From de861c62fa88619631037bff5500c571c8add43f Mon Sep 17 00:00:00 2001
From: mhjensen <morten.hjorth-jensen@fys.uio.no>
Date: Tue, 1 Sep 2020 22:36:17 +0200
Subject: [PATCH] cleaning up typos and inconsistencies in project 1

---
 .../2020/Project1/html/._Project1-bs000.html  |  69 ++++++++++++------
 .../2020/Project1/html/Project1-bs.html       |  69 ++++++++++++------
 doc/Projects/2020/Project1/html/Project1.html |  65 +++++++++++------
 .../Project1/ipynb/ipynb-Project1-src.tar.gz  | Bin 192 -> 192 bytes
 doc/Projects/2020/Project1/pdf/Project1.p.tex |  49 ++++++++-----
 doc/Projects/2020/Project1/pdf/Project1.pdf   | Bin 281744 -> 282485 bytes
 doc/Projects/2020/Project1/pdf/Project1.tex   |  51 ++++++++-----
 .../Projects/2020/Project1/Project1.do.txt    |  49 ++++++++-----
 doc/src/Projects/2020/Project1/make.sh        |   2 +-
 9 files changed, 231 insertions(+), 123 deletions(-)

diff --git a/doc/Projects/2020/Project1/html/._Project1-bs000.html b/doc/Projects/2020/Project1/html/._Project1-bs000.html
index 4f89ee7c5..44db7bb78 100644
--- a/doc/Projects/2020/Project1/html/._Project1-bs000.html
+++ b/doc/Projects/2020/Project1/html/._Project1-bs000.html
@@ -45,12 +45,14 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec0'),
-              ('Part a): Ordinary Least Square on the Franke function  with '
-               'resampling',
+              ('Part a): Ordinary Least Square (OLS) on the Franke function',
                3,
                None,
                '___sec1'),
-              ('Part b): Bias-variance trade-off', 3, None, '___sec2'),
+              ('Part b): Bias-variance trade-off and resamplng techniques',
+               3,
+               None,
+               '___sec2'),
               ('Part c) Cross-validation as resampling techniques, adding more '
                'complexity',
                3,
@@ -119,8 +121,8 @@ MathJax.Hub.Config({
         <a href="#" class="dropdown-toggle" data-toggle="dropdown">Contents <b class="caret"></b></a>
         <ul class="dropdown-menu">
      <!-- navigation toc: --> <li><a href="#___sec0" style="font-size: 80%;"><b>Regression analysis and resampling methods</b></a></li>
-     <!-- navigation toc: --> <li><a href="#___sec1" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part a): Ordinary Least Square on the Franke function  with resampling</a></li>
-     <!-- navigation toc: --> <li><a href="#___sec2" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part b): Bias-variance trade-off</a></li>
+     <!-- navigation toc: --> <li><a href="#___sec1" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part a): Ordinary Least Square (OLS) on the Franke function</a></li>
+     <!-- navigation toc: --> <li><a href="#___sec2" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part b): Bias-variance trade-off and resamplng techniques</a></li>
      <!-- navigation toc: --> <li><a href="#___sec3" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part c) Cross-validation as resampling techniques, adding more complexity</a></li>
      <!-- navigation toc: --> <li><a href="#___sec4" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part d): Ridge Regression on the Franke function  with resampling</a></li>
      <!-- navigation toc: --> <li><a href="#___sec5" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part e): Lasso Regression on the Franke function  with resampling</a></li>
@@ -173,8 +175,10 @@ MathJax.Hub.Config({
 <p>
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+<p>
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 <p>
 We will first study how to fit polynomials to a specific
@@ -218,7 +222,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 <p>
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 <p>
 
 <!-- code=python (!bc pycod) typeset with pygments style "default" -->
@@ -263,7 +267,7 @@ fig<span style="color: #666666">.</span>colorbar(surf, shrink<span style="color:
 plt<span style="color: #666666">.</span>show()
 </pre></div>
 
-<h3 id="___sec1" class="anchor">Part a): Ordinary Least Square on the Franke function  with resampling </h3>
+<h3 id="___sec1" class="anchor">Part a): Ordinary Least Square (OLS) on the Franke function </h3>
 
 <p>
 We will generate our own dataset for a function
@@ -277,7 +281,7 @@ distribution \( \cal{N}(0,1) \).
 value decomposition from e.g., <b>numpy</b> ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in \( x \) and \( y \) up to fifth order. Find the
-confidence intervals of the parameters \( \beta \) by computing their
+<a href="https://en.wikipedia.org/wiki/Confidence_interval" target="_self">confidence intervals</a> of the parameters (estimators) \( \beta \) by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 $$ MSE(\hat{y},\hat{\tilde{y}}) = \frac{1}{n}
@@ -301,9 +305,14 @@ $$
 $$
 
 <p>
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library <b>Scikit-Learn</b> (make sure you have installed it).
-This function is called \( train\_test\_split \). Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of <b>Scikit-Learn</b>.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_self">homework set 2</a> for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library <b>Scikit-Learn</b> (make
+sure you have installed it).  This function is called
+\( train\_test\_split \). Similarly, and see the solution to <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_self">homework set 2, exercise 2</a>, you can use the data normalization functionality of
+<b>Scikit-Learn</b>.
 
 <p>
 It is normal in essentially all Machine Learning studies to split the
@@ -313,7 +322,7 @@ is no explicit recipe for how much data should be included as training
 data and say test data.  An accepted rule of thumb is to use
 approximately \( 2/3 \) to \( 4/5 \) of the data as training data.
 
-<h3 id="___sec2" class="anchor">Part b): Bias-variance trade-off </h3>
+<h3 id="___sec2" class="anchor">Part b): Bias-variance trade-off and resamplng techniques </h3>
 
 <p>
 Our aim here is to study the bias-variance trade-off by implementing the <b>bootstrap</b> resampling technique.
@@ -325,6 +334,16 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms.
 
+<p>
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+<p>
+With this result we move on to the bias-variance trade-off analysis.
+
 <p>
 Consider a
 dataset \( \mathcal{L} \) consisting of the data
@@ -355,6 +374,8 @@ $$
 C(\boldsymbol{X},\boldsymbol{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\boldsymbol{y}-\boldsymbol{\tilde{y}})^2\right].
 $$
 
+Here the expected value \( \mathbb{E} \) is the sample value.
+
 <p>
 Show that you can rewrite  this as
 $$
@@ -365,25 +386,25 @@ $$
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+<p>
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
+
 <p>
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the <b>bootstrap</b> resampling method.
 
 <p>
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
-
-<p>
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the <b>bootstra</b> resampling technique.
+Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
 
 <h3 id="___sec3" class="anchor">Part c) Cross-validation as resampling techniques, adding more complexity </h3>
 
 <p>
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -391,7 +412,7 @@ data and test data using for example
 Implement the \( k \)-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-<b>Scikit-Learn</b> if needed. You can alternatively write your own bootstrap code.
+<b>Scikit-Learn</b> if needed.
 
 <p>
 Compare the MSE you get from your cross-validation code with the one you got from your <b>bootstrap</b> code.
diff --git a/doc/Projects/2020/Project1/html/Project1-bs.html b/doc/Projects/2020/Project1/html/Project1-bs.html
index 4f89ee7c5..44db7bb78 100644
--- a/doc/Projects/2020/Project1/html/Project1-bs.html
+++ b/doc/Projects/2020/Project1/html/Project1-bs.html
@@ -45,12 +45,14 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec0'),
-              ('Part a): Ordinary Least Square on the Franke function  with '
-               'resampling',
+              ('Part a): Ordinary Least Square (OLS) on the Franke function',
                3,
                None,
                '___sec1'),
-              ('Part b): Bias-variance trade-off', 3, None, '___sec2'),
+              ('Part b): Bias-variance trade-off and resamplng techniques',
+               3,
+               None,
+               '___sec2'),
               ('Part c) Cross-validation as resampling techniques, adding more '
                'complexity',
                3,
@@ -119,8 +121,8 @@ MathJax.Hub.Config({
         <a href="#" class="dropdown-toggle" data-toggle="dropdown">Contents <b class="caret"></b></a>
         <ul class="dropdown-menu">
      <!-- navigation toc: --> <li><a href="#___sec0" style="font-size: 80%;"><b>Regression analysis and resampling methods</b></a></li>
-     <!-- navigation toc: --> <li><a href="#___sec1" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part a): Ordinary Least Square on the Franke function  with resampling</a></li>
-     <!-- navigation toc: --> <li><a href="#___sec2" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part b): Bias-variance trade-off</a></li>
+     <!-- navigation toc: --> <li><a href="#___sec1" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part a): Ordinary Least Square (OLS) on the Franke function</a></li>
+     <!-- navigation toc: --> <li><a href="#___sec2" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part b): Bias-variance trade-off and resamplng techniques</a></li>
      <!-- navigation toc: --> <li><a href="#___sec3" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part c) Cross-validation as resampling techniques, adding more complexity</a></li>
      <!-- navigation toc: --> <li><a href="#___sec4" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part d): Ridge Regression on the Franke function  with resampling</a></li>
      <!-- navigation toc: --> <li><a href="#___sec5" style="font-size: 80%;">&nbsp;&nbsp;&nbsp;Part e): Lasso Regression on the Franke function  with resampling</a></li>
@@ -173,8 +175,10 @@ MathJax.Hub.Config({
 <p>
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+<p>
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 <p>
 We will first study how to fit polynomials to a specific
@@ -218,7 +222,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 <p>
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 <p>
 
 <!-- code=python (!bc pycod) typeset with pygments style "default" -->
@@ -263,7 +267,7 @@ fig<span style="color: #666666">.</span>colorbar(surf, shrink<span style="color:
 plt<span style="color: #666666">.</span>show()
 </pre></div>
 
-<h3 id="___sec1" class="anchor">Part a): Ordinary Least Square on the Franke function  with resampling </h3>
+<h3 id="___sec1" class="anchor">Part a): Ordinary Least Square (OLS) on the Franke function </h3>
 
 <p>
 We will generate our own dataset for a function
@@ -277,7 +281,7 @@ distribution \( \cal{N}(0,1) \).
 value decomposition from e.g., <b>numpy</b> ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in \( x \) and \( y \) up to fifth order. Find the
-confidence intervals of the parameters \( \beta \) by computing their
+<a href="https://en.wikipedia.org/wiki/Confidence_interval" target="_self">confidence intervals</a> of the parameters (estimators) \( \beta \) by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 $$ MSE(\hat{y},\hat{\tilde{y}}) = \frac{1}{n}
@@ -301,9 +305,14 @@ $$
 $$
 
 <p>
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library <b>Scikit-Learn</b> (make sure you have installed it).
-This function is called \( train\_test\_split \). Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of <b>Scikit-Learn</b>.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_self">homework set 2</a> for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library <b>Scikit-Learn</b> (make
+sure you have installed it).  This function is called
+\( train\_test\_split \). Similarly, and see the solution to <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_self">homework set 2, exercise 2</a>, you can use the data normalization functionality of
+<b>Scikit-Learn</b>.
 
 <p>
 It is normal in essentially all Machine Learning studies to split the
@@ -313,7 +322,7 @@ is no explicit recipe for how much data should be included as training
 data and say test data.  An accepted rule of thumb is to use
 approximately \( 2/3 \) to \( 4/5 \) of the data as training data.
 
-<h3 id="___sec2" class="anchor">Part b): Bias-variance trade-off </h3>
+<h3 id="___sec2" class="anchor">Part b): Bias-variance trade-off and resamplng techniques </h3>
 
 <p>
 Our aim here is to study the bias-variance trade-off by implementing the <b>bootstrap</b> resampling technique.
@@ -325,6 +334,16 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms.
 
+<p>
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+<p>
+With this result we move on to the bias-variance trade-off analysis.
+
 <p>
 Consider a
 dataset \( \mathcal{L} \) consisting of the data
@@ -355,6 +374,8 @@ $$
 C(\boldsymbol{X},\boldsymbol{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\boldsymbol{y}-\boldsymbol{\tilde{y}})^2\right].
 $$
 
+Here the expected value \( \mathbb{E} \) is the sample value.
+
 <p>
 Show that you can rewrite  this as
 $$
@@ -365,25 +386,25 @@ $$
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+<p>
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
+
 <p>
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the <b>bootstrap</b> resampling method.
 
 <p>
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
-
-<p>
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the <b>bootstra</b> resampling technique.
+Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
 
 <h3 id="___sec3" class="anchor">Part c) Cross-validation as resampling techniques, adding more complexity </h3>
 
 <p>
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -391,7 +412,7 @@ data and test data using for example
 Implement the \( k \)-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-<b>Scikit-Learn</b> if needed. You can alternatively write your own bootstrap code.
+<b>Scikit-Learn</b> if needed.
 
 <p>
 Compare the MSE you get from your cross-validation code with the one you got from your <b>bootstrap</b> code.
diff --git a/doc/Projects/2020/Project1/html/Project1.html b/doc/Projects/2020/Project1/html/Project1.html
index 90e66f282..432642e56 100644
--- a/doc/Projects/2020/Project1/html/Project1.html
+++ b/doc/Projects/2020/Project1/html/Project1.html
@@ -44,12 +44,14 @@ div { text-align: justify; text-justify: inter-word; }
                2,
                None,
                '___sec0'),
-              ('Part a): Ordinary Least Square on the Franke function  with '
-               'resampling',
+              ('Part a): Ordinary Least Square (OLS) on the Franke function',
                3,
                None,
                '___sec1'),
-              ('Part b): Bias-variance trade-off', 3, None, '___sec2'),
+              ('Part b): Bias-variance trade-off and resamplng techniques',
+               3,
+               None,
+               '___sec2'),
               ('Part c) Cross-validation as resampling techniques, adding more '
                'complexity',
                3,
@@ -128,8 +130,10 @@ MathJax.Hub.Config({
 <p>
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+<p>
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 <p>
 We will first study how to fit polynomials to a specific
@@ -173,7 +177,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 <p>
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 <p>
 
 <!-- code=python (!bc pycod) typeset with pygments style "default" -->
@@ -218,7 +222,7 @@ fig<span style="color: #666666">.</span>colorbar(surf, shrink<span style="color:
 plt<span style="color: #666666">.</span>show()
 </pre></div>
 
-<h3 id="___sec1">Part a): Ordinary Least Square on the Franke function  with resampling </h3>
+<h3 id="___sec1">Part a): Ordinary Least Square (OLS) on the Franke function </h3>
 
 <p>
 We will generate our own dataset for a function
@@ -232,7 +236,7 @@ distribution \( \cal{N}(0,1) \).
 value decomposition from e.g., <b>numpy</b> ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in \( x \) and \( y \) up to fifth order. Find the
-confidence intervals of the parameters \( \beta \) by computing their
+<a href="https://en.wikipedia.org/wiki/Confidence_interval" target="_blank">confidence intervals</a> of the parameters (estimators) \( \beta \) by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 $$ MSE(\hat{y},\hat{\tilde{y}}) = \frac{1}{n}
@@ -256,9 +260,14 @@ $$
 $$
 
 <p>
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library <b>Scikit-Learn</b> (make sure you have installed it).
-This function is called \( train\_test\_split \). Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of <b>Scikit-Learn</b>.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_blank">homework set 2</a> for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library <b>Scikit-Learn</b> (make
+sure you have installed it).  This function is called
+\( train\_test\_split \). Similarly, and see the solution to <a href="https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" target="_blank">homework set 2, exercise 2</a>, you can use the data normalization functionality of
+<b>Scikit-Learn</b>.
 
 <p>
 It is normal in essentially all Machine Learning studies to split the
@@ -268,7 +277,7 @@ is no explicit recipe for how much data should be included as training
 data and say test data.  An accepted rule of thumb is to use
 approximately \( 2/3 \) to \( 4/5 \) of the data as training data.
 
-<h3 id="___sec2">Part b): Bias-variance trade-off </h3>
+<h3 id="___sec2">Part b): Bias-variance trade-off and resamplng techniques </h3>
 
 <p>
 Our aim here is to study the bias-variance trade-off by implementing the <b>bootstrap</b> resampling technique.
@@ -280,6 +289,16 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms.
 
+<p>
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+<p>
+With this result we move on to the bias-variance trade-off analysis.
+
 <p>
 Consider a
 dataset \( \mathcal{L} \) consisting of the data
@@ -310,6 +329,8 @@ $$
 C(\boldsymbol{X},\boldsymbol{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\boldsymbol{y}-\boldsymbol{\tilde{y}})^2\right].
 $$
 
+Here the expected value \( \mathbb{E} \) is the sample value.
+
 <p>
 Show that you can rewrite  this as
 $$
@@ -320,25 +341,25 @@ $$
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+<p>
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
+
 <p>
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the <b>bootstrap</b> resampling method.
 
 <p>
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
-
-<p>
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the <b>bootstra</b> resampling technique.
+Note also that when you calculate the bias, in all applications you don't know the function values \( f_i \). You would hence replace them with the actual data points \( y_i \).
 
 <h3 id="___sec3">Part c) Cross-validation as resampling techniques, adding more complexity </h3>
 
 <p>
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -346,7 +367,7 @@ data and test data using for example
 Implement the \( k \)-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-<b>Scikit-Learn</b> if needed. You can alternatively write your own bootstrap code.
+<b>Scikit-Learn</b> if needed.
 
 <p>
 Compare the MSE you get from your cross-validation code with the one you got from your <b>bootstrap</b> code.
diff --git a/doc/Projects/2020/Project1/ipynb/ipynb-Project1-src.tar.gz b/doc/Projects/2020/Project1/ipynb/ipynb-Project1-src.tar.gz
index 2f9e4fabb171d1aa41717738832e6db601256c27..b2b4f085b5f81d55169e68713a8186bed5c11df2 100644
GIT binary patch
literal 192
zcmV;x06+g9iwFQzuufk91MSaC3c@fD2H>uHia9}<m_F8mE?fvjyg+JG8y}gPq+oAv
zAD}D6O%WmA=4Y5;nAxqDtL-{+cNfeC5z{z^A+t0Y6BkoGVk`s3vjtB$vc7?g^8^FH
z`A&LitrMm{r9Me*rG7itk5$!&J<}`j%s;WOm4!{-xtcU+a<U*gLvH9K4~|8u>2v`l
uNjo&b$Xlz71fe?+WvQGZEiPeW=xD^%M&YmD@jTD-zVrZ3#WFDf2mk<cB~yj~

literal 192
zcmV;x06+g9iwFQT+D%^o1MSaC3c@fD2H>uHia9|^nm*QoUAPd6c!AWWHnnDIl7hXx
zeSoeMH${Yeo1ZX0Vdju+w)-rwck8W&5Rya@OqnL}oFrV{GfGpSEDafp<FN^ZGL{ez
z$a*Wi^u}^Mu2MHb8KK_x##&W<_-8o<p7|$^wKTBH2V0{8g;o@Z7sxd>;$&$cyFnFD
uqA=4Cw7B(B3*c%1UY62IRJ`>&jc3iTCh&iKj^j9v^Q8y)_NuD@2mk;sSyzVu

diff --git a/doc/Projects/2020/Project1/pdf/Project1.p.tex b/doc/Projects/2020/Project1/pdf/Project1.p.tex
index 67151ae52..cc7e020b4 100644
--- a/doc/Projects/2020/Project1/pdf/Project1.p.tex
+++ b/doc/Projects/2020/Project1/pdf/Project1.p.tex
@@ -166,8 +166,9 @@ Sep 1, 2020
 
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 We will first study how to fit polynomials to a specific
 two-dimensional function called \href{{http://www.dtic.mil/dtic/tr/fulltext/u2/a081688.pdf}}{Franke's
@@ -206,7 +207,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 \bpycod
 from mpl_toolkits.mplot3d import Axes3D
 import matplotlib.pyplot as plt
@@ -251,7 +252,7 @@ plt.show()
 \epycod
 
 
-\paragraph{Part a): Ordinary Least Square on the Franke function  with resampling.}
+\paragraph{Part a): Ordinary Least Square (OLS) on the Franke function.}
 We will generate our own dataset for a function
 $\mathrm{FrankeFunction}(x,y)$ with $x,y \in [0,1]$. The function
 $f(x,y)$ is the Franke function. You should explore also the addition
@@ -262,7 +263,7 @@ distribution $\cal{N}(0,1)$.
 value decomposition from e.g., \textbf{numpy} ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in $x$ and $y$ up to fifth order. Find the
-confidence intervals of the parameters $\beta$ by computing their
+\href{{https://en.wikipedia.org/wiki/Confidence_interval}}{confidence intervals} of the parameters (estimators) $\beta$ by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 \[ MSE(\hat{y},\hat{\tilde{y}}) = \frac{1}{n}
@@ -283,9 +284,14 @@ where we have defined the mean value  of $\hat{y}$ as
 \bar{y} =  \frac{1}{n} \sum_{i=0}^{n - 1} y_i.
 \]
 
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library \textbf{Scikit-Learn} (make sure you have installed it).
-This function is called $train\_test\_split$. Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of \textbf{Scikit-Learn}.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also \href{{https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html}}{homework set 2} for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library \textbf{Scikit-Learn} (make
+sure you have installed it).  This function is called
+$train\_test\_split$. Similarly, and see the solution to \href{{https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html}}{homework set 2, exercise 2}, you can use the data normalization functionality of
+\textbf{Scikit-Learn}.
 
 It is normal in essentially all Machine Learning studies to split the
 data in a training set and a test set (eventually  also an additional
@@ -299,7 +305,7 @@ approximately $2/3$ to $4/5$ of the data as training data.
 
 
 
-\paragraph{Part b): Bias-variance trade-off.}
+\paragraph{Part b): Bias-variance trade-off and resamplng techniques.}
 Our aim here is to study the bias-variance trade-off by implementing the \textbf{bootstrap} resampling technique.
 
 With a code which does OLS and includes resampling techniques, 
@@ -308,6 +314,14 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms. 
 
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig.~2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+With this result we move on to the bias-variance trade-off analysis.
+
 Consider a
 dataset $\mathcal{L}$ consisting of the data
 $\mathbf{X}_\mathcal{L}=\{(y_j, \boldsymbol{x}_j), j=0\ldots n-1\}$.
@@ -332,6 +346,7 @@ squared error via the so-called cost function
 \[
 C(\bm{X},\bm{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\bm{y}-\bm{\tilde{y}})^2\right].
 \]
+Here the expected value $\mathbb{E}$ is the sample value. 
 
 Show that you can rewrite  this as
 \[
@@ -341,21 +356,21 @@ Show that you can rewrite  this as
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
 
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the \textbf{bootstrap} resampling method.
 
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig.~2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
+Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
 
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the \textbf{bootstra} resampling technique.
 
 \paragraph{Part c) Cross-validation as resampling techniques, adding more complexity.}
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -364,7 +379,7 @@ data and test data using for example
 Implement the $k$-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-\textbf{Scikit-Learn} if needed. You can alternatively write your own bootstrap code.
+\textbf{Scikit-Learn} if needed. 
 
 Compare the MSE you get from your cross-validation code with the one you got from your \textbf{bootstrap} code.
 You can also compare your own cross-validation code with the one provided by \textbf{Scikit-Learn}.
diff --git a/doc/Projects/2020/Project1/pdf/Project1.pdf b/doc/Projects/2020/Project1/pdf/Project1.pdf
index 4ce5df235b6d2e1ad87c609d4c97ee8b5a1d6645..a001b89baa930b0790ae94966a4040dc9d8c5576 100644
GIT binary patch
delta 58976
zcmZs?Lv$ugxV0Ut)3I&z6FVK-wr$&aV%xTD+qP{x=@|ce&g37QuSR=SO=?u@+V|cS
zmyY}|2RV)mjDa(89+?Vwthw#9If4v)(d)mL>JNw6=LO|h3?-{(UhHtNEWdW%;_f3O
zBY^^i1R*`|y|uGLB;pq$!9OitoR7M(b#VLW_G%q9cjyem*zFCIZQ9lCZS$rr3=R{b
zj?&fM%>xL<(^WC%r;HzXHo6mjt}RU_qx8?dbqG<i`R|p~w0!}!T67OSTFM-uy2cwz
z{#0y69?Pu)HO4rki?pleDwb~B#~Wohr+}R|7l;UAI}AhG_5f1sc?xFkJ^H1{kdQ5(
z7oK`j42v{Pi}1|k@Hw@}WZ40haO{7%4RnJi;cz|W4G;CSnmFe~m2B{?S7SN&AfB%7
zPw%^GmB*40YO6q?%0QY_6zc|Ta-I;K-=Zd^N}(Y3Xq+F6%pY#-H=Pyd&MwUf7NT}b
z^d<JK?>2HGajjNN_%|_?PszG2i^v(DtQz;tl;+GofrJ+)zSN0ChZc_mnJuZc!T8~{
zvV!1!lI3&egGai@Zfm;b{Mz=LRw1<FLFJ4K<JxB&QHH>7bq!Ps!N&);H=D#(oB2#>
zSEO!So3=pC^_v=F4Wq8fPFn0-cs5$nOeDB#3fV71V*V<OU)2i?F@Da_MXOp=@)=}n
zCL$u;9Agpqf#UW{uA9X$VF)-|@x?;U*5?rCz0Rz)cAZ}*^S}qKn~Rac^&=hCq^%zp
ziM5P^omL=sHMT_G{7@C>W{C+O=K)2_xZt3fVKx(&NHg6FE+sCuSYFoL2#Nb92n-!l
z!vs{eSjeFDX3VMXSHrAd_Wn<}^waY+I!?cySdq58QGna<3EOEbY6qih$AL{!2*%{7
zzq^-oH-kZzr%BiNMv8LzNC8YLz@}giUqRQ;F%+m$Zkz^|#*FQE6WMU|j}Wig_XO$>
zO26lB%iD<1GYE^d&>sWN`G!>~ch2$BHRB_q*w5~%RX>-H5j&X4Gwg`iD1f3tK^I$A
z?D3n}j+Tb)ZL7%n3+Uils)sBTG{>;KG(3J=zx{3_%1c$0g)MXe_D@m;hjUjLLjn%~
zCKy;B_bWMqL1Jl4p($Hka5ApyK6h-Z*+m=>Qgykl7~noM{#)qgS6kv`H2!)uSpS-f
z7JBd1V~PPe>>;U5{D%b5VVbs`<e!jLB|#EL32g`Y$Df+(j3`6Y_5w;mMI#Vs4CfV@
z?S!%@@>$_JdIJc2t-SaUvbhsqnM1BBm<6E0jT^SJK~_<?eNj~`t&l!=%)8No4PHro
zDhq1R-CLGr)Nj&#kFQmxcv)Ue_rQFDH(jBbYL}%^vVcG@i6Typ;VNIp(QU32jW_k>
zP7=W8`A}45irMjHfNBl)TKqfS>_1B-qW^AcdYt&a@HeW;?A=tanhTTPISImQ2p$NN
z^thmnqSxj2sEc|kqgo&QnSZ=IsC(E2g*R6yLvmJshuSWKuXN~Q3&F0l*rG*X%iJ4A
z3z+?|7^Zl+w}}x8Zd5$_*JMU2;McXg;Aj%R{MhnXPV-=TiV6lg_6{H^n9S6r*EnT1
zUOF8LTXtKc1lV0d`L<Y%J69WUUITGC<b=x43Mkd^C{f7^<>eDxL);hJ8&jJ0a>`zl
z8TH{S8WBzAZSloG?AHX$PL04dc!KHHz1EBu`eEtaW8BbRmZ5nB17G*1{>6{d8MlSq
z3o|Jz1T<PqJt2YyA8?MMVQ>0vxy<Up_!WvwR{sJ4Q1uH2WP$(9U92(^gaD?x3<@EP
zPF*7DnWLsYobPQ~JMyjK!-I4jBJFG~P0tuT6J})e6w|4)$lX&Y;Yw~_Le2NuAc_fL
zmwM~wyWS#ERz;$j#U#KBkQNasrxBgj(wJ(3(ATz$ITY~W7v-cr*_#XO;)gi$b?SP^
zxeVGxR_S~tGW3NsT3%Dk1OvI!&*Uk(Q!=a*o%$*bM*m#B$t@P<Q99iF2}M7|D`r@0
zy<zP=77}p31%zHr$nLT@i%A|A-S%_@4_~Kj#5h$TY<$IVWwFWF9}Y>zefEDCiX4pM
zASAMs_h%O$YiMxIc18=Ld+pd=BwXQk+=;0rN5#~P^-$A6w3(8B3j*yA&C2I#0*zQ1
z+HRe#GuL6nQ1Dv5vO;Se5*io@KVqK;45T%9He~9`KjSqd1cL#2!w!j*RG0=o-d_i)
znBbY>19NDKsRJ5z-Cp<KrRMehl#0|k*Eick+k%6A77Fccp7(A&rm_<I(Z&II#Ms%c
z%6lbchDh;La**YN^}vFE?mVVs##{zPhdG4QUsLzLk%|l}mz_<~ro*6HS$R%!hq7nJ
zxL{F@pgm>_qzjLTPBe^bR%hd7F}cSB!0<2<z$wr*_iyzdl#<0G&dYhEs1;ziXU&73
z!;;hHCy_(zHk3E&GjjA%R|O3cX3;MWtK^#*&0ti%=meu#ZGkl>c=-n%7TPpkie=OY
zu!^(4ROE_c<w3IInXh+BmT7WDDU~w6BYa-fx0rdWl726HbY(vr7*x_f8hO+ZRyKaE
zWm!0$jKxyo%6z^8h9s%M?}Jo`Jt;_d3{=P9<5E9$yjA+E?BzHTeEyzR!PeB6aap~$
zCwOHgXYsl})&q}9nC2yesE4zdwWoswc}P*B0inoZITQsRYLN~re+QRU|Dfa-92AFi
zHGd6ljy#|^VWTz`!IdpA4CJ3hI#Sq?FQA$IW^HKrnXSSGO88+G#XQk8XU$}=HNo7W
zc7N8I%KQ`2K(&R$&mmr7ZlKW7KlI4=O<=u7t>KP`GyzV((qw)Me;d+8XUOM_UY)Z{
zXvw&<_7X>v@Olrd2}6X%uF$DEJ;ofcFJ~3b&gS-x7eXH{&f?@E1Y?PHlUmyk1QB>5
zG|s@1=pAW1JEeKiX{U3ZJPA86Xasle3D_BiT*y6rTnuG~_j)b5&?KC{-q0?U8H$qX
zwF=zIR{=-#6Fbd7-0DkMU6|cguStash4P2dcIT5_NKEi5gHQ(JeMLt|ZT$$@?3+G5
zjwc}InOUeMV0Goxe0;!A1wE`|jwpORb=Nq0J5Yy=kyUJ)5Oz8)z>nCMTrB2XZ>Fy;
z_T5-pdwPbNODW-`kr<~8%LI)K$O2pKYTB56kAV62XC4ToYUBc&HeO?GP^iTMtK5$N
z;wWypdZF2+x8xm7^XMg%Qg`98&Y3+I>S2t8-(d+6OLrmgZ=u9Sd-7zPdrIeF23fbq
z6Gl<Z;r&ac*xKn~R5QCYl_Y*s+*rD=^ok?rsjV%Rh@ux`y`+zy+~z(qRtkdu<kE1w
zt^p<2`+i^9pJI!Vs>L$V&2N@3${a?uP|u8nYKWPX-ltfYAMSST4+b5m!Q*9|W}{KE
z;}S~TB*j|3nmdwlKPaf*$?G)NbabEzjDzdxLyCOzU%Gfvk(3G{x7SV)_L9Nup?_J8
z6GP3B`d(_)6v^$3>WD1z8HM$ASG7yp;{d%EF<-`4n0l9e^n{WJ7Yn7ir?sDt-+GwQ
zrl$Ant@Zf{nf<J?u5tx)NGh%w9XqStIQghp3BM@!(~sGMS>iCk&mH9U{ytv5-1N(q
z(BMW@21MSntf9rZ-y7VbrX68=XKbHrCo-a}f?0IoJkvct-{J)9K;rtc*}ervMgaXD
zrGV!-L@>T2&?Kh^yqp3qBvBH9ahU1n8K;A3Q8H;`h2)e2Y0`NbuNTJAN`5dGBa?lP
zXU<gb&=w#M&<#@b)VW1v@7W7RWc|K7PMVFfIv{Zb6pmz_Kr8~N55f|48R&5i6x3&k
znGYsiG8j6fs6yxIEr0*PPjqw%d!R?GAg6Y5&^;<8?x6SLex84#dq=@;E7FT7H2s%c
z$Q=Hc@z@g|wqU3)6NK)68z5M|Ou{8C0me#yULI$Sv0%TV%s){{+<$e@>vcQ#ew}x8
zs->pByu9r9lnqw&6pdz**wH=e1pjnaRW{r_oYWULuYp%ioA?_));UuBR_(2qW;j{4
zY163)%SjTh1L!3fjWD^Vrh1(0wy|!Fb=xSsbl#S+_Osj0@5xmXUR-o~woYcUQaVcy
zS4RX26=j=Z)gzvm)X!A1<(gx(mj8g+Su>HO64OGUCaS@sfwQo({x9cet8LmIvLX5G
z)*{KH#WykvDZGGroK62i)6$>EITLDy!3u9_pC8UUi@vVCjl&yHxHgXL2|^p^G@}a@
z3iY&#^(5R;zyoUbbm~EbrzUb6gtuyYfabFV@?1dbB88BZw5JYH8yH;CTHV_HT++n&
z=w1u6XND60Q>)JSYWxIV`y!hfr4f)!i%%3#5A9%g&14pn{T?hwwwqvg#|kc_$<Dy*
z=^qS8E7IU*^^EsCBssKIpR8q0c`tCt&BoBm8b74DgG`?;q`LdbFv+K%k_Ev+zf!(A
zWd2RcBcOeqAvlQv%D0%7LAHTdfBLTl<;&sjCrPqQG{aaIWjF|EGNW-9Wa6uZpV5-i
z1y<IV7UJ94kx)fO3>zfXZSp3G?kE`O`;#1WN+!MVJ=F{!=!SfWxOOD`?MRxm2cS)I
zx;25a{gpLY;mB;lWTiOWtW`E$uVe6S8`5@F#%{iSk7K5_ta&Mz#6_-sk+=5d@vaY#
zjp4OAZhAct6y^{738h)(V5P20jP1JAWLGh)7G5=Y1M9``T-;zv+RoNWh`v_`Xz!qt
zqTnio-q}(G0}9P0=g|&4=Qa2TmN>7|Lh*#Z6Cs-KXMC_Yi3zIMx)wycp#jm8jwi_D
zi&eENW)laM=+glTB8-jwdqxm$cnD-HlnQ?7I^uaX2(rL2`e(5GGIR>JZ9~nnTE^B+
z*{mA~7*iyce^Tu!8HdSW_|$aw-$~k!673*z_$7ULG|JG*LfWE!8ZnPa!r{L_8aBO@
zGM+5Ex&n=<F#L`KVb1fQh%Tz0&&wF+bT9P>0F-4;W(HQ#xN<j4c+wA=xH#5^^Z6fP
zU>K)DG*>|G6q=9|{Cs0&i&_GZtmdeh>VVxmbAPfeh1QTaD?P5p9WUbN%?g#a_lEdj
z&L&ReBFccCyg7(Y2N;I><j#_3j^I$riR@cB*{N*Ti9)j3nIKF`&~PVsmvfY)w_F7I
zh>_`ET@h`;*KC)K>2Hx57#|Vrv&^XYIXKPdQ>j4w7#W3Y$xP~3CZFjkMSw>6c7+6J
zAJty@{FL$EJFU+)0+*yjvMH}OYT}dPDiI^k-=i3?7ORW0R4%~gP_=HtTTBVwj*+EI
zp_dt%TV_m;;tfx>SdFW}$j7qH7SEIp(&l^^saDcUo64rlpEgXL*T_mxB1p1W)KC^)
z6(^t#$e%ai7U<Jz4Yuv^C@pg_qaZ0gs1GT`i1*$N6xv9$!dIuT*al>6AM0N=S}vz6
z81m#bKT8ubQ=~4zrcn!57uN<j>Ww~|>~pH?v!n91_&n!9R9;vZ33xb3f$>&udj?}9
z{;cjNQ5Wy;xrvWoLoo1T9ed9{^4FLK&0j!!)D(6j5UI<nUMJVyN3(s?lb#MKj?y6V
z{v_Lp6>HegwUrg2h<iv5MZrVVbMmEqprFi_pu^4m9;7L4$F%17Lf*|S!s&O=-39rX
z9JMMH`Xb2?81CTM1GLpd9w#GPG~|&oG^(xS!Yc-zAB82-2^-D8ffn<)-HVeck{!@4
z-RmoAysLFc>E_75@j(!3d>r&y20HG;he5C_G(>n74`yyMkn1hu8Q#KO)U)5hP$vGC
zw<GUwB&#=I1=)rrpEaRv1ebUtsIgRX6tYdQFleWIWIgKQ<GX2qPIVwMdCZ~(Ga4Ol
z<%%PGKjT7v$xd%uPUxo@G24cbs~LC@Otd<{<b`V3Ls^o~TQDDlJ$k6KoB8&fUnT*1
zJN8R$ZvF}ty1$+~cR_hBP|vuW9r4TLBsxS7F2Sw9TfI+;T9)0DAL~U3{`OR-bf0}h
zA#2$Esyh|GeFsZ|^69WXp?Mji%+nk|=2@7VP&mvo!uqpNF*?CR24(WlkY*U{*HgR%
z(69V{V_Zp%P0;9m%aL#Czy^=i{TPpRea!~_5%zL=wPR%dp~F#RV|;aayV3Kq$KHlB
zu{Cybb~G`t`5$FxXbH#6OvFI+KZ2VZj$Yct*38+Qh?$9j?Z0Q98mK81dpLsJGh2%k
z_Y1OW;|g>A$q*4?j}_1TaEO`Tu<F{6)IoJfGD%R9i<$Ga^|Z34<P}Sjx#b1zh=eWk
ztE{Hx{i4|Y*#Kj<r#)iuah6?8uS=#ugQ7v0yfL#geMz$@Jd>OaU2=1M+VYwAebL(>
zIl4!)x-I_EsD1Ow5@@?~lJC$<muh8e`{*-RE%$b!9`tOmnz<GBHp>nKrT#@NVmj#R
zb_~7GXSAGah|50ueuF-1x#FVX<Bj?Y_wR$L9u1-$&B(YbZLtW2NBP$YQO3WUoD_>t
z?|aqpop45h(t&k`Lp!JHf-c@E%n%%w^T{`o@W!(bPZd><FW{J758z=0abf;nZ>)!o
zJ-f}*m)oi@{ce5btA{!!_xm-sRKed`-UU&EmF;>>f(ae+G~V<lJK-TX;Spv|=5UM+
zW9G}II;PBRYk@nylbDtK#+Q%<9vsfbax^wZ?uVW&>yuwLI#2m!5r6s6o>D|Rji$Fd
z+a(J>BD~bzh=Fzy;K26HU}6t<z8RAt7Upry7YPJmbEJ4qo}z>6EeqaWM+JX&RQ%3l
zsL|+R`x*FB)WE?MTSF;St$k;vtj9ago;V2{9@+VgO*x7DoGXglcGP3xiv`9~#7ik*
zP4Xfo%y1ST>||Lowtj8;f{O=6P4pHv)Hp5OFP<gx4&V@nMUi+i&Wz{Rlq?1E0z&Cn
zF8!mFCaGa%Xd(m)P2>DDj-tpPVj1Vq69*y#C1pS|jEQLFwoZwrg*_kcw1;R_(ul<L
zSs6VVz{FuPjd|x5yI+xIE2gI2I*jXYUN{+x|8pZ-vBKHq9E3LOe5?~?{^4gesMc!d
zq3Lk#FmRB71NY=O)091!R5wi;&XPr`zY{Q}ZqTlBPCxE;b#n&0RkZ<HX*jI&o0u<5
z8>QIF&#e83ak<8Np!+ad^zF1m*ldvgI^4b3NJ*7&Ljm$drPDLhF72K1d?SL)N&3CQ
z8o|ZG>*~ld&m2j%TDsC4nBAtX4Ru5de^z8#2aLd=<Dq3DSCw?$)sv&qhtMy^C!#Fp
zuM`_jqCrM)!AGT$!pU7U&Pxk=9&{}cZA~F>R?s2naxXiZ--NZi9ut)};%sHCg7Q@z
zW?q0V6Nd%7<05?wm|VaQz%0>>&G4H$-E5q+RYTwW0W@IccIB-`r$XU0T+TA+_SGLa
z024#M<(z)8gCf9mLNkaPz4_h7cc}749NTt=<_+Mm3Dlu0U)Ai`w5x<69xMkDIt0L>
zdQEQXHIrxohdSK3Wf>~Ok|p=u%bhuLw_!K3Wu<0~l;nksIFsukOK-@})kRww3p~3w
zF<yMlzv}nX4Tifk9fVwN{gpJ>43o#1fa<IEUQ@*dA$EQ;O(%dkZ%F8-ePSf;qkj<}
z>{m>0#Qi1(t~%ajkrdig)q;%=(?&x2V))fzdDyn5kLuSg<aZHXlNZ=VR(*4z*NBU$
z_J8XniWvGn`iuI~ouHbI;{(~#Ln67{)+tAD$eU(D5@Wpnraa1v&*mE1-0aEk0U>8J
zyZ=}camAo_(^jfrbvz&Vh*S!Gn*Iyb*!<kjtX%ivb7`+MFIQ2Z{gpHF$A!w0N<#9L
z5M%$WPtJ*8ag)c}D0lU(rLQVPf`paM?^Is3?Fk=8$5hEu1PVwx*2;pKH4)@Mnbb+H
zNcc988i!OyE#HL>kEE6`FxWq$0Yq`XGl?-`t8J?_SMbZ$t=?QbJJX3*+cM<ig#q}9
zi#9`JcFiO2>HvLCtTp~PlOT%?4VVyzc|gC3&DsC%ofOfN<8Nz7V`(B|Se>Uwy+pi%
zttU!<o-)j@2E8z)L%N@&?z(vQ&0f<-jvV^u>fX^3oXN_<QFlz|If$`J4D?-S464#J
zYpChzX@7rNrp5?sj$4?5$AblITeo+6{y1@K?UNcrLxJpk<ZE3W&)YM1_Ple2mI>K5
z0W40xIC_mDL9oll6r)D@Pf0=3VMRDLaDx+_3*m?6oGBx<yt1E0$OsZbpV!iO!TUeZ
z@N}taw|uwwD8&b;tb6kKfkGh$$3wo79O=y;-QlG-ng>@Y1m&zoW=G+7Yxm2h;85~$
zkjRYb*l{dRQ9;XdJGSz%Dx}2HQ`E|w0PA-*1(r3~`a0=QwCtwFS?LvmLlaXZzV`gl
z<Qa@^f!fk9>~m3q#Q1QAq;_-KA;CcbSa)v(@N7t^jr}b`3h~fc;BSX{5>RZPg;~X7
z!-UBFf$0UWS7BB9G7?m<tMBI<!Ot2zYZYl{)4L-<n1@U1X%g$o+uklLrIA(Z(weu=
zd!{ii4i=7FH&n3MdGP`s{w{5T2z$(I@H4$`&cU)tHDf;ui-Z{0jHO87N!V9=A-nCY
ztv5Di`{y{Q8b>GyAfB#a&$eB}`0L1!=vpL!B!zRu_p1yU#|8iXngG}Of_|!%yHmeh
zrD{LFk+ctu>~s3<;j)D*t7a&-IDIo}z7l+l+n&3uRnaPrm#&p!ylLIqcHT2XWk6O@
z#21Kg-E@+**YzlceG4*!X{v>S(q3WcUZOGh(_7RHznshuFzupe4F3+1GoKPJzIj;=
z-9%q>yg~!o8hwMzhWEL=?Z6mgF4ibb80b66yJod-F(6Z_OHH3)?Y3@g1aq*T6M6@X
z^cgpdo!(IbDhdU&9vqV?q<)S=0IZmwsGto?{qtrPG)1Ytvi8X7QMf-mAmqBm?;3tU
z`#-2v$IS+S1s35ALlsNeFQ7`q#29%!kWbWv_W4Nx=R{BaBY(yFzdFhSJvv`pezzE1
zau-zz&Z1sAHn?vpp-t=qIsvbBna*=rC>27Bm`1OAWU^~?-`wf#-)xZ(k;a96&kH=~
zQZGlscy0j`HSjUS$H#pN#2qO_f~4jR@`t;G9674M0-M7t$31gSduhJjlw3r-?MN`O
ze_Hi@(RF<xZS&s<q4q(Tq})FbQC5{=0;q5IF*Mwq)qIMufmU&IsJZ>lynzz6ms?&h
z*p!2JG}K&eTX-J)Cr&Z$EK<2F#%_rVS#Aa!e`$n|rEf^?4m~N#kHNWCz9*@Ng?|y)
zA-ePcpRh~hQR7XIJStfNb;ooSf=r=QM6LGD7cxYMN~)W*$5sv8$JP~RZjCzo{4wR_
zP+VSd5R#ii50vqo!oe3~biC@Uln5x&lNlRllmnE(0*eN>lX+ol_qI<>&{5u)u0*$k
z<z5R!0E#@4qsk2l#8&j>Q;T<D-Sizwm+>_~e>l~rhN(de9T$Gl7wu%2z)d(dUT5?x
z8nVGf`HVE~2=XDA7cn&Ar-P}#tT!7uiBWUE)UwW;)DkZcZJ+f4r{rJuvsiypJ`m)E
z93>0H>AfVGJTv%`wW$V^v^;@Vu3V{6Y9o8%`y_0~HZngl9D1p7_JK(<A&WJs$D<Fx
zy4RYq%Zv1+wgBWh*CyKY1-73xW{>Im=}Wg&4g*+fi(R~j>5}Qz6zBxi{?Q(d*7l#6
zNniOhodw|v+AgAb>Ej)a-+wPA$rN;D@l`F5-3;D)Dexf{hv#)$DEH`)+1TiHArHV3
z4Do*8KjcgqTzkN_`C?8}B5NjX+Q%<})1+Ui1JKJ7@&YVaPAuUlm>NE(wocaV8AwgM
zN?hWaG>)K=p94K9Mx^L1Gz1Zz8f*rUjDjMyIb`%@R5i*`WveaA2{}XV{Iwf!a~xKL
zdw9Ela-G(|K7JD262RJtxdiYz_wg#xrmgWS{1f^lBNvzX3eoH%n69CMRrD5tj7Fql
z>=;i+n?h7=ET8bl7cDS>l=7Jk1Ty{P=_n}QW4n(sSg&+?3#~31!Gckx494o(OL#Gr
z4dx-uq=;gL?=q8ISV~|NwmEm-u#yw-eaM@tfwKIy=w-!<gsp)?)HPre7$09{b9ccc
zBOtuT^*qIfiD#4##enx){_z*!=vRM6O64LeAeZXT-y;V8n1Y)>-9}?c2%x%rHf2iM
zkbLL0NXH53$)YaxG(!G~Tx)Fw^*hMtFP{y?&+)avUYaK*3aH;^0@s`0iYCWjsdv<f
z&QF}1dL??X&4-UgcgBrM5qHYOjg&W=D-S60-YO1iJ=z#vGqDHNyq-TmXxf@5jR3g6
zDSTVBmCR0)Q0)!XqK4I9J<oU^dGC7l=?|YDMYvSl#U@Zil~U>ILmCL8o(%ibPMz{Y
z31$0aK2V0864*qpEaCn43#xHE$t8mtCx!xl@BI($xa0-N3z0XkoS-F{ZZi=n7{7TK
z6H)3!lh1PKx%}3<rvE*Q4k<(L;}G#6i57Xbhzm&IuAKKH1-bn|GHamY{$DG@!ubEN
zGK`!o|5+IT(DA6{zNR)UOE8&-;hxNVRX3~3sC30av9f~9<b!`&9;Fdak%BYjD6p4j
z+uV<g15e`Ey4Z2d1PTI|zi=0^By&@y<84vo;-|N(<Mj&A&(1j+&5&7{`8zeiWW+XV
zCN0j>sd?8s_w%1(MA2EuWVNJ><Cjpy+i9v+b>q<_u;Q$hPY(yOwz6P78G^}qxM!<{
zuSju0c>c;_F4@PStu|B3dnzG+Yk$X0=HFud)%tO)8u0rHV0!{nuMARi9%7dysbNX6
zgy6T~HhSX-K+lrLxBrLd+ZDMWtsc+|;2MgbEV?)cnxFR)#(Yrnxtto*oO^1{)`Z>`
zUJ`Esi^ouU+X60=LmT?b^DXH-0^E+_T!S4X)S@g1ZYXIcO$heAez&4ZNc9RKQvNsA
zvj8|wyN|GI&1lr{Y;sov$Xb#XkhRq1tX&0ZO*BS^M_^`*MN5CW!HWF5JQVkGWnBGW
zt?TK}@Aqpq_IY^b2<XH?OyodzyS;8#3qkgPd6upqDugc(38U%*D4`}8!UsxeOMg82
z%I(cnRuSHkjy-ygTr!x4yX{$McV7H$du!jf>qmwBlVS!ra5=0g5ISF+5J>SD63oS=
zoey<3_r8VRIOc@CgenIkCt&-L%N?&^k9q(v%eG}3C5Ucyk$^L9qlvlS@1*tc7^OtO
zAf7L`V_Ny@P#LVS6qNB*`+2-@h@_wOFIRi21q6r(d6w5_*j1bX<YywdzC04`Saf0$
zXinSxNgr&4e@}ofTtzZy?U(B@#O|)BW$Drg+eSbK!hGCchb-8#4(uhN&)sE6?cY?j
zu8;*syBJ@gIUzxK@HP7}B6`;m9nYP>fiS3D4B^%aYKsFM4d>`ph@egzuDEL<pz3dD
zigZ;@;%^g=#g0qMU(P|%o1{EjflYz>u6OVJ%p$1Yk?-YVq=9lV7y9(!7Yy}=8uouv
zxU|=VOI-=7u0k6Rj5rqCOJ94i;LyPLe)3=U#-imk0-mnl{Rtv0ZZL|QANhTNE@c-5
zh0vpfuRj$XCM+mXPIw;q0qM1ozGaf!@+IRcF59~*Z94PHzA^5ho!XjlhZRAqUP$t6
zJ10kf|Cot-^7gHZg=RJFJEwa!gI8>EIu4cr?D)C6*|gBMKRWt0`8@nuh~w+~QI3R_
z*pDHn917tCjS}s4d#@FCo1|-iH=i@JOoFCU>a#HFQ5$Ii2t;Cwc(k2n&{G86y2rw%
zY2`R|sKa}^yIgx1Xonqiut-sUYU=KCn%R`B8eOwD6`Nc@X%%;@4h(qys|F(18Nack
z+%W_s%ToA8FRGq-!A5C!D2~lb`1O2p$MPak0-!xqM%bVJJI_dWEzdmQz~<~jsgb;z
zeLM66M)f0;xk4=ddeAO1MTUgDw4-H6R)(9z3>s>uO2A0XRctfgCGl9h&?0UI%~Vr=
zLMsH1hI8}H@1q6WwoSithxco8_UA4j$cW$hmAS`0KGK<?9>P)2GB-$FgoeMdTjXxF
zhye9rI}8%n76A(_hpQqG1lw4nz0e#JVAknpg#71M=jWgi7+vmhdJ_VwDLhIRbrTV@
z^^rGWUEGSbDZia69VC?%FM(@gLo`@6w%u<5lM90@cu}Uza*)R(&<E@h{=b!^Q9sT9
z9Y3sdj_4u9sUjl@>+e^MCO5(_ntsu|ZE0Z9=eB1cjQTjVA5#Fewh<vFGmvZ9&m629
zL;*RBHDZ0U7Gf`E14@kpRADd5DKg&Ti|AzbWZ>|6>*R(f*AXycAVsWauvkHl^$%Tu
z?t2c;e<AK5OV{pHfdd#qWVdB`qmGJ9V3qIs0T~V8LLH@}nKe99ON##074N0<>hjK7
zwc^mw+WlmLRkT2j86*EF>SS5NAJb?V`adD2?Yzs8Fqz`?1ZvOx8mx^neQI#T334m8
zrdVCM`W`J60x6dgEvk-4w*CxYFc6Mu;B*KNR`95DZs0f<&7_6+%oNk^#`<R9@he9H
zeT|4ZE<SkcAA8$@k_aYuoV`gpml`|nb-sQE^Vi+&!ic~-Cz1UVxCV`pCuxDcUr2G`
z)<PBjfDkWgjLJWG(g=!t0@=U$l7oIn&l>euP<a--B0}wH{(&P93c3;kd3})tMcwYP
zgoA5aEDGWG%`-scZF4%u<i?xWuW?bh%Apuh{xvHwps&7`^x^PA5#)IKU_;&q;V9VM
z)(d?w7!wOzHV{tnI-LY>09yiK=~~a;=ZBFH9>H$)a2}X`6ahC7*j7ZPEeFFeaXYiX
zQe=-slDmnl0oh2>=`CR-Hj;xx-D5b-3}mRpyui}tw?s}1hj0=_5_|9Iu#q)n5;I&2
zup-m~Gw8f<m==dASt#y}tq0>tGSU_}=Pu5~s>BDzIjqWZbeTL^o<b_b4^Pb*`gX9)
zs1RwRq2;+>a^t;T!Yynjt#fO2e5TS^PZ2cLnRdKj;GJ-`_$gpAxk{ytZWQyDVEMmM
zt%d%)y9km9UldfBM>3yH#bTi35V^#xpl9=Sz=hw}TZKKbUpq`hF|OXHKYxK(Y|$It
za6tf%=J`Dlr7-&G={jq!Ws77>-$XTkffl9ytxJt8YqV=ki!9vp9PD_M0BhTRoXFOI
z;v`EGG4_V}FYL@=1RMjl3UpyU7vA7zO3N;Tsq76G2ii}xX|e!gYq$j&CaOmA+2tB-
z&iBt32c?k;6!1r?ND)%m%;Lyyp^>*hX?CEOd*}hB02Be<5vUR9P{~eL?<?-ImC-JX
zZ4lFK`D)crjUU=8AcxtY5#~?C53x9^CplUaPd1V%Hk>F9yzmSI6{epT!J}ae3V;8;
zRu~`Ti~jP22N+50ykVEbDyeY}?&qW+(i`#uEPg&3WG7D5i7Av5g*nvUH0s@V7f2u>
zgR44;J1B8;h`e)xF^aTd(0rQHl!e5*zpt36!Gz#E1kJ`wo}jNAZ<J3b3G|A|ZA$UF
zIC*^zy?^Z5%)TyyNAAjkIZW+R`Q3$xs^xU#%_#4wFaNho(MK(1qax)z(a@A<L9BK>
z^6v-2c{A4c5C#Z@Y5TsW=EgHF*<|2^SOs}<C!=YJcmr=rrS=S70AprhVc`-((0QU#
zpCoLH@L7BWz~zk3uN;IlKAi*H8^}(-6&fcvEm^A$_|vs}Sr_&?r!tsE2t^j*dn_6^
zV$meyId$+Y)ljLIp!23+%>Dt>P?>dAu{u`68NXdzICc7TPp8O1|JfC(3<nH|{_WaA
z)>ie?C-NRmW2#K@?87)&y_5zY`H5XxtaEL(Y%xsE1pb!yD1Pye;M{Z@)R^th-|R<m
zZo0pa*bzm$(HI+D{Msd?eJ=n-7}zmonAd5=GSS%YLu^i#J+9#hJUUk<)r86i`P?`^
z2h%1W-99qW<`X0FwYJ5dfH0tqOb?FbXNcs*59BqufYuy`WAH3`gsN;~X>bAd(&=m}
znqWqRf*%^Mkk^O19wwcfGLECysYvpD4R(p(n~(zCndY^{)?(<>yo<bn#<9TYaZh8B
zwE4{?vsJU6lvpi!3RnZC#U~}=<acDogli)~&o9Ob+pC#QQE#{1<Uk;R`QKF$Ev7hG
z|Ei(3tG!-J8^V)JBrTa5=FA96YCc@O#w}dBN2uC6Mj8B4a{izO3t`>c&FP5>9lHbm
z-$#ZKMESk1#Mz}r6-$GA=mq+crfhbMLMpl;UWWsoOSE!oysc)eOdi!ib;h&}3Ar&z
zy^D0rQObo1a+`KAUr*r1w}auznrX9u<%!J-xkajteb7Ivlh%5pL$?*Y&{7TbT^{-x
z_kd0`eeq+_BK73PeD1<U<YOLRTLn6=Y2N`i)63El9%T`kVG!?q&mgZ|Erhunsi;Fx
z#`~{bx*hf>`jO~v^G-wqih^SWv9;EQQAEa+i(#j+WyjeR&Uzp}lKfpM-HY_9AfNX|
zxxScb(BnY+XVSR&$m0BT*Ucq)^zWK_ipEb%YP2$cHnYlOZg4X!Ud?x@OD2DA(ty6P
zAUv=W0`bmA>lB&3!J0C+(GD0uxB=(+cj?oY>T-~d6<-@P%e_?ziT7WjPSeAh4~*cw
z7c)NOJ*ck;qAnnxbS~gNFJF6BB<DX6_-<!v2a8vw*!kABPOonNG*NJ@k_gC$nhs}k
z%|$x9M+FOg0-x~ks(N{HKkXgfORM%*+IZ{iCr4zt{$SIV!=!f|tS>uvF*na$6Rbg>
z3_AT3G+EK)WKgcHZsitADBmeCO+LKuoQRedIAq8G^C5{NCQ(L)A1e5PLRdkb{=eph
zl`$R(4UB>5f8+IJ4XM~oQIwt!wX-o$QJd#TnUj>?hLn*IFjt@f!=8wE^~L*|B<Kkb
z?hlu(6_pwEQi)gns*Ud|I<vKw3)2RPwmZENZ!d(1b~`;vj!I!9O6w;Ez44mAQjasI
z{^mDrb1P>8DSNjql9RUWTbJpiTyqrx4(5r39WA%5dN)9(#H7)VKcO8q<_hbJ5i>8x
z=iXJB;PcSEgM&F4@3E1$xAA8Q4HfSGGO{+OZ#mCR-5;}qeEy5oNd<LN({E?JE8sZ~
zj%vo?eXMr7<JRAB@qgA*Y>z}w%Hw2hnT&!b*byl?_V;0H<wX=%C~>&oVkBJTdWEEV
zXLWe)mL`GOjuKwyXO+XXry;zqbvV2L-F7<<h4m)ztZA|3WWJucSG<+h)5-d)0aBAA
z_CFxL%?(3$f)Sd~)D>C9?z2sIRB@8Kwr<DwXMloRbul)rGZ<4@Uw6Tt$08IrEhFiN
zuly=p-do10SDtfl7<zXTFJfDfj11OMNx18<eP1Av6R2y><$kzmD+lQ2-`viYBKB?6
z@Zf1VPfPmIJA(A5JC0uU#d%CT^#=!;;g1`#G^y#VsNUb{$y5*jP!|NT#svwz5pmJd
z)H&qP>9GIC$<2W!oC_8eHj|P^P$2=@x=x~du40;R<)=D8fMCDw@WY;mJ}f@8r^Rhw
z{)3eYot;-a8dn<*Hd?k8`vK#ppm@USX+bQCj7Wmv5(PRI$l9Rh-<g@|v<Bmr7&G3z
z5_DhYc}E}&cY60CzAQMn?X2zgiZ<4^{#HU)-%(O>JpwfvYGbB%cG<IEnQ{sX0Vt2h
z)Va0LLHCxBu9hpAEg#S7U&ZMJh_pe0rKmt3K|E<AcaXs>t-`Ee=-0LSOb>)?>%w`s
z)gC+ivIVrEbZFT=!I)GgZR3R^VftcX3uLHX-_Q8}66i*wnEgs{p7(D@JaJ}@-R9C|
z-4*gE`Bd{lRmz`^N!oOTIXCiLXvR0~9i}d6IIXncV;5s=T3;TgHSA!4Tt;4f2mo-M
z<7JHf4jZj~LQd2?&==RSe^lqIouH-H56G&vl!Kig;o#6+>V<O)&<j`iG_^@AK}t8h
z|LLO-$@Xy0$Qu3~m>zmO-jB$lPb~L{?-N~eoKljUC=Z7r$qQ%VYO?uyoXl>Rwk&JR
zS-C$)BDNqJqmR})S=33f>#k`KJOjLbXcC*!k&ya9<QWQoM9-gfR~+#_96zQm{1#fM
zXO;rL8C%eoC-_K_D1-|82snn-OVru#uo~!2(X4J5C^sCQZe2NBcgJ-m!4b+B0QB-x
z)EZ_emnB%GMAVOe=}1AA`;XrnIvjUZ(=aLYx&OL46=?}Eh2mEazCZ5lv;m$tKjhaS
z<ctx1f+|W8M*Z$RUtK_rzeDg>R%6S(XDN~Khv*fjYqOA4JP3JcOz-lybfk0~ox6PL
z+zKiiGH^O4wk4qMqh^4c`WzVU3a~{^P8dx+_=I2BS}dx?_Upe&%X>IdM$q|69&-`r
z(7WT?{de`>_>0|xsFc3Bfed)aKPt(1;y(zUwRu4kcu0eDKfGvmJRb%7^pEw!XM_X$
zbiCU@`4N7d5Pl~6N3j)C_pJBbR_%p*^}$AtjD3oe_jA`Z*FsdM-%8TBd?!sha09E(
zfbOz8)`RYqVB`X1eByBr+t|Xuykr}7zqEnSa%P@Rr+CY~c%siaS_i0$9bjU=f{Wz4
zk%Q^CKkdjhsEp`(H%iXoC5Pq>S680e?#c$q71gitIXd>zO7u6xn0e1Z<x(|VS+qKp
zk`L50HSXW=*;rmVch3`RquMi|3Eh7!otQZbO@lIK{me|NRBF8_hF-?2Lcq?UG+m(i
zoc#PR8ct%wQiHIa8XWKxd3qH-n8MU8F^iDCtHQu3%SZs^>|d#}RbMq~-y*CY^e_=E
zSZ7D{i*!W%Chhun(jTuuzY!mKKFw}VIxF9lz0@*m+CRy^(i>I+C4aK~^cX>V>GrB~
zcn$x?=LNHIRLdqlJAi<kw&#>%lNJbjsI_m4EH+s@IJ7C#-y;EgP^^9d5%TBRlSkp3
z56fFTMEnkoj46sdbo}B>iZqRjf|(iAnaYtH;+z$Y(XBiqh#5#q4DQrZ`@4wfTwT_$
z(%@qh-)tgV`n9~?vjsF#75YH`;OXCiM1%>~6+&{d(|HdEy5t8fXp{Nx^=C5-bd$7d
z3xEJpaP4q+mb`%=#3qjIzn3zoe{r&_q4h>OOPVpn7&6h6SeQ~!hxt-W6BSOI8kIYv
z&-ufM#CMPbfQWO`1jcA+W9O27jSc^BwiZ18D9>O#*vx`{+UY)|LlVu+XV(xHJW-(i
zkv_nN<K(zcZ4fgOegK0*hYyZ`3622B!%&#Xmtp10GYSBSX;Ki`6VkFqk-<F~;2f2%
zgcMrGhwb4C#X%`F{vfTCTcN!};D~3J*4;|&ph@eBs0}_+O!*xl<27~l<nQ3ZQ=1Ov
z<wqp*a@h8V8ISss{nKpyiw+V{fG@lr!YdY|FGpSvJfWZ}!6-w7R-&Jd4$Gp&4M~F#
zdKh!n|0N78e8TWFFUxosvw)Giys?mHSD+zN;ZHKTy{#32q|tAl3RW@tnz*Lqr&3N@
zV=P<ucI3vDV&Q_M8O!kt_aqq!a@__jN*fdPzAXb4sHd=2WdqF4hi#KoLiHbm!t)Qf
z4fk1ZFX=&#?v!kgIgQ6n*#yDd$7{F?=)uIRy`q4ksk{NkF%y3aRH9>+kfWwfdUjBQ
z86_~_v3<WIOF`Xpq5a%TWO8yX2Xei&tCrGm>!5|O7wH&gHbhknKyDq|*_#{Mde;$&
z5lq5B9<jk##!&?)V}A99ZR8AA95NbKS;5q9*Zvv9ns3xN7<>~sz{NG+Z6ZJ5B7!za
zi97?Y26i`5lbN+|J-k%DAiI@Hs;kk&YCIQ2n$Ty6SDgeLD@`36DW=FFA1nFg4}B}j
znfLtKgX!^EZC?F2oG2y%s8=@59KskPqXr>Z_tpQXeZ%-P5K+pKP|zpj;${*4MbURz
zkEU&&Neo1r@2C@7z57LkB4nRigq~^pSWE`QyZ59YRtM04(3wPyDa{b|8fEoZFvy=B
zhy1!NhaN_EIfF-0GV&E-@l-!Us^q(HADT%28ev!Xs;eaAw))&^Hq)F7E<(pqd(v8n
z=oz7-XJ!s`|7&*I0lASRDWLIbFL-s?1esCBp@NWg5u+XFa{~W23J}W@!6R4pD%1mf
zU<8MvM1=1L7h-aaH2##qC~;0SVhW|rDn($VH<%?*)eY}};-SS1HdTRS4DusgT*Hdf
zrWz}kk5i|5)%g9P6d4Qy8{dJA)yIG-iD`n{VgQ~bJBqatd_x}>IJVb%KrvNM3&;t-
zLo*w<zkd!X9Uu96W$RqfKGQy~YPSQ@49W1!_!Z4yMqF?*7Hu!#<7B${ltq(SPK~f3
zVLH%pHd$BMavS_3D_RQmB<c+|K{6{VMK}$D*8RoF)T*CcRkU`H9qfT}ZxrhX{oLEX
z4HC<?ZGaj|xYO(LK?g7;h`}eruB<9G;CR{X_W4y#DEsBf+md}dRLu8uFy{;Om@Hk`
zwOy?vQj(vKrrh4GSS?y4ae>G1wIwmE2&kUJY?qyJxi}&hKJK-N(Qh)NUnAyc?mWYV
zzPtE9zGO@h1d?y!*5)^VA^97u)64cg^?a}q$%9LoDvEPj>0g6-)^!@Z5}u{b6B_Tf
z2bfFL>ErH62^GN+DBeE>cLIRiBJM+w+$a(1dJE=I7;gy)1(3{h3sXCYq&cu|P^BHn
z*G1G2XGsc^=2fyYQOnW5CkQ4SG7_HD1kG$X%WHQfDKbk%50aGFefo0#5({iFDy1BE
zg)RY+S(l@Bq(2J1?F6~NOY;A|X*Fc1e=)~Ue605lRR&&YU%7$6Drf-7Vu0{DJg4YE
z)pxhw!wHeGgmn{MujqrmDc-x+Erg&<g^YVIA4VTA1HRozqMK4`v4=PFX9Y^T69V(l
zjC}M#F>@PqDc`SmC%{dLDER+N`)o}AH>zM_U}7Qq-<7zEOahEn*NDYsNAZ2BHP<IB
zft;G#HqJ%#u&?9Vp~>R7gSn$S2%aZZKsn?^YuGe8UbRE7T<aKV^e4y=t5-i&6Cs~x
zlPqYhMfv`2^ZxhFOIx716iicaOd%;kVgLwVpi+@c5<y4zpe~vyG{OABs0|IKIV|nL
z&-iyBi!Q=_!vaKjaNmylP0Je78~H=`^Y2}^q!g3EEnOl$6?1F%!r*O@C;L4|y<tKm
zSsU*&sjO<*&>{8ok8Xl6WG$3IRSx8hgdsbEY9mfbsnz7sn-z1NXu;B3bK!ayjdh9&
zN%_%R`Luqg=}=%J>)eV44!Q=Nv52x9SEJ_GpB;a1$zfn6RIsYXXImRMqIpJ|?QLlw
zem6<m(qXwnp3!d~U!JH=f9$JklIYNCo4*-yGf!GcrU;RDK0bO&l%C{*$12X0N(!Y!
zM>s*wfJVNsnYLtx>}4d>)e~`Homoc6XDX#Sv4jX3Bd<>-uhlXpwljcA+>*%^e|Jjj
zb2PpMi#2d2ET$s)9<B-Mc-1>ZT0IeV66@HB#rmDK(IzyWzL-CR6~P$G=4wvo4c3XI
z)-qm&b^NwONh~<;0{~@WlwvGB=gLp+ls@{kbdAU|Wc?uJMGuE3xi7CfM@(O^%dXS7
zYzmV|T3cv#GWhl{8Y#g%YK1%E3smg1^>*`~69G7S*{_m2YGB@*HXn|FUG}e;#;(Vr
z^!ppgF`b~DtLL&`rL52S_nm#i>y!~zY$w?_uiA#Hnf~{v7B%v>y=8@*M9vnsxBB-d
zw$T0~eV~seLaq|%xi7faxzj--JYwhmIsI5)XkVG3QA^^o`|*o9>9YCA3hFjZdiE)g
z_cAa=l#DETF*8*?ea1g-43M_Rw$`%#V-MbE?|%Q;+TyWPoqD{x8Z;b#XHWJy8;KXl
z?SB%SG4z^00q4h5{P_V3LM+wz>Q`)$W)HU36R3dGYL#(=v-KI_QDI?X7dooy7Bf>I
zaPOtLi-hr2;6|RKky(wMnW7d*`pF4{#sD-5DHBR{Uqh1c)R4_z`;QRv&xnFjC;iHT
zG$pt_VdKX6f4yr3a-5m7{j+*%FMSGw1i?fC_-`!CY(gZ3_d-Mt-FmVvNWvvEO)k9p
zAig{_hrB5f9E53TcC&(q)|ORXDCTmbRv4>QZS;Hiva03%@dcUZPpyr_BkJ5uum>iy
zb0ipS9rRHCog*l8!Y9I}0VS2P;sYw09E5GJEk5INbGJJ@cG_-g6<D=IpX+s1P8i+O
zbO6p|8j^IJd&`uoT%^u=>NOXsl+_#P(6TLtxu9=5#}z&B?Vpz(^O2Z21Zu;PDAkZ_
zP~y?8SJ-Lst}^Gnhu7w4$LNff8i7G0klorzqi3VGC*c&=8PA%J!K8#{x{j~1F<b>D
zh|jbWiEK0H^ShufaZjPFE5H*){gKyp+!~ks8qwn9F;uow$5U`xxqS6RZ;t~n927(f
z8W2T$mj1b2D7}>hl=5X+c7XQAJR6ILw`<*#3)23YC|;T$?<g1lIP=VqJCK<_gNG1T
zg4oekyKOn(c{mybgSWilRlv!q<(|o}kkh}N+%I>Gq%(TpZg5j=-Y(}(s&QKox3i$e
z$QIZAB_S7e_GwieJ71s%TNkK9g{^d{jCsF8N&`Rn`qde#rTMA9q9r3F$-U;Y%E3N|
zQ^a?68g*N*|7|>ikKE5R0rZCwJ=U`gQh_a_%=eNLM!tISHDu_^DTmhnA~2l*O9((t
zOcIz|l({q?V;aY1L@BcR%VQZglfg&qQYL1s3^9>NBVr1=&QtSF6P)NBR0ug%C;8je
zRGt7kh{X&&;ZrfV3&|wq9sq|`0Pr3fdo2vb>XHihQ=JMX=pv^O2^7*y%p*9I!wGcc
z5+x7=;1QxMbc=Q4<`D3>O9fezohO1<4R!JSxj9va5L9JvJC-b^K{YuGa5I&{I>;rJ
z58_xMbGy1bZk6;p3!QN{8bn(vB6<CdqwbmrkN+25)z$t6;f}y_ex*uI0?&sN;SJp>
zxn5C1fx6|htX%DO8c4W^9cNZwlCdB=J;r4wS4VSG7FAP@Va#iJexDa0v8)}T0>a83
z(eA3N-<1?zCfSl(Ye2f=Cx=`y$+7bFj=?#N967nPOXN*5PT){Cnu?HXch=Zo$^u|%
zGJ|+eF3M#muE{sM^XePqO>uy_D#2iV?5Y#ni(I*z%gp-q4qQ_6$WSgw>0jFC3HFCK
z<jz=$a!N+LZ9_TFLZD|MEDxjFmvF;A(RM$>L!<+L8RK#<3|NygVnY>V<NI-aed0#M
zp^gw1vCsjDTHfCc=ANf$BBGq(r9PblGh~`~(J%)S3gsVguq!I#FtFo~Iaz0P&ZFWS
zfCh7qn`)g(0Lrt{I`Y3MDyMmcJ@5%42RVN=H22sFOwWKq@OPMIRq<Amry|jcXr|01
zs0l*aI$zhn6J<^3%m5UUIK4qfFQ1A;NV3ZYz%uO1>IfI-77a43ew#dU*d=rtGg#xG
z`m@UPF5CnNl>XPC@T=u33Hk0*Fc;xccT2Vu(1oNf1-3!w1G%}aNtM}LZJmXRrF`6i
z)y#dan)PaS`>yew*AkCQ@G>gq8c_|G<-^OCa<oH!hV9cm;KxCe2<y3meYSH#&E&-d
zSz&K;^&mukV@GFB#zI}eU%UQX5hBkeQY%QH3zW%MM#j+3nE%v!|Kr<tWS#o|NrC8p
z<p1=t+5aD{1|~)}wp0fp2x{PekZ{O`;+w0N-#W(MstN64zst5RZ5rT{!Gc6X@sXEe
zo{6pDNM+&qzFSH+k#a3-iU<KE-5?&XGW@s45br90<8@!gd*1is^<>{h5#sNHh$ckr
zCXs?ikqQBt2X?Ur0e4IPC_^;ckMF0$+aK`)j`4h4Xo<KZ73Cx|LOkGJ;Ar>ymYD^*
z?8)R)Q@h-xz2DCtJYGhhwTYO!Naa0O|65<bFTRD0QJLhB8SkCF)tFu{wb(>fH||%4
zyOj|cY4;|%!<~+f)`kPI!;rwb2f6{(P&*5)P6@$9f{SkiY~t{xbYX0&Vi1+0Wl17@
z-`(`Ub9SnYF*Ky$ja*=fXG5QklAH8vX}J#K=$fw^rBn<8)JS#n8)MQrRO|NY=s3sv
z|HsBVHRr;G%c8Mud&agiW81cE=Z$UKw(Vrbwr$(k->Orq&c)tWqsK3Jx>`Jkr8n9)
z$;@tp(_l8ppJ{0cri!T(rg)*q&q=2FQwU2Lr}e~={YwQLq_3^TCo%)Q<pejYzS}*~
zYN=N-EGF=X_R5@05iYr97M!mYkuEwAubSAZo`a!}fZkmbNlvbIAB)Y7jH@}}zNsP0
zy;QX+$U_mwzuXfa`m&%>rA;gBIR7dzvB2n}sK+LmBWaM|a)`f=RNKLsLThp7f^)@7
zm)DZ_lbXyX(adqYO{5RFRybziAlq+SF-8m%wcE>qUEiia6CpOt%_;gKb_3f8#v}q_
zrS!Hl0GI`1MP$a%GSjX7+!Vh3NY4!~!(naInp_NTwSiUs){@k{f%*r<-ESwVrjgG_
z)KLp$rfPomNB)#50~0hZoHv?yXRT&7*?f_Hk&)j82Ajol`tmGWK}q||U&s_;6b`fm
zO)v5^mGQ<vox0d^Ea~HPS+h-KQjdup<e&Ua0P4Ae0U7TVW17~lz^KrSX&?Q=GnOCh
z++6X+pMuWYhAi+`=w^v``(=2&WK(FCA4H|oMWYyMbhdKZ>aJ~|mlgrW?+}o~DNvI>
zxU?z|n8sN?g)>l3_N#duj0W7;Vpz0$JwXw)GR0G?fPsJ)-+RG_ME4Tc<P`A7AL(o=
zKqn8lp#4EpR)K&qEwneyDrF;lH(t8wpd+g@Py8D^2VJnt0n><sP!s}QpWacpS1}O8
zX0^(~5})#Bik$s=Y~_SRi4%DBtxb(R*2WH`c+=bQ%(9;wj%?evm-l$zjNXBSdfF65
ze>Fx<-))wiT|j|~`jlqnfI!$f+k*Wt;LJ4qCd~Rc6<=8rtv&a{zf#Td4gVkCI7(f$
z_`D+wgoO+aisNfz#lu}(P%@^tsv@+aSOxRM<Z#wpUgvBJe6IbSAC3EvgE1Z*SSnAc
z?2V|MgTbItJ^!@e%SndgnHijAk<SI{VOegG-|W#00qs6edSj3N1`Y)zf>Uk)0M5tK
z9R^HUlK?%UxF#u-AzC_Q)|<2|@15Z-1n&^uspCR`Wr$hrw+oxRzpJ-?kam%Vi#7P$
z@kT@3fM`!vj9f!MlirZS`7VNqhajClyDlgrGfWvCGg2(md>^vk2$J_Y!Z^ccz-rY{
zU7$hJX|O#$;E7Q9X1ew%Pi>e3@CMo#8)j<y_F8+0ePD3Je@uN9js}%w0^@YFHGY+-
z!zyOJlbf$0s34P+1l&?Vm7r%g6S!9s$1$W{npBQf%>6`D1T^#$=;;1M!>p-B?*E<q
z*KBb8%u#DDmtH6!Z(9ddW-z12<u7TgU*q;LbeIJkIBTH@QSPxbQ4DG=Ae1*KV{bp4
z$j$@_BngZ;KP7mt8~98*))`+_t6D=Qg(N>!LTM|U41Dd#=oTE;4EqP7f7l4#VG!vb
z)w2*GhQr?;t>uk|4Do$}5gmM8eHJk8VfRZ~S8!LA@YF5~$P`VgN+tuTK}Z&6Z=0n6
z50swPZiMHEmMC+W-t5di07;i%1w>1_o(&}hu2fZ^4Q0)QL)#fbMfy>K)p}es=nO@`
zC8zv*Ewy|(M1-`$V8C@6`o^cg3eX|Tdlha*|0rgnNq~dN9QTj-I(?KNZ%r@vrf$dD
z`|B<+cSGOwiG9{5$kv1%my+awAjYICo)Tu3s)1pY{#<#2iN2B|AS!D7-eCee+@`~l
zjFz|}HPA2D<60?L&pGP?#0n=u<|N8-ujs^X+>#&L#JnGVQejciGP8CZe|a$wpGusx
zIPC%l)xrA0l^C}C%Ksngs2&i2f=b`GT}e?%5`AP|JYXHF2!?ot@%BYeRRVeF<0QC;
z2B^H|iOg9Tk?5Z}KyJPAA-z{CIZzPNVR;sd^}Eb%Yw}`W)o+qj>YMZ*#R1UeH0=;+
zyII|ym$~>G&c`zumlFfnBo0%pU8!;fBx;U|ZjR5`-wJLP{GHD&mvAL%qg~wQUQ)!9
z%?uj414=s4-P1JH_ftro@7<yqI=)Iy5Ltb*+T?8l?V$HkKogeMt4u`)SK<kdR|3^t
z@4I6uFN<Ej`;eWeYhyNH8~=E0QucYCemM*`o_RY#KhQiK4G<=5KS-gB^uQ0p5kVZd
zLsPOn>z#x8KTFOT_?auu(i4buUcQx-Cyg+ZR#1wrt9cZZK8i+}JRu?XEpX9mBrrol
zV;TZ?4150zKuK>`$mAJ!B1zyJiKB6{QcHvDYL7$YQ@988Ca*=<PnP6E>*Jr=!#F9Q
zK#4)z_=e!?3(J)0bHzcQ)*)Wz07l*#gG(1ueqy0iPuXA%Q3Z8aN>o}$DW<`Ku(xpF
za+KLLsjud2NcbVig9cpagOjJf$HCO}@c7d9xr}^L0NoQN@&H(1Y<_~C|Ki$<l=73X
z^$n5xT;(8yo!P62+$mG5^0YS5IxUmlQF@*qBAl+wO7ZVVH$e=`w&I&(Gqc___9ksG
zRmG77<TC30xv%{cg2rH!e2dYbfXC<@q<97Z@%=pP@WnK{v<lkL^$zHR++pJ34ELGO
z@7swxJ@(X<;{Q)t{+F|GGX6j04OXWANn%n0S~UJMH^-3wGdC$YV5}9dAjeO38r{s2
z41jU(vDOT7K*g;jrxc3&$;GPxd6l(RkxOH7(v1D}yZ@P+y=skX42l2h?Elj>9^TBp
ziIPH)91}{N?%TDYN)C8gqG+d$D;EA8y4{OE)zi8xxs5jJ*Q(@-b+s0S)?Am~Zj^2S
z6adzj7p4^=CS9;G%{(7&FJSP{Wp~)7O(i20)~~u9X!G-QzmIRT&p~4tqB<tbj}OBU
zV48wr44D{FG8t<Y>AZm^agb4EWDb$r>$Ock3IQmM&64qIBEA||Aqk#HVj?L;5s96I
zy8;To(uFRzhsV^@4`;j5q=jzz8M5nuxs66QI(Yb*g}G0}TMICHh!_k>o%9HVoo?nR
zSz-xjv=Nb-%k9!CIm$#i4N!6Xx`5Eh_FIN9dsnT&8EJrpXTGKPr^x91ulWWib)ho)
z_wxK6bo5OOfLCogcjTgR9gz_s6={BFq2jcJmIT)YxWPUdnVTNXI~Wi6^S>kp#4Eo7
zcV~X~jvHh5z)S%6lUR^o;>HW7`v_o#o%PY~`=%^!N9ohhk`d(-ftH9QgDiAJyZ64R
z4g8`49x`muqvmpjFxsmSr%pwJj8ih2I-Y{sUaf@0-6G6O+7=_3do`(NOnc$Qr@k9G
zmOBMs6SEU7U@jkuwR~chaxNW^cnJx0FZmW^+M-uYW>U_W9HWntuzU4xHBn4H^Wc(z
zgEfU#@ro^5tBTwMN-BiWO{;uO3Oo9d+6$SNhhtn_kN{h28y4#o-@+gA0}Hg*&&AE6
zxP4>EYy*6olzoSJDcYtVeA1_IQJpR|O;Xhz9>VUNfanWw5~y1}edGcVRI=M`w74Hi
zNpaGtN1+R1VUMwYb~=T%G;K?_G}!4rTX*5J|KWCl&-hg8e0Ha$6dI{yjZubvFd<Sm
zlE`&lCPbQ<8m3;eX0)`n0@U&=fI1XtPZ!a*ga1_{x;e;@^*q`mvKT(2%HjEh?W7C#
z{;HACG(?35*3bCSC^i5vS7~RU0rgFBsmQr%+d-$6v>(zbmUBu@`>1a&rST~6_sB-_
z&WJMYj;<#NyU~iWl0(HH2P$SY@^#sjSDj>pY2g_5?}zlMeeZ^qvHAGX<eiweUuY~J
zTjwL3#9PujyCp!DO`b>AjjVE5Mp_GX^%-25hpT`|w@<)EPT~Q=n;Tf$iNUs<($KyV
ziz9*nfh+n*p2!>x&Xf(Wjbv{FeA?1Ioh6&d{tjO%n+x|iQ!43R6|KR0NL%AHbRm%`
z1{`o21S%+jc$M!<>EjkQ10!;!l0ZG`BPSo=xH=)OJDZ!YZPyLeP_^LwY%*Gz>1t<0
zPgPfvBGt@p(`*3VBe0r3f9FdHTK5g0k2_04Dah|~Cg_W7kJ;-z91!vc?^8o(sn0>z
z7G`JnIFr@WI`8ikUW(oGpl^&z>;km1NLHZJly0|qJuq=L(}~`Fara8gAS^8_l-1>R
z&)+dcU^l?gDtkoc1PH0A`<Cc<c2u1H02x8tcvcQE>i-4c42Hs|Pk1`-fU@SCHpq2E
z6Ks&8_4W8l8?dty1sA;x(|g%r3i#6Y&@-NDJj#Yd+E{o6H3|i6kjijsNfeyXT;2Z$
zB(-4|0e32o$9_Pnc{73HpKYH67|bD)3&E-pBjy7{<0|b3T!HW{*r=$IxO_q~!kQTQ
z$@X>fiM0Wft9pE2Em(^R^G<FHrO$ZzVXRN7VBrtNJIdHGD(mpBw~TlGRiH&Ay=siW
zq3H5Q#N@@i68D3wsa#jFeZq0W&QQE<)}90D5$>e0;acR4L{H(_B7@xY*?6og;=DA-
z5a|(0Ga#V2I2<%>+(Hgka-a-gI}qp{kWE7I|BL~?@f=f%ppRoPC*X*xpg=4ih*K^E
zLpc2KX7U~8uyTp{dvE^=6;`q6pgwiiwe?^4tbc)glE=XDqhND0G%@CTt;0~gk$f(&
zir_A_MVv}*nDtfB9BVVoI|vkt1v|!$wGOMq8<Hb)V$a?Fj%UXA2h^)fPa$W$c)+1~
z1)c$zP$SoeSb&++Mn}++6Z1B7;}FMLen^uu`EhWq+fp}?I`}lUo4N=+guF_WJ*3R2
z%x_~AQV!!H=b@zdn;nlx1)9Q9KWXTcO=vex9jVCSt?VKM<9I}i%j*x<B<jrDZssam
zOA)&9KM2ejh?5G0cFzh#@%G%*LGo4pu*?Jc6{3h}L}AR}g(@0zN`}h@l+Oj9u>%)-
zj!`v;ZRF7gN#AwdOZp06-m?aAyajelkjtpD{C*f89Z{xasLm3mq(z$;R6XNjDp}R2
zc8-is!DbJ53>AizD(XvJSo*`f^8cvp_~>B`cPDQh%TZGu+{q1O@$KI$pi~3<<|zY)
z?nQ2%Ae^$SHbO@E<F5%KIr132H-`Kv$d{d$>+;E5c-J;ZTJSkpkJTyZdhje?btzLs
z5%i}~go#pZwv^fU&CVUc<7Ty^cs)4JN_&;;N8#!wV|otrgho9dNN?s1V-R;&Atj-i
ziYXu~cqj(6xfYpeNsb_1L5Aoa-h~18Ni9trzP@Y+8mg=T84pYG9>mYU8n+^atPQNj
z$+Fv^aCbi98!Dq(F|C3c?ao%nRX$5lAg4Is?g+s;j9#_;Mx$?Yix9bbPQ<{VG1To2
zbTMRYLLz=|PoK}n7h<f8?(;;ief~^l$z40lKurFz^(+B0S3kW4yNIp-sq292c7!N4
zvl))u95`~6yICxR<Fc9DKG5HIVCueg2v>S(*(cZ=y>l?Y{17J)Nr*rM+1F=5UVdED
zaY8xmOX+ghit)ea$+k@x@dmR`UyX%39oA?DOU}sf6Wp%WF4%M>qYtKmdO)gpelSNl
znQ;ODvg1digea2(S^WYzhX`=bhw%LN6X@di4_6aOk<2kMYtx@e^GW;Idi;*U)pxD?
zPfa_*0HsjjFCXg0Q<Ad!IzJH+oL4D`zC<&Dr1~@K(-%z8e5V%zIJ8%ngDYeBEz)K1
z2sT026K?RS&A%G*j))MICVhQ*&^{^Ir*}&t1Xcw=RAG<|EfT!kS{4Ahd(3=_V^9qm
z-VB}NbvI<FVr{Qb1LSJD71<(em`pyi-*gD}P({C-$FU9i&x>At1Z|Y4*$7Q9f%$sd
z!4$l<UtCI=UrXC}RsQ09OD|QXC<Z$9BrS>Mw;l5+ZUK&m7+cccaX2W5?z;tAcA{F$
z#i^CA>-*_nAkGSv8yM98YXX3aEeV+u6_n|}_kZ}=w3>9>7A4Yut^YgQ?FH8Vq^1`~
zb*V;qs)-l6bx`9)JJwc9;9SMuPddsi#8(TpFuWHdVMR2;f9i3x=Sj5hU-AEb8M?v{
z^DI(Wiz^<e*RN5v(1O)nl~S^gL^G~BkF=?EyVfy8_aRLlXK!A~NM+$M8iUD`mxC91
zD>Nhl58Uf9_4{pn7kTWSIT!2l56z2<@(9Oudi3PKF(_q<lCmoihJW`bORJR(mqG)P
zqZ%0~+~-hZ;|ib2b43{rFRGj%S&tBuu+iKe+Sdbfv$?riU|d$!HR$1gNf_=C)CfY5
z<q*nbbxY%Rgd+O?M0@7qtY0`Lqvl^K9DG>;6z28*0?VDghW%TgC|{wvocdzza%dPY
ze0t6wv_;@-LBY`7@)`y>z!C6CI3Ngo&gqBO+9xS1Ido68!1mL70cQ^_^sukS`VNni
zi6P=m<#r^_Wh=}i(Sf%fC159Sop}IZ8%CTbnZ7OGl4*`$x|qsR>yfL9K^xDn3dk1%
zmX#^&5WgT4&P?g1lK=Ik-(|Quz;D5pu`B#*W>%8(3}Bhy(Pq$ou>ipe41z7`-bR4b
zzm-4<NOKj|*e1Fwn(skAEH5Q+P;uoBugD9CbEp=TbYP^<(6h6pfrto{<J=eD$_fxn
zYF*GehF*s31s;mI^2Y#D>!Dq$@;M6!>>sIE6JjRIHieFJQNe~X9@b1O4Gxi-F1=P8
z&`AZ1iaX3L2^?(U2M;AE6vZ&()0N+ImVZN=t0RX8BIiFlC8w>?51j|(lNDJft4d1k
zWJ_{e^xdMpJT4_nr!wW6NnqAs(oxK#eH&K8AfxTf4F3T`CVZgM(BXS`w6zTZe3Iv4
zRGSJ`@_H@hRwos$e%?`)sY9a$#wp3aAQ8S+>m6^>nM`Na6jB51RgSI09lVxB-2Frd
z37)nH53K8{<+Ygv=fLu+3#aQo-i*-SP*T$7?h4~uXA+M71Z?PC=c?)9XkbQ*cGA<f
z2f=4VHh~!CjYA<0&6Ej2(GpJqQYlU68{s8^&Bih@$3FS%=*b?6&H9Dn5EIEkl@Jfc
zAq#ec-s4QqcO*O;e#n6vL+`j#+p(D~r?No+1vu$y5x^{$NFeHRYDUvrC(*3y7Zso_
z&yJx-{%X}bJczi`WXvDaQs&9d=`UC+)1L<(a#w)w2&tVlI@EPGN2+{)F`R<Z!W_=L
z?<^$%5?wv}rN}74zfb$Qddn?-P=;o(B<?rFJj%?j`x*4^ialO!U$L12I=*G;?D`kD
znV}3wE^7^P3w_9_;H1l6RRC^{Z~?08WlqTI?4y!-Ayk}@us8*CVIL@~lX*L6mv|{u
zLol_zx^{4(s;dO;AA2ldfE+i<7q^mFzUapM*e@r~!6qe%JsJMSFk1n}me2&<@H#P|
z(QKI(Jv*VhVus4%ss2Q=V~uitjon0{r2YYpyH#c600hF&jf$E7S<9oLJ@!7mr074#
z`>98^Z3DIO3{MUad2)EO*3w+i0)}S);v<@ZDq$$i=Zk9D&<x7VZ)&)LV5yt6hQc4F
z`+(K5yo5uQ&v0%(iIU8H_7z||myj4L-pR(nP3VsJ3*wqb!k>l=3ycQLlBSpsjMikP
zrcMmZ_W$7pcSk~Cj{m`^G|a5b?EkxP!c4@>$;`p}zu*5RQ2rMZF>|qVr8#neP&HMh
zIe|iy5LJ@Z{cR%6q67x$i<8b7+FF|-e36>|2%PV3{`MKofpKPXX>xrPIZx$F34DVF
zQh$pRrVE;K{oX^W9@|(S*}NlyDvSw0uL*hvq{^B77Q;R`zC(h!4n7kispM=Dj)7E;
zd=ny}%^Whk^>rzqzqUNBR!lybcm3@8eq*YEb91ewTWVzl_ErJDhD3It&P<MLC&9Rx
zGX$OHE(Lzif8{|Q1j@bzx<y6swtCLi+c|-rXt%C_%n;^Cf;L7k9tC>I0s1)r{MCHt
zV}ZBV9&kn$XAYJCeLV-pefU;6yEmN!np_T57}1$&2d!AT{6DCW9og*5p|5&Zn+OpZ
z27Z5&0NkRzLD>MbxL$iBf8PQOPsY^iOEKrc*Ku5B(*i=V-iKa~v4g>)bsV^hfb$3G
zpAvksC2{8Ujtx;8mUB8sx?rH^d1J+!)|oKM)Ci8|Dl(Lwh13>2%C<nR6EPVIZI{<S
zONiGV(#qEc{cVs_=@LZ*`=*(iLtC?JMa*-t+d1Xj&EtR$D+pI~*s#*Mgp}<!54TE{
zm+*@l;1qX_b7`dA66p0_?&k<1rg)&wp|ao+O`*ue;~UE-TXpM4THgXub0;>P?ngSg
z6R5zCd{Q{p`xVuqp6+tWce!@%=m-!;gwiE=BD9Ioc6ECsy~fAvN~1(z1K0r=EY`Kv
zMr1jg$5Q}ero;L>p{OI0^fe#=)*chMQ_})wDq1X*G|pgZh_5S}QMQ2;M&xr^RD}ES
z{!Mr2EkYr#dotOcdu=4_IDV3=#{DY?NmD&L8jSQD)RUUEz^+8?{r2=b3C7iYAemCg
z&Xb*MeP0ud@{ORq+>)K2CcJ<DMTHq%d87GSZUvw<(G&ineEI^ySwHHK*7@%SWhiv=
z;{qkoxe41}afb;fdvYjBfNgV+kK#~kYoF-2r^3m*%Zf$69qN*#eD1+&kjrE-l_X!T
z#**(QMEjqzvu>36{f@@n>g}8b;Kk)O13u)`yrz@q)NfJ<sGjiVo0G)5Ua~d{v74wg
zXC5FpqTg*&p)K*44wF$;^Qnpv-&wx0dzc;c`K0vV-w@Qw&y^7Vgb(0G@@fFG`nWOo
zw!Za8*Zmmacpsc=C9?@rr%HalxqZ!|O(@!2E>G6p$Q&K?73jJt4tebpjC$yl|19Sg
z61s><;$__g2O@Vz#)Z{4N|ih??`RZtY676gM4m>1I0q9qua{w{S<*DSER78Hm*eFn
z0lIxgmHi+qR`-|2kTlwjN7sZ48oCNZCRUf(HHflo_fd?5_OS%c-rqk_s|pNk8!qz?
zkphEbFP4{mfIcSOw4C{6FpO-%bD1^S@go$5sCLB3MFeUM&bEQ?gV(mZ5>mZRlmTYw
zAgH=Bw6&9UO~EGOEQTDs<NFSU8MzuO!mB$-WH>byk~ibtc+S0^C-Sx~jtVh_LPCso
zBcire7n}+zBA!*FPk-yOlnZTDp$-u(2&dGgr18_>dC(-z<7n0Qt|C}dbx_tbyTjuY
z7_6F^n0n8rM>xnD>8`p_s+%D(`vGHC4gT&LZ7LlcGYu<{nlscEJ$a*^XmZSYEAwBB
z`eC@3SvTPOQZC>@e;q;frMzI=vb}g#q~%q*V<H1!?h+O}61qHzF>SXXdxNj;Y<uGK
zRYn^Ha|nNVKRk}f<5hsJGlsRxdv}H>wK2#yfu;rjtP2J`hA8x{!@>C}%K++o{@a!4
z!^h8(Bv{rEeLBQAs)@TJeO}7i0~cZTa4E<PFUOz6yKt6Yxj_CKhdK>Z{ay1j`~ss4
zr|I`VZ3v4=!{FWN>sa=cv)7^|JYaMv>BDOo6yGsC>TVh-nN#ATPcn3hM(y<AiYsTV
z__E~eZAyulUF$3`n!;F@+yJ<dQoHX#bmQtKV8)5rHsZ1_!6>OvJtk~WMpyry+f69?
z_Tu~X*e_({m%-6J))gWSN|S0$XxZ6o=CJH^&r6V&#ZPvstI*q*o)gd9NYKT0rQ}`;
zN@M{0+hG$l>?8O7o|GdUCedl&fsMBx3)AN#Mt40~P=OXuXl)I$;|)k9WXKWaXLA;e
z&hStIIy4rY$7eO6>Sf*yB3RnllO|u4NdwopYg24RUybpM$p&3@5$bj&y;ORX(}itu
zz9l^x+6gV<4=ZT(X$64+R}NnZB0>ljL66*Pc^S!%8>cwYHy(}yC9D4zOy#o-RbW*a
zIoh{e0c9%E>8Fx>*$GfedUeJ18jYj^X(~PVtM=Wo_~)ddeZUuWR0~HSC3V~)A-*;k
z@wG|U`+%O`gyO@!=qlPMAVaBOOH|ac@X)c;rwkYb5r|Ko#g%TFD*Nck64mw!_2|#s
z+j_9ugQKd&iQ${oos{s}rb4X%fEF7<)=v+GFH~MEG`jMqX9BPlDG&tKoNm&w4f$VL
z1~Gd)DGcJd6X!U751+}RuXZl;3e*=mxns6Be~~T2Q!!LW$d7<kF{G+V$0=da<Y{hK
z`&l`qE!`(Rm3PKuAF>JdR8<a|T0h{Rk&D2%z-ZalLHsZnBX1TKl4x`D_)Q_x#-7Dc
zI;t9rbK-YSejh-w@itf`pGck!D{GV%WghfJP^re^@}A94iNf@AzczBNUXAHa^=3QZ
z+uwd^2M;@110rgv5iWcj2Eo0tV_d9dNp*Ba(f*CyD(hfhGyv<7ZU8(puD8(BNc^_{
z&x1B#skajndNx@Nhxd+w#l$&`0Rg$`XaY&PSe8BS>;zy8vHON_M&oU8b3Ou^Vy8uP
zny_%qYYl>VLo@MmUIaaK6mZY^P1-dBOgK_p*=Cz!0}uie`>a@@?AOGoIF-84h{O(Y
zU%Rnv!C#Gc=A^@|zQ7x%W;lat{|Yvv8#PNQWS7+tY7<Z|nTrbetJm&!dw9!9i6#z5
zeGos@-3W-o5~2|K6iL%;LU<+y+t9m_Mv@Ch<Bj`=6X7e{L{$=I^bDs|dNBGOku~qA
zuIf%;sspIx;Eq5;OXx?08e-Ynke-$^+_u94ZlY;hRuWJqcXML3RRQ0FDgoM#?&_Vg
zy3yRL!}FPP1Ti6eU3$X2gy@Cxl6bPv)WM7nsQ}E6c`rKLQi~J%KO#iIS2)k|X56DS
zXP>PF+WuEosP%xV+=+g}oV_&pAptW+<kLMX04Sr+Jsw})3PV-ah4S7xH(eCK8K*?x
zl7r;c&)ue;Cw>+DJ<h0ZBcQqs-rC`WqMP+4$LX@R@TEF|q5n7ZBFIzrZfrMJZ#3}f
z0R`a0qDCk>zc16{su?3Ao5xa;B|MGA<2j8za5SChk&~?t>nPF1JIl;-rC|krNVmok
zGmfyTf<svr>k>#syqq0~ps&Z(z#O{lD=P$OE4DSmB6?{gk?F*LC}A=GA=#k1m-7_$
z-nmOQtsdBu;6~zsCZK0n8j&+tuI}K?jsS2*$EY8!qn_(_qZ_I-gfWHaHn_oKi@chu
za;}XB6i|Gf`Yq^v`T{yhd37gNkFMW`gXTz4uODR=25d)lJdhpvjUw<DcSgz~Cd(os
zW~KG59BLXh)&dnUSH%^_8VRq?85OGdUtjQ2R(e>dT~rlb`Qde!i;*k3Bq0@y`~h@M
zMT$})({SRjLE3%~(;1IS5emt$(GOjsD6@80BNnQjJ@na+>&W^=Xky)Df+xb;6-hyr
z&8^Rt75zv$>&;-d1Iip)*Jfgd&hgLOeBo`8E|%Kk)!tQEE-L5M{QP^tgX$bB8GaUC
z5{(=*IH?h4!m>z8x>Sqhq4R=AMgZaf#f(rN5mcUUA<+FZdRI!$wlNQ|lU3$}w<rM`
zUOM%-o69-+^ddW`YD2B?D>(EDsoZDxd6S%(uy^&1^G?ZqnUFX|dOJqu-l;wak59F;
z!u$I;{N3$)^ya^qUSwyn8h{PcaT~O|qEJ|P-PS0VflZKwH=rTcPam;fCO|QxnJi-O
zWSPFn0WVIU|7B~b@c9X>X+|2-lX6=l^lFr$Q5WL4EZ<Mbb35C<9SKSde4Uk-;88y+
z>647A&+m-K7JHFN)s{J$DHnj<8mfLd|2g0ajI=fdu(mbhi0^4XJ8Y4S5-7A%nbU3X
zp^MGFRqkU0`h<!8g3@sJUn<#@IQ6nD&>MOP@;LyqefZh|KjuNG3}=N3^j{5Bp7c=C
z+<X+D^?J6P+5S*?K4EuIH-kj_aPPEw(XhZSoG@g@SHMgrlu3;SwHeRo-#fYNMNyY6
zJa2s7tp{q`G_u0yuVhhfvs~vv4jBq(n3s+hP%+IbPVdUVztfqLcz|>=bd~Hw8jdqr
zvce@D{aPuHCEzu|^Yf98<fsY8nM;V0LUU#=A8~l`827uuYv+6#%Kgp7q^SwsE@&+|
zx2&9umO-!b1H4}6r6Ki~Q-;3E$Xr9Ompl_@%3zwl`9-H$T4AWqf=0y0)NDhq!@L`H
zHw>Ihh$E<1HMzLI9RSZ|?ZD6Oenc@R(<^Z|BTF^OZNyt{q{mK*N8a}07*n~)%tOT4
zel|vZ0VP#L7YTy)NSsp;DcGX(8FwNG<MF{$JBa+4VEFxbhSj?&f2cq%w#XBi)!+uP
z?X1uH(xbxzUGVpIz1{ZS)eT-^B5Gggs)oGscYm0syG5jR5WrLT+_7!f*f{(-Jm+6w
z-kYV(*%c|zJtK0Yjm+A|<*C2WhCUmzPkJS^ldtB%{{+Bx`DtGMoB4zKbrWLdyHom>
zEJj&y#-%%!Qh3H6Se?GC*>)r+2^y;B!}^PBw04lM&z9W=KN{G~{t~6yayg((<e6Vw
zVok(Cte76SfCbo{8s*%=9NYq5?2m<*jw`%$w}O(n2F8qyRn$*jDou}Gw(&5=Ri<`f
z4{2gncN4!C(m=&0&t!__9Z9+%@S=bgJK({k2?LZa5FFELy#M}9JCqRt$ai#XKk<)O
zOQjua;^1lynUPf8(n3MgtT%mi*<67xi1NG&!I3Xvl>xfgp!*7FfWDII7y~<4C}3Bp
zcOt&|cA~1{y{V%OHr|trnwS^8U6zYhA3a^?=pw$$+4B@};@K1wU`FZP+lN)L+G5yI
znihY@^dC4y<XugSB}W%Q*j|_fwmy`NR0r^wf4oWvZ8T^+V(iI`3LZ|3%zs(-Rn!-{
zT?;jbbO556p-ysHo-#ihX)AKV9qrWV!W;qen2q7JPvlxgW3!s3ZbFmd{Y&AUq?w@q
zRhF;ufn}oSLO6?HesJ!Z)cc-WYTo@qJ@ldS$tL4QE6eE9#N37nm))T$qbjcyfni6V
zpD-#?YsWepGch%*%Zo?}Zar@yNX|;<!&Y3C%mMqh==P;s{!4-B>iX==`r9Mmy!w{n
zJZ$jWTkUzGGhAb0iOU~2k=CAFE^k|M6f|Jo7kcuLx3^&0tujLzQBh2<Yn`M`lOtqi
zN0(Qc%5QN}aZze|!=b2Qw;S%RP2GoRJHVeeW{*k_9wypV((d15L3Oq}isYYb4l7_E
znSfs6R4n7`q&(jbVll{HJ3jukXPSeYF4gf5TRW|g$gZ6XdxO@SPl$sP7t0U~dt&6@
zVXBFv7!ghwP~2uKEkFs6_$8~!-Qew9?B+r=rF28~o|=q($zTXS8de|F-owH!VXQ+y
z7cJ~aI*%RdMdF@;e$kDf)Vh)zpTSKQK7dYk-Qes|8!V*xD4pks`&WX7MDe2ph5TWv
zBl=Zy!M`rftJ-aT(nMb6ia?Skpz2u~xb;NMqf)$v=571YOxc1dPeK6Zsy7BX0jEhQ
ztPr%8p8K(E(8UY6$e;7*EHJ6-d>t?87N@4!CgFk;gQFxMOC+Zhzs^hkkCThmB>;k>
z11v2{^^ai&IldA~Oy)%~`6-PZFm?|dFuIxxxAyl}%a5sSXu-AUvyE<Xf_vvz6V(V6
zuGAXtNlVpZH#E)liO1B_{;<mat_q}ula;C`^kTl0+TuJcMvJj>xORk;EU?E%gT7`U
z;BBa%zRMtHL$B}^&3qvWd~Fn@0gwvQr@XvkYnhvq$ytRk<GGFG3imF}XKAftfnzd@
zpOx7Q<+1Nv-}iQcMJ@iyf1>wegcG5#{v<83WrukmpoT(Qp)hNLqAP-k&>yHPv4`hb
z*23@)qT3m#Df&nfm-G%Sw4{MT<hc05%t@?P_OaJ{Ieh9a>znZj0Ri4!7!ap~lMpXY
z&uWprfAk|{>fF1XpI+!Aez?id$+c87EYo|N_4e&_Ko@;_Yj(&gJyZ<$2ya)*4@uSa
zHdXV{cYJ5HnX_%QdQH|fkkj(Wp+>n`Nf^FWx>TJB2iD4I)iY1Md3(AwPa8^yu?B7@
zy@9p1%CPGi1$j`Ry7xYZ4uB+s*+2@DsV39&R>-STZ$RViu^Mf8D%U0~@zY~nRo1PC
zN7r7483XsoLB7ARzWn9h2X4KT{B$gW)*qk}OB;tL&9^Tap1O*7lWomQLwGa%8AKI7
z<Cu};?ZjIcNHY?kNE_%u!R*5H5&iw@MQvO?{!%_m>?qmm{&B@@0R+nlgkXk2=!H2J
zzSWFT!#i6hQ1&3!O|*f@n1FkFQcZpy9sY3pTK5AruYeT1a874D;Yg*z{L?0C^RgFR
z^_-}=7?XR=GD>Amm1V!ft9w~~hzYh3dMnKdW}~q&f$CQs7P^18u++Z8N5K%5r_qvC
zp%(dN0?w1-5lR|90DMsv956VrZdXWN=nq!Fj*a1x2p|<=2FnjKkt;k8o#{1mPA((w
zHa;{r4!&Rp$(Khwswy|)P=w2|WW3r|BJF3YMpkrH>7<_##h2^f>Qys%$#_ZeN6ODg
zN;X%qilGA7Is{#`WROWYnzeV$HNSoY<Bp5nhK9Md3UD-`0R+k;@ePAL>rMp<M>b)p
ziK~~}$$tj<*;?_f;oc@0*=Q!d(HQ%|IxzWGDhm6rVMF;_qsbHAAqj2t^4#?bx0ezV
zS9ec|{NQ%!g26Z7ZtC5v?9jSBNb+z%7;rVwC5n{Lq9Nz!NcqsVHzeQa|8?qdH|T8M
zQPHQRM2My50CMv8C#zaiaC@XZ(wsm9;I~QfYtOD07ZhFs@Y&r>C<hsCQqg1C9ho@}
z<Sx&4l=&V`e#WH1(wg;Z_GV>jQLi_Za;0!kxs7hhK~l{cn)#g!{vkqWwG<0}CcCm#
znwb1vxYyaw?9%rNlOYLg(%Z^bBTRD~GgITi;NIsw1JZfLH+%5p0rG+kCH>{67`Mah
zHB6rvXwoH|?~P@)6`~2!5c%$&o;n}#QJJs+H{gi(XK%j!`k(LAi)E`MEqw}cJ^#Od
zFrwG2vTf%PRrycFWwwbPlmqd-i>r<lI#2mFp{x75?Y(R#N_XsMzOOi!X!i@+7rM)=
zw3cjL0N54y4pB8tCh##R-D?YQpej|LO|bQ%_)xt(@jbK5>g@6G>&u@)@4AkT>i>kY
zy}I}F_#N(ig}Jnlu=9*7QTGU38!K=qUgP1crAMhPJFJotuCyXZpF^V;b99tq%e$jc
za#U6#J`PX=F;8k)2=&Tzb{){&ljq1WBi@F%0i1}d6z>tpuC$A&Jo4~Z)yy{<XV5)j
z8s4g#gp0A5<ItGHamaX?Yl~vrNPa&gF4A^0h;n+b(a_Uw?}Hncwom&V%3TA_(0{~o
z7YXET@F2E6IaLlD%5_d6gX*<f^TE|(1meTRnSOLKVjq?VEGoX{1N1WigG6^h*?V%0
z0NdQ=Kc)5Ihq48*4*gqgy@R7Qld5bc6@H-0$vriB3E|9mEfP9=-{Ow>IX`J7MB&9A
z1x%Xd$D6X#rn~wA67`?luL=y`kZ+~!=E_xRpFqgC8Ac>L`XpB?muzNdo|k+Yg{G;e
z-hW&V$JP{H*Wucl@S*g|`H3c)?WTqJ0R$+$oBJAx7b4h}!<?5}%5Pcr{vlt{6;Kto
z2;BT9f7}o<(bVZK!L>a30`tgSbu2l3f^A9j-=;&5G7FEVyxZ26OZY9w2jWg*fRE`!
zk+U7Ma`jzQ`C<KoJr>t42LtP>LSb%1^v9ccT*J!n9}_ZqUvMkr)6q6RCF`)}0VFgj
zRWR?tu%jxai$W2ZdHjM}&Wd-n$f?hesvaOqu&|8jY4e|B?7eW#z4N!{D5-=iN$Pt>
z&q#6!Jh{}kd}~Kga)}Zbi?1k!9<3n(=ZR3%T<oX+<(5siJEm!r`q$(byze``qcE%{
zhCEKm`MVSXA8@hI;aTea@pp)y0C{>3mbR556%`GY2C`ddh{{_m7D4L1E0j*Rej{DC
zo@5gI1pK#uuiw_6FAF4ua*$Bsot+oKo5qwLf+OWGC+mlZM8!3(-q7LJ?G1pBEO_4w
zl1o^bT5NnIpB2_i4t!O|<BH0Kse2lk;?BEM_NPE`a?BVhaaB3gkhOfl0K__^)B>Yi
zk2^kYP!<D0ZIrm6o9HFmx9)pg1N3TkM=>M7tvs6lwnDBeB=YMz)$!d^r}_IDJnsfY
zfQVw~W33JXr-TjoJ-Q<K%QGJ2C^84y>vKzdoPB=z9EL<x$c<wt{o`P)A#?K&-4t30
ztO@(?EyaV)W5Azj%N3a50qYeW{cSX|S<WLkatPr_c0wv#QO+2n!p9axI3btE##{d5
zyXh=8lR+~0aGsE%S^PXS$9J;C?#5|M8k|o2@mFkq=`HFrF2A}PZ;0yiLo(YnNmP*$
z<51$mMf8R#@)OKSTp7(7a8ED#6DcH@8)0fKJ4>Nsh`n6OxwXtb0QL?I{x=z&R==<e
zwvEquXats=ge#joH1lR)_2Pt;R8?#8-mgMKZ@|xEv{1Jta;jXRq~3Lx&5Ur&HKFC@
zN=&1ZJber>=8_^^>oUC}*{QzHcNNrxV!|3H4&RZi117KlM3dTlOi{yavwd`Yl3pvh
zQGOBTewxSIEyL>!U|Z9%y)fNc4WpMW9p987SF?!e(nDctV~B6Jq1}J}TWf^Rf+Y^=
zo@SBKLPb}bbi2WksXc_sp<N<lt{K0Go>_&^A%&%lMzr{s`9R*(I&s~_)JTIj_TJ+B
zQx>$dCwWP()zF5gWK-1$L8o5>`i*TEe%kl!@BDAev1*(zAR~<a_C@CrDOwdZ(E~cg
zVJ41rmPBl9TvX`CvTlh2uS6Z?lV~oe<OLjDx3cXuwZ=U-A|a*AD!r4lDJcoc?RzBY
zi`yrN-p)fEyq6437B_a?_vTJ!{<k#ykx}$cDg*M}y41@<fKH=%c${Axw4r`d$R0|m
zaP(A@mp&r{K%rk;p<v5#3cIZsG_nx?m~UV&33y67mfa4D+X5%J{@l!UzEEg#{bW{z
ziY^{UmVh*@Vck`~WT%b^K7}jOKX#46+?eY{7md8MB${44R9>QZc14-Y9y)^<D@=Xr
zg9}iNQiW)ZeZF}N^qz3E<6ZkJGBXqrJG;72ZLu{7K$X&~G>%3v=1q+0?8a~!2*)ej
zaM4@7))5^HlcKj0G?VruD(<LJ=bR_XW$c>s>y0?lej<qDD)@Qoy)(e|;a6Z+oQz%a
zdDhJtI!C?KVIIaLLDs;@VR5bH8d{eniL<#U?WpNieI1Gw&QczB!IUk5ZpPEbizpch
z%!6D3sAFltp-7@7zP6t3?Go!VF;BMx$%Bkm1YpQdB$1r%_Ef^`Q<7BJ6%C15iD8ix
zpPO{Q{oO4e^0q$m-W5<{*15NPMH|I>&wQL%WY<uBpG%!v%T%_sq;hF3C$KBV8pd(w
zN)hYJaC1?ds0RE)A=a6xGKt}dv~=O);&0OfOy~c#D=B)WISlWZDpC&ym95_$b@kv=
zz7a2qq3{|T1l2o$l@{Tgjl#v)fSD}EJk_^!uHJ7k5_?>=Rf!d;LaP!Oq$oXfKR^#J
zIZT<t%act_Zc(4I*kyO6KD0|%pxu21yFd`b@$=AspPVBC{g?kpfl4Cv_pD~=YX+wQ
zptkv$ydt#9G@+7Cc%F(L^72OTqMfE4bMe^Tf|yP)ap&-dlfSzV7N;>hx_iUxKOri;
zVDejCn0$&0jtsEePhIAfouG7}sjmpmQgX3v8T+O%5LiKO>0|w!eaHUsw;8cU%Y8rN
zz|>HrgJnfn3izwiKu%h7gYg#L-E`9&5HOHs!}cQa_24{hJM7%w)s##JnRTMtj_iDc
zJ#j?QjJ=fdNzHU;oxCKL<uCv<px~O{cJc{ckml6Te(fl_k{(i5<p(16w=G&0q%Q*~
zXsxaX<p)CIl?$)Di{F>3Coo&PUSc{lwx)w_U?zk1ltjv!Ci<$fdu0nmTMm^IAi&8D
z&A)g*;ZvjP!|c|p0GIDoTdG(bHRzB5{p*zatAdO(QhoYg>_{}%na=&TGkp3%6%ZCd
z99nqHL<*9{AF|J$`et1dp8mr-YkVn%4$dCq;&i5BRbUuY#f2(XVEj>AMm>3jV&s(P
ztPf($X2K>XxZ9~MaaSH?nqH;^XmI^Baq@Imo-jK?I0+8Y65N+~-FT4(OJ$OM-Bt)d
zAO6wH`Ck3FdI`SxY2fmQtHtG|9~{xu)8>sNGyaM_k{fY?0;<jd_P!lT{rCy)|Fb?&
zR$_z?F~YvY*C2T57^rs6uGx^+yu~3%FB;X^<42J^u+8-`@?B9EVLr_V;5EU#X<G8m
zNCjT)+mW-0CQ!hnl#+PX$;(g}a&@P%Px-I!zoT8;mQ=XZS>SpttR5Gn+AC}>BOR`I
zjcI~!ikQm2fry%MCCc9Wv$1RY)AOHX<_J3EbON(c98+A#M%h~@$~mkWqkzM^;Fu=+
zP-u?mr-z@NgVHg#t(@5n;C}i;ABpK#Y?q@wPrQ3Rxjc+5!+_%nL-RS9LeH5R6$>8f
zn!V6F9duPPCok0^RsGGwQ*)<8ud@1j%EHb|>G<bhb#{3O1Mzb*b`{N!i%mT+CHchJ
zF)zIL*{@f$aFU=~>D?|@FDsNr7QUV_7C-AVAZ!!l*UXDGd1rwN;H#INx6@oF=|*c-
zVIsB_+ajv&+_VbhR}~n*qhFE7fG7S=TAO~nQFVPbScsEGDk^GM+UU`N&?r3v=>&DM
znpB$><?Ju1p;+5}1qBcR9iew~zybLl4?z;hy`vxtJ{GDecG*F3fV5H`8?ER$Bb=mc
z7wL_a6CH!k<L_z#1fmv@4LBW@r@a2H>*=bYh&TO((LBkaIg_GBUj6mB+Hc4RDQYcE
zr?q-pUVxH`fk>^;`R@L}%eR$0jykBixV+Af(@fEvV!MLdbDk^f*HyUHU0cY|S(x8|
zxoMl3o)wu?F3_a2`QiQ`b1$}kH&$SO2v9U$70;|@AY2Osc<XXCi%FBh-ENC!ulkE1
zfoNj&6$pz)E2P7FPQ*$cN-;)@pS0_ia9u30oJe9*dVc6Re1#FxM=EHf<2TGlcyBIr
zJ<=H0P5ftRS{4eDs$qpVhA+1*b8O3R#s@FBe%Nq=^k~fVmoAl<Ra)+UfOrhVS<3iS
z{2QxH(kbZ#7=grht`A69=+qI&Vv)Rk5fVmua5hY8t##09N$R9Z&0Dddv~OcT91Yu<
zn9OIni=XbS*@v*?F8u>7mtTKLJ|V9*K)8-U`lgLKH`Ly$fGM}f(_5wpWi9tM+`b^%
zYA|Wrz}}^Nj^H#{X(~PJ(|;brkO&T$R1l>C<&@$FIKPWfXqx>)mOhOjqp74=An<ri
z;M#5LwE(Eidnzmgx7Wc^?ql%LG}IDUGl@YNg0XUFs@rqexk}>S>847UFrcU&-^TnS
zip%qVXB`KR9CiM%N~)V?{co)7t{wt8KK+rY8wG#f(6EONNr-J3WM_KYSKCfyZsaqU
zVi|=Qzz0)~v_jzRdMN+tVScqJ7M!Z5!O+*cYS{BAmG+au%m>Y7A3#Mp@}s-vp^QqK
z3J*;vb~Ir42SMC*vR%ZjPMw25#D(}cqe3&@3eqpvx_VANoD^kh2Hy#4Q9Z}d*No-m
zdg1jPi*Bn<kXx^qYO2m?yBe>iM-^v$fVMsz;2T%*Pda!z<_TjDedKP{agTML)K*gp
z7CWd-=En|8U@&NHLz#kS>QiqcEo)a;`3g3Un_ty1mR4o+y2Ar8^lx{Ha>N{Vj8;W$
zwr4L3HBx7x_24CZJx42NuQf_iSP%bk{r7vttY_guox%%Clcl9n8cB-_5yTQ*QFz}H
zV0NMN{G%S5bi@5vz~^c=lKa`E6xLhRoDL!Ge!z!qmjsoas{!i*bP{#OMKLZ<5+X&%
z!p4X*KOM1zn2{T@rD>J--42(U3V|7r*?0f!OKYbx&*1y$M2n>6F}ucj>luq0$)Is#
z+iYY%jw~;x!Nx7+vE?uvBWw$Qe{~NJNOg8_HhAhHqxH0G7e{Pk{da<R2rKbNc5{1-
zZd68R%G263Arq8OYa|$Ia&}cvbW|;sLzx-Q=agBa25c<O<%&-$q2Xb;(I$sEh2U%2
zn^k{YJopiR_dfaE{?u35oA$9tGB)fF@=S-HrDaU}DAsiKeQnbC>lb?)gJ|dhfRS29
z;|MeTkA-#^V<svbIw8+E&Mh9cDMJ&ypLC9+w=o8HerH;Ub4G=gq$Efh5rN}^yFve0
znyttR+>!#9uLw3N6hg`eZ>pfcz32VL=QDrEGmoOWY;x6Jq>U;^$8Ky*>_Dvx_m&4-
zrzgUESjo2kICH@=bCU0!Wpn`|03~=*OCC>C;$2Kra3QnkO}u>N{LX?G6`ebk0~Zv%
z&Bj-YJ_7a+OsqTZA2l5WGSo7Y2gJ8?A5va{+WUDWZtxgl)HUt8L9&MBuatRy-C`y>
zV5UhYIvb+@PM|wDQwO|5{E8_|a|Izkvn>hSRGnbZ=57-RCioa4Zbd8_Kq-hp)udkU
zDiTC4j2jq7!p<eVu8GyHTyliOj(KcczZ5Ziw|7eJPwCLkdN2zE%jd49F`gDjhuL~A
z(t;98jX~yc*MLk$Q5Wl4bF@Qr85u&nSZIUqv?jEPSZl|7g&Xb0INaPjdN;@@7bO~4
zGaZ%MtLMrQaYW<jG`Rs9z&D>zEjp&*?Kvk?Fxt7ie)6JGlD@N}|G=6|39gxF;=w!0
z&FJDPCIsPj8Ap*bSqx1^n`tDOGo}5}+WjcwJS`BR&-ahx2Q9%$3&_BS&vsbB*Nu`V
z6sz7y2`XF)&tm3`uGrf!!~;g2$u*S`0VIO(t>8fjd7sa6g)q<-;5a|zV7;n1s_2x0
zU%Yc$nB)0yj?i#PTcF|-lMB1Hm~eouGla2G;~~_7aTm$M&bN`8437H-4}nTTeHLWx
z6!U~iM+rt*oSM1OIG4Y#0Fe7d4)asB{JTc6;Q~YYau(jOpTHT=R@|oYB>^5?DO(Up
zbDn9J<wp6kyqzrqK-H5x3sQUkJruMnry6b_6zY$qb|OcVyDiPgTX-+U4fyUcdgecV
z-&0;XTOCpBV1Puo3(C1V$lzsiCoLzKRW#NBuE#f6PE5qpL?XbB3_K@{tAcU{w8Q>g
zT~_F0E3b9#Ow7o)9n$P?;h=@5A$3*qV&ZUJ@Pjz1tFQQPm64q^xLjaW*#@sJ`uO@*
zDVy3BQ;i=zm)GdDOfEG;3W+-|>_39vQKfWQ@_XMv+7`75=gKp5Ph3p{0@ng_Ts())
z;}w9eY<@Y`)||pTZ@ef3H<h_!J(@qi=pf}MjLv)ThE0#A=w;T<C_!b(2oimlHr_pz
z0%J9**@<EV=r&$~{>KlZ!nQOXK{4kV;aM6b9j`rSi<|LDAtdQW5t$09Bpeya&aK>I
zYa}CvCVNwQv0ZRM6s%u0PF-)n1inU@gcFG<ES4rdyr{L{mQ99(9c8*GxjnA=4n~6s
zo_5lhY>Dq@(jCIlpe+t%XVdSUKqKLw9Rb!tMZrD<fV_@*r>sLX!w!)>RnKl`?-a2_
zC>Lve4)LI?wul(I*??W4Z5Apcrwke5*7)y&><u=Y>JEDP_^H63{`MD=ic8En=@psC
z$|MT110kf_VD#6g+Z7tPU6}bAYo&`$B#=&qWdyknAt8j@{6^~4PIou|zOy~fkg$Mv
zxB!6y)S<R=<H$UD-l-~cu&lhkzU$P3gp#o;@P$P=(pqma$473-iKCz3vESabf7}}y
zP>bvJRT+*IBoab=LQ^`SGSV6#qVcap>qS<#mI%PepNs|w5PA(qP~6YbYdR?n=>P2~
zT&G5rt?<99Te<1mXF<<q8(~N6#UBFdkBr{{g)xOjK*syI`7wnj_s)__*4|I&s-Q__
zQIMt^zclUX+4AH57IjW-liEyj@OkrFB|;)nkrRPn=P|?q=TD%XxS?0-`Jx2)`PtSV
zrxt4b4%t1p%kIYoJE@?jY_I)E$4(HBi#lU;VjG%~s8m7E0sYjI4J#dJLan<sz}!~=
zVGfKG|HX>*)z^2o8tP@dZc*_YS~W<vK>Z}8yDZHNWPH!EI1uO7Cpxc|7HGfG@%@22
z9RC7W(SMP*Su?#kCH0iVeCd||2Sz}-zx5n3H<$qnO&gz$5BMt7zw|IaDeJgUuDkwF
zy0SlR`dCz63G}%0V(xHJ42Mu3EVe*%56e##2Y)ke<m4N13{djrbiE`P{WQ$P)<%=c
zE?>zB)g!^<#I6CuX4VTAKn!OrBA?zByFv>XNP2jr|2}B0>DP7s*xn;RS%KG;{E{@*
zt#h5xra5EDZ1U{zj0-sb2}RU3dh+q@8r``)lH`)iq5)O&S+%zf>fX4A7QgN_haR`c
z!hdPtCh3BnLC0=CvIB1*Tk|zP+>j@ebhg~n<zR%kq=0-N+Dq2|9FQ^K;L0vCBLaoI
z{i5yP8y!82|JIo4f%=o%*;0Sm55i^vNr44l9buWZK(03H#Kg%hdQ#UGkr7N$B;ZY#
zIvtFDWb-8EC>EyN7tzB^Jyj|y-3zC^SAUUhC63is&8n6zYdQr{uHg>tBaCViUJbeK
z-qDvy*%}=#rt!&-bJp67%eGTCe6We>2%)ew4Py{<wtajC1x~ZC=;2RwL;9Oew|7;9
zOf}0dzf3MW!g}DfD})Y;)^6Rd=!tiFI_4QxrKU<eL|M;C$$A8`p@<|0)oswBaDOnS
zZv@_Yb@L*SZ)MIIqFKwg@5~P0=d;iUIqWgKM<b$#QTB6yUM}-px|~v(BF@5GgO$&n
z4c;_QM=bTPE#<EOt`71GntVHO0!&O@MCJBqwL&LDue>1&!6<abD%hggLFnDTw7_R-
z?432~%Pr-s=~gQckC@3P!)cw5J%32n8%o&Z1RR4ea&fG!766AF!agKuMhZoqJ!j@A
zQq$LvKa<A{NFXw$Uc=M9_d!q)zzOc%^zn068-f{Vt3VEiWta|ezKeQ(TljR|EZ61I
zgf)<_E`3V68uFD#O-1{YE;SFoZ?L54y=>TWa*LjqU&~vJ$x>8aKx3pTVSjD0kss>`
zvw-88#`Sr)Gjs_fZD)u^3Xgw=H96y?!H1q*y8N=HJEsif>ZijljBf?s_3AW3ldwlr
ze?`X_rP;|IIw)Gyv_b1=lrfe~e(Jao2U9o--B|zWB-hg06@1OPct7n|QgcW9Q8&2S
z(CMslMX1qv(AXodK>K2GD1USIXDyX9qM9}BVtu#PfZcHUFE|Kc1UcWWRN;p+q4x30
z9>j>PeS%`?qx2&u|1(Y`E4aX@ef)fVb@z^)%fKq214J>kCZC6CgrLD13l3Igq-<qL
zkgzR0%nWZ}slu5zA#iVAK|faA%>(#t>5HF+Wsnc(fyUL)X*c4CzJKTrtI~5RWq7Or
zRmai`xY9il-i&4-LV9kcADAjN;MW+;Sz_^NT$SgAgu!J0^8GVg>LqN;`-CfnM9Z7Y
z`CYzZ_Yxnz@=ya_Eblym1Xsm{IQC*@J$m{v*cJVooE^=eBgr|GEE~bVR7~3}S-s15
zSQf%6ioBLRicjo9Vt=7nV|-qZByG7*l{?hPU%v5B1$(}JNgZQO$DvS3B5jvs;PHZb
z83KnoGxzVg*<o<8?)WH==EApoMVe$>{(K-51i+leO{9@^mXFcXfrBn`jM`^U75Qqg
z6}*=bx~b0oKp&Nx9GDZTXB!Qy8nQ(aw8e{vxSMV)5|R&McYh4FLyA#E3=}-8wBag`
z5*@G23wnLeaF{0D6G?BPL$#VL#MUWnA7i^zBz++0qTyi@O={Q!4MyeMY@c}{P%$vy
z{7w9U3dy593rY8~%dL~%*pB$!yt{nlc3g@Z-Qr-ecSBFOpH(1lIUa(=P?$>eik4oG
zDnA^pLs$gH9)DdtT<N36f~t`C9!axE@vYclC|@@4y{(GRBpt<hVLb#pdG}&Ckl49H
znD3M8M=k3QV&+^J%L?FymeHw(+D1~!5S@JVvq}4bCor_H_^`fe-_}eRwPEE9T~+Mi
z&;!0vY|DL&QSj*V6jR^*IuMi|Qi4c&PxDsqmjW4t5r5VTPN(!XAyheSa$i@En{I?$
zN0^`m*sP<ILY;Z_x^l0<T1_(w(+OLO>GwL%8AKf(E5YYM;a@3D7k3twdUFXF8Aih7
zGjRAWAS3AC^Gs_^n6FDHVccZ^FE*iSu>FbU!vX59DHdXA`#PlGKs<r-r-%@l4|B*Z
zS41c08h=qS7EL+sZq6dfX5=4H3T!kZ$`mfu<HZ<I)RQL0Vz?VbJ(vYC8JkW?gL`w_
zJ#2iJCDEBANeeXlx-N0iurG(2D7ALq6Fs>j5x1gjD@)OxRrL@sTXe>LkN^`)DM&lt
znQu);t&ccD`Z1Q}xoU0u(Fpa3DS1tVYvB3Cn19p*f`%@4@3}|@exDC72&>biw<SF|
zYu&&Y(N`Q-G)MnbOy*Q7k9khX|CN`2lbX&^Qjs9N<+Fsx2Jj+Y0o+x(oS`Fajxh2t
zlf8_Acuo<+6<3=q3sot=t@<s3c!x0G&Gvll*Y<XvePCNP@<-)r=AwycC1QB~0NXtT
zsDIZU*LuM<;td$}^Zi@+wr?`uU%pH)1i5(&zhH(vpR4S2l_BU+5JL596P|T9swj8f
zYdYX(^mpwM^K3g^f~_mz5r0GkI9gopCnn-M8M5Th;ll$~B^t(eLs0rzS&?Sd_N{>A
zxg07}xBkGMUy}<=#>Nv6Pw(+B&v@V`M1M==F#rn25KXE+@?Oe5N){X3js9hpl`|@6
z)>PrnIoSB|D$Dx|cV?{iza&wy$O)!C7tfU7(@>*?7h3LWtT5eE<8b%#vA#;b&Kc%_
zvxv>2R<wi3IGzm<&w|y}u$CQdL~kpN-Cm=7PYE|*$>g)9%<0uK$>2$c-h`?ddVf9h
z!E8}8a#Y6@Q~sq;E1m0T>mg5iIbpN>C?7pF?=~?ntrPuFCrx&v37g#XYz@S#>!tq;
zL7|~86A8s0&y#Yvv#(0SS1e>a^n>%m!IM03pz2MTy7+{BXfQlf6*)t^TF@ZBS*+H8
zr2_2k`a2RgKS^hcyv$R1S`N*JLw|66hbm6#ZZ?{mcPa9@rVK00q?{e5G>V6Iqg7TA
z@LhQchd^q}lqZ$L1>~m7cWcS;4i1;6Y7m0T+jv$=^K*Q#@${8KZP<X#pZt2)A<hYD
zx3$F6>WElfyWdH~r%`*P^k7>;vJEhQF=X8(!Xnb~sLf~uhh0m4GuvfK7=I2UjM!(K
zLBD4?oNuY~g5C8L3D*m`!;S7ZWg7$wEdG%)?&!lLp8jK`S-PBPAK{CmtI_QY?0b$O
zTdbAgg2aG16LKpaf7}XoRk=B`6Wt2(uQ{zZ3uJEY&?5W#SvxSU1}Ul&t2}l)0v@S6
z?}r}}E}DwL5M8Ge_M2Q!c7MVclda{5;d#!ImZRC@+sjTKz?3ARJrP-q<osr{husAD
zAm{875mf=K36?TLpLB_Bw9#m2;8Y!J_vASkjd9D(h2G_x=u8Vb5xIP0?jUP4V*U|V
zk8zw+(O^?}0XevELd84qZrO9rrY7n}CM-$`*$Lu(60>bhcjL*3*?;4&qJ*=?v)8C3
zp^l0Pd6wl)n0|P@+;M6c-b77Z2TJa52CXX=lAQAV<~l}2>N_Qu{!l{og_3t_UK;}|
zuY-g<J(08KjoFhhj<MkhnH&6<eU7#2c0P(@knyn}z%6#EKRE8%dHmw#sLL-gXx3=J
z^!Sh=NC~~MTLwF{Sbw%tFl4xVJhwQn8#(Yw5hf&rurU}}3(pvz75Ayqqlo;lnGHuU
z!JYewV-8w2zX<PO&iD&i95B2EJ&o^aD+)#MRSq+$0KC;vdNa)=&{<GL!Y~6Cc37!%
z0I_#I<-Xusz>@`_>2!Xed9hR$$)-Vm$XZc@FkxBpLtrW*-+zJmG1_9arz5vuLaYwF
zk>%9b4%Zvu7LfNO@2g9(wJ`)63f&q%(x{7$7VaEgQbKq8`F^REKo}IWC}apUrcU$t
z(7e8?fehV8mDF`(moEa^?L0<9upxtmkgg7oW)<p5BZR&PV|@c+x;%2y23oqD^rgh~
zt(bXEAE3N&x_`)t9=bk(d1dKLSjh@6E2-nCt`)gtWpdEV>&~PwmTp01aRE9^R!Aw=
zHPDic*NXQi5~My;yJ<!8T}($zxWrA@|AMR-?FrlUaZNh^@?mzO<w?LIk=HeAm#9#S
zUH;LR!Y{F3;376C5GoUS5j<qQj~mXlpk`#ia&F2cZGZTl#7&4#f8u^RexTa-8qPhc
zu%{^{^P_d4#Jd(7U?8uy6S5wOMwS&}!WX0A67|wzM}w|pAwn$VVYOags~dgl%Luk@
z=V4TVroFr4{`{TJXzSBdJNBP$m)<{zu$ACKd;2Rq&|jLTnnW~3-L5f_8FSGq!xGa9
z*ahMCpnoLq7axy2yqK|UTE|10XLMB3l^H^|21G|W-&23e5q0)a{y2EDQddbhnlo^O
z5=t-5aZTj?Yx%gXISFB|pTgB<!oyh^7$zyD!}}qsta)?JOfG+E`k0>)+@9geF2&~^
z*c|2OnGJ8mn4cFmdiR2S41yUf$1;2z+O7m?qkm%*TR5G{RjI*!NWpsF#CT-F3rT1*
zv|Vf#G=BW@^*^0*UNbQ6fbmQcVYe=ItW<-vKStJd6bEn1yh_L@SQP1ZSKeO7Dkj1B
zf^+sBH3zvEaMhQ4O1;HVpr2Bq$mo``D>=rj*F}23>ZqGMdvW)lpYJn_iB~3BtY@wr
zW`9nMC*^1zJ*-JPa7BgF3{||)b0ld85mXvXcbJ$>#d3IutxSrU$9|sX@C8FAEE@)N
zjLowxdzzYuKuIIS4~*8Ku7*Ej*^^m(Sjhn#L&T057vyUyROce?m@4i@<t6Bv@rhnV
zcQn0x|FN%LQM_4~vSGOCv*sW`H?*7v*?)=Yz#Wl&NR5^zGjRrfQuvV{kuBHJz(hS&
zmxmkCF}OwH{K$Rm#*9|p@y(8L)%IX=k-14)tN?*Xvkpz~&|G<zort@xgUFHXD6$fL
zlf*D$m#mKjiTL{Uv7jShzWO<JukTvHHkI7v1P&%s>4NjR8SiYoPvK6WA|Z-`4Syww
z8L+{h55Cw)Pk3Xq@ieZ=(qrz$Y!qmpXkrb9M%p+2GOjZ*M<VythovZJFEvkw-Lrt1
zFoW?L(eapeW5RMZSH^IY1Zi}lmUOy2Dy7TDw?QafeU!BU|5?I|0MapKElfHMkC*x)
zA~DY^rbsW<GDuyzDlNi3mmg6{%YXjIyJB&H@V&+YjPB)2f4aBPsZ(RfJJ|2lqGg-&
zd$(#u_+y((3JB}el8l1YwaZ^%>Wyb(bPzgzRyq)_shQw=;vJww<=;t6WXlJPVL0|3
z?<s-vJQ4aCvfx-Kbwz0^SzfWd+0rlz(y3BRW{wi0NR7|Ajlj=?t*P{$S%16Fa{vSB
zp^09CS&p1ccJ8`t8YXS&J|i3#3%dW3s=@C)1jj8m5E;QeK9PgJWoFP9V21W@qL1Fd
z%5~1aXfG2rF7C#3s?{8*XAP*FwAfTW<+x`bUVx}^-&o4zO(Ln{dg}mB8JvV9y9;P7
zpdvY`4Jd2Xc_x6|m*e0Wbbt6PZOiu1XPOdk_fc?}d2fqf8dbxu@q+Fkpa((nLc{`?
z-F|D(;F?f#aO1<)N1$0q>Fv5Dp;2N%5`U21(=0T{LaKE;YN8YF6f=6)Se3qg9&K}N
z7XAwf`NP*lRD^dPasfwZLJnxFytZ2;{(i1#o8Q?9EH7(4#TK^rz<<i1bu(*M)jay;
zp!(Z`9KKmow{dD4u|?=~Hk~)Iuxtli2`4j?P!A1eqlbTg>dro@Fo_keXYp>L-p@uz
zN7i33Di4!cU*b~sPs(tTd1)Onh^s$5y3uZQR-KXv>fT4uD7Uw@OIc%(%SWl&!H(>o
zDUG{{0tBb}cBW$HkbgM8AxyILBI2p|@CLN{w+3bD&dxlJJ{$@e1(x6QC83@fj2#Ag
z*;{a^+fN63!6x|YA6VVGGJMkgCI_bfj6&J%$cP=m5?!*blVtEAr?Vdc)@Sdr%pMFH
zYeuGD1HD{I9{t5-cUXD@G7A4Zocs!Qra`rt=a;4Vj~EACX@AvLi`ZrabW!>9%dd~e
zSzIdd@okVN1!g_Ru`LqWb=a`Eb+fPvpF1KZcnfkiFEdXcaxy}T43?EY;`DC#?h^6A
zhu2uU2dlvAXigZzp(yHn%QB6V5%4+Sn6MkpeSiSZC_OG7jN!$8EAsd_WwKV<iPzgI
z->unUOlo0XuzyTC@1@Y0g4N8jSlZY=!zeuvG8I#S15@^R$JxW(Hbpzadf-f<I@Bxm
zJvaH4J08_A_f&sQUlI*gHi@jPcVu6xAx)TKJ&c+8`Bp50*os+;OsOBIrHtMQB2_c9
z95Q!uOqK}3GQEkJKRBU<Pmy?uY!*TnYm-8-tjHZtA%9rU&nq)lP2Ei3;W=#hnf<&~
zS4^R}K7)wo^tlkdwkq6|cnR&cEU}m4D%N(QY+CSu?VM-*ljB<kfA&G}xs9nKqu|$u
z0B=N8j+*tL{#LyKYyvlu+SBjeW%qF93kYHevRI$&%4v$`j!%M*Tt$V;A;TBYbgvc6
zA=&o+zkjx^X-5=A(T#1}w!Ox-ZQHhO+qP}nwrzXv&V4@L&^AR<1~&pLFbi|{{kR1!
zp0ELMPluBUMG8cIrTX2Z?oASA%-klL%}*F2d$UZ#9hU6Y*FrcCTSstadBS!<^6Dc5
zlPGZ8vf^?V$YGveY*D1-;-tjkdk1ElbnSa_t$$8#)QFt-<x95RzyYGV^n6!_Kqqlk
zLA#V+-8cf(_OGi_d;U4@6QkqF-I7J+_x_ofKO7mzA|XkHPHFAkx(4kGNh*A}&nUZT
z;h9U**fS-Ax7l$(2jOB7Bu$#JWu&i9C!q`|)c(3DbZ$&*l1L)50r>|q=KX22?GN6|
zrhj~jML>u(yMB|zEDI*qwC%Y*!*MFYXPq&pFWWk_U{17`=~jHfBt?{3j!6RxlDCon
z*g{2W-2KsHBdcJUXBL(*w#k%BUE?6v@2R@UJk&6ub$bUm6HZ`s`3ScI&X9$F8dq=E
zqu#gPVmP|9-dj++a-BOdr1ng;y}%G-K7YV#6<T;RB0JZaflT@tRR66`wJ2G^zIIQZ
zug<5nFjqU7o)V6M$M%onwwu@EINLPc`@D)f-9#%|N2-F6c8k~gLJZCM<m#W|Yd5pE
z(uo6D_ZR+L;}q^Xd$=B5ue{m=++azQRU|%)bxlPF=o5+MHOx122v|oFL@Xgp0)KgC
zjRsMXvWdb67r5Cql`WS))bOy7q7Hlkf5;_PC@Fhu9Y{DmRT>r~rtKNH<Mi`6W{UkN
z__UTaMkhmConGE%iZXi_<o$D^Nqurd6u<Xt&b>+ev#4-1uC58VG4?n#o<w_j%+6wp
z>FiQ3M7gcD;1^qHbB+aWL_*C(?|)tzc;bU>DopzValq-L)Wf4BSHgY1764no_#68I
z9h&#K5^P(loQt%&-&pF1<6LK35_9zX2e5xO*Zez{)z(|<sY!g&k7$W8b0j9-IL}HA
zT=*-idH0|txQO0}G~omS%ES`3noz~^&+cWxsQK`s8<{0%lbjSEfYbUO%73hu(x;DH
zLTBdi)$lL&T<EmXe#YokI3zD@Wrmg1tLsouUr?C&psTy3gIBGQi(LI=!>v>XiJcO0
z|HhwdQb;0igx@y1Z;@Ymtd4D?mAS~qa*&=Mqc-npR`xw<IfdCcHu0^tB8-i^Bdjr%
z^~-cvY1-iv$+ddD*or-Gf`99h7?-^#H2iz3a>C`T&&haUdFMF!C_)$+**b9^yui5~
zt?X4s2tkQQ$n2f4(Pe1Ni>|kj57tIU;vWMN=bOAf#52^@Mj*<5N~Szo5StEqDKKGd
z*Ox5hd;^G>5BM(&fEvfGrM%VC6R6!5XJ_2f{_1^xuO)HYlJLfP)qe@=4}Wx}Llpt4
zY@3grqBvT-?X%)#ca>^uWwg)skbqS>E2BskKOmX-yY(#xQ(h33squ2-KgXEa3}UKP
zFckh)4fBD6IfY^HsnmCtuuWElRr?Ma;tD(8R&|eU+?0M#F$&1wx&616Y8+!+rn#nE
z1%-<b4a;UL#ZA$`8h`!H<GRe6qV*)&)a2gpup8I2LeJ<tBcn?g5kr9FOh6}yA5x7e
zltXd}c8Z`vLlhWUGnAJ#t9q?6+m8~KZ@GY}iEazLZGr3fk}^m@5e$JeDCC$Q+G&Pa
zbrgGBvB%vGFd$jDEIuQj>wQg@?yofbR-V~&;H<{DrbM}VfPbh}v10M^S<36WaMbuu
zhMn{mO8KutnxWYj@%|D|2$$s%a>cJ5TecU&9R<P*ztL5=dejCNNO2)2_k42Nmw%`7
zr+}#KZ%*>@_=%%sv5J(QYZ)*kW}N{#;y<pqcbRKkdkM~iS|l`1rHn;+`sEGhb$14R
zs}dq!bS^uKz<&+YlA2%Cik&=amz%hh;1K0R-Hb#na*6YR#pDk{X1n=s@j>ZGuAlm(
zLhKkZO~M-tXaRqJ5+DtU*30IKRPt%26RrE+d-Eb)!Z473H9A~D{6Dhy^6p4)&<GZz
z3>cu&?X})@eb^&<N%_GjC$lvwB8cT7Y)EFyh)6-M2Y;<$d_O2o#4>1d;zfbMox<tJ
zZUzgP<<BK^u|}N}^vIyOyrlGB7AvQYrP?%4P77Ph&{l;<9|&3H8)_-~A`J$N$e%Fq
z4hr>$iRf|po@U<+bEB>bkZDmC^%c{>5hM_D_fZ6vy;jBxG3TJCTJb`5^R=pSKPEOX
z$kkgf)PJ&P0d?aD?FLxVXDBOlLR>s@`S%2DqAxr<*O<xf#Bzg20z!)eD_7sZxH*+z
zHNAcXG-YL&SjReuV2`idzNw%Oc8h^Dz~Au*QREE@=gV*FD#MrOyW>=4Zbrm$rla+%
zkVDJo<+;_>V{@?5e4tqX7~AnTnLqGFzjKkG^?%8AsLVNCZvAver?TD`S*EmyQivog
z{nULyK2;M@GwLT|S5){@n2v42o`facjz>T)%W~;XRZ;IjtFs!U@p=DY8;|g}h)GjU
zsliqdCHP@=5+Z32JQ$><pX0a}vtU#-iE9~c9Z!d!YWaFuOo60!Rsiv9AsOw7U((NQ
z3x6(TWuo8c1IFD$L+L}>+{+w!q8a+pvqdUC$|7wePR9Sr7i1Q1?Je5g<!kknuq?~$
zRl^flkhhkM=n5*=i!VOdZqB`UN3&?NF~zd?Z8j$ja;hFW>A-4tniCYDE9oYR?{Ape
z^&u!2C+%ce`w(2eT-ORDFI|dYjm}!vqkpe3abZq7iFceM2wVGbz{&Dm6lD+1Q6Z~s
zPE11?U3-2$QjF6q_t#>I%ADbgfFlKXhTgM6pB6+#2II`eHPxgKvZ4YnI7!-+a|e4*
zki{=HB<fv}Y!-#M?lpi~KZe|=2Ru2irrGnL;=H+d-<9!o*^=)@HA9ohCsnar9DlPx
zShan5^m3?QI)jAi$I=KrfDWmL2F5F1l#wNxO#%wDkCWx2UCULsafNs<V}EG=u-v2(
zIY*Jg6k3AbBxj&N!|0ocP@X44Q!|Z~-20LJb>nok2753*<kw}-fA5p*Ng04jrgVV-
z!pB;7I9XCby!cpKE+5nT_yED1>VK1qx(d5T#Jz-O?3rpm>l~86soPW?tvT3wI(^xz
z)5w9Kd|Mb9o++d{HDR=qLJBbqh_<~~310b(&74Eos(_iX$S=_kt4n6X=jNLK3ld7l
zFltSOPpu3J$EWRUr_G<U+89&u;0x{^JGU-5k|77rk-c3UR*@7`hEhSBvwxPXq|VX0
zNPh<z?G-LrsY;>aZCocx2E2+Y;DHUyUV>|Tffx3Xx_2-+PV>V>s4f8RCqGa0DLGr>
z&7tRnGVZ?6F^@oT>-by}od^s6xB6xFT^xxnDv>WQ)u&ZDl5Mr$*vHrHrfV4v+=Y4J
z33SRA#c*|-H;EnPCGypX1AqNLo>**4{RHxm+*)Ix&f6QU`IMf!&Z^m{!k=0<m@ZIw
z0!X~eT_@bLH)Lg0+cKGdiX}C1*owVKa&^x_{S_X|ajv2KPi2`Ex~w3a0CU{qt3?uq
z43!VWq!`O0jDQrUm-u~YIFrZk(r5zSh9?ndK$x3oT~i=wL;-ujrhg*eWEF#at8^p&
z<}7_@v4YO)eiFD*wRiVv7<n7__Z#>a2M@_xDLE{C*d5gt`v&hzck%Iw!cP^s{N+>`
zUE%c@OGO{vr!GZHWca9bo#MF@CtbodUoPr92`}vH_vaL^&-KwE>1y3aQVAzm#xoXq
zs*uDfY~k{Pi{f*09)COlqDM-J=zcvrS<WcgdWwQeQ=~j><<jHqYgW|7=(r9Q<Is(c
zoqz?Nk(ozwwf?|}<gV3ITp`i>xgufh5CUGxr@*EAK|NcUBu|n7S@X%kv`~y&%OpvQ
zdy)J$%KP#;>A)Ii`=W;lf%o?dZ!4f$f>t;e4ST0KIE<YtRDY07KxiMOu=wg{GLXzg
z{T>XMX4i6LPqURoV*aB#z-r04K!QT;U>l<;H`l@*JE3M2099i*j*)JK4r#e<AyCag
zu*dENPM&OGFjFMcGZf|WKCJhNpX!ZaZhkkFXYvi;(_@2&Fby=@jenRhiWp>2EAj~c
zKmPIz$F6Xl4}Z{qg$QB_m4!1E?^1VUJYITc(AE?=M(6TEBn`Vay$S!zWF`b$IQ6~1
zPiSoR5K~;c)yYa$lzB6+DPZQ@>te0nM#usx2P3PfUJJ3El`*mSiY-=VRL*NHz7cfB
z&q!VC!W%;=e1aNbk_cKx4Mm`++zCNP{Q8^$v44`^p??*25sk0G8cXV-_%3NTCUM^0
z8(agBginvdQ1Ug(nUxbThUBo+(xaiMNNBraE6Kn(Nb2j7P^u!l`I%JjYID<&SrX!4
z(31t+_IYVSmz|mBF4qk4U}!#t@>~-Jp;lEB9YlPBKsJK$J=<mk;@}*)QPPY>h8A6#
z@RQl^Lw~sgza85p{)XtAkIVOmod!)@_kg5em=#o+{JFB1NA!?1j)Ub{e>utL-BQ_|
z3CX~rz)U18`QruaTH4Ouv_9aYBaLyzWnqF}7w|>RF?=_}Un&Z=;HvbNE8z&O3pEFW
zHVNB6&6y=2>wh2!&Z(aT+2nyggyGTaqzBu2W`8w5$?X&r4A!rvU=Qs-?tj)P93P84
zQtLVYRY(BXX=Qg@<6#R&E&&B3)!tdou_S5^7kl|L0-ab7-MGz$6jXQ!GQWOwnhL#;
zpN>-QhcL!knQHNoitjbtA`#X7s@Y9o-oArRq{3o0NpfIYi0t7%y*_iFrQ!N6YU{?7
zc7InM5k`EFJBL_bLGAZs{?9FFgz~oIG%Cj{m)R8rjF<tGbKnyCJ7j9-S=>S2{6;%I
zHIjInDP1E_)mGc`qizagSl5^Wmv{DwDYl+yl%t#eHNko>-Go2P1SS0TILA!X<uqy?
zVP0_-Xv35CMiRdZw8X>Bp8f84KGZ}M4u9_ub~OmuTJ*Fm%C4+B9a@qG;7U#ASn$4A
zoduDkfSqD!y;5eylcEXtuNnsH=?Ap%Imp>zZ9|Juo;YVB-0+4Ku~l7=hhSmpy&!8}
zH6e-0a{UH8mEtZ1=TAnq3|40BT+Je1*0QEQG-i_0x}9Y%c|Agd^dVktd}V7gU4I+^
z<k%d3Zi79|WgWtq`JpxxOCZJf<*WK7m8igM>{5tPN();6ko+LeoGu~fn}nS*<yjB?
zX>VU9CSj(Y?tX5pu(m{^WESV065QEzViLOWhQ1dKGg(W6VNQh7GT=*s)(Mi>ba%z^
z7+t4&HgQE2RQUPznYP70nLg(Bh=16R!Mp3XtsOm_W8Gc1FmlX4|Jq@eORTW*kDR3S
za>}TbfBr#!CXc#rfOYi!Ln_A@zuJgDE1=>i&)VC`z!Y7doYaGzk=^5RtSGLKV7D3v
z9ZPU&gA=osmwgu=s5V&{30<rFN80QuKD_=a?}n`0^mI_irtD)vncw5`wts3r<yZ8=
z9&u;jVdG!A;{=W5L8fUj)C_^Oz5YZ)L<O|RcByTkKUzoyF|B)oYqW1_{M-2(TXZKu
zli}A4e|Yb1o0gn5N5yQ=<iS9YaqbdV3E1M&<g>6*zpQjr#(P#IcUNLEP7Y4F+OfWk
zLjFa(dS=w)MwPzxoRr*W!+*XD*%9-Ml$Q(bUeuu{H9|Na+okRw>apv3aRP@vba`Aa
za&xV-(&T4b0Bha1rX1i8{F!L0Bar9<64TbJ>wIJXRA!G;3)o3sJ;{Sh3EfM>VNGsC
z@$2_OQSqCfhC-5=>CF`5zF^edGmNg@j6@fHBECq6S_t7~@MC*SN`K?&WI4qm$Z@j~
z^1XB+G6-iCro#syM1YHj2>0dCeMn_tRIx`063h+%B#DJ>E(bKu<OQF7t;7*9b_-hI
z+}q4=-m}Dxfxz@J3_MbSbqJ4DsQyXxr&>IhaJn7Ze$(Yd+eRfh{pzf5ruPp;ATLvQ
z+q{Jl+aY^+kF*pzynool>@Gip@=f9?`RAIE6$j&ff(v`b$3SXI;LYe8Hgcc2QWuUl
zP-%QT%JHbmO<`l{-Z!*JcmCD&0n=m5SyuU^9wMBzRkEdr($MqK*pHVxmetjh_<^Z9
zG(PWV3AdRbmQf`;%D@9cP1)OSs!7ws0N8AnA9^z&A&l;ibAJ-9)D{fow=Ci}XICxH
z{C^$0L}r`M`5?5_NNRa<siR5A6o1zzAvW#5m;O~%%-BC-aUIl@FxmPdv_%zIdsJ5n
z6rE3!82qU3<um#f5aYPqO{x(wT?x|sR85EdCS<7?cC5rCiW4#%Q$1l(eT22#`L(Q%
zX>RgiX{*p-I)6ug`|2?0GS)v1$=YjRVWDi;b{AbEL<wa(jP@#2Ovq%v-$&clWvD(B
zd#0nRrsEjOSN9AA5P&Gc@v~`%*7!jKDaQpyc!<}lmYaGwCRW=MPtPd085@KcZua=J
zpFR7}E20L(Q-~Yqc!e%I=}$HUr32aYl@}H_kFB5$t$&+ISv62JF!OMYkD@Id<)d+R
zDvy<5LCV&T1oLF8{^hab1pe}dnie{@1v!IRuX4X(H9H%~WD26d*CKoRi8nDV)7Lv-
zBgG!(6$lG-bYD3h;2<x>eqlT3u{d{}aKA7jTk7s5B&KRkrtce}jqxlMQc7Gm2OsD1
z)OM-<bAJxKPn#q2CBBG98}OeB`Ju&?H@cFdrVUXhGx>SCdI(1KQ+6kzEX1P%eLo%T
zkX6hgRwRJ4&OqOZ9-x2b@SSOmTpW>PM%Lp5>6KZMq#Jk{8M8;~lO<yY;R*IZT-$Vd
z)W7;11&=A#<J+?Uj>wUmpCA75<{9ZmKX_B@IDd2*(#YGe7bGP?k0GebJo{I8^~ghQ
zJc>N=ii5vW1Wf|F@D<>;r<xMc-U~Mt*`{MOZbW}gCKhtagVIuGs)Y~fFypXW8_*)e
z`dAH~@k9k(*I!(0F`L;IQ9b{wNpBL?X$c<MG#deof~zsEe9V!Fzs1M!w?|0GZoOfT
zd4B-yfe=t)QNASuPc9rjFnbPr9-rhvmxx=x?HYqpw+1T#Pf*NI?5jFeJ2f{$uiA}O
zteb0*#WPdc;Emzzf^_06-JrF-wIOXI3eu+uC|w$>zzFy?k&+33{-skL4n4c=qLAOw
zBZvCM0%@{AQl+7ub-3G(+FKZBy8IPdB!8q+IX1A&O=P3g8qm9me>gS3Dydkrr(H5f
zq21C6He7%YnSR(Z_<*mm`N_1ShtAV%2peH=ucH+-3)eLC;`Ufz5WXk_7^UR<Q=Lj<
zO|qvIu{of)%Um%g0lmj<!}2gnSbyP&a-uC;_w}_DG*c+FS}VQ;NU<kmylmLa#DB-!
zjIgIIOZpOJ&B^{&w2r2`C+$4dYDmFb3KM5x^RSo2_M4<RcB9ulD6CG37(i*Qt^<<w
zQ9ocr@t#426!vutXgfKxdsgp=krk@r&7I~LeB$26Y4JXDD@}yj2>0o?b>^l~iCd2Y
z%L4aNGi}3OjBEOuwb{a@S?Nhb)PL$AdKZeaQ#_A&6x8l|e-=vou_|o#fG#{rMR*Ej
zkdFe?Ry`fMh0gf!KmzZ9zb6Iw>-kEnsv)(6A+i(u>Vq4e241Vwf872ybo2DipIE`b
zMmwKKIS$BNfWRxMq5v#lpq_*S|Bt}Uo57x8*TeUoVu~JHIL0;^(h#Ll$A67s6uX4=
zQ*j&#t!YX*9&qunrD+JK^TJJ6Mvn`aO?-lFL~I7e!wL~bZhMjsJK$UW>5_FW7!OD1
zlJac;Bheifmka1`9CVe~RysEwfqpiu@&bxq(6Rj~I7tq;eHma#ah4%G?Y6sTIrf*D
z;Uf&;8LjSsa4Ieygqd6)p?}+}V~;DsI6B^#v+DJ6?$2IGX=|mgwMEuAyNvmCDCs%m
z8oVt-F(6CtnUoOH{WUa~2bZj<NxYY^SOJ-(rCn-^OmmEku>S*2-w8W<@UC0|u|DX)
zv*Zz8-aUs~{G4Rn1!T}datA~7DiHbN6HRliKd$fej8X{J4N`p8cYhaHW?I-80H~(7
z;O!D~dMa!~%{Pbl4Fj9tg~cp_rI?Mj5gNun(&n#^ovAt8qS(KJY8j6gN2RMcR+CIw
zM^NaM8wcI2-1On2xzWpb>Yp$|AT@5SR%pBkD40_cP(adn6S(n2y~ce3Xs)5$TVd8a
z#n3A^M+_<x1r+8NhkuIk2@R*mm?z1~cfdu*&?i}Th*G>JY@m9~^jD|HRa98^8vS95
zd>+o9;IEY=#;t}MWt^kU9HrOw7phkGocqa3$z{CXzBnN$17~nO@;}ZgMR+<dVf>nu
zDkL={uL1_9F0U5p!)=Y#=7U%0@MQX_DDv?E9hNp1x8ldiOn)MZT{qF#%((>S@3KOS
zliGw&dvMJ}sxmdA`@?0wG3aJ40-!g39F~9<%A+G^^D7Vf>q@Z7ihu6*Ji>wINXx#V
z?x{>3pOvZr!({QL^ngv6=nZO+NRylPTZpdS$A@6dU7*|=K+xVc5X<W39J9jV*h37q
zaY`ee?K*FQ`G4<+oK<?wX5}|E!nQt^IcX)rl$br>0FP7QzXC&rsEf}j8;^}uFi}`A
z=ZO20bc*>`9nTpt{6Ag*IQrI~8jIqi8-X1VKYz9ctVOW+Fq3{TB<Gx+)R5PJok#;l
z@)a+Ks8`}jEnjf&5vm)~)j7=H66d<?Q50ghrw<;htbZODiwaT^8LRHRa!gH_NMnDE
z7>iu6_=O*$&&`-egFX*#IEMS{keXOwmSA4`6cfl;sk|7i%v4P}totP34X_}4QJ+Pv
zT&skfPXF5MIJx;x%b9wKn5O6YePV``R9@!h2i6L4xyzQ=ax-Oza{ocHpAf#g+jot!
z@~iVtHh)=QcBCN1h7Kr`9e+w(r4k*h1O7o#ChJ(e{+li^^Oo{KD4=A&+-Qsgk3Fy)
zKNm*m@N?|W#1=kYJ7dZ50(>F||FqttyUS2k|F*65nkrm%FxP)KZgRs3T*yB6gNm8k
zA$wGbpB*0+XZfDtdaB^7+8H^yKt7a9mbpK}!hh#N)6exf!fQt|Tt>6{DWD!S+|Aw_
zMOoX%9<e~kj2)?a)ZSd?zV@tLc<tOU#`N2^N@ojfgrr(`Y*4?~&d=?Cfhy~Ud;;_I
zOYZTC#)3V|Wo3HDb(<^zULOgd!(AZXNS#z(MaP91LC-^a@*l^*`hl{>>K1&!Pkue}
z3V*xY7xvxUtp_(Qim4JzbZhfb&b|jONHOL6;zDBm9VQ~#k7ZyOaqwPzd4Ct4F%osV
z{s1GU>4Np+9g8Yud}00GF|aV!*1*&V#-z`XT;)NA9=oT+vQT&+NR~6pq*r`D2-W~1
zW{K5Z;l{0E!O%9(=dOx6TFGt7Ck``Xn}5O5K*3$XU^`pOIi;Smf1G`h*Ow)%EK#oL
zyGEVqt+$-JGYZVtN9Y<T&@p4Q&L^}sLZ9fd2M691RzayD=z4c4IymELp^4)w-ysRW
zGu_PGZS0vWnR*DA7k}gk%n`)O2X53?oxXWY?vod+nYpj8qh~#91hb-p8a>>F(|?Ok
zKG8GTaXTmDPIVI|i;ctqwefE<cpY4y@TR45+fM*#DV~Og*u8&9P}ptY?oEYgW(ZjV
zV0VmoIKRVwUk2b!d1tJ}^(&-CJ>4818(43ES6*Tn_=-Loohw!{r&RO|Lv#7-nNmg%
zwe9y&9?$ew`jRG4MxNC3PjJHy#D9yUqqT{)tZs%<!t|WG;j#D8Scw`o%RN}f9l8iF
z+vpOivb8cg0q*0GF?i*LIsF6pU?a?x(<niC5kh`|bHocE6E;8k_Qn)=zY9jUaPyX@
zlUYdiQ-;kf8mBsz=3<|N)9+0*LIugyg|rR-p&?mg#&0np7%1n$(NHbW`F~Vb2wWxB
zmi59N3p@5=tdqJ6+$Q!+MHOVMK7Q(H;jK0G7KAx*Lm$k$g_32!T5<pgD)ZV7tSM{I
zk(+A0y<a%rP{=RaZNKU73L?pVMoIMqg42qEmrmE=`ZouLOxQ1}pMB<M$Xn!enR~|)
z-mIU0(u`tO4H<a1`35>ybAKN83Bo$p1<lxj`(GBL1{PNwb-Es<q-=$m_2mdW&N*kC
zI^qq!KIq)W&`&?pr_ja^g|Ys*>i6M8L467!wo!KiW2^5<r+2A-VuXCJ_1u+Eo^Lqt
zCoWmv{)<*d$7A+5D7OOexFv13gwYn2_Li4gM-+R~iHKO76z<)%ihs(_Q$V#{15q-k
zZc<1DUGb`pjX_`?0}fF}HtI$YY_w+V%jJK8bADK{X7Hn@%KJ&Z2H|Z0XYRKi4<#)M
z;N@Z8S#+2T15iv|awE*Cn?FXoJ;&PjSpP!*X~1sg7Bq}`W{+0Nsx|aiZW+QQZh@g@
zb*={~8symiUaCnoR)5i`E&;XkTTNMrmOEDlMCjbdGb?@4+}hA!?numOUiUSq==Vcm
zK+GH`y<Y*z`E)MG<LC;{!8Zcb^k=Q;!!3GeLX<U&Fd{B8?(V|_b$OhDtN!W?v@vn?
z@;CcuRUYwzextqAX?lqRBsUIXdiAvh57O)e=~w5}(K{i&6@Pb@&!c^ZsLDxOAJ>AC
zB-effMrVt#BCkTk935AE4E!y}fH}PQ;)T!JKCq`lzIfy4@L+u$Xo_K=t6nEAaq1Ni
z((Auf#9ihfx2UuCrEHY(b=Q_uf!@>#LXO{N{MqjCR;j5sc|Vuji7}(@(nkX`=O2UG
zM0#}-+Ax(>N`KAKb%rY$O7eD6SPflqigjfTY389BF{b&a6kw3ccn{9GoB~J7elF<K
zq|3+RY)~tO!>p~1S63pAmt9FX_MdSRZH|7Xw|)K;0ggFVcY8vSN-<B2=FfsJeyMiR
zlk~@O!<?Cuz{5pf{3Yc|VHug2N~2l)M0<D+d*;wfB7c0V(U1ZFz8?8BwBW2?L>tFX
z&8vTmroq=6AAC>Ww^E<1yez;4xOMcrM;_>P`|4;?ne_X}b$7*eNbT$VQFF#g{4YWJ
zo;VAd&XUNdG@=lmsq{`=2(CQY4+M{Qpu1yklpmLPp!B)uz4w*&kyyRyl<#k>MwJ=_
z!-%UdEPtK)T*epJy(Vx^yG-9w#-^5}sj$O28xc%a=FI5Kalc-yev|9g6C;mtS+5K+
z8U|b**7zbSqxFj9*7qv+Eu^4Y&2V7xnRG%Q&Ae61bA)c(s?MAob9d<zd~pAq_?<AG
zQwIB>Is3(7!4c5#DK1Hf?WfAX#YVh>@W7I?#eb_!5}}9{i-WhU=fEEun{vGs$Y_N6
z-!J*ije&&nc}*RaD<Q)mn%<`G2N-}R%@69!z7Qci33-Oy0kX<hb<+GDMFyq>R?1D%
z4Vqp3BbujfEGcyK>K9epC<PVV?K{pooqj}rNe!ke@pNSxLnu&Y;RM_ugYQ~%QkLo#
zo_`!Sbaqq_{APh^K#mv!4~_7ZsBpdjvOXz!?mDX)>uEm>0^)p1YcQo=a3zL2;P#bR
z(nXBJcY3+{7_h|39EFKw0)5nIr_gPc68%)Md#wRkHle)e&ctvfL9x<~gWn#jliXla
zUq&0ZWe^YcAbm`VZS@&Je+CA(8jtX`jDOK$3!3QTw2dyItR;E+>e@*58Yu_(*tGJh
z3&k;GIk0zp+iVTMMj-zww|=z+rpJbr5XITi*X$Eh5j2jJ`7&9nk%fUMX;SI?1?w3|
zw!DCkBYI)w?xb(HLQg?NO|7|r4zpccXBV*H4Vi&0r4?XjZ+neXQ40}EM9Ee#41Y8l
z`J39cnPyKmEImnWqHfxlvMt=~c=8@?J#9)cUJ8m4QgaRegS!iwr}5#Jg(43F#AW`F
zvLVA%2{y}F%dNLix;B!IS}7~#%NVmM<ZC+bHT=55Jnv&brsxrtS*X9;Owe)@Z#@;*
z#1;RqB$*W<z^|pduO!pvt^x%=+ke{x7G>b|F}^qO(>u-6t&{R|AOpR5=VYXE7n_&}
zX8;E}LAfqd3z`SNB9TTzFL2#)4b=_x^IERxW>KHB*@|YDSfCot?-07_nGkBn-|N{F
z=8{(487~)u6f06>rau5vUa;-);rN%#F^~c*JqgBy|3fpUY$VRTX&*@VMSs06;dSaC
z5?h2qxFWpf62)Mc-vEVH;<%L>%-|U#p&eAr)#>2(+l2)O1jY)m4sfdO($$;4XZ|2L
zX^y&6nTyP?vwE>;_4E-RV1-k^E}ywwhX)DC0{*_4c9mCm&6n8BF_s}FO#lYO!N|KA
zRlu;bu*+Uwph<iBiq|$b^M6nJ!$;DwgMAzJ_prl}e5LlXlY@VF5}#Otr_>+TOQ3rx
zOK_$lBQ>o^OpLyzPz%bu)Xvg{d_lgI81ek@zy1N`(q8gKg6E$;j~+>^61InBBjiw%
z;UUv~Z8n6f{Py{Yy|5-CL>A+m5xYehBWPvHGUAl<;ecku6_W(%8h<5rnjb!u-wJik
zJ!gWTJX3{Z0v_1IqS_{kY1BH0Dk7GCcnl(0`LPwKRu9L+1nv34aJsx=)R>TmSX<r_
z$9+O!tsUkA5te1Y2wfI|q4(XLY=}wg6e-hHN4+fkdOBycu?5)b5$H9$p&hiQC<g9w
zqKG|;EpB`zxNIspFMr#Mk&ZQ%#bUW$VMT?qY!tsfln=60&RcnwrRSIwsI@+gEP4cM
zi>eK=34guQmxwV7t1_GY6t;81z^i=-pr7NNmO^fLO+=pFTk6F)oaq3}a*$Q$EOC)V
z%ZV0~(0ZYKWFj16$w4ybu`!rrN7xJkEMNP8*@)ElrLhW}2!E#Bu{}1)2MMlDv9>yF
zLu!W@#gu^n?L`_vGQ%>nu4U^vD2|lLIoqI&oduDW_hy>Cd7U5#>E=p?yz@+ir!xay
zgR=7&-U%ikKqmYm$Xu(QVF6zUEd>eOY==-}vbl&vpKNUvrwY+_(S*eDXRqUW8~hW0
zv-AVN;j)OG3xDcS#z6KR;uJ|5X+v3<a?Jwzsd(9Ioq|Mg0xG`q6WMMy<g!Y|$TTR%
zlKFR3o<RY^eVxh2OOfCFIt(~vFe_qmk}MD|p+OGvWUfR8t#mu(Man6}-7ev3HHb&a
zzn~WHQHVST2|e!d5v$OvB|W2B%@JX>BIjH|JJ}3mEq~hqib5a**1YPR=PNw7iU*w{
zeU(=;DhBAp%r|>0pQc91^sh*>%v90e$i#4BY%D*+>7Fl}jn|G>88R}M;Cs)PvePK~
zgR*Pww<6zq78d)T_?Tb~jK(5cB#tKzBBPnOO>5tF>Zlo)y@2G1-bZ-&tdS-FDGw^R
zSmH7;Nq^Agzvu4THZZmcx~s%S7_-d{MzYGh)ag=ersE<Qh&^I4uIP4~1G52|31&J>
z`g-X($awjh8~@b+%3?h0UYm4+;%OpqI`kpv!b<b+@9n2Qpq~9E-im_$_ZsBI#0|mV
z+&%ldjE0G?Hppzcc!Bd3mJ(9)$S-xFEa-5bynmTF-;m*E0Lt_@Gm58KtzdPiR+^B(
zR1a=32P^>=bC|NQYwweHyX;!W42>ys&Hb8*BYtj<3jZ;Y><;4Ey~lkAFR8?ce5m*H
zP;cl!4?;*n_H75&$k_}7o3LLB@(mGss^G5`9T7b85{zql^H&F6KdTe1>zTh%UQ03I
zM1Ov$Y;n$PS1o`YMqFzj;DoUx|I0|9KsoAc!6B3AeV=|B_>$ZKZapx}e;yJR+{%Cz
zqesLw`g1g!w|A1xaS$nLZyRaeH{eE`74Id~ujVUvRpIQi<~nsAI4V=l&DBfz%ZbAa
zRqkOp(-Yy$XtWAk)HiJqpz|KgaimT`Xn$|ELvM;WGk$QVZH%@0*EK`7v2`J5r!}s{
z`=EM}y3P<a)qg?8!CEPR{52a3??&!Nl~HyuX3V2!`(Ze<51hD;TM+dVng2k|D}_Uz
zS{2cWOI?+0-IBp<!z1Xy(FNN)&{33c$uP*P^bU{Ail>+LZHWLL^H>KmRW$l1>VJ}Y
z2UHd7b0kC#Af7=6&7$l07yr0bPEZPD@PD%kp3b%Q-mix}RyeEoYz}nm^P!I*CPB$e
zhSs|o633b=r6ii(6uuFczhB%7geNYg%0kSK7`bB%vVwQkedo8?<*#gcl8*r}K;J!_
zR>~?^t*LzI8}C+LnBRszH^S1CwSP?Hdi%M$N_c-N+TK4peqZ>W>=7)U0G1;K7c>Vp
zHh=H|AgX;A#V&hPLHN--hGoW{#M5X}X;4^U@!@AFl&&7Us6W#VvxlYnTkYvIu{gRN
z2(;B~*%4=#zNd8ka`J3MsltlanA_5s2Z~)2@p2B49F0n&86~EvNqYD<NPk#3lzUAJ
z!yxDbm+zo7Y#*<|C@wF+3qTPFKtyA_bI|x1^8|U&(gqJVguV6$S9diUK;~H%dp~;r
z5IRTk&69jpWAj=g58+Vb<o;IVzUVR~1T4Bdu}^M33D0?E9`p*gnbSbN7mAdmYvdt)
z<-T@tVqM6-^rDRdX|TxUVt>~P!H2d{Ac`J490VF)hYK_pm*rn+G>a&wuu}|ND;`L8
z6yu|THVTX?1;sV~k>1W{e<OooL(hfK`T7I42a_)@PFudU-?}`6$yxXQTd*&k@24QQ
zZz)d~r1H%oyrh?XkWwtmm{5d4-rM}SuKVcZhEy57iY3$(;f(uD6@QAy9ioi$TJpvJ
zuXfECi$bSabQkTWTQ9=sG-PYnMXbPzZ{B`PE#22#<_BbqucWLagYU~aLMlL+Ta+Ln
z6gkXB7kC1EEjP67gW$bN#xqDZ5Up<uOVTcs>R?>lKbj9=PdpBF+FGwvB1g4SdC2h{
z{L>cfXc^tcNW;1K#D8T%{ikh+wlT3(P3`H!5)@qJX@KlwgByTXF+x$0-hLf=LrHpR
zsHx>6u2J8sLr?2=r3YQj5+Jd3YdggkrGa>^dKkP%+{tLJEG&tjMka97{ikeZEU=ld
zSd{ngSYSh1Q*jRj`6oK{s*}uUp~#!$24hvZ?)eAdQ<$bt+<z(7{qc)M47>sPiI3I5
zHVrXidpOAp*UjUC%`qktIp28DcNo45J;87Mg2kQIg=IwCSp3^l$d7T=#JoMP79gE$
zSTZri?k)YtBudme|DLU(`FHrf<@n?+Y$6(kcN$W1aD@6m=oD3_%cqAHz>ipkBsOsx
z&0M-Y4gkPhJAa<r%K{-N&}~fNiGBi3d0~DGI9`O~*F}9SQ(i=>D^r9|q#+u%rsv>^
zLl>#W)B1>B*Q*C|QNqRX0pFf_(AT27$-)(vbSo2-D9Y|DBq(PRy4lhPcIjk53@$KK
z1HF~s_ci+NadEV(U?RomuI+C_<@O;!Vm~@P55OXPOn;MnsY1ZWB5piW4`|AT-BWm8
zLDw+g*l29qw$UfH?KHL;?Ivm1G*)9<jjboPou;vE_RsfzIQ`G|%r)z9uDxar?p<H{
znLY!V^%hyn6QiG39Y`+_cTFWj-Z*xr!bx~b?b22(>v}XCZ~fO%qA&0~kdOB02E`8d
zLU}h)ElAwYu+JR~sCRk_F)69;gd^LchCjzTtA3FJX9mPZuvlfn;Uf$(^6WmJBvZs|
zeu@SxjtBh!pipVa?iS$QEEkkOLaues%;3VyzzE4O&=0!VyYaETNq<%@8fiM*>GGXy
zn28>mXPwu5^pF5X1@kMTax=EV2>nHN(D#rrG9-5b%2)nRYWKVQjW?$48^7{d14wE!
zLrqP-q+*^0F7keWPbi@stQ$zFk+6YTU+<Gku^@E;)-dec%L?dwT!WnAC~`tT>QF|F
zIx2a(V4v$IxI!JxR@3Dj76Id$Beb-)(4mYc(=iWH$Hp-Mio||A&4H1{j*TODYi0bl
zZ$Vu(M^7mku2NQ2e)wXnyzBH;Q<!>Yk+~jaU!PF}rlOn<E55uv=pm`noMj+~&$@RS
z&UqMUkM*+B%A`_yY?%lOooau|h>LdWL#{{eN@QX+?kL>4Y!}GKxc_W8)E>WO#O@?U
z_{x)%J#bw!NNrluD~hJzN>NaAOUSoU&;K%CwvX^KEL<yn?Xj=S>K}1Rn}p&S`$WU_
z=;V}5ca^ZHGj1HzWe0iQ8N^8-lb;KY_Q?Y3d<esg=(ZNCZoZk>u3M{>Ufb2VgkBJ=
z3BMb}Tn+p#sKX))1r`{;UgKhHX7<Nq|N3HtBoO(`i?H=3RLMS<r0bejX%8nAtjwZi
z-L9|B*>CoeHYqiFnHFNJ?x6pR$F|=Z*=ohZhruJeZ?-H5T8{l8A1TN!Jdi?U^=T2H
zYlE!WfQpEpO&o{J{S-53p?#%|Au_SzwDxfI4Bh(0{kV91&vd1uBB!4<)Km`^4n7xD
zjj%rBOy!&CSHaxDtT!{7-Jv<Mh{h2=$<U}0Dei3-X`HA!_Vg{Qg8E00h?XbaCUPAW
zlhk~ZtrD9?nm4=E=9prv1ZQ^9v6~&>K-(0)wiDn_&g=aw1z$e+ooC*2n>w8aWYM!b
zG~l$?kjOc^JblAgn~%UWF+qN9&x+MKkuYx*&8JYB0IJtD4G$N`rMZMU=RM(`rn`Z~
zxuE>AfJs*3_+;B}cem!p-*VMYd7c;c!9;8ZH%%m-UE5(0WLrrtvw~JwuA70V>wE`0
zGJDE6Nc=3-(syC;v+_8s^0YD|8JIX_IetHiIjb+<3L9k6Mv`#^^U_Ib$x?HjiDHS6
z^n(c_bD^-+d9IbXK#u6RVeHu`FDqvhlbWWQTL%KxUE6yZ1X#uBNv`lCi<W-Ht~ber
zgLGIa5sf^yVBCbFI$AAa?r`8!u;0_%`|t{~7pna9&}@KY2adlV3m*LGg@jmoiLBYw
z?6XgwOE#}poTGd?_)^{FH1hHYrV%z@VXO0&_m5OkzGZ}dlD7I@No-BB&=;&?gTW@^
zS8`VOwmVg{Jh(19+iwULDs9xZ#`6LrFyA44-0EvllEY(;mp6k5^cMl{_?WK*tSd8q
z(UiG{;}|b<98_MB%0dBY#%2FBt9K&P>JOC;i6k>trhEsE<$`MU@B6#?8z~qtQRMJX
zl8V8Xs>YZ*!P+wa*4K`i@r)FrD1<2znCwz`BLoSY132siNiWb?-BO-7bt$bLYTDAm
zEw@0`fk-5fWJ0+ANFD$w9?f9>!0^y(?c3Q2-s2JKv)JF{13SaoLuH#69H#9>Q33hf
z`ItE8V`;C*!;#uiEpXeKs}Sv1bGlxS->EUkB1*hl(Z0I!bbW5*9dIBRRSt3+2RTm8
zs7AlWTP=$KcfHI1s&+S$qwmqg*p-T@mx++7TBxZo^Z~sU{#t?OwJ<r6of}$Pw806K
zl{}lb{!^R2Z)DYL*QS~3gk_f1g=m|mZ{+`$Ga(%7z73t!!EFD7(CQ~1{coR8;6Dyw
zvj8&WuoV;+$A5&vW;|EOF4X^~3$dDy0wKSA*oDm>Ug03>|G0(COx2L^KOp}TjNL3F
z0JSFz#r>aPtY(Q|s91MMuI8CLC~TAeF%O%cHK9fM|2r7FS?MSA_fP*Dyq^UPLQdsk
zfnh-7_&@KEhl8K{gMDQH3~k+s_?q6TO=ifx*p2=iMC*`ON>2458uO~;cindD%r8xu
zw}`To9UXgp7I=W7mQYJ-_1WJ8S9~|c3J4k$;rE7|uB%L!*U{H5P8if~ngwa+)yeDp
zuA9tna#Jcb+^eiiDhsYlcGF6FkFX?8uaM^z)MV6D7c}}nUmd3bTMapTXy3}hhsU3j
z!5+8l1k;n;$#yPtrSfunG1Jm=cuXWrMTOPO=R}e5dG;09gWuudHS~tR?lIY4o>^IO
z<dSoEg*y{5h1~z1kx@2BKs^>AjqIIP)l5nTJBJM7mwVyIay^2C=qmyW7{e)iLwrm3
zEi}IRhnL2bo`ALizA)yHnvU{lkVUDrV#p@$CIliQK9=L&m80ui=qSH1pE&EIDlAH?
z>pi_V4Wf-44dNbD7-K<2a7Z}>d~7We-%P*VHzJBJx#2RMcX3jkOKnJLrOV4yC<{EQ
zF`AlbMF?8la5%Q-Hr2vid4~NCf`=F!-0;Mdnh(N}hdnz$ZyiNkv`}Azq|je_NGewL
z--Tb;HL+YV$fL7^Q@9^gXCSc|@_vM*U?$=xQxgb(HhcXGPkQkdpG0buHkik#`X#%6
zrn}*r9AiW8`gXNsr?zG=B<78zPQTWOTO7NfI*Bg_cLw*|Z2zZN>LVDKUokCiJ<@iv
z%d(uNl^&RuKtf@idm1YuV{nkFG$Ccx$C1v^=5>IaGI27u80BZ8jM%GcWTLz<!JBY-
zf5CBQm_TP)|Blmr-t)_*>x=B`tLM7*w&z8@2qwA=SWhTZiAlvxW?M;!#7O1&4Eufn
zXXs3C$&dY6z0+BB>sw*$F&HCVWg5*ZzW*FOoa*)(Ak|iwsy7Ve*Z)d6MjQ!WFrp{4
zb)JsCUFJhex((0$Djs_=RsJwt;q)!!uBnNga>v6lBC~f%J7Im~?g%9y3UA}GpyeMJ
z{T0TwTkHfR<x2Ud>nkXy+i=o)=O4!Q&B4^&$D<J8R2*4GIi6?I4LVXuI-M)Ed{_r2
zCEHid03G3^ga^3(eyr2WN6loi8V(4BdO6A^)Cw2x0xe>q$--*cKRa5HB(Exj*Xu3S
z)U*ZVB6DO!+?@29E(ZRJw5Rb!HPl16dpzyHze|9;2z<$kF!0muc6LB5MxS{F)5=XG
zn3}Pe<rqW|)*Qnv)>1y>zm`WN#*PH1D}w3(Qh{zbuvV9~{m7Cfs#hRthRaj>nQ5R>
zj2~-A6Nd>HzWk~FFQDm*8%Axc0(wity@7`IxoyD|gYL`_?ldD9{lkq~<o)|6jm@Y7
zACAH_FC^4Z$t#7tZGZ*?%LA8BGOakxsi%9~9utw>bq8%)dPga5G1_H1ZJRz(bH)Oo
zXvTY1`@a5hfw)jpB;63<`7?W__m{KU&39*baoQZ;aNRbFv0_BmMq8B7`}8^(Sc`ey
zbJ6|xLp$NFSHZLAx##>0BjcRK4O!xc=hThbfSG$*u#*qPBq$*L5im#5fOMXV7@Lxh
zo6G;<S271pf*)oqGg_-ziLr0`=Mn~}<t<VpJIw*HxPk2a@Jbd*Qhzq_4(IVM|D9db
zCRvJd#6>G_n46VQTWpGxP;YfHM_O5DQFb^wyl-%&T5=FRKWI^`=jo})zafW08kmST
zgD$1<Nh(iYgII&`M`Y)ew{V8U-{oxv;Y3-5LpU#e<8z&IU?kpG&QKMiR@Mdf<{_VL
zzuV|v4woL=%#)rs@clAsG8qh+Da@K*nnO;zK1|f*k!cq(IBjxRA^+h}3!3dtuJpo|
zYC6%;)A3`j@v?Ln>6u+eM_W>^Dms`5jIOgK=EWF$f(AF%7!`&#R+C&+NCYca7}|b<
z@-D%nX(jW+!wYEwHGKNgspkx=P8P%W?=R2il3EWhLNeLzKpTow%^0FKcfTO`C9-YM
zDO3BFXarg5RAxlrtiDSneWzf#OR9!`jhW3Yqf!60`s?2?f`l(!bJL90j02eqxv~>Y
zl$Gm7X>C@e)M_~F8}5pWZE$mje(Ub?-05pNmG_!}+l5oQ5t85^Vryq0jX~kXB*jI4
zc12)(UDSg%BqN4Sw&U;Q$S&*Pwg>k!eLnh6G|@EE0C)aYlD%Q_=0O;lXGfld(MJ<}
z_O$)xSedSsKt%rv=E{kl11G7!<{};A9{1FCHcpre;cpSgv;4uL5v7Fvg(T8HK1<d5
zN@x<<1mN9gJamN4p|QaM8@K9;znxBU%MY#dMcU6FT5P-dUgtBSzzp_g8Wh2G(hD|v
z$)K`-$J3Eku71~p1_(!1N<@ik!^d&KyV`XT+RDyLHR{DqO~f@GNIP0<4wR1ijO$&-
zYjd^RtNSYRF9azn7Dy{y3FjOPSabLVf(~mQlk5048A+3zf1|kpA)1ZTczIJ6(JDW%
zF~BW+^1JyB&AKQ)Ucg|cI&zB(OV6OaL<V9ywczB(7y-JIFhOsMwWUnK)zSsKwv-tT
zRs{t4ET<0#=coFcKW7b_0jJgva1BGfzg9dJg-9AUExyB1p`cNDH1?S<>PAmr5a{$i
z9@)5w;s@+Y1cuK53<e5<m62#=%-UPSeFYmN-B(L1HnZKY#sz(ppM6oogJ6gXxMq8Q
zUj(fTdCg|bk}j4H#aP?Nb~2`mG&MAGm@(<u`=1dt+IMw0wq5-(8#xst?rbJ)-Shim
zadu=`QzySLEE>NumuFKtGL*YGA_$>8z<^b!zBymmR%3P#{Miy5gG&>=rM3-E3u%v{
zCNT_=n*2$OM}rwLYLM_St6_j6ODjJt?d<L+<%(e2SaXyXtK6%N%n~2BhpB&h%di$^
z<Ay%Q6ppdgyz?TCL{o~5L1A+$!ApN6P1|<_J;XSR@yS^)*LW-)O`K-%%&fuV)CRjx
zDdad;Ze5B85bQx|ArO`d(@K@mkl{g<6CABMY)Fv)xHH^AAw3Oo{>^^WEi(G*dnUBR
z!K1-y%f5}oL954-;A}r3_1RZtB?|=4Y`ySDNxCI!Dk3GSTMv1(_G@W`V|-)d>-jtd
z?UiqTq#@@qTMr9y$!Bf(BDCzMqall&R}0G<V!k7o0Bfesoqvjx)Wbj58xc7#c)qXt
zm-ycwXC)U`cxEfnCVO+wM(Of9L&He1T2YPkUE`H7hADC+%h7^^ezD)wsZxpv$)^W8
z|Dw$O0aZpdK}@e=iLOaDOdhv6_d17$tWB{?4`qWU3icTJ^(hoP2NzSpe}YXU&g&<&
z=vO4GPrw}N43*v(h1D@#a`wbP`<`dQ2v1njjVVu3O3FfLaJE)`l>YOsqijr_)gvM-
z4N1^B_KHr#XO(q(yPTX{1kMwLq}T7%BUaJEUqAm$+%@CEysUj1Wi?2lC-4o0&DpE*
z@Yd5YmuMVKK%;olA=Qc`ekidyR>WGy){~qv5CWL@S;!`-N}*6?;w5Pm-~Zi^JwceQ
znC3IDVGO}TIM+EVElH@+tIy102a7YN>J_(Cu;`h$DSM+cXo!_5sxT<5skSIDi6*dJ
zYyUp_ik9gR78RVbV~QKMz06z&y<h;g)G91+PZvRbp-UMNaYug=R4wmOqY<MK<D(6a
z_zdXiZ7CYkt1%{3{j}1~-HsXgS(a>HHBPx)^o-(F?ydViE+MZm#98E@z%hFXGdRGV
z;+-gYl&(-GGAVt}WGm9G52q49%T|`KsYqIHyI9@CU?s(1{_e2p@I$$^rPJ%Z!%B&w
z<Mm8kd2p{PAL+<AK~Yqsh*h8&S)i7-xDcCDuC_dr>7WkZ(l@=NQLqzsh=4Wqnh%Bz
z;1L#<I?4;dnfXLtIuJ+tId4#4k~giqsJuA|)^bo)zQfr1c0|XOGDm?_gLTDnQ6VD*
z*D&Qy-#q|IFrGbmranmZi03G06vl$}hv?WW6#=VQ-B#}b+oZmmUXxp<ZB=TJnlYPI
zUEXW*s}ZZMn#5RS`;~FQ`i51?Q~+Bxpf;{=%3P$Vk!Cu@6mN;ZfZTGU14gAx<n)a9
z-O!Y$go9;kLeUSU)@D1i+|r}$6i-d)(@l1wPxjG2Py9Zo#ZW{Ls=3-1^F*ppF;LL-
zR@$ic_@s-AGC`|sDF4e({YN4-o=s4OhMX}?Jh4@G0-lMx$WiC?@O|%Fp`cR+;HR>#
z9K>+f{kK|Q_|N3qDrRlbXc1;5sdvhkf|kV*TV`s`Z)MsnF9=0utc8i+t#@YPUlg#^
z*Oicx!rFGFkH7*wu3Uu`f#+Wck;XuYK!JBZJ;|PnwdWs?v=_{!%V7R=W=9d4fm&Oy
zW=(iGFeL9PH@u~%_O-8!>r~7IaQ!TM)t462K{8Riqj(HD{A-SSDDvcGa9PwjpGGHi
z`H_?`(yzK)U4jtM!`v9i42?H?Gt5PlB%;XCGOwwvKdY?a<ZV?l9a-8aWaxd-W2K?!
zy_8pB5w_d?h0sA@js+GlVPauVR?tj%C<%ZZN5uVnyDj)Ij4^&GoFPsH@Z7oyOcf{_
zhcyStqFF`<VI}vpqeg8K7Uot6(5j)?Fgw=b%NWS4ObSm~)rlHoVL?ZT*gjCaJW?j5
zzS|2h5i0av0>?p%^?72g@cN8#L?6GSiy)^Ml{Q%Vdytb3RA#O`k=ZD#C{q~-(h~8D
zUiGa>0}J?3IE`y%R+<*L?W2Sod#<$b)0h$vSOU$pDBcJpz-C+NsdqV;<hF_8Y9%~?
zIJE=SL2=XcuB>C<dwhRcm-Gbe1n1WVufg>K_3;V0N}zO5OSPk3bqOZlFE9HkcLILK
zfftq4xC*7Ek#GN*#J3!7&+W{p&RF||Qx((Lc87atwIsv`fS>KA7`H8tl-p{FlIcj5
zmM;lRkkrJB+}S9d{vU4f_I1C69ZWqsA|<uGmK)44|E%!{D{b+7@9~dJuZI5F-ZrY8
zz_`31ar9M)Ijk{B@d33nW^OvzSmuwo$+{d;p*`ynJ&j{7-bvW7eSzln7BtOJ5anDQ
zt5Fd&f?!H5z+~_;29ESlai>;TYA5v@ADcJP)1RbwQxbgsTaexp&Id^q^CS+wnNi%`
z_#>-K--_df{d=n;Y6Ib{fBFakcE++Os5$9LaN2sw+!hw`zU-)1z4%U9^qbjg?GN4K
z%o=rKZ`Yuojqd9QLmd97f(q}*S^hnVnRb0tR>41~fbKcNT`y$4wj4BV6tdjHUwVHO
z=x1<58%8F=jnZYxuNN=Ptj^=dgcZi5DjJ{<HTF{#GZ*ujPnO5IowfNol9hlENZZLi
zY7&#_6r7vI5)$qqbLLNv(GgyKj$LT!EJj6A5b5M#w_8(&tkKjmkK0ruca#*H!mnJe
zU#Iy00Cp-8Kf32|TW2~j1})W)vDF>t#}UgTGhf`>Lmk?j9qQ=p%u>|-*(j2#TVGW(
z#pYa33cOZxJa$%W5@JkD?yFr6W(w%K*TtUJz&r34mLWtTe+C*j<+dTa1>C0Ej<dJ1
z1fJ`akh@vWRbFo2BYwpBFB6G5w*OFgMLCz91}qN$h{*BfO!$UW1Aq^Q4U-4JX<rkV
zr_oP!#mXmHDnXXoiAZws{&~EuSMO!%_H~Fvf8;e>k?G(l<L>L-=~IEsRPNMOJ$wds
z0xwfxp^$W*U*V7;O{ZFs_jW_powQK4SICTN6x;Smd9W8A>}IF`<MZR*rZI6hfze$U
zP}llk%FW+yq=IF$Pg$xs7)^03siA)MqP!7sB`%B|+i!6tEsXW=6I=gp9~$EyeB08m
zDA<8mBCd?j`C^3AKP(F>Dzf>ps`&8^9=l(@R|N5s8L9M#wqU2445gl_dTbZqM=)_<
zQ*jVBnKKXGp3mlr<P=pPqJEBElj9S_0)U9fw5X$8zD{f|j`qxksutXw{sYhHH`ZJ+
z;QqGvWC(eB!6?@cD(l^U(HCwbHFq;Nbp`E~)(Pef@^4@h)E)6IlyP4Q`&qx((qHNx
zh=l_+xZ{dP9U+1$a~$v9<D)rmHe(F!q0W`PSd+uhEGKE{fH`ABbl=1^GmWrP;P|j6
zj>XE!E;8L#emWhBv$B*=d|$r5ff++6>-L#!jez5@p*`)NCE?5kY^T$E3;knL3*SFa
zisB`ROyfR*lj0qTi<7$G_eW3IsaKt7gYBo6F*=oOu_+%~t8Un0?Ui00T7Piw5erGC
zJulR=>MO#t)p2}@!n%Db=|#t1;7_i%r|%*(<dA!TRpJ@zw53xE-%qUzY$(!9;eeO>
zRGgmI4WE&(-wHI}+k5sFKrILnTjjn&u-|0<J-*Gn<wQZal-S*>koD9KS09q0f7@i*
zP%~U$b=r$LudVuu@E9Kg(fRjY7(O`lGrbQ}AVxPUl%!cK)1a(!WyWny4KN9RmfD!x
z8;f^Fw$vG9(|y}pPAoGi)cemFiOwC}XW>L;n(F=JFMJ{&F%8ro5{to*69%XWvZVsO
z3@7%WHI>7_)qqxv0REhj^E+RkMdjkRk2PI5^;M0hbA@-5;*>Fo*w+W&$uQ^3br>Z}
z)z>M(PXyz=mX{eLakSwsG{D``-8xG-akuZ)OGaz<7Ll*psJw6baA9-!$<aA^z^L!O
zv<h`M)tZ@8H3nDcUhq$3_z_JZA(Hikb!u}f)(z?9Q|vFs*~0!e+Z)7NkEq`#GUc$V
zgIxIWw^{`Hgh$VJSlG|l`V^uVZyc<%M_kbAk-*9`J6Ex6o3FR-TY&MLDIW%l<wbdj
zP=w9r)7p%(dnV{O8|m`;?HhVQ3)=PRfRgF!%Mufl%#M-xo86OzBOCTN>uuZ$zs%;D
zZuOl2NC?WKz}`6r>po=mpkV2C1iTga9Ak7i7O1}5PRW1yMA_b8r+Y0!PpOuZtHO4D
zBdzqTj#Kb~Ga&TtN{N%^yoar8IV+|Fp-jJO^hny$FXUqOSx3u|e;I&AQ~{U$SAi1?
zisyf7no?!ZHi)T0QZOi~hPjXk&E0j-7Otr?Ssz3IA2rM?0yGCNU+Sv_OavS!Cl@y-
zUuu*j49Wj~S4qOeeAGjQrC^8&_>tH^wr=ho<Xjv)|5vc&5Mbv>b(Ml)0Cer0w}&y`
zCTk936NN@oUTh8YtFRN|hQt$s<Y*HY{Mo0!UEpW@o^Yt^e%EfN;NJf|f}@YvS{9fs
z>e*KA$VhO7ket9iS$i9K4Vz~y4yB<kM~}7qmq8MCQ61gh%Jki|YD$>0mgY8q(go>w
z1g_{i$K-ElWdc(wI>qZaQ$VL!l8ekN=oUfeyB{gqHDU=8hv~?KZ-PtEjq57|DYwlx
z037GSEOk0WkC7B6{WoEJNd50rt~!DQA^Wl=u9-B9JO<f`6anG9v~MO4J>e|!C(%h-
z0x(4s=1B6d5DW%acuiNzWQo{DPH|>UIZAMOeV<sL^Xt1GbXUhXHxQ(qlJcbufA^|#
z{2FAnDXrlk@N~a-UH0~S<LE;%ec(L%B8Nfjedm)cU2K8*6PfZf!kWvE79e$xB@RDC
zmxq?~7kmBlEzvx8F5z+3zQ>lkeg4Wv9TVpgF3n15jRVy+y@FGIyH`$7nS0UnbItr5
zyoVYzDP93g3I7Q7DB#ydO=kM(AVeCyZbiYERFAj{Uq(Yt4R2$R4Ct`U$#Ob5dhb2u
zQs6p1+sx}}jh#8@KNxP1S+M)TiL$|>y?=sBU~yt_n?m&QdMavm4@0H|^6GRDQA-Hx
z?O$Z!KE4QYM)|Cv0rW#K1e{(%(PO4E6@D<>n>#5Dd$8K80syAZIA9M$R5Z56%92F+
z7fKe_gRL&RqbVhrbMI-~zk|*BA;-hj(=XZP%M`t(5_z!CwXQ@n>^LCQTN?L#Z27yS
z?Gq!>-?ACz$>y6w(Ih1r+SC6VHATDeh_FYgGX%3eB4=;GzK3;tW>@(scMQ32B$kBV
zoWy1s*GqMt0q>F=q{#06t$(R4|BM<*MtXi0JiqpOl=q3L!S)XD?CL^qY>mC~h(q1+
zh#u=NhzIGzw~+&!WS~I}cFEYE<|x#;Iq-7ajq7MHkvrrlbP9Z?(yl9;$G>&rHM{Db
zbxTdRq)!-k!bfigbr$WD?HJ#%uWtHk76z-V83!;v16A`83G|EBG)wN-r?7^7IELVo
zP!H#=7#N4VO+TM+HJ}0Nq!?z|!-C46=|6*IlFg0Etm(&cl85D!<|@JvupliLc)PkI
zot-~xiIEC95LD#Igpjov^0JOwapm={{#`xfKoq9mmvtj3z`S*r3pUl_i2k96+1Y_g
zV>P{G1$>u>5@9l@CepxyP5KProBfffi7vqidnlq}Iv@<!)HJ32u7Seh5@Uuhh<wcy
zr8Lh8li|^lk1y2(&mc-Le#yW*lXfXAbz|@=Yw)O}r?M3{eS-A)hPu}xxH^6BN>eH`
zW}BPUM4Zl7e-9qkaU0&nBEfy2y}JLu*uFp_0B+Ef5Y=tI__8S*G5Ys{ExYdh2&sFS
zmPmXc%~SGn0J{EQeJpihwYlu@61K)~_M8~ZH-}&7Quw;hR}2(`CFo+zcc9G_AAiOI
zR|QegnW)6P_bvil4%UemE30&QMbe4V>Eh6ym6ZXo+{KP<$&)VZP$Rva6G-cOuy!UM
zP=sh<od?%Iwe(DgK87ebf72Ac7jUv^Y?PB_wDp`-v9Aqbht;$CJg=A21-uPq%(YLV
z7T-DX{!|}7nP4Vx3c{E-V+Zh1tj#*?YTE3(h?egbtZu6BkfP5Z20I1WDZHR8!Xdow
zkB&}v>%q&%w3<UrIov2COb$LJA`nf$*rel2S$jY3X&9|83E`$3kqCk#U+0|Diz_0_
zI<!&;xZn>(ZsVkLTTVAWX=8Ib%QP=qfvOPtjs$Vr%c4<87+*`9)E*)CKL0IwrIE2{
zle57Oyquc@^%B_Pi_QZxDUWPu4i}Y>a;n1XH-sD~R@UkoJcKHhhdwN4^hR!AIajrl
z%(UM7t8?cI()4DCGZ&7Jy(sU~j#l711w?k>`|7#pl8MuU8hvm8X)&ocJ~k1=z(2^#
z=XZkxf0DnXNJY))IT!nGUlDx%-MZ$YIq!T^J-<Gfl|c`xdAHLY+F80vTe{Q{9@PDd
z2nMc>%_s*JYqzdY)t9?7|E^R6V-$DjL1)KJ?pHp|{Y^4Q<~PQ>b;awt_q2=W`M+5&
zwL~+h(AJl-zUt6dI6EyIs$6|CVJ<I==;Xm7x`+&wJY4}ND_sRl?fNh6c4hU}G=L<p
zRIozFg6_A!bYZnojW1z1yt<^I#`sw!eUKfKuUqBwbDM`4YNNkbFs@nwiyJCT*&$q;
z_BWc<E~8JYSKO@j_AGvmf6I<>g0Hex{2QR{r(9y;vg|4Oi;0PPS3H_(F)}-^TTN4t
zS=t@Q%Z^>icv)GVw;%>ch~l4|{JssBT7*aVLsm^g-$eZ*3(1ZW%-pkcYDSf<;6-0b
zzWa*GQfVBy3zM%M>DeI)5Xxb7XovU98OyZ6^_{g7Rh#{ef&7)^i85SVSsxl*{;X@M
zHImh`sEeXgd+N%bGc@6_f6~HolyCVX;bQKrDU##XRII#dH+K$!jd!KRY-;~l(IBc+
zpIU!zbtH`N963OUEs>uG{>iI&F40cl`q}M8jQ@p)9m&ec(%r)?6(1QA1&WtH^;iyu
z5)d##V$-nowIb)>6+mLsAvfS4=OX9$=u~%h_W0=JC1)pBM`Dw6vUdJx@%>Lrn%qE`
zgXc3JCy0kjLRwNlT2ewnf`?y<LyAXITAGhrLJGtsO8);JVubpC`2jZ<5BbN2COtqh
ztKj@o4SVqF?(|`LR_ZR7wMwIz@i>rasA?g<s;S&&!@wU3WKA7SgVLWJCySr<2uDrb
z$9D;v#$5&(iSPP8wVQg7*?I5lU8gPG-;EtFIk1)^fk{oDD}gZ~O5fWd*ouhTXHM_V
zo}(G0H4Qap%EKJ2ZBvleUZ$pY?gAXZ5;Ie@;tWJiMzWU4sg<==6O@h>c3`L6<do5u
zcN+~!To364ukrOEp^X#TXnAx$pdMqQs|fQw#h*<>+M97X1efO@{HUo4N3F#WG$T3v
zel}fnh>GVN>a+~C<`dWLir{M+Zhl$+!y8Px4C`k`b{Y<?8Ul8KiRt6uhXmkpuwvJb
z`EMo3Zov?lyRfK*8q~CD%8B@Hgw$VQ!%Xofdk~`g@IO{e_3UT|rcje$*p||?wu@#(
z#`6h5C5G%x*EZ^8fZr(PJkYI;Qp9K9L(6wz+e69sVdAJw{Va73wJ6+Oc}^b4=AS4{
z^WX8%7})7nZr1GZKlU&f;10(tBxXeQueo?z;{BH-&y#AIB34h{ic~3X4a-?DF)qw?
mGd}KKBL7x%QUwwn1eW}jW>dc?!jK{H32-CP(0o<_A^kt8{FM#>

delta 58244
zcmZsiQ+p)<lWv2KZQHhO+v=E|?AX{H+qR94ZQHhOJLj7-b8(*M%w5$lcx%<FddfkG
zOGk+#2jgT<RVe_a1|I3kIUfw8bYE!YJ)_pr$1pSTFDD0Hc^Dkoc23KN(>2|J%88QN
zqQ-&F>K?D0T%3WzD94R%Jy9g9CLjCfr2DxmAr<a2h6UU(7NA|NaA9fvQ%4^ar?ysb
zaE`1&WrZ{O6RcwG&;Y+3EX4U_TVwtzS9dxqV@usuP6zl&0Ww*<cGFe{Q2oa8R7+mk
zI=Vc`ya7)RQ!h7)(X{lNF^^|j@qXostns^<rpC>wOWmxHmS`ONh46|-X=OFK)v|-a
z!#XOCS(#Rrz5(q^FsVS<*y!_Uc+WU;T#ndNasO1qpog2z{4a(7YE(slC+&1>Z>|z(
zFk|VA^LytL1L(a@7lp8nlF};%K{FjgpjC}YE7J=~!<z2F9`Vt6#wo#}rX%gc!CPhD
z_T?y%C~P<P`+icW;@&-jE~EK!bL1|9>#E%?kM=VkKxjdW;0&2*vEAC`MR5SSk)W<E
zKiPz7U4XN7u}`@1hhSU|;5}9g>CgNp`JjzGJPZ|W5_o`Jad`CA-1;Y?nZY}G45`Y!
zkis1#(f)d6@=p|(L<{MW+Ft$9fMl3xdpTMUJb4@kD>E;K?!hbqVwaA{-|><?N+!B<
z3gh1Zj|gljyOxZxxZAuw0c&3Ekqkw)uhg33C%YDd>rbklqY}U;%S6G%1jh?7cg|Wt
z$6gQe4-gXHft$se+itBs8($@H(Ar3yKF?WRQHNo9qxc|Nmefdbgp$=~!3b6TuqYXU
zmoF!Dy_sO(_fPYbBk#-)U;SL>%8C^^kRZ=VCK_QUWuhZMDCTx}bu6M~4pLSBCP(}?
zt{Z94=S8PoHhHv|V^4Y7RJ4qa`lI70X2zUR0q{p0zBC-Lq0Zk>##MB7BNL=icQ_V`
z%If-;%&tGRMmtWSs^Y%IZtoFhm2eNA#u#;o``?72^+y24_MP6Jzyo36sPwYI21grL
zH0U*dpI0<RWC=hoC(vZjxSKK6^JLUrJ52g1EKQwU6@oX?C<un^et}*iD1-&|uzF>a
z3<%0jZfsico5h-t8>9y+hid4`pYM8`?fb$&!+zXnD+bgVpc_JqNl~qTYbJH+k+*g6
zRv&aj3UhoR#qMEY<3wI(d3@EZ>!KJbpA-T|jrePL+Il<@>IQ$4uAp<Q((r6{Uy0%0
z_6>&%fE#=qy%Z&>{PFLmMNHWqFPE2gG7xUvZD>H_8mrDHvH*%MHCXGO26yjhe{{ln
zgO?&GnZL@aEoN^$viaego^&s>ODiuNwL%XC)nM+?X^iiNvT&P4lD>OD@a??@kav=a
zNxreln%j)v(ntbySu3QAjaJnoGr2HN*l}art-xMH!|d_a3l}rW?lDZgikPB(2A;=G
zmuNX^-({4b#vVhDW{U3XN4UL9%OCx#8o0(@Emd0~h|M13jr8p=C+-jDe)x2QL>P;I
zG0hwfcU(1TJz`?_f1P)_m~^;ih+C!?_-BoQLNZX<zPIPyd{dXaKskM*_Pjt3hjdW@
zUUE&~GPk)rHJp1X#LnR>_L&nO3G|ELBnbZ{5@o*7d>a?~DNKO_#bDENy_;knPjXu&
zg95(^ryq;}_dRY~0#SRP;eAN;jvVi)TQ7Dfx+5{zFSg=rUk-S6O>cPs&cn>^b?S2W
z(r{H!bZX;z({Q}af6U@~RY@oTQwaFFs4>e^0#6mD(b}uKkZz|}FB#qvfT6bIirefi
zp22RL?uJ|a_W%hyEsxwy0N17?4|<+Gfu42A>#kiL>9Z+9hYXlaTx&CYvQj8=GER{~
z&hv(s!aIV_>!$?L0~XCZyiPub_2EO29umi-34;?|)^4}jWzK3CD#yZr&ns%UaF)Zh
zdt+*&=!ivarkTbYTVlW=5a#8tHaV{;^p*Z)3l~<8!R47eYku6KoUt7)Jtx+UUrYdn
zV7xp{AJE}%X0Pyo^&D?&_bZ<v=Wy$-WF4w3p5Qb$yMy7ZjBv$nsSIHZW?{$$Xnewb
z*eJPxh^%KtgRy92)~vDxZ%|YL%r~(m7a1h-Km$xnDQVMP&OxFj&_cz_=<(3HW@Uhy
z3IUQ7y*}uE^_!f*q^KMIRhv>)VzlX?p$HWw>mf=zNl~~v3waVHcy<H6%ReiJ7_9{y
zrbsZQvO}f@_s`63#kW{%MsI?n9e-dYFWPcj165nbo=Np2xt5e}wL`3oqs3uT$BK=B
zOex2ic4dzdo8RCK5Qc4;*Uf!js5kCK&o_sr(pf<#*KYh-zqnzfgg%$`h-sHO8TJUx
z3j!-+{4&Vo?3bDJQ)pUch>3OCecIbvQddbUNv_?$HK_&w*BGxe^Qq}_bh=3+I?h`*
zTh$b7dG1~~3uJ03nQ7u&V-rOh_rUrwZ-S^A0US;l=N(xf5RXM(acD|iJRLD<P89UL
zfwXi7mPV@NN_gg5$<*?m=l5?5Qc*&;7=6VN#ZAbu2_%RIAw0e(%OOQ=%pea)t(mZF
zmqZ21ws&r~VI>`BH*~(lhQ{|gkVAFmCeduEt23kX7F^UnLL_PZ68t5$qu{Tj+=$1C
z4yHsoX|^7qKpws@zX*k!rWyW9UrFh5sC9Q2+)3u*_8-u5e6D+)5)2@`;|vcypU9?k
zucWYLKEX3F7mxoF*;_Ziis(y=0|my)7gCEuIaCuH?i(zkmp;{<@I=2x@D@hq^5WED
zb&ETSHGNV1!9xtwKQTtKmIU2vwhtIWq_V!ZS^vT<;Lg-BXH-s~3?1yfMd$eh8dYC9
z)pMBr{mDssH2tzUN})szHlUxY1`+i8imsF0itpvu1N%cqhMuAIKJ6k}F(rnv>ifEn
zqG+`G>+utZdk6;*7wPvOW~V3A#;uB698J>#`pXQ06%-hcjW>Z-<HP46%bdp{>+|>X
zF1|+`AP)|Cc)0TuIIm!qA1Ud&c?PfAV<S^8kEHJAdO;yY2ik~s1|gFUW;eSwL!^z!
z5Jy>S<?&ECaI{aHskT%@VMv{%@nC94z|yo16Z2U}*P*tqEk=wh>_yP~G&*n91*=%v
zengV%nT>Kfj)*8P#@3_NoO1s#k?WJW|CK@sc%69?jozl0h3|Ns6iu3O@fhZ5Rye!s
zo6X_ox||Nb!hc(=X*77_`nx&Q$*Gx%>V@O|fD0o0!WiANSnw6ZYr)QUg5&89Tx<Nv
z!D}`y-}}IcV~_r`g6Sc#=E$!yv^%3>-d8~IiQ5@8ZO_-zkhdjtq*4I%IPLE$smgL7
zko4=9svVHDrP537c0u)_@`qfY?z25E_}nShYo4ri^y}SDk6HbCL}Sd{1D9UTq0^`5
z2~_?rkWr*X;w33IN%J6ON$84uXQRhK+H1qk$z-qQ2x*$mB!|dv<(ehnSZPprP=Av~
zN-EpIbUeXr^)BOND^0dadTA~J4Ip0(jH9=8Z5)OX4skv3+_iRL9b9Fjk~ER^I^#JC
zL|^J-^GYddGi=6RHlbpakHeSrA@BO>RSGIHom1u=osX*~X5soKnYf@}GmxT6Q(QUu
zCF0jOguWtmQlgHzbRa>sN=ShWVEtXs=NOYPO1bMI73rh?%bbVH{m~g2TVzNKn2V?U
z?-NPq-!9Zf?&;Z}K>BE4YJmumBN89tJs&}v{OfW!PGBHA!RumSsCqg`D7IdV>j(DC
zKltmn6Sn}p$G;hbh6IwAeP2GgPXc{+@>Zwk++svm{&LA{CvSo?qJRzbFFfj5Z|u%z
z@6@t0tzX|Xo1ThZ$5gdfdtGBcK-iO8G|)qvx8avpDhioa>XeNV?)P|P6WAIFn6g4J
zG{8UDI$y;X<INrX5;#f-&rf7!tD?*Y=-n>}VO=EdO%yFH{cP_7^AvOE_a#J}I*vWK
zA2DhdD50SGFQok0d?L}g8;aLZ<+Rlw<Op>mM5y?=4~(DJP^5e!DFh%k5F%RaWoQEH
z8cV^z{pbGtd~Z+0c0yPE38PCgMZ#IbmiBY!&>-1etED4l*#xPoS&PD{1j*UG=tyc{
zE?mNu%Hv_`9Rh|^WpBg8;P4X@sj4N}yR7({Ye)Idz(0%GLjV!#vC!fY*OP@BJKVGk
zsEznw1|iPt^dQkW9N~o%qKf*JcGk|e{wG#ozy3}l0z4HMD^KcNE;vo1Av`RQSs6XP
zfkiAn!5rgu;CUY+0$sZi@_m*8OtQcZ-~ujVzwhH4x<8s?Mg{s#ZZn$hY{88cuGoym
z`}XO-|8VjYx}m|HTST7S@MhMP9uzc>{(MTF6_k}|)}1za5X4NVw0v}!Yh<t9ls(Oj
z1m2P8cUL;z(u)3{fyli-ZJi0&v}t1<RYCXjTec&Rp&8oA`9Pu~qrY;AdfQR}xut>2
zYFo%@tr5yRd4e`^0l3lwzJqc>O=-I)<88Lh5)%N$0bx4a<5SEc;Z0mEZd^(agGh{+
zeYAmbBSjr8ok0&`lO1^UFKOs|mN<$<C#Z@~j9*A}fNaD{1eKc^swN9)Y^DX97a4dU
z9!Mo6^4>Iyy=fP1obNxhdW}Hnq-aGGk>W}>wxp?}VuW41EUM8;R)wvPwYm_RZ_JSS
z%!%rbcfT3PI8IN_e}^dS!pGSL_YW$+G%4Q$lpZ09`pzm90mN+71yh3k)w_K+`$4IK
zx-){>M>Kk@e;JdDP?ryQzZ;;{QqxEIn-?K!Qka|Xy??lSHc6ahTe8}@ybClJ&(*Q)
zTon|K31T~L4pk&5w1(jR<@!3S)08APvtXIpcfEX)U_Bco0ZrIFpTEB328@0^vm@!s
zF(c0Ex?Jd0pYu7*!#2tNj+Bxm`tz<S773{vKX0|)w9fWf=SUi;ORtY<Ya>ubhFMht
zCyk*Zptx}^(nZU6Mtx9RF&^Q&9RgRxSAN$xp=aK;I3i-xOEYE8g{HVq{WRDSwA;-f
zE-WXf)!(3>T<?T10>uV4VXVBk+_Q*RWY&Qy#5s8F*^xoj<y4ZW18rZUY9M0oVxVZr
z5*b)eb8TmU6*~^Z1!HPDYTHpUSNn6ZlABY!*VBy1BS#`*+PEmB0eb<De(uj6T4H`N
zSR>bF32t{55KNd__{|eog?~D2xI&?69?SSJLV2xD5!_wLX8gzZz?`m?+W4n_ohfgu
z^Te!)DRW@9uEKMJN%~)>-iv}K4=W3*D%=#a;yiM7S@jOkkT6hMxOrv%$iV%})b@^F
z84RL+N}P|<<x8P?t*dkYO=CwmGNM+<V@PPQ@R3fdr5o2Wsn(Nw5U*bY0cCfixmbFx
z2RS!s_BF^0aDFSKU{DkcHDjAHf7z^iG=2UsWjq-=@R(^DapV+OzTY0CZ)|^R{k>Sh
z3VSpZ@)`$JH0SOuArJrO55lfyX}sJnq3v^pdG_=W#S^)*GN~j|%6h?mmn7tgo#n%T
zgb(`62s-$U{Y<|tNs~_oMpDoJnrQETk8+Ar^1va|xp4|_b4UlASb2lt#O-ZQY7N56
zZvwS|N`7S84RF_8@2crkyzvvtA-|OGpCON*Sz`oRuLp_!P4!1gtw&6J&tCs>vhBU9
z{o_6P+g_Y<jG||A8$^d6ZAIlq#Tq1fCN;B$C(<H4nLJSBo}9h0(e2USon9f78G?O}
zDtTk8zS2TP=haNPf7GWfAe>r>!nbkuPuS@eleg)f-ho7nzR%exofxCLq)Aa1kVbv&
z_?H>*sjfOMtUE<N?|FXRWxV)R^Wvi)3Dwe^StpCQ$R1-9!ZZ``$$j;Ze^{g?)NeU^
zY-9V%&z|d)wF_on%4mZhiUfvr{b9GFFb#RK>aKE0QL`0#rt;=Qy`^s5(pN!tidu|k
z!B<@|Lb2%UK33vi-Vim@ttHt*kTvi^5E{U@YX&-l8H#}Jrw4pRl0hy=gf$14IlDNS
z8{7R?Wp82w%f?L1O#EL3KR+y!yg9(q#fq4Xjh*?w_bXSp6M)0vyPAJ#WKg$_Yf7A5
zw<M5m;85G#;lId-uX*z<bm>eiM_GvI6cE3k)t>^*)Ux6)>3zmUaO-uJ3=9lz&Rw1y
zlWV?xlHYf305#oyh4mVWPD6|5#{dZ)s`Pm}3UXsF*9N^VxtNcb#+Rpo5A~8;sTCr?
z_x%%|B#_GL%w~YCfWj3UsOOx5uF$d#?IT=dzl@iBemG*Gr77X;Wa!5%ap7MBJc}IO
zWm7mGk9K_-7U<MDp|y@UN1MJFxqi3VNCth(zr5a3GhN@W6~T?Cm6c&MUzO7FC6Do{
zD9NOx0pE3z&|)tnKUkAJP6(5)0wDf<ooQ;G06OLLi+%|3*YUqu(Jr88JKDSMyoPd2
zB!rH4T}IScDV!+Z*dVF4<{Cq#kP04AJsmn^gW}Lqo=oXFk06wwf!mX&PqJ@C$0gj_
zHskS2_7v%AtBz)YxUx(iR6d1N(DN|N@#-+4i64EAdAazX(Tuqo@j_?k(UtSWv%70Q
z1BGYQWQ;AXDNVC@lHbr((8f_VBEa-kN0NxA!U`JJcB;8p+Xi_uN`uDuhrOOAWqBOr
zEm;u19MTQx&C9Y-5FO`%fwTjxELb%ZUg@eqGb!C2@{AMM&{9$NQT%oCD6$O`RIY%w
zEY$q7#HXBisuat-r-ndkJ+mkmNeG@_zz8xaj)=D7wl}nZ#jX!X%G#ZqP4eTrRw%oY
zzm^CHW6r!N7L9qR^x^8ip(1?s3Q=ZjJBR9Lh5%Z;Xv3Z<i|&4lJp)<|&SG{@tHMW^
zGF?!aIsgKBjnrz#*sWTSYji0|xbOpCxaFC^ZEvuOb{=$ZOOE5iKpf@rldT$CU`I*k
z6%`oA5(AAewfJmx4`UY1B!ve^8uX+KrlJd{d_B2JRMd_kgFJDV5iiKB1nsW1=uiSS
z2?lGghh_df+q9@vf{HR!JPL)U112$%Lb#uZa!OW}X>TcRHkhF83M&K`=#SfRJ(r1S
z9~kRzkTonswv%vBG3(j7!+u^&AT}?$gKjLKtV-EJuuzRv;i3ps_?!lEloOg_a`gsI
z=-ZQlLfX#$n9)2AiLr0-4-=7v9#KpguFMu9DMMmMiQsOcH2j-%w=OjPsadJ~bV>E>
z)(bPsXI9Q0dYKm_o7C%b7z2r0$L$869SSqS{QAe9ot^bgI?vk6Qy;%JaQ0a3Lhv0Y
z(&j{X)d=*SCldP;7H9+hIcHOEeO>i+dx3skdIeHt#ecXhA@v*>8*|&nLPRoLM|J!P
zf2vU@)TDSbRCZ=u=LM`d7lV<6mf{&7g!qX?iPL0xik%TxFdjj7d}=+AL(JL;j8M+u
zF{{8van2%q9u-Ng`U>0uzR+N~W5e&uZC%QP59^oEdaA9H2`m5k*)`VFg6?KpT8`y*
zJ(CBAB}N@#$vOdqc+zL=8-dhxVau8ao#@Eld?ZCoY@pME$CjHk>Lv9r6qI9nDzd*&
z1xG-;t;2vw1xySJ@FoMs!8kYkj)vjkX&0TbMtFZw@!Cf_i*JSkkNINf*qd^Z5#}~7
zRtia?)8l9hMEp4UDR^H<UiBP4taJ1S!cPyM0>UkjYb#dOY}Io<J8OjxnTYduZCF+d
zX!cY=fX)o~mos8c%rRf>wY`+Vv$pm93vDD)sC8nXYz2w-TP)VnVXv_C+#cm~R0dt$
zZSO}m(6?%cmJC(E<@}d*!kTSgX)V6jFv;M+Mj`E4aC!^ytgrtN;0>d5<Qj?v7)5^h
zj;8V>{OBlK4GQuon0SU5bh?BxX|^y-LM-E#Zw6#2JtebIy$+{yYi}$Jl>b|LN+kmL
zx&YdDlQ60<=yr%oe;E`V*0QPi0)N$uUAxO_l5!k}dR83pDH}$KObDo(CxeU-io{A9
zaGN5Rj}$N5`^&1oA|0JJzu(KUIQmC0!m5!4an4PAE-zQjg7*o9n2morTpeH4ymIQ-
z=~De!ffpt!3}}Dl;^GSnmd9@g7~KGYM3!qIOzBV~=)vF(bUWF1oeJ#;ClsYkz@G~h
z|0=%Q;)PzIkX(09M_dvc!nU9UuSiuJQalhrG8iTOE<E2eUL~EVL7(^dVPm4sDu4fL
z&FXKkQEvGmG-v0g+i>m>_THM-#U9A{Jqir7P5QV69m9DrZo^f$Vi`9+mRKiY{Z~UK
zIx@+z5$J&!J>p20&U1vksmh)zb%NOY#BUGu{ZADj^=jopE6C!s&l%>gDgd#~!<}{$
z9)Zh3W;rkf{LQAf#^S+f)^f;=gh$r@sL++alyk8nn)e?k&Y08f8kzFy<g`(ldN6mM
zHmnb)=9h6i!T@PPw>+F)y&!wDV4Ro2)C?9pG^N*j&)Ib(7}$px7J9JAHT|2KLii_j
zJBKtdhPoFPL$*q32*ZuC?|FbTXy0(VN$QUsy@JDh`qFr$pq}CL8VFe34Y~Mh&qViO
z{fXa=weXloFhnaq%<`3}aJi8M<9-|~DqjxR7poyCYW@stj#vEd7!#Str9?pL&)t9A
z0jbA67xeZ=QaO`ZWD12WX(g+#AaszeI&={*=jbG`7r_)>A4-=<JI$NMqm?V#-5#lX
z(q@7-m`V06o@(3t)+3Yz4k5kbN0~{|Dg%X6cL8hT)`}`)lm(n%bJHa~IzYw=x3oER
zI^2+#N@3mwT^(zpmy>#EQ6*~%ziw2)ADWRI#X+z~$Qj>aU%PyU$uef%Jm!7k&l?di
zm!~3~B~H*8y@!l63n$8GR$!LPGC)sIE|WVLUD|@t<_l&cru$!+L9}WIZt(?JD`P>m
zcb~nEu|w;PJEaY4GqyLTH<rDL_x*I?)<<})#97Gf`x_y}=}{qE^;N3g;$mWMZO>JP
z$salSZch0j`e#Y1A~hnPkL3HjiCc7Fa65PD-QcSr47>u}KhQE~;mH;yF}|h5^>-(x
z>n(~bGPIc9Am25H`d(M-nWZ+CmU&V$-e#|9Vk@Tcajo5QLxj-i>Xm-9Y3TBzeVXR<
zGNzV=*(}hoLc7?51(@_m1xlYaOuz8DDy2lI6~)`$d;kK`@Nc#PM2Smhy)h>sv(~qI
z9I14<z63teb^oZt{p-H&?W?D4i<18aS?y9X9z&}+Jo5$u2nV$<wb857U_?vto3pjI
zUf4cd=77yf#hJYSz{xl9^O0R(3-mM>dRc!+K*U^4%Ta0l#-~w4i7tQyae}q>N_9F#
z0)YoTmolg7DzmO|H`cMQ&zJ~UauxBl{krbAmfwila+Ud^+@9NfDUK@H5(Zt>p{;~u
zn_zVnc+RnGfqpa(z@a)z1O9l=-9@qpHTQQJ4WY-Ed)(0zGYPB0eh3Zx^mP?uJCAz^
zr5N(IF(MsxTRl^Cz8f`|W4zOiZ=lvtW@}<sz4-1{-dC$)*hW2b4wC}iYq|zL8;|GM
z^>$1(xOtg4q;I$6p=UWUys*)DYKL23?b2HZrvuM|?UNah@p~12O0j2%ia_mn5z(!>
zvI>YT!u&|*pi*T+9(*JhNdm^&V&1cw0~sw4!yP46Z(_S|^jB8Hk7fnjOjFEihqMc<
zqAv=Ph+C+37Pq~7TY`aqOj46?MD$>Nd4BP2ebuo0`OnTbOH#VD2<Yw-a0{|?T|3t9
zE7=jrku=APFhD*`;V8qQxm>Twp*SP(;L^7T-y>OpM=jX4ObA{kAlQ97oZ4F5@d?iD
zzKw4=OipULPFf-T`n_)jDOQ$?ExH(G9Y3=T?$G1Gl)uSqVJ-tSZ7#OwddCB;aZU6;
zBs)X8KKeV;#fQm-nwE)>N_hL`#;c*Trab_p<tI>^RxG`P;I7<rY9`7EELJq9J`Gj|
zPhxg3gak&6|4{AdF*fTtyDZ6j{CV%P-){ed?lCg}m;hWam)oD|wJO$x0GKCT`8iX_
zVvq7qC%-3~lKg=#pz}c-NntK-<p5Zj_+@s<*XK43<}PJ=$$iVq<FKT_tt^thkX#*D
zvsiPTC<?}A&g7|kzl{*|kTbUCu@lZBCp;-B>FI9)M?ZBqhxdhAw)%RBvI)haB=0rd
z&V{hjl*+ekNl4JG>YfE^^$Z^^%&X%E-SAlh31mF->q%fqNM}mo#7nZrI5d=&d~tKJ
zjNDHw_8|+XWHo<ppgdZt+(BH|#G&{0YmDZ@Vd%BMWOX5IqJ`X&gU%VuLQsG$3z57j
z@w%fW^cM?ZU+XP$)JM@!Cwk}_pRX7OI0te9&P&f9<MabCC1Kbn53!N$L;mZL;HU-_
zjW;>_(L^BiiZ}gu@KorQIlebT5?m@w7<XjAO6Z4EBwz{r%GSrGTK@vX?{jbcKyT)6
zXG@j%#!9<?n0!7#dOFnt=X$ZUEiD@65M9wpzy3Hj%6hIOM}~DwYHhRc25&Bl^vh@3
zf3j0(#OahGs4VQ@3CewAuMV#tHj+Sa*Dj3yEMAeWYlUD(0_?<`j%+|<f9wmzT}JFA
zsIq(xSJ-ez{|hUoHwW&2VJe>gg{jzC|HoNh=}x&94Wsm2{7cIU$l;*hdW2fFM~Z34
zY|70(;lGIE-+*$1v>u)Qi-|3F-v{)rXs__Mz?P|di4$XFn?>JR(&%|VC)L@iNm$xp
zME(Amcp9o~>94xy=9;gnvDcKzvb?3%2~U<@Sk16CZ;Upt&{_#yidAvn)E}o`H>d$F
zh)YRG{n&YKq!?&%xcand%1Yj3$fZlYNb7#@`)$af7=SiO7mBGIEJ*pRh)*F*&E3eh
zm)ZE_3uy{Wj=-Tmr4Rw+dQO|qIY(;f-o|QdJp)X3I8ucSfq|K0Pd;3-Bdmo}wh_$|
zjH(12ttBUm(T@mU_RkQPm!;$ynP5Pk{ZfgS{6e{X1jg+F%*dK`^bdg<SS^=30}+$T
z^90ClhREicb3G;ZjYf?gK9@REeUB0iOg$1U$TTI2%3ZoR*s#93bV-OWJGkAl%eyNr
z`vsqZIr*XVL70CnxA!gLD5v`5b)&_ltGkZ|Go8gnM6Xm|MdJ(-4Q69zDfmF;f~wr3
zT=D#b<Z$Oqi<rb!>^9v<?IYS<N(?nW)6Ddxpwp-RlE3B*AwvMn5*3g`;}szaUgM0j
z;RWu7@dbWX|7-{Nf(e)2yB@3(72<B&@jEQ>B6Hy)v*Q&nZS9G{b^3o4Il6Gol+(5f
zA4lD-`f#RLCupS8lf5e!6HUNIoknsxW7B-7{uyh@7t2qBt~_d(DqX&F5*f_2-pa*=
zuwWf9^KC|9Zen6Q-O~Bt-45X*(&;9PLH}rOZX!NWHf*l6tGnLaYj+8vT>{2+QyyHL
zcnyN+%@o%)%&rF!2k!;W&~TlKGC;+3yd>hTQZIz<#1|OvuUl9`n<ik2byg&DTz*jk
z%JaX}#m6e+Q|^^gHxvi7)uDlY{4fo+bn)Khq9{SV_4M$)H)*=d0tj=_*j3)88(lME
zY#1;wd>hPJzQ@<nM1Sa~`G~$CPQ|I(f)$IcMyQ)kwJ0WEy)Gfd^)4tR1sj1R2?f)h
z6uSZ3f9X{hd-YZ=mkdBqgCi!4;1~!ChD%EmBI|K+S_g!1$YSh1;*?0tJbA=CqH^w}
z!SE~zG&}lVk`i-=e_G<nZDJS<gw4wR+7^+Md>ZArA<4HXjq<iMT}ArX!k>{7z1uGZ
z3E`w6PLGwJ`NX@To7q8N{G&Z!sj%TIUE2sOPM#v(gWM5!bOV9Ry<FVeksW+-g7R<5
z1dsp-DZpM^vJs@w-qUdutWRyS<{;_B00kgr|7bL2vUeL5ty|4gaQu08jsiy6D@EKz
zs}Uls)V6~nb8k(9)D(t<EsX%AI>$m52Imh48b-b|DLU0pO?&k~)WC#JIrkVg8S!Sb
zm$nl+tL7O0aVXH%pQS_$;Mh|c@`yrT%%69^%@<VbZnj~{PJ5q$*>$|MDWXNsi$M+<
z=9v8f0n{_{${y4KoEZP`FyYZccU9rMbcN)3Dl+Y^n&XC0PRcd3sRYmmo7oxMa4Yzr
z$QQ}$TVTjLhGd`3vj%sF2j5^YFd4kbzpO5kzh@qEmjK)I=yQ2@o|f|8U8OC&XZXEj
zV(JeuUythTGM|6B@`5e|W<yVuftT`3@OM~5k8yydc3qU5VIy#zAZ%L1@#Eb_mwC9w
z7&4S?4eMld6SJS=?4E#X!?IZXo-qq3%&-9E#B#zk0yX_7W22MR7e(1@wIw#~a(Dc`
ze=r0Bn+P0+y?0E8V?On1;?e(nb*^9U7yD%dOWo>T1T_gAm-}@d8VOXE!49DoQiEKs
z(e`^1g@m*1izE$wABR*I!2l`GScHQy#=b3LZwOL`;Dk46PRo$n6+miYF^4JJ?cy-Y
zS7p{|s?7=dG`+?7mw3dla|roUr9>elqLMF`at|a{SU9s0b*6NM<VmYLf4B1|a&V{g
zP9bR;je~wp4p4*V#-)udrymo|Q$(?jtDcls086}p!RXL&6-@VS_zPZZ3+=6wmF9z-
zsLz;5OrZ}IG*j9gD|f_xsB@0qZKIiOalD3{B!Cgn3+7_<!-UnCKxIvGIK2)}UkXP(
zX9NVz|BJccP<DH;xkN*W8RCCCFrA&V>20q7s@vCX)W|b`Te3?Hb|)eJVz-2Zk*d@a
z2>=CG#<p&BCYTWK0fBxF=B{OOe>gFU^Fzg6RRRHja>8Ck!=5})6ih~v!aobZ6~}iK
z+!Y?C_D4w`JdYK}%b^SoS`O*6D7mFy(gqsHrd?P4tXwl&&;5j2Dnj-??eM;2nj6|Q
zVDY%uU2cKjc2uVSWU}N?g7?M3VF%gzRaF7GAeWHwfr5de;KtH|Cr|#=a^U}daQfs1
zB`uW(=!VeeJ{^4!9zK~sZ0+uc(!OAv!+r)P2W%-1c4zO(15<;SFcE)Ks^<zB<N!Y+
zlqhyBlJiVXrUtgU9Vk-gl_@)rE8tTFnGa`B(UUc_;V*(NtS`>3J?2}l!8$cr&Tij7
z#$6S5Q3!suN~CNreKIn1sTUUWNJB6yajX}I&}>wcap6ygnlb}s;3@Qn5a0keNmmAX
z&)173coin8rHbZB7Q`z_=8r3m!@x<b;Pf3_{k!=!7%kx~%nhe4ZVoht9=4?D>zEj8
z*4(y67cQPQ*O$y=hZZKHikygDwC(&XPMLODhNLCU4<~&p@CiNJ)V2#B?F6FZxoo|v
zHKFCHeA;LA&pmMIyk7)GM`TSGMFdErW8N>MPr<{UU+GWGEE7E)utA*s*TD93f$YB)
zres&Yz#%L2W%x43Ve5G=A3)>Y*o^{0yB~|&En$%Mpm;+zDqC2o?W1BOSX)I`O=$1R
zpn)HVjb1HAe{j8V#bCBm3_#y;-*vu~QN^8~gV=+C*4nN&n<!)i22P>K=9XxJZM62d
z?iMD8<Oji-d|vDQrna!yD8M#!>>PC0I^bniK8_$B`?nb)Az8tS%P&47>|7{pNM7Dc
zht&mVh+@iT3pNd9!7>teo7RE}vAaIIsOwwh<|p+bZZ~h+bTB>2rTld)zxj~w9nBUN
z1JK7I^f1rIn$!1+K?n_01Fcp1D>n_j#iitt7q``!OLo>|T1AeF1K^c;NntZ>OK8CQ
z6y^%WcF`@JoD7}Z9vwWk^+aYp!v#VP@f@@VjYDS*!4$i~)h<()s8N_XB}5@ksP5x3
zF$5GH_Um;h*Vb7+hd0=0?$Rlg73yr6RwFEiKyYZ*fS3nh7sV#-3p^@NY=<qN!b{J%
zWt{~o7CORgLw-J`1$g=_43qPRMd*(bnsgI7n_~ZDfq>3Hr2)jv_O(@<Z|&pt;M2U+
zXF-Z^-%tA_xek<OWb#F{oXbCGDym)n{(-LZzy=OgG0Y_eC9D?n8?5}F`-Aum4u^$1
z6~~2}5HU@*jrhqtz{G2jv$O43)=v;I>)Z~WVFGUt7oEjT8;}PjsR?#`{p+}u>D|dq
zJsfop?V{qj;ks?Q3*23t2+ajf4UqXh^Ti_<w=AqW!uzI-6x^cR**v?D204HZ|1whV
zUGb6+QXYo}xsY3FZ#!FDlcJO!DrSL@o(oImrdf%Cl;MUTzf*|7Ir5bP5Lutb|3!%X
zHyGeyUpoel0CeV3Jas%6DuEK7HT_BMm$4MX#rQLUoi;RB3`wA0vE%MTIjJ^cDRzru
zf$3WWX)c|zl*-iwcectwvnfS;=~NG$X;{PKw3%e<3h{+7Yq5as!*M+yJX1b1nKje7
zFEda^!M2kUIA_xM{DdmZaxWb`73L$2balExT%hQK2i(qAiytQbc7|2)SUG?}dFSpF
z5g`3-WUzAMdV}yBz_k@)6u2@HLo;23_8-m4%pVcAws5#_ZDtBV(+3+ML%B9g`bu8G
zO|jlxWT9=~%b#)w))CJ4Qr}Jc{f){^r|TlUEROg5sVQaHw7X6chwZ|;wA>=<KiXzq
zIY2&&3v6MRmz3j^G8a+VU_cX&7vVvuS-{s`+e<~y0xL~>I;4!qrSW3DyzB)kmfsgm
znX?%4&kiAyIZqTag6UqLtl!Pfz5_U-RM;!dEDkb*T_WvvLe5~rNIMejuPYfOu=s+8
zJIHWH6UKw4jPXj(kBZz1`YVYcrFUxXbCGPdfpXIXvAP+hMZ35fk4FqeSo}mMPg&;u
zf2D*zV#TL;I5`#R1FJ4zSbAfB{6uohB~<0Ig)}<~@}<yOAgFk-@tPC(Rt<1w8#i^2
ze;K}C8n#EF91dr}|2(yEQ?ZkOlRkv(cZy6eRZcP-87d-h%fsz+gAqzeBE{>ioY%Y#
z0qd<^G}fGC?*(`XZ!R}K9^chxxG(WFrk^}o_rGF<#CXaRf@DpN=ZWD?B#*^$Z9eC%
z=%ak5|8D$Rugk)f$)T=V)YdMZQy2(#fBUOP3$=Qxvx;&$gNpgKBJD*@!x*!}$R95P
z&F?YMuOmp$7i?@Do*O!4sJp}dan->61T>VNXDC?`Qxf^*ko6vx<ohdSx4Sk)WY5v$
zYjt|DY9L{M?OXbdaFJz*foT2J-gFm?7PLL>(AZ(6{e~gNdwTqi$CrWAH`NtJOYr*4
zC4s7s`$^@tfM*jUsB_PI`)&dS4g(5XBc<ae3;%_d*w@+jj%ebxmGXZ$jqLxCU+n)2
zsIhXh{Ez%f*3q#);70YkG2D}a3d6ZWE%w|nkSB`7F`6aFt6SZM8ZyDR+eipa=-~Uf
zXo=8R+~dwk4;?hSqg{T(NUnCPw0*i(814b~0G(enk1JJ^+QzkBV(DmDA;i{XYl@cp
zEskhg6shgBUP708R%W=yt2@+M*Q^i_E`9CG8?Jx=&e>`K*B-6`w5E_!o4jSyQ$((x
zwx5$NJ&Kdi`|xq|eheGW{ZhM!CLGS5SK7wLXO57jW>L&HC(j3Isg~`PZcHw`#Ez#P
z04zj{Q8!(;a1)V_Rb~EL$jbePB|pFname_PmP@lN;Q5;{gJfH?P(OkkE6|BXo6+kF
z#02Qi@-Fl!Ru;ovOQCxu!B355Zai;Xf~*)ozN=jQO8aw)I6^xJ4u?$3<@fc-#*qeb
ze)%aJO@067k+;q1;@k7)W_q`+fY5TSODNJTyKKsG?&xN=@_rN_{*_*)|GUbGBG_&4
zvqmhWNC94it~%wyVTjtyI+*d&gble|lNzY{IYa7ejms93<VZSkXp)%z$~COb4l8#L
zzCB8vr*icXBUAVq{*n+3(<QUReUo4|!$1>3Mb}qwiNj@vCv(m#Ll9K~A>;Ugy&u1W
zTYF|^4*TRN60?h8zPEe$@gBQH1C@Wqs(^|0yfMm(MbXGL9#Jr9eQeK_7eNbuY6!G2
z0FG6Co>?UFb$zcc@+lq0`eFI(NXZQde(AQ8qHb>|z45AFX>CSZ$CkSBMafAswnDpA
zAddJhTo2N<&sg3=R=BFJZpO6=?Vv+WD%Pw8re5zI6T%aMJ{&88`d7h?Q6;EGxTxU^
zz3zo$c*zA4V!w30l|$8p`;>*uegcV|79_Lxg^0YKrHm-#ud$}Aa0=!Usu&=&(F_o{
zgs$lHlTT{1CCCIl&zA|OxFOGL>{rkHe?Rdnb|X}WNR58yIl_PIEo6HX{Ok=hpLtdt
zEaXT7Sp0^63|?i)Xzy~OSrqAvbpO%6S^zj$kU|ol1Z7nyb_SHgHv~0C{DI<S%gB+|
zMc(B79~DHnN-1%v6Tesp%bl1FOjD(XrE#(~^rx8??hDZ8(~s`;_6Gt9KK&x5x6vT}
zf!W$TLUb$X&4d~Svv5ZMS)g%l$<PfzK6qy{{^&x@Ov_D6?85tvb6u-;m#^FWY;sNh
z9_mq<L009SK7C?ocig0@Qv~8S%*~W2hStkm<!)Th{@tFJWi#+{OT6Sq*M;Z_q2Ps_
z^-0g5Tt;`UwwMG^@d#ykXigksQ!#(o7;qV~R{P%0?RfAIx!(&YfOSL+=RLsjjG1Y2
zAEnp=$C7-9{tmmMipB|cF#hpN*&0**EmMK{>;UmcpyCne7@u`QehL)v;lB#a;ibuz
zo)yK)*0j91HZIFZK@BqtAE#FNtzp+Z#%|aA&7O6B(g;JOWBSN9JbSvaZW^+Fue|ab
z#i%`sMo#7Q0xXE0Y>AYNA<I8LfhLO@Ka!&ETt9+jnq^_u!>YpSH!_x(T9Y1SVBw<Q
zLcZPYe2!l!e35}7C={@)*1r#dniRD!2J_b!`&|SS9#oqSjS22Ar0XD*>xfNQ2+%5h
zMxU+Q;Rje`RmRi_;p(>fIA;{N!nvfir0KluK{IXng;9A`<S+RYG0Dtm!8bYwL~-iz
zNd}!aXAgiINg)WYxg}T@a-T%2xf6mO%WEdZ>DUiqLzoffUtu7XdwWnAYl!GG1Ul{1
zIvcuE?E#ZU>~PrOe|j9)WEj|FtUP4+={#i~(R`->(Y(M3D`;}6NOJ;Hbw+aei1}7v
zgbpYv_lTMm1+2rEiGOQ1kkTYr2m)W;aYw6@dc9M&JLE$l4|I%-_teAh;zQ3ybv8s&
zPJ_ag+d(8U2sjc${?z7FjDkv^g+Ky><q9`@#<tFl^=A~&Z(;-bVUnBzt9v9VJzavJ
zFd4w&eW=F_S)k%K(1EYQwUqOKe2obxpKFw%C)fC!q;X@<6bw<-s!$R%elIbkfE_%J
z$``_+CtqW!7Hm-(2l0lctfDlSG|%{9o)WpacC-e}2R=hORKs@Z$LzyGb0ekU+e}kn
zUO&YAvF)Dx(}vg|uB=I^4dTA0>HR1LQlSuwa<)j7HyN)(HCQp14kj|@bA2PAf}MIe
zm$G-#(~1Z3>WM7vz5eu48p2F%pIX6j3@P8Qyxa(v<qRlU#}G}XCluyf`V8-V5sQ-G
zDd;r`0fJ4v!|a*w?lY0J+&C`#<vJ`=BuRbFRE(l8J?PvI+)3OAda4(8dZbFffDl(B
z&>egArg8@UVn~(y_5l?#v1g{g7(e1{2t$%jz359!SU-Nzz6A+i(y#X!Z}X8KzNbm_
zk!4QeMhWo}eIx-RKm9;%cVKC-`m>KFacE6{4Sf7%hJs~OEu;O+bcD>$1G&UOg{qcx
zfy5Al<o`=Mx7meU9w7(6NPZcD9Tk+F7+cIIw+Q0~{_#y6HKz4D%cKeeQy5E})wuQY
zGb;XF!#kb_#IUBTl#80Ng5~ei;6mCya@;j0?I2hjp@8b$B;a79$X9y_DF?$3ek-0b
z2$<HqKY;P2$zEvR>%q^)Cg$00wBv=)ayBV$N)Au>89WFPQO2W#n05KApXhsp&l|<|
zvM4Gr7<EK#`j!M<7%LUB8^A))A;IexA4;=_o;Ga^V_9J^E?-L_IQAR5eOJ38s9wLW
zq5#{ZGhW9FFHWCsv|KYzo9<Ny=c9W-7Z?;Ijf9Kc!;B}JYKE7(*Fb+2L&E+D9$z@R
z+j8ul=Bu^DpMXpold!da#(<O<{&5F*mUpfg70+on{;eIBxy?<M#+yT3hBX+MBj6i*
z<LXNvOKvk+l1f~2u<5l@SJ&<s+CWEAyv70E#vmWatj1v*d>4>P8Z%yZZcI^31>Dw2
zqCydRg!VN0!qd-&X@Kp}u<`SGIlgA--5cvek(*Gr5<nQJ`#HNtR28~>KWM!PK-yot
z@v6J%-ny%K#M#5gJd~AYyLS1|9V@YA%rZCdu|l(`D6CeF)i2#ub-hO|P?Zqso6G4k
zibS5WO|-x4t+bbF95Iq@GxI)R0kXFcp|V%K4_%jTho*x1Z`UTV6pUKfYsdcOU)^GU
zM|zxiMM}8Q7h$K+t&Kh{rcgz|D))RHM55;xcNmo5M!nEBRQVHv`4$&Z0KqmpH(E#}
z$&EG>rT8S<sOB)%?_e!}Q<=zS>PmYCF=`bD{irW#<4Rk>aa#~0{it{p4%8O76o|Lh
z9tIgtqRL59R339C@MH{<U9`_1x#)32OBRVS=cs$P#8-HJc+HB5sc=3U1DA##N38>%
zD}H!+X44_-?UfcfjSZE?YM*ZXL>LPh{xH~b<$|yjvKakrb$f#t8~zJ+rZIDKYkU0b
zA_J{MBAdN9hJhqic-FN-3ACOe46XdLa}B@W(6F`hcKtoDadBA*=Ah)(jQr&uOJq^C
zq?~<E$EJeD))x!kGUzDe6105!_P(!oOU)4n`FHE6PVC_osy<{k8N;S6f?>aZr2n<N
z$S9d!7CKvyFC(2uxKt+g!QU}zZw3s!C&+o|c7tt^&F&pLXot!zAIO^}^1>pl<nM)Y
z%LXbz2BHMx*pp|pWuJM_z%A@Zks;>N7b#rJAA2-!6ZZDmdcR%zx$@Y6JHfJB;O5ol
zYPvO?SW?+`q{T8qO26`HjPrK!wqZ9hhK{;4%XHq$Vmc`(+#_8U_b-V4jig*?3M07+
zz&MLxX1Vuh10GhYOd4ZmiKXV(4SWDLy_^vK|C)n?Cy@jm4UC8F{~p`|qqXGh$+%Fv
zKeX6up;|yPtrNcZ51B<+#~>5snK79=Mt_0p7LX5rDQ8sIB^&l=R27iR5n3uuN!4vX
zX}G3TCDOm$(SD4-bi6Hb*-y|~i>p<zrvF1NZz>z-rHPcKHXwg6zxO_GzhYKw*bv3A
z?zw38*2SHX+E^oc$o1a?RwotEyw2I+^R~NZ>*#R}PQ|BQNd8G-5)dptpKP=XEq8q}
z=@?mZKHG5fGVp1!6H*$O6>J@gM8r8gDrU6Lsao0j0WS@tOufM}@TqLHDk^IEDVT5%
zG>%^wzVDAIbo`P7dy_@oYbou1Pr_Jx+6sMR%1yOk%>@z+OEvre>4y$AR~)DeO-6V(
z?LtO=454l-wA-Eljnwyls^V>&)bPS;P9_;|3A<P0HH*lrkBU={;*&4<slz)9-~a{A
zQ_mY0Pgc(>|H|(dGrTt-4Za^}_r9fHceG#%1>=u1Tk83%G2>}GQYcvCJsD`r6FRHJ
z6Tu@9wLiT%I;PBlIoFmIvY`Zk2y~BY3s$?-#x3m5PgXTon=>;$#Fy$T{w|5oeRm$>
zEz=!Q#x_5NP?Drqi?d|_D?{$EaFBZ(_3p6#fGKJ~bHMR^9`B0JZ}b9dY3EszvLoRI
ziqTnKxzaq4d+gI*L7XzG3+tAPRP1}<a2xBTUwbZ%xa$MJr`iH0*fxk9;QEj_mFnM;
zHu;?w;?yDDV<p!xIt!llCOmF$3AL4z<HNR5b}5rNM8VrX^E6h5rS5sStM}%bN#$oN
zMfwDv(PnAQ9z<c!E|%{U573bPV<46B(|&J7m6K>4<)s%~uO26*OS#Q<_F>ai(HDlv
z8d&9d)Y+N9qU2GV`qQhb<?<#5r`cWnKh_e*)^31oMt8XGwb0qCfOeiiWVcuWp$MEn
zJot<-9jEvmS~S%<t)0f(TCT*8fO2C!WrJcX3yl)QKz+@T%@y<TwLrW{-_Ac<;3opt
z0mrvn!P3dt$)~@$kJJN9?MsOZWwp$Nvg*E|7%7&4a;qPEMv=CPpXn3iGd3i0)30xo
z<hZ9TMi9oHME48xAsoW6)09K|>aOSdtI^APY>cz(2-(<w?R6344@idc0~@_5xv{`3
z2(=ra_BwF$rhm&LZY7D1WmWlW2M8h)Tu=!|Dr<UrNy#LDr^cI1QAEN2i!-3SH`Awb
zeA^oVRvx|xzZxxHEJqB@vM?wh`*=^!z=_IHD9xgJo06P|p!~F?YwlvQ*V8Px{?*pj
zJo|8W=W==NV^~Q^QTKBAnD!8w#Q_0XJwf@^|2?OGM;c(d2rdvc+!L|+H%gQ)S@;Q-
zp{;%U1{+b(oc9J)y9Y9(y)2lKqmVK^u-8%ncws%JTMlkxCT+>fr!#Bs0u=lD1xLyX
zc4W=TA?1LyiYtrp`#v#^Ss^0Y1{(>!f-F$`!d6@}3okyEG`2SYo&}r4QvID}7k*sT
zNEWX#u6#{(e?MUwJ#B=TT#XqpaC>kT#3n8>^<vescf}IIB|fUCnlIQZH_1nQz*%bs
z#IXgyPX;USVtMxYueEZ3%9IINiD%ih8qB%l$@*Hoe9gA)+R~3RaP?DRoAAhcY_bW`
z5BGiO_4=eq>9z1v+(D7|$`8hTnat5r+=mDVtwCh8ooyL*SX$+jro78`u%Yen1MpSQ
zM5L-+WDrP@Jo!8=>g%M;N|yvY=au$>vqCd}e^1Dx^L#Kp*1=FD^+tQK)!6lPiYSIg
z4VmIPB8$eGMQTG^Ee^hf2u98+0Q_SRqk6j|F5wsl*`nb6US7EmS+D<Y)ksW$F=e4?
z&(6OMO@CM^IsY!ND3?&ty0meU33Gt6+3Z+W2vwZX0X58fkRMBrNE1EMH7Mc%wt#z$
zL_Q1(+{et7k-V6V14L(k&kLAU3CQ76jLTD1FiMnT=meM>8Q_a42Kde|7{d;lkUDy1
zfghv^kKg$el?(0t74|>gVsXu5^6ZBS@R+5SpNeqMNP*b(p|HUo1)Bz1VzN=Mgg#;G
zVzFO?34=7UEdUy4+Q)dL=}0Ppuq94^SKCPd<j<=><Y~lsNP5OfQWYLzPnE1m8MDNq
zF_j_pJ??StmAx`P-iqTh2bn$9xjI&gfVp%(<1d`R>5QDci+==<K7t|2@P$3pO=bTo
zAYQJrks1U7IKhf>kO~2pF0Am{)V=EmjbC0rkboPbVM+H;gh>rY8>&vAO&*d;r2q<K
zfr(*zY%ypOk)vR+_lBU-v{h8l?}P@P9@udlC!Cp7A_5N_$m?$%awL?l;94zQ!8quM
zpP@zb#n`_y0Th!$!Tj>hed<*urQs#qw2OQG(<Bu?yA8bA2i$z?%$usF)eE`s@aoSY
zoJ5d>$mLf9aPBu;0QN4RX+$o=XHzY$5RB9AMla^ppVVysZY(`*Ka=<n**I6m2-g3+
zD<*z}{DefI^BWJAgMM34p5SQEoLXF3L$@0z*g1T8a1IvxC=d+U=JTevlpwixQdWBq
zsf8!rkRq4`YY)wxLJFx|N^PvC!7r+U9X_V$EzuSzU9z@vGXK4j=`uwP8ftKvRgKgY
znO@YTo)nl&zsH$E`p;JusAu54?*AX)T>s1MH!C~mf5&;3&Q{_<GXk)ur{6xF%2Eea
z#&=V%eMR9P)0i3O!Wp~otZZX(4n0p%<Mj6n1hjIqt-aG%T4pwjOdP2Q$S=QM$;vNU
zUEd$Z{bwbmo90-V#)yMjWm+A9E6qA&<CNn{b#r6V#q;vtUu3U~|0M9EEVS+Os!MHb
zLkZv0VTXbKx-BNWRtuWuQblC{eo|{BJHAgA3|nFy+iCkr6;$2Tu_<~X|D&Wd4viU&
zpFv8$CGZgN%J^5S4~$ORjkC@A#!#yVDdBPyD&56d+HGPl5US?bwVt2&+3*qD|H)b4
zQ<hH!Lb^=_q9JVB<q4#_J!ZZ#SY7l*iSHml-uwbi8XAzu40MK_!M0a9yg&+~1pERM
zLzfOCnu%ByjM!IYWu-6Dx7#OG_mt#)rHE*#wBJrD86RL*t7IP+MUnYGWSv8hE=;(j
z+qO^Jwr$(CZR<O2+qSLKwr$(CJ%7ZV8?%^QEoxa&mGM^Qlg7uNFJ{w)Qc~Vl$ZMfD
zpxfG;6U`DBNVI?4S((@~x&650^PvU@v))PM5oUAv=hBA&`U#MknrHVO*?saDAOO=^
zwE|%x*n<v<d%bjY$@~1#&?8Nq!mfitqk#apOcV)*w<0U=#*QC7Cx?0(!7VRMw1ZxZ
z)`<m0O7@*N`Rv&yu8H7bLNP`6PhhxeE~Ibt;r+ER;Kn@grpmv8S1O2L9pF_iK1-Rk
zYw!{juw4h@ClLU)2W47D&Ne>-DZqh;r-B(J$-fN)VxpWIQ`}~f96k7Na1*Dw)@t#d
zO~J_kEn?i7+>3s_-ojG*&+bDeVZZ?fo^;u+27h7m#sWKRf~uLqk~h&}=f9#vMekps
z2Mign!TRfQ<r6Q@G#{0Z_4y0?S3qkTCGCTadB2GyS@^nw-?AJVV=@|-6Cf4sjz4T%
zb*pae%qx5rakViOUTIrRknb2+jIL;}pNn6YrKcUgE>?w6(1aSqu4C1$HgMNN`G=bs
zv58ZQ>P}n#$>7f0UheMTQ-NnCy7R%%4%YsdZ(jUy;sYm~F^gtyqUTC~^k3m09#>Ru
z*2%n%QP3*>*P|yNVvlR!W`Mkk-EG>+#$5ntUfH#$i5tpINg=6f0`bc1LpL+spKR>u
zPJ=FePjugCj!~_7fN8K6>8}Mzayt~drGr(lFj`a;+TQc^QbK$m{~nW7E@1$=JnUzL
zuR`$rz2MWnGu~7(E3V8O<wiM5zrqI)6R5loxyFb={8m&xP2M;)6+mBC4+NdCiBpxj
zQ(Z8Tr)Qsw@X|AVEbz7?)*Kfy^@Cp3D9iyk?~Mv7L1nbb#g=N0Yw-oIT)70;yK$hK
zDPO%Ls%SN9QB3jrrS>(k5^6Erjp+%t325;L{=4=YI{^+Cl5v3O%?e&=xVt@YUq8W^
zR09V0Qk6S9m<8TL5I}18zU58=jw?YzsYoabb|A(JnTY4aHuNVNY?GvrQ-%<X9B=Gz
zDrn$6hN7-ZwqO<3MSTr7qGmmV)a_d$TC_9!TQV&RtUtKnLhN6(SuA3{F#WSZ5a)7^
zs*4r%B(obMxUJn~m+Vj~tb1nZZ9%+0`R6kWLPjUdo(d{gf`IUZag!p1+R+~yiA%-Y
z`Y22)e=c<cl8y*sP{<N<9OaqNL|!-c<Rh*6JhL!WxCcqQb0RCy?6~Mvr=8(WUY;Us
z(iwedNIQW#$}Z48(!SiSAhuK}*Z-m^7*G-*H7^2@?!<%YzD{kO{>D+a1VA~>z49M}
z-bn*Z)*eTdoPa||j41qw@}_C&`ucahHl^Wye0nRZ1|LsOGuAft5}FqoNor^*_(e$@
zB|?Wjk%UsC6TIplIULWTk8j*%Usfc{K6nV8Z6aQ@@x-n3g{if9F{|JwBuKBpe=B4<
z_x5{!o1o+hN4~&cIS%g_gVLex9wkS9r0MW1)X2v0i-4Z{LSY!36%LE59MJy;d!r3F
z-ayp)l$MiQCLg{SFx+^3Y5<Y4a{)WM!KyHh_Y`Q-9@@RxSWsEckUkP^nQj=g>pQlm
z$%G)&!QfQVLOHFNQ&mKA?wJnmpZI)be<SYZ*OGHi>G6c_c}qVDY}r?lG$drxZrWQJ
zvjzvS8bGW8iE{)e1iwA8hVU%DPRxC&&3p^HC{+a)_c+l8hE(K-yXHu{o<o=5ZRmHP
zwP!m{#%$N&m<9WjZOCLG!YK3-C?R2&1-Mi~El@(|(5}<3#}R&9$~>2!3n-$eSu&V=
zrDJJ9Mk2Oh;x7BQKRNSyz`h=KbMV@21Z}4R5g;GuA~63YJg`2l2R*LyjzjRd)gOSE
zUclYg4U({2Kv!0&rsKX%6CxQGVbLlV<bBrrhsz!(?*RLQau7uncdI}8kF7>*l;uq%
z%0~$vxeQp?!8S~zyopXV{j-BUaHDrTM-cM}$|=$yu6KMKR-f5{^~})hq?ICat}Czm
z3cyJBm|8Wo;;Xn|5ua5~XJ(Rj%ylnNDD+jr;T|!=FIg`rCInG~5T@Go4+EK(e1?Mz
zt!Z>}i%WNjX4ljF{*sc5^od@ZM-ROV&p-5AUu*5!gl(A`6v?wgQQCOE>i{;Pnm?2|
zThRfpCx3WI2=jh~_gv7SG{SW;_&6fqQ~()u?7|^V*-5Wnx|m~M{eM(TIWdGF2?1=_
zYc+*V@PA*yC-ZKsh=%cUsHm}p--?KJPMjE*#rsX?B75_hsSBtCvl=8+<fF8L_w37J
zoDye@(_E2T6HeL9Q$WqM`lpak7eu872?Y2Tu(qQnr&TixFe9Z-7)6MHR~q+0A^<S_
zf*)^#ph^%fCXu?>F)n@G+50}M{L0Js*=(t3aU_j6&=&V?M_`S7++-p^7QzD2nO#6+
zGiNt<gYWq1CjO%-WYo$BE_{wias8*yKGn8?ga?MRe`o_nxOUs<A4f8H2Xz)e`AQP_
z;=5Qi(BorRf>QI?K5G-G=hnz9$^n64*MqIQ4~M>@Ylzw`gaMqAg_IN@eH1@uru&UK
z*>)oN?Hz$!1BmVj92%0SKyHul^pm~e6WF(9!KaHR%)JeVZ9TPied-d9WE^JjjRLQ-
z0=^*!9<*XuWmjDO{}323`$fjhWXW=jq-t3VH<2{OQ7<E$BWhvkcP*ZN;Q(_nvMzz3
ziIX^MsU(lkfCAl&?+FipaP8uiygX4Oad4OMxHC(v#HUdwUyl6#5`vL{Z4L&96iBBT
zR7y4|3Glf;VtVPiww*rE;F=wey&InzeZa;*rytez3nD4j=PPwGj2g&SIHxQz0v<&;
z#Mu-^AF?=K&wA{KK&yZvq*Pbn`WAR3Q#sqc*lZ8)8;ad*sXJ|rnP$M5X~ZAkoe0J-
zN1%SjDHr!8naQ`u#ZCYyv*^s}|1a`5lYj-$K-rl7ulqDbQzq`{A9Bxa-C1@a#J?8e
zic=<WUU?8uge#`KzO10q;TsHfhVZ4>&)v>QHB}kxYjc-`H*>z;N@LeH{K7%R+rsJJ
z&QlM9TvaE^0pgVN{dZ?1Cwy~Z`~_0j5^03l<^9(^6Vk>*K;AFg?dKio!f+`iEets7
zyb!KvVi>?Gl^4Ns|F1z<DFY4f!C(^F<mFebaEMZ3oyW7DOBtU*ms}*=7~RU}l^lVB
zv{`?795GfbMF!-zrPY;5@j>Y6C-U#PiHM3vo@aSk!Y49^Lf2i`#*c~jJ5VpdWZLte
zQv1tN+ywn%#@$E~dEpEeKhKk#^8to0`ZtEZ#`pks09URt=gWCI#SMb)@!60W1SPi(
zHg6|MVB~+7)F@>HoG(}nBt6VtAa^T@PhJAlj*8U<8{c*<4K@L`9=1M9G+*tz?x5IM
z5meC`e3EpCbR!0NP+&yh>c&nznR;WixfX1yTkzw$$xlo*Noos_R*=5bR;|Iaot6iY
zBrO1={}G;?O)j33rOZ#Ec@2e5tem;<-3XVNvDM$2+In@nJ%buh3pY_(0C55TE3;sx
zxO@T^MPYK$!_rl9*d+bI*3xjvvIcW3Qe@80>#;G2V9oHlVUl~4T2c;W((mm!G2P`{
zpu{C4*1xNC<tV$VV~v!S$u0NIoO+1I69PzrWnAKn%>VVSnw9L2={O$ClI}WxWReO*
zX?|RIGwS)v84PcMl;U*YifcQj5J?-YQuwpd>P=bq!RK5TSOJO+*Meyrv1;tJXm_p|
zz%>}1XLEEYyJ$iBtnRKc1SJAHj)cCj9*WbaQ*Z>Gb>WEOl|DMm5p=X-kO=o9j07nA
zz1nkb5XL>T>lS1@2<H_9UfF@o|Lu$2c**AiLP*X`J{q;{CN|J%8p{zIXiw{@U^Vzu
zyu*~Gm0Dz!TN|433tMR33zkm=QtX(wh~g~lTM)M^9l>tyqy>gt7GG?1S)Y%@TtITO
z#wT?N7%3spvF0-9|22htVw{8>$pv6e#kSdKO5vL49{picI>Zdb5Zi$!FUo2EPpb4O
zXW^=YA1*aeHM)!fgQ#%Gb~#dxHZEfX!;oXl6d0wF^_iufgo%<FWm6GfRonpSnO_IQ
z>su6s8idt?{s^kTVp6VoGZC`FZ@>uuXlj$<lI4}}mG!UKtxt^}1S+_BlmG%Q*Azf6
z5m~LV@|{mwp^l_#hA+<3sAa(*IZS|sxr?LF@JIUrvF*mx=vBBm*P+R7JAP7uk;%V`
zL+uWmxnUTV_|Ejers4IHDCfLaKut!pTsH3|y=x?30UBROq5;qIVsmHBwXCiE_H>fn
z%?Ne7rMNPXhNXva%yE?FyMS(_;>Kqs!c3y^KkTtzpY{%8vRBbtdJ$-bn4mA0U1)j`
zT&sF58&n`-LU^MSwHxivfy&G9ZlOv#QB5AVqr{sJ3FJnhSfReU(tSxAz%cPM2{=>1
z5o9VttwN+~-b+WXtenXRDgps+nXkSeF2UE}5}w7~@8yh13QS|$M}XBcy&zB>gr-0|
z2*WC;ig6m9#>`_`{<FVRN@7iUm(C2R;1sIRJ4ewL<zak0l!TM2{&H68o{0N|sYbYF
zb3mJ8q%|m)l8srMLZy%xBLP5l?&q<fs0;{aJaw5H0caPbrZdlto2eMgcCZau3Rt?8
z=OxJC^b8dO%P(L=Mt~!sDxSr-{Q2OcF_~Ms>z$AV%@Lzy=Xm@+G?~;JMIW~|KMGBa
zDPkqkL<&f#LuIXt@2f*OuN58`YM{;{hk1poRnAX8pZH`$(fb3!78exK5G&em;sN_P
zhND(+=0m|SBXJivuY#?1FDR#28m9gX=M18fkR?dVZ6>h{5kL)N*19rZJW$V#BCHdg
zdXp^@mb*M1;dBlc$e@>I|4hIhoL84Q!MsWBLz}E>dvFN9E1rv#HF(t5dD`w+5k-*%
z%b-vXEaV^4rBaJIN(OZn5T@Xnwk9CCtDloXPY9*#)(wCCuzhU4QyMmkqKFA%DC}&^
zuIP_v02t665Fo|wpHQ((67L&=N8W*?VaQyONe}jwgejDLpucz;T}k7b&Io~a9MdXY
zT_v^zlXa&b*3z9WS~RRRXJq^dw&ygyF*PB#xTvu4Gl3Z_)jmY3DP?0!o_)dPs|rXl
z7qU}z?mRT^?tc961?<+l+|C&^@GVW(5<WyyS$|v$EP#BGV9Cw=F}LkN5UdC)Vnq?>
z0`9J=VP%Dc5UxWP{AbS+W3l$btWh_(-%<teUv|(N6VQSgeyO+u9iu_%b`CsXW;dLQ
z3S6XLob82(x?1dYcR4AlH7EKFBL$h6PKd30<j_7FojvM1Y?~6USFO3*3mYcWC?r+@
zwx(l80+@S=mCBs*W51+{E@)TrD7e`gUE?!J5(8FDn9=ZEpE+8+nvz9dH90m^(6HOZ
z(!=FF$STyDD)Kee(ju+tr9JT^5d}?CK_0mE4EU$ZEKK72*fS_di3-f+_#p|&fnEt$
z+4D)$Bse2;(vNO&OMs1m4-S!A7(nu!v|h??4WMYbcc8mrDlk{4sf7i$Dex_C@fr+T
z&7vM4J*FM9tk$p2l|hCUK-^<aV)+Z|h90w4H0lXCIp7nY1y*!oHp52Y3R;FGQWQl2
ze)Zu!X@_@q3nrq$fgPMQ&AinjTn@fK`5Z@x_{UKFc&Pp#R3dE|5#EyB(dRFwFflk1
zIsoDDc`r-AWg^WEh4K%&Q7xzQPYq4MLFGgZ&K}MR@2Pi7fq)TsK1Hvc<0_|)lS9xK
zQi!o`U@X7^<)>SU)>hhc7qfrXz<V4`4CCaFhR@)dLBWWfUT~@vKC78Wj`VMNAc=e*
zE1UlFc(#nFHg37Jjxmi(T$E;AE)nsHB|u5T!Vv~Gy>cXbdF3(4EV4(1-9smjNRTl_
ztbvS0fpE?<o1_*+)~AiruM3<^(^PmGfw!xSf0Z02uR(jzgfH*!UrmFXC7m`x_-4Tk
zwbqaQt%%uCU67U%W|hAy>$b(oygNGd)w;m&rRR6XwY8E$q;XFE#!qbYtl?1VM*vu|
zdz%!*%|&+C+mbKhls+JtrLCCdYx#7LTh!Vy*Pr{9N)Dn6Gzuepi_0?B_Zs~czmx$m
zz2Ym?-6S}NZ`bYy_5)Sp^lO5U>4i;4jBgrwD!HH=-G;&V!J73X$KXV6&6kyG$7C2+
z-stIPnK&!q2u1OBK!&l&5M<4}I^cCpjQ&anVBI%95&-fvX3f7^1gh>JYPsZua%Tf}
zBKx>$JH<PPlerSN=m8HjSruHu<MyO^mS>&pQ#pUwPS3^8G2>L=IiGb`8w}%8k<ShW
zHRZD4)ua9R{n#l0j1DsC`u}S{F4q4IBq97i$^#oCGvWVG9)?vWV>b!Wwohp^jVsLi
z40-hbDO0EsC@*=hx(nfE=W~vLBr5&%^tqL}6}v0NEWzX6M7bEC;US4-q37IkTwm-b
zsFf4Rsi~G<BOhijEn};h$-(Fsy2te#*r``HWY7mSYt3+t&OGtcSbw3RFHpzJH%8qi
z@lrx2QdD(o08D(LswI%@9eym$x3d0=2gMnp1^GGb3b6Qk9!=jqfmsC$di6^VS{uH`
zG<xrg3`#`^vhxG`?`%<EBO`909mtd^Y@7xqy6XnRbZ7t$lSI<9LwKb1!xge>xYUba
zmyL+<8Q@hv^D0sp#O-`!j-utd`FE70yOR%p1du-S0oAUr=)_7$BPFC_YBMHHxi-G4
z^559pK!Zp^5ub`4RZEURWzp^WO^pM}h|U~etgrbm5+!uu)y^uZVdfPxmo`@$ww?1o
z+h&k!R&}`J&O{fIhY-qa?kBubX^Dc-iW|jS>yOH*feDq?iarQmLHAq>9nl*$nw+6>
z1P+m}fD3FlYsLgQpp^iS)OhiUJ!JR$ga0sld{-pH7A9)OV|SNxGbOX>KXSddCwEOA
zh!8naY-==<$NVvcL~N+>BLwLKXk>F4PkF809`_JSAbE*tndUZp?&?xyqo7(dK&Lj!
zVOFStDD>E8eoggWo3;KxC8S)-|9ei-+MR)6L6{lS3VK1&no>11h=7^@XAR=zL;%d1
zb{_<ShQi9o`oDk>X2SnWk}Mo)!1cgr;LKcXtZYg99289(=`NrUq}9cRMPf}V*_6Nl
z6A6+zqwiPo-#V=3l~{9mb%ftDWR~($h2OvVnLkQQEagC_0zZU5GylYAQhynNLcWsz
z;w58Ro}0cX7!wkt0R$DNYyaEJz4arAp*Vf>-_j8u(Eqy~Cf&=B-#epGDuDC&Yk$&P
zPdzu@KO(z0f%a5$L_q!^E=WP+p$8zJ+aH5V4ATF7SNH{>#Q$=w{_fKeAMD+T&5jHM
z=UeQXLo&MjO+{VeSI_&sXM6ul{JJ!R$RYgg?{%R3n~~9(0sZw57zFK5@8&#siehlz
zRiRAjY#+MlYzs`6!8vx^T*2P;YP{<%IR*9zn)$Uz^?-L&?D^z}3VH?GG6o20ur=Vw
zMD0=7NvDSWW_(C_9TvgB#GhU<RQ%7AWAMOmO<2I6I6kvXZ=5CN5Me_ERvw2R<=pDV
zFWMm6RiH<faF+6~`dF|XeTs?0n0vjjlcB)ha#SbN8WOaJ&u~N(`x%O1aSHpye*`}z
z!2Rk*bh$uoMvbMsN|Tc{Bo<J8H8R?yQqm})>Va6^H7IQm_(m@`_Oe`|02UDdwno5D
z#yW$doK5ecP~_05on>(jOUIp5wSSP~;KVF}v;>J~lpoPri+z18CN}OduxuoY7Vg4W
zVh>cH$UoBU9dwtNbE=A-1P7!8rQVm{-;)~SXq+Vgm<yhq`K62u(gaxl!s>c<Ld{JJ
zs~DQIGUIxInc&~7n+01%;wcesIM5R>WW<f$Vvi^UdmrhQ`Y!gf^V5du&RBIH-Ug3#
z9GH?~NfC^x6hOPtv`hd5Zu3;jd0<OLk)0=-#>Ww6i3R&#`^4m$?zM!#5L(L;E7BKA
z?K~P+lyAhE^SE0v`v3uap`A-~?rB);x+@xK@!46^K+*sGj_v8<>wi_OL)>!xSv^32
zq`H+$+1ymidLIze#1sli)j*vg%KC=}>}D-@>BO)_ott^7%^USP64>h>CjQw}<}+(c
z$tz&9yv_Bhg8}mQ+wpc3c|AzZS<iP5oas6mf%)fvOlDg+a}=ODrfaud*Z6lytmOy)
z6nCpEEfF#ruladALbUkyZ82aj3S4o~Lts<K2eSHX0kLNsGxQ&s5l@$NLA3{<ch@um
zce_@+WNTrC7vtu0-Hn2->la=-dDU-2Vi_4l*&+L=V4WU_u{~-{^9`j*7DZq(g)T9x
z#!{7611+5dt^hF2-%>BFQ&3XGjt?Pz{FeQ7c+8Xgz$4j>P$nMQH-N`8f)5k34Ns=r
zQ~R$UWZf1v4G9%y^i61Xb~jHG9^58M4-zj7iPTlKuJ#0U&fwo-=A%&~l8MMu@_6I9
zWH7e;JwG$ihZDHSEkZcO(9KRv^AbxQa}`qhnX$2xo<AUmjFcx6x^rI&Aq;i;X{Zsu
z<0X`E-cnA^Sz+fe{c)NFU|kdyWQ~T766nN2Z>^#-8c;|)p~F`S>AjteVpk=M1rYo+
z>`~UfXtP}_5R?X(wu4wG6gM0tj;$Pt`38F{6~#o{%Q>+35GB|dou(L;@U;Pnt4D!%
z4R(}FbOW4XTlmME)rBs<d1s0o6Ym>iPr}~0x7yZR8K5)@X|TW|ao&^{RO%0xHXHQ4
z3lHp6;Z*;|C67dHO~bZZ9`fIjnT9!BhJclDr@~)+y`9dF(r^S6gILaz?g&3Wlk4v^
zGOj~dB2bSbU}rM~e@vSj-enmFUUyHDe_o%h$pRkAN2m)4S_vKef(a_hcEA`26Z`*r
zjzmSTHBS`td>Yl`fV{IUAq=k_o+<umq#KVtBxykFv}t;+FJ4FS9N&V~rC`Hg;3ym%
zO5p@gk@5G?sK{Q>mVBc8;V)B6NUGa(#Exn#%iW|A@!E6E`Q)k%_9}^X(`bym^{b!U
ze+I<QTV7#IYSFY%%GFT+hUZkaU4;F_;$E*_zK_CUhTI!1oKfGR!2Ib_ZH%dYhA8Wn
zTI~Le*0sLKPxg^}hA{}_Us#I0nkW~UFThXsB?6i-!y!F!nV8MH(&ZLi03V#(gYr;)
zJ7%%AGC+_kfkM;MLA~2)`tm%deW=KQH3m!#C4pcO&_v$k6Up4CyyHLRzd#r;x0`p+
z?gjS7G}g4)PuU#dww0nD)Zl2PWqMp_m*@Bp5D~nv<34l2wH!C0D^jKlH{*jj5t3Fa
zg4MNfK={e>GyP(d_~Q}=(2>jTSBXQZsi^V()X0lbk}NzI73!|0WOmhU&he<rA_LOd
z&><TyW^?G7;)daGvN%^3-YS}9#iN2<$$Z=Lt~XFU?umHjdvVMx&|s&MVUFpkBT@0<
zSdMw%kvxIUyfQ0ntmOAe%M`LKP4WmC0!C)Zo0H<Igjo;=HG(v-2X;c8zu!&987&_k
z6VI4@nVC&hpKsMp)W80a<`0ujUIX|I<xqV0WOI30%U72v!@zFCg+ARvR!j-bWb&u{
zUchr|$ETv0k5C>S^CPHrbB+_EvdN3Wes(JJD!Yb{f|`$<$YBN2^*HOCr4`>{(9)*w
zv0%}w?1Ol)DVpDW^*Mpocj!mM7EFDzpmnrOhx?R%!u(+b=0`B~{SsOCbpe(row~d@
zN3V)4nFW)lPeKXd_i>}_^1a60Z8H51us5J1dLI9&5PNh==|N#!Y5~@{`4@bWWoa2A
z34#9W*?PM`9~^l7zCkcpNBNh;n-utdM1~O)=;D`UK1t8g=7qfgKn773kFQgBI+;eO
zCb3=PyC??%%iD;Dpusw*3qXMbzua{ZgexxD{!|OK&Cx*Vp8|m#7CB<{=sCf<8^RkS
zxmvefmFXucG2@0pi#YI1qa>e(@~H9@W!Ip)Le+I!X!fAhx}n-Bp}LolMBgz5s|^=>
z6<Ukc07}PqJ9UUY9N5D^iw^-{RH%T{SP?nQ^d!Y^2#K3$gQ8<96W~^kNHq82Y!K)u
zT)gK+yCQjIlPO2VhgM|~&x|f`qp%en+gaK1NO}07(mgL#oF5M6FNl@9zoKvL4s+=~
z2$inaUX%Kf!Q7BkWK~2BBtZ#&X1?`ortUHe0yy3?iA5#+Y%446)#SJnn-eLNm^|n(
zq?L^k-^mK~(ct<30%Ut=+Dhj5oQy}8<sXZVV%WO4-MJ2nC_bAH^?K@4JYo`I`J9~#
zBYtO9{>^1D01yZ(`mr4a<YlQLnqgJRZZRUi1&8>z%BCG5f>Y0$c0WscQ5?OD2w8gF
zu*|RNiCoGsX*g2h|3lp1nkH}&2}Ot985KF}S9HRXixYUB09c8U6Ge=~TYqqR9IbJA
zpvBRs+_+f>@=eiQ9|*2~RqGT$$B*g`GGfc9>X^lsvu}Ji_-X<W<Xr={`#P}nfBDeA
zHzPYr^gQPVGfDRYm(m8}sgUR8BRnQ(HXcLEXieZDBo+eJr0yKO3fSF2s?4X4_hpBS
zY`d^$b0u#N05wpZ-dxdCSCK6?V}u<HRRjI<2ylG+0<TFg8@Hc}c>CwdJ>(9v6!^jT
z8J2k#i3s*QCj(L8kH|7t=`ZZ!Y?}OXvcCEs%0V_M!vh$>9c}!P!l~be3Sj}-&+V<J
z<;^!aN(EJc9j{hrmE<v!+j880Q9x}IaAK5+t<5>L0a#lPNe(mFY0|-kx`xTSbcvqs
zgN));6PIrL*|V{4*d>eyWYEk67xFo{swFKo3NqlKdlRK(o`|^}^VS?JFj+y>tLH*p
zYQ-{}(&pPb+r>rF7SH#<6lgsi4Kq(7iz12LTJ!a?tT@(z@fXq=;%w0f#F((q(G#pk
zB$;dH08DHs36Bbq{&uku62{(C<WYk!t$nL@Z&!mb_kLVA5kvSzVN;88NwtUS-YD`*
zSN^Hq1^HDu7z=9OXlJLOgnP3ew!Y0)Va%8~z}4GLY+XKh#~g`}$I!<jy^ML$UjjB=
z%d<wkV%Qyw_z_5~+RamzhZv#|CL8ZNJ^pWE0P5-d8@O&O8=f?W0p(;@aDHA#id>sM
z9O|Z@r2WM`L(OOnv<GT}rLCvhA}ldvE>?j><-w=N$l<>~WID(EMU<28O<E#<YanV6
zO)0RPYQ6?f++#1te0<2o50jD58r83lz!z^P>i5NuO>pvi%j?s^Lk#w*U>kG=(6?l;
z0Q}{MynST$VV9Q4nB2?t@_#7p#olzWbAw~`ue2slT$y1B#KFGz@Q(b121sA*da%iU
z-p#SN$BqzPS40VIhLxE%xgY{P4zlSB4^P%E_x4jYtU5`NYfj5(7nmNiQG1NIGgRtl
zTL>5@j7Qg1(7~)%WDojCEAkI;4cdhq0K~{8Cm+sg14RgZ2+|51GbOsJp-%c7l<0g(
z<NG|;oYq}InD%*~dP2~<xw)q5C+l)^&W)Q_Q`P7zBIu`gIYszAjRJsoHMJ+W#in1<
zBq1W*_n6#NAM-!LYlta+>8(d_i#Wfoy^T$CnjC}@ZLQAUU=!0-J1ASH;Qow>02&j4
zMpT(5=?j0))qwO>zKqj;GCNuq?G;>rCq!+)`$s~hcfQC&pV|^9rlK~MoJS8R6P64H
zzXuDD9jPr;m~VTi0k|mARD}|&MF)8_+r`dS90wTz<}LNu=<~<ObN9*#G_caSjAqQY
z=rp8idPrXA7)HKCFulX3n7q9tz@ZNq`@pGe$N!)D1Slp;z}$*!PMtY~G}Pguec>y$
znO2Vg3GENj+ly9%25E-KTPtefc|h8Q0&z-w?xrP`MquSFRFA#B!*whR@*mvL>wcgO
zq(KbG@5=oyJ5G$^mIj0BN<G9}ka18lIz7+~=!gNd<dHXX{*nD2q!apd0E9cMr9YZp
zuZoDFHDa5iYxGB9G5fG-wfP{<X1a1g6IFIfdm~RI=0N5d<9xK!5mT<}>l}|1Bg*c&
zCf65B$_u4M|HLn32d)M2htI_Hb?K{=xmpM9?84d@Han`8pM{X5=6LS$u9*ozJpD#H
zrZHo|2<LW<)Em4W@LrxTAOv|ThfAwo{a*g+_Fi8%WJ|>eY;h(JEh{Bsr>F|QJ9oA!
z$X@D7qUAk$;cs<7pfaa>`yg5EK$~52LVC`R#82WVk#{8tjtqA$MtMWr`~FHEN)^mW
z{QWA;x8b06zl|>mGHeP5CeLRcbXiB@Ne${W_6hX@YqHD6l*z*?pr_d#dtz9vAGZVA
zw<BK6wS;3HMZ|xGx(QXAAEZ=#32z>EE+fq97Rt~#n11Mw+TH5kyzY;hI=kstmZ48!
zATlj~UUfdVcGgCcpYk3VK`m9YI2cXN+Ok&37FV&Al;y55V=+9ud(T-P4kRIt82;wk
zByL!TDo>VUh+>Q(K-WiO!utg~{ocDY&qjsS*~!Q^t@Df!dt&v<s-o`XX35~GJGU0m
zn#!v;w<myNt<em7L|ZqyeRKscLyw)DiNwHSHgf==>aE-SK+K-a>ATufm353)c89R$
z!I71gI7+gqt=DlWf$wYg7|Xmq8rnoYxB3L%Fob23_jR=hKxE>6^SHe4uQfov-KvpZ
zad26rR^57@@da&i+vPywD3hUS_JoaiqmWl3cOV^5!dlB$Tl}Ee=F%=B3c~acbKgoe
z7uAY925U9;ra`BFD(HI?14>K#`Ed+}-7?c+9NBWpQcMGb{KR%cRGvs}yaaaHy|T2L
zfDp;sC8jI}kd1B;t-ecWBS$PSq%bB}%$Rk}CPMi7Et#}L>67~u8vlp9f!3`tfT?F|
z9Z=!3*6Y8`<;W6_dKdjrGJLg*4qr+~=SIVGbeqo*Krda4!U7*6lEBQ~lAAqA_g%Ft
zF^f_iZD^x{-y*$Wz4l!8p-w1T)aGa#Cu`HKriPvbu%5>!S}v(luyGkU!H%7oL6z4R
zF0Nmk);%-WFvHm~R9ajT$8;?o=gBe2uPYAhFxKhioXTlhag3UjB2tvX{FZVHCaF|S
z$fQH4HD<=@*O%jE32#ZOTEp}$#(htqd<_ynxUYlTEWMy~mvJWP)AsRWgtw@^<y8K&
zmwCJbEDdS7InG>$mYa+jTuA903NI4++^j7<iS@ze1$_^5j4E<<yfVE$U@l4qPm~%E
zAGmjhBQ_@*N`-inC!G;{I|5-`34@%D@znTHvG3(9rkjx}Y+waYFw(~G5m8etb0o1i
z+hx9PSBaa3?p}ozcsN?N;dm!6ga`XN7YGdk;3!@(sk0uhA%@Y8d9Zi==cfzo$;==8
zc86}$P#uhT0US7GTG-@f=?;!A&s>u0!xr|-zjR?5n>SY&o%R&#@Hf>ty$!<a>feG#
z>?P*QriAswN|!O}>;3|#vF1#+4Sn>EItwFxJ^t^tI=n~<H6XKLwVZ@XbD^B3#$@+^
zji3+R?#Pn8b2XmY5{c~(%IC+z_S}m=yYjV4bwT>R%jj?o1D)ShXvBIwg^f5Ixi=Vj
zrVd$qEMPwzGIMi<A&18AF4w;Q>M?RTg?pXk59W{K5qi)*x!U1wQrgzK_h>#&UvuKS
z=}+^7)28UaPoL{L%gJ0nsJb{=@3^A@sKX9o>Q?D8QKIL5o@za0hQu^ulQLR66a=rU
z=Q^Me`7IbwEu&U5#jA>qs+2n*?aMwi9WyHSY7*`pyJ$_x4YrK|r=Bf!sSc!~r~@3=
zq}5tRBqBhy#2AsY5;B{CxAFB@-rDolg~~)9anNt78_1cY7^f1@=zxHtb9p&{@-K?K
zo}l;+(t7|>H=!?aEK@)wf%t3TuvTsCTWU&K3i`iZEMH-{?7GavWlka2%<(P<l@O@a
zYGv7Nz2gB<$N<rA?^k2Ld6NZ|2PI!;dFevE5~j?zs&V+jMdL`U_JZvX{>g(^L~mM(
z798@Gd6?xf>AQ61&}j6V=~P%i2(blEb8?sdZ7D7COGU%Arb(YcXoZ&2P1#>y^0Dr-
z26=RFdjrO&1~b-(DS<b`$B(W@l<U&e<gOon$a|AFdm`F7bH}h26_=vDVbu2}B+vD+
zN<r+DLPOs0Q2g_PvlaOP52}^tg^1&gPz9&<YFXBSt1B>g1O1wkjWa#KKxZ=Qo7EZq
zxucjxUTSC~;#GY+i3qa4Q){(jb3{@*vWK?BYwkq?$E$=saa-}TH8{ZBImdE0^+9zZ
zeGb`00==F!s^mEVl*XTNu96IXqpZ40iHzh^qu0mJ<;VRVk2v6M-QCK@*v+zsDotr$
zaV`9}di02E<L21&!v_WsNe3>@*Xl{}&njlUHD8l@csz~IPKGX0ml>#ga`JE?Om|85
ztqE3gs&4lZI_&wjRkcn%Yw3fV>j4Cey1{Tr{m?ebzx;tq!1w|pArIwhjYN+uY=Q4~
zUEUD<#Bb3?h5gyz<;G&Pz#JNJIOT*-ZL!e4Xj1a}ua0dFnT;hNd2cw?94lXYJ5TQ~
z-G$)YcEFWiYI|dV?O5#&=EQAeR3|mY!O|i8?^8g~EzZPz;Mg0kN)FI#d6L{RXM32b
zrVh8d)TCNgwCSoeZUu*F_RUeEv*@v2@>y}C{1Rp!QWv8~&iftr@keM?TTw83yW|kT
zO21bd|F}YQxk&~fcg@SX&oISlR9NARZD6Bn<0sv{OH#g#&_8-_ZYY=iBJ}PNvp~n$
z#ozIv!$Q{b-z4UWzC#h>Y|aLJt_&s9L-vcO@gd{h&S(L4oY*n525#_jBRE~X^;du>
zZ{+>ygllpgg1F3a9KzSa7b-*m!4a58qTORL9aD%Ht^*7}?=R>{&oer|y<Tl4k<2qr
zdFhqiceH5oe!18I$gMym==D=CEw~WrFl@}YFx{@bHT~z$hfv`5YI~UXxj-MsFkhB-
z_ZDRM+87$@`a6nIt9~7m$gIw7KMGwF?=DRRj>Jf4$E<l<&b8p3@H4PdshbVP!+reS
zHo|D4D#Zjql4L~dG<RUv7nJtMLRL)D&tvYMiJ`$|uDH!wi(zi!68Vd@;!t{UpTrgi
zc-VnHcn*!uVyAPsmi}3Y1=^q00glpvDsnG+^XaRw>Ol*LbSP^?9}~k@k4@E);zK%j
zyMl-JIz4XcsbyEvZ=iT>N5OkI97C%@S8a?!_dn7Ta}I{|($^F8LS|H;M@)USK=}Y_
zzP{$){;ud9iGTNI>5nn~3TgFU^|@Ujm{nnpkz4nw>u3Gs5{`+Y%6UC$NW)EZaI$Hl
zisKDjr<l65mos+y2_Pfx-s@4T+d9pPD)=Kj9QW%XD9O8h<(BCI;8%AD1;0;;3N|Wa
z(sD5XHE^4QQqS~*Tp^sUize{|pI4S0zgZHYhx<D(NbAwrS_4Eel#Af_zu8$?^yy-$
zi4>~7+6JL`5LDa)#-nwHWB#D9gc672h$XZo5fx^KGL>Ny^ys%$(*C$&c*^G~F<!E#
zHE<UAb7!nULhI!n4UuyfBvAX<O7C1V8LuaRCJ{kc{4TL|thHLoYybb+72w?)oJky5
z5$d<csZZ!_6~sXNmb)+7nrvU*W2S>J$w7duT%A-=F-`G79{V@do`j~8J-FxLMnaxx
zkFyRxZv$}13z`4ehywenYVzEH;k5=eK5Kpg#VYU+^V-msTL^_jT8+sMgZAiIZDg|m
zRveuLgB?)|C$zGgIPs|^aF!``83+y`WPe7ojdbu$FAT?c7`L9*wWz`zVA^pyqx%OE
z9s^A4?gy*?YZdssT^IqhMdF)Z*yqS9$;ygz663o;SNSTZW=xG2Pn+hq{wmj^X|3`%
zS3Z}ri~8S9XFRy3>^F@MKlEE1G(PhHrmE-vbAmgjpGTWod=_&S!M72|H2<sCX1PDc
zgx~yp%(Fl{kc+B!MUi{3-ZgW<x3>M2)M>R(PVsTJf@{l<wQi_!k09pyJ-9FsQ;dxq
z8ks*MatQ2%dSgm&iL}R!@hkH)3C@0MWJ3nU>xSwb8Uz5^J#J<q2hoRhn;P!`HT_qV
zk9SkpF<PTR#w++Y@3F(+(KoBiH$qun!QloO;zC6i8`$TuF;V?=|8Y{qCW1?^vD^|0
zKK@o^x9d?6f2Sb>sCAO2N@-9B51gVUWOf~}=dMeN-5cReHmSC2QC#*wMx^`di0mzI
zL?9!7>gejcuO@k2y;@H*!wjndnejt1<-sogxLz1c!t(Se4+=gd{QH#MiEEGn__Qqy
zlxND_-L1CWB(l-fTWW(#!x=f*qV<;?xksz}IId2Eoz!<^1(q9*Q#USQ7Lh+3$Jt?2
zff^IMV+JhCgP&IYg=F$CT-f0udKlg!@hE)!vj<M&Q@roQav|28Tn1-=SZvnR&}>L>
zADNG$URs9>eBUGbJImeOmu!`=%$m)#2w_7~xmZDND4UY^9_>~oD^gwy&Tr~Eovyji
zml{fCT?J69T>QL3{))OJ%@bU44nd5T;$JhUYVB&~dbetiJj^YXSeL_o2x4b^c~{?_
zRCBRIj9kSBWjqsCtiuF=?b2~eJgoQ;(sJozrTr;23csqe5-hXM1Srw8UH3o)#C`C3
z$jhS~>2Bex=M?mF*=q>o=H6pKV}!PQ@k!O~1+ILa=4m2P4^l+m!dZ5O%k-ZadlWCz
zD1vdj<C+cvcifF^;?hn_YPtO5wF607HjFRc`ksL~<9Nx!_^ws}ALK8>dt`#u2)HK-
zr#yVKr*bYD*YAy*8k7XGlp78nyE4lF5|JFm;V|TfINpv9$?oS*LgId;M`s!5$k=MO
zM()4<%!#mn!6Pf#(lRmyO=8cEB*TO&ZIczOAJff6QOf4uG^tSg&iZ6FTS~G;m*e;C
z&(;v&&NO&627e>~Q)c+x#JHA$-~6?Mml@6i*mk`0A{h90C<dCu177-D^>gZ);@zCT
z(pe_(b?7Xaz4k@Uq}nB#-^lz-c)h4~g0y&brYCQ<jm6d7#^Rs<DpdY>L?xTQ_fRN$
zFfQB;eoh#KLj0kG$PvguQUVcnC^_+132)ihDQm{)h*c^D417tpR#P^QzSi${Y+0-3
zjn??}Jt>BU(e&T%qwCGZi<urRuS~|i4Z!K$^kOp`%@`5BS3#(AEgwcFqs)FpQm0d?
z-O9TRg~4FRvPm84T%HEE&O;XZSuS_L&Z05El1$_@;48!IX@I+EPb`FbIkeGi^25UI
zm{v7N9Hiv|Z0t#LL3eUWY`NB0>jq-dU1k-@1q!C~YqvBa0t||Z**Ah=)uBQwLA%*@
z8KAn8qlWF`Xc|Xoef6GE89_SCY*F5(!LWw>uT%cBUw6#z3vNH3%bcShZ3T*1v%A_(
z+d;d}m~vq%Y3uw~G;`c?g0beD+4tT*FLjp@t8rojs5wY#cnmQHT6STJA=X0+os@k`
zYx*jSoMSZS696HjeaN<H_gNLQO|ARjk%ePIql^24y!J!|*V-^-U?Vp4bn1n#qftp&
zZ&jQmQl<<GkE1)C!3cx5=)tF^WuAtV^qKvMf>lLS<plGwvMn4ap+`H&n-W>GNs%<!
znWfAQAn{<Nz+HlYk?2_cXFyyZZPjrb3R6w^@_KRAyJFEwq52v(4Q%STz2}6?e)F(#
zphWuT(1X;2IaLJ6@EUH-hv47RD+L(+I08Q%`XwXdzk-_xc<ngUX6e@P$nWCg9%yGO
z?WOByo&va=nR$*I1wT1no11;2PRG2Goylf6K-<r=yRpgB-wr_~;mXsGsbK8dP93kR
z>g%w(9pdgBw2~G08XJNRZ`|q!JZ=+h;b`k#!b?h@`sR86^f6P0#{1{hAN$AKZ(ZzI
z9IR9pjX82VQuG?3dq27XqFU?Vg}sE>PTNFGKA*eCi@mGlv5=eGg`?#H1}CMiI}*1`
zz*Z!}ySzay`HZ3W{FsX7I%chFe}k?oEo1dz@{1k5DMN_Aa;2K__3pY=NSbp&M~^^W
z!@f`i4RWu4E&e~xH3Rpq(!0ruc9iO3*1V`s(Wjc%M>-U}m%Wq;kH@DCR$VLWzadUn
zYg>rCTNrN~i$w)bsq*KK2VYO6esYZ_045nSmIL={x{l1OZiqV_bhxc|>2KfvGJph5
z!?dgVb`qEv%Wb$+)Xpp^!;^0Ddd~duB8<JcuHCxXpt1vz&c|jFZzR}^?hQeUNM{Q4
z!Eg7$0z^c#nAT(^kEwro6tjf+<*&tlHO2eM9^721h3(&-)@#tSp~#?9NhE5z0Fd6-
z+je{bw4WMd%>Qu;G?XLSenw3Y<ECkGTiAw8l?Lll#SN`l>t&BGPM1%C>}047&<xZJ
z93cIsDyNKVjVyrjg80F20=;hsQfH15V^`E2d)@5)$+dG~rmc-0SLZ3(N&F@Wf*KPL
zqp>^r7fk)$`!KdHCkoiKv5h=$0rUr55Tt=fV*_vSc+|$h1GZx-?y}e#fkwQ{qu12a
zN@=-F3KN3ki?4x)0$+|w!2<(V;W6@Co2ffF27i3i?2S(4P~ou~1}LLNs(WgB&iyw+
zNt0_AEe4U#hMghUOX%5D<X~4Z`(XWLO%7z}#NBg_sF%@E|1M1<hzlQj1D-7hkor%5
z-!z8?BlK1VX3XZL4;ob?)vFTvuKzI-Qn-_lMD29(aKDcD(`4Tl)8h;JP5dik<uC$w
zv2+MNv(S=7+&qw16ZAX0tK?@Y7TSirREOM;CF@+Cc@wWt7y@p}WyAaZu9WyENKC66
zN-h1TPw63Jl7f4jnI=|~8=yd{rNo!;($^U`_1^O=_SqCU6I6-N=cgwK-Fl`K;r6T<
z#Uu<xlHWew;oa#}C3Erjwo#PWC!w4g;6MxE^#wLnCPlBvo4cry<!e_e6rdUy#uol}
zC_8RH<t!Wl_R=l@sj>p&&i>|R)43>+edm?8A~pV8Nid}TWvvPa0N5DiG09^mZhji@
zet+&pxXjV$c&bZ2mm2eel@(%oxg8kE$~0`Sm3(6)Mns@-;;hseA#3#M?zkr`gr%C5
zoeosxxqH+#3Go-kO)zx7i5E%lxq!Q5$!c&&>roiB-1moHxF?n~@<+ckKeoKCS&&39
z4;a4kk)_iozOpZF1pw7ChW5sSmPO!OO)6`TVRe~3cnrQF;}5Oma<3chye{-?ec@*s
z8+b0=SHQTK7Hrth4*M{I9n@q?2hUh8NxL0SJ?1C{%%SU{>O1`HZWzATOG@XxdZr4y
zo>lOhK~3=wY0S%zlS%48t0)0w%D4X`_EqAax+1q4HjOpk1EBfcH~CdxswiX8TchE&
zaq^5tJgIue$fzHW7*-_E%yle{^8S)#E<X}BK^+7h_N9amnCB|b={_0v*i@8qihp1H
zimAIbaVm=RKhH`t38)|N85!}Kb;(xr?RR!4Lidr0^<*BXoL3q+3?;G0wYCYjh;yjV
z4$1;8+9fj@2gnTadAF5Nu>-IS1gWhyZp7`m%nyubbZDZMaL?wM$wf8&dl0C(|21}H
zBpU7Ur1e{q3HleC+|EZvr%<GtUH(@#yXUY|^e)X$oPGNiTQ&vYxy{-z$smw(=B15r
zoOza!u^HkUu{4kI$dT<iZEg3XVnz#7+8-e9?V}ai1aJq>XB(Ektx2tTo{mxaWZ2#5
zTtsi;G|d4ochhE1%BsmoqqKqDhbQ%bf{ORpk~Es<hfbW6OkfE)mM@()z<or_My>AD
zy;4>0Ms!#g>Vj*SR&|VjyW6_KxtA>NkC@C&yM*)8UPH*HLf52bTm>nZ!1=`!>aKVn
zj0sP00487(S8#^~gj?jpBn~sm=Nh`!vC7-bE*G7pbSae_yEZ8s4x9-yILVR#eF~>I
z`|nOMajE|y5#TcVcD&KaBWh0SqF>I5W2~UVwNh=2K$T$OJ88nl?fwQe)2rmmecvTm
zp&A+b!((s#<f6Ax3i_`}x)sM``&JMKJsWGF0UWuPpyvdgiw1pjN*YAum@%A^gyesi
z{RF>W%-pG&@Lp!fv%NCbe`l}{=Q_sOCSQW%Q!}l=7fE;E%_nfTOZrRAf=zhIoc^*h
zM_=2_NE5O&PERYen9R$Y|DCDZL+6|kDj>|alWORgZ6ZgMO6Id?2QE4T{jN8*jUKMM
z1uVfCXN!dBsi(;~&rgG67?30w=YuPy?&>t7s#rMCx_Z{Fgbx(378-6uGL;Th4CNSf
zlFZ8XqPf0Z(?l#<+}GElL#>=%^1~f_q)mwA-!s$a7?y8kthnAV4^Xhjx4x0uC?G#_
zI%0VvqA;+&A6Loaf$Y&k(xJ`jlj-5|0R;Xt+Fj#!R1k#yKo#wLNNx9yjXRQkL)p*M
z1t$zAa)<`KhxWN#CaDn_%{iSqN(yy&WAyR&GAisRDyMTRCnVrKKE1<#KOS)?-5J#U
zjPlQys$expNJPWO4MBI2b(b7SP#db+!Kmp<t-5YDi!{{ToRDVTNedG+Py7-f0-QzU
zh(^#cK-^UmSiBs)Ii|moXE?T@u5Qt*Ua4jw^9sd#+5b)Wag{sbT$y6d9!0d{2t9ko
zVh_Uny#}gEO+Fe6yc<(Qll}G76-)md)C20M4gIJQa?4pvnF6!JnE|QoHnP(EQ<1a~
z)n1YG!omm$k$VPh_(LT6-JWZF2=FwcbpWIts9JT_KUCZuA@c~hHBPAp&Ux`hYr9YB
zrD!vJ$Hn2b>us5rZ~lozG}1#hELWqX2&rA0hpkTM){*EbsY#B&&%!V^Txr(lfr!Zs
zRe8Xv+(RCnbu0*9U>OV%l9mcWHVp2SJu$0QRpgmjNI^QpYm%pYLC&u%0LaVp2Gq(-
zvFParZUx$Jj%^7bOE>Z;HS@+}E3-DwXI$8!nlKkyI&-MqBB((UeQC?72Y^RW^@(RJ
zDf(hDckmMDt|b0DB%L4Aq|K7Qke`{L&}^q{V@)fOy}Tnir0a^<<`#A|0o{qx6#_Yu
zB`ufZYWTx%=`bbX<;OTl39#kvJs4`4CzEy&WNWNp7_u=ElfiXg)vq@%K<8!?qQYJ2
z4@9a2#*F2=MVUE#EVAbOg~Y()5)Mp##3r*bPK0X&cI-240ty{=MFnqC;kd+E9H3Q>
z#ns}U!mY*$brdxH)Od2XJ1O~|t;_Jh$e<c40mW()-h1&S9p4;R04QA#DJ~w~0=lu$
z3Wp@Gv8X%Ro84Ar+5I`7jfx${`bKvjdZS`;2t&Rry{W>4BWaIVs+YvZx}w<(uM8Oi
z$vRF-UkDisFkxRD@8DTd+eAifm&Jfxs9r*?TXlI`UfO$=n?AiaWj#;OCMChbaDRiy
z2nI3j(9l7#U|+%P0BCG+c`}9in?(wB*y+4_yp)SS2K~cFzS0n)J}I4gJCRP-Z_y|8
z=)}s(su-k_4sg)2WdviJjHbOHbPs1h$GVFG5={>QhgaL>Jv4~V|8P0$+97jk(^%5;
zPK5257G{*QCI3_-f~jl8w#<|uHGbQ}&lwSPG?as7%MW+;0A^cl)GAK)=(wFN8o0}m
zzcP~pcc&(z&&GKC(9)Yq9OkQP<nDOn5fDRHf#7OrsMURUU0<$gL&TuZ`(>+6!`u7e
zG%YvYY<>GQ3Vuq$-leogL!7st!xmD$ofbz6AEfg~b1+@PqKr9%KMWhdGkrAuD>>7h
zNGNv0P5ve_0Lm`O`p?RY$p5its=B{R%a`Lr<5Rf9W3e~sw*eV*njpVF)vlpy_X#cx
zl6=1kxybd_K90WN15dzo)gGl4H`wP~Mzc@<kuguQ2(!@bdsm<=tesDab67o_xlSNW
zGsS=}&z!C7wp0ACajyBzt&Z!dO$<XGltT*h369(S6Oe%C3A&zH_r4#>n^D|!oJ$W$
zGDHS`1Uj$Z_;dO<8W>W;J>d^B<<T{!-S5uXnbBA4h%Dk#-quX&$D}_JJFFxO?b`*^
zWHq>RixvgBy#0Jq$G`As!$@LahmNIl!zYp~3MPpL#^avw|As^r%1Y<@oY&ovwlq%F
z|6#~mTmf2(RN>C>9+m|E2Sq@-zhP99@M_3)_l~|y%GT&`F^x}toU_(uT(+IE;e$;~
zM+k+jX&8f;v+d(EC~%s6MGt?f8`9r&y1lC+WU5(y`DJq15!M5*T_JQ(w07%uMNhob
z(=pGmDm7K&A<BAAO4cKg4Mik5sBVJ}g@ZADBk<O%n-_t9d@FO#5Y1Y?eP?#~KA(j?
z$YGD+JsJ@`jIy5t^m3W+(&d!O6mb^j8mxToZ1AReI%27RZ7F{RaCMMh(B#{J6JTQM
zA}Y5>s}(vKdgTpK2u7hZR>2m{4npt#r3F4yWACg<Uv4R9O}AQsc*IOT8BXhb>_M{L
zP{JlB;23;=k&9z(wE#He5cVNKGg2t>>^U<}k($1S{FyvvKmw5|^%|b;y$^zd08Vi4
zrjMVq+7Qe@TLp4BEW>n&^Ig>Q+rp>wX1OkxCai&cb?H;m)sU||YAV{Fbg6mxeS;-U
z?`6Z5lUww>{94{(OqQbZ0vaP#32Tdu{8&$z1svCZG_KFXouNw@X*)wSQh59`tjQTC
z4L<bj(&d*m-8p3-S3ezgVSFq2u2-iSnuI-~`YSrdD9ujx&_U6nrVUz0ql~d^@>9o!
zIGDmw=*Id_C%Km1uHb9V#rtW$lA1f(kGjFthE8XdD?*LVgT@|t1=<&jLz%NbYpJ9W
z)vRfM7wfyV2JD8*f5AZrBgpw~r3ycs3AK+`_8>-V?GqGBAEh5T`JZtjS-}NH?c?X`
ztGjpXTn1JF9UzLSHTgVDBLofBSa7f^BV{W~f`o11VP<#(OBK$%34wd_3i`3?ZXUpI
zOJDpnEQ5SN4>YcZPP-9D^hI}Am7Yr}!(#=1s5+Kjz?JTa@Mbgv5z=!j{lHYI0l&sz
z&Jv4H<ElI_Bn&3|m+zn1QZHd!-X~lsBwF5F&hPRSyO;Rzm4_PeVtMBgB)BRr#IYAM
z>(SGX!LI1v<m_k$9ZAliWZ4J?refM=$?9Fc!?F-oQRKDkQG8+-5(~u|<MVnXY0G_o
zs@$PQ{_>57D%kV&OX?VNIu3<O5^1|61CJNf%Mdu!nYn+@%?^W$b;n0}G#9?zE7Bz6
z^5+AgAOPkxZX%7WvwV!64jgomW7Ix-s>oM^t>C?k&`ovr2l}Yo<iMOzJ=<tt)sQWc
zpe<fR#NBjbk&t{4yJN5&Qj8*Epx{}5r43hkl<0VEUeN1%hQl=Jo=AEV9jeu2A+}Co
z`xx7$BIyG`7Yz@SXi~!-XfP`0X8X(wfr^3o=5OK;R7f7>SxCB<U2dK9#&*Q-=H2BZ
zx8qXW=oSZyy&HPM{j36c%kdB_hQd^$SG4qkRQcg(9l|0o_UPi_N*^^ARE5NU_eh#W
zif_dZL;13S?`>6lCg~{73+o}+$-5WBfyB-o!hD}xKWbTj5Hsh(SXKZpw2V$Q)Hae*
zhUny@pH12iJb|Ho#fSA(`?hAns0}M;=&E85haT{aVq5NGjDkm>r<nTg*MXq)kP<}F
zdz!a;zZA$IjIdsCI;FP>p~`80ll!`Q+;k)4I>H1kz-Aqt6za^Y*Ohw>)@qtjm`>PI
zOuyH8&LHaWSP4E43ja!Jy128b)SFAd$S@KnpMk@70U1I6o@ZKX!hBsq3F9sUc(Dmp
zgY8c&9}ZA=O|cL|+t(re2I2{vKShMde3(OSxgt6_*NBR-Xv%SSa~4T|HY5LtQedMQ
zQKoRI9xuj#qMkG{7Q@{j>cK3C$=GyC8r+-X?qTD*EQ!t}Nm`)U*L8`DhJ88IM5(p=
zp6JOPiMSPITUm<ktg45A*`hP{g9MmZN<rHB&U|Y+YJJ2J(vPt$&sA&Nk4C6ROv!5^
zTm#Q9#-tt)G<3Op&qXqS@cVptL0Fw8y)Eg%S?dPIh`!>uqB;7fVlt;vdCYT4{;$0J
zo78lUl8OZBEuSSkHh>rL3gE8N<qREZbA*wHne1f@#B+)muDIG{S*S_@Zq;uQ#5;uf
zZno!ZzqYsY>;v1Xkv}R|GZ#%nD-px<2iWc*K)v?3)(fr?Z@{R3pYPwow|$fO{_<sd
zA;`^J_ysfU`CMhEs|-Ppf)J`-oA9i=QAN4)Uef_TqrYp9m}lGR5^P-wkN6`Zz|rDz
zKQR&C$&e*~4j&${D$y{$8-mi$%8E3rwr>R_&*e~=y7dS4{F+=~GB%!oczTb2dBy`j
zAzCVr0Z=f8Xj1imk@r&eQL@<JZuBp+tejCnv!)7n&cViyS6SX)xHDt5|0RivMNTmF
zxp<}opN1MGywGx2V}<FS8i%`=kM&jhb<QvcoJDLFwW1wN#_?=`cowX#hPCWyBYInD
z?DiVvdrG(gOD3N+Wlpb_Nd`|k^d?lz(Ce8GW{aAUqdKO4nDQ@$TIpOzTMv2C%L$w1
zNBQWfdAEsqX`SeYI%%>SP1xk7XKNr{T`&D-2nr2#nMf%1c%GEQoqbgrzG5Ncp&y(d
z4xZ$R166Oz)Ws+4LxbU=s>m7Q)q)23&0@6%EEQmP*WZ!2`AIrk<Yk`9({gA&9D?gR
zRB=jov(emtyi1YKHDy>~CgtofrBOVz8?CZ}fbYsnI0RB#raY+}E+98uzFSL%cW}5o
zRf7;z-o~?1nxEr?ji;{^YQqL>{^ZxY4slLMyR9XjR!79@+Wk%<K8@NVr3c#*l5K$b
ziy`YS5f+h-M{PzUIP6;To7pZ?!f+U2#6IH;`aR2k;e1P-7woR5NVs0e9d2~TDcc}e
zVDXQXaYr8}@$?@f&C=yW`v_kgU5##MVBd2L*<!5>7bFJEnUGuY_~TZvtIExho#<AO
zf6ZyVSs-(BhZfn_&)R`;HAqpNSmm+X5%5Umc|ZJ+aM4r@hUhw-u;1i*vJ=LbY%NC&
z&vTZ4v>eSI-(Gg|0H!1f?TN@@B<DAqJ?tjH2RUb#h^Pu+O|X<1`lL&2qm4#G1E=a(
zyC=`VXpCEKF7z(nL}yyiiOA&}a|c<Y5%Z6@dW_?oiUym)3&_ER6Dr<;cgvo0HZ@T<
zGGS3l$W9RNlbCI5x*JbM%pQLgC7d;$y+$Q}33XIV$g?bW!t}%I<&IOs@Fr^NI#6<d
zGiY70kmQu-H`g&LQr{`L^oJ6vFO<Af^V%3#c^xF=>4}^*Z_J*Aaf}U5$lTz+>~pMD
zxARdPgN%>;0B*5M{lRh9&f^y^M_qo2L9<2!rpJd2K}zV2-7?sr#j>4(A;ab4xy5;Z
z-N=DgiZCG|gpI+-T6o6zthi5=9!2Db&1^V=3GUoa9COgJ`9*jKbH-oD;(*~T=xKaU
zTTv*2uX3141>miY(wk`}fzE;|5{4PDu)|871BkuzDfb260-h`YO{en%&5Nb7NHz`f
zL)MBKgbB-%9|BVm`3}sF(H5&c9k~U66JmAXjV!0ecDUXMw}8ASd0$<Mt&JhrQ0Uh9
zkw#r~v~cI}k`lVx&-Y8U1j3-0MIl3=F?E{9hvxNF4P@v(s-&(PyL=JQZs#!?f(;og
zgmiUyG^<cg8X@#W80#Al)8&zqHqg@Lq%S3=Z^g`W`T*sH(?w47(Dez-D@$j8!b(<n
zSxFs7b*;!HE0cp>UUw#iv2+V6iwn?UvO-F^u7Q?pyjHwFks$S%+D$8x?_xS)!X<9H
z{ugA$XiwO#k89HTmk+ZOEl&a#iM*~^yF`Ur?DCJk6n=^Q0vEACfl!&qi{K&aecW)a
z1vMiBmUB}sX~Xv<ZbE$e6Zg}9@dMSq*KqDpg*{CvnIEkSCEm5z00VioosjiNG_tG+
z6TTP?m#CK(I~sH)3lU-=53BY1THWYVUq-NPI}f7@H0|9T_vi0)Mq8hz+Ohw1yY&7!
zgslV@+S^~@f&S7w)g+=R>UNEZ%$SQ_8J3t<z%B^42PJvG_;}>u#f)Wt(>fl~JfowM
zuFMd!H6S|5`JVbyj;OPb^2foGmAXp8(VT%Jlu&wcj%y<CU(3gB%}EGz{S>Y?6CTdW
zz%WTE9o`R7WzCy&W^(yU)5rXb;Pwnxb}2sZz~(4F&un-j#{9gn(YqJqV-U<>IhNt$
z&~_zA8y%zA!s%45N)7IRLkiaWCdMNZUPwZlq3vR`pz-6Eum9<k^O}Ki2aIQu2)lKu
zW2G9T{V}qxqd0h5=2b#Q!J<gNyYlurRxt_27o4;As5!{RfUCaTQ|c{_0{xT<MMk%j
zUCA+Ky)M!NR!80B*^9gX{CuBbOuRD5Vm)*1Fmqx&DM#z*VNKe9fh#JUW~kzYo+C*+
zh@jG7y2He5Dwe}LY-Li+JofW6hc6g1Vc9UCV{D#n+0)cK1WFnqeqgi?bv67M%bv{Q
z!%7a|7$SDexFBCsp*k07$5e4QDlb9Tj8F6`x})jk`;UF~isH?>lnuj8pEU;ox}oJX
z$WBZL?uhI|YP2+enTa#-lfsYuh-|rz1}5sMx;)&7j=?Po=SS{iH)gc*j&F91tF{M|
zi_A^ZVg(39nssP;hvv$&>_psk9Yl_7N0F82n<R!2yJUSNNW|B#j|Ck8^VQF(dwtgm
zwyESUCvY&CN*A2h&3I?yeF}F16$w!kY$!p@fDQh9@Wn=ddcqr<ji+%{mL792W}`s+
zL=$T;G}6BDmvNnmITE?AJ}gB+d#QOc?4AY8gc*$2h>pj!8xxkRxiW^EBuJwZwWQPK
zQ7K(Ez70a@>Z7a;_|Fnv1dxs?Yhlu9c)ZjX5s7(TF-3Z*mO<*$RcR6Sx%`MqTJ}HQ
z6^jdm?==>GV015E`qRCQPMsP<-obva7A@PH-@8>S!XMjQQb1UzmShyHu3i2DQ*S&Q
zql3`#v(kZZP0a+~6Yl^eD*sMmB3nLS48yVScuxtO=ZVnIkOjv=sVhoT$?}Tr&6b8)
zkWQ6iGINv|MQVJ`Z3KQEY)z&2%-VgP0~kmTP4p6f%yQ&pvUAsE(=cgE_Zi{1SkV2K
zR1JRbAvkWifyfB%@rfM#Ei;3@05i0I6MggsR<3jYMSGd3ad9`MQ?2GeJ!?Saq{XK4
zDaSqg@B&1Q`^HixZxTrr*INg8%HSj<*<C<u0TszfZ9rM8&NBhzz8nY7pu=ZrTegQj
z)0B9ByN`m)%zIn>(x@7KjTdwW0X+zk7a|tG?DkuO2G@j|gBu^VJ_5}`N^jRK35^m9
zlK6x4o@Svr7E-O-Q4^hTr<l>Z#;Ww~^JtrEv+!R?$REBYq9VNWkPA3M6LLUX<+a@+
z@%M8@+x*T>V0l^VDYme^2UZ5Hn_0W6=Fu;I2i4ym<nYa!x{XuYh%G{=v+2Bvg=IVF
zN;sLBgnDQ&8$JB{Q+M`Jg-NV%J&Sh}^?o)&I<o$PQF)lm`VyD2e^Q2%%uDNtL0tXe
z(T#Sav+9&YQ1?EHM!CJMUCJ7RTs}(G4t8Y!OljOr6d*X&w=)$phs5~}VUnd65l_W`
zhc}?rzcnaJcXsA+^x;s*D6ssVFA4R`VC*o^%ie-R-F`aQ3pT-D|G?_jmEn`_H#so<
zXB5hAM@H-jmgtggog{+~Ii39ous(Z_W%gjuSTi#H8tCOx^5`!vyTj5OkWu*O;pA7a
zGYzWEJijc>f5bTGN~^Y7#5N<Ki^`vWUw(Z&&f-#uk8guKDKP6fj%|_1uEU1St(%2a
z_}mdO!CR2Cd6{|okdqNwWU#FK5vO;<cbA9{KD@@-Jy-=^M{~j$4n<MtTb5~@jDXJp
z$AsN*?gIpPM(J_!U<@zzTam}dDU-F*PQ2b$`EJb)V^RzAf@RWqFNMw&tY(&f#nQ(1
z8Aj=Wkg1po9GJ4lJI)^Nwkg^X)&pk>)uCRg@43mZ-0`S}xu^Pb`jTj{vPooRy(9Zl
z4Qav@>tW2y&$nV3#8%8&WJ>)wEoJmh5UHA(<&e3PW3ogLmg!B*{J{w=e2T<NWU~;u
zSeq1rWkv3I3c-4QUYW6K>Sh9e56@x4&+O-|x?&2&^%+Dwr_Y7xwN>G!#7k(mWr@8U
zSFyGeWz&KOZ09`dpB&#Z__Gg!&uvT{83n&K1b8E&a@4E`^|$H`U=z5J)SiC#F1v>-
zUqBE;kj46BS58wjcYG3j<SHs$4jI0HrhBbm4#~F1<OZ;TwzB;H+V;kO-b@&V+1j>k
z+t${$ZDZ?RwQajw+qP}nw)NgQ$@%vDhMCDA8IQfVc^$s6en>B;<8fswWFeKhousY}
zGFI%|M!Jm;I1)#TOyg~~?3R~81W$WsNLEGSHWA9ILnN~(Nc+;_GB@ZUz8_o(l;q;1
z#GyMUR=afFJ4u}mUi656oVTS5j-9}MlG^nAzf6HHlIkK3DL?vgMCxsym;W4vW_gcI
zjwW_W7F6H*rel8b<e-a0rI9*hb$4p(bu*-C2oXM_9HzvkFU(_4Rg7L|#(<r~ip5d1
z=|-1OK0jPUGhopBY9}#yv2DpBNhtai?<rXKrYyGJ1uh!%sTP2Lp;jIG%o4M#SlH9H
zW_yjtXo#QmMqNJbYcYbkF`lPd2t|^V(Q3FRjI79BhyP;>4Y_I8d!wDal69U%SjOlE
zOD=7-lR}@D`UdM@{kYD}E%0<Wk;%ma!ZsvR7U4-;okO=~?^?6*$o5)KLCx}2?)ad_
z6V28<Q;cQ5fK6zB;q|cmY)1wv`A1OQmnO}EbUEkh9c8|zpw9ek%|v=iI2Hj1Kh;f_
zfag)RdAiS8C2zW!PPCp(IWzsHfbIDo4A<jJe&v@gRv(pPCy1_3!r6vN{58&S1BM<&
zjeCTFl17_ILO9#%@^<h?GV3e2FV+x<_9UoSVz>m#%xW!vk|I?zrFR}kiz^y?9)Fmj
zAyH*L!~&s^3!G4L&X!uRa7LOm9B6F&Q%L8jr!(vn#}UXW9b2pprq)`6yp0r9&Q9pN
zr$n>5<oYNfpO>6Fv-l?ov1ojKGe}e1F<1hrwz8P*g%tCd#U7|KdtH%Fj?ktYEBuIr
z>hYf4G|0q%d-+tjwt3QklLwjm2Wg&!yL=rWj)1XO&Upq5pEDKs)>H*ISxvxb>ag={
zM{5#m^xHeAe>TtD8;;H9Ys-mQeA2f>i79I&Ho+L*ay3Hu3%h04fHkDJ!LTgxI1<|U
zBCdvL`O^2!MZt*W(1JUKHFu+ej3AK9+Ai9Rj>?CBuR=mc=FsKP5AJN}l*wMk$YnUR
z0DMJ;jm*p6!JyutFw23z9@b9YH70HfbrbbBG8tqJDx`huzkicL5(OdvTbzDHfb>{B
z`vx0J@%5!30|2wGz)4p29eEj*#TYK>jjl4BouV_mDUI#(RM@|?gGaI}&ALC!jsgjP
z7ys~oIC{Y%zO|?(T+H|$j}?}6j8Tptg;7wfk>()^pE=OWUuJ|5m3W5C+=`i8gvLDU
z`-uADthXorGNEw4DjGsP!CbBfqV1(*Dq;k2=wbc?C64X<l!u<H2m9j-`NIaJ!F6M;
zX!H08?y$+-5x2Ova+lv@P1?FBwtiN5%>FHZ6kXv|NrW!n>g%8^i4kxAq<qm;soqi%
z?YlK7Y?IE;EZ!*uOd<JZd&9+&7ldPOy41k$95a(aO0xoiCe)&3*}p%lGz2-B`o<Qv
z!LGF8*iJ`U?%>y=>A8iU(&zPu3Oaap@3pxK&lI0!wlP;p>HJ;Gy2(a)Ln5%+u;ZwI
zHnX~DEr~ugx#uhF`tNC>S9G3<$pxIaF;H?QunW{Txz;4w0i_ftRZyWZ8l1cZ+ViSS
zoldF!dkNc@LcrvBmleU5@Kt<CDKxM+mT(#jYRotN6w{0*nxp-nhn;o^5P5_wK@;Dr
zJ#DtG&osgozL_)7tcJM8M1?w_sFgo|#gb(+)K|6P=<y#++v(5Lik}B`gEP;PeI;H{
zZcD?I%Aecz9M8tvO2p@YkrjkG^m;dFNl_P%d`kLf{u9L$VDz?E7sYtO#F5fiWoob0
z3^+22jsQK$@4xuBnX5dz39bV=WOOe7n2YiZ%j(Z+Z;g6aq{O`$+_o1$>uIHbwLfW<
zJNPs%Ht?w-p~^_Qm`R!y66b)6DepxscJg22gVIs{ei)LAa$?0aimfwY1pNL=fz>D4
zE?FwmD5jZ@x9s`s&WU%5!9o94>+u8$eaqh|dZ4_*B3X?vVS&rG)%eu*;tm@m<p-l3
z&s3|4BbSA6pja#+qXhlkZwV8B{6=#jmBUbwED8+n5KBjOH=55Zdn%cYHR%{<L<P?k
zAZPrvT0U|9r%U(fGQYV5Yg2gmj+9llu90FWUT?&V`T>XNq*Qm1h#8mfW%0!{JMvcv
zIxWhou6!yuf(%OGE{e#y$HsI%<_!EqCtlQHu0~zq+sqCQwQBR3R{k`9pmq$Y%?L;K
z6m6M7l!s3t|Bi@5;+b#z3Om_@RAJyySahL(`SJ@CKc@nsx(84|S6YgVbEJn1@$kay
zmkNIGun<TG`W259MOm+Ow)DEDHgs{eGe%SDZbBMoK2o;=J-BpMmRnUdItwo=2%ZIm
zwH1Gz`HfigGaCtBmt2c~&YIKd-p6owBJXpaWlnz}gG{#EN820ZTR9#zt$F<Cl7?^+
z+qqTDi@2oA`4GfyNg>^(GU_d8Wk!oUKJPzl<B@(Bu<6REwKxi*MBZ(VL!=!+2ZD4A
zbDVc$=1qzw@U5e5;~5B3tzRyRsZcad3!r|iq@%qEOZs^2Aw_L}%nTcRA$Ys#sC~(s
zdRQZmwL{;#H_0VO*ktV_DTH1GgDm20eI(jC{cIi+mgHHztN8*8@>Y|P|AH&@5K0cT
zS@JI2(k<AnPqOWPS<K3UooI$mII%mN<OBuiOS?-D`WvTqzKe*&$+}qA+y~b!)wX~r
z%Kk&L#bmGTHdGpaKewcxAUMhqfv>sW=VtpVigJYIDwo%_B&DN{t~om!F2-w8_-!^v
zXU*_K!jl0$#q3^YObeo+fOF;GnQSx!TUJ99nILb>xrM(Y$`blBDB)9{Y!QXL=G~88
zH;US806H<Jq1}D2=DM+P*O~Eo(VXv2GfkH%C{w;v9J5Y;T)A~|_<UgaZyE&~z}5iU
zj|pvn0U;n+l#wNoO$H9PhnMB6Tf<Ykeu;c1=XhXwztpG|IZKto5?X@Us9>Z-$LyDg
zRF)@4S3QN3-1DCOdF^t!3co)#2<UVay7Nu;q7Fc(P&r2e6J)PFm?$YHU3jP|Q;g|(
zxQ7x*_02_pUxD8x;a$Wq^-6V|aSch})^Dte)*fg%nY!rFYv4jsy(x?g&lFXk95>lc
zA%_|Q#@O1efUJ1JWzC^(QNqqx5R&SH*Qc-(bpM<G0~X4_G-69bNUI8tAgJr-pevNK
z(hyUA?+57-JG&-5oS}fomAzFQR-P18idIgavzn}bqRG{=z<3KC?Hw*%p-!deV_GXg
z0lI=N?1>A?S%PnWP7wB<y1PFyM)%D_tSJl$P@E(Ake(^=;WBVR8}pcNpF^U$aegX^
zPJ~DNUHP>5DvrdIkjht->DBo+oNcq$&@0&Gu5TR<+KGMc1#%)7#dLX-H-Q`FE&kbn
z2m3#No>&}<eME}TygH-cu3PIZ`P5!JuIkz7Vjnu!*lsWc!YBeuoyWX0*A%5RTXLDd
z$|cotxXL{!3bjw7edV4@aeqUF9!oRJ_1VF=fo6HfR*IyInJVr{$+4EinSrTJE(m+m
z@FtGlWHE$&jE^HQK(IG3Iw!%>NCI|)&BeceD9Q%}R~Ux<E!lccV?|uo08;o-HMe(Z
zSb6Jrck6^$`}fJ4DLHJtxb0O|dq!`IxAF0bVvpsyLS-}=o#Ay@i$(7~CvHWH6olvu
z9g?}@$DLx;pKhAl3D2BscV|>DPj%5D=^9;!G6}~xrqfn=>d>Sq9N~&03zD<5o_s)m
z5{D{@n1Jr>ELXH_17(rLNpe1pGTAZCRU6u3Onj&EG1vy@4&VZ>$jn298h=n^%D+{U
zJR#A0x#D4MP{Q7-C!qiKg1R>|$sVNxvgVS5>0y{Rm&lS9b|Z!CRQD8f(m}OO_aqJy
z0`KnTUzfpkL~QWR>vvCb@R&Q)XrLQ^K``F`!4qnp%0aUh^?5R3TU;qnJkC^*{P7>r
z1Jy{*1rZT#gIFI)xxNzf+zz#%0;(Lnc8+u}bV|!@4S{I_fj{ynaPi^@gPSCqnx-mK
z^ku(G{LpL&a|c{koGR8sPK^#6z}3@jHT+`3DPvK<EGr`YzwwuED0Z3Wte^3JGeiVi
zv^1Qhc!#z<<Ke<9gTA`RIXYJWDrv~0@l}jJla&~F{>1O@E}@~x^N;fCjb2u=vfQg>
zbpb2)ZYO))7E%^S83aXn)oO_SjGUR(XKb-5vua*T@wJF6VMgj|C&4IM;Umm2i&W4O
zdMFZA#dZiL^5@4inB$}3HocgCn?!sy&S+9M)mKTIDVgim?!YRLG-7%jmWrQA&WwVv
zDKwY0jsYE2c|z+YM@a_Oeo}9zlu9M>_4kBkXREuG+@dHKlYu<&mhW>Tru_62Z<%(8
zCsWf2jMu6d7_GX7!~pUm6sifF-{}@J2p9M8wTgBuDy+ofI6!W%7wr~*@@900^b4wY
zE-v36ehNHs%@dl6X+}hC;``E35!qAPG!C9`?fE!gU{h^pIwS*+3OkXw<W~Tqb8#zs
z!**Ygfjs6fJ{t?-ny?>wj`5oX;bKv+6;FkaLJ3!BZKx##tXbGPdd>_HMc+M9a8BI}
z*ajcs0UVz}2P4Gh6T9Vqac+l*NU&iQ6=!JQQQwnZ;n-;Gp+@)Fk5U5Ab_=KTDj!EU
zY6&<nx$gE-jx|YB_@C!r6Y%l1(Dj>aXc49RAj_)<m&wp`#i=OOJ}6V1<;iAWnfM;#
zO)?4n&+457)~#E_L>e4cvm_^u`N(deldDsY89JV?qSh{KSr65JVKL-;g)^wNW%NET
z*8kjsPBd>TPOD<9Vu@2p*n|~GH3uP~uU)QYj?Dx76)@8Np_L@iMEy4cU1Oy+KkB+L
zhJBSKaA|vwlxp*lPBprbzY*SN@jCo|Iw;|%+cjpqHm5=75c`t5Ko^m`CzA9{xH%qS
z=JaRV>%KaoaA=!<xU*i=-m1HKL4J9~<-nRe0AFSz$4cO>@-&Dn1>yut=Y={eo*YB0
zZ^by+z%Za$&`H4oXAAZZ^|5Ov(zQTX5l7`YWe5(A!85AvWh07&JkL+S<3Iex;QWcm
z=7EZg?aLX|iyHRy`-V(%diT@JMehe#uwLZL_0Mc=mh*jopd7n{k1dGD*{lNu3jlh3
zu@p*tZ@xMpsYDHKeTV7~wXB#G5ZO2C^vNP>zFF7_OP=lEukO}GViI=h$<D|6GJA6*
zT4r(H3DK=x2R5;rK<Hc15R0uW1on6+Jrkicc&!MTT~}uupUG9KS0hhUK{)`>Yu+0F
zX#SAfE$%pfis)h3x_WqjhI4z>%*-|S^y7eCCbi5VG<=-a!!4&)_VEk-kv!tD4$<Dr
zk5Yy;cDWvZT0p~9mbJT`fi1B%F<}5dEx*g-TwYu*#c4AJKAPaxitxu)QT|P=zshWR
zICQmwpS;OSa%k;E(H&K_@o~SFL)F)eI=|cPb;WUi5>WKa8F6dmY3KiM+XWWIlS12S
zun7uhYweMaga%}n<3iWSaHNn1YD)hY-(=6+^rzz|w&+%hE(6encyQ-opO&09OT%i^
z=*dKsapo3R0o?4?=sUk&x1@4dDsWn)a9d(FMhQv1(!REZM)^s)a%$4;PLsa*l$6|S
z$GHQ4-5&FVl9vnXQPgfAGfX@e+o|av>bdjx{1^dq@ZzXW{Q637xe;Jr0B_s7sv6)A
z`jKd_C!FX87Sr0J?|SX{Sn7yZ1KdGbHNl5Z4ckM<WlL#7^#gdOD*wq(LnF(~^kIqd
zm^bO_9>UaYLSYC$mRw*!FNE?o`nEqJr}J`uv7TfT;ksTA`C2>|A3(4P(-Q;|CBi2_
zM)-8<I-s#KDc>aq3+6?9l*YleQ~()c@kY$PQsD|1y#X(9?P(IS>|W%=LSlIz0v#^L
zIY7iI)chd%RWF`RIN1tqyY6(MZ>5o*dT}+hFyKcM&db!_vTSC?b;=&vCI1H-UTkK6
zahsn({VMgC{C&mDj)!$O&VxJcYa}x%{A%(AAGya`p^qRCs4_Md<$PG_uC%^*=NDR}
zKlk$Y9@}%&RbKVD4l10zMY_40+Su#C6d=GG%l_Ak^q!?EG(PWp5x<EjmRT)3%E%K+
zL)FJ&vQgX92-IRm2zDbNA&lXddjjEqpFITHPie$Y&W=W&<^MW%iO)2h2}0>=kk#<z
z(nga}DF3WYKy5gFE%H~EPdh%~@a$KYu-N+{wMG@#dRA2k7oAOz8GUQ+<}(8dNb%fm
zCp1V{E=B0RE2qMK60+2c+m~Y!C5f4iXdZEB-ou)20nKZp+8cs6x@rvAu907VetJy$
z%ykcg@{U?KIB4tkT}4+&QKH#SBRxvx<8s+=chUB>8S3}NUg_xS>3GJ9Ro(poL}1DY
zLL9oG)c{x!)wsY2Psuv<GILMo#41P9sc9v5Q=<^$jc(tz(<lEqW%Ph}DoN8E@6aU|
z!-@K!bP&7Vvclq~(PfOmH47Pkn|i8xRz9Aw5sbydd<>or)zK0hX!*L~V7_ei-#kvd
zz#oB7^Fr6wAXf<672a2zCRZc5Oc6B18dNWUWFyNGW1SN&O6);iftYZ6*QN759_nK3
zC$4KAn``?q?=v&1wf=5GVygB;`ko2KDBofswbWHp@KG*bO{Y4)Yv^5n+AM`1>3KXx
zzyD;&H$A?h$)yY}eTXWB+4tk+eK0ye)q{+>kbnmK?PR1~UO9_YnFzr)19LmNpYe&y
zZ@MLNVOW|K)qoqUM{ZG?q5pY!)Dfjuo`MyOFW47(b;ISs@bY5>GN#x-aMubnB1d{|
zZs^;GZ@3F{|5dsDz->@}D{tLVgq#dBhNw34l)v!ufsfX71a<rck8rsNmJELWGr)aU
zJtd;82Vpd_RnKJ1gz<{vPsj-$T63+r4k5JDw9`&aK(i?OLltDkBMoeAUvaV3OlE6D
z)!dIZqghyoHDqYxOav$zzSfxP5mzSRrXbVrE-^8u?YblOKCCByQb37S*`^#KrC50X
z%o+Sye3B<aB7WW0-x##oRd^{xqGG0EKlRC)$=PW}^)8%Z{amXozUhj3A1qfllw()f
zdY!G!by+(JuwHFo*?+N0%)p=HDVac+pL$i{urph3O8M>G3h19~&}Qpom0Frv2Rm)(
zJ%w@ROP`@dqIwm7qy0;~Bz7t-0X-Xp2b2Bm(#qAlx+Swzy3HLBLj{D;=?BdN_k>y-
zA1vDjn0!sf@DWCLdOATf2#tf!?hge<;R|xWQ7V4lRjG8gWV<>M8~w^VtmUIpu)Dl=
zZ1*F?b?44#$GY;hpP!3C(}kidHIj?KRJ)?4OU6wsf~-w{NV~f7<j+yI+?=mPYZ&^w
zvaXXY##93T;Nq<8?su~|ev-6DuMN5e#5Bo~1E_5^^}w>;>-tTo-ZE%V!ak3H?I)&p
zPV1bpvO@KIc+;GNk3D+1t=^_@WJ%B)5Iz8#r|w!6_;q;jYzXhwQ+Axi_~tKJ8_hi0
z6<%~CElv`Dx1ng;#dFAqL2ZBUPD4rGSHvvtF~vq`h)-aQ^3gzAtENIXF`4i0$q+pW
zcV&Qoyk6+lwPY5t#J6K#eDTB6AZt|mj@n)auOHup63dx5=;sot$ADQ2kOZXFlt6`z
zG?VZU`H9?pm>ijQJpJydCK++XV(gQljnNwP+?hsya7)-fl*dpo8Yflbffo*%8wc?^
z&fWFp40u2}B*!_1|4hSq+91OzY)uH_27GBgUa-#w6X5AxP`~zLC3@iF^ML%0fv=F-
z%jTvdG0ucloWlr-IJZ3pCn<ooEddQG&oG6j-E?&?#s1JTy@w$^VbtyuPsYWAu~Hf$
zb$Pdc@A5>LM#r0SSH3*V{yK`PY%ce<G|QW2m$IG=COw5*LAGWn2V@yMkrP9EyoBcR
z;ZqbfO7;*JE1|MAx5;c$Xpd45_q`(+y5L3++?FXJ*9Gl+l{_FSdgSm*o{_D&felzm
zZ(~VZ2BMySU}&%Q#r2+?Qj5a7Lrc#1?SRUEO^LYzfmHVtyk1~WO@^&&_~i(^V&M`!
zvspy26?4!xz{2@Q+Wq!&vNVNT75kUdED;FcsdW~|YE!7{iHN@N;$d2pncu%RHF%p&
z@{1t_(&E?XgvN`5L%1Y?1SE|$LYj`(Y26io=Nc=#7G}Ls4ZiSl#h|lL!C-%KshJ*s
z({X!_dXX)E1)R4Jeo$nGsKjf-2WrMleRg<WMulasG9EN5=HdN{{Mbli-DtT}$2r^0
zQhQ&0qU&_cdK^EOT*L$RB#FV9xP$9Z`MD>R5gEM22&+>nQ8Y}v3z%3sy_;nZwzO87
z_FrJblNl$YD98Hs*jnA(iytO3NvL*z+$CZ&XA@YzN(-@$YZ5*jA+-}}N;OFC4we9;
zuuVKfAg@ArYyr*Ghlj3~m!6DQ6%ZBW{2q>cVu6+@OMamqsVtr!73x4k6!HHUfg7<g
z>ouTJCN>;5k^lM}9YC;lg7a#D!1~xhEooYE&4`8L4l>!tsSJCy>Ai~Nza4OYR~j@~
zlwDVg+51}Oq?L$KV|POWJxqrG2oDybFFd8JKQvUrMd84mA@5BvDCb|cKV`%SeR~7p
z8QOkmEl7^62ew0f|Joa|7r_(4O#tA?&bT{hp|5~DQ2I?2%byR>FC|r)KM~#{)YoOJ
za#+8l&h$B>sQ%!e+<UIDdtxnrD9J=*tau10ury+$jQ%!YE%3k-7QRb7HDMnP_};tY
z8Skw@YvY7jLwFlfjichE3ShOc(lqL^?~y^)!-MTceH69utPpd%@V7c}^9r4mvGkCz
zOwIQB#tf>cJ<rbduNLC-mM(JSX37ud@<VeT6Tf*lc8;(MX$nm=T4A?;r=Y}!_N!7H
zeMnuVk{oFQ|H4ow>)E{UPZe1B$oL`^P;*|aH^f24?pu$Yi6ONEoVzk{#g0}_*>b#r
zA1NU}Y<C%MGgLLd>}$Ly3s;;h4c|-~-SGnFv(Er9F|*qg4{Gr<V<VDmU(-B~<${&l
z!^h{S2MWn@cc(anJQ#+5x!#8aZ79Y|7&hMpv;)RF*}EfXt6R9kR!Et#!?h2(8%w-b
zUN!SC9qXppfGwMJj=%<Jnl<No%{$%v+`ea+(k|#n2roc#w|6uS{An&b%NxG?L;>j9
zZ~z1TJmq@ogz5?=KHM;79?GNt7#7Yqj4e)=$UR~5%b|DJ#h#dd-^NZIq-jx1rAVTC
ztFLPI9cV#{x!@-c3j5Cx3E5sO6VtGh&%*QDo7l98g!|Py1Swr7JV0PH>L2qn``5OS
zm8q^4wpK7UV}|q!A1dtV9W{=X(mheKf<-2y^4or}76>U@to|}DekB{0u4O)NWz^wv
zZeu=am<7i)jusk!{xTNF>1xgi?WE(w%(J4QJaI*dYI*M!`gBj7_3W)lV7?(zXMcg7
z1+#5Fv8@T_c()@Y=!TdYS~XGUn_JQTDPJ>PoM72DSpb3gM&?dK_e9C$eZZXL16N><
z2u?m|gQ5D=^+R&6qDb}hU0p3B`)LD&4Fk-`!485!eDbk>f!VhE83k{uyBI}mBp$e(
zf1}aMz}mPEJ&pTb0$6kL6fD%v-F<@6PCai=DpV6w$RZG@bHx4GE$-VAkU+{Ca}7SA
zkQV)9V{Ej4tsYTzk*)tT`gCNrSjCcB*((gg?YDbU6*JVn&sTLU(_iINmPi$K!oWYl
z9XF65j)C5PF50@P2}T9mYxbJY(N}9ZYRDpYe+_@|JiK(hQ?%0F#^e~Zmru^<MF8&P
z7wDaXI9EZd1npTA^&Y{QAb>*5^6<+CTlnoZ7}Lt#N0C8pJ{h12pIJ0Ub0o{dIg4P}
zlW2ktma7kKAO1~8w#rJ_Y(~^y#)GG&UZD4(sT8<>LaHn8jXxT8<jq_wa~rrt>XnKv
z!d!Lq*xf8pW9}mYckGTikaq(k&xEt+1Qb;2y%ktpTCb-t*>ZC?f3~iaU%JzF-PajJ
zmi>g5>IH(J6NM<7uE)ba3xP`9CvBL03NRKZa=FO8WeaaI%s*~Iv#5d&yxn*OpRGO%
z`v7BqpY4QYZpZ&Gi_wBgDv!8ajZjm!z|Ht^1s>&`F;5-}1YhlUY+)IupBhr>5{AOr
zeqZ+a3ZkJu29Vlmx<GK$b*3}ARXs98ztwo{NU6@%@B5RMtZn^9t6~zcdhVCm0D0b!
zw^_sKO33;s%B&&(dDM%DSeOv&*|CYr&r?Eww_gQQv7~KON(5i_u8NI8Vjl$!QAIWB
zLK11PW$w)tdPZ=)U$$ieFw*3Gr(S^x)I+lN*^Y&h7X=9Lac(a-O@sj{r!Kk^=hV&}
zVceYI?0K$zVt&`-Ht~uW$2@UHt7O#}`>Qq&;*&PR(XzYNffWsKZGA0Pr<$s1Rh58$
zI{;Qv=A#wPRDqE?_6RKgJ!)^RYq7Q`<}|JO8I|_|(3p@j$H?!N!E!!a3-Y)+!*d8t
zfVKVE%X{&Q-dK?3Eh0=vi%h$E3BcVRrV*+>I|A*@oW1=m_-)D}o-wa=7dy-^@IVyC
zz|1c{Hxa>_T%ZB9F6}+zlACd-`Fy&6x5(<;^mTE~Xh{lfhY$?*NXv?9B&^YKRY#y-
z3QX8T3(ww!>}~zKdXx*-&QABXM}g*8M*5nyl2Rw$0U<s9n?<~(P6`WpyPv8i8K1YU
zN#&T0Ent*{t)?GsPOlZ(1`~I)$sJhJnr^*xaI^j~=#Au;*P->3S^sFcI#2O`r9;VH
zj|;0{%TI7FZJ{kZwIfEge^mmEa+&WSxtCH9=sC|sd>i!z*<6ikWbjyZbqQ)qBnh%B
zh)4f3Zjz0WkM!1$?;_w)=c+C*7;+ib@sa!)$c0b!PDZl6SYEhOOEN@+==0yCTp1h_
zGjmxCtM6z}uOY7-Mrouk4LWjvAmEon-}+{}wex7x_{llV_mMQjI@A5Hsk;{1<K^dh
zgaG&U?zhN&gDyWkZ5p#aUxlvDxOSO6y<b}HIH~_7NWWuO5%U=`#gqm#;#0Mr$#aqA
zN5}r)v35)k?Devv5>K>VH-op{f4yWjuX<&B>nl<JjDq1LHRl&kd@tgE3mje&c&FT^
zZm45ZOVZRh5nN4(CMt5K^=5fr&R4!D_3KDcM|o_Q2bqlnE)J^wkX12y#B=L<RD0)B
z&~2u<a0JadU=OEXs}#9H*KgFPkB@je4T;`)zmEaO%xBcWz8J26I2;5bIzi<{sXu$E
zatN^zFJOG|<Q(xD6C`MVe~Kj`o7Zv>k4(*ZUJK;3LjCU+0kflEp@QC%hh-|LaLDF2
z>3ac2;7M}>hVsuONROgkVYk5Sij^I7zlV{5DS;IV6AS|um;5Aiv<)SN&fa|z8tebS
zMRt0RvQDNRFrQO{8A`nVvWy}XsIqYbuTvm)uDYnobcs!j8M``vD~Wuu!PTQi3_^xR
z_(_$!o`cvPmppZzR*rVJ9Rz`JKc>~2)6V}TMcC){lUmeAjw5t=zI^Yu#>pIki(~<R
z&}yU7Z;_GsP;+>x23s<tKJUuJ`b&mpqZ@~~HCijZ&Y`)4F=o#s8SF{^kQCeEJB;}R
z3TZPI;b$GA!x1!p-pg$tT|!+$_W0Skp6oqb2K2sR<6RqyXTi4b=={3T5`c?D`CVrF
zVh_rQ3oj*sw{581`$tX0G*a%<Y_(b*4yvS4t@j6_yFc0b95IgMnVq+TvHlV>1sOfH
z`W!aQeqoJM*iIm18vY->Fehj0OPrcU$e%>CY$fABli}Zg$sM~X&Sc|%$El69jq8*4
zg&XaUJ|itBjVY#!K{28lf5U$fZiD9Ne1+s;D8qp9S>I*sD6rLnEpk?K>#S6+Ok|^$
zOaBRGjM^0nHlFnue_mpr^)jJSb&JW(*WGR;=(tO^oCt5=Oa51qEQ*j2SJOR~lj(C;
zK!Tqf?E;H`G6;H^-|B@JUFI0p$c4C2L0)}wGE#Yq&CJ9yKm%Q1{w~pqSO&kKkVnJL
z^W5?b)(-XwSTE~m(VlVGOJtW=p&QR_6T2Ij5$nd^8Q2x(l2_cCE)|0nD^q2r--A-0
zbL<M@`IpQxkpnG0io}HTV^~r*kmg>u^(XwGUzG@dxbzMFS%5*hB);O2z+#$P2ZvSR
zx{(>k;2R~QA5hNKYZn6Sz(WFq;sn?RxKwrO8_eCYzLOm{Mct~+M&{SrJX^JR`AYV)
zBWPZgP2a2`f`w#(eqB%fRn+{OFSU_lDo0A501AwUm3KX&gymr6ki9lfm-hG>uWM=H
zpY%(Ah@$5J|2pLF>3}EwLhJ3IfO!8XIlhQUZ8&6*!0=p};7UV5ZeE_47=1&f5tMnM
zo23u^jCvzA?Dg({^$pIWyXc34C^U5zJ)Br6<_OP0WChseY10aTD#&d!&J0j<aZPk@
zV`*-s@Zb^o(0k4W@H{gbvYGHBL%hsvE~VptJe^T0>QGjrw4+0V#5zlrluZeUSm0Xk
zXsc16;U;;H7uh-eQ6};juS+{Y<Zc^PGM3@LVZMD)_|b?MX)n3scg0qP6q^p~5sAL;
zO;zJK-n|k-2#Nrp<)17l5HRp=z>Q+%$$Cx5V;4K%c;lNbv4QNO4qqh}(tIdw#3S2(
zyD&6fK3Gxn8uY?&20{Gv&m`2{nX*VTS|a`+-_c{?nUk6=wBK@0OyGk`GZCv_@Lu0S
z-^vfk?fsMyGB-I(J$?T!^CTCXrH`bjD1~#W1tHYnCqK-rWS2<t0QDtJN-jKiBg};c
zXz`n<51Fp$HLL@p;!Z6m?M!U;lH=ZgAiECk5$VJBfRg1=u&w45{ghFPz4?Wyu+g0+
z0(J`7Bcx^09(`W@=Dpo^IGXFvxNsww)vyY3*6ip<1`&M(`Z$!0hA3Xn!KRNCSe&Q^
z#I*R0Ke%kfS2GkI0ulX{UQxBG(s3h;qE(%TK-^oL2+W4?7-JD6-~Vk^abFXEYkpbh
zj4Ou0;S#gokD~Q#m!&_igx;To>j=p=T|Ki+DRQGItLcoMFyq|DZcUMN=CxmgDc<1D
z1FS?-F9<3j_ETnDRs*t^QQ;=4zWI`kRj<<#un$llQd}W07kTg8`J!^>!07bPkeS-)
z3Ig9Gadxb%t6jTX_eWZEel=l#$Er?F)9ALTkEsN}mG_<M3+ipM7~S@9&yp8&a{`+u
zeuHnr-sLYd!d=-FMOfVXUW~!dGQF+hKpra3Zg^BK={LEw9t`1+Qz=JZ{C)hO@ka4%
zs_mn_k}XNR&Yh49yp<f|Suw?U#MD|ZQ94R=B+g(a51HH_VulX`s8hOsE#0&oY$iyj
za&MC>8kdpNv3++Da#-XFiJgaq@x8&$B;a<<&=xjS@J#>WV?MiK+iq2u!1znCTCXp9
z{QtbMscD;cx09bTdd0<Q^lmy=(D4_m4Bkb?3pbb}FeE^k_lnYt9WBCpqJ8c!stE`k
z*rIA8goW;P&G%cSj`(nY4LU$+*VaaDAl=?^1Y9iIo6I#!mXHOV(Lwf^iHf3z_V6hY
z{-Lo9_d;cnIGz-`2$XxabJyWh@+A8?cAT^`IZ*G-)<G`r91p*>f=#9kNV*u(Sdl<4
zto)XYvUZ9M@x-pG^m}fcf&+_Um@nI#<uR<02hACfGvsFaAk6fC%VPmh!qQE8fO5Dl
zuE@PvvJZE1KJq1PCA-M})?t04LkSxj#H*p8T2=8cZM<6X)0yuda!WvrLn<g7su(`}
z-A10ip_&)<ZJa7&4-D>Bgv>>1)@jV0hqBV>!~5vQ!*o3m%xEkNt0fz;9<Mj$vOLO&
zSmmIC3#E8k{9~klhH0iJTQUu3H3EB&V7Qj^JlKSWtZGAc18$yEV{@RQpQ&5!4d{S5
zKKK{F6|%zgP>{ary;}?fb~=)fO`AU4r-vUDztt}=a*a0QmvWnsNiO?5djfnlm0-LA
zjzM$n<T=!f4}5ldwNM{-Fggl+V6*1g6LpwOR5GCcS<9z?;gcw&Ms|dX81ez`+CY_`
zgg2v>1^6A6TVEtKUJRU8ak~fbt%%GW`3xI?c64|26kYt&TjjdJg~q>o?nPGaKqK8Z
zMhscUHUNM37w+ewNe?9)p7z`{4es0KOj+HjKazaRAtT_xTEFhRquf8ITVeTXC%a$F
zLAHP`gEC)#G``cfK}LK@OTWnio^uQF;Jl6Ox#a=JA?Tvxdaih_rrTPvK5IW%c_ebT
z43=UVV6O4k!>K7j2->D;WPnevg55C--pc)s!y^wvI1UoFNZV5ZDohc@*$n#d<yMs6
ztVk_}`qvd13*Uz&$R0HXx!whB?|u5)`{$*BvI-G@d=8v^zgi-EG;3|$Y{Y^6ZK$Vm
zQE3JyY*$yqo|EHffKq?nRW9W~-3Ll*n|y^o7+gfIDyUzyq{I6)ic#JxuN`5C8!j9Z
z*JY76Va&0zSf#ZNE{a$pZ%o|9;laV{p*_$Q=+S>6*A$m&sHOo;eA*tKHGqwr=SKKt
z11Q>mA=w<%4!j+`<ING#ppB%2KgZ0<qYO8`Bh`o=GANPwc;B<k<XphhF6)I6sbap{
z0GQyuUfBJT$szrDskWEAQ&Xw4$M49Eq4mrIyK~c?17S5hZjMSI9+R_!82)R2o-1wx
z75BY%)$%|L@?pf0ey4W5?`g&$2=Sp9P7(coOpWd2q<C8N{J{*CF>CC5bpzJYzwm@A
z(#(&Mmre<(x6(5#&mVsnn^Ge1S{<_FYA2Y6cZL>)-uP?CVhZ*^6Os=WNxZ4#-;cI_
zw6~-#K+?UoflG3XJU^64T@^$(pFYk5!(CCW9nPvo{E+Y`cQGw&O>IL!6saMTf({CQ
z|Mb%2h+uY)y=FQ*hp3fSAx$gkB&KNpN<D>q_l55G5)y$LV6&*8G_b-~GY3A9fW&J{
zFN?<Q?}!Y66=M`wQ>I;jcD^U#Vil@O^?t7g%k+@=333^4sD>%u0+-A#XJp{t&YPOo
zuN)rX*#()UqiM~{GUJQhK}PHW1I!D5w$AwIy=)0TGks=jAB;*Hra6L(MGp?ZC4y{h
zWM_hEzPGhRKG%-L3GRTCLEF^~`BO!8$g*nk)|g#aN5p^ayGSP!LHH>g7pbsAW22AM
z0?D34_wy2MBV@m;9lRAi+6BcVp5hc9Dy>b=g|3-2wm;fF03Bek<$;8w^ZEsUom#s*
zv0Xg?fs0qzEQde`$)9q?$aKRS9zeiM2MWwRp@}_{&}84?4??UPRlHkp`sE;+*zwJy
zGCh~l>UDM5+eSTLZVBEoax6mOgTXJVR!{tMkzi{zhkpK<w)=D4Bmp%=71bg~G_t$|
zu;s*koL18Bf@_#chS7O#zp@m6iDe)b6bBT}E;K!t;Jk>}qFXt_YOX-NXHpWmscT_O
z`1;+guA+i+2&n%uee<DBa`%iJjVjTw;$ro_B@1tnVA}_%7}>gFkpEo0io5(SAlUxk
z)zbD=wM8XK9q)cCY46}=N3Y2$kb<4DpAAD5Q%V7G#*0tQnty3t1r{KG811)=lHikh
zJLPLXJyJLJVh@c)EsaDLMU{T%PbZ_?X@@UEhA;XX1ha%QP$}A$mpCH#H6aj&i$aRp
zlL?rpaS2aEB=&%BzjKM+CUDkS8T!#bPTq!h>hvJcR>iFifWWIddYjm&eQmw!t*ag$
zM*d3(#rBk1kNasP>e3f~S~*(n;}MIuAkha}+DM6lGs_7NSSG80iP3^;yASy}`L>*x
z2BC;swzM3bjJgwSc^we>nnNa!aYe=akcp1z%91MIExnbP&%a6`BIt#kRaBf&ldjW1
zXrQ4XxJz(%4I13tAxN+eF2UhLg1fs*aMxfVxCeKFy9bwWnE7Ys_FV3&de^?&wN}-x
zdR$SGA$fXFsN^?&GdFrYOQAa4L&N=%laUo#16~?AB*WW1hPPmU{<UN)htC=(2}b~w
zp4umioaGOCTL1VmKO~3%g&Zo_`m_#$jn`SXr|ikQW>?;JQpa~)2q~?lP_jYAM%{?k
z67Zg*VtPPs{FSl;FJXCU(>0Yug0s;a{`LQyON`vvS<+AY$6|-5<F!}E+iWcRm-Z;m
zKIWDzFep+?cvW1F-GG<Y;V;F@^a1X8d7dAcP{cK@6*dBl;!2k_kFk^y3zDXdxdNo#
z@Q=oDxO5c~dWb^#`tU~4mX?BepyI_Q_C@GdZ^Kp_-TRFg$Rvfxw(H#uCC}y(45qXy
z(=Uz{E)TL;s7_72>cyXu@ethVz7@;Ixzl$CbUReV6!R};Y~0UUOE~Q70CfpJyUo9E
zRwVXn;>@PWUE2rQLo)zn5Dcqu6Sn*ADOcXF1Z4Xu6ze{_0L3S%{o5C)l$n0c-|OYP
zwY=!cwaT`560+%$wER*`5k0J&fJX{sCq+6$6D48FZ*;fG%gTRmXR<!qcD#c#8lqQQ
zXnt>-mO!*X72!06Z_0}b0|aaK6-t)^NT|M=3g>7jFYro^vpAmkS-eX}`R&#x9O4uG
zZvwI~^_EP`eGu=fTKRX5XKXV{btMCl6E}g|jH}{I<KP$5mq#D2L4Kje1<?2|&*(W5
zG1g!Mj_)vk9)rWQ3`0B8)X}d2l}L%|!DvEFlad~2T9jq&BBjPJe}K4c#z!6-QO}6h
z6VxoXw}>s8HExaJuYu)&3Zv)N=`ZGX%P8kjSZsfa9gM4V#Vsy=lCl&B5o6r3kAhN<
zPb!L!DGBzv{&smPHv2iu4E?&4EM(oLN=p7rI`x+{lBD@>H?eD(wRkAWW(f12z@XXd
z-zz6SL!@Qmq+t9e&H$I)b)cH~g9#^LaL(25nXr^Pxb~5$G_vmJuz?Sg&GxxXJuqwn
zDo&b@G*inzuE*!&B3>BD#9^;MOw_29s!^gkSB0S({<>33g!{YIp5|_?Rg0EL`{B?t
z2?XX2SweNyLlmOEh8wUiSjfDc#9v{aUv6H%pAio2i^LcTGC+A?N0FWn$=N{)`9KRl
zzd0)tkd7;z1Axt|DGyi0Xf-FXC^FNH1h96X=hv8JR^z^(`fyyIF4I5CXs4I@e*8$a
zaPswrT3?et>CL`y$5D%p_>aTL%RgDEUcw@>yxjdK7G4Y{{i*R+GwCgh|2~WKlb$xX
zsm*QUjmpe^<e)afz>b@+6uygm>@PgvsQ+*Q&CW|@v*vz~jn+lqaeD$5q=q<CYMnq>
za2!1UP5EnnMg<L+!*THaH~+7B$qCen24ee<tI`i-@xNriX1aV(9oqk(0h`aNL5L9`
zp8wL|G($eXZOg!M@-@Fw1M{6<#s5VW{9bYkXY@Z*!R8JvcwxR*?SDC%F=F7qVf;sn
z+gzRlF9CsP=i+OA^?4)z@68-T@TQLct1cBlctT1!D}$#;{XZZfIX6d&az2OwI5Aju
zBI1wtnb)4#QgCTeX@Msq!J^RbjGjG7KDKKn@y#uX@8#}OOGvaGn?D;u2xnR|i@!V`
z?+I4KsF2Hz1=F&=kr~QoaLF?_L4nzrKy1z4p_IG4gB&qc;Z~spF=i$OaY(8(sZ@nd
zk;O)qqu@Yv4@9$MN++{^wMl;ms$%6HKE7@LpergtXbVw)MO_-LJ5G5HK7JH{K~Yb^
zr6_@}?|k%K4Bti~Pb7u+e{-uwDTfplGNS&_a_>Y7nu5+N=Gjm~<l$W&er`DbRH^L2
z5NFEl!BEW&o0G!hq9Qrh6@x`Cn+*1g8~Pcu+Z|m%De&^blfcj%Az@KKLfYjJM2OA6
zMnyS^uE7)jfb1TkQi4*zjuexZo{NH>loc&n!zQK$23>-N5dv@bAthhl=b_(+qJzXs
z-q`sa;t1TPV!Y9p5G`OZayOWtA*&KU3)GqXCaHi%UV0AQf1enN!k~?_i*ojxEXE0^
z9}j+vi3a{QGq9n8{D2XNLXl$b5Q@c_QOor!+~XhhL%{nOG=iy&lqA8ZYqF2X5#O&=
z&n$*Y9R_s||9(%?DyK~~rr1rY_Vy{(B?aG<dG&3I%MSCZ*jODZVqr(VwL8eO;f>E5
zp=`=9pMKS-JMf!Qup(;x?i-<83%_==WX(ci@&u(x5<liFv>AAys!jCa9itG4yqonm
zUo0_IJ3Kda3!M+SIlCiA$vV{cUn|D<AosO)TNFnP@0psCgpYzm3L-c!cVIiR&}KK@
zuM}%nEJyGcg#?O6lP+rR@gFc~uM1mAZjX=^B3X&&_X5|pcJ(i6f4%iAi#ePIrN>M{
z6O~Ov3RD9>3IB2c4m^@al)(^Xvo!dN7J{7l&!<%nRG&gqHj#xz6O^y+{x#guDbg`o
znRPJZSYQY;ht=`isUgC*PA}KfKVVmb3nn(t_j2x$H)}*lxF{VN*0S0Q=l@g{PiMSf
zZXV57@0{e)OGl+D*iV43b!O%%6X!b)N_cVb+^{Z=C)AmNN}Rw+kLMFrp&v}u#q&K~
z;{MMrDoG6RHj?6Cu-4BTdBvPS#;AC&0|#M}tC6E_6%|Q^fz3@<zRbBl#hY4TpKyMr
zsjOuhxkw3F{(KY>@v|wrazN^=qmH!rwwuEv`z6^08r#FwjKST>Kw@Jqc13Wfi9wq~
z&vBw4RY$W3%yoZGIx>r5HW@NnZc<8)J3_qd?10K6*^+2sIFod$H%<gZs2=Fz$Gdad
zJ$&ch1tU6KRbbfruW=}+Fo`dkPpf$1EvE(+kGiU}gKe?M9sUiANYEQr(&5}_FxZV%
zOHL}V5t!gvFzO-)!}uQ1jK{=u%@*K7NobhaDtf4Z@ET4J3EgQ9^th!eljY2~M<pqP
zY}Aw63{;XP-kTZ}HdAYY{KfDnlAm;qu}7jy2(Dla1pb0Kf5Nz`3U&|igOOpXGvsz%
z(Au3)PqvpzuegP-fCS%;4eZ@XoN=PV-g<q(eyx5(D!IW-$Hi6IKUYqh{_B?ZECZ$C
zk=+PDClb}dt0|S+Mod6kE%--NMXjc8w?K2;Y%xPr?Q%_x^)AHXk<zEg&?Ia%fttrs
z8fL`;$jvySo5fxiiL+D{uxS3&%fF;%##1ntD?DoB48@F|cx45c7~W)9JFaEi&|JG*
z`bE-dQ7^pJ5G9g^6u)`k(_>B=KJC6(yITg<<rgDaCC_Z$7gm~vX{_}<htMHQFxe!|
zP~-9naw$T1%rV?YfnOLn$1_uyNPlg_!IGo(Hzr*k*(;IY0(m}^^n(#1$2qPX$t%#2
z6%#95av%BtcErIH#d~8kJ*fv`988nvE;i#9g0_zD^+$r3@otU%)0GLD4y(|Jl}f;@
z5)pSfY`CmJ&NkuE#wqqMs<*QAzIMD+v*FbKNj#+stYS@YaL@m&s8F*m+Pv4Zw()I6
z#Kzy7hM;zpUK9tuRD&>Cp?@5HWu|O6*Y~zv1B#CwlPPNwPjAXqIwN(AU~FPiYbta*
zom#bfzq|RGeK61Fs4qU5np8E%YyJR!nU&fmhh<!jfu*m=Q^Jn7d?jw^zj(dWpvCZ<
zYC!LKJFG9G8oFgV&D?l0d7ot$iLM{U_S*-PF2O1|FoLL|*grQzgNaKqr9UsrT07d)
zFBk0k(f0YNzwq$7$<Lt9&dnHLkI6DTSp>}v@znMAPJKIhn4L?UN~<)}l)we-n~t>U
zx9zQ8fulom^5uawLy13V{@O<%PcmhwX?Ufp5n9U#Qx>aA-ixD-!1~%yQIPTD1q#<3
zn%B<SA-o6Z!sx{@!8)I!_$o6LyexYjz&d#6^ihf|q67knZ_tmkcY{jPxV9op{w9I%
z&nI=Mtdmprdd&m^CyTV~ES~_o)8erf;JQF+M-c`K5lo%p7qIkOuGl_J^{E=@kma{Z
zHc*~HU#iYX`}16UJ;|>f5KdOvOqUf^1YXzlcd0JyArLiOV90gBAt@WgtU}q9@T!lF
zmXRaGy+nEwa5p*p-m*_ZHr};+%1|;MdXb_O!InoCC)$lEyGA+pyA;qOPYX5Z2xXg`
zA*~Hv!Z(^SNv}+XFmLcVBQv{teq9c>azg)0NK7n6Nc;z^MpqQbqdL6+ztYBJ^(RBA
z*zfZ@SV2~F@&1qG5@|D92jRJoVEQPA{7MIYE6iJ!C}=pG9pXf+{}HJoiD9YClt+M+
zR-6Q7!Mdmr`Nvk_Ux<L=L9(-1B<=R#bHkl}1@n-fdouHCV>DMS;!(cf_XaCvqC<bh
zsGAWPDXny={cd*1`1FvjBnc&ze71ng5Cah`4X#CQJD>`ae9DnHRP%ku)1cp!S79Ek
zFiZ5?5zj*7wXGce*HrH1#tSXOz2-5u9}}P2>wmjkq+2G)?g#-<4exQAZ2gP}iz=ve
zb$qwnEeEmH)66)CQ<0~WNvNmlDXG148PpM6RLQ><ror!o9s5^-IgprKVgDvYMpKi&
z8&^{a_qYtDVFZavPZ$I{=fXRfA_mKNUt(L37c=Hd*CmG}mgZ=9^5C8AUQ;Dh<ExDa
z&Z_9?Y2l%GYK;L5Uk(zQDfrQTuhU{Dx*+^eH#4Ge_1yfV{`iC-8LKOuyG=_a%Tl|P
zPwQ62Criph2|~}Xc3rEBh<C*4+Rd3-y>Dc}b<xzFqM&*iE;xKY@JydBp#GLeGWoZy
z^&ZCd9TL|R$mf7gt>oL#OV>YUR&pjrg+%iYn|E*46>lnmQc~=o_c+Xo6i9_mY32fM
zu%cWip;aH5v<F#pPr2p>oS-9(+c@vI=5WOc=XEnq4eSPDaJ^sBx3Tz!p&(<k)QSda
z$RX^8>o`Y;=5|zQTfp|Hs3bloK8GklTmtS-$GSo${iG(HKG6u1Xdg)ekvimS<tWoa
z(GkJ1%c>?o$0-9l)G21nF^l99^mpWmT5Xff)7Xo*iQ$ZUmaDu{Bd~gR$?hSW^ki|F
zdM-ZFXe%dcR9So9=u^)uuw(i>d*kt!k!0iEGR295Wj>tYp&0FT!BDnZO6GD9u@08H
zS{jv(=6I6xB7vD_dW`=x*?2Tgh=MX`fgw3DlTR{5rwN`4&=eFDKgbKjdHZScZDG-Y
zD)yJmEt|m7CC#sji~ghda$(vdk;3;3O54supNFY4M!&$OrvgS!zgFIR!U@E&B~8}*
zs~m72<P0NxX8GXbDj#Fk@)(bUF5Klf7e8XhIFL5?g#dqNR*i;e*na5GPCPD7LuL!v
z8P4#kk;VPWEpSYbr3;BHN`#?c9eYs6qf(fIZhgyI#LU!+S=6dC<YaquU%@b9<*~ho
zty!aj@`cs(MTKm*hP;V#g3PzFze?wlrXoKw*bYSBlggJb7Bu;kCaK=JGI0i5V9~M^
zjS7|a$;Lw(evXw=RmaOj*BK2(RBI{g&|cc6ob6e!06e$ZEm?5thd{p}w^zIE^Ad4Q
z>#!lUj?gL7gw*${gy~dSvp3V4+-H<7XwEcQR9r+c(JMyjvE!?SqVp<ynfnh8P=^)N
zgEISP1xb@dt4n(t0VQev%*hKJdx3_|Ud?v$Xd3xxyf4!!+C!^T+z+Nq;03E`;@DZ;
zp(Bz$K<Go~$Pr~9<a5J(w8gX|O(V8Iiw?+Bxw8_|nG9vRGy|`3Q7F9?XaPyX*&_b7
zd*8Bq`2bzRK7ygBAvmrl(y}vpE>$7nSQ^UGia+L$3XC|>BO?cEdw$tfrJRshAH&GN
zAHUa|hg0X5ruG1q@P1|*t0Ejtu24#|_58H~;Gz;c6Vi88X<2)gg{cPqXzfZ+pWDIP
z$~cDidDOBe!V}69G4)T*sWx#=IHy&_%=h-bF;!mj85His8gBb%EzC$L-+c}o`fJta
zy*QwAX_NCd%mVFTJ<=Q(PL*RdyZN_#Vo5x|60<sR)uN!WhJ0;1A78SXR@v4NooV!e
z^uceRQmKt~b2M#Iu5E@|InsyC)0<^n&ySN0huxOlR18TX4&&YE3646|q+);XZ>}(7
z!Ihh*XF*Si@UTf3b$F4|`!f-2Ii|RNY1+2!d|{{^ZD<r~=5Wvmzs+f&(9Sp<@*`6#
z_EkIhynW^NiExc(R9mq6qk0K@L-h$fz&HN7*%~OB`Vzd|pq%kDUSC~DjhIhkS^3Mn
zPyUV~e8WfwEk8@#*Sbj;7XH(Ah<Ft(r)k>T*nM7l*km7f_&>e$@l{k-!-CH{PyD07
z+$jpTibZ@xd4Jl^5g(3~)W~OB3kwn({MU4OB|Z@q#iNta@k6oGcIVZYNko6w0^bZ5
zmU76H3cb=Pb<55?k0+B$yX;{61%ujClVGx01mSjMe8*8RVRnqg`?H(=a9~SZsweur
zR;G3>QC=3l07)KsdIi^cGRHi;t?O8_R;+POIw_O6-q|)helc2GXzO@u@@7i}r%i>e
zqkYWh7O22<RjLiTd%U`$d!a%NaCz^X9<gFaBY!*H9limd$i?~B3z@oXLo#E<d!fjq
zkii~Jq+(RaDa{h*Yn9~s#%K5`K5SYq(VQeBZFlT2ytxj<E$fU+b8r{XRFT>&e3HwL
z``Q1er9=akowJZtlS>j5WEFrqG8tIp!l12w5RenKG(IWw*JAT2+p&EX$PU2m`p)cB
z;6WKVly5;-Sgy)RyDzuBlCwN!Htq8{K}Z$CnX>7H{Q59ns_)sPM(e~J$^wEdX*^pl
z+Yj<*gl^H%*)P8C57P--183$zM6XS{LN?h=cHntE?rR@qAFb2XBQK<j1xoAlfEf@G
z>pj)xp^DB@ea_wjFBSvjdBsqzdh?&4fuDOOcWPLEKOZ!7{>xH&hu-_Rf|Q5oV{qN?
z(*DTHPJ%jd)b)K&geW&jK}1ql@Mx`KqMFr+grwt}@Bk~By_(EY1;+FFIS57Qy^vng
zl~mGNM7`${*7cYfn~jv3H<nujN|_<HxAWcZcF$wZC$b!_`HncCf>g*Zc<<=CkabH?
z?W&QF!va;+jf;BL_jraRd|)$-4|Af|`9keUy&ZR8Q~N?HGgSQom+pz)b5r3;vWs8}
z?;!o?D4P$ju>}L~PUJ3NNyUJx`LiQ^RqyQTn5GaW{Dm%RUn72)6jy=XN+{6-+-NEy
z`72K~mmd!Zt4#$$hiZ6RG1b{S2<IDHaEyBAHA}`=RJuIxH!m_QIZI}JRJ;PxeZ8%O
z#828(SmTCmz+Z8AKR)A=cNx5C32K68ppcV+_5#&@GJc@4YDL6DrKfUI-Uyvm!y3<v
z&F3Q{<npbxtHT4{2b8f~3eW7Bl;)l_cOjcpbc`9z*S5dn@caxhxB6GHa)DSAk`55Z
zyrjYii>c+-rx#}-k~zHT`fl^bIG99En#Rf4X~+olkHzhha*C-F<mZ7k`(EJCSdANR
zm#7GnK`-ZVz+Q+^wg_;|u$7khOq!}^-mH~&PSRZ5>|$NDgtU)!fqgYHL-~^2qCN4v
zA{pY<vj$u*dY~h;!=jQET^+(`DmPe&%GZW2M1*}VPFK$$e)o?T@wPez2>(KU?#xQH
zAcaomdwytrE&zP_1fD9lZq78Ld_m|v@7wap4CNcwqc`(i9U0wRVr(0uU+CL>ABGpp
zkN*J&2!3vxZWjYgs4mBj-S|UcX#LO2M-e8^wg9!p`IhAKysbUmp}kKHgLF)U|N2Wt
zfXor7+UxoChfx>k$@nr@@%{SP%ISP>YR1M1g8y#yvW4DvB8Bw_$d4S<PG(86ann~6
z07=&TmTVWnysZr8i{5pU=l8*@%lWOnvp)?schCvN-x7!q-E9S|+Vp>oIqt^HWD>+%
z;(>>9&r@8@Xu#)rqi~kfTUuF3Lsl6p?doR3^M{vAj#F0he%SWyHs9zNj%jhtipkcf
z7{~1RM6h*yhD60^5zaTaaL%qwVSW!GdD<;E3?AkLIs^EPbK`GYyGYK7s{<d5mvVwq
zur-%#ipR|OLUjGQkFkZag`Rh|o0~~)(*eY`Fe?f#?S}|TbP7FM*M%{S$3Tz_*ZW!B
z3!ZSLG-UZ7XV`k#Pd5Wc(+R}<%4by*=^NW^Vfsv@j%!i?T<;_FO=#icTjlyR<mzq&
z6wGTT12Zyj*AK*VNJu>ogwGM(`=)G!3TM67#t3dcoTLV<Rn%ijpKgZA-2hS%LjYmd
zi~ntMa@@bF(7Ny&Y+nSH3sM7}Nl6?xr<xF;)8}I2IzRl6V|}8Y#J|5R{!L0-F)EIn
zxAX+~bI-fGcI~N38CfeU7en>v^dD@b#@y;@71k*gdy8|{wWr238yvga@3#~ri)Ms}
z)Wa$1!<nj*y|CnXXmaL*6S1v$5p|RA(_;ch_HQ0gVT0>VFp3VERSI+5>Fh??iS*D0
z!?MqAm!q^HWWK<|;rFfSfdL}k5@Q!B&7gb{1RPD;1m1C*tNsk!IEe4y!Ks0e;mqe<
z^B1rZ2<h-YrBWWQloD}-P<l3A4i0v1Mm7#wHa1#%6jnuhGYJzHb8<Qf0S-0}Ha31X
z_Wy*W=t>|&zt%g?%ixI!Ik{0-rC?6ZF6111JSeOR=604YR^;s5{Cp{@`5-!=BURUV
zNd{x!&pr0vR2zon^f(JD%_24PpJ3<$bRl}HQV@yVZ%!aIx_A8hA5!E|a?#r6l6ILx
z{vW`D^6h>uA}wNG;yZP|Z-lS&I6;<jHecS&Q%6&Wnr|W<*Wu@lYV;s;w?_3K?Hu|&
ze@tZ#PJw=P#K^0%eK$=N$&tbaWU%e#>I>@A+&KS22>vNEip#e^F=|f3&d-iUJWvAx
zH!-=Gl(two2fM@3+h(3&POB!wEq%2ybv7l44(Nsq#|2Yzan1xqUE$_Ch3F*r=qI-I
zV$vYuAl4GYw;@m`k0g0Xwd5f}61y>!j(S3H4Mdo!{I?dAX~YDr#duQzC3A+)Mi$#-
z7CkN~({G9?J|Vadkdr<Cj#&4K+jioBQn2LLD%2a}Voe7JDfkfV4H`zt1TG4O1cgAw
z`3*m5Mtv&&Q4iMnCW+{aDM>n-gX>Jd$STI+&vaGC{}c~{lVzfKNfQ-TF8T&jE0nnw
z|A;I2kp@^N))k0?C*G$8Ofa#;#C%!XW09cJzk{JJ8~94=*yFk2y>sLTHa|$qH^xmi
z{9i~_8;Eurkd<!jO((id<h8V%!uS>*cQ!;a8ovz_L}#RRBX>WH_n{R2Z@ojXIR^9m
z$j-4yH;{cz1o&bTn=dM_FSgFEdFIrj0@I3<>-tOgL<YTQTe-A>)l_q$1nYXsUUqUu
zm2E+pO#|WGUdK7K1akdvYttii^ntabLJ9)05lo`B^t)WfdXcoG#e!ACOqWP*S`b~9
zD!o`Wl=n9K7#qLXYi;qF`dF#8l1dT!>l>4GG--Rw!cF|8%P|Q@eyebFX)Mx(Mxa^~
zJdt0EAJ9g8=hD6f*jRq4?CLPp$OOuK#83|88_W?6=1?25f$U=~`$Kw7i%#nlb}!4Y
z@61sn#kTM1cZ<!BE08O8>Ikvr)t9iw%PVMCSk!&9j+;QVU(C-!<*y?ts#@dNi%=xF
z&~`o-wI2l-4aG=9L8P|sC577ehR89}c~0$^WHDB6Hzei&1io%;>+TZ#y`oY=7-Z=a
zjV8Lk-R;$l)m0Ve>lAf?CiHB0R5?&G+Zc6)9F3#*hJ5UDd_`R`|E-U5b0<QAcQ%DL
z|BdDbtJmJD9eNMC=SkMZ*vTEy#(_Zl--mN;Y3-iJHn-<)wy^e8!n0*rzD2R_Bka*N
zaWPoXMWzSv1<eTEb?t+-QI=p}t&g!Q$!Oh8Kx@dGBi|oqq&{=lM8mr8JP&v>7MTd`
z1_?9x5SAC$>i`jhT7EPuJCd;##tcV{MHZ<nlNbKUm1iqlpT6*=5YZ3(o%p_;b6O24
zUMA~jqRVBH^2eGd<PZV;?3F(&@0ZBK5{eDB^vlu!Z}=#f6<GD;__zuvMR8f*`MWh>
z&`%)To`JupHAO&}U!WbROWS+9qK&m?l$o8K!-`zaZOf$nu$h&t)=8ar@A>m)|F~Fh
z)l#(0-~Lh7*ruY^(_65*_AAoQ@jce(Ne5&;&+vZPRBl<Ti((up1VynQZ)IM<>-S^x
z-4`@~ap1S))m}@8qKGR}dqMTQ<?XX#d$8f+5(CDU!w1+U!8pZTu=H*7z}^{R!Hw~J
z8~D#K-^LfgVVw@oP?5XCJlF87<zSN;3CC5DCvQuyHPnls>o-;%Bi#q9vnZ9Y@&#YU
zDQ>I>r5nt#s`j)|z2CAZy+7N2sO2Dg++*benNReo>%5X_%y)X~S#rF+On<R^QfNPA
zcXIRb-tyxmiG1(vV-5Yo@1%{Tt2=Z!5G1_yD`Lj-cZS_&7un9I^e$>mKi&=lDPX!;
zA~k*N_inmvCv|kuBJPr>h_t>2D344+>{CP|VPl_q6C~5tdd;SR5%h=YoD?U^?maOU
zkRIhQTW<rS%ru`d(el?4Z;}~yZAZ7Ws&WYUZA(>utpbf_A9GcmD`H8znw@_(3hyjK
zB(cO%Bx?$4IOSwxDSwn4XwtQpaLn|iY)ZRcl;cvhR{wP&aDezt*j1!TEpE=cAijOv
z@5`kCBLl;sv#MtIjgxK3epPLmlgi7HOd{`fTb00v5$B{A4SYq9wh{JILB8m467<~Y
z5PR;S3eDNwAH|GeB|5_9nrGx`GA%>f=5Zln6)5*k6k7!x#2&8CWW1B_YyKx!ai{bt
zAiM{73{hCsVV>sX>^!eYN1I%qot%@L{ne;uZ}0MI<RoV!S3_ZywX?8)J#qcdNrGHo
zkcU^Ci=B;~omWDflTSjDlaGy)lTCt4Tuhu(ijDK5xG?$uKg1C7e~FHZBgIh>fdQbN
z=2uS?!w%Z;dU(WkP!a7$!$B}0z?O`2?0L6n@Y9_5_8AX9jUbgse^yf~EgnBrx4#ho
z4hwa(l7nr^1AR!{vbfFuPx>;}y9-vyLKS|VcT{2la7);eJ3*K{4TC4Y5pnj7O(;tt
zO+C2#{v<WXP2Ea7f7u%u1v07>{M&&v34aaMEAQ)%v^BA^twc?iBs*BwYR#s(6B$1G
zWG#4ZOCtfwHL<SZ)%iaTv0T#kSzF5uYrr?Z-VlHD?VEY9uOSOI#+NcCBlZ*HL;dU^
z!~VQH5dDizV*{!3zG;gMYH5NJO^*xQFCg<FV>zHG^-yK{J3ElF{N$R!j)DSsEl4T3
z(3IBs&zTe+o!T;_w@gC=yT+ur#_y_raW~!h8-x(mAdEO**>&4WIxiNT(ye(rf1wke
z$vTxwGWlM7C6IY4YY*_PrU*0h3&jdo$K2$p*{&SKO|_|hACLp)GP6spo?JZ3HlGWp
zIdRCkJ_u>os(tm$+Z+j>mQP}!3{<Mt7@v_zYmR&Lr;YW1OPjP<6SHVYJJ_R0y6&p%
elDC$60g3uahNLJcBaor6@o=J0Q%fpKq5MCswj)#k

diff --git a/doc/Projects/2020/Project1/pdf/Project1.tex b/doc/Projects/2020/Project1/pdf/Project1.tex
index 43a7ec7c1..db8035a2e 100644
--- a/doc/Projects/2020/Project1/pdf/Project1.tex
+++ b/doc/Projects/2020/Project1/pdf/Project1.tex
@@ -136,8 +136,9 @@ Sep 1, 2020
 
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 We will first study how to fit polynomials to a specific
 two-dimensional function called \href{{http://www.dtic.mil/dtic/tr/fulltext/u2/a081688.pdf}}{Franke's
@@ -176,7 +177,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 \begin{verbatim}
 from mpl_toolkits.mplot3d import Axes3D
 import matplotlib.pyplot as plt
@@ -221,7 +222,7 @@ plt.show()
 \end{verbatim}
 
 
-\paragraph{Part a): Ordinary Least Square on the Franke function  with resampling.}
+\paragraph{Part a): Ordinary Least Square (OLS) on the Franke function.}
 We will generate our own dataset for a function
 $\mathrm{FrankeFunction}(x,y)$ with $x,y \in [0,1]$. The function
 $f(x,y)$ is the Franke function. You should explore also the addition
@@ -232,7 +233,7 @@ distribution $\cal{N}(0,1)$.
 value decomposition from e.g., \textbf{numpy} ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in $x$ and $y$ up to fifth order. Find the
-confidence intervals of the parameters $\beta$ by computing their
+\href{{https://en.wikipedia.org/wiki/Confidence_interval}}{confidence intervals} of the parameters (estimators) $\beta$ by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 \[ MSE(\hat{y},\hat{\tilde{y}}) = \frac{1}{n}
@@ -253,9 +254,14 @@ where we have defined the mean value  of $\hat{y}$ as
 \bar{y} =  \frac{1}{n} \sum_{i=0}^{n - 1} y_i.
 \]
 
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library \textbf{Scikit-Learn} (make sure you have installed it).
-This function is called $train\_test\_split$. Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of \textbf{Scikit-Learn}.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also \href{{https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html}}{homework set 2} for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library \textbf{Scikit-Learn} (make
+sure you have installed it).  This function is called
+$train\_test\_split$. Similarly, and see the solution to \href{{https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html}}{homework set 2, exercise 2}, you can use the data normalization functionality of
+\textbf{Scikit-Learn}.
 
 It is normal in essentially all Machine Learning studies to split the
 data in a training set and a test set (eventually  also an additional
@@ -269,7 +275,7 @@ approximately $2/3$ to $4/5$ of the data as training data.
 
 
 
-\paragraph{Part b): Bias-variance trade-off.}
+\paragraph{Part b): Bias-variance trade-off and resamplng techniques.}
 Our aim here is to study the bias-variance trade-off by implementing the \textbf{bootstrap} resampling technique.
 
 With a code which does OLS and includes resampling techniques, 
@@ -278,6 +284,14 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms. 
 
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig.~2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+With this result we move on to the bias-variance trade-off analysis.
+
 Consider a
 dataset $\mathcal{L}$ consisting of the data
 $\mathbf{X}_\mathcal{L}=\{(y_j, \boldsymbol{x}_j), j=0\ldots n-1\}$.
@@ -302,6 +316,7 @@ squared error via the so-called cost function
 \[
 C(\bm{X},\bm{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\bm{y}-\bm{\tilde{y}})^2\right].
 \]
+Here the expected value $\mathbb{E}$ is the sample value. 
 
 Show that you can rewrite  this as
 \[
@@ -311,21 +326,21 @@ Show that you can rewrite  this as
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
 
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the \textbf{bootstrap} resampling method.
 
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig.~2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
+Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
 
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the \textbf{bootstra} resampling technique.
 
 \paragraph{Part c) Cross-validation as resampling techniques, adding more complexity.}
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -334,7 +349,7 @@ data and test data using for example
 Implement the $k$-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-\textbf{Scikit-Learn} if needed. You can alternatively write your own bootstrap code.
+\textbf{Scikit-Learn} if needed. 
 
 Compare the MSE you get from your cross-validation code with the one you got from your \textbf{bootstrap} code.
 You can also compare your own cross-validation code with the one provided by \textbf{Scikit-Learn}.
@@ -408,7 +423,7 @@ Austfjell, again in Norway.
 Feel free to produce your own terrain data.
 
 
-Alternatively, if you would like to use another data set, feel free to do so. This could be data close to your research area or simply a data set you found interesting. See for example \href{{https://www.kaggle.com/datasets}}{kaggle.com} for examples.
+Alternatively, if you would like to use another data set, feel free to do so. This could be data close to your reseach area or simply a data set you found interesting. See for example \href{{https://www.kaggle.com/datasets}}{kaggle.com} for examples.
 
 \paragraph{Part g) OLS, Ridge and Lasso regression with resampling.}
 Our final part deals with the parameterization of your digital terrain
diff --git a/doc/src/Projects/2020/Project1/Project1.do.txt b/doc/src/Projects/2020/Project1/Project1.do.txt
index 6fd9ddcb6..53dc0b901 100644
--- a/doc/src/Projects/2020/Project1/Project1.do.txt
+++ b/doc/src/Projects/2020/Project1/Project1.do.txt
@@ -8,8 +8,9 @@ DATE: today
 
 The main aim of this project is to study in more detail various
 regression methods, including the Ordinary Least Squares (OLS) method,
-Ridge regression and finally Lasso regression.
-The methods are in turn combined with resampling techniques like the bootstrap method and cross validation.
+Ridge regression and finally Lasso regression. Ridge regression will be discussed during the Friday lecture of week 36 while Lasso Regression will be discussed during the lectures of week 37.
+
+The methods are in turn combined with resampling techniques like the bootstrap method and cross validation. These are discussed during weeks 36 and 37.
 
 We will first study how to fit polynomials to a specific
 two-dimensional function called "Franke's
@@ -50,7 +51,7 @@ beyond the second-order polynomials metioned above and explore
 which polynomial fits the data best.
 
 
-The Python fucntion for the Franke function is included here (it performs also a three-dimensional plot of it)
+The Python code for the Franke function is included here (it performs also a three-dimensional plot of it)
 !bc pycod
 from mpl_toolkits.mplot3d import Axes3D
 import matplotlib.pyplot as plt
@@ -95,7 +96,7 @@ plt.show()
 !ec
 
 
-=== Part a): Ordinary Least Square on the Franke function  with resampling ===
+=== Part a): Ordinary Least Square (OLS) on the Franke function ===
 
 We will generate our own dataset for a function
 $\mathrm{FrankeFunction}(x,y)$ with $x,y \in [0,1]$. The function
@@ -107,7 +108,7 @@ distribution $\cal{N}(0,1)$.
 value decomposition from e.g., _numpy_ ) or use your code from
 homeworks 1 and 2 and perform a standard least square regression
 analysis using polynomials in $x$ and $y$ up to fifth order. Find the
-confidence intervals of the parameters $\beta$ by computing their
+"confidence intervals":"https://en.wikipedia.org/wiki/Confidence_interval" of the parameters (estimators) $\beta$ by computing their
 variances, evaluate the Mean Squared error (MSE)
 
 !bt 
@@ -134,9 +135,14 @@ where we have defined the mean value  of $\hat{y}$ as
 \]
 !et
 
-Your code has to include a scaling of the data (for example by subtracting the mean value, see also homework set 2 for examples) and a split of the data in training and test data. For this part you can either write your own code or use for example the
-function for splitting training data provided by the library _Scikit-Learn_ (make sure you have installed it).
-This function is called $train\_test\_split$. Similarly, and see the solution to homework set 2, exercise 2, you can use the data normalization functionality of _Scikit-Learn_.
+Your code has to include a scaling of the data (for example by
+subtracting the mean value, see also "homework set 2":"https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html" for examples) and
+a split of the data in training and test data. For this part you can
+either write your own code or use for example the function for
+splitting training data provided by the library _Scikit-Learn_ (make
+sure you have installed it).  This function is called
+$train\_test\_split$. Similarly, and see the solution to "homework set 2, exercise 2":"https://compphysics.github.io/MachineLearning/doc/Projects/2020/hw2/html/hw2-bs.html", you can use the data normalization functionality of
+_Scikit-Learn_.
 
 It is normal in essentially all Machine Learning studies to split the
 data in a training set and a test set (eventually  also an additional
@@ -150,7 +156,7 @@ approximately $2/3$ to $4/5$ of the data as training data.
 
 
 
-=== Part b): Bias-variance trade-off ===
+=== Part b): Bias-variance trade-off and resamplng techniques ===
 
 Our aim here is to study the bias-variance trade-off by implementing the _bootstrap_ resampling technique.
 
@@ -160,6 +166,14 @@ continuous predictions such as regression. However, many of the
 intuitions and ideas discussed here also carry over to classification
 tasks and basically all Machine Learning algorithms. 
 
+Before you perform an analysis of the bias-variance trade-off on your test data, make
+first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
+Friedman. Figure 2.11 of this reference displays only the test and training MSEs. The test MSE can be used to 
+indicate possible regions of low/high bias and variance. You will most likely not get an
+equally smooth curve!
+
+With this result we move on to the bias-variance trade-off analysis.
+
 Consider a
 dataset $\mathcal{L}$ consisting of the data
 $\mathbf{X}_\mathcal{L}=\{(y_j, \boldsymbol{x}_j), j=0\ldots n-1\}$.
@@ -188,6 +202,7 @@ squared error via the so-called cost function
 C(\bm{X},\bm{\beta}) =\frac{1}{n}\sum_{i=0}^{n-1}(y_i-\tilde{y}_i)^2=\mathbb{E}\left[(\bm{y}-\bm{\tilde{y}})^2\right].
 \]
 !et
+Here the expected value $\mathbb{E}$ is the sample value. 
 
 Show that you can rewrite  this as
 !bt
@@ -199,23 +214,23 @@ Show that you can rewrite  this as
 Explain what the terms mean, which one is the bias and which one is
 the variance and discuss their interpretations.
 
+Perform then a bias-variance analysis of the Franke function by
+studying the MSE value as function of the complexity of your model.
 
 Discuss the bias and variance trade-off as function
 of your model complexity (the degree of the polynomial) and the number
 of data points, and possibly also your training and test data using the _bootstrap_ resampling method.
 
-However, before you perform an analysis of the bias-variance trade-off on your test data, make
-first a figure similar to Fig. 2.11 of Hastie, Tibshirani, and
-Friedman. Figure 2.11 of this reference displays only the test and training MSEs (and bias-variance analysis of the test data) while indicating possible regions of low/high bias and variance. You will most likely not get an
-equally smooth curve! Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
+Note also that when you calculate the bias, in all applications you don't know the function values $f_i$. You would hence replace them with the actual data points $y_i$.
 
-After having produced a curve similar to Figure 2.11, perform then a bias-variance analysis of your test data. Here you should use the _bootstra_ resampling technique.
 
 === Part c) Cross-validation as resampling techniques, adding more complexity ===
 
 
-The aim here is to write our own code for another widely popular resampling technique, the so-called cross-validation method.
-Again, before you start with cross-validation approach, you should scale your data and split it in test and training data as you did earlier.
+The aim here is to write your own code for another widely popular
+resampling technique, the so-called cross-validation method.  Again,
+before you start with cross-validation approach, you should scale your
+data and split it in test and training data as you did earlier.
 Perform a resampling of the data where you split the data in training
 data and test data using for example
 
@@ -224,7 +239,7 @@ data and test data using for example
 Implement the $k$-fold cross-validation algorithm (write your own
 code) and evaluate again the MSE function resulting
 from the test data. You can compare your own code with that from
-_Scikit-Learn_ if needed. You can alternatively write your own bootstrap code.
+_Scikit-Learn_ if needed. 
 
 Compare the MSE you get from your cross-validation code with the one you got from your _bootstrap_ code.
 You can also compare your own cross-validation code with the one provided by _Scikit-Learn_.
diff --git a/doc/src/Projects/2020/Project1/make.sh b/doc/src/Projects/2020/Project1/make.sh
index 740fedbc3..c050a5717 100755
--- a/doc/src/Projects/2020/Project1/make.sh
+++ b/doc/src/Projects/2020/Project1/make.sh
@@ -38,7 +38,7 @@ system doconce split_html $html.html --method=split --pagination --nav_button=bo
 
 # Ordinary plain LaTeX document
 system doconce format pdflatex $name --print_latex_style=trac --latex_admon=paragraph $opt
-system doconce ptex2tex $name envir=print
+system doconce ptex2tex $name envir=verbatim
 # Add special packages
 doconce subst "% Add user's preamble" "\g<1>\n\\usepackage{simplewick}" $name.tex
 doconce replace 'section{' 'section*{' $name.tex