From e0143528ccaed2dafd7a1c9efce59d020d71954c Mon Sep 17 00:00:00 2001 From: mhjensen Date: Wed, 6 Nov 2019 22:56:19 +0100 Subject: [PATCH] added xgboost example --- .../html/._DecisionTrees-bs000.html | 90 +- .../html/._DecisionTrees-bs001.html | 88 +- .../html/._DecisionTrees-bs002.html | 88 +- .../html/._DecisionTrees-bs003.html | 88 +- .../html/._DecisionTrees-bs004.html | 88 +- .../html/._DecisionTrees-bs005.html | 88 +- .../html/._DecisionTrees-bs006.html | 88 +- .../html/._DecisionTrees-bs007.html | 88 +- .../html/._DecisionTrees-bs008.html | 88 +- .../html/._DecisionTrees-bs009.html | 88 +- .../html/._DecisionTrees-bs010.html | 88 +- .../html/._DecisionTrees-bs011.html | 88 +- .../html/._DecisionTrees-bs012.html | 88 +- .../html/._DecisionTrees-bs013.html | 88 +- .../html/._DecisionTrees-bs014.html | 88 +- .../html/._DecisionTrees-bs015.html | 88 +- .../html/._DecisionTrees-bs016.html | 88 +- .../html/._DecisionTrees-bs017.html | 88 +- .../html/._DecisionTrees-bs018.html | 88 +- .../html/._DecisionTrees-bs019.html | 88 +- .../html/._DecisionTrees-bs020.html | 88 +- .../html/._DecisionTrees-bs021.html | 88 +- .../html/._DecisionTrees-bs022.html | 88 +- .../html/._DecisionTrees-bs023.html | 88 +- .../html/._DecisionTrees-bs024.html | 88 +- .../html/._DecisionTrees-bs025.html | 88 +- .../html/._DecisionTrees-bs026.html | 88 +- .../html/._DecisionTrees-bs027.html | 88 +- .../html/._DecisionTrees-bs028.html | 88 +- .../html/._DecisionTrees-bs029.html | 88 +- .../html/._DecisionTrees-bs030.html | 88 +- .../html/._DecisionTrees-bs031.html | 88 +- .../html/._DecisionTrees-bs032.html | 88 +- .../html/._DecisionTrees-bs033.html | 103 +- .../html/._DecisionTrees-bs034.html | 120 +- .../html/._DecisionTrees-bs035.html | 129 +- .../html/._DecisionTrees-bs036.html | 145 +- .../html/._DecisionTrees-bs037.html | 119 +- .../html/._DecisionTrees-bs038.html | 163 ++- .../html/._DecisionTrees-bs039.html | 167 ++- .../html/._DecisionTrees-bs040.html | 138 +- .../html/._DecisionTrees-bs041.html | 170 +-- .../html/._DecisionTrees-bs042.html | 166 ++- .../html/._DecisionTrees-bs043.html | 108 +- .../html/._DecisionTrees-bs044.html | 101 +- .../html/._DecisionTrees-bs045.html | 110 +- .../html/._DecisionTrees-bs046.html | 129 +- .../html/._DecisionTrees-bs047.html | 124 +- .../html/._DecisionTrees-bs048.html | 127 +- .../html/._DecisionTrees-bs049.html | 129 +- .../html/._DecisionTrees-bs050.html | 114 +- .../html/._DecisionTrees-bs051.html | 193 +-- .../html/._DecisionTrees-bs052.html | 199 +-- .../html/._DecisionTrees-bs053.html | 156 +- .../DecisionTrees/html/DecisionTrees-bs.html | 90 +- .../html/DecisionTrees-reveal.html | 62 +- .../html/DecisionTrees-solarized.html | 105 +- doc/pub/DecisionTrees/html/DecisionTrees.html | 105 +- .../DecisionTrees/ipynb/DecisionTrees.ipynb | 1275 +++-------------- .../ipynb/ipynb-DecisionTrees-src.tar.gz | Bin 213983 -> 294061 bytes .../pdf/DecisionTrees-minted.pdf | Bin 445114 -> 528514 bytes .../DataFiles/ensembleoverview.png | Bin 0 -> 81932 bytes doc/src/DecisionTrees/DecisionTrees.do.txt | 16 +- doc/src/DecisionTrees/Programs/rfcancer.py | 35 +- 64 files changed, 3356 insertions(+), 4148 deletions(-) create mode 100644 doc/src/DecisionTrees/DataFiles/ensembleoverview.png diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html index d44f3e9dc..532859bcb 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -249,7 +251,7 @@ MathJax.Hub.Config({
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University

    -

    Nov 5, 2019

    +

    Nov 6, 2019


    @@ -273,7 +275,7 @@ MathJax.Hub.Config({

  • 9
  • 10
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html index f9d50de42..b7c6325e5 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -278,7 +280,7 @@ given some assumptions, make predictions about the target feature value
  • 10
  • 11
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html index 61ed7cf48..a8748ce44 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -256,7 +258,7 @@ This tree was produced using the Wisconsin cancer data (discussed here as well,
  • 11
  • 12
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html index 116fc194b..56f38313c 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -264,7 +266,7 @@ node.
  • 12
  • 13
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html index 6d46d19cf..b5bc1fc5a 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -265,7 +267,7 @@ Then we are essentially done!
  • 13
  • 14
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html index 43a694004..4f74844c7 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -344,7 +346,7 @@ plt.show()
  • 14
  • 15
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html index ad7ba5d22..1ba0164b2 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -277,7 +279,7 @@ within box \( j \).
  • 15
  • 16
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html index c2aa59226..3663532e9 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -269,7 +271,7 @@ better tree in some future step.
  • 16
  • 17
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html index 6e65bb774..ad64a1ba1 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -302,7 +304,7 @@ region contains more than five observations.
  • 17
  • 18
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html index da729347a..52d193278 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -271,7 +273,7 @@ parameter \( \alpha \).
  • 18
  • 19
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html index 7f434547a..b2f5cacf7 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -284,7 +286,7 @@ subtree corresponding to \( \alpha \).
  • 19
  • 20
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html index eda8fd1f1..68a3f02d3 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -280,7 +282,7 @@ MathJax.Hub.Config({
  • 20
  • 21
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html index b5c8465ff..42cffb0de 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -272,7 +274,7 @@ fall into that region.
  • 21
  • 22
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html index 096e9532c..5733e08f7 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -277,7 +279,7 @@ than is the classification error rate.
  • 22
  • 23
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html index 07727f977..88ae28e91 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -303,7 +305,7 @@ $$
  • 23
  • 24
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html index 28e41e197..818a8d847 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -294,7 +296,7 @@ os.system(cmd)
  • 24
  • 25
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html index eae1cedb0..b5710e68c 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -285,7 +287,7 @@ os.system(cmd)
  • 25
  • 26
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html index e21a486a1..3e07f6869 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -266,7 +268,7 @@ We discuss both algorithms with applications here. The popular library -Scikit-L
  • 26
  • 27
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html index 22fb991f5..62dbbc302 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -258,7 +260,7 @@ MathJax.Hub.Config({
  • 27
  • 28
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html index c1b270ff4..8b19e19bb 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -258,7 +260,7 @@ MathJax.Hub.Config({
  • 28
  • 29
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html index 0615b8301..a915226ba 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -298,7 +300,7 @@ The table here summarizes the various attributes and
  • 29
  • 30
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html index dafcb108a..cf72c7947 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -329,7 +331,7 @@ os.system(cmd)
  • 30
  • 31
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html index 081cbcb0a..e7b8aa064 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -331,7 +333,7 @@ split = get_split(dataset)
  • 31
  • 32
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html index 8da93e536..39dd82763 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -289,7 +291,7 @@ attributes at each step while growing the tree.
  • 32
  • 33
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html index c1b3ddd9c..f4ea06f2d 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -449,7 +451,7 @@ MathJax.Hub.Config({
  • 33
  • 34
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html index d15b58644..f66bca351 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -302,7 +304,7 @@ deep_tree_clf.fit(X_train_scaled, y_train)
  • 34
  • 35
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html index e9148a901..5a7c5c524 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -325,7 +327,7 @@ plt.show()
  • 35
  • 36
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html index 361b898f1..968b88265 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -281,7 +283,7 @@ plt.show()
  • 36
  • 37
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html index 2d4dcc7f0..999e10fd3 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -275,7 +277,7 @@ tree_reg.fit(X, y)
  • 37
  • 38
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html index 02dc79e85..763697a01 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -331,7 +333,7 @@ plt.show()
  • 38
  • 39
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html index 26491eaa6..713e3bf13 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -267,7 +269,7 @@ MathJax.Hub.Config({
  • 39
  • 40
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html index f08c9f7b7..c6626c3b9 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -270,7 +272,7 @@ However, by aggregating many decision trees, using methods like bagging, random
  • 40
  • 41
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html index 97b823338..ed1d0ac50 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -277,7 +279,7 @@ We discuss these methods here.
  • 41
  • 42
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html index 1d96f5fdc..7332fd504 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -230,21 +232,10 @@ MathJax.Hub.Config({ -

    Bagging

    +

    An Overview of Ensemble Methods

    -The plain decision trees suffer from high -variance. This means that if we split the training data into two parts -at random, and fit a decision tree to both halves, the results that we -get could be quite different. In contrast, a procedure with low -variance will yield similar results if applied repeatedly to distinct -data sets; linear regression tends to have low variance, if the ratio -of \( n \) to \( p \) is moderately large. - -

    -Bootstrap aggregation, or just bagging, is a -general-purpose procedure for reducing the variance of a statistical -learning method. +



    @@ -272,7 +263,7 @@ learning method.

  • 42
  • 43
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html index 39ad277b2..095abbcbb 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -230,31 +232,21 @@ MathJax.Hub.Config({ -

    More bagging

    +

    Bagging

    -Bagging typically results in improved accuracy -over prediction using a single tree. Unfortunately, however, it can be -difficult to interpret the resulting model. Recall that one of the -advantages of decision trees is the attractive and easily interpreted -diagram that results. +The plain decision trees suffer from high +variance. This means that if we split the training data into two parts +at random, and fit a decision tree to both halves, the results that we +get could be quite different. In contrast, a procedure with low +variance will yield similar results if applied repeatedly to distinct +data sets; linear regression tends to have low variance, if the ratio +of \( n \) to \( p \) is moderately large.

    -However, when we bag a large number of trees, it is no longer -possible to represent the resulting statistical learning procedure -using a single tree, and it is no longer clear which variables are -most important to the procedure. Thus, bagging improves prediction -accuracy at the expense of interpretability. Although the collection -of bagged trees is much more difficult to interpret than a single -tree, one can obtain an overall summary of the importance of each -predictor using the MSE (for bagging regression trees) or the Gini -index (for bagging classification trees). In the case of bagging -regression trees, we can record the total amount that the MSE is -decreased due to splits over a given predictor, averaged over all \( B \) possible -trees. A large value indicates an important predictor. Similarly, in -the context of bagging classification trees, we can add up the total -amount that the Gini index is decreased by splits over a given -predictor, averaged over all \( B \) trees. +Bootstrap aggregation, or just bagging, is a +general-purpose procedure for reducing the variance of a statistical +learning method.

    @@ -282,7 +274,7 @@ predictor, averaged over all \( B \) trees.

  • 43
  • 44
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html index 4d249c000..3879ad651 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -230,23 +232,32 @@ MathJax.Hub.Config({ -

    Simple Voting Example, head or tail

    -

    +

    More bagging

    + +

    +Bagging typically results in improved accuracy +over prediction using a single tree. Unfortunately, however, it can be +difficult to interpret the resulting model. Recall that one of the +advantages of decision trees is the attractive and easily interpreted +diagram that results. + +

    +However, when we bag a large number of trees, it is no longer +possible to represent the resulting statistical learning procedure +using a single tree, and it is no longer clear which variables are +most important to the procedure. Thus, bagging improves prediction +accuracy at the expense of interpretability. Although the collection +of bagged trees is much more difficult to interpret than a single +tree, one can obtain an overall summary of the importance of each +predictor using the MSE (for bagging regression trees) or the Gini +index (for bagging classification trees). In the case of bagging +regression trees, we can record the total amount that the MSE is +decreased due to splits over a given predictor, averaged over all \( B \) possible +trees. A large value indicates an important predictor. Similarly, in +the context of bagging classification trees, we can add up the total +amount that the Gini index is decreased by splits over a given +predictor, averaged over all \( B \) trees. - -

    heads_proba = 0.51
-coin_tosses = (np.random.rand(10000, 10) < heads_proba).astype(np.int32)
-cumulative_heads_ratio = np.cumsum(coin_tosses, axis=0) / np.arange(1, 10001).reshape(-1, 1)
-plt.figure(figsize=(8,3.5))
-plt.plot(cumulative_heads_ratio)
-plt.plot([0, 10000], [0.51, 0.51], "k--", linewidth=2, label="51%")
-plt.plot([0, 10000], [0.5, 0.5], "k-", label="50%")
-plt.xlabel("Number of coin tosses")
-plt.ylabel("Heads ratio")
-plt.legend(loc="lower right")
-plt.axis([0, 10000, 0.42, 0.58])
-plt.show()
-

    @@ -273,7 +284,7 @@ plt.show()

  • 44
  • 45
  • ...
    • -
  • 54
    • +
  • 55
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html index 38dc2899e..1480014a9 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html @@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source 2, None, '___sec31'), - ('Bagging', 2, None, '___sec32'), - ('More bagging', 2, None, '___sec33'), - ('Simple Voting Example, head or tail', 2, None, '___sec34'), - ('Using the Voting Classifier', 2, None, '___sec35'), + ('An Overview of Ensemble Methods', 2, None, '___sec32'), + ('Bagging', 2, None, '___sec33'), + ('More bagging', 2, None, '___sec34'), + ('Simple Voting Example, head or tail', 2, None, '___sec35'), + ('Using the Voting Classifier', 2, None, '___sec36'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec36'), - ('Now Bagging', 2, None, '___sec37'), - ('Random forests', 2, None, '___sec38'), - ('Random Forest Algorithm', 2, None, '___sec39'), + '___sec37'), + ('Now Bagging', 2, None, '___sec38'), + ('Random forests', 2, None, '___sec39'), + ('Random Forest Algorithm', 2, None, '___sec40'), ('Random Forests Compared with other Methods on the Cancer Data', 2, None, - '___sec40'), + '___sec41'), ('Compare Bagging on Trees with Random Forests', 2, None, - '___sec41'), - ('Feature Importance', 2, None, '___sec42'), - ("Boosting, a Bird'e Eye", 2, None, '___sec43'), + '___sec42'), + ('Feature Importance', 2, None, '___sec43'), + ("Boosting, a Bird'e Eye", 2, None, '___sec44'), ('Adaptive boosting: AdaBoost, Basic Algorithm', 2, None, - '___sec44'), - ('Basic Steps of AdaBoost', 2, None, '___sec45'), + '___sec45'), + ('Basic Steps of AdaBoost', 2, None, '___sec46'), ('Figure to Illustrate the Iterative Classification Process', 2, None, - '___sec46'), - ('AdaBoost Examples', 2, None, '___sec47'), - ('Gradient boosting: Basics', 2, None, '___sec48'), - ('Gradient Boosting, algorithm', 2, None, '___sec49'), - ('Gradient Boosting, Examples', 2, None, '___sec50'), - ('Gradient Boots with Early Stopping', 2, None, '___sec51'), - ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]} + '___sec47'), + ('AdaBoost Examples', 2, None, '___sec48'), + ('Gradient boosting: Basics', 2, None, '___sec49'), + ('Gradient Boosting, algorithm', 2, None, '___sec50'), + ('Gradient Boosting, Examples', 2, None, '___sec51'), + ('Gradient Boots with Early Stopping', 2, None, '___sec52'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]} end of tocinfo --> @@ -194,27 +195,28 @@ MathJax.Hub.Config({
  • Pros and cons of trees, pros
  • Disadvantages
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • -
  • Bagging
    • -
  • More bagging
    • -
  • Simple Voting Example, head or tail
    • -
  • Using the Voting Classifier
    • -
  • Please, not the moons again! Voting and Bagging
    • -
  • Now Bagging
    • -
  • Random forests
    • -
  • Random Forest Algorithm
    • -
  • Random Forests Compared with other Methods on the Cancer Data
    • -
  • Compare Bagging on Trees with Random Forests
    • -
  • Feature Importance
    • -
  • Boosting, a Bird'e Eye
    • -
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • -
  • Basic Steps of AdaBoost
    • -
  • Figure to Illustrate the Iterative Classification Process
    • -
  • AdaBoost Examples
    • -
  • Gradient boosting: Basics
    • -
  • Gradient Boosting, algorithm
    • -
  • Gradient Boosting, Examples
    • -
  • Gradient Boots with Early Stopping
    • -
  • XGBoost: Extreme Gradient Boosting
    • +
  • An Overview of Ensemble Methods
    • +
  • Bagging
    • +
  • More bagging
    • +
  • Simple Voting Example, head or tail
    • +
  • Using the Voting Classifier
    • +
  • Please, not the moons again! Voting and Bagging
    • +
  • Now Bagging
    • +
  • Random forests
    • +
  • Random Forest Algorithm
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • Basic Steps of AdaBoost
    • +
  • Figure to Illustrate the Iterative Classification Process
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basics
    • +
  • Gradient Boosting, algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -230,53 +232,22 @@ MathJax.Hub.Config({ -

    Using the Voting Classifier

    +

    Simple Voting Example, head or tail

    -

    from sklearn.model_selection import train_test_split
-from sklearn.datasets import make_moons
-
-X, y = make_moons(n_samples=500, noise=0.30, random_state=42)
-X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
-
-from sklearn.ensemble import RandomForestClassifier
-from sklearn.ensemble import VotingClassifier
-from sklearn.linear_model import LogisticRegression
-from sklearn.svm import SVC
-
-log_clf = LogisticRegression(solver="liblinear", random_state=42)
-rnd_clf = RandomForestClassifier(n_estimators=10, random_state=42)
-svm_clf = SVC(gamma="auto", random_state=42)
-
-voting_clf = VotingClassifier(
-    estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
-    voting='hard')
-
-voting_clf.fit(X_train, y_train)
-
-from sklearn.metrics import accuracy_score
-
-for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
-    clf.fit(X_train, y_train)
-    y_pred = clf.predict(X_test)
-    print(clf.__class__.__name__, accuracy_score(y_test, y_pred))
-
-log_clf = LogisticRegression(solver="liblinear", random_state=42)
-rnd_clf = RandomForestClassifier(n_estimators=10, random_state=42)
-svm_clf = SVC(gamma="auto", probability=True, random_state=42)
-
-voting_clf = VotingClassifier(
-    estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
-    voting='soft')
-voting_clf.fit(X_train, y_train)
-
-from sklearn.metrics import accuracy_score
-
-for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
-    clf.fit(X_train, y_train)
-    y_pred = clf.predict(X_test)
-    print(clf.__class__.__name__, accuracy_score(y_test, y_pred))
+
    heads_proba = 0.51
+coin_tosses = (np.random.rand(10000, 10) < heads_proba).astype(np.int32)
+cumulative_heads_ratio = np.cumsum(coin_tosses, axis=0) / np.arange(1, 10001).reshape(-1, 1)
+plt.figure(figsize=(8,3.5))
+plt.plot(cumulative_heads_ratio)
+plt.plot([0, 10000], [0.51, 0.51], "k--", linewidth=2, label="51%")
+plt.plot([0, 10000], [0.5, 0.5], "k-", label="50%")
+plt.xlabel("Number of coin tosses")
+plt.ylabel("Heads ratio")
+plt.legend(loc="lower right")
+plt.axis([0, 10000, 0.42, 0.58])
+plt.show()
 
    
 
    
 
    
@@ -304,7 +275,7 @@ voting_clf.fit(X_train, y_train)
   
  • 45
    • 
   
  • 46
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
index 2ac864af2..4f248f6f5 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,7 +232,7 @@ MathJax.Hub.Config({
 
 
 
-
    Please, not the moons again! Voting and Bagging 
    
+
    Using the Voting Classifier 
    
 
    
 
 
@@ -239,46 +241,39 @@ MathJax.Hub.Config({
 
 X, y = make_moons(n_samples=500, noise=0.30, random_state=42)
 X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
+
 from sklearn.ensemble import RandomForestClassifier
 from sklearn.ensemble import VotingClassifier
 from sklearn.linear_model import LogisticRegression
 from sklearn.svm import SVC
 
-log_clf = LogisticRegression(random_state=42)
-rnd_clf = RandomForestClassifier(random_state=42)
-svm_clf = SVC(random_state=42)
+log_clf = LogisticRegression(solver="liblinear", random_state=42)
+rnd_clf = RandomForestClassifier(n_estimators=10, random_state=42)
+svm_clf = SVC(gamma="auto", random_state=42)
 
 voting_clf = VotingClassifier(
     estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
     voting='hard')
-voting_clf.fit(X_train, y_train)
-
    -

    - -

    from sklearn.metrics import accuracy_score
+voting_clf.fit(X_train, y_train)
+
+from sklearn.metrics import accuracy_score
 
 for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
     clf.fit(X_train, y_train)
     y_pred = clf.predict(X_test)
     print(clf.__class__.__name__, accuracy_score(y_test, y_pred))
-
    -

    - -

    log_clf = LogisticRegression(random_state=42)
-rnd_clf = RandomForestClassifier(random_state=42)
-svm_clf = SVC(probability=True, random_state=42)
+log_clf = LogisticRegression(solver="liblinear", random_state=42)
+rnd_clf = RandomForestClassifier(n_estimators=10, random_state=42)
+svm_clf = SVC(gamma="auto", probability=True, random_state=42)
 
 voting_clf = VotingClassifier(
     estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
     voting='soft')
 voting_clf.fit(X_train, y_train)
-
    -

    - -

    from sklearn.metrics import accuracy_score
+from sklearn.metrics import accuracy_score
 
 for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
     clf.fit(X_train, y_train)
@@ -311,7 +306,7 @@ voting_clf.fit(X_train, y_train)
   
  • 46
    • 
   
  • 47
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
index aaf42f6a9..17feef8c7 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,63 +232,60 @@ MathJax.Hub.Config({
 
 
 
-
    Now Bagging  
    
-
+
    Please, not the moons again! Voting and Bagging 
    
 
    
 
 
-
    from sklearn.ensemble import BaggingClassifier
-from sklearn.tree import DecisionTreeClassifier
+
    from sklearn.model_selection import train_test_split
+from sklearn.datasets import make_moons
 
-bag_clf = BaggingClassifier(
-    DecisionTreeClassifier(random_state=42), n_estimators=500,
-    max_samples=100, bootstrap=True, n_jobs=-1, random_state=42)
-bag_clf.fit(X_train, y_train)
-y_pred = bag_clf.predict(X_test)
+X, y = make_moons(n_samples=500, noise=0.30, random_state=42)
+X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.ensemble import VotingClassifier
+from sklearn.linear_model import LogisticRegression
+from sklearn.svm import SVC
+
+log_clf = LogisticRegression(random_state=42)
+rnd_clf = RandomForestClassifier(random_state=42)
+svm_clf = SVC(random_state=42)
+
+voting_clf = VotingClassifier(
+    estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
+    voting='hard')
+voting_clf.fit(X_train, y_train)
 
    
 
    
 
 
 
    from sklearn.metrics import accuracy_score
-print(accuracy_score(y_test, y_pred))
+
+for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
+    clf.fit(X_train, y_train)
+    y_pred = clf.predict(X_test)
+    print(clf.__class__.__name__, accuracy_score(y_test, y_pred))
 
    
 
    
 
 
-
    tree_clf = DecisionTreeClassifier(random_state=42)
-tree_clf.fit(X_train, y_train)
-y_pred_tree = tree_clf.predict(X_test)
-print(accuracy_score(y_test, y_pred_tree))
+
    log_clf = LogisticRegression(random_state=42)
+rnd_clf = RandomForestClassifier(random_state=42)
+svm_clf = SVC(probability=True, random_state=42)
+
+voting_clf = VotingClassifier(
+    estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
+    voting='soft')
+voting_clf.fit(X_train, y_train)
 
    
 
    
 
 
-
    from matplotlib.colors import ListedColormap
+
    from sklearn.metrics import accuracy_score
 
-def plot_decision_boundary(clf, X, y, axes=[-1.5, 2.5, -1, 1.5], alpha=0.5, contour=True):
-    x1s = np.linspace(axes[0], axes[1], 100)
-    x2s = np.linspace(axes[2], axes[3], 100)
-    x1, x2 = np.meshgrid(x1s, x2s)
-    X_new = np.c_[x1.ravel(), x2.ravel()]
-    y_pred = clf.predict(X_new).reshape(x1.shape)
-    custom_cmap = ListedColormap(['#fafab0','#9898ff','#a0faa0'])
-    plt.contourf(x1, x2, y_pred, alpha=0.3, cmap=custom_cmap)
-    if contour:
-        custom_cmap2 = ListedColormap(['#7d7d58','#4c4c7f','#507d50'])
-        plt.contour(x1, x2, y_pred, cmap=custom_cmap2, alpha=0.8)
-    plt.plot(X[:, 0][y==0], X[:, 1][y==0], "yo", alpha=alpha)
-    plt.plot(X[:, 0][y==1], X[:, 1][y==1], "bs", alpha=alpha)
-    plt.axis(axes)
-    plt.xlabel(r"$x_1$", fontsize=18)
-    plt.ylabel(r"$x_2$", fontsize=18, rotation=0)
-plt.figure(figsize=(11,4))
-plt.subplot(121)
-plot_decision_boundary(tree_clf, X, y)
-plt.title("Decision Tree", fontsize=14)
-plt.subplot(122)
-plot_decision_boundary(bag_clf, X, y)
-plt.title("Decision Trees with Bagging", fontsize=14)
-plt.show()
+for clf in (log_clf, rnd_clf, svm_clf, voting_clf):
+    clf.fit(X_train, y_train)
+    y_pred = clf.predict(X_test)
+    print(clf.__class__.__name__, accuracy_score(y_test, y_pred))
 
    
 
    
 
    
@@ -314,7 +313,7 @@ plt.show()
   
  • 47
    • 
   
  • 48
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs039.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs039.html
index 2f2ba5519..ccb8fad5c 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs039.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs039.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,47 +232,64 @@ MathJax.Hub.Config({
 
 
 
-
    Random forests 
    
+
    Now Bagging  
    
 
 
    
-Random forests provide an improvement over bagged trees by way of a
-small tweak that decorrelates the trees.
 
+
+
    from sklearn.ensemble import BaggingClassifier
+from sklearn.tree import DecisionTreeClassifier
+
+bag_clf = BaggingClassifier(
+    DecisionTreeClassifier(random_state=42), n_estimators=500,
+    max_samples=100, bootstrap=True, n_jobs=-1, random_state=42)
+bag_clf.fit(X_train, y_train)
+y_pred = bag_clf.predict(X_test)
+
    
 
    
-As in bagging, we build a
-number of decision trees on bootstrapped training samples. But when
-building these decision trees, each time a split in a tree is
-considered, a random sample of \( m \) predictors is chosen as split
-candidates from the full set of \( p \) predictors. The split is allowed to
-use only one of those \( m \) predictors.
 
+
+
    from sklearn.metrics import accuracy_score
+print(accuracy_score(y_test, y_pred))
+
    
 
    
-A fresh sample of \( m \) predictors is
-taken at each split, and typically we choose 
-
-$$
-m\approx \sqrt{p}.
-$$
 
+
+
    tree_clf = DecisionTreeClassifier(random_state=42)
+tree_clf.fit(X_train, y_train)
+y_pred_tree = tree_clf.predict(X_test)
+print(accuracy_score(y_test, y_pred_tree))
+
    
 
    
-In building a random forest, at
-each split in the tree, the algorithm is not even allowed to consider
-a majority of the available predictors.
 
-
    
-The reason for this is rather clever. Suppose that there is one very
-strong predictor in the data set, along with a number of other
-moderately strong predictors. Then in the collection of bagged
-variable importance random forest trees, most or all of the trees will
-use this strong predictor in the top split. Consequently, all of the
-bagged trees will look quite similar to each other. Hence the
-predictions from the bagged trees will be highly correlated.
-Unfortunately, averaging many highly correlated quantities does not
-lead to as large of a reduction in variance as averaging many
-uncorrelated quanti- ties. In particular, this means that bagging will
-not lead to a substantial reduction in variance over a single tree in
-this setting.
+
+
    from matplotlib.colors import ListedColormap
 
+def plot_decision_boundary(clf, X, y, axes=[-1.5, 2.5, -1, 1.5], alpha=0.5, contour=True):
+    x1s = np.linspace(axes[0], axes[1], 100)
+    x2s = np.linspace(axes[2], axes[3], 100)
+    x1, x2 = np.meshgrid(x1s, x2s)
+    X_new = np.c_[x1.ravel(), x2.ravel()]
+    y_pred = clf.predict(X_new).reshape(x1.shape)
+    custom_cmap = ListedColormap(['#fafab0','#9898ff','#a0faa0'])
+    plt.contourf(x1, x2, y_pred, alpha=0.3, cmap=custom_cmap)
+    if contour:
+        custom_cmap2 = ListedColormap(['#7d7d58','#4c4c7f','#507d50'])
+        plt.contour(x1, x2, y_pred, cmap=custom_cmap2, alpha=0.8)
+    plt.plot(X[:, 0][y==0], X[:, 1][y==0], "yo", alpha=alpha)
+    plt.plot(X[:, 0][y==1], X[:, 1][y==1], "bs", alpha=alpha)
+    plt.axis(axes)
+    plt.xlabel(r"$x_1$", fontsize=18)
+    plt.ylabel(r"$x_2$", fontsize=18, rotation=0)
+plt.figure(figsize=(11,4))
+plt.subplot(121)
+plot_decision_boundary(tree_clf, X, y)
+plt.title("Decision Tree", fontsize=14)
+plt.subplot(122)
+plot_decision_boundary(bag_clf, X, y)
+plt.title("Decision Trees with Bagging", fontsize=14)
+plt.show()
+
    
 
    
 
    
 
@@ -297,7 +316,7 @@ this setting.
   
  • 48
    • 
   
  • 49
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs040.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs040.html
index f14ffae64..7ebfb7732 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs040.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs040.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,30 +232,48 @@ MathJax.Hub.Config({
 
 
 
-
    Random Forest Algorithm 
    
-The algorithm described here can be applied to both classification and regression problems.
+
    Random forests 
    
 
 
    
-We will grow of forest of say \( M \) trees.
+Random forests provide an improvement over bagged trees by way of a
+small tweak that decorrelates the trees.
 
-
      
-
    1.  For \( m=1:M \) we
      2. 
+
      
+As in bagging, we build a
+number of decision trees on bootstrapped training samples. But when
+building these decision trees, each time a split in a tree is
+considered, a random sample of \( m \) predictors is chosen as split
+candidates from the full set of \( p \) predictors. The split is allowed to
+use only one of those \( m \) predictors.
 
-
        
-  
      •  Draw a bootstrap sample of from the training data organized in our \( \boldsymbol{X} \) matrix.
        • 
-  
      •  We grow then a random forest tree \( T_m \) based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
        • 
+
        
+A fresh sample of \( m \) predictors is
+taken at each split, and typically we choose 
 
-
          
-   
        1.  we select \( m \le p \) varibales at random from the \( p \) predictors/features
          2. 
-   
        2.  pick the best split point among the \( m \) features using either the CART algorithm or the ID3 for classification and create a new node
          3. 
-   
        3.  split the node into daughter nodes
          4. 
-
        
+$$
+m\approx \sqrt{p}.
+$$
 
-
      
+
      
+In building a random forest, at
+each split in the tree, the algorithm is not even allowed to consider
+a majority of the available predictors.
 
-
    2.  Output then the ensemble of trees \( \{T_m\}_1^{M} \) and make predictions for either a regression type of problem or a classification type of problem.
      3.  
-
    
+
    
+The reason for this is rather clever. Suppose that there is one very
+strong predictor in the data set, along with a number of other
+moderately strong predictors. Then in the collection of bagged
+variable importance random forest trees, most or all of the trees will
+use this strong predictor in the top split. Consequently, all of the
+bagged trees will look quite similar to each other. Hence the
+predictions from the bagged trees will be highly correlated.
+Unfortunately, averaging many highly correlated quantities does not
+lead to as large of a reduction in variance as averaging many
+uncorrelated quanti- ties. In particular, this means that bagging will
+not lead to a substantial reduction in variance over a single tree in
+this setting.
 
+
    
 
    
 
 
      
@@ -279,7 +299,7 @@ We will grow of forest of say \( M \) trees.
   
    • 49
      • 
   
    • 50
      • 
   
    • ...
      • 
-  
    • 54
      • 
+  
    • 55
      • 
   
    • »
      • 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs041.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs041.html
index 78b84e253..09e205e9c 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs041.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs041.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,78 +232,30 @@ MathJax.Hub.Config({
 
 
 
-
    Random Forests Compared with other Methods on the Cancer Data 
    
+
    Random Forest Algorithm 
    
+The algorithm described here can be applied to both classification and regression problems.
+
 
    
+We will grow of forest of say \( M \) trees.
 
-
-
    import matplotlib.pyplot as plt
-import numpy as np
-from sklearn.model_selection import  train_test_split 
-from sklearn.datasets import load_breast_cancer
-from sklearn.svm import SVC
-from sklearn.linear_model import LogisticRegression
-from sklearn.tree import DecisionTreeClassifier
+
      
+
    1.  For \( m=1:M \) we
      2. 
 
-# Load the data
-cancer = load_breast_cancer()
+
        
+  
      •  Draw a bootstrap sample of from the training data organized in our \( \boldsymbol{X} \) matrix.
        • 
+  
      •  We grow then a random forest tree \( T_m \) based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
        • 
 
-X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
-print(X_train.shape)
-print(X_test.shape)
-# Logistic Regression
-logreg = LogisticRegression(solver='lbfgs')
-logreg.fit(X_train, y_train)
-print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
-# Support vector machine
-svm = SVC(gamma='auto', C=100)
-svm.fit(X_train, y_train)
-print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
-# Decision Trees
-deep_tree_clf = DecisionTreeClassifier(max_depth=None)
-deep_tree_clf.fit(X_train, y_train)
-print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
-#now scale the data
-from sklearn.preprocessing import StandardScaler
-scaler = StandardScaler()
-scaler.fit(X_train)
-X_train_scaled = scaler.transform(X_train)
-X_test_scaled = scaler.transform(X_test)
-# Logistic Regression
-logreg.fit(X_train_scaled, y_train)
-print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
-# Support Vector Machine
-svm.fit(X_train_scaled, y_train)
-print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
-# Decision Trees
-deep_tree_clf.fit(X_train_scaled, y_train)
-print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
          
+   
        1.  we select \( m \le p \) varibales at random from the \( p \) predictors/features
          2. 
+   
        2.  pick the best split point among the \( m \) features using either the CART algorithm or the ID3 for classification and create a new node
          3. 
+   
        3.  split the node into daughter nodes
          4. 
+
        
 
+
      
 
-from sklearn.ensemble import RandomForestClassifier
-from sklearn.preprocessing import LabelEncoder
-from sklearn.model_selection import cross_validate
-# Data set not specificied
-#Instantiate the model with 500 trees and entropy as splitting criteria
-Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
-Random_Forest_model.fit(X_train_scaled, y_train)
-#Cross validation
-accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
-print(accuracy)
-print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
    2.  Output then the ensemble of trees \( \{T_m\}_1^{M} \) and make predictions for either a regression type of problem or a classification type of problem.
      3.  
+
    
 
-
-import scikitplot as skplt
-y_pred = Random_Forest_model.predict(X_test_scaled)
-skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
-#<matplotlib.axes._subplots.AxesSubplot object at 0x7fe967d64490>
-plt.show()
-y_probas = Random_Forest_model.predict_proba(X_test_scaled)
-skplt.metrics.plot_roc(y_test, y_probas)
-plt.show()
-skplt.metrics.plot_cumulative_gain(y_test, y_probas)
-plt.show()
-
    
-
    
 
    
 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs042.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs042.html
index 9b3514df5..8cbdfe4b6 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs042.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs042.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,24 +232,76 @@ MathJax.Hub.Config({
 
 
 
-
    Compare  Bagging on Trees with Random Forests  
    
+
    Random Forests Compared with other Methods on the Cancer Data 
    
 
    
 
 
-
    bag_clf = BaggingClassifier(
-    DecisionTreeClassifier(splitter="random", max_leaf_nodes=16, random_state=42),
-    n_estimators=500, max_samples=1.0, bootstrap=True, n_jobs=-1, random_state=42)
-
    
-
    
+
    import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
 
-
-
    bag_clf.fit(X_train, y_train)
-y_pred = bag_clf.predict(X_test)
 from sklearn.ensemble import RandomForestClassifier
-rnd_clf = RandomForestClassifier(n_estimators=500, max_leaf_nodes=16, n_jobs=-1, random_state=42)
-rnd_clf.fit(X_train, y_train)
-y_pred_rf = rnd_clf.predict(X_test)
-np.sum(y_pred == y_pred_rf) / len(y_pred) 
+from sklearn.preprocessing import LabelEncoder
+from sklearn.model_selection import cross_validate
+# Data set not specificied
+#Instantiate the model with 500 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
+#Cross validation
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#<matplotlib.axes._subplots.AxesSubplot object at 0x7fe967d64490>
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()
 
    
 
    
 
    
@@ -275,7 +329,7 @@ np.sum(y_pred =
   
  • 51
    • 
   
  • 52
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs043.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs043.html
index 3ea5fbe0c..0b45318a1 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs043.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs043.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,11 +232,25 @@ MathJax.Hub.Config({
 
 
 
-
    Feature Importance 
    
-
+
    Compare  Bagging on Trees with Random Forests  
    
 
    
-Example will be added here.
 
+
+
    bag_clf = BaggingClassifier(
+    DecisionTreeClassifier(splitter="random", max_leaf_nodes=16, random_state=42),
+    n_estimators=500, max_samples=1.0, bootstrap=True, n_jobs=-1, random_state=42)
+
    
+
    
+
+
+
    bag_clf.fit(X_train, y_train)
+y_pred = bag_clf.predict(X_test)
+from sklearn.ensemble import RandomForestClassifier
+rnd_clf = RandomForestClassifier(n_estimators=500, max_leaf_nodes=16, n_jobs=-1, random_state=42)
+rnd_clf.fit(X_train, y_train)
+y_pred_rf = rnd_clf.predict(X_test)
+np.sum(y_pred == y_pred_rf) / len(y_pred) 
+
    
 
    
 
    
 
@@ -261,7 +277,7 @@ Example will be added here.
   
  • 52
    • 
   
  • 53
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs044.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs044.html
index 5519ff351..159f3cf2a 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs044.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs044.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,19 +232,10 @@ MathJax.Hub.Config({
 
 
 
-
    Boosting, a Bird'e Eye 
    
+
    Feature Importance 
    
 
 
    
-The basic idea is to combine weak classifiers in order to create a good
-classifier. With a weak classifier we often intend a classifier which
-produces results which are only slightly better than we would get by
-random guesses.
-
-
    
-This is done by applying in an iterative way a weak (or a standard
-classifier like decision trees) to modify the data. In each iteration
-we emphasize those observations which are misclassified by weighting
-them with a factor.
+Example will be added here.
 
 
    
 
    
@@ -269,6 +262,8 @@ them with a factor.
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • ...
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs045.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs045.html
index 4afd6e9ef..875fb75aa 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs045.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs045.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,24 +232,19 @@ MathJax.Hub.Config({
 
 
 
-
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
+
    Boosting, a Bird'e Eye 
    
 
 
    
-The algorithm here is rather straightforward. Assume that our weak
-classifier is a decision tree and we consider a binary set of outputs
-with \( y_i \in \{-1,1\} \) and \( i=0,1,2,\dots,n-1 \) as our set of
-observations. Our design matrix is given in terms of the
-feature/predictor vectors
-\( \boldsymbol{X}=[\boldsymbol{x}_0\boldsymbol{x}_1\dots\boldsymbol{x}_{p-1} \). Finally, we define also a
-classifier determined by our data via a function \( G(\boldsymbol{X}) \). This function tells us how well we are able to classify our outputs/targets \( \boldsymbol{y} \).
+The basic idea is to combine weak classifiers in order to create a good
+classifier. With a weak classifier we often intend a classifier which
+produces results which are only slightly better than we would get by
+random guesses.
 
 
    
-We can then define the misclassification error \( \mathrm{err} \) as
-$$
-\mathrm{err}=\frac{1}{n}\sum_{i=0}^{n-1}I(y_i\ne G(\boldsymbol{X}_{i*}),
-$$
-
-where the function \( I() \) is one if we misclassify and zero if we classify correctly.
+This is done by applying in an iterative way a weak (or a standard
+classifier like decision trees) to modify the data. In each iteration
+we emphasize those observations which are misclassified by weighting
+them with a factor.
 
 
    
 
    
@@ -273,6 +270,7 @@ where the function \( I() \) is one if we misclassify and zero if we classify co
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs046.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs046.html
index 11b689d88..4240a0897 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs046.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs046.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,42 +232,24 @@ MathJax.Hub.Config({
 
 
 
-
    Basic Steps of AdaBoost 
    
+
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
 
 
    
-With the above definitions we are now ready to set up the algorithm for AdaBoost.
-The basic idea is to set up weights which will be used to scale the correctly classified and the misclassified cases.
-
-
      
-
    1.  We start by initializing all weights to \( w_i = 1/n \), with \( i=0,1,2,\dots n-1 \). It is to see then that \( \sum_{i=0}^{n-1}w_i = 1 \).
      2. 
-
    2.  We rewrite the misclassification error as
      3.  
-
    
+The algorithm here is rather straightforward. Assume that our weak
+classifier is a decision tree and we consider a binary set of outputs
+with \( y_i \in \{-1,1\} \) and \( i=0,1,2,\dots,n-1 \) as our set of
+observations. Our design matrix is given in terms of the
+feature/predictor vectors
+\( \boldsymbol{X}=[\boldsymbol{x}_0\boldsymbol{x}_1\dots\boldsymbol{x}_{p-1} \). Finally, we define also a
+classifier determined by our data via a function \( G(\boldsymbol{X}) \). This function tells us how well we are able to classify our outputs/targets \( \boldsymbol{y} \).
 
+
    
+We can then define the misclassification error \( \mathrm{err} \) as
 $$
-\mathrm{err}=\frac{\sum_{i=0}^{n-1}w_iI(y_i\ne G(\boldsymbol{X}_{i*})}{\sum_{i=0}^{n-1}w_i},
+\mathrm{err}=\frac{1}{n}\sum_{i=0}^{n-1}I(y_i\ne G(\boldsymbol{X}_{i*}),
 $$
 
-
-
      
-
    1.  Then we start looping over all attempts at classifying, namely we start an iterative process for \( m=1:M \), where \( M \) is the final number of classifications. Our given classifier could for example be a plain decision tree.
-
-
        
- 
      1.  Fit thus a given classifier to the training using the weights \( w_i \).
        2. 
- 
      2.  Compute then \( \mathrm{err} \) and figure out which events are classified properly and which are classified wrongly.
        3. 
- 
      3.  Define a quantity $\alpha_{m} = \log{(1-\mathrm{err})/\mathrm{err}}
        4. 
- 
      4.  Set the new weights to $w_i = w_i\times \exp{(\alpha_m I(y_i\ne G(\boldsymbol{X}_{i*})}.
        5. 
-
      
-
-
    2.  Compute the new classifier $G(\boldsymbol{X})= \sum_{i=0}^{n-1}\alpha_m I(y_i\ne G(\boldsymbol{X}_{i*}).
      3. 
-
    
-
-For the iterations with \( m \le 2 \) the weights are modified
-individually at each steps. The obersvations which were misclassified
-at iteration \( m-1 \) have a weight which is larger than those which were
-classified properly. As this proceeds, the observations which were
-difficult to classifiy correctly are given a larger influence. Each
-new classificatio step \( m \) is then forced to concentrate on those
-observations that are missed in the previous iterations.
+where the function \( I() \) is one if we misclassify and zero if we classify correctly.
 
 
    
 
    
@@ -290,6 +274,7 @@ observations that are missed in the previous iterations.
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs047.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs047.html
index 416aaf671..e2b533008 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs047.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs047.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,7 +232,42 @@ MathJax.Hub.Config({
 
 
 
-
    Figure to Illustrate the Iterative Classification Process 
    
+
    Basic Steps of AdaBoost 
    
+
+
    
+With the above definitions we are now ready to set up the algorithm for AdaBoost.
+The basic idea is to set up weights which will be used to scale the correctly classified and the misclassified cases.
+
+
      
+
    1.  We start by initializing all weights to \( w_i = 1/n \), with \( i=0,1,2,\dots n-1 \). It is to see then that \( \sum_{i=0}^{n-1}w_i = 1 \).
      2. 
+
    2.  We rewrite the misclassification error as
      3.  
+
    
+
+$$
+\mathrm{err}=\frac{\sum_{i=0}^{n-1}w_iI(y_i\ne G(\boldsymbol{X}_{i*})}{\sum_{i=0}^{n-1}w_i},
+$$
+
+
+
      
+
    1.  Then we start looping over all attempts at classifying, namely we start an iterative process for \( m=1:M \), where \( M \) is the final number of classifications. Our given classifier could for example be a plain decision tree.
+
+
        
+ 
      1.  Fit thus a given classifier to the training using the weights \( w_i \).
        2. 
+ 
      2.  Compute then \( \mathrm{err} \) and figure out which events are classified properly and which are classified wrongly.
        3. 
+ 
      3.  Define a quantity $\alpha_{m} = \log{(1-\mathrm{err})/\mathrm{err}}
        4. 
+ 
      4.  Set the new weights to $w_i = w_i\times \exp{(\alpha_m I(y_i\ne G(\boldsymbol{X}_{i*})}.
        5. 
+
      
+
+
    2.  Compute the new classifier $G(\boldsymbol{X})= \sum_{i=0}^{n-1}\alpha_m I(y_i\ne G(\boldsymbol{X}_{i*}).
      3. 
+
    
+
+For the iterations with \( m \le 2 \) the weights are modified
+individually at each steps. The obersvations which were misclassified
+at iteration \( m-1 \) have a weight which is larger than those which were
+classified properly. As this proceeds, the observations which were
+difficult to classifiy correctly are given a larger influence. Each
+new classificatio step \( m \) is then forced to concentrate on those
+observations that are missed in the previous iterations.
 
 
    
 
    
@@ -254,6 +291,7 @@ MathJax.Hub.Config({
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs048.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs048.html
index af9e80be2..230d60153 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs048.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs048.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,46 +232,8 @@ MathJax.Hub.Config({
 
 
 
-
    AdaBoost Examples 
    
+
    Figure to Illustrate the Iterative Classification Process 
    
 
-
    
-Using Scikit-Learn it is easy to appply the adaptive boosting algorithm, as done here.
-
-
    
-
-
-
    from sklearn.ensemble import AdaBoostClassifier
-
-ada_clf = AdaBoostClassifier(
-    DecisionTreeClassifier(max_depth=1), n_estimators=200,
-    algorithm="SAMME.R", learning_rate=0.5, random_state=42)
-ada_clf.fit(X_train, y_train)
-
-plot_decision_boundary(ada_clf, X, y)
-
-m = len(X_train)
-
-plt.figure(figsize=(11, 4))
-for subplot, learning_rate in ((121, 1), (122, 0.5)):
-    sample_weights = np.ones(m)
-    plt.subplot(subplot)
-    for i in range(5):
-        svm_clf = SVC(kernel="rbf", C=0.05, gamma="auto", random_state=42)
-        svm_clf.fit(X_train, y_train, sample_weight=sample_weights)
-        y_pred = svm_clf.predict(X_train)
-        sample_weights[y_pred != y_train] *= (1 + learning_rate)
-        plot_decision_boundary(svm_clf, X, y, alpha=0.2)
-        plt.title("learning_rate = {}".format(learning_rate), fontsize=16)
-    if subplot == 121:
-        plt.text(-0.7, -0.65, "1", fontsize=14)
-        plt.text(-0.6, -0.10, "2", fontsize=14)
-        plt.text(-0.5,  0.10, "3", fontsize=14)
-        plt.text(-0.4,  0.55, "4", fontsize=14)
-        plt.text(-0.3,  0.90, "5", fontsize=14)
-
-save_fig("boosting_plot")
-plt.show()
-
    
 
    
 
    
 
@@ -291,6 +255,7 @@ plt.show()
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs049.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs049.html
index dd5ac1ce0..2e1feab4d 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs049.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs049.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,18 +232,46 @@ MathJax.Hub.Config({
 
 
 
-
    Gradient boosting: Basics 
    
+
    AdaBoost Examples 
    
 
 
    
-Gradient boosting is again a similar technique to Adapative boosting,
-it combines so-called weak classifiers or regressors into a strong
-method via a series of iterations.
+Using Scikit-Learn it is easy to appply the adaptive boosting algorithm, as done here.
 
 
    
-In order to understand the method, let us illustrate its basics by
-bringing back the essential steps in linear regression, where our cost
-function was the least squares function.
 
+
+
    from sklearn.ensemble import AdaBoostClassifier
+
+ada_clf = AdaBoostClassifier(
+    DecisionTreeClassifier(max_depth=1), n_estimators=200,
+    algorithm="SAMME.R", learning_rate=0.5, random_state=42)
+ada_clf.fit(X_train, y_train)
+
+plot_decision_boundary(ada_clf, X, y)
+
+m = len(X_train)
+
+plt.figure(figsize=(11, 4))
+for subplot, learning_rate in ((121, 1), (122, 0.5)):
+    sample_weights = np.ones(m)
+    plt.subplot(subplot)
+    for i in range(5):
+        svm_clf = SVC(kernel="rbf", C=0.05, gamma="auto", random_state=42)
+        svm_clf.fit(X_train, y_train, sample_weight=sample_weights)
+        y_pred = svm_clf.predict(X_train)
+        sample_weights[y_pred != y_train] *= (1 + learning_rate)
+        plot_decision_boundary(svm_clf, X, y, alpha=0.2)
+        plt.title("learning_rate = {}".format(learning_rate), fontsize=16)
+    if subplot == 121:
+        plt.text(-0.7, -0.65, "1", fontsize=14)
+        plt.text(-0.6, -0.10, "2", fontsize=14)
+        plt.text(-0.5,  0.10, "3", fontsize=14)
+        plt.text(-0.4,  0.55, "4", fontsize=14)
+        plt.text(-0.3,  0.90, "5", fontsize=14)
+
+save_fig("boosting_plot")
+plt.show()
+
    
 
    
 
    
 
@@ -262,6 +292,7 @@ function was the least squares function.
   
  • 52
    • 
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs050.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs050.html
index ad0af564b..9d656303a 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs050.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs050.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,30 +232,19 @@ MathJax.Hub.Config({
 
 
 
-
    Gradient Boosting, algorithm 
    
+
    Gradient boosting: Basics 
    
 
 
    
-Suppose we have a cost function \( C(f)=\sum_{i=0}^{n-1}L(y_i, f(x_i)) \) where \( y_i \) is our target and \( f(x_i) \) the function which is meant to model \( y_i \). The above cost function could be our standard least squares  function
-$$
-C(\boldsymbol{y},\boldsymbol{f})=\frac{1}{n}\sum_{i=0}^{n-1}(y_i-f(x_i))^2.
-$$
+Gradient boosting is again a similar technique to Adapative boosting,
+it combines so-called weak classifiers or regressors into a strong
+method via a series of iterations.
 
 
    
-The way we proceed in an iterative fashion is to
-
-
      
-
    1.  Initialize our estimate by \( f_0(x)=0 \).
      2. 
-
    2.  For \( m=1:M \), we
-
-
        
- 
      1.  compute the negative gradient vector \( \boldsymbol{u}_m = -\partial C(\boldsymbol{y},\boldsymbol{f})/\partial \boldsymbol{f}(x) \) at $f(x) = f_{m-1}(x);
        2. 
- 
      2.  fit the so-called base-learner to the negative gradient \( h_m(u_m,x) \);
        3. 
- 
      3.  update the estimate \( f_m(x) = f_{m-1}(x)+\nu h_m(u_m,x) \);
        4. 
-
      
-
-
    3.  The final estimate is then \( f_M(x) = \sum_{m=1}^M\nu h_m(u_m,x) \).
      4. 
-
    
+In order to understand the method, let us illustrate its basics by
+bringing back the essential steps in linear regression, where our cost
+function was the least squares function.
 
+
    
 
    
 
 
      
@@ -272,6 +263,7 @@ The way we proceed in an iterative fashion is to
   
    • 52
      • 
   
    • 53
      • 
   
    • 54
      • 
+  
    • 55
      • 
   
    • »
      • 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs051.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs051.html
index 9bfb51dfe..20712fafa 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs051.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs051.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,94 +232,30 @@ MathJax.Hub.Config({
 
 
 
-
    Gradient Boosting, Examples 
    
+
    Gradient Boosting, algorithm 
    
+
 
    
+Suppose we have a cost function \( C(f)=\sum_{i=0}^{n-1}L(y_i, f(x_i)) \) where \( y_i \) is our target and \( f(x_i) \) the function which is meant to model \( y_i \). The above cost function could be our standard least squares  function
+$$
+C(\boldsymbol{y},\boldsymbol{f})=\frac{1}{n}\sum_{i=0}^{n-1}(y_i-f(x_i))^2.
+$$
 
-
-
    np.random.seed(42)
-X = np.random.rand(100, 1) - 0.5
-y = 3*X[:, 0]**2 + 0.05 * np.random.randn(100)
-
-from sklearn.tree import DecisionTreeRegressor
-
-tree_reg1 = DecisionTreeRegressor(max_depth=2, random_state=42)
-tree_reg1.fit(X, y)
-
-y2 = y - tree_reg1.predict(X)
-tree_reg2 = DecisionTreeRegressor(max_depth=2, random_state=42)
-tree_reg2.fit(X, y2)
-
-y3 = y2 - tree_reg2.predict(X)
-tree_reg3 = DecisionTreeRegressor(max_depth=2, random_state=42)
-tree_reg3.fit(X, y3)
-
-X_new = np.array([[0.8]])
-y_pred = sum(tree.predict(X_new) for tree in (tree_reg1, tree_reg2, tree_reg3))
-
-def plot_predictions(regressors, X, y, axes, label=None, style="r-", data_style="b.", data_label=None):
-    x1 = np.linspace(axes[0], axes[1], 500)
-    y_pred = sum(regressor.predict(x1.reshape(-1, 1)) for regressor in regressors)
-    plt.plot(X[:, 0], y, data_style, label=data_label)
-    plt.plot(x1, y_pred, style, linewidth=2, label=label)
-    if label or data_label:
-        plt.legend(loc="upper center", fontsize=16)
-    plt.axis(axes)
-
-plt.figure(figsize=(11,11))
-
-plt.subplot(321)
-plot_predictions([tree_reg1], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h_1(x_1)$", style="g-", data_label="Training set")
-plt.ylabel("$y$", fontsize=16, rotation=0)
-plt.title("Residuals and tree predictions", fontsize=16)
-
-plt.subplot(322)
-plot_predictions([tree_reg1], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1)$", data_label="Training set")
-plt.ylabel("$y$", fontsize=16, rotation=0)
-plt.title("Ensemble predictions", fontsize=16)
-
-plt.subplot(323)
-plot_predictions([tree_reg2], X, y2, axes=[-0.5, 0.5, -0.5, 0.5], label="$h_2(x_1)$", style="g-", data_style="k+", data_label="Residuals")
-plt.ylabel("$y - h_1(x_1)$", fontsize=16)
-
-plt.subplot(324)
-plot_predictions([tree_reg1, tree_reg2], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1) + h_2(x_1)$")
-plt.ylabel("$y$", fontsize=16, rotation=0)
-
-plt.subplot(325)
-plot_predictions([tree_reg3], X, y3, axes=[-0.5, 0.5, -0.5, 0.5], label="$h_3(x_1)$", style="g-", data_style="k+")
-plt.ylabel("$y - h_1(x_1) - h_2(x_1)$", fontsize=16)
-plt.xlabel("$x_1$", fontsize=16)
-
-plt.subplot(326)
-plot_predictions([tree_reg1, tree_reg2, tree_reg3], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1) + h_2(x_1) + h_3(x_1)$")
-plt.xlabel("$x_1$", fontsize=16)
-plt.ylabel("$y$", fontsize=16, rotation=0)
-
-save_fig("gradient_boosting_plot")
-plt.show()
-
-from sklearn.ensemble import GradientBoostingRegressor
-
-gbrt = GradientBoostingRegressor(max_depth=2, n_estimators=3, learning_rate=1.0, random_state=42)
-gbrt.fit(X, y)
-
-gbrt_slow = GradientBoostingRegressor(max_depth=2, n_estimators=200, learning_rate=0.1, random_state=42)
-gbrt_slow.fit(X, y)
-
-plt.figure(figsize=(11,4))
-
-plt.subplot(121)
-plot_predictions([gbrt], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="Ensemble predictions")
-plt.title("learning_rate={}, n_estimators={}".format(gbrt.learning_rate, gbrt.n_estimators), fontsize=14)
-
-plt.subplot(122)
-plot_predictions([gbrt_slow], X, y, axes=[-0.5, 0.5, -0.1, 0.8])
-plt.title("learning_rate={}, n_estimators={}".format(gbrt_slow.learning_rate, gbrt_slow.n_estimators), fontsize=14)
-
-save_fig("gbrt_learning_rate_plot")
-plt.show()
-
    
 
    
+The way we proceed in an iterative fashion is to
+
+
      
+
    1.  Initialize our estimate by \( f_0(x)=0 \).
      2. 
+
    2.  For \( m=1:M \), we
+
+
        
+ 
      1.  compute the negative gradient vector \( \boldsymbol{u}_m = -\partial C(\boldsymbol{y},\boldsymbol{f})/\partial \boldsymbol{f}(x) \) at $f(x) = f_{m-1}(x);
        2. 
+ 
      2.  fit the so-called base-learner to the negative gradient \( h_m(u_m,x) \);
        3. 
+ 
      3.  update the estimate \( f_m(x) = f_{m-1}(x)+\nu h_m(u_m,x) \);
        4. 
+
      
+
+
    3.  The final estimate is then \( f_M(x) = \sum_{m=1}^M\nu h_m(u_m,x) \).
      4. 
+
    
+
 
    
 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs052.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs052.html
index 54ea53871..94de8a221 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs052.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs052.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,66 +232,92 @@ MathJax.Hub.Config({
 
 
 
-
    Gradient Boots with Early Stopping 
    
+
    Gradient Boosting, Examples 
    
 
    
 
 
-
    from sklearn.model_selection import train_test_split
-from sklearn.metrics import mean_squared_error
+
    np.random.seed(42)
+X = np.random.rand(100, 1) - 0.5
+y = 3*X[:, 0]**2 + 0.05 * np.random.randn(100)
 
-X_train, X_val, y_train, y_val = train_test_split(X, y, random_state=49)
+from sklearn.tree import DecisionTreeRegressor
 
-gbrt = GradientBoostingRegressor(max_depth=2, n_estimators=120, random_state=42)
-gbrt.fit(X_train, y_train)
+tree_reg1 = DecisionTreeRegressor(max_depth=2, random_state=42)
+tree_reg1.fit(X, y)
 
-errors = [mean_squared_error(y_val, y_pred)
-          for y_pred in gbrt.staged_predict(X_val)]
-bst_n_estimators = np.argmin(errors) + 1
+y2 = y - tree_reg1.predict(X)
+tree_reg2 = DecisionTreeRegressor(max_depth=2, random_state=42)
+tree_reg2.fit(X, y2)
 
-gbrt_best = GradientBoostingRegressor(max_depth=2,n_estimators=bst_n_estimators, random_state=42)
-gbrt_best.fit(X_train, y_train)
+y3 = y2 - tree_reg2.predict(X)
+tree_reg3 = DecisionTreeRegressor(max_depth=2, random_state=42)
+tree_reg3.fit(X, y3)
 
-min_error = np.min(errors)
-plt.figure(figsize=(11, 4))
+X_new = np.array([[0.8]])
+y_pred = sum(tree.predict(X_new) for tree in (tree_reg1, tree_reg2, tree_reg3))
 
-plt.subplot(121)
-plt.plot(errors, "b.-")
-plt.plot([bst_n_estimators, bst_n_estimators], [0, min_error], "k--")
-plt.plot([0, 120], [min_error, min_error], "k--")
-plt.plot(bst_n_estimators, min_error, "ko")
-plt.text(bst_n_estimators, min_error*1.2, "Minimum", ha="center", fontsize=14)
-plt.axis([0, 120, 0, 0.01])
-plt.xlabel("Number of trees")
-plt.title("Validation error", fontsize=14)
+def plot_predictions(regressors, X, y, axes, label=None, style="r-", data_style="b.", data_label=None):
+    x1 = np.linspace(axes[0], axes[1], 500)
+    y_pred = sum(regressor.predict(x1.reshape(-1, 1)) for regressor in regressors)
+    plt.plot(X[:, 0], y, data_style, label=data_label)
+    plt.plot(x1, y_pred, style, linewidth=2, label=label)
+    if label or data_label:
+        plt.legend(loc="upper center", fontsize=16)
+    plt.axis(axes)
 
-plt.subplot(122)
-plot_predictions([gbrt_best], X, y, axes=[-0.5, 0.5, -0.1, 0.8])
-plt.title("Best model (%d trees)" % bst_n_estimators, fontsize=14)
+plt.figure(figsize=(11,11))
 
-save_fig("early_stopping_gbrt_plot")
+plt.subplot(321)
+plot_predictions([tree_reg1], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h_1(x_1)$", style="g-", data_label="Training set")
+plt.ylabel("$y$", fontsize=16, rotation=0)
+plt.title("Residuals and tree predictions", fontsize=16)
+
+plt.subplot(322)
+plot_predictions([tree_reg1], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1)$", data_label="Training set")
+plt.ylabel("$y$", fontsize=16, rotation=0)
+plt.title("Ensemble predictions", fontsize=16)
+
+plt.subplot(323)
+plot_predictions([tree_reg2], X, y2, axes=[-0.5, 0.5, -0.5, 0.5], label="$h_2(x_1)$", style="g-", data_style="k+", data_label="Residuals")
+plt.ylabel("$y - h_1(x_1)$", fontsize=16)
+
+plt.subplot(324)
+plot_predictions([tree_reg1, tree_reg2], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1) + h_2(x_1)$")
+plt.ylabel("$y$", fontsize=16, rotation=0)
+
+plt.subplot(325)
+plot_predictions([tree_reg3], X, y3, axes=[-0.5, 0.5, -0.5, 0.5], label="$h_3(x_1)$", style="g-", data_style="k+")
+plt.ylabel("$y - h_1(x_1) - h_2(x_1)$", fontsize=16)
+plt.xlabel("$x_1$", fontsize=16)
+
+plt.subplot(326)
+plot_predictions([tree_reg1, tree_reg2, tree_reg3], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="$h(x_1) = h_1(x_1) + h_2(x_1) + h_3(x_1)$")
+plt.xlabel("$x_1$", fontsize=16)
+plt.ylabel("$y$", fontsize=16, rotation=0)
+
+save_fig("gradient_boosting_plot")
 plt.show()
 
+from sklearn.ensemble import GradientBoostingRegressor
 
-gbrt = GradientBoostingRegressor(max_depth=2, warm_start=True, random_state=42)
+gbrt = GradientBoostingRegressor(max_depth=2, n_estimators=3, learning_rate=1.0, random_state=42)
+gbrt.fit(X, y)
 
-min_val_error = float("inf")
-error_going_up = 0
-for n_estimators in range(1, 120):
-    gbrt.n_estimators = n_estimators
-    gbrt.fit(X_train, y_train)
-    y_pred = gbrt.predict(X_val)
-    val_error = mean_squared_error(y_val, y_pred)
-    if val_error < min_val_error:
-        min_val_error = val_error
-        error_going_up = 0
-    else:
-        error_going_up += 1
-        if error_going_up == 5:
-            break  # early stopping
+gbrt_slow = GradientBoostingRegressor(max_depth=2, n_estimators=200, learning_rate=0.1, random_state=42)
+gbrt_slow.fit(X, y)
 
+plt.figure(figsize=(11,4))
 
-print(gbrt.n_estimators)
-print("Minimum validation MSE:", min_val_error)
+plt.subplot(121)
+plot_predictions([gbrt], X, y, axes=[-0.5, 0.5, -0.1, 0.8], label="Ensemble predictions")
+plt.title("learning_rate={}, n_estimators={}".format(gbrt.learning_rate, gbrt.n_estimators), fontsize=14)
+
+plt.subplot(122)
+plot_predictions([gbrt_slow], X, y, axes=[-0.5, 0.5, -0.1, 0.8])
+plt.title("learning_rate={}, n_estimators={}".format(gbrt_slow.learning_rate, gbrt_slow.n_estimators), fontsize=14)
+
+save_fig("gbrt_learning_rate_plot")
+plt.show()
 
    
 
    
 
    
@@ -308,6 +336,7 @@ error_going_up = 52
   
  • 53
    • 
   
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs053.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs053.html
index a01826cc0..5f7afcc1d 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs053.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs053.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -230,16 +232,68 @@ MathJax.Hub.Config({
 
 
 
-
    XGBoost: Extreme Gradient Boosting 
    
-
+
    Gradient Boots with Early Stopping 
    
 
    
-XGBoost or Extreme Gradient
-Boosting, is an optimized distributed gradient boosting library
-designed to be highly efficient, flexible and portable. It implements
-machine learning algorithms under the Gradient Boosting
-framework. XGBoost provides a parallel tree boosting that solve many
-data science problems in a fast and accurate way
 
+
+
    from sklearn.model_selection import train_test_split
+from sklearn.metrics import mean_squared_error
+
+X_train, X_val, y_train, y_val = train_test_split(X, y, random_state=49)
+
+gbrt = GradientBoostingRegressor(max_depth=2, n_estimators=120, random_state=42)
+gbrt.fit(X_train, y_train)
+
+errors = [mean_squared_error(y_val, y_pred)
+          for y_pred in gbrt.staged_predict(X_val)]
+bst_n_estimators = np.argmin(errors) + 1
+
+gbrt_best = GradientBoostingRegressor(max_depth=2,n_estimators=bst_n_estimators, random_state=42)
+gbrt_best.fit(X_train, y_train)
+
+min_error = np.min(errors)
+plt.figure(figsize=(11, 4))
+
+plt.subplot(121)
+plt.plot(errors, "b.-")
+plt.plot([bst_n_estimators, bst_n_estimators], [0, min_error], "k--")
+plt.plot([0, 120], [min_error, min_error], "k--")
+plt.plot(bst_n_estimators, min_error, "ko")
+plt.text(bst_n_estimators, min_error*1.2, "Minimum", ha="center", fontsize=14)
+plt.axis([0, 120, 0, 0.01])
+plt.xlabel("Number of trees")
+plt.title("Validation error", fontsize=14)
+
+plt.subplot(122)
+plot_predictions([gbrt_best], X, y, axes=[-0.5, 0.5, -0.1, 0.8])
+plt.title("Best model (%d trees)" % bst_n_estimators, fontsize=14)
+
+save_fig("early_stopping_gbrt_plot")
+plt.show()
+
+
+gbrt = GradientBoostingRegressor(max_depth=2, warm_start=True, random_state=42)
+
+min_val_error = float("inf")
+error_going_up = 0
+for n_estimators in range(1, 120):
+    gbrt.n_estimators = n_estimators
+    gbrt.fit(X_train, y_train)
+    y_pred = gbrt.predict(X_val)
+    val_error = mean_squared_error(y_val, y_pred)
+    if val_error < min_val_error:
+        min_val_error = val_error
+        error_going_up = 0
+    else:
+        error_going_up += 1
+        if error_going_up == 5:
+            break  # early stopping
+
+
+print(gbrt.n_estimators)
+print("Minimum validation MSE:", min_val_error)
+
    
+
    
 
    
 
 
      
@@ -255,6 +309,8 @@ data science problems in a fast and accurate way
   
    • 52
      • 
   
    • 53
      • 
   
    • 54
      • 
+  
    • 55
      • 
+  
    • »
      • 
 
    
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
index d44f3e9dc..532859bcb 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
@@ -89,42 +89,43 @@ Automatically generated HTML file from DocOnce source
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -194,27 +195,28 @@ MathJax.Hub.Config({
       
  • Pros and cons of trees, pros
    • 
       
  • Disadvantages
    • 
       
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
    • 
-      
  • Bagging
    • 
-      
  • More bagging
    • 
-      
  • Simple Voting Example, head or tail
    • 
-      
  • Using the Voting Classifier
    • 
-      
  • Please, not the moons again! Voting and Bagging
    • 
-      
  • Now Bagging
    • 
-      
  • Random forests
    • 
-      
  • Random Forest Algorithm
    • 
-      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
-      
  • Compare  Bagging on Trees with Random Forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting, a Bird'e Eye
    • 
-      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
-      
  • Basic Steps of AdaBoost
    • 
-      
  • Figure to Illustrate the Iterative Classification Process
    • 
-      
  • AdaBoost Examples
    • 
-      
  • Gradient boosting: Basics
    • 
-      
  • Gradient Boosting, algorithm
    • 
-      
  • Gradient Boosting, Examples
    • 
-      
  • Gradient Boots with Early Stopping
    • 
-      
  • XGBoost: Extreme Gradient Boosting
    • 
+      
  • An Overview of Ensemble Methods
    • 
+      
  • Bagging
    • 
+      
  • More bagging
    • 
+      
  • Simple Voting Example, head or tail
    • 
+      
  • Using the Voting Classifier
    • 
+      
  • Please, not the moons again! Voting and Bagging
    • 
+      
  • Now Bagging
    • 
+      
  • Random forests
    • 
+      
  • Random Forest Algorithm
    • 
+      
  • Random Forests Compared with other Methods on the Cancer Data
    • 
+      
  • Compare  Bagging on Trees with Random Forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • Basic Steps of AdaBoost
    • 
+      
  • Figure to Illustrate the Iterative Classification Process
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basics
    • 
+      
  • Gradient Boosting, algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -249,7 +251,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 5, 2019
     
+
    Nov 6, 2019
     
 
    
 
    
 
@@ -273,7 +275,7 @@ MathJax.Hub.Config({
   
  • 9
    • 
   
  • 10
    • 
   
  • ...
    • 
-  
  • 54
    • 
+  
  • 55
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html b/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html
index 31bcc4388..5c05bc76e 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html
@@ -148,7 +148,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
     
    
-
    Nov 5, 2019
     
+
    Nov 6, 2019
     
 
    
 
    
 
@@ -1464,7 +1464,15 @@ We discuss these methods here.
 
 
 
    
-
    Bagging 
    
+
    An Overview of Ensemble Methods  
    
+
+
    
+



    
+
    
+
+
+
    
+
    Bagging 
    
 
 
    
 The plain decision trees suffer from high
@@ -1483,7 +1491,7 @@ learning method.
 
 
 
    
-
    More bagging 
    
+
    More bagging 
    
 
 
    
 Bagging typically results in improved accuracy
@@ -1512,7 +1520,7 @@ predictor, averaged over all \( B \) trees.
 
 
 
    
-
    Simple Voting Example, head or tail 
    
+
    Simple Voting Example, head or tail 
    
 
    
 
 
@@ -1533,7 +1541,7 @@ plt.show()
 
 
 
    
-
    Using the Voting Classifier 
    
+
    Using the Voting Classifier 
    
 
    
 
 
@@ -1585,7 +1593,7 @@ voting_clf.fit(X_train, y_train)
 
 
 
    
-
    Please, not the moons again! Voting and Bagging 
    
+
    Please, not the moons again! Voting and Bagging 
    
 
    
 
 
@@ -1644,7 +1652,7 @@ voting_clf.fit(X_train, y_train)
 
 
 
    
-
    Now Bagging  
    
+
    Now Bagging  
    
 
 
    
 
@@ -1706,7 +1714,7 @@ plt.show()
 
 
 
    
-
    Random forests 
    
+
    Random forests 
    
 
 
    
 Random forests provide an improvement over bagged trees by way of a
@@ -1752,7 +1760,7 @@ this setting.
 
 
 
    
-
    Random Forest Algorithm 
    
+
    Random Forest Algorithm 
    
 The algorithm described here can be applied to both classification and regression problems.
 
 
    
@@ -1783,7 +1791,7 @@ We will grow of forest of say \( M \) trees.
 
 
 
    
-
    Random Forests Compared with other Methods on the Cancer Data 
    
+
    Random Forests Compared with other Methods on the Cancer Data 
    
 
    
 
 
@@ -1858,7 +1866,7 @@ plt.show()
 
 
 
    
-
    Compare  Bagging on Trees with Random Forests  
    
+
    Compare  Bagging on Trees with Random Forests  
    
 
    
 
 
@@ -1881,7 +1889,7 @@ np.sum(y_pred == y_pred_rf) / len(y_pred)
 
 
 
    
-
    Feature Importance 
    
+
    Feature Importance 
    
 
 
    
 Example will be added here.
@@ -1889,7 +1897,7 @@ Example will be added here.
 
 
 
    
-
    Boosting, a Bird'e Eye 
    
+
    Boosting, a Bird'e Eye 
    
 
 
    
 The basic idea is to combine weak classifiers in order to create a good
@@ -1906,7 +1914,7 @@ them with a factor.
 
 
 
    
-
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
+
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
 
 
    
 The algorithm here is rather straightforward. Assume that our weak
@@ -1930,7 +1938,7 @@ where the function \( I() \) is one if we misclassify and zero if we classify co
 
 
 
    
-
    Basic Steps of AdaBoost 
    
+
    Basic Steps of AdaBoost 
    
 
 
    
 With the above definitions we are now ready to set up the algorithm for AdaBoost.
@@ -1971,12 +1979,12 @@ observations that are missed in the previous iterations.
 
 
 
    
-
    Figure to Illustrate the Iterative Classification Process 
    
+
    Figure to Illustrate the Iterative Classification Process 
    
 
    
 
 
 
    
-
    AdaBoost Examples 
    
+
    AdaBoost Examples 
    
 
 
    
 Using Scikit-Learn it is easy to appply the adaptive boosting algorithm, as done here.
@@ -2020,7 +2028,7 @@ plt.show()
 
 
 
    
-
    Gradient boosting: Basics 
    
+
    Gradient boosting: Basics 
    
 
 
    
 Gradient boosting is again a similar technique to Adapative boosting,
@@ -2035,7 +2043,7 @@ function was the least squares function.
 
 
 
    
-
    Gradient Boosting, algorithm 
    
+
    Gradient Boosting, algorithm 
    
 
 
    
 Suppose we have a cost function \( C(f)=\sum_{i=0}^{n-1}L(y_i, f(x_i)) \) where \( y_i \) is our target and \( f(x_i) \) the function which is meant to model \( y_i \). The above cost function could be our standard least squares  function
@@ -2063,7 +2071,7 @@ The way we proceed in an iterative fashion is to
 
 
 
    
-
    Gradient Boosting, Examples 
    
+
    Gradient Boosting, Examples 
    
 
    
 
 
@@ -2154,7 +2162,7 @@ plt.show()
 
 
 
    
-
    Gradient Boots with Early Stopping 
    
+
    Gradient Boots with Early Stopping 
    
 
    
 
 
@@ -2219,7 +2227,7 @@ error_going_up = 0
 
 
 
    
-
    XGBoost: Extreme Gradient Boosting 
    
+
    XGBoost: Extreme Gradient Boosting 
    
 
 
    
 XGBoost or Extreme Gradient
@@ -2227,7 +2235,15 @@ Boosting, is an optimized distributed gradient boosting library
 designed to be highly efficient, flexible and portable. It implements
 machine learning algorithms under the Gradient Boosting
 framework. XGBoost provides a parallel tree boosting that solve many
-data science problems in a fast and accurate way
+data science problems in a fast and accurate way. See the article by Chen and Guestrin.
+
+
    
+The authors design and build a highly scalable end-to-end tree
+boosting system. It has  a theoretically justified weighted quantile
+sketch for efficient proposal calculation. It introduces a novel sparsity-aware algorithm for parallel tree learning and an effective cache-aware block structure for out-of-core tree learning.
+
+
    
+It is now the algorithm which wins essentially all ML competitions!!!
 
    
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
index cfe624d34..e09f2ea36 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
@@ -109,42 +109,43 @@ div { text-align: justify; text-justify: inter-word; }
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -186,7 +187,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 5, 2019
     
+
    Nov 6, 2019
     
 
    
 
    
 









    
@@ -1461,7 +1462,15 @@ We discuss these methods here.
 
    
 









    
 
-
    Bagging 
    
+
    An Overview of Ensemble Methods  
    
+
+
    
+



    
+
+
    
+









    
+
+
    Bagging 
    
 
 
    
 The plain decision trees suffer from high
@@ -1480,7 +1489,7 @@ learning method.
 
    
 









    
 
-
    More bagging 
    
+
    More bagging 
    
 
 
    
 Bagging typically results in improved accuracy
@@ -1509,7 +1518,7 @@ predictor, averaged over all \( B \) trees.
 
    
 









    
 
-
    Simple Voting Example, head or tail 
    
+
    Simple Voting Example, head or tail 
    
 
    
 
 
@@ -1529,7 +1538,7 @@ plt.show()
 
    
 









    
 
-
    Using the Voting Classifier 
    
+
    Using the Voting Classifier 
    
 
    
 
 
@@ -1580,7 +1589,7 @@ voting_clf.fit(X_train, y_train)
 
    
 









    
 
-
    Please, not the moons again! Voting and Bagging 
    
+
    Please, not the moons again! Voting and Bagging 
    
 
    
 
 
@@ -1638,7 +1647,7 @@ voting_clf.fit(X_train, y_train)
 
    
 









    
 
-
    Now Bagging  
    
+
    Now Bagging  
    
 
 
    
 
@@ -1699,7 +1708,7 @@ plt.show()
 
    
 









    
 
-
    Random forests 
    
+
    Random forests 
    
 
 
    
 Random forests provide an improvement over bagged trees by way of a
@@ -1743,7 +1752,7 @@ this setting.
 
    
 









    
 
-
    Random Forest Algorithm 
    
+
    Random Forest Algorithm 
    
 The algorithm described here can be applied to both classification and regression problems.
 
 
    
@@ -1769,7 +1778,7 @@ We will grow of forest of say \( M \) trees.
 
 









    
 
-
    Random Forests Compared with other Methods on the Cancer Data 
    
+
    Random Forests Compared with other Methods on the Cancer Data 
    
 
    
 
 
@@ -1843,7 +1852,7 @@ plt.show()
 
    
 









    
 
-
    Compare  Bagging on Trees with Random Forests  
    
+
    Compare  Bagging on Trees with Random Forests  
    
 
    
 
 
@@ -1865,7 +1874,7 @@ np.sum(y_pred == y_pred_rf) / len(y_pred)
 
    
 









    
 
-
    Feature Importance 
    
+
    Feature Importance 
    
 
 
    
 Example will be added here.
@@ -1873,7 +1882,7 @@ Example will be added here.
 
    
 









    
 
-
    Boosting, a Bird'e Eye 
    
+
    Boosting, a Bird'e Eye 
    
 
 
    
 The basic idea is to combine weak classifiers in order to create a good
@@ -1890,7 +1899,7 @@ them with a factor.
 
    
 









    
 
-
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
+
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
 
 
    
 The algorithm here is rather straightforward. Assume that our weak
@@ -1912,7 +1921,7 @@ where the function \( I() \) is one if we misclassify and zero if we classify co
 
    
 









    
 
-
    Basic Steps of AdaBoost 
    
+
    Basic Steps of AdaBoost 
    
 
 
    
 With the above definitions we are now ready to set up the algorithm for AdaBoost.
@@ -1952,12 +1961,12 @@ observations that are missed in the previous iterations.
 
    
 









    
 
-
    Figure to Illustrate the Iterative Classification Process 
    
+
    Figure to Illustrate the Iterative Classification Process 
    
 
 
    
 









    
 
-
    AdaBoost Examples 
    
+
    AdaBoost Examples 
    
 
 
    
 Using Scikit-Learn it is easy to appply the adaptive boosting algorithm, as done here.
@@ -2000,7 +2009,7 @@ plt.show()
 
    
 









    
 
-
    Gradient boosting: Basics 
    
+
    Gradient boosting: Basics 
    
 
 
    
 Gradient boosting is again a similar technique to Adapative boosting,
@@ -2015,7 +2024,7 @@ function was the least squares function.
 
    
 









    
 
-
    Gradient Boosting, algorithm 
    
+
    Gradient Boosting, algorithm 
    
 
 
    
 Suppose we have a cost function \( C(f)=\sum_{i=0}^{n-1}L(y_i, f(x_i)) \) where \( y_i \) is our target and \( f(x_i) \) the function which is meant to model \( y_i \). The above cost function could be our standard least squares  function
@@ -2041,7 +2050,7 @@ The way we proceed in an iterative fashion is to
 
 









    
 
-
    Gradient Boosting, Examples 
    
+
    Gradient Boosting, Examples 
    
 
    
 
 
@@ -2131,7 +2140,7 @@ plt.show()
 
    
 









    
 
-
    Gradient Boots with Early Stopping 
    
+
    Gradient Boots with Early Stopping 
    
 
    
 
 
@@ -2195,7 +2204,7 @@ error_going_up = 0
 
    
 









    
 
-
    XGBoost: Extreme Gradient Boosting 
    
+
    XGBoost: Extreme Gradient Boosting 
    
 
 
    
 XGBoost or Extreme Gradient
@@ -2203,7 +2212,15 @@ Boosting, is an optimized distributed gradient boosting library
 designed to be highly efficient, flexible and portable. It implements
 machine learning algorithms under the Gradient Boosting
 framework. XGBoost provides a parallel tree boosting that solve many
-data science problems in a fast and accurate way
+data science problems in a fast and accurate way. See the article by Chen and Guestrin.
+
+
    
+The authors design and build a highly scalable end-to-end tree
+boosting system. It has  a theoretically justified weighted quantile
+sketch for efficient proposal calculation. It introduces a novel sparsity-aware algorithm for parallel tree learning and an effective cache-aware block structure for out-of-core tree learning.
+
+
    
+It is now the algorithm which wins essentially all ML competitions!!!
 
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees.html b/doc/pub/DecisionTrees/html/DecisionTrees.html
index 9f4a57d59..2b45b7595 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees.html
@@ -114,42 +114,43 @@ div { text-align: justify; text-justify: inter-word; }
                2,
                None,
                '___sec31'),
-              ('Bagging', 2, None, '___sec32'),
-              ('More bagging', 2, None, '___sec33'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec34'),
-              ('Using the Voting Classifier', 2, None, '___sec35'),
+              ('An Overview of Ensemble Methods', 2, None, '___sec32'),
+              ('Bagging', 2, None, '___sec33'),
+              ('More bagging', 2, None, '___sec34'),
+              ('Simple Voting Example, head or tail', 2, None, '___sec35'),
+              ('Using the Voting Classifier', 2, None, '___sec36'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec36'),
-              ('Now Bagging', 2, None, '___sec37'),
-              ('Random forests', 2, None, '___sec38'),
-              ('Random Forest Algorithm', 2, None, '___sec39'),
+               '___sec37'),
+              ('Now Bagging', 2, None, '___sec38'),
+              ('Random forests', 2, None, '___sec39'),
+              ('Random Forest Algorithm', 2, None, '___sec40'),
               ('Random Forests Compared with other Methods on the Cancer Data',
                2,
                None,
-               '___sec40'),
+               '___sec41'),
               ('Compare  Bagging on Trees with Random Forests',
                2,
                None,
-               '___sec41'),
-              ('Feature Importance', 2, None, '___sec42'),
-              ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
+               '___sec42'),
+              ('Feature Importance', 2, None, '___sec43'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec44'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
                2,
                None,
-               '___sec44'),
-              ('Basic Steps of AdaBoost', 2, None, '___sec45'),
+               '___sec45'),
+              ('Basic Steps of AdaBoost', 2, None, '___sec46'),
               ('Figure to Illustrate the Iterative Classification Process',
                2,
                None,
-               '___sec46'),
-              ('AdaBoost Examples', 2, None, '___sec47'),
-              ('Gradient boosting: Basics', 2, None, '___sec48'),
-              ('Gradient Boosting, algorithm', 2, None, '___sec49'),
-              ('Gradient Boosting, Examples', 2, None, '___sec50'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec51'),
-              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec52')]}
+               '___sec47'),
+              ('AdaBoost Examples', 2, None, '___sec48'),
+              ('Gradient boosting: Basics', 2, None, '___sec49'),
+              ('Gradient Boosting, algorithm', 2, None, '___sec50'),
+              ('Gradient Boosting, Examples', 2, None, '___sec51'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec52'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec53')]}
 end of tocinfo -->
 
 
@@ -191,7 +192,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 5, 2019
     
+
    Nov 6, 2019
     
 
    
 
    
 









    
@@ -1466,7 +1467,15 @@ We discuss these methods here.
 
    
 









    
 
-
    Bagging 
    
+
    An Overview of Ensemble Methods  
    
+
+
    
+



    
+
+
    
+









    
+
+
    Bagging 
    
 
 
    
 The plain decision trees suffer from high
@@ -1485,7 +1494,7 @@ learning method.
 
    
 









    
 
-
    More bagging 
    
+
    More bagging 
    
 
 
    
 Bagging typically results in improved accuracy
@@ -1514,7 +1523,7 @@ predictor, averaged over all \( B \) trees.
 
    
 









    
 
-
    Simple Voting Example, head or tail 
    
+
    Simple Voting Example, head or tail 
    
 
    
 
 
@@ -1534,7 +1543,7 @@ plt.show()
 
    
 









    
 
-
    Using the Voting Classifier 
    
+
    Using the Voting Classifier 
    
 
    
 
 
@@ -1585,7 +1594,7 @@ voting_clf.fit(X_train, y_train)
 
    
 









    
 
-
    Please, not the moons again! Voting and Bagging 
    
+
    Please, not the moons again! Voting and Bagging 
    
 
    
 
 
@@ -1643,7 +1652,7 @@ voting_clf.fit(X_train, y_train)
 
    
 









    
 
-
    Now Bagging  
    
+
    Now Bagging  
    
 
 
    
 
@@ -1704,7 +1713,7 @@ plt.show()
 
    
 









    
 
-
    Random forests 
    
+
    Random forests 
    
 
 
    
 Random forests provide an improvement over bagged trees by way of a
@@ -1748,7 +1757,7 @@ this setting.
 
    
 









    
 
-
    Random Forest Algorithm 
    
+
    Random Forest Algorithm 
    
 The algorithm described here can be applied to both classification and regression problems.
 
 
    
@@ -1774,7 +1783,7 @@ We will grow of forest of say \( M \) trees.
 
 









    
 
-
    Random Forests Compared with other Methods on the Cancer Data 
    
+
    Random Forests Compared with other Methods on the Cancer Data 
    
 
    
 
 
@@ -1848,7 +1857,7 @@ plt.show()
 
    
 









    
 
-
    Compare  Bagging on Trees with Random Forests  
    
+
    Compare  Bagging on Trees with Random Forests  
    
 
    
 
 
@@ -1870,7 +1879,7 @@ np.sum(y_pred =
 
    
 









    
 
-
    Feature Importance 
    
+
    Feature Importance 
    
 
 
    
 Example will be added here.
@@ -1878,7 +1887,7 @@ Example will be added here.
 
    
 









    
 
-
    Boosting, a Bird'e Eye 
    
+
    Boosting, a Bird'e Eye 
    
 
 
    
 The basic idea is to combine weak classifiers in order to create a good
@@ -1895,7 +1904,7 @@ them with a factor.
 
    
 









    
 
-
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
+
    Adaptive boosting: AdaBoost, Basic Algorithm 
    
 
 
    
 The algorithm here is rather straightforward. Assume that our weak
@@ -1917,7 +1926,7 @@ where the function \( I() \) is one if we misclassify and zero if we classify co
 
    
 









    
 
-
    Basic Steps of AdaBoost 
    
+
    Basic Steps of AdaBoost 
    
 
 
    
 With the above definitions we are now ready to set up the algorithm for AdaBoost.
@@ -1957,12 +1966,12 @@ observations that are missed in the previous iterations.
 
    
 









    
 
-
    Figure to Illustrate the Iterative Classification Process 
    
+
    Figure to Illustrate the Iterative Classification Process 
    
 
 
    
 









    
 
-
    AdaBoost Examples 
    
+
    AdaBoost Examples 
    
 
 
    
 Using Scikit-Learn it is easy to appply the adaptive boosting algorithm, as done here.
@@ -2005,7 +2014,7 @@ plt.show()
 
    
 









    
 
-
    Gradient boosting: Basics 
    
+
    Gradient boosting: Basics 
    
 
 
    
 Gradient boosting is again a similar technique to Adapative boosting,
@@ -2020,7 +2029,7 @@ function was the least squares function.
 
    
 









    
 
-
    Gradient Boosting, algorithm 
    
+
    Gradient Boosting, algorithm 
    
 
 
    
 Suppose we have a cost function \( C(f)=\sum_{i=0}^{n-1}L(y_i, f(x_i)) \) where \( y_i \) is our target and \( f(x_i) \) the function which is meant to model \( y_i \). The above cost function could be our standard least squares  function
@@ -2046,7 +2055,7 @@ The way we proceed in an iterative fashion is to
 
 









    
 
-
    Gradient Boosting, Examples 
    
+
    Gradient Boosting, Examples 
    
 
    
 
 
@@ -2136,7 +2145,7 @@ plt.show()
 
    
 









    
 
-
    Gradient Boots with Early Stopping 
    
+
    Gradient Boots with Early Stopping 
    
 
    
 
 
@@ -2200,7 +2209,7 @@ error_going_up = XGBoost: Extreme Gradient Boosting 
+
    XGBoost: Extreme Gradient Boosting 
    
 
 
    
 XGBoost or Extreme Gradient
@@ -2208,7 +2217,15 @@ Boosting, is an optimized distributed gradient boosting library
 designed to be highly efficient, flexible and portable. It implements
 machine learning algorithms under the Gradient Boosting
 framework. XGBoost provides a parallel tree boosting that solve many
-data science problems in a fast and accurate way
+data science problems in a fast and accurate way. See the article by Chen and Guestrin.
+
+
    
+The authors design and build a highly scalable end-to-end tree
+boosting system. It has  a theoretically justified weighted quantile
+sketch for efficient proposal calculation. It introduces a novel sparsity-aware algorithm for parallel tree learning and an effective cache-aware block structure for out-of-core tree learning.
+
+
    
+It is now the algorithm which wins essentially all ML competitions!!!
 
 
 
diff --git a/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb b/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb
index 3c8225c59..666e15bc7 100644
--- a/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb
+++ b/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb
@@ -10,7 +10,7 @@
     "  \n",
     "**Morten Hjorth-Jensen**, Department of Physics, University of Oslo and Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University\n",
     "\n",
-    "Date: **Nov 5, 2019**\n",
+    "Date: **Nov 6, 2019**\n",
     "\n",
     "Copyright 1999-2019, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license\n",
     "\n",
@@ -101,40 +101,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "2nd degree coefficients:\n",
-      "zero power:  -0.28499417022929907\n",
-      "first power:  0.05490242306549711\n",
-      "second power:  -0.0002577121418274489\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEWCAYAAABxMXBSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzsnXd4FNXawH9nN73QQkKHxCgdEiGAiKKIILYoVtCrgAXkerkIlovyGSSKehULiuWqWEFUEDAUFZRIUTSAEhEEAQkdEkJLb3u+P2Zms0k2ySbZluT8nmee3Z055Z3ZmfPO+55z3iOklCgUCoVC4SgmTwugUCgUivqFUhwKhUKhqBFKcSgUCoWiRijFoVAoFIoaoRSHQqFQKGqEUhwKhUKhqBFKcSi8GiHE10KIMQ6kyxZCnOcOmRSOIYR4UgjxtqflUDgfoeZxKOqKECINaAUUAyXATuBj4B0ppcWDotUJIUS2zc8goADt/AAmSCkXuLj+w0CYXmc2sAqYJKXMcWW9CkV1KItD4Syul1KGAp2A54H/APM8K1LdkFKGGBtwEO0cjX0VlIYQwscFYlyt198H6A885oI6EEKYXVGuomGiFIfCqUgpz0opk4DbgTFCiJ4AQgh/IcRsIcRBIcQJIcTbQohAI58Q4gYhxDYhxDkhxD4hxAh9/w9CiPv07+cLIdYJIc4KIU4KIT63yS+FEOfr35sKIT4WQmQIIQ4IIf5PCGHSj40VQmzUZTkthNgvhLi6NucqhHhGCPG5EGKhECIL+IcQwiSEeEI/h5NCiM+EEM1t8gwSQvwshDijn+9gB6/rUWA1EGtTVoAQ4mUhxCH9mr4phAiwOf64EOK4EOKIEOJ+/RpF6sfmCyHeEEJ8I4TIAS6tqjwhRIQQYpUu9ykhxHqbep4QQhzV/7tdQojLba7PhzbpRgohduhlrBVCdLE5dlgIMVUIsV3/fxcKIfxr9o8o3IVSHAqXIKVMAQ4Dl+q7ngc6ozV85wPtgAQAIUR/NNfWo0AzYDCQZqfYp9Eaz+ZAe+D1Sqp/HWgKnAdcBtwNjLM5PgDYDbQEXgDmCSFEzc8SgJHAp3p9nwNTgGv1c2iP5mJ6DUAI0QFIAmYALYBpwBIhRFh1leh5RwB7bXa/CEQBvYELgEhgup7+OmASMATtul9hp9g7gJlAKLCpqvLQ/pu/gXCgNfB/ej09gAlAHyllE+BqNOusvPzdgE90mcKB74AkIYSvTbLbgGFo/1tf4K7qrovCQ0gp1aa2Om1ojfyVdvb/jNbwCCAHiLY5NhDYr3//H/BKJWX/ANynf/8YeAdobyedRFNIZqAQ6G5zbALwg/59LLDX5liQnrd1Tc8ReAZYW27fHuAym98dgHy0l7TpwAfl0n8P3FlJnYfRFE+WLuNqoKl+zKSX28km/aXAHptr9bTNsa56GZH67/nA+zbHqyvvWWCJ7X+o7+8CnACGAj52rs+H+veZwKfl6jsOXGJzrqNsjr8MzPX0va02+5uyOBSupB1wCu0NMwjYqrspzgDf6PtBa1z3OVDeY2hKKEV3edxjJ01LwBc4YLPvgC6LwXHji5QyV/8a4kD99jhU7ndHYLnNeW7X90eg9f+MNo7pxy8C2lZR/nVS6zsaCnRHs1RAe+v3B1Jtylqh14Nepq1s5eUsv6+68p5Hu47f6264RwGklLuBh4FEIF13MbW2U1dbbP4TqQ2aOEwl/wuQS+3/E4WLUYpD4RKEEP3QGoWNwEkgD+ghpWymb02l1ukLWgMWXV2ZUsrjUsr7pZRt0ayIN41+DRtOAkVojbRBR+BI3c6ocrHK/T4MDLM5z2ZSygAp5XG08/yg3LFgKeWL1VYi5VpgAZo7CbS3/EKgS7lr2lQ/fgzNVWbQoRrZqyxPSnlOSjlFShkJ3Aj8RwhxmX5svpRyEJqbyww8Z6euo9j8J3qfU3tc978oXIhSHAqnIoRoovvXPwPmSym362+X7wKvCCEi9HTthBBX6dnmAeOEEEP1zuV2Qoiudsq+VQhhNIan0Rq+MsN9pZQlwBfALCFEqBCiEzAVzTXjDt4GnhVCdNRljhBCxOvHPgFGCiGGCSHMemf0ECFEVRaHLa8A1wgheurn+R7wqhAiXGi0F0IM19N+AdwrhOgihAgCnqyq4OrKE0JcL4SI1vuCzqINEbYIIbrp5+CP9nKQR7n/xEaeeCHE5Xq/xqNoLrhfHDx3hRehFIfCWSwX2siiQ2i+/Jcp2yH9H7SO3Z+FEOfQOke7gLUjfRxaw3gWWEdZi8GgH/CL0OZXJAGTpZR/20k3Ca1P5W80i+dT4P26nqCDvIzmhvtevx4/ocmNlDINrTP9SSADrRP5YRx8DnWrZQGlSuBhNPdPCtp1W43WqY2UcjnwFrAerd/lRz1PQRVVVFoe2n+1Fq3P5UdgjpRyA5p76wU0S+842sCF6ZRDSrkDGKPLlIHW0R8vpSxy5NwV3oWaAKhQNAKEEL2AXwF/WY8nZSq8A2VxKBQNFH3ehJ8QogVa5/ZXSmkonIFSHApFw+VBNBfSXrShtg96VhxFQ0G5qhQKhUJRI5TFoVAoFIoa4YqgbB6nZcuWMjIy0tNiKBQKRb1h69atJ6WU4dWnbKCKIzIyki1btnhaDIVCoag3CCEOVJ9KQ7mqFAqFQlEjlOJQKBQKRY1QikOhUCgUNaJB9nEoFArPUFRUxOHDh8nPz/e0KIpKCAgIoH379vj6+lafuBKU4lAoFE7j8OHDhIaGEhkZSe3XxlK4CiklmZmZHD58mKioqFqXo1xVCoXCaeTn5xMWFqaUhpcihCAsLKzOFqFSHAqFwqkopeHdOOP/UYpDoWiAJCUlERsbS1JSkqdFUTRAlOJQKBogCQkJpKamkpCQ4GlRFA0QpTgUigZIYmIiMTExJCYmelqURkNkZCQnT56sVd4PP/yQo0eP1rmsQ4cOMWTIELp3706PHj2YM2dOreSpDjWqSqFogMTHxxMfH199QoVX8OGHH9KzZ0/atnV0FWH7+Pj48NJLL9GnTx+ysrLo27cvw4YNo3v37k6SVENZHAqFwjUI4ZqtGtLS0ujatStjx46lc+fO3HnnnXz33XcMGjSICy64gJSUFFJSUhg4cCAXXnghF198Mbt37wbglVde4Z577gFg+/bt9OzZk9zcXLv1ZGZmMnz4cHr06MF9992H7RIV8+fPp3///sTGxjJhwgRKSkoACAkJYcqUKfTo0YOhQ4eSkZHB4sWL2bJlC3feeSexsbHk5eUB8Prrr9OnTx969erFrl27HLrkbdq0oU+fPgCEhobSrVs3jhw54lDeGiGlbHBb3759pUKhcD87d+4s/QGu2aph//790mw2y99//12WlJTIPn36yHHjxkmLxSKXLVsmb7jhBnn27FlZVFQkpZRyzZo18qabbpJSSllSUiIvvfRSuWTJEtm3b1+5cePGSuuZNGmSnDlzppRSyhUrVkhAZmRkyJ07d8rrrrtOFhYWSimlnDhxovzoo4/0S4KcP3++lFLKmTNnygcffFBKKeVll10mN2/ebC27U6dO8rXXXpNSSvnGG2/Ie++9V0op5dq1a2VMTEyFbeDAgXavQ4cOHeTZs2er/p90gC3SwTZWuaoUCoVr8OAicVFRUfTq1QvA+nYvhKBXr16kpaVx9uxZxowZw549exBCUFRUBIDJZOLDDz+kd+/eTJgwgUGDBlVax/r161myZAkA1157Lc2bNwfg+++/Z+vWrfTr1w+AvLw8IiIirOXffvvtAPzjH//gpptuqrR841jfvn2t9QwZMoRt27ZVe/7Z2dncfPPNvPrqqzRp0qTa9DVFKQ6FQtHg8Pf3t343mUzW3yaTieLiYp588kmGDBnC0qVLSUtL4/LLL7em37NnDyEhIWU6q2uClJIxY8bw3HPPVZu2qjkVhsxms5ni4mIAkpOTmTJlSoW0QUFB/PTTT4AW9uXmm2/mzjvvrFIx1QXVx6FQKBodZ8+epV27doDWMW27/9///jfr168nMzOTxYsXV1rG4MGD+fTTTwH4+uuvOX36NABDhw5l8eLFpKenA3Dq1CkOHNCWurBYLNYyP/30Uy655BJA64/IysqqVm7D4ii/GUpDSsm9995Lt27dmDp1ak0uSY1wi+IQQrwvhEgXQvxhs6+FEGKNEGKP/tm8krxj9DR7hBBj3CGvQqFo2Dz22GM8/vjjXHjhhda3eYApU6bw4IMP0rlzZ+bNm8e0adOsCqA8M2bMYP369fTo0YMlS5bQsWNHALp3784zzzzD8OHD6d27N8OGDePYsWMABAcHk5KSQs+ePVm7dq11ns3YsWN54IEHynSO14Yff/yRTz75hLVr1xIbG0tsbCyrVq2qdXmVIaQb/JBCiMFANvCxlLKnvu8F4JSU8nkhxDSguZTyP+XytQC2AHGABLYCfaWUp6uqLy4uTqoVABUK9/Pnn3/SrVs3T4vhtYSEhJCdne1pMez+T0KIrVLKOEfyu8XikFKuB06V230D8JH+/SPgRjtZrwLWSClP6cpiDTDCZYIqFF5OUlISUVFRREVFkZSUZP0dERFBREQEUVFRTJ8+ndjYWKZPn14mrULhLDzZOd5KSnlM/34caGUnTTvgkM3vw/q+CgghxgPjAavJqFA0BKSUlGRr8wBmTZ/FibQT1u+A9TdAdkY2b730FvkF+RzcdZD8gnwkkoSEBDUhsJZ88MEHFWZgDxo0iDfeeKPGZXmDteEMvGJUlZRSCiHq5DOTUr4DvAOaq8opgikUXsDOUTvJ+CIDgP/y39IDf1SSoaDcJ1DQscBuUkX1jBs3jnHjxnlaDK/Ck4rjhBCijZTymBCiDWCvB+oIcLnN7/bAD26QTaHwGs78cAYAU5AJYapZSGxZIrHkWQg7HOYK0RSNFE8qjiRgDPC8/vmVnTTfAs/ajLgaDjzuHvEUCi9Bt58v2n8RfhF+NcqatS2LrRduRVqUEa5wHu4ajrsQ2AR0EUIcFkLci6Ywhgkh9gBX6r8RQsQJId4DkFKeAp4GNutbor5PoWg0WBv9WjytVgvF4jx5FAq3WBxSytGVHBpqJ+0W4D6b3+8D77tINIXC+9H1Rq1WbtOVjSxRFofCeaiZ4wqFt2NYC8ri8Gq8YT0OI2+vXr2IjY0lLs6haRk1xitGVSkUilJyi3JZ8dcKcou0cN4diztiwsTCPxZiCa2ZBvBJ86E97Tmbd5YPt31Y4fjlkZcT2SzSCVIr6oKz1uMwSE5OpmXLlk4pyx5KcSgUXsarP7/K9LXTrb9XFK4gmGAmrppIboD9tSEqo11mO+Yzn/Rz6Yz7quKQ0tjWsfw24bc6y2wPMbMWrjUHkDOqdrulpaUxYsQILrroIn766Sf69evHuHHjmDFjBunp6SxYsACAyZMnk5+fT2BgIB988AFdunThlVdeYfv27bz//vts376d0aNHk5KSQlBQUIV6MjMzGT16NEeOHGHgwIEV1uN47bXXKCwsZMCAAbz55puYzWZCQkK4//77Wb16Na1bt+azzz5j3bp11vU4AgMD2bRpE6Ctx7F8+XKKiopYtGgRXbt2deJVrBtKcSgUXkZGjjZnI65tHD3Ce+Bn0kZSje49muLA4qqyViD0WCgATfyaMCamNNRbTlEOi3cuttbV0Ni7dy+LFi3i/fffp1+/fnz66ads3LiRpKQknn32WT7++GM2bNiAj48P3333HU888QRffvklkydP5vLLL2fp0qXMmjWL//3vf3aVBsDMmTO55JJLSEhIYOXKlcybNw/Qwnl8/vnn/Pjjj/j6+vLPf/6TBQsWcPfdd5OTk0NcXByvvPIKiYmJzJw5k7lz5zJ37lxmz55dxrXUsmVLfv31V958801mz57Ne++951B0XCEEw4cPRwjBhAkTGD9+vNOvr1IcCoWXIfXe8Dt73clDFz3EetN6LFjYMG0DhaZC6yzmhIQEEhMTq5wRnpeWxy/8QtGpIm4yaSG2J0+eTHFQMdxWWpdLzqMay8CVNOb1ODZu3Ei7du1IT09n2LBhdO3alcGDB1d9wWqIUhwKhZdRIfCo3q2RdjCNQgqtEVVTU1OrDSVidI4XFRWRmJColZOWBqGV1NVAaMzrcRjh4iMiIhg5ciQpKSlOVxxqVJVC4WUYVoBAa1SMxr1jx45ERkaSmJhIYmIiMTExJCYmVl2Y/oT7+/hb80VGRtK+fXuXyV8faKjrceTk5FjLycnJYfXq1fTs2dPh6+IoyuJQKLwU69uobnHs3rMbk1/pu54jQQsNiyOiZQQXx19szXcs6xhtX27rUleVN/PYY48xZswYnnnmGa699lrr/vLrcQwZMoTBgwdbXU22zJgxg9GjR9OjRw8uvvhiu+txWCwWfH19eeONN+jUqZN1PY5nnnmGiIgIPv/8c6B0PQ7bzvHacOLECUaOHAlAcXExd9xxByNGOD+guFvW43A3aj0ORX1m0qpJzN08l9dGvMakAZNY57sOWSwZXDgYk2/NnASFJwr5qfVPmEPNXPD6Bdb9Z/LO8NC3D3Gk2xG2P7vdabKr9TiqpqGsx6EsDoXCyyhvBRghR2oa4BBA+Gt5SrJK2DV2V5lj05jG/pT98GwtBVU0WpTiUCi8DMMLYHVVGXqkFtMifJv5csHcCziXcq7M/tyzuWR9lUWTnCZ1kLRxoNbjqIhSHAqFl2J0jtdFcQC0e7Ad7R4su/7Zkb1HyPoqC2FxzSS9hoRaj6MialSVQuFlWEdVCVFmuGytghxWguH2ElIpDkXNUYpDofAyygxYqUOAw6owmbQCTVI1AYqao+4ahcJLEYg6dYxXWbYTrRdF40MpDoXCy5BImuQ2ocn9Tfgo5CMALE6Ii56UlERsbCxJSUms/WEtgOrjUNQKpTgUCi9DSkmfv/vQLrUdUYVRAKSb0+tcbkJCgjVMycuvvgyoPg5n4g3rcezevZvY2Fjr1qRJE1599dVayVQVSnEoFF6GROJTog14/CPgD55u8zSBHwbWuVzbMCVTp04FVB+Ht1BecdSWLl26WMOQbN26laCgIOtMcmeihuMqFF6I0aAPvXUo//r4X04pMz4+3hqmJDMjk+1sd6ni+EH84JJyL5eXV3lcrceh8f333xMdHU2nTp1qlM8RPPa6IYToIoTYZrOdE0I8VC7N5UKIszZpEjwlr0LhLqSU1gZdmF3jSmroneN79+7l4YcfZteuXezatcu6Hsfs2bN59tln6dq1Kxs2bOC3334jMTGRJ554AtCUyd69e1m6dCnjxo1zaD2OHTt2MHLkSA4ePAiUXY9j27ZtmM1mq7Iy1uPYsWMHl112GTNnzuSWW24hLi6OBQsWsG3bNgIDNevSWI9j4sSJzJ49G9Ci49q6oozt4osvriDfZ599xujRo51+bcGDFoeUcjcQCyCEMANHgKV2km6QUl7nTtkUCk8ikZgs+judi17tTGbXD8etzjJwJY15PQ6AwsJCkpKSHArtXhu8xVU1FNgnpTzgaUEUCm/ALM2ACy2OBj4BsDGvxwFamPc+ffrQqlWrmorvEN7SMzYKWFjJsYFCiFQhxNdCiB6VFSCEGC+E2CKE2JKR0TCXw1Q0DqQstThcpjjMDVtxVEdDXY/DYOHChS5zU4EXKA4hhB8QDyyyc/hXoJOUMgZ4HVhWWTlSyneklHFSyrjw8HDXCKtQuAFJaR8HZtfU0dhnjj/22GM8/vjjXHjhhda3eai4Hse0adOsCqA8M2bMYP369fTo0YMlS5bYXY+jd+/eDBs2jGPHjgFY1+Po2bMna9euta7maKzHERsbS15eXp3OLScnhzVr1lTpBqsrHl+PQwhxA/CglHK4A2nTgDgpZZUDnNV6HIr6zLivxpH1dhb/+uZftPt3Oy6Yc0H1mWpITkEOmwM2YxEWrrBc4bRy1XocVdNQ1uPwhteN0VTiphJCtBa6E1AI0R9N3kw3yqZQeASrq8rJoUYMDN96Y7U4FHXDo53jQohgYBgwwWbfAwBSyreBW4CJQohiIA8YJT1tIikULqbFXy0Ys3oMAPvS9nE+5zu9DsNVBbC81XIO5x2m/Zvtuf4f1zu9rvqOWo+jIh5VHFLKHCCs3L63bb7PBea6Wy6FwpNEbYiyfl+2ZRlXcZVL6jkYdpCOmR0JTQ+lG914///eV4rDDmo9joooO1Wh8DJMxdpjuSZyDVe/cbVL6hAIJkyYwPj7x5MZpXl/R906yiV1KRoeSnEoFF6GEbG276i+1hAhTq9DCPL98tnfYT/dBmidpH379HVJXYqGh1IcCoWXYQ117qKhuFC6LK2U0lqPLFHdhwrHUIpDofAyrIrDhU+nMapKIktHbpW4rj5P8dRTT1njPNlj2bJl7Ny5040SNQyU4lAovA3jxd+VisPG4jBmkTdGi0MpjtqhFIdC4WUYFoerwo1AWYujobmqZs2aRefOnbnkkkvYvXs3AO+++y79+vUjJiaGm2++mdzcXH766SeSkpJ49NFHiY2NZd++fXbTKSqiFIdC4WWIEr1RN7muITcsDrBRUHVfnbZW2C5pW1e2bt3KZ599xrZt21i1ahWbN28GtEizmzdvJjU1lW7dujFv3jwuvvhi4uPjefHFF9m2bRvR0dF20ykqohSHQuFluNPisK3HUxaH7ZK2dWXDhg2MHDmSoKAgmjRpYh2V9scff3DppZfSq1cvFixYwI4dO+zmdzRdY0cpDoXCy3DHqKqyFWofnlIctkvauoqxY8cyd+5ctm/fzowZM8jPz69TusaOUhwKhQuQJZI1g9awMHAhK8JXMN9vPm8FvsXyD5ZXme/4x8eJ3hQNwN8H/naHqCzYqa1O98u0X5gXPo9Ro0bR882e3PjZjRSVFLm8/vj4eLZt2+aUOSuDBw9m2bJl5OXlkZWVxfLl2vXOysqiTZs2FBUVWVfjg4rhzCtLpyiLUhwKhQvI3Z2L70++tMlvQ8jJENoXtadbfje+mvFVlfmOvl26eNDSVHsLYjqPbi21iX+/hfwGQFhuGNEnoxm4aSA7Mnbw1e6v2JFRv1w1ffr04fbbbycmJoarr77augrf008/zYABAxg0aFCZtbtHjRrFiy++yIUXXsi+ffsqTacoi7esAKhQNChkkeb2SfdLZ+/dewlYEECfvD7cPPLmqvMVa/mmj5rOxSMqriPtTLaO38q+0/sA2PjZRpJfSmbiwYmc738+XVt2ZdfJXVikh3rM68D06dOZPn16hf0TJ06ssG/QoEFlhuNOnDjRbjpFWZTiUChcgNFfcF7387jt3dvYZdnF8fePE9s71qF8J0NP0i+un0tlDPQNpGdETwB6/rsno/uP5reBv9G6RWuCfIMA6qXiULge5apSKFyAtaNZ7+A2ZmdLSzUd0Ho7bTG5v8G2juIqKReSRKEoh1IcCoULMBSHtTF2cJKdcVwKWWauhTuwKrcSWXaCYA1Rysa7ccb/oxSHQuEK9LhPhuKwxoOqxpAwLJISUVJmroVbsFFutbU4AgICyMzMVMrDS5FSkpmZSUBAQJ3KUX0cCoULMBSAVXE4OslOVzgesTjMdbc42rdvz+HDh8nIyHC6fArnEBAQQPv27etUhlIcCoWTKbYUs3H/RkII4Xjuceb9Oo+wk2E0oxk/H/iZs7+erTRvh9wO+OHn2T4OC5iE5oyoqeXg6+tLVFRU9QkV9RqPKw4hRBqQhfauVSyljCt3XABzgGuAXGCslPJXd8upUNglJwduugkOHbLuWtk2i8R2EbzES/x56k8eXv4w/0z7J7dyK19s/4JFyxdVWtz8rPm0ox0WYSFg0kNw7El3nAUAoiACeAq5bz9iWyq0BHnHaMgMcpsMdmnXDr78Epo08awcCiseVxw6Q6SUJys5djVwgb4NAN7SPxUKz5OSAqtXl9mVEQimNq0BaJNt4Z5foXuGZkEMOGwitIrXnqYF2pv+A1ssDN18DIpdI7Z9zgEgC4sRuXkAWA4egIPulMEOf/4JmzbBVa5Ze11Rc7xFcVTFDcDHUrOZfxZCNBNCtJFSHvO0YAoFFt2l1K8ffPghAKafv+LFewcC0Dq8C/Oe28G+l7I5tC+X/PQIeg26n/Xr1zN48GBWrlxJXl6etbiSAl8AHp75LQHt3BWsSkMcKoERmeTQmklLP+X6lgfY+q8TXHLt/W6Vowzjx8OPP4LqbPcqvEFxSGC1EEIC/5NSvlPueDvgkM3vw/q+MopDCDEeGA/QsWNH10mrUNgjJAS6dwcgcGGqdXfyyc0M6X49otXfwEH8TnYg9YOdBBXHkbqviFbFQ8oUY0Yb7SK6XgAd6jbypab4RZaQbf6akJIQWp9uQ+vTbZi5aCaTH+3uVjnKEBLiuboVleINiuMSKeURIUQEsEYIsUtKub6mhegK5x2AuLg49XqicA/Gm7DN0FkjbMjvrX8n7m2ty87cRLMe+tGPfsX6jPBK3FBSSMzB7rU2AMxBZiwLLMx6ahZ3n7qbDukdGHndSLfLYRdlcXgVHlccUsoj+me6EGIp0B+wVRxHgA42v9vr+xQK70QfUhvcJ5j4G7SIr23vbwsSSrKrX9g7JDYE3xa+rpSwUuJvjyf+9njmR8+HdOjdq7dH5FB4Nx5VHEKIYMAkpczSvw8HygflTwL+JYT4DK1T/Kzq31B4DfYsDiOsiM30Wt8wXzo93smNgtUNY/VBw3ryGO6eBKlwCE9bHK2ApfpkIx/gUynlN0KIBwCklG8Dq9CG4u5FG447zkOyKhQVsac4SioqjvqGVXFUF1vLXShXlVfhUcUhpfwbiLGz/22b7xJ40J1yKRR1wvBGub+bwmkYExA9bnEovJJ6/E6kUHgBdiwOq+Kox0+X1eLw0HKyVpSryiupx7e2QuEF2HGhNAhXlVCuKkXl1ONbW6HwImzfjI0wU/XYVWVtGaofBKZohCjFoVDUBQdHVdU3rH0cylWlsEM9vrXdS1JSErGxsSQlJZX57o76FN7Pb5uvYInfEr5u9jVhH4UBcOrMqQrp6sv/avRx+E3wY9bEWXW69+vLOStqgJSywW19+/aVziYmJkYCMiYmpsx3V+GOOhROYOVKWYy/TCa5zPY938vrBl1XIXl9+V+fGfmM9Vxu9rm5Tvd+nc75mmukBCmXL695XkWNALZIB9tYZXE4SGK7ZCAXAAAgAElEQVRiIjExMSQmJpb57o76FF6MlJzz0zozCkwFrPznSu679j5uevQmgu8KrpC8vvyvP93yExu6bgDgyiFX1unery/nrHAcIRvgaIW4uDi5ZcsWT4uhaAQcWvohfTdO4ouXl5Ptn831j19vPfbUZU8x4/IZHpSu9lz36XVEvRbFzb/cTPQr0XR4qEP1mVwiyHWwciUsX659V7gMIcRWWW49pMrw9MxxhaJe81feEQp8dMPdDJHNIgEI9Qvl+i7XV57RyzEJExahDw9z/2KEFWmAL7j1GaU4FIo6IJGYLJriaB7cnP2T93tYIucghLAqju1fbGfTmU1ljvs39WfYhGEEhXh4dUCFR1CKQ6GoA1JaMEnd4mhAPYaBPoHWIblBvwQR9EtFBbHs+DLuePEO1wqihuN6JUpxVEP+gXyO/u8olnz79rop0ETbiW0JaO/eRXcU3oGUpRaHMDecRu6JS59g4xcbrb9zW+Ry9MKjAAT8FUD7Q+0pSC9wn0DKVeVVOKQ4hBBRUsr91e1riBx66RBHXq96+Q9ZJIl+IdpNEim8CYnELLVRVQ1JcfRu1ZvQLqEc4AAA7S9qzzUrrwHgo3EfwYea0qyMzJWZnFh4osL+iFsjaHlDS5fIrHAfjlocXwJ9yu1bDPR1rjjeh7HwTvht4TQZ0KTMsbM/nuXkkpMOLc6jaJg0VIsDKBMyxeRX6oeznmcVneb7HtlH7q7cCvvP/XSuZopDuaq8kioVhxCiK9ADaCqEuMnmUBOgUfhmjPARLa5uQZuxbcocMwWZOLnkpPcEglO4HYlESL1xq8+xqexgqwiFr813vTGvyuIwXLvRL0XjG+FL8eli9v57L5a8Wg7RUq4qr6K67rwuwHVAM+B6m60PcL9rRfMS9Ps8ISGB6dOnlwm9MP3/pgNwYP8Bu1mNUAu2+apKl5SUROaqTNa2XstXvl/xbei3fOX7FWtbrCV1eCqWYm8YF1mKvfNrSOElHDkXmWPi7Xe15WNy8nLcJZpbEKZSZfH7zt+t1yIjIwOAU5kVQ6oYGDGuRr4ykoFPDmTe3nkA5OflN6h7pNHiyPRyYKCjU9G9YXNmyJEdd+6QySTLK7lSBgQElAm9cA3XyGSS5bMtnrWb1wi1YJuvqnQxMTHyz3v/rBC+wthy/spx2nk5A3vnV19CajiCI+ey6rEXrf/Pi81edKN0rufkqpPye76XySTL231ut16LCT0myGSS5b+7/7vSvD+2+1EmkyzDCZeAbOXfSiaTLFeYV9TsHomP10KOLFvmxDNT2AMXhBzJFEJ8L4T4A0AI0VsI8X9O1WDeiv6S36FjB6ZOnVom9EJYmBbMru+F9rt6jFALtvmqSpeYmGhdcW1p26Wsn7SehO4JWFp50UQsG+ydX0MKL+HQuVi0t/LjzfbT+cPObpLMPYRdHUb2gmym95hO9GPR1mvRuYt2nu3bta88s36vtmvfjsjISCb8cwIAQQFBDeoeabQ4ol2AdUB/4DebfX84qp0qKbMDkAzsBHYAk+2kuRw4C2zTtwRHynamxfHH7X/IZJLl8YXHKxw7+sFRmUyy3Dlmp9Pq23nXTplMsjz20THrvp87/yyTSZbZf2Y7rR6Fc1g5ZbZMJlm+3+kVT4viNhY8uEAmkyzfueWdStNsbLVRJpMs84/lSymlLDpTJJNJlutD19esMsPiWLq0LiIrHIAaWByOjqoKklKmiLIjHIrrqLOKgYellL8KIUKBrUKINVLKneXSbZBSei5IjT5gytbfa2Dd50RLwN7qca6oR+EkdIvDCEPeGHCkc7zCc6MPHPD4+h4Kp+Co4jgphIgGJIAQ4hbgWF0qllIeM8qQUmYJIf4E2qFZIF5DlYvymMqlcQa6cigztNP46mXPXNZvWZz7+VylxwMvCKTFlS3cKJHzOPXtKfL+zgMguFcwzS5pZjed8d/LRqTVrS+QVZxy+efGuJ9rrDjUcFyvxFHF8SDwDtBVCHEE2A/8w1lCCCEigQuBX+wcHiiESAWOAo9IKXc4q97qyCvKIzMnE4Bdp3bx18G/yhy3nNKenIysDDYe3Fghf20oydJe1Xaf3s2eg3sAKC7RjLsq3/DcjJSS1KGpFJ+u2vC8KO0iAjrVr5HbuXty+X3E79bfwlcwKGMQPk3tPC5649moLA7diqjyhancC5D1Rai2U5686N5XOKg4pJR/A1cKIYIBk5Qyy1kCCCFC0CYYPiSlLP/6+ivQSUqZLYS4BlgGXFBJOeOB8QAdO3Z0imy3Lb6Ni9IuYhCDeCL5CX489mOZ40O2DyGBBL7f9z1Pf/C0U+qc+fdMBjOYJ394kvUZ6wGYd2Ye53EeWXlZhBDilHrqiiyWmtIQ0HZC2wrH0xelU5xZTNGponqnOIoyiwDwjfDFkmuhJLuE4jPFdhWHNBSHaEQWh6l6V5WhVIy0tbY4FF6JoyFHppb7DVqn9VYp5bbaVi6E8EVTGguklEvKH7dVJFLKVUKIN4UQLaWUJ+2kfQfNKiIuLs4pd+ffp/9moBwIQJfwLlp3vg2dj2ujS8IDwxnUYZAzqqRlgDartnNEZ0o6aK9nhmvgVO4p2tCm0rzuoNhSzLd7v+XM6TO0ox0WfwspE1MqpItYF4Ffph8b0zYyPGY4ZlM9mh2n64DA6ECKMorI25uHpbASxaC/QTcqxaHfj0UlRRw8e9BuGkuJdj0OZx3GhKlUyUjYPHJzaVlmQfSD0TQf0ryyypwnuMJpOOqqitO35frv64DfgQeEEIuklC/UtGKh3X3zgD+llC9XkqY1cEJKKYUQ/dE8ppk1rau2SCmtkU9fuOoFwq4NK3M8vUk6O9/eyZBOQ5h0zySn1Ll92XYyf8vkueHP0TJeUyKfJHyiyeMFM9QX7VjEHUvuIDgvmBWsIFfmctfSuyqke+vsW3SlK1O+nsKsLrO4ufvNHpC2dtj6540Z07LQ/rWX1s7xRqQ4dOthT8YeOr3ayW6arwu+JoAAur3VjXy/fAAWhywmLDuMnGVlJ0qu2buG27bdVnWlylXlVTiqONoDfaSU2QBCiBnASmAwsBWoseIABgF3AduFEIbV8gTQEUBK+TZwCzBRCFEM5AGjpBsd/RZpKQ0nYadz3F2jqqTQ9lksnm+cTuRogeu6Ne0GgNnfzJ297qyQrkWw1ilukibSc9LdJ6AzMPzzJoHw0/5jS1El1954kRaNp2G7IOwCTnOaUL9QOjSxvzKgSb+B2zZtS5Gv5vp7dvKzRB2KsqYJPx7O7atvJy8nz/VCK5yKo4ojArCNoVwEtJJS5gkhahVbWUq5kdLxQpWlmQvMrU35zmDEtyMYsHcAAD//8jPXXn1t2QR64/7zqp+Z1noaQ4YMISUlhbvvvhuAjz/+mLvvvpuBAweyadMm3nrrLQAmTpxYYZ+R9468O2hKU1I2p3DNdVo00hKL5g9JSUmh99DetTqXpKQkJk+eDMCcOXOIj4+vVTl+e/345LVPaJrVFAB/H3/m3zS/QvlBJm39BiEF0tuGg1XBH+l/8J/F/+FRHmX9gfUElgTSmc6su2gdOX45pE9KZ9Iszbo8t+UcofO02J8llrqOTq8/hAWHcZrTBPwZwNzoucTHx5OUlMQLj7/AfzL+A5ngZ/EDYM9De6wBEpOSkkhYkWCd+Dd72mwASgrL9phLKfnu0u8o+LmA1k2upC/Lqm4oFG7HUcWxAPhFCPGV/vt64FO9s9yrhs86k/5b+lu/v/jFi1z7VFnFYfLVHojzC87nkROPwGfQl77wlHZ8IhPhKfid3wkmmEd4RDtgb5+RV+eD+R9wzUxNcRSVaG9sX3z+Bfc9fl+tziUhIYG0tDTr99oqjtAtobQ/VTpjOLUklSu4okL5/z70b87jPEzS5FWjwapj7f61HDurjTQvMZWQ2jaVzkc6E1IQQkhBCL8s/QVmaWlPrz5tzfd3k1RPiOsRDEv7VOYp672UkJBA8M5gQgm1pjsYeLBMcMSEhARSU1NJSEgAoDCzEICigqIy5VvyLPj+6IsvvmSf7kExIfjWo3uoMeBQyBEp5dNoI5bO6NsDUspEKWWOlLKin6KBIHT/9ayYWUx9fmqF400va0rhkEL+CPiDPwL+IDMqk10huyiOLaY4ttj6vfmw5hTHFlvT2dtn5M2MyuSX5r9w139L+w18fDX9fuutt9b6XBITE4mMjCQyMrJOoR4M//83Ud8wqcMkoj+Ktlt+x0htZJuQAov0vIvNUaQsjXYbdCqIgY8O5MiHR1jYZSEAbduWjiAz3IpL+i/hr7j17hfWU+i6oGVYS+u9lJiYSHQn7V7YFriNSR0m0WxBs9I5H5QN4ZKYmEjzllqHuL+vf5niy4+8cjwyksJdVGtxCCHMwA4pZVdgi+tF8h6MdRZee+c1uvXvVuG4T4gPw9cOZzjD7ea/kiur/F3Zvgpy+Ghy9Lmw/JIojhMfH19rK8MWQ3F0ie3C80uer7T832b/xlnO1jtXlUQi9JZxYP+BxNwUox34FdgNzUJLJwIaDVxWYBZBQY1n7W3D4rjl5lvoEt8F0P7/AbkD+HP0nwy7fhgPff5QhXzl75Fw33AKrinAx1S2Gao4ZLcejchrJFSryqWUJcBuIYRzJkfUI4w3T6Ph9jTeMKrKqgOquyTGjGEp6pWrCrC7hrjduQu6a94iLFZl0yioJJKBEaBT+Dh2LYzRWcYLmpVykwSVxeF9ONrH0RzYIYRIAaxj6aSUdX+F9WKMG9pk9uyNa8xK9gbFYTSc9mJ32WIcN0lT/bI4bFxVtudovQdsvG7Gm3GJqaQxqY3KFUeRfm/4OnY1TCYjZk+5cizlXVVCDcf1MhxVHE+6VAovxV4D4km8YTiu9aGu7pLYWBz1qo8DadfisMYls3GjGN8bm8VhPA95+/JIX1Q61NqIW2YMGqkOw5K3DnnXUa4q78fRkCPrXC2IN2K1ODzsqrLGQfKGly5DB1TTTpaxOOrR22KlFof+dmx7LlbFYbI0qqZN+GvX5UzyGc4kn6lw3BTo2PNiDUdiKb3OJZYS5vw4h/6Ujmic09/EVFmMb12EVjgVR0OOXAS8DnQD/NBeAXKklE1cKJvHMRRHmUi1nsCYZ+gNFofhqqouFIRtH4dXaDzHqMzisCoR21Ox6ePwaUQWR9i1YbR9oC1FJ4sqHDMFmmj7z4qxy+xhuP9sFUfKkRRe2vgSn/O5dd/rA0wMyNnF5XUTW+FEHHVVzQVGAYvQQo/cDTSs5c7s4DUWh/CePg6rxVHNJamvFgfYd1FaLQ5bV5Wl1OJoPGoDfJv70vmtuj/+9lxV2YXZpYrbSGcxkWup1TxjhYtwVHEgpdwrhDDro6w+EEL8BjzuOtE8j3VUlYc7x41WqaaKQ0rJXxP+Ijs1u9ZVmwJNRL8QTZP+unFpdHFU1+9TX/s4pH2Lw3oP2PwFjbWPw1kYVqut4iiRJRVGWZmlGUs9slobA462iLlCCD9gmxDiBSHElBrkrXdIKVkzYA0heVoI85p0jiclJREVFUVUVBRJSUl2j8fGxjJ9+nRiY2PtpqmAXn3WHVl8Zf6KVQmrKi3XtryCgwUce/cYWSlZtd7OrjvLz4k/W8s8maEFJj55qkKA4rIi643Ccwuf48h3R6pMe+SNIySHJrPIfxHP/PMZx6+LCwjYG8DML2YCcOxY6Vplxj3w98m/mXrRVFYGrOTAWweAxqs4yt/LNbqnKVXG4SfDWX3FapKSknhk/CN88OYHZdJ98OYH7Es95HT53c25zedY23Itq0yrrNt3fb8j6auq24yqOPTyIZJDkvnC/wsSJyW67dlx1OK4C01R/AuYghZgvP6EO60hJdkl+KZoXXF7W+2la9OuDuetLrSHEXZh9+7d5OfnOxT+I61zGtG7ogmwBBBAABve3MA1idfYLde2PGNcfQYZzD9/Pp/M/4S7/nEXe/busebz8/OjsLDQ+vuC87XlTvbs3cOlXMpoRrP5583WSY4H0g7Qn/7s37+/SpmbXtqUzBVaIOPTy09bw7DYI2NJBiJbEE44H7/3MalFqXUKi1IXQn8tDZmx4sAKhjEMgI7NO3KOcwgEg7cPJrggGIB8n3z+avMXD+WHu11WT1P+Xq7JPQ3g09qH402P0/psa8zJZhJOJeCf6Y9fiV+FtPu/z3DFKbiV09+dxpRpIgibyaK/wn+n/7fW4YAyFmUgcgQRRDDvf/Pc9uw4GnLkABAK+EspZ0opp0op97pUMg9i68ceP2F8jVxV1YX2MMIuTJ061Rp+oTp+uPEHrnn8Gj7poIVXH9BvQKXl2pZnnIf0k9z70r00GdCEe1+6l7zIPE6GnyQvMo9rH7m2zO97X7rXmuZciDa8sl/fftYyO3XUwmifF31elTJ3fKwjf170JwDdOlecdV8GG0/WiGEjHL4urkCUaJbDd9HfMfStodb9bZtqHb4tDragCZrbbmrkVFaFj2Xzx38yPqvBd/lVoPy9XJN7GsAUYOKuSVpoHZMwkZiYSNvW2nXe1WoX5746x6E2Wii87t16uOYk3IjxPH7T5BvGhI3htFmLdTZtyrRahwOybauuu/o6tz07VVoc+poZM9AsDZO+qxh4XUrpmSfbHeijZbICs5AmWSM3RHWhPWyPz5o1y6EyBYI8/zzueuAumA6dOlRcA8FevUafSKeoTgyIH1BpOntyxMfHc/S9o/x1/190bF8aNKBFcy1cesuIltXKHdRae7NqHd66ynS2fTcj40fy4IQHqy3bZehKLPrC6LLXSX93iL8mnowvM7BgYfOOzZgXzYexYxvlgkP27mVH72nQlEWxWYsqLKTg+uuvx3LQAlvAv70/8fHxfDBFW6qndUTV91B94GSW5t41XWIiLD6M4ieLIQNW/bCKgdcNxGKyMOvALHZs2MHjlzrWfWyrOG658RYmjXPOukDVUZ2ragrauhn9pJT7AYQQ5wFvCSGmSClfcbWAbmPPHnjsMQBkQSAwHovQNIhpzFiwYz67C9P5OyEESFoG3IhcvQZGTqs+47kWwF2Io4dh5PPVJi+PONAdGAbfrYWR2o0sD0UD5yNSU2FkhUUby+bf1wPoBFt+hZHLK0+44xagnVb+m/+Db36vPK2rOdoJ6ITcuxtGjrTuFoe6ACOQP6zDkt8ZMCFG3waHdJddI1QcdcUkTCC0PiKTNCFLJLJAH3Dgo2lwY2VFLxiJXmd2ndhFe9pzJPcIvx3/jbOms4QTzu3zb7em+Tr2a14Y9UKtFEdli425guoUx13AMNulWqWUfwsh/gGsBhqO4jhzBpYtA0DSHE1xaHerWLVKW0bKQ4gxQAjsP/kXUcDRzEMc/n1Z9RkLI/HhLnLyz/KdI+nL15uVh5lhHD112FrfWb97tYPpJ2Br1WXKDvpb4okM2FR5WslQDMVh+X0H1EJWp9FJD/Z8OhO22coxFBhB7mEzhvkhkmwUZ+v6/0bsbkxCu44lphJMJSYWTF5A/vZ8mtO8tGXSn8H6NqTbHoZlHds2lsnjJ1PUoYiiBUVIKZGnJJZUC+Hnwq3r7ziEjUI99c0pirOKaXNvG3ybu3a6ZHWKw7eS9b0z9PXCGw7nnw9L9IYgU8D9pTO2TR9/AuZgj4nm+/dMyE7l854Wpu2DNdEmnh9Zfb7o42beexv+biG57+6a13tlagnTl8LaKBOz9KEQ930niAPMffrAnJuqLuDNNDgE9L4QZo+qPN20UPhL+yrvvgdu9FykfvHBPjgAstsF8EqpYjBv9oXnIBtt8IA5yIKYrx/394crrvCEuPWaIF/NlWlMEO3wZulqgjK0bGgbWdIALDpdH4QGhtKnTR+4B20DTiefJvWKVMwWc40mzNpaHCeXneTkspO0vLGlxxVHYS2P1Tv+5jSJ+jpVwTKYW7mVYt1VJa67HgKaeky2qXsDET+/QrcjWidz2+C2DI0aWk0uaGVqBUCwf7BD6cvT/Xh3ANoEtbHmj2waqR2LuQJGxlRdwKKPtM+wiDJun/LI57cCWQCk/xFBjtCUdPMhzWk9xs1v8l99rH2Gh5WRucU1Fs5rfpjCE9pt33xoc7g6zF4JCgeJCI7g4xs/xvScydqoHht0jLxOeVz2n8uAUleVV0x+rSv6Odob3m9EFDZbzFVaV6fyTvH0uqc5na91rI88O5KmNGX/xfvJCdfiz17gc0HZkVsuoDrFESOEOGdnvwACXCCPx8jMzeSjVK2ha326NbdyKxZhIdg3mAAfz57qiPNHMOL8EZwQJ/jz3T8ZGjmUSXdX3wmWtTWLrQlbOT/8fL67+7sa15vun87Ot3dyecfLefBurcN63459HFp5iGaBzarJTWlsuur80zbHs3/NJvtXbcJi+qfptPpHK/eGfNFlkeayD6/J30THRxvdygIu566Yu9jgt4GSAq1VvWrGVbQY1qI0gclQHJ6QzrlYlZ+dwGaG4vAp8alywuzinYt59ZdXrb+vzLuSpjQloU8Ch1seBmBEkxHOE7oSqlQcUkqXxm4TQowA5qBdyveklM+XO+4PfAz0BTKB26WUaa6QJap5FB/coE088jnoA3MgLDSMjfdsxN/Hv5rcbsJOhNaqMG7UWkf3NdupT7+nHSrTuHuqcdkacnac1pHALoEA7L53N7JIIi3SrYrDGI7bcKe3eh+2/2/5kOyl4XYajquqKsVRnasqr0jrbB0aNZR/9P4Hrd7VvArPDX+O4g7aCLWI4AjnyVwJDocccTb6yoJvAMOAw8BmIUSSlNJ2DfN7gdNSyvOFEKOA/wK3Vyyt7rQMasnY2LEA5AblkkIKTQKbENs61hXV1QprY+3g25e1wa9lI2ivPqsZ7chzbNTroMURfls4oRdqE/D+Gv+XplDc/abpYCwuhROxaUhNfuUufANyVVlf5Oy8CBkK02wxV2lxGEqlR3gPxsaOZZPPJgoo4OZeNxMYGegCqe3jycejP7BXSvm3lLIQ+Ay4oVyaGwDdUc5iYKioNixr7Ti35RzrA9eT7J/Mj11+BCA7r/YxnlyC/oB9vfJra3gC2xAn5UM+bFy3EYDTZ047VHz5sCXGDZ6+LJ1k/2SS/ZM58JIWZsMRxWG4e86bfx4v3/gyveJ68cj0R8p8frb0M06f1uR7+Y3SNIXFWl+COxuM4/OP0+EzrYM245T9mcr2Qrso6oZtQ/r2u2+Xvb66xREx/yLm/t9cT4hXK2zvk78e/Itk/2TaL2sPwIn0ExXSrdugrVxx/onzmbJkSqXl+u/2Z9HsRVxz3TWsNq2m4IAW/PHFl150633pScXRDm3MjcFhjDGZdtJIKYuBs4DdHkkhxHghxBYhxJaMjFqEJ5BgybcgCgW+euT/n/J+qnk5LsS6gE5uHmlpaSQkJFhDnKSlpfHyyy9bw44AvPfuewAcOHzAofJtw5YAhMSGkGPOwYQJUSgQhQIzZgpMBTS5qPqI+tk9ShXvyf0n+eP6P3jJ76Uyn6N/H82hPO02+MjvI+sxi9n9b5qnVp6yfv/2zLd205S/Roq60/zK5gCkk84LC18oc32bh5fGC9u1bJdH5KsNtvdJxpcZ1mcn1y+XVemrKqSb+e5Mss3a83LJzksqLTd4WzAts1vib/HHT2pzy9JI44V3X3DrfdlgDHIp5TtSyjgpZVx4eM3jBoX2DeXS3Es598U5xnccz/0d7yfm/WpGDbkZ483sQtOFvOP/DrMLZjO7YDbv+L/DO/7vsLDlQuYFzWN2wWx+HfQr/yr6FwCdIivONLdH+bAlAZ0CKFlUwqO9HuXcF+c498U5Hu31KAWLCmh2SfWd43G3xDHn1jkA+Jf4Yyoy4Sf9ynwGmAIw611pPvhYjxlzaIqLi2t2keqA4dqbddMsWt9mfzSXvdAuirrRfUF3zn1xjud6P8cDDz9Q5vre3PoM+ztpI92iOkR5UswaUeY+0fs27r32XuL/E88Foy6okO7JZ56k6GNtfROzpfKuZeMeXRW9ivEdx/P9o98zp/ccJj882a33pfDUxBohxEDgKSnlVfrvxwGklM/ZpPlWT7NJCOEDHAfCZTVCx8XFyS1btrhOeA+R/Xs2W2Jqfl6t7m5Ft4+qiRflIjKWZrDjph2E3RBGr2W97KZJ6ZZC7q5c+u3sR3A3bSjuKv9VBBUGMeD0AAKbucd3+8fNf3ByyUlm3DqDIf8cwlOXP+WWehVVMHo0H20y0+nAfewftZ9xC8d5WqIas7HFRopPF7Py85XM/nM2Lw9/mSkDK7qjinKK+DHkRwp8Criq6Cq7ZX304Ed0erMT+2/az7gvnXsthBBbpZRxjqT1WOc4sBm4QAgRBRxBWyjqjnJpkoAxwCbgFmBtdUqjIRPSO4QBewdQeLwGU2jMENontPp0LsLkrxm1hccKOf29/b6Wkmx9vozNSC1jNE1JSQ1m0dYR26VgG2OYdK9ECBDGUoueFaUCFgts3w6FVT+PslCzmi0njwNgPnwENm+ukM5UoJ2gyWKyexyAU2e1z7zsytP06gUBrp1C4DHFIaUsFkL8C/gWrdv3fSnlDiFEIrBFSpkEzAM+EULsBU6hKZdGTWB0IIHR7hs9UVeM9aezUrJIvTK1yrS2QzE9suqhzVKwCm/CS0dWzZwJ5VxDfzZtx+Zmg5GYsJjzyYhYTd+8zzETyOZfvoTzwDT7JUh5qUJxAhPwvRa3q39/u68uMvIOoDfs3A79/21frj17tEgYLsSTFgdSylXAqnL7Emy+5wO3ulsuhfNoenFTWo1pRcGhqpf+DO4ZTEBU6VuSMezQUuK+Rty6FKywVL+musJtCGNeg/uMT8fYq68sERUFYdqYnU0nRnPegT7WJD90DCbGZMJsgZT22r0c2jYS4ipGlxZSwFZtxUMZ19eu1SsK9L7FkCCIq8Sr5O/6eWceVRyKho/J30S3D2vev2JYHG5VHMpV5X0IgTB8VN6mOAyv+dNPw51afDWfaK0j/0z7MzQ73IwrWlxhHaX5yCWP0KxpM255fCL4VYBCu5sAABiqSURBVIx9JwDEDwCUbNqEyadivCk57kPYDvSJgY9frXDcXSjFofBKrIrDnbEmbFxVyuLwJvQ+MG+bPW6vu1XfFXZRGCWLSxjYaiBHLUcBSLwyEZNv1QNZS0wl2iTAYov91tlQnh4eD9tghuMqGhbK4lAYGBaHo6F23IY1ioJNyBRduQl/7dNSYCkN0+NA6JwSfSBApedqPA4uDQZVPUpxKLwSq+Jw4wo+xsNqPLwKL0AIrK2lt41ZsKM4DItDBNgoDh1H4rtZ9KCOJcWV3INKcSjqSmXhL2zDkDgjBEFdw2yUz2/8Lh8ixRZDcQwbNqxMeJXy+WorW/l8eybv4ex6baijxaRcVd6ESW8tz/v2PNaKtawestrDEpXD1uKQ2vcV364AIH1BOlC6tk91GCP6von4huXvLmdty7WsFdqWbEomcmEkABmZtYiO4UyklA1u69u3r2wMxMTESEDGxMTY3W/vmDPrqW1+43dAQECl5S5uslgmkyxbBrS0prGXr7aylc+3KXKTTCZZJpMsQx8Llc+uf7ZW56pwMnfeKef1iJRfBXxl/X9WmVZ5WiqNW2+VEqT87DPrrvnt5stkkuXgpoPlMpZZZZ4bOtehIhN7JlrzPNTmIet3222F3wp52ZDLnH46aNMgHGpjlcVRj6ks/EViYiKRkZFERkY6JQRBXcNslM9v/J46dWql5RoWR7t27aznYS9fbWUrn88Yivtg/INkBWUpi8OLKA5I48bHbmTUzdo0rgA/L1kKyE7nuGFxmKJNpDyewj2d7uGeTvfQYX6HCmnt8d/R/+Vg2EEArrpSmz2+LXAb93S6hzXT1jDyppFcP+16Oo7y7NowHgs54koaasiRxsTi5otpeaYl0Tui6dDdsYeuLvzU/icKjxSSPD+ZxL2JPDf0OaZdMs3l9Sqq4a67eG/nfO6Ph6vaXsW08dOQAZKwtDB6t+rtWdluuQW+/BI+/xxuuw2AhW0W0uZ4G5qva07M4JrHugt9LpQ5r87hvPTzSLo6ifiv4/mt92+8dd9bAGTkZnAm/wz/vfK/PDboMaeeTn0JOaJQVIphcRQeLSS/SX616f1a+2HyqYMBbfS/6p2TalSV9+Cvx7lce2At05hGUVERfd/py9GpRwkPrnlAU6djxzqt7eJpvSJ6Wfs5zmVri6+eKzrHnlN7yqTrHt69VuU7C6U4FF6JoTgODTvEoTLR9+3T5KIm9NnUp9p0ldanu6qMTkzlqvIervsLJoQO4VRrbba1WZopthSTmZfpWcVRxXBck7l2LzHrxq4j5bUUik4UMaH3BHI25HBV56sY/a/R1jShfqG0CW1Te7mdgFIcCq9kff/1DP5pMG1C2mAWVYw9lJpVkrUlq24V6qMfDYWlLA4vQQia58PbLcciR93FujvXWfsRPO5mt6c4jD4OU+0Uh6/ZlwD/AIooopm5GTnk0CSoCZ3DOtdZXGeiFIfCK0kansQbF71B2uQ0OjWrfD0RaZGsM69DFtetEbGNU6XwQqS0rjppkiaQVLk2t9tkKodVcdTS4tAy68UX1m3pZ1eiFIfCKzEJ7WnZdnwb6TnpDuVJOZRSa99ycYnmSD+ep4W+Vq4q70MIoTWiFi30uFvD0VSFPYujDorDmGFuKbSU+e1NKMWh8Ep8TNqteePnN1abdo1pDT4WHwa9M4hin9qtGLgibwXBBLPoz0UQQNXuMYX7KKfAhVkgLVKLIOuNrioj5EgdGnsjr2FxKMWhUDjIIxc/wvu/ve+QO0KaJVggrnUcxX61Uxw+QnsUYtvGEtA0gPgu8bUqR+EiZGkjKoskJoupyntDSu2ecEWja5SNRWJBIEssUKTdd4bFYTbV/sXDanEUKItDoagRD8Q9wANxDwBaeJDJkycDMGfOHOLj48vs8zH7QBGceeEMmXmZBAcHW9MZ+RMSErj22mtZuXJlmc9PP/2UnJwc0BdxW3fvOsxBytrwWvS/5pPXP+GHLj/Q++6KczmKzxazvvN6TOkmCq8sZPia4U6rXpZI1nZbi3mPGZhCsWkSPrf6ABsBCENbl2P9xvWc1/O82lWin2PGIi2syIFDB+hClzpK7ly8sNtFoShLQkICaWlppKWlkZCQUGFffqE2z+PwwcNkZGSUSWekTU1N5eWXX67wmZaWRkZGhvVNUT0RXkY5V1XLG7QhueFZ4Sx+fbHdLLm7cjGla39k7g+5ThWn6FSRrjQ0fCzau3eJKLFuu1ruYs7bc2pdR9g1YRSLYkooIY88FuxaUGe5nY2yOBReT2JiotW6sA1bYuzzz/SHLHgq6CnyS/LxMfvQJ7TP/7d3/9FRlXcex9/fJDSUCAISESSaKNItVhMxh7riWu2BqqwN1d11lW2Vdj1uz9Eeuu6eri5dpDnHPae1aNtVW6lyxK1H665apmCtaH9ti4pRwy9F1AKrkR9Blh+hAsnku3/cO8MkmQkZJsnNzHxe58zJzL3P3HmeXJhv7vM89/uw8dqNANw18i7Wj17PxAkT+WD7B11+btu2jY54B2WHg/8Kxzu4LoNj6qNTaX6+mYm7JnLdNdelLZOaknzUCaP69fPTpTt/6qzv8vjeNVRUVDB37lxWrlyZU6qfqlureH3y612usoeaSFKOmNldwOcJOgjeBb7s7nvTlNsKHCCYZd/R19vhlXKkuLxy7iscXH8w5+OUnVTGjF0zFDyGknnzYNkyWLoUvvxlAB479TEmfDCBEatGMH3m9B5v2fs/e2m+uBmAYeOGMaN1Rr9V53DLYV6c9GKXbW03r+XKe+f322dEJR9SjqwCbnf3DjP7NnA78C8Zyl7q7rsHr2qSb85ZeQ77/rAv5+OMnDZSQSMPJNLCJO696S71qqC/F39Kd7zSXO7ZyFORBA53T02o/xLw11HUQwrD8KrhDL92iGRMlf6V5n6aRFqYjDd9pqyBNBiBo2QIznoaaEMhVH4F+EWGfQ48Z2avmtlNvR3EzG4ysyYza2ptjXiRExHpXyld6skrjgxBYbCvOMqOM71IPhuwFpvZ82a2Ic1jTkqZBUAHkGnawEXuPg24ArjZzC7O9HnuvsTd6929vrJyCGTMFJGBEf6Bn66ramfbThb/fnHy9eH2w3zrN9/ql4/9qP0j7njhjh7bi7GrasBa7O4z3f1TaR7LAcxsHnAl8HeeYYTe3VvCn7uAp4GeI2EiUrjSdFUlrzjSdFU99eZT/OrdXyVfe9xZ9NtF7GjbkXNVfrftdzyx/oke208q79+ZW/kgklBpZpcD3wAa3D3tRGszqzCzkYnnwOeADYNXSxEZMlL+tkyOcaS54jgcP0xJ59GvtVIP7rloj7fnXIXux0449ePjcj52vonqGuteYCSwysyazexHAGY20cyeCcuMB35vZmuBNcBKd382muqKyFCRDBxHegaOeGecYfFhydelnaVYp/VLJt1O7+xy7IRiTGsWSeBw98nuXuXudeHjq+H2D9x9dvj8j+5eGz7Odvc7o6ir5J9YLEZNTQ01NTUsWLCAurq65M9YLBZ19SQbvcyqin8xzor/WJHcHovF+Nk3f8ai/1rUpfzS+5cmM83mpAWWLFnSY/Mbb27M/dh5pvhGdaTgpaYj6Z5mJDUVieSRlK6qTXWbks9jdx39Q2DhwoWM3zq+x1urd1fTsfP4kl+mKtnU/evyMB/nf/nprx/K+dj5RoFDCk5jYyPV1dVUV1dz6623Ultbm/yZSyoIGRpenvUyv6gLZvBf9YWrktsbGxsZO3IsALE/i7HqX1exc9ROIPPNgtno7AiuWl4+7WX2L9/P2ed+g09zA9d95Zqcj51vlKtKCk5DQ0MyMy7AnXfe2eWn5JE0XVVmllypsfac2uT2hoYGPox9COvg3PPPZd6d83jigWAWVK4rRMLRezgqT6kM/n2FOaT+/MILcz52vtEVh4jkFcMy3wSYmDwV/kmcWEO+P24ETB6j+2B4Ea4WqSsOERn6UsY4Sqzk6NrwYXqRfav3sfvp3YxZMybYEE5+6u2ej6yrkAgciT+3o16BMEIKHCKSV8yMeEkQMRJf5pu/upmD6w8ymtEAxE8M9h8rIWI2Ep/lpeGx0iwdWyzUVSUiQ1e6MQ6OjnE8sOYBrnj0CrZ/sB2An8/6OYuvXMyey/cA/dxV1ZnhikOBQ0RkCErpFpo0alIyIGzetZln33mWjkPBdNuHz32YFfUrmHjyxOBtJQMwxtE9ThRh4FBXlYjklfv/8n5eXPkivAQ31t7IvLnzGP69IK3+T675CSdOOJELJl0AkLwy6YznfgOguqqOUuAQkaErzZfyiGEjqD6pmvd4j8kjJlM1vorV7avppJNZn5xF2cijX2uJMQ4OHd/HuzvxA3FKRpTQ3hZO2dLguAKHiOSB1auhvDz50jaOBk5kW+M2tjVuS25fd/s3eDr231x99dWcd955uAdf9h9+/kOarl9A/WenZPWx6++uZM+6EQBMYhIAh1p3BMvZ7ggz7uqKQ0RkCEkEi2XLgkdoLLXs4JvE+Xhy2xhe5ez77mEaJG/Oe23G33DGe0GwOO2RP8Aj/57Vx+/lmR7bztj5DMxLWXd8ePGtPqnAISJD19e+BkeOwKGufU2jgar37mfNmjV0dHRQVlZG9fTpvM+lNDc3U1dXR1VVFS9MWs64/5vKJW9cwodTP8XJ9adn9/mPDuuyFO2rkx/mmqrhMPP6YMPpp8P55+fWxjykwCEiQ9eUKfDAA2l3VYWPdNsT7L6pyXs+xi74N5jbMwlib/yx30LKjKzxs2ZSdf8XszpGIdJ0XBEpWKl5rY5nSm7391hp8Y1npKPAISIFq9e8Vsfg7tBtFq8CR0CBQ0QKVuoVR+pYRZ+kufWjpExfmaDAISIFzOiZ16qv0uW3KinVVyYocIhIgcs2X1XL/hYuXXYp5/+w52wpBY5AJL8FM1tkZi1m1hw+Zmcod7mZvWVm75jZbYNdTxHJb8czOP7cu8/xm62/YdPOTT32jR+V3aysQhXldNx73P27mXaaWSlwHzALeB94xcxi7v7GYFVQRPJbaldVfH+c9j3tDBs7rNf3dHR2UH6knGtO6bkk7OTKyQNSz3wzlO/jmA684+5/BDCzx4E5gAKHiPRJ6hXHlgVb2LJgC4e/cJjLnr4sbflYLMYjdz/C8j8sp7yjvMf+5nXNTGDCgNY5H0TZYXeLma0zs6VmNibN/lOB91Jevx9uS8vMbjKzJjNram1t7e+6ikgeMowXp7zInvI9tNEGQMtzLRnLL1y4kFHbRlHeUc6RkiPsYx/t1s6BsgNsYxvffS5jJ0lRGbDAYWbPm9mGNI85wA+BM4E6YDuwONfPc/cl7l7v7vWVlZW5Hk5ECoCZ8dqZr7Hl4S384JQfAHDaaadlLN/Y2MhJo08C4IWaF/h69df56Gcf4U8699Tew/xvzx+Ueg91A9ZV5e4z+1LOzH4MrEizq4Wu2QMmhdtERPrEwlWXPnPxZ7jxyRt5fcbrjB09NmP5hoYGDvz2ADTDlE9OYcvPt3TZJ4GoZlWldhJeBWxIU+wV4CwzqzGzjwHXArHBqJ+IFAYLU567e3LlPj/GOhreEe4vHcia5beoBse/Y2Z1gANbgX8AMLOJwIPuPtvdO8zsFuCXBKdwqbtvjKi+IpLHHMdKwshxjMUAE4HDy4p3oaZjiSRwuPuXMmz/AJid8voZSJMQX0SkDxJdVe6esnLfMd6USE0ylOecRky/GhEpWMmuKlK6qtKkEtlyxxb2/movACe/fXKwUd+OGelXIyIFK/WKI9lV1S1uxA/Fuyw/W05w/8ahU45zofIioMAhIgUrccWx+0+72f2n3QC0d7Sz/cD2ZJnOtmDQw8qNScsn8cTGJ3ho80Ncd8V1g1/hPKHAISIFq8SCgY2Gxxs4c8eZPMiDvLHzDS68+8JkmYpDFaxgBQf9IJNfClOKTICSEiU0zESBQ0QK1txPzWXr3q10eidjK4L7N0q8hDFlYygfHnRJVZRUANBpnYwpC5JYHNx7kBN2nBBNpfOAHWtOcz6qr6/3pqamqKshIkNI24Y2ms5pYitb+V7t92hubgbgyO4jrK5czX72s7B2IQBr166ltrY2WaYYmNmr7l7fl7K64hCRopAYHB9ePpzGxsajO8L7OkpKS5LbFy5c2LWMdKHAISLFIZxUVVNdw6cbPp3cnFinY8y4McxomAEovcixaPRHRIpCpum4iSuO5H45JgUOESkOiW+7bilHElccVqrA0VcKHCJSHDIkOUzeSa5vwz7Tr0pEikLGJIdhbipdcfSdBsdFpDiEcSHeFqf1qaOrhB7ZfiR4oj+j+0yBQ0SKQkl5EBnaW9vZ+Fc9V2hI7JdjU+AQkaJQPrGc6m9V09bc1nOnwSnXnzLodcpXChwiUjSqF1YDEIvFmD9/PgcPHgSgoqKC79/wfRrQ/Rt9oZQjIlJ06urqWLt2bZdtxZZipLtsUo6oU09Eik5jYyPV1dVUVlZSWVlJdXW1UoxkIZIrDjP7KfCJ8OVoYK+716UptxU4QDBhrqOv0VBXHCIi2RnySQ7d/W8Tz81sMbCvl+KXuvvuga+ViIj0RaSD4xYsz3UN8Nko6yEiIn0X9RjHXwA73f3tDPsdeM7MXjWzm3o7kJndZGZNZtbU2traW1EREcnBgF1xmNnzQLqJ0QvcfXn4/DrgsV4Oc5G7t5jZycAqM9vk7r9LV9DdlwBLIBjjyKHqIiLSiwELHO4+s7f9ZlYGXA2c38sxWsKfu8zsaWA6kDZwiIjI4Iiyq2omsMnd30+308wqzGxk4jnwOWDDINZPRETSiDJwXEu3biozm2hmz4QvxwO/N7O1wBpgpbs/O8h1FBGRbgryznEzawW2HefbxwHFNv1XbS58xdZeUJuzdbq7V/alYEEGjlyYWVNfb4IpFGpz4Su29oLaPJCino4rIiJ5RoFDRESyosDR05KoKxABtbnwFVt7QW0eMBrjEBGRrOiKQ0REsqLAISIiWVHgCJnZ5Wb2lpm9Y2a3RV2fgWJmW81svZk1m1lTuG2sma0ys7fDn2OirmcuzGypme0ysw0p29K20QI/CM/7OjObFl3Nj1+GNi8ys5bwXDeb2eyUfbeHbX7LzC6Lpta5MbMqM/u1mb1hZhvNbH64vSDPdS/tHfzz7O5F/wBKgXeBM4CPAWuBqVHXa4DauhUY123bd4Dbwue3Ad+Oup45tvFiYBqw4VhtBGYDvwAMuAB4Oer692ObFwH/nKbs1PDfeDlQE/7bL426DcfR5gnAtPD5SGBz2LaCPNe9tHfQz7OuOALTgXfc/Y/ufgR4HJgTcZ0G0xxgWfh8GfCFCOuSMw8yKO/ptjlTG+cAj3jgJWC0mU0YnJr2nwxtzmQO8Li7H3b3LcA7BP8H8oq7b3f318LnB4A3gVMp0HPdS3szGbDzrMAROBV4L+X1+/R+QvJZujVOxrv79vD5DoI8YYUmUxsL/dzfEnbLLE3pgiy4NptZNXAe8DJFcK67tRcG+TwrcBSfi9x9GnAFcLOZXZy604Nr3IKeo10MbQz9EDgTqAO2A4ujrc7AMLMTgCeBr7v7/tR9hXiu07R30M+zAkegBahKeT0p3FZwPGWNEyCxxsnOxCV7+HNXdDUcMJnaWLDn3t13unvc3TuBH3O0m6Jg2mxmwwi+RB9196fCzQV7rtO1N4rzrMAReAU4y8xqzOxjBCnfYxHXqd/1ssZJDLghLHYDsDz9EfJapjbGgOvDGTcXAPtSujnyWrf++6s4up5NDLjWzMrNrAY4i2DpgrxiZgY8BLzp7nen7CrIc52pvZGc56hnCgyVB8GMi80EMw8WRF2fAWrjGQSzLNYCGxPtBE4CXgDeBp4HxkZd1xzb+RjBJXs7Qb/u32dqI8EMm/vC874eqI+6/v3Y5v8M27Qu/BKZkFJ+Qdjmt4Aroq7/cbb5IoJuqHVAc/iYXajnupf2Dvp5VsoRERHJirqqREQkKwocIiKSFQUOERHJigKHiIhkRYFDRESyosAhkiMzi4dZSTea2Voz+ycz6/X/lplVm9ncwaqjSH9S4BDJ3UfuXufuZwOzCNK53HGM91QDChySl3Qfh0iOzKzN3U9IeX0GQTaCccDpBDdoVYS7b3H31Wb2EvBJYAtBBten05UbpCaIZEWBQyRH3QNHuG0v8AngANDp7ofM7CzgMXevN7NLCNZQuDIsPyJducFtiUjflEVdAZECNwy418zqgDgwJcdyIpFT4BDpZ2FXVZwgK+sdwE6glmBM8VCGt/1jH8uJRE6D4yL9yMwqgR8B93rQD3wisN2DlNdfIlimGIIurJEpb81UTmTI0RiHSI7MLE6QnXQY0EEwyH23u3eG4xVPEmQ1fRa42d1PCNdV+CVBJteHgRXpyg12W0T6QoFDRESyoq4qERHJigKHiIhkRYFDRESyosAhIiJZUeAQEZGsKHCIiEhWFDhERCQr/w8n1ZCPZ2JaPwAAAABJRU5ErkJggg==\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "%matplotlib inline\n",
     "\n",
@@ -545,469 +516,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "     mean radius  mean texture  mean perimeter  mean area  mean smoothness  \\\n",
-      "0         17.990         10.38          122.80     1001.0          0.11840   \n",
-      "1         20.570         17.77          132.90     1326.0          0.08474   \n",
-      "2         19.690         21.25          130.00     1203.0          0.10960   \n",
-      "3         11.420         20.38           77.58      386.1          0.14250   \n",
-      "4         20.290         14.34          135.10     1297.0          0.10030   \n",
-      "5         12.450         15.70           82.57      477.1          0.12780   \n",
-      "6         18.250         19.98          119.60     1040.0          0.09463   \n",
-      "7         13.710         20.83           90.20      577.9          0.11890   \n",
-      "8         13.000         21.82           87.50      519.8          0.12730   \n",
-      "9         12.460         24.04           83.97      475.9          0.11860   \n",
-      "10        16.020         23.24          102.70      797.8          0.08206   \n",
-      "11        15.780         17.89          103.60      781.0          0.09710   \n",
-      "12        19.170         24.80          132.40     1123.0          0.09740   \n",
-      "13        15.850         23.95          103.70      782.7          0.08401   \n",
-      "14        13.730         22.61           93.60      578.3          0.11310   \n",
-      "15        14.540         27.54           96.73      658.8          0.11390   \n",
-      "16        14.680         20.13           94.74      684.5          0.09867   \n",
-      "17        16.130         20.68          108.10      798.8          0.11700   \n",
-      "18        19.810         22.15          130.00     1260.0          0.09831   \n",
-      "19        13.540         14.36           87.46      566.3          0.09779   \n",
-      "20        13.080         15.71           85.63      520.0          0.10750   \n",
-      "21         9.504         12.44           60.34      273.9          0.10240   \n",
-      "22        15.340         14.26          102.50      704.4          0.10730   \n",
-      "23        21.160         23.04          137.20     1404.0          0.09428   \n",
-      "24        16.650         21.38          110.00      904.6          0.11210   \n",
-      "25        17.140         16.40          116.00      912.7          0.11860   \n",
-      "26        14.580         21.53           97.41      644.8          0.10540   \n",
-      "27        18.610         20.25          122.10     1094.0          0.09440   \n",
-      "28        15.300         25.27          102.40      732.4          0.10820   \n",
-      "29        17.570         15.05          115.00      955.1          0.09847   \n",
-      "..           ...           ...             ...        ...              ...   \n",
-      "539        7.691         25.44           48.34      170.4          0.08668   \n",
-      "540       11.540         14.44           74.65      402.9          0.09984   \n",
-      "541       14.470         24.99           95.81      656.4          0.08837   \n",
-      "542       14.740         25.42           94.70      668.6          0.08275   \n",
-      "543       13.210         28.06           84.88      538.4          0.08671   \n",
-      "544       13.870         20.70           89.77      584.8          0.09578   \n",
-      "545       13.620         23.23           87.19      573.2          0.09246   \n",
-      "546       10.320         16.35           65.31      324.9          0.09434   \n",
-      "547       10.260         16.58           65.85      320.8          0.08877   \n",
-      "548        9.683         19.34           61.05      285.7          0.08491   \n",
-      "549       10.820         24.21           68.89      361.6          0.08192   \n",
-      "550       10.860         21.48           68.51      360.5          0.07431   \n",
-      "551       11.130         22.44           71.49      378.4          0.09566   \n",
-      "552       12.770         29.43           81.35      507.9          0.08276   \n",
-      "553        9.333         21.94           59.01      264.0          0.09240   \n",
-      "554       12.880         28.92           82.50      514.3          0.08123   \n",
-      "555       10.290         27.61           65.67      321.4          0.09030   \n",
-      "556       10.160         19.59           64.73      311.7          0.10030   \n",
-      "557        9.423         27.88           59.26      271.3          0.08123   \n",
-      "558       14.590         22.68           96.39      657.1          0.08473   \n",
-      "559       11.510         23.93           74.52      403.5          0.09261   \n",
-      "560       14.050         27.15           91.38      600.4          0.09929   \n",
-      "561       11.200         29.37           70.67      386.0          0.07449   \n",
-      "562       15.220         30.62          103.40      716.9          0.10480   \n",
-      "563       20.920         25.09          143.00     1347.0          0.10990   \n",
-      "564       21.560         22.39          142.00     1479.0          0.11100   \n",
-      "565       20.130         28.25          131.20     1261.0          0.09780   \n",
-      "566       16.600         28.08          108.30      858.1          0.08455   \n",
-      "567       20.600         29.33          140.10     1265.0          0.11780   \n",
-      "568        7.760         24.54           47.92      181.0          0.05263   \n",
-      "\n",
-      "     mean compactness  mean concavity  mean concave points  mean symmetry  \\\n",
-      "0             0.27760        0.300100             0.147100         0.2419   \n",
-      "1             0.07864        0.086900             0.070170         0.1812   \n",
-      "2             0.15990        0.197400             0.127900         0.2069   \n",
-      "3             0.28390        0.241400             0.105200         0.2597   \n",
-      "4             0.13280        0.198000             0.104300         0.1809   \n",
-      "5             0.17000        0.157800             0.080890         0.2087   \n",
-      "6             0.10900        0.112700             0.074000         0.1794   \n",
-      "7             0.16450        0.093660             0.059850         0.2196   \n",
-      "8             0.19320        0.185900             0.093530         0.2350   \n",
-      "9             0.23960        0.227300             0.085430         0.2030   \n",
-      "10            0.06669        0.032990             0.033230         0.1528   \n",
-      "11            0.12920        0.099540             0.066060         0.1842   \n",
-      "12            0.24580        0.206500             0.111800         0.2397   \n",
-      "13            0.10020        0.099380             0.053640         0.1847   \n",
-      "14            0.22930        0.212800             0.080250         0.2069   \n",
-      "15            0.15950        0.163900             0.073640         0.2303   \n",
-      "16            0.07200        0.073950             0.052590         0.1586   \n",
-      "17            0.20220        0.172200             0.102800         0.2164   \n",
-      "18            0.10270        0.147900             0.094980         0.1582   \n",
-      "19            0.08129        0.066640             0.047810         0.1885   \n",
-      "20            0.12700        0.045680             0.031100         0.1967   \n",
-      "21            0.06492        0.029560             0.020760         0.1815   \n",
-      "22            0.21350        0.207700             0.097560         0.2521   \n",
-      "23            0.10220        0.109700             0.086320         0.1769   \n",
-      "24            0.14570        0.152500             0.091700         0.1995   \n",
-      "25            0.22760        0.222900             0.140100         0.3040   \n",
-      "26            0.18680        0.142500             0.087830         0.2252   \n",
-      "27            0.10660        0.149000             0.077310         0.1697   \n",
-      "28            0.16970        0.168300             0.087510         0.1926   \n",
-      "29            0.11570        0.098750             0.079530         0.1739   \n",
-      "..                ...             ...                  ...            ...   \n",
-      "539           0.11990        0.092520             0.013640         0.2037   \n",
-      "540           0.11200        0.067370             0.025940         0.1818   \n",
-      "541           0.12300        0.100900             0.038900         0.1872   \n",
-      "542           0.07214        0.041050             0.030270         0.1840   \n",
-      "543           0.06877        0.029870             0.032750         0.1628   \n",
-      "544           0.10180        0.036880             0.023690         0.1620   \n",
-      "545           0.06747        0.029740             0.024430         0.1664   \n",
-      "546           0.04994        0.010120             0.005495         0.1885   \n",
-      "547           0.08066        0.043580             0.024380         0.1669   \n",
-      "548           0.05030        0.023370             0.009615         0.1580   \n",
-      "549           0.06602        0.015480             0.008160         0.1976   \n",
-      "550           0.04227        0.000000             0.000000         0.1661   \n",
-      "551           0.08194        0.048240             0.022570         0.2030   \n",
-      "552           0.04234        0.019970             0.014990         0.1539   \n",
-      "553           0.05605        0.039960             0.012820         0.1692   \n",
-      "554           0.05824        0.061950             0.023430         0.1566   \n",
-      "555           0.07658        0.059990             0.027380         0.1593   \n",
-      "556           0.07504        0.005025             0.011160         0.1791   \n",
-      "557           0.04971        0.000000             0.000000         0.1742   \n",
-      "558           0.13300        0.102900             0.037360         0.1454   \n",
-      "559           0.10210        0.111200             0.041050         0.1388   \n",
-      "560           0.11260        0.044620             0.043040         0.1537   \n",
-      "561           0.03558        0.000000             0.000000         0.1060   \n",
-      "562           0.20870        0.255000             0.094290         0.2128   \n",
-      "563           0.22360        0.317400             0.147400         0.2149   \n",
-      "564           0.11590        0.243900             0.138900         0.1726   \n",
-      "565           0.10340        0.144000             0.097910         0.1752   \n",
-      "566           0.10230        0.092510             0.053020         0.1590   \n",
-      "567           0.27700        0.351400             0.152000         0.2397   \n",
-      "568           0.04362        0.000000             0.000000         0.1587   \n",
-      "\n",
-      "     mean fractal dimension           ...             worst radius  \\\n",
-      "0                   0.07871           ...                   25.380   \n",
-      "1                   0.05667           ...                   24.990   \n",
-      "2                   0.05999           ...                   23.570   \n",
-      "3                   0.09744           ...                   14.910   \n",
-      "4                   0.05883           ...                   22.540   \n",
-      "5                   0.07613           ...                   15.470   \n",
-      "6                   0.05742           ...                   22.880   \n",
-      "7                   0.07451           ...                   17.060   \n",
-      "8                   0.07389           ...                   15.490   \n",
-      "9                   0.08243           ...                   15.090   \n",
-      "10                  0.05697           ...                   19.190   \n",
-      "11                  0.06082           ...                   20.420   \n",
-      "12                  0.07800           ...                   20.960   \n",
-      "13                  0.05338           ...                   16.840   \n",
-      "14                  0.07682           ...                   15.030   \n",
-      "15                  0.07077           ...                   17.460   \n",
-      "16                  0.05922           ...                   19.070   \n",
-      "17                  0.07356           ...                   20.960   \n",
-      "18                  0.05395           ...                   27.320   \n",
-      "19                  0.05766           ...                   15.110   \n",
-      "20                  0.06811           ...                   14.500   \n",
-      "21                  0.06905           ...                   10.230   \n",
-      "22                  0.07032           ...                   18.070   \n",
-      "23                  0.05278           ...                   29.170   \n",
-      "24                  0.06330           ...                   26.460   \n",
-      "25                  0.07413           ...                   22.250   \n",
-      "26                  0.06924           ...                   17.620   \n",
-      "27                  0.05699           ...                   21.310   \n",
-      "28                  0.06540           ...                   20.270   \n",
-      "29                  0.06149           ...                   20.010   \n",
-      "..                      ...           ...                      ...   \n",
-      "539                 0.07751           ...                    8.678   \n",
-      "540                 0.06782           ...                   12.260   \n",
-      "541                 0.06341           ...                   16.220   \n",
-      "542                 0.05680           ...                   16.510   \n",
-      "543                 0.05781           ...                   14.370   \n",
-      "544                 0.06688           ...                   15.050   \n",
-      "545                 0.05801           ...                   15.350   \n",
-      "546                 0.06201           ...                   11.250   \n",
-      "547                 0.06714           ...                   10.830   \n",
-      "548                 0.06235           ...                   10.930   \n",
-      "549                 0.06328           ...                   13.030   \n",
-      "550                 0.05948           ...                   11.660   \n",
-      "551                 0.06552           ...                   12.020   \n",
-      "552                 0.05637           ...                   13.870   \n",
-      "553                 0.06576           ...                    9.845   \n",
-      "554                 0.05708           ...                   13.890   \n",
-      "555                 0.06127           ...                   10.840   \n",
-      "556                 0.06331           ...                   10.650   \n",
-      "557                 0.06059           ...                   10.490   \n",
-      "558                 0.06147           ...                   15.480   \n",
-      "559                 0.06570           ...                   12.480   \n",
-      "560                 0.06171           ...                   15.300   \n",
-      "561                 0.05502           ...                   11.920   \n",
-      "562                 0.07152           ...                   17.520   \n",
-      "563                 0.06879           ...                   24.290   \n",
-      "564                 0.05623           ...                   25.450   \n",
-      "565                 0.05533           ...                   23.690   \n",
-      "566                 0.05648           ...                   18.980   \n",
-      "567                 0.07016           ...                   25.740   \n",
-      "568                 0.05884           ...                    9.456   \n",
-      "\n",
-      "     worst texture  worst perimeter  worst area  worst smoothness  \\\n",
-      "0            17.33           184.60      2019.0           0.16220   \n",
-      "1            23.41           158.80      1956.0           0.12380   \n",
-      "2            25.53           152.50      1709.0           0.14440   \n",
-      "3            26.50            98.87       567.7           0.20980   \n",
-      "4            16.67           152.20      1575.0           0.13740   \n",
-      "5            23.75           103.40       741.6           0.17910   \n",
-      "6            27.66           153.20      1606.0           0.14420   \n",
-      "7            28.14           110.60       897.0           0.16540   \n",
-      "8            30.73           106.20       739.3           0.17030   \n",
-      "9            40.68            97.65       711.4           0.18530   \n",
-      "10           33.88           123.80      1150.0           0.11810   \n",
-      "11           27.28           136.50      1299.0           0.13960   \n",
-      "12           29.94           151.70      1332.0           0.10370   \n",
-      "13           27.66           112.00       876.5           0.11310   \n",
-      "14           32.01           108.80       697.7           0.16510   \n",
-      "15           37.13           124.10       943.2           0.16780   \n",
-      "16           30.88           123.40      1138.0           0.14640   \n",
-      "17           31.48           136.80      1315.0           0.17890   \n",
-      "18           30.88           186.80      2398.0           0.15120   \n",
-      "19           19.26            99.70       711.2           0.14400   \n",
-      "20           20.49            96.09       630.5           0.13120   \n",
-      "21           15.66            65.13       314.9           0.13240   \n",
-      "22           19.08           125.10       980.9           0.13900   \n",
-      "23           35.59           188.00      2615.0           0.14010   \n",
-      "24           31.56           177.00      2215.0           0.18050   \n",
-      "25           21.40           152.40      1461.0           0.15450   \n",
-      "26           33.21           122.40       896.9           0.15250   \n",
-      "27           27.26           139.90      1403.0           0.13380   \n",
-      "28           36.71           149.30      1269.0           0.16410   \n",
-      "29           19.52           134.90      1227.0           0.12550   \n",
-      "..             ...              ...         ...               ...   \n",
-      "539          31.89            54.49       223.6           0.15960   \n",
-      "540          19.68            78.78       457.8           0.13450   \n",
-      "541          31.73           113.50       808.9           0.13400   \n",
-      "542          32.29           107.40       826.4           0.10600   \n",
-      "543          37.17            92.48       629.6           0.10720   \n",
-      "544          24.75            99.17       688.6           0.12640   \n",
-      "545          29.09            97.58       729.8           0.12160   \n",
-      "546          21.77            71.12       384.9           0.12850   \n",
-      "547          22.04            71.08       357.4           0.14610   \n",
-      "548          25.59            69.10       364.2           0.11990   \n",
-      "549          31.45            83.90       505.6           0.12040   \n",
-      "550          24.77            74.08       412.3           0.10010   \n",
-      "551          28.26            77.80       436.6           0.10870   \n",
-      "552          36.00            88.10       594.7           0.12340   \n",
-      "553          25.05            62.86       295.8           0.11030   \n",
-      "554          35.74            88.84       595.7           0.12270   \n",
-      "555          34.91            69.57       357.6           0.13840   \n",
-      "556          22.88            67.88       347.3           0.12650   \n",
-      "557          34.24            66.50       330.6           0.10730   \n",
-      "558          27.27           105.90       733.5           0.10260   \n",
-      "559          37.16            82.28       474.2           0.12980   \n",
-      "560          33.17           100.20       706.7           0.12410   \n",
-      "561          38.30            75.19       439.6           0.09267   \n",
-      "562          42.79           128.70       915.0           0.14170   \n",
-      "563          29.41           179.10      1819.0           0.14070   \n",
-      "564          26.40           166.10      2027.0           0.14100   \n",
-      "565          38.25           155.00      1731.0           0.11660   \n",
-      "566          34.12           126.70      1124.0           0.11390   \n",
-      "567          39.42           184.60      1821.0           0.16500   \n",
-      "568          30.37            59.16       268.6           0.08996   \n",
-      "\n",
-      "     worst compactness  worst concavity  worst concave points  worst symmetry  \\\n",
-      "0              0.66560          0.71190               0.26540          0.4601   \n",
-      "1              0.18660          0.24160               0.18600          0.2750   \n",
-      "2              0.42450          0.45040               0.24300          0.3613   \n",
-      "3              0.86630          0.68690               0.25750          0.6638   \n",
-      "4              0.20500          0.40000               0.16250          0.2364   \n",
-      "5              0.52490          0.53550               0.17410          0.3985   \n",
-      "6              0.25760          0.37840               0.19320          0.3063   \n",
-      "7              0.36820          0.26780               0.15560          0.3196   \n",
-      "8              0.54010          0.53900               0.20600          0.4378   \n",
-      "9              1.05800          1.10500               0.22100          0.4366   \n",
-      "10             0.15510          0.14590               0.09975          0.2948   \n",
-      "11             0.56090          0.39650               0.18100          0.3792   \n",
-      "12             0.39030          0.36390               0.17670          0.3176   \n",
-      "13             0.19240          0.23220               0.11190          0.2809   \n",
-      "14             0.77250          0.69430               0.22080          0.3596   \n",
-      "15             0.65770          0.70260               0.17120          0.4218   \n",
-      "16             0.18710          0.29140               0.16090          0.3029   \n",
-      "17             0.42330          0.47840               0.20730          0.3706   \n",
-      "18             0.31500          0.53720               0.23880          0.2768   \n",
-      "19             0.17730          0.23900               0.12880          0.2977   \n",
-      "20             0.27760          0.18900               0.07283          0.3184   \n",
-      "21             0.11480          0.08867               0.06227          0.2450   \n",
-      "22             0.59540          0.63050               0.23930          0.4667   \n",
-      "23             0.26000          0.31550               0.20090          0.2822   \n",
-      "24             0.35780          0.46950               0.20950          0.3613   \n",
-      "25             0.39490          0.38530               0.25500          0.4066   \n",
-      "26             0.66430          0.55390               0.27010          0.4264   \n",
-      "27             0.21170          0.34460               0.14900          0.2341   \n",
-      "28             0.61100          0.63350               0.20240          0.4027   \n",
-      "29             0.28120          0.24890               0.14560          0.2756   \n",
-      "..                 ...              ...                   ...             ...   \n",
-      "539            0.30640          0.33930               0.05000          0.2790   \n",
-      "540            0.21180          0.17970               0.06918          0.2329   \n",
-      "541            0.42020          0.40400               0.12050          0.3187   \n",
-      "542            0.13760          0.16110               0.10950          0.2722   \n",
-      "543            0.13810          0.10620               0.07958          0.2473   \n",
-      "544            0.20370          0.13770               0.06845          0.2249   \n",
-      "545            0.15170          0.10490               0.07174          0.2642   \n",
-      "546            0.08842          0.04384               0.02381          0.2681   \n",
-      "547            0.22460          0.17830               0.08333          0.2691   \n",
-      "548            0.09546          0.09350               0.03846          0.2552   \n",
-      "549            0.16330          0.06194               0.03264          0.3059   \n",
-      "550            0.07348          0.00000               0.00000          0.2458   \n",
-      "551            0.17820          0.15640               0.06413          0.3169   \n",
-      "552            0.10640          0.08653               0.06498          0.2407   \n",
-      "553            0.08298          0.07993               0.02564          0.2435   \n",
-      "554            0.16200          0.24390               0.06493          0.2372   \n",
-      "555            0.17100          0.20000               0.09127          0.2226   \n",
-      "556            0.12000          0.01005               0.02232          0.2262   \n",
-      "557            0.07158          0.00000               0.00000          0.2475   \n",
-      "558            0.31710          0.36620               0.11050          0.2258   \n",
-      "559            0.25170          0.36300               0.09653          0.2112   \n",
-      "560            0.22640          0.13260               0.10480          0.2250   \n",
-      "561            0.05494          0.00000               0.00000          0.1566   \n",
-      "562            0.79170          1.17000               0.23560          0.4089   \n",
-      "563            0.41860          0.65990               0.25420          0.2929   \n",
-      "564            0.21130          0.41070               0.22160          0.2060   \n",
-      "565            0.19220          0.32150               0.16280          0.2572   \n",
-      "566            0.30940          0.34030               0.14180          0.2218   \n",
-      "567            0.86810          0.93870               0.26500          0.4087   \n",
-      "568            0.06444          0.00000               0.00000          0.2871   \n",
-      "\n",
-      "     worst fractal dimension  \n",
-      "0                    0.11890  \n",
-      "1                    0.08902  \n",
-      "2                    0.08758  \n",
-      "3                    0.17300  \n",
-      "4                    0.07678  \n",
-      "5                    0.12440  \n",
-      "6                    0.08368  \n",
-      "7                    0.11510  \n",
-      "8                    0.10720  \n",
-      "9                    0.20750  \n",
-      "10                   0.08452  \n",
-      "11                   0.10480  \n",
-      "12                   0.10230  \n",
-      "13                   0.06287  \n",
-      "14                   0.14310  \n",
-      "15                   0.13410  \n",
-      "16                   0.08216  \n",
-      "17                   0.11420  \n",
-      "18                   0.07615  \n",
-      "19                   0.07259  \n",
-      "20                   0.08183  \n",
-      "21                   0.07773  \n",
-      "22                   0.09946  \n",
-      "23                   0.07526  \n",
-      "24                   0.09564  \n",
-      "25                   0.10590  \n",
-      "26                   0.12750  \n",
-      "27                   0.07421  \n",
-      "28                   0.09876  \n",
-      "29                   0.07919  \n",
-      "..                       ...  \n",
-      "539                  0.10660  \n",
-      "540                  0.08134  \n",
-      "541                  0.10230  \n",
-      "542                  0.06956  \n",
-      "543                  0.06443  \n",
-      "544                  0.08492  \n",
-      "545                  0.06953  \n",
-      "546                  0.07399  \n",
-      "547                  0.09479  \n",
-      "548                  0.07920  \n",
-      "549                  0.07626  \n",
-      "550                  0.06592  \n",
-      "551                  0.08032  \n",
-      "552                  0.06484  \n",
-      "553                  0.07393  \n",
-      "554                  0.07242  \n",
-      "555                  0.08283  \n",
-      "556                  0.06742  \n",
-      "557                  0.06969  \n",
-      "558                  0.08004  \n",
-      "559                  0.08732  \n",
-      "560                  0.08321  \n",
-      "561                  0.05905  \n",
-      "562                  0.14090  \n",
-      "563                  0.09873  \n",
-      "564                  0.07115  \n",
-      "565                  0.06637  \n",
-      "566                  0.07820  \n",
-      "567                  0.12400  \n",
-      "568                  0.07039  \n",
-      "\n",
-      "[569 rows x 30 columns]\n",
-      "     malignant  benign\n",
-      "0            1       0\n",
-      "1            1       0\n",
-      "2            1       0\n",
-      "3            1       0\n",
-      "4            1       0\n",
-      "5            1       0\n",
-      "6            1       0\n",
-      "7            1       0\n",
-      "8            1       0\n",
-      "9            1       0\n",
-      "10           1       0\n",
-      "11           1       0\n",
-      "12           1       0\n",
-      "13           1       0\n",
-      "14           1       0\n",
-      "15           1       0\n",
-      "16           1       0\n",
-      "17           1       0\n",
-      "18           1       0\n",
-      "19           0       1\n",
-      "20           0       1\n",
-      "21           0       1\n",
-      "22           1       0\n",
-      "23           1       0\n",
-      "24           1       0\n",
-      "25           1       0\n",
-      "26           1       0\n",
-      "27           1       0\n",
-      "28           1       0\n",
-      "29           1       0\n",
-      "..         ...     ...\n",
-      "539          0       1\n",
-      "540          0       1\n",
-      "541          0       1\n",
-      "542          0       1\n",
-      "543          0       1\n",
-      "544          0       1\n",
-      "545          0       1\n",
-      "546          0       1\n",
-      "547          0       1\n",
-      "548          0       1\n",
-      "549          0       1\n",
-      "550          0       1\n",
-      "551          0       1\n",
-      "552          0       1\n",
-      "553          0       1\n",
-      "554          0       1\n",
-      "555          0       1\n",
-      "556          0       1\n",
-      "557          0       1\n",
-      "558          0       1\n",
-      "559          0       1\n",
-      "560          0       1\n",
-      "561          0       1\n",
-      "562          1       0\n",
-      "563          1       0\n",
-      "564          1       0\n",
-      "565          1       0\n",
-      "566          1       0\n",
-      "567          1       0\n",
-      "568          0       1\n",
-      "\n",
-      "[569 rows x 2 columns]\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "0"
-      ]
-     },
-     "execution_count": 6,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "import os\n",
     "from sklearn.datasets import load_breast_cancer\n",
@@ -1053,20 +566,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0"
-      ]
-     },
-     "execution_count": 7,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "# Common imports\n",
     "import numpy as np\n",
@@ -1154,78 +658,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "  (0, 0)\t1.0\n",
-      "  (0, 7)\t1.0\n",
-      "  (0, 9)\t1.0\n",
-      "  (0, 13)\t1.0\n",
-      "  (1, 3)\t1.0\n",
-      "  (1, 5)\t1.0\n",
-      "  (1, 8)\t1.0\n",
-      "  (1, 12)\t1.0\n",
-      "  (2, 3)\t1.0\n",
-      "  (2, 5)\t1.0\n",
-      "  (2, 8)\t1.0\n",
-      "  (2, 11)\t1.0\n",
-      "  (3, 1)\t1.0\n",
-      "  (3, 5)\t1.0\n",
-      "  (3, 8)\t1.0\n",
-      "  (3, 12)\t1.0\n",
-      "  (4, 2)\t1.0\n",
-      "  (4, 6)\t1.0\n",
-      "  (4, 8)\t1.0\n",
-      "  (4, 12)\t1.0\n",
-      "  (5, 2)\t1.0\n",
-      "  (5, 4)\t1.0\n",
-      "  (5, 10)\t1.0\n",
-      "  (5, 12)\t1.0\n",
-      "  (6, 2)\t1.0\n",
-      "  :\t:\n",
-      "  (8, 12)\t1.0\n",
-      "  (9, 3)\t1.0\n",
-      "  (9, 4)\t1.0\n",
-      "  (9, 10)\t1.0\n",
-      "  (9, 12)\t1.0\n",
-      "  (10, 2)\t1.0\n",
-      "  (10, 6)\t1.0\n",
-      "  (10, 10)\t1.0\n",
-      "  (10, 12)\t1.0\n",
-      "  (11, 3)\t1.0\n",
-      "  (11, 6)\t1.0\n",
-      "  (11, 10)\t1.0\n",
-      "  (11, 11)\t1.0\n",
-      "  (12, 1)\t1.0\n",
-      "  (12, 6)\t1.0\n",
-      "  (12, 8)\t1.0\n",
-      "  (12, 11)\t1.0\n",
-      "  (13, 1)\t1.0\n",
-      "  (13, 5)\t1.0\n",
-      "  (13, 10)\t1.0\n",
-      "  (13, 12)\t1.0\n",
-      "  (14, 2)\t1.0\n",
-      "  (14, 6)\t1.0\n",
-      "  (14, 8)\t1.0\n",
-      "  (14, 11)\t1.0\n",
-      "Train set accuracy with Decision Tree: 0.73\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "0"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "# Common imports\n",
     "import numpy as np\n",
@@ -1312,73 +749,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "X1 < 0.000 Gini=0.408\n",
-      "X1 < 0.000 Gini=0.408\n",
-      "X1 < 1.000 Gini=0.394\n",
-      "X1 < 2.000 Gini=0.394\n",
-      "X1 < 2.000 Gini=0.394\n",
-      "X1 < 2.000 Gini=0.394\n",
-      "X1 < 1.000 Gini=0.394\n",
-      "X1 < 0.000 Gini=0.408\n",
-      "X1 < 0.000 Gini=0.408\n",
-      "X1 < 2.000 Gini=0.394\n",
-      "X1 < 0.000 Gini=0.408\n",
-      "X1 < 1.000 Gini=0.394\n",
-      "X1 < 1.000 Gini=0.394\n",
-      "X1 < 2.000 Gini=0.394\n",
-      "X2 < 0.000 Gini=0.408\n",
-      "X2 < 0.000 Gini=0.408\n",
-      "X2 < 0.000 Gini=0.408\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X2 < 2.000 Gini=0.407\n",
-      "X2 < 2.000 Gini=0.407\n",
-      "X2 < 2.000 Gini=0.407\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X2 < 2.000 Gini=0.407\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X2 < 0.000 Gini=0.408\n",
-      "X2 < 1.000 Gini=0.407\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X3 < 1.000 Gini=0.367\n",
-      "X3 < 0.000 Gini=0.408\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "X4 < 0.000 Gini=0.408\n",
-      "X4 < 1.000 Gini=0.405\n",
-      "Split: [X3 < 1.000]\n"
-     ]
-    }
-   ],
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "# Split a dataset based on an attribute and an attribute value\n",
     "def test_split(index, value, dataset):\n",
@@ -1482,37 +857,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "['Outlook', 'Temperature', 'Humidity', 'Wind']\n",
-      "System entropy 0.863120568566631\n",
-      "Outlook\n",
-      "(Sunny)\n",
-      "(Overcast)\n",
-      "(Rain)\n",
-      "Humidity\n",
-      "(High)\n",
-      "(Normal)\n",
-      "1\n",
-      "Temperature\n",
-      "(Mild)\n",
-      "(Cool)\n",
-      "Wind\n",
-      "(Weak)\n",
-      "(Strong)\n",
-      "1\n",
-      "1\n",
-      "0\n",
-      "0\n",
-      "1\n"
-     ]
-    }
-   ],
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "import re\n",
     "import math\n",
@@ -1712,32 +1061,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "(426, 30)\n",
-      "(143, 30)\n",
-      "Test set accuracy with Logistic Regression: 0.95\n",
-      "Test set accuracy with SVM: 0.63\n",
-      "Test set accuracy with Decision Trees: 0.90\n",
-      "Test set accuracy Logistic Regression with scaled data: 0.96\n",
-      "Test set accuracy SVM with scaled data: 0.96\n",
-      "Test set accuracy with Decision Trees and scaled data: 0.90\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:757: ConvergenceWarning: lbfgs failed to converge. Increase the number of iterations.\n",
-      "  \"of iterations.\", ConvergenceWarning)\n"
-     ]
-    }
-   ],
+   "execution_count": 7,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "import matplotlib.pyplot as plt\n",
     "import numpy as np\n",
@@ -1791,20 +1119,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 8,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from __future__ import division, print_function, unicode_literals\n",
     "\n",
@@ -1881,20 +1200,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "np.random.seed(6)\n",
     "Xs = np.random.rand(100, 2) - 0.5\n",
@@ -1927,8 +1237,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "# Quadratic training set + noise\n",
@@ -1941,24 +1253,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "DecisionTreeRegressor(criterion='mse', max_depth=2, max_features=None,\n",
-       "           max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
-       "           min_impurity_split=None, min_samples_leaf=1,\n",
-       "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
-       "           presort=False, random_state=42, splitter='best')"
-      ]
-     },
-     "execution_count": 15,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.tree import DecisionTreeRegressor\n",
     "\n",
@@ -1975,20 +1274,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 12,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.tree import DecisionTreeRegressor\n",
     "\n",
@@ -2032,20 +1322,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 13,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "tree_reg1 = DecisionTreeRegressor(random_state=42)\n",
     "tree_reg2 = DecisionTreeRegressor(random_state=42, min_samples_leaf=10)\n",
@@ -2135,6 +1416,20 @@
     "\n",
     "We discuss these methods here.\n",
     "\n",
+    "\n",
+    "## An Overview of Ensemble Methods\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "
    \n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
     "## Bagging\n",
     "\n",
     "The **plain** decision trees suffer from high\n",
@@ -2179,20 +1474,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 14,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "heads_proba = 0.51\n",
     "coin_tosses = (np.random.rand(10000, 10) < heads_proba).astype(np.int32)\n",
@@ -2217,24 +1503,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "LogisticRegression 0.864\n",
-      "RandomForestClassifier 0.872\n",
-      "SVC 0.888\n",
-      "VotingClassifier 0.896\n",
-      "LogisticRegression 0.864\n",
-      "RandomForestClassifier 0.872\n",
-      "SVC 0.888\n",
-      "VotingClassifier 0.912\n"
-     ]
-    }
-   ],
+   "execution_count": 15,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.model_selection import train_test_split\n",
     "from sklearn.datasets import make_moons\n",
@@ -2290,37 +1563,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/ensemble/forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
-      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "VotingClassifier(estimators=[('lr', LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
-       "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
-       "          n_jobs=None, penalty='l2', random_state=42, solver='warn',\n",
-       "          tol=0.0001, verbose=0, warm_start=False)), ('rf', RandomFore...rbf', max_iter=-1, probability=False, random_state=42,\n",
-       "  shrinking=True, tol=0.001, verbose=False))],\n",
-       "         flatten_transform=None, n_jobs=None, voting='hard', weights=None)"
-      ]
-     },
-     "execution_count": 20,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 16,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.model_selection import train_test_split\n",
     "from sklearn.datasets import make_moons\n",
@@ -2344,36 +1591,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "LogisticRegression 0.864\n",
-      "RandomForestClassifier 0.872\n",
-      "SVC 0.888\n",
-      "VotingClassifier 0.896\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/ensemble/forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
-      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n"
-     ]
-    }
-   ],
+   "execution_count": 17,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.metrics import accuracy_score\n",
     "\n",
@@ -2385,37 +1607,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/ensemble/forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
-      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "VotingClassifier(estimators=[('lr', LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
-       "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
-       "          n_jobs=None, penalty='l2', random_state=42, solver='warn',\n",
-       "          tol=0.0001, verbose=0, warm_start=False)), ('rf', RandomFore...'rbf', max_iter=-1, probability=True, random_state=42,\n",
-       "  shrinking=True, tol=0.001, verbose=False))],\n",
-       "         flatten_transform=None, n_jobs=None, voting='soft', weights=None)"
-      ]
-     },
-     "execution_count": 22,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 18,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "log_clf = LogisticRegression(random_state=42)\n",
     "rnd_clf = RandomForestClassifier(random_state=42)\n",
@@ -2429,36 +1625,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/ensemble/forest.py:248: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
-      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
-      "  FutureWarning)\n",
-      "/usr/local/lib/python3.7/site-packages/sklearn/svm/base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
-      "  \"avoid this warning.\", FutureWarning)\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "LogisticRegression 0.864\n",
-      "RandomForestClassifier 0.872\n",
-      "SVC 0.888\n",
-      "VotingClassifier 0.912\n"
-     ]
-    }
-   ],
+   "execution_count": 19,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.metrics import accuracy_score\n",
     "\n",
@@ -2477,8 +1648,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
-   "metadata": {},
+   "execution_count": 20,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "from sklearn.ensemble import BaggingClassifier\n",
@@ -2493,17 +1666,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0.904\n"
-     ]
-    }
-   ],
+   "execution_count": 21,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.metrics import accuracy_score\n",
     "print(accuracy_score(y_test, y_pred))"
@@ -2511,17 +1678,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0.856\n"
-     ]
-    }
-   ],
+   "execution_count": 22,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "tree_clf = DecisionTreeClassifier(random_state=42)\n",
     "tree_clf.fit(X_train, y_train)\n",
@@ -2531,20 +1692,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 23,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from matplotlib.colors import ListedColormap\n",
     "\n",
@@ -2650,80 +1802,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "(426, 30)\n",
-      "(143, 30)\n",
-      "Test set accuracy with Logistic Regression: 0.95\n",
-      "Test set accuracy with SVM: 0.63\n",
-      "Test set accuracy with Decision Trees: 0.87\n",
-      "Test set accuracy Logistic Regression with scaled data: 0.96\n",
-      "Test set accuracy SVM with scaled data: 0.96\n",
-      "Test set accuracy with Decision Trees and scaled data: 0.90\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:757: ConvergenceWarning: lbfgs failed to converge. Increase the number of iterations.\n",
-      "  \"of iterations.\", ConvergenceWarning)\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[0.93333333 0.8        0.93333333 1.         1.         0.92857143\n",
-      " 1.         0.92857143 0.92857143 1.        ]\n",
-      "Test set accuracy with Random Forests and scaled data: 0.98\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/usr/local/lib/python3.7/site-packages/matplotlib/cbook/__init__.py:424: MatplotlibDeprecationWarning: \n",
-      "Passing one of 'on', 'true', 'off', 'false' as a boolean is deprecated; use an actual boolean (True/False) instead.\n",
-      "  warn_deprecated(\"2.2\", \"Passing one of 'on', 'true', 'off', 'false' as a \"\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": 24,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "import matplotlib.pyplot as plt\n",
     "import numpy as np\n",
@@ -2802,8 +1885,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
-   "metadata": {},
+   "execution_count": 25,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "bag_clf = BaggingClassifier(\n",
@@ -2813,20 +1898,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0.986013986013986"
-      ]
-     },
-     "execution_count": 30,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 26,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "bag_clf.fit(X_train, y_train)\n",
     "y_pred = bag_clf.predict(X_test)\n",
@@ -2940,28 +2016,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "ValueError",
-     "evalue": "Number of features of the model must match the input. Model n_features is 30 and input n_features is 2 ",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
-      "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0mada_clf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mplot_decision_boundary\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mada_clf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      9\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[0mm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m\u001b[0m in \u001b[0;36mplot_decision_boundary\u001b[0;34m(clf, X, y, axes, alpha, contour)\u001b[0m\n\u001b[1;32m      6\u001b[0m     \u001b[0mx1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx2\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmeshgrid\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx1s\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx2s\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0mX_new\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mc_\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mx1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m     \u001b[0my_pred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_new\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx1\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      9\u001b[0m     \u001b[0mcustom_cmap\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mListedColormap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'#fafab0'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'#9898ff'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'#a0faa0'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m     \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcontourf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0malpha\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcmap\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcustom_cmap\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/ensemble/weight_boosting.py\u001b[0m in \u001b[0;36mpredict\u001b[0;34m(self, X)\u001b[0m\n\u001b[1;32m    599\u001b[0m             \u001b[0mThe\u001b[0m \u001b[0mpredicted\u001b[0m \u001b[0mclasses\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    600\u001b[0m         \"\"\"\n\u001b[0;32m--> 601\u001b[0;31m         \u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecision_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    602\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    603\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_classes_\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/ensemble/weight_boosting.py\u001b[0m in \u001b[0;36mdecision_function\u001b[0;34m(self, X)\u001b[0m\n\u001b[1;32m    667\u001b[0m             \u001b[0;31m# The weights are all 1. for SAMME.R\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    668\u001b[0m             pred = sum(_samme_proba(estimator, n_classes, X)\n\u001b[0;32m--> 669\u001b[0;31m                        for estimator in self.estimators_)\n\u001b[0m\u001b[1;32m    670\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0;31m# self.algorithm == \"SAMME\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    671\u001b[0m             pred = sum((estimator.predict(X) == classes).T * w\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/ensemble/weight_boosting.py\u001b[0m in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    667\u001b[0m             \u001b[0;31m# The weights are all 1. for SAMME.R\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    668\u001b[0m             pred = sum(_samme_proba(estimator, n_classes, X)\n\u001b[0;32m--> 669\u001b[0;31m                        for estimator in self.estimators_)\n\u001b[0m\u001b[1;32m    670\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0;31m# self.algorithm == \"SAMME\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    671\u001b[0m             pred = sum((estimator.predict(X) == classes).T * w\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/ensemble/weight_boosting.py\u001b[0m in \u001b[0;36m_samme_proba\u001b[0;34m(estimator, n_classes, X)\u001b[0m\n\u001b[1;32m    281\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    282\u001b[0m     \"\"\"\n\u001b[0;32m--> 283\u001b[0;31m     \u001b[0mproba\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mestimator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    284\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    285\u001b[0m     \u001b[0;31m# Displace zero probabilities so the log is defined.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/tree/tree.py\u001b[0m in \u001b[0;36mpredict_proba\u001b[0;34m(self, X, check_input)\u001b[0m\n\u001b[1;32m    828\u001b[0m         \"\"\"\n\u001b[1;32m    829\u001b[0m         \u001b[0mcheck_is_fitted\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'tree_'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 830\u001b[0;31m         \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_validate_X_predict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcheck_input\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    831\u001b[0m         \u001b[0mproba\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtree_\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    832\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/sklearn/tree/tree.py\u001b[0m in \u001b[0;36m_validate_X_predict\u001b[0;34m(self, X, check_input)\u001b[0m\n\u001b[1;32m    385\u001b[0m                              \u001b[0;34m\"match the input. Model n_features is %s and \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    386\u001b[0m                              \u001b[0;34m\"input n_features is %s \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 387\u001b[0;31m                              % (self.n_features_, n_features))\n\u001b[0m\u001b[1;32m    388\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    389\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mX\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mValueError\u001b[0m: Number of features of the model must match the input. Model n_features is 30 and input n_features is 2 "
-     ]
-    }
-   ],
+   "execution_count": 27,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.ensemble import AdaBoostClassifier\n",
     "\n",
@@ -3047,8 +2106,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
+   "execution_count": 28,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "np.random.seed(42)\n",
@@ -3144,8 +2205,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
+   "execution_count": 29,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "\n",
@@ -3219,29 +2282,17 @@
     "designed to be highly efficient, flexible and portable. It implements\n",
     "machine learning algorithms under the Gradient Boosting\n",
     "framework. XGBoost provides a parallel tree boosting that solve many\n",
-    "data science problems in a fast and accurate way"
+    "data science problems in a fast and accurate way. See the [article by Chen and Guestrin](https://arxiv.org/abs/1603.02754).\n",
+    "\n",
+    "The authors design and build a highly scalable end-to-end tree\n",
+    "boosting system. It has  a theoretically justified weighted quantile\n",
+    "sketch for efficient proposal calculation. It introduces a novel sparsity-aware algorithm for parallel tree learning and an effective cache-aware block structure for out-of-core tree learning.\n",
+    "\n",
+    "It is now the algorithm which wins essentially all ML competitions!!!"
    ]
   }
  ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.7.5"
-  }
- },
+ "metadata": {},
  "nbformat": 4,
  "nbformat_minor": 2
 }
diff --git a/doc/pub/DecisionTrees/ipynb/ipynb-DecisionTrees-src.tar.gz b/doc/pub/DecisionTrees/ipynb/ipynb-DecisionTrees-src.tar.gz
index c02d536127cf72d495ba4488adb8ed1b9f07548a..1d7d28defc7119419a8b41b49e1dedd52d4833de 100644
GIT binary patch
delta 93109
zcmV(vK5J@74
zk|pPybIy4Hk(@IO3_(OBizLaCGb0%W1~33h&O-*tISy%H;0^je_k8EOAOCaKdiT8h
z-g>uJ(+%CbYwxOE^{cAAyElS@lY1AKJW^`bYja+!(2+6&Eku8{T4K?`5#O9_APYc}
zh`!meR+@i7kSXFVK2$Tm6Awu^u9*I#udkq^*==FB^{OC=#VDy^XP3wxR*_#n=M%ls
zo2Eud@m;HQgLusaVH-2`rINAlDsF0%t2RKqYqPhd`h2@`K&<;1a+w-1H^Bv8H{P~~
z9U1*JI(<&rTQIp_nUIjFMsb`!+}3@2fQzkJ+N6KUtCpp}G$vaw5`a}#z$SM->50@G
zO4R^svf|(b1%WWuPF`34(W~RDh-k6MW1{qJY#_cWO6b_)MmT*ac{^1CmUku3GfI%r
zK5k`d_Rgk4^hQup^EuC{Phb)KBwL4xs?f1=qe6d~!h!-#V`Jkubj^xxIoZHh%)yLz
z@7{kM-9~`6A;tr{-Ibyl4`8$eMwOtqrJJ%z8dna;mSa4mZfjObiUKY!?#6Jw5{cbh
z3+RFO)WP|}>G5E0u~oDW5Sf_l9|sa&S?~IEwQNl_T+SE$qQt28wo38&Et7Rv3Rgs*
ztq^`{eRe{E)TH@fomDTEV1k1yC1yy#gGYa1LToX9(Z~6DLws}oo9rqE#y`5ca+bT3
zhwB`t*0KUGHA3(S{pF<*^5tv;mn*dvjTLjSif&);O{e?sk^ml
zU~3(CrjW8*SNasEK2yX`vsAadNDmBnNUIvSkcJ{kpwxe+
z_i=9jgPbon-2ZWA_wHXayR{cQHmydrt@zs~rCwWG*IjJdE!8@@OJP2DH2w*!gEKYo
z%w=uh>4)GqV6m$=w~Z(H;ArE;mCr(2GV(S?*jjt@Wt5VJ&Uc&F8N!DT+1G^HrMmSl
zrJgxGmf?*51Q78uBS1&LkM=M3U}}Gw`H`@I`=dlU96>=?>FGT7kMJRGq~!)pvq6tT
zo<%M*UuJ__GFEgg0sfZFqfwxGrv>Q+*_GVlEG@z&QNoET&osfl?ryEl(^#8t9V*-x
zh_v<9tiZP%RQMQJ;%NVBr)I!y`$UyscVv?PJ@#^;J#u~FGQ?)RYPrQvJ!pT)U*qI1
zgNR7T`*Q=(6A1no$^JAM7o2GR1;(h!^~u;9l#dAqqf{O##lBg&JVi4eOgiURfI
z&_I@g3K7IWjqPxe_wHrx
zeGfPwUgPRRula&-<$|?3DzYR{KqpDI$l~`US>eB?U9Nl$-U2Z)}kQ6W$&G6CqR8DL>=cLgvD4XiVnn
zTC-Z_av8v@V!hj{);U)WWKipos;cPF&-LL@vBAW~Q7R@r{zOwUn*6bx86V4;ADPKO
zEps&J)co2pu&nQk;RJu7{p%e|0G$*bV*KT@Y<7RK({`fDvc`Q?b+X2mh{t{`s;k4MJPm)(sa}N|U`woa^#V+kO?Kiphj_n!{W{`xvNJQ%0`eQJ03L0~@QgS7jmfCfT=$7Bg(nM7
zuqX&I;8;H+>pe?Gx{uLviLghk<;L19cO~Xma5-V!L<9XkJ4x_i3rmjpwiMQQv!$SsX5Z^I;gw>ZRqU1PP
zldUFt_6BjXYx+or;fO1l)kHB@Hadq=$V**XTDth~OY}{8jM`VIgUApo1DJ~Dni%o)
zY6U3$#7^fhb%{v(go@yXzlRg4D3Y{e+|1YB8#JpmKgoZ)wo(z|K%ii)&EV@KHd9p=
zqdL4FkZVWE3R$C>Yykk97*-2ngE=BRJo~{j+0#8n%lM>@HwxbpkcYJ4}4H=oAZmp-*my@gCL1He?f`x3wg8StqG5<
zaSq5~pAhm?+g*60ZptaQtn39!U9);Ls1t7T$EdxH5xrsBH3FL?Bt;wmcv5RS{I-8U
zIYlzr#LIwPZ`6F|5sP8oZ7xD0B1U21X>&p2n;UeW!4K7wrpJQfxLxnsI;jGcSqy7%
z3@lpAc41&D&#mz|(TlxroI@B-?*5GwS*A?q#xvA4(&ASyUe_qVKq+xfQ)
zx;><%RB?56g?CUm57fc{n_zz$tP%Hsvt<(C#eU~H%0qap^pFX9@1#d?v7@r4pc3pm
zH}~`xd6dzHXp|OdJqY&P94k&adzmikwM9xP=A?P{^oBmAwdLX)7=1UhF}8RcX5VRDfS-{6J>vl>Yeffkchz}
zm?=2p-r06}{j4GiE%H%L>OK2VcZp6ZJZ$`?*zsD!uR%EZfsgQ*7*?J0oYRQ}Dyb+$
zB?XhvM-lXX@U`q|!iiG9moIam!`Cln-Kd$?*P2Pj|72ARBPAup<6!A0Awg7BR31t(
zhKbaKH5LquGT&2wp$dPz;J~FYiZlDVP=WI|*rQ&DWj4TObm;>jz;&3cQMmPWR7?_M
z5fv7s?KCXjL)va}1y4COMF*U3*Aw>SrV+jMG|{2b
zYwr`fhQf~)PW}yRm&L($ae_{u#wU~&&N&fy_
zdSfsbyVhwY^*ZK6dB_bzgR50$=sj|7`CE@v6%})m0YHBR9j(ROQG>pU9Xmv%Kw<@*oN&M$){hKF8C-x{AZ0pz2e6{S1b0{%lNg|A09exX1=iaMA
zhxgoN(?H>-W8`|WMc}%QK5tZ{w&&2P4e<#7&&q8cmgu_C!qGlkBP6tL9S)nGfX#uo
z?N4kwtB5&r0X%fq?@8YP%$X=11(HiN1TK8XB^Q5M5k*+cJ_Z(9W0=4y1>0rc+%i`|
zn&y{HV<2o7IS?+(8W4EZ7x*u^EOle!bkjQ+P{ciVgN;$|TFJf}Axa#T-I)#Kgb3#4
z=WBo?AsVf~<+X{HTE;3GNjiS<<yI~=$`wwCF62!%5xf;&l&F^O^#uA*?((kX0;(k1s
zo6f%-nkXdaxiNx@l83tT==>N&Jm4Dp0_!qI;i}r3Pi&
zk`1xT%<*G8h~+Rfj^E;Udz%V-m$T-wdU+`^Ny%(YHX}3QZa)AAMb+WrZ=a_s
zyV*Je@0t9pK3b(O7=FiAr`cK0!u5;bfad|o(0PzetuId{tZJ=1TgAq#+Xqsq^g4oU
z4+(je0IK^WeQa8$nBTB=hud7A3zBD`Lmnt&5fNJ)5eqM89hJgQ-WJXm$_&5k@xIU?#qN2bq;g#HeXpgRdpF=jm=i0
zvQ@bl(Yp2c*j05OeSr;4zPdHdVY@ivw~~{A0chE47`t1dNm3-dWo3V2?U>YY2|q5c
zKu@Jot0V%))Bx7_P{Lce0H_n`kDL4>^%9OYP9r(CYbflpDWYOi%VCdEX=cwGg>mzvq%23$LqL3iiEG&Fu^;kKqrG7?#IoV
ziEG+w_r?U%Ps3`}8&Us$;)>W&t#nZ_juv#d`k4UfpN7_8z8ZghzM$Bt>ox%k2nhCU
zGgfeBE8m9Z3smSZrr)CyFibTn-w6hUzt#ZjDt4gX7?Cwdpdwa2>RXB*7E=lts&e)L
z392hA%vhBV+I{sRU1vRVJeL{tgo9@pyH>VzlZ*isU8FUw@iy`Iy0`&-5lSH@3oFBDQx@ri{$&
z5gHn`-1S-jC(3F$Xr&F>%lrxO{xst-qd%h$%7
ziEP^vR5t3+x5BP|1|*p^rwE+YP&0;9I2fszP>e8TKH1SqrbA{(Qi0|3f81HNou2O)
zDXuv0_o#O;6#V_2prJ3^XiZ1$TQ$gZ{d=`0HP1~TTJ(UyJBD_IiL&e{8ce9v-C2{(
zq|tv7Gb~R%f$F`gz;lM{SHLefx|#KOU>DxZ+ddUdoO{W%T8wjbH7`m`1kyt`-Zs7H
zQs9N}E=-&|j)UH_y}W00>xb3cCFR_dd}05RO`&3K=rtDR+u=3f-h?Q5xTN-H__hg_
zre~7JxVw>qWGO$99bb`MqkY53&G?>*Kv#b!&RQ^~)=GzeHtX`{{Fk1~t~lZ%C6UN}
zkg1)D$5X+5-A{TaIJZ*)Go{3u!8eW?as+>flZphJv$f$QSSU4>+f(G0>}|a75Y~5z
zskpMEh`z8B3g3r!rx}tZg49=@LGEYxFb|a0V?{&f7@*`vOn;}l>{4RyY8-;L$De=V
z?{kSA#&{)U=!p||C+k%&P|~rshgP43)`)E5UB{aoTe)KT-*GZXg$D!nNh%T~Tl80F7f+jBBg*tMTa4mmBfT;{
zWkxd%*V_*lz0~)Gv|3YkF~7)*O)eZNp7tl7&3>!Mmhy9Jt)+LthM1y9Z%3p!^Jr*Z
z)%X_+vYV
z$oUqqwnoO8-$^h*hRph4{anLQo89`Sk}g1)k3OX?&q_n+-riOPX}+F!R+z_)>B*-D
zLEV(m((jzOst;{@)PyGVO@r~zju<2Gk8%rXhkmB}cd_xyd8?`AcshTqwZ`HfNf~y=
zlTNf+EJEe)H2pjS`3b53k|dLPV{^Hz@2-7G5nA`<{Qwv#BJ?aGQi{J)&2IH&7k@R{
zq6Pmt{VH=`6+YkY$q+dlyF}e2r<$PyJGufjU%c-sR9$v~G|4%B2o7Zyn^7)oFz)Lr
z>zA6+yX{K0_4#mcXsv%9Y&!8KptT-zW~#dG(nEY($l_rHu)K%spmTp-qx4M?jR;-g
zY7x?fs^L8o5XbB0Avf{s5l`x#s=gBL3WG<7cz!9x{!*)08-3NgVYOgYXV&R{*<;~|
zSOGcw{{HdZ9}=!Iue_k7`@79W-0||e$b~_S_WhD1~w;|Q#8bHruJY&oolM_Fzvhu8B&3WUt`wY?cNx=>b
zV1r^BlYXvcHe16`#J=*m9BZ*0ZsJ!RD1EGrs+beM;@oF+?bi`qFZ^7aa>IN4$MEt;
zbC^qpk*P+;dEaH<6q8-O9Eas$H?2Ovtd&PL$lJBCdencU3>l^($r_s_KZ&1#lgW^&
zpXI2Czr!_Hv#L4m*Da@7Wpqj&*fS#!Shc}B>#8i$Tme?7@(u$Ea@b%1w4ULmO
zH`!q2?b-@5mA4eYw^)^KyS_30W65yvNp`W0Afmo(y-MBVc_*
z01@w#%B`Gp%Q|>=ymfJYAGp!@A)gq{428C?Aw-#GtD
zro`EcL-E1NfI{0^`PA-~gAk2$jyp2YCx5-gUAxFO{?XZ)nu&A@ed8uU-Z%ao1A&q{
zU_gIgufW5u3(!`sa8$^w?xJ|T_4$jl^O!hYsPJNWPUx9}gw40NM`7n9`2;BF(z_g2N*tew!NcJv;uznImw$5N8o+;P
zqn;66vQ4H0laHDx|n!wu30wW8)xZv
z$kik)oc1h1wx-I)NbQil3nmCR9O8*71e$p7)o9L51tuDob8N=cjm%(%pj7>+M&zl;
z&7-Rp7TL9!?(Dew7&^@D0}SFZ?%_ftwj;$GNNbOCa}IlYl}cG-X$SIYfCQ=wW18~4phQjI~IhvY3u__5Ph?F
zJQzTcFgJp=Fkubd-U`$rAq{npM1yZUqI-Vx^dA`4>}?ASM~XfqL#|(4*@twr_@yg!
z(Y9Alc&L5YXsbZ1_lpXQQaFFVLcze!wGDuDt=@>SKAP$s2#F8lbe}MJ-#*uhwV8<4
z63jZltiis)GcZNlb<;4W`Uf;T1Z~QE6eb3NOVFefK9ei;6x}_|(^a?=Z_(W8kylB-
zDAIW?U7T=SFYv-<4p?JjeH+w>V0T1$aTcPS^~}s#0h6EeC4GwN{lR~J=V}x3u_vW+
zQd2s53aDT~fhhTL-ptli;1Eym+GGq*y5)oWFboB7H7{lURfV9kP)vIE=~{$`@1_s1
zBoIW%r&=NXHB?@FwQ*{Q@tIH-WGptH1e-UCr|6cEwEtL0C_&2ZDq%1YWo{
zx-I{3eP<9>e`_l=%h`Wf>18uaT4**sRmL0nUV7AmX$bpUnubpKMQDu$ZxRGW#{Msrr%n^Xrv#CCIR?oA3`
zk7`sZVW^m-D4JfStvp&FNXi?)7AEETuuBmrV=dlW)5PBg21ieAuFoZl+#4vB@7MC0%eS;1nG)1Moq|-(MX4QrghymrY!!;KN5~
zB^3@7In~;DoETm&p$6qHdP+IZTYTl`BBv+2KAK_u6#qsDSVKjt;KBz>=jDVsHqO8p
z&PjV6r28taq5C{{EtshEVjwo7Hj;)Zq0kzeJFi58cx`{(^kkgpJJl$0hu5GMtXSVepERnqI)}k#G3^6)L`3S0jJkRtlP;CGcbt^s=TG5J?jpi>_3h
zd=p(CxWxcSG4teTBaoBu?P4<@>feXOq!4PK}=m
zyng=DjSP9;UyV4@44#tViK(3rN2LpDnc;sme>0nb3J!9-=DT1J?OLM{MTZ)KLf>$*
zgl}By%Un^>7c}m5l!bl;^cwD&m24FXJDWOTnMlW2$Vy+}OEY5kb=(Xi%Kbs`Pn>%U
zQxK~fFZ~)vz_ZjWg9CpJ3<7Vv&%wI>xtp|$%A&PZhYld}+>_AIweLp3fJoecs)c|1
zP|iTA^ER*|*m(9Mc@;CQw=~yex0zO~Y?9D-OEo$I;acUrvTZU{%F`9ieRIk~j~3FB
zDT}k!{o?W$rQHDhNG?I_{E0(DvcVb6^_m#64%A3@Ww{Sz+Y8!#aMaVfnE@y0~B
zKBbgO^O|ux_)NoGnB+zO7KQ1Jx~YF;2w!DkRbG!)A+%)9RVH-$jE}8%!+(7G(jir-
zfLEtUFi|lZ*!8SlB2%~@&sItTEau2x5_NXH7+eRuV)KEQYvaS8tOa$HtjEb${VnlK
zvAR#l;UYkG!kaKUC@2`!8+`Yz46K)9x+xhK_l`$x59S!T(~yg54G>$6N_&5uJJwlC
zjR`Bm!{ELMfC
z8?||m(!d>x+^5p&XI1nGSp5^AJ*pS0gDvEUIOkk{Gu47HHT@jxw5=Q2VhH*;Q@y0o
zx}u|VOo(+9ed~Ok_1@T2lLCM7lpML2LxTMy6ySdl^U5m~?;TRLb`&(sD?|kZ+#jbt
zaY1Fn!Upfhtnc-vep6h<_Zd*ILhzM8Q809w&cQR;La8K4>=(;gOJhk4<9hijSv;GtK{wtHI9A|iot|@d&QwmgGNWQQT^yOJqf>vq+CJAp%E=Q)
zU7xYh!6b#eViA%~uBtS)lCVGM8z}HNfm3p?6lQE_b7NBxlQ14n&(?s6lnyu)pZjcl
z18odFe^PVmBU6%^BI4v?p#$*ka6R5;3l3;++l)!5E@>3<)*m8WPDQ2Yd&a2vj$9gO
zv%<
zG1M9^r}sk5b#N?LroH}5*ekI*3>%+wqOhfB4gCG&R4H>~PP~5|ZMXos}00?`c8QX7rk0YN#
z$S4j#v|=c4E#iN>ERJ<9OEMVK^hg!JYJAh;)dhoYwU)#5=ocs-_IhqQYT-%y)mNx&e7RpYV5}2QtSs0A1;1
z8?Ax?EI_Xg!-`v}0FzN{*o~u3ZAsz+ypJG!qvOY_kJDnr{S);KMGNqC!kMN*NlJZh
z`>6&z4Gl}A--qrp3v~CNhGo$b_Ok{q2n1C$j6aAYMJG(Je?gmudDq+D`B{c&{GtxW
z^+z0~deMIvH_!4t53Z`cc>COKPG|Ck6c=F)9FL-!qM8b6I&R%acXn&F=qe!^?$Zk7
zt}*f4uIb*+$smI-Z+x@k4d{Mj^I;x(J=*ivs!c~D1{=Kr`(Ov7XgXE_`KAfyt33PE
z!CFfDhMp-HZqz{d17A`NH6v)3{(RVDAD^8%kFI|oeT5ZA7VGW7(ef_|8lL?_^s0f$
zne1*hRfvEUK0bD)x&!nKdoXnKG(skt@y?np)@gvi?jo36c0PF4GcxLok>1UAa*`Xe
zJhg3i8^qdQ=B>wDXf-bR5=3592&+G0p331GKyZU_e
z6(@hqCVF<@l3ZWv0fYv8ejX}1kDl;v%Px)oS$2u@M^l&eegzSHzQ0LZd>Qv!>R@zw
z=SX5P{HWFTX|A4LlndpJHtcYnl(591iSl+UI~l!6(^MPgvmcVTTJvKvEmO3_biPHH
z3;sddTQZdH*qQjNp1!|}ujzNA_HGq2c58pxXxAHZL;P40L6$L&>=&&@w&t2|4OxC|
zr6`Smnqy~uT@dSJl;ALtsf>7wk2pCl)AHym6S5i&H}K6izR4OpP{1-r8dLCNc_hVw
zbJ)QNziYX?G|vBCmaviuec7mLQ|%1sMHuG1;rTBTJ<;&30^0ja%BmHxcJ5rj@OXdk
zU`lnUN1Q5Obg4`UB;qcmYv7Xd^Sajn@U75Q8)w;j?Qg;antHimQ3Iv6Od+pte5nNQ
zI`JUYX=JsjWlEfO^Ij07Kg^R??Juz77+X|T8F%p)X6_db^-%xdaD0mm-4RO6wZW6J
z`0(AD=QD|S7GJxY^3x=?=xo`~dq{r-GK|CtdZ!h_IQxe(NDQscOYhPl`x&6AH9&?(
zHz`#(uD+3?in02`UY_7d9-=k&{&dx%C4jmnOdtsqF@51qYkc%3k*)A`Q4(l9Jz3mC
zgZayOTeHOn$!h-u8N0F6IMS3Plsu3W@%P&0kaj;Otw{Doq_kXcbJGc0w9H
z8F_=dtz@Bgvgb&$cF2QzffRvTBb;AEX8cj{((g%HE#}nA$HN@dQD-^Ix$q6oZ{oLa
z(t)RPXt$Bnnr8#{vIFTe?YDo&%!&-{d8Z@Qyf_|3{_r}w)Eenh$#pg0{MaNERzWWM
zb(7|!26*9^p-R$s(yM_#>5JI8Y6%3Ao0i4K63TsS$y
zk&M7iVAvytb{llK_I2|E9vSDG55w-ZN_I&w-!ra|$}KS@vK@O{)J=c7kUZgQ%jk9B
zaa^!EU>~@Z8?`080c$Rqfm;bQim*g}h^dmeDk*Jd{XlkvxZ0j^v@z0o+_`URXSb*v
zmH?r1)MM}+)ZDA&)E9cSmXOT`Ph|7wc(Y!|H%#)HAtKkEN!gs;h5x}@92(5Tr#7Ao
zE)}_(hQ~o;jwY0)4=H~*t4qpF9MgQ&RvZrp^2~i93auoQeouCwtLgn>U{$9+19L~f
zxxP`f=nmEJuS30DQ^-eLgm}5!@k+z`0%H5@^G-uAOIEG$2#K}SVc<;bo9g4PRZKTB
znE!FS^|K%}3>@nJct|KjvtLWrFlJ_EK0@@FOg0uOVP`pcZT){Rc56S{{xQ(dL1DGn
zwG}lXd=!dJQFGF%0S&x3Su^@&LXqa_EOngLJL$Nzc+nB}9xyYQUs6(1MzuTA&~oJC
z4p&OD78M|&%ne*FM|fzZ^*q5|ktflscIa)Wb-Fz6XOjijUd^nRr_a5BvAPclB)fVJ
zHLTtJpIS&AByN9;!=7zQ?%LImoW>_sib=Hvo3*e224t`;p5JlJ?oZvL+%`-b8EI0(
z8gXCm&vR><&kswhDX#{QfR&5Ocf5R=GLYXs5S&Y<7L|-9u^4SXeFgG*XP#86&K_PY20-1TbmTq|`t
z^A{cUF_xLKsc~Ln8!dC$EI!wzJ>saG&)5wd&ZNf-Z+AUQ99!B)MvuB$YH_&>`Y`30
zUNscxjHS)2zP?&}QyX*MbG5nMX=41T$J$zQGVq`Tp+=SHTP>;=7QdKkhdftafPCzC
zi`aa*Scrdzj&)xg?Vk?8+Nd@aeqj7{0D%bJ3Xv#xa!~(nIBPYGsiao2>ejFw4Twe}
z!7X)TQvI(aTm3d?+~x>d@zm&$DVfj;DKQg`#3^Es(1oTCQ>y8(HcPdKQg$pKL9ESO
zr3OAe{?FZ$)n?xX!%WxSGwe`ont!E8v%9DH
zc8Wa4u(D+~P~-sVApv9G_sH@Z^hE^^Pr85`H!jDd&*547_CS0wOSA)
zP$ow;@hRDKD*O0&!QnvB*!g?=_s$r-k{W*#DnDMBAV)=k>Er*oHK@b$JM7t644Nxv
zKYuD5!QnP8F6vMy)Xvs6Zo*~sah$v~F){IAb$99!cPYn&uNGBW-KL5!-MYs~8sz)4
zy7KRk@nb@8WO?^{}j;9kmQpbmTJstWcHfeJYb6HylOLsO_Q%fgzOUIWE
zmd;+5ZeF&Q-mET8*8eN&U}tCN<>r5;zCQEu^87xtbN>1J_r}M|P0hi@!;31#&d$w4
z&CZFs&x1zI{=cH~|7`Pg_b_)uHO0p6nti7m%b|){S^W*2KdeyykMsXmwp)*|(J;`^
zRLm`?)wHPfssGxpuc6_hemK$4(7&S2sK3$S|22{l+S;0be*68z9_ozsPep$)|6E7A
zkA_BW;q1t2?&9KL$?9P4?&0Zf`RbLqhvhSMt-oko9~#imX#c&Oqos#As*Jgyy^6Gg
zhoz&QjisBVtedl=`*nF#Su|lZG~$0R@9JspX71!+>tu;mBlgYYnjkhrnDYAk(#65n
z-GiMy2OY!ozKD&7hl{%)8=HTlt(!BdmRGi(j;ts$Y>wtWY@9qioNOFS{My{|GIl_D
z4`)+deswn)S!MfI0utAzVtQZhe?0#Wk2<(nUKT|^^!wK
z$_7QuQb>|VQh;5WM}mcmUqXO|o1Ke?MM8pKnuSw>N1B6IQj$|vQu03#`gi+RH}{dZ
zG=F93#`eD&2mUkrXXoPmC;R8%;o(4|_W8e4>3?LmbNrM4=iuXI=lnnZ|Nn_C
zTwPTT{|@=JrQ<8eOQZZY)&mYyzZm^LTZr)3{q#}y}h|<2=nyx9Q9qzaWTAheyqhRaD2e-K)-OeL
z7xNn)7kk@#w3N1!!yYQG`gFgY6u)1WE7LTin@86bq6Pi>5l8pCVfTT*eIGUa_jU2L
zTfc7)-2Htso8v)<3-$gJwZE(PKT`W|VeKz!{{`0m
zqV~U;wSS}bznOov|4zmDyL$1s8C8Ct>vA@lqjEqKmU4EfObK?k@LvbGyST(bPP;IP
zUU$-kso|bPW-B;23P1sB%-2tZvldy8VDEq|Q_R*$#i|+mKSLIn|wQO4#pPm|zs?
zNE7D`g=SJ?1d0aP8d;ZbA9Q}F7j?I-&B)uOR8x8W;YcSu+>e>R$kHT?U^;2K}77$)2{3|qlJSsauqak@o}lgECb9_sjluj)u2vFk
zYwOpyZwm&mDNz3rHfc*`G9B6$Rw3W-fcY&8@VY01&^vA=1tI@5vyJ^1jpw0Q>}?TvlG!e>+LY6DtKR%S_8&hz$D<
zy`FV|&tFXxsDz2F)aeX=z~5bL-%pNRg|nNVw!4F)?OSm)DRRUbez~7Jm`3kM*M*Jg
ze(-;l_$>oj<(1k`qsO{Zp@ulg$J0exS(HCDoXbalHvj3=kfg70FH;dHi!U;&>|Mw<
zM}58GMmY4(&1bwx1k>8q1y0rtAO2i;}U=&7{FspT+@=7mYVq|Jt1
zY~gZ%HK|jElH11G2!9qeN%;NAj;|j?(yIA*aG*=tFk!*d90d)9%bh2V67o}qx}K-v
zm(FLsBD!tupo5-7Uoq!zD0nAF?ssjGjT
zpac{6&>7oe)_Aa#YPvHqHsCqk!&uL9HxBn{f2h!VO#C`L7gUL)wOxou>ocS&$2>$nW+K(_LneMAn`u9G;hfkmmr4
z<71=dy)U?K%!xJcHk0z(pNk%=+rhq1rG$M;7I7P#xlF2l0f4`(|33AlH?V&Xi?8WC
zVqzO8sdf!3>;oyV7UulHv18+JcF{6keEa*n%O;*6)nBvEmVH^sSNA?o48L5Aw)o2W
zn2#3O*0K|77y6zb(zK43;S_NZk9|j{;^!a{D2kE$`T9mUb{!Ra{y;+*+r5*GbSUGM
z*%yvqqI9%OlkHsucw>5X?hSvtj^4_uyr2@6iOQ$;?yiyVVp`TN@*YYB6M_xum1mAd
zX1CwkzIqx6VV57{2&wSjY}w^3iO3N7P#A0QerwmdbB^;j0=E5#f?DxJq`+gZl!o+E
zIy7)H^nxNsC+s#2AAT+Og?-@@?VYqIhn_Fh+q*dhtW4j<;|i7TIT3&UTtDUeRg&h7
zF8%WuNu$=)XsRpUX3#Htufl2I#A3M9TZPwjGG|o$w)~dk$mXT+46>u|ezCi!6^q~c
ziuqU?*(Rjyr0l)#8H+v}^}u&oxs)>%OLFqb(yR7f-0!A>!kMU~hE(+*ReKu&&ix1o
zOcx*bww5(*haryZdBT5&)dCh4{Q54aS510f0e&O9_*$(JD{BBcE7Bw`LJbBuD;`x~
znjJT;oj1NZ4$w#m3mf3mPb*>BA9Da+geCWMJPGjbKepDi_RgD~f9UEaH6=91tR);3
z&p^-uS)A$<;P#tE9B1k%Wo0eU*7Cq_PYd_fTfDS9O9oR3IdFgI9FkDw^+T9xAtHyJ
z{jmrATv*P-&KQucX<+0*A`eACvH4xGG#8cM+OZc0yJs3wPcz!v)u%i@ELnTze6gnq
zUDlTwT%mF7^zrsK{U#o`i=*|NP4HMn2G&mO-*9X&Y@8ED%#9B}VNWa=Z$4k#NfdKk
zBGI~+EOhRHmrQ@E*6bD4;
zm^qKwK37iSQ?+{gENm|>L}x(>I{}IO$0d^o+hRKbp59ViMi2ggcH`EC08}$P7GOH~
zOX5}K35X`@i8gD{z>xn
z@<6xNC8xvhr4@ZIl-Y_HEA)y-3^{B!j>0*s1>k>U-NRE2{Ltdu#zu*WEJZJaq?T^+
z-fYT#2fK%TBIS^}7uAq9e{qeSyRj=R?@3R>+&lXsPdG)$_20xI7q^cLz$=ej7+HQZ
zNPUz!%EJvH69QJox~fb4B2jZf4A4Z&0+L9VEv0?}jUK^rYzV);XEza^nUi`D@M}2lJ^RbE2F4@B1!q$?wT6Fv
zt%Z+aSw?Q5)~&j^3rE{Li3I4!R_e4@?@*w-Rj`AB~VUZ1vp
z3Yn#t5GE0vYHo;_mJ=1zu;dkA*&I#rspY{wPJ$hg!&s7a4kx*c>QcOWAii(trSqP@
z?Fd6Xd|%RFW?;V}ztTPW6*{_wW)6J{J;1S-jV+(@RT|Bo7$y19DKw!?b=6!Ajf1IhHzXvqLzQzW1Z~xEVpOk;n;^QUYCg$X*
z4xf9tIj6MVXYq`x5N8t1K%8%9CtDh-K>(>&UFrVylGubmuhnkHV@Z@4}Vt9pkvzsWtVoB?q66vn?18;!K?uMqmN2=uS$v?4CK=k7TEp8y>V&j|h}F^H)Ev;hYG>8YU=i@C!e
zG;INYIe?E@71V0CP{(
zAZr^LrKjcSpy1Xq-X97>*N^}59qL1<+
z@OZiRdU=J~48)_vy_V^e_kLGrW-2;)tZCw|V>~c^m#!?22l4FKd-Z~ahjaQx62rY+
z-;2cDa~twM!EV!M)6^chrbbA#DPghZzPtN3|)wgAgI=VW=B$h$E{^Puj0
z*9U}86QXUM!Jxf&RzW1qX2%pEhN6FB7
    b^@FntSm=GdUdV9&vmfm5SK*Uic@7?-
zFk71&m%cGlTAmht@#Rb7=i9FBejYW-S%ci4l~kWB9sG&w-`>i9P(45Huuj+*D|WqG
z(4Hj~0`E>3e>i-eD@5hb;huH>32ye`Qm^G9TeChtS%=NYBQD>bjMKI#G_|MP0x4jB
zllS)J^P+{e#D;@XXMdu5c0<4UZx>yQEeD@a>76ad>zLV=Td~J+-px8&rl1~|2hJ8=
z`+`qjcQSHITD@a`Qt7ojJjP2G^^%qJ14PHYI7Q@*4ot0b_tuIe5&RZyke7fZnOHt(
zXKmg`Y6Dr(xAST?z%AbUyXv4ZD6i+G_Jp5l##OoSgw~{{hIh{B!75g0WxQ^&&e1?0
zNOf50uXf$f(HXtyD+pXag50G85K{@)`R(L{BZhJY$>;2UbY_go>KLJ8lg2r)Mu*IT
z!oMU${hfZww8EAt_8uu{@Ox=_)RV2AO|yN*@!cGQO0BQW7q1APzL6c&PBf|V)0;&I
zR@*isSyLZ8UU;tUS$kfe^h%#im)FzW<3rl_#{B((x38flvUG0%0xWa_>N^vUp4ODp
zCSso~RdhRl*{!vVaqh|#|Kd=U)G>x11tV@hHLZDh(HQ5Zm+i>FFnQ*#;rmitJ)ZOV
z6FkWlxuG$%`A4t*qg*NZn
zK19@w#s6$da)z!*K9>gY2>~U8O0{uR9D4^b4YJ68-g=jTmr3I8`!3fEyfSS}(=OF5
z+K6O|(P7fDTHGWccF@`L`|;j=VAIwUC6!N9Uds+X&cm(7>Bdwo@IXsGgti5)q#t~D7rCEjkRqb)e=}~o$0Be!mOG{EjclYoU1`m1;vD(ExcKUSsJ5RcT54Y^6ly^68g4D_v`|X%;%>zW
z5Zr>ag+g&FZpA{8;0YS6NO31XAh;8PM6kds?eCp;&igy(>`r!PXYS0s^C=>fmsDR3
zS{$Ggt^gXWzJGWbWMCyuTo>-$oHE}i`GM7G-l3nGiIk#i(}CRerJ)^2s*ZS;S>oA$
zQu^J8$%yZ?p?x;dStY_$qsQ?)sr0s7iW9{z$<^KmEcVDvJ}kfY?BzjKVX`!p<#a7%
zl!FZT2VEGjpkN&mMUQQm+
z*Kg(-e409kqT!;R*`Ec47neJHy}6zi8@z-hOFlrA!&Aipn($6v&ZVj9@vgSiwS5JDzB@1@TLIX%lBQGpY2&rU_M$hwizyJSFl(!KOB7Mj|ZY%
zd?<1bt97TJQ|8YC06Q}1O`
zD=1+@@LXKIY;iaAvrD)qzGr{9Y5q&VxP{i@*ciYQ*9iD(%lX#j84&on*U@Ol($u2#
zu1bACR~D{OOp5u!Zp4QN%))_xJR#?e0ub
zkm{}{G3|R2Oa<4=P+&T&(oIR-ftFTxws|8Mz7k{l^6S))=eI-ZfB+w5`K54%K*#YM
zNme>($?-Z@nEdGz!b&~r1jD6?IqMkds`Ws|?7RxD)^?*xO>%W71YM|q&RckAYuB%K
z=d}cSc0!xcOxNU1rR=-Q^>q?L-f}#e211U1l}*CzNkP{qh1Y5r?LilX+Vhcs81O)W
zsq}u|7K)DlIWZ>^+q_m~MN8LQ3X3eG)Ce0MN1tG7DXbo6tu%Vf%;}!<7cJW_F4$GJ
zHM%0jhlsrI`OZ6rV+m`2L^?65WZW{rt{T-M$(~D0SUURL;-}L|?r!}r;h7~gh4(X&
z`!6jlc>8bpJMkRw>r-CP+4$0p{rq?_W*vrDD7@>%U@H*dtZ49n`q|rG!0zg(pcVcu
zE?B=$A1fUwRB)nE=rQB+!?Wbn00t%XB!7FwCI;Gh(6*ULH%Os>Jr^dYr;Tq*4(UK9
z-R`4xi;TH>dne=V(#OLzBp;$E(P5hLK~B**x{Ws~taczv*ULfB|0X$l!==8{cs&WPK}<`n-B7hWmH$se@LFP7=)(t9xSba}|Z>fKk?^!C}X8j9Xh
zWsQ@G#XK;@BKj78+OmsR5N!0=!>M9=yzSN||sb^@^(GPPA9O
zLgKq3nqkGoM}h4>F7*hXr2HF}6N|e>-(MVAX_nE&k58n3PsNffBf;1v_mT)s?pXwv
z*kG4O@@cI~o_=-9QdegPqcNefx~Y@aqBuQUq*ugr5o}Q(a<%*2TN=_{uyXT!k-8qL
zSb~aBL|$u(tX7Usp!U6do#M=$~HZG`ZafOW|gA&+fz9
z`Etx-O-6MGqL`C));Q!eK*@DGvo|P!7h9OFXhNNTA(gSi4hQ1<_)3EBtxIh3NwV0n
z?93J!7<{iK-9@RwE7+0$u6DCJ;t`>ybTUH>vge3N8n@RD;Yeh-D6m9X{&GvYnn(Q~
z(AbdWt=7l&`T?5F`D}w#!5OL%n`Vhlsioq`G{Oc?xwqCe|uTk;gd?jju
ziOk0|Y8{Cp;&h?sRe2jR$)UCLUzLQ!CQSVdnAeye4>?X+L!|XT3Xbm1pRJtyw>WFD
z!rk+;m!=5$yXL)cwc3M={0=uRRyIQ-F#)oFqx839*h)_HJ38DM{gqYLj2I*;N#`Cs
zLG&to6pJK~$KLVJoEYXzPUS07v??C$!UiRW4hbs1^69THsr6MA$U+(g9w3(2m~B-s
zX$QAq^tu^a6r_cc12lDD#G1iC2FSFuqqOq{mGpn(-<^Ce3R%YlcNG3@d;Unvs(kK$
zBIh1?OLkz4(Hi>>3yx}X{I)C*607o964PgF*r3KQY6aUQA6hK{DQnwvpGhf6dTE`e
z);8aduwfCU0mAY{Q}6{o1N^tu!u5c%HCf+N4t}*^HT6$-RTj-0f1o~Ig>y$0L31)e
zDSz-0j(G9_$$C@Z9QMKG=rgGoI*dGjg?fo{f$OcN+hSO;B-GwBk3UA9>M}*?unC6O
z3Q6Tn*io^V#y^spF561V0=>8MV-mxIQ&s1{&hV|?w`2Tay`I*o*`OvHw*c*KLjt?-
zkdV}$o83StM(`=p-Bwl#%XHq}_*@KcF8@$e%dXz|+h5HlZLFkG!LLrO`(gTjHTgMj
z^Y?x}t4W!n0HO8LX;Z}|DD}T0LdUQ95Ep)u`*J=-sSx#&WYN}4yG&}l!o48}qyNC=
zF!93ed>1v~!BDGsaI5(|>rNSe0sQ14K)8f^tD4_`9ePX2m@A_6H?g2b396Jmui@YJ
z@mJ4~ssf?(3~|xi;?>fbXDiEp8q<96W#4TTN5^=2Z1oivQn5d~X!5M06P>7X#dJ%U
z;Cdn4Qfhy1^ys@;OK$EQl$Zj>&~BY>#0dHIOPNdvOwTz-VSn}|H~Z9^N9yF=r-i8I
zsk0XYswo(aiH_&yc=mtg1}}N$Y$fqXP4&S#9ks-hL=l|$0r|;5x)N%C9%g><^^SP-
znpIv|<5FYdfYs`NeZ9@6a^*?U93f1msr0TvW-pMRQmMPIpHvX7d|Cj4N>86HH}#1v
zRgSy$GUz(mJF#(1Zf&n=w35FFN6Ufec6DO={Nq#>PO)qkcI`
zx~!9v9t}^(|HUwOJh&&Aocz^qx=h#5$@=G_tK%VG)uf<~*~fR|jb?+MUwh8=pj-?y
z5K~?|`75P*(L1N}Q^+-GLw^)-^)LoFl`BBEuuJob!`@Hh=;sH2%_T}1w&E?&&qAf4
z60>FCEn~$abHb1{F3#6OOEza~UBG{GxU)-4Npc`i@dZq;H(Qf)<>X?3X;e{3R{pf`df#2d5l3Hf3`6+<%v^|KdwSspT7*&ZEGa
zMQ2jUNR#z{a05Beyl1QFeAAAfESOtl?{^+Z*eiRn-Aej`r(>ALh|tMOeK9PeXV}&l
zIMsg+@W1$*UrB;;(;m+6g4Axh%r!NtiSCp42*Q)bnNtgxQ;AuSW-zqbbZlMQVX&ma
zZZv5!K54AvwJCFrZdcN6LiL}m0VeCQt?w;<6eB%AY%bA{K-uFRWwX9p>B#Ppu_)=<(IyH4_$r(_FULMk<%C(eX9Mi6{b%_I^ZY2~~V`b9@={
zIGz6Qf7%$&xAg`dy2)%Lt2VRrN}IR#dAYb2#J&1Hwc^{hc}WpJc9EJrk?~hjk#X+z
zSuT4Vd2mfqR!wJhcRyps!Z~MRX-t=xqG*96|?k#PJ24WLQ$wxK7UEDv4UVR656pY@K%=yddKL4sy
z=<^4vazuT6*PhE>c}NO;bl0W!tNVUISl_}&EACr32@VY4~b~Z
z{VLZk(>}^PKvqlmv%Ng0;mhYt2g!GCbgEMf|Z|T-Jfk>k8UiT
zv7XnSvnRf5)BSX1ohy-OK5_wnb+rDsPw?p@-Zl$C&PR71CN*QnW%a5t853*
z@t#5%rt>u|wIH!rG%!B~K}GHh&(itZe}X;VPNsmJ8LqA1_pHiKK-aV5#8#4+<`Sx3
zyg--uZUND;)-K@bl_`WN0%6yP4_B`O}-K+m-@WtiRAf|O~%ah(~
z{=YVv%M>EujhoXt+J_w*)7A3X@D&E)S!?&41R0+$w0mWvUEqxzBViC$Wa8{&+)w^7y^oRT6wrZavP-
z4i`2P(wn67^<4>K>_tUlP*AJ_^A)GEN~x@evS(N4jO`Lk9!&^2{_!3IBTWs&Z0_jB
z^z<-Ba^AeZ1TQHnUL{Nq#jo5_G9VJ4l@E(
zPtUgD81IIElXJ&o03#o7qhsv(5dBQYDZ
z3i*Zm!(@H@_bEc9`Zxy65?V=UDdYVXgDKybmnB03v
zEAR!gZlr~*VJ-<~YI-FAvrIG+6SdykT`wJFn@1g~K~$^Fc9=cWy~pkl0xGw;p-FC>
zFI%e=#|}B7FPyqm*zarFMWGZJL6MwHBpE))+r0-{X||+fS9FVBpb&tLvh$gm1bHRE
zG}%jk0jx}&sCd*J0$u)r^xXc|$PzI}YNVFS@BI>69w`{iU$M7a}^XuuC
zruGdCnn4myvLmDF_&kb42=+8q-Du#e%Ht&!)o;OVY@+CuK}f3fduCN(9`9j=k1>d?
zWCV3!<+cjrt&lh1i;sx(Kk<9zlJN%VPS5s#zPS?OU_sYYVycDg7y4vg(KAsy?pKGC
zd)i?&=0tWF1sn1({VNu-q-!s4>p^Rn6oj{^P09e*4yJ>8QI930YL>D7Jm;;r)P0fg3s0pNLC-N@R0?
zT|!m5+1+LqkI#HAxR=xO_cUL>C$qFLd;lYf)98CkGd^CRhLb5r$?87u*Y`uTbo5Vde7u3abj=~sCHnAdvQ_0S1~&$+?IBx`
zsEhGGV5N(ThFWwktHfYJud79f49%i{tNQ$J*y&ynX@pD@b?neRW|(h%h4UZRwYBXZ
z`I$ae1E1Pgs9TCWs^{5OsV_ZR1AkXAzQx(upYp!g*5Q
zou@omuExQpv-*31QA)S|h9i-Oq)Rnd1{Ikcj=d}Z=#Al95#8LDR*QyhW9YmhJcFLQ31WCwF|5ZdTe48l!dtY?Pq&tq(3^=_h{I
zCsJ2~qkYe-rcAmJ+KJ_J8T$%DdBe14z1yqDeNJC^4W$KfR~gt=^^W8jot(ThwPts+ZC;6ZPBVKeQ*Hj}iq}glt0_+E!E<+FrD>rZAEH
zO$LR&Mxs}vnFn-V0T6?y(e8{B-NAB_o;@jM7T8QF{x<<DfPBAG(`
zxAR(lsgyc1hwyg42OvEjaBylF_{(Of_0DW`z+^kMP0SwvrQxsS;pcpCysTSQeb;Xd
zN518MRV{nPYQ33nccX-tX;yjr^JR(*sqazGDcWy-%E;r`eb+#L#oT8V++Rh!+n~w&
zl1D(W9Wno_N1+6u&T-ghuSB_4P*l4*c~^(jLDomrt!lr@RNiYlR3GSfX!-PrB)x~X)LQQDG@Y;IvnNA+Sv~90*|4Lg~w9H0v2>`~#EX{?SiU`x`XX&tE~NC3i8`^AC*7k*GAO9&sz#8ous46Paiq-o$Y@(
z_tdo^Y-?h?oSj9)`OE#+g{rHQ6S*huZs7xT!urQ1`Z{-iI6iIUBg-K@M+=q{c55F(
z1GR0xDfxVt7?=dcq#3JbLJgigdB^>BbnQ>rM_`m27`@Q7yne
zTh&Tqk;h#)8r)QSjt;q^-@UtjqeLmqy?t)v1gq-g#jUFU2=jbfOm{fa<(4pXvKeYk
zun{!wHkJT?jta&(c58*doD0Ft=$6xGKf!RBzWr|?Q^a~A?iT{cwwYB{E(+{|=1Fp=
z3!mlIbDR1H*2!@Pm*9ALM7Y*A?>uZbW@tQ5dnt61b|?@^0MAudS9Q}mQDMp+Q}5bW
zb{wv;1)O`Hy0K2k)Db;8YuvtA8gFI$J71s;LpPd#O;>T9*_e~;drQUBf!GAnDkM4V
ztS0orEBElt@Rci^P6l-nt&jb?)$~^V4&+v2&sP_JkX-AJV<9o)m0xP|zB>(gi?R1b
znmD2btVTO7xHr?h#ocuKH@wOf!qK)`>9@y2qfRV+5FLOh)dCp2*x1xq6NpHb4{zB#
zl1Jo!3zfHwnU-gH#U*wCR<_*6ic{|t80+d6pzCPWsZx({0}r_$l&c?2xnKryg6Wlz
zSU}R?cD%Rv`&phw?=G^TT~i~ZkwtgZh4b;oZW!XdnA6q@Z_>N1fo9r6+A<jc
z(`z3~`T0sYN^I!aJ}t9y#@Xl#VYo;lroc3R{pX(oz~*lO;Ct3yC8U
zC?%%2^bETREdi+Hty&b;krC&u&eiuZj!!NDaO=&}&pln6U*1WJ79JM26o2^w9943q
zl^gw;!pY*1|Ex%DWx2?zL`Vk-Gb7mWL<)8m6)$}m!c$TeTdk|oxvA!-+ch*b9q5dI
z_-aXhC=}jmz~agz0{y;h?hMStdvLE=WS*(uG})uq5_dRDf#ZrxdB4_T@chyH27Eiuk9GvnHXuG&|IE>G
z5veXk4FIeoWA>dt1dR3efIT@m*m~^0M=Fu+3k&Iv){OL=Qdp85P@T-3Q>f`!9s682
z6+I=#Blz)NIX+5n`VyrWP7B7hIo?tj!$;bvRl*eyya=`}_#9~I*52o{`Ga(S5c~s*
zS=7VbK0_^53F$edr&nd;h6n$bldP9MIp#K*Km5MvgP;~>T2p`A|I6?~M10e#4D!3?
zo_}+q;z>-Dfq;;}ByvLA)-XM-Pn8NiZ8C|}@#6qo|C9g7R$N3r<;);XmDT*{X
zo|@vlItt;dP)wnmpQnE}{KuDn?k<&SJvbuiK~Q@a-E)28!9~jLaOAwu4zb}Q;fsDC
z!@AXPE1E`%COs*$?TwY)`#YN(m>0q0qEDxQZTTrL)W+;L1CSXOKIG7bV*TgATeS?K
z*^2xT(A3T$aCdp5L|d%fuVvYw+Q}pX@t>o%U$veIK;;5WGn*%zFG_KLtR{6Xco84@
z;tb)LtL$0vDJvz#oXCv=$%Ki)ct4ORZ}nf^uatSO$P`SAjh~&ojYVG
z|596PH-A|3Fha_|t`6||esaDE^*m#8;{8F2b6A(3(ZSS(p`dwxZp12?<~!30BjaYr
zQ)$E_YVr=+#hup6GX6(;#nEw>+b7CSd6lqQS=mK5P*m!NC9fV^8^iF(J3^f*6KaOs
z(Z&d%xVaxsxB$>(oI0h?hSQDPmWIAiu
zgAP!IA%2Lghw?3dKh(D536SI}<5tsIyEwmQvE&FWl^3FPhXI0X_gO!s6oIRg4V9Qq
zdP?7nLkTU7RsaZ;PX0z`Kfy&{z{AUC3B7mTklCm_pXk=w>tw1Te>W$->tk_Oe-
zj0J#VQyP^D;?~CpOdvKWKx#)hX0ea4e#7a)>5fX~MK(_u0&8Qr-%iT)nM)@
zEeI$`TL7h0e=KYQLFEqE4Ee2gIfWsPEcAvm##?yt^q&N)8mTX#P-|CT`%;r+)Z)8p
zr{$jAhda}M>83VZW14H0V+Gc<7U7x`O!gJ@C6o3irahW)@PK+vk`#Vzh)^I%D47}s
z4~oI?H6I+&^U|U^A4Q9w)`|>0e!9xYKrP>OoNl1E!XOqm25p^`m!8TjM9;8j=7|lp
z45G3#Eef)Vy~y2BP8?$GeMz+E-|cQ6H_SKB?khxpbWUM+M!{pJo(|X^STZ1;)8cA3
zSE~2K?`g!~Y(e(gd}MCl)O58|git~73Hs#$$_)Ks6#;>PGKs?r_twin0%;ayo%SAiYXUdZVaTvjc4d*C~pk9NlaEh
zyWZu0BDn<&94rS4k-MqgRuLHxTna~3-z*fqR*@WZjRbFDU$Z%-B?jtZTMyxh85+4Y
zbPsdByQZ=HrlxvG%e1oljPZ9fBP!^S@HS6_y+-DJzN<|&(}}Ua^>q=TXwep&oSDf4BWW!tX)p}33K684TkU7QW=GyoNvqnchIIq5I
zh0ZVA&)VgqursWeIq#3cGgxy<#Ml5#!Q}En0xejjU1HzaVpP>z8x&@O^rwTK6a61MDT5-#+1G?KV5t
zV~#fyEJPVBEY-LaUNU~%zFqw2S_)-vx^
zq9-kK&;&AHn$=?h+H?n$SLI_SUq(jZgvBEaEw(G0GyuwlGizrdjB6SIN*VVIwqmr9
z3X`DwI8x{0v-X4@emYwB>pf2Exn0bCiBUX^pGRzUkX56BZ0vE_m7%G_z-$A5EcVWQ
z=R<+z)S=!3UQ=n$sgDl-6HFHnpIS!mALgAMs=~5;7eH3wp)RzmZMt550N77Spg(A+
zxL`CTSzu-WNZp~8{@kiFLJyokk?#pj%LyVA93Wn7D3!R
zG}u+KTA@i`1IWu+SN*&RJj272#yB!8Oa9n4YWgTrW{ywP6V)Fqw_}fgSC=gM!03l=
zWaVqj^rK(6Z{GAlr_zVeJ=Z7dvXUI~-H=STv@g>io9oDXXHRH=IXv25bHoGug#sZ(
z=eH`m7%%m#oEvHV&2+swGlt=Kc82T5X2FfZ`gecpKXbPyWk_fM0pXF`eM((KnchbU
zqgJj=ZJE&-mmwX6g`>$viqjLtp3)o7X7uNOepYg*Di5@_Lwrhp
zOl9ZCnJ#kfSqSEFk>OrRkZgnol_wWv*NQ@(3>a|tZV5ay8U<}nkB`x9QJWd)IBCme
zmjv4X#T_IfeI9sNadh@iq~yJrpGFPdOKf5K$5hFVTgBmbOwQM)tZ`4)qeBw5d*5+W
z2+?178%s?SlZlysx?O|W_s~E;(_}B~3{cI$2tX(8>v#5PuCXphjyX6d$Z%`>>3!}Q
zljds=ZNk_&`eY=&R-AmY)GZwBd!=n4KIBrrDjz6OGt<_#e=>eLm$ACzWAhe|l5X+L
z<(_S*8-m-PGd%RWYZ!}|{Tjn~u`!6{(M-?8G2MfDe|C(2xOsSM!((h$Z4eFkGI+XF
zbSFj_`JrCZ;Idt@5XmHeMFisS=w(72V7Q_t#NSEGoEs=?+4!&=W#3EgU|LyVZWQZP
z6VosEEugtLIiF?S$lD4uo~JYVhuJC|`y<<&GtF^eZcJ%klj}5qU!14>
zuBnv&;-c}kIM~hBls7csd+7*_cIHh1|J>byPVHQO%=WaHujaNsS6R%l*xFWxT!{`&
zs?VJ+k
zqCw$*I*kUf<4Aq1cBj&I@yAaJ-mvZFrh#tGM+r@q3~tQa$ol0*9%-t&5ES3!xakZQ
zsoqKQV&aW>yD6NQxYbp~cA?(KV#-fOct+GWXbhjAYtC|?#*ih~pS&dFPB7I
zaumA%x`}N)xwp>h3c8ReN_w_@vOiR&aN3i*J+^ta^0SXs-g|O={!AsyuD^3#wZ+Pz
zJ4$n%baU{yX;|c4oJmm{J5V~s9oM+)b*F4wBLXXQJ_5+btvl1N2abs3;2-7gYU_$}A5qfTOeDv5Rf)6KI
zTY*X{|Iz#c&Ta>Onyq-2Rp;C9&&0rQw3#BtyqZ&Q{Lr)8$cmLP&f3pEZoK3%ba%|C
zdFsN)L*AC+PK-BxX=RYVrrJ$fcUnt-)<00_0g=S29?mL+8Ju3=N}Q`$xDvKpA@OV?
z8v=DacXpc?bra9J#n9~Czk#(#&9v~y99yNvNSPxkiQM5V%aWa!V-8<)i;yoM{xw;X
z_6Y{5eLe2z;120tw6BYQjcM$I8vg3;?P(_sLBn!d5Zx2>~d8#rb
zxGSVIYTcAm8W3-EjoF&Q4(F{;u4jR)x)Xx)O>!va?wu5#oP44EYVZt)bEVji_n?>c
zK9-A+81uK2m=&YoPUGTbBihd8j3dEOTL;%CVRg%Fi8mK_xRM(?--hsV_ca1MtUXN&
zGNV;MceYu4XtKz-Vj`Q{i}ntG@tJ3}k>Q3P+wYEbD5v3-+;Lf>))%iY@=0pn*qY%C
z#g(ji$OZjkZ@g_eOWrG{P3-SeK_O-BCx*_GgttdiqCH)3jfXRCKzn-*$5CJJ`qRcV
z2812pg6JPMcP?0FSA1l7sVvg%Ka!2reSCo!^tNgo$0KADUKLAutIEW1s0RwFg>dav
zO*y$~3#DI$`8o6-W%G^tmLrQG_Xl~R1+I`_e{-JYY=&{NL>f%23bOB0jA=3&3J~i>
z!>CWhvL&~l(@`%iE^V-XxwyM0gXF~bZ^XRyH8l10*T*wRQ>74me#4vMKYr?1WtBBvG_gsbUIB10B$$eH;wMP8B{Z%nPPCZ
zQUF`2N&R`qxdt4oCG>O~ZBh5~FjHj+R-DHAb|5Xm`w9q20{~%kFrA^_iZ7>
zV#4OkAg##+JY;bs(9iOiE}2$$YkYEp>)7DU#*NmTH>TyU>j9VvPp?A*F{>rnea1D?
z82stzcE9SkpxchK2so;4V|
z4uvXKU(L)m5Pg-qyr*4lj_&F#KdV}hmgYVhqHr@kTFkw0^x=h!);z-eT-g8Lfggw!
z6#b-%!ksUw;i*zpEQ(w3rX|8n$?9`x?7NAPnxNz+mx|s!g%U@ogF-=99l#@UtJfW9
z9i?G|YG*osuiLDbWLBBkjZiY&8_ALE`u@YNyRqAK8
zF3e3e?~M_3m`qW%8POMgYLNFz@LK~!Qq3#-lpQ^mQANfp+yYo5rUnMu>`}Za=M!F`
zN)3ok&2Uv;PYFlVA|4YM3tt5`tn9n&8n4}@r}o-^p6iqHdx7g3q(`0|x{L|2hqd_k
zNG^7%g;_8LIQ39wNsce;nI~=fFO|v<5zuv(k}l&xl7rg&=vQMbfS)-&YOcfC2tm87geXKyzn(DPZ)QGOod6RQ1rtpXs#YxFc$sy|%8r0a|Io3p~
zXe+#bBe5qJM9k*of_vgS^j&`qWezYo^-q=2<{**V>nco@r!yJU0U`y*FCKo8xGQW(
zC4(H!|IH2jy|o$8S623L_60XLL?2>p&SQ%Omn@Mng!q6wO{(FrQ$Jy2mtD*+vII61
z=(M($)|UsOraG8sB&SqT{C@6KAf5_Z{9CbqC;rwVDi2xyT1Q9_$YuXB1G%}nLchY%
zOQ$u*#Im!xMgQWhPf2m;I+;?tapu>G2Cn6$CM%gdO^!6-HQ|)ox4b;iK4mcN6q=R$
zc6AyCws$z_ymScO)#WIFtsXSXQ2D{`CfB$mB3~DI?wYgdA7C|E4Yh{(Cxp=g#1AVVv6S-bT(
zM|Ho}37u0lBWndyL5}+VG-O)&c~5QmRp=684mw$4-m5vYHhY3n+)h_aED+0ow5tid
zT|(hOXHc#&L3pX;ucYQuo7b9~G>J;9tvHE=%5Ivh+Z%q({5m1-2#7}kruvFK##;hN
zb~7R4Y$()`TBd3v+iKNfruWd*c&QeM8bP~Eg0;0``9HOYD3o3JH^Fkr>dx)YK5okF
zBLmMR+MYKV-kMlbYbg8ezjC;LKq&hrIA^c2wmsn^Sq58cz2dPm{siNMo5Jlz&&Bj_
zofV7@49q)}&4C`}*@SpzmOLZG#auIK5zDDo(_TN&q*3kb}X5&6SCalOYi!q
zC3ikX@WYL^9q^SHU;Mv;FyFqbwGNHD@L!|`4TEFfGmi(2=P%IT_Pr>ha(h1Ni>gUf
zt|rq$C+f3A%+D&{nTsNS1~UV_&AYo4FhA2HY=@aJ~z5Tu{cFU>^Xwp?dB%6j_J#Dr5QrlQw5)nH7W&;})
zzE&FE5GCmEZp?iBNCkI@jZ?^UZ_tTqs4J_KaSS-A!8=
zo2~Sm-t?@7zJXhX%-HFd)%%|{pGlkr>lElv79`UP5QFxTr>wju1_M*gD$S5fXZ3?U0*z>kdD8taoU3!`fC14E`cmPYMl=i}glM`*+8Kn{qd?
zg?3z}D1E)T@*BcbIu2Z=6|hXU1K(6guCLA4#2bQ>?6e>xK8<|GQEa$n3kSOVGRCJx
zXInE!{4Of4%j{<IE7}8p-R39v6piT>Kbzh(7n-HlD4c{M
z3bV&bJZ7Tr)C>3)Ow*=@G+uv81bt3ugg%pbzQm+m!r12|ft&egdFuNM-?S7ATJP-m
z4Pgopn9vHSWhW;31w5e%2wcQroX&M4s$MaFG_W5XjjNvY!(B?@2f^P{mxob_!3Sd&lze#`!qjVbyw`yU8WkJ0__{L3X0!a
z%J&21Y#p%J9qe48p5fHIwd4vpL!XHU7qV}Ez+WRpK6tiAc58);lH+Ste%b1K7ssUV
zavSdx++Nx;bEt%oe_5uNFyzW5iSEfEBC=%f1*w4p5WnUu+&|5JC!}$zo!92~jWmFD
zawz$5_+?QHb!z9#!B$L!jA<65d5W-rNUL~b{h+|)FX^7giFAr?j1mqa^~HjxuV)s2
zM=XF5LPq381HA+1O_h?3SfFqEIkT<1%odw~3T63DulUFJ#qvt$5`jh<@{u0tv@hn2
z%QgcY->jjv5Tj>?p8WHjHy1gI&OGOnFNRQ=t%5FV7uNz^S?|R=-?(~9_}k5&pGcp5
z!*ehg5juxLqMaNYPv;$fYoXnEl#>?}4fAY-CQll2qH>q6JbOZ*bdq`h
zr2EBj%lr2?@4bB8N|H?gm^8(w+N}AkrKi4}@9*aijyq%;QO?X~VZUk%enu@5Kd#yDrFQ1)wC0uPt
z2sBmG^hXXG0wTyrl%KiNQZ7|?n~n2EHnN)t0bGfuvV(U%jRZQbdyqNs*Cy(^foz+8Lm=x2Q+6eUymibp&9_|S`jpN5Mn$nh9J
z(_qR!P`=4UHKx(ZH-S%^kOUL7{$uE_B)PPxOSbE@urz}`;2EKXthAtiQmQbs%FnJ!
zk3KAx>Dj_ev#bTULwYjx&;-70_@JH()}OHb%e?B=$fiQLbUL>87Uc2dN-X>Py_d$7
zfyHs8YEytX@Zxt2vNF79H<3GNLRj>O%l&qEaD}`Eq#&Zn#HQH3M#4$>J^aaW&g-XM8m
zspbL=i#6ko@%t){TCfRh)nF|@X@IGj>YT}I6te6$4o$Okg?=5U)W&iwxab;fWRUP)
zNs!vRZ?49DAe5Nooem^eab+?3?s$KN%G=_v@mozAev5*l#rm{=-DJ6?-%m(xT#Lj=
zF8>C!tR2x$uFP&~{YU@p*VN|AZ+BUE5+>^%y3)RXBZI$@r*S_idjkxBU}$7F2bQ^k
zxHSb|_#32gB%d0efoY13z+-mFf(gx;5rA~i;0{R4cOhXyS*-8eT|mnJ(m_e`weuvo
zf%#*4qMea$Dwfx_PFy0x$egX2$_3@>`}^8(V_MfwlQi0lh9E%AsE;6enZ7(CHwd2FjO&CoJz99d0qZ2F3s-e&903Jtw)trInQNH1$n-Dvg;QTJn
zykM7;{l)5s|KM%!X(Audpa$-7owZ)SbQ=~poO$BEJUgd~siC0lvgFUkQ0?NpsF|B#
zt4%DN%D-7?4@F7K<}qk~6U7A6d5y9ihcP+UIF
zw!~o8x>GoR1;OzHqV9kIRvR`479XE+IU{{Z1KoVCVL_G6HdDJg47Sl}Rwx`XhFa3*
zRHo7kCGdwUG{)ubJhWB>23F1RNbb5-HEefQj%FTVuf$@nw&qXf6h>=LJe1*k{-s+X
zLFF2@Q=wK|Vk>~&(eGPFx0r^$DPBf3vnwFFHC7~l497Mw8C2CH;ANm3>}dY(^oAE4
zXjgW6<{+1qv+fLZn*Go#uJ<%ynGvk4YS}L9$Snr^FQsEQrE$@9pnf!@55(hF?a;Ng
z@BTyChHq1Ox56pp_2}Ps&(auV1*Pr1$BQSwe=tWFAqThdxA3%Qox&m&Ge@ePY49Rb
zNhbn-UAM$YRX7}hM4!1@QsM7LNB-*aDW7=v&8lyz0|W}t43Lm!M>+!^)-mK~M9q+q
zY#9Tq?-dqK3ZVAh*@?8}iIhpA$K>WrJb|>K(oeB_r0dPxy{Gs$j%<|9Mu<;U*`bYh#s*okw+uHT6
zv6FY9P1OoN-9qi2Lm0{yn<`fj&jpoXdK?uhy26ux
zSsRq%g_~QRYzn!XvZH#s)U+!lgJB(qppUj!nPdL99y0-r
z?>Y6xK4rOUg06vQ6Y4i2X)?m_6e)Z!$;D_4>b
zPvgd<9@S<7(O9b67Qsz#9`qB}Bi6XGs74ACsKqiBZYB&aywuM!a{AUHQVqqIv-WO>
zm^1@3GJVj_-|a*oH@v2=bepGJ5RX;A^F5fnUiU4lQb;l&E?D;4tBbP&;Ru$0ao{_8
z7DDq4wU*<~3H^VU+Wpj~`yH=+=yY{Qzx7LHN&+wV_?)J#jpVHEho$kg$zklYVEeg=p2Kkao
z3d>4l9KdzM`CEHk>A!xN76;IJqL;T5e{Oi0*Yv><)>ABTzx`sipq+NYy-?l$f))tB-}_7hvqKY|^NA+y&)8P0QrT9f#dOf1$1l
zT*oHNu98z$4$cJ5;pN~w20cAK9Mxm$lkwHbfIbzRG)P|D*mS#a+Nu8>*S%y+O&Oq6
zSG#Ix$cDfBoLebGr@49d+e`8xP8K=X+sIk~DQxR%p6zeINdmqEW!7eI3939vH
zdU4|Xl)(u2fwP`l{*sF`y_41>yQFDQWmqq40^Ff7><$M))ac#_u37Gj*WJb&!N
znzvq
z9}}enUnsUd`wpu!Iv<(#mBSp^-5>2|4|s4T)JiI({JZeZP%@Jyf0=g}^!t-BQ
zTqFg1LG<+vShG|;e?)raR1RTZ+N~(Mc+HOlk8+aLJWT|;I|heN&Ajt#7J0y=o}dCZ
z6^GvD(;qji?dbLP$Edwm_@>wLMJ(`OM>cdD-8~Proje&XTYeTb{vm#wxXEn}3(B{P
zYb#!2!00o5dql^-IzdYJOb>C-WRhuQeS{XkbIL_Brq<0kpuYzV?
ze5hS*ERa4}zR@53pBy_6@yZ{3@zNYE#B|#_SW!an=wMR`HJr=RoTL=eS;d~t{8a8B
ziL*DF{My`5e*_ad{?>ktN5-@HSBdqrW{hpVfI8mOyZ>q}UCe7tl&8RN>4P1mew
zBk?fFi4z(%s_xt8a$i>nJ@(!c!&{`9h;!|)nXX0|vC-Jp$N#Q_5J0mRo1Ir1<6S2
z)ZLC}$IRF?iQ$ghx7={FOs3h>m3(110|Rx-v-QI?jO66@8N1xggaK>X$ly1u9Y1Zi
zEe|A*f62(Rt;%%ya6nHQpoQGA*Lrcp_ZLz;t}5XjquHWktUD8((Ar(ctup3k=en>+
z#BlTW{~_xwqvGhIwZR4wLV~-yh2ZY)!9CEpy9T!qEI0vzOK^90cXxO9;NDC(@BQYx
zYldIFSas^`v-R0hRS8|$wU%X3=TkfXG|?26fAKIAW;HJOcoD-%Dx=2_Y+Wq;;}i9&
zfAO>%WqB*Xx(BbNR@p8W`uh0noI*Y`*|}$==S2T|J+#pe4QrA{Yn-TX)Jgjnua(lb
z)DQlvlRhA!U5ndwfde
zf1unnTVsC-LJ^3V!BlcuIU?_*6cHd6ypn*$(G51P085*}
zahk_+6R+{`8eKhf=}lTRlgTwcBIfUP#g@Y-$Bv5WVjoJA2#FZ?;G1sEM9plr8&6^~
z$_wJS;)JK*0TV_F(zTOfd&50z+GFqdf6<5ww;%R*N=c8me~P+Awx`aX(m!2>`kK!n
zl|R?o(EkXF)NZO0LQR=8s}V74PKv*n3N>GrC@f;oel^Lw@!0-Ub+-}(#4e3G9OE!D
zRgmLxVab2>L-m*ENm;v}zW{Tdvy#ju?ICC>AzorB-V3BSCz{`czGrk9@0*Iae_CnK
zl(B2h!%Ssk7n_UTB4oFWSiJP=eLI{I4}{vW5!CyFDGh!0(coEM_tH;gA1U>y=(x2%
zVVq?=v*?TVlg&*Jh>3OEj#tm8B?mM$XLSC$*R)^h+&s6!9kohizNl|^^?W|sOn4LT
z1>AcDQgI|qDzsid95$7ClYq+Je=Fn!2j4wuq$Oh1JJ#k|ACKP=pSb!ybTulh5sjQK
z4N*2NWzjI2v>|ep(iMMV(J{5_@_Aa)`Mty|8|g
zPmQf;s{1#c5NUhr`}a57_KhAj^<5c*yZp+~>|>LtCVqR#=R*|%rt6S?e-`yn%~sQ2
zAEk8@y?jM?UF}Yug#1bor5OPL(j1+9K_wjpatc=MqlR~my46)nLlL!q_&6AsmRX~R
zum7@YB4o`U=C}CRn!$EUOYw#RQ
zK3YM{_gFV9E^xcaIT>vf*h^hJzqL5Yj_+L}VzE}tusgwgRd;`B_GWxBTh6~j2YLRM
z$SxgrdmLZqcWl~GRCaqDsbD+FWKdlvqxNIDlJPl;yVfy_x*mp!f2R&BZTwNDGJW$x
zqUpQ5KHUPV`Dt?%{(3
zovYk?2f>jnY;e78W?Zk(8$K}ywO#WS?t#uz;
z`7NPR^x&1-qf-#0f1PJz*?EWo7~%tB$Ni@8(mJ_VCpZU&VI~`j;j
zSkbqULr@V?DAKZKACrhl_p~Ij(!MmD=zG^@q?TnQ;Oe`h6CKGh%-+yAc=d-c&U%}r
z9?{n@BrUb&*ZAT`5-}=`+@L1u!8g*0j3jU~mEv-B@87kDf6|iS?8vtY3;jgcWDJ5%A0zf2r}?46{Z?E@`>+cP}ch
zw|SkU4gDg&Hy2-kn2Mq)VX0bkGgEL9pL3mhLjIx=O%^b?#fpew!1OUQ?(|boa0a#i
z`6Bfu(`&j>f0}m>JIj4KtYI6fkcyT|lk)y=oZlx#{IvLS6+X7NpxC*~HPrHDvlcCJ
z&}LT{h9n=QwM)Q&qaVMZF5YWwmj`>pdbPr$Neex@}
z`&A(N)5S4ABpA%pz>2B|rlM4FiSRTcbwk-x2&jJ!f2`_hvAj@TrXfB(g#Ocg)2dCU
zVFLRt(!RBY%X|1>Cu4{EK%0ZTD?W;OKfk|nS?U~=f+&VVcZ7X3jEVSVN(WmH@9Feh
zWITi6dB?a{^F~dOW}4!`R5ruINd95+XR?zXk>`2l2sZUZn@g!+`)&DnefgO{$PsP^
z1Ix<{fA8_=uQWsNLM|7Y<(5|V+9LKI73Y)dr<5z?I}G}VBf<9yJ{5+!l%w#UfIHqt
z_Qyi*@BO9OGz3=bHjU?PDd4Wr>Ti52bQJ6*w8%|JWMMNgsMlXF0{LWuog_^KYxim!
z&g_16ETZ~R%h1iMO3jU&9kP_x{qv8ItaEPXe->?lA@*_{JR-qT`kZ1t+U)6*N3S^~
zc=HTwT#TO=j|g_%E%$sW^BhfZLAi)iWA4Ah+?0-)!dV_B`+uHr-sk_Saas
z${&U7lQ-SqV+zh}OEdL0yH)9v@wVBA*c{=9*|}yU=nUDL)3HA_&U(u%y2o~(fCBsp
zpp_AOr*z^pTpF4$ow1)*un9>4Ka@<^l91wNQAFa!q)`z8I6=DYAxb{A=il2R`v3HFtPtv^7r>ET>ktWn}zROrKHOrwwpGG1qHm@dpmV&9gH922`F2q
zhraVIl5nK!srTP9Z6$=XbII)N!ftB0u3?{eHDR*LuPWOZzAWsGOySMe0UZWObTIY^
zye=yG*iT)HBt#dVT3OZ@l0#qqf6VDj#IC=@#?F+RxemAg3(Pjb+*iG;;a5a4Y;DtW
z+Bl$ok-^qku#h=;uokqoJ3}x7qyO(|MHc3VGhgqCE)0pqU^)nu2OQSCf~t!<&?%~A
z-+M^L@`)QBTL7t2&dmM%H4OOLjt}b8y9k+fXHODb7oxa4NnIZTS4Ivif35Zd6s`^r
z**Og7R(#n`SM4*641K1YY2ccDM~JcWi@NQkmzWtV97Z;VB1%Vw6>0gIb_yJEL{ukCFY&H^LE9G
zR}nv1wC83_Rx?e;l+;QlhTE3R=fDrW)tI@B+b#DtSe!~XTRTd*
z9A=E;)Tp;`=7C9KKlkHGu=``AM@||*`H$TAvtY5JSv9CdZ!y2y<>|22U9NXw29nb>#kLS)7PH9dlj;B1I2rH`4O2p!$vg*D`7ylQbWahX={_TX}cwfs1EZdij3liO?IIp
zDtv3c#)EX}x6dIh^~h>k2p(l8{>as07`Ph;Rlv(&
zsKpw=f3(uxY<4YO^*XyhIo7X6H+=G38hDBEwEZ(P^@o|${Rj~l1!t;NZev@#pkc1T
zVz8vVjJr`6TFQ9xk~427V&K|y!`EO{f%zSjzE)W{In(N-iB25k`z~XrsBEq$9Z0Nd
z@+TOc{r=7moGSm#NYnW8ku_{q;Hee+HUl*Lm=Z`D;@?#p&3-Jr_
zmtt;-4W~y4-;28o|0Isq+I`s+TbvfCs@Q(eTZP^%JyGN^3|S__T1A!3`3P55&#p~u
z;J8e62U|FF?3SW36}NCOBbRDumwxtZ>%=mr#7H%LHCug7Ren&UQ|%J&@A&jGH`kSV
ze=moW`1RUvFA2n__sxdN)>-KU<(^X+UcTdD%i})jI4r!17(df&0(gpDFb=1UMwYXp
ztQ{-~WES`Xd0aPvPcB1VEv1abpz8ft`TY5@ot{u$I@TigTN=f3WldtWK80V&FW@Yv
z(`=}?h-8(f{xc1S!G5Fm^9S5^dj&>7e|V^enwhOAe3(2Chf0_YuyQQqwSqXZMr&Vt
zS=x(DysRO&INH4nPezfYi(6i{dvjNPi)c@q+AjoOPP0(aIs)=S*87}{MU#ZO$5?IB
zq9ZLKX>>=Y3LNj9r3<0ZpTj8h?vr81;C~M5Y733>p{WB3=*+Z!`zlEV;&0`+f2h(t
ztmHgGY#I&bz>RQ=yBIJr`vYwLekJQpV&idVHTvKk(K>X9P*;||^u*6?DQp=%uuG5M
ziuYvaY9=$j+=fMFy26>O=T|YX-g;YHYDJ+NU+-v=TbKVVs!~BndiWl2VB{+OXHslR
z1o}MzM-~)-Txq_jjXvk=FV=EmP{<$BAKc
zU-)e@zeQfHG+mqc@zjav-_552^Dj&8KD0gtcn|hucVe&mD;#kre~_=?f6-}eS66!L
zF0|Wg8OwbsgLns`6vmzzb}L2>W*5$@qkTOGZ)4GD&kkqpD4a?I8>{E(7MmOdI_|c>ew9xY@3f&x3k8xH|66YrXiFm(5rV`sjw{R_do#CyCZ6Gzq<63^i@{fAZX{v)`Uhl>Fr?
zg22}^8F*?LA&6zJE`L5t6L-`T*y@evo=0M5bD%wZ9NmL0@JZ9Tt!!=C*D}0Z@0Ag4
zx?`Q0in=XW*63TdGke&p=h4V@c-$GCRP?3q9nn-#%1prV_Fz5co`Z)#b9t2ueS-~U
zzq-#UX>Cg
zGVQj?Hj$O6)ST-`TeDYfdR4cex^~peDvjQ&vF^d~^fSGeVd!rPhS2sKp*7bODOZrTdH*+0efpw*}c@g*C?-e?@*ajr^QvmOX=>cO$xi
zppD)iGG{N*E)kp(*0PaA52aTUb{YcHKV3e55>!>9-0YKxzwWjB<6o{gytN76`zjp2
zl5P(7
zi`^BY&vC*0f9G9|_4SALpIJWhBE_MXYrfA{cfclAQSS_B{`$S!l<&1H*=k1lH$+fh
z=Z`;Bmy~$w@_zp-Jbe!PzBTLfc{R3xj&F;r_D(N8@#E2G^(9ZRE1NmH+{622-6szt
z+B&8?3D(UR5rBQ6POsWJc3l5MSbAdY#8GJ5A9}{N)T5)XR}jx!+I@+vcV7ujJh?c$C+B$r&LsvvtZEG`Lt?oNq@%r
zZD(c&0c_VjxXJGoCGjX#
z6v`YqxEELW0KhdiL!=Im(ao5Xp3UIW^_C5BWm4+Jlw&<#`OMtZ;f`~IjmafHr1*Sb
z#d{rTc5qPAdQ`gaAMfBEp`p2mwOhD7q#_!sf2%2~t_kBZuJP{n>aoB)9XcrEg3|lE}
z=gciPuzcqG1+@|`uDKRwCp77PEO#X70_-EYOvMYc(#B2mzQP2#lSD$n$X`b)&4odA
z-!JbTust?~n|+fV8>D?=P*3k_*8xE4e+ou6t`6*^#!7BXDZ>4JT(HKifDK0|kwcqm
zA33;(tpQ%kQZy(i0QKgjDZt(@+2tm;Po2ZRjY(A`DCH8m<6FlV?Dk@p1!o$qpgKkS
z{qE!)vGz4IG&F=q`r#)GWn=+#aDUvYG`oS&qjJO+aOe{gcH9}}#Zf>+18Yuhe~mN8
zw_I-87j(9nTGhQsSy)Y8?zGO{?*sc*Q4f!oKTTSqf{m#A%;&ZMf1%Gj%ftWkqfv){
zZ}Ld-_+7Rfe_s&sZnc-;45}mm|Ilv3&F`fPNf{;!SlWkb<@s#mJBdNLk@O
zRB*FzaV)Ni!a8vr$K0=QyT67?A%JAaa7gj9=7@bg0OG<9CyL=D;xEjve*z(~WH=J8
zf0H=t23y7%WZ-fQG`gA8VMbd9Bl_a$uK#JK+|lX%5F^(f9n?yo1ClvV3*t$>8t@beU4JmqkX>aV71-wkSsb{a(=Y`
ztx*$Yzas;TMLF#?-CV~Cn#6BeC}ICNx!^?HDxkE}DNLYW6T-(8rJKQ3q*fy5ITE(7
z^UZTpe+~Jp?y-8M(aa+8mogbP85~$QI(oyYra}LiPm&DQ!FH$je-|5Ae|b@tAOkj7
z*tA>KB?Mw*1{dARA9s(0SKkl_U^+evb~*5TEM
z3K@6IyjvvjdN9OEAr#ugpC_Z4dG&8*731R&fyx%{G4i!{l~6wTcAd<@|NX2b5&t^Q
zMP`mFd+;D&kJx`K>i89gaBl~sI#~!-s%z!@WN)8
zTaQh^B2%)wQL{O9>mv;(h>(~P^-oXb7c!XLwxNh^p+zKQ70{2~yjuYEp>
zjt!r(tS*PxZqW0>0RsSpuYesGU)*c#4CmGzXZ!G9f7aC^n^}4jJwoqu!(JLM*$t79
zOsBaqU)dwm_DcvP_?VR?cauQA-yqw!&SG{2HoA8JKp8iE?t||-|5QPV4u{drD9!Oo
zMIhzP!(wmyj#^roimKI3+Wmn3K_Py`@YZ`c|80|MQ%N0cT$JEw$UP@7FBXTnFY`>4
zs(?n_e<+9mfPF@Z<8-8%c>tg5^s)g+1qB6AA#^?pU}B3POullKljS*Wsvr>qpI}AO
zSZDvdr#EvfQImYd>+>=h6b)2g&H2o)|LS^0I;E}PPIyYfmCW&3<8a`qawb!
zUD3I+*^C0)zo=DQWH^%m0Km1|xMI$XPbQxge`)XK8^yS|5C8!7HVC7jD!kT7E|o3p
zfjc2K0H6ag@Lrrg}C&sw(xB|x1vJW=Nq`vH>TwrKACLVKI{^NB{s1vN_w5H+;G&Yh%6)(rz3f_^@C@
z1t1)V=s4MU1|vtsqU}VPgAePr$r!!De|#q_GN94FW_7L#j*eg+0|=IZS~h1jnnQm#
z#TdcLL=Dbe`lMsklIvaxUU;*h-Ui|Pwy;bb!lq(-Wn=1g?eiuBH39j(P^5T#gP#0cYjl;NGqya#WMFfAxk>
z5!S=O3V@Kt#m;Ft89w2j#&feDye`7_c7p2#6>S?b3OTHUUQ<%Yo<1*@plz?Re~xnU
z>L4UB7=2{}vUdO`lt76Don@i?-~R&=!>x4rCYNgZ1-Spc)|vR?QOn(8OZ7EC&5oM5
z&zhCZ=vXSTo8RDhkue#w;RO#~e{6(zuFMpri4gZab3Oo>e)AJ$NiUJgJL)Zs1MimJ)DE-
z))5Y4`!h=x0Px~`Q0SY%m=m6N!T4F`neVC0wMN71&fRXnIS?G@!@$&
z(pKVz#XZ;P7ns4X2}oqpw}1G{Y2MCrYhXKTak2!lqk(IQ?aoUrk%6htG~jFzfIi4(
z@zvwRap>>5*8yVoB$Ef`e{B8i^ZAaTTu;CbLbo|m-fP;knKuy8#Rk%uX6r9L`lIDt
z5G~lN@g}f$)s|x~ZMuGc-K`>&)7<4L5f0n7VB8Qk)+8<7i
z0(KD5N|Ik2M5?k<%rOBY0l)}}?!6MZIN|fl-FCBzJnVQ1~0N>P^pC$bV`gcnLzd
zfxRbaau_~K73%Me#6Ha`
z`7V>n4xRKtmBeqJ&{Dwy(i46c*6L31rci(I(KG_X^6QH5Up=?T|MNAnLSNEAZfg7y
z(fDLJZ!%~ye=@}L*6CxVquSs3^Md_FKHSUiKo+Na0eRs!m@ystU@Ufo0LEe&B#g*!
z_ss+t{QD37j!(L(;ZZ_EGa39X#{-b?_+KcB)06H5Nu}<+2mBbc+N?<#(;#3hY5g}_
zc}Jvu9sHYqWK^M9dVb|vi=~|wyI{3JNO)KO4sZmP|)k{
z#BKQaf3xzvuE7OP35|)j0m(aD^ejklrm8wg?%R{MA<0NU*$4xR0g%&6$F1i=SO4WH
zjQpIhZf86@OJKNJ9&*eig(i3yjic
zlVX8yV(OYNxY>Epr0V==-dVP}C)a)QeO|ove;(=7a3N$ZzY-|fy#zSlS8>-g
ze?LiaJ~Tv_HEGe9(GVsaTvnnDe+TBPu1V-^Ry+q$>~A8pUl4{|S8|+A(q6XEUWR?R
z^7)40d;CppnbUa_ZZ8)$kSRo7xA`fi${z)ds@0~Bt~08iB0#wYOZW4`Aj=xF{Bsu#
zJb5b&7lH`Fsaox#rUnvCNQgtNOThawf2mIqS6A0ZUvUu9oaL~chLGcn2~~}4GK;E?
z^nW-0$aqC8%$%l37m$UWEGl-A@g1`*ak0=jBStg>4qSvsU0qFEz?cxN8sQKbu)^%4
zQZ*kV&c^8_gA6lEewls*h>>fq#
zzif1wHp&+_SzB^^Zq>1yy|&u=e|6;eYR)y;HN)Lr9d|D6@6B?i8E4>~zxdF-`ZTjZ
zDESVYn9C~P9F4)>jMBZ0>OicfBeN%(`^qr>faVm8y>R6EGL6eb4@L^k`Ey(Ij}TV(
zAGm~zK~+b7y0~#_`q$p$qp9;|eSx^=w}Tf=%wE`x9zL+``65q^T}FC7e+CBY1m)B)
zaceaWy(&I1wM-jz8gqP$yxX;uj#uf8jpyOyMZ+)Io7(VlRQ|(tln{V)dmE<;0eM&?
z8x;`T^TTf?si`xVk-?5`#c8u$bih5>g*?w9+MT`$jl3ye(siwrrku-XXQ)x#dt5uF
zL2tAsWtj0Yic==At`Ju3ewu4J;!
z?Ub?+xP~>!NtHJao!V7J^U5I?(EkdJp88Ki(O@%{S4wp*>|Q8U_ypBWZxhn52r~GGTj2EQNGdU
z)~=^(BRRf`e~??Rfp7QJ$>HipEW&Qa$+IW$yfFEw#PXOwGd*2kRpTnucuqG->SmW1
z?@?Xjq4##a7||Peyl;{s6YG3542qy@t(l**K+iZmtq4r1w>#2X{cQ+DR+kO=yrAaX
z=*yOBoypqnMY2oyQF{%6>ChOQ5YUL0WPH{JC!&zIe{=hf?vxC>?M0j_oVWgKKO`;O
z-1Ueh^Y^I5H$cb|e8+QyveRp<7wEAdhbV$*70bgK$$3Rk-O;)H>yu^3
zI7+(ajM-egN~!z*UMnA!hj
zns7+)pmQw+R6Z)k;UW!5tH%NF7-+g{@BN+UTgfSRrI?PcQ;{m`brAkF67YaH
z#+-Lr2A568L_!zL#qL8avQa`CNI*whe*`Lv!)cZB`UzyCHv{(Be>29OmvCq{wIe@?|!
zIvKY}>K%pd_(u$`-_~Ue&qEJb!^T$`O&RvCGm2ZMTrTXsJ(rUQRe$Q*9Tt?`y-ufi
zzHx@CORahql5Pxxm|}FM*Bn1rIvHOL9E6+wkOuQ{_s>OylY7X^($PMDYUDN{B;65|
zTvM`ImTX{~{`HjyFMmzNeE14Se<_O2cjUS+gif1x|XRh@4PFE%>^z78P_u9Ck0aSKO@ol+I@90`R`B
z<16tTmxcByaCoX1BqwiIO5
zU^tZH(m{PKnxP#0)BRy~f1J7(SBIr_b4^)ba|nv1rwE9{W-Br@iMJ1lr+X!<)!AKC
z3pDL_Ep{oc?2dqwzX!;E@(eAO3&;t6cezOSW+gk|)W0d*jEin9DMr+6I5ON^GrV(p
zg9wQi#l;P)IMaBWGgr=qyg{HVonU+-#;HknQ@<`g3gmKM;Emi*f22EQt|?-z7}pcO
zjVe+zIs~P)kXc`bt}(;Aoh$!(*-uwXdkpcp>mnippwt%^XXf{-YUbAqi;(`iTCIzv
z!Z1EAo0h2?69iapeDTrsE8gQ1f_pex@EjVOTPr+5NuMdUwqyOaKaIolBs+^=g1v}$
z%_CWDRq}$M!jtxLf8Q;6zU;`H-#w8(TSWW*6?7Vj&8X=~m&f+ysS$5T&-<~tJ)^js
zl}pQ{H1F{85AhmkbC%jkCRj&e^n~Lyoc)@@n!KiRZyNl
z)jj3%!0VIO!$f=+Mp&a*zJ;P95|v}4Ar$FXxo*Jt=w?`hPXHSY0bh8&Rgr8e+0`W_
zY?`h{8qjmB4h5CddZTfK^|9}R$3qGu+H?^w-Y)$<9sxV!b`#@yT*wISQtx~nsBLIP
z?oyRFi8DnMe+*+w&5Bfuic8z8Azod3LS+432pNi$=TFtMBe3xbfAYf}DEtLgnz6%p
zCNn=>FuF7o0Uw(}`h7Ya9N1EEppxpSU8kUS!Efh@p+TZUhy~aAERX6zjZ(M
zzNzUn%M(*=_1|yd+JP_3KG3)#-ggRoU%Ov*>6$2+NHwi*QjxmVXJ0A^cYp|iqu;=J
zmmVbVf6~N&Bwm1U2F9{~XAkW!Zm$__!=QOKi_4s*zJ%TlExk&vu^rukF^KPO=+Sgn
zF$$*Yh~*^jKz^gihP7*NG(_ur(S65WiCnB@FLH+4Dh5lSxKzrPDkYlM8-=Ej$6LTD
zCrw0`U_a!4WG7wNsYIXBkX94zI)1&h^5sl^`QI?CM%PGB5z8mLK+AbWbdl|ZM1slj~h^cfn%M=c^Ybz5Fe|bgK+wTn|m@@s(q@lzh3J{_d<2)gLDDk`Ox##@!
zlw_T^Jg+@h=NPG^J+wULPeotj68$KQpCxq4xBJphua9lA=3$p!Olu{*W(iDcN&6VPIMsvJ}{NgpqVis~&vLArl#hk)lC_dJg~yjq$bt
zf&crd6AQC)X`<(|n)reg##wG&Y?8x3DAmFG%_37A7>Vq%uWj^HCzA$@r1fpNb*pUjANkCT;htn!pyoNVITB!!_RX
zvMY%5DLv+kO#UbWo;>8`V6s(WRr7+r(#rRtVwyFaQIZI0r$tLuLq;Dg74&k|@&txa
z#e(S+KK=T)`E;=;Yxq~SD3t7Jmn$*Q{^1bb2utkgog;3*6b*2Be~2;uWWXZ!uKe^1
zrJPRK?&b7!P9+gSTGq|U?y3}S!T-#k{9|f<<4mkt1P&7%JMjgRvDv`tsS9&)xe!kS{`pBM~hCjvZCP3e|*S}f5voXt9c3N0d
z`&`DhOX0IeLHy-5zN=?s#`|BB2fAwZ?U46eS~+r5ZQVoBk9(f(C6+zY??^7uL~z|zjMotd|KE*)Z*#g@mj#+V;kPsA`J|+)R9tp
zC44T@s_>Ztb1N#kW=a}<>Ze~e@7+6n8)znYQZmmb3T^iBix+<4Ugeumg}$6vIaV33
zPcSzDf4?a^l*QA|;v_<~u_z3mKR@OT?YW3h2Xv~&d22`x)c80#UwPgZa);Ep{(pw&
zZ8h2RKEGVdj9|;3oKFKnQK-Vk*%SjhDFT%1XmN;q(h=IIof_U0EqJm(7^UNdvyq>2
zmP%ww1iFQbt_KC|t8J>*Do4IkSea5Uugzfje^guy)SYpf{c_##S76X2>>>MP61%N_
z6K-v(X7wxoXyNdVbD#K(M)7ji_fz__e(zTqM;wg0fu|V*pqY}0U-?b03*D7*A=kxTNn3~EuN)xFg@LnuZ58Sso1~!RY8j6Sl!y^C&AOzqY+8_yJuP8QmpUM
ze?z@vHGO#)9`gB+-rb|lomTqY=k%&Cd-rW*!4GKv??Ano20nCakIgAnruBcP+i(6v
z3gFD2IFO%Zo3Jo;2*7|uw8T`g%qU9DCzq^b*E{1D(^BUQ>*o{w7`D&tT0XUugPgve
z+or)PM{5ET755UA4<%F@BqTZc!N$wtfBivyU288?t{kWLg@b3USCk)&?*4v)?Jwm1
zZ=&4eZK8b5_VoV@lwUz-vyxMaQzF5r{Wrdzk*PRWOWQogFG?+w4_yxbh$W!7&~Dl2ksTzn|5OTQZ7o~L2J|UYixU<`
z1JhL@%@%;{(zaCL!uq4K+6)P2V?iySZH_F)<}s;0YSw&=rtf}92>S=WG%
zE6vBCIxZfM=4=Nm(!Muajsu+Kv5
z29k5~#H3o==V9w>^C;=8f9K5x!|~@oZh}xDABi&h_zYKwe|z#=%SZ=2vhXan8)9r%
zT+?37n>|D@o@@EA1$!ovll5(p_GJ>1bA&$K=B4S6i+yb!mDZ-|b^jEKi?>?F3fnPZai*?`
zjMKwMrRub#-kpL%e->i9myY+{?c}dJU-U!5Q8VkIfCO<0F
zey3eEwQrIkF^1@=$R!s4)Tqd{T(wujb6hXIH4HkK;?9e}PAFbJ<&X`}MHLd~&a4mJ
zm#+-T|G!hK{*01l70$nA*H&D&e$M=bIx8@AzaR+zlTh>sI$UYXXJ)kcw~1)zf@D-%#t(R8aL9
z*Q{s6X8PLZbLcyJ#8a9wssL?bq8`HyNm|IX0e>XHG|b@EYdQ}rKY8xcz3%^>)?Az}
zM~h+=DlVD7NT(S9;`7Y2Fi>0sMv17Wl^5ULX5eife{)XI?eT_wBz~DNZi6tophp+@bkA>J8Iz#FOSG%7tOYL4kL~ueY^hqzME$fjT@kGE4`LbQaN#|96OU
zNa?yJIq&(WO}Te#$Qp*h-ba-Xl{iwgJ&Djj-V#7jYK(8!g
zw|kpJ8-Glst<{SlOPSj2ny!-}jL1U3!Dka0poPIPW^k|D3JEIdC%Mw>9$B2X>CJ<$
z&#Hg+kL5Lo{v`^N5h>O1ri*V0k`Jg1m^s2t&avwe73l8EAgzdGa%#x_YEo0-?YXHc
zOM68KQMC5aDp$`fD!1XG%paJ@W(#*F)=A+sHh-8GoeImr?sJ`ba@HN}yj)T(bbAS>
z`GpC#IbPtM1ME+5GC3?Nf7=L*{}UC*W;i&U&0!u|4*zS~
zgYOr;ak(eFk%eIo@A%IkYdaa2N*L2K=k!i
z^QR76&(LBx(xgprcKwz+glBI{5lBOm9Dj-r!P~R%hTi_%;~tfe4T99#>C-VMD$p8lpDi(M|;rYyJKjj)
zq6n%ODtHvfg@q6>A$D{9BJ1FUMriRZ3TdNr;4llD|xM&O*EoCVy
zsGLZ&doE7m9$_N5d&{^7E9S^S(tqL+^NOEqo6u)(zT-8LDu?QTW1Id;5~lm=_u>!qbT<{w{VgKUqas*p$WlwC6}UlF=uiNv^NZ&gD0d&e2@T7k`VEJ{3{isxLbM(UeLoC&3UV&n!JewOx<86XuG4)>o}eO0B&j7AF7`q~ekm
zOna>R#nwtfNx*@PYt($ec$}T@+69
zXEP-H_BC=V&Hv8zee4IdAACm4f@0`5b>uV#Wm>VqP_%BNrt#1Bvj)h2PdjQoZ9)?H
z_2CsuPna)ZkRo8pB2){5mu{PIWLWkRbrjRS7lDTvcd2&iOW>tYvwv*|>j)7Mrw4rA
z<$^Ao$SMD&`v30!EIyv+DD(u)#r;Fl9b2T{fqb
z6rf8%L9t*(Eq^m8&O7*m3i7>28zDfxbC|x^^r54~;VP}@%&h1wq+PFYiSWY`Fj0M4GdHJ2b{TaN;KSnh
z$T5F4w{i0;d$rD1lviB7*t!OrI-X1aDfyJ;?9M92#LaU#hBZ2#u#>U;Snwytm}+$}
z)02UVpXaeJIe+w34d?iybbOy6<{;|#;_5ovX_Fh_a~qc$MT=^*ZDNc{O``L>;4pPf
z#9#)sHk(V*oUViQ+WBZM>V0ehD?DqykL(9|*A4hwHMVw)2ai+Q&JXgn^h~_{Wh~rH
zZ)$@U!n9fF{Hl@ng-@-iT`0cTIzFY89uEFKLd1aeXMfMnM7qOk+`+Keo+P-BF>6Zc
z_PlZdHB=UCAjH`B1VQ!gTv@!Vgb{&9+h?w1TOfrY6LAG6TYng94;$NxiFU+#bY7pz+ovI2P_4&>yDkC+!_#P!YJxrJt>CI_?9;kN+?zy
z-?<{e9`r{gg>-5N-TRHuzhv*aQ-(Mm*nO>*BsQ6N(7sgHJ$#evc8^^pSAcKtT}}9k
z=)*OZcEO{@9*P|HHKSeiMaRE$xx;i02A2r8Tz}fNq~$0#ESFJj2D2e7ziK}x)a*{3
zR?D+AH90Rx)Y9gUWDt$KtczI-hd+bHqW#*=Z}gyFb_RFTw$<_}KowV6tBZzZQcq#W
z1yw;qNnEfnq~*g;N8IEdjdC%^_2|}fW}HLv<$)6={6+{98+*65Yr7$6Ujlz_L%z)l
z(0~7b*m}#TxT0liln4&N9YSz-_XG{@?(XiMKocB-I|O%kX-IH)cXtiaIIo+WbMO7e
zc>j_yHhb5Snl)>!s@dn`Q%mY$_~+<%X#R4D^{zu^MJhRp9ogeUz-P)02k?{JGR;Xt
zFZuSRF~TVc>Gv`g66X0D-58cG*X2vG2Y>DFr$5p8sw}LESu>6wv=^H&ARwA?F7j$)
zV$_a51xhH7T;|I{pZF8Zm>&{Uj>)w)3os$0>k^gmnjM#`G4wGecB~2lGkV_fh5U><)cf~dUcp$AcwKwl7cFw>zqqR{>?*a3I2!pqvdhRstd$k?oQI!CljCLt
zdYosES_^)BxP0R}bvj!8=LzO563hjtm&nF}rongEvA1D`y<*b
zd%B3sfvl3p$=nk>`?M`pT99r2zKvNtC}Pg#Sa=}(BLf98M4%k%GJPNISo+B#4{W-T5*pWX1q5nnwGhCyyiGP!Yls{^^
zjQ*`l0F)VQn5!M{=LIoW*+$2lg7hwaW9_x(==h-Wq1
zrbTV5z3qUrc=WyDYO$+=Qb~Vw1x!2B3tCS8{y3ApIYqo|QgX+~!+wUESv+~GL1!pq
zdP9UzS?^>X+C6S`9ox(IKsR;FuuJKTLBjG8k>_d51r~j5lo-w8!hfe=YJb6VG@kh#
zuymZ7lchBgBfi&^)#3QfKPp9(K>vdbKV!MWlv=RRIk8e&C62Y{bfO_({9IeO%bw*R
zdcb?t6;^TrX)yo2L{Gcfy9-m@lfrNwPwnP-6JMK)Sv8+h&)#}aJ$v;Od5>Uk&+(4_o1
zSa%Ji^xVFtz9K_htb;?N9EdG9=)*tV&lCczZ(eo0FFd1Tv0fdVYwJNDb7(?{u9#GLLh;h{&
zSEKS4Gd+0fLn$0Ih`rOGEA$_6xnERjT@?Q6y=q-Ojlp!}dXL4GwUsh*ee94Q0>}x{
zHCIkhZQp@g{(rzF>Cm$L)2T_zQ&q}_`CT97Kf=5kllAvL%O3d`tANmq^|Nx>9ZCj}
zRnDm}WaZ*q&u}_pCKZQX7n^zix-jyF1TJRkMLt8V(v-EC=c53SvQUV_@dIgQK#{~P
zW^`jqM0i8WmBdi#>LjcS6B^6R0$TRnVF3^;%(d8&8-M0?wSQhkHJhZejz2Cf@i!#u
zno^k4)4W7k5l1btXeIBvu%eS%j2(-T(70Yd=~D1jY2BrvP8vhID9vqQj|KdE4_f$d}gGk;eFYkooLP91rY(P;tV7KK5pPa1nE
zQar7*d1r`$OC_|jQr-d)b`~mr+Nz+qhYMbn8-~d!IbGp&HE(alUKH!dA&Nh?+F6N$
zB-dj}bk7whlRfyOI64?aEWw6?9@Y09AJ(j9uIr}`v^`pf#X}_r9;0&(Ihr{;>i-U_
z9)C!3wGgix&m7NqX%E&?g144dxWBL415a
zg@f>Sty|@CG1ytt#|-N9vLOY^7qyuG**y~?sMm!lia;MTCT{iHGE>x`#(W7v!)@Kp
zeq?cx{E5j<0}6xy=uE+$oHWF-c{j(9MF4N7O-NMqi_HiFV>wh
zfHB{AFFC@q(D-H!ch6-a(|_qMpy?|(v>QVb7^I(EDwzH9Nz_c7hcjlS@NeF9
zC{ok&O7};84Nd<^kwS?^@rnGamxR=)^(^5WUv1Nq8KYe3UQgkN;E|{H73&}S_|C^w
z+KYnamDIiyl+6^e#PUZCjD-b?z-E>DsR^J_CYh$ttSj|?ZbUDItlK9T&48w1ptMbg!~lZS`?#af4$4{
zr`0Vy>?{|dLbBL`#V{eY176}4(foXX*fJ4u=ITc-N`HiKAA;X6x^X>*G0UIw+jsJW
z1++$Gntuw0&(sDEccD`z95d{ybL^_b^oue03H4@wVX8mDd_Q>E
zds{daO6}oFdUj8z{f3qyT`puqZ*C~gfX5{@Mm@r2UK
zS_!v}8)fUsY&Sr|X4Wf?W^9x2Yk5(;18*evu`BQj#d+Ta&C#urEq?=BngSUclvbz0
zsNI>z0zZ)!u@}J^s1F{U@7aZk5IdQxrng+Fa=V9O#**{*a{yDE$BDbgY&_PgvN_^!
z@hpyuhZ~`&Aywimrl?(Lp;l&CT}Y{Lu6jGY`%oxKxf`PJn>T&GSmwrBn6!C`viAwR
zyu97>{gEB!5APq{*nj8S%mZPhr7N=ZOJS6yD~c(g>Yccc=%WOWY_65e)5@*2?C546
z9Nddmtv1gd^tJ|&Vw=1=I0jpECP7o*Gh7*Fq3oaVSkSbY)Mxf1_7bj@NxK}lA?F2?
z5ul;mw2DVdiU>uu);sXO?J^!v|E{)gtXBE%v?ceRh!URCtAC+l04w&8j(<{Z-Vvev
zrEtrj7$0lQQKT-7%>(=-%4a2vfD9)_DtF##Gw`NQ+@O=)B+eN;G!=a4s3yT*g*d+v
zYZ=$1YL+TLf52W5;M+bd*nMjo-nna1qf!0#&-={2jQk_4tmNQdADr_yZ7Ea}M()F1
zWidKMsudyiM}JBBqrRrqX=t#xR@kj~NzVGVI#=F+x001F4)vEFW=l#XddX-TEi6#7
zn6z5^9B0#W9r}|h0+^}><3&Rl6l++j$J8+pRRAblT@z%}p8C6xcSzBji$0nn%_{LB
z3ZD+3&qbU^$o7yxR6X+5P!FF+~Nnr2sn0RTe=lKa)MLt
z-xe`UL#~@jdbr}yH9F+LuU=0->}f0}d=!4&eUp;jc~4c1tHYZhl8+h;9Iz>7!FL)F
z=?JensH1?NSPCokrz!Sw5yx(TgS<&Gr~6%)fd&@crFtB4$h%lQ6Cg9{2kmM(?Xp&3
zYkhQLV1KKO-Y>gYYh-jm*fdC4lSU(lvzn7uPCMzD;%$s<+h9os1
zW{!6d{)US+%!v^I)in^H`$3WkY%4I4qW8G6j7M9$lCw_KApjK>0(AQ^ZJ&IK{
z1=VOelW&U53=-wnv{L4ieuAv)r+0sxAstguKYz|>0Z`&$Sdy`1tUHoH_o;QasGJy(
z5B)AChq?_3L%q5X3xc0wmIsEezn6{v(8!a{CvD2j`3k2#;6T#H)2OAD
z6n}=}(NMLnS$HjmjzR9&F~*mCqS_+(!36?a@B|~C*sb6vuJKjT_wso#b!I>b(o|$X
zUf2fdURzUz^3%?sXDG2>RnOK)MD6-zmKN5Cct4mROTw7ALzV!=-yDdp3R
z>BZ$~*hEp^{(EwW%U6UscxZB9?n%KnWCH0*gA)IwGllpq&)vmG{Y`{MEd`p8Zhu9|
zEbiZ#$OSav*8u}EBsW#i!3hO*zvqQ2G}-NVzLGknBqvVd8GXV(ckl$yI;4LM#@x?l
zn~7qEB64vPMz-YZH^`Sioj^LtZwmsd1H}p%q$3u*nytJoT}!A5{=4W9%apPEB+h8#
zaArBwt*@%zGQNf6(eXHtGs;qg-hU_q7+*!5=~31@@Y_-3hD0I9BMyx4P+z^kCK(Ub
zbSL{_g&~%}%_2T2%V-rzy@Rv22z+*beQeYM{rxqbJ;C63*!_>+u
z>7-M^ORHQi8@W@P3<_9~79$HSbD#se>2j=wFJHvbP2;Uc6!N~g)eGs?xN?lYz+1!u5g
zVP(QxCXc4~YQ#8FLWp*ymu*AUk|s8vVtUhOhqD$@_zcuq!LzNCF@L=p5;ZV0h>mZ1
zNM(oq0Nc)MV4+xC2L3OM=GXLZNmw7X6z1wGg&ke32y?(F=Kzg-u#(26l+|Q;_;{Y#
z;V9qaCcrX43-a}T%wXNt{ob3nyp?j1D(ls`USk0eM;wZZHm`SmhCx)?(ikOuPYm9A
zo$ygZGM;Vgf5V#&_OUs?Mj~
zo0wF|tx7L?LGw!xh=}!*&UbbSlA$N}iGjj3MnQQvE*h&3eZNvW{#tcfymXtL)S$ZG
zK!
zw8ES{azU*&W0v)Uw7ObtII3gAf;ewLCD$VV!;mN&t64$aPFH9hw9pVOssuTZjO%RY
zeNUo_-}pG-$bXyUr)V}0%>~Fg@Es54T?ytvS|Q^pm?K9;XDiddv4XdD!hmP+ui=AQ
z{hMKr`jQpmk3PA>TdRN2(71GjwhC+vt?kTejq{8QpdB@d{%~<(m$dlg-TNGIEH&Zr
zYe{Dm)GUpp%eqaj!rL@xSBZhQrN#1k6HSW$+&)w3QGYZ$dzLK=I31^q@PzSKgbNZk
zhON#OB~d2p+=Yj6-=nRpAq6N2N)&Jq#?jz(i;?U?R)ze;bj`z8D)~XxctOISXYq-L
z%<8)F%~evRieko!zybE)8|xxrh0sZ2X0`1s4yU(v*i|zo7!w`eFSEblVhCCe
zSNQ$4fq#JzTa
zI*-jSr4%EF;~`~?`*N%X{Gvw6=l1KS3`DP8<>4no7^q>5DU>m{4@WXExn$NOhlFB;
zYGi9i%d38``|$9Lj;KB&)gQVWKF=;3GGk&vU4K3@Qoeka!u|tJT{b$x1hur_V0mJ(
zQ_l}BxrLgOEP0gq@b8EImPS$@tEakK_zglo>rP|);&LJX79aECFLVCVJmL1qP4tNv
zqQ{F?J)yn`u2LR=KZJNV%Pc~Xam)->XxX)Jk+_FZ*!V6sqv@I~#ec3w7i)NwYrDwG
z^na2_0rS&$_2w7f)(k})c`v-*{hq&PIwBt<w@)=+9Lrau^AA$EhCN;MeE5fs3JV5DT0iptI(WYMYbvCJGeQYSHhu9t
zPTm8Me_M|#^ImDVP$?jYi6`8rER_8^$iJ+mA}OM={v#fSKWMDSy>YC}
zPN`*eZ&WJfB>(gMV7i04;w*OD(SK83L{g^ivA`Au03coI{hTT!aWy;HkF3k4C;YWG
z=;{tCXZ>rt1bnv@hdZVkkz@OlYH_wz6BJ039rMYFP6
zIcwu$A7PQNBPo(Ah+8X7deH5~;VNScaz7qHE=d|F!?;XPf@i|(<0G+Z<^g57
z@zD?#j<;?}MNPx0a1hOJKKS0Vn>TY;ZzI02Bbb`TZkGnsWaQU?so5*ZP?3&`aC<3N
z$h@hP9$OeUVUIC*s%A?QZ1ov_x|s-L&`;q1dtXw2a^Rw}9Q%{4FMkziis;6dJrSbo
z>@?M2j<}@Uxuxvi`b>;Rq+Tj67mkxvqHwJVdpdHd-i(0~=(M!v{kGK^@-}7gq?SMeFh|eT+-nqj#*#rv~}2
z(H}RBH`Z;;pAIzw9e=urIvbKUisz`Z7WEG~m~tjUi?_GRqcg2!dzy{xn#A=D2koC8
zvXR%!rWTJnu8BMB4uI8pCxeIPSxi^vG`LdxV9t}aD@EohBkoySQPht2TKMetov>lh
zU{b)T_LC5fy5G$f2a2EPl6g|J+soJMSwD9!y0oz)(=r?D-hZX0mjmJ~n!08o-pl7u
zvQxb780Guo!2MIrx9S)&=Fjy<<+~MZ$2_;)JRqV1T)a~BZovMB`1q#P%~2WcWmlU^
z4&_e1`^u9@DwO2K)d&xh#G@EFvmGvb?979jkmt*2rlyZ^($bofEeMsl0SQ_5zS}C^
z-$e^2lUiZqf`8pdg^h1`D&{Yr+LJoZUwg#hWbj=*jgUx8BqTW?jS9>SlLz05xm=F5Tk0O0(j~a7YG7s`T)6PWs(0_b&r|hXTS6VZ*uvvUo_$u=m
zA`R6BznRD{-gJr1ciN+
z=XlRqFHOCJYBI^3l8
zn0LwcfHQ2QemhFX;p+h#BT&W*lK@=G$kakBm3^ewJ|*UPpUo9sTUmo4dKDLJTq3rS
zwH}e2OpogZoFvM8B=Rj}dTV1_d*YNq_6nUjyE9y?a@v)xOd(RY;--`6`RUQ>5gf!u
zlz%c(&a|-(B5qiLcRH4k4L#%`)B8Bz?~BE;By9qcmz97J@gFMR=|TPZL@xMUhwRW*
z8j)(2WGN?9r|JDXN_$*>%hiFc8^u_3JQ^V4+IoY+QyQ>1I#APMt>=^okrZ@~TXIJ5EZ!9U|q>P{KgMTYb
zB?-Py!|g9F2b&SCEM$Ut5`ME!`ExoWp!ZQ8qtQFQs;
zeTN@!>O8C#CmH%~Ppe0?7BbPt?&tZq#PT1%hK>pGar;`wJAbD5G(rRq+0PM($VMYE
zAw+UDwJNV=AV?WZQEl0_mNMa)b${eXi=UH#Pa(o*ek`uLan-(~M4ggIj`=CE3ZU|m
zZGE>PkKpjO0pBz8o4j~Wy}cAx!zjK3)tOx{^2a6VG9rIOiXrp75$BbNK{v)1*fDg3
zozv&DpESi}XLghy2uWXcEy}*eD$r~<>CD*Fyf#;eY*<`WI44L1Ca1UJJbz~jQ@>&I
zpIdmB5g4$5@#1j82HXZ>*y%pcNFS{|$XfN~cVL5L^U4-~oU)CjVc|_4(Y~9L@T}f!
zd+_|KhzA=I)}Iu$~sovSBRoI=l+P7?tFF}O5~r$czm?@F`;1*mvR5>r%efQ@*YEQn=`}DIna>;7km^18_I_
z@;&2lZT@s(K^i6HWsk4Dh09xnm8#}i8J|h5$?WM-)_(QrzB9RM7Ehxr3VLOZjg|E*
zy@TV-W7$7{rV}qI}e>41d=Rm0wr>Y1^8WxO{uB0?Gwh8nQW^ro}jCfwUHePh7dRl^mRp
z%Wk;xHT`{*ow}gwS=s5y#Bg`>X?*Ky&OTjDtu8V}J@LBk#6~R#W;pe^uC1Ac3>3^;W%A}4C-HqG}z+|))
z6i-MYB5VXpoa+R(dmf;5YjXTbydT#aLmxz#xsRA{AaIQNIdjB0HggcDU21r5Ikaf&
zMJUkqUI0vy0svzv)Yz#fQ9vvNNCDNSZ*DubjEcL60b-3K-Yy~Ps^_PTFK54MUQ{GY
zZ|}O=qK1spN`LhiwdXwLsnos>MoizGM$DO==!7kJf7cWGTM8=6-z9$G_v^iE7%5xt
zpLdW7$Ql9ZU)8uQJjYh#&baK|cp-~)5ik*QA1^QrZ%8T`QJMATiME3ITuAem
zEqUoDEBi)tw~W;jMPlBLTzpEpC_-?=Lv<*P%Ln{?mw!-CnMMQ&-*9kZ#yw-^f>Z#f
zxlSr>dc8f^QIgcb_IbaPHISdK1~|er3%U75Zk`tktvo;I+Eer4OxQl(C7v-RUEz|H
zF){i)TvV>w9c#>5owihrhJW|=nHEgZaW|{crf*kiZ`ntQ_J1+d69Fc1FQ$bf3i%$^
zKFBwC=zrfBJVyY4z0Z$Ep>1{OikLIP~4oJ
zoRk7~W&MTDOB#qE|!9;8)LcAb+P$(gJI;?t0jadCF6oXFj$5;$88h
z+WX{{_f3}F%XEx@54M^N8ekzasCRldjA$B+p)_YQd8h8^Wl};dJ#6_B`8&sop4N^L
zln=~4ZrDR!ikIco-@>4Pvu`aMGGZ>W!k44-)~xDC^lF%!H&MFR7uGiAek(leW$XD}
z{(t0p36=n~StL4%$h?X*!`0zNnl)zB*5m>c820@mWE3?!-QEufeyE7#^7@UTU(DTP
zy^=@;nTG}En^3--BXHHc#g$;~6yAKlOdI+9wcg3mhy68uGU%!I1pi&0k|hDoOSU#R
zW9UtN8}VuX@C~khg;Xd7;FvLvk9iPsnt%FBi~fiFhj!r*=acxQtxhFW)YsNCC7Ccx
zF~>jxtzoKdtJiK)gLeM=GjSj<390eSHu677C)_X8fHKJ#NIo+#4mr(ZhBBx7u$OpZ
zKDj`EBOsaYA(Y`})X9E~aRme6(0o(5U5$DbDu>#mv`5Z4ov%l*YnBm_S;U@ksDCwn
zvoDByAiqGQ$2uftqhEzM29HEHn1h3P4Wp|{c&}ZgN%AO9SrcVLiKEF`?f4U~0$*g)
zgvA=ajZ`eJpby1gH7g_@nCyVmLL;V#ryA$D|3s?98DSDyNF^s{7wBHW)+c+i-@pGU
z8P(~bZS5jZ_#qQ-F3G5q14ss20e|KMg}Z&?woEZtTvk%}95ljjAoDs78{G9g@jZ!w
zb1Hp0A#Q8$ssl<1^&8qd2nJxL$!W);%EB!u2GDtfO}(!^S;kRy?IF#)Em-Kdu!XF(Z8e35pAmX^Oo4;rq`T{d
z*5G9AlrzgiyW5O6oPxmcd=;S6TOmACXnZ&I;j*)
zrj)WB;cA&M}ND7?cTqA`?|tF
z)d3~p{y!y)*lEaMVEcA4eBk2ebdD9*!_8s%Ct{o89U}||rOp(eFnGG4ED}&VhmcTc
z%B-L3b9PJOH6Vnxw_Mj#TPpr8z+l4m@*|PxZp7!e5tv|-`0Yb!oVN3dFBU@h3Pj@N
z_idg<+3nr1Ip8PQ)PGUKi~lT{TVCk<&}@ESk6Td*HpB?WezQhDmvz(Dt`=%%-2!sh
zf4XnRBdOqBEwJO1_Aa*uQki*20x}je4$}D4Y}o4?0`oihyS3MwU&;*{xLt^9`w@LR
zC3n>^kOTJ!;(y-#>(#rze=+}1!%oF`2#w1!K7|OCK@F%|>wmFQPZ|FHG7EM{yE+Y|
zEJB|k)V)QT#=fBZf3j`{!>RaK*nbR5kqf3cu3u%*F!YJMTNNwiLi#m0Jm3BEncl0U
zw?UK7!>bvlufhwxfwfq{@(Is_d9_@OUa?Yt4p@F`1VZd=)nwVsVp1Qd|G}@?C0Y&l
z<_*+078zrk$A8EEG{Dd6+Gl-kAzp)>O;VEAS#s*YDT>FEjJ0P>JfnU+jsI4C(h;o3
zk?OiIiFY}r4qK%`Rig)Xtt~F7-{E)FiaM0QGD-L1*M#$1wbe^cdC*@rord)&=xFE$
z6NKzfZOrLaLy$MUFk4QW!m~gw=bCD8z(!Lhh(C**_J09>F6})<5o6-dcWOp_62<+2
z;npl@;K2sv(MT=mrJmAO)ZSd9i9_5e0;c~4^p4!e#6%toMr=;F0B
zQpRkEUw=TO&HJtoB71heR0BzDyuK_bl(LaRgwM=WikMFLUpRtFD%#bIvV!QDcCR6smb+hv;H@KL^
za=(+{f*^DVmLxiUf>TRD#j1rcM8XVd9?UT7#eYfADD*O%_kLFF&w55PAJD%i{DB`g
z+TyxO2DyZT!u@23YON*%SUFhM{4uHr&hf7iOqzq4YzwDw!N@LF_~C>>m!cB@nS+dg1tdt1E{)g`p*l#`ybua%ajxj8>K5|bQm@}t@+{iIY`8sYR
z+e#a2UFGF3%posHnNu=_qEY&IZkM-}#Ld4){ro@0ruPAxF1%>aQ2W~T;xC}Cy?K|#
zaA|V~+}mm}MG((2?r9YgxbpuZGpb+VFMrV=iTjKYUyLB1aLV*^r-U6Lu0Cj;obXX@
zh1y2^rv_>qMTQ`mr(-NP2FV4QoZg#X{?%{o+(a$fTP*d1jC8FCAyAhq9Vl0S*WC*E
zslChZBgVSg@W5kED*dL3)Ik=!h9<5~^du(!Vv!4GkPu{m>ByS__RSs>7CBiBW`8nI
z_hkiTN+;jhD{QfYC0~DQINlRhv#lfnDv<
zWtUC3Ul;M?UX{mx)D-n(;ZPzm95~E|6jKl&&J#2C{|iXt(K8VAtYB?^2TFer^$qPT
zcsUv?63giO_76M|3I3}ZqQhDy<^X#-)_mp53CpYta<8K2(;j*9-%
zm%NG=J~%!WQ-?@KYA|4TQNikybbHRkFKzypVnDU0;$e^=82!7Ie}y2K|3c7LeStf(
ze=2j|gx2uE6o8zD#6+SPqA&0bZQrJ6Lpi^L_2AeW8_Y4*U2t9>%-Zb@yMLn>H?3{2
zs(^oeQ0#y4=VQov9~k-|)BD)_ODTy;3XWvy)tShK{ZW8c>_;_uY`cY(5XldSUdec8
zQ;`7-u<48PY~HK9BY<1sRyq3-DT$%U`H%5GmCtYAji)T7H*?`hmZVEwX9HdZ{|;0r
zND5dMJad2irSPAhCH+M&8Gm9nQFM1m**yCV5rQjvpNY@UG{-}aoqsQryZ&7A$o)Sod(_^vPZAlnA}?OSH8
zbZTY0`}*HD^<3!dY6h&6j9GseVX>BHe_qj;{AFpwl4iS~{Pv}4pWP6Z
zPK>-i3{~om@PEDNY>~gW{u^BWsYiw)PT0wP1_3{acOP=NIJ6E$h4{Q7O96Wjaz<&+ytN7GJXv+za#;|CVHXT*{
zbbO8ra8OW!GtCn-xC{l&wBR4ZGjI2$POTQH%m0aM@d^eY)<#2v0~KrY`L)eg-SVpbgu~5aMDxLFPSAr8F@;N
z#yS&J_kUy$8EE$2IFm5wrL(4{s~G7MgzCClE9caoRba%TAjO2Rl_GM$IXKS=J4^#L
z?f}0JdUW(NiP?~>m$~)f+~qgMY+s~+J~AmAw!KHXWaPq>wEr&*r|t36Z;MC#{;@aY
z#8QR#L-{*BjwFPG<7$DEV6p3w2U^S_bR%5$!+%fG%6M9wIEh;(Z+?vKcK(I4`|0s0
zpno4w_EG(^(}7oC*T3%=N@!~I|5D1~Z3?p54GJm#UvNKEt6BXy7QS&qaDgQLBX%rm
zXQ{<3QH2_#B$Kb5{5NKztD7
z#DAchvrVs1RGtaxjImG{B1QW9r*<5vFJGgBd$W(B4`w9m`A}&I$NeS@@+Q1UdD9oN
zsl$Aq(xjsqC{~HT3*Q-3p8aB+*;I)Z=#6h=0WYKY9FnV%u^{Rx;`AO7A
z62w0=rTe&I_z?te#1vZL+e)O9;UEy`|BG!w-3Z
zS9}AVw-CR(r+KkY6uHxHUm!)~{%(-rWJ9fF--0FDXqW-!!^Kr!ST92LXQ!@+2Z_h*
zAI8Y9bY5&K2a*OV;c1Od&&q~>Al4wc7RMOHH1L=v=vy&hCe&f}!kXJX8F~3Gm4Cld
zyH5j?5t@tq3XoWz^A*%6qy>lQRupLo944(BHn7#0WrnxT{4BYVgB449jBdBr=>Xi7
zsZQEH9<=|IqSwPx7|>pHbQq@a|5b>j$G=nhT!(9L-Y(cs{NISPtLRj=Vet-^rU>p0
z)O+j~n{)(#slFEd$Jz;fC_Z$6v7EXFuIi=}F%uk%KeEB|TTi@;S0IJ=I8WY=ct
zgw`w@h+BclD~$~+lQ!{m^?#;|8-DW*aYOrG@j6Aw`hL5^btrtNJ@0hOf$ZZ3oeTn6DP+}FD0i=~z}F{If7E&HTM1sgcN9*4jdJks~PR
zRlFc=^sKw_j7aD#wHcJZO$E0Z*m7O~mS$UR11V9b#S9hoK
z9)xIJdMg%_`pqSHx|YcUf<8E`cK^+lR+%6EIJOF@>v`M*|Bkk%1_j$U*&iTpR7|^1
zZZD;%R;C5>?hzclAJr)~vXaDI+b0K!AM$>UFZtkJcuV&w
z*q;I{>3{bWFZV6xzOH=rW0$a|p4jYwgZQSM=%3>h7PP+Y5?#~3+O=_LsNWRb@iQ@S
zfYf5RJ+=fh=8Xd?uCRiPlm4O=;hyY
zSB~b|O{eGeH%~k=>!D^zQn&@H6#_M~p9@Ml4q-1(lQHA)df{_oZj8A0W@z9IJ(gtn
z{rXcq-Xz)CTUgvz4>I@LP6J)b(?n|cdH7jFq(XADvoAEV%btAa7CkR>=gpKI(ED6B
z`+v78qu9{AwAFJ*T$-c`2?BuAHf+8==lUNCjeEMUX<`<*OlwrAnjdo
z?qGnR71vMfJ4$+=lYs81SmYmJIk78+_^pj0Jp(D&8R(+57HOAqG}7K5=PzO@7B((~
zqe8m!Fj_|pekR<^kz!MsyGX4aDAt1E#f!77
z&nHeb{dR$i?^?M(L(u1BTdmp6OGmS53gv9;VhliT7n9c?KRTa1IjaY=wSTmw(5JB4
z|J8NbS?h5lG)vLzqHA)$nVtqBy!~xL0hj>_Di;bgF_xTss*;cKf2zyz
z{yu4qvy!l!$e5NtEcASzv1hJ{@$B=kO;c8ZC?l=y8>s5Ky)V?E-2ZlhlvqPMxr~P(
zNhq1(oIiDnilS0a-QQuVqJNQ}fWv$D#lxP7!F1(#Z8$XC*p*4ZT50i+U|4nS#BVbs
z={~4j^LB10){cuIO_1&Z#`tM<&K;;7>@WNVQjcTsRm?H4P&7!BC`w=lLG;~vm@}%m
zKZYY}ESu|KtGer0!7?evq+kmfTehtnF``R%o}Z%#C}PeUnn6u}@PCTc(KrWvnMeud
zZpm~uO7(%=71OA}U1xnLN)j~ev6OsO?rQva(Qs!CG-+e}k_}oB^o9p&J80>z`dtUu
z;Z+{x@!Z~9n~iLOmSwtiiH#g_1y+gYIS*}gZSFxJ`GeStmamjDj~4A!zBE*h?~S7W
zGJ)uR?tE;&Ld@u$Z-3gS;dNzqALGt=mzKWbylSVc7B)_Voi;J+`vY*b_w23+daPpe
zjxX~zJJ~W6P;0T+NLyjmn;WIkYQKt(YtEQ)WHPF5^|H6+tZn>}tixBivOt0FY~JB(
zA05&7PK4inVlCEp$~sOY166A5uOp&LMVTCOd=hL?CReUpH5dv)P0u~iOL%%*~kYM
z35G3A>m28RW_HS?4RUeur7WkazM%M@|*LW;gVdKrvnL9}H-I+L>v%CS%6vdKQ?
z!h6m2vP&*)b(!Z~1>PfB83GO1Bel}@3QI-xuI!X^t7??h3Gsik@o9KGG>V25kD9C$
zmn9;>{v>&AS_~+=b@N4v8v#)dy4n#~uDcD;@PW9Dc<+4Z^#A8}UZ$0gm)$Rd*h-MW
z>QHU8jR&Q-)xT_fJCoB@H|G3Luyf6FWvW%w37udObL{PHeJN}05(ZnU!RWgw`L>_l
zb!_%BUibKtd(?mMt?hqxT+ud6O!|6i+w%SjC}cdU`XeT>hptUU&r<19kT`y@|*z3H~chUJkMdZEHc_6q%fWH|PqpD0h
z1S~3HUQCWnZe~M|)6MEGOW|gCu`%a4E?TQVYn|OxAQFFkQvL!A7#K=Tsph7WVc&m(
zg)sC4RH}73I2549+?BQ1$l~;THoqd&Hz-1Vd4_QJEaGh4AJL;kz#&@582{wC-W>Gd|*)Ac-W%}xJoEMme
z@tk%LT+M%eWXQh&USnF8-{{FlTc}J@9&bW|7hLeE51k6aLBZ0JHPM|pwtQHdv}A!<73s4HgIt9
zY|wy~ftr>vZrMFc{_>!H7hG<8cFN#?vf7Nn$9sR(ggBYJ#HjrtYe2}Sp#CqDP!(yR
zE#e_oKIn-Z(B;<7{r5*hJ`0D{X3gnyS3n#lB$O(1CVoXXVg=j=Jr@15?KG!o}v>4T
zZAORB41c3_-A|wNpWT{)`uFrHTzi287<{bql81}uh|R9P^^L^~7NmcEVC=2Fv@~5R
z@M+dxw@yCjYgM-hoJGbmZfA|SMmz3a04jfkkX!vd9a;-c#(LWxOPDk}DG+;xF<{aW
zAcr}M7P4V=90%np^zvbb6BclAzx%P`+L+78-muXitSrRSNyGzV2`t8VSvObs*DweAUd+{}Nb
ziCIlz^L@F_ZdBFF>dy6VCEKzUU}2nUsH@xyk_WY|*OQAKxS4id$JpAM)`|jLD1n=R
zgVTv2hWeM*DKfa(fs5J`XJd*92
z#p{?pRI5c)Jv{d|kFn;#+%GbtD~Nyo@)tW7)y5nL*Qi-f{g*3e+oskp8#(LKSsfK8
z_k^YA59Uqcf@!L2?b`8kJIY;&@|Sy$jIHG_>8WkZjAc0`3~3mvu_MkS#)6Fp^J>wm
z8TSrH7>)W*hwN`wUh?J^E-Cb#&P?X!KX`AgCdr-d*X)&acpRj4X)12UYF&T-(%M$<
zE#vmh_8HV^E}-rFS@=5)$9
z;Cgx*%1Nz|wj9Dd<#q5ZYd>QCbuyU*lW~}aafT{vmw8$BlYr%8RQ-Q<@d&x$z14-9
z=fKTPJL#Z{yo!suf|#(P_KNwYg9SJ*dI?W@$c?
zO*E9Z(PpA&WHqs5v~q+4xkAhjFsFBSF;;2OGN#bhG5u7F$wy0X_FLjhVD@FfjM-SU
zMTH^N^lfY6+8FVbdX|5F-{LJG6HEgstz+};Gh3Ppb*i}qQ{{S??4Zb|uRK8czE{n)
zN-0{Ct_oFw=efd<0uP0)A3V+WT}XI#7NSYEgOI+^wV;?ZXvqSuQbCJtKug7^Jtyf7
zx=rZnsUSj@q(7wHHl&wE3z>yIc8bF8=EXar#d8EBsrU@!Bt?H&0y*fdyHMqZ@HMo3y#nV_2w>K|)#vaa5<94MPF1b)wDInpbqui$Z+@eg6?-J0
z+rvkQ+jqH%Ud?|7Nr;g=xz9$RzdzpM`<&RJ%AI)Z>gpa5z24>Hi8GV#bZZ8>Lr0*E$doy`PBPeQpb++
z{dz%{=Pl`}J*po+n2X!rQ?W#Br5N+U;;WkfYCtQ0vE3u=t=t#5TT7(Ag#OdHwEKh&
zTU;2s{Nh>IR&nyZ;?iFC_T0qqVS_Ze<7#0jODJ~D)-F1Oh2QSZXiW_b1OGI}_ctzx`OzIWVp}rb%SwNtwVvV1u+NXEzU~jOeJ4~H9VYMlvPfPblzr2*nmn5C&9uOCOW0LuyDd+YE|PLRNHElx
z`5cexG#5a{O}~oEJN*`whL>UbK&4TD>1bS{L#M@bg|C*6u`*$q+1YY66i9Sg|0Lmh
zf9ZceV`gtznccG?h^z61-P|9?mwV;pu%^W?@kpb@8Lf3@5=0tkq4<(ku$DR{yiC;Z
z7CySrz|zVjb}&C2P$cnOXXEMcVtt|mKy^Ck#9%riZuP&dg7IL1dXl7-SoQ#pjTDLtrJrWW6`7?i%xZCIvcQt4mWh8G7c2Y|gR+VxsY?I1s
zH>7S|5~D>RQvBtQyIATUj+1m@k;SuB(XXA;SI`pjgjYNto>mzaa%m>)z0BIMw}PBp
zEi>9DY`B*CAA{nSLB*$jDb?2BdlTQ>TzQp_{cenLHo3Z5Jf4ld>9BFvz%kX!pc#Mo
zgCgV0SomzUhN4`lzvh=>Z87M&Ky~x~QT5gVQ9WVXFp45dD$=FW4bm;$-7UFvcdC?<
z0@5rXT?^6;(h|!83rlw_EFDWN3!na;_xqmjoqy+?GiUB|pXG->fH3WlAY+U@h?XcM!e<#pk{=`l9Fg#->Uf~k7eTa^F!X)5#UIuWDW;m8f@;yQAkmp?3p3m;=Vcwc`u8FQu!
z=J8?W|EWyZP$~}ivAT_aln^L~G?65sZov2DSQuG&5fm&PwB9h$4v(M`gHfGq*{p9}
zbI-1Vi6i}7K?u)CSI>na8!qxPn7zj`W+vy>LlTXQ
zZynN?ovb_A2i$4z8h!~SoNIUY`9GU}7lzMJISYPG!wX)9I@M>~wv-5py9AV>xYJ#3
z7C%GU?pz6jLIQL0kJ0Pe^X=&)E893VZ2>E0Mqd)#5--ZHBGSbTq91=0K8FVlWBmS=
z#qy01tJcwb5N8DIALq-7x)KLgx%;)$0RUc)J!K~CXT2m+0g0T!PD6zj)L}mG*05u-
zt3(b3w}7BoJd+rFLrhQj)K1L{5y&i&-If^Z5t^n{{n&C((Da+-5v90;fD35qTHjeK
z9(9l_AsjzjZ)hGYt^a?$21$M*fN1vV3_)bBJUL0I8)ys&IuFa&cDP~>j8Ou_H(P_7
z-G$pj+TeHOSAf$@!N8U4tK{D08J3KkoSjWX#r{$&u_#@rG=$txi&A?oo`VX~V+6~!yb&P$d&fCo`cA56BiPnGcE)#{PhIy^=rsOH)
z1WobXpRH&lQq9hZ)rTx~lK&f;*XMg-E0-KWSXZb^azgY}HN!oI+n)h!?B}NJs6MDF
zNn*Pyz_jT$$|P3*#?>xe6r+CA?20pCOmUU0uUeP?!CR?yJJ~Z`+Qb^!+|YHfV6l4&
z?(JAlD-Kh>lsJDgu&uwkL_-DZQ3%u6;vu3r=`_WTnrd0tn#6p%y7D27p2rekma#UZ
z8uF=ma?Q_QK=5O^L+rrGaE}4DhdKW??D}NdjOLAqYxBHt09y0xpfs%`M7
zGDBE~{iKU}kE$u1wVX8zNy9>MU-~c1O*(h+B}1%q`SxQF*)Q=vtxMn?U(FAkO)dLW
zcO;PCCkB5pg0}p!qia2-&z|{l-tS~R=^8@qr$K^0%HMr?UaL}ScW(`vwhq+^1r`_n
zeluydPx^2tSxImYL4+q(FJ0}PnGtZHx4RRQQ;ePAqe5O<)
zJiSoFAqx1P5<({F#p$e*b@#dGTiTqn=Eq}w;;&zY{H>U1oy
zq+&TqM($wemG<^WwHER1{YGARkk?(XP_99vY-c+eZNu64@i8v&BTMARbsWJ}rO5q}
zGtBfhBL(jYe>Nh3mbFm<0~h;)$&*Eg@$FwrNn%HZ@d<$`-dO@N-JzGglGG7SgXMoo>lF=pksblm^5nu#FvQsjT#PmA9nv6x46vZb>xrAwjJ6`uGO)gkm_uU%=FI9XSN;HsmYK}wdnp;>k%hPck$Els_
ztfCTIKVAun@H^6F+b53T*qDC{^t9q}X`C%(Kya*&WjkGziv+o~^k
z=vh!qxqS@b)PmzGc{fYYm+*a2@Ybu0Sh2WsUy0v!;0tW4$quCtbkM3aA!42?=`Fx)1m6f+pK
z$}XEJ=ZD5Q7QuFg`I;tpHfDgLmsjC00OemnL=Hv+3SqE$U;MN;?}X1pST|;zfb?%=
z=dCCqmzOk@v|0AFe;Y(ejw=p@e#$4sh!Ph73NW$%zOZp4cf|b2ja-xdG^rSO)0z%{
zxVqaLApzE_>Vkhp;;NS5g67o?`?m1ARC&iQgYv0nQE|;_8m~>%)qSFUX>uRt?}hX3
zc;G4Gypjv`yOq|Iyw33BUc%KUI=MhNOl=q}W2-a?%sWJ#_)bG
zvI`Aa{{?G$?Fej~|7Bj|y`pdMaK9r%Sx+G%Tb6YYcd>s6E1eak_hVkX3Q7$2Z&I#4
z&fvBlCm@9w(OChtRf-Xsxs}EW%N#(EfdkO^ShE;UuJVdRu#v4*UaOq%a(!ybZ%9J7
z!epaA1rTp0RNyCIy&APcLaqm84XK1W;G0QU3&!^uG%_6D`793YB~z5*(N9<0=6qR|
zxILuKw2proQeSF5)i#cH9oT-_N|`&4R8DJTvI8m~ao6+;u8K8TRQO~e!0-0?-=tbs
zKuc-SyVL7!3GF3K6~l+`{QlfOC>MTVYTOx&IP<>dNFhi;OQTqsRX6@QWI?Vr2+zL6WT`ON&Z;+
zvdgq*(aUl<>C!mZAFjU;x<7b~I8QnNO6Gsv#to5lv|XrlWe;Jm%<8+f#eB~_+U#Sb
z&~r(t-ggk9FbhdZNyp7R?a%cHJz6Qru5pa4t2WLuuiOtD6sv{M*S0fFIyGd)j8%9e
zHWI#5D=U_spPlAVtRF3tG>a5%JfAl=D|&UtC?EKg2YV7>QUqo&uwhu9QK3!yBm93>
z^oJ6{sr*arPy+Ec@eFopV|paK<`0lG1AMtoG*wlh=cDgKb=9P-sc!=_X8o*owO+4O
zzZCF_79tdepQmE|vLKCb=PDno?R36kS*VI`_vxpf+v!zo(J_^b&jKa+cm_l7Hf88`
z@kp?H=c^APlt`>deUs&aA7t`t6iR>OltCtM(&fJzvn;#sC^uQ^I!9Z?)#YEcg&mS_
z3CpXgu5b@6nZ-Jy!|s>1*vW$I=kczdWq&ab>0_fUr5QZl{NrgA&#;^NhjnG;zDT|_
z(0hfANCpshP;B1m@JL@Aw+^CzD8(QYnNWqXmep#hIx#tgIM?%OmW{l2Uygs$UwsRPN&m<9Jy!mT}b#OR1i(<>GK(+2zd7$%o
zl1zoOM2G7acTR<0{!{(W<>r4M*qzB4k%uC5Vm2|~mA3Mrvr(=I=oxU|6+O*Am
zp1sSgrf%b8*EYF<16)#W7Ulm#Chrx3$iX7+lTO|Pi6=F@xAhzxh=`B%cyD>L-2ixRqGd^^?cs8?5}R#iYI>Z9rpoRir^bn=
z)xvnviGdOE*Ye4Coe+P$nZs^v4|vPT&wyM_-S6ifsmqkgOX3eA!qUz+H$`To>#YnR
zSHql@vZ|yGWjA+U*OEo=v!!F_CQ-0v=|sK*??vHZv^t4_yo4YddmQ|(UxpKeLM!e7
z80L=T85`A+hm%)$@_l&E{bX+k5|fMNn}sp310P|qyCx9kRaSq9>g#vD(H^sYAEkI=
zZ-LduwyfwZ<>sW)X9CE`GbMEnHINtu9?L8&3s7g_%Ck=MlZjMr$){Glq};}qBf+D>(&QU^@!
zNRkJ>kr3=L@K%3%C$4K6;N-n~!rV|$J4Q&g=lJcL5l~}ypEF|z{_>A)Tcet@vtBt4
z1qDReIIC@}tGRs6d8O)uR*~o;ICaQKx!l-+j9-GXVL-dCWde3iDox}kf
z+<*~y$WX3r3kGD7rtdr2CpYJOUpQ*5Z9S?N6IQhTwOz2`ntstr??1QVkY*|71S?BbN=Wpw+tkVNUr&(ib7hcxO||MPs*QqWW9eJ@H|DQFGVFiic8xQl`Ca|7#Ls7
zG05V3#jhuGzjR>DCrwpk4}OhaYdpP_+jtQUlFBoNnL;gZzvuIWQNm{i0zNv+j-iei
zAZ9;IZ7bx4iCjs9yH$S^->O*|9B5|-Z5JSI9yMZh+A0IJT$-n+{nPyZbnkGtjC*1e
zDs+F8%ygK4oK0`!wqO47L7#w=(tgrG->s&BEFMJBDP5@N=>4Km-|}|xa{jhP8{~w7
zUCds%Ev}NJ>I1l2Cg#gNknvGvMI05pIE}eQ-}!atJ-B*vh+#@P|L5WdOH^;ioxW4E
zSL9Gq2u&B5f~4SNL7sWpucZe(1Nf;@v<2bV-C@T$J9m8V@@`ijGZOma+
z3_AvA$uKLI=xslrSCI(ppY88ku&k8JxT>qUppzTuPFWb8(g*+DA793FN5c-7m==HD
znV0L#f9KH9i1aP$H@*L+D>p57JK4Tn*>I~5vR-pc(47&{A*|$osBDm{FX^T*
zTeVa0@|oO(V__x5#;!z?=^7VPrid=2b&L#%#a#DbCYx-2Z<(gglD)fN0ix%MrF|nu;`9YJ1owUd`YRE$?PP
z;NXxxBh5B>dBu?WNrqUq{YrPo_dF$;1GM+*`Jx#&qz@TU4tCc!^3Y!tGtJ9M$@%yq
zml%)O&CGeX?~x4PjkiHT%511_iO?vOn5_ZY)>pc!*gnU`BCw+Wbz*DKfH8k2atF~%
z%9RP2HU6IF!4kN%`Ew_|32|QY`p!)@hXjEq&nmBHOgD4+xpLu1y~?L(Wg|4I>`78x3z|*IE
z^2%tS)wHkb<~Hutm)tLVr5?IIdGKIiQ~mzKJ`_9ssP=dIeWdJcsjji~ay@U^88N=w
zkiCn;!u-??L04sEEgEE+jvwC=KZsvL;~MLqU1=d<=4c?coj#we@hX3WgA`k=RsLHX
zVPW1UYhd1`VrXXw%pp>5YJyp7QNj4vCl_aj#+emD!$VdbI!pLxr*_8vG>3(Iz5QE%
z%v8D3%y8`l7Wc-%TG;Oyu_;b>{%6fpB+JE;O)j54+r7DXzl}F}XF*`+s)3zLv7N^q
zn6~`YkL1x`e8R8?e93{H5dr_6+wR`ck}oBi8{BU|
zs=so~@}|$&%lQ7{&)%JXLhbLaA^n8x{PtwrqbRuc(u68{e=uK^kH9YLi=KXa(62aO
z=r3}b7_CC`UR))E^PjpZApX1{n41K
z{)XCbVk$W)grpRvDHe_i54bRcUNT)`5U}4XFg%NJ6pSykcy`G_tTi2MwtJe6%^G1d
z5EVTX6|F+)=xdu!t>caK%yMO$xA)e^omC(7`_pR7Xh1{&J#<>QJAKQfk!qYbV<#b|
z<=hE!DRY0EG(Df8X5TA}-Fm>$AG%^7qpDEg<
zaB-$*K~JzVRpZSIb@fw~AZaWUK_vg;#V3h7&9*5uGt$MOyfcW#R)-iD=h7XWvL7vk
zvG~w+0x_)z+%D4+ApiPjhH4m5!#K0un-j(?qd$Li8+#BsE|(ixX?I)vdb=xgbvish
z^dl!JpFysEjE*Y7TjxNQ79DstCk#APnYShXp-KF#s89lvIImHUlQr${&0dJyd^{REn-thHh^A?EShPlmJn9>F0bz&Gxj_
zS%;&knRHwwOcq2M^hIz+WbcGzd
z3=v~-M((v+y7~h*#|a_V7iw;aU+Ct4-&IY*I~(o>xTDUwjE<62+-`=QAS1L4;J$y3
z`;Awr_kpi@EBGMjNP8{*!rF$CS5=$a>Tt@zOKU%NN+PyK6)Mq
z{Fzuz)!k`oRC{!N16SWMOJ01pGnesEGWzi%mrlFwTXjZvb8_{{0yLkFbzyJ0tz($r
z>>HP|+pGIH84wVTff*nS&mx#y>2QC2Ec}p{p@={HK2cd5IH@
zPtHQYex42@EYsO9ESxB^;s+&)J@tzQ?e>O-pfov|e0_J1>7k+@fvOI>+*5P0#FHt2
zz|LiM0dXS=73~>EZWna*k(l7m
z?87no;Hjc?&&zI|LR$RBk&J(AW*We)+SihqundbOkIa1&ynb*niz2juy4a-S6J1--
zpw2!)XXoKTRPE<)U0>cDICl1#=e|-j<8>5iIUat{GZs-L#F98)h>zt!OSo^8x(|1)
zhdAj@mQB4>sY&vTB#D0h&rP|1!CBRFMUf{sCc8>z&IZIyh9$#QNfm!XCu5Ti9}MRj
z*fL8OfwjCHY0tc>9+)+a5T8~|DW28dX76W*$itt<{Dd=q4n!G8kOZp)na-2ox+FBsn?V&#*+WiARHqfQ<
zsGha6$B4n{{vLc(6Cr=*7S&Z&-RSVz?bA0)-cmWRBrxn_Uy_d#Y1kG*xvv_AV`_mw
zlomMebHTo*6HXc4oQ~HCkA{Qa@&I`b!uzx07hM;6VmG0wjcRlVt!MxBlpZo-w%4z>
zFIAtDc{-ot@N?rvs;N5hwU1tD`RtT{3MXrLX~w3G!HNt^Z5)5Eu&%-gjdULW@JrnX
z4zOP8(AaZSatu;cS^Ke0wwdOcwI#LB#^%mZv(P65^?X@Vr7D6nkMqO~VUBc){oGQw
zA-0=}7c&@&69kSYpSgMOk}TX{_aaSA#n_G|QA!~$FkGq>D%ikRny!Cu9?hFMnIHvt
zWrY)%UBRY*|670JY|ZeTZrxb}LZ+Q1iksF~jjXGNnn1VJ&OEC;rH^uM4bAoB9n9{l
zFKFGAj3cd$AU6StwNE?iVtnh}gMAhhTX+_jAH>t?1}h6L4!>
zoU0vXPFwEWKZ8Ueb!QdwCbkvz#S4a#=Xi#&FZ+xTO~+O$65y^Ml{&h!Q3iaqRS(rF
zdB8Ux%X=SY>ffF~XKJ<=F=6JwE)qU!_bdoC=2UeH-%F-&J}-2o6%=`J*0%_ell09A
zs06fmia>uD2Uq>p4f8Y!@Jn3)6KvTgTEJyz-WWX$br>h!LGD)Ta!pUB%UNOi}OuT
zz%udCIQ?xw!N@Yw<(nt0i9Jv5NFIgrv9Y~{|rpIctplNF=#;+2zY-vNgyJzbgRVvjwXOZ8YRcR!zv=H;znP5mjS^0kwj9`%gGp~OX@TQ(s-%;nS+7*-h>47rYhGdZT5;}9w@4$sN
z6=z`Gf-9bgms(yumcZ?*Qi}May_xETMh0!S_d_wqDox14adX&sQT;E#r9GF!aO$p7
z%T4lk6!Pkc+jU7pn@i=n3HnlcAG0&!hb_7$|4Q_Bnni;H$?Gz
z+4Yh~t&lpqanDq3@zI;3YNN1Ba@Z+a>G`+X*!0R6!Q;?f2>z^@c8Nqg->*aFpO>?`
z+X~$RZj6nja!)n1(~cKdm{@qt`7~rMQbq(M^pt1%^dLNkEw{ysGKNsp;j%lx}o-dx~%N&FMqAbDTY+zY+btU2euMR8RxRA3l)mcp~
z=*}>-WJT#Lx$X{-umkv$)40rEb{>oK(8hJ%u*0h3&@JM#=dM=2ux5X=alqCJ$zgVJ
zG2jl^rgTWH^M&pkGr+_eu`nCsSlc6j@ugAv`&S~MJqLNZ&B;BUM`?6%_x-KsiN~Vn
ze^#K02jVwtM=6yAMJ^n!MS<^MUHeDjCneZgemyy(w02p_uW;F&YZB(+aTtx6OQ>b@
zd1==>%@W3r*L|z20m^@weY?@Pb`uRIHN1?W;Y^D{e9|v85TS(NQ~R9^E8)N7A_fd;
zEO~aOZoih2n
zWOg2yiYMATgjY|RxN6nkr7tRUFsi*UTfCU&ldaU5N32)C$y)AK{c3!_`hy-9&<0wc
zi4|G{EhJGxLw~Q;$L;ZQl}4+_X8eU7?RXZ9
zH58N4vCxlg0J<5rjyI^RwtsZs+?&@n2ED_QSFfUE8LWTrEOdHSKZ)SOm$Qnw4+5~y
zsJ`JXZ43E~E|kn9wQnY6=G@UMQqg^_%OsUuNBlte*UUeki
z5fcuoxhsDc9FyB{15+(tM!9E!CVj@IMc2jA_z#ZpO7~tS^m-~j#0=RQ>1l{LL5;+0
zQt9PGIyr+16Q%v_p9_Z^z(IfR?Z~kE%Zt;c1Fs#n+DE&E9`gUkLOz5{#@3d$llB>Lq9E!siWCXpi2PNHAgmk2;AQA(5@Ag|9+l9ornl~Ovv@7JVk*|=z5|8sbOztP
z*)ab%SWOpco<7SwN;#(+h1f|$m|w&@GdsBMRdH?W!!2D(H_OF}DNIfdxQfONru1TbxhxsG7um*dl+YeV4Cpzg;}9VcW1MWM9J(Q#5Y2MI1nI
zmjUQJ)Kk7Jl?+5JJu6_>vY6hg{l)|aeMKdYK>FfECYR5a*9pLhQwF2H0OIH9kYl_O
zw4LKme`bEPrsE>}{rN?&jn0dJfRzytafr}c=-zXny!TYyY*)MgpAN}GmOD@6!SjDK
zIN!3)AWs$mL|EO7~Xn4ILYTIzNBsK=KNG11{JXr4m9UQlW+j44fh0mR7mzU(3^Nw^?bq
zF3&h5Eu4)c`j%QgQwr+X-!1bXfYg6PF6cS$_xn{WxF9#(GHip;v!&8iJKB1+!0l36
z&pfpAq}8a&b7^>B5slLUJO0nfZxw6V7PR8fruK96Qiu)5;5gg4Biw#|GHX1zZF75c
z$Cq%3smAEEdRxn3Lr{3>c#PO9{-!1)N46L=f;yPlgzc-tzeTUoK~`GCK1qLNW$l-5
zEPE}BAxarVkpp42x%^?{HKI*|sCyqj(EBT;J)%x_#PsK2Dh#ZbS%R3qzl&-8=B|2H
z!R6GOB{*KW`Q+#4_2z3s%3p{EpV_8{LYFIw$Km2N-`BEJ`rvBLl*xzh%%zP557?da
zU}0laNyzEq#9~b^t*$uGp1!(_X+D28e;AocTy5JQ
z;@5g9KH7Q%>3F;kSedRpr`{YWR+31g)}3|V`Mw>fi)=O_6}wml;-8ipffb@1L7?RM
zJI)%wwDW>vg8$-(CRBTFJ945DXuX&KJAucVfL_+@iS)ut6LcP~pBg?kv;3p;4VFgK
z7A5KYvPQ`od~47hv~_?!y%M3p#QRsFtJK3CqwuybrPwLl0g-4RxSmxQnlQ6_hS=>Rgqwaheevs(q
z_;D&bCdf{N|Npjw;mLQkS|D^
znLR%KsDECm>?Kj5S=cLoj<0W_?OCVp)TJ@<(d$b{+i+X|l5Lg0ybad~#!tF06A2>Tg&Va2(SNPFHUVl$r0MhO((
z&`uzeg&x0y5;yp*<@dy9q)9-ii%)YT6%+nX$1s2GDV#Wxjk)a9v=*jwkEZZ{ZuacU
z%S~P1Kp*Q9%-YXAuHyuJF(iARMLj%l_^G41#pBy3HwE4X^umXU5Nyh!3w}34
zA|Zca=<4}FXkeC}x?W+-b=JnK+r!vRU$>SVYmTk9MXRdjMm*cbz8#MnmY1^@?82Gm
zo}QbN1Tc0}_{PQ;86$<}rRxZe4HuF7g|t~mvcI-`$if&ES_h4WPThBS)|Asdm3*j$
z#fn#)MImxYcNAS=!Flv5TO`f9vzzUz21S2qj9Pepl^Uk@>6CN68
zMkYD*A!&av(L!&<(r64^a$m74VJ#qW?8j4ClIzacf4vyP%69KJ@4f&&%DfpPR#KNz
z+qqXUait1l!nZqYU6mZlb1b#!3LSr~XXMlH)!mdc5jM0VgKbKjV@g|t)ronVqaig9_MOHl@;+T%e&M+
zghkZzXDe<|)3T)o8{W^P{_SH#2+D$~B*Rq7z9t2}Tag^tGo;_69RO)f?#O?4R+W7a
zoGS9ZUWw1^=iU(s!Q?PdL?+7v+C*$tP`ae=-;>(ej8oFqdn(TfB^|yUKG`a_m^lvP
zlbM$+o$gd;ChM!GRM=EkB-<1+8comU&5oM!9kx9+00&+|Lu&3q*HJeNO}Zt6nr%0Up$VP=Q2@8$dy;!3&q?(FT|Y*r_5MQqEHPVDkM
zHN~LyT}ga_6m^RtP-UB`#a9r;Uqh_@pu&*9p>cC!QRIJbGg+48djEe|8UZZ!vesnk
z6lv^&lpj!!{SnVdV!I?TZMV0o<4?=@{)#fiL&EQOSaZ%+hjY&sa|K63msz{LjG5eN
zci6C48Fgs3bdHcPgYRk50gy7GDrgU;vCD^*VEa0?pofIVA&csL&BMj~VOFB6g=KsY
zdPk$UEq@BAvI<*FrW$`&wIJbxUL*neE~CIhH$(~=>3b0p=H_SM`@|9p#cwp
z(TvESe*IB6LE3M{yJtj&y!gUK^B*&0`3WW3jI^^A9Hz^doBFbAMCSSZ%L>h?T?j89
zpv_p_Jsy~Ijeioqg3+~o2fAHaU&J4Ig(s#2)2eVAms+<6bRt|$OU4w<0V84%xY<~171BD3}H`?`=I@oGG40@jf|K1`7zUn0dV&;6A9
zj_a?(qEirG;I3x?OKrqbZ-2(M_+F;a8ZNcJ?Mn;{3{x5PTv6ho_-?kZjw(h(uUGjO~9b3n*hIw0iCGJf^9&tEG9`
z$Z2A17<)gcOt}r|Ig^-P>6?mw{IwYb+pBt}WA5%#Ox<>863R~#_sczaC?|DgXUU}gZ*Cjk|ni$IhY{|sTQFL{cM>ZHMP
zO`yhH$B};{EuZy?)@b2s>wEIk8rBtbwL^S=v=dgDJv!aAM;kk+ANwyJU{pUZEC)~5Ho3(hXTk?-(32~Fzfa1yey%arvbdN!qEAKD
zhODJWP}8XeJe9P+%^5j_GLaky?d{ZP_w)ENkYayD)O>A{rwk15y-Yn^ruWYK4u7nA
z!@IN-|MDx4(+AMFUeqR1EwRn;4LQ94-v68`coekPG@b6M60)9_!o@CjJDW`)&L4Ij
z(_dN`AY*Ap8FYcgH1wfYG;qB?Rld;jZ>TWlk6G_*My^g6*D=nCG_xM20Td9u`(}2=5_GvsG`|Ui*$^vHq7ik^wpvWPtaKmrC_(n6(enR9>LI%
z2s(O*!W$oYz5wo(WC;YX9xvS=#cR~nY@2^fDA4h_VSW1Pm-T4}jDZ9kJzj03x<82EeGjXx9MQ71J3jcT&&J3C0-b8<%7Gs+256zdq0
z4+s=)U4ac0x8#~uQ}zoRlh&W4PoHlBH0os4?vyGGsI}M-|MbZVZOLBN^Dv}_!qvEO
zU51=m_u1Ei){Z)@io(|CHxd%TKW+J26OsJMuc#8SULO5B@cy@ET!$VWaTtGUxBt;l
zQha2l26frQe7xoP_4>VZ^qd}@+11`8ra`JQSA8Rq-8BlusPJrFp`&f#Ziry?0OxWb
zFVc4z*7D-%j)Cqz-}r{1I@8RCcSp)f4w))y5X2g(BsRwpN;y#kGlW0`eBdH
z^37_cpck9i4}&A_0$~@6XulbOw7c^UAeTPC^TLw;v7!wPpqZOv)i}H#A?09*DX9>3
zE)wQyT`eR2tHu2CCD$Y=El1scbn8){AU<_{x6=4_Y@zhv5SH9i&6jSA>Ly%&eOU4T
z{oPpa*E$Ae)b-^80B+xBD0M*~IfmlZd(V$B0|0@24t=-#(ZQD*NM5B6>_l%hD>bd_
zAe_#G*dS9GeK%MbRIR8N$V6lz7UbI_m`=G!T*Y}jxr&p$pjMM)z|28Z=7Z%(p6dHt
zlv0CVU)T7Y&xg&0KH}y+BxN9fZLwn^eDOX5*)zU(j21w5x}-d`9!4b+1ZOk;>Xo;9A_RZ;5n2j>arAhrcfF{6dv6Q?KDZ-RpThVlTm$ph9=((?w
zX*^|kc4Tpn>iyQUNzH|Jt5;;T}}HZ!99~~j?am-
z{8f!Z7}1h_JVB<+P)Fo{GjgvkM@ji6ZjJZGdo~yRCGGVa+)du+lcN+{�J4JdN%n
zblIj4SMnF<{aGO&Ib#WZGJC!CX$bug_sr)Ug#}QBbXwQljJ4AsQpq<0Tp#zF^N!IQ
z#@)472cs&tX(y6lcjA>BcP;r!8p{{3UyEY2*N>?6|OzxoZ%AV(vZm_-dnSX4Wb;
z_l4BjY}5}!4T}obAYQW;WsWTq
z!xw76h(CwLkFZ)d
zq&7P;RiHH2QT-4pQZc}V{QeJ=F)#|hz3k(rZP1n7_uc=pmOw6VsGuDnCWk%md)fWUh{4waW0!L2-%fV^+6hCP7#KOD!|8uWpJa$moze>KIoqh}
zOQ|&#$yIxQA(5{JBT$hzF3T4=qGIepZcXWA0$#?`cYhhR09An7;YOmVnZq2`f2t7N
zhYV6Zj8EdYB8AMiVjKsNvOl*i7~PgyzItrts2YZ{`1rJWK4CvKgVI1MN0>pQtH68c
zMp01h^Z$f<7#JV^h$g$t2bHzFyS&+_>kUx1sF&VwU4qjXOpINOI
z!A|}0s!twdP=0A`EUX)>CeB#W3jCHZ`E{5;tvXgvWCO`6KOi|~nfQ`t@
z={v|Y`4(}w{Rr#5yDmm?x!hdtqQbh-UW~v<7Yq!-i^sp+69TlJV$AjAKcm>@KzSL<
z5BIKrBsZ{DZUh%M_>X2H{Vc9t!z3^+r~=*o_L|H^iwg_4uLxx)>P@lP^MlmBI&QPI
zohv=BFBkVe9Upv)j2m}hUI+s>>Vwvt%kUMEv;Z|q@u%xxEiNV&)ns7c46YsqhRm7j
zBMglE&HxV;&eU9%vf;b+>!HN)AX|IdVRv?aOfuGgYfz%B{=`1QIxg3CxTuw=h~&t`Iv7XSUx)f1fypZ?lP&QLm~qV6hx
zc6Of?xo>J3wSmV;8~VW^IC`He}^v0l|-NlUVY?Nqvm~+y(G??x8DBk
zh2why8&~Qq92zF|!H`hgaN+(~rp;&nXcsP%kgw$j=+2|}e$IsYVo^O*)rrq3zH(%a
zZq#HU+27g>-KeEbZMLLG$E`niQ<76%1J;&5K8
zQEtZiBwD&hIamh|J&nP#w@x5pD(&ZlQ8xUg9MHW+85ax6O;G`Jv8N`FqI$?T26P6m
z69$Co|A|Hgf?HgLf-T8%dKzluMIbBOpTYU-rC3;ZFBrow5AcYvyh&6TCAa8*u&aKM
zJotug5>|aqofRK0s{h6C`-AnA5nD~rE9At}h5efD1Y?X(Z{-R!cVs(cQ>gP
z!3SBX#_><}9=l=AScf169r!@ZGfQ*nYSOp#HLP?sBG&R3@!k~Hih6S2nb|d82i9)=
zSjq2;dW23;0!VuFIqAHsR?z#W0JfCh8sxP9uC%4B`xLkAG}OF#Fd}|`VSwO@q^SRi
zKHmsik*726L}H2DHK^VC_QUFCBk8*6k>|ioR!N<_ExLb#Tm~flMb=Vjz24C_HMvCD
znPwvKm2`<%$oXRJ=Wk^(uB~M8E&VtcGU}14_h&u2@#={cpC0uVLGOj#l)rz9uJ}{&
zgJz6VgQJ2aeZ-_uq=SinGB8X;Nk~=x`!he~memf!)89zbnWx{X&JGVN`9)^vydKj2
zLX$lVO00A_-`kfy3Yp>4EAwIflH$IL?AqzIYq@#xc(Z>vg>EPAG+r+CKY{H8C$ri6$M?sR!p3~{(UsYd6~(BTL7`8$L3V3xJ7nr|_GK1iM0VUQJkdD8Lb
zkzi+qx*K1hbC?mx9CkN@1w+uf}4FEGJ
zh09j!?p^rr5C(dG<H4u?h|x^i|QYIQGl*5A9S(*
zDY!BI?T>8xmsvnwQ(>Pb7`I>385mHoKEW{6`d5d+ZN;8<-|`i%W1QYCie{R|`u}c<
z{10LSNJ(cq$tBaLw}1YR;O}pVW7fJqJ>N{Hz`ey&fH`=7JN|9zKh*wv*LGTw-sgf}
z#zxz)X=EARe~SeE9ew-azgefJS!|nz<42?0@V{34hl&5rJ^t~3qoI7{|9J%rGOe8f
zj&B#SZmy~K?i)mZ-|hcxW#E5@bWguQpFY$2->1F)Us${EkTQ>b@#vo!|Mj#$H9kX>
z$2&^7aLIpvE2MDp>;JE6iT9kmE4wV8@%+P%{xsHa`^_(-X8%=fm~l#8?R6Rah4XYb
z6ustaf$OJb_RL+T9$op2Dy%hnPI`uP6Q~Jrq#WRzzeG~#=EfJsrvXtvmH+7w{+2ms
z3FQ3b0r0O?(p}LJ>ejdm!3=dH;-q22HC=If7-;5y5lPpci1y&8pa1)({}mTL5v0jt
zURkQ%c&tU6^GIJfFLdn8%Va#baK=gVVd9vKZ--ikHind`khM)d2#uW8`k#)RX#|_?
zSxWGfzQBJyvm<-MYsFM|Jj*Cy0ef}
z)zE=|k3xV3{h?S7f476aZiEBSCGe^+i4d0z%bD9%_`Fe8|63>J|G2=1?59L&(f$fb
z+sWWk3CaobL-p0F$;y{MEXj7I3Js3cSL4k7&!wOCE@o3NKaLnP0)fWv%%JOj3~+JB
zx^k!w9r46pICfDQI(l
z>Il-r%oidKlKaIUej>jT@jdi=3^j@UsC?mJ({h_(I^Lu-9mi9v{URS!U42&TU?3II
z4S(Ums%Iry_y0y*3He)jHN)y1XUbC1)1z~j|@rI1IOO2(6^VM)}0$pHuQ5OQ+jh8G`z
z=MLtFl+q^SEM*sGW@;8^7Gi4ZPCIFC+6&w`{TIiQAv;lN%*a90UI+1=dc~YSSrN`}
zHF@^U8-k1p9wpeH-Vvl6R+U|3V!bl4!Z}T-@;?2t+xfw!6&V9GWZ89(oL>Kb0Wa_B
zV}?KZ{?{1|A$J~LujM(Ek^7l#&df2`4N757AZ=;fVnOf>9qj=>Q~9i
zyDIWUHu+$#ui6*c@rk;@cBcZhaedbomnntaa9|!zSG)*wBMM$#UTlsOKFTgCi#ko*
z`0ZPRxjGi6eTYYw+O045_HUPe(%e8)>J)v7(@j+`e*UJlrcP3bt=N!;=^4BH#JfYB
z>|{H#1|%*-GyFMhy0u~N_C+#`d}2pg9#*iQ{cbv8Anoa&px*Hcvg+zg!T+f9Y}?X{
zG@osK+Jy+rEuK5sncA)GBXcoP2qz^m`@(d2o|x5N|AC344jS!1O829G?!vr9g?Z7t
zg39VS3CX|QG5#W?lTMC~%Ua1o_ZlOz9{SaumRfP-4CLvVf7?vdDp@6?-QbjjNjE8}
z=}EVDrrb7mJPnfNjvGH)5bW0n-pmbl8|#*+-}d9Ogg*_J7v^A-Xp&`XlHEA^
zOv~u$Cm1uV2<=IHRy3Ht5F~xxR{c41!~c_u*TnG?-@j%)($Gs|Eg)>XZXlX^-2fZ-&zTHzsY
z2>yl+!B`jtJBqIMnM1F=c$r$x$gLE=>e}r*RNyjRW|byZWMuuwW(2`^vh#!d>adxm
z0m+a66g+^I?+*3rIE{3lHfT&a$7llfcvTPaDo%u^LEUVVn+m(^AgtoD`xDe~Y)MhQ
z!iuvZr*RI8!{@1gnIFIP_)99$CCknV_|=)9dS5c4qKQFu@L#KI$FOQ{@hAJ_Fx*SoOd`~Ypj
zeAwytU%x~j>;3L;Z`=l-{jf|mqM8=un%HKTz1SMHyVIY4>)kSJI4`WWB#&GUXcgA6
z_10C*9uu6;oO@3!mRMQ6tZAeAI9S4|`BHHdmSpcYWaqQ~FiG&Nk$`&z}zJeG=969i24I
zFUV>&<3h220fQZ!4BVfR@b2pi=V0ayy8&mWpb+#v&4dC^-N9eBe)6yw$%Txi)g3JA
z?pA$NikiC9V2^wCS#Nt_igSFCz22_Jn>@QUd*vP*G~WS`Y>|>M2HKsW?G_?sBAy>r
z5Ic2@eVL#`?N!CQ=hlgd9htP@rKchavzP+9iMkYjnxgh=$sTrTozCm#s639Udp%}8
z@I+33J%P(!4%4kN1Te(n_c9B@bQHrwJc_RhpOKxjkDCtyA&mhSvZc1c>teY9LOS6e
z!^301cY7J;?0XM%mPSYL@%=KX74sC%ov<+c$`Ff=az~IWdGXa|>~IVHhOk0iwkJxF
z^W(68DZ&#rl%E2Xf6Y_xonH_Alt|5GiXQJ|at(3M`zf|lPkqNy517}FecR$vMhSM~
zfw)D+h9rchKS|&@eGsW8=^FW(8Gq|nCEfQT4`9ei9cQ_<3du6
zNxx)>#q=kMorr)NmN}tI+w8HnReAI?*T=U~d=
z=A9RL7#_f4IX0D#r<5&yAepb&JGsx!$W;eWL4*qU{!SvkK2IBvl@m#wDK@4P?YlPd
zF?$)7U!I=@0PI1PS0d!cg03qmDR_^+F!s&Up4iiLC(d&RadB}=exDsY>TVNb;^VA;
zc$3Ze{hpa-f{LB|DDa@Sj`r+EzG@W}Rb0~=u19$9eev@rY1E=
zj-wsKdlfI^tK1<@C+G=Fq!M)4<#~y-&2_4;3t;#4&SaGZR7-z(@a~5D`aYB?+ZT~x
zPXhF<8x>^1#vgpFd{|*MOP}MacPf#8a@^ZQJ@ac>S`TdQW*!dGkbA@bJY@2bRgpX&
zu-0c2=#edU<0LV0YcxjeB{yp0h%LxH3pzYMU$+ZZ8~-Nl+(zb@y`{r}K5UF;lmxX?hRA;hg7?vE)~OeDCdY
zTre?B
zxBwu_(PFOkv9OFg;G_LDrExlcj7mx8zyL4J%l(Q^c26c2Qn*qN=hM3n7suH1a7R{%
zj+pn|Zf)M+i^|m8xmFt1dR0|3H&KicpKHlMBwQn)ONqcY#gm~_=$-}lx()HrGe@aZ
zD#UHPe3N7})Hld1X`vGbPpjc+H;3qHlH_S;vMyHD?B^*Xr>cwOrew{39Hx|)K0)i}
zVAA+_iEK?BhAicu^Pv#h;C+aW=tEl&%b?Yy)Q>ZUe%6O!buXl(4isbeqzKTp56<`L
z=m2{y5+8*$osY|+@RD9fZJD#x9mhe|eXs8jvj}`>9Nu30I13V*TfaoWGY!}V&hgxN
zf%TxGeNEeAVKR^jcr0vx+2~URy3DXw7;sD9+Avd^&n)danmmk(p@D4s_je)UzWnwd
z1dT&;ME?q^0O3<-Ykd8TT(4_t~X!S`;
z&iqNm9Z?RG_}?{y!nGRpLQVdj<4Za@4kGM*D|uPwwSAD=zCf5vZjDIpr)n_Y=vlcdxR(kGpFK8lGiad#b^@12Rjo#=vDNn;$p~j@C
z#o9i`K0#yzYroRE|E4r$Nl4$W4+@^a1EnSC!z~a1&KD
zT4vFIYEI&JdxguPFJ`Ff1Q;30u?Gk)imUL}!L=ymi5QrNdmmMb-TA&=wx>k1JDO-%
z?<+Eo*BgA+QZL21@>*W1^sVZrJ=o_`U%csbO62D4viW$eRP~66FQ-QltVIBw!7(+h
zKhe)P={ufY!yx&oj9J}S;Oj>VZo!04Om}8~Wj{>4?x}PRFFkuN5p9{g3Rv$HOYE8x
zxLlNcB5K;U<{Yd*b)F6w-LDVVhLx~(KwLMrO%~#^oM3d$X2Xfy)
z9B@fG|F(B)<`TPe*Cd8f<7%5prEk*@Te-WXx2FNMaXeH|)d)#C!=;V&S>B2V^9r7S
zhwx4D8b?~{!^-1DxYaajglpC12A@m@
zOxv5&EX}_zCRx%WeDwH7*;Z?y?DIF#PU9BoV{kh(bEukAz20Y+VE53&a`#`m%tRBe
zdAh8a@Q;)pF4D4BGDa9S{_+h>%ke>fRPThb@@4Zg+1Q#47vGF*1=Ae`?`czv=iC9CdVze7pxdp{T2gd;~!>&|@
zfbsJ$98l52Ku^^YHBO}y$2MbpkJJP&G;=ORWL!oqVNl$=y2L;yO{?U@1c?}!`@FJL
ziW3DZYgMXbBv@u6@yC-R>$~rNK7*!{92my=hANXj*4=Y!9;lt
zLQ=-`jq?gl;&?dC<8g3
zSIx_d>0Yjf(K#xehtb#KKiB<-EL|3j-R+3sllG8Vh=z&r&>TqMLZzF3y(bXQ9pFR4
zm2p&JbC};QvQnj%Qc&geg2MRhk(C;i86l;$4FwEOXndJer;!|U?e4DofPXJ9s>nc(ag6K@iVOi>EvGBbD^0n5)H0I
zm3OZI(miNx76xqP4hI*1?})+O2vzKS`LpsSrI?Klw5*IF{2K251)iK@htjM1!_?)E
z*bDTxQp3VXS>`^fibY`pZ+ZTr!*GuIO){6#8absEkA*3}z22Slx2APSk4SgBQBY}{
zNV*@6Xy_Au5Bpc|6nS~WkT%2R0iPHbk+8UUz1aqG*@`evfg5>$B9|`qIZ{s+_ZS{k
zkVLrtflX(~`uqJpAk7npC*b=*Z)E?8S%ej~9nCU+4a(Nrk)%^Mae>Q^2Rke4LqZ~5
z$ez$&(8(ocfQ!c;Vqpzz%wBzZ+!T6wBb<6C1e|Jegp{MwJ%wXiym9Ah_}Sh>!$lR$
zu30C)n;IlU?=3NZa`w`k&wrib{#?RZer#*4#hZW0G^9c3Zy$5w23c43K!T3)OK=nG@YUsVtF9^>6xO_fLzn_VrXEPeXs&1A8&thv{i}MB47T+69
zQ4`J&GEle)rn>R(!a93>GO;!KNt=wbL;Fe|3qvO0=o`p?qnk+c4K9wx$F7axEB$CO
znNG`!ci
za*gjbGtG)wL~cd*Xf-YcT3$GdZ|?=Swvn73O_*ZNz><-fGtIa_%}t*lGBY}`IUDcs
zYm(iyVcPtEiT|p)FTIQdH(d-L$OZtBqgg;n7WSpfum`qjrIo8A4>pr`1HL~c>Y3rN
zvEcg-FW3HO3`Mgpps)(eN_&}joFTvBq4Pcf6SnR=4ruz)wZmG3$2=0xa#IS;Sz_0>K1<0zODU&m{l%7o>)n3f9q!&
z=#o-+V;k1#om4yCck^V*(|>Vm_LQ06J-hI}On~vwrVRh$Cr)D?B*-DxcD;PnFn>dj
z>$w|$@6g`z@jdvX^M+&VmBnOmNfjX@UYJ%T5rXCax;h~Z>n`mZp7f<8ylg`la2CqH
z1@5GL5$Ai7ASn1ie$>D)91eo1`+%KVU4FVi(w50Nws!^ZQ#5N&=YM@)pUI-@*@?ihLc)G&wqXK)+9g@csS60jj4(G$~TW
zdg<*c&;PY4i!(~*lpjrmIiyNrx^wJOqZpfaL6o&AvhUAaSzmE~
zjI=yl*XwuK{W4p1`nLba?+*ZOmo`#SFFhUF-=kY8XdHEmwWJ#r@&qBx&;=oQ_z7@XGRfO#BZpMc
zZPFL5TVsvwuwhwHHcS>|Fs$}n4JU+uLs074$4Rhk)CiGep0SDdZ_6~q6kMLWJH}bw
zX|G6i-h8sXdl_zOC4HPE5gf4*`BT>Cu-WT$Rifp7*x4;iUDH4!YPu3nWqoIbx39BY
z%CMNT_th9chEkF+(q|;pV>Zi;0brH@RIUF2DBOoKo2T94xw#)lEA2aRoP_RCx6%-&&G8
zP_D9lr7`6AF!ucial{#k-SSx3dqYI~1yiojF5TB@`;dp5tCGmN?ZGR6O^sZRjnUfo
zW>tW{tB6--MU0I=9LQVW$j}3Sb(bcWzDPt
z4q1dGr#xC4EUl&;3)!Hw#umI{F56(z)lDhZ`B90Ui=LLuDuBh_9C9=Gkw(H2OJBrT
zP(I#vD|d)zrceYuo24Eh!X61-nBxEVBNO>KO{V^h)jQRme2(f>HSghn%>!MQxeuJx
z-JGx9u@rRk5TUDB;{>6a_GGl{Lm__?5hEyT=}Gcfl>N`E!nib?Eh*ktyjbZ`%xKH^~+l
zm&W?u9vRK6Dg3BI_cgMA8%XgkYkMrW-prCC-pdF1CiKxlRFC{+PnyAQY)eDQMjZQb
zcTbx4V+JPdL@AjN6|YvCm;+pV1|vNr_69BI)plc^^Ntm*jaAKbRz}Lm=T#78^e%0@
zk-vA2tTc&iKCiQ9C)I|5Pd)EAus@mTx#_th^t3H*_=blG;N%*At{(ScK-vYLXd;WN
z9qZ9i4V2U-=D_SHJKNS8+b?P56;A(TNBCFb0M~@_7Yn&Z3+XgqIRS1tOy#HM9NUsJ
zD;vg(PLqkBdeNrJ_Ai|eb#f?ls&asTVliwC=?@2?^qjib$2X(9gfG($K7QfGpSpIyulMq
z@GV=A0bNIfi$rv(H%GHS3vU-T-0j2K>)sW3SvF>Uzn{J{@G=y7IcgJq5hBkJ;F3`W
zOSD10is2e|T*qU$DX+6EP5H&Oj*0yKirs`oc7y~=%eH=Oe5p8(xnsuZ4|et{Dp8R?
z2tB!!|~1mO5SVkh~o)=`21#ePLYS%ahJ9#D%N9GaQHZP
zImvFG3LNKbDYKsHO*&kDHR*Je+s{aGax5n3x;j#8Q}K%2Sv7sM;=;G0jO_1J1mw#r
zg#{y=yq+-w#kxw@?T02^+FbYbF(e33@}Cyl!4?E
zP7srS%`vWG?B&=Zf<^e6sCDeNfs2_w3^X;}AaezSUOF9}_PsDllRa%jWK}+9j=Olg
zFEGNn4zX>9{IU8mdmM*fukt-%_n%cZPa;#Uq55FF_T4$7V_P@&(D~@{No|nZ{Y&ZR
zVKR5HsUc#?g^W+7PC8hd86>G9k=q5mQZP!uXr
zZc~wwo6ZEmp%IY}d7a|ikeOjj5KJ#?Ui_&&yxP(StiHEG`^YutsuVpr84v4u`bDYE
z!ExJJQ*KI-AbZoXvA}UQPKH`abl8Gdo#U`kp`bEsP+8}wMX^n>dhloe+G1u?Xb5P3
zQvWDcJ1~~h)pKWRp7FGJB(fj{qpV%b%fYJKmCzGdsXs2owOkf_*$r*HxssRVTf7ws
zM;+%L>4bj~UwG=^yXj`QP@sJSohW9G9Uv__`(E)MfPMx$=uR++NE67zVLkUUktdCg
z<#~+{D9t)jah!oy$oe`!2`W1?InIcGD0<|^;x?bJF-Yv4wt^K%J%j_nNgwvB3cEjSN40(tiK3skP5eWKyETps4L53c{Bt-r0>@FOoz?lxP<$A1%m#4&wJE
z?65KRC2S&wv7A)p1_4qfzEmt5B2}b)#!u3RvaR9XbXAiQHS=B8WGXdlX2rP@6*6#p
zBsTWAH=;aSi5_ZWoNq9KR5NTgb~D&@U!$@38$Soh6R2W8KTs6|s@S1?e>0@zg^weu
z##luSzE8PP?KNT{fE9k{w%NFUd9pdbLr8`bi8uLGx&i{yrD*3Jwlv-}(KZ$$t22-7
zegQFS%lA$)l;)LqMz$^&qMLWs&=aZj*0{5%SPy^nbTegWL_fvbaG^4&zY|yGRaf7R
z7_OrsAHm$pAOKqWo(V59@VUi-;MHEU`(B&NtS
z_ZI0#%BI0lgVpKp3>owyERx#ehZW%RK4(~V`ZEXeR8$PLxo?Rp65QY`eV)TxpM$|t
zrNoG$w~%w$@%>+&q#+}JsP5r>AhOY#J<1U_@bXDxBVEQdIMfD(c1J2Dcr^G6L{&tZ{yHo2xuzfsH*
zY#+{7K4=%8n9C31>biW)W}MLS;Cs{Zne_KQN>wef{Xs(-8=%{N^p26)IqGkfX8yG7
zcR>S&6rNu_z9-EK_O~UpVKheQ>|KfMBaPU)pjbbcX$EdA@KRt
za|_UjiqRK)V{-6++(k1--~&xXFW1RIXsH>=zwJKIDrI$$m*aNp7s+PT@+>IHhs8dg
zN8~hjXeA-Z_V09ex-x$XX9xT>wtK%L*5W_dH{h?)iT@I_+kY9)-$eS)lYi^dKObJ4
z?LSWb=Oz9Z>t3Ag>p1?}AC`kRw0p-OJBmKvq@PPV3@!#>^@!#>^@qcFg
M4}^t9{Q$HC02OB#@&Et;

delta 12400
zcmV-$Fptlz`4QiT4F?~K2mnCx!GQ;b2LXi#0<{MNl
zy7VT!_YTs#Nbem2p$Ld{ktSVwCsINQ)dEQGgbq>zp%Xv?ffJth{nq);`JO!I=lS#A
zYu3H8ROI&^#4O`ff&t}_1rJ)XTxK1Ob^yoxzNj{j}Oqz_T
z)fp_yEX;oz{hqi@yQqfL`N%p}9+v_fNYeppcPa#(sK}$2dh%J*{N9c)h
zukoK3G=s&nhc(H`IU3ZbMdO|Qw?}w6TV+jKe87L~#pdyO`Y~pB4aJ-a7qi|dy|D}}
zkTwSaK}ZM?2X^+Z?q7O!dL11n5pznGwTllRR!0jRU(%Rh2rX}y8ng9%>5J@gRGgnj
zrG}%cxd@YS+hfZm@3}7^F@sclr%>lSFLo>V5_J;J0j{EO+nT
zo!oy#0(T)MBm4a|;@J5lXKkH$7d6kdSbDyhw$@VW}PX
zz<2KGV&%{2XhE4xoF4#{k`|ZEoeG#^pL#>
zaYkcaae$78>lSm=_%Rnuq)Vn>|4Qb$
z%M&@Cq%US-J{Ba{*!QviwuREsFHe6&2Hl?|)8!5c!OP0xcYH()@pxQi*s>V%B)AJ|6?nk#8-z740
zYcn_aJvS{e4xS{|-!(dB!p<+W$qpx`ML!d+R=Q%gR<6SAwrbbg12jX{0=0k6?lOys
zgO5F)r5sZhoOb4W8FqcLMrkC
zK(0^cToG3R`!(h$@I;9(r^95vP!WxgbR0E*p~rf^j-?c~5})mrhS-0hiDjNJs3B0D
z-cd!kWpmg(jXGuwHY_v^EiYaupn*3Uh1;WqW_-QOOhwGq=K@R?OV~y>6c)m!~GVAlc#pdr{RLtWWvnD^7Ma+9f@vrK$a(wVk&9MuyA3qkht07}ylFk2k(
zkL8VHVBb2FAPB+w`-Xofg#8#1X7cT-a`AAr$9|^Py54hBeYW17jNfr8wy(~;A%qxN
zOTJq9foCj7aIx=Pn)|@9+a$Bf$8>M}X7saZ2>MyE)aUSP8q@L3vITbpOe;eoSUm2D
z(AaNQynr+Nv$Tzjf+MAzp(CTPe4pWt7_TZ}a
zN+aRjj9tB8)4e44n85e%-zR*|_7*1EfdP}%fRk-G{^_QFFd0qyn>Deg_Gae~6$c_m
zoEsP9eHSSoAL6uMAsvyhf&{zuzLcVBUKcz}ePG|`q6j^1XCKgIPDaNCvKAWy={Gxh
z8kd4v(xcbwUp9X(0@&6!Hp(Sy)X7%pF^V-rvh#i?5h_HB{EpEhyrCR173bOdJPq;l
zaOBy(`6D^z6W%lqQ{@8rxO^HBA5B?V*|OtrahUWN-8Se#D3Kc@xXPB=I7v(z#c2J+
z&*CNl
zaikm&82xNJP~_dO+l9Ba>@ar3e53i?cXvWV$U_xqfyqb5$6G&0Bxm)Q9sA3lxqPt@`q^w%vUKxsHFMy!)JVd`6rOgcyE)Um0&%Mk2XB4l>Jd
zFm07E$_8qc&U^SmKv>wQO5P7Ks<6xRP3a$&;P^p^)KZ|ZRMe%S#p%wB*UmIIV~QZh0YQPFuzVH3;)y3ZVdUP<#)VM)Ti51l=!4IA>oTqpHL5dvM10%j^565nihoY_2VWyUW-K^DMaEzq9&)*=uu<*t9
zWCexuQhRo}K4|avg5te@=!^5lWMY5#a;#W!NQndl1@)WfxOKnvm2#0tTyl{=d^ob!
zo77*5=l2pb%!aU``)HvLhf66~K&wr?M&&5VD7W|>CI}$Nlw#?vt*t)P$kBTLWM9PC
zT4=yOm|ma?mN){~Ptlwi(us~>s+14ck~GG2G{oe%HYlv35)w)_a{}UX{8oRWR+3t6
z`-isa@BKq>m&#~(>saajajCX^@cYN>B4u*G$)r9D6H{a=8u#H)5R*GMZVltXau`Xu
z?%m+4^AZmQ1QHHn^I>6OnId$@!z?#VM!aQ@64kG(c~|sY&a-?Fcy&vuNwdd88X7fs
zcXvcLwd)8NY6b_<<4t%5ov(jW04@)Ew$L8JYh!>?*mo~0nwJY5HHDPp-@&}n-xSeC
z8>UrJs`DV!8$MN*e*Pv)+-K)8jj-1TGr$&uQL_ib`sOA@(cP2m)>cs@npwioz*I6Z
z@w!X-v$d^l5?A;?U_!DO+NS%>@R>tN2tOnm*`zyuN=cD3ZgeRMLLq-gPoUypcv6gsFVK?eo1ZI7av
z0uZpgdD5ARfH!aQ+s1D;W?|G!+Z)ZKeD!ow0w+B^-Ro$rmy{$nHntG07~^CbqFO6P
zr8ytzzR?C>auZS;Ct81eU#TYe2kg;oz_S?PG`{kKkPte})+^omJ}DuMvx=q#nR`vE
z_fYmbyrFY0Epb5?yN#p+`T6;Z2LI$G`clzQX5CM#arP{9FjFuIFMyoz?%e{kc*fAG
zZok|aDjw)yVY|=~98CU1P+Z)MJ8AY85zvwlBXvTH&iF=j?7@Gc3z2v)K0EsO2t-6i
z>SPOh+pTuTO_b>{(Ta1ws&!dvZ+oK)G860K5EM+i|5)&qSBw!x+!naOZR23(HZmN?
zsmi<$9|#=g8mc@f$G=G7GL^Y2X-LMNuh^8xMBa=3jfRbrw3>f#u$s5=Sd+i;9{%Ea
z@e!>m{i*B)yK{eyY*1D-ihVg+XI~ZG9Lg^6Sno@8S~sz@v>dH@XxPGt4zGRB7_tj4
z)-$SL)Q~qCMock>T|wTiZPxq2cc&L?NH
z#pAT_aiLjg3G@_PKU?D@H&Ce`N8EsMUUYbUHQewgns^eSU`OcKoNKH6@y1|@I`L|m
zHTaI#D-<7p#Zx)Ub@umz(%frxi8>8$;K`|ZGuiTI%+9f|_mEkhU}E$ql0?
zH`^Zj#6f?PUz+#@Y#4raa^{GZ!LC6A{deo0M5l6N_Ege@&GGlZExv&6KSp{$D8VJP3y!IaZV@E#ms<&pBpFOT&vSf0|-Wb)^u!Vrm&p?+zyN+k}J+M!{f79hkgI88l4X7erww_}({Ovz7c4!55uzl~(VfLeewc7!QIQrX2BOsg
zSclEDgIQ|nr5RsIu1{`dwMqj5dM%BQp7?bO
zQ!s!0nXvOab>Xa3UKk5)p2z05b69p;LOi;9bpJN6?L
z!y~D`rgH)8hi+Mz-LMGW{fE$d$&wNbyiFH~){l$SQ^_u{toyJ@!p}!b^F_BKQbZKI
zw7Lf^ESQ*G1mVD}-pG3lHe28zAjN+q@o-TzQ6Q{k{Fs}o(Z;zWALfvgAHaE(
zz-?}lwDZd2eFpSHKFn?N>Re2Ty0kpKmOgHysB5He(y{DKSM>YWxBz~<0{!$q(jYT(
z0>Lm$fAZw!Pk2(${>t|CGx*n~*zPA#5)in3mZBJ+r^;6xX
zx~QfkKR&<|SSj8FR&?5)Xg0)OuAkZ7RLlXcpjmCqu-+senfhT{>MhOJCHpEjZ(qqs
zpZwP#XQ5@TYhK;?CTt7)WXn%ayzPHa08iXov=ZF2H(5Cy6kX=psblA>1oBhORtuwU
z(l>^lpMxGxz6S_YYty$0GBuP~k5)-MEnLHmWTvP8o>@t7Gf2QBvl}nQL6ZJ8$cy=A
z&ocMGaE;h=|637=?aMZ=!@?6URGRVDO~oVCwIS(obwbWifgr_R2`-a|y(@p$FgzE?>a
zhEAg9&1WyN2YQ*(Lp#LR`4$dQ_Z-si%Y_;DjPnSzUfa6V_8H|)E!Lr1u!2AKA^L!!xh>Dk~QtUMUb{*HL6423{tr9>AlU1IXDt80JYGntH9so*IM
zGh5;|(pv>)Z7z?0Ve*f3YXq+enyK*JBcN9;(baGrr+I<$as`dtqAhsB$n&&`f*$0Q
z{)~X-kMn>S65+s~1zD$2D(v?C?yTHbEuRw7X!G^2r)>od>$bJ(*Yec$pr>D@Ivif#
zVl_9*u}GA6`a+>3kz{`>ly_l%dFET`xErmw{jAzI@R{j$*O>lY{eWzKg!02!zzM{X
z4i{6Z;^T945PiQ)lz*3CvV3CTfU0-E=d?+NLa@vz**CXXH=7C(z{in8Xx`xvZi*YA
zWwYsvZ2UNLP428wv8tTN06gCON`eZ^#A>qKjJsG-?$Ps@fd+pCg?e|ID7kW0?Y0#O
zRqL{3-J=yU$}q0l3pI;+rv=he?#9N*$eN_kj8%~Cmhz|7^b+RU{6n*3^^FZSyqX7H
z{`xWQi(dKO>&*J1p^Gei8~pi7;(N+*pQA-JKWf-cuWNWUbuDvpgs^^1G_zTpi!P*)
zxCSI5svIEKs;Gaq>eGVn{W^dRd;MM$vq&?gW}iC(ZIIWZq?lPfG4`gOVdl?Qe1f6M
zqgp$`i+^~gD&8FScL0`HK3Rb1sqXkB@T!2v!?+D4lQn-da|2xQD;%qasaG1G
zipwlez1D_a&G>%G&`T(cw(JTcnN|A~w3;{GI=h^eamar@H+i8k3flwMzvy-`q!>+H
z?BNUfOGw=U0&BN18O5EyjDo#Iglut~wQ<5Sqc3UIw
zZtLgwTujwva&SQ2>d;K{#Fu|%%SqeZE%F7qNMDS3+IPP%@Pf*Lb)U8*SdUj6pN*P|
zIq2*&>e_$FmQeBe1_-ycThK1dRy_iOkzCeqH?=x8x~ju>#OGs@Wg8oMkgcT
zzs?C7`6G;Bx*9*~faY62>$K_kF@b3517_cNh6z@h%9A({sY-uOeIBb;cl@{_-3)s5
zYJ-0-m~Xa#Uv2lZ>GMM`ec5*Xs#|yt(in7DmKy3`mYWJ?g>AoYdD*8VfY@J|xp1Ba
ze&l>}&+gVQo24t7r8⁢WfJwWpKm|78W@WH4$P$ltRMCj_1Vo$=2rQ(x-&`F{6~}
zy{PW*sJ_XeaTGjhpgP#yg$E3x(b?z@%;SGphcAB{xavzJFI5qX83vj=sChjTKGgf7
ze@1XS!)&2~TsstF){r~&Q=&`^$da>@AlXW#rOJ`Ip!{I_W4EY*TYUAk19jY`gGkgN
zqCeA!G6kr)@f>nL+mCIeq7g5yZHc*!%9!<^beBs;;zPYt$nNwr;zM4E<9MIsY<+)8
z^8Pgax)mBmj;@Hh^N4z}U7{Pl$-Pq`Vepe6`*C9A^PG{2=R8Bh7urlixv>?P*jS$G
zO~9yl(e5#rukf%-9MRPkH|oK~n|8wuGPO0Rv%SlXjhh9NgxlNRdmP-Gi}_
z-vKOmie|c8xU`JfUo>8=TlCG@3-*6#AH0qot0)LFqogqTm5`g-i0t6PsLo>Fe4DEmD6bcw!43
z*eA|d6+5jB6q%MehpvQV)M2wmRkIuHfsrz}TshEXMAwlwZ5~WUjXW;CeaNS1!WI
za+J15TNhf+ueoe{RrJh61)0(t3T?DR?j7t@KQ7Yu&5iWJ7@mB25YkT*C;P#Lx9-?}
zKtp84z&w=r{DdW%_@tnOVXQYRu#fYVg0F@~zPA&sBZ2rt#;7Oh@l1b*)oPpKotEBn
zV1TfiS*moJKtchp?OoWnbdfE8flp==rKH}aWGYG5>UkahT#|1m+jS7%=ilZWsw0-W
zyqROh6V~Wj6x6eIp(oeC*2|B5CF<*LkQN2!PoWWP5(}y&O(sKqmBTV~`nTOFcfKBv
zj=>s1<}=|z9gVmPb9H|WS6-64B32Kh0aXLMM?HtjS{31?^kR%9o24i>+NO`J00JM(
zDL2WRj3?bd?NB*iwc(>YI;7-qC+tvszttsg;*rFc83d
zp!`fM@MXQj39A)AM*9_k>1aa+gyUK%WIjrTWQ^Q-#eSQZeexWeeHFH%L~X|*AyO;t
zA8B3v#D1aB+G2T9;*@;7YAF$`Rpse+p@cQ8U8~yg3sm$-4>TseUeMdv3R!~(lj!Vs
zrZ*UV0N$rWCb$@blK{NRs@zlUWs}bP%yEvXp-#*o!!ml)VcvC4d!q>C
zq3VSKN0|a)%6Hv1rUW~6371#OOJA{Hzl-jB8Q|Vh5H%1uMN~AI&sIK;$}l!5{3#z^
zMtS{a8d`|oviSfPtcYq-bm-*ps!ty?%0N*zw@QDJJO_WJQKB-wD$tVuL}+p3*6}!Q
zS6SlmWJ$Yf-HD;~HjloUnZAng1iZW+u+UR?lzZ+>!oHBL1of#@UybCOvV~8z
zqvr;RqL#9E?WitB(q@n=9Gm<(qwUC1j)!{fPfmUt9?}Cd
zbhUy`?sk7=l#y}I&M7;EuiB8)Is@4g7LaQj#1fAo7xI
z%Zi3T%_a{8CFmEZU+YIS-d=!9IXi|u?H(=
z^tT^{X=QQWk!ySU?_J#BRnFfkLR$rF$wa%$zL1@nA!~m`1=GO_gw;a3PfYW
z7WIEtB^zzGUJhqabgb|Mn8_0DgJ*6HpTWmqOQ@1p)6TJXJ{wbcNw^BjO;1wD7G(1W1>6v>FkDX?ATr49eKe@j|WHQGV#(hUV=lYbd!VZoQed}>$#iWWG9^WvrfE$KZ^^cJ5gt!oPt|SK1AyQ=
zO##Md8TD|3dr|J@m(3!dqtEGb@I{#-pY(^+wM-V
z4#ndL&loJk_9Mm@n9F}4LW}q9Py!jIuxy24O;x|glVxw*+g9xd^+m_J#*KH1B?iB#
zL1{#bs+9~K+&j%s?sePowkLmcgCk){kvyI=rXRbOI`H5rSnZ)4BWzk++x#PQ41Ji=
zn3|ukh>#5F%dxmPByJ(IF2tU=(<{1RU|$u?++~6%OCCsC=Ujb3aVGjeveR;+-#m3
zV|gx;3z`du55mtx$$U)
zB(-n^U-U8Wr+w;RIjh;U9F-s9OVUFA5e^pJO91bRk^iSFz(apl4@in{7LvY!wR>#a
z7=2FGO5Prr!GR#_2Kz$TnQUD|mC+C$C)|qS90n}OOH_(zC+l@QuUm4peco)gcNv5a
zY%HE`Vzs{wY>k4b8Qh%Srsqz+kZCtaQFh+}na8%L@l}+^Hu7Z7nf}->o)||ELs$Lu
za`d~%P5(7XX*YjDY6Nw++z4XS`L7d4zg4!ihUHVXs0E3UI;kZir7mDQuQQ{YE7aQf
zN}tgz3zS_4xGCt%Z%t;~J|o5m0WdUM1(iHlyQm_~w{r!=^UONxqC7VVjXan6z#y`U
z%aMd^a11?ba)~WIUtzfxIc&@PY?}Wk?OPdG>JLLB0f&EGaoH-P-QM#N=NtM)QgN
z#`*X69F>Aru2i+P`iBpntlGV}{WSf2rRrKo8yZ>FuYxV-I5oa!s;y-g=Eb#@0?t&3T1
z#Kq|>Y3G8gI*o}1%@p&Kq5
z`a%0A{!xe5XnwV&LB6XJGRHvT%_{6;%OoV0DLj7_SED=|9@iMW!)%sr;mzGiA}t3B
zYjSp=4^7NPdJE#$iK)1n%@~O6x@8VCjks|b)y0h!Rpo9YHh#sfq
zdVd*RInuLz@vMXdr8!bj3_6e6Lg0X|K6W;^nofqhG=C}f>HV!26ZUbq4tb&-Iw!{;
z4_=N!hYK1xQT6{&%|Mfbe4ph$DC9NRI81-psh*_7KZ-K>2k+K8Z*1Hpy=MbWNkB1^
zmS;{4XSLEETsJZY<(vT77z%!4LGHOlm~BjRI11_|xW_yPv8nelsCPDdo{?*K6sUzm
z;_L7=R4=dq&akQ~4z4|R0#X&6MMS`U8i$(25C+w*+;8IvroHF{DT7QFKU3ARG5CMV
z^3L{K=p@Leiu|szwvUUaQ7Y_l|AuT5~c$8|TBLkykl{>>hu@g65Q6
zF{DTFe-9Z_Z4$BhAgfCb;bC#k3m%KQGj+X?1Mq>YkF~oy6U`*y6Z1HP@m<
zKBqwCZ__06NSpMM
z0svFR0k&EBcbCi+L~F9E_p>BX8M?Z4n_Wxok9qhL=^C@QyIEyW*X(~HvT3z7CN@%z
zM?)jUUS|jzzKxRX?bm$xwB!^lr}K;TATpIBZsiw#+dqKYV=tc8|M8P6&qx<@akJ7j
z^Y3;)-Q^4o>gt5YC)brXi})IhJzmd1hv)|;Xnsc}3$WYZU}I3cyu}S`8H2<1`HI`N
zg!pbQRqXItcSBmfrUrjU|6rA&z(?bLmz5hO#@ev*{Sc`Ryic8CS~*n*fP$8~jF)d4
zh5z@>aH&4>&5KcSU-1eTauW;Yj|MtA{^ax}lG_c5?XpDJA{6vrYPgS1h01j`o{RdV
z)J5VG^URdA55Pb_&;F?7Y%fW6VNG{Q`OQtl<|%R%Fp5#@KO=vnhdbE#unYVtj)mtf
zY#i$SJKuHp8-IeYcl20_SrG~a@C}&lkw@*cO1FfmsMES#ia1P$2*A0j`?{SQ
z4a~jIMR!%sBiUF%W2M`E+r5i-zl!4^bR&4_wDn2MGYBR15s*Ow?XAW9*Cp{T^S=_QvJ~I)^%dq9CM((`
zA=Y{N3MZZZ7&XaF_KV+aZNbA=v|M=e-h-rEe
zvre6ri3T}NFc|DNSdDiH}-$3h0BxXX)#xVgqA!xb)~%n@Q!>i
zhPjN8gJrS@v&Z`rB(%Q@qLN<@UG$ELJ!fI^u%DgfgRIZ(I@|_w3|IQ<3zXQ5no7LT
zI{6%p_;z43Q4e?+_wRdr=oX8tM8=o(4p~LGm;2rQKKn~P&Lew%MZ``M<7Rn*P_dOFX|?`fLxY$q>uKEw3fl34dga#%1+Pq?92}I_#en=pkLL+I6PBZV_eX2Owk4wz-nHpY>n%Bi?jgri(#;QEX%nk<#!{~aGzky
zUxPZhRl?G*WB0nzhz}CLo&vOvcjmh6Ft&fU)P8To{(C1~W%|n!7stEe1Q+9Ers1wrf%kGnRZN*G
zC)HcP3&59=xXVTtzbW*^qjriJ?ysq;S3|q_3e3i*2S?NEBD@mS1>-8@$|2ErY21H<
z*Hm9LyfXvcirBPsmA}^&9x2o^$Onxbsjz1adxvpT3DHg9@mRN+!w$@vGVkHLBFuDL
zsHi?%?7%&>s;)Ne7AVR#EE?gZ`N`?@7ALlI8#(VbfBNdf4>0fNQXj0o_qP;fO6@S(
zb6pHj2xS{fk_^r(Me+=fWm6d0TvUJDWkd}#w`J4=m>*$MswhGOV`ViH&4+{hp|kvC
zFs|V&_0lynO>L-9DlmHf(v!jDBs_()BcU6)t1a(tpZL0<1JINKTPKIu6$cW_u$jZB6xI_oq0k7^&(@EC}^g(c^4
zB2qX5xx=>_1ji}fF-I48vZ`8gxxISuo`4=V<(ZxLid%I7+~_o9%Gs2r;vqFpU3rzM
zbEdz>hV$`Ap`|}Wse^xFHsI-A+h*3V1W4Uw$k5W+?83k}PJEAc{P(ebfjQ(eAyTqR
z;dG;EYX!M`{&lZukUbYHIzeG8a~!rwAz9%&tS&C
zO{p`zU1d%)2WOqvRxi5~KbkF!7L}KmSJLiJG_{}jc_LI&ZN-HsXbOVYtB_tgnFCMp
zHxw!K>zoFgz%Ey(!<_OU@b$u0Ro2o=D2L~mP@22@%h@?FMSDrbQdtXyp
zrl3RxF6|C<(aL`c5R}cidU3}+Z#d(CX4fclVxmO@Z^CnHxX`0zxhOKTzN*fQ0;F12
zwddo{nvMGLiR3~$qqKZ7mECyv*;}B`8p!_i6ElRk)l7icf#~naIuA)w`~7PFA|X%P#a7
zPPl89v-g9?bC__Wy4=rGrq&Koag*-WI=r63eyoM&*GOyq|@_E=J8vbUrga{k&z9iQF^$LBB9#hFD
zmfuGZi12@{FsUjRC(WNmi#Fr9DjMaR9!BVnqQGDi*+xzzS!%`hK-&0goFl9wPzl3=$~qeVJ_oj=#Lfw=1Z#wU+y}wM1
zvVMPHV{c(0*|M})w72K#rJ)mNWsGr%&;{P8cXusX#mPCCz*;@AP!V=}6hc^=H@5
zo_PK8dQ;i}0jMx{b+P%A|2Z1;>%|?eyj*`~?Tz!^UZoQR!p_Z2v#qVo!QMV`#%=OR
zqM|H0Ir(T^f5r)41^0}<4sB)wT+N?x%j+x?^7DB^)lbOuDJdwXs{aK|Ku31Ne%ZXafL-OtJR>

diff --git a/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf b/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf
index 438eee9a95af5d785c02924d7e6ef0c0dcdec7ae..85c249036108e603d88fd49322d03d3534e02db0 100644
GIT binary patch
delta 133625
zcmb4p1yEc~(=LG|5P}3JXn^4E4hin=?iyScKLiaPAV7cs!GpWA3qgZBixb=(7KaPT
z`_--QuebiXm)hc-owH}Ud#0!RX=zL#49I*_&JPbiC{-m0AMxzz4eqf)(vv;}tiVOY
z&_%>2b&H7bk+DyBVllAbx!IDIvf$|eP1%@Pu2+x^%}pAF8Vzi5K?QMD*V*StO4YNl
z&!_KXs|PefN&TRsm2q1D7TfO@
zL*+F(`RBKcDVDpcQ|&Y+or=KjUHmD6q{ahsAr|(#o~viBEF|6P&D=9XFZ;$M5-RpB
z+TkTG)Sbrqu~^SNHs^ky|N7!I-IURf%h*`pRT@`JK5(BVKRgfmKG$bKFHUkz@fAZN
zTR5d>NvjJiA!M-?X`k$wvzGP@P{o@uLpgAEx9TTPM7(}M>?D+Io<5Me>xk4FVLOm3
zXV#@4d?_On22t6h{}L(LCotJJvtkCbNC|LfAwC+XzWEx)Bh9Y09qGaWe_w)wLa2$1
zX~)9UAd{vJv2&uyF7XFo30qWg98I-)zurDLtSX`ai)SuH%!^8|oCL^MBUzCoN;1Cx
z+B?TCY}YSZ;pg@yq72Op(9ydTQzuu>qRNO)+*ceMJ%FMGSu}QZ-n9X|D~|
zG6P1Z#Iev$xpSACc!{L6|U8kw&;0XN0Rh*|^;k{5l)E~4+YU#}Wf`V)x^
zYUFy}IHKUd%T6{V*Kwy%vf&Sl)iFLzeM9io{fi(GR7am4l`%_cmqUT3t2rPm0)cMI
z|A1vhMug4rRH=OB9T0@@`bGAO0*>+MJ2LFMx6R4L-0>3}y5{?gLCV7z;VEEmxAL68
zG8a7eFXA|~WO0?4)hE>O@$wqo-%hdMWb*>DMWBJrk!MRR$$?bnv9DyR4H9+7wD1PI
zw5wzP<4R&
zn|MA;`32Pv?`?JLZ?`ZP%Ouf6sPVZ
zED{@={;%!@t9+tt3(C#LXn_)BsQhCT%o&XNr30OayQcMsP>=Z_HsM%|m?tDI+N=WP
zPVe+upP~?YWh0&!KBK)!iV#Eqd<&`~V+g1fWhdapSC}KVAVE(Xn-HSkPn2^l?tH9I
zS0rSYYH+vL0+7yd-?XgNC>cZ_W_X6P^O#x{|_o;L
zgZSQEbwPJ~s~FmPFl;!iwkII$E}YcE*PF@)w#YmMLF+p;H
zqQp1E5Yk%j#+)33Wwe*2#)Qhl
z>#e|Dn2Ue;#impcxh(Qrz2y=TCKgv1;I4;iO^e|vQ6)mB=zE@v_j71v9dvQ&eSWum
zJ2BuuQxNnCQ@K2j^z>*7P4dgxeEwVX-XHy960_Mn8V0UBg>mmX9T8agP92Yj@r5Wm
z&S(L6G1@BN6!xoNZ10dl2>=9m?ax_OGa24m;!Cd5&scAUL~Qe1B(vH&UcZuvSTgeAh}MeQC|xJ39J&oT*l>4mZ#1_gU{NF)++<7+F?Rdc=5Ux_>9RhxH&nF%Fv_D
z0?wQ#A5v!^d)4MwoyaUpzd!OTykgCR@hg})Wm~F(6y@GpN{xG^%x0z}ZwEZ}#6R6R
z9m^7LoFtdeW^)qSn)zA&ZWEWs^J#_8q`yl)prlymLm1^H%cC{g@vF+T7+LVUD
z=WLAS_}EKhFFGl{!;L;O93_%P#F!;v?0$iwrNgeSIWfJx9d;gF-?5WOOI@y`JAm@C
z{9fR_8FMoyK=sDeqQKR53PT${w>6jI3vSSsS`dKxJLOWqF%g?qQsgxkn%XN1{13%|
zuJw!K@bXavta^sm=}N4>=!U;-py=k7rF8```4u=>4y$}u;^qe>%EG5I9p^IALXl~I
z*d@Dkkv?aPk=ACwL1o~@r&4AaiCeNWBm(ytCSzzo{-M`Vl7{9sK#mWF8#Y;
zv8N(%G_3A{j{YWHMcduBEKZ^;R*H3dIB;x)0P+&&C
zv{w*;^*y%X@NjdVFqA?EO1y^2Zz871c)sc}vgsC4RWpJ4Zf{f@RfYqb`K4
zsxp_yTi1r|S7-xL=523XRN%Yt*6e`G>6bNd`dfSnVFlEu*U$PsyZ#i)bhh~xD>J&(
zdb8nGZtRR5$vNj0&vI3|r?dU*MTgQevi5UgwcwKqPxTujQe)_z@EZZwV*j^BFPY5-
zkD?6KFih*#3lN?W)t(0DzsQvej^k(^RW1360i)17l;V!uzhD#?Rmnz#sIvbA5J}>A
z4h>Z4`^c4|CQMKwWkaE?_=onBYBByl=FfZ86F*{q3hMPO+l!xd9Wp#uqSK2^ATmv$
zxYC!<^e}8!(`@^J>Mx$5!P+!)-aEu0t^PL1Wf&?f%y#1&G3H@Slla5bX!Y*(q8>v0
zFG)x8r2^qvSCWjSf(??-qIP%QK-YU##QG8T0zOoR!)Ka|^CY>~1dCXd$y{|cC*|1+ID~`=U`%k%*x6c--O1B&c@A!%q(SYZE5BDhKHRa
z{u+%DkTbWpbhUcJ^WcS9k(ni|Ke(DZGfR9hb~P6>H+3*GM`r$eRWf$Aa}f|=7IpaG
z;Oz2-{qe@l&fet>>%*+fBG#@hO6Ja@4t9<%cq~u~r|nx3i-Ib#rr@aAUMR
zsOZEsm6tn6c!w$SB51)oS?B-kAPDSO>F@BL)&rMvYx@`;@@c({i@
z|D|8CRF5}sue_c-Uf!Jk8GpC=?D3ob@XN;w0GEmPc!d*Te*FK>=zkBw1sOa(gF^!U
zx%@NTzYhK@jei~dhXn8s;r~J7FE{^V@L%Ql*TH`k<6j4X|CPpnOZ>md@!zZZzpmeZ
z$p-w7RsC-o|L?;6M~wg4&HoJks~CR=f&W!I{*mMVL%9E6Rh_EfUB4t(^Q9ePrI>-LRDNT3TRs2j@-}S|OwGP=}oySj>IF
zWKH>p2Mi
zy+80kxV2Rm06Esk6n!{YU+c7eWk^@zJpXfI*9rVR|M67f8wF1^XyZQyTHk(pnV>yw6Ou@oGhKfLi=%l$PvP21;Qx|RL_l992;GmcVT
z*v01M#af-J#~bx`D?>i1WPitC`J(m#a28xSwRrdy98Z&z#e3$>tw-zdobThN^qaq%
zfDfriqJrCfn$L~!__Gj3uSav@|D@|#qW|q*K3iFq$lXp{owA9IqSGI3g;DFs82^v9
zGCkN(&v7k-s_SVd9{L^|N*dou0uHO6-GL75?j&`i$ch^HqcHv$Sf003UCb!%Zdirk
z?YzR+w%*^E7ovgl
z37CdxuntlZZ0?6@262zFL&-!>^WTjROl&)KHxvXjdfrABe+t}Vt*evS4g>#vNWqne
zKBs1xXKz^__je>n#=G>hAHKK@xaNJ3@!ylc=GmfJs_S~eOZ#L5vhTM)sFFFs!Hu_$
zy*F);XzXSK;0SHb@axG`TMrZ7-lr|
zN+N*AO``WWvG9!f?#7Ff@JhSwuhV+Yr@FnjmwS-n>Fpyuu;Q@c*n)#62yT^G=5!9z
zft===Nkd1~bj-UihWj!aZ!97@Y_Y#N&fj8Lx+Sx^-2shn4NPG$-R8reFP!NUsz2={
z=61dnf+^WRJ7$wZ+mi&HN9OMms~7--Q+tD4FA>E@SI9D)4296=
zj!eSz+HIo5-MNpBxYqRDd=-`!DAqg~2{sPPr^ifIqD2H8k@FI
z((J=;;t`S6OM6DJd?RT&-fnG&A=Z+z_Bn~T)u^j(-ma4E>01jq9L0%IoP5fG0U{;kQcYtK+p-ZwdhglnyJYL0
zKj$^Q5@sJF9Y7_1Ny=l*V?K4;yc(AF`mDn{=HQ}qrG2n~o3Vj;TGU3=7KElJL^0ut
z>bQO+~H2d|sZr2Z}vKhO}&QpFOKv+x_+$ys!kf015QATpi?9n!cmJIKr7y>JCfSC`Gm
zSf;GuJbez#2@KSknwVcEm67s)U{ycCBh49Vq^E2YJn0^aIUZtv>M-U2-`G3n6>*%v
zN$6K#{8Bj8QT~sH9L7S?34<)+3&YAzja4pqQwTH+(?FLRMC?7uU%1+@D-cgBO&%j
z1L?4
zE9-dCfg6-VL-)11Mn1+;V#LWA+t~?MK2DA{eP6GhpzpIV9sYw(k4O3@uLnlV6QBlETa?m3CNv(Bo9yaX<_-ZFB1*|C7wH^-O4-
z*&O9qSsD?>-qzba&<&{i?EE83hW{Y?%-v0)k=yraXM`+V=JJXeRMHlZ$MW9Fa#RKw
zv)dqn2Ua+9yeGWx=LQmG*99>CICcDn-5Xx8N?446;Yhob=E>Qn9tQXjx#6TN0=MZ5
zeC!1doTf(XQw9%BLwFJfUEh~P6p1ZDt33GVr`@#C=AXmFv+X7bTfqI*-syH%eFLhh
zzWpSwt5_V@h2~Wo#GaSa2s^Vv{N)vY#PP>S_8OWY*eU@4=ayCbla8X&K?XVIAGD>K
z1>-ub*4t-cY*oCclYL{e^*rE$?1qN-(-|`E+KG_9cLSNkLmzCg2L($SYZi(=J=bb);qdBjKB~^a)N5gV+3C6){dqqDy#+
z_E*E?QGkR&BL14WXpfgJE;P#uv+XmBzQFvP_-r?mjvdV;kCG+9=r_aW>!7Sf#)RZ+
zFN^)1bcn3RU<1>15yc#}LZi@|^9q^T)2b(+zW_i;Y397d_l!%yngD;>w6|G2!|qCH
zWXCtqj%Z?R8JSLATZN0t>%+OWo`}ZVU&X)|Yd7uW&b&e0vKL3$iVaV8Y$tCIn)pGJ
zw*6HcAP1^HFs8%cx?k)zRjx!GgrdYf>6&JKRsDu}HlRd4qKJln(&UCHuAaQMj{#D2
zS8G)vyGYw)DkX0Eu|SoN?=nP-FjaP&sfw*xpsciqLF#Cj{4WA-l3vQ^1~QKtrSe%d
z+hXKuz$Ly
zL0O%()0!SQ3H4r{7l>XSpf7Z!#BWZ6(^#d`dG@a)yY#iSlQk#j=*}a7sG~Ek1cR?9zQk!o
zmM{KdD~xw0J3$&vo!b!x7bTS^BuY#MN2*xsUop$J<}Aq?7YL^X@c3q%>%-?Hg@l#O
zx!$esOeA~Ma3aDIp=Sh8#w7KV8Ft;;WKe&jXArem&fAc#(1({_iD=IcAC=}-xI~#f
zd19)PMV(9y7}yOkGv^YUrIP>IX^I2`2ZyqdF0YabcVs(TW>q@5I{wMR{t#}zvlTZ(
zOG^-cejP3U$r&QV1KfoVs3+zY)O4vvG(~M!@8W*RIYE$t|xK-NKQJOex^=
z#p*qg$i^APw{Q?qZ(G=`zvs(waHzWiZ$Hc7ZqkYcW{HfJuRw_#d(Aet*3L#Nacije~dRAe(gWNnn)1*
z4`eNkP3nI+1)%-~?rgVXi{>TCgWd7X1*iFIT(n?&GtmMhmKkD_wv3Yg4EELMa
zCu0vtfiouIb`c-gdN>tZHX2-zrb}M3Q&}nlVyv%Hfv543pEYyx)o5@ga0bl35u^P%
zyGDjWGcXl>7j?bish3p6Ih>qd0OfCvJ@KpN^_ubY>46)%jb>LnKsU2+z0cm#_@u>Awo=w^2v&ciQd~i
zr%rFzYPpONj&fNAlGWqCct7M>n!+`#%ksW~#&Ov_`_nF3BjFUuICzuGeDputmi!jVx5m*K5xxA2`dM_m
zf4ATCu0W2zKZJhPtO3yrkY_QlSWvV2uAy(gBZh;i`w<^mKsPOih{Y#5`2m{~7|J&x
zcL@=1gQ!@5;o>T1Vt96gW1vPb5&h5V`n&U*zZTB*-QWD}A-a8AI=FAP;O86<>QETb
z^UdiuQJeNQOuH`;m{y%pQ3hpA9B(`gu87wtP(K?U^i>#>{fDtvS?UuvgL%N#S>sD8
z0~}I;TJQa=u&dFm5rRb<^?BXmT3YZVFr$|RZTOI$m;cWqz3HHyH7vEJi$TNp9qA}4
zi6q(W-!VF(ojS3#5!wc+R@O=T)HTGn!HnrXL~y{#n0pxlZMeKC1ox`o2)@)S49Ao?k&rGiFZEljjlPit6O{>>$bkE{Qe;f+&kTkzN4nEQp5H+
zD>F1Ks3P*XyWB|(tyz;Z|EUWM5!`NOHCUXDLakEK==qhJ-uWx$D$F1qQy;*xugaB^t2Vvnn9-J`_PDd?
zU^Cq#_*G`MgO|??Id~c`6TR`ifzp?10md(W^6oeL$FbDMbf)gsX(NbBut=#!E&p>Sh}e?g
zxgtBBV6_Vvq6?S6*ORLg^;s%-n1f?ddhTYP@_2n2OgXrOT*C2-xlMmm`7&TzVu>b5
zOeg7FJ^57{p0w;~2#kTDmnQa^c1987#Fq>u0Li$WL%)T<^itoJDYnTr*ot1viQK{A
zW3Qdwu9=}}F{*rnUsw8B6Q2;}z4eGEQyvxF?A2|in|Vs>G*(#}@wx+D)%Q2@WQ8uA
zdVBJJ@*2BS=!lvdiKIvF=Z2gcr3-n4r?Nv?2#f!8VcETmuYLnhltT(Pdk)cwwF%U+8}yAo#bAxCb99p
z&sSfJw2jFd$Tbu`>f2r*WPj)nhy+d{>y_pNgplIe$Wk1Yu;KB&tAjdpbNuwxCbyY+
z%{%sj6`Q5s))haSoM-umZkeUn?mC7)pReB%^5G_(t(;p|Ewl*IX5GdwUwmxw(da~T
zwfO7)jN+Ta`$^G!mw*5E@ZI*cXF6ur#|Wy`|f-
zYQ?qGVLy{XI%X4*`v_bjV~~({zlklIoTSXMK?@Q-Op{<{7ToRqWR41}8%Wm#ec<#(
zLfhN4Ku3$@@WYZP1nB}#cM-fHMOz;95y*1J=zl>NOg7Wr*L#JXFU{Sl_jrVCW$1&s
zL@;$~+gw;RdT+6u%GJbDb{_iD-itl6E@sX=J$yI&^6BxPFJAgY9aOPFT~SUWIhe#N
zjAsewlVTHKfm;g$&Xz!ty5QmLVl7Q^G;htI<1L*07@i%l0JPkBClR{RDwHy_2%+@o
z+v^!dvEmlJj0QDGCrA{$9+p|c2HL*AcsO(ZLcWbF358vwQui0;s-=W+;}fYV#1r$F-50P
z(X+5z-|3Q-)g0nj^cP3*h+ivym`(sMXwx;3ojobYR{OR!UNnjC{i-IRGfGe{N0}~o
zv;L`iHn*C!>UwPF$9R6{569xRwSJ4~lZlD|S6Y`|qg_kEv4v&u^R90#Q?ME?UEkDF
z`^lZA+OLe1t{HtWYcIORQ-e{>fhfu(jdeP474M|_-MMGz4)@y!UD4chJ|R5>qqTXc&0CwV0u^#yNvKVhIuG#2w2p;NMYngP*l1zBwQ4
z$}QLfA;n*%os6TdL|c%SV)#irt;mDCaier047Ns^fAz>+vJBc%wPjdRnCGO-zCmSR
z1v-llj*1IsD^dg!C@)_tb^r-H{Dh%*mD!syNuhO%!LnR@6NWxIl+3Am*z}Y8GC;8zAl*kRV~F`rHFZ-)8WEEO{05z8(*?3`zN*7igxRBak4ww
zo?Z%p4&KTNXC1$R_DY-ZMV_O5e(9SwF-)J7%vK;{UO3uy8kiIc3~|b_diFOI)p;p!
zq|fRB$Iz8^N-Oz;lw;&tBF%5x=!p4(1F!4zpzAsV-@FZrJBmBs;tD?>|6=cXpHDr;
zxhwsb?j5$8dD;AZ<}0+83?N2topzV%40m!IS=6EAvpl}wL7%>Eqar)6dF>Y3&|0py
zoVpF;jgYJ$08+b7u7hAiSW@x6cBswa{R+$V9AHOQ4A*iHtxo^)^|<}#sjBw%zWkDr
z-S-6(crasBK}}|w_tyu+bas1%IZKViaaXJ4bny`pv=@f<6>k~mH-ng}%TidOV2zD?
zHhYXD?xVlbpy3R!;}XiAu*Rh_jKdb1F5ON{_m%>zeMtFWdC3a%d5+-iUa>LuuwHkI
zS&dle)%d|ri>Le
zN&D0*c4MU?NX~m0OZj-NC8qJ(a%~h(8_F!#;Qb9QG<+Z@XW>^_EQ?=d^A}j)maG6H
z^-`hna@WSp18q#FlZ1PhyZwa$H{v>(;}?*-^p~aVx!@}-Tkayp?J9QPjhYv-`V0}p
zzixA@Wou+J7FB$InzZ)}DM+6e-$3snEufapY%5Xf7|3zJk;~c{A<6ZndGJ|C`O(~x
z0eN}nUV{7+!)@+!=X>s!Vw+c^7heFQmaMG#n%l2=2L#*Kn=xEI{Xzy49MkiTQKvup
zk|3T9#*rG?7z8B>*~+wqwpHjP`nPs>#xUByaR@V7+FZr$g`xuH1|gonn*^RS{$sSO
zfmGRr#aiRz;FtW->*m>|P0LLQ1LkW3HVu~kWpb0enOp~HhQj+gX}t~rJC1C3Uq2$}
zp4@e=ca8A$%}R40-*UycOE0meosB&WLutu~lpa_R0YMPVPxqUZN{abd{9dK3gwaA-
zWH9zPFNa2SqwqX9$(OS!vj<#45bhw%d!H&Xo6p-hJ_{afvA-CK+#v~1(29T~xS=?|D_GqYW@vV8&H$iJ
zGmXWjix9}|{0zyuyyA6%$<4sWt{6#lV0w(*O~!?!p~hz`?PVzmDdwD@01b|M!)0pA
z>>W^f8K9gTXM+=zu8&s;n@OegH_aCs3QK=j?R9eqCM;SG$h%Gc
zKygF6ZKni}~Kb4jwf7%uKk=1xVK!qe`q{-gi
z{`3VkTgiL|BTukIn&D?NCW}7U&@mUx=+CrAmR2e`kZogAmiVf@QTnv5BQAX}YgHE6
zq^T>vt(c>0xVVtFRd6GeG;6RCt|RWf=-z6$*u3i_n#ag<_$%9+8!8io8c?J)GdgamwUe%H-Sh
z5=dSR#Bgjw{lj2Ux%Ft`WL)A{(K|!RYR#@h-zJ)K61k9Ude@c7b2a>!O7)k_R?>;0|t5dB}r?O60Gc8ylw?s}DUmg3rm6OcZ^365z
zYOKE9`ev<=tC9<@<%Ny8g9qnR|2W6ifSL1)thY}P~0plw7j#&eQ@WC
zUT)IUVD5ea?ooXUw}ADT>=t|AZbt-@`A_PVM^tF)Sj5y;-(<5xoHl8MV6&`wcjGSc
z_FK^oaH0G&J8({Lj?$UBK#9JHemmm0mn-3iHQ65n2l;v>!IkMDv|_El*TQr3n;89
zoX#G~cqM52CSpC#;BPC!H*LcMz&F31ig}oRH$1rx?L&{33bPS65BT-NtP~>^%Ls&Q
zkXRlW2ff4Z5o%oJ!7g?zA^538f27(R&$04wIB^l#Oj3YQ^$Hue_JN$7b7NV*Os#tT
z+U5WV0%8NMmdajh>M;;>xiYjWX4G-e8eN4~ffX~Extbbcoah`-$$Mfr_
ztwHk_Hh#0vPzO^<$_Z*G$g7(E$`O;x#_($UJK_7@oe0R(S;(v9ZC>Cs0TFFfC9jL|
zdjad0ypG&aB*Dx-b!&=XjeO4s{7M&PBGHybTXBNu@ZRaYsS0^sKHhHL0P9l~^8D^+
zQJKVEBEa8e3U@dM8!9F29bi;CPZ9@$i;LFhwiw(p2Uw@RVP}i1c{%T{!ca-5^jx0?rZEU{t^nhet>V=DMcoM
zaZKMl)|8k)UN^kP5Y~(j-j|2cr3anw2=F*>3weQ@7$`297dTzxVI^BxYAJS-&}(mW
z*H>YO=4DsjPcvfqRuf6*Gf%71%3hFZ3_@1qC>LPYywtHO{_FvS+7C#=Xlit;`rw0?M1g%8W)dHT
z9~mnA-y-_b%5@;=hPv5RFG5+$0K5J_xxvT>`nOxd`Dkd-kK4P>kpKRlCvr670OR9S
z`Oq}syYQgj*aT1iCbinTsb}wX=e0<1SZua(Y^_lLYAgz=x3k}D{7Gt+n)^xsA`vkq
z2hu+s1!3EH;S>A@7$U`AJ$1iZ;{S&!W08C)xM^!zL;a*)
zhbVl14~H*h4G#w^$Z9wfe^i)-#b6KOC)4=rbf#BfJ0d0cqy>#q#eoHl}`j{KJiPf>RG7YXTLD!x7
z0&C|!%y4}7az`zw{@?=5Lcd;G0W}?YRik%uNJsp_LT!DW$$6VudD0bLAXsg&3i!7X
z9D=JEEzx=;a)Yy1K$#v_M}*Mly0a)$MP6zByMwr-KrqE6(_`xmFjdez(-nhe$z8
z?=!F42w@?Q=C^>I-fHCRkZ)uc;^PeY+X;n9?&ibY5kFV<--2x=*fw}IxoP+BeJryB
zgbYxa(34N9H2Y5pZ%)1e3fDI~-~*4wt2?_(TM{56ooNH2#8a@-p9
zyP{`Fwe8l#3R8sHzp6j~bpv8wJrIo-Xlqiw*t=Sfv4F%}rnnfV-qz($#^Rv1Ns5Z2Plif@G@z=`u&3c`7BV9X(&@o15+
zdEQvrQy!Th5it}Qyuu>j2sqf#Q=6@(ED9oRX62|gOVH!vwK&}0C?2H&78foQOBAY%
zb}8LcJ;q+1`&H~P)+D;nv$R&qjGfR$-`jV|(_*MvM=eN`ct_Hc!hQ3cMeaS`PO(Bn
zd7@eHb^&VJQF=F19WNJ%OCafoG|BDc>ZX6WcfBzXrzw8M{1d-VyNS~
z2d3%fRXX?c`|wuHR`7KI(A>6put)rme%}>Qu%n2AERVk|ukb18Ck^l3
z>R?H-uoR^NHX9kabMXjYZ
zHq{*PQ&S=Ds$|;H-kdHMCo0#2#zqd_M*mr4?QM}jxI|##
z5s{Du@xRT0qGobhty-Y4%08u%+QJITy4nU2U-#nR{Hp9F`SO+>sXH#&9puSfPg$f2
zn?e1BeLqXtnub=)N7L=k?)*jwJ}BDju5!wt3#y(_HvjENIt*-S7&V=Wk($B5H!w?h
zUoc5y8O7(I(oY1q8$Wgl5`99z-@4X`Z1Fz8^0-FSuM)88`gse?X?WK9Xj8yqR=K?w
zX$yrge!p%sSufeOYuC9*=)2?9UPzP=kA_?T
zA3__a;Z3`OH4Zvq4~hxCQXCWPXs1Fp;T<)BMvT$@XSQR4%&D1i5lz#
z&Q&g}(8iyu3dBF(wo-p3Mfu|x^1bwU;iD-BG+Aw-gV)inBop01;)3oyUyV#p(}dXH
z``uWrOqa`09M;Yb^oE>s2~!98ZRMzsJjot9W*oV@Sad#L4hBa${1}+3KCc;2R~OZw
z7z4`Gb$%)eJuDsmK4Jmf6!iJ-JIUo@ytK@7v7;Va;c>s?YcrtTa;Wv?T(sY08=hs%
z1Ex~(m38xRJic7ftgJ%ySwBPD_Jvo>ST)~(uMqZ
z&Z8$kOQS_L3yW-PNohGy>INrPQeGH8bcmH=l2!L&QTGW8>uFO3sTvkx5@I_7`#=0?
z9H6zYgG*~F&a9L>%erlQuSzplb}N)OI=#Fbd607}%j=JK5Z0*s^U%?d>rLd8a_S^{&<)tnrH7i=THui~Jct89I3n$4I$8q{2Ow-@q6yL*4*S&?jtYuo(D)v1C#h*3cs
z`+#DHPjfiZ=>>PqB&5dT+>%qjTVKFolrz?@TP>VrKICjhvy3R?%>e^oDE53HjmKgl
zwjJs~vz1<2#_P~klP$=Y%6*g7z-Z_T)Qd9)6`e6N@i44!y~JwMCvLh+Vd1(;IpGLB
z&zrBRs_Z7P$2};;$J@85=r~!Y@w;=scA=gSslRpWtab@7)8EeUb-drx4c&wouAOye
z99(6*S}vRhPU8_PXVHM#o9fVeP}a%&;U`b%?RDw}TJe3m6}8rUj>XqL->ofuhePOp
zXo)jomRxT3K%KS~JFxMZHL%;`Fdyx>XWUBh;CIpN-vpIOhwoXbr6P}qMqQbCLOX1u
z6mo09h5Cm2st(X3$?%q~3rT1WS6R!LVOcsTHlf3Ib=zgEFd2B6tFNh@yH`)3jGKIM
zHh{&5S*Cn3<#aH3#+h1C^4T_Va3{`#Uuur2$)gKpXx~t8R?oCM>fZ5kb3Y6!#b>|0
z%A6>%Jpds%At=2Ci!XD=rGbu4aXvmNTi_caKO&%%`*t(>mK)5IaFA=5`r{!mu>2Dk
zc-^!WI2GIgri3|3mv!W^7hhuvi*HW1YKm;-v)4@X>rrkO&F_?t4(#4M!s1q>>yJHM
zTLBWny!j`EErkIA4x_Tp1mdGVzS2{<<&YOBuC5fA7jbFK)*78#GDUKB78EWA48d@4
z3(Yqa2wfC%Qmq@Cn~ya{ywn7h^0~JgPtQsP03RS=Yj26His~!ho$sWXzGHlHXWwPo6tH9ZiqTmq=NZBW^Cu(dv<&
zYjiV|W8ACxJ9I@2^GnaChP>BkFep0=l<6t^zAr(D_&Rr7olXp0A3X9CPd;4c7<6vc
z0?Vu^-ahQ_YJY__+)sR?XC@BppJa-TFpVexR1MUH~4!%cbELM78sg{8Q9hZ5K?IC
z4=&;CgwHN=?cO$;oCjzF#2eT!?
zmf!3*Q%|&-#zJk1pzS*1&}o9;WzFX;PQH<5OyA?KE_vNJ8{%b@5ja=743NJ={`~2lO8>(
zA>(}S>6C|KMK=gJ>mxFiNAB!8-AU|k;et>Uypv$;+NqP5Wm$FK@HbbEJB<&#phO`Q&Ma-g2X-MZoCFtdScAIAaET3*C#u)(oQ{??5J-(0=l2S`B?K3dk(Mvrn_>A
zrB6GL>1*>bRNjGZT%_K$WGLscW!HM~q#7$i$hW@U_9I48jse~xNm2sFAo^Wxmyh1@
z)V(gJaaP0{imHTRIiRNxvAY(AV&kE85G?JsK8E?O@7CT)d-_
zm6kz0R+hTqkuSMAwvM^{Kj?e302aY#>1Kc+%V{Oo{dnp6)bf#CO<)dp_EZkWBvXP!F1Wqp;(d&
z$Fw!@I_`Cb`7~
zm)2f;LwQNmhbmi%Uqd&9e(WptLfyjLOsvPO+C3L$Y-s~J`8$csx|y=yoDJ?
zf_BJe8KeH$$bG%{SCCxR8-%nwwq-LUUcPr{inuOC-4aVvT&+xCog$l8)#E
zrcQT9^Tw{-Kb-c|CfTOao5FW9BzsK!nuZR}R>tF9N^
zyI0M|`fDipA#*%}sgPa=GIC;>M
z@+qgYteBD_7p9>fq0BLR&3vNT89;;6eNs!F!(f>M7n+MAvOLcK=W_T$71BfMraUPS
z0S;zrQ@u5%#^>Edvk}4+Gc@vI=+(j9{<_M_`|7G8HN%ST8Eveyan4}B6LmdtTOMM?F#*=Z^JwHF
z&a6GPoTp7OxHvaGAlc|>%)2U7^SzwXqx5|?bZg!6m**KQZv6olgN%|@wu{%ueQw$h
zebJ|==KH|unJMuG>gvSqJm7J2EXfS+&A&B$*H-xzJbN5TcJkO-U2y!TwR-SkUJ-2F
zL-7gsP$IoM$F~b|6%H@0;2l9wm}S(7U^#3?vfTM=?3ZU&LdLN
z-6fqvcPrh}Fhh3@NDLj{@cX|P-`!lC*=O&yS3WCFf(bu~iK!BkEE}mYaQLq515zqZ
zwC&>UE>+njWc~U9mM6>k0#!z&_MA1%`ClPPmzF5sS-X$X#*!A9gT;V;(Is?&o3Q(Wbx?GIb48~+MUp_v;
z5}LYfCS8UAjUh^>v4|O8k{;lcZYtvXF?BM4!QC2vibQ*stdU4POT%dM2{W$e@v>Us%
zaDTECk&@^enDA}n=BNJHR=-g{hYXK)@Eb#4BM9nV#THHiU^QFq@uJ(Hb4XikA>yRN
z3PEbJwla<00Nu;9V+W=K6az-W!8>Yx*{q;W<_n$~%_rFA7=~fnx7;d+Vj3zwVPj%9*=m*BU2$Ao$#Cp64Wfe2T{HNh=T~-Xq%@UaD
z)qdft;uh)uVnZVr@=8d9|JA%(-oPDRDNe=+4$z3RO<!-|?QD`d!ZJmp
zfQt?&>SS3)k^lv<{!->>UVxl!Z53c{HUB*YAJ@v4F@x{Xy|{?QNt}5rLAb3UuqvfA
zr->g#KBmJwe85GiH|2AWzQIkgJ17fBF?ifrFo%)vMgt)Uhr`m$7(B#Jp86t-4=E2>n*HNBm{12_!xWDT
zu!EA0o(++xx9|P;<+kQrDVm_%K;45w3M^)@K_?QnMHoH9h`3O2_3d|4&EoM9c6A-W
z38$8A89=BW{QLLm?abYB=Jt`7h2-UmXqS5)^HOW`M2Gbw3AQhqZtVP0Fc3rfv^Rdi
zqMDHfBg1U)46utpK7YICHhru#4z>9)ez#GI62lAm+M=rS3}~!HlF1%Lvnx)NY8rWKjgD&K&DrN-PYbZ6efCBUTjh5MND_{~
zZk0=A8d!^1NySmHIfv3igGaA~Kk~Uv1=amU#JMLP4{w6Rm(RwFEZn@8y*4!?-^miV
z-+{Hya}*d(Cl49HopD9XClw>@&Wkpjf-F^Nh9Z9J>-vX+MJ|?rA#13AV#Q=5Ay}-C
za~W-{Pd)GCa8bZp^-!CsGUi5L=O9z6TmvRDVpW0lY(&?NlJ)0*VTy$uaao_G?zde*
z8{IS(XX}jLFO$p;7v)@KFD--lg}|O&8*A)gyj!G?wav$lH}om)2ZU$m(>hdwifYNB
z26*ll`y0{G#lSd3vQ({5fA?I3PLJn1KU-UMtL#ex6^Gty<>{uQ8q4)xB^az>hn?+X
z1B@RM+D%DZXqX`_n{6zjxXmFe9CI^<;0uUy9|HBPnBX77j)$a!{@RaE$|JOfoTyin
z5866y;j=`Y*{+Kxs3|SC)OzW)Hh!%)LRl`>=ul5{RUm)wlz$mqJ8HdISm!-mN-~D_
zg~+;~i9^d_Q?JNKk+f4>?PBp=apQjyj36jH)o}W${o`)RaI+aAwD^Jl&Hd)>=|rXM
z-B8})^#1);$q1c{=iKhjR-`
zO;ISrt0m!ylh!d33k;uEp^p|f*Z39@95CUg>XqcGU!Bw)^j1X_IbZIxo4rT<$VfQ#
z_EQCDwsTwbvE2vs%;>IW=tenb^f%ll(5Cf(&V?s0HyKL|lo-#;jjeIc`Z^a;%|$f{
zKY-``1Al|raNe)ZXN)(JXkBSWsxpkC2o{hY;JFNWfJX>#D=
zMfX2`uAl$T!d;w>yjKbp=PNW)cp4KA$7|gcNfKOTQ`^*#rR2s0+ni$#7B2qg@6PRJ
z`tb~>XFqqSsSU)~kF6Yib
zilPqAUd}Z(Z+=WdUO!?=Y3q{=VPytJ+HBm+-3<$~q7{5l4r#pzvQe30B0K+-{yV?S
zx^IdM*ERlwHr*?qektb)%bqfS`t($Q&?IL04W=t7XU>Ao?Hl__vg{6r0V=C?88I
zKD-SHRUYb>jCp7+I&7h;jm3FeYoI$Z)Zo(uV)|FRVDF;NlkuX_*LL(ehhxgS3Q`Kf
z8fS^-LPWS2WW3Ma4`ZZ_)B$F?ard3IGxgJ{h!|-kea)b}5i10SnH}*a{xw@wK
zNC}?4ijnlzHT3q=z9bRFO}+9O>}ZcS{?RXUEA`ygkk`;q`|xw|pED1Ntx8zFRDhzQ
z`FzQ{;4k@;2VBeX6+(PFHxYkeP_s)t9fv9w(fuYovm0yr1r_t#Awb4|Tr16}e
zHOWcHrP4gm<`B7
zygzS*SGwF(-a_*FSmq8*4jQ8j@$qKeD$iM54|(R-6ZXO5ga)%6A?uTXujvg@3Zdq~
z%-kN+jSjr{y*pRTuCT&&%eyJya8z+)s=ROUAto#Ehu8`P
zY-(oCv>OhUaX+0`VS{1Yp4H5^D}|M(h}S#*yFm=C2pF6Nm>DnvGRxPv3b8mm_`A793iLs
zZ!%Z7Kk=7Ux$&-^RVcdtnHae9i_O){97A-GS%t6JE-y3gs^?4u#!j49X)h9vlf+=x65YdAq@a5{$T)ORUEuqxM}{;X_?N-yQ$&xP4oCyq{Zw&
z>(;5$vHlJkF`mca@`#8pHLQP}7;=B^G|fXF*5x-Oyxy-AWX+qq{V
zYmCf(bfJN*KK6gz`gx)M5CB^Z4Jt?{Jvd_zR
zT=BRKWrJUa;@_^X){6Vs%%tGCD_8UYF5#)L}4KE(75ufs{&SREVUZ4
zb>-h5?=#H^&lhMy9-VXd{rt`5>TI6!ruUCq*T_u%(4uepLB}yzYj+P`^$R}?RzwS$
zTby)W6;_N^sy0{L58#wXmew9^&hkRJ9=>k*K+ffoQwO?|`|~zhbwItdBR!&jd)hY1
z3N9BobCxt0T6*b7)3}H3J%GCZHFALXUK3$4VLvySm%>RJdFfv=a#R=}H(9qHY~F$7wpi&s98avWEh>wO|ReMg_=}cZn|9
zRtcukCVS|{{BpxwxzVsHcxUN$<;!8lm!v|0EUWs^cjXvvL^@R}vsY|#esW4qP5Iq<
z$#W~|P1U!t4HEkX@b-t`tl(KeJDd0wn}v}wx0x<~gySsG%*PdYnlnqW?c;XsdY0$J
zi+GVPkkXYp908^4uHJmD6cM#z75tBq5!^Cq3+jN!uJG=5(D<0D$j*Na|HL1txD__{u?=y~a{ipwRpBY>O
zJNw=vFPd!IjyrnZ=(LVcCuZG@>#sgRf1Y~aMCA<`Yx64!)oiB_Hq5qU^BLXONE`9b
zk7og%Mg#p{q?#f)XWSOk$b?T%N$x%%RZF{HiDSUacfRXi2X>N>9P2jlg
z6DKXVdHViXACOQOQzfbe3AazMQ*>JZKbqRA$|FK&#P%+t!gngdTcf!BT=i*QUMs-P
zFXCjgTw67wTAM3t#O?fV>n|P667rs=M@-Ms9*>WsTU(3z?YOgE4K4}q+_dnc=sUT=
zew*w21R!z-wj*Vc1pS@o;UWH5%7sc63(j_WBKlp-JMnmhVPXE;9!qpJeag!m$SvC*
z3_CNyQBWY1jm@zzcKsHqxhuSW`Otb=d7<&hhE4=I>HLc9MD73~B6>WgcJtbcPmyS+
z3VqO7J^lUOrz($&?JKQXc({PV$#iomm%5t$f!`TDDd=o(y)*l9lBDVR*b3
z2ia9v4Nib2PRjKajiQeF80RZDF`{8;X|(Ve6u+?b$r@SX9_($8mM~-Nc{~1PeZZ+v
znZ4f6&L;v13w_-ox#-^7`#K^zkw7WVR6OP1mic|(6{?trCpRL
z;&5Tv<7fu2J5gQ^RK@(iiLjE>)sM(HHS;3DYt%T6?YYVb#T3R#KXgG
zZkw0%7KIZ%)-yDxb>|?av4c8DO9WaJ7wZXG^c#pTYR}uhD7Ip%Skcy6u6o6bt6|Gj
zQQesJ@yt67l;>>`ob-wtn4>53fn27eAK3}~?K*(@JWuHOzV2xG4HSR864(E&q{5-P
z9jxrVTya|G<5gc=H!#dU)Nsg&?#>@GEO@`TZ^W}itWrGJFkhTAUG4^sMrq;lE?gu`
z3u${P`QQ_f(AGdHPPIX%UQRktTv>z`@ET7ZPhEO|od3;HvKOtc~bY|2kjoDXI
zh+-&j=jbSEkvhaDz+|Hu9pOv4>#I3_Q#a=0W9FfFG#bmzy?y9BGU7U3XPj^G#&rdn
z|Epb^B9X4X7+~ulkYNlpA*3jXRy-EyK_jd8DpY?(DEsASSJf#XWodhHadfdCekf~ka}6aav!i0+kvXhPu=UfbFTIw
zuQLFp_wK7HPZKLy-P}$)Pe_0>9cH}as~2y1e
zK<(|??pE!Z8U`*`>^&NwjqYs?DRa7vD|s7;l;l@9_Trz(2dHnlK7U5WWc$>OkaJ~Y
z&>o*=vE#Lqk;b+9
zbq)5?g|j9nNceWX=C49m#EaQk)N_EFa}Nwo0w&x*cnuRlt0CuD)jkvMv`RO}>}C3_
zaM6|8Ym$K)gy${wOZ49rS7g%rQ(kqURutBV6lc}?TGeV1hsBanu7NuxOp1C%=hDs%
zIyQCRx4?Ud5)ONeYTr)o%Z7_)TnxVj;m>Ww(trNPre6b;#l2Blu!$>L2NF*pHQ__Y
zNz8$>y!_WpuJ6Kws%3QAbhYotFNbb(av~f^@i}R+dn7HT6LOWc1>Z#!Yu9dPS2W+b
zd8mw;x?7BQB{X3@{H!h1-OC&HV9HeEZ{#B*Cq`)aR=fA%cw
zv;9P)Wl5N8vF%S!{E6BjCPUzQTPfz?4>}=`*dNGWrO`rhEZv#;~5?5h_h0uL)w0`wcy%krbZ$YH>CJ1wL9AT8lJ|C`Fy6DX&_
zmjuIAmC7UAX(HI`cn)ST_@rj&m^BJB2J?W8qLp!CY=}wE8TzI0w~_0oeovq>Db;CPASCe_}h|pS=SD
zD%ysZ;)+@Q{?XmBd||MUV{?JELtGv@@sIvlJj@9u4-vU7LggYcHV4hw2g
zBy5rv$t#-vk+5SHaCn{Cdz{*~?mX5q6*A((;tSN-_8*+O{*t%g*q1-9wh#F<^`HFH
zwgs9s5DsIqUZP{2EniKd7yah!V)>JO*BJRFYqFM7g%y(aW)^Th5O|}&VCV4QwcPZE
zH&O4}Pgy4QJL~DUI%H)Vu0kjYD5TSlNXLtA8WJ#HKcOd4NkZq1NXv-9^R)6xsiW->
z2qxh-qzxH$gG!Ctj|5rZBElu-`;fc?7in1uO%PL5VhfByuF3=bSOJEPYItlT`05wE
zJug_J+K0H%;Xn{ubT($Q+bQlJHv*D8)WJ3!
z7*tgx-Bz^=7@9Kr!XWEam~j%~q_b|*_~TgNLEY>ajiB?Ri_5UU%a{M{F35}{&aaR=
zD3&PSY=I1M~v-ok>u;*!$6k
z-H08gY~0D>1U!K(S>yy@e2q;zThIPy7hhe&Wfk4>(o}MAmSg_4lJkb-nTM{$BME&`gA`IMvA8*>R#$xf}@s(9G_Rj;*&n-|hyz
z1alY+9}me8-w_g^-W|s_BnEI0milX|`*^<*=9Li2JnPU5=j0I4&6LV
zc;#W#H1aho>s?ZySfTYuRR%(_mv%(F>9ly?v^bK30(z=vj;Y#Oec#(AqNosN?Tm}p
z%e_y*Zv&Sdm)n(AfDViKH&;jAhspCr?3kGMK18Z~@NyRuI|o^A3RCUGNX2}Zh=8M@
zvzLk84~N*EoHtR@%!O4Bs-`J(N4CcRPk%M6cfZl9hEYPQgRCf*m4WfbhvMdZN{ehn
z+*>2Ffr5KZCuMza-{r~vs3jn7U_&1d$6v0wX85XufHnj={H5
zRJTiOmVeVjXl>JG{icckJ=wq+I+_38hNNP$04S}k|66A4Ks7JTYKCqSvjgb*@uKS_
z^X_lZ%5?0Hg$N~7jNQ}|OylsPASVSchTfO|v2xPtg3e^odq*vof(<7}VtdQS%}SG=
zY%hzI!Ysh+g3Y1otNnZmyu*uswXAV#fc1DBw^mlSxRXvF_j5c5;}XQwdT|nBpnvIl
zAmFjHi~V~^y)#+jbqrtEuiT{F_%p?LHx|0P_i~>KyOh5X~8&_69YU7D6Tc$0ig64%L3=k!MS9
zVwrUSY5O$o!r~ICJBHCE<4A>H`Xq0@ir?n}DHD2a4@Y>qwJS@W5E)`z2V;BR>$yXz
z<+5Dl`LimC!Z|9D|4dlPFps)e!jBPCCgivaLB76N3$69zt$!DMRvc3+L%-CMAmJYo;tdc|yVC`@_}vsx1)#V8$cYL0ew9Dfs-1lc
zf=)>rL^3U`BMDnGvUKT#XaLia?J7GP#DR5_=6CV*Ct!
zEyCig5Z*hL!#_=TG~3tEbZmD}NkjRl*%&$hxAQRNPkw`SWz5Oc7VuB|y%K~FCuLUi
zR7emllDsB%V$a^)Dlj?Ku9Y+$E}99
z1iZ(cq}Q1sEgf^w`zdIntd;pDaCy@I@zb_d+uI*|vD>nUSR%fnA{nx%iC@%})x3^A
zL9Id=2XV|;5<}{KRsu!0cbPmi>lQ2A4p;z5E*fD>Ph}LVZ{e+b
zg?VqjtG!wZ(ZvJB?2q%9m@6p=xN{Ug1z-g#3S_KY&!57H&s#(3o^=xfPDz#qC;L8r
zos%eT#ojRi&XR#lwB;J3)|_oi0d`tDhlE1x_D>EFv~uv1-{nX~vM%D5ov0xLj_ix$
zw?_Bh49)^nchNHs$__gQQYJVia?2R>?(AT_k8GTq4uXdM?FElhb0#NGJFUAlZf$c*
z{4vXKEl+eO#_heo(|hxyO9vIeQXfW)?k!_nQR01%oie9Rw3fqt>5vc3`(HK*P|^Gj!g
zx?1pXR6~*w4=WhL89Q?Ei&GmlZ~l;QCd&*2b?*5*VhEZ0A&7s{2Cfe#pu-cuSK`*w
z?!2k};~RYHCNg*DDQH9bP+MK=F%;7!$^mTW7M=&~@Y`#K_A?PxXEQ7((xDkWV2$GM#ULE?qJ1rT#F
zf|N`b0c!bPsX7gNAlC*UtJCeT4L)`Ox9N
zRDS)HD&5K}z{q~@F1eMHBOAr8v@DK>d8Nap+A>80Kz
zJ441GIKtZ|8>{~IiW{Fm*r7|8zyb7;)xoEw@+8a%?)oDnOax33X8q>l{L4Eyf;8D&
zrl`N$cRL=Duki?O#Erd-9!8A%j43N`@ivW&Vn;^?H;bY9=$l_dPG)6J3c)uoY|;Y-
zmRT6RlkY$n>#Uze<^gj(OB%SdxE>&iCS_F~g<+)`qO>rHkEZamfe843p53diaQY*D
z=+S5~=P7`Gf%g)We%`G)uyS{{*QSg(-R??|DywR;dzxenXHeXkOD;jn#MG9yR;!22lpG}&Tx1?mJ!r;K1`Ta+qJ1!8
zdz+gGeHRj%Ll=0mNVg8KJ!P9@`So&hTxY*F?!@|}s^i|OvKxmVcVr^WNR(LZZ-4oi
zFn6|sRRytV*O7W${}b^+Mu^7NJ!sH5Kv>j(@G0SsH+S=DozkLNJ2SV;@%_g$SwM*UJQ(}#^icTOFwE4yme2^!#H{5L7idB6K?EGBMKnGJ*Y>tspQ=FU
z$)Ni$l6qhrdQ6&TueG8bJKFjDM)tOHw)pq-c)0&0$FT1f57aZ0`1)HmkxIo>9UHB%PgG9S7
zt0tPNjzz)_2!I?Gj)xh+!?B-P#(t%I?(~rM4(^6B?p{^5ql>$YbBwDG%r^(3uax|&
zjSFeUP~Pco`5ibOe)4)1D9fcOvfQ|CI&V*neDS^s22|@SIZEqLnUTxC&iz0KA1D_r
zBp2!;V=mgT)6jhE@S}SXD}YXpX2dMF1U=$$J=+?zA(T+ss>pXNoj;s7J
z!*;CA$G?bKdsq~9uiLViJTDa$>aktr^HAwCmg0_inyAh`aD}t}gz`+!QFP6jQkYs~
z4BN2GTIZWL2yCQ!0yZHqo1=3ZI-qlSLk)pfd?}`c=KocE27&o_&bjm~K`*`A$)F91gwGY;`N=NHd21oc}*t^!@|z!6ya>6
zuG0{?{)Zi6-_z=TZm36@w8Y|F2geEvkRJY!G^hVJVf7&)VW!H$W2Ea}aK0JNuG&?-
z@K;RZj!qqywLQ8AIXwLZE4jTJTTxqwV^j-xRN$IbW@WiQ^W~cK)R;^XshdP)&}qdt
zq^`Uho4QW^mA6cSkd*PEl`d_@v1LG@rjbzR;a;Q09}>3h*`pMXg%~~`3Xhv$U}3n~mo5y%=3E+Y*JLZ}g`)f(C^I9t_E>1?^(20ni&27p({SY4GF0r#9@~>NP
zuAOlK3t=sqkD+rK+r^}n3%IuQ8rgU)eNOxMXI`S4QnHGst{|_x>eu5Rk3yQF^?nO2
zZiWnE`7-rAS>*irNR^r{x3aPlK+JzuS}`G5$Iw|dKF6dXVda7MFl?Ub_tuhr9K4_G
zqNb@J9kZ6{3y)`22(8`KpspaGt`tdLVfQp-Oc+vJTA^J&eVjvr%VNs@=CMk6Z5aLf$v2jR
zsx;22fvBLOY2c9lP^6-&ed$>8=&151PAIFskfMlFGdT1u!kRVSVTKJ@u)XCdksN^E
zlRkRyz%pB+`bYbEm{=c>DY!gfl!jAh)>zO~R68Wb9K9nj=@P-lYDWt!l&h-ip6}85
zT=7DVg-O4ep>3Y`IPYK+fB(MQ!2?wmA8b5Q$^ai$*R`0IFfQrZj4b`ZeU&Z~(toK3
z$xLbM>1=PhXz`eTuyb;~AipWjZ*Vd1BS|61A|x5NfwWu@;@Jbu$*aAnB%4+i8NQD2
zsSMJSq~}&EM3jZ{0;Gm-5KN2!|K*4}$>dCp5|(45m|~X
zEx#HRB&d^V21aF5fs0;c3v&}rNdbv2(gxQX1+cs!DMhjjh_4&Aq4+0;m0rlteFnDJ
z_om6z&fEKjhzlTm^oU$q*_@35#it9}Zz^i<$#mLNqU3?*d5FRx9
z^Bn)1sufvwP33K|6Ij)Dm#6esB_Npea{boKZM;y^1#Cn9-ltP|U&e8br_goT9(7hI
zMYU%D0r0UstMcTNsUDwfmy(N|Wf
z{Z@zYQB|Q8z;uWy`|>`ly$M5Ir^<*n@7Ex=s^?rqPf26>xcfA@D)FuS-z0^G4@O?5
zA=M~y9{}|P6QmAC*!myvd7qotopaN2Z1X!G@ZdwdkA2e2R1!O{*8!St46?aSm7RV0
zf~KV=hY?r>La^2t4q`oeD42H;H}dF(2{qePD6cK*{TjWEqfv5XeI9k8$HpF20)LKjp9UnZFAkKc;
zysGm3T3BJn>m4Ta;N>nniLbMAT{*>-sPkA`?l|P9%50IHD!w0qA(vyTs>bDZKp=1E
z&?z~1SgE#!Cl)?_NnMjkP&$-TNISgEJo}~NFN9Zmw2rxstEr&2I3(Jo0Zb5=oL%V&
zcq}#fI%OoUHfTMhkU_88j8$y&vWcsFrgD6Ni3rGqe>Ogc;MWg7vh0Ec%H2QgPn(QF
z@?vbAtcm3ogo6b<)^+dPhke^YtmTh&y9tWL^W$5+;et$@W!#taU#nEK$<+Imex^Ub
z3NKHTlDVV5Og^{&wq#v_g1&Cmq3Y7)WxS2<3dMqe788Zhyp;XZ?RCkO%`eGPn`7Jjl&$}xDH;4CJZK`$Z@VBOBtZJZX*X^(^(D5R(wxFC2*XuW9d^uSi_tSD9zkc2i
zcGsm^lf~_d>Y!{rvRkMN=|9-J5=#cKmp`Mw5#fy^<(1^q?h;SpQ4)Z0fr8Cwzr57f
zc&gXcz*+^+ga{7CD2?B*$RDm50r
z79?z$pLW*G%QU_Ga~PmP^xL3q+ZQyb<)X
z{qA7>om~)<%t9xEbsT`G@y!O72-YVF6z^JB+;=H4zRln?q(p_)-rt&+lsE{2_xK7Q
zoFUmc!u;>Wx2&P+l~fvoIoINf$ejG2@sE2r4I^y7HhSQW;H2E~DlOli0T>5Ux5_Tq`W({}5uT!8@AkyVCItna70
zSJ`RlIz#16wDR4!Bc6N?MlT{0kC{Hlo&i^+Q)L-5L5G_|O*lJF-Xd*PweHN#b~FPT
z{!ogB+OnbsAlUey#9A0Y&M?lKT&4K{B+nE;(!|gv?Z8Ce>9K&SY(kugIxrZSVPsr
z9tfLrl4K+Z@PZu`Ok}&ho7wg_vcr-t(+2mIWL5pb}lnq8#waSO^RWph+pOp
z-_%u9_!tPf{t}x$dHt#*@8d}wo`^os?xDBUM?`jeFjjXa5bD8Y$*pkn3Sw~QZNk{d
zb|uZZ9xsmISgPNrzKWMPa1)W8ny_}S=K*+N0=a?g=Q)tdoccqNJJwYrr0?v>5226{
zVFzuA-$alc4y?1$aD_NTr%g-H{1a8r10fB(MIt9mZnjZHlb!&Vk2jH(ZB^O8tMfRK
z278M=P*Nz*vI?5RcHS&TDgGfHEav@@ppj2=foEqzt_-
z0RDoaTcyYCZS=@VEnahD{ucB_-4NpXMX_c6O}m!_Zl@#hpJyTZk~ryFU6G7gORd6F
zKyZ3agn)2?#}chCU;-O=F|->3Uu@Ws)ud&fPB}LTRn5#z?Qgj?+nJpSKO@z9Hi8$?
z7JGXY9S5M@z+@+{C{67}-0jlMVL(+P+(27GLmSC`*y(KwFVc6$tddHb=>yXj6Q%x?
zaDPb{qmTU?zM=5&3ZIKzwUTFk?IlF@sTUm;0gjEYrIUWi80?%P6d$bpND$q_F26+O
zSkR8o7P*f;P{29Q>YFTPLDG5eA_n!kZ+i8d$8{Xrn&F-;6?hQhlYK7=5Qm$5nnOXT
zq=bcOrTJn|6ir}s6;67PlOwQEH^Zk1>6shKHf-A)2t@HjCc=~OFWhCsx#ezn
z!mnt4KvXB?^n!0;_OCY6cSv`d0fJ-HA9JbFysQcHqAy4xUgXh;NQxKH%5xEjJ=c)!
zJ+9lX?NOy-7_G|D2cCx97hh&_+=+^2Uf4#Z*$BH+LQB;Gv>w0_g
zicIg~#fuluRQeWq#8OEa#DtvWS*zWn`(jLi%nBL%?_UT09KfsUe2{(CZ{9?qdo|89V;l__RAHXx~;f
ziHx?|^j!fRj*ZE`v=(-2FN^*UOo-o>$G%IIe~A>rj$ah!M|&=k@qQsx{mD!h|FknMKT>SL+0a&lyLr8n7Wd`4j5MkV|Ib%q
zd@NG%+pH|+#m7h*c}pQjSxBXqsJ6y0%7sZ0Jjo-<Epn`y
ziGuo%A=?)smoJjiDxR&}CwLi^0uS8+ooiP{$`xZh4LxR7;-C~y51BmPz&Cd3zGki%D|UZLPK8nLUd
z<+J{61>?B!wr;7#!YcKr3MC#ui413u?!HLc>Bs*`QJNg?-sW#V2D{gRiW2T2MqIC7
zXI-j+QOPh^+zo2JU)_?NYoQV&b$>Kw->X_>S7O3n&Op7<`@MV)$RXdXk0#$X5;lc@
zGVPvuxk&7LV~n3cA;v?;@r|L=3+|17kI1UDvpx9x((|n7&2_6akf!?3@zl)L7t!;l
zixst!q@Axqltn)6;3P_To|1m%DX@)GQiw+!F<-3L1aS0_R@u|KNaqoDJ#J4Wx=3`F
zo1@Ji+S72-5`0qT!PZpoE-~Tcx7ig9=R^H*3J({k-XXA|>au)jw1Smy{E-Oq#bcRU
zO-R9|RJMkx+d(~neu_xqDCsGQ!0b$3Vacsv%Y}rWLMgOC^(757Xm-6pS!j9<1Cz-#
zL7n~s?Ij44sp^DOvZE}IQk5m&I8rAA-X3G4Ax^+O)Oh`S6n8fKzO3ljPzO(b7EfAn
zcCAGDYtc6d5+0>Bu#@7mAmX|52y@Q;eCM&x4JvL0dj6$AXG^30a&GZI{Z|rST!pnp
z$yq0?5Igsc-j{xlNkz@8S8h2pzDYX^Iookhl8!+_!k5k)*@c6-9V&l%N8a~HSM8RG
z=~ePyUFoy3?TzBud+{#NErB*%%C6T$FQ24jtg1`DkJO|sp``(-*lxn&&mn-`$rv^s
zN(Mwj0S)Lf>V4pWe_1PX39ib75}y)gcsH(iyGDL-E|=NOGMgEa_?;N{(fi7;iHGI7
z%#8!`2j*R8&o08-ygUV13e|42eE%wk#W`H|ULqh=@-r9S0IR}NB^CNSCa^L3!=>tA
z8raQZU)Pp;R;8+%4J_+=@Y`M~QPjxBE95|6)2z;1Mjwv=Gc@*j(aE#Z#S#9qA~WS$
zU6US6Gz0|4oC@dZXba0AA&+T@(W5FRCPMY6lQ9G~t|*f9b2kNf0cdMAxg`7uOERXP
zme9h=BWX8xBpmkWN&Zj!@m+qoSNgB;2qfydDS}Lm{!kO2+0+5qVGItQBO#HLO_749n`VMTEdP=KGhAp
z!K+VP@V;ua;A<)7oEt`Q`5cm!0x>}Ozk03A;zCqy9B1)CG`=zTEEap)#fvt0#xnzQ
z_~da$26s&0=S2W4%+v=3FJ?X%8U-~WJiqlyxVw;Z4sN)HzL794Wofz*A|gI-&(ZC5
z!u}_1GzwFzMAYX8dFil%m*Q))G0*zqJOt_);N*wzu3J5`qS9e@??=1P#(f9;Xb6r#
z#D+uEyGVya%N6f?KRBhHANwXBbi@dv0D_nz;wLfD;luhcEV;mbDB0$mEyZ%^+leIb
zS5O{zNr{dzC>Wd43`L22_y8SWHaR=3>`XMkH|+V?vqn3=v+q*tBd1IlmGbwf$u`W
zfB$hReew6>)kLHeoMFDbSn%D8Qtf*~i3Ai0*BN8mRglVlUDc>NzXu|=ync!{e~qMm
z>?iPeh3sU$ae&X-b5Y@g+g_B(z2K3xRHAPx>OE-MGQ
zJ*xF6NT7q}5Gq|r{QJ$kwFY-4v)}<8vgC@7`m;YnEU0Eso{Ktt-vL*Kl1T9RQ?#!!
z(4VW3-i!MS+@rI(Hsz&A*v5bW&7rM^NZKAOW@MTe%9&;{On8&UPy2LKGP8&
zocc%)4hq$c1cpo_r|629kL(89J^TwHNn!P{S5>@C8+CuHIL@#>Ey0`GtMRT`%
zPHS@wlpiGN_?5(p_GtzI&zJVH&U*Zz(;y&8LDM~ZRlvL9-q*eyebCf2H%o%XX$Hy;wLnP)#^L47rr2
z%hlLW!KKV1D~Xq0AV1;P={ei+?G=~4-O^Wk?YL)nNF
z=Cg5AeV6`W`&|-VMiK+svu61A!i0U{kRv;HdhgDqi?MXV1NsM!*Et?z^=P3hKw%4|
z0377}uKFLh!%N<@QGG?9>4CQ+YPKIaOPk;Xw{OgaPV4bway?^3`2E#_j4Afr&$sZW
zOrz|a2uF{Wic0OrGZ9UzeRp(99Zd6U-65q)SZMfN^7^H$eEpKh2;JtIPB;{%VxND#
zK{14jIvYLQRKN%(9$}%t0J`C$m-#|_>2KM}p-=LIpAeQ_jbFU@)U<;>rtL))sHppa
zH1c`0KQUxIda4UY#S&2#{C5qlfQd1opMF0l6YcI+(gPlYVx(`to;-ff
z0>(ldoHX6v;O^N0D%?HGA!o(-Z{b`7qksQ%iYF#L)Cp-`ykIl>RYizEE)aP7Rf?Hn
zE97(L&MSoPLpJO68RMG7tQ8&qW^1lU^)5o=m`BHy+GOXKE_67$81YKRfn6Z$9G;e?
zi?Nh9~n+Td3WgqE5Hh_ujv)fUm$cB?2h2AvU|Ws
zQ>;GTUWc;BOU8wsAOlkd7B|77zE?+{BZnVVt__S%@hj-f{EWz75@6>&8*AFyqxAkA
zMLY7G6wI{<_)Q=b^fB=px-&H~+(l5G2sLg^~YsjTeZT@!C^C-_$Tte-{381pG`;ft7i6pOC^J_TCU+-@y@SFZ0U^c=a)OLbhM
zLyZ^0R*GwMWqQB}CtC~Jaq$bEW$$4%=eJxK9lUUH`o2D5XIZLICZ%_I^o{*l1$?9U
z|9xWnDl!i!*~UW-wve{-x#a@*7q@%NB%xcgHXhvaG>>L)Sc?yCAk!yAV|P5ye-&1t
z#(Z6pluinQM$Xk
zk#3ak?n9?^cWjkz3F(lQmd*o8N_RI%mvqCo55MR6-uHUXUvQnX_nH;6X6BxI){6KL
zQeLWucnJ=zfqD2vP9ir+d@Kn@C~Qd1D!C6v8BbamPaZ&|UWe>`U(y@7n1RDWNZTkwlRWG-_vcv{2~7
z!klW|17DR(zl}OQJ=>MVL&|Vdz;)?EO)Mc+H?zwwuG%$T(rietGWjbx%;a_rTJ_2X{Q(JvYlV{e(iPjy1s68i*?&GF~66Q;D&Ua9KhsiUW
zwa{JX$B(?D?Gt=$)d@#30ezemN3Gh(0``G|LLBr_GZ+I|$
z+Oj(c|LCIkxaplQE~~dcd|RRDZDYHsp}(>5GEoKXi-e^br!F=B7qx6Fby`z`vjXe2
z)Xt}wjg3cEa?m9p?DqkgfcLc|xB)FQhd64q;_+U|Mil
zZ4>+B6#}?Z7Fd1tVs65qY$}xUSSq8h;PwC8*Qn_?tQXf{I9QX`&w3KWBNtRx1g~+O
za;I;}&{rEB&#)IMjQVKk++rtcb8zf{3cb0^sCY=uvPp|2PrqC#Lu4}
zGn_FXVgNgiJ4EDYT85sb`T;FFCP6qJXVF+7
zl9e2p03w&WAac2YMhvs8f4`x_BXLdkOviOeWi(4?AjeI#X`V^1{p}A7a4wpmqtQVOAd+
zKG?S4jvGhS59@zy-Y8nxmm@VCs%>Gh;k%bxEl#r6-?Q`IBlT0e+Q~JKhWz2I$(pa)
z8g5LqKz0O?MNm~Zaj=Ocq1aRwq4Zs>qoUbMzgLBCTV+2I4rwv
z)9!n-1ad&(;zvQq20EI6wGGf-{Dm2FxPS=c)Jit{{NK`c!e{X2dFhH~A=5jDpVXS!
zeQS-oG~zz4XlDdSftEIp;!3@>pqUrG8N;BJ5WX<*Z!vcwg<;
zG%jLyUr%b-I#Ytb`{!TyCG+Li(Q;
z`D}an?;aKj)7(o7n5#lT(C63yBE339VdE{Jkcy3V7Y?26E2)W9AZ@!raO1$m>GV@P
z(ni+)oe$)wD0R2gdQT`jGgEj`>onYK${tebD*$=X}Wu7J%@3!0j
z5XA(onVtfyFtZL0eg>u0JM0=R#+pD-HRZ$TW;EOy0nT*WY>qZxvJK)lVBV8$#|#V<
z=*3F2==8uS6%>XyVDN8gzvpJufzlE5U)v6O(dwEu?bwnrp^-r`boO^1|0{{SNQ2WG
z1iQzzNn8e2UfsliMw0-sQin#4LXmdWa~bS2ATWywuO&m>CnL9ex~>+@WH~IMm->HN
z&z+6!pn&c7Igte~%t5+!{!=8whK0UX#6wj+=dzkMsFP;UaxRABQtJJsCSSzW(&ev_
z!(%27d?ANT^TurmOY4vuv-;E)gUlCd39Q$Wag0H^T1mH3roVh~ybgc-20&~vzQ1?Z
zJ4lii+q3)HhZ43J#H!e@m&lHO2G{K#DL&p>hYh1;>P%QpwZ8fYrMXx)GSDB6>LDWO
zFmtvqH^A0MHs5w_3ZlUjGZUb<9*h@mFx+~v?f^8Ntw;Wx0A1;T`*Gqvk?r40Y4C={y}t$G4^DJyRrka*
z3WV2xoc27r|Hwe{WSLji8qwQvMSL?_$MB}|FVxO13VQV3%?`HlkHa0wi_MBZc||L}
z`Ysv8<4HX-Z%)Ku`WyUVlhA8@pcc$)J?Kn92pxdOlN(Nq(H4;SMa-^Fq%0v-JO
z(LFum%Ip1H`L#z=7=8!J!x=wu2dX
zj9G}P1cD9*eg61In0NK-A5)sY)(=2FqWA%zeS9a)z_Fm{p%CEFyEkwe%|J+4M%e(U
zbrhUu1yMFKrp|kme<7IPkzI}aLEm8|!ox1J{r$bFHs_9vw}8Ide@$KNUEu%c0~pg=
z+cR3(ZB?h7ntlGSHPr7rk)T)k6K~%%sM)2YNCFw8@Q{W46c6LzR@otzj2mHb$9OetS-Ik?!8Q71WY5pX8TSF@n%v8
z)*!$_?;coodavWlEYkiJIGmIibZ;B^%!$eWibyS~*vv~caE^z{@)Hy;seSA}#Z!r4
z0!b?a!JwPHP;;mFxg(#)*SEB5ZVS_9cfGzPZhN*62;_Ztl3i$jW3t}9>iB=3Zq=c4ZvC=)MDG
zun$<)5tQogW+280i$-C2MDS#OdzHJgEyifv=?y!fhu`I1kP?k`UHIJT(WEHi}>0gvH@L~iV)qpjbL4FiYfz8)q@b0-^#SkbHB&VJd=3(S?+
zV3oncD{XJkUSPNl4`!_?$!omeR!zu!=x)&o=N208iE!XCI6mDetD9=d0Tm%Qo|!5z
z0hhDb#5~b4ERpGvqE>%nRwKxw$D_o(1gPwYLQuSdlYi?IUZN0~7xMCCmgz}B{(^7-
zh*a}3Ev3YXS@uUqe$G$q#?_Moxd_0?(h2`uuYTd1s`U&wlbuB#EU>@cQgTKM&l1;()2-mwaK7B(|OGUeA
znG+?p4jkHB<9PhH@{@$6xU&Rj{NQ0{rT0i`{2BLG9a2a0&hH94MhYj38DBj<97N-?
z>i94fa5CIB67(7R-88oWStS)5yt?LP1v@9-NS7W~C%JFAXFe+4o9{gmm)q?MXx(lc
z0S}Xw8s>hw!jZ5w&|yz*b-T5U@|d%)7T;Q_vVK{0FU8_Zh-
zL|m-vO{_-=VFUPcUDI_Bt$hm$C+egrJZWN}KU-#5tX^DF)>aMM)b&P0m#^NVJ|pM<
zR(0nLX*>l8M_*q)MBbyzu>LZe$S%wj`C68ZM2Jfz^C}a>Sl~X{&`%6j*v^j>sV&y8
z>`WwgP32v7_hNpI+-t3(hE~}6wLY8g*b}KS6MK7X9P6%_&JE&
zz{eII?Rv%73ZLBoO8!`cf*teY02Q%L6u_GgZx6hN`SWJp0mnu<%pkOt_XKU>QM=@n
zXQ3w+jJo+Io+rMEk$d%&BiYo}&~H{GrCIB_TElk$2RCy&iMCFoV{R@I3|vQ#q7
zxRLqK^tEosG7*GrWSC#74TfQcurN#q3l@jLMY6D#|2w>HyZ3y|YV8Lera7%n@|wDn
zx>j@zD|v_h8vGf8?cEKXI$mm~AvC|@c_@FO!tgdcJC63v!7yFd;;UaEW)N@Pg_`BL
zh{YZxA)U6SPK}}aB%%eZa0ZM)>7#;;z=tA(nm2{?=-QNZvy@B~a{emi^4GCG;tBgX
zLO?Xe$OPzfur2$j_J+4qz32&a#3LlJeSq#LQJ}!tzXEw)GzCJS%}PQ4Ezj>i4~mZ`
z;57F4Y-U4BwvG}D&i=Vn2OOrddD^b?|CC0+Ks7j1+sX7>^p~)77hAYHt`@o|yC^jY
zlinbxNo-TROeA06R)ihjBe9JFoNgbTcTaPaNs-cBjv*;b^(jH}pWNa$=~n&S)K
z%g0y5UV&1?|NeaPAFWoa{RuM#ql4=m$Wa(@w&y+Rb(GF8rt~<~PBgB@0~|Y(<`lOhI@%;&FH9@@?qF
zG#k^=oW-b~rddzz&`*xFe`=DtKVUsk0m^a^51}+i60JGFeW<;^mP@$r{N>~D0FVt9
zDp7PqAw{+*o~`_Hc@}4Yr(UROK>cV?)D)@w3*qq@K~yS=f?G&IQ@K^!ClL*r#&11N
zvuZ&oT2mnLvL=RX4Dv}E90I-ZbdLpBvwUj%nH^1E6Hnd{V?`}-wf2SrL5*QfgFgWm
z_!XMD6imV)ACi$CKWsjyg0dw~0d+3sT!wg2{DIH)B8T
zQL+~e0uC|WFn`!*w8Rar{39V|YKq?%{EwZdM_B3Pos9e+pBd^WIjs-A7zTI?2dmJ69VV~&h9#Q
zi^65g>SHWdupIm^kNyoQ@;M{e6|IX_mg5-OX;S|J`3AipSAFo^9w2G
zjOyOIJPOf2ankh2&%mpW-VX`1LLmQlIg6ELg$&8vVNK$XG}cKz0bH`3pl9m6^{d4e
zcrPS@+;Z7gW}5v`V^;F|R`AH^jSHT|9(I33wa&Z}s+gR^H^hdT}_#?FMghkmL0ZqA8B}
z?Mr9;z%hErd>?D#{vWIO%ZdXAY6bm>jg#^5yh;+JjGXiR*f~KH3cU&)j=LA?x-!Xy
zT=B<|$%F|q$$IT%#oZlV2eRc_%Z)c(A^)cw1B3kkD#y47Lk=SKd3tHrP$J+X#BT?r6ebLcoYg8t=mGO4mz`zPapU}F(X
zoBC_}r66Mrl=#m=;7w6R7@I9LqUFO^C#$Xa71)f9#+wjE36#RoeP@H-rum1=9kTV>
zwcv_IzkYd(3eH*aKYbfn7yUO)Uf~=T$37??>GW1Qr91r1mKTa|dzK#-
z=dJ_G4=xtVk<*ey&hsVpZA2rtd8Kw%M7`Fxk<)QzB}L?==z0*|aRIWZ@@LZ&zmgrC
z#WV|d8@ksnL
zky>e88lF~L$mov`Xus}xz=N+QX+%NdSPL&xCmI-3*i|i64!Ba;SkTTdP2l+dJpNO6
z$YUAmu@a!fqC?zC@zy+kUGqHB)>^|Rv~YK3=aOfO^pIZpWDOc
z+PhnkgT>teDaXq@dB|KmuzOwaT+Lh&L4bPHXLxy|f2o^!Ntap0uzA%=5ps?3|8w*$
zwFsX>JM9i=vTgtC-d6KlayXvC5nx+!l5@n$%qb8HX5W(2#j|6mb#C2rQavusS}e-k
zvTX16jp8`z+I0PEse5?%!VB3twUYe{xwfmv8v+iTqa1kCe)RyTc(+v=#E$6I*w4}sOhR{fIcy^Y9c
zW9U>Q`opgiT8RLr_UpI+{1dOD_a8#J_@;<>Z)6^yOW?ZC-Pmtd-xIM=ZLLJ0P;94i
z0bf)?IH2VVxp3X8v=YQcUm=<5utXZ1{M@>9(aiFVs>TEvPvehTLHj&;tkpeoBlMi<
zI33m44xPUz5!M)%eB=A{I*mIIXLk`IcFhg>Er#LSzA{_uR?a5nR#nP-2==lkE~V~c
zw~1^*$EnnKz9~0|%ij{uNJZkfK$8N1dCic%jf@22;w(D#KT@!$iN*6>}jU5th-aa6HJcWvLgwnu*+`hrFjaGs{5IOjk(q<{1
zLGR}hLct*vTI6Tu+m1QN9S}+${`hX60i_P@S9YPV|
z#r|C_D-&eb%BRGkj}>TBUea4EnA|}!n`(Kz27f4;m-BCU80Udcc(Z6@ydnbqTrU78
z?rLJZENPjEfkd)b<$cEB&8a`NF?i4ucZJm-8^R^s(0*&u0bc11C6@{q16+zbCPD?_?{Ai4ozbcHb)a;9lN3R&_m
zIcz@U3ClUe;VWay#c&giB`Fe9&
z3VR9=h(hk5w<5@E`b>k~t!cj4}
z8zm~H?QjyftHZ3`q3w>>{HY1$|I2?3V3o4`=@vS%@X!>yyc*`d2x{1dG6zwf1ta4cvzdx$|sFj;SG$AYGwJ_82>GYkS*VVKVnj7FDoQRdvIy4q0G<)1
zO@Z(?88AZZdIlWR-&e!7Z>~)AZIw`yh?GR4
z{oaTHI+{i6_JMX?y8Vw$6{~4oYlmu0L@EPBDv@P)`Lyzy%Q^@mh5q9t*2EGp|#e0{A`z3_#TIHdz$=5LD67JKC;*1DnUmI
z#iwI4B^hjTDk1T;-gZ{lMR_BXjC+~WD8wIu5oJHxZEOjzEa<iIFr(H4j?>_tW3BX
z43M`woOn21u@e1bSe~`fu}-E3ifM0iFT!4~F}Y;mV)8$%i&+s0nnQSCs=9k4Rz4w*I;mn)$^QcKp2X2s2D<$FSBXNrW%iq9VhWNX{ExZP^s|^CVy)|t$U}+GiYB0=Rf%jIcn&9Sk`}pFzW#_*{Klr|{D<_*DR&uZ$5q)HfpBQA)LWMeXgc-P#{?@Au+Sq9
zK)#2OFFN8EhSYHM|9#Ad*~x1sQ@+mWD~2mhey{y>oqYSs$j(&M*P+}PPnP9hZNWD$
z#}VFrJPnsyR*u@LKKGF;j-rX9K}2(&nF)jCU%Qh$XG_Bi7{P!s<>jpH%V#o2Yz=hA
zYl~TEPWO`$_!ul$E#+wo=sYNlo9-?W-VvhsoAdbF3m;K~WhA1emFa4mFelGl2^z^&
z!}UQ+XS9%l?X@_()hbG4G>|9GA`WsR3^;6~XNsJ~_=fpDq(}-TbsrL&6Af4L0?L{t
zJgITd{ay$<`M@yiD@=SgE|4%-e}X0Y%#@7WOwsI@dM7AS@`lm<%!UDcv5piU@3nNU
zQ&F@na?=oj)zW-XK3#i_?Ic5hRZF*;7=9JQ-@vcbD(w3quyLN4yN+o+8}uU1m!LCJ
z2N}w>r=$`@phRl!86k|oo&kfsjij`sf{RA4kh29TXz1%l9K#QSNxf9yT|eQvSOQGz!~^~EC+LINj$KIn@9^hp=NN)c
z-Z$I-;us(y;qgW+IGHiv6b)4wAN8uNJy3oov*a>-z2pOQ-S4LRIIu^VRHmM$MVdbg
zvz#BW5#_^Yl0=)ds-CUwPQ<8Qtj^}oc(DD>pQuDXp+FN6INTeKw)P&JUA%j*qc}X{
z_9UsLzLNHHD4I0py2TRzk;>UI#SSsm7B-V6-nG&57zu<*8zxH|W-A$wQ+yJ_h$L)u
ze`xmVxG1o}s-J`ykswHk9opQ=C#32No%~tx^W@&%BgDe_NakDV+Z;D94skX<{*xh`
z!QrI8S@SnPzQvi*EcRymu<#1;-*l&jKWgCZy^%@m7QyaCcP*)|vmZA<|8QjIUZZSP
zjj>LORjEUAR1gxOse>HCqS0!1LY~*Yy<9u}m6sOS!WFh5uoZm6wOw%5K*(ET@4&i!
zGp6Tutys&GjjF*LLoUJaA~-)uR{{I%&r8Ug9L&sj-~4eC|Di3<_uTyp
zQ7sLRqlvMsZLaXLkRMjeT9VtqHKUDnF?X2vW|@>*2!-}9roF~d!T^uH?98uu;7J-p
z|86MFLPuzQE~$cMi7Bqzvq8hOufJ-!LN|OZWl33A{Z|OJkAEhkwwL38ZE7$X~`M90dmlI@+e(JMJ5;iO&JFP`#%x@7bxaasG
zSeVr#_W@FBgQxiluMPmhNy}Zriebg(Mj=*Y41MBKevADwP39h!#I_~jK<#cnPuYd`
za*>=L#j=k-xCwPdG=>!AfKp1#g8Y!zulpn6kGxtp7BNqJ5u}hlzGP2~sW|27V`HlaPQ%JFO7Y3psa+NMAziL!L%Ulx&(Elyf1__ImFYJh
z+v*N?x&k*gIJlQDy}d)QqwqVnI**$a)4%gpezdQE{NSn=^vo`$6thuNHgFlZr$~;U
z6zX!B+G{QdMmc>3UNSKRRQ~=z*!6+D=(bf}WAd1Xn*Iy!luel)zQZ2F
znFC{FW^Y#U-B|7czC+r&2E(OY{%M4j1PL=WGFJt6G+hJd%OnKMgQbneZ+|h4glv4X8&Z4@Jx3QZgf~2cxxf$VOU&022ZL~Xu|?t
z8D;mZkb0#9E|m!oO?j6olgkno(+yU4fFCqEyrqSZU#Je;oPLEPT%$z;_DA*Gz@ZT7
zlupZpxLu`>J#-46sXIb5Zb?`+>9@X+d3$<6)4@sbk;26x;%S4b2LV3L1Ki}?!EBuKZt0D
z!@HfBUYli`6JLZfsgw`o{U-2V@6_NM4X;%S2ERIx>m0~fu+c_wCLvi9UJ)tD$pvNP
z)w_iO)uvQ2oQaR~-!5d=@@(|KpT8r|eMS8gOY5_QkeZD*Z{Rz6&-|E?&3W=ZkI?(HnWutgsGpGWW
z5`NLw@)I5KI6k?tpFX`;a#+?mQchSqYkOIE#Q4r;sc*Wa3w?o*xRoq$(5Y=05#a4r
z*DZ*3(;v~K^^q{&x$i+O>@$kH)@CdO23>FFRGa5SrG1ucDkm^m_uOuK@6cp_sm6_-gLhi&`3FzSLFA$I&xDMPcuoV}5cL!V()Z_onf=#m#|7nMJT~2ksyW!7Mr{@!Cx6
z9$PFTYvp9iV(*eCmS!DNasq33M4PF-OVg+YFUjnJigh-jG*AtPmE#0wy?~B@p!$$k
za?Aze1PA3b2P_%DD5RV4J28y}oiz4aB+dEH#kxdao*EAb$y^ey`VOZ*zUM{Bs2dnR
zCwjSbuf3Hb%hx=WcZ3))U(BE&>mw9t|54o^qH!62d(5wX!8{hNs4tqX>Eol)jcOCs
zPyO2tl9eb-emav+|}51z5J8Jg*(!fp|6#NQTMu-o@nX
zay4?d)t(G!?8xvmk*=6d?oWC{`l@Nb#p@$j!X+uz@SXW6pO%&enmBWuGWfw(F=TO-
zEk(DFfWTjQC*no(dYNJ@PS(Uevlf#=Xo1>sHP%yy;+96m(9R1}RG_{Vth`!RrRHcs
z^+3K9k?~*s_CLsyV)>(E?MBpZzq~rccSW9GL><$hyAaPWwLj9xuPQLh11z~1kqT&D
z2MwA8y{>#o+>68bZ5Ft!xy*C_PKs!7SpQ<*
zrmVcJE=tQOn(KszgdQV8+3ooA`<;T^3W)!Z3vz-aktw`r^%9&esns7(Yw!J-0B3qf
zGk-{JJ(p?*G`}HJ!>)2}aXy}SKn8Wwm3gV*GLFCP)U7|nOXEwJlgmPG7=CFonr4jD
z#TIMwz~@y
z>F$Q1avfjZFRkO_WVwnIQpOd`g^OtJ@RKfz=jR7XED)1sE|Kui0KnS_U&3FH`te=H
zu?uARts8lwLJ*4X-(c#Wt+$*D2TsX@v#7fbN1*5U-d
zk_`Uy&g0he!tpRVPdAFA8%D@82Ih3d&_RRg{&*uk*VNFj@M()92YxgH@d{g`Me?Fd
zG-^eos*B6TyteK%&A`D_XP~z2ly^M+@EVbH*^dNA{wUskH|QCv%eE`JlY0e6296vx
zGR|dMjXDdYJ@4I{(X_~|NbUebuz!9>$7Uj&lx{jciY02To+=shE}#Devc`KJcz8}F
zU@xgzBL0%h;yS*)5Q*;BAkAWp-h3s}%=TIzJkA#`SF4rD1LzM!EptZ{efF&9`@8Ak
zCRPYParQQ$x3`Z+K7i~vvwL&r&N&&~y6fL}Gd#-Ak
zR%WAX&p37K=<#F8dhO`eV7(V9uED#FtFK9K>~j2bh8yz~oWni-J9G%E*5r2N*0*yt
ziVjC!&xX(
z=mz2Mg}A>Et67$1tH3)nP;eH71pe+8Y(BS(Xa_dUsN#%aDz0!x1N&U^x&%Tzo%)}DG=%&Qg~
zB%`l3y@ON0V$j{@I+~bnGaOSF!qPAr`O%L_y^O7~PZtY2iU1Bz-wfHjtM)qd1yal!
zFz2fy)~KEkstm$6|HNDdDYuX=X}g$KIc5zkf7a>6F)pd>mV}g}Ag<@QC-j9Aa&0Iy
zbt;o^gAw_!kFl)%ZfmLrc#^NG^r)cH-uJ-{^dCokmA_wql9k(dNn44h$Nx<%A1wsh
zYg_b=zC=pS!m^)+^U
zP|k}4#=0a;5jL(DFqaMwdw?7B3tUI<7k#j36%-blNZq|(QOXB_RB+b-;*MXyiZ<}G
z9os_Y2OiGHW*t;FKE?T#|=Kk
znwZC(<>ed#q4JH~T#5Z=WUy8!Zv`faKLe4u)*nVlGN_1Q
zsH>XXd#GBBNmSyLWZ<{F-=1hKrhOXpl@41thb&2~kujhqz{pC!LI=EJVP}sMBZF_J
z91*XiOU2d8eGwL(r}2QKY>j@L5z3;Kd}0C&2}uks0I4XV0CYBl4N@0;GxkT
zuD)8&BDJfj(M9yji>L5TCZlH2MW4DHm?5*KjsXT5bbq{#HC|C(jpVCp(a3P)CIUxH
z-~Q7IFL2;tC35hV{>WCp8cZy5P=B)BL3t0aM_X#Xo2f@sex82c3@O~etKme>q4-_`(r<
ztlZ9k`N4(iYl-VH2|)ou^-v?g9@9Eg?#6`Mr*C2{i%I3aSX6&9aa$2
zJ;+jKldz-{BQoOkXne-n%r4`sSI*C%R;CcOQJstcbB;lsVoyi-94JL;;i3lpU?(aF
zs)+%5PNN3bG6f1-dh^1kq^Px5N?~sWAdt3ZJttGi=U+fsKGom`1MlzLA|8G!Gi1B9
znlkX-xBEzJp%YL+b1|~gXx{c(eH0E*{sqYt}nEcO#}zYZI6M8;NKR
zY~56^!O<())uiSPSK8mxBj{$;i~DK)qYz3XMT^z*sPVE?QCxjS)TJ>@u6!MlB>>SLfJ
zVt+IU)3`mrBdV@z4^g}%1=nBuo9KRdua=dQh=v_OB;vO~wpoGT&>NKZsZIP6u;$vT
zEF|IrC34B@V0}~URA5}vBip=)O_f=%@MaRWJiWXmW+&78%t=k&fA2BcTe!?3tc1Wr
zZ;jIPJ+%D#&fD_pA-|{L)l(_
zRM)fs`Y_9yb;$A%Udd$NRnZ(|7tevfN+3x3SP6GRkw^-=s;fQnAT+R<=*1q43hc#2
zr1Gu1&Hb}exfAj>;wSNQyt@hIX$RzjY6we~&8?i4W_1LbQ{Aj2|Cb7$Il5BXQ|Y(DzqOP7Ebd|21TSUYKjEQg@$
z937LbMvuS_F7R)?ULoHHcB&0820Uwv7fDHca|hO!exsxF=!vWs*c#i|TU49onHart
z(jorM!;Mqi)(fG?;kmr+nPKbWrIKDPA6whr=B0$iIP7gieRQ-JAh))HVZ
zhLbbP?j0De(?oi~11bnXlBTaKbH&NkDBCyRzlR8&y}stQXLa1q7R<914rZ^9;q
z>cD*2(@!?}R>O2w3gDZ&=cBYfuRpTC<)%EH6`-Y_v?g?Xz4M8Ej;LJZAThJjZVH#%
z2NHhPhz0JBPA?ZYpYbpUEkq~-zohGEs4t)#63u;o!9;2img@z$Yv{!$@PK!5pV9AK
z9pa{fETyLWdTcBl;maP;imZQe`6V?2*-;9%Lp
zQ3_`;f|f!|q#0UC!Op_yT)TlkLUI!wH)Zk=+wKXW}e^XOEz6qCP`!ZYCIYTzr_zgJk$H`q)f
z_27;4|k9zFxmXkYd?9Tx8z6(;c&@V6LOc;mT@(884d)B3;x^ADaoBN5Ll9z
z@H&FJ`CInPX+8L6G^{hslD0zHdgc`xr^+6@G(6`8r&`+l&tG4DHLW~b4cRP1w)E7=
zQ@rS~xm|WRgyugoDcqi}d6Aq~Z_wLeQqDrTWvXwF%PIyvzP;awIxtYK
z!sj7XMCxp6><_X&hk>YcbZi
zN=NvVyJP%%!vBI#B9OXKO|y)>ey)e;ho2J#lG`P(Zkk-5`=jGo#xUyrmaBT3
zl=7VnRt&c_9bA*XJz0p#N8A(=^m9t;Zh`=PD`N)-y~PS8B|J5Cso*aS=e2cIXnAG?
z7+KXc`*s=3qfYMbNHsEdXv$3Y`mu3+^oz@@>JCMN=zj^k={~x6HhuOy@&o5vbMv^r
zC10vC@~c2^?aAvce=JkZyr_^Ho*gmcj5WHiIC6bB}xLL)@=Q^(s2T!#c
zF;N4Y=a)V1=b_~eO0gS=B5lI0>)iTpW<5GLv&yC%8o$r8g-l;){tM*ut%B~s?Ozua
zm5Zds==0B$#HVdBIaTD-T>{Gh(|*H9ea-63VShc;v4gn1E@`!Bnl@|BaF>fUXEdzn
z?|NBGrnN#>qltZkq@i)2!~JbG^0LMF++N!`X`B5Hv@-9YZ`U%5^~{nUPj(yhg3>l+
zDZS(+y{gN9wBnC~nwT7dCR02_eVhbe9GU0hcRt<
z&%D&urhC5OafdXEzNS%x|MVe@@({l>R_&%JVEa(#xfZ6p1{>smv2CpgWq3upU>ijV@b{MiXk3DqNW#oA_RUcwD2Ixi7inp(f+QkNX#tO)_;#n*x+a=q6F^zOuZ((bgUb>(lBd?#Fp
z^Ir+n9CD`8pXp9kzs?e{CQzTr5UZ;+`o#uT7!By+k=Ev_yBB>>8!eSXSS2R|hqohF
z)UTH>VRp#qtz|VP_1#NjwiubvE>UL_Q9;BDgK`b6qmA4-3&z+n-(pqKyygxyLuI$o
zy;6p2s~5e%N<-E5)N?zL%#!(r8mJVHXS?m^Q>kf$tye|#6nr
z#&Dc1suDA?^)_c5msZ??HP()`$lvqk<&zrV
zY1?kbptJ}k*nGqv(E)~TbO;N3X#U|pWIc|<0}#`aMK?xMpNyRPt+Say3sGKTz@x?E
z$JN@{zw~H9ZkO|zo_Ug9u{L+vEw)X{E$Ac%h5v@?e#FX`)j>GZ*|9)8nyA!5YxQlU
zrz$A68Q--$KF)HzX?x%Z
z&$^6ZaXnLn!o8^BAZBJ_sB#b&u^zRQf;(-vjhOef&B)*aLc}c%ISRZ089|FlJYhC?ykRZnR(aCSva;wX*-Kgp{2zf_1hJ@
z;izFPcF(lIjw5dfm%pC#Yokp0E%&^fe_8o?oyq(xCF@-=<$!_OQfdV9hH!liv`o0Z
zzTfgiI0hJxz_pc1ClLK87E*(Fl)V!cOpf{lij0H~mAT!?o<<;_R#j#4;J%75xUe_C(w@@4b~
zH(FxyRU%Q&B?}X$MV{s=2=ie*s=lCy-@x5Pji>e87<13nVdbFiY$nF=&5Qt#ME;$0
z*szEIub)kV%X@0FLE<-T0Jy3n5s^&>V?&7*tE$zXxFkp!tkEsmc2@EcnKe{8U_
zzYXV7A+K~U_>f~b4I6)KkKx6%lvm|i%dMBR3jXELHs4G_?P|e<^fcl|82fN#Pa&$t
zw8tHOx(jf$A4UwMvFz{7k$kJ0gC+*5
zL7(P{qWQf1L*bnBnvm!pW*5>o8-X5XA3kU7F3%i}&dQ;tJZ=fJHu3m~vC~$as}Znj
zHdx%>DL5?M-?S(HoWj>GjefN_&B4xolwLOh=#xPjR(Iwz6^uOw7X)d?Zl#2WMs%(n
zShn?C#;7ir8muxuZ9Q;Otwmcpv{bktxo^}_PO(H}HF>sdaRg3c`=~EY<#?|<<_<2-
z^w?i*VJX(9iYv-I9FSHlp43yLR=plUWB;oCZXr#c_ej?mQeqW`nNt|V8)8KJ*{i8DUvTK?cA!D83R2EttMp|!>@xEeMw$pG>-I1BG7TrEeycsbV#^`%H
zc@sHPN9Yv$b#jk;cycEIQeu2#)jwzFO(fLuQV4XKz61`Z(BY&WL_@LPiUS~$MfS;k
z!;VE|6Y+~g{h*I)sFudR!}`ag?^TcLG9_2n9WBxQCTS&xbC78-C0b4CzQ~E|!^mli
z1HJHBpU(y&CncAq`J1GVg8tp7b%UiVe`XwIk|L*dkFy3Z4bQ4vXCLCqb0=N5F1(S&
zItW>bc=u=bkKvxNbf7}Y-l5~bgUdqbidwX25WWTRHWqk-Vl@CRH_VJw%PSNwQ
zsDd!Hs<$?6qs9P_#s*^0B;L8S2t*OT-RfJVI#0t3lLy7xp8cR#%PHPE5mk2;G4}ok
zrDL})u>%D+?fuJ-d-HAFx8q~>2h(=%qOH-@4RbKd8o1KxICpOp^I=U)?EuKf@PVag@!NuD-$imYLimIhb&
zfbh>Gux?asbo$)B^+!XbQZlR$`(Wv=;GIMv%r+oA(}4Qn-&>xl=Xg@=?V@Wh7Z{=*
zq-&i4CtuE|$jYdz)(dT2&Su&z*B_3Fg%@X1Y$>e*AWVoBObM9lF!NXi@f$`jq
zdV_NHUC0o_o=Z;q6I#1$6&aO9>J^VxK=?$Qw4
z-Q8*2U0;*)&b{v&pyx~V)Sq08i*yvreb`*r-fwCkMNmJ{pu=R5{Nx(u+5s~d|a?I<&w
z@M|DXQhC+^gFDX=@-~Owlv1JX@*#kYSe_Hv&mV%@4(mMdWHjLS!S+5Z-@hH$Y4Uiz
z(t;=39r9~0hNRL<>w|Yh;>a+M!+-GCT3u-1@k1wqn4dtEB;ze+EJ2G=}NDwjjdYjv>?GZPDgf2ZzX{;4qKw;&I+V_N>Hj$5nd@Vo=
zQYh^rrNaDg;|@bfY|S)r#xp4BdK;#ejX~eO{$~+z>T#sm$PP+5V@bvTXs$#X|7)%>
zcm;{jV2&rOhcwJ!v}xtPx$}tUb+B$LY0AQe2Gko5wF}w6atgq6%2y1c+^Zne(^@cx
zk%9<=INy_j9VZ`$WAvC#PB#74unn?T^lwqgwZ?gbR+4q#%;J#S`adF36chM~b6N8e9{F`ABx-sm4f7oR0&{$6L964g7eTg567-Dz%SbJg-1x
z=p0pTcVITGMDgq_4J6~BW3=iQBIb;Z3
z9tV~{#{UE#y8J8AgXQ8L`?4xI=slt(TX&fLs`*<2u$4BdGp&IcRC}v~c$M2l%Hd%t
zbzdfFL#zAyo|J&!7nP5?oC4f>n`^{G{|=W!3uay{to&k2AAdyoFM01=;Qof*l*?<8Ly9C^^Z!MozU3BAoyCEtELt@y5pN=aH#J5d
z(!Vq@rk3?T*8!dw%_ofCGd&me>MCGa&@S>gu}9&@G@O|>^{_7@SV5>R
z_56BnEO2hMl>8p5Nlkgp^%d%vgl)oKVV{K7o^ziN-)+`_Wl%z!iOWa!Z-3;9xawpA
z@erdG@k=+++eH5~9k^rqL_km%ZTa?~9v=iNi85|k@mz^2YnT6|Xg0s}7nb*lck-AY
zEMnBh$)VEZe%1Y~!gf4g@f`Aa2+n>lWFdxi8#9_HDIyyW+5k8cC9dHP0S7qsRXsh|3(`X!=QeSyxhT
zD53g=mnOha!6N+woai(~QkVu#3Pb3*`yr0;;;TT%!k_#o{anjE*M#0_Q*z=ju-}pN
zw%xY=?z!g%w
zd)Y-wE3w~d$T4IfeiC>4Xiik36BJpU7k_~Amo_edh*@)kxo)6=wguiZ=w(Xz%U0iX
zGzR?E+~V~TWnQYe<1!_ddQnGgCjz7=LW&9Xr~dAtc)5X
ziTnGPFP{PJZ>SNGfHEbcn)SOvi|6VfmZ%HVn(DtUCal%&xkF^6aFPmu;a-|U;Na>%
zxjd!stSa|TTP#97+6V+NnO+!Zz&@t0jto3fII10+Nxyv3hi7Dp8vg%9rjbZkjZvtl
zgt`9JRqAUvUzDTZg-G;pbb}uoLD*L?nW5)$-7NfQFqeRo3YNpke-=3KMcKCMTvg0&VFq^gJuzTT
z$BGl?zbC<4T_HELV#bxtWhL+bJt_LXKt%9!r5gr<&pUkI^+zFrLK2>6{@tcpYJgH@+`{nhgRaJ$!OlqE_wZw@3SR|
z+q9bz;6ASCG)azWFmSlX`@awB*(Zg)v6r`2z09bDVOU9-HMBXefCl5FgJaMK8TPN_
zoC08y0aorAMgU+uI6LQ2olfS58sG~TCT46g*z2I2RwP)=RHJ4M4}TphL}~SwP6(M-
zo;o>+QZ@wnBY;mmk>HZ{lq{6KRCz4HZW!<=5dHt9<#>L&2Fm2soZ^4uMt$ZYl46!2
zmi+q7D?D|4M+6N{?=Sx_4VO}JDh)cY%tnPi3gaYfgx?NFXg7Z;6sP>+OD@rVm>5Hj
z;t*^dj*)V6B*2?BMX4g&u+y-EE3PN4gwuI>wv0B7p?BpACmG(S!Q$?(<_mRz0k|)b
ziYe=ZF+d+EbkiNZlWv6p(q%`LTG-sZ&fk3?O&3nN
z+WoJ?^&hw-p@XLmH-HgRQeVHr{dh21v)S4AU50CyHmZCg5t;RHsb
zV0>O-c`Mw%6}s6dX{RJ
zYmI3DJc40{5Z;%KsvJ`({2jHW3We0Hlg6KrK~**zdfM7K{GW`kCGGul)+Rbrm{BV`
z*PU2>en~nKz{HYP9duN53#Les1|L!*(U165-C2M3)_blViR*RISW?oI40iLwwVy8+
zv+GWMr$;9vMtgQ&@<;=YVG&TsZpu~dCItA==ho80AZkUlQsmN&d6QimwQ-VkMZln_
z-}L(989fJ@gzbM3IR(l~yCD|#HwfDAz*L4Er1*mtCaCyq?KEH^vgpO&9W`42X9FzO
z-7iv#*cvc)!Mf3lprNg{Q+S)-Zufk;w|+$gs%LF>+`8I8_dcA!_|OjU|M-~$8)PJv
zt7MYAf8c|Z%b5YTSzk_A8qb#~mH;CfxvkJ_3co~!UIJ?$T??%3uX6Y8*^)nSSNU^vJ>ljAM{14m*K4%m<*qd~6MdTPh9?|CtJqxA&{jC{8?9Eea
z=L*z&>3gAtGavUC7O>r}y+vGmFI-$FNHV^g?R8jZ;fz2qfRO2xp!mObA#J71jihGP
zB;~y(VNlGQ#%CFKthcD08CW`HLZ-Y#s)KQ2pXyS*oKbyngIA++EpTil7(UTsO*Wd3Q+Hhw`ru?O}J~_vt}~i2vMt(eaGu5K#eJV`ySZCiD}b8qtLq
z>JXZq+XQa+q8=lj7EBX&dh1~D>4#+YV&yi~ThOme#DACypnuK&u0kfo-%qnBPmOCg
zW?8d}p~5IVuzuuY&WRW-U(jiA4a-9XaF!-JXnMI(!;CcS6s?3L(V;wPY0-`2{D)?e
z8v9D&YZaE>aWj8S{{G@eu2tco=IE!u
z=>b_hn)cU5JN)Dz@8JxJe6-Z&HofxgRxcICHO7b`S(KM>@O3x#QS(xTQn=Nfh{;M(
zdT>;*3yZR#X_{s5v{Ult?7DbIYqNeKIFj`Xu?>#lI#hg#G@uB4@%6QU2^zk&p)tZY
z8}+MUJR8fYFAQD}7iSb=o}M|wKAysUdB6pO;h_NFRZY;1^=#^2=#835p&Hcal+4L%
zfT;UmEEGE&AJ{1Qj$hDk!WlyTerCm`#U`N@>Y7U0EmHbe^5~l^+}w;L?S2a>J6+i<
zwlb(A@0rY%OT8Ue>zsKHtDgAI3t+)_^mFiSl8R@^w2H9qpSW0Cvp8lIO9NCJ&#n5*
z)Wq~}f5UDQCh7j`3g147*9LEJ+myXEK;DE*X_
zlmuww%nu(~vzkb1%oUn5J>48d*;bP{vdl_Afrn1~c$%6Uvpo}2`p+!y?X~)7JX*Z2
z3C^>}dRhc7P>42a+*>x^1x6&Ka)>NkeY_tja;8~IQ>)u4c6@h2#yo>@J%<<5-D+03
zX7mQ;bbM{x{KeFwGLfptKCv4vueCt`62^OCROhw4CCmwJXfm$tQI-iYsf_DF4z$bC
z2fXM9AFy3Lcs-@k%tfqmT*u}|*1qgt>JMW;E+nn_yE1+;VOvLb%A~V3tLm`vS~~wv
zWFxE1Pv&z-$vgG7p`N#EFgYAi!%|a#CVUX-a~b!Q=4bHD2q^pwSY;CFv@YiwM9SSg
zpO8_>CP^nlj;mjtE9;SoUn>s$b3}uN4Ti$e3G<3zcW|DQ3z1?IHZ|ZgGlg@n+rh21
zC&PWC+O&1efLgD$qoQNm95dJaPrj>~=JT6FIX66%cAZ7DG2QwCY;E(zUVbl3W}E*;
zQA><>t@e!qsyc4Bkkg^YY___mlzRA5?5{pD*U&
z>g7$&*P-E9FT{BE9OwH)dRM89OEM0VCphI9VRWy$(_B}>dsVVkaDtd~^H?9@UDltG
zc`wYujVDdAVQDE3N{S63h_g{(=d3XcRrT9V^pfA*l$I%mZU)Nf%~faj(`X+q4g*_E
zS^T6KvoZz+*;n2bCk+)7LMe*XX$|X8s_j*Ben(F=Y=-};X}|%UqBEiPi%
zaPcz#l>C{Qo_?a1Ui9ESJ?DOwIb)(|_qp47t!KS7f(6AxQ#EtYsZKH%*AFsb#p3OC
ztotd~u(RWWDr%O~sF4w`fPLxn+zAgH-%r_Z3=+0p8*};gN&0v?^AV#NMZ-LrHkj6&Trck{{43HC$yuMOzcC(XTYtfGqcvgkdlX8+W`c?Oc#z?(PPk>CXtQ6zn!K
z7jN^68cBw@k15EtRzgk&-@YT4*KQ;BQQzce_WAK!aQw!&A*b~^@au?(Mhptch+fRa
zX{`O(*_(s``YcjumU1RbCFMyla}rH9yLuuN@v}V(wQ*4IciiPPF$RUH%h>*<;(eJ}JVNy*9hYI(^j_T*ENkgFZ0
zM2tH(bA3vVYww~kQLnltv7<}g4E2Q_aVOB7#y|)W%UA+k9|DHn`GaR-T)Uo#OF#k_
z5i?VjB#L2~8H^jts-0sz^902}>B6E0$6IJ^e`P#b@Fi9=aT*)b=2CnPQVUOg{ix;1
z1dqy414;NYTUK=SS@^PTKa0lu_q+07G8h9uc;A`K7F^F+XT34o;O%)}yUa}$D130w
zslp3Lc3ESdu^$%N7kh8vmunB9&+4B%Im&o_VOP;@<~#YJk@@@Q=V=+B(UR4)a44NB
zSJt{JO7F_$WbA@~!13t8Q8k#Qp(%+riP`p4+xZmy=zXkciyL0)b}2AL*5#yabi0s-r%7-EB}6
zl^{qGCLc&!PH?ajRLRD|*!t23HGjjPs5e}u%B$@MAJ?~cgg3%&j!16FN^E?*3x8f?P@`~Lz~dR9*b{@aE5)?
zO55u8>Pl`W8aQeAMlOAC)?DUIMPdKiAo7&qiUyX`Z9Yc;bLDYkWI
z@6R>ISC)jEp0txvo}#!@&2=+UU3+aIjjOvH6=sWjWnQy;e9Ssm1yGq!1S>=SPxcVO
z=NP7bvUjYq`T~~nP^FOa$G)C>QCC5+#y>zLdpF-SR9SwzyGwXxDd|kg!nT@|zjiq;
z>|qA%>ta&bE-$alQ*$vCOE`g5r)UaVxaeZkXK74~Xn!3vX;o$jO$upzD1Nn1Bo&V6
z479|i(v;$N+h&ye~Fm&gwnm6zLvFNd1
zdC}fEFunQ`12T}
zudk~MnJec}S(5dJUX9B&{r0S4v6c3?#gTwgzHe+k)pABzH8Se%tZd3U<@*kLsIGOS
z5?mPPAa^-zFd57?%znde;wHIY7J}ix%TsxOHdl8*1qHO
zKPU=9{CfM7lgc@1q*=Eg-aUsfamW-avhYX{wdr$9(ZRWinRH&*+!r`JT{*f?S1{oQ
zKG*F`Qbd${eL>EX&Tk(_ITf|}{+{;|vLEe6G}*=2xQ&&(18
zD73+Q-&|sIc>xwJDZM)C>I+O>nXG(IFjhuvCxA!+l+
zmfR(}taqFRgW_yKAbLhN_0hIA^(sg`bfF!e9^`(x2xQzx=*HqWUguU&DAVJ4xaZoXl*%_ej7r5Cg2XpJ-fqj;h~{3cr^V6c+ff0e(NNlL;$f7aJOsBIT-G0
zx-Vc*ZzDtK96)`W`r$d@BX1@XQpL7YtVAmpqCaW|5C69ZJ*J7Vh~xze)rZB|SQ_zI
zNHp%*@QyV!iFLHvd|;tJ-X~N#WTR1*la>cSa+OHBB%o>m)?#ZuhNq$R5>dDN(%6(Y
zcCRIe7;9U0{=8pHr;1uiVDWysNUv4a0e55qjU?;Vd`$G?HC3h1K)I`?m1z4VJlc6*c!Lmq{%tX6;R)ZiU%yg
zT;Cto=OUKL46}$fD&{Vtx{)pBkae(Ko7{%$`!ZqKmYO0vY_!bGCT2
zVQl%dnz1qgZYepq#Vb6%Gp!TjPf=cO){LFrRBVrzJA>ZSHx@soCO0wC7iARCrJycF
z4>}GS^4IRns6;A*ZteC^Yjq!X0oE6bPg&EmXJon#M@G{#pFG!>5@ZjzE1(4}ZaXRM
z>hf#R8W(>wHdMO`xo)Syx+lB=5qO?lG+Wu{msW9c6<>I-8@Yv#iWPx}Gdf*QI*o_N
z*4s{+Iq{$}ciy!L51V$Ehz4evgs(}sR|hs7Pw9W#*x?vaPY+1
z`s@ApvCMjSWeXKmz|roviie7p0_ewF2bvV^?33A#xk$#JQ)uw3&4w%+eJrZy@AGv8
zXdJ*xBaJ?29V?eOpIsUP4vLW-44^wlVUNmxRF-FxF^2Pk>N6#&hY}K8bjvnp<>XXm
zAl>QEE!+Wi2ha}3!$W@tkX#~V-j8<3ZRcLpyvO+GU@YM+Xg3A*2wBJ`?X2uKE)zjS
z^$)Qy*#YR%Y{jGh+M112;7L}=NmY(NMa#Cj)lKK(VN+=T#{)I$4hUKl=0o0Lv(kmB
zKnv|vjpOC~Od5;GFYa2a(ayo8_=2I*K{mt^QJ;Wmotu;4QnQ9(;CoZc#6u+-4>hgH
zUvWYI^s}5vli^6S5`Bt^>&E!yVZw7&ux|Ib&-Ak{(xti35855!|iXeI8jB0e*iP0Q`Hkr#3`&EFKspT-?b#-Sr6$^(;9
z>&^EV$Y09`6A2`^{BO-6QGmt07wh(ydD*^-|EwT<+~j5iR#6A7{!I4VAtZT+=@8-P
zj3PN-$A~M>Xzgd4Pc7nzfiGPJN>_z0-q==uX9wng228oa_-{Mm7=ew`$||LYwKuOU
zqX@0Ivui~ySi=K4+`I%hy%*|eRjfV=(32*1Tk-Yu#G3t>7Ts045sO}0g5uNaoZTNd
zGHBm#LfZx%SRUW}!v%U8A<9ihBaL>ZBnv^UiIFsNxWB%U43T?C#4#}?pE}&V^DAZn
zk+%6U!CSUMW)~mz4UaCeyPHCVg<8k_5)Q37-e)J=jhq;idU(-WP
zEaEuBS0^^0IR^gTl+%$8qiIXQ+W^gRiiNo>?uPY^oEVK)mtM~)i;!Z8Ux4?K^k@n5
zP1DSCi!CgZ-BUD;q>{3j`!X*39V=Dg+%50+>Iot5XV?s+hP2l*k|>QwSkkPMM?DAr
z`O@`$Uk!Ld)*WX$%y)0=wJk4t{H$LI
z^QB)9{Ilu@ap&7JVA90ayfnRYl^;t@kku5x^yXYV*sW;riQiKza71aG9J?a+HC^fOX$LR#~~*Jy{-T1t2@@w4pKpdxVX^ZN~RVdz|#O
z`r49|#+5Zq2JqtECs-RB%($3X7$qj6W-(W7Oic`i{rU=HAh-fFI?=Y)m1dPmF*x^NOIlObOPY+9Uvzb(*
zwjL%;80&!!&gLMXdDMzyzUMwLX5lLT&?l+f64piZ;_}?1aQJU+l%vu4`P}|g)^gT;nViFQlzdL8U{9{
z#;NHWH_PJLU>vTyJN2x{?1dvHHzA>k2Qhs;x>JgdrA%u;#r(bbuqWGX13?@SXqa|Q
z>#okU%hI}zw7=9w$&41G5F`!qk>hUcCO!mv_UU-m>{p*>J=Get`56g96GDvIGLg(f
zIO3EIH*lYW*EmYF3U2k=-oQ0j^CYb0Hddc2^CjGq(`AhYeF6uV_3fXY!UQ>qZ1>pZ
z%}<9B9)ReDrT_+h)h8FaxxTp%o?b$pt5qY7&;UH%DZJwiqt%UT#_8oj^gvge;$4S8
z8;7|pBL?ieDKq=US4s36qd!NN#1~3?*c%jlcpvO#6tiJ02%j!*!$UQPx6OF%rtkJj
z59^2+qiBb}^)obV*fTm0spfO4AI{4uh{Y^9w*ow&gabt6-jV7Wx8@0pmWB}8PCLT;
zsxr=qGliCRw|5i7z$dc8=|MCCmchkw%gV&tx*QHZYxlf6#suq|`5(g#_cq9$-X5uc
zjviJOW*ZawmNpSe8{C())%_xDBhT}%0uuOCgP+8I_uc~_2C0SuI!Qi4u{q+{X<+nb9U?Ev1=78`S82_G=8q|=}I*n
zFCmrRr5D)8?7Owj5bxcjrPs$16)-Z0CvXP9*dSN59w8E@;kAZ?wRT*M-VIRLa3a%_
zBo2?I>#OLt#Yw8f)YPr@-GaS^dUPHVAE9AvHF;cx8Slf?urK<<*~BGs`{hcYR__1@
z3R3CaTAOD8M=|L(c5E8>jSt>V%eJ0yDh-x%%Hx5?0gfvyF<|jj&oS0V*sy20NWAG8
z-b(KU2sSL=7N!Dgq=5CmAy;H7VH~5rkxsM=VRR!K)Nr+xw2H4byxY|Z&F)5>5dz}BHII4+_3Kkc`PSkNxy){7^
z>Kg4JB)ebUtk=dsOrw1UmCZO6Bk?bUr-?9B}
zx*mtE4C4mXXta9TDtU1CHfrS(ee{rLxECow*<+sd(6pq!xvi+7>##79OPKb!8Gi?_
zCV@JiI{AWt3g+UA%+A#
z$fk=~6C6-+4@9Y!@?Bn6gCK$CuNyuv8Gy%)^w}gq2yF?F-|K1$VI4iIHI{bru41I{
zoTf`6f>@CsU0`pspQbx6nc6K|1QXCd6emvIl7k?O5tw{GbM*mlE5%G55OP$r$0;2r
zh+E_Dly{G?%QR%_G$iv~Nn5rrf$$IVdqcd^td<6<>jL+G1!*H6qldUO%<&5(z
z=wd9IumRY8|7%a#qiekM9=JSw#T&x>^y)grYVgp+wQ|RJ{2K}Bi_b#n@Gmh!-h6+7
z-oG5NxuG<{v$vC<2i;L074GmOfG3toKL0E!1iwQBvF;;J*|(`Vdg_gzT{KdZ5bC=+
z8;CJw;W^?tYGSEmX-RTjQg@?hiE-G7g(`v#b6NnGn4VK53+ahaP)@;Ol(6Xj)@OxH
zJ()Vb&Anq-
z4FulU^wZ61erKPO`uX7nmIXhj+h!1Y{AHMv;NO0vx4d*QaT3ZJINGEuw+J)wmG6F-
zQc~sU?ClJdgOqyMK3&NJGu}*_C%y7jF3{u>r!~o3ls2PNx74zgKb>r3!&XC7+w`?1
z=vqT=KV&pH#xJ>v({=xPhrT)_5y_oK5y}O|O9~u5>6Unt4tz!nAd_7poU^OZSlwJl
z-&=G(DxR~IRh3C0)O!p*z`q5X`tN4r)cCe%A}(?7(oA;Z3|=#y9e7-hp`UH_u{PDH
zZ?oHS9*cr+38iTG(=^
zBYvy}EpZsG3OQbjJIZ;v^~xGci5e?rrfreiU7gxvYOr(YRB40xv_qb9Gas+DQBdD7l4Qp1vQ+NLd2XFXqP
z2b(ck@*{Qxj)R-7R#0VPW^ErU^2MuU4qZjIz{0EeSy$|KRCt4SukLrAA^B!N2=dsR
z#q#in@x%K>Sw~)V6UKc_%0gCd(9%e`H`6e~dAsP^q2u0A(wTXGEmXyFE;4sKRkX>W
zcUXO>aoirKQw781LC4T@8AWr0%TkLG{bZLFM9xr`TFyV++)vez(1UTZyIT;U>@qHe
z1TC1Bua)ZFHd=cks67(uQ8ojT0!wT;GL0m6kSO?x0%%iC%a8<+d9*wiZH{)q!_)G(
z>B}2-jm;z3x=pV80S_Llb?-z&)(u}0-{)VK(FMP~`!r)w+H~8J;n9d<74;YPvTSoG
zsjQ|6=;l(Xpk1u9T(rr^sB
zvKb-ljdub=-@1K#|G*{+^09;1DQN$m8`)x8ywYR5SP}9a{T6mpp8$Qjy5H%;098sW
z759Y|EkHSRN~-n%W9WUHn1x@LSe$lHSZ%x^bQIeON@fR2Y1JA_RUJ0PAE<|%J%lNT?gGJKHjTkQ#EK7
zpz7H&ol>VI71gK5En(bB9Bz@kpU--^IyUCo&979Txz-V58m`Tm
z;mU5f9JGajtvpWcT{v$3Nt@r0BfMR;n(XM_dA@rm8YdT#bfVxk)o+>q_5eT0FsNI0
zq4q>UBN(vh+KqP8DC9d0+KvrEN$ah=B?2!a8BMGB;I5D5@%R2o=f#a;BEc9X%
zpWw6*=x`3Cf`=?Ff>cc@-Zt>?W}bCCDor4LOdhsPWG;wz-4gfoNb+M}N+8^AWoKiU
z;OJKtjo!K+@&K>Zg}Y`}*VJsygL&@62z9u+ypGDR_ji;jaCvg>T6k>h1V(s&S*!z3
z_kgVACtt2CXz
zdbbTI5{{U?GVcSeGFbAILrBO5Ct`o#aR{--MHuf1x_gvIH1vmK^ys_L3O6Q`iZECz
z5ty4DtbX8p+<&=y7PSu&0N;jnV>C6KO99O(-SA7(Dz*(Fztaxa+sSd1tz$~|%sFwi
zy<=h$J|>-Xq}lr%F6E?@!gNteG?H}+_dL3I%ZEwJ8!1LDtCB+o3Y>P=B7Wmbe#<>O
zJxRq`JzT`7<<44rKdYmi_4$-s%;OCc{OGP`)*zXx5!vFT6k&W17ZFda#I9w&Upden
zf&Pmxkyc29^a5Ji1Hw!kyj(>cD=JWSR&mC=s*(JzVl<^e{X^zzwNj#T!TmGgFemT{
zEMYkV5iz9s;aPT|W2g!X(>>81QYgB?rT}
z7jFYqG#4{svBXwzB(QNkHNPZ?{nP-c7VWkqYxI?@f_1|x{#-Qx2iO~2VlvW8jNJ>`
zp%xFo59I7>za^UIDxDv_6+*xu21m1Xg@Rqj@y)
zh9t4Af0pfxKw`PcI9x!7c+klzwNS0&YVB5osA8-C2#9N<)QYT}kttohR&NfvuG
zm(~MVPl+cKZ6uW;q+Hbw^$}Zpy^-%P-lNWVxN!l*3`jKQ0dt^7MRV=u+{UBRoybgf
zI5~-D>gntIH5e;YZ${i0Jizj
z_mE-b$%7UJduZJ;knW!LUFr9kecU3h`<}Bx#2f?x!s$%>cG_+yR`wn6H;=6}$Tc%_)hhpT_vKI%V
zSBPTmi0@rRZaX8Rv&Cw;;NW5OS6Z70WN={tkBSQ9i^6~*z#R1L*i83r1I^+$D?wXJ
zsdi1%M6ed7m5(ZaKj=t!Zjl{7`J)*11REB2;f7dT0SfLWyZ~wodnF`M(N0eGhi|*g
zM<>iPtV5@_F|w@G&dM>&g%wP?{^%^x56QRA#YN2D(Omko?!M!~J=-TS&=-D=R#hhn
zVm>7;cS|Q=`xUiJby*B^_`xbTj{~xN1>-@}
zWXyy;3$y6j)ZcT5_2u=4mAqWi26u3>=x~$tSCIgr`9+0&CVquyYDAZy~9`cGdvS*I;=R?A0Zx{X6
zZa~%OAu7aYKA@cAt_JvOBPNoUk$(!A`0$>n(!emmgJ*&|^ymrbeN-B_xf?xelo5m$
z&0zJ%26a`ux-p)jG7%L|=`JJ-I=5;M9Vc#^aUna$kokmEOwHg$(wOeeN`J1Mf`+r=
zmM4LB$0KV@1NV`&c*$4Nw^hW^9MOSi0=VQ6u+PqIjFc=C0l~rf!BQyUZG6|GX?{&_
zXQRdnQo4QyuT-C0imrX|FBSxAOlghl-~Rr?1mHp^JKgoHL=|3o(`i$y?V>Ot{6n_Oxs<)(ttpBa1iXOg8E00n-W9
zjAo0mUsaIkam_}}RcuSEu)>RRAVQhS7ET|kRrGJ?FK2H{6^bqIrp~9&ZReLU;#5o-
z>qchtzGAWBB?la4d^icYeYkgR0YBL|(eb_#u=;n<_ZRSphAe;M(-x%S;&4u6XcY@W
zu{<5XDQH5S4(@FDX+NBi)GPtQ#$Bb^NoTG>jTZ^90o&__-dqUF9Ov0(TdpCRcc~$C
zADRP0+z|URxzI6)skj<}s9%FpQ6`JwQaJNF=_xZO8fzDD?yT6clsBI{AQ7#a&2N*WXkE
z0`B&Cl^W(>N}>~@x1)`lg;lpI#fB>u5lWNXipYg*PlZ*Yl{qaq+RG;5j?Sa&(6?`+
zLPM7Vv6M8jabtMq;vthWelRvW4ETIyifYC4eyfg=3drlQO!RF*DVkFXb5n#wTuVG*
zv#>%o(VVlvD%E>Gx`B(sfUmo?%0zU)x>nqU`a;9gCi-$BqfdSd*)9SC<~z(3O_VFL
zIA39Ok;W^fEtj*HXu9H^&+q588Hd|10`dp#RMhQN=6SWisWGW~AJWhf(QUP@w%cC_
zn|^Up&4`)yamnHA$Ky3teK2+rD#|uXHPZDc=s=6C_w3YobpdScwqY_PnNDl`j<=`s
zSXfWrN~qaAD@D7v6-mX|MHHhJQ`R7vy!=r(cPLxroV7HP;3-%_PjB@$yx+)9&51r{
z$e5{4nmf36rRHJ&GNHZ7Yo!7!XV{w1jrOho7WYB>;&tl}!SM`472MaK)jON6$@+76
zVeZpI`H{`vw{XBSDE)06L}}Gc9bR(HHu%O0Ur7_tGjCc`JN}Ec#YAOZ#>_*$v~gJ}
zt$L?2=W)?7?!e~tlRXuU^zT?RANYjB^1lg>fg;nnN*Y3ol`MIu=$~x6&8*GmX2))j
znhOidABwdSK84qXcYP`)unBd8b<*-{Td4AGCd@|5eFnVgFyVRY#eRh$XM%kVLEy`8
zJ`nFI^FZaX5lZ=aag8!xYb*2W$t7g918PMQedrI%M)Hn$^SMXmy<4|eaiY0N{t6LP
zcGX?wQ-3FU$LQ_Y-qqq^EM^PTSo{7k`Et&@i8y+%i)3OW2cL$s31;+&U;OEc@d9S1
z;ml0^z<{YQPi}#z1lEAQg3(sNUast5N^MXp}3{)b((4vp3dD|j;IFNVu0!PxI6V=XP_a1qzs{2j%KM6t0khA
zdr{5rNiT}{Zi5jm?Xsixo}7sr`c${8A%qlmf`^qjGTpE?m!W?asFNgsE&plTSCvJ@|KKvt>(OEEjdAYp?1;_OnH@6KUd3
z;_nYVJNIr8<$s&IRU%UUG)8|s41y{wj7UH1b^YOCMKVeDQ&ws8EDLiPFT-ZtZ-WEX
zyWeXk^bN&MZQAl*%=%L?{1N(hRe>-aCLJNi$~nuSW_Y&AO##*Fz5bA^j;iusyi%z#
z$e1`5v2~us
zsMgseKJD7WHE@=P(OE)Ycky9lSWDX+DW}NW^KQ2?5RipWipOF{bsi&J@aLm==lL2w
zNjh97Q%NOGVmO{^#1rG$cKGt&e{UF*(Z-zb2A>YgZ8Y&R&@bGRNV*aZlV=~;jO
z5JhPuqVbzY+r^bA?a`TUYKE^wYBH$FTG+Mkrn{KUN>+95<*Njd!l552LcF^h8*6~>
z>YgZ
z3vt#Dm(@S|x$kS#8FU25i`IEoC+OZZekxsDX8bI)%*8Ub%r7QwshJn{bm+w-6?}&B
zbp!#9QxR*Qxkqjyzx`n_7ccDliA2><6IzBn@+oMi)GvOuRHExc%ajILPMwtu64|P~F3wNiylS$pQw#)a}YQ
zlvFxzjv~CT&t+^Q{YYm2-WQERAyxOCj6r7%>W5KMwl}?&z;GWS*
zAo{RR#<&dqC*$wEp&`_;?%me>(8|_B3MEs>MW0=^SVfZ>h}38#zyHhwlpdLn*8mzvWuK_=lq_V{v$2~0*D963V%`qXF8)xn&b
zY9i%>t|d-#co%=RgGyGH$xe6oI6-O>Yy166(s3R=khJ+W93Nnz50(fXm}ke4GpT^)cd5Y!7Yvamj7%nJe2N%-)=3}uD5yB+fr#XZwy7M
zG|C|mBl!LES!FDPQp=|CTpS~4ZcAuesRHOVYB{|{Q3c({1Ec0&)n=+_l5*!kuQI~Z68kCC!@Bygk{Z&gP6k*$
zJ`Nt12JqSjHRqL7o1@wK{?cd36&(}+0ebCG&X$C)p4uPCOL%fWXh6z=qpm>;vzhLe-M-^*sg36;#FiX;CgrF(#*#N;KrGQ4|
z@h|`BirN{;s+HU@mO?Vm$N2UV*nFmO&GAi{Q->l4p~KW$H63L!bM42Hb3$7QjX*>7
z;T!kJ@;8u*5SL0jFXy>$bxd=V!&V)iUOfkT(D(jds9o*>L!PI(FPcEf)FE(?GfCtkdQ1#d@QxLDlwltsk^uP
zQOq-3aatgzXw?{tw-J%{31GMdmC;I@Y79PCXs2rEidHSR?0A>cX
z+!`ZDw6tNu+S=x{-2VM0?>z4YEQ@ZTuxy;wS*8sk2ltbe3ZUyPD&UgpUTE6jz&%Xf
zwd}g82L6upDc9O`geFCkAwlxeXxbHD?UFt2`-=VawJ)bFL9lFCSg}c$g71DJ}
z2iW{ck?j0cR>PIuh!7uC2)|eb*Dxh4Iw&*cWkX1^F?xY`fhd4ZWp(4}zKDJZ6IRK{
z=v%z7esLC-2dt!Idf#)g3Y}j#BX-lrC3W+#kx-lI{7k?E4~^hzOzvh#n#c7u+lOp@
z_gVkmSiax2u_SO;A9>ky-Q5XZ?~G~pGO(lFOUo~8O%9Rz@Y&L)eIJ!O)cdgR0P-RY
zsQ+bgQ$*Jvch&Gss8m$(;T~ILe8O$jZPFNuS{?%p0=gW`q>cMKg^i*6%`4(>8Kln<
zF;AM@q`61;HcrRxD`J8RhPXgHnw$(|lrWjMIQ02(qrm`euxsuXE-WDa|VL;p0Q+&)d3U|kbb!N4{*b=BulOaDh!#?op&)~hA
zpmPR;2Zo*SX6L@&4@-CTpDax<(1sq&#M#g1+nr4Zv#^v>$EAx({me$E9PDmy?{2r(
zQ1A47^Wm;wAx5=+hG7B+webVzRwQ@v=uCMfU=puQJ2vZ<+x@wJ*D-rGLA!W-TuU3?
zKAR_~RV&gLlbjL^p=>n~@F=Br)+XhX=e|>+jkun+W-_R~e6T0)4i+f;=6W&^*c~Bp
zOM-q@DrLIB)CJLIKg;Rj)6_oRxZdnbf9lrzJ>I?RBO>(b#{dU-GYfB1Uq
zsHna#ZWu)nq(np-1f{zhl$7o+5lQLpt8^ocbazNMBQ4!ScXtduFuVg_f6u$#wVw5!
zzh=$Mz31K&`|SPMpS?HuVy6s9&eP7DnA-Gqr5fbx(sPIzTvPkfaoBPrcS$XAvM-Il
zD(s;=fA?$0W#5V{4y|j&fH@q8odli?56kAV_#^ITGdEBAqnELp{l_XM^aFZKbobbre)XXP!K(f>EjYolb-BQ8)iMIzsa2=
z#n_rXX!wDuvYMc=;ROL^g9k}~U$3Z3JXmOZ?VP1lzTh!Wfr)u=Mpp9~MONF>wY;pX=KUel5f$`K
z*e2~0v;YGJ=mt<#C{39nU#(fb4jjbOya*v>iVxWRuAZ&IM>LE{;&R+8g^A6KYuckQ
zZ{L!zR!{*dg?n81?%KxCC%tXpVi=d^MRA;(HE$KEj#ngBSkL|6T0K}I8+7B-s`s7K
z(yi&rtKBFfr-PwOX5yoc@>DTxIjiOIEK|PTAFV9CJpdt%g|?d2ySpQU(-Q&e6KW@H
z8q2P+aGafkt+G)aYo)5&ggMD(D!CVWb7vDAV#O*myDLRo`1Q9-F6GWauBA`2$lQ%i
z1+tA&4TPaRJ@Egkql04I?zI^l6NT6gvw?wcWGBj=J_*fLXHEvhp2#GJaObO{K(&F#
zO#tWs5Om`tpxJP`7gCZmJ2MlSS{l30$!ye$${#t?p_A*+L;IB+R4klM^fwP;R<#mpjVgvidgH+OEwTh4UqVFNUDcBP?RnMcE(fOgmk^kTDZ`}f
zTlV}$uOD~W!oTp$>hNfoHq~-uR4Eij*YPVr$;Q>Tvh#U67{&5^H!0Hiq%JN$&pD))
zDlcspgXxfT*4DmM`G-XdHI=AiRlO~whAGFc%Zes`K%Au<_H
zWZ~eQxJHqV)ZRws>;YqHw=2trKLEQpJDFct6V2-mj;Sa<7n_|K4*9Z(!QG=8HjM{6
z;)dGf8ViG*m-oOzxU`Ir9178E`-$`L_>KSId4)F<{ng-EY5wyXa*O0s
znc=VMdP>_`ZL2fm6cnXDvaV`7f0>&`x}oe+?l1ob4TG9;IDcfrK=pv8$g;dMl8@1aQycS;66=<~G93S6o@w|Dv
zs{a6Izw_5mn4;+?npO?i-HhMWKMS{{F}L0+Vct;RGPKHHD-_5j)H&W~&Kc62pbpVa
zN%v5Sq#@Mq$3BVp^3ll105Sov%IbuDSSKK($dEVN$emFzuAbvDD`yPJ8PZ$Fb;G(%
zG;KLhlfK9oc8AVC&tgzAm{_j}rXDN}f=2fZcZBneE}Sl`U=2o&Y4kgr;-cPp9bx3%
znOJ1H(s14>TFf!PAZP7gRbIHcE%ue-dhw8t@1UkiIF40LwKlo_Z59R&Xm9P$_fg}w
zI2Kg8S%E<|uNCGrGe$iX@7dLZ&6!^Lcq#cM7HC}XrpHL*K+M)+A?esRVK}Dkaf+y;
zGG)VeFcDum3n@WOXbvNzEwpT)nUTULSfTH1jI-kE}LRKZ`zL8I6
zpMGaK`lVk7Rbx4_YghAV#{o1G-=4Qac!f2
z)A?l&b-C6_*@lw&DmU-^(EzSP_;oqF!gX$GA9R0m4ZJ766$}I}y&GPv7x?}$DQT~8
zb-`glU^kzN|GWz@PU8fI%K2-#p)eGc?OEoJmAWfUqTjpKPE)nj*;bc?PkjZ;e=aA-bZp645=9?S=!3np2>c4AMtTePs_lQJH`g%cgsDvqDq)F4zoPR==cnoP|$DD<~HevsZjG
zmCZlTF*ffxv7%nB5$^}Jx+;o#(d&5p?sWd~wzES(i0Z=!J&jqf`gU)oCd{oV9Y`q3
zGC|-Xp^d5a{6UoW$zbUinSy6nw!v`c5<6{$X!FgM!xFG)PImI@BBbs}|GsZ5a;b5z
zhf@h#u)pDYxcTXx>Ec8Mgk-HNS4t?JM0Lt`^XG=U>Uo_Gp1|2cD&|SP)}UmdMQLgD
z%q>&7>4fF1MTF~IpW?Xk^oIXPajMZ=1o(I>RHqcXe21@nD?dWz{t8L+sh;6qm0)l@
zPGf+uB{onlWi)uB0rFUf2eMK1{3kG4>qDZupL
zU%NPqsk2O6F-$*hv?ZgX8&Vh?w2Q8CzvI<0RRs*mNxkI73-qVVvJws>UR_#ex^q6k
z)sr!Ra#hWheMxWCPJb#1_6w`OGJ-Fth-
zub=-0oPFz72cHYcpWxJz9SL1afSi>``i-R`%ZhNg(N@G=Wu
zP627_>iNIV8A}g6E31VKc=LACcXvjK&%4sCfnq%k<)SfPV{l7If0ckR+4_cz+|SgL
zma66#<8+qBmut^;==Geo{bHymA?z25r6sPi%$7sGsDsJyc*V5nj^n(jTg~kFTZcmw-9mYNw54%4{rLKd1K=8Ryy1;fVHNj}qgEJ~
z8PDDGam*&-{rnH|8dQl?;
z<>=XTr#4XIG?!u-BscsSS$D4E26$UJ)%g=sV`f?a%429@Z&s(1v%8bgF=EA!HtQiQj*vGyKl=SaoNrf24(RM1U~Z!odz^XO+0Z<>y4qK=t>i*G
zr!#1qiE1_nbXw&!yZgCh+P)z#;hC$wmQEGOcWZUkM^sC)ZvG0+aI)fq&Bjkz;KN7rJ_(_pXQi~?dkofWUtahQN%+13
z94c6J8nBy&xB1B$>`DiB&(4VPo;$6QmmkBmO%8*?K&)wDlh;Q|BR^&
z$w{5J`|vUxb6j_j*nB~7>Pt&w&ZHKZ{Yd;q^;yQ@zV^tC-vS32HUERni!
zUrOg-)F|$53m5DW^r;`;X_w}yk800PE0;=QgsQ;58z74
zf0WyV$>>-U0{}Iq$w9-Ca8J5LzCo*Ze=(<3+w_n|t$QnIU+7zh5y-gzUb1$vqAn^&
zk=x{aQbXHaDLSe7POU0>;77H>wk~Lm{~HS2FsZR>O=;-&9u@Ebx7tDE=8KEM7Z!pBEyd#;5aW3DZWtuZvJsJT^8wldnTWWL5Uk(>zfJ?GjG?qru3zW6((AUcbeO4^#=PKXAt#}rR2s~dp$SC)<#_88is5k!
zO5u*&Cy$kFVN9+Vd4rtN;znJApi)BkP%Hzn#0V1Cq&DAZKXI~s9W98mhCLP8P0e*M
zQlxI-t7#oB+$S0ME0`EXe}SdjWcHE20=XA(uZuUCMg1W!pMU5o2|$UXz)q
zp7;p}=oKg+@lF;=;o+rpK7n&>h$2cncEAeT9LN#Iuj8`X-msb{5wn%wpUoU3MOp&}
zhT$G}n+my&8Dpt3OW?U^;vrdsHyq<C7_Rdy&7Q_~w#K&2C
zyWP77Qk`eIbOxY#D3c3WM2EBAm#C2zVlYRlV(r>0zC9)~X?v~Jv%-psGeVfMX36kDO-3Vo>+N|CLtUPjQJ
zycOMxWvr2>CLId!-yc-cMt(NP{;gHZ^mAMGBXnR7EZigxX(LhPH$@Xaj{xLa+`^l*
zzFn7j9t_Zob3Tu9g;i??aHfEBO)YEEYc$2qlht7zt>>kJWf(?SbbUJ6@K&Worzb1JHL3A(Mep5vElh1bw%M|D{fCXh(5?dhGDo;
z=&3hLQ$V8&g}kgLxZ>cg25_tS6qksDtZ`y?OVv*bhmW?Ds%dc+s!e?-U$HQQsB2_x
zs2|^V5*Z3Gbe0xKH=f&1MkW+H$6+1?)s=#GN}j9e+dAcvG~ViXbCQPrHdmheqLFSF
zIdFLODyutu*lK3RZtY^be}0IRv7$C;S>D~X+qK_xvJ25E0_vN#3+xu=W_nRoX{omo@E+=TRko}abbs>Kyh;aMC}e)Xt#Rk!`XmPhMHYy
zI`)m9=XKB!2b*q}Zl2pG^kesHu$B&9ppx9^I;FS{2e?{IypU}Vk7EU*Fi<}?7uTmw
zOx$>0@qU-&otzthBy4t*(swXU{pQGHXR}X>p8>mgZ;_DZqnw(86JFimA}KHIuj+l0sIO;59a-9j*r#jG=zv7#CJ(WR+f&{
z)5CWc2)F+oF~=dZgYCk!?t@E5^RUBkh05{`oe{}*95%?`GhLFtZw?|Lx|0gUh9$Fa
z?WYWZz*8*{m5PP?UEq*D#m>}k7g^aENmQHO1e@8*nq9ESYZwcrPnAH4=~}bv5cRbwWzxN~G8tl-|71f7
zx>eS)faBA{`9i?kjcY>>hP|a*CM$usli?X~avA!G4^leANMG&?Uc=)nUH$y6GxQUn
zwC=WAammqh2`9O)*qyJgk))xhYmG#k@vnmEzD=FtmIHG+`Y#I@amw*2URysC_>(|$
zkA#~$Z^I(E3tksew~VNhR~xVDIAH>^W~xvrCU+}UtU;ngzx%*;o^6bFFq#3MD@a}r
zX;&tS%YjaVvKpEzm5Q<(P_KoAJb_GO&W6Zy&gfUfk;vF^Xat_tPBE-m4FeZ|Zg)QE
zi3<1sr9e_JG81kHxmtmV1Wv2H)4SXmMbwCsX0EEiF}Z?5sU)AzNH#al-u7Vi@1tFG
zrTaTCfa_l%ZEC3QaSW|$%2Q9SIyT2Frr^s$yG!Z9)$_Tx`KocYt|dL20S-saw6w-C
zQdzD3oGKMGtXy2O9!w+kROo;yL4+VHyDLRz`d`*nHI5Y6R6r~z6!v@R!w!ex+ofV|
z2YP{6jYF0!UMqt@mr1VpTgZn}s}57t?7Yr_oK=NXJsXRXp{>r{kLH?)!jb_HKET?j
zOjPi9z5WF@^C%t}W942T&~VtnjY(1kk{a3w%@*nIK^8|+#I~7}*J18J4u9O`$ZX5a
z-JjcRIKl
zzohSOyR$vYKB>7=;}{5dIJa>p5Y6G1FddC9p(W4Cl}BmNFakIQPRJGnan7jxL<*eh
zs@f0^Cwi;*e%COsn6CJ_=2_o4SmJrx(9Y2v*IkQM%zeW_KqQuDGF4`~hoR<&v6q@_
zUemH(NI&4L4q24zlXc*$H0^~>6~vU(?aN#~LCrZIb#g^f?agic5}CXGo(w!lRrB=F
z00wOjw$L%%0QgcXH6VQhSJ~jhrF|Kl)N{hg+wlwjy7f&LGc?9w(CyB6vQSM4Z%ggM
zvsy_9e$UAT4n)o#bqmRH9svU2Y`f)%lvlNQS8-!D@3w!b<#rG;pLj2~@p!Vw`{r20
zDST)~7tTE1pEp>#yPlM=9Q5r+8bpDofor&|dspxbz~@+6ihrQNoaUFh5oLiG#(A
zxfxfw;m=H=FP(n1JF1hublHW`Kp3;L#w8QTthW=FPdxC1-*Pd3-m6PLx?9x-Rb1}(
z%iP2R$HKs^VDaj0eTJ06!WsDYoB-LC#l;5`W|Z@^W0#486Hu1NNe
z!fD4h)uwu$E90aF0Et8zpS69Pg?e{aHHyl090V<%+XJC=Pm}9ufakHXT{LmtLvWWBZwi1>65(JcXA{)Rq5WD3Q4iMMZrkL+T*vxG7I
z{{=zTw5YR@j+NIKc7171}
z{(*hUWD&h5aC?&CP?w!5Y
zUY7ibn-c7WyWA-Md@vf*qA&M&bv}YLyH^g1&_?)9E}o^hIoYp>#e!nukK74$kEqSP
zSg!ODT!7mMpAo+UKh{M4>d53#`QM=L2ba(b4p48Zw*APOO`*AWRk+MN|I4>D*r9;$
zS$oXVXzKT5VTu&Aaz6bhxZnlctYJxkr@9vowdKR~U0FmOrQp7x(LJ`c`B=|=FZQu!
z=zEpjHeiW-P)ez?)fMvGP+z1BkNkZ(-Mex=BZ;$cM?#}dfSUME8V1GJ?iK5?iy0jO
zPwpb%L1T6u-&82dB-I|s^oA1a-ArabrPifYkU@g
zV-amapH2;5Gc}Prz$e8cEAjKWi*)@`v*yX4bKOa#;F42#(F;z#$#)L-aITnx8F(C5
z>SCt7BcXp7UO2kIiOwg+cKf__v)!cr`o+_=&fb`Jz-HJ^WB4^O%1MNF)73i-4MMtS2)as0@cI+*Q|${wI;=iuOGP^Y#D7OU0nf|)?<)UC
zY3qiEH`PKQoG`JG`5$W#eoYvS66^%BmOzMpgCPkvceH@_^(esMeQm_&sQt+Lt*x~I
zo;WYSc#J%NU-It7qpJ!ji5^}K^ss+p3qFp(17HPj60!eZ!6@PX?->yAl{UK=gU>>3
ztm9y|t2pq7`(HO*|1bP|51EcV;&;Rq=JUfJ-HMt;z3i4r7eFFK9*_uzk-s5GfdQ8HXJFOaU(Vk!f
zeXsbM>k6sBjHXrBt~G;7hOS)go0{gk5$MREzqo0zt5B5p+Nuu~l3PHg^aFVT-li?#
zyvx!3RQR|gSl{`{8==p7Qdaf>f_Ol0d25&&!W~VYuyvbXlPA)T$G!b41E1?KC?O|ffg*G4F&NGEi>cNjy0gpLf^r>$U^LHLLsKGl1w2%NDm7KxmZtQo6JanH
zXkL-_cA}zSL-%E0*D#Ieerp2|GRSWp
zF&8p_Q0vH8VQGl+wkvp^|8OkAn?&l25nec2QjQ`g+%H!C<1Z)!nVH7tW1K-uu=Kx4
z{#<1NCS*!ZSF?keU6gqP!_BR^-4p(}$mqTt6Lp_{tH2|v9`5xU+g8kF*fr$k>E4XZ
zO$$V)M&6%VjuqX)Mcpcy_`m&u@H+-BW-Qh{!{OmBQBKIdq_|Hz5-J$akhZo99%miD
zCSCy$0Np4{=`E&(CgV{qj@w`dVPm8GQT)OGpilzv%>ESwDNs{kIELJ~wYnMI*bM8(
z$FOEahy8Cg^97worjvmtK9AV&oVH5j!}mrQi`n8s*&c-K1xBf2xI(ER9a@)HW?;U%
zuuHe{MCN-YqRZpP`dpnoj*pes{K3H(F2%eWZi(Uh{=cdF6ZDGa1<=HJEhO?c&xnk6
zVo)QMERDJAGm$0-w@^RE@R>gSzg|X6^ykN9ato0-f5EA$>!S?-sP8AKR}*(4TtW+3eACtQldn)-5^E;$Kj{es{VAX&S|HG7t|#
zLpA(GinxJ7;j8uDz)o#u#s-bBvl+!5c$7Y6rG-%tHfe?Ygp_&LGha@$MF%^(9hDUJ
zw{79MG(JtR3NzaL@j^|btfM5A-+NdDhje_!g6&fBnu5LksW0{HU;)B(bhAhWOgVA*
z1=xIytQX%n2`iM5$@nTC*MY6=@PO%+b(YVr%yTw?L>YW+Rt~(S6($G_~JN(<#
zLj7X=+xotz}cL-=p~Ak2p1P!AI}eX<#`=Bbp=
zpTbP_X=pkr9QNP27uv&*;hZ|Sv6yOMzB}#2Mgkj}YNJlF#_5mGhz*AL^uZ!rSG3zw
zAJlpT1nPJ>H=a7W7G>?`wRRU7YT=3EYD8B+bQm1-yR21w*HMB_FQ42dacQ-4S(7=<
zLkB;@rnC?gtA1))k=6h57p^ch1?7Byg4{d4@akr|bH!09PI3lV*v^wLe53Jk_fe*U
zKZetzz7o#N+j7+m^)76F0CiP}Fpbi`=y!foTwdg-`lP5BRzE+_^-*(qthen%AI?kz
z9Oc3r2sE6NdUW!(x+5W{V+CGaV4V#P`E4=Kj$b#2nx&VzT;43(M$Y^JI{M(x<$OKX^GtyXm;^3ZGBrj-BM0hdl2=&RD9l!!a1M)$ukOa*E%ADd
zj)u{aF@>Vnkm2C4dBJkx_+hU4zmEShlIw~D&89bBZK*@*f~6vtUj4C#eofRo8DFBC
zgNNH#4S8Ybb?5EArHU)Ep-eGJO_|(wosNs16bxh|j{bw;#
zixx~oxX7UiIGsBb$w9jmAYqvF6_!xY;xP4K(7`TYYpRke-FMaG{Ha&&^hv;6kw*+$go@71=ar|#p&
zgwm=va2)%KLxKCcB|QjtR`3gP%-L8n_Z_GHpB=im
z?Gx%4%eq3MVtFgU)1e-Z#hIv7sT$@y)I4D}^8hb8@2e#nj8}?gOup))uexUp~I5Z7Mh&WqWob{2!RC_}oqhV+khiJs;L(?q(KD
zK~iyGC)wYQ7B6r!HuYT(`cq^!hr7mrszAu0>%m1FX&%s$ZH>wgG*7dUU&dbr*V0s*
ztr6urYr=R_Z%5kkR_AvFHT8{LjKVaOM82`e1}%OtD-&bpZ}L9kcgy2Oze`Fwo{r}J
z!Z3VzLlxkf8wcLlXol#^&t(ZZ_F&oN9jnpKT(?D$2H
zI}rwGq~1r}NSi^^1_Gh!&b8NsmiQyr`eE=00C?N>DSvCb6B~#^E}V8NmKS@$vQAR}
zQT?P1nlE3w_blXy)2++r+-C7inIe6UrRB0CMm!!tjdA8ie26m823Lp6I9~IeZ_o_+
zV*E)~+{q!hWO64xb=FWO3*f&zj{v15?4&T+)Wt1-^=XY85ju3Dbl-{Mc5aO+&VHqT
zbFnWqM|dXlyMKI~3IEGQ^-n9&h0Rp@wCjz`!RxJA>O$y^SxaB^_d%B)uSZ%E{E0jI
z_bWoXk>r#MlDJ9k3ruN|IT~5jU$@-1$72o45Pk|AahYT!B8voMYQ@{-NR7NdS&7s(;YrH6w1}!y|i7W{l+AK+L|BP#0FBZIy=gm`ABhHon
zyAk3}5P0FP#zkYPK|W!&DA4&}z{A^zOZ9(vQ{OPj;36l#^Joq8&zq~u=MndjpwAnz
zxj3GJz`^jb5mobTtos4*;uM9P^+y}{WIuRKgMIzrDFWjPQ`Cu}4=v38G=QH^;+-J@N92l}0z&N=^oaJS}T@$L!X2=efY!#Zep)RHhB$4L}}}1hjy45VJSa60VM0(QeVP!Z7$b=wY1q;
zX61{<_1Q_CoEE1Pai8f(Q)03g=8%K$N54m$3;{tf`cY>S_)lj8kbPT|w;frM%ljg2
z7VnoECpemvTG`uuJ?6I6@@+V&N8t`Cqoz^!mjFL~PQ_NCyFq*FKyj$~oYxGK>0a|Q
zm_lpQ>fm0-yT%aL)Yim;(DIQ+x8)W2|FmNT#&npNx0A{LtN!U7a}>~>8r*9iUw5#6
z+ynqEYj3~U+2>(+3N0=<{w(|$G~euVC(cxJip|t7ziNef75O>ioE7Su@t^LB-#(2#bXS!QH5;B4#N+M>juCCi
z4pjZB^6pzj>NdNURaeHMl+0_&xIzs;Rj^O8FdC=C!LwgD!t+RRY?+dk{Ha)rC=Q&5^{~{=L}GzkcxWTCuXjh(
zB`2TLR^}il{)GEnvfpcaRPyfmKDIrSwyU6R(oGe0$(Ex8K`uw>9@J@=3m{&}ucMC>
zkX&Fymp<3QpR=oj%3r;GP+L)6b>cn-NkJ6!1GD0Sh-Y$kjSNRT*#8-SgS
zn#z#ULIbSB6jAlXvx9zz^cOCR`s`n?{akm+QuE?oD$glPbZr>gVFg@Ta@BD1O7EH8
z7c|Z0zgmFyHD5Dq0aB%u+2R8v@Ep{W;b^=hTa)>(oWsmKIEL&5Lh1M2hO~Emd}4;O
z@?|e@o%xOCH^u+ikpd&;VQ}APMgv>K#(tx&_mGyQ5%JF-{7%hv3BVtW;j-yUlWrqa
zpO?GzHlxv#b$9#`A++a|fn~APUTV-TsX|U|qcE=!dNypkGm^s;gzUL{Q-Jp32zCY+*-Qf*>ZUIo4fNEU&)
z%tm6qpOs_^daN12CJbx|KS^J$3IA9!9;$T=t9ZzcL{KRvqpu0OL~+jd-ziJEsWqT!(r9oT6N9fyQ
zzi-3~S%1z89-fmwO0KZYyzXWg6g*j>Y>H*Xjgt%KBh=YGJXm9@e!n5^fvs#P^?t^U
z)oS%j3?j@hno@HqAJeyPnT8h#R(tZT9N4E#Pot^5-KsHL8J?owb3$hp$}U#*qBBhQ
zZH`Xia|@fk?W{cR_7|4@7UeK!k~v#rHA5C6@6f85=oIK0a$V&*4?TR66|3fUL{Le%+IQcGKBAJiG*`N}4G&}(AHpx?-rJg!%VE?73h
zQDA-#x!>}1Qg?PediO&D3rul2>#3Y`Lh=a{JYEgIwbjy9e%~2@TE^ECceg8`T>8=~
zwKCZWrg@oz&w}$|#3FINXdvH2FywDKA4zGn4S(}p>QlMW~gsceyaUJ(9=_?{j6TVXE`6`q)
zPBGsyYm!}7ubXPkNTl%Rf+9yhFWSHhb;i;_hkE>9Rh-DZxP(99&fcKG*9h
zC9Nm*q5nJmB$${&K^EXDcumDEP;qR$=Gq@@;p$KiR`f#7r~kvhZWK~@+`!Y~d@as!
z3F_#6@cm+xA79i)RR3MTH{t(5UQepw7?6<9A8GTG2(%>(sQ=E#iT>Kwpr8mREK#C=
zw&pe&rIQ=n;xOlL%iEK@Z4W^_48uDO(oZwE0g2$M!)h+Oyr7TRIsCx^1Y~eoHJNkn
z-E#XMUaq&>By=PAkFS)m#E|%#qUzStT!pFV>F1z7j}9;dggY1a18ENyW5(Sl{{s^K
zuG6E4Q6iv|{0}Y|98uKvwHI=`e
zPKEnt0<^_3^l{PusSU{eBi=cejatb^Bx{owQf`}*2KrIxpc@R3(IGu;+ehcNu-&m4
zI8?zGDTnoM2ce4$@aJUfBX<3AKLnGxhg#oR
z1{CPM?EpyExJ@FCqlQ%4w*5{2VUf(|iVUi{BAWQ268YoXx`vMUbF7UR-kn0LYQ^go
zUUSd+2Ae`vS(5&D$^%3~HVlN(%o-My?jrb`!|>3eOd)Fj9+mVe1<=(|3dFBo5_xEq
z*209jzNSj`f5RbRIKU5Z$P}BY?x4&O7gf_ZUQnKeo-At#6){;3@!H%JQG)Eb$-Pli
z^PWDUANRwHEgv|&C|!v`mQVep(6Lzr-Ycfn(r7BJ@JM=wX&$J0IDd1IFn5_CDHUv*
zERC3(SyMj=)DJy?v^}!w~p6fWC-7>IWC0!^d^slShjxB=`n&dYziz7s`hk>
z4mIgzU6>}%cW&R-e0bDAj=!2L2L7&lLUPYRAd}>N$&#_RAXq$DF$!J;UOz#!c;fef
z2`sa|N(FxLFpDOzvAq|Y*47ED(I{5%)_3T3`s?JLzJpDB0%3cH)1kUjl<~o^3O{_(
zcxZMW~>NR$-$tUOF5nfDX-~>hsxNZ(&k)aK0JI!bbJPf6OZ45mO66E
zLT)oLKQ`ihw(LMeFYf7QavXTA?9lEv{*^%K4`o1txlbBDU#*be=Vxin#l`hFu4Ea0
zszAABW^XEX@=Kc8Z`Pzdre@eh^GCu^-4r^p>7-(tG`_RJe5!wuU?+s7w3#^tH0MzB16oH_%=>oVta!3EtFPuAr)udlVj
z`Tv9FsHko7krU2VNBRAy80{5r=t=n8+xD%7d&DtH42y(ouVYG`TG+0CQX=vtdwCz`
zqe-|d0yFB3=9hgIR>F^0it
zM=^)PL;fE1_0b%d!4o}z~Z`ZrfhnHA-t^{gn97uu*18W>7
zFUX%}o2**{O5>oA*Qg{!*$J&@M^$>1`s98ThNC_UzwF6xY9Xs=O(}WsZ)UnwmbMe;3l
z_v<_4D1DtdJ#(s~Em&hUytH(6+=k!0&QE5g&wKMBcD}1>n#W5I@F4~>xsN^w3
zWh5n4b3K>^lZ>mSoHJ{=3gMj85pkPKEWw>Qpu}{c)^BF^azjP54SVul;G-*b^7S8d
z4TE%F?a5098Ifn|C}P>-l&!G{4km;C+aAlo+u~X{xmY#EM}Q4a1qj-ql^N`AHJIYh
z`lEeho<)@(LoSailOs)_YRz{nO_QGP(%aST917WV@$X2?H!%-+ff13Kxog{_vv!2F
zNwPgVYI~!*Aurz9JTuxxtu4T`qw2QPjgc#Ya$ckXB#aWc%iZ`2-*xD_P;F19=+a)q
zzPanTh#p_C0=Qn2wp4%JYbWz_@}7ur<2|1+-H?Yw%QDTD?a$V&_SZ%Ti4|5OXOCUaRHfsMY23!xv=IUf5tO5jH6gyjmoDk$4WIPn~s3h)MS_
zwj6&mHKrQ|rgUgu*FD{O9
z@bWf?cLNl78=@l7j+wzL(oPvK-&7?M7u#uzmxDgPGS{OY?AyMci{FTSLpE#ByygDw
z9(jRz#{`Stc{;F+QyuSUIxmsqr1a05QQ*>A{SE7v%31SwV?_P_h;eaK0z|t`n`8_zE6IBfD%^CpC2)
zE*B?U7W(!&8fla}=+vM(m|_VMIlEx3-vupeq&&I}1m0M_f!uMjuw4K>_Z4RbCiyw&
zNv=@(>)o34H|yNmHuLW7C%>2O
zpG}g9is(tk-k>XySCge)NCB(<(Ih`-VUhB4ac~x
zv4L+3c9N|&aP?yyO#z{UG%qx86;d*6n!!g`VsSP);DrK6
zfXln5nU70l{pv`G)07?*pN5xwwZA&?1Zg$KyLlt7B{WNd$Ez`Xp`xN?D?pPGpeMOYE>i9R)}Lo*{pYCYh?`!|gl>8^+ktgU*`$jnq}6V^BzhzoCjO4?
zZ{@Uy1Zga4-NdN+$I!vU$CfroVCBG6L`mwIv}!pMErt3(!jONh)|>$SPGumX%Z%O;
z9Qy#uk0WY^ZwqeJ!hZ2=O4rV@o6v=y20saAdC{)d=AO0UCgpZnyCvD?q)-lq}CNj4K4oZ80Z
zgf;5LV8_5L>|wDNb_)k4fp-5Sl)_7|nb`#;kp5}4{Oqfeux%WBtEm@de4y~(swF!&
zo5BF8Lv69ond#Xy#;|d5srg-6C(}#(+7NOyj;`
zc>!;=P>nU{GUj$9yedzMD9%E!NP7w-uhXIDq78A~$2a+J4PB)y443=(Xw(r8m)lfb
z90heLdfN>*>dDC0T)aMP@SHMXz2vZWIcht)JK8t}o?t;a;|=C3b~X43Q&kJEyV@`e
zRQ1@gnvG%x72OScioB9_1eth0VI7DBsTU&Ycz~pA_4>=oHScdEk5VV6v{JovHfsaM
z`k{56&8@vH;AVX6sWZ%M1b$QC0hGBD+#4}`SD79WhW09J0X%P3@7}NzGX8~}
zoE0D>M=z}zIvgvJ4@!Ggu8KxJD>u1yF0N-R^3;gqg`t_9KBfDF-)NCy9D^o`!$PW!
zF3HqgbeTs1CR*_8_K9ijy_I_6Y$8r_KGnJFYW=EF$Kt%qPd3n`ptO|PxG!9@_ff_mAUtj+1cb*+5fHFY
z|H=Ds8TTju-R{anUzz#%_ww0;Me(QSsTe{Gg7-SggPZWr8Z8ftpA>?yJr=6`qi@1f
zwm72DqTn;>@-P`M)sIII=c<=y0az~zpdUcIrP532dSBnZ^T4J19w{fE)cgKVIu+&M
z=U2Lhk0-!!Mb2N-F{c^INpV@E+qZ&D{IsV8n{Bx2k2b?nfAWpQPk}_YKfZ6?V+|$W
zAVXM!5Smg=gC1VH-V}KSxULL;{{ZSsZ&;`>Or@zN
zyzu)lQZFvO{XHxHHNPy<;~3y`aTAq$o#o#%l&#+AUnxs?IxkMdRT!fF?|1LGI%x$B
zPrH-uUwj>MRk?AfKek30KVaj(BH(jRv9?z~9mx61k>~dxa^*T^cll_8J&y1L|Kr~f
zKkDQN2#*5@r@{Sa3;lQE|KGp-Pnkh@oJaU={+qD>>zDucE*|=ogkSjo*YUsGWz5|_
zM?kp0*RirEeqn6=)xpu;*wE_Xlg$^47p&|Y3FM_uUq9vHPWbHhlm-aLpRr%BJ<&60
zYql$TpUC6dK!?>7F(p+L9a3cut}FTFyRj=Wp%_qfTvwu-$D
z9{B$0Rqv-H)la2N$%NB@Td(@TKiOzAH_+)_0|1$1c$~_siOn9v|~2
z@t!vR=oYOxqUTqAe&h8gMjyq4XUh6(HRekCiORVrvj`ubPwlf*CtIT5OwfJSl**5#
zWxapF;J>%ON?uFsZ~}miI%>6YRIa%rbuNuctwAk+7S`CQ%!`v46m+Sa=;J@1hZfZ#
z9I&Y}4mv5#KFe2?$O4iiKp*zGSTv`$5~?D`df~`thvZu_CyCz
z5Z)IkiF!KDg=?%QVM-9V9DKOHFw=tuF6bq_ca7C2HP&gp^Z;s?!q;Gn2jAx*Hmi3y
zK~5N%g3ex?-zzMUUpN`(rJt3-^p?0e(
z$Jz6wN_*Sl6F!S&}TNC
zSp~GHbz-yUM{0?$*?89O{Gy+3xe8r>f66}WgXrXe#-x-2Oe&5#+G*%6+kX4{w0$X?
zPN8{~aM95GI4V3(KWO>gg~&a_R714Hx``Ro|_cwv_Y>|S{?FJTcPYO@97Nm>?3rFCMVnNi;BF|6|-Q*YTN-81!ZJ3PcojYw0p;&h3Hw%5Zzoc|dJ(
z!pAoyQJYL@Y)FD|Re`6>LKt{4{Aoi`qmj`E)SR=|QxRV+bdIYw?=&~@bFJKSiO9i?
zfP8&)Ko!5Pk|6vOXD?)(NLoU_d2*Y;rWxFz1rc&72wJH5LBTf!
zfVY+lyUr2RN2*X2Eks_P9nQq#u5c|uYd)9zed2Pb*@=iwq=f5yF$%a);K})sg$l|J
z;Z+(cKx2hEuvL!L1W|v1GX_)(!|=I_Y^OVF9LHC2WYkj!ys
zYea9TG&i5fTC7~vc*hOJ$W*pB?vVTdATfau2}Io?!JK!vfwPq=%9Nr+mW1eGYYcpL
z(@F{*sL)PNK9yWel;x?HFeSn86wy2vJHKw|w*oX@*&i=)Gvvky%U#ZjXf`mcp~=9x
zEwm@giCN4FU%ZH!O3rScZQcbjY}(S_!L
zzKgC6&{gTvuV`GxBhmqqIMT=fuEH+3@;L^>cVBVKI85zV|GxiRB&=$z+%0xO2!7m}
z_CKppCfh?wzwtfwpV`#=YVaDD6e^d}T)pU3ldQR*fSWxTV~8@agblB&KY5APzeQ{m
zlW}Qv@G7b)J6`>PK9JN^ASQa$y$D&XzD{A>hP$Ntj3Gp9j92CV!X-Hac5WboLUAPj
z%rlrDbi?%5Vr6OzOEZRyMG$v95h}owI$OLBcuynEy4*V*hAJ}%I0rO2CY8r`di@-H1X
zR%FDIo@x*XktqYx?HB5f@0s}FOgie0qi{^Y7d^%GxFc-{qtwsMlkGB?BB2(H
zveJn8bk?)cq&n&&f_3oPOHQD$UOlX(9{<@CpP?f#F*3j?`9Q!$0F0vljm0b*`48>JLJIbX
zOm%-;oGIe?dUTNk+_iIfgRd4tP=?;d64g=f9N?hzlE6L}?i)`}MrbKh*y4p7
zTry26KuawflyQR7nEL2nSLUWt0KtZckPwbly<%V{<9tBT08Gm^c}LQhRt9c
zvg3GGX9%YLtyXSOv#wl+ioN=lvHCX;RY~ER=&O2`I~d|NQ7c1Ao(ts9@T_TSbDx
z&NX@e3d+NzBea1?fUt6=Hza}30OTF;IN^Hd>h~DuBSg){&C<|BE?MAg9>OCVzaQ%%}a
z49v|^45`|&pkXf^)d!X)0IXJu|AAGR$N;;t6#648&z@s4T{#$rP^`Li4BB)9Jc6oz
z)V_$u?{Q^BmlhS;N`?;W0xi%S-aE-6B@Vf6SRn$dnPpU87J}`n1|aNwIHM))i%Z#J
zrewmoC>Eix*$)&2#>K-r8njL=uGC(EE;+q~x-x`v5Ju9tfZ~x}#R&ovxxy0AhOmKC
zsa+OIl&5VBiY`h#j&wVWFCF(`8K?IdcIQt+i
z39#OXTg~j9(InU9^dsBg@QbiYUh1k^DnmSZupK*d0;<^yujssPvWF^3Pp60v7Wb<{0W2E_qLP6Q+}Y
zX6|-RANRlS0kjvpi`HKJTP`Ryndf$A^ed72>K?Ef(tQ{0_3
zv0Ccs?-tZ9HFb?LrK?Y|i%Q|w3RZ;vpO5+bU!%*L0Oe=GRU)Q@?MulPDl^UKq!V?x
zi5ZcNFabSQipD-J_(Eu#POd9vd8J1NGy5Tp8FpfEz+O#@_Hk0qeX_<7eQqo*>1r{*
z3bs?#qP*E~UwQTOL9M!J9Lcy32?}!A`xGs0H{tXM-3)rVs=B;jMXYf
zO;ocjz)mZrWuwhdbYjhA^8v}&k$6I}n#x2tPMp1t3Uby&P>(+8DvJJ|ud#HVuFdXj
zD7=8W1|y<2(!UcMZ*SDl?NRpu&tc_4Wvp%vG0l!{4e-ex~TxQn@-J;`cIaKtlohM#%
zxbcxwAL#(zy!1DZC&*r7EGFKEf8tC8iDz
z3GU~-QFi|>gI+bFx^1=+MBgY=8Bv6ia9yL>n{fVfgT@uI2i@!v;?QZ!$jPVaO*V|>
zYK=*G68!Rxm;GNjEn7O#f0P!KlY`~|aoSvM-S})N4F6k${q_+Ry~4VFF)u0eBxMbq
zTKBfn$dqYNy9V452*y2EZ~yN+)l)QxEce^lXzHl0oJ)^Bnz8C{nm^y(sB?L@4i0|+
z%g(8h)70%<+qI@Bu;rSw6e3pLUH?U{7u7e)tpc`w{4!oXiseRK#A*rz|4gLHnE;HM
z7rscRfBkZ+D)by$CW`zvYt&Kr=XC)FtILz;3ny+1c$K|28%>v9hPr7~gO~9%__x!W
z)})~^7MnW{A-2b2X=iAZT8T8R8&XCrqLcVm%Pufh*-=^+OR2XHb#zh8bBW~+v9T~Cq?7X5Xm|AsE`_;;eaGmQ{$i{f{Y?m%!^NB
zukwdoJ$9*sQIJDG_zwprSsO$k1qmN0MDP~QyfURbPh{=#LfvT{3`h!k#JoaCrXe}*
zfH{z4WzmSg|0{0MG5suv8sAlR8_4JDtK@~%m$wy@B97mOxT=X{KA_3cXS^HXM9sj&
zt}OtIpxCjstDsPKd^Yd&^r>Ph+$sxlKS@q&BufEl|6m5uEn1E5l=s6srp=|(0##RKxj%AF5vQmq|~n%Mt94J%pBW0PL25*S)t
zDE+rOLH;H5$$i3_&GK&p+I4AYFw<-B&p_~&rBcrZ7jijvYy#t6)B=gTg|*64U}D^R
zdm&yW5GPUzzMB$WaVE7Ng)(+eDM}*M1Z;Rzb2!-*)dq6|5&)f`^F;+MF`MmbJ>W9#
zY0**Wn9ZEqXHPU-kq;lyQY*XGCdZI?%GA$P5a};`*ykpwg`D>I-QXADz&w>bB
zB{z=8-#)IqXM`lq0nrfLnmvSKQs*u~3>yiPNw`4gP7e(uUqIvA*sGU#hgp8EcCL6>
z;xEzu9x%*2JRlGXvH^|l%&%`3wPqli6N=A4*^%(>Clzc&V;*6%AQxeS{^}t0l;&^@
zz>|lRUr$DC#s0NKh9K9
zx!Aisk)gFLv=OfXM@iy~1vt}XZU~|#wO^_VhL9B065Rf!KNIzZe|Kfjaiv9`?rLKG
zv0X0*lQ^|(P!;#A4qWl%$Q(5(YTJ
zg1Cuhm62<0A0!#{S>kM0nmZ5
zb$5xI(BGj=`b}~mdw7FJkhy+ZUWVhs4ICfUsffTF=6z4O=}|V!5DLp3D|0}r=qVks
zBbXbFQBg}i5e3!?A${HB7v>jf^M-;00SSgu2#0Mzr??f#w@8>>kBuew2@QT+Dvu~j
zZ((uVVDB^ekw#P_gK%;YC-Fg;9>cf@meLe5EuRV1O0#EIHqdN+z!m3@@NK|Q#@Qh7E4}?Yw|qP3#L4$SuSKp4t$ztrSf|j
zb}XE^#dUT`S`~{EzR9t!(od)Y06hMrJ5SNv=b$0>y9}*Ab$~xGn&i1kPFJ^P5q
zYsoMbzedeuW#tvW5$;@j(01>v*o7^zDPe*6&}Ax(rd`I~2i6ecHWw9QM?)V;mkU6V
zc9$jU!AAV;@STm1H$W|!LK?+pdoq3pqXhKAl?ZeNOm2wj|Ni;XF_?PHP7Zh
z{x}T@E?RnwW~&3%9iQjGA|&UX=_c#{F$;kvnI}U_`PYXM_glh**zg@c!upWi9X$D_
zVrmUY~%riX(9xIChLHAWog8+=200S*aR{4tf5T!Hb3
zym`;zFZx`yOVVo1)gHcQ#F_`8@TgeWy)W(f{>WUE^aGJ9nF$QA%-U6}NVSiB!
z*ODX)`_}gbBYSkU0FZ3sArNW6Se~69!^Z*hWOI?jF)F_&L)S9^dk8)S
zwuE*x8V?$@`KasrlPSXcs(^ScYPN!#U_x&&z6THg@Rh#~KjHui1r}?qi6F5!j~Yxz
zdh|+n+5`jOR;>7}eVwvtgMM~93iBxOATW)|y7=&BFGStA+TKIV^)9SECxH#b7OJu1lBLvK&y
z&(rDgNoVDWvt9dPbhzD8+cCy(o(M=m@%k>EOt4MU3dHYDn=MTxr3sMd4{SB{!o&Z9
ze>s?mn27#I-QnYdVU#mVXQ%~%0Oe+7`hRG6TSqs3lLNy)-(X)7J3@O$OosX)!8Ms=
z1Y=X&yb-pOsx|@SQg#I++w^qS&-eH;)-z$k+ks
z>+E^E$wh-D*2*h$cJk^)l3})K3YDf$rqS1{!z4`R6xU3jN^P=wD6D_MXV13r?)w9v
zNvw(Hgzt$eIsZ6zNEh^bRag_BbtZ*pn9|N0Nb@D$LYTKw8)bn9*c$}{R==;E$yIZo
z{u|9uQ_&k>Rm>v)+M+j0Q)4sK|DH_US@GMxN+UV1c=cy1viwHrhVZ$qY_E-a27P{o
zcXLx$j2#lNMW78+=irowHMKi`?y>?@-s3=%*eCYXS&ZL^FfmBNSvLaBYNx>(-=VCl
zbjsEBP!(#|R}Z%Hc#{~txrS8gaSt{GuT?wPr!P|U68ZsPX$L-Q0kxdCe<2ghQUC52
zR<9nes4u3DjQ*}eHcghfk+5FhL)j$?)$YnX8j
zumH=|f-QAY>{tuR#HQolm#YD0`(?^y|B9rRMRJL&z{mvMFL`xGuqwE-Qcq3CBw~^U
z`aDgd;+`}VW*~dV@y>vwlFZ@@HfCZ#$T^K}zn*d=H)u1ICo4=S1?!Zm7<5D*Ok*+U
zk||ynshD@>+%}<%QZU{dof!=c(JDtS7_)%e;cCrcuhi!}x|r$=sL%q620a(~TG*!}
zuq$nG$btmTVJJR`Z+)CSG1rNRI@UmM_0{1EjWKmX1({O_loM&oYCwJ$Rv|+TNN&iv
zWJ9%U$d)7)8i?q)aUjBImLh7VE#WNxAUvQY_G&oFMXAwgq#ZkE$>(_WB$#H74~o4=hA=y(qv*)ET%PjZLp
zrh3Y^t?=h0uN;9XIfBFt0<=4pHolcx=?O}2w$4bt0(NR=g^RPm-+)yk$~BRBf#X{Z
z3y4%ouQ^%Ke0G$Mn&HR=9jG3-!-~e07SHVuTl@3-u_S_bBufDKEJnkLzbTTKlQjq?
z1SC(EX@SxbXPID3Yp
z?z#egSBW{4f{y6ne~1lGBD}Q8`i-?5h(oI>^jPQZd6WRQK-UJ2I#!Zipg}j7PS_Ct
z#1^Fe+b^EAZimRbrVE}cWQIU*9299#T~qh25so-Z5N(!Zq(yVnf%9@!1b0Y8;J0%y
z-_YS9k_UXnU+%bYh!#5Zw~3f(%nr|@yxFQAWF{V~MV9&^
zdbSlF;bj1)wIT*zg0)4$ggT$zDAzKr+~b&2>%85CM)>y)8JoPX>vod2M
z<9A>)m$jP8^F?!8wZE8~-(@gX+EPl)u*GBks>}W(ZtAP}AYoY3%Ty@G;tTES>@M6h
zWW%%2VFUdJaEIbURX{J+OTP~D__C3*#@1l7Q=8D6K`onQW^9H0bRs^6T4arJ<{xqEM0(`Zkfl6TW
z;#N~f%S0J}%eGI@0byAx!Gqy}V0N;6pj%2YGF4fH3iDg(T6WN4La1<)(^q9&X6$IWoKttAP9_
zv>nM{q11;O8o;jV%wgmaD0yO5ecQZk5?%=f?$Rqg3dwg)h}5qlD#>?Dh?F2J@gkrc
zwdku8Ix&wztQ_p?p(VVmQ&dh%Fri!qtJsC+Bxmd`MiWH&tO^L0fHhZ;?K=ltg?f~N
zDvu%TE%;y)W9G-9H34;}KL7apg?oqbS8wMxJVkW6{pH8?;9HSWZhsW!K?f+DJFwBi
zllurIkq=(;4-V?TG!#5X8P9Q+w%SAY;
zc2EUo^}w^YwkZSx?`|?xrQ-o$jFY@kl>n#G4M)*-Sy89Yh{KnYytf@>m>Xqtao8&3
zv-S4&Sm)-8e|g&D4byda)d>L~r51cSB$d<#p(Im?d!1-diFyvQ)>J_!nT(M|?@8Oi
z09_gMZXhS;>~$T|FE?o`0(_GlP8_{HqLxSZ%;(}Ho3pX|N+H*>oNGUCQTe+8%Y4aMvE{zHssY%6$8
zPZJ1mB3HD4YoN`+&Xfcg#|tJqE$yfn)%fgmQly&U-F7P3{yLhGA81NkfigbP1(Z&d
zAewzw-e>}9quf1CGK4vE?9QXiM#)+klZBgUcr>M6nnLo7|L!`707-4cf&S?DP;pqdUPSv6%6ti~Z|@9zm!X6c~f@w~Yq4!Joi&-DP!BFcWUQaC9#+Qa@_04+tAb~>yosFRMb+gZi(
z+LAR;{#skfM@+r7K6}Uc1;SP(&QnM^3uA1^<&i(=X<9#+uJ+b+Yh=vCQ-zT0_qMps
zQim1-=2ykN*_Qy^L+=jv@N6hE8Z+W;YH-qfN{W8t_3Ehj4zveB1Y>egVO(u!AZg|V
z)|}6EcV!^dZFG_7y5j0f%8Qo(9f?D^FKIFFYg+uUU>5}6fAvZJh9%cp{Y!!A;lOvrFzrRb&AK7vGZ}S*e|zP}
zUr0k!be=v&6>L(n*%KUKGtIm8-Dn@C0=v8IXA3AeVRUgAB3Qbvowh=th)S^RJMe6E
zoa{lS3KT>heKgRDsy&@3yGoK2eIP$gBp3UD0*~ZFY@dSSK?#h8yIHLFsO$#^;yuvi
z8^Q2nKfT~R90;y1bd&urxSf;xe`c{zK{=TIul`64(9w0+YD4jVtykJ{ccRhgiG~J2
zqedYW2hHK|he~)7UY-1tYpEGm5_|pg!78evBYkqSA+3YeAQp85clt2n*dgtTPrmT3
zeUM+zFzfF>WGP0?tkQILKvR^1TDvGEEsvGHFhcXu!R64Tu5Fv1!_Z|m-Ky79wJ(k7
zwwbyKxXjlx(9r)Uzh28Z^_Ri_^BsOAkywH`bc$lQM2IEwEDuCcnr813f~seCdyCE_
zZ7>R4;P}8Dn*Ngk;}QY1UP
zMj^(;o%=|gIKj+QNfa}4rk7h_f+E81BNToBxOqD$&7Khtz*A(Cj)_ZA3Rzl~|E{OB
zAQ6u_smx74l2%7!B|-CudOkR?Ws^e!#Z}k}P?TPDP>&s@S}V92n+K}Jl8XuwAIZX6
zE5X-q$lniA3&h5!8Spz16TJ1?({ft4QC`+5!ZgH9>~ahD81Ueb>R
zq#qYR;m1PsYK^L$t4j2z5*0j@JJqqps9iLY_48+O{dE-uGU3v^q-cEx2z!?njA1K}y9YkI51_m96WUys30{k7^|Y8WijmMxpRy94W}9D1e7icv8t>Cm&`|W)a;`jsJ_y#Afn9b$9Nx7-YTF1`uc^2I74z
z+=>vR?@2ED4=Vr~t;FsUdaS%HVf5h_x49js7JbHn|@4zU4hmx_^^?^8vgk}=#UHFg>amZj$
zvTMFr0^&evk+$D_2Z1YGGY(VlVM|zPuzn@4pro+bus&>`4P_b>FNwv3?y{!26G+By
zCy_>DSVXBxMqMNT=Z%k$rp9QqW2)M}C>uCra&mlW=+F`XDPtpv{v5Ot)?{uE$_k;d
z$xxn_ebLBTCF+Goev9o!QOR3l>I!rP8%RXXT^x(k3Tgy0KbIf2!iPYwjN*vD(55B{
zsI-stEAaq2TRE^h;~>Asam9hF*ljW?fY`?av31wp9rZq-$c5=@*2>erAC2hGsB7Yo
zu7-N3pt+v89kM34JI7`bcF=#yz%J-zf|{1$h!uersFSEM&csJ-T`+Yz24$iuDzs$f
z*n}U_C4(IN9iJEjzr<|1S5RLT(&XB38lm?|NImA`)r1sReI(@M3?jA1DNPk^K4Yva
zTZ!_%YM}x^fhWm*Ffo21^lL!iu9pbu$PWSJ0~xD8!7#x8=p{cUFoymIb)&E8c*uxu
zX-X@yfLB<_+|lzUG(K&~i!2^Hjoqgtdqu?fG2
zPRrg1XJ=%IWVyS4v@vYrcY#%UuWo=ltxepS_h%0P9-ODgv<43pd(Yob;rW9fx}842
z7K;o4JzLRtZwV@h%>1Mp`pzf;_W=c(o~(NCCBkSvsc$rbH?3c$*ty{2mu_cYRdk&?
zPBbC6UWwkd+UhW6fBkPF>td(3@Ae`aw*X1!0hGbL&*JEPk!Sa^yN;S8G#e1)blY<$
zDYgvIll;ESM@%9g4?2I7Ny?58{U~Wb>uQ`9YBJ6osIuRjF$y%{q!7N_a*1%^3-619
zkN+n5(}I*ruOn1WU$r-e+n@B{kZogB+HKoi))V4VecN%PDUS(xOU2bu2~pkh^?H29
zZ>BXqFRuQFL)+yRr@Dx~zC8`1#g@S$NqGi3lfOsSmLW#YJ25+2*o4DSs6MlrS*ukrW_kbPnSIg3)o$`GYr<|K^LuSV_YKr7Vb-$L)zUbZnAMP4_#u@yAQ}C|3G|$|A;SeZJee1JAFT-)aC|I4LIvh
z>({u^zx{O=f@n@NKIYNu#Pn|Y{4e{Ac!Cy;?~4mOXdw38
z7geOI!C||j@-yGF15Ac7V8vI{>?J5l$0TujAJmr)Uo`fyXazyB_(n&T*JU}{uJ>0z
z!k`JvjD(F!A=m9#Zb581ED$`cPlbV8mA+KjQU3&I06n1qkXQqL9tnp*
z3evxVch*XO3pm{~1%ae3)qMZ&
zeEBDsYq#a>SJ*)ySny0fdLMvvnLUSO`rdZ>+t$$V)uyHWJz-($iE8_V>3fNJCz$Xq
zf4>=)E@TKQ4Ipm1Rx9mjJ4MMb%hc70o$+u=Uvy$Z}45QjbHHHrXR58z6G9)
z4d6fkc#r!28>Vmd;5p-)M(Cpf{eR12#<#YdOM|;wn0CM(AYbZzH@*m9uM;LH`ShE&
z@M|az$(sPKv@(5m{W)Pk%<_TYZqASe=kSm0F#8*D{e1h|@*5rSGzb5y*_MDbK9ZYu
zeC;{U;^YW|rM(*mK}yQneF`|n`MGp@pF{t4PyfRG_16Fn5W*}V-a0@`Aod9GQV~^(
zOt;)IrADKs+?M91$j>r`J2Jj7rM;0%yrMTI0`xf)E8=JiNa+O0Zqg3?>r#qk0~&fT
ztEC3?86Rw%MfbP$rN>}06jkg=rNQJj8DvcdDO&L|9~YWx)OjOMx$3ycu>uQu9Z6+0
zw;l=vS$&}&TaYwrSUt)hq__;w9eTL(%BrPj6>+K{=y9{P)Dxo)Br`=yGeqAE-^3GjAW-|y}zTh<+=KE>JHOepeJ`(9b
z`$*53#Z*5{tBP;6R&xA?RaD2|pc1ol%9wBef9Vdo{L|yxMJ8?_$%DuPtTqBj
z^wI-6<8_8Ia7cfY(#pJ*6c)1(>)DH|jHt?%)iT!Fne$Ufe!}N_is7dWY-n{QZ&awKY6N
zKIKkl9#9~qrxGC+HC5^qwzMlBJo*Nd8zZrWp1M2Mc+892IvJ+0@b)tn+YT|Rpko_snri#3r*qHGa5=AEFz0?qTB{CvQ(i8Ch
z?sWgcUBK}lFtgWnq=|TVWW8r%WrA^Qr5;N*1rbcMe_Q3ua}8PdUP;!CENbA7>Ocol
zT`$71jlW9C#fC}aM5Ix@{{aE;X@)~#zX+|y{ip-_YY_dznwaHOuM@+T7}8l4(b;ib
zwH5B_qdT23l9r!qy1=M!a3>&ejSL)D7cQjqf#w32**hLJ@4N6PaI%u^wq
z@~&I{xTZT}k+aUqYs_{Mq>-}Ra??UmR0?=JBg;kwIIc(>2E`O--CqKbRiEk1YkfeD
z3@!hmisrq~=~(Eg%464`qLCItI~|Wl>mQPbu&TvRAEgcn-{4Peu4k2QP3m`8tSe0eYfz{xq3V#E2PCYe6U6frj>B26FVYHv2z~UI
zPcN>(??t@+8Tc&J?flbiGU`3L5fr5+D$^Q#e9TvWc)=1TUH-hGg(@Y#Ov1B31h-=u{3fD0_RkMScJ%DUzIrpM;s>tbpBT
z*_UA{S?kf7EPcUtGB+SK+pSMu8bI5DK;^ybh>}95Z}f1zZMq5$3Na+hWq=!!#R*$c
zsr(ea$%i^r4aY9Gw|;Rnkx>6Kr2Aq66q{fY3U8L*lf}PFkt(MmN64f6yM4`l)bjK!
zJk=IzbdNhF2u%gVb*;gBV&+5Ke%z1Gx=`{O4{fK(vqm9jDt~Ka7gF7tV5W@Fk6{&v
zn*odYPHOJJI*Xyq*UFGZKTU+=!)|soG10^S%iZB>(9h$5;`MU)!|qDVKTOh~t*5`!
zs&c2s*>oRVz#Zs=c!95tS3L8EDACTbl(x64Na*UU`^k1uFVDEhFLC)O~2djaf!gm^u8RzM{D#(!|!sSzyS613#@
z(S=t1yX2T_Zb%ZpLXEvr`9j<{h0c&m^Bd%JU<{Q3CAbulX^tuO$IIx>WSf2E+YoL7
zh$AB_LMK6b4nyw(ey{+pdwiad3=NJ;;J_o6qkGr@Bf+n$l2;F^_tC?I?L(Z|rcE@Xe6^Xr
zd=@@_qU8Q>ES>Cs5lu@VcC
znKpa?k6Zm;%;(v!=GJgljDO+eZ_3Q71W9BK(2GyHOJ&)GXY8ZH5|!U;xFB@wKkLR4
z=|^I5eCZ`B<#6G#U0W$DZ#++*Cy=}~&e@?O(kh7{3!6y8su;6s0JEnB3eC7`_+=XG
zC5FHsClu2d@w`WY1L)MHL0VT2gm25vxOxkKhTGETS)GaA-Zhfn;kn<(dpRoKyp*Zjlfh#3}a2PId%
z6zB616vw_VH1f|-_}@kZJMr3-x=AF-B3s@aF1!#AE$#b?AQ11PGw9BLnDz)^#)RL1
z$#U6G4tuAF;;@4(`u(oTA~9rKohBj`G8=Mi{uJF@n|!Dp59c-9?Q$9@GLg$6
zADF9+fqp#9lJ~GJ2>bpH@l*jX8#V{PNa0E-__^AlKw7=olW;7d3$&r=y|6qw!@~Su
zwkZ?NS!Lk`+ciDzYdqu&lzP*EC_X$D)coa}qi>v#aw}r6V~^Q7XE&Sc>&zgfQdZ~m
zHM9IL$!2XCHU{fb--d+CBuqrvK1Z;!4~b$ZrG&|t2S^on)^|b+
zj{mtDU|$~#v922JQE~?LZA`5%O$IsJZEmG40jZmm@95~RHToJIs-W$02XY5CrFa~6
zh=u0sPDyfu?qX-X62;x|$dQv76Tj6HSg@*#G6N$57Dq3;8qX*}B?t@1BRAv)ow9l+
zz`74#H16+$S)3~kwnJlfT9L1G@Mkz%#K0%M7M|wWB_P7H`$~Bo!DnxR*_CNiHLw5LbC2
z8b2246RXh2Cjsqqw{PS+2j^KRDp^BsN&xC{ti!$qJS?-I-$QYjy+;cqi)dE|96ZaQ)sf_=d$NA}J
zg+;WJiGD31no~4>5xbd^)0NdSjfQSSlpPDLoJ_R*8eV3BF^hO+Ie$1owVe`p*jld9b-vsP
zE0@R!k|B<_Q6p9WCT<|~4>M<_Ss8yi!gB-WCTe7GD{J0Gcu2bjj$s86H3UyXiJwKu
znyW-%5v+pvmP4-3toAjw{us3=M~SE>jgG(c^>g^uYZmq^o_lB*@nImIf+C^Z*`6Yd
zd9q+{kd2_u$UZI<2u`EC`w&kHAj5$Y21(Zc-TPQYP~rp-qR4~+dO=Aq%ty5W^k^g$
z6r2S?Q~EZ8UHmyoBACQLJ?L9tk1VXc9G%o%vEB0gX~@V9pGd+8N&HL#zcwdI>SCmk
z9g#4FIA%4B;@e;)9+GS=n)b4Iw!9(D>B=`QLRYL|Wfg;r{rbJv#00l<(y0DUo7`iG
zwmaBTa1H@z@{<@u{a$J?IIQ@?5eDtr6iz9Ex?wh@udG^k$PsPMip+Ui8%{J?7?twz
zs%Y1b6`lS#gb+~`6QEv20ew%)P{$*wm~|JHz&%X5)CyzSB7y_apy006HG%RIP;mNQ
zcfZDD&lxPg9j)I7G%zq;v}88@pw6P@{{Z7GTmJ%(BcCoH$8Y-DFU}*2X`OYZXzCU_
z3w4QQJ|?_TvTBFajM%)8l*owo)CG&a1#H>Xggb6b`~HFXddA-R`Wzfb^6-Y=W+m6v
z1l@H-rQgyfFR(1w>^%18wUpo-nD62(|2NrqgGf#9XJp
zko*Mz>-xEF&*oCXsh`M>_nW^^53%{V>~6^a&iQrKsX#-py#}Tqa*%{O3+v%}>qh~X
zZb^gPTclaXEerfy1Vscc#$)!0SnS~1C=x-QedP|9d~11QS4$(0#jAEKo_a?-TMp(x
za4IxXDUHW=yj>Wfhl~p#6bOV9&Y`NK
z197%qy>+A?Tp*Jzo;Y3%)8h!8KRFeJ3`x@G8}Rm1Fx}yt}h{f5B;PA0)iDytL`G7wWK+-57YzBv|bjWxVlS`3eZLfc8)AVubC8r2m@
zeTVM0B`6-MC0VeOhmy`SNaOHRc!B^5B65F!e{|%%DsjHCw{KpTJWP&S!97zDnDenC
z$fOy;x3rbeydH3M8ZCle&+f}D42dFZrq@xH7A9wufpSZ^y
z(E?jOtQJ((HMSI>sJsXUP2A1`DVdA^VO5yIHOZ}fS8*Y$2QovPEc4lC&wWJ!MMvj|-)jm3PqhiPXhb0z?#V;8jwI%=pMquv;34-%5ZO*YXf5!BA(U
z#YG_2!{M69!hwW!T6Iy|bagn!_{Xc~bOlq;=@XgmpQ#+E(Y)+vLqTeFJ-*#?rUrPv
z1Z$?GjSV`h)ijM$*4IJKdWD2`RVV}wSpYSJLsj<|v?TCTG`Q?72|Z@NuvinD^}i@f
zVBs=W)FZAfBd6A*>>vP%co>a*D%Iqs9b_-N*=*^SvI^lbb?3t0=81j3#?9=U^zJlz
zA55yz$V~Qf-dZPn0wf>cHJ5-)LlRF94DmpfMQyj8=O+fSK`$0W@tJFm31U&9;z2%{
zSMp1Y>^Jk*N|sI9f8lsIwxYtu@bM*$3{AqDPiscA#Xyq*dAI<%{@I{{$I_`V7L{<|
zI8_!@fDSuv<8Vj3uaZ}Be>UQUn0Qe@U}ODAyGSdRrFQH4bBw0Y`d|J*A^#uH!6eu&
z(R~Dj^N|F5TGETmPc%qgYP42!1#8y66Y!W`q60YMk86ruhK#>$ms_x2L?_BDl)C#7
zR>8w7mtxlVFY^F(o0oa>@f&7q&B^+4)Q17cYNI#vlJ0-cnrdE)BOG&wFbU^N;ej4J
zu1|qnoK?Z1*#%fAw^7(feHFxLp}hRD2zsLRYmbGl!$R)E{C0!gaowNp$#k}Go5ea_
zyA9iUesoM(GtyioAkq%GZ0%t&z_&wszg-7ES0m^;1e<^l{srN#HP{uC?f(BrQYi{O
zdOjT_TGZSB0XIO%zc8Z6rJyS)hvDN!5-_I|uEsg{q?<975uecz6ueBlz7p-AJp<1?
zxVd0gbU*b6I^VO5DyYfy5+Lh$w!{bH9@08fabWYKt$R3e)Npgz{yZ$;{l3&%y@zTLCAzge^<8+tLu5j4^GAjNC0od<)7bbLbo3JTRW$76(OQ+3DS4UiOSrpap=<
zP}0s8X>cha)tDlq>dY`IoTvgHZORY!k|-+`6T5#wp9e@Wl>Ku%GKCv5y@fx=g>tZX
zbydRYi}1Yce+$}81e_+72jr-sWM`_Y$*EmO1?pjduQL~6v&>YJ+JmkF@+~Ny(ORWQ
zW>YfKq{d_dcO-r-uW8;9J^wK)si$jH{ZydDP96DrS2m?<*{}SQ<`)n!wpD#L7LMTvfd7&gr1Bc>Iz^1exjkdloOwe+%}Fx;s^a><}zq{p=Za@Bu+Vhja=slS3z0-_I#8B`hT}M8_V}1^cjf_T*=tUtlkdgRbY=1fm1DhaZXX
z0)o){e`*50(G{R8+z1RiZhmF?*l=V=bvj!jao2^od{z+^bBwm%qPt8*OIu9QFq<*B
za2Fch{h3=x7@umJlrlJK-BeeHl6^>1e$vAbv#c{9;hZr!muy@lnE%olvUkRGb*qVC
zcA~xM`NO&SNq8Ba$ZKry&+9!B6?J7~bNYvuf6lS5$twsiz(-|o)@UIXl$W^1xtDli
zoUtJ5Da{G`W{D}qUF(wYko16Sp4Uy9xz@7_5_MqZ^3T$S897m->5wwD!SW4mmUW#L
zj=)tz>mV!v1Snkc)$FosPxLg7?D?G?7Mka28W(VvMf>5KYT$udAflz^{q8VDvyo%K
ze@)j2R^iynub}ysrQAH78vZHgXjg7($xP_-`aYB|Y5bBE8og^K=^Ewv_h&UaXi0Vg
zX0SdGiV_WbXc&-Ul>A)OtUNqf{(F;%{Tn%H{?*S8u<*RFvnf5*gnV;aAJ^bx_PMB@
zbt8JmE~RlY;NuR92Nh#(`j>ET!E+zge}BeJ9L?NXGB
zfw3eQw#Tm3arqCgQp2bC-MtU--WlG-6Eiuyw75kS`!ErY
znYn|A(?w%B@QhmFWWnVx9~6(H!qYsAMPG6l&gYlUxstg92C5AP8`a}5
zRfT>nZLF&zVodeys-_&EB2Y;f2$u6fFhC%kXj6S;THfs86_S#7w?X{{}VH==|dYve@RE2Sith7-X-kmi_#eJzGJ*nrrP$HW+yZ+mbwiG
zspcAo!K%w`zx!Z4vt-XzpH>rw+{!MUIJfy=qWWs1Yo0XweU4qul
z4ujOEN;Cf}arkZI2E+nM2Zt>=C_Ca#?3r&)z{2z!yT_bkuhYIOf3@lv$_3p8_%i=z
zuv%(*#n)4JvJ{e(Ul;_MDnCi*uJlCdbFjO48o!kPN=NYF|h%%5Z8?oaH)xk}|!LVP0*+!cfGBy<1o7*bOTW
z)yR*umG{0HCG{)KswzXA8Tb9cuk<87y%Fe>2Go0b2@;*BaR~y2|yD
zdir&5Ejoa>(sGKiu
z|NLAA`z((sH1uj~H{TsXhS~F3O@v%F9}t$49Y5!Il(Dz1Q)u(@NJh>}&AphA7J;wa
z03opZTW3G}e;FmRqf=tfB@?t9>>l6f^0AkR~UJp)+l9iv+!&`gO1AgvI^TtxGeN`$Lt70uG-Ps%pnCDeDSOtoZ}aBT}(k
z#ke8mm}}A1kG(|lU{S9t8O1tcdD!EPu$D97lgoyzPm&hod;a;2jF{6&14h+!W1#x4
z0UxZ;&s?4YYTESs$i%f+3(Re7kqS?UJH^Vie`;I%R={9@Fa458^Ql{L00$-&%H{nI
zHldSr_qNx89DToGPr|O8PcuUXVkz@h_FwXgqPpv0?I8&r;hDB7mzWMc>ho?a(zOaN
zyxOb8Y_Tsll})UWi}Tc#o0p9}Yu@=RX(Hcl=XvTjtb)NAYn+`4rrc_WA1Hm+Zk#nB
zf0lbM(=FU!KNJyy2Hl+R_4P}645&NfTX3P7gw+B0dR}kvV`WFTxl7c6EJ5JJt~Hu_
zz6TZh6!eW!yc?3#+O&;WNF=XFhfwSIhWGemldPa{bb8>xxl;778
znn>#~-25xo1-;-MNI&gAoV(OVyF0-9e+l!a=^RxPhSSx29z>Ky?+q(WJYw5SDJnmEkZ!;JW3>6o|2x9Do@}C>;TFrj4^vXE
zy$9#9Cwi2qi?{cAW(PjQ%g^BXfdRaw4}3q4HJFvY?qe;7Y=01%ZK=BptuY%ie^y43
zch#6n-&7@*ZP9|dHmN3Xg5T+zE)hQWp05z(TUXK${$NH~*$tM#@$@_r>9K8|1+8M}
zJiXQ}U_F@k{k@N0&=(Bs324y)bE@Wu}fe)Llke@rtlr5`kARx7TB@~3b;J1%5k<wtG;Q2$0kBa|~@(s&Bavxh^cx+_u#
zWYXQ}I0GWq&*`VHmeS?3ZZYoRQ+}?*9h%kARsOzt8S)KvajehqhJ@(le;F}(WC(Xe
z_zFWH)fO2eZk#a`-Sx&%ql%S2G7`CV%p!)vob5C7uS)oO*kwHlF>y3E)vx@azQw0#
zbC7$@FwD*O&2jjoN+t@)uZGdH_cLV)w7#?h=SrGk)Dpz4Yz&;olcrOQ16Sa=_m9An
z(F!rl_`PE8tdG~=)CzH6fB5@|rUJKU2Yq93FRC6Wf<@6hv69V-`Sk-m(bsGv(uhVu
z;##n*q{ph13sv%(WcQincExs`3Jr@Oz1vf#uL~X@p<6Hk6cZV)r`(sQeQA%^vu;MXi#vM!Fp@PBr4B$L;rrW}%EKmePdgA5}
zGXCOuKZAo`2A|BgtuxlEB#4Liw8O^YivPV1zhV9vCx3)+INk_gT52R}0&Eslw>Me}
z@W!I0M556=3BwS(e=|ME=gk7)eU&6Pc)>-F!%P{i$b9M#LOR%D5>tPFO1j|F${c7t
zotFBIF&-@IRU7{N{V?{0LH-U@8ZZG&t1B-=`K}O;vWqxfc9PsKP!c}Od=2BcV{VMP
zY-CY!TXCTzX*Jdy`wE&EpN+OX4@UQL2=#psfK=e@;%}T
z`35d+RFovZAp<)WHBkWnsZlGX;P9)KXjDjPpbS06f3FJPq1(0{lQgkz9g%i8=K}iA
zb$V@m&%pw*`O~0U^SgkXd<~jn@={WV8S4uec>NKphhmjIsbw${wAl6#Ask8@jb79z
z(I;;G2Y?95a$IJ|L&#JMqt{~toAk&*>kgiT?d^Csl^P3l=?C*ykrt&hzj5Os*zb@~
z9QFHae?ARsP}<$$xO~;Gxd3^-%oPLPNg5_z8>4uc&|^CD-fjOKxOe5@l{J}V0pL^<
zCz1|!9zLaDK}{mK|56^zUu
z1ou^iKHXCCxMEkhGT4fX}3!^
zwvNr19);|^HHs@Rm-`VG%DGf7N!Qma8iV)TBze*YE4Hl@>gJ(CRHn%j#z=q5l)XkL
z_Utl99Wfc?aG72v*1m@sSvg&Ok_dgvfM9{VFt}Lf^+6KFy^Uzk4B3pF%6vz;hR3;_
ze_kL+Q8=1N_tdaN_UMB|@S^f}4+U-{urY)KuF#RBfxX?iE;pgv#AllVL`TMq$$nyl
zTPc0TnliB^8o~N$HIP2C_Hc%pn0wWaYuY3943R-e`e$ti+5+4SJGLQ@?v6MqYvzHVaFhl%Q(6g}soqLlrj=l%0HCTOEq=~f>*cw!aJ{0`PY^rCMGrXd^vldWIGT=)PFe`IJ@3Sr(k!K)-f-+Nn3IU1H?W
zAovGGqeWHyV)?nw#b+Q;srp_4r!4zRYa`Dt7YJ}$J_Q=5Sa#qZqNeL{Vn>=$et171dyxSf
z4#u|qg>)0|F^YiG@qKHP8&$CC*8u<5kBB&~%ytD|sHV2!`S!ErAa!xt
zk16uhbyoCqE$NxTwB%fA5N4mx5mZP+ey)mN#z^3}CD3i&Lu!Gkg>s~S;uax0(XKkmEe^pH+Tq(wd=RX`_pSxgUT6*n;ZfkI7{J8ySBQG%x0s07H
zn%6Bcv(zdf@*n`9!|l8mxd_sEF>2LZ4O%$pU@UlePTe`hdQTg>0P2~2l-
zEgtTE#-TLesi81?<(4F!$|V@j$*I7A#Y+p-C&YBwd!UCAe^60SU_QiVV=j6u2PbyL
zk71e_BP#v6o8?&iyokU0OHr}oL*2fJB!wlDU8io7^HLlv%Q>*
zi7?i&+t(6HT%BZ=$<;;ER3Xzi{Hewki8Xz0Xpbecz1Kalu8
zoMN%pzcNyQcdcDfv}G?2=-}lB6`h~gT`R%X1YWF&<#h(S)*8bL+R^S5?ThVsnXSdU
zKCIa%z|nU2{_3fLb4W#Jd#QOBc-na+WRT+#fi+5be+n6%Gv!B1SP>eM{SNct`G+NM
z$DUv!DzUZ^&2dhNah9Jx`;SWjysimrO8g4QD4ZL6Ds3y4N+unORlE2g8->a32wbp<
zgw0@Nqj3%b)%j|q3EfYY8qZ1dzm|^gU}-PaODYe=Y1;GR#rJK(FdIv0CcIWe90<{c##wf>@eAG}IL+1#$H99oE
z+Q;1xD@WJd!zZ+Bu3Yg~9MY&HaB+;q3}`{hI|fcYAYfL7X)tGmh8)KhutWq6e&YkOKK
ze|QhQxM7I-&Bt?|>?<0y3I1@ZT)`6lV2z)L>c<}f&*dU9n`3g!p!v_P`j?r7FU!4^dQ#0?i7*Yxnnj$XUqar
z!+ap?$2$e;3@zgH(78mTS0amg7&+cqU2$fV3(kI`C>aXBhXfICADeQ9V>uLuxKCY*
zp9{cDB=dweph?cj5;px45xW~RRL{O0e{=w#TwP1TBvH&}EKSW!#9;%=iRXhkNiTb)
zS(DiPGBn+m9{C81Qx%dEnfy(c+WRID)O?lf^*nOhtd~%c2+FGbIysnBM%Zi+BXN)x
z3}9)rKaF$yu$M=z@)ZXAi$z7KiHyK)($|<`Q=LiPCZ6s}uZFoJYMjKW71wNbfBz_p
zCHL+Y6iy+#pvKpdk|6zs3~a4;Rq7qlNEPazu+Cn9w{&INe;svv`H0eHgPQ!yuvE;#
zMc5g(VAIsbs<~~@dMhL^qb4|W3o1|Pb>#=*|KsQ6Zls~7A9|z{!(_vHHdmJ%@m>iE
z0yIo~(&QD_BhX@9VOK8D@V=+7&D64^L
zb*1o5?9(v}zUo$-ZG7;A_Lr->2h*sWrWaR^A|#Z9NyJA3jvfmIm%+z4qP%^igsa-{)DsoH5Upo6Mr^
zr?0Xql1Q2iV-0HRcDtcN&aK^;@R?D2N@FfK3`I2%NuprmXUb}{OOSbQxigNtTA&_rDP*ffD0dDI
z4h)7)nPMn`v~>
zf;;r8@6!N`Kh^~vBI9OG&0b|(fg%P~e0KIye{0gW+7c18a3yr}9}{=40I)r`;=$q%
zwq5+0)-^6`ypPKCiuhtUgb!IWbEEw-CLs?Z9z~WC@-zUW{HF4roTJ8F$t2k3TAR)R
z&+XbWq-P0Eey3)wt>mr(hkk92SaWmz8^6(U+3uRW-ZFxMmbQ`r@RkOZ3J2e4GOUDx
ze{xcTCdj^m{BIwy1_=Aq?f3nyjPV$!!zr2L#!!Ooj<&`2R6O~=S)hL?oa_c9sqlge
zP%GMnvKJIjoV|!#my`29w)H_U8$x8^up9b&?O{Ec+6B~W!Chnr63D6#APBE?+>e@Z
zYn}0HhG}e1Q@%5iZ_L{g6H$fd{H#Sve~%WJ7T9k^WzC#91o5fRaLJ%huc{?X#|zpt
z7J`5VzNWYfmbNWP6$=9)szKa^c`{;k6_8_DK)Kw9grANC@jZGn&*V#>_0)_7&_i2F
zF=uXlZz$cD&Wu)JB}v8xu!WX42ST`dH;;j3~OTe`eW2WA8k+y>Q{R-gixX+y+u4&}KN~gPFum^p{(m
z7e;Iw+CettN8<@2qMPb?Jw6;_c?NfL_{E3Hp5O_bPFaCZftgG$ut_9;Ke&su&{
zYgD>!hpL5YF`k-B+oJ3xy}Fo_Wx9E7)JA2;dUY;u1>jL3I;|AlFk3&j&3l7)CnoTJ
z?lRA*!Tq=sk>OeHsfL!^P!x*SGv*Iv-Sp(!cjP(r$@g>0(K|twe?TdCCABf3AjO_N
z2m!VquY-J#%1n|kNM=g<%_+O53@{f-;O$I^_^DAhK8vnjBpIZY&j>e0oWAMB}XC
z<6$%#464QtE!m|^u~D?8i*Prd^9i*yk4pGsIEY1;_uJ)`Fc!pWk=UQ~h_$aXJ}IL~
zQr5|%?_|301K)$xhkB7FaH~Nw3ME*6a*Bu-=bGWpB#iZleI*uIbmYXpa3d$;8f%--y9$rRts&22F6Dqc(tHEw^T(CA5b0_whU0pB&QWbsf1Q?
z46xNG#RdgHsebtec-G_|tBGS<(W6_2?s_`sL~K)nDnCp&X3#f^A!C{{&Hd^oH@H8N
ztF~{2C0N$be>RMI{5s1a*XQxgG^_J9@|VF*!8DSxvDF7O)#?P^$P{@WJ{(U-Z&E{!uFqjF4qU60HB*4#i5_LKc}NX+U-aKe`{;klm;jJ?^jNw2|TqNno%3{
zTRRH;T%6ixKL`!wWjpn^Q*#iHV`5LfT2nEGu_aW*_J$a{Z+^(AAJ>a+&CJDxno**AZVhz{aTD9umf30Zmw5siu4ay|8U^3UmV_WF
z|JmmUX=HURIsb&kSMSX&$uG;kL~bmQEDojM-zVCGW!F2&N$NGB`j^kLdS
zeGgQXK3wr(D&hC;A{A7Y-2^pdD{64jOa)wZ%lmPW{o}zSdcFtWc*MU0ia@GPeyQ8|
ze^_LtFX7O2`5r^XNulNXDw82O46Thf_g$`?d&fVies^
z;SkB%k!^=gh1Q1&rZJwFv`zH(<=u!SfATvu!3N789nyC1D=!*lN)vEp2K|lz%0t~@
zHq%hE_)*^;@9g;v@TD7fwkaUic=C
zg_=L|<3hB90M)>D;2{;_IWR|DLyr^(NHi7pbgvb
zgW!}F_EeG&9~4r+HKrVM+r%b4Ez*v{Cp66Z`qFU^2U9JLys9Cw_hTdutk{Xu%=UOcBEu1)Y6Wmwe}2%#M1qgSrs&`X{
zSn_U9ZKASX?%A?_TB6F>e^62X$$FYq_gVR^#8AaM#jopX`M5>rCLdF()jhN)R4uXVnmvA^;P6-umxk@-yuG_(+wq
zUZhT&I|(f0=9i;^QOu{RU;81X_MW>$*e*1eeRM#rLs=Np>%%i`;w!Zfg-f4saqOUu
z{Z{K`MA$pxQ}4PTe=>+wHZL^9puNkFPb?{y@0&ly|CCKIy}q2=;niSf4;f9t)h}Z|7MI|yi=JJkyo|8=
zho|DIITIK2ezAbh7)&Qcsjcnqf+W=o+Kwy}GWEl7i{guCe-7_kJ=dSsO5vAZr&e3J`g+n#~yEY=nXNW;?}68fZ=;>AK`&P)#nfxDhdIWgs|@wga*V
z`K&8M$a*aLe^~2+S1mujv=2lWVZz%jF-kN_D
zfBa!<>9g&EuA{aa64V!V!|!^T-lZ*L7p!h^Q5nquoyi<>?c>ec0@H)?tBqNdIck}f
zGXSut2d-Z9HFGJ{dg3ufY@lzE28IgkZG3iwfZ)!b^E2llm
zt!BB_W1-_KvN!Qku~81P9lxTWVt6O7AsQV=7>MPBMgjFG=E`57G*y3XdUMU7@nlP2
zd1~eBcHq*g0A)OuY}Zn5ra0CxHcS~S{$Muvnq1lJT%yPS*S0ymRpPi`J5RPb*><}o
zf7_Taxyg1r+qP}n_GH_(ja~2O`usUhhyBt>d)qe1fDm@Mp&KOEX+nRhN$f
zX38qSvBL-Fdhj$hsk-RWVp&OaKXG^&f3dXnbv%`wtlcz?i5G3cT^1iAKqkZT7~=W)
zhpWHmqC?eHGhGNhO;3MwxuhFtMQM1iVTAp(TOlyMV6w|!K^o(nTDbhfLzJps51Z*i
zhoKT_Of~5U1Mbr5em}}eH?tj8wIq-d_p`nojAqbjK`b1~tqr$xD&%I;Kxw^ff293Y
z^UjN42z|#cIeK4@Yt&B8sxor`88s8d-FDi7L_a|H)f`9W-F3Rl5w}|A?@naCbIuG=
z>+HF)uxF(vmpe5D9m6lu0yMcle{!au;Ac9;OJT|=y54PCgeD7RSiB3}ccX}_AegJn
zm&~UM=pzBDOwFWI4nGboKh)dSe{iXSgdNuP(VT8lkHX|@L~uk)b6qOCF47@R$!*>k
zk~qA1_U;EO@Ue;NNM=sLGgUJY6Aa}QADJR{x!+)V!BkCjI9^>yfEloW|0~)VVrE=t
zjG{c}1TJG57E;P}RQzkDTczgKM@l*i#d+WQzr3b&qC1W0*X#S
zLf_Q4le%>RLD`JNJtEd`-;xw%oqN2@TXs~$$5_x}!rQgxY;4DqpJ*{G6p89n3u97H
zzvbkI8-hwd0VFG;D!g=de@Vm6_CbHlf%02-<`{tC^KC6_Vhea4bA4KEgQiPGGOm}w
zD#MX^S$sue0`sdZg$_C!*N8V`^}7z+$H*QFkpt#KPRJLK6`V9fGtPZ`FN~#f<5B+~
z6QTuEQEnoH^12#cEXB8>&#qlXev|W~KwE%{WRYl^{{t@>1r|38f26P`37ix-{&;^B
zyinKPz!&NbVopY*MVb$pKe5HNp4EFBZpW`*v}}nhRE?a-=(mVox5jKyorM!^yQuLw
z<#LmTGnlTwA##b2)PxR?EUb@rwXN`e(k1-xE)V>T%=%=tb&IyL17jkZc-g)p0xrM1
zI6bfti)ETi&{G2tu
z48ZeHJ4&v_fAP-fX}lk!o^y1yIxcAS_?GY@FfxrHqjJih5BvQ0o5s{+~Tw&t=
zp(0Z`G?ev`(@Q$*T_TdJr@kA{NnB#rX!yf86g5|moD70%eJZ}mu4Ci!+ng5wu{*(>{=d(4Zpcp4MgFTN*R;O5>>rD@X9e6x6pBn_IIHewdCpH&!o`}I^&Jzu@6bZ6F4YocAK|)un0E(W^HlFta
zwcHX9>)-e5YqsPgLuZ;Qe)Y~-r#K}m-Uql}?;>Wbzs;qpWz1K8>!8Y4JVs({!b%Cp
z3BMX2Fp=4|+;Mal@Xm`l?K(}zlsk@ttpQbbe`FG_puge9ZTydeV9+r5h)9MD8A;X%
zFC_|`b!<-Jf}lu0rp53ADVcj67!@tcWxDu*W%~b&gs2I!{&~pruhz@9-}1vggMQV1
z*c*F`?Is-6mEF(S&;DxCP1`8yo~zqAbh`oRt%ALUsw@#Q(Tok=n!Y}P?lw2m$DRL)8{-xsCe+EgNMbj$5*MtS9%u>;{P*NnZZMvc*rf`U+
z2>ExN?)xg%Z-=Q+9EvA^%*E7zZwPP>A}g$1;t}TNnjTyVx`!f~5^Uh5jJ
z_|hdZXL7p1G__Kyce^*mzzV%>NgTYF^)0ETll~!cE}@%TtV2?khKR(xr*H0je{HZ+
zcL&=o9;sNFcJb6rrdX^lbe&%&XfENg+6(1TwN2;o?$o{P5590w*SU+eXpLxivS;+e9tZ$k#%V@T0>I)(D6ju>Ihyf3^^XuoK2UR
zRJ-v>8)JVwmontN&xoFzm*UJBC%V%_N8#P|nW1ydYks51Q^Xgyw0
z_s>hNl`cW;L;0<(Q*@zn-y?&l2w`qtYV|LCbf-8*P!c5yU%!4RmQ!{gqR}lwoV&Pm
z(XvPNvHiSMg6B};e?SsCYJ^A!v`Aq>KBRqwlgVHz&TFkf!p~Pcf65pHuY#cYC02{{
zwN8?aR_%crjOc=xl_9MI7f=Iq+jMHfA$FNx=I9F4=Lm=HgG`0Sz_TLVxz&g{o9-+N
z{TV8(*o@l=z!RbfJYoZ@>m)XM-|5HW(e9h?SaYPMfhy*!e_FofAUILkvK$9uKD`z~
zE=t7`(wxjhd(6vM@spUP5*T_I*&xGO83Q?@_H^A>od~}|pJDc#BU69eI`igrHi?1h
zA=BQM&KxGwO0wx!bWxhB19ht;^=G~DJtp*97%Wx(ZP%d>LRPN~4D{mHI=g9=R#FlQ
zZSR{kOpAvhf9D-f|HW$5*qn>#dT?&5-8K6R4XUv#S*5Xf@=FK`77y1iq`=`vHOayD
zHe!u+A_tG7J*<}a^)3Hk$hJKDK;*o^F1_fthh$LT`*G@{7S>qqjVmJMuZ%?m8rD)#
z7lCOS&mq{amnRIIbla_9Spk~2vS8Cu{66G(Bu)|sf9Rz3v83lXz|aq3rXf|ROfq}i
z93g?AefPB@WXi2D0baWqo-AF(Odp(DkxsI_iN3jy(v=kcYjc<6yRTO$!a$AQVe>>>
z21rpg2z@0$Z#G*!bV%hzny*8YG0n_~)aPwTnzJ3jl9%_AQ(A?v)#y20RVDwf$v_<=
z9qg!4e|Juey~iL^Z`7CWG>;3;jkj*J0r4D`f2k9@LMSy96dhQp
zA?9t&uV*5Gl*P@sW2ruivD*?$?hhatI6
z;ffMg@H*wE2g0vZ!4v?5+^ONtaFag(uvEx@e=ifbPvvDCRI=b(%N?Yry|bP&Ok_RsR9!yRFmF1iLP#phTfjf>Qd?p}kg
zYe~j6YJgSdG5b-hOQ~JpK`Da8G7+=S{d`B8aV#EMH}OnaT87^cl2A^!RRT775S7X2
zf38kSokc9~b5x54e0#Z&u
zQCRX-CDINm7WL+uE6{m*DtqVGGz_=-e_&l?iDZ}cV|~QOMENwwj<8-CyWGIN^*uZJ
z4GU`W$XU)S!?ub`SHmU^P>9^-vC2Rg4TosXYW(5#u9$7mY-*Bw#p+HdZka2%L
z{bVT#?j;(cBt7_X-ay|bY6&NSg^*|=H)^dZ<*pwA5p&>~dJU9-?x-tKx@!X~f7N9<
z>fg(TA-OtMXSitp9e@V$w*H5jVT+V^5=wVwq-$rbh^HUfbVuwUo4*3Rule?-Fyms!
zj+YoaPXb>x+%Z&>g2-k5AT0~9Xo-yX)@0Sq$=>jyT&@h8gY}lKr76^Hhaphv=^o#<
z$Itnx)s%LYJ>~vU(O=g&PXadpe~{|MkhIl#otWQa1rT2xnmbWwa0}(GjCTGr(>byi
zSJ!h-_X1s1vF_VcI;wD2JccG3L;rTyoJz^?=5OGIuc_w8+I8H9528duC;A_xsx~{d
zkL5%Aq(*P>V%1=kZ7!Qfg_1`=2@%cg9F(HHX)+K>y*Qse-xB^#K6j4benRb2Z
zYz^s8nH;|pSTyv?XefSYji)F-^i}*5fwi2vg;e>vfDYtvNtSb1oCP#+sG;(a3pgES^?g-}wSk;%8kQ@E#(JHaIZhZ>cRj(yN-jT{-mm(dh75^FzaJ@et?
z#hUK&q`NW;0`0{htXQM35WL
zbv$4Wc+*Yr
zj`Bf|q~bK$XeA{E1Yvy>lme$X^xNqPxGgN+C&x<;z!C^Ue+aawt?7<#x2oFw4~U=%
zzEEaD2OezMMh5v5rXx$dW+VvMS86c{_rz_Z`JS2$DH_stQVNWZD=i4@FwZOd-_S1$
zIiJ#!i1{CIcVp=cB&k+V5+x{T^>L0rERS7}sQkEzAlDL|D&eA02urYE*g93H;LHcN
zaBV}F+$Kr}e{NB1mg_uR>6vXyDFvGQ#ii;JJhAk`x8Cwyz_Nf2%szq>Bh7)AND3*A$Y@dF3Ch
z)Xdt3sAj?>Le#N)1SMY`iAs7+4GwCEn>AHXWQ9<-rKp!<8aBYM_h+aaOiU&ZQyYYS
zBVyS~PkFx4RP7QG?1~CP)hiA?^P;{Pt@B4+|QRx^PQk*l-26-7{RJ)t!
zWf6+6e=Lyt;;;6HjZT#5;9QWsTnGo>V6^)IPI-ToNUQdztIy@0mx($&@*%86b!kiR
z<1OP)f0^aZgCQ>2xe>)I5^RDUOoKC-_2whJ+RBX^xUVkm0J*;K{PrPNfPsoxi00k6
zUwjaEsBeklW!gj`n~qqOol5t%GH>Y8m9i=QfBO2C%w6mN4Uo{oCI315s5(hT3vnT1
zfaC9XxJ&yXNt$b?g|q=Rm!sA5&1+U_T~`C7tTP~_eLip`^mJkjOGelPpt^q{K
zXKyLlOG!sW$+V)Z4zSHbPC^$~{9BS6d>yZFz5Ea=%3t6$;gJJ~ChzCRrz{LI2CVFo
ze_Cw(&_c`rmF0cIP!k+jPrL_Ko>V+?13ALKvk%SY`I1?l6v5L5D`~1WR=zz(xhmi1
zUYfnM@`1stGXGfJ4X9@N8&$OQ7H3(eHPX+D4*K<
znWrejKCc<@o7}hX$x`TeD@8>KszMmaf1-@aq4aXE@3@DWlF|gk1&H9iO83Adc+FZd
zF^33Frz^K15zk_4-8A%%?394k$oUb}908Tz4`l<4^#>m6^JWTn8Eg_s)s4;36w=oi
zDig=xmy_{E%h1>-d6%ZLfwMq`RQ`R^o8xEHTL&z3R{=JwK6VVh;)5BP*qa0Vf6+#@
z)zy*?+?Oi`@?15wJe-lr5k1*a_FGO+UQ~Zjy>CXP!z37fwAc<>)s3`p+mTib8VH*w
z%fEXSwF~hV<+4;m!#K&x)4@|iBCAk3d{<@?UjmF39)>krf1hX27X8XuL>l?7*WKa7i(UkMQjrSJX9#kEiWb<*
zo9aN?EoT45QN1SWyAVNIkT(h(27oGkiW(8G83Jw~ZF=p(hnB{9kdp6N(==
zJ-wCxu-2$vwyEH61#G`|62={ir=dp|S~xVwPsQnAF<|QM^w(`qR?qyEe`A+CM0C!O
znLjeEi(T*@VCiR#1SD$7`$X3U?oW7bH=H7AHd7r-R=RVIr2Ak9jAgLuzg&;L`sg>m
zHLZoToVoPIQs;XOnLV|Lek#H2g}q~NWeeZj}y8mvis#CSs>D~KRpFY*S_g=qKsO_2EYLNfT=DGotq9G>80#6aDl~(l<
z7nzVeu5`^|W^Th8AnfSU6qGSA$V`E3lt)X$G@`8!mR&xnH)hV2QF!S6xpoDNdVGG0
za55yv5$JUjbNum;`kvC1C6tW~UUM5QrmoXvBRNs(rJ}r1laSCg>2bT%%rx
zucdI*3wZ(oo?E2Jm?TIOd@w>JldiBgDg7uvr8=DiP*wV&X<_O!yGfsD!MlOaAL~W#
zA4uGdwIs}Bk~6WPmadqz#MsP|#^k)~v&KO#S1hW??Kr*xNRI;~R}zP0$nRp>YHV0O
ztqsJYzHaVk4fu=ms@(4AYG7~5wgp0EdVgfVE6ZhR4
zXQl>~dpZ&&aeRlx%kHp-NnId;zfgZD?#sUHJW0mXi5#X(e|W+PSnPM{9f&aJ##RtR
z&!-33^d0lJ88CeLA}@|i8al0k*!YVD8{Qzng0Xv@$tKTMKJ|krk4Kh(
zFSpMFuU4{g`!8a@5?}NeV3NoYEJ&mtVK;6Ghv88p=IHtgcQ?G4QpFHbVkZWiqG1^V
z4k!Xrd6TzhzL+7-yCHUlIr1ATvs
zbtGY`%m4W~U*Xbgr34@L17Wc+@OLeNUT5HNMM1n$s3N_BbgwQT_FpTM`T+mRXtr-g
zI}C_mK+uHsL?is&sLF!uGNIWwKYqc_uIx9FUzPAh*|^Qt!G6Gk?4RIZ`6{shuHPxO
zPt<(A$gQj|_Xqm3jJFXNwmy55wFVdNoJiGYE5fv=Zt*{5cAh22#aXNS&q|KCW{3Rx
zYbO>e8G~%ym3aH4P)13bMXy9b2b6p!q>f3zPB}hDP%mj65_yv%q>H;(?0rrPY{g>CjKIg>;*k@jqzStI
zqHH|KnCKh%VfD=<7h+4k%ggO4$!Nr+X+7|Xup*ZEOMTG4d?BZMjL9FNorWll>(+>h{-CjOgG^T+~swR3>fMib1{VCyFIiUfT
zswqW?Jc4`DcXi6Qk2ttV1gu8g+$ZGOXROqW1gxp2i&jD9Vhx(%>sJw$9`e0Ndno5{
zdR}=n1zodxrdv2X`7I>KRfk>dA!h(beN_c?7hMA@UG7JpIX}DdhOU4v0%CBch)na`e?1D`
z_I_9Xz=-HUchw9c)d7Cd$bxISF1C<4mkJk{&zNk{$AZj~O2_(t72idA9DK#=a
z+^kc74trx>n1MOsXu@4Z>Ic()O}G;~tHAgqZLrRLa9+#&Ael_-ouv_t_5WYuyN_f3Ln3v*KUXL%HQ`kfJ`SpS9
z>Nb=yUN(%E(QvATnL~wl#B+YQuEaLZ1*KVMC)C@A24_
z0t|AhEI&RHE2)J*=GAJ<$|HzgB-DVjmXJf~eA3UFe^9!Z-Ca{fgugFMhu)E4e~^Ye
zu=m+G!9t7Pr9&)%NyWCNb>QW_F#{S~iubKsVTl(mv2(wMr%i>Y2%-#;TN7!aIr}`I
z>UxN{nz3^o88#^&;70xwfZ2JhaqeFS47GC(N`1ACEhoRUalA_JlFj;Kdq4tDfeuWm
znT*si%1MP3xn-~r`smNrfss!ENZ?87Z
z)5FVXHtoZgF6@St6>7ds9_xGaW^T(YY120=ghATqANsg9an!Me>>C3Xcw6*T`~
zt=oisJy^^4rET={kauR&
zwVPSVao|42iM5K4^fC0keyy|qcBjgpcBy@H^+L9TY&6cyWe#z4OfCdfXi@SZU%(p!
z#Xd#5F>yn2Idl1?3txU4a{!XP?u2UAiU6mAcUHTV
zXc`Lzt8>&PT6oyaO)K{;AyS?3qet~4_4yi${m-$mimOEnr7mXiq;CfOS2l|_x{l3<_SS{&K
zGuYvF0JrTFY=HMum?(OGtVv9xKaCo|2sUJATHYuW*H~+6XA)=^_`NoDfJGt?>9fBR
z=}4MvMWoC=#>9klp&vB5NZ)K4g9!V91HXc@^GWl*gH20@33xuK>B^h4XCUR4?Z>0wV`9WR!#_fnwqzSgVL+LQG7K+e9*7t{D~HEWm#L8a3Et`s2o+U
zG3hS9DU*~+)obR(0UMdgCG2;b7#HF-6DvH<2JR?iFD~Yq7f56pf^Ixcs@mq1gd@+M
z{KjKXAVAqJ_ZryYCg%H#Zx&+eQKR`!61H}pD1aKc6Dnzglr=lc=wWRUr})RaZwiln
z)-?+-Kq{#CGKMb{K#)V)L0eg4P65W*ik<@>KVHoUK_gY75_KM97BaP3x44Z}$U
z+lcZ(j|>$nScnP49_eqK;=3CdZHkc7*_oCT+P#LhhM@yl2AOHPqk0s#{O!QyQ?-RZ
z`SSq?Cy8nDWCGf-98_ISD{<*rg`z%!@>57IzKviC9Gh#=TgR^_qJ>^gfo`1B8JplR
z{jERYWsag|DCcrrwmixFg#CwBNz!={e?>+%hO{``oXdN
z0&De}=m(Yd#iwU3>9!)d>DYEdpW+En_hXht45=#!(rPt{p^;LWzKCkV%)zN*
z#~hi>@0&hiG?tts-LYZvgC-&TVq-p1`d5~@1$llQb*uc;eDvA?6HZlSed7|k>3}A3
z*h~n3A9i(Sv1dXiTIhlw);F8^v7NVob2gi)+(V>b5bR`H{
z`*#6qO+G!|GbRX2B&N7xS{l?ldvF<{L;RjLb
z?Dr+rN-uB&Xfn7_h+sLM6m}Q^sV_+eU#)p+%=U&Lg@gx%luy6`CF8c)z6POQsFl0&
zcs$uQqP|nTL%G+Aekae_1D6as;;)=1*cc3%+P}y%quv2Lk1VVj$ZvMsv4xc-`W7^%
z?6BXQSW#|HG>ObRwgQO$4_J>?^
z55~E^*D6?bhRTYo-x00FnnB*YMkE@$Kl{wkBO4IIX0{+MpL>L2Bq@}-%w+l1mD{D
z4S4oF>Gk|>U;D2m@$UzLitS=AUj0?nXdUR+Pm^ITUvRvo^fkA-qdp&$@?<|ob~4`gqUg8q$80ETbM_XI|%|vL<=v2ME|NK&^AqLm!B7X
zf6gj5Nhy5-6&xWC$E3YJ!L?s5tJYxdW;6DXy#NHZ3HG$%&5Q4r9&G0j=2Ya!%^e5B
zZ^k)FEnlg81V5VfilE_s2VRvhd1u=+?v;_$b9^Z~=WBy`6Pxt{EvlzX~c6RLT-wFNqb1beWS++kjpCs-o_j5Q4n1bV_=hQV_iJGj=eka+^}ME$RlzCmS_5iQ3RMU?B$C#SqM4xgn6nw*9FN
zGqqBV2^(r#Zll-f-Bx^>yrHe(-Ct>D2#^F%00py_qX&mTkA|;^zlsIBf+zCjRKg>o
zkb=j^V)ea~g1o?aGNyM8+|MCBW`7~rdY6^~D*b>zjH*H!`=!!r^iaLMeku<7L?|HT
zUMy(wqauiI6U1z(oEe^8UqdjVPkIV>w&WE5)3_05h>mMPG2j6Y6jQAXd?vCPTv3&a
zwg!G_8cB3kA3W7PhbV4+nt@K!G$4jmIFR5|8`4iIOQ}xPpw%h(JUHu67+cB*_3j7)
z*tZY2VQjie`uei0gqod+HP$@5I!@u}qDT)dhW$`s`DnDdk8B&|Ue&`k*2i!Qngx3V
zOB%Dsy8d8}#8Nj^QyF3DjNUYdCRc4US+NtlNu>sRo(do5?!%4DDL9lNT)3H
z2p!XKsYU9x+xaXCK4xMw*K}Nb&p8*)k4;p`@0V+MZzu6K6!gAyrlCPn42pl%-743j
z4%@Fwy#9E}DRxbli*7{E$f1h^ZwPDP>Yvq8CY{~zNa~xbdl%l>m!&$%-OanCP1{Q4
z2dRt@)jt=U!P5dO8x2he2l54W>M=ap1VN1#Z%_gqsTdz->eOMD7DpDfEwXc|8Lzm1
zD_%T6S$?7v1z2;^HInWUEvx__q+%jdGXiqZd^aw{a3ypY(y)Kwutjt!>9v7n3>1)428;C}Vsk>$(E
z5^@t1Bx34$nrG)ufhQ!NV)KpIi^3FKt%ztFFBMVSW$+cD2KSQ|&k_gwDJSYJmp&EV
zU7Oj83-rFVe{)VzxHL6TTNlDK8C`^`iLmrW>Rg0G_|a4)Up&H+x1nQEJa2Ev5F!kt
zYKhcM(B(7mbs^zBhLV|_XbW4*0E&b>X1~Rsp8J*+Yv84?yZN6Sgf8v*o>@X`FE2Jf
zjS1xlpet=<1?g?fb}o9b`0t?#=HVt%aJHR2S+!1o-V?!jMli*MOtEt>dLC)fzrNTXvAQgf>_%6l
zp&Pp2=+e#P5=-LhmxJ#(UN+kTH^3d+e^*w<{B{wmU5fXJ=~vBdVS#}=UdyxhH;!|r
zD*p5|w22lv@tqbxc#e(0(%bSR*ZYD?{*U
zRB{+0S*s}0Hc2l??=Tw*khnUK1O8wnj`T|N~ordQ1B(tQx#
z@s=^{JS|7r?P*a6q*Y3phxN>_KdMO(%oww!D@r4Hbk4-hsI@MOV4`g{P)kw6$F+xK
z->0=|{Z@BCYdh+v>wc)Kg$g1Fp7TUX2U*)&*9izay{br7QRBMdkvFROm)rQj@pE9F
zlRQ+$=h-X*Yl*zS!8T0=nm$Et#0jU2=38M@#Fm3mgfYB;8tACpjL
ze|+^eQFTNL^X&1EIp{-*zwGAygYv4hn}xhgt45qTyx3lNI?e7k_yq`Y>;H(9p@0Dx
zQ~s%xp*B;GgG<7I0RYX;i{Rq)U;vi?&~xBFtl+@_Z2z&8kbhudf^sxlQ~y8>1!Mk?
zYfA3R5BTQ7x*tcPV9bF34E#qaae*=acaopX5YP%>%xujl0T3eQ|NJ%1AT-RtSpFM&
zNCDEE4UFZ#uEHUZ7D!+$|8?z6g*^BL#`53z?S+u$(Esy}unCL;ffMlm3a1gWuracx
z7`TH`{acofD3W#F&($Tvn=1m3pN)dnKOkXYF)?v4G5w)5ZLK%s$YMyD&48arqgIle
zbtcU{p2j2>O3uz|sz;?0%9@qQE>e_vS?E*qAxLRe6?s4%NJF$mU1T=#w~4;E!Y8Q~
zk>^hIt@sZ1KJo6qooKx~Yhj4E-X2J*#nXiOz+q=prg>J`BwSsnc-3rdFj;>+E{(DS
z5wteOsjE|g`_#Z>0V&Nk0tpnH+4x*0Ek^PxV)A7cjlg7lh`Kf9+bwhfI+spf&z1XAPT$daQZz)w6&%lsC7k1tH-s8-Ps;lEnrz
z@PsW5QEFkN+Kw2Q!)no1qGA~(q7<@T-1kAr!juW1?LS9`6H^LB%V6}HkCKXJHKkbn_4}J`{HIOa+NDB&{yKID?C6K+fTY9sz;FfU1BSx2
zAo4I7S8`B~Ofs?#ib2y{gYAeDg82GMNElx4_Vd1qc#&5ZBk~~M?f>20-Q(p5XU~K3
z9#$`yP3Z)Fo~s$&E$ZdIoRG>eElIc0(H`uNS^#*vE)b{{pWWAUrpr%LapM_s^rmz0
zrr+A|^T*KGzG3m3J^xS@S_*6isLPmyO{ca=$U^?5;7y?DN97PI^Y1Uc1V+8q5{m5Qn;HY(_bCE2ieYeK-0!kQ-?{p
zKE9#sYp%NIroXb{NxW&zql;{#N`a3$=7!at;b2u6H4dF7T+#n@)22P0_KE|VGFx^x
za?5H;NFv!e-q4SWp<(TB^RGm1*9ctRa_u-er}@tw1f5;qnyfEJ4$xK*+o6AN{%YMH
z;q!-57B}1z>>k}_gRZA22Uhi=%16%y*!B%<=8<5^(ZpX;S}R|*M)zRFN9C3OA)J${
zDqyV4fqQQEW7+$we8Fl|fz@1IV-=#!tftLmS+o7Ef;;3z0&l!Q1%Fs@>n0@Zee?sh
zjqrQmcx{2Unc|sZe%@-#xQyYh)r*K=9)jovLJRh2XEp@mE{2{$?Ub1x`}onUmddTk#L%JkuyQ+UU{i#
zd!{Qa?B*!5X<0?*351%oL4Klt`g3E+8$#H7>cBVHD|rf%SWLKRKiRC*KKZy$6ex}CmrtC)UWVVAR?Hg{
z9gLcgDZkf|>fvd86Av+D|I{*?ZwedMKbGRrot_F?{&5JiU3R9QjWeOC+{^)<{C34kKQAsr-@0K8%E-S_%?bmCW^Z`1@^VCw9WOL
zkrIfJp^b50IoRK(u}Uu0s<cQnxM
zc9KMqcJMp_@wk-h{yi;0qO_Xc0&AKnP#Y8gVgsMaLh7=STFV&>%Zl9Y=0KRC@I)Iz
z3pJpWT}BqCoE8pEok+J%7qy|qMp$lnBL-|0=*IG~L8ksxXX%d|Czaj$HN|QIiH{zz
zD%EKs%&(kmFO?=^u%(ftN7N+EYc{^LUgvo2B;36%!ne2Ul4>6xXx@jZMzdR<`
ztL&wA4!d2g81q#ITvsZM7f4x^n2n4-ZpQ%9q^ez71ZzMCdZ3LUNjTeK?ha1rym9`f
z|9n&zxlGHOLve=QM1ML^1z{&q$n3q#tsd)1OXkUm{K=v$0FpL|l1x5mja
z9vf{Ys%>R?AM+5QKVQ^;)0UKeSv&M#CzNBT!`VOh%mZYMKv~Am#k9
zpij=Esloh3>JyG34BYFdMnMXp5r}a6l0xA~?jOLeK1sBq5)PNH{Dg_xL!hs|;w+}Z
zPDC+QF3HIs)|4Awfu2MTnoFm9U$ate*;Z=WaP1nP@@Ppy-KP1OI-n;wt7s$iM<5gRYwu$vPZPtet2lW18JNA3p^AjXDW;v$R
ze>x~w1=&6gl!=ddPnj;?UFFm*uAyK;xZ<5@c{SQ1fP;z|GA{_%D%o|@)Aa%ozaEWo
zW|yM3NWA|-mG}Q9=O3%SlhPZ%1~O>j!j@K3ZEH7n`>xt9C8UIeRO^52FYeAB
z>xj#{;SbYzQjWJ7rps$_$G#JQBXJ$E#+-2}cXjFT6n`D4%^l=Gyc2+x!l*8OC#&-m
z*U!%>h-vGD;l7l+QQ0GigODPYlMp)Wp(Sh7PZdO8nSBWE6rXZqGX^Wh)xdhhpE016Yy80@vz*M>4H-_c)(
z-Rp#Zbp`m`+?`%%1o$qdr~mzNmfuDmkpg+jrmDm~Pt7`_0>t;61&`w%)6q8MSZCGX
zDgu{5C6%^9yzRs@ApNI*TsY5w`m7WGlw!vJ6u2;byuSaPj?P&Q#;CliORpOe
zVje5vwit3LOv`Q3(#wQ}hkW+XUL3ZB&@?w7_?ew
zH=`JKyMMhrUfXFc4X$sBE+6f}&Ac-)q*osG3d65*hx4N{^xWwiRa@b7p}=+f+#aQv
zLq%nssTE`mmYMkkVaZr)$!u&@C^b)>p}U+&4RkN6cmXaEzDkY?&d$A1=$MDqfO?PO
zK)CRsZH2pS_VC)c53bg9+jYp{etXebw!Cv7GvE_aDwK|(o
zWbYaqFO5Eg<@1a+)SAOaZyPhc*GHCbC2Yr4wiy
z=o5?KbftfPh}y(zm?e(NNSu|)mG<2h`Au>6<#h9Lxn2OF^$Opx(Z?RUh@^E2{5v~#
zrq+Y*)GKNhYY65G;ce)J^maL62)6d**zJi)LEqYfqqDlSyG$yerdZ<
z9i!dPTCfh;p7rzQa9m#e{rc?U@a27Q)P$v-d31UY^!Nq)@7YRCdR-cUw84gp>xsUe
z9K~`FwcMNXRvFC}PpewfP(t}Euq2<(Lq&H+=gwN|@!!>ER_t~1+Tb6VjPt=o3%(9q
z@09(sNCa5Z@60Fpkbtj}s6#og_lukPjBUM!+724DLOe3h54Is3p2=1O%I11oV6ckO
z6I52hREh=@rA@i(-lllY7Vr0`>iQY?3P!2Hlbc4;jSSCr$l`jx8Dl0qP;Ak_;njPM
z52Fn(^3{naH4}c=oP{ncWF%<~5mmq5
z-JbjA#o-q%7OvgTOce*0VRt=AL3e45LZCA-r9VAR=}(t5w(uSH!=47o%1~9(`KdtB
z=*keTN#XswUEy;oc=J?!AQw@p%6lNaQk#A7HCz1BVXX9qoVdpBJlG5vfWQt*8%gK<
z&$w@BiEl_~le?E0?a99?0KsZ{i56f2sx{^I+d^_S4&Kw{0L=rX>G+4!Z65b3
zGX64o`H|6d6jM*%IA`J-u4~ePZ1Kh|>)XaJ-Yq)yMh*OY|I{3;hTryb+8}2CDkAN^
z1~O(&9T6fsg!8_Z>FZkv7kskuWaUi1xaQ!D-Zi~8^X}`alT?)(HkJD2c-CDQ`Msnh
za?>s_
zyGO`%eHT7+{=8`O(Xty<*jAXiuG+zy_F8^y7i`gbE%twz!Z#gl+2up+*sr~8^Jf=Y
zK1TN~>vgkg_xT2mis3>3&zToHTS{yXWC#=+0Km$~%AA7V3yJ%GhF~vbG$=>P%s&o`
z9fm>7(!tS*keP|~|IV72SQwf9=d6hmSeYtqx6Xk$eEo*Xv($c)j@#vzFOXqfzuAE<
z*(T{S8<2q+<~IOueK=#cjTK%edFIzJLKMq_JN5CB!$c+OFqGyXgV_s``eTX(>&7T8
zE*N^e;#kJwD9MO0ZD)(=I1N!R7wfP8JsM?$L;=>0Qyz#eE1Y&4Xe1a*V0~5*5U1hJ
zpqJkeZwoUL4&ROz+!P6tBmh*=Ppq63Ifny0Cf2t?>5t&Pgta|xND+XgFqH+SfH>R<
z*&MSU6@$ej4H=&qXE?E8#F|H0D>r743@2JRRYOw_oe=3nY!9w>M=CL8Apdsg=4`Xw
z%H#Iaa;i8-ZM-c#n;%%-Pl6>1Wc}n=pRf#+ML?*Y=F|J|aq{9Rzn<>q_tECH{sHu4
z>-O|Xtt@8*0=rr@8M4n7v6mTYDa|ePE7%{mKCLI+Sb5V2J&AJD_2_NS-{Hziu$TE$
z*vr=k@1mjY0uDdlb3k}Fy7)RjGQro7_;`o@zM{(;ly`t_g`7upe{qiiZ#RBs)7=Kj
zJSK#By|KQ?@#TDZ-$1`ksm0VEYM9pc|6)l>LRY$MD+oIrv~xpF)Jo|y&n8J~-j>-Qc_RTL7q
zJlkGX$1%WAXnf35<{ZX-WpEFUBy12kF4hs~-)v#k8fSz1`Si#@+f6Q&N#DL#dbpC=
zSSHNEBc4W$y3mEak$Aenc`Z@srpeaFv*A8dHnOPb9x9!5kRO@=0z&CB^jr?Nx4-D3
z=-Uwx30d8TuEROxLJp!_A*1Fd#MePKdDvuld`S7-%3d>jb*T-8kpD`
zW7)_n>q%|u%n}NLQL^e3mYq%Ap^47ttRP~0Eh;dQdo2wX@dKk;I`JHuF(fHcWNHlt
z=5MtVCdZ30MG|-^WdY+VRA562rP&f$rx6orZ4%Zx4Dg3m-drs5Eq`b53HSEW&9yRb
zO39qsr0G;-OO>Th9wtJ{EOQI39Bs28OG@sdDQ+3bRt~U$C}mE#^Ogxb#vmjK<_imD
zKG?jgZx8diUs%7+uIE=4n0UyyC+yA!tzL2HklieW?Hd`QPAuPM8SRgjd)MLe`7Rke
z%thj?LqL=2@Z=YIT-1EiITf)bBjq%iqbJ;dV3}G2Be+SdPNfjbI%#YuP2Wk@uueP9
z-1eNRgTpp~q^|hn#h)7~bF;};+ye{U9RV28b@2KeQ)5*>+naU*RtDnTV+@pm;dQYb
z1Y?DUGnOJo?G8`}}t(c-?WpY1&N@^*p3Z(J)S67i*rA%ix4vH?#
zFbtSQfz%x)iO6L)`Ljf%hBm9pP6u$;!~^kf@XEjE7Lz$yPp!tc7XB3o3eW$0%+C~-
z)0`JclJIYA{%@ptEPbPbvbF(XS!XJI>aq~-MH#>{ozRC-=yePtvtT5AZ6l!%m*6}M
zu)yx|ltvDx^Ao%N_H3^)4tY-;9+T;#Ypxw@_=S9Kh%Mkn@7L``z)SXzp{_s2G7$?$
zFQzv8@3aHL7}g6beW}UbpttSVpSho{kXnD`H)`GSE5QmP!ueLp&
zXUKRxb^NT+n1ebvo%ebrDl{&G`vjNW0CV~I-|snl3g$2*F_1$ShC$iV!;}!f&H=-q
zPN>a9$V|xek5sa=bNWZJ5;77h!7xbLn%Vuc*#2vY5Nh*q2#Ikrv9hv=GKq)*I0QKu
z0fK@67IsDvR#71~W!xk|T!&zJQa$My4W#T?GdE
z{J#6A7ry0bdi>Fadz|@>!TDD<7YkR*&o%z_wLeZo=a5F4>Sj+TiC2uUKkS!*3j`67
zE}8or5VvKdE7d6sA{qim6$07*6eA-AqP)kdX{DVt7pwvzNhU3&uGS3Q4lm@OCKA%GYYB*5S&Yygr%{7LId7!@P=
zY{L1I2IRg0SSI*!FKG`rKi%vc-8v8W%rHs(Fh--<5$XM9kWCs?;&G+)*u6z@MHyPS
zI*^Oxv|7zGL)M6HLdooe7y^MqZ3)X%adJ!cN
zrGW*W9wJc!QG!8?fA~o!q4PRQUJ#T2e=y2o1DjC%AN-iL^d$=?8>3KDQIt~jQtav^
zjv|hl3o!y_HGzBmBmO^_r4RdUe%ZE(-2BI9x;aLBz8a8GaGBQzlF_`}c)8i0(

delta 49778
zcmZs?Q*@wB)GQp^b~3ST+nU%;Cbsp&b|%Th$;7s8+t$RklmGqBS^w4f?)L6p-FLlq
zRdrQRK1q8Dek~sun4(OB09Z66#D&kDej4g16ngLubodT5Xv+>XSWF@|7YQ?ogNZd9
zKR+Ckyt(}^S4$FBHWr>VBQG#&ph`okL$D7fmZ#;vPYEkXr#h$TeB1M2bSHW?BA>B>n
zMwD(*?<;}~NT{hV2oEv{@~Xf19Ht0n`4L{gQnXLB&pYX$91AHd-wSf~Zoc1=hei$4
zMeIp^dNu_Q)WzroyUNc4S}qp*In9RFC>)iD%Xc)I9w~?`ce`%F67)>>+0mY2Tl)O>
zCNpin>*Tdh8>$O`m(s_V_m_yVf+=#pYU8I*T0fZ{^=N2H#$}8ZnU*u%Y)+wNy~c_
z>zf@`_o1O;N3A}MXLT-gdnzQoQ8>X`l=5*ogVe#=G!IW&^=197>LlC|ypoUTmMXyV
zl6prSTE8$vE0Zh+N`l9s&%bvPWSSQu80T@W9`U?*u~>EF$&yd1h-%oaV_VMnrtl{E
zJ8C=D&FtpMx%o=x-aB4~cOcuqQEnPc9&IB6rjn5z#gAp^Ids#&1*Yy#nusky+U=~=
z;yrURt@|?V(MM8#6n6bqqBuBSE?mrL_
zV!b@4;u15O#Kg7gd|bD;<2SJ}5bP|+Y&e-9co7DpKC9IMbbQ6?Sad_Orr{=o6Qk|i
zd5(`KNNjNcnI2RWR)5pA<({Z$zh$pF9-h?qHpYRk7sp}&p|w~;?D2LmHiLR4-160x
zu?+IHA2zFW4kk2WKsiV52fD}@3j>+^8>kV*lEF{G<3K~nBWK);A(V#JU9m=ljbFM;
z(><|lxnQqJV+>Lko<;1=?v3+{--0R+`xlTzpeBL1|1#Bj)S&RK3
z6@{KuZ&vCB
z!nrwM51HVjhQ!>|>U0fVE9VCk$ty+Kup%M0W&+V6tGgJUpV16g00a
zndT!}Qd!x_6VJ$h(gWu7KP4wHu!{OQCKb504a7?-r}m)#b=3sR(R2NnJvJ9N1CgeH
zhmzL2K*61b{0v;qm=?hJG3TUF`xEDFf@6lY9WTol(FvXo>U%z7oI)>eoiG7mKdkK)
z=sJx!vlILiG?QHt#gu;m7VXG=FrYP4-&&JdGOJ+CwVYyQNho`lPxiiv$gbTFP`_((
zKWsQ)^bsb}i)S5I!f2TPR5@iL%~ANcR<-%*c?I_{kqDJzqb0OrBmn!hP5PgiZViYjT
zwVBA1v|^0>;%G~hOg_M?xL($xMrU;WC@le`Z|n(*t;t0tfB8rL#jcT*cDmP(4=@m|
z(Ke&$ARoXm2p}}ydBWaU^LJ)*v4K@4!*ZrYLq)uLsXi?H4KkoJGH;uVUabrdKnC||
zd=^!FgG)QDwB!}gqidW`#0V*?fd|v+9-?Y)Wm_?-su=jFWq(`bGy+7aoAeq&N7`(
z0JpI>6==4ATIf;p9#_xXu;fGxy#4yU*qWHRC8asuq4^am3x1}))^=q;dV=MaTc?ae
z|JObX#Lf?_$vkws{F1oaB_BOY{K%d51UOb~$i6Ci(4rdRtb)$=%ap3CP@#iXQ@*6K
z(i>+LPC4Y7e=KWr4T;G{yh7y}WpEXK3bAl0(f$$7QjKQAylZ)$bHTUPZAIkqlf|1`>2yQ}iK{*F<
zbfZN}q`Na@$eF<~!`Oij;K*5GhldN9tK`N}&4dgt&&^;%Osj^a6jGRtoxSOlV9A|#
zd+@_%%sT|_e?W>`xK=0+0g%N;a^9Hx(EOYBL1xNczQj@aOnlR%Bu$RuDH|!zgZr
zKhN1bN*
z@mn41knlx1(+r+3dax3oSwD}NV%|ZJ35fg_K@w*XOr5#qoHZ(dT>n5ZX9@~z)CCZ{la-Y%<$Au0q^)?aNZ)H>`|fv<
zifk39^kVPln?}{)K`)JqdiMAx84u*gtCo{v|C@KdZpbPdUKb7Gn=8EdX_MdHEPo_3{~NgnUR)2aj&`1jMr#AG2m
zoM^>=?VSIPLseg0A4;?9W}T?S)8A;^>&8u0L4iq+@xVahel1Rj3`HkC=pw0=e*SXH
z4GX(yze6eb`SxqNnD_1RVG()Z@E;{I7r-=sg0-5?$~(`IdmH~P3P7h@QGQ_kJXDnT
z+}zpY|1QnkK~GjE;r-6>UUOTwc{zkep_C^yM=pmWx28&krPf#Cx7iscN5v6K(T)5a
zD@V5cGE{J8%yy9xhol4o`9w8S_(db{mE-i7=4VZY0$O&pJM!4SJ`h@NVU~S>2Pmgk
zT1g)LnCt~xfH=y;TTY%;yK1Tq1j!$n{gRqc7|$bu)}
zS!&$Wm)01vsd|f^m)P#;!u0XB0Dv~p>CTv)66t%}66x~5Zc3WT)yFQ+cg7@Vnd9?Y
z+B{UY=TYqb^P2QVtC^qUZY#s3icz6#+(s<%!V^QkR(X9BeF15l;8x;0qH&TW$9PU1
zvvtS|Tx@_gD@oFUDpvPop%g0MRtTjPVud8&irt*$29#589D_ae
zSK{~I0@%%dz}UOq@uR&hI*=Nt;w^;*kum?__vYSb49eKRRc{6>vZ^Ep15etv;ycj8
zts97y7bsY&Q4TAld$59cmX|M@*y{z%pKnTjQlD70x?(b3)me#LI}(ZD$mS_Lo_6M!7Ksf2Lf7LQ98nr~blimN}}Mi=QepR&go+TyapW^#yZ8
z2v@5LChLu9cK-L{$7;)N4y6}D*+quqDG`RA`2*KD?IA~~n*D?PK7tyf_`(pq4)GJH
z5X7YJ7ta7!3`<-#J%I1Kg#36^T9ym*9dOkCMY{SL#9t6(e^qYPFb$=o1235F5G
z_OjfKM{@WRtRX7vi_`|z+i1))%!n|(*;OFNjo+!)7S7*wKX+FI$Mc_fi@dX3jlfNy
z3`Thcq}Yum?X_zi{dIE?7G<-}q}#W*KDt)K9g&V_Uv+__m-w7p!HvKvTOrdyVp`qSrK48}T)x~O
z;gODFF@_M0600s1z=smgUg*gWDu@A7qJr1czu$FECEQ!xO>x}qr!$Yu{B+TUSZ|C0
zrP5$EbJ03p0cxjL#r8TLlwV$@i}-4C1ki^J_|Ueau?Y*O`B>aI`|Y&cwy5l1d0RZT
z<}D~ZQ?)0aGjzdf*r(@RJ){(rh7seqiS7*_#m12MX&9q|u!APDbWcOSx|l#oi}Zqa
zh_v
zWMC5ksxHS^5K*oZDU80uioOr{wd@HSQG?Qp`uGyhC58H@Y6wueLI~ZEdia`cEA+mu
zTeY&t1wZaFns?a*V>VqlOE9svSRZ*F^toNE)-%T9gw%BPOIec@-F72EY#m2tPtt7)
ze%(hcjXtxkVc4uIEj+b3OjneMeEq
zJgIky&U!{8QD?&?9E~=1t7pN={qfIT(+8U;SHj({M^a?uvp~+Oo+g#)v9p=vi!6nJ
zq##t(5F*zgx!o%T1p_<+=M4|*fc<(DGC@T0GMZ}Ki&N1ewggG0^mD&DOo+rfo9<)H
zK?4KHa|GV%%VKSN??stBk9%kIL=?9C5R}1(eXA0rwKOWG0$T`VRa61pU
z_D?}LVQ-dh6IYKH=6m5O7Zfuta$e1Vo0Twh@{mk5Bq1}dR&6A4JU>D3b>E90*~#yZ
zUqdxM12D$%W1th4-ZSIRyNi$5z3hd!JeyQEXcyX!rZSDpSgFLP4Iyc6(AETfSn
zn!>r9Pf&sdKMqK$2F=gvlWUd+vHUW!rj^IQ%Yw`Zi<5ZES||sXJ>nvaI*r)b?(g@@ggMFn60Vhh(*_8%{h+c%j5oV6XJT!w}B87YzkXcaCXX&x|&nI
zx$=Zdj^8&c%ahOvy_mD)LRpTB-ro0{jm~{;oN5(G*s^xO>T&NlhZ+L|=om;OG~7g>
zm8R8GsA7qHsJVTU{7LF+sdf{HbWEBL63Y)CR#ly@riiRJ@cg+UE8!gBZI^45WabU4
zl!rDi$nwN-`FvK2B9VWoe7E5obXTqzyOyI1L?jWIa0`4^@g1S}G-jX|O3GU1?q
zx0LC3hUSqfc-A!178CvV!DeU$i1)TZxqlw>C#1O
zhTb;-6bXQXfHSu@b8&SxH@5qq>VK3r7PkLQb7M^l_#X_JqNQ82!HMSopW?PLE<71g
zSSlAN$gWU73Kk6(imjh_3T`*ojH0h6Gd*@%bB*weT24@q>NMKxHX&<*i)*OjfY#vi
zU48Oka61(4Ryq0
z?VR2dHTJBs40DSgbFukP%SNm#m0>m+`E~`O`|E*A$&AJ6yS}=DIaV3F_ksZ{A~)!-
z33$|KQh*dPMyKvfWDi?~o0=+~mZ+yv0dK9l-ZG+Vg|sUS2s)9@heOFe9JXWfAQG5A
zo1ZQk$`)c0XV$=97ml2Rl9V8R-DT3#TwLuhz_&KJV{o%^_*Ah8V
zsQ`2zP&`~3j27wqoD{~0Bbu-YdmQkV=HhQ*^}Lv0jVX!E(&|Q#i__Tr$R{_J?J1;;
z;-m__Xaw_fAf3x!K@diPk@}zmF-f&M7m__AHioW+qkBLYo|y(^jo#p
zbWehd_F3Xo*Q9Y+J{xFK(BpfW`o;mmPR<-*SNd!BpL{cSZp{&Zb0}e95DL;s`)JV73f?$H%EO6lX
zjDi90vMxswCKH7bZ(HU|gxx1ElK~2RRef)>fZbsHqE!~IbQ!$BunXKGN99#C1{=Y+
zXK*O>-}%E0bsx;G4}@b9MlX(70vUUG4_W9mG4fzT@i_T+((^Y1+4w0~ug`K%yZYvR
z_nF8#(K))934y>L%@x00yGIbR>t)^UF!=3zrxsV#f|A93K!;?P<>A^{J||
z61CqUY7mTT8|8;M;JYma=E@{U@y%6Ve{Ve8_OJS(Kq{v??IJtAcNYF&Nh;ylT&!vu
z4n@;~c02$hNZz!Hd
zg0E0$U{Tjz;6fGXH?uECYYfEH8fDb1*$1Fx&PBr9Oa}xlGq8mUj3)Z@4%=&ex-dBB
z>oRB=7cf{Zj#)GuX-f5~%BeIVeQCE>Akdw#cYS12J|4xRzoeiPJ<-rcnA`7#eYVUi
zD>%(&5(|`2eM4Q6whr!G0T%_+9u34VC##d(=6i=!dt6NL6&XtTVLV)r28g8uyOR_I
zgGBl_j`{#3vuTuMP*ck08?szw#dTV^e3DLb9XLnjZ(9_(GK_iM%clc)lApZTtl9>>
z%Z9DJtk!-}>dx0H%0cqrbFYVbr7AG%(H>$?J8O7yoRCrdU=SnQz>3l~yrax5_%F*+
zY`wgQPy@JU#A;ViDE8Ak7|aEIp-vu>
z&>zRg(9=ic&`a|G_X9Ipu(`|`O#r1yWODh
zrYG-EY^?G$nZz)007vC*VOX|b7O2$0XeMN=LyvPaLZ4Cj&e0`sj7iI@1MH0pukUwx
zIs9)giVOdQ0Zg@v-gNV>;Fec!XzAzObh6wQ_QosY)BVXT#ypVm)*Ef{U@wM5SgYAg
z4o?^N$D>bG3}p$_n3HJLIJ}h!nh0gwt)E!*mxrEr_y+kdz#YD&y(=yGpa*0kglD(sm)fWjdqJcP^l`zz>LbGOhbY=ApIh=>s%c1wsY9D^
zK5B(l(EE$;0{Ui8>`r{RA>6s!Y4wLxtTOiKcKGU6G?DzoDDIgy#Scj>nY^e8he7g(
zz$CnPFIfp|V7UXiQG{EPBl(?P8MhL0tXsN40zX88U-Jfn-cZP8Uq5wIuh+$RC}1D%
zqWjtC@~C6+{_#DayX)Fsx9Mlz4mY@7(AcDcOOZUQO*Pc|OPftz?hr&=ML1k0PL>;n
z!fzI%A3NB_dcx3wdVRkVwFv|?KFyoAl;`@c#;`Vh0HGj*D?%)M`8<;L!>rhJcv5dZ
zx53&QxIm}q7@yvE3YMsb%;lfZr$;r^-b3&++QSQgZEXTn{8=CauJJElXloL9@$1?0
z?r&bY_L|FOdwW{l)cn6t3?{PgIg~F9RxS9o|GX6lbTnj^(!NcKcZ8~&VmAFssFTe3
z=Pk<>gIMVjJ=V#-v78Qo(ykK1#QRlh+r
z&gI~;_qrc4px9*~2kJ8M@FS`|Z2h{_Lcp3wEmQ(4S>;HSZo(sXVBl7449#3gjfEf$
zkPB?6p8OUaCWb;?a$Y{l+EW{$^pnh5FZ5rifj^Reip*Whqzr_NtaO|by=$)1%FyyJ
zsT_n`98L>JTz#lzLepVc`Rx>2@k`V=a_C12Z4h&2J3AIB00Ku-Zbd^-KY93^5D4rD
z`^^y_cnFPTJ+p+{!bpi6N>DA#Xq@4Cvc_q-d%AQJPSQby4hzY5G+%n*oLQ!KtBh;Wz;jl8%}C>Pee##*`QMY_ha;41!Lb
zBiB;Hy`Aa%_V?TdY5mmFwiJ`DVt|aWP#)>qsDxq}KiEx8?pr}1+OQ6)A0N44Mc+m6
zKd}9gy|sH~DlLM?Vf)9Pt2IY3S&@fHNP9nFUmZkO6_Jf21zmDAB!*1W$&1}j6$CdE
z1$7L+Bpw?&`tVahp)hfyFn}LYLM#GMVTjmvF#JR$aXS7?+r9;t0%K)Kt4;b3%;+oW
zaiRF_7_#*AuG}9Y(~OHH1>Grg!0{Kx-U*+U%cj&@YM1}pdG<8kyTWF{4}w^IA&UQX
zL=f*+^RrNVeDDRekI9KODI!-sWpK`g6`S^7jtnb>QIWE9rgV{-TxPC3YWKy0^~QI*=hjRXhlHU}5P(E;l$ViIn9<%H2yuW+e%q0G>uD
z2&%bD-mX`kq`6=2wO^|VKcJn2B^yMgvd0*tN}oZ%kfi`iM4J>aOOIa^or
z;OGph64=o!-lZkXB+d16hq1d*{Fz4pXfGOlN!6n3lsrXKZ%z)+e@*pXC?;@^WYE#suGG`G|GGjWqncEOP>S2mNMgK@Alyj1BM#qr@>$
z>bHYfpDpI4=1MObi^gWpnswyo%bvRl6Qs1wer6}|Gw}VJ#&qk+3S{|rWX<1PU0k1<
zc?xuB$unVT!nA`@4ZNQ&U3mKocJ(*wAbs$Dp5L!ktZBZjqTB>KrzoLM=9-fJYYW-R
z=r+6=xCR9t9yK38sPXR*CZVk%r6fniHjCtOgE&E3Qlk#as%!Vihtv^|eVjBzvE
zF5_IZykGkgXc*^!KBlS)v*E?pl)RAoRx}?JF|QbpFrCy$K?Q5#%e?%(HGdVDGUuln
zTOD3DSEU`c_#Ee#Wo1tLL^MMQ~rtNSEK*q)nS;M+7eIpH8295k&se;;6
zBu;{(n%d{eR7j7Z-ad-q-q>Vfo}t<8WFRe1U#kK|w`#_(YE^Er+qtL2xl+A=FA15#
z@%%3Fosqc69&HtP9X&*axa!|eRVIBl=72g_4;^|pjkxezg+M)IELBJDUk;yYy)Az*
z-U@mJ8b&?I=Qwb8#@>Zrm#PNJhX?*?z9UzRf8K3G4#}xGo%gr*VLmffF5zs-5--b1
zm!Z>9n4R8>(3v!JuVUy_Nxaosn$G<~J#TFmEYjk!--_M9t`7dUF4+Ef<_7*=Yvd99
zj4`=oq}W&g5J>19*@T`c@K1}k?@FYI4Yyhu`0GWiCA;nI<|j%M7k})dO+?Ext(#;j
zr0l%;E#U;-!JEndTkoycn%ZK^Xb*+KBZ%w(u?wAELJad9f)42{D@O&lV~P2z`k{Aa
zzGL65)Z0u~z`4}>AL?1+Lqre-UyqE9^gO%;qRA%oGA(zf4HY-Mtnn2@M;=D;FP?o#
zp^uA
z-6_&U`>i&vkw8W@uY#f(aUtXBvXLz|6UgMOjn!Q9bdLU&pJ>=%w+!x%k=2%7&<47bs^8^2)i(6|d_i1>=j7%dGf)MaZGSRP8WyyB;)#%}J4pRECY`a~
zY~5^USE?f(HkYf~DUI9lKgou>4s&B-e>|Ns>qsUp3msuvB$_>b{_ayF5rH-%qY4cuB;@Xz)Q<2ax?sEA^V0H5B
zq=qpkHMl8(SQt@*bJZIPkxlT*nYcKZGzp^XH$#nozrbD~V}mZw?@GU~PM|DK5ziPZ
zotdzZJ%yhvj9yeT-KdCw^5M;~zNYvi1z=_TOh(!LLp>cXco9QK^rs4uMSe*BKJH%N
zC2#8Jl~;4EqR$Rfgn1eBZ6v!UC9@h9=^~~;4U&_#WpHpn7W_V+e7VE=@~nk=a;Ex1
zfWwvceLjWA#;`V*zk}~W*iy5UhizG08b!LwRHr+)1q{sA3N`*k95Xvvyta^s1gNo`
zRKnI^QJ&j3%Z5LKGQzg}6BWMqiU~!M5-bmue!y*N;VGPu-7&SF)D>@rK(q#5~
ziQt?hVqY9X7z*F1zhtk*GVnkHfT^++1D<=&r)PHbq-razN6upSc$#DqB0JJB%J@dR
z3tu4RMFXvX=3Yy=T_1N&4oe9FyF7Gzp6;1=-xv+<+i!gnqE!)*khxl_fGJR)Jy$t<
zXXb=fL!=j?L?RaF!f(x;Exk`?4jgkbkiNbj;LXf2Od1;%@j?qjc>dxZz>nSs!L@2#
zW?AL@=&AjIB_UrumgRC|%07@HoR)n?*?pM9(Cf6%?Zt%Hw_HkDx|@x*qG0cv)3JKd_hIw=g2
zEQRbNEaAP`4a{CXlqe%eN$o&nzou)w|32~MG2~rfe<_cTGKdsFv{)l}*q=(pc%|vd
zc#!M$DLU-I8w&*emg-@;*uC_!i-#Yu+IV9>``PRtrSIszx~_5fw$@ZYJU|J*Ugx+P
zK9ucI+d#mEtufkjc|s3`Z-K4dLd*3vBA9R(0!~2|
zN>Vt>iI5T0Y_ucMIYxFmv1Wrsyv>ni(F^JDOcfpz&b4QsGvI#446sx
zX^b%ujg7vC6MNIi`)Y-L^fu6Y
zM}lFQp^GwI0*YV(CiGn$Z3Hk+T;BB^TxnR~!o6@r$Qe`0eRa-1x*ah^BA4lj*wY{r
zMN!3GT~qG06V%?o_YZ>3CQjqY(ovGB2a+ba>4@en|E`6oq2Wj7#j6G2qgGEKcd2?F
zAO1t(byeqpJmC!Wf%}9LF4(O{1o6LmdU+1<_ObkD0
z7uDg$YX453DUg>Zjerc&5t@Nq_%V}-G#G9jle*oG5>y0^4nt9pcV=lSGS8jWF#^b-+%p;?MyE1;MfY}`o#dXXBq(4!Z=Hl65q6c->3TfDt1>tC+H$3q4!TdlT
z(m0h=vm~ljcWFuRQjR=Tqr!g12XRqD!@v(I0H$-k4IZ_aI(Xt`jO5n((ctx3C-2rB
zEilW=o7lmE#1j4FPsM4!I-GUwa>9%BJM-EMz47KD=KDCP1@#h)EpY#)Z4du7zpkOz
zJMVT2sVT1>z_Ji4z0ZrV(E(|%{8Bo?_=hBjfU<}X5IQ>m@)R5P
z4LpiQrNYl9!=~3bGVHOzvz-SD7Vy0NeT=Cqi?)>C;vv#?h^}n<;ne0-#fK3)fHl)E
z6;$;O3g!+}%PSzK;{9#Rk@T$CyOSMv^nwfw?8+M7I>b<;It6U4kTRR-)_ZZ`myVq~
zcGP|_!~dWiHoj`u`Q4Q3)atARs$f9f3()NKS5%cpj?47-dGcZm#HaQU#9Ts5cGkRu
z7$p3HPE)Tfb664FW~5XJ;w8+*AOO7r3Ap(&FnvnDh_t0l5cKcoV*0Nia)L5r*8L|&
z_B}AmZ7h^@1t);H0?{An^@35-Rl>|N^DA)w%+(?dI+
zZ4TO5^t2{*mOl#V_w>laClP9*w~Tn&TfO^S=J9uPt#M%%=Y22s7{nwLrW)LN6
zTYPxa6{(BJ%GIj#5pTa*he5sXZNG~79o8`D2vtEFvC>BdO;-kl`d2*p{U!VBY&4xk9Kk0r8B!ZXuYWAgLUn79GPjrPJdt(^_yyj>_Vzb4
zCYbxUk4D@-Fh=G6BNA>)-q36`@;GW8DRN@uTx5UzyllDd0ev3jsTgQ{TQukO#%LR<
z&pQ{{nvSIX=@%H87}oE^-I
zkrb4&iDDrjYuz&q-}HMF)!PaI>)lmE=SRCmmzEP$T1pX)1!@7<
zNXQ7{i>Do#tQ7z#bPxFI|H9iidH!GC#>&Oa`G0xap^mo0hBTW0bImMSXsq`tS2{^=
z475zb5X7G?&p?nd{#i5zR8)%4!hKKwPYpK?nj~u(_ajeSoZ3+fH;-&Xb>dqyo=7T(
zYLjmtKbOY=n?9p$t)hqAq
zW%b2?o#AVEivm*B&`R0AM29!8>?OBbnpA5f!)AQz&!+x-PxhH}@3)Pqp1`8?h;(`G
zsW_ztq{khc@K!{7jS92KC8KHr
z>x1+%2u@4WKUA{BwD!ytvnFMz*qX69zV4`XEry1L=9Tn+5Ajx<6pbC1_AV8_)X(@W
zr1}H>$a$v670kx|*8RU;Dzho-c6@bBW;RTceFTVovNd4kD5&z!E^TwFR$fhbRVm}H
zN{qVs-(wt@@m?gb3BA-+&znitX$>_16hd9=crp}B-U5i1R8$u6J*=f51@rC7Q4y?%
zTbo40$oulqG3+gI4-w6|`P#*^!)$ywC!z`aT5;6Qn`<{y_gLLn>Ry?Eeyl-Xz%s(^X
zzO(Xr^{bvN4h=l+pE2EpEsjuNiyXXWXdDvF#l`Jo9vq37_|`hXTzNCbnOu0(2#_+K
zuJJ!4T{pU@GJNB|jQ!^TgkoEeH;l0dx09$2%Z
z_6CLCkys}yrA>-Mh@M_C&R;h9BTYnOj@mkymGIc?hvxVRijwE#&Jdrta}W&(7+I&_myzlOK=DPc;H~`sbSDZ
z8G&q0H$~~@G%%+pQN6-J6#RD3-nfh9&KE(p37&*&Ovmapn7@Vvr_4am|$ULs$?|~By8H6yPggOgqvWWw2)o9DI~_36qA}F
z-JXrpqGTb+kwM#JuvTvL;nT_$E|K}nrSD;r#KwB{IOufAYT>yJw6IZw&}o2aaS`)5(#5cvQ&8UoSrRBEu~e6=_vkLOrZK{kK=4$;P=6h
zL%&N#&yWrQzQZ?<1~zOw;(hmZCghL0nN~z7Q=Gc?2U+E`cP@a=ijv_owvT|naT*8r8*bZl?)+iJR!%g1MccPZFnpA9lKuQE3`T1a?M`PM=dh@lp2QL=XY
znHNY)Z*y&Hh}(xePWXOYk|P{S-_nwH2aGkobAOyZ$DzL+-;4{gh&>`WaYHQNz~zY4
z>zcMi7vv6n7@!LTa6uV&q=!^n#X>gW>df+KT8L%D+XK6o#l1^vx9>R5&RPz?8@rS3
zY0@D2!o>PD$78m75wej~BzhuXym+iSrCf(iA68An41THLiZVT;Lr+pe9Au-%HA2u1
zR^j#9Z^}k>F-j<>)s)owuWI}#HnrGCa-MKZI0pyNbF^pP?t3G^9%fZ~;EU9RiKld<
zer10f*Au|O944f=Qxl&F+8i7U^U(^kpJ>BT7dpj?jWupw&YAyy*noxe{guo#7Owwy
zSV>Or-+r99Zh#RP>SU%yAZ;ty0?6L2<+>jT7YUJ+v}O*|U?F5|&@dEP4?>>lh^z?-
zM=URp6(HMB;1ENcpT*Mmuwx3XS^_CW97|LQgJdpH2s5f5EJ#HQ@fi*mB3M9bP9S^0
zQt+|)22RU@bbgRu4Io5fHHepv)c@@}RkEU=glSCT%-U!vs6HOV?g_i^HfYtT37~Uy?X!lpoNs)^gT===AgC1bF~v
zUB}u^KLnMSwM>UYak%fDERkNkSr~Iw5aBu9Y`0!%&abSL-1g{-a}wJ{X*(GkZ-%dUa+fN!k>B_
z`dqVAaz%!JCya2ns&Y|SxULG1Ed>CTMg8B9TzVm)zaAS29vMkI`_U-A=hbC)8w0^6
z+U=aBfV+r3q&S@ICaLGH5;nYx3PWr0N;IR=;vqChk`MnBemf
z9&R8c_`HwT$@dQzdKglj1)4QyL9TQ0N{{^9Ru4*=q~me*jlUr~osgnN8$5t;^U}R`
z<$jreZaIp~y_yM5_SM~6;h>O`_EkR$IqY}SPpq13&0|IU?W4vhLOA&zo`fdLnJikc
zv-cUW6lEHAr5a-_V#1~&a^tg$^4{2jVM&bdU;UoOs@#;2La4KAd==*3!q=O$@1beC
zXW;RZOZ~qHk>*aVt(3ywd^Xf%q2jG%^?(A#z-b2p|q~^U1PJ_gtb#3V`D$;B@ES-KkwelvR{brKk*
zt*l>pfQtLQu&`Gyvwl2=TE!}4-KFmVeLzz_ycOq_$5s0ALBN4$m*bHYz%hQqK@vbP
z{K?|H*YdzRzpC$%57{vKgEw&R2w<@
z{{&LOn7PvCq#-d{oNyt+h{2e7{-;B4LEwRa{{_nOe;QO)c4p@PKcLD=@*m1gi)Vty
zhTvxBU`{jk0;6j2&x7y+hmuecl~fN*`z(Mk2f$8-zDX!Sx(0~{5%Ru2pzAz4J_?5-Osec
zLI?=}+o6BEn))3W{t{xLx9zDSf!^)04-#r~Xutdlav;6@4sy^2g0y!m3FR*XdV9dO
z%N;-9yVdvpyRM@DwFc|I@($Zqclwd(MI&@W=%!~T@8ddrM#zJ1AXn@aP&x!^{{|C|
zf|XiZJUqqzt3ko_gXUwyRtRf}PQ0D^0-oNUf%R_*;kRSOZ#rFRn1B0<6L;<;mg(Hw
zpfQyW;em;X>09?b&WPWS0UyUI{v85ev4BuEh;SbKEQDP#Wf}ci@uV^>_2zPtaO@P5$3#AJjLlCdHl}glWXfv#C_~v
zQFm;KS{0366wu>sMya=a?b#IN6JV^0_*eaA8+_p_J`bgipZ0wc<+xp&B$KEzmYLN^
zW^NyJNjPZzLs8L%1~bu;@lQ^AfFUbtjVwYzb+yUcJruuwOq9;6QaLm00Y0p(b~JpI
zb=>wQ^2DPEp7uOe{Y8r7#oW795`Wk9f4s3q*!q!>*@WnOEw)x^*5CPu9^9v#;&F7j
zi`>$P4hMDXS7-?>JU+%>&Q_%z*aIO?
zNKMOx7kTC6a{*}U<6K)g*IE^%(^<5bmjz$%
z_C%ywM4=_F@*879vWWLHdCq>$`M!G2fBbS~t|+qJXj8sn+nd05fdOajeZ+m%#2=%Q
zL@bQ_gdeP>nXL;fGP`2M!yRm_cGrJDJReELd98gt}>Q4J6E`-b#nBG$`=dQ
zLkf4AF(FYZ#o!i-Y2WjkLHr|wC2@xfB!q@JlmG;rv|IY<
z`xbSk=Cx8XL{l|HBD*{?eE=hb{$M`W$(?}{_n&#w?1Y~-+<{M+3M3h$o}
zd9maGIik4SZ(k-*Ag4ks1i^>IM(Vp<$Y>c_NNf$Kxa>)Kz$$Gby(B6Fv7Nq^u#vlV
zi}H3SQqIezDFHkYmLEdns8K#EuVD1@r9j10U
zo|q)uV@kSXJ>&`0^yjP;@MN{zj<*O33E@`j0wQIYs55ipdE}3o
zk6%@6)n&yoD=txqvmvd{|Awk>(fV=Ar7Z5o_eh@O4liw|RO}BKx9SKc3B_mg^t)uU
z_pnQzMgP47R|QL?B-eORUGvzGoLg|Sc}84VwAZ-6YyR+@LZx3lsXl=7^mJU)uJ3L<
zVP~yF{ul1J3mDmH+Uhr{)eCFDVz5K3M64N+HIa{!DwW&~q`SMxDfh*9F`l`&Jq3B@
zcb?aDS!!ONGpyAbI6vp+W1y(g@4ddx(0cjA0kF{3JeFcyBAztORakD%?-c#_9_q>-q~2*f6d>-YQ4c`Q7<
z(DN*ySUbmakHSiSM_ud&yo*nRcl~%|=ed&|vEgpZrWwVE%&2~=%}|v0u?x!LmSYUi
zqOt%YKIJR#1vw%K=vF0?Mf&8!j)iSpS(uiF>rQVR*e*{Sgq^%MuNu0L9%2#LEUqr^
zTB-Q%;%~3VoJ!gm$^Bv9^b-h`@X
zdm>I-XdBy|UE(Zk9yaM}wqm%QEOVQC9+b73mwt}ie$T~xiW7%Yq6xqdY9(AITiy;G
zpNiWG!yZq6Z#H^57Vljg2*ncI$RT-bPa;_4cl7{2@`rX|bWo5W{2uZ0T&uw*TEXc$
zeO-#x|7o~RGcb|I>XeM?RQr(Fn$vF@7celUh)
z02@qAL@mgqXNjV!fTa0o-Tjo1TU%ZG##Ra8z48NpyXS+c&kbpCggGy^&%hLvvgD{X
zO!*N30=coA)uZ)ma0HyGo+xm#!}0m+2FU~4l+pKg()&ArjPpW3$
zPZ^#bkV)Wv3?0C-;NI-&s9jM+*(Lt~#KzLYfZHpydL*JwjM#jv#6I^Mf5$xy{1jB&kS&=xqq%DGO(+zcTu0lPky725Oaz
zX_r9%I0-VwBF}jr@r250;T5_&hkmqN$JaA|tGelZbk{L1@OFqkjh)P`_FM|#Aca{Hz+zT
zO1bo62{>PVijw`phW}WHwxU)_S7nGJQMMg#91g-etgLucK>Nbm;Y*QA{UXCy-ZYi+{Gq2c;I+tWy;P5~20s
zJq32{U|^05(tJa112G@_r5d)3gwR@;hf&xw^IX_Dn&j(iyo+
z?6Ob(quDGa8W*)vK3K?g$%}_2UpN9&eWOu?xawo^503iGm(?CWcDo6V%kxx!(Rmf#
z+MnvOTy{;v>!o7Qtxe>b+O_h8`%qzEE%ra>iHpWuW~lO2f4TXRf&jr-yVjA1E04@?
z$V7>#K!edAr+_Mp{G|wkGp!BBiXF1GW%KneQ)MEZ?-nZ5A9kQ;8sYI7d0jUTNg&m!
z(uW9AyE8pus8UMy7i)o_OtLP2$Q2pUMzo26%p372?EtXJ6oW=}i1KmAJ%FhVXTD@C
zhMxM>KeyLLgY2orJ5qFnezBCUB$5N?U9+?-qk}EUPdGR`EaUsdLZ_@dcma5V<)?DP
zaSRGRyAu)jin%tmYQ-Gfc^l45#z$`6Kv3%~iDm^M0>_pm>&F)(l%EBE&l0emf#tM%
z4~+OjX)$131?lI~!8VGRxay|VH$NA2X^W1K_8?}RZx0ZTHh#+c;*J_wg>NNTx-=#g
zfFamdFtw#L(ODaIM7TP;sGlHtSBl?c4+r}~-zXW(fO{TcY@3&N?z5yv>izgBoJlLc
z(zIx437A4P{uRS%GvI%JZl&gJ=+8NIkSgJFO1_KnGC2)8+e-wmZBv^P>vVn?Ae_@s
ziXMy+`WYboeUFv;UB1L@n;Z1P2@=>|O2CQo#<$?-K)@Y430XzCzTr#Ex
z7iYNvOGo~~FEM;2@Tn`F;9As?d8QeTr5yFdM8>`1jdYF*Yf!d-xE%hRQYAn9yY%yY
z)U(C7v98t+inzrN(V8--IQF0VO$r*38-QOjz0l=1;F~7N6WE=&hIP>}?zhf8=dAXOuH%#;8`2>>6)nG{_WDEid(Dh{4!c6bknuncr>74lbnqty
zv&o-c!L=W);_wXw0gc!uyI~#LKg_oi1&AfTe7$n1A0u;r!Y3Gu0riuXARZKLfcY>W
z!9h&}l~VlXjM05%ksnEy6k$G4w}_xD>P5+2)Nb1PeXy6BhDec@gOGZW2JGjdk-0Fd
zO+?7$hvPJ22|n1gBEm_vAxnE{bZYIS8#*3jA(m0Av^2(ujQf6YSr`Uy^Vq8Vl}TPS
zif>_B&~yWT**+uTiT-KWuW?i_8^Z)T*~1f|3GK4wuBa*3ahpZ6nTVVB)TwjK$0OVFyH2M!+gnlZ5yc7gwF5Crx-m$QPbI?c8#|9WJ|K~nkSE&R8n)|@ZaT}3Cxgf3$y)MOo
zQtl*z_%lm2M`|B`bwR{H>BkFSp)>D|wI`4RbpB3a(ta$xjMb!w2p2AdNg59
zgBzz<6e#3ITzvJgib5f!1kAnOJ$@BEBmXRa8v0sN$~<4i0|pV@S7Y4KVJD9u$aI9&
z#Xt@2dp)fr`MHD1GkP0)uuI1ex&50bAg~cjgJy*(?|y6O@^NQm@a4A2koTU<{21^*
z2M7_uHq*>^HG0%NDwGyT_A3A%&-|@aAK)}H#GFAA*daR_I%|~5-
zK7vAtA%T)#AeY$GRjL!a*1sHnx7eJiYq048j}l?=So%Q6RrYPN(jZWU*Jow@-~fa-
zr_Ql61ic~_xUWVO{_rjy{$bBiAmV9Y&m$hlxPF^xHZiL_T3Rr2!NOaSx&xtS*4Qei
z5x!ygZX*-<)gzgV$zjQK%n`>9N*zIeCn2Wae}Ao}yi|?;jBfY)>ANxV+5$XEhnUX{
zM3SVJ0`4p*iahxZt=Kccd-V8I74PwBy%X8#6c4yX*;@px*&__W7G1(bp_$TKvdG{T
zj9|FT(Ma0$?a3a8_p*1Zw3Z+tFFH&ACW<
z65VIM@U3J17*2lY;eiqnZt@F9s%OwuY8e`|>N^fovA0(nr%BcpUojKf(f@fhRBUhz
zeNPi$Ix4^-Ba!|XGj%u{gF|~Ew5I25V>qFNmWU+w>x;+4LV|Y{ga1^2A{^@B36J62
z#546k4Eb?vp=faV
z0LLkri9fBHomgVrM?R>$QQ*x7b)q6sz`)qo;#;9w479A;{CdVKpzPlsVIV*M0b$7n
z-Jx&?hxB4dx0Mm~%zcW3ofL{%GbgnfU-JUSUJ!AH5dC(OZx_s8+jZ`RJP`9JnUk$>
z#I!-;>UAQu4twZ-sx^CPGi3J|wJYUU0+DbaRtcq>BGP~
z9Xk3zoEz1_p-oT>iFd^ea$nRCio%_qMhC8;SM2#vI$|I@;+{6)UB}&hyGa%{hV4cU&ps9y&E_N=Qh}tmPMTXo#(?)>u76!|fA-=EEst$~{hsI`UF?PGv2I}68Oc(K
zmWof70}4fa7*XU>&=r)!@Npvvn9~VYjjnJI{#T55hJC;Ejy1Qd`;#_J#3qC)GGVAOvYhwys9<2TVMQUC}&EM
zQsMn(JGaTIu$-S2sn`-2d#}P&wz1iN;c*biko}Z@L%5g~g4|M&!oqDI-5k-*S(up9
zX{7rQm>o)%xoXOMz4Ns)C|I0xg76GU)`~d)F#U3d4dlP<^lmXPd&Viy0>EY{X=jTx
zxRj7;Op#G_W*8MtRDq8+|Cp83(>1DoDo|plj(oiw@vOtcmjdQ=
zAeb_sD@-xpbu$RU`*K=(G(E;4Uk3@I@LFj#d
zH38q~3eXj91cn_qzp{L6II^QUoh^~L>q1;UtB8s@M%!=EU8bU?Ev9Ig%@|y`3k~o7
z%q=90Pqj@-86354s;fiEKBOr>>0yXj))|m+&X}A_HZBs(f9VX_J7c=K)xn6=y>)8c~It=dCUkjyAIg_Be#r`r-ZhhSjq?2avl<<=Bs&2!SRV*Q
ziH1Ei3`j9belBWO9-b`!y-CFWjhr<9>SqU7cwX4ql%8rrzB#RrYw$7qTvX4x5xrxV
z(l{CLafii&iZM6+OSre-xsU39KjS8jW_ERf9W){RQPrh0Z&Qw~rOs3Z&o%lRM}BWfF7%GHG5dpGv1fZ6o|6(blkcJ$0IMV!DTxL9!1
zbDDH=1Bc{tf5xAD1Rp=R!7mGkf>Cdb)RU2A__*sVVz(pIC)~AigeXM@B%FIJFqJ*}
zxW@X6cgNlTi5b`Qp$(&dq$5r&VEIz-687{(X^eQ^Fn3Ef{N!wnGfki4?B;itX?*lAhjPreWWqndtk6jsfTud=+~rBs82^SEyv10
z0kh5EojmD1t^8rDME^&h_T1$wbzDmA^yba`+pxJ8N_HpmQ?6N2Yc7)x&2K4y+ZSWN
zkIX9K9#eHG_vj6e0os0(??6qBE42PkO3JCjGGyJp2|3rjqXygz)lRcP1pXJd@TBNx
z*_JZgI}6k2EhIXBFHcbi4K!Lc;_7+l;ccD9i8CB>HP)tI&%Ya5)r3VxhLuT|(wtTk
zf53qEakuAgki1dHoh6jq?YG3%-p0q9RK-I7Cb%jd(qSDkVYHsj%FwDVLF;CRLF!Yb
zng5kI{5Em}Vu7TC!!~|g3Q5W@3<6D+pQLkFdZP3>*xfvhU&?T9guiQ
zvMlzZQ9^O_Z6%@Nf4O%9V8;RD37zG_Yc4#Mk{iO*AdD+Uu0WlBe3c{9r{qLjT`&5!
znfQze*WmAeO}yfvtfFtVFQgY`IJGFwaveWOnO?~-ueM=fC}PCkt*dqHhLwkEB;(G{zIndF6lEd|DFjqDj+<@!iH{kpf7
z9*8Vw14YDCthfpqx5}%bcRcsZ+GXfB5N`cFWz5?mn2
zwdm@{ULtw0sMnQ@VjZzO?D0le%bD=WWkc2{Nel8l|NKTq%;}^7qiVV_Q2p0{4_4@B
zE>8h9ZTfv=;##Z)<~FuSg(t+FV&z(YwXJi2ZqE1m`lUPu)SdAyxX?_(>VSMbuQ&LyvZLGFCF($yAaG*W8qGc5g9?2L
z`bH_<4M}g=@8px>fK~SPIWfT|5AwJevTv}!S#LHnmHvA@qeqoFxjsS4@9PLnq;(i>
z{*~*3Uhod2pY|WlUFxIV9pL?cg!$8Sj;aZAw_W(4ztvK{HYl#LJO2>-4BbqCv$i_q
zZ|M@l>1sX?BFduohLt8Bv2CUlm7hIGw_pD;+I;2z9pPM0Hqp*-i)G)3DJj?9gY(!E
zJxbKY+xtAT1E1mLXYl;M0N&CEz8}XL%t~MPv6e%&KM2jX)ZK;Fm<<_!E2GG}YRsi?
zsuIh#XhB_@R1-MC@AOTV2%me;R|xX0D`^OSFr%#O2Fu`hdLD`N*f!6CRxxy*Uh5XH
z9?bjx-p4QK3x27qK
z0TJux^wU>M>GD~(829igKUd-o&FbhXf8V?e`G&eU)@OJ_LiF-~jF>z!ggYX9g&~k?
zi;NLB&KQdBdgG{3#Y!I;iCjBo5yN55_L=!tC44>XvYv#PIGUU4SN>4n;#0Ia$h~G5
z=H~n6IDAqi6NThg!|2)jnX&|0U)q6lCCxBu3F1~Z2F~M2(<#P*EAZU=N8rh5g&1c1
zUNLvp$7^tEg*Y&O{Cz}Ifm^hLzA?BLRSy)wqG+C2$!5j;`hlM4Yqk+-M57>aEm&64
zV^zw9DtS$^`^<8?V!KX-hDDIx?Wxn(1&@!=Etmj`i450M?n~4@xV)E%FsSPID#xv$
zIP_9_=IRtQD7edDy*2x&)NHDtbCS8A;e>dKHIYK7am=fK(`MVWM=TUzt+pSNKG$~`
zDa4?!$oS|O7cURiTwJi?_Q*?*GfUTcPN|;|?_pyGa2`6-ZDCUuD1sq9aq|ZmfAPGZ
z!ND(sPv+a!8S7ON#KU{qVPkQ{|6YgRF#n8`KSDShZv-$cH4-%eHjAp;8!ZKRW6@F~
z(P*B8VTj#-nI7cxW`Xd&N|GDA;G)N2ri@l(KJ^D79c(d)slPuZUGQmT4z!+5OZ~?+4S)WA82iE?e+McJm;k2Ll^3FXSBOX1MVu}>Np2S?2_I&@hH>06H%49Z@O@uC
z2@}6$4wC+Wn5^PMHBdcF87d0MhAq18o$|DZgPC1%S*w!
zj=;_am!^G4K&o1%iwj5Fa$_-LMY9VGnd(#?)BVBsP0MEgsR&$7Jcbqxe|xY9Tvsgs
zhZ)TSri_=h{P{c9)+6I^rXE9+%~>lGwqAN3$pf;MeJgftSoJI_{AU`dF`N#oN(7?b
z`dE~Iu05W~?>8KF+3~%->@(;#7aRu_wKF{~E=)^N2LWFTpIOLnUn=$(A}W17tSTmb
zg>6>-j3P!ov>`JPQvC5v*G8hS3Y#>1NdgP#W2T#KGcD$QPjRm^&gZZmSi_)3jxbYC|cStCX`u#P3
zpN2Ik?e1_~zUtRpfIMI3iUIE=4HK`8QM^p(F`aquw*L;?yYleLn#{5QaHXLI)Y#g7Wek!EM&=KK`>H~p
zaApZBr5fT25I^0MUKOvkyCSG#U{_*)W=7@QTGEGf8Y=gbK*dtSF}A+6+oc;@$7W2A
zLiXMo#TA&#{Rj)?Tq>8O>+2Pb!Fz6!Jn4fK+tvwn^UxtG)8q+bq`zg#UZWFxb{V9O
zm<)2bOfM5_-@}ZooUT4eguZ1!us~iIT&(l@Ac^AMMzm*!Y(`FHzN1{j<6KUEFA$_C
z98IKqYFHwB^g$wcQTe-v0yh%a7{UQp=t$DQ-tJtNn^11zvrPe_BV)#7KQY3sl)hq3
znb;DIVEwcjNFP~yI73a$z3Rs`?Gbv0$e<+sv$g|m0q%yK-SJH)WeOTcX2Y#bv+nY9
z5vpH3&-s~#*2nkWG~5*JRcdvArv)bO&|*dT<>lA%Q&f0@6nNeb$q9NCpndde@LnU2
z@m8$;3LOVuu=Ff5Y${LpB@MlbpKV=tejAvSh|I6~jyH70qK2uC60e~K-tEfa)TMNH
zN1T*3^FUBIN&}lItpv_gZzV3%N-$CYP*sr@Kj*wI&ll>uZ1Uk|WHJ+f%bHB+`S{DY#=
zqN;we{M>HP#JArO`dFBMC0L-#VvfdA`9L>yOUyMixNQ(N(T``PjkX7{jvQ&c1V5-Pd*R{o>W
z2(Q#%;^T(SPJgxfo`gotl|seFTa%TNMW17jIcOpE*^@O7L^$48+t(Z1`qIY76nW}8
zEBd*X^vqyda;`K8v(M)UDx@JlSH&-5Byijk=r-@6HQoe^h~Gwbd?x5im`S7cP}rb#
zFPV^6%5b@}w$sReswNVy6yw74AC9okT`(~%z4k)4HMlc=+mX}d
zl|+W&qP@VrV|DSQ(Ut|v&ce`qzPGK
zui4vl;~UuDp4p#GG7)V_M%k)+c*kc9+6o!l5-R(A(QDURzAumG
zT8XG(O^Azs;3qXLoPV#M;a%`Eb3M!QI+ZCxl2MxCNAlZT?;%t$EbFP;fnirYty4vG
zHx2p8F1>8s-9B*01D~raJq;rcX?lgkzY{0?kZBzLRAY<8nm#wQ$CBCJYbAk}8Z`u_S(^isBJN2b^u*iy-D;R>
z+Z5#a@c9n6u$JiI+w;9hT~q^=0cO;%@98=k7baPM
zOO9syL#O(bTJ4E~V1kd*O=o|-0~Rm4^Z77=T82Gilw(UiYN(B&bA;O(9hzV5<8FwR
zqigQr6WTRbuJ|hsX;c!pIL2ZIv>@dj1E(GkFss5em@`5{j$;d0qJ@{;I4O?Zu_tkg
z8iGovvSc}nc)K?ske36iBBFf`!}4~20gwJSNJdWbhlpG?>@oCpA@A?CJuQ?!yoX-g
zFvR@k<2g_E6%E=1e>hdHU$LF*)W?u6Jdq+rs&W;R48?HF~@*
zUKP%6e$7THsB9m>*N?vge+Q;9$>{7;ZGFoH=BnjdPb|azeFoEaUQtQN3zIE>68HX7
z6}&i{`|UAAmW+1Sg1BqQVUW2dxs(W~UDs!eD@p-+5Nls|ib(O?F&m*XW&x^UK9Kd}
zodR`+7IAv$T%yq{kwra>9Pg~II5WxxXFpMt429oAf{3?|O*zA{9EwBSr!K|M1z;wU
zc|se|B?3zDo9b9=UDSOQ=W$WmSHi984-BY&M9II7kZyur%7A
z#<_ji%cEBL3WNQ{q9W8pM&LH-YfQ1J&LnRWPj{tP!`u-yPU6&xYc{)of0V_Ndv^;8
zr;uGx<7-Jtkp4mjw${5U^^Rzy3iVG|XD`58x-#v*jyk@4L}{}@P5xzADrVs#>ZUsP&;1Xp#Ha0$Fx0*748lhTHzt%YbO+fRXJP;vtJ%RanCHq(X3_T3SJ|Jd
zZ5f|?0p|)@o%QT)r3r%6BR-|}FgZgm*XW{Asa~C4^*6GAnKKXjY6?BA9&If;Klz8E
zp4&&=@5u9fuBbn=6q(+|V&$zsY4~`nw$3z^V}0;Y;pv_}Bd3OTfJAWj`~D5F%B2uf
zD74UVeLLeuEU)A0lXSE_caE2^n7<6!*eXWG2)eP
z_KjG4Q8CPag|oUXI#?rJfxW&o_Tj4j(wW#N`MPFNebEHiC)!w6tMp-67|?c%G!6)CgX%v?
zBu$2~ib1^<&Eef3=n2GcvxgL+duEl+pI>6pIFHePYbmU2M!V=&iMl(9N*R*0HFyBK
z-OwTD)^1Gr%&0x3F&7+$q8f-KQLyndWi{F*$h^1Q8OL2MP>;A2vehD#JBJ4c21BPz
zF_^HKr$+OuoU~tr_YHz*R>Kfli^9uQ;jCBE_gf*p{!V^L5WR1+wG;$~+co_%HtR62
zkt*tWUnD;YaOi4=T-McE{h41app*&HYOD}wsoD5*!i
z0XFlBI^`|ND3Q&yFFmK1U(KQGQ8+k&?u;(Xc9$!w=RkWd7%Qh!i`)+8E{#LB8$S+ZMPRn
z7EbBAnkL408I;WaP33IY0NE`0*W<%4MsC7_(9TuMk{b7q&?RV|->~M(0&YA5HO%aP
z6ixLL+EY&myqaDP$&GGWvbbm41RW0IP7tR_*LOg-f+-N;U=|Lm5zrrYU*HkTwYKxe
z1iN7A<$g}`9)-jDpvaWki8uQIMc5rZl|JA0{rHR)Szi3nP_61w@1iMv+-*q&SQVDSgrF8)mG
z8kaTRM`e0Nd@&rthpd^o(S8||kOvWuB1;K*8h}xLQ+ZF$QRA*;5^Qs=P3M5;c5NBb
zvjiuOx|S
z8c3OI*Hbo9kdEM^Bkcpti@7ST7zy7}ev)gW7b~xr?_M&8(lus+2P;v|jfzC+g|6Z3
zst#IaKGNN+UN}-(tv-AV`C8CU;+Y4-@d*2F=Q{3T6;(vctl-X)%*FJV2$=2M0r;G)
z)$(pLX>2fkta36&6x~*TvuvTUcOKhbxbRxT&7D0dkqZQi5kzNytHzYU$@CK(LF+~s
zTZhcIS`@PyU!RhzmK!;3?O#}M921G;I;weC>Ian9FNn$oMZgGd);g-vmWG3=4rdlD
zo65c(9}clRgS$C==898ca33X--_D-63BAkyJRcZsC#LZ_t|_TYubtOf-sThD0XqXx
zUtyw8&^j25Uuv;`-0*9P_4^V04VK!K)+d=JN^B~)yMoU_CF*qhlpo}0Ex)NXDqXij
z)k3uxPtB!mQTCEvUChZc-Mltxqq1YYI+wQs@Td@-R*G(zt)JWGy}`Q^6Zk)Ondj8t
ze%y)3@GSRKLrZQb3dQRg^M|r-dh+c%@*Mi)`#I(4oghnppcK54+L%y~V$U9g0Nan(
zLB2<2Cdn5hGbR1zl-*MXn2RLvb|ysp)TpW=cyP@X$0ZltawU9}k+cC2m8cGZa2VbO
zn4Y63BmYM%+uR>AY-woa_vS`MKKN<1*38z_R;1FeV7o?jc!YmV4laK;mWoY2y(BuKan|thFq#bp
zRpW=2>{6!KDB99RxEs&;gj$+MCHyfQ#G=dl?Q%;P3u3iM>`!{c+SeJMl+h$9>*UdQ
zGF|wA??LKAy+{+d)gT##5-dMCMZ}A9&2VQD#(Knm5$KajZdqLMrSgyN8CwXb#Ov*x
zFfnIP0F^p$D)N?Z4v-A1g*HS3<0BBfTF{?csv?LFD31$U1}J5c(+Z+gLMu52*lLtw
zg94ybzkCBcYjTg(#IddD(Je!FJ)LtRwkbiCAEp~K=o`h5F-@7~esz-@+#ktR+qc3J
zENf_g8%8~To#l}0^Y~_()%hCv%V4Kq8cEsM>I0f;bpmf>io6dWjxMow)bx8I|2dol
zQFA9uw{4M(9MqG&`;c-)6pR9cBFxSwY6(XgOmODD<{$fo>~sgs15q99R+?a
zPVKWFgog66o%-9UIf%zGu_s@xsTjl95-MVQLyX-wKV;O8>qWO_=Hfz4sRyw{57SKi
zARXZ)JWY{CvAKQN_+HMi#q=C`qgy?ZwCPDBrCbk2Xc?3wF{P3>PydS6;LbFt<>G69
zSMljKI>j9JO6liPhU{1?WI#2yhPs8giS294?6mpIJb?^XGe=jAg70KYLXeby?$hw|
zXb`?gw5a3@gK5%xQ?FdprXHPWD~-@xMf&wh;{YAQv6Oxx*=EhvgRcuDbQc?%1thgE
zsMt-NKP^{te20iw9-K-Kb0^c*#Iw|Y!pRHkl4w#5Ckh5@(2(VwNXy5sWZ+KmiV1d@
zO_QukkAp@VfOp+GN8u-|EshBqxQijWac;p*9glc0bE$lnVrgfjla+J&Fm0f|2dYXR
zuJ|yO@OyWW3M$KPf|{}wHMnS|0xkD=nE&~km1$&eg|*2bIrF4xXIWo4v0y`j)WTw`i{`ay8C5KXkvYT-Zl()jGdSvczd|AzY1eRK24&?)XYEiteXyh-B@^
zwnL{v>%#=o7*9;vCVKnwZp0FQ`JI|zgXND7X}kB87mYHd3Ai$Yen$Z1q3$r7X{cHJ
zsPB*7tJfYVu}K!mqj;dazFb2!rk~D#CR-F=&!W4w
z5H#CRm55wXV8}c-0SXeb25%_w>vbktm#4qLTJSB!DHAWuU^PaBAdQ+ltvlb?bswN?
zaA3cZNzsyv0oI%&Cnu?cb4U?hc(g$04SRI7tJZ=uR)Di8{J0_X3cZ(!-Xnsm7zh@(
zY^hG0sP)2RxU7@3C`x;OHwR93WsL2XnYt0VJF9;+O737rS(2v4Yvl~tpyYAz@^ecjOaIlR891dDC%hO|CRJfoU%ILkR4?GyQ_I>}CQZ`saArT1s
zfI36FWQ3hd-;C~Wcck-|MWgNBUlTP50e6k)_fJ)ZIK~jB&JbLG^2H0#hHdyka7qh%
zD#?cr3Mt?kQ;xZ9Vw0X0X-DA`8fJZc>9~i3sg_1w)sWcxF%kz>?8IqidqlbFYZWJv
z;RsQ+0ywQdKj>m2!N+29a(&1~tIdCye7_{9tw3nYnrGMVC-n&9S@v!pPV95)7=TcK
zBM9V>QxuA(rY?*u#;%P4T0Cv=H4f!B)X=+2IGir0gbKHD;ErJV8Fqerq)J&YQm4(G
z1Qv4h%hA9n=2O+L{SZ=n&s`#H7n;jHI-u5}ER5;(;h8q^m0F0xrBApxc2LKDtMxJ>
z>>cr`cij(v8N@1^7aC&F-sQ(9mXyo)%^%}`$|jgzU(W6DYA~~hjHck~mp4?n7;Yo}
zTswK1+bCbocaaO0uxb493)Y$x&mx5FJu9(3`IWwI^5ABROYqi3&n{D5Mp*sBQ*qUt
ziHmu^Sioltrjw%7)^>M6lIjI*N0teh`eC?5@x?QLhj%-`3z^WOPEz(?3=Cf%6;jqh
zN7K*gdcY1!zw*2Eo}}^f?;mjXZ}gHyoMRWz($zk4gTZ{_7f51!U54-xPMz!t_?6=wx!@)U
zF$sk^2ocJi*wKQQHj(>I=K!cfCyS(w4CcR=2pQjAnq&WDdFZ@#bxT=|TC`#w^MlwM@$y09e!m
zS1~*v@(DshbkBc24s1382-Y7TSW8Z>RQmeMm;6VHv&rwtSu^g*4^<@*T`5l(fWDW(bY^G+_Cp#`1IP2z%|zVN
z>;e;W56HQlZ&-Ni7)Y+~Of)wif|x!wU%YUne$k-h3oN#SQDyhNq=DT@A#NWuD;^eTrpwI0Tm(;nqkvs~-3
z(D4=7n|P_%C#BxHTfO-^jMTFdKYLu55NL(Szjw3wbSu(lI*KIjl)D1;rkJC7{(`
zR<`L)%`di{W!Twp!^c(T^km{V0+)`$`yxqQIJ}Jo+PYRlur+SG^#dzlfiDz*m|@ma
zR!xj}n@dg!GYW;DdN{MRQ*~V_6>CPl^)!59Fu1_F^{D6B@3hH9q}&PYrQ{0(l?XUt
z8GwO_R{Eb31hB0x>p=C+&i^)lb(vsUj8}cx6Nn&h*BfN(E^ndN4aqgl)9hL}9hX+I
z2%B{(6H<$0Cd_gM0d==|^2aeM&)$rz)jgJwAhEdsD8qBpM2Q4h0P-*1(T;$Ak#Z%k{eo5Btdn-6(z&S@WCf?fD%7fSke&b)Ge7_CZU-;1Gi
zp3fb((CTDCd#hMBc_$<)EIuKk5lmWIp3e0O&$~sZ2B;*w`favA%%e&$eWQ8cj2^54
zFRWO|C*0L#Lm0tGw%(0_Y=c
z`7D$)eK_Ir!>Sq<9k3sN&*zDoQ=1HuVNFl;E{fnc`3AoLQ)OH<_;nw6ng~1V&}h*y
z$MXU#j?Z?_VvKvyJ?xIM)0U}RQh~2Yw@A_2#-_WYMLGGw{(R2AW?yfA9DL3@tOK}`
zN!5D91mtvH8Xyu3e|-xA69=Wsc*1^O=K%7P-ms_sEAD!Zas_~Y
z>4p-~f~4sEK>sK3%w4cOgqc8d9AR`I%S^ozb}LgO#wH3RqkYY9m#wHNZa
zzXobiv*p2BWgXLILpL1@sj#j&5o6Yj3_5npxGG$!SwsyOrSX7kIyJLJfL$0h@Lww7
z!VCvPhDuP=eb|YmmnPP|94x1tt$&Z754WH(7EVoF6oeLFx
z;7B~%+Uq4->yTk@YtJzR!ZgRea@Q0qv`_>j0ieKVe0<;jPlZns@D&E_%e%y%#V7by+_L`~o!YXdfW)5PTFba{;#F#--ezZIHG*xPg3
zx0i==s?^zLLQk!>Z^$FHcL>wk4;*B4r`Q5pc!p{FE*;8Iy|IabZV_o#C7+NXM7K=Z
zX%P{lepqFftXPFEkPF|iW$nZ=ALlr?e=&T|!?K8f4n2m5dUP*$PO`f@AmBzovEy9X
zfQ6q_31{KUmDV2h%iqoKMSq6>)Is7N{woWLRSQC7`|A|{b%AP?RPzpQZrxaBzV2(i
zS|b83?}2YYS?%?f85!iH0ckfi$(J0=19A?fo9JE;TXLz~KD0;jA;$#AqPp8DWh=&7
z0Coz0&{{~uRd_+Ts&8-^h|~IRggtme1M!S3gPOEV+<^ycZr;I{44R3&lMbPwvTLGw
zZs}D^WR=CS-OqKYvygG+Ac~w}hZ0sn7w>EB9pHucAmpGaE5|LTu@F#$eo#^MLVpU~
zg>)nuqFP=*%ekJyyY`Rwj!dkq+cJ2DiV;_Tg%3!j;PRQ>ka&h_#H@E3dXz=E#YnFm
zT|rquYVxu_8~wJK3SMO;F(&ec#A`v6E{|crDc%SoX`F-Ug=Z*GCvL1R`z>1O>wpVf
z;C0B00CO#^;h+k{x7_k<*_W75Cuq2n|8`pW-*?<{TOrP{n1TU*q+qS0ix-jELOEuC
zesA_9u|((HFJ;7F-s~*Kwy&;{grgyly|{|GNOazS+sK!e6%#~04N5+n6@Mh6K1z61
zP9L;nc-+J7&2ceY9^hK3-FagOic_td^PWPewq9}D9oaDali;l4425<{1y9DG0~pd$
zCA9#V!zHKFJ7UnS969n4=Vk+L
zuaC$p(MY9spEeNr)w-aPJ|fJ?OiA-h>rpPNm8^-Kdv{Avcv+^`F=DJJcLDMcaX9(x
z$mQhu30)_{5W8da(0*kVFB!ll$e-k-f1T;thSyr1k2}?Se~w}q0>(~qV_r^wG#mzl
zV8;X(NVEbYMH53CeSz4K7f^2wiKadNyNJlTGPil*5}D_u+2`2PaL>u=rDczBDrt|g
zclRU~fz#@s$QxE3ePL<)3~1#7IoCBKFq5kKRR5(4Ja=10c@1aG3I|+p*w?1tlnx{P
z_;BKdjN8JaFgb;qcs7g{^WPGa_|5BriFt3ix
zam`MFU1UT`{1(*ew^6i-+x%W%bB@Hl1IbFV@N_MHfM#7F`6E+LmV-Z5%7tYAHM$>xO{#us%3%64~?(6x_rR}6a1*JxbpOqt2S=9aa_yXFn)
zI_&{58G*H}NJ)gfV{m7|^F18fwr$(Coot-l*w!buZQHi3jcwa@{`>u@>Us54S54i%
zGt*P|P2ZYRr~7EOQ_$Tb33D1k?WZ<7@LpPf?5WH^O+StwPm|v2&f;^dGTi*boBp>CI+zB%y#EmkE5jyIZGz_3O{9$yQT!;)uw=A_8f0|+ygenr_n>@JZJ
zwNX0j4VG2=&Hiq-N7wFqKOw04cp|W&1TooYhx@q5FEX~|VpO6|c%?`m=)n*vG=9V&
zH%2ftiA`br1wiI9u4L^GgvoO|Xc~7$L?87pW?d*LLC!2{1ZGP^nj*>7UYY>c^Spsg3yA
zrg(DsXM+VVWEdxaZh=egV`yOGPOlUej^NPnJTgUz*|80gn=fV$BcK7Mz`nv+!S?;ve%`r|
zdrsLg2oBl(LvbXdbL^xuj3X*wRmv3Mt
zoEVP~S|aRXAWN{^g+?XkRo(RrqhER~qTnF)iCH2|#RUj#Qjoh6q(^yyb#^-Us6^K?
zV-U$PpB761qd<7p7XEi@=H6!VkAuo_V*H&?b?>h}af!9`Rl&9{|3^U0Nrk!8?`>2x
z{a-WmU)pAM<@VZhM@VkTsRjCfqMNS0jBM88{CYh&%pBF&6Fp%$oA0`~a1PYc&vmjo
zw5F<@qqv^PJ$T+cvkrPge+{pSj|B{*`+2M^cY
z>#Z;HUV7HdJ$I~`Vgoj<(%9%`P^8I!5sFfOCyWh~xv4ZH{i3&bz*GXUGs$OUKh$vg
zKv8AU)-mg%_vvC4tjZeF5>w29>ygp=TRK2{CLof}%3o$73{bksDMEv*YsQpsa^?i7
zVa@LXh-lPd2Ow(KkVwTN+KAK)9Khg-}Qw#+gkD)CkC8>+Xl~Udh&SO?ulcWx0l~KYAl!6QZR8!A#@HvixI2J_M04+b-3%;lZ=J{J
zfjEvXI(a-7RcuTi5a|u-4U@}VP`fk%=O5}7jBdHY!^aJzy7p4O)909HlpDik&{b=V
z;n2bJ>Y$7?nsl&2&|kVH(>MQ>$YWmuh<6p1P(IJycd?>Cu%?9&V{LEihvi{cF)y}p
zENY6J1a&h%>ooEEfONf9Tr8`8I~S16-})w9DgtljJ<{PjP+J<3-}yyhXYuUR`(|<_
zTI6_;ObRO|i-Ago>4R8lO7$(sP`Bg>yL}`EeU5=K>#^&|)+H<`bbC8^UJCF5kYEk+
z7zWu$#;dmxd5}eUyG~zH_jbf8lqz&H72^`S1$!52|IjOM3zuiwoR@~aE08Dx-n5f+
z`Wo?xr+EgHlubD#De=xf{7BsbTxvza*CW+WB@_HkFWY&uye8iIc<-Ob7=yNXZgT_mb1{%A5`v}kgY
z{Hu2!E3f@N#U3yU>wScpqvn|o3Z{7oTrnMkrA@~BOReYZbY`T;4?gxxunz$k;mw~1
zf2H7-C#ygE(tS9!rXS+OZrzNe+`(sLd#*q&^SCabo|1vfeQI4GMaEkMw5Ml2LA5eq
z_$s11mD-Kd;;@e`)lVUR#p^?7LJndWOjuY~E0wd6V48iyjTDxOe>e1}nMtNNfZw+n
zGK-hdE@@Rv?z8%aUKejV!l|B~S@7%u-HynD;|IcDALd$7ZgE!_erD^5;L+NiH2Pph
zF2*|_9dkFs$0}U7i4tc6)?-(lgdaHXP>zqKdxX4$|H;Z{{Ou7Iexq-8>F*R^I~c-1
z&*!RtQb%EyiBb<-&Tt;aF*^T2>}YT%vOQ@~p^ZklC_xgA>}E8T)JGOX^s~o|y-0Y;
z^X6ngS`hGWzM$YuaTA{dfyAs0Yqg4&^B72)Cg07(G*T`o#RDeuL!WwF;VIAMSPl3S20`b5#y
zkanS2!Qb=x2d8RKQ9v#d09kpCTS+!!TeILQk+chtm0z9o4`yWbxElpR@$9eH7f;W^
zAUnHa>Jfs{N+hTR?9O}pRrQxC3#6yKZ&LnSeNPpA>n96)X+IGr%Yi5Rr}`E1hI`oa6*i6spe
zuiz3GLFZ@~dB}t>-HrT}6Q-$2iI7WN+`!uht?jj4bVA??u%Mw@7s0DRrG;b1!b|vC
zCNul9zywPh`0XWRYd}a!j#=Th#pP0)_Ca|3E{9N%|L;!Qd^oE5@kmFQY6T`yf{0vq
zrZ=yQ8&jrJs@No&N~)3TmNG*?<|TOD8<^5g$ua9Zl&Q^gTXBTG_Qj9Ot%G{>O3lYw
zy@|s!Uh8`gP=dV1*e^|lE`9YP^wWr*;$aWW@ozXgSmb49vlJHZ*s>Fg><7+MzhAqb
zcSEd+n9`zv7~hP4)V}uQb){w`wIU`D#;NlTGCH+FVyFaNM-?(y;6AzYbaO1Z+El~2
zqWY(QQ@^Um{1zga;M+WH1&~2b6dPARe0!+7Wgg)GWYVpII}ptI^Dn-E;5A~#_)884
zvEwP@M2sM{@pLqPO^ZNZ-qe
zZ-Xsg;pUSq$PZrXyuzY?+uSwjX}qj6y5dm387ymdp`86=Qf|(HUR3q>jU#`N-z1qi
zo+a|Pp{8cZ_vBH&HjHfxYRB3DuxMknHkMj63lTexM8xo=8-pND+mgbi|FzZq#v;Hv+4UwH!XX(QJe~jZrh5CpyJ2HDqTY>Z(n4_9H8To}=v!aFmB~
zCf1d%;=@j!LkKK%+o1pMcZ)4;hgrS4iVTA500-z?tFJ)Nua(!POW+cQH%`@|W$oS<
zOqcyu;%Bjsavg;2s6kBT*Ns3X8$UZyhFMV7f+7=#H=df0{L+KW4Zj9pmg7pId%}go
z4&-zddgEf3t?IUHloEEHSWI#PB9+3wbhTuppV(b-c*UQX6-^^*XW@@=pW}^G-ni08
z><%z+bcss0OUr}j2BkR2GHxQaGqF^RTg6yJCRSr;@GkvH&D!eYz}IcIDH_eLcMc0%
zMr_lIsuzZLKz!_4`UKm7$)y=NGs(5H8LljV6?PWHR9N!_5h7^*uriYa_^Y9$BJuoZ
zrDDR;8)U(3&(>;^w`*e<$=Ni6&e_ZhuLs|{UCj@A+^g(96EN~@dt-$AH<700aXsmd
zp6oa>8rapz#9&^0hg*)x6!gXtdn#2>TxOB{i3e8P_bv6ioPTAnu`Exue3x_}+xyhy
z6AogcM0T-!P+$bi99Y)?ZbkWi1WN=ns(&3{EJg*Ylq
zJ?Ae@khE78Sb=q0%p~xI83`#=26)V5_Ed`XB?_erVG4_)HXp#!6bdJ8f(aW
z+jH*kHI4Xxr;(-Afkax6q)ZWD{+L}mjmfOkn1(6vc>P5Ig?B&$z#J)yxkuDqFo1V@
z{GL!^Yr9#B?Tf-Ad>D<$OK2amX%5HJl>+TUmQbATRlmHDs+>2(>aVe1c>;^mV$Ht<
zX&5CQX-BWz3*}`)>|Da9HhzU>EN!|cRIPJ)6*6505rmD$ElBJbC8=ekaCRUD<`K6a9O?4(+ALQ8epkV{
zo8$H*K$n8F5LL*$6AWFT7vHcFtepQ72{
zjB|3&v4@A?hp&|Gs&hnO@l>$%(lIMTX18{U*l3#*2PGDcV_
z_r3nzMb*AtP2kc?tsh@-tU4n-PA;LeDqm3QXK#j
zvzQuV%L-roi}NGWhiSY|s9rk8x0Xo>w*dvm1r#OjZ~k^?ab-Pg)4opYR41+R2yc^<
z2CLbv?Zr0-pyXJq+a}Xd%6e&uV7n15@iWs+)s0ZrFf&lcIMJOpZO6t4g$ObnIw$wo
z);IAE>?qyi=^&f}D}Q;3x#KjSSwMyoXCX`_v+pIe{Ya}~2>+a<`~cRD)RK47S?&g^
zJnfAjww`$7c=@?bD(tsnz#kp{g)0c)qu1HMS|`C9fKS{i5AYbAFSh>G^MuH2BLn8|
z6I6ZC-eSU^>6#pz)ZaLgELJ8^i@2Ouy_{9e%z3V$ZtBw3{BM&-ZNzwGd-t{qsuw2G
zR=bE5CH^AtrJLlnc`}Q%WH$o-L6*@-&tQ8n%djTD7#;uC)72}7{L5
z=x(;60etkqq@%R(mqqeP=!?pEZbu>B1hbT-Fx+k!#ay@FO-%PXrYpe<%BGj-_GQ1#
zcn)Kb6VYv}3S~}DoB<=WmmY_`4)v?uV=&h#0`&TKM-He~9yP}ma8rU-gLQ0sH>7_p
zQO2j5HB4Ic%Ev!`_bT}s8c9ws#$V#?BH(jU>?}acVkeN+9``o2jDR5Q77Tkh>QLwAd63*gydR3BA=^{MB5QN
z?F?z;wuC9D0cqY?=BfzW%i`UP%NxE@fR!vsJkdQ1qg=kqPgdx%%MA+RqJA3O;+E93
zn#;_Q+*{MIjTt!bP2RzgS8pUW4E;1xXQf=ty0GEIcyHL-Fwb^x6w_5mI)&?W0H`D*
zP)HtIn3OD`D|uZ*ay%?BF|;_~by!UjDO}XR9N7xG7~a4%bXAZO8{nnHS4nnoD7)1|f1$
zT07QBf#`vi5j@rHmLQj}b=ff<0`Ot+$}3oo`;$SOqmRlmR)E3*3GKfuGlVsa|DB&9
zQhQv$9AW!35HlqR)odGVdKawGe77-k2Xrw9!=vAVA6V;pdqG3^zX|7f$cN(ll@o?*
zPo!drEpy0CAQ_9!zL?a%2~2aLe*j2|4{rTC&?imC6(yvN
z0fMN7iy8!)V!CK|<*>?+Pe~>?l~KOyJP!#43}Mfzf)ZS-dOSO~gR9Pf+JihWa2hhT
ztNrJs7{jo@(5OF^g<25LV2PivfW9F2c1+|*hu4ZHfRZGkeL9T$P^e+Z6C$$OSt!}M
zPv>zuJ|fu_Y>{eacdK5=z(Q^#THfd`8u!T;%>~7LQdc&
ztePJa4C~8Avg7T`Sxo89@BqrAdEKUL||AVVw%^w%-!I{pKGu
zT3foBv<@o*m_{7yy`i-j$d^5#f|}ER4}FSF8BxYlCCGSqah`LCT9jIXk&}kL_^%lT
z&(H_gPurmc!K&aFlBc-xH*Az&cB}$pdLg`P()21E-)932lE2oC2I#9IM|otiHA4v@
z?UtfMcxo)!nSKN6CzBuy+X&5_gXNHnC1EHTHJ_vaE4)_tW^~A6h#+FjEYT^rj=gH^
zVem+AdO6aglCIp0)s>^Agz;7hRL}lTnWq;2@mIv0SP*
zP`-$Oycf}Qb;-qsN}?i7P}E9%fI8me(FuwfKO^dAJj63w=pawo%^c(A$k$$O**vas
zx*w+k7_9w#JbPd8hb<^UM%NWbE%{7?5Z0uXvUJN*ar~WWhm}J{PURTzZf3h3C?ErR
zo*nT3K(FxcTs1g4(3e7bnW^@bPadD~=(jk~Q*y7N8D>yXPX6s6nSZ9^sG)hq7l7C$
z7i{5+35YeJe=Jz&MaB{f5x|TRnwD>Xz7z2QiY%n*^xghA(BfV*?m@i(w?L4d$I3iI
zxdgEMgD)m%>lr!SEIjrP{OWZ2Ci-#&VsMm*$Ac_|L)pVhts8U3d5|egCG7^AfhmlR
z@tW3MF8TS)KtU43t}OAg9gHWzQ5dBbjcg!+t<9feU{DXN@rMgK>FGfll6mtX+_*Qu
z)XB^hPq5rF^8<6#_Z&Nt!;miaX3def+(H%B^s?66ND@S5cQ&m+kripv=1H!CneIJ0
zshtnPa`RQOqt;?|)K%!ZEErUj{+leB*t`#?A>)NG$)$UR`8;Ro%s+C}#%$jdd7G`3
z95m!GhwUh}V{VBobC0}MG@JB!^(ZDFK2}_ii)@+jqs?FwNC75a#KRX6PN-iSx&!1C
z#S7o`&vi@9&mR3pcFOmZlnRdg9Ff(97MJ8JFTaO9aPS`QQ1hE1m9;W;v
z#Rry)Zag778hJEnSIH{StP%ZwLYs$vHJ$>#n-X^w^`|(wWX^wnq!N2TmmezO43YnV
zezZD)#XH>#Lys;6Jlddz9x701GZfkO301&pf&N+bdvQ^}yGH@_)#Cvm4HDDD>F&{n
zC~&O`QWbNgL1@W--&=t(@_DM`c!F4UM19Co-j6q#WNT&g`Nc<@kBz)Nk#MMn<$D=-
zgCw%
z9-Q`+J++njgR12zqc4>B7Uc<#(gJ$85it0~*;iY0H#kNs#ajiv7>H1bqH%EnpZTj2
z&nE@MKW%44DmsN=$xUR1%Fn?z;THiSPUnvOZzgW71T7vBe#>oudzG?=EB~az9NU%>
zjgI>?srXB03>YlQ!#_67Vyq67qkshE#TxY~<`X&tVfFL&S7?!4ewF(gcE^0AJt`5z
zW?=IfSvPy5k1jA#1Ezflufg2%-zuxG>v^!1J7q^MkDlC-Z(E{(7IaCdujtIM6|J^RZd^^<
z!M#ciOvB^Q!QKJQW0!PqCZc(1QVwH*`KG_GuvS*T{i>TbLuK6-
zTAa|$rN%A+tEy**e@Oz;tXNfBw|jtd4oRMCOId0<-?mjf6Yl_B=vq-3DsD@#$eEK|
zWwfN9CDp$sd?_B$#D_*TWWfI3R3?m`?UVJ3OIHE^@g}p8d}Lq_E1>-wJxutCoQWR4*W~+XWHoEzyQh@-t2Q4c&TG
z5G_tF!fX^0NXbHKso&?ONN2Y@mrQsRG|oY2t>x3#hlpcyIlYv85!
zRWSLjt}S#1u_1GQzMpQAAftmOFXp+pyAd)13tpssomtLjR1GqtLyoh1`ZerF?Y=~H
zb*NDSDk(SOz0@G8BZoFwf_aBYA2DHpzLH>d<^JAu#O7PHCRci^;uEbKl1E5SuPlNj@&iKRNL5?YKw~lrS*Ub;T!^fIKy$^n_ConQfpz
zKRH5{_mtq)6T@6myGufRS-(*cic{oJuR@*rj3`0lPsmV&`e8ecFuRwRa8w83iie-7Yn0U
z>~O(xFX<^7_JB4p>bg{Ls~$6|@OFE^?w~trBxmWj`cB$r5A{AigkB&Fc)}R2V~T!x
z9~r@ySD?lOgYCmJmZ)pS$f;0V2LsVNNfg|cY@hFY0%&tD{Ufw4W>nBiG257Anqu2(
z+K(1OJGx+D{Bv!5r)jjt@_uc&n9U+@13DM?uoz;YG^`U@C2;%$hbi5j1ZVuRs$m%P4=z1!wfi9JO6u=RGY4_=u;XFMZ>W2SwyhQJZ_#Ip_ib?
zKuX@m4wgun85@O0eI{chmw+3f)(};*jaq5W3zjCe
zPHg|T>3hcOfJ4z7K4Hv#;IJs%gU7MvPmYXa=TX^e?=8Nl&(%v?4NNh>Fn=HpK5^8C
zb>-`5u!lZdqBgq`^d6>3Yu4gQ+r`AE3>3iLf|Vn%-*Ngh{xG}X$|Q;L8F0d9p&fsG
z3N}jKNP>UEA`{Ebg@?(?*Zekfv}f4*ieuY3mnpQOpkVddLH@s(Qh{D;X!qBBgHBud
zdu)xDSPn_tbSOZsiPSV28TsPJUl3Tskl3
z<%sVxsa#k|Ei$2mqrf5jMLeELcNeC~VK@7dI`*DD&Iwk9&0au$8@Ft~xEPIJ5J8s0
z*qa(A+6_>>O5bTQcTOk8SX}G~{CJyH)(HNzKw5;Yf&fq+5p2QgY_=eY0((MPhdf-T
z`qA+iV2}3y!hq$c*HT%Lo$O8a)uDULtM9sb8u%1+C$$l`Bm2j?{+b9!6mKR?vP(ie
zT_lzEuIJ9y$QZiXO*f^y%9{{=y3M%aoiRGV*aP=Oda#w!q4EnWbX8Vz=YZVt&WU}X
zLnu3!XaKZrC8*x*VjcP1iSkl>MSH8$88kn5McJ*mw}N})v+!a|dY1AXd84pt3~WF#
zW*b=R5&E_yH>#ZFkqyEcqdk@Z-8c3%ZU|mEp8;1@9JY&HA1+6K>n%{LaEdMuG6zA(
z{HxLeTa@EyZBZIG@N)Lk+VsBSLePkWA!guo7z9MUYV5^E2Kaon2^vTb5#!Xvee2x5
zj+0==5`Rw_LrBkDxUBXe-l3c@zq+NARTu+roBa4HS_9+@?oeh4aVi)F@#L
zcL2`YCXP`0=DtKB^gz?$O@i5a&59lF1E3@GNd3l1U+npyA?%Q%M3ty;%z2B|)!TMQ
zYmN}p<542@4bVT~2zeP;>UtdX?Qfh5I3kIHkdQa)-A@W5=_}P|JFbR=UIf(eIhf9Q
z$?wL&`lbtt1ocvKCGiYH5(rlc$=BRpH~_=CC#f(`!++ug@jCMYqk1k2gF!cE_Q4mT
z_`X05j64TdMDSmqD}w2%MG@evCgL3cO(qT1_ym?itDlq#CsE%&Pj6MwRU0!;eyk$F
z^r;#dFn0i7J;b@I>K}^DD)C>joN~j2ljxJIVC~xEjonm*lEekWpcBdOvm@)+!vNRZ
zB8l8R84PUOdtozK|Cw&}PU!lo!eX@s3)itC1jDYaDW{@jS!V5{b+pkCqbDG?9{vW_
zd4QCzCE>GL`RJgmM#pgk@eBQNrj>g&md75Q%bMg2U0WO0UoDqjI=C4DRlGh4zk>LM
zLeIk(`cUMF2g1&7%}sRBU)JHY`~ZXoL}|S6h;N>KQl$tow&lrRe$P2&ty@%IN2C;w
z%HLpc6tq0&qV#l(;9dal3>ewmUw^E3dQ{u0RMLwqsY>f43eoXm{B~^`rIDXKs(^Kf
zx2f@AVC>3_e8%D{la~k+9`?07!KcBa(YXp{&d>s_GUQossLI8%5;CwQ$
zJAlC7pR4bFv$Aj8*c(|_=^|W%$?5C6mk{-SUr%%+%&y7a(4CNsTwK!XTq=o@Dh;Rp
z)Q#M$iWz&>S7w4VR2AcZdDCq!iB^<(q_UzsgxAFe1aLAs+Mddp0Qh6UPEZ6uiL9ee
zeoT0gBl2OKFjd?2cLDc3HX!)yb;xfkY4+9Du&0G9v#s8RZbifrz-D|v?s$FH|
zRf$|!e3e(RYoF50Ia7*C$RS|46|S?xf@g^Da7pj>YnD~>PGLxfMU=Q9>pA}t+37BshLtPN5aw{Sd8`LA+=n2B1ct4nq$%Xx;64ecj1%^P+eRn?sW<4fqt>`5dTTuHl
zY(M9reeqm>eUgv(-*REhtpCvjLvIy80+)gTVP^ktxv)$ea5e@IX3o}sbG`^5%-pTY
zn&9+U|05vQnqUVG90tPjUt7^y%U{4B#X(s9Yb#pnKM$d`pC3Yu2ZZInq{uK@@!cUH
zl|WejM~aMl41)Q8rN~-+86m}ggRruu{)>;^x}FUAn*)UPzazD3OhK9>gRuTr5cJli
zEyx2{5Z3<+g4Q~H2KfgXPpVmTC#N14ZM*TO|-{&dU+-re_
zM%pSl8q@qLx~f8h4yjjD7_-|s44VB$+NP48r{L#s-_QXn@0ehR*DM&?D}rOKbJ|i^
zMOvCak$x^&v?6_N07jB!hx{v9Fd_||CBH-FGlZBl73?69Gq6C(>d_)>!
zlyURucqn<|0|#0h=q80(b14c#iP<`^RP*dpsTH6N6bY&RpfmrH(y(|Xa|&l<)Vw2B6#090KSyVl?QbP8v7#gn>olL
zb<}TqsZA*y=@KS=R0J^pJ7m(HIth^@A+14WY7B(|9F`Gjb7Hf-tdQ%9+j%1}1Zgbh
zV7CZ0k|dufBgF+B0b~W8cal}9oRs=7m40LCYZAP%M-m?S)UjPGR)|Axcx=b~;9)Sg
zdsF{qG|}-P0DlaYMfp;ObppYhs17wYHnC%@inE)|7$=%v6wVW$A2XXMd2~G#Cu$!K
zaxwmff*?Y-+E-a-;r2q?h-0NF?iSXN4@O!dN^TBT?f2JH}R
zdZE;7U;GZgHF?;9xX1ScUB(RryivBsWITT`vwzq5hVZA&AkCMQKnGg2%UREwwn`K-
zeOqRgY$@N`g7QLiRB5Ogb-95woKpw-;*>|Y75%x!#9EfANe2Hc(-3X{&73JCE2^_Q
zVLtN&;C(v%`^Z#EdWO(ods;txs-rPs3<%U~-;gZ}T`?D9Ux+xfaTQo(X6gf3S>p-b
znDQlBBPI?0+=4xuSF6ZYxV?ogZ`)6z(PCfGBE?ZG%6md)Ry{ap9w0hf)2_OM~yazIPpj-=E3-p_MMLrpvYbM
zd+sKAwDwuCe(}XrG&I8V9xM4uSk=m-XqzT@05uvugp7}zCL6h|pny%eL{t(eoyQ*m
zw(Z>evTQ(GmL*4qEtVBudL784LPt7bbH0Br3ZtM9U--PB*ocuFLqcr!9$6OW-pB}j
zm_3`-JZUu?!G}#|6kY~4)pRrqaP9M=EW+IMI6fqGakxB3O#vpO2qW7B5p26O+T0@8
z(IR2{P>~|Epq?oS0*oG~2flQP^#t*cFPI$yrVGn`fNrfGH`d5cuAtJ>p&xy>Y3Xvn
z9BAG@vl5EONA|S;jNk>4vg|}|MX?fX_~6dWQ;t;jB#;fcI}8ZPY6&BLy`@_&DyZ%rU=Sr4i9n6ohEX7-e`)5v5U0m
zRIqJ!ezJNTbX4a&jttM
zflr-f@vluWzxqtX{T6G{lvxm&(dll0IqCyW11~av1qllOJkJ^hAhIBqb*CP$KInJJ
z5G$sXsVS-#NElKDIwHsg5A3mPc!XI9@MwY}m724_7*jc@1qE;xHIZv#ghLWlOwQC{jK!qqxZtFsgvh~>
z)^``)654uDnemQ0paea+-KCx&0y08tWUwwMedamG#-yYrDQ;DK0OKZfJ^4d7#G63v
zsTw2P1!I=vIALScS%1z!HkranWi+(G2S0;SZL#Y
zs@O_uL*5A%z|usQ$Wkm!W8BBmuZj=HN>i0+;F%z8$injz$1pP-rMm%-(#jWjuXYAi
z|LoE3q81S9fh0QB)ESl(s9a+qa@)V340-ihmEmpGH^_W%G%LjL6&3c&}Tvi
z$c$F@e;@McCZ>Ag22ou`m(YI~2F&T1oWAjR=zqn;vexT^*M~iSDtMAh3k7SMnHT%ZHw4^70aPe&q2#r0IVX(r~YyfVbp_kbq
z-#9S$0NmUS(lYGhy%Kre3aj~1VNW_^uOaKb5A)>=!W|K3SKK|LC`5$hPm(0K3C+0h
z5!BAZ=Q}@UNhBx^^gG^<3@hu88+!P}S|5SHrjENJ!d}=sJQ_ZbL^#Ap1rAI{L~{Y5
z-mC6r!7t@tzj^1z_HhlE;K>Qm%7Tjq9I67L0zxOX3*Yi2mssiT9G}8{`XEz(wXDW>
zRmX&QSG=|_n>WsnAic_g{&LL3qhe@{fguPC#Ye8AA9ZA;0J4H2T5*>G34doF#1*HE11mJpjmd?Et#H84mSMU(aP*
zD#zD3y)pLGKg<|mpJ1xuvN`gtHg)|@
zFELkUh;k|M9jxsW2?mm&MY}g=!LyRBz3mFN?y>zftvXdFOWl)#Jn>bLeBbnZ4F;I;
zF-Th2E)vX%rg-DN?Yw5M`E;~z`nj^3iH*^}p_SHz*qXF3Y8(^jZSg43yAvx+=ugEf
zli=P$ohjk(>beUj%XJ$KHT$`b5RmAlv?tW_DzLwMs^UF;$$@8tB#)l$8?Wm6hKt!o
zHK?>u)0?Hy+=9*l`w{m>$K03Li~tY<{_5a8TGXWEf77^RYBO%*;dgs|wYEx*7G7K{
zjZ=OU_P=c{ubqzbdHaK$iAAJ$_>>zF3fQ(NGE_I>R&QX2?ql9*;DwtwkO)N7aG~-$
zFvv8#7cepmC%3&b_t1nEFn*K!qJlcmhZvKM80#s#7dkN`t1^@NeIo&CL;y*y{-+O&
zp^^qMm`G}Ji=m5RcrJo%)FPU>u_T=<5hJQ2@8>)7TvCw{O
zg{l@AkO;fD1>R2tguNHDe*wC?vXcj?w-O8P+Pb>1ohjW4tLgXW9$KfA9?r>EHJ0mB
z*o(T8aO0j`@}et1`MZf{fd)MQJOs`FI>_HgS5+2mYs0sLii`%?cp4#ps#~<>*-a?0
zEe&_8IG#_>|LYKfH7x*Y4j{wdS{cK(1?Q7Xn
z?_2#IKl=qar3=Dtp2KVJCph`kt*R|O-;SR%g%AbT`a)*lls#@QT=o*3$O?UqTTQl<
z9ln6q&#l@U+~wepL2>otJ=o}26a87-zoJ2S4$g<}W6D4;vGuC#@H^o_JKerNla+z%
z0xq?4%YBufK5AW!>VS^^!N1isEgKcrMxlS%K6`|R3}|QsGM&A8lN*ISm^&j=(oH6J
z>^cH*#19M%ZhHl(Mj`f%omtnKpb6B=y9=>O<(A9^`3s=6n89`yfJ`|OE7EIo1&
ztmLHs$@7x?XpNNn`zZL~q_zVx)H^v|wrAG8AaHxP^iX>vI1T9b25bvBAq;Lmq1y)D?<~(rS-@?q*
z$L}@3Y7I`ZW@ADyaD2Mt`7k^fIR2a>&SI@#d5jU-3J;AH==B%`htBSJ^}c6+PFd3ut}L@vz#9}Dcrg=xsu%2L530v(9xVWfz$KG!mGIiX?Qc~
zNN-0jN8;$_BB|O{n5U@U-J8$RsYk6>zj{5KotCwfnjLY1D_6C!1@?lw3n4AFIJR;}
zbu1ZG_8LD%^z$O`W^(Stya;M6O^(N2G|m}zG><6`qXKH~(5Dykf3vHz3TpPP{@t6!
zkEJZ!kVF5o1#P!PZkMw;se7>Y*HCF5gP*GQOn=`|#nobEDpBd?c-sXam3i^>*KnB-
z(KpPj0ZMJ*G4
zX6LNXp#v}+&SR_I4tLA-`EY-|*eF@jUBxUmbaYokf3piG7(D-Lv2da28Q8#y^OMFl
z6MT#?;>~>-J^e8#+d-clN);6HDstl2H0X0jC+;x;By1Ji6z&_|sz1R}pfPT7o*5
zra3^nu#smHeSImt!QL@vH@H#0WKegDR~xOFg9<};Eel}-+w>@8jMF=Z+S+=E+q;E1
zBC#TOA}9k4Jl@zI0h(No_4N@<`Rcyg&D1DN=QYWhO6KH!30yXe=rC
z#(um8NOf1_8Z0T6M`Nw4gPVLpQ?&o2-&FWo1yvSR(uQ#|z>dhr=I^GAwp3O%q;xQ_
z10^+oGR5ZaEC;SxW>mFnakPoA@Zl|sC|%d=wk%9n_}&S8WmT1z5{P!&j3%HzU1kK{
zrMQLXmY-n~iuQ(HnrqmG{UdebfmYVPpn|o%{X5Lks_V}JM6E=xcHTqX+mB!vK6d?W)P-VNBkerbo`
z=fB^xxjyy}ok6#0~u+TMW6&#zj|R7sW}
zFzvE25Ym52-|39RA0|E<_!-aEFJX8!{m(svoI0ZAFE^Q)fv>!0{!VEGJj_s;`zC;i
z$J_CrTs!FNPehweJ?sc2p3^edF>{X?c;H0NwI8!*tBeCX_3F;n!8sw?cPcD
z=Bf*De3@1SZ~ZXfQ6WMXRU9KJBliI*LMh`)B+=?AhND@MwH|bMCChYpQZS>c0VJ@B
z&_XHFjA)YBF$f`aOp)M~i3=df#oLK+WHEcJVE+cyf5X$J$i7w$Pk<$>qM4>-t_O|$
zx|V=hMw5g0=D$HtP=1}OCYQnnk1)IBuET?2!t+rFh}3~ERV0yX+W{63e$dX9F?6Sg
zcgTXbHS6b-HR@QIxWbo0>;j?*8Ooh58fC;qroy`|@q08r-klBy9N`GZhb9BEidqd+
zjbL-Oq`$n6-lABMjxE#EIDc(wm_yf`!>1Ye-`A9S(8vMhh!5T#2*}vL+2UIj1{_RC
z+o;Vt;ixf9HTIpKvcv)cj3o&&QbMQN5w~1=*KB$Vk^Sjx$`0G;6(k;=OV~(;v^rq<
zrUdRPu@nUhBzla6@<;IPSqgM|aDZhn)GWxh^4LgJXUVUEO!hy)SS+z+kTTwt)|Do9
zWl8ZBk||wdi+j{U3x83>RF*%mFfZL`A*-e*pzHIr()2jHCRYMTlFii>^l26bkF$;_
z(!yU@IAMUy=G9u&xjV~UX@BWhj8>|ws12&4r@9KHiC>Pt&mL1-qO=@S7`aeu&|G!s
z^_6#BSVoamF^q<-dTLu6@AXVBp1IE!Ym}c~IMRhUvmXVpoNX&HP$ga?{O3^t39~)?
z_x=|c7qr!+CBR5ZOP9FBsWEAzsIf6AGp2PSWqS6AMuZOa#>`rF;$6i~c0vREZ$g8*
zcw><@)e30#KrwuC-K15JRND;;YsDW~n8oRk;nl=IKbxmd{5Wai9a5qlAmUp3u}Ww;
z_ve~vMh26|wBBk{--^$m&#AR)?VjK%j8!~AMxMF@P6NG?ziw({qf}gP!Naqf)LHKu
z77yg;_u!!SckoZ)e~YtV!wAPuAqRiizWcbfT(7uwZv1|o<@LcupoI*%*uK5U+>&4E
z5p+(C{o491YY?1@xb-}mq>#q^__JT_T=CyIl#?lS8V-sC!2An_QPslJ>|ZWqhGEns
z(q$%MCHl`x6+63s36g`0h>1uAhEdw~x7|OF`+qz!B3(XqQIUVs!V;`pq8!}ftm15J
z>|+0NuBbRC7dwZj7>fYW|NoL-@c);8IXF3q{yYDYCvht#3!{bfvc61*hJgf71vn1l
zi=pWuueDCnl;vS=_Dzfo$k4*V6ov<5NZP6r{K_9B#t+A!Zso(cbszHVg**_TKW%_XHFp?UWXpQ!J6sN=qU6wh4AJGKpNh}P4dtK0xa(peW0LVzK0wPTb4~>xaH172x;O(?
zNTh(*X&f!;i)Jo-qNbb01~znlhM0z#Tn3R;CxK1|35Yie}sbW(}Zo`89_9GPMdN{CWJ1IUMxF|`G3Z41FBW`&Cm-k8R!0Q>IR26!{b0T(?wfOuo}^m}wp4pfakoJ6<1Qz4|KH}Z
zss8@mw_LGv?f>|;Jx7if`5zXl_p6qgSn{cLd7x*$d8O3kFBgOTA5YTLt(-OC%VF_J
zftw)kW0S-_iT?J$wMb9{%BkGSkz~W=w%=_b3{IQ;gw;t1pxZ2{QUp`

diff --git a/doc/src/DecisionTrees/DataFiles/ensembleoverview.png b/doc/src/DecisionTrees/DataFiles/ensembleoverview.png
new file mode 100644
index 0000000000000000000000000000000000000000..dce581ee65b3cb552589a0ad6361ae775a6183ed
GIT binary patch
literal 81932
zcmcG0WmH?;vp2Pe3Q|J~1xj1ANGa|ul;Tj_NpVRbcz_U~XmJX~t$1;F5-3HAI{|__
zL4t$?3%o#|yY9Mo{n!2WemY^!IeYf(y=Ue(atQjUBu#vm>Mj8R0kN#idsPC0o2LW>
zHzaOfzo0+`q*w_Eya{CAzf*Ue+?)n$>ua=h?B3vRCXhtKnEc+lm$bgDj@MpD{HcPd
zuQmX0ZEWZwoSmIVo!NotvNm!PMFn)s+lLZ&{AM5tV(1&ciMU@kM-n77XU)}@vE-wV>6jv7s&dbYx
zBog=;UOjRfc6Ir8wEsx_k1YNp@qhCDUo8Gk=6@0YqZI#<_#dVCkHr6z#s8G||B~YW
zTGIc;;{S(e|Ha~eY~_Cu|DzQDk@z2__>V+_|4WMhF9+lQ5{-z5Me*`v5>;oQ8bdRn
z6K2xZiLBqPZxQ==I@;NNt3T|z@yMrxF;MvqE+k#{vyC9kQ2Ev68@^eM@WVjKG_x4<
zfyb0@GbN%plS`lfc@>)TcegDR+~;RFU0~T``IdKy1ek*M(dJX%A9BU~*L}ua&vLv5
zy}(TSvw!Lf%x9Ihw`dSoYIdXKV(1V95K6X@Qvdj3j3;YPy4?C;bKAWajGLSFz;;z~
z1)tF?%Wc5>0M;Uh>0hIp_OKthm+DCQ$$zv?Cwsl}M!*M6D|e%VpxaH_i9S~vOXQDf
zfhCWJrhiwdjT=BdQUd*R@y0*eE>B3ATOzwy7khkY2Zt|po2WFu{f)ajS7nmEVA$V%
z4EU;=`QyRs@9Li&L>DpU7Y`8#G}xQJ<6K+6Gz$L+eOZ%z_E$A2;bn1*1e-T=`EOi4
z7QpU)X+^xRwcm>M#{ZpW^XvKN&ihXGnKYlmQ!j1Ar{=+7;h!xf`q=)?y=pc!Crld!
z>DC*f6rueXGA>UWv@Z5+KWVkVlIcosy&pVs!!x#X61g}($`Af6DE=4-N#|)?Abi_%yH7QJl!u>UWa+f_VC~Z^Z9Z+i{0%^}Eg6>bYw?v61Rq*rP
zhApWhM1&4AWL0F(wrOqN$slz#oe$rhfsT4bHJjU@dp)%Y0e3Cc!<>n4N$(-(tmf=x
zt~Bz~dkZ=Igq|c+Ia>B%Y;U8pruuhzgkV%h1pHIo-cll9dtz+Bd8&t{hW*~RJM`b*
z~z
zQxWubJdD2|{;XM6n_R%Y&u@b@1b%{EA*~;sK5sdS;3Od}e=6Yop=&QT|A6!C)SZF)
zSf}N$oBBsc4YjqEjZSofc*yoHf>!OC$S6c%&2kI84newbOtz^kVn|)rtmAN?JFWiI
zJfs!)IN5F*|H#5Qfy?>KtNx{-2^Oo@xby3d!>gFeuN$%1Z7;>JAFUCs$b`V(aiR``
z(`T_2Y=#))npWhmUa!8J{0%3;6I<>Q%9qaX^*{mw{wjNL46Fa!I!wxie|MK}*}xg9
zG(YoV*^QlQb*FS<*myC_q=n-tKLfrQzU^=AA1P4Zuuha>8+;l~cvqvibMO%~l!fQz
zIx2{;n(lS>Ky4r=2@aJEV>$oyi)&ttk%4uxt&5asOsm?l7HtDoP~wB;vrm-K10C%{
zzJ|lsPO~0MetB4;Q=>3_Ffy|hU}Z}0RsULMjLWy!V*`$UlOLQS8kHNN6S;}D=$L(T
zWyhQUgu={ugA*_j=ki0^3GFJkQ?$H8MklN{bRV}Axgmb>i}g+dsbD8dH9!Yj&y}fr
zMBM)GdvGG9HAuG>iRXTrOfSPF^qN+~=+505q4PkO+$r~o#UR@NSszBJBRT;q0kd&@
z<5Ez{lY>@B1p2Ue@poSyKYJbLgrv135KN*k_F~kD*ly)a8Xz8P>zbxj0bbngrh{Qv
z=_H4gw8<;LdWI(uQc}R;)83}ShONM3Yz=SVu(IdE0;KQs*{VUWso|AJdAn96&%xn&
zjRSA+Em#?0c$7CP%Q`czUp1$Hj`dVY2n-zH*G|f3-yQqxei|6p(@yIN?#EiFT7a`=
z<{sNSNFs%2U#W?NMl+Mb>lcxIf;^C!V{EF1d|KK9LlrOPpDB^v8WUr)qqq_};m=$e
z`xJCp{q?UH>P7cE`Xl!Gxo?7oLD%cMrrbmJVtA=N^Nj9^C)p`posOMv2oivA&GXUT
zu0DnFVTr0E(CLmUY*|}saOJs8hbtHi`2E%keOv7%rw~?A3eiU9QH#|X*3Sqe;~~c2
zUdQB&H=ZnR$B5f6QK*r`375D>d1sQ#7r!Nm$1c9wl55J5^h)q-=Lt~cL*)tY~
zojg0u|K$8*!r?{+IcI6vBr#nR*mYrRaUm=X_OQb+MgW^Xxwj?0?dc4btkR2SlHq2uJA1$bck^zk;ei^guF$pK`=Gz;oaCyzXI*y5g7gJ11^ViIce2Us4m|&h5*uXabNSPE>t%Aev=kF9s89=#3H+;bPDAry9}FJ}@a(9T8ox
zTURSb%Q#xk~sB{L)z;4S_Lt+?rw*K?ylhPtaLQbF#aiAvZ(^w$I|Tv
zIywnjV^gfS(6HSDz8J&!eJD*zfyJG_Ku0DZL@0!ijkUp3(qiH&W_)i~HbxU%t9S{qv4{gI1bdvu{v@|vbpp7j-p!Bc
zeb&o>_P~pm?@Q=R59}6ampO)+Ub|+Zn!%L7WN6dN$(c=Kn)Li?mtVvff0Jpufl2aC=C3k$GtZVij@{sAG}71mDar4gH*SJkc-(h!C!7YOU}
zus>!;7DLcGU&w86Z7@d!&P&+1(W~OaplNqG>s}gswIGfpHqNRDY9J>yTDA!?JX~2n
z4qxw{J>Ny^-QACw&K(=05#Bq#G24_?+EfxhUCcz{u1k2aUz8ug9;L0c1q{SQ-|xg^
zWGD@vI5~h4n(nhZ2j49)d2v1WeXZMV`Vn@ibuvO?eA64NAGr!VJQ_}3FC!Jm*I3^B
zW=ISo7xGbsB8h)$1xUryHS#p1))@#D{aw2=8u~{O<&*a?@+8>XcGk)bz3egByo9gZ
zKi_xz0dV-5l-tlMb=%C%M$;AwpSE=_(QgPhi1>@S71v;qADSZrXXq?%7Sw(tyMznP
zn5l)^rPTUY7Mp;w`d2_%A>wAkxG94GhAI4Vw-B+WG%}YiPw}
zcK^@wX3u{eKzw<93va`02F>44>-ChSQ(_IjSmqG3<=tjaJ9qBO8N7y1;|z_QmAox1
zrR3?0u8seDw657TE)A3fA_RowY-M0+V+5Oo7n8bBhjnXBK6t~CAU#H&(U9g*yo@g<
zS~^lID_5PFY?Q3mESZMnN@qU-#3r$-296r~1LOA?3$u8SUtqzjr|i6Mrr2VcNziVm
zF_|ZpRDZ#4Ln;2SpK!mrGkN-g+w5RE{`g(0wrTVnkkiww8WbnNVTvY}6Kb(r!6I&h6Hda@ZV7piIvrm{8(j5D6@r
z6U((C(;bvN`it&g>q6#!pLQ+6_kLW2cUD|J;WO}C5!i3wMvYAXw8!oN$i!t=h~B4z
zg-62tE30C!^*)v;SNIDgDMTn|2wby
z85*OfeK}t12le+D4aw+4sv+AMLB~TGgH*HD8q<1()hw{FN&O5&-RIPt+<#Z-St}C~
zP;A8-K|%=~Y%M4XrQPh=__WJ1j?U02Q)_8FHGN3$_hC>y#-JRcHFGRfVbzG|NPO^g
z;ibBB)k#gPsWzu3pRYA(8dh2_V3=scDwAJI|>apm~hFB16
zMdIaqLw;d*3GV{++ln^5gEw^2o&Xp1bW}tv`6JT*@cNUcvM+nw$t6!-I@hW#|TIh~F^LY2^ga=jG1huPK|Gqf)Q~cci
zZ0If1s|xkU9tR}=KwiaD*R!LI*ubh~1+!mzUj0<~wTwFRgW;?gt9HU}cm3@8wDa1P
z33Zb@5)cd;jo;1ENf7<$ftz@EkPOe&L7J*asvS)OJy$iC0h~vYg%CtFyWrVaCQ=iO
zuX}d5(v;fpW+V?+Qc!W-cfOaJzZ9gqEc56>K7#53Tefqjj#m<2$+$rkiLQa(a6Fmd
z%7hntU14A=lDsvFaSDIBkVO0_4)L8Tu-hU$wSb>|=qQRK;i)B?+-Ux55@i|B`A(^^
zMCLR{u6;8Z-)u~l`
z>_FP;y|x1MTB@u6{q#(byT{~n*aSs7Y5jQ}1C((FD=fADc8>eYaMERxQg0no59Sjs
z_*kV7oFEJwse`rVzgs)2WVUzo-R8l9-9Hq2Se~YPWo?N@
z!eN3|+4GgN1AhBZENII?3TW5U@^mUemo1tVeE6Ptb=RLJe(wrLcWH;Fd5%>
zIRf6#i{g~IDMTwBVQ@Wy7@^K=-LOa*{{bGJG1;lpX@xy#&a*xBpg~BC;oNG_jU$JV6?u$LcQ#9Go_JJ}LbC#5t@!aw
zkv`v%SAR?KDr>PhPD-e?8cVn9f&=ib7SH5RW8`m(YV|fXJ2B7dS7Xc2fY<)x{Vrek
zDLg&h6r>k|X}oMk(nXo5#YIPI9jc{Ip5d12W{%NpYUtC}!S)(wWW*+yUxkJvc|xp%
z6VCTsY1YIwO`~qPBeTiNrL79SMR!9QAZFcBqN5od@SgSY1gPlfckx)YH6i5Y82?Ho
ztu^F4S7SC57?IPP4G`b$M$J%jzrd%5V&E&~W>nPh!s^gsGS#4gk*#BN4XGJr>QaOA
z3dPWK_j
z^Y_0&9Tj1|OWf^j)je+AOw^D--q8j=O4`Lor*R3MH1aC3P;0qH8md{y#wWuLh-^zn
zwUo97_E%g$D>C+OH(#O=GA$?q48_Y;wC5P315Cukq0=bdf+;tUI_=}pxFf98eW
zCfQmuCnX8x@qDC64c5E4c>@A3Sa*FDa#z8K^e^JV%BR~xA&cBXDK7S1dxP|g{h!S~
z_%kK7Oa)bvb`>h99*@pv=8(>B-Ptm2=j6@Nr*yV0of!Fb=blFxy5bGOi3k(Q!y#2>
zgruI1`xb-sU7zW7u<$C-5Zyt}Rx=ld3s&{n;RzInDQsrEn$93`)J`l)#hkCjh&l~y
z^#NgAJLfO2ZvqJ>TtbUb3cIf)o*#!$a7U=!MN6OR;9Ae
znT6Um$Bc}|0K0;}ph-aLMEUbX3~WY+wSoTNT3)7l^7=@@7$xL>RZLr$s6v(stN&W<
zO^r-`bt|=%Z*5
z8mi;j8JyC(u~zP^vlzo6t==$(IL#GSDOE^Z?dH@iQqVaH-5!ps)gw3fyvd6AKC%?;
ze&-T4d5h-T6rWlmz5MPsCE3fDg8gi`i8bIVh{w3f!m(tbq8%G_?1;5S0Ba1*dA8|{
z>2Jr_>zNe)fDH5)->Vz1*7ZJFpH0(U=A9%Tw1A7Wkq@UC3U4SoUq_l3p5pr2v-7sV
zx1*0!_D6^p!_Dc6A9%{zFDil|ZnSYe>-w;3}p64x>29on8f
zT{`|}e#&Htz4K?+lRV^~N*VqN<%j24t&VI=%zAiybkU~#RmQNH1n+UUzd7_fwHS`*
z!^?|j>^(zt$r=lV;z%OzK~qKtUd1K6$&)k7A8N<=#s&8CDsZ{=)D_gYEAgZv^}Y5G
ziz^J4is;0>yVXy%l2N3DxuU&KYI6`Py1mH|Kw)cP+e>ose@4FBx}E1T4fAfw{$+di
zSlz5-`aJzUNmH6vg#HT4CgTD5*vMT;P*J2Jr6{`FK(AhzThOd}ouq#`8=|0L&3-B-
zFA7#aNvMH82(e%kd;+S@5{$=YyY+J4RSwn$NmizQkKgzt5@pTw`
zSYB0n3MBplA06P2A0nA<^PE+Gv_Jep2;K@JiKkc~n`3*g<~SdeFIr{olF1z*O@8U3
zG?{sWYxu-6N3F;i0a1^|CW=w9sSp_}4>}=iPJ+^OjvWlu^?sN;sLRDXAq~=K;pnN+
zp@(DP@8=V{*x5CNT*wbnZ^K`C|{YhUtV@RE;tH7qI(f1t`1iNggAvL3a{9gmW3
zNKd;O$#}Re$EFAZ`Bn=j0CEde6n}#VO%M3~LGW?L>)Cku+?m$|^Om19ip54so$Axk
zIuGslW6mAVwr6^sX=>#5?!eDd?-g@r!;T*T`3u-l72MF(syp%qY$1hz@Y$8}Rq|=G
zs?cA?zdQStWO0S3h;w)oxT!7EQj#eGzUQ;+xO_lOdvavve-Kc*J2h|kytHjAM)4W|
z#gB71=Wi;szCU#M{Zms$#&i`vJ_k*OI$4Y0^XL&X9OapqwhP1j?2d!G)ft6qWRh#?h}~R8qVL;YP8PG|YCg_C
z4$9k*-U%zB3I>S_o+o`k<_flrAYpw?wuk+p8??bO+TqbM*%GWVw<(zF8XCMz4|}A@
zs8f%RJ5^nZ{tLt0b7mi%8z{1bxkM{`7y91
z7qvGdE{d73n2#rWi`0#DG6rjAGj;jJKP|U9ctm4nEqWE5$9@tyli+_dfCF6Eq{_Bd
zg=hQFjUJPu&Qm_JHHL}Frk4CAU-JeRraiD_^s7G*8&;RKspdHZh8O74#1iKwT7K%R
ztd*wrYU%jAmu^T-UQ+rb_hvUS;*hmFKZ2&Dx7vu0lP&=Q@qU$8wVohkIC$-mMB{a2
z+E~G7YgW#@NrTDXp~dANLG-A{3`J1&XCJ?==t{|+%UhO&HfZT7qLj1r0JAeW>v^YQ
zaY)0}U|lK5tV=UswsF%#GKXDY=T9btADlMdVkUmh(K_%%AJ@i2aXuiRt@o?JYrN+S
z2tB{@uLQmsiH~NtAAF0j7o
zq$_vWobb)C)?$CK>Y;S*g|^PUbk2GADS8Yn)+ewUyYw|YqVFv81N&1mo0ce(b0|G`
z(myY*5+4?CC4pOndS~o@wWcWai_-51S}SSXTs@1$Cl_oSs{G3Iof4g5aA3;&_^RL)
zDrB#VFXpEe{gqv@Tlp-qXY*m67M!q(Rzo}AC*irRqzDA+>KamQeG06FTxL-|6h}=)
z1Ge1sZrk@Ji$fZY2A?(WNM2QhJ}`%A;<4>s`BnK7nf+<^MS;&kR-z35b`m7(0MO9f
zT1L)+SDV2JH6))q-Bme1@e9D7TnzeRux#N5^&9v%0XdNV&W{jvzEXcjo7Vx9Nl&k2
zS{uo-bOr4@O02~x!5O>6jhef`_72(KxW11sx&KfCE)CVxh01yP+CMKzN@S??7x
zAGo*t9q(P&MhJZTAmIK2O3(`vLr5~Dn$ynyBaiEQPHXniZPE0r((?;{&0Mz&sLC2<
z{Gla-rEH%N^v?dBiR$y5T#C({Uap&}&vQF&g{9N{TQ&O-48p+xA7i%A56T_JXnbLX
z1uIkQY|iPuT;s{v99q3~(Joi1luudRtfHsDV=8IF4r=l#PU=&=>41~=sq(dXtOa&&
zfmfT8zLhc+0*>Iz3;*9Ntz47r*1O=u~WZaTVxE7K07WAIIN
zx+~7-Oq!oCAsQH|rR~rUJRMV!QgVl0vk$(gz%YY75oG$N
z>doy#t#bh!K8+8pSE4_=f5@{4P$b(cjS!lyJ0Jh%UC2na&u1dj?EK9DGkchmb}LHS>EGLpze
zz#7&kG&S~H*Cr~Y)1-a>HV}O}Jlq|~y+~rsY@ZdZoF$AAT#eMc^G)kp-S%9Egutoi
z8El(_mgn~0*#_Lnu0Kdo>_<6Gwk^go1t|S^M?fiKb)5j4mr-}{$?70zGN3C~)Bi&O
zV)$igjIVE`EWH!M#p-?NKFL|s==?bBO@
z+q=TzVlCZm2{~&SIkIK$U~F}vDs1i(!3`%2v_8rpiAQw_j>1LI0?U^
z!b@D@Lv%C{XH_Ybc-{9stNkf*h($6Qa^Q{+5f^i9d}*V<6pHMJhPD%44v>qA$&Yg}
z8|Vo6wYdEf2K?}5RZxqcW&7O2BGX6AkcfjcVZTDF2Sa_jpX{Y{a=HOSyVf6{*&G3B
z-*7hAME;&3;DSUy(rL$d2
zwGU=)bCxL2MrpLPd(*!LsjtydWD}vD-_3VOJ=*TqO~FR2iVO021VgJO-JrXqfnwdP
zy`Q3+38=`TJm&q#-WeM4NTpqx;S+&pmbMJL#MRU6JL1
znpN10EG;DT6&*pc8}e?~9xBn2km#CL-pd&Tu(kMXvNBp)9ymetng=6MD<~Xx4@aG~
zCqBK48eG7qiw)MWdB`j3_TZ_!-j(>EoQY+i&B
z@Kw`Cs_6InD^e~fDt-0)#VokB)K`=sF2kV2&*3^CYaD?<#UUuX$~F{f?*zQdIX6Zm
zUD)rX%ZfKdYhtSF-L-(vCe)n;0Ci*!pSMwI=V;A{N9}{Y{8>0^_T}xfI*Wt>#48p)D}K%Sk)jCcyG09|QyU;4>%T_ZKGJz>xamaU0Q!
zZr8UD0@O8ij&B*?+PXZN{e%logKtQdm)Psx>NB@4LKTIbk6eJ2cD5%Jg43x5x})0d
zO?*UKj9N-Bu1sOaoIpd-1YXl>tMe23Yj^If@v4S~^)pZh3mqP#Cip+V%
z(xwfWGd}r^yJMSf^0&*Z`wV8bJuT#`>Y5*3+G}rahi78f{leYP%Et}c5gIWi)5*KC
zK`;wI*hJEUq%yPjo|@Am~S~CxiwSB9`sH
z@Lp4zZq*(-FR&3y{E%6ee6a^}PwI_ViZ*S2>mJ5*p9$*B#3}}d!w(nI4JvP6O!o`DQtYX
zEb#@@b*iN2?ZWA4Icz$Jbrp@=mUDW_@x=9NoUb`}cv}#&zdL0MfGVu}-PZDCE7^H9
z%VWfvgr6ndlpM=<%K_5hjX;U;=tUokK=)%vt-Fx$-J;_3GWr)K3;WSyua@UGAhMN#
z(E}^1ai^M@K)|n^w%u#Lio+$=@(X~~bSykXHGQLtiFb@Hy1I%V(pPppV(Jzb*Vke6
zQv(z{q<(C-{nMKngs6VvUtCpoV5!np(gEx`E>2(EEK^x+bBELm+|4d8t=-$WvCPzy
zLrQYD)>z?VR!vAH?5ru9%vLe
zGCg>VQL+WIE9pE&zt|Ac8VI$&!(TN9uQJD3@EUX&h=7K8zu9!C2fv;UIGEHbd7AbN
z%?5b?Pj8BV`RKRb2oUpnYH^7msJ$vvls$?6G^36k0QIVsV)rXJ;N%csTUo#NsKtP$
z;Vkhr-*MtTParO5x}u`IgUXg1T}(-_ZC%#7zryT!=5pf5G%8VxcW$e6^f5I+r9th^
zxAX$n;DF_Wwlwr{+Wm$62`@}EVL6hd`m{3e9GtQLZs6LrSGKyfBF&W04rT3Sk3Fg7
z$g}179|SjgE}GaraY`>Vx)M(S-=nSFk%l%~JZ3|!=j`i=uELI5J!{|++2Acp^`yHa
zfnmp{ZirT3m{N8%EZ+cNpaw$3Ne4HrA4(&#_)40F0VS#6Z!xXFC6wcEe!{(M11+8G
zty(G-@`S^K-bd^YOH>ZW?a_S)yh&w6k-*r#jVM=PnJJD2*LI@*ZGb*f-=rh#-0o;?
zI|v~oWQ$tjjQxP>g;VWQ72{!23%pS&;9U&amun?6LjBKosTdTJPlsOe!vtc`*?^><
z7Xg69RRG|A+LC9VcWOeN6?atg{M%uCM1J8ZX1%HasF=BIl3PoJpEWyE*+ttty9B_^
zw@d7@i+FkXtZ(*CMme*un53L49;mT|
zuf%M=Dg=$lEoA;)iNJD)@9OgYhTz90x8;W_0TOJ8yMvl=M&aBgzac9JiLXsuOM%J?
zy5(jYLMFbgVJUVt3KoqcJR{pZc@I<1)QWO!Ve60}FAm-;$J+d5JD1&bx$KA%+=$=K
zCzHg3a9MX7>n$4yRXyUP`OkDsdx46=nO>lIXvD7FN6+D|&K#FltjwL(-$UhzcKP|V
zhAIa;Pe@IP_Q;Q?&q$Ru&5qn|0t8P;un2A}INeSDU3ZKa%}Mr%2J3rL1K8j%3dLZ#
zJuuF)37cM3h}!vbI=hzyfqk4|5Oj93PF9YTN4kt_YnNL%VsrjUkJC;X8+IJa8j#8R
z$g7T?P}3QK`V7p4L^Uo;Ait~aK;bcR#}Q$=JbXN3$WaYTy`;o$CGv}&(=nu`2P^RA
zh5sa4vKR19pG0f}9@_}X2{hWA0I=ZJT-b6II#R*yl7U<{+K;!AZ^qZ`?RsPwEbXneGUE6Z$yxO9P}$I{-yG;
z^|~iA*~E>csy<)mg&(ShCNNEoJESV1tv{!uq(S6Yq}1<4N#B~WSY6?tLze#P&v~k(
zyZ|ZiQM>cP117^-d#r$)bbd1b%JU76Ot{^io6BfrXKMJE
zrkumMn|%(MC2Jr2p!=zb;@!4(%tmZa6CeCV-b)4Am_`dg{V0_t+0@DIgWt^kwOj!G
z$2M;7$1&@wU{Ocu`Rc|wuC``2PA#>6NMgov{&Ht^?zhSDW@Bj>_
z^;Q)XS`OakA9gWX#?D_e;-3*Y@geAz?_d1Q)R$mqU^WOopFaIkb?@;MMbBwR`S|N@
zog9w%#yx~0h;>Aw)~TpJ>$11jYIeW!QHU6{wifv1L0pz0#Vl=H%!5ACv+8ya{k`#X
zJzk@XkYysZ4}emA1GvqJIN~t{Nh{U-X7gnj|0AjN@TlFUJh
z*&46m>fHJj{!ul$)B2sNUh*Ks4}sOdH;a!nOgTJ7*^1efHCN8hZkR0CfC{Dg$egNy
zel@#H-tl=k6>)m23=l)1O|?MXtV#Au0%nmnyr8Oofu)uxKWu*P`EJx!pDhacH7FkUY}FiGAwLTi!r%
z35A;5yIU3-MZ@MlRM;+dZa>ya(Y!N
zW5uv5{kUviT*3HIPM=UUw(_IR)~vXorm|hENAm5$pVWJDAd
zICAv=4CIX1Y7~;xNvpFzk88>S^_775NF0^#Dhl-SE(Fh1+`dT5l*akqAi$bfSFVq%
z3qjhMR|7aA`Umgy-$$Qquc$7*tE}i(2b6W3(*A+d&iL-*-e;<^R7%KXIapNC9UDN+
zZwi0}^H8v-q*^)|C6LIePO0135e4p@9y#*{K|J())oa0
zb?M6*<=EPd!Z*zeH2!qH0?83af=xdsu=9>2c}CpN&uF)`;4JVr-{G%Inb^RkjI%j9
zG+bcT@bHIc6~ty+S;f1l+iys%^h)pH(q23F_#F1JOIHAAPwsX*>s)j%V~EYRy)XwR
zG^@l*qnp#Nr+dSRxVL@G{l!%3}LxOeiA
zzR#K9u#-*Gq%AJ6*9b9Mn9`;LS#)@olxLyGUWbMq@C%3NnQW9bssa^qCs$4bXjfE$
zWD-uv%=ue#-IOfn9j!2n}+Z1XP)jWRj0YnLCk^!?WvZ3X@Na}
z)jvf+P)L7%cE9p&6jmIAYVg7kHQD>@y4%_G9$UDaTZ&0ZwsrJ-RPzeHWNxTjs2sa}
z5!6st2mV7KtcQ!t9-LJFtkais?6;3>h|In=2%32kAFHFnpx?oCGhrLVhPr3?it5kF
zOF`KbKR7#vZ$#_-V6m4J(C{g}gucstx{-M<4$wf`81O#AeKAVlXq_h$UShOGr=D}j
zn&psJe7y|8;ZU_f
z=NunWgc!BQvgCY}*z*#0r1@9CYDG!}&C%2(+pYENTe)=~uHru<$7jXJDo-53Iiur*
zB9APoo2;5_`3W$Uk->Jc=`Y167ceO{b7Wn_9-x1%A&+#
z&8z28Nj}5YdnqL|wJ_C!k*{_89ZB0_)7`&ndPTEyY)v&3f#;4t3p!kxz7A^gT8iy#)(ZV3_p(qnH^t=ztDE)8Abe*3y*X)lq=$zQ*Ak4H|Xlb1K$EZ$>h
z#G70)*r)1i`(f5+G><%p^&$~d{t>k2Yke3FwWO2-y8BhGUu;5d@84N-9^O{&+XDLZ
zoh;(n((vD1zF64e`A+*Od9V+wdvVy}3KQV8b2l50M*;5;|pd}3(?AD
z)%@i~n`KevTw{^DS&u$I*19+96;__mL~l+y7fCd_xf@yoZYA07PUg}%06*9nbg@&Q)Jy8+;}w!|CYm!F6?e}m;iDM
zbCeJR#Gv!@1{=aTj+Ow$a!UvaLb8a)%0P*AgyMRp|>%I8(2HaM)v*yHTh4I5_n$fPDG+gS`^rWE3q-)c{
z99Izk675-e?Y7+!OPmXeYHxp7_l2;MdTgK}g$rhVAv`>{5BWY%J=8%m!t25TL3V84641H`CO%&5x`A^dyd!9Z3rHvhNn7Bpo!w@j$i60mII;nMI^R58`6r)^cK$?S7S
znA$4gdf!pwfWU`uhIxrBAn|yogNAMJz2XgluOn7|N*7H9>WN*6i3V0ohTZF%8ryZ1
zo=-G_9!0
z=FHA&`MR?g>}bE~K9jA4DAbJOtNy;xSMH)j6YV4$
z-=Kb5O1@XP!65=Qjv2^Y9a~L>m~}+^Wf`WEPTxPyJwEoK^4EQSaNt0?8|C~|()CCx
zL}VChB{C&M%AUx^$&9z0P9H($grW9sj#bw#GRNGW-(-txXnP;P$==-nbT)SZWT%8H
zLhfxax;;rHVv7icx8&{YV^dCRLWA{;TZo5S6%w)XP6w$&=I3wEvj}S5S$;au6P7pU
zAmR0hy!F26G;SxKDyFAfaRw=AJ=%X3%fB%cAMRp*(6B%02(q?jwHb1EtvhK*q(N8#
z&+*Wp=`-G9tNcUL%g!RLo_n+z}I6cj!0q;>VZ9=;%Mo
zAX?3UD1N1b(J$7r-(&t+`1nphph{=^`v`n}&PLsf@<_5f<%W9womG&UBDS6Kaa%_X
zzNG8b9-n&-)7}iZmmu?s9`td9^IRjq{CRbnz8>_7F;sVSnU`gkbXbj6j|X2T7({U_
zlqS0If|_DseqoK--pMHrA|5){gQXECkH@+Ffi*@M`?O0ROI?0+kWp1t
zj{iEb}1_wEk+DzDK@4v
zTXd!h!>IGoZ)};=zTUs6Mpfr0GQICjjCQ>SK|XOm(A1M!=TFEf>bDtk`?8&*LKb!_
zXE-`D8uYcw#-7ZqLjmb7vk+YDcvO6h%xL2n+b}^@hXE)lMxBfIIh;255cARNlYLZ%
zqp*PaK`#%}BkDLRE!4=^8rzZXMSFU4`a3|$+d3e6)CIh+D`d7Hxl6l37=b+*+UQYQ
z7%FI+z(=I``n;P#PMVsTvw1>h`xZWgRrx$FVz+vdfP=-UeAe8APw5Wdtg4c$xSo=x
zE6A0xxT8^FwAJ2La#}toF3x_~PwEIboX^y
z#RhZ1w(2xJo47VvyorTF1V
zYemTn&{Df9RZrpJ_Bz{$C?ZP>dFGJ54)rt}s{o#746kq3&(WE*Fp?}nxB5sq%U4mi
z+C}ky3d4oXEzodWc41GkT5aKJ2U$UAUis$YDE}4@{^Jq^xi23Z-_aD?p0QA`Thuz*
z{abwCcf%rG7IvIFZAxpru9b;4fw%J%T=Dngsv_jkDLVt*bkn@rV(t@C{e3&Xqw!eQ
zu;`{-HISpTUhKzeb>4dkC3>qSW#!=6BUR0_^}=s7$7tSJM;R+T!PkrbaiN#|M+_S^
z@x=cPX%i%;P1@=7K}=5&?quO@_H>EgitiSrDY{eSWqp@9Sq+LZU1_(*b}PMnQqrti
z5jo`kQ;%-x=Y8cxuD}Jvbk$gaQJ31}%GB|U+(wdIOtw&pRb}Aa0#avc-4fMN+-rHL
zys~{wW^+dD*i2$g>2YM0lvq=dGESpdgsHi7@gL=ouoFq6>S?|+#D5H276zK{Cd%-cV~1(
zxxV-hbZLJMSNxTC+FJ4A^Dj5i;_8*=YtEY^&(O99;|JT}(-A$W)9j(%-r3K^(~!rR
z76C3P1-&bkryTVe%${$3-360{WpoftlrA*tA_MqOSc+St!U-)!Mi(@;_n5;*d14HFi;QW#tx#tU^aVEBJb
zePvV}UDGXugpdR$xO;GS3l718ySonV8Z20Fx8UwBgS!TIcMmc+!=1_Vu64hE^JBXE
zbe)o2yXsWGy_W-B-^5C1I=8AtwKiAQh}il**5BHj#^=7yjGA1fg3r%?wYC-x*m7jP
z8(J0Ee`w+Pr0d`U>NeB$3V>w`Y=_Gv4EnRc$w}~~gdLGMW;x5@6~DWPYZ82iVs6%L
zhbFXSN}0$9Pv
z2cJ{8cpSzhi?)-8zG!V+cB6Y$<+3w>r&0+I=asvdX)a+`bx|W?H6e-q3q9FpA+)Bk
zw5p`3ULVH!o&=UT^F7)FN7T*I8fn
z_OFr@WdEYLp^H@5mRk-@R!m-$>B<{Mo%Yi$RPLbs0wJYRLB>D)#?U2hWQu*UvpHQw
zjke}W1S)(*}f0UJnUm?K%eV_2AYxnT`sL*6QnFxK+v~QEg2CsN$QFd}dGym_G
z3{G9!?kdf~1VO-!MX$Xnq-8{LEs$-;*u<>p&AWT~h$u#yF^^{pkmWoUxw_2OiDm5F
zue;;>!&6R97BidN#1Eg?kYn6Kvs?F%qZ@$~K^lT*g|RX2D@#6uafR)j>5^yTZC
z8f#VWxUkf0=*z1cGrwGUrdH&7ng=Jo?lIZPAg6r$I1-h^Ut4GwbP#S4jO;5+x&3l8Ot?(0t31#z|St}F{oo?R}Xf6nu
zP8{j9N}`WcWuQt-#f3L<^67#AWfjIFJcDbO45ZI;sC
zBje!q8iBSOfM+_G(!yEf_zSxnX4UBkpOI9lqzstfPe(~I(j^^_VokU5$j?@I|4e0J
z`dQnRZKEGyi?-z1btCz8ay_$)!(smw&d-q=HO~Isn-5$+5sLlalNr&eRV>dm9^d|1
zYixXJSX*B?v$Jh`+OTbE7`$D#b89$jbZu)$p4X;Z&)tS4Be}!06Okerq2F{#WZ4&N
zz#$c>ECI%&E~Ad}X$UC0&AGl}9tJ6m=yNL?y6v~_>^wiZ#lLBd_tRHU_F4I(Z5@FK
zBTMFthq+xdXfVqZ*+!$sY3=x;FFNtbKOY;WE4NUW@N|zSYr)^=0_1dnwGULZ^V_$o
zep^*D^R*nMb>TK4A1x4s8m5G4;mOo(;OEj%;`JcPu#ub>J8+94uhMv%HM_K^VToA1
z+I2c?ic;5LCs8nGd;y1J>uvTfbX}x~g;6#8;o1#_4VNA(5Yl;s+p5ntUUfu|HKWwU
zI(M5kCs25&0#7(t0|Q;rfat(eQF#W9FWFruVtGN0U{PkRw`HvcL0Aki+2(SGsBvMR
z(0uB(9&kq+y#;c%C;p_@u=c~$k#x9F`pw9P3Y_`f7#gX+%(^vxrLpf77Oi6ow-PQ^
zYQl%l6Bz>MxOndAoj-;LRZD8MX=^@D+zvlxXGZ{taM-9Ydc`ax;&T)=`94MzY1Zy#
zl{Y`RxG9aBxSCG@;}1HN&*JQ#_MD@NrX;S?3FSd=NFalANK6zQ8p2`>n4-Od`PMhh
z{koK>s~%e$74+4fC6$V(F54!Ts0KoVXE9Z8Xg3|ybq+INm|)`t|DqWxFkEpaCb_2r
z?o8>kh&)f-gBv6*ZsFKW=e2)EHJtH|syc}>2uaNe9KuCj2|H!!{)t4z7(d@)Gmw_>6A>sZiR=7WDGczfxWE-X8sA|6fcB+5fx;`gk*o@=TKXyG
z>^L+61#`;FQ`SO(6Q^Fho}2M=vb>{UXLdm!sDi@et3vAPfc%-E+`pSXs!``)7isgf
zHm*kF?>_D~SD#WZ(UPi^3KyIiOR5EJ`d8F%ugW)bbek&5d8SCUEqwo=DU&?-ak(F;
zH#MS9%#sjmzYgc;`KI@9k6l&1&}=Txp0hq>Y4Gsewav;2?To%3Ks4P}b<-i5F6DMl@Ukrdw&Ye)jFh$rgR%j@N#NW2{_}~!lm0{b5AkMz5)Dd
zml*}looZW8`fH{$?{8pZAVf$0d`{6X)j13iR`MT%((!U|H~rF0eEi401h(Ek`v!TH
zH1%&qGw5zqdDdfU#wob;Ltf$7vZ<$lIWJ*jcsqg3f9FKY5%hB2bUOd*C-g71B
z-1{g>ruZ{m8q^^uTC*3N5mtnB*y3w>P|vIcEYwfxh*lC(d%{xDp>V#gzf%CVodgr|
z=u?G^xtxhl*o_96W5Gg9#(!8Az)nU_BkG@9^S%2`W5>0u
zR_#Sl066Bu5SokLDgM|vTjr%`q`6odvMu^iwZ$=J9;utA#FCKR&{{yvls@zXNZcLC!7h+<$Tn)wR?De#&1W4RVJ~>rFwJ=BMs6nNo{KtUCBHg3
z<%y;sJIUHArVB1PViIaM({gyIx>efd{~R`8S4x0zY#x=l`$Yp2hKX*f%hs$YXNzw6;>M{qeR##NA0Ew@OZ&--P|}OpCj1GX@vLf+kcA-GG&eRDc}f-!Ot_D
z*w?fpp|X4I+=kDxv3c5c4m?{+EluKpg-u!@jhzwv+>{A8yTSTs9Mmq1KK61Du}`0c
zHB}Uk&6_EW5I?cWteK@__p_6`uHm+ddgLTTuLF>g>2YSCV98za&TR>H&@kQjABUO`-g*|x67=0Bj>}C
z1W$Olh)?G+4G96PcqP8-s$QP&1-L}{Gp>N@;cQ(ajs(F!miVmkT+en+;@`O$HjRGI
z%>0-bC|qDQT9pn{bu=SGUU`1&qitE?jY0o1XiuIc2NhlE|yeW1Au)HJHeSX
z7&D^o;`+WtwCKWK#r5}Pqj>#ceXm_bg{wac7TpI8K~;kDvu@+GCUH|qJMou^=fYB7
zN^xIbH*wFJu$ub(#{()jw1*c#4vKGGWC@MUI^d}#Pw|T@OSI`R8eeqSP*cH>ssygn
z$z)<>ME9@;sa+p+qwhb2vIi+VD;i4*&cuAjC*^ogRJwCXvLtrUOo4u>N0jXXBBCk#
z`oD@N;0nI=i#>c7dCpCq)M1OQ&tHU=!xuXz`N5
zMwY8&zIryi*+i6|lJDLW6C*aN<3b#}%o(T7>rZoaag7Z{;n&57^|zj6Gqp!$Oez`0
zHniaVU7>Snzc*gX6yKZn+6siOcHWKsf)}S@e1V1F3uePJPU4SdLErc_?yrty*8k*E
zbtK7z^w-j)Q<92bCDD!iLWWzLj3B1D{cu??fSRH+#58G*S`wChy|=ee-{9yCIjRvC
z7bi4^NVI!zKCsHu_HvzMN7oQFt>`)GfTqrQHY4SF!Birj2yz^&U#eTz`GRZF5;xRb
zap$%Ader}>EjGGhyj}G8UM8C9gFbz`&Y%5z=}ior%n6O;%Ts+n_Zy9^^WDEU+?K4<
z>%+fQ6Ia5&z{Kgnq;#e6vh%peA@ZI*{v{#6~jrE-vVmHED0?@IhXtXAYO2(aZe)3r>
zYGk|*T$}O*f8Dce`|#s1W=|RxP0(9NFkKok;hUpBx}1<;90b{-CPD@adveu%pN>dcrsV|cm>_p_Ks
zu6k~;W$=&qrPEsb6)k)_HH#q8@nR^%!!7Ajjnd;v!LSHKEV+48H;?Pz+EHtepgkUlM@^qcb!hR+6iR{_cg$Qj3_C&0nv
zN!H1nQd*rSTTmaEnQaHar|;XI51yVkzIfelJ*aVMn_uOLUL&$tG5}mZUyQdAf|9(o
z_g}TCIFqLofHx0Et(86`;L3Lj`JthAPwJT|pPL*T3T;j%?}$&`{2qE*6xNBx&Xz|g
zTbFZbm`n>-ZWB^s5`L8uuBe(AE?+0H_Qfc*?A3f+5qd>iNO2N-deY_;b6d4$SWkKefut8R}a801u&kxdi
zDc*jfdv1W!r>_AONYaq*ebRj0A|XXx1#${Do#W=Wj(T;q%Og<@iTs>Q%PVX#qc^|V
zG!b(bPViHSrfg~
zR@b@>M|TRb?&TC*2kmp&T&~S5nlphekXv8Y{0LvwY&~q7%b7OD@Au0V6Rw-yoyox
z$H@%=$JX8NO74$i)$FG^j4B&s)B&v5GQP(NH@fE0H=}S#)Daae{*}rzpcfKNzm<)d
zc6hxnT;Ahn+FC-|$)X@$lu|Yegrjf5JaKbO!GFD|aXxa)HSDd7pb8WmJxJg=M0J;K
zJYa6jOqB~OpZkd~G_!aKji0Xo6bF*6bKejy^ncVzc}Q|x8DGH^*fZYV@U@fQ{#t<%
zx_Wzj24(_ywNze&8A2dFO#EcPG+t&mH`^bsp;5T0=2CgxgZ{^fh&(m$awAsE?XMBo
zC@EBFS@RF6#H9OLl31Cv&8G%F4cQr$*~z#D9vDQ&a*T5z8VB#eFs3;l^NeGLddAe{
zj-q-${9i;&ien(ORQ*pa^de))Jj@li+`R{Po#8a6JZGn4MQoRT$T!$;E7GpJ)Cbp}
zt`6H2VQ1Ri$xbaR_hppxNTnr2LzVQ}P$}7_y+@Z8oIOhRh#w63#w`gl1C*UMx{>F~4RR
zXZrMUu-<3E8+D+4Rn~HCRXT{pi9IzIpu8rY
z!Cb6;0?P4EAGLRVj_psrdb|siW>*(n
zYuqwfu%ke@L2m;2RqHC)OK6dslE}j6e5QsBZ42g?33ZY*6KdFRXg&x0=vqPxpq8Ou
zP?4G+J3nHrX#5)(CE4iG$s^hhN9^r5d`yC+_$A$Dyv@tEkU?`q2=uJovXrzS9u?}g
zSLO9m={270ih7x#$})I|x%K)9D$Frt_1O}TN@O(ih~#F+`}Z)+L^{0AhYQrIx$8$%
zhG@a9PoAC?-OEU|!n(YFGF?6MLQp$6)LpI?VB*(&`r=X!(=%-xjE~BwDg2~CbsY!e
zDI7b!nrvyMqn*p~cc$caa3t&^ZGdiNzvbIM4_C(2m8S|(6JoMDVjk6hziEA|-=j`$
zQO16C_RiKtxA&?Al_Hcr>;{}~c>n<_3GR`G~vs;
z>UWoyTdN~455>+eCSGz9dxM`|&70cH{oJ32o+4)2>ehdoDAfj{l6~}{AAU^7nQL#O
z-r=yWcrw{Be;=PO{4h7)hWsX57IZd|Sns00!m4)y_^2%?kPKcOvv*e(6$Yb0G0*+Cr+iwbzR_W(5;-Ov7$$*9|BM}7(`2{BXl^f3w};@>v7;L6gCK?LqazMZqxn@o_Omtn9AhJ24N(>@
zD@FOGsR&GJL7}Vp9hui30^=OTw&wW0b&HO*vrfzWXiH|}$ZSTo(Wzy;&Mke0)*h=;
zsb6{)M7zkuswnI7bIYoJgL~WxC=0g+%r!V@(+OltRS%^R@)jbMYC2s?ON)hl=OpCg
zgSGS>mE*FF8{*erxK6?r=(~4Ubz>n%yl!fm^3zaj>A!KhSA|gNtRuoyN>OqnuBh7Z
zhsDHDg-XuTYj=FJ%fd{?$8%mAoX@p#eX?ylL=+=0vc(s(tlb?DVwqCuS&<
zF2B5>L-TUz2bfJO?2~lsWxn>7(*zPf$R_C`_YJNwC#Zh0s|N}9XYj2J8m3~_nKtG(
z71j<5GyK};opOp`X0oO7TP#yn)xJKY_PXO*ITs)@Fh$zA?seS9An^A;0CJuwiVQWL
zDx@!;RM)kb6w|G0TaT{xV7*I|3>mo9S;+NW7x@mD+c(HYGz9D%i%4={k>nBXc
z&BP;|uwH4o!Nay|PTJ^0B;2tyPxrQmOhk~FBD%I*$0sY0i#2>t%eEU83EKO`LAR<$qvE_GzoG3*S<=~ifir-D2qPdwdn;5TXCsBj*
zgWR&LJ`rh>WCeE@Xj{G~nu&(r$8{F8)c?N8#MaaE0iPWY>=wDYzB3nH5tqhiwWB<061N8IWkKPfi*{
z`Ip?}i%{8j^Ez<5{!&q|>(fz#hg|>SEHuA2Q7N?OoX*{4!L_XIDoY-yiia`oWA9ea
zX}nRQ6GCRpX7}fOMd(p5^O)7mp{$rnQ*nk6z0|
zFL7RWiL=v*EL}VmvvoD5
zGdOF#retas=%k539OCyjd$+W5zAsCgSj99k^b^PZT>ymN{hghuPWyp0q9?FLHU#7K
zIR?VM@3{=qTD-OvQJhT-or@A=z3$vqd4De`x8?E-3%p*Ou82
z`zSM*XQhZ7MWD!J+bFBByBy=n>H|8&hE6I~cCm%SCvGWfGH^?VvkRz3_88{A0spLU
zNsQGo)Uh|^*A|8Ra%xz{jZMm`baz{A@^(m1+Gx;tNhUtKZ!=P|$;~3D@|w=}_M41Y
zne@%VVdeY&`A4R05O105=c5_pv6b9t8wV=_nMHwMUbijnC)W}0_6nvlaNR+?e9^+h
zZeKVbJzFWqEzS2TB~4=00fnDwFOV#!+kB*~lw^&j=?e{~;X#Ye^LyM*d&qo>$Z$_p
zb34&b5%StNRKn~qt0!N*SCPin>FgV>D*CaBSJdT}#(VcZkx}I8;Z{}df$r*Vk?iR*
z2ETr)qFJm4j)A>V^uMHG(WIj7Gu4^4>q<*V8sE{YXixUf(T7nO%zrNN8Ia-lEbtuB
z(-9u+OVg+=s5=YP_$EmO765Wy)@mNra2+GHj)!vMM!6?k4w;%KYPUIG%X*O5dfr)&
zKlnreM~)C1E3=lL1bD23tzw4u81R93PXIS_naPz7EHblIt^)ml>Y)wbZCQmigH6baau5xJXCY?AXHZAhz9RX)9EDX8gLTMvmsq1(7g0JQZ*?Qnfwj%sl
zSeLHBk;(Uog
zlTSvA9sd0|cXlT5c}AwfXMpRo?R|;HR8hOUYDJa~@#C2j(O=NVVhh@34_`W8L%auj
zvOBTYgH_Ii(?pb;PZ)G|YpeZ@mpYvdOjUlAA$&v83KP$ad(~rybBh;sF@9ddxAEw7
z=SOn@3a5(TmbwM{rB;V2r5f|!i=s9RulrkOrRpBO*|ZeH%2|&)nH!yryNc)BFTs}6
zU3=p&o1V5>Yb9y$yiq10@A-Ue`~g<}EPSZM=2{U<6Dum*bAiQt^@+Si2Ug|JoeFdx
z(%JOM5J9!ik7mWi0KVlz?t&M`m8@(59(0jit25O~a>_SB1)sZjw0~$O4B~GiW0)IP?Nuyqe(dK3kLe&c@&ZWY!7utPh$G8
z1wU%KbX2ad_*q3(>Ay0e&vb3DP*Hb;${K6rxo||hdL575L?)cmOT}FI-4RWfrq2eQ
z>-
z9H#8!AFr^zuBc88xnUwHGjD6{QrL(pEVz$#H2YO&*7S<&8ph47Ga0;F8Xp|bzA$(j
zg~w7bhIfL#uDhk9055wOE?98sPv4V&_x@dN_2{fQiC{XB41z
z_HWaj8(8yHjf+4;u-_I=-q)3vd+G5x{}!IPfPV+f{c=%pm#GuNo3
zPJ@fZ&6U8!yrHP)(Q%$YiLuKPD8`>_QFWHheBCTeCh|FLS=iHns6KMk`m?7~e-L}U
z6+w0|=r#=lAGL={Op-=>`P(P$Djg0_LogeKNX>)Uhl+2i`YX=gjy6
z=TDbeTxmUXZ1QHZ#Fq>Ljs#HpSqyOW1xD7noouTW-1BO0SCn7-kP%@ZZ1-jw(YN%W
zj%hrRq1rFFzj|3@SIO=3+i{p9kB|rR*%p|*lZ4F)JD#0{&pe)+-a(78_T#&=9CIbS
zFZ|rn=t2bH#$V?@!Bmog!ZcF7QAi6XQ928z+(<}Zm?@ee!+Uhh^raj2?DTk}I3p7v
zRrEKmoH*HbQaUa<+wPG4Sr`=3O1`||UYz@*N$wM)99=3=0tSZo9VPUAZuFPQU1JYbU1p{iFfgQ{$A|dHlZ@M(PCg69L
zdMvpz>4elMI`8+U?ud16-n@AOm61W@DPtvB5IsVod#z@#_M34zVoL;!KYsu`*%oEd
z+DL{rTs-RMj{kDG89(s(Rz_{_5@kspd6m-!$Dl9#KSn(=N&YN#nF>Ct@e9BE;-^ak
zmN{O5A0Ldn0{hd($|eu0mpOQC;F*)sGpw->PcZEzk3*P{J6}MTC`D3*p9H*DIS$0<
zKW#Gp3$n9WaaMuD?0kzG$%DAJA{5>zV(0?{6dQ{oK0tdJu8witJms5Eh)mFdwRG!F2Tf
zXeP!KnDkYFBL#Q+cE_^XXslDmNz8+4cSy-NodB94BOoWuSs)Gc!4Q{pIZ=$J64SE0
z3Wmj#;YhgsPUUJGZl7e7K`1a(?`2kl8*d+u8c3qQ`Ky_J`-OBDqJcJQ9L-xDR9J7f
zBqb1yd49eV=4KR!Y-ME;7Y<6|=m%06vI26Q1eHx5A$C;K;3-ZacKaZqKap-|;2)+R
zNQVZPBgd=hNhx~tnZMyXeG`*Va&MK`xO~{+TKfyX?7l>A1A{Z*C>1k4;O7BR+f7f&
z(&J^9NBday`e^%I8SQw~vwpL!4XltU{Px9ijt^6dPQ*Y##ocaU0)zT6e(q?!Z0=Ik
zayhTDhyz^>uPuXhl=H^Ny44nQ%aor=WY}Z~5Z&nNkEEIb|7Sj_GFXQ@-9EIo@PYE8
zt|5l(@bHX9=67k
z9vktR!oQky&AweC@P07DNhcNNq-Hgs?eKuWZvTkPN?V)bKX2Wy3g6$iTBRz#0AHJ#
z`y+aLI++kF3ER2Lg&1X1k1wJG7RYF3Ujy6N#P~TyVDp6gjQ#B0B$N&{ZqhgfoX^`+
z@NW`aW#*~!h7Z*oH27W>IWg2#yNZq3c&rbE!nqNDTtd?Zq%tDXMtn(qoKZ@ycQY)|9D$59b(2u#*rl8mIgG$(0$WKl4{iD8QYE
z(Va?!MI%A>Ug$lHs9Q-5=%VN
z0vcl^Op}(d9Y==fCH~PVh`IHOEhk0~saGLa`>u%+(Qt4$61k%X5P}w{)YCVL-YZdc
zP%5lb$#Zw7%fx&*hHdA;wMe~MvF%iPzbSP4DlTbRUE)7lle&tO>X(AyA|L{V`Z^cm
z7}%fESJdQ?It}|?I3b^)id=;_F#d$s_*t&)JFd>r-)w88wsQ=o`h-3gM*TG2vYR4d
zIZpEvezM1AomWsu@F6!>4wOuO&@4N!!D@aDF}k-fFqJ$E1rL530@KCix}3(Kahj9W
z>R?LH!%~0eu4-nbvWhh*^M1(SumnG9bo(7b;ErjXnWQc@E^25D^qx~!mrJ8OSB0jE
zwc5tLs7Np{_Sxl*Gr!C&g81ELRt&+)sHia2UoXaCnAxKU)2?0QWO>h8t4YKlZ?Gz9
z@^uv{tSpjhXI~U>NJIEPpT~XsSsho7LKTl>d)ts2a;{2GjcSej61f6U1_qCHLWmxy
z+ZCO!oXak@`;Au1O@=cC8CmAmYf?RL&M#BMhP?lxK`|-*6$S?4Z4k%7wRpfOZsl$4
zp}Vi_ki*SL!S~_}*$;Fgve|^|`NDpR|C+2aiV6@lF&#yQkhsSo15oU2mTuZ0b<6aX
zlUKJZVh9d_ye^Ensp=5l&Fp7=!=NS@=tJ)Wx(e9lA)#v6fRJuAQ{4p*78bfXdl&fl
zqpzAsq$ZXLh|oWBQKJQ~#?@w_UiHVi@z&EqWDi+fmug0NC1CFGM?3%aGZ6A-F!p{C
z+kmK#k$@A6bF=K%yhWcfb&=XlvP?*K3aa
z-V%eBA`h;pUq(5-G-8
z!EE~&hpF?KH5Ufv#pSTXFPkYpvJk{ph|ppKH8d)Jnz$qC2$UgAOad$>RuVfztFa1R
zAmz!KL;*qf`vogCjtmCAV`>DkbuU#Yd?Dad(JRYAoOuR9Tp-k5sNhkMp*=+CBEGy&
zsXB^0@VFQ1gMzd8^+74j2KEoXxhy()Zw>9{EKirA*kR~KVyF8`OJr#J3k@V&guxhQ
zxBTXL>NxUy!}|~^Z;IIybFOLb`C?Z{t}kd8vDX5*@HO+<+(#S9)fQTrW*;m&PSo-#
zjuGlt|F>ap>#QbTI`je(<%!84Y715aO^RQD{noNx~In=x|6S+8x
zIlx4#sR12OU6Pezi3=JF(vFho-7l9*5WcwDJ5V!CXR;aaH;fG7UtNlT=xIjNf_Ivm
zbbf-M+N2}I)Ns7GipoEI9^S3M6v}!T+xyh16#D*icM5Iks5mM#oGE`y`@Hut5j#DR
z776MR4Bi;CES|7t<;)yDId#&NOuC)*K!`Z|bF^L!WVtiYVhM!0^jH3;=a42fO+;@Y
zi2A<8-WM`Gikzd0cYs7Rh60DJxN(`@@j{$|YCgJTyDsP!^pvWVUgKlxP?3|X{DA7b
zNwxRBRtdl)uhw;?L;(#6r&Ct9go%4V3<0LgOx*#K)I?b7_xn%!5F2NvfrJMT?j#r#
zI*6jLg19zf
z-!|nU*C{23ZibK=;(wmdO2HCZ6Mh>3^dR`BP>J~H>cJ63jivanUfbmV`5M`;w3%Rc
zRe`7&e6sw1d(diZgy*9>z(!AXu$%aTO)DSi9dIa%)4PbW7z+_hmjMKe9V0@pST+e0
z%D?;Oz!?7h5B^S0xvAn&zInrJ_`3=ZhJ-ip^1C<#>28Qr#{N4P|6%K``t%8P0;Y1{
z-?^GQBAuJiScb82g*NGh)f+9=PCD%33BQBPYe;F8I@&_k9@?$6D^;LVkeBQIc1lp>
z-@_R2a@N{5Eg4tHi%z9nhndJ_JwjRUCw{g%S@yIRDBB?#Mam^hQgwi&k-+N#@f~o*
zAJCC-5FqlCb(M&>^s|IGX&U+KH!xlK2cy^;~7nX2|QZD3#CmLxkFbu$984KVrr^gQ~m^i8x+
z5#$&Ajk}Y1+1Iy5%sLSnpETDSqC*Bx`y`c~+YxfUMQ~nJRt_9&iZr+58q`9avfwyf
z9w`>&CT4C$LR;OJttu{$7TuLw`*OXf&I{ted*m~t#jy3F8g0qmWtfZIMa|f_MJefL
zP|futI|`5sVn_a4qqY_$YNg^jx2FayP}4Uc{lDnfu8P?0jClJc8-rid@!Dc>`8S7a
z^c0>;vnDnq*ffu2ZCi;9Z6nac|H6RIdlyh5V~r!nvbZ7CM(*S(cd`&Y3>aen;h{Tw
zFhy}OGD4U;W!aM593dQ9S+4Wx4kA}Q)9~BeBu=9ESR(YFP=VakaGp)kUA5C)MZCZE
z)%fgpqM^FN<+6paUjQG>944>V_7n%H3ZYX0Z5!#kqah6tO7&QJUmk{8*IDGBduTq9
z1L3$4MG((a8SMF-5drpjGu03=a8GU^{V+m(Uv}TM$WXJ}0fJW7J
zf;5+)pIp?x{u8pXA?9uni`<346~_30f3LAMjJBEyUO~C*xam!0fKYyrJnuWJ`V2zt
zy54nRA-{KgkPN~U{9%?*$N+6Z+zzuM%L$UtPGQ^ddE=)uUcGZ
zjElrg*Owiifx3XX8|&?#$BwTS+*3WXJe_q37t(<~tmm2uhCW4054~$ovx|h1Zy|}f
ztP<#W;uC0`{$HyO)M~nN`eJyljgk&&&LG$eM{Xd;q*C;7toTBp09bU4xOVWsEnEh!
zJr2;rO;9zs@tGXYSTG+5#=W>5zHDXj#%A*Lh3_m9d1~n~*7r3u+#sl;hD%tlcj#C4
zg==8mY*e4;U*g+opme;>YH7KMBrhF($=lNTBu5oE+DHinNVm5MDo~JzMY359#XbH3
ztEsKsp-hYb`c|2R
z9~lK>x!wHw4B@C!A`>8dY@6SjSQlvP
z@qm@@(lP4J^}#mn*fWMvO_RD7ZF?n~r!vI)6AkT*);upVTcyc%x+k{2C&WI=_eXiA
zzqE(g>$6vC4<=iZY`8+lQm?1d4*kz%nLLC2CXOVM=k~x;&7Pk=F?HTHjp|0Vzy4y{
z9#7VAn+0q*A5x!sG6t~$VdAGCd<<|Z;qxK&?JhKU!vsbkrPVOo6?9A43f{n*=4Z&8
zgf1MN`zyDx7AJ{T+4wcY)XKFs^E{r0(a>x_`?OB1nv$6^`o@vBdkw%s^sF0053K~weFqO}m1Mtz2Dr$pKB?C570
zhf6?b=;%N7Uh1jr9%K|Z&bu>%0uwU45&LEcd#gzhHq#DSqkN0)EugPwGcBnW(p%lk
zzjx;3aP1$DxR-tU?4^BCl6G8fbs~_Hl_j{QejRQyua_zX+9Sq$R8xQGzg;Lp^3gsy
zFin?f
zs+Y`1CXj@HMzlQpiw-0ag{*};aD2CX)O|ndOyL6fxATa!WNTZ4dV1Xu-$1VvxYV7`
z`%XMRQt$c^!S;1~4x5&VTPrEB#VAm;9MHm6B-En>lE%CV%BII@&s6cs%o|u}Y(O>I
z2um2I{NKa#mSTI!pguk~vJlP`(i2_qFJEE)jJ93EMNzrzw3R-tnXcCk62)>V_xw>`
zAmnWBv}*u(&EkSDVzX}BdkSUjm~dfGnLnVF{y{1J`MpFOU2vXy@{^=-mp7Kjd2u&{
zGK?P`KK2ZjpvH6`c<>^JDuQS&>%%(9MRiEy@rC^BlU3LxYL@1#`FzJa3n10)j)9);
zWYi!5QLl-kU8NqTE|Te?b89dahNuZ2jnyAZ$vVcnXo=2Nf?pSWGt8V6hXGJ+`xzDk
zESTG$X=jD{B!i*Q3FrOC;K`2L9e`iU%}!+XECe$LzRVDg2px8>r^D-)4tMj#pIvD-
zD4mF977FeZUk|>z7H1;5Gp>AB|D$u{{OQ9~BkhNpU7)RP6!7A!i!IdI&%j*BiY|%$
z!$6zfx=xv;Gs%O}AFf29h8rThkMFTF)&MC{fi>r~@d2+XquWs;X-0U$w==50Mf1pl
z0KTJW?(T7wO+L=N8s%p^HDbDYPNgbrHzD{p$l8a*aTa_tGPvxzrV@G(DfS*=k&P4D
zLIXPblD3jKf+jjm!HfkIN_G9EPxn3=R935-eT%HtRpNnv%;=`EA!BqAe#jOv
zxyEG1xPOyf2Ap=i1ZX^0k%!cM?AaR?lHI$>qId?mz&2*oK7W;NiGZ47Opf=w0CyG{
ze?0;;w;qHRm`{3tEFqrWLu5&g{^er}k0~MPu8`!qqV&h0R*EovN
z^pJj&v5>_Qzt#F@`GDR6E^qJh{$+qGJO~m2V5G%_RSqz$k9V5d;|`aKg1A|(a*FI8
zjdprA?BT7*L|bnsn9dPTeDg6R-B+AO?Jle3Pm6igHd|X)h42bVy_cz0e2NgW>oEoQ
zPij(gYT|H3P66a0UXbG#!D+58dZKjDM2eN@pn
z-S)=`d6|B-MuB%V6#EDGS{-2-Y5k5dl=b|Ug2eOLHVI?
zua@ZlS;;OW^@s4#-1O_IaiZpqX3J~}G}jp+M4wsmHgRk3kvnN}1YT_qB)gSX~5pEwsS*4|UD481;iKm1AR`5e(AR%H48J2I7HixDj8
zH@RMz$?>g-X5S!o8Up^vBI{Dw46^GhO887Y^-OKAiAGd3E}PAkF}BBn_nr^wkZN8j
zAKo6r0UiMd(@rbXMMBsZ?sES^Be-K^RqjfKIF&116ar%_%u7{D%PKnSpk7_)kI2UT
zR}@%M-o)BxN9~qt{HgbM;K&y=X{Iid*_@&*p_qysM0{)t>33NWG6lxpnr3#q%4X$H
ziSk<0=Fx0Q|9t-4z%WKYes{vPQmXQ4r;I5
zQY2HTW(-WLGqwlpD})dZk)Uu?1Cn>?L-Q`p&(OpRCX$h<^6&XW=ZpJmcE>1qf!*>d
zzjYwFe^X1phI?XHZ)gJQyPNwoJ(P_@sk-91$h%OWq#BMbThIP@n3i+teHVluh_~uT
z$#!4+%o;2%mAEs&IxCMv`La#3V(+rEI#m(K56*$T#Txw{H^yp2q+
zID$U*+7Q~sz-RX}fxBF#xWK;<81
z_Fr4~w|+84e5eFq{H2khYf5mf0E@XmlJXv|@1
zGpoNVXEsIOK^76;9dPwuGpu^gKUNfy5j-4eUxeM9c+NRE?r<^9M|X1kY~Fdq#txA>
zi^Co7KUEhT5(-4Yj=A~uv$Za_;jCRr_4gD9Zqz-)hmZh?BBy2>Ay}b#ZSLN9?BnGE
zYZ_6tmJC!!QHpc>%yA
z<3$Q&y-|pf?eaI9T~4rZdN?XYYI>BfdL^xq-@6cAVez8UkmX!L3tKBKJN<=Jsag^T
z9hQ`Xk=5rR9h3FZB-7gd8V~{K%w)K&I-3+w+ASWbc^kQNhZ@SOiz)Xs7xJr4vg-{6
z-r`oN<&n_~h6<-3z6$TZp#UX$sE#&q?;9za*?FJIAKzVCF($h$3BC?(?f-k~#L~P69T8X;TMe5_50*Xci&81V2PaW2EH$3mZ!vwFx87eW&B;$87-xpZp7Zz%uId_5n^U8=rnej--t}7N|WswPDM*PrHCW`
zGBO1}K{`dJhop3%*W>J4rTRw8V}I!XKR1C+`hU(%ya@4CesRiS&k%R!nABcX_$s<<1@qW$P85x3B~5@%Vbv+RLTw)44Y$EC-`+&?KPt
zm~KEl?*6hcJe;976L-*8p;J-Cm(XxI8jdbW)1oUhLnV@O>BitrCcyvy*!t$UI^)0Z
z##**k%eGxxwtcc~d)cj8*0O8aShj5}+pgz2?fKpJ{owqyz0TFSzWVt6d4E$U>ms{y
z>f?%83_;$lFTb+9r(+h;t<}*dcGwN97Q(0Z19{3Vy_b&(4R*wn`#P$3tP!@IT3B-w
ztz1ide*~D)IY(7&+-kj~3?4m>&mY`O-P2-}(G}GTb2M(O9
z%-wv?a;`!j@tmk+E)`S(F>mD#IT(#Yv1Awda}28ex<3qPXl%`9PiqP3lX
za{hk3>@3hJSB?N93=165B{C#j2>{Ms9RZ1NyrIGv0PEZ*>RbXOHO5H?4!9(^CZ27_Uk9R%A7-vf6@mQhd
z&gbGeU#yFKL(6#
z%)b8hpaqQ=d_!!1Vgu-=rb+n$+*dO@br&KKM_5*_R17%~n;VfWu1ukO{XXlfJ!Ufw
zc3$(7p;g1}B7`=K-Bi8`w=h#O56<78+q+@g!8@jqJ)ic5PJh(t7MHNYM6K<6n9>Cq
z%kh6Nzsfb6MKG^ZBV{()`RUPd0jf~#;Fz$&HWLe?p&sxy)#oJ;JiT4!BTkOhs(F0i
zJ6hf!60x~?lmsordhg%W*;Ub$hhZa}3~1lnYu>1*-;k$QQf}Y25(eEN{}-U%SmpEE
zmM+U9(oBngv)gHcCV*heAKR6lW*##&u=Pg;TeQSf!OSQUwFjrHWak@$W}{Na42#D@
zofu~FHg&HWl72RC_f4Z9h5Z%YvER4C6?Y}1D){(0`9TJYLcM`K9V<_y&a6kbh5g4Z
z7bKASH@~1i_7-ydLzKI{BFa~+j{XBEzj&L?L_{J)0{>3wAACI{Q+B3?s%gSywgf4h
zb7>NWKJtjJI5RxYAf&Ag664?aa=hNu&fa0P+9>9*=ii?~(dk(z-cWxv=libPG82yU
z+^M99rU?wJ>;aUO__=+F)5JqOJmXhdo`Q=DI-GQUqnSBKEx>UnZu
zIIN!L^m6#yhQPdCGIN%?G~@9?;~a%4yYxiJHsq#1lY+wq=7=mY
zRh6#E9Mruofx(5BT3o9uYW{u}p4R${hkGwjyMZ*81Bz3X0QP5s9P3&M(3}09#HNO0
z=7IKo<4h}{`=zFX=UR47w;$nkFAH1guBS3V2(`QJai`Hat9ARm3EL%@TK(UPKXZ@3
zlfk_)+_lMq>jV+8amJ)t0MKV04Nf`D)x7B-D4zWL^=BktR`mLx9-#_9Umrf!&{2c-
zP2G#F2k6>lS2ULMruSeCW||>4J|6SuWSx$VqCa!+ZWnG%mWLr-Y51YX+)hoFCah9Z
zei81M`J=FLbRU+t8*=iiL;#GXybj%i#b&Sek
zH4@mKWxUl%!{Nf6){vioX+iW4`I~*71n(H3Lk)n5e9m?>G!P(tz2DSOP=PKsbV
zcWklex_F^S^nasP{pduDe>(>4mDT?xU
zK-^Ns<4f{JhsGZv@DjfHl>mURlCqY=hLg!&%W@D))0odDv@h2&M=_epOOwT`qTA|Z
zw#MaZm5!TPxq72tD}v_MPY(0aqa_#lXNEN`sD`dU59jwPdiC(eA&*DN8N4nw4@?y1
z?~anYuH0>MkJNXMk84utI7DH}-6rP8HO|=(*c4heI!~Aqh+PL1yqhl`X5+aBTJ|Rb
ztB%>H-6Pi1SJodx-`l_)QIwJTYv6t9(pwXu0*;e2PbALdyZ()rbWSE7qTGjDt^bME
zoS!U5j$-;+Trzu>PSFSY%sIMnmmZAk5T+1hUWy%Cg3sbe-Kaaeip_)F>@+?y?XrZAL;x*Sn%yOd`HJ4$6@OHW+c}U{
z8#qn;M)*apnk!vsgO8|Bq0iV3YJ7%8`%8h=js*M?e2Fd
zmwLH!Zc(`eBwaXyU~aVFo134!x|z$Zovhz~MxsLQ9Z!b9{jbm(MW
zV?u~_vE4ps9h@<&@V1|`$UxWBIx_(N0Q?EWHOZ596QI0$ykVl0CN3i6;x6<61+L3L
z+yr^qLo7i8<kNaW+Y?766P0ViPK;dvccx5fXDy3jP})DpmtryUXO(
z_W51+Oc{uJyn~C@_k{#tjyV~9sc>jJ9ioxU}RAN;&VyEiLU7{$0G=>xg(Qa-o
z1Q6Ay7;xVpf*CrV(ZO$^pB9Kt5Cxux&a^NocOiqdTqqI66Z)g0zAF%0Y|C
zB66LpFW09H&2m769wrydg1zh6@w&}`ZVJ>+`wZ$~0pJ{WCf&f+}
zmhu@L)gjCOIrCwD1h)?SKiVtg--{Pr|($jwIQdf)dx
zx9yU{Wxb*8HL_zfc+NCudPm%{iIm1O+souH?U1(La?}XV9}M||uKkTWp`RGICnr=L
zm5X{4cW6-Z9BCNMrW=-%0c)XVfimFyYm-$w)6kbMY;G`l=W|-j{K0aQw?AHw;iCT<_O9;FnMlvEoS?a6p&LvK{4O&$l<1O?YZSI
zEoIH6j@rewJ@fSpvCVfc87&Wy<_C=rkY3dv)bZcwi5^;>cDSFQ^(IAD1u=OIn`ID7
zE$~XkGFv8-N~M?dBuPyi&Z$Y`SogYbe8RL~6@4ovDOG*t&OrPg>c>}|o}H&gZ~VO#
z`_a1<_wJ|tRCjB}2)&Bri54A$=%Sm2Pn6esN5Uo<@Ob
z9SmM`YzyvBEW3G^^`F_Rt*q&G?@u_D45#_eqAQtF5g7zy-TeAu-L97s82B~a_Kf)QvTJXrp(W9u^(^(>s;Bh
zxR`Jw0`@nLoe4LdWd;mgX@2}k!r=E_WnzGVe;V}PEm@&NB(8TumiVcC9hjS-E`@g4
z_a5{q3niT)&I=>{6mfO2_wh7Hz9ck;G|IT9sod2_a#T8FMPduXfGpbH(0b90c3NU7
zxMgrT1rl8b86cSK%B_B8OBhD_r7cQJ{@LtGSPs$L&4E1M8{f9V>#y45=_0<=QO1{5P$d3R!1`I8PiaJ2!n=eBB`-G!5^^y79(IaXvzzOzjy;&)-Y1>=iBG*Z@710`TN7Z(2&5f
zl(RM*NXdLcfc|p$<;0#l=^H~?XQyoO!nRtlbW7*74LTV)hip@?>#!n!U)HS&dPJb#
z>&8KdonY#nj33P>{ZZDW^=?|^iNU!S;GBhZFe;$>=6EF`zY()}yqArV(f_kPhSFZ@
zh>gdXp4DN+Kj|CHS%q?h6xwmKn%Z;FNB0C7sYV6m@|rX9s`|sLFV23HN~{gm%R|sc
zBF@@HbrT=)Wb(~vfvc7*-{h05bwVUl^CvN+i)GV>MD+k&7{CXS9CKwUs2#zYmDw|Y
zjAguV$m*nS*ZY_UZ$_x-Ku2#YNs>wr|ZBly_0){Uz98aSTkjSGW
zct|E^9(j|y^R6R5>EE>!kCC%7G{vL*>@j4tIeZ-!{vK`|bs!OT+ESY#Id8swe7S^T
zEMxcac&+c)DTu$*nqe=h-+a~l+ykmqBByK_fu9p-%bBLIg=#2CG
z3~RVwj68hP=FbWxcL-@MjM8@b!OLf-TDntd)5#drn&|YqSNRdw1W?hFFEVM$<%Wz6
z9A9`^Nb~tvY~cyheG5cxQKRY_!^3KmE=2~5S0*4G=#Utu=aDk+_VZfMLLCcj*+0JS
zHc&)qFi2(`dt6-Pu8r3;Br&F@c!@N_i(F(-Puz82L?$#DJrpFMaJ+ibB;oq4ewTtc
zq5r`_es&Xm)c42xn^pMy{)bTVm6=@}MQ*jBYhyH&?E8kdN^eqlBDzuKK2^-4(&W8O
z>8vBaGgYc&)?Th|%Q7;|@)zp$3wxdMN0`~m>!hj=m%b~kq}nR`?!tTN8Q6!GfZ~t}
zq*Y6bRi$Qh-l65&D8pYBU`U)=nYT{L@~>v&Zz6FkNg?6B!M6-3Qrm7JZBepbJNs#;
z;n(RGP#5$;f+lYg`>2fch&F8;)bsflE~yZ?C>k`Z^5XX*bsXQg3z2i
za3!Kr_m5rRzgc-w-ANMXXqw46f$>``qLdPM=LxqmQS<^S-NZheaVlQZOhig+@~0}h
zyUX<;m`4nf{ILXO#PbqdjV4e%mmN)XV~=2Hpx`nD>GHZ%+_%B6noVETOzwePnudf!
z#QGnjvi4aTSlepOhm`ij*c$QH^rsJ}-9Q6Xq`wIEX#sv_s@{{CRJdEhr
zRc=cR5d-RSMR0WgYIgL&2@B+oPju*zK>L0;#&UpLTt=8sBEJ^QEwMUL%B{>Z$niF2
zp@PjLg$x+h3xKS6kK2z<6*upFQ+<=;^a~G0ZzDad(v;eXGo|~?MVqMj&Xk=^+3&E$
z=6TcCs!djQSCOBLnLpbW6Ba3okEqAy_pe7OiM1In0&Zpi>oJ2FsaTBOEwWzldhEVV
za51NLHgN{k)Un8?wdB^PD_K@7EHB354v8UbJJ77u-^6gXT{sQ~xhnaF*xD~;hhUW?
zA}U9TTrXf-K2H@%Ts01V-E$i^AMLiJL=hUYDn^W>)tPMaQogbn9rS!C`{XElM(p2n
zQR2I7l_k69KUSzYsq@iz{k_;Q$9&zJC5$rh5YHWl@pQX8-+EENwCjTd7}y_dil}`u
z3FLIF`_o3npWeJ_lOw0*l<12P4N3h$l0=F`f=>*j*~CZ8XK-e@s_LJNXr+sHy7S?K
zhM!uO%>#F_?GJy07I;g`$vwwO8%SdCWDaU+^Yi3d8Wd|L$6NHux6#w2ewH9?hPcCL
z)huE&4~zV*$#64D_M9C!rgxhub5Fx{<01%;w3GZob9CD|)x6~q!A-l~Po7>NQL#O-
zzwceBz8`|Xa(45yvWbPB;lP(q6qC0Q%BQ@?iMK46o9i#Mgo~HH@`a7m8#>J6bNh>C
zY`1Ro5`J#$Rt`T8h_L4mP`altDr62l+G(RGqxnh>VD>{M4Opg`m1dch@TeD}vEyot
z&OcV*fBe4pvhy~7GKAd4k?`c63iO7OCRI9kSZj7LR)@nOIrzuNl;yD_Pg1T}sh#mc
zNq%Zlg~D<9<<%l~3n$X1qnR#$Rf`$7Sc=gN9MRH(pEjHk?1zpmmk9Q|4oJ36PX%Liz|8K
zlu~nbE2^0X8|Q)*vyGDnt<8Sen0mK1mVri%iJQsqX^u2A5Y|sv3`ih4mFeB^owzFn
z!cH4@z}pR<2)`N3PAPjdB?*^PZn^{hZI$+bc)rrQzVh31t2wdfND$|kS{V_grECZG
z@CVt(9WKH-iBlTM*l1mr9C>j}j{gtbTt=L5@N}ZZv*+wK{H}Y2bvl?0V(o#cDPn(!
ztmplekI@cOMY}3q3EV^hkG{;q^>;sS`>kbI$F@PGYQ@_>@6&tJau3il5`%uh+vjdr
zk|@Ou--kI$p>zmT$N@hsoS;48X;_g0q!rB2o9+^THe8KMcmKCyboo_Q?qT_6N)^jN;+c&gD51cXiA6V(O^~p2uw`_IK%FmyWn@QQ5y>$
zY61<4KZ9j~)Xb@X{V>rE>qmScv$ZxTVicS8;<83mE40
zPpaAkWj=bAcVPbw6K#kU#TTNp-&YgpB?ZRX5>L`|SYE;b0+q8?3)=Vs{bs;tH@X?$
z6ClE)Hw&m#vnSpVm>$5*t!yIACBz55&BlNC#~$1<6&ZL!=?@_siY69A#JD9EaGzX#
zi^z%seo=cNF~n_fDB|UrP{5~2dYP7xwf9m{@O2!iToQ)ttgq_S12AN*lUhE}dWJEV
zIa$ZSk`1gZ@q-5V;Fl0Ra3dSXKoEU!b9)$aWzfYY&}{J+y-n?pCJbAy$k+IujRHLN
znh{g2q7>tY;!ssGubh7cLr254ZRul+J&|pG4rIFlM)34eIKE5POITG@&hz4VAbGl9
z9^6!F=Zy;o{5a8=Ug)mOkzNnUKkxErb3=j
z(34A2=~P{6VQC6FZln+J=GV+2CRYLKz`ntru`3DJpaHlm4MG@SV-g@^c6RVkc@wT$
zMS{eqSynWIc|IL6kHr5vVgEF-4Mk)ipdjyaJyZHXbUT(SuRw
z_pg04_Okd(M>0(kv9R$`sz}ra{ATOZc1AZMrdQk8y@WVE=
zpMcIfU`YA{m0gLRXrVAgA9GhQ`{9Ja-vS7V%lpI&`5!kOr&+K%SiE5zbgXvXhDX=(
z^F95hP=w$WXOHeOB?kMVQNsF*hctjZua1lABUt=F3yBd>8qg40W-&d>!bd0k1Q-(J
zN-bc8xh5*q+bn5h&V&Kw#DA_H$$qXY7^M;Ws8LdN6*
zI6Cak4R7e07{%-~$~Y+%N~I#UsuH2VZZIHBjHyuH{Y8k3>=jzRS@LrO0-}YK)u6<(
zN(fJ%+O*`85V7$=CIp%b1Zr#2u+kACUry3*h>!V}X^KV(oOvhGzO7^D?ccC35
zY~q$|c`KSHj0&MODJ6m1OZ|i7=lLcrwn-6e%%PCCq
z8TF?7kEiKvwleijJPiGmH=gc~X^ek0zxTv1Z6=*1OSyHdRhsw*@Y~udjf7(Sg|Z21!84QY
zlA>DCz~8=(P8QoNccbPtzPP!;WyII`&P+lu_~bm^pTA1WD+9$wVFus(E4l65ti$A`
z%jl>Q(fK;3*U10uQ$?J%*9jmd9l<%y#K>XWvCQ(S=S8?rE%fhA6qStzP)eKB@N=L(
zh-E%jk{)FiEdW^)e*mzt7HSJGln`=TTXW+4S<8H?A8k0ozaImJ*rWC#yKk>zF9ql=n
zr~qMo$ck)10tKRu?N1+_cPPs%3H{}HMegYhj9U?jPEEuy~1(^gVh}rdf9*q=tSH1OU!TB
zXabhPWZOk+NlC6-n1q@J{sPs|LR5!qKv`%+|6~I$sJ^5)xIM;5e)8sP%&kL1Md!Wg
zAf%1`cBtI)O_`9(>DP714K?VuiY!^1N&jLNWhU&waY9FTpUHZUPC9Ud-SI!R&Cf=X;8XC+D;q2E7;?`y0hXs-zk3JGzG||w4-gn{v}RCKayUTbagj(e5R&tecD$^xQ;(V=AS!b)4cdY
z>~`~^^g1$KRQWQ{+tTH?agPL4hf)NeJ^*fOkn_SweVS%@-^b@0-zX=ep4n
zEBVHioe}Y*quj6e1F1GDax>_$2TwWS3F(@LJewr`{u1Tx&&hlumopQ6aGFe7{GwF>
zmv<0ZYoe_ppSsLgoKej`&+z2vgJB<=J(6Pn^uf7fSajr{UyE>5Kh8PdaR~X7G$<&S
zGS)5h;uLt=62dy(u&bv?3^={mT&9g8JnhJ;h4fa55g(9_*g6$EEkc%pnE@)CiI>pg
zMBvdfOfOxiWv2X_@@^Kj_)|MX(o>12G<<4sHq&;fOcMZ~d>9t!OI%Cxo()?4~
zT&9*0u;ZHL$Fd!fY=IlwnEJ$PS?k*9JRY4V$r>fhOE}Jl%SYQ9kM8iG&zB-{VYYd=
zJ#7+e=Z}wgO6hx~rTPbh
zXc+FAMP-$>$NT{l?OdOFPOjg~UcL?g#tdy}81uK-w=yla5^&94oeV{Z$Z)5ZQrYzD
za*5ITF$3mkou>+>6y7F}p{MKdP#Wzx?(_Sinxj1j#if`ZOufl1hA>WCnd8Bl_I8uC
z#xRQt9h(Z?P0x5(xGKfM(qR}GMY30FkjKLpDh((Ies+ti&QFUiW!5s$TQGcWysew8
znxE#KyS6h*XRI53Ez$?gUMrvab9hv?96%i26&9B9r9^3RE`Etk*`c;A&n3U{UZp;)
zAFHcgpF19`Yq9CX?Wj#yFPtUISkT^Qq01T%Dg3)x8kKG))!m?HRWGcqJ7E3vkO{YH
zG`Volc7@kwwbxRSb2PAToI!VKOo1uB3-~-K+u}s7lESW4Wd*HR$<+^B`5askDdNt$a%tn5${Li{l124-=~xxyyeS{^!s7m5mWe(2}FY1&cxl
z*M0d>1Q|l&!b-S{LHt3qw9yuuHG2A9W$^Pw6kYw7SP2QWiAL!1Y~Q#HYtO%m?%xIT
zClZ=IN(VWQ@atc5l+9f{wI+0&zV?VgNn$&?>cJ8ih={R(8x`nlC-!^_*&HHiSZ`Qk
zTl6MJUkf*ln#3Rzw5E%;q$~7txqtsE0GO&4+fne%+&|4tu*TH9NAA@==l7DHVm)WP
zOnwCE7cVAR<8)$Lr)4IA;Z<9a(cI~Uw;Oq`}ha
zxvIrTAkBp(-h`tz2kxlAOB!G<)0nk7!89wST;5FQBXlaPKZ=^08mSn@fT=?$A!JP%
zZNp`U+-E|NWubA6s@EQ=BbH!@NpUx9@WRQyhT)0d0ufZe&z{NcB1gt947
z(h=ElY9ELE4x86fgY>5H&;#QV0<+!q^VDZLJdP%h
z{mpmwa}52=`4b037V}tyDk?M)e;m;2PwF?K5A^FTnQ}%jIO~{yG>DZwur5k?m6l!A
z>MhO_(LELs4QVN?B!|Im@zzzhl=9Zq4H~}-MFzqSh_({ZpEw0YJPIJ^QU+e`#0Vc?
z#IQ&$Vh?+nvrq)HQmgfMmw6EJOY-)a(bRkOu_#3Lj{JY!(Zul*(tflKEYlUmc|Hxb
zzBue{gf}q|@#ctl%{=AKY7F1JkNoCHhD)%H!_T~8q-VR#)>H|iIj%#{AgL!7*?N8M;^VO<6;xaeH0BD<>O-aH2-P;l>~nn_Y?R$hr__>4Mzv#N>^4X
zzQRCY)950bGc8Re!_upX4;H>Av^)j#pLj7iZpT)5j^K8P!P@2~#mL^27X8(B8gvN?
zYwq_vF}}|E>8iDp#Hbs|wWl<_?MD2#C{cpz4MQ?$oHJ~{96sPg`vN(N483*yeDZ^$
zkm$sU6ds50mDi#~G)9)hcrzH+nAEeBBbKB5wq;g6Y0*{$SQ_!#`4k+Kvinvy6G(WhC%3~hTgtXa>S5}k$F3Sk_r=*>q^nst7}O0_>Z48iqIp*=iU_!3vU&=T)|2pQv9zPbk5
z7x+XSi5T<=%ziTW
z^AZS2FFRbVjco1$jAWHp3fOeY^+r#RQr0U^_Z^A9XRuUDB0nt8GBGlqq}EPp;)Ckf
z_ZHKobe)HnxXC6SM0iKW)b6ZlcQx!Mh;LTvP1AjLo|%a^B8{z^%k5#EwrYuHXu~oZ
zU7B~8{HM{}6_;nS+%|2phnMGit!{TvrR$OfrNy3)@XD9Z>qrnPA;%CIf2cAnrAV@$
zsOy4?O+!$!@&nj|bjZGIttD&u*y=h~GS}rD+c+L6QOEK
z7fUQmGel-h;fG~YM*Pz4z2Z&wjj1k^-Em5^ea4OY{LqmjJ4oKf{;=elEmzIkL&2`|
zLk%M{HIWeJb}ofyb>#{E<>bl&UE~v|<5o=Ml3$u#ujA^ffnpIXD=nR5-*sQ}5gOjf
zOQX9Q+#+pCS^C`82=P722Y2@}yNhVti&@QOUdiNU||~goIE2oPV@PepC2@>$+;PJol>}WNKZsul1g#YkCg@
zKCC9WsfATOmPc869hFQx*hdfL-FFPGz8oyJu|7;rSRIjEA0HhR`)^Bm^PLvgF7_wX
zNG}66qBF=hU+7IZUip{NrlnsEygXh?ptrEVl>Bt7K-x?9AYft}pJpu*pMLF#+;~@8
z?9Pscp-3TDe2~Phm}7xUo}g^0%(&}j(&s2mYM91v`o+2IMYe}ELnnG3Rl
z=!3tBF93VtX4$arG@GX{@Xt_xq9lU@h&
zAj9Z_
zok!ej)PaGV+L0A`tk*8m1Q~>shnR_yUtf-j;eoI=%hJw?;Y5BtL4v9^mm
z&{?&R^9>uq9IKvU{v)Xpru`V#NGLWjGf(xZY&|k3`n>y|5)thlnpe;K_#e`-W)t)}
zSXzi6%K)Dsf4f)Mk}evJ%}gAdg@XSLcwNUJgF2tbzbDYJPNq)9#cu9gwn4}v`XIdn
z>j2agX;2KJ6x6(&FO@ssN8VQ)Enz4*b`z%G=FPXAS$?|Y(&N{r16HkmBA&SGx5<=m
zBWaQ{n0pK2GLPTWbmy*|0(GAUCc}UUoSO#k0;dh;Wx(0E|Bs{HLm+>@8_;d^e7n|$
zBi$Psau7>Y<)H!X8JRdX%4PSebS>5Rhn}IkISg9{#t669^38#i+c9f~i%O}Ss9=c7
z&eD3Xp6E9SVc1JxFU$!nwsA~zIzwJZ+f4pr>q|qmpR~S(#wBW6=rQ5z0H8}H!4}jY
zqOqbI%o_wzzbCaIgU})j)nIksvpeR(m(ZgFlEhr&mS4M}^f{xYPUImX-SF3FDVB7S
zP-c8xwc{`%YZDOmWEL4sf8)%mVVdaIe`A19izCiOv{THRNGg2=q!Rr7Z~KhS{fz(>
z@G*gY&Tzf)=C%K1!z01NVBS>(NdpA@4Td9~{8m8D0G!4ZZc*x&=kNPdw1A!h2Pih5
zv!Oi)FS}E;*lu>{8xk
zR|4X-bVqQbFQgx!ir=8y12CqwCw?~vpzaE_s+gTH%DdscTSa#j(cpx8INm?${`D%|
z-@h5dS2B~)?tj2!7z0?3(#ZX*SG&zrl7_y&%m4yuRjmq}f#VT?xU)b}*Bg)vR2;yr
z(NHor8u}mIVz|6XmMieuhoflYx;M#{O9%IJ2Xd;|Xx(DO|LifetNMY**`&s>kwK@@UlYi!)G1g21?0$m7$o&A9v}Ns{C`|k
zJ!`Y`aq4Vs5E8teE4wy~f}cPw#hNz$g#0xd?^AKq7No_J>^MJxbup;|`CFB&QVS5R
zO*V-3u)7L54bqkpG3Ub9fYYbK?4`Rj;G9XLb}jNl)Q4IF=uG@(`qbZp;Jp3|z3G(U
zCkAkFuD%k;ZT67H2|o)Q_xk@>+`$WHVEr?Uv
z#BbWDtM^9@xM@ILlYZNR>qbaPvm%(R)%4m=K+Y@P!EJH4gkBdrj{?N`t`|&ucDh&z
z4s6`M&C3=u5duxW^hSyWbVuATECKv2#_O?%nOLZtJf;jAu$`vr6NCWjmk#1KqE~My
zR+w{q9?-J-!T&=Ku#t)w!HuiObYaa;B<=ciF>@=_IvB;UzY}2tYv^D!F=XsGyT-h-
z6%&9t52&C6bxNKZVFDx)tu*_cA7#5Uu2J-R)bDWuv13OX9ao6JMK}mdd|gCyWl8_#
zy(P845iKCbzk)Vs2%@vhpTq<-yHGy-5sfBEhd*RIkJK1kh$%T4f8V>o^mr%~8wYT-
z_#Fg@HgGLPlDK~DD91)lNDv(Dc=N&R7XYF8d0Z@Nk*$?qA74l_i>=^07k
zFntf-MbO`TS04{=XhJJSwAJ3)t
zH^~w0GH;Q-2+U{bpM}seamRE&wu+dcV`~D|h;bgJm&q-Jf2bnHlBDqxx!T6CqY#`S
zNo&3N
zA*UTnivPHiK=T{8dt2?9b>#y8&in7egj}F^R{C%1*UjPbRKvw`8>7ykD
zg%E*a!(ucjmjn$~NYSJJUxhRl1(XW;`^p#_-rc0Whwwo<30jImi$K%+{x=Zo2B7!>
z*WK>fNEoSOJ}X)?rzsY^4y`}b{Nftz@oVk`EA8_;zY(Cs1?HE$jN;!rJQPv^Q%0)N
zpm&l%Y7=z1PXAoo_>YSL(VC2fLIADzuPXnlf+YW=f<(1>?u-CS4jgh<$gH&r9gP#CziW_)8Wr
zDPT5?U({m3M&ku{`=kvsK49Y!w`DWO$9Y=WZIm3_sAuX|u}up9U@K>-a-CHL
zsIIc#mOHHlMO5f#U2MZ%h`^*|9Q=WneGAtt18>9O`31L{OcGxde;Th05}eXslKC@Ns!qjN_5GjuV*XD&
zjXV6U_is5>hxi|#qmelXBp9U##lMPrM5Ipc38F&iwDXQqaVVCiQlYJ~P@s(ih=RWn
z_oI=Tt@yN{jeDO(bR-qAyRR=+QD@P0
zZhRobBL>u&T>VsicI*JH?L(k!%=~2NO&h>}*Bi5!X@L$<0jj5xD_Gsvv|H2=p(5H)
ze>}=0QLBm}3^I1q#&JD3&DAE(>f+P7AWh@k2xQnEw5vGko^A4@ziUQVD%
zi<>`b#Ojv8k$6#!b#^dQSk=@iw500__O?^Ri|n*$Ql!wHRwXHL7ecf
zcu;Fw-PibsqyC#+<{oR67JMl5$o5yX2#&M{JA^>TdiBnrbl3-m-IJrwfV1fHM?H?3
zC%1r0qf{dnc~GA*P!ElclTo3x*WS-Nd&qzjv8+5vhcT`$7@L?{MA=&J=jN
zYrhe8`}4z&vw!gcnmAJ+7&5lcGy;vxS`UpTT8c@^V-y#M#cA>t20$fWE&=jrPX(?+L%@89DKcz;Yo$#u-G6(lJxUZjx
z==4yTlT(!q_wqt?Uapq1YR|~hqLC1yg1wajV*gVh&I#R5xvAS~`99#%)<-8~L9kZh
z)QfSKTNnNJEa~P8oq}%jd)NzFHdHa||ES@V9d7Es!r|?KJA-x%zp(-pzEcCtg<$)p
z4g|qMSHlmKsDsFQn9Te55(-%A061XN;7#DjcE>rC)eo0P9_@SIk}oP39X6cWn%=#K
z5PXv(|I?)$`b$DowN4_={R=8kse;j)W!{G!tQQD=enF2x>?k&w!7Wp!6~jD4(*STU
zG?>${0{wc<7F1=e&f*I$<}MLE!KoO53Z*@mrvQfXpNbCvXAHVno3-)?1v^WEjs2$l65{+ypAaal-KRsUD6fUY~>3qw~$M
zt;OizX)&jpQirXccH0M?f*j?xr;}oE{4v@$RXZ`b!9U>kvzE%8kK5VRxX>e#*sk5P
zg(PBkJ3x1$wo0sb-V|vhR0s9p?6Nns2fE^`U1#`%$YW-pKHRH2FEp724g=+})JLXf
zq{86wR0*zxQASX8T&6zvF6+?aXaJ0}v)f0*FW<#;m#cOuKBk3a!@X)!w6D3c$|MrJ
zgH+3Ml%H)T%xl-tmFXpiHcz}v+2MlZin{e~ffSQJdwH@Q$isyau*$#zS_@60L3+{9
zpqa$}hY(5q`A+_8HKxvKD{pNf*y#mZoNYO~l68}Jm=p!T{mk#t)5JBS$qD=q$>ZQQ
zs&P>AbWq-ma`S!9K%sjVG46}x{tl<7C-at6iS3_(@e`
zxUlXuR~tLHD~i=2K-Amr4AnV}(*dP--<-QW;<8uNg~>@^r1{5~>qRYGBppoDtBQjM
zgo0ibWewulQ7AS590>Co1+>yRNJ=dQuX}TAKp3LN5b-!-ML|*O;_NIm-d9jQe3i%s
z+Hp@c?FVuy8fl*?#o-7y>Gyd4rvc6`J$@&J(D}v1<8FeNca{|mr_Xao=jrrHw(&-c
z3(K4@Ey>eT`rN;sU3n2f;g9c6)%dEpU^+E$RDJw`X%~!Yi|cp(srX`~zWqBDK($1c
z!zU#rd23=Xj-FUD8;Pqg6q_=<+?_;Q*AUq=&5P|kb>qfU)!vyL7#Y(dGkvnv7@%@%
z^SJ$TnLE+f#&?B8@VC~rZ3{9WG9i^saOnp2aje9VY9(E@ey`LX@{ELW4*hlkC${&G
zN!5nI2f)Yp+O!qI(55_XE-?mPTd^DM-?%Qfc%^D285h(HYj&+CD~X!W01
zX@&mbk8Km5ik8cL%lSxCazK!Ez4ad4dfAlo#NWjvrSg;@&YjN(FIj24!wPM#i=|u~
z{oKigT2ySC#pqxsadAMff1T2>EbBOVid~K#fN0g9=R4~^s*tQ_BnUgUP7L5ZdfH^|1YAxGN`S!3%7Er6lp1LEm|Ck`@xGBcW7~U*S1)T7T1uX!Ge1T6xR?S
zKyZgZAh<)|=6rYN-v4`MPu}cV>s`-!B&MaTv2=RFFzd|Cu+HUA1h;7lvwKJNBz7nH
zM@;Z&=#j^4cRHl4oA^al+&b?sXl5Ne9`#aF+1Rq>F4EWd~8F0OWV
z*EK^t1J09BZ*aD3t+WqBV0To-5ok@?4)(ZWz)8Rlq0ns#dj=4x}
z5PCzx_Gi3FHbobcPq6)2qcV5ut8AhsCmAn`PoIfunzr$W*se+ohJHvo2>m9#V&(|K
z{OE~K!jFv3Pg(>~LYjW|4rY=9-pJM2WuK_Cssz%`93?T$t{%z6|LiU#hQPl3PCZA`
zld{4u46|Kt}N^62hJwK=qJ^b>4k(o)N
zNt~&6S2H`4sKH}e&boIK6jFKEbdyL%e9<>WkyQC)bSU;fG@bJ}8A^v=%@^T|4Ih0L
z2AoOu?YpP11W295&rH`bXvY@h^K2~Z_fCk-(}JrO7LA0SqMC;5(wL&o$=m631$Z}?
z(uxT?#9CU$9gil(^hR45;+6#plNwIqmR*PWj9$DzYXX^Y8+_S4Zzfk@i*U2^#%ybF
zAjWlBlr=&Yb1SST9;)2uOm-r;U+xYL-lC~79VMCldUWrp7W7U~&%9mY=(|zD@1Jjw
zY7j?$+w##oR*<@Led3o3pQDL0T3XKotfzhyUu#h2nV
zvNa^W3%c20t*FFPRWS|@)Aioimg>+Rcr;E=XJDLBAwrcdmBDlH%~3G
zOGsa+aa5?x!~gg=Uq0b5ZjfArAx*r*CT?@H675N38%pd>l*H#f1kv|~7cbBgPe`^h
z@Z~wH)8Tpb6o$CV;)HIFPdN3=hfY7bTjsx6pwW7IHE%r~uY1Gq{nnz<*JNclUg=`Q
zZ$9G`>}`2_)Oh6xwQA#%&bwHY4157Kb~7?x3OW1XLQ%C_C~|r2XaifjSWxXTrL%A+
zmsp~k5kxzgI$d8}XzU~%wSHn&y|rtv4rXO>|HmTXnEQeqhtsy7;sN?*nzva+z4?9N
z!hpxwIg6f4F4}t0V>G1MBJNQlG;e{jJW+2CuN}xQ|bbIrRO5bi*G-MSmRBN}mTOBmJEAi3nq70=*K)W&;
zcJB_A{%=%inHrp^YdFIJ8u~0MVHB5h51VUQ)+9&XtKZDGjn>s%qtSFHj>_JQd;+V*
zrEBL1*#Nseb23)f<>i%mHW6;Hk{3{an!T(;gfrQ2mfgmR;}@pQq%L1(O2+6V`0?I*
zl{kFw(2kTG_9~j5D_|UvXtGk1*z-7C*)nv@_x#bq&z!!>sGyJmVItN3Di5>IRImjx
z_4jkA61PacUXWzBI#^|fzhhL(2(pzywKS;Bd$rZ)9(ET8!hisYMp9UvitFe3k_K;n
zX5^9{b4_Z>bV@db2&?AT%3H&aOebZsQK9b(+l)8~x9Vp~G*^Gb-)J_E1~
zO`fdM3kV@*pZ!)FSdnGX6G0n4<-K}8=^LH^bsMiC7q!7JMjKMeCb{}~4N8`vKF`Je
zqplkkn!wGj9m#nr%C-1JU2V2uXh|8*Tt-qx3*Vb$Q(#X;rZDHdldZ^J$#Ugu$n@H0
z8cD3SH1x;da8_oGFuN-MHul+ngGwq9%ZmDQLZ|`q)KMNbH#w6hj>}DXKy9wU8RHA2
z+n^D-p2>_)^lvFEzBl_F;;Er$Fh4x-LJ4s$v~gXl=RU%S{JXd{(T3kqpKvp_b@xZI
z#cveI+z51@f2p1uY$|YAY#lAK-$8XcP0Lwy1U-9e4?Z{3NU+nHm^hq6ff^}{%^8m}
zy^Yowrvt*`h=!~BkIkA_5~t0y+H!!m!*Q$_+uEd!?SXE!23p6CUv7dvNFI?n~oq08utz%)V!u@%~2tDhlj;u+JHaS!y8`6FqIlo5+~w
zG9~~HN(B`+9CJ%m(}SGx&?H3{SpM*CpLXHn$i^aG0f)8Pz1{x)@oBin=^rb}C|E%g
zvrkg>AcBvS*Cz;SkG^$
zHGF}Agoe*U&qi%$gl$;rYPSX`YDNtVS=QCCIog)2AKo!8@2VQ!G0}G(yx|5c%pR>h
zT2)-QURk3mATKmTZg9E%%Xyq&>bs9Mzw>E9RB-GyxQjCH3wcVC+
z8+I|_^^Oof&3XABn4@W@7F{o}1J>cti2gcm_x$mm>()oh#&+?jd&44`xp%8|3X9->|L}PswDnjF;ZKZ)xe#ghU>nzdP
z*}6_H+>05y7QVATKFry04?!?IpDqSTN}etN@V26~qMWH3Y}&k2eM1d564M-kBA>DA
zbMebemWA&(YUj=p`-vRph|DN_+x*}yLj^d%co8pD<7o4!#!_JCj0lJwI9WW{uy(v(
z&0m=Wc2r`nsml*3oHtBr@NBEO6?n|LV+<_WWu
zY}%R8OV9;eyX_J;nPbuX4;SwXk+UaE<{k%D$Qgpb^`&(6{q0&*S%=?FcDJGCT9VP(
zAEOQZzH*UkB+&dwJUpH{P=tM}==9tvHMRDg_$5SC_5iFMiap^p59Bn$-gRF0vd<|-
zR{M*uO$NAh`^2|?R7?Lilj;KF(sQ5pw_5-=6j5u1|FC**?pi}rR))Ff@BH-waj3Q)
zSJ%PH8h(KOREw?cC_}vhd9Y`j79GvL0ufrj$!4c%up4!Rgg7+J-U~$++a6RR5;prHya+t`Ms}f?Arb
zy$f4gb5V&nDy%%JFAihr*fw;!>RsG#iy3}}F<|R~zD*>?lK$)fRQufL84yMk_whZ$;tS>1jxwtWq!ay^HMYFy;Qo-
z|6d_wNb&g(l3)L_uin!Oo0Wc^I<*;3UeUwxb*5-;ADXqpyN~yCMw?lpXUtb@yy=Y#
zL5CuK#3Ns+s#Te@C$0^j1?PW;Py6B+YddkI5fG%VUe$in)nms*Iw#ShCV2<`xbPmo
zASvPCg=P*tr&m&3j2Zn-5(5J%cHfcmXq_vCq@^Vk4TssuEyk1E^mY@~1@qv$d-A_}
zz{M`}VbjyfsGB#_LE2<;ESLGg-){-r3A=>!$;p)jJ^sO^wXV6tWNLNFCRiC?VWpx(tiv~G1&(}>wh6f+xQeI;TK0PfH`i-!R+I)
z9xJ4y^mVu~f@NW@P_%V@qd3{<@p;gHh(t74=~wW9>UibHtG1c{w5~X8dY}SHsF%G6
zdX!r9jGwi9?pl0%^@vXV4K6*XHTNGi<=2n{a#jARg8@ufscOs6zZPQA>z*?`_UP*t
zQ^&J`AI^`ZKnc+r!S{2Fl~Q@<%}W`hMS|Z=
zINr|^x#;90B34n($sL_03GUr7#SW8J>qW6TF|MlA1s+fPrC(5*lLoA!_w`BWl#Q!>
zRbKC^B)Ne!zilX4u<#-VUE3Oxwrf!4`5H1caX}wur+J@UyqtxQS)dv4i5Z=3^?h@dHWw?G~
z<6xDML0tH;eq(wP7W?a89D%@RbXU0MVuKJ?hczzmC4*%Pf=xKF&I=45SaNgnMYd~gg24EfRBvcLN0^K<*6KYqP
zFNb70+6{TnvYc-Pav@p@F}F481lymg2>Tk@xdn6})5t47NAPSErO3^VVPQhi!U3+D8ke>RN74{WqVcet26eYIr>uv4)c}
z;{WDOWuAx-qpR&A3upb{Fe`YL2N>fkZ!#OuJ_cAa@
zCUCWW41x-$7N4d*+^|~Rxa2`D!$=~0UBGCM2p5mJLMv{Hl4(2l#mAXk8{#K$1PqaF^{hm$IhY~QABVli8`Cy9(
z$GVK`=3*fUXaABLo^2|W0ZuW8|MdFZk24+DrdcUjAC=LbDP2SL8bmr1Ohw-&IB1hm!
zD2~W+!4VLxWE*%uGg%%Pqh7cJz
zg@GoO)=cnzPVBp{@FMz3jlCKjKxc5c2MePry2<6gs{Pug-`cZndFLl^W`~!j}wG%n6*}zdK&iNT5c}h}Q)#Kq+uQTC5&U<$zP95!DvxkRB30QO$6R
z=J8{K8Tq;@JE-@mN|9PG3$m=c4zq~Yy>qci7kf~*ZhFBLKk{jryt`6I;NDBAWi!bm
zUB=k*tf{_zZ_a%C2-elMnpPC5d@gycZ(VnBjztL7-Jnfkiu#FXWYRvhRaMGOuSm|O
zt1Dw@b=?h-hqT{k-3_zf*!4{wCC_&P*`v%!UDNs#%MQkAC^)
z{QI_*E$v+ut$q9D86vVAhwU^w)Q#zErEFPew5*hOp#R$3xKlfS65K+Ee;2o@dX4{W
zRTBT;Vz&2qV$r*@EuP{oA&?oo;hP;*;~{hW%$MtKE9;;3{+r!2c+eMx+aE7#R6sU&
zmf%Ut5bY2^QNiDL8nn2tMWj^wzZ)Sq8UEcy-fX5c
zl%ukA(AO5qKYr3=t?CtA&%+XkLB9KX4tGby&aQ(A)BF&vle`9b@fUM=hNjk3)1o%c}Z
zA?EBom&Swp!L->$D{Up2p^KA|WLrM51F?fPzB;~+Oy4C#KbDSUkBt9Yx9d*v&80Z10O2T^EOC0SHD3H3L&SfKOQ|ecm?>v8u4Wn
zPiR>wVt45HaeAwvqF0$8JF%^%rC~l3H^;s4KlAoUzZG4VZ||Vl`U&P1NO!@nzxJW5
zJ=ZE<#?~g#sw6QP_4}@7EQB~0rOz|}8yxd^pFs|Ud@Tb)RF-)Vm{$ff
zj=Xsr&ZNFXH|N%9w7R)Yvb*Sg0G@MI*H_D?`*IhBp}7XyhM|in8^3iI;4g`y*{6Cb
zVV687n1J&Ml9R0={eedq*4hQRTg+}3$L?fCfT$>5-u;6uWK!TXioU0?5_ZhhhjED*O{qx{`0gTi|
zc@1JW%9)GQX7PO^ezfqVZ60-Z
z%Qf=lA9tCmH9OB&x#40sE5E0VJrEM*Y?9WSoLSo=Pk8t^OUPVXAfo&tZPu5foA|j!
zcToSc!01xDXrhTZhh_Ln#QtrDx~I6I4bPtOhjM;Vz|vSvFfW|@v|D~{-*a~~^Td9*
z38m*amjFWK$hUb6!VO0u2zSa}JyIL2nPuSeYvTlYWcb
zqN(zqkXVPp>8(Nj#xdTtDS+7dMc7N*(w|^rd&6GE(CJwpg0y$^BFCa^Yg3K@#`kiw
z7F6)_b6P68EIT^f3vZ-GKlO)v`-2PLlMn<5vT*!8wQ{9!c>IOuY(?hV_@~&bmUPtp
z#qG`zDWFbO2Rsy8v49dXtE}I(M%|_=IQ-~SNHvX&ZA#O4YpkyB9pyutdp~<80=(r#
zB}?#0&DCv}TT<~kp$@u;mjBVt1)!$Y24S++O5*?^=H}4Py$Ml9uDkneAEw_X
zBzU!Yy78?8pkcP!tjcRiPyhaIOO~pRQdF)aYcKY6ej0=nWAyzne-W4vy`10i!q(TksEee{Y{O79^6TbbDI%URd
zJ;}Clro#EYf|kpXTci}a2)5vIggt?&q@_??w|)cD!L9dv|4tGm2%mAX>^kSivg9>}
zCet#qUwxtJNXsb7rFZj53svsynQ~eKiw!_|pQoxzXj#0`yv#cb4Wr)`c$aEfh5*rF
zw9Qesr1Mjg%~^=(<6wu*@tob=r3OcK7zg1csXEKBp&#ENe^#9W)D#>9U)
ztcTv;x_Y8o&LZqzFHmrZs(}&Zx@GjG`7z>8sEfJZzbR6fc5amJvZJlTyfv&8)*5%p
zHT(PJy7je^C)uFdiwLKDz7pMN1}4~K%6DoZ8UECG>s@L8fSUN$;W*NPuRa{IkSsn5#HTrk%{(3hvrLJ>w<*vOj
zrD?)iaNu^PiS@~{2pLh+Nu?E62?pU7{JyPa{UW@-tqx=bU
za(t9Sxw^kd+9X=I_F~q|wD8R_vqHcV-sj_J<3bpdz7^BrlnPzifXI8X6eYA{>5rQJ
zc#>Zd85}Z3jAy8-0RUMB?0l7YqM}UKThE*NqEY#?p4GGw+jq5#)jFlR1^+j6aN$q_
zpo-<|{51Zpi$70Wk7mmjg)11gpM1;POs`;%iY{k<79hpX+ZS@XF3YfuPx`!Twh{xU
zI(r(SXS|r7LawkvsYF2)X#6f+;inPnqT7~o<5wN0DD&9bKNqc`7>W%M1vS+rp8f^X
z7zb?V-NFV3d7#}a{>8KGA7;Vb>~tX7zJv7v4~saa?bHFbrKP(<1yF$35<9W1ZR}o=
z8Pxuvp2VwKxE@BDNjM_D0+)f)@>O+gd;)!<>)j+5ap|@gski!%G`HFgT8S@0>)h67
zoVL{FA25;6j}Ecn(@8u4%X8}_ClW!eZ9<)7qb@UXDK|v!kk7Kl&je#yCyDptCG47(
z3@M#|mffrX3ZN`XkY^D%8a!I!w{`~;UYxxko(b?138m*IWVa8W1Kerbo3?`>2gf@J1%eMA
zNaJqpq(X78fo$nHmA^USU%sx*M4}59UO>{I4ymcGIfbwZm=eq(yqx?MGwHuMAr5&ihL-*`_8Y2S6-!b|(c5UL2j~gMV;5=y2T0C*4Bf
zx>J}AEq7G&VV-~PXPw_C?x~AZ$_oAd&rIL!oHf2{gAD7~J
zuj7RSr$#Z@XVBOmd%n{GOq4pQzJjC>J4Y<)wnvr=e1rY8Wy>_PufW`(euf#pz*p$T
zf8i^4-J6h9q|hYt;5m+{%Ha}E1S~HT)6;{#(;C4s@;)8fnLll1U;N}G?dqt~ZETwX
zG$C^e(U%+s9LUZs3ese~Qec}DAQvs)P)IF%O||)4inv3t4w|9urlJ1*TesRPFR~Nz
zeXpm<>ilfpy2&5QYkAGXNcfVlS=Zj+ijSY)`V1G|e5QKvUrSN|ypt5_)b~=Nm(Vfs
zcl6pm{7|1?GeShYr&x+JcE@*Joz(t0to}g&hN)D`8fKeCmcHv?
zm(-N=dv3p_re(iQTtwBP#L4cHetpDf;?fc#C_++|OeQvt6K1dLy>+B69}vhKcq}pF
z71=pfhR_pim~_CfOq-V_uRmS>l;F>ql@cw5_tpTfAX=F~=Wxq#-v>por!T>l+Wc75kyTM!}BVapu-m?L3ej`1fp{Y|jX*Uo-@D{NwP
z$b5K>&`o3Lzb*=LCHbYt(8-rC-q%@JrU!^kvc(+ShYsH9j9lG}pFq^4Q57@f!zqz{
zwU4i?r&(>}6|?)xnIcb}x?>RZT{AvRr=)5GgkU!AAO=WYFhAF+Y4%&n6~7Xtw>M(y
z*6a^n@q?GXDSzvf^R&nJ5Su&~lX5QtK@yb9`r
z>a6hn{Fg`3*VtR%4&c3u7pM4RDX0H^;(H6!??)Yf({PmKP*Dm`=9_?}QRIOh!0fj&
zy&nDoG`Zis2@M?SAZVT0DQ^*&(Zh6H>h!v0un69=&}mWimT$E2O-?lD%l3$9Km4Hr
z5fUgG*!8Qi$hopA+X=O!OCp9f@w?tIBD9#Icn3_`%MJ;c^>3r
z^HU4u67I@MZp8Q+7AiN*p!BcbcFK#y~|mMaeNI#Z(BMCAha<@NG)#T}HU%QlLh-s9`2r%w}OVwNJvb&QIr
zlEvoIpi?tHa9ns?x%;M8(WF@NTYrL4O4E#Qa%c<1(w3g8
zMSnYCuRG0i;7h(m7N7W^*Rh8W%-c3LA%5|9*yzox*#&4;L6&uL8+Pi7?-o5%F)shy
zyR)*Z{!XzAd7i#s^Edr2LJkSiF_KxV<10BPA#m-tbGDzGowy=wFE6jbf=$v1;+o_7
z2&!mZV*GHLR!LI_eeuoo*(8lO!JK5z#ak49#S#?&zgYr+=bu8L!PE8;x)Wm`H0KqK
zs=qlq**8or5gB0Ev>CplzB@vgd(!OZ?sRvr{T`>vgEGUk;#u7q`f8^CPKi%&xd}XL
zq9$D|l4^AR_TA>)>Bmj{@mq638yAh|xs;ngo`AH)pT49IaeeuLo$U{x?Kd${nYI(QygDY|1IxTvvBDtki
zDuw#86|?20(d;GEXQ*RsceUqbSg@`S_V%`ZYJMU+Y9fe8(sPxVM+|!E#vATn=3}U;
z{F{vpSOUHKMLdZuiWF?O5&mqj-Uv%PWi>j3V&f_uen)&0|BA-nQ$Z|g{T5>;rqu`C
z1I3e9TseLhqj=eZwAhtYCiB=XpV9yO&yiYk(h2qTMg+M@`hFr&HDXb69M25pqIi|o
z@r`W6R=XotbKG5{p%|-ZVZR}+ASY2>1~u7jQ(4QpR(Nv61H8!0m$O{fn)+nj$V-28
zBXn&=rt899$~h$p`#ZDQv8Aa1BA)5ntv^$L^FzebG~2^GjSMCz*(h(j?j3A^!f=w0;-44hb00
z;~5Z^MKlZ}kexq1=Oodb3^Lt5N`KB4Zq*wZ)gKw9LgnCNolc|eb>@-f!ai%~rS}S{
z-sd~eV#KUZOo+fZ&fOlpXVFMC0#4aTifcMS!OkTP<0dClG#qQ#%g~b&(%wq0=ChY$
za1&E|!s3eHz#DX3M5y}n%rtUAV($sMg)j-0XMz7b&s2>wE6~wPRf4Cnjs=qb*DtTo+doO~+?G8CX7+
zYs_b3>JT)&LH2#Lt6Vj^)vT8$j6=O*i4N|EKpI2)8%D<#%Ggo{j`T}4S4TZ8QaP7au?NBr75QJDH+&x!y=^q)Hj`qmSQlHJ
z3bH{Uy%sH
zmMbIHH@$bNFZ(3Hl%9GyUv7|{t+ke6SQ-{>7fIte(Rw}6SETMEs9}cIIf&Jn7=cZY
z(WK~)v`itUxiDu3nl57hIB$jm&TXdfkyt~wnhhPj-m8Q7;LB4r*MuJovwv?Z#!=Av
z+g_f?6K=!(L>1R7Tvq!K9TTj(?QZQ&SC>JtX(U%Hfi)HP@DbugFe;i2r=smFjj8M0
zXVI~Ro(;JbDb%(Z02r;tN{3GZ0C@?e)E!V0!`B;8!%Yv>_Yll!FweBH43Io6*CLYruj{T0U|XQursdl9PeoQk5qr{$j<452;WORx$sM
zV^?KnzVedRlmm}5wsK!wC@&i`!WcABnC@}jp!pq1ZmWZ|2|WmY80D#uGrZjp7FMlsgpiQ
zqe1atMPgb1;mElCzXmh)?3ti>Kn-78+B46J`wtq~S#%eiBzY~ZJNPt|-_S3!8~8>t
zTSTRlzSK|zOB(Oa8b0G2Sp0}b=e`YVh)~ShV1mt55S@r71|CmZQZ%Nu%iRBgA7qic
zwB*Zmxa*CHb)*o)1UNVB*Rer64VfJ8?qK^>;o`25?Io2B_HSLk{rbuWl7~qFLceq;
zdOMPZZlIOBE2r^H%+Y92zSA!E^v`sn34^Pn(OQw=FxY!u0PkK{PgdN#%Uoy7IwG|}
zjRCFs?0>(M5k6$P)1$j9U6+%2G@E0ew{|7nSefwFTNhM{oDfvuVhaOhtZVBp$+FhO
z@`>mu4$;a01qNU1+;hToQ~O6=+$2SxsVZwd(#|&hd~ErZ#(QmjYrjeO8=7Xeq_JET
zO$OvTG)0@8ImUc%u3Zz~PQ{Pz3&9J7#ZgRMeRNI|skeEVrlw+K!)rj8JYF$4VOVw2fRk%wiM>keuEb>-A&e25YR4WG
z9y0qm*9K;~3ihUVm8W#BN=6ZuhVU!@gqkPN+Gw9Tw;=DiPtCk@AK*?s1dsn4=1?!f
z-m&@IhFXf<&%HAat(@Ku#T1W}6|z1|1L_yAv>R%tm!~A@JxXk`wkX^*B+NC7+aUs70nq)o!}cx|JY@Qe|Bi0CIxK!
zQmbR34YII{jXH#8@jZdr2=HU5i611M!Ne42QEh6^AsRMp^DOXY#IfQUg@f6gP6}M;
zghcP1^vr|hrF?Sy%WYddMB&VR%f71yz|VvPAZOb#_H1MBbmem^q%V!ZxnSz&6?f$2
zH(^)WDD`>=Z>WxsD8$1T+IP(v>hdRBx%XFfqeu}1FD6tfvKUZpIznB06o1NLu
zbTZ)R7?sf!I$zm7b*SAvdh2t@`e#VZUf;%&)J;pmdD~@wFL1p#xjV?hjbk^rw7fGb
zTIJ;%N0;tBV$qo3{bmgGVQy&4FNdoN&f(OH)=x53^4i#23OU52->Tn~HHx?<85P;*
zVW(<6+$(F1+G}4?e9EnQicfab=BFw;zPoWeaa)@lRW>RD5Hl9$X8M4$QcES7M_9qa
zg$^o@g#I~M_}Si4v3452V9F*l)0WBi!$O$f(Lc5^U^`0W%^QlK<=C+oOD&`D|%)X$2)9dpM&;-UxY?}f
zJ6BpmBoonpvQ(`2*BR<(lbqf_;0WA{dUo#fEnKy4_-JX8nUM&R;^V7g
zOU;v4`B_Br0oLE$-QVqQWY`<{w)K}`iuiu)vtO%FFNL|VcZyb-a_WT;1rXCznQ
z+;||Lo^Ms>w~Eu}JC)hRM#YoJtq?zEm}sx5XEJPE*P?qLdaZWBqHwgQOui-+sI`F2
z-0|IWrc1@^T{Y*9Bj%&Q+07BUUX)9de6cfCJ=Msd>+rfS;#{T;zCUOR9WAV@wmrAw
zwjWI0R%*UV`h9bD@sI0eaeb?E`H3<1TxRz%G(5!`TlMEc?0S+_gY%5fbws7PW8pfg
zuG)XFt;yw_SFM01yJ5#fZT{iA{Yt~oOp56vtkR2LH8JVs(Lx6y+i(J;saCNhgufc|
zA@3Ziv#Ho2=*rwcCjUf3EA3#8m4%hhj9)|cG5}If=o$Z{m!dRk5r
z4#Z9dmJkVh5?3qx$>oHF&YUiV0UzI7`b82X##?{=d3a1^>Adi#%z1mJQG}P*emHt2
zzJ}fVwN2L~YbXbP$F+_II0yNDtzqRV3Pxsd9!<-Y7K#3*SD-IS1t*~KJseabc+E{>
z+pn?U0Zm^osp$aYRw<&`&WTr3A+ESOO!DDnd8)39NFA+lZ~|d$|NTIt>0QRho6S#a
zAz1I_Tiew)=tL@S%;*I-^+9(PK+Z_bc?FSgD?0eSooBG0@`Hnst}_Gn!2I|~oaKnc
z`}GH>-ibKk9So{++}K64?lyg1v5i^nrRn_XB)?p__AGj}3`O32yX;%#^V1LfD4#CC
z@>smU5@0TM)8CJKR_|3uivPV4hgXK=$NhrP5l6Zst)L(2ZFQ$O1#r+)ntj|h^|N@r
zdEj+q2S&nQfa1fZ_r0X1qTQ*TByxG;zA&%VFcCC%YRslKT#T2Q{j*#ONHtJ=Vpq91
zn%~;G!98W27v%D<@z&Fp{IjIR9>ocSR^7IYOcuZB%>bnCXp(gr6r~=Mfh#>)ajXwk
zZdeQtgi|(cu}#yf_uEytAV
zJ3dfr@ONy1R3@2S6WIr@E!{#D9eR!N4f2|fZ0mYA)rI06{Brf7>l|frT3_GaXZ%Xb
zwHIznzu2A_Z}@9#8PrOpmyL8;D78>E>1yB|zCN@|8;Nn4>N(Anemgrv;H~aiOX?Lq
zX0MvNbjmrtiPAUG^gb>Ak6-?58(i-?@_{3+BVcDWAzbX2uv0MTS;KpW>|+(U
zm7b~QwqfVmEP|dM{5}7&u_x2sHo-S1aNY*KGJYm1<0$rnzlncb*vy2oD)?XJx=`Xy
zU3g;Ox{(jRuI5q8M|_s|;!0L=nvd7$G6K$5D0WY2S9kT~E@$I?)MopSxpoXB$McfA
zBewcB+M53~FK1VPJE6Jed`z8@?2lgT2d61^1>N>Z>zoy;Qu2()e(XyKvNCk|o+G2N*s~-syL$yM@=_K{0VO}Y@rxk
z<#Ui!z|F$5d=5?X$&H#{EHLoTo1`Ilcbw?>;_>1tAuM4+f7r*C+*NVV>je?qn1i(Cmf}4h;`ucV`FWktLXW7
zxk|@I%@qSXn&!`+tv#s=>=i73SBzR1oj`rlPpIz%hc&mzUsf+pwq9qYM#u?vA=rwuZW-VvRyKciz6>j~7Zi#83|O@k&X2Q&KH!A!X&DGuVg28qtu+Jk}&jUGDbps98SRL&=Kmyv&o`{
z<3YNUA!?ROkDDXy$&+$bU|=8bP0rBy$FI;e3qbd2+*Fc=Ryj3ajiaq9=&aD%zW>OX
zWvxMexTD?gvurTCabR_i&-Zsd-QtqW|Nd*LKNsBG6~YA}Kp~A!VxF&1L@iOe+PW-Y
z5<`w);0Zd7px6R)c+WB)YmIE%HOg~&$)4`$&3Sawf!Y1gM8b0GZoh!$Yl-2OD|p+Z
zUE8I}niHD!-XbN*G#VYG+t%;R0G+cYV>0p6MF7DO$PlI&;p2jh
zcLZ*nNSQ5vc7PVITQ;
zoFIHXpPghzN!Q3R;wxQ!U*?|ZI+A%j_YW>8yd-un$w6nfgdk9C{roh!$th#?EAX(@
zjgF0NR09TcPpS*RiWuAKe0@(FtS(<-hP2C1-;aND;gT01@Q}n*#r#IFZh==<+sr?)
z!7yA5=uDvjB~e18Om7I2fu+Ujr@e(77kYa)?@~~+s_FrQgNUnV)uBFNO>3}clIlmq
zjdTZ+R|*|InzX}QBc8io&x_J8R7%YcepB7wXxFT8wO16|;7s+eh7PM5=#
zxcNyRYJZfzPqKK36`V>uOm6mhWV+!!8GI8gr5zwW?u>u&=UD)V*lsday{}2o?o)HN
z<*3fTqo+A&dG*OzwSCM!??Q-%guf(vvD6hN(Yqt(ihf9{6R}7JF=7E>V(#4gyC3SX
zQ^>3FA>h|MY_50taYJC05YP`1!St+ZO~YkG*LDc>^G)apEd5vWdNJN$)$}Xd{?8q2
zn~ENl!mzP82?dT<7aNBZ;dWDuBA;f0BkcGZNlbaq8zgV|`?msE%yk9qm3Uz9zy3~W
zIy3eUarSPCpk^V+a|qR%z>6hao5@a1Yi2odZw&kIGkf;q^}3EvfVbu0~K8sDh4zn9!Op2geB&~?SI8CHEj?-kJ*
zg+&wL455RUwG&(OebsL|B;s1B)&<}DccBJ}(d^10bG}#oqQRlq%Go|dK$fn$Zb9^A
z*4mqEOw78EYjchz=SJ(iMMYBszI8+Qmira!Yos}cNT!*G$ND(oG=~XlZS9Awp<>g*
zC61%yEP6MWhIAmuo#lPjgJI!S@Nme)U7JT$DZ>+~`>-5-pn)a-X=nlVGm9^wn?#R(w~s
zEomu;bI#!{FV1yi?zx7`sr1vJlu{T9&2yfes+^
z_K@Ln0q4${QT@h%_KrPC$8O6G>*0IF#-*C3#KO-)HfK}%Mgf{hS#4O|x}=fs^%~nI
z9qVF_&vCOzgL8XM%W%lJ_>bv!_Y<{+^0K(4#ci4t5mEI&$g=C3NjZ@In%6Vj*6Bek
z5jbmFC23lvwB9rw&5-WoM;8lTh^!?~>p
z(lUO(p-Og_^!*#!l(W(1)VcAYjI+Mo6k;c9Du2`wIw)R36Ot{HBP_z?bA%(zP{mgS
z?o4ZJ^FNKZew&)#Ny=-VMg6hr{&W_TmEdCjH7*dlrBT%SX9A$IJUyR8J*sL>%8xir
z1n{3n!uqd>!P(G?({I?_)3yFZxy6TRsV*V@_W@DNXL-Iok$8bx?#DqQhLx=x7
zV#@LrPOut+u;*jaWz9@{I5Z-FzW?0{O=+Bo&L6!><~w@Go&>_^#pjjky5<&FSEhz%
zxi(FJ;>i-jCgNX^sgY{0I$9e_@SgMPDsJkgYE)JQ-Xvyk`L`OWyTt)=E(s3fmL9Zk
z-h!_eR_6(Z-aM9Dr5B7LGK`@wFw(1}=*vbW4H8=G7wam=go@o`YFQJ$m={zVHgf&m
zR+YXP*`1bx$e(o5=!n_k$)ClmLH*+(hK^U2L70&b2!Usbes}c-kpel>MYc{2zy<@k
zvuwS)?)EdVL?yl@A=}Un|6|d9AL36N9=pkRZI?d>#U|iBfNc+d)|&8zuAYobiJeU0
zl~*)=)~|8CiYBr+G?k=3u7mxjg9-<{R`O8@42SdvkliwEycb=VfUq9a-UKpk2tZDqYq8{Ma
z=QnIS)%(~x{w6FJjV>+93udk>e)9YVO8e9$p62oYY3r-QqWYqC2NeWC5dmos6_D;6
zN>Uo68$>DTW+s{|&
zYaUyYC}+$$NU4gorsZC$3B17ru|kO2S@Fq>?;fdGrwVD;2)RBFK!sRoIc7s2!F&Y4
zeOw%RZF-sRZ*cY?XK-y@q97H8!6h0=T`qW;x?~Q=CK1>Cp;AxP)D*N!gPgqPxcKD;
z#R~;@bP@COgs$yjx~IDy>nlZXTiIP`h%o$%f-9uRAb!m!(CwL*PRZX^@5rBVPOpUF
z1t+pNo0ZR}*9a5}FVlQE9+~Xj{0!yY@mQ}IiM5gQnvaQMWfD9ei^mn>_Bs68nw91*
zZmdt{afCtFA=(Uv%(p_N(u{$lLWle7e0}Bd+<;CA%^dvw%x1%Kd<=Tq6m1GpNzQH2
zk0xX)s1C*Jb=sJzdwJmjt{wiF1wJV-^!VW8jHEB{V0dwY$~ny0w<>}biR?)cOc@Vv
zZ;kf0Q`XH+@DT$Q@pf`6O;w0`h>LynQQl9c1lTj$y>q+MsGTsSlA>kZ0qGZ9wiusMUE@Bj^n&g}D21a#<2hHi68d29vHC;R
z;+d-rpB=^odo%y#qqbh;dXSB~#H`-vJ3AtW`k5_Cv{AXatMaXDyw*e;Ka<9bmpQdY
zM`tjf(9KQiY<#DP@Wa!X2?o`l1{q0g?Vfkx+u_Kky|Fp&cX$N`QKxN%5o8{9Peb)1
zN?IexDw1na6Pv##XDlCa=2)wR3f2pA(N*`p~ARv%}<`JiF7oXfFCPKBl>hoen=|74uXGG
z*OjQ`rd4B|2@89se&DW-IOdLgLLPyE2jCt+5;^72I`xNyKFGCJw2tp#0Y8-}vIi!@
zj1N!eQDQ;k>M!&W>w|YSp|Y%{<@lzjNF=TF{R!#1n#qd}-0m&xleXjlml=4~1N0S5
zwN0+R1uaFo(FLcvsM!Q!d1QM5LzqSu@0vgv{_2^uXC0T5R#H;U5T(3!cY3KRHf{zX
zRmUR}jYMRR8A-ShJEz<8)Z{)OQ?H6-25Y`=|?
zsTCAR(qzn`9E{+aX1L}FFJX2&TfRm8$$8H3J@vUbXiUSF-wRzr&7ZQD7wu^tg>Td$
zbohpX9xhG1v1C((`7FDGbJ+2d3Ps;FUoqonJ7ZXqKwYrFWJ=r`YC4Bp&WzSC{m)Qd
zQEf3(b(4Nms2|5oee1iIJeT#>QCHR5ff2Z#@wBPU9>!XtLyxr0pFvOr>q
z=3!UQX&QWIZc9!#@tAbl;Jx3}=xOYp|Pi={`B#QwMF7zZ&Dn_H4@3gd0P@5lhHp=bz0^
z8kCP+H&Cw*nbUp^OUF{#to``A??EK;jEB`|vod-2azO`HjM(ayJCE5D@wpJnTezr7
zkx`mCg8!Zpq&l@c5jADSJWl%CPYC2L4J*_xe8rH@Xg3Khjz>moDH&!eIut)IFGo1;
z8^sH}9aHDts9}tl8+lAQ=aiGR^)Bp#fz1&pmCV92XM^t}ugj-z-)PmLRYq9}WVvd!|-B8VQzFRx$gN>q=WwN>qlBT4s&n?|lUg@7Lhnzzj6jafh^Gi6fe(%w
zXDbc;l9LRZ7VfzSXQqg^x*1zb(7`Tvu1%4OGrt0?zc@Ys%9E93sknVCklV(3{OwvP
z1DQ#`;I*((m8nT=x@`LX4aWx(T~s|EIrsb}rO(c!%d3mY_p8=SP!e;(|4oypGaZFmO@2$ay>8T*Im-VgB5J{9k})q8o3*9sNsl7)`P$4}b*kpbA&Ge-$v
zK_(B|idbSEN1c%iz@z3NcemYMhyWND<3NUCz;wj9`az@)`Xq!SkoM|aka;UWB6Zo2
z)^@ro*nTlVUdx-_*}2By9@D=5FlB#U*Q&*I%Dk#o=I4ls?fpCbFRE3=ySNIN%s-TR_QuBsEU~VK7bP-#
zkbM{Qi6x=Gr-6}k%5O7?>1pr`$&M2(!0OC}cS%WVScq4+q%+$_`!rviW*}Lu&ra5m
zKCnwi+Qa~Bpd>f|INOE0tDu9%n^NwRE;Az?b)-6*XCuoP@)kTZ%G|nt9=jSM2;jn?
zBXWo>@YTfDDy$edE$K*8uoLNGQ#6rzVd!FYEo*FYOA^ivk8gE}mgP8XxwPnf={89jZb{4ou?}ln?QkVF~m8-*e=k*#IpnaD~7pAI6hrgD+`>g3~)Jy
z*UQDN4apY4JO+a3x@dt-Bnxe-?Wf^w{4cLbqdcBfHhiAO#0gDh#bE;dvuvMxQ|7ZU
zFfJa@`kib&#K&-YDn~1_^a3+K`stO6_u}LH&(JZUP9Tm((MW60!DW~eO?s%qbrw58*q-Z^(CKC
zsE2Mvt?Bcx$%I!|Zc2#M<39p(y)8<_o)vp+HD>;3Gh>vn`Q!1%FWCZ)_k|NJRazh~Y(%
zfQi&mm=meV8+FNVub7kxAjM13vnefsXqVx@oi6*9kE+iLrPZ&Q@5WL~?JVZsWR_&r
z&~dW&y$vec{lg~Pa1l?_TkmObbup$pSm6vrWJ;bE)S8@&&MlS`&4YIw+RoxKOQb$r
zwhoYp7l`>v7|YC;T7G~=X3APo4n#Z=%M|oFoB}u@*=C5Mf~0<~4&G(50X7J^-FYKLI)Bez;EU=?hIwQigjVJwUesjqM#%GO8+g
zRaGYz2EL>l0rDN+b_%#rvF=1fWA!8=u_Mc={|J9TpU_t-sQtsmSW=wxhthBhCrej6liEo$lpTU>2{TD4$3sAYzr3Hs&u!A+|P9l-&Y8@NWN&()K@)?*5ly^!qCR^E870QERBW*9IXE(5|Nu
zXl%da!+wC-odRXkrrnmj(guO+nE&@r)0~GS4e{A2w1=WA8Y38u=8L}F`Zv86^jSm@
z>kXQKI*+wQzEmdQIYrMlf7PI8-;hEp$55pHQC;iB0CJ!=K+^1soA7t<#RXqlboani
z*=rsG(8jF*!sOsqA}}klHkba2CODIxG$ma^L_VEG4Wa6w%U4|@HckGup6H)%_Vjfh
zYc7Ff-KpeksCr+BVU1N+iy!B3&w-aQCMvu8bcL@>G{aYpA>gM1{`iX86je<(%4K!xGkl-N
z==d=~ZqhQ#k-b^MWCJAIigWpcc|ZmN;(Ix&Vl2{xuj_p!%JYiyMvQ-eQtn04F+_qIBL~(FPk~$u_}iB=r*l7
zMlva`?6T&timJDy%(GR0_qNZzpLKl@0!=D$!h|RcJ1$+B8;Ecfef|SW!C-ym`x-hMk;*r}Re`Hc*N^&*XD8v>@Amca
z!RCbRdHtboxncmF0y~bStA=$v`M;e3`W*#m7t1BjFyKG@kJ8X*=jE9HWsQejxZ9Ib5AjKpJNHY-t7Oq)cSc$Bu
zLv<5Bv|-0V{kJOwf)65+sbC{U1Ganz4N_Uat_Wo_pS`Ym14^4=mdOKbM@`v~%FIGT
zi;bBLhIur(PpNlZZ?=?CS=ul$Zf2dR;|D-_c~xJK1N;5ox-|wrdG(-{`Ak^sPo@bK
z{YbB7B2^M=n-Q577q4*l!~P=!#@j&#Obw=o`A6VBnqDHHOvi_Y1
zhgQgRLBx(N6+BhyR?PMASy)&QS%Oy)))5&*$$a{Db~;OS)RZR>_a4PZKcAgEE+)pQ
zuTU@i7Ojz_R3!hgy{1lpleS2gnDPys#EiokQe?IhR-?(sNeurHGT+{K{P-H31KQX(
zx_+~>GP-NtbnI}lG$7Omo`L2)-HBJNbsimiTHz*(?b1E<0og@Mmj`o)wS&0MhLWM!
zSjImo?=E9L_tSo9XsVvCYUbz&U`jM?Gs*du;<{|m_+RKAKg8qpPXfgz!uAj8_jqQL9Ux{MY*rP<>
z>e~IB{($u>;Z5Qw?y>E2%dvXa%Y#-tbhxpZ4%R3;)L?i*qW7zB7b@IsUZ*MX+I|Ot
z=wcMy`6?Oe-|EyYHoMBviHzix-E|Z1he!P`Y>?b=l48)$qPG1px6t}mi#5qZmpH|?
zAO{w$abQ;btNZa!LJ_*>G}GO&3NHl0%IX-7l?P?GshLor>uB!;24kRC-_3eI_a}dU
zPdYJe$5Y|FS5dwG)>7eh5Wi!~
zouUW?*70A+!4IcjW1WP{nhbqO*_~URnV)s*Z)Ga*9&O^WQ5W{6OCkS{YGwmLnl6zB
zUX@+Z03MgKp^ke>w4**JY^j<9;t@Sn6Y=wG9M8=cobh+^@=h;@U(E4)4GxCVQ?Z2L
zlvCm3b9kZBph!=m0x5c=AJa#Am9Jc+8>@{m;&y^SoeA1
zWwvK-Jfnr?+R`MLCdN)`9xjmda$-|rE{{v{qU@vI#i|o|Rr^{AqjHL~PhN7Q|oG#M5b)839{v=KY%l^dneMO+=H=ZC0Eg$BD9
zp#u=$_43-WjDR6c{?&qRIH_zI->Ew<
z&r@8_i9GN_=2$@XBp~ev2bwG@6*|
zrUdvxIe&N$8i<~lX@PTw5O!_OB?Jbp%n;Y`sNmSIsp7HBsF&f20xsbnsIDGOSD@W4
z2E@9_f0=q#NR;CDtMW7dh_vxJN!zao)N6a*+xlkC_xz&y3&3%2`W{gHH$QNUkS3MYXW7?7k|z
zq2BBGr>`GYlPZ2o;imlnSd_(D!2O^%nq=h4vuORQ(hcHxIWc1flYzpW<$pU_J|Ik4
zF>u@NPLNyaZyPENI^1#FK7mqZ`qZb{V1a$=#yO}FF{fWDUX__Il4ZGQq4*Oo2AYW$
zrq_kE3`|^2LN!&yKC;ON&x)ECO0a@!y?4RxnY=icadCU&k-YDi`ghK01Kl#9@a5&Y
zLj%RBUqVhDxb~TQ>hu$5P2VYm44^}_8DH%P%!Xj+{8!}1&X()phS-$eyY1*J!Q+O4
zA;~TkXQWod17`-I09zkuYraL#?AD`$FsVh7E+jIePuP|ys$OgCHzBhWD>mB<@pe!^};WBf_gS){}*(B2cAE=o}4&o
zEcXkHI1Yy;#jYo?*j7USzV~f_4hZi!(?Hh0^SU%d<)u9_I6v8vnIb)s``tY}%tHL`
zr0kou_{>ToW764jYVX8#Chu0082zcR^?ycgGc65ce*>SuSSJ`It
z7+f{%8U*!M%`v$Ba-?s923$v($*3aaDkP(K-hGVM8yy^>5-?SdE>Esu?lggDy~WB=
zOMkvvS#y#j3!AGh6q^$@lPM#tAXZb6HBguQHo#h6*43@KgJ3mF-55J0l87)U=;Cl*f_wSj|)?GwHwtXHP
zU{bSxZi4S`eOc7xT-v?^VxFhj2QM){QAW5El6)ZPGc_DervWm3HF;NR-X%XS@PDLj
zp_?wmYMlIF0xyJ+yb*W*ZxtuzU~XqCsA+Twn!`pE69X3AI85Ue;pSu6$j`l`qBidXbqg!vtCKO(Nd@CA|CX%&{i^bM+D$r9p-E
zMy*$W_;($Hw2&5a6Wq7gFL3zl*vdf4Y;h{Xk-ebKMQ;9`=GoDHrGhrM40)H?K<$rs
zuSb23ey{y+5HJu(DDuWY6a3FWYlvTSVv8^8im;u%`dc%?Rjbm#+3e>rwWgkB%S|^3
z#F|WM2O*Drz>Mj|t2q$E=7yfU5Q{0V2^O=>x_c<4hT4VR&6XE6U)>TLeq@If4b-^*
zomc#42P-irqa?hYP5;N#)8FIDW;igs(%Cy}VSl~iqiy5uA3b?Llt`)GH9de0{P5LB
zhm4UUwe&o54a9rxP_JBn7yP4Ke@*_PS-Pd!2Gd#3N8CD7-*l6-X$zt)xO9YMNWmZOxfCieQ
zsV3AVofC=Q%-PZB`G{RWW^}UmU?VZyH6~YyX3IK}tmrugw-P?S&|Hf^X0>OxNVw;&
z^3Y!zcIrD4_2T#ddFooqj>MvUE`Upi7=GR@-x1sb&a0`t+{7c0sd#?^2Q2FL=`M~<
z%%;=eHOduL4e9#bSRo!K*$%96l&fS3Sl0kQFh1LaA$h6=oW7xp1uk3Lt|%@l-G>Yv
zCfpVBhqDvH$S2Y_OpFISIBy+1>Ros`@{Xi&dCAv;97Eil1aXboqwQ|T1_D=M)m#5l0PAI0h>&(NZz`Nf&&(zKm;?Tnj?I9
zt47bePut3b{M%=+b6sVuPvgV>!toN*b`vb$#~X~cgOQ_^m*DWP^v5(oh0$eR>c}>k
z9Bv(xP_Od}-X~StCC4(vhvRX&4L;n~m!_pBTRT
zxMVp-HAyKf>?i6x?0F%L`pBfBDj@w7vp!R)HRJ-{N~o0d*;BrlGI7LyqkOU`@_Ft<
zUxia>@pWDzk`f6y11%H+(qMzYd2j6aK@c`~2|!DaGCRV!PC{cBWpT{dt<&
zXl6pFLYM%l?%K}wB1_rJWl0abx5hFrC*0Ys7mg$jMVJN?$`OiDT}xI;L_u(Mw2wtw
z^vOxDDlS&b&FA~a7&o19ScTK_)Vvssll|%<69n8t$1hq-_SyqPH9j>hw&&@Tk;&O##h(#FVdVq(BowE?&-K$hdkI10i~
zcEWVRSTRkqJZw0i4I&vFCwD;xS3bKeD3bkZokJ?@vpbH342t02dHsFiv>5F>N=;A(
zQ-!x|yf?364|)Y1nWA|Ya<3v-Q#)2Q%ekg53nm1$G$el1ah>ueCo3;5A$>y?Yul{f
zNkjig5&kE&%+KZU*CBJMvW1K$e|Oi_u~}llEnplYhO9u8)A+Dfl;r6`$Mbjw`Nh|93dpL*cg~gUqFUhyGQc;0y~ieJPQ9N?0*=
zgwLz`w~l(U>hDWKSpD$QkBzlrB#(itYK2@{tXU@P>SA_m`_SQs=6t*@IgrSxBK}L=
zGuM{k+F7O{EtF=?lBTxUlz0UsaZ+19C;w7xlZ!N{=)ED#W6RF5mfvD?Lrs5`>B@Q6
zYYO{;z>z*J>z~Vm0A@p3Fui=`vIMWj-G)3;dzl+CN+|-~8*rBH@{K&6DEnYMA^0wupTouuTyh}%aokawT%><#csrdylnPVF@atoX?TRW0R%wJ-n-G=@0yKMLx(+kTqn=Du!p6>7$@iOo<^u|LX8^8XL3N#$TFt$G)w`pT>7T8StfF)YeGa=
zxBNcx#jQR;Ax2XxheDJj>^C_mVez53a)UqGPlc8}UTS
zLOj@Ua=7o@o{8s534HGGaY=!C7eH+~LAF1g41$TpZN&{<1b!6x56nU<1MnST-$q%B
z{c!BLhsc2S*CX8(4Z*?T&bZ=ax9;Q`6^*ky;8<9In=&`0FPgvJ-3cW+2sTJEJco%9
zYM{yyZC=P5H2nI-G2n$|R4y5pf14Zmd7iCNFbojid{D|9(~sZJOE0V}R9SNG7zH=p
zgt9=OOIOIYtOwf%<_)y}aDqQejF=KMARLPS;DG)C+eeK9xA+Bs(%b}Nfw7FLuE9{C
zeNqWha5Wo#6IJtpkJZ1n=-GH85Niq4k$h-Ch;ut57~KQj>4+Nb_+5&{(GwZ>l~F^3
z?>MmYhcJ^JM*61LE={2uLlXcJArPT}d%JgV#CpK#X$BZAr08sXOdJH@nHul*vC0U%
zVN8|E_nxJW)-io3xhtg2rBz_I{<9X!MPAeJz3+XhO=|&Ww-6)iu`-jIo|qPKh*Z|_
znx3%}@f3T_L+@7M1@*ioORuT>0==~%YHV@;o$5Yh!nRDLk*u1QG!QZ3m44v%PpU9o
zK!<8_sS@mTH{sB~N*beYf!@-TwW_>S?Z3&8GVTF$$>o`=tz%A+=a#*4I-vnT-oM(y
zxhz(F{I=)0G%yEVYHzH>%sV&O##R4py(qUAjT<@4>W;q*4j!A}OXZ}BS6V9beA1qw
zTH9(KE+1W`EL=w@3WRD$3&N)+7B#j*XGWd#ktH6c4@GSw@f7FP@c)-WVKv?@eceepjMVTyc@e#X%5kDVsAw
zdA-Gh@L3-Mw7ZsQ{?|VLU-l=7Klxb2V>vipN{s91hL&sQDR~<>wmbiEh9s}!(eEE}
zyufc!o6k)VH7+F%8~xBXxvr4f^^@IHM^-CZj5dF}SHZQ2NRyJ^Y`{ZxXjWxqswNA#
zS0g?=fzM3@wxp+v7_(GZ`H>YJ{xVIzC%lvJz=#?js30BE{xqP(r{MEjDb_Ao;w<$X
zuwRoa7`P*KETz|sc
zeHB=JO-~$3#AlIf{{i)#G}9#;AC+O)R{|^wvb5NSquo+{8Ut$o=f;D+Gd~@u&npfW
zuxk@Cfn0HK!9sGQhl7cc`n?rX7}iw>)!vAAS<4ohQIB*zq3!$PshS+
zZQ*N)x8Yw9jNnrnK33RpGR{*-`e1kY>x*2+ZcbptVwF)XW*89*tO*=
zh9O(DQ@=p0dlLfKKQ|9dv#Eg}DrEAcawQ3tF8U26X)!V&yxrW7k%ud;0WCkWOf9}X
zco?3Tx?$I$ySR(HLa{bEXm_r+tSH%9H!;{mrz6O+uI9ep{xCxf^Eg-2Cw7noqHyjf
za@k_wO1n0iphtfa{q(ZsByxDhnumb0zU=*GGnK!y_ei)q|M7^~vf^Q+Jj-O^)@0>E
zcSX3cL{1q-+Q8b*M~P=QkqP2BTJH~;i?fPwu(Zb$TRQQXzK0$MoYEc9a2$6-n#=ph
z2IS{qA?K?&;F?zLWS_c1fPfk%{n1H+lU1$0&6;XX!TcY(dcO)J&3(??)dHsR0#cQ{
zEPvN5%2JdG{l37$#*Y4~<80iwfe9m7n|DrU>RI^em04gw7UMQUdmG5`DEWY@ihzK4
z_9TxTwM$9Jb}-lPeKh;P&<{^AclO%P0!vq0NA9RI_0-2l^0P!iX_$PB%T2#AD$cO*
zyW!qnuO8M>+2E7rv*{M8cKlo1#d6Uwp(060K3`T|nq0(cRa=B-udvJZF5@~<;0DSW
zGkhm46O!D100D}_>|FLdCt4O4i*!o;GB4|#^7v`#5Ar;F9myh?(Gv@O#@?>Av#FTn
z=%-YZhIOluk5?Eotm~$@V3+ZrLT(MB-SNyHT>Ax$PX`e;8c*5Zl}uW^7$WNqxC@2e
zjo6*I6_Gdf2EWJ2%_ekm<#e|YQ|j*U=A7=5WLf3=T{F2uWrwk=boKnyZTs3$F~xk$
zCiZElRyw*$LLP2F1NwD3nP?U{>Xu_Un&AqOxwzskT?YTvOt=XH2tKuXdU(mr#&Obd
zReWS*kltDdg35rOzdu-=eHZJg#gF;b;A=DyOureT}Fy
zr}Mj)wSlfL?xfy`$}EGCI4{&Z_$1Py&5%DZkES|9RDiqESiY^J!lel<{w
z`QatmPaYa6&fl^fa>8Qs>6+!xu<%zbj?u2L^bktO(ZKVZ7?e$UwVAhp((wj~s7n^Q
zAQ>Jm^YfeS7oo6Jj@-|fbCrPv+89kv229a1kk=ASgsgbh*~9e;IArriWYy2xq&lRk
zNB#_Lt)@4B4c461K95ohh++V)bj>YOpo+%A^AnF`)rwf?pKJ8Q_61aEPYE)vmjvB)
zozUL}kPV-rz3}q>i|ljt(4TxO(zZUkuDUDvYB&>dye5cY?857M#ejl-lXu!1YZ#Wy
zo`Xbq9H+xBo9s(-nqCqZcO@Yi0`B1sc4@}S9V}!y)kl!Qc2@UUee^YX9n^Fz-CCcO
z3naVe$#(0Z`hWVU8~Ui**6}{);%Bcj)2^!75{(L?@!CoAy^22>x-hcvJ_wJGvmN2$
z*7-n9TCR}?cMAH2*DvwFerDS=$o^I_DLwX@n4D38861=;PCm&GUBh><9(zzI01NxA
zmcM?!A~JGy!f%a;wE0cTtsBJv6cO_m{Ov0#8w0pU`e1L|FQ#mqcjS7->k6^(zro!W3)aX%*Y$wO*y(o{7HM6bOYbI*j%g=)>8?}+
z4s|0dJ?k1e>*008cwior=wiqx>FCly9lk0td#h}|*adSGmBZ$JYX@gK}phB9ekI?g28
z>QRhcMizQxKf?Y(&^$7(vpHWyo<_z^Euc1aR%TE-=#-b4`o3#k$q;
zVR8_(e@{C#HF(Rq_xtbBwd2s(aYq6a1@d0Pybkv@F4FU{`OAPr9B{D|Z$b^8@_FGo
zmf1#>fBUKUX+PhLeA>x3vy>ai*mL)SRzJ4wWmWU~HTmsRgo0vR=gY=qTAE&?Cwj&=
zT7R!N{g2s@^90SPq&)h?Ga(i*{Q=2J6QRb9Z!p&#e=hY+@P~V&-_w`4eevfQhit*1
z+C;P9YmccxQ*FO3JZIDISS=`uS4EjGsb@9*BwN?9dMPy7A$p8_ok
zr5RAH&#QyXqIkqlNCY4WHVzsGJ?VeAGX4Kew%d?B;>OYXf9Chb_6C7&W)C2){quPK
zo$&wvAOD#ypxZUPecJ!`$N!e_e-ClHuK)YT|2^*Q(^96c?t?&ASGv{?
 plt.show()
 y_probas = Random_Forest_model.predict_proba(X_test_scaled)
 skplt.metrics.plot_roc(y_test, y_probas)