z<8o6{ZUJ)m5e3$cQt8Y~nyK4Q=FWOGdrc-Wkm06%J6P4PW``MOK_=ugjHT!QwHv=q z742__HJ1U5ewa8n{@_yCUzQN)z*aJ=ni&wt>QFvNcYNp(D9XF9($-CttWX@HAA+_b zQ|v9Op=TEfg7v>2d3p{p-WpkY9q-P)FzwqXhFH^HaYq_oO>TwdP%(L@U2ru<$1jPL z@R22ZsbY%gmGLy0$~9!pG<2A77(k9=9nyIow#weF^OFNHhQKYQ6!d=ib!#n~PEK>NIwQ;gYVqHkN_aTL`=d}Ga&^V_U#Y|idnKHzFMvX`ng88!uZRQXZ`mzZx z>EDa{RsXZIi|K@y96@d`uenP3uO#+`mx5t_V^Lyru7;V4<;Pc?x6O=(Mt@x4xji2?&+2CW#>0$4jYc2{ zBDY5bstyl=r*)e@!_+YFlsj?mjAr!XANDr3M7t52^az|u8f)q>vWxn##?Z0YaH{J3 zmqjMr2#q=gT%c195O`lurQ8D88H~r>jIWErMvZ_=z`Q6+Y%^mby3#Ds-s_j3^0m+G zwfXr!G}=63w{NyB$kUmvF2(T7?Q!d}11A*Ej#+QYs7^C1|MThm4p@RG|zptdcR9Ojg+tSDMF3)TWYPunTBeP6_(4p}h$qMXGpQu<69Te0Q zV2U^TYXQtE5^94jqwAGpk}{#sCt5tny-tZowEmghB^ET*Sj-n9q`huUkFTskGrfZ?2&QmtqN#((Jsm$d(=KUs;JsaatWuGmj~~e zB^TKQqA6AI%iP}NK`E`Ln=Gbn4LG_Gq9tsx1{Z5wj$A@S^Hh@VF_YQ_IlqE!fsTkV zB{nZxu8A%PObb%ld0y|=bn(o*(J)J*F}&hojbb1Przl;fmxcoU$viQ|=3n9#mncQs z!6lH&=c_O_%dwiCVc=3irY15>OOCds_|l7S>9D`GI{tmv7X{^@Sm~lYXeH>H98@0; zON0yRmm{b$*d@k%KreKIV){P3PNC-w#Ik5Yc`a=d>2^DDu4&VJF&eW=f2_Hg$b-H3 z^PRy$OjBXKH+vu4>RoSmWmntvy2<5G^JyCn>I@5lttlvUp5OYaJphO^Fj=*wV`iF4 z;5J@O`FDBud=VkCp;`f%fy_|Sa<*$u7S~b9^O{e3CWn(FBh?5V?^Qpgc$9KYPW*WE?lW3hayCb@daFquSICdp0uTD z1lBl5>~MCuC=)zyBz&D?{;yDqM_uy_PR?85R5{G~tZuKmSmJG__-3a*{4yeHe;}eE z$kMkB9XQSCh`<9+UQTXeG(ab*-_~6sjkl)MBVY)zzKOCOn#k?Xd=R#&6`Gut=^heN zltwTT7`XqOrG{oprSgqmL}(=Z0G$EONWWq7F5g2z31ebj308@;y`Ef_r`2L&bXL5% zmT$%shHW+om0TOyP+Z3dP0aBM3~!kKfZ&!XCzEs23*8{T?_5oyWlX1i*lR|&#iOV! zyPWBNSMPdX5@1q`pa8Ajm{M;Hjj5q1oE{~a!BXPsj9$ODj)qcZcUyLuhQX`G++DIi zhLMkkXWPD=btmfB~Hr$)w>!kTBf!EUVk}D?&QMm3l z>9&w~yQ|?;Z-d238yc48h?7M2>2bsLhuQD)(nX`>(<7F`>=mB}@@;?LIk?EDZEA?h zSR;=u483^0PCL@9ACQXw1GD4rsh$AJBBqSy98&cBJva+#OEVG{#~8&BmF{1$ND*Jh znMN%>0XX`$DH1mc-;{t=@b&b(Yza${Uf6)utd=0&;iXk7M}#=Gjf)#4&Q~jU^x$kn z7?*iu`l~Dic+bM0-5~2G4>GD2_ilK(j?g<=1Lgcqz^IhxF>40o3#V_UTRL zf0tX5)KWI^#+g-iyDMUV@6Vwk_S|(}7+qp-QzmysadA+rM2m(f4qhEUh4OGSK{O!0 zs)O8NXO(VQ+O_VsH7SdPY9L|yr}cc`&*Z($iga9GAr2(8SJZv_z6)UjI0-S>4ioy( zqA5{MP=xxD=Cs|@+vu0v@g|ZPC`=3XLf<&jV$!E;t-EJU{OnH;MH=V^t7Ha)6JThG zCNR1DX0Ka*FWvtPP)Pky(iCj1mZEw3T_|>he7#676HYWc@RyuT^slYLmIuvH7I@a6 z!>*@O(vmqh;YzRQx9oJfbHFHV?(yWl-%RZsV85Op-Q`!W*`L!lpUffZmCE{-MbA5U z{3>(2ye{17aZ!S+!G^AI1Ec?CN@MQxA@}Q&jn4oS?>uQ-lFqXi)wOsvY)S+GX4ov# zs-CVyfRp&LHdG9NFbQMK@r_in5GHD$;6e2JrM0LC)^|+oemEG)<3?4ITs>_~bic|{HgfK4Vu1=EewAGh$vwd-viU$x6rZ-Py z*G(@N@XWaR4>BPMp`#m57lJ1$OM5pSc1+!Mk%f>|fpt2bL@#?VPh2jj?hNGndOUuk zyz-V=RjvtFzRcPF()Bgwh>q2AF+)bXRDC_&JG3o5u9gO_bjy+RNT0nfT`UG<%wF;r z(8(3t?X&>cbhBT$68Q_vDYGgKcN7qrpO*ViWgF8kQJbd37#l`0X}{9nWJ**?f?lz} z?s}t84ClvctAhZ)kT`-3T>mXF*tq^%V6d?NUwEh5+NnqXOJFFelhO%+kezol9sN`8 zX;|^8NlD9xM1Yo_Q7RiY9X~QGbZ`qT;o6ip5ZgXKx`Iu~9g%Pd@S=SGycy|vB~LchDK%VujbtR2TEKdBs7+Ud%4K%b4z?$pNvGIC`!i3V|VV z9!l^1M(0?^(L*MBJ4lXZ{bkLEs?xYff2^3uCG{MA>Z+o^hn&Fm`f^EmVC%_?Gj4rp zF~DZJ=~kyGhY?lfLI}`KqWVA)pDx8x@@i~*R8{$T=;uxNtix%tM_YGQcn>p`6SC3F zo3BloaspG@hg%WRN>iHY!?$$rbq!(Abk~C_|70&@rwB^eBg!A=?yWn+0!halCC~%9 z%|=v92LhP;<9&Qi?qbxuuU<+jJf;~EpDEVzW{vN5&bm_BP5pu1ZsmZZv?uiNiP)sG zE?ffiiP3dW^V3#Wcl8MgBfl(bd@)uT06N~*X6e&CPXV0yhi>3?DW33NZra2ZF5l7h5!AYqdJ#{*hbx6_*dlii_i#UgrLbwlD+4nM5nLLB z5}YV`9Ni$e?&Riky1B@FPkm+v&p>DO>(`qH(?*=U#+l~0u^Ea#O@Y&{H%HmzQ(JYm z8gy#Q^OTp$8^$bcNGX9;(6%DX7jaD0PR|HfgvzgWoLw#j2%~b|_W1`fp4uYV2rGCc z>?AUF&9-#-;UUJ_gsh#%_%D&6!cD%VQ560QLmi5p7L%Bhmy2wcmF0P$ST<+@$_!hG z^CQ!7L3N8(F~Jp#brkDmvg^ZQnyK>TTp#`iXa>@GWVr|=8$brBLiEd|0+T&Dx|WF0 zX8>mb|2IXlSf3NVjcOC)W@uJ{Mt4eDm)2~&I88qS{s2qIm8-&FO*NDsvoSbgWE^)( zycd4_O40PF+~H)hk{zEMa~H?xKR4H$!DIj_ud`LeR09Cd5WnthGupeNS*WpWk4eW1159uP3S68lJt_l^z(HQ3g-dxP90{}b`C zic^>})NByK#p3kDOw1Sp`E%4+63{ujqzbH29+wTib7|Ku$WfsVj+XiTF2`2Qd2JG( zhWqr;YEgPM*-6`3!BA=tGQGGU+FIU9m7^8M7 zGjjdKl>?4dg%uZiC-x2o(~6SZBwUGJP7Ixy!0YEC=&*&8k|=zqt&<^$vq|fuok$O& z{6yM*e>sSdxz0E&*YKB<*%{Q$UrHNBKIVJ{D_4j9LY#SB*MoE^xL%im{O=yQwXjRA zbtT74G*Ko@q7m_Mr)&@D=O(;hUid4Z7P|n%BD}u?uq%4Iz&Ij(gaqU=MtQK)1BA9R za9+}`Q)_-AB0r8<)g88%sML1jCJ4!5;$~PhSkGT6&NpJtC3FxSb)$;~nQY--dg5pw z+n47`wMWYESlaQ*J$cLS7mW8aHRr|wx{8(AcH>rA>}t5> zgEQ8G{$z~9lnOY%c8`pm8{UWTy(wbJEqISQKXW|*kubhaj@ZthRlGRJYP;xqp}QzF z!1!EK7KaD0Xz@Co%Ra8-Zp55$X_6p#KXfF`gd!Cp_;Y&8o9!RkG||5IqQmj(1LG z(nSC9twWqRP2cRJ^-tIDx=fI`Pd>OBn>g7ve@G7Odt=h_)AN29)Q8DHUauW7g@eSQ z5@6EoR}QQ@ZU;RG@xD=GaLY-uJ6gq=!I}N=s1Hk6Rq6CXgyJ+ynn5LB`uruv(Gpd( zz0KAGx%osUF}k^}dg?NTg~$FB{pP~6oy4^@cxbtr8aU8snxS8e?7CEI^>i$J4=M$_ zhY$isF4M$2S$vI$b)+8ompTDR()c`!zUL6RQ-1UC={zDoBj9N;q|KP&aS${4c}QxU}{+5HejDs}r8IQ!UV z78A{1J%H(gCNf8b*-s*NZ_eGHC?n&O9Jj~hl=(ygk&-Tx=`&*%7 z!js%_5|!aj8~~MwitPcvzxpT~$u0NR{1q;_!)m%d*+2|p4|al}XXmd_0)IFU_zQqg z8xEt_r__$dY)<)FRhfYJXbhg?BTAC%YbXw{bw8CEwZ_mYFR>^dCCTH;TZx?eZTx-l zIU&s>(d6Tlq&RoZr*cWm>bX{$#!c!!q_HTS*oO!VyUQYK4BQ_u=H|jLN*FQMOE5Bf z|L>Z16e3;7F3ID)eSc%ztkn8^$f1Z?EOe@5qTj<82l!6cwOZcQjG0X|fLqKetBmmn z9m$1$L>Qr;F)1NTbQsyq?d%&a?(RbQRo+Y02)N0(zWXpml}`28MK4=1MaxlO-=VI} zPC!$(KhW#d5XQTG^uJM)gZck}nruw0|5Go!PfIrOxE;Cw4sgCzbATUDm4ed7X~}V9 z$G{QC96~fFOxo0wtt@NLZ^8E~pk7z90(fXZHChykrR$IR_Z2f`(<7rH*eg5NAL#An z>C5<}j*O|DQ%}G_TP&9s>z01Or%-Odzfv10K0rZY`1^Tq_td@Rn~cvGlZtjeB~us1 z;hb@0{!Q@rcfVlGnqcE!{EVF1VP#Z={#xQn-Hp<}Tc+-Ya_XH3i1GY#3EhNc9Wj1S z#&JZhb$$*H#qZ{NMT@LS6j=J7MAeOb|DZkAhV%qgQMnO0RC(X69vF)~DeBTbe9xT^ zi1U9!X_|Q}^6Scr4WqlPKiW725V{kVDZI$9@Ej+X5`&!If@S%aS~eLkl#H{I`>A3L zDj6Xu${Th)yYW5zi-yWL1^(wmpWbY?FjSKQZ9)Yd`ij0Q84c@hkc_ashw&`d&zxIN z%&}TUEw7*bDT9D}2D7@TCqvZg-ov!0Z7E7kec#cHD4hwO<6L((d@b@}<3`n;_i4@N zYW<#ncod6g85B3NfE`VN61;1ntvskQ8r_n=lqh{RT+vbKCg4e}C)KURS31q9-d_uz zTm5jQM@G4`Yr3*8?tfE8!?{g0aFK15ky;A3YO-yNsOaIGGcfk^c2+dE8`h7zV*4Xq zwI}$Mztx5KPYA^8Q_dToi-%kB{c@Tsl<*TFY_IexQ6MmFnTCTa(F5nteXDFV< z&$=21Q{e?1wK#v4c9^)LXb7V?UI2AN2~aIVAoX_Br|xd(@o@UA#(cP5)t&@d|5iMx zA~Q24FxrK`}Q&Y=<~tn^t&&|Z))XW+`|6c|*576>aM!oV;gTrCTpP_!oPQ zyqQ!~Jn4i>t;%&I1h@Fh;J9v*MX&)xU6{(M0x9|`-CR9knn9CCdbzIPW#doq2*Gtn z#w>~bjyO0EcyKE(*)X_%uv#b)M4}4tiE~}$gZpw@oJBJE;#ku#cjiL%GKCWJoRzUf zn~}8-vXnd=(Pg%7H__Tb*1rkKo^Kf0a@a;>mfiw&#I;_9v}_k;lT69B@Wt*llGCKT z4nb0S@8>zmC<&FwC_xePIFj#pQX^3<#LGquo&3-3E}+IpSkG9X{dp)KM!;j-6J24r zE$8bR{%~|$8D^R$(hw5dxPB5D*^&dm49=$9-Qc2TvX|i;az#l#%(Wh(?C>-E&%vLz zJ{m}uF*A85)v#5iBy2m{JXCU}C5UXJz+qU~ARIRPK6nEhY=|e&Jz&X)f0{lp-sP*= z6x4XOl3qB{a%Q%aY#yz6rxB{NPLnSulkE`mO8t5=#{I`x-$QJYxFpH1J3R`N!+?@T z`bT0t*8_+^@C%)Dz=wmGuI?)z^JSK`qP}XMcEV%v0Ry0ZhXJ|Q6aVBP98NgUY7hvn zlEzHLNG|PKVCd{#VE1adc1Fk{bFT3bBYj|t<+oyvP*P-uRd(2x^w1`8@0Fs-FtBNt ze;j=8l|PN=Y>06uNzWnjpoN*`R@?4jt*h& z@3~BKVr7I!FNclEE=mS(FqOgxYjE0o3ad#OkN6B2If5d!ono9hSOHggRg~4s4L)07 zz(C*&Y5WcrVP|w3Pqg%KzO!rR*`mjyl9O_?ilq7Id^IXq0M5By!rD$mix&G{dwJOO%x*e5CEZV zDae%tqJ3IV3x#?W=Dx(7p``-niX|jq0%RC{)rDeMp;}48cj0XZk?4%frmp0Sl&+71 z)EV?w2 zpdulIwi#(i^!i`;){uJd(=fJEn6km;-n%9#sMMT?1+STV0B(FvL!wz>X?mX)NQcTZSnT3YjZbaU78AhC{Fz zqkz!I^AWys_s=lO0Rs2n4%_T7uIk_+DL*#@yQ%aB7jcWS6WXgR1L4ll71k0b>ZU_0 z8fcZ>xKSkRnXOUyJO2RPA}l%2Re@?}P}G{kgEs0hQ3?GT01W%gMl{b*yy~PpdPS&! z&E-pZ;%^Npi?fB3QRV662`r)GP=-ix!mYzy|7U{#6M?WGRXrqh6qP)R=yQVN*VOdz z8c&40Xh|Cfe4lgJf|GMNn2S8~-u~MSCi4b^7&Z1AToj$%HFv}Oe;VzzpkgVplJG2o zcjbh?NS3voN-%mT+p1Z=*pN`Eqdb;;|6pbGfE$0nqKh8h6;-iAM)8f!MtA7r@yNMv z{8xGB0tas%h^4Vbsf@HsnEg>g??FlLI%*fbus6U#yXnZwkc*(*|6DoXArd+|+8(lb z;7VE@P{^C-Wq}~2AnK82Pq{m9`CnAagCC%=WQ^B|L5RS#O~8h|rxMLSbc;J4H$P&& z9(|?lcAN?KSA^9)kx9nU@NKtZGf=x?t0J39Fnj0zDx1`pI-iLXK5LJxF6 zq?!v62{;Hdb%PWCs?@E!nbau~!MgemIN(6@7#p9xXX=SmIMk>diAMix4rK)b*Y(qX?lE=_C z;~$`*5#S>{9N7~X&HcS25cKm&dM`#Z&#;J`m?l;ZpGoRzA9Q;+?bE_Ce72q$M`e+|iuMnw$*IeVc8+~e< zULr&F&f9K@*YxF^B1?eSB2=_Sb{~-nn{QH!|6YD5ON%2&2#+AwAui~Q)8mA|^{d5&jC zP)sskLEK@UEnVw289qM$ttS95sWAUZ3WWn*%JgOeF9_{%af>i2lacZtM`SNgmwyYt9dA_kn(aXiiJIkf5pA>)Z2I! z=Z1c|ZSh-L{W6LNDm;$9v3GrGyyzfnd)}QM6HM-o&-3{{x(TdFi12%Hc0IlC@!@FL zP_(XnTWRsOJxJlDq-9ep0&r&T6HuFRsFsU_xB`3rTBcpdsIoGl##XqSU5@jy2hU(C zd4Y^z)1CCP_krkM9H~@=^*P-*v~$kCkbWpisKSbH;V=r*rgO?Rfzi|L!Hwm_;MoQ= zC?~Vh7b^d)wJxNGIKy5~pw-o(qQWUI0svUJcxJmTC~33tvU%j>n(;9j(KfTmlT0>Cm|d z#R{C+ZGByttQnc0sR>l9Da}GC0xCKk3nAs^k>|J$%bjms?CLSf)P`p)VCk&XL54Ug z<+$*6Y9YznDM&yT*a&t*)&e0l+o}^1M$&qk#eRMpGXW*rxt+C$z$SJqF8Yl*bKr+@ zVI;kYx*!k&6!LxPd@dYdF#t1OeHHcr? z%6b_R*GJM4$_~xmCOgDWf>A_Iz(bWR81Qfbxn=8Q11G#1hV_)hVL! zE=$avVj@G@sTNj+tYn4fpZPUVZ3z|3pH9M_X|FUTmHS8&;x#;cCad~-y2Rq{I3qVv ze@9anu+6D(CD)*=n=u&;EX*6T`bB^z@;6KXREMRj`Z>Z`$_1mWB7L z_?9TrAv9&mTkU=SLNwao2bB76fa2=Z$N}@Ji7WmnA z7Gq_D=`49wm7g~*%1c()Hm~{9$}m)p6Y1H!e)>h@K zg(ye29Arh38C^U;4B$gVqKpE-TnD!;SKY3rig!hB^#GNetiz&CVF=lZ=gH>n5YGgb z*I$O04`B8igQJV>4Y-4)IoFfAdyb=hx}xP7MI=2%kxFcKJN*c^_kNXX(Jma&Yr~L; zP$Qe#k4Je&@!VVZkm-L0!~*vOX|5k0K2L%`3Y z&Y5PkB)l0+W=+JV`Uo=+a-N=Q5t1b2_c+j&K@4hKv+8$9RhzFQbn3%WTI`XqJQU~; z4%ltMf?LFf=7N*p!n-vI7eynO_7k!0!7!bZolkf$sJqZf9wP|@E8+_9&?Q1bO)wFP zG&mbf7nm?Fk(w${!sa?#K;-j&_&hgfcs!m2>{yvx`yZsk7Rl&glY{>CA}EIXf98V( zDaQJZK%_G)i$CML6sqMJ4P1TJa{5tCFoTB}771F;#_z^*>8=rO-Wl$8_SM*wnJm8e zX|3GrvBfdkNQp~5f%aM;GB1;$O3wwr=s)t)l{XrvR_dOAoRkT(Ig&K2vhoNY`r?Je z$rkOU81hYq5Yg!No6-TRnk@?28p>4|T3?Qwl3I_2LsdaV_y2}7w>BhcirP;E1qnUZ z!+eRB#J}Ox6jm&BgS1r}iip#&C@5gj%|m1cy&v4qGXrNY~GGc#GR~F(J z@)nzaTv(S-Se1B(r-Vw;AtGz;=py5g7BcNc>10zi+PDOMGhj7@LqJBO2|3P+N8vZ8 zY-@$xc13-0k}ys4%4_yBhy?G%vjwfaqfWJp?1E?iXe4!2A2JeD5Gv09B>NFM)9A$H zmZ{s7h-i1gN^S;f*E5r*f$S+E_uiES;lG!{!DT=Mgoa2H#L_Wj8aO?{Q-!}ue_Cn; zvt)*Z8Y1D_cgs>n-NKdPo)m7k1Cn|D;Vq%@z=-u)YS05Mz>2~vMeTPmc^dcpj1e8Z zEqy~7Bj2%8h7E=8*c2_FD_)UQ#!Z2|5~rl42nV2uXSe(e6tci}nG>pN0UliLKgpAI z{NKJK3#;+qN5?JjuO&>tS=ohewzx*JC$mn3$Y$EN!aA-;qlP6YB?g14UuCYu556el z5&c__YZHhAz!X;~^8XD;TrB_XX)$wg{a<)m+KF2nNCESJ1zmHa`l!f;N)qwZ=SdLZ z=5yI4$a*o|>q{mvZ~fOb&j``YjNT~-ai#0fDdfZhygk1b=SKL${hL^?htK7m*i1() z95m(RN4Cx_FgXfk;#4B}=BQY*&e~8#;#iHq-EWU3+#eF`IIi-8$PPvjj4@{maylbv zg7S?qW_-M-8phgU!1RXkP-Pp^_1ElNYw=$5gdqV!&uWfkv5RVp$w4C_t@yG|sS(b< z?|`Myz2;)l3W>%2Z)clx(gNM(eb7q~Mykj1LnFw7)ZHHE^#t>^ivU=gGVk?_Zx2HU z^lcldW^9RH>+PDh>OUIEMGTcFlGp^Gp(J%_%-xZFTew+gAOcl7nr~p=orm7%9vS`{ z1eE&OxBxR{x;8Wtl5X2nWcQiwBK7MS1SUs+D}_=Arze z9{pNQ$()-%Fp9o|n7YD{HTLRf49<~*H&qm3z`8b5+GH#$%^E!(gZ7d={IHc)Z5b+I zcgj(i=9RHpQm0t(UaOL$=K{$HbMdb)qv0c-)5{n(y7rkh0ICkOgyI0?4)D;O^TpYE=UrUOh|Dv^(E0|>=!mh^jbfNI#qb5OT?s9KG9An@=V zgy0#2%Y|MQSkJ1A!pO9?kaEHL$A)}9rSwY1ca-IX7&>&rs}mMEONxwghq42_O>Q++ z5^_?ubCzYkhZGZ=HJ*Hrec!Z%=%dq7gZC*u7|v<9Y$-o6Q}+-5xl0$gCPWni4h8D3 z8IkR;p5|Vvom@PYUR=+fk2NEBCGb3gQ z;u@j)Z#T%{GE1M7te(KTOxd$ki88k4C9N)8%R;?>ZKdJMD;tN8%6LXR+$vX!+P^58 za%O_yL)A90-nk{dB$9+fK3-v^NShm9e(RGO*yIST8Z`BSzbJ^hA?Kjv#(oSeP<;@5 zr~Rq| zV*G{tpc{egk-gcr+0iVe3C{RWvzOY4&qSNU^E(7VK)g^aSt5_SHL&GRY7$h66CD-e zF9>3)7!mv*m@0LsR9_UFMqAxwFVN&;35)^fFgftR*Kz1=2TxcsKV$-*9JSFN8}lz} zQP29g&&l(NHApjkOvltDZjwx7MZM_#4EbQ{DR{F>ezM`@W2l3e2upZp3+aX8%IJ{? zzOT25`eJ0N`z6^xjdYHLjS|Xq9{LcmT5G=HgyK(>LUimzv1>e#HEPyuLL<^?YKw!l z927{jDyDnyT>WGED9+FRVIKWB3kPWyVm zQ~H$|od3k*Mk+nD8e)+s7oRJomO`1PX9IZTCX4ZgVMPpQmu;~a2rhs6{`A-yoOSyT zfX$t0Q<%2S6}hE0z^y3P+jLg+Aa3xJh4fYt zZxH*5PgKdb!2=@vM?%xpY(uo{Mo=aGG>i)b7)IN|BHDwI*0rCIINDUh!M3H&_03A~ zrJ)5lIMoBdgw=(E)J3>PGU4ol^o_v^pmr2#?<#;l#ZKh!8+N5Q9PLRHlP-YGORQ^X zdBs}2Z$Y+Aj(m7|+{mJWiBSe~XE#~5fyKVhX9VXY9}>Qf9RwxboH;+}CGM4CH7?by zxqN4zJ;4|XgvqX!f^PvH<&INC^C4H@<*q<(NRX#$ z{%{-P_-#ddCCJ|n`-DVOomm}%#jrjqw>tCaK7ZVf4ZQE-@cj2v&CK-QW(G463o8f9 z{|=3^5V3HwGqL{f>3^aZ|GOk&;o@du{vXCkTf5xVR<1q-=&1FVxI^4LWjsAO;bwtE zJv}|8JxSx;q&zvJ+@z&pdOKfV@m@EsUhjIenDjgiJo)A=|H8fdilh4wtMZx_b>dG( z7i0%VLlV(e+F9CynOKxywzn{YH!?9XHxVbL#s&pTC}emLDU(tXAqFi^ffN}XK3_sX zG%_(elO}pm^Lmg|0X1F{F*_)ZzpIutnLom(@eSLZFf1Qz)mDN*=K%-!$YGZ(dF5Q5j z+0j?`0@Z_yG=EOrfV@{Vzwh_}5d%9XqiW%5ZD?az*nw+6Mi8AqLkSrJ%RjNWW3;S; zgf@T%3nSjG0TiGry1cw=Z!BzmeSPeieEfW@>%mjF7mmOO)wV`Au+GikdSC#+s0jWZ zFwPY!XSj7?2K>c4c$W-hMlT;kB)RWM>D)7r-VUhubN5N-{mxvsCCv+%Xno(p+lK=C zQN)2Iofz;lM@KgYZkNa)q9mGxGJCW+H#h^) zFtRM_I1@zwjYZ1h2Cm4UPyopO_$VR$#4v!fw17JO(g&D760ZKn7hVEHWxjnxQQmb|*2U{tdu?ccdrRm&bqpZhlXTfxhL^tp1K{ zEd3{>2X9qlbzx@m;HmkWA3$Iyo~xef8glRO}1Gccb#do|o|z>_g>M+d~uiaRmj z2I#Refd}E?0pa@b0PTO-H}o^a{mrEhQZ^7%mKDx16}9m_X1mZ7=`VH9veY2!F*$U$8!P4 zkpv=P1}Q%y9zrn1d}BJZ0Ar2)7OhdS1R_BODZe24#8baQ_Km0gG9LQGm%KqX0b!;8 z5@`UH{vjNNFjD&tH88dL5(dn){={|~x=REu*a{p8p06$)4@7;dTbfut@f7|_t*-~j z+S&iUv9k85uzf?CK`?>*0vl52eh1S59%ku&$&9|?^?!`#KX7JeP7{Y5J$Wtf1ws0| zvwlJQQJUXG@7iWDZeK8j76j=G%rkz20IRzHS;??^LVWu(|AJ^1VE&Y-Gy`bc_?CXL zv4#AA^gPvUehFLwjK4?#FrW7#KL-xiHnu>AGCY3Iv7c;_f2oq+)0>V6BQ=K;I3yA< zewn=DA1FyaZAmC$neGea4E*!{bHG@)cSt|-+mFQT-jAMN@ZVQA>YsqRFLzs(Z*GU* z*WPc~TCU)mx?r4)3cwxvk1>hmF<86XvawI!W0v6wK=Ain@h%W0>9?!?_kf-_O;$Sd;M8^uLPbwaV-r`?OYdMuUErj!ag(h?-)FR z&Ri_b0H8GPi6P-w$axoCz6q?2&V&J7mHw~Z_Yc~`UWiNGWT=kibLB|uWK_l?3DoL20p1Ev(X3Yig^Ghe1%;|4RM-R7Ifl23ofh+{_SAylu4rjuyc~ZmSMY24w ztcI?EdCUMFa)kx*C_XtE%C^P>~d@9??5_UZ0SBQ-LzZ zN5?l?DKEEQHtduqqd&A*fd`SOFF7Z(wRpZiMEg;vh50U5cQ&ypnQdYFnP{aWsOYVX zCJSVz)t6eg<~i2JK#N&Wo-@o`3AiBxJ*wd7m~}jh!ocd1oKh(c;-d1V#_jz_NzC0^$?)TiEaSV@zQu~bBQU6A~$)Nw1tONkTcz}xLx^1;7YNnXtDV? zF_ElE&`ezaO@%^AS24>!p9 z;A+TdG;WLKw2fMXre5Xah^>m43jWv5Jgw)e`G+6Q^5er~F|}T06w(CgA2zbro4~u1 zrn`5a@~No<)9y6Xs(pz1$YN8KBhI&y81p%lR?a-LN($3G>xwJ{7iiVdV5pSpn zD`!+QA`VO?i45_Q_IuI#f-aa`vYZ$2 zAYVq=R$Mmb)${gUGKmS+jjYPex^&D{&%6#%4*Hr7`|XC>3)zY&`M3_fVZ{`FyQXaS z;d4lrbL>NDliMbal@wt*EtAnAFZuS{qW`BrMi%pq3qYfvDCNda`O@x zv3%0fAE+8}mFHIc7{p;kB`R(4u_lS|Ia4PO3>tORE1@AfW6^If(OWw`6*=?Cdnt*dES8TDw&A==X@Qk1FkIG0V=Ppx zI!$iqoUmf^1>pb`kZqnfrOCH%OxRMwD+cy*-}TZyl59lq>XBv$<-?NN<`Y<>;KI~v zgf5@a2SyCa8*LVK2D?dL= zo~KQ0=V@50@WT^J0(uyCe3(_;vLT;2_?YAQKj`nO>Mb@5L_hSniw>!A&pZNoioP6c+dyot4xX&I3#eR=_?6CUXWbG*u!o*3^wPhm@ay{U zeeFjKZ2Szp@9S#WOBu#09Y0J0Ibp8XBi$|WU7pG#Rd+o~4vf;#JP0eteJT>SV`z0D z>H;ar!-lVw*Q%c%JgiWJn+T=*mWsn@di=3ZRx>IEStOQB!`rbl&(^WCbX%e(0fz)>#(W;5Ykx z1Yt40_u|{8ea+EL7LOAXAxjZ;c$V2LY=LNt>!YY2a_&KN@6c*P`g#w`8{QZdyU43~ zorJDx&Z*QWe@RXpdt zEw)A}gmgiQ!iToqAPZAtW#j#&`klbarUhg#ftt~34>Ky8os8$nexhZRbNfnG!%M@Y zy>i#*qkT>oLN#OU>HC)Z6cDgvSUQFcElE&ir9%DlTWusR7SgxM)6fs-q9@T31^oG+ zI%!&?6cwRe)^zz}TT1!cj2@6zY`*KlV$W0E(G#;}w&vCZK0%UU} z8?p_b1xHg%{A=i23%3tGyqV$8Wja5b^U(dNGTfL=dOzAdC~1Dm`h z55hndnXBjx$>5R*gM!vc~P`5 zz2Po(y7BA7jbMyMMN%h~b|#b90Ce zGSFotFH+bI-PMKi*_6%fS=JB{dop7_b3D50{NVJ-wOQctW)!wzY(}2KojdEOpuZi@ z%8Y4u2A+9K@WWx+dyW~b4!lze(%PiS9F=dRQe#1+0>JipB*rKivr3|%fGm9hCaF#YFJ>roc zoeKJ;cTs1-wNz(~BP*c&aaF~Ca?I@tovbK3sKl{gJ@@2Q{3E18Go3=a{WSQzx=;HcXxKBu zHw58Tt9BGGkFlB^ff7?Xaf&m$Fy-*RjMw6t@gj5l+k6q-ihkHRju4Z4GF|5xarn>O z_E|Q($WcMF*0wAe?Nv3Oj;XgY+!lhom8gpF{OiVRJ52s^L2&mP}2TVgH zpYJnUtu4%RrB_Nol{Ym`reZ;?CR#mQ=g42+&t@9WSj|`RL#+->I zdxeGS?E_KGC#*kG>Dsg%>Q8*Gpmw8-^H2+kZy)*pvubOkQ<{l%s!G?@eT|G zuFP(^udsTYs--Et+uS+7?dboPeUg<&yXQg~T90Qmo!wV}| zx{)o@r!^=*jYd&Jf>k(*F|T^CA^Fa1OPl2s0xFN_`q|%SliJHfJ)`w$9QGFz2-qsM zl&k$NqqlcJ(w4%IEgZYcg551py=<=K&3x1^7S1dHL(YNyJb#}^4`B$eiU#{5BgVnz zC(!vh))1~};xC&8$1AMJh+txZ_eeYy+Dc+X`QD2we$QVzD&Wd>_=1*_xdHVTRr79n z_d3^4#YtyON$ttidl#K#kyLl?gTB(bu$I$m3vA_AJ|XJi6p~5}qK~P8{pC$a*|_4d zHW5*BFX}*GEE0~d8tqzUQ>MgG!5g#gScW*yKvIZn>UC{~r@ljT2f>y}{hEH43Ml-b z;1#cWvHLT$I&J4{dpL-w?^=nT*M#x{Bddbx=PZWT}%+D?Lom$)YrnCCW zoxqQPf3G(gS8CDk}0Hw?K;FCn`cQn5F}ynBMLY8yZUan zPu!Y#ar(sko9d%Z^PEQB5HP{^en)*b*GU9Y;0Pex`Ob`?WG~+250};s5Mgv)QwSV`jB=F z51IH;;$f-u8h`0jg2a4(`u^o@CIb2{1?UL2%~n7EO{?A7PEf zMa;xEdyfEcMH`Jdw*>WV9Z|`qVvOH*DDMTo-50^wx`u^q?tJ#Lj0Gr{4G2NAzkEbj zFh3d>vEFtLEKm5VrS@$K$Xs+bYr|BIHDIeTRY#ecqJB{lRDc1q|7&r)+9GAJYMWZr z&=%SpGv4J4OJ;2Wx}c!QC0^puwP^TCdGn4}&Ue(>{j0nJj!SdEs8Bz@L&1DS@Wy9k zv&~|ABu6!p%(o2^SO%-@#hhbP7NzEn1&>Zs_};MdsdO+eLC#TTFPlU|D-1BBDr^XX zst79++ln%4&DR!IP z@Z?Wnbr~ZT@!|!;B8IZJ)hMe$wyAt8Eqa{6lFe8{&GV<(Ta`A(v{Gu%GuuTe<}z&W zkPvw+uhKFf?s3C&iiM-Y+xvEkS_73|ijN%Q9ENhG%qtXDYZ7#oppw4Pd2<~j$4M|p z+>-#FM;3U;uq>L~j*Icbm1B(XhpRQq{wMY=I^+HcaDmPPqX+@q8I@lxu=NpkAhEH7 z43)Va<9qclX&M3@!tG~w?&J9NBH6OV?@%qn;H2yUWBD8XvuDU(TbIx@ePCsozC$YI zfi$7)pdoPIQI8@c7G5YmB?QTOHW%HJ-nkm5M$N&J^`St^3b`N4ZE60X%LR( zytMa!b&0F2fe_I^F{mRu^{LTWW2sqsWFgHZEAn;H=MuPUv+W*K9;)W@C6dLF8|lAA z{t9)h7;HXxB|R~(t)h?kdh085V!pTejLNG=&>inlxQ_${@7lvN>l+~EppZXFrQzAy z4n!Vw{K!RFjCG$iWs7@zai(x^s<8Lnh{OTdpqKt(=|UFvT(bV`1x+9IdjFacb0oTI zCAE-358G5ACphW{xzxv21E8mFt)82-8YpJlL+u?7k__6()UeB7iK!7;Cvs#%q#!<= zp>`$1Q@64dxz7lnB;#FZsYN`uGr#!QRk~*qns(Z5RaYbI3oyQ%=0y2>;T>)_V5`)R z*Xz5J!Qq2KmoGnG9y|pGfsDQhc_B5IUeVE|qH>B<t9APCZK&;|T)xGziGt%l%c>+%ZY^5vbT?NO$e8E)KZ;$@k>jUc}=KuP6bg3`Y@q+WBU?CWg*Cx z5-`Ir%z_1zb?rKWSr_jZX&n|ue$^s+{s2i-lC=VE8`<$bwmdg9A@TgIMSDlk5zpm$ z{EhN?c57t;olV6vK;)j6w~o7Y3JqAx6bc8%GU=II$FKe(kA%+s{W8O8>~`5zs*HTA zNr;h0T>u6GE^Mg^{%PumhE-;*rePih@{!yaYfTv1$~(jFfDlHy;w|!E$R$BKQD0J1 zn1Eq0P=6HL!oqTI#T4sSm6#wn9m2&1-%It9OnA#;+vwr#;w4(Vd5j>#;737ImKdy- z1QM>Nrb~%VgZd236dBn<*&4AMp6g6jg!=UmK4#6X`LhI-fgcCn&)ObXynWR_G7vI) zLi_tFbXeKZFAzY^Xr^zam3z1zXE%x-y2^G@8bQ>9QK*U()!?lJxc8s61{tSY858$q`j6kXZx@e0btJv_NZoP~dy)%%qT^ z8xT^N{G@2`GLcANh^|xG^0YHpDGmycV7dG3_WXGAc`a7|xl(mjX>-K4=Hke0YT*gP zNx-@{)i^tdH~`P+#kX~Q=)i?4G1?Y6IOe5M|{f_uQUI#mzIoQR+9dh`?bE_!!R!GYe12()MY zj)61rfzg)*MUUK@{Jg`67OJy>oR1&U6nV~3YIwFEn!GrC5bo84EOil<%mB?J%SDd=4w!Vi5peF`?vgV$s zX}W{KmWoxpPj6Za#KcepS7Y^?E0P7A2y1Xq3erDyha(85TpiZV!5nUM@px^TUmU&b zP4$oLp2SbOST3B_3QieV`_%OS-7S>6omr^?Rm#LYl9HzF(nQ3G{gSJkrdKZJ1`;BP z7?@Grm+N7n=NA)i3)}(65Z>)bfzemKX=y){_okJ&}I>(a2fgI5VBOM%zRx9{M{SY z`Y7v?NX;tK+`nIO8C%kf&CsB3%`fh9T-?$Eg`i2^Igq~j9gVAD4-fg=&)Du>#8Z#& zHA8E$oxDcAc;lLstgWblL7ETu#o0*ly|ucPbA|aZy0d!9h=Sh?I&lgG-Q*QirEZak z^Q{zf94tXkR>cMf>ys8KX4>sQ3oU|W_XXi6iA8g-DxH7^)yT=?Qh2VZZ;mTQ+;F`x zKMQ6w2fflxab_G>2MssCji~$)(8Y@aOP4~QO1wC{P;A_>7`{ev*u@DrcqQ@Z9t>r9 z*6=6pi(j_5*{OexIFgFIbZ_DAo3bz8v6MWOWNyW3*(Nc4E0$XUB*?7f5z-i=Nveig zX#u06SWIA0=r9fr%(#!i)rJH|Qi&$$!_ zo@GPIh{Cio^QE}nWCu>*yJ{J5+ z?Q$^cwUt-Twxoy7=22-U5pkB)L))BS`Y$T|KG5wWi?sW9StO-Ttn$Xi<)!bugpyRJ zF4}i7oXbtrlRLtA8Ldf&Ql~j@J+fJ9Ee(C>uSS2clu9Ayl#MUa49{YgSe2g;#!UCq z$l7Y2TkvOm|AOP#vPXp`VYc!GflaP|mpKOt1}{gy6=EKKHEw;oc>wuva<3uG+#w))!9-)j!8X=J{(Hyk5-wrvf4hK8AL}H!K?fn_&EII zOGd9D>N1_UotbLz25cg`wYoIi{E39T*IrGGZU*c1PTbg?iRJq@**yu?5ZHHiVI_xV z2&Lp7)ulNO4apYGKHlQ~y2nH04e1LvY@EX_n5$M9mrUGb)k7|PSq(l0yc_Pdpf(XV zVY`*7#SRP1L`s7uT``L4>gMq%#(bFhG?zE(W5gj*CQv=7&dweNqN1h|CN*u`?h3{n zLULbosBC4GVX^3a>`wgoA~tD{oqeu;1XE>j=C{c4LC$#sc!?w=K$NQf87|YA=ACk+ zDa9BV7Q(sN=i*%+wRG6X$Or#k9=6qOKwPU=`A-a@`KQ6gLrV6RlEpsiHyyU`1a{xB zE!mzesl4M1!6KQm7Ih>mi-GoErS)U?z7jF^^hCj)V#-39eGTk--69f+wc zqc|@%GvJ}>o|BCINm{EykqPs}zY2L_@aB|o8JNubLx2AipDXls=Y%@n`xJqkQLq^P z=esi#yg9pTIrWFLiHO7Zy>|Q&DTLVk!<-RjH5WY z2M(a3k1R4l@~pIl7Ya=qP&bTj)e5R4fuQwHC$ZA>euz#?w{V7hPVQ{azGI6r(&9%r zh!%Bc|6;f;*}0ZmCUTiWIPLkA_tR&FCu)wK1|eEAJkOWWax>V2ggVo36jTjb)FOtS z98kCuf^PGCW0pLGT)&}lOO0bV=*#uE<|-F&+bVbV^1VMnLfs&9cF0blvJt-R zaAK7^PFfk7F?(*6PE=v?)L<-HZTUn@|31jW{zG`3uM$eh#hu!BUwGM5T2!L|NU6^} zkNITqSMbu*if54NiHMIxY;_tphP+GjVz(<2go7S%COj%lv@Xa$!yjpfZCz9KY#2Fe z?qYKk9@e_3wxDymB5n7zgU_Di8d~cPhoGM+hnkKpF=F(l*imT0;;y9$jX!aqmJQ9Q zgL3CT}E1j2>`(JA;9h;-fG}xNf+x^A09M^Fa-C#U4f- z_~m)Hl!2+d1hNygonG_1#$h=T|G|j+_0(kJeZM@NidT1ytYlC6p-03cb)G9SmngoZ zjVD`G+!vCjZjXZ@Zc#UI&>8@1Q$ngeH>*B=OtD#I|;53X)S9r^A0 zpy%ug4(x6;usRRZ1v_|EHJv^bohALM*XKjPhie2}|;1I)4cES;7==e9ZU| zeG=I}mvTjJ`;-2fKeX!2%mJW}_Hx;wmOIEoFyf$WF=S3G2mjY)5V4I;e`dkZCz`xZ zy{Ilv!YJM*L0EJ>LeAQ44&X!k;^c42y2@{B<6b7NdzQE`t2nqxz4#1}X=4+lttHe)7a5Z_Xk82v^@U)zgtu?*% zMtF~pg|r`Oi<#kQal9t%-1(0JWhilyhLZ~J1Bm4_Rrph?aV-0k~M58(XdaoV08 z`204w#1+wEX?M0TLJx2o2@}>l6V_c3~GrvipkCwuUf*hZ1V&;KO>D0>EGXEcG&77 zwt$JyV6aC6vsaGgo*{_vB+lnT6^vE4jyW#_#s>$apm>9t&T>bxNuUw3 zygxc_@L!>Giqb+S`v&k+;n01CFSwlc)c!1>(9AT&#;E}CPPwyjT(RbbB7lV}<~oFF zEvk(p?t6RMbzM9j)%##J(kFv$Sieg-qQuh3y|B372prrOc4 z79D~V3(v4?wv;tVH-^rco^Al^#x2)HsA*;*?u1)>f{(1#9wToXsW`NE$#^Cj&GJI` zyUzoikZ|!xl=LN@=D$4RPo`W&X|TY#MPKEZu<>k^Exj++d;_wYf`I+bP>Z#jK~y)S zE2_|~ga6QjrB_?v-!<0ida4)8*-EOUskx9^C@Fe5U=A@k)9`K$@svXFc7-vd&j4lHi(J&CGwov<2tx9v?5vu)C)84bzv)H%ON|YkF1Gtf`1WTHKWCYF9a*IHHB*PXGZjvjijZ^Vy~Z zO#Rh$m`^Q_v}(y;5Uqt$ke8h%i+z($@cK?z${pl^2$w+#c6VK=0E65wOGy0d^;%*{ zAj-9r0Dr@Ndk|Cg(b5>qS7kSH3@%fRc9FKelJf)7LL{y>63d%-4BBVOy=OtF&Z!v% zd*9dNa&Fq!hM!o*hyirVaM(TE9{wx@pz3|-WXS4k^cAAf?KUXUN|%3A@Nl9_s zysMaaw!i^43YzDiFir5s+$;iO(~6`8sn~f5)d>&!FUDx@7$TRpYilp3oHE_~Na4n# z+#QYFwYA|@Otqx!ZKSMlD$1yr6lT$Y=(F{d3)Sn1%vl< zouXWn*V^#_IoLqI>!zsvL{g!!ATTg~@e4x1tPW)W`^$aF+k=%r5t7ZWkK~wQwxB-Q zVjJkL-0V=;gJDs6XN^PZu5&igm}(OGtzEyfsl?|>>pSpkX^jx@D)2fZN{hh zE^qvqL*%lL1=$|BDdiZ3P3sv2_Qt$5o}rlwHcZIJ6S5RAOzDn!O{88k?YdOXK&vt+j?q%440#t7_`p{u2Kc z;~G60^K&0vDih%6mS>pb%wczMTP-EXdnvS`aE8Da>(I051MH_y!eNVpU^g|>ExZf9 zqE^9R1(^C=d0tKjfD2`+IbJmxbLu|j+$fnw^AA%3ll}ZG4*Jj9Zg-TU>;{^q2(s0) zBWgbNRnDEl5paI~FCeGlQ*T|h@*gaX0ne)Etvtc!n?1ee?y<@w)iF+F>9ERC;FTce zFG9$>l1-G>P+zLfmy;3j3{0O`;0NYB#hdM8gEb4Jy`&_QZn{q4a?6Wid(&94?vleX z6>T0j(tpJ7Cd35x>ei7EQ@YCbBEwO++sGfVdiARFJ~ZbIW_2?1-Uz@(mcuYbF3^_X zNA(8m?c4z?V>hFJTKU1g0e!0ZjoVnsq(&9cw2q1NmRc+!Gr_qO70MB@_ELzgO2u`YSXPL}um zsFqii>Jpja-Tbq+bkkG~;@63XucH_{2VF?~egqDdiZEy`$?aU|qS!s!m#c2k72lyQ z3DKJF4yTugH4*t4T%aI21&FQ)7o=yf!|EP*H}yFtzXm6P^Wy>@`aU+8YV7gVfdOl% z!~ScdhAKM`;M5A7h8ZT`{J@+EJX@V*-c!LUfm_|sgnxa*RCg4Vjs62m8g9t)3*oeP zj@&Z0u%g-N7#h_Ex78b}dTLwS%|)6ODu~kW$e*E6hm8!73+sqIhKw=wKl;nN;S990 zrhodX>^>c9FpQ}9!W0JSYOk)lBN#{DVChJ%Vy9>Z%VBl1ivqP^Xds;ywIH@0^Jd5m zTdad4rN1ff;!Z%{Kz@~(MYBuNUXm!bkQMC54aoLge^=F(`u0j5XK_Cam-uulT7)jO zdD62cjdjA8ydSl?I&?La zdiR*Rb1f0m5fmq0J)84@1#7*G0FLNviAd7=nK=V%`D;VWpY9ZIvXX8L&-Tb`g5ClD zeQaMce@VTaxS07RT*@w5e_(`IZcM>*PfNtsjfP3koO&*>nel~EkA&mxdiHc;)n%_t zk$70RTrDJz_KOPNqNTnlDqe1Jiyx$kS00@?O@&xS&0~pTo|RPkd{M@+LA6~GUWY1n z^snS;{N&4g<{xFwAuvps9zrN&cHM$zV|d}=IP`8@wOJm!W9JAQjOZ=|L~rApU_d2X zwwl54gf{^&+dvfZ11Zy*PM}o9S(UPBF6$%pJ}ME;Binuk4Wi#kThNGT59fsx`40`V zbEFT75S;hBp~^oeI@$-B+}ms2^s4Zw(e5QbV<^{r8)kN<>QEQre9|yq|1O|=gLFy- zNNz-mzbb4 z`=MvKQTxYQ%#nr9w!6+=gTYWQl~lsCwfyfpK0tYul!5rn9<36`p@}lsez{rxN*J4= zQZloSM?4`x)F$0^GC24l6nj7JWQrhKnIysKPAhesCR1Ucdv^Mn;3uk}vHfHyOWa`w z?>ua&Mxe`WG!sx=!8a@K8}o7k!H+JLk?E@Ey>xJ3FFY#jen#Y8>j|6ZsYj1a1}{>L z$1iSVcIIq4bJ@4GLp}vzI$vSh5cIROa_Lk z8)nPL((Bz%VV26N)v0%3uB2E9!iAqdZ=~1St1+W{DyzgT;AQ+m3*z48)!MYkj7MBS zKxPh?<&$d}hV%1K!YLCcOxSQJtv_^wvPV&jL;kts*|sz`CHGz8JCPbe6Tff+Mql^` z7l%O4K->*4Qa*bbO0D(=N`S%Qb&vfi3iml;Sxy(KraN*|nz|MSQ=?6SR|`AurlTZr zg(7>2Z>G(V)hbtH~ce?}z z%b!PqT>&t~UQH<^{aIhk90%X_U+@R=hpOlfEfy`6&ZvC~#_?iy6UJF6CKy#m(q7_e z9NA{(IYgT@Ea^4F)07sIf*(ax;w4yNGiW>)Bh^i6Qx~pm`_Rm_(flnxRJn;kDPet_ zl0vFOzPZ^UIMJ>TM4xq9eF82rv=qe*9hz2Ql4s}OcVHU8`q8a3qo+jm5$j`U@H+ki zO95Oc6*tQN@`oMj!okZ=aDL zFM3r7^JYJ0^By?R9xjwj#)QXN5#iTdV+y&F@(wePG(Hwf%{<2lJH(Dd%dCbDF(&Nr z&6zb+7)LngTT@`EPdo8P_4*RMZvt51ZWVo+Fk%8hWBMoQ__}LR>>YQJZmli2Zk1I- z6?T=fw`_b1`_JCkecRyJDdUalUt-GL&6sD0U9{ralbNOS!?2i_43NfF9c%(T-DQ>$ z2eR!km?XGe&ie#9AbZl*j3V8ncSs)6_utw&CBbwsrjmjo+S|nor{Tv`*f&2n1JK8K zwol)?6uU5_L+qs>%`E48CO6?~OCa?D3CrOtVw@le=SwJjl`Ideb) z$tteR+3NcJYh)6-SW+Qzp;n$k^l@fNjN8Q|=A<)S!d-b!yJ zY?xxhH)oUJF~O!u^YtJ;F0}iMoo-bKc9}fboOJtCm21P<{h++AB^NWDr^SmWxr<=@ zP|+5kQy$+rT>?uY^a6VdflGTG1Tj^O~DKPC9?TXmra zK8-|v9(+3a1W+ZzTML8u!X{=QAB8h3H|*2fD$Nz(DF5RQ%j)K=OTNL=+Pe0SUPUjL zLLBox?}5^dl~QM@=7J7RQM9`-D0VqU2qeRiFkfhC3aX!KCu{=9g*dduDIEiIBfak8 zw4$GSu609u$3D~QBYYO|$>f~D^lua1%whX_fM^t3taKN~b2XyYsYeq3@3R`-mnP;`A7P?JMAAO3VyTwT|k7S?4oJM!a zATcPDwaMA#+KxoJ>!;05_H29mX!Ei4u)?zySDi|!+bGdw?+8!*rR|z24}?rXQmn=* zvTlbT&`KV7VRw-R!CdP^j7PhjOIFX^Z;s7e9&wPTzex|;5;Oth=kf~Gu1yuC*3dzp zj7yS@*cqzq8&L{5_&BBc&`Kb=fcs=NbUq|C$?}@Q2D7~;__@YU(lQ0QUl>STKfPOW zXz7%?qR}XHkw;4MV_QazZAqEO;tMjX1@>c!xbJ<;dSm7#5=6O!#+~W;Ljg|9&^2(m zom<5DI>EB3q#A8oXf69{A~^I@|9zT{EOzs#g92!9dnp%xGiVDm)J>&QJ8~)9GvT~G zSa>qmwy{6;`Et&~1&o>;+m?Bn&&=>H8pQ#%cy_v6y!@nx)l-shcx~_}>fPY9j)H4R zjlf&o!`6nw)WF5&~vVKa>mD|;n>mNrNlG$=U z`i`Vd*PME=vuB(TrCIl66l4Ct%`2pZ zLQ^1U-s0v~!F~G`*dYL#Z5*D2$D&Kvq#31XAgOv}Gf}CXNLq@`z z6_iDAIIrozqVc|R z5CzA6oj_BkQx_}L=L_w}yL+KH!R1K+YVzMzru%0X9BP*^wI)s^5^3xOwce|9yBM^Wak7Wwp!mgRtY8l}umg6za|Cm<}3dn0MI9JJJZ>CH3?qkv7)&z1{Q zK2569--huNYsrzW8>RR?*jcRhs+rdi-mb}QUhKs_pF3JkH-RhXgt<`??YTD2?p%Q0 z)WL-Dlx;k#R|ym?FYV^7wWJXDwLc+5wkkymBID*EWdP3s$s`CvpI;&WA#{q{EQbes z*9$AKJ$g1!unI=d8!g}Qh5OymK<1rG{P69*^OrztfnhV$RN~G6FKIFw*Vazl?c1uV zE1>*2l$69J2_Jj7f^Wsr5@$$xJ<*BVQ>|Q+fB1p1BW0}JFZOXh54iP1MoRdUyyM8Y z48wSGr~8U`*V?}nYF&*vvftfA&Q@O!5h90=rkHYOZNBhQcXciVqRI}UKzth?kJJ>R zJ=o#y9MPjWTF_D`L)&RC2E9~>TH5+eim?|UQVu~Asg9j2W;k}$N9_9uJABYj31&0*!)}hn3W1U$%7u&#N!ED4GSTxY-Tvy_niym9CSu(g%KSME7*EA{3K7>={I|dgKeKr1Eyq) znl+4PJGnT$p-s%!EOB)A9oz%YZ4q%gnov;&hSfG~RC3NYIl{jg`DCe+a?h60>$8 zr#BZUu@n&7phOy$^GHjHy~$*g^U~XY&jVpzn3ay@*_K5&Sa)sKeHK9)qZxmwwGI#jcb`3RpLb99Fp3*5QJREGru@k1) zhLQ-F8LK-=0^XR6JXQ{%D(!*R^&7A1LXPa#&^|%i*;BPAws%DtXm`W>qTx|9iL%2y zDauowHKyAn7QWYVG*0r>c3WE-;H)PfZuiZg{1tEF@;MVi7 z+39Tov0bEZ&|8PL8|p|bBtypHCgU0D1!EhYvWYeXvQp>vN}olyCoNyqk89~ku_6T& z5O`JD47Ek|@OszXhGzX_eJIi}mKEJFc$d=nJiaG@g*(75l9G81G5^5Pye>J*tVlcJ zq-LC<=JGoJbXuLr<>dPczNRw!SnAeE{&ToUv0sp~tQ|a|FmZJ;EM#{}9lM&1hS1E^ zUn#+~@~n{LR%!m|A)0Bfd{SF?XGFyfdzTn#O~o@b4(FM=1H23rht_s5>T~I23(SuF zz(sOW{sUc@tY`s1=o_x4BR+khg1xw<%iHdP zqR4Uw-hmZ!94bqgg7dl5<)jG`9WX7u&XNyRmKCw#~+N8ry7a zn~gE6S z8ljcRtZf)|x14FCdB+}NXJ_mNvcB}1{Jfy6PvM>(YgV^W&eJ^v#o(cu|He*Q0+06Z zOJ}i)k~@U&u^aWp+R~V$WLkm7Tgj7sIM^t|FtoB+20d5%`-rrn0dV*dsL)S@EAxiv ze*!&V_2UAJER}``h^)fSoBS!VNV8rr33b>0>89exR0(Lsd`}ceVa^gOf4RL(ZvhuT zK+Aox9)`esD<9tF-Q((XM{%f4goq;03g}0_`d4IiO1JEY%*1aMgaUq#Hd~5Q=l&(K zYYc?>>5qEwisPibmQBaIlhPRBLpT(`S}8DWuyaG<1|{f}6q&1;;NVr9!s=^Ug=;qu zl!gMH7v}7Z=ZJ^Cp-_edp}hXs{SAgo&N)Ji^30Rze8nzT{#ugSN{eDUZvI`Na2(6A zhmZRS5Z#APK7zHI)~$}$tAmG|n;Pw93zjd~-Iq6w;;?e%wHbVw$N-hHxAol9u6_GQ z$$`Ga5tAoiX&ayx3e}_P99g`kIbsUxE}i`6bJCEg77l=@>2LI7bqd%PhRn#^I#ej6 z@wj5>JjDC!busRrcN(%r~iz z8S5q>a&Cc#lLw~|Gzdwg)zS0N!F8UAD$!AbD&)SV=eK9}UPqKo43CHWewfw@l|zKy z{qChxqjwvcZ8#I^^ipxC%EJ>9i$P4fkNMoK?PXSBt~$sDelyr^!=1`Gw56{P7pVir z;|bPxYA^ivx)>;TT@VYhJ@XG?3fnp#%cL_9(=sszjno1IYH#73!1EPV6bVjN-+}jL zuLAHysFj#``{i~9>RVc?c+Q1;zu=8yREK!k13ZpfWsgoU&T4(}oNGqhJ3Y#PI%{ge zYeCYJD2$gcKpU2kSg4c;n$@<6_<=4qA-g~3#fRb~!jyTNgW3f#+mlK9}zJjgH4>bu5iWWN-KH7)b zJw7JSK!JYW{(C-N(uZ$ni~a^PZR$m{^1YBn7>2Hj@~N55!Z6oA*2oF!itAiG}`Mo5|6CLCW;wa!9IY3dEZ!W-Upx@ z@sBub_)dXH_#K0^eT-r2G*)4FQPN_*n<{Dzsp@L}Re@cELdH zeu}?rqHhX>83nrHISrMeQDlw~%bBFquLqSu+J#CJ!6H%57!Ni-`FdZ1nQ3@tU3Z2s z*F=$rC~MJIZqG@^8yD>dM@&zjGW6Tw#;%gErJE%94CWlRX)&s~VeiY$^Xu2m;00TOHKy~)kgne3+z7c* zY7Bpe0WD{CirR-)sC&zbO>|Y6gQ^bOG45jMpXFkioI%UD!e+V@mMG;87kTlx=H`e_}_o3hW9Uit;pM0N!5_2FjPph_}Y$w*6&TUR` zI7}@{8+;vQ)zD$1o>1RSd8H3{ZvwT=a#j!bw%{^kU5Z7FoimhZ!J0S&<`W|?tW$||WYZJ_~ zxJD^^;CsHR8AcQI;59T9*?-R-UXISVNI|ctKoD+(xs%s+yvn3nn0Hg^rP*Qv;`h4c zn;RMbqLeL+w~kA3P0(A4%bhqzR^P4;m@JR9B|}MPU(lN<^H6IH|0le}hVa`4o}xh` z3H$n4F>Axq3keGof`I+V`sVbn4SSC(YcR$bMdj}N~^FJ$22IcHk zF|hqOx^EJ$Ks9N|^_EBt4vm4E=m0*?mH}U?U)pdkU2|PRdqT;?iC2Hsu}Qah zBx4ip`{j$?Wog|D3&6;%-fP<)(Y1>b`bmj0w&oC6g%dZugpaH^DrvF++W-7m*>A^? z8D18a8r4*VtBVl{N!<2`;c$(sorw*&ib1)Ter)3%h9+5lH;c7tIpO}*Vt$%|{!afl z6j$I8e)aKn`D@lbta8lOkHX2z5PCLLhLtly16R^qv$sltKsvt4F_!E#NW|@)fjyw@USd>d`oxpj_l^iil2cl zM5~`plsNv$Cj)oq+rdC;cEy@%^Qw&61z8|_I#vvmwlr{--_(H#VaRA2e6LLEzhK2j z+RCUdA!A8AiX%k(cXyWz79tZ&@sH>Td*q@yt0N$#r{+LLPL+d=VRL5_*7peu+pP8K zqr!|T`wIt0X1K`UgJ=o4G!K>OC(sw)2qVxW3VMI`H4?J7k4+H>B#sz^F87gpa}(yB z`!ER|tld&)xd|_lug}{_X=E)g|IQ8Lu<{adJ-w0p(D#Y1>%WFjeAljbH*z*&XCdbuEy2)w-;(k71 z$hvC3*BW`hb=IPyJ3u9b`H!iZJ%W{!cMvsWGLG3<*B9o7lxWTMq~wbh)Y4^UC;Abu zLWeRMB$F+-z5anD5YT{%bx%CN!R23mw|Nho$9xuCjW@Tem9`sq%N)A=@ZW`5_D&OH zB~1HV4wp^9ZjXjF>9utR`#;MCc4e7$;z)lnWO#&px8q7i!3ex#{p#|^`5Zj`^%XW( z^^T=!g_<%ypu3h|A*qg|V#SEntOG%_%h;(u1Zu{Ng=Q33#kwal`-xT zwYh@Hbw5lDjui`W8{1LjFQ1c0%yk*5XMy8bD+c8bRulbE`u3bcZZVh_Ry-UuOrDBs zx;XW432`sRcU^cH)T5d76_{!`JdINj_m`0jPC5=!k<5#Zvw;ZlBXfU}Yt!b97xJ8O zj`CN@TxXz3xET7h^N|aWVMIY=FH%L{dVC3h)jZ5w)5^s6%W}YQ7bSc)$Kye0uW|)P zn+AifqVxJ$Bj<{mIsjqy^tEQ6V$CI9=a31^y3&PLsY2;TlLI&VODqq~h#OE$*(%lW zr>s$?y}vbiCUF2GHPhmlcIi`~r_~B$7d8s7My==Z!>b^j^ayEv3vonHaZsrDBN#?% zw6*z^fV9I?iw`#U4dNYOS$rZ_U(F$1QCu(mgW;H6FH<#FhYOt@n+Cr+8kp@^)3^4p zDEGl@0{x+E%WEBep^ZesY+){5T0y9FCRB%Ua^%V-yNI7@pGjJ5jCE6{1J5~8CL?Ne zARS|D6WRx7Dh=BI8=dBkONIf~jM zEVB@5PJ1N}VG%5{kb&NsS~xx^orahl!l4g1xXOQe4~ua-T;JcDC%>**S7g&m`swxa zXsY9}E4{-wU@H8y;Y;=*s6Q`Kp6dawhF7zl7bHVR(Hr?ICv+6Rq9|?|qz~f-+#52B zwdz3^xxo&5q%Lnp8c76p51`S8v3ESw>oKuLF=Gi0UQ4tU*h^4MFPSO0bH zWB?7UD$;E!GBm+MR5?}eMx#ZVwDtFS${}R*P55+%T_^2|oYayIh^f9u*8%x)6X_Z(X-%&GB&Ml_N{= zOIMf8)`RP+EjiD=o0OvDC zM}K7-C0J;YQYr@SMgKRQxeRT;n$zKwsh2vjt$ghft}D^wM4Z@Cq1n%0x+O4iR;3be zrI(tsu=YZBWz((Gv*(+#BW2!Chu-oO;uz|x`|+6}HB^o_iREs$d-v-Eh}i_RQyZge z$B&8Mp2rp!Fn|_CeF;_B!KY=$5ZS&P?saXrK6J&|y~~d@4Ejlx@Ydb@_y-*L+s6Dm z*fSu;3f3Ndjb1FllQ`qnl1fr!HmGxkz(&gk%8jk%UaY{^TsDGC@{n)SUJc4UNBllo z437ixfxet)J05a4VGK2DJ_B6?!6)?E@bvOC9UwtdAjt?W_~)WPfDapj1)Fw*H1B}j zEeIz)8*CrHV{({L5=zY^dKUBQGXWPsORYpKLvJIN}MN zBY_38cRk>4@H@X?;%G1^Xc#qlA5a=D&Rn0u8;)(u8~T?CdcaN;KtkRt*3dyJpcum# zFr=XFl#sIU(p4z}yh-{7+qka1@FwRjHA*JfN#a%Yb?&VCW|c>}Fj>R-J?S_=;2o98 zXG09mdx#t~D7#S6M>pKm*->_zuGu?4L5LJ)0HFuchmbIB{M3})Q}S%z8QaxrcUt%& zjmORLMrsNJ6gE+IY&6oSm&HbdA!NSAIO{eH$Bj`}YVlDc;}f&G5D+ZQcFDTLQACH% z>deH9XoXm;5OZ2|c4cW!+cX4G!~bj+#jbDyVG8o|EO6Jedp-9dv&?R1*Kj!DYa+;3 zUM$1F#fYKvgXJw++DopWb9GQfz0Omg_PqA!@5SYIB6H2{H1L{S+uFbEJ*B1SjhWE6 zFckzRnC*8`$VBe(1|(9CJnX8t*~TRX2g|5!^i*LPgsUR*tOzT2Y<`P((tBAP>j9mB z?=h$(zp~~4GXslp&>X+&LZKn1 zQWmnNRLumG70yZK-bqWzr24^4|4Nyx3Oo06E8wD52pnf<|KaHTbJx^7(h&wq`hg1! z;&{q$N9Nv1E&GgKovjVZPTk@)SReyfYAH%XAA^?nTN7zjYMWa(-G|I1$bep@%0;FA>QB zrw7@)$~43PY=RpbPNVm$AbKFY1A%|bY9K_&7j(Y>Kt`?y2Djq9;uWU0ZmMt=2TPS| z>`U)FiQ}B|{oZAwX|ZihNowd6w__{BrAyeLB=!086reBk2-DZpzt7|7VX$| z46S1VLG~Yj)o$WrO<_fdf{0(rqaB|uT#tExN0LKD+t+%k<5y{WVglZMAy<3&J}VYj zuG0e8w)$bf)4%>3Q+toYptB*u{&@tN6dXZc5>N3t)p{+NRslG*VwRs1+4Sv#e9xjz zxWkE!cPH^oz63PVUuq^(G1}lOGof;cYB^E~uoR&@QC}2QU1i@vr^g%)dC86T$&q`I zR>PTc%rfwhd2LHt4wUz;&Kg+fX&hyU%p{qLcob+dY|qPI27=(Mnnz2H3WF9bMPL}k z2BuW&rtr5u>o*(LU5KEHLRGwfDIuaa|>1~&!zdSzcwUh8Z#}|4;zB8F#EHe+? z@?Xg&*Qj?AF`Uq6J=Ur(s89R(|CsxdE7K;g9WJ7aY~fk1pi^8(qZ%PDy^E7p`ch%9 z5_l5Lfvbl*?vD>1QDtZ5E<7U{xx-S5YWxA!<;G~-iFfjBdamt!M5 zxpZU;@c>0(7#Jk>ypfTt5i*aoIFE1rZJJc%wBhJ6jvg%i1bHVs2p4i2G{6w0yun*cipuZ7i%XnqI+50%7COQxLeV8B7eOiZl$cv`355J(vG-M{bG#MTW5dN1u9VFx!*nuc-sK)0@ERVeH1jC7T+SU7Wu^EPiZy|%ZPzrQ>p@o(= z*zFV_Tx})Om9t*0A&AzDEJ|79+DQ{ihS0yD<|(y?mHSq=xcpo-#|$qj;{-{im<}1< zDIewf%=Qj*SNAmPGNhMVofZ25com@ekp3}QAg{#Y*iuRB96ObwHgIbT0R&J#h_X6- z(6O`Pam(n%xkh;B3#M$3BK!bi)9fG(&z=kcl;MK?IgqWuJ8=WtKbx8+eyM~v89Uur z#+{_1llC5fc*hy6W8A+Z8TTbPD0 z9LhlhjZG-dhl zCwtt^qU8$Dbcc&#Wrf_^?}~W>6@x7bwuI@iE~^tf2IRZeX=gQ;QE3FGJdW0CHe)}` zx%@|l&knhS`9~RV*%g}Aki}#4?7T|7YsE6LKBRpNSrE*~^_J zZj*ZLu;*KpI-S5H4#voX_>8sRNGU@t*J?b)XqoCgppx4zrGre}NSxz@cN+Un&~lHy zEi%EEaYrA?)HGyg6NKz6HrE(e(#n9<4fvz_WYagG6xB%f2saH{jd{R z`|62D6OMtAY_yn2arcfl5uZse2Ek`uVLcBFm5a+xX>K&vv`<82-zTUL7~6{;Qkqm2(v7J=8? zCz=kth*3)vZ=Dd|1!%mUq}`cdF6hj%d7|3I9cc@78te%^Dz_ETuPjBjEK*SS^~B-L$xE+j4*iO4oaI@S^3-97^?Cic zw9)r9D{hylM(&K*DS!XRZ0MuQz3mlNG~H)ljp#^E<|Y3|hKWVfLfKy*gmf|~Vn_+!fhySWvT?oKB$UQHDxPO(-q(!&(;%qqZ4Cg|yS%Ll2l*2l zGMbRpL!*aMoz4@JaLPKM)TBrxm0h*^NI3Y3BJY~75xqR@N}G{ljsJ5vG~mek1S~J- z^}20YAO!e`J0Ybt^`FT-B!U1tWZOYtNypG9P5Tp-kJG8sxbU)itk<#d31pJt zKVdbl*ODPzx6~)VqWXAvAJrH;6!9e{DE&JNNIUM;9DWs9dXM^9n9DjycEa3%7^JZl z_z8gy+$C2MhQ(sGGeJo{;DIN&K^&wrTc#gV_*Ld}G7$1F>ncWr7@#HPI+GJiG72i# zd+@(7i6rld_u+)A>mToF#ZK~qiC9qoYy*wt02ET!j~<7AZ2{-)&1Y`LZVBS*l7`s> zBocXsrS+@cpzK@h&>54oYi*F027|x+p{MEU|7ny(8J7dZx1L;fb8XVJvho5b3TccKsnyBD}kEWUiqK{~bU5YQlWFS!9 z?1t__JCSG>9&jxG{<2f#ZafVR^{deDNGPFgVh1%r_Ubf~hx=7qVgbt+Hx{RQ9swaME7M>Jdz za`Y(zjU%|)aXtSsBOw%bwfIZylNdc4i0A{cXrRp+B=LM2RNSLXxXvVg`;J`UOX&7J=i+Z#$e+&gD$`M>8nBwLae`lDchSl6HFqDUEl1KNN^a7n$w ztSiKTyj5te=EjQWTJFh)mgu8|<@5*KKWV}DkssZya-U~Pkn~*pj4ZKFujm*1qM=o% zTrjpjf)o}>!t#tl5>pEH!rHZb#AcV|uVo11oo7PFGb}E5nQo@xVXVt}VKmiIbyTD* zmrfj=1Tlih1}U!zDFa>}ysBx+1OlLg?WHr(!0No7ite@W4!>8XdSnX=wo3JhY;XzX zJ51VWzPlyelV^BJG$Q=#!MXByrBP;YBK0sdPDGi+1wg^aMkIja`*rA}F|CLmOrBC1 zriOTD{@+Ey*XNnhjW|%^gtUV8X%opgbF(7dXiYbsozCu*pf;>SoGcNEFR*$435N8K zoB*xQVv#Wf_+!}@T-65c4k>n%c=1C`Petea`#^G0O-rib_m>(#_hgl?aP4H7LMjVoqbD2lMS8ED|<1WRD5DjO6cpF$tjxC zx7>n8<1;0~`Ci<*;n0Mlee+%5FJerTsX2>s5WWoEFy^1FT=uKnK&L-1q@Z{l?XFM&f-pGZN&Xq_F zT>}ki8&Y=M*OyWs5grc|lqDr84UIM?Hx7f0JFhv<5^9zB`V7GlC2Et;O044%aQ$Ab z2Qp%q7=LB%Dr}m#ehOUJf={J(c&VYU>Y9LL~*g}^!xlg3@CjsEkllg6vqr(r17F0}4DJs2Ne=#OsX#*&rgCvOwpT|n=x%;U9L4SY^C z7zf0g&?$BipYcc$D<a90Wd}>ym1lNz+iZ;c?lHIK2J zb*7pqEKi$+Jc%&}oaN~qxI~3hf*GHd$}#Ocgt&H=2*D?lCZ^w^(`foO@Bj#dpB0Sr5k&rj;=Oh z;q1}W2*)ACh`&z@qQa*@{w$y26%}bz2zs=plvq` zRj&D}2SBD#L%c_Pp8k;$6d6b5!=KB!nQ=Kl%WX?PV@jdG@>s6{{g8a@@oo{fEL`%j z=5CsR>(DK@bC?@9;#cNO_mLs7R1kH5mgZnEe@=@00vO}P@(-qKDSYd>C<*S1YQ zQN#BXe=H=?+kSMn{<`^83Eee;YLWQItPA7Dh=&5K8#t~Zi@1G#PNmYhhkDV9`)PQ# zIJ=~4RrUJk|qBld3V%-6HJ?bA5 zfCSX3-9}6Vmd;C>p0vDC75`IGtW?{?^`!Iiiqy>J*bg~!AAyI`68l!IPWvqdT!5)L z;R$;(g^VNUZ~+ zV)>#F!&r63l_qoX0LTZYHDXDMDr2@iNA2^nQ)|@{r2w)3+05YCxCXcB=Aiq5Y{JJ% zF4Aay66pSgl*iwL{S0ZVK5TvPAoj#F=~cHdZkDcWW`(RC9C)BV!FE@y7-|r^WwFc& z^8uqEL$Rf%%3#Zb+OBUh)(=i(-^a)1c;g@^T`kbHuGIn(+>SZFqL<;uDBJ|CO!_Mm zKmLG1ZSrvELtO&_`yB0BGGAC9kD^ohHgqK$oCLwYou$Z%zePgN@Z!r@39*;^yGrM2 zS#VLXojgJ(ZHj=^+EqE%$52QTwBG*NpFaLE57S6QeedHPWMrMuR^|D!Nv;!%J7JoW z;9nCtgUgmelhj!$CPBRW?7%aDDZrAMRYGScYrFvDNz+5$rqtt}%cSorv`Ucf!Rq^% zQ)sgD>en1}q^%BGRy4c}X%|kb+)^l>vHdYoV1C-a9?=HP-8VI?$*OVN%&?Y&MceR7 z(fY~RXH@f$Ns#GZZ5QRv-Z*W8oM8hUueih+icl4&e6`&dh@>%G3gJcH6$eJe_z%I^ znQE0IU8Q!bSP$7lI`?#jrVURgEv2y7SAdaC&;|>Pp{w!QBOrkpq8|I>FzT9oq+t9U zItlrd_qN?YkMP;m6Yg#C6eEys&Q8jQ&7`?XzChnqWY#F#S-0Uh=HTDJ*_=7g=dHkXeDoOPrgb`YUgXs3zjUcOJx+|8Vsbr1*;v;y++NU{mp)acpTKFhpsbSFMbK>ma>EP7pzdmTVju>hS1a2EZKdeasXjT3529mGD2) z|Ij-=U75}SMdUoxN{d&4n%?#CM&VxD?jG579RMrTdGiQvfS7-ko!Negf*!Z8V#$-g z$7iiZpN6`3kaTdTG$v*6=-kkSn;MR%^fG8MYolIrFQ8S?ND$r6aTa*9SVmGt8~r07 zj~_;G(OplzQIX^-mJV^;5J4CN;LCobV?}_7%S5*E5Z`~v2o3?`KNl321AgQ%|J^VP zPnJ=d8ps++c^TW;>)^{XMd~PF;LV;U=CUKi`1LujhZDN`hGO@-zy3{I<|Ze}f>~p0 zAM2kT|^;U$_X~l)z-s|E(RT8lp%G%WWHyBmL@?>-E4WPb`l<){X{TMCX?~>A9Ag?LOLV*(U6!xK z@zR+xvn#5-Y@Ts8Hs(KboW%$~V;4;2t??F8QSrMXWw%ns6u3rg$<8{cFLT>&(E4Ye zS?DJxzDuxEOs@0z>%fAjX%hDVE*fS4&b}$%CL`GqVa1dY+$MxGF#PgQG+bTjjtw_f z_ee?_;%Sn<|5yg%#tld9pp-Jqa$-R=LGbAnKojr==toYwSP~=v(HWhk4}mu6(9xE3 z1)kF#Hw@t3TuF{aevwdPI8>gec}WeCIt zy|uvY%>f8avnfZmccgadiMO^t#Y7(MVUyc0RbgvW`1B|&X>Sv|{PvU6ZIVvmDZ#x8 z)>|JNvND|Ww@%1mHXR>2BVqEF1g)=rqCtyj_56Wf!c{5EOR1E+#~orJe{RWYh9-6| zPN63Bb_91Xamoxi@SD-`F(}BlSXOvKb&>Ny?}8>r2Rz?TC40ya2715^5XCsK`@{9# zX3zVOuH5&=jYhOlS1N<)UO_TzCx2|Cq52q*h!9Ib;^rt`l&YQbvPvHTT3<7UN;SmG z>dC~2Uvf~4zBJy!uqY%O-sI{4X$!K!Y`XhW87RvZf4k&3e|bqbz6eRW4|SA97g`Qh z5NoQ26QLU8N-j$6SY!uIHUyvLXha2q-NzVthn3)-k^>Cm>eJmh;q?p-P4mX9RN&(x z`glhL9#cW4RT2~sDIh{*MblAt=!-I#%ub5f$kpBWDULa~uD;2s`M&DrP!4`uGo-@$ zoJ7z2Rhrb*jX#UlFEYbahXA3V7D%J*nzNy?3K*Ki`uC`u{g%_F>6j1I(5Ez^sitr3 zZsI+|_X`uKzF+;-V;zCe1T)B3l+ZHWi_DD|8d#4oG!q|V;}L;- z<}sisj2swCAnEHF&fyPmrW?-O|6x~TVEA7qipg#te_s7ETYWGcx9x#b{=gIe>3-iZb zEolj1aZV-P$n@J041lW-_UP*GGuid8U7@=n*jwgCW@US0;I zWN;;P+Hn7R!0`MFoex%uDlLF3xGDBOZ6Kx*^hYIq&5MccF^E>c59*$r6abWJ34tg2 z96lLEs4vC!j^z8v(;x8P&^0Z>KHt9^Ka)6nKtSJWSW)7;`(z z!_F0ht#$Z&R8R6+!3$cU=U_nqxH<#8x_`^xY=g#Tfax0QT;V!TyBa zYVH5lf!6sXC4J{nO}_CR{u3+~b(aUGCwGJT1x8{6IJtm%5H)m91AwEW_ho)tCr$I? z`NBf_B@gca{!6XcIe~rxP2pxC{xUz%a8$nG&-Wt0`>7o8g@wZQQ9iSI!0RV}V|Tw( zd`0hkq5O(=0jL)L66~~+Q~DCa0aR}6p3#(i_#*6Tq3r$pj2PbygYUEWz#kfa580vJ z*j>4@vNfOj0bB>H%KPnWU8wt6UiZc5LS*>GcIX;rYyAlQh)y5o+tU)9>|J#1CX^GC ziu>lH-ZA=p@|`sMhJSNm&h+hTZCw8W?jddc#BS)tJ;sk;-;FuAgMSmU&gg5em(1@o+`FZ1w@#h!As;jj0Fb+%G-B$%C^jfiwGvi< z)Hlf^*`$8iySgluU9j`fL0;+t*=E=s-;tG)pD1&- z&xV~Wbq;&u63PzXC#5H1!DJNd@s+>xPqXscU)n1I0=^`v*`-=j9NF_otAbE-I2?j+a9wa4wjP~z<@LkLYS72!r#NJwc-$mP5WKn)Bm2=J6m)zCN^ZYe z{=DQ`0<9z-mv4^cJQR|~x7~@z<>;S`^RC#|w!dGzu9e~l9D3x({O8WR_JG0Lqbz49GGX-a8)9y7xX2<$OlLX$JVOJjnax2xNQ}NE&T&0;}vb^ zlcp>E4bJO~Sh=UW zY-NA4J_!yv6bOP!15Ou=aW)PIcVB)y{TY>Ae>s#{XcwiX^l@zHG;1`qHah*#PS=ge z%<-&A2(yka*@|MIDVsTrGJz!hqbEuuA3`-(K}^?x+eGCrUJPxTWlWHYu)Pd|v;Sw6 zu0yO3#-U|EYC?l250I_XkE08%wxsvY25}BI#d^l!=q_^?%N61%4ohQO^Nz$^pux7B z)%yy1^NnzdM}*I;xyb?YLdJaW>r|?O&)*L(866%hl(pvW^zW7|wdPTW0vK0tRPHnF zS{2iZ3otraDaJKcb#0dS3}2|(SJ)1l5IU*2E%%dA72SEPq&Sv;cBx4f!TdY|M0nE{ zi%J7z@L7HVayr5t+2a>uOH?waW-|k!F*XV@pr#cP(x-DLMB&LBd`1{?!&)i+pBi0o z!Zf*OHJ>TV*XNckH#&FlS$c~y^A>R;BEHHXk12hHi(Z_vaML0zg`6XpV6Utt?JpN% z&fo>7;3ddON^3k?Xr>$_tIxr#s&;HxX<=ZoTX|M5yDV+Kv1Ox9{gbT{@OYhGM6A|c3s(|a;9bz-41e-@3F zr~bifVL@GsU2*0^gc3nVGrgP*fBOmR(&eTaj-H{`!P}hD^Za`!imvAz{|>qXMQj*P>09 zj$C8zCMslqYA;V(0hy*@&sxJX8Gs5@2S}?rx;J zDUIdrVa#Z3R}?d)F@<>ggG~VE<4D812f0f&PrJkq!A}N?oS0Ed&^9qkI@>OB@e~bo zmqu!8ONx4^_wK`9l-hn<8?(fBzWl=bA6^(h)7B)J1)pn+XQl0H5z?M@CJ|l4DT$AZj@_CczbERqo!&dFBTpvB4scmw; z;1|TLYHgouh7HpDv_8=5E?EMt5)WlE-lX6b%$~3!hEdpqm9q+h2%HeSR4-0ppm8YYtazWO{4BVSVWzT`430H# zveH-GdP5j35xd=d*bP{nhU&6zvU#kEAyh#tgz&hFY8ivDoU+bEh|%R z!BCE}+MIp-*pSs?g4bS9y)iN@G%*E``#0s&> zZfDNrQewpwT>}5;ObkQsmxmH?v7c3HPI&jeZsF0D5UP!DsM~E>nD#lYa>SUWhjmqQ zM;e^1kfJ^&ef?S&zB8#QQCzbNXf+Dmsxm_Ue`|Gar^plq&>4>N#8eCi&gJ9>l?Byk zM)oJo$eIrQZ7jDvx<4ye;wijYk@m z%1bjWkM$$tE-N&oN1q^H*pCqEM-szD?n40ku5!f{X4~aEQO6|Tvk=s--V;wNU6z`S zBd!!PrYxB03WlM9lDgC(AA=GVFy*{|E5>WR;3kkGolhtQpR>bd*FBzt-%hlX8t9JI-VFTYd+HWHIN~+tG{-}Q4vK-gQm9RpgSuQ_VVL(rP>My>Y_aVgkQon$J`M^qFmWY==E zDE-#%E$A@nV_|T)2WR7AaW=e}359V`-6(&;NgB{U5V`A4gnXt$9eEwpyXXn;xFLYt z4T(g|Ug0(bs%07oCOA?0TCl{f%Z1DI((E18KEHZ(3Go~i0|_mlmW3nTjqIZUy02vh zqIf`Cin`Lma;wH*m)DQmRCjlh>Ir%kBXvCIc!f1rcR8Z zvvA$3lz{}~d;3hJVuPD;?5d`Wo)hqA4Zwn@abr#zK1mlimWA5>nyy1q*21pjan{di z0@|G{UR^v`w^@US8}JJ;w+tw799wJ`@OFF7%QCw799o36G3A*7>W!V)sT;;0eUe<@ zfx*YcxFIvw!uz!CivIh~h$4@t=VF7kxid+c?t6e^W$QM+{FzSj`E2F|*NPltAXp9)J6% zkU7oFj|^cwSl2yA6-4|vwLiE>?b}fz_AEQ!o!3C)`OXIp`;--T-W=u1TI=2_RF1%t zmgUuzsmVKmOH@ebR{Em2+XIh6LaQx(NY!fXCFs@fQW&0+Xqwv+TMUM9FwpixkqZ)1 zP%&LZfyCzE&Faejy>c^ z^|OCurUFYt_We92EVND?9XvXS^d@pBgE`+gVs|n&>uqTGtL?PQhqP1|{@gtK9`S}X zT&!88ZvRVTAiLy=ATqHpMxBpvY@pAOqmHvG*Lo3~HnSRm@0@PmPOx`4ONImYnZLK( zc#+DoF46UPOVLw>#Y7BYTX3*MMbG)gOO)qz)6K?k$3MistDTtKa^uwV-$Bo&eYbX0 zs_@`?Yq^Ho@vdT1GRFZ@V4meGjHQtZryiX6@DMKd63dkqW-(DK)^0^@7H1)dd0FJa!8Q`(}XD_BeZDT3G)k#Ei=BObzLR;g866Vu5)V7S=B{X zbM5dIT2*1U6~qq8(x&(k)2A9ecUOn)|va;ksj2HjA|_1efQvZTebI<;h@GgIuDNd zK>mc%e4`XbK4-Z@G3}=C7ye3m^Pe~gV;%;m#8boPjH|0h3!)nFf7%OlPIhQeWfAzH zH(t~0B;H!`3NeQ>Qm+6cPVhSNPZhDp0XUfkV+i+u+Q>Or7Ly^kz}-CVg(L-`SLXYtiMFH4p1Pv#5e;}KyCmLER; zMSu3Km;{xW$D79%6ciZzd~q!?s@O~83>~ChKs%dQN(*Lr;H4&v@~| zpc|*i&fTvjH`G#Ix<|e+Bh>ES+Qh#Qw$XUhd+R4UGlyZxSUyVb6p%B{1#q30b##9# zmNTY~fBN0qw*k$?Tw3sVOj1KxtL$2fp7kLN9v=Z0m~LpMm>bF{So8SG5U04C)vz64 z(Etpr${q9;_s*7Tj9!S4=4I*x#bNiuce%t1KuafKl7lD~eh~_w`$xU{s8XTY7Vb3f zDGtx37D`{UMAoun&DJnN_B0Erl1dSj(XLOZ;E7SN(<~KEud|?2>Xi?ZYxy=EydM$? zyH&!DLl(Ero~DUri&A1lnY*L+nRzq2skxWT7k?1Z!YVD*XeL!secPmK%+Bmf1L*)%G)M5^ni( zh#d5DNJF+knMRrM`--psd+*}_Mj=DX12>8gf(icD%(YA791#AxZv77cH$ce0JXEL+ zxNTqjKzFCh+=EbOrA@i8#;%%<>G0^6U3gApk>n7AMqC1ztXWrFp1GuDqj`mN>xOk} z70Tn*=T8JC;lAwho$)j0!|lD%aax${{v|_P2I}e%wqxHcPHUN02H|i4T_^NzD25U^?z3zUCP6QB9cQWL7K)x?LAK zDJ?}}H9;dnCH$NrsL+Ttq|jIIfh|B|dT9Z+^lrJvRL$>O8Q$7MhZVGaP0e8Mm+;%I z3a90&G?7?n!yhFQzSqAQthTv#mL>2odB}*urW{L$OqCTw0EL>2A4TvakU5Z$N^iT| z^yzhbAD-)Q&+#X~nA0&vYp;Dei-+VKbFJ9p+g?Pr%+NXl z*G%0hp7?tng`FcS_}rGJQ%U^#j;?yA^wF#-d2`|+HP`QsNo^_q&`G(2`h3>LzMHaH z{;(}!fw4&w6nPP0sxo^b;p0BbyzUFjzn&e&g zxa}R=0qCy2=Z<=s3+WAUiy}+Gt2rjp=dpSkhSlV03$Nk6-4#}{VvJt?CU{)|Ge!9R zVorG5#t&tOr3Pm&ouqb9|ATUu9^pebjzKMA%YH*486zFe=3&$|gQgPX^SrZbZ>_d4 zLIqaUZT`N}LcQUl1bPS6yAY#8;a=S{4~%4n$#5(b_b-GAvUVFD4LeY*#&l`u)~F@EeAdhGJd$6cxiNq0 z7uG)C7u!7~n3o@J^&+d}>u{Ps_pcF~@zB16s){xt7813a8Q>5ARxzy>5YbF~%l$QW zZ_NrfS% z&W!bk3zsxH7+1qlPL!*B3Q3^T6z+K18EMHviQx=k&$fT2GXapz;IoDPb)fZdud$5j zN+hp5dY4@AVwVfc;Df4q_00Y7#?mvmL?cYY-lb`=3su$JQKBBq%c*ATVe5- zEUHhZ`QiiyI`8fsBN5hiL9%Q16$Ph=nU9e=h6@dI-b&Hudn%+zlZ|P0UD2~`Mp^;y zcaxS#`}DYW?_(gJhZy$knGEngOE}$EwV|VU*xdg-QJzZ=c4 z?n2!YPIAiGL_bNZ3sJgl2=Dc-F;?0fVHyu$6rdfvsC?Qpn$UosJSb4eJLM}Lr&>eC zwGJ8K_O3^qh-qaCXIa5}rSU~0@6pjN*(@jc(7wF&Xq;_AK$l-g`b(WAFo9``Jj2k^_tB803mjY|5~_S_+8GDl4GTiVrx!Z`+x7V-|A)7Q}W ze9Dlib!P9Lz2&7Oby{?2w=MxJLg=gdkB!KU2c`SuT%M@Vo{TCQYT#f}hHGZF1)VSI2Whr{;EJwKP=lE^bO-M4odP z2(J6#I!0oRCA@|@%@LQD;wpIv`b_B8Y!rJ?qyrEEUvzJ#M){=;uZzN-M&;sw2)Pbaj2 z-MI!PL$i;O!O%#zgy|XFX?MMs$DYjJKoL*w0n0yTWf-M_4}omE{CS^BmO0* zOfoyXie_so#@8%oLJM|g6D=8IZVW_pW5lk>NSYBBlL|Sh+M(2tc#{%7H~dyDZ%1g) zoO`D%X|!p_!hMIpyIt(7V)hdsG=x>AQGsYm9{CWtJN261EWQqnY4Tgip>qT{*$X=d zHE79?Qy*73g0Ezm7G@|QGl(?vE~RfN%z^P41)U6Pi*rkitpMS@k|5Lk#*Ag^TBbTd z$@!p_QGv@eluG_#JFNxmsSh|@R4j7QXohkY$lgpRFASm^%DLEyP7{F6Y!%7%pdTz* zTa!=I4eM^S1?sSJl4_&0?#ORr9jx_{|E9k*4kq|rWVKom4sE6Hlg1JT=sdzfPbAvq1_@TC3RLR93Cl0gP9$D-{YxnubI6Zpi+Uxs}-} z5xEzkULvGUyXm;7j&SQ(g^@;>=iZmNoD^NK?~`2A_V>KbrP1{Z#I5f7CdgFEo)cXQ7M?j~+AK@jaT; z?kL2{iY)V>GUs}}yF8#dOfvUj;g;fEX%{b;P=h zF|4V{3YC=MW951huJB{MEJC zY?DD|pmDhoYdU9xpGjrF%QdlW!htrSJXm;U@gjn^ih{Aw+-1Fj`MLDI)Kqt{gjF4IO{7fi#KdfRwmaq2TfeAf9uER8xcK3Vj7eo^#5jCTy0nB1CSWY&Ur>(ql z!g4GR*&&@NoHiJQmC^I`u_qg`Ci1nE32V(L=ZhLzs>kk@*4EyZ(}Q;W)5}4j+^NN) zAzfluBe51(QY^xb+wzMulGouqD)rCN3jz>G7Qi^w?@DXkzwnLhJbRZ&Xfo~PSb<(m z)K8(EG}z@B$!ZxdD{FmNj4O(eemY z%LpN#E=Ql|_zwv!+^HAeZJTSYg>#z=u3aqTvQ2M>9t+jgZ+QJ6$q1En!8=28~{0?S#(^f_S|jd!5rCq{?|Q(KiP zY0PtZBtDu>Nri^E9zwQ$U8v;fIX!&G|IzN{*HI@_bynaSL$i$ZB|m?it1pYCzk#2o z7RtA2AfWFej_H;5a?rM^xXG5b^PY zx$%!;m7nA&*p3`ymphMZw?q2FjUffSZjr_@_^sUz5|#9CLrDqRkGNMM7B3_Qie+)% zq3>?mF?JnNX)1s%GN^UY;IfcOzxT}VYX5!^mC}|#BR*9RB{{E?^j^f+N2l5Oe|&#Z zxtiO4{b)F^U00@`$THal9QA`z6)d}KJ;ELI(FcCZ8%K2WEWOa|H$!ZzMAtySU^25? z#sW|3yc<8yvKU=VX^?lu=D}GFORYfFbiiE!cMHL6n!LtF=m=@9ieKzomqm*{DyM7f zWd=+n(U-#A%wz5Iz`EZ|r*lRuDdU!Pe(~XpH{`(!?Ah@BEFk;c;!uV3wQe}}gCpi|KdB?oH^PP_qzi)ELukYB|gfm~{ zI8J8Q`QER0#?k?{=E1IQ*osB>Yw&ngrx=4<8j8j8+=EcN z=lHSy+qBxK8uC4kXq+UP6GNg+F{NGEVi+8!b{fO2Xb5>5V`XE-oupZl8L7VO>1Ic* z@1^q!CNAc?`5MjAK}j&lOW-$BM-Qy zjKprs5YM`CsbC+*N2{rP2pvBL{49RN5S~zFkOAF8O~#F)7vJajgSEaNcd#W@v*=TQu`_RMJL@2ste-R9 z@LhEfW*Xd(#~t!4sU0%7RV@K(8Qz@#qS#uG?;=LHNE{~`5Iv2}aZ`$=ohRN)h`=Oe zj{hprP$JM9dX0Gp>*OkJg$Kdf3Efn0t%WWnNwV8UmE(KE{c{@_I3Hy|p#5=nkTwP? zXxXG_saMIpZjQ?WhW26SJ>5A+!}Atvdgd zW)adzYE0`pov7&1KD{U8Bl@gbDpnIINI#kGvZY)s`$}1}Q`^vsw``9O{3?s(oV&1R zLEj0KW} zrA+%Z6o%e$no`aV9KRJ`VgYIvpK;OhR{ETTAMvfdJ5eRhmvdbDphZS*Lvu_pRP4jm8f05nuR5ceD>M&sIk`xvr zPDp9X?1y1?xzu_V5fstHb?9+Fq(3pb3WG~1Y*fAx4>&h0Q{7Mik2ILj3Fkk1{~B53 z-7n)sOw~T%$m86DlyFiMh5u@XeoJ{yi_Hw@l$dxfN*-_`dr}zKE~qADR)yG<^*h)Q zJC8?xs1L!n3@54O;-SMQ@vNpST7v0E`>vNZN<%tl_qpDZmYMsN)*hb*jdRU(wb{q& z?+Njl;2fRfhu|ev)tm=aXc^UQn(P;cA6KEdB=<+u#4s8U>I^#*HVVD*aTN0hDuSv4 z-x4@WA*ga?aD>PFoYg-Mm<3m$2S>5?#gn=qC-8YZv%_MlR=}lL&M`W!J&%ibD5rbn z*2fDAWRHl|W#>9FqSx%A;);UbKc$n~4OEX48*Z#D5v1g1CGFA0hQ>HTZX22y^t#g^ zijYRShT(W9U27l(Y%g=Mr7&0DTPf<@xb1wXageZEeHgh8gTS z{^|={eAK)(1G_qC;+N3@OPz?B7_{wknAY*s>u{A!Gg`0oXG*^I3Njb%kc5>!3(*fT z7UA?*vKxxpnRx?V~lt44CESDr`IYmYp{Q*n(H&xCJ(^vjFk zby3>N`yocat%mPOl9P@wP>$KbVpg=Q*)YAMf%#S$b6MYe*$QVSo~G~pVmFXfp&ylh z1H&NJ=fpT;kmAu}3s)x!jB2&EX_7DxO=0DAfYCbRaPKen64#=U)0h>N3?IyTWM#xD zN}mo^2kT?ZpkFhKXW>KcD~UJeE!*ywhSgRaeyXQtWNrk@;d#SR-5C#wN}S0%94sz`c4q zgQODGW`>8W*H4N(!2ydmad`LG9ittZqmJZ(=0gN4kNk7*rZH8K03FN4f8Vg$y`g275)mp@}VaKDLhVC)4V>}*(? zaqXjbH15ocK5zW(K$i%Kk+e+xM)%!9$bACLHr$ShvAQ2#AZ(ASOJy$mT_`fnf%@cJ zau0N6#c+qv#~*jti(>qlg6PhJWV4ADtb;RBFvB~UBpq0+SvQu`%477%%ZJ{diILT9(Ft=FyaYfX;tl8Rn+h?C1lVJq<_-I(6JlG& z^`ZUa#eJsUr%^BY3L(y2nuVn$cuHfQL4L;s;{)j8)13szYD6p^dPFUTE87FloT}bw z+2*RUBy`_xy}lgmU19W3uo&$s*TpMY)W2IekApJq<0iS}hns!#O9;{!$@g{LbRzOM zZT8Uj$Qm#rQ+fPtQF&)R9{oI=EDE30w5B7$wUhh=zvJr_A(qy)jSI{ehxO@h*&5t* zRm-Yd{dH>xhRiWKmy~8hd-(RuCmQD}ZimgOKbAmyCyGDZl~*wV^*90h0$hJ4R3yZztc`8a2P5z?CHxC>XhM1NUAY*=nWTdG$t1lZ3QR>wo=U1A;|(QyX> zV7>z>7b58$5OM730^gSsT?3`B<64MvrW|g~h9|Srif#p+R*6w?%!ec>vpE2_EG zM3zG5%*FildEu{MU$C-3ynsjH<1~MZN7V`OpOxMX`Xo~fBLOm6c?EU>qp>KjjH-zV zLr43}&b*YTNobb*HjkRq%#XOFBsO;6&_2QP-;d713WNprgl1RQ z96+8{f(t9l4Q5o(n|uj~z#tW{{cCR4!fjv!T)x`{JZRE_a;ex1Ot zvSE->FR=u~Tod}A_0(?*OGvuK$a!qrd7oFnx3>nU4qSJ4br3^E7tx-XfdbS!i1XcI zMSnvbQDB8GWZgS?B0tPs*OF~d?xt+mOZqYE1iQibuG&%ACi|vQ6o1lTVH69&v@w9! znb(p>W);xY&yEOZhpqU$5;ShkRv`MpfmKiPm5R9_fiZ=h?~?A|h>k^TMk0FJu<}zx zxszfB!i`sK;dEpddy_h9iv-SaDq!r;Gh#yi)2f}h?bJDB1l$iTcrv{hZtl^|yo7t0 z6geaw$h%^O#OTnvI?By8KRt@{5cTnjUf{cQ!&nefua%^tIpKxG50wT2F6s4}3xXDI zEQ5?`FwtBAS`AWZ7xpswhsUkKwt~6CX|U3ZsvEyeyI65$KrQXqWrX`kjn(jen-PWw zsQP`_eXSe`dFhz-M=|?nO4>DO*o94OPdule+!<#3w#40JFO3@vp5$|GP~VBt`!*j* zFKa0Aa-#8tI0~Z#Eh<9_ux^&XHZ~L6r$LnKpA|3-Qf{%c6-3^zRv4YZuc`(>c{DLD zXnRmgDnW=bYSYU=qLfpSGMA|qd-uW! z@wF3WwQj2;nuG7@W(QG#lhFmf3(mx;H?)ztOe>D?-l8RiqxOi)rCRp-xx3<~OAwUn zD(LVVCR=2bi=x~{+9f3Qw&FN2@|^r)YbmT2abEl23p;uAsA2SBZtD1JT28yFTWpF~ zmAGO_iUm?d7DvVI2&xKylMGs;4_=YOhhgAq99XLq4=1?9wIgCHZY~PxP95z#PwMCw zb){XN!%;_@f}W|nZWt|)jyCd5s$bmAxK!JBB@7SLz^HzRaiL-enn*m!%||6;4~-cn zJGDa*s*>1dL!zcA-Z(8KwT#!*qsZ9slx?wfnTsU7H4xmcUwlBCI#q>1Z(4UaNUdH{ z#@(<1)l)%m<^&sYvb&dCfBQ=N^>ggzvX4_2;~=@g#M}>b+Jv2w3NQ5p%0iBq2E=m= zYMGz0OMCs6U#1txQ3Tholwl>cS1t!9Y)=cVyRBUABJtrd>Ia)OLVnXU|rabO?=4zMP9KaF3D1R-l@nkt=*MtK%yl+gv`jA-N@tmc)4VEjr zLr~$mLy&}mQiI05!;Oesr*K%5lXchZ4?26l2prm{z|(#XrDKW>}tW zZ0Nl&p-AP?POHmC>Dr8kyFIp0_BCJCobaK4`$k9ZsjkTwN-qJsL}uHH_PqZGG$T6M z;t2Do$tL%g67~2AQpZOct!0VP9x)JXLD`DPg+$WTlZ*Y8l^|Bocazw0Lxk#dI`9cO zWy6UmJl1gJfU7o~eB#iU87a&RSf~IiV+?N6jm>81Q$HUgGHNvJd4e?e-N9EusY+3@qq#Tfr)tcYCpbw>$M7K-2b{)?c0@png){4iK|TAvJNG< zNoD5xvYG2XcU6&*AERKDAU8@aa6MLfvDv%I_)~VAY^3OAM4QKK#?B{URZ= zmLADJ@13)bZrilKXOwGzXIf^ia@`WcWB_X}CI7uav_aPnTI@Lqq6}Zu*pph&yE69Y zcXbvSkF#HKzl9JR9$UMLxFYY9?R7Ie3+t)Q>W<^C2P{W^h$= zE)bk(w$v5onCgtz$td4<#a?htYmj}8Pc@s`Wr=~m;u&BNiLDZ`3MrNljv+TI)bl?1 zQohrZISUFJ=80XdS%oM~qD5d4I>t+0&U5JE;0clC7g3dK-R+cl{e8-Mkx3&bi!W!8 z?L}*?m5!puoKYM}G|DUjQ?Yj@x+J9L)Dbu6x2gsF*YuVbQw=0lS|!V?j-gwb(w-j8 zqVLb{KW@rFb{E)9<%N+ownYf4rLo8HS6ehfXkeU})&U6;bJrsoVNLmp?h=myy%Lr$ ziRy#kT;`jq0;f?weuC+^TXwR7l%P{h(DZk`%sd;`lR3+g1bcF7+pY1Q4kfsV(O1`W<#&rX#XsR6i4RvtL%V(DxqH#lXv( zcp}KUe&IYn{u(CSnh&q*bhfnIW`$hb@LP45Qn_p)-KHAc#Ml0LtXwNhuf`K`UgmRAyevy+&_jO}A%*gc?y&RQO-;a+a^mWDD zdZ)VHso&5B*HJyJ1I2i%p)}p3kL>WLPV_!{Let^XUm`ncP#~Ih6*XnN%(|78dz)5l zlkc=%6zK0{e14>f(%`3imiit6ZL9?QjbIote|Ionp&Z$Y$hu1;wPa6}tVlYmi!D8D zAe1KsUsEQ!Ud3v&Y+#of$oT0W}u{} z>^DyHT}^Hdr8Gj1#jv64BNRn6irGt>Y&DYpgBmFwN6d3V&UC+YC*uLS8^^$ll2&oQ z{_81+s8&zSG1oap(C^wbBL7u*Bn$P|G_oqMW*J1mQenOk-Q4?j)9(a|%uDYO7|vQl z(Qg8whpr)51)ti`jFstVja*d1^ffPVtGVbjaW6NC>P3|9y+P!Lbd2NSoI$tGYr6hk z)J#oen;#fWm%ykI<$`H#g>l1+R1XJcxN8TCzdf6%q|@889tqZ-}$VCnDFopBX9|o z%xv6}pbbLBy?;ZrHFo?g(K+b)N=~K=FQgd2GWea+(#Mb`2Osw=Ro>aZy<2^&um`;c;G$T_;BV86p;y5>p11Z{aiYORm2e9$9Rc^ zQdfHn0?+yFf3nZ2s-27(Z11$tr}s@@6!{Wv4cEPangyRDN6-(BA|^t4YQ!P%nf}aG zgvF~qwYMJ&BQ;M9k)QbL1OpAT6^MDDBBW6)Gs{&`$Xg;ieOqk7^?6@?tJ@o&(1{E zk;aH23=Ppb+G3(W(-imnhFyyb=9Wmk9;K*g&X$-&!bE>#pqBR#IPJ0Kzz4@?R{n8r zt?)Vyo072ahYt}UN3HXT?)G@g{Aq(`P*euvxHL$fW$J9qs`~IDwI-906huZc#4FX< z0FkOBx|1}vT+if2L@;E(#V{mtyKjYfLd(%$E?qb*?Ix9Mt+`>DQKZ88oiyH21C$GBk!aUrTz|Mt9n z)Q=&eaba7D5&r?@qjy;oGe^qM_$uJux{wb|;UJhCKY(s3l?b}m~vH)EW&bWjVmm?~#a##{mKrm%P5Wy7T z$Vbg;ifTctUkl?@56blbPbnKl5~qf%`gP;SfyxSDTz0I{ezhR=^hWuV+)D zoE5*%Ekzz<`c_-UMoJJlr#r#;riBmSUOhQTw(K zGJPj;cygs6_Z46V(TGXx+0jFgw=(TLP@4&5jbs(QTvqt4AoRgEajYbOx~&L$Hsxcj!!gDR!Ez3|AuPB0mC;=1}k%OFk z`6HW60?gPJBi)X~Z}>M)E)89^o{RYL&Dh#IM-F)F9LforI84)#-aV}eHSH*B(gOkt zboB4xE?8ixhw>2&;hE_jrTwn9{7)uW+!r?#`!V|6S(&B}+60qOdrd>l8{|dfW&)`O(QEh>A4s(Z$6VQ-8KAY@1^>VWolMg6~pH zPMI_DJ^4w$B%Wb0U18>NOD+FE>8d3P^BRen&Ov?+uFnv#d-HszW$^$G1LeJJ`u9*G0=G4{rD*Q(iw! z@Z}$qfzady+6)Y}HOvW0o&j0vO4>_%( z;0~37#c)nNtTPw_H%x;Fr=ivR7WT7W`--?#y?K6t*)3NynCD7v>bz|L=mnd=)najf3E^(1M;fu8A zbJO&_OezWk4^=apgM&KM)V%iIw@KTMtZAGU?8OWg;pEu$`B^8Ic?5Ep0%sE=^3b85 zC?yAjv&op92SUP{qKrVB(>Mg;*|NcMQ^eGOgW_<;Bl%c2nJVd#`(U5L`eVtMOBJHiHjOH5RyXWy!X9xH$hRUEtpC%LHu#Zp>}O}j*-upL`P zp`Z!#$n-u5C+O_VgC7QvKB}N;Y^$tP-s#d}^~7?RL%1Sz{XEEILzzB*F|+(v>CyOc zW1hFkH~cS4ZbYM`{F#}?QjgVk*m~u3e(^yCyLkMHL%twZ30*m<+cMfDAe7zqP^SLe zL$;th+~T5rd*CxwTaqG!mWxT`TY_Geip`{91DW4-7;gVm!NE5W6h+PAY0(HaMBF`X zXOmU6LiH6$f!l|{@GMg6JNHHM{gt4{u1!!wt`t3An8!4rSB(ti9_w!%b)~Vi28)xz-NJgm+>U884 zty!|;pH|5x^*R#Ezow*#aRDn$MEK_HPl{+*#+}|62G85%-<&Es@mQ77iIJ;U7mFj2 z>zq>Yd;Ypmdv5$339Jz4)Dn4&wD%}Ei3_rQgnfO2Gv(jSXrt|ZK2tT_t`)epROY(M zXGM+>d7Ekm1BSPN69U5dg4(@@;eTt>umVn0*IwDV^NcXi}_t zgz*%Qg#h5eRMa==y-%*`mQgPRaJ7V^{CodT21YE6=MU5k1N+Qomqm6i`9M4&msSq{ zm~Ffk*bUgm(b^O}Z&AUb+ueY#|1OZXFV_`;t#`RFacMx`t0e%;J!{lSX(5AcMW_WiV2zRIs zYDOb*Mi5LCj;L~4TlZ!wJ(QIxCPJ*PB#(M`^(>sAY8e9MBafISYv`=NR#Ar;~vum5JL_pId|UL6{>w>5v{`bFCY2d}2Ft15UjQEm|j&aOIplvrP~ z5>Sp7q>(=2WZLqNtZ;|Bu^Zgg0j1Mb4{7ieT{}d+CPktwu3D#UUEoym@@yv^{WX-@ ztO3&ID0a#~hM~Yv8M>9a)<7n6G&-7<4l}S9xDO957tU}tdy$Wfi-dbN*hm|c;?31? z#0S1Nv@*@;ZCf@;kn%n@!nl{YR|#^0(Q(6{xG;ZWar9$vwJal zfuXbQoXJhTgKWmuo=H*INbhop{4%p6wd48j z4@@mD%fk7wb5dL@N#j+$bMINo&2_nII#oXwL3@W&t9T`RwU2B82`Sxt3H&$CFJ#jH z(jGuUi^#fk;@&G+u9eG$u}(p$?q^4(d}T9A=a+NA7HgDa`*^fI`}9;iPn$E5wV6e2 z;qU+0%U;jOtJ}2B67|(5+ENU%>=(Ys3RSFQ!YFmyI0ogM zqqfYL;r_z0EkH18aUo>4#P!}9Al3Mpv~HkCZ|ZHuf%hWa+pFOwck@<^q|g!+K&}Xk zw^nFt^}Mu454rTH-P;)mXx;3oKc((5vaqIUn6Ni|l_ffoFRI?bc>ETP09B#488Y*w5T??V4t2L}0Lyx%WRA#6(v#VWr z9(9ZvISdREHAZqSKRcx471lyEg2<3{$QY~`6kt}zdX1_y=EU8XM`Nf33RRzpMtUJ= z?Ej8ZcGw7pjDHk<3ZR#uifC3S=)C=4N=~spv75(%O_zue81Yv*IGS7$VbxkeUzu+m}^2C$v-I-_#NnP;xfZML8j#cWezgn`Ykx3 z-g$a-)E?y;L80UdPecp#pm9d;Y4Umm*h!yuq1$IpGuw9rx0M$x2D%-C1$%t}4{Jt^ zN&0_ez)E}|bFc1LO}`uH;U~(}+zo4nT1-G9;R7_{tlzEk0MGGLc7hOhJcLSbwc`84 zJ_LEcI@<{C)PH3g>+Fg-G&KB0^=El^5GO8QZHYZT3SKE04(&5^gwdE;%4*ME&UnE> z#I>X7pTA@9OsAgNlfS4nXR-&l%VjYY6JiPHPZTg~Y)@0w_1Q-p$d0X}OIZ}H2x{|+ z&)kP1Er5jcIS$W0_okwJC09ql#yNUEc(=5@b>)^?c*^TaQu+$@tRk6{VJ_M4 z;@(Z@ufqDTISK?nd*A(mhKG9PM9z(P(|!cOS0RWzdJm+3z2;^l;S04q`}=})H0nBs zeW25reTN?hEBK(@U2GPp-W2GNJ-2E7m0M3wu+_!D;j}9blH3%gFoc6a#UUX&66ZxH zUzs_7U+ZI{P>6vPqHSpsPzuf*o9=H>IOZ?|0%l@dHf!P7c=ZHma(39nJTn1~r+tkf zspQ&&hl(!Kn3||Gw{83r^Y{$gD0EaRvJN`^Py2Y?-Q({k1sW|ht&}+U^+WcOjW{z4 zNVxZlVthaJXn%0k+qr3=nb}efYquWN;77!UmrhBq62=E)OHoT92G_);<7O!>7N=Lh zZ6t{q+jGBm%*elJFpbo*d0HBB;#n-U!1ip~jT^P-EE+~aK!B@)zEsrt0iuNBnblIo z3GldL zY2w4zt{OF%Rkw+enar?n*m~8H*2nbz?SHqZ} zeb_WOQzlq@gkISMA)A9dFz2cEh0)q@ zY*ieQT%3(fN`+@UzRtDFzlF3SIrfvWx2Wd0td_c`p`aR{x&SI2YI>ZfB0HR0SLm2P zuaGmh33(hbFnsq=04AixNT8*(JKxKo8T@#`Z!~hnFZP z9sAESmH4z`kYW%u;CrH8lIZDB{N3>ld)u=;=o%cln>8W!b=&ecQyW(~8CbjAU%J5N z<}vh7_7fRV$0E2Xl19o6bz3j=F6g~DGA^~r;8rP)-5%|kMenyXtPT3P$!Q_m6h3t` z_nz;hQS=A2ILyXjh78s}(QcSHpn>vK@q>(JvhE=4Zr}%@tzEwN<`vOiU%mN-@6S${ zX~Gi)3miI!16uZPl2wVAy4CX1I}&AxGEDKWYc;30`hiwq=|-jj6H?n{m8J4kHO$3s z>m@@W*nD@r0tExe3ZbJkEh|aao(n>7TR+L%l#c0Jegz>W&xqPu`ho=fev|Tev4f>a z0uqYxT%p+&+yZ(Kt*p>wKfNp#25h%ma|NNp7`RO7&&TYP{Mi5#qm{AB9!I@+v@uB^ znb8~`H9r@cF-Pwa>dPYq8Jr+v%-=wY5J;lDqUb<^?MvX=7Q@?Nb7$gqD~`EGJV&u^ zJnDhZ$e5D%@3r4a3WSyLQCQIXSPyooEL2pg_tN5 zkPyIADdntI0kORllA5#DRl(Q^}_$WG;O4I&Fp*otB%V~O42?ogkQybmO&9CbuGi8v-4qP+M_%(ahNR++Ur_CV!_R z#w4obV9$K@mLzlZejDLT=>DPFvC>vRKSa>jc-=Z5sfeT!w?pQ@Si1TJy{{S6U(ii6 zf3Qn;6k zP%5wK-v(}mNXHD25xQ7qtZD>1ea5k4I~J`r9XSL)bJO_G^-q|gp*40F@#@@TbTzn6 zYl%vCTytkiFjv#BQ+D!MN4xp!-}BZO*P$770G&b!o?==Ol0%j4j)f&A6$HyIAtGL^ z7#?W!M+wK>#T)ZBh@NP4omn%^KEwAu z<3I>qnc#`@TlfeK@9_WVb+l7~frcMr1kax|4~IW8+3zoI5wA8(XpmACXcc*#FF3^_WhHq$5~(Y!#z>WQZ1dQ~ zsXvQ6CZixKMJezW61Fw!fxr`h{E0R}ZA2gOzBK}nD2NCTe9Pov)P5#3VM%YI;eayu zH$e%w+nAAuVGXju%^!TfKQt?#2B+_uY;n+ocTPzpG{^IQA&zzK@}n#tjsQ5T0D;Xg zCNJ#-CR1ryL>m`$tdSS*{Fe1@2ttKTN*2HFi4!fK$^FYczgMZ?f)FA`RVCXXlUC-s zS}i#IIHAi`U+i^bn7{f?d8d0^M}iiYE6LHOI}-D9;{N?CdkYk8F@Cq>RaL@sl~R2l z9jW-PL$XanrFfVDu51W?)uSnIQO|%`07soxlU(2SEjvE2u&XG;+?#hG(>~a3S11$N zm5}u*0v&>`%dWs6ESApvtyp6d8v+*c7l~}Y&rPey`UUIVk`d}V{zSpx(8AceE7QOd zBL=+ot6v^7^alP}#pV0@L7CDyq@b(uRZ_#MiijkW=6{Xm;p|zqLf^jH0PL^Mtki~1 zM|x(j|8!+sj4Iw*j;lhBq1s`FO$e=`z=<(Bir@0p?+v{JlN#rf>;NaRR@cz7>)*

VKtN<2}2n!*F#+4PP8e`&3??*=@ub`Oh5 z405K7y^J@P<6HxLC&I|lmReElZo)c1Iyc+q!%eqe{=#aur_fjI> zV$fNGP{JvmgJ%I^>DFk?VErR7doOJ!An`3`=dgI=yjMm=z@*9&G*Jqb&)q= zkYo0TITDeaOXq^TRY^4T{^8*|Xt6ge;pO_OSZ%!@J>c-3>gLZ39Iv21oVbBN!LL1# z7kQ~4=8JoTC?5JFts@sxN)|^%Fsz9hsQJyW0EZ~OorE-U|NDM2YB`afdt11TIn{OCgUg&e_`AC(#2hgC#WcA(LH z1J}Vi=}cS6d?hz#oqMrB*CiT<8Q(BS3s_88FgZ}6om9%-CJrUN-J5=$Z;^WxTAx%o90PrRfw6cln%Wc4k3I}gu@7_6$p-k>yOGA&jCv5fb z1HCXkz-j_njhPH2&XZLf3uL3G{qbJ*g}gYUOynDpg;nrgr^Fw#Q+F@f1CMwuTR0>4 zm$RPVJD>Ue@t3|m*xVBKw6w|bObK^LQ7@9G;m+EDMU_ z6~PgD)%(5t%?0C9*t{j<)Ep$ymhagli-9+BWYrg|8Yr239F-~FhbZ?t)XcYGEfFVm z<6C2fTBlVnl;VXge-`h$b|zvcrAy7-YnDO3|`6k zhOjEO6Bj-ZgNP9Jf{(KDZv=O7XqIPwf)gvQGON5`*S}^ z6M6;HS+rqhC)bMSHLg!|zK0{)*Hqw13TX_ODW*Eh0M?sQpIjGLs&7H&q zo9XKZjeL^}+u@}X`!^d5oXYXiqM@ImcR!3k{vI`PK-W0B>iSW7Sd1D_0GwWx@p4di+9;;twkZe(#_$F#MBHWW@tIkH)a9)R4{fD&ri*{e+41XE6 z;tr3A*#V3rDc@PXJ=-7K1ToS1>@13&%D_op86LOwkhN*U!XP%Rd3XBd){&p4Zj`|T z$Muumy1~LSUm_?2P^XhRU`$L%KW! zY8#FzxiA)41t?wnQIl@TRwV|~g_uEjqb8rk=Ub zf|oH4AQyKC<3;V+WNhzTv9QRJ$MhOACaD+~*e%$IqNWf>k8ER)*}t>}D}>xb@7dFm zQ3q|2^9CBVADFj`Ls1Br7WJgRh^|T5WZCsER`gk>5HugyHT)^|?X&hOzOFUyHxzSz z4b1YMrM%Yx(Ozd_et>vzbGk^Whh#;i&2fvZXM#!ze$O%GIkFhJm$MV@vGS}ed+vb< z5tcPDaf1Fc$%Dmnu|NEU<(MojGQP1~fh^kD#)=OMk}AZ~J}w-+jyMEOKa4*ABu0ah zPgW0H2r-le$zNx@l7&?iPnJ#@IXTQ1;sS(Ga0w$7+6)_+fLfl&*^AU+o&-2DNDhxi z!qYOsGjD~Edlqwz6;9j8h9C8qCgfiA*|uz}C4j{!Nk=P@J$u}jvxpq{$V4!eRQ$k< zPs9z;ZN4Y20Xm9ctVNhM|52*CBr}>4nm8MLIEX;)%d6QXCI^gB4gZwg9gOkKXF7T+ zv(QVTWEh}PpfXRvH+yQl99MP)KAcUW>qlheh)A8eQUYRz_Y^W48{^el&*uwG_f~0pYA5`A^0B;acQ84jT-I z)o?c^D4~(0mDxSpzdZ?*D@hvE&+Nae)?$zSe>e$3jeY%fd9u&VH3 zE1KW+M0?Z1petjzs=b=i31mIE$YUh4u5bf>wO_sO%09GIm-fo=FcZb2xeoL5UoF z;gFQ_O5q=$=(V#Spkc5k47=roE3j4*?x?G`cnQCJvXX44`y=O;z_=N{DBJMlk=I%b z>HQ9?diSq}v`<6f)lP%@|_Z=<4IU8c_I#aYiCFoYhGZfK>Xn#icHxCcP1*uoNx`*eEd&3@k#F6XfJx%?N9Y+WK%fH<$>EN8@`Gsx0C<;41Z6$6GQlM$um49Nb% zTEn=UrAc?thC{XmA{S6xqa4V`y$h;L2*E(F)i>yT9>RbM&)$!wi(nNVxhb!k6Visg z((D1*0EJW@Z+I!04jX;(dEiMp-}9+7U1L1m*+9P-a06iDXo|)hO2%l2A*n3z{=7Zt z0Jlk@$0^LxBzicGl~oL8+y=Vse3~TvNJj*8kv1mn>T~F-kT>d%!$`#06MceZoj^_K zTUlInR-{ARWaZdcPW99Mw>rcXy}S@1nY&4CP?69!CVPsx&Dphau$`w!ZA(ibIrzm2 zU%-^f?AvwjCs+_ZDJD<-Fs*pR*>z#=Xb6bxtCm10I-klj;$2g8OdKuH?C?qD1kOD1 zm9I{F*psofm{k${RKr$M3}@Z%?-u@nxBBfS6-)G4tTfAk#d@@d{5$t^+h6D2E;Kp} zgU3TB>@}Ur(8@Aa{jyx$W?c8(t=GL?4a2S}>oAD;4@;T-kG1A|613gl+|p@EZ=W}q zUB@z2ofC(HSOjJ-c*jwa>nkVOkn4JR_RQgYyA^Vh1xpyer`0tSHx`Nb`Uce0fH`h6 z7j?a01lWy4odr97h{rGJ@@r%Zf*xlZosk$O8sF4@Dha0yjl(!nKf68QQmB8+9Ob#P zR}1MZXugHF&0t(mIhz>d1AHBkE+C?|(71~qFjrqcLT7}d_ioD1AQ(*|FDelDx9F(a z(I{AM*U+n;*sa-V2KPE30%|vI)k3-S4oX6dYSS`4GN`0>>zIBeLRY_ykW?|SDh?jEOv_~GCTxwk`n^yI?BgBtDYXYm3KpaWU7X6=tD;3k|EOQ2%}S)4Po3{b0~laJ>AZj$fehSLD6D(O9OFEiZ3) zt;51txtp?1Y-*EUTYw7e<$b}%K?IxLV#<%4fqqBJA`Br*MdvnJoY0@0z|2ecM#*iP zJH-5_XYg!uAah_&Sp8;5ZW4ALi6(ehbF^B$i%_O`k#oIkcpX#=|9o`hcv^& z^#6cn2%#CoEv%eO90(c2tqhz@L`{rrjZL8W`Jo-1983(Xq21P-)GSm`G_jovtBf8RJKb7P3Be}Uqmy?_;}iVlJK z7qsRAN07uU*hFvHz{Q+)La$TvP4{qdC69kJyY75lzsR)~m^k>vMaYC0koMh*0mlLm zNsblqBwWsK_x=4WlF#cu_<>^5gAek$AsKAUGJFN zDgbCHIB?(z9}{5&X?csf#?NuB9MQbhHBqH?!TVtpA%)CA`pc~;gbQJn zKS;U5P|4|6!Q+A$ zOPOCHDAggSLu2`MT!@TGNa^+b&R*U7VP5{w>dX@F=3bn~=kY=5px=C^L5G1CU6gLw zz^tm(=1!`lc@qHY(ql*?NpQ)am!GswJ?_i5ofF)z9MyVVB@(5 zLrJOFGmkerZdmzHYOHBWHa?Y?RT3=Nsb7fk`GKXfrpmlL4%~&-P6B(Ea1gJ=w)PAJ z2~|qcg6ZJfd=`jE3ou;g;E_0x1th`&CBxBQ^_i1*$ikB4cEn(A?_ogViW=Y}AQpoQ zBM`>nvw+gy?V6&3l!OxCMS#F~CLlIR`&IEsQ9VHkoF|Wms!2gsl+l?t+{CCX?FIlJ ze}I>7UfPO72yqi5kb2(5XCfkKX8%HoC*f0A9G*A-r_`qv$n~#1-vm5dnPRg$zAm7h z1^uY$!>a>0FK<5CoFZz^YV-y>q%?x%OC1o%g}j6XuPl`qZO)ET!?|fINed%Qf$E;! zF`JHFDn*(eOT8U8ufHPZ_8ladwQMC0+u*5H#xkOUhAsy1cP~_%!M;mf)*tpy+0qOu ziR9Y@^+js~m1IIs85A)4C$@3hnd&b6b590~#|dQ{T{Tvm?f0h=i26U2gFkN0>fI+p zuFW&coYRy%9)F~J8m(7Ee}yQt3!QOThp54ItZ6(^9P6&c9l5Wb<(jrsNRmZfo)>hp zXO=goeS5e$t)*8!6;F3NwZ6VN}JRfHW?&3$1YesvmiZvxku^?9)@e_|}qdrYcj zy42Kem7kWgxFY9?UKr!RouHJ8d2My@!+Yz$6fFLNG@op;ti0lf#(<$ErMrrqEHjBt*V9oc_RleDAM3NVd*FMF~_4?sEmno)yz86#@)*nST(9>8OeQ=NM z_+2!Ly`93jQx5qxwa$LG7*MX1hh;FQYURhaHJiMbjqs#iIkk44WErmnDvwR+<-6jO zT-wrUUa%4}4D>k^>@jh(1pLupTxp+@eJvkYj4$V+rGIwVZtXofN`%+V z+l}^_+7}q>S*d zmg0W9srdzDn(`90s#R2C*K^Cz`3*hSi00z*ro4h}>lKH%)654cmaJ7g>((8Xu{Q5m z^-dqJ*bdr|G@l`S)Y_-pTUq!R19S@?*kYgX!a7|C=$uV~YRg^p$8t~_LZ9{7oBz2N1z;Q!yL1o@nt0b>ngN#X|ARDV4r6$U3FkypNy2x{ldtj zvs0*t4iE^9@R+PHDYz?^d5-eisY=SXhO2%>_X(QmAr)FKoD)|zqua!#$C6#z2d4~%p5)nB_Ol)2Y~P=aw7Sa5d*{=N6EHei6Eof3x?Qa8R$rqe9K$cM zl#ScAk~iiLT)%X`5$12Jp8i`E#`a&UFgEu8)Ls98e_Slg|B3$3DvXtph4p{q{{C;O zFjZGx)t{f4h}iK`(O~%ck*<|iH7d_ot9tqu9h+*$8qXU0F^TcYIHVSP6I8e~MeKLh zC@W!!yiy5qq6=6097VqF90A|h56+j(6`yYXnv)A}vLUNWROdQT+u1NkMKnbz3A*0? zd1R3%mLfw)Hv&|pIrK+3;@_hHDM_sG!<3{&ia<>2Y)Gu%rMc*S!pdI&o;Xo3kWL|H zK!>@5H1IGXXo3L``fV{#3L%=2Jc&Q0Ab11)Dk7DNm@Oq-)Ie&3h#KJ##$kxNaPJ0?oPJ>-mqKQ} zebckc@Qlq(F12-ussSh)VDS>7_xc)-SC@EEXaEph8cz{`lL{aT@A5Kj=a~@n;6)Ws zd1=7Y0Dn0O+rrwo(&SplBuezx44@#$C{etdX5u?Q4n_j<9Sk12KRW351nd_@kSsMX zOeicYoa7ZqsX|7)rXDV;F=BQTS^^*vtrr0Fj1dCq*vcRV{e!zcfp8!*mVjvNdJBdD zGY3Y6;S2*>lkE^IOTIy1Z9tRL5^MM16$EcnnSC(FJt~8vqrnjU`b1O^`vv)HNzg># z@&S4m+^;}&^Edz@L4hzC)8*RF4bktC$(`A!u-mas$UUULaXSq(}xY+>aIpx1B10omP(z-OtBqBVXVlaO_{wb>*ApC2TWQk&;@4($}L?%$^9|jtD1Oxj?s^5VUkPwgr zn4sy{pJ*Tztv^d|0V)6}KYsw``9bLK6g@L|NbyX8C?G9Zo~XO#D^rA{!khKB$iZ;tOQooX<+}`1_!8^&5La>Efr@W=6ed9 zWqE`Em z2_9m>gN1=1PbrY#2upt=#wh_p+8Dlkad3`1D`%0PFG0S#K^OIS$uf@+S+^Rmw!0J< zuZzZBr;4PLe5jS70=_>EKSh3rJ%1jQ8?*75SalS|uoP6(0<{BFmH>-OaY#L)?^c@< z7nPw9cP*2h@BlSHt!>gHYqZJ#ezsLEHE}>j z8o0bkzWLng^vUv#jB^gLlbk?k+|t9)h-y2Fg*%p61NdCL0pX2Nn-fN*8%{%tV?-e~ zYdE@}3OdgPYsnS#D=IvvFG@Gz!0@s;k8FPk=&}`UG!JsdP+e)dIpQXelxBn5I}&A9 z@z*C>PWraDn;~cyn})JZ5}4$jJ`TCNstH34_S%(vss|h6l^cfK7|Kgm0lSKaa z<%{Ve%g-bEu-o|@O=+e}oP`ny+x{7UWc@UExQEjg{y?w8{m|!ZsiFt7JhB=nv%Zq# zU-_-!#+cUNGlln3T{3f0ZIr}n>ufiYZhrfCpRK1>Fo&0Qz7lk+K1|-aDw~oqrRxco zeBkk=FR_A0Ph0WPt*`vBYV57nGWQVt^|{aVv9EsCR(^kJO7H+5Wx9QOo%VuJeI?42 z0Mfo#lZD2(h)vbyKhVrCQPGwwy6fCUhBjV(!fjANW}Hw}G$s-gD-C)#|gWDVU|QsCBKh z>ZZNjk8c2xhswoG3F1n(iOOJb@7dD5xNfWg>{0n?#6$I%gshQ6h@)y0o-WqR4S1fz zdG_MN5^2G9Ce!3d)%`Y{hGCARs7>U673js>vWDGscrL%qRXQV_B}S_tbzT1XkMmvN zyG~&$dV2B;&)A1p{p>T|^)^Y}(fD)Owvi%LROKPoF2r~xaB>qgUJJVK`)I$z_il9Q1Xk;)28cX3@F{5!?`UIrr!7MFNkZopqOw=$`iOB-F74EWe+i(v)`6^=%Z4dp_giheQ9r|g7D+u=JK?5e!YnM z@GB17$qs1>>AggTDEgq2XGjV(W(<^#?Y;S)$764A_S9eb{k2zKYsZLR*1uW^k6_4< zUhMeAOx2&mYM0j@QU|ui!NDl*8An1iuL2Qj$#Fg7Q!O42HdJ^f^qj3FSGa~=ZYF3v z=(-)2pOU+U^)hK-shV3UgEIIvk=J?m5@Kf*k1-}t485w3Y^-|SVm#)`Qx=)** zVs_?nhN|dt=*|U;X&rc82fD5q1xqjtPXkYRZykT3YwrI1x_=Vp)(r2ck^6C9&Bz%# zFLG~wd6JowAuhP?y{MH_Q7h33_V*t2+SwYCC$3S-bRnCRF0H2u^1g+Wo$aPy^MZPz zPBTZx>COz;?yd+^=ebLAKO|-_I;CDYU8rSXo=-koK<-jf(_w@T_JwL!Q|q~$0y6FyHccg$amXhLJZxBbg@||Z%at0 zB)3V;@D!i)!~3-xmpNE0)!Yak9$?)zyUvqHX;ieo-=~Fw65x4^-4e0Z)tVOYqGLbg z*>d*pjEmeh{UPPuW%oE3o|XfPYo#4aAJJ#gk?SP{Y<7xE@Oq^Zj5>yL2tE8+=-Z3{%P# zNQ4O1$Qf#)eed0k(%evRHu|1KM)ls9`2HLLXS6z|LYH)a+_ZRGa4>|B-ph5NyukO6 zu_ZUNX~Z`fPsCsp+Z0+3(8*`tc&=_-b=Oh!J4hfORS^NrRe05T4YPmFiZ&x zVqA&u=H(AX9W;E5N%1^Cy^K+#KBCh-1^!6XG^Cu?dgLzip|z-y3D{6hJRlJae|8Z)jXXIhr9CK|cksh}x;69B$yy9_vsArI-P3BDSoj*`im&RW%>{h(MR^C| zr=gp{u%{0HK6m#~$E1Ipi9VFeBFAJJUK+0T_k4dWw7fl*8cWn4jT3>{2O(;A>zBbw zHk%2FDZGOZu+$rKZP%DLCYe}<1y^rOH38OI#O^ZX{SOha`G@lBjaTKIImi1NB^LLo z@j?di2W_YkQh*jxZ^s3Fs(|L zqMTH&nU@~6{D{*^&jK%sA{&)rp`i4E`;W+$2G{4T7pw^vF{?+IxBs%bTW?pBl^UgL zZO+zEXe-?0dP$l~i$7tXjGkY&r_l9Kstp?2q+Wqq!BVIqdKdcpmO%L@k;c%hS=!J`(Q^`RW!}Y&D zz%#^koQF8#Vqkh2n@aUi8}mxBqgve4#)8AG&iCCx;lQ~mCEu*o8x{TTs37AY`6g41 z_l}oxRj}+n^wUjt<8D35546%(ZxVl_>@o(?6~^L;)Jg1;Vuu^*x#(~)yg0^GzQ*?R z9tYNsLxyMxmJUh|EttUzM;~~~Z{W-gm#;=63z7s0C z$m};=!_|7i?D&#!{MFfp?4+ephcAE(3>H!I zGmQxaAV~O$7+67qh=!3#!f*;h1|j*-|NaGp0cqwg4n>GPYcfTzqex^kY6=C#5d=B_ zS@^4PH7|e?79PHDh)o6-ZYX(htr>N9a6c$NzFz^e7LndAzf}Q%5=|P#I4g$K8zQXX zg)ivdi*gU9hyQ$y-#h1>m|78vA53|PM30!eN}zGc%iD#k4g#KaW`RI>v(!5`F`)eU z9E1q$du!SINcy?YFPh7_HME_0eRg6gUl8#EFee1=532nq%dIv-8Sl!Qi&7(q44C2o zwh(bKG!)C?V8M~2}I7YfT9OTpu&I+Vg$a<(lYHqrskp`_#Z=uAwX~r z;6fq^?~_6}O=F2GNRlGCd6odjn;{MdAmIvYK!!pN zYPz)|f=;c65ezP+fj|Ov*ukQ_*uCxde=i)vIB6%XY61kvQ_$Y7!t3Jf3?B7c-#j*O zQIYzEbvV_LfO-ESgwMeYnSqluuw!sL1_pM8mgKSI@eR&`k*x|*wnFAaqpuXOWLEi= z1#{)5{ROF$yF#kQ!$ud+MNb%fWcnRIfjg96>RWT802(c7_#K)EL86oW3fsB|T#f;3 z3m3}H_$@&sQz&5FhX@}9XhahvmV;E?bW`InWRbNZhx|7nlndNzbQi2ZI~f>;KqWpJ z;RlEb7O7+Mb@u8ao?8acPxe8Lp%{szS9JsJ<@GvD%MiZpzl*@O52G5UKZeC0$&R_ zhR^`D*EPsQVg7ZO>0OPC2X!pc8)quT#LUEz{ooc(5iKu~-kS3JZb@b=5Wr1d-`m;JYU+Xr>H#_yyr{d#L_Ma3v0I85ve zq>TKkJbyM*@j87nhvVu*%{gQlAI(TJEWtR){W$Iog_x&lI;@j~h=a=qZ7Fn1O(ne) z>goq)sl?Ms%eF#c)5GU-5q1h>`ayy@et+(omUUcLN+7PYiY7;{pNGWr$NFQY>J54C zO3qZyJ)XV{oW6( z!qnT6l=}&SK+`5o-&#sZt(1I@C%}rc=j<2sX7>fTYVqr5*X0mXBi9+<#Sz2f0xZ8d zVWxe(2&nUm&so;34smhMK0wW6nHFz{3Z%>6_uA#>q8idKb9;Yp*lS2nmxg`%REZD_ zaC1J}*M~WO&3#)E6 zz4+s#FyCof9#%Va$XuXD)}fe@U+ptopl#`W(lCGK_WxFF8d}f&TXBD9Av@obqD8_gD+kGWCgP+ z|10q9WEufvjejM7U|eh*uPXUC=WO*>A}Yi>Vh~}O{pn1#xkkvs={13eSxhf7-%9P# zWm@{ew_}?|u?|-UcU>#jT*=dnIPhySG*eAynWlhqXt6V0MSP1c{yBeE;3Dq%`S`3v zHsM)L-Y4lU09>ZrYW(jc{m69Sib2DP>{XE~LWS+wtu=<_N)Xpg$93Uq*0&dc1YU0(M%0`2{WP;07S9*p#oqH)qt#Q$O6tfHC z#Zxla@|da`IhooA#l0au0>!vwQ?^I4%+`l=W;IyB{b{oUlxpQ6z5nA6UU2Gj{?vgeMuIB{ZJWsp!-eZA_@o9~R>mYP>p`4g6~k z=AOfGD`+9}{;63H9nk*{Vb2y8j7wF=Ul0MMz0o0;ZCdq;P&ZVW=W=Rk?$ujfl+P=aGA+V`GrAC`*hvm5UqW*sVZczp;0T zAZL)r|7~K6f3TF z-np*7<(cHn-|2}W@pb)oSJ3!An?_8I*(5ZD^87png2o08lN?<}UdBdmkUi#O zJ1~RF$};_Rr5i0`1Oi1}W6X1Nk}YOcZL}k+v&HgC{ugP-8jxD*-v@d4mo|`*1-hAG zBk`iq;0XsQW>+rDt7T{YtJ^Y)+^dtjPGqc#ibek_Y2c%aZTj|P0b2wzXKBSrEM?kL zRcrUt?zjY5qsmAF=unsl>aAAo(a8DVUpBr*Q+Qq^=)ZW zUVYq=-sP;hFuO^dJHyy9_mehheN)#yPe<*QWfrVai}8=3yU_B7;o4ZKL>r^%b4!dU z>J)a#rX-t8_KK#<51$dprd{{8(1d-1_{D(MGPoQ8=Q58y>@wPfgF{TbjHv|YHwy)St zscqk|lrNiDsjSQAb{$)zplP8bZn{qI9Fw2Tj!^zm3wqtoPZKm}vzgBa4u`x5v_>q4kYP{>%Xt`?pHsH)V^=Pc+6M2wDRF8(2GUmZ|RVU-}e3 z@JGtqJY(m0rxS2|%@rdW93|e;d?1Cc>P0=CRX$H7sdLL2)Lss-Q& zZ&83mj#i5qf&9M~Lz_-Rg@L1fbnX^C_tkS+DR_3t*fp)9h*JniR-={ z=X2y6Fe*7ggAUsRiQgpxabGO6sdZmAqz&4N3U((&))w_|sUU3vnDO8!!b$WADk-hT zj#6984ml}V8CrSZ+z_6zoTyr`lclPgpC_@l&Y^#qkNlV1S7|Ra28ArKYyW$Suo=CXwi0gd0WJwQm5Y%;@lU>_e)o^pa2kSP(*<0=sAd zCOd&51UZ!9Y$*tF{FWaXa`#WqhmB3op zqJrvAdFcXEAWTIE4;NSaCjy4OGo0+dqBE#8A%qlAIs_TFtg`a@r3?ZI{Lo?uN@?Nu z)$`#8;#)!^Wf^99S}svv_o)Zs2e%94WZ(3*_1x2{7?=k1tpHM@)y~QNxdagWb9Hb~a{pbbUU2d1^RxY5apB0QUz+-LIq|5)bfs2HXhoO%}y73QSv)!FGEC+~W$C z``ZW}v<=voLXzwI$_XG1B?r8*^(_la8??}U? zxK5%TUY|c8SbsaN9mf!=qXc#IRdY}vj^7b6QPExM|5?v12njZz0>6c!D&!%{KBwJ z`Qcl2A^asDck{DX+X$qo@x8p=zZEWqVtjB3pOpNXV^<4x7CQxf3jP}mkOd^j;P@wH z2ks|5b;q=M$FvslyIU?s4j`@Ush-;tO@{dCe|TYe4g}iT)(Z0F{d4vzA2c`&Qd`U9 z3KGlx+DiDPs*7%X2D$sO0c(2Cu<`>7bfy0*8Qx2d4p-=T8vv6_h8&qtfa6AL{_~UB z_;XHnWMf-nd=4o5(og?(%`FRz?t1YJ`Bo2Ozva-@BKw5_q5Sc4zt$?X<_1KkVzoJ0 z^^@{!+fkW8<@PNmr1mU$+x%%V|LI+L+1CK$^S%1%;|270m9y<<0s>#!kq2>h39a|d z6&sOu(3c6k;wOW_PfSlieZ%wULFjBS7kYX> z1{}9qIzK&MI2gZRtiw}qIUvrqjyDimJl!}STS^WLKe5M_GuuJ9-$);UZUEYSU&Mcx z-bdI6sg?M~xvK@%DEua52Sh#+48s48e+%RU(AN7RQuv8oBLF+b{{;5xpnelFVhjmD zPKurVODTSH?1MQfz6JJfEqxQ(xjqR%Zt5N&0K2Yy{!2CePwMJ_QYZhD`r_CJ`IY`H zv@6jd0EuhyZ953|oW`>r2yv*7E;qe{4iPHwW>vX6r11$7kPP-+1~`6Qnf+c%-tjSW zRsQhR=Io^bx9#aF8xSb>W2^fFS?g=0wf(`_|%fd_>TG|;{6Q3=|wFNaJ2qSq~OisM=t)%}UVt-X>_&4xsh5KWL zNTC5q;dy)YT*AM)|7qsyL*klyp+6JsJMu|?czJya{yFn*W%V2HCyyCNqfsEa&4(m+ zL_p;TIJ9Hk?TZLxxqoSUap|3atT*eMbr-m1Vrm`iH8vc$W@l>QjsR*G`})Ukx9av4 z0<_}nEmjEF%=6!!Pk8?Pdr8l4BJet65U2O96+d40WuM(&KkdsJKfKUCCrzgemDk7n z<`9d`b$}k^ZxYeD#i^lT=^QjKiaP(R-5TF7(^mqx(r=xTpHXxa$H#khR*#scu7&FrxVjF0uLsDvE z;krL1o_2YcNu07ar${~|!;kYM1S82g)nUVy66uFymN!D_cx1jIfsL_RYOSgnc8RHo z^(BJPU8IEk6}uYO69vRKZs-TmA7950sUUn1TZ)CL)fiG_82US=8CTm zRFSv|q`uTrvy6orUH9aBc#&xBNUR~m_;qypyp8xTVoPa@LK;~3G=j8Jda(3n{7M2< zY94BB6Jicu@OkzKhB=MpkK(B5QiUuzI?^|i*yyB3uUpaTI?!o9 zXfqKTJHlYD9}Bg3TH0KCav=s}w3<&`M@BrMYy*~@w9WBE_{+_HIv$@1>(pNyQHtwbsF-pu-2U`Izpk?4i$cDZ@`r3J*(p0qN7+ z(tn)?EX_nC;>&swXzdTro`Lfi*f7 zuWy)1HtlR0)}tc!mK0TD^prZRIvDsSA{zN_TU}Y`$D~%^E8=o1N~V!lyZr{&)+Ccp z8dP!Wmf~X}&$l{jARUFqJm)1w)fs1Jq6f3;zH5bphFuB9#R+T31xX1SgQMX+L{_0p zGJ+hoOwy)s)8S#NW|QdY zPw3j2n>#XeI_0_5YD!O%e5>iR=(#+M$=x*mH(B3S^!@Y`yQ1-xoL{$}rXSbq<$sOq zFs~>(t?}1{duRnIX}TbhA_kt8bkWr}g@G$x%-{oL?nodK;gu&ZmKGm7rnW#!6R*I= zKNA*sAT@NsNLby5a7fIjpkw;3C*E+XXC#*=Z=Rr5E)m$Kbo$p>P?+h7bFgrD7T%U$ zai2v?0*Ua@r<-2Oc2Fj7tcV%@2HZ#5@29=u$j^V~Fh>u|xyblTEg#Vw1aTrO^$5VM z$wck3T^vJy5 zxIEc>Y72B79~i9Maq^e(p*mmj^%U_`m&%SdT$ws^^vsXBp}I)53odYosq0TVwCJ)c zX+%ar?3CBNxX)CSKql5fqeb(iR&XZwgsI0)u2T1Mk$4aViqU>**d?GwLLycQ~H`n zoos>A?bhuTjK6re6AeqI)YIvCHB+KcC_Hxs!X`B=jHNIdVtrFH`CX^UhM63V;qZ-c zmyv;Q?SdLorXlH!;Tp|Tc$ z)T4#S&=k~0b;3ib<*tGqc@c$A-zd#Ud$NViM8$&J8wHDPgH7kTd#4}?7+2&M7YS5x zwvhEF-5I&FHwEh5OnqjcAk5?Lc69U1YH4jlU&8$lI97Nald4%L2@T2`^)KT(a9oh< z`WZ9RI(0M{CC{~G>Iw=tPW)fTsElBbO@BaeCEJ5op)4H7Azh=ZTB_CBrxqr1_J^lF z4Kmr2yo!AKPP5xhF7ghqIy;|1$Fq?~dX|lPZkOiSr}ah8ndFYXpT>CYh+7^tZlZK= zS^tedF5BCR5vDgCV#mJd$Qc|$5Tcq<(MoguCR>-GA{#sv?yIK9vUHVZ5ll6Oi{9xt zE;oxbVfa(;5*ta(CPaD-z0gcnLry5wzd7XZoRO1k38GI(p-0rhYTGYF&^Rl#uhebY zJZUme%94FAtSrPd<`(18LLJei5-3>C)+Zp84sQL1?ELomBW6@Z1Q%Xs! zZYkg4*=c+=$+4{ILf_h&y75#|-vndsC5kX1uA7q|5&>2wh+OLlQ~PZcnyY@kRFsBV z&HkcLnx40Sny6=Vg3O4m%(t{Fg~ZXh186q?UY6P6B~sGou{B&IdmAM80^n25o5Xc9 zxBV7QLVk@hQ%!6KY|;wp1-*GLD35U%uQ6Po7wu_%BFcckL4ont2sw9W{!HqYq>7I~O2WgnredPJ~t*^7o)2to*TI@PV-yHRmzRt#n< z8?TT}Z?_5y`pdcq)~n>v#b-ifSyGLw=O$0H{C97pJz`3}b}US*qCVN3QQ{91QryNT8&-`Zp2nL# z_LXfld)nK)WB>wMY?*xO9|~8m2;PVyzK(CH{`y(PnF&FJ6DSo|VA0GZDBR^hEc3_-`c*>L_-2TKb;%;tJZJ8* z5ND2zAvRv+0_sdGV<10u7-KrSCnNT8$bl8-Yc_R)@5S`vZQ{dZacl`GZo>&TlsjSB zTGTf2hi6>9!Iclt!dDw;-vgDcjDDhiX#&RGMht&)Up^9av69om=X++EHvi?%j%aTi zpoBndy5=Li%p;r`D8KS+UcaOBghCU}xaQ@+KIRki+~I_!B(_pdJu+LAm8auO31aBz zrY#cLY6eqBFae^ZZ#Rk`we^-I2X+JN^biUY-~ES(IZct2LaOL7es|R5$@$)y$_1{K zz!%&O7q!p{NWomhZ2ozZw&73oD)o8XPU5Yx`l(@%MdQI~4rC%#gO9={Mms}t=;ckS zV#W6$n!PS<$cMS5U0kHQ=Pp4+9Oh`{!Exyxm9lRIolcpS3SvwlA9c4b`o$#ItEXwjgxjdQ@ib*{q7j-LpOgXfQxp7VOajjZwbc=*Ea z#@haJKF%ZMLT{8LVmN1sV-b-D?~#|XONSNgH~ zmFpR08T0p_6Dm20cJn*^w?BGMZy2PMf2gb7fV}2oE;bWcI%XH!t)ZwdX2VH{y;Es} zZ6oX*TPyBqP@&(x-+I?iC`zQxZ!oyss8!D`qs5P6ib*jX(VIh<6*;KewLB*9FhD5c zC5DK~nUVZKd~2w?k;mSc8zG9qX&s&)x7S7%YWlUIk0Ghvp-=gBAbQQ=i4*q90GeDfJZVG<^F*%Q0&`WE*u(R zQr~djcLjY|7GW`b9VW@-tXO_;_)k>zO~K}d3o6$zt#A7dXXleV(zOL12j#opbl)i6 zLn+YS640OZKli!H@$RPE;P$!K-IqSqYzh^`W2VL-P$rKKS?;0(*SjE8PZ{NpDSl#2 z8{Fz?OmO!zkXLz}>|2AB;*K+)j2wGUDr}CvoY{I_*tMPd6S7*0ndJD%n-bRf;UCuG-N9-GFcUWPHWh5(U8i!Hk# zsEn96NpXuihceYZQ0c>n(sm#4Z2=42qm}9~HZES3I|)&FCPm1PzP#YLXGPtx+xWt0 zpHxG<%%ewyzF=MQS6bLwxSr{gR=35^J1IzB?*Q( z^hXp95h(-Nj!KU0cHYy7h_i2x_r=pF_pz=YI?hG4KA=o1xoIs95H)M@3K3ME1G1dZ z73W{QeFB7&80<7RD>|7bj7PfERf(xz(g#7f{n@qy&BsH%&X`|4>H`6h zk7-=P65a9%S-xLOQRdcuNk|u3Ii{0g$7w~<*zDB!+M|7ZZ(C=?L+&g!C2P=l%gX3` z8`YkP)e5LoUPcse{NPbgni*=?%9w~lzq$oA)&xB$uB^C7XA?W}0Y~A1pgZ;Kn7Nh{ z;tN{Dy@#&Lc{#NY?XG+qO@uS!&bZCs)H=e~V&We)VwZsCK=K5Bap__;3SE3j2F*r& z&hR&*o67Tn;pM!si{zR@%Dk4yvWaos=bm-cy3Dp;6YewK-WT+m&csTdvURPeW>(-2SYvIF?L1uOC&WYC1p6hi5dOPV#+!JSINS zz1ozGPHP!lHQT~ywqOzyde^!wMoH#{qouW)wYDGFUtu~$zVLtVIUvLS(a=gYEo_Ld z|KygYN|TP$9ys$ev8@znz27i6yP4#ejwmv>7I4vkrUkcB@~>ytAd3%Gf>HQ_@Ks zd{bA>NPN)Bsr^EUl2ShBMZQwg8X8onhW2ziqB~ zJ{vU7T?j^X1wna=Lab;(A*t)G7+z_fXo<04Na{3m^%B5^RCZn`nHNUMxNJk%s;I*e z#IiFbh{+Pcq2230d1W@C1hjxF_pfji zR@+Weydw)pS`5Y1PTH-930R|x+j(B|z*At5-12qc{8Zka>V;T}evARW$=(yj^9-&M zX`;0$ZPun8D7!R8mAalR21`eK;SF0!)RYS%Mr~h#Bn!*MEv*~i!+FQUMlFlc-fU5N5x;y} zUBxCn8!qR%PGXF1nfnSJJVdYXFMfq_I~s1uBu=XCTX}r<92Y`gSYwHi{ILYtPJ)%nf#nh$^jiXCly~;woAhhh94l%3*q~P2$I&rMxMa|q{pk} z$=Z`K9c~8&+o1XeN4ZS|yTd;RugMh&ET%zSbyNcU>U#*z5XEcWvS z{evQE69gj*p+|;NC-mgGe1@o<;-qYB`iam(%u%12p!1Bh7WK!acZP<2BPjw6*!u;- zA$Ha%Rb-M7vEika}xd=U*wY$Vou-jkN z>X}scNXcd0w}HERslhMW%e{Ks6Pz;FD$JBkSe`fet)~L+s39Q4k>!9VO}@BHQ)(1Z zO6Q&gHdn=1_ifEvsxb9(NjM94!q)3_%*;rS+?2udu47^S?tsRe{=F6$3$es__Gc8J z^+~uS{xokd$IIMoIUF|+sh0&Efq<~P&{?{eDQf|Hr&_2MhZJxj_SG_iR*2!sB?u>M zrjJY1er;K1&v^4!pp+ak)4RJSm^%x?UW+=a7y4Jby#NH7U8ltYjexPhik^M|#; zy|0&J8^w}xn%1Ow3->Z%hxFhA^jmwpKQ);g+%0fvrP{w+cW8gdKap*Snt+JA*8El2j?`Q?V z^gmYR%-&k!5uvQoWn6-O=CJp zQSD{6oRrx}RV9fR;HwT1guCl%@Dd=;DbwlDct}fYONz3PxB0dQJcAyBHH_Y?86TR> zY>mPe+0BdJa!;{7F`zy#m$*Qb(YYO+&)P#fw%rb6eoOFkK8(xqLT&8{UuPz|ml0b! zB1T%ycc`tOzi|raWP1s;+(E9U5}Zk*)5|Jg7g|SN#YAwJrZrvJvKc5{HIm}U8{|(; z)a`W_H2|ttN>R$FTpxZLCG+m5g)&Nc7kxLpEB3K8e!N_cRrR1sSzNYQu6lztAncs3 zK5rWSyN(n|*J9AxzFs)|ub)B^Lac2hihHIIJ8!np(RV6_Wc)IJnjTf=9*<}9Asub6 z2cGsy*iN_+F~Loyr|b0VCFn~(>O@mM9+6vAXZtjTx3Ts^xbxb7#1sp;7+q8!*uvl$ zbnFiqipcbJEX;`LpJ?!A%{+8%%ZO=vQIN4+%B@6sLpD{a!z=B*N=HT<1rf}5DP`H~ zQSopJbxb6UD23aG`VizcVTx0AW)8KBw2gs(Kgo-)|)-6@Q5ws&TVErhPVg)A6t^ zM0y1{zDSoXpsOp&bkH;Rp)Dw z%7MH@fYbE$t}Z1r4$B7a}d3$2z*4VYg?3=7sk1SE0j-Uul@PS<0S@E*cpoBqD?_JYo80 zt=G0&Sm!1?Tb|K8ZCoxliR5ne#T?{`FU<*OXW-oj+A?LVYmw^5Si?ndIc|QbUfNdA zo^8;&eTecTkz9q>qsq*j7#$G1bK{p1C!?3^i&V?x$zUjrRv`7>aFrf( z;Mt%rzbc2fQm=Wae2Hd1qft-r4~%TU#3M|+n6|Q0w!FqpME@j^ns0;eX1 z@Jc|(&44`v^lfYHe3c?)P?YH2x9g`8m4p)|hLraoOn=VKgw81l)Eq9MZUjI)0@SpK z$X(vNxsTU>YdCN+7E|@P_NGc1SHY!VuH`{JdvIwm^76(iJt+_AEWqN{SbD7hj^Rsx z^^C8Bg1~}L@$7T{s7~|8*q~7qP3p2qY937#&-C}$!BfG@_Fq+hW$nffc|o9TUsf@Y z$GkS(A;Il%%GG{C_;ozsf526#8&ouX+ufy-E;AC=TW>UmzD_I$C-?()nA{eB$`mEN z)zY-P`f@@x#^oF0BN5!))3Pw$@RJf*#?7rDB5-?|{!PTWw(sr2r=H#qppbW{_A}R9 z70Mt?>JpwT9^746`W%Z!WSJ<*Q-40rfC@W`InjEA&*zprI<_}>TAGX4i8LtvX1w1~ zdWOO*?7jGM;RKOtBs`5Au$|6hScAvyuaA>la=4i<2GI|U+kTmPo;GtvB=r*|}F;Cm}ecBXbHiCX`r z2Dx~?5Rna_gVBeA2!l(#F(L21(8^lK3?%sbT7+LOHwG++koPE$yq^j5 z5Klx5veKL?Z@hmcpvQ!LnzOL{{AM!#;Iy;$hdyR_vcvb&sio@t*oq;!#!xX50|cp@ zxU}B*w;4gL?R}~tA$A$xvv9EY2D-d65@$bjx+COfGYx&3>;^pFK0BCP-;{g_J}G%8 z$lpYxZwgwj=We(PCrBmry z#&X83WhR9O)&Qbo5ocLH=_aaG`k;|P*-Qyc5bjd0Yp41v!os{_*b^TpK)#_PF&yDW zo0~?+W#pbh``XkY!V?i&$-ExT-|;U!qgPF~t#Q57@SG0b;3JsW-Fw($v(!XILueep z@{%*SSJgj?Al_7e>wQ&)F0HL_ZP|12++z~YE)(~_if_T9dFoN|-12ME$ehKteshHh zQBSZXwa3k&RybEvV<-~ksnRLu#g@*cIM8ETg3+k;!=z~-#jfI}fRy_TLIdh{rM&XT z;|yx7@e-kRB7w^;Flr3pDB^T-gS#2O))nsC8X^|=&YR+HeQfr@X)Ve!!Dh$)mDXa! z<9zAOM=D!ggve@lzT~AgeOWV?bghvALfm!-%mD2v6355MJ1pRg33p!QEkS-W>PF1 z?X_7(A^S@0c40e^ig2TML+`$%2#4)ODnDDR5hf&Gw^msF;WMo$PY+IOsUqzkV2B;ES)sn+-*Ji zB>7`k8v2*dLR@Qp*7Z)@MCAI+QJC1^-o@86E2gY&+sZm&yE4DTgFE*rLsa2hc=ha{ zl8kQ^UR0Q(4K}{K_=_K`0MzxE%siPv!BRWRQ2-u&MspU>l3aVFxNFh^%#$UV$91O|c zceTJhd0Qx!PY;J8Pf3&-3B@3ArTW%#!D4FktUEK>^*#{ix+t{G^M!1`giN{=1{OeF z-cS(Ljr6N-J?&i z)gcg+n@MfG@AoX9J>Cm#6u8=#{5I^E<%G^dk~N*_OsV|j@1=PcnC|j=Mi67xKSf1K z_R1z`4%(vgbXu|=(*?QusBs>Wln>B{UnOk!e2i*`5xGf{Yn%JbS@qsc{7|((Ts*`X zgS?VuB@oTG{qx@2Ao>Om#Rvz7(lUQJ8*WW+ZQ>t}DoE~PD5I2JM_fVtF0jQ=kPLe} zFCXlC;zuqGGl1h4IqEAIoK0uc)2N$T?~ml!_W10@LJ8UTN-${08YZAD$)z5`j?YWc zt>lZK27j1+3WHpfHdNG112x4DHXqG7?r+7(6VOvl z%G20{g1#qD;Lqpzb z{t$G(Q9IFZI+w_T(FBFW#C-^TuUew?lCUtG?mPRmm3zi9hICujgJ=-^okVSj_EQ12 zj_e_AUYKJ+-SmiigHDPIF>&tO~U@-fZ0W6dCuBKXN z#S8arE+YD#@21^AYq5O8HN!08s;8G&D$g%73HH#D@36*MJ_C&jY)TEgr0E`QHnJw+ zA35_IYqF9ZAGl4tpnV{WZZL@JD(WKhYL7D+Nv^uh&}yuP5HvT{CNM_uzMq|@rir_? zfpYj5RJjFt<=a;Z^X)r0+S6ENqF=TfF9>z01Dzd5d6ymzdEOGlo||3Le->==hhd>r z97;7Tl$s`_BNEO`;Sy$}P=~`+Znhag7eh-=pvjkChg>8UWDieB{nX>fly5$GY{S=3 zZ*R$glNz96`S^41oE~W&E)kYbfAH|TbYPbHo{Pr0biJm1;CrDfhw>Hrfg@G=Hk1PM zW$V1t#66l(fZk=q{T}OMEm4E-_@r-GsS-Y$B-IV4Vf-F!4Y4w;;7JXuU1RfM9^Ab@ zmfBvF&djYst3TD+8!<%@|C&Or?W9iaV!CvsEO?LUcsC%=8x7c;IAK5p1n<`)K_k(k zA3u)vQOzKden}S04K!Kk@MpLMX*H)720jGJ!?`Tg2hx@bX>Sj)VR%U6?*fML2g$?^ zm7NluVNw>0p!oLpL66x5;SkuqLe3~L-}w}goI))n03JIvw_fb`NtraIs_!9@5O7Z0 z&y{1opKaZm3c`)4taV|%;8SJX^mM9K0|)M6U}7O?Yvu%2ezl)*CU@TQiU@P*F0h5p zRle0Oj3jGH!rW!4B{dXBUT;06ygVA}Q|MTkIZ|yOBz)V;cGB)TfMHFrkS8yVqOi^- zN3rUeZQaB=V;-u1_y{G!OXi7u41$DhE|%cu5scebr(lJ{jmfm@^UM>DY}wzTW~t@g zx*??Q8tF^WWv|Rn6Q(MoD2oknCAIKMnHWzUkz=V1Vbex!>&GI@9-gEw9^B;IcChP6 z>)3yCDr@A#6;Afzd*aq`4^VubAnkz?{VdUkJ+9|wei4#K8Wz&Q?#OgqXFquH)ItSN zqR0=lN5zU*)3_PxfG1K)kH=FeFk-SnuI>Xf;T&)|GY@ z^FDv_q=!)**~n2cuEDBuG2D{W2{KH1Mf^7T#!tpKMfOwbQ|H&@1FSxGk(R>oDzopd zc?S?4IEua^8f(fwN#g3Uy;XQFq^kadoRA9!Gbxv7-sgIi@Tp;8Z+7be}R8sy;(- zzAx@2Z9bR|qjz515aoH=Vso6MR=JQqmR$@}xzGz9b_1SElmJW!Sgpl+F| ze0gQJ4DGMSxir)`GzVsNpeF+cKfFJ(=n6-k98lKsuXE4Ai`ftFEDgYJksZ`&Ui{e= zhhU7r5>}*Qo^C7FC?#l4SZ7(r{6ZzceaoBf%w*oai5`MB#}FJBg98_m+*w&NI(Nlf z8p7Px)OqAI3PoGdwyi;jDEUod2C}=ydw+E)-|~Lh!7~desscmIlLE8oj2%bJ9$w7# zJk%a95_;W*7SUa}R=+Fq#eBK28q4DxKNFupA7Z8H?lh-y{y|w$3Ue$aZ~7;^XWaHJ zozAOB4i5)?ackuGcQW*J2uF41M4E0njp(ogApJgnbS8jDr^mQ$HDt^qRbM8p!(B|+bV2v zTe{kps!wPixCBo=tF^i$Vs#*}Dfd1@UEQXknZtb{$0|bT9%$F{Ks_kLB`;^vta6j> zB_XrlQE68Q8-~edE5Wky88qea;eySD7I?UKX#O}jK!OM%kxnI2q35nhQYPX_^(ut{ zHygJT=Bf3Vfg=e~{F*w<>uI?1>d3nib)D~!Rhvxh87t(D=n-MN(zOgS=1?WSSHfM~ zkHjPSITa_)nASAuTa(T);a8zh|3svUNj16+B)`L{%~GY;CAPPIE2|r(@YNnCj+=XT zcT!TZ|H2vjrF#BTLXhiHPC?0;*RSbIt}A-8OHno0a3>DnZtxR+bA)f(?OWe;d3*uR zIucSxCiX%n>};$UO2pYuxyc}{AQ%d>E3@o=f>%!n2+|?mgD_!zCPT+|l2D8J*09h< zDSVFaxuvp(I0j`YD&-do2mb{LEo7w7TuDH4m$$UY`fNzI7yPP6S zzFfc9dz7w9!r3&Hl?vw?1E%?-;kzCA^-6+RO(8bnrrt^IXdiWX%wb~{%_i~469q~W zsF_$G*Lk9soUE)euHsgmtw$GpeQPmbTFIV9yeIp5{IN~f0QBRvfTe zPRKl?dw-H5M{RrFFm}1gtT8FA;jE!z@G3B5V01Te@b^@@Cx&!x%(3!e;`Cg@Y)0uX zrbLC43WO+^F%^ls<^e|dFJ*})_vh%ptKa(8)5`X4{~xbGsYkjuX<^=SP?2y(;fptA z>~+Kj8}>EldZ*4F%-blzO}w^z^0}zwSGe&iw)6J`uVGlQWr==--!&r<>MGV#gJg5R zSie7Ly9lLLr-6@)c(&s{-5@T`2x2S#0bgBnuq1`n|FJ$hm|Eszp-yY^r^L4#9}Npo zJ?TEb`U_rXu4rCdqPjpmQzRS)dUQk6yj6F9mHAO6z|2w8W>x%nh*U+Fg~aAI%U~PQ z_eEqZWAD`Ug_k0tvi%{<(3|GSgUT4Qt&RtY)K^Wh<5xzjfFI*ib7lEmj=$Z#?~522mhCuPA9 zgit)SCpk|qxEu?>bfKmnbssYfjPut)O26w{GFzaxaDhT~EX4DLg7zNGP>ikBP1d}- zT9frKISRVHcC6+I9jZ(|St%Bj0>G<#?Z;7rmq)>CVnCX}MicNW2t}Lc<;Ht0IRpyn zmMcSZmxIz5CKqPI^KcTaU^4?ZsB@7$r+^m%gLz!ZiWVXbI?da&{iQ zR+lTQ;PO+N_*{mpNC@uyBus50{vey1SySNyyQ#t6*hov;DH2I>=_r+&A4pEF0Gf4h zL{N!#38#g5yS1vfUA64z+q;5Cu}C|ky1>C@qVawGjd$5J%01^dIX>YF_%BeD*SRo{ zr-3}!}IZEWN76)#yDf#&jL&D&LHfBkYi{{yXZ7eNlp+z&l84tQ=m&4v_(tA z(~{tq%WaL+dQ^@5t9w0?On`R8gqY{&dh`tM7bl=L*Vjn@4{-0W2G>LAJ9QRrW8X=o zUJf~{-&O|78`Ncm4#DxC#u-dq3uj`+9}-FNrgUH)l(DTWnC_U{0+m5&vCRh`h&R5d zz2?rJ&m1?&2-5r17pzg+8oA28HO7|i&zyeGvft9iiiYA`rdGxpq~#fjH*W;*{DTHb z%2JZMr`_4!N~UtrINIWP#3&oRO3I( z@aJN(;nqN5TdD(Uj9g7+HvH&|`?!ip!L;BEKz%@Wx`EM9c}J#cE|6k%IA`&yg@>O2!QGlm)gSBD~W+zmADYnjq!w-czdx*VS0d zovLDnCVonUY7|p*WCOyW%&O`jP)eeh6#%^2VflfdN$=il?3g?6^vqG`-VtbTF`BlV zGEL7R@H^MLv`0NtIoz8saL;JHe1@`ady1_#kt)Tyo}Krk6Aa~Lw!}wULc%!H*B5A{Z`wxL zLX2>Pj|BEr*LXEQc9vBk*LbUQ@PHb6!q*7i!*IaJetm(+%4yqEp+U@{4X& z{L#zkw8i33b4no96lP)}Soy8J@30l~Cy$<#|g!KILK#VKDEMSAhvF=-cn3Tuj)$vlTRB)cBU8Yo(2FD$D68n(S3#(#beD z4l`v#@RQNAO+oo+?)%bsxte;ldy#ud0c7LdTPuL^Mun67HOA{9(kFoT+|Y!5Y+kHg z0P_m2-_507rN5*%IvIY}Hd9nG@~nPt2qVd{-OQ!y@;g*xB8J2RWRlv){!~Si5#bI9|qH_C$W-! z!fm4`%-g~^;PesW9cBt1Idx_OV`mOjlRZK=wTZsP(%Pvxs)n$MeE56=kKAI6dILqt zg(18K@tWo>Dsj%qVO@@kd~y&tkI@ym@eY|begN55;UgK%0ExD&`Hh+B~_0y5Tsqk)MH$B!zHAy&WVKsG^j%~V&b{bDr zNl~Y-tD}O1doFP`G+3UGKF9)lR=2DVQO%d4jMwBHsd(Fhfy)M~P~R;HMo{ufR*>J( zUM`)(dsv*LJ<~0#QL<-TZ;AG3Al{28A8 zKJO<9b*Eu-NvFiF(cn*;Dak|GZDaJy>sSYbPY=d+uc)0jj!vtdA57SHC^kmY%b{>@?6AaSC!$~y6c3E4I^RXKD?(0hGEFeg_~Y2oc#HfP9uz;GmNzB z*n@d9Z3!PWsO3BrpdS5I$|BJuZa)p8(8huMz2+y9Nn*_ZI4w z?_>cuj|OUoiO76>Yo3TsJ`3u@pf>6}fil#N6IvAa?3e^j415Xry`B#9CmU%dFGgUV zmzfpm&(qH{qCb~+gJ!F8ubVlRRVTi}^HX$$QiLS>Rh7Yv2c_b4ii?%N_`&a6N9_rW zU8GXf_)7BH;?1djPr{e-a<$}e=}Y!`wprju0TprXPYci4RPD5 z1v&9`%DYv3E1F=&rb9IJA`YjsWSybl?){3a+eco!XH4FsH9@t;!Fxzm0P&r-qMP3- zxg5!q#^bS{X#V?9jRqRH&*RIVKN%o@K3{sc7Ev_sXAv?t9L6^)32FSm!_t2_cjY5M z%@<9qH*>Fyip2*JEdY~QyhcjS%sBU&{yorpIBKjre5Cw&Vnex~74BQ^+_3q#eELA4 zEe_rr7FyjSi+b$}{=+2}#kVzx;>-bGgbuD`G*K|*No?z>o_f5c>40r^8NI}BzUFD~ z2VjbO^@ZX~r23LBB0L;RtX_23fz`YJoyHOSo8$k3R$nG!>p}?{S-ZS|HwEYwo|^w>PC5q4`{<`H}9KK{}ASg8Nv9h5i66DOUbU9O?>yON=)dY(6R5K?>u_Nts^N&`o)U2ul< zg&3DrUxH?^%T$`LD7Hqt7`OY4Z|BCSYM6^p820kN6d?cdq#^^#1~vO>w&X0eY2E-g%I6#D!Uu&Fx55b4_D*T$SLJ@H&|;O z@&TqZLIoB^>E4)0bD5aBjw2}8q)b^jkN%XaKNcXyG|!+T^94{lqGMe$k>TVkN=qVQ zVaF$W@wla*iVRsuXZ+p*HHfFf!lK+?h<$p5HWanKs-;2xr=9k=j*g;7^a3z{d~q zcWTV4b!JVDPbT>JR4eMg_)jN$Hm&FSmiV8`5noKQBYS1de)3-^OuKTZcfdLm@3U{@ zz>m*rLm0N=e_$2g??<^kbt-?Ax^fD99QQ_mYw9d!iNBB~x0Yn)p?v)p#?~a^y(wPE~#IHV8Cup8Q zPq|g9t4AljQ{CU?pBW%<`oUQt#dZo~RqTT_KURz}X8b0;2)bp($+z1kzoMgFE@q*{ zT>-iK`b5*?ql`BNvzeQwU(JqO$%PpW-x8!rpLxG-r28xnRM8ldbW+E0%*O{|R^lGT z&42xC#VJ1j2C6hS&DW-{PdLdlI%KQKIA%_)7dJv=y@HM(QJ7MD%J#K~Y3N0|$x6$= zPp5HBY3@u2-}>nAmdK}4a>|@&o0rvn^T9o~Ktf67Mx~BMo1@Bq?fWr&FUS|W7O6#XaQTw0`evOZ<$=uKq0Np~nYVK?;>pr&^uhU-88bQMI{T&j`Bn zi~5YscNVxutaMfgH0@1v{MnlG@La_MRg~R95AvYF$ha2^pvUw$;G#uePX?hh_pMy) z?1STEkR}ut4Qi>Ej}j(C#0UpeFx_?j`|6x21AavaNY}vByco4pZx2^*=yj`ocgr@Z zW7F9(huLha{oPrET)fkKob{5ArO1+IR(ESTY`2{7%)Is?zeIz{yNll297L|}EYo#d zeWAV9HbW$WErtrv8csT(xF@E}swlDUH~D{ToYRgl0AmH$wr%rU+qP}nwr$(CZQHhO zd$)hu^rA^$;^fT5OhAc1V7ww=6KcSMH5PA-j4-iS7v1r->UQGf42|4Az&aB0`m1qe z*RSq?BrFe3aMw-@i2Re64Kl<|wt%!Y!eN2Y$Q+KzrmQvXg&o4Irq-JNQso1m6P+Ux z?)%GS%f3eAGF@x2je!ZKpBSmQAz5EYtA!C%!(I?%x98OPiIRxxhAHE8w+K2FW)YnE7cl8 zD)v|_j+^GP3o1z!*N7$HF)&%9m7B)fje`F7YJ5s!u|P zD$ujZLXCvn+cBTu2LWK}rPLJUea2JzdJUh=BbiPz1&e_x;!@)@FXIRZd+&a3@p8Kt;7 ztmQ)){0Z)f$q;rj5FzBVjsMZ|2+VCtYsWkF1FC!@9(PFg7|j@lNl*n~ay!PjPY$DV zz!PbY&(GFl@u9c=_#UL|yU~oIFrb3eoBuT#IeN8if0=?E~!DSbUG zZATtPTJ^-a zlsJoK)g>%~*&k5CY}ZDqkl38UGdj8zP-GCP{@a97+@P=3nVBY%F-@7F;5xgjXUXMy zYFt1Lx}FHfN~xhhdGuZHWdPQej$o=ENV;>-;n?)$IS?lIh$lJsUVs3~ z%m@=0J}LaqZ|rrOt$@8DZ)Mi6V%pKZxLLYHxf;XbkpK1v{9UM$;4)(eVm2>poJREk zi_~mcWst}5Zhyr7UksWCaz1lCb9Pk@YL@}hMMw*12Lx}8dS6?@U)6EIW#r&~o1n+w z#^+JSmVQ`SNC`eor=eUdtmh;t!n59MX1|=?8ts>e;Tc8_`6kW}8Lxp4+)f*7Y6{bP zN2JK%DHu9vJs*=}dW=?Q=0Zp+KRs<2y^0;k$=~JFgs+yARg9rML4jEQAsy>7q#4Jo z!Fy44dMRg!BWq&_pXW{nqir!3;mwsC0{>peF z@Dr)R27FH&w;mR%N^J#l@{Kw5GBsW`h;Nn(iOa7Uo^gJ7aPMiwnKLj(*!fN0cP!R> z`4?xZLeQMf+4|nva0_0D1HiB0BQTmvbXYn?S|Y@xVZNdFVr}rY1Wm-WRdyb}u0~K3 zx(oOWjU*rl1VeW%1`6Xw%3j#I?bM9Idn>n5Y_J!)#T z(l7L}N;i^s^m>0CaEoZdjp0`MERqJ2qZ#jg3Khp9Yt<3 z8#cJeY32>4$=l!Hq%~#U@k3WW940a@R91bg8k?NwdXdi@n@QU4qd5s!sQekA1a&bk z;pdCj!W(wn77u##!g0T(HBYef+xnl$c1~;8UfN=p8ytU`JXpGPv3f+E%9HHUXY`AU zt^7%@;NPdS%7K-*<!BTUvFWIo@e*0tZY=4m)CX8>a2w0tegD(X2I3@amS&85c z6@c6VQ0hWeD0d2mA=sM;H@b|1WVQgcT}*O}Tw{?<{XrVSb(s_xsF#=ADhjP^0W8Mg zXQF0PXX{lHZQ<+VhO^>B5zL$+NvIVO+>=#Ej+59gRECC0KO_~!><*_)EJ7d~+#DA2 z?nJVdLexq*2th13Ez5XFx1kcTv`_t1oCy})$qrbgV4~6vuIEFCI8cNKSONrR56lEu ziU8j(M3BcF5yzSLs^(!LHL>Pbx$|CAJcDN(HY7Zsi6#@_JpvV_Ji9YrLB{%0+BpP+ zM9Sfji8?jPQ0)BjT9ZUyia{KGe1qLlB7_#6IyusNHVZ~v2NMO7?ntvyrHgpAL5hPU2!-T zZkC*7jFV&mo$XdDI^N5>oi##X?Zs;SUvsEHp{>wkND#>&9U!%9_Hnp`sTGlV%1oH* zW$Xf|?%rg6?AxjYmiTz9c_dqdDz>iVV+a4FttX{0)}MPIgDiz}=VsY~GKE zB;oxq8yw#((7bV}(9)~qtyvc6Rh4(qZ_Y!t*wjCwB^Nv^2jVXWeqK}oWlm0ob6z~o zL@R3(l@oR}oX||r_8jre5^Skt=Y1bUa#S2~%fMC72DD`yhP0A{y2BLb;=x4^36nBs zMfJJ~s5;~BtbAmimm~<+X-?VKDd+mmKa{Xh)e_<{Bgoa2n%gg_P1VX@9`px0Rb0c% z>iKzU5wbp;5)z$iH~h3 zd!I2n3iw=!ph+?i5wgtkd~|(b#IY8-zV1G4qMz15)qymq(^Ir7^ONlT31Or*=<^wP zk9a9=9*Fd!2AQDp09Qb$zdPKR4-X9zeLZo=A8tTpV8G1H0bWA!`G>z_Hj;LDxd}Mb zJ_*M2lHOW6QCZ&m=J7z<1X;Ipr2|Fw3O=R9^C*py{D#8XrThe>?th`3vCKc42V0+x z_MYJ(tus}Vve&_-maD+7=hRjrt?~Y)1f+^I2zRNw7-^QO0%hl9xYs3xLT$C7QGvBy z+vONV1l_;_U~@{O@=-NEa^2;jfZWG!v;#na+D&&eUs0iJGO7kAos#j z!m&n{vr19ZY~@oyJ*VH*CU$DT-DG_7+W7EE=l7vsJA`mzSEB@> zzChQeJjsA-?e`o;;mRcBKALTozUGoE>-Icj)_}rqKFaRt5f_0WARphrn5<)DAXEp; zgwk}v%vgumuDO0*gyn9Ua7}^`^<58${zxAejCJu9_>rs|lB-733brYegA1uBdAp`c z&6JJY;3#g(rA3{ooxMu|xx_Rjt97d5IfCaq9N_Yy#o(+46TX^~E54aZR7Qp?_&Pyw zYR4MTF$q010VYFYf`n4cHKUDtHy~Zd(2JfKxl&rBS_pJ0n*fSZV?qbr%H>h`O%BZ& zHC06uy0SW>$M1dTibOsd3M(!#<7fCfe?39j42PCMyWZjL_j6a}a823nF(&7a!)JRX zt?_0VA)uqb!qQCKCpPJ8uG|Fp3(s*o4p3#hD)2jR68Zzl%v6k$ENYC*Nm^;=-f2Wi zy^D?R|4fS6j)h_PiqakivZY{lh%71&X0;y?;LrH0r24)g$+7@?mAHJ)eN~j*GI2p; zE>P-mQvQ1FP8%Rbn8tagPv0pLI1q$%SpR%C%OhBNGeR80alm5f=Ry^-R=f(=qU~Zd zs4BHR`Zbl5{HR)47OJ%*-@92j&lx;$qXh%Cu2x!z?h!0^lmGQ?);RIRc<)pNPte?z zzS(ZCfzKBPtxoBKc}dDd4l8Z4)7b7Q0pi{H4KqP31!p zpARK-bX!7!mMkPv-K>7-m^*Uj=hm71^7K0Q;KOkpDpYeeBcbW<6l+G@qskS2QHw5E zGvF*6!R$UrUkyNZ235SrBi>`8Ek&aJj756_FV+@?PM~x}v3kw5UO*0E8;nA78)<7p z)h<)y!i2uX5dmlPuXAv$VLGH6lxb)A0uulF5->an-0!9b(QaE!>MLKpFN%p(U>tGc zl5Jw_`}SuBhkmCyHh>fomG=T`tf$@(A`PY@>K5}U zgj#hY<72Yy-b}K?${r8c{gwRG9nsk^{aS=yqA$qEZjV*@GEMZ@m!7e=gKt^HHX`t* z;mh{mJGM+vS;HfKw913)Z>U};eU@HyUkliAYo~sgZ0&pgo1h^+&UC}W5qGZVWua;i zNSrc=m;hDM&b7wBtg&5xns;>FTCV2hqf6EcvHV%gF>cp9&P!rLrwwJ$qwGhpME1JC zNnveDV~z$Fntx<=ggfSYTe+s|qch0I(9FkNz_2255NL3Q+~7tNcVyBM$;zvYY?Ss)0EF94 zvAy%rBkj`^!~Yi5isAn>>QtI>sE*WKK%*+WO~<CPF{<>%N?y@M#d9^sC@Y9iG6Ez#(c~%85}U0u`23G(^2H|{;5{$=hIe9 zoh6q>RM79Ec5>nD1hd>AbhWGod17x;Hz8GnG82J{^qeCx${uQEEZT?ID zkaiV?A=MDsfMq)Hr1I@bG^xJ91+W3-?}ckpqOos$1(O$;Fe8eA6!Thqc6QXDfEZJ( zh_DG$!txB0`9drKlFW)Lh)~X$|BO_!e2G2DmM%*)GHW!3FR~RWzPu)>S>kZ{M56jM zh^r&Rz$f64bX3)s*+$8K?~eXm%FS4@wnJAlD?h$%imlCRe5?&gEFw~A%wP8`+$Q5- z08@4zJ_$Jyvny8U+NhvhiD5_mM=BpQ;zu$r{H)-0l3aylqD5lD*yM#;@H54mgdl+k z1W$+($E?wrb$*lc%6y0qVew9y2#(zw(|Dm56?-1`t)wI++d}RghTnzVq|&Dn9^Oc; zTRwKuwRdm($G#{Wsqc?C6tuFXJ=PyN>%GbDwzhfqV{Zf*X6?1ERJVWgUa%bI_HEly z8#-}xg*LD#4D1-@n4cwtm`K#j9{zB#+dpzn;FAfRxB`zj{Y+U!K*zO9QqLGN7-0p@3-+ITXzH>>1t_>Lx=7vwlKeAFM8Sl^4Oc7LUsR0*ydPfq7-e}9Wdv45+MPsW(f4E7M9u^?w@i- zgm*i5+{*3#qPUFsHtFM=W#c5Nr@>%d8ex!2cR)a2LL^m&&%vh50OZs&!KbSn{|PS9 z!WiimzUW_!q1Q5PJEt}E@dd3@N8^ zsytrJ8)F(-jWs)~spCWFeEH-g?4Bu%?QZ?+T+B^#nZIIq!Qx@aG(HyI(_Meu%#82O zVwtD$W>T=$-&Ox2SPHGFryJtsD)FT5MnSaUU&3aa*a0nJ3v*-Eau2|&nO688aQ50B zY9h4}4A&aW&|L{nFr2eoetzqF*xFG}EF=jGpb$u%FF|Jnim#c%pf1nOx|<2xQmNF1 zT38%PFLVc}tVw%jK2nr@qm|a(-(&EPzHUjI`|^d7^@vJfFHTBW{ZPbi+ti@MBF@GH zu@k-J!E=NKMleqw`s+S%Id}JUA5M|F$D{xHz|Cmy%`(J!o6D*JH?T6)Qsj4&y2$80 z71XwbHZSMqy`0XQoYoNWjxS)?#h#p#a9uieNK+$qwK#_(d-JcZue9%5bqaXKD!QW9 z(PXP=tQg=>KMnrh3=k=8BE@4kp0pb~)-kPp%=^g%vvb|v+Y+OqpC!7VHmY$5@R4jK z+Tpogeb%ZGS2yZomv<`ZbA9F(^o)UFermysLp>zK zb2pQ>(uo6D_ZR+L;}q^XTeu!wPf=)PD=S4zSGcz5h(mDte)&-4Q%;_=|4<(qN08YL zJr-u(CU;y36bk{Ro(N-q7-$+{NeIVXscdgNO7BrncccHV)e>Xdk`%4Ra$}K)j(gA@ zsNnGx?I#U;ww)k?t*yXna;ClAntI8VQK_uAm3^K)EwQr~($~E~shRX!*)dxI2$v8j z0uY__rvhDZ?#)n7&pNvA{-wxAHiX0Y=4YJAD4pr$y;@dgE;9t~0z zt&M(~GK_XW`J7UXv2bj{zkaL^DKNz@p#RLe*Z+n2Kf6+8=5+}r&oLK<(WMyZzl~>S zuEwh@DNCE8Q~T2et`Qu-&)=nGS3)*B3K6vT+tD3+CLpG&^|ZC1_E;it$tO;nRmu5RR9l&ccc((*Pr zVUDidyq@HftOG)8OHig$UG>ok5Zj!_+kf$R%=kaCDYNr7%6=$C=ZD52OKKT6SlKN< z5%PELT}NBp0go*`YN`ca z`dunMoba6WsLOB8fW$~JA^0KTZfp6ux~>TniqqSgZXKQZat6zXOONC_$^_&-_g@Q8wD#^HVNnGWP98czDVc+Zec!)&`pC3zn? zxe)C|IN2656oji6PvpN2yjO&8qNyiHPhQFtey@zmtm(@bQ-YgKE6CnZprGdqXiIo< zfJel@S}q1(>=hJw$Hzj@>3?>Y+BH=9;x3olHO4N<&6O6`wH;pM|3PLv_P9JgFeDB& znoffVi51OMp{USn=?8YR2R!cr-Eeh$K>&fdl`f zp97*R?4bbMU|60PuDiV}7L<~ok)C1)+$+!q<68~RZS|PYyTy43qA7S(%&7^Z`KjS` zZ7UU%uhp+mSd!aQYYgThP{6pDGYZM3cZ&x(*3N&etUD$^0V~d!;lj&abU01|Fc7zj zT)b_DS4;VU@OQ4ep$YzpHEvJQ6t~I>&k9@aA`m7EKk9R9O*y@K&rs2#5!k>Ki4~+XA9DJ|4&BHrub> zzY4@!!jom5bA|b*2DqX{(68n-#d^eu-SvfHhP-7|q{yp+?e1W8WswA?kqxCQAz&vP zNL?=(RUe4T>tupVPIV}rqtg*!UirZM`Wot-`)+g2>Z>adV7~km5d~}7%dAdFQo&@l-ItpPM2f?-wY z@}0q@bQN(rOE6iXx0%!=Za!se4O8frvT}x6MZSGGI_NW9EX*-E=bSnH{pf?-krb24yh1Mr!be}yrNKwpXd-um+}zB6}8iSl=_?X&k+ zyVlQp=?+QSYteh5ml(MkqpVNL%o_U zw3216<~j)BNP*#xmPW7RA*;znp&<3FLXNJpng!NMXTAAWoDe#83j5tLGD6IA?_T4YE^$;e#xM zDM)%tQ_oCOf7b7~^6ESIt`q@CyNG3xQ#*k$#zXOL#xb7IA7U%W$;-Ec1k{es7qqxt zUb=G`w#O?N-vj!DtCtrmyNj`=st~-YrOg~gshK8RBUB`#(SVO_;-4_u1-N5PZ^RW* zT0SXZ4m6pXm&Xo6dIrSt7vClm@M*VicQjx8^~vTb%x#4%nqWPl9baFvi(;-$2fyDg zEChoL>mM6faW}+{X?oSQC*kJnuWBxFIf#&jijossaI(<`l~jBn0VY}nwe-4CJ*fpn zq1&L#?QwEAbS&WRZGgf5Bb5XyN)x~zNna<)Wemv1C-b+6c%F$n5>0!vA#4A2(``NG za`axKPUajq;>udCANQvb+o{yqb#K(cp}7gp_rM`zpQ{$chc^B}ccOUx0l2Fgh4YXG zCcbN!Q9J#vQ&#Dw7dJ!b^&f~Cn19;rvVxKKVw>I!ex|Hc4?tB!o~!r4F`863GsKCg zX_aJxN7%c+oo8%{VAKj~Cx_-ND1`Oft9O?HuOk_m2`Q@O7zNok70 zX#wE8R!5lBJA|Qfsbd=B0JGG*9IlgRIFY4Mh+|S}grHgSkl+3m6G?#=Zt>Esm-&ZV zm$x@>aRK3`qJ@{~9ZCRv(~gEkJQSW$5w^XNi50@XUeu67qnKE2l5Fq<+E?0EJgMf< z_^Hd(lir$gnKyecd^?&8lJ;zy{&kQQp6RMGY1B?$-Hg%SK~_}aTO#QCA!+vj(V?hv zn1c@V!;w{46agofvRTd(-ktT94NtFL2E&W*AvT;~e#mM1Q}A(fTKYZkZW-!!XqV5A zqpS*!&D+q$7t!XJxCKh6f}}qW8cJ^_$!qlsMiAYd(Fr^_A#LLE5DHqLP7ut@dEAt??Y8nPHt=$?mHm zJq>H(m0SQ!*}a`#)$^7C7Adlwmug+b}Fu#DwJ zOV_|bSCr{bSA1#qlIsglZHw&Awd;><5v*ETjoW;Ccd-gK4K^}|DvK02?!s&V#ux?B za3jVI9q(;d=&)Yv3;!ru?X^$!e3>mRF;Gm9dD!R}{A^B{ycc@SLH?yd!8uOaqTchJ z46kX(MTU|WQ1a>oh*ESMkF27tC8q-!CxXMS=vt$qM$I&JCIP2h?O5MNA^##?Ju~WY zqe|a;PCO+8*C>UZJdj<#vzul}rf`R`=XZlpO9mo(w9QxQjsM`*lJD)V7B5t@X&M1+_9o3%&_wEewk_@5B2@WG)Z% z>s(LG17np%fWtX5!+_qH0A9PgA39@hp|5lSroY4D0>uyoGAqMZR>`MR50gTBi*J3J z??QrjZNMTVR$9|HVK*rV<7WJcMxl&UN_p{sWGFxmMQ81161!eahaL4_63W6fKWjPL_ zL^O?+R#BhH07)t6;NzryFmu<&nRC(SL4(}fJ%fxjc9$R}xw+a)S52&SbITfC-U^wo zNUPxg>abZ9;_g-}T|}pkE$|`5-NrpWqYRc}5%_u)0y9b1{|V~Olep~^i?^J&KVyIFEh(ewbB9q+9(bR!vnzpS+? z`1S&IfZhUput>aoqRry3gC8AX`>!@qr$r-|lacyz?+rP`g4fEW zuV2Uzlo-kQvVNSixvh3qY?Wtz9}$}YS5)Q~h^^8HAg^Fm~GdQTh;hbwt{ zi)49AbEYL$uG2ellk(Co_4F2*+WLg{`<)(()9br%E@*6Qdh(1VVeH^BGjbvKsa11f z%Nsy79=}2i;$trVS+iEF+~u@w6IQ%*SU?r7k3#Tl^t4Jfm2P^4n&*Ta>|a6h`;LAM zAp1j+pcb@etf;SldVl@9MC!(|TwE%-)T`rcUgvf}49+cl@2r)Vr;Q^@Iu@9F7a@DwAE=7;j$fVMIxuh2 z3E4d_)jMNYLWM6r!hZ{?_Q3*Mz^&UTiuTyAeowG}haO~a-K?wHT^Z|VIu#*uLvOVN zjdYI1s6GrP!aN}#2;@Nl2=9M3tAS97S;w#u?m6P)DXYR1C;BgxE%`*%td#F$@Z$Ic zo$~W!eiv6p;AX&$IWaQ5y@V=E#yFtliFYC();A@E;}Uo6c7z`i4us);{o7G$`uPmH zFFDwlv)}Jod$T0Qt3sy2EJ~rkdxu!}lifmLz1bH$x0)ZP4N-es5vz(T0J4FHU_2-9vyv_qP`8l zUG7NzDLRiv@bIlMs)-Qcm`q#C z%?yawC{cBiX8Ql9X!x-hh>x-0NAbHUYo#$1+|V4kwK5EI(C)0p%AtAko9lQzNPDE) zTiDNGvfn#0hkl6?mLgEUnoPo^F9;rR`;XLt5Ntn2JHdFUy-J>h4fKTr2?hM;otZ-m zea{S&cWLSxoC=G`-%vmEL!$mT9ZcyS9ppWgE9%{6Vp97Q6qBwaGR23dmg$)E>SLud zKarPcdRQPI!h9$Npt9`Q;#7$vXr^Ppjivy#&I8S?#5U4+YA9*LJ1}1#p=+Q($NYqf z?K#6LSdWiZ5Bhao;^y!A?Z}n;Ymv2nh^c!OiPGQjU^5f?o56O|X0?`?F5|~|lyjaO z{b#PtOnEDsBlKpI|1zypbc}P&8@w}i(JV}Hs>2fnCSXg!R#_G#fSE17J(mzctF#U@ zZ291=N`_9PLf&lqV!@YSh!(TUNbOXzPnqzCDGt5G@Is~eipX&U!My?4!x73z0Insy zz@eS?_^#<1s2?!^!CF-0zH%jDTe2AWEw4BL1Os26v; zm22R8W+=@`mb7D=QgNaIklgN-%Ul`Bv_9v>DN7PTP`nL21cI?CfCdCgYY(eh==mCR zDQqdqeQPt{Sc|@4;)wj*QbNXs;49^~HAh2koc=8IeJ!HL!K6kHfO~c_{G*eF4O4iz zs&XvRHQ{K=EFz3S7bt~BF0faeWXB^vRQbR{FG9VxfqK|z-W<`(A+g(Kyox)(v%d30 zuvmLz4+#;X(5(M?spY%$&xSf@w`+&2(?th47WB{Gb)pT(`=lA81c-#ymEEONrmK7m9JSZY;8a8In;t8=fa zdDcy;WfwuVX-!-|T8v)A@;}eeZ=ee9AWVajA9~KX6?)pyFZ)j`_9MI&ZqMHo zimbUv6w&8o_6VAz>R=$26Oo>o8X>0i>!%T=30VxWd+WBchYOL#8nUfOse9*$M20Ku zbdp3Fvu;{k<7DOmc6RhbBBEw&rL!Wdd>q_!3M~_d)RQ+@p zIYtw|rjaV`>HxS1z>fO-J%jR$&Yrn${(rj-|8ny#o8)&VpRG(F#hT-HQq8bhB;wxk z@$lLrCg?}N#4+ZuQ*~bGwu?sfx=OoHN%0hKu_K=7`E&2{cfO`*@I}5wULz7Wu`$hP}cThS01C32#g}KGLxh zwl@*&f3a|m$}33Sg1KA#TiB3Alim+todsM?xDerJ>F>4ICxL2NIl)7J3j8nY1kh28 z^MiB%D>7A??2H@)%C?{nrV+)2y|z>6T&){LYU?uPI_o{WiD95j>@mrN9v8dyHRC^6I7ge>x`H%mfkM5oH>Z`XVsE=9Io zNqU$kz|AgpLu_7udaRls-<2X&q-=$1zEI;td7+oLHM3mB4SWW~yS)ULHH17403nq_|&jy7Dgyj*%Wwg8p#Xf_tYV(N4xxDBjz(%~|z1Y@YiL zM+j*y>IH8m09KnM8%?%rw;2AqK+MuDKaD(z;1{v6z;aJi82>@Sx6WoQP}90cl~Pr2 zQu9Jrbnf|D5P;x%RRsPGL_b^#v*lH=zq>sgrXh`|%$>myo@ZwVe1VKLl ziL+=NGYSrG;TRzM`E6UF+x)JW_Zpw%91#*9c6G#uNxVp*L*m6q96pNUGr9nIS)Rd9 z>WB1Xxly`=d8FTnQW8f#Qkz0V(fhI^E@0dHiWaarKt+@ivsgz^ zw|Pd$6DsoaB{sAb{EL{cDyMy?;TqeWV$c~rQ=RjVw>Um%np^$ zYoWXyVRjIoz$X8Ks}e>#=_6YuZ~zsz?-UpU6gcfWZT_!)R*mLvXr7t9nGMf_TPDu(Ac` zYP29$UQt6;lGXO%F56FIT3)4GjuisACa*o7Z^n%8i>9r{njz06h?vZyxK(1zsyvku zF+)<2gz9H2gIp7vAmlWQ7pqG>;wae3e`~!NiO?=jD##GFXzma4QeG4}fqOYW(9=dF z?;3&izg*S;j=!k>W^OU8`H{{c;cqH;sEk!S1tI)umk zpx8wH6=sHNOEVg_YHTX1kfXc_$D;l6^p)}fXRhCIS}F~8_^=#`ND-vFsfe>6A$zxV zA(&a;SpT7MoU_6hnf2_HmQOx?rl6I$R>U1e7x}KLFoG=7fW@MF5lJxW_8f}gZNhQP zTHh2agyF@12jM9$v40k)y^5#wT7ySnFnkW=I#@SxoO)q?o1 z1H<7#itOCbY>EkeXXnp)c|jd14vXD3YLl$!u_?$izwneZ^%$={m*3wnhbH(#U@qmQt$qPKj}jt z#TO%dKj{%{xpHa|OJUJ2v)Ho3CRNI;$ZQ_^oTXc@*3ulN9zNA}@Kdd@%y>%YFu%l>lC$Tt-(0uI8sRrF1CCL#Gc z2@kz@rv)_v@-$S++j&moXVE6?mnzJ(fXp!+E+I2SyOZ_f@yg8YRqP4$X;|(4eLghA z5l{Wi$AZ#M$MOvSmL(KyOIHZ?-ozcHZLiLH+k9R<P+?W1`$ z%j2`fC#nU#jYFrNNN&P3WL*UR@ye}knjD|J7!Vyu%N3r$d9LO$JBcFyEzUyDbroVv z6(K{b@6DV@;PBUGjnQZH!uI>4IT{YP!-lCFS5Ui+J?PV(Dob68m!gr^8_FVjKv(ed z)LuPPF;CeL)Gk4Eo&dn6r0uUnYh0Zgiv8)ih?mD)fN$jAV2!KZHWgUoPpxv235CzSHeh!S9%+Hncwff%sxwIB*ecOQN z7_%QA;myB^)=3AMi|OjaMi=g#W{oRjmxPJ0+uEdC%RQ~=$x*&duwVCA^i7u=U#-c% zO{Q`{<{fVqZacGFE%s$EdXoUjv9or;i|MD$zQ2K`TzL#AWsqSJA+4= z7PLMrTZv1Ep;LS*(l}_+iPHwm=K=$Hw|Z18KFDN`eGil#qeN$P6;6GT)!?mNKA3SHJ?6Xt6!0MVcc74v`y^^3;4Ikxv4sW zH8yNAMg5y1Jskz*fxc*Xk2`*g3&PS0tWrTXe+-_QHI@a4^A^hSB8?pGzhZc61KAo@ z)>)<=p3{$6FAsWa^L2U`WqW}m4vYhT2WUbrU_kf^O+44QvBV}2T9}DbYT7oRCYnm% z4M{Ds!31Q2sn9{@LF1=m3jF?)>PQ354j}C5YLv+-Xbr!MM@jM5plkm*hd;>q%%~s{ zQBq{Nm_7}q#TU~_0$mZ*1hp~mB+{*_m>c2OYj{JmSEY!xH->*-V2-MhcrpYPrj>a~a}WJj$Fqe_B>1skNsD$a%=@I-nB z$RXfVj3y^2R}SbD#a49fi+kre%+YBvP2JdD7=o}1A0@RMsgU*&$HLrkZ8mhn?_NI$ z&!G7jD{ALz_QkGk@VBTIG6xrAs1(a~G~?L}I||TAS4VFGd3LNg>Dmx2ac1jT(bZGw zv?3Z2K;)sEZOI6jJaL2!5}ueUi%I#DKa3PugY8kO{|>}RORBiJ%@=DnO` z2eT||G$pF}g6fRMG#twT_)SY({;^h*Sh?4JJ&H8ZqwE$!>6*+5Mv3XqI>D3(p^POC zYnHd?>SaQH_S9+Cdn14>6D!fMMGXtzA>H%+5tfn<{pE$da@$>|RZT)JaB6@2n>QWM zd=Y#E8^=mMSJjS4x)Fk_*QG1YEz=O_&o@ytEZB(dXW1#W&|eHgn`3u#a@-7{f>kci zI{{6v1JqITNT||r+{g6uj>N~he*etTJ>866x)lV6tj6&-2mVJJvTB|*RfBYXX|42M zgll!@3$X71>xkH(d&3_b1e%HfjVOE%d6HOb&wS%%Fd3mo!~KKC3jMl>-YLI>l!E}& z^~5W~0)H+!f%QG*G~rO}BR|8dP6TyOkxrb9@p?QlSIKzL`=I|t`H|Ajlli_lsc9IJ z@m{fhmj@R`gY?|@r+mb)6oA+f-r`2dg@=#W$kY6o2pKhf<@6@SF23z!HLa;2EQQ4z zWK=7HhC<|5nz%ZMFd*&I1`VCa|j0LcumdtFbS1y?x)qMwrZ%Wl+Ffo z-welwVeME64Dt}=XCsJYRT3ip9lH}Z3;FX(CQMBqRxo93>CO=`2hIR~r&X|rQ%K** z)h;w{h>V0+Fi7+e$K|{AyqD1&x6a58b+K~RQHCcZu@K;p_^uu5PV6^L+ghpqmR|$e2z5YK(r&~8ZhB~}QI5! zqEV8*KxkZfk)A~u8g;tzWS>^N0_29>(zxJwK~tgNBfXXV{?mfCX)_=BdB6qV-rG%E zU|YNd%hQgWX@F!YKK`Btmpt+F^)3=Ef+WObJ@vvw@NWc4t@Eh%i3&5O(0~OR>Yx~X8XH&)a3D+~+M3i2L&IHq!5YNw6^x}*` zqQCHMPvn}>7v{_DJwiYrB;JnOp((FPZd_-xC)#SB2jxuLDJS~q;5KRh8kRmDp7M}l z!g7PB-RJdmPZ9CsnyY(x@hemxR-~C%zLRixR8HxPk3noH@T|RxE@L~x2}xv2R$82Vhj6^8opw5-6QEYIJs>z^SD6?~LU zw{>Bbq%i}EfB?rPyt240gFWW9=YFR}Wz|v1U#Z+n#f>IGT1%&|ORCF7lowdsES&{V zb&g1tRHj#>nf=kV2YC44tK_X0a%c7}sJ$KK$h*gs@vImG2%$`Hoy=66&TlpSXjdILlCjm0B znB?8HP3~KKqb{Xhyh(OyFMBR6y+~JSl?W2qNn%c`TW5-VqsfDrICBb{gOnN&+3ch} z(jIm$R{~2ZWZ!;DWJx*CUT*|QO`_N4L#xd3Z{MU_rUXM66O{ihW4H6(I3#Y)ck`jF zwzy~;QVVqit4c~F47`~qJ!R?Eq1oS(ga(Am$N;3`6>WHZSsa@9_lXCBm&8XnUpHpr zmBojfzgupa4KW^LuVBk={q9Ao+XbQAOUGd42i!grag1|`cpCv}vjx$~G)nOANnVVU ztX+`v7973|%m6F?iCKk3TU^j~j#aUGX|Qg7sK1?3FJKv*qSOQJVRPL?Tt`~Ca<%Hm zUaatsjNP05{wobLx&B6e?a#q0jT%toV&xF0c_r|Y)3LiTXm179@Plb8hcl99gr9~0 z5f7HQblzY3vMn|{S&^+lyJ@6irw)z%l@Z^e5~c;#OO_hS5I2MhP$8{U#MO9Bv6T9+ z0Cfe5`lMHUILOpwJlQoBHp+|c&iCW81@_&B#O&{*M}L(cV4nKh_yNtUWi4paoleM{ z$-d3IwcuJqr*~Q#asf~hG5C&CE#(*J{5g`z$`lcxG=ZJ|s1#*mGTexelCHQwB!f%c zSvtd2oeirbs5Sx=*b^pVd$u)hH~e|L-&zv#K{|pDMql}gP*Yf4vNZ@HnyXk5+KVxv zGA9@*+G`t(&QYj6OS1owfuYg-hs@&Y-{8?@0`)V5CdKevyXU?HSoeR#1$|a^y1YL zWO8@8&N1T9Q28ap4%M|!XJXxHc&jC!Y3XnqZ6~JuQ=qNGL1UfbP596;z#KW$M7?ii zEE1bcPrAb~GCC$Rw*W~C3V>pspa_Kc3w<92*VVCB3UBH9(>TOOB2&5;rcYkqg8a z{we%nW!!gSw<7zk^Z_Ry@tM$)EiHmx62Oh0W+=phh@d7+YI^nuj*T-tSxzxBRtKW3 z>I@c+mv;=Z>3FsRMhs*4{}SQ$4W%33Cccc{xA+~68?`;8HZA)jK7xbCZv!OU1MT;NRrN=1i3QC zRT>|Kp)Gb*MNAJ_#dJvRIu9)Ordlg6y(FYWImll_nE!o5faVr#{Wv zDsfbQXY%$0tokZ&Rwjm-#6rGE|6;@AG`aRI8#Wx@b|Ny1TN& zUvBe{XFi^*Nt!cc+S-;x+udwvwyPdK9s;z0Dk%-WJGUf+my?fu|24xB%P{zr1Kru0 zzPGe!FT~|#+BDX52EO(Nc}QQtHX5t6ze)H)-1jW>GJ45EyadO2ExRFyFrSH4I z)sIHt>UjtT?LUzeM-?g|z8N2$Ch7tcy$jil(bFR`BJH`V11=bMFS<@zvxhA$iC|aO z=Jj(Sf?VW1e##=9#t2TFjc=R34TWGSz0_b-%CPxi+DNdqvf>1iui}$)9_K||S@=;h zup?!`>by)qW{Cl8_=F7hMLejFs-FTqNkEF}|MT^dyA#99inc2K+GOhSZ<=SB>RdsL zxJ4*B$Wsn-0p6Xq46Q9yR_5V*W?|fsc}ALhidV9;7Vi}w0>3+|2@D$h-Wb&yzedMz zrpmc8ar1?f*VU~@Iw2f-PP`6o0z@w6)8P67&Y~or52{h&Vks_D z&ap{z_{i+VLRT(d__nM(Sr6BMp~ekfa?p)3LW1~Hm32?iRbnj)2j`^<-_Xn;*Q+Q` z3{7OuC@+)cgQAJP!gJO3(v?Jl{!gm~scFJIwNz@}Ws%$@ zLn{hpZe(+Ga%Ev{3T19&Z(?c+F)}eAFd%PYY6?6&3NK7$ZfA68F(5HEH8lz^Ol59o zbZ9alF*h+aHVQ9HWo~D5Xfq%%3NK7$ZfA68ATc;HHZUL{ARr(LFGgu>bY*fNFGg%( zbY(I55Ntw0w|e*K0-e{SXg)w zC;`$y2cXM`(Hvmv1yBHjOw_%cfUE!-lYa)vj;GP|VoU%z6Ej;!4_8}jfQf@SK#oa)383ic@c~)`XdE2?ra&tbI}3oL z1wb9B4bV`NR8<2=t14+ItI;xjBvx~Ca&mO}FD~M0>Kf7v00~h=bx8nFlK~*Dp{D-# zst$Dc;BUzQP*nf0|E=@k_%~cZQe9MCM_H1U`R^P6tN?eQi>vkDvj4%2>Vp~Jf2e(g zTDUmc|5E@!V+8^^@i8-dcz7^by19aw99=A#ob3MLr*38K3h;1ru?2kGU4VALe;VWF zVE)lgkQMM>3H~k$K;GI6=->+c8zklUFQffOD?frhaM1rD_R$E?-<)>;atF8qf&W&< z%Ea}bSb1e-d4Ro%wF3y~VB%o*;RrGTxw!(2|G_@KKy&JU5d;Fn-CSJ$)=>Cwlgs}p z^WW6P96x5JZ|CD@;_;s=X5!%H>ixfZ^Y7C(b98XEb_KcqDyJKx99_Ja|1VwJIyibb`263vg|&mZ#ox}GyE!pyI9NNo0c9or%liXD z_#ey?2m-JGfX)D*rMs|MWAlx3=^8H_!j@&;tJDN~7rL zVsB#iA2w@ODQi!lxw19L%<5l_{TD6^GWpQHsDq^)@MBT_fi(VZC%X^bew;nle~&Hz zBP$Erf7m`GX=dvHbae%=bNpihezf5~2!HhcFI@n$x`v9lj1>LhF?f5gV}KNSD#sQPED26Az<1!`HFe;m{Q%SFKi z&=Oklegw7B0DDj3jPEr zlA@#8noJtO=?Y7MWgTR+M{?PDI>j6_=`C?B!~l)aV{zWL&c|rBiFfqtWHSA-mepKn zkHR_DUzjd-7@kXL`0WHsQ|;iO_eadwZzesW%XG8Gd12V2axhNaouBrzS`LM_9sMVi zkD53#@!ch^kAxphRj^qM_#%H@95s%g23-BDpv4T4bnfn*SHqw%Od83xp=U>IY z!NA0*Q#6V3daUoML1QGwxfsyXTw>?a3zUmKB?jtVs?tiZ+r@5 z4-mU}L(13UdZF2&P642sq@rcI)~IM5{K=63S>wUnAA|7bpw!b`cAOO&sUj_mrQkXl z`hi+m94@mxls~-N*2GVaqpc55QJt!WK}J`G8%un_M{)TTO|z9kCNwGDNy$+b-lXKO zWuWC--1mgH#(UUoF`|rhmNy$;ua*rqT!E|R$OW;{4XbuD%l9yL3K(`}Ra-J5_ELx+ zq3*vwHIIp^qddYV#j}4N1j}{|C*4ilI%6Qo#DADX^!sy3pTo~@ORYvTn%Kbmd#ddj zY^GbUtq`4&oUsX))Dl-u`YVSfb>DW>Kgp^*ijMx(}sA<>rH!cM&7Y9DM+ug z1mMrT_h9ND@>-8}e_T*n_^i--6%G5$xrphibzm*hEG}_Y@(_n`-!v*0zfj0$@T+@1 zw~)qpJx;bkf-?3J;@H2b0DQpDxEh3sNpc`BNnTYjB?fk7tUsxIl4~m6;teLiva;eR zYH4=Zyrj+Z{<-A7HGhJ>UWa&}KSt6+>`ZqyS1R++jk0oH85?)<>7dnUg(3b{f^={j z@nEw!wuPF(`aWMGVNGz{RK2U+%@pk>7JN~wNeJ+r-ECb8PgxsY#949?GghSOi6|TS z4^?ksQHy~8E-fnZ*QmN(sIjuRw}MmVRz6qR37-nzWyqJtT)aaS7&f=6c>b|_)1Vz z{_l_|x3Q5fee{L=n4g2m4jG^J7e`l+n5m+gI^Sth=CLe=6mXWlFrhFVa;;EipSsoI zWOo)su(;yFq=I#0)IW7w~gij2Na99#@pL;p+(;f@#IfY`Mzgl#q>^~Q;4?G z8F_36y;Efm?VNo0VV8#V>aJ5r;3^B6Bh8GKrhI>njQ+{cmE z*D$CwJa0xv*tMD-)An{{0G{}|M*3+9kxL$M)@A#E^9KLmVGM@HZJ^s?ALI2{z++L*%TC)3|_~0+nm9^H~+CAS|?pAGM?UClopB+#P6{&}N?c zEa(MB7-evR*waiX1=u(`6$!p}0M~tx>VJ<{k@P*p=Uv9<@9M!gWF0^c1f>vW6dA3T z`ggv7*H`ZZJr(DC@kX9Lo<>CQ7IsT(FGBjPr}3 zKnVOS{V_dlJ3ib|vr)3rA67-bFW|yrP>>>7PeFeg4#6>b3~W`m$7v35KId$EF_QVO z1(&aVilo7)Fk~=h5%EkW+6R`O75I?KyZ%fKu9v=aelHp6dBvwS6{w-He&?5?oX#U2 zZXRYNnGMDw%=a&VZTpt~o#VtSaRa}MJI(a$`IBb}a|tDSCvG^8H!Wwvl|?V#I)}bZ zGahNP5SjHCn(|~m8&je?U{_3k{klM9tJ;`Nm>@eASf%V#iGhdtNgcXHUIOeN4#5%K zqrlWx%;E zqgn44?#id6UZ(fLf)yLH#p)<+nYp7ZV&_SKeF+bo)YTUaFAM5#iHN^oE`9K} zNf9xETH{JWh!tcKhxlFqFWn@EY(Dx`c&cS;z+Ht<@r1VJIdIUtupll|G-86E-_7BMT& zDyb>VrCX>GSL{ObildxzwUT}+a-W!7c>k>}{DumhVpZGM;KK1T?kuI$zL4=pAk5Lc zG=Vo0tO77Y585POju{(z{6JmD@ryOYx6!0abvA&~rl?fwC$L{s+s6oTVW!}l;qjO3 zb*Z)Y^t(rHB{O4LkH2A9@+vJLFYmF9yNCd7;z21iMB@hfCKq-C6e;>l6d%9ADAixThq(~k7d!v;99clesT__qY2(wI2=`> zQoh~p`2=qotr_)2Y2{xA(pWr%oC}!AyrYVC&K^0YaK8&g7{UqgFk&1C?7H=h4xJ*n zzoZbygH%?^C*5icb>{`5mz)vC>pX4Zvg#Nd%`eX0UrYG3`D)0A#`5U%qA7?_!Hf_yMXs<$Up5M$K!Xp zOIhpx$4N^??oLZ%FSDSUg;hB$WQxoFBJ^D<`LI6sxomlxO{fgfRp^}1+W~0pDn7#N zG!b%Yo5=p4fF8z@UrWx1k;V1L!bWJhyK1tHZw8Eq!(&QAvRck?PXnSVy+QsAXI=j= z(=;5rlbFS(1a@rQ7I&>P^2;K#a$gXo2k7gxpYYULwN687d(}3b%g>yf>?g4{v=lGl z8L_NG>F9Tb28flhl?B4WhHVtW7pTGNlPk5{s7qRg8N;eH9J@G`$G+2m21xK!^y>y8 zSVlIvHFp`6i147-BOj`#RLs?d2o1%j=m6#}SARttJ4wbhb@l0=wEOBLvI3C0>Ero| zo2VW+lx&&%ZePmc`*k$7rhu%4=d*S!Wmp6DT5~S~pL7hL4m!7=_tygv!3_gf7&6Wr z9eiJ#rIO|hLW8R~bYD{JEVy9kAUHn6Ou+3>Ca_pH|VBG;hdk?9sKM6zQeh;2yvR;C(mkw@&+xQp)@7-RFALrkMz* z--|CcYa`s;)ApO~LfM+6o|X)>K_8dJd(q52AB-lMjinS5zi}24EG)CIkmFFrJOvJ$ zPMjHPhMX}-`H)iT1%i7)C}kU&L+7)=cD{crj1v^-w%}1H)X78VnCVKi3s=v6tc-Qv z@3mJb*==meYx*S?-J|y0;HL;X`X%oGX{-WG@KcgRK&U6_`WIE>AE`UPZEop_XK7vtrZ?x z%G!nc;6B$Zq$f`WFq-%;b$ma>G8=4{pblK($UQ8VeBa?vV}5toQK6&(Yzx9s>5L>e z`}qD`gA2|$aO=LC$-|M|$0ocYW?JQaS7?I_SU<&^{M7@8rDF0+%J-Q^l7L>95QQ}I zPCo8uX9&A0XGYF}j|VnSWNCP<#R7d{2>xBg>aTr7gw-PHecK?C;LGTzt(2^^l~e|; znYk?%pDY#GU?6&SCbMwb#-y;RUN~n=kZjpJA`YdZJ?%Hu6m>f*hwfxezg)Cpo3c~F znCU(`d3)_kE1|TV791zE`fvWXQ<)?B<2hi*{uS*_TqXAh-ES%aL7_Kx(VD(dzm?@H zxmbx|_!fJRbJ^ZkntwL5H5q zEWp#}ig#zz=$DLt$nU@+&#;QAy%dUQ0d87iD)>rCeUb<#t;@hrc}Aj5nsfL}*ezGPH}$yNK1gumowIzI=~Lx{ zQE?;Ps|YpDuvba#u))1biHWA+pQ{V9t@ZexYrY=jqOoYlefs?e;!ereIdsfwJ&}Q~ zcwR45TSUQyr%7l=CPI`tDVH+8auEipUvk-E(%uX##Y{cawFoE^Mr)>z(cj8W z7v%jytbW^1Qd<_Y`B061;uG=1O0Bb;v*Vb(OR$seZ1rH7rMS^)WoL6Hua%1UrFB~= zfey=jSpwq=V5m$Q@<^D!kG*DEr-0^M*!eA<{f%pam*9$?>8?j+iHINDX8-qsK5qEr z9mwU{r)wHrhQW{m_EIvMhL=0??a$(V>7hl-ruyBv$`EajV|qjdR#s$gpBx>ShLq*x z&xZn>8L8`U=iikJtfR*@@1pb&l~N2tWld;NS-RR_mh)S45yVa4E*F}7-lYzRejY(KE4Jdq-FHz z6&Gcfc=mlZwr`lHEW>V`b*rv_&d#cE_gOF$e)e6cyVQ4q?{)+(SChQK6`S(UgTVW- z28}RT{JkLdb8uT@CqA}zK)ds;99?Sb^mx4Wux_7gg!>;T_sFPqgdOkwn)YvB_bd4y zn!E}qCuWscmf3!;G7VBKU<8o=pZ81e{nV*FID-N^_G^1_$RwFG!)3#!ogL^EVOas{ zi>gKZJDpYb^5B@vKnVT{8Rn>jN{y?Tfc|^PwaZ&O-y$dS31}v^bVK{Z{&`eRaWx zrK!Y|MX5_<8K*OeJ_>dE3_=Pn$Ubcpzc5cSr3Tv%q3jn&Hu`t$qtJCeCyqGpd2Otm@)e4a5qp$eg7(Mhv<4&*9CCz>SC6=9;iIsE6l}wlyk~7IMXq|&Hb}U*@}j(zh7Lya3^taY zwZ~MtwH5{G+)2LrZKY7E45eZ%mqKB3i}brq=;d)ujcuw#^-`rp6*xJT-EkDbFhM;9dyI zx#gsx2&1o+emvaztR(jYW3nRl=l<^RbbJDMoD&v(VE*gYz!tF;gkbZXv`nQ7rf(pl zZy_N%Vso%wK@P=)hR_pP_sOk~5MKqp@Ma=q`)vr8oAj6YXMEs-gHRb9GG zs^7_lCZC$kh2VL5bT$)4Ju2bqVmG+G2X z0aQYiuT9x06Uw^$&c2(T`SP%1Wv{kpWq1<|M~66+e=#=O04Urt3CgB58+|A9$(WcSg4i z=h6E4%8Xl5(L-?T2>8AZFH%wT3a;g#_W+E3Zc}Z74UU0mWwN6`vu}@BOpu4~ru^hq zu0^~|AsS>gcV8HWSrx%^(O$UNK-nY2;E5+8%}w&1Mb`Gc_r=mVP1y)u2=Oq9E}x-L zcq+m;^w!+ZQtrGDUr5$*y{9Yp5s5_Ue(r^;e=GYP`Eh%f%{*VNcc5%P+f@^n*2Elf z5OsAENYEdVMIqX6`yE~=SIxP!G(594pn;jPuyCWmp!)T(SqY)iD}?b+*bjt>N4`l=sAU+wohl^4++3yu9}P_BvvY*T!WaN{0@hk|%2 zj{9lMDp-lAfm6G>*{bsx+b%H{w&yQW~SGZt5op@ZJG_HE9j+sNmJ~NWeUwv6&cq@Q%meK*f8R6QPZ^+ zbPc*r9jYyqf0W>^FPU+=ng$|z$J*%!@*H%8=bYRt;OS^|AGH+B;)eNCJ|!EH$6NEF z3ZJ7RaDd6+G=&wXxPSMS*1?4YppM!GBepzP=A{LN9ZZ_?VO~i2h9iFBveu-{D*Dk~ z?7#Z0nz9B7uCPv{^A__+@ogcTMe0g16q%*%mEJ}+FOF(5PHXP-`8Zt1%oqIGt(8b? zQF#Xenecg>c>-dx7k(#0{UXfm!sEQ)4O1xk)Ut58>n1~ALLoN<)h;KRA7rhN5TO*m z!(;$WHWmqE42`d;21zw*+xQHZCDL+Yk!_uL=?`o$a^tKBz zbZS(D_G^&J(97z!i{Rzr_?hk**6A7RZwXmRx?M>2NNZrzy;XlQHMV03Ijb-IJw*&! zHQz{v2W}>8Uj;+_3hr>5NGeNkG1=)&nd*^zJ7up?yf?_Ik})8sMc4S$J! zZe?VH&76ZHvpLS`TW#9F9<#KF^vB|kLooqe9+&21hlqTMo=nQfbTR+B9-+DnNfo6r ze}4*&J`ZXUjeitk=0Mx$Fy6{`BzmR6N0MvZoP~NiDQ;kdoX$8Y`n|n*zl++frGpV| zHxkPAlgZKceD}0V{r-0uh$v>U>uWt14CXIn;#@_an3BSF*6mk00pO;RAfJ$A@Pi-6hp zzUp(X)ajBE)Z+496H^tVdbLl+OS5*Px8ZKq?=+)h{G71dt7rDGG_Xl-J&mUCc4-I$ zMu43Y~vq{b^8<7=os6RrL^HOvQk7;DE+}}vS?CnFR!I@eh6qHN8PFyp%uHV{$Z(i14SpKIVApL zJn=!}`z@E5LA%x3`q$(q3P>;ATC8IL=fcKW*)kuj2wK0X{jxN4@F4@b&rD&&k2Roq zjWa?-)#K-gZ>m-VReOSV@x&f$tuwsD?G}_(XSEb!V<#aH>xEl(dN;viRz=@C#kB4G z3~pfBA&XSRsPK8L<|q+xr4htPW7_YVAw(zVxt#`VVg!$uD;t-~dlJ^}J&6s)eW~(j zs`wEMJ{J^yi%qmVo^ykl`O`qHYD|ebZHL;ek31@`(^|!c#P9A9(ogg>!=V*I{tIPJ zZs1~@w6fuVvivaR#)TnPI@FSNrj8>(MKOifttv#@;s@8dpI?n73pDZX7qpquFAaGU z(9q+Qgvz44{HwFF$M8ZMJSI!f28+fyvWkJoKNgso zU?UVa3o#yqaTDkvwTQ$6mauMUBC~- z0q+&T34S=FYI8FFSzJeXnBBH#=rBLG&aGai>(_x_sWiI0!i4daCDPc*nBy)yX@g~N zAk4CBtD%F`Wt+!4M1})`xFlu6>;u15L4lFj*OcvI)`D9+tvieQ@1F#LmLaU+ zR%2O%y&SL4-eh#wDBWKzv^XKZ^^_Qx3xQYM{^*Fij+o)8bpmQ}Hb(fxzzBUBSP!P_ zFlu^?82SmaI-z+rL~4KB79_a=oLr@P&BNLYAil7z37T6={_K~Nh5!(lSj>!cIGa0 z$pdN6c_Ax2*B^3fq+9OI}? z4VA^K!#lX8!T={v>;6}fWs#ClR-}dbkql0v^W#^Sr0+6(_)MA|_Dgf~VADh;8m-04 zM^4b#uMjNR-S4#-fe4neqCRr+>WF5|DJ(LPZ`%!5C-|FDEyM7;Pi+X$o-Fu~;5y*1 zd<={8Nm0X=sN7_#eW(p+qcgMaDk-p{-o~;8mERVuWChg>2ksHsyy}Z(isZTi2_xxW zxYJ~E?12L@$6#mGzZ*U7j*|nQH{+eKMb+jXdBGajXl{@-r`K7&$;Nwc$|L3L-O|Vr z;Wd@q=LJ_<_)V87P+RETPTGyQ!^*%YcyzLBk-k4`6fAkQ>UY4*U$-57iWdUM2swPh z^3OaJimEhS!9mqVCpkQ79`CNd_W3=Nqxf?eIrH4Tciq|~!V#~b#sD-5AFDY@SSOk* zVgc*NeQED$_}==~p`8OY$kYhh>rI?m+K}jWk(&8Tnv8W7lbB>t+m64dQH!gXIip2b zqJ-`js$H?r0`_Czq^6PqY3uj(`3PQ8;v~Fq(_Cgb&2N;IM6;(GvwW6T7;#Mui?1Th zl!RqXqk|jQ+R3ZcFzZ5}7v%{{zT(Hgb)^|&!OS~V;HpD0Erl8h8m3Kl>H;I!=1ga^ zr-C1#Lg|xX4mUxKAUa&>ZY~a_dC5KD87Z5)LO6C$EM8bsM%Q9wqM-CNtgZyQ0Nohk z5wn*x!VE;j`JZu&z_nZkjEfO5x@v6Hse;A?$$Vf|g%kBgcZS1WBJmy9DsO zFWa!gp$Lz9_jU@Y4$YqLnmWKMz)SY4qB|x#N*vLgyw*S>}+G2=v zYr<>Sho;0Hy%DqtVCkM%((5=F$rCMr&>t{AFfdC|b%Pl}N{cWTeb+dBfXNDO9r8o| z6S7$dQ{I2L1a9ENGkEWZ(dK9G3g`#BrMEfC7teOx^`gj=#VtAA+<2IMm&uao=S9p< zL*8@2fN&yDRG5BBg-v&Gei1;Yqp~a)g9J6{3VdB{LK2*|mp3bo!@QVbW#4`d7AM1J zimVx?XKc0UP0W5A#bAvAGzJ!;I3cD%|DwhvmQd~*PMmp)Rm`a)7@%_V(EXIIePX$d zJ=M9fbn)0nk{3AcXnIdyF^b+Ipf+gwuo&9bMY+{q5f3JES5b5xLrL&Q$sw?zNih|R z{w2tle1M_{`-QeSN1;eZAjFpBmpXf`+|{263@5QPKW>5IPON6on0#n6onB~s2edKO zP_LdElAX=X6%Pr#KI9{=8iVM4Q*NO$z@ml27qa7~T3i2CEqm1>&X5e<$pv7l+bOLf zW*({V3|y>K{MXa#U2knyXOU#uS>s@-mGp&0GoA%_TtC_?1vq)S!7gAR_+)Totn5 zJ+~Y@c@mNIG88Q+&)WJx?~)t~HG;tfiTaDAQ#ay6{CzqL2TElqx82(nTbt8Y#kJ3v zJVttthGXD8jMui&GNo3OI=ekE5d_-D<_HO5Zy>F(?|S8Q zfLaIF;+I9&OfhlU*0pC{X`PjG+bPwlr7EAcH6vm`Fr{iDR>;I*sWMBy>M$0#uv@xCrD>G4i9DXjZgdIwo}XEMt@1Q?FS zeyoqmwVxTji+##8^qb?w*y{TI-5Vz?vf_p;1IMw)_IX>Q(m37N(){`Hg=ziC1 zU5!`O;C1@+ltFCIy?M6Mr)F#-<@{y2D@e5z3gX@B_ ztXno$I{-6~G%0U0ka@^?Bkq0(Gw@@fI9se>E*?VJG#;O zf<;Q&AzEWkk@m(Sr~;S507AFp7J?j8`G?V#upaY7;ZSwQf4*tUn7AdH_TbEd5oy?=QH2u`Q zjJb-ZOhl!onx#XSL{YwhmR`O{M}U>uw97R*&U1=VjFS27wO{gmeT=Kw9t}!q8GaAs zq?4<@ zOpGG*o@(f0D=5CkC}1cnd1|NzytxczFcpNg04xUORAGvkaqKIP4!>dBh5j;(;6se@ z=?-2+ARb@LinlhT^y{u#spNGpvw^!CNDEJdtDo0#J+CSJFTeKrYVVSQ^){FwTr{(i8xk z-}lJ_+et&7)ax-7tXsv|YvF~Z^nrCJUQ<-b7{Y?gQ;2Kos;?${LRzY>l<7=qc8Sjo zqS6r)|HDs+V_9Nx^;lo)V?HVO+2b8 zC<_lq(2>`Ht+Z%z{-aax)26Q6HKtL{i>B-I^yOv(07ac@-9yIa$ap$4#2M%xF zThvfVfM`osY2y%^vf?fNgI6)%s|T?fA$M1Td+)frepX%e4$;f-SHLfRiv3Z*zHrl5 zFSd!fvsZY?q`IQUR`MeyTn^ia-oZn3HZb9fLS_&NA6P{DVg7sAbTN8(0jie!&lCcD zdKbTT8Y1O-?!q6w;ya(cNz44O@O3u>SV$qbXuUZHg9R!iRru*r(d&{S>#?vSN!|it zOET<_khM~Ml6brAL# ztj*fsvW1i4tWcYt^1Z9#Jjzb0?W&v`)#hR4dhZP&;vF0Pwk|R#mHSgU^Ap^IV*%U9t0aW5(V{dD(2S2Na1L zBHMP=Z?+@fhWtC*4E>@@&#-M*eg%2=0`E_pBlxpD^(tIMgAzCEH&EQI&_KL*afV}M zR@}?HJ3scLuLHNKcKi-4tZRY^^DOZVDu>Nbm`iYiw`^z}f!G{m3y)*PC{^9F-lkx! zI0=+}S44}>7d2-pS!Aju=+M%y9%`EYtGHJb@fl^0PH0+n3W=>zU3q?}vs7POwk-4y zUEAIa;1Rxt90jtx*^(|kEl$?QC4XIowt1l9LO8wet5XhUsEDS#JCUHjb&`I^wWX>e z8lx+OBZ$*67Q{&}u7}u1JT@QNxF=0UwSs4YP9rt&r z&Y~Ip91%fqUX;v^6VTqZe_(qkwv^pCc*r4y75REgpDyMWkTu>RD*DLS;MUzL#Ie(d zjg`w&bI&t${!;H?zn~>u8YCW588l>+Kk2Q3&wX`0Bp{6)2Gq$+j7WYK?Tyw7vq=PcoR)F#-n)adcK-9i(w#$qe2~Ju&af$BE#i*(;}7dteN5Qv zfcm0Qs85|=yOXq#erdPtD^Wdwji4W9^o4rn2k6UoYcVb}fU7$RUh6-&Ula05c zVjLKOj7dE*a-0M_h89MpW=o-hkRr*QZdVYK*0X(k>g1F5)0D_DPMrwpGwU^61jp4vBg@bhsEoSVT8o)7&8Ye+HSt}<+_$us^ub+e4J8kLU!x#$zSOH zRH$Y*wHR5w%`q-Ol$y(jbklUptK)l>b4aOmN3Y%a8>(~e72Weh(8cTV=Vko~A}Ca7 z!wSRggz9xW=KrR58&VD1Ho;`Yjjs*bjSiF%0Uer;Y&d^M57xm_#k=-v%!zGB7 zWv8*&5F@e`mg!okZ^2E+3z7Z$W}n19f4joX5{em&D`WOBc2kI%$mckxa?x}_w%h&1 z`~C;2XJinT`dYr>x>hJ5f7FTJ^uY!xaRQDZJ8GOl7lTCVw9k{i$-Smw7 zr{URNUsOHGOmE}>|RYW&3#afmida@N)3>X^Go^8w>^Dr z1u9&y0LjFXH`n2vOAlA&j`+77anz>r!0*9YSfAgmZ6J4T2TtPxj50qhs}df|9xI_U zolY2L^ffT0z@>bjihVow!xS9z%3z1n@+v63iw%xmWJj=e(QPwMWP$*;fu=XtV1hBFq<`c#@xIZ>KAB_iZzbeLS`R)fhoSQJ^z%(5ljO{l0a7B)M;e6(qu>!N!yreAYkf{Mz= z1R4Nue4>>m>XpHK3Ci6n#4;!~BY-K1ge(k1gJ^~zHG77tg#_uuIi1CueQi!Tz-9QN zYQzmyB#uonh0vlA8kLKILRBix9FSXFBwrdumv!7UHMbYbp6tHZn)ayFf$>v4Vy^D4 zfi}3c<+{_b~X>P4L5JiXqf>-))$9KBNh9nAiE3mGe;g(qowh!v4`qL zm`4|{_v2K{o0%t`xgn-i9(6$Ry3Fs?jR0m$<6`5BBh9jo@Vj^4LmE&bmsGuX=h0#n)(zIU81e#JR(-wfa zknD4@TmzU4bCKX{X<(oH<%NQQ)pi3)lLVR;Ekn}|izBFD@TEs#zNqpG-{+!7$NhkP z$eVSq?50`Tq8Me*zB? z2x*^ne~hX!AA~|f{(yTXNV=tc_B+CZjB_yvO5iLCo+sapP$a}8!h|@*DlV`6#o|0P znsXand_X#WVJbnQ8c?0Bq!~UT+~JZ+I(#oX!J;;fyGhgafD?z%U8H9`U3NtF!VBw~ z=|7O)PJn&(83r((_$OBwy1M%T!YwJxm1Pou+bHW=8+X-d|s{(R$hs@-$+qyj~B z#QQ9I3^eV}(5HFOna6H0?XZj9=LpNWW9NexDdZ9=Z8-55j=F`|{jp2{m1jTFDayWW z%!X8#g@@9kpIUxQh_OUHi|TIupF-=K^{``XW|1dK}*g$AA?^ z9}ug}>|_aV{~(Io*?rO)AVn@ksouG%J%#V^%jJ(zq`Yh35(LdCowJ zI!4rkk_mrnd9#rguQ%&3J4>!?|FRYxpwSzaTq1t)c4eoJsQGa<72OVl;#{p*k7Z!c zR2wSV=xDY_N#2=jKV4XyT|>Py176^|hiq*$bUF(sbN_rlq=V19aOBKNNePjcu#~4h znlA4Ef|&N5_wqXIDz)dk`DmB2a6#J{mnfA{kWq|PP+37UQ~@NRN*0zbx*g`7d(|I@ zuofG=raHnGx~=`626*(9z5bsgo#m*$dT?Le$d>5wJ96*L^lC;i`!@0>q*66R;?Yr^ zswvn|D0G4#i7A)_5EWB5$OYp)_}gq)7gGf9&Qi!jt@S)8e4}5RU5$ zL22tE9X(c>r4WWalrQV$PeSHJeHjRWu4f0bciI}5;;~Lj-#Dn7h35FXB}E| zE$u#QWz%_(AXo>pb7qK;LhD}L&~$>eqkaU#iVb9OJVEn=UP6WW;}C$AO`KF7WKy;x#Q_%T#$rGe7%n8Sb34f)JcE8oZF2ZM1M`>b?l zxTZcr!`%&A+-`E~#fFI#Yq^?z6<<$o>N`HARK!<%4A-=y|Mq1W&5QgSgGUT7zgJ^h zZP?`J(TweT2oszezufryKhEttC{6-Nn-lMm*T^vbl3&iP{bwVgOD|^c@hJugl(f zUmuN4=e{9lB&4C-U2%w)Vxe-D^_%jK{Fo-+;hra= zgU)#o&9h8or7X7h3@g~*`vdw@+FfzAbQJlWz=gmR zODBkpQ-9BjwKL1&N=Z2FsE)HlT8KUPqP_d9QT)6zNo3<9LGZL8UUH~PfIx6=t-#($zD7HZhitK4^-R2q{SWxMv)%j{87hv z8Z#CA0xKdyL67@vX#M(lJcBp1C2eFNYg}B7TpPS&un6(rV7W2=2HHIS-aZ%W!HGfTeP571^AJkLqcuTyQAH$BI z;>l$tU3l>L<5o$U3zS#f6~FTx^&BND{j2uNyr9?nmaOld8TsI@STxhtsV$+D-j1eK zbxtop0Kv?T!oHT~n02`PYltQdD85j-$BB}4VZi)h%pFXC7e(H}K1lN^|GH%Qgq}Kl zKo2wESDEaq543#@0XF*L5-zbcT-nugd(3?!2=6)**wbw`Zd-_@URTnrLtudkwtJri z3qdzN=%jncXQbdhrSI&9xktYFYZ(&u-K%Rdbl8QyPYmBmcN1w&R-)LuWuJh0=M<{c zRxN)N24?nN4gpuMQIp@pHySRyoH`FjJ>iv))c82O9%}-T2Ma@dIe9qgz&eCbto}!= zOvh%-`T%ZGyMq*{@}LV?7I;bpPJy#ZXV0MUn~=2R#7B!WyKF_C{rsxc>PMF7YY6x{ z;k$*s%vGO}oet?uW5Bogvjobo=!!H2`y#XH9(eheySW}F+!k^*0aLB9V7)loC$H;r z=@a$qemBpd&$+rpXije>e&(18L?bLl(+dz(8LmS$uyg4@_BIFI(!WRD)I4#Em*aqB zvTS;t%p1^!zgLzg$Hik|+^;9F2e3M!YrV6AY0go=Qz1lH96Q%0n^Wq~;+BlNe6JR^}EYA|%YzX=8h2-yg8) zB-=`1a=!EY@LiLizs9T%3lIXUYj^$ALH2r|SW?NKV#*N%G&}Qzq%7(P4dmjsQ_`s_ z!GS}8lHyr+UvrJ_Rta4UUbQCs{{k8Y<@vLtMBMLKMpEtd7GedLO#-Oa$zN9#vWRU8 zI3^aC>-nOEcUG-jP<&032HlHW=S?Vq&zE6!_k^WrC*5*ixqL@l=*H}@T9+a*BJp1y9_s@3;AQhfol1Yh z5D<;YUy5Q&$zek&&|yb$rHwDs15WfKh5(b;ANmpR$YD?pSyL=V2^ECt&v#kxJ}%w5C=3P3D=f$V@e*_yv@KAKb6G8Bbw%fS)eskvP@4{nR@MjslY>I>AWdk2)Qc zo-TlIR_ok}fFaGJl92Uk1k(k9Q7BU=iM25o3`-E~(8+ed)1h*`>-h`2ed;Rt zL;b9m=zEoY{0T5Acw`hLfln*X2tEs2%zOWB2=*M9^RUUc?@v^4jxd(!KR;D8v=wn_E|960`M@go|!14ug z#dqxk)-Wi=k&}24kY$yK=@$-h97j=1ixHSB1I76#Pp zjDZri9o5lMWlYTlN~nNHrLAsv{YCQqN=IqmO0@Twqm@mm^?q z`0O<&MfRl5q@MKm2Mu83?*oK#Fe_N?xK(WgldTQ0$v#2bd;I{k++5=|T2QUay+jmE ztTh!c9_N5^Q$L9S%sLopzDO`wrYKRfs&Iut#Ag>cq2SCG%s>`ETM8*DHz0y>p{B=_ z(!bxpDa!r`tqa(sDwx5N>3M}NuEEwJNh;>lK_rvl&9Nt*&-eK1BY)FXtf2B9^~>MQ z??r!v9tv={v@MJe+q{F&26}*LplIi>6Q@3*bH(_WeBUN=-(q&C1>gG#Hg_squIY^3 zQ*bU!pf%{&wr$(aj=5vowr$&XvSZt}?Kif~|2u!xshX;pi|LE5>ifQ2>sdkSkY;Y5 zqg-!HBF0p6s91<|=#+B%;T=_d<8&59>lnq`tLE*94N$f@xa0jv7w91A;$$|4qsOV@zHv)2qf($GTXEZZKjuZ-@Q<$JEcMJP%9{-V$#Mz> zw+|Dc!5q(&q;WHKzQ{{ z=RHrVN3&OD?i`B_FTt9|bUHX<9jaQaf77g5$B7$Sl=}y;X--pOD1<_w?$mdmlf78nl(ziTD}`ZO9sYLLzXQ|Y~Hq#!*mW7pX& z9_)nECZV=ySX%-!gNkLPicx=(kWGCR;;I8)>Ua`~*KcL!rUR}VUn8Mo3@n+5gsl>F@teb~Yq2T)N!XVvLulDNXL=b^0@4W=?yp2g>mps_BCL zth4hlF1kv4{YSvL$wJZbtyw#yO$C!4@La~=$&*y_o<2{u*e{eNLT|LHvcFU*I9m6hdx$sQIW7A_`E z*8hF}KQSK`ZZ%_`o)@baZ%7DJchWLdNE_ya^!ovNl;;-G52!p z{$bDpC>M^^yL!k*5Q~^!5E?i6*%*pZ`CB`p@j)1{5?=-6wx(By=Ei!LALakCd=yFr zH~z=+F;=$b<)pNx4PYLJBN{6i2UBcx_U6)veIzOy56}5k6b^+(^#|3N?=oe)a+~kF!yovlyANUJ6S2qVJ|EIC%7i6a5R}2XX$=%WAYf1Eek74ZB_R;>r$nBPPsqOv1 z_UdJi*9^|d@!>0+k6QX$V zb3YIZ{d#u=n3=na_ZR)UJg*`IgfJjH6t;hI3fRDtP)RBwvpu)wn%9~#MDRgMqv!I- z=Jagf%l~8($K({^M2(2`QE)Q5W1ZDHs2Wcvpb_O4_1jM5ZOaVU1>!%-2LY&{ z>mT`JzMH??6d~&+;bRxvuKv~a74XR@hrEzIt@HjTC?Gnyffi~1>I~%R=Rsl2p9DGi)(aa9#F~UshKCP^lo<>*;DC3?L49hJc3R;s!{N zsZXN+v3z>#d*_<5=fCTufSFhWuX6U3fnFG9cY&OFB|-$dCPttQFusjHFb^P@qJLvO zfTmG?quVusrj`7QU-{nqkTC-^0rpRNV)y@Zo%`~dvT?O)xyUQN1w9r~79ez(4RRbPZ8XuJB-=YHhZ21aha z5!l`#p;z561=GeqX4lVeX>Q4KY+`chev$9=*5B>ZZgXk^Ui6IJ~@C9hVK=`sUTQFR%Ijzvg#@@KnFOk-v%b6lSMCp^1mzP%Mr2jlde* zyQj~(9`d)plT{)3JLpnnQPY$t_mve#P5-X89)SydBl zjS0mFMbM-c0M+rbS^7(8m|uR){ssIgQggoIAXv0J(Q1_a~*-vr#SVrjHwy&3t~ zG^kUlN2)rM1^EnN@a{weB_3aeCYhcLIz8u-B~7Kzgm=&iHW@q*>Uw!NQ?1$1!?dqTN@3KkjH!&K_=( zLkOszyc`}of&dc4gAwZQrUO}!&Rh7+X%}GIJ0;qv-OY-pI0RPkeH?7IdLz6 zhqLXI)zHIm;pY>gAl{xTajv*tPrxR`&@h^_<4vph0GjGTAECcZ36Wjc#ys-Qx@fP5RsB;33h^~MEGKnZQ@hfX68wiyg$@nKU&Lt& zWj?p&1xk5|T(ZjuJMlTWWQQnYWN|r(8jBrw3yV%oNV^Tm~3pDPRDSCBnt~-2khAbM!F?>h6Z4g)rOCE3=BL zw%zl*NhdB^rXO2@5iF5o!KZZFGml8{1}8PfzB5qrmX-9s1o^x(vT8n5C};k zg@dqgsE`32u#f(ZvvoFg*pJgqH{f4$%7Fld5 z`2okp6yU0hwPbOVhBy7^I0FzPM1^@Ml)*;z|5-X2uvi46not3f*fmg$scG z$(2TZrIhSxq7xYEmJ28ANQLjUYH5+Egn&S=#cpUX#;V#!NnyKTi-j1nZA2?u8*~{`Zuh5#c9W9QYE~b$T1< z_l`!}n)r|XQGl;mAc~SD<6xe0-RI00M zW$S)G=XG7;9v^3CVS+oMG^0NPB*1JaROAuAvrYH1Tx!DnXkG+zm5djOeV+%fF?|%l z@jDAhjOJfZp&w5xj$pkxXdrdemSVhQ?oRQE$t4GI7i4r>YNo#E*ra~(xEd5s)vuA~ ziGlBMwQ2rTV*#P+a9=LuQf8RF!Jh=~LZX6y^%wB!n!ay;E@!&r?Fs#Uq(+Y6wlW=t zGH18AW$V_wsj{gvxt$vct3fW~^pr+6NFGXk2{eZ~{i_g7&BUtQ2E<(HJj|>&&|g5_ zxWIM|DT0^NosrI0+Jo8FJ}ua1Uk#;eiZ-@&)Qqqj{lH~r^6pk&Vcu=0f-h=Qtzq){ zxn}R0RQ2yTDn&&JYQ~I738S}+4Pw!wRXShtUCer!` zE00nGscmGp#D!y=2inJPI2>9p1e);W|Ye z1kI$iK1%nLtSa=QA}k3l4rMQ4a}|=gh3b&2 z7IV{JLyPVONyj-&Kj7O4$X6l;X;n17MgQHDw@A=k95Uvbzh9ghp?WZf*lcL%(jzy* z^wf~;>ZM*Smtu3UkRK_0tRsreW>==fnSWG)D20n>+yXmU=yz+Cjy=6oCS~)praIgV zRecODR0bcx2XmQxmd(uhjjL+#3wvMAXuFC2`yJ)2NIeXZ1Bx}}!jgVzbmsTM=31k| z2D)+-H0{0PQIQIj=a-d+ltKZY&V~9N`;5Lx`y9V<+z}ctsv!p6n zC@zEXOY)b@n!$OLMlVhaDANI_(wsX*@n>jfgV;B5B`QA1?i&8W41Z$YJLL@xG|eY$*B+XYeMG zrvXWw_;vLQDzQJve?X`)C{>myHe zF)l4sm-h?A13%<{kR@JiB#rCqmdCAli_^4$lLuabVMKdPqgSWqwyzuS~avjD0|2lpX1|9l!if8|MqAt|_V$EyZU@nj*m zmAi?x#3xFRjxnM#nr!=Xq`lQVML6JcjnKK5boBI$PotQ`i)y(`$AoY+~L( z2nGBeHx}O%vdp*<8Q(X$$G&|R;@fjoYa^_u?blec>r=Cl`i;9`3vYe^3roxJXxP3t z6gq;>W^&jwK$J&*Pku2VJP%Lc&4m38X_fN%QS>;WHR|!`Uo#jdLAH%qI6;>>#~PF{ z`RHZh(~jCG55AL+2)-svt{Ro8$}54LmXw`^;eQS*-^M<=hNTV7h^h7cy6FJ);HAPq zQ-Mm}r)F{XjnyGr6hleWKdY2pu z$i!+dA8e(LKqf3HDh-*}CU)=Q_b2U3QNm)3>0u4o1iWxLh>O%m#D zVck?M#_(ZZE%2GQ(WsGJ9=cp&wrO(uSq)F%@Jt|M&<+LC+)v=)c=nnaNg@>E@bdjD znHr?Czcew~=7@fI%Z{i;_U*)M9@G|4OMIyF5Uy``1v8bLq5RiqN-l_kOkTBYXk5y$ zC^>|vK$EyU?PE_weVVWjuJ_hK|cq zCx+>H<6X4O6Y=BAD-|MI+L+1Rhssdlbcx~l4yp`Lwm4x(p4vX`X#E*`AWn)JKo{0G zAPKr=gdH2Zng1v&7RmY7J^zYRZNRZqmQscwNTpFNQ-?qXKgUAO!oq6%8G7&^mz3Wg0>j;)$_u&lN?~HF1J6PweaKi#Up`%|Qe{&HIucbZ$3r%J?b7tJW?}Z6St(Yw z;?x%W!D{K!NJyKh6!Foa}Po#zC! z{mPeD5mU#dkO47v_tc93Nmz^}Qfn3x5-j}@n#R&r6va72azyvvxC3=#y8LFg=k0ac zpAS-aU2*22S<+#iC+d(CS_93y@JN3S?QCoKg(I<_CXWrun+Wkuj%hSOMy6-Wx9azw zG+@|oUCz$X3~CMA{D2z7dRp}CWaPX~7C5r({9%8+>1Nwd2*Nln3-A~BHJ8=@_uD+Aon@G{f#|3RVbp zC^HT^H{CWXY!PqQe~}yiFv7h1PqsAl;Q@Hp&-|$rdV9U9ULy9-%cK(pGIiFP3q=@R z>ZmGbH)Ec8j6~G&iBP3HZg|pr|1@S5vx;tsC5h8(Wn5s0^o4V!lUqx&jcSqZDw;jX7(d4S-8>=9*<{5-n^w z?o#%8$8Y`vq9xYnxrH-QVYXxEeW(V$V&4Wfvrq>&a%94|Dk73;cR^XBzcSc@Iplp^ zm}*R7nAbqX#gUYG5vf8ev}01M{EK_KPZUCteUBtEnn8D{H$eF_z-oBRI*s>z_6*Lb zLf55^yuW5XUe&4R?yqTz`9Ycxg25{@2)+j@jJ7*3D(F?+cQx74K0=(rV}4k=AV?BF zn-=&OCY!IIZ4L|#KbVaoq&X~Ewkn1XLnlzXGb@ZYMW@QR(Z4+SKx$C)x#JSJ{FU<5 zNzlF;|JGy<|FaDNos$}jvBG=^rfU$TpeG-fBB>lNfpulSM(;>LHXDYXy#Mx5{s}wvVX(XAiPst{8t2depq%%u z^*P@tNNph8Q5ye=^(Crk+JzkZNQ>HWIOS$ws30Rn=NhNgAEfE_MfsBfTz=Uz!Pv?7 zJQ2DFhoUBt#mqob9|5I%L)vlbXhC6UDP1zJX72pg(x)|_UsYi%wlqW?zkop*f2VDt z9nWz^Ajr6m86eShD+e9iXR}iLGN;XkvN4!BiU{)}L22@pJ@}ICrb#JCeYctI$bJPH z8eCZ|%mbG%ke~L32AI-Pq1mo&nkIkvO9 za-xwtWB0>+i)DdgXA}~j*dm4wd zwokOFfjdShhGZe10DfTOP?tfcQ-C5dndgK6fz-Z=jy?pwg5Q$x7qG!_5qDjQ^QPx5 zdg`6kWdUsH*oxxJi2VB2h42%+n6^T`RG`Jw&{3p92F=b#n@30LjNkNrlEy!8FU$g} zJw`h1GE%VCx%avPk6ctoZ5}&X7+U>cWYFEtNkOC#xbI`p-Pw-T$fTDs6|J3{*NeA{ zcqaX%7AXGfQHDCGLT*%CSBH4J-@9W{hLMRdoBF#s7qommb^wv0TU0{(F|x4uMT_hr zl1seyPVvwOydX&!WtI%boxB#?uX(vjW7_@9Is7P)??vBhJI4&YHYSq_U;u)e-W!wh zi(>TrLOvV-MXywS@p^r6XMqdPI5NDy@X&2Nw*rdsFRN{#e*1TjSZaN0E~)TscC7aB zp(TtQYcs;Lvrhu9Ev6|=3&~C=V}^0&8JfUi;A=h#va^NDUX(7;dDGP*29(^Erxys1 zBTpz0l9udf7g}=u6XVW?NzFJNQ;k$iR9j!HNh{ZNxA`B26$}_HXFQr56{;$T{U$Mxzpt?N+1Cn3>fMny0pQYQ=8-+BvdjraiuA(bA&2m;fO81Ku7y)QBJr1Av+#o5{g z1G+iS!R4q;hmJwF?VMR++G$l$#_%Ndw&;(d@r@$e2S|-F_s<5!?GlE;UtdVjp0*$Y%}Mi{b0egP4a? znvDFOGSwv=?)fv+Ew{m{9f}GL(|-Q>3AjPVNq;VwA`>2`;NH0?HTUF#m=?&|90p z;eK(FE(sGKjh|_)q7+3f<(HZGIJ+0(Vv2dfP%8KCBg>v5B*#3Ikz;4Wz3$@>ULG^| zDT|Mr{_7CjtyBb|5|pyNO6bPZB=GnlYb7}EC42KuqVnfbvgYGWmy(8$FL2;hmx~Lt zD}Kbfb~HfN4vUew=#}J$EIpm>fOGNx3D3>@k(|G5BhQRm<&)AEVGa>VB0m|p$7Jqb=QrJL7OEXzx;eMXPzkmm%hZH{AuW{QStAt~#UbYEQ2c1xFJFz7Wn1+r+O zjIH~;6ISkBnzTb&yJS3$hU_>RVcL~GW&#%3JfrL zrC{bV!PrtnDXSaNLA*X;hCla?j2mkq^Lxi=J({Chz{DfuoQTW=^Lj;`WhB%@gzRWK z_~gp%34Z|ndYl(0qB~En=lvnqVJ|o)mc^@-FMXDv zLffAJ5<1a|b~@|@!QTaEI(-wBq|^lOu>!(oT_8$h1AF3Hlsus&>7M}%1z%*mF^J$= zgS0Kg1~7lzhnljeWw+5nWGUjjLZ}dQSs(0hv2%fODFs!`w8bgG-@3 z@%-iKSOg!{*3B7DK?2aKjf1FSd1)O=s9!j;h9Id#{Hu$$Se}S@(Oqk{&fst*gOSyh z8qMP0n35=J13}ogvcCJdM>PaHL$SfNjmLLx;Giv8Kv2!aHCKgePSHzCjv-g9MVM46 z%I`~zQ%xmQ{6zEcI@iD=*r9E*)=JMYMpV*3&jg)Fr=Jq%w0w? zZ1vC|IddhJXC19N*>Voi?x`diB}xZQygiGnPmr3XYg{KliSlQeEM`w-p%$_trfq&_ z{b>(6!61R7UT3C$btw?8SnV9*)jle4F_Xm+CilmQh|1WYHYQ(Td0Ma#fULCCn}dkj zZTkY@;_>JmOnjsBP8xZh99#akQOR=WJM50IPD8N3X-3e1s%{c&!I9s~^iWPMg>UHS z$267bZc;HxeB5$IMQV#Ep?=thS*iwpt8Fi>UH_(}%?ttC1elaL~I z8ZF`iOCRI0k8T$~aJFPeM%GXmlF-F5F}$1{2kxo?cssmf=U^_a%Tj_-Vpe6tP~@%4 z`g%FGtos(Kf?k~L&QC;7VQu>CxJTD1_q?LinC(104ZeG+fVmaIvDM>3LNEh^KL?MT z*W(bS@1vA*H%m*xcNC4V!{Tc&j>f0tGR7!lE zbbvk2VoOLdZANgn^qgxU4H*Zv=k3lnWrR#4^lWxRRH(D&+ zeN5bS^2G7no{UVXy!m_vl=XOu>x`Qa0fwc(Zaa>rZwWWCoSv`@PQeab`U2!w) zZwb|w-BeiRAd`}+DpVJGCX6qci>1Z-i!Dm1z-le~MMuuUA_ODqa5+38Ddqy9!m{=b z^;X3UeDtA^RjYdv*DxN`&rMQ#y0&F7tb>SHwRGzag*~ugcEvW`$a@*yuM0lhJ6Rdw z?Kt}?)7|h&A%GSeA{id16U#SK>v6;w7h^as>?gUCVsR#7&0iGj(C7Y;XJrR8K~@;k z=3bn4lR3++3w5DvZ$?5cF+nIeaQA+yBA8uHT$1xH13OZ~V+_h%mjo0W!3TucG@QL- zG!be0Nc=1F7hm6fkwetW6X-_g3KURN6||?$zfTr|sv?W=PPM^fwlyEni&_**zbh^@ zF)!(KmW{l_egqo;5UhUFhU8$1CBn%>1QED_%a{lQr#psANdbn|2DVc~wd z_Q^#O@Om|!@NP!eNZXJx|BGCg$5(}$^z`*;8PIJNl}<8P$~tjqB)nuL{nBJ`&kc#x z@`n9+*Gi=ImX(leZh`(MUkkML!&}+No8G>cd4#rt%Ua9mXJQ~j2kWgTxu@I)dg&n$xAN4KD{YMaDX8YSqD=@v+)wzGsII;bU zYK;ntu3)F^^Rc(RN|Sk&Zjo7s&*Q&u{(I3B==ZshW&<_Ai09g%Zf%XQ=!?RQX+5nt zyZ{D;-`P(H6I=8%6HF4Md;UOpwkhMinX!%>A;aF4Q4%;=pAAV=V z{~$YD^L1g)N7-k}fO3;obSy3l2v7D6i-Ah4(V^ws5oDE3PM`dH$rO8aJMi7Cza*}R z*%x~SIPg#Q6;gCnciT~?LDQvu#66ojv3@-5Ol zt#K>XC`mn&ku;yY6VBW=cAkF%+Ig|DF%VW*3lg&(ZWm3fsX>i8)Khe`?A+LUsbWe%3S4^ zE?Sp#un7{&v3s*2WhZ1vR zMYHjGT-Mhs?l<9yG3!LKL~)rObBXlQkYP=yLgIbK&4H&~ny<p5Zir*f|-wSQ?0&qhQ!mA5@0<#^){36yp-XWpF8Uo=ZlKYdv(W zSw}j;4p7gMz=o$E+}rkK!lb?l(0||sfpewGx-MqgT_?bhN9a#guc#|%2KWxd(vjQW zxLURn)-0$nq*GH`S9wQ@&%ZLqmkZXD8UF})Y0H6;(!wMm0ouMsI)?orcSpU4&U2i4 zavJ_(6ZJ$**hb2)AQQA!OD+9&W>zzDGP4|51|0pjP=n*XH~ioJTs(@U z^iKRu5T2P>_RX885mLIj@>t}S*r#QvkeCnV-friabal7J^?j>L3bk13f^JY$tqmmj zMdQjay^X!WZ61shEa`A+KXnG_+Ww)?!wG^_h3f`$*WxAtHACUV4>ys8!Ox#=g?#{( z(6cD)C#71~pP21x(Lh=EOocAR1ssa5f)%+DD{6{k6MO>CD@6$F6vT{v%!lExPFy}{ z-N%NfBOtHj(mBC+OH37eBNHl_bbNv-e;jajV&U}|0JfTn-)+nIwhqw^Hypi|4KgFR zX7-IpYd1|`yvaaAXE5kvd!2ZiR}EqPKX8GuKT_LzN&>5*mcH9LLs8OOL5YVPO$nm( zELJ%+B6b#T{$8sx>&fA%+%vi+zM!{UIS6c z?Ghv`$HbLF?o3(lm9=$hMVm!zT1nlvUmDP8Q$ISwm{$<`rs5{CM#4T z8veQ7eA9?krW!j+Duyh`n>= z@xs<&z}R9|N}TDPrEeTzC?n2!a`?I%v9P1WDfvDL!KRmKw|Ss7AI2oXA@26;x}fZflsBDuYvH)+Zec zDB6Hc&tQ5G-e77ztCkJyx{UQ{xAftYHr1cOrUClyjvaiH`CBdIMZyG|9{3wdEHWO? zzx)-55lD-|sUd9f@49bkgu36l3Aw$>AZmeo_&gztGb=tb@DCN?67{eYtd40j0#i!` zQ@6?i-_|$eY#U3UaW6ADd>L)8juoR*&g9)`yYN5GIk3E4#DIR_Qp>18g{?}*ARyyt zm(g$I#W_5YY}Id9T@wqTH)x11XB=F1+&eHlPv*oB0|r1Js0ATHx_?jE`yVDVrqEkRBe}#xmmLMiqr5OoQE0 znLft8kBiOaX{8{Wd!nbs%1m*V7UXE?%; zSk=id@{PJd5?5hz|0wYcUp0>{v`;m-;MgAWxin#-t*FeLW0HFgrp2|xnxr~mF~o=n z`&yX+<-~7Com#5F8+BjH0x6*yy0=DpKmCB6?9M)DuQzQ;aC$-)F?b(04HXha$S9jI zmq1cSXLqKQ@U>;h7>`?w%hEZXVhuBj%R}Q9bFb#=@ydCQmL3gujG@8-k#UkWc^YOH zdI3)?lAXtps%GDVt)iJV z6X~ZG%C5~RAuM$DV2OATxTI%zn&r0}1A58;OPppA7amw+IygXUE+%NmBw${58 z-upwHS$42P&y9#*x7<}&e0IvfP2T+rhqeCSO*hRZL(G$WEpRqU!{_R2IV2|g>Z+!; z!4DawhmAzDFcSZ>39&%)e{xNKEVw5g?e@EKFs9HRyi>d4+ZJ6*AxNJnRtz*OtzeR< z|I0kJU)HgP;qy)-{u9Lcv!RAsXhKDVc$|grZz1{MT>mRq1lVHD#kbyOy}tL2d;|g6 zVi2)G-*ZD1$s0M7T`2d<3~x*+896<0+Mj;`AxY*_H_B9rw7|Gge#d;jZ{3I{eExp# zP*V!GC2)nBNO=UKgvC6HY_A&EIo5L?GBEHkEgvP81#E3@c>r5jBUro_c3VGMZv8j_ zzTd!#D9TH1xP5|p;&Hd6Sw5;2nyqA%Lo>q3@Rf9;&iDG>g%*yvZLnT7?5wZ7#NUjx zY_Dnhze2osq9T2uM$VF>j{{=SX2!+jj}Hwn)QAZeet&Ae8+z55b}a8b_{rTnyGRmGo02Y5#cY zeHf3PP{3P|=is|Wc6LMfccW#~6;;kKT^zGx}jj;DkA;hTxP4vyox|_^Iqm3yK+s-wN8!5phF6dg|HwgkM4gNQ<`+he|jHH zs>SOne>@pzvA7AlY5cFo7Veg>h_PD3lL>8Ce&|o@b}h;oUE!V0IUFO!nYB|x@~Di? zP~ddIzdeuaW;GIzF@+QGx$>L-_@nvn4Lu&=P^!L>%jRnisx+u3CLPTS-6=*-%7%#b zG-=^l|HIZ;;}}O7$)t7d)nYYC0sXx)f-ugTxTob^U%cf-u@+tr$*uw-p%#NLR@!=i z&g;tKYa35@kR#;cv)#}NAt#65TpFByc+odju=ZpEHZ~FKq#N?!#5&=U@O|AOi zez=2}qj4UlNV|}qzPzlP15!Kic~>u4T}SaJ;TS9BD}4@&>(9_PoPESo=r`!67!`3& zE(uOrs#1dmwBMzD+w=-Yp@rVxRvLT%u7^=WR;#EPuIPe9$JO*-k9*Wx@9m^jb|$F-^@6FTy}XycFc$WMqtn8!;uSW|*uLPpf6E z5mW_UF5_u*)93rR?F!m-ovS{om>nW!|Cg%~(req#Wi`$#cp38!v9mGq>kPfOq)2i0 zxH&{iz_)O(D@^=*)}g|^Wlm0Y-w1+0HaBf5os5!R@|e!=c33rgi9;!uVw z5)vv3S{pp$FOsnK&3ev;9Plgi313)Mb=Zqm?EFq(>XjiPyVI}2t@CK^3CJ=Fc0fEo5!R< zf>WP%+lccaA(|FjesqlQKl1j(5&PDxS-3*tXl6I)8ICe*>MpjOyoA;zd21E>Qw+V2 zON4*ixX~xx76Y3$%%ScBT&t3K|L7CX#j8a&laC6=u?1@}NMtQdxFe+yh3K&2Y+Us6 z^hx@DqRfq?y#vUux)ElG|HFvMf>u3L%C`vG)Xrk%hK1Sxn{j`6RBw>Xp~i|Xgrjs4 z`x22881?iE-~W#vseU!pvekBuh5U2+mPHjzZL6s4fNq4E9R@T*Vm_)A!h(fSMd7_@ z{Wct6jyA+Wc;SzIGDxvxs!yduH5cA+Y;H||;6K6eEjpUCZ0 zqmKk<8blUnibchO3DGvVOt7X`#;yu`tE)#b7Io5dD|8x%pO#LnUW}a|P?dR}>ZvS6 z-2|2`Cc3Xdwez_Z^1=Ge$iOV%U+*CWzV{IWK+QHas0DOSi){M2M%I=J&$1^hyW|)b z7OtoZCW8}Ks6>GXrxX)nxi+|m1%LnQ2#=FT%$wd}p6-+1$-XxE9-sU##oOazxhuDd zx`D^GkM86$%;)i&sPDy4kchRJGVQ`Ew4|XgOwbP5q7mW~*fRIf*)_9(nGE>43=~aB zBGiNT2_IZ*99&6wZwd8DcE+~Do<~|VTV{U=obhyB^lCPiw7}!VK+b7KyZ1!`nk<_K z{=K^3W#JcI!v@&rhtOL-jQ=4%$pNvnyoR%^*{k+gVMa4i!5kzby*nd{?XnDgN+s8^ zM-H*Q5R`{sJ42a=tC8_h>Lp14-h0P3Hm=K;+P`u)N zVdF^aBp85sVt3RNnN>bXTN}5uhyQSb0Q*0$@r;f86R#q9FU?Kl46aF5eT%-O7wCY%6tB%Q(=Sf(Vkm8BAJ*-RDNlHZJ5Guf=Y0)_ewnW#~ zguj!83~go6u%Nx)wJ~@UOk$jmRIlwFxCn|TZTN#hbmw=o-*ujvz`#njJKkR;Ev}1p zvhY?(=(W+bFE5(lx0V_=dynSR8J7u1a8e)bn7snsx36df*Ep3B3mY*LNoi4mzgeBLnQ)x>^wk7 z*b$yFUlqsM`fkEuE<#hyB{sm1{DQ9^6}{;sBEjPZHmFh$Ai-TJ4a_c()O~63R#cf{ zNaQ1Z51)F%bi5+8fyaT-rVYcIFYcaJ0`-QVzmDBBN4Xp_jP8C`=tT-3(T07;XGj7c z+!T`G7JWY|PwElyYA*N>yP8(k&=HPOQH$ zzweTV>jx*dau33vZ(Vq`@yo_5YqWtPgS`nLvSDAKX0ze0BWOSp7`(KxhN=m9O{z}f zxOIquFb@hYpu|Ti@uQ3J5)le6J-xC?X; z&8w7*3lc+-qV{yyjAxwi$+$lb-rF~Me2|NJb&*$A2%Q(9e<8uS#Wuv$@fdX%QK?!? zB^&E$xRlU+FJaz>*tu%Cp!rL7p3<4Pke{MDM7zxvsgWJb~) zBiE#je`2G^Jnz3Y9*jta^!C;vMcrN`v@O^h2Lb=VUUGP5o`-fd4IRZeRzP6;^ZS31 za6$KdR{Oc7Q{af;muyiE&z7$3jF3!pSwosFfSQ~g1HEgrmG!eyU{Qkg7E{o~nd`X| z{*YzYI#irXF|q`GrAZPaX*58Txae&8c|swqp`n1+=@L@KnRr6vvSh?z2J2dnmyDsP`| z43c&)T+qZh9uP@X{bHmBY4bu6)y2I2vJZm;^cheV5EZCFxEu?fDr*e^2-{W|*>_k? z^o&@7K;D#eg4%nQ3G3GgamWLn!bq;Gt;kPi8}u!w{^WU%O~5ZwjImX6Z`x^7OLfI< z@g%hDPNPu{q%Fl zf|LNbFRmwi8{!mEd$_W(a&x9b_Vmy9?swF)&i#v&1<4Y$773d`)JAYE?;N*gGOMV; zmC%<)C`l8=N!Bo29bI!tpT#|MLrq32i&5KcMoQ&@pE+&8F5bB1xnPkskGcZ4t z_m*8%a&Lv+w~3Yevtwgt-R3#gIe7BXIl+0e(aoYS^@l<*G&|bwue+3LT`2=p6v69a zTQ`f%;eb+9f1PV@LJ#Z3w5$>fE!sh6lNiC_De2#tGyl(*`O$^t^Ug8{Y--vEBAxkW z$Xj8FXeQA%sY&TwT7f)Ve#dV&l14T6llQJLDu}2z>ySe=Nwg+09zv`c_?2+r|3nE& zOF^A&!*+qc!cE+&Dwt3^zwx(^{4^-qC$*e>e9HazfFwV$uVH4UFuL4;^56X%@-1%E zaOwbR4t=|Y7A>WNfymv#UoocC@ul0wI*%ZG>5 zK6XT*0tK}ZjL`CLx3=i@?zV5YKk$pQ%Qy8+LFIlPvp*MwKlt2ObJ>J7r5L5|ix05L zZ+{j=JRWo9)G(qAj|!X};(sdBP}(+rl!2!(D?509EeDX$j)Ln)# z1AS;F_c)UBwjb&Dot|SeS>)htiQ@4s6(x0y4SusuYPZC&RePR;AFE@+m}BA&>R4vs zAB8AqEIZ)*kqzZ z<%{p#zkdim(vWLdytVT0Sj*%afcTB1=2RyMP=af1+v#sK=DIpQEC_aPZPZKNqt43i z^)$v$YwbN(km?JDUjG5z#_fdFQ1Y(4#BvEH_&7!H_4AF{3BAEEEBjG@@=?rNff>>3 z-a4P8d>HBM)$yC>ighCYZrVy_YpZ=v+7)Jo8&5}H z?6Exo+A9pkcfZ^R@??Gw8Uyv+<{$tYZO^y?@U>&Ayb6654=*ufwm^p3@{N{l+9Ep^ zNK>Nj^UaGTA(96n{E}o&T!e)UpEvOu*bAo?ZQ@I!WvzBL;yY`wD^#DJB9~J_ zm&{_c7H2S$mHi8Lq}6yj7{8e)7T13H%f5^$1lI;)F^i%T2TGx6cC;fuA{$;QEyMly zT(63?-hXkVMMfxyLaq8ZQ|$IesHAs?^6>{Rqt27P1{u|^;Eer+rue5Oz(#{yzS7gL ztCka5gguq^auyvXQNvePX(zl(ot+eimK2S0?N1nYMWt^ClTBjyB(4*n-+|yIa%^ms zIOW9oIZ7_}F?s=JJrj8Qbe%)2GmXS3#8S7X(@$Hw85&#lY=^e+Dn8Px4mokckyVla z-c<9|*}6xSj23nWF{5B9Jxp1e2=svi!c`~QKuKRN1Rk5aYTzt-6IXJTL*yh93AJ`*oT#v+gl;1KmP!K}918k8EY#Y88)v3(z zG-H@~NaA7)DNgxrs4F z7#7>>C~Lk}4(|aQXgsGO662K;SPQ-*wCitw0ofYotW>DmZUl4>Tjqhy?ShBbt;Ay{7ktlmWuZglv0ADxbO-6-J)=YV_L?sK z;5p1j3p1|((yX|s6=ylWlUnOzD|Rk~oLUOvm=|&plX6$qnfN}cP~;4zmc^8=b7~bx zIu9CUE}g6sy9jWr@)``6qruFk6jJnIBxZP?Y$tqP4m}CYT;SRBmtU{8CLC4yi+)eyaf` zGvJ*Sytw4S=m4pcT`gOs^h&Ke1 zb?YV}NDseuUmrb9Pc@w05B4)*a#caR!Pf^yfMIu9sjOwb5Cow|XOL3!_t%?jM;&H$ z0&e$PtHNeg9B@d^x0^V@%5cd0?R!AGs!Qpv8##);X4wc(=V4T=`%sVtv^lH1?6*@7 z`Jih;*39CWg>aasl?}Pvlo7u>5KM#s_M9cYl?MVJ=w1!}cF^B#E^Z~k8pS>6U){}< zHBD$>g8hw^a_h~OZOzQqmVc{tXLqyprpv4J%Ocq)K3RJD6qDBa+MuI&b9wC9g|VH% zL5r2smMi^tfpJD5wfUae7@#;YIHc&awS!>55-68}u6h5xw(tDAgepCRSNaz#5ee+( z=e+Q*dSMV_s$Etr2gb*;V^=IhbNG!TgVkpIVQyYvB-HhidL41i2qonCUe%W}!O^Gd zb8o3ITC%VAgqwf^gBrMbbB5$0+&7{t}y^dEE>sR@Fiw)Pl1Ip%l?5A>f9I zVD>ha{7!w4?}@F@xV*L=+fR+?6%H8Frg<$7a&P>VRht8FP3t?G16(OY<{}NfSFb|? z*?R5Tp4)O-3O-Yjr3saI2NPsopOy%|X*RiB^nG(?1fQPoXf?|nl z4;%@{{7uIP*`^oW}gWb4ds0|A>MLPhdmF}40JyD>iw%vD~Ljm>p zxfM)8Yj#sgMfqLrBh88CEp(XV)rt@*Q|($xoFbZL)v~n_Iydk>@#U zSiJ|Shq#fPbJ3^kV@nBf6`!h*^~|u5*<#v>TU7 z7{%53b)IFJ#jDns?tQl!$Es%vm~sY|XlJ+uMG06FNQC$;nQA5G{Ad^Cg~EByasFuj z)5=dvE!uqkENbWp8h_=QdXlHMM+lBvT#Vwy?MFQx!`AE|z`Dan&h*+jjj!V>y+#G! z%X5_d(0YOkHSVD7JqA#m`v0OaI5Xqy8Qj^JJH6|rPo)y(9&UB2$-6EcsCMzsF)iBz zvo(XfbjzPRl||6;=GSLz{S};_E)+I4o2U_*`zGRoJQ<7{!O$t+L&6$okp-T}ZHRtg zEHv{>SdjO={IZVp)2QsnBTU%e$EXtjb3PN*=|+IB0chbq3_}Jn$~F3;zbe0oFs-#n z<*!dF|HC@hFX9T>Z}G6R1Cm=bsI5S49;NIY`4g@8tL1dutrbE-DIYt_oK|0oSvD9J zR}7-4a#AJ^NNcXueEZbX?(>V^M{CPYDL6O{#Pkvd{>fWZLlY!}A$79YvNroL{}YxJ-A5l#y>D;^L#S~}??me%lBJlFsg-Cb8EdKs5PlyY1O{NFyM=^V)PEM#*HrMD_~*2nrsQdmQDm!jjCD3Vr%kH0R86Nhq(A;( z%GZY3v9mLX3A_z)$H<|nR}GK+C_(Ni)y|SCB)idlGLilOGFIznfS!taZ^XMSY^|xQ zyB>_gOS6`gM2xzJ!-blBb+d;j?_EX^o;G+^2Zy$*fZGQ_Q=FQb^wT5S3IM}WvfK_R z|Ii=_#^taj(uj?1)Z$uR6PIsfd4)$f==v3Joz;FOaGd2fm%93^u0)0_7Mx zutHrITd0@N9sg4@(V!jar8gy}4s%pdA{j+`&7f5VLNEf6A*GZQyEZU}HazUO#!9!V2i@UI?BER(OJq5lS&IM9fyf0ACbMRj_2-dDC0p)fW%!B08n|3cDGWNB?Vp3+(MWx2Y;kd;CM|>+^ z3(tp=D3wr@a~gX&goXqhXR6-rtY9`3iT`-~SjiZ9+T-t0{ymInEpCj@A6|=Mj9j~5b z=ojJ6SXrqjxCsF<6tsrW-BUsht+D=E1Bp%;uHfF9@tWGlT62q2eAY6Qq5G`(fH@|o z>FHVf7#j~mYDKaAwSgy6{CEEJequ&>J2B`8^`%Zs1{LoPF9bdKN^qqCwhBBQbbY7R zV|}C$1e1+0NGpz)P;0r=ln+s?x=V;@_GlHXSVM9@DgGEBtBkO#o~FWM1w=cyw>P~``_1uJ4+|3;oRE;SU4hov726lHSffmn30|ScxwaC zv}d$pKKoe-o2on#s5f}QwCO?hIQVNBOr~Kcj)D#zL>Xesucl4W`*P)Q2a#pWFlfMC zB_z7pCwy_@nR{PhS$APV^TGnFpuCcl2xK$^CQd)}Cq^%g-%o|+*Vz%(?`+5Rk zphp6X(?Zq`SE-B!k1B_dj(&JVP(Yo%gQgXX4&^Lz*LeBY3UeZt6gyVLwu3vRhQ?x% zTSMBC2Y0ASd;g(qIl%;}?J=gJ^xs+psQ$D;B)#Yl+E~yB-gB~_y8tsCUyVa{N}Gz` zKBWzy1y~f_#b*2>)flm7$qf0Wst>7;`-g@10ymgz0=H$g*UiY=n$D~Yr<3r0bcHcy z$Z9*;1n1f_Q)yJs1W+}2)ZCP=GLBdv#to%apO#~~yyIbio^(uX%T3j3-z7MAQ&laN z;_tJqcLV4L8||>Rx`K)5QCab73HMLA0ry%&%j$scZ2V}K@dFXixZ>(%LUHWn2WT%W zSvAYau0iohDP!Y9MYqYXWcggzcdc#RTWtD>1_@u-k)Q(#s_9&;ouai`dUaleD1yx> z60aUwRX}K1Q=r`{wk)k7z~lH5B!1MlQ6pxmxB$woTU_rQ<3L6~zdvdC*5p}iQvC7i-#$4fqQezDo;54EeqM!z6r<;^^2MC7 zWvrZDkomdkY?*yT2C&c|H0OxxS{`rPC!aDc;bA4d@JqKeKO%QXkmhNM3G6WgqSr3W zuCD^cBuX^%yPC6h<;a>se>fNIx?J@do8cg{MwY!H2vZb!O)@X8h%h+hKzJ_FBf|{S zSD_ICF2!fY17-k>h7d!mS5_umpvva^tdp^l$~pkMOwVh7VP)Zdx)7k+eRhXzS%17p z>uRnzlJ>?2@Dc%;v{*h*u2jY&oi2e2^g)Xfb1;PpIGi*UZf_!|vI9rgdiOJPG3h*7 zzxq^SlY|dPPXtUiBKl@y02 zn#<>&>r<$L+X1OWN44-k&j`$=?V!bpQ5)geiMy$X>&W@CUp49pb!aE#>?EA;`Mrb3 zNy6gk>&Se0VeXHHo}a==NyThqs6x;$DXV++r_SyOLq%%l9Zz8ZBI=aGEi13)1&guL z8h##Xj!7Bo)KDvs3R2UG-4hNSvup5{-YFYor5;= zC_cGM^W!e2i?I)TL~9_YDcxkN9{1CYWxSI9>*)DiMVDdw%b$j7ZYSRugTy|ttX*XY z^TrIrTbW7S(u1-DMkm-3$C6uk9|xb&6lP3%su!wg_)hi0sGCc>@GPShY1TTz2CqZ0 zOA$unQT0a!JFv4_!(f9oxQYJ*ALl+7q8xdl8Q;}y5d;Czxf=2SY2+vFTso5*jmY4P z-oa5H@8B8Y@8>jQNtZ=7!JOTg|N2v}^74p!oC^g?`D1hzR2<_Xx4-6bG@ZhXbz00b zip8Q!sOYz=1ft9I8u(CW{K?{v2eE|EZSBqKQpP3QK`gb%2$VUNzmA%_qmP*9y;*$n zVQgUV3g7OCluB8a>z~xdh=;RpZC*$8U3_q2g8e@SwMGrovBOO~-6)--^;TzCUJp9| z=Bq7xBOWA;eARC?jYRdN^9|$zFlycz=!DQs3g~pTnwe%E&Wx83dE|r?)G5C)(S|ZP zL?wjr=xW!9r%NkA(P}lMb8EkTR-8jrnmXzL1<@4QnEnhCEvu@qKL0aa)wgx z7osqA+Y2PRl!CS&l_Bz~3)qfCc9(p1`8Nn%Z@29uO4~6Pp1mPzmo~L8%K2dyBGjJi zdEvIH+*EKpGdRl&OfB0|A?J-u4M;5 zo(2ITVcB6@RaKi~gl~i7xM0SUd&W$M0Vw?7mcVGUA=NB3BZaTyuu&T=BpS6S z#OROD_fDLb?xl$5+Bs1RAFV2^b(L5?zr+(}M{2FQ_@xCECs#9W`#jxT{aA25@2Y8gUT!Uz5r6Hq6rqF|# zJjRrnac28JK%NUCaxgj&rqN;QOMZyVSS^`U>6njK=oA&>>ot7@2sziXY#LgX7>UkE zA`6Zr_n#+B(<~HK=+6N50kuc-R>XT4$Q6;zEo`b0vDdWH0hUNf=gqU*vQWS+1`~!Q zq*}wm=$EruMOP8l^UZ3MRlPHKXvQZ23y(a$bN(}y8=lVU0*mGBz&_$$$ZIkm=Oi{* zcJS}R-}9_@sqzQCGp8n*r}*jC11~(uHAgst({meLO3v=afzfbm_6*hcHcs0F%1Tt<;#d$Z&KzJG`^!MDAbD=O z^HrLq1yOc;hD`?Gw#es5U2OU-@=u@$YpGV4O;DZP%0f$$&SOZ8O2h?}8M8fRJ@{GZ znoGYK`5!04st+#|vu)fa_#M`lNp!Gc^}Mfb{Rw^dj4yOk5^8fH~@ z`inE`j$i?=Mmh=HU_@WqIR}rnP8ShpKb=A6b*%-LnRz~Zf1tS5!Q42g&GKf zueAam33qPS{l5347w&1zwncvxjMuaq#w%m_6d<)6z!82jr!6oVp?@q;q)@{Ful06x zQ#h3ASr4f2m^!H7ah30PI2!zkL3S|N&Wa#b&0zDxsW-LqKJSq*@IXFw@NceGaGoh zM~TCe1gj9%;eF)wD(K{$3kk(h%|;@bhgZ6h-E6nRm*3=}cPk=h*=>*ZcpIr^J1m`A zB~PIM{@H%OVTgH4j}~QYY}@=}9lmDofQNCTGlUp_bEZHP&_Zg&0uLCRgD_sm>%DU+ zUjh<|(Xsgx!$evX-dAUejyBm^aGsH|DxPO3&f7Qq`7blKahU#F_o;O*%fegf`Tq7(-vtmu*vy3&zM7=R741D__|Ej*nYZiRDPsY@jZh(XB4i)gkQ{~#t(B+k|>Cj z;mxp4uthp~`V|71+g;@a)}HeNGW;BYH@^-B`wFeatI}2FvxPpjX~EZ+`6P zFJHMWBk$$-p~ccOvJ$w^L6tvgmnPO%&Mf*3yGW$sxa6ViGY81KB-)LQ5!T$WOf9Ck zf^9LRUJm+hf5e4lmOZ^kwVIL41#LCR-Eq+6aZkr{Y^*Tgo6-RJOqnK85#qwL+*R=Q zUNxS?|H!MOSEf){ry$s*k;&u2w=xt5GS-Ir*R`3fVIZbnQcd?Xh#hNKkXKI~f`BB( z2WuGc1JZ)>E6gpf>sYucP7%lB`OCrq9iE5~96hmQN*~E3vYRczs?c9?woW(H4yPdl zodnYeUWWxQK~tk!tzLVf)$7qU_P*_)Yvyr6hNPsPG;DcH3+KbTeMoDNnqnAe74ZvL z@kP=z_QVQzO_uTn^RS!^`)$=JEi;c__)GW*uPnU~DGWU5daH5z}Sj<)Wv8eoQIx^Ic}!2}o;@g-V_ zR3VU5u8_$uSR;JKtmvTnGMtJB%u&V1K^{DGWV^Y}t>=W*u^HzR`0o%!QKx@qqi#86 zIe6OjJl@s)LHghrszY!_ZA^YpHgh(zv_`%l7W;`Lk-2Zwp>AiFMNR2)Vc{^`)mlp{ zAs|h#YG(aqAR&?#<4tVBM2EsrI(1QrOUS9&ljR_Sm=+q6Y^NsqMk-?&F z)gBLHYzOc@gc^Tka(W_u-p8Nc6x<(g_9+LkWM>z$`1h0UNX6$~%rJ&9?&K#e1 zqi7f1nVEiN=j~_ox+|Q)=p~%U%@1(T{9Imzbv}0&pu+reAFT8;ud|FWi^PqsqiyEE z<={uueA#T1Ho_0p;2->h1n}K|5H9iLF~}nw*93cK4eEh5YHarnm8=Un^eSNCptJR! zCrxl&xD8;Y{P)$_tG^OOR5VBcVCL*XOXB*ELIFTJ+IYG#MpE;VmA)M+d%YR%o;Sf4 zI$dUI6c;1O)?#G9rrvrJMwij~!LBH3w6*os%yVEjbtBzo>89EVNl`A#B)M zwvu1~fkxHo=64s&(GRbLE$V&e@!<5rotuFKiI@hcLX2Wq8W!j#V`W@6{pu>S(H=l{ zrkj-NcUo)Zhu6O>G-?4gn&OGR*d_?MMkTx}{u+2&MBMH6C8KIv`u)l>Tx14i25GH6MPR<8;*~SNK%y!h@e-SLgq~`1C+B1it!*ly%qxK^n7Aj~~mIeY~p%ex=G75P6H=NV;2DA3hIR4j%q zO`sLm2Rd}l?#PY+ER&7r8JtbqXmarrZrqpxfP;s?fK=?W2lEP$K^-q9NL;hjEvkwpr50q zXzy=Z^Cf7EO+3kK;8GZa_i&ew;8*xvRt-RP^(G-jp^f*eovZyA*%6?{XQGoPow^7(q;L5 zR}V4U|WU#vw}Qee;loHz)?le7{_Q%do(n8dqT*0%TPw9{V=$Z0Ju%6>Gkh z@B+BmwG4Qe>ENuwZ|xQpQvwZx(=K%AM_RLO91ssS96Dlbr=GSe6n)izr%VV%Mra*m z%5be8y#}aYHj$JPlP0Y`+u{kvBkrBl;4LaZ7Pe=or{48%NgTwddkgf>zxPumnw;=( zMA9ygq<;h56{c;Pf`meUbx#;t;Y-=Y%=CE2!wP4b=pNJ0Jm5jk*lfSiqTB{J9?P|_-e(z~x3&X8G>rj>y$b0H!5^n6k;jAINE zWe^^&=7x{AL5#PCiK1_EeC6C9(91B!qHa^e&i;VaXdRXVzlnTySLN~_I+%d+;wmC< zjHtWold!^n>^tmHc_oIpmZAahqSHhnVa=Qyc+|{iXr^O^_gBvYaFiSx^J_OD`KcKy zVAJzkm04-o=YM-2f>0{XrW`42PrT7E>-WHxg%zTB4x0K%8;QPl5|B^r5z`Iq)$4_R zBC6PT61Ep;68Ut!Bp*Y?$q2s;VnbdHN{ioF%b(Q?QTA?3GhFKxsm)U7AN&!PU&PrX zyh4D4%@=rE+vQRK>q+6|6vyP)%sZB7oL76kIYRXYKPN>ujw^4^=C(%VQm>MEF{LI5 z_28#e`xK9{J)RAN6k2Z+en&%v!Jq$eM6a4Ly+I zuGvBX-*cdTS{w)(6E0_PM-JSlm0q}ZFh8L%!`tGwUMwVcEhsEcc%Ql}zD_5jS!{cL zZ4w+=+C`09VRo=Hwrq!{>n`PC04sPyln)P3{%iOOrG0^-ZdYozx0D9k3O4LGpgqIf zAV=%q&KGzn%evMia#*LD_nKv(dxgxl;)TFN5&I%fV8Nu`7_H7L7YAVK0$m4>-2^d%2`_I-uTGHS|E8qCM99AvkfDU@+AC2v>r)HPD4 z%faZ8b2ZzBg730=EK@_nWiUzWqhp8U{4e|9SI!Pbqy)bB&!E1am1N;BGMi#Zs{VEY z2~+u@)zPT1C@Z~seW#q9Snn>B{vI9avjt-0sfv{N(RWuvm3((5-X6EJ;Dj3;SOQu> z%vlsXgYKQJmvwp(P~SkPKXjR(x(K&LN{iDS>u-lOzn{gsp-A27bN3E^$KjDs^-W3% z`#0lAtn70G0N~4s5RvqaM)oLC6f?m9(Iw}y1C=9$0o-VcQ=|RMV@m#T>H@Ya%2ugu z9iGG}q9YOgAPsxB0%BGwp~w6yiCDMW1o{9h&Scn&rmMITPNMBaG{ zWIq(ix)1fgK(eCvU&*_Y@ef_S>#9zz50Tuv^t?{iwx3W-Wf^Uo);_eL=I7uyg_LOk z&p<}h!^8xU$ptRWeFATr7VR>$r2dS|bH>YoLzG&`Ai)Q?Ham}7FT1}*8u@__RqztC zXE67+QOJiZbwIDGX}g(5QXmNURA^J+6b8t`bu`dg>H?4jOsDYMGCSn;A8i&fJggxn zQ2P{P91E<~#_Mxa;ObHMJ=o76o_%E<4|O~1(fN&kOUKn&h*Gc2P5<;z``B!vJw$k1 z)C{E%qL>ES3!gdD!vP0ul88}+r$wuKV{;J|hb|Ns9T!lih`i`crVy8#U4OU&aSe_m zElcK?$VbR7M4$moh1TP8?n$8sYZQELL)p@SuPOIUu@69WtN?

g`YU!2+LYsE!{CZd!6 zYs3-@Z^H3@l{_I2)mVXlr}QgE8C93CLPh@lGS{)2jq1AU?o`?Nbx4| zj@`Cg0sx}U%K2@fF?X(dMsMw^GQ*hv7QK$SR!uaVjZ7s)UJ1A%!G6W6MpJ~&P?R`jM;?X0PNb* zFRWQ|a$mT2(bof@X3n0>qPz&^t-mhB3<$11F5>fMEiTwYW3awNMfGM)UD`1<;8^;x zXg-Y79vEJ(7G(BQ*2FpX1k@0_)T*(opF+}2et0{E-HQJliC@7W#FkIfVEJgK@WoW? z%&9kI=&IzCi8tG&J}g|7UMz(H?=n7nQ8$jq-_m#C3%;IQL+0c(lzaRqZ&;jM`J@)fX^KuMz(f&A(WMCtemnXN5 zJ-Tg%X4GyPc?y>2sAGtPa&;NKE-;#cmM^uXsy~rD7I$9FIwo4s+=bvm*q&7t_ zOTil?H`U6)X~-+qSCdQ%gcxU`{ON0RKFCiM&laEi>&7T)Uz&oXg=8l9)lYn$9P+OR zZ4I-97c#U<{>`pxa>SmrLsM;(wQl1>{q6?V( zz%vyXv%RZGhf>f&6n=!Fj%_;tqDdTrFRG5}&(V=DzLqEbSlU+9sJc$V&7xz<{sQX` zGWWwb_0FEV{dc=j{YF4zMK~E!oRJ&ezCqCT|81`veKdIPe90_?Gjzk;Pa%xh2G3&! zWn-mWC}MdX&R6%Z46*Rem$WqZDJ8Dsv^*fjVkuCX&1x^3C)K5WN~>*GxF(;9PFueq zChPPl;IdsDKzkpQ>pca!y|KTjt=TUlD1gTuw6N04O;67N+B&e9q(|2)`K(7kZ}h*N zNeLF!MHl7c0n#sG1RE}&=+>g$Ze>P=#Gv<_o>xnk0f^4lTCGwl$#GWhTLl?kyyqeC zPCQNBwx`!U%|X!dXK+y{Da-MhgRzvfCZzgPLYG`$KF+bDTTs=c|La)A`AH(vA-Fu! zq^q@vo<r^JJiYu4pQyh8eg67zX_G$#yNLwP3QNwKI~{_Zj@yKNQ+ zbJtR9t$N06{j)HHTWF&@g=VGLy#rVl3OOIx5Q;~+;RE^fs1$GMTM;y)Mni=eW#OZZ zoRG#wm2*Q_l!1$!Gdnt_FCxntsUPnHjFexgC2FYzbyFbT&~`<43UKdlfo^ai2SJHNj2hd7tJLamg>Mx12`w8K~oT@6xJ* zv+{s`d;s%oBlJ0K5)rn%>4Vk<(Of_P#+91zw3WTgue1eGoTlLPaa6QI2OrvaA!QB69(7>3eVQQe_NUIcLj4os z!uSikUjk3d8~2YrAyihBJ_LJ=)wC}jiO&p9ocv8l@jQGpB8!FkAdZJ|yxAWDqaw3X*qh0fKqvN^ zZ>2v&Gewwi&7sNy?3Aj8sqXBPbTP(sP7?Y*RaZPBo%cTOu@4ZUF-h((=;JYSiryjV zn!o&q&JIf&^1-~S2H(b8D_lifj5*`Y(f!{qeM0) zedM=MDMH3-zGONv3ZQ%EkP?~)Zd>Ky8V-}XSl5(2N41d2L8bgwJH|GjK_(FBQ1z{) zQPgPPwcbA$fAT`ck3ly9Y_dj%ql=m0@Bah&5N2HN3Nx^QOYwD(r>8qTi(omQUqu{A zve5O0oHgt`g6VpXXr8dVuVnfH=2HQg#UZl({et5tzc(eEYm>wwTwh@n9+b%5CU=ruk6g_KMBaLTgZU0Ay5|nAGG? zdVIh-#XyozLLNxcFZg=1^f zin3cQJhCmhius3=huQndLK_=Q7>buY?rU9L98L2tPDb~F$SY3RMLFcHs$X!W@MGD8 zS(hoByGB~aKC#!JINTm?DT&>-#Sed42Eo#q?dnI;D<_=fXo`6gLIbE;<^eDb@LVsxEOj=wF5{00?T9qkC4)dcG8{JSFR~rHA!w0<3o5h29jufMKf$EF6QzT16icB+0mAmM zGL(fL-PJ^K+f7w@y^AMpS%&L1?bvW~6+Z0zt!E?Z*KOgR%i9tH`A zYEUxsk~&jzCt3}*hCmQ9oO?O738pr6Y#Dz~2NvTm+=XU2<+oHctvr<@>k!3!;W(PW zny)O6a(eAR3Lu#k-cyxo2hp7edE^C|{W7H=n*K7{`?X}Gh_8)iUc-nSZs|8QRonZx z_wjc(&`YI%`vR;dXvq9GqLbyn5Sh zB?qDThjDIiXAk@z?cyY3W?Q>VfGgFS8Q94Z&vs3LzsZE;ceMXaItm$bKl<)VO>z zGuMM=fO&O|pcsJDu+`FlE+qh{d0>3X>Pq661T`eJ6?szPul02ek=@hh@ksM4VFA|Jf30D2&8 zK)Bh|KXX5+tPSwSe!h~QVl2*0jqi~FeEK(6RJC*#Z53`3-H({t2B?0Rb?5BXa{Qm}Zt{mS2cxR%cfSAoOpOFZUoB zale8D2n1ILXK$f`FS`GP&iwaOC&JEHuEqM72L`9lTU@3l=YK+{ruENbOZ2r>Hg#53 z_D?+#f}Vvw>{sE=`D~8*+K+J}YEmK!va(8k^1InzO%3(V*o(D=|0Oz2x%m+>X#v22 zvHju`LHq9JiwauE23%NK@hh_Rev@Q}4hm&vTw?H^ek*IMt**DMx_?@?mj+gbo@9@u zwXV*Ut!e%loe4z0Q9u9O;Y|}TCouZ|H9G4#{xO!l9mlch_p$GVFg!XhE-wIzf|lfY zvB4B~hd`IdMrTkU>|7lH-d#U8Z?YhhV{r8K^e#VUPqF&`-@IW{mfep6=$#HR*u%?kAPt*m}x2Yv==&#m6(KLc7!R{CP4-zSSc zOtY$ZKVA;6>faol0L?JdF@Gmt=MzR$TKLGx0hmU0Y4=1iehamG(fANR<0rFJe*bG4DDu3V)jKJrxf3Ure zFn{3<^xygL7w>I(iGhjz@ZQ+QK=RnWvA@LVf3d&x8s4zK;>^C;-ujInZO7gS7hY`d z0>^ItQ-1za?*CKHZR241ZomF*y8QT0d9fXP;y=1a|C*IndUFi>X32&5-C6d{x$)IE z{zqR@<#hy)kCJch3K+p=aynaGum!d(7X4I?d3M$E8u&K}TrlJF1@X19$QD6NHCuy++<9^}I;Hi{hvSwb0Xw zZ4J-d5%%kfGOR;CsU2z|lHCxo(D>NlFe!x@{K=sFDh>qVTNBXN*F%Slfd25O*@8;56r( z=9c^ZR$F4Z5a4~fd#S?nYy%E*I;A4l)-C~r)}qlWnSaM<*gdk#0z{@YpEar%4mcdz)ZmnmiV}RS5e2DDtd(>5KJCSt zrXGJ#Q6AxLNR=wu%qs+NRA(U)^V`pao~4Bo_&*rX5DZ47l461YmT?s#a<92aS37 zwfJS_dw^G$S0LuSO}cfD;5KzZ{(ARA|5+WBg~vQ;vc95_%{B(b-LBv(tbYv%Lk#kz9X&SNoZhQi|ILmhvBVO=?{tIO8e_aQuf zk5$b5uZ@3fC5cNH(YZ;KtOG=X>QiJvQc-Zi&Yr?EToat~oNjj!u z3j}MI{S7J#->YL6FWQGhI%MN))*$HOrV;HB&6AwjXK^eUHq^dNw)nxF@A%c&0*jgz zC#0AIf^D!_u!iylP_YJ)27$^XO|Xi)*^bj9_XhF$xzyjh8v0<4=?OJs_D=`m5_ca4 z8oDYi#HMGfL}&+GsE%Zjt%?pZ8A6``+3$jW>OUfy|yXO%ixeI3p{3yD*7OMo*%N5-~%uWBoMn{kcM4X$d?p^^Ktn2C)4P06##$zaLBJ ziI4|&BhyXxK1m=)Rpn+oF+2qQL|~6p=~CEh_>1!>I#JbdCrCEGSBtX(*;e0%OxSA%hgS z5Esk}8;8AGBmhmPXJB?)16o^NhtCxr1FK3qWHsRhrdx^fCBigFz+FW(NE>aFy#WPN zn!%JVDH3Ig;dKHVq`e3;VQU4F8P2D63Ge%+@YAzdyUdoO4MYN; zeIuw-I%ofFfD>g>=5rxw2Ga!$d6(oex@rLzS$tk#kMzZ`8|@hKyLt#Oa8+DR<}~f3 z{0QlURsA>RsX;lZ+v<09lUU9;EeLYOqI$A$0cluW@jk_n$iY33PGGr_=%59LvW2iF zC)q2+$Zyx>tG#xZ=XyrekokJxRoT@^l@4-Om9GYWX75_v33p7%NE^|61dG-Me*zky zg2lU%GaTa0HMjC7I*Suwl&#_;UDt!v;g2Z)B;S@6%`$zjXUQ1>P|TP|ddNCI^-2cfkAsH_AnSx|q!wQxR+ zVqpNZlb>K!%r=5hOKAcmq&)(5>$a=W^J|yWG=s({=~_#VXU<$I zPuW*%@6&m`CR-sD#@y0_==T!;%`+U@9pCnu_6kT|8HmaSI1xB$3XeSA=^;X`z*+w@aqU;?(z9eWG$F zcgc|S1?_0`kvzKmOIP-0z6AOVdHNpFz_A*#+WX*?@PUFQo<%K9VfhA0znZ9S^}?5@ z!CAvX$zH6DD4guoI}Y?QHCp4TJo+&6ajx|-1+mqKpy80c&nqhAaKGE>@A@FcD8SV| z*Uqs_Nm(+ymTs7Nj{&v?$;zVrH~kCC*D)>~J+hM&(8RKMU#lgqt{4^h+L_6vtHNaE zTQqR(15Yp=pu6rLc4gxuuPNd!a>zZo<$iSTRk%uZinMdkkhd60E0}Z8xSEI%Rb!>+ zys@XD#69C=5RW)9j;d$o6c!baWFdK!7vmhHVsThpQiGR;3V)H5Da;+xF|v-LQV#bs zznJFTj$?a%;aqFWfH-|*#7)+~=5=km30wS93!+S14?CYZFD`)aT*UqZZ>fQhDgUsy zjr%IUcj{r_A*1QjGS(?F1Dt~A7e^6{_*2rWRxeF6=vEWzx{fvWPe8E)Wj{BozSfL@ z0+*%q{l{l=Wccb@JAnW-ZtU9cY76R<4mBM+oS$BT)3MdTG#kww8D1~;AAF5)&=92q zWp)@rI-%Dro@`HTDc(rYdo$wU1mrnu)Vo|ykD1+78p$E6Czh+r_lpRpYX-t^h2Y=1 zMpk!&P@P!q$a}oM$yR19m6X_zm1^LUS#LecnbXSO(+*XNOZ8o0K_?;DW;_?+-}kYz zzY$5?;)d`c6T6@7=j{q8VcZw4=#^8lr^E$;09L^w?EGW*R z(6aT`$kQd_k?Lu9dN{2ja9L1}*xOK!Jc;E%DPh|Qgn#-znZTe|t0Y-;eMIRz{T&Qa z=4JF!b}TD_tO7h#C!+3i44bOL%$C#auL zqwP~1!95bLxE=1u=SUf;Up0Qu^LqS>es-HiQWJc;`VrSt${DdS$XxQ|lKCWK>VbJ>T3ChPtWnDa*}r&GZb*l;UVM((jSY)9 zmcKY>PglGRGY;=PpKIU;NYxJ<7Pu*)@ zVpps~9;<-u>gUcvvW!X&)WLQZu8)+?ZwM&>NiRnvU5n!y)cQ{?OWhZU`D=#v;;P3+~7_-7Is)?3BF{c5Y4|DjD#h0&B7KdE1(r-yCHkEYgXL)t629?7MYExz@ z=%Nb64dbaAwjHT=)R5cul-^S_F#;no2W(aE+S2y~ksn!{D{HpR32STxu#xQ!Qy{f-QZ6JEhaDd7vkRPn!f&`J5IYp z`U#UCu^7@0*0*)8lAW_!l5|fKo2;`MInv28^lAEm3udycV7FU!7C*kH4mEFcuQyG; zHnrg<8N^Ghr+@yKVV})%pH|6R<0gKqTdIlgWKFn zCt9i3tEW>Eihuun`krYq+AR{7czlY2C>EFfF*ni*q%fhT9-^^T z>=-uvo)$YHE~FLY$$(1V86B5*SpvrE!Fd8l5jQNl-CnyUIzrXlXo?X2CHD8Qdt%oC zS3-3muYTv+n-p)^nfEv42cWG$^m-&uAjW*sP#enWxYTUIx)Zm3EZ$nf+EVJuf=|p6 zF^n~Rm-lxKflDtI^_u`Ts-3JnzS-(eEmII8!%`{t6IAiu%bfleMNTPFt$5%q8tQ`t zGAUVNGPrdPL_q^oZq@G2ZT5T4@(~jox7q?$LuP4%nDk=J9Z8^K5HN|TT;z5vK^5_f zryixgCT4T+V!vxO$n%Ys#8fh?n>fSVJx6OCA5Dj$x{vXE{OOzZ1x%YcAnYu>O0#6U z-#RhOZA*9xaWKu6xR%+YivIxOkK90jY8-C%iq+J6Gky2uI)EK7$Y{oRQce#w6A&f!x@Aoe=MUlE z%`fEk?*F_hU__Cr>S{dCyEo{5pU&S46;Wwd>;aiK6|?J$Ff!1o%wV*dia}cP7fg$FE-kTAupce}(Wsn#Q#>D=#CW zTUnMf#$xIODUbdjfFWPC&%(qPC@rn>cP{^K33-CDheRMRIeg8{hq}U}ca%)MEU^K7 zN7ve@-5-i_aBG6wL>&Ub7*cI*Pk$=heemO_(YZ9yqKl9GQU8Rj5GaQ{5&d!ePPJ%6 zmq)R!H6;x?-64Ot58eEx$&3<62|XS(lzaZSs0e_uPW%A+5h8{KprwigyM-Z%;Yaw2 zk|J5?Io7$>VY@+FsuV1Zv*fT5NBf+}3@y4G3ev5o&@yqa@cJv7%!aJVYiAu(zwBxp%Ijxo4}#E%ihEq)g@jP3y1QTIQOnfMY38 zP@V25ps!efFb?}L8eTzqMHM!)tw{esaDSn(W9B(j{VQayXOzmH&=5EJ7!uu%=K?t* zyen)MM%~f)umKXMS@YWl^O)J;qLiG|Xaim0*8`}q8W_ZJm$iMPekzi6g39&hC8Z6_ z#=L7NV#C}bRju3p#bF~2@EHB$pM6st>r0rQgKPBmIY%GD>-yU0S@K$vn7ddCL`xyC zBoK0KzEARbKoI#`6)qxS9}URw;$HaoEF}vntn$(LRS+A0qEsj~!QjL;$nc$#e_&+g z9bes%6Nt0(>p$jQk`$4Iu7>Fsxk2L!W8C7b;;9m zwazjd)l!UGoQLwpq54MkE!<@ePR;kH=X7aUYZhsuS7*;k@)tapfx*;_w2S%jFsPChMA{ zo7emdF$cEdy6{k=u<`4Z#e^wqVmB)KcH-HunhiUJtQm8^tcLjcMQNxaGCju^3^HAu zwT$IWp-(&Za#m?Ag5N8Ng1%^=8Xth8L@LRk-EV=~mpwUQR+3g@Y(-Sd507O2Zk%tm7n z`aZkL=x-EecuSWtOGKxAk+Tq@C4)k#<@ClTSr!D{sR`M;!fI~u`rLNVLKZbeExS_W zbjF$xsYJpk=?3lFg=?uC*%u#79Kh{#T%B#BOi2<0JK`D)+w63s^BMWB*}|qT z9=o1d)GXz?;%X($`;@2i8>(@k*s)>TxIIQ*dYIQ;XJG!taTupXLdu~Bxq!_+#^reC z9EeRjS3^#ZesOQYLn#lF)Gwn9vdjs$dhCwoq`XfK>a@43`+LF5?s8pd?y%(uLnv}(JEPRt#G;!@#K;GX`EF@GH zCP_S_(R!E_jh%P@oa9x{?(ghhu(1z}|N1IcWWpTz1%HHRz6vq?p4ATjz7hKK*@8e5 z{rym)$o2yuXXNUt&y6$)y5~$^crCFZbD(~*EznPKE4({fE9?%tY5f`;=Z87@o)(z> zuj_@^4hDI>9zq%mpM+=Qh5clma``tT$aQ_U;Y4!@UCz|c1=Z*`!SxGSE``+2-sz~O zJD-1uy+MOo^A2L-IB$TjIDei9g7eTmkBEIrm@#Dmi&pDj5oqE0X=~Hlc?H%4i3~cb z?Ig8WLN`v7JVjRW6d;Jo62u}W-38Uzgp!JvDAzZ=7CHo$_}*v{Uj)uXRz4m~=l4K6 zNXdnkIc5~-^{v1ZwsOg>4vlU*vrhc>2Ok~2QX2l?eqJcf-YV0eI&TgN6IcBuL`kjq zlzKYVmm(m%V`p?hdRGg|-OW(ANl1~2@A7!{UNkJj1{e(jOsJP*yB zj&Xi#NXs5G^ZTZ4in z*m&&2MImN7Ap=zo8kyrWmeG$zU20=bl?qYrB9`mwu^>-H=J|d;5js4gln9T3NG!{x z!&H0*KuLUG>KYEdMWf4%5RaC&e{Ao4@f-bu|Cs}D(pogKr|;HWDV4 zT6V9M_=Dw6R&eZk7FJyo3*Z0|-<0M@vc2PM8uj=_u8*+lm9*OGv`SL41YSM7r^(Ks zdc0$(${&ZxH1${%4Om8p=X>M!`Z{foD6JLl;HpPfNsJ-%Eyz$L(ALay(k zd>b0~WR6gH48b4(3%bLKg(=%^?IC{HrMJ%LhiN++7jf)E5=!tYtOcwhb#I0;yeY$V z({da&CWhrL#UD-B-jJ?U%N~(J9(shT7Xog}6Hf+yI`W;T@0w7bu8C(E22b1;BQuOp zmxS4JkeUZuo|~w>L*%4e1-;BDqPpbg)Cb=L2)sBExV0ZlTh&#-o;~)v zHqsJx?Q_;EJ;Vr?CYs3DtX0EjY8F|kH{P);NPR~}v8qYM3EQ$WDTzNS6_EQR+L~t` zvMUD{OlH!}fMDvDcWq`wnopV8cOYLyuQ*E>8G^9f>xVHHUxK9Cg>8+dgF-w`XyaVh z-(C!yw}g1JB!?v3jBE!7K)QeaCbCh;LfSIoi$~Hgmn~eS)^4Sppp&m@7&O5SzXM&~ zr-jILgIIM90q0PXP4x_|uQKs%n!aOP{0$pa?{@$#aI+TUm64z9J+4<44o2$iE?&h| z15=U?7F%dJ48ay}HkM4>W)b_@T9Rg<^hLqNgXwD*N{_bX_30ymrsY8^$F!}4&JpIw zfIEn4TZ{$OrhUx%*vaSVLX;U1cQHcS^7zk_Ribg8vR@B~KQXJi$29_DIx>hpPhAb{ zsw%;G!WiwtQ@syOCk$%tXwFhlr$n!po*RP!Q{I0um83Nvf#sHERh0S^Xe(2XVGbg9 zH;wPF!M-q+b$0!Drli_$M`}V#5zx%a%YAOWrHvPslGwdX=iSub{5raWW8wR46NAt| zdh1IOLEK4+>;m}|+~Lt2cb~pUDzBl7+}B^mD$kEH z!LgD{ihqbE4&tGWy(luX@lcd~A;SwKUI^e#;T{FY@MLo&0vBZsE2RL=rFqlLwmDlD zL85Naa?bGk8JQS%Ve|-w1?O&GA7PLbaV8DiAbLpa7iK;ygONc*YIMxQdEM{{e44O) zH7AE=SzJ%U2^V}d2!0V_`R5pvEiSjS-d=E&4fh6jd!TcEl+P~mM_ znoc@h6#IoBVSg?%K%y>YziqEkk{(weRTb^N!bXIR?&a&YKM3;5nOR;CfheE6Qpy|N znfZ`~M6X+dI2QhR94XBMp#vJoBOL4~iZ%8%yyiitbsj(ElDHtG1_k{ypJ9Ok#C8J3 zb$Q$MJ~=yYjRY_J*-GqHZr~{ILLX#+es|qzU9ekO{3uVyN*Sl}ucGQ9d{<-?@;uw1 z*V$5xIh-Tu&trma@D|I9eIlm&f(Ww_o9_Lrvt98`g8>W2M!Tr_^HulZ1m8`k*$LaT zaQrbzBQ-%p2GLqQI4_qs83(A0x?!vWtgd#|qmoyDK`O2Bzu$i8CaI{!&oX0HHDL&^ z-QsLWD|OTgv0F~od&iR5M39>Ny7LJ&)T-oM?7B8xz`XD2SI|@ZHQwWuqKH0wwk_!( z#fV&+BuB?v?47whEJm4qI<2Kf9E)*7)J@17Y!{CSb`Bm6`)%QbyaZK7o@EYER>=cR@eg~%t49=Tbb`>QzOAtupU8y4R*L`{CR}M`@59cHeyfhq>`oQgc zvN;CA%QDO_8kxL=yu%U{<9-P%I5>f?VW}6Vl)+Xut+JQX!`JU%Z0hLa-)NZ`s$Lxh zuS4ch3XuyY`f0UZf8&|Dvs)EVb^J^dtcF3V>7#7Tl|N+*;f6P#9X5#7l+V2M*7gx_ zVX;WXHMUZq*3vHYQ#Axq+{s!IqX5B>J{zY z#*$Pmk{v_YI>liq<9ah=@-sr(u1BOa;KX+m!E!9DXk0MWsyJD>vn*COQuv6?XlIAz zv+mm!C()>t9LB5g`R`pdM!yi*1|1#_SOrUaX5isco4|gVxq=p`6ZPBH)~RI}bY0vN zI6BpHFq_gq^zrDz(Kc=a|Bv6ahy9wBz4g*yGaQ_DY+qE}C7IkEPW-PnhSfz7yE@qM zqQOgSz0_M=9wXW&vk{dgqG9d6$}4{oGI)@M*2|1)=bv_$8*nWsO&#vMmM})&i6Ri&xNp!$qx!3Za8EepBDCj$1$1wYNRJG)`))r)KI; zFy_j%6091To+;KkFw^O%QunH{_zGce76MUmRBK+2eUnQx*NTS=W4tj+3U9!|Dc-XT z(s6%Jzw$uUZMa`1``MICownYtaM1M7+PSGu9v~Dlt7Fr@>TethnYdJlBzTOwU;QsInvdxXz2q`W^!6zt5=9GXg^~MQ>c&;tcc>B6xIVcWGQ1BmmvqMqP%}fO0 z_-TzjD20s@p|3lWM^Y?A6uJcH+%~Uq$6t~2Oeka3%@)FOO}ZR>k;~P6f6i-yXwE?n zEZ&a;)8Tv^;w&+rnAg`L@Uwt|tf7hdVRukc(V*?BSP7pa9$dW^*wGj4zxc zj0hz=Mq3Uu$tu(vIR>keezIsH=u-Q7=tJ6T7p2Ww$ zAm4(AgUmz5+yWvA2kSov>O(49%*Vu$n@7|?Vue$`D7=#lSzKR?O8o66;=vBB@*B$@ zrPd>No`Gz{;bMQjA}+#4OD5ZdXKA#EfDs;slkK->VT^f%sIM54;raR_I0@^M@CEx# ziZC|wu7nXs9oZG;4Jr?*(XJ;b0ws~pCkb5J!|K4Tcph}P1*ztVA&kqAVQ24G!^murMLXEnT_?N=Q@0>MfTh zIL3$h-`HsDwA>%b$5}OQ7mB-T1s<;99RdwnZq!OW0jXgJ25R_&Y6hhTC9bN0X9EY`PW>%{2f(=<7xF~VuA5Q)%? ztj+1f_$=&+OkbwJD>L9LOrUJrPCF3akV%X}A=8mqae5Na;*_Gmyk|6zdsU%xp@kvR z1>LKxZzdrkdG-6Qm?Vsb_Zo7})8l^gnuHCzRg%&+Mwnf=AVB!#1^8`}t_IcnM|a-a zo8P(^5KM$^xCI}-@Nk|l$ZEbO9i&YcD?sGQj!UeP-C;DVTOMgqwgXWJ3AZ;kgZQ)R z^5B#h2Pl`nop!T~0{HB((@dtpEU3pN+K55vM3;YEsJ0tkR$3?!;6BP^D}1|8Rvm3Y zwQe$XCSc73;z?*SvBAB#v>>CTn&%ll_PH90DgkwoS-OEU$A3A1$$zSe6$~WjpV)k2!&JKVm|c)xLI^%O6>3 z>_q0M*%#$wuPJIm`G6^5ZG(qj%60@b3_CMj00;y}xC!|>IuWfS***s&$6>AV{ji9% zbQbQ@p~QJu%ON_B3xG(d9VIvH1S}sso-@SrbT4CQNRNhBd^v{weXY9g?N6ec9fOuL z$2K|&>wkf8c5R@3Dl@ZjMRsQ2kQwgJp12_5LTGJslO0%G6JiKdS*MIXYE*#G&wgYs ztZTwogVgH2L}t6Fhy}m)$q^e6=1zx$tvCA_g!LxbeC`1hSi0)P=g(x< zG3Jw?nplRVrJ1Q8dc{_+s5K?49MX;&BiiYmg3c}89^~MZdS{fsru5A-H+?dr+aNd_ zp)if0M0b#PcsF}~Q&$^22d2m|MWC3hTTjMiu}#QMFoW@XHs1yOZ4-q&uJ`XzJ`NNx zc*;G8hD?e~-|({2kjmoUcQ>4iQ3vlmV*A}32-A&+>$r%}6yy6pze;v7i^R|yX{;EQ z-+Fu=vi2pz7!X6f{h4W_Ke#QudRmdL`>5Iz&=7XB$Z|k~&qqhy>V)pO^@CIo;?#7J zNi`oDUep!T#Lg=;y-N)WK3-btgoGB%e25Eav>19@pF*8kAdTZ8=!KB zCOxQ;Wx_3*!r0lBI|5#bNfR7B3A0nr&$>S&FuXJLYj_tBKafiRgvpCa;2})NqAxLL z^}r85{B@ni$5ey*-72i`8bX2WbhhK+$FgDy)7vBU+Y=wsA7PTU?k@gR21$ERrNmv0 zubl~!m&_%8FD)3TxIEt64LSAlS(r6ha&Pf6t^K6E4>_{bcu@#5!u>!) zxGvKay^_U1W_N!3_#(H90KPCWzlbrv)f!|f#{KHm5e#bzCe$4H0^OoTNDZGedkvMK zn}3BUHs#Qx<(+^^y67iBMs9%R^W5KW6efarlhl$e8}v|bd)rR{enrCZqc@q`{cw4X zF|R3C%Be;wXoy?3#;Y~U7afT%j%*!;TkV3r+#ohF91)*#lg1*lA(JrnhgTbFoWSOl zh5~Ch@wzw_=-by5^pSJDbjbA|*hXI=RoS`4LYebIJK;bgx7XJ@s#9Ip(GG`5LY2?`mkB3oAq%{(21hlHblrGSlZWyMO=q#tP*BdX1f zR#txf-C5}8*9KwSHDhl5DnNerO*DGRPXZ`WV4hqvOv(z&nMvnl-nwo;e)7bLDr|5Z zvo%JnB6(byJEKs@GJnf_;!dOp<90u+67RTVDQ4$*vWBFm>7=71 zniH$AYxM)W)b=b$1CL&lA(m3jtgA|)**i}`_>lmX#LvpH6CBlMW41DY0#-HdG+Lff zl<^^_)e70;tfyE%LDUqDjqx4d`O32gop0%$|e=&Zb}vzb>|}& z-m1b5lX4;g3RMLW%6C_F>Nc2L+>(zkef?j$+Ljc8%L!YtAK6Rb5-At&TVwMf&~HS! zpsyn07+#XY8G@SL`6B{K&(Y?=589!a+n?IM;*l!d&he4c%?a_XF!|#lW{hwDUiJ0< zG;0{J#;H_fE>o^K;LZHdu6Oe05u~JW3M(OvANqK z5wHejnoVIK_cwpBHsD0mAy>$83?S_PAj6kng6WkxpMPt#Gj8h;eD+ zvvb+1BDi_Xz?j9K`a~q`%y!5TR`T}CmrWS`j=iC8@cHT`@y|{bzk1h* zxmZ0wH*@SHWaVkIPF~?9Vy=Q~DOJEZ^Kn9r(c^nXE=#mn_ZERU%iQs3aZxpf9Kj)i zNSf6s+OJtJJ`8X)$d|x2r@dfv-WL%?mNI{RfjVD9x1mZG1XG%n2-e=SI*^t1O20(H z;!GSzC<-&hTlq!)E^X*+oQ%}Bn|;QB;fZ{g>Pu>x8yzSncZ@XU-KPQj@>s`Z4MVh! zjmcHP#Om|}NzX(|sI39o+<1j+XgfF)F1d&UhNooAdQDI8zzv?(q6(;xA$Q2P`fr8T zI8S_44YKGLwMVs)bqpBwJ`nDZ0sWm6ATa(MV)p`scmya?_d`um^Aw9U)lw9pcjs*+ z(T6p{$;JlDy1JF|Fwa*rW9JX+Rw)>GIlkN)F}tE> z*a1n{qE)P7$pN8akb++j$$a%umt^lbn2a9%G&631yOyA+oY(0cnP09dvkMaX2zG^g=HFCU<}J;DRi%9IOEI`=(EZg|GK~UCy4(*?pzWh>8Nb>!gFq@nWt7tJbkGud?p~ z8j~3${!lxeyx&l1f7Oe7nw@>D#A0xToMz?DIk*@sb{rUbRH5BDSUSMCwAkHUgyNR; zg79m3-()~;lu1Z_7;5bRi4~fNeW-YB#y=jQkw>W+TClkcURt7aZ`C@0cC1(0yv}v7 zWP7?CDMZ4t1dX?M>fT$X^LDquq#TGHYKJzN0v=PJVHqdwl#L=yBe~!;xrU`=Ze1(5 z4JRW$8Vu5q(>4v~?ti#aXPlLvZ=;vWg=|(<{xxGGRmXq8$j0g|XC=ufa}FCpX}ff~ zS712|UbQ9N0n(`PjkghiIBF)lQt$@4&R`bFJ_Mo@Hqnb^=&E&>0%pME^LRGSyMx0~ z*O3?9Q$|VoI~LC5?qzou?3>a0!E4O%{lPCs$w=<GDFBLp{QLR}aZM97| z`xY5~aqp#_-(_{77*(?zk(JzK(GleebdaE1l->wg-nByMhk{mew|c)1H9HM^H)8;oXS`vRxOtb09$OCasYkct{9#mWICE*oO1LeuQZ^+eiKt7XO>jCL7R z0%}&TW{vEuWe4|r3Y8Hrkxe)kUZmKPK(zo)$G>OjS#f^DQ6hfqWMr%n27KaIxPylH zjjCh9JU6ly{I-3>lBJAh>5GjPIECq2GSGDMVr-{Bu1*bU^Hf1?2dR1g#&$!2b^PM= zdD9Stl$7bTm7vZ8HzqgnIjcOs?wBV4u6Xhu5cS5K@MUWwgm)R=G;9s)PMFFd40 zA94OlDWqLZ9bI4geTorP&e<0rS5f<^Y@a`Lz+j@u7n)waB3eCESU6cubOoZona`fI zna#;opRlkuPC-Z*tsvlsAs=47lfXsFzCa!~z=G0u#F2&GCE#8Ku!Z2F1o&D1@^gG5bpE& zQ#WQRurEqkwXGOQ5JSqs3%t~qK zG@U36N}F?eD=I8lgRahtvmp*q)$-w|KIKwzZ`;`U{5)g}@{)eWocl~Y&``XEhYZeg zuLWF@s3dGO@0OSb=^1JyH-R~(HBIPqT?Z7TNz8*BG>ey!Nq6^TW3vL1>k}Pb8|Nk7 zmHFDi;*m~|ZN{EhkG!JM^i5o^U(6Po)-hM!DGW!tsR74xU)j_==Q!`jSfb{Nt8*?a z;I*_$hO%9yqJJrg$d-gd*H%Nw%pv{Kf%5xaviC)>;CG|duqP((ff9eVDFV${F6&vLh zX_9d?X*&EAy*5_dn}84YV@wd=;kH$8faewSo#gG*x;*yCJWtZ>X@ye%4EAQ7#{zbY zwm0f@`?{(DSD4z=`t8iEuzk5 z;Y%5~;7+ID9>(~Ys91D8U?;?8;jYfQgdTa!NP|=;mC%Ir0!AodseeFtRt`_Ht9t9R zc%PEsuoj<$EXvupaR|o>Q$OZ6`)uCTb6qlv>>jO?ee9?+DQ9xjUvQYk9x-MH%WwNi zDg4-8H|#4>>71W1Ig)Z}!M+vU6Fnll0a$d<`C6~@BTnpjNKAPb`@lF7wzhby`=q6j z^{Vh9E&?71Ez4`oi1oQ95?m7(GFwHPb9(Ai($J2&$`Echa*EHxREOHJSV)Grv}37_(FBCyCUw`mJs(=k}MZk0%Qn%i!!w#BZaGHo-~_aMqn3 zIr?qU-kkkYHcm8Rae^RYBT6aCW$~!51EwZ7j)VQAVs0M9O8ZtO_BAfBI@|@lnL1?< zz1!Fv3iYe`sr5l`Dc7~=Ui;76OSAF_p290yQkpe2i0yr_*(cy5qjwR)f@950f5g~x zo3LQ7F*N!mP|&TG)K_=5x%QgIpn7V~^hAscO1xLb)CCRxh-GFcQMW}`-5bg`i{`fN z<4-~vaUB6^hE#9f)f{=>DI+$HFGT&-O33o?!+nTMo3JcE&NkV)xlm@3ulg`i17pX(XK-|wyzAPBXuTgWJ39ZxBUYGbA8 z!qM>M6tOaz9&1lI$RvJQiwp_gDL0{GzBxMG-*Jqnno-Qu4 zhFh0!)>)zv{sk5RK`7rIgqz_e^MSUTOUUC_sq;Dq{HV#GW?wZd&Jf+BlwW>Q9U9JN zd(s(B{^+2TEo=Z&@vtsXG+)E9=qUAF+qA%%flSWsSZ;NGo^) zl%%zsxf61Fy!6SZIZVx+X&h33S89Y2Z6cKJawSh)aUuumyKqpTMbM+>v-Ca<$>44H z?44tW+91i>VE;VYQ5%DG+Z6(SGY5jd*ULjSRwA#Q+*28-`Z~3t59b|0xh*^s{lx)e zm|KQJAOjjF+y)ad8Md$CjyNZQ=sBr%;Fy46g<~D=^;Zw!WbHf`*v0SfCV@sK%+hev z(LY9}!fwfnu@3vAp^S8b7VlB=tFpm`_#3zme-Oo``D4FteHrBW&Ad#~5dTI`OBMpg zgu$z0gtXvf)$Vy^fKpxgr`_*3Yfe4;i!SRS}$@I zm+#8z5fyoHVLsTXT4<76N5*PK5U*VED}32gVbAFDEg@xDmY`rlec^rHK!>>;k^a8H z-q;y)27*(?UNTsu+CH*+Yq^@`3Un+aG%{2YeQ^m_ODZ?qS)B(EQkd(1CPGI4(ha;hM#jbQCg~S47nW|% zVAth8ToGDp6C^I2zH`{2(O|0$Lyc%R-}FH3d;dL~DdYSx*jf&n$lzHVE7h`q@CvRx zYtL$v5KBNk{IruMT`0qxgm!2-@UR9L8b++VUjEv{9cMiBB%&DS*8 zr>G*W2%{9y-&saP(NU-m7S;XEW|%LpyP5r>f>3)1`B-xPzOz zL;L(0$t`|K0%geDPA$a7s7+3N)jF04yy1yY;Z6%2U0hRy=;FO{((>U)s`&$%B=KeG z??m>magQd{^NO3cUkc{wH~(K7*R)Ux!)P;iwr$&-YO-xNnL8VkZQHgz*=}cRs>#MV zH{ad=4(nohTLmkcnpF=W)E(Cu{7_lL-or(^yoV~X4-$_*$=1$7`)t; zg>_rw-(Ll-g6XcXvzk=-FXHEf!AA_n&f(U?3xV6oC;Gy&EUd`@`gSzoeoy<~)2#wW zG`+9O44Hnrt^PxQ_$=Wk;tt&rs)E!@g2cC0IGJ$ zr$*~uZOkJ4IfXX3HgKyPfaV`x=(9Ig_F~PZt4ZTXaoidPtCsyetPeyOk76JAVfdY; zu%IfQ4FgbXoO`}Kytc1AY-Y!z!ZLcQ&~|D9YU&B z(}zuqC_#_ub`$O`#8Z?e9iXXBA7)hR(l||BWwR&$D?n_adz?m*5}!J6h2)-jQ4~h9 zJG87JV{3d73&RZq93m^5p%n~ex54dzyiMFb>d~XL{jLE|89Bbw;OdXyOwFTd9BLuj zb9^l7i(!W`&}tV}O;pE;oP+&|;MNjOTuNm13?ibRFbxygDX!TH84$ z`Z>DgfhB#nAs_?CdD=2NlayS#t&PvI!vlxMhR$9A=X$4P=?)=Fxr*OGfBXcX`w2IJz*^2d_+d!p_@~?G1KC2|!l5NyQzC;x z!B|S4{|6*eclB2b%>&>DYXiiEfid0oxAbWF#3Ld!uBa(Qy_-pVXul3D`L_?^vX%v* zYxwrbZFR;e1&Xa7MB=hypjYzp2&QaU<>>3@FWizu!+J~|=C7TzEcs6FLmDz=%9;TT z`M$(YcS7otxGh7C|KS2xs1F-(c;Xa*U74DYTX6Uv`~(iAG8}o$z&I_;<#fW?quFbg zYcA@$f@z1r1*SroUG=1nQf5_r*J1_wvGz%XGQ=(lej9WLYaOcKawGdvY8eONGU-|f zbJjeE{6i?;NBQ@u&+6?P&Pwt>dL!8zjIXYE_m_mOk@pj&pd}ZBjSJ3f3w>!VNl{U@r)`DGohnyI zY0mv&1A8fLPu-{^mPs3lXL$ z$`hDuEk6-;A?3TkyoisJ-W96VVrY?2hX8$o{6wyljh8C}iYs!Vnxe8iAD)jB4Yt9` zvnzx;2KqUG!daNF)|7S!#(!Q+SsuanTxZT^riI@Jte*9#faxn#FU;t7Y%zfU-9p zWX-D1y-HNHF_9F+^>3#cNPoPWf=*fbL@P!fiPsM*Mw_{0UY}DJHB?ba(%sMWz zoq}9H%e%GdJXA!#;D2)7$&0gCt||Rtg&E6Ro-i9Lpbrq*=iq&0VK|DRQX0fp+Rz_> z(Hv-zIhS|KUdK4esc2tKE|65)^-c60-g_{*N-6L026FB#8p%z*5x)}gU-{wI) z73PJ@X1X^;pwy4jBhOlbD<>&H1`Du#e;?GCHcL6B8qB>x@oSL@+WSJA5!Y0^3%_p$ z87{qd9GeYIn;&d&vsDLX?Y+wtLdtVKtkIsIkz!ycdGXOr3yCG;l;2Sj5#e~s*KmK= zl&B1lSr=BF2F}EJp53x)>7ZV+D=qG(+kKGT&{5g7K**4$?hW})79%6Ii9p;E4a`QR zbQ($o3X~kfAIN$2p_ma$dpZT0?5+e0;K7Qgap!{LB z|G|5$oeGG%o*)d;C9yaUDc7%+Wp2E=^~Vs2Jx(+JYI zEVgsUzdEZ*r)D0&;W8e7bfNjNM?Yp@>CPXnzGKKB1H-i?fiqZ8urJAOzJ5@5svLZG#A}3YJ*^|?GcKGxKzcUA<_XLrnq4ontSV9ayk4eaDv~HKdxJ(V9ICfLC8k7u(ka4FjPTjLFFk&p8l7T(->)dZPq=wcg3o?ZBS(!5*ptoR`N;bKQ)2gk&1CKT zsRV(duz=ZMnIob>`w|hagIpG&>+?9ddM(onTzWvK6ucs=}FAITub zxV+Oc@+UK}wj4d0Y0i&5;6iKOPqN38_0!*ClhR?Aiea>EMozG|1G{uhED<^p!!Xcs zYn)`SYrLCleQ5Um&6CR;V#wAb;*+Ct7Ec{?15xr^H?>gQYJ+tA)prFFTjW_t;F_3& z@PT()fg{PDU1#wQjNMa=C_t1h>UQ^T+qP}nwr$(CZQHhO+tzN|ynD{ff6^8JI)gMWZlx^>47rb{zyjrY2i0&rXAqb8#PC1?F)Gw$tlt>%{duRWz3 z-mkKD$>$-hr~YIEb`daMeSxA_tDCu8<}OOMWG}+);8le#AubhglUyR$Ns{0;CziCA zGhi;FgngC}7R%A#pXhDE#41vOExpNw-qJB3QrYt?3}mrnYwjQRFc z`Vjp2*>%6GL3U|O3o=BhF?5~XJaT9tTnyo#eGuM~<+oet+XzW8=fszzE<)ohv_viG z|L!d_%?-H(v-tC&ln01O#&6E*vJAGvOuL63DEs;qiBBH|?=;NN zu}CvscW=+n{eWR-u3`Z0OW$#fBhf)9a!RP5!+LMlBQ?BkMxb!lx=kVh>#R9JN&R~G zQg1u#EM@IRP@sK$#mTgwq&A(re@y5NIU=I9t0&T`-IJhZunR~-f!P^ldngAu!zN`+*4h#JyC+Umpjbw1i^i_e` z$zyAI@uJCIK ziBEWwx%vcp(Ev5=M8sGprUj*MwEK8oAr6DzFwvlcD|@m+OmS?A4=qTx@NW%nb7})v z#;f#0A48z1&idPItVQL?mE9sWIHQAnUBm3lsZ#uxB&V#jAm&^iUW>kRvnec3Px0NG z^o^u{0=XDvn%H-4yH=0*|Ff)%|12lsH5+6s;+XxJcq2B?05-_TVLZ1dzSxx z{(c+$REo)R^&1(cE+kBm)6-Xx2v7do#Q;K1xWrFY3-2)OZ;||_gl~L|(sTx0+~l*D z#Y1n}Rot=CR#i2lZQEOtDX+uiw?E)_sbg1?07r6tM4h27FQ7F6CtOI09ZFe?)c{d6 z(S$E&=$Zm6wv9h7>^7P7I)pr7T5XCyiWCBr2@;i%5s^d6Gw!)AvQL;W?-2l4;Rfmc z&%XVgHVB8|)M2`&?I3jx6y#Bm(#(QjuS>Z@3aVd4QZcD=hhE=l+2iy}Yb_wjL}Cgb zSeP=E##+QOe0?`zbRf@`o6l~UzdY8!Ip}ZRl192yjGi`FCaR^0exR~Z*jGIJm_i*{ z{iJ$OP;)x{a8*79q0BbK&YKW97MblO-V-6*CCiLAJ% ztYw-KMIPTlS0DV!JRCv_aL)o9*6bW41-`U^)YBht_A+G#$92=_Npz%0YWS1yqku5L zoIaYn9ij^-b$Ep;U>v_4*`#GwfPFPKJ$jWdIXMD?VqI5)ZXY7T1SRPemugB6D2-jBsFG3N*wdy zP#&vvyGJ61q|JsFyANpztFmWOwAaNEk$?*Ds~b1n&)Qdb1=rwWs#%oCoy#^_9P3iQ zt~z*UqzFr6Qr1N1C6>qLa!~j+Wz@Kn&u(h#+Q54hH6LGE_L@?Xy6sP|6+=a=K2Q(p z$E0qX4fEFP1011LX_;OKx#@@(QSxv!Xi|q=Be9$9ZcV;#xD5sfvAT|_2$f)j>4Hch zwgEs=(P9Lp7J-~gP+desMHEUJmK_-=IEWMem&0)uw}_oD;MfDYtNDH!jAlk$x%Gt_ zx7mi|PPG!-(~1Y?DTLcl(+PRaPzCd)3WF}LY16I zXyGX|gP+XYBCgRlPqJE?&mOGZrA)dxvwjKp6fX^Ms%!Ch;v62*wC&ziHab`%61EvF zl)T(@Pwr|W z`*WG^XrD2Jiqkk6rdIV?W>O0>m*3JD=lq_W3?w(XwG7cXdzV;?@5&vH%1N6mQcSY` z6XCE+-M}L-fAeJ+mO-a=kFX`QYLW!c= zoIbel&9)28LA}9LZOwXblwKbClZS%sVP9>;fY#*3N33SFgVLEtz0o2@DGfcZ&@Ld6 zSSFE!sYil!(SrW7E%n9Ht`)F-G14-_U~#X|JyCXWXZ|?znsryi=FwYaNGkDU1^$Qk z#AiwC2?^tzQyU(f>yJ&cM=!B|L%T?0vzZ$&Jf9uhI<~s`65nS|zp+;-T4KG-Tc+yb zml>eg27Eq*>@cD~G$V$hRIa4oWIY~X7+2(%JOJ8jR5b@2Kdvx~#q=r{)JQ}i^=7Re zz(m8c@Mqw?+`~`*aMRw5asHw03YB6gXi6@mMYRzyAWscb@8yFgoWpha8PT?4WsHCv znaC;XGAO-Vd>Q7$a%Ov-<9uYP%2&55?kMmkm^whni%4aH*5M=sQS#*?O1t+nXUOZg z+2CWJGY(0TWkmOTC>o;(;qsWJ?4?c2>oiNMamQ1lMdAH+4k2nL?GlijADe)*t$$F? z?Ar=RxQ!)*AgW3YD0Z%6bHyFrzThH?)yiW5s;-%bV)-?-Hh_r@8i-)=pa5*#8 zN^_&h0 z%q0y2%x2Ccm`+lcjt(7BS+n-STqf-SO^hV{r7 z90nm^xL3*Yn5B+8f;vm5Zy~b+55TTt9u}6?X2rGQdG9zv0>XM`U$gCz=kzV|mjgo~ zFW6CYyz0CnOX{Wm1oYE~q`_)Eo5*Km*6cnoU6Kt#DQj}B2pkg)JE2KyGyJ>-o;?iP zQK@(SvXbS6v44{N;#c+?XGl({hyjx1iQB*?Uwqd+AF%Ii`aFxz=^jy*-1|)=V(!YH z5yEs%_Vi%FPEd8DI~wp{g*SfvSsHSx_zN(#P%!5_!N&BcH$jJdGqH<}+83?tCC9~BZtGrG9v zfKY#q%`>#f8bPwqaH|g0$^6gl;qEGb2KBqZ8nTdhbMR3BK+Al|F>pVtA6vy{j%Jm5 z>_)zf>?AY~@g5CJR*+Fs7 z*~JfIm*Ri}{aNvc?i#eQV?&F6&c*~US)BQGGa*qUgj`z7*y7|3JkqN{@+oz@`7Qjs z8~;pNcN-4mIuCft$2wFF_^2Br)(1$-@VIR3L!;{Zl3A0J&LEuO%7Iy_pXvyzCuS7! zmD(c>w@-F}K7b_Q&^oz|o)=B}#)leSFi%U~sJ#UH)<-|mxRzS#cf{lu8mdEEf(!;l z4ZDh0EL}Hiq^#PbmdZt!GfeOtlu1F+b_QE~6BOzxYmb4o01QzxcIN*Qdz@}`gP_J*Z zmJgEqxrvDyAKwMXA@pX1kls)%DaIJ*J}~4mX8R&%>FsFtl)}XOCVZ1=J5RlLNQ)%_ z(h=8ycFy(rWXzP;*SjecebE+z586>Xa7+nzFogTxW@l}D0r4`#H^f4(RYRT%N5V$5 zMRBz{@Q#z&RVbV;jkt$4ZBx_AS8q?htPg>DnWuGR-7^31XD_xuFzl9LT+s zkk7)a9Hfh&$P0y-WrF8E_b#wDgb&v7a614gsq!0q;v4F?4sf^MwP}98mE#F_w6UPKW|3}JS?Z@Zt(6_<2gM|YfS!atu~|csx-3r_y_PpsbSO3 zUqrP@o~sTWBMYv5XYFBqXv`}+_0tR7%8)(!g$I`kMq^Ycv$Vu*B%xIQ&d+A7iR{KJ zHIn}7SzXMCn-{pK8bh1zU$_yRW)6hjdfz(^(j`z00vVtP*CXUm7fhJ^3m1xnPv-IG zNqr5-7Q?v(EnY7+E|nCUGos`w8sK9B*{Mh5D;aRP3p;^6Np$>jVV|d8$vQ^ce7y+A zV=Va9;LusiP6n&rphvjyu4ueu0VP?M;|jlw-YUG>gGj%*E!5u(w?VR8GpwP*kd#mn zU8`ak^g$>;rbC8TBnF!kJ99I^w<<2eG3`l*IEo0 z_%_>!ZKRl}pWZ!uFU%sXz`HUmSLoabGgffe*C~u&UH6<^h^tt^{2cW{=>q*erL*WP z<#g#hZ`QcGm#Gdp5uIR6+-2P^#Du&|eGnbC_*)glk-kaCs^f1q8ViPk%d%X(URh^U zcrst$*hIsP0^qNc!!_2>GgR)mtB#;znfV#_Js*pab@hmg(SKdIICw$)bJ1pdcf|*S0{4xQg;Pmtn`A9(R9r3(URTsw zfWn_!pst%$E_-QT^X;zp259njFq9nqB#8qGCR&y5j&Kge%ZM`1+sFI@ARt>WV zLy-BvSwEH&l7G#5X}<5gJ^FzJ_>!;+)3A$kf4{O?_<#mDvh>5V?`6dS{vz}FsC={x zT*+FqTYxJ5)teB5F_t|sSSEluHQb?3mP7T+0>-=Qeq;wL1VwHPJhDE=UO!>ibN*%R zUs8*B=7Nd9WeIlwhP%~=-)fU;07O)Po?7e1QvbWTJGk^i!^EdzVcN5Mwojm-3Yz*!%6tvARp{SL*$S`QW^47flhpG!tudhmK$=T*ps9n-) zhyFK^?$#IFOF;c^zi2J zvH^lz8eph_KRPfdxTR75#6!2Vh$rDQD1oO^Les^STl5tm#V0*--U%kx%E}LvlLs6k zVJB7N3+j_o*M>WioIbST`(MD^Vmht=3B_moA5eTodU}@sR`40|8JO9bnE!jpgirt9 zQ+x(idRF%TOBBDQ)5*DbtCI)2$;eH)f%D%NuIxywjoWsE%~npak=DP_bfo8b>ovFW zcl&nFZkc>K>@nw33rk2#6iidXOvnn# z2mlEO7#Ip0Bcq;Zetu(cR&{tDV;of9jTL+xi_5Q=6DS=Wo&8`O;E-P{|0rw;ME@8t z9+VTS(tRBm128$%4-k}#%uJI=s0@y6esy{fRs1=D(AwgwoN$mn3_SlZyoT-eb19g!Q^9{jU!`qv!v&D{+|6ZjW#bY%5dNY>10e-HGB8j^o> z5Z(L|;H#|kdG7##4F8UAGaZZbJB6w!A0N1$p}F1# zY!h=6(+`9bi?fRz5c;RlhdY22*$)W>C|Cy@=l3c8TQBXXZ}W@J<)4aA_1`K!%73c( zVXVI%8>6nRvZ%7NvVLkHK+-hT0eDwUivuy_Y3|b!m00)5I&B*D4@A@Y*mCtmP?-UNhyZh?; z3a~VERhAzQN@;%#D?2TzfC_&9>I~@V{-f-O4*{EqtABWK{oQDuj}>ZLZ-vB0<{BW| zFN-hztM1ZI)c>I_M)J8qO6JPCh6jL@Uy0q5gX?cxK>pXSS>CrA-G8h2Cw@{Ndk&$j zsH?cY81H%n7yW$36em~MHqbS{Q)UOmrU$Uh3ibA^tiM_teb;`ki2w2$z<|s#G|te| zH`0Gz%6~RgSMeS{9l$icI0k>-Hh!jF88ZKi#fKXD$Kr=!|B8GW!TijF%|lg2OjuSd ze&ucak1W2?8#=^)viM`x|H0yCqW^zb{8uxT4RyYYUxx3H^$mb0#$VAZ0M89Sf>}7@ zH@jiLe6|m^Ly&o_U#Nz9@BBEQ`&)BSFff3g2|oP)Q1RjZmx>Q=3^t$j4eXoD^y_vH zH+YKsa>~EuC+xjW{R7C?uJMb-==<~|Mt>w`*X7qAxBK$zk2iz;3&=;w^>5SJ(HGns zLBtf_zL?&cUq76}_76-3?}z6X=&w@_%1^eNpN&bVAKlmQHpI{1*nru=;w< z_*%obV<`BI`o#|aPba^v(fogR@+C$k+%zNpIr+~hh9o|nNN8HbZei%-dg${Pc>=oT zWJCicms!~QTGaxm(FWhG8>z}HE>?x)^;Tz8mr+sVlJ)FUQpy5(nwDtfkQjY#CB+N~2OG8W0z{hS^B;}eqqo_(YRlLCJ*RSn-* zQd_=`j=zI&_rhoX^a|82$F@*^bBEx$x1pXzp-nT=S(00eufg8*_@CNhB4cuKot_R< z3l$U&4`}C3UqDb7Ptc-fWO{8ElH)&@Dj=Z?HISD- z|9p{mzVh9qaIn-48j`;BqNt%bdRi%&S!qC0G}Nz@n~TS1r`Do*7D%79OYzm|1&%9C zWELCO`U6q?PRI%$4({Uk3b`BRMZevr!95lZ#^Tg8vQDJKN<@jsQ}7ryKBEvihJ@x! zD^vLGp-YhRPmw~}lT-EK}$h!ndkr%67-b*mbGW7r4;4#vY z=QE~89B)z;kc{ls4mqbJJnsSG%Mr9Hmb>YANZkt4s-ZlKP!7t>jX@;XWd zX=C z3ZMw7K~9V_obN@xFj8{M?(&nI`6D-Zpgb_LjT>}cIYOsixczEIjNL(e0$AB|sU;7> zQhD7mQ@v%Fn@3(#W|L4#kl{*-?Cf~-up&KiYAu%)p+MU?t9=+~XWUybpH$$#6x8Z} z3MyqXeM0Fbg3H8>p=4H^%e2`ZcE-oVUH{#c`Q%^!h($UDHB1T;7Xd@0;NBlI=*OnB zYS*Vis}&nd`SGwE1Bp`7%p`hmv(=Q+TRCy{3E4avK?DvDMlRC6!erkPJTM-7UC%=( zaH12Cp3T$xs~f2f`)TtFgm_rK=n-i;$B#yQke*7-vylQ!^K8c`YLX18E4&#h!)&4JJjWLUAJ~oL1XaQ9!E%oY71n39?J#aeGN;H?X$+%K?NtdyGfs<3 zP^$@sHIOWxaR3W0)dYHVXuiz$RVk+#4H#cZEtpjLC8%TH34t7L{Vn{n|e&Y5J0U~~;7ayJ`T znw(<_B0rX1-ff@ItVXPYW&zRoRE0t=na$J~0!jewa@^c~EFsS|dW30(f+9EBlAgK zn|s@fSXsBx)A8obbQ3uBY2H*;&^_4K9Md|hzx`rDl0iO;e@^=OgqZ#pv%WEF0e|obD$wMDz%Ls&TUs_3Cp|2Qwi00`Z z7Kq4lOB_6aRo<{K4{qrt>q@gyB68hzR8ip&S|z3w4ph4Al}X3@IFQ;;Ci-=)nvMt0ai~voQ!(J^xnSZmrZEyTEgU2A)6GpETxfJ~5mN4DZXdwIMoOA7 zuY9(8{Hk%nnPy*@fhaR#wfb_O%WM{4wzf{&V~!=MOnm(cOZuIwPM#h8bvFN$DH^Z+ z_&7^4ATl%gTZjldn@+x`hkm;Fpp5)nGcme;Jvm!RIv=DKq4_;zPe6zfHK4mfps#Yx zZ-7b>hfqrNcsU2k@UyN2Z4=OGsosl1cJhKV^A166r`%zFEXj#yyt?fOj+Jjod_njp z^3)1b6^!~+?RDb3qp0x64P>9i;`9mWuUQ^o&j9Y9$X=o^bY&S&k`8D!jqinVS2TEt zA*$He@TN2d;tHVz%32hZ9h$o{yA3*|a!+;C8(JH=5nZ?IrvBq*d(%T^?NZ@hVR)=F zxPeHDRAwAJ9vcxJL}1u6!}+R>BV zP$I9>A51PmgJU<~z#0#gB~tY-51pqV;}+}aiS;FWLyW_S8zsP3ZKU2HpeiZrDoolz z%WjiUYgy|gSw1(dO|4>w$-NT&U6Szh_k5<7U&pI_}aycZz(mPrhvkrZnxFEf;FBU6_O5wO141=eDFbZc(UW9;Pl5PM} z5bMsIq#J&h+t)0RPB*CR<87QLO^DL_*v+Fam#`rtNJB<+@ z_dq8Ovy*=uV?>X$>)>VnBfP=bQv54 zpv0D1J6=lIYnHH5`kXq=3GxG|znE{P4k}lL5@M+nR@`Nke5FDRTv$cPrd=2dI<{Pz z7-}!uCp^lr)Po9O7w3=xn+awBcuc09;y3CM0F|uPDQfp|^K}>SW(Wgo5 z`>a%td!B0*H4-ZB9-63YITA|okr%^|QSeR>jkqCOniIZ*QI+IL=xFmVerBhZcoUYa z)PWW$gw0cNhs!_P>~WXM(fVm>zVXXSihfJJS+W_HE0LMOS=PhrKLQk!)Q0^Ayyd>B z_18~Xyp>`U%QeSt$|9q71)YG-<`L+}qnQ|Wy;_w5@b5N@)3XXnAfT%FjH{GA z!;iO&wbHUk#LdLK4B6RFl3xoJEr($goDZ4Etwg+X`lh=u=hqhD$0B2hZEIKr zUyI%iPc6pI1Vn_qu^q(@lDnD5HDLb>!&Kg+nVKiIxN(w!8im2@(UbUN9U%?^=R;Xf zb}1lR(_T6#C=N0e`bYR)mriJ}P~ixNr>O!uRNpi1VOFU-k*>48Oo^-HW<5H*b#hqAiP{e|F$4f*Xf-q$&tnG%VfE!g0J4)!n4hM8-F9fovlc!XV;coAj@ ziKGe44DdD}P(feE*&|M;lmKBqA`1ruRF;^%$KKz+N`(&`=n=%`+vKxnb9VNDtvuwp z=Qoer3~u8#RBD;Mtp`g=8rrQKN3V!O)zEIq1VExBdK?9sr@if;)>{3`VwFkM(y!RA zxpx9|xqLh(^-~Q#O{YB#D8v8vET|?yq<@Zj`B}4Rh84w zCc)FaB-aZ_*|gJD)Ij6Jp34G_*=7)v=~Tjl`JQNY+S!&^ACO%gLo~PX5@-*!+n!iS z3pUzhK0@uPA(c%CXEm-X%2n~2P^gu>UF@bZpq6s8i9p^C**xEgNJ*!M1}p6q_waTC z#eTV)pswJd3%DIT7fORb>lzqD(K}U3y=X^&o%y93{0|qvAL4yrc}KF8z#g)$r$yqR zuN5RHj}K#nI4hcu!F&qc=C1!CzosgXeD79-2>0y6TmGmJ6Ly6lunWPIpRb|7h-fco z#6F|GOUsE&(a*9|(Sy4Q|<3bgCt0;L= zn0`LcE!c46W(QN{h@b@EGD%%7jn90p@wno|p}c}i`Lf(iJNVFwnF;xad@?VGl4E(${-s*E zZ*^FCT)lR}7b+S$tpeD@y>V?=o`j7YwOO5cPsoR!p6@UZF3y1QG8LKm5i{de-B z@@Xt zFMsomNu_=PDC~Q_MA)Wz+Y+23oOVPm)G&#u0VE4hKWeuETWkdz1;<+9H50ds5we`Y zE)h;Mc+9s2TL+q~HV{I-6hrtPp|UZHuV<_kO7p?dYbjY+vU5YSG0HXVZ6}U$PI0}@ zN!2gv{`VbmEz4+OoGHZ4p{#0k=01XDQYh3Gk0>8}4Y2V`6rioUr+O+<^UsOS-MNiQ zC%DFL`%_Bfq(oT`Y7KX_S5f=lV}e}z$Yw@SI^G=ffOHcEV5&W$&jqT`ZonACWA90e zyjjVs0aA?QiWZo74!rJ#ZsWsk_>xIor8gD)dPj|%F?lLpxIx-*RCb6-^RuqnG zx_bmTa17zv3UBzq1aC~sF@aFD#~z)6c8Xsp5>&t;{2k73jCW$LdQln&WPdUBwZ>!e;)Lo3VHnnnzl~IpqLl%@6rA>%rl^DXT zKTaL1=^c`1cStF64r8C%0DzG}P|zML0@R^9#rw{niXFdXO`-UB+K_}+Il;>#ryOGFT9E@cJ zDtjuD{l;9>Bb@)$5`ivO;GR(bP`#uqx;NtGYv!0>>)j?*p`+RY77!?`6`+b3kucey zR@&CS2ug+s%gEZ+c9gYkjI>8?R^1g;74%dgc@|@iKK>Hx-9LM zaM`6w*k8stfvZ~YG~%5N>!ZQE3e{v(!n8B7$RPtVMEXqg&MYRTHgQDH$lxc12Rwt9 zDYy%G)OB`@Y?aT6TVJVvnTImetUec1k5Ur6FLf7naxe3mfAsd@uBC4va;A4^snz8k z37>`LcGa|Ax}3p@_40A;tY9n9wx78yqGg&O4uY>OxB^*UBI1cyfO5KrX+-P`%iph1cK=8+uDR_=wnmNLbtmo3M6gX{vnq>m{zAQQ|+WgCf?25VltLu*+nz zOzrz#DCY2S?&5sak3xY?2qfaacRwlq^*JIua|qwOmB-0hONBSO+=JDkew85(&`qZ z@?7zZX1`ug9}nOvsn*26WpHxY`3OM(ILMmnESm%aX!Me5w*c_wG=7=#EZxC+W=Ue$|ctA zfe7OjJeJCd!w75`K!77qris})WgdWxLzt8AUn;zA^S`+_dIWs*CRh#m?dh>k^#Bub z*5@Wpj(cHy!mx-0Jn3r6Gf9a`dSHn0XE=8|yIdnK#9smzjxu};d= zS+mC*iE@p-Nr12L47#or%UlEuXi!@@R~O=8=KarA6&O3Ooy_LPFsyO-MS^0RTzw)IMp! zWdBQx)E@dl}e<{F&D zYHB0;#9C<#ofURT*CK8lw5SLU1z1?=uaY+9ctOIwtT2cb2MC#%PeO9(#9d9!9d##m z1Piieb2r>k?|PvvOnR?*i}KVyX{asMv`V1}e?wm5Pm!LOuR@6YV7#pA+4>dC5ArTzg?tYjd@N;3fS9y8vT)+FS z9J(b)WPl^YLwL6^Wr_)sUb;1BwVHiRT;D&_tAruUf{c|_6xD`-7T(iplEz`*MZvtv z`sKZT zY=zL=P5Q^*6Tk2FM121jCL^Z2c#(JRvB@}WnCS2mCGgGiq`)C1%Ys=`m|@3l#FwE6 zlXqWIe+L93@wu{B)Bh)pX%7Z6g8`X}-MJUXmg*C;-Kk4FZef6=Ahd`_Gy^&pV*$)h z-5J3wq?Cz(!dn~Ag`(;#5L~Skqo%j*R{3q)Yw znQpPVk{X4YY!?l+xfXx+5QF4k>N9r9t8SMu-L9=>b}MwMh@+Nxd}-EEWI(cT2aC~2 zPwD}b^(!M+7Hhe_SPw2e^9cHqG0O|>&!i{36{_P4%~KJQY&y<;N?6@B4{kDiRvum6 zRDufcBfCHac|(8HrCgw+N5N|0W;9WaEX4&;yOulL>h>gu{SxVoUFgM@wT&XGJqN~l z`q=~QmXHB}A9SL@>G8I8O=aTT%L{J`QMKludeFp$E$jIXkAZD<20U4K;#JF)^`CGx z10QUngu0r(!boyQ%6d22wFDSb$*CU!*cf*1)64{lICxkodxjEH?0?G9TIT5hnWTA$2CW>p4WBQGVmC7N~S1OxOqVix&7+Tk(YR@ zd>OgEj&^OBd*4FRmSE{o8;X_Mxq(j&Z~6#Xjv+10mq9;;GDYi5``C#N`4*mSt59N!hN9{ zoma@b$UU}%l?6FGSzR2a5|a)OH`lkukWK??=D{uEF+Rw*pSNOW_OEN|>9z1?hZW2a z_2hVsfJXS>fx8;dbR;!4b1DrI<|^nL)q3@c+PGGFHs7!&BH*>9gF_4Eq*}B3LR*!S z=<-NjXp8RqCtuVV>g3l?1tbsJ!<<&b7yJtFb338)!w~5(WFRIMt|BGn3*P9>BIdFb z_LIVb;;!0l**^_4IJp~3TRrL0be9BB0^Uz};m=yxB%7xPPk-jksMi?gM4A5*@Ss(9 zXosMB1~IajWxJk+a3AxS2k3t`5B$LQOsb!eQ?&5qx^g5W?NsQw2Q6Swa3=pF*HVY` zE#m+{(uy2Rk7$nYGaS0-^8vaP@2C_N0soP6)N1>9 zg1%p;G!s+@Ev%`7C`H~kcu5m>&m@$1)NHvBSn&OArmKnRnkdzclj;Mgm~vHSx_7K0 zKc1nf8kkS?A_qpJ=bg*s^{9qH5L*vZn=S`tv-<_1Dfre%kD-n$?mf1m(3^c*X#Bo$ z8F9)*<137jY;A+loU;#?f|UwLFt{4}SMal*0DXl2twpllR@0<;gXv zB;c2Y-yFSzz*d({gTRwe1N0h|MTCSDUt=~GEo9}TxYH5AZu$A1uu2?H3WnTp>{62c zGo;ZpsyV$Ad55$-Ylm>_66+#J`q({;dQrdX@FW2go(*e)Y0NoNsQ;TQeNBG{SiGU^ zjsT>Y^H>Ak3=hRim6oe(KqQW+oOHB_@T2M_ous5PWVebtTOLN{+aFeLmyZ@$JPb5m=&tc`q!%+DRqlgnBx@Pa()hW-o1b zu2jF6g#|pQNK<&8SNNa^Lo#FBy#T}ZN+m1hvmaG#<+D5FxR?yY4rzg*igp_+yG>^q zilZCcF$0M+?MC~27!8g%u*cyLh7$3n@!CMCqF%+HW|ZKHq-~+2 zpzmh3(hAdPiCA%*={f78gI62U$1t5vM>u0~SUdW{B1^L@p6pAKN}l3%Qs2r*y%}(D z$3(n*Xyvklppow0Rg#*XDQFK1m#S#U*wDX1mkRUp0jW5tI3oG68Bq$e>CeqcP3*YY z39Cd3zF__i&#J4Y>D1&1?zUU>TIp0xWH?W{gv+$0*s-4f1Leqo^Q!+B*KZl^2dHoo zJ-aK30^)g{FI}VtE*Ir?aY`vjw=z=SDyChPZC+UnW;FYqiIjtBA7{|^f;|+Sd2jgM zW$-a(=$BOaWh$S>IbAfg8yky=ZKg20=0kd9o30AIxP>L&sR!qIg4D2EYHLOIzTA=* zs#8YHB5BD*=r=0G>PYZlCB=Z*CV5UXK4DYcGfsxEB)hpey9Skf4MXjwH^W)`;8UWo z0eCh2X$y_-513Wwn^6SPIS+o{6K89AWTe5_H6h9DIzGBzZ|7DXRhRxmvyL$2BN{%W z6Eul>?@4;8hb2rPE2xuGF-8EZWB^qTX8|!F%{B^9l%QPN{mU!%)n;Y#tyYT`FAqIA zmI!Van^n*Gws^!k1Zp@>ivGTTQ|6X{uAf71?z!tXGiT=~Jej?6dFi%-Pz_gQ==}1d zicdlep%Od4Z45I!XFe+g?fIItp}ZCfQ1hMuOT|5npatgn^eaZ5fiQE9%0*(n34QA6 zqeC0Uk-pwh2i{3wv-xx}dyKyqqWX34_&?y8 z!|=~F)fR;#eR{|XumTaV$9o>K8`c=g1$s38RH#d-gpM2Sxn+2du1a4G)13zT}GLT>lnf?hrC zp-z#~=efr=Q;tBWc5~emZDXO4hukZX44etqIS}Rz6!rN*nk(g>U+zOioA-1hEjG%rKybTx6 z|4fso9j@YEnfUQ!t^Iu>=;7dht2hed>Nm1mgUAYWsEABT7r;z@Oor2)m-Q{0(s2<7V!o@D z-PS-)9^q(1cRTp71&U-Y;Kr8K49r@|bf_JSV3lZ(za@{v!`U2Kb9Zx`S7F6Itt2n2 zh0;A}Od|iiIa)m*`l+pA5m(XVroR&gF2U+=MMPSXpDhUNbm>A@yL1d#iBKX0#qcpu zn0o`<3g`{Kfx}$T=cqzm7>moB<2rQw=xJJhWG^~w8;7jI^M`3u^pvV}Kg1+7_hP4X z@+PDt&OPl5G1RzJ(1=QpJ^-z1F&5VOFL;LJkR^1?oJ0aZki$6T_1RXm+9v9gc_k#e zYkt&n*>hK?d#;-*4Rk*Obea#A{*Cua;0YK2R}#XLHSA`uLKU{p_L4*{S#WJGCzma| z2apj*yge~ZI8xu{+VZILA5cqbVmllOt+P@&i!5s1a;)<$L#|N|ao8M?n0l*OF@D6s zqKT?YlrfChbgU^}%6bxVgcfZ665e28+a)KyB-a);@sGaFa(5NP9sXpOO^dNY=%{zm zs_I=E3-21h2&kEd?yau=2zQfL83+483CDU^{0dOMMhB}y;YVD{zV{MwR{D-&6X%2y z3GXV-|wBJwQ3p)UPu`Zh$_j?}n#aqncj^@xwDkKTlN(lTv??(?9VxLC2f_`r zN))hs{U)Pbw1Oq{IK&$G2pYf$4>|0%buGm2)@5F>$_5&R#3k``dT}K?&y!zFO8Y}l z3t_esTP6~W^)dUf2=6URg!M)PhFS_yh?{5aRC_rf;YG_WyWe+eWQY#BH|&RANFihN z$PIUpez`Gc50iPkknslVdkd9F|E_b=ALda*pZhKd$XeR#GX0FibQ%ACjIKou{q2^4 zna7s$!qVAc+J;+rakC5<_~Y%ER9qHp!U+K{D<~@<4#8mkKJDwXJcHs2Rdd^9r6^Mtou4*FO2Er}72&SNsA(_k~V5UuevALI+RNH;gg$DPm!TR!ZU z&fU>BgPZ0zq-d|6!Pvgj77-^M%)yODj(MxKlA#_sd5yMNo!szlwC21SwE)#g5nOJ7 z(y66+eo56|N1Y?El*!#V{$gFm}@vToeQK8isX~>_#+jVl}{3+x`(`H;JUUT-Qm!OYsricyN za9yDS+k`i6G*Uqmly;aSMU%4PJlKngL^O>B7fAN{Fi0P$JZ#KLhnXEhQid$4P>KO>%9ZxL$$@;t;`825$Qy-;u9J+9J0mwv1F9Iu+4@K3D|kQ5CO{Q6tR zi~IjDc8;;V#P7DRc5U0XZMVC&ZQJdy?XGRxwr$%scWt+4=O*Xm-sGR$Gw&wz=9|g9 zSkE(SeO5Zi`H9T|s>kT;H)&=XVr}3c^=N*6_2Oe_xn5IidDteeV7$gxoS|iWWny%z z!K)haTu;+cMYzoUBiUX3AH^AwDq^@R;Z##B>CCv3@JXET21LKt?Un5w+iJ} zX&85&rqzu{1k;8F$dLL0cr->X4Q{!gbqXLo0RBj0lM>qX}e@T626h>GI(v{iI4{)5oAFFv4lFm}P$Ut{-%JD6H6x zfOp^NYHcIzE9+s3ubZM9t_gih-w2Buy|m*dA0U|&FVH%04M+`8*)AwP)=Ip)gy&SB z|8|y%pqdO8QoRhIWV|zspVoc1R*kUSG z2ZbBI3k%O$m4KffaW!EK;VIQzcGtH=T(0R~DSmxwgPspi3@Be=KU?E8Cmt{3$a;&m z@QHchl#NHoX|V%yA9%+|x_^!3^*COdF$3)E(MBY)VMF)sk1SLU^rNkd-yMPxjN(wC zsDD>q2W>PT9LaJy@59D|f5&(lE+sNk_86rcz@dJYPD(l_Rprb475$?qX(8TE(6ti> zI(a&R;P>(j8BGG|qvp7b+9v@P>!u7jDW1dnqzMayxJr*K_RP-F{q%Gi=24F9HU%43 zu7B@Z|GG61<_N!!{O}YO%wuPIhye+skT%R#zYIg>u)Aob(G!!&z@-3FNV2eS&3&aO zF+Xm+dfEid>uWjo2frU*vQuKHjD0?)qa1RYmP9P5m&_4=Z#8?8=v8I%cG@aWqJ5XW zc+hpVFSm~nU1{NOA0x@e0Y4++5_r(UMxZVO4t{E&cs*$5Ut5-5gPNoRH0FyOw^w)M zBYA`V2gNi*@z!4O@4+Ont$&}MihKV#%5E2%ZKqsefbN9aI9juV!M_HCHMQu|}j-O~EBBy#{vfZ!p5w>UJ*Y;XIr z+N_#5Er6m9L(T`XP*rU|ft=sHKQ8D#D|Ysb1>4ptC?VX9Q$^QX7VCk3P<63>*6p`Y z`n-G_T*lXJzsDR#&cPl~VT@&$k@lFjCj@mKFRylTEraCOxfXPelju%Il;(JrqBh5D zRtkmh=)a!;>;4eP(2!=9NQhzj@bEMwROj~f#(c5Eq1Jo(1pWQCjpjN?*hejLNY-5g zJUlZIgb1dmZsg#X=;AW%kXI^{K8#I&H=H53AuHw*ii9SWgk(4pAV#ad>sN`H@X3O?AdaMc*!LJNBgob0NqbtE|Y?O zsVuD^PaE8S9s|r=B4Cck_ z;D|D}M5;g1zE-VT_?G7Atd9q zUm=iWi0?NV?cn+gufTs6Ay^M&1TBh)gh!hQcHfmF8HX}W&7>B~z5!O^+l0o}Mc*E5 za^lbi@@}A7LdQp)9RWDzO-?s6qG4#3jmXm7YFs5hXt`DZ669?Uaq{d_p{_G|C+6g^ z7x7RMabZ5(xcBI^`XtjNd(!knL*kQBKRiv^e9OLj@ne6&kgh0i=EO}j#+*iZ-u1{g z;lXsHb>#>OJ_i^XX>}n(LDdf*2A*ywYxkFe%UxU#g?<-)K|@9rM7wiSf3Y z)|qpal)KOplGnl07&Ymzj2^93Yk9oSS({=p^*cJDfIdj5sOYranFR93@wewjTMB-R zhaOeZ9577bn7S+s(Zpu^J{jtF6+`|j70oIHyRm`qn{;Y-lb(3UDb07~lRkjFx*5C+ zvavTM&S*6Atw*WR>rYYqh3$|`@<)6&v5ltK&Pz+FjL3w8Lg1ssP84{ak%*Xcw+(>` zL=Q`TT0Zh{F?5Qo`+^H5tOl&nLuPzj#pK5|!dC@rS|aXg(!Gr}OZ#wC=4Y@z=Z1{B z(RO=RO9TfYn*q3;T=Ghsz*djdjb+$)x>JD=oFIh~#dQ`xG)S6DVQJ2Z*l3iT1I z=g1uexe_ouod(Iax%Xrg6lPvodXfo@@8|vfVygb;a4Vv8Tt(>6-)j!p3gBvphYvm{@`TJ?oCW84GE{eysXiXkoQduUA)Ti$HEKjtN~C$&u(-I$s5otn z>M92~Sn&M?fbA^nWtGENo}3$GkDh=(s-eC~UBO|!?nG651${L7Z|i43rd0W}(Ldo~ z)R~1*EMzZ3iJD)GrZF7(LrTr~<<>{c(!$-nblT<6`4d1pMgU^K9P(g0;b7gFi5GBb z#TY>nl!#?gRs8#BA8{=ej4}KU(>`0-8%DGGW@-HxLqET+kIQ+FcbdOQ&RTmHqM_b+ zvl-xueW0L9x3=>`j}#c#N+1y@ib2NTAf~M~6yK z(wnSY1etP3nnPRzCRzPooWWPor58RHPCF_jcBbp?J1xEI6?szNl}B|6+C^Tqyz?{K zmL5k)Hz#6WT{_jZ+oXoy@#GzY!LNSE5?sV-zjnbLeKaV&7}aiHK2!Q`13gU>{zu=P zdyK@b^SI3va9DJ4fApIib0VJ%I0{DC&5mem1zVINj|)TGt+-EkghSby522r)->O{@ z=|t4>pZkQhayYBqCh;@Ch(}?pT#(lzr`bGq%fTc|) zEImvj6VZkY+sg^CWq*1}U#uGb-#di?vRKVXCw~&!ne|EL(&;r>)r^F6OG?OBfO`vB z`=AAGDh4c>jmpkpq|nON|V*|s^v-i34{pB5zZf^72FD6kYtn5-7T?LXSRDW_qI7aLmt}EJ zra4yB8vM>khS~$KE3=7k++^NFhsqbe1P`%!35%w;SlFYOpj<|)syFZlwi~G{f~6px zfr&&1>L4CuCGP)^p{oc&O01H!G#sUTu%gew*LL*wYHR7HHwagse zTF4IJc^zv;EmJU@{hhQejl-Yy3KIS^0Sm40$Uk;-<)aXVyQrs?8uKjBf|`v?i1w9zj$`_}-P67{VZ6;3+l;?w)oCdIa>UOXFOUgb+>- zcn>@Naa27`k8obp@saW*u^o-fDOyI*krn)eX`i$n+l)LPXG({Y`&vu zcCZ_$afthuS^m5B$1_+mgUDw`iv1)7Z)OrC@ZO+4w!Olwqlogg$E)3(nytRdsa0I& zaDHK=t+sJ2333TmxC-y@q(H_>`MI!$h~CY$Pt;1M=azA@PhT2~fH31=62T%TZL(mY zDi?F!Ap#w^kD5c^l-YQBc;Di?o=_$yo|O=6=*uB&MUP$IIE&V3-WF^ZzwQZLm2N@? zR}{6xXVFht#{|?Qy3V%(*Zjy%+3#iQqKa-3#_7xIRf6{6~` zyxhbYTU^EyUZCq%3k?oOq27)W6m4FWP~m7<};_=q#R4r+u0?*_*)5`IO86z(kz=QL1l z$^q;t?|CHhtFf+VcX>`4Lz<2xslVW{zC~$@3t-l2i(it5lrIjvN2#!a{o~jBSROaB zA%|e3^qSuEtq>q-)J4X(P;bFDN-7;3bdAW844Okd7>m4=y;Vl(6e<-24DE0j`qI&S zO>pPkJ6d~S8#ju*DEsH550e6m@Y&nf4LCJf%&i2>Q~8cA5Bfnjs#7U7)Lo;`gg}N+ zhRUeQx1^s-am^2!>}7WO(!iABi$*+Ka6`jTcJpA_^F%$P=;V2?pvSOx+`{xK1%^8qn3s4HeWf%7^l(|;V9&8F2OPrB zg}psQMvv973+tWPA>uXgJ(^QeU{19ILarqWn|Gr@_m#V!lnvpsE&952_gZWL;ZNR- z%dwKbuz%FA{FI}^tNzr7?oh4ng7d7hnJIf}36GY?-c}cP*VJ>8a2k!+8oE-P)@OxV zr4=GwE>~`$PPvyE;v{)s2B6i7d~{{iIxpB-H4 z&x`eCZ1L0?{uC4P<=oywH%ji>zSXsGeMNCzArEYyk!hb*pk@T1qtvFhuja3r`sW64 zWH2%}fVX}_Q*jq>3;;=*ynAN#n8VCa$vA0U7RJA8anjT#I<`G`QBgXA0jwK zo#Lum81TM*Yh`cYWch_wBOo42r#^_j=nk4uSAhP?1EX`MT3T&VYEwNcd9$3Mgzw%R zk=-xU24$Nxt>_PPdp9~#K3^U)W25)Oy{U4@p`?R&WZ9PZdCY12W`;EP5#lhyUv=P; z_t9*)*{zy{1$dpcT8%4~%gTl?_u;OWx@^$gZ*M!PL&hp_>!&E~P*aHlndZKB#Du7j zQR$`CIwV05`2jzdn&$qx_RFJyz54<65LJHoPvLp4{}7&MV&`J|?}+@r@H{))|J?oG z*?DGmE|&ijohO855Vy2(F?Awj5VtXOF%>m6wl^__=I4iYc5yN_w1xKAXjbv&Rk0+P zr8bq$tP%Da3u0v%F%43&Ljt!JZ=y6e%fSjwcOKy~pMw3ZQrXhbF@~PJkX)@M_7a(3 zR2-?K@}Ae(nrypiWV`8Ywxq^y59rB#%i=nH)9syodHcpMaSif6Cg~S*ppQ|crQaM! z3A%rSOs1Tp7H!Kq;;tvomm!mY>@(%z;bKaY5da}ZjWGU({KW^HDnSMUtfHhq6bP0l z!iy<`2+1Tl=Ldyo2~40ArcfEAV-S0Wfy60QFA6EuNklHfbS7B(x-U%^Fo8^^3x|RJ zTU{#kY7)FZj1x;eahy_+5em}mVVw|GeNh~B?GO{OUsRO-R~ss%A#4~jm?*TMfr*Do zP+x#726|bNbe2c#usmTDbQb)l??4bR&`d5oNFkW;b}yn4tKYf{B6-nQ2fs@meKrB% zBNBZWU8 z4k2QLu;rB@Ex%mjpL!~|;@FkthdYEPFkS@pzG8xGIZ(!Mf*OUnGO3Ulen8Sm_y82$ z>2GNH5eK%EZ?RU?VN#`SE=d(>In;|r~8R%2$S zmeVbjm=}`UkR_40pjZ^Biu9vgkfnnCiB%X$zLZB#X~0+UBo&y%CUfxdLVDhLWg3sAegM*2S;5mITW+@Qf1M??kMOx*Nf)o#hf3d!7o+?7}VeqHPzyn8KOt?AibG>*M zxS&v!7R&0(mt6-gzIlTBG+7OgW0Fek{Vj4A%E$!1JqMi+fWXxU7<#Bg9i~1O#lrUu zwSf(r^sPrjAO;FH!BS|E&e+)zX8*3#2VwI<1BS;Lq~iWXsj31VcK{~hzM!Ly(1$et zxv)@x7d4wNdRC&!z*ic(wxe(S-t13&Yl0aqefIk^5fPgqz{)7wV%x!z=j-DZ?MD;z zL}DC}aCH%By`VXR?|Voku*1&=M*c>>bhM@Au73_f z3#+(!eBg9#;aGi};8{%l#qMZIacxUbu;ZHO?LtdLP;H5C>bFtwwqXhSSNCSEYkABw zH^q_zS=wo26#|E9HX9G^Bh^$=Q|LaiYh=~zRK49b>HBkX*?EM-EVVW~s>{G{Pdw$e zW5WRNVYVo2KNeZdW}9^*kFJZ7Xl97q^@4A|uo;2zMZ!K8mhVbi)fA_X!Sm|QOl#j~ zv`uOTL!YijntfU$kwwp-N5#8Vm&>zWa}Fd(wX4eg2=oDCBf9YSD)_angVo>jbGP>5 zo(_MR=J_o^gO=&y7%Q)@bNXtrxP+0L-tG$MNtc)`k71DKZY5m^#UZ>(PA#*hI}bo~ zy5#kZaA`GvfL3u?jT|LK4PlcAg5^EFD)(XI65*pk(P*KP@!g-K&xpJ{JpJa2r&Ad)owx_o~!|dj$sB{6T?uMh>?N7ZPa}txwIk~~oRQGPWNYhXIxAkKF zjgJBJcEnp&1SZ4lrE7Y>N##9gOu|k^bDe8DjMgswY~I2ghIurEaseMRhDlb6#+=o2 zmR35FAiQ-qXU`gMEn@(Azu;E26CPqF8jZ8Gy7B7+76v_&F1NwA@;sP5(RD%^#XkQh zYtOJJB}G<*l_Z1Yx6z4CyY{Nz+w2P8!;IQnhxpwTRcp zs$>3mIeo!w@^`4i3Qi-A#4~2jN>`OSHW{XE_s?PlM>Te3WMlL?+D=t(E&bw7!?YyM z=jMGk4zVqV*HOU=FqU|EeSsq1O7m{L};A7w5xLT6*+4#sc7aCUG0 zCcZIXDPMPu8|5^JchX;rQC;{gVx`uUy1{g&&}-hz*`c}B73xOdH$0wveY=+vE);(r ziN{#&66hS}fLgwsLe=8g%z_NU+v`5CpMzK4oGfk(9Ozn&ZerA2mS5F7!@BzNxQ>qC-$i)GP_no-G-4ZI0sb)E` zsBe|<{HcMKp20Zfr$Pg8ft@vVA8gsTn`sa+H)>>yjjf?{QIwylAH1z);7Y=Wd5r{g z?XOO*HVWT=%`$k85h)Lb80oG*N6O9EjJ0bIW({Q^9g%P5x-OsG4fVS&4&WerTbZHi zR#@;Q?Je*rcQ5d677>|8E){u!^S&SXz?Ru>p89fsZa#Lpf5O*hAx&_1_BoQ0Yy@%u zoEPVqNFJbZ%-AzfXRubdmJre#jz_D;CJ(8*S4`x_oTeHvPu1F9)Ho2N)@dWBcu`mD zc6R*Cn2ZBe=aP{dyytK{y7^r@gkCz@He>}828+PsqkB*3P=~&De6KbVHVCNoP7Opo z=bgVFLbGXhuH&BQ^&cAdld|n2*Pnmw>@Js>fy>cyaH=?7oF}?#d5#st(5ZH_Xxe_< zkQ4mgy|~a(U_V#4t}3b8lYUqp3K_P@+9bmOez-pS_XCDAbi6cMratVE@c zKbCiY|EX{m?sio#tH;bX=%dmn^t3`;DB#;2r3usTKkIU);287;Y=_Cd9X{78`S6Vb z98Gg!NND;8ls3goQCdmNra3ZIzd3pALse@pdsPV4&Hsb=WnudNi(h7r|15r47`guU z;@49JWffmsr!KxAFHo$ym2fNtP7_QA5|JIMgj`6JH6QVWC#ljCM-by zYb-Ay;N3(L7!|qC5f~}XAN9UcumMVv@Xxp`I3<`}nkNyNWk_b!{1_Y*oSdM060$V5 z3~G&(;Tm>gyG{#oCf)$(e45hll6fs$gCJk_9#9yu2q5#?@!A_l5R&U0?I5U{HU!TR`zC@rIwb8rCJQNbER9z85BEO=m( zOy5dM53-z1y&h3<*Ou1tf9pWuoApX zq^1WyfBKri(#1_a6V;&%F(#gj}-F(pA2jJdJR_N~J7vS6`)2m9woG143wOYEXaoFS;RB?uKr{263Yk0o9r zg*;;;A-&PPIodm!7I`>|2~iV6=0;HvVhbhs%`Ay)_mi7`u}ZU5wm59$b@u!+^c*|# z*f#UXarbp_B8c9qO)GWIq$@xm7kYTEKsJnJD$Ow7{Y(ii!a_vM$_?2jOps&Hkn3J6 z(Jz+?KaQ;{%jw-tBrxehK%i8m^jk0kJn?IaJ}m7#6dw;F67N>b`8XbmDd&ff+_^4I zA%Qw;4#>VpBdUNAMZzET;OYW~IY{!AI?XNxaqj^#L5Ql)z@))Hqloj83FgZOSU+4@ z8c(I+bga%!PM=rFO=tQqiVx81H9kwNDXiE(B!f<{-Ce9Oq6M^kE`{I%5;4zv_fN)d zX}~1JG9U>LS-8a9fdU4I1L#VOGV1{5^9!`$-Gthk0kzQrDfVXdk`)wkF+XAPk)|A0 ztHoe)9&LF2;D#X;)(vWJ`(xhE%jfZ3NOqGx7n6tlvQv$M)7ah4Q!f6(j+va)ux(tau1n>jXJgxCl+oZNccLHJuD=YH%{&xQ&ey$r59`JNi=NGO zf5VQJqH|8vl9z43^=q9rJUBNM(BfFS-u%`u%)K-mUrv_4m~#CxR;XCz_6{Aq4;+~1 zQjC6j;HjH4j(zf*31|uT1CX+PqS^z>9_uWEBvw1D)wMaxTWa($KF5*=ZMks|dYth1 z7Nj$iQf+6~%_{CyT3fES^{U-5+jy-Wdph}YX>~T)iw(>M3R7eIQqUlz>2&0Muae>J zT|sQ6Qhx}FW49aaFDrvkSUTGW7bE3maNDGP=j~2LB(28h?VH#n>=+AnQ42hoKTlGt zi&w#WxiMc0zmM)}e+MfY_XQmv(zpy~EB2%-t@Zb zcBT%p(M2S(uh+F!JgNH|>W28}E%K6=4?AV;J1Zyu_%IjJRUlw&(KGX5d0HcVHMmHg z?o4|lBSd$jd~S~r_4CtN=OV0gSpfzw;tqium1ax7+=riUcM1H9H(T#I|D?bj>Nx4K ztp7BY;xyLeP*3Ca>dR;4{tWZ}$Yw67KgzaGW|z!*aS9fu zxQ%qIT)MTZa;S1QdYq(x=BHV>YhDBB8G2)@kE-n)M^aE%CII;E#Qp7d`p#}*aof_v zN+){eZx*wz`EYZ+ht)*NjyjlJZD+DM5P+vg;@RcH%MD-nBl<;;hZyo(_0Y+Z?zi%g z8u#2`G+(m~oi~<}4PG_}bH5kh*$cXr{&MzUP(1EHbot4#B!r*DGy}YOw^^}mkHvl4 z;3bcG&bIs(wujW}7m8uncI8%QkykwC+VkpyJjW62R$}#E;r2$_Dd!Hb3jy5`YfTEQ$;wc6J z;Og2A#@Jb-&@i%l5Nwk#B4g6L(|*NFO6YizvDjrbuCjls@^3Q4_^^`Gg!M*);Hi%1 ziNQHm1t`EB`bEm~<>WBZs_`eV-`>qY9=;bzs|0g{wAN91d$ z$kpB)*!rKZ(2`q`Tplv6H(K0eKAEM)yg(0DqQU{U=4svhlKZOr&tR(>s3lZ3N@u-40dNmU}?^^&IJ}0#yg;+qhq{FytMQ)ReyB5 zmXR4CbZ~NVc1|ZJMf1G+%H%)YY4tR55Pj!XPztQBUy@&821W-VBoLAyR{=8waxDJg z7=rj`R#hj*kS5>?SUn z(G7vdFf%teJ=)eZfTOOaftX?MVS#=@#3B@;`A_uY8-O&iwy;5PrhzW`YXP~cI-;ri z7ey3wBv7*rN*JngNJv1O{_GQ1RFu$5K_H^2DWE_Av7G@W{H?G8_*DR33A}300i!Pc z5496R9Yt+X4cXl6^8psf%>N%0Y51@FCcvTa)d2hzp*dE?*&L#?7aORIR#z9#6BCP% zmluwdb7`DalhOx3dv&G+;-XQ` z>xQ|NPY4?9&2l3s{FHpJ>-=V@j}>bBF1P%sb?}EleATz5jn@61NLE%#0A0ho@|xCuL;-1g-^PtV9Uy5yf)qe=|Iib6 z@}0d)_oGvvqn}LR0t6Be68`P=)i86j6S%K;Ag6~$m+)Y1+?~K*KHt}`_F$t^P>hYt zu3w1nqYZsus=6CHTflu^3=a^AfX;6T5ETF&_0wod2G`oA7x0w-i9-}Ld<{Wh-usXF z?Dzb?&Y~5y{xx{Xr`h;#mYKDgiIt09;vb}ZsL#P<>HhxpnYCvEei{rY8u`5RgyhQd zV}0e%RLza$SCRLDXY`i7SeCcRqHps|8v&CK$0v=i_I9AhDcr1|TVKkBr4jtI1CYJF z@5x`EX#)bUeo@~D+CYUJ1z{O&lvh9TQ~HJn&n58xUNt%px*NQIePd%i2JT`gyvEu( zF#i-m0z0VJHyJ7r$)D~&7ZT8Y^V7=+2G@JAFQ<&_?Lbi_@GS(i0#g9ZwDdHfC^NvH zDIoE^T|yZD`A_>exIT+Fjxl(HrgxmrGP@6gu}{S}`{=ZP$jwhUKNYu6{jrg|0RI^^ zctGL6=ls;008$o^IEJ=yYHzdox1`e-&S#jz2hOL#;Wxq9{rt`a!|U!J^>^o+&t@f= zAFlA9_PC#cikcaCc0pO1##oNR-rzBuPHt__zs$aupb!6GoEN*j=RfM8_diwlyA5KZ zxV*k+>uc%2k5vS*{o@m#=8w33Kgm|_wUu|Zs-Fr|y?43ZP(WZ_AhP=@0PeK=5Uo;< z(A2j`KsJ3S>8TlK`4H+{f>e;kOxAVZAo}Ew7zipYv=!V`0i{!1;=%ljJX{0A3NgZP zo$sc_@A8ZeCS~*ursvd`Q4wSe{K-WF1%#OdeQ4Sn;(Y$3ahfH1!^~+bc^e9_vRJ%u zCwCWtkp`KbLY?CHv*gxVG(j()OmjOS_ZG1yo%f*4@K)AssK4*uVfX~bTQprVEaobh zF{_u_T=ec#U1nQ`b>hl z(1letw#+5bM=2F>{a6w{6X7b#8I)>CnJSP`_$pBGE~22Yg7BFlV1{x}RpD1>6MkS? z1nYHwWgmR;jV0zwD-_S=*kBDNanDs}E@q%wpRdaI3{ZR-p!=sQkw!)cT#mC$SEezd z^C`Y~+c7nGvgDSlADL_8aPF4~sphrBO`84=uKEmlU9lM}cJxogVR!Tv1YG zfyX&#Umcj#KfV(Ti32cwfNOLF8IsF3-{$Zyylk<0<@o)>>ygbwqDOG;H+>ah9;7{Z z>aJJ;**k}pZ;20=X+~~%xM7g!3uisQlB3T00v-pg1Qe*A0==JJQ2Kk^XWTD`a@TGo zZqV(qiIPbh3wPl0O@vN2@dl@ge~Em4VMA!G$#MQu49%)q(DsJI?Ctm0Lvy@&(TknZ zRZLL$!vR9rX@O$gt!80n1;m-V1Znd+2Zi^)*_=Lt;tQDwGkUmct;%Cn)@me3&0&)M zh17r0IrqNMpWj?W4w7xlEH2!Tt9?xL@b!60WO{49#91)og@?m3gMmXzyYx+dlNzR) zTS+nk*c-Njb)p^|nIDKP^QU}6rIAcB(M^0^ml&L5Xb`%h>@z5?OFPuF;hNM5b$F(@ zZdJjp0|x{PQ@=Cj2Xv=|srn8$EpqDs^?w}b;< ziQiapGzb?NmFgGeM2!SiZ6?rEJm?r6YjSDbLkS zJI@(**Ux*Ly8nn81Q>WNHdE8yeEn)W8CCS0)on}Tvx_&+|C#_Y)pYP zGesV8VI2&j09oVjk=N4U=-K>1)22n#VBkN4z935cb zXUQqZ5B;wj>_Ujz#w|FNPCY+Q6-#(aJQ#c%97e36s_XX=dfmzT)ZQSM3|Z$_KDc)m z`oThkIX!itOq)rTXx1J@H%({`_yVNRS0g9bC^}zh$1bFQX6P*I;D_(|Bs(yRiV^K8b@W0gPdHtn*a>EjHF_EN{+)38|cVLrVA zQjVnn^5^u0(2--V@h>*jVp?>63wVHlD_~2%oTotZ&yF@8r5G zEf_kHCzt3U8}taMgn$>2jz29Q;37ik&)&3Kd+_&B6g-(0Z!8IrOM{|zCS?d z&)<>)$JkZuaL^LUZV_m(Vk7-Ei61?#^Nn7~sbo%1L20H*7nn57XfWA>-jp<`DNJ&Nauw1p81!7G}y!hBU@^Pej(konP zq!XdEWX?}$KLEk4uQ&(54#3UWg2Bw_2<%`Q96?dJ{uN2_@5_?8W8`m0-u9#SwAS|c zrK1;9s$=bj<_?w>bz@}@wn-s-RxYcyK#JK|&xQe~Hv4_{D+=zRcN&{Ai@51}FStaa zHdxD(5Cp!sbNu~L#}9`V$Mem_0Ah-%QMM*z*5^YQy8a?;F^AeqF@p{ez@SC>QDVkz zlojf*hT`OP!Rg}_P#rm8pM|>7jHO^Zk@qW|CZdbu&7F$R}-btp-;Q>E51H5YZ57#6%Dbv$aT1{Fx zI>z}xHY6RfH0s0_$tV3ZL5%%vm#+JV@FRCfTxuX%nbi`6g+2C!2Oy=lS-Fr|4gh1G;#u5k5}f#0=Ob(`ZddwnXu~{GOC7Ns`U^bSR`LbWj*tOtxouN}}yQ-pQc4Z`sFF z==hhJWP|Lg_Bzb}Yz}ZIn3!xSM?P$A!ej=Ug1FoW4=lCl?)r=p?|3nVj*}^k!lm@x zAKGtPBZ_@n&{~n}sQsjXzal`_Dh!uMX1;f==W2Gd_z;pRORb#0)W8sBXj=;+tt%@A zc{Ctp0;ehR7b~+m_!Kar`y|OS`=-HGI_Tm=145YXiwptvbf>S zgPY@XElUH|{3Nl1m1Z?h=t%^JB((aaVxmZYX(v`2O<8eF-3xMw-PwA4S%Hwf z=|bFdjzb*r@w0#ZP$wG2f50bOQhK6f+8e#Z*XQ`Tuir7-87f_-E{@Z(YnAUue2DsH z*a?q=2H~M^F07w~s_`v$wv`?zB>ApC<_&q3(s)y9u`N)l;eex@=`hKeg|4jj-nEcY z&%3@rg?+|j6f!TG^P$?I=Y;l`FHbiTR=L40v*=lxa-BJ=F?kBzAm&^VE*#o{^cITn z1!~(sphNQPsAqsP8uA($`QSw0Z=3ElrF6q0<0bB)*=8>ecE@G$KwrDo^zWX1m;1&^ z2wba-`=aD5-CfdrnlSpFh0-6xyw#XFP2yaIh1z;6ng%LdTdtR*9~sE-f~s{a{!v)f zXS{M*vX;J@=B|f*V3c|x?sgsC_G$Zt@ZNM*M!*AUNG;4CR|d#w)uKwhRSL%0NBBB_ z7zy&3%(*Wix!_52Sz^k3T~L`)>?3azUy249yyk_(LXJ*ZoUwU#`GQqnN3Nc0GwgYYIjf3;eW?_wJVm6JKI3f?=YOJ{h zt4l&>sYR!yDWi5y&f6=(_SOx{Yx{em#>sHFpKSE=shyfjN(^v+ zt@l^EJBM0wpfGy2H;8CPflBmxnvB%R{W&U^FHe8jtLe4Zih-#g|7|KjF@rlpalnkK z0Z077Gbmj`yx1k{`90U}0}kn>44-Nxv})bEWuIxd^v(Cy&|8WZl%%FCRxRxQ3u&VVJ zE6|$edzrwU6K0O(u9y=Cj?ltoNbg+rS$Ocf(@nb$Yyl&|Six$>SEwtYT1HYFACuDM&VHE+ z9!YaE7%SZ`%FVkYdnUfIEgP}ha6@ElwqZnHiG9?z?{=f<2U5Q0^9(AlBSD1Q6i<5U z?V{SFsb95frX?q`Ppu>(`VC}_R+tV+RVHkAy-*i`*yG33CZ3rj^#H9P5{6 zKy%v187GB!tD1_(&Gq+M1i46`2gySNksVExjw|5aN9h{AKf}cbnEi!e>Z5qb)`l5I zHal@8r3lkE=N3<%MarD*Sv{aKS8s0ViHSqb#7d&(bx|E_S@9nxTYMEM80VaG2l@Ff zZgmUNduC_*!(l;wUgH0X7R^szx*WW4S7)W}RSW4#DSvX#v&^H5^tw*TvAG}1qlPEy z%bSugtxaT+-JZNX45wPJ1r6)^3ZidzssC!=0M47D0HthLN475W27~Y#Uu2j?vB8T7 zU+d||J0bnjX!gj8crqXlc{G(r9ANZDJYj=Q*o_z7oU3imQ$omN*Iuz@BT^hRp7~9* z=h2ZECuj$&*cLrm$__Lv5IFEO0W$or+o*jzJaoS?%2vfc&IIpHK;oklOwzn>n}SQf zDAsZHC@n>oa9eifE96(LBrFj1ORfoGBtcJ>XdY@I?MK-uW`;ondu3w}naSvz(YUFv zo>~lSBhF_~%WrN{_$earj24R4_*9E?`Wx2)_Fqn(^>wk=c+`ih_Af@Rm&P5j(5Uvz z1jl+w*g1l`tG_;9-WP!2-g|K`NFYu{ok+Ovb-4FYw2yJm3@~G7dK-74%>^nsB)AD6 z`uU1pqOR07;h(o|y05}}Neofvibs5V|JkyviSEDA#!MPDxB3M0HQE1|vEO{!d`VD$ZWVO1U+^{Rv1+&h8h8=Ht|*+^ zO{Tb+`h|O4|}umq+U4NXsI3-oFm7sHMYld{~MFV$+%;RwE>5G)(Cf!R~G~t zJ#_$R1hyOrlZFlX0yvpeH1N3qQ{_nRzlF>tV-KbLvgduY)sOWsvb~ z2d-c{u)fA@VWtVmA8XnD`M8K@W(6`S^@4m;K`&|=nNVyQO`2BmF6cza5wQ^562JSB zts5_V;D+M?23V!PyPXiTgR@8=WCLRpHp;A##`#3-IoYmfXWmX2NR4SW2ry8)C1YC zhQjq?mGSnjpC`5VQBu^zD?$~-FD~sIrsfe7@3)vV1PIfS!70TZ4;2^mTb?}uT>FTz z%fdqMnyD;ZFg%d^4hCjaoflcf0ZP4IMbJ;1RT}C8S|vU`+uY!TwbX20#pZ?ViHZ6X zx)A?r!s_BKN!LW@(yk;;NCSb!B&hRiwb%L#@uG84t+w!PXM5tXRP>-^F%3kLPcRdk zbneF2_oSYs7qH0?7WL4WjqReAku!9yXDbkLyXSauMa8H~(VgTzmWp_1-z{=>j&^$3 z-CnJ86Q*vvclL^A!@gR>{AG!DZ_twPlLgT=U>42n=X+d zdbXP4MF&>Ykz@tFhvyjglq&oR>frLJgk{b~Y=jfF5-aY(t~3qd1n7fEd@`$G z&mg=gK1(>8WNTb-zdTT%oSZT7E`#D^gmhIyA7K|9>&?|PEG%MKW!&2QU>JhednM-! zSe_v=VW)^MN%hhneB0vRo{`AME=yY_Fp`{Th1jBb+wI<4t%Y90g%K)%VJWbsqHded zFW-s@KYc#yYqNecum{@bYeSm9rX3_dUz>f|poQtHQvxu0^PY0yCx0{8fQEdCSG~1M zA3mBYHIpTwOq(9Q_jPgHr~_2Rq8cy}OnMQYAtlJj(qh&W!Yrf>8NN0Qg>CamdY~wg zzkLCW`N#?s4EDJb^VE79=hV!pTZ9q4$j4e|YuA^5hKv+y*_+D*l^D6}E#6t+=)6iB z>wbqRwXYKjkySXRe+c(t{yyEWWQdO1iU3Wy&VF^xfo^iwIBMT8gjTB3Nk(GYf{(|b zqKN{CpXVuz?cpOKf6~D=Q@EG^!?N2!(be62cy4*63_wi=5#%y}0`{DS3{XiNZVja1 zmR$=7;*&A2VuDOxa6$OWE5aQ_e++*=Sv9Lx_hyppb_!;%Tp|~`(k)4+)Tvm=@ea~g z7kj`z37<2L8*ph;iES!0=Af2?5PO_;KYfdw#DNcx2-wQe%qfsG20aFn!DI}P&i!J??h;P~w+h9Yy_o(#IJ zxAPWn4p#1jls9&FSfKv<9Ny$*+-LTsNN2SxR3N4)?Bv@mfj*56UFOVXCGWvIy>@lQ ztoV5bX7H%MX|2L2D=D$yv%rrHmUvDk#UJtQM+>{R3T2zWqv@>8ts9n#3G|qDr;k{P zoTcn_Y2V0wJ$A{)do%+0Je7!tMcF3Q8OZ)1q4n|$a~zzuBI_z(ovfqIlcD;=W=!qx zo$MU{{@W|-HBG~mq7qQ%?zT;kXUB40U?UHw7W}Z!3oD|=br(FSJxPM7J9q{Y-n)=Q!!IwlVy#*@Y4t;&84uJi^xjDK8ma%8q>Oz8{iZeab%GV+Ep zMKVGZ`RC)wT!gU?-B9~&#c>F;w9Xtunn9Fa?>BxJm1~G7GQyO?t6IsgtlNV~N>-SD zS`F7+nSCr|ktXMGF$_E0;VJ8z$41fWu8lI_b~S!s>miCq+7FkM+c&!_GcklKmpT-? z#l@2R^}0>kcV;J+3I9mrHG!+58Csbpb}0aIHHAvX`@D-Soqn9>Rtj;uQ?*_nA~1^& zO9$3d`Oc9X)yf!+__S7_TF4iy>95;;+#xJ>gj_&ux2U!kqifdwKrEEoqyv52f6vht z0IqAdBf}T&XFz8C3UIVBnUvExS!U~HNCX8ERUgh+T~Nh)x&y37-u3lR-f?l9k^6^9 z-5Mi+(VWD`m9;1?!58rYwOKUam7nG~|12zSdMjWbOe+HJz9#Y!1mph2SO2qRG`n{U zPGIMf^=4#Pk7Ga%Q8?Mf}LZ(J=Vu3!(g{+}JTBisI?qI}xY*c?t$b6nfnn z74aMe0{@VHlsLX|gPx=hSy~INqC2bw!21?bp!(_nm?GrzYBl1-ds56!`PFM1rLqF7 zOrc48%uy?8u~yxi^2)(c7el|dbslE)&Uo)+Sd$ZsnW^lFsVY?N=V-QU0T~`1Gb=wr z*<2UlBUBVW2vG;NBx$EPw0^xBr4GqjxaP0% zjo%wOAJw)1#X(eVAALNHeqOF2^f%K>SGDkL;f3GG!cGiOSqlJ-m8EdAQ40SO?t5xy zGThjKdd`E1Ju;^A@Vmv3(urG7CL(We%nya;<}INnG+e+M>i7KbqdCu$aHbuz3>*e} zamM?mGfj(WM)o&3DygZ z+=JjR>poaKNB>5gw4@xM!&QXu*xGO1pw_w|W&NG_fX#e!JEujh|Dd^`rvI(zUd1&Q z)_u%i{}h|ZOns2aqfuKi^QYH3Bg=l$edbq<>~)6?{e(mJO`d&bwl+fr)u`^Q*i0Jj zFK3*#PcYQWG7j0F1vbrwN1g2xZZM!W%6SGx(CoV#>1#+5vr6Juy<6+{WxgCme|}x^ zde0vaOqC+JE~OP*w5U_wWuDP$2V4*kE>B3wRjBZVlj7ZLUdvwFb_TuMU|EBd9P11A zCS_?`RHBfToX_I=iLzhV?t!mTCW4YQI^|Gr=HgaYxkuH(P*k%l#Gbxt*#+b)TPz$a7O_)OONt9K+iZ)t^3H(Q!GPJgc#iI+->m}mZ zfUbxTuAuFvRD%D<79`}T21TTssU@M*;O^6k1Z^be5Xk`hcu1r(ox5QXE?X&#a&l9T z*MsZ^=8+K68h%h9^ysSa*5pgM8_(G?2sPB+T>M!l>W_kM4;=AnP;~Ee@lxStY3x3$ z+FQGAS^|AAkt)PiGmheiuce0hq;$VtF&$@ z*aWEsr|EM2&`O=_eXz}HNvGVByVi1mss#Q{pYjkzdn2~~sCe~WPhC2=aoQV19&5y2Lse z7x78chGZ06QKV|;;^XgB#xHE9hFq$~cB~i*gHUo**ea`!KO|>vhxU#bll|_z0~zR) zLKaH&R^z4*lRH`Xt&Wz0sZLcU{(E6J0@ly0Aeuc9dFg7a{wm5cEyH#$B zlsRy?eUS0FoMWj+l&&eHHQAb#2?Dt1Pk+h&fv2-y05#KauBkFfV%t1ea@rd$fqfk{ zAmC$P?bx-Po<9Mc4V*+a$%ee&hg);)q`A8-iyW#Ek#y4-vC^DS`aft}9SCAz$0ME8 zU~0qEh>@x)+RHB{SI;Xc;|eP_jNPE$lf)KD9v&l@kQOZ#PCeYB-vec8q4XFn^n!gQ z!6v;K^e)X>3T2H(G$#hodGHuCd{2`#DAU7=($T#}XWl~*&57B0zBkRfkg~vn(W`1O z)vr2~xN@9ny&Y#K-G&>@+(Cv&kgLp__3Se6I9~Z-Z}zzf+nXPZ=F@~Q zX?E9QnJ`9@-GrCNvG9ip7>^Ms$r={nxzOhGfA#!Q(PaSBokx(4e0lCtmyX--1Jhnyad*R zIVkd~&?CGXG%>Wpso0{BK;C(Z=BH=1HEq1*@X73ObMO-|h2_@u&@2mVe|iD6Nsjzc`%GT=cvl*CN)pXbA_xD{f1HC@eIzWh^)mJhu?tFILo@BT~sT^ zfIMk2?MC|53ihLPRXJ@_z9KjFs26m?c{5xgB&GVR=Vi05zW$qQx&c_OPV%{ez(zii z-|MD-_a$Tt5Z5uv3(GLds;8-`$6%!`2yyggdY0snh z3@xkm_F?&vLYO3Rp+mn{XE8McU~39t z9wYyRI?`8BSJ=>+qclrsV;PDw$$amy6%~R&!uH>kJa?_U1cj=uZGI$Ak$Z~}Y1f>HD+_QToi^fhAvpT6<Yaww08z0q-RaCB{yx?ZUw}CQ+-dr?%^#k z!|mS@v%nMlA7QthLUhXB!Gfx$8R5C-iF{AgP05a|lbOWR!vi^zq_1)Al7zSI6HU`@AboT}I)^`qE1ePH^1I{52t{`5Q?p>m-m70F7S?W=%FSe{a0&cjp z{C1EaQvUIBqN}!HTSPWf5~BNgt3y0i0f7h378F${aH3t}DmkI+i{P57bjqvTn=@zn zeP3uzY=WfLO;J6mHx0}tSt*T+$JBQX2*KzpDRFG~fD&-47_a{Ta&yMA^r$RH>!I=Y zDPxN3`l21_Yacxf|BJSwLwd7()HKON@F&)0MIIXi9Lmj65@yj9kI<}8L9EUqam85a zzN&Ts?Mjg9>ljF-RHQ)V=?xdioio?^d1#%;VWTZc3DE(9p~Y_e^ADnj6DN_?1MYz{ zf)P|qK?n6d^>m{A;>Z>T0oKg(G7!HX_x|wvIa{(_75oks= z)Iq)Q#|9omYFRo?>}KiPW!qP38yZXe6nJYiV;EHvogBO88OZ`j^Xd@62^rcpTqJF| zM@$M(Po2icO33l{gCoP>*HyW)Q5uoOB9pX2ZZmC($Jy^0+D6BrGC0Kj`kh9qRx+Aq z@mb>%;|-)m9BXa1h>2rDW)d_xw;p{a*wgtx8Zx`X8=17q{fERPzT**$ceWNJ5~Bc>b{z8a`NL!xpPl zJwAKQA;?0`?jxv64csH2F8^FP7IWbenKHC!m|mxl@?nAQlJ%38W(#XKckRN142q|S zfA6tf_rgyArO00JQ$dtsrpw{pNI5v>fJiwcYfa zQz}y0?k-!gGjz7lQt7#^E;pOUu*=e(_$;gJ8r=j2mlIZG!sP1==={0DTiU|mZ9!PW zO6_qk@}_-b)_qL9M3-eG#l>_~CCIRHCrB1HS4)@;RAEu5QBde2;ubkWdG zMJc_x*U~3-O05J>H%eK%cOu>)(=gDwqJuVh+H5)s;Q6~upx914o!Oz}bGcsZFlsRS z3f|3DQRkp?YVOKoN$q-Gv*WCD1`f}9wH- z;mitt3-V><=~^97?8YiKT`LYr!^7eabH+s3sVEYrbZt!R-aGBtvoCSB{QWADOHo3V z$45zY*-5}}mPFTkwVZxwvWb(pII-r%wYv4}o5N7akC7wN+a5gk9kQ6`+*5xjPx*oea!>I@Q^|hkK}iuiS_is!Jr5S zOZ*rK_?Q34j3?Q)8M#4fxd7et?TrebB`@k*NokO;G8$+;Qeq+uwxxcsE;6Dy6EU{7 zJHL0A^WsFE336T@5y-C>`8$HHANvPDK<< z<}?0vDo1r20NKpYyOpqs_IXj2`gEK^!w~5uXX{WJB@n?) z+37qUWnhy2p-`stL!nV~p5ov?hBQqWVgF}C61D#wq8WnwEXx6<|8UJ)wE>~2=m9OG z}gk=fHMlZ0ifI&QS1*jrW;#jrWXnS?jZT_9So>$lD%_g}?&M@8!98&_UMS z5m6n}30u#7DmI?jJGy;fb#tzWvM7aXYIgLCrW1BB;UCwr=v-x5FxP|6hH;>DZt3of z)lesGT%F=q#8Q$haOk!sQ@KiL9l*(S-L>>Wb&I5tL%ysCc2V)=gZ@gC*H%YWf(uqX!| zD@oD5a=P^9^oaN!;HDjXOqdfUwv@1QcXb#C#Sr}!brBwej6$JeCG~R`-taXW zAuYxDpmr1@)}HjxWAsYdVxMV2>nq^SAfk^g*iy(^2RX}Uno}YBdxW=q9i|@C8q3qn z1LSLL$9LmPt@3B~&hpbk?0>ICFzbqId6ixuglxtvVMgJBwRI`y4maOUCE81Xo54U- zf}r+ZBfjvUk%yursU}e@zb7UZ%VSuH%yqa#H3XvYAEro_bmd2AN-YEKF%hd&3@8vV z{^)fnj$ACyd$epwgdeV`m|FJvKg;~U^yQG|beQ^;$&W@4xp`d9bDuu_JDB;YFSUpx zq4J1}1Y~_+Aq;jblY_#DE3-s@iF=8`eIw{}3~-htfy*Se&1TZ;e$Xu*cA}&{ej3jX zTd6Y5InuCAFMC{|+~>OSM8Sd0*K&XDnEzFfm-a!?Nw%zmZc%Y%ZC5yWvR7o7Y_+tO zn!|;lOq%l}p z;hLke>?3RTNNBbxPZNAPts7li?=sboQwinAmeJ4cD~`Sif~%&}@;P2^%}`D|@bOi* zJ=UkJ?HSciNTZ}YHAb4Xr=gh8v%lk(H-wQ9_nn;9@J6hZeUT$6+vt;|cbABi! zWxKMA6U;4bM5AhovYyvqC+D{vtjet%^YUcsOq_X0HWV%feNDfrs8a zyH62??^jdWdkZ+`<4h0rQ@$jih1~QQ)4K>&0Ef_Iz$vJ&AfwCVRxoaeT!f4<9%}*| zpwUmvY2w8ne&X(I2}0XTFl{H_H`n|8Hs=iI)!X1Rl&oWv&BO^@$%*SXvfKEJOC24I zP#lwDoQkEfUsm^@pFzz&B?v)>*@}SyO%?sZjeS8TChh;Ofg3S48xQ`7iIzLL_#JI^c~pMS(3&LY}0x(DZPeD>`OI7r{T5WM%}iA2?Mq5_aPROP@))bi+_7#3ra7qONL zla#9bx+E^ni!L?mSBXy<UHq2Bk~%G03RjBP5#$KqB&9W}7n*$u~A+AAFDB1Q!$ zN+m9QIh*U6w>@RR9Q%n0FWNL%qFq6#Y!S!PU^a9?rTWSBEc;+N7DAjND49gj21Q5 z0KxB4we%M#ntbMHl&U$MQLx!NV;Ke{q1VHPmI=ih_oiz>WkJp!Qw^4ac zl}6s;709vnF7JA{#VV~7`m&L1>Ine+E}I#`vkdVTh*~bnH{wLcsvm?+NsP>+E~pioGV8!>&e%; zXkC6$s>h}I^f0>d@Ldgt;jsX(_UKoRb=eIzqz7!8@y);n#5jo1yr+~?|1Axa7oJ*P znIoKox_>0>j$eYN+d00p?G#Rcs+Y3z3u;L%)Rie*j~b#3!I^Kl{aRfX=MRwpoHA*u z0P^t+ifyK6I$n0l+hvBjIs*=MH>7i-ncD^EOh13Yn~OW!;9HX+dNEVxvigrY^@Lh$T+Gz3% zU?XyKmJgbV^L?3lwWD03ek9kDg=f#Dfqf*5TwTGo>z5uDxcoKfD0{O0mkGniLsuQ( zWr$?~+>fXg>emLajTW_nk2es0{t{8SFsU#(N!0^8^V`pq^e>K`Ue@eDbDSP5Cv7sa z-_FuvGeJIC2nC&7^nBF8E?LKXJ_r*-#OjXLHog~6-mg~ncPEvgr>z$2AMRc^wC2A~ zMueH1GduyYdnhS5XVYA-m0CXOj0NRzt%lBTV&O)%-a{@_Av=1+#i+I(MaHvt9qCE;l2I+J&`>2d+Xn zD_b#!y@Qe*Z5}VWByscIJKH=zb2{dYpJOzLvO7a4OiAN;;C~gQuu<(74+O8HH6$#4 z4_;LgYy!+v8U}<(CGKQ3sCrCo8ZTcTDdoejwv=z>NA1)q+lv28l&$v_jcVIuVjvrJ z%BMn^f;6Si;pDLK-x--h z>R(9*AZxC^1nM){qe?0IE0`=w0Zw-LVkU(9*$0Y*5`M%aJ*zE+S?@hNtdKlrA?xw6 zxp~(Ikze!ZflA;L_ZHHGm^|B#JxR-WBWX{JqGVG2Tw11RYU$R^yQLDWdn z(la{#c`pFYt@b<@TK@BP5~}NQ#PXq#X}|YF<_Itw1J-(H=^~E8Ccic<`N~xbwaLjmZ|cQ*P^7=tHYI3WQ^jvo*4 z0m&I6$u%7M>$nDTPyQa|__u#3+gRtAv1Ex%pdhCKr1Wg+;3r$;r}ZbTPqs|w6K-PG zlY9>F#{MiAN?DtaN$$U>YF!YeO-TYBL)Q2$K$Z*V&oN?^YmKKQt}0GlvVq2ykK2IoQp#o>esxbu2lnPH z_g!tSBq`P)6Z1#lrRi#;-w(%(B~`P1lWDO7r^oQHjO8qCV+|Rb>#$?o$mKC*p7|z= zrqUy?s4;yi7*mr*$F)KAdQ1`kf#s))-F|{?ODS5gR*1RL(!10T}S?&jV56nr1Hj3 z8Qq3Y6h-EWR^i3x{hndz=oHDNjEf8;3C9(!Z%D^#W;L-8;*>6AutiNdy=cJCK8cwx z=^!uMYtsTpl_ED{@y6jJxSzkJL!1}Krxe4>#_i#>aR>3jSDVH0pC(j$`J?vR+h0-h zkSAfB+XAJpsT#`nODhgIxH$>A?l}5C3g_n$9kPy>7XiPRZ#{B;d9BQk369>meb?Nt z3$0UKYTUl0s+Oj}cRqLXBIRX9N$>LxRHO2ysH4C7Cd~(3XQ3g0aE~ga6gb4f$iMmp zNYh6cX?G=i5<_kdM|~xqkd!siMK^fZyTL5+YsqWFf+LjnPo(lRb&S_YXfkal@jRXo zwCqpJouD>uRQI$iQMr4P$;Kc#TYpLW8dOxZR>G0OUw^^q);bKcX*T~9AE>o ztvdy;R-JrYM*k&_y5=vloX{{;=E~IT+}tZ8OJ1#PPSCXlrDGZHp9rJZezfmj zUacJj4m-FjLdZaD1drKfn=jQxmYo)F9RXRRe$nx_kkjm}GNagYoz+|VRog6DQ7Bif z6V?_NRw~m0SNo9GA_A9QtdqWUY-tJ5vU+}n{j_{;?KcWOba-19Z0fz$)2ov#j_P?Y zuCm;aZ+V!?A`&#rk){VYrZM0jDrEo3@qwb$zM%OklfVPQY2_Oxs9Kww1(QEp!s?fO z?RMW~6^*u|CeD zf0HwXrtjHoOu3}Lbf-3dPyf8u7UaaO@Uz+k+7YmOL7Oh_`m)vKQZQnWJmQ#UNOwEQ?D}Tm4E_^ei?MEk<)VlKJnBSeRquuT<>1g2XM?H6 z;0z$!{)(cKmhn}z;SF<%-MEn}*k(xyxVZ0d0btMWkiUg4dh&cwVv?{&2v9Bvs%O*R z3;Pmi6{*I%hM<(+Xyacev@71+obd{VRS}C>8$8GT@-dRD2o(l&{fJ53a6I}De_3ju zuf>T7{fcVYK=V-_JHGdJ_HdBJKHYQs!pjIopIi8TnD@J8pRmmUa1anD*4g*e_7+f66Ulq5Kf#Js2*j(>#K=xU8B;CE2WRK3n zxZ4~`ZCR3LH{UEg!EUKu-o$Dnh$ZQUp3F7>@VtLSW8zwog zMM=2@h^m=2m#*7|`-(9PK_fml$qh?6t<5hc30?8O7lP=#ATgTyq~6<9Z#_{twXl?S zH{Qmw*4h0JYE|j9GJ-Rx*?>?+@x962c4FTPJGnT12y#&G=1)njXR^!$Ss}{z^z-BxvwfZYs%J!Ah+Lo0)6jhKjvX1@Fw-co zmN5}!hvLVBI)9hPZst@ClM{!Mp18G;mxt!WspM#hr7AiH!lV4h`3x7_-k1^G`5E}T zT1QR>e6j@RO@E_A+&Q<6`D!{4H`=ntqx{_r%0&I-w%ge zLYK7T;p%MO=_7!g&-P9w7pdJ^@DhnJ!2PfD*Wl%++Pg%&)c1JMhPhB$KGrmEHn_=Nbk! zQr$U~UgpnMR_25!ShLXRymW0?nV%E_13Wr?RVV|mNFvWZ-uh*ubsxDcVLf8=6^Y%p z5#H+J%f~AA?8p7e3t$Fv zUj*iMV(r{~vQpR_gD_7iYhwaq5&{+rH1(y7K@PrMzJhITZKkv5)xPT}TJP7>=x!%T z9mt%wMq4x55EA%ai7g4fR`ggeO_ffUuX-RVDG#D=;?wcRqy5^A~W zg2J}c5HcV%;&SB47+gtZn2EG@%FPPKUO3jSA?(nS9)a&C)4|wJ5Y3Ve6O~ zhmXDP1OMe0@(c7F1(B%{JfyY9aoD;s5&y$CEbmDEQ}<=M0Ay2B0n#a;jrTyW+Iv`k z1zkU7K^snj9oJjL8%Xke2dlNSAWtOT5*J7`pA{)>QN7*IFg)qlOc*<;U;g zJr3`eK*xq^nAAieOr;y>QPn)V?peH6 zZ5GpQ_IGPauofZgRDO6+I+EMy(l=?MGP>c1Uw zJo(^B6VthllmF+As(^8TDXy$bA3&KjNnENIB)dV+?fyn&S-v>?P{dDEPF_|_tHmbQkZ1DF#vB#F(bn^cyILmir2XNA zOf~OD041>^@I6lbv7xr;$JbC%_2S--!ELU43cl|6_$ruipAdzwfgD#DDz{pV({j{Q zKi6USi0lP1AjQQATudZG2nDw!_j^`LnMI!s@?ZmJwGe|8X7um#kim_vWyu1!8O z+5op>xVZFizmG7%v_XmFvYqkY1@FlTW$~Fm@qO7|xYrlrOC-lM9<)n`zDM}H`7a%sx7F6*qS_`96^wMl%H<=r` ze&evT46rkdl9T(CgNULU14Ta<05|J@spUPQ^@>)2E0CVL4y(x@kM^JuErIf=pC+?t z<>eAXqU*?4n^2rQbQe3#z8kK%9@`ds;dxBheaLE^2J7ioKZoNE%lXG#r|pR~B)cQ3 z{KjSfq_a?Ble~@06dKyBx;6lI5|>F@wg;EDSVziC=}1LLZfga}p^LI| z)}^sgL9AXNK?V)mDI)KXzAQb___-L(1TMXbCc@CgKUjeegFMMO_F{oM!6LeTJh92E zrY1K7n!XDyV_T7$z?0a9x_8s*1}r7$VK*)?=FeM?r^69id%Q`tnJqL)GR#0>p%nIJ z3gP)iH>D7lXyS}v^)F{^q3G;_&O$)87o@;4jGk5yz~b-M(s}em>o9FG7Y3j(6zb!r zw1E`=@bf>-y2Zr5DpW^Lv2rt%sG+uDx5M zz)Tos8%|ZE^Ey9*o<^XD*u~kErqoe1nV1x&@;%cV6$L>8RyH{jlzDEwRVc5n2ZI;1 zfp+nmNcseQ#hCM*2YsthH)!T+1t@=51Qi-A7hxSI@MZ+FJRQ|ruY_A)PR1(@`mpK0 z^i6X~3w_(Ns9hT)7wto8T$eeeyhHDeJ7d>#<=-c?|A=7=)l0=6^+^qn&4wE%Nu=aY zjunc+CxYgYIu|R7z5yQqP*A3JCjZB2^WW)z;gwAP55!W%)4`O0Uf#$`*~J!$UXFl~ zf#HA6I8M$k1RPB4|CeJ)z{0}B_Wz3xzty(3$L9P`U8TQ>?~d59>645e3~FS~GgLlO>BbS!J|ZNwk`znRqEPJ!7&o zJ!7`@Ib*sODpN_bLG+xNqGc*&u9Qub{LxHSfcVdNF_A7=?pmgB*E5|`5_uE`Wp5IO z*c7Dpsepl?*psn3DzoWgrNS%N6gDg8#zG}}C;;|noC!=BMG-Mvt7y?f=2F&d!K7r4 zp|lNf9^;4>WBdccjiEHL=^_I{sh*j}lEtGnL8Y6h8A`Yz4zzNRE3IsRD`QBb#+did z8)mFI_y)tMX9X8J4JgAggCDE{N|+?WK|=^ECOQL4NYEbvers4stSZLEdLRe9H-yaC{oH__-304*p5w_*qs z0-1qsBw=v{af(Qd7*7a5`hY$l6et?Jfx>j62*mn;-AJlh2d%M%T87-~XPhfzikm_f z4TBoGNHz^qJFE=q2G9$(1+%FkLj#e~%w>aHa{(NCD^aw-%teBUQZk#mkb)A{ zLohNP96bp}1~qC1fjA{-j1)6_h|h$lGm52o2!fXP)gYv8%7vsx)(U=;R3xn|Kro3` zx)|uc>qZh5SFZ{rsI;~KIHIzFzAqSU(%PY`23VB{q-w2O-Sc#PCdZ`6{i=0UvOE)f z7M)8DfF(;6vq@X#&(_Uc7kfnI}Lz>1$28&Ul`*zt5M156;_~$Dv|wqyDV7 zX-ayPoT&1nOpXEI7H6rLSF=rUOCh}h?wa<_AiUV+PRNG617}Q_3%N6M<{8Skuw(h7 zNBaH;s3S*vwwkdckseMzvHX4*v*yWv5Rb3d>ppxnPmZKln?3wA#XUE3W4oCg z{Z*rf-N>}HsKbu^KW*x9^iP8~PA}>Z4-A+G51eEmNT{7;B@1-%SZfEybjBNUVOT;g z<-LOs(wXqyu6*~f&rSU3>BNiob6u{S)uR18_jBUS;;WygttG# zpnxo=i8*zMBOeLa0z<)|oY`*Q`~Cj&^kL1-RbMLS!%{jf?^jLS*|Af-e!M8XCiOvA zA3iiU<9ij_>%-56_0r3JB#rX~9<(ZS?_#F9#5#B~qC^VX$YZ zg2TktDd7#TF-5zC<2q#{ib?Dd?%j6oj*j%IP`~-t3u+VNdhBj6b4=P4Aj#y(y zmyT9ju5wd0+(PEb!!tc}{jzoAP1ldA*4BZkpIX30)mB9oMw+B{u8i6}u1xGG^M-s9 zL8y+Og@`3XUWwwrS{8}qvSuv|X#mIbWGzfxI-}Z$tg*SaAFeJ9yn#!W1UYvA>Nz4V zL3@H=m2ZG5$#lY1j`2J+B(WGJX_j+zlLp4_LTb3sI)i4^fHfUy+R5liaGNJpUe~}= z8MW6ta(rvc@~`eX_UWf5D;Otn!l>OQCZmYXt5L;2d368d;MTzq!5(?W{8Q_p>VS(I z;1DOWhcV3Lr^*pkN!b48$pyHcM9E22KD#v{g18cWc_=&xhFnAEHb?(W*-}O zT>^V9nYlc(M%IGOc3iX(gGMXaj{!%;`JNSPS!fce*@{4G0&PXCbisOPV$Er9zCdl^ zN65?VftzRJgUPdn!nYGA0gLezf2V%nC|47o6F@j>4{P`PH08+BJC%6I>MqVdu$eoH zvx6%>`KHaldHitp44&;UGN#uix1~vx^HAJ1)4LU3$*OFfp`Mz_%$>Oo&z3rGXRJ>P zF?2Q!!H(+|I}MG*UbKo&p|KH>QdHM$#-yEN7qjohW%v36ZYQq5r;CdY_w1HlALVPj z%q|BJ<%7IGwkzKq+r~*^)II2kRxr7+M)A-`ynQQ#6%XXPkL%_De(UBoV}m}UDT-)K zS)^b%2h0ngMaEB~5LgVh@?*?kbWXNh&3`v5TE|sSuFfBtYO&rj@WG`BsJLR%rt5+i{rw3&(RLxl5u3c_tTiEv*^knb&&t7Eq z@89N!GgRzhjZKrSbnQzk(niRpiDBZuLL(@g-b2R+C=B`l{4z4dDWM^u12O@|1k0cI z%&_7ASl?G6)=uS3naR@kmS*>AuOiDW7)Lw9?M7#$81oZmxTUKQwNYYTvb8eh!L!L{ z28S4jDd7?S)7W(YMbWHVB}+JKpkqya_kq1k{v@I zrVYHKonq6ieL%40Arx?Cis3Z|D>duhIIg|96gYSPC?(L`&CS;6W&X*ol=dWr&%`9p zaEZ_eM`y;Q-V4qG-p8lUTFXUx4Y`4IdJecASTPM6&ZF#S%0w%Afu{usys&3x3WrwM zx%Ht;ZV-1&ig6|xcQ9#nH>J=Ue9)&0{r5U{5y|n0vh0QL_Zly_*1e#(m!BikCQIJ8 zwnoU*uz28#&dk6?74r!t;3=`XFQJ>>;&g!;g%h9kY-mR~N?%hJTbi;P>M@o~ZakU| zNhmhMMk8e7uK`73^uC!r1^VN-s zGLNxD=fS5sx$H0BCy_mEV)~5;RUe9l zP=70j-kS-jDfyP^NgpLg`%TBy6b%&zz`}8Q$j`j`-<(UJMo@y4soNk6 z$-I3}C8bX>lcRs7f7L)xeRZD^W>`52*=bih-Y2KT>Bf zQ@E*@tOu*gr+gjxdW_F81+Ls$o|UwEe3ffzAa|_+F{-Zk$=zECXU(90!%bxDm3VB+ zNT<|`GGo+>A`B6}Q3StkC^u~1H;D%mvm#n@t|6u_x77zGT__t8h7V(g&<*j=lmt>I z9rObmAx{>oW=((yH3(3=2NZ=h5H=>9K$j8ari;aSC}pNnhz2^Ye;l@1C9}YLls7pd_CeC;z69fG0DS6)KD@a@4{! zd>-X|#4VYTWu?}aqAmxdYcAn=Ro3xDTp;P)+c+e5u%~3w#)`tW2E_fA?pai=xkKaE z9801w(3pY0=E8F>!HgeA`2;C%Z%hZ#|3NwRvxH>|l`M{~azh!HGN-hfjHmsmpJ_iP zg;JHfHYu)po*_1;y?NB63ZpR}Phr;M+AsL5pWx+nN$@jB1%{0FhQoYYkd&8H^?1*<(1 z$N;Ih<}>GU>h{tH#!e!}I|qa7xtFBp3S`!VXR=c=Tds%L{9L1Azd6TDJ>NIxjRm3k zFsd7K$;_-bvi!6<7MAzAN(bO_!)bmvX_ zN|A@&Qc+9ET(P@|d+TrFOme{RGl@F^Wz)%c@@A85%dVE9rF)ZRdWl)34KelQSbhm* zw&BU~w?b{`zUZNxLi_5vdY_E8NG^Qs&k+`3;$VEb^+~9e!KeUbVPZ?ql&3 z!p!wyOAQW?0cBsROF~u0d>OHE#aTdKG9yxhfhsrK69&>@L7x6x#IM#ex@lq zkXs=!uLT!?{n(q4-e9O*0T3$5iAo--Be$z%3{Lagdb$XU;HalPrl)#JZy!I{Tvz1J zZJc=iYkT7D;uL}ypew>N5~- zY_Rg#^J8Pkda=e^cbp!q)V}5sG;#h@IJ?D$Y=e7G@g{M*qB>Ld`Ki+P$B3f}xeYrO zz$fxL$uKUa@H&@tMTuc2x#!;4weL>J5LLsr6}t<88XL9ZiYa1OMZ>dWhRawnmZMSz5b1 zYkpX%bT}xt2@$Y1e4Mx-4dwYhQf}xl1dx<`G z;4!xjKVM#C%4-cBd*9W$_Pcti9uMHSTH*-i}Y zT7wSiEro#88i9V)7Pw{}*w2(l1@*3b3K1VJUHlCQpD$0~>XJYGL3 z@MHeo&P5wDG8!)VaogbZ3d=&vUh@}%oWz5Pug*=}bLaPyf&2HBsR4d%q?Wx8sZ zl8Ov-&3acR(+3%rFymt4HA#!lF2xRt*hRL3c7OcOzLTqYF6nwwEcB2NR=8*is@iFK z6&l7Gk&+|(sdg)4te@?g&yN-!F}jVwt0@|bH#=zAD)X?)RWHc262=msoOmK@C#t|z z4%R*(ASmn~(v7O0FWSRfjP!?#@pVO#Qd7VC&PWGeQcj&@VeCt~ zr)O`5v;YKPe9-_jN$BzRMxp?o_6{!gPDpVXBX18!KL@0@n5QF(1OtMkp|a8-xVVfi z5*Og%?dT&WPWx9+7!P*@$s-vchPV!v1u2j;A}|;X0f9l1vS3kJSy2*B-^1}gLCn1E zJw1_*B*SHUS05y;jFFj!g|w!htLuGxcXyHzvzQqg;{zaFf5jsOxP$ceAq@?HNQ0mt zm;x9M1A?XD|B91@C(TU{1ekoX@@=f8z75agfh z_kX1?Fyvp#7wCTtT>$cO;J<{g|H3i4BJHlntnB5s#RKsjFV~HwP%qcAax?Um+b;-g zv+oJfZ254O%ZLa<$Vm?CXyYXWQoTI$(^1gRgZUZ8l#3TPzVMTagt79MUJwxX-$e15NXBlSX%mTU@OfVRobg9;) z=-H;S4hQ=la}AD8bvgp-n{5o1EeSyv^}$^;GZ$^9l2Heh2SXYLRA2g0e6L=jNBqPDE^Yr+_w@zQoXT`tr zwkv8^%G?2%=825;tkoK~D!g+Qu#mCq_H*l&>|*ogp9VSO<}t9I;{HI@?DhO}F3 zQ|=U20%o1pplTN_W~i1t+f<#P?)7zHI0ct=3E=8_Gf?zH(7tEO&T``1UI(>uD^=BK z0M=qUXj~^Rl;Yh=SG7vu?9zNIFn1YeG8dZa@Hpi2!K=g9;Jou_UPV5Ck@>_Rm=kZ6 z&E=Fi?5FpFT5TtJC#utPYO}#T{f=9nHDA^$Cor6jFkp{_IBTF>`Z@iS+b)a9<2k&X zcDUAYW-D z_nmMNr%9Iqy@{(M1Evtz>DgNqzv!AAw1Nl$a;mV2cS4LIV>(#?#( zH)?cHjqj1i{$e)8v-u)lpk13^U7`g0uS=B~>gsf?r#9lN>ntO;gXz<{`(Jz3h(&|H z2?+DnJU}Yy#q{W0*K1u(1G$mS`Kn0V(cf=&Cv0o}_Dl9qXtFIlN*~U1J{fp08PvCG zXjM<`PuaFizC92fiae<&d$+X2DXK#K{MY+_b5w|cWB!$`uG){h!uFxbLj^?&kmHx0 zxLgf#E9^P>vSA+k|Hq5)fAHd8@3n;q5=AR>%iR%)1z5`hKu|ebS{YMJAo5R^R>mA) z4FW&_AX24?hlekz5=!dABoJn9KG?RaEXD9=O=(wv z_CBEWntTOe~ z)}OS4G<^n^(uXZi^m@Z>dLNRNhPkH3V_a_2*o#Hn7YjG(jRuQFOO!s;N!^Wum4;jG zyiqBS@<&Gorf!(NbP3Z$N1)K*jBAgi*P>|X)w6W^i(qTvc54qqxWZJqB7(WX9k?RN zxgM%>MPHJSxRoA(Ne{P7e~7aP(@53MN_btk@9TiEgnjE7#XB!V7TunhHP>}1w;g^q zd3#Y=%B_?WTTLiiu*@q05?h{;&RflU;D*sU+n3aRyCK(YIcxfRC+BNY7|_c#)ep@*rAaVGTn~+W!H5 Cv=QzA literal 0 HcmV?d00001 diff --git a/doc/Projects/2019/Project2/pdf/Project2.tex b/doc/Projects/2019/Project2/pdf/Project2.tex new file mode 100644 index 000000000..1210054f2 --- /dev/null +++ b/doc/Projects/2019/Project2/pdf/Project2.tex @@ -0,0 +1,403 @@ +%% +%% Automatically generated file from DocOnce source +%% (https://github.com/hplgit/doconce/) +%% +%% + + +%-------------------- begin preamble ---------------------- + +\documentclass[% +oneside, % oneside: electronic viewing, twoside: printing +final, % draft: marks overfull hboxes, figures with paths +10pt]{article} + +\listfiles % print all files needed to compile this document + +\usepackage{relsize,makeidx,color,setspace,amsmath,amsfonts,amssymb} +\usepackage[table]{xcolor} +\usepackage{bm,ltablex,microtype} + +\usepackage[pdftex]{graphicx} + +\usepackage[T1]{fontenc} +%\usepackage[latin1]{inputenc} +\usepackage{ucs} +\usepackage[utf8x]{inputenc} + +\usepackage{lmodern} % Latin Modern fonts derived from Computer Modern + +% Hyperlinks in PDF: +\definecolor{linkcolor}{rgb}{0,0,0.4} +\usepackage{hyperref} +\hypersetup{ + breaklinks=true, + colorlinks=true, + linkcolor=linkcolor, + urlcolor=linkcolor, + citecolor=black, + filecolor=black, + %filecolor=blue, + pdfmenubar=true, + pdftoolbar=true, + bookmarksdepth=3 % Uncomment (and tweak) for PDF bookmarks with more levels than the TOC + } +%\hyperbaseurl{} % hyperlinks are relative to this root + +\setcounter{tocdepth}{2} % levels in table of contents + +% --- fancyhdr package for fancy headers --- +\usepackage{fancyhdr} +\fancyhf{} % sets both header and footer to nothing +\renewcommand{\headrulewidth}{0pt} +\fancyfoot[LE,RO]{\thepage} +% Ensure copyright on titlepage (article style) and chapter pages (book style) +\fancypagestyle{plain}{ + \fancyhf{} + \fancyfoot[C]{{\footnotesize \copyright\ 1999-2019, "Data Analysis and Machine Learning FYS-STK3155/FYS4155":"http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html". Released under CC Attribution-NonCommercial 4.0 license}} +% \renewcommand{\footrulewidth}{0mm} + \renewcommand{\headrulewidth}{0mm} +} +% Ensure copyright on titlepages with \thispagestyle{empty} +\fancypagestyle{empty}{ + \fancyhf{} + \fancyfoot[C]{{\footnotesize \copyright\ 1999-2019, "Data Analysis and Machine Learning FYS-STK3155/FYS4155":"http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html". Released under CC Attribution-NonCommercial 4.0 license}} + \renewcommand{\footrulewidth}{0mm} + \renewcommand{\headrulewidth}{0mm} +} + +\pagestyle{fancy} + + +% prevent orhpans and widows +\clubpenalty = 10000 +\widowpenalty = 10000 + +% --- end of standard preamble for documents --- + + +% insert custom LaTeX commands... + +\raggedbottom +\makeindex +\usepackage[totoc]{idxlayout} % for index in the toc +\usepackage[nottoc]{tocbibind} % for references/bibliography in the toc + +%-------------------- end preamble ---------------------- + +\begin{document} + +% matching end for #ifdef PREAMBLE + +\newcommand{\exercisesection}[1]{\subsection*{#1}} + + +% ------------------- main content ---------------------- + + + +% ----------------- title ------------------------- + +\thispagestyle{empty} + +\begin{center} +{\LARGE\bf +\begin{spacing}{1.25} +Project 2 on Machine Learning, deadline November 8 +\end{spacing} +} +\end{center} + +% ----------------- author(s) ------------------------- + +\begin{center} +{\bf \href{{http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html}}{Data Analysis and Machine Learning FYS-STK3155/FYS4155}} +\end{center} + + \begin{center} +% List of all institutions: +\centerline{{\small Department of Physics, University of Oslo, Norway}} +\end{center} + +% ----------------- end author(s) ------------------------- + +% --- begin date --- +\begin{center} +Oct 2, 2019 +\end{center} +% --- end date --- + +\vspace{1cm} + + +\subsection*{Classification and Regression, from linear and logistic regression to neural networks} + +The main aim of this project is to study both classification and +regression problems, where we can reuse the regression algorithms studied +in project 1. We will include logistic regresion for classification +problems and write our own multilayer perceptron code for studying +both regression and classification problems. The codes developed in +project 1, including bootstrap and/or cross-validation as well as the +computation of the mean-squared error and/or the $R2$ or the accuracy score (classification problems) functions can +also be utilized in the present analysis. + + +The data sets that we propose here are (the default sets) + +\begin{itemize} +\item Regression (fitting a continuous function). In this part you will need to bring up your results from project 1 and compare these with what you get from you Neural Network code to be developed here. The data sets could be +\begin{enumerate} + + \item Either the Franke function or the terrain data from project 1, or data sets your propose. + +\end{enumerate} + +\noindent +\item Classification. Here you will also need to develop a Logistic regression code that you will use to compare with the Neural Network code. The data set we propose are +\begin{enumerate} + + \item The credit card data set from \href{{https://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients}}{UCI}. This data set links to a \href{{https://bradzzz.gitbooks.io/ga-seattle-dsi/content/dsi/dsi_05_classification_databases/2.1-lesson/assets/datasets/DefaultCreditCardClients_yeh_2009.pdf}}{recent scientific article}. Furthermore, in the lecture slides on \href{{https://compphysics.github.io/MachineLearning/doc/pub/LogReg/html/LogReg.html}}{Logistic Regression}, you will find an example code for the Credit Card data. You could use this code as an example on how to read the data and use \textbf{Scikit-Learn} to run a classification problem. +\end{enumerate} + +\noindent +\end{itemize} + +\noindent +However, if you would like to study other data sets, feel free to +propose other sets. What we listed here are mere suggestions from our +side. If you opt for another data set, consider using a set which +has been studied in the scientific literature. This makes it easier +for you to compare and analyze your results. It is also an essential +elements of the scientific discussion. + +\paragraph{Part a): Write your Logistic Regression code, first step.} +If you opt for the credit card data, your first task is to familiarize yourself with the data set and the scientific article. +We recommend also that you study the code example in the \href{{https://compphysics.github.io/MachineLearning/doc/pub/LogReg/html/LogReg.html}}{Logistic Regression}. + +Write the part of the code which reads in the data and sets up the relevant data sets. + +\paragraph{Part b): Write your Logistic Regression code, second step.} +Now you should write your Logistic Regression code with the aim to +reproduce the Logistic Regression analysis of the \href{{https://bradzzz.gitbooks.io/ga-seattle-dsi/content/dsi/dsi_05_classification_databases/2.1-lesson/assets/datasets/DefaultCreditCardClients_yeh_2009.pdf}}{scientific +article}. + +Define your cost function and the design matrix before you start writing your code. + +In order to find the optimal parameters of your logistic regressor you +should include a gradient descent solver, as discussed in the +\href{{https://compphysics.github.io/MachineLearning/doc/pub/Splines/html/Splines-bs.html}}{gradient descent +lectures}. +Since we don't have so many data points, you may just code the +standard gradient descent with a given learning rate, or even attempt +to use the Newton-Raphson method. Alternatively, it may be useful for +the next part on neural networks to implement a stochastic gradient +descent with and without mini-batches. Stochastic gradient with +mini-batches may give the best results. You could finally compare your +code with the output from \textbf{scikit-learn}'s toolbox for optimization +methods applied to logistic regression. + + +To measure the performance of our classification problem we use the +so-called \emph{accuracy} score. The accuracy is as you would expect just +the number of correctly guessed targets $t_i$ divided by the total +number of targets. A perfect classifier will have an accuracy score of +$1$. + +\[ +\text{Accuracy} = \frac{\sum_{i=1}^n I(t_i = y_i)}{n} , +\] + +where $I$ is the indicator function, $1$ if $t_i = y_i$ and $0$ +otherwise if we have a binary classifcation problem. Here $t_i$ +represents the target and $y_i$ the outputs of your Logistic +Regression code. + + +You can compare your own results with those obtained using +\textbf{scikit-learn}. + + +\paragraph{Part c): Writing your own Neural Network code.} +Your aim now, and this is the central part of this project, is to +write to your own Feed Forward Neural Network code implementing the back +propagation algorithm discussed in the \href{{https://compphysics.github.io/MachineLearning/doc/pub/NeuralNet/html/NeuralNet-bs.html}}{lecture +slides}. We +start with the Logistic Regression case and the data set discussed in parts a) and b) but train +now the network to find the optimal weights and biases. You are free +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}. + +Comment your results and give a critical discussion of the results +obtained with the Logistic Regression code and your own Neural Network +code. Make an analysis of the regularization parameters and the learning rates employed to find the optimal accurary score. + +A useful reference on the back progagation algorithm is \href{{http://neuralnetworksanddeeplearning.com/}}{Nielsen's +book}. It is an excellent +read. + + +\paragraph{Part d): Regression analysis using neural networks.} +Here we will change the cost function for our neural network code +developed in part c) in order to perform a regression (fitting a +function or some data set) analysis. As stated above, our default data +sets could be either the Franke function or the terrain data from +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. + +\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. + + + + +\subsection*{Background literature} + +\begin{enumerate} +\item The text of Michael Nielsen is highly recommended, see \href{{http://neuralnetworksanddeeplearning.com/}}{Nielsen's book}. It is an excellent read. + +\item The textbook of \href{{https://www.springer.com/gp/book/9780387848570}}{Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer}, chapters 3 and 7 are the most relevant ones for the analysis here. + +\item \href{{https://arxiv.org/abs/1803.08823}}{Mehta et al, arXiv 1803.08823}, \emph{A high-bias, low-variance introduction to Machine Learning for physicists}, ArXiv:1803.08823. +\end{enumerate} + +\noindent +\subsection*{Introduction to numerical projects} + +Here follows a brief recipe and recommendation on how to write a report for each +project. + +\begin{itemize} + \item Give a short description of the nature of the problem and the eventual numerical methods you have used. + + \item Describe the algorithm you have used and/or developed. Here you may find it convenient to use pseudocoding. In many cases you can describe the algorithm in the program itself. + + \item Include the source code of your program. Comment your program properly. + + \item If possible, try to find analytic solutions, or known limits in order to test your program when developing the code. + + \item Include your results either in figure form or in a table. Remember to label your results. All tables and figures should have relevant captions and labels on the axes. + + \item Try to evaluate the reliabilty and numerical stability/precision of your results. If possible, include a qualitative and/or quantitative discussion of the numerical stability, eventual loss of precision etc. + + \item Try to give an interpretation of you results in your answers to the problems. + + \item Critique: if possible include your comments and reflections about the exercise, whether you felt you learnt something, ideas for improvements and other thoughts you've made when solving the exercise. We wish to keep this course at the interactive level and your comments can help us improve it. + + \item Try to establish a practice where you log your work at the computerlab. You may find such a logbook very handy at later stages in your work, especially when you don't properly remember what a previous test version of your program did. Here you could also record the time spent on solving the exercise, various algorithms you may have tested or other topics which you feel worthy of mentioning. +\end{itemize} + +\noindent +\subsection*{Format for electronic delivery of report and programs} + +The preferred format for the report is a PDF file. You can also use DOC or postscript formats or as an ipython notebook file. As programming language we prefer that you choose between C/C++, Fortran2008 or Python. The following prescription should be followed when preparing the report: + +\begin{itemize} + \item Use Devilry to hand in your projects, log in at \href{{http://devilry.ifi.uio.no}}{\nolinkurl{http://devilry.ifi.uio.no}} with your normal UiO username and password and choose either 'fysstk3155' or 'fysstk4155'. There you can load up the files within the deadline. + + \item Upload \textbf{only} the report file! For the source code file(s) you have developed please provide us with your link to your github domain. The report file should include all of your discussions and a list of the codes you have developed. Do not include library files which are available at the course homepage, unless you have made specific changes to them. + + \item In your git repository, please include a folder which contains selected results. These can be in the form of output from your code for a selected set of runs and input parameters. + + \item In this and all later projects, you should include tests (for example unit tests) of your code(s). + + \item Comments from us on your projects, approval or not, corrections to be made etc can be found under your Devilry domain and are only visible to you and the teachers of the course. +\end{itemize} + +\noindent +Finally, +we encourage you to collaborate. Optimal working groups consist of +2-3 students. You can then hand in a common report. + + + +\subsection*{Software and needed installations} + +If you have Python installed (we recommend Python3) and you feel pretty familiar with installing different packages, +we recommend that you install the following Python packages via \textbf{pip} as +\begin{enumerate} +\item pip install numpy scipy matplotlib ipython scikit-learn tensorflow sympy pandas pillow +\end{enumerate} + +\noindent +For Python3, replace \textbf{pip} with \textbf{pip3}. + +See below for a discussion of \textbf{tensorflow} and \textbf{scikit-learn}. + +For OSX users we recommend also, after having installed Xcode, to install \textbf{brew}. Brew allows +for a seamless installation of additional software via for example +\begin{enumerate} +\item brew install python3 +\end{enumerate} + +\noindent +For Linux users, with its variety of distributions like for example the widely popular Ubuntu distribution +you can use \textbf{pip} as well and simply install Python as +\begin{enumerate} +\item sudo apt-get install python3 (or python for python2.7) +\end{enumerate} + +\noindent +etc etc. + +If you don't want to install various Python packages with their dependencies separately, we recommend two widely used distrubutions which set up all relevant dependencies for Python, namely +\begin{enumerate} +\item \href{{https://docs.anaconda.com/}}{Anaconda} Anaconda is an open source distribution of the Python and R programming languages for large-scale data processing, predictive analytics, and scientific computing, that aims to simplify package management and deployment. Package versions are managed by the package management system \textbf{conda} + +\item \href{{https://www.enthought.com/product/canopy/}}{Enthought canopy} is a Python distribution for scientific and analytic computing distribution and analysis environment, available for free and under a commercial license. +\end{enumerate} + +\noindent +Popular software packages written in Python for ML are + +\begin{itemize} +\item \href{{http://scikit-learn.org/stable/}}{Scikit-learn}, + +\item \href{{https://www.tensorflow.org/}}{Tensorflow}, + +\item \href{{http://pytorch.org/}}{PyTorch} and + +\item \href{{https://keras.io/}}{Keras}. +\end{itemize} + +\noindent +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. + + + + + + + + + + + + + + + + +\begin{enumerate} +\item For project 3, you should feel free to use your own codes from projects 1 and 2, eventually write your own for SVMs and/or Decision trees and random forests' or use the available functionality of \textbf{scikit-learn}, \textbf{tensorflow}, etc. + +\item The estimates you used and tested in projects 1 and 2 should also be included, that is the $R2$-score, \textbf{MSE}, cross-validation and/or bootstrap if these are relevant. + +\item If possible, you should link the data sets with exisiting research and analyses thereof. Scientific articles which have used Machine Learning algorithms to analyze the data are highly welcome. Perhaps you can improve previous analyses and even publish a new article? + +\item A critical assessment of the methods with ditto perspectives and recommendations is also something you need to include. +\end{enumerate} + +\noindent + +% ------------------- end of main content --------------- + +\end{document} + diff --git a/doc/src/Projects/2019/Project2/Project2.do.txt b/doc/src/Projects/2019/Project2/Project2.do.txt new file mode 100644 index 000000000..f29d1dd00 --- /dev/null +++ b/doc/src/Projects/2019/Project2/Project2.do.txt @@ -0,0 +1,239 @@ +TITLE: Project 2 on Machine Learning, deadline November 8 +AUTHOR: "Data Analysis and Machine Learning FYS-STK3155/FYS4155":"http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html" {copyright, 1999-present|CC BY-NC} at Department of Physics, University of Oslo, Norway +DATE: today + + +===== Classification and Regression, from linear and logistic regression to neural networks ===== + +The main aim of this project is to study both classification and +regression problems, where we can reuse the regression algorithms studied +in project 1. We will include logistic regresion for classification +problems and write our own multilayer perceptron code for studying +both regression and classification problems. The codes developed in +project 1, including bootstrap and/or cross-validation as well as the +computation of the mean-squared error and/or the $R2$ or the accuracy score (classification problems) functions can +also be utilized in the present analysis. + + +The data sets that we propose here are (the default sets) + +* Regression (fitting a continuous function). In this part you will need to bring up your results from project 1 and compare these with what you get from you Neural Network code to be developed here. The data sets could be + o Either the Franke function or the terrain data from project 1, or data sets your propose. +* Classification. Here you will also need to develop a Logistic regression code that you will use to compare with the Neural Network code. The data set we propose are + o The credit card data set from "UCI":"https://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients". This data set links to a "recent scientific article":"https://bradzzz.gitbooks.io/ga-seattle-dsi/content/dsi/dsi_05_classification_databases/2.1-lesson/assets/datasets/DefaultCreditCardClients_yeh_2009.pdf". Furthermore, in the lecture slides on "Logistic Regression":"https://compphysics.github.io/MachineLearning/doc/pub/LogReg/html/LogReg.html", you will find an example code for the Credit Card data. You could use this code as an example on how to read the data and use _Scikit-Learn_ to run a classification problem. + +However, if you would like to study other data sets, feel free to +propose other sets. What we listed here are mere suggestions from our +side. If you opt for another data set, consider using a set which +has been studied in the scientific literature. This makes it easier +for you to compare and analyze your results. It is also an essential +elements of the scientific discussion. + +=== Part a): Write your Logistic Regression code, first step === + +If you opt for the credit card data, your first task is to familiarize yourself with the data set and the scientific article. +We recommend also that you study the code example in the "Logistic Regression":"https://compphysics.github.io/MachineLearning/doc/pub/LogReg/html/LogReg.html". + +Write the part of the code which reads in the data and sets up the relevant data sets. + +=== Part b): Write your Logistic Regression code, second step === + +Now you should write your Logistic Regression code with the aim to +reproduce the Logistic Regression analysis of the "scientific +article":"https://bradzzz.gitbooks.io/ga-seattle-dsi/content/dsi/dsi_05_classification_databases/2.1-lesson/assets/datasets/DefaultCreditCardClients_yeh_2009.pdf". + +Define your cost function and the design matrix before you start writing your code. + +In order to find the optimal parameters of your logistic regressor you +should include a gradient descent solver, as discussed in the +"gradient descent +lectures":"https://compphysics.github.io/MachineLearning/doc/pub/Splines/html/Splines-bs.html". +Since we don't have so many data points, you may just code the +standard gradient descent with a given learning rate, or even attempt +to use the Newton-Raphson method. Alternatively, it may be useful for +the next part on neural networks to implement a stochastic gradient +descent with and without mini-batches. Stochastic gradient with +mini-batches may give the best results. You could finally compare your +code with the output from _scikit-learn_'s toolbox for optimization +methods applied to logistic regression. + + +To measure the performance of our classification problem we use the +so-called *accuracy* score. The accuracy is as you would expect just +the number of correctly guessed targets $t_i$ divided by the total +number of targets. A perfect classifier will have an accuracy score of +$1$. + +!bt +\[ +\text{Accuracy} = \frac{\sum_{i=1}^n I(t_i = y_i)}{n} , +\] +!et + +where $I$ is the indicator function, $1$ if $t_i = y_i$ and $0$ +otherwise if we have a binary classifcation problem. Here $t_i$ +represents the target and $y_i$ the outputs of your Logistic +Regression code. + + +You can compare your own results with those obtained using +_scikit-learn_. + + +=== Part c): Writing your own Neural Network code === + +Your aim now, and this is the central part of this project, is to +write to your own Feed Forward Neural Network code implementing the back +propagation algorithm discussed in the "lecture +slides":"https://compphysics.github.io/MachineLearning/doc/pub/NeuralNet/html/NeuralNet-bs.html". We +start with the Logistic Regression case and the data set discussed in parts a) and b) but train +now the network to find the optimal weights and biases. You are free +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_. + +Comment your results and give a critical discussion of the results +obtained with the Logistic Regression code and your own Neural Network +code. Make an analysis of the regularization parameters and the learning rates employed to find the optimal accurary score. + +A useful reference on the back progagation algorithm is "Nielsen's +book":"http://neuralnetworksanddeeplearning.com/". It is an excellent +read. + + +=== Part d): Regression analysis using neural networks === + +Here we will change the cost function for our neural network code +developed in part c) in order to perform a regression (fitting a +function or some data set) analysis. As stated above, our default data +sets could be either the Franke function or the terrain data from +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. + +=== 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. + + + + +===== Background literature ===== + +o The text of Michael Nielsen is highly recommended, see "Nielsen's book":"http://neuralnetworksanddeeplearning.com/". It is an excellent read. + +o The textbook of "Trevor Hastie, Robert Tibshirani, Jerome H. Friedman, The Elements of Statistical Learning, Springer":"https://www.springer.com/gp/book/9780387848570", chapters 3 and 7 are the most relevant ones for the analysis here. + +o "Mehta et al, arXiv 1803.08823":"https://arxiv.org/abs/1803.08823", *A high-bias, low-variance introduction to Machine Learning for physicists*, ArXiv:1803.08823. + + + + +===== Introduction to numerical projects ===== + +Here follows a brief recipe and recommendation on how to write a report for each +project. + + * Give a short description of the nature of the problem and the eventual numerical methods you have used. + + * Describe the algorithm you have used and/or developed. Here you may find it convenient to use pseudocoding. In many cases you can describe the algorithm in the program itself. + + * Include the source code of your program. Comment your program properly. + + * If possible, try to find analytic solutions, or known limits in order to test your program when developing the code. + + * Include your results either in figure form or in a table. Remember to label your results. All tables and figures should have relevant captions and labels on the axes. + + * Try to evaluate the reliabilty and numerical stability/precision of your results. If possible, include a qualitative and/or quantitative discussion of the numerical stability, eventual loss of precision etc. + + * Try to give an interpretation of you results in your answers to the problems. + + * Critique: if possible include your comments and reflections about the exercise, whether you felt you learnt something, ideas for improvements and other thoughts you've made when solving the exercise. We wish to keep this course at the interactive level and your comments can help us improve it. + + * Try to establish a practice where you log your work at the computerlab. You may find such a logbook very handy at later stages in your work, especially when you don't properly remember what a previous test version of your program did. Here you could also record the time spent on solving the exercise, various algorithms you may have tested or other topics which you feel worthy of mentioning. + + + + + +===== Format for electronic delivery of report and programs ===== + +The preferred format for the report is a PDF file. You can also use DOC or postscript formats or as an ipython notebook file. As programming language we prefer that you choose between C/C++, Fortran2008 or Python. The following prescription should be followed when preparing the report: + + * Use Devilry to hand in your projects, log in at URL:"http://devilry.ifi.uio.no" with your normal UiO username and password and choose either 'fysstk3155' or 'fysstk4155'. There you can load up the files within the deadline. + + * Upload _only_ the report file! For the source code file(s) you have developed please provide us with your link to your github domain. The report file should include all of your discussions and a list of the codes you have developed. Do not include library files which are available at the course homepage, unless you have made specific changes to them. + + * In your git repository, please include a folder which contains selected results. These can be in the form of output from your code for a selected set of runs and input parameters. + + * In this and all later projects, you should include tests (for example unit tests) of your code(s). + + * Comments from us on your projects, approval or not, corrections to be made etc can be found under your Devilry domain and are only visible to you and the teachers of the course. + + + +Finally, +we encourage you to collaborate. Optimal working groups consist of +2-3 students. You can then hand in a common report. + + + +===== Software and needed installations ===== + +If you have Python installed (we recommend Python3) and you feel pretty familiar with installing different packages, +we recommend that you install the following Python packages via _pip_ as +o pip install numpy scipy matplotlib ipython scikit-learn tensorflow sympy pandas pillow +For Python3, replace _pip_ with _pip3_. + +See below for a discussion of _tensorflow_ and _scikit-learn_. + +For OSX users we recommend also, after having installed Xcode, to install _brew_. Brew allows +for a seamless installation of additional software via for example +o brew install python3 + +For Linux users, with its variety of distributions like for example the widely popular Ubuntu distribution +you can use _pip_ as well and simply install Python as +o sudo apt-get install python3 (or python for python2.7) +etc etc. + +If you don't want to install various Python packages with their dependencies separately, we recommend two widely used distrubutions which set up all relevant dependencies for Python, namely +o "Anaconda":"https://docs.anaconda.com/" Anaconda is an open source distribution of the Python and R programming languages for large-scale data processing, predictive analytics, and scientific computing, that aims to simplify package management and deployment. Package versions are managed by the package management system _conda_ +o "Enthought canopy":"https://www.enthought.com/product/canopy/" is a Python distribution for scientific and analytic computing distribution and analysis environment, available for free and under a commercial license. + +Popular software packages written in Python for ML are + +* "Scikit-learn":"http://scikit-learn.org/stable/", +* "Tensorflow":"https://www.tensorflow.org/", +* "PyTorch":"http://pytorch.org/" and +* "Keras":"https://keras.io/". +These are all freely available at their respective GitHub sites. They +encompass communities of developers in the thousands or more. And the number +of code developers and contributors keeps increasing. + + + + + + + + + + + + + + + + +o For project 3, you should feel free to use your own codes from projects 1 and 2, eventually write your own for SVMs and/or Decision trees and random forests' or use the available functionality of _scikit-learn_, _tensorflow_, etc. + +o The estimates you used and tested in projects 1 and 2 should also be included, that is the $R2$-score, _MSE_, cross-validation and/or bootstrap if these are relevant. + +o If possible, you should link the data sets with exisiting research and analyses thereof. Scientific articles which have used Machine Learning algorithms to analyze the data are highly welcome. Perhaps you can improve previous analyses and even publish a new article? + +o A critical assessment of the methods with ditto perspectives and recommendations is also something you need to include. + diff --git a/doc/src/Projects/2019/Project2/clean.sh b/doc/src/Projects/2019/Project2/clean.sh new file mode 100755 index 000000000..2e5da2c72 --- /dev/null +++ b/doc/src/Projects/2019/Project2/clean.sh @@ -0,0 +1,3 @@ +#!/bin/sh +doconce clean +rm -rf *.pdf *.tex ipynb*.tar.gz *.html ._*.html *~ reveal.js Trash README.txt diff --git a/doc/src/Projects/2019/Project2/make.sh b/doc/src/Projects/2019/Project2/make.sh new file mode 100755 index 000000000..e1fc0d6fa --- /dev/null +++ b/doc/src/Projects/2019/Project2/make.sh @@ -0,0 +1,79 @@ +#!/bin/sh +set -x + +function system { + "$@" + if [ $? -ne 0 ]; then + echo "make.sh: unsuccessful command $@" + echo "abort!" + exit 1 + fi +} + +if [ $# -eq 0 ]; then +echo 'bash make.sh slides1|slides2' +exit 1 +fi + +name=$1 +rm -f *.tar.gz + +opt="--encoding=utf-8" +opt= + +rm -f *.aux + + + +# Plain HTML documents +html=${name} +system doconce format html $name --pygments_html_style=default --html_style=bloodish --html_links_in_new_window --html_output=$html $opt +system doconce split_html $html.html --method=space10 + +# Bootstrap style +html=${name}-bs +system doconce format html $name --html_style=bootstrap --pygments_html_style=default --html_admon=bootstrap_panel --html_output=$html $opt +system doconce split_html $html.html --method=split --pagination --nav_button=bottom + + +# Ordinary plain LaTeX document +system doconce format pdflatex $name --print_latex_style=trac --latex_admon=paragraph $opt +system doconce ptex2tex $name envir=print +# Add special packages +doconce subst "% Add user's preamble" "\g<1>\n\\usepackage{simplewick}" $name.tex +doconce replace 'section{' 'section*{' $name.tex +pdflatex -shell-escape $name +pdflatex -shell-escape $name +mv -f $name.pdf ${name}.pdf +cp $name.tex ${name}.tex + +# Publish +dest=../../../../Projects/2019 +if [ ! -d $dest/$name ]; then +mkdir $dest/$name +mkdir $dest/$name/pdf +mkdir $dest/$name/html +mkdir $dest/$name/ipynb +fi +cp ${name}*.tex $dest/$name/pdf +cp ${name}*.pdf $dest/$name/pdf +cp -r ${name}*.html ._${name}*.html $dest/$name/html + +# Figures: cannot just copy link, need to physically copy the files +if [ -d fig-${name} ]; then +if [ ! -d $dest/$name/html/fig-$name ]; then +mkdir $dest/$name/html/fig-$name +fi +cp -r fig-${name}/* $dest/$name/html/fig-$name +fi + +cp ${name}.ipynb $dest/$name/ipynb +ipynb_tarfile=ipynb-${name}-src.tar.gz +if [ ! -f ${ipynb_tarfile} ]; then +cat > README.txt <