From cda1372a19aabb48b5eb88f838c3a2c77ec1ba9a Mon Sep 17 00:00:00 2001 From: Morten Hjorth-Jensen Date: Tue, 15 Nov 2022 09:33:04 +0100 Subject: [PATCH] update p3 --- .../2022/Project3/html/._Project3-bs000.html | 16 ++-- .../2022/Project3/html/Project3-bs.html | 16 ++-- doc/Projects/2022/Project3/html/Project3.html | 14 +-- .../2022/Project3/ipynb/Project3.ipynb | 82 +++++++++--------- .../Project3/ipynb/ipynb-Project3-src.tar.gz | Bin 193 -> 194 bytes doc/Projects/2022/Project3/pdf/Project3.p.tex | 10 ++- doc/Projects/2022/Project3/pdf/Project3.pdf | Bin 247226 -> 247706 bytes doc/Projects/2022/Project3/pdf/Project3.tex | 10 ++- .../Projects/2022/Project3/Project3.do.txt | 10 ++- 9 files changed, 87 insertions(+), 71 deletions(-) diff --git a/doc/Projects/2022/Project3/html/._Project3-bs000.html b/doc/Projects/2022/Project3/html/._Project3-bs000.html index 6a86b4453..119db48b5 100644 --- a/doc/Projects/2022/Project3/html/._Project3-bs000.html +++ b/doc/Projects/2022/Project3/html/._Project3-bs000.html @@ -65,11 +65,11 @@ doconce format html Project3.do.txt --html_style=bootstrap --pygments_html_style None, 'part-d-solving-eigenvalue-problems'), ('Part e)', 3, None, 'part-e'), - ('Additonal (optional) exercise, adding 30 more points to final ' + ('Additonal (optional) exercise, adding 20 more points to final ' 'score', 2, None, - 'additonal-optional-exercise-adding-30-more-points-to-final-score'), + 'additonal-optional-exercise-adding-20-more-points-to-final-score'), ('Introduction to numerical projects', 2, None, @@ -130,7 +130,7 @@ MathJax.Hub.Config({
  •       Part c) Neural networks
  •       Part d) Solving eigenvalue problems
  •       Part e)
    • -
  •    Additonal (optional) exercise, adding 30 more points to final score
    • +
  •    Additonal (optional) exercise, adding 20 more points to final score
  •    Introduction to numerical projects
  •    Format for electronic delivery of report and programs
  •    Software and needed installations
    • @@ -160,7 +160,7 @@ MathJax.Hub.Config({
    -

    Nov 14, 2022

    +

    Nov 15, 2022


    @@ -175,9 +175,11 @@ MathJax.Hub.Config({
  • The University of California at Irvine (UCI) with its machine learning repository.
  • Or other sources.
    • +

    As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL":"https://link.springer.com/article/10.1007/s10614-020-10078-2". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester.

    +

    The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is:

      -
    1. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as Principal Component Analysis and k-means Clustering.
      2. +
    2. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like Principal Component Analysis and k-means Clustering.

    For Boosting, feel also free to write your own codes.

    @@ -194,7 +196,7 @@ MathJax.Hub.Config({

    This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides from week 42 and/or the textbook by Yadav et al.

    -

    Note: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise.

    +

    Note: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise.

    The basic structure of your project

    Here follows a set up on how to structure your report and analyze the data you have opted for.

    @@ -330,7 +332,7 @@ eigenvalues. Compare with the solution from numerical diagonalization with stand

    Part e)

    Finally, present a critical assessment of the methods you have studied and discuss the potential for the solving differential equations and eigenvalue problems with machine learning methods.

    -

    Additonal (optional) exercise, adding 30 more points to final score

    +

    Additonal (optional) exercise, adding 20 more points to final score

    This exercise can be done independently of the other tasks. Here you can also choose the data set you want to use. Furthermore, you can use diff --git a/doc/Projects/2022/Project3/html/Project3-bs.html b/doc/Projects/2022/Project3/html/Project3-bs.html index 6a86b4453..119db48b5 100644 --- a/doc/Projects/2022/Project3/html/Project3-bs.html +++ b/doc/Projects/2022/Project3/html/Project3-bs.html @@ -65,11 +65,11 @@ doconce format html Project3.do.txt --html_style=bootstrap --pygments_html_style None, 'part-d-solving-eigenvalue-problems'), ('Part e)', 3, None, 'part-e'), - ('Additonal (optional) exercise, adding 30 more points to final ' + ('Additonal (optional) exercise, adding 20 more points to final ' 'score', 2, None, - 'additonal-optional-exercise-adding-30-more-points-to-final-score'), + 'additonal-optional-exercise-adding-20-more-points-to-final-score'), ('Introduction to numerical projects', 2, None, @@ -130,7 +130,7 @@ MathJax.Hub.Config({

  •       Part c) Neural networks
  •       Part d) Solving eigenvalue problems
  •       Part e)
    • -
  •    Additonal (optional) exercise, adding 30 more points to final score
    • +
  •    Additonal (optional) exercise, adding 20 more points to final score
  •    Introduction to numerical projects
  •    Format for electronic delivery of report and programs
  •    Software and needed installations
    • @@ -160,7 +160,7 @@ MathJax.Hub.Config({
    -

    Nov 14, 2022

    +

    Nov 15, 2022


    @@ -175,9 +175,11 @@ MathJax.Hub.Config({
  • The University of California at Irvine (UCI) with its machine learning repository.
  • Or other sources.
    • +

    As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL":"https://link.springer.com/article/10.1007/s10614-020-10078-2". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester.

    +

    The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is:

      -
    1. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as Principal Component Analysis and k-means Clustering.
      2. +
    2. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like Principal Component Analysis and k-means Clustering.

    For Boosting, feel also free to write your own codes.

    @@ -194,7 +196,7 @@ MathJax.Hub.Config({

    This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides from week 42 and/or the textbook by Yadav et al.

    -

    Note: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise.

    +

    Note: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise.

    The basic structure of your project

    Here follows a set up on how to structure your report and analyze the data you have opted for.

    @@ -330,7 +332,7 @@ eigenvalues. Compare with the solution from numerical diagonalization with stand

    Part e)

    Finally, present a critical assessment of the methods you have studied and discuss the potential for the solving differential equations and eigenvalue problems with machine learning methods.

    -

    Additonal (optional) exercise, adding 30 more points to final score

    +

    Additonal (optional) exercise, adding 20 more points to final score

    This exercise can be done independently of the other tasks. Here you can also choose the data set you want to use. Furthermore, you can use diff --git a/doc/Projects/2022/Project3/html/Project3.html b/doc/Projects/2022/Project3/html/Project3.html index 0ce8a7793..7957b5e9e 100644 --- a/doc/Projects/2022/Project3/html/Project3.html +++ b/doc/Projects/2022/Project3/html/Project3.html @@ -143,11 +143,11 @@ div.toc p,a { None, 'part-d-solving-eigenvalue-problems'), ('Part e)', 3, None, 'part-e'), - ('Additonal (optional) exercise, adding 30 more points to final ' + ('Additonal (optional) exercise, adding 20 more points to final ' 'score', 2, None, - 'additonal-optional-exercise-adding-30-more-points-to-final-score'), + 'additonal-optional-exercise-adding-20-more-points-to-final-score'), ('Introduction to numerical projects', 2, None, @@ -194,7 +194,7 @@ MathJax.Hub.Config({

    -

    Nov 14, 2022

    +

    Nov 15, 2022


    Paths for project 3

    @@ -206,9 +206,11 @@ MathJax.Hub.Config({
  • The University of California at Irvine (UCI) with its machine learning repository.
  • Or other sources.
    • +

    As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL":"https://link.springer.com/article/10.1007/s10614-020-10078-2". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester.

    +

    The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is:

      -
    1. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as Principal Component Analysis and k-means Clustering.
      2. +
    2. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among linear regression (including Ridge and Lasso), Logistic Regression, Neural Networks, Convolution Neural Networks, Recurrent Neural Networks, Adversarial Neural Networks, Support Vector Machines and Decision Trees, Random Forests, Bagging and Boosting. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into convolutional neural networks or recurrent neural networks, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like Principal Component Analysis and k-means Clustering.

    For Boosting, feel also free to write your own codes.

    @@ -225,7 +227,7 @@ MathJax.Hub.Config({

    This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides from week 42 and/or the textbook by Yadav et al.

    -

    Note: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise.

    +

    Note: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise.

    The basic structure of your project

    Here follows a set up on how to structure your report and analyze the data you have opted for.

    @@ -361,7 +363,7 @@ eigenvalues. Compare with the solution from numerical diagonalization with stand

    Part e)

    Finally, present a critical assessment of the methods you have studied and discuss the potential for the solving differential equations and eigenvalue problems with machine learning methods.

    -

    Additonal (optional) exercise, adding 30 more points to final score

    +

    Additonal (optional) exercise, adding 20 more points to final score

    This exercise can be done independently of the other tasks. Here you can also choose the data set you want to use. Furthermore, you can use diff --git a/doc/Projects/2022/Project3/ipynb/Project3.ipynb b/doc/Projects/2022/Project3/ipynb/Project3.ipynb index 996f643d5..06aa75f6d 100644 --- a/doc/Projects/2022/Project3/ipynb/Project3.ipynb +++ b/doc/Projects/2022/Project3/ipynb/Project3.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "markdown", - "id": "0c2e93c2", + "id": "4266c21c", "metadata": { "editable": true }, @@ -14,7 +14,7 @@ }, { "cell_type": "markdown", - "id": "2dc0413d", + "id": "403dffc3", "metadata": { "editable": true }, @@ -22,14 +22,14 @@ "# Project 3 on Machine Learning, deadline December 15 (midnight), 2021\n", "**[Data Analysis and Machine Learning FYS-STK3155/FYS4155](http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html)**, Department of Physics, University of Oslo, Norway\n", "\n", - "Date: **Nov 14, 2022**\n", + "Date: **Nov 15, 2022**\n", "\n", "Copyright 1999-2022, [Data Analysis and Machine Learning FYS-STK3155/FYS4155](http://www.uio.no/studier/emner/matnat/fys/FYS3155/index-eng.html). Released under CC Attribution-NonCommercial 4.0 license" ] }, { "cell_type": "markdown", - "id": "92237bb9", + "id": "e8360cac", "metadata": { "editable": true }, @@ -39,7 +39,7 @@ }, { "cell_type": "markdown", - "id": "9b6ca5fb", + "id": "1bf77954", "metadata": { "editable": true }, @@ -53,8 +53,10 @@ "\n", "3. Or other sources.\n", "\n", + "As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL\":\"https://link.springer.com/article/10.1007/s10614-020-10078-2\". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester.\n", + "\n", "The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is:\n", - "1. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among **linear regression (including Ridge and Lasso)**, **Logistic Regression**, **Neural Networks**, **Convolution Neural Networks**, **Recurrent Neural Networks**, **Adversarial Neural Networks**, **Support Vector Machines** and **Decision Trees, Random Forests, Bagging and Boosting**. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into **convolutional neural networks** or **recurrent neural networks**, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as **Principal Component Analysis** and **k-means Clustering**.\n", + "1. Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among **linear regression (including Ridge and Lasso)**, **Logistic Regression**, **Neural Networks**, **Convolution Neural Networks**, **Recurrent Neural Networks**, **Adversarial Neural Networks**, **Support Vector Machines** and **Decision Trees, Random Forests, Bagging and Boosting**. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into **convolutional neural networks** or **recurrent neural networks**, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like **Principal Component Analysis** and **k-means Clustering**.\n", "\n", "For Boosting, feel also free to write your own codes.\n", "\n", @@ -74,12 +76,12 @@ "\n", "This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides [from week 42](https://compphysics.github.io/MachineLearning/doc/pub/week42/html/week42.html) and/or the textbook by [Yadav et al](https://www.springer.com/gp/book/9789401798150).\n", "\n", - "**Note**: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise." + "**Note**: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise." ] }, { "cell_type": "markdown", - "id": "d3d26ef5", + "id": "b983bd08", "metadata": { "editable": true }, @@ -91,7 +93,7 @@ }, { "cell_type": "markdown", - "id": "a03830d7", + "id": "deecb4e5", "metadata": { "editable": true }, @@ -103,7 +105,7 @@ }, { "cell_type": "markdown", - "id": "3bf25ce5", + "id": "1e46b462", "metadata": { "editable": true }, @@ -115,7 +117,7 @@ }, { "cell_type": "markdown", - "id": "0ba7777e", + "id": "cd593738", "metadata": { "editable": true }, @@ -127,7 +129,7 @@ }, { "cell_type": "markdown", - "id": "129fff02", + "id": "8e82b99f", "metadata": { "editable": true }, @@ -139,7 +141,7 @@ }, { "cell_type": "markdown", - "id": "da24bb44", + "id": "531cb90c", "metadata": { "editable": true }, @@ -151,7 +153,7 @@ }, { "cell_type": "markdown", - "id": "b4d8bd6f", + "id": "df454e93", "metadata": { "editable": true }, @@ -171,7 +173,7 @@ }, { "cell_type": "markdown", - "id": "9fb8b554", + "id": "1891abfd", "metadata": { "editable": true }, @@ -185,7 +187,7 @@ }, { "cell_type": "markdown", - "id": "9c1a12d5", + "id": "805bfc54", "metadata": { "editable": true }, @@ -197,7 +199,7 @@ }, { "cell_type": "markdown", - "id": "81df663e", + "id": "04191794", "metadata": { "editable": true }, @@ -207,7 +209,7 @@ }, { "cell_type": "markdown", - "id": "3d639530", + "id": "48917dfc", "metadata": { "editable": true }, @@ -219,7 +221,7 @@ }, { "cell_type": "markdown", - "id": "b4a3d10e", + "id": "62431c9b", "metadata": { "editable": true }, @@ -229,7 +231,7 @@ }, { "cell_type": "markdown", - "id": "e72fff12", + "id": "eccc5397", "metadata": { "editable": true }, @@ -241,7 +243,7 @@ }, { "cell_type": "markdown", - "id": "1448ffdc", + "id": "bef3bdae", "metadata": { "editable": true }, @@ -252,7 +254,7 @@ }, { "cell_type": "markdown", - "id": "4a10dc7f", + "id": "1f76902d", "metadata": { "editable": true }, @@ -264,7 +266,7 @@ }, { "cell_type": "markdown", - "id": "665a89f8", + "id": "a3ca949c", "metadata": { "editable": true }, @@ -274,7 +276,7 @@ }, { "cell_type": "markdown", - "id": "bfb1865c", + "id": "28a0daf3", "metadata": { "editable": true }, @@ -286,7 +288,7 @@ }, { "cell_type": "markdown", - "id": "caac2a80", + "id": "856db5f5", "metadata": { "editable": true }, @@ -299,7 +301,7 @@ }, { "cell_type": "markdown", - "id": "864c151d", + "id": "74f847c2", "metadata": { "editable": true }, @@ -311,7 +313,7 @@ }, { "cell_type": "markdown", - "id": "0e7ee4f9", + "id": "50c6a638", "metadata": { "editable": true }, @@ -321,7 +323,7 @@ }, { "cell_type": "markdown", - "id": "e40b01ba", + "id": "732fa78e", "metadata": { "editable": true }, @@ -333,7 +335,7 @@ }, { "cell_type": "markdown", - "id": "bde8abf8", + "id": "83b8bd01", "metadata": { "editable": true }, @@ -343,7 +345,7 @@ }, { "cell_type": "markdown", - "id": "fbb28a61", + "id": "46d68c66", "metadata": { "editable": true }, @@ -355,7 +357,7 @@ }, { "cell_type": "markdown", - "id": "88fdfd23", + "id": "423e0b70", "metadata": { "editable": true }, @@ -366,7 +368,7 @@ }, { "cell_type": "markdown", - "id": "ae0f5018", + "id": "9416896f", "metadata": { "editable": true }, @@ -382,7 +384,7 @@ }, { "cell_type": "markdown", - "id": "64e90ef9", + "id": "9484df9f", "metadata": { "editable": true }, @@ -399,7 +401,7 @@ }, { "cell_type": "markdown", - "id": "ea02a377", + "id": "dbfc73ef", "metadata": { "editable": true }, @@ -415,7 +417,7 @@ }, { "cell_type": "markdown", - "id": "57f27bd3", + "id": "ea92bb58", "metadata": { "editable": true }, @@ -427,12 +429,12 @@ }, { "cell_type": "markdown", - "id": "684f4cd8", + "id": "2f3ea4b0", "metadata": { "editable": true }, "source": [ - "## Additonal (optional) exercise, adding 30 more points to final score\n", + "## Additonal (optional) exercise, adding 20 more points to final score\n", "\n", "This exercise can be done independently of the other tasks. Here you\n", "can also choose the data set you want to use. Furthermore, you can use\n", @@ -462,7 +464,7 @@ }, { "cell_type": "markdown", - "id": "2a6962e8", + "id": "8bfdf8aa", "metadata": { "editable": true }, @@ -493,7 +495,7 @@ }, { "cell_type": "markdown", - "id": "4c047e8f", + "id": "b99c4ae5", "metadata": { "editable": true }, @@ -515,7 +517,7 @@ }, { "cell_type": "markdown", - "id": "d7750944", + "id": "7a32c59b", "metadata": { "editable": true }, diff --git a/doc/Projects/2022/Project3/ipynb/ipynb-Project3-src.tar.gz b/doc/Projects/2022/Project3/ipynb/ipynb-Project3-src.tar.gz index 5f3a85de08ae26ff661515e38477a0a41ec35ee9..eefbedad38660b6e9391f58aa1be78d0a5904c5f 100644 GIT binary patch literal 194 zcmV;z06qU7iwFR-PIF@b1MSaC3c@fD2H>uHia9|^nlxPtcHu%O;ssKh+Ekm=Bn5kW z`v6@jZi)!`Hb27*!^|PuZ1-8@?k-pjLWofcQ>HAL6U9qC!#D*@389RpF(ZI5Nhl5g zt#{H(Zyh(|DQ(>dWrcb-H;z@+hds+H@XSAPtfhrrKDb&dP}+$?>kH&YH{#?-B)dTs wP-*cp8`ytPsI>vufQ^Sm!T06{@wn*ay^0CR~_6aWAK literal 193 zcmV;y06za8iwFSdbaG<=1MSbv3W7io2XN0m#XNzm?yBb?4;=!DUSQ2N*K)^pSID=| z52#boMIzYW<2uT=&DbpmLlSE5B3&RvbMk!-yL}LOZq%n;= zAnTp<(i_Y5cuHG0LK&gn&5dJK^fwryLpzj^0i=9;tm1y!q7-PIn0 zoRxr_NCC#in;w)0P6MRjtUIprJark79>_wOW@)RQ=~HhaS(eSy&beWSco1V`)^ny& ziYKR7_5@7tl3I)_wlgARNJpf*_9P0dQV6_0LVgSHGjlU*IcfZY882qxYipC|p@)M_ z5kQSni9Xz*%4Ag-7ie3h%%6Mq|Eoin`E^-|3X)wG9bQv%n*cQE^?tNC5|VDUH#nVp zxXR|+Y1iQAd`wRY2g#6FFsiFjWlrE~OPn@WZ8#Q6_`u^!p80GzshL=ROdxJAp0{k- zyUfCD?r>9 z%xOEU3LrfO8UcKa>zswhYT#sYibhUfZ*@NyB;R(AJSk&!`~HTz3{lh-CKIfTbW+BG zF)=uk>oB<+4_uB1Y_xfh#$YVgYOZb0z68X`7+2#l!QR(bU!xws6%u_O_zmZa&%c(i z`7b{r!)KOJGWzOy-lGZ}YuGKZI0hQr;M&2PJ{v?>Z2{o0C4G)E^0ZCy5^x7TS+u#b zwVs8TDSzil5g>#TF|F8@ZkBJ8D42nrY1{}#j zcl74DoWtm(1cBmh;8@2I#!+T{^qB0D-^~-g^>KOz?Ns2A1l* zHhFF)lA06z14{=@Q%+FO(Y-#qRw%LNs;v-e4hEQs1xg|6YSghi`P22MD9aCczxlSU za$Lz?OG)>syUGvCc}Qj6(lTb{&?ZsBNhO6fc{Roz3yB+H@<9|220~ODE9eoYupFA9 z4OKN%^bN1K7dvqf$M-xwr|N+WO+4S8aVvO*dX*0|ios10z%+;WL6kX_DJo7#v*QQD zWCEP9p(GlCiJrdLU4Mu=z9zu(=hz1s`lvD(y6%3eNI{EY_k!ls{tPj? zl271sVbY|Cwo+K%(XZ`HFsBUEPvtI}<{3&Da`R`k5I+7X?0T=q-o2ED%;cRKWomMb$8^bWL~4~*4s@iG|WDmx*A zY7e_)AUI8HMxlmRW8N7ildj=$q5}3k9vwA=LwP@z@FS&IC`z7<-{Xe{oHy_ZXuypg ze=cvDypU=~emRYPmplVA(#`GrhK+Pwlk{a9L!)UzvKfmovQBZUW*kN27&i zo1-%`5tf7-opxEW_cA5$5+{wc@00A&$%%0H1vJm@ZW=_U-pGLgI%@eC{{lj#2`Vxy4+LCDIc1L!y$wS z*n)dO;USyEQ-e{|=Vm$(IRJ!^XX_s`kBz65v?gf$^@$aN9tjRo&@Qb5Od}_}v&Mxu zR;#*@=#XRRKZZpOPE1|WNTvH=L|GtamJP2TNNYOh3Czy2@kGZ*&u7ui2v%qlm2DyB zLhH6vLU=k1a@cCqAnr$a@4dL`{8O@nh|;$YPH5^FNtB`V37e~9Z-7tHO|3~2MNlU% zAA?%%c+1f^5HEPC6F>CZO)jit{t)_!oSj4T<9BV;$MfGYe(9Z3 zckM51qgXxt{BKJxadW1O`k6#DBJu=<^Ei^AeRQ zI?J=Z#ol0+gTrk5_5+io3!$Bn8_+yIsew>_3ph*och`-IiMoyG*HSxe!nZANtYPZc zEV@<|>bEr9)&y3C$Qec;;c@k7Lu^-g-S1UFLopjpX4u&!-@zvgWBD2WJCQ}C;z2mQ`>Sngb~#p*O`SyPyv>IAnx&&BhXYFIns&Vf-(KOpkWJ?F5TC&VmZb>)Nx zu~90C#SkI--PCuVf}Qar-@*5?lJ6jmz^7W$gp`u^yBlPC*}3u>1~eeVHMfT{eAQ^D z0dq^(-j5oT;VDq3X~&0kiB(*PK9$CMjY+^uaGV36b(#3UswR!yFAVH$aJmaeiHb4jELOnSU=6iFNf@n zNyq-GM{uNdmi6}XGLC3BbNGuVRi0U6)Mfzx-SN|Y+{OKwXUC8S+#_3avu0C!%yow#JzwBbUc{e)+G^A@jETrRLxDw(cSU zNO{ivTPF23kBlDEkNu;;ZxH5LkGvFeSZZ({UY`FEO&soc@_Os-RqYI;#v($5*G-!I z*!drB*=qDJIy-lfl+-BI zj(d*IjrpKsoe{R@Jli#At4SiUnHkWhx#=s zYI951Y8R|tzYU-|fvLrEE7abS`M_&=|D0sJ#ES7OwICW`{CcQd-ynhe6p7ambObqm z{W7MjJr9fCYW|InFw@lGqMjKfn{Hxs!Ryc56ltN?3xa}*yu4Mr%g{iL>CfH9#{)dY z8}}(`s91#-4))5VuoZOFPa0c~IL@MpUB}h_F3Ey4&7roro2Dl9cQsky!Nn#x4taE4t>XJ%&-{iVqF$=}Bj0W<4Qj1uA7-4B?`L4imSA z$a^+KDNoOkPzI%FG7^EBafE|opLe^WxjeD`#|m1SAf~A@A8|JEuDW>Nq(JwOmPGfm zP^MZ3YbM$WRyWMx>N$T3nxWaNILh}3LgnS>_DQudI{Gbl7 zs2@o5&#z~Qrk|z^fc1JQi)oCyXc|Z-FlC2qMD&Uc9l3vXjZyAj1Cv+f2p+;cyZV%wmmaz%ld=J=GvY`_+22b_`TgN!mF(PSq$;$0+H zX*biA^JQ;u*=$`u96X8_W%J=TxriqDJt|MA+uMZ!VNVA+a5I!a9>=mClESYfR6Sk@8bEgcz)N$HJ1*4~Uf4gQLvSKG_H*woUv2xW#I-n(ICREOH{ILUhZ zc2m^^J_wpNEh$#~jcjAwo^0!btrenHhI(8w#fg$DLWkC{Q6VM);pRf~8&Pq+)VDMV z`6myE;vbx4prGg9tUUXLjd?d$O`ox>I!s0HCmuN;xfjwrh`I{f%QO0mu*TidH5~&nr1b}7 ztz8(WW?$IH%ItY@7Ih3(wY(tx$6#LpukoIgG6T8LtMw#yZQtJoB0n09KNd5-SWp#a zbzlPs;OC_7Lh;Isk0>Z{Lb=|sUxV;F%G)YN9#p4)$%VN1G_nR&LQuV(WYIVy3o9xt zu~ZKO0b@^+vtSokrdo7MUIZ*otg~=!Z32g^pg$Rv?0?V+kvnTmq^}kW%-rjt{<-pl ztM7oi-~88l9H!#sRAcisX?A#giV?{PPS+-|qUtSOPi<%b5l=LuL0)^%R21bt&-5Jd z?#HkTjR>=nt)Kp)8=#4t4_jME5YFl)sv=4oK#K|dImTQRo=intzZ1YaxzgV@i|RVy zO=jXQoed?A+8aAr{JSh0!#{Lo@-K=RPV(!E?`bP~@1kB$ar&dk;@d$)<26NlV0c*2 z08I2qOgNOY;6&VD^^s^aww)nVtrxBN$HrhZQb8j)=KU9FqzP=0piOTr5lxV0DQgl! zrcYAInX5mgr)jb!v$e8WBx4Kna_NGMQV)tb=mjy}4tY?nVdZmwae~C!QyN6gHk)nQ z$MR}AoxQ|s`e06B@evTlbgt~V@~B|hfX`{3%;Hr0c@W{EYcdM&`nJfa8=JAiXqrf z?>NWp)%}?ZC6#6(f>LlBw64Q&AeJ*xI~?J#;;u~S+Fm%E#yy@wC;f#ZC^@g~8#*#% zEjYQ3TSfxyJZ-bO|Arc7D*u>}EH+PQfMuFv{*Emb$P2cxo^fXAuDB5!mc&J)zIShZ zVczFWP@+?G-sd9dwX6U_1U3Lrm8T7lFOYvCt)pV{4xO?<&83DunoRoYS{ZlobHronVjrB~VQ39m5zDDqH_GBGi7pL*6VQ+2f?h1c0X41OG zKp|ZA+H%G5fGoXHLGI{P`voyB+hj!gv@tEQe->GCwS@GMqb@H_Iyw}~X zZq8V9*l*BGR?uME^LXrf+oj=|06K`DTTgYUyf#@uYpzc)z^Vc09h26FKsP8?!yr)- zA`X%JkN!SXUm>eqW`BycPBVH~a8+V}S7Nw@`>FW)hfU3U$C`+aoNrj0hueaRlq2kD zL}Q`8{k&WliPQ=zld%#3au?0>M&$35y2`&kaxH@ybTy#iRAv`aqbRIpz8$Y_<3-)^@sM2hl zU9lr+eyv1F zE<5wJ5lidH(?&2W|%k+A5zqblsS?}*b;*>)UOw0d7aAM@G7 z=e723N>jtiK_}p7juqb#H!%QS-$7TVM3%`8NN%#VZKSZ97!k-DR#uf>c-NnJ+iY_= zHrjd}FEnQ}8h$#p;R73_yH_mh-ZZ}_xzfl!ZcjC^{g6MJ|DEsexg>|BN7tCc{<_%L zmMt?1Ah3_?NG4Jr>xu^5T1yjvcqLD`IG2kE%-T~wO%5}SC8&x=3kz_YOm*m8T7TabUdn$fjGreoZFwL&O0kND{;EajD?T%7 z4aE{oj-rvxg(Q02T;m0WKi$|_4K}%;DzomeI)?7Ozyr1l#qaxfB8hFXlmpWX;jM1+ zmO+Gnr;N+sbC!jOtg_I*H*C8vRzr@`ZO7{MzwWj=zmAy(LJg}M5XnDTfHXK@QA0;y zmpU>wW7(s96)zn1qFtGen*4dyUAkI^wuV z{X&$r^I3pF^QJSHyOrYUcQ8`V(o>VM5%w?S2%+9q%W4#t}HD-KUT~OrKrK$ zEo@4H`)zd|Nv`}Ba!(VhjFQYTadyZ#Mun#oXAJ0F(zLG)b~C?k1<$;Ma5-D=fBB*;|f7z1}2b+i{`mV6^mW6TvbWZEu?ELu+z}Mhplx4bJuwwn4Zqf zo!Q$sai>9vNl?jB(Ep&K3RITxMjGAZS)4z<-87s&{^9o;VrH!~??_{(w zb6C7**O1QBA(5|X#FQ&}R)2Lq=`0?Y2!LV?nVlNZLQx&`<1rH-mHw04exv^zR((Qr z)#2eb>LAHtp^$39l(+U!QUaX?(d^^qf&#D)6?U?~P+>!nR+PDO!4#Wh8(Ung7UQf% zf@(0xUVL=$lZC%kUHs;pGNDcbXP(nN*fvM?M;FfYB@7 zk4)VCqu^UR!&VqEstwFIeks_@xX@KXbV@zQ5>w9=H995nQxD0Bo+Ym_Z2eSKIorT7 z?ZP~j2^T%5wV-y4Y6su8$nqm{RhIv6q;>7M{?4lI`ebmsTDbsf5c#gZVlsGIond zijcUO^OKG_=Cxz0CogeZwI7xw>myN&JP1Jvi}HZ!R1T|E^#yxsj>7KA1~9jJ{I^M= z#IOZJfh@{KyW(SGHFl5WU0gK*84~>Ux@1xv9(vcYY-1X~l~&Z?CIXk-25xXBX52@Q zNzsn|V5dT}%L!z-#MH%MUZ3_K39`?eZd*bQFeI1<-^XkX$u2Q>F?3muN zW(Pt9=BP&BNtRHts6!gRze0E$em0@6uEp=QMzcQMkuVs}u*ft={TP{I*sX?=f0IMS zz24g%BK~t~XMo=``&m-+MEjGpbL^!Yl9OXx47rE&3K$-5xCF?919!p4qCe3QdSi;CYl>vk1>a~xXo z+?d%^H<8>e5z*wu6qmvKz?5Dc!&cWuyY!yzj$3TEzls32>x+~h@ImZNa_ihVAU6WZ z(>mvQc$YC={g`^z00J!vy*eNy@DkA4TH(UF`i!INela)-_=kx$;donW;l}p#7xcNt zWM8=s(IeQRO6!__VT9-nIcwkcSgTZVjLrH?)yyq-@d#==)_vF|soUaKOo(G)UWH_`@frA-zd)$ul^~*Qoa47P+lkD(fUc3YDz?uvg2(>zIuJ~*zYSeEG z(te_3`hRYrsm)zH95pALQ4Kn;EFmzcwJ2)6Ax~C$TLWckKZ`NRMdj~;Gn$7M6S%c| zjJy@)IJBctXxSn&pgQ&!%cfoGhN(*M!;1E1i3E$d;+igj^PS}?;7I*RJX74~je4!$%&Eob+{5jis-q-^X5JPN{CjjxVB3)SGO56$;Y)6ZJc zt5Q=&!L-bD7fSg%Rqb-36e7x)y_6T{SswkCn(EB;bE8Mm&!rZw4wq@bFkO6Iq5BDo zk515Y*=B-S4~yaAa`KtQCgE6w!NjK?{;eRs=zJ&_@U=bI-N5mFWc@T z{b=7zYc7<9PB18<7X!|VkpIv6U8${y`WRV_2Yud<|LU*J=s`i<>Jp9H600M)m1P(9 z3oh^XQ7+6$NA*XM^2Td#{5^-ssA9M;h#(L4Zy$OMo;wMawlPVKS^@4cI-K90oov!X&wD>1qIE0=MmC zP^6ZCUSq0?PW5T#H5aI9tW~JR-f06OY~YF8k1b?r^KpvV2V{~BVaQPVLU6w_)cf!n zHA#q!Dc}$*T*tzSqnpSBlFmcsD8O$BM!A7EDDkGu)e?48O^DdlSolLAM=hD@*qf(8 zIOUb+B5mPq*6*U#n#Zcn$X4jjh^H187EEkG#sHo5<@jap=giUPO#4zc*R}>FNDP#g zVvH#M%kb0COy5yHK_AJn${_P^Ua(rAmF89gkM{_?N*z;LUR>9|KW3E} zYALBfYBY&_IE}hvJz9FWxbTSR%bY?by}K=E4rc642DVjtMS)4&&g^F z#Q_6Hd-+#qZS62>*UhG9`n*21&EsWpgBFjL&Tbn$ork4HljR-`vPMKQapUNDn^kZBBEn^-4Ft1`CW>oXcgb+4< zn*IzFeiSrqbp@DQOnOs3#GY)f6mDptLe+E$4-SOg( zAYX2y(klJfw8?(`(L*~UJSrn5kXCY0zAG-(tI+B{b7KQ#TSiWMC`7|=hZ^tw2*b(? zqe;9}fuYI*K5HeEk8JfNE0-!dE_<}y1R4xWoya8sD|z5eL-y@S^D~cHPod;rfghQyVZbm(i933RS zVow@XE5tfX2-<`;%$WL<5Jk%hs*ef=pYEGgOcD2|o(C!(+9w6!IFo>56;X;`-cUA_ z{N8iMr>l|1?tMRu7P&&7$y(uu`-s26MrvL`HSWb<1Kqc)xX;=aPflRTxuvAK*Njo& zGtr0jhxSAFfU(A$Ht~U2coD$?2yc4G{Kys)!^okoo)L!LMMp*=8(&<6(&8A zC*)IGGnngRhfSd^0|;=71D{-nTTP3qn`N;)eWtuYTyCsMKmLmz69s`)8zO7`X6^mZ z;?iANf|^1Vo-Bf3UWWuK5n3lZ&YOo7wM=L9=w9@GzSP+QHzBvdjrZ!GIRPtsHnJ(l z50$M1uprxdo#hyHby#bwGg+_$|YFd$ge zAm2!nbLmX|d-7KYA_SBQ41rcDfuD|nLb$y)kMB+<#1zVtH_5_ZXvzd z$?HTvHK=JUaXc5$p=2KDWp7jNT&TWRA?&;j=12mI1bxkjg-b_^272iRp9hLt*;h@Pne>JvA`IX#GQE5C41~EpKGHj%Q&X;f>-`!aMVEF*Y0h*|* zZ|9J_NF#-}DJ~EpzU#-)Gu6(|csCcT0o{(-)&A{v)tKL4<1JmUeUL8B`(GGq5M*QH zh(t9(?H%RG%BDDnj=ILnb2+Wtj4K@t2;CobY_dtPlc~!+6ELfOx(dPReRtKqw^7yp z_Y@PiJc+I1VL$0jezsvi-I99x?%(}uhVTafP4~f>Oby=qil27d7o;8U2XpqRmrqRG zXJiQZ`s)`SSg4anySQ1&-H6i3u;E=63gG70rKc*A-r@v zjf%D~0NHQq%Sk)btoJB({nHt;jj@=eahEb$hO+JjJZ2LaI^lP+xlrdCBOE%(ePy}4 zCpY9J{tWCMB1Lq z$rYN09n_gh;p(UXDA zgGywOT*i`BtH8<@KNE|UUmj{LJ%kJXYmVdHkRgcSE^hfE)W(pbIYAU|k!*O;Bm%ql zpoxLE>gb{XQ}DgO+`LO{zq#rpbP!(i;(%!z&dsY;S?`1in0;W^l~%3vgQ1)ei@H`l zso^T`$(M&j1IQQLeyb{bCtw<@gz}bQ6Tw(1a(2cn%vk8EpZr?d^ifZ9+Pc1%Vt)zC3o85kNF zen9tj9q#}CbubtA|E+^<;RFQWm=!D>Q?AgV(r>sx>402a`GkWO)b8ur8CApyYnddU zLo}ZHWskbRkPDEY-APe$m(om4c|x^2uP^^f9m(c5)~FQu6ozCF|3dN;V?@R$9YBD?9474bPVl`dOC0Sh!-J0lw+r*FLFZVKywRUgMTDaUS z=Ay*aD}Xow@O~-ZGE*P_Gcn!f*JQ?z;MCDBa~+tna5{_YG-R+<-{N8{9Lp=Vv^X zS+-SpnK-`{EJ9ACbn3=ZjeqLa8my^TCTJ+u`NEdOW*1F1oZVBxt7@*N`S9rEAy?|} z1RUE)?7HQafsnBhs&;yah>&WQ9&zvUog;!2Cz3gsAm@RX{S^K08pSTbOBOvh8b9Ly z#3886ouy*U((0uY&It$s8TeSX%9d4S&c>chW6-q{LhC;WII zDu_|DRAZ+(IinV1lq=|?8CY(H&}ng%r*N`7`!f_8qW#}QfN27EBeukH+9|hCFCq4H zri>Biz1iB)Yo=2g9va+TY4wew-m#HBxPlOJ2EUFf$yOLT_xG3w6)r}36m12v0Hz)B z!;0ceV}+H`s0|-HsKN=B9CcZ0k;n+v2?ydDu=Ii`^1<~1RAvkt3RD#6_X9mE#76v! zWq)cYlD-rU`83~ zqAEken26V2_A%dHD8Co25R;A$HTpvK*Lb*Ay=l4*^Gg!}t&83gSRyIV$ zomvR4y?2#JvuA2%o(*ewI__d0!RdQdU?c?Li&%?r1dTFQIv;Vw~s~``$t;obID-ZB-cit@d zf&+y!X1+9ftm9h?_;S~@|1ANm3`eO(qvWg=~6Hyr*{A;)N6TO03 z)*q$=g_(Y1dyd1L&igw+?5i5X;6xOew*1Zo)Cew1`56<0O3__?BRwRmxx1!Cd%5<= z(LoVJsZcndoACvtHuPmPJS5rs*K08-yQ7uY>jh|m^bx@aw42C0qZbOCFN@SC-@IZ3 zI8>}q=$NOYPP(KPj3gL@c=c|e<6R#Yi+Y{^0`!EABlR-AA;ALxa|zG)Ap!$ZtDPh; zKdKhg=nrO$+zC^9S(u9mMHl&vLg7tzgo0UWF}o5yc?&;F(H^MP_@$f_cOz8CYTW%K zwdqU`lc($fks$N0C)cDI?(F?^{+K_n`4ir$$R5r*AA}ygGHv+*3Y>!p)*B@|BjR0c`LSg6?QB&Q&uH)?CoG|4DFN1_9G9eV+nF;7e5q|fP;bPGzM8K`6*YzJ@RH?4t$5S`N z>0}|Ayrp5~_pW5JXIdLbo@hDju1Ul8B04xJ5tiZ4^2tt%Kxj*x{7qO*Y=XFt7^{4%ix#vjWXee3%HfF+M5>( zUC#s%ldiQ?*107E?ly~6L8*zvn|D7DD4x{BhTPPXDGOp&b-)s78Pm#|+&IBvlR(AL z=!=4b8++GK%>;i*%5kUyULX>J$CU+tPr`K*|1-{EG0?W7Hnz_5pd9$Q1+EefU9&qs zL``mOVJJ42MmY6jgxG&d-kET6%v{V{LKgr!r5^<#l-HfxYa7@RshJoA zwE@Lh#N+%BOK6{13-_ReGvS!yeLMlOU~@^!>bzd)C+Vtewe)qEQ(0{{OeC(5yQM*l z)98q~%o-%hD#wE8kfyDIHP;|Ixe%eV2o`Rv#38b_=OA_s@4@W))<&c(5>!cpv}yp9 z4+3_2^f-N_2j%g2sbE`y4)L-m=`PxmbbmyGnx_8ZLFS*z1EC1`WX`rO<~9uM>@7_z zRVxck3f=>+X3OFLr{&5ripR7>&V!P^_#ugH{o z9a^(I7d_SCwNUza51wUX0Zwyo_W^)exX+TSwf*#r`&SvW5#gyx!T0oAH*}ab93Bbj z;rAl^>=G z9kcpfCn?hH3M%`nC4qD%_omoFu%|By`QbX_ z(AfC#G&6;vk`W%UN(jQUZ#(kM0?ROh%{HSbadcf_gsTggG|?}iC?U^Mo6++1)yPl- zel89>5tX;PtL;XhfAI5lYfpFhEXzT?Vizo$ z3}R95V-s`$?uupbb+z=Y3mgD7j3OTQ8$kMsN!Md1fVpgkF+B`U`{tiGh6}Mm2sl`2 zrVXI=FcWZPy;79WL5-P4d-35EzZ<6As|&UA-;!I5QR<05czJCj6PR@-`D^w=uB2V< zJZ0Ga!w*t`ku>#;ZQQ-_*L-$smuR|SV+;Gcfzda&F6j2-DrUnx+Ww4mJoQ^j4nx=shJ|3?+idh2m|{J}ytxNpgfdKL-PWe|V~ zRv>ruBBN$2lDB(R2kjZS1CrIpf=2-iqC3l%gtR^D6+4+<(LA#y`C4Zw95U| zhb*WXCpQPAPB@s{+ehaFe$j1iTyg_3-7-n`-vQ72oi3hL$%Ca7Oz;KN0De8H4XV|v zDFFWtoSa#%WnadrSv)kPNiHaZPeSdNTDoztdI;bJ81uhk;~XAJ*OXWjD>lUr=4Etr^O6qu)# z$Qgm8hI^#d_!E`T;XwL(7t)Jnl#EZ7>;U2)RB2WP9-^?o9Z0GvQ02QGtDh;|IZqD? zO<3t{J867gNt%rB(p+-bF>h>p{wJN7=n46OKReB!pi{$;%(Pn9`{IxFPXj689aOp) zCByeFS1g9_cUPaC|YCkKlrAyC)F zCK@w?)PD^6ItCR`Ky3U}$HG+0%q+i;`AOHm5ea;ag+MQkyXX-2a%CRMatzP-$M3Be zrXsX1n#nPq$*F);AK|&3|3I=;jGrAufGcCZcL1pilQlHezlrGllg6G4Rxl?)ntkhPIvOG0IH^@gQX zw;6=onk$Lf%`g*ls3iKReyQS8zw{+T>ipW42>;mYP zK+-rJpoCKU(yL^CP>PpNa4IV)+)E10{RvR8@%HzY5zJyr@Swjw>K8U3b3jNEqM+>J zpfhwkCOHSKeI0)x7sTiM?7HmUWa!*$oh<)_W|JGI5)INXI<(#-InS~wq8>}np|dtx&_1G&V+ zd}4r-eTsDKaY?r`4B-Wops(uBOaK2wOf9A7b>VMCny3&zu`^mdf)$ z;(++mAm-=jlHc9^>{9EiSDFsWz8$R@=^f@s89PxO1)h{b_r&vnR}L<}c;T;EGz}q@ zlmx*M@>QwGRt5Fbl!OJB(wIrag*GzGRDZ|6lx*8R5o791QY;`3MhW0KTJwI$BhNNf z5MfwqA62eRbft<>IjBJm%m^bEGckk!`-7Hg$dgD++}K>}h%U0{Ry1!Z!@Y5MpB*1| z_a~WSN^iAs3iu%V{=R!Lb$87{>bLRR{IGtOna{0(Ue;ko7{n<-o{C~%^F^yjmY*cp z+^{lnVJn5ahU=LcD@s5-?;doIV!Xgz{ zVTj8KewZ_wkIc!zbt72L>6G%5Qb}tt!Si47e>uXBpVGGn>9q&97Lj0+g)4gg7|QQt z^4nIgbESD$F1@vQ;Wdw-N7Vq(sKqA85-IJi#l+EL%`7>38=M=gyoJFI#PISo1mhTQ<=M?OGB3zt?cg=Pq_<7 zrq2~WJGNB{VOXo?IFu~vL3Pk#KxFd@$&AJH3#BvMaas`Lgx8fh0w-8DI-{{H3e8Ye zM75hQh!7O}0J>uwJrY3d)nH9aI6#QgzqxGU&-R8cUHP9uAFFLpnYO8u zaep-t?zG6qZr0FCGMPvbU3*=f6!`^)O2-8H6N;}BweR4MX@6JlpTVn(l<=Jv0!x53l=@PJ)?-@O7fM-l z=GgP@5ztIIyaPU(u&o=swNDlq- zQ$|7hrP6?phr$FT({GjZ4D|D>19695^21h|58$g7Kl1wmXr`&$?{^CyM(+`AQMJ@r zpn>cx)BflqAHkB}P@8|XkE>`IXX}|(D2BVkz`RJDtBZgNaH<-YCd2cvq{lqbCbLny znBawn3_U>*dLk2O20`uBC(oW8aaBn3YmUi*zu;aSBk4h4jpHIHypX{(ADg9!K_K_8 zZ*^wKEg>60wkgSCnV;Gib^un`vn&sG-N@FB9kX7P9W{De=)|~IrEmLnAkP6Z!4$b~ zf|XlCrxF0wi&QsuOP$OIqIy=3%M{K`=tueAtKzI1a0-;6&K_g*Ud`MG^I{meGA2>Z z(9v)G?)RPLR@G^%@Md65Qk(~Nl`qX-Fl2`6IT zpc?nRCSNXjbXQot2W911^$fw=A>^OceT_7I>;NH}g|ORxZ-sY?6yndyAabxqGxe=I zNxK%fgy8{v^^kp~@hKX8cfXQ4Rcrf=NE>;kac$dB=U_OaA0S1|&VL`}rPmHL{bi=1 zmvHW7Xwk<7yPgM&H!OZ|4+H~Y_LC4>uSiKIW7B7Ro!`{{hM1@zzWaX*Orrkco6>e-p9^I*-nD+#;kkw_SgHX zle1`hg`~j!wXE)p9jM%2jYKgEyzXE~OKXxy?bu z`O8{!Q);aQY~4HU_vFw_-c6HBtCZdrnttE+)J4=Eo#NOKXy&iZG*T$Jf$W*E2T-Q ziln01H)qJi`p(2E7b?e_0(g4Yui>Buc&WI?1gK27-0U(fg-(Azr}Z4&OO;PD9G>Kyn8ld-*jW+UigF_w@k`N zVLhDbKD%+(STpe8Xj|V6V9=b@$V;zh+I??8P%o#O3AknaLdl#({_U$BCDr!X5 zq>^v9l}fr{ZYzKFM7_cGjoI53`_x*L>6e&C$29@Ij|WK!7}B$Y)XGUR6NQ#>k)XF^ zS_xI-V*>Lq%UDaIgU;dEwz(-$0X7Fhp`rf4`TfOvNA>_E9M_O2D%IQS_oEr*O4g_DQ^!@XZJXC~%u0(&C$xW_mQNVwI0hm>iSpkv% z{P+_&^u^o3Pi#=FvO$@c#&^sa<~Uj=AZE~sc4nN3-JkXy=)6o0mbZ@Gv<`3{JZD7_ zk6fP1QWQD+-^*a%|4Rq}C#cH%U1GZ{beVjOn|A;bco8C@Ox`{92?{ttf8LsNV?EE# z0}zb>G-!YM;oxx8G`tU^8lH=)p{Q0@fzzgI)+jU2^58!~So@1P_l;c;lN4FG@YE%F z{k+T5lr$9Zdt+;K{lG-vb6Elo9Wk@iwJr-G^#f$8BKh8hXma_5x&b&~Se<4u62T~v z;^?XUw#~t_24}~}fiM{K2LIq{B7{EJu{S(=LG-S1;7key zAuxkrelNu;E6#U;>{)%kiI}@!?<8ZEoF>7rcc7Uo4cqcVNC3k|8?}<%u&cm9fMdrm zJy-%L@U3n4&iaQpu-Q~q0d5q%!Q0*pW<-AAE-Ht zJ`^Uwqj_0yh6{Q^u#2bB_g5EQ^GaQhne510d-@NGp{pAU$|)xuu7OaMWyrR|YvwvY z90xVr1A;WmsYq+i?EU`}a8ZI(=*@82#O9Bj_LYkyN@2B9)+F9nt=V^wM=iZTLGgbw z&4r%x^iOmF{26ou@1m{`SQ++~G3P=o9A>;Di6r3&qB@n$FqK8u&ypV>P!a{#kRM1! zNk4R_!JN+_O3R{38q=OG07$sj(@0P-O${fz^Lv9_ivtHU8f=w#3^eLw4>8dpqD}#D z3Tv2y!@PBLZxMS1`i^O?quWD>hy;IKFU15HsDjPUugWp?v4j~cV)ULy0mwZls69PV zpUV?y8F%(P*6|5QiSBr%-pmDQ;0&Uj?AhL&JC8Y8NTQZfLawPZHGz~8$m^MFD}j|9 zKr`^gRA7;TNFRge_%@Zf&>$h$kS~Yb z>Yo~LvJa#HJwnZ!8v?OL5Fvl4HK33_NMP3F8uJno9?fr1I!%s)wYSsP03ZuUBRQ&H zfJm7*xx5L{VAwmuk+}(e3-yZ?zg`UnWZyC;wU*(oN4mj{qvfFP+Q;BH@%lf$=PGQl z3m|5yysNy3-m&_Nq5qT2(l3DGvRaUsJFo|QTohX~z$H(=DbSA2x$}RFKR}>oEWkrq z3FhI#zog3Ycgj}%VY2;jeWuR?oC-XILlwdK|JK^y?XXN_r%wYi7vLKA$a#LX-rj&m z4NC0tziU)3C{Nyx)g_NW~0HeCNi$TwbW{-DrJ?93_fz zrBO)zfLM(8k%ZD^O_P7AOvsopp*M$5PAwFmmIDlB4k-jG3O-H8Syd?V@tafpsubf4 zvabh|@jUrtnw-1A3i?X<6@}*69{$vYzx-K+cg4hIs{B{;z6?JEcR> zHwrIIWo~D5Xfhx(GchwYmp~%{6azCfHNj zT!Xv2OK^9G;1GgCaDoPx5Htk1op+zJlfBRP_YMYh&$3yos@7Uh_kfCAMV(R9!pRIM z<>UxrWMyLE11Kn}IXRd*va&F$0j=EaO8k*7!00~hg z4M_k{ivb|5sjl(+uLjT&oWB(VKuH6v|IGt7{OztNsUfPNt0KwD{Cf=mR)7c4)y?KN z?SJJ)4bBYkCpECEm)S-EAAf!t);4YcPbXJ90Ql7vXb=3SFz${P;Btbjf&U8dyDI<% z8*`we8}PT2l+(YI4&YLPoxpO?zY_x&0`xnk{lClsZb0CF&{&(g{o|{kqM`tBFtu?6 z0Ub>p&B2BsQ;@qGz~mnp_!nsL?q3N40pjkiuD>}H|5xSuZ<_y2U4P68yfOoOUw>22 z|L&Nnqr02WU$yzqWt%%Wy4kpa-2N322(YxV2ma39?RU>?9RHCiiYm!UNvdlwDu4&i zkx|hJTpdRykQeA5>)-K2B^3AoJSecenwfB=xIJJ8?vZ^eHiL{@Hq zg^f7~JnG=>g7{B#Sw~AJ0PnxV;7tBk^*=*E^UqeL1@BV}Cx1tKZ-52R5|LTS2?QPq zn*V=x+JA9Lx!c<-nK}Sz{!dN++tAd(#@_osP5;Y88~8h08YL%J2UGk1(%HC4*?0jh zRBS-z*8eK)zvQwYQ}FzYI$GHS!EO0Rr1^VG*@I^neE!({KE42qtQ=hbr2~(pxt$}> z%?-fuFE=2#5P$!bF}V7_^93-gN~x>JXwm=QvH8bN($U<>!p6}Gz{bf1Fm-h`^+sd? z&jK4KC%~5#JlGaMuYZOGz|7?61OmGNoZUhG081xV#NQ3&;sh{@{uccUaRHdc{vd7u zv-lsx17Mc;gLnbVlK)2BEC6PyKZq5;Ed2+u0hnd}Ab)lMv+N%PW|#ki!0ZZt5SU%@ z4+66*{Xt-M<$og{FuTeh1m;uwgTQ?1e-Hq1^vuplAV0NuP2+Xei2Z7mj{va^B z?!OT)*wXY50$ZB>LEuW5xq>Gf2(kxSg8sC=Kn^V zU;=X|dw=j;{jY?B#D>$?>nH zu!3`U{-Xx_I-7!Ts{iOI2kZYx|Ffguz0d3nbhUB%(+>{tx}EL8ci^A+V2R70cKk-} zK(~KRkAHWBT!` zTz^V8ds8>-ze0eS{t$427Ynj>1^zWv;G%#$o&ItF$8i4(g4^No7X+8s^RHn7dwcx_ z!Qs9Cg5dJ|{7DWR$p`5AFRK4MG*x~-IQ(-Xvivzk|L2M7AFK{?b+Q9$+gN~~CjT-~ zGzITNFMSs9b;b&ogMa?_o#B5FQ2n`y|9|op6La$NW#jSB&fQw+2;R8%405 zrHicX*sk`NUMuK?y+kXsy^x^SN9=@H(?QWyx&@QsaNJ2bIOlazfCGF}L!Qm;HyIu878ZaE9dB`ruK99VK6ugS3xYPPkRKaJ!cvk|%!}og^SJ zr|49!>8B#kR1U+f`@{Bg(~+Ko?j+*NF8o95stJ5_Dyd`6X+IOxWJ&n*1%I;~q9k8B zO3(zPShMmr(LJ*jx)C&nB0h;NZz8lF*Y}!BsiV*ed=`LMl)M;R-cqD~MuYT%a=)wO z);)ad0mLC$P1&t3&%nes7vy*J&Q8ZpQdDA{rPKyGPqNS;DW?q(kGK>P-z&Zqm%oJy z-fsDX9a?uKeqNSgH7t4U6@TT=9(HT`^>A1mBSUJss3SEy7Qvi8Y3vv|q~`XFgBFbG1Zq&~Fb=!T(vpCfS=E>h!7g}dbbmxZ$cmVUK|`|Jo1 z?p;o|(T6wcC@Lo}>3{vUAFn+tO0+%jx%RWJ$;&p6YY?GpXD^djn;ga2*vOQZa#dS# z)R9gIn?c?`i1Hm%52O*1e!xIKx03Y5`ojtm8Q;fOe1iNj?MNE0Eo^i&7|cc_+b)FqPQ?kW@ru@RR3d?|*q;gLXwsqB?K(r9*GI z)7c$)_RF;j>UuTRkkq5C2~82Y17n#`>6^i23E5ABD7~BPK_lXMlUzr1dak=-y2l>o zwA>;T(l4k05)+nqf~}H7W&tkZzNMM_P#UItsnRj8>yIMRrIYKCv7xmB4~iPapE?~w zUO&LW6q1e~y?;lZ(fcL&ELL+^uiJ_;BzhO0(Eu3g_u&~{8=C@96BN1q+7L5p{ zyS(8CFn5m3boFfxQ29(-O@2#GfSP9(L>*Sfd4qy$tcl%R`Cq<~9q4LVEB5 zLczy9T2bO#gx0$G%tCier8LcvPc%np7!D?8xcd3qUC0rLE%J@lyc{iyVyP;{#QBAy zxqNx~B7fMQ@pb}`i#w9=L@_d|Xce=tm2N34x-TKpa|=qOLrv#cb8p&TtJuG}M3o7? zbVq)EF{dK7^v1k{<+O@yM%Q&=e3_y%-#k(6870CVWIgL~8^N$9{8At$mS;%a$7i_^?d0eP8vtxzFg1 zpIFJeCFk5tSgll0>b-Z3U9p&Y{D{R>(8~23CWd`Z2HS9bxHQoFD=>PrDep!2oaxS& zrf8FP4%zo&?GU@>`#XPcK`SMb8~=i+xq(I7TNZKZdkCfwTm;uQ%59IZVPghMr0Z{6 zs(-j7w%a=<>5`J(20^;CfJF0>6u)l$x<}EB`$?6D>RWD6i`bfmq@F9f_24Rx8dqR*N{~lXBu<_5KvUs{acX@BZj?T#o7; z5oA|La{6$eN|~pHy;6@a(t)Wmd|>1`c7MZe6zZic+v)C}FKX25HFN%%;JJ=W%&c%g zLCg`yyc%beO(`QD>JA-}!oFP&XFDo$twVUf7Y@W-)55 zW(_vV7Or6f=lDIW65M?T&Yjh&*%}aN%85KuT$;G-ras-~5Yu-oC3JO~zo&+lbbojB z=Cd9Vm*|3@xJs`-s?Z#tQSzvJr|Of#H+>=Y1D#AFKVhEVc5N6LVi##^?ALtOz0$Ai zXLKr`=Aun%TO&7FY-1eh;q19E=EJ`Fz-^k)n_c^<0jeELzrB8mRx4`Z%svPO&V)mf zMT(plJQhk>7Fg!#lX}EbRDQll9~Kh1$%EpthBhd+D2ioy&5T}te$jE!|(So_`e6};AGG+3-g?% z?VJ#vf$WQ|5DD9a9WUuszlCVnEY}MyiXXFG$CCBDMU>K(4p&H{ApyFhu5t+CiPJr@E-Csx8C#cZqsiNCTMt6u zZURdB6%vUqI5O4hb7Zh~0=)L4;_uE+ez?R77@rb9mFfoVe+Iv%;?(q(8AFh)Gi{Q6k zBec;J8cpdhi+?6vdWc_q`bjshB9Vw^vwqophgw57Bb}H1Fke2}YcA92NS_^Uq2L`R zi&=oiWWiq)w}=;Hb~M-e8tNr19OlWq`6#~%!bn*`G_-YKBBch|rj~nOh!DmYHWCZ6 zSQBDJ)Z%H9{p`lLy;4uO!{L@PGe<+{|Hv;;tZqq-YJa50_iLyie(LH;mr?EP!n-2wFGaZ7eZjz(6 zlw}+R@n7XcnL4-Opw}2S_^~g{Xw9{+Bo;|EiJKN~4f@9?XO5IOmg}zj6epM4qh?PhT4qD%`Kr!+x&p?byq5zvxy6zMB*B%2gyhRjn|k zp?}$0kl5esqXl6?{%EEihE1U4SyjkJDT~>H|CQR^vfjVL&;2qa_@ha9tpB`Z$4Fz) z-UyRR^`yH<)9#ZW(MI`&cRHr!{3Z&oXa{!3<`Ie%)9~vnjhjvH3D#WQ`^qyue`C(H zTblMX$eYLaoew+%oOF+~G=cRR+-9=;9)ImYvK+4^mvXpixAsVd!j{40L-Ry$=+qWY z*XS&c2G=4Oy<|MYjBqT=iQs+R*?9KCJ?f{9YqdfMHk`t~)mxw2B$7?=+2)HEvO?qD z2V>M$5mu%t#EKS`L&CAfAPgAN-iOW@%rFdeX8}lqRNx58!;mzDEu)FKnvGx@o`3Gx z9+@NF;-bP^ihsSX@HDwii(os`cv@_(RLz+1LMQ8aCtXdDG!!1D>TH`{lv%8^OZ`~u|( zdCVL^U;IQ&>9g^c%o;pHgK;+^7xb=W%3(s90^0t-I))%`5Xxl@3+-bMW2b^j(8Z!W z)`pkRup8SQL0#O?Tz+;7o(L2Ait=`<8_hoA!AhGwP%o06c^sujs;9Sl?0<7*o8U_E zqp6yCG`v!IM;kiTHKt7RKiArDdX%QfoZEg@&4X$D7-MJp~y1G!99~Je6v&+ zL{5?55@P+1oG2m(PHj1VBShW(oNvLG^~+mK6vq!LJZ4T^svRP^Wq&_&n6!)zAzZ4R z2X)%Fgo2WoGS#MevK{=o(c{d9$+o2d_HUSOoN=%T^u#OhsaHC3N9VW2oZl^RM`kYY zPQv;NspsoH2lgV&o(Jqyb?OK+`1bJH^#L8_89OGj-x(-lT9v273j=M!+r~c{`e>EI ztkvf#wNaniU;kXlc7L(NkE5R^$~Qf2Z4xpRp!NDl@9dp2s_@fMe+`OPLVkw87a@UCoqt)4wD>J+u&ZmhD4=ti zAwUS~9tW4y&%ft0UP|2u1b?rT6B#?L+sm+68eFsthEw02!bJ5m5qi5X@pS6c#15O# zDzK{>*7Cg@nXKAh^p#7gnU0-|;+bMQyH;#lDP!fZgah_7+D@kNR2@ETCaM{$NJ8RV zX%qasMl(5>(|-#Q0cWT78b>43Dftxv579$4aE94=nm%w8$kf6%uPOAH%Uf0q@p42|Z7AAKiK+H`Ci zdGlC&Te^54(Vm@%vCL{$rE_8-n|qeoLX{2Gz@>S#o|MhEF~e9@8-fn8>-Of$Bwg9r z?SCV!-A9SGq<6}`VK&Cb0B;E7(Xz6x@O#gIOlrm4ytm&SMeyagGKQ@8x3f{H4RYFI zPp+RhF5%q(-#N7N^6}@ki%{y@NSJcqehzN2FEpkb!8SFUV~|Nc^b*}XGOFjXrK~^t zDFu$azm?Vq%=YC+8UbSwPlDzRplx zO@LiD7gBsLf2CJTHZ^Xr8t3cW0orXuXo>^zYq{3YJDo5y2-DNrL>%s2p9TOb#DZa- zd-1gFIE+_gSiBL>)u(TM-qCi*4%XOlMwLdwt_R{mC=QIm%Epa>R%L zElxu7qY~0H2qJUfO+O9`QkGGeS+if5^>!)G*n^TLH&l{t>L|QsMve%55WmSK{=wJ_ z(@WAYEyU^TYd6ROLP*{0z;Sa&v42sDNjX!OM;<4V!4CB;MuIk_9~*Jf8q>XI+&elc zDK?*9Wec+xX`Y5o;lmbCZ#Q@q+CMeI{9yip1MDGQIuaM>Ty}i(X2@~rVDKzJ*Gr%< zHCmV%PyZfv@g6>hssBZsp552uV-Qlj!4Wa7d`Mq9&0J%;g3$_7!Jr970sA7a{vV z{4OY}Xuxb&Z?ER6Eqz~N%3+7#-CJSrD+bK=d-s-c{WYO;Hxk%Xuo7?Il) zx$ea}j+!u#__1n{=-Ew*ntyd_zZ}P}`3fsV;v8JfBY|2D z&-6L2k>NHHRT+cIf{cC=7U_X>#JBvZKfJ`N-tP5?X{QKhI+ezvJ7zX%?b5R4ShrVq z(}pDgkGZGzNUX-CT{674khqG@%xU!7)=DMNr9$fv#7*~&6%UzjxAT)Oa3s~E9PF!J z1K&?RO1mkh#1Z8<>VGwV@Yj|Tw0_~iCyPa>br_Ja3VSlk29=!3f~x8`+})DrpxUFjk*)btLuDbCy_EH6;p4EF9)+*f9$@ z?K0`eV`MZ>-)qmI>2Myk# z7m-Sp8yuL_eSi3fclzjn$BBE)nD6Uh!T80O0-f#R-TGmr&Qxav8i^0#vlu_oP2#=? zs@)<{&AWsh(W@Zh5Z9Htv(gfVuT*<-X>eL1JCQsx-YpH}-Mb^JicsFcJOshk%aE<& zwKF7Z`upFm!{`0oNcT`=TzGsy9&5X=m|4qcTLqGLAJNGO(3M#zp;{ z#xEGH0~gDZXPM$#W}r4Z)b#6KI=Ei|Hc3jm0}%YLv1T3HOJh04nP#|6RbiXwLKVjQ zWF4Ha#d^CF>DkDzjOOFJd1VHPQ3Eryamda>ZcWQrGO8)U8^~EwtyYg)N*hVKNHb<# zI{w+=1b+t!5t-=2iB{jREz@bQXWDecNaJ7Gjm;QNb{1q1NQN@Qi(8bR)=e-pk|NI{ zG*z?X9i0`o5D|CZ$LGitQgH2DvT}I6`C8XL_d*L zR3;0*b9ZxdX1+HJ({Q9v{DT2fo}1c~J$Y{vdIaxL;82?HNSFv63kRBPPv zi(Cbonq?47X>1(*M|oW2$MX_4Ro(OAm~6uDF3 zfq!4guf7n|Nag+R%2CjoV-;2yUXdTNBq7b(xdlF-hkL3*ge4U+=>8(nBBQ&VL*G^o z1blnmkBV5v-FR}U@2jF|C=Nt$${H8Dwq?JgSJGx7sN{F^LS3Q-8iU9_oc?F-P%DDI}h=PwDFh-TT+X+Ks$c zoTtSTd`oO6V$w=65q4gvHkfi*_+ue+iL0)C^>xUKRj9Z)-!cwF=!|<7{IYn`OWh)7 zJGhRwyk^{qq2uKh&rMDlkg>3y!C?Zvf(7A3H_ab<=aIba$F|7&=~hD}YYJnwMt`~4 z5&10T$l&*HgaU%TN#55LD=IG&6FXiZMr*cBGBfEMniRIt)i2(EY`39#kX%41vg1Q0@r~v=yQyt697wm5oiAy6o zpZ*G~D7AF4w`|Paq1((u-x-)*|rL^ykaN){$ zgFrx<>Bm^F4jM{Vt^GXl$Bdv4ObfQ`$FHRN!gpNgaCv%c8@bd8(1DcpyH5=Y97~WC z+8%HhJ~F!95ayRBJ=Hofjek#f^~Llx!;$sKm4@Tby&7F(SX+Jym$jkAnq6S1?Y?J|lV`uhzjBkC?IAF6fp^>Y|Ti=zz^kcJkdSMR}3Fy1HN8^=D z&xpb8nWX_1>Zs&BVg1u*nwLHm<6guv=BB?Q!d#zzE|ziuspaa@;g|M4krh)-Z;51w z9PQAOn{a-Qnv%IQ5r2_gC{S?;kaCt_?2#mb%3lbe%0Z3YXa9uuF4U9=Gfv~Ccg|G< z9x`~yYPYd}nrT>fJlg{S^i8Nea1%+Q+ADWgP}v^l@LdD|R|`9K`6*_ssBXK|?g-w{ zW6r3>+|m&q5@Rt;bz&c9emg02EEo?c1W9JFg)vTGPU#}aA%7)IwLD3)Gjw1*dvv== zqNHExzT3>q7gUpy1)fpw!u!+b(ucd6HfSXY#A}(9j6`^1Mro6;eaQu9RXI0ls~s3w~A=>a%2XmqVaWfQgi6wZ$wO%-4wIqO9Uar zl6`KCdtHs8tAC>1wpg#cl1$h$D8x_NLYewX&9ka+5Mud#@OIbkNFplqjr%@GB_R|fT#sp%-?*!fFsgQvAlyQZA$PGE3+7q7O9$gSeyVLAKg8ld#z5S-V1|UG!S1du z7dt!Zy?8T!vQ2~vI!ADXU{aXXG58$z62;I_(8X}t(3nfOFn7yNT|+fJ*9Gy}Ad1n# zrvrYj`hOylIVRp8^7Be0jtGpel;dWFDYACB%5mEXcCTa;u8~ua{Lz)nWOd_rKP(Mb zXh`_U^0*@g8Y~rsqBA66DPQ)=qOB)4DM~mymEMw!ZHC4V#+g|L5SbRc}lMY;? zUbHQl^#D&I!&t%v)7Go)6SUb8SfbwE@>H5|e1DRthZEl-0<*ozfOaj1w63gTWMML+ zxi_^%A5sxh=9>*n%(2mrHDvI5k6Oq*wU4yqFP@fElGh>;UC$}o5haOq)?tdS+uywL z9Y`=o=g_+etN>q5|^?<@^-`aaAfBRRjYn#-MpO5#S8WqK8-1@{qhxWFE|#XiQj# z%zyZP_DP2boi3I}m>(@G?F`XTwbO$lUq)A~)L=hiy9_q(w9w z6xYzk=h<pQe8uRO-I(kAD#+*V3$`*OHzzm__aJ;H+C0hA1^|Gjro{ zQDNvkVIX% zTWd!XL~~u4q#!M=FOn0?&qFvo7`&4uocf)khbDC^j@nVi%8}!zl{Q7|6Va8YUXoP5 z&OiU89Yc=F$A1Y|I7$=;H<_TQEvJLPj76dRYa=2m#&ML-O znf&Xbwa{c z%~X$xzlGg*G3H7UN@BxWyyItam-Xp_C!XbbqQVNp&`+8*4taArU?{!|3RgZp<#DZD zfrk=x92(SkD8&P#OU%R)Cnn6cUHe&t;LrZ5a;(Suc^fZ@?PpA@WKz$~?tgh+(q2od zy7MMVvCpTW&|!wA4A?wG!tMJ^p+uMh-Cg$VAA=r9C{Ulvf?^sR1{A$pU&naxAce-n+d>JsnM;|yt{kqNkAF%)?k5|x3Spo8CRPHDIELUd5zD5NU{8P_+J^zp0g?Hg zjk>v}x`=$=%hNI*+g`RK4v8L$K$Jp#=+h_ajnp3!aid-+?=hNqk=*F_skZNs@#U|P zW~-nKgCwUtB({JIk)LRBTjdeSY(C~4>uc!aBZU*32ZeekYI8cdS&=x7dQ$k?{Cf=)^pePnow$kDD64)x&*>?}(?%g|?+xweXKf5I6) z0Y?FPzHjlK4cUmevkX6Ql0`;i_?|M*d@l&VuZXz?<6ZluqlteU;<7G0L4On>4$IGwwR_^%7+@Hq}t zi8+`kp3fX2Hp|Az6$Sk%Gh%ODZThd_y0WpqBXt)dywt`;cI!fmm`tG15v4Gwh1TEJ z?KVaDD#HG#F4z&U7^DgOalACg#J>=|q?tbHqoi*<59%MzSdfQI4Kw2tRqz>ik-z1H zoA=jPR;pdAMt}blNJ8U+!#P~Cv%X)FMlZ5bva`UBeWWwIb>v zu)nsulIAopatduT4~MKu^y58$qc*db`)7?hrLX7_OMg_NR>fo*!xLZID2rvTo1g5R zO;tT^2;(|Q%_%kVl0<6fIbth#nu|VJ7c=Ooa!kUQ1^A;RArGZ=CI7llqjQc$e3X|~ zMI$+RB9}UAW&9vm9h1KTvT@^W?4uq`Cl%R#cU%f#_}#2HUe=F-5DBrS+?{Hr+U7%n zL>8Rg)PMSyN|f5?0etoec7egl=8s_%T=Of_d-mCkGq}|aov3-@$ik!Nd&f3s*=+T1 ztPc$(pdqfh6?3ziquM0&vbHi}P~Pu%RWik35|WaNGwX(SZ|cNuqHk4sA9j@=zLr_* z!*$?y(5fGwu;MiTD)Iv{=6y$*murd296Y)#_ zA@;rydLSG|*8oMRlfm(Mpzvmt8w<>{`i&b=8;{;C7ju!@_v3!UPj({Qt9(A7pEtKx z?SD0|bDW?koG_;Y6bbs%XwOI5Xm@@IEE~OD%X2KD=`ZYEuhW{GxcnuWYbH`Z43M#1 z??mJY%b;^U<;^5UH{ZEG07Qc*QC>mC_ZlyAtF`AQ@G$LO_7uK{(*m6ZSVt#bYg!q? zie;2UDkzp!*zniG7~Gl*2}o&q3yPVAvwtFl3pjSCNruDO5kF$yCz7wN65^^Fc?CMp z9F(aiplJJFQ@V}kuNO!tRj$4BLF`!(=bM>(r8VK1EOp7n`=Fd-ti<~Tq9@hR(Ezen zVI-qD?P~(AVgGlZ&WhF`&UOUJ&M9B~@puB(##u?qso}4yn3|nJFMy#}^_odqx_|J* z&OWmFUxpb>=<Fm^2*tuNzF1j zd9KlZQC=R5J&Uc6JO_*mS_e)ll7u`>9!C2zKZy=!F~pW2A3({EGU(PBV}IYRE$jC3 zlH-VcqzxSCE$EzYjZWHtLR$5JW0`ryU9Ur>fRCsLTq=JFHv47=o>gY4I7HgoFP>g(s{HP=ATh8 z>5)3Kkd^{SFRL!_VwYT2KaO`J<$yTCw@bvoga>af?;sX5_X3^7Pb?mHYHzrN9Iok$ z^TPw#^^~*Tgxb{LP8Z}Oc=}O~*j8Y~Oq&%Jq@4-{LEAT{0NdGD+<(&&Cqoe4L0nx*3 z@tXBHJ=bS(t)}rT^IT3WcGP7P$K;TdYzWKKVaO%0Iiqj*!1YHgfr8Z*ggCm~?KdpT z+2dmtUo*`_m#eBxd4K;kNrSpP6207^D8dxQNUENKlQ*a$9|$p3F?Sl^++ZxBQcI9M zk%E#UKLwLjtF}nsNb{7+YotgbB$tUNe%dx}peWmyJF`TEsx?3rPe>P|Lp^oAr$$eC zs5qdnB6V0_@8Vue3T3^^_wc}h6p%5}I zsw#p@DVy6bH0=5U^}M%Hg?&Zkb(*4OLI7b$=t|UC(oI4sFLK=}O>;;pJ?4<}@`DAI zCi9G_x6vINF@IEiWMwT6RVO?YbZ$(oE<5aj)mo+P>KhwW291jK&jCOf9-1JVFEn39 z#pwl`*$dxjz?63hVnEn=2+hU_Re!>jIFzC6V_;TPPHq?UIr7E&4n9^)iC1NQplSa2 z1(MC7+)`1yA&eh^5PbRQMh7W0*ZJ9`9it^xTj_hQseia+{mI>zYD^PL%7r6YoFd8c zp35t*76tNG0Sr}g1$t=31^xJ}wlb=462=ycbW>{m94`@U_iieh?Tjx>JysIMOP>r* zDMYTQrH&Ss7`XzhFByb2_4qT-S>7NY$fNxb+s10K9sgWWa1LvC+%TLjcl4fTlJ_Fo z_Zwq1MStBmn-0htlH-cAP{JhJy{dkH^T_{X#@&vSiSq8dolcc6gtce;H&29c`jl$+ zgEl+NM47h}!Q4sM>N|5osyjoNU-08}jDG^Zcb>hMpnq$*$#9XDoUX3QKgdUSda|w6 z&4tCt`$41?y?(b3r%MGEuHL9MN8jfCb59WXZFU$g{Q&UpfK0i-P`8YJi{b_aTBnyq#64Z|@zi zcJ-rOd`*`!-qw4dz$KeI4`5;dEL+AG0)tPY{f3N}L-mHGwOoeCmpu9~RrrA`5<>80 zunW7min79gxB*Q(^4mEl1hjEYSrBY9(73Z_1?)KKD(8|{DHe)LaX*1mi5huz1U_=9 z?B^D(gFug>;ki&5yjHR=`XQm?S9zY=+%^|%m^0-;=LkXqqLRx!JO{7sC63-nU(_15 zRZ5UqE9$O5%|;4iQ#rPAzI|Udf)0?&;w$A;+ignkQNHp zyi?hKTB{@-NJ^sfd26YQ)uht?pH;P+CdnSobXH{B#3NJ@%0px`4P#D^DfLziA+b6m z74?fs=CSQDaI*RSPwal}v;{wnS0-+m3`Ix)dr$P2E_Izk{jI9X zVXsg^YRA{kwCNauBD7h7OK)t$?T78eA-T$KJ4QmYKV0PZ7y zHUD8uzi;Nm1q94JHmi>iA2hpXf|}GIFI0U3TVo_jp@myH96LU9N+pBx2Ua)IX2Y&s zB$0mt3$n6Eg6{nEaKm`Ej65H#Bg#Fu>KtAmXZdX4q4}p+et6*~>JEA~ydvovAHMUqjTC8As?=(TEb&Y}?*huAVA5<6WL+yxp zu8XwQ$6ilYegi2hxl;XBXU=U!0bGO$pVY-}y*Mr>n63?36ttyk8trYtVRidaNzU2e z-e1(*9xfw&|2LrFvtRT22C&S3$QlS_A`&OL9ZOJE{MwaAxv#O=GK^qB1U$Wbx<$4A zym;sWSQ~I$!T5Tt+`Bx8HTqIKT-bJx4$H|#@)SxK)rpF;;ip7$Dm;_{lx?_RX0L>% z1Kwtjd6jezOD2e&n^V$cN5-pq$hpEvk(pZn`y;!Dp@B!t&4XgmKULv>k@Gk|=7-}r z)KJCygp5IPHSlkjDlc&V`9#^cqXZ`{ev=)Cx;GW*xoQkCt# z?uiFYbq#f3nY5@3RLt+pEJ-jo z8`}?dFCu)8roK}8UwsLGU|s9YdQFiH-}D>iq6?+I+c{y>kV&<&F=Kx7J~4!8axUwT zjW^S}ccD`7gMdhIbAh#;rB4W1c{*lK$8arO%cz24fXJS*iWUO+N;reFRkTJPp9s43 z#1Kcu(O^EmshFo2tkEPh5!Mj0Q`4NwGV!9&Gd}WTs28~FOf+eKwrpab#kk>K3L#+I zFls80$wzw?$a7w107*c$zpMNTGEvIXYtjx`^{n84(U|($n7BnZGYd7G3a&4#l!a(IbpjAk5$+?CZ{zQ{X zoAn4Yk68zrX!k~ZCHE5Cf8E1yIiQ397#r!hi<;BK8x9X-LBqOVzhZ#nDeDHwUbSrrxwA)~6j1%U5){WJULsf^5YK=_(Auke+e!H!Ri`pU9 zhyp&F-nu%A|5fQZrbHO~QRFG?29iOy(1vO8^-Rz1ooY&do}eJre;_|3oy3(OOs}(>0(23ro zUq$J^eI${!aUMULYM5*add-pv(BsMI`wvzoAWWY*&nujO?nN&qCV{fkloZh_7s0R# zGu!YwyB*Ou=cdT+^8v#g zWk>xuO%t2u*4@%|j<=Al@gpWX7TELkA_tnB)`)bR_%%YWfBdl49$aKT5$!?Pkh&#w zm#>{zBUj4-_XA48@U=WNaxYKOYm^qUa~M!l{Fp{wz1J$4KZ$E6dQ^o5+5Fu_JoGUQ zGOW{3;=2T@S{B1sinK0BcRH(c3sF|x0AOr zNDGVg=#E?c)}v>v*7%Xfuk=WY(?EZt8X5jF>O?~0e-;wk6+a5~1im`Ub{!oQdi$W^ zJ?Xbuwk@{kMVp%6cz55f*^yBnMEsvNemk&>(DA5+FoGfXc{X4Ih-YKXOJ>zgUU2;b zwE&hl8nq@2=tWm$kR+XrTDdQbEi1-S^ol_8DGP==SdqVJ|Kq)qdyzya#j@1r&Zk+b zNQRPZmR#Et3Dc;Sh9BZkhOF-vq6u}#htIRDf5Hh`iHq5M5m0~)N>s1!MZe%)(DA6VVF71YZe>fGNMagz0*z& zf1&Ut(C=Z>2YQnRJDXqs#$#AS;Ebu*nrnyxpdZe>f56!TqBT1`6AfVF zE@F&fmfSwRW)pBQQzU^Rz-oZ#OiYT70rrlBQL0{;^IoJstE4Ufb{j&mYA#)<92bD@R9~ZJ z<0{}T@%t#yqjuZM?nF&s!F|mXe5Um zAygw`h`ge^6=p1suj8`W7I>>mIaEZ+K@{R~!?fAGHzh!jHvcp^&+M+9d5s!Ism>X3 zg#OXC{^*$0Su!l1?v7+@Vf7JwP+ew(?WB_9G3j70p(bmnF71f6j7DH6dim+-&S5Xom zoWUNteaLM4O)r6qHeXuS7|^hShj|#nehvCK;@~%mRO8T?Qt8#fPx$MUEI0iwMKdPO z9Z>ESvhAxOpy;|2Dy5VC#6gL>?QSK1&oA4e6S^Sj!&~=~qq{YRe}-Mf?odg+2VZ-% zvJk^5PN~w#D^^hi+BTThMf3a{nJx+iI+rwC2wLPjnx++Yr!coDb6Q1ICtHe4=|u{R zjU=dGvZ0pm)wkqdt=)&-IAm+$jdK~wlkB;z;%_eRcT zr?_$pcyJape?h!fAHWXJTz=X&RkN?M8d@o z;24(UZU#K}CLn5u^QKqYa{*Rp=z7(fs>cQ^_VI zXB_h6raH>2Ftm4g6~up(Qq(-2qYdIyZ<0ny9|WylfB5t!tKg{@@(sZ^61(n# zM;=?F58_Uu01HcFwI09eC zr(I83QVdw}hQ8wlvN-{g)II&dfALO*Si{kUE!D#5M+an(8^*~Pg<(}DC#hdV03rW5 z>21wze-+osln z*iLzN%yJ^sy+ziR6M@qZtnr}zA$lm(!-p!9f5-PT6mszhHsr^kOV@i4FhXx^P(V43 z7Gl6^;!&VYf?)58{s#62xXC9(yG#F zwq7HE&PQ+OxWAGTC@AIe6i4oF6J4b_yY~7GGjGa)9W=n73l5?iJ4PBaf}!R;1>xkq_;=_B!PFLf5XWiTG1jv6&jQ@13H}1;1e5g$`!~(a2 zH(tB#ks{f%MClgsnq$2v(Xxm%R$()c5gKfKFY?*JfWk z#41zQNH>TfG)^)(!T;pLB|QxQf49vrO@%Mm2q@6P9cN^wA9J3_el`CQGq47V$@vav z=&Qw5seT%~yvbd+59feE=j?La=u{W?%z2qJ7wjes3F2LUDP+m&?3|Fz!_H_*JX6&B zf$W86#fo^AvR*Znon0kgd5OK2cF@75(yfaUXp^{+Q|q^!qtXwrubW51f1L{1YU9|u zt=0JlB+Zs5|MQqnscd0{t31r)^shx2>keRm&qOzEwSBzVp092@4qPPQZ?*Mr4tu+#GHr=^$e>zwGEGr18K&DP5f*gjHEBmb1yQL)x*@bDk%va&2UHr5*9R37Gxm zv-4w>H>ZtSvF)W@x!r zvrylo;r;1f*rjHp%FtV4I=d(BiK9&_A`9Ci_ao2) z1!XNCq!hKX=f{&n%YsraW=+q*L|L&^5NteD`fbt2Xev&dG@?00aaj_`BI*om-B(rS z1t5c?ZZ5A>g4!cMe=4hsf^aqSv9D-iEGj$^0=iELxYIp8L&%*wY}IMQbn zpsesd+`yAIc!*C9ZyKP~YYO<=|F?l;RoflTb+pc0a6DYyf9aOg@JYgIKD38^{HO#T zy9rxoLV{2s4djqXeVO?SAA+g^*L;qNPr(l*!tREoBI> z{^8XIOpGl}@WQG=zC^$T{eEbyto{~&VPgohr$9GC%pW18Ftz({%R0~8kZI`d0RGDKIb(>Wugtq zy2aEPpEb(_S6rK)P3R=4vns8^*qvs)?73YVe`LL@!Re`hLgzNe0**+%oJI9toMe6; zKiqQ~YR^p7@?Cz9$v~$3g{X$ZXhTB#(v1tH`3G_BlzJncCbQL0WCPx)2W&-YYg2ox zInnrqBum|MMD*J|O`FfzGCCi`J;AR{uarNhUKVe$au;7jQ|= ze|nB}pJrhHzcsinrzEVuEnmtn*jM0)#iEc7)5mv5RMZg3N|N|}Q*nNeNm-@Y7FaTm z8+|txsecrJc9u|ZhIf)OqMWhS(W({DSWLs+I%zADP4Sl=Q+t3b8d=J*&F?>FChx&6;1C^b4R9|R;m%)%1Kh;1LF1K7_s%GKK zD~wWx_+>Taf7*bn=d8eq4;{}j0pg1Hc@`_KV&F5HMTsZ`2I{cRJt9mMCkf8Nw!S4j^m z1Z9-Ka*D@TN*A&m5Z=vb9yxd*(3tFRsRRf=GP~Gy7~m=$NAvb#dcZ@R9X0!BH2R83 z>57*=1nO$dqRK33`7}`3@d#qG+@AKu-`=ivh9512_zwg6&KdR&T&rYrSwj7ertW>f zO{!ZM(M>XNKGd}>bJ;wyf8B{Vq5}u1jQ%rAc})SX4^2Z7b2V|{nmp1)FwO?NK%jNR zm88ukj4M_{O`4$3Umx1=>O8lw=M;-OQ%?`Bz#uh&)6`;>e^L+^URq0{ON4h< z)7 z4rth;XKNmmHs%K^e{qW6-|XXg!kUa|2^@JYPmFx0<0be={qP$lQ3|P~xX!oezy`E% zycUYMv5{rh0=t;obyI05g zP;0waa)esbWFPKQBXYAZ8I#nHVOQ#%XQZgxO5 z|9$gl+R={~e^g=1X?*xmArG3}eXPwb6auM35$dS%_PH!jy9w|!wQ^*#v43q3YN|P- zIImI{bSd+?+PNx zenv_41cK9wf|pL$;m(@Js6cOkX1akmdz}q+K}lu;kpTWwd40P_^43f&kkwbc}EmKAX??YXyzRJF3RTs!tcF{7{xq7h)uXZ^kp^3D zuT6jE#n8qIaM>rwFzjIgGvQ1(;S%!Oz?e)IAj2i491US*F%_dcoaQLTZ>$mn` zSkSPLe}m+Avn>?1(|7>)^vMx&Gj8XV6{@V&fAsCrQ=fhb3NdB9+~Eq|CoOj>Mz#Y-g(gV!)H?n}L8QFc>ke{3g14uDvJ< zKmOiHV}B+*KXKhM^Qn&1G0}Qi%y+_v0DDCFm_fI=6LrjC9S?KOViB3IU%fiD83V9D ze;db!E#axiF^4y*(ry+H(xl@eiBXDLw8%N*Zmbh^V?AyCb2FH}M>l3~o98y3FQu40 z^5GlVG7SM#WeQI}18Q_XMa(v1%oH|%u1IWhy@&4Ac<^KBCre=ysw{&73LGD zy5yv295!cs)W^$DOYwLvLw71duzbnHe^_tJb3aM3VYsg_j&#H_gm|W(yV%C*M=kL?QwE0@MfC^GhkX1G`0HLdCD4adto@vsK0; zxn0Zfe-zD#cvx2+yC!MS;TZ%UX|G~XaU58y61Kx_#Oz6$sV5xUu};>Kyb8llf8Sp2 zbjtt_*-X^X@#!5akD}E|93DVV~ zX5{6s#44T{c{ex3MP|%-x7271d5W>V{4KJGl$FfutnXIQL~W`^0>_zB!4drZ$#^?x za$YyDN(T2|*gQWFhSTNku}T1Ve|lYfRD|Lm9Y`A`Zj%>g_wZ;byM#g|TiWakq|=L; zb*~3vgp);-ia3g2=38q-`ko`O-v{5KjO3dX0YzLxBFXkrlYy!j_CH3`gEO~GZI)GV8c4A`pzKOvzFY4GFScp2EGOQf4h^nDJ3-2 zvz#lYqesa+t)mETif!*< zaxRo8mLPbODX!;y#7`|tM!k0hKbua8%ji>+MT?ZE0eZlrA+r3$G@1#<%47&7R0YMv zaZ3$B18q?m{g;wkEZ$$_f8gl4u7fAmIpz`>mv@t?Jrwq^z7dq`?^kzrX^r6Yso6s< z?Y=jBt}DgW2leeJhB9LORCV-JI4;S9-MVo>HGFSNdVTNjte^)P4hX^~E$FVOAh`!q zzz%e9zifGxLJQX9&C_jwP*A#yP~EbanQOTEIjBLQQ}}zzjxhy_f4*BsdzI5J+Sca< zr=s~kg&Cm+98>I%jzLZDrp>j8ZCa0-zEOrQt&w(TKnn~b5A%?Rgt)EfDKu)uKi$V~ zWlPyzyBMF?kQ%$OPBQ0#G=wLIgQCN6+@hT?1vT#izP{l}LWkP$_tT}X_w@^)U{pv@ zdBO%%I6;vU?n%rnf9^6$Gn?c4y&X5Kh9jH%0MHfpJ$Up?)>q`x!$3$XfnAuMzo6l3 z&~Afq-r0j(mF71HKRloh*TN1T{Z{ovlc&gWoJ(C&!Ts_^3F zC%hq2x*~5R@-hil{~b9@ ztqNFrqd)aWB?2P>m9fxoqPSChv_zc^h^V67-cdZtrTpcEllhqNtiWS;Z6*eBrhBG* z`XD}W&8U>xwGw`>F{E)H9)7P@I zC1}pgB5(D97?ZD+QS|lnQ5UwF-kbmOcV5`<>yM-*e}iSh2a^KBndHXJ(n&r%wFyg& zEzQ${L?!dc*F#`ay%HxrGxDIwS<8!a=tlr?uAgzD6(<@}>eBeeZ+zI@fFi#~+0F|O zmLG9iU+9ROwulRFt{4E5lL(ukqezT#Rdu%T$+iTv zjHBkNi#LH$ZQ(h6!1hcxVa3waZqkXiGzqA@qbX&4p;fLGHi1uF_mcOC*6ASTcKA*| zL*@eZI$Lj$11eX?EIPDRZ6EMtvCnnwL>sl5e@tpfN}fMS7B>x}z@u1#ML_L#J@)a? zFO8;JHGab2Siz??I*C*Uq3y~cLqRK$LJXIlz4k;;rvqTKPf-7*U&|8e#^V$;0WP9ZOn^!+RRv>i#O$RS&%esX&lO0Pu_~`hAFBe_EK7 ziUr<@L~7=MN@7#raOAqfw%W)Opy#i2Jc4ON4gmm72N!Ryk}dIcssv!s$l<@xOB*9_27Lm|Sq zzh_Ewl|rZJ>y&8w+dcQAky|ghf4tL3mlO#%oa(|YNWf}E&FMoKO=gBa>!-Nh@QmmC>Joln0S2II)=C<^;pQ@74Y~T|3LmG^QKj5a zgZ5)HAq%kvOkw~cA{i|T(cj2AR_=57Y7~$4G8-s@QGX1)4OL@ZL2w8;e_5bA77+CM z3(2t{7|>PX8r&?nX3mF={Z&1z2y}0?&n`rkkm%tdOf@VCY|egZy9Usb@@)eW`F)ri z^L)uUJ0zW+l3KI5tH&sxF@N@jN+6w&Qo>j!{to>Br;Xhdz_94)nLGP}?u<`=jGSg0 zTyO_z6d0bBtO}L6J&Lxye~7*l98kDUb!MRNZ^xE&zFI5v0ERhx-^`y;I;tmbdZe<0 z8ufW{ndO%2-MNvtjfgvpIPmYBtFM0Q>eLL+h*}}7TYo+$#E>nt(wWlEJ+ZbHv=Q4ktX8aKf;b9*fc-KiG4HSKkef7f~0wfY(Y0koP) z^tZe<{Go?QmQ5Q%ay)fLmQ;-hQFYdc;a*Q7Dp!0km*@$Vcug@&eF6F$W7|R_-1ZW; zEL{%N1IZ%yPd6&74$PRMG0mb*uKJcmww(i`HChny)G!mgd*RmS%N3V{r!EM0W4M%y zX28DwB)T%tg%xWOf8Au%iz}_#Y7({BdHg;gNY(4mdUL)JA-#-4{H3j4wG3|~w*has zom%|30RZ3JwMk`74#TKtd)!1z^>Ax5&Z3d6KgmW1zXQ%q5LFu4v6ud*$5H<~0?X~I zPXs9NFM*odDzm^D6MFTmBze{a;1)T0xFV}laCEC zNxqpJ=vcwGf2kzon(=NKqq2Y^%@>rW9&**^W$8#yRr$pz?SGP@XU^B|tHx zmD)$wz9Q`UhK%%`b4Ub}ej|ORTDCnk(WIswE-j1In5|xUAk18c<&g1x`?pVSl@a@# z!O7=_NBvzF0Y1#bwtWh=zMfYrsLyO4{pWuBoX# ze4k*oBPbWxvJby~0{YpXyuiy~JDoAvJ71hVZ9dPp9Y%c!`R^+eFgIp-ig8E;Wvb z!Uh^Qe^I8BDQ^X_go7Zh$8AWGRR)YJleSCbKLD<}IZSllCX)fN(j*{^`Esr(Tmy%} zJ0k_@d+ZIj%<`FQ$*SlIgPGo?{EK^V-GY+RV&X!!plo7lT)4d_q zGd$4)l{K3-(wIYNd?qdS-UupISo6Nu828l1e~#%`ScIZX3xegR>`$<86XwS$jeK2- z6+?#2>CE6+&E5h7r$TMfh^DD*(K^|RjH0S2DQ77};N8Z5*8~RBZc*LT4QhJ$sB6pS z^LOX((DBe4R~)R|N@acwG>Iy+quP{cC8qnrKn<9k#i0dU9_zPc2( ze{3!x$>E1bq3sf07y6B5q<(p&O7Q)G$%3+D|4gXtUNUWa8ply8%8|LE_cljh3^(Sy zEF9?$DM|4X`K5jIe$EV=G=uTRfe3^^5B2G?6KQ@Pdnmh1d|z50#Z6@qE0?l)!qIXJ zobQH0_|^JlVGgb|D$ybcYN8``|}3PWh<0 z=hS`PffQnJImtbk$i*Hvy9=lNv;m;)yX2@Nf0WOe4U%2tgBm}Z_S<&{Kw9f{e^;mb z^DeIg55GE(Cejvfez<7F*h&MMl4)8WrJEnJ-29>m);aUoW68}37Tbso5Ansd(!f^u za;9lLbMz(WotDOev3M;?QH#_tp?^|bKV9pejR)u(0}b1UpPoEDJr4VI*2tQHFfWa1+)Ag!hy*JzKA~U0E)rWbP@#T@E#=-crH;#NcFr;78BR9MxSiVZ?4r2Ylsr-YXAtJ8P~o-*!W? zK_l|2rbQQJnr)fWc`hn1%Qk<6C$f{I9vrDl>$d=&s&e&-VcoeS{@@BnF9mF7oMZ4w z4zS%ZT3I|R%jC*Sx=lGffB%mSo0NK!M_+>L;uoVU2Rh0nFK+zS-4=F64r&v+CLiPP-~Q4_4JZ zL6$v&%=TghCp7c5d5p6EcVcu6*~vWtk}iE-V#HgOA^OA=e+?D`e@TuK;OFC0xrUd4 zTRuC+4@@;xv90y0Sv(iH35FK+)P6@%L=H8{NCy4zX^*!2`1MyA6_Hi^p=K#uTu3{! zc+ZBN{_Blv)VenYSQi=r#axcZ4P%hvIV)#5VQuHj{Mfmjy1$%e{CS%;u)y*-B)c*sLALyN=0uKKqf zQPIe9##Y|BV3RKu(+2e-6fYstvl!r&vFX5W%&X|i!pam0e=*&vG+XRr<-|QH+R0h- zeowJ=N9|ql==`kxdWi`a>c?xV&T9M@MfZi3L;eLWN4e?)_8~r_P4abjP1f#`?hLXw zJ2JH|hL>E}1HAhnXIq$X2#T@J!j3>J9WZ#I>T)z|;wftDv0*-OJ2Cu0Ge4!^eSNx_ z))NR7uj;r^eJ*2G^^;#2@U>In{0>g3f;sbA1%95PXrH%Nx&Q_~e58?3-up1!Utn_$8#a}*8WP?2o!n9j8L2~ z_9FQ=ojv*#I&>6)#%yCXLJD!dj90KG%^rNsl%J4dlbTNLfp&iACod}!jZ9^o5@gU^ znM?#QeSgXd>_jD@NilpI!#;M|BU1d~Wyi3W-Jc3cA^5cn1|LL(wE>Bnb$+#z4A6X2 ze*c(;*Ut#c4+QBR=mjz*Gz2;sZ?_M@*~ZjoJ6ZU@2VBkpb$7l_Zm5lnX@jV0 zYhw?f4Vf5`!3Iqr7B<`7d5JQ`(8())awSkk3#im9P zw_!3!}NDc{7J2>+C$@1S^}6XR0kL*Q3OuNro|(tK6!vV1;M^) z*{sfHR&uZaapJW5^|0_nJP#~Qf30A-G@!d=oPHNhAld3?Z6~jJTO|4Il2mgCjURrh z_^Sc6^ZZfhG=E8k7;q`yG)>W{y-J|71#;y)WLie~OfS;{xrqBX#3o4mfi|BUhc0;5 zxGapSl%L$dx2Hdof!8R3u>qya*Dkql_>XY?;J$#LJ&dyiioz{s5ffHDf7B)bPT0ht z;*TaY1tQL;rf(4DhE!O?09&N{T|DHPc6)ns1@2gSwqE*Ax1@@4>pS$=9VIA$*{D?g zWYc{q4&>2j+U51z3mP6L=A?wQq`_+9HMfTD4db1y3!)}t+Pr7Tx1(`8&XiDC7bvp) z@$VXxkM9&n$(qiWJ2MQ`-+;0 z)lB84`=?*t{;B;s$UCMZ6uCrG0U z!KyS7M%WiQr@|0)B7oY-nr)&@tlJ>YE6~I#B=x(zo}O7e4O%NQf4Z1K=pdOKvyxRv zmFv$l#=O-~n!wx1cjNJHzX{#ETN4_oIEqdMWbgK%z-j#E0oIUR?iccqJqUbfm0;$Q zr0#U~vUmuE=m(l@DoxG>d8&{)vT4GwBM)W1Ct4%#Yq3ezv*m|I&VG;_19voIYDamZ zy=6_DQOf$~4r$U}e^hZQ@5sZVu@eWf4c2k&0JR4i|3x>z*y{?b&~a}#BbLFi(oW5L zA#}Z~S+=}3jakGE#MZu(=q2ETZGausA&JL2=fOFn3W~c{-WQJpV@P=Da^&Fx(E@us zM81}4iMiylje?1BgLVR40zBuli>e-)utN~OxgshAMcJ?Se2TOXF1Ig(QodTrGwK4Jc*QM02n zaSSjr7p1alXnQ7&BDT5imz)8oD`IAed0shxD^=n(oYBQG(se$E(^Z&(nFsj3 zD*i)U%SURvfBXtJFr?Iw@zXac7wE=awY3Xk23}K^*nU8}M3mItc?pzQfp$+2@*lNj?Wk*mn zO>gKkzI&wGk&QmC2s>e9;R&FTeFs=dnzBgk(AMZge_0amdN2if(P(10dg`y{4HGxc z<@sU0|3eiLsSru4oZ1C=n8`hF`;(s-@f-f?AO7-2{wM9l@6r-vULt-OYnCYk^y6GP zJZ{x$hwef0okcV+TEGq`%m~1ep^hFl1g=*X#4PLBex5@>3G3isAcJx7QQXPDzy&>? zltG-xe^!87@p?ORL{V+zdckqU1DFX7QMAdv{=psw@zkfmj*T4n5xt8`ENPX6ZE&xq z6b8K~|B5%oy10FDtz|>QpaSL1GMh$MlNn8weFj*W45uX6z$Ao#zOB2I#i4G21}U_k zY43M+{@~hwDkmf=cn2q39|J2u;hClJZ7-~7e{Hi){KTfEUh1c_O+W5;Jz@*e+>QOg zg%1_q9}!=4hjN;HhAy_ll|+(mQzc^q9&WQtKUKCA4FzJ>XqJK`fV zf49ifntJh@cuvKn03?0EvESFu|6^;Zmhy{TDnpp{rkOR|c|4 zys-Nw6JGIDzbMFzh6e>G0an+QUsvP{e^iJVw*+nsY3u2U)pf4&YM4Catv4u7Cz91H z>fKk_VrhDljIZ0fwG2n-W(*1iuj#IMrs%%p=b3>N46qD+#w)LN`%g=?pJ)l)xNdD? z{aq7)g&&>OJQBm4B_AR8+Aws?wB6xFgQODKTzJ6Zx5t)u19$~rlE{w5D(kbje+txT zz;tojpZriU5Nbnqs7{Sdp+dDy*BjcC1R4dxszaL)T~0w@`=F=#)7=+OGw33cZIiRDplh&u=tVb0B{mH!nBm?ca3#B>WXIxNfT}ve zl!+UQ7n2a^jO5&##!$!e81f%Lf2e4`0;c(DJv^9ge5oVP-6#!2A=|Axob^DFwIY~(o_ zbzJj&VkOy|*kqovoi8^0vo28+@NfeIe{3HS-r?j< z$58J5O^BAO=8z|>({C?dO65u%M6hswnhQ%*Zd^`>n@1#qqt=hegADM9B$e$li89pq zq$LFwDD^yHin}O4Id#Q2mip>8V@*Kg!t!j#F}%($AMGaKRSUP2dkAg)5q8kZqZ{iH zXTeguA(5p)*d0C($ho?ue@YiV-TRbC8T?cgg+K;vF{8HqKcm&AADKIbrh;)0rbj1y zpR=$AA6~7m$=kKIl~pj3_}Yq(q^kdt64BX7M0t(>DPpE?T3tnf2PNAq%VYtd1^m`e`KOz%PA*mE_TBQ z_nTBY`cwxFS(nc_nD@cB9~`>b{TcCfl>?R=!WtAV>leeaVbL##gTRSSe<6jjI~CDm z-(b#Tsle|1b|4CJ`4S4V@@!kwg^`&e#m$~ORbfjiDBh;QN`F*+RgB+^;7STujZavw zI*SmfTz)+ydq=(uf5WXDxk9z@85>9$!;`&v9D$v-|2wLixaW^=6q*?NxU@aD>+f&9 za~axC3?66^n=c>XOM*|+eltclw?b54f?i39!9k!CX2yqI(yW?UMcY9S^Co+Af4$ZI zjp8;Mic+A~KvOGF={FoK0RxM)ao=9Qdkec3h*Z{R#0cWLf8{xc0t8@OVX4zi>yb4Y zTq8!>4U}NVPyEB?qbVbDzt7(X;FY`e5qXvq$*3-T!=?P3WRU;vq6Kl4if1d<2L1P# zm4KU<>Dj=C2y2Uv9m*cmgh$RG%rYHNk*9pne6(<48_k({>LxUh@|@RR(+hso$+89>Z8Vei#Yf{0 zHHa<_fU_2Pm@f4dJ#^NX6!c?-L;X)$$D^0iRv0dFe<8(IM15c)AQG2?Cphh6xiOyH_JsM3&=5oVr_=&(TLgCysa zDpwaR@3t^HW-=D)`RgSx5KLXgel=kN`Vf`5HMFatQgDrC_?G|VdOD1@bQU3qdd(5X z*K(tIe`hsfb}ps#qeuS{55h`YV`oDR1iC=zAYH$yeQBCj7J^C2c~fBI9z#Ume3NSw zvJ*5B^*Kfy5QEXmMA>lA%7ebScJk!LuiEnUXIcB&^w*O!>-aRrhUQT7z$acNJB);n zo4)3vtKmvR5viQ*ZfoEo;YzjFt`1RNzLxc~e-_O-x6HPf(tlsC8dKBkWgns=)Dh{~ z7ym5^7sCd191PyEaY0h1^_o{a3xV3Gr(|94?8+ZuYf7M(No$aQnC(uykl90_ntT@u&_OMbrNX z{3d0a8s=I!usC1At7Sv&kjVns64xzr;MZaLm5Fs^AP=@$qDi*y@L$($e`9m^#`HLT zmdo|{fbqVpvQ0zoR7A&SW)dYYU2-dVrf;G&SCbj_AA9+*7Wfe%?$WAb9(16r?KFd z&Q$2dx~gJGsY|!2Mbs@%|N6Cg=L2<$!*BTmVfptzXtM&C*i7~fSDuuLrL>!Bg?<{< z&JjYKXX`CR)rj*fyWVFzeM*4Mf3H*>HXw@>1Ld}_O0CDB0b)gGP!y(bZ)41T_=8h)ZBp<*PxLpe#yo}x?eus8|6c%G1*H170Khp=et(ce zJZU0HLks-%+tTDk2mPtSE$w?!;AWS6?rV%?68Jcr`P2{!{D`aU(LY9bf29!k_Oj5< zr2Hw6n0{{}6IGQ8t3{gfZ0Q8^aoN;Ddgf$fBQ#!So>|7(9(feGm`z;$phXq^0%55Grp5@eWt(4FI}Zjq^rT$-!gbU>r+3tqCEL;C#kL6q=# z|2CsE^qMEu^4;qr$|k3tFDNF<3?YA1G}!s3Ha>buO3+VxHPH|We<91_d!Fe1*I)fS zM)WGEa$;G)`<3S32(Kz`n`E#hQ2-$ZUyXf@BHUqVvd9{aT{h~HW$IC8>_^;%CR`#+ zFWd5XL9a3Cvp!OZ3=qmNW3S0oqtht+AA)+485HW}i zfm0@Eq2T4UtPWuh8|TW<E+fn)E_pGYSsii zobo%HY`-jCz38b32)p-JW|cJ?kbg@+%CZ@$m@9IK$eCZ>)h}eAd!4bDtepQ%!b@cu zrAJvm-~d6qnhOLbmd8*>n!9|zFs_VK(WFI6vVj4MY9<1t-0o8478jn?me`wF7jEyt zEa?`L6ELg4e*p>5&uHxcm^w@dXSPm65_$3H7Q!=v({#=K1*)TT9(bvwQiYQ9D&V#? zq=l;XKxsXropfop$MfR#ae8JvVC=bP`C>R`UwF>D1~%%g3!N$UZJTNtKKVSKR(*Y0 zLq7KFo}(E7z@nKZ&GI&dFJs@$=`j9Ef3n6|YPxUiD#fWQ$_vtOd0ALX z;i1~(AtZ;CN*}@j##sPx;`=69zH{@RZ0ZpV*?ce*Nb2(PZRYLW#Z~E)&mQg0}pwuY|K&bf{tTsfnNGg z_P9Cnf0v-Y+$@CqJZyOHwlZWZpa6q~RP$n&n}*W=^PoDTPS||Z75`p}tisLgi-Q^y z@t5P@DjN;N?sBc5M5L*7cSH%t3vy=Y#2$vZD&7(kt+QcdWFAwaGu!P6CT$e^4#NJ) zif*5!;ePHc;kp5n?SEzVBW%3rA#`G|s^seif7b}EZEHBUhWN$Z2``{nRsXrj!ATp} ztc*0zih49yL-I0=Sll`;C)^5HQY2}OtG(sNIvMqC$t^hDlx1t${KQZT#gs%^zETS{ zO<^tJtS$7$FG6Ozy_<5|8l}2>_lr3i5L{&@K^jbE!y0bn&l}cecoJKE*H5kw9_T9V zfAq5Lu}cG^&H$z$h@T$`P_NC7a~i^+GzHSrGY4l5KYesrLErGQ!4{UDGXNtsgz_a9 zRj_nFyI}JDoGu*f`XlsbZuaVmk1;N09PAWoVMJ=Zo0h0Kv^sA>)u+MTB0ytX^F+c? zw!G~jdoPPR;J5}_DBF-Lq#w$A_3RaLH0W0K-B#?%`*jZsGR#re_EubpSBJx2a_}@S zOU$xa^3Qw{WraLW8H999EycOXo{^5W9haas0ThSH%mKH_%mO&B0y#IAft~{>f9+da zbKAy}e&?^i<2kiu%>A~NI#sr1#W}GPM@k%LU4Eb_SmH57FeGSO=hyGoJpc*nVnd{N zvkzM(4hNW?neLviZvdqVS(w78!WPQK!WG8Z!WT~YLZ!mTP^2PaK*=IW37R5VQ(!bj zKT-krbt)8%2VH0oD3EA{Z%o3lf0nAy7&dg<70Mb@Xsqq6Dh$XY*1!SG7c=PYDtb9G z8hc6$7VvFjVTgyM3lp$1E=*OFZ=B4Bc2Ng&x;)tcF3t~-ST?f{*w}pdNOw=P!tOd{|uoOb!F@G;Y zW)J#^K@V1wgc6A*rwi}FQc?x6q@pQ2SRsCc#CMRR02YkGJcn;m7rrf+3SiawP(V~0 za2(jmCNLGSPKvk&OA<R|JU>|(rWpzFsKdC<#Z)HJf(gj3%H^THK<`oc#$sW(=HCB0B z@HI@8TF)&&PG9dvJKOVwk$-qnFVCwbuPxt+=i&$P6c^C#JK|;16mOM{rF1||aHY{e zpEso?G{sN?SfD#ypEX!hyqL~Eh{xZ3$2*9}lV)1a#i{u1m{sMZz7pr-X1uDJ)d3uaSVm1g#Fid5YjJG@ ze1-R=H~!%qj+c|m>Bp*^PFCf5GA*m~wYZvPjh2hc#Wzr&_ggxcg>eK7^FPY! z)}Y?}@n2-7CHb%*U4IFmGGEVT@9v{COzC|n6|8musr9K$Y0#?HYOETDkeZ}R3MWHI z16D<(7(!}gY2=_)wen?x8SjuD!wVJUxsJ)nsn%s7WWwFFefYP|m64mYZ69GrcQgo5 zI@4fT@oHI5PAlLAuI}i#5WiNRo10GPh7Z)~rW4eAdO@8XpnvuUxjqwQ=_s+#rZfoP zR;N-SplbtHsN%zs4qvfYTwbpT=@--Ha(z}#Yw>an{LiZw)p$9d&M(AyJrRrbnfO#y zAFLMS`r7`XZZJkzE)pQ6%0zMiHk9vZn2|I4dBN3U37)GZ`@xHaIIHUqA_c@^iQpR% z6nD*6$HDfyOMjZ->tK~$rD3$nz^4wQwQy<{Inf|m?O~W<5G_H-_+hllLk%(BYnnZ? z>zgJrHPe|U;5<0W?Z>Ae}8~^ zk~yGKGCCmR!l@=iAS!j>Zcv*1_3j3^wR*7m?vi%#7akOV9Ym{AY3?w|yJ#KdBsZhl zeY9AY)I5aNMtDj+j9-*zr3`~ekvEsxs{1UxZ{2;X?+54mte)z*jb}iO*U==Z{|@m6 z?pB53{C|Glz_rzhJ;)nqq;g#Gd0Rq5UBaw_EtA02N@Du3_?<+VQCG#&3AVme!uct7}5c1geN ze($8c$Nk<#YqUhf@`v~{WkM13AzBXp)*}?VpMwhdw4*$*dWc-V8EMNH&9DKrpc*rSF|!j};ARP8 z)1Omg5!GXuJL|4u576?|81@gMTXN z71JPENsqV&kvfnjJ&YFV9qiE#qBUSwn_=u4diD%s8t^ppzQ-qn6OXq}Q-=8epRau0 zjoyZU+6>ADD~U4$_kn!ktwkoMk&;bg(EGqR-eDbZwuniZ;XbO$zTk%f2?*yh8}hokdx<$r^M_&0k) zS|vg5+RyN50~Y1x?z`Q7+dDnt?X_n$=k`ug9b8LE@PS_F5w_(A*%nvP5Ii!ePZme% zGt3+o&r{zmV|pA^j&^~d{An&9Km5wbeIv+(ywDs>d_BVC?G`7)&m0q>)Gz+3u3HtwwNW8Cn?iJJ8SVRx6fa z*O}1#Fn*fE$*A{PGGQR4-X!2A2pxnF^@6G zjC|F#bf_QM$10vp(Ll4m*4;XJUVlgDa}L|Bp6R$%($NX`Z+Tw4uZFgt50hzjltjPf z#454vQP8EIc5X*o_0m0hfYn~d4O?5vJZ>f2u_U{vcYnm8H-+e!b~>_i`^L3T;MFontL=n_Hsjj~Vt5xldz5KkaO8^LqW(aqw>MaH+v+1HUGzgbzRqXS&fCz^xaA7g zI&S%7wqsdK)bo>SUX&a&f9IzI)^G%7DC&(C z^4(+@aobbjc^le1Ze6xknqjw8JeD<-`)w=byWaCodSG@`ttQLqqN$hp0oqUFD~vq( z<$vXmZ;t-s#mkfWYCKo+(UW?14uaYEVpZ5o`Gj9rJhIUjkC2iSiqNMNzk`3gn$T+t zINOnUI$k`lrWcp(C2dG!k5s}!KQ!anbnXWYw=Nh64!sv+Blv<0;NL##qHO${L9PNKOMi?Xjy+r%Yn2U0bdtb%YUJ# z)CjDw`z@sW;t=QoVNAhdqc72ONxC& zsLiVNvZq<2`kH+Kq-HH%EUPj7s()-&tyZL;W;&joPodQoaf|5&aBJ~-M<-f6L?_>$ zJbCl@_Z^*tFRMhg8c~!QeG`oYeXm9cIL%D~&4cXd`1IBD*FA6bMd7TGed**@HhGT?`E;*XoufLcX3d!0)%W}^sdHIBZ=BX|a9h8NH?8YBTYruxAF76} zrtN3DIa|+WRl`O&qc`RoQ{?SM_2+s#6V>O*Y<#tAop<`ae?7?iEoSQ#4EN7VMg2rVYdTs22%^ATR?hHxcKSx>B);X zovzbYbZt8zSw}^JjtA>H-5)59S|9NnwA%$<(&XhvZ8fx&L+R~A#D6>}LLC3|)r((V z^&Dts_CCnP)_>-I!P_5lz4+kn2OWt$>tNCBO8KBLarFG>qu*b*aq#r|>a3ovz7{M1 zr$@>2zpzq?-xC#aalcS#??3bLVzI10x1M}fU16;*O~^MXSIE7q=^QTkYCUVFi`k8{ zcMr>$e*XT4r=*p7UVnczb~>a6hEzfSpDUH$>cj2k`fexJ*IGY3f{r&ng?z|Gi~NlH zJF;?)FT0Ak!|y(uPwMmO`~u7&*MeT>^fRotI&=|pRv6QT(as4KCiNrTU6B&Z=O~#i zwe@f%m4>$^iK$L?S)kwjW8iF(! zaK!@-vg8s;FwqpQAcsaGE)XMHAjuGaViXlM+BnPB|a>!5#xG^%I;6gCv z9x1p9Y4GV&fWJS_EW)ShID3fGNr1cY?keV50R zrxFQ{X9=)~5dUZM3}>mu<*y7XkBj&>QWV#54aN6)H2%ky&I2-hDb{W%kvN7qBshit z5*K^YafR>aNn*i+GI*TLo-2QUjnQXWS_XJDzb`A%DQSRDqQbf-2%lW{FYda3_WHwY z8l{wcJjs%Wpc>Y-#W;^=WtIb12<-A~h6}BILl}~k1cfMt96@&XKI<52>oENLMe) z&mGR`Hs@jZ+EBeTtrsBod%z!Y8BN{k^Fucqe;k!jULK)(!9*XjIb4UvX0xeBiB$XaerSBTD zRG}wBmf9x)hh`DFu;Vh$(m$%%l}#PYaJ`uLOv{jKb}g`fmxth4wW2}xiZ+!{0>zq2 zh@?zY2^s2^PnTwkc8z1rcn2~RNQX`LFu$gW?D|hNbdUva4u&K|v$~K&eLmWZtB=-f zx6?w}LN|QFV9{g%e$gx>^k}`cZDdeKYo;rNdKGbB51cpm@u0ry#Wh@9>u6MwU2puR z92Wr;eJk2SqK({#iG4zQ4q)#9JNF@@wa9lOIN4>Bpk7s9*57(Y+$WDpbWi)RQ{W$k zy_%OvoVq3WYG%{pDjZsE&;J9!@(vM~=E4JUP&q_IMlvxtMKd)+I5I>xGB`pxLODV) zGet%>GDSu*Ha;LcIYdN8GBG$sGc`jvGDJ5rI6^r>IYKcrMMgI=MMg3MjCy)X>7ReCdh z!5!#>ei(m%yD$htZ1&f=1S4Gf|2qozU>U|Z@O}45-h*+NfJvBwX_$crFbi`k!}I># zhbrHU^g`O%|PJo@4&`tU^73k^(F9bEU^72@P0cj@UbrN>1W_`M__k6@NF@$x57T7 zX67}cTt>N!av9|^ifM&F&5W8Ey)t@bB^vL)466=*$tm-@Oa+&b#RDD-H!v_b3MC~) GPeuyxny_^M delta 55018 zcmZUZLwKNFux?}9wr!_l+eydl*wz=@Rwo_Xw%u_$wr%tN=Q(F~XEoe|s%NiyYh}eF zcSIv6P=Ilv_h6xq8)$xPPBG%wfI*dHU;o$g#(Y+nnEP#|3Kn8J!X9w_1xf*$+8mk0Her@q6vg^E#kk>KBJLX>Oy*Abo4cMxz+YCOqdHBY zd26-=bbBu<-HZ-W9|&~%KqGWJw}f!m7KY-BjVc;5lO)RshO*M&Se5`wx=It5<4tMf z*Sb|w&=Gpl0F(^*n6fFAl-fsz#yElQul+$^F6`UzPQz4|BfM?A{SfJd`~m_qGXr$V zKsb17@gzCIF~`mZ{Fa)`aWbk~8_ZUEqHPCDv?9hg`rOy(A{#J9ii9A_Z5bC*Pj$#BV3=} zY1tXtKms{D3QD6e`&kr~Gf5!cnxGQX#8De?jNH^+h)d}6N;JUMND|L58)mE)R>`lG z9&DO{Q}|Dc56NJJP|((0q>NiYjk5`~6oPSYI&u=~N!>V^jPq~aGLq}ZaSyTfr4Ip> zGpxH|g}>3NsS*nB9};Kb%9_o)tsaUEHMN)!Kpx^XBtRU|MxElrq;5>NxHZgir_tEljIJT=*F)k%B7> zfW>A*)<(i69MSn3zq$%-TywdMW73DJ1f+34!I^yCA2VTQ3`$@u=@1m-}4Ck6-3=Sjib&9Eg~&l8{vqt{48}jDz3Dz zrmlX%+3tqSxCYH`n@KFlYPP$|`tCL!z#O|-K6;d>c=tOS&lA z@V^_nQJ8a6abhsq>zzfhTaL`BW-b(ET&CJLEuk~+q*+oBjf8tVjRxz~!P_z^+-3-@u$4>;{X^(og)=T2T$Io>>dsR}1@vPq2Wq zp*@GdYrTdRodM_hGTF};vz_Xo;|28~E1~rDV2r=@dg6^{d`~Z>FR9dfgI`(4BSU55 z?8oyA3eOg)n)-yM;27+j8(k5x?A~+;LNj3 zm_QOzi1z{ASntPQU#Yot{lmyG4?^6V8)5vObiQJVBy%CjoOvuP9eHHGm23B&PV3G< z+SlbO7FH9X^zNso8BfJc0Ek2^tG_3+j3fAJ7`RW?WLW++IvcG)T5`bElZ}ZrH})zC zzTMjPZ`m%uaN+0q%gKP0&N9M_Z$RQb09qX*0b|Xu91h>^ z3EAX}hD{h$Rt87qWya?(Z1;>%UO(KMlL5Ai~=fHWE^p@1nb{in{7WG?eg zB1&_1{x7r_S&VyaTvooqv)=J57`df!)L(8nl!5)?E^>HpUYb6MgiC-1w#Ta@MRu(J z9OjW8R=J4!AJPvbKtl;L2SrcK_^6*+!`|W90A9Q5rSnrVSHwxo{57NqGE_yvSUt{o z@x85{1~j_O6?U7a^{ufk!RMqM$Eeif(gY+Vm*qJx+l>HUyg=_-3`Iek$;P(8n9LM2T^ODT2r zpC@?9)A>}ra-(Vf(WJHEvc{*h?;G}Bv*>4K9BASGmqt#ySH^0ZhMEvt7k#^DX&4xo zL0C?8kejI)faPof8sfwBN9%0rM&>fp0~<4oP!#1%QO@l+v}~8rAX2>KHA5wn?>C)N zt>C;IJ81{wYvlzqbvbIF_^iT$Nz{&CxO2l=V!ow#W43$eJu<7E7#G7>nAO*dqC2Zw z98MF9Ocar6(5`bFpJ1`P$s(?P-m$J0*XQt;CJ9#nkg5Fh-@Cu#=3$PhS(5d{T#t_9K**C%nt*|-9PuK)W5H(c{OKgGILv057lDQqY7w_9$+MwqDGF)J@q)=9-QW4 z;-f)mP8TW>xbHJOeb)ZY++(QzxtbR8VMX(H6!5bR@dEXIJu3z%3Booj&(8;U6D}&b ztk!JqESe_VmhIPf09_;nUcFi(AHHaQ$2C2zsd>NM4#idRUnWs={hqIPcArUqS7a9> z7tPxHMu@Oo5wT@i;bmK}zRj4!Tz&cL9p4a!y=x^{~ z%QEU@P*`ekE)L%RWu8zS9hdbnRR8DNKHWY#*a3pJC(6Uk6<+J#ZA%rl#a&(j0jQCf zFf2R_PJrJVUcw*C_s|Th2L$i84{sdkbrPhnH>$Mb_-BW!t!y5e0_Fd|+^gv$$TyDCJM;G0Db);zWA7n8n z5jQzfty{vOcR)1}tgB>(`iDMG;da?}UUW*ZbX`-{LHB?fl#}|w)1Bj*IMqMjwgxd& z1=KTq{|@Fdo#owYj&wU8vp18-0bNmuW>&bL5v;;>24#5xyW7N)xjP=}p2ypt!y1Tx zM@z0VSP)sZfKQJA1pV zgxedLijn@NWRXa^uLc>T#im0s`h(H-N~(lv+iY0gxP}?>3Ag91nvHLY;%2LqAXj=? zA&n_D_RYIUrnqDb2{Uxo;h7o&zOO#I9#dC`X_kRH>d13HiihYkz$6wz@;R{FPM}WuAY6y$zB- z>OCu&(h7?*Q@i-I!o5a67H?wKMKV8@?_{}}-r@r!0&z*j`kdFr0fo8?A$yKad4VgH3Mi?O3C zG$TlvRE4y>PiMUZ%c|b8Ki8v&>W8Duff;A)f-+G?xtAH#%LPqwErA(J+mchu-p2=> zwwqaFv^FQUyfzw>8^Y4xn3ET^#&C5WVPI@&e5Cp*#ZWqPcnfg| z5}5^w>%t5<`UG5aytsN8ieD1p;Ra!{%`uXIH7V)9G$F;^`6~LnZCa%UDSOgXdVKH` zk}_%kxv()mY5h*x#KTZ~$aZ={9-V(H;RK!yM%n;Fjwfl#wygF3`IKQla+8$B5^z+v z%^l9Do>}-&{h22LNHd#wjLpI7@FF%$u^b#*C$n6AJXF%E2-g$ zI&ov7FDO(RtG9nv$$j_9vGOl`aW=5;wrva?NJ+6mJ-q87-V2aHso#QD0As@+f06Cj zuR>ypfBQVGhCX~BU%%VmT{hH4bJRhdDbvyUO@@{A!6|5{z9}2?2KDXd7T+fpqv)ct z+Ymtca9=_!`z(_f4y$a+6AVP{flc$pu>FxI_sZ$|gYJM!pnc4T(6}ZkdpEY__DCy- z%XI%sAC4lHNMi79B-b4)9H`RLs%=CNgOR74tE7YOMpyU77b{ei#`RlQtxa%|98z6u znK!L-4&dPVVJ{>vkCSds4z0J*vcZ3&rwNKj_uTNMteA}d@wFd??1Il|!J=@nMQXL= zVc(@gKTkpcBabQuO_MRB3$Okha9EuMVa8?aZdG)a85PjQ&;2$f0l0YV+wAbPE}Syb zc?hh+lU~?x>h{gbLQ{b=KVt4{U7W+D(_Q3?=NBsfx7C#0W2oz#env+ovp7iw>h-|q zkBAPSIE!Xsg9K+MwtqD0{Q#CU$coeB2j+_2M7|1``p)rn?E{n&=TvE*&RxeI8~IGfaJTa_DFA=yVG46oacb_HS)U3dnm!%T{y~Ohx!lCx@2i#(|ZkW zeO>}XN@E~wNRkRYdr=z)VKD}2VRtb>8L#TC*9%s*nzAg=zwcmJ-n*FxS?R%?RYK)V z5dC>qzYfPOH(O~t23_ftr>}7zKU~XRNvbXjWT!liMBQ=%V023W+8VpyPxsrP`4JG< zV~kAOFSSKWbP-GpaVdUokaoK|pi3=*E4y=vJH$#E3k)ytO<-Eg4WBX!_f{vnRY@t+ zkR3)wCP+&3MxaznMeP6AZKn#KhY%gZ9LurjA}f^hj#;S7f`3~LA7N9aWEkM=^tB?B zStX0eDO}V5pIWSoCC*eKE9y~jg--vIb$=g$@T&6X^^}jNd8-8!fiOYGt@l_QxLv6} z7QAjn3y-5TlWDI#BEF0)ERR+tz-HcrG!!n0zOf7F0lWhtH5l?&i z>Ecx1b@H{Nk@H7fNZ&JY7afBQsQ1C$iDUAE{AFm4v1@U@gsvWy&+l`ZW`=#C_i_DG zORY&Z1*09`Rb)#6LjR%%y^vnhk|G}=wU97k{)*fZ7`gUuIvZhXR}TtoYi9ehES0Ni zoGmN>!2ugwOsK=P^qAXLi+ii=<{+??U`HW&SS8DcxT=V{ac`B2gH?$2Ua8aK&xUi_ zO7Tp>3iWd{t-(%AgUYD~gcr`~3gl^o2buBR>A8HE6oOA4?t~@|lDMG$iTJWUQXSro zioMEjeun~m`u0rb#CK!LnD)8|9R{*Eh*)NTBS|9Y*>MLA6l0HO2EJXPP@w_B#H}dh zx3MBVNYmfoF4C!i`!*3AIqXH2!XzXJy3=LBYc?NC+wD)gooxae#(PI3BpRzGGd>O{ zfehr21Bx`p;wW@|=8a#X*|u)kxv)8ZAr<$ic?ivaZKBAF7iM0Ec@m#l*JLq-2{S_g zNKCOgJiML<#&^=Vwpf1eSrgagZ1~0XDQ)p(w`k@qG<1AU50^glso*Qpi{AJxPDEe_ggM4{mj$X8ibrvjZmTG_jQLP9>wFmzu)@rZY{j zlHjyk1h0=-y?czjNJL7NAh9A+q65@{=igmLQJJ;htteEKHriJN%x(^9)dL>?{Xhq9 zysv27Xvf6Z@8GbF{haHS5cA5Ua8KXLAWP3!hTa-4nulrlwV(53j!~#YlE*7b%w>7` z17RcPh?A7ed$-DSgZjLS>w&qQJ}&CcUZVD|W8NlAy@)`yftRmO8FbXZUxwjA zc}FT2wzI%EI}jY&;>P-d{^Zbs_EJ>W>o;oeAE5O}CGbsyp~HL|R-j#Hwn3dFZE}qJ zmvWAVV{<{}qD+SyL{;Q}CTj3n{QQ2gqAg~#*eLt+bYw2+=<32`gzuM^$3a~Jd_;kx zHTu;AgnH5)h>L6WOPY*2biAiLkaN81!_@N)&>wOCt{|ZYhEkO&yi4wmksv$`oOUFT zQqV$wPTV3b-olnE_Y0bR2q#aEdR zJ!-Fq+(sun9m1T3B?S3HNSK8JBpRh81c}Yeyb3jxH$kW3LM}f8$qikya`NNui!(gj zbh)M$niWk=YxZvQ8^l9Psk6g>;UPe$t9btCm)04#!5TVLG^y+?J*YJ?vxv0^n|4xLSz$h9*o8SBNq6G zdhe{#fIhq^Sc_@!P>lFA)_|zyh-x7Zn-D_b5pZ~`I38M9qDjJd;B#Gk=xQn!8Q+_& z&c8tFmr73^jR)40!F9}tWcED^A)+Y<*}@KA~Pg!68M9ZsT? zdGCFui}jhEV8>7sDTn_bh-_&`(PlrQDY}Z?&*T0>lVfp`REzQ}RrO&puZQt7wQvQ* zS#^#S3tsHRJ;_^NAgGt@C3yIEmr1@StJ+5xl?8?R(0YU0q`XK8iLtAaXl&Jdib?B= zES=I0O4bhS=h}|UwoPVK&<5Ce7tCV_`_%8{b)?}94(T;<4E~zGdm-gz+frY*hP`?c z#1~n;ZlepC-eBvHrf03eDi`A|9!+^2KhKCIm0L^l`WaN{KJshg^m)-`k z`c}Gm@5=B@u~^bcR^PUR$gQmjG3y2s4>>(XLgV(-;*9AZ`LisuHilSU{R%{1Cfbp? z_FqiwzpFIwaFiv=JiHzjmkTxqT-63WTnsZhiEW;rYwJ;aB=2ggIYQ|W2fbP?iq&W# zSA73Ceqxqo0{%FtNmO)UhI;?hObCDKl1edGyMwhv9Pfyy`?X~!>P)~L@-2&X3vnp* zWY~9-^OzlH^_y>>vniLLHTq5npRy_Iv%N`&0FDN-#znMGe&=wk8uJporGavz7`HXK zT0RFI-Kc!n_WJv~njO&@RyID~`*Z{r5Hg_!A^=T4fOh)93VHOkVNW}0J!!3^2#HN9 zh3Z@oaqLH;m2=B5tkaU(*3HDW<9p)$=f|b|Dd2fe|CXqHxQxeLXhr|Nw54|!bE_`k zz@FU3VRwWmi2Uy)iwt3!`i*k(7UX+#_wBa%hXGYvcd@nCQT4Cg1#ww*LO5o3iLOR` zgWc?C0JFHKuYgS@2D`ntjrh}U*rnhl6D;8rK;Mmx2FLP&&wkl7J=<%|b} zNJ`m~r2|R&uU87ybWuYxX`w}6;8rD)tr6A!@RcnfG{|x$pHmlDDV9># zFS+<^jr8w*qxPHBJE1JHkV%vn@RgQ!zOF&6A&3Vp-7QEDIt-O~-msCwhZ}JTnAC%_ z`nQha-(DSx8*FMZ8YW;Hr?L9IF#H6-AKCm^f|WykaJa2EgVm~i#feBC%L|CF_J*NR z8}JmBJC2g^QMM-frZft9-Ab32q!78l@)k`(KQaCJcQerPOh%O^7ZS(&=M88S^vtA= z{rc*w*Y#jO#(2|>T7VrFn4fCao#mj5BO+tbh-qwoif+0+MBm6OnUq`4n0qo%h>v9N z#^jHnS`>-VcGkcUp%)wr8@KJgfg^4UC-oQW6U7;7G?hdwDv3QO_KQ9CG|XoCZhfhF z^Iywxuakq$u)XHxo%x)2J1Bm%ucA%9z(huutAzuix?5XEaUIJuyaQOcU7Jqu1!q;Y z#uwjm#Wh8caE8j~PzruzF-RrQ(>Q>I-^4AT#Y_}5s%OZNxS>AdxCuEo4!LRDhQ^j5 zexy4vtR1=~_@pft1* zKJ)U;Vwv@@bC%QU&viG%1FK`mtF-{*%ELeZx1{S@R8F_z)OOIlCF^e+wWK$uweRo$ z&VKZt-+Z~|!o>4-;6D{rM;W(OzQmkD^X6?O;_`DuIk8UB=-Kt40wXvM&IR;QKFE}W zME!ZlEmEFTd3UklMNvXfV&(n@mFZo3x4X%?5=e)~3P)EFDju4^!rkU27;p$B?-J2~ zas40nZQ@NjAGF-o)b1pz8O(X2%TGX$Zn^OuQYdO_l&4&RbBf6{@rq@}rykTQapN5% zU>uQNj+7s;i3P<<7H)5XMB9^N`MLCeeCO3L{Q!O`q)=v}Qgw7hhM?a3B`ckqVC>_nX98=^I5(*uo(Wa(AFn&r=W*&m~5yJ>Kz=D8-yiAi3Inkm+ zR-s!Z4UE*bupFM%XzR;P?`8Y9UQKvoYX>9Wv**GjPoain2^X5eR?)XSKje>h_S_b> zD)i7sZ^Di6znEvHFnkDFzhGe zpw4@&qNBJv_MGGj^)J7#KQ>x=^o}t>)Q7_)qekB=^pGv>&yI{K*;XpPKU!*ms4&gR z%Jq&w1qWz0%WpYyQC3Tquoy2^+X*bQrPSWJq>bo8tHjh?b&)6%DV4!7yH`*dTK5O< zSZ1Y%9CQX1o4}uajrmF}EfQlr6M`4h%;($7Am4obT@ zWNX1L{d%70mN4uL-43tP7Aryk3DGUl;AB>+4vd{0Hun2juGm?8>CWfbbODt=v{GF~ zATBu|{ZyV~9*(m#H+C`=anO)%lFB&7alC{qgYt1iU+gRbq#(5Ls=xZ3&LlW8jFThX z>007hNDWx{1C2@VZmlTJOu_cO1VE-x&|S@~Qdj58AwmxOzOiGs(cz3@Ek zroebhEj0Jf9pP3Ls5z!VOD&0O<}by5pe1J(VNpcvjk>4gZoI<7dm7oI}E(R`+H6Kt@IT8t2M1a|`Z4eiLcsm9ml zN{v#3J|#gL2Pj-NG6Osia7J*yi;@;QD|I1-m&#%{-9lWCa?dbi2xlcSU#IdX0i6J< z;J8dR0tRdc6bZsxQTT?z%NL;x346NXR}Mb`+Lu}hzgSWbb$x9lM5!-*kymX)A#On? zVWb-85H)RyK^yaE-Hhl)(C2A#QTg&qz(vAGoTC%knd182(Gdk8@)i~$FVM5kHx9Qa zaaxLi)i*70Z52kIigjmfdOHN#Xhxs}q-1T7FSP*sZ@^|5AUy ztoW9CNcH)_P74FvQ{nsYsb_#_8mkk7#O5MY2EUh{7Vrg@vXn{a>?l=owCofJVYly9 zgpo-mD@$vJmg4cP&nvb)8aiGPaJRW1-85{`O)`Jh$ulga>!bycJ5K_=1CnyVn-n`6 zoM)P779SDA6+^-_c_yz|yezhP9bCCwtmj&hZTjSY3F{|4u1oAH>OaX%Y^8!|@Rmu) zawOKLI+TE>m}@KZRp&e|g{r5^8D40Iu(Ga;uB$}tWSnkcXBvjsOJm2|oka^5A|(b2 zoTvI>-=yndVqJM*mL{E!91j!LKqB^Rr9FlX#jP2zu%GQK%h_sjGC0R?5(pSZ3qh;c zY$#DO$vrs$?SawH-UNV_!gikF)gvh|pxvpDl?Uc|c$Hy* zIRRES21%r@q^XUOijCNZxX}AGA!sr)`&V7sA@d9!Il`Y~A$(q*qsT?Ar$vQxu%N=;CH(qYr=t$#LXn{Oaq}D%@CLBxIt2Pm+)Df%+AO(Y2CjM>o zS$jCVnW6QN4x*W#W~9UF4LQzx!WyjEIQ6r^<^@wo6l*Rz1VyqqV|Fj>4T=%UGO0Ap z2~wy3>Nb~(%#+pHelbbbhsXeyZkvUBbWr|rNGP}UU7Ne(=>rRTiNpG#6SbaQSHP`w zEYcYlgpZyF1fLP7;qC0{GtcPV5Djmh>)8u_6k;(Je7Ew?JqUd?RT}cV1-Fw7-|aXH znF?%Mk2}#PjrER4v_Hv*+z$;kXDxrE(j)f2^;sh({vmqZV<3@&S)VS_Lk(aZms}w5 z3s__}(%0;9{#$){SuSv>GwD0lp94r&z`qu!m_6-%$m&N!q(!_qb!{GM8MahdC78EX z!EB(8xXd7Y2XWrxp1_O3Q` z+vg_woSx3Rv2LVkLAUJC;ReCDtw?xDxQuJ0QRAC&$e^v>%X4uM-FJROPvA;E&mqfO zPU#s@=XR#fE(i;nJLM-I`<5#9jp|vT!;K^O$|q&2Z_=P8FbF0B@3CgUZFHRl^YM|M zN*}3bQUd)sNszpmK1{}VXL)WIR*<#@g>hv`puABi`uf|kc_WI|XlRR3CDyZCc(a#j z-EF&rju*e;AZ+WzcN4h71h^X9;o;kDCllL?jhVGwE!Bg5SrT@}_WtYJz>JhR8qqtl z@YBR;;K;u{icwnX*{rVgXHUxx49c#!6e-Nb+~}c7ReL)W<`g#N^6X1`{)`G_6S4Z+ z@Db*0E>;o~e=XG*m|oBYsUKkdJP3E6S;1_%k^UKu;G>6!16AIW3ow5o=~B9B{2Ln! zhkWKXs1u+f>Qipl?Ng?MTsmqFE1vxAyl=lG-{6b-J^8%drtD(BOE2?2(|k(NG2U-5Yd`KZ`2*RT4rGN5`qybG?2xs55k<7K zqx4;WY-y9#@ZC)5s=Fl1-xPtTpPvV5$S=HS1#|446(ycxfPgLB&y;IssAi>gyE#sy zx_*YZ%gSHinnp(u8MCkBmNaarl7Fd%KPeR%i;Qv+YVPHgMtSjUBn0yytQS+_qc2T* zxCMM5kG7}-Ku$_@-FW$~DW4hTY>)P>4V$*y08a0QfbUc$`$7wZDAp0-0H?_(UG+c< zOBJ@+x{sw{pkZ4$duM^gA>POJlr($_5d(Q%O0Jie-?rvCKZD@jk#MT@E}~}jQz%qH zqUgF~T+E->5j-v!W>Y1{V{*RJyn1g>S<5vFO3jpY^sWZNT{y_TxlvVm;);|GE2V7^ zq*3uz=WGotH<@Rp1+rtyz4O15i7-lYpT-<}BrH)MfON$|4w}9>AJOXL-98x^{WU-x z)08lqI((!&KS1e0*d~KjQUnj3QQM|m8#Y!xGSHsgj!EgaZ;nh$<@IL}`$m%G$7Lsd zy&+J3;rB94e5~ATJ@ZM*Gi}-ofnHb=-+6nSS^?I*`q~-bd_Y?qfT91)q$LYezy_8^ zz4V4og2)N(<>Pkiu<*40`2VNGTs;4!#28>a|F#M%xvyBs6WHuo>am8C`fHd;&LuU&)dSf0;Zm)@gdizBE>tv{}aC9E~%nvlg zLuVGfucG5861ccd5|14pw!+6Y(B|>(qNWn7Q7J6I72B&*)^WG zRpf!2p!fIJy=8O$C|rBJ7{!#vu+%cg_k3z6@5MgrNuYI2+>JL?9`ADq{KZh@dvKin zib)4@3v{V8)$4)8`snUtkci2gNEx|{w@uq-Uc=Ums^00DfFj{Aa2#_&OZ<`O24L-} z&?KP8W_octiGLwS{|%*ib=^Z2l0h-wrRvU!j7F7Ct|UXd-Eyqn0*eyFVxY)Bt3e_s z#A0q6N8P1dR4*xPn;o7{fX6)AJf-!9xIdk{dnH=mt8c9zN>N!xl=Su-)K?3ogVfOF#e# zg{;84<+olMFq%)>DLb~Hywn4Ii7<1VxCcwfbKClO$OEZvkt+HthKY(l4`29-o++tB@|5A1{?E3aNb~&0pI^NZQqO@&_rGqGf{CSI zmIR<{y)8czsd2_x3lz~;umCYH7*)aCv8$nQWCyR&!sqIkp+CFvNK+tTyBTaQMm_GZ zl^S3u6#+~&PECG$yawyov6%&!eM@9iAmGt0{V^oTiRz1R8yC|4qbtKf`pztx| zaBer2+NcKrQr%wIUA?f$!@n{YwLY6~Lb zW_z{2w5P^>dwt(NGy%=*HyJJ!FyS@oKG}E+%-l88Aj!0}y{fnyrW*O&G6pXG^B6WJ z*iZ>P@HWvvb;?eLWPCuO77LgZH@5!ZYk1(XCRb86ON)l4yC_Bk_ zKo1BE6%?^57C2D<39%QT(Z5o#+K*G?9;5Rh{HJ`{s=`f=J2(ZvLbf`KChpVd?Vz}T z)(j9!k~$TvFagd5I}E7PJ&PoIxe6Lt(YQ2chpM%>(y;2aq>=O!42yQ)YW>nkCYoYa zXh(-*dhsq=p^oiWsn!_XD0^s{#@nQBeO}D9-{ir3gszS|@{C8p9Fbqrs4+p0Q)NhQ zi4#p5u=}9)O(R`0s|m+s@N$3BZ9s)ixh#{ff*EmrC1}KUQ3| z(ipHMCmZGAvGMJw!a-O>7T!@N^mcy<*@>}~7G5F>PQasM>;%{7puosaw8C8iuB7|Z z-{J@}CRJ!5^40l>J!E5hyMkVW;8(gm3V-V8e9>o2tolH@%O-raj|-xI+0647QZ@L$ z{eE=!2VjkQ$Lwee=BKdy_y{s6l=w#0DPywYAs92FKPPN2iJDx?a)x{G9(FIm2P{ZK z=z1oy(*+lkmRCoZJ(nc{n8%_m!af8AtZF^dzY+_nKA1))2oe{KiscH|+y4~l5RzC# zm5FxVktI|$L3ar%_kLd*kRStD4Hn&qfr9YPDnL~sv0@{Vil7*l9Y5K~R@}7v=3wK; zP}d8nP3xl*{j|x9??#>f3PN+&Tiq%s$Pa__M@Jo^C_@8K6ut(Fxm`thi@aAx% zvi=7p#GU*@{|E-KI36I7UeHE;AH>u`4ZbNpHW0UlUd@v@MiTp1B~~WYX!JzddoV}w zGl8&rY`g?$%X^|sFxoC&#u<>69FzgiQ1}kzxPCGKUX-)bk1mwaH;btr9x^JsJP$?e{ zru)0RJGm<$O=)eJcr5q4F$J-GCad|*={ z6`?4@IF6|>X^H7-C`ni^jZ{>zW?puP;aU||3{>_OgpzfBxt;-v=+)NBx-@l5Z2wM% zk{GTg^~ntJ_jAB%$q-4kOe!Md3NTt@{i{j=O-tJ9qEC$ULyOJ8V9-<6t%{yY2dumZ zenZJCB|XrN@&!-mEKZJvR|V}+>^-Rbo_k3jXIAPLU85@1H$IYU?7$@`F{PJX zmZ9QS#%x%^i#HWR!tgIx-|nnGFUe^2zga6{8&W&5@%=7&?j~Iz1?Qxx9Dt6#iRup4 z$>h<4(Nj1tEVxsO7CQQd;@?{oSBPB~s?EQ>Sg?t03Jmg$KOMm5L{0Cu$yCG?(6?N$ zNkqAREDVKm1yJ|EULPtl9G&3hS|AB8st{&wIlR{S?VqX-q^gdZ6c~P@W}k?j-8r)1 z6XA8=1tzW zqjF(RIQg$HOD?0cq;4kG8me5~0y&udKEw`TEWb}Vze7ixiUa6-YrCJm+4FEY$~bL) zbdF;4-SHhC@G_N%gG2>uU+Sq2ABc8%n5C8YTCUNi3@~9pvb6oZzyZ#w4(u~NH?6eu z#3*+7oK-WXh8z5~#TGyhoVRYuxQ^|L@hou)Fi zA1LT}mmw7QdM&SQN^5CJ#K)ra=$>D|%OQ)^9i&Tb1Vz)Gs*W~JT*J_Vy0=@jYb<)FK6igM+boeRTj%bq0)zA!Ye)nuUhD9Clv`Wm82~)(qnDXSj9^v!p zTxyMTc5}S5LHUoJ^g4ITg<{FfG_NkKbwUPk!1zww^?5j!$i0f3qO! zZD*Qs(L_!HV0UK4nsoFggZe~`S_SsIy`1orM)9=MgE&N1wa%2tgZZ&I!|2-jn$jn_%(v6{l$n(o-r^gJ# zcCVs^FB(3%q2&CLJGIvAPy>*Cabv@tnF(s zZBFLCXgK6v=A2RHk|*B|Rxw#=U?}X}z~Ga2uGtjv_Ha1ME^Av1d-S`*TS2$`omNKJ zYv(jOedlu+OPwnpl*MTB4);sozbVF>tOg{ZyiGGqj`#$gH{7~Hz>?E=wC{S*xg}X& zL%(!HwhZdN?g#0#5p&LN&OzNPU32gSCU_LMzj5F6aFS-JCzE*=yhVvG=SDMzMt=xa z zc9H?w84~k|!tZm|`VUUN7LK(8VJ~{$yfb@WAcyVi@9kL$wj2LM`@L9IFoHM~M56lY z%97jB2G~GzOP4ydb)I;4{rAs4!5u)Jt>##ScNQUv_%iY0n`HZC@>O}fE((%*g8Hfb zN^&#^6sb=zmy&xnll)&-Gi35ew(X$qA9*tzk4eWngF!z-!uI+~4+yZ#IDf3?)xWKh zG=g0GbM;PJ^oit`E$0$+tg-2DmamPi1gaKt=#g*wlQSUF4yU!K+K~i-@^(NPNp3#| zZjuCXRf=e%UwD-(PY3KuG;IGkF08QcjF<%9l-TC6Blij+H+UHI!+-5wzBTdpSggCa zKU}LXNmZbdVp^^KOFojryC`L>EKTyql*gxI1_rycrqzKHB2Uy`X!h3?@HksdxP)x; zwcz*>+|uE*CEI`>kcBgw3NpY1aktoECIxxd3Y1sVsI9L3wJrD%XOvhph5u=eVycl^ zvIT?D!!Ud&f>%9y=oT24sZoSRR-gw(((KhI?Gc2|)7(>ct80z4*c=z^8Xd2*bQjwX z1L&{4d)eui@S{@V>o$d(d1Wea3{*JU7Tz95MPWlHy$H#?M3T4LeP|%fXje{ah{)M2 zoP(34eHJF}bQ;lqeVJO7R>2H94&NV|7Av{MkYMKRKna&t55kUHJ1j8nTj;jWirmVg zV4DxiTlnVIJB1@oPr3OW)md9q!e_EEH?wHMK-R4t7yO&{Q2d32Ck-m_kLy*f&X6J# z9ABgjXl-B*tA;tp-vR*n99igtWX{V&ErA+Ikl_#6Zwkmsex2LoR>(_qis0Xa3B6pg zN`uD%NVP40l6hpNnjn8fW(7jN05H<#K?Q+np_$zioslpT!t!ol8J_0_yn#qQe~=fS zd#796s_qDkBX^d+F|kPw*v)aXg)4sWD`ifeh#guXp)fz@O0WTYP!9quhE&8hI&9P{ zloHj*`2YwA0TT5l#$TutQ=FiMUxK&TrIr4@aJ9f4zXV5b4ZXa<2;5lX>?;cb^|OEX zkGz>%Z4amQ>QCM+YaFUa+i_DI-LV#*Jr0{+@5kOYn;k?;?%ei%q|K5iKeX&W5z#A9 z8lXIrtxnVq>S2MQtC$dB&lZw+%Vvyb#9p9A=ng(0zqeZ!D-a0a*tN8C!?;GVfO{^_ zLAMZrI0R%6jieZenlPrJaW!0>)vXnZ+F(~953_F`G2NiZfJ-qT%-{JTIy!J3E z{L!dbED?<0{%4zum&p+wk=F7#9MS$2;d!IsoJ2s8GvM$eyfWTR?a^TlD!IflTo_+q zC)a?aCByno#_$=RiT=9m;M)6YIFze{x2&m|VS^3ks%RYlBgrgsP>2)r?oj=VnHDN` zRmYw)5bsLcU)1TeNQp*NSqZA=ch_3+oD#r-kAWg-W{%~Z zD*Nc^LKk1bo$w|$tzl%N$+{e|F6CZ=vTQ13FQxavY}tP93BzOzVWD>Rwa}nuQ#I|G z(!g;ccTjsM|I8ZEvbjZMze89Y-_Q$D&fH~~O8`4*d{~D$m`(=L$bu&)oGQXRQ0F#m zem(mh?@zvKbSg`rn1}e?I(t3W+=^)UG6A1{mW1}BB1}_qtb@(kXqu2qCUn8xs@u+- zTA1LMq0t?ifol&aAvV-Ae6dgJOd&+|B#kn!R3IH$gAri?$R+kgSvZmu^bSd*)sz#% z6;K&f%wIXsNpS>;ysD$~R1o8NY3iLk;zgQiyMC(=9W;A89Clx|<&wz(n;w1ONpk*2 zU4BK@xbLL@u~~WeXNWW5kbtK2MWMiGV8$^{g#ETmI4xSOw`R%yK*pbkknHHmdfBcK z@Az%S&~$7V%)<;wNPOK^=~gT8*~MJRQvhZ!EJoe~klDMj!4<1At zHkspNDi$^wiVI*KpRgNsFKxeq z=o$|;u{QpH!@g1+#p>?NAKXQxy{adswR|}fJC5xu{yJG+Qn2nvI1kECL(2{ikL@)z zbt=NXwroGg`Db85Z!v{vOc5axf28V?{6G_}*yj+bE{5yrFE+5wl&?{x1_*?!(EHVX zuAvhlQ}q*AzL(*}Q@3|hz_;58IuvX{uM;Kon1@NYAqSRHJ*o+A+%ZG@R$J>XKR#rK zVH9jUo+nBfLaIrWjET0xYas}+=bV-2>f5f**22(?fXF$iMA&gf|LE=gzS^cQykq+R zV^SPmRuUGH|Ium%1mKwE&Fz1=T9UAGb8!AIola@X)UI=(^c-uwXySTW3`wllgC@Dk z%#k<+g2*8Ive}_4r6H(zGk-n*-c2xgSR{dTK9M;{Aeg?J5-pD2B-8KmOANcNe2h+a zFzYDOPL1}h-(qS@sT!M<$iLXr0OP|>$ZdEi-yrX8nZ$R>B}|VS5pPkiYXraWWMUdvGFl>j$8m8;{`$+R7Q=_ z@0{K4`Mk=~lnmt28^or`5ygU72E#)*N5p?jEuG~R%OLtDfGP9joo$iB_?4y}HsDs3 zatdTmwdhtyY|%x-i1T794-rqeqZ{c&tpjyL?9P8> zz$+;vSB&#wBE@V@ygmi8^Xm4Hc62?(CONb4Da4bV1h{nHoya<_l7H^r#Se~i# z75@&k>|0QCEp24fv2aqQ;54Vp&CM~U(zg%#pMQ+w10cG77ju|fI~ba`I$g4 zLrrI3z#<~bNiI~CWTh%e=U=yU;L3l+BH489PeBEWUxElxeC~oxH$ep$&k^G&|LCp@ zX3#;`$l^ioOjr%E0*nC2HmJHWgiuJgrmd}U&vXcWQzd6*B}oToj=I+a20@u6cgP_- z=WtPoa}EdTHbfbgG0pqT%wrJwMF{Dz%}2wCq2L|ki2Ag*UYk8loxxxaM~{E>^@gFV z8w18E&NiGwp@L*n$>5P|8z7Lq8Ez3lmJOXqs}CXk|5K7g2~wsO{caJ5AMhkg)8eXr zrLezBJc6<$ zu%KKAgA1m3jD4Pbc!rWFxGaBJnuXGL?7LmB4+meQMP4S2Zg(R960UVb@dl=)erHy$ z)|hOOu+f9YQGsk=P$e}KMT>~qdB6e0d$R#tfUx#4VlT(o(amYFHPnchp{uEwB4<1e z_6=HP*!dX3jD?;`)kFm#AHhKF?vCceUJo?V_TU1o3mh;f+U<#ES0R5$gae4SGY4}C z?ku)szK5DhiK$!_XOL3=c=G3qWPr{>DS)QW7d4XNO*|oc^ez-Q^j#m8NQmIbd(Uy= z^5c}-MDCk_in68<09CN4JGsFeX6h7H8D7NZ@j0p|1!@JgqEwT5Aig)M#6(0{3MTTL zM0`y)+;`({AH1P*fu(=8qwGJ8HeAK{QaQ&?4W}_QkmU2$Ig*g#p^e*~!wRV$b)H#( z>$AQnie02ka!z8O!3d5~=UX8JVhF1kr7CcCJ}JPAQ1fVepI9S^5Y!4#=(}m(HrYns z`;JEi9i&Q=ZEwtl`y2oyd}S7_J3=N6nKL|Zd^zZAt3z#WeBgh=`^AJ8lu)cnC$9F; zzVuFVuE(7(b-f)16Uw@6pI66oV*le^c!l+5fvS`nWcd;I``*7yMT&q{C)y)3=G{|~agKdtHaK8?DiT98zY0y#>ySdFqUzIhp{??dd|04BOfQ87)2{=Ug>_tt$uv21z3~!af3umK@tJh= zw};;<;Z*Eg&M*Jj#t(cAd8UdrbyiXtTLMg5l(~Is@A}XdKKN7S=;EWJ?u~}tWH))L zNa%(<+Q9djQsn#MX3y*Oi+=!!@zJ4^k+~GJT1ZI#5i&LkFHB`_XLM*XATu*CGdGig zNE8J#G&V6dm*9K@Cx5iMWmMeHvNencch@0EaCdiicOTptT!Xv2OK^9G;1GgCaDoPx z5HtjMC;xNrNzS>?*LSgCW>;6$uHIeU{hL`(k*laPids0C0i~QAL5!?SEPMb3MKvb} zQ%6=7Mm3<7yS=F^fQ^ZTg#(d_O57D_3bJu>lrRMW`2gG?Ykz>UIS3pJ_F!S*MWg~q z108{`;6V$3nKwWY2r||1b_TKnXiWbBDo$=7Ml(}4@G#KP%El2$3l0%?a`tw$v9bpJ z&cV*e_&d|@XfY;$oT<5;lc$@V4ZzgV0wBkv$OKSw@&p@g05ndH05hPqsl6q@$r7Le z)B$L!ORA{@q+HdMHC5DUnZSwF-JPACT>qyQadiz%X$F9VsFH>x0I0l=AZy^i68v5jfP#%V(9sR}J4njuUxN-{ zDZxQtJLunqfrSA5&T0QIcYqrZ_#ZUZrf&blDyXO^031wh96>-wQ%7^KBghow?glXV z#|HidTD<#Lfk1$`yQ}MO4#oc+a{V{W|CTO)<^-OZfxWN4spo&Mn5m_#n;!X|@;Hupaf6FIf1J()T9g$u8?? z=>*{Yml<5i{|^066VUv#RcXQd)WXSs(cT+i0klM9R&oM?TY~2QpPlw!TvG1#_DZG> zK$`!f>3=(#I@s8I|EKGJxo87_S4*Sh`H$Sm|gkbhzHEB z@&|$W)cznapZXuf0bthngTU;Xe-M~m>kk66YyUxDcAY;6%&z-y#0&N`{e!@sW`7V^ z2{Tu4X9Gd@KugdcJNv)w|Joz}JHW~T&TR(nYT#eaV3Ya35hs|y+{qq)+*ki=;o$gf zad7xk1S< zn?G6Ee?yNyG_364A}s%;p!ysFOIdhvo+B1FAuQI<}V1I zi0xkxybyMOL9kHve?f46xqnTD697+k@}GKl#BHmp{w#8@U7B{y9DVbOzUNs}l(L zUme2=mhCT5S-~3p$;|m10X_Z_krV9hX5;mz7_e$?_NH#uf29C_GyNgp1kVd(?F#&> zfxzG@aRfTO+sg5dPte?hR)K7Wb>XYv8M{)_5A&qtNt z&j|lqek^|u%>Q|K`Uk6nT%GKI+BO#82gko$6ivZf&P$&Kd~>ma?cksPeP{R|1XO=6 z-M^y6#GJf*89BgztIp&{R6MVA&FaG@y0RRHMfaZv+ zU!2VOLv1tK!YczLOJ=KJsd$;rYH)>gEOVK33jY%e6}^$|NE z)^t#Gm2SbLI2?CU4$gU?A8Q{nwp93649$KH-vgE;&7;fKn&$Bv{V2CyBw`4XKuY2& zD`~*H6az$mT+i$wMZ7fV$Qe|~qz`U=?(d6bXT{&;hnx_!V$7}l;AC)+;pVppgW29 zvJ3yvx@rO+ol5GMbK1`YHCYn=e8DV-D9M+O5;OsSDb}pKO?1y}g>D3mp@>gn%bN(T z$MwDDQtBwQ0-psS79}qRm$wwDpV1(_pxo~&xpfcUdH``qR#SGX%QG;s%?0@#y|dG? zlN6O$XDPLT&XX)ONXlsg#3L?+#P^DC#pQ3Ig11{fVTaaTiJzBcSPe^Fdqw%PhuxZf zJscK)$H)*LYnNXxGjT<2oosH{NQY5 zN>2y1P>b*rX=06`C+*kr#+lC(aUHY+O(#@nO05cC8)LLqqkkk{L6y*L&#q#6) z8RjOI7(b&3K>&oaP^QLsRkCVGro?lh-rv@LT^_6yEhJiTf2pMj7m*>^(kgR#8d4?n zKDZRd5u!F;5s{X+>|pt%hahQCRNvP~`N|Hg4OUp+LBwl_=aZ&k_=b&yV7Yi2JgE<@ zIJ#kI-{(l2g^Sd9Q{gVT|7GDUfTdsU;XXS8gnO6MZS>)dI*Q84OM1WU$7|1u5^WEE ze6IbhYx1(q;~GTh+S$t_)+R@BHa0TlrCilk9Cf4)B|Zmq#rQQ&#feV zvHq}vM8@~=6`vq~OgoasYYQ744Fzed*9!?sRrXp8ayIg1TN!H6--wXJbm`=JWNc`yz=NVj@uyD5kk=1zFomS!NAHnm^nOV`i`5)|*6X&S z3<>)HSKvyX-)6>1!0=l5(ie{P=;c&-rJKb20cHw~T`rLPe63c->TX;~ecM**pGQa0 zlv#Rf({EVKWeX>o94u`2ocXPj+eidjszoD0=`L^h0nD8vGhKa~15`fKR+Hb76QJgq z1yM&?W(pb5a+OT>N+m$YRzanIY(MTAA70;(LeF)-8!b($y6fx0d{2q|(K3K(Ilbab zPwsI7y~}DiIo4!vOXf9t~z6kbbyqy5#;*KPLJW-5{Dq6)X zY^7TYi|$K^^xT3H=}^--*4&%+*DCgJE>UHIFWr%!U(Bh9Exj@CU^%TKo6&V$7+<0n?~Zpk@!6ILq~lzQ)7V^=Ju9zSAn z6|{0ahlydIlfgDzA1)2_{tAp9ZOVHQK4-e~r77B^okRA$SUbe7`Tow|ThL0$mIx}>Ckw?U9DEg;dnB*m{= zKe^5ybY=HND7QHmrJ@hUSh#*T%j3Q!_wKDT#k)T`9hakeM+DgwlAJ!=r&8u=VXxHVi*#VB3?CSI zj@_^ug?cH=cDlQN=ZhNkdd-}FCU~wR6EiCuP!My(aW9uF6`>#!*{-38Ptshap!xooELUvwpol?t677MvW08dz&U;os|0tSfpcfIYPJRh znsOqK6qhD0yQxpNImGlGO9@?_=I^PYCEXpp`K(98CA#2$C$7@#k190BXOukZ-l_WJ z@J(Nc{Xi#^$WNH(w_O`XhS){g8v8Y$b+7d6`WcyxexWw=VyG8h(q;280I!^+PYP0hj+0R_7wY**b42Bt->MKn7Ar08 zt+r9vYp+JiDXS-))A0L!4F2x{Iyf0L%)&fpX*(x`XCV7xD@4LJVaH2))o&phHp}%w zi{i&@*Rfr7lGq&p~fM5|SXQx$O=2h}3Z*p)P4jBY8tG z2C{wqr=>CYymB2n9pOVQNkXVkzQ|B2l{5)|fSZ7leuYG$3we0={Im)yhDp_r-vmGR zeZuVEJ!9y94V4KQ7eRuOuH}R(i$3Q9m~5O`%0Df?-C!W?&-X7HefHXyvGhz*Xz(Tp1_V#;L{eiJ^lN(` zCAPa~UuPN)8vQQkZlUN88jVQYS4z^LGY&3o($!5qE#=7bb0kUi;dXLCe5X?GI@ne` z3h%~(BCMTALuLB@WOKlYL_=Kj`D${1Md>+zK&Dm%o{OO@#_}KExY;{{;nI zEqX8c+5L7F2NFJMGA|hBYGnS{Ng^P2o?jS?cIOH|s%|CuRWxOA#qQ3W-BvaPP z6|g0;c|77NE=<$#PydTw1NoJ6JjWSdft#XvhQlH z%^AZ}F@4qb9#D`fs+sI@s>?hX^iN^v28zU8@XNr6q>2r3A#hSwDLs9EpZ(&UobJg z2*EQ7-?F(3@gAWa9@6zG{izB1sx?|}BS%haG|7{5xJS~3f=$)Z(Kh=PD8Rhnrkq&) zZ2_iE#u}1RbkAPtX{?Ta>gGei*Gz|?s+;7fEoB);LHt)aQKrsqIOsKo4SwtkGg@=) zD~Ux?P2#47TZ8`b$(iNH`Zjv`AWDUoMu#M9SC zg$noU^st{Rdpq{B+%LM-f$!#oymA!@PgN^SX=t_Tm3I1pj9_v3Z*)h@>v^T=!Qa$M|(zN^JN3>CX;hm0YIlqa* zE82nGv3Z0d#Wei-O5OwD+Mi z1~Uu;-B|$AAQd=*@-QR~VasS@u4W^chNnBWN9KsPxTvsyRi+ftTy6|}Ym?87S=XF9 znal0alswer!LPEAVc(BrD57djKXC6oCk?QHK9{XI`Cb&}kXsTv^9u2Ql-qFTR!0o0sC7*#Dv`+T` z)AkY2hV)q3gVv76y5`b)ei5d@Eu4WL;a~o)vHtBg&qs>Edi}U|byQ9ga(TfCm?37b zRSSp}uV2XxajVLkZz3n|vhY1zR-`NJoww*%{U6B8j?FZKOVJ4Auh|LEgai;9DpIfF zJVEGx6e>F!CZ~XWL!R=VpNShv9A4~4LkrK#x0Jypyp0LhAXt&w;%Nv*!UjRh>G*48A?Q zc6~rcdB%>3>~{vrm{#Q}@xnlx@V4=fhCW*5Fl+U>N^R7q_SZibvRy3k-I7%mIfCsgW=S7r!Z0dOoZOmSK0(WuhC2n=Jd@G*VM&kr>W?FO~bK% zQbUblL*9Gzp=5J@3$&@VN);D;X~oqMHMoc3Fo!z|vMCg9!ais5KY=1Q>5qExjzQF= zxu2G@>s@;vt|j2D%|W4Ut0u^9A4XC2IWkzoXOy5Fhi{S_M>gZwaDYp`8sYVg2rFuM zNvsL9e9y`{A&z4nr5OOT5)b@;xx-i2-C$INll|YHem*Fxuci*OF)8bt=2e|_3Ix6~ zq+`HhZIovoTii!e%hWA4^3$nxR1b{N7Fl8}$z<63-NI){rF zH?vnt_8;_Z#1g}X)!$`BEJLHY;YZ&Ilr|mPM&3LY-ONVI1sVl1=TRq32q$mX79 zwoqk5HE?MjttVykZOkxM)rO!$?7F@AGD%l7BBFZB zxSxYt>54}V;kBsVhY$@xHeoBEO?{B4bM^rxTFWML*&SD>uYi;fu z)N)*BVRhe}Jrt0Cz$)Pk-`Ea*i?-jT|u|K#P;m{HTQV41&lUc+-!=f|O+xX4dQ% zX1!g?GxngQ$qkjHn>q@wnUNzxAH;8RiGMKm!t|0fObc=P`q~Y$fDlqQJ8<0GQEZfA zQqI)nk;jRDWUxbhi;BmN#w8nI=8TXD(N{Y?rSJ}esMVhCfQ~0n2)Y}bSh4xR4 zFh7`o-~fAwmyX26IhP&Zycu#_Iv6|)(Df21OpO+1#?!xtUA%|SVd{Sor)T%|_!xu~ zZ*YXqM+CtQ^7vM{5||!#F1uo@Ge7FhJ|7m=wv=RleqJXTpLBR)L0XcK`lSEL&uN)U zEmDTeEgTXmt*A+)19Q27jD5u!8;1}T>qW>u5WfqGDjG1`)!VDNYD?djm~z-5c=uM= z`-%axJvnv!W;q^Vu4*VJqnhlVXe1%32S(&JMXr0Xj-w_FBz~+~Bzks}qGp}iFURp~ zzQRg>kvIoe^GKkU19`zI|C>#r#%eSb<|9%FPN4R-N`^-hh8bfj4u(?Gk6%hp@y(OUK1IAy7!3gGV(0y7K?sg&Xd)dRom`xr}wW^6cV3L zd<@UJKmRs<(SRV%jOfnNFp#=#H69TD!DtIo9pf-Lzo|z+>*IJrb*NX_pM|EhMg@GjkgK zwzX0Tbg9rf1aZ@SW5q+}+wJ_M3mi%HC{I%r-tzUS5 z@X2BkY8?h7?09<8TY;*0>#UtoXVh=a)jiE}YvVQHeK)X8xHvIj(&W0Sb1(vU`9}6= zfJ)kh2aMI{TpdZh^qeJ@RZWS45(|ep8+ObBPPi#6quey-)}a&Pt3-i|ABhi#fWcYkb??!^d&5qv6Nz*LD19CKF$c zchcbMm*6lv!>Win%I^_+yR2YW(E1S`{Xv6w=tZPb+pGhH_|;685bTOkjL8Y zD`wU*8W%gFbVR425AR;DjAKoIhzx8cl5tT#r|}C$>%hgbzOtk zG1B-~c4IS!lbr<_1d^f5@ZuKbr*#tyjiku42u;=Oct>Z&EkwlK_whM0g%n&nm#iFK zZ@$*G&%Mw>)*AWJoQP*Qg&p^;V}3mw;!xcTfMvKs+dvFcg7#?pc19CUKEU*PFb%3M zj*Zn0lYMoa(J(N;V#Ps!z933vvei_PrK0CNHkHZ3@7&$ooSE+p!!#Tz6#rm=lqV2g z!$vqH96@5a2%E8gja-X7NW#E}5sdKUDb*Tx{32I@re+xgQyLpb|Iyw!899Q7e{&it zkQ4!94nub?IjuysE2nRTep)1WKs45J0!8jrc;Hv^t1rYfQhC3ByK)q?=2(RlhF9c= zEJ;YSc5Z>s=i#2J5MfD$47$Haw8-dg=g_y60|DQj_oE_~aW|fv>ieo_8j1rEoU+En zu5H=x=#{it2rBtvh=(8JOOa?(M7t91;zV%SMT&ed0;ooY&2KR<6-uhDpfoK9e@r4n z{I%>LHGVOv34V`73XR31m6`#ue!A6A$(q8Ltx;}vL_SM7GWh*}8=-)pZ<6boGn3?px1qAl;VhnP0!U#soBq(STRx>J3>EhZJ}EeV0*7c{xrY zN7WtliO>WBrjEPqB!?K!`D)lOS>>bv>TTpUUJ#9Y4^npr%T9`3v>C8sBntX%Pm^24dI6++j2ULK6 z^r?6~;R}&s|5`CMh2}c9{Xum4PaYpEcfUpDM-UY-go!(5LVaUgh-Ku^TZ~ zDqbQSEmFC1G&N4Rihyg?u!&GcieR|gHHtJZ#=_+v)U2c`vE z_TyJlec?MUbhtb{wvAls1n59Y``xDo1&$?13T+R#3m+NXZV2&A6~^{M<> z*t)PyrX;`kC1C=&p%2W(m;d_}-7xvGx{}O)NBqkWcjV=GRZ#F`T?pU8@-rnf}0LymT6$xS%FM@`AxnTW_P6sWiaNI6S? zF!o3iLFF$5Q01V;?z4YFdlza-gc+xC(>v#?0S_5GWVPGaKg~3(JD%-<0Qx4>9=M4l zQSFtxE2wM_bNDU-fUAWayZjV0R#dlLYIg*0=rL#1Vs7aO4~elDraG~YGrye_Iu?uv z6oMo(*uoemFsF2puhl# z+GIG(SV7X2CWeOLpW%+6@|3lIOOof^5<<=N9(ExhGOouo%WvG(M;KK*Nf2%!$B?_& zj0N+o-KB%^9Y57Jj~`-jAY&kITrfjI(qMO2my4Yp^KJ?u zdx>J`DClB1ZD`CTT$sCMr>>!zp6i16Y!Jn0;nM*>SA7x5924&k`FSOO5=R8aSITj- z!W3CMT;;g!1iM$V3D?LeNdD+bX0p2RyC0T@D>NkhWO>{X0}YmnLeUwLu#_)*Wzp7? zn-nFSol0*>Msod}ubL{~rEM>EL@Yj`$VmsTQ7_t-%zA()kzp+1f@$m3_6geT2rN-= zZ+R+BI6g_#!-;PZf!W@FWI(%?Lt0l>F|sh3(cGKbq7SKvDf7(+Cg#}a#~Lzty+e+h_2_9?TC^@I_oe+*X?iK_zonPqjTupgw=nq7EyuugmV6hfw(Fa zjVgix6=Tr6n+R|QX3@hdYI#UrC^8RbO*AH~LuPzG`=mpJP8UmmBg~JMm3D^csM_g4 zkuRewR%);xv0Vn6dP{rss;yY_7G1vh%=K!J0g7vAluy&Y4k~ru z_s0m6YiZWeYe`Rk8qA{hcyQLO3qzC|x0$)|xTrAnp7I7iXfuWp^isAZJz*5qTYoX} z`8evY$+)H3Khe-@nOtEZUq=7DlJ8`%?5K6VA@a!2;o3f9(t{rTGCB%OW(` z)2)l)T!!mZ+kD5d@mCTFqbMjll_p4iQ}r_At+k^GqPebrOj3}R))&bM=I0?C9t_^e z5>EZj(LE7CBt817{UutZ{a)sW_Uwmcn}EpdH*kFR_oc9R_QB z%Uh$D!!(tDh7P?8R#dco>8?@z+;tUGvN|DQt7fXl#NWd1yBKq&2qm#$E#C37xXb!< z!4uE&JW*kVVdy8#8i%|&9WWH%1%)dgpYpiYuE0ZyIt~qLJe1;r(IsYLi4zlM+phgA zLhxsQRXNt<{k)Br#P&0$RWhmPX7@ZVX|E+!-FXv#rP$}wQ0Op2QwD4vBH{LZrcfeG zf$lDQ_K!giBowI6WkE3w4g-qbE%GG5nOLUIz%N0{Z}mMZ-dlP^iOmFHtkme%XM?XkZZgb>oZO-c9qhsd7cm$<7N46K_?}D zj6O0vMdWB#35WV{1$LIDo@MB>%UoMUw?E+wpMawPJ>R!@&xUM7+*yVnILRWTF?>%M zXucTmZfB>j6whZ45u0V>vo$Wd=+7TR2S?BSPaqx{y1KmW8zi!p-|@EGyNnRil3jB%yJ^;T$f1*;(H( zvDFXoxF-3S;T4H2fM<2z8gZkA5O=e!E92c+naEfvX@eMg$;W_#H)vxnSZ6hc^-`GB z*TE@Jh<5x%Y-*9EMTmFO)vjTR%3Bg3QuTveaoAtmT}g8q7&(QunTJDGCHnE6zfqgn z%l)%PozhqIh$Sjft70;Z;fb$*ZIs0_*UeA%&Zep!H-vGWq~??wc}XI*^Bl1iJk3R) ztcw}+R5>PL%mVyTl8}c|x{`n0r_njbB0kDXtD=#dJdsPCwK9GXtd7ZF0ol0mHug~u zrjv?nzdJ64F#K*-953rfL5PG{Q|?Z+Qf>30Kq3pyZfgBYB}(n{06zPF1iQdsW%I`{ z3aw2t5!EqicX7)XCuZJWzNu%8do) zS^dV1sEtSOmW#Q_?fY@R;U_y0?o~b?(9fIOtM;1LIZjX%PMFhw0g43uX|(4fZL~YT z1eT57uH`wF(DWDfuGeWzPF(&H%{3FL9|p+Su6H8xgk{h<;D}VCG0F0RTUql9oo4pK^Zu_sp-xdPpGooc2ALLZMo%| zQgl!9;NDGtP*55tAyY7K8~YXB1nV&?5vlwDakPo2bM;UbMjIr<5mUVl1$#Fz~KGFsb^cHl^w?-#zKq0Mqz_HA{ z;;z@BQou*lom=Tk#SNtgs~XY*8la{`g^iW11jR%#&-2H)UvXU0Y>zjBPpTl(rmL{U zi^{AOBbOFoJ?~bWu6^#{BzfL29D2Lk-v!@zeyIOj%WyPBpUf_-m?3{8` zA-21J%ai?5kxqRUm)ro8`}q(NOCiC*r1 zP!wT`VkA{h!O0s`kq?BJs+cHewa?UV?F8ujA@CXsZB!9|ywkMAbAelw04`BgQ#mHj+BAB}8nJ&4ip} zsMzicQwUu?C5uMEPW3^|D6KE-v)XYkh)@U_7gZI(rIgL>7aDebfqLHCsKUOY@;Xh? zG9iGlBXlL|Ea@hploz?~l%_c(l^%1*dHKNtOOts<)Z6HejTkCEva*(ksuLc63OYBY zR+k<2z-q11cJ+-7DuYJF`sV;33=d6^%@>+4qvG^}&FqD5G+@fR1Ti4&JcMRrgsMN` zN*u~i_AxLkDkrxK`W*RUeFq;aro^i* z(vH!Rs;%@r*Hm1x{^agUHKvJwCFR1AEKZSRdC%pQSBnDqs{n>7xdJ`3;(~sBR$CcW zI0<75M!G4revX$2wtF`f&348YrXDMa;-yaprxYSr)KW(aON?9r)|U*zntJ@1=PYlK z59HB)h;3uF*p7d$C^(0;J8l?GmpgjTGs$}q?fZ?fnxbx;O$THR$#KPhStwzW?Os*C zzj@?;GUIN?$wYbg-A<><7sA@J{hKF3IDJYr`$3x>W}?j7iD2#|Z1tVFA=RBB%rE$H zI>tYN-#gFVOVGcy++?^&OHNl;gK{?at`3hfQJ5Xt9-2pv9F;A!YpaC%x^v z|L3~5IC%j=0s*R0Y?%nNpPmiPmSb6$hR zSrpO^6W4d-^>G%n<_a;OMRB}J!%|-!xLw!cLvjxkuYR|dyUfoRG;`%o_0hA!-l^%z z{yTj7YoBJ28nivLQ-AnqpoO0l0c~7U z76jW2H16zK0Xt5*%DLoKiiP4*+)v|crH{1ua)eJen{x} zRi39dx6K6`=1h6eIf9UYsN`}F&%tYZiKBPY7qy0Ml@esu$~r9Jj_AgP;dS25ct$LN zC}{K3mKisc6o0P=kbv@zF*|J&WCYu7#_I?6tXw4DM;Q#fukUZ*ZYFcKIMJ;R!5Lm| zfQauzzuI~1^HDyx3G#7$Z3N}|2x;Q=<^TdVUXz{vtR**4c>3TLlF|d-V^<$i_iL|=3egX7#Y%cN267Ls67MUT@6fXXd$AYJzmyWn&MX`&hMG> z!v=O%;D2`DlJZjcx%w#P;0SFyJQ&&m%MFz+c8x=cwZgLRlgPC)kMq0__-SuM2zg@v95|CJd(h5Xf2*o;*ejHf+VQnBZ8}Dv2yIs2(i_|G z_;u*>{mOT#VhL`;9f`f7#RUN{7o|Ui)GEU-fPec)&3_ux@0&Ss0ReN5&FUk>2hHx8 zpe8lQ3ss-M))jksz^4Q(oF*j80QGk0)YhO6=^^LP1a#37}dA6~eLdIf@1 z(ti;)Y{=#u{unurrBz6?3rW&kI6^gmy#PjFFB*a*8mlPl&v<$!-cLE-ur9v>@No=I zZ*I3x85E#D&TOQl~G8rL2gFOgUSSDN3)KAjGXn85FHmy}8=2thbRs z2UZ!gFzH_iQ|M=&{gUB!qmzVX0*~6%9emqISvDed;-$2Sru2jF(nR8oF z02g7xCv~w~FOJIzrfWkM1#PLCMtfUuSlxb9l5;k=_ZKy{hs#Lc{|#vP?AN@$0e>tr zvIYW~h{Q>5#}ZT(zjozO?rUte3?o<&0Z%WVZc(j2FCMx8)&?9`Fuoou_bv}&jlL8Q z7q;D_!*a5bJcSZQb)w>I_$iT`3J+xfWg9M-*(;&xfVbIWUM1bbk_lqx=9Dzqk@2b? za;|VvWabvY{>biOXy6fZ^PpJtPk&W-K2tRAsxbd*VS;T|*sMCND>Df2T#lp}$rd z{PR{b1e#1jge+T;qz1JYJ|o&M74thYxzT_eOA^e@#`c5ViwNJNsjrm&SASmuSl4>9 zUQ=YlH~ogW=t8OQc1{>IWKykc%$VQ2PYhw2oXa|7qCD&{E$Yc$DBgf)ch z)HLU^OuQ)cjF0>n>ILpP6Ms#bEt}Y9F>bh*LI~J4jG78$^3h%e@|>61D!+nEl(O`i zv_n=sD>z^@roJ{NZqd!m!qVXwaA@tW=r!IZbNA%Z2&bRxTSmg!bj=P&p516;fPsu3?5h(WKI5J;Kam)`2G4y%Ar@y?+FE_b^-zC?NpG zMmp}I<}~q!!vk5+uq5ypNLc?!FMWY8_N zVOo4W)3bZ0n$n*qD1S&b$PbBUkof<<6`kp0fS~z=0oqe9^{c$k;Ju#0zrpmx42=VD zB!kYGw}03;+`|PiedFHYWAk=S zt|@Bh>m@6{h98Tz{kH%yK+eAvZwCG&$!)SPCo)%Pwb%*Ehe)H`pl1~qZwlxA`_*?x zKBu}lTKmE>tnQMsS|Mxyx?5qvZbvlExhb;ye84bA*-<}E)5NB^b+>e#<1J)s{D{ep z1@?Ts$blxOH6k4+evN<7D?hBY2N#)7M0*f6q;3h_$Ay$Y_B78t{(TOo=b@9umjn7o?_3C?3J>zO-? z&B9!NG^?S*HRCtO?r&eIpFD4#daymcT^BOg?c}Wt(!ydry5m;A_2?O^HGbsrD?O6p zG|=CuMuxwPI+1_SxP`=a#g9Tgfv?W8T}KCn-acq}Px@_^ZHp~>(Wd4%-rcurc4X8C z5&x%+-wx~|bUbPyj9|!po(-4);@Md9l38^VdFw1y$P2i`=L(C>aQ=;#rM;<3M{9c4 zd*e`ja*)p=AhKtxuo7>J^=yDp!T`YnI^Lk&Cd5#iivoXZn5ZX)Q;=rwLq>v+i8ZXK z^ev8SCM1rqC}Y~oN;FK&Uq%I+r_adlKUyJz7hL~9Er2DCMy&}0deK!GBuQtZR_+U9 z%ZjlSy&{l&%7UQ|R^)Hm|9G$DUL+Apu`Knu^J$hUlA$DJ1B++MYCgVMj8*6-__@Nj-$N1$#YOMV7tMxnm=811&&vzui4nZZ$2i zq?nxjFVpaE!gLfWcL@4N7-mlMn#D%AjOfx=@3enYLnwR+^n2L!f!?IS&gR!YIp9bp zKDc3}@fa2nIAbcd<{F{^=!Y}!A8_`7Xw6Q~L<88kix^{=CAW{S*#sQS6iJ{6uo@sb z6O*DNd6;}KUiJ@cm9e-QlL7!71(-rkZ+5%-G3G&TH5DO#f!UDjaJ0Mgz99B0h>9Qr zm$83C5#QG;FJNZ)=})Hg#-;nl+%i3EmZ4vD%Gyx`sGBAt=8Y+zOWz{H$JPOL2k(we zfW0GOl&Tlzycg-uDya*A-G)%CnoAcd#|5A})z_%mxC*#S{5}fwsNJ@*J5du@Fds3j z4k3gg-k2x?&)~u%ZM|y=gN@hvXLKwy)#HDnjm&Xrc84D$*ZL3$v&Dr0OPoN5*)Qph z@rtElmmw@ADkqlWqzMCOGcg#`LNs2PmNfp5zadqNBoN$TFkHZG${My}RoSTlr8cP0 zL8_`oNn)!JW>=2|bH_6PYNf>7-vgT^muj)k%w+5Sen?6!=W}g7ykot;kvgdsPFG*jg`%fsv9p}n`|iPFJo|Z9%n!MY+Y(`nU!o_D6s*7-?m%EZFq*;X-={~W zDeF@)?d<`s5NcAPB2%T)jM-}wXN;O+mO^sUXDi?>PGjk>sOVrWj;rKKlzsj7Js#^4 zMw)-SWghF33HNeYvbW(5oj%f|8p40&3i+is8p$C?2-S!fBCqIfg&B+E>$q&T1>WjX z4i!;y5QTW$Fl~14O$iXB%|A`fGrMbNUZaLls&hshp?|ckKRPCL7WC>a3=2dkMFNXd zck~2^FsShZ#$6-k1%8&{Q)VL7323^=NB$k~FzM{vbP8LbJmjH?3J8!&jNE@2MKyui zb`s_|8GzWl0{?(tv~@FPMfD+s#SoRVBCOicRg?q>XRwEEA2Qp1(@Wr@&6k!n1~jbT zVIGFCUxPl5IQWes)i^Y!RC;yr6aG3S%T2#a(Ts_62b6n-Z2M{mD7x;1O6g=jaZuuJ zyIaZM^UJpAgf2+>@YcQL=x%?Fpt|Gid7VWwhg9r(LDb~ zri(&>&Lzzjf)@FXrfG%UDax<&=md>$v95Hy^(X->8)^_Zg*eK1fYK-teQ+~(fdT; z+v~$l*;_337ow9Y8Zq^$zGE%$5!hlAGE5-eD;nvFHoxp`(g+e4>M+KgbUecO1fDbH zIq{>62GeRaIQ|Z^?%K^#PD3&5Jxu7EC$P@CibsWcN9*`aY?4;7S`pVGD;L%6cpQXj zd+UMsfc3i_Xod<9Xv=>rFFbe&4-FfUGnbSzk#I2tIELl8n*quNH2?nARI&-l8HYT%sgCk04DB6W#a7n-L6FO; zVcdzp2e*`ss>`(8P9KQLZ27vf6!z5@jn$2x!afc!Q&bTN)dZhJ1| z$pB1^aI5B9W)y7kEOX40X*6-6QxF(BdN^USdEShftj3BV!GN*^^JT~bF}({CJKa!{ z6g7|MXoL9Fo1{_F2SKaXA3nXwDtPLJd_(Y!#IC#Gk;m5PgSe9@z{1klZHj_;z@;_i zuB9#7xDi`nzvO@DZ@73E`jA;a{Bvmn!lZZ}Ptr6nXK@h5#c=M*4c1@YPa?|1C9%F# zGJGSxrYui#jj6nFb$K!gSi9xir_Fi>VsAkej=UTg7!U<-=&GBWQqrp9A2u zge4f!$X()At$jL;h!6i|+%McHS3yFhWjCNfeYW!(ga{?GYV zQK4jAQgnX`$V!L}2Q#$Ev-puhqAEBkynaoUBRSv2;=1SL94ezUKwCvD^?wTVVJ{WL zkI~%LPoGXOZCycNy%s~;nv@d%M)(aMLRsXOw5oKPt=9;k^U>Qm?ysZ-3QBoA#gY5l zL|19fuDyQ4%$ssx2MzG&f`jPB&eB;p{B~?pZ(bqzgHq7%9 zq57kmU0O|;>OZA^Lfh53Od4_ekA7^i3KGnJ)M5ma8?Q^c0usnj93w^n_(ehs=`9if zN#K8-=y39fRf04Tc640E>K+cpFs zo1t2ctfAMESjhb?m1er3F^(4BVm;02DK>g2x0VYPG?D=j#T^U5N9=gi-uHao_3XMI zNh8f#ApP3Q{f9X0t&Qn#~GRF$DAj!U(J8S46K1-a=ybE`f71is-FfgZ*te|!#QBkIlCM; zI@QHJb6)1m1-l7Df_T?o3R$u`J11oGurrzx&lL53Aba6iu_B(OtXEBCXIIHrUShAM z9dxj%bnBu7+9Yn|)cP&wsPx0@>*jyaaHm4H+Bo)ZYjyqsNwek2|2*bXDq9#~YjbtQ z3yc5PsGi4=tD5=%MuKv#B|~`?Tm-8w6c-k4D=*lV{bmPz0@?O!YG+&zWajOHs4Qma z72}pC02L~iDDJ@9I{+ShYmHS;-YbpLh0h{5Eph{Rs3xL0QWODMhX9`SIk?vY?cUS<`bc zQC2J!1RD>Pep~c0nu^mVjc86$T$Ti~h&lsX_f?g70m$H}o69Sep!R?@iWiwaMKfbLTQ?sSjO5OU`ZTXov-I5yI<$RS$@A1C7wirK9yM??S`Gn_V0d_G5v zc@R4gGd$8P;HT(XNWhygXI_5TbR<`IWA?n3b@1FXC@Z`VH}Iql9^#Y3n+7QLngagz z|7{>y)pmz-9j!AL91nk2ce*7te3G!55AC5JKPrL8Zo<|Xk%5zlv|m{X(|bwQbEeG7 zQxsk!!b~ICD8VRmyC1nyA>@>|XsHt`WwJJ7OBn*Je|WV46Jtvgys&DJFA*?7zaJVa ztG@+c*cig>DbS4&^G8T2Ozr;Lvd%L%)ZzqbWK4_qzvVr-a4V=wk3+WDIL@dACB7aj?G9&7tak&v{KznP@|@ZZUPnXU#If71!n`6kfmxZU2dV z`o^&IQHi}}WJb^_wqs$;*Kxp_o=>|LnH0^0NNS` zqdrfahho*l%E=$7lmSNMqF=v4R(v&X>^r81N@+?sW8vO&J&&~EIHTt%M`GxTK+g8> zRc5{`dv1T%23hZFaC$1B(7DaAfFn{bXHoqZCz+qe5BHpg+A~wNe3##2GLR{MA*$gp z+K|w`bmKy4{y|(jrQV3A$!s+g*?>3d0b5bp+SJ}^PBeZY$x`g)8=!wjLrvf zPw;EgD<#yDU$e0OEL6E)bl)#elF7`8A9Nt|1zdlUvz}w!ry1D)Zw;=?DGBRu%a`&C z_7yl{u_&a&^zq#h6*WY%k|ciLRGi;qQdViU1(wX?M&FG^>K_H5oh1~U;hm(6C}(VS zv}y%37SnLIPTI<3Q~agJ)E?l9MwW7H^ZU=4$@}nRVQ?;ozwW2_MYsQRNE@g`#^G1G z&c=U-u@3daKxHQ$)fZaeWiaH$Pc=}5%Pp6fs#*B*3Zs-EepyZVpEl!QuMEhEXUIsT zd4{<)zrjE}WjyPTTd+taIr`kpHF=#0Vr5stWwC4)=1@-uRs?Blh!cE^C;XjUlJ6Og zpoZ88sBacF+wqFyUL!gf{9_0g6`h_4P5FO;mbbJ^e?NSSUZqWyr$l~THo8OQL{vL* z7Yt}zL5aDPLou;Xi**{&K;3B)#;YuuM#3CZ8>8jl-I)I2iL%ORwRd_|VF;1;h>^>dRS_d#>;b}H~7Bv1M zTmLr@nG?K^Y~moZ>N-(uFJsgm*KVM-Cnc zG$#97DgnZe%r15v2DnPc(Y(Ew9`F!nN6r2jjlN=1y5gk|fx23=s4`1hJ`GfMJc8IP zx2Jvax3{aE;YSN0{=YUL(`DNTuofKCXaLxjI#kR5NI87C26w>JkPcZ=@4Ru=+P@;4@%1rr^YQ|^p22z3*ALv$Ra0cnLmIKm0~XltL;guJbKAumLR`uZ1FRY-HKBKriC;%bMcWV!w)v zum)zL#F3_m1uqrYVPjv*;I7FoyMLr=-@|fi1nrhA{&a2b~UCR8e)9)aJeYawm!)6Ri zHoa~JyiLtZ_0bHYP!z7Ru+er^2ig!Cnu3^&l}?__^q>DAw|{37j3sPy?tG>_P8){= ztefI5d==aA8PtCWQVGurpP}W8yGewrciV6JyMjovpHWghf#9^F;HA@bxU=RlD$pCC znQkD?US~sHP?DKIB!E9vUf=GKyfqU`WOHKW`!)XzZb4w**9~>^*+wEMX5c4XZ8e3P zWkuXed+seFRjuqS*N(kV%qZ-FXatz^S-64+1!C^2ZYS}Ytx^3F|@G)T=oev40~9>OgPg` z_ylt6x-j-ad%Z6&qXs(&0Z#O4vq{w?9RBIq`mH?}7Bnp6-yr$jYzu|$G#!WpvZ(#b875OZ!3 zDf2InBk`y_+u16B81SU}W+31R3`UGAzlrXiYcGnzkH2@)*q;f{Ph7Xme5xaLOtfAW z^PTV^z#dUPX3#C}L>+Tj$HQE+SVZRQSFcWO#sGgT(8jT0OL!`B%;Alyw423)H0iiV zVw9p5EppDd8|y^fSWjF3+zh7g(T&;L=DCgMODSfLeE3GTOhW)wnZnc0fEwLT5wp!0 zGlk8cD-s)|EJ?H|VgZL1NFv*m2rPpI@9XDch4}=kE;%V0hs_xu_3`r4Qaql^(4C49 zEMI>zG1lAi+)q+$815^KBOS2}A)e{yZZ_*A8OyL<5`gh_E$$+d?h71xF~bvNT$rQc z$@kPWkx0P40QCX({89?;z;2P4Q1NSMoE=cfY?bjyZr3vWA4fAH9@dq|u1Okncm}~o z+N&5;90%5_gzaz}F?*6`>Iuhotdq4Qufl)u)3=v9-7>&KHWM{;e0s;qqiD5~N34@P zO(>$u*Yl_B_w4*1{lSVXO7>;8p=KLXd_)#ef^@a08F~3Dv5IF#-px&Mkr{K|Ej1cL zo?@&oe~T<4WhL`E>$_DnQJdtP+0! zo?aIp6`}Y?2hv7~+vJ7WJv>^A#mNxqW>GWb|-RprE;bak|B97vh`PLedzUK(+ z_rbR)Bl#vpKoQrFNV2`uWS}aB{f`lIn@SZLGdt(6NoA3hW)Oug#ADK^nMVoV%vL|+I=y0 zJx&MA>Sp75PPi6Dk%B7UV3AG&u`kL6X#-TfkJ<%|oC_t2B?#VRit9Na@l(r^QSV*B z&!$u2GWyhH(IO>kfFAH@h%7%bjb?(eG8sY%RY5Uv+)_i(KwDHs|E1&>i}!yQIXJqm z>)?rXj=4m}<=tdz4~0FfZv^G~`_-LYS|d1pYW7e|yYCI3>q>F;L47-lp^O+mRUJJQ zj!W`jw{Bce4d2_6Uf=sWE9ilS1A?$g3%V;RNbbQDumc_3FI!%v(1JC2^K=^^6qN2F zRJSZ<<{GYk4r);76#kyFV@!X6qVLwxUgfllw)J_zsc8OBVMeF{#}xacV^Gt(X>%=N zo7SVIZuWxvg(4jW`{dDQ;ef}^%c4FFc6YTU>Bz6FKGB0wA)~uclIDxrTGoQ4-e?WwXnlS zzg0caWH_4sj67lVws)# zmwjFHzxNa&P-b9T0v~@ky6uLU75xI)BtZL&MNuOoF0ihXBAqN<(DxfsCMEObz5}RP zp>&+tyvmbFgE|tMEl9zI?a&y&GMhGnyi9`Ce@6~es{)qZ=uiDoiNHueWi0fYDDD&= zEm3C!BC2S&cNC9uDSvt4WIiT5EAZG|n~6c3>7FUy_~cqC`Eh^tF~cAftfFfHQYCz6 zpJ@26WTLo-5E8^%f+Pm$C4IQJ35s9}_;?G=JM7hV{8J-p^)26uFQ7_-nDOWVr_8HP zVn)r~m+m=wKMB0#8H~q`MnR|a4E!x3KV)q3^tJ4437T`W$Xk6N#^h^d6n#B?)P=34 z_vXL+ofr1|`Xhg7$zYlA!KA=&Cb@C5bdnEGZNgGxOY^iKQOP{=^$^%puf&PZj65iE z*7D*U`Vl~!>t~#3#fgTLx-`D=8y|Kzpvdo0w)4V+7dj%RE#ktP>xU{m`EDnq z#xReFXGQH>UT>o3+IX4S3AoPI(y`r8vX(=05Y?f|k#T>ki`l|eV0Ov3;L+KO3JAP< z?PprgCM-`ILw!0D!Kne!B*JFsC=z2_Rh=z-vMm8E~mc^(MEr*CX*VHlIKs7#ZAK~@FXX5`1b{c99mQh{3l9qOEvoH1M9C9ex>NrT8NLufB{ML7>_He z+@BLPcD4vKpDb1%!c$#vURC`9d}e_l0XV8o%J6>+85jXfPUIahD@y!nMCutUB4RuF z1^10dTxeE%zLhF}*@bBsR2=>`Jw_=^a&P)@BQayulZtP8ip0Tr9T5g`H3lJ$Ik`rfhfdg05Z*Aj3ZuOaJ)G1be2*!%Be8q{M%1RLhL{0o@^E@+ z$CB3K@SeuEy1$Bb)kCjADi9c|JS(H3PKNP>AsD@0rqErO+w*IwjivcF+B2|ElNf-ANJfi7^f$7O zmHQmN8pUJ1%m#{J)E@(HL)BPU5FCF(P8R5n1q8kRLUJq!26UCU1~&_?ne$;|e^n1F z0^M8fvkQ?WBzkxVQw>W3o3mfqt^u^9eA|FTejg^sJYRCo4oRn{q}FWi>M_b^%%6Rs z5=iHxlrWZwze7L3X=679Ff4j{=FYyLJLA(IBd6I07u-P_1%_uOt3qXNkD`BVFQV@R z2NbSTof+u++p#5`uht4ZfML$wH}hwdj_Qe<9;xi0Mtz=KX1V2hcWxwZBjOGt4*Wak z>Z_l+IyJ*HqE<+2`B~6`h29NY{n)6#ae?q{eTircc$od9XNo}PI!g?EnNp>do6s{| z6odwp#!Ya|++K@9cdEs1O?!Xc<8@wkt-gjp0Ig;c{Vi_|f9PS7Wz&X`98aB*B~>Fr zRGl?qxYv`2$`v2XC3-?7UQ^6cUw}Tx*tXCJx4pzIOP2%nK(fgF(~Zii12d**OtYwy ztG;ECZRfyfjTS^aHOvI>Ubyx7a>eD~sSCp07%t_a8L+QEiLMNEVa0!%L^m1r;!3Nw znnW#j9={I=QuR8t-kfhlNH60Me`%{%EyLT$ZNM9DrxrhM0KhkQZBkj2!!YXE9yifa zJ>1%ivuI@NPqNX$?|^d?M3qK%?4|$Zan%1nV7Yzui2w!uB~Wu)WfnMNLa&~cWPK=B zNCg8Qnu)`V`!xOLRRe!O4-QNa6PaxEy=)UHRu*efpx+=ps|nc3oIQKFenihlZmIHX zgWGJoGkel73`6l)z6|e8_2>e z6+ssdHFli=pXmf|i`9ncM%BzquGG;^Kt+;r^06T%$v2Y&9V>tMHkE{2Gu|y@R2ERA z`GV5aL$3O~EFI~oD!&+|{ZCRf?dx9O(;)(~1Sp2IQv2xISA<>PkdeM~4vApWZ=}yu z%eJQ`n$)zzrDd@iv(+mPgqiEG95TLd|MtnPGGd=IIQiW0sK4tXz=wI*wok#<*Yj!x z^_lIX|NJk9wg!LweBIHJS9W>no_W0b($s!5{a6r;?-Q(c1myx-_TjfrKtKDF7kC+L zr!yvd=Zmwa+{w9FL!SKlHGUpk78)jxH87B55vx8Q6Yiri+|=iSj6@7$ zo2|5H@?JU>L-DER=@gv-FHtdgn@Bo2lu4D?rN$9a*g${dCdza&<*gu=a1f;RxD83N z%7AfY(sqgb2f$S~hl%dnWHKODngoO~U(OYUYv3?=XQUu~kGTd!9_z)FZYI5R+@rN4?Vt z`0EMy@0)?>;g?sHv<^PmE+hk-?vMdsAN+{MDIYcWoVw3DkU|VDC%Fd`x!B`ocj2_3 zwvk26QOj(WI9;8+t9fjv!{O_0CdO4khRPpSm1Y14}L1=%eG21AX?d#pFCh@@i=`Z z!zu=qSv?L*haTSjdd5jZJZVe`JCT1vrcBxClm>;cJ^g|)e*~eSO^~gkE#NOP5Gq}N z0(JxuXhU>)noTwF4X7!^u&`KT36)@_jkwXo`4}54_T9estsvL4LngJaP??CX=1|3k z7+Y!IfD!B4h(0%;-#%Ehg;X&GtR4#-yhaMx{yS7?ho~5cQOYxH;MxM6mFRz3s`Irq zFLP-`S6lFINsVL|Fr@Q!JHHoa(`}{>R5PzT1`Zel(Ee>p#I6?HIMuKH6{nlJpz$sQ z&%(R)27JuUV@=qRrVK{MlygD@i<4^GbS>g9`H?hEqfN#e2_f)?gSvnSv{X;dw4FTD zy{Gr(Q&>G)g~}Ghhyb%Bs)BzAI&)HjP)+zN;Qt;S@sSt2U!qW5X~_f9&-mB1cMWXn zU{h=hSR8knm&G8vy#}l)^!;S&$Ex#x5Xkyxy@xN>Mea8_X|(dfLS+0|Td3qgoBcjd+-=lGX__VZC=3GT*1FKl>_9;96`c$D)7X<;iLt_Y)xq zZL4S_Fl37u%L9hFu4$k0i&Ji03e0PGZAEWHl_0-nbmSM*wbOB&8B(m9InP)#S6({` zhoXoc7I$zUr-@8e=8aKV3XTA@sL?MRC?5hNxI^`;3mSYX>A%|B}wh%A${-4|S)xC;aw zpNl=2;>f1Z6;5A2h>anGsKJ-ai4~g~p8}zrxyk!k02^`W5M>t#*7kfk~RVWV;+AGyU#W9S@$3S7+gjW z@*-kJ#Ky25_~fCgTZ@Z?vtx(!*p z?RQLJ!bStnEHxTDGZT-Oz*ux^Q$8kbpzVg81oB5 zio+?Fn)VZ|UsGmm#p)8+c3J|UpCpt~83)#77j+Msd0biV1x<*3^-kq64V2}xiDYV_ zcxxSrpD=#_dYq& z55ass)tEUhPD`}V7$8Jm_~;$YPQZ@y^*Z5#?j4v~*I}X&I08m?p)T~xm<>mjKa`(j zCkI+y=+^Am(tvdELS|y;%ePDA7F(dx$90&L8C@(k$A! z4Y4_Zdf1At;`U@omidi-zYKvS88urTo_%s(zXh8ADZ)@@le_Qg9)bv77jq51h5ALR zH;u=>6Imf!r0E&D%9(V^5-UKWiI>}>3Il(rWCF3P+}A}qCJn<4!)6OE0wOE9NI?7~ zhw1{gTYAYJ34%XY85Wj{YV)*FP|^anj->*0;$Ub%Wk(mpVs5EA_T@W`GTtNCERWqB zm(TZyMfKGojmkZXMz<-yF6NW;>#C4c@C61ZYn{sSWM>!6u^jNPZ4r0A<6ZG9Sz>?M zO{oE(TRnk7>47c1?`{-g=!9}ljer0oW>S7rj6m42Y^3%wGyNdQBuR`^?Xny2xSANr znwLVySAJcI3WCd)oU>Li0VAz9$8je~g!5~t?Y~T(20`Bsm!H=X4i3O&^>64o{qd)& zrdEzF_^@=+()kr-+Lw4BAwW%xB3yrXMGha(1gGaHT}jrn8kg4vwdg4X<%L5u1h5^8 z0_Ur<2A^oeW`+4Wj3$;kf-9*`Rv}?A{6bZ@Sdc1+k8*EWNtqV353rFJeSioTd)Ja$ zWyw|$y4Fl_oxzhS6c)t7Z#X0C<*0!>IOka#LhOz}1RgGw(+I!7s~(O!$0>gk*Q|P@ zrS>LFE3pL1W6zuE^<}z z;$vTl^j4HU38@Jc9Kk0&>FfVhBK0IZpu(Jq98R@}ugNq`*7#`!hjcU)85WOVgRhB^sxp2yk9G%vg6TxzhM2Rx;eHE;%X$`PI~R_+a~T zlZgBWD?s0m<%p-#GyMDVazkW2l}fdhZQv8NE*7^f-~GPlimO$A!eFauORG0`{itkZQQb`uz>A%0pLvBfMx_Z+DEICD z%Lj>s@XPw(VhCZt@7A&s*;04Bk+B*dnt~Y);9{|jl@7{+SfFI{*urd&mP0DA9id}h zbBkrcK^-8-%Vb}otrBbn`^Kt_odhH?87$qB(#p0|r17>K6-9sNDe%IpbqTtQ&Jxjs zSFfFj1kSfjF2%a&T#$Pz(CImybjY zJ{;s6Gecp)%kzKSVI%48cvjN2#*|^HK8Pcr<$9r5i?Akr5z9j>IOJC^u*5}(YMm~< zQP0MmO56P4737cp!>A-A{E0eY8n`|6d0^T=ecA?Kki%{t1;O-{3A1gx2-s}Xx}HYN zi??2y*VK1RXwN$r7A9tCg&u(tIYGV9_gN2zNp6_8c!Pg(7WXf4Xt{#&Q(@}rlX^)6 zuvN8`8n>E9a75a~Y_UGo`5Q-ACa(Jq_8Hvn$k8ptWCw<-xZY@#m3*VGbXS=9x%=k^ zv=Pl4G! zddFHr+dh8?-YYkjy`t4ld5G=U!av@~c#ciccRl-~nrnp=qP96?;6P#JZ zJRv)AJ3iA%;KOK!`Etb57MIeoi(o_c0TA+sFw4p(V^*RTNkZPl-74(qGNO5z>xpLE zU1RlPamlZ8D#bVJdf9DUk>^{V%uTIkt23)^dQ#Th-hi0r{I^2QeNMoV$Kz>iVep2plVbq|oPE)Xxa`+9VV8gMg!Q zs&jvmK!V&tjJDEGRI$6;Z;ii^o9fG^NtwiJWes$@zSngX(b#{YVj!rTO>`j~udYR6 zJh|;5bI{8%IMzu>>JJm9x~fM>)P~3BL%jcL1Kp`FM#rwd&T=Tia%I8JdVwOw`9pcV zojQD{5j&7%JH_!2bAL_1nxWkIqXlEj;Ff=XHtpAw(0YDQJ5id7*SAKVt4!~5Js3m} zIP5V}%r-G&^HSB8t*0r8Pvd!>(43`B3~9ZuHT>?z#uB(Z->g&b>XzL!9B`gcU225+ zY9KO%xA0T9-q&}?nUl55{bY)-N!df*Lb7n!;)S<#0|R#bs&NBS%1dJf_50+)eXD;i zU;x18oPZO@(XZ1STm3xza$B04C`K%H8E_xvtKVGT*(VKD8(!~cJCS|}^-e>CQW8d^ zS=H@(-=uTpkWVVjm$`K$m8=F6v{{uDHu9Y8uPMrj%cFgr@Qs(52i>&ec%|rhXS0}} z&MT#vL8`X`#oso?aK=0#jB*w!$=iRkaEWmuj-x zeT^#|?$04O-YXb7KgkK&A6-!nzgdwW$H*%(0(&Hir z46DV&bruS7`N{IP`~Z2mj!jBkR%kjQAOv{sAW892#|OzMCr>EQ`$4C0Dj|OWghwzm z4&j;M0TsHhSHvQfB+R#<$5gBztC6e>7c6nw3A#@RnJi9MG-=!*+_$yxAEA2u<1iej~lk>3(s+y7wZ zxms?_xTcJIWQ})%<`<~3b0O*NW+_Z5e;`%rl5y@-a!Z z4vk6@3=@*8%~`tDgYZZrj7Q{VSl zv7Lj;7rm+<7B2C#b2w|4FnZkdG)C3aS~<~{kO#~#!=L{#0|X1ym@1g2)UB**2nw$JfrZY zM6M=VH?733l<_wKwVGnJu2^j`ImoR_)h`j5gs6;{;*0L zP&-Z8DdzIKvXu9n)K}zMJK%0r*_<0{{W6LQaPH$klbfghphO?<;iy-n%nw7HQ2|8( za)XX4`GJUGkMG3kpAicN8_{r8ff)6WmLL(!u-jsv-oChW-NAV3#FGrV%zh>j5y!yw zTl=2RZlsKL*;#+x-Uv;!lLng{SO7GS&GAf(o2AhZIXtkL5~OzUUW;u~c64uB%7c`i*s}YQ#j$?kV~b4XUy}L~P45h8m^@R$wE3C?l@Q8k%db9~ zA;Nr~E24;|M%B&$fmMfafD;Ng<)HSfg%xjmuCZ*!TakZw9mla?k8}i3N2v1C`H$rz zkwkuCfRfZCU(7hC1=A;L3%7j_nCkA+8AOUj(Rr6ovm?dN3T5|@4FdT2OnM*}df6t! zUp+%JgxvPp=?q{CA-$XGh&nJ{e?Gc50>}3($3#;>X=X|36@>4muJ8~Ba<6`}W2<3D zLvIhr5g>mCNqynq#98SEtIT$htHRo4thPE}qP`TWe>3&l=?YSR_NoZLeo^+t+`~em zvLWgT05f&U3-xMIL*tw5tWLQQ;)mS)JyO|x%$gNh$~8_e9p)V{v(?APeYrPJRE9t` z5X~hyiPi%oo04VKz=lpZC=ZBr+5*;YmaL95ic)`tAE}UhLuJe+p+m%4dp9zKV%hJ( zfr0yPse)j-azKQ|aIHq)tK>R*JD@ST^m*DN?~qLt-Xd z_I}}IxApN{?fe1En%0S#_))M#K;hplH|i8>VUf0s!_uH;kSZ{QoX|;bC+^EoKT=>` zX;z$^<@7dZFGzej-9+s)q;2kDZK^PQ3I2Zr1vaeN^3qKFiiEky-!7)7#y=vX;!1%N z@NsIqmywR)WQ+F=;Z(t5*7-OZDTjyd8Nz?| zfLdq@|8|n0;?J}F6&u3F_|J>lPYtN>joG;KW zisQP1z@cDTB?u!>QHkBJ-0KP7kllZdhj>+PmvqWER$?8Mq=tcMP{hW?nuZUuVip^# ziOP*0xdVFY!H{W01W6@2ke+^Sk|Y`=9ectDKPRS)_s!{DWD1pDZ6$6n-W9_^u?0k3 z2NbC^B#5&wwh_is_CRV+=?8#U%?4&WFbEL~&H8uUGLCd4y;lsLgTO6rcEi92_hC*$j{ut&Z9n=f9tCimuakf=Og^uBgd z`2Sc(N`v-C>ut)6Albt@*i(e#s2yp~03=H6-jt!@qN40p0>PmV*Lo1}iro<>Ke_q# zHjhiuRu$m%fc~!^TbY$cz+8XXeP*^UrY&qrSvtHYd_Z&{IhRB*!_+K@Oi2>WI=y=-e9=`%e?);^U0TRzj38D*%WR`E`*dSZh`1u)xihdX6x!GiQAS$Rz~E38c2S zRIKD3KzkH>*0!!Jyy1Taf-glOJnd}-J_mF_bpX~wuH*W0qIoYLF?mLTLk}9)U=tHD zmc@$sW2UcMJxBs}A-Jo%XbAd`W1yaN-12^hvKVcX9rVNt&S87GB=Oy2i_2fmVF$y7 zl_ZKXgXJ|?v62Qkx&Ta{NJP61`6DazBrHh7KM_3c$wpk2DB^z&&U)6#t-$6#l~Mhk zQT`^^wqlEAt!YmN7Z+ZypdAK#8$$enr#I~BG{G#wNygJ!_l~k@xP`r%NfY!y<9GTO zrsQoA^}_@D;~83mGc8tN2U(j+nk`#2mKjP|_eW#Rs9;HLHjvb-8nghRSSN%B5lhV; zL}J{E{qseOl1hJQoIY#9WGOmhG423~t7k>WOzM2$SkSY46q4G}$=t_2I+mq{)19p? zOO4}67jZqj_N6UmGKk(v@E%ah-r2=RgpK}IRywsR=9|5I&?vb0XP6^!iOF;BWF*J% z`BPAWPe&yHD);=o=qG2u>c4pc*ja8d>4%xZO&;t#4{?7?urVPj8=dPbZrkkTXJTf` z!ra4PN4G59>wQM?S0xBu^Oy%zP+_vnnN30O_bS;QZ^T|99sH}k2Rwo`TpjmqjXm9>YUMuk<{qg>YrvP@8s4^DhdjJ|hnmJ(37&*b>aUOYNXOwl_J zjm%(h`-6W|7*Ps73i2T$xVyJI+%bLC1(v&JGwc4N`pMGF+8?;U)V4}znaD&r(->^% zA})t`F|j4=Cr@l~;+m(&WSl?ONu7A8;7-=F@~m)jFpD2@>Ak1!s2YG3#Uc*0Szpe} zdmN$mCX}8>QHDTG+2!G&P=WK?0O*rHO`v;TuU%5+2;xGR0U6KAY=__wE*1A%2ZXcv zRO~4Qn+Bh~k*AiCpg0@h@a8BLLpQV|8NDpPe?{z_LM^MZy0###1cV*}wO)S#2+l`1 z>YO?>#KC`NV26>;0k@IP0!*#~I60RwnFA<)?OWY*8@CaE=U;J;?R0!Nd;^_KXKc$# z+{lR|C5}^%A1LZ9v8H$od9lCtmW3&# z4TUYVHH9mzbcKwC3%-a&z=W)dSXF>2qR|CrV{jwsF|J}EY2K?s0YCyoNql1wrd1_> z3x#P-!?ut{>p}syGqTVC55R%lWpLFrHsFjm)WUyrb7slb+sRAfftXyG$C)%noz!Re>(A5=?TOug{PJtv4 zX+eT=n8(&SsF5SmLRDCh;Jq&#)&=l??659i5?kn4g~9p+qwaoNHc5ao4YEEZua>&==s?2~+Vh(93^9A7U{sKY_Z4-(n3rKBB~fVDR6v5K4($!B zQerbh2ti$pWDkrK(nik_GVmhF+@zsh*x)?)l?t#h3N@0nsnQWJWM|4Lnq>>bev)iX zYr@PQ2i~N{3MeenZ&;`d)WZsYW1s{=WIO>^Kx8Z9N8fx?h?7Em+x*-V;-Gk=;X}uT zIGna`i@*I1(K0XThH575eRN1j_XykUQX&2{DTKx)MwN2ljc+L zrYgvqpwXz_j<9@*bp;q=u*dUx0T34N=>sfp3jPgOrPgr^Bd4#oqn+(}!ALx5f0k$U zl5tmW#dqSlc#0p8#ap6f(iU%|31wwrHQtsQ@-j}BhO8_2QuzS5ll5r}oZ@IYe=iIzOvFJ*wyD)Y(p-wOaXVhs?P?DSeJF#X7-C8XA2#@HJYs}%rPJ>I6pojZi|L2DoK9BddNM8R zv$eRKC5e`ci^W$ErqM8fNa43cr~&C@82Ud->eiuN|Mm}Z)6zS{kjj>be<}0zZ1(mp zNX?MW`BK7b2Y?zEOA63KK#c-cKLk`qRaVGRLqI*SB2x?jHC3srL0DA?r90X;+% zDkyRtQjn85WKkGaT_5qS3uP20P3|MTs161xOC=bf6)%^~q*^8Tity_JmoDV+Faa$@eU+c7N~ z&$_|WS#yeNX4Rsoe<5E!0!iy-y%MwB!MVqli?erocrtkos(bBS87n$cAX$+sqoDCl z{_&9sc58-bYIAFv zDro5*n9U~BdOiU^OYFhko<~rmw$sV17K`aroKyjsO(~pcV}&ED+~1KA5=y=Xp>M=a z;!62HiV2vl^@F{47c_f=B+s#fV6`-*a&izXl`JUY2f^Y5S<-#L9bR^)2lc$B?|oYj z+V!%&x6%cJf7`1Db~oJ2-OJ6kB5}ht=x&i7(j-bEJqVVLDq6cChl;yO6S@QQY9GmY zcjF4hJS;%@q`^?_d3p$Ze$&yFbc@vmD1&6BbaM8o z%@eLf#AH1GFkXqTaKsM=ETp5ojC^xhs-XU>q{HOAe;sVeyP!E1;n<_hFj!T&(kML$ zR-#~!UfpH&J&V3OjZ3<^jW_(^fi4i`br%S6kPAeA%>_bsfkbmi+=q0W4k-eJuwRLu zZOj;7R{0uP8-!ASZnCk zDGq`)V6&3LXrTP6{C%%N2G<*JEDm-5|9tWDcJ#)3gl>FmIJl?Bj-}kIO?XnvxKZP6 z`w1qG7}y~U7f0~IOIy0aF}4GVhQexs8$V^hX?`vCg#Hs_J@nyRt7O<0}4af2G;&SYH-2N`b)^L9!&6s=nkH zpOOSy5oH>#jZpT+Ammads)#$(p`g&=z)4SQH8Dw%8R$riHqb(=NNiSDAlT|ii#6OB zqOmlBGlD6Ca>TGn%1_EsNnJ@%VE=k=e@H7UfZO~TzH9(eZf+m#_S=s1fVaOLi#fHo zf+}w_B$*H7LibLFZ=Vc)1x@CWKwU&kPjE0>AW&>%iM$(q8)#)Rq|_!{sZDa%70>qA z*@%%|Z7{SS|LS3<-?386rDQ+qDKfRNo4gGH_gTo>vdN^@_9Ve`J@RfN?5>Zxe?f)s zXvvnm5f6zD)~-9Y*o%=4T?cAOF4i- zFG-mJ+|o-iIN>1zX-Wveu{TgyGl*E^B_!2APmZ!mqemH!^SvfKi6=uRB38F$IJ$p> zuNjg^Ac*95(U>d|=q#BQX5EhIr*~4VeeYUWjj!u{5`>>LO_uoTx7m9 zYnCB3>&F^Smx_3Fj%IOkeNWM;N)E;swL( zh>8)X=+*Djq0R5WWMQXqSCzgKQc@km^x+eXfpd@o<3^4B*~C!la|?rTWZo1iKAT>) zj;c{lg+DYK+|2F&;6w;f`bp`iR%R9pu6c@_ce~@3KeiG6?ZNigU zA0vC#wlx^n-H2ynO%Wb6U_8)^k~HTq2cA){+6!-t^pn5QktFg-hF&{1$rstUW5p#5n>z(* zd=W@s=cmd*=6-OL8Iy^i z_1bNl<&lamdVh|8Ctc)q8(5k*e1bKWTh`2WE=h@c_>k@`6+slmIFg~@>srfW`5wop8mB$WEOt*W=Z z%SgH;JE&Ka<#f?D%XDe>!}t<2Pkwsw{p*9j9KAShe=f&!Sv`8v%+3HX8=tQVlOUh) z(}YJRxZ)8?l0rIuH(S#SppRD*da8h6I}%UFi|^{``9=Oo8`9h(8G-0|JDyD^kLTyJ zx~N9tq^&Ps(fivY@j5R;Dgus+@se)m{v;lYC*r9%5YNP+_)dH;j>HS`gLo;9#fkV? zyb`~Of3Y|f6MCQJaxBh7E#6H(pqYtx&3Y-$#f7-IT3pm~F%|E{Ok9e&Xhbuw#X>CT zS;(w@*QVZ*_7{tKIc?6wQmn*kHeOwbReg7Hkm!$mD zi&sA!zT8MzeNM`rq)hLgTYlYL`<|48*^x5+e;gv^Dt<=F2axY`mhMZI>5^2Z;-A9=`a8?AlVba^$&&meuEf8@zv^YPE7Eu$kv{$VPtRVENMDTGi^Mjp_qaY^ z&&JE!MXMDg%Y~cavaN5v*95CSC)h`1wMn$D>Pc41zGRPJsY#0G%X&<2h|TKNisaKy zf5)@4DWsY$ZZX{$+**9vk%^M`k;%8mPhLO%bw?)t^CFR1B9cm@uObnz?vw}_PJ5j} zeIEvoPJTK3^<^(weU>?+D+*ZjSb+XTNvfVJRekQNEpOEwKIGz^Y;}R^pq{m34p-mu zgPtyA{kREQzaVV=DqiQ%b-El+-q$UAe@*Tuzd2pcW_8O>IHL#U8&~A*Mg9AFJQMY& z$!vVN>z%j8zJJ~;`YmSb6&&~Xb-m)pOTVuhm@Y$bV5~m?>1z7vT8R9jbEns|hwAzJ z@1LDKe%9&v3pQ+mCN(QWOh^AJq)b0z?&hU1EHC-HwM`-fx4X>UD;fNF{QA|Cf8&n1 z@CD|q%gmX`Hf;B2&V9y|pK?CvOgEYBHq&m+andZVl3C;&!S;8<_>$qxrx4o4`jVls zVV#}MZppXFn6fF|Fr<5Bi62g$9v{8#WF5aCYttE$v2-?2VQ*Q-yYs_A4h_FRx?R2{ zNnUKk)<9Y*gWjw^^t}?p;on~#fBn>tn~B)FOfKZGse{|g-1{oi1kAbZ_rhDE;{D^QR=0X5M@;Wm;4OnnXeWe~nx5%e7e^ zcDKa0J0-qS>i&syyoo6kL(a41C%eC;80Y-5D~Ma-?z8!%Ih)SUK^#gf$aPLHN4-(L zOO&%h>n?kCL8#ENpXqK(kRU!K$KnR=Q#%a_Cr11kgY1ich@xzX3PVQE`)zxf8cVjsv(bDl{=TFf%hX zD=;x9Ffb=63NKA>WJFf4!JZ zZ`(K!hVS|nJT|-N@|zTaqCgWT-8OMPoVE$D=s|JVL|9q!S~9lTU%#&;JEbMtv3uwt zp=LOod4?R3v^dj*5zZ|BV*D{8j3HbQDIjv9oFiNjW6>c=tQ6rAX512kmODl)45X2f zgsi0|8YY&bg&aFgq#}Y@MYJG7f2f#Pv`TOYL}+Ir5MiOLA7i881PUiq#I?o)MsjV4 z4?LH5~l7lc!+uQ z=1oLDJxx8K$N9LpPxFU}4$^Pm{LJTLUy{EVouv7^B!W)nS(Q$+CuBu4e_xjVpLvqe zhnY_*Sl0sbqRDhRE55g)X4GeX;`8cj^!Dw`u{a(0tV%#nrwNpq|5mk6v-IJyB1+O} z1{M|5|NN|=Meb?!Z-JSoW%`{MPTRPS)4N$R{@Yiz2jb*XY~66CX^J&uG{=8U%N^;Y zp?C8v^YCCC9T(H5#$RXje@T&75xt(@RgG9Hc|;GB($_tN9$pPkZwJ4h4M)W!$vJy{ zSY-E6Gi>0?@hqKI#SC0A%5wyLIxDNA$7F_*i0E0e5V(NToAkbVEbAOv+a)-0i@~3Y z&v^=+M=7&LwHNBu@XPthc~ILvt+i0 z*6cwmcc2yf(_V~jzZ{O1wDD$G*_&FaHdX=O0aor0dwzX1I=fxMx-GDx3#&H6svTg} z{;($>FVC*RI`~i}Svp<=t9q~|gyrT3EE7Ax>iuDlK3=^4a0J+4kr&%)R>m*Dx@TA% zTQ9SkpJBZ^9@Y~(e@v>!c2c*#WxZMudZiXvPz$Vs%$e*n+x?l}j}C9&-K;R%Ey%r! zz(KKznWVUKxCVBOt@{>O}WE;xtfW#tH zz{1H;A)rv9e?mgBNRghUleFptd^YP*wH*rMP2dm#+WYH5hJO(Bm)CN0>|bs12jnr&n-2P>w}2<xW@dM~Fu`q^& zf)EoDLqS1dFo8x28$#&ls3g(zdXc2HsruzI0Alf6pk^M2W+w9e|*MeJoK-2y_ATe37B0X= zhW&ld!6mo~^(wW4e#d2KfGZ3qNNKnMjVkFPpKFE|M*OF1f@^RcZb2)w!EI=V4(NlM zDs5}Nf2tF@pc{Iim(BijzJh)({r??+JMbFra^PF;k-P_kFa*PJA0EI%cmyLbs?xLL z_l~J7*L!-bnmTyFC#p?Lo}Q|9Cp|q=UH|RrxkiOso?dA5VbasMMt|OVdZ}^p;EoB6 zQ@frfHD28ejr}wc_`DbRG833?49siN!c3Lq|GoxlkudH0-m9JsduYa-+8uLtSmvP4f9t$=#GB64y IB}Gq03Rk;%(f|Me diff --git a/doc/Projects/2022/Project3/pdf/Project3.tex b/doc/Projects/2022/Project3/pdf/Project3.tex index f29685e09..aeed32dd9 100644 --- a/doc/Projects/2022/Project3/pdf/Project3.tex +++ b/doc/Projects/2022/Project3/pdf/Project3.tex @@ -123,7 +123,7 @@ Project 3 on Machine Learning, deadline December 15 (midnight), 2021 % --- begin date --- \begin{center} -Nov 14, 2022 +Nov 15, 2022 \end{center} % --- end date --- @@ -144,9 +144,11 @@ For project 3, you can propose own data sets that relate to your research intere \end{enumerate} \noindent +As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL":"https://link.springer.com/article/10.1007/s10614-020-10078-2". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester. + The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is: \begin{enumerate} -\item Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among \textbf{linear regression (including Ridge and Lasso)}, \textbf{Logistic Regression}, \textbf{Neural Networks}, \textbf{Convolution Neural Networks}, \textbf{Recurrent Neural Networks}, \textbf{Adversarial Neural Networks}, \textbf{Support Vector Machines} and \textbf{Decision Trees, Random Forests, Bagging and Boosting}. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into \textbf{convolutional neural networks} or \textbf{recurrent neural networks}, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as \textbf{Principal Component Analysis} and \textbf{k-means Clustering}. +\item Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among \textbf{linear regression (including Ridge and Lasso)}, \textbf{Logistic Regression}, \textbf{Neural Networks}, \textbf{Convolution Neural Networks}, \textbf{Recurrent Neural Networks}, \textbf{Adversarial Neural Networks}, \textbf{Support Vector Machines} and \textbf{Decision Trees, Random Forests, Bagging and Boosting}. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into \textbf{convolutional neural networks} or \textbf{recurrent neural networks}, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like \textbf{Principal Component Analysis} and \textbf{k-means Clustering}. \end{enumerate} \noindent @@ -171,7 +173,7 @@ We propose also an alternative to the above. This is a project on using machine This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides \href{{https://compphysics.github.io/MachineLearning/doc/pub/week42/html/week42.html}}{from week 42} and/or the textbook by \href{{https://www.springer.com/gp/book/9789401798150}}{Yadav et al}. -\textbf{Note}: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise. +\textbf{Note}: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise. \subsection*{The basic structure of your project} @@ -278,7 +280,7 @@ eigenvalues. Compare with the solution from numerical diagonalization with stand \paragraph{Part e).} Finally, present a critical assessment of the methods you have studied and discuss the potential for the solving differential equations and eigenvalue problems with machine learning methods. -\subsection*{Additonal (optional) exercise, adding 30 more points to final score} +\subsection*{Additonal (optional) exercise, adding 20 more points to final score} This exercise can be done independently of the other tasks. Here you can also choose the data set you want to use. Furthermore, you can use diff --git a/doc/src/Projects/2022/Project3/Project3.do.txt b/doc/src/Projects/2022/Project3/Project3.do.txt index 03ef970eb..8572bfc4e 100644 --- a/doc/src/Projects/2022/Project3/Project3.do.txt +++ b/doc/src/Projects/2022/Project3/Project3.do.txt @@ -12,8 +12,12 @@ o "Kaggle":"https://www.kaggle.com/datasets" o The "University of California at Irvine (UCI) with its machine learning repository":"https://archive.ics.uci.edu/ml/index.php". o Or other sources. + +As an example on applications of the methods we have discussed in this course, see the article from economy that deals with machine learning on bankruptcy data from Spain at URL":"https://link.springer.com/article/10.1007/s10614-020-10078-2". The data are most likely not accessible within short time, but it is a nice demonstration on how to use the various machine learning methods we have discussed during the semester. + + The approach to the analysis of these new data sets should follow to a large extent what you did in projects 1 and 2. That is: -o Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among _linear regression (including Ridge and Lasso)_, _Logistic Regression_, _Neural Networks_, _Convolution Neural Networks_, _Recurrent Neural Networks_, _Adversarial Neural Networks_, _Support Vector Machines_ and _Decision Trees, Random Forests, Bagging and Boosting_. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into _convolutional neural networks_ or _recurrent neural networks_, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attendtion to supervised learning. The methods we have explored as _Principal Component Analysis_ and _k-means Clustering_. +o Whether you end up with a regression or a classification problem, you should employ at least two of the methods we have discussed among _linear regression (including Ridge and Lasso)_, _Logistic Regression_, _Neural Networks_, _Convolution Neural Networks_, _Recurrent Neural Networks_, _Adversarial Neural Networks_, _Support Vector Machines_ and _Decision Trees, Random Forests, Bagging and Boosting_. You could for example explore all of the approaches from decision trees, via bagging and voting classifiers, to random forests, boosting and finally XGboost. If you wish to venture into _convolutional neural networks_ or _recurrent neural networks_, or extensions of neural networkds, feel free to do so. You can also study unsupervised methods, although we in this course have mainly paid attention to supervised learning. We will only touch upon unsupervised methods like _Principal Component Analysis_ and _k-means Clustering_. For Boosting, feel also free to write your own codes. @@ -34,7 +38,7 @@ We propose also an alternative to the above. This is a project on using machine This is a field with a large interest recently, spanning from studies of turbulence in fluid mechanics and meteorology to the solution of quantum mechanical systems. As reading background you can use the slides "from week 42":"https://compphysics.github.io/MachineLearning/doc/pub/week42/html/week42.html" and/or the textbook by "Yadav et al":"https://www.springer.com/gp/book/9789401798150". -_Note_: Project 3 has an additional exercise which can give you an additional score of 30 (thirty) points. These are added to the total score from all projects. See below for the additional exercise. +_Note_: Project 3 has an additional exercise which can give you an additional score of 20 (twenty) points. These are added to the total score from all projects. See below for the additional exercise. ===== The basic structure of your project ===== @@ -171,7 +175,7 @@ eigenvalues. Compare with the solution from numerical diagonalization with stand Finally, present a critical assessment of the methods you have studied and discuss the potential for the solving differential equations and eigenvalue problems with machine learning methods. -===== Additonal (optional) exercise, adding 30 more points to final score ===== +===== Additonal (optional) exercise, adding 20 more points to final score ===== This exercise can be done independently of the other tasks. Here you can also choose the data set you want to use. Furthermore, you can use