From 1dc910ff058fbaa933337678e80d144e07100f42 Mon Sep 17 00:00:00 2001 From: mhjensen Date: Sat, 2 Nov 2019 11:27:24 +0100 Subject: [PATCH] Updating decision trees --- .../html/._DecisionTrees-bs000.html | 127 +- .../html/._DecisionTrees-bs001.html | 125 +- .../html/._DecisionTrees-bs002.html | 128 +- .../html/._DecisionTrees-bs003.html | 125 +- .../html/._DecisionTrees-bs004.html | 125 +- .../html/._DecisionTrees-bs005.html | 125 +- .../html/._DecisionTrees-bs006.html | 125 +- .../html/._DecisionTrees-bs007.html | 125 +- .../html/._DecisionTrees-bs008.html | 125 +- .../html/._DecisionTrees-bs009.html | 125 +- .../html/._DecisionTrees-bs010.html | 125 +- .../html/._DecisionTrees-bs011.html | 125 +- .../html/._DecisionTrees-bs012.html | 125 +- .../html/._DecisionTrees-bs013.html | 125 +- .../html/._DecisionTrees-bs014.html | 125 +- .../html/._DecisionTrees-bs015.html | 125 +- .../html/._DecisionTrees-bs016.html | 125 +- .../html/._DecisionTrees-bs017.html | 171 +-- .../html/._DecisionTrees-bs018.html | 198 +-- .../html/._DecisionTrees-bs019.html | 200 +-- .../html/._DecisionTrees-bs020.html | 190 ++- .../html/._DecisionTrees-bs021.html | 359 ++--- .../html/._DecisionTrees-bs022.html | 234 +-- .../html/._DecisionTrees-bs023.html | 211 ++- .../html/._DecisionTrees-bs024.html | 327 ++++- .../html/._DecisionTrees-bs025.html | 178 ++- .../html/._DecisionTrees-bs026.html | 247 ++-- .../html/._DecisionTrees-bs027.html | 159 +- .../html/._DecisionTrees-bs028.html | 152 +- .../html/._DecisionTrees-bs029.html | 208 ++- .../html/._DecisionTrees-bs030.html | 160 +- .../html/._DecisionTrees-bs031.html | 154 +- .../html/._DecisionTrees-bs032.html | 184 ++- .../html/._DecisionTrees-bs033.html | 189 ++- .../html/._DecisionTrees-bs034.html | 201 ++- .../html/._DecisionTrees-bs035.html | 182 +-- .../html/._DecisionTrees-bs036.html | 184 ++- .../html/._DecisionTrees-bs037.html | 188 ++- .../html/._DecisionTrees-bs038.html | 188 ++- .../DecisionTrees/html/DecisionTrees-bs.html | 127 +- .../html/DecisionTrees-reveal.html | 123 +- .../html/DecisionTrees-solarized.html | 184 ++- doc/pub/DecisionTrees/html/DecisionTrees.html | 184 ++- .../DecisionTrees/ipynb/DecisionTrees.ipynb | 1308 +++-------------- .../ipynb/ipynb-DecisionTrees-src.tar.gz | Bin 195 -> 195 bytes .../pdf/DecisionTrees-minted.pdf | Bin 268642 -> 274455 bytes doc/pub/Regression/ipynb/Regression.ipynb | 32 +- doc/src/DecisionTrees/DecisionTrees.do.txt | 65 +- 48 files changed, 4469 insertions(+), 4148 deletions(-) diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html index 71dfb146f..ec2e98fd3 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -210,7 +233,7 @@ MathJax.Hub.Config({
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University

    -

    Nov 1, 2019

    +

    Nov 2, 2019


    @@ -234,7 +257,7 @@ MathJax.Hub.Config({

  • 9
  • 10
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html index 564622280..faef31394 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -239,7 +262,7 @@ given some assumptions, make predictions about the target feature value
  • 10
  • 11
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html index a656feab8..f9b5ef95f 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -193,6 +216,9 @@ MathJax.Hub.Config({

    A typical Decision Tree with its pertinent Jargon, Classification Problem

    +

    +Figure to come here. +

    @@ -211,7 +237,7 @@ MathJax.Hub.Config({

  • 11
  • 12
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html index 3b71a3c6c..a927e32dc 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -225,7 +248,7 @@ node.
  • 12
  • 13
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html index 6241edf93..667e72b4b 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -226,7 +249,7 @@ Then we are essentially done!
  • 13
  • 14
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html index fd01cc25e..578b184c6 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -305,7 +328,7 @@ plt.show()
  • 14
  • 15
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html index f4928d223..c5f8301d4 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -238,7 +261,7 @@ within box \( j \).
  • 15
  • 16
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html index c95e0e19f..b6521d733 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -230,7 +253,7 @@ better tree in some future step.
  • 16
  • 17
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html index 3fa62eb1f..f4cf44cdb 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -263,7 +286,7 @@ region contains more than five observations.
  • 17
  • 18
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html index e029c1c2a..9b05825f6 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -232,7 +255,7 @@ parameter \( \alpha \).
  • 18
  • 19
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html index 6429bb8bb..6fead1309 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -245,7 +268,7 @@ subtree corresponding to \( \alpha \).
  • 19
  • 20
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html index 6f3e89514..95e3b593b 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -241,7 +264,7 @@ MathJax.Hub.Config({
  • 20
  • 21
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html index d990bf268..56a8b4c24 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -233,7 +256,7 @@ fall into that region.
  • 21
  • 22
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html index 7a5788c32..00541fd4e 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -238,7 +261,7 @@ than is the classification error rate.
  • 22
  • 23
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html index 04481483a..adaad6832 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -264,7 +287,7 @@ $$
  • 23
  • 24
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html index 852f360a6..e061afb47 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -255,7 +278,7 @@ os.system(cmd)
  • 24
  • 25
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html index 42c56ab71..f071e4f4e 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -246,7 +269,7 @@ os.system(cmd)
  • 25
  • 26
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html index 2bcfc46bd..cdc844132 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -191,48 +214,16 @@ MathJax.Hub.Config({ -

    Computing the Gini index

    +

    Algorithms for Setting up Decision Trees

    +Two algorithms stand out in the set up of decision trees: -

    -The example we will look at is a classical one in many Machine -Learning applications. Based on various meteorological features, we -have several so-called attributes which decide whether we at the end -will do some outdoor activity like skiing, going for a bike ride etc -etc. The table here contains the feautures outlook, temperature, -humidity and wind. The target or output is whether we ride -(True=1) or whether we do something else that day (False=0). The -attributes for each feature are then sunny, overcast and rain for the -outlook, hot, cold and mild for temperature, high and normal for -humidity and weak and strong for wind. +

      +
    1. The CART (Classification And Regression Tree) algorithm for both classification and regression
      2. +
    2. The ID3 algorithm based on the computation of the information gain for classification
      3. +
    -

    -The table here summarizes the various attributes and +We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the gini index or the entropy to split a tree in two branches. -

    -
    - - - - - - - - - - - - - - - - - - - - -
    Day Outlook Temperature Humidity Wind Ride
    1 Sunny Hot High Weak 0
    2 Sunny Hot High Strong 1
    3 Overcast Hot High Weak 1
    4 Rain Mild High Weak 1
    5 Rain Cool Normal Weak 1
    6 Rain Cool Normal Strong 0
    7 Overcast Cool Normal Strong 1
    8 Sunny Mild High Weak 0
    9 Sunny Cool Normal Weak 1
    10 Rain Mild Normal Weak 1
    11 Sunny Mild Normal Strong 1
    12 Overcast Mild High Strong 1
    13 Overcast Hot Normal Weak 1
    14 Rain Mild High Strong 0
    -
    -

    @@ -259,7 +250,7 @@ The table here summarizes the various attributes and

  • 26
  • 27
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html index 84d46cc6c..839564a4e 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -191,79 +214,8 @@ MathJax.Hub.Config({ -

    Simple Python Code to read in Data and perform Classification

    +

    The CART algorithm for Classification

    -

    - - -

    # Common imports
-import numpy as np
-import pandas as pd
-import matplotlib.pyplot as plt
-from sklearn.tree import DecisionTreeClassifier
-from sklearn.model_selection import train_test_split
-from sklearn.tree import export_graphviz
-from sklearn.preprocessing import StandardScaler, OneHotEncoder
-from sklearn.compose import ColumnTransformer
-from IPython.display import Image 
-from pydot import graph_from_dot_data
-import os
-
-# Where to save the figures and data files
-PROJECT_ROOT_DIR = "Results"
-FIGURE_ID = "Results/FigureFiles"
-DATA_ID = "DataFiles/"
-
-if not os.path.exists(PROJECT_ROOT_DIR):
-    os.mkdir(PROJECT_ROOT_DIR)
-
-if not os.path.exists(FIGURE_ID):
-    os.makedirs(FIGURE_ID)
-
-if not os.path.exists(DATA_ID):
-    os.makedirs(DATA_ID)
-
-def image_path(fig_id):
-    return os.path.join(FIGURE_ID, fig_id)
-
-def data_path(dat_id):
-    return os.path.join(DATA_ID, dat_id)
-
-def save_fig(fig_id):
-    plt.savefig(image_path(fig_id) + ".png", format='png')
-
-infile = open(data_path("rideclass.csv"),'r')
-
-# Read the experimental data with Pandas
-from IPython.display import display
-ridedata = pd.read_csv(infile,names = ('Outlook','Temperature','Humidity','Wind','Ride'))
-ridedata = pd.DataFrame(ridedata)
-
-# Features and targets
-X = ridedata.loc[:, ridedata.columns != 'Ride'].values
-y = ridedata.loc[:, ridedata.columns == 'Ride'].values
-
-# Create the encoder.
-encoder = OneHotEncoder(handle_unknown="ignore")
-# Assume for simplicity all features are categorical.
-encoder.fit(X)    
-# Apply the encoder.
-X = encoder.transform(X)
-print(X)
-# Then do a Classification tree
-tree_clf = DecisionTreeClassifier(max_depth=2)
-tree_clf.fit(X, y)
-print("Train set accuracy with Decision Tree: {:.2f}".format(tree_clf.score(X,y)))
-#transfer to a decision tree graph
-export_graphviz(
-    tree_clf,
-    out_file="DataFiles/ride.dot",
-    rounded=True,
-    filled=True
-)
-cmd = 'dot -Tpng DataFiles/cancer.dot -o DataFiles/cancer.png'
-os.system(cmd)
-

    @@ -290,7 +242,7 @@ os.system(cmd)

  • 27
  • 28
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html index 0c7d2dc4a..d493197a8 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -191,81 +214,8 @@ MathJax.Hub.Config({ -

    Computing the Gini Factor

    +

    The CART algorithm for Regression

    -

    -The above functions (gini, entropy and misclassification error) are -important components of the so-called CART algorithm. We will discuss -this algorithm below after we have discussed the information gain -algorithm ID3. - -

    -In the example here we have converted all our attributes into numerical values \( 0,1,2 \) etc. - -

    - - -

    # Split a dataset based on an attribute and an attribute value
-def test_split(index, value, dataset):
-	left, right = list(), list()
-	for row in dataset:
-		if row[index] < value:
-			left.append(row)
-		else:
-			right.append(row)
-	return left, right
- 
-# Calculate the Gini index for a split dataset
-def gini_index(groups, classes):
-	# count all samples at split point
-	n_instances = float(sum([len(group) for group in groups]))
-	# sum weighted Gini index for each group
-	gini = 0.0
-	for group in groups:
-		size = float(len(group))
-		# avoid divide by zero
-		if size == 0:
-			continue
-		score = 0.0
-		# score the group based on the score for each class
-		for class_val in classes:
-			p = [row[-1] for row in group].count(class_val) / size
-			score += p * p
-		# weight the group score by its relative size
-		gini += (1.0 - score) * (size / n_instances)
-	return gini
-
-# Select the best split point for a dataset
-def get_split(dataset):
-	class_values = list(set(row[-1] for row in dataset))
-	b_index, b_value, b_score, b_groups = 999, 999, 999, None
-	for index in range(len(dataset[0])-1):
-		for row in dataset:
-			groups = test_split(index, row[index], dataset)
-			gini = gini_index(groups, class_values)
-			print('X%d < %.3f Gini=%.3f' % ((index+1), row[index], gini))
-			if gini < b_score:
-				b_index, b_value, b_score, b_groups = index, row[index], gini, groups
-	return {'index':b_index, 'value':b_value, 'groups':b_groups}
- 
-dataset = [[0,0,0,0,0],
-            [0,0,0,1,1],
-            [1,0,0,0,1],
-            [2,1,0,0,1],
-            [2,2,1,0,1],
-            [2,2,1,1,0],
-            [1,2,1,1,1],
-            [0,1,0,0,0],
-            [0,2,1,0,1],
-            [2,1,1,0,1],
-            [0,1,1,1,1],
-            [1,1,0,1,1],
-            [1,0,1,0,1],
-            [2,1,0,1,0]]
-
-split = get_split(dataset)
-print('Split: [X%d < %.3f]' % ((split['index']+1), split['value']))
-

    @@ -292,7 +242,7 @@ split = get_split(dataset)

  • 28
  • 29
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html index 5e0032142..bde376c17 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -191,39 +214,48 @@ MathJax.Hub.Config({ -

    Entropy and the ID3 algorithm

    +

    Computing the Gini index

    -ID3, learns decision trees by constructing -them topdown, beginning with the question which attribute should be tested at the root of the tree? - -

      -
    1. Each instance attribute is evaluated using a statistical test to determine how well it alone classifies the training examples.
      2. -
    2. The best attribute is selected and used as the test at the root node of the tree.
      3. -
    3. A descendant of the root node is then created for each possible value of this attribute.
      4. -
    4. Training examples are sorted to the appropriate descendant node.
      5. -
    5. The entire process is then repeated using the training examples associated with each descendant node to select the best attribute to test at that point in the tree.
      6. -
    6. This forms a greedy search for an acceptable decision tree, in which the algorithm never backtracks to reconsider earlier choices.
      7. -
    - -The ID3 algorithm selects, which attribute to test at each node in the -tree. +The example we will look at is a classical one in many Machine +Learning applications. Based on various meteorological features, we +have several so-called attributes which decide whether we at the end +will do some outdoor activity like skiing, going for a bike ride etc +etc. The table here contains the feautures outlook, temperature, +humidity and wind. The target or output is whether we ride +(True=1) or whether we do something else that day (False=0). The +attributes for each feature are then sunny, overcast and rain for the +outlook, hot, cold and mild for temperature, high and normal for +humidity and weak and strong for wind.

    -We would like to select the attribute that is most useful for classifying -examples. - -

    -What is a good quantitative measure of the worth of an attribute? - -

    -Information gain measures how well a given attribute separates the -training examples according to their target classification. - -

    -The ID3 algorithm uses this information gain measure to select among the candidate -attributes at each step while growing the tree. +The table here summarizes the various attributes and +

    +
    + + + + + + + + + + + + + + + + + + + + +
    Day Outlook Temperature Humidity Wind Ride
    1 Sunny Hot High Weak 0
    2 Sunny Hot High Strong 1
    3 Overcast Hot High Weak 1
    4 Rain Mild High Weak 1
    5 Rain Cool Normal Weak 1
    6 Rain Cool Normal Strong 0
    7 Overcast Cool Normal Strong 1
    8 Sunny Mild High Weak 0
    9 Sunny Cool Normal Weak 1
    10 Rain Mild Normal Weak 1
    11 Sunny Mild Normal Strong 1
    12 Overcast Mild High Strong 1
    13 Overcast Hot Normal Weak 1
    14 Rain Mild High Strong 0
    +
    +

    @@ -250,7 +282,7 @@ attributes at each step while growing the tree.

  • 29
  • 30
  • ...
    • -
  • 42
    • +
  • 50
  • »
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html index e67e33add..dd583f072 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html @@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source ('Classification tree, how to split nodes', 2, None, '___sec13'), ('Visualizing the Tree, Classification', 2, None, '___sec14'), ('Visualizing the Tree, The Moons', 2, None, '___sec15'), - ('Computing the Gini index', 2, None, '___sec16'), + ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'), + ('The CART algorithm for Classification', 2, None, '___sec17'), + ('The CART algorithm for Regression', 2, None, '___sec18'), + ('Computing the Gini index', 2, None, '___sec19'), ('Simple Python Code to read in Data and perform Classification', 2, None, - '___sec17'), - ('Computing the Gini Factor', 2, None, '___sec18'), - ('Entropy and the ID3 algorithm', 2, None, '___sec19'), - ('Implementing the ID3 Algorithm', 2, None, '___sec20'), + '___sec20'), + ('Computing the Gini Factor', 2, None, '___sec21'), + ('Entropy and the ID3 algorithm', 2, None, '___sec22'), + ('Implementing the ID3 Algorithm', 2, None, '___sec23'), ('Cancer Data again now with Decision Trees and other Methods', 2, None, - '___sec21'), - ('Another example, the moons again', 2, None, '___sec22'), - ('Playing around with regions', 2, None, '___sec23'), - ('Regression trees', 2, None, '___sec24'), - ('Final regressor code', 2, None, '___sec25'), - ('Pros and cons of trees, pros', 2, None, '___sec26'), - ('Disadvantages', 2, None, '___sec27'), - ('Bagging', 2, None, '___sec28'), - ('More bagging', 2, None, '___sec29'), - ('Simple Voting Example, head or tail', 2, None, '___sec30'), - ('Using the Voting Classifier', 2, None, '___sec31'), + '___sec24'), + ('Another example, the moons again', 2, None, '___sec25'), + ('Playing around with regions', 2, None, '___sec26'), + ('Regression trees', 2, None, '___sec27'), + ('Final regressor code', 2, None, '___sec28'), + ('Pros and cons of trees, pros', 2, None, '___sec29'), + ('Disadvantages', 2, None, '___sec30'), + ('From a Single Tree to Many Trees, that is meet the Jungle of ' + 'Methods', + 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'), ('Please, not the moons again! Voting and Bagging', 2, None, - '___sec32'), - ('Now Bagging', 2, None, '___sec33'), - ('Random forests', 2, None, '___sec34'), - ('A simple scikit-learn example', 2, None, '___sec35'), - ('Then random forests', 2, None, '___sec36'), - ('Feature Importance', 2, None, '___sec37'), - ('Boosting: AdaBoost', 2, None, '___sec38'), - ('Gradient Boosting', 2, None, '___sec39'), - ('Gradient Boots with Early Stopping', 2, None, '___sec40')]} + '___sec36'), + ('Now Bagging', 2, None, '___sec37'), + ('Random forests', 2, None, '___sec38'), + ('A simple scikit-learn example', 2, None, '___sec39'), + ('Then random forests', 2, None, '___sec40'), + ('Feature Importance', 2, None, '___sec41'), + ("Boosting, a Bird'e Eye", 2, None, '___sec42'), + ('Adaptive boosting: AdaBoost, Basic Algorithm', + 2, + None, + '___sec43'), + ('AdaBoost Examples', 2, None, '___sec44'), + ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'), + ('Gradient Boosting, Examples', 2, None, '___sec46'), + ('Gradient Boots with Early Stopping', 2, None, '___sec47'), + ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]} end of tocinfo --> @@ -151,31 +166,39 @@ MathJax.Hub.Config({
  • Classification tree, how to split nodes
  • Visualizing the Tree, Classification
  • Visualizing the Tree, The Moons
    • -
  • Computing the Gini index
    • -
  • Simple Python Code to read in Data and perform Classification
    • -
  • Computing the Gini Factor
    • -
  • Entropy and the ID3 algorithm
    • -
  • Implementing the ID3 Algorithm
    • -
  • Cancer Data again now with Decision Trees and other Methods
    • -
  • Another example, the moons again
    • -
  • Playing around with regions
    • -
  • Regression trees
    • -
  • Final regressor code
    • -
  • Pros and cons of trees, pros
    • -
  • Disadvantages
    • -
  • 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
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • -
  • Feature Importance
    • -
  • Boosting: AdaBoost
    • -
  • Gradient Boosting
    • -
  • Gradient Boots with Early Stopping
    • +
  • Algorithms for Setting up Decision Trees
    • +
  • The CART algorithm for Classification
    • +
  • The CART algorithm for Regression
    • +
  • Computing the Gini index
    • +
  • Simple Python Code to read in Data and perform Classification
    • +
  • Computing the Gini Factor
    • +
  • Entropy and the ID3 algorithm
    • +
  • Implementing the ID3 Algorithm
    • +
  • Cancer Data again now with Decision Trees and other Methods
    • +
  • Another example, the moons again
    • +
  • Playing around with regions
    • +
  • Regression trees
    • +
  • Final regressor code
    • +
  • Pros and cons of trees, pros
    • +
  • Disadvantages
    • +
  • From a Single Tree to Many Trees, that is 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
    • +
  • A simple scikit-learn example
    • +
  • Then random forests
    • +
  • Feature Importance
    • +
  • Boosting, a Bird'e Eye
    • +
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • +
  • AdaBoost Examples
    • +
  • Gradient boosting: Basic Algorithm
    • +
  • Gradient Boosting, Examples
    • +
  • Gradient Boots with Early Stopping
    • +
  • XGBoost: Extreme Gradient Boosting
    • @@ -191,198 +214,78 @@ MathJax.Hub.Config({ -

    Implementing the ID3 Algorithm

    +

    Simple Python Code to read in Data and perform Classification

    -

    import re
-import math
-from collections import deque
+
    # Common imports
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+from sklearn.tree import DecisionTreeClassifier
+from sklearn.model_selection import train_test_split
+from sklearn.tree import export_graphviz
+from sklearn.preprocessing import StandardScaler, OneHotEncoder
+from sklearn.compose import ColumnTransformer
+from IPython.display import Image 
+from pydot import graph_from_dot_data
+import os
 
-# x is examples in training set
-# y is set of targets
-# label is target attributes
-# Node is a class which has properties values, childs, and next
-# root is top node in the decision tree
+# Where to save the figures and data files
+PROJECT_ROOT_DIR = "Results"
+FIGURE_ID = "Results/FigureFiles"
+DATA_ID = "DataFiles/"
 
-class Node(object):
-	def __init__(self):
-		self.value = None
-		self.next = None
-		self.childs = None
+if not os.path.exists(PROJECT_ROOT_DIR):
+    os.mkdir(PROJECT_ROOT_DIR)
 
-# Simple class of Decision Tree
-# Aimed for who want to learn Decision Tree, so it is not optimized
-class DecisionTree(object):
-	def __init__(self, sample, attributes, labels):
-		self.sample = sample
-		self.attributes = attributes
-		self.labels = labels
-		self.labelCodes = None
-		self.labelCodesCount = None
-		self.initLabelCodes()
-		# print(self.labelCodes)
-		self.root = None
-		self.entropy = self.getEntropy([x for x in range(len(self.labels))])
+if not os.path.exists(FIGURE_ID):
+    os.makedirs(FIGURE_ID)
 
-	def initLabelCodes(self):
-		self.labelCodes = []
-		self.labelCodesCount = []
-		for l in self.labels:
-			if l not in self.labelCodes:
-				self.labelCodes.append(l)
-				self.labelCodesCount.append(0)
-			self.labelCodesCount[self.labelCodes.index(l)] += 1
+if not os.path.exists(DATA_ID):
+    os.makedirs(DATA_ID)
 
-	def getLabelCodeId(self, sampleId):
-		return self.labelCodes.index(self.labels[sampleId])
+def image_path(fig_id):
+    return os.path.join(FIGURE_ID, fig_id)
 
-	def getAttributeValues(self, sampleIds, attributeId):
-		vals = []
-		for sid in sampleIds:
-			val = self.sample[sid][attributeId]
-			if val not in vals:
-				vals.append(val)
-		# print(vals)
-		return vals
+def data_path(dat_id):
+    return os.path.join(DATA_ID, dat_id)
 
-	def getEntropy(self, sampleIds):
-		entropy = 0
-		labelCount = [0] * len(self.labelCodes)
-		for sid in sampleIds:
-			labelCount[self.getLabelCodeId(sid)] += 1
-		# print("-ge", labelCount)
-		for lv in labelCount:
-			# print(lv)
-			if lv != 0:
-				entropy += -lv/len(sampleIds) * math.log(lv/len(sampleIds), 2)
-			else:
-				entropy += 0
-		return entropy
+def save_fig(fig_id):
+    plt.savefig(image_path(fig_id) + ".png", format='png')
 
-	def getDominantLabel(self, sampleIds):
-		labelCodesCount = [0] * len(self.labelCodes)
-		for sid in sampleIds:
-			labelCodesCount[self.labelCodes.index(self.labels[sid])] += 1
-		return self.labelCodes[labelCodesCount.index(max(labelCodesCount))]
+infile = open(data_path("rideclass.csv"),'r')
 
-	def getInformationGain(self, sampleIds, attributeId):
-		gain = self.getEntropy(sampleIds)
-		attributeVals = []
-		attributeValsCount = []
-		attributeValsIds = []
-		for sid in sampleIds:
-			val = self.sample[sid][attributeId]
-			if val not in attributeVals:
-				attributeVals.append(val)
-				attributeValsCount.append(0)
-				attributeValsIds.append([])
-			vid = attributeVals.index(val)
-			attributeValsCount[vid] += 1
-			attributeValsIds[vid].append(sid)
-		# print("-gig", self.attributes[attributeId])
-		for vc, vids in zip(attributeValsCount, attributeValsIds):
-			# print("-gig", vids)
-			gain -= vc/len(sampleIds) * self.getEntropy(vids)
-		return gain
+# Read the experimental data with Pandas
+from IPython.display import display
+ridedata = pd.read_csv(infile,names = ('Outlook','Temperature','Humidity','Wind','Ride'))
+ridedata = pd.DataFrame(ridedata)
 
-	def getAttributeMaxInformationGain(self, sampleIds, attributeIds):
-		attributesEntropy = [0] * len(attributeIds)
-		for i, attId in zip(range(len(attributeIds)), attributeIds):
-			attributesEntropy[i] = self.getInformationGain(sampleIds, attId)
-		maxId = attributeIds[attributesEntropy.index(max(attributesEntropy))]
-		return self.attributes[maxId], maxId
+# Features and targets
+X = ridedata.loc[:, ridedata.columns != 'Ride'].values
+y = ridedata.loc[:, ridedata.columns == 'Ride'].values
 
-	def isSingleLabeled(self, sampleIds):
-		label = self.labels[sampleIds[0]]
-		for sid in sampleIds:
-			if self.labels[sid] != label:
-				return False
-		return True
-
-	def getLabel(self, sampleId):
-		return self.labels[sampleId]
-
-	def id3(self):
-		sampleIds = [x for x in range(len(self.sample))]
-		attributeIds = [x for x in range(len(self.attributes))]
-		self.root = self.id3Recv(sampleIds, attributeIds, self.root)
-
-	def id3Recv(self, sampleIds, attributeIds, root):
-		root = Node() # Initialize current root
-		if self.isSingleLabeled(sampleIds):
-			root.value = self.labels[sampleIds[0]]
-			return root
-		# print(attributeIds)
-		if len(attributeIds) == 0:
-			root.value = self.getDominantLabel(sampleIds)
-			return root
-		bestAttrName, bestAttrId = self.getAttributeMaxInformationGain(
-			sampleIds, attributeIds)
-		# print(bestAttrName)
-		root.value = bestAttrName
-		root.childs = []  # Create list of children
-		for value in self.getAttributeValues(sampleIds, bestAttrId):
-			# print(value)
-			child = Node()
-			child.value = value
-			root.childs.append(child)  # Append new child node to current
-									   # root
-			childSampleIds = []
-			for sid in sampleIds:
-				if self.sample[sid][bestAttrId] == value:
-					childSampleIds.append(sid)
-			if len(childSampleIds) == 0:
-				child.next = self.getDominantLabel(sampleIds)
-			else:
-				# print(bestAttrName, bestAttrId)
-				# print(attributeIds)
-				if len(attributeIds) > 0 and bestAttrId in attributeIds:
-					toRemove = attributeIds.index(bestAttrId)
-					attributeIds.pop(toRemove)
-				child.next = self.id3Recv(
-					childSampleIds, attributeIds, child.next)
-		return root
-
-	def printTree(self):
-		if self.root:
-			roots = deque()
-			roots.append(self.root)
-			while len(roots) > 0:
-				root = roots.popleft()
-				print(root.value)
-				if root.childs:
-					for child in root.childs:
-						print('({})'.format(child.value))
-						roots.append(child.next)
-				elif root.next:
-					print(root.next)
-
-
-def test():
-	f = open('DataFiles/rideclass.csv')
-	attributes = f.readline().split(',')
-	attributes = attributes[1:len(attributes)-1]
-	print(attributes)
-	sample = f.readlines()
-	f.close()
-	for i in range(len(sample)):
-		sample[i] = re.sub('\d+,', '', sample[i])
-		sample[i] = sample[i].strip().split(',')
-	labels = []
-	for s in sample:
-		labels.append(s.pop())
-	# print(sample)
-	# print(labels)
-	decisionTree = DecisionTree(sample, attributes, labels)
-	print("System entropy {}".format(decisionTree.entropy))
-	decisionTree.id3()
-	decisionTree.printTree()
-
-
-if __name__ == '__main__':
-	test()
+# Create the encoder.
+encoder = OneHotEncoder(handle_unknown="ignore")
+# Assume for simplicity all features are categorical.
+encoder.fit(X)    
+# Apply the encoder.
+X = encoder.transform(X)
+print(X)
+# Then do a Classification tree
+tree_clf = DecisionTreeClassifier(max_depth=2)
+tree_clf.fit(X, y)
+print("Train set accuracy with Decision Tree: {:.2f}".format(tree_clf.score(X,y)))
+#transfer to a decision tree graph
+export_graphviz(
+    tree_clf,
+    out_file="DataFiles/ride.dot",
+    rounded=True,
+    filled=True
+)
+cmd = 'dot -Tpng DataFiles/cancer.dot -o DataFiles/cancer.png'
+os.system(cmd)
 
    
 
    
 
    
@@ -410,7 +313,7 @@ MathJax.Hub.Config({
   
  • 30
    • 
   
  • 31
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html
index cab8de1df..3aa96b7f6 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,51 +214,80 @@ MathJax.Hub.Config({
 
 
 
-
    Cancer Data again now with Decision Trees and other Methods 
    
+
    Computing the Gini Factor  
    
+
+
    
+The above functions (gini, entropy and misclassification error) are
+important components of the so-called CART algorithm. We will discuss
+this algorithm below after we have discussed the information gain
+algorithm ID3.
+
+
    
+In the example here we have converted all our attributes into numerical values \( 0,1,2 \) etc.
+
 
    
 
 
-
    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
+
    # Split a dataset based on an attribute and an attribute value
+def test_split(index, value, dataset):
+	left, right = list(), list()
+	for row in dataset:
+		if row[index] < value:
+			left.append(row)
+		else:
+			right.append(row)
+	return left, right
+ 
+# Calculate the Gini index for a split dataset
+def gini_index(groups, classes):
+	# count all samples at split point
+	n_instances = float(sum([len(group) for group in groups]))
+	# sum weighted Gini index for each group
+	gini = 0.0
+	for group in groups:
+		size = float(len(group))
+		# avoid divide by zero
+		if size == 0:
+			continue
+		score = 0.0
+		# score the group based on the score for each class
+		for class_val in classes:
+			p = [row[-1] for row in group].count(class_val) / size
+			score += p * p
+		# weight the group score by its relative size
+		gini += (1.0 - score) * (size / n_instances)
+	return gini
 
-# Load the data
-cancer = load_breast_cancer()
+# Select the best split point for a dataset
+def get_split(dataset):
+	class_values = list(set(row[-1] for row in dataset))
+	b_index, b_value, b_score, b_groups = 999, 999, 999, None
+	for index in range(len(dataset[0])-1):
+		for row in dataset:
+			groups = test_split(index, row[index], dataset)
+			gini = gini_index(groups, class_values)
+			print('X%d < %.3f Gini=%.3f' % ((index+1), row[index], gini))
+			if gini < b_score:
+				b_index, b_value, b_score, b_groups = index, row[index], gini, groups
+	return {'index':b_index, 'value':b_value, 'groups':b_groups}
+ 
+dataset = [[0,0,0,0,0],
+            [0,0,0,1,1],
+            [1,0,0,0,1],
+            [2,1,0,0,1],
+            [2,2,1,0,1],
+            [2,2,1,1,0],
+            [1,2,1,1,1],
+            [0,1,0,0,0],
+            [0,2,1,0,1],
+            [2,1,1,0,1],
+            [0,1,1,1,1],
+            [1,1,0,1,1],
+            [1,0,1,0,1],
+            [2,1,0,1,0]]
 
-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)))
+split = get_split(dataset)
+print('Split: [X%d < %.3f]' % ((split['index']+1), split['value']))
 
    
 
    
 
    
@@ -263,7 +315,7 @@ deep_tree_clf.fit(X_train_scaled, y_train)
   
  • 31
    • 
   
  • 32
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html
index 9ba0bc5d6..81650dfee 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,75 +214,39 @@ MathJax.Hub.Config({
 
 
 
-
    Another example, the moons again 
    
+
    Entropy and the ID3 algorithm 
    
+
 
    
+ID3, learns decision trees by constructing
+them topdown, beginning with the question which attribute should be tested at the root of the tree?
 
-
-
    from __future__ import division, print_function, unicode_literals
+
      
+
    1.  Each instance attribute is evaluated using a statistical test to determine how well it alone classifies the training examples.
      2. 
+
    2.  The best attribute is selected and used as the test at the root node of the tree.
      3. 
+
    3.  A descendant of the root node is then created for each possible value of this attribute.
      4. 
+
    4.  Training examples are sorted to the appropriate descendant node.
      5. 
+
    5.  The entire process is then repeated using the training examples associated with each descendant node to select the best attribute to test at that point in the tree.
      6. 
+
    6.  This forms a greedy search for an acceptable decision tree, in which the algorithm never backtracks to reconsider earlier choices.
      7.  
+
    
 
-# Common imports
-import numpy as np
-import os
+The ID3 algorithm selects, which attribute to test at each node in the
+tree.
 
-# to make this notebook's output stable across runs
-np.random.seed(42)
+
    
+We would like to select the attribute that is most useful for classifying
+examples.
 
-# To plot pretty figures
-import matplotlib
-import matplotlib.pyplot as plt
-from matplotlib.colors import ListedColormap
-plt.rcParams['axes.labelsize'] = 14
-plt.rcParams['xtick.labelsize'] = 12
-plt.rcParams['ytick.labelsize'] = 12
+
    
+What is a good quantitative measure of the worth of an attribute?
 
+
    
+Information gain measures how well a given attribute separates the
+training examples according to their target classification.
 
-from sklearn.svm import SVC
-from sklearn import datasets
-from sklearn.tree import DecisionTreeClassifier
-from sklearn.datasets import make_moons
-from sklearn.tree import export_graphviz
+
    
+The ID3 algorithm uses this information gain measure to select among the candidate
+attributes at each step while growing the tree.
 
-Xm, ym = make_moons(n_samples=100, noise=0.25, random_state=53)
-
-deep_tree_clf1 = DecisionTreeClassifier(random_state=42)
-deep_tree_clf2 = DecisionTreeClassifier(min_samples_leaf=4, random_state=42)
-deep_tree_clf1.fit(Xm, ym)
-deep_tree_clf2.fit(Xm, ym)
-
-
-def plot_decision_boundary(clf, X, y, axes=[0, 7.5, 0, 3], iris=True, legend=False, plot_training=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 not iris:
-        custom_cmap2 = ListedColormap(['#7d7d58','#4c4c7f','#507d50'])
-        plt.contour(x1, x2, y_pred, cmap=custom_cmap2, alpha=0.8)
-    if plot_training:
-        plt.plot(X[:, 0][y==0], X[:, 1][y==0], "yo", label="Iris-Setosa")
-        plt.plot(X[:, 0][y==1], X[:, 1][y==1], "bs", label="Iris-Versicolor")
-        plt.plot(X[:, 0][y==2], X[:, 1][y==2], "g^", label="Iris-Virginica")
-        plt.axis(axes)
-    if iris:
-        plt.xlabel("Petal length", fontsize=14)
-        plt.ylabel("Petal width", fontsize=14)
-    else:
-        plt.xlabel(r"$x_1$", fontsize=18)
-        plt.ylabel(r"$x_2$", fontsize=18, rotation=0)
-    if legend:
-        plt.legend(loc="lower right", fontsize=14)
-plt.figure(figsize=(11, 4))
-plt.subplot(121)
-plot_decision_boundary(deep_tree_clf1, Xm, ym, axes=[-1.5, 2.5, -1, 1.5], iris=False)
-plt.title("No restrictions", fontsize=16)
-plt.subplot(122)
-plot_decision_boundary(deep_tree_clf2, Xm, ym, axes=[-1.5, 2.5, -1, 1.5], iris=False)
-plt.title("min_samples_leaf = {}".format(deep_tree_clf2.min_samples_leaf), fontsize=14)
-plt.show()
-
    
 
    
 
    
 
@@ -286,7 +273,7 @@ plt.show()
   
  • 32
    • 
   
  • 33
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html
index 1cdf2975f..b2a17c4b1 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,30 +214,198 @@ MathJax.Hub.Config({
 
 
 
-
    Playing around with regions 
    
+
    Implementing the ID3 Algorithm 
    
+
 
    
 
 
-
    np.random.seed(6)
-Xs = np.random.rand(100, 2) - 0.5
-ys = (Xs[:, 0] > 0).astype(np.float32) * 2
+
    import re
+import math
+from collections import deque
 
-angle = np.pi/4
-rotation_matrix = np.array([[np.cos(angle), -np.sin(angle)], [np.sin(angle), np.cos(angle)]])
-Xsr = Xs.dot(rotation_matrix)
+# x is examples in training set
+# y is set of targets
+# label is target attributes
+# Node is a class which has properties values, childs, and next
+# root is top node in the decision tree
 
-tree_clf_s = DecisionTreeClassifier(random_state=42)
-tree_clf_s.fit(Xs, ys)
-tree_clf_sr = DecisionTreeClassifier(random_state=42)
-tree_clf_sr.fit(Xsr, ys)
+class Node(object):
+	def __init__(self):
+		self.value = None
+		self.next = None
+		self.childs = None
 
-plt.figure(figsize=(11, 4))
-plt.subplot(121)
-plot_decision_boundary(tree_clf_s, Xs, ys, axes=[-0.7, 0.7, -0.7, 0.7], iris=False)
-plt.subplot(122)
-plot_decision_boundary(tree_clf_sr, Xsr, ys, axes=[-0.7, 0.7, -0.7, 0.7], iris=False)
+# Simple class of Decision Tree
+# Aimed for who want to learn Decision Tree, so it is not optimized
+class DecisionTree(object):
+	def __init__(self, sample, attributes, labels):
+		self.sample = sample
+		self.attributes = attributes
+		self.labels = labels
+		self.labelCodes = None
+		self.labelCodesCount = None
+		self.initLabelCodes()
+		# print(self.labelCodes)
+		self.root = None
+		self.entropy = self.getEntropy([x for x in range(len(self.labels))])
 
-plt.show()
+	def initLabelCodes(self):
+		self.labelCodes = []
+		self.labelCodesCount = []
+		for l in self.labels:
+			if l not in self.labelCodes:
+				self.labelCodes.append(l)
+				self.labelCodesCount.append(0)
+			self.labelCodesCount[self.labelCodes.index(l)] += 1
+
+	def getLabelCodeId(self, sampleId):
+		return self.labelCodes.index(self.labels[sampleId])
+
+	def getAttributeValues(self, sampleIds, attributeId):
+		vals = []
+		for sid in sampleIds:
+			val = self.sample[sid][attributeId]
+			if val not in vals:
+				vals.append(val)
+		# print(vals)
+		return vals
+
+	def getEntropy(self, sampleIds):
+		entropy = 0
+		labelCount = [0] * len(self.labelCodes)
+		for sid in sampleIds:
+			labelCount[self.getLabelCodeId(sid)] += 1
+		# print("-ge", labelCount)
+		for lv in labelCount:
+			# print(lv)
+			if lv != 0:
+				entropy += -lv/len(sampleIds) * math.log(lv/len(sampleIds), 2)
+			else:
+				entropy += 0
+		return entropy
+
+	def getDominantLabel(self, sampleIds):
+		labelCodesCount = [0] * len(self.labelCodes)
+		for sid in sampleIds:
+			labelCodesCount[self.labelCodes.index(self.labels[sid])] += 1
+		return self.labelCodes[labelCodesCount.index(max(labelCodesCount))]
+
+	def getInformationGain(self, sampleIds, attributeId):
+		gain = self.getEntropy(sampleIds)
+		attributeVals = []
+		attributeValsCount = []
+		attributeValsIds = []
+		for sid in sampleIds:
+			val = self.sample[sid][attributeId]
+			if val not in attributeVals:
+				attributeVals.append(val)
+				attributeValsCount.append(0)
+				attributeValsIds.append([])
+			vid = attributeVals.index(val)
+			attributeValsCount[vid] += 1
+			attributeValsIds[vid].append(sid)
+		# print("-gig", self.attributes[attributeId])
+		for vc, vids in zip(attributeValsCount, attributeValsIds):
+			# print("-gig", vids)
+			gain -= vc/len(sampleIds) * self.getEntropy(vids)
+		return gain
+
+	def getAttributeMaxInformationGain(self, sampleIds, attributeIds):
+		attributesEntropy = [0] * len(attributeIds)
+		for i, attId in zip(range(len(attributeIds)), attributeIds):
+			attributesEntropy[i] = self.getInformationGain(sampleIds, attId)
+		maxId = attributeIds[attributesEntropy.index(max(attributesEntropy))]
+		return self.attributes[maxId], maxId
+
+	def isSingleLabeled(self, sampleIds):
+		label = self.labels[sampleIds[0]]
+		for sid in sampleIds:
+			if self.labels[sid] != label:
+				return False
+		return True
+
+	def getLabel(self, sampleId):
+		return self.labels[sampleId]
+
+	def id3(self):
+		sampleIds = [x for x in range(len(self.sample))]
+		attributeIds = [x for x in range(len(self.attributes))]
+		self.root = self.id3Recv(sampleIds, attributeIds, self.root)
+
+	def id3Recv(self, sampleIds, attributeIds, root):
+		root = Node() # Initialize current root
+		if self.isSingleLabeled(sampleIds):
+			root.value = self.labels[sampleIds[0]]
+			return root
+		# print(attributeIds)
+		if len(attributeIds) == 0:
+			root.value = self.getDominantLabel(sampleIds)
+			return root
+		bestAttrName, bestAttrId = self.getAttributeMaxInformationGain(
+			sampleIds, attributeIds)
+		# print(bestAttrName)
+		root.value = bestAttrName
+		root.childs = []  # Create list of children
+		for value in self.getAttributeValues(sampleIds, bestAttrId):
+			# print(value)
+			child = Node()
+			child.value = value
+			root.childs.append(child)  # Append new child node to current
+									   # root
+			childSampleIds = []
+			for sid in sampleIds:
+				if self.sample[sid][bestAttrId] == value:
+					childSampleIds.append(sid)
+			if len(childSampleIds) == 0:
+				child.next = self.getDominantLabel(sampleIds)
+			else:
+				# print(bestAttrName, bestAttrId)
+				# print(attributeIds)
+				if len(attributeIds) > 0 and bestAttrId in attributeIds:
+					toRemove = attributeIds.index(bestAttrId)
+					attributeIds.pop(toRemove)
+				child.next = self.id3Recv(
+					childSampleIds, attributeIds, child.next)
+		return root
+
+	def printTree(self):
+		if self.root:
+			roots = deque()
+			roots.append(self.root)
+			while len(roots) > 0:
+				root = roots.popleft()
+				print(root.value)
+				if root.childs:
+					for child in root.childs:
+						print('({})'.format(child.value))
+						roots.append(child.next)
+				elif root.next:
+					print(root.next)
+
+
+def test():
+	f = open('DataFiles/rideclass.csv')
+	attributes = f.readline().split(',')
+	attributes = attributes[1:len(attributes)-1]
+	print(attributes)
+	sample = f.readlines()
+	f.close()
+	for i in range(len(sample)):
+		sample[i] = re.sub('\d+,', '', sample[i])
+		sample[i] = sample[i].strip().split(',')
+	labels = []
+	for s in sample:
+		labels.append(s.pop())
+	# print(sample)
+	# print(labels)
+	decisionTree = DecisionTree(sample, attributes, labels)
+	print("System entropy {}".format(decisionTree.entropy))
+	decisionTree.id3()
+	decisionTree.printTree()
+
+
+if __name__ == '__main__':
+	test()
 
    
 
    
 
    
@@ -242,7 +433,7 @@ plt.show()
   
  • 33
    • 
   
  • 34
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html
index 6c0dce319..b24d3a1a3 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,24 +214,51 @@ MathJax.Hub.Config({
 
 
 
-
    Regression trees 
    
+
    Cancer Data again now with Decision Trees and other Methods 
    
 
    
 
 
-
    # Quadratic training set + noise
-np.random.seed(42)
-m = 200
-X = np.random.rand(m, 1)
-y = 4 * (X - 0.5) ** 2
-y = y + np.random.randn(m, 1) / 10
-
    
-
    
+
    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
 
-
-
    from sklearn.tree import DecisionTreeRegressor
+# Load the data
+cancer = load_breast_cancer()
 
-tree_reg = DecisionTreeRegressor(max_depth=2, random_state=42)
-tree_reg.fit(X, y)
+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)))
 
    
 
    
 
    
@@ -236,7 +286,7 @@ tree_reg.fit(X, y)
   
  • 34
    • 
   
  • 35
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html
index 0f91c9af8..0e1700f16 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,79 +214,73 @@ MathJax.Hub.Config({
 
 
 
-
    Final regressor code 
    
+
    Another example, the moons again 
    
 
    
 
 
-
    from sklearn.tree import DecisionTreeRegressor
+
    from __future__ import division, print_function, unicode_literals
 
-tree_reg1 = DecisionTreeRegressor(random_state=42, max_depth=2)
-tree_reg2 = DecisionTreeRegressor(random_state=42, max_depth=3)
-tree_reg1.fit(X, y)
-tree_reg2.fit(X, y)
+# Common imports
+import numpy as np
+import os
 
-def plot_regression_predictions(tree_reg, X, y, axes=[0, 1, -0.2, 1], ylabel="$y$"):
-    x1 = np.linspace(axes[0], axes[1], 500).reshape(-1, 1)
-    y_pred = tree_reg.predict(x1)
-    plt.axis(axes)
-    plt.xlabel("$x_1$", fontsize=18)
-    if ylabel:
-        plt.ylabel(ylabel, fontsize=18, rotation=0)
-    plt.plot(X, y, "b.")
-    plt.plot(x1, y_pred, "r.-", linewidth=2, label=r"$\hat{y}$")
+# to make this notebook's output stable across runs
+np.random.seed(42)
 
+# To plot pretty figures
+import matplotlib
+import matplotlib.pyplot as plt
+from matplotlib.colors import ListedColormap
+plt.rcParams['axes.labelsize'] = 14
+plt.rcParams['xtick.labelsize'] = 12
+plt.rcParams['ytick.labelsize'] = 12
+
+
+from sklearn.svm import SVC
+from sklearn import datasets
+from sklearn.tree import DecisionTreeClassifier
+from sklearn.datasets import make_moons
+from sklearn.tree import export_graphviz
+
+Xm, ym = make_moons(n_samples=100, noise=0.25, random_state=53)
+
+deep_tree_clf1 = DecisionTreeClassifier(random_state=42)
+deep_tree_clf2 = DecisionTreeClassifier(min_samples_leaf=4, random_state=42)
+deep_tree_clf1.fit(Xm, ym)
+deep_tree_clf2.fit(Xm, ym)
+
+
+def plot_decision_boundary(clf, X, y, axes=[0, 7.5, 0, 3], iris=True, legend=False, plot_training=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 not iris:
+        custom_cmap2 = ListedColormap(['#7d7d58','#4c4c7f','#507d50'])
+        plt.contour(x1, x2, y_pred, cmap=custom_cmap2, alpha=0.8)
+    if plot_training:
+        plt.plot(X[:, 0][y==0], X[:, 1][y==0], "yo", label="Iris-Setosa")
+        plt.plot(X[:, 0][y==1], X[:, 1][y==1], "bs", label="Iris-Versicolor")
+        plt.plot(X[:, 0][y==2], X[:, 1][y==2], "g^", label="Iris-Virginica")
+        plt.axis(axes)
+    if iris:
+        plt.xlabel("Petal length", fontsize=14)
+        plt.ylabel("Petal width", fontsize=14)
+    else:
+        plt.xlabel(r"$x_1$", fontsize=18)
+        plt.ylabel(r"$x_2$", fontsize=18, rotation=0)
+    if legend:
+        plt.legend(loc="lower right", fontsize=14)
 plt.figure(figsize=(11, 4))
 plt.subplot(121)
-plot_regression_predictions(tree_reg1, X, y)
-for split, style in ((0.1973, "k-"), (0.0917, "k--"), (0.7718, "k--")):
-    plt.plot([split, split], [-0.2, 1], style, linewidth=2)
-plt.text(0.21, 0.65, "Depth=0", fontsize=15)
-plt.text(0.01, 0.2, "Depth=1", fontsize=13)
-plt.text(0.65, 0.8, "Depth=1", fontsize=13)
-plt.legend(loc="upper center", fontsize=18)
-plt.title("max_depth=2", fontsize=14)
-
+plot_decision_boundary(deep_tree_clf1, Xm, ym, axes=[-1.5, 2.5, -1, 1.5], iris=False)
+plt.title("No restrictions", fontsize=16)
 plt.subplot(122)
-plot_regression_predictions(tree_reg2, X, y, ylabel=None)
-for split, style in ((0.1973, "k-"), (0.0917, "k--"), (0.7718, "k--")):
-    plt.plot([split, split], [-0.2, 1], style, linewidth=2)
-for split in (0.0458, 0.1298, 0.2873, 0.9040):
-    plt.plot([split, split], [-0.2, 1], "k:", linewidth=1)
-plt.text(0.3, 0.5, "Depth=2", fontsize=13)
-plt.title("max_depth=3", fontsize=14)
-
-plt.show()
-
    
-
    
-
-
-
    tree_reg1 = DecisionTreeRegressor(random_state=42)
-tree_reg2 = DecisionTreeRegressor(random_state=42, min_samples_leaf=10)
-tree_reg1.fit(X, y)
-tree_reg2.fit(X, y)
-
-x1 = np.linspace(0, 1, 500).reshape(-1, 1)
-y_pred1 = tree_reg1.predict(x1)
-y_pred2 = tree_reg2.predict(x1)
-
-plt.figure(figsize=(11, 4))
-
-plt.subplot(121)
-plt.plot(X, y, "b.")
-plt.plot(x1, y_pred1, "r.-", linewidth=2, label=r"$\hat{y}$")
-plt.axis([0, 1, -0.2, 1.1])
-plt.xlabel("$x_1$", fontsize=18)
-plt.ylabel("$y$", fontsize=18, rotation=0)
-plt.legend(loc="upper center", fontsize=18)
-plt.title("No restrictions", fontsize=14)
-
-plt.subplot(122)
-plt.plot(X, y, "b.")
-plt.plot(x1, y_pred2, "r.-", linewidth=2, label=r"$\hat{y}$")
-plt.axis([0, 1, -0.2, 1.1])
-plt.xlabel("$x_1$", fontsize=18)
-plt.title("min_samples_leaf={}".format(tree_reg2.min_samples_leaf), fontsize=14)
-
+plot_decision_boundary(deep_tree_clf2, Xm, ym, axes=[-1.5, 2.5, -1, 1.5], iris=False)
+plt.title("min_samples_leaf = {}".format(deep_tree_clf2.min_samples_leaf), fontsize=14)
 plt.show()
 
    
 
    
@@ -292,7 +309,7 @@ plt.show()
   
  • 35
    • 
   
  • 36
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html
index deb24ecb1..1a5e76321 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,18 +214,32 @@ MathJax.Hub.Config({
 
 
 
-
    Pros and cons of trees, pros 
    
+
    Playing around with regions 
    
+
    
 
-
      
-
    •  White box, easy to interpret model. Some people believe that decision trees more closely mirror human decision-making than do the regression and classification approaches discussed earlier (think of support vector machines)
      • 
-
    •  Trees are very easy to explain to people. In fact, they are even easier to explain than linear regression!
      • 
-
    •  No feature normalization needed
      • 
-
    •  Tree models can handle both continuous and categorical data (Classification and Regression Trees)
      • 
-
    •  Can model nonlinear relationships
      • 
-
    •  Can model interactions between the different descriptive features
      • 
-
    •  Trees can be displayed graphically, and are easily interpreted even by a non-expert (especially if they are small)
      • 
-
    
+
+
    np.random.seed(6)
+Xs = np.random.rand(100, 2) - 0.5
+ys = (Xs[:, 0] > 0).astype(np.float32) * 2
 
+angle = np.pi/4
+rotation_matrix = np.array([[np.cos(angle), -np.sin(angle)], [np.sin(angle), np.cos(angle)]])
+Xsr = Xs.dot(rotation_matrix)
+
+tree_clf_s = DecisionTreeClassifier(random_state=42)
+tree_clf_s.fit(Xs, ys)
+tree_clf_sr = DecisionTreeClassifier(random_state=42)
+tree_clf_sr.fit(Xsr, ys)
+
+plt.figure(figsize=(11, 4))
+plt.subplot(121)
+plot_decision_boundary(tree_clf_s, Xs, ys, axes=[-0.7, 0.7, -0.7, 0.7], iris=False)
+plt.subplot(122)
+plot_decision_boundary(tree_clf_sr, Xsr, ys, axes=[-0.7, 0.7, -0.7, 0.7], iris=False)
+
+plt.show()
+
    
+
    
 
    
 
 
      
@@ -228,7 +265,7 @@ MathJax.Hub.Config({
   
    • 36
      • 
   
    • 37
      • 
   
    • ...
      • 
-  
    • 42
      • 
+  
    • 50
      • 
   
    • »
      • 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html
index 830923076..767d35e1c 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,20 +214,25 @@ MathJax.Hub.Config({
 
 
 
-
    Disadvantages 
    
+
    Regression trees 
    
+
    
 
-
      
-
    •  Unfortunately, trees generally do not have the same level of predictive accuracy as some of the other regression and classification approaches
      • 
-
    •  If continuous features are used the tree may become quite large and hence less interpretable
      • 
-
    •  Decision trees are prone to overfit the training data and hence do not well generalize the data if no stopping criteria or improvements like pruning, boosting or bagging are implemented
      • 
-
    •  Small changes in the data may lead to a completely different tree. This issue can be addressed by using ensemble methods like bagging, boosting or random forests
      • 
-
    •  Unbalanced datasets where some target feature values occur much more frequently than others may lead to biased trees since the frequently occurring feature values are preferred over the less frequently occurring ones.
      •  
-
    •  If the number of features is relatively large (high dimensional) and the number of instances is relatively low, the tree might overfit the data
      • 
-
    •  Features with many levels may be preferred over features with less levels since for them it is more easy to split the dataset such that the sub datasets only contain pure target feature values. This issue can be addressed by preferring for instance the information gain ratio as splitting criteria over information gain
      • 
-
    
+
+
    # Quadratic training set + noise
+np.random.seed(42)
+m = 200
+X = np.random.rand(m, 1)
+y = 4 * (X - 0.5) ** 2
+y = y + np.random.randn(m, 1) / 10
+
    
+
    
 
-However, by aggregating many decision trees, using methods like bagging, random forests, and boosting, the predictive performance of trees can be substantially improved.
+
+
    from sklearn.tree import DecisionTreeRegressor
 
+tree_reg = DecisionTreeRegressor(max_depth=2, random_state=42)
+tree_reg.fit(X, y)
+
    
 
    
 
    
 
@@ -231,7 +259,7 @@ However, by aggregating many decision trees, using methods like bagging, random
   
  • 37
    • 
   
  • 38
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html
index ccafcc714..294420e7b 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,22 +214,81 @@ MathJax.Hub.Config({
 
 
 
-
    Bagging 
    
-
+
    Final regressor code 
    
 
    
-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.
 
+
+
    from sklearn.tree import DecisionTreeRegressor
+
+tree_reg1 = DecisionTreeRegressor(random_state=42, max_depth=2)
+tree_reg2 = DecisionTreeRegressor(random_state=42, max_depth=3)
+tree_reg1.fit(X, y)
+tree_reg2.fit(X, y)
+
+def plot_regression_predictions(tree_reg, X, y, axes=[0, 1, -0.2, 1], ylabel="$y$"):
+    x1 = np.linspace(axes[0], axes[1], 500).reshape(-1, 1)
+    y_pred = tree_reg.predict(x1)
+    plt.axis(axes)
+    plt.xlabel("$x_1$", fontsize=18)
+    if ylabel:
+        plt.ylabel(ylabel, fontsize=18, rotation=0)
+    plt.plot(X, y, "b.")
+    plt.plot(x1, y_pred, "r.-", linewidth=2, label=r"$\hat{y}$")
+
+plt.figure(figsize=(11, 4))
+plt.subplot(121)
+plot_regression_predictions(tree_reg1, X, y)
+for split, style in ((0.1973, "k-"), (0.0917, "k--"), (0.7718, "k--")):
+    plt.plot([split, split], [-0.2, 1], style, linewidth=2)
+plt.text(0.21, 0.65, "Depth=0", fontsize=15)
+plt.text(0.01, 0.2, "Depth=1", fontsize=13)
+plt.text(0.65, 0.8, "Depth=1", fontsize=13)
+plt.legend(loc="upper center", fontsize=18)
+plt.title("max_depth=2", fontsize=14)
+
+plt.subplot(122)
+plot_regression_predictions(tree_reg2, X, y, ylabel=None)
+for split, style in ((0.1973, "k-"), (0.0917, "k--"), (0.7718, "k--")):
+    plt.plot([split, split], [-0.2, 1], style, linewidth=2)
+for split in (0.0458, 0.1298, 0.2873, 0.9040):
+    plt.plot([split, split], [-0.2, 1], "k:", linewidth=1)
+plt.text(0.3, 0.5, "Depth=2", fontsize=13)
+plt.title("max_depth=3", fontsize=14)
+
+plt.show()
+
    
 
    
-Bootstrap aggregation, or just bagging, is a
-general-purpose procedure for reducing the variance of a statistical
-learning method.
 
+
+
    tree_reg1 = DecisionTreeRegressor(random_state=42)
+tree_reg2 = DecisionTreeRegressor(random_state=42, min_samples_leaf=10)
+tree_reg1.fit(X, y)
+tree_reg2.fit(X, y)
+
+x1 = np.linspace(0, 1, 500).reshape(-1, 1)
+y_pred1 = tree_reg1.predict(x1)
+y_pred2 = tree_reg2.predict(x1)
+
+plt.figure(figsize=(11, 4))
+
+plt.subplot(121)
+plt.plot(X, y, "b.")
+plt.plot(x1, y_pred1, "r.-", linewidth=2, label=r"$\hat{y}$")
+plt.axis([0, 1, -0.2, 1.1])
+plt.xlabel("$x_1$", fontsize=18)
+plt.ylabel("$y$", fontsize=18, rotation=0)
+plt.legend(loc="upper center", fontsize=18)
+plt.title("No restrictions", fontsize=14)
+
+plt.subplot(122)
+plt.plot(X, y, "b.")
+plt.plot(x1, y_pred2, "r.-", linewidth=2, label=r"$\hat{y}$")
+plt.axis([0, 1, -0.2, 1.1])
+plt.xlabel("$x_1$", fontsize=18)
+plt.title("min_samples_leaf={}".format(tree_reg2.min_samples_leaf), fontsize=14)
+
+plt.show()
+
    
 
    
 
    
 
@@ -233,7 +315,7 @@ learning method.
   
  • 38
    • 
   
  • 39
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html
index 19bc01de5..fd04d1e03 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,33 +214,18 @@ MathJax.Hub.Config({
 
 
 
-
    More bagging 
    
+
    Pros and cons of trees, pros 
    
 
-
    
-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.
+
      
+
    •  White box, easy to interpret model. Some people believe that decision trees more closely mirror human decision-making than do the regression and classification approaches discussed earlier (think of support vector machines)
      • 
+
    •  Trees are very easy to explain to people. In fact, they are even easier to explain than linear regression!
      • 
+
    •  No feature normalization needed
      • 
+
    •  Tree models can handle both continuous and categorical data (Classification and Regression Trees)
      • 
+
    •  Can model nonlinear relationships
      • 
+
    •  Can model interactions between the different descriptive features
      • 
+
    •  Trees can be displayed graphically, and are easily interpreted even by a non-expert (especially if they are small)
      • 
+
    
 
-
    
-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.
-
-
    
 
    
 
 
      
@@ -243,7 +251,7 @@ predictor, averaged over all \( B \) trees.
   
    • 39
      • 
   
    • 40
      • 
   
    • ...
      • 
-  
    • 42
      • 
+  
    • 50
      • 
   
    • »
      • 
 
    
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html
index 044ef068d..21e98113b 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,23 +214,20 @@ MathJax.Hub.Config({
 
 
 
-
    Simple Voting Example, head or tail 
    
-
    
+
    Disadvantages 
    
+
+
      
+
    •  Unfortunately, trees generally do not have the same level of predictive accuracy as some of the other regression and classification approaches
      • 
+
    •  If continuous features are used the tree may become quite large and hence less interpretable
      • 
+
    •  Decision trees are prone to overfit the training data and hence do not well generalize the data if no stopping criteria or improvements like pruning, boosting or bagging are implemented
      • 
+
    •  Small changes in the data may lead to a completely different tree. This issue can be addressed by using ensemble methods like bagging, boosting or random forests
      • 
+
    •  Unbalanced datasets where some target feature values occur much more frequently than others may lead to biased trees since the frequently occurring feature values are preferred over the less frequently occurring ones.
      •  
+
    •  If the number of features is relatively large (high dimensional) and the number of instances is relatively low, the tree might overfit the data
      • 
+
    •  Features with many levels may be preferred over features with less levels since for them it is more easy to split the dataset such that the sub datasets only contain pure target feature values. This issue can be addressed by preferring for instance the information gain ratio as splitting criteria over information gain
      • 
+
    
+
+However, by aggregating many decision trees, using methods like bagging, random forests, and boosting, the predictive performance of trees can be substantially improved.
 
-
-
    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()
-
    
 
    
 
    
 
@@ -234,7 +254,7 @@ plt.show()
   
  • 40
    • 
   
  • 41
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html
index d0c8d30ae..bd6c12724 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,54 +214,27 @@ MathJax.Hub.Config({
 
 
 
-
    Using the Voting Classifier 
    
+
    From a Single Tree to Many Trees, that is meet the Jungle of Methods 
    
+
 
    
+As stated above and seen in many of the examples discussed here about
+a single decision tree, we often end up overfitting our training
+data. This normally means that we have a high variance. Can we reduce
+the variance of a statistical learning method?
 
-
-
    from sklearn.model_selection import train_test_split
-from sklearn.datasets import make_moons
+
    
+This leads us to a set of different methods that can combine different
+machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are
 
-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)
+
      
+
    1.  Votign classifiers
      2. 
+
    2.  Bagging and Pasting
      3. 
+
    3.  Random forests
      4. 
+
    4.  Boosting methods
      5. 
+
    
 
-from sklearn.ensemble import RandomForestClassifier
-from sklearn.ensemble import VotingClassifier
-from sklearn.linear_model import LogisticRegression
-from sklearn.svm import SVC
+We discuss these methods here.
 
-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))
-
    
 
    
 
    
 
@@ -264,6 +260,8 @@ voting_clf.fit(X_train, y_train)
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html
index 11262aea5..f614c844d 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,61 +214,22 @@ MathJax.Hub.Config({
 
 
 
-
    Please, not the moons again! Voting and Bagging 
    
+
    Bagging 
    
+
 
    
+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.
 
-
-
    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(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)
-
    
 
    
+Bootstrap aggregation, or just bagging, is a
+general-purpose procedure for reducing the variance of a statistical
+learning method.
 
-
-
    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)
-
-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))
-
    
 
    
 
    
 
@@ -270,6 +254,9 @@ voting_clf.fit(X_train, y_train)
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html
index 91e4880c1..7ce207a31 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,64 +214,32 @@ MathJax.Hub.Config({
 
 
 
-
    Now Bagging  
    
+
    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.
 
-
-
    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)
-
    
 
    
+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.
 
-
-
    from sklearn.metrics import accuracy_score
-print(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))
-
    
-
    
-
-
-
    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()
-
    
 
    
 
    
 
@@ -272,6 +263,10 @@ plt.show()
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • 44
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html
index 4b7896efb..8bccb0b35 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,47 +214,23 @@ MathJax.Hub.Config({
 
 
 
-
    Random forests 
    
-
+
    Simple Voting Example, head or tail 
    
 
    
-Random forests provide an improvement over bagged trees by way of a
-small tweak that decorrelates the trees.
-
-
    
-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.
-
-
    
-A fresh sample of \( m \) predictors is
-taken at each split, and typically we choose 
-
-$$
-m\approx \sqrt{p}.
-$$
-
-
    
-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.
-
-
    
-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.
 
+
+
    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()
+
    
 
    
 
    
 
@@ -254,6 +253,11 @@ this setting.
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • 44
    • 
+  
  • 45
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html
index af874b3ba..187f9c1c7 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,20 +214,53 @@ MathJax.Hub.Config({
 
 
 
-
    A simple scikit-learn example 
    
+
    Using the Voting Classifier 
    
 
    
 
 
-
    from sklearn.ensemble import RandomForestClassifier
-from sklearn.preprocessing import LabelEncoder
-from sklearn.model_selection import cross_validate
-# Data set not specificied
-X = dataset.XXX
-Y = dataset.YYY
-#Instantiate the model with 100 trees and entropy as splitting criteria
-Random_Forest_model = RandomForestClassifier(n_estimators=100,criterion="entropy")
-#Cross validation
-accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']
+
    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))
 
    
 
    
 
    
@@ -227,6 +283,12 @@ accuracy = cross_validate(Random_Forest_mode
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • 44
    • 
+  
  • 45
    • 
+  
  • 46
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
index 462ed91d1..3e4b68888 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,24 +214,60 @@ MathJax.Hub.Config({
 
 
 
-
    Then random forests 
    
+
    Please, not the moons again! Voting and Bagging 
    
 
    
 
 
-
    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)
+
    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(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)
 
    
 
    
 
 
-
    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.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)
+
+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))
 
    
 
    
 
    
@@ -230,6 +289,13 @@ np.sum(y_pred =
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • 44
    • 
+  
  • 45
    • 
+  
  • 46
    • 
+  
  • 47
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
index 6f47e526d..cadb27ef5 100644
--- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
+++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -191,11 +214,64 @@ MathJax.Hub.Config({
 
 
 
-
    Feature Importance 
    
+
    Now Bagging  
    
 
 
    
-Example will be added here.
 
+
+
    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)
+
    
+
    
+
+
+
    from sklearn.metrics import accuracy_score
+print(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))
+
    
+
    
+
+
+
    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()
+
    
 
    
 
    
 
@@ -215,6 +291,14 @@ Example will be added here.
   
  • 40
    • 
   
  • 41
    • 
   
  • 42
    • 
+  
  • 43
    • 
+  
  • 44
    • 
+  
  • 45
    • 
+  
  • 46
    • 
+  
  • 47
    • 
+  
  • 48
    • 
+  
  • ...
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
index 71dfb146f..ec2e98fd3 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html
@@ -64,40 +64,55 @@ Automatically generated HTML file from DocOnce source
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -151,31 +166,39 @@ MathJax.Hub.Config({
       
  • Classification tree, how to split nodes
    • 
       
  • Visualizing the Tree, Classification
    • 
       
  • Visualizing the Tree, The Moons
    • 
-      
  • Computing the Gini index
    • 
-      
  • Simple Python Code to read in Data and perform Classification
    • 
-      
  • Computing the Gini Factor
    • 
-      
  • Entropy and the ID3 algorithm
    • 
-      
  • Implementing the ID3 Algorithm
    • 
-      
  • Cancer Data again now with Decision Trees and other Methods
    • 
-      
  • Another example, the moons again
    • 
-      
  • Playing around with regions
    • 
-      
  • Regression trees
    • 
-      
  • Final regressor code
    • 
-      
  • Pros and cons of trees, pros
    • 
-      
  • Disadvantages
    • 
-      
  • 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
    • 
-      
  • A simple scikit-learn example
    • 
-      
  • Then random forests
    • 
-      
  • Feature Importance
    • 
-      
  • Boosting: AdaBoost
    • 
-      
  • Gradient Boosting
    • 
-      
  • Gradient Boots with Early Stopping
    • 
+      
  • Algorithms for Setting up Decision Trees
    • 
+      
  • The CART algorithm for Classification
    • 
+      
  • The CART algorithm for Regression
    • 
+      
  • Computing the Gini index
    • 
+      
  • Simple Python Code to read in Data and perform Classification
    • 
+      
  • Computing the Gini Factor
    • 
+      
  • Entropy and the ID3 algorithm
    • 
+      
  • Implementing the ID3 Algorithm
    • 
+      
  • Cancer Data again now with Decision Trees and other Methods
    • 
+      
  • Another example, the moons again
    • 
+      
  • Playing around with regions
    • 
+      
  • Regression trees
    • 
+      
  • Final regressor code
    • 
+      
  • Pros and cons of trees, pros
    • 
+      
  • Disadvantages
    • 
+      
  • From a Single Tree to Many Trees, that is 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
    • 
+      
  • A simple scikit-learn example
    • 
+      
  • Then random forests
    • 
+      
  • Feature Importance
    • 
+      
  • Boosting, a Bird'e Eye
    • 
+      
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • 
+      
  • AdaBoost Examples
    • 
+      
  • Gradient boosting: Basic Algorithm
    • 
+      
  • Gradient Boosting, Examples
    • 
+      
  • Gradient Boots with Early Stopping
    • 
+      
  • XGBoost: Extreme Gradient Boosting
    • 
 
         
       
@@ -210,7 +233,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 1, 2019
     
+
    Nov 2, 2019
     
 
    
 
    
 
@@ -234,7 +257,7 @@ MathJax.Hub.Config({
   
  • 9
    • 
   
  • 10
    • 
   
  • ...
    • 
-  
  • 42
    • 
+  
  • 50
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html b/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html
index 758d6151d..5b1289cc6 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 1, 2019
     
+
    Nov 2, 2019
     
 
    
 
    
 
@@ -194,6 +194,9 @@ given some assumptions, make predictions about the target feature value
 
 
    
 
    A typical Decision Tree with its pertinent Jargon, Classification Problem 
    
+
+
    
+Figure to come here.
 
    
 
 
@@ -691,7 +694,31 @@ os.system(cmd)
 
 
 
    
-
    Computing the Gini index 
    
+
    Algorithms for Setting up Decision Trees 
    
+Two algorithms stand out in the set up of decision trees:
+
+
      
+
    1.  The CART (Classification And Regression Tree) algorithm for both classification and regression
      2. 
+
    2.  The ID3 algorithm based on the computation of the information gain for classification
      3. 
+
    
+
    
+
+We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the gini index or the entropy to split a tree in two branches.
+
    
+
+
+
    
+
    The CART algorithm for Classification 
    
+
    
+
+
+
    
+
    The CART algorithm for Regression 
    
+
    
+
+
+
    
+
    Computing the Gini index 
    
 
 
    
 The example we will look at is a classical one in many Machine
@@ -732,7 +759,7 @@ The table here summarizes the various attributes and
 
 
 
    
-
    Simple Python Code to read in Data and perform Classification 
    
+
    Simple Python Code to read in Data and perform Classification 
    
 
 
    
 
@@ -809,7 +836,7 @@ os.system(cmd)
 
 
 
    
-
    Computing the Gini Factor  
    
+
    Computing the Gini Factor  
    
 
 
    
 The above functions (gini, entropy and misclassification error) are
@@ -888,7 +915,7 @@ split = get_split(dataset)
 
 
 
    
-
    Entropy and the ID3 algorithm 
    
+
    Entropy and the ID3 algorithm 
    
 
 
    
 ID3, learns decision trees by constructing
@@ -925,7 +952,7 @@ attributes at each step while growing the tree.
 
 
 
    
-
    Implementing the ID3 Algorithm 
    
+
    Implementing the ID3 Algorithm 
    
 
 
    
 
@@ -1122,7 +1149,7 @@ attributes at each step while growing the tree.
 
 
 
    
-
    Cancer Data again now with Decision Trees and other Methods 
    
+
    Cancer Data again now with Decision Trees and other Methods 
    
 
    
 
 
@@ -1172,7 +1199,7 @@ deep_tree_clf.fit(X_train_scaled, y_train)
 
 
 
    
-
    Another example, the moons again 
    
+
    Another example, the moons again 
    
 
    
 
 
@@ -1245,7 +1272,7 @@ plt.show()
 
 
 
    
-
    Playing around with regions 
    
+
    Playing around with regions 
    
 
    
 
 
@@ -1274,7 +1301,7 @@ plt.show()
 
 
 
    
-
    Regression trees 
    
+
    Regression trees 
    
 
    
 
 
@@ -1297,7 +1324,7 @@ tree_reg.fit(X, y)
 
 
 
    
-
    Final regressor code 
    
+
    Final regressor code 
    
 
    
 
 
@@ -1376,7 +1403,7 @@ plt.show()
 
 
 
    
-
    Pros and cons of trees, pros 
    
+
    Pros and cons of trees, pros 
    
 
 
      
 
    •  White box, easy to interpret model. Some people believe that decision trees more closely mirror human decision-making than do the regression and classification approaches discussed earlier (think of support vector machines)
      • 
@@ -1391,7 +1418,7 @@ plt.show()
 
 
 
      
-
      Disadvantages 
      
+
      Disadvantages 
      
 
 
        
 
      •  Unfortunately, trees generally do not have the same level of predictive accuracy as some of the other regression and classification approaches
        • 
@@ -1409,7 +1436,32 @@ However, by aggregating many decision trees, using methods like bagging, random
 
 
 
        
-
        Bagging 
        
+
        From a Single Tree to Many Trees, that is meet the Jungle of Methods 
        
+
+
        
+As stated above and seen in many of the examples discussed here about
+a single decision tree, we often end up overfitting our training
+data. This normally means that we have a high variance. Can we reduce
+the variance of a statistical learning method?
+
+
        
+This leads us to a set of different methods that can combine different
+machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are
+
+
          
+
        1.  Votign classifiers
          2. 
+
        2.  Bagging and Pasting
          3. 
+
        3.  Random forests
          4. 
+
        4.  Boosting methods
          5. 
+
        
+
        
+
+We discuss these methods here.
+
        
+
+
+
        
+
        Bagging 
        
 
 
        
 The plain decision trees suffer from high
@@ -1428,7 +1480,7 @@ learning method.
 
 
 
        
-
        More bagging 
        
+
        More bagging 
        
 
 
        
 Bagging typically results in improved accuracy
@@ -1457,7 +1509,7 @@ predictor, averaged over all \( B \) trees.
 
 
 
        
-
        Simple Voting Example, head or tail 
        
+
        Simple Voting Example, head or tail 
        
 
        
 
 
@@ -1478,7 +1530,7 @@ plt.show()
 
 
 
        
-
        Using the Voting Classifier 
        
+
        Using the Voting Classifier 
        
 
        
 
 
@@ -1530,7 +1582,7 @@ voting_clf.fit(X_train, y_train)
 
 
 
        
-
        Please, not the moons again! Voting and Bagging 
        
+
        Please, not the moons again! Voting and Bagging 
        
 
        
 
 
@@ -1589,7 +1641,7 @@ voting_clf.fit(X_train, y_train)
 
 
 
        
-
        Now Bagging  
        
+
        Now Bagging  
        
 
 
        
 
@@ -1651,7 +1703,7 @@ plt.show()
 
 
 
        
-
        Random forests 
        
+
        Random forests 
        
 
 
        
 Random forests provide an improvement over bagged trees by way of a
@@ -1697,7 +1749,7 @@ this setting.
 
 
 
        
-
        A simple scikit-learn example 
        
+
        A simple scikit-learn example 
        
 
        
 
 
@@ -1716,7 +1768,7 @@ accuracy = cross_validate(Random_Forest_model,X,Y,cv=Then random forests 
+
        Then random forests 
        
 
        
 
 
@@ -1739,7 +1791,7 @@ np.sum(y_pred == y_pred_rf) / len(y_pred)
 
 
 
        
-
        Feature Importance 
        
+
        Feature Importance 
        
 
 
        
 Example will be added here.
@@ -1747,8 +1799,17 @@ Example will be added here.
 
 
 
        
-
        Boosting: AdaBoost 
        
+
        Boosting, a Bird'e Eye 
        
+
        
 
+
+
        
+
        Adaptive boosting: AdaBoost, Basic Algorithm 
        
+
        
+
+
+
        
+
        AdaBoost Examples 
        
 
        
 
 
@@ -1788,7 +1849,12 @@ plt.show()
 
 
 
        
-
        Gradient Boosting 
        
+
        Gradient boosting: Basic Algorithm 
        
+
        
+
+
+
        
+
        Gradient Boosting, Examples 
        
 
        
 
 
@@ -1879,7 +1945,7 @@ plt.show()
 
 
 
        
-
        Gradient Boots with Early Stopping 
        
+
        Gradient Boots with Early Stopping 
        
 
        
 
 
@@ -1943,6 +2009,11 @@ error_going_up = 0
 
        
 
 
+
        
+
        XGBoost: Extreme Gradient Boosting 
        
+
        
+
+
 
 
     
 
     
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
index eb1f7a577..160958595 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html
@@ -84,40 +84,55 @@ div { text-align: justify; text-justify: inter-word; }
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -159,7 +174,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 1, 2019
     
+
    Nov 2, 2019
     
 
    
 
    
 









    
@@ -200,6 +215,9 @@ given some assumptions, make predictions about the target feature value
 
 
    A typical Decision Tree with its pertinent Jargon, Classification Problem 
    
 
+
    
+Figure to come here.
+
 
    
 









    
 
@@ -675,7 +693,30 @@ os.system(cmd)
 
    
 









    
 
-
    Computing the Gini index 
    
+
    Algorithms for Setting up Decision Trees 
    
+Two algorithms stand out in the set up of decision trees:
+
+
      
+
    1.  The CART (Classification And Regression Tree) algorithm for both classification and regression
      2. 
+
    2.  The ID3 algorithm based on the computation of the information gain for classification
      3. 
+
    
+
+We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the gini index or the entropy to split a tree in two branches.
+
+
    
+









    
+
+
    The CART algorithm for Classification 
    
+
+
    
+









    
+
+
    The CART algorithm for Regression 
    
+
+
    
+









    
+
+
    Computing the Gini index 
    
 
 
    
 The example we will look at is a classical one in many Machine
@@ -715,7 +756,7 @@ The table here summarizes the various attributes and
 
    
 









    
 
-
    Simple Python Code to read in Data and perform Classification 
    
+
    Simple Python Code to read in Data and perform Classification 
    
 
 
    
 
@@ -791,7 +832,7 @@ os.system(cmd)
 
    
 









    
 
-
    Computing the Gini Factor  
    
+
    Computing the Gini Factor  
    
 
 
    
 The above functions (gini, entropy and misclassification error) are
@@ -869,7 +910,7 @@ split = get_split(dataset)
 
    
 









    
 
-
    Entropy and the ID3 algorithm 
    
+
    Entropy and the ID3 algorithm 
    
 
 
    
 ID3, learns decision trees by constructing
@@ -905,7 +946,7 @@ attributes at each step while growing the tree.
 
    
 









    
 
-
    Implementing the ID3 Algorithm 
    
+
    Implementing the ID3 Algorithm 
    
 
 
    
 
@@ -1101,7 +1142,7 @@ attributes at each step while growing the tree.
 
    
 









    
 
-
    Cancer Data again now with Decision Trees and other Methods 
    
+
    Cancer Data again now with Decision Trees and other Methods 
    
 
    
 
 
@@ -1150,7 +1191,7 @@ deep_tree_clf.fit(X_train_scaled, y_train)
 
    
 









    
 
-
    Another example, the moons again 
    
+
    Another example, the moons again 
    
 
    
 
 
@@ -1222,7 +1263,7 @@ plt.show()
 
    
 









    
 
-
    Playing around with regions 
    
+
    Playing around with regions 
    
 
    
 
 
@@ -1250,7 +1291,7 @@ plt.show()
 
    
 









    
 
-
    Regression trees 
    
+
    Regression trees 
    
 
    
 
 
@@ -1272,7 +1313,7 @@ tree_reg.fit(X, y)
 
    
 









    
 
-
    Final regressor code 
    
+
    Final regressor code 
    
 
    
 
 
@@ -1350,7 +1391,7 @@ plt.show()
 
    
 









    
 
-
    Pros and cons of trees, pros 
    
+
    Pros and cons of trees, pros 
    
 
 
      
 
    •  White box, easy to interpret model. Some people believe that decision trees more closely mirror human decision-making than do the regression and classification approaches discussed earlier (think of support vector machines)
      • 
@@ -1364,7 +1405,7 @@ plt.show()
 
 









      
 
-
      Disadvantages 
      
+
      Disadvantages 
      
 
 
        
 
      •  Unfortunately, trees generally do not have the same level of predictive accuracy as some of the other regression and classification approaches
        • 
@@ -1381,7 +1422,31 @@ However, by aggregating many decision trees, using methods like bagging, random
 
        
 









        
 
-
        Bagging 
        
+
        From a Single Tree to Many Trees, that is meet the Jungle of Methods 
        
+
+
        
+As stated above and seen in many of the examples discussed here about
+a single decision tree, we often end up overfitting our training
+data. This normally means that we have a high variance. Can we reduce
+the variance of a statistical learning method?
+
+
        
+This leads us to a set of different methods that can combine different
+machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are
+
+
          
+
        1.  Votign classifiers
          2. 
+
        2.  Bagging and Pasting
          3. 
+
        3.  Random forests
          4. 
+
        4.  Boosting methods
          5. 
+
        
+
+We discuss these methods here.
+
+
        
+









        
+
+
        Bagging 
        
 
 
        
 The plain decision trees suffer from high
@@ -1400,7 +1465,7 @@ learning method.
 
        
 









        
 
-
        More bagging 
        
+
        More bagging 
        
 
 
        
 Bagging typically results in improved accuracy
@@ -1429,7 +1494,7 @@ predictor, averaged over all \( B \) trees.
 
        
 









        
 
-
        Simple Voting Example, head or tail 
        
+
        Simple Voting Example, head or tail 
        
 
        
 
 
@@ -1449,7 +1514,7 @@ plt.show()
 
        
 









        
 
-
        Using the Voting Classifier 
        
+
        Using the Voting Classifier 
        
 
        
 
 
@@ -1500,7 +1565,7 @@ voting_clf.fit(X_train, y_train)
 
        
 









        
 
-
        Please, not the moons again! Voting and Bagging 
        
+
        Please, not the moons again! Voting and Bagging 
        
 
        
 
 
@@ -1558,7 +1623,7 @@ voting_clf.fit(X_train, y_train)
 
        
 









        
 
-
        Now Bagging  
        
+
        Now Bagging  
        
 
 
        
 
@@ -1619,7 +1684,7 @@ plt.show()
 
        
 









        
 
-
        Random forests 
        
+
        Random forests 
        
 
 
        
 Random forests provide an improvement over bagged trees by way of a
@@ -1663,7 +1728,7 @@ this setting.
 
        
 









        
 
-
        A simple scikit-learn example 
        
+
        A simple scikit-learn example 
        
 
        
 
 
@@ -1681,7 +1746,7 @@ accuracy = cross_validate(Random_Forest_model,X,Y,cv=Then random forests 
+
        Then random forests 
        
 
        
 
 
@@ -1703,7 +1768,7 @@ np.sum(y_pred == y_pred_rf) / len(y_pred)
 
        
 









        
 
-
        Feature Importance 
        
+
        Feature Importance 
        
 
 
        
 Example will be added here.
@@ -1711,8 +1776,17 @@ Example will be added here.
 
        
 









        
 
-
        Boosting: AdaBoost 
        
+
        Boosting, a Bird'e Eye 
        
 
+
        
+









        
+
+
        Adaptive boosting: AdaBoost, Basic Algorithm 
        
+
+
        
+









        
+
+
        AdaBoost Examples 
        
 
        
 
 
@@ -1751,7 +1825,12 @@ plt.show()
 
        
 









        
 
-
        Gradient Boosting 
        
+
        Gradient boosting: Basic Algorithm 
        
+
+
        
+









        
+
+
        Gradient Boosting, Examples 
        
 
        
 
 
@@ -1841,7 +1920,7 @@ plt.show()
 
        
 









        
 
-
        Gradient Boots with Early Stopping 
        
+
        Gradient Boots with Early Stopping 
        
 
        
 
 
@@ -1903,6 +1982,9 @@ error_going_up = 0
 print("Minimum validation MSE:", min_val_error)
 
    
 
    
+









    
+
+
    XGBoost: Extreme Gradient Boosting 
    
 
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees.html b/doc/pub/DecisionTrees/html/DecisionTrees.html
index 0d5d09fec..dc47b1555 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees.html
@@ -89,40 +89,55 @@ div { text-align: justify; text-justify: inter-word; }
               ('Classification tree, how to split nodes', 2, None, '___sec13'),
               ('Visualizing the Tree, Classification', 2, None, '___sec14'),
               ('Visualizing the Tree, The Moons', 2, None, '___sec15'),
-              ('Computing the Gini index', 2, None, '___sec16'),
+              ('Algorithms for Setting up Decision Trees', 2, None, '___sec16'),
+              ('The CART algorithm for Classification', 2, None, '___sec17'),
+              ('The CART algorithm for Regression', 2, None, '___sec18'),
+              ('Computing the Gini index', 2, None, '___sec19'),
               ('Simple Python Code to read in Data and perform Classification',
                2,
                None,
-               '___sec17'),
-              ('Computing the Gini Factor', 2, None, '___sec18'),
-              ('Entropy and the ID3 algorithm', 2, None, '___sec19'),
-              ('Implementing the ID3 Algorithm', 2, None, '___sec20'),
+               '___sec20'),
+              ('Computing the Gini Factor', 2, None, '___sec21'),
+              ('Entropy and the ID3 algorithm', 2, None, '___sec22'),
+              ('Implementing the ID3 Algorithm', 2, None, '___sec23'),
               ('Cancer Data again now with Decision Trees and other Methods',
                2,
                None,
-               '___sec21'),
-              ('Another example, the moons again', 2, None, '___sec22'),
-              ('Playing around with regions', 2, None, '___sec23'),
-              ('Regression trees', 2, None, '___sec24'),
-              ('Final regressor code', 2, None, '___sec25'),
-              ('Pros and cons of trees, pros', 2, None, '___sec26'),
-              ('Disadvantages', 2, None, '___sec27'),
-              ('Bagging', 2, None, '___sec28'),
-              ('More bagging', 2, None, '___sec29'),
-              ('Simple Voting Example, head or tail', 2, None, '___sec30'),
-              ('Using the Voting Classifier', 2, None, '___sec31'),
+               '___sec24'),
+              ('Another example, the moons again', 2, None, '___sec25'),
+              ('Playing around with regions', 2, None, '___sec26'),
+              ('Regression trees', 2, None, '___sec27'),
+              ('Final regressor code', 2, None, '___sec28'),
+              ('Pros and cons of trees, pros', 2, None, '___sec29'),
+              ('Disadvantages', 2, None, '___sec30'),
+              ('From a Single Tree to Many Trees, that is meet the Jungle of '
+               'Methods',
+               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'),
               ('Please, not the moons again! Voting and Bagging',
                2,
                None,
-               '___sec32'),
-              ('Now Bagging', 2, None, '___sec33'),
-              ('Random forests', 2, None, '___sec34'),
-              ('A simple scikit-learn example', 2, None, '___sec35'),
-              ('Then random forests', 2, None, '___sec36'),
-              ('Feature Importance', 2, None, '___sec37'),
-              ('Boosting: AdaBoost', 2, None, '___sec38'),
-              ('Gradient Boosting', 2, None, '___sec39'),
-              ('Gradient Boots with Early Stopping', 2, None, '___sec40')]}
+               '___sec36'),
+              ('Now Bagging', 2, None, '___sec37'),
+              ('Random forests', 2, None, '___sec38'),
+              ('A simple scikit-learn example', 2, None, '___sec39'),
+              ('Then random forests', 2, None, '___sec40'),
+              ('Feature Importance', 2, None, '___sec41'),
+              ("Boosting, a Bird'e Eye", 2, None, '___sec42'),
+              ('Adaptive boosting: AdaBoost, Basic Algorithm',
+               2,
+               None,
+               '___sec43'),
+              ('AdaBoost Examples', 2, None, '___sec44'),
+              ('Gradient boosting: Basic Algorithm', 2, None, '___sec45'),
+              ('Gradient Boosting, Examples', 2, None, '___sec46'),
+              ('Gradient Boots with Early Stopping', 2, None, '___sec47'),
+              ('XGBoost: Extreme Gradient Boosting', 2, None, '___sec48')]}
 end of tocinfo -->
 
 
@@ -164,7 +179,7 @@ MathJax.Hub.Config({
 
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
    
 
    
 
    
-
    Nov 1, 2019
     
+
    Nov 2, 2019
     
 
    
 
    
 









    
@@ -205,6 +220,9 @@ given some assumptions, make predictions about the target feature value
 
 
    A typical Decision Tree with its pertinent Jargon, Classification Problem 
    
 
+
    
+Figure to come here.
+
 
    
 









    
 
@@ -680,7 +698,30 @@ os.system(cmd)
 
    
 









    
 
-
    Computing the Gini index 
    
+
    Algorithms for Setting up Decision Trees 
    
+Two algorithms stand out in the set up of decision trees:
+
+
      
+
    1.  The CART (Classification And Regression Tree) algorithm for both classification and regression
      2. 
+
    2.  The ID3 algorithm based on the computation of the information gain for classification
      3. 
+
    
+
+We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the gini index or the entropy to split a tree in two branches.
+
+
    
+









    
+
+
    The CART algorithm for Classification 
    
+
+
    
+









    
+
+
    The CART algorithm for Regression 
    
+
+
    
+









    
+
+
    Computing the Gini index 
    
 
 
    
 The example we will look at is a classical one in many Machine
@@ -720,7 +761,7 @@ The table here summarizes the various attributes and
 
    
 









    
 
-
    Simple Python Code to read in Data and perform Classification 
    
+
    Simple Python Code to read in Data and perform Classification 
    
 
 
    
 
@@ -796,7 +837,7 @@ os.system(cmd)
 
    
 









    
 
-
    Computing the Gini Factor  
    
+
    Computing the Gini Factor  
    
 
 
    
 The above functions (gini, entropy and misclassification error) are
@@ -874,7 +915,7 @@ split = get_split(dataset)
 
    
 









    
 
-
    Entropy and the ID3 algorithm 
    
+
    Entropy and the ID3 algorithm 
    
 
 
    
 ID3, learns decision trees by constructing
@@ -910,7 +951,7 @@ attributes at each step while growing the tree.
 
    
 









    
 
-
    Implementing the ID3 Algorithm 
    
+
    Implementing the ID3 Algorithm 
    
 
 
    
 
@@ -1106,7 +1147,7 @@ attributes at each step while growing the tree.
 
    
 









    
 
-
    Cancer Data again now with Decision Trees and other Methods 
    
+
    Cancer Data again now with Decision Trees and other Methods 
    
 
    
 
 
@@ -1155,7 +1196,7 @@ deep_tree_clf.fit(X_train_scaled, y_train)
 
    
 









    
 
-
    Another example, the moons again 
    
+
    Another example, the moons again 
    
 
    
 
 
@@ -1227,7 +1268,7 @@ plt.show()
 
    
 









    
 
-
    Playing around with regions 
    
+
    Playing around with regions 
    
 
    
 
 
@@ -1255,7 +1296,7 @@ plt.show()
 
    
 









    
 
-
    Regression trees 
    
+
    Regression trees 
    
 
    
 
 
@@ -1277,7 +1318,7 @@ tree_reg.fit(X, y)
 
    
 









    
 
-
    Final regressor code 
    
+
    Final regressor code 
    
 
    
 
 
@@ -1355,7 +1396,7 @@ plt.show()
 
    
 









    
 
-
    Pros and cons of trees, pros 
    
+
    Pros and cons of trees, pros 
    
 
 
      
 
    •  White box, easy to interpret model. Some people believe that decision trees more closely mirror human decision-making than do the regression and classification approaches discussed earlier (think of support vector machines)
      • 
@@ -1369,7 +1410,7 @@ plt.show()
 
 









      
 
-
      Disadvantages 
      
+
      Disadvantages 
      
 
 
        
 
      •  Unfortunately, trees generally do not have the same level of predictive accuracy as some of the other regression and classification approaches
        • 
@@ -1386,7 +1427,31 @@ However, by aggregating many decision trees, using methods like bagging, random
 
        
 









        
 
-
        Bagging 
        
+
        From a Single Tree to Many Trees, that is meet the Jungle of Methods 
        
+
+
        
+As stated above and seen in many of the examples discussed here about
+a single decision tree, we often end up overfitting our training
+data. This normally means that we have a high variance. Can we reduce
+the variance of a statistical learning method?
+
+
        
+This leads us to a set of different methods that can combine different
+machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are
+
+
          
+
        1.  Votign classifiers
          2. 
+
        2.  Bagging and Pasting
          3. 
+
        3.  Random forests
          4. 
+
        4.  Boosting methods
          5. 
+
        
+
+We discuss these methods here.
+
+
        
+









        
+
+
        Bagging 
        
 
 
        
 The plain decision trees suffer from high
@@ -1405,7 +1470,7 @@ learning method.
 
        
 









        
 
-
        More bagging 
        
+
        More bagging 
        
 
 
        
 Bagging typically results in improved accuracy
@@ -1434,7 +1499,7 @@ predictor, averaged over all \( B \) trees.
 
        
 









        
 
-
        Simple Voting Example, head or tail 
        
+
        Simple Voting Example, head or tail 
        
 
        
 
 
@@ -1454,7 +1519,7 @@ plt.show()
 
        
 









        
 
-
        Using the Voting Classifier 
        
+
        Using the Voting Classifier 
        
 
        
 
 
@@ -1505,7 +1570,7 @@ voting_clf.fit(X_train, y_train)
 
        
 









        
 
-
        Please, not the moons again! Voting and Bagging 
        
+
        Please, not the moons again! Voting and Bagging 
        
 
        
 
 
@@ -1563,7 +1628,7 @@ voting_clf.fit(X_train, y_train)
 
        
 









        
 
-
        Now Bagging  
        
+
        Now Bagging  
        
 
 
        
 
@@ -1624,7 +1689,7 @@ plt.show()
 
        
 









        
 
-
        Random forests 
        
+
        Random forests 
        
 
 
        
 Random forests provide an improvement over bagged trees by way of a
@@ -1668,7 +1733,7 @@ this setting.
 
        
 









        
 
-
        A simple scikit-learn example 
        
+
        A simple scikit-learn example 
        
 
        
 
 
@@ -1686,7 +1751,7 @@ accuracy = cross_validate(Random_Forest_mode
 
        
 









        
 
-
        Then random forests 
        
+
        Then random forests 
        
 
        
 
 
@@ -1708,7 +1773,7 @@ np.sum(y_pred =
 
        
 









        
 
-
        Feature Importance 
        
+
        Feature Importance 
        
 
 
        
 Example will be added here.
@@ -1716,8 +1781,17 @@ Example will be added here.
 
        
 









        
 
-
        Boosting: AdaBoost 
        
+
        Boosting, a Bird'e Eye 
        
 
+
        
+









        
+
+
        Adaptive boosting: AdaBoost, Basic Algorithm 
        
+
+
        
+









        
+
+
        AdaBoost Examples 
        
 
        
 
 
@@ -1756,7 +1830,12 @@ plt.show()
 
        
 









        
 
-
        Gradient Boosting 
        
+
        Gradient boosting: Basic Algorithm 
        
+
+
        
+









        
+
+
        Gradient Boosting, Examples 
        
 
        
 
 
@@ -1846,7 +1925,7 @@ plt.show()
 
        
 









        
 
-
        Gradient Boots with Early Stopping 
        
+
        Gradient Boots with Early Stopping 
        
 
        
 
 
@@ -1908,6 +1987,9 @@ error_going_up = print("Minimum validation MSE:", min_val_error)
 
    
 
    
+









    
+
+
    XGBoost: Extreme Gradient Boosting 
    
 
 
 
diff --git a/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb b/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb
index 964cf0fc9..3739cfe22 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 1, 2019**\n",
+    "Date: **Nov 2, 2019**\n",
     "\n",
     "Copyright 1999-2019, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license\n",
     "\n",
@@ -47,6 +47,7 @@
     "\n",
     "## A typical Decision Tree with its pertinent Jargon, Classification Problem\n",
     "\n",
+    "Figure to come here.\n",
     "\n",
     "\n",
     "\n",
@@ -92,40 +93,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "2nd degree coefficients:\n",
-      "zero power:  -7.159472988729101\n",
-      "first power:  0.07804791022278464\n",
-      "second power:  -0.0005586058040523115\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"
-    }
-   ],
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "%matplotlib inline\n",
     "\n",
@@ -536,469 +508,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
-   "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": 31,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "import os\n",
     "from sklearn.datasets import load_breast_cancer\n",
@@ -1044,20 +558,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0"
-      ]
-     },
-     "execution_count": 32,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "# Common imports\n",
     "import numpy as np\n",
@@ -1089,6 +594,21 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
+    "## Algorithms for Setting up Decision Trees\n",
+    "Two algorithms stand out in the set up of decision trees:\n",
+    "1. The CART (Classification And Regression Tree) algorithm for both classification and regression\n",
+    "\n",
+    "2. The ID3 algorithm based on the computation of the information gain for classification\n",
+    "\n",
+    "We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the **gini** index or the **entropy** to split a tree in two branches.\n",
+    "\n",
+    "## The CART algorithm for Classification\n",
+    "\n",
+    "\n",
+    "## The CART algorithm for Regression\n",
+    "\n",
+    "\n",
+    "\n",
     "## Computing the Gini index\n",
     "\n",
     "The example we will look at is a classical one in many Machine\n",
@@ -1130,78 +650,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
-   "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": 33,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "# Common imports\n",
     "import numpy as np\n",
@@ -1288,73 +741,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 34,
-   "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",
@@ -1458,37 +849,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 35,
-   "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",
@@ -1688,32 +1053,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 36,
-   "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",
@@ -1767,20 +1111,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 37,
-   "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",
@@ -1857,20 +1192,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
-   "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",
@@ -1903,8 +1229,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 39,
-   "metadata": {},
+   "execution_count": 10,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "# Quadratic training set + noise\n",
@@ -1917,24 +1245,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
-   "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": 40,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "execution_count": 11,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.tree import DecisionTreeRegressor\n",
     "\n",
@@ -1951,20 +1266,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
-   "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",
@@ -2008,20 +1314,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 42,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqwAAAEiCAYAAAA8kIAFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzsnXl8VOX1/9/PTGaSEAhbQLAIKJVV6kYtoyJRXL7dfkXpt61t1bZ+DWpbbbXiUq22tEXUWm211liX2kW7oK3VWlQ0gDIuWJeoCIoIKiBLEELIMsk8vz+ee2fuzNyZTJLJrOf9es1r7n6fTO4993PPc55zlNYaQRAEQRAEQchXPLlugCAIgiAIgiCkQgSrIAiCIAiCkNeIYBUEQRAEQRDyGhGsgiAIgiAIQl4jglUQBEEQBEHIa0SwCoIgCIIgCHmNCFZBSIJSarxSSiulZvTxOFop9cVMtUsQhPyhVO5vpdQ9SqmHC/lcSqk6pdQmpVRYKXVNpo8v9C8iWIVeYxkVrZS6Km55rbW8JldtS4ZS6hql1Gtpbv4eMBp4Oc1jJzOyo4F/pXlOQRAKC7m/CwCl1FDgVuB64GPADb04xnFKqYeUUh9Yz7hvuGyjrOfMZqVUq1KqQSk1rc9/gCCCVegzbcAlSqkR/X0ipZSvv8/hOJdfa92ltd6qte7sy7GsY7Rnqm2CIOQPcn8XDOOAMuBhrfUWrfXeXhxjIPAacCHQmmSbBcDFwHeBTwLbgMeVUoN6cT7BgQhWoa88BbwLXJVqI+vN9DmlVJtS6kOl1C+VUv4U29te2s8opZ5XSnUAp1jrPq+UetE61gal1M+cx1JKnaaUetV6u21SSi1XSu1nvQ1fDUyzjh15Q7amv62UekAp1QL83C0kQCk12XrD3q2U2quUCiqlplvdS2cBn3Ucu9Zx7C86jjFdKfWEo333KKUGO9bfo5R6WCl1ofUmv0spdbdSakDc7/ms1Ybd1m90SLf/LUEQkmJ5w25TSv3Cuje3W/dhuVLqVqXUR1aX8hmOfSL3t8NmzFNKPa6U2qeUekMpdVKa5/cppX5leefalVLvKaWudaz/ulLqBaVUs1Jqm1Lqb0qpjznW23bz05aNbFVKrVRKjVFKzVZKvWLZjIeVUsMd+9k250rLPu+1bE5lirYqpdQCpdR66zyNSqmvx23zI6XURutv2aqUuje9/0SvznWtUmqttf5dpdR1SqkKa903gJesTd+xfqPxPW2H1vrfWusrtNZ/B8Ju7QS+B1yrtV6itX4N81wYBHy1p+cTYhHBKvSVMHAZcK5SaoLbBpZBfRRjMA4HzgZOBxalcfzFwJXAZOA5pdQpwJ+AW4BpwLeALwI/t841Crgf+D0wBTgO+IN1rL8AvwDWYrrxRlvLbK4G/g1Mx3Qdxf8d+wNPAxo4CTjC2s6L6V76K/CE49irXI5RBSwF9gJHAacCRwN3xW06CzgEOBH4srXdhdYxyoB/Wm05FPgUcBPQ5fYDCoLQI74GNGPuq2sx99Y/gHXADIxt+Z1SanSKY/wM+BXm/nwBuF8pNTCNc1+Aude/AhyMuffXOtb7MXbqUOBzQA1wn8txfowRTp8ChmLs3I+AOqAWYzuvidtntnXcOcA84GSM/U3GTzG2/NvAVIw9v10p9VkApdQ84AfA+dbf8jng+VR/fG/PZdGCeR5Msc75FeCH1rq/AP9jTR+Fsc/vKaVmWeI81eeKHrTzQGAU8Ji9QGvdCqzA2HmhL2it5SOfXn2AezDdK2A8rfdb07UYUVdjzf8MeAvwOPb9BtAODEhybPsY8+KWrwCuils2FyMAFUZEamBckuNeA7zmslwDv45bNt5aPsPxd2wE/N39Hi7H/qI1fQ6wGxjk8rd+3HGc9wCvY5s7gCes6WHW9rNzfQ3IRz7F9AEagKBjXgHbgYccy3xAh+Oedt7fts2Y79j+Y9ayY9M4/6+AZYBKs72TrWOPseZtW3KKY5vvWMuOcCyLsYOWzfkIGOhY9nXLRlc5trHtfRWmS3xWXHtuAv5tTV+EEdu+XvwfenSuJMc4F3jbMT/D+h3GO5ZVAh/v5jMsyfH3At+IW3a0dY6xccvvApbm+vou9E8ZgpAZLgWCSqnrXdZNAZ7VWju7UJ7GeAs+Drya4rir4+aPBI5SSl3qWObBGJ5RwCsYL+drSqnHrOm/a623p/E3xJ8rnsOBp7XWHWkcKxlTgFe11s2OZaswnuqpwNvWsje01k6P6WaMtwStdZNS6h5gqVJqGeYB93et9aY+tEsQBEPEHmmttVJqG9DoWBZSSu0CRqZzDMy9Szfb29wDPA6ss+zXv4FHbduplDoC42E9DPPiqqz9xgLvJzn/h9Z3Y9yy+Pa8qmPjOoMYGz2BRBs9FagA/qOU0o7lPkyIGMDfML1CG5RSS4H/YIR/T+N90zkXVljG9zDPlIGYni9vqgNr4/18O9U2Qv4gIQFCRtBaPw8sAa7r6a7drG+Jm/dgursOc3w+gely2m6JvJOtz6uYbqS3lFKHptGW+HNlG+dvEXJZF7lftdbfxAjYFcD/A9Za4RKCIPQNt3sv5f2Y6hjacrF1s7297X8xXtrLre1/jxmw43GEE+0DzsAM6LG7uePHAzjbq61jxy/ry/Pf3vfzxNriaRjbi9b6PWASMB/YgwnHetH6OzJ6LqXUTEwo2FJru8MxoWQpB+r2Q0jAVut7v7jl+znWCb1EPKxCJrkCeIOoEbVZA3xJKeVxeFmPxXSrre/hOf4LTNZaJ30rth4QQYzH9yfA65hYsFesc6Z8607BS8DXlckg4OZlTefYa4BvKaUGObysR2OM8pqeNEZr/Qrmb1qslHoUE9y/tCfHEAQhv7Dswt+Bv1s9Kc9ivIaDMDGrV2itN4AZYJrBU09XSlVpre0X95kkt9FvYMIFxmmtn0x2QK11G/AI8Ig1eGwrcAyOGM80SOdcxwAfaK0X2guUUuPSOPZqjPhNRVNarTRswPyNJ2Fil7EGfs0CLunBcQQXRLAKGUNr/bZSqh5rcJCD32C6an6jlLoZOAgzmOEWrfW+Hp7mJ8DDSqmNmEFOnZjBSUdprRdYb9onYoTbh5g37QMwRg9MF9I4q2ttE9Dcgy6q32Diov6qlPoZsAvj5VijtX7ZOvanlVKTgJ3A7jivBpgBYz8G7lVK/QgzIOJ24IFUItyJUupAjNfiIeADzO/5CeC2NP8OQRDyEKXURcAWTO7nEGZk+R5Md38VRrh9Ryl1Kya8aGGSQ/WGMuAu6yV/f4yNvsMhYCNorZuVUjcAN1gj41dguuFnAmGtdb01Mr8MeA4T7/ll6296qyeNSudcmAFxH1NKfQ3jrDgFM7C3u2P3KCTAGjj3cWvWA4xVSh0GNGmtN1khJDcBVyil3rTadSXm7/9zuucR3JGQACHT/AQjIiNorT8APo0Rjy9jAtDvw3hke4TWeinwWeB4zIjT5zFZCuz4zd2Yt+2HMYbxF8BCrfUfrfVLMHFhyzCDKbo1anF/x3GY7renMB7X7xL9e+/AeElXW8c+xuUY+zDGtNpq+z8xBvZb6bYD0yU4ERMjtg7TbfgnUo/oFQQh/2nGeOKex/QmHQZ8Wmu9z4rDPwszyPQNTCzrRRk893JMb9RTwIPAk5icosm4CjN46wfWfo9jsgtssNZ/hAnJWonJXToPOM32DveQlOfSWv8LUxDgJkwo2EmYrAiZZgbG7r+EGTfxY2v6J45trgN+ickgsxqTkeDkuHELQi9Q0fAaQRAEQRBKDSv0oEZr/blct0UQkiEeVkEQBEEQBCGvEcEqCIIgCFlAKfXbFKPRf5vr9mUDpdTYbkblj811G4X8REICBEEQBCELKKVGYuLX3dijtd6WzfbkAqtS3/gUm7yrte5MsV4oUUSwCoIgCIIgCHlN0aa1qqmp0ePHjIFXXgGPB8JW+s8jj6R1ewt6xw58foVv1HCo6mkeY0EQSpUXX3xxh9Z6RK7b0V/U1NTo8ePHu65raYHmZhg0SMymIAg9o6+2s2gF6/jx41n98MMwejTU1EBTE3R28tq3bmTyt4/HSxj2gW7Zg2f5UxAI5LrJgiAUAFYO4KJl/PjxrF6dWKU4GIQ5c6Cjw5jTZcvEbAqCkD59tZ3FPegqZOVs9/uhogKAXUuWUUYYhVWEOdQBDQ05aqAgCEJh0NBgxGpXl/kWsykIQjYpbsHaYVXP9PkignX4p4+KrNYAPj/U1ma9aYIg5DHBIJx6Ki0HTmXttFNprA/mukU5p7bWvPt7veZbzKYgCMkIBmHRIvOdKYo2JACIelgdgnXql6ZHKvqGBtfgf/Qh6dcSBCFKMAjHHosOhxkATGQNofmP0MhypteVrq0IBEwYQEODEatiNgVBcMMZPuT3Zy58qLg9rC4hAbRHy8b7DxwjVlcQhFgaGiAcDRtSQBkhdi5pyGmz8oFAAC6/XMymIAjJ6a/woeL2sDpDAmza2qLTduYAQRAEm7i+bg104mP4vFq3rQVBEAQHdviQ7WHNVPhQcQtWZ0iA12umnYJVctAKghBPIGAyi+zYAUCHfyDrfv1YSYcDCIIgpEt/hQ+VhmD1+6PLxMMqCEJ3eKLRUuXjRotYFQRB6AGBQOZDh4pbsDpCApq3NjMIaLpsMcPs9eJhzSrt7e00NTXR3NxMV1dXrpuT1/j9fmpqahg8eHCum1KadDoqQ5b4tRoMykArQRByT3ELVsvD2ry1hQFrXgRg6KpHouvFw5o12tvb2bRpE0OHDmX8+PH4fD6UUrluVl6itaa1tZX333+f8vJyKuwBg0L2EMEK9N9oX0EQhJ5SElkC2nfsQZmsq8RIJPGwZo2mpiaGDh1KTU0Nfr9fxGoKlFIMGDCAmpoatm/fnuvmlCZOkVrCL7ZSLEAQhHyhuAWrFRLg+9hIQpg41hiJKoI1azQ3N1NdXZ3rZhQUgwYNos0Zcy1kD/GwAlIsQBCE/KG4BavlYR08cRTvXP8AAGF/eXS986Ek9CtdXV34nOnFhG4pKyujU67R3CCCFYiO9l24UMIBBEHILTkRrEqp7yilViul2pVS93Sz7feVUluVUnuUUncppcpTbR+DPejK72fKNz4FgLdCBGuukDCAniG/V47QOlak5pFgzZrtdCDFAgRByAdy5WHdDPwUuCvVRkqpU4DLgDnAOOAg4Mdpn8UtD6stYkEEqyAIicTHrOaRYCVbtlMQBCHPyIlg1Vo/oLX+B7Czm03PAu7UWr+utd4FLAS+kfaJnIK1zEqIIIJVEIRUxNuFPBp0lTXbKQiCkGfkewzrNOAVx/wrwH5KqeFuGyul6qzustXbt2+PCQmIeFidDx8RrIIgWKy75k+snHM1r9c/E7sivzys6dI32ykIgpAmwSAsWmS++5N8z8M6ENjtmLenB+HiYdBa1wP1ADNmzNAfPv4y+wF7H13OwC++kHh0EaxClli7di333Xcfp59+OpMmTcp1c4Q43rryHib+7JscDLQ9WWkWejzmBbcwBWuvbeeUKTP0okVSKEAQhO7JZq7mfPew7gWcuZDs6ebudmzdvpf9Hr4bgKr1jYRPOiVxIxGsQhbQWnP22Wdz4403cs4556AlnVre4fnrXwCTp7kMq2em3BqjVJiCtde2c+1auOoq8xDqb4+JIAiFTTZzNee7YH0dONQxfyjwoda6u/gtupqizgUFEOpI3MiOcRWEfuTOO+/k3Xff5eWXX2b9+vX8/ve/z3WThDiqphwQme7ESr/mN7mb8ymGtQf02nbaSRKkUIAgCN2RzVzNuUprVaaUqgC8gFcpVaGUcgtPuBc4Wyk1VSk1BLgSuCedc5RVDwBMoQAN4PNDfJog8bAK/cyOHTu4/PLLufvuuznooIO4++67ufTSS2lqasp10wQHoz6xX2T63cXG2xoRrHnkYc2G7VRKCgUIgpAe2czVnCsP65VAKybtytet6SuVUmOVUnuVUmMBtNb/Aa4DngI2ARuBq9M5QUW1edjsG1DDh3PPxbP8qWimAButC9V7IhQIdnnVk046CYCTTz6ZDz/8kGHDhuW4ZUIMjuwhU04/3EzYglXrfKqK1++2c9IkKRQgCEL6ZCtXc04GXWmtrwGuSbJ6YNy2NwI39vgkllekKnAoVQ/eZpaVlSWGAXR2Rh9MKQgGTfeYDEQQhCKkvT06bfe8+HyxA6/iX3hzQDZsZ1WVefhkArGbgiBkitxb4P7C7sYbMiS6zE5t5SQNwZrNUXCCIOQAN8Hq9ZpPHgnWrPDaazB6tJkuL4fqati9O+qFdi5TCg4/HBYsSDCKYjcFQcgk+T7oqvfYD510BGs3ZHMUnFAcnHvuuSil2Lx5c8K6tWvX4vf7ueCCC3LQMsEVN8FaVmY8rFBaoUPt7bB1q/ls3AiNjbBpk/uyjRvhH/+A2bMTUgqI3RQEIZMUr2C1PaxDh0aXuQnWNDIFZHMUnFAcBCxX0vPPP5+w7vvf/z7V1dX8+MdSKTNvcApW23aUlUVtRh4NvMpLQqEERSp2UxCETFK0fVx68xYAdj73FpHSLm5deml4WO1RcBKLlcfkWbDczJkzASNY586dG1n+yCOP8Oijj3Lrrbcy1PkyJeSWVCEBUHKC1R5iplJu5cDnS1CkYjcFQcgkRStYlTZdeMNW/pP1l9YzYXFdr0MCwBhbMbj9QHyqsXyhj6PCJ06cyLBhw2I8rKFQiIsuuohDDjmE+fPn97WFQiZxZAmICQkoQcHaRgWbGYICho6qoHLUENi1KyrqKypMqNX27fDBB2b+ySddDaTYTUEQMkXRClYn6oEl0EfBKgg9QSnFzJkzeeaZZ9Bao5Ti5ptvZt26dTzxxBN43a5FIWc0v/0hg+wZZ0hACcawvs40xrAarxcWXpAiY8AHH8CYMTBsmKhSQShRstm5WbwxrES7tfRp88yECNb8w85x2ZfPqlVQWWn+v5WVZr6vx8wAM2fOZPfu3axdu5Zt27axcOFC5s6dy5w5czJyfCEzNNYHGbBmdWR+/QMvm4kSDQlIu3CAHWIlFQMFoSSxM4Fkq5Rz0XpYw/4Kdgw5iD3fuNCEA0CvY1iFPCdPg+WcA69WrFhBe3s7v/jFL3LcKiGenUsa8BD1oDY/+YKZKNGQgEmT4Mwz07iVfFYJW7GhglCSuGUC6c/Hb9EKVs/0aYxYvZoRzoXiYS1e8jBY7qijjsLj8fC73/2OZ555hksuuYSDDjoo180S4hg+rxb9mEJZfTJDZk6B1ZSsYE27cIAtWMXDKggliZ0JxM613N+ZQIo6JCCBXqa1EoTeUF1dzdSpU1m5ciUjR47khz/8Ya6bJLgwvS5Ax6ixkfnxxx9oJpwxrCUkWNNGQgIEoaSxOzezVcq5aD2srriEBOyYdSrvHf0ldjOE4fNqmV6XX146obA56qijeO2111i0aBGDBg3qfgchJ1RUOt7d29rMtzOGtYQGXXVHMAj33guesI9bgXCok+eCedfBIQhCFshm52ZpCVYXD2vN3ncZ/th1aBRtj1XQyDIRrUJGCIVCNDQ0MGPGDM4666xcN0dIhTMPa2ur+S7RkIBUBIOm289kAfNyK+AJdzHnBM2yJ5WIVkEQ+g0JCcAkx/ag8dHBziUNWW2SULzccMMNbNiwgV//+teofM03W4K8uegBXp5xNo31jiGtTsFqe1hFsCbQ0OCMAFCELJ9HuKNTSq8KQpETDMKiRf2fDSAZJethDWOEqlNGdOJn+LzaLDdKKCaamppYunQpr776Ktdffz0XXXRRpOqVkHsa64NMv8KkuWub/ycaecr0qLgJVq9XYljjqK01Y63sOgudlOGjkyp/iNpaX07bJghC/2GnsLIHWGUjZjWe0hKsjhjWlikzaNrWybidL0eWbfjFEgkHEPrE0qVL+epXv8rIkSP5/ve/z7XXXpvrJgkOnD0oPkJmvi4QFang7mGVGFbAPKAaGkwMK4D3Hh+0tfHoQyGOEtMpCEVLtlNYuVFagtXhYR00djiDrv8ufO5zkWVTv3JoLlolFBGnn346p59+eq6bISRh+LxaeMxMh/GY+XA4Nr2dhASkJGaQxV990AZHHSHpAQWhmMl2Cis3SjeG1euFUaNi10t6FkEoapw9KB+d/OXEcAAQwZomwSC0tEtqK0EoBbKdwsqN0hKszrRWHg/v3v1UzOqNP/tDlhskCEKuGDGh2kzEC1Y7S4DEsCbFjmfb1WLiVl98VgSrIBQ7gYApKpKrbCClJVjjPKzhpY/jrBrvWfpo1pskCEKO2LvXfDvjV4H22+82E8Eg7NtnpiWGNQY7nq3Tiip77hkJCRCEUiYbGQRKWrDq08xo4YhoPf74rDdJEIQsoqOvqKE//4X2mv2hvj5mk/JOI1L1mjXw5ptmoXhYY7Dj2UIYD2tghnhYBaEUcBOmdo/LVVeZ7/4SraUrWD0eJiyu450Ft9NebroGDzjv85HVjfVB/vup89h66nm5SzomCEJmccRa+ro6KN+5BX311a6bKogKXBGsMdjxbENHGMF6+PSohzXXuRoFQegfkglTtwwC/UFpZQlwxrBa4nXC4jp45l545pnIw6yxPsjk+bMowzykwo/cjWf5U1J7sI9orSWBfg/QWne/keBOMMiW6+5ly2bwnX1mdLBVfLxqEuxfXillRKsI1gQCAWBUGWwnYjvzIVejIAh9Ixg0orO2Nvb+TZbaKlsZBEpLsMZnCbDx+823lQ1755IGfEQfUDqUo6RjRYTX6yUUCuG3f2uhWzo7OykrK61bNCMEg4SPq2VUZwejgPbn744WCLDiVSOC1GX3LjzsPexYBs+cCs8/D//9rwjWZPisYgGWYM2HXI2CIPSeVC+dyYSp3ePiJnIzSU6ehkqpYcCdwMnADuByrfWfXbYrB24GTgV8wDPAuVrrD3p14riQgAhxgtWZq1ED+HKUdKyIGDRoEHv27KGmpibXTSkYmpubqaioyHUzCo+GBlRnR0SMRkouOwRr56Ah7KOKwc0foKqqoKUlsrt33AEMfmm5mTnpJPOdJ4OucmY7k2G/UFl5bPMhV6MgCGmyfLnp33/rLTNfVcW0D1vY2NpBCD97W6sYdXIL+I02Cvj97BhWRefuFipUB/7PYW70qioCLS0EOjrghugyWlqixmDkSA6EsX1pbq7cN7cCHcB+wGHAI0qpV7TWr8dtdyEQAD4B7AbqgV8Dp/XqrC4hAUDUS2AJ1ul1AZhvLRo5hvJ//FXcBH1k2LBhbNq0CYDq6mp8Pp+EByRBa01rays7duxg7Ng+3d+lSW0teLwQNl7RTnzRksuWYPXtV8PgO++E2bNjxCpAa5ePSnsm//Kw5sZ2JiPOw5otT4sgCH0kGDQDzeNCz6rjt9sbOzugt+fbupVhMKK3u0MOBKtSqgqYBxyitd4LPK2Uegg4A7gsbvMDgaVa6w+tff8C3Njrk6cZEuD0ppRPO1isbgYoLy9n7NixNDU18e6779KVPwIgLykvL2e//fYTD2tvCARQc06Axx8HYMO1f43GsNoprCoqYORI1939779DY33Q7JNHeVhzajuTEedhhbhKWIIg5CcNDQliNd/JhYd1ItCptV7nWPYKMNtl2zuBm5VS+wMfAV8Dep8stTvBao8gdlZtWbuWvRMPp2NnM7u+eRETbjg/rVM11gfZueQphs87Pqa6TilTXl7O6NGjGT16dK6bIhQTbgOsbM8fMHnetOi2TsG6cWOSA+poCEF+eVhzZzuTEedhjSfZ4I1U9GYfQRB6iCNeJ1VMfzJSSd3+6jvNhWAdCOyJW7YbGOSy7VvAe8AHQBfQCHwn2YGVUnVAHeDelZpmDGuM8d28mYFsBmDoL77Nem+ZySyQgsb6IJPmz8ZHiLbHKmjkSRGtgtAfBIOEZx3HqK7O2AFWu3dHt3EWBnAK1hdfRBNrXDXQRVk0hMC2GfkRw5o725mMFIK1NxkDJMuAIGSJQIB946YwYOMaPmA0IcoZcfAQBnbsMtlUKipgyBDYtSuaXcWxrLmpneZmaKOC3QxhMLuopJ3hw6C8Om7figoYO5amFSu296XJuRCse0kMk6gGml22vRUoB4YDLcACjJfgU24H1lrXY2K1mDFjRuILQLIY1lSCNQ71wBLoRrDuXNKAH3OMctqj3hpBEDJLQwOqqzNxgNVHH0W3SSZYjz+esLcMT1e0OzusPLx3yS3RF8z88rDmznYmwyUkwKY3GQMky4AgZI89oQoGAF/gX7ziPZKF3zSlV9PhdcfLZVkZfPObcOaZsH+K+3WDUpv60t5cFA5YB5QppQ52LDsUiB80AGZQwT1a6yatdTtm0MBRSqneDTXfHhX3Lf9ZEc16GzfoKvLtgl0dKxUR7wzQhTdmXhCEDFJbi/ZEXz478Zv7LR0PayCAd+UKdh43lx0jptB03Fy8zzwd24OSX4I1d7aTJAUBUnhY7YwBXm/6GQN6s48gCL1j0ABj17SnrMf3mz3AcuFCeOopuO22/n+5zLqHVWvdopR6APiJUur/MIb1C8DRLpu/AJyplGoA9gHnA5u11jt6fOJgkPCD/4go9AHvvEZ49vGmIECaHtaPjjq523AAiM0y8NFJX5JwAEHoLwIBPJ//HPzznwBsWHS/ud8udnhYW1vNd0cHey79KdVA+7P/pTwYhECAmuUPJj9+Hg26ypntJEVXfQoPa28yBkiWAUHIHlV+c9+e990yDvly6vvNLbY82wMsc5XW6nzgLmAbsBM4T2v9ulJqFvCo1nqgtd0PgF9h4rH8wGuYvII9p6EhJg5N4SgIkGrQlYOhR03s8WlHHPaxHu8jCEJqzKDGBobPq2W6I7fv5JodMHMm7HXkYnn5ZZp+/Cv8LzxDdccuAPw7NhM+bjaeFctTWtxdG5oYCrRf8APKFy+GUIiDYUJ//V1pkH3bSYquejv04uqr4bLLEmLdArt2EWhvNy2x49/a22HSJFiwwPW3lywDgtB/xAhP60XznPPKYJLL+oCZv/deuOsuc//3JB490y+eORGsWusmYK7L8pWYgQX2/E7M6Na+U1uL9vmMSLWxCwI88oiZ7y6Gdd++np83U7lGrZHQe9ZuZfCkUYxacKZYdaEkWXf1H5n6k7PQQMdj5TQFahlmrzznnITt9YIF0fUWCqAzlDJIsrGqTMvKAAAgAElEQVQ+yJTnnwCgfOcW2LkFgGoYkoE/o1fkxHaSpCBAMAhPmN+HdetS7O3CmjXG7i5P/cKQKSTzgCAk9pQ0DeuiAiKhT/Hrb7oJvvc9E0llZ8BKJ7a8vwZPlk7dx0AA7/IGd9Fn5WvsNoY1V4I1biQ0ayD8yN0mnEGsr1BijLnhe3gxvSWaDpo3bE65ffwdaI8o0mU+VIqgrZ1LGvCkTN5SOrh21S9q6Fv2hFDqF4ZMIZkHBMEQ31PS1tJpBKsV2hO/fskS822LVaXSiy3vr8GTpSNYAQIBRj8YICELaPygq3zzsMaNhAZHOINYXqHE8BN9oQzhp3zMcNgau02qvIIt/qG0z5zN8Gvdu6Rths+rJfRYGX6i8ZmlXJstoave6XbtDT5fVkZVSeYBQTDE95RU+oxtu+W3ZRz5hcT18+bBypWJmQC6u3/6q0RzaQnWZFgxrB/dfj879UFMmHeY+3YOwfrW1X9kR0MjA782N/WgqkwI1tpaUB7QtleJaDiDIJQYZRU+M4wIWH/7Mva/74aEbfYdMJGOvSGG7tqQsG7gjQsZ+O1vd3ue6XUBGlmB/6brGNO6lqoh5dDezp41az7qdudSIBCg8dcNhO68lwkdbzB410bXfI0xy7xeWL+elgE1vPPLh5ieBeXYXw9PQSg04ntK1OeMYL32ei9NvzLr4ntSpk/veThNvw2e1FoX5efII4/U6bJtzpe1Nl5vHQb93tcWROZjPoGA1qtW6a3jZkS2baFSv3r7qsSD2vtccUXa7UjJZz4TOWbzxw/VepXLOQWhFBg2LHp/aa3XTf584r3a0aH1xRe738f19X06PbBa54GN669PurZz1SqtKyu19nrNt9MkrVql9c9/nmim/nHJSq1Br+SYhH36k2TtEYRSpmVAjdaga9imvV5zj/QnfbWd4mEFOtbGemE8DU/GzEeq4WzaRPjYWYwMmxQ3ijQKA2Rq0JVjJPTAuSdJn5YgWNQMjU2p1FVZhdfng8pK9x3srCBCn0jW1Z4sZjQYhMU3+vkCJqyjvT173fOSeUAQEvF7je3sTR7WXJCLwgF5R9tXz44ZWjFwnwmIS4iD27ULFe5KiGPLSmEAZx7I/CgTKQh5wdCBsYLVO8waxF9R4b5DeXk/t6g0cCb593ph06boaPx4IQvmuy1sXhb8dOD15v8DUhCKmTJtbOflV3oLYjCiCFZgwuI6thx0DGDEafWu9wEIo9BEhWtMOVeL14//TuoYVk+GfmKnSNUyclkoYeJ7LeIHSQ4ebL6TCVbxsGYEO07tnHPMv+SOO4xndfhw92pVtbVEfns/HdxyS/4/IAWhqLEcYRdfWpZwL7pWtssxIlgt2lRlQgKb0KixvHTUuWz7zDfMguZmFNClPHR6TGaBT9xxgVlXX8+eqZ9i7bRTaax3/IczFRIgHlZBcCe+ytKWLRAMsn3pavftRbBmjEAAxo41/wLbo7pzZ7Rk47JlZrtFi8z3nfcau3nQxzqocxQNzMeHoyAUPZbtXPyLsph7zw7rueoq850v96XEsFroef8L1z0Rs6xiykEc8eRtvPabFYz89z2RUACPtn2vmEovv/kN+tvfZhAwCAjNf8Rx4Ax5Q8XDKgiuNO/qZJBjXu/ahZo9Gz10YjT+3ImEBGQUt1H4dsxofDzryj+al4UKb9QrLnlSBSE36E6TLvOqa7yULYree/maCk48rBYTFtfxzg9+E7tw1CgAdvzzmRjvq4KoaPzoI7j3XpS1XAFlOLooM1WD3ClYxcMqCBH2fmTut5iY81AIz+SDI8tjXvHEw5pR7NAA26PqfLDFP/ieecH67R25W5PFvAqC0I+EwyhLx4TCnph7zxmfnk+DsUSwOphw/XmxI4utggLD59USoizy4HM+/N45dzHbhhwcc5xOfNGZZEUIeopT+IqHVRAiVA+IhgRE7k+fj5prF/DOgtvZ/LGjaJ7yyegOIlgzTiAAl1+e6IWJf/AFZpvfPrSvIxICkK8PR0EoaixNEaIMr1fF3HupXkJziYQExFNdDa2tZtoSrAkJxMeOQK9cidKaA99+nNDbvphDbPjpn5h05ZfMTHx8XW8RD6sguFLlN/fYyhN+xGFb/kP1pP1hgaliNSEQgMV18K9/wf/7f2YHCQnIGvEJxD95iBGsbXs6uOqqaAhAvyQZFwQhOZY28fq9LLwm8d7Lx1RwIljjGTwYPvzQTPuiQnR6XQDqHjQzixbBihWAHQIQK0onnTAmOpMpwSoeVkFwx+rFOO7W02Hyj923cYpU8bBmlZgHX5uxqX46YkIA3LyzgiD0I5Y28fjKuPzyHLclTSQkIJ7q6uh0sgdbbS1hX3RdZ7zu37YtOp2pkADxsAqCIT7zhv1SWJbi/dt5L4uHNXdYTgAfIbweLSEAgpArbCdYKruZZ4hgjccpWH0+920CAbzLGwhbeVm3zL8mdr1TsIqHVRD6F/seS3a/gnhY8wWryoAHzU9/3JVX8XGCUFLYdtMlv3y+IoI1HofA3PvQsuQJyAIBPAMHAjBu1rjYddu3ux7Pjcb6IP/95Hy2nnpe6mRn4mEVBHfsXoxUngIRrFkhrXyq1u9/2UUdfRKrkrtVEPpAOj1TeUbhtDQbBIPolU9H8jZWvfUy4dnH41n+lLsbwP5H24O0LHYse4UaeyZFSEBjfZDJ82dRRheshvAjdyc/lxQOEAR3euphlZCAfiHtfKp+v7GZHR0wYEBk354MupLcrYLQRwpQsIqH1UlDA9rR3W7yOaZIDGg/IOME65Anl0RnUnhYdy5poIyuSP7WlOeSwgGC4E5PY1jFw9ovpJ1P1babVi7W3lTVkdytgtBHJIa1wKmtRft8sflWfSlGBSTxsHpweENTCNbh82oJO+vwpDqXhAQIgjsSEpAXpJ1P1f79rf9bb8Sn5G4VhD4iMawFjjWYauvcc1k3ZS4fzj03eRc9RD0FbW0xi7XzZ00hWKfXBWgfNykyn/JcMuhKENxJJyTAIVIb71ndzw0qTdJONm7/LywPa2/EZ74mNheEgqEHIQH5Ei9eOL7gbBEIMPrBAKPT2TaJh7V93MEM2LgWgD3rt1Edv5+DAcMqYWP03EkRD6tQ6gSDbFt4O0N3NUdqyTWcsojjOkLmFTGF4V3zx9VMsaYnzD+RRpaZ3MpCRkkr2XicYI0vLgDm4dhdPGs+JjYXhHzFGScO0PiXLuqgW8GaT/HiIlj7QhIPq7dtX2S66rXnaKwPJn84Oj2nqRAPq1DKBIOEj53FyHDs/TLrsSvxYL3ApTC825a+HBGsPjrYuaQBRLDmhjjBClHxmU8PR0EoFpz3VVmZkRBTOzupA/a1exmQYl+3kJ1c3ZMSEtAX4j2sHvNz+rZ9ENlEETYPx2Skm6dVPKxCKdPQgAonvtx5LbEaViqxoICDYf87h31UEsJLCD/D59X2V0uF7rBf9F0yqMhgKkHIPPH3VSgEKmy0x57W1H7LfIoXFw9rX4j3sFZVQXMzHh0VlBpP6odjuoJVPKxCKdOdlfSmNmXT6wI0soydSxoYPq9WwgFySXu7+T7xRPMErKqClhbo6OCCTj//21VFFS34ujqoXgTcbG0zaBCcfz7U1eW0+YJQaNii0/awdnWBz9IeAwentp3xITu57PHImWBVSg0D7gROBnYAl2ut/5xk2yOAm4AjgBbg51rrm7PV1qTEe1gtwepkb8341A/HdEMCxMMqlDKBAGrUKNi61XW1dpRKTsb0ukDBhwEUvN0MBuGNN8z0rl0Jq6uAjzsXNFsfm/nzzbeIVkFIG6foHD4cLrgAvJ1Ge+g0sgTkS7x4LkMCbgU6gP2ArwG3KaWmxW+klKoB/gPcDgzH2LPHstjO5Lh5WONQHe2pjyExrIKQwBu/XMr6g09m22fOig5NTWFYO3SKDAHFRWHbzUz08S9Z0v02glCiJBvRHwjA5ZfDzp2mY9eL8bB+tLdwOtpz0lKlVBUwDzhEa70XeFop9RBwBnBZ3OYXAUu11n+y5tuBNVlrbCrcPKxxVOzdwe7DZzN45lQ488zE1xRnSEA4HImDTUA8rEKJ0FgfZOpFn8aLRr8N4Sf+YlK+paga5wu1GAudD26AfqIo7GZtrbGbLqFQyV7DEyKT583LcKMEoThIZ9CiHR7gb+sEDd7ywhGsufKwTgQ6tdbrHMteARI8BcBMoEkptUoptU0p9S+l1Fi3gyql6pRSq5VSq7dv394PzY7DFqwpPKy+cDvVL69A//a3cPzxia89jpGybt5W+22ptUVKswqlwc4lDXgs+RJTAW7fvqT7eLtChI+bnftEgf1Lv9hNyKLtDARgxQqYOxfGjYPx4+Gww2DcONSoUbSNGs+mYYexgXFsZhRbGEXrqPFQUWH2/+EP0w4HyJfckYKQLdIZtBgIwE03QZkymmLt296CuUdyJVgHAnvilu0GBrlsOwY4C7gQGAtsAO5zO6jWul5rPUNrPWPEiBEZbG4S4kuzxglWDZGyqwrcr6B2R8hAnBB1lizc/L6UZhVKg4RBij4/zJ5tBuYkQQGqM1Tsw8r7xW5Clm1nIAAPPgjvvgsbNsBLL5npLVuo3LKBP//gJSZ632UMWxjr3cJNF2yAww83+37mM2mdojflXgWh0KmthTpVTyNT2dQ1motuGA0HHmjun/HjYfRoGD2aUy8+kFvDJh58ariRt+4tjBskV77gvZCQT7+a2PB6m1bgQa31CwBKqR8DO5RSg7XWu/u3md2QwsMa9nhQ4XBEtGpAueWESOFhdb4tKSQkQCgNptcFYH503rP8KTj00KQvavZSXeZDFXeNzuKwm93gHNEcMZmPWx7WuCItycin3JGCkC0Cz/6SmZ0XReZVE9AUu40GaqwPwEi28/XfzYYzl+f9TZIrD+s6oEwpdbBj2aHA6y7bvkpseFP+uBfjPawDoul3PYMG8c6C29kz+AAA1CGHwFMupVdTCFZn/jPbfQ+Ih1UoHcrKzD2TwruqgS1DpuJZkf8Gt48Uh93sBteyq3ZIQFyRlmTkU+5IQcgav/1tbK+uC/HrFeApkN6pnHhYtdYtSqkHgJ8opf4POAz4AnC0y+Z3A0uUUr/CGOargKfzwkuQKoa1qooJi+tgbAi+8x2YNSvxYdrVFStS4wSrMxXFfjeFYZu1QjysQqlgC5W9e11Xd6Fop4Kdi3/H/sUtVovHbqZBQhqdykrznaZgzafckYKQNSZMgHXrUm7i9uaqfL6CeKvL5fCw84G7MDJsJ3Ce1vp1pdQs4FGt9UAArfWTSqkrgEeAAcDTwFdz1OZYUsWw2t7WFFVdEpa5DLqKGO5fi4dVKBGcI8jtl8IkHtY3DvsanHd+KRUCKHy72Rt66GGF/MkdKQhZY/p0ePRRGDEimgawogKGDDF5j9vbUUArFXzEEKor2qk6bBIsWFAQN0vOBKvWugmY67J8JWZwgXPZbcBtWWpa+nTjYQUS6mYHg463/mmOcABInZNV0loJpYLTm9rVBXffTeuVP6XSZdPpP/0KfDb/DW2mKAq72Qu27algJLD+9TYm5LoxgpCv2E6wSy+Fiy+O1RsOM1lpfQqNwknAlY/00MManyNt+ZIOPuk8XiohKoUDhFLBKVjb29Hf+lZy41omJqzYCQbh1UcrmA/cckMbX/p8QTiDBCH72GNi/P60crIWGrmsdFX42A9LW0C6CVaHhzV+5Oqzy+OqYImHVSgB1l9azzsHn8L6S+vdN3AKVpcE8zHY95dQtDQ0wL6wCQko62wrhLEhgtAvdJtb2Paw+nxp5WQtNMQ90Rd8ceUg3UICHB7W+HQtx3yyByEB4mEVioD1l9zGQTecb2aue4z1YAYnOnEKVsfLmX3Vx4x+jb8HhaKjthZWeCuhEwZ6WwthbIggZJx0PKbb3u8woTObfNR+3iU9XIGTlodVKfVbpZRWSu3vsm6SUqrDGo1aWsR3R1ZGOy5bV79mrjCHhzU+XcsRh0gMq1BaVNx/T2xKlQcS68K/c9+zMfP2tjtGTKGjJtYErX8499VGhcyQqgb66d80Htazv9ZW8N2agtAbuvOYBoOwbKnxsP78BqM7EtLDUdgV4NINCbD/tKNc1v0SU33l6oy0qJCI8+5sWr4hMl2x+R3Cs4+Ht94yCyxXfSAAl19uXTwvvBB7PBGsQpGjjo7NwKRPS6wL3/Lk84k7+v2M2PYG+w46JGbxxxZfQGN9AVpeIYbuKlONnWgE65iatph9CvXBKwg9pbvcwg0N4A0bndHa6YsUy7D1RjAI551n9ivUCnDpClbb5REjWJVSnwU+DfxIa70rkw0rCOI8rHuefzNSjypSA/11K6d3R5w3NRgkfNY3Yhatve/F5OeSkAChCNj/rJMi09v/54zEcABg6GHjEnf0+2msDzLo+SdiFpfRwc4lDZluppBluo23i0trJaVXhVLDtaCGg9paqFBGZ2ifP0bQ2vfL7bcXdlxrujGs6zAFviKCVSnlA24EXgNuz3zTCoA4D+vAYz5Bxyt/oxxLnPr8MGMG3HFHYs7VhgZUONajuvs/z8JVX3Y/l3hYhWKgPTrQcOTnZ7puMubQ4YkLfT52LmnAQ2xMaxgvw+fVZriRQrZxLcfqxBasf/wjLFnC5PYKnmkdQjW7qGhtZ9ApwPDYfJPx+ScZNgwuvBDqEl+SBKEQSJVbOBCAXTNDsAqu/qmPyY7t7BdC29elVGHGtaYlWLXWWin1LHCMUkpprTVwITAROFFrnaIvu4iJ87CO/+whNB7aQOjOe9l/fxi14MzoNg4Pa2N9kOrbH2GsNa8xHtkhJx4ZWR9zjEBAPKxC4aE1W+fORz33LHvP+jYTFs+P7Wlob3ffb5dLZ43fz/B5tXQ85o+8EIaVl/cuuaWUigYULd1WprJcqHr3bti9myGYMl8Rms3HaRkTSlNu3Qrz55tpEa1CnpMsh2oqhg40jrHJ02Odac4XwrIy+OY34cwzCy/NVU+yBDwLfAaYpJRqwpT6+4fWelm/tKwQiB+h/NhjTL/pJnA+QF95xXxbHtbG+iBT5h+LlzCKWAM78bTpNNYHmTR/Nj7M9uFH7saz/CnxsAoFx/tn/ZAxD90BwMjrzmU9igmHODKqJhOsTU2Jy/x+ptcFaCT2hXBCoVlcIUL8AzllZarGRiB5fXSb7tYDsGSJCFYhr+l1DlVHHlYnxVKquCeC1Tnw6jigHLg44y0qIHb85wVqHPP65ptRU6fGGsO40qy7/vp4RKxCnIHt6mLnkgbKCEWW61AHPPVUrFdVPKxCPrNiBU2XXsvgF1bFLFYPLIGJX4ou6KGHFTDeVPGoFjw9fiD/3/+h4wepEms/XWukux1rXuJAP0HIF4JBuOYaYx7D4WisaVoi05GHNZ5iKFXcE8H6PBAG/g84Brhea/1Ov7SqQOhaszZxYfzbe1xp1v2On4paFtd1pZQRoV1dJh7vseg6jWbnf54nJqpPPKxCvhIMEj7hBIa5ZLzQp82LFanJ6sKnEKxCceA2yCrlw7SujnfWQ/t1NzGYXXRQwYiDhzCwY1fkmmqjgje3DmEwu6igHd/ACipHW9ts3Wq2O/ts8a4KeYv9ImeLVY+nh7GmSTysxULala601nuAN4BZwDbgZ/3VqEJh7xnnoyHyARLf3uM8rJOON3kk28qHsG/8FNTcuTB5stnmiScYf+N3Y3b3AsNW/jP2mOJhFfKVhgaUi1jdfvLXTEaAZDGsCxfCuHEwahSdj7tEGRWpAS5VukvR48aExXXsXvUG9/58C1tXbWDgupfg3XdhyxbYsoWXH9hA/bkvcfHcdznQv4XRrRsY+f5LBO97F774RXOQ2bP77W8ShL5iv8jZYvXEE3tWUrXlI6MzXl1TnAVVelrp6nngEOByrXVzP7SnoJhw/Xms93ipvusmBgxUVF3uMgI1zsPK1q0AVJ5yHPzTEqL21XjFFQxyOU9Ct5Z4WIV8pbYW7fEmZMAY+ZlPmgk3wXrttfCjH0UW20bJHowISEWrIqO3MXXJujWdIQYej/HcxnSner1mw1S5rgUhx8Rny7jmmvTvjWAQqt8OMQ04+1wfv5pU+CEA8aTtYbXSWNUCq4Hf91eDCo0Ji+sYsf0Nqja87t7VFOdh3f7bvwGwe5Oj29M2pnHouO/oCvGwCnnA0qU0T57Bnur9adtvLEybBo2NeL7oEiPY2mq+3QTrksRqVwmIh7XoiCmikgapCgU4Qwy6uoxojfHeimAV8gy367m7XKupaGgAnzb2tSXkL7gcq+nQEw/rD4ADga9Zaa2EdHB4WNdfejsTHr8fgOqXV7L+0nrTTRonWKNCVaHQhH1+vCHHg148rEKOef/rl/KxP10X7RFoBr0N1Pz57t2uqQTrpz4Fq1cD7gNnABGsJU53g7TiPVM33QQ7dzq8t3eJYBXyg2AQ7r0X7rrLXI7x13NvB0fV1oJfhYwR9fkKLsdqOqQUrEqpYcApwCeAS4AbtdbPptpHiMPhYfXff2/MKvXAEnARrDtrJqGmTmH4NRfACSfEilUQwSrklA3fu5nxf7ouIVQlMv/224k72YLVGbdqTwcCcOutMGgQavhw9vqGsLvVz7Cu7VRuscodi2AtabobpNVtiIGdD7uzMyvtFQQ37BevtrZoR2mPsgCkIBCAjuEdsAP+8Bc/RxZZOAB072E9BfgzZpDVL4HL+r1FxYbDw+r91AzYtCrqQbXrqMcJ1ppNL0GlyVfZWVlFWWtL7DHFwS3kkAH33+maTiiy7IAD4IMPYneyMwK4eVibrXD400+H229nIDAQ4LjjQASrQBqVsEj0TMXkeZWQACEP6O+KU35lQg+PnFmcMf8pBavW+j7gviy1pTjxes1VqTX7nzgN/gZ7B41m23nXROuoe+JCia2Hc2N9kGmt+xKPKR5WIYd4PnEIPN4YE2Md9pVTFmqHiy82o7afje2IeTnYSmsQAk7BaovYPXvM96C4IYfOSnIiWEuang7Sig8hWP85L6NBBKuQU3pacarH1a5S5GEtBtIedCX0AevtPnTRJeZ72qFRsepYH5m25ncuaUC5RfWJh1XIISNqDwGgedBomo6bi2fVKspmzzIrTz45ajQdvP5iK3PmwIfvpfCwVlfH7iSCVbDo6YM7PoTgvc3iYRVyj3NQ1VNPwW23pRarc+bAVVeZb7fBhglIHlahTwSDkbipshbjSRr67H9Yf2l9dBunYC0vj0wOn1dL2O1fJB5WIZd89BEA1T+8kOHLHzQW1wphobXVVbBW6FY6OmDb+y6CNZmH1eklKFKPgdA9vXlwx+d5HTNOBKuQH6SbHcMtbrtbxMMq9AnHVeaM+1MPOFL5JBGs0+sCtEx3uarFwyrkEkuwMmRIdFk3gnUArfj9MHqYy6Ar8bAKKejNgzs+PdD+B4hgFQqLHhfX0LroBWtPCwcIPaW2FrxedJyhjAy4gqSCFaB62gHQGHdM8bAKucQundoDwTr1wFaW/Qlqro96WDteXUPHxMMZ+ME6s+C550zpTBun0RXBWrKkM+DKjZhBWA+LYBVyTDAI110Ha9ea5/wuq6xwRYWxpbuiZYapqCAwZAg7B+8i1NJOuQ/Kv2q2a93qWFbtsi8YW1psVQMQwdr/BAKwciXquutoffYl9oar2PONC2NjWJ2DruIfzAMHJh5TPKxCLumFh3XcyFbGBYjJEuDrasP31svRilZ33AEzZkQLcIiHVaD3VbFikLRWQi4JBmHWrB6/MFVaHwCazADXCusDoJtcKmGCyYW9fHnRidachQQopYYppR5USrUopTYqpb7azfZ+pdQapdT72WpjxggE4MEHqdzyLiM+fD1WrEJKDytVVYnHEw+rkEP2vWVuwfXLHbdiN4KV555j96GzYNu2yCLl+ERwVr2yhTGw95GGNEcdFD8lZTsteloVy8auJiSDroSc0tCQkWtPuXxcCYXYeG9D0spwhUouPay3Ah3AfsBhwCNKqVe01q8n2f4SYDswKMn6wiWFYN227iNGWtMRT9Q+l1RXgpAFGuuDTNu4FoD9F32HxvFTmV4XMN1aYFJVuQlWoPrVpyPXcHwfQcTwzrNCZYJB9GOPR5ZXvf0K4dnH41n+VNF5DXqB2M40cKa2aldergERrEJuqK2NVK502r6kgpMUVf/icDtGuMzHWXfV8rRLJa1CJieCVSlVBcwDDtFa7wWeVko9BJyBS3ECpdSBwNeBi4A7stnWrJBCsDa/vTUiWCMXpp2/UhCyjDPVWhkhdi5pgLrUWQJskRpvWHcMHE/l6CEM7NhlehIuvDAaDtDQgNY6so8CdChDJWEKGLGd6RMzWEuJh1XIIYEA7487mgM2PsMmxtBEDeOH7GJoRfIY1q1tJl7Vj1nWQQW7GcJgdjG0sp0BlS4xrBUVcNhh/HHUAp6+I5C0MlyhkisP60SgU2u9zrHsFcClCDkAvwauAFpTHVQpVQfUAYwdOzYDzcwSKQRr5RFT4a2lgMPDKvF8Qo4YPvdYeMxciyH8DJ9Xa1Z0ExIQuXYdNI+ayIh1S91PVFuL9vmMSLXxZbAkTOEitjNNnIO1UF7oRASrkDMG7l8NG+G7nt/yRPlnWfbv1CLyXUcPgcdjLt1w2MiFhVeZEJlkHBwE/+97PlAx38lVDOtAYE/cst24dFkppU4FvFrrB7s7qNa6Xms9Q2s9Y8SIEZlpaTZIIVj3P3FqZHrHiV8xEyJYhRwx/YuTUUBnWSXrb19mwgEgpWDd/OXv89HgcWiIfCAuU0Y8gQDe5Q1snXsu66bM5cO550o4gEFsZ5o4U1vVnSceViG3DB1oBvx99cyytLrondfvLbcYaZBuiqv4tG7FYjZz5WHdC8QlXaQaaHYusLq/rgM+k6V25YZUWQJ2745M7jzxK4x44n4ZdCXkjh07APAddEBUrEJKwfqx75wG998I9fW0/Pwm9rWqxEwZbgQCjH4wYEpqCjZiO3uB9ohgFXKMZRe/coYP0hSQztRs04rwq+QAACAASURBVKf3LFNGTFq3IiFXgnUdUKaUOlhr/Za17FAgftDAwcB4YKVSpjMcGKyU2grM1Fq/m53m9jMpPKybX/iA/a3pcZdZHlatIzndWl5ey562cgYO8zPowrOjMYCC0B9YgpWamtjltmD9+98Jt7TEdt38+99w7LFQV0dVXR1VQHH48HKC2M40cQ662uYp45cQk9aqx3XaBaEv2Ndembvssq/H4cNh587E67IYBWhPyYlg1Vq3KKUeAH6ilPo/zEjXLwBHx236GnCAY/5o4BbgCMyo1+IghWB9e8dgRmFiN3xYnqu2Npg1C93VxQBgAMBW0POfN3GCIlqFfmLtMzuYBOz7YBcDgsGoBV2xwnzv3ZsQZ6QXLUKNHy/XZQYQ25k+zkFX7eFYD6tTzBbTKGohf2ne1ckgoPHNMqYfF7vOvh7b200HqsdjpIBcl7HksjTr+ZicuNuA+4DztNavK6VmKaX2AmitO7XWW+0P0ASErfni6dtJIViHfulk2qgkhJcQVuUfywq75mNz5rEUhAwSDMK/rnwOgIqNb9J1vKOw+5tvJmwfk5ZFrstMIrYzDZylLVVZrGDtVZ12QeglwSC89YZxOJ1/gS8hN6p9PdrRfuGwXJdu5Eywaq2btNZztdZVWuuxWus/W8tXaq1dyjuB1rpBaz0muy3NAh9+GJnct3J1TKbf6XUB1t++jGdOXsiG6/9uFjq6FJyDWIBoHktByDANDTCt6xUAPOhYi3r66UBi7sDIvFyXGUNsZyx2cYB4EeAcePL9S2IFa4/rtAtCH2hoAI82IQGtobIEIWpfj/ZwFo9Hrks3pDRrrgkGCf/t75E3h8qNbyYkSJ9eFzC5LnfsMCnAy8oigS6tYw7Gt3kjvnAH6pprpNtV6Ddqa+F171jogi5UrEW1rjt11VWRalZ/PeZXnBx6mKFnz5PrUugXuuvaj8T9/T5WsGak3KsgpEltLfhUJ2hQvrIEIeq8HpPFsAoiWHNPQ0PMqP+UCdKV1fHf2Ql79wIwYP1rcMQR8PrrcNppWWmyUJoEAjDmrFFwFzTPOIEhv1oYe43W1RnV8M1vAvDlZXVQ/t0ctVYoBdy69l0f8t7ELAEyiEXIFoEAtB4Qgk1Qf7ePw12uO7keuyeXMawCRBOk4+jeT5Yg3e4v2L3bGN6yMnjxRfBZsa1JSmIKQo/44Q/hgANgv/1o3f8gto4+lPWX1gNwwEiTyH/IaXPcrevQodFp+7oUhH4i7a59S7C+0dhVVLXVhfzELUylssyEBBz+SfET9hb55XKNlSB9y3X3smftVgZPGsWoBWe6ioHnV3s4yjGvOztRs2fDhAlmgYtgfXPxP9n2+MsM/dLJgCZ8++/Yb2x50nP0mGAwrbYLBcKPfgQ//3lkthKoALhuPuuBCZ1W5alkxSuqHSlCPfI+LPQv6Xbtr1vvZSKwprGTM+Z0P/o6PuVVf6XAktRaxUfSMJVu0loJ3SO/XD6QZoL0lU+rGMGqwIjUffvMAkeOQYC1ix5g8hXzmAS0L/s5PkJ40ej/QviRu/teOSgYJHxcLaM6OxgFsCZDxxVyx7/+lbBIYZVWfWAJ/M9EszCZYB0ULbjUWB+MLS4gCP1AOl2pjWvKmAh46Oq2tnq84LjpJvje9zKfAktSaxUnScNUbIdSip4neYFJjbhACohjj4v9d5nwAR8MG2YWhEIxXRHhP98HGMFhi1V7nlAGcmY0NKA6O2JTa2XiuELWaKwP0nDKIhrrrb6rI4+MrIvPQKFPm2cVZiepYF37yFuR6Qnz50SPKwg5ZNonTEhAGV1JQwds23nvvbGCY8mS/kmBJam1Cptk2SmShqmkUThgzhy46irzLaEriYiHtYD41MxItlU6ho7AP/sYWLAArrwSgDde7WTOZdE39mdqDzbpw4EQPrwYsZEyTrYn1NaC8oAOZ/a4QlZ44xf/ZuoPPg9A+2PlNLKM6UcfDXfeCYMHo4YOpeu99/F2dbJp/s9MKVVrQFUywbrlmXeYiP2S1MHOJQ0mw4Ug5JDJ04xgnTKpi2V3J3qvnN5OrzeqKfx+k5Ft5cqoXc2UebOFTaaPK/Q/qbzjScNUuhGsaQ8gLGHEw1pIOGIC/WecDg8+aK5oq4uh8b+hmAt+S8WBke3fur0hMh32V2Sm2z4QQJ1wfGS2ZcJ0CQcoIKpvXoiXMF7CUXHZ2mpWnn46bNiAd/w4AMZd/L9meTce1uFfPIHWSKELP8Pn1fbvHyEI6WANuvr4+C5X8+QUC11d5r1s4UIjPOrqovlcM9lt78wTK+EAhUV33vFAAC6/PO5/2k1IgOQG7h7xsBYSKuphjamIZb2xHTqtM+aNffJB0UFY0+sCMN9Me4cNyZx1rKqKTA787PFidQuIAWOGwXtmOiIu9zxjrRxgvuMzUHQjWKfXBWhkGTuXNDB8Xq3EsAr5gUtaKyfx3s4z48aO9lfKIUllVJj0yjvejYdVcgN3jwjWQsI56topGCxRMXlCKOaCP+iF2EFYEZzCt6/YHjmIyScr5D/DJu8HVpzU+tuXGXG58AmzoLLSfNvG1Ta23QhWcBS6EIR8oRvBKmJB6Am9ul7SyBIgLzCpEcFaSCTzsDq8YDEXfDAqWINB6Jf7QARrURDxhNr/T1uw9tDDKgh5iS1YO5O8xJMoFmTEtpCKHolLrSWtVQaQX66QSOZhjfeCWWx8q4Nx1vSJJ4RpcTtOXxHBWlzEC1b72hLBKhQy9nWcxMMaj6ScEjKKfd15PJKfug/IL1dIpOFhdfLe2n3RGVtoOI8TDLJj9qnsGjqelgOnQX09jfVB/vup89h66nnp5dUQwVpcJPOw9iAkQBDyjm5CAuJJJ+VUfFqjZGmOBEG8q5lBfr1CwilY0/CwHrRfxKfKEP8+aLNmurpM0v9jZzE8bBnwj0DPn88UPHgxwjOtIgBFJlgbb1tJ2x/+RsU3vsL0uqNz3ZzMEgyy47Lr8L76Ev4hVVTtPyRxGwkJEIqRHgrW7gbVZKu4QCFR7CEUffr70igaIHSPCNZCwilYnRd+Eg/r/oOjgvWhPzXDPKLbNTSgwl04h19poIyo6NShNJLBFZFgbawPMuX84ymji47gb2lkefGMcnd7QXkXEobfJQsJEA+rUMj0ULB2N6gm3gPrVlygGEVbMoo9hKLPf594WDOChAQUKk7BmsTDSktUsB45aW90eWdnjMvAWc0oZlk6RQD2OcIOClyw7lzSQBnmgeYjZPKSFguOF5SYymTxiIdVKEZ6KFghSS5Ni/icmfPmlXYOzWKv2tXnv89+NouHtU+IYC1UuvOwrlpFx7/+E53f6xCsoZBJ+m+JjlCZybmpzjgjusnQkekVASgiD6szyb1GFVXS+9erzf8xvtxqAkk8rFvO+RHNU4+CbdvM8m4Eq8TzCXlFDwVrd9dvfNL//iouUCgUc9L7YBA2bTKmsKd/n30drQ5az+Y0PKxiO5Mj/ulCJZVgDQbRxxyL3yFNNt94P/vbM/Z21luf/8yvwF13wdFHwx/+YJZNPbh7q9vVFSuSe+C9yEecxRVCNaOLJhwgGIQ/fP8tfkNUsMZ7V7ePmMrAslYq91iC9PHH4aST2P3+HgYDoze/iN4c3f7FRj9HHpz8fMXcPSgUID0QrOlev/FpjUo5h2ax5rGNL9l7zjmxRSVSxbU69727rJN1wJ7WMl4PJv99xHamRgRroZIqJKChgXg/mmdlQ3QmFDLbhsMmLtauauTMJGAb+FQ4vatQ8B5WJ+WDyrvfqEBoaIDakCkIoDDe4/jro2bHGsAhZq+/Hj7+cdq27GKwtY1T5P79y3+lY8Ul3Za5LMV4PiEPSRY25YJcv72jGAW781oAGDs2VqymEpfOfcNhc93t2O1jzpzkQlSuvdRISEChksrD6hKfqo44PLq91tBmpQwoL4+myGpvj26TRtfFCyuKV7CiU3ac9yuN9UEaTllEY31m+oRqa+E973gAwijX/61rXOuSJfjHjXI95kmdjyaN4yrm7kGhQLFfwDduhKlTYcwYGDHCfMaMiVl20Y1jeLVrKhsZwwddI7h4ceI2bvsxYgRMmgT19d02R7p9+4dM/66pbFl3ca3OfSu85tncSVnKGFixnakRD2uhksrDeuSRER9ah38g5R172e+zn4RH7o7uYw/I8vuj8Yi2iIVuy7cGg/DLuU/zV+fC7dt78YfkKTkSrI31QQ6efzyH0E7bYxU08mSfQxMCARh93ki4BfZOD1B90ky48caYbZx/beQ/P28eQ1etguej29jrHi6b+//bO/f4OMqy73/vZDeHJq1t00paoK0UCi0WiiCwIjQCouCBYh8eER/ri2AiqOCJQkUEqYLtq9KXg9AoIPUAj1j6gCAKFNIWGhXkFFoeim05lJ7TlJ5y3vv9457ZnZmd2ewmmz3l+n4++9ns7D0z9z1Jrv3tNdeB8+uCz1eMtweFAuaPlqXq7obXXkt8/913Yz+WA9OIv+Y96+EY47cfADt3QoMVV1Rf7zsVue07OAzGdU1my/oqfebc9/DOHviREazJhKjYzuSIYC1UknlYd+8GQI0ZQ/mMGfDkk7B3r3t/OwnL6WF1VBXoS7A1NUGkZ5V747ZtaSwgz8mRt7h1aRPTMZ7ucjpMpYIMxNJOGmt+tyPOPR0OPTTh/e7QMHrHvJ/K2pHG+l58sfnA/ec/Y2NUfX3Me1R3+bFJjWkx3h4UCphsp60vXRooWOW27+AwWNc1QjORt5fA1WuNh966ExmpqGD7ISPp3tHGsJJOyi+sgJEjoa3NNSYycmQsWXVC6bv84ZvNRJJMTGxnMCJYCxU/wWp7WNvazPPIkXEx6qwSAHFxWl4e97A6x/Qh2Orq4K7QseAs/VpTk/L0854ceVhrZtfB4/arDFYqsH+3VVW+Gf5lZ9XBo48m7uf4O3v94I9xJEawnvmLs2k5cnnRJKYJRc7s2fD4432Py+T5AujLMyf0j0G5rs3NMHNmQo1zm2rni13Bh7HvTlX17uGchTNpmVxENb6ziAjWQsUZh+jt9255WBk1Ki5YgzyszpCANARrJALVt02JZdXHzlcs5EiwuisV1GbOqNlfUKqr3W19bYLKVDn+zrau+jdTMIY3TFfGvL+CMOjY3s677jKKxuEFo8LjGfO+9hvjt23/fmNnP/GJQO8qyG3fwWJQrmtTU6BYTQfleA7ZNb7FdqZNzgSrUmo0cBdwFrATmKe1/oPPuCuBLwMTrXG/1Fr/32zONS9JFhLg8LDufnsPI4GO+5dR4dzf6WH1CwlI4Zb49MkH3Bsk6SqjlFekUKnBj1/9iv0/+QV6RyvdR89g1P/7UfzLSHW1vzgNKmjtuA6jzz+D9idvJEwX3ZQVVZ3aQkHs5gCor08qJL2k3YrzZz+DK6+Eo4/uc6jc9h0cBnpdE37ndXVQUjLgzzbnp0kPYbGd/SSXHtbbgS7gIGAG8KhS6mWt9RrPOAXMAV4BJgOPK6Xe0Vrfn9XZ5hlvPLyWI044wbxwJl01NdF+ydepBPb9cw3D39sCQMXWt9wH8Eu6cnhY92/ezVvHfJ7Rk0ZQO+8r/lagiMta5cNaug50U7Z6NW3f/hElr60hPKqKYdd8N/mHbnMzur6eKuulfu4Joqc1UfLRU8yGgJCAwDJmjjJA0+sjtLCc1qVN1Myuk1tauUHsZj/pS4A634d+JPBYX/q2vN3Nb24S72ku8P6O0/nS4Z+0FYHTTzd5IEcdZT7z+uGZV21ttHcq3h49g64r5ort7Cc5EaxKqSpMZ/sPaq33Ac8opR4GvgRc7RyrtV7oePm6Uuoh4BRgyBnelsZmpls/HzK/gZZDjjB/+LZ37N13iZ5xJpVWv/iq9zb7Hwj8k64cgrVqw6tM41V0C0T/+lv/rle2YK2oMBUG8kDkZYwceVibm8G+yu27DhA65aOMsr6f672gGxrM7aUg0epJLlEAPd3xbObqan9xGlTGzFO3cnp9RG5l5Qixm/2nrwxy7/tf/nI/EngsO/zIsm6uXSYVALKN93e4aBF861upf+kITNqyu/4tWACf/Wy/51cJHNnvvQXIXR3WKUCP1nqdY9vLQNJ7KUopBZwKeL0J9vv1SqnnlVLP7yimEksWrUub6LV+ZSFnr3tLbPQsN/3ibey4GV/p5Zd05QwJcB6jO6BwnC1Yqyx/3mAI1uZmtpx3KW9Nmsneo09KqcZhRsiRYHVe5nK6UI7fXqzc1NKlwQeYOdNVS1UDOhSOG90gD2uQYC3w7mVFxqDYTWtMUdvOZDUzm5vh+uuNU8x+H/pRD9MSrCra0/+e8xlmKNV79f6Oly5NXifVS2ANVGfCqpBTchUSUA3s8Wx7Dxjex37XY0T2PX5vaq0bwaQxn3DCCbkPQswwNbPr6Hy8PDGGcOVKAEJd8Vv09uJ7S8N0VIyker/nQ6iPkAAXYX+LvefmuxgBdEVDlEHmBWtzM9HT6qjtiXfg0g3/TO5hzBRa09LYnPXb3+7LrF0/OeujBmKHiVj0VlYTWv44XHaZ2VBd7a63a5Oih1XIKYNiN6H4bWdQBrntlevsNOarpMS8P2eOeTQ1meIntthJ6i21/ofKS7opJbnQTTs+th/kqt5rNtbmh/d3PHs2rFqVetWAwKQt+7NSBGvOyZVg3QeM8GwbAez1GQuAUuobmJisU7XWnUHjipnAGMKXXkoYuzV8KOpTn6J27hx6L7saXvIIVrt4dkBIgJOSpxMt3ZuX/4JJL1lCuc3qP59pwdrUhOrpSvAYJqtxmCl69u7nyIaZlNBL1+PltJCdEk7Oy1yBEepRoLP8fVR2voc655z42hsb2bPoLtraKyifMY3auXPggx90HS900BgjYq2/kbfueJSJDWcnnlgEayEgdrOfBIkR2ytnd6k+7DCTN+X8P0xZ9Fke1k+e0c38uuSxstkQkrmo95rLpgj273jJEvN6+vTkVQP8hLVf0taBHfsYBrz072pmnDyoSxD6IFeCdR0QUkodobV+w9p2LMG3+r+CidE6TWu9KUtzzEt8YwgvuQT9/POuEICa7q28fvYcaiMROrbujvWDj7HOuquYiof1+OMTNoWXxXtcxW5bZ1qw1tWhS0pdYQ5Acg9jhgh1xCsg6ByXcOp9Xw2V866Eq6+OZyA3NqIbGhiO5V57cyXRR++h5KFl7p07Oth+7sW833o54dc/5J3eHhJaB6SQdCXkHLGbA8BPjNheOdvDumGDiXucPt2MTUv0WYJ17Mge5s0Lnke2hGQu6r3mQ1OEe+815773XiNY/X4XqQrr5mYY/+Z+JgJfuKSKuydLTHIuyUkMq9Z6P/AgcINSqkopdQpwLvBb71il1BeBG4GPa603ZHemBUJDA2rWrNgNZAWU0hOLcd035+smlhHHTeYxY8xzHzGsgO8t5HBpXJxG7T+jTAvWSISST3/KtUn98IeDHw7gIaslnHxiRvXevbDZSqCzA+zuuw8FrgfdXbBihXnfjlnt6CD092fdB3xqeeJ5xcOa94jdzDy2V+7MM+PVi5zxjmn1dvfWww6gpsZ4c+3wg8ESkvba5s/Pnqczres1CDQ1xWOROzuD41aTxTR7xw3T5nNxd3dVzmOShzq5SroCuAyTOLcduA+4VGu9Ril1qlLK6er7MVADPKeU2mc97szBfPObuXOJhstiotQpsiYvqGfD3MVsPvhEekZY3aiGW2FvzpCAIMHpEawtjc3UvPWv2Ou28UcH7r/+u7ez4fCPs/6qfiZLvc/jG541q3/HGQDrF2exo5O3VBgQjnahb7vdvLDLpXg+CTSYWGPbG25ft44O1OGT42OA6CfPSTxvkIdVkq7yDbGbGSYSMUlX5eWJQist0eeth+1Dc7Px4Eaj5lyLFgUfMxMJU5GI8TBm+7Z8NkWyk5qa+MdQNBrcfDFVYV1XB9WYf6vusirpSpZjclaHVWu9C0hQH1rrVTg6nmmtP5DNeRUskQilK5rYsnAJWzZD+OI5LpE1eUE9LKiHD38Ynm/lzVVvMwncIQFB2CLJonVpEyXExanqsgStR7C+1fATDmv8gXmx8Emee+QZqjatY8zIHt5/TYpFvL3e3Rx4/LJaM8/2kDpQgLbDImwP68c/bj5lLToOPpzKB5bEheqIEbB1K3R0MOpDh8FzsHPsVPZc9C0mX/FpWHyN+yQpeFhbGpulfmCOEbs5OCTrkmSHEtgCMjCZKAXB6o2ZbW31H+e9Zf3Nb5ow9Nmzs36DKW1y2RThxRfNddXaeLCDrm+qXbEiJ/YCHWil+POTlWmvK1cJaMWKtGYtJiIRxi2LMC7o/eZm9AsvoIAJG54GYP/yZqpmzkx+XIdoXPej+6jc8CoaFYtdLR9dZXrpeARr+cMPuOJqT1j7WyO+9rjribY0NlO2aCET2l6kskLBjBkwd675D88DwZo1mpuJfvZcSvApRVZSCtHe+JcHu/2uReVJx5jr9cIL1obKeADbLtPkeuzPrmbsnDmwc2fCqbe3bInFuTrZs2V/LMtncsMZWUs+E4Rsk0xoOasJlJbCbbf5CEdnAxef/e2KA6nElTpvWXd0wEKrqu7jj5tn+9wiiOKsvfmvfOnOG/gh6wnTTXe0jJG3VcGiffEvEWVlJtt/3z4i3d1Eysrg1+a13xjbQaBKSoiU/IN4lezUGlHkKgGtWBHBOpRoakJHo/GYR2DYxjVw8cXJ97NE45rbnubo6y8EwClNVbfl9fPcPtYnnQQPvRwf53i2s/1bmM7Uho9Sah1RA+rNN+HRR4230ePdLWrB2hSvo6uAXqC17GBKT/4wNR8/Hq69Nu5htQVrVZWJPba3OzusVFSY7babwS7L4uNRH/XYfbQ0XpogRtcOP5GTWIkCwjlOPhOEXGHHRkaj5vH1r8cTs2zh8tkRYVMQ1+Nh9Sto/+KLyc/nTJjS2l0W2i6SIoLIQXMzU79ztquiDABJeuekRW8vzJxpPpMikZSufT4koBUbuYxhFbJNXR2aEldNTwVoW+wEYYmg9/70RGyT0zBUbPxf84PHwzru83Wxn189/NzE486eTevSJkqJupOHwBj9pqZED2sxx1TW1WFfAROHXM62Wx+gZsWyeLkqW5C2tZnngw5yb7efy8uNYIW4RzWJYFXoeCMKB1X/9TnaqaSb0uwmnwlCHlFX5w7zjkaNebKFy7XXwje/4x8S4BUuL75oMth/9Suzr1+MqjMW9Hvfc79nF0lJNXFoSJCNxdufSZjSWR0dya99rhPQihERrEOJSIQtXzTWT7seCd9L3fz972y66Fo2d472fTv25d+bdOXsQz/P3dJOzZ0L9fXUzI6HI7iqGITD5j/cFqzV1QnHLHTW/egPrDzzBloarU+sSAR1jGm+u27qZ3lj8dNxj6edGOcNCfAKVvt6+QlW+xr+K54wB+aaRynxFaPT6yOsX7ycZ8+an93kM0HIIyIREwYQCpnYyPJyY55ct+57EkMCmpvh7bfNfrZwgdSEpp0wtWABLF4MZ51lnu1wgKEsiBIS0urqXOFng9L5wvpMam6Ge+6Je71LS/2vfa4T0IoRCQkYYhzyuwWsP3gyFb+/i+qxFbzv5Gm8u3cEB/9+YbBsveIKDgE+RUVsU29pmFBvN92U0kuIEJ1JBSt7PA16zjGZ6tP/4yhogB5VSvdBh1C5/R1znD/+0R3DWl1tYoqKRLBu/PYtTFl0BUcA7ct/Go8NtUTmkXfPg5MdVaqdn3TNzbTf8Rsqgf0726myt0NiSAAkhgSsXOnunAVsGntcoBj1rf0rCEOM+noTBuCNW7Rv3RMKQycxD6vztnFpKXz1q6Z7FsRrhaYqNOt9clRTTRwqNvxvx0di2VZq4kRjBysqYORIczfKsovtVLCbkYykjUr8xyRsq6hw5VU03RT/GFIKvvKVJElbOUxAK0ZEsA5BYhUDLA4BNgGH/H5h0v3KiIcOhHqNUV595nWMj0zkiPlfThSszltjXsHa3W0szw9MFYHecCX//tF9TP/FRfD66zBlihlnC1ZbbBWJYC370x8AEmNDnYH/TmwP686dRD/6USqtaz3sDavLWbKQAHub7WGtqyNaGqKkN34t9UV9xDELgpAgQJzdlca3huEBYv/DTu8rwIQJ8X0zITSHasKVb2zoh3vipQHefNN3v0zF/HobMthfQoTBRwSrAMDeGadCH4K1l9JYcpTNzL9dA2vWwHwgGmX9VY1U/P4uKiaPp+bDk+MDvYL15Zdh7txYElhZ1z6ObJjJgYmTGQbxrltesVUkgrV8ykTY9A/A05jA9pTaJXJiO1iCddculF+93GSC1cYW/ZEIpatWsvPqhXRu2EzHhRebLzGCIPSLe++FwzpDXAu07+2mkuSdpvw8b+kI0KGccOV7XZ12L4BMJUENVc92PiCCVQBg2xOvcBQkjWZ9+eQGPvz32+IbwmHzjbbEhEJ3vbOVwxY2mPfeBf1Mafx4XsH63HNgiVWs84boZv+eTiNY91rt0YvUwzrmQxPgKfOzKzY0yMNqv3YYZFecli107etlVwhwsPaBNUy7Yrx5EYkwZsWy+IdksxheQegPthDqiJovmR37eqgkibD59a/hlltg06aYPesixKF7K/kq7YTpoXsYhMtDpjxde7sZF4q/nrG/h81d0EOIjvZKRp3eDmHLNoZ89psxA37yk6L4J/e9rrv6Fqz9aVUb9CXC/tmOPy6Cy1oQiGAVAKiZ/TE6Hy+jnK5A0frh00fA3x0bbONgCday99z1PWPF7iFRsE6caMY4NvUQpvTQ8dC2Me5h9SZd2ffXenvhiivo/O9lRPfuR71vOBXXfx8uvbTvxeaAlsZmWpc2UTO7zohThwfVFTval4c1HEaNGAF79rB13HF0nvufTLpzXqKHta3NVBp3cPi3PkVL5YrY+Yayl0YQMoUthKKdYYjCsHA8FCrBk7pqlQlm9VAGHEJbfMMB69HW5h5ova60HtZG6MA8PONiPP10QlmmQvEQSEgPKQAAIABJREFU+s014bo6k00DSNczmsw+iu3MDSJYBcCIphaa0HfeyTEvLvEd0/rMWlyd7jyC1YmGeLF7gPfecw94vylTr8rK2D9+MpuGHUnXFXOZ3nQrvPJsoofVGxJw+eXwy18SM0/b30NfdhmqtDQ/WsHccgvceiu0t9PeG+KDW98CIPq44q93XMnRH6rkUL/9LA/rbY1lHH+uwwjaHtaODlN3FRj31j9MQtWd84zlbGykc/5CyoHOpmfxmu4Q3a46qlInUBAGji2Enn8oBAtA9XRz002mSUBrq0cc2ZX/c4FVlqmZSMGIrcZGU/O2t9ckrt1+e4B5t76od6gKbk7SjSwoCcpPFCezj2I7c4TWuigfxx9/vBb6wY4ddp3qhEcPJe5t48aZfdatSxxbUqp1Q0N820c+4h5z3XXmeeJE9/nr6832O+7QOhqNj/+v/zLPv/mNGTdtWsI5o6D1WWcNbP1+a0+Xn/408BpGrcdDJef6Hr9r5FitQY8r2aorK7Vevdp6Y9MmM7a62jxXVZntu3aZ1+Fw7NjO8zgfnYT1K4tXx861erXWlZVal5Zq97mEpADP6zywcYP1ENvZT7Zt0xr0dsboEstUlpR4/rf++Mc+bUM04P0BP8JhrVev1jfeaP7nwTzfeGNOr1ogq1ebKfssIZHXXtMa9OtqStr2LMgOJrOPYjv7x0Btp3hYBTc+3tLYW56Eq2Qe1tJwCMaPj2/whgTYHlRvYpDtSd27N357vKws7mG0PazHHANr18ZCCmJhDHZV7VzywAOBb9ldvo6MrvV9v6ejmzDQHi1zf3O3r7UdKjFqlHm2t3d3u0I5FKYbWcnUqexvJ+7BdoQfSPKAIGQQK4wnTHesYEo06vHAfeAD5o0RI8z/sKOUkkqlvFJfY/y2bd1qtv/lLxCJUEf6sZy5oKkpsU9Mb2+AN9P2sOrytL2eQd7SZPZRbGduEMEquHGIT2d8qRFaCuXcmkSw0t3tTpDyhgTYgtUbczR8uHm+/np6rruBENDb20upXUvUPuaJJ8L999M9fBTRzm4quvahzj47P8IBTj7ZVZzfex0BWkpmcGT0DfMiGo3HAVulw6IlYfeHiTcJyxasnu2uRKxwOdx1F1WRCEcGTFXqBApChnAI1pKS+L+16//Ybvhx/PHw1FODNhXXLe7PjDGxCcccAxSO2KqrMx8PdktcpeINGxKwBGu3KqfUe81TOE861RxSeU8YHESwCm4c4nNfdS1dHzqZsm1vM/z1F+gtCVESddRWTSZYo9H4N35I9LDankKvYH39dfN84EDsj7Oktxf98J+N2LMFqxXrWdZwMdTWmv6FRwbJstRoaWxm+oCOYPGxj5lgq+HDoaYGNXIk+3d3wrZtVLXvYs2xF3LCqcfBbZYntr09VgWh1Kpv+/3ryzjtTIdB9Fyn/f9+lw2NzcZjan86Aj3DR7G5ZjrlM6ZRO3eOWFRByBYhY7EqQ938+IZ4DGtNjSOb3E6Gsr9wDgLehKC24ZUmnr29PTYm02JrMJK4nMLaNx7YifVZc8T0cuZfEDwuKIGrEAS8IIJV8OIQn8M/NAVWLGPLeZcy/PUXCEfdPbKTClaAAwfiPweFBHgF6/btCYdRgNZGkG1e+W/WPXwTR1W+SS0Yr8awYWagwyCnw1sNP6HyvruZtvfNfu2fgJ0o9pnPwO9/D2C6UV1+Odx6Kx+86ET3tbEFq9YxIX7VD8LuGmOedqrD2ndxZMNMWljB9LKy2DnDX/w8E++4IzPrEAQhdSwPa0lvD/PmmU1e8dhy+W4mg7llP0h4b3Hv7zWC9cXV7fz1D5kVZc3NpmnC3Xeb82U6iStlYW3ZvxFjymPX3m+uQclm4i0tDESwCm6c4tPyGHS9/Jr/WPt2dJBgdQpI7enuHORh/fzn4amnEntBqxLQUQ764y2MI0ovpWZ7OGxqDnrPlyLbPv0VJj56T9r7JSWoxIr9urPTHS5hi9ceRx1F5SkutmKF66Vdt7Z1aZM5rn3O0aMHPH1BEPpBqWWTtIZJk6Czk6M6K3i2fSQjaKOivZOam02FD955J+XDpuu99N7iLh9VCa1w2UXtPNeTOVFpC8COjrh5z1nGvDN+NwDJ7C98gjNshKGJU3xaBrjn/AsTBSSk52H1EpR0VV8Pixejpk6la3Qtu0dNYtdps1BfuMBMiagl1qxofKdgTXa+ACpWPZH2Pn0SZDydgtXZttaet73NW4MVzKdQKISG2KOHsOmQ5RTGNTWJ+wqCMPg0N8d/fust2LqVkW1vMoOXOIy3GM9Wyrssu/fEE6ZmUwqHPOMMuPZa8+w8RRD2Le75881zVY2xj6HudpdYGyi2ALTFqlI5TOJKodOVLeRLS/M72UwIRgSr4MbHwzp5QT37jjo+cWxfgtXH4xkTvkEhAWBE69q1lLduYdSujdSsWBZrNJDAAEMCQmPSvzW3/ru3s2X8cbTOPM//E8TZbcpJgIe1/fRz4Lzz4JlnzAZvghWYT6GVK1GzZnFg0lTWTZvF64utJgAO7/X2J19Jez2CIGQAHxWofB4xli5N6ZBer2AqRCIwb57lQbS+0A8PtWdUrDkFYHk5NDT07bltboabbkpNeCcj4TgpCFavkHc2AcjEnITBR0ICBDc+ghVg+Ienwf+64ygH5GENCgkIIhTwpzoAD2tLYzOTN7+b1j4br7iZw275jomr3fIS0dMepWTlCreVTiEkYMfLmxlrba7YshH9PxtRjz4aX5MfkQgsW0YVxLP+m5vRO3bEPgjHPnYv66/6CJMX5EG1BEEYSlh3QVJuH51CCb7+tBNNwLKPP/9JO//TnZkYVjtMYdGiPpKhPPtkomGB73FSEKyQGKsqHasKCxGsghufkADA1+vX/twrVDY3w5Qp/sfyeDw7ykew75RPMOapB5J7WP1wzsVJPz2sLY3NTGmoo9wqI5UqZQ/eHxOHpmpBd2IwVAoe1o51b8Y2x7wudkiAn4c1CD+vzoNLQQSrIGQX6y4ICxfCiy/61k7t7II9odHs+T9XMDmFEnwZyWC3BOvUie1MPb8f+3voq2Vp0FwzFUPqe5xRqQnWlI4lgjVvEcEquHEm+zhFoo8hqHh3A9GZH6PkkYf9j+XxeFacfgoVn5oJTz0Q97AmCZJ3EeRhLSvrV9JV69ImwqmK1ebmmBUriZwMD/wTsGJJQ2GU1+3RVwxrRwfDPjAO3sLd+MD2zgR5WP2oqyNaGqKkN+7V0Z/Lg+YJgjAUse6C+OESerfC8lmpiaMBZ7APICnVjyCR15e3MsFbfGovfHcu/PGP8c+DcNhUTNm/3x3T79j2HRXm871VVLKf8t5uqm8CtGX/Vq922eu+yIgHW8gaEsMquHEKVqe31UewKoDuLnj2Wf9jeQ1kTU1cjNlxlzkKCaiZXUfU+vP3TShz4vBijvuPU2I/K6DkyScSjWMKIQE1R44BYFvtsegS64vBokXmOR0PayRC6aqVtJ42i82HnMiGuYslHEAQ8pD+xqMOmAwL1qDkpb7WlxBD2ngR/OIXsGmTaaiwezfs2AFvvmmeA7aVt+3gMN5kHDsYzW7K9u6OC95XXzV1sFMMSA2KaxXyE/GwCsE4BauPiNIA4TI47TT//b0CcvToxONkQrCmGhLQ2Ah33cW+ti5qd3fSVT2a0L6d7J9yHNXrXgzez7k+bz1Zq3uMi1SSrqz11M7/hjHar70GY8bE15QOkQhjVvh7dQRByA9y5s3LoGBNFruayvpc3uILVw14Pr6keW9farAWDjkTrEqp0cBdwFnATmCe1voPPuMU8FPgEmvTr4GrtfYW9hQyjtPb6iMst836mummdOyx/vt7BOuuf21g9Iknusdk0sO6ZQsdYw+m90AXoWHllH/keLj6amONGhuhoQGNKeJf5ThE9eknQTLBGo2y5bxL2fP6VsaVbmOE87329sSuNakIVvvncDheisru+Z2uYBWGFGI7C5P+xKNmpIOUbR+vucbUxwJjTysqjP3qNSUCuwnRTgWVtBO2ywY6xvV09XJke4hJVFBOOyMqewndGB8TaW+nraqK5Wddx6ir6vuOaz34YOM5zTRyb79oyaWH9XagCzgImAE8qpR6WWu9xjOuHpgFHItx6j0BbATuzOJchyZ9CNbaZVZHJVugefF8ox/17CNsrRlnOlQlOa4vyZKuWlrMz729VOzcbH4+APrhd1GPPWaK7v/ud4CnrIzNu8krBUQ/dga1vd3uedv4hSEEZaz61WENheLF/m3Bmk5IgDAUEdtZoKTjzctYBvv69ea5j5CpsPUIIgS42pL4OGzLeY9zHmqAc4BIffI1HHSQeT744Jho9iap+W4LGjNhAkybBnOkJXWxkhPBqpSqAmYDH9Ra7wOeUUo9DHwJuNoz/MvAz7XWm6x9fw58FTG6g08fgjVGOmWtnnvO/frFJJ5NJ8k8rP/4h+9bJsbWyuI/9lhYtcqd5GQzYwb8+c+Bp1a93f5CF/zXmE4d1nA49nPn4rspB7r/99+E00gcEIYOYjuHDkExoWl7XHftGpT5JWXpUqivT56F39Zmnu+916jafpIRL7RQEOTKwzoF6NFar3NsexmY6TP2aOs957ij/Q6qlKrHeBWYMGFCZmY6lOkjhtV3nBPrW7Pz/mPP8JGwJf5aL12Kamw0zQKSkUyw2l2gPPUPlXXuA/N/Rnm0gxCghg/nwMhxDHvH8ad33HHJz221hfXFLy4sFcFqe1jXr4e//Q2AsrbtAIT2thE9bWZifVdBENs5ZPDGhNbU9NPjesEF8PTTSYf0FSNif2HXPtt8sWrMJo1rtYX0ANpJSx3VoUWuqgRUA57sFd4DhgeMfc8zrtqKz3KhtW7UWp+gtT5h7Nix3reFdOmjSkAsEzNIsFpsvuA7bDz8LDbMXYzu7Ek0jil0fEkqWB1doDpqJ3JgWNwAKqCqfRehTssTuncvwy75ovsYw/3+7OKok08KfvPAAdZf1ciOsdNoH/cBI35tV8gTnravfh7W116DaDQ2V/tZ2fVdBcGN2M4CJd2OSt4M9tbWflYZsNpdM3Uq1Naax6RJ5s7SxIlQW4uqraWjdhJba2fQUTuRrtG1bKGWjUziJWbQXmuNmzQJNWMGytrPdayRVtfAM8+MOSCca1i0yMw5tn5LsN5+3+h+d5nKWeUFISfkysO6D9y5K9brvSmMHQHsk8SBLJAkJEBjYjtLn14OJ5+c9DAHf/08+OjPAVgPsPAZt2hNoeNLUsEKsfqHlWA+Fa65xtWy1MXy5e7XVVX+42yC4meBLbf9icMe/GXstd7q8DzcfDMcdVTce+znYZ0xA/77v9Fd7pqwvvVdBUFsZ0HSX0+gN+a131UG6uv7vItVaT3AmNBrrzVCsLQU5l9uWr0m5Xe/gy99ydRCra01dvXAASKdnRzfA617y9hPFVUcoGt4J6G9bZQAD/3sDa68bWK/vKNSR3VokSsP6zogpJQ6wrHtWMCbNIC17dgUxgmZJklIgIL4V1ql3OLWzkq1cYjNyQvq2TB3MTvHTOXApGmoxYv7DgfwHMOFX0Z9XR2Ew6awv+cBwHnnucf3leS0N1EL2MeqevKh4F7h4PYe+3lYjzkGmppQX/sae2acxs6xU9l12iwJBxCCENtZgGTCE5jNmqFBtVaT8tJL5vnAAdi2DTZsMImkbW2U7W1jHNs4nA2MYytllljVwCP6HD7U2Zz310TIPTnxsGqt9yulHgRuUEpdgsl0PRf4iM/wJcB3lFJ/wfx9fxe4NWuTHcoEeFi7CQGaEqclKymJZ3pWV7tjOz1ic/KC+vRbhyarEuAlEjEicMkS3vv7WqJvvsWwkk7Ka0fDFVfARRfBt7+d/BhOPLVXne6psmEh2JMktsvpPbauYfvb24nqYaa0VigUc6O8L/ksBEFsZ4GSKU9gtmqG9qsl7Msv9z3GgwJCdHN6SRN1dclPEpRcJXVUhw65LGt1GXA3sB1oBS7VWq9RSp0KPKa1rrbGLQYOA6zaRfza2iYMNgExrK+cdAljjpvAxDl1cUvhEKyd0VJcAQRB3tF0SMfDCn2LwHDY3fovGR4P697h4+EDhzHilWeo2LMDgP3DxlI6oorKWqvcSlWVEccO7/Frf3qVqUBFx26ib7yS2rkFIRGxnQVGXwIwG5nu6Z4jbSF4/vnw5JOBb/t9qdeALglz/u11TE9yLkmuEiCHglVrvQtTI9C7fRUmWcB+rYG51kPIJk4Pq+O2+fH/MRm+9z3X0KiOx5eEW7e5j5MJweqNO7Xpr+CrqjJt/yDtkIARX/i02fcVYmWt9p05i9qHGpMeZvsTr3AUVktXrKoDmbg2wpBCbGdhEiQAsyHGBvMccSFcT2QxJruqvT2hVqpy1E/t3NvJAV1BdPoMan46l+l9TCZpeSxhyCCflkIwQVUCfESWjsa/PytvHYAHH4QPfnBgc1m7Fo1PjGgmBOsrryQfaxenttm1C5a5W6Ee9PCvWH/VCSbcIYDR55+BfvKH7usjHlZBGNJ4xdiSJZn3tg6W4EsUwvVE1iYP93Lt8xwsB/qaiiRXCZC7pCuhEAgQrDse9SnUn8RTqK+7zrRGHQgXXGCO5d0+EMFq89xzfdYhdLFtW6wUlRP1YPLyXNPrI+yuPcqMtTeKh1UQip5kJa2cCU6lpXDPPSZD/4wzUi+B1Rf9SqJKgf4kk/Vnn0jEOG7POMM8i3d1aCKCVQjGERLwzh3xTlBjnryf9Ve5BWhpeVw4+oq/VGqtJqO+HrV4MfsPP9a9PQMeyrVjT0OHylIXrYcdllCFAEB/Lnl5rpbGZkZsXefatvWaW9KcrSAIhYTtUQwSoc5M9698xRQQyXRd0Uxm0zvFd3+EcH/2aW6Gb33LzP2b34RLL82cmBcKBxGsQjAOD2vvymddbyV4Ex2lHXW4LLGUVCq1Vvuivp7qFX9xb7v//vSP09yMfuPfsZdXXh1ize1N7D4hxfaAkye7SlFtPuRENsxdnDQcAKB1aRMKt2f2oEfuShD/giAUD6l4FCMRU+d0zpzB8YTa56ir8xTvTxOv+Ib0hXB/xLP3Gi5enFkPtFAYyP1IIRiHh7X3/C+gFzbFXid4E/fti/1YuqKJLQuX0PXiWkZXdTD8iotTq7WaAmvvf5lpjtf6yitRI0akd/ymJler1RO6V/NI6w+Zd2U9fD4guctJebmrCkGq5ahqZtfR/XiYEkyTgFhnqweXpl/mSxCEgiCd+Mt+lZNKkUwkXvmJ73nz0q96kG4FAvsadnQY34jWknw1FBHBKgTj8LBOXlDPeoy40p+bHexNHDECIhHGLRscK7L9sReYiif5aunS9ARrXR2UlELUlOFaFT6dn9QBm4M7Wrnwa1ObAtPrI7TQRNl1VzNl68qUQwkEQShc0hWhg1VXNBOJV6mIb1sYd3aaj5Dbbx+4v8K+hkuWmBjfnh5JvhqKiGAVgvG0HE+l4H9ney/rGpuZXj84grXm/NPpejJEGT3xjemGG0QiqHPOhkceAeAnTR81hvuhJIljOERyPwUrGNFK/QrWX9XYt/gXBKEo6EuEZqMOayYy7VMR301NRqxGo+bxjW/A9OkDX5d9DefMGfxrJeQnIliFYEpSDHF2BBKVde/niIaP0cLTgyJajZdyJWWLFjJebe5/uMGkSbEfY0bPp5uWLVRd0n0AgtWmX92+BEEoOrJVFD9T4QZ9ie+6OvPRYRdS6e3N7K176Ww1dBHBKgSjEqqe+tPU5BJ2YbpoXdoEg+RlNV7KZX0PTEZlZeI2H4EepSRW5D92NSoqBnZuQRAEi2wWxR8ssef1EN9+u/Gs9vaa7/dy617IBCJYhWBS9bDW1aFDZdBjkom6KaNmdt3gzSsT+AjWjX97nQ94tpX+bglb/vQMWzf1ctzzvzIbM+BhFQRBgMIviu/nIa6vN2EAcuteyCQiWIVgUhWskQglK01lgC2bIXzxnEGLYc0YPoL1nTV7mIT79v8br0c5YtkdjItGoVQEqyAImWUwKwNkg2SduubNy/HkhKJCBKsQTKohARCrDDBu8GaTWXxu64/6z7NoX76AMjoJWWEAh87/Ki2HHO4W4KkKeUEQhBQo5LhMp4c4FIK77zbidTDjcYWhiXzyCsEUszAbNiz2Y0ujSRqbXh9h/eLlvDT6zFjJqVJ6TDyuk3SEvCAIQpHh7HblbARw0UVGrGa6U5cggHhYhWQUsTB75++bONT6eXLDGbSwnOn1EasKwfW0N6wiTJd/PG4RXxdBEIRkBFU1iETMe/feW7jxuEJ+U8QuNGHAFLGH9Z0N8TqusaoGFran9dmz5rN+8fKEeNyNj72WrWkKgiDkFclazQa1XXV6ZAWhv4iHVQimiD2Jwy/8DO0rFhEK8KKa0llxodrS2Mx06+fx/+9qWqadlP+JZYIgCBmmr6oG3njcbNWZFYqf4nWhCQPmpUc3Fe034un1Ef6dxIvqxemB9Y1rFQRhyDMUPIlBXtQgknlkBSEdxMMquHFY2qP/tYSP113CTU2RovxG7PWiJqNmdh3tj1cQppuuQqgzKwhCVhlKnsR0qhoUep1ZIX8QwSq4aWoiiqIEDWhO6W6iqUgFazqYZKynaF3aRM3sOgkHEATBRTY7VhUShV5nVsgfRLAKburq0OUVdHea2M5nw3XcVJfrSeUH6XhkBUEYWognMZhCrjMr5A8iWAU3kQilTy9n05ImVlDHTXPEuyoIgtAX4kkUhMFFBKuQSCTCxEiEObmehyAIQgEhnkRBGDykSoAgCIIgCIKQ14hgFQRBEARBEPKarAtWpdRopdQypdR+pdRbSqkLk4y9Uin1qlJqr1Jqo1LqymzOVRAEIV8Q2ykIwlAmFzGstwNdwEHADOBRpdTLWus1PmMVMAd4BZgMPK6UekdrfX/WZisIgpAfiO0UBGHIklUPq1KqCpgNXKu13qe1fgZ4GPiS33it9UKt9Qta6x6t9evAQ8Ap2ZuxIAhC7hHbKQjCUCfbHtYpQI/Wep1j28vAzL52VEop4FRgcZIx9UC99bJTKfXqAOaa74wBduZ6EoNEMa8NZH2FzpE5OKfYzsxQ7H+bsr7CptjXNyDbmW3BWg3s8Wx7Dxiewr7XYzzC9wQN0Fo3Ao0ASqnntdYn9G+a+U8xr6+Y1wayvkJHKfV8Dk4rtjMDFPPaQNZX6AyF9Q1k/4yGBCilmpRSOuDxDLAPGOHZbQSwt4/jfgMTj/UprXVnJucsCIKQa8R2CoIgJCejHlatdV2y9604rJBS6git9RvW5mMBv6QBe5+vAFcDp2mtN2VqroIgCPmC2E5BEITkZDXpSmu9H3gQuEEpVaWUOgU4F/it33il1BeBG4GPa603pHm6xgFNNv8p5vUV89pA1lfoZH19YjszRjGvDWR9hY6sLwlKa52piaR2QqVGA3cDHwdagau11n+w3jsVeExrXW293ggcAjhvZf1Oa/21rE5aEAQhx4jtFARhKJN1wSoIgiAIgiAI6SCtWQVBEARBEIS8RgSrIAiCIAiCkNcUrGBNta+2MixQSrVajwVWIe28Jo31FWTP8HT6olvjy5RSryml8j7bOc2e7x9SSq1USu1TSm1TSl2Rzbn2hzT+NsuVUnda69qllPqzUurgbM83XZRS31BKPa+U6lRK/aaPsd9WSm1VSu1RSt2tlCrP0jT7jdjO2DixnXmG2M7YuIKzndmwmwUrWHH31f4icIdS6mifcfXALEwJmGOAzwAN2ZrkAEh1fXbP8FHAJ4FvKKUuyNos+0+q67O5EtiRjYllgJTWppQaA/wV04GoBjgceDyL8+wvqf7urgAimP+78UAbcGu2JjkANgM/xiQ4BaKU+gSmbNQZwETgMOBHgz67gSO20yC2M/8Q22koRNs5+HZTa11wD6AK80uf4tj2W+CnPmNXA/WO1xcDf8/1GjK1Pp99bwFuzfUaMrk+4APAa8DZwKZczz9Ta8OUHfptruc8iOu7A1joeP0p4PVcryGNtf4Y+E2S9/8A3Oh4fQawNdfzzuDvT2xnnj3Edsa2i+3M08dg2s1C9bAG9dX2+6ZytPVeX+PyiXTWF8O6XXcqSYqJ5wnpru9W4PtA+2BPLAOks7aTgV1KqdVKqe3WbZ8JWZll/0lnfXcBpyilxiulhmE8Co9lYY7Zws+2HKSUqsnRfFJBbKcPYjvzArGdcYrZdvbbbhaqYE2nr3a19Z5zXHWex2L1t2/49fTRMzxPSHl9SqnzgFKt9bJsTCwDpPO7OwT4Mub2zwRgI3DfoM5u4KSzvjeAd4B3rX2mAjcM6uyyi59tgb7/T3OJ2E5/rkdsZ64R2xmnmG1nv+1moQrWdPpqe8eOAPZpyxedp6TdN1wVVs/wlNanTDvKhcDlWZpXJkjnd9cOLNNaP6e17sDE8XxEKfW+QZ7jQEhnfbcD5ZgYsypMp6Zi8RKAv22BJP+neYDYTg9iO/MGsZ1xitl29ttuFqpgXYfVV9uxLaiv9hrrvb7G5RPprM/ZM/wMXRg9w1Nd3xHAJGCVUmor5p92nJVdOCkL8+wP6fzuXgGcH/75LARs0lnfDEws0y5LCNwKnGglTBQDfrZlm9a6NUfzSQWxnQ7EduYVYjvjFLPt7L/dzHWA7gACe+/H3AKoAk7BuJWP9hn3NUzQ+cGYbLs1wNdyPf8Mru+LwFZgaq7nnOn1ASGg1vH4HCYTsRZzqyvn6xjg7+50TPbnDCAM3AysyvX8M7i+e4ClwPus9X0feDfX809hfSGgArgJkxRRAYR8xn3S+t+bBowEniKF5J5cP8R2xsaJ7cyzh9jO2LiCs53ZsJs5X+QALs5o4H+A/cDbwIXW9lMxt63scQpza2SX9ViI1ZI2nx9prG8j0I1xs9uPO3M9/0ytz7NPHXme6Zru2oBLMXFKbcCfgUNzPf9MrQ9zO+v3wHZgN/AMcGKu55/C+q7HeGycj+sxsXL7gAmOsd8BtmHizO4BynM9/wz+/sREr84QAAACnUlEQVR25uFDbGdsrNjOPHpkw24qa2dBEARBEARByEsKNYZVEARBEARBGCKIYBUEQRAEQRDyGhGsgiAIgiAIQl4jglUQBEEQBEHIa0SwCoIgCIIgCHmNCFZBEARBEAQhrxHBKgiCIAiCIOQ1IlgFQRAEQRCEvEYEqyAIgiAIgpDXiGAVih6lVKVSapNS6m2lVLnnvV8rpXqVUhfkan6CIAj5iNhOIZ8QwSoUPVrrduA64FDgMnu7Uuom4GLgm1rr+3M0PUEQhLxEbKeQTyitda7nIAiDjlKqFHgZeD9wGHAJcDNwndb6hlzOTRAEIV8R2ynkCyJYhSGDUurTwJ+Bp4CPAbdprS/P7awEQRDyG7GdQj4gglUYUiilXgCOA+4HLtSefwCl1H8ClwMzgJ1a60lZn6QgCEKeIbZTyDUSwyoMGZRSnweOtV7u9RpcizbgNuCarE1MEAQhjxHbKeQD4mEVhgRKqbMwt7T+DHQD5wPTtdavBYyfBSwSL4EgCEMZsZ1CviAeVqHoUUqdBDwIPAt8EfgBEAVuyuW8BEEQ8hmxnUI+IYJVKGqUUtOAvwDrgFla606t9XrgLuBcpdQpOZ2gIAhCHiK2U8g3RLAKRYtSagLwN0xs1dla6z2Ot+cD7cDCXMxNEAQhXxHbKeQjoVxPQBAGC63125iC137vbQaGZXdGgiAI+Y/YTiEfEcEqCA6sItlh66GUUhWA1lp35nZmgiAI+YvYTmGwEcEqCG6+BNzjeN0OvAVMyslsBEEQCgOxncKgImWtBEEQBEEQhLxGkq4EQRAEQRCEvEYEqyAIgiAIgpDXiGAVBEEQBEEQ8hoRrIIgCIIgCEJeI4JVEARBEARByGtEsAqCIAiCIAh5jQhWQRAEQRAEIa/5/44wZPZLKTRRAAAAAElFTkSuQmCC\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",
@@ -2091,6 +1388,26 @@
     "\n",
     "However, by aggregating many decision trees, using methods like bagging, random forests, and boosting, the predictive performance of trees can be substantially improved. \n",
     "\n",
+    "\n",
+    "## From a Single Tree to Many Trees, that is meet the Jungle of Methods\n",
+    "\n",
+    "As stated above and seen in many of the examples discussed here about\n",
+    "a single decision tree, we often end up overfitting our training\n",
+    "data. This normally means that we have a high variance. Can we reduce\n",
+    "the variance of a statistical learning method?\n",
+    "\n",
+    "This leads us to a set of different methods that can combine different\n",
+    "machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are\n",
+    "1. Votign classifiers\n",
+    "\n",
+    "2. Bagging and Pasting\n",
+    "\n",
+    "3. Random forests\n",
+    "\n",
+    "4. Boosting methods\n",
+    "\n",
+    "We discuss these methods here.\n",
+    "\n",
     "## Bagging\n",
     "\n",
     "The **plain** decision trees suffer from high\n",
@@ -2135,20 +1452,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 43,
-   "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",
@@ -2173,24 +1481,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
-   "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",
@@ -2246,37 +1541,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 45,
-   "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": 45,
-     "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",
@@ -2300,36 +1569,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 46,
-   "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",
@@ -2341,37 +1585,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 47,
-   "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": 47,
-     "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",
@@ -2385,42 +1603,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 48,
-   "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"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "LogisticRegression 0.864\n",
-      "RandomForestClassifier 0.872\n",
-      "SVC 0.888\n",
-      "VotingClassifier 0.912\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/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": 19,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.metrics import accuracy_score\n",
     "\n",
@@ -2439,8 +1626,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 49,
-   "metadata": {},
+   "execution_count": 20,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "from sklearn.ensemble import BaggingClassifier\n",
@@ -2455,17 +1644,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 50,
-   "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))"
@@ -2473,17 +1656,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
-   "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",
@@ -2493,20 +1670,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 52,
-   "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",
@@ -2591,21 +1759,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "AttributeError",
-     "evalue": "'list' object has no attribute 'XXX'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mAttributeError\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      3\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0msklearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_selection\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mcross_validate\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;31m# Data set not specificied\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0mX\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mXXX\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0mY\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mYYY\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;31m#Instantiate the model with 100 trees and entropy as splitting criteria\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mAttributeError\u001b[0m: 'list' object has no attribute 'XXX'"
-     ]
-    }
-   ],
+   "execution_count": 24,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.ensemble import RandomForestClassifier\n",
     "from sklearn.preprocessing import LabelEncoder\n",
@@ -2628,8 +1786,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 54,
-   "metadata": {},
+   "execution_count": 25,
+   "metadata": {
+    "collapsed": false
+   },
    "outputs": [],
    "source": [
     "bag_clf = BaggingClassifier(\n",
@@ -2639,20 +1799,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 55,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "0.976"
-      ]
-     },
-     "execution_count": 55,
-     "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",
@@ -2671,35 +1822,25 @@
     "\n",
     "Example will be added here.\n",
     "\n",
-    "## Boosting: AdaBoost"
+    "\n",
+    "## Boosting, a Bird'e Eye\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "## Adaptive boosting: AdaBoost, Basic Algorithm\n",
+    "\n",
+    "\n",
+    "## AdaBoost Examples"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 56,
-   "metadata": {},
-   "outputs": [
-    {
-     "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"
-    }
-   ],
+   "execution_count": 27,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "from sklearn.ensemble import AdaBoostClassifier\n",
     "\n",
@@ -2738,35 +1879,21 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Gradient Boosting"
+    "## Gradient boosting: Basic Algorithm\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "## Gradient Boosting, Examples"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 57,
-   "metadata": {},
-   "outputs": [
-    {
-     "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"
-    }
-   ],
+   "execution_count": 28,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "np.random.seed(42)\n",
     "X = np.random.rand(100, 1) - 0.5\n",
@@ -2861,28 +1988,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 58,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "
    "
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "61\n",
-      "Minimum validation MSE: 0.002712853325235463\n"
-     ]
-    }
-   ],
+   "execution_count": 29,
+   "metadata": {
+    "collapsed": false
+   },
+   "outputs": [],
    "source": [
     "\n",
     "from sklearn.model_selection import train_test_split\n",
@@ -2944,32 +2054,14 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "source": [
+    "## XGBoost: Extreme Gradient Boosting"
+   ]
   }
  ],
- "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": "3.7.4"
-  }
- },
+ "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 866e8bc6fee5f73d5e3c883649989c1a4a087323..cc84654c373b9919009e62dbf1803a4e4fc3b795 100644
GIT binary patch
literal 195
zcmV;!06hO6iwFSnS-o8V1MSaC3c@fD2H>uHia9|^A4%6jUAPcLdV$nVZG0p(Nx|OU
zv;|$Mt0F?a&CkHkFtcB;itReGcN;8*5W=zy#>^R=k|gsyiQN#Vl;#P|a+Uyz)0D9Q
z$Z{vW^u}`SPH9gl4a&DYS8JLNd*)Z*nSY{eg@Mg(u&t6HmBmV_GZc6bMkrZoL=%Z_
xkQLMl&Da2qw_a!jE(hRsEvzf@OW0|3)ck6q@Yg?ip67YrJph!{X}|yo002}lTEqYV

literal 195
zcmV;!06hO6iwFQN#k*Yq1MSaC3c@fD2H>uHia9|^+az5Jb>Tt~=><|dwegYEBn5kW
z`v6_3t0F?a&CkHkFtcB;^6ffucNZ*%5W*>iF*DAlB*{Ea;$etW#&|{($~cfXO?ego
zS?;8l&N^X+Q<@Vhi|TD}sMYm{J@YH@%s){y(!yrfyGAQe+F_~n8FD;GE43;NqK#zR
xs}d@Wc5HybTPF>IP(6sMl5Qx9OV}EHG~#Nb@Yg?ip67YrJpdG^W*7hn002fPR}BCF

diff --git a/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf b/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf
index 3038a87577964dc6a4c39ea35e7b32959395c399..233d1821a1f00d71edd4ec85cae6ee08c788ba55 100644
GIT binary patch
delta 112678
zcmZs>V{@Plv~3&PR>$txwr$%Tcbs&RC$??d=-9TCj&0kz@2*|Mz=%|9^L8dY=b
zHK3&Kpu~}bv9hGjK|;{PE5ZX64NQZ95b0TzTXHI@youF=+~!VsG&*|>;b=5TI%gU^r5cun(`{jueSH7i`Aj2esn8CPx>`4-@N;iw1Mu%H^X
zO?gg3J3rF#=vdx(FJF-5ThVnXN`CwH%vklrqwL%<`rPwD_bs1r^YPPs{Vh_;t9srs
z@i22E-{s*N|1rNK0#4bCmTakI7$#jiDtMOVaLVw#Z;4!^X20h+qz+X`2NbN5#;q#Fb1-)4&sDiw$uN|}FY?mnet!?{GeVQ@`BdJ<(0pHISQab|
ztZ;oFdyNPVH}}vOQ3g*oT4lm_j_D+RM|#T{f#6^Qc$>5CLO|lUwix?o>ApiTc*A(g
z*f&ouhiM5f`+kN?!jhc+Rv+W@^PV>5S0Ei}#`sXh)FBtM^J#
znt>UPGAVS|Jprte0&XMsIRr21zGwLal`-&o>E
z4@Ao?>5lk{6ZuDDCx!!i)5jK7?sB%GH!z)--9&iqTp?Y@>DT2%9(N}*C!78*(uu8W
zqpkEK^{+Wrs;|e#H;(kS_TmNF8X=J-t&NgluqLLLW^z9j8SX`5dRAi2!q!ZI5(mPrUWfc?2M?4;FB`K)&F7Y4hU6+9olb0tVf)1ZmA4?ay5YFmyDq}
z1bJUF7GN2n8O;YD8oVgy*W0hJ9F6bK%Y90>YCG
z3v&^^1oDh9F;F!d5UTX9Yg~Mwc@hnHG)Wm79<|*oK|Tqq66MkctK3CkoxJk;Q$Yn;
z-c-;`S>ub|bP`TUYD5FZtC*U#{=rWC#h877zMBA>4k)p9orS=SJjJFcAL=)5LGn|q)vMc{*d6P!qiV%
zx_R+U@*r(=&}_f+RHU7inV&l1o$0G8%VTIAQ%l{K*m=6q0hx)xi0llTfNIap0dVZ)
zarkQ?8cC+Xwk`XgjFjSC?iBTCS{0xce|F*3
zOblbZfbC842AvwF#(t|2xZmcmI^xfy6#6wpx==yO3#Squ{+#c1dV&m2P6!UK6McVs
zVEA;U--rJ5V$t@6^Qc8J;ir)wlWR_QR9^;3k`(Q4;*}gYyoisO^3(_Ttv2g}!)J1N
z+OuG$c)7b|$mVPs!Pq8uX*&O(wDAU>n0ewBJ4eUuJ)LeU;aV9Jz(YbcUDNdmDK{pw
z@v1M@L;A?9C80rA-pIc8z_vS_2sUI1sH^SNT+XSX(&^)u`W<<<1FYzxk}~mD!>96I
zZ;$-TKMWnA?eQ;`hu+g$39&qR?o{|MaP`JqS)(U(@M8fexMU)erB&b~YhpB_pE5zA
zkyO_9ul!{(Q2R0uY|yPV%j9WBHixGP?0YMk;o+nnE2pYY=0=+-W9V9c`i$5xRK0ovMu2~sPAx`b5Uo?IGbASH4)k0R^Cqkikp%>;{5L6#jld?WXL1EL&#=g}
z>^l#km;YfE{Azl;uGLMHP*leA@kqV9e}8wszGgv%jZf0CoclB-FOrip5NqR*rgU{N
zloT8x%IL(DsnN8P{(gV8
z`FE)d50L!an+cs9BzCB}xoo@4Jt^eR9qB`hr@r@7#9WqlMfuI!B&rzXX}jY%vT(QX
zMt^L|zXQwoX;TkN+nM(Lr4S+HoR6MvF~`n--fv2BqiMXap)61cV`
zo`C=37qyl@ySIq0a$uDnU!((MGMrLv`0}Edr%O?5M>pr
zVHNt@4P@lw;sWxlfXq7HlDfH+k%~dVlK!InhYz|Dy
z)Av%z2J8z^>Z|H?c=!weFE8K#e)=1SGnr4Ha7D8ziA!bo?lq@gz_A
zAl^B#n|VDA8U^HqYSC$I9&Qm%KSV1%_DL$8;{Mw7zaf-gOF{%&fRw0c+}wPnX=_#;
z3?a4R2>bTZyA)(5i5swG>kb9y(Og(f+;HiDa?}N;={KKFE&UA1UaMaqIj<32S73}#
zHvGir+%<0jhe}H6juzoNtDovY%F5IpSHtfUAHxlY?8&d4^H`H
zy|kCm03;;727n!7DOt;CA5+>?T>*;fM=(ry-=tL6V2XhhAPrSBp=8pSkw%>Aa;LTO}(9!?JVZ7{k2Zv
zf*7d(b0@^m`M)QxGgl|R)o%uG*%pNMEe0ru{t8G4YQRvlX{~QXGdbSB5W&435M0m;
zv`jy*w?O<1!DRGC%P5|WmMkJAWaL*4(1zu1(FK7jzk;8}(T1=G_ut1c%In)TYB4iH
z8fCS&`Le*&lP}S|kIvQacU+C%2W1~@n_sKh=UNoq{y2RRkY-(L$!i~lLqUgFe~o{50!dWGeGFjcIR_N|Rs}K+A2Djlz1}!C
z-tJrx8y9dGG+9&{XpJqhA6P`({850nq!^%sn-y|Ov;3Cpo~-uk}1IU2d*YG)`uRUQG-W+P#<9jJm5C)iwTV#Xdq4X%?`b
z*~Pn1+x?x%1H~r0IilWV*(1mSLzWe!fRlJ%h8vf7?4vFvBQMtwL%$MB#RgY&gpLAtfp7-+vyI_t+D2VwlRRZ|f
zqGGeq&Nd;4z$rdVLpe%k-zN4As+=6nXj(9}DXuh&
zzPudk>LO7nIMXxw)U}QsNtU3Vf?Qi~KK8#hzd=K~TCNhKP^rO~nf~`R=}$IO>kzL~3K6c!&4H})S=ml$;p^V%+$ogwbQ3u=oWx&BTVRhyiYp(q
z5#_*V&k$O?UJ!A5dVGC89aXxk%^)yGl&@ZXMkQqJstOdcj&sjdq|uba?mM}4WPc`o
zHfFPC^G{=o9IaLisp513CN+1?6V0b@pT-J}=GT^Qt}lOU39Y@ezrEi#HLtaphw+pe
z`r}AbX~uz2MC+2Uxz0L#ws@IotRxsRiWNKnW5I-15n3Lr1qEZxc=bQC{g2AU
zZtL?g%hl_vZuV%4EnoW;Uuv~`8$1^)mzC;W7wbN~9Y0fzr8l+Md>hC9Jq~#NPGRQb
zhHubaTCX@IPH+kYs#QcKWMRq~GehNli4h&4d;t
zDpE)G8rb+3`A6L@kCx(>TL+RMp$-+8UwH>e%j5O6GXnt;kk)a()__710M9_>^cl6|
zOo*)~)IGR=V0q8=N*aqkk0PTg@aDSu2Qu55ic@UXNN2shyyfw`%7YtOrhn(1sJlL8
z0FrO+U1DEwV>PuDwtxmZh(*>_nC%dDRWEAP0Hy#VmmY^X_;$iK%P$oj=?TjyY_9HN
zAjqN@EzLO>_!gn6Q&xZ@s3xOU9VND|E*r2IvP@k9#limxRVODlLb)RaHWC9LVC{)C
zz!Z!*#-DP;eRD8~lgu(CDHcdGY=ME+Z_;w+!*CB93GS>T;jOBJT0}kH$a^O)X+$3H
zK}7GBdyKR>#o5A1lepf
zq#nhB$L%CVHn?rAK{?0?PqijsjhZfkSU~CjImqjW79vF=80wLZU8^vd$poL&AD#*+
zd!8T)h@0u+^N$br5F@-eMhOvhzu&M5guMBL-;V{ubvC){H+{3EvWPx_Qm2xxM8~@J
z)P#bc;e1g(470_^qv~5kAQC8Mj6jn6`MFUutFV-+Pa8Ll_NImWf!zPhl>b5AKWFbh
z8Mkmzf6IDk-{LTFnAK?eIjl(HP4twPZNa4nFdM#}cC^zxEBbDb$UHO12$Z6DQ7cTl
zq&xSkalL*o2qk>x)4W>;V#uyxM3Phs&l)lfWt*g-jaD^x1jZg5LjBw{+uRWSD+O}a
zA%9qE8}@h}A5^6Ww?k#J_HeMPmK2=;sZa=7chOA?
zybpBWH8LmmGZm3t7Q5&OkweO&W`!4mtIIhh;Ahy&u+o~KYZ0oT#evb-uTv_)1YTB#
zQYm)2gPf^ibUq(4Dj0lKM-gTB8GY5KBP@&H2Ovx)qD8|_EYCDB8j~^NXk>)(&0dXS
za9|T&yRfG4y~|NMvI+ndDTn#+9o44*h&>9V3~p++X~6|}Tp0B12X^Je!$fz*2NnL5
z(eD^go5|AG!d_TvF*)MEWnH~)+8xr^AX^=GE~9?#@QV5)ss1RYZ+Yy$o^Ly^w6Jg
zr^Wi@Pi!BoBNr$B23?mdiKInNk^Vl{K*l{T-t{k&8o9=BngFdv?r=zUxV
zeZ$L-(O15l92DM_b?`EsjUQ*g)Hu;sTJR&!tcK(hj`>3?8Sy0;rMugkG5K#@9Xs`7
zMUn_1Z3ubp2+u24tK%XK;U{ukLXJa$)^_^dicS1)s1{@=w{<+hHIIaKLlKyXMkoE`
zN+#QWBTZh@u*fn&+&lkhR0?hp{hLh75eOMv_dIPR4p4YxdlJ4$oFA1ycdbd8tHLiy
zWgTXsldK{4^J@t
ziTLbrb@OT6AYES$d=f)%ysD^~gJ;jwUJeXn=a#wzVl>?NF@A8o>WnCAW?Fl082`@x
zjT(~`iGRh-^#nEzOQ4+1ruB-Snl#^6TYFe+9|PX
z12iRjM$^&Kpl+$*LU=+>L-YJ|c}<$)?oL?Iy43daFdtqUBI!KvweG7yDx;pN)-_2V
zzY^)?Z&2Z-V!(e3JF7wD*VU2cQ4NYXX=@l5i+N~p&JRt82O+`oMb
z5-_rb{G%d(UByOCI8}Z065o6#)oXR+^3cHn-#qX@5Ps
zbF8QH?-HrJ>@MO)8QHVUw*nbRIg^r7y%U$=^n7sF>+dpSm;o)MtxPtxI>#GG0)Hm;0hAc2)043!W;fZpz^R%pzfZV~mU$Uh$bTre|jv
z<>!o+XEQDd89bS6u9+cUw>+g0sObQfs$jsINXc|}P(3R$Mf7^3t+pEAO?~k)
zzSMJ-mSSQ~AS%hsRbtk5Jl=c)w3FpK$(8lIyA8}H;`88?iWEqo
zXNSH4>iNB)^_rD^lRgyZI+#-xoAv>7sc{PfbG%_&_@-Ag)In|
z5wrN4|C%ow?)BrmsQ=l;7HZ2kGh2~_r?6>+->%)Re3qM-7CzN*S8Op8+WRvz3~aeA
zFYCGkc2NXqXc`Lr2YT;G)=DM;#M`R?HPvjl!i
z+EFVz>O{Hr{Eny~=0$LxO_O^YvY5S}PAk#qMjrRp5c?esTkkZ>RRhlY_E*>)@~d33
zLkXKtjV=CFqVevah5>@E|_`Ar0JQkTo2`l4p_RfOn#
zONI(+8nL=d$8U9nXT#fT>|csc9^b{G3avbD{XUWhB4Q$XN;<`5z&a4|BCT7BV;`tI
zKaX*jZg%eBZ|fkc2yqLl5|&@m5t4|Xg>;9j;#C64eu17Pb+A0v&6$sQq<_g~+?Wlp
z`XfG(s&_LON9KWDEmPT^>eF=$dmDcrIQ4t`tT{QS;Coq_PGs>F1q-`X4J+I0&FEGR
z)e|9V%Hc5xnznX2Mz^H8jvW*yJ(4eYN|QKFVbM@UA?YWwh8!pE+(I^}hVn(TZn((@
z@NBKF=bRi1#|@!WT`Y+omok(xfHeXBG9*4*Dng71
zmj2?6UzXdp6Z@fJ4+5HOb|*29yMht8R
z0mnh&`wwbR_S8CFP#U0EN7i|N7`6Lc^9HNVq(7|>1bpx?FuP&Dm6pb`nznTO36xYA
zO&cxlHcsQ~@ztvx8iHEs+>CM0h$+_B_tURn$r!Wx%ME+%dimKs7pT0TMZQ-nt$hEa
zg}XU`sVv3$+{w(SZxMdzz^#R8JpVgj4>DtCT&&3P(e
z@lT!R364*UH1}oFt)%Ev{ZWk8UGJ$c&Yx3}^9vg+
z(Mahrs}X4UukQ^fC6Z}7dP_@fZ~7b)G?XtRvD-J3&Pzn;FyJKi((RHYzkpTI
zsxqee<*Xgfe+5O&7a^)M1?;GV$m*)@yI!WYc;3rZO%#W#dEQQdu1W;SnC6e`lS=1P
zO2onWNAvo9PsuQ4a>TH*v(FzgD}4zKC9Lli$(CjEE`VF4Q+59e%kHLWeQ~6G4s0+^
z>9Fi}(eK$s7RkS}wH>R765V(?9^~vPit4N`rPw(YZaLkC1uIFsIj$*(=1I8tAAhHl
zdOGZ-&}Tjz>8v;-FDXTp3GpZ@PGaq(p~ch7Kso2=%clZsk;K)}TfF-1%vQOLx~^a2
zFG8RGkN`%bB6)oV{|F`&N`~>PHTh;DqsM)i=J5}x+7RW(7kSaAhk|Z-nRhob9k1lL
z4$9vPJ56(2rjI=>*24~L5{{L|ABY>NeCb|F&++HC$kwqh?JK)&rG2kZLud~qSJ^+c
z(J13@bBJTquDXedA|NdN;LKLUFDIi3yTGix-2ffmU1&-7VHe%msm4ILoR@n07niwP
zTe75$u{Qai>+hz@_ec!`R1zv9bL^ZPNa%^IoKvEj&5Gybh!Hu}n7RVr8m4V<_vCX(
zeVTE9(#-vuems55BC*TPc}36p*$&!uhkp_GmZ=9c{$27Gw9%~0Ga7Wwxjh&FIbxLR
z!3T8Goznugn^D3BThG@EcnGpR@=&m)BT{xYf04lwLV@X42cjDzbKIylUQcAiI#Pxk
zDuRP4$~HWtiK&qaz})q)pd|;@qx~d0YbWQWUyE=I*tt;q8=4SGV9vC{Vc9UI+C=}R
zawHXlq?M1Jf<#FaQL}c3;-NZPmiPnln-I{6{dtw3xop4|1mZe3`|~}_@4X!Db2P*y
zIixL`buBI^{aer^fjD6@y&O{KjDagB5q41wWo$CX}|!oEd&W71C=gS!l$v9?hhQ#HVw
zf-+GQMER@`!B0)^E|ofb!!06^kD@G`GeX~`gYxIjZMZ2(TV3l%u#j9ox4*HScyE8L
zwv)%>&Q0CLpK0_g$I-6|ERqIyq$&7FUsTvAaV_a2Y2zwdDzEQHr5+x4E>|*9#TK}Q94e4UO_lyb
z%$
ztq@1VH}W|`hZe!oFjT)DM?OqD->XO++hqR1VAJA^!~v@AuYBdmc`nimawUdpDB#hijb#F0R#8gZIcX_hxdRO92ZiX9`Wv&
zj(e3}6P$`Ry=`3}wr#Yk#R2~A%&f&SrG;`3a+B-g@>huq&_jxMom&`CTGT--#)b4L
zAz@ezBQD^B&incsaFOMXeqc5K9DX|BXx6oY%xI@O3n>TLMVQXK>Tc_Kznoq#lEA1-
z*BMGjVok1YKJLf`qBVFs$$KJE8FDt~S90DFm@&jyRM|Vu<
zBAk>_{&V=zM&uFSEsp&B$RESnQRcn1djGPqi%bWs7GG1XcLjp;$2Aw6v92dXf)T99
zT;b>|v1-V;55Ky5=&}k$v&nt?p!2D8a2+E(f*~p8h))sob$B|c{w133DN&qSH%KvV
z1Yc182={Dp)Qp(ao&m1zNTVzjatq>=m86O~|Lbu;@vyN)bkDW9KQXwRl*KQprX5tj
zt2Ftgm-B*?rr*PGBYhctg@GBI(W>}9=2@)$BZYEhWX_qO0x-?L_6}5stVM$ldqZDH
z)tih|$d@2Wf0`dvbkC!D4wx1{QU6_QsC?hnHJ8UXRKJ{2m4M0w$%{W?{1@6pohIwiugopzn@E~gvsrW64uHjbX(t}T@E9tNnf??pp7xbA
zn%huM0ZWomhFY8be!;#-L-ih-6h~{tHq!yIX{y{54vwUP-tDkNSfA)Vh_QfyUXE6B
zQ)MQs$P7qGjB~fQZ)#^tz^@eFx#yw0y);K?)f()5#P&F0%Y3nCN7PAbtn%Pfwn-C`
zqUcjp1DLl{$gDq^#GQBw4@pOG7ZtU_(9Ov1`iKK(1TwRYOcLs_ETXL>FYe3Fhk|3iRG~haFMA&cp)e8U8Dmb)h5|Ul3-GhI9
zYm6?3+V@@i#H4sdE+A%Acx{>M=-tJVy0>g+>#?p9T<%=CS`MwO1_Ky0OF;v5e
z689Z*H}ig@lboI7&2~B!k^b1*()i=*F*5C*ixmIbN_O6$z=&OC3cP(E5Fyf(#LP)|
ze3~mOTIjWY^19dF9i~o`Lhr8~1;ISLtZ3+ZE^vMv)yE~IQ9keZ3uO#TKJSfjA0EJ9
z1Mr(C=`uiovKJRNd#>}um(s9VHxWml-eNu3k&}aYqM42t6^Hqc1n&61
zjNe_h_=Nm_#pw;ff<{GY)Kszi!_oIA{&99U{i1m7x?Y&&gdLhvaKiJgnuu8+Y`Z2o
z9xp|Oe2!IkTH10IqiJFhD*U5O=Vb}QR=KFP=V-^8TWK
zP0vTQz2M^FIY#BdwU5I|6lv<1<;XYab7?Q+M()*JIUezjc{YW=P1(o;CIjpN>s|`4
zo7Cq7dO|DwYQ0T+i77lakZCnaKe_)MZvp;OHu1G4KauV_Bg%Tj{xUh8*fv8IzfF47-i=$vu9ZSdTV@i&kyeg5Pl)HQd9m*Eb(@kx9`+{o-IP*!En85dmeU+<3Gft9!;y
zIKg_*Z0_N$NhoAURhJ1N$mOl{6^Kbx?Z8ReLs*em^o+|1nhyz3b307Qs6LAI1haQTp6m;hN$4ZDQwlB-5beefmjKeqch?Nrt&sr-Yp_
z;~GvRP|=A5q4O0uoWYL&gEAkfEN^ABxPniIP
z%EssWX{qs&M1)!&_2y{w-?q^WvoF}A?|C5E9eK?{TK9^s$gh%!>)!mE~VjTc66PtH2L
zHW*djFL^c$w3RG>tjX}f3QC!UD`nvCX}`rf_5(L`td_buZ7_3HMa_o?&AC4JQf!Ql
zXw7X?hwbgfRgOX6s!$|k7a_wVk~)_yXlS<&L~J^QY!0kVe0<*Mwo3W0e{@AG#vwGa
z&Si@cld>FA5{VB@!YVCs|Bh6&PD^h~)SavXBu9%r^zy|2c^pM2Y4ls%`}hI|yS`qn
zENGo0li2(O8T&ag+fDMz{C>~Ei-eUvs7s~Pe}7BB6i}|@pLP!uiA70$?p5LA<=mCc
zQJ$J)gN!g!zS=;}-JV_vQ*x~|&Qp8t5L3poKR2X9`}{^R%4;IG3}d-Ny-~z$SZPw5
z4qq%kfvoh>==Z?4%v0b!db^NA{29Eg~EX5CQ9>3EVZ
z+Zy>}V^jI|G$f)*Ao7+c(?{L3;kkRnrOS2?NJ=1M6K$jamZw;Q83lf$j~8!GiR@n%
zrX)fl!`)$AaLhRDu-VsJyCKAcsIJ|~Nf(4k7ckD)8jg7uY^fF~W#Iu_;us@jb1m36
zH95m4dNyzzq|36pQ>_gqjv=96!K!GI8MrKX?=?AC0v-eXRNWOcrz-iKp`icX1xw0)
z8T+05HE^I?ES+>t;#Y2B<6et^>%N7>EZi?Xk$scm8%R~vzWKP?FSeTCkDsA0tKHCK
zeg`YJT}J_7Vo&f$?q&tqQhkSbG++^N4*;fVeoYj&
zG>0*^l-IPBi3{-Vd3-#o}
z858IIh8+pyDn#Zbn!t?P2!qp50&;)cE_#`i6F~`Fq3TPOp+LY|AT|S;0xZxRmW?U3
z-vU_%-VeWzXwn4fgw3!Y0bG5?#$qqm9QYN8Ejd_y4^t54WtE8@zx|vh6<-nta?(;d
z$M6Eu3TeJpvIV&F6^E-bgAR^?GPzLXPVDUUL;G&ZGmUU_WABB;m-OSOhSM2Vp?jm8
zNWeK%XXKERAo9Q%f?F&6C>pfKH#tik{5lwf3@Geg*S#FMssAq9GaBSAbO+nEBh{kW8cIJx4rWlZu}
zWLcB^e*G*+R9z59D6j7jxXp%`%s=|oA{w{a41ORA>pw5gYrq~DY(3>9RDC8+1R=n0
zx1j_!Dg7!2)=RP`L{o1Yv7Bn%N|bH}W#1Y|$#iBbBQZdOk!ucLar85
zju_&f)u^)K$8M?T6j{7Kp>6F+4ty%6o#7=zQqiq*RNRS_X4OxtB*gW6%IlR8FuZ6_
zMaFem#lXE4cFrmyahUK*W8d}g9t^683ua*Rqb7IiM*e_zN;Gh$j_*k~l!U4gHzUk*
z4n1fVhAEfkca~VkRP~`Jo*9aRO
zQ2)JR)=Jy0jGN?^Pla<|e!q}u=JE5S-ol4y5CWIp@t)HlNIGjB&uq)Nu64zGYr}@L
zUUUBd=={Ddfy7V)uNmzqC{Zi4lLPV|&~B3tp%x_mHrn(_>r4l66u@R?MIaL-a)eZ5
z(_$G0n$U3n>`&n+hX;Afr{2evT*22^B#UdO2nvcoDLxUUoEJV}5{ex@_8=?JVCr~S
zDJy$5_^Fe!Ukrkv+o`98yy-y?vpSp!8QWKCcSk77tcOJg1P1oMc#8>PwE`Bu
zSVR7xi_*fOn+9-3J>0~dP8~OV3vCsLk|wUNx8%makNus?Ov>@Dy0BT|CE^;ISIh#d
z-y23@*-qb?>E!17J7m8R;05%tk0dc*JNbeh1Wum(BOvkpH5^kWTmJz;C|~}^-hkWX
z6XX5He@YHrKu6`+c5I1-rcq9c6EVof(bj3V2(Bf9*1+={k6}M*u)y#Kto@FzaO@IV
z5D|_9Zl_dP*#O*uV>Wo8`wlf*#hK4*ddQTMhIAADt>}y7iG-9?VGt0JdKR>ZA|D)b
zq<4AJ4lAAYM(J2izUU){BZtJgR~MBzx`G2r;_*hGH1sXDwa^md1bV&9+;AS6F}vZy
znhlrT@D<6LSL!hUNjUM?X|?KO>}|cKwI(|`rq6DtUN~^`9hvi8do%VruaVCkW|vF!
zW;jE#FG-~f-@@!}JuaR-A(#~siSnUJx_#imJ;q>VsS0>|-5Xw`))xKWB+2>zB1smG
z|45Psn9|lwT>non**5Hxw5AjZ<2{KY((9P!`o$C7wctiM2Clg$GbxiKqoVBq_z38Y
zQncz-FDtP6ppu~Ua&x6#dxJEZC7O-3U~7AP9p3Gow?t7UCFsed=owHYITs{LCoKm3
zPS>kR%$;BED6p%{Wo6J1*ZgrdlwgzYCa>)a9A>T$1S~voJP>^}@?TX51gh5sX
zbk;)txl`XXQp!?aK~7_^*l43S?tu8HN{?yTN+QKmZtZ{OJh8_<_d#^x;nOaBq1osK
z81ch$!F1*Gr||!yv%HURcHh5yjy!wS(G4`9Zvwhf*;pN_tG?#$4xORt8mIYNo2x00
zs_G8j_y1l;Th1HwLP0~{{W*=Fny!J0H4iCWH{agVl^NKHHl?d7@{7HE=W=3hpqwE$8T#Q`*Sp5iH{8Oi4fOdl8gp1a|A57;g@BwD|hB8rWq*5
zA77YXx3QcKGcVE1@>Vr_?6JmR7Ya^O(b>e(5}pl3Vh1Ge&z!8?_~#Qr!&U%Z?WUd2
zQ{7*hSozhICf8U+oqxx)vijG*9Dv#^`Txv^8_&pNbzP1yh~#!}yZsj9Ylcce5=3*P
zuJovR$98&J2~TP&XIgli+4=Ec-b^P`=7Xpm0!mcR&e3<=MjO)P2Mg@J3&pY#zcs$&
z+xZquzjHYZN8#4l+pM0MGE-p3;FiyoT(U1r^U8-R@OspQrnNqQQjeM!{;4m2VcMNU
zRoCHA-_`VeLf?FDkDSFZWCxo6o4pSQR+W0@;+PQKe~c-hL~IB>ks}9;ikk})J=$xjk?^s03xn@fRL9vz_Q
za^i;6_HuqW1pKEZ+5onClq!w|5b_%Dk-5HO3G7=wp#@XQO%9u*{9MrSYj5_c-mNdp
zDC4&^Z^d*sN{18zP+W~A2PT~~;w(ZePkH{{_A5P>Sh_L=z51yLFy)_hvBpgZZ2@>g
zKL4bw6N(RWdB*XkB_e;H)F#XkbDU#juBH>+_)1`5NJhn)UjmjUy?(u8K?N2@b@=R|
zGzosqkNt$gC<;+2pJ_9p2xWE$fy5a{h^_eE_(7YK)^w6@^Q8|yzX{2_e9384N@iyA
z%cxVXgbt>QT5b<0Grl_I{$Oj1OjEb}<{&Lji^
z2SR=*scMQxVs7>HThJ%S>b&l85W)_4Okbh)O4Kc$NI*DsVMaEz`&ammJt3(3Jw$6@
z8%%Ej$d9o_?1Dc%DN|BkrTrZ*U~(P>nOVDe5^z!;}Nq3@Q5OQO37*wg_i7jNCA
zi(gn?^+i4}M$4Mx6tW!(oel9b-o_$Y5@sqt7J`TK{1R@v#_vurPltK6-
ziwmNaJ24D*jBg4IWDqs4Knr7aXExPk3lVQbB(AS2#~DHTjmKR4t2;d-;Xg@#
z=jt|%S*ez=2QL~S1i&&_NH1QyIr3}aV{Mk=3bPZY%zd~%Q~NK
z8clUxZ$y7Fdx~kFU7;Pvn+yu}I821!zpkBsfD1tI@M6V$+i5PI?*>X3vV5@W1?HR8{P
zhsVlic(S^m&og_Mlageeo`+B5R&U)KJ8Fwaj7G@6E|0kbelX(o%MD8vk9b~yGmt>Kk5_l-)IOAzToG|l`3wVa$s#nDFseD
zIQcfTW1oV*DhueBcQrbm^z{XyON;l*@m;R2yG}npJu&w=?RD%#82&1)Oe2_(vAxweC&zpCjnylTXv
zssamyg8Pj*FYejEedqPxW9n+@bv^bUP0D{okZa*(LL?op@Gs7&MBnFwGg0N;q8|v5?bwET0GV6d6+t<}s}d8WX4}kTiJZx8Jp{7(Dyr
z=0EUDN5j?qymq_R+g-V~S#>AFF|q!yZCi{p;k3?ib5*+&pHYvxC+wG{fE2wze#s_pV?!BA4y{u_5%uaQJsR`=
zxH%NSLZR9`p94dO{k}ECkYI&LG54NG?!7&!XwZeux0iMJi?-m^b+Fh>!ZKa~5uGLa
zXyQbc&oH%-_1FB%$8n@~%^vqBp=uV`Po6co(KtPIQoDOL4k|9!^4cYgaX0Tb`w(
zfckA_0b{BWS~7^u+t!rI5A19^PJYm{`DD9gjmwAwu@zVCRz#1Tq~<_K$1Vm(XHIl(
zVULX&54|nVBzxl77E7Tjup{c%oYk%a31<6%O#uFw=TFmbiijdl@1ABen2iVzZ3e$!
z{gl65hN(Mu40)smFa82TIFJw^;}*Deo$GwnDa1ay&_!d5w-(I9uAGgG5l6
zt!vA|@RS~PzUB^4%Drr#(c)-RaP-uP9!miuo}C4o>~#FBy>u`x!FVw=E=7zPtMtwA
z4?yLlC=ci;rlz`9fONn0+*_Vgu-%&<$4;Or=T>79fk;%MsbLqJu?RG_e=^fvac>xq
zp|!ulmL*Nx>HP7)?K>>wJ-P{AQ{}+V8%Bfi;+Z3IBxkmI%NMNT$g|B~mjl`(ZKS9R2B!Arn40xwKd3(NiTRL&$;(
zPB79+L*X!+5i>C1PQp!;hB>vf$K?+yWWHS{mp^Kh8{91Mo-!VRRcO+Uok#aGMH6@g
z2`NARm+yVvTJTW^4Y{#pr?$^||yH*IcoPwQQ{)
zj8p$2=Qy*u(g!k>
z9Tq;4J%h>&QeU0d^2DCnp0IE02k_2A9zcWXJ=an5xRuIM7Y{cNa(KxPbLc
zZI~W{0N0w3PIM}8|FfiKSK}S}x*$i$o~!Q9VIL73ln~;rA2FKu`cK4d!y_sn*S1NT
zwL~6|CQ&qeqCni|?obzvewWHiA>jOq-}(nES-OTtT?B*TSHYG|3-(;ZDWC>c4Ei@vFJ`HY$=lT=3#0>}UE=ZvEe#`9bWop0b_4B07y5
zW#8o})CrY*Dne=$!erD)N?GFGr|6;mBV}}7vnB%&EtVgBBl$`^(Sm<2nNlD^DVcFI
zdVb@CnZm-oc;b>qAD4qFQv##XFX_Vd-xzplalL#K?$mpm@+#IzDt5(krDL=%C^LKT
zGGBE2x01P8w5WTy$yDLRis9Lz6}#*7HxEQ4&Yme0GYbg5bAm#uqC1ybPV>wb^9iKT
zu+yR#J#wBZ-?sgSG%hWY+Ab>pK0t=_g{M}kyw|QRm6n~lWwWUak^tCiB54;zofzrC
zB@*xpG@rgAQ5_LH;X6ey8yk(+G?^n9&fnb9_spq9AvatCD*qo_-xQk07jys7b}X95uBi>qi^UUH=K}-+tMqYU+v!EomL6c=Wiu)59=r>
z-);*GU%p~`R~Op7DEM#)vJCBxXV=ruxTIE*3{ks9#uiIaZUpIfz@OD-y;HF}S@-cxLf1vS_yy?kn2r4Cg`J
z7RSGY6+9^){L_MrA1R0Ao*Fi%LQ8)eZWO#9dU1XdZ!Krt*Hg8L!eu3kDUpEvNFvQ+
zUGKpmjRV
z>m_4rV{yWn*>OpS101o~Q@=PjCBgUq9OPE|SW8k15;hs?dW(PxLrjS@i}4-fST1b&ASTn%SLlIk#%2pRy;!fNfMjUNMBl5AUxg0MT^k03l0viZWy
zS4S!RfAN!>`9H!^E>8CU0%x?QOzb8HQqPITD;21fxc8|G%mRf@k&u4@dz|P(p!(m9
z0qq3R@%5wK7u=b#lqUJi#IgOjtQT+9rj6QyWxHCEx97f>F@TSj$`H}+Z?lh&
z3$NswdnN_v14by)=*hz5j-5XZOpTh=4wwk-gVYWfczW~f{LyN4PJfm)Y(!b%rLq!(
zL!0Hh#fVbr*Q92gxK*cryCc@5AP?>Od9BvCVDND|P6twU_$;%f%FQC>%u*ibffebE
zRV6Y_qECJvU%vYnJpj%v`IT0W4A0S@L%ao4k2g;J5N2f+;Y9y2Q6L-iO#{E@a$Ro8*QC9Cc>QP-WE3$8XnNeF1RO)L1nR-4R$d~-y@d8CThqDq5Fo497?@K!
z)aZx`h4cbk0f6>bs|Afuj;PJFzjbFZgcMuZ3)g@rnnAYP;_hmDKG|B$$kN0<!UTAY~H^zkw}2q1DoCe7?&%}c;_>J{ya)b`p0
zl^$A{6-?BMcC=dN&%`MgRAEdiRk7d4B6~gx(3+|nz6>~7q=+dn=q!ObAVn9s6nBPJ
zZ!dD}Zuk{&eh
zVO7Zg1yF`hFFf(9bRAo>c_JpwWOLaJg%U+4hrFe6NSSp&jdb8ZgQQG|m>ya4s^kBeZ_9|cw^MEW;08ZoW&%VLdy
zjv#LqSFe?E9xjy7G_|+oTas}gq@qarf3fzb13m#fo0^9e^ZQxA5ewBRhW)wxFk{!e
zv0g-BRxlOJ-2Izlt#tdo?yYfIS;X+M9$BPWeh@m)<1e=t>5M=FRJG9qX+UVu-r!mFe{vPGv09fHPQ$_kexPnUR84x||0j=2b^Qr*?2Q|09kG
zAPq7dfg0A>Q0pueHwp45&G3@Z^YP3?Gcp_>EqolAMv?ci?_@9Fsz4_cxmhMxtka{@F7sH<4Y_PLE*p=!v7%@!c(94=Of^137!$jN(
z5FY{C8PSf9vbjJN)(~aND=e_j1;Kc1k@aW@eCYTUXLylCtIFPsz~L~a(OxjsbcB
zK4u}{!f=uE%l$l<)?V=T0MTt0t{K3qk^k9kfsED&N?YxQUdGe#{TExhz8{u>pZ`Yj
zQfT4jPT_RGu}1{)Z`0TN;_#AMfXcspm*4vHsqy++G$h*g(C^CcrhLayv85-q>qnLp
z4U`>Hxog0I$HH)s!7Ig<1w91e)}3^t`DECvax3NmjNY%n;w|PxaJWAaIq^6O!6t|r
zRZ{w1mJL>S6RvGYDdPxGGTmVLtpxI(5cNH!z1HpUo=@s;WA7r%BWm~ou?mj<7+5au
zYYT>)@~@Nm8qW(Ck`T4`@FO1;cZ_rbG0#l}V$|my>l}Bdxq6$7zUWDq-QMUpjOe6U
z=7VUET|JzZ?-vGDc|a=MyLa5mXGW8VKOBJE4`?5@e8$4*4-_k?}F?pt|%MM*C{
zXA!2kAt^va%YjPO|^%uwL`8_I!KwlRvA~
zY1yeNVk)GHmDoD9F}0A2Gn~}`H(l-(=1Bzepy)WE9gRnJNG~Pn|3+$SwkqCn&vvx*
zHHp68dN@QwLj*u+x$^zs4I{%?JPeh7jDfTZ<{phN!*>lpFhXJ-U>KOgZQ5)!*Fl6FSD$
zAJ6ZLuA9TsE?qn6#;$;H>qI+%qh7Iic`B&9ecwY1%D8#;?sX9r3ahqt(s*ih84B<3
zaq*H<2-?`vr3(t-j^x&%q+P+Z%mZ3Flm67(*+Ov=!C&)xQYM7MW9htfSlreove
zzP@*HjD?Iu#J_ipw@@-9T4KwO2A8(((REd>&IsstOGkLu#_g?IS9bou10yOl6b
zqGD%95DW3_rj{4BxxDuv8k9NX&4&ui*Gd#le2WTy&b|_8XM{maw>vfjorj`hcGCt(
zPS@zOv#1r>kP-nio{xM0r{D02Y(*GO4RWEmI`n=^Zi4LEX=x&yoEqcoasgUl4XC*?w*ys9YWtMu2_Um
zkKY%#!fEW5!$ShjLx;r$gka!(!3km!?xhpQLVx~FZw9V|Th;-;$IW{5jQkMfMJW>Ly43X{jfn7t}EL((y2)6hO^YvA#FWE!eiXH??
zq}WB>*0n*F7GfVTI+OW4*@AT1s!W#1{mGHQot2M9q*dJTy)9HBQp#ft{>gW_&7}uH
z1PdlYghZn;KD4*;$FUAnli4B_yHs_uAvGz?&Ka(<5`rIYnpe@^4INx`Ik4tt4T*oX>+!jAS<|~4
zXBW#eB!>YYnL=o@D86^G-nDwp6Kfl0DT1q^b)fhwD>VO)EMIpS>Ozh!mtrgk-F2r5
z+BDBQWv~B;6<6zxRIJ^Xs=AA|I(hi4{Mr}JeRCKU1YQT)5!@0OX<%Vb?$%or2mVnY
za2S!FkePWpuE!=cW^_z8miMhmxEVkFWhcNceJLN{M!@)bP#StViLxfwFv4w?ukTzF
z&R%?K%M^*}#C5wTThv!y_Y4MYokXfQO7tZdH$8qi{xoe*lt_&#?PAuN
zA?&v@J_H1oy_0e37sDT>;QZPn)ZX-CGgTW9ZG@n2I{r`(I}~V&K>R9iJvL-DEq&9~
zTZaRnrkYISyDzi&7vyh_1c8;*G+iH<`CaZ^L6^>{4IYil|B`p$N`-dqKRJgwUQ9Tl
zTZ7yq+<-;W)59e5WQ#!V8qMY!7QP1ep!~s4aE88Pr+vV}8hnr=bUcT6BA32*u$yPE
zd+`*z7w;S<7$&lo8%sxRsVs);4`Jf>Wq}Q
zm3Wjb@|cPEJRWde3Um&mA91Rf$U`3>7D<~ls8-V*Sc!hUYC*lzy=E%LzXtqo$$Mw5izd#_$HAkZS
zao^-Z+KD-6knL~F=(T-OcnfIEaG5!k1H25Cdvei0Xjngh?u#0mxMGhkBe~6th46ET
zVF=6BSFZ@!963e+=7JbMXWS|@nYTsWG!0ZtLW@~ocXWOED59lteW$!9l?7mY|0c(e1(9M^h`OcwhL8%S
zRHS>~SQ^MvKvz)X`>R2Gp*nDSZeY+{jj`{5U4k(;a-##-PkS-f
zR;5W!)+P)T3r2Z$A`FI}GY(A8z^vjzUrqa1{=Ro?obF9chK53sW?ry`3V{u_8k^~-
zJ7dLy%L*S^LV>e)-se0O^A&AbKa*8TLsVgo8gzv@T6EjbT)+)=NZl_(6
znvN4=n2+!6=+T`Wcohc26$E*bGVTRsIkon3efKDj)`F5UMa(^AT|>!!A}YcWz(LcB
zNU_<=1$DbJV2ROTy&}yd4y0sOLH$apg7T7455nbcG@P7x`oeVvZdNgl3QBtBQ(BSB
zqO;@(=q!2<|NA!si0F6f65GgSl>N!;5)j*^qb|n~5M#kOeUd!BeZ0_dTS`yZmJk+U
zzN)q<-cnu*mPTG{s}lM`FAcrFPqTdSM7YmICON?K`&NH3=L_LKsE^6a_Wz-Qvi*;&
zD_Bb__NWE<-|zi=tfhf=ilh?cJ<^*0YPdf*h*tvNn#Z1GmMu6B509(c{e=pq%cFCx
zeo47ed{v~rnnS_h8xp?nJJR?K;_N?Vd)a{~UT84$`R%7?STjN+n_v+IY%)ja$@#6$
z?VToB|6J>6qP(-g-7oOomF;#1K3@t5z)70ndHVaf>o0srJFlEO_lQ8sqT_z&deqwT
zUwUFnndDs^(Q~lDpO4*@J?04aXHN=gKqwz;04Sy>
zJOs=4n*l9-UnBZtICKW|etSI|&l_Cal@+6o1ivUd^|tn&A!ku%H7R24rSa%Y*J#DQ
zBhc5x8zquyFY2#_nbiDv`%cYoBq8<1(7|g?^(iUw5Sm3c;6io(c&hol@-K|D`S+%G
z=R*EXhtP3@ZEr|~k1Z8D>3;zK0)!9M5NdC-*>&Z)wBQm}AQ3BN4aEENR;9v8Ogbb3=#MaU~BMK&Ojq*v1Sn
zm5-J)*bF|7tmhm5#97`&|E6FZkc_m_Mo^0@e1G5-CG$udJ~wEfi(+980;I7p-e$od
z#Mf&1Qf|;wo;gNMhC`CE&sE}dl@j+*@%(9UsY8ZU-peMIXXchoHDSW*%Z7|e?QoJ^
zyH=;`Tu13;7!!wpBAq)JfhvbKCYL~p%w`@UKFV|wTv{w1?gajL1YP`LjYW<9JAaq=
zi#zNj8G5f>U4s87#fMXV5TGkR^}2y*n=ZpUaF9mm4pm|H=qw_E6{&2ZH0=WK&poSNdM~79aBW{5$9Z*k1e*_T*Q%oRICI@pHG3M+=KMN!d
z;D;y3K~dZZV^S>5gQ&~TC~m6xf^U_Ouz!%_$pP$D5$-Y8efKktxa5)Re9U86LwmTw
zWiYQ6!4`ndra9!uB;qjEx#Yr?s<*P-*puYuyp?nM*#pH5ClRSrwSoE*oa
z7C)vj$o8y-u+sbG03Fd+T+oF$((Im>D)&`6*h;FtB7qK=Gvh_g%7^_C(q$>1L{Akflq{vIj?7jQE2t0>5BN8
zh8BNwQoHOEQN?-Y9B#q5$L;om!UxBX+mc;hW8Ow)AldqpsxyY
zQ2>uEpG}jDgtD1~>mWM{X+idMre^g6<)SLwra-iOfV4w<+uD@`qT0tIx1>?;adydi3Yk
zuxrAh4rVX%_i29*-CNX6pYb4SbX4$DGP!uU-DqRNrO?!0TQO=X{P{{yA}339!_<@k
zA76ul>#QI++r)86RlG5q_ks9YLfsRxdY8-G1`L!3kZ4hcsAnCp6QwiD7l?8AOnzN3{%B7_`VFrNB$jrFS#KAA1
z08~%xMydxW6*&D^d?#LZ%xUJz?uymxw+vQ;WbX90EWw&x3Q$*0F+)p6b=teM4Z0i!
z6P(!QlMbS6-f3@-3hkz$HEVfk=yHXq~)b3Fy6UrR@1K5yO
zIwM?Z(%!LMxfajg5iqg4m~NXh#G%Zz2InB_>?=1(^|$yl_y_p6uqq^Pe-w}K077}R
ze$lTPZ^4YgVQlheU`wpV&~XNQitk5G4a0ey>5vE&@KoXM=t?N790K$?&9_pD-ILrz
zX^f(WT47R^+S3>Gsko+qTxphyEPxux!amTuR!FKX8ccT<+4}duT53mmITZ-T*5-
zIWFE?iT+-^>*=320s@P3ky$=$Z~_$XJ>qcCn|N&s*haCg0W7TXO5`RpVZaZ|9V{?X
z^{1^aJ5wA;ykffzeK)h%eFdt=UW{XSeSmQlTEU5a#h&dBBPFUY;N9aB0@vu)~XvN
znK#JgDWVo3|3vXt;e~JH4d6BN!ux?Q*m3TOrbw7(g-&_H;lnfJ#xv5AMql;fi{}y%
zm^Js+N!T9iP_AYBU5H(k^OxM07*rUemW?2+r$7V+y6ACH>rKzaqz6i*j^Dr!X;@t>
zKGDUUj*qO8AS~n01#!VhmL+MPji>^_y2pX;!-wl=u7+hCS6EKkKT|Iyuje1U(^Z6V
zJ1FFCz}>x#7)&_IL{C;Hv@&yyYnPw^;8pN0J#qNI85|b&e;FJQw*O%pL}QJoY;s%*
z=;!Yoy^KUqS>}6A9+6vkom%0Rrm{iCOxLfQ-hITE2j
zk|0@NK&NMMEZ3d2X;X>=)}Fh&Yo{#f_sveD&I*@#lEq(fggrFRY-cO=HMB6!Esyj!
zFE7;*P00sxh&uL@Gc4Rd{o)wrd0Zh{;A}>JdDk_Azb`*XlkZoTMs$HWs8$T@?li*b
z`q6D2hl1{am|N>`2?qzhg@lgrSreM?|Gs%?S!Y!*_MZh9r4K{{Y&MaW>b+zhO4XD@
za!O3Wx;N)d`0fz^mvaam1)wh5G0u>&g^;tN9y1_OEaMn_NBdrznj5ez?7L9rUN~;1
zo7mf#_+LARQnNwZgtWX&+L5=lugv(nk#Wx{dGNxN_*Fr1u6v`zMRD_P(-
z{hi@{?0PM9yNY`Q=Xo=l@uLG<(9gilN8(kX=RemJ$RtTI&3n@?A&(sVM+ng?!eVVV
z_M>#S2dD;G&d&62nr(cws~fpS3XeXiwr{&{*8CXpUc5R1AdU)xLm#kV&FuO7P1$Gt
z+6)pAM^o~P$`V>Lo>^r@<6_{xU~9^eJ73G!aMe7${>V*B!kT$%pJ^Mi{zQ6o&P|}*
z%h-ffJXQuw2Yid4(9=q@Kr@SXsyOc%HqH$b;ke`lx^;`h=)WdT8O=ASx#_0J-nBPf
zN&?DOP=kB`sxz`n!zI=cRWvkW2Y5i!8`*aoHtJZDfAddM@|?M9qgpnQH5ca$@Nle#
z+Cz!X@fGD9;IGZS5DYC7(COTfY^=_BA4N-Je-$)6c=f$^4eKfkn1ly#on^}yN{{Ua
zbs`^w<=i>j?}M4f0K;~;{?#d9lCv}C>7-_`uagr8z!mHk5FmCLS?%Oo*gHg*$Or8a
zy^9^4Hxu)tSx95HURuNvcrqzKc~bePaftxknC8q6K0Q&__~q3?b`(VwHqoVoE;I5b
zNBuzyw4r;Jbx336(ZgZ~n0rwgz(K;jun42Ib#8OxKK)J0odo%%wxEQVeTc?>v$#5_
zsRniqfG@31UB^HQ`qNS5)z$a>$s(2Ti$rjj?Jine2qP4?BQb+(h;0l33}HLx)8#~k
zkv@js4h$C6@aYmTwV)4y_BXN|mJiFaaG3sLvYa+J*Ezl8@29&>)@VpulxlxceH9Az
z>Chu&#`Dgz!FK~n0v%1U_CjSXRsT)=-lSFsV5_l?pKJM^OGxao8052~465)!9sT{H
zAY(3FY`h)Dt)t=y>dD|34LFZ_sy{0_Lu4UxLsAtXycDVk9$o5w0
zXLW9HBZ=ZE9V;`Gp8@c`V94A}VE-;Rz(j9(tqNL2q=M)Ei`vTP6X+_$-;kBO7?
zzPS9&i*d~fGL+%;-GMvmWK1p|?0qoDfD&vm7uja;5Jj?UKZU$M8O$lP<1<$909~Q8
zPGqUbIbj)%VG2z@KSj8Xc3x;ZM6QqWWA0gT6q*Gwlhgr0q-{KJGb3C^QH;fwYMREU
ztBjk=29@elv2cueEjwgx&fJKMS^A4V_ryX(gdH?Dn&Qx%$EpoNJmorTpu;lP@6dSl
z8)SW;(M6D1ib^4ilo$279VBd%fXFHj&fE+2IHnY~O;S6NpKpQH+owd9Sd*y%Ur+%v
zf?tQ06Tq97Q5)96si`ES`R{>9Oj0jHPyyk{VCppu`k>>_u+{uAu1rm9Dl!W8Cm%r0
zz!gPc3A}}l1WBg*j1X!=6=A)`62y=*0IN(aP61pd
zkFThqkez~x$ixHXuxssJ5x2&m>a0=l*gXkEBec(nXwXGS4EK*+v%vX_fnbj?NN+eX
zuCFJvBX4Kc?*KdpN*4v!x|uqlE>g3{-kOskspI$GS^19+OHl)i+=@4G!v_l{61(;B
zqS>_EKLJCdwh4MY2fD9sq!o{-ee9WMQZrWSXD3YExU{L8c(;0dZf?pz7ml-nZ
zGK|VvB3dXyzqa9Bc|inQHMW6_02Ru1JUuy8(YEjFwD3decvAV=zSy;=17BRu?j=Xt97l?Mc?C(94R-p2m*Y07QLS8uzZ;mJ6T@S6H|)ljWM8Y^HADDIx2;GlPs8@dg+9a2s>Y0N?kqd^P)0fl
zMB~xXuLuVn3I66xS_
zR9IBA1MRSEGK~ZIf@~Zq%qYpu1XY$=vgnFODV}BZG84dv@fX#gCP9?jo=eT%@~e7j6iglGeVNR?Xd=inM(fndaVlx<~j4Lnrf$U$+c
zLPn6y0a8rC7{F58#(w5pf!4=73jxB@sAg+Ta0PctxY)q)mOg%lEhL!;tmySo)XU(#
zelk~s0Jzwr>^IyQCVP%9L-SFX0YgPIC^XO`?Yn5QpnskUl
zgEMm~;GDQH0v*g4B+f4bo*!Cg8LreM0?=fT@|@aL_EKItL~3a=AqbFW)iur!$2O0{
zN~t#@lVI#5pU-VUaBYr?An#db;Thb{0J8YrZgwo%QA2pDdE@G)4SniC>?U82D9&IH
zqNkEYhkt~mFyVy$QQ$xcj`84>5ZuO~ekAzUjLDp!bZUV8y786=sm7z2du;;)WCWqM
zA#Rh+@t~}UYN8t~i)hwR1u-{PaeF-HmPBrM@fXB*^9h!u;zwUvnrSe;Oy5cBsru$c
z`X0@YjkZ#}L3<@~(>W`#fi>HX|9xg=)qG0maD=POX=p-H@1SV8Cvl-tmdUA1e->hW
zlALa8Gu4i<^Zi>FY*k@OOvk|rkkX#IUh{B2!NS1J&?XB3ho62WcSVhbJ}nQnP_XZW
zziSV8fDEu=?r$fAnmG)3poFr;uqvFihtqLfNWVUA47jl9nh&X48_s;jM4rF4+NcS;
z9&$0c{|G$#h#$YXJ9?oc17Off_Si#N#RRRS=~dQM3e$BIbq?39&%YZ1G{sA=n?viw
z>%ebt2LC3&$UxDee|sWDs_Qw?VUpwo8P`kMj?2IruInf_rv46lL4O+?9&IPA>2Zj0
zTe`KmbrwHxDzo@7G{Q7vR$J^nU=SR?rIkMkWCB?fo?$sYM?LYjZ
zu)(o|(*QiYEYP(0nkG7|tEP{F{sLy)_P^0t3R3OP$iNOKMi8OP@Z
z63$ZQM3WARdv36T0j6QSZZ!w5a=C7XzR+M`tSXloG@t{$b=;R;yru`qwMh6$JYUi
zB7ELLN28|K7K|i=grsS6|1#XH5)qK(dR))A?
zQQmCpPwK}O2nGrp6Vv|!7Yh*!Cld>6qM;c&7#ABmQ>wohC^g`ZVwT
z|MoV!yZy^ZzJpun-t{$u|K4>?|2BRA*a_(iAM3bmH=y^tAhY?ns`+L8L|1RQNbEkM
zDL{et)WUv=R3LXYp?eA(QPwMASD+FqhqOZzp0Q2;7=j;rmLbwxh~^u8hiB*lZx<3~64
zHf;`(2c!iG=U4KcG6EFGECgK!p{S0grU({BJ4QW8s|;A^rms3RHwJC^5TqZ7ke)bT
z140(oB3A-Z42B@8sVaN?(1SMrD%lN!AgX-)`!oH3@ye?wp(>>)rT052{@V>6NWu?w
zZf56_^XkW()Jp~8mGp{!c@4qj2@%+DV5z@9Br<1beLa0_PvPp%c`)|l_w44r@8Nj9#}FH8>ITmzc+D6=*LmJyZN5}Z2J0^
z(9ka0{y44eXHC3c+E~#{nU>`P55GEeBde|P8Pg|T7!
z1^UDHjg5mCxZ^4-S&2|+D+5F%
z)coH8BgeT!N>-wA@#)Rbe&NaRd3I){H!!2O9H6pQRX}bdtx6mq9iOaZK%P8Saf3QA
zeIxVzs0MHifgj!R)>S~B{ktUqVh*70Il>`?Jk0MvE}%ye01~C&??=I~Y+xbAXGms{
z)8QW?3ZU*G!Xdc4pHIkopi}anJOHQbnc!)(-d{vdX$iN$j1)1rr
zd#i~c_j+vu1l#R8uD%28NUnYY?6d#@TD#`>vR8h6y&yXk_MfmlYWx4TQ#tTfi}9~}
zkLj&5^Thb79=s`d3+SEA`@9LA6?CH=%oTkDC>#Uo3Ig%RK*}slAD^R-YLT#gAbD9a
z0DgEQB`)K{_*kY+Yh~t@IOy}
zpC`(`o&4X?0Kse^p*+|*B-=a4u{hp7zH;KSAsIHi2DFGIK;m@~UZR2=3xp%HO9T2#
zq2x0LbDZFA>pW#N9U&QgfALlFo|Rsi1dgBKTf+*PKC^?(qqu?2uFME5rjn{XnM|bFF2Z>y#1B=1`(u7ri|@;t}fL1(!K1ZVmBC!#>Vs
zIR-cu0R0UdM|?8n&PdXpDAV2{JO8R675?U^dkA~TnoKh&jmJRCk0WL&+ahXeQ8B##
z{>Kc@7;xg{TH!q<;o!botUPbYRmAXQ;|$fXOpF{={$}QUOy){8#3x80AKFGT?tgjz
z9s5LYPT1)L4yOJGf*%Y<(j|TDQiGPwekA2(2ynfCE?-d-lrODS51O6$yDUywBXTq#
zG-r+lw8y9Ofk?0Tc>*Uv;J2{xHDb|Sm17pGlSgLO&LS@q&wIqVs>f>SFm?4Xk@U*w
z%psH@>f@jJF0Kr$m{ZAolT$vL(BGs}Vhb{oG5R;#00@^5g1a;v4PBZ1PTuKH*(lin
zJV4dNJ*k#3;@^yKHhJ?b!{*=5q^m|2l?cYwK|)Ec|5S`eUSuT_gJkexGD+Tyz&Vwo
zL2zC~R=<8K1PxXo1x44fO3I%l2Q5-3(@P^V;5+Er@SC{mwkhs*Go{&%OV_2b#Euk@
zC2W&wz6Zz>Z8!7vaRSpJvspHRUByt4H!9tnOi#rX96Tuo}ef)5vsdr!|K#?(X*MS`9r-r)(_sh)1FJdj>Wd)_TqV
z(nsphXl!7sVe1B@&17RF3Z?gh>8>s^iv6+O4ChWR&jIfFT^F_8<{CE_^y_uM9ba=enEmbTy}NI13oym@EC(Z=#Jclqy2
z(DI!t5X?&S5
zd^39I>my_Ht`^-BM!k#T%O;?PD8esZK4v`V_`2pE=<+mqN1b8^q<`sMLUctc1iF2>
zo1C;K;@9unPL^N~hfS6FR!%D>z0|`_9Hbb-%@s5Nkq8{q-hgwHK%c6UCy2iv4aNl9
z=xU{_I^qb7kk{yH7;d!7U#}1BYDcrAH?YUQ40j(Xh9SLwI;^P?Mh=n)9|Z_E_bwapT%r0s^F?gobIiiux-7j--^+d;A9
zB8dW?SM6d6S0YS@*WxUOCkrx9AHstV-_D=A>AXnfbtN{cZXO=BSZP|2A
znO+1*%^9h_F07p~?l&eMpw%x)eG)2>7`Xn=pd}3U(ZxJQz26cu>PATVuo?
z)(Rw=M{Z9*g7^PoeP_iuqECXmd)>-@3hn+g9v{7k9zh}Nkw0#5ECF=SvfKKu_CkZf(pN)H}mbUq&OWN
zy^j%f=>6_{(9V;<@>zHb`yg({mf1eEq=ut14%s(kV379utmVQdD|@i4@*FDrhQYhH
z3y5*h&7}Dk*M$$>B=19qv1lcI3q%%4b;`
zHKG*C8kogwzWz_Dd6lt0Dr8EUcte<|J+E2BQZm=fhCPT;chxbj~`5`6De
zXxY8z3zLq?3wCjCj|*_)p^sSy%OTVBC{8LK2I=-kBLFm)zHdar*FeaRW@t-+y0pfh
zFjA!!d^Q3Mm!GtCNMH@xhcv45;!+Pv)#RAE!Q9eu#r4{K-XOpQmX>ha4?J5SuHQdoAk
ze4J08vH(i5l)9908s?FJmQl!JjI}BibTE|sr+t2K*DC{q7}y1GQCm<>gRKJT+%EQv
z)`Pk8fdJ^3Aa}Mj8EY}z)mv9XhA#nnP
z@^A%9w|igsuf-;$fO}pujrNY_=Vw{rvIXpR8Ng+O!gt|19XSqWv$BBZn9F!%D7iit
zh>B04Qc&Bu@f~8!`&8;OfuL(Abj;*~>bwW~&PMKegRZOG;6<^~T(#U$#zymf(1)U-
z0sXV!NqFRlu4&o58le3xApodfvvVarls0zCZcHq^Oek=ku+)zx^0OE!khL0qGy=b&
z8Sv-fi9zKLN%jo}%@Uz&o*B;m9kz!aC!!S5a`{3ilyWC4zhRlO^fRWFf1Ydu!x%dD
zyhKyD{}k>W)fG_PI9i~DOW3``D8FAhyjhP4T}aORvS90~MBrulFt4(6kZ*lb4Os%{
ziC$w9U11>oxB?d!FWuqCut;eYuRE}N0N}XPanJz{MxYZF`6Z&AquFv8Z_5LBMt0ZSO*pxdn8Z2>Unspatt5z_$3W&FL(+
zOB~Iug-;NN!WHvq4gfIkyN)hP2?VBDJFv{J?*T?;Z}Qu6+MGS@rFf4L
z?3iM_ry;SC?t^IO7#J(XmNvZaEM;=R&(6mOpQXm(w)&bC#E7^Ean!Z!wh)(jrXxg6BI4E^P8anP@2(I7ODp6N>VT(hHvsM0HYD^7
zE8UE2gu+TTR_Dia2qBH#LP)SW-l27l{7_T7m{`0auJ>VGdSrAT*iTyK63b$a
zOsMBtuzI(*Pw_B=A#W3K0RCjo%v+QNOSpXtMI>->7+?Kk=>@86lGKsey5x@er3M`SI;cm2zTZ$DxO8M$iIlO#qSssb*psFAHbO
zt0G(;0u$mCrOGC@k+7%_UwbVq2)l>1iqB-J-)s?fhT3z^pz8vp`{CafYqbuljJShf
zJnMs~`OsGl$2Dcts}7msjp<>TFDs%*$BN?8Umv9$22sP4Z-$U!Y7%`kYKcLw=$IPW
zg_LscqGNam@n_oLYyfRycu-ABUb<})Xn@e%gZHYtHCAh8f64VQgTCLMfyw+Ci|HE;
zHl3gcczdyiR|$&gTmpjD_pOow%80fZH|n|`$>Tu#NZLK3TUpa)AnlON6DjGGco%)Z
z@QdIz$BHPIji~@5`1d=U?YEDiNf>W;$aQ)`b)~;uM>xiH6#zw%ZSHEzu@|4MFw?+%
zJ9{qsRl75>*k{8ub;ymUk<0gCdJZq82J<-b662}x$FEx8pZmxRY6#5(xrqVmN0MPi
zPrKb!$c^`!7pEfQ#^u(Ag-gem?6>
zbTeffsPnhH5n#cWt~+YEDC}^QTKCL}f5?-~P(Bd1R1=NrNJ7{9sV+iJ#f$5E2S`0Kl<0<;L#0#7~oiA$fH3_ctFo
zE=A?u#|R;h*TC*w51CM3KP8LW>yT$vQReh7J%Ip`aAN-ieWKL0CAA9Fe^xtqzO8Fn
z^c4r{VV>0LmTbqeOu
zt~9gY_;v=I%=FZDMfwi9I2_p+64U23r{XKxF;krIx;vc}ja{81(LWqNlyM!8n{Yz*
z4Dc^l4OYA)y;kTT$z$_CVS<-qJ@M?irOc&l1DSluQS)pzA+fVLQ=srLGnr@f$K-o6yxve(-#*+6v+4o05w!Me_Rbe)#
zW{Hi{bk2*QQs1h9TPn4Jq0fxEG5_Pi4p6nWhx1WDHQ=G?BO#6d#$t8h!VMJlB5o=<
zu{P$)4@wCmd%lPq9TEv}Ll68>CBZx2hmPIf8e)ACf_rnY&Np`;WuH`2()qhA3TyJ^
z#b>ID)$in)KwtYu9?W=lTD&ediH0HPMg?mPfnbHl5b~rSZv6#KW#fcc!4Rp!fC(DG~
z#T!{5M15ApbK~iWNwUwE9$N6%6_9xpNBn;PGeFG0piyBSrTda~i%op9`&60jmFq(e
z4Yr{mv4%&|SQpu#ZF=l5=TQJ(U}r(l`O;-W8L7hn$0?bKKdqXbSYq5qKB&A=
z;LQhhq9Reiz}VN~TcKJEw5;0vdd4fD?B5BVaWyEp>PL>^kPW2l@ayKeTsvB
zofL{%GbgnfU-JUSUJ!AH5dC(OZx_s8+jZ`RJP`9JnUk$>#I!-;>UAQu4twaTHG60?
zWcL`gE9F<@3HjiI)i8WB(%tL&E2=!{Vw>l21<3U2!@xToI{HAI8`Z&~O;8Mpcf|~H
zU(^ta!kwN*2d<%4?Do;KlsUB}&hyGa%{hV4cU&ps9y&E_N=Qh}tm
zPMTXo#(?)>u76!|fA-=EEst&ep6DQ5?1k#FZeZFO$x?}yicgjU3PpSvQRGt46_ms9
zaU%(s(+OAOoO{yEn97LHXb1{kCSG5OcF>-IXCB;Kuq(Qs`U9QsSw+lJNkJmUu^
z;{+stH{$ZoZ#AJ?5B;s3)42+)Oy2Pdjw6y{v7G!WI6a)YR%u~7&xqa+!Zrlpjf}8T
zTiOivh4FAF)i^&O1~T)vsGfU&>vYhwys9<2TVMQUC}&EMQsMn(JGaTIu$-S2sn`-2
zd#}P&wz1iN;c*biko}ZHxR@1!+)|Lj!fhYj9MR5Mn3&UPr27z<9ZHtDYRY`Q^R+T4
zSe$c$@C-@Tia7r;{c?s46=Fu>QDi?CT{s!8oZR{{AJ6whd_QY5n}8EH~uGJ!i1zn0fD?}(oNn3dGi
zHL89pP-3T!e7!51(zWb=SN=)!3(Gzp7M2&Y2BGVjy7O60;OrJ+LvYXAeb_sD@-xpbu$RU`*K=(G(E;4Uk3@I@LFj!o0pI8f&=qb3h8;J*vV3ef
zvZFekEs?nELR>zph>AHz+i%fbrlO@Srf8VW7+kmu4e$QUEhLOjwM|MH9JOw$t3$~?
zq$xk?VTf7Q8IW*)&X}A_HZBs(f9VX_J7c=K)xhCMV4NHI!&E^1aDo-F^p
zNyPq*oHYM`>SqU7cwX4ql%8rrzB#RrYw$7qTvX4x5xrxV(l{CLafii&iZM6+OSre-
zxsU2U<0g(~c6EUrG$JdgyPdfv?Q%9=<8gJH2kpRE5)9j8*Xp?Z5p&%ml-i9AFd+^L
zwe|Z6Yk8^RQ~d7Uhj{M{@8XG>99~-7B8q*Oh{w!-+(E?YqOlx!My+tN;PRIbipNpm
zX&%O+FF6e7^ULR4$=m@0)dqu&Z1%!q4qodU&)2F#zm_)ERS_|!dUjP)4p0%OBn$-0
z`5+i0Y8zh4)r8-BH}Cdb)RU2A__*sVVz(pIC)~AigeXM@B%FIJFqJ*}xW@X6cgNlTi5b`Qp$(&?
zBTg(}`BLu^_Vh(*jCkKMUMW*;drY$vniosmhJ#ddjl*EoWi!AUj^AmKGE^A}-Crid
z4G{8>#5yF>`CUVam=TAkVZVvnUK+1xulCV@dr1%|^V!k%`9Mt
z4?B;itX?*lAhjPreWWqndtk6jsfTvx*Q8LWPe(E>$I3tfv(4b0Jn22H{9&s^|3{zp
z+~q5ETuSZq=FR)tu(=mXb|>;vu31rQE|U(;Zz+J=7h}MW%qrp@Q*|l#=nann+J2LN
z??6qBE42PkO3JCjGGyJp2|3rjqXygz)lRcP1pXJd@TBNx*_JZgI}6k2EhIWGPf-UA
zG+H&{>UrnkZJox6GaPa?)}~+2zZ+WBghfV%l}VSbmf#X|okxGEmfVI4AIw4Tj>%FwDVLF;CRLF!Ybng5kI{5Em}Vu7TC!jf=ln9q9@O6H#SyO!d!zyyBs(qHnb?
zq!(p4wJ6SV9Y0B#Udb@8wqao?V#MC9t99&#m4|BNN7~AJUyYLbm1b3yAB;(G{zi=|9<`ba(fy0?}dh%9FX&t^qb7Klq1T-8eh
z`Kso?aI0CejuA|A1?qw*i{?6w&iVD3y(v`A7q@?YE`xoR#}pcRwY8gn?+ziu?D?!F
zLN1#R2+PTipL0CQ*xS}Aw0U_XBj=^&UQ9@fz*la75ZL{#v!DHp64}uyvFDNrS`PMx
zi83!I2R$aYsjGLDR;Lo*Rpulvr=W|nv^l76OJ<6;DxPUTqWMDE_m!NSZ?x!{gP(N@
zf=>W3=w+`5sDp*bDUJPqiEetc_;x?L$*ZTu`ktdBBcS@o*$G2R4U$FT6d3n@`3Uju
z>4YDCthfpqx5}%bcRcsZ+GXfB5N`cFWz5?8h9ZTfv=;##Z)<~FuS
zg(t+FV&z)3t$izCFu<37Nu>GItvG-KlM3bXeg~V-NxFO6>p+ga->@fPSI(!IAp^0L
zc`N%b`9)FP^|1DkgpTk`Ta`;phaUBLHx}tyg%@7!RbsY(*q58iCRWJBdFsl|%f_BH
z?|hatk#D#2JarpZ!QhNF&dvl=ZneV?ls;=W&KeNQy_e}0Zm=JU2tk8x&iDHIr91}I
zo$)QW&`iSWfP6i#H~6u#qubmi>OhttaAMaQ%{||P3VjOtMk(G6NpIQj{*~*3Uhod2pY|WlUFxIV
z9pL?h`O|ccstIzpUHG8C)l$ATD6XDc9bAgY(!EJxbKY+xtAT1E1mLXYl;M
z0N&CEz8}XL%t~MPv6e%&KM2jX)ZK;Fm<<^#qsY5z%%yLt63ez|L0y|v6F9-|^i7ut
zpL@?&2=c8fX$XHXqpa)(%iwr=9*OkWHqU}qF?60@>lUyc%=`Y{$1ms$hJReIeiHmv
z?-4+M@84hXf|Yf24do1#4C$$sL@%+D*i*OoN?Pp+^IH~=Vz*a^)$V0@88K-*1?Snrp;O&|6{!O<>27qK0TJux^wU>M>GD~(829ig
zKUd-o&FbhXf8V?e`G&eU)@OJ_LiF;Cm^?CsJ0g6AA&_c|j1f1^7>e$CdKHIYK7am=gJX4|w!EEHg^wjYx|*LN5x#GtRp_~;lHFAvsST(IKy
z$V-niOV@f%sh<$s1ne#KU{q
zVPkQ{|6YgRF#n8`KSDShZv-$cH4-%eHjAp;8!ZKRW6@F~(P*B8VTj$C9^~_8f$+Xc
zk{i6>qQ_yTj8sxY%z(czdt2i@M&cZw4P2&{l*v%mi4L)fBt?L`@$f92PzGi
z0H)QI7ovPuh)3B)oGv>_ZWkzj2_I&@hH>06H%49Z@O@uC2@}6$4wC+Wn5^PMHBcg_mD|O2ZV$gzu8%Ca(tu9OToI1z|IDjrhQ02s#>Ot3rE{>
zV=-ezvkMHF>Qo-n{lWK5%Vz(n2wYA)h87Kfd$0&xS1kaC8O;NxjF+{4{P{c9)+6I^
zrXE9+%~>lGwqAN3$pf;MeJgftSoJI_{AU`dF`N#oN(7?b`dE~%J)X($Hyn1^@x8t5
zGw3!K90wJ(Gd(UYOiNJ*0bdKBS;%i+D)ty6Dt$eyDkgn}ZC3q^B1S#5Au|wDlO?Go
zKRfuwUR<(o^EJ)ppxtzTL~-JvxbFK@L%%%v9&v?y1D7@`N)q6ZfgOvQD1iUesFhN1
z_*F|ZDx@?}h92Wrh40X9+m1<^ShtQyJDhU?{pUKpw!Y_Jf!O?MP_6l0z)ijeO)+^X
zsl$x*1q{6Yh}A=}%AV9R7ztWzdx#JYrHw`}>XYabxBde_1Z6pYF0#>1N
zdgP#W2T#KGcD$QPjRm^&gZZmSi_)3jxbYC|cStCX`u#PZhBYYd?r>bb>epO=JYVLD
z0q-OY6R(X?yiDjZoq6xJ{|?-{^6<)<%(4J*stFVV{EuXtgo5f=j`t?$(1K6U;TsM~
z+IVyO2+-HemA-O+wVhalk>XLI)Y#g7Wek!EM&=KK`>H~paApZBr5fT25I^0MUKOvk
zyCSG#U{_*hM&;aE(uZ^!D)*B>#ZtpDw!XC6r5jtvW=xMl_TC!B6`0HY2n*$0Dwm||
z>lKZ`dv20E>4O#9)(Lg<&>zh%l^qZ4~}8KjPXm<)2bOfM5_-@}ZooUT4e
zguZ1!us~iIT&(l@Ac^AMMzm*!Y(`FHzN1{j<6KTJ5Tqy^O{9BjSR#A$K_Ylj`MZY#
zHxk$w!U0$4NYcRG?p&9fP;TP0O#z}KW5#4ZF~Y5szG6+8*bc=(h5qgGy$e<+sv$g|m0q%yK-SJH)WeOTcX2Y#bv+nY95vpH3&-s~#*2nkWG~5*J
zRcdvo1t#y%VnzAo<=664RCt0Ec-{}m33?Quee`PZUL%k3R;>LB9S2{q^ei)MDo^(%
z4ZVt=ZC!VM8<>=c%&+;5H+03KhN+Gcub~Fs?aJYQ)TMNHN1T*3^FUBIN&}lItpv_g
zZzV3%N-$CYP*sr@Kj*wI&ll>uZ1Uk|WHJ-WnoR3bA{Q*lQr4+yM*&r?A|wDk`KE>6
z^0eYYZgVIIiTaVo-*=OzA(bNhG2neD;)pt;o>-rL4lVWZH%Z%H1`+}IyPQ%jFBY_s
z9y2|ELkk0RUo7c-%Btin3r=pJU%Nf+)SJ~VF>+`S{DY#=qN;we{M>HP#JArO`dFAH
zSfI<}mpDaNu>i0ml#*D4313HC=Av;i9&$DbSH&C6dTs4*jVv
zfdA`9L>yOUyMixNQ(N(T``PjkX7{jDR3rWpD!KVq{-eb@8OpmIces
z!q9xax2?~gEJvoH=CNB7SUv1>7Usu)W6@vo)#qp#Y-IO>ui4vl;~UuDp4p#GG7)V_
zM%k)+c*kc9+6o!l5-R(A(QDURzAumGT8XG(O^A!&Cp9gcf3Kh6
zUGOt=J#5s;VOKq^Q$=()4f)9~y=>jxK5)nbpQ|f9
z4I>U|dWFQmd*Dy)On461wmiIl-JpjeH0m88N(RZ5FjFfNrfPuu&y!#Ja|I;5{0HsQ
zaf&jrIls)7pUi6yR1GRx%ZqG_s-X&nAk
zV~fO^J~y<-lG)yCC4rV2H3X(vn*);~?nxl@#M}GbYM5%<6y*8v`3|>#u$JiI+w;9h
zT~zDZsndt|;2F7YB6ka)XM_
z&+D$0U~2*|R>bl;16^x>jo}6DXm^VC#rC|+*5X|s*6b7DXghp=_0+&Qq@uIE)VvEk
z?K~1P$nl848l^mi49}VJqa~~e4at6o`SAS1lDA_|FcFnl+lb~kr^GnRPoMqAr2t;n
zgf%681!NS?jXjmN6-yq^=0cO;%@98=k7baOtj%NEqr}~sy?TLb5f{)To
zXMeo|7B9Q=`7nW6hCO4HV@p12sEwg>gxeY&nqTeXZitnmYwqC_+BH|M_$v-+R1&y2
z#$pDvAmtqcrydZ0Fss5em@`5{j$;d0qJ@{;I4O?Zu_tkg8iGovvSc}nc)K?ske36i
zBBFf`!}4|kkN!7EMo#jFh+H-7G4ypI@9(ueEtEgJhhE$;#Qf&tIZyT#4cY{MI90A-
z34gH0&qMX&4}s@$k(kXfIp$EVcV(#C!uf~c0?3~=db}=wUKP%6e$7THsB9m>*N?vg
ze+Q;9$>{7;ZGFoH=BnjdPb|azeFoEaUQtQN3zIDp_x@8Ayf~cu?J-1_jCR+8xNFE^
zkhvzglnAI@*Jq0>N&$KhYhQPYNb%e;8=*610jgm>koDu80(FKKaeC-nqR}gnMLmoh
z@2svkGs*>jXFpMt429oAf{3?|O*zA{9EwBSr!K|M1z;wUc|se|BZUsP%Y
zS9O(e3A|bYhg{5)(uSq2g=i<+Pl0Aoasv3w?$gl}r}}+
zR8G?kwuR+JWJ_mn0He-MV6m-i16XQ0^m@6k$g>IUjYKbJvHyDz`Zcy65+glJ<4W6q
zHlhxGm;bOhj$z-m1~}
z<*h(z_;{>7G6#r-pWbL~!@}{tdCpr4Umnw9s*VJL5(yujA@}lXSE_
zcaE2^n7<6!*eXWG2)eP_KjG4Q8CPgv$`!hSR<32
zI(AWB3tssF1&M`)wAOwK3|~z08RjfsqPk6?z9|ovw82p^1^FU$qclqz(6t?by}mT|
z;i~@9nb;@!x@J**(FE8h+E`Y9tMp-67|?c%G!6)CgX%v?Bu$2~ib1^<&Eef3=n2Gc
zvxgL+duEl+pI>6pIFHe5DXeQoyXaSmx;u$V8IrU$cmTWI&>`p6ZcO;hs6C}I7aWG7
z8i*uOu<Kfli^9uQ;jCBE_gf*p{!V^L5WR1+wG;$~+co_%HtR62kt*<0J3MZNYY~ub$%YL(
zYDYrrVb%>R89PIX6nuh3*(8z-e|p#194GoKg8JSlsYks5HuH))~+a(&i$%gG4xgBfYRXX7CWL`^5*4FXgs)zeLvTlWW*==pa**8(4~*68Cd!
zpvmJ-Ze&%7o3?xQ7FrT!abxQig-va7C4_!HQhDjLP!7I8GWjza*O2fk?vW$|ThbBg
zPkkZKD5#-m5=DNuE{X+tp#r$Vja)<-a9M{Ui^A1yw--tlPU*XUnkL408I;WaP33IY
z0NE`0*W<%4MsC7_(9TuMk{b7q&?RV|->~M(0&YA5HO%Z3P4yGnQ%?xInqCgcjc!`9
zxM$l09S-775T{AkcR;s-DG=dc77nWs&>wbR;1SEUw)4jXyI|?%eopcph0TSuPLEib
zS}cE#q0~~ujDpvaWki8uQIMc
z5rZl|JA0`$>051y2wJ!jy7`ZZyH^0%o?G!?@dw*3{!Hr{mo?r;WqL(?F&x5&teLsd
zei@UH2N91VO9^=zfKh%^c~8z!+yzV9mZXY>fe_Uo?!r77
zvAPP#u`Hlm?nA;)M}qhsJ(*|nCD3|m#scV}Ev1-$Gq=7sl>NSSNbQ#MkNj^Lvs?E}q=
zxhky~3ExtFl53+EE3cUEUNVQ$HD-bbD^bpkibUy!uHo#e4q9eD(%r0HI8s`zK70)M
zTF_2^;+Y4-@d*2F=Q{3T6;(vctl-X)%*FJV2$=2M0r;G))$(pLX>2fkta36&6x~*{
zY@xAt9@}2H@LKP?rao>1DH3Qi9P+_TVki2`tT&7D0dkqZQi5kzOJ#+1Ry^b;IG>qZz`hs?KH6tfy%pOUMV8#!(5
zUs!M)6N%(Hs(D!I2b9+@h{^>;zzA;EI;zr^hJ&dNXBI4*%Dx^S4zWCgyE%O3ic?{K
za33X--_D-63BAkyJRcZsC#LZ_t|_TYubtOf-sThD0XqXxUtyw8&^j25Uuv=3@N0_o
z`w{#NmfDrpCz&QnY$~|Bg3m!E>U8^*ALM5(zo|7UUAIHkLbVuA&82No_L5#*%*is{
zyf$j1vSYnEm$w4&s1Ti2if)*#pWEhty}`Q^6Zk)Ondj8te%y)3@GSRKLrZQb3dQRg
z^M|r-dh+c%@*Mi)`#I(4oghn~6ugq!m{5>n&mM#T+mF{lzDH#y$rmIuCH>}<-BSjb
zizM)NCPe(ysH!4(aLpCRB^TXtC47{Tv;h#6s1AW}7~Te$o}(xu|3@s_+#fQ3Y-woa
z_vS`MKKN<1`!{c+SeJMl+h$9>*UdQGF|wA??LKAy+{+d)gT##
z5-dMCMZ}A9&2VQD#(KmN=#xrrSzPg@@{jHrTL`Gc>+PH{F=tQ!l{#=L@|JH7kPNGZ
zHbevCBM`h=(4SkXB8U$tj|*D{C}on<3Zhg(D>(+(YLsGw0-#jCd;>gxYjTg(#IddD
z(Je!FJ)LtRwkbiCAEp~K=o`h5F-@7~esz-@+#ktR+qc3JENf^RMm>I=<&f+1_-2~b
z`5O7lV5eXjN!i%y1Da}e0&irBybm9aF0pph^m`)zIh+Jhb0w{4M(9MqG
z&`;c-)6pR9cBFx|wQEX)ll}KAC(;C-S`N*q4f?Ge1%570?Xw?*hVrtV`rD~Fh{rLp
zCtt0p7{k~SDq?#>jNLasWYmxAMYm?=;zCWS2eCvC(@gvz9pNQ^JWY{CvAKQN_+HMi
z#q=C`qgy?ZwCPDBrCbk2Xc?3wF{P3>PydS6;LbFt<>G5s@#!`?#T@oZ>E}|0>{u&g
zKsC39x`nuj?Q6^IwE4?CfecqOM^}x4?_^6tkd%M!)9~|X5WYyXsN@TSY0`UBuUyln
z9-U|_jnG_0`t?eG;{YAQv6Oxx*=EhvgRcuDbQc?%1thgEsMt-NKP^{te20iw9-K-K
zb0^c*#Iw}G$qVa}Xi^O)3I=P?kma37%g3)|;7;+133ixGldMaRgGL*GcilQi;U}yu
zjtLsLiy^vkZoy6+k9aV1seG4WX=kL9m2>(qZJ@pfs!AV!uJ|yO@OyWW3M$KPf|{}w
zHMnS|0E5q2i>_a($J_kQ|2A
z#+&;t*Umj%#=o7*9;vCVKnw
zZp0G#otj{S<&O?&yZ4nBjWVSPxH5x&M*!ua?l7Bas9F4|?~mTA*B&UbNfzQsW&Jtw
zw`am4?;EELW!2`;GAr#onMZM1?S*H)_zOr(XaIkI;A7k=hG&{yVyhB+FF0APeP_mn
zAJU0Fg(6E*!I}L^HqX>mqj;dazFb2!rk~CxTNGch+I)%$UHXz3KFsg
zZz%EWbtYPur@z2j@GZqD6EDnQHAaLWjhZ~IJKxxKAE0b-V84<{(UOY+)|?|JC#i#T
zND*Ftc(g$04SRI7tJZ=uR)Di8{J0_X3cZ(!-Xnsm7zh@(Y^hG0sP)2RxU7@3C`x-b
z2TpcnjO~`0x)HfMtA900?qEh)lBULM=iSVj4Cg}4ANg@1+Cc#89HNN3?&H4nD}r@!
zu#E&94q7tH(_>y#xS$)#=)I{AJQ%k2egBt#QZ`saArT1sfI36FWQ3hd-;C~Wcck-|
zMWgNBUlTP50e6k)_fJ)ZIK~jB&JbMk#S74eZTLZON(*}`$%hXLDc~AYj=61Olb#l7
zN8uA1W_^9>xQBzOmPTIHkl6b%5(if7#A#-GM7io~6(^D52vM~HIITZF=wc$l$6|7S
za(&1~tIdCye7_{9tw3nYnrGMVC-n&~rcEfKY%V2;`7c6pE&%E{rV3u8jd&
zJZ>>cr`cij&e#44K?8e-7i
z<;N$Ml*{+cALD<@CYWAd&h7AOFtdk@rr_$AH&nP7ZX^C&J9(PhC|}NZkqefvY5egE
z)|wR0B82TdE3rQLmA-EB;AV?|OYqi3&n{D5Mp*sBQ*qUtiHmu^Sioltrjw%7)^>M6
zlIjI*N0teh`eC?5@x?QTcRRofnb4w6QubdA3|}7=Qr1F8)6eO8zz#~k^1Jk&r1A6b
zA8__>^pZuMK%S^V7$Eq2-|6~EZ9}fi0cAGq=fg%fsZcf?3`;EWHv;Un#rrRGCM&Mp
zCjOtxz9)qka1b_ZVm?@?s>^TIYTh>#{t|D^KZ-y8u(kBrc0t!s+YJfo3%lWWy-e@Y
zmaz*~x45W`W`NFQ4!QR6=52xLLHX6jEXo|UOv@PnSkwboFZ!B)xfE(W@t7iV>~*v@
z(DshbkBc24s1382-Y7TSW8Z>RQmeMm;6VHv&rwtSQ=M5XvCivEm9!TOl=Du*|<9>$f^9_3cETMTFdKYLu55NL(Szjw3wbSu(lI*KIjl)D1;rjEpw(Yiw&_jHFSea!*x7Ky$5rO^
zWa2mimyW{wB1v61yp09gx>iH5HEz5011n&GFBE{7Vb)VtO^kS(OHK(h3Wc9~IJ2}<
zbzLbHYev0)^)!59Fu1_F^{D6B@3hH9q}&PYrQ{0(l?XUt8GwO_R{Eb31hB0x>p=C+
z&i^)bnP6FrSAE$Nh#+s*8)WM)Z=u%>$u-T>>{>V-msYU|n{_G^Qj26J%yI?+b+>u)
z$1y6;-i)l(J(iFlvAF;!!*kO_i3E4J7`x4e@62F-P}-=*x%gP|d>h0P-*1(T;$Ak#
zZ%k{eo5Btdn-6(z&S@WCf?fD%7fSke&b)FMtx99xi=lL$&mFhW>SRHCt5`O9CnPE?
zJ|UtJOj=r=&h-kO|C*0L#Lm0tGw%(0_Y=c`7D$)eK_Ir!>Sq<9k3tI
z=ZTwBn+%d+O;7YLir_c-2EPDPWn47)bsu<|2s`W0Xwfmp^8zf6&vwsZjC;{N?2fY2
zmZ@A)fv-uoNYUEHrn{p>Ir+i7<&xwXlRtBTA9!h7!?&r0D%X|0nRwU9dfb
zoh=@|LTvf}M<1gej1NZ%F2}vxq7xKBk25M2W<-uBI9n)q*HysP9
zu&y}~W7doeI(EyrDqN{qL=70F@qlZ8IyJLJfL$0h@Lww7!VCvPhDuP=eb|YmmnPP|
z94x1tt$&Z7m^(3kYR6Y&oKnT
zG{?Sj*Ay$XPy{3apulH*eBb_0g-;Uj6$b6gyTqTywkY;_5*)a>c&`l;cCbW$hM+Y|
zWFR9xNmAzbd){ESFKcT}`}5Y0Mzbg+-sJ6&ua_VS5VM&f5@mg|9kcylFWWfohdh^A2uq-B@P6?rXhTBLXh(fp0;7S?%?f85!iH
z0ckfi$(J0=19A?fo9JE;TXLz~KD0;jA;$#AqPp8DWh=&70Coz{T1doIctN=~YW)mBq2$&vmJ@ka6W8
zikxAG5>`PM?`!QH;Dz^pAmpGaE5|LTu@F#$eo#^MLVpU~g>)nuqFP=*%ekJyyY`Rw
zj!dkq+cJ2DiV;_Z4@jlp@|oR`c!p}ktaloEltsD4NUt4TL0Ld*^0Gf0{kEA3US%aQ
zCh~^FYeAGQk72+m-UuRToP+6wXDCo7Zmcf*En4a8fD2sUb;yf<0CO#^;h+k{x7_k<
z*_W75Cuq2n|8`pW-*?<{TOrP{n1TU*q+qS0ix-jELOEuBZ}udyMCaWvWyE0K>@3E%
zudb1Vqal#JxQe+*bl!m5$d{HC6GT1@N~lxo`okH{<0NTqh4HW2yMx}cIi
zBFxE5N%KtWQ7)^Mtcjg_cS}%sS*F)9Vyq~40rC)WIQi^<$mQhu30)_{5W8da(0*kV
zFB!ll$e-k-f1T;thSyr1k2}?Se~w}q0>(~qV_r@)90r46#{?Hhv;reV6GIz)f!L83
zP;U;2rak_c3|F7F6zgLB4o88PgWm8X-52=LYQ{x=-9
zYo_CJ?^jp>3t=F6{nIDKN9jChBM~XW4$o?Uq@8^=b&HCxj9S{VYOLb5fp>1@?s(z<
zh7(>KLJNl;j-3?b;u{h7N^T`YC_{4`rx5#p6@>_T^c
zR|JaOjzBMpbx5p2__-n5Mg%N!14#Hf&i|*0DfG{^%-fOBw2>uE_FvO`5ejo#qIIA2
zhl8`KT;j
z5t$7XD?m>3D{IC4TnZrdJ?E(UMeNv|#%T#<_B&a#gtFz5upj?5)H7mN1S9A=;#qOB
zeKfLwvo)=NMLR^q!O8pW)z)CWX7wIeVtrmFBGM~${sB-Lj(LVFAQ`0qY6PEu)3Z=?
z*jw6~VO?A@Lj?k{lbk}drJ@crNC)6XBjuojuy0!O6CmzN$Oe^;{Atx;gAFsbJ^E}h
zzqKX%wC;OhZP}23zj-!ga#S^clX?p`weh?-8sk)%>ROz-Sj?<+j5zfrc{=%eoa4WT
zUB%c%$b>;UMFmrE#CZhXnony|3?9_$9B1d@Z0bRU#jo;Q+SEJjdalv-y4hqIf81-B
zG%U{&NhMt$Ln;1}$i*agQnHj?EpTxcPy%mbp)WP$R&)CjQ!-gvWxOqaeY`V2tenJH
zL{FX!zDiilr@|}gK0={0J*a6xe^Bq9q!bLWwHHmabTLF?ndgEB0q_^$G`{
zpQl%oA4l)~NrR`N5b{`mR;fFAXuF8q@1tryAChwNC$$gDbO}Na>oL4qKa(**7NY{jAKti@M`#bjKNxk`XXFK^rR|K@p^Rl
z5^CdCiR@|gj6G3MWDS7^blfWguEQ;kpUWl(l>(%xIe-&<>ovT~3H;*|_PssH_XGc~dhtZnVLM-5P}iGXzPutw{$7
z%=ZXvYhoJk%o7?~JvPBUJEYzyb(W*9ja>vwjdG@5L+sor1T*F+&
zB~0rnb+iGY86y)pJ$v^{+Sy!fi?
z)>Hm;q9oGrew6G_0odqyX`Xo^%Rx_-L<2~OUIJ@Z
zNWjxdDYl?Ncl(os{Ky0`Ts2xtu#(eGy?k+c>^?jY3tylv!!#nvt=Jymum0#6!eNNN^a#-@
zI?jZJD)WQaWXKCEl^o3}Ty%5kVNX2;$o3xM27ktXJE77A_$E0`B(drJTl45_EyOn&
zx7>>--yJVB_No)8dG4vs%t{QO^|#X_5ggRHnVa6@fsWc)NW&RAtqo+mwvcZ^TN0xB~^0;6zMW`#)SgOn7Gk%BerQC81
zQ|Re`X*%Hm@Wx989fJ+-i-YAvUDJB=dw~OjS}mo5J8K*FJwnv5kC)qfN&R&>yr1Nd^au-(>{T<>n1yvcSX?x
z2N$o1w`+||dBQJ2Wq`|QoZ6e<`MyCR9AGm(tiv
z8TO99aV7f#k_2X{LntRrdobN;;j+WZaPs$`qLoASCcLtk>sP8hzW8Svtiw+}OpAYi
z2dnBEt0d2EuG4l#l;D3$P1wtUg)cLC@y7rXdcmEdq8eqA0s6f
zs(0Nub+4HFALjZ@4x*v&m*utO$5Z?CEFYI^1V*M05wPi8(oq|Zs9L8_v_qGFBUYbx
zmd_p}dq3{KgO<4`Hqdm$n^`*Ws^t~g{XH8wS`+2?DLgZ++!{K`uYU4S2ksW)r;`b%
z_!KI?s_|R
zDK8^yAti+xuzZAJ=xY!zs{`796F^_nIp^_ucXFof+*BIvD6RX4vDjQO`G!Z5ezneP
ze@YRPM9Ed;`SY;wF~&BVEZ7Bc9?^k4)cFD&vlotmN5zCS9CB46Q^1Br@V*`;Kuh=&
z6M3LlCS-=Z>V6RZpD>w%7ZEfRuZ2pE!18eub5d;LfR7!ECg3>X8;?1E^I}W-uyu+F
z#9a&DtCQy@-E!Ydp>*6PGh}NA$zH(XDh#l2`Yv|f|BhQ$BiUhXUymU8SeFjd{Qrez
z#-(dOvUUGYxW1I1%u57;H+G{%ALd#|(&|RPrS8ODdviS7G@(xK{4|7s4M^&$X~EuT
zc9iwWsJo4-pI2D_Ff0%g
zU$Q?W?Bql_lkSwHv;l8MFY05xNe&x?*v8&OgpuwAeHsXo*8GHj4m!t}k0-#@-QUkF
zmJG+$;utM4`EQxPmX@eEBvbQ8d42VO+71d&6Z`TDCExdk8pAi@U$71qBIrgT=NNpX
zet+SOTwNGQw9+=N#&JfJEwAO>`q_J4t-4n6Bxq;@9ZdY$KU%8LV10MNpy$-C`T%%iSrH2v+d>&x7ub%xDpesyrr`0{Ta7
z7y`!)7DG1lgEX_Q(1xewn44PnLBk)ytEukm46(}ER352+)O;3#Q50MkVs7Re+ih>g
zQg}6lkG*fw*$J0;f~leC4*S?6c3C9p|4Tv@o>_QuO$c-$TS=@u8I{ovgQIubE&joG
zkEgi`*zN6$gbca`e_HNU*9I-wsjrzv8y|_Y%&-)k;PRr*sAf<-GH{G`!wb5u50yPz
ziV-Bq>E}^@3e!a|4*ju*SryslL
zY0Ec@P>KCpSV~FtHFg?1MfEXHWQwoQ4dxs6kzrE#IG7@D;}#lQAgA7^sfE$QNH!#z
z?~2}=7*b-lL^49k*p4b3?13X?xUrB0+HKw@vt{mo-{hmV#%`IUt<%Nd0)20`z(2VG
zB}p$gnDjK8
z-mi`?B{Wxr>NIGh?!)=Lc4*_ic&u;yk`|zDzm7q{+z`%98+t6|^=~$svNcsCL<;wQ
z#kS@oPF%>7f4-|y5_QlJfg!@zc;aeaymVci*)rD5w*IEN5paF3p&!E+5{tYt`z98G
z&*N;a@IyTRJ`>+I*I)i8X($RgSY|oc;lvo=M@X>*(*2E4hH_5K_Mz0uBDt%gYXrS(
zpL6QDTnx7Yan5zCpeS#{9=m&in&ef>nLheLue1h8e+rcT*kV>n{%|+q;k3qq{+J|~
zumX!|B9_b^7yo7MbI3~9u4IT#I(y9$0}XK$3guBmc#5yFLp
z^T&#ylDWifwTT>CNCUtTMo-SaRqX3XtI}W?!LD%E^#qVuGokDPvd1CDxfQG7hdVS6
zyVV^je`5FvFQ>HfA5*Y(_&{XV0zV6p8={lC$0DyUCpq>9OAqR%*ut@~H-Si0xrgvG
zp(DX;aFV+wcToTFA|Y-0I9I%VH@Xjlpdq>AtDl*>TxLaS=9&!F3qUw6rSb<@}?@t%3@U)R&HKGYzF(H~Wiy
ze=y0wK(R@Vey=c48?q7_bpsNR&7t1ArK_|ZWpmqayRY`0;`OD?FAsndT@@`I3>tji4;18;%fkF+Zf7~Un
zpi0y$B=0_fgldft_5|Sy)(c}@FLKSXEU{6CW^1-3@8s*GsiN}pgujLYX_w|pL<~p(
z?6C*jTkaE+?xi&Vn>vceR2;m?Tp)}-Sr0`6W|8rx|D=$yjy*QLwQDMW`@jmB61F|4
zi#$eNLri+!j1TXHGN_g=@Y`XIf9|>Q%Fe%XnNSTsOPQuyz@7Qedc_tcAS~WfK%36W
z=>ZP8r{;+G#KBbZ7BR3-T*2VR<0#kkSIJHrmW0@aT5iAFNnO<^k(Xt%%W1z&k@Wt*
zP<~=!B5dcohrqw_8@7b&7#LinsjK0EY{s4XtiosG7at^GmKZD
zTuxn9q}{}58$qH|W0FLSbojvTMU@b_$K$I$MKBE)z3Hb8pBm`w+4_N0E8ipsYBb@H
z75!6!e1ruLtm2);rIWM{e{(H4h3Er!3e&AWxIXQs-R=L~2{d-s5UwcbSEldgad=B%s&XV
zZiQs=g;urk+t2X;>(L52UbY>BuXgb3Dj(oT;TIrj<{IY2e|;wY_%{kT;X^R4s6>eEk
zgmi<`S56is|BuiJy+i_$fJ46Dcep^Zz6&w<8ADTj{pzqdIkTWh5t2C{tkTFoh`^l?
zZ#%%Af7F(}nt1|RmV>M0$L7%`(G1tB$10igQP?E-Z8OY?y>Q(@@cuo+VcFH~%WR0G
zWQyio`4qNaO{nz~186P&#Qa3eVJDF&l1h1dH|jP+X{TxgB_oITf!Sb%&n<^$AVr{5sZ19G_?po5a6d0ih5Sn
zZ)E;|vCEVOV7MP>bcIY1Vn+%mFIV+Q*E?$+Wh@ay4+2CnGL4oSa=%nHhz$_=T%JkC
zM|L?ct*ghUB>^H~7-cN#MI8|P)O(nbVwY2Jr0yiaBq_+E8V>19GVE2dC-*`E`lKqn
z47^N_hes0-$|kQy`TW7g1+{09De}#4f;5)8d?eLKwO)baI5us
z{E-##8VSry_=lQh2#j-qxdNMK&X6eyJ4Mp~HAmBK4W%9-R|)@VGA(~X$djWE1Ywf2
z*6#rRlO0boBbhC|ff@-#d(U`;zkhdw0>1RCpkPdr1fk>oFbQnWOk12BHTxZ{nJu&}
zsSY6p5{mTWoKECRtQgMLSkcCm2AjQ|D?X%aBedM-pH5|4@)aX*eGb;O?WJ;9)UAx|huTX72#j9&
zOC&_neK%k)>p>#yc&cd9H>b}J+{oL|z|1AMZfi^l2gNpS$0>hupQcMOzD-z-G)d~M
z#}|>0ML#$s)*t)0U)%sr3ZX)Ieh&;l!C48q0_YNZ9b{+jwyyu-E&H8>?Xez)$ZS9R
z0K7}5yD*k<855lhTH<}^IYkqcBH=gq2;BcO)CMvIgBd$w0pa=e2X!|>s+Gn*
zP``kgGJ&9BJpY&lFdJ)h+vWoiT)I`h?!0l0`GIR`;h%FL&qrh?AJ=iB^z{qT#o7{#
z4-k!QCXG^Uc@dDiz_(+;{-Z)wfeV6xtW>bW8m`gQXlUb4Cxc~wq~^W}{5
zMGrA6+iM@2g4gFxJe
z5pMMpK68K5E%sCX)YM+RzD^VPdGkpEj
z5VrW~(b>sZ3qCG&RNo&&Pb9{7LfeLn!j^p68-;&j#N0b%WumVY`CD>IVDWtLR{Hzt
z<)f-t5q9yb$tse=^e;gHtcU_gguocb(BESn$nVy}&U#DeZc1kpSb9ocK$H25dJPh`
z1LWI+!}T}7udeaK$SX)YtO?YE4yzJY>+21hU8@{@mI8s8-`7ZumLAo2Lf0sOac+`b
z9uI#SyWenTFgaL~SUN&IYTkIPQC4<6%TCcW-B2o8c+Mk5_q&b+3nKG=kL%QrS_pDJ
zMknjpDd-pV6)qpG_TK%6q;H|Dc-a}c(u;5tAO_rnwdo|iHM4zN0U)v$aE3cy(bmnZ
zYyF*?01M!qB3jik;M8i9fD1V9GobsBFPncORrD$H-u&wF<`xXXu@m*lQ(XnS3xH(nQHX
zSlh~(a3>`6utQuM+?v_nLIUE`jMJ6(PLD;x1My}>VYQsaIO
zJV6x(dD-EG=RXz!8(2-W&G-vzYi56dt+d>SOLx<@a}{Vv$n4lt>Ez8Kf|zSFQ@7r}
z0iNU{_OGWhuDki(Q+ku}f4mh~jG9t+)e8}u%Q&bjBmAbVUCMqY^b|$J$5u)N&F`y{
z$LZoE!9_`|0cU(9ZAbJjF^Hln>?)R9+-~_-+Mo-Bo%SDpN)4^k4mEFtA6(Tdh9^BR&Va=5d#WJR%hU?nqaH|02Uu0|pO#6Txpbz=
zd7`t>%2}26{h|UY``}z;kgd44HNq5OPr+6km%yqpfY{m7yM#0mH~jkO+#(Sb1C`XE
zb$E9lF9AR6nEmN$(weNpF%^FY$u)X_vf_;3_l&%{R41z^+mU-z2Ca+4+%knb#Uu0=
z3jqC;3tLRh@4AnnX}6VXCmA;nb{quuAyn0j1WzmP8*vR;G5*z0Ets&@8{N)q#c9K@
zt|vSyyrw%lH+89m5IOvudBd^e70llFG@`NC+_JUU-RJiS*f*a}U@T}1T3UE!=QHNnY;^!0Q$k3np$)FFk>{BB9
z?5X@V935{sA`Va%7nxf{%GCgRSBzXimJcy)2m|~!hSuRw4BC^3b(g)ShJG{t6Aq=8
zOcU*gbHHII1Ci*dFS>scL?8-bR6SRZ>60ci9bn4|mR5gof4BrGu$KjEzGWzh
zxp3gacoT~^r~XynnB2Z8;vEKDg9-(b$QyE&Ua7z&9{qcRR?04J^Y$1+cACcMK#;mF
zAo?PV4Bbn>e)^IFwZAR&30oJ8g!iR1$VZq@xY(HgSPr=rKckWMuaUcCjaJs;uZcu$
z)H|{jI61{c52Jrc3v(o^0g>}IZZIJoR_fR6V$aAuCvy$2`90g!sC^7P&wHLYdITuH
z*4aTV$Ee<5BSCZu?H{E>ZZOB1tej${d%)NKZeLW(+wd$4rVk_^0U=9P7L623`L()b
z0gqXs{@f#3ETg3SNemKJV5{~*KfGv=(k%|Z9==~lO>KW~3m{Q384@a0&zi=c9yvK5
zl^eKG-+IcD?iFKoO9rxnd?Zx{w3!a`@-`OkIYF|54G`K=#ct5e+f|LaX#Vdda4BC(
zN^XkGJk`luW=V%BpC5%5Z^2l^D79@=r=Is1S1DSL!Tx0Wn)t$Tq|qH8j`D-WLk#nI
z=2!ooN=JXY<*3-d@Et>a?f1B@3Z-D7!jKJ<$9*%`{pG{HEuBWMowEb-3ba_zY3*5c
zV|}Ote>Juko<8Eb==I6-(9?D>M;82o=R9dyk>=&c
z5$pefKN`|luSA#7HDd@+q4>i`AAUtj1rWkFUYK7+NuZwh5a6V1m8>r%?r4&6s4ofCtd?^ZzcYm$eExT#a>MTev
zrKNvz0rc>ksIvw2(<7uAQVvnZKeiaVwP_-Z1oaZy2jto)|5|y%HI0G&77INsSW?$Z
z%fFR=S}gRSvApl~!u&sYBW*=*B+-(K{mO|CMYSW8Ze>&pt17a7fDDIIhw(A}+9Vb)9+!VY+u)&$0vU7;ic}{GVBd7!V{!;yXnr7uEdb7m
zv_G9(wNJWu!UM+A5W!-U{WYNjhBXI6AgdD$F`(3~t=@0t*>VQX$&5m=vI+l*4xpMC
zRa`bLx%oHK8bXsl2M6!kz4cDUnb)o9-*#Xhdmbqbo)>hWmHJ|ZUS!ix92SIV@0}uS1?vnk&-gxYyXz2b3=Y-_4%v`tlQ7^fHOjcEtgiB3h{q_To2Lht^X$6
ztVLaqH%!GJ%N5Hr1i4uyv%P;Pqemg4#RJNf`q0MO8O3&=q(-0NN){Pz@Y9IDgsmH@hR=Vbj
zA@?ojn`?Nj7;|!b{q-!yF?N-F3A+a`I5rXh#W{MIjvhnWoNJ;5@gvrv7kt4zr&+9Fd#e$nrR|`q;Bc0;H_xXN
zA2vQ}-HBS?G8}(k9dYa}cU{mkSL9K|s_7}3G6VBAj_OS|J~!nQ`x~LIP$3~8(tFD-
z{g_jtksy{_o}@jXpI1v2BgV5M|HiPnJ*5`EYgga@#{7;JK)@+!oh4gQQ@CKLDriS?
zHE}TEH+}!Zj3ATSOM24xi@SJ7m>8c;Fm>Mx+Cx@RW8r@n=D;jOX^ylh=PO9*ZLQ9<
zoH$9-&rw6%WAVZ&Wf>oR?qyTDN1MT+B;TYkEPo^LDm;N?6|fgcupBtcqeH98&i{{t
za3c5oKWed;JVNOz&fj-Lx%QjEO`7z7S2Iq~ww1Z>DdY)^YckD{;l#cOl59obZ9alIWsdilVN8R1Tio+H%gQNQfZP{SE%!bM&0+
zegFKfxn`bOcdmW!wf93$t)a~(WdpSW%0nSAHZFEfVStLNswK=$31;a4w&vnw(*oMM
zIas;?xY)TlIWg$zWn6%kFfbG%YYBe?3In+KVRis*pc4$}Xa#fuz(a&E=m82q2+#%Y
zw*gpr0aSr7OC2vKAQynq@;A_cy29A3EM4JlAjB370W!f;WS~x7E?`?b*h38-HnxXK
z582Y}0A)*Sd#H!2Js4mKu>mNvtFi;spdN4|7{CaH0IYy^mJT2Q6a>%#8UTNEwdJ(5
z0Sa2`x*FO{?C`?cZca{6m;d1+qphQ>zzUF+Qqz$G0QFb_3cA`l55GD<2%NtyD?m*L
z?tkb54}3^hmD7>ZG1QRb;&@mCfD7Obba4egwEZt`&*97ff2hGzK`v0o-wgnab}*Qe
zFb9W+hX=c@n=6bR>SD|8i1Y~bUB*#ZA4@US}o
z6|glB;tG67l864|bcBx*o&>kU{!I)%2-rhShkwEWu0Y^_w6U{v{hh0#p`ikBv;;$7
zK!_#88XgF|G!B3=H{gQymNhR+EGb@Af(KLl&H$jNwH?Pp>)*ql)CJ>VoB8++y_y2?02)|BZOzg{|S|_D?hyC!D{{UyuvH0sITXd4c|d@a5V51>xoF{(|rY
zg8zc>slEIQ!in4e1>rOu{>&84!r@`}|M=m=9sfWscmU)t2&W1C3&LqS{aGcvu@le*
z4E<|eTyTG?E`LEd)xVl?!82fgK{!>nzaX57#~;WIr{eh+gj4bQ4gXWF8V|Sm@5_nv
zkADB>HTN59!(5>DKz*h=9@7i4@0UFhH&9MMp34|*AtRB^S*HO?_GG6RpLInFx^Ocs^hLl7x2oWCr
zq^M$F|Ev*n+;M=G4k8WEkCsWj=JGmz~OwU~6f9P33<{q2fmweh30pq&>p
zOCP4^+@;N9x*mhkyK86R3~Ue|)=W%XpIo6>3~KMaDQInG&bnUDp8>bQef*x%{pf$@
z*({>`CgFrY5efL+_C6Z2;?2+tqggrb@LBsFh7QTXj~!=pLP9Tp^rlV8cVE!q(*#jP
z+-2tSvLGw(z8Bq0EuESrmmW&(FsPtNPD^qjP9d7@HY3>&j~KoW5T$-V`x*k&?K?3h?@DJZmnhgPZuu@qoaf^fZVf6MS(6j`2-
z;ktw^qHavVllX)rGGPA%YoeTzY302FxU?^6IjS-Whe6S=AZW_mw=sWe^=!xwrhT7W
zVWHYU6AUeRG=7lKArTDtVSJV
zC0ky`L5f;c`~kmkd^(I)L9skxK=a;aZA?X&(bRmP8F|L{Yl>OOlemoTJ$CxuE|1l{
zJ%ixK*toS((P=WTvfh6zA)@V~Z_kzaVuz-We$Km@)b+)t0rMV=EcJ|5)syaBcV+&R
zROVk$M$Nipz8g18zp@(JP+X|ABd7&|~?7^98Yh&8(!sk4%=n=yh=Pp;IJj9@h&8k**;>wJ8j`*a4kR2wK
zo`kQuH*K==!N++Z8B{BNF4yE?Dzga;6?)H~j^F7qtMGp~4$(VJL$cA^OY30i+8fy;
zy*od!sLB+KQEPl^&GVjqFP~EhI1P9f;U(xAdy#uk3_y>h
zNokblQ}U>!
zUVeK=DyK0_-c3WH4O0JD=8MI}RMYLPSecQd26g{Z(IZOPk6Hn%!5Xa<6xbqd1FTdl
z!k6~j0Y>a=DFI@3(N+sf3o?X>g(ozRF_2^jz68I#`$Po_b+@fy*mCk%>%G0%-_+rGco
zRXBCuSZQ<0?O{3gc*>Yfpg>K%WORm(EnPDk&vQm2W$JVLjfr%3^91N70>V(ZRa?{+
z);E9V{#INNmyDuM$4eFO#}&D8^`YFwtH9e2MhYYE$$j)HqectE~pL
z4T-DSp(toj!_*ps&Lqm1CSxK31zldbpLo(dzcA;lgBQt?V&too^_TZ{7WBZJ{bs~q
zC9`G1j4QpAB=3!5B>A7w+-$<213^TeQzU<_K$f&Ho^D6^fku8#3y*?Np^TNxN^8I}
z5PRiKF|Y2=;ab)$$#IdVpJGV-8JV8S_PlfRn!;-1luK-LXR~J%c$A8+vf)o5-qhF4
z99nfH;tQB=%sh%?SS;SOAFcP4_^^;OVOpA2Ax>Ok9CA#M$AvSv>WRwqC>WV}FliWQ5}pF)3<~u}(pvb){eqL+hC*&W4p#zfEUC^%6+6!H+ES`|9r_E)
z4Z$Db5b)UMS|?~5J#duP?WqKnf>Dact9g5^u?r;jAOKWZR1Gmz#+|~M;bNYqp&KQ
z^H$lHlWWfy8CS=(ZhBJbZFwVgs%0GgM$$LD83`PDMPyzJc*Ck>-mF`vhf9BUhX;1+vdr`4+Kr|U8+4H6_-wrYEgjg(j?PNmxvPNyR=w=XZrTOpfkd;|&Vh|Hge
zqo;OByRopA=VTTLw&j{R+J1k*o_)A)U{n~<@n@$pmnUM<8$}wiMA+WoE
zBkNx#Xsgg~K9Z(vngV~y+siJXBuzu6opE!BZ$fryCtSKG+G%V?m2nU1IR$@Rh59s5
z5$VPtrQ;M(mX=JH&RrY?Zvig>RRhQ=qAsdutG5c-pXf*O{WM>;B)5uH4IP{2ONK6>
zqSVo!>q^Ys-&7bsW}MCWQlMHt_U)cT%b8X_?FGe^e#VCYGIW1?H*F(r#!uHjmr`5~
zP$S%-4(c!~rkaSh^~?#}ZXy^_$$)j(@cmA(Qa$0Ha*Kz5$W@>a{v4FZ$1*
zBsb0PB@7(%!72(^Yd9UGsG;30P)&rLcdwtgA|AZFptQPd$WC`+anpjd*3#tpM0Aa_
zYK!v4Q*|g(yHkI(Dj+37Tmvj#1(knivu9G`f0r77J~B_t*(%!;t^0A&o$|Ww@FjP{
zV(%hy)vGn!iz0^$iHF*{SYx`>cDqm490z_
zr2CDqz0dw<-%=|+e*J;(b@Hm>%a1{TZz${LBvpwiDU*NlOuaf??>?0SeYbT^N(k{=
zhWKh474aykj|oMsMDkQ5&+mc08CroQP)|@5L4((}03#ynlut`z*$_Ha-7e``SfCrC
zxo*oH*$U*N(J%(zGT!rb&uo;mi>C)V{`#n{lGG93I(PfMG>h%boE0jaJ)Gq4i#^?b
z#jF>0f?R)wHgQ%HP#2>vcUO-BFuYu~Y^{!VIJbgaiX*y??0NDoF?o$0GwdsAO0fG|
z$$2AFI{ZX$tg9z;WK=AAgE|oXw}zLWhYf7>De&=~5n#7hQz93#3e{epbn}Va$J}Je
z3dIE0@s7E6(-mdR8DLO=n46MGH*ysbDA@TZ8x4PWR{8QJH6=fdKHk*dBCWKPuN{dd
zKVMk3JubU>V@Ac<@utiV;vnPkIr6lKXB08?>#d)XY?j5m*H^`QLTk=f#Qkc*Y}u^B
zWy9l)D!p|LHsV(s6mOk+8Tl{M7GBG$<>YE^+PD3ySU()hOCi*28wg2<#@iPL^+~cw81|^H1*W9V;kJ?EX
z5Yqitkm>}BzDkAc31YpnklId_o+CkfjoYl%RE_NO~zYgSwWH%
z)mBz&HrD=U$4Y(!j}Vj!@Y4jQc;f|
zepuDf#Y=f>uqm;A3iQusLjF=#G`nMVkx6-bdpgI&r%ZckIbZg}|4xlETUwEpvUn9>
z(2d!2CsjKzxA*rj3y
zOCOvH9T$a1(?%B3O_et-+aR15bQat-#F_;qyf(m{a2WXhws?@8q6CSc_sa|Lt@kYg3V
zGB5!@RFPM!%7LsU$W2_0I|QY%NI!$eS0E;;6OnoW>RIaysQ`b*!cVnlK_-PHZ8wwY
zKdDnPRYBT_jFXeHv5WN1npS_j%Zbclm+ia~bvreB0Sp18jhJG^Xee&CYHPTQ92123
z^Qey=&&2XXu*FAN0ei;EnqN5G$qrFNGG`h|Omd>Hsg0Inc>
z2nn4Pb}048#hp=AZu?U7t`H*+h4AFxW;=ZI@>_pKqYuh8D0irpONev;(9IM^P-
zmqhZ~;jq~KW;?ou_jyLw#I-n1&RTdpG~M5c(5|tAyfKVm(&U#^Z89yrgkexqw)@nG
zi@{HUr|KUS&-@j#s3vih1_m*!0BW019!ipFh#q8b55pozXl8#Wia|VF*KuoRSY+c_
zI2K;*HY)Z>%wWfp`C{r?dRZQ#3$?~BAU_!dB|KWALPYe%E@!u)oGr^#Mc~wULs|i0
z5(4EpJ%t%TP{_zvB)15H>pgD+zpS^3RZ99|#k8}d=6XKiO|;wgN8?z$*OP>~Co$!{
z!H6x2V6N!SsY8F9g~7EQeN^mk`j&=7lfX~;V~19oGPAfC=i)PgBu&}(c^q%wNot~$
zyHXr1Y)8m&e&*~yENj|ruK$+kN&M-AP*JMQu{)OqvzGK2BAv(F{xi!kvXh$k?K(LH
z*SF2L&L&%Rhw-c!d>y+=$BA3s5pft8yL3X+{zPvfn1g?}u4NJL3(n2&sWw_&c0Lcc
zd~6s~|Mg-?!~jQ5rqI9Gv3tJpJM9nWbLQvjm-j>(XK$vn1_?%==fuA$KrsJlni1$q
z<;D`Z7W=9r_{Y|np{-N)XO3NwqxwuQAQB!b>qkU0qMs`RP(g?v9>EK6Rv;gSX{;1A
zc?79c){uV{NwV5&zX}JoJBH_@q^|-_u^LOhlw;Ykx)Kb>De54q#vNx+m3a)^8gxxF
zBi$i-x#$BL%e}NV9S%f|uDsIM)e`D?K1x9-MBab*2)%xs6fC(OkQ!HC#ouZ1mEDX%
z>BN%F=(y~xgH$QjxK@x)b7+yvguFNPNf`fYJ|ch8XmO{QLKD8Hj|$InX1td)T5XV!
zahCR)D50sNWX2RHvljPw-kr?(zPLyRPF>>>0fDVp!eOWm{&ykX(EK@v$=c6eu444D
zsc7t8P0px8aws2=Iw?Qv7~q2#d&Decg(
z6${is$efcP^hoTrJH5}Ie8mmiba&GYWsOAFgLlKUB=0Hr=BK9eUP=(ZMk+83sZIg`xk@nM5E?;T&rwLQiCTd}Do!c>3k
zuhVvzHjl*u7);MblCb-48W78T(^S*y>ZR1X_bw=1xf-PL?T%H0i37c6$Pm&Llj5V+fDLTb)a}!@e4Z^*)JA{H=(4
z3}C@3DvPiFn)xYCU2f#CNKr$YHqgs>xC@;;$&ye+AQA
zx)-92%1yXVY9P3fnV5?wC$4{*brV`JAlbyG=1D&n*BzbuR1+QMAYW4743vNL?`m$J
zMVNabIb!XIUv{iltnRXB!Fjlh5h1&sPWvoNG6NRD>jClwBuRf{)~|@!FcP9Z77?R@Gz4)9Ye0^rg6|>9Nec?|t|MX>`w+2^bHPSv@q#T;qEK_XSlh
z4&E*;5zS4+%NjC_2zEVfmMMQbEt*Q^&wio(-BJZV<(Y8#jVbXt(!SJjU>XH*QgQ%x
zXv+wrbTN#Xk#jy;tt5Z##2^LOx5YdcZSw4eZ8cN3u_c-1-SU@aTANyO+g&
zf%wQqzfwc_5o;id5WKNQZVLW+GlXSbIBG*DNs^&ud2#I%?<29`bp4M5ikh~6Ylt20|@~W@|wE_jD?e&^)Wa$b%Mo$)W+z^NFB#B~i
zhMKo#>9{593)=^6OrC$X3_m)z>%;^VfufX)6uhTIoJ2SQvAs8|(g|AkyZW+XQB|fN
zswK-p&_--`JJ8YB0T=*ZBOlyurHvKfh7>^SSYDz@uAghn^bU5?aY=Y$yzrLTr3
z{LNZtU`OW+=xw?2d
zvp@+p=9a{x!q})$RV=Gd&Y|bcW*>(GgD_|u`tHK3g4&ki5POftEm0~sq(7$s>xS37Hk9ttFqxJWr#Q$j3a4G&l;hZn0lfGk{*iyd*4Z5tZUF9j6^rQF-6Cjo
z3AhqEFZ8|2?Js>JUhm@}DT+g2{gL6VgZztM2Z{v2*n(ax=+=IfKAKh;TXu0c7xJf|
znsakUU0bZPC*C?$o|vI!eBEd{b9U&0nEOiaj|UyC3mF>D&pIH9{=P*E-e+ISZMOE%
zY=m9rbLxN855suZnk^~(2g6@_!Vhxk`yi`^Q&R4wLgTsBnUJAZTZ8^>j^L~??V%%)
zPD-;L$$R+%u5a7*ygb$)KOt4_uKZ}Ga~ov=!|od9QkL}AAAa-nWx|1jHIUD>D>%M1id*aEp
zrTKr%9i%WCU%BbJ9#ldef6|WHPV$q8Vo@xuN*@R;pA?AYK*kuqBbPZx4HWOr8t`Q5
zE9KlV#TMXb($*BAMmRbENd`$(}3b*w~iBQWBN%4&ZJPp&;xWmWuqK>j&h+ffsf5AggXAqI5v
z^5T7`5Q)tMwk_?qz@tRkZTkH<0TfwG=U;td2c@D*JC7a-YG8_TGMmMKTv6^jHbZ}+
zuHLD{yi8%!ZH2K=mp|DuMPX5M6l=2<9nC?~p$9n>ylfyuN!Y8hp1B;|im5P5`M^T_
zRK18z*(;a7{aFVvw6igoR$2PnbhAoQ0STEFO0(U{`l>J0*iW%VuFy=hs>haiB9X=1
zee!Y#H#_b{cd#q16WS|oOwmYuO4)zt`z)H+d*NNj2&^x!zut@+BYQr^GJ}skpnDsk
zRdQD)Fbv2~Syff{d_+eOHiU4y-l?DlN`8{}xjXxSl)Rbqbfv^%qh{g5B}Sqs|2Rh2
zl|5(CR9$g*)8$Bj_j(Rv%0A7=ML#K>nW*7HVKKgJSx|wO{)pU(QzWIwF*SdIxpYS2
zy;Wsd_(2hUk$`KooYDRU
zjB>ky&?So;jMf-RUF9@Emx_ewrWtPv69V2MR%T-$w}VX+XPv`4FbMnASUfS6
z*Lmb6kj7V*XVR7jnK|{D(l|`th!YXNyh`Qc_)1P5LYGLB!`XL#6nWS=y?%Ac=we5X
zgf$x^f_9fnD5`U&Kgk<~W6!M(b>9aVvi#FI>mA7+d9+
zKqBw`a_Kv{^L=G2)-Onzf`|5W*`0GvvD9NelbL)$886l!6t#NsQi6(|29Ege%m+tV
zoVW}DIj#{?V*x*jZ+q=b2Pcx>CH3+VCp;eblH0psQR@ElaC8<$S#kHibz_n4aO=P`xQs2-R
zJ0D+Tx~H?+o|U$j7x=vqGh_{;oUl5zEAVPOLVwQ8iWlm7y{37GmT;au@ubg&P1*Qz
zq4o4h-a!y0%We54(-Wh^1DTo@)(Fo>2%#uziOSf6mbfA8Y8rnPbgPOum$5~DTMR9H
zWoF#pkjd+6Oy4~V5&MzuOb~DCuuEAv4&{G*j%RM}<5fah%IJpE#rVY;Z#Pa9HP5K(Y8w9_`890}7B?h%hUk-*
z=Sqn~uc{mGG%+q%iuYFU)eMfTC95y~PeZn(((%*HUybrc3|$huA8_dO*SlKFg`fr{
zPpsc{3z1I?P7ji3HWma;=9Dt~OztqQe6TBPPk$XpdE$Rv@SM#4)B=mJb|vIc;NEuI
zuLB*QgC)okh8wX`9WzaOfB9*?k^n4k$4b&Z!X80ry1aEJ^uhO9t=i5(_StZAc3tBK
zSR;+P`i;j$tZS6w7suS-}nCTc3XxA
zFZA0?!M;bXMqNGWAXZ`?!#Lrxx%vJqR%2W6iykH#CyF4JPlb_TRoY!j+-1aei|Zu>
zl(_`A8McQqKGFQrcNfAX?@!L6VgYSf8Yj6)Uq63s#{;!aBh6~)1K{n#RgN)AyGC70m$%OMwr_#$?lPwKqEJ$$kdx
zYMxExedOhB^=bcSX5LreZsWKjM@9HfU}(7P_*J&5C;dDLqKA;l6ZQ!3v2oGC@vJb<
zRIh)=SX1k@YnNk>XQLvX6hj`%mzOGEz(5vJye-CcR#rEy{Oknq{*XaWwBDPrHa=W?
zEfL>2_6(9{AZbQb!>jpw6G@p=Ck^MIB|Uk1@fVe=%%Z3w$Ot6#I>>7oZ|zdv&y+@0
zbTbdcU@ML&yS;{?#K=@v7XynTilH!u@
zbAcyzsz@^7($q#3d>^_7I_QTSupQQ7_~wj|&7}F#c)flIkuj3%0?oV8)x~=@71T)S
zME5%&^@Cho?HSMD)>scZI!KS;bO=vw}S8Eo(o*P-5jSme4`aE&4qI;URG84qlKs
z8b^)vG(u1J>f!!GxoVWWoegD~rT;*zngok@cR0BQ>4!}KAt44am?bBpm~W`10`yYH
zRNi=&18cCusxR>luqxa!ta1$g%)*^jF6Oy^P7t6iDzl6WA&EOJxhl{Y7>aRT-~{6b
z)7J6i5qJD(x#i5DL)Ch;Y`t(j4qShIG7%Cs;;UHyj%3>YZfdV4RMA%P15#>&n?o|@
zk@)L{Av$#|d^GQT!CKSlN_tU>qQs
z{o==uiyyH{6g~1TJNQJOc%kb^K&%bfp
zQhG#a=8QAYup2&uZ$T4>6=ME>RljLe5!4ZqwnZslJLRR_U~^yp>@}=L#<#bmu`o^7
zCcYna)V+O(*sV}mu@f49{={B!Vlrco{8Rualkq$$f3;S5Jd|zQ
zE_)*q8e6!LtukYaP>C`2k!)FG%-k5p%$TLIWJyx8CCecDPRf=gDUpyOp)5%vlthb&
z&^Pox&(qWQet&+y`I&K?>v12~d0sQJQkK?AdRU@6XhDF`AdfQ<9>qHfkuQ&QTj
zw3)3B15D8vA0nONg9Fe6EMN*l!2k;)ooU1Y3Pb|n4tk;S9stn;f3N}V0b6UNl{H{w
zb;8!tS`o%9Y)vDPh~)opL0H?^8XW`-^ek+U0BCm*FtW9_*?hMF2~7T;2LTHkrhl^!
zGjKBWD0Jx?H}Can9P8Gs4-JL$VA_*27rPWl}gfv!|8N7
z%#%i;!iZ!~7zw|{f6vAXM*-+WvJb%gB!hTxOBfmf%aoJq1^z0qc{qSM4g(S>;AWB`
z@t4z=DJ3(BX{Y{8j41?lvnKvmI6wixf3@*KQ?_!=EiKIfUo?(D1qo;Zh8ajjQ)v{y
zZOg{|1F`bI2!a5DMka6eK>g(+|J~*<>iR_HlR4vqg3s4Y?@Uz%}hapgGfXY;DN?dz+jvQ
z$ozl=QP5{VfJ&x;!9l+}{*53ingAAup)yCFxi*lk>LvsaBB1rl%w+PH>p#PwpbBGd
zlOl6vu|xts0KkGC5V!@A${Y=a|DTQfqn9BKkGDYke}W4CRo6d*(Y`o*!2b^aBkUBo
z$y&jJNcKhJ|M1}`hB$u^Yl)*`ynadi%l@m99>Eh20!k|CFl99j)n6Xl%}v8IN1Ayp
z;WjTereFC_A9GqUJ_L|L0o2sDTp&}IKgcr=W0N`nN1Ew7S{NVr?~}9@izHx(SRBC<
zP(7ppf1t@^bO1z|IUcHq4gosQ2s}_fB)RATiBXPCi;MWpg5LAd5FH-MuwjIni{2FQ~tn8k&
z;(OEVJcnZSzH5yMjS_(E<5`4OQ}ar!YR}KzUs9x<%Nj>>+3AG?D)oY7Jiq@CpXgn&
zq29aOsXtc)<73dRdc*ts-SQ%#k4-s9e=W@d(^-x0j!h6ldMxL|X2&Jm&QQU3-Iogl
z^n{ys!aLvbL@H!+q_hqSn_h(v#1loe(CniU-IUwnt`X;{puAG#9+QR2Mo+f^=r#M{_yzy1}}7Z42|?+NVV!@
zqSExsrg%e|!Z?vpyU*Rt)l>f)PL@ZxM0okA;iQ3%XYlAHe{{#om!LJ9hn89o3W^F?U(pjQ5SJ&DEYN#)ckw
z_Ct`2eiWOCj&5^Ojs5~qe;!EN)y-XE%Nd}%cVc*Z!rL##w&$Uhz@7Jgg;2G--AfGj
zH>ZRXN^eYXLsu;JwME{5Ut~*|E_O;!mJ5BlI^M;BG#xv_Z41t*)#%=^uoy8KPVB%P
zn(7)X?@5vN9#$Tj+!$8?xNUnNbwOWMP;`;(R4{-5XB&OF=^f3aeumb_=%Q_L+7
z_`ay;Q#i((&KDSqc()Pxu7NVs8%Kwipc0rn?BCLNKq{+seZ=gS@a
zS=k|K`-ROeQMU{GM2v%1jx+q;5?L}i$|qjlB}wMxB!xrm4!e#63wOxq2m||#(2^Mb
zrNR5f$*u$2Uq+=RNtR7u984Izr|K?cKI2>pA0D-H)d?w#f5V(UIq&ea<-VTD7w9g@jXMg0c|Pi`k(Ab+JMk*G@NF+~eLE+sU5PGc$+M8s2t##=v0u
z$@UA6(}|~cJpFv`5({hMT53y=zL=8~QX;UQtzeZm*RA*K3%@%@$(N_ZjTm8_&kp9B
zcG22*h|n)|e>tY)SLewP`|}f!ROOV90cRsb$9mP{K9wL4{w`M^t=8uz>=ZCQaet5N
z^*{%AWAI{h#v91bkkFyjl7Zb=uD2-L@eH~Khg-S2V>;dKhlpi2$6|fut4m6YlHE-W=cbsRu6%e+bIY$@%V9k816qhJ<;z04p-q
z2w-Ps>x58dz|CHW(t2sNJmu?&4I}$lPnkAWU%e3o)U3R7uAFdvEYav7|DXH-I9GJYq84mp<
zWy^Q2e|VL|%kq5>Z9rMwjbpu&5d3&zIk##}Oiw4R&OF5Y$XIi=?5fz4OOD3)$mr9dX3s{vQ+&tVodWBSoT)M>jQC%xL!ix*#
zfBs^fu#3Sa@2zR6V6{t1KqLH-+!B}N;-y)kby?xK=gVa2sfv7+^J|Q8Ppw$cv
zxs*fAlz`T+Wp`K`-Ta*pw}%%X*Di83>uc2RK3}lLn-dIEzHqZ}D@xgoe
zhUg9r`2D!f03Hv
z3FlZUTa9v&pM@Bj=@Z9F?`jWz_m#Fh-O)1EAlVUK@_;anUwnAyW*0ag?XklYv4{*$
zQQ$qVb;;?3h-Ur~g`gbt44<(UI=V=L_
z1<#LXb`{PnMqNs-d&D8eUJZ*aoRa0}Rw~Zk-&7Kj&0Y#^}pze>-ECReT-4
zo~lYw&27Iz-8;HIvOmfABwRZ&MQW^B=cQJ1Y}@&-Cl^A3D=YLrVzQN*E3}6zPZl^g
znxRzV$}ek7zy7XvkWWzb2J^T60^Pst9Q(`ep5;vGIZoNkYbftqO8gIWcGe|8j+5J{
zb~iVQ8erRpUXe0+azs(kf9I(WPeX)4(hE(e^)%$VV##G}Fm{ayq>`uJ@oy8+qM~*O
zTzaimw1=ucel?VolAqr_X3z(v*@SYJMYjb-(BkTGv|`g;G1HQhFxVc*EWtdZvy4;A
zL^!swqn}s0SC&n8QIr3}tQPK5MwJ$ivi5W7q7FxM*i
zQo%`VJRHK?l;soQ68Rjf)c}XNN8h90>2x13kX65r>YhB|&&fB8I%sk3<@`Z)JYblRPjo5Ee{6H-5u+Xn$*y$qP@}dKgc_xngw&BWJtLx|oSfwGA;{?DcI1(3
z7tJ?fe$L*wjXv?I*267GGFN7!qQx;!o+N+vs^H-(ULX4V7BMyFpY{v0^T%0nx{`0@
zkX<%Lq%=~!Wz-kXz0IAG<+k&>OVYbtQHL;?d4|Z2divtne|`Q!Xz{+H8(oL{@JIS^
zr%_*O=Bqmz5ac5@3Ub~(<7KBb_xb3am)2;qCEbHO^H~>AtF(EY;P%Se-yzb+(NE7T
zi1kBnz|Zi?LKbvf43RH&8qK)5T9Lg_?GXdLWi$9lpe-vzUB1fDhP77E+pE8lvpy;?
zR+HZ(uZj(Cf4`<{)s4_tYP^0}Lo2~((Z|8qdE1QxbuGs_Gv+G-7|3?|Q1?mo>t|x$
zGJHXiD47(c8rFn&qx2#uXS@CFNE5^!KrKnq{>e7ssA==i6>(!=UtH0J$=KG{>&>(G
zA8Lt4wp$sSX4Jp#RmpS5U%I9+1vnt}vmK({V;PsMe~>c6-;*WYmPi^1NQGs}qD33n
z`xdNk_DOlKX_VjpTui4oF9(e?8VwiE-3&~lX+=*&eJg2*6^c>N>quLaco!{?#eFDG
z4hZ!=eIe+6}k)Bk8v^SV)9iQw8w7
zW?j-3f835fi5!$zm~bwAq3imJj$)(l(pu45JIigFO@N=tw%y$(QXV2hK@(xY)gQ2v
z)_v}QR>##et&=@Ep;lb;9bRc1>gCSkmpH_2mmRyEU7_8>g|Bv1u1hsBBxMgY{M?&g
zX<53$Yw=!AI4G%!*0^vf&Z!a3%ZD7!yasr`f7Tznh?pX$5ehDP&DH1#EX#kbFi0xq
z%K(puzn?F`?rattmER4$he7haQ|e=_ho_{z*ex(qv?G6+$LweGkyQ#R&;1Z2wXXR`
zFp`abb@g(@zMDG=M)yPK?%DV{W3EUaYYi3q2B+;8RZ-%J=(nBDFIXeKf%#IeUo;R>
zf3OTcFo2Ji;M9Rxr*Zie3gyg1qx#UJOztGrX3JucP@ms2gb
z{lOwKnag2tQ{+U6n{M03!7~xkwR5^E`@^WBFV8}D?DVVBPn0sAyo4%7#ytQEc6BIo
zKfn8MmF1cR%;Z7o+>hgeE9i5sVGEypf2cnmCVPF_ZbY9OQirao&F7ko&mP}!E)bn#
zl^IK7%q}BxEs9qqy^TbjC9?8;v}A~E6<munr!l0mCbe|D~A
zIcLAcM&~AKaPko4X*oab^CS$2GJOGK@VyW1xW2u4bW+S|-T3U6toEbzIa7t4x;5j{
zUT;2j2Kp~vANf}B={3t|>{BNVMRS_CQ#{8WYNcgy`mUwPy`%%C=~cMyZN{o69QpY@
z?ANk1wG|$JVJUlNcQJTG*VXTsf16`hi*QkcE+tZadUgX=S}5{U_LKL9D@NXxcSsko4Pp`2=4+T%eV60VOp9Y{+q5|Po>pCh=kY?Y@*lR-
z$lhs%Z7Q-a>la&}E#sz5f0trbZD_asiyoDSax>7TpH94abAXM<5tJgVi_711V+M^zy0^*YjJFW)w5h!M37jIf-R=
z4Nfdc*d{diI!!wf^W1!)4zJ$Ld#s%kHH=Lfb^_EAhimYrju6Cmf0Ktp3ZK_rq*Xv_
z-RQJpOF2`4)FW>w=Ibn1&1F+p*ZFa)9?b9kd;Ha{eV>mmTRMLGV3cDt74%3AU_E%{
z3db#_8`TNi1;H$}j?Y!d0e9wC+wn`#%T>a9GB7T0TAF0sArI}vCMt_j)O^Ixy8$Qe
zYy>pvRwbO5Zkv*F<;dn(O6ghsC49#9q2R2>b7s0m_By?qo%
zw;LSvw$EN(f1iFX<|z`$sEBm%61(|XvHDxl+g|NG-bdye~@-}(2uxMVr*9COTjKJx*kB2(3160>kL14=uB+?d#y
zS@{8qO6radrXV&}CUu~dyS=FkfSs9@l@pPQO2P$b>Sp5zk~DP#@&kC>tN|+KZhzoh
za0DwWA0ib%1_%PWfSncqGjD(r(9KlS+X=`9pf&vqs5-j3F`1dVg55xnl?@0;2TqZ2
zbn35~y+2YIqc~f&cM^9Hf8-OXu0wB+<#0*e&^aLAi0JM%EfEm!*
z)ZP-{XbI2+z5{4!NU3W8WYkr(RDU(-n8AfL+?|{pUH*@YgodV;3?o2NOj%P30Mup#
z$Y^P3{{F5B1cCWmF#?n|!T#TUz=6Nhm83MqG<8*_*jRqA0l)_E0J^x^{BHYS+^E6K
z0Dq`~Q!QN_9sX(nptW{$bK+-V@$~d$wsLoMV|H}0Vs^6si=U>ojVr*@(SOAb0RD6V
z+5`U*#vNnvYfP(h9;9DcN{(Q4K+JAlZhuApt|um`$PeIUpjU^ELf#~aM
z>H!3}xwr%UegEnBPlU+E1F*0$cLR?)__!ecRb38b=?LKa+YDy%FW3JK0qtK$l@5GP
zEgV7i-T(`rB_fNmqkkKCBxwKtIcfjZOWNJuUfI+ENc;b2`rm=34mS4Q{~7vUAv(a{
zY-yDpT^vm9|I26NDsAHhv{1EiGq?U*+Q02`Zl>V*7Xw+@1Hs$!mr3jQk+KKRF8KLl
z^ZW4yU}EFs{x2VREY0meKv!1)=ig~Sun_;n7_9zpz5o^}Nq;>_30;Q&cWnNOlLDDL
zTG)WB0PI}c08lW^z{%asA7JU|g7|ww
zxw!x=V!us)BW?hT_#earV3GKPcmXVue-Iy-*uN1ED}Y7%4`Ktb$oxU<02bLlhy%bP
z_XlwTSmgg8aDR8jKM34i=??;TSN?;*-BtdLc){IO{~&Nb^*;#QPvZ{)_tX4?z-!m~
zgTURj{~&O8oqr=fFf#pvz+BA!ATSs6ejnH%0nFpCQT(IzZ@;swmWV1KGi}EN%W&<@gOf{?X=l
zxTCwvKMH_Dto{MPysiJN0<4v_x05vx^p6Oz&E_8vOx*S#5UiQqKOk5b`+q<%y?>Br
z11s{UG8Z@%1U@GJ@Pj3A{6hwu?D!vU@M4_)xWUDoOu;vaJ<#%xikxi!XLb1xalvPk
z#R=$Q}W|Bm@x$l2Y|4fvn++1S83{zGLp
zuxNiObAeal3Usjf&ynH!9Ru|Eht^!+BCa-Ge<*?#aF3-akT=Ox;{;y!2VY_YE7^4*vSr-wgkw
z0o9+Y@*mma;*MUvOq?97048=m@Y9hKymau)@c92nuKC|@AAj9-;OFtb@b9Z100epg
z&3_RWW*yB1!feyOMU(|d6^@s~Qt>e#eZdiaC!YkfkU8E;h$or9Ne&b#3T_O_qAqe&
zl;t-JNDT&+yrT-kwtrb|`cOG`WuYdzWg4IqK!_?OHk7B$tQC@_G#^ynPENNkpI4y!
znR7C$IjNZhpfz+L!PnYxAICoW`t~A?+<&0FdHEBpXYn-KcPtls46k`~f;PhW@is`e
z*L$plSkrE?1^OwIf(V=;c{r!84$N)H*kX~V7~1VD{yR)5+I#2KMXh~!{V3NqBq9jn
zKuV%=D;WSyiUA^yXJ(HQZYuQqF;vK;BG-2J-~zdE37Xu{1N?@!6Y~#z9G8B&qJNr8
zA?Z>_OWnWH?I`&(9Atddvm&foM810oA$ba<(n|pnvpyZle|=Em9m!(6cE4MH{`yPL
zL3ar8W%JEl!-C11=no{ItiujwsG-7$=MxrrL@E9>l%PRK@j8`t!dn(AbR+1uN_fP!
ze2LI{+~;3TrGJH$;;{n6qZGv9a(`yt>!;VqOsRCZNUh$!Y1o0-BmGixt;^drxXulE
zj^5M+Y9c`;dN=c-hTfAjEJ)gM*^O618j1fE&x+gMLKS!O^PnACOBr5HhG7LPS>+k!
zLo0_h!%_q+wvizo=B9xB$HaN{CHj6QcZuUT&VkL$_wMm$4cv;@wnAhABY%k|QH(sA
z>jOJS^CNoiQ1i73ev%}9G4!Nc`&>IVJQ&wV*Y)*)3Qf5|ajiB+dm;L1D*0+r?jw)8*F(6gcw+pR5(GXF&O(Ly?W?j?O)@2}Gxhem?p$}7SU%yr`%5Klgs3d>s&9`|;}1zE|;{ue~(%JI`Awyz)wc6L&gnQXNNYz?FXf;u*dAwuKZ_|l({m!lw(cpVX=U)>?>gmT}7a)h&g^gtLMDi#)m#4yE-e384V
zdT@q>H3E!$tOrObAu{?Y@Nu*Te?dDVB37F;n{CvaYjSc&9-qBXMO~_(>XE*;HKF~4
z?!Z(cTnyhmQDrX5E`P`1MDams;{F+V^Y+TBh-9t^?I5CM2FXIdG(afpV<7?*hwH1m
z9Ik|gBxd(<&KudH2M(rpOfj#Q;}gQ`+euJIo3oVp#V5BAu)2yyXz)tcM$lU29DSV^
zGuK>IDb~#4`h+X?`&wo$VXc`-8Rzex85`O^@~!YW?1)VSLw`EVo@w(_hZD?fts~8v
zg7S-S5}EiLj?(=fGmVCiwoIO;YV%caLL3RH!%sujlVfOux4%wE+
zkd0cL^xI2lL5y$dY?0b}IJpaTQyY-Q52h07f)!&%@mGmK)$1}IC7Zlv`(%uy5Tbc+
z!>YLA91S$tmw(C@whWNt&=;~K1&EsHROS__^;6{E=u~beV)^Mc9uB&W#PLb*plImq
z_|4yPMXf`gZZ>MT`IU?%qOeoa1)3rkUVb&p-E>(?Wi2%B!%}DO@t)C06(M64Djabx
zQ*|nPk*NGRX{!k5jNpYB_FnZdO8_do|6z%?(3uF1Mt{KjaY`a48*lgOmsMc@F$==LR9eXbN1w&G-fVINF;+I1D+A-pu<
zxJ@3JedLGX7;^mNsn5aq$_&Eap~jCmkg4|b{T-nXDG^Y)aP-rU*a
z;(v)oe-wPNG4RXujH_K?&@{ekN{M6Mhsv4-QV$g%K~qp0)rt(HmAMMGs{O&3ncn_RlTqqSN4ECUCypck*I_Cn
z*hAA@2(AzIAXX>~#}P=^=r2u`>aF9mqd9wn`7ilKQGg4can<=<5gzoGraR`
z;E|bPqn_TTdG=|3(Q_ubrT;-=ymG*;fEqViva_I%ERf5FRsNF+xlR1g7ab{sLw^WD
zOe-o{dHRRR#zm;;8c(_VlIfuwU4>aBQ+2_d_eUI;>p7Y*{Bhyr1`@MT(QZR8G?S%}
zBT9{ahy1Nma*_=}^igT_$QoE3``JhuXXVz#+I5>pEe1+CvgLw`0t{nraUN~dpSn~6
zpBC~sD7>umRBGL}`ZzVKzhaesP=6rLm^0Q|msS?9ozHi8b{bhqaV-6Qu5WEk-Eg9$
zZ}N8fC5kX9u9K4=@(rwP5V`gvh7MK~nyY@kbd;ug_1>INx}LXyx|nBllI%}g+5YrP
z#pI#sU(js+-7Hgs^Q5HD!z;Lm_BM#_pMZ}!$jPf_ZhK9fg#4PNrdrqz*ngzuGP8O!
zTu>e(FkZvBJ}){G`b3lgfqeoauaWZZ&MH%s&-0dI$}?vxjIYuTDA(D8Fa?KUYb
zZdyY;iqfo)Vmvd(9&g^`2hn#Ptx6|`xAv5LP|Do3Tj}FCMeG0)PU#`I+Wq{QPqThT
z*E`35`U18;wJHf;g_GuCyAyHJ*69sp7u5`8GwK$TV{Lm
z^uq}4o!pT5#`%SwUw_>bVY@*47YBL*>;70D?*gb6Js514v^m!M7Z!mOoF_3)!>fz;
z{e)`1BO$aMxz##vNOhL(=al_eaV_x7(*z4E3^9jbA<*q|XC^>MQqMzH?gJ4T<-u!F
z#8-(;mEWpVotfTMf@E`%(HOLf6Dbh~8DHLbW_+?$Quvj*I~|UX@SOnV7~v
ze(EsBkL;d|*oPs%tTs3{-~uCwsKb;5?hp&r{iQXLg>)CEh5=c22)!w0fLlo
zCyF1n^@b$}c0KFF016Y|U3}!UmS}1LRrD~wJL-qAnSbudidnA3z!%&$7xmClNWol$
zZ2lROmch^Ts&#qXP7=+r`e|X1g(Jb~4rHQV`W{3~jJ5{k(aXN8i5K00Xm&buAn&K=
zw{a0~pF0E*aG0Z2`bK2BR7?AL+B2{_-V}2`-@rAJu-fL|Cgcu=%Ex7A@i=5@WF0?l
zcVt}p$$xW}y4p%uXB)jDzQom6`T>!F^xtk+A>5@5Z?!Lfa`8fHc{m
z`@=<~nWgJfbs2lDtR8kG_mdGnsOVT;iASGFd}-o4ijNU={nN)ojZ4>4%2MX#<5AU|
zWV@NIA2++*C)W(pD!Ur$*C4N%nDg~ymbR(6R)1?K>hq}x5@PQ(+F;vAd&lPTI~r7I
ztndxtAB18=8vF)*n+@7^+_KvID5e+`gOS}igjw(VbUPM?1@3zZMZLrkP&qR`#7k@p
zbT;tV8*{&jqHtP;=f~}~k%O9et?yw-sdMO2dF_o}ad_l}y|h56_Y}-Hpe!H|D_GAF
z34b+yu(SAlnj;T4zF2GY#qF2x_w88K^yi}hj}MrXd%r}0;{DUPaA<@nJ%c^Viu$lD
zBI5Y(n4~^r#qxW@f2OMY9&Bznt9k{~jMcM0HIw4;QAgmRPoWdJ^W1dh2pvEsN~ak=
z1bxvab0I6W$0d}x+7~rcxrV&1j_haJ{2tG--b_nM0Ss_*S-
z=zj={od_l?rpx;Fi(vF6+AxNd!qJ)IR%7)-wq8V#6!U7bL&N*~oppinq8a17NVGEh
z;7Q;2whc+5h6lVhTq#iW$=n!w|9_dTjlgx!09;^U7$FLw+m#OcBJCPdBO4$}lF~#_
zPwcoQ4Uw$@y-;8ntIxVp@D~3g>B6tW9jg*Z6o2^Awx(CfuU`PEh1qU+B4XQ)jM}XY
z*}GM@`#8D_nDZuK3{|~{24sPK;hpw++cJa6{la`p#kQ-$f4+N_#iCsmDtisHl)Ek_#PbI@CT?;XlxRBB$6lRv)Sh6e;{c`ydFpQDx#)5;jD>7z99`A~eoKA^B0PGrA2wd+06
z$9I}~(;}EMe5<;ouS=@)RgI?zfkWv{N%jn4e;zO;3%^5j5ch>P-9tWGk;9!HUlafPw3~)pvjoP)*79c6Q@XBJ}mc#oF3s~w5$#~zj}Q|
z5SqUCf>QaK7A-@wP8)Oai84GyO(Frd?ts+Vj_VAb>1T)fx0du=#dX`51Zg@q!0Q&a
zZ4MlYgvNEo22~9uWPje-RU97sFJ2k?1cDD`1o1PISid@ug@2P3JItMbNr}q(5l3An
zL83H|LOO>5qm_N!PVC9MU44S#&4;s!CMGf@p9idjfs`nq-cz%&
zz~5poIZFn9A%9#7M|KE!PBjh}iOKIox74XOXwKv8phS6fE4bBsnOLYiHhBZ{-OHXV
zGZ9caV#bn%?7eNShdx%9yQx}Xp^Wd9c~9_1fCPDAkTSr_H18h??V0|r0aZ+u5dT!Tf08C7a1
z?Z>64WuphiPz6?Wk9kZPSN3R*rr69h22N{JZ`2Av59Rj+bgQ*V(Zaeg!klLchzJrg
z(^K(JP&HD!Reu!erYJVqWgQfu5HDvl$Q5DS5-&UM2@a|;J(r}ukOr6CXxiq7AE}9z^
zLD!7>YKwh+;{JR)AJf1B%TTg&p{3mO<_`Oyi&xU=b2cu6eB;9hZw@MkCuJ
zq3~tYEv)Ho4~w=$ir9B3#%XbtW8_;92W{aD^M7%BZQMMJG#+tfG!yyQG*<}vJuV0v
zh)}GBOQ`hq>$$^GErTj|?%399)a1A;AzC-Fc%$bRG3nC@QE`}-QAaWb+l+$`t16P&
z_1nmr6^bIYt61Uzjfc^R1Aa}7eTvtClgBYd(QkYYm|kZ#uzkbec=6fK>NGG^?V-+#
zynm!|$l6fC)UPHbn?L0|668}BmI}vB>os`LI&L4KDZPWxdRr4*=48s#!uurgQ$@G#
zay=5t+U(-n9xleC5LAUEP0Mlect#y$(hO9D33Y_y;sBd+#K{+q?MV(M$#wvj#Qjs}
zdySDHs2F_xT_S0GAn!}qjX+v8S4+SLi+>M^$l3}D2!sva+-i_-zMfaPWA~qM3(8Y!
zm%_e}bPhq;MR6{o`+n%4jD-YK!qrF&QwQ0{nXTPJa?!Z1J;Gc@wm5j`o{c-^UlsUS
z8>m)txzgrDO`B)-7j!*ff^DmZB1yme1B?1
zT7cFf8MPHRUEHnfWlb(h@J&T+32r0CB|Ny?<%l05nV5B(NvIq*Coh+QYB86Lr=>>K
zEx=!mCEJS5{yWzX964haO`w_Dg7K{CU{p-jg=)
zqDqczpGtZw&dY4pxE+Mw%1NgSf~15dbZX6xl18?lPgSZ7I6qG0j(@E!kSYYf=6Zi~
z1BHXuAiqkKne|mGTyO4ve*m0&5f+W`v*hu3^azYf7QO4eKE;9RBxVmC;(r&9WaDd1
z8J<9fQOEA#EGLs-@r@ko;mPG|>BKC^+5O%qf2$wVZ>^7WdC9Na{e{rQse@9@0WWbt
z+nlB7+Zxsgo*Qh!(tP2ZeWxAWfZ!_)v8w=NhrJNe{gj07V@5$HNQpBR6r
zx9P$b$Ke}4#HzZFVtm&i=<`e(-wmTqv7ln#KU@1kqF_N*#FIh*KE3un1Ek{nJWM;)
z77vj{pR-$v^KUg7gRP-1K7^l)6D$B3Y(FeZ1hu`;-=ogH)f(O
z0Y85zfVFAqU3F#{v&BEO6(+oOOzeB@_SKY^?x?5sMBD+Ujjw?b|;)T$dNMf|co7ZP7a
zVjt2DO|K*|sNYsv+-Mrk)%Qh8EgjQgZ7N
zi)Hu_OtDk2l1ghT1(q&X~XC4Uv%_$$f^=2!FPGWDzCYkfa_
z3
zY42%s7se`yv;fO>D*^&!zdhgd`2@7aUe43cSB(ZR^DVWmm!D`2PjYNf$qf$j@=sul
z+U2>{+kaO`sMY|7Bx_t%Vw!}I^_wc8#zPSi2h6woUFbkpD-~9u(*LEIeTm;(%LSz}
zI-aqE_m2f)>THlai-~Mv7(VHHM+Szr(VBeu`GXT3(*{CzHb)iX>XL*N@=9@9o6YyM
zn0`)n$UWPR8b>uz@;>|)+G!bGo#Az34vwa*V}CZXRAc5{0-~d2KoC#<)3;FD@=pIE
zYI$X23HC^zFG1{>mGi=k&@b@}}xk-RAWdGffQ**d?Iu|JQFMVPRyx
z7I%$o*7*ggsOtDSV7^(&7E{!>=q$>8YpPM`fgk^^O3hs$Y*WSHa$%<&ILE-p$;tq>QX?kOdP
z;~fS-ktdRuJC5AKK8WH;3Eo_IDEg6pC%yCagM{Ea!;T^)mTZ(wdrI6XvmeBBuW-h+
z97rq&9{SchxrPNVhBw4Q+Fi9Z$Df@E-GBS4k6yz}ldk$2T!Z!a%e)32YX}wjdi;0n
z3M`q)IF()d#bmoq2Yi0ixv^HiFT7J@Uxc6{)l=qQ$dy#Xcq-KU#Ihz
z;NFK=ffA<~A<6}r0qKi}+&uW1g~{;kJ-kUQtLvoh5$SrkFRVXgR((%bznY?odVd-v
zlropLBQh80xe)`tzo+toxSUwazmP!s*_h?m%Tzu5=cBC@r)osZ?+-#7VK6Aux?g6;
z@|wSlDJ?a~4GPJ+Jm49L-Su%(m5(6c6j{Vvx3{V-HmR=k>-ANrXk6+q+RgiKd9uL^g!H+opE;15XCgW1zRGJ3fm(vr$GUWsB88NC8zMS&FKp72
zvnK8$GlIy?fW*uh7=>M6ZI5K7UNS;~||5
zdo)S*jO~(tl%RlT)E=a*(gA1cc24dq6RLX06jC7(&dvC1-*f=&7s|4
za`H!8%Qj7e(cxrvrNOV?y7dH3;{v3pBMBK*ufKZ#yboF#@{Km%>TvUWpcO}LhBGMC
zyBFUb>$Y%E{qTWm5&{>(cz>^H)Yc7sixRSMtsmj#`
zj}sSxSyQ<_Tg*mOwy(Iiqo)CyxEYfIIwr{gg{i>3*a{7wvnt#7?MbskS-B|hm
z3?xcTEPdrJ*M06&#h6b`7%eJ}1f&8VPP#_dQM|9lF)d?-H|bt88-FmvbbCh|f$80Ikl8u@_3O3ANY-dIFe`&JtO(d*PPAu2!MqS~
zN#mZY3i}on2I*aiwbpVhN46MD8sr=kac|;z{SE!iNqYpug=|;7zNl$?2n{JuaIUD*
z>v;`*>^6F6G8}wP`b2quT}As#vajdhb|Ear4klor4o;VK&wuZD*7Ypa3&Mt6T=6|<3B0=;opGR_6gT#<5+;IDT&6ysYW;JM4HDQ~@?wJB}c~
zaT{t#VR11#aRZre)xf?FZB5wf3^LMNE(ja8(u`+siP1qHS*KpsZbFZ^ECCA(K;
z3fq8>8>~;c-Ct_-69UzI*Hk#M`Y6fLzXCOwe~DCa@`hBxr_BsEmt0t!Gx4FabiPZq
zerLgQG=uoQf6QSxS)N%{H6SmPd=w(1-K)H12eNzU;n49Hf<$LJTw*5_fPem_iGoy*
z35C9X9)H5M1Jjz9t}-2>4xlH&ijz)oA$v&WLKhZ6W0I=GxIQYQ!2gbV6{g`ATHD3j
zdMAokU!S}sJYtIuJ&wdvx&vEm`&wHO;@(dvGgpsOc%m|Q*9`EHhYC)uYVUqx@yqCl
zV^UDWYHJjBlGaE~vTV4Ehz`mkFfeJ4<@E
z!r%FQxn5l6x$P9
zwO%eykQ?OMN8$SUMAfJ}A}@_?!p;9Wv|^M_FZjAZX6ABZr3v!C`~-8s-xju)mcLz*`CSKlMdS7;Yu=COtp$6WkV{Q%j1FUY`UQ<ZE6!MKq@)9)xCNxfbiG%0Y08yONE-c=kSV}*C9YYAZ?
z3Z!O~Jo_fxR>HsCpNrg^Zgnl8gP&(8CeO`_qhKhAOYoyPNls?-$b3M#*N9VW%589y%uHhbC;0
zA2gwyOSOE7EBdJ6O!Hla=Bd`=0t@y&o@>lg_UC)@1!;7dcau^5W6_Hc;hPmt=@yFL
z?uVvkp?xAkH4;s{=pTB`W2*RFemZV*71V~gGkw`8{6!ZA=)Ht}ULnPLwe9{9Ix>GC
zD9`p-bf9MV{g9qH6&m_{>?54c@Q$i&l0Mabe&XBvmyO@+APC)A2c2>$!dRTv}_;z
zlZoRVJ*bmAXfj@bTJPYRxl^X<&0)D&1Jt`ljdA_1v?t?Q^8`hVlaLZPgf#S%BjhPs
zZT49MM!JE!aROACK#H?W8pSsofB6cCpBMsbdK%`_y~j4Dk#NSgVSCw
zUJMN8#h+#GSfIi^8@J2hbE1#*e%K=GnEZK^
zB`;0+Gk&jc&a{sBk+s0o$tRtB^#I5@
z6^m+==SYVb0l8gvM7>u!Uz)_dZ^NZ~=j?4>UHsJ5ud(gANz&MkB>#Q{DkE&Q>xADHQ&#nc3*!{al&NRV+*478y24$WNsdne1$!|tI8Y|7niv7r2sQapwiR9
zfb+C{=cs`n3DsbHa(aqbX}F&WySM{CroE0(kI&QH(DVh%*TlpwZO@5psj`kC>4#Ord^o#@FNT^CdAgd_^6(>tB?
zy3l*{C0=WdtK)xK#CBmNZraKC{9OGTLyy=SWkNi*mgC|MY*ZdD>M^a)_Je^%fzNaE
zsc%RgiPPMF`EpV{dgjRw)_*;NS^|6tR@cr-yuJ~?9C{5FX9Inj#yR|@(R(H`3Y
z50%T;1hJ}KNPI(1X(YuO6j!847a}-EHZ5UlY@z*^sk(og#>jEoGX>ahoxYD0-UPaH
zy`BVSgC-=HfE%pgGVXCmB8|N|A$`?j-^(kEsH=pN0U-+=AZ0reVUG7_AorpZH-5sl
z3NL$bI3}aFefN#Vo5@_r{uKT+lsD<@D+S7>7^&`8T4p_Es}9N?y0T0@u3g%~2wyO(
zZVPvn=Nf;E<|WYEDIp>aVaCpxhe4%nDN%9~(L!A4lQI0+HepHro=#ES2>u`Rbbq;C
z;}R_4aycc|Bu7;Y+!S!E(>v>b+mu#|dXj0nRq{PkA!#M89IvS!ZU31RfUE2`{dp?%
z!l2EWl7~y8peu}TiJXSgm$it2438KFhXVU9H9g6?f5
zje0r#EJH~}x{)(Eijfc*1dHa@A$6T52k*plGTDy9Rc${Ij_qmBxJF-Nqp7Xmqodue
zrGF)PL2Sj0@Z3_>LYKr=!iL@IeJ7io{>Gt|v!{?X>U5zsLD+t=ubq;f>(m&V6M5wyZ(I`6%mLQuoqyg&6qEdQn-8BwfCF(wm(xNmr9%jg(NVzlyJ
z^{MfL+)Q?uILxrDJ|j>B%1q|&yf3}vxrFR((tNnXXY0HNR?MDt9b{dci-U{}Uc5;`
zQy77F!R^&YqP&SgEz
z9dCqhuuX@HuyS+}Jt>bAQAo5-K2(%fG5M_n;i>*7Q*fF14@we}tHG%Ly3aB+mGpll
ztoBipQTYaNo0j|MxVR;(EimNb7>V2nVLvivy07VNE-8pfN-g1s1NQP5z9Ac3Rb%&v
zQG8|*f2Ly2d<(3v*dq1!cQo#oc)P7S6qETWzp0u#Npe>XR*(8qBm!kY_1F2!?v*ab
z3AX-AsDk@P;L&iUG)}V1K4x6$ptJPqMkeo4s6>}F&A!K>TPuUqg!a@HAsaAlEMw*B{0x5*K-z(f
zJt=i;*JluxcE)vk>ALm6z}siTlRdAoh-7e&%BW_c1_C%SvFdPT%a4tBA36piqmEG*FxJcQ2|?nM$9m>
zR+_ha-N2LKJMAAM6y2FlcYA+CTgmg!Wr4K|2#5+2mda5%*oqbg)3%-f
z#iG%??^-wlQLV*~46{)TsnWi*2Pm4c``pFbduaroVub_A~3Fl~+2l9JQ(
zetujx$LTeYq;a*+HC+ls8-`<2W^bA#s`JZ`S>t4AMMH~vk@2I%OyoD+;Hw$`bY)+^
zm?6Abpyz7Y-u37KK%rrD7AX9NCEyfDWoD?ceAcwdYuAAQvA%!)_E+WOZGsV*Suzyo
z$FNmqt=Ee;vvX`e{7ro&Pbo_bCnm9ldg@XF=8Lq!PiQYNW0CHj}z%L(GxK{7sKy(YA8YONB2H!Vqg|1ok
zY0g6nW_!`Tb7+4J`$e=|oK?$%z=DVZ111#r$}GX%X=-UpU&Mru7_-~4M&aZGNjnK}
z!8YhNZ`J(CpI$y!15`F8a7$3{W~!7aj)hi*=@Teh$3D%p_Jb}blD5&h=}E7KqK@le
zPB+Wjpfe=Htk=ld1Iz3skBIV|saWJAaCk@A9gg!-Qmo@Kg8{`d1Ri!Z$RwHYC<
z2S#@%(9CeWH_O}jp|V`Dl0+~i^2lO$gO{@Kt1A~swXJN@1+mF=ad4+6@rP3n_AFt+
zN?~+Yhg^R@Gt}pl(r=p?JA#lr9#PQfCMAk%hx37LkXPObeHn~v`nBHkLCNdbx|eQ8
zj$FIy=y(@Wm;B;mWC)Eb%M^5tAuaw-%P8)7Ht;|J`&wN?sg;A&~
zsf(pT?_0BOv(Z@ed4k(#PUH2K%rz57j^YZR8Eb!p_QyggPPQ5lwj3KwglO
zC_a_WJRLnM$79>ELy0dL{%+Umr5ukL8$e*aw8z&;?7(e}K}~nr$DI8tmvevxpUpOF
zu#kW7*`}?R*OuCfr-HYsK_w_;B{`&Jbd}O%(Jr-RHAh(L;R2I=pvdub=sK!3|EjrY
zW&x;eH+Zn{utg->VX6?yxgZDVI}UNEEJ_xdD3OH^p3hx(#)F2=_M~Da>RcrT9ei!v$7idqUh|LWHW^>LiY{`Y+wd{
zqf^$AsrE_R)71QagzD31vcYzoxnHFgu}OZa(aTXSKpO>yup{QrHO`b?%c_9c`itiq{!Rb6ED^A)0zXkhZ^=t4;l-!1U>Q7=KVbX1#w-
z9_h1A4A!TW$wrVY@O+B_f&xD>4H_O0O4b7`Mpq#?xI1i}mN&U*{=g8%UiQ`JMlsxB
zv8yBd7fzkrXaJp6VU)wP)AX`jhhac?3RK8B)UKoaQQ+N}I2`#QXZ_X+lii%Z`1HAj
z{Ukw@<<@rRnOc-%J;5gVTC~+jkWqiH8ii4RP(GEESmBEdM(wz3)`323#BtxcPY!lS
z(%d!AuI;rwd2w7E*N)(6*O;Tg0Hz$@2yHc4x}A89t+{iyqy)oqfON1@85(SLaO#LU
z-_d~WCy7PwNRc*T|Aw_ON>v)9@5XqzFi-r8*z{TaVX-}^L_d1euvT~BU6p?j(0foe
ztr-WU9O}#`JMh)3PAW2|vcm(KabF5>Y7wyU=SSJ2=LsLJahAOrqBV2a)w16eJ`M?A
z#5vDyb^TBf^yOLSU%OH|<$~%Oc!BuL=+DAmDW;!Bsy>A3+(qGaU!WNd(e~Pn@&L7*
zAE?uCj-ygRUc#dMKYldQ&i#KPb)v9@6W1J1bC^@aB{nynJE#uxFh-DO!O)H?{?=o?
zh%@nYS&yy&#txEAXlN7nteoooc!WA_!rm8m-TBd;R+Adm1_vAK(mWuF(7b~pBntJ3UPmZU1OP90tR(v
zx@m)7+VSh%Q7f`KWS$nafmV>R=)RmN@4yc%hg?xa###294MYp6VEucHzyDq?SROxN-j|SfR~itp|D1Rj?oBl|QTfp9SdTY*qw;DR
zXt@0OmR|?sC2jEVv*M5_X7qTc3hQT(nNN2aE;Q%1rZAzREJkj8!DjEWQhdiNJl(@U;LD!gauM%B;(^B&}l16!zxpSdm
z!A+?2h-ejv-z+Zfts16}!ZBmG&Jh{IH;W)W+m>^xs*JEZJ5#yxB&wHbO9o<`_
zQbW8R1$=+|432(+42i>6kn3-W%iyq{B&rHamRW8h-Te-2u$Sg;TRG|!E)Ol{>q*QD
z;?v_!IrDuoGkKljHj-i9Nk?7tb=U`GU#mO|H)?D#<+=w?!yj%HvD9jCG+J7RHGTOX
zZ+POy*yhk{w%$FE>u?Yiy!ogHrff)@g$SF_PR)O}T@CbbKJOKdEP%zqvgy3^Rk8}|
zsB75OPx%^_L2A3O`_@#hu}VwBQe~#zWwHKfJ_4VUj5c#RHmbOnI{)XsJ;&ardvqj#
z28qr(R{U;k$~DPdNcble9(I-3kA&;L6_m1Y%EbH8J
z*Sox>+#dc==OdG!Bwf#Bkz@9|<~lzT6}*4H!}MoVAs{njl)<{i$I+;nn}oIhXz)ku
zPu-l${Lwg1Ma7X<9@c~OHIb3g84bM=?c3G%&90;cObxAp&miXLfZ+WlL@4Q`UZ&p|
zL>Uk@0DI<8)_z`yHx}!ozj@qogNGH)%t;6P^OMa-nQ@yZdY~1t<)IXjAU`Fk%VB>9
zqb6Dfqg?9pqOeE%$Mm+<9x4Q!^Q~BMb1zsQf28lZ>c}UzE35{$9rvlrfOV@AI}!gE
z#MT4KS)mR$2}=Ju?4?ci7=9-l-{jco+Nish!>**v7>wiQ0Xyg#yj%aUbFNQs?80S+
z>jLcFGrz3#?`a3VwbVTQ2uuH95_Erzk26RO7o#BUzVp#MFBg=ii>giDZMKtx$LM!vhx+$;n9~M;`H@Q+@7v@lZ*+TYK??6CEr;!l8woZf*SD~d)y_v*W7=mWoE1#
zBV?bq!z420H&6I&*RWcEi|Uy~2$a>Ew-70lzOJ7uGK7@!XOb(WHKlxSHk$P*$?#0k
zAcMkXT>}JF@8h0;EaeRbhv`cAp5Lk@`?ev_P;Xo$tsZs1%~gdxNX;@0jPgE6p6}AovgH_o~bJzp@=Hd?{};Ubyob3SCT)i)^5Ag
z0J|}(MJG;SB0m^y`eEKVW{|Y*IW#7YFu%iJ8ZN5M!K2~#`BC9Ygx`phb7hc)$uS?E
z^+QTYnfyF1Zr{)w*lB}$cUXgNIfu4O6IRCL1Q`+P=Y~3;Ne7)q&K7@Y_{{vchNhvM
zvB{LDws?FU$%_LNR>$7v9{T}}ddy4|hoiQ)MB1aUZn_pL}E3PjBh(pM5iHYR2B=%h(V^HE!xljP-6^Jb|2P?oGG
zIJ%7pLn!&$D`PTA^38unWYNA!l!r0jbk616^40Nig$7*v;zhI*$a4Q|-yoev!Fe_2
z+v3a3T^3HF{E;(X-8f@FWU3uj;2-a67aJQpDh=Dr3qXJc*+|joH}^
zlgT_4qLzqCZ_a;Pf85dd4{6yQ9|R5kp8#@@WUQfeLei!{PCAubW3oU6?Ya^9A<8`4
zLyR)~2yCBciNqPo`~+s64;WZV{g;P2y$Ky@*AVk4H;V=d24w~gJqIQV5`6B!|j1lqU?LeUye8D?ZqqdRqyh274z#{cz%koYPa}JX}RMpxu4+$zesG
zey6Q$=`p2nGZp$%?fk5rnTyyIn$$yt$W6<$L|o}?Qb(Xa_86(RCoB2(@JXfge#e?1f*R(X
zSP|cH;|*assdXYMju+eQ*rZ1KDVT@FQ#ICFQo4U^0W}mZZ3QcPtY($SzxWeS6|US~
z%PyO4%;9?lHGbkEmccF~*#$o}a@SM3QIbzQenaWXa5RB35pxBRiD|t9UtF`kiGI~h
zyjg>*CG^XE^%rBGXK=od~D5pkdcajXUC6l_#6Qr*a!jy&hfdx&Wjre$l
z+A@C+Qfe^>#T*e;*;Uaxm~ZLI&u7{pQKztZ?#@O0hGr4%cv{eHeOyA~;@thld3PE2
zGf;{Z*{i~=N67vggK!ex@%eSR4Y&QrnJ88-9C
z6_Tb2j)7O8?jjSkiz7qB0B;tUPmNpCxfXwq=w+dJqE@^P;egO|GN0updK5*IULj(O
zkg-&q_Tw=>NwWQ~%}!UMnz(7c8T~htHeLyY&zkL#G2#>}vE@h5u3HN()fH!wUTXH;
z23@|6PAt#4d}JF=jO9%rYAJ|hAJ1iyXKK2!)&R@$|BsDxSP}(5kYL-kZQHhO+qQpg
z+wRx4ZQHhOd*?EHnt!N6WkhAo-I`8QG84x_VUJUi?oY^!!|&B~H%VgC1^yI{#1uyp
z_$VV4UCdAIX1~i@=Or%`KR`(m`GHurTDJst2{&8;Q36r!^w;!f~{2lS0NcnKNU9-qnAQr>>eW9jAIq(;Ws(nOo>Z{!4kH^)Q!446oo8
zxg!LsONX9GXmmn4ECF%m*_-|FQEl(jjh>Meg4yGV;N
zu}44FI(f{dKY8RWFBtmq_eK_)Ne(0!r&`&)48?O
z(6E$!?!x{-Iy>SbX$EJEY;0Y+T8a!GEue(g-K6IVscDH5TZqOWDSw?RV>8y9JKGuU
zi5(jtrYF2{ZW|*XUsYWza*coT$B!54G7FJFL?12dfh)^5^tJ@dzJMR2DFSSncgf&f
z8=O9b21AC?gGgG!sRXz8ijD@3nAg(DA#}#G4pu)JRF3lf#r>^3RXs?$-|nT7>grRsKyyv65|
zRcbpXl8oBm_HqwZ>D$^feF&?T2)6jRsSSJ}7HMpKCQzISow-plDPLEEuOf?p7
z6ovzA6b4Ve!`1`@2Om*H`=MB{3SDfMyj|5
zidt1Wo|cA4&fra1la
zLL*7L7j~{#Dv(3#!isBzI=mBDrFq>o^puOs$M^EkU>_y6r98AOzIgFU#zO!(V4$_)
z8UV66MG~OhuwUV}tc||yQ=;4Sbyh%8*ui?g?cm2jVToj8X81)fn;hdNtt7$bR*@Ri
z@Ci)3*IDtm%Ta%w4>!(G6UI?TBQA8YY6MatKN_a=qSN8?&HMP)<`7&STSj&+*h{2k933Pv5Gq4F#=r0Dro}A=FeC1yE8MSKbWSzy@GMKZMe?
zgJsHTl~8|3d$q?1JMQlCGY^FnRq0NJs$qJvX4O{Jx^6n*i<5*)_p3VQ#YF!kUxWG2Vo8F*Z7b(;)*Sl8^&UB%pguvKKsr`+gHGq>0`BQgcjvTImjmj6u#P~uK&6s3&8xbl`iGT
zywgV#Bl{6ufLdfb#EkT*t1ep(0_D&ZsRMrpun)*NNGu_Wt1j1*e1wfwf&5~rBFxQC
z6({-Z&{vTxKLxC$;2!JU3}Rp68Aa^1b=?XA%m)bO-oI2GGnb^Sk-Tz7`Z8j0WGep+}5+
zsu^z$LG`Hyhq6%AXfz7lD=Nt^-%}-Ei2k^1X^6c%8
zx)-hY=~m0?WC%+{6V
zrbjACwmQ*cY+Se1wT*w8nKB2(SR1|K2z*ktrxPr-uR*WKWWQQj6vldi7Po=MqBf^$
z4R^h|zZ5>DqsX_VpC2P20{r6h1IH776|s~@H;9@
zViSx(Eq(6LCG|!-?|fUix0lxWpam5mVCmPuOY~L|x&MsN)kA;Izf(NfS(#JZem<=W
zyH{W`GYPfB{e2DuvjxLP%Q!mVPd2l0@q-KwV|xBV-DN
z2MF#iD2P(qbzOgL)%j(S@hDo9uA_wZ=Q|A_wPTu
zyO#ZwG?yW9i+jKzf24;__u4>nEEDh5u>1$D&Upxakb97#ttejQ(|V_jKPl`-fwC;u
zL#Jz5!eIR~fCrd!ZhrxsjMHlkd$}Azm)Xci8@^z7FqD6J5WicJn>Q_*MzpuZQA3|0NV=?lrVqfuOnU3Se{zX2IKt
zq+Z@XC$E1W0{Fp9WnLMD$L^-c*Y@Mh{c0ZH2BDSbr#%rE7>?0+&^c86qB_^lJ397w
zzWhjP&2L+!>k#c@upW&c#t}SeaJ^trp{saC_L;@J6#RGg4K|<1pSFh=B>rFfw%G@O
zT8-;((t`b|4R?~?al%|OTeg-13ZFeoXgJgE$I5@ZL7HrI>7_R9Z-3O>Q5B^xtp(fi
zsPl;Il+K6GQ(KeTmK++B!SBTR7lW#o@Pzm=7FPHb1GOqdGG=wDR`;z4rU`IR1cXp#
zTuG1eJC>XL4c#5Xi4n8o?7oDwwy5uZ>S0sdZ0x9i9DJCn2S4y@q*V^THR7
zxnPha&UOYFA#wG6O8d*I<;kuBl&&R2zDjltzN>t=_+GAIV6r5Z_2BrH9qaI3Qw8Sd
zA}t^0#`DQh?1;(Hqi<~Qm_ChZCSd3RiKu_Enzc1LGVo2_M&LXF!$V9;$Kpf3uuz24
z2dGeY()?-LJ-4uz^wRh!1#Ps?81oq#+*m77b8m};p-bJ(aV$r!#A^ZO_qg8Dm*J9Y
zxj5n3mf@a~g5>Aq7i=YuXi?G<7#y7K2Ft$9rW?nY@j738E+q2m*R+Hiw-VR(Nwt5P
ziptSS_5^h>geMU2Pjq_6sr-Y&Oo6mNaR`%Nt*@Z9by8}59HB74u};)I{x)h~6mwve
z;8|(B1I*&k0ovP19$xki4aWEa&gAW_P7z2n@^{jW3P^y2jDM~BYF?HTRmNecj=DZj
zb{a{LuPB5p>nw*X`-LqHDBOQo)_Z?`%3J)i6?vZO492V34@no0S_x!_lC3Ek0xeYz
za0>%H^5+F=fY_~2A-2EnYaOgclW~{`m>{Jq`3*~p&Scl1-Wb8#-;g@9ut`7%2^heq
zG4{-XzhL{^6%XB8C05AEID663f>l0YTd&nn9!+&-u0{_(17`#!-5v6Mca{@QYfE6iEEMceRBG7hV}hEN)%hfU0N+VtC}bV6-T2zr*Cg;0!s
z;B4$P3X=kUt+{Xr1(nzXN!XhegQD+(XCkdGRN@;cy++pIb5Qq$2qRS6*K1R0)iQ!C
zG?Fm*@f?afckYKs>I)`?N2Y||A{{3|mimjqYaMM+zs>f;u
zp;9HdzmpoPM_CAdvr&A#ekS@O$Sy#M1)fI*oY~7OlY&dvL$4p!1iyb{f*5tYjjSUr
zWgxcP&Qfhep7jY0$sn5$2fE7pde|k#tov^27B%d*vai-HI)CN|Jxed8q6*Y7jC
zjpwS^Vox|1AuG8{xv>JSc^6F(+^V?-QTL}Iurh^pAj(a_*#PdzgQi~Yg!9~)X{F`6
zpolm@KFx}#kmEq9HSB+n=-_G4#V#(>jcnprz40V{oPxmWglvUPg)$I;LJuhMisbE?
zRgCRS>I2IF6U8m+k=|<4%w}Bw{HV9J6idjA;xT*m3~a!29V~NWWhuX$w(UIip|b68
zQPQRCr=V+2v1UuOD9Kg5QRgz_2(U+_$BoyV7xFn2Ea%>WXK`BkL?&8GsT
zP3|i(8KEqr>C9PtbW|@X_Rw
z@rcyeGMCA}tVVy*HuR4D;N&H!hl8xcC;<+BtI4afZAAT2S;(+?;pt$%HE)Q7o8=*UxALOle6lDvsAkl515|vHxa!Hz!p3&5*OZ9+
z8(TYEqvd}_$N)v12zcGV9}fjcP?WoOyJ*i()bvKPfSS)8hL*z#L{$fJy90iXCuVVF
zB9W5nuHOq(FVAs&sbfQQ`I9~_3Vh4WAukrA^GEq9W8Zo5SseQICrMv8lJc8#lDJ*l
z0No{m0Rcu_iGW3s!|-P_Jg$zPBRQ-F$}FuQy6%7S{M0;fKDpZ4RjBtD+HW({sACWdM&T>5UpCx{(3SQ%s(s6uQ!_pgmR=tmQpG
z{;7{=rdJA6t1ap-UwFg5PA^{Q5v;AV)Z2Xw00|aCj>Fna>}*~uU=%e9k00$J?O%#W
zl6;LBlpN6{`|CQU-n(DJEoT+u7k^0Mn7V)ef>Pcpp*^-hgW3BVSSAN%Xi528T`7*P
zIt0{L$J1?4Kqhm;PH@*p;AlP*+lF&J&%eDnT)n5`CfRt{D9&Y@4yB{N*BK-c!&9a>
zmzEnWPTM+f$#Jw9eTThHP(4+JHLk2;IY;7&wR;?)C_+Q
z{aJ8$*O{hq8yNGGJKulo@7w>*_Dx4^dhCtfY%k)1(8s;kit*zzg}o+aAm%oo2&r~?
z)`ki$3Bt8BT$1MTi{EXwR1nRc^m%ubX*#L?Np!hVBM4{XWD&hS_EY+ky&S)ySy5%a
zKRPxxTn7Zx2%Io|OHB+%JWzRpM45l-|249y)Q7tz$A0t9-gQ;md!F}9RdPCYS5<+b
z`t~(Makd8>7ojt)xz4x8e?F&yJ3o
zs$QP^9>pyf?t#z!6cyxc=pNdP*W$&L!DCXaFsRu#V_`I4W-Z^5+{kmwUgLjIE`J32
ziYU(&c6g&$x7Xdw;5!x|@H4O2icU8|5Vc4CUQqf=1agdpAMqP$iOXSFuYxmRVem1R
zO;M<0a~M9YiIt@j&RMM-Me~qNY|2_!0${CN=ywk5HUS`1onk
zH*3n|Zh6itb8uOZ9_-Rhzn$8VyfN!J1SOMz
zjBi$K@4CkKT`y+%luhFSpw+v8=`y7^XB66=8v5OI^kHTbcTW=6H$Bve42UKmJ&D7)
z5xe;@a+64+Odo$@R->Adk-dS}oP8L@9$@Ke_fa@hYyHt`<-Y9-Icp6E#UjK
z`IS`UTPPGE#Jj&^UOEP%d}3BU<#6G0S8R4h3UC(`jO-#cM=Mnu0sWkwJuNOx1%2UVEitesHqep4yZ>dZdx)(OAThdRoF4jC5L6^3`=Hgy4V(GH;!
zE>h-q1#um**sJ!dv$*F1;?om5D&O$9?jfO9bNE-r8k_qZnYOM371{r!!Y}BMb#+Li#qgYSk%8-bN#CkIr->J3ZAd0
z4BU|7nPPt?P$h|_Gw1pe`JD3=Z%Aia#~8;^N2P!J^{4o%pgFJJ#-E7MtdY|?GRzW;
zrV$^r93EIWrNDyuj$6G**W@$BiOh@5?#wb@L
z0?tyOVyWhN3(+^MF(GvU$?bw4-I4)BMs$%jud5PsxtH%U2dDfyxFOy3oouv?nr((D
zuTFnACf@-*3JT0gs{114%+8p-ay`84K;uRh0a5P=V-&;}Ciy->W&r87>Biz3e(_;z
zXc9M@^i5gGz#TZ|7BLE>x*Dl;!0*Q|l*a1oRC^J_Y(H!}qJD`~BYyz3!o9*!>C3j!
z*EQP>n{mgp^57qL+FA0$AcZ`hE&2;1S_OYE4hs@WnW&>?#=_3O&IoNdxeY;|`vqlP
zSaa>2xzu3!wSaKtcNsQqvsoZuL%bzN|eMF1m-ALJ1TY#d>Iw!zn>ByPZ0q%zK)nE#au
z2+gIG$w@bwDmoW7|NQ##z7f&VC_U+Zx~=z6H141cQma5O9<9IIun{4%fkS@=u1XDK
zh~`+0Ouay=^KqEfG8$}{?TvZ6TTYq#>W#gBt8?Hg!wPSYRjc-PX)L(nJmLd67R6{2d7Rp*nR}(7nxY!1OzM#37?6CegzR5$w18YV3|*2`k1>GqPMkz_g4CR1k&#e
zg~Bmu24r#ah7PT=k!^7}i&ym8ux&DNl;@Y8w)d4?)N;~&diSrxXx
zk%6b^TvlZ%8uIa3iA}UMfYLyzBqGBx(k%v8d4G#RJ=kc~2XkfGTci#)dVs7KDEIpD
zb<1LtD^~ljYsICLt}j*A4(JVJ+q;fsKC_?6!(nKxV6+{J4S|24&Z@18<76Kp03o7;
zr|qVts2+-xWRpMgi!jJ>+UPPxHYm+vGKo(Fz*bQp+x1do*-BFz02T1F>})1xTVq0}OVp{If68RD
z`5KqPj2>2uD!hMcR|qB^^q@oqj#vJeaF8}|sF|w1EC)c+eJMvh=ZiefzR0npDT?odq?%2{~nn+wf5d
zR55)zpadz!L4CGgvT~#Z-Nc994D36=j&9FiI;f7`cKyWKrvRfN6H2B}g|k7ndN&(|{NMU~2?9A)hpcHxP7l-KENt(ERq5k~^Pr
zH>T3}jRV!nPqj2V^tOe}{O^hfv@tTE5@h99oCQqfe<&VM_XUAh6RqxdDOR)rw&|bH
zI%%O{t{12Qa%_RA{pi5el`my=3`XpP&)R<`z?UF)JKsQ`U
zvXhRKEufnMX=hrKj38E0l3Z1duRwnY%4wV(6gfpcxwpKsG(B4G7JjF0%uhqafFwO6
z6^qmBPJiEnRB86M*y(&bKC-LZa%%&TK(;khM&+z9+AvGOlboPt1Tc1Z>Hme49@FK(
zXw*_jGfo#KPv0@;oNZ97Q4xOUWm_hF+WDljLPzqwQXyM#Dx*+NA`_T7-zI;i$=#~|
z4bOny5nXnmoZl+?b=y<;U&LkrL1Zk}5`u3tnpLnx)}Q1jigDswvZ|($0$g)6h)-xCODFfmUi#?t`cubAJt=y5qN`YOx<
z2Ra?$6fIK{6~dIc_!*tIJ`I1uT&&(Mu{jXh)v5c82=T>eYPLX5KiEM8MW82N+-lY+
z(KtFhn#bX4LEl^xU<^v@aa>xYhBNC`ptCiMN+;)k#~sOr4yotJX!#ZqcnU^E`jK!;
z|Ne)5a2I@e0>*!_7oH%70_98y9%1wscLlqL6nN7;1iN)|=Nwgrnyks4vjO-$f6ytQ
zM-3qWD&+4HVY|_WOzg<;lW`{M?lNrhv@Qhvg?&?WV8OC>I58%+ZF^$dwry(%6Wg{X
znAo0ZCbn(czB%{Ym;dQsz1Ch;)wQbk)9%{UU)LwJPXBy!tG6*)km%NTIah6lw1(ow
zBFh^xu?hNowbj!@a5wK$?h2lfPk+WOD(JxR&W=F2O
zrb3kTJHp>34E}`BA6&xFky7fCRiPI$nmvKBD>GuGUidl-*6&^~GR(0$D&LI9XOQtW
zHxo@0g~S=Lfav_}M+bX(R-?eI!)nWpS-y#8b+N@ZEx|G^Omw@1S#r$Rg+_*lSY)(q
z8iDO#lE)0Q%vE&e#4a3m-VNxQ&?M(XEfqD}f9(oI*|&m(@?L|uOmFkC*aUj8Ysl@j
z;M1IKF+6;F-m2$f!AB36SP%Ra4!5vT-uvXplfwTR0G{gvw7WfchS_u?QEDCo2N^b5
z$c(sfH@C}D=U_Np5Tk*A?jm?>N7>)VKLq#Uf_YX1sxxCSA0PvzDc@efv~Ajje`>f!$|t|p3Tyn}#`Z=LZB*HZnKwUqzUc0*sDq_QR3
z-=J}k%X?bUQVv;3x8A4ZI`WDk5<%Gc2#V7#1c<9YJk_}~)>A6CHOM;$I)lMBVh6eB
z2UKret^L_rjkD1_3kiWU2R5q7E#!s+ROdbU)Kz%=ek&Xnz0W>r;|1m9hk#v5G&|#Q
z{+a-bnEPa@DnECaW~rWMLzcO!iZq13))In;E?7%$Q^;UvE(Faz>w{0_
zal&>8TFrj`UE*d<45f&#W{(fFq_*eNSbZw4)D#vU>}h3SMNq3SH0hax(mlvt)P;ca
zzCbm?JqPu}nr_m-$ciyfVYPbuOl+DK9#~;W$NoS}kJxX5(molADQ-pkslaSA2~fYw
zyNXH%uY3_#ZEIipQex_Id+K}R#=@6siTqRS;@~<6SxLn?>2;eDqf`lBLWT$)b;t2k
zKzdusZ**>l1xxWet#+QHd~MAn+7VBCKPE2x=9;-nz;-9)YylqUfFLW=K-%AD2$-=Z
z^TPgEuES&gJ;GXvyn5GbIH`~E*6i81fQADO9rm!-^ide
zNZ-CT<4IVRNi^6k(01w>*$3yzK)%c!Jz%fN(Y>cT9xQ^tW--)YoVL;y1SyR3^Tv05
zd^)*vyz-}`pj#@g{;}kK_bjP3AHY*BVfoNQ4-LIbOSW&N;$v37$9;4^BsR)m23qb5
zA6am8oS_n+ad9lAWf2!#g8zdPcX)LtWSq>Y0B(l!qYZx6Hp88ChdUXsT*;maX}Y+V
z0zIXRt>n
zFl$h~2f^mKjcnvu@TqNHCDnIb)q}%HSK*-&dd3SwP|GgpRN}5IWIyIHnmIi_qT?8`
zm@lUEF=?V*AOe0wCKGFpd#;;%ZadXkC+p2`2Cr1m(Y&}FU1nZ$#MjLv;W^eCLz=0m(|Er)v;zhs{H{TtjvI6jYN{6L1mR(i|@B|h>wuhxKEmIG+
z3~&=y{JQwH-t!myTz>{pDd$y$OvjSjX5sEud;Y?US8B{y5T3pAOczw=LgcX#^=uZ+
zjg?_+<|r#{4}CFft{Sl>;kX<`i%VA?l9$?b5EY$sdwXXA3tJJbCXco#ARB*}{j%5-
zEldUqWrtJ-hFWSG&@xgI=F(dE4UAe4D;39CB!D~7D-TJxA<7*TBQxcB8qD1+FXr5?
z|E$vr2kCPge+hC%VEpszz_Ij0ZYL6G>fY<|E<^$~-7D9bB#{VoYPs&-=akW$!$(L|
zw)UqFa3AC-d^Ggicx8$s#rZx(quv}SCtGn5i7sa57gN9ojJNjr!=T{2R~x%04~WB=
zCA8c9^F`kcy`!-AgN3FHi~OX5rd!LDc|1ha$&?m^k;b#FS`(u)dqq_lMJ}qO_-Vpi
zL8b^94#S_e5>U1YemqghWXOl0=+gAN%8DiCRy8ofRYr~URQlg0gVQs`9SLkBbEPwZ4%iYOf^C7s|~
zyH&0;QMR7uyD8L(COh`ockF5=X2k)bp;)DCX@!#VK<`Qx0@ry_h1md&_<<-XbY++@
zR6rr(l}G{STMsC=AtHKJz!drBeXg;pncp9w_9fwKL+k_BZ3wdsPX;vyZOC<1DF@}V
zY542>(=e-WD
zk7a4lfeyk3f+dr?jzj+uY8*?ukiWm}m9MsmLC07t=+cQ+@Jh1YAN^v>ouWx%ljF#5
zlsT^^(qJTfV=SmK^GGQJ2yhY)T)^=m#^?P{1fR`ISs~$T}HSLY7@@os2nZK
z&P+=bnYD%dz4Woyb+5abVd-yw|5j0Z%NGmc{q^UbxocsVa=A7)+tNE)Mao!mNx@Y)
zo56)VVgfJ6fP7^xt75aWN>-`J@hYpd!;^mTNT8JeqP*y^!Av<^xf*qf8`>M;Nrd2!
z0oZJ0VxxY!vk=vO)U3h%Mlp88;3dUk>_DyhjE9k?E1QXwr8?zxUPCEw(<^9gK!VdQ
z08MdcKwM{2>%>y#t11lqlUP;bwd#*XjsYy
zuCOro^DtT)MNqO^#l&s+IC*@3)+QDn0kn;5ri6-iNoeMG7UFh@)ZBvnTH@8pHsjb2ZKD>il~?Mih}nvyNOwky`id)7j^<}l(H!%5F;ZHtLE+q%>$*6
z?^PP1!|+{xuGA%bmM{>1GS!NLZ^E#>@|>k=0u{Rthogrl7j$_|v*Psa?bBy5sECHX
z>vSn4AbhJ=htd!z=MXw7e-FOO0+5FJbxN*b>Cn+hYc=8}mZkd4hNl`nZL5jvp)+TT
zF3FdZr9DRoZ&6-ps9oGkCc#3bGx2@AEgOL!x>tAGOD2*Y@^F8H@M;m7E3_xk`yIgT
zFL0eseX(W#SksVx?r^}Mf0hk=T>;l13k*VQ8pC5(T|P8qFqcCy1Jcd9ff$7#WBZrf
zHYq*n7yZ?Y5Q(n6eYw>Y$9T~%Tq&g<6zRPh?(ME_;sb{|J
zq$tgt%+UB+P_gy&AGvxpUZYUHt4QpzaVjLD;e&6i3B;M{E6V1<1t5lXBY4;Q6Ucy3YR;8m9o5E$3gsDA=0r85Z~n?yQ|F4`Mdm02?_O6XB3ui;fb8Qz*3
zxRK-k3^#c6U<1K75V!wh;}&EWq_-f7=UQyK&)0K9ra~!rzTD6t29k{+QYp);v;>pd
z+n~Wgo&3rVL+OB7WuY<5BTIlETf=b`>)of^9_RXS{dLl8o!m7XNYebibQeo|Rf|Me
z6rDi$p;ec%*5u9!#967X~QmAgTWiVtg*E)VgqqUC=*)HP15q{CJ0j<5inR3c~^8|=54Eo0V(gq
zP97b01R1#|`Nh}*#Q%yOyz4hnSL0{IG%PLs^cQnslTSF6W`(qM@kn>zI(a&UvHvId
zBXvqmjF?$AcZj8@X8}fTBnFGJddIdFW2@3Jom9xf(A|2Go3hPFOI9(5UI1nbvliRm
z$Bn*k+iuGV-TG9AndpHfF(^*0S-2;=>U{>5)UN|Ei^v@vs7YFM0VCdp>ao(2b#w{=
zE6UN9}_(|=j*`GdQU
z&<9kc$eT#q*q7(`HJR^Jh9@36pAGjiQ>?}rMssnDODY71qn9Y!?~exL0LV?46nBq7
zhag^Jw;{7>fC!pg&I1o=3!dxT&k`>w>RI%7Z8cNKn!>`Fz!T%=PyAOtkbupse|DO+
z4+*@R_k>cNzXnU58>Ia*MtU%{K}JDPx_Z72oA!3>*Cs1v7#3>K6)yM}x*U}wOmYLO
z!ap|V5?u!d`H;I%@Z1I!FY*fVssy(`10Y7@eC}Uk$`6p^EQSc$iGZ1H=
z;~#rd>`yaRW-Fx0tN7MQxHm3Oyv+INAKOUic@a~XoStjyer;`BQ`zhV0`}li
z{Kjk-v|>&4bENf0?c6V;g9A7cf_AC5q+s+aX;ql4z|Yr*pJwvOKn2;K98JdZznWv-v53(f
zgeBPw5jz13gEh;J3`8%1{WRD5Lio;7X?L>@jk_Dj(Xz4>D}$z5o}*mru{ImfHzM$)
z1n8MnE-U*soo(Dp@prI}b5~AeV^Kz?0ArS@(_Z2VPV=I4a#vZcJYF(Yw-ZW#uT~{0
zT`*^mC4q~GbNN=kLXUxs2aRa1XsbBi2+uvKb(tVcxWdCmF{7{5`=Oep7+>QG_dk}c`0$|`8U0i4of
z7Cg6jdhbu2{>qeQ(rfAM^u>7^lAr}QR_F=Ts@R%}%tPxuNxY7KO8fa`%YdERm6v@Y
zSs(S9r|amV^OE~|43TwWCP}TDt?VD5Js_qO4vgCSNhxGE9Vyb>w;#f6<$m;fMYxtZI%{euGq6jMYup66f&<`kQ6;OACd@~YsLc&G?E6fcq8Ec
zuY9yQFS_hXSa$Oj{t3i%2Ks>dukBd3vVnew#q`Ctv9hl#stQclB?@TzJWZ#V0P3`(
zVa!LpIb~B=&HMpa(isUV@WhhFH+ukxLW1ZA)lmZFHf?KQZm3DIW=z&ZBF7<#I(pntWI
znN6`JQEqWsQhQV^Ya`gt-awQjK6I$SAK>(ldjWl%?J8OZ8y0=|zIfH&2W&&>^WNDc
z2qg&O^R=pY_s7?L#5U)cgSXPhgcRB4@eph;AC?XJ;!@51C=O3Dpuv@2^;I~io2PBJ
z2r}viKRx3)6!Yo>3S&KFjq-ag4bI8mB7Pa(7zN)^-N>9NCPIxz5&BU1Ybg3eJBtWk
zth=Fn-2WGX!BvxrnfgYl)}WCzx3u6s=vwpkkwtj;&w4O&aR_o72+d6s1i{#c
zk4atnPBTT*z>5Pb3ts-5A1>9Mnc|kTkeVS{{-$vJn=V-p(6zde0Y&E)IqtP4YtY@-
zwj^{TfYIj2>$AXhUK0yRwpPUXCI9>`S6a77MxbTE{(x;5Km#7sziLC9uF2=m)O$6>
zc3BHPV`wsmX`_zGjRf<-wqLuJ;dH&UFOXaS8V+JW=g%p(=cDuRGkWjs+l}nI{jrS^
z%SHci?nyU2uq&K-d2Q2S=aajSZ7iQ~?{>g5(_O&3hTb3jMQdAl9p{s=WV^49sIOXy
zeLvASn1gTAvnF!-R!YfG(1^*he=^!yXPdBmM2uY4r`f@G!~$1V9|_9uDyjY_{yGbX
zw4pfhQ}g`(#Z##U2K+rFeEocl_jA>IRhE9Bccow#=wMaeBqUP^g*ILb6WY8D*6{)s
zQ+@7KXtolKGi>D<(Iq<_51Tk-
zl05heR5uX4x4>$EW}UW9AnDzmQl`x7;A2ITlU%U6qhRAwsQ%-++M|5BBt+)PIXO%V
zW{hTBb;XX;+ex&ojX%CPtaDKBIv&#m-Xz?m(uQb4M>>=-QyTB6!N4o
zBzK%`QOhwquwez@ytBB402)-6A|n>~+}cXG>>a@J|d{u$_-f*u#6Xcds(n
zN4`P+UpOuNXH32pSq?3_;%)=oaxloGGVqvPt6mO{sKrak-dH-ewbG{WxguxLn5Iq6
zl{f`gypiqpp@F51ks}9$A@xRI(Al!(kS{NSDu+1wPtB|)|C-Zjb4p{bfYTiI*Py3I
z`@G#w*fPjtRh@|9cNom&Y33O;o->!%iArY1KIt445(k4tDfE1b9B0)2a3p!V2ev@<
zu=Gv-EkQ+gGudbS;{hqfcm)%4&!u13tfv6oWv-JkTA-$Ye2?Z{!Dlp=VgjU053)|Z
zBb>WgoHl+cYWs99HNQcCUp~~&cB3trIw0JX<=eXAfn&~27!u?TxZSZSHm|t9n|SVP
zS;7-Mvczkgr<-P_c*E*1D!iRv0n~3Pt}KaLB`OVrAW|m0^2g~%DPZjpTG9l)cq@*Y
zmCf}jhw#KU_OzFKE0|-wlY-?Xf|l}VsK&a%-k>s5QdFUyEh@b~?|Mq(ig$n0eENV?
z>S2v0;z=b-+8h~QWPL*qG<{Nai*t*sa=PScu)x1zQYTcDzwK83)l9gK0Tgk%w1heL
zXc{toc7IPEL=fl-HPoAfJQ&!M<0yfj>+8bkdIlN?8Bc5;N;>~RvQ>rIb*R>g?}qze-Ko
z@pu#Anl#s=!Wqq5liz7^T407+YQl?+4{oV$$}~RcFKt;Q_Y)gRFIkMXa@ZOS#x1^p
zmXUqX{86eMBQ*DqX!3YIUP4fg#Am0w{0z^K)vOl*#3?!M7Q)
zb%=`P(rx~Fkt|q~qXpR{uM>N&A6t>2ar`~yKmXc8+Ou7h+MSn4<
zx}Fn6uW^ZqcWh#${h&r_(j_*(HvKzSj=fYXJ-|aRKy59j_{}=;Lk6-zo?7D{6xD&K
z+2AY(h8MvFI$O=|UZ}NE+NFiQx!$MvLQ>%tQ3&%jBaj2XN8tB|E3nKSY>mp>B>x?)
z#&+NzQPK%1F8YcCv5I*wVFj3HC=;4!9^QP@NMh>sbl8PhrPHY4
zWBa$V$jBsc=a3czt%0J%oUVNFP5zlSifQ^L&{YZ;AZj353K%x48tT;!auGufylUtc
zHKLY&`ApAm|A;`>xE;z3CD3=u)*3HmmL%vA2ewafLfc+hMOgvszr4puFp9mPi4abO
z8PZ(o%{*}^BWfgP`f{NSu%n)8zKjR6J7
zXpPtPX47Dc$22^Ms>40&LOPBMw6x!q2Ilk=OJJeS1RM$^O$z$~1dcC?lk|8n*q4kb
z|0rYsH>`N%%Yg{^Vg4@(e4fSEc42xW;NcAS<>?-T>k)|#t!aQcWUyi(}|yjb0@Y%>D
z<&m4N%GTKvLuH@P$>7~Wc%k8_ci8T)RFYO+3~~?_n(qp!hxN68%{2W&L9+aQGsK33
z)|u{}v)%2Y0T;F;${p=KlLH4vH`o%V(R*V#J*!(G=7T833$6mj)1T6rW9k}Ta;klo
z`$OiukuowDH;5tDquUsOWq`f_GJ6c)5U`L|!Mq)Qch+y6Buo4`QE2{Lwy&w*xY4RE
zd3V6y6Vok~J}>4Y@DMfqMlXH^fn1LDRCI4+iG*?-O+S7r`AIu18K3iFh6*FOmKeyT
z$=vr(8`ZzGfkQqA*25+@x6#o2P?bOW&{#+ky^~HC^u#FH5M`JK&D!7%p!;cq`LNct
zRb^@tZL*mU8?;BpnlabjJD9(IsVwQmb0PW+E~OX6-?aH?wlH4wee72>-TVS#UvNdG
zu$0cTwebFg)BxQahA@?yqu3d#>MZt<)OvXk%-`RWla#BJ91CNfFFw&Ua6%H}{JjeZ
zJDTag%V4?9lmZkr?b-U-4G9zqM;FvvhNZUg)|YFEmf)a1rl40}D)Rq+S3oRpR0wB#
zWd62ESh9HctTA1@M;^oTr!KYdnl+AURdAQ^5XBwgYtPc8)9ZYafGtTaZpUSxeKH@k
zB?AHyb*5MLRnv@=!B9cD(u738(OLj_uzN8uF7~u-V^FlT0C+IOmULw>Q64ZZ&i|M&
zTJEC3pcKEnx4!0Ekk`SOEWo&#|1*W!q9zJ1W)8~z-#|4L;FcU<-0W#aPN3*54BOxW
z$Y9+68w0H+_!?Xr=F9Ob271fw3%CUg7z=BfE(AmvBnvY$3rE@!1O&nViExk*abRp5
zNt1@SINU6-jN(?#F0MpetW29Gom^b-bWdz_^Z%11@i(|6|S3-y}WF
z@mEMLETY(5vLr6dzip&Qk?VC$v^xBWz7!b-J@@F;5$n#Q&O4wOl9n7DnpH0+eI(>k
z=QK)E_!X8%3VYRTY-nunDJjaH>M1v#3n{6d94YRe%_-mSs*Y>sdF$(Q%@t2+^p=St
zNL1xfG*D^F0ISdvW>Ab&0*DxCn{tyMZg_5Z?5X-4hVI2Bi$c`iC@C!I+u>bKSi}*9
zUrA;t2M#Hj$`w}KTtp+{H47|!PEB+(m@X|Up3_5NJ&yZp3*or!MEH%J#+NMjTkt3HYEv}Q2IRn>UEA*WTD
zYDt#$0ch=<)Tvq`!D($2Xh%5GyVxMApZs;^!ZIgJf-7)$Tka5O_q}xDmRnoiWX(Y#
zF*ga`XYL~^NEdtI;?pb;FkY~G`*JzZ
z41@FDgJO5Q+)h^T2UrwOzkqBEG}nZ?y?gU;d##RDG7dO==2ygJ5#G|&v|CqHbo0FW
zv;TNi_tUD@x#R@mdl{IKQ_a{)a5G7#Z5}e`
zX0_RbVJ{eyT|?nhiZuGsfeEs!mDf2I~7bOTDFZBDzs&CG{nrG0S+L?8X$l&!RAi`y__4U}CZ4_Az04%rr`?Vg@YAlhKXA0O|(i$DIqd@*C7?jOi6XX!jUVu#{%52P0RHPjF$rG
z$Awb9e<#Z1ruie;5Jw!5VDV!6^_iC7bt)wVqKta?LA5jjpu%F%r%QeU+
zGg$J;W?4`QdY=f>OXAgjR&E;0u)WVpN_wJeYq}pb;^8gu>R6P>P!Tg
zUv=wdmnvLJFjP89Y9Q*~@7mmDG@dcr`qfH@Y5(A0deA$^^C+=|E_nr4sx8x08mdtl
zcPURfiihYB!yXl~GIJ&mpiy>^PV3B!yV6{tSk8;*YZtH~KLP1!N%81iRds2iv
z2XjW5_k-^1jg=9jOO8M$t&M^THr-X0Ci@nP3MU%S@i#to2V#FF)-YB2p&8PY8=<-h
z?r!wThO&SSNHOPVx2v4sW(hYhqE(8E}$Z`!rwck_0x2CQ2s3h
z?EcgWN_e$uG0VhI9pgHGH#2BRk;#O2$MFbSfmaToi6|{rU<*aze-4OVI~k;rG$~ke
zZj;M$+=xt5U(~MwI4p7Gy3~K%5mu)xu#SG-YFm{E=2{K{)#}=J>B16cL_91#UkRU`
zGSR(C$FE#fK~+oe9)xqe^#>u|d4EZFca+hi8R
zqaxMJag%4(_8=CF;f;%CWK*H+B|^@cQmb+XPFs~;w-jtKe+!kL#xemVW(>S2C6l9R
z{3t(SL^3PgjgaF%C?T-IUkGKRy`ft(jK$Y`=2S>I>T1Oqreq`d0SpPy8=&}|2pj`J
z$UIz#q{Lo;0vleZu9LT|v$ALZ8-8I599<{HIzAhn4mJB?=y9BsEZHsz7F0wII&K{l
zg(>AE^XYA_ls$tizfpo4BLrT{V9l5|=~La4K(N
z(K#;E_tCOQ0~_|Q#)rX=zNLY5#0>u`(>6&4;b>hBLTjEX3CAiC(%spQJOMcyHqWm>29#Y}Ztc
z%lvsA7&dl!e9J!4DV1V`_$I$)d((a49D{6quV@qdkg;T8`H8rO_r@*`bAsIWOcw?h$UFSj<
zLhj?IulLioB~083L>v?!Xdael=wHk$h}j^hxYb@#V@5tCn0$$Q|Myn@UwO!?lh9R|
z5R0QP>_QKzf<78OH3H^G@)Wzn}i0)!B1ML
zgut9Ug)EBx*P;)t&ko4CG>^ci@zc21nhoTEu>R-gUBcS~XK|ZEN&QYlY6$I7l%r+#z;9Uc5)GBtwdZpBn_}}wJ`VbL_!eqO
zcrC|LC3Dp~DnGYbMX{an^Jai7Uu(>SR5$dlFK{J~p@p}6_0OeXIE23%hj0?wty3QE
z*`7CiLfzW#lFz6k$3mp{(PBa4_u5MVV~ca4?MFL7Y!qJcxrf3G
zrRH;8mXf|Gl-KNLmlBBM@8;3Ra&rm4*x9|*i32|I2Y!_t4TL!n7&Y##^%V?x7H?PoL63{)q(^o7@le7At4r
zufx4b7&{!I!eu7r;5f%yvey4Nnb)(dV~J}&z0h9uJL1t)`V!zE%adrk7`*6cJ|wYg
zOdRqap?cgv+d8#er`7;?JYQb!>%YaePmXilKpF`$*u<%;nvA&PN}`+r8a
z9(~1PivVM{+|XD4H+rXe?wle|X{e3<6ZjgXo>g@LMktvrwxJTLE%d_X@
z@M)RdmObsYWsec4j9n9d-t*<4s&Y#!k6KdpY9+8LS2up<+;RM8pt
zSYQkYW=EG&?jD#r-@B|#--hZwJ>8f(^R)97nMrK+B)RDBM41>QAF|l1S_K+c&+sgI
zq{NzCyQz$e|9#9(a=w}H;(5kmxpQ`rX~Hh^u10UO--TsROU*vR4AEn_rlFk0(Bkf*
zgQm4`paIOSUbRZbXnv_nz1Br`yR9n|+@xZH|9(ZU#_zEYF@(1HdwfnIoiUN0f&jON
zlF8-4a9`E=3yG#?_Bmyz)PMM8`uaD2wdJ*4V}HqTGPN)q3APZM#G^(facJ2N2RcQfi@IDx^uZ}Tr%OD~GvOV2E{
zcK&`SzI!b0?p$BJNi4LS#m$4aa3R>yuI;LAhx2=Vfo2rBkQ$g<4Gtvn*=0N-z4BU>
z9;^eX^6=xZ>Ek!wAp;ikm5_`D{d&18nOF5zzoZ0;d3-2)#^`>Pt8joH@8y`+#e;;2L>pYilqRl%Up)_L*wU-mThRZO$JvYE8(RZOH|A+l}hYji*QG#a0@e%{2HI@a~bRH@$$8
z7x6=#eVd=`v?uON`k*;=EaNr4>XW68c3GP&-MPVWU;VT>0lw!WWBX8Gu~kFsBN+-1
z@Y`9QlqgeuC?zbST4-W9h9VtysN9=UtKjVUfW1v7tW@33XORcYUf83?Dh~_>Y_&nX
zXzEm@ntv`h!y<0n-Nd~Q4_~Z4$Hz7`;-G$9N6YtvKL<>d?>BfVlA~dQize__Xv0(C
zK<@|mE0z=nBOWq}V-|T?{2^b?R6Pbl3Z9!yEv@p&gNp)F;d8%fvq0d4?+f|{fYs5?
zJO3e{A7@F3)y6nW2I+7XBWG4-D7QqhvmJy6EmFIdJ+*czl05m7xMv;+jiN^)7=GmJ
z1V%QW@tkobX_T1|#-jHEz^d|^G;9B!;!U(2ri(_PuI3=L9^qnnoB^9MuNm&H1ET{12S1YZ82W!sI!jtxcM&Fjqe7jLX9CUIa=pfkE!q^
z?hmsl(IK>_|jPaPZK55;p1Wx;}U0K;u2-&7GY;+1g$Y;4VRFu#LuU!&7L{!;+g@b1F
zVG(hOiVboFx}d4z2@Zg!jtc8YH`UeF^86m{+h8=@=d8UVwX`u2I@BHkX5SXP);tV0
zI($sb9~dn#R*j$%2oFJfu2hW;)JHx5;K1sDyCO%#AXwkdh15f7?kTt^Ylt=7KBR=7Ynp
z%^QW2^_x0gco$o3-vV(JLAn!0e_5}%bTfebW{4bWh$U`_7TcSy^Pc7&-48Kmh=SNh
z%mv?RU(vxHx$@_jVuJgai+f<;`0LD>fmtsMuxeRmKi{H7RU*qFGsxw7C}v!IOjIJN
zFQ&Rn9(9DQ?qg36F4bwsr`!z4s)fF_hfC6Kg>2bIr0dmdg?iiut@@L=DjnARJ(ae>%!yRu=?E4z$YH8{!F?jBw{IzM53MEah;@q{q
z(C085enHbPh*3qfz&Q%7mGYupXf_+(7nf`k`CGaict?odI+Ed?ULA~WFOY;alF{?p
zk|KvcSSpGbGXsR0XxR`XgFxeNj%lB%3ZqF#sKm#k$)H3ga-{36W5RNjQzmm3CYwrv
zWJ5VA$tKMj7m6lr>&<*BDBHq8Cd&|muyf<$KR(?gvg;jKckX8U!mcu()~&8`zVAPV
zcaHtLCk=^1Yobh5U{()wlfL1A?g2zY)i*2mKV1FUs=x?c)l0s7-!fvJ+H?2k-2R^V
zKDqU%k&i82cglB%kKJ2h?Vd^B5|U8Ou44;s`ARm8f&=!g%?fg^-m@ywp0Y;QQ`Klh
zu1)XrUre)(J27$uJZ$3P!Yt(`>)u$`9o7@eWNBmZb5*lNNJ0@J8KZk$2tw+RX=FHv
zUS^EhePCKCW{ZVS8AR}p0`Ia4{Bb$5!Ir-{iG{9oBDG^ac>%omRFO3D&Y=3u*C#hS
zY6Y{V0zG(Bn>b9?+}5VOR}<0%>F$%l{!+)5Kn_2HXZ-$fe{i3PD{m6{+~SdVnIHe$
z*0XH(XmX~}LQi}-rDL%vNB_66=oy|8y7g8u6!tZV+m4f#(#o8~s$VUps2Jda=CUg;*Y)KA?^tLyPia0IRL=}MWf^5wfO
z=fG4aXzLnd6mZS~M_Z*jNRs?pVF>K8*$j#(jfAu|FHd{S5jj;$3~SQ0tDc`}()Wbd
zqcbQaVSqDQhWtQMO{{(wBh~(i%@t-2mmdiv6AEOG(<@LgPJcqvr?6Wr*k3
zB~}eF%TJwviTH}+)QrdPFW|@R^+|Ms-cqV~{Aivp`f!;5mdd6>8`EA1$D`l#0Gg=Q
zQ(X_8Fki=ixSBB8T_7(3xAQ1ax+tD5c-jZavBLE`W2AIusT|{cpy!Ewc;5q#9Nu@l
z4H3LzwU-^p%>GKUCFtbd@dzt03^23=g~Ly|p6p6;7@<;=zbv#ZQK!3MN25@&dN-V)
z^+QLT3Sj6%`w$$JTx(DCQ~(o`y0F2=@!$W1D0?;x7AW7Q
zs$wkCCP|{BC6!Cj#i7svSZRJWRcRXm5r&5=N&LF-S{Sg!%BxJ(w#U$9KyeMnMD^tz
z5#o12Wc#j}@t(K`@i}iT>fvH4N-F*Abtk{f{B?($I&^a8eKK-T5P|;QMgSqT-%KPm
zT;kzdQnF>lR7ydVa>ioqdx)DB?DhlHhfK8LfTEkbk}Jl&NiV4=EWIA$3U}9-3=)Sw
z@0U|1L*=>u6gYJv>9@oZ;3hunFPw{lZT-iO3PAfyyrj1R8DC>kl4s4@8CzJ&F(a1m
zdt>jzAH^JfMDt*E-|oeUCr2^O!o@H9IG`o6=I1T4rTL^U-bR$xl}WyaEzP>xP$}$=5PH&8@$ED
z3q5hCM{6!EGhHO?QXm-KFiES2Sr}JXL0DeT9C-b5EfDqh-MN&gSApe)*550wC=aCd
zmkzQw0bOUIRWX>chf;0r!F?Bx?t#x&=x~1_#=THhIU0O4ucEMU`r7kZ)jzPyHMaXB?@&v$2^nNU
zr?;uUfpqz}R${3A-cL_;59g@pa)#)aEJAF}cn!*5f6ph}%Q_Bw;41MPOy`=j|`
z+Tl==zXF;cQaf)OIl;L0rXvWWPbv7WI7hC6D~oiO
zgmA{^dW7>A!1B33dhxuXQ+a=6ak&gD}ZoD_O20lJmMp-j^
z3s*}b78W+{|DgrYqa{8-cAvB37k`M!)_5ibk~vzBw|ZccaPEs;%Wr
z&P{n*cMWerCBwsGZ;$J06oGTgJm+A3!o~B;>0v9{^Y-KIYNUFlGR8U4LJDJx|Ch3s
zktvGVdIZ_>g5J+hk9&?Z^Y^|k!ZEixeHm7*kYY4}mk9&F&ciCp{%nzdSJG~FOWl6m
z&Y0u6s_5eM-jU9#zD{pP!(m_ba!auVL1_cZn`ix>V$TMJyUN{4)hX1nqgG};OrR!ajTSk36^Dq-m@mU&!!WoLPugC)L=?Kg-rNweR
zK{fxw(o?t+)aQ-xa$2BRhM=d_`zv&KmrUsD^NI-AtU^2_FMqFN}MNx+&1dDnW
zV|^o;EJU9)>c(R-;qvwh4UP&#VR1Y82#d)-Un~Kvi=lT;QN(WY`N}m6)#k3-GTk!K
z6ekZ*K3GhO-%1RcZ{dY~@LOXo?KDOwTu#ONw7*%0Tom1)2kYqDY^rsHuv?G2?_red
zu%DDzH*8L@cwRpLp$eO0omc>4_@bCEW
zgZ%Q-7O%9DDKpl!W{>ZwpHR^oLRcwx_Hh_ko=@#UErNVQD*&1J!8(UWR3kIO97xNf
z!F~WHhe3Ia#!hw9mt5dE`CaI;oO)bm5IW7@pMf34+n(3i=!T
z>C2zG3(4g{FJt?0VAAb!=%C@-7g8!wHiXUmO4S+B&9G~eA?O!2bxP{+>0Cc{JJulZ
zqk=xUBAqOnGZZ(cjd<~ieg^tY6bCj46~E^VCLI-WfdAEsGv8xbKv~$lUJrRAKM)sB
z@A%1Z?F@S@s@qmy85)mPYZrC7fLnxWq+3w_eCY2hQjmUgH;h1{0=Ripz2$$WNBiNDqxRG!m9!I8)^-t6Vge5bagToDxOBi5Gb21Gpk++bN!!tY+4B)p+x$FkLIsc9
zg4c+@3vg(H+0{wlZz?1UdDArkCg(G9o-I54$)!#-sfcc}#-w)N8oKhl>2VaU+MKpC
zs41}>&;d;K<$z?KY>8&=ffX~|66})+XUO+MNp^%hv)ZvN?h9P=O-MpElZ@;DarZxZ
zI7PB&WH}wWV91h64r@lryh^yXt8S;KbB}}a
zk;AHxIEspg!bkdTp^yHDOH$KPA7*t52jYRZ?llF!Mpr)U&P
z>{f=YS{Hy2=(
zx&D$h55?1t9Sw>Yj1g(8u0mXe&og2YH)rF2@pX6KQ|Ri`m*8ntF^8{dD~sTHwIHd92~yW2aCmBgBv}7tZw|N
zFFGqiP?5$-`0WtdDisns=s3EVH5;Q5rjbb^4D@UvmMKzG^?$F;M$D=kL=z^k!QJF*ZJMNrtHh5ccPk)_5HAM-exA*TNfJ=(E3ShD))9aR
zW#>sXIb$6utFdf3*e3m5Y-m=rFRzI-T#u2(^&Vr_F)JA7(-SxCB}n9q2@bR?o*=tD
zK#BA*iXb#1qkVZ-T&1zBzpWAy#>A(KU^`KLD;%N=@(e+%L&Ez}{ibYDxinE}tdNk2X}8hG$L=AL*j&j;?QN5=x
z3T0G#5%>kYz&Y>hv3R6CT0Mcw!W7uH>_Uq00WLW0#Ux4BJyWoXA}bzbhm7Vj+8WWX
z*22su-h)EyahG4oGBaD=0yNb)Gmy~cIoOlOzrK78l1QZ9|repGtu1hQ&C(E*S4F|wC^ZRki|7nWEQ;UWLv08Jo!{9Mc{d9sR5vPCb
z^OLjH5ev1_tQsbvw3vgAJ*%VX2eJN7I)PGTp~F44UJ!{=5hOT&GUzpFutM2guv~NO
zoa9n3d*iI(9~K2L^=AF~F$w0RulgLl1iyT4Z!i(7xRh<6l
ztSWLA9gy3DBKCo1&3CuGwo`!16N`!R$eJ`O#6iVMy$49@A$9#*@@jRy1>^I@@=qU9
z<2^}kGl%8hoFWk=eT+^-$)X+RQcCaXtrv0U2%J(X9uMDkf5dqW=P}gumOZ|HcbrZ>
z8U0pOen1Zh%JttjS>DlTF>nPJvRka#JKhi4Y&Wu^7)sS50gp-kX(S9t6AAIq&7}U5
zuL5gMqcz0E;>&(d-RZv-C--yNyvS-XoG`hxh}v(ki@MalIrv&vXgPdZ0=uHCByj9~
z_`|8D*R0^biT|v&Lwb6Q6@+Iyj}5A!w2`_9HV$Tl50d)5)Iag1is{z@XzQfvk`~{W
zob6SHqQbARtZ{_zS3&{$9BvKasrHRk4uLIZ3u>J2iwci&ng$-rh@q=*k$0eLx~&QX
z43n%1S@x^2DqbZM;M{wL0~rrt#ofHI&W)NnzmNZi;&Bp
z!?WagwWv_EBlPh|o?ZR3)atsRgf18kSyifQrRT9JBwij06(9g6lAIh?(BRYIR{s`o##%QA43=A=ip5+HEL=?sg4R8Ga-MpH;R9NN
ziEa?!iDjO~jeYO6!RNd+?1ZzCH42l%t8VUll9$%4N`UN&DG9-Lt7x3vqX|DLCVK^X
z!_BvJrBUd)>KYi08Lry``)az^Ucz6j#ga$J#LP)CLn)^Ri0)T;9~lQCRl%$k#4oOF
z$8qMK=Wll>5?8ZBe@KIZF%T3xI=I!|b2-15;MUKDieu(uoH?!vM0t58*5n1@sR&?G
z)V&s9r2z0a<-`)aQLxZUi2~4*_WmS0h4-S;YD5*Gp%x(6M#0y8!&-3NCnbAlJk(6e
z$pC{F=Pw(Vx;*wE$q~IMM*{Nz8MubTv$;LRu6`*C>BPk;n(&Hnff6U(_H^a$zHHCG;-KK>5^zJC}?UfDg=fC5Q%_cr$QvpAvm1m0s(YCZ8Hf`#E1zxNh452NAL#ccvTi)7!_r$V;9$0*
zF;fp`y~ZfC;=Z{to*W_K-6F~k$&exwkQecXhxs1GZPH*-=^K{4*P4!*BDet+T*pY43&hH4B0L!2Cx+p)NZP*I5+jo9hT=rf2_5tGnb+@ESHzuIQ3`tC((CJ-QE
zSShF_zv)ZAyHM_Y6rY?qD>@WR^-7Dj*~9s%{Xl&?asPJk21|W|Lb;mCs7;sgLmed@
z-nPOo!_veO(A0F|`TZ=xadx9~dhz_Ii5ei!P_-9H(p0YD!T=@}ft8An|AIJ-
z)}N@LCyudR9yA|9QLylqeMj#~?@3Tpvxvl<-}KKRIx8KHDT2up#aB1jF-VMP0zz)h
zd6yjtLXdXBY}E0bu5*Ng+y~$k+d;Q(3=&Uk@DHQR3ESORfj1^fE&h|sXS-eo7Y>|b
zeoDM@q2!p_+AQEsKv3u|iO0Y&E#3}vl%?nFZG(cduM^)uo<-roq#Gm&LFe7S6t(ci
zitxetHT=*BSypBHGhnUo+f3w}wrV2olaW<*@KF=y89mWL7ljMb&eiv!Je45Lc$-=+
zNK}5xcpS6ncSB&z);u<;00?GK-Kjng>chCu7rS7J9+fZx@JjF=}c6!K6NK_5`mWI)(J4>_0~2~})1pP1iwm`WwN3@nj;cU4n7~?4kDz$~^+Ck92S1QkSA@_fZtP(@c
z$TDRU3x;i!z^03F34S+vD+7<qyixcR%PboLA_h_DbZ
zI)6bmeh6fyhZ!u}oahj1#XXUfjY-Z)y-ns{l)?YH0FXV`egl!<+OTVz==oWFX_EE;
zS8b%%CuJ@>m4KfZA$7E8VUs4cG7czb&lk~8wX$>GXu{9%^~99KAJreL2Kg&z>q*9K
zcwnn}Jga*{F9;vKB81V5zH`iWOWahRqnf9fh1^c9Nm*zd*2?WgL0K~DO-s^A
zHFfMXz>3gaKK*XC0@U)VVKS&_>P1v9^j2M=M^!})jMgts2x2{PY*+QzsGA{$owv-}
zDKu<$KlI8F93CmS9wv(WC2js;xN<*Bq0qe6cdUVvgM0*WCxMiEySgVHwyLW6&RWFM
z>|qDK&TS89@)1{*zS`A&1iW=)TU+;Zfy~-&QKVqckn>
zQ+epK!fF;^BN~zbN8#oBrMXX8kG_$Hq~>yVL-7yoAUK35tS*Cix8~39ell8wvP`-h
zJ(ZcYo=wd5?Df!es-PQ2C<$TnDPS@|D+xR&q}(!@a%WR`JZ$L>rX>^GXW$DR9r_uH
zox(INBSC8}WcoEQt}G073?n_8lDzbu=b3-{Ch}2gv|rLpi(mQP)QiVMj!B8g;?
zy1aNZIVR7SiFf5$-&XNCe}CX2K2C`#)hZmcbV^8WbNcm0|GC?7~l6p8zJTP1P6Vh
z;q;vmE47$7=k7pkWZ^(@OuT$qJWJQ%N*3YgJVC9RujSKoW0wu&3VXqxLH5_%N;ZJ}
zZRO=3I8D)M`o*hJciIfq_KJZkhqKdy57`-&?I*bO-r_S<7e5@je_LCYPU(MZn?CIh
z+UoU0m8|0rY_-YuxI9n)Gc}1#4zb^uc-kx6zK@HqoDf*x8N=l1gU=I>=2+j@cFV7|
zBUk~CJ4Ic;o8gT;q^Yb~E?Hy?T1C|{TbP0L@jr|6!7%=bCLTP?z0JoO&Q@8rIIU_r
zMZ)a)cCXc;J@QEJsxwM|OF#YYrS1(=>JXPZ&CV|fbm5eH@Lh%1+FucNdln=};4I-k
z^xUr3-(ETu1je=1ZX5O{ZA`cUlAqs;Y-qIboaa>M)T{B!QRW#R8ak6gy!U^bgw
z;kkN_n{K|!qNnCpxoy3=!bggXBCOfK&pc?=S_w|(P9j~ZpDaL4L!@vx@721mUu)ME
z@42u9UcGj_W(f}yEXMNKe(2AM_vLl~t3hCwXH#*N5A#BWzf8|KLA9~tXQ4vfZ>
z{rv))`-CG5oxK#ayA5p;?nX|psKFT_E?|S5VF97v<`a8$I!NL6SRdN^p`uUMoubE(
z*Yn-=1Q?0^Su_pW4+C+w@G*^)6WQ>+1FYP+AFwEeaCcJE$vh|9!nK2-;0~{I3q8=9~!ro`cy@VG1+ViN_X#T`&
zNGXq9u$$!!W`O84mhao=B7ApM&541+EZH|jc|VtRYX2G3k0=DB7hBkM2l;}ik*Q1T
zLcoa(#Q638=VbTMdar|h8;Jz-GUo^Ighj-Zuo+Tpt##$k%fk#)S1ATjPBHO2VjCRJ
zrf$t)(*mO>5Ph$r9L6bl75t+!95hxkW`ejt;4l7J?e;<_~r+g9Rtf;e$qs4LZRw#e_$JrdR>p&uCUzJ-K!~DQu7!hNHW2ikRBQ;S%l(s6t7!
zZ^Dkp6oM-o>0mP<#`?XLON!m?2SWz`G6`tth)JU3U!8z#XnL#HWYGK*g
z_bD*^T+5eMx)-{S5~Z*W5^W>lG)*;e+QDXD_qYXA778Vfc#7zUL9KDX?k+L&8P4KR
zGW+qOhDe0<{#92i!&%v=fh%4PlWv*JEHU&x4+D#Yw$ZlA-M_6;Vs#jYLk%}h=Fjn{
z723EFOcH8=!fT_`nQT7xam=ld^l>CZ;}oW$Nd3(8fdTDb5+zOfiDjuO>wyxJDa?
z&QJkY!9JJ9IIe{cscU_8qOmv{_kANp@Y@w%`c}K7{#e;(|E&p73*4q1`K4Z>uf$qr
z->axn3b~HfeLN$BE|W$<&C;|M_6*lM3Uk6%d#P_Uk|u>{o3Hf5!cGQGgYCK5!7lyf
zBmYl2-ds-PXb{c}N2$jHT)|mck3`KW+nnDKC5SxS`%)xOQ7$PYDCvh(x3rXQ=EHeu
zP9O$L$!^|fDJ>}wU~ZDviX-wvFf&L2(yyw~$s`KM+ryQlZ5kmPnu}1Iz_glv{pZ}s
z)tSg1N#DfX0@ty;^x;wf2S8^{5bR`Pwt4^P6@D$6of+w3q7#Gl5OkW;f5QXx-!2uUWMvb)n}I@Eqex%z3ft6_
zWWIC{#S%21Fyg0X8rPY#gd^mqMDcrWtQx;
zuigQVomhxpguGW~`yyjZ66hs_%Xcf4srvQSmB8P=5F$mMBzbnh2WK|yO!i7?w8oIN
z-mcoM+r7T7+QS7qjq5|v04A^N5Sud*PbXvFbsA!y36uot!XveE0R+E9#m#fF^1YRw
z^9~u^Pz>i%5ga8KLKsvJw{H;AfGrDZV@yDcQDyI`!uJz8E^c676!%F!OS*aQD(U(>
zC+d%HnTJ%|*-Z)C&uy1Yl{%Ga6x8U?6qRHU?F
zm-MeiiOMLHp`I|Ormy|?3*%HKJj;4z?jLxx9$Wwk+S?2V$9Q)za*=b%VuUMNUoa#9a$vR#^QUOoOdJRoGH+
z=SnTOyP+(eI3#f^L&+e)p>nF}2y^B;&L**Oqho(uk%LF$
z-!X;CS_^gj_7ID~mJFFM#hzk=*(F`hr_7Gzc{*S%D&h|errvh@dWRf@9fA+p>oQzlE
z3=BsT{cQ<1ZE|;SbB?4;73F}6gUmA0LTMp8l<@cKTtMgz!}0MWSsgCk`DD*sQ%B;P`zu*|i(~-3?;u_-El&b=jzXr?`1e#gi8>`nwCRjJM@ex(EJFf3G{1
z4EB`LL+f9SbKC!upBnp?Ia%I4n^Lh62{f%fRdAZs{D3Njc+bK9u>!|>G$x!jo2#z*
zqzo$K@F_R-eESqpseHJ@oblYC&mupK<(iONjv1j;_vwd?Y-I~mjoi>EwsvRny%1Kh
z^*!*p`DE-;MX^kZ4`>cYGl~^i!$8b*ie9a?3FzXm+@(nj@fYP-UR$%hM#3AV`JvJ=
zT$0+#UjVL>WiH2Sh;ulhdE)Kc@3dqH9rBidEe;)+Un2`Y1okni&Bt~vxk?d^PqGWo
zr1ta(XKf9n;?K}}wgNV9b9M1f!oe4=8FzL0V;9=~qS^kcOmEF{G8@^2-}s7@mkcy-
zG*>`#*W>TB_YEGe9*)jRV!)jA!7}`MQhe46#>nIv{R(Ry_2cWHUK!V81k!_d8G%iB_
zbHHeyE`F33(*1=2<%d7myTg(0KfsL!^La6@_Bnl2=P;g3E3D#uO7aKk1*dEaFte6Ws!Xs{vfBYUKIZHm8DtXl;Vba8a
zHjeICN=YDYxw$|xXNGdsW>A9UN~EG=y`7skSU(wMXaFhn>A2oNj_7pACQaOV&X7zj
zi~=%;$JaOzd=^Kw(p)$bU3eh
znHXJnh)Ewj7-?sW==JGkTsNy1+aiypmMt(i`3=s=qe`z}f1QOO+&U4!>>Z(p${MU_m%c9rt0Ugjr=HTK+5h7CPo3x!I;
zI2nZ%wVF-5Sj|`ytFLSqJ
z80XLcG~*5Hy}888IVAeviGV#OqMSA|zFnt*2L7MfLVjbT!ZsDubb%jEsczt=kn)qTBGT%VVK!W%o
z&DkI0B_k@6kBeCGXte+pD5HSOi3XCa%>5H2XAER~6z3WQ4JQ<`n8~=sUZDZBt`Btbio4@>R
z;-=+9#}h`LZ5S%|m?+JE&Rg4qGjM99h~Ie>b(;*$Y{x$B!0VKLCh5dakzLoh2ERkr
z3tF7^--~L={xUpfTHO*rIgnzNELe8V0q=Dx3OU0Do$YGv0HqLg+HO=DnXT#_IB}e7V78AM@p3QUnbLgY&Z&q5nOTRg`xX+8T<{3
zRlAx^uBN@?e4fv!>)>pd^UZtbvTlkz({71D8Ude
zKa+;2>A!RtPP#R{b^D6Nm&;ypA$?-o$m|?Yl!u|^>HEfl!h6P)EK`*+ntog49UAW%
ztsMzwVR5+4k6cV6$m3eb%x64WhQ~aO=Pb#KWv*AkYoOOU5Kmfo0Ap2+Qks3e>_dtX
zMqlfS$qtOS3!@t-Is9dbnpVHo@Yc_4rBbn8X@1LYo#xy`VrkW+0wE-rX3Dn@;xiQW
zC!)AkvQS#6PdSc5u{2J3478I?U<~(XOVC>Bn7G)?v$j$b9wUU|>c7K5?~5{>f$k7_
z&jAxx>+GH}O}s4sgZ52;EA7eVZNrjhWvCcN9ilDbsuHn
zTFjlG4vOc2lT|!Yf2;Ah3Ym_NAB1w1NSnRRH*eH0d*IobG`=k4PaO*>)uAYiTQY(O
zp~t(cDZ9(a?65foO*!9jD^B9ZYjsFJxNd$Ijaj1)BE=NL>ys}n8H+i0&j$|h-k{+u
zmxteEhRryu7u>}{gLU6cEf*vp&Y)vcl!Q6bJh2`F-V)DplK(b22V??;o1oURzE3tQc!^}40Nd-lZ@>Itfb9#v%@LP(A)*6@J)N^9uuHs~%?r)^=5
zj+M!c@Ea;$omo|9eT<-1-GB5AS{%#hj2#iV`T3s!{NXW#ezx?RqX(7LUjBbB!(2-f
zvZ|K=1r6aPdW3vvU*?`d3sqJGC8}6Kqt3_a%XGcA=7dg!_en;dt0P9;^S3HL_$9`S
zkk|04@Oy9Q2Bsct>Oa(wU;_(tCklGHy*e;O_{6+mK%HOW<<}W|tl_
zOG7az+&`XdXW~`;4`eL%6qt^W&$oRKYeZmy^sZN&$3PSJ@XuFT@Q7
z&+-f3_K+)#(qr8!-2uH!5euKF4#aR#W{GgP3=uRRldP9Ojq6A>rAbp*L+&g`?sE3-
zEwChp*y=+4GEF1jv@fXXufqi~u~%L^xL7bzq@&be!Yu-vqktipcN@^
zIJq5l!=<&UX`}CpN@-G4i!$>P%CS9Nvy40mWrC2pY`j$~(a%pM8R|Lqe3UQI7v
zkCzq>8JTwoeAs_Nv{UbVVc~4cB9=GXA(GHiEM%lZ(ATWNo3bQw+|>;z*rv;Yfjo
zbXWNgb1cnR$jJg>0Q~{huA+aI5aOntCHBI2CnqH=2)G+T;_1WM;x+X!@z>_Fx~P?=
z-bWfKCMhWPUW2b41Rm*p0dRw%@@XG_;@IG6==o(a2ZqlfPR&{H>ZH-7yQAF9_wpAC
zjGdvg5@fMm-u?1p_D(e}pEtbAB4-K1iy>C)3IBvuU6
z;fy8YM|3;3i~bx-7Y`})p+NVd$$fOsii+wf0$_cp&Cp>KOz+j-n4O|#u>4OF8p43ElG+fH1ZfZYlg(t
z8djLCZOwEYA2K=&1PY6YC!t5!zE}Id?$eQ_1Hzh!`J%NkkyTB0HX*>!h~WFqK?~w3
z`*dN!ft~ux+M>(v>2pSqxq-YZswO5d5{*1BI9CfEI56&{7J~&bTMcf7FaPujTT@1X
zh_u9>J}&D)+JuG1v>_UK?N3}XoeI)iJz#U>%4XUx=$oWSUae8m^~B`b>%!J5;Cgvc
zRcGA?4`ZTdg|#Y%-pB^Wkg6zBTE#}&8wM$7I)=3#zpntb#^rX~(#oLPB~n9s;;L@Q
zIgzOYPTyb6Vj>`tRI{l3YT}t%sM4;hZlTUTkicrl777cGhgkug9_X%zHA
zC85@*C{PR{H!Nv0s*o1S33b(O0{XKfA){9GsfzS1aoA+eJu#@5(5~K64F6e+X`}sWkfI359jf<{rq;HF+5|e{LjPBK$L800FL|G
z)A_42nI3&Nbg7aQsVR=m(4dmEyi6v*!nkXAgPYp0?)&%(aa~k8TXw@^gS#>V`I~J%
zL~9914$$A`B~Eln$KXE5bgUIz8YVwS0VIhpC^>hWrxH^MQPvrxH6w$+V$D(n5$9_}*x7lwCACio1
zS<}rAYBu7(5`m~%98^2});}9QP|oewY!KBu3~DsTz@uee!P0p6FngHkC!!D8l{xm-
z7=&Z}I1_^%!qLf{G|R8<>RY0l#pcStL`B;nU90!CV9C!$Rej34P&=cL>$iW^Te6v^
zvS;qk5?Ea2GM4a18a?%)>HaR8!9i3gc+bf11+z|e0(^ht#Jb!l
z$D@5W0!OTRN7|rPs4tz&BTIbRhN!~{p3mdFbP1t)d2K`sK&mHKKL
zuFXwkW_X4Oy3N2*>fz|wA(vy=8{~Yko{6cBYyYb3EGb@t@0-2?m*ieJwW1)Y6G08u
zdEHz$)a5s3nK7a$tVfS5naBc>WN_|q2_BYb2v`K-gtg<%F_!I=WOS46F2XcWtLQ0Z
zGI{6iM>%A(gtg!yV4P6aAx?sOb%A$Vy&-$vmk(bnw`KJ4?7UaV3I4~vzZGA
z(JtCMkcFWYVUAa#|F3u-gWusHhR;Qyu)*KU$mxUz%@`UWU&4MasJjck-Y{8@b9>!_ET*v8Yy41dF?
zK!e8iSDT*!5&*DEDu#51TC_tSFdU^mL8b
zQlwZtcLDHDNEgi0oz5>!j#KFRQCXWTdr1V3+rE6eX7s5}vB})~}NCNx}2pw?uwUu$cZ$}7t!nx*rjim^#Nd~zV&4muCPn#1amzA#2eEsXp47|)?9aYsE!?~osi*v___;{^$E1326Kd*
zN~AU7cE4uIAW*$-io-i>r7@aLHYH%(HV{-z*fpCtFVzRH!G72BJhadd>SBh+BK)8yMXOBq{DDpc66gVuv`dn5{`YW8lH&(wA^ag;Zho&9aWd2g
zP40p2FBTcpFx@-F9mBqVfR<SMMX^g*6P44<*Iz$nfBeVURcf0
z7240l`t4lrB+ZQ#4N99ZraDp0E~fkBKMM~nq@XiaQtp#cl10;66hTn=oRvN^Fi*az
z(A?E1N@QATCQXdo;yB)Ag-Ig!N_%7Is5DBe{iC!?oe;P5aHxtN(|sZ;H+&>b8xY
z4m-A!j_ogYI<{?f%rCaBj&0kvZQFKkpZ|_~AI>>7My(n(c0JTf)tYPVIoCX8%XP$V
zD0EqKg1+hPaBG2C`vj6UcxvT*Lc8qo|37*Qy;lxofX*yZv871hE-
z!wVY$KY${vej07XFdSXnEBV>l167uuVsf~*zX5sZ%+pd8Idq&8<`4XEwHb*YXcknv
zD&a^GFJ|WaF>3*{+gpQxq+ZV!TD@`E2a6}ide-M?H-i-8+zzuB%GUoZEyZ_czb&40
zE>zXXzJPMM@@3Bt0lCak)IUm@Z
z$g&fnN}%&SumccSNB~5Fm~ia6HCQG!%GbXzmD2YWDxkFUa)T~AvaRFvZ|
zk8ifs4D>`Q;>Yi7D>h|=GtZ`-_Gh8j>K2=OL3w6G8=Ry4g_?ybK7G2tB5bnu>Be?b
z=WEaXt!z(48S?*-wMPu
zibVGSmO(7M&qQ$QZ5VA%iq47OT4|`S8=hEc6PiJ=*tjs|!HEH@Eo&n&x9cb+`0fc~
zSXeJ583L&^*-uI>aU30hWZR|)DlIg}8%OgO(|7NSUt^%@%ECvwRU^%Qc8}O2JZd(e
zfMa)vb4VU|TRIQXNtBUP?};+fMhOe%1m>Y~dHIs@!eu?mq8+X1M4L{OwU>0=0EUHW8eFtr7Ri_$^dn*+q0;_kb2p~5;WVxnHS-NCP{VYjD4Y!Z)
zP)54m-l03Bxz{ebzgMRMrvmtjV^r(K+yG|hY_=vS2>-7YJ`+0&=YMvA2Gyiv*8e5Q
ze;QT}9Q9v$CC8-_3AS-ybm2<^KV*@#!5_MFpp
zVv-FJ@h9zSql?cOljY0Lb&+Q+^m8O@)}(2N4u207;!+&S?#$=Ky2D9yKbu~CT7Qmm
zO8QBWV2^f$N8CjMkv=kjQ*Dcy)YMJ;4Q7h-W*pe_S)ndsa+5T4qRsX!*FD=sOd(0=
z^*$$vHjPEKQ_&NQ0S!E4N(S!drS-?m$5T0XIcG_I`sSK(U2WS@34h!9bQKI4vsEp~
zBh@I-klK0=_j#FiLv097dD2YrGHt^2U>VDrck8-mqnenec$hSS#W%Kb{c1yv6(hYF
zQ!|^-*XWz1Ni$A!G~i~BMQNM(Oz}xSHA(}sjz{_N%p9vUDNOf|dXN)6yOP5|UwTw3
zbIDuf-Lp|ZugDvxg|k!~sOL*!nWy68wOR+
z>RyGfk-S4ACWx-<%@M^GxsE?W=v$)hxbn$V4MTj&Av|NEeE>pBeMi=2eCu-7%zv(<
zT!zCi_mhRuygmXm_gP(2jy+n!FP3#D$*Qc}e};BU?DHA{%eRo!9dkgFB&c1|`!pf>
z@Gf0R`9DO&bz$e9oeC@*hgEBGBJh3@`$F@brVXg{W}*IpBKTMbL8qB2zhefmHlL4F
zioj%0!a9&LXw5X`yF?fc^#5d1(Ab~UK~?Fpm#5zs8U^`XVy@{}3F@HxRXR;8zOZ2c
zdDo}QiNU@CtQFvc<}8F5wv({`?(=zYaZfvHGE+uuQI3s{BG6nvlb?z>-gIX&1LMoj
z?)?qR0j>?@b@KXTixYPU+W`5}Io4I!r?*|46DjwFjw=WTqcWs22=hol+8ClNAW=w^
z#1#SR?*%)MF_^-yB`ASfPD4&?_c%kuSmaK!b*g~|EF6uK03#swpWhobly%#Jymx5S
zSj6_F=kC`d)Fcs`2@d+!KeEMVjfJDhnCw{*KytubljQbfsexz+aJ;`7o8p57=_qD9
z_32;8_J}IV#85l6;LR(NNHCF*H8oRML4*b1nkPkT;h*iucJ;d1L2okb-As*$v07KP
zmhQ3wp-BC8Ck<1?b8o14c6~9Yj(wM(<&}=}!AZ>_V-_AphDDOqT0v@Z7IcNHe2WIuOXcvmDLt~~F~AQbdpoQDn+eQHIF%Bb0NgQl
zvU%v57J}m1+I?#mVW+Aqkk1P>c^la1*R(~ur7t?@8cPqL=n2-d=JSJ0;>i!riUoX0
z*9-ZQu4lEv$cb$a?ZE6f2ch{sI_^2}a<46`Pt)BC%n$!qBPz&`OnxK7Gav_ccG6E}
z7^s?PlP$XI1h6o?!B+GV9V)%({hR@YU9QJ!O{?SW@$eSo(#(72Ynvfmeb0spfg?h|
zO)GK?00d}e(x6PqCLgmEBLo#pnaO-PqO2%9Ilpm7sQOW)Co!9Ep$5iLgEHY#QD7<^Ne`9gKe>Id92w20BV*Np8HGvT0+xuyJBdYYH
zZsNW=NUf(^l-c&Ef%dM2=5-kK&J^)RbkoDo%*Y*UI3vf%s5AM&s9mSec@KqKKN39f
zR4-AWK;N=L@dl_d(0UM8+P3rRa(~8PL+R%H-&hKG|3)h?F|z*8`i+@|_5TVG8~@Q1
z{8|&xVRQ6jt-La
zVulVw`~G%0)~S2H74F-Xiy`Fq!8oOt-B+Zp_2j)&q~0f4ZrA=<8*BOOw(-gF*5z{a
z;+!2Z;^ax4H=9|x{(XI`cmJCZ2pD%#8#1B)(=CmWqS53j;Y{l+nc*clXIM``mu$5s
z0>ZZn3XmFh&vj{CsGZVp2zvZJ{r$=K`!BZ5ecdt%_XS}cYs6457z@%rj$-}gg+2~e
zuq^?;CyDE)L{!9T1j$<)CU>0(Hp7N}QDZ|{t)4@&jd>1W_L+Od8wdX!Xjx%_cMaBI
z%ZcG$F`mfDuEK^`BxG9<8JS;w{wc|vU+E_#Bak1mBrw^V1%0l)>Dtl8GnZG!GzCie
zm?C1DgzYYdk#_D3;$cT)e0|sDPf$ZGD0uVpY#lvupwP#8n=Hg(X=eJ1-$WTyIxljl
zO;}Ry^B~j*?5P5~xR<#Xh>(nvQ2*t@J(BleS*U7d-;8<2dr2to`4hf2(kJ^YE+Yzi
zD!IwPA&DrYs;!zvrHs=+mW{|Vp5>Nz4fKI8y?E#gO>twbmB#mHWoXK^jS6&UO1(2f
zp)>v)y#%Z*uXG-Ps7NqhjzJ`^S3*%~dmE3^wYiw^Is2cy*{rSuAp4Bl(>pt2*llA*
znf6dA+0`oKIl!I;?fVo*|enjV2*
zW$j0IetwwjCoWgfwN3~ugI7znXVQ>GFBoOQ*(s)um9e0S6TA9g0#|k#RRBR?-y)V)gqd
zx6{mLe7k-Otag9ueXNhIBZN;%d}s4TSqz~a=1nX4
zAc2)AppFAg(Z2&HSZ|!G!$(*X4iP1Uu%^uGQeKsJuY;())je@8gc<$SD!^wU<8vmL=fzW0paxRWL!#wxF;9p
zYxnl;n`A^|kUumArPF;v!XO6J=onWM>}GvfB5?`$P|QYQjFsRr^AlW8+d{}N_LDh!
zErzk?f1PhR*0sJ;b}7bpj)HUtUWAyoMKRn@GBiI#fp#<}rjTYFjQ39PuL70$3efAR
zv68RSed?jxj349*qlNr><~!=(1BNoJhzJF&K|*SRBpu~UzN6aq+Ih->HU9R&o2l`P
z`W23(@5sMuU5I}gsj}6Fuid44@mnWi#bYpP81wrYHz|JLWgky4kQ#rzlZwaNnSho0
z=+@B#fY*)M6vmPd+A)}xjYS!xB0^kDy4yjsgF9f&39CEaR1i{cV@FDLFp-B&)@_{Z
zZoOwC96ch*CfddodM+*s=(DASC7^RQr7AHWj~U({75D$%!WdgRr>Y6ceEruhL#n<9
zsLj5vd46+^<_Q8#wsj}U-p@a~RkHdcxFGCUz#8{HFinT()hUMt
z0>-%u*_pIHBZ~>i%!u|_bgJ9&2ab2RMer4y9
z@ZuV(zO_rZ7$w^}3TTJOkm>fK4JuA(T~uA)5sJLwM)LS%obtn-XOAXA8lNCAH&
zdZDpL#8>JkXc3xQ(<2HDj<=OH(!8b>H)dbjZjb8y{Lp=-hxE0TIjG$$M5=a$l8-dezQjP?pCAG?M%bTc
z+TTKU1JQtr;xILqmp0_|tl~gamx+0;hHln{E=FI_RhA>qV7WHQ2}c8PfEP*14}N>l
zS3?|T99OrX!CfEhRM{;0YsxJbxO5J;px#*Jgr;M@vx_2fNU6cixLa+
zM}63-A(2<@%efRALcS@bH*m_XOZHB;*b{fKKN5}F^R==)nK&AV#xmkLE3Vmx%ekni
z1KIn)y`hl`=)Ezz_FCGrgC>>}3!w_emLq&$n`3$*nEVhN8~5HQhmB*C$GqxGj*!>?
z^xTqD@ojB^8~Pz@L3#F(0m-^TqIV5e_nAm*fZ||kPWsFGWU;QB{wCiT+8D%2;|c1!
zudpKZk?sKdGj8O8Pwv)R{1ObJ!@SsYx~#1z_+2OJ+~naN`{M_Yn&7sNx+jp*wWPeV
zdRBw|_UD2OOK(pD-bp#Enw<3W(N?Tg?wU-!GDKP57Vf-Zl_JNROegEYDVX@
zY+UHPKv3uZAf{Vyy0y~3bKDd4hAD;aSm#cz+gCnFK8;Bh>7Re2BlNk`Va?oU
zT$2Sq^Ywsr9foHN@2s#g&Eh3oD-P|oO`boMZJoR+DWT7wTjCsG>PQ&pSyL_l=pmT-IOP#{L1(H<*otF?u$DA@$7w(A9n+h(6fE0!OqfUfu
ziq6(&WqUWy4^zUDF+LXAQxnS85K9ss>Dj;f`UjOZmj{>#(Gv<$zNh&?QR#Ds=xX7N
z;m@;7cU)Qts|H$b2J#+b_o+XB@RkyV%CwO-3v62hO1xlzY{J;h&rC>T`=)+25$-O^438-%wDksht&(Rno;
z##D6ojH9W9j2YmgG{50G@MnfhdZ0L%9|ARj&5hu4g4IH?rDfRK$B^B=F=4qPc)u^y
z*pb$wzyziHD{)NQvO&
zjHi~ONb5qAweE{WATqge_H)RgOJn0ns~bLvLame!2;-(`Y9xDS#&3Vf!LZahvEuIo
zSrKkrLWr8$n^(r1Qz^*n2SFKW0^nv(VIW=@v6q=+C-Y+{yuLXVFBBO3OIbI6qVDe
z!v?n?1OV=wj68N<32}R&+hV{FFUStibD0aOwW>)^Z!`7=(RWr)kqxQ}gpVSL80~h;
z2YW*h+Mq}j2_FAI$K-}^jOLIlNZGUjVGlvMNTK$Z5l5H?j9ghrkS9Z0Sll#xw2s3L($jYu4Dak?(W
z5Xl-_Zl{v8f2nA=JQ(vws%&pG_ZI}b&@?G&n@OyO)}U4;=TI9V;oWkjbO4iM3~m8o2LZW)yX`Tn3i%o!~zn`Sh5pbIo@c727fOYF`I0G
zTG;ufXD=^Ldy@9;YU7U^;YJ1EKyT%6TM2u&u|{v|xeCaFZ^0RPgHK^(daYnz-`Miq
zvaP@j`L|UW=UC4^YhCXGo7_=C=FPH4c_6%+Dz)c?VRnkBK2B+k{9tg1wZr+x#S*p_
zRiqVs&@^<%X^<;Ja+tLpBJAm&EOhBTcQCzF%VYc0@c?bLk>@gP0)izlNZszW-89+v
z7h*l@<|g&eWX7#!{;fcN!U9ogrmGkieMtia^bQo6Bax-_;@s+bOW$v2
zj1UDH5A7PVTYu3!yr1WK+BWVHcz5c^q}
zS%7@5^q;=Ge7lO6!L>hey?@d=;{?&%#evTv(_4MTH~gv$7c)BthV-%{FAvFHE&jqS
zJW14LWzxhXo_LK$a%zFmZVnS|M(}>}#;lsCD#f;T%h|Imn!E5TAN9lZ(g(-!&M%&h
z%9;X4+B4Foe4dAy+>$(-TgwK$zo&4NeRsepQ#@dcS2NHY*@=!}k%G$SzwC(2ak-#|)Riysf*`b*;udwRS2GpDTv_aL^6p#+*>sEQN3~943_$xASBK>cPtQCg}rk
zy)JR3Zp?Ck6jSrYo~A~g_T>o<{;4PBkS_%-F#BZy#l6fEnX*KuAj(oku3bk+YA^xm
zoPbuWM&9g>;;5F9KGtzphjfAVu0w_LN6?dI^dW;eFfy}22VyY$rtv-mVfD+(hokw=
z$@!>mk#bsESEdGQoCWnpj_z;lOG|a3M~m1^78`4zEA_(c$>8MG{UpQ3LUXPrvb*MK
z8}M5={!uK~PeCtQDFX5kAg%;g*wr37l?!*2=~r4h<7a2DI=kRXJ9q4q@qI$z|{ui!5>yYH%;rZ*KPsq!jazE
z$#eP@=(|A~;5*txGoja_-C!?Fu&j!jteOB@<3{FxMunzt0TK!dq}YdWAPNj;r%x%!
z2%qqhn~s^WOTo}lelG6;h&lmpR|C$vp(K=E+jlWq$5-@x%MoPKtgt?Y=&Y3;^YnIn
z9!b{*;5@ThU~2AAC`#2^Lh`X|0CftQ{30-M4Z@*0EIBDZ;
z)~h=T@K!fR81)WH?wjh<b-o{PcJBt;;x*3@N@kOf0l&F|NL0;+mvw}
zF48l1x9l<%lpK)#g=(|3`9tFGzp2$G!8^;6J^c(oj;U$a`{((IYvUE+*3mc<&=ynI
zhwKISy@K2;OfxDUyeR89XnZR76=MgB+I%(Z0mww$+VAJQEKIqFPFaU1{sjo&JI?kH
zA7VFY;zJwJG!x`K-87c`ptF@N${p)CKxbL7&Ww{}X+(GI0|9=SV1|A_XQX@!732B5!{F
z;*1Z#@{vehs06*p(a5hn5y*&?RYpY&w@n9T=SR!#5()F%y18C&c8W{JqM;lZkdCwY^!}$
z@474&&ag|B$J>95(8U5FMb<%Hois)ry&eGGWdh!rO#40=jlfKfNw;3-Vq`$3jzzd>
zlR@7Lmq>HLAxlrzBbzgYjuG~=O`AZDdFZHb*~M(z)1`ERl#9R|>1Y*gL)qv3pL7EC
zB)Y+f+9eIAlMDUSg73XyZyyWE{&>WXzA`bG8yM!L4?6Fnb7GwHUicFUs*B*WZz7A{
z_Xs(&NEeFeYEmrzLi<+qvFGgl06lJ?xU*cDCQOwwHq{}bDTnnIh@8^ae1g0Q4}tt>Gu5lL2)L>i~S&PPaG7QTZ
zIfdPV(EX>h0PLGKE>F+KUUavdRgrUjCsdVl(nS7&qb2W(&r48+B$6gSkV8}gHIEjX
zp6JBCr;dVVkW|b%RDM@kkfh;MUccC_=0wj@sDTVoFo)*Ipg($7;b*JwX1D|?R4L3)
zP(@J!;9M$Zf#}@~7CDrKL?pA4G}R(n@Ei_tgEI3?jna*|gpX)4?u9ZT^fvAg(@HH(
zfO|M{q=^vc2$SL{%o3cGm!O11KJhaZQ!2J1;twv8;Zp8f8El=Anb<|BXvWe>cmGO%
zegg3-BTDu15q9xZs|z}RxIL2tMN$|8C~Vp-AXpQYyUA2ejJtkh-d?02bGTD3m}wl+
z3X}4Lmi%m@n1bc4pn-IRCmfe~5F!lJPsZ`8Q4D8&zuuxPumD5{mOlkYXMPYwaEyex
z*9VZ9c-YRlq>Vwjj!@#FC``hgqDt7h9XW$sbXyf$)#kV9^l%gtJerx_+In;2DKGBo
zz>e=@7s-fbHNxcnz2e155BK^nnMsHKz0JG(v+>QxJC>s9m7DRq=@_+YRlQS^IZi5-
zgmKW#>lvgnJcES8kXxV6OIKHV$V&Punv)-d3V$uitiXwT#vwkt$aROSPYST>3
zoU=h|p!&NZ+Pr*Cyt6^0P>r|zY|d(y;`(TF61Ml1T
zHM3Q#|9wWXe}}%2zGR#k(`T;H)C|8jIQCwqs@CY>^3C6d%x)PI(71q|<-z%M`O60h
znEvBci9ZUEWc&4$JO5(Dn{Q>
z5X4J6X*cAiSkh6r3U`*vdB!~r^ZykCq4pbH`xH|<&W`hLLjPxRLS7@%FDf&aU@mJK
z49!}4+Kj-^{jV6**(!S%m~FJFi?A~h;stRYaXyFRl20D-!`68b1dK``QaMve1HW}R
zD9g77%zE0a=YjLY6`#Lr43wNy6CX+Yb^828?8TCtOU5cU<2ll_p?EEo89le-Od_p4
zrK%?E&{~L!Nb(;-3?7|KuEA-8Pt)LEE`&{ceJShp7rNrnLfe%9&xRFsUfF+|?G>!&
zYi>t%$XvF^Rh5=lR#yWc3ry5p*r30fsY|^EEe0?3M#!LZwaR&4k!kOCBVParAx1x1
z&=3HO{B`rkKqsIYjpQw26R4J$idePK%wi@fqG!#}XD`bFPMc9bP=&^y!1-
zaruDwfFnHNv3Awqu)u0<(MH+YPgwPylA3n{MBN-Q%;Av+-RVy`^a%(*GsrvoWR2N`s=;
z=UcCvfU&Xs=LVI9o%#PI5N0A~b`D0al-);gbVz2-e;ulXTq;UnL#v5=rp^+Z3=A?^
zizj&R<|bQ9yYT)@OFI-;FL_Hl+Qufn|G)4zX5)T>jOFp`Gu+?QM(g~de!8`)jDkE~
zE$=5Q6gJqP0DE*pq!WTRhy}K2Y@;6{DOx4+$QV|S$Z$^&n6wm)H4>{mEA(n87AIIM
z$VId%2p8#avk1@=1PvUVl9pNk&m6n~=!gT3unEE}AP#QAp+!XM+qXkVg!IS0fQ{AO
z%@N#2X41wt(T=*0NQ=YMQHf=En=RGC0;u-(mrevamX)y%qLouU9UQ?r5;We4*$G4=
zc06MiPXM$I6boz;p+W|P2+fyS9zQ3KtgHkASuIg1P7Rn>;Q6cKz}O1D?i)bYCpAX1
z#|nZXs9~Z2BA*0BRFzlo@m|e5b0phMjVPe{I{BS(On2kj7Fkl4Qy1JpMf!Qo27(1T
zCEy8tzuoYiA9W#%y%W>Wtc>Ov`v`ylfnyOD*P{3Ez{JGlR-%L!|0mR-{#&^ScuSm|
zzz3lx@&rovtQYgmz`e*BWm-pWubrAheW>|uYk|eGbAYXJf&eMQ{Jstwe5?+&<}x;K
z0exFIyD#lUi`{*oKLmDwv#Hpe9|^zc7#I+1*cqE0Al9%}usR{RaX7iTKpwtpJ3qmq
zM!(6CVUS(wn!hKB@9k1Pf#zRyIxr7S%r?J}7=Yj@-yU$r^>mNo5D2bo66lDc6zdqUB*c+I0ny*%7xZ1eC&_3IyCVVjCYbETLDnFVUEQ=Lw`ETTx%YqP01w1H=8>ZdEi#T%&|-
z9UzBGU7=bnQ?nzO55{Zv3J2N+VFvXKN)HIg8mMoTnCU}z=~r6fw&~FJQ8R{{`-;Cc
zC>8w4HV6ijV+57UUiCYlw%tm-X9inB+KoeQVo;7xn`?XvnARvuqm~0gx$R
z+IHf{`Xg{%?sl5E(2FN`hzR!=hcD9%1Ta|_}s;j=nzgA~=Y)&9;WGa2_cr}YlI^Wo+AbfujM{emL
z_0JDZAQ?ed`@gm%*i}I~ckJf7k#T)>Pmm0uXW+jD@$9(2h<<}~o&b?~zKs>?#q1Xc7uSkZVGtyszK)Z`M;0$j21mVLF{X5~n?GWWRbSD$lm!KVj)W3y*
zJHm%1%_P7{l-wgA^ge2O2gzff=tqQdS6z!<&J3un_$Y*8x9{)@>e8coIVtw!bZlVXWaA>jQOzNT7
zHFe)_C!4)udTIUp=ifwKGSEfZ?S0b>zEBW6V0tN8zMG+L1$OF4{rEH2(g#&^Yy9fgad+CR=%~cT<_Uu6H9_`3|`0mP`4#
z9v%ZUQ}k!=KLMwK>%f}$e7+IX5>wlcr>a}E&`7`h&en2_{`I8c7gON->obu1TU6lf
zxasScmITbfnc(;x`&okwUM7gxQZtj8v9am%TeWsRknioQ)pq+T2)MsJjSqr+0>|pF
z^8(Af_oG3g0-pX3V=|2@nC!5Chh}g`DJ3#kT_Vf8cMuggx285;`x-XON@)lHkD;nupYBdG
z(JwLgEN)P8NRlV+X=i%EgkLP{6UF=13_0$+l{E+8uS&{Ff`bBowk>{M9ADp2Kpqkd
zbgqbJpk7ua2DCLm^nY(z=VX4
zw2sQzXgG~nSZhO^((2Q34NCrq$pD7~IpBbS3x|EVJ6)KGGTnODPk0h?t(6_&1u1GX
zf$yXic;$G<57txM#nTHI9aVnuP%2Z44!h6ELdfLjM#Bby-PCLG=D%rWS;ZGPa>+{Y
zXod-lA|d7+VWPFnV#9H%bKc=tz(Ybuxp&UhzXq=NToz70}3X$$RRZ!(#QHj4x015Z@*}cZ;ZT`x+Jq
zEjTahxf~bdnFWvCIlz-WT{6f(c@khvKUi^Z!C^jb^CTw*8|Gf+*A~tc=ZxaU`v3rW
zi2A9$`TOyT8={F?{N-?4r60t_!)?IvvDSq?d#`c#sgbu>^^DHDP-;h9IH
z?lpU~>SlEYk8W>v>gtWDSkKuRYoCvOc6)j^XxCe;>COB~kVnelOHu0w#jK-KMAM~?
zLfFnP;&c3Qy>%FO&Q3sYh5kwB+~=z{mNjb)x{R;+g=t8Z(ZYA*D!Y3DNGzZ1m
z{&PsPz(brRuU+OtDb_r7fYwCO6TXXq9=XYOWzzg0@)iWz`h7>F1P0w7`@55dv!Fn?
zX$kHm!r&B6i2O>;`vhL1-2gj8v5Mfd*ZxK>mEC}_oB55#DsOEFn`+=u0`CxMl!&M%
zGP|U8HjwW)_h!cHx!zatk#?Go`4_%(F3$};J>(SPdC!*BFBZee{%F}>@hD*m`(h!v
zM9}4Eb9cc;=ra}ENF@A6zb-+!h~Z9D$YQ%sGH`x(M3~W=6b&;lecY
z5P9&8$nR%UJ*DdZ*b
z4qi2@y0B!2Hd&pYMo->2WwUTnJGJ~2`P@sZlrEhEg-ZL`9FpoUU=)4RthQB)ul&WV
zfP>nj@G~Ex^a3Cu?mxFJzLf^gxIB`Wg43OpVJ70{=-{cwzs$T1W8m$}0cqE477~Gw
z3{7_5WCh^|p;IAq;9zcl(74XUGnQjv)LFb0c0}6e@0Z!)p_?T4-`MBgc@DG_m_nsB`s()8&CZJq)FhV)K43sY!xMln4^@*JMNTBry5XX|
z)K=CCx8qs5OHA=fqagY9!(+~dhF85yd><3M*&koH
zHHm!O;X@#fR8XWwvA{MKEb1HBv<1U$%bCL$c{+56V50W>GTM21sSazi0{uRzn(|!bu#^Y`(x9M&M=d2Mm1V#Rka#=;Q^glB+5*-
z(PG%-Ja%=Qvrba|>ENCNI=#rKZS8K2rF_a_#3_&_(wPpDL$gDH$y#4TQ_Ic)
z?>bOV*GQjkd^khTBwQ7JA&*nBr(jZm;U&pRE}U(Nk^k3y|}
zZ7G=^MuV5mHCpG6TAwVTP~yHqa|{!$lmcjB&@JIdO0~)Wrx9At4h6iNaX9`R$`@D_CEyZMNwKO_Kg}^~@D^m)mS=~@d7WNhf{lMEL9cIc$s~J)dH0v)
zY6oHmCOJzgYG{V~e%YaW}h)jdFfj)P3oB=80gSG7W&VQ2)?
zeHxV!U$C&@4Jq;^fO=N{uX}0n9C4ErRtt{l!ZPN)M*}2uD9^Q-w(Zjn(h~i_r@eS&
zWk1JWr8X|=ea{a5SgK|>$J{Dbqj*beizOcMTVjNs*^i(XL7iW?zXLq+bL@bci7pp)
z1wQlN?!=kDe|2xP=$k;MiZHcC9W`7caW<+xGr{d-Zb&EF^HFSz0n<{!{$7yzsx_C%
zg*&(KKA=rm(ugC9!0SLrEnn9tc8@2eS~!QK1B4*fUkTg_>YhjsUTcn?!t{t-2A~>y
z+3m;m1Ehy0ZZ61p>~Dz#UdVyz)iCt#tMJ+xZ(7kB=}JuEbWCxQF>c_06ZU8d@ae@t
zij`-leSZgN>qW^h*-nh>c|vX&zL!KR8IDBLNfZaU%9Wh6Yc5*2c5#+hNY7xD0u;&%<~I|Z@iP7
z*J@_!gfk!aZ%X}fmcGOq@eZV%WPE_xFE4rb075EOy0CKMeC6EC3
zuvvLQQ5&uPAk+D9lfN}(dtcoxW;zSmvrd2Kjk_D*<-|rA%DTT2E34E0`7Wqc;If`h
z?>*cN`og^ynDaaA;wu89Ozm8+llbmb^S7*a9J~+wIta)
z>byl4*-}fIEBRi4i)`}Z+Oz}e#bLAvEm9Htj1H>oABdvqKAT|d_E>1rs~JI8
zk~d3nYP3}yu!WD(qmnT>q>;poPU%3|C~%`l$PPq*$dhbWtL`rQwckKpgtkuvTLl1s
zcjT|8=(Eh|lEyH{q6ECB;t#5m>pvD<+Xz`p5gc`=;nzSZx5}-t87|vKxdWaoU99#m
z17lF(<$wyJcEL}q89OuIqUGAhz^WmyA1Dh0{1%CZ`nzhK1UIc})3P*Mu}AzTUK6Uu
zdi1GQvfWD5HK{ytL{|e+n_+7l0>a3|5^uK?awh|3Hjl6eFiGu6u`Rg53`Ts!dx30W
zzr4$07}^)l{RB44&ogVMk{+dQx;*NfbIo)7$M=?_o!7KkF|
zZLH}-<@?r=dvSRxCJ1>&P{TM`;YQ}%`;`pCdagRqC5~Dxbthmm7wIK_>%TX~`@GCl%G<2%PfZX*efDeb1
z?%xpcQ>@Q4KZ*DU?VbQgjWEjQDo9Um;}8O19Rg1F5~bhY-xe`iiJclf!F+u%*1c>p
z=;73jTmc9!TQA@#kTjARe&DP7GruXarL=)^wX5i!y_|b@D^WcB$EQMPql=~}Mpm!X
z*CEADVgdW?d(!!hPgK$_Bd@TkyhErK2U{^o2rOXU8KtHr8Xfh1uofQYi;7-C)P@5q
z#E^vN3C=$9wslxNLK*D>ibkf)*y(GIFL9Ic_r!9TcATP0FoV2%x-&mP3P;iWs7rvF
z8fr%Wnhv@<;k1vNzaGvcMfkbrvSl16>S2nQh8)_cOajzXHT2Qn?dG4`mmmXIA<7QP
z)@Rh6Xj>%QQF)7gzW5k?;7NHLpN8R~*m^{eF7oYAG(tdX-r~N?H`OcQmy0gpsIlX(
zZ9Th_D_iTM+la&Kvy=3YkjA}X@wEa0Xi>tR6b)t9hE}Xqk?FovF6BF1{Djv^y~vDd
zAQ5%I0V7tPFPTk3Ss$zf2F(?L$v+x1={~PB2nz5~!;yD_H;Gxs{hncHTLV_!pcnp%
zi;M7YJm6a2;H4C#rGGE;RGl46=pB)?=f^nRaf5aQ5*mbv-Haw*A!#9J08l{M#)JlC
z5NOcZu@+gLJZy-#F`3vyw~*1rOe?523TAcNA{m7>Q;wZ{az~m&zswIRxzQ-qX^N0y
z{caW}N>_5NJ9>#oG*Tl~JE{8x62>Ehn*A4p%&?tJNixVk3gUwn$MfwBsPt%DBWG6}
zVbK;~=_VL)9t_wVD+9~xa)-dQt<|mrd3~#J)N(g6M7HfSr->i8ef-;+vzf>qKib!Xkcl|VH7
zb61}B4%BlJb)4AlBj&?hurd(gx=lUMrwl4s8k4f^TjqY>Aqr)b4Rrz5Pc^{T1v3g9bs
zLrjS3i(@{oolZ{z?5h@`pq2$^$@op?B|75*M#5d+(l1-YwwQ_oN?`;qV0gF#T`ygq
zZHlm0PF$cD?|cl_H`{@!`mE<@HKFeD{B6`%&K95i@2j9}LEdTb;|;tnjXZhaPE2N&
zAs&MN=J-a4u9bCRM~|R87>I0R1TYZehy1{m=*f)vher5NtHo79yv4hS_6zqPF-{?_
zu{;iYrh;2?U#)m&UjG)T2>dR<0M$5OMJ14FleC%`@RN;kA=m{D#0kriw>vU|oamG;
zYRcV0olN6H5)2pY_DTc}rEoa^>i#1`kSV)|0?H2~jR`0d;vjg_#;(quPF<~iy+Vjwg(r>pn%;83-4}UKZeG^r|Vj7dNar4s&rO#{HAK*
z!qWXVyC{=hOePq3pMAPQR^H>}u9{~UVRq4qptLz)+C8{W6ieJ58vw{a@n4Zo>>IGb
z)gW9NO()D=w8?qLO2n2*?lK(mYJ6A1!<
zg4$zBW?>h?S%Z~rEflBfV2FFVi|i9z~O?
z>^Rs!)1h^oY(SPKUQo(TSjSMHw)SOgiDe0q#tfjtDbhz)GZ+ZUU-nWX23t$iJ-|NE
zHGr}o3AHX`88y!#DMMqkjvfS*MC{me*C|f3;o>5l1xaSmL0^^~55tl3rdA-N(%CnC
zEnJ^KX)rWKi=+lY3X|@N*Z6Z7l-w7`)!Yai6uRdX-G{qs+hKIlf!$C-5~`_d+a|~}
zXLf<>xWXHnAk^tX%(Bv*n3}EJ%XSL;vtN`f6GBsYDXJ>5n?gH(p}{7*NITgHD4i3P
zZNOvA@K(YHF;xRpY0AZ?NGz*SVJDe<}^yQ<{VSi+Jil>R_OBTcF+|nsm5_e#KrQtvW
zn}hJEqxY}1=PNa#+r_nCY6$3)-P>y4_fX)eBn$;h`Q9@|RM)+fstZH=G;}Y6+4lgI
zA{f)Q^exUsoWUiySa3A58g;XS2IaCZ;*PMu$M&xAOTr;PXf%A&myu<_+IA7K-xBH-
z?tff9K#(E>63#y6naLhtt+L+oZg~VeG2@y&w4xWa$BG3mo$H^&9KR@!67M?2DW|J%
zj%sy4@nUM&a*%4Rau}|-t_Rw{@;fh3hN>W<1<0hi1HL>YvJQ%L)N3jeGve^n?KW~d
zNaGdkR6qJG3Ib(5JJ~(&sq4Y9pqYz`Reu%aR*@6!VCV3Z)XIhsB=-TR4>YHG_6)Zv
z^-=fz8x`{n=t!pISQ#ijY%=&HP54ZyeB3B9_}Qy7d;Urt`@MQ<{TlivZ1#nc-I@HD
zYev+D%d}nV$9KTZiwWRodIfQ}nTC`{)S72MZJ%j9P)qX?weOR%O7f5lS(jgY)_+yc
zh#@yarSnWMLBQD!94XpKrj-o$*8J3IGl}lYQ{-M9jdqo|M$Tz?Yllg~G>2T3joH`J
zdL!$quy5bON~DWvjw^{jqQm)mIB?fVUTfga5K3@8?O{Iqb+mnGttl&wR52mS#WLJ!hSIoOj)*
z)lQJl=+3~G_&FG`<<-NnM9zc{0!1Yvs2?&8D%>@R&hCt+aIU-NC1sKc7@Z$Yt^YVK~|Gvwsdp7<*bfgx1dw
zWaPZnJqihF;rU7p;e)za=Zpcj1R>U#&OEjIS_`Q;o^NkcdbMUiHzULD_2!7e=2I^uWa7tr;qMKST
zyxGlc^zLr9x#wt43xBLUaB)VLQiouXI0nW-FC8GBENcnSkON5LH=Q)^Xq$6**Y9}GAxvw
zsBgicLY;Ul?bO^3pJQdsw|~CXgA>m;&Qkh4K)Apr
z&b4{)SCBQDiWlxK$8Z;(S|!E=v~|EBLhUAh-Dx~x@i=zx)Jo?5SmCLN!zYEJ)_yBx
zQ^tuox94?0DweGjJE#(UCAxyuLnIFt`MR7|s4JF(J=OqYH61>&WW@R;X-U2lklVnB
zF_qYFTuC;v$nPZDu1
zc_S9!$fQcSwA;=mbd>7R`r4mm;6LO=*qQZddeBhp`<%6dLT*81S1pVK1fdfg(?x9521P6);#6&65+s
z?>9Q(d&-}68fFZM<)CG{gzFpzBfg+QHRXDLRVa-Hb*23X$v2m<-Xs5&(-ZPo-ri;5
z8o4J+5H!ARgX)p%Nrg5Eb*&udj;O!nfAmSI-#T;il$c?b89*%JS!u%<^2bBcbo34CNKdLET
z>y%d5U4DvvhH4_fSzQ?nuyT#&bhDTP5oOT(z(^Ah+ci;&%FpbjI{f-M(sUV64}YpJ
zn_zFW!LsYil$h<{$$9977AflL<8zwcj?eIN5q~n*-;cNWk?-fBCbRO_UCgB~n;(T{
znrrSttIP*YRFLJ}G-p%S)re)AwIQ!eD+!$8wtAEVBk(`mT5k1wS
z=zlq8B75=%Ur~z#VQ%xnLCogLkovt0Z`|MjmcO!KilG_(fC;mDVKrm`h1UY#QA5#Gk@0@z{K9nDA(@C)guC%_d5xp##3iTN5TZ?aibU{cQ
z`;PPM>DZy+hS&~~csDY}fPi^1_4L(Bx_@-WJ=!CD(%+4^U8^#xBET;vO}?%smh~CV
zh!CwbEjot`{*DM=aS)`|EMv@#Gm5PDt6{{rY`K?=M6L~^fMGvt^Tgt-GQK``Nq2m7
zEX{T0D}Sh8;W6qg#7@%(=BE3mSbS1tQ^llLqo|qt>5_O_Kid9NWvwu33E~zu27k`O
z3A0JY{!8%e`$yo>NSPQ$+)g2P2G$igwPGw7{w{)Vp?!y9-2zDe=Ggh`yeAe^GX{WSJk9Nx
z`y8bgHs@tL46-t=!f7Ko7Oj|`xqtFIDkSWAi2kZWWO61|@F~gcMK~dzQq?yhlvw7K
zDf3O*0~QLf7CWrO&$aEwiqUAxGQPSdg-ZifXJ@Rq-SX07%+l3flNv|FJJ=Y3ock_x
z8`zY2N??dj-2A~N3SRfqIQS)SNqn2Sqdm%kcz91+Y%Fg0^)>i)bI&-r!+(TBamE0%
zVq;NLV3VkZgYjaZ4<;=oB8}Ej7`oV<*UDEaHr5i*k^{hTaYQhcrfkGoG&VO`FNC_#Hq3qiy*Vp>4bS
zQ~)j~9!-mizd29nwZE7M}+Uf|dK
zXBP6Cm$Ds(h;lzq>#_+yVcQjd?FO$uT-j8)`(!+Z#TYq>Gb~j^PRO&2H#UCwR
zMVggQ{KrfNVd^0uIcj%Ted|^sb-KcF`6^$t0rGt5%Z9uYG)%m<#&I&ChjbP_n*m#}
z?@GhVtI|sX!KucP3GlJV)(HhQFrDsA(VzsMpu*Q26Ls)rcj2L~n9Kd-syi?TzKKVM
zQe*4*moP{w8h=|n2=1y0eZrX`ESIW_%|rNfPkLFn+U5qYfsS2{ksg_KV@3a^!$_r%
z1Tuyima*lf%|6w{CMIoa1fu80I5y8h?q^si=VGZOU2l(QG~RQg7BmE6i<|>`ov+DqL#6+;;d1{GR+YZLJ@~O(B1b?U-26#)P`GJKR?~jtm9<4+>
z=1At`R2Ew*RXi@G^a8<3!cjze$3{i62OlLu7F6mz6}b_?M&b9kLWdLkceZCc-G#Cf
zo^A6GoEXz4`iS9gqzsg*O2igv1Z$_%K?X=VLuu+_9u+^YXb;fRLocpEYv1{@6mN+SE`ktV0q$V~~N`Dconf`KYy0zdp0YF7rO5Ci=FL}OD
zza`U;*TWO(m^NhE=MvdqiB_`Cja!N+a%Eov(UPv4`K^x2&g9kygAu78X#)J#dFqlW
z!XE>nLlK5G5cI`*4YFvd4}VD7-5N>+;%{?GHNRNWe)F8}9-Qx|Q?R1*Evb;R$~(G-
zdVlTmve#%*KS$4^fe#3dLXE7rWx3d_)53Sy62i*Q5X{r#@lTkfE1L({6G}-edAIcS
z5vNrip!;Rd(qO{D*fteN*YO@B2{@ggTYnnesY2Ag1_r!hA>g<%+vh1zO>V^T?Pkiq
zH@}COq#6#8P|eP@4j74of2CH4iyb^UzHRY43XPmChKz}`AuA?}I>jD!)JE)eAZzN6
zaJsK_s5QLtqm7F$@X~Wp@^>rhp2o1^T&@#lpUV^6O
zgW>8z?#^{ks`G2%P}egKr6Erhh50MDBk>6zd
z45ENx+DzW`54q`UA1k4`Ysyb_>SyY0_ku(0`CeY?YZ`M%(<>$f-2*Rl(&0E@T66HW
zgC7b|skekE86=m(%&bkBDt`eU7e}`Svw0*v{CjQEu}U&AS+^&`SU967!nLvrFxy5S
zX71rj?yktqr#=zJICc41VTx;z%rLpRYMCjfn}k1A*&(u~&JOOdq__20OQ0r44uWY_
zXMOlC?nNN<#M@JEJw&x>2J(9Ne21G~P4w{N`CgqU)t
z!+cfRmVG8jxMiGTF;}-~-+^~6osqO9FOF#7rG{l4pMSZPgRKg@SQE?Z_IIu}gy*%P
z-YGd0I`A^vhaPfJ)9
z`X%!n#>4YZE8g}U!2}dy9b=lqtRj;Pe*^ZP=K^@0<2ID|We|}#*A7%V)-2^rx)dw+
zaly8V6PppZVB_)YAxOq!90Y1}m5AedpR6>W6X$Lh5AR@T&wqOF1RC_+TliXmRp`Uv
z-AbR>0?nykq3PNi<|kN-4rcm7CwrA!9EgHHgdC)r&D_2N7A|@4`7(i;hrFUyVv4Xd
z)ko1d!tIRrEiQL)*ThQEw07_b?VHM%1C$0e%L!baVlV=m5pxcKlMnD16=9mpX`x>Z
zWAa#{gqPeoDSr+Vk_WGh{gnc)Ql#BP|70L`3=ShvjSo9s{lsjh*H95xA<@
zqv>nDK;P?lS*m=5hFVxN!ua9qHAnUp70MKUC|NFV5r3e{-%}0ir@(WmNc8%s9CIkw
zyAqU5;oSXj0i@5G-QH)fiYM2%Gm(m_n+I^UWADJQW0tZb60PY
zNa@rm6TTyQ9x#r
zDZ`;0vVUXjr%t8Md0?iJIYMhtB&THY>j4P}T@7h!CqMSu0g!HPMPZW2=F?VY=BDB>
zL8ZiVA)KV=J<_a+?EYz5?u(CngoVk9NeN5=W{YjTuB`R;_paEQula=@JzFPIg9(uPr4(dW!^XsdZEC8P-e|>KnJooQJb=W7>TkaeBc*
zZnZ^8x-}{mvvd`9fyrApv$bw&9kAK>l9N^ylD+|%qx`!36XE~yb#fQt;L}fi((xg(
zA%A__%X5x6?|4N48YVtz^0KR8D6!74OIIj3KNMZloC5v6^f)2BZ061~bd2)amh!4?L^
ztbR;?`541wqNf|m8w572Hjno*4K%$`Tz?T>?CYk9m&m`>cPIQjWHO9I9_3U=A(T(V
z%AmXXT|y|Mj%sB&|4!`F;Rk%Rjaa+5kZ~P_%ex1&$gIX|Y)h-PZ_OP&fsDFaL4|g*
zbzsSY|s4a=>Y2)g*40Yl>;H1YO!n^ZuDzB`
zpMKa?SL|-_Y;D%P;2(^9ZX5BqBhT@@r2foOV0IUSnX?S3>FcA`GF?}S`O#CAr)%nj
zoEpjz0^Y;F{s%&ZYd(fhXui|0&9rN=oc7C4(oqiFS>D280aB<2rA#&uc7MU<#Z|l29p5GSx@uW@)(F@oT3b>t
z_hp&y*Kzu05*XA9*>{vcnt$}cIvV9hG>dnQpgRb^)d50;?wM6OcW#kI^E6t&Ilra}
z^{h`d^6n_|`=F$q;RD#snl3rFPDA`>Mx9B`*^n?~wID=^ytRw(D^ae&7CogdI3DUj
z`ozT$Eta9&Sv)u&&~;0cLI|69sy$|
zg6RF4Y^2_!yI;{SVY3eL8mj?6wZY-0xfKA}7H!$EBex{9A78G2E@El%lu#DFKoa>gD%YU!3hsd<16$$&%7uXtXarPOFo7(;T@%TIG+zc>
z=0+-@3_PzvmPO|3vfl|M3#as3NfBc_4^CqLp?b1yh-990`+xXw%g9Zb7uvC6RaE7H
z1yzLV^#gORB=FiRNYmV2$;=?WE%}JRyYXe8-1xdVgL|e`&~Y#J2w{r!*B0nT@ca94
zFiXdk2&j+SFK`H@+MBtff}Jq*au*Z42Vt{cTBe4r%`BHbM^kDm;c;T5fwQLE8b(SN
zptwHp+oJz4i+?JZ_kddQd+PrXfO&?8z_^}OwNoCOr-V)wmzlZPlK7*#NCY)p8O`G7
z_}wc2Y{$KDpzxzzCx5z4mFp_+qYAwez8DVSL&o&%NS}=9mj@Bg0xJo58h~+bV`+ER
zLBqCWB1}`YZAZV?W_1bTvjiu=^9sjLUPP`+$@w2!d*3q~z0bg5Hwy6H!F)2a53JROJsLz_phBRfU}}w5^G1n8-`Kerp1VF8VCR4z>fPt#@~$Que2S
za@BjUPci(X$I-XN)_jk+Ys={sc6y#5OE;PsEq@N62XQ4}ayPR?8TnCQAOn!*B0guw
zk~r=9J~eu0goQ8CxYfn$5&oD0Z+npi)PEoTWPWUE&r`4#5tH%hc7|q>_*Q3y&g@Z$
z{C=eQb&<>cYlI3GA5t_oW6k$#2iUTQWOfq9undIOi(ii3=RbRqe&US(x_VV#4*M*w
z;D6A@O2Z6WV(eE=iOp;nKS@5|v9$Hf1{tkm4X`Dtvyi{D$6_nJQso}paE6)eUKhiQ
zb28a`w>+>(merM6;SmZYv*!Y##Wvyf6O&hgB4V|>pOSFeLT9|L-9r3^JF{AingSKl
zd|D}J2}^3=ECCnOJ?*D-63ZDh+70!Xc7Klc54b66zi#TymWMRQ+>3YmO+@h(_y@$G
zcuwGWBOz53Q?pO2FpJjiib){Ux#t+?r`0T$+WkxO457N!{Yi#~jkZe^s~<#)s^b@W
z#Jkgj?cQCmSy|0oDTGtHYssxO27ZtHg!qh3U~Q_Q@+yx!epE`5Gh|lR1Z~5b1%F<3
zr=P|nxSaVWbu?BF7pjxJ5~@vyoG985L#QF=V#Q?Fx!w9qgl%WrRb(QZlpQKv=Szta
zajxn5TSZtNhquL~0}eXTdM!EFWLG5Fk^B>*&e~wcM~NhnvS~D+Py|kO7^)>KQiFgg
zsn+l#EvT=P&=oGceFSSLa><+9S$_v+_6`;=lotU}KW)*@B&dbLga{+3dKR4&=z8+HiP}%2)LdLIS$ek1MB067
zTY5oZRNpH?5a8@!DU|p(&VNOt3@w|HS$x#d*IsWMh_`13f&(FwzX27YrwQY8<-S0l
znXla7P_4UKVyzcQpITJ3Yh3DN3Kz0n7iMXqgy!^L
z%wZ5Dz0An$8huE@iZ`8PWV6vjes1sObPPC&bl)Vcs;s2jJWip~C4VQ;hW%iMoK_jt
zf~6!m)PcQ8FL@i>8Vrd|f99csBF6#=jclaWxt&+r$ROr*rb7IVL_$?ZpR^#3=IcXi
z=2qPn-jYO=OVab|kPd>F_O^3ee)p=EYRO2~mga6n&BRZcI1zI(ee}Q($M6q7@|0-!
zKV)a)3m;DR+@f=#X@A^M_-s!sUNH!}$zG?~Ig>8Jr*Lz(BGEw!9$;lm`_av#UUkGc`-B8#JRiifar(`+EOTG8vu{>Y0oSIhj>b~RbUuPi-n_-Yp{
z%?xQ6YoSV{U~rBENhOh~BMsW9pDS$4YWgPh#Xg^(V2__fA@0&1FP2Ssyqs?Z*d%h0=^lBua&AMpQX2L#eTS
zP!FnTDROLuN`L+5epaUfRkJNFKGT@8gCgj>jhVd8MHCo@X_Q~!7*X%ge&v65pGG6e
z4zCE6c|l`iv65((>7ReKZc%gla$iHsj8iyi=!d8|$ncekg`X<5i>_eaTY(!aDFp#1~d0)RJ?9b$j+Ospq_$K
zE^nF|Z^D=qYj9xo>E#_#JcROu7=H6Zt!$co_`&$=4=?2aBfFCTQ
zF5whLj?P?*9q3ffDb8yhOsR+^%JXb_F}!@%nTR*i^Kvr&Xt~89Q&Upw%c+GpdCY2l
zUw>Qq3;@A$&24peos*7M!PP;Q35e~*OF*Taa=TR{z@0sbB|-o)dt*3OB|%Bi$@ZKX
z7>bYwLTy$U9bR2C_nxa<5_~l?8YNdIgF~@9SeJ{Zchc?GcM^%FLm`(Mw`b6(M|ql0
z^#k&$KH(z?kZ7p1X_?44nGF>4$sA(1(|`I?)7j&~7HS!v89}z49uLqyCx&hM;LlMT
zKc@MukaS=pK&34A20GAboGel}ey0fMS8TH&VM&Vs=Tq30F9wdpUUg#nzK59aS@1_S
zsn7iMKQ!1U7!0A`r7Zt^?;aEuE4@&1BBkXJZt4B2hR8b$3I*duO&!xC?SAeq$baO3
z7B
zMZU~29a#dEJwIkPDO%DuhQ1bG8a3=uyAxBqjq1of-E6ewD=q9YV3)y(HtMl0{UQH&m
zsA$)PJQ3lKuPNh=pN#EuJ(d*xU=(?$0`We{i>WJ;5U(7EOa1WUO+Z|eIxf_4O7!K%
z6|;1)SdFxka;j;8v1>;p3ycv
zBEv9pvsLUe%M_CubLbESC7|ZVORGGf0Ow`5OP^IWenYDo*?;1Vu3|rlO7PvXuP#$#
z*a|a*rSu@e<4mx(qMj}O?nG~Q4ycg$sdwJLcXyW~hJ1AKWIrLz$r=_^r2Q-b`5p{$
zuQueo1*RJO{bvcAHW{%d@Y4;;1W$|+w&gF={HnF@^vBXmHP$QL#3G?l5zkz1EJh$c
z8}m?V)UfEEB!9Y*Jmu^MVxy#m{y=?791%V6%R7&W4U9}RLR38b{9^LGImp@EiF?x#
z$_-(6WOvUg%Uxml`Uf_wR&8$zJcpKkE;)W98ohn+Ns6r`O(nK@JvO&{aFase#6qib
z&DFzKY1LLe0q59ccU{M3oI@_M_Q_8d?cEeh5P3&%ihmp83tDA5$z#**-l>bhHR^=>
zZdv*vX5H_3VgJ9j?P0MFMFF5_Y}>Z2#Nb8Y+G-1XYV`K-pDO&
zYx%dg&hBDzMoEKg<6<~b%W#*t7^+xCK>=^Lmnh71Ef4ZfCoi2;s0MphJ>*?_Hb_83
ziLRdX?|)>eBE-b{W@kc%jyl#BM!rU
zG>BM>b?MEKWnv}_3=aQ;528Zu%-6ma>8^&P2z4h=slskv&L`$~6etZqDi!=>quR(Q
z6{j_Z@ztG5mz2tqD{@YB@y;k&Ecf!kcqoTv+<%XA@yNgFi1{3dwU?0RCyDzw_M17u
zV&*(qzui{
z1+WJ8%-{yoV|Y!+8@;8A1RZ^4Et$o?@@<0!;@sS!B`kRYzO^6j)qYDar;7wvfh4#m
zaep+MW8VNAQp8HxA>V+FK5A^a8hT2CLNbA&R@hqoR&|)mK|Hl6$ZQ=
zsV>fdi;0cbyL)UjJXMR&7T7@{3emYU?RAPLh>{VxPvfvcRYv4xt)FIBk0W>f#FqRa
zqNYb!A-NrOfVMGjfZyuiGw=;@I(_FQcz@^6CB6TJLMWu54M&}Uy*5XeFXMQ*TD-B-
zMJgkVMSJZF#TKvUj0TeVU^9XLM`aJUWkpL84D-q9DZ}ICwK$0?;9V#FTkL2nEV|lp
znwhbc!lIsaX1+Pt^jK1N2UeIN;p5)E?VHQA`Xj(L8mu{xWSkH))7t~9`?DZVPk%!p
z+_(Ql5ug=VMy*=f6-%$1UxKOZ6=l@7*9sD!Hj1)^iSH!8T@
z;zR1TUwTYdN<=&+7c$u<{+YhT`a1s6+=mu>9*g|PB>YHoEzD;&*yP6(?tig({?MMS
z)%MwWZ_)%~r=X(C_XR-Nx(WP*%r*7C=g7zf{r7`Q
znP*piD~hBY%2BC*LFmP9v;887A8YHON)SWYCPNHk0QUXK%Yh|N9j>
z8Mi9xMLfsB6`
zCz{i8`#?UV9&}lj;cwkl3jDU~Rj1RHmt}|3`wlYvL}&04(
zvRa_y5YSQmi_IBBxqtLfwAAw{zG9h7{1;(5N0Xzu7H3NbarqrH1wAX>ZXU@LzchjvDERcowlHpiHdAs?>YScJu2tf)Qs
ztsfgA(rxXH){gfhtwf7Aa@r@p7>U96^WYMEe@IsS<;1%K+mOtMlsx!0Gy(B_H_
zqoSptsK@18E|PV}@+{4QJ`XuKr01~{r4zvP+qFU1GD5oj{aYxkpN2{l(q*u}v=YLp
zeaUIKx5Gi@k50i~)Z}x<1iAuCc{pEr!uQX+QDME7W#3v#E4(bL-s!;=Ux_p}G;y&u+!@rVej~T_`m$%0N$l=V
z1CKlufE(gvkxq#b+RO=Y8dX4TIC3@{cyml%AQlZqmDp@qZsavJ)KpKa6n-_{Zy-1K
z)=N2vqJI>YJ>dv7?2=B{rrMV&RAp&8LxUdU07F%W{YA)wVqJPE%j)>)IbP6CEAE(M
z*^lYx7uRX3ATDtzFP}OUp#^hf*_(BDR$!(mV91Z|`9!;G_tfDAKPS9ehn?0H|0fu7
z4AZdsmA*$ou$y(}6Y;|!7Q>7AD?OSAWuBu#=6~S+{OA#O*0)2?eZVeh?&;ZDp!b-W
zFuqS?xJ}EmrtvoH28LJj%|=s{3b1PW3ESkK6C*_@>#)+aP|P0
zJhZB3ZPH>u9F_p<?A2E*M1Ig>~=|8^-ID
z17nL!EYe2heaZ=yEEyi@pzfY4-xl5nVSl>YsSgL`dqJ|C#B3m~u1y`^ps88RTNqL{
zwtmQ*<3vH5nB!8TewbnSu?aky(Nlg8{Q5J1g
z+AdBqe8G>CrT5s~Lc?#rQc$(skHz!gzO`sVDkT=;gQ@RKlZs5jN>)N
zXY3dMYKVBJR{USZV75O4;8mYvEbCqZipTD5`;f#DqcZMW^8jSugDXY+%>NldV*}Q
ze}O{8oGl!-%Jx9Wf|&?Q%ko%#cCW7a&PSD;yd@#NA8Q;3whBZGBm_TaRe$AcA*}Jc
z9@BRg?oD|XDR~$5<2#o`;X8BaE&D~cw
zfGOtzyo1od-GQ9>GfVZoNqXs{6JR=r(J)LZDTdOP0ciJ$
z8H%exy4qI$UIrT;*Gu*5Iasf@7b8b5BeeDzp?z!~OA}g(Nuw+(n7>U6(*M)qi&WgH
z=jFWwl4U&j|42mYZ~9L_;6gp~oF{Agm{JFeT+gj#YRPDx@M?Mhn}0!ryWcL#s_Tmd
zWQ>e5N{Jz07yAqtZlY18!tEh!5Qg0dIFmd`hW0ZeWzOc5ozWLH@Y(rPRAx@4^mUf&Nts?RYvdw2rI{SQ%n8STn5
z!q|iAY>G#>v-9QULppYo2Cq%2zT|p>3Pwrt`0(PFaZ?w634bud5sO#YzAYH<5RwCq
z4GWF)Z1GrN`;Ggq{9+XKv0YJ!A&ms&3hFfW56gxiHCMx~wSUzPw7qlDG+z9vQt1F(
zPe^OaN3(P$cVhA2rN2lHCBh!sS5=fKEj4_cN3hfub{lv2xs#K1I)U1HE45m$H`xN9DffT`TXI-nD;O1u>b4QvPyBI
zKVU9iu_sz_G9mI@#@D1>tNYZ5z7QQ>iqg_PLiyasM6%r7$=9f5sYNNeSJT&jTX7No
zL&o=uO~8slJ~95-B>3y8baac)@#F1Sk?CUZlrCj(Wzz7fA{I$zqH2D(0EIc0
z*{Pi4IrfyxU7plo-BVrRJ&Amm<>H0_L@Lg70b2h4(ZF8pob
z_$jNzlQmR(F`Y+Cjl(pJQk`!bI`(e2n-8-%GTM_HOrUEguY!bTbo#$m8xwCJ+1K#
z3q9Rc5;M~gg04NEQ1g_Dj5~fB95;S04VubY8()A_It_S;<&Oi1DchER$<&;#TSl%&
zQ&b(|a1G7kbv^DSf+WP@HLN@gOMR>DG=HXXg2a(%?uY%L|5SF%a%(PB%iMR_=#Esf
zwf9nSt{|_yGU>RJ{l_wKZ~Z>DKR~vgga2pGRbGu&a*P~LTnyj=0!AF#c_C1h&SWBO$nW2ZFl!{V;AmmY?-3u4PQM@==!IBc;wgme}A<(
zceM+o@kYl3LR#(kOYg5F0{tTCW5|3LUUMz9hW>SM)ty=>L-6j92{nE
z3(oqYb|3A>c?EbQ+xF8=L9d1NhEs`d(egBhS`wOpRo%pE?St=9P=R6+k?;tJO-SYFq)Gv-dViiFiJ{j+EBc0~ns^mIo!2T@rdB-{Muj0@8>8=I(tip!!6lZ1
z89L=eRRYt&-uVk~im0K;Hd;Cth{DKNq76a@QAatMvwi&ujoVZ9Ojr$pai?PKTbJQ(
zp(rN{*lCu{XWs|a6QqQ@pBlbj0}_(!S!+TaJ;9Ya6Pn_GeqrPzNr0o%tzaY?d&1$M
zP2jMI6!R-ufTTm^wLhNKY=7M6A0=m75Zp9-5}>McRmgsC%sF>CsatB=XeCqUs>2H|
zP_whf3{Q<%MVNXZZ7%Y-XW?3-QBTo7%C{Dd7xJhmRM96<9N%$sW*Y<9r
zmF#YN6aUcgi6(T8?k+cwY&5(E`R4h>minPB%9!9KBRC62H67UC|$
zx457*%zuuVV_YfN-Cix5Ow9jkL|Tb2>-F#;SVc($z!qpyOMm|rvtk$=e<gh`QEEHR0LJIMD|R&fuoWj@C4}YmCHZjw8NdPz$<(1C1>K!^aWk$=G{`djt$)&o*T@@q*=WXH~iC1Wh#NA+)`KM8RxnT)ASfW9m7ai_U#
z-&Jmn=>~XppElfZQaYo`GcYE?Tb_1_m|i0p!mjz$XQ3k%%}P1d4CCu|8zE)K6Qpz?$YG}xFMjy
z)AT7h&zH2g&TVCdj4&g~U`NFRM~}DbHi*lH;{Ged73LSszz9kJew;1j
z=Hp(h4_2t5#-;sx{VhO3FDhQ$og{P+9gqnd$y)lD{=g%V*EBxVB%os
zSqfJYiV(5Qkr2|f*thWj;`b;KJNmrV&gp4&+|~e;((N}ai_DvT*hYIvzXJMUZisYM
z;?$M^74y>FdsTjmh^lt*XY#M%O3wh|;Ji6mqkr;B{^BZzm%xq9#qYqu0lLKbK1(Kt
z#va@c;{%WN^xNxlzFT@N{#Pm!Zj>ZAsO-qqe?1dxB9~1up?XK@hrjj|$LeXU7;pU`2uGb5BYeQLQ+6Mt10
zz9tK+K4cmK)8|=+IoczsOONlNF~H$O>o$GCrPyI}%@%NCG451y%4Z7;KtPD=PP
znL)(55OkigF0wIN4hnDYnlR7pJbxO=>FPWbVTS5izfrq66;Wx3)M6y1O25vx12vTD
zQPZ2e&_%JwE)mB~eI=KUe=GMiXoTEL%+EY8Fa|=vX51bKE;XLNMgzhpo@=3MVO`t9
zY*eZw_1Xj}Ra6|v{FmG+f2p*~S=*nDS|sBq7FQUiyC_g>B0O{w?{D?=@qe7@PO>7_
z61I<^vdyFj6lv@|7q!w+N62;P*CQHNtx(M&8r|l?KK>Mw8hML0!o3@26ANFxFJY=?
zs(4IJ+I!-oQwLosy)7)FG=68=#g=0Wd%Fm0>jdX2!r6|Su=g2=tqI|i`9S<=%MO2S
zx7xIIxBxpQxL_dZbCMs+JAc>vb>=hK4J-e+7%uB>s-%Bt3jnv)5y-AQ$fx|gfWr1G
zo)oMcR#^DHm1ahIhB+pAvR!Xa$A^uiXUy@KcpCKN-irW7Z$-81Qh$>D!ptwcl?|-eiN$0k8gx_&p
zeZ?Mu|3M%k(iRT{z6jfz{Iw6JDkK<|!Py0<-YeSnA_|IA6%=A5&Y
z&vYJ!Nl18u7#3SCL|Of;okIGNtML4^T}QMEznE0r((l=ws6q)@YH3+%bm!&>h)9fu
zN5P+Az2X?8C)sPJcz^bq{b%!wvTbop={Fv~Xv2`{XdPLL@H2~?)`!;Lly-4lX|(IK
zj4n;U|jA3x?;U>;Z<{&zJH9<1tysAFg*CFA!uP{
ziSe?2;B7kDCi*&6T7qs_@r~u6?*gd)mt+5}&0fK*dy>h@m
zR(XHK+sn?fXR|gjBn*n?mO`~DY%okQ!_X*Ht1zq@hBr@)7opdbutcGn^-*4f*l%*P
z+U=%Nis)J#-G7xL&E)O1IZ5poxAD0?rzHbXL+OU4*GNtTHeV_&T@1g^qsHPnVSlo-
zhWp9lY_f}3n2{oD!Bw@t6FiZDdJ)%B-@hkMGBZ7mij6l*5T}W~BtHL6T=C_XN%I*k
z@`+I|P(q*orE~vXJ17@Y0D(Y$ztH1X_O1CLz6&P7Fw*bO{f2+)pTh+sBg_}c-^6(6
z|E&{ZM7o96MH&%UJUu^*4K0O)Z171u8Va=*+*6PVIXqRtj$x5Jsvvc@5`p6^J>=2Yg>oRVT&e
zF+e2^)cuOiDS&*Wa{xvqM}BJb
zkYRZCNqV5v8-jz_dyy|Uh>Z#sXFAMRv{Gqx>n#xu~J0{j1L
zAxW8me^86oJU=K!S^;%~_07Gwpgu2#9s==s=)p?Y>=HyFYanC*jh;L_u9w6?Eac+4
zf-DgF>x40|?v<-+hXq`+zp$|EgZW&`8ve06IpN)io=^>Z3SE%FdZcEh2EC8hvu_F|
z2quIg>|=k;x5ccSU;>87nk+hRY58Cn=;K=e0dO6Zaz(?xomAVQBx_2KpjqH+9l$P>
zZYQKgCj!k$W}rnUau@yREVF@Dh5ce`wLNbDOi%v#`#b0y5MJB8>ZCrV6@tvH?YjKC
zK|0bxDcDk1;NKM#=+WzZNWF`o0D5mRgssAsCe?o{ti$&_J*?Ju%oR3PHJ3y$b*qT?
zJ$9f410;o_SLi%4c&0JRQ||jhSu_GidQbW9$NE65)iS*SAR-s!Hg}NkKGng-XQ1BY&EMNZD|J_-N#swZHu6iiT
zaj%8bjv2+%luC{}6r)Pl^@{xjbtCaYMM{73#t!-H|29zQ83Px4{R4w!fJzIFVR0hf
z^=(15Ajh2Kw~I@zE!q9}R(2~sJ)mzTAj*thW0+{$Z`qc5;LG-&UfMeK
zSW`$8iWZ+orL;$>oRfrh85%-z+)G+(}_h696HaTx2g$(4v^7oRvoFr|muJx+f_
z5;3h~`3IbvV$otfz)MHYh?iUAa=#!}AF=UV;689u%iO!2K6VWHV_T?mwiWI*I5
z*t2sUsRDK&pq%=VhIZN#jAEzH1ot(%B2ko_hMiBMwyrNHs#}S{4QEA)@^4Zfuy)$M
zP$$E!sv`QXpt;ZOEWzH17Q4?+VLE@_$X88g=Gi@;v{{1*q@S$L4#6$Y;ZTbcDU{e7
zC+yD-M*hRCkIq#o7sJkU4|tJ$j!a!yJDjtSGhnZtCls5pOdXSP6CF6f9G~-`)
zd_IIK9-tQ>g<)ZNz>qGB)4E5bcdO#5*sDvFU3?@3i}72~z5@|XGLMBmtFD^Mf>&rc
z5_&`q3e^4yXbG2uw=XrVoBZ2Yc)NxcA3=`RMPTl;PL2&Bp*O_1bs{3g#
zLYNrb0&n8aOd|=8Cb!5gB?^C3zki<_UjTjZj~fi_d$Stmx++=-dWO}LjigQ5oPr4B
zJzq0EMzFMFg<7Z#$C?;yRZn3lu{#WPAq?XKTeo^|zZ%k=OVa(hAa2;T?P3{%$X`RG)beFslst;QNhcCrc0m++*;&069c
zwB(bv7>4&gD%p;;+3|H^=X;r39;1u{G?QToKXMu$tob`j2``hA4?K_7xE>0;2@Is|
zQ#-R8G15v)8)yWeaKe8@^GcqM{r`y?F9{ybUQ@fofzIEJj?8&4fINNxWKDW$*Va6z
zT4}l_Q<96fcjF3s%BCz|?Icg`eoa3~9c{We-L5Ts+!!U6oVY8M`*vMnRwP@@Eq4Gm3B&tF{
zaGHW49u{>4=`X2@WHZzK+B{adnnvQ)l%l`opO=-rONZRu
z8>j%pzuFw1(f5Cv8tPn3eR1tE-oSmy32b01n=lmCBst!a=(M)d%6ZZ$6)XBYehKKU
z-V*31Sm09YYJb2CNgtUMhR72!VJF;LTEXD$`m03-lx*qH`IcelXQZsFV;nj5!aDvI
z-?!Spb0%3NW{iEVOwOHcJX_hTGBtqRa*hXw|E`j-EB=33rQn3xjznAP3=GWT87vTO
z?4>-C2XDc^vEeBtoC6_0TS*j~4-B42E`LuMAAR=^4KP(L1i+4c8DHx9n+_P
zTWlkYl+`$a9;$Tdm-G^%zyY411F->Gn@GzuJLVsg{_m-Zhx0tE{&3=%RMA~bs*t5`
z-?5r|ubzLdR0OEOh9h6=P=-RtlDK)zt_iBTC6r$QRd~Mw6
z3le_=;UH~NK!j6x_#^h?U9E!d<1>!J_LneGYMfK<@ft6
z=Fve!k&Vx{ok%3*<7TFx&6%ul*il{8gkFCzVEr8iHpgO*qw`Xhq=C^HznBDz3aghh
zoY``gndwe+QyZP9oUz#-AxyBBJiy4Ci==oA3;*Hr+2(h8b!v^-8Jt1Agvg?@=b+GX
zOi5Tua&@ubgVkuWbBJG-8>SL9M+IbaP8z2DI}8v@2qdn~Fl#u-6JH-987&}Z>o$Kk
zhe#X(flR^|^10I6{Bl|jHGw0swaS*Z>LnF`4e=c#K%wBfwNWGe-2U$wlDtlSJf3I}
zA(PQg*W!ldBh3&ZzDmE{(u-#Q>n>|K6@G{-&}38M3O1i&8M-Za*Vk&m^1>rk>Pyn^
zDKYgugm=i@MSe5*;#KM>(WZ0Clns9bvnTSpRa|ijoE9_`F-Sd~Tqz^H}AaR<^IERoVhkR
zu2KvO#2Kj`@|)tDQnR)%CWm!T2QLgB^TbrK(QgF=x%ILa2SO41C-mW~zn*``;%iW$
zj@Ym^Ke6G86tc0}?(YzTIM*a>z6!u)ZWOu{*mVmxJzrX)XfIt00rm?9S0xwMRSrWa
zZMnPHI@@0;Ld|Rc(z5R!n!Fj{%*Yu)cjgVX2BXeLJv)5SrQ5x6e+8c6^yA9E%SkP_
zY)Wnw8%g>ow|PoAfnwdncN>4A9Llse2x>N1l#4*B?QPwLAb|B1Ia*T!*wsBlsk1{S
z=G8tV?{ySAqA1Mo93%om+F*o9Bm6@x3ORny4Xpd5vq+kKZ!Y8W1bn#{?;dYvm+mij
zig)}}`Q5Z_w|xyx&mjgcK4
z$~oBs(FYKrOs(A_6OelY*^pS_GAwbArsEAkvsR}y$h`6=UEziMOhm4Xn3a@1xPPf8T?TA_RP&fsaeQENP
zNcE3oI*8l>vqXQb(7~_!k%VaN$@E*TX?4sJe?3kG6u7ey(b_aqj!0e4>NDwUOC3!v
zH)R~*ii=C$0<{_`E(V(id2^S<9uC6}!jr{C#5QwlHHT;?rvcF)b`iTHcyM!WC%Kf)
zV*16kvYYRU$nmijz3UQb(h+Mqs*~rxYY?-Jk}QT;@MeEtkj>&b_Zk;(=i3a-4v~y!
z+%pke75;i~OG1OlYH_KY(hFeVYxnJD6z1+bNERldQ5Xrz3iW__oWqMzQ&5ANSFT5<
zcUzSouZ}|3w}Rm!LrH|la9CQR-0h15a13ZfC6jayCR>XHX5pnsLg-g$brPgEj~jk(
zYrU=njJSW~=0@QTmLK%F29_hi-e2#tNw1&;vVr}P2O#Z5j@bZZ;#6hp`3Z9_D6)3<
zD$zeBP~PZP2)XVLwV$Y|P;E=Vl1Tz$P|gIK28d
zi};5)i77JY6j#Jq&Sz`UxdYaaflc)`wx4uJJ~>=<0#g4
zS}G(RHr4v9ZWykdSgG{jPVpoXMKwoao?Q0e;!JuyDvlns9-C1ojg6C!dC57
zBjrRKOdr$&Cp`KuN7uVoZ>}+ZvbS^iBP2YwdY#rRpmS5%_=qMYm6UpN&KK+?1$p=}
zJo~cV?Ot5U3dcR=m#rrjT>OjQi!n6|Dv>HE`acHdH^75@F`-UJU)v$WMKga1Oe=7k
zT_k;_{%llC_((OpF(Z%WMp)_$mFG{;NU09q9zxwxW6~YIgv6B)J?exjg<{bPhpoZvpgC##-uN=trlNU8H
zZgfa0_7s%}KtG8Pi8P)W^u0fSqJXDM%Ae8uAL(`V#|mX`WOH~^i5e5Vm5IHzC3NK7$ZfA68G9WfGHJ9M^0~G`}G&nbtpeQJR
zwRLwq)qVdzvdhZIC_YxUW3Q0C_a^Ib4i3jToO5g~Q!^El^op0D|QzsCE+$!cIMpa?-aK-G{4jDWD9kPM)ss|&`sXkx%{
zn4_?efDzQ$3l2sB!h#|~LZqCW$|xv*7z0BhRKOUh3?M9laRH2>9vG;*0~7^dO{7UV
z0d*(>io)t4fP){P3&ns<{5+t-05|v-FhHU)0uEp_Rt-fs!w^s&tcxcEMc4
z5ET%(h;-qtBnW7M9o>*VXg3%DMnC{9L0v&W59xyy!T@e00&sx3fZssev(%AU3eEmxl)u^&c+E#wMof{D6w0o{1^|HRA`=O^r=1eodeVEPrQy
zK+goLzsQ3%yl~f5HBmING*A@=U7P_B2E3stH0&bnzqsARG6Vil!@4@5knX<{0NgGZ
zjE4*ex{7+5YO2O20y@~)LkQ?1vDHBc
zVtg^btS{mzs_4i764GLTF!p;fr>Y2uGSb~0%Nk93Q9czIwoVu%$`ABE2HXvS^g#st
zFRc>{0dczM0>sM$WQu@)d3r%LRsQ3H6_NgtIYTjk5CHWApuUbSpo`SMCg(ydd?Ch`
zBq+cG=>a%_;b>?O%n6EpkOrW^-cSI8@`45h{9W-+L@F!^Kwyp-?9gKuhxAu;O@tE?
zkp3;kGWoCSKjXkHB8c559_-RWkO;UR0D(G@g7lCW>~L`Z|Jk{Jf8|p1g2VN|?ojUk
zyRiSZ2D`)He*cI4zwFGR7reRkkSKRB{9ig4S`Fq4g&4pvjxN88{9FDzlp?|z4g~~+
z#RP>!B}9H}OfR+#jvZ_4)dah^+%EJI|I%S6*3k_CMWX>Rpk8Q^VcL1bk
zYNcta#P>gj>6fv8D#8&7fgzj$5pf9sj6#9^NQJNyA|fsh1PEgX9|HCLH9`PL5P`&C
zT>uX+Oc3CNM3G*!SwaE;DP4$uAxTjHr1=Ml0U)hENE`rZ|3Ok%c7Gu$5ddWL2Vwn8
z{vfQS#a~E72n)e*4;S!XA~CGFBX-CBNQ8y3Je>aqu_Rr8{{|%iklSBSLI?oCFY5V6
zFAad)|3G1^0pf2E%hlsga%^Z1C<=yz{LKP;(!Yfh!TS6D4GIGwzhCg5T{F12)PC)#
z(4RK`=eGZajWH;s8`K;I!M^MMW`b=k3g&x92)kgySUL9d-+wm$NWl5$ZupzGk`mH4
zKtNOk5D*c6$M#)FObXj5p`d?wI{vU$cSb>W-IX9Pa^L&CK*DEL+1QA)d*iLHpAJNDJ>E4_FW|=I-#(4yUdR+$#*1^YPB?FsCFd*i6nLz?d>J9twV$UsCD5xWOG=G{
zyM3d7TyU<*Aa3{FCULqGLx_2-a@vWo-!85<6%PB2n-m`%m-Hq-brLVUC9N?rOx*=V
zUC%g^^7(qJOU-wb_}!@Zc1%B0U#TSJ
z;>XL^b=i092tvToVbOnf;wkf(%t`#pNL)gHm+A|-IHJyT8>iWkF~?f7U@nR}@6YGV
z-aDs^qj+iy)Ps@*^swxubz*#t(_U?>QB{%1QMV4RcM2~m-yLvBOS^7#J{neQKjffd
z3uTEq%X}urhp)AgE4z?ZG(7Ud?Rw*&a88{2kYvFFT2-0(iXtaS-`k}6bU`?URmn|%
zUwkF2bt4&ijSLw9Jj-wTiMrzQ1Kg|Tm{nxt^v{9DTq{NS{1IEF`D!QQObksK$#>Zs
zr9TrzxoWyM1MD0u`s!o3V|W3EL6sFzyZnKuU{ZPbkWe9XK0i|$dCB^2c%+?`t)zz6
znH`^|nLi>fg3)mFjlSz{)SL6ec)SXKr`_NyvqM7w;~WCtBYbW{Uc#XUCwktr^z3z-
zy2gh?iT9E%N;TT^Wq#9M#d;{yPc7fc-BWNrXLgP?f_7%&AK+1lzFj5Gjj?^$`&sHf
z+^Kns=@$6AnNIKM<$etdCrrIk;sna2&(Z}kX32GO*ek+e&gXI$*N;VEn34T|a`x4&
zQP<#yIgjGL1INNzRl<+X5P|w^@0)dWe-w9YEwxt|rEAZePhF!as{a(GZCPh|m?VtH
zEH>A4&=GI;XwJHNM1a>WB$+mfe%+p3ZXt#3!GK(6kyAvzD7#`KUJ(E(#5tLB`^j}v
z7GI5;>!eU1s%Ajlmv)anI%Iu+@5*2aGtbvtby!hX@^nmD3^|v^-TcsDyP*1*nSojy#E?o^z|kZ^W?uZ-($_jseZg;N1Sd$n42i0r2Y>q&^q=$NJuYJtv`8aZ>HYP^c1|=ppv|MC%(&O6+%UE}z
zkh>OXZ?vx5bz=E36R$|J8d(kjvyS;9nwt%l5U)fBt&=_~W($D+zh@
zrmScKQ%=$Zv5PfL+mz$W2Ok3x!o=krjiX!C2z#_sENYd#8v?9Fc}|;0QwIEuRbB0n
zZ>dzyc&e)fSd1@2WOr|Ga_
z(m|J9`of`qK4)iFv>e6F))XPCOwcy@nN+(VdqP*5DOdpbi(vqumGyV
z^~g!@zE~y;CYah}iJW;56+ZRFHk5mdV~4KBIV$+R{;YP!{rbz_-g95_d9Sds&R%+a
zd*a%EroTr=;b!AUk#4PU#62oPSGd-l^l$v1hP^>mWaJMPb0lQ+FfPlz0wjdiX%rxU
zMI^BA8$&INI*aQpeEC33
zCahC0cP^SllFfj`B_r4j?jD-mDCU04Nn+2>;`3zk*>$f~>iAb|7tKdRhbJp)1Lw-E6Eo
zCY#aRXG5(crEFV*eMj4fA3LASyGW3Kn95c6DlUhQiYAOCqavS&^m|b0<_T5dDANH7
z2USV=-A7mb>We7d)sD`A$Jw}NiA
zO<10VS(nI3U;82hCu}E6%nwMWf@@QjQHc+qeS0aP=Vl*&zux-L+vfYR@xd{FJ6jPM
zuchCQ@NTr6Eo^;*;k)LI`ydjUC5*7qa9EViEbCzK9S@M_vcAPcg_Q71Jgs_8P1032
zE_YvImM)|B@`k4&zC#SzI2f;!#jFxz{MV*E4x-;MB3?;8`$LgoH*5S-1R?$bKPP>S
zdspvMT7TerIFT)r>NZDj-fm-m$VLbT2nSNrb37a0TTr`v1=XNH6Kj{-%L0%z`(Gt@
ziZuy&HW(J@DpJE$)})9^30Xjyj%z_@zT6#7zK;unO~PWx@TBjlQ8xNF8F;m+fX3dP
zL~PZ|w^-=v^{M;0wY&g`ZRF!(1)ZA|#CMYD7rLUMmmUsQU5iFal%?-~XYcC!qWByM
zOXz*
z>*4dn+^n=~TpJ&{#$xC=2Ahv8Ky5qsU*TrgHX8PYK93IjPLP;}Hj)|H>d%jMLkBn)
z%E#9_nk;YA5g
z#o04k*JX&)(_hB4+SF7=u{^rrDNlc^jMcja3egRb;R5%n8Z*j&_NS@Ymn}cqihW6r
zaa9+Pa@BgO|k6qib_l@fD|l*%Ob5s5v+4MK7D`+TW?ztUlm*S^ej-L3EE
z1684DmIqoaxyrh0H*_Fr`Jd~|pEbBm582#yKlMnOq)^I#pKl3%zti?U`5I%m^iRR3
z3D(nlWbHy;tKo`ffFYR>T2o$b%U#5S`3`5|^M}?6eY}GZd3E`COS)0k+E3{(c81NW
z>NMfu-VPO0i92eQ9903;hV9EL1X>&~E*6G-bz_l+i0*06A^_gkd|oW6ma_Ifc2<6HW*tOFB>Bm1;H&h!
z{Mk3sv@QsRvRl#>g?okU$=U@p)@cy1JdPblZ`>4V$yGPULZ$+EHh4}OZ1vDCHF|NR
zjVgNkEsMHc!(LCAcK))v-3Wu#o+J5zKu)=TIOot)!jt07!YoGI5YeaJEvpA~N#CQs
zl%GYk%i`c4xW8(nm{ozLRGF2fM*1oDk`9r~FMTM;?dy8Pw{eh}#|(MPGJwnNZK7h3
zHaZ^XbA1wiC`a;c{A6U^^kz~6VEuI}>00yhbS9m`
zNE?29yQlFgZu*;A#~
z=r4eNyl6akD7wqNv$WSNN+B9y)izNwB;FoB`zVj*Fhjl2X9B;OnrW-Spepy2c~f9{
z31^T};p!vW3<~RyA>fbQp;ttI+k6vy2h1pMELrx@ys1jKQzxr7>SDy|evOU;;yT~x1@*^+x2k6RdQ%k32sXkZt@o5
zX1~sT!LTqKUm5N-Q3_^%cpVotf3t_`&R+S<-NG395Q-Vc+hd1E&Mi#(buLl3xC@?$K$%
z_`gI|_i^lZX7k%~AJ?{suVs9)kut*yL5D{)W5kCEc%!Ij|E
z{l202H`>wfn2j2LD=1POx0*wL-jX}K=om}
z3SL!@b#k1^YN&fPWk{Cd5eh$BOA{4xX&&83X|q5C2BL39h|^NC$ToiaiXc$
ztb)1EF%f}(5i_0FjGz9X9%{|~kg&m)$n;!P^RX~v;d!og)f
zN1HW&qq**AX_j=!4vMnVosaED5IX2YJ(J_8YU~WYgq{F!4
z;92vJKpB6x&CiM|Y4a+^5v*^dcpev`_!%bA&jgBCKqQKl+;>fB9=V!$D`WdwbxhcQ
zR2LGPrPy9we$%oAIm6F#ZOAL=dhG|XhWvxC{i{ho^o)LO
z6fLF+YipCdrn0cQc+9hSHWji9XWNP3ntFY-k7o%pZ!b;9_@QO_SjV1FOUa2Gr*S5-
zzo%^s?U~;zpX7AXUAdcGQj{0Y6CVSAW4=pf81RG7sGjs*@U^pw-I?aPj&u|Com0}N
zvDpt5w1?%JBbD7=H8uFR8d%21Y;z%qFu$X%c^-NFZ`a?&4!ruD!IQ;KrEgjCI+`Qf
zV$2gFrCdiA@sZ{zq(eM@8F8pFSmfn`TX6PrX=iufshfw8bC#aXtCN?Pd%6XG9HMh|
zX5)4iCV$4^E77^Sc`yX!aL1v`vZW*Eq8ycN-dewIQi6zgG3Z_j5~_C9KC}$4r6Y3Q
zU>{}SiCx_L9?>ZZS5xM(^Ul)VaiZ*!XRNf>Un@fJbREqRvAuNN2{$!*__}L*%b41v
z!2kQ<=7t=CAUIoL_kjH`;VXuo`;45SPu=
zAd!^kqM|ym#XJslD-)J}67jxP%WH|9eaw#6{-*1}qX7LFvl(=KKzdJ|X~(;7Q$7dN
z*US!ApI4thpxk}DELlKx^5zYal!Zz!HgSNkpd&%e?G7$?;hx3
zk7f$A{jid&oiAG#PB3&l9t<%vDxa!<`qdQ5mQTf1LB7{rTVW6TA&V%Zz8RNiD=ufZ
z5qb6Q1dx!#bZ&Kzzj---Rxby_NHWObTu7ieR3OK&4~)Hf?H4
zOqKbQ%LLHq;sD3R=LH0RbW=%t3Q5U>M6vqZ6}lC~>D&#HdW<2D1o>i1pRMCr(z&D=
zkgdl0ewb9ytYWZgtEH~{eCI~+=^l!0hC22n^J&lZn*e>8+MJzuX@dyfDBm?ApBtUG
zk%!8jub#!W8-4NRdS}e>ifKmQEPN}97eraC!JeviHAG0N;-=<*niPU?Xo#`I3b*}T
zjKtod`qyF!GUW48<#AN6d|Sj9yBu+`4N?K3g(3Uu3_j^LL8Aw)iJRrnEb91qVvKAg
zqMGF7%dR>>Ci639CsUeKqAVq*$0*wD&eKbaPBC-$Lbv&+BOAD9JUU7wR?p)J2L2x}
z-^E`FWo~41baG`3VQmU!Ze(v_Y6>wmmw^ui6}Jp_0cs8cHZhlhp#l^UI5IK{FHB`_
zXLM*XAT%>GGMC}?0~7=`G&eYtpeQJRwN-aK)&IXwWE>-vJwG<#3?nOwV;+tzl;Sw&
zgOlT&k%o8C(MwCQLhz8kZ%S?!nD68Qf`h8dTcmKSP^El)An$OSk{d&ER
zoRqDDsv(Z#37V3KG*tu?t_xV9(HNSK1r0;MW3^Q6LGJ(ph5{g0N^ndz|6_Par@H|Br^DWs{q!H4F7f?M&Ncj+QiY&(bd)j0o#5C009Jo6e@nZ
z?Qh)V8O(scsWDQ$C?vm~27saujYigm!GeN5b(m*1Hh-C!QFth+F;IU(4yg{78FM=R|44_c9d!YYxQT}T4
zCv_tdNX1iW)L#`rzza_Rx7ky-&kRr8v7rsE
zElf=u9969tvqw}#lNjm{p|oJyPV{y?Lt`slKtuZwfKWdKY|p6)5r-uC`7u~iA=~m9
z;~6^9NR$xRf1ht(A}NR%`hPgR@I;)~_8H&;$S@}&-ai1eF#b<}3d02X+vW|@05|~p
z17I-L2e#dKXKuF5h;1`Nkg!lPi41sQ2vjf(?*%fxAfZ%DAPCSX0bp3@KOO%@5QG+h
z!((ZTk!P$8WT(0X(TfD={4z6`{OS77Fes`)8QY}9SXms2NC*LNpce#YO`B+W5)$D
zboq@u<1n_V12893TU!@<<^MiOJFzB2ED47vdIM^jhX4$Jg@Orzz!~GArl|>pA{aA|
z1A}+w2Y^9|BpM?HAP3OG051{+vVF!{8UPHnZTf{Y0hr}qNLvkn+5e3gk&b^M4FJ
z{&Sh`U39A=#ntuO6Nw`UVSLq%(
z9q4?SfPo7tiDfXe>xrn4aqc^5@JK<_!^ms$1tcqfbKR5SX;H+RE^;yagrB32Qp-EP
z;Ovf0V#3kkqC6&st=Er1ouaRy2O>-Bq?Kk-*KfEMX}r1iIO(wj;M6*Y)Ok|B5wF&<
zwQo6H+O72QaNeGv!XCr}J_=!f@Bpt!t>{2kEfZ~PgD)`-(`Yz&u-oIt1%Xx+E4iWm
zz;w=k;|~V!NZ}o}i;;8V5*~px@I(9Mf&+#^Pq<*sAGj|lUS~~j>=!~M!}=2&qCRE@
zl)1cpvQb8_?7%#mDT3h8>(N*rt%DcfY*#K31w{kjjY&;+`k9)9b(Zbke=O1T`8q|4
z8pkB#b9aSBfW`{!$LC^RXOtG7V@tgn!yN2?q0*-}WF_?0<8`p)?U8#uo!7fH?nscr
zcqf#Nt;i3)ulWYU5)T^p9pN@LGmIlwYfwn4({G<7ng%G2ljzm5o*wSrMqlxA-0)(d
zHGR`bV?FP%;fukTrnhTZ7MfYTBd_THk=w+L~5b5kXn2@7m{BmOml>?O*E?#JV
zGd_)t0PSa*bvPq4C5476kXQA(w7(Qots^5Fm5SSVGKh0Mca_;5Y3qB0KIe^0YT>w3
zwem`0RZr#<$-`XaNcD2!oXAOWfe1?9*(=umO~&rE<&8hoKby5%TKZtaSt#pqUIr4d
zYMG9Y28u%;E12~%&kqX_e_meUh#Xme|C#O?c5{$_K8ZzUa^X9VW&Wd*SaI9J+xJB+
zLB-UouFoG>TMw{HO5AZ5-wHnSLc+r163@j(zi{0d+?HM=Gi5457~T
zWr$e+H;tcz-)6F$U!3u)2o@Q5U3uuIA!Y7ksEpAi7Gb?3^{MxaJ^^KC0`|6la~36R@Y8eE>aw>hVR{)Qs{f`-LgQZ0}CU(BSC{A!lPm
zM!Pf;mWq+c;FHO(HXr0AaUC$XdCKp8>x`?XIrv9x)&yiLJYwK#anC*+`+Kz0cvjFM
zR*y1`j^%oq$NWtnlW4g(E0;lDXmDWIDW>Mw
z-rboasAP_R(aY#lSHG8zMV>vM)ptA46$dnjDT(iYB^MG47u1;&@39+T6a8Er7sGs@
zEY*|OeIN6wI{LCNr_1b6oz`ro_#-vd+98HGH!1@`?)hSZaz9K8Utz@m
z+XT5tw8i+y@!{ewwY9Lp-RztqVM`S??evBRP*1|-S^OZB59w+4^bY#F?aiU@&%$tU
zE5DnLrG~P-LH)EFf<2ec$ny#cLiy}xpeCy+Vixlm`2%tEpI>AA$ms`2=^g2#t*fnB
zW%BNSw0A)ztS_Ei7d^s!jd^aL=M#3X1Rar4!fk3oyn-FpBc|?obr_^6?OlK7^M_;S
zQw7jR+oV<^R^p6hif&F=#}+;eQl8?4b?^=ZxrS$_+`pczWB#~tbVz5>r9A7CsG2)Q
zird%M01MxfB~z$*y}W666bb*
z@wX{HPSGq(*@wf>^*@`DxjxsYJx=`@wr=>sn3(Bm9caJr#mdXm2PTAm^cU*L_luv4
zPe-iuR(LJ~1M3&WtI~~sgnLB!+DQROlaCI{&s5uW6_~#SVN4G{loyfy
z;tsW_H|8T3Ci08jxJb{w^7$Kk=m*%63SIuz+c)soKsawM(t+e!gA&8UVO@{2$_E7d+~p6Hv&G~Ux4cs$>}H%tF}UIJ*;
zm>0cwlJ(`WYAcT=j!9QcNPbqJeq3?d8e~^iH@aRh<37AHUk2jz;;?Srpu&`$r7p9*
z*fY1%Q(LW3d2P|tABx;iqjyn%7Do~9rK_u03ZAkW!fBP5DobZ2gPz9%SAGsR#)v3Y
zs&;9ZpC}5Swi$~)PxRqG7#cCJm5T~2NV}Gw=wxllY0lx^`}DIYPuRmo<8{vKOZwzSJv#RtCBR#0R*4zRi@hGAYEb
zVSXO22rwI+d#G%eSY1>UmxD{{P+yRnl=Jn(@a^XIfEd@MswJ#F!g0+(7fM@L$_E|0
z_p%X9vjR#5Q;}2c&tGDnt&SdAH~W~?QqIp)81s^+a#ZEEsWK`C#!rtbwG-EbZ#hlg
zmUr2Zf4ICrbgdDcvbxwaqp5+WcWG71fS@xW>$C8DqX1Cvb&Rv=b#-CPtC6|}^=$FBpES^!3QC;0^K`FI=aapR5p9;U7&K;4V
z&n$YLe5sh;?0R^4Q`gm^*v6|kC3N{t{2SR}E<};MU7G%@PeEgOBfFb#;omxkJ=R(5`U+A8D_G|S~ZiU~jb6D0$CofwjqW21ZGuT`Z
zjr7{Yofv9dSiHyO*yD05fBM`yzhygm1&x?a3*FCCTsb#ZCN_qS35f
zt(QXIz-R6|hL%ocfA5C{cXhtEv*F0=^XGJGQt!!6!4nieqK{5
zY>`~Pz%{JwXjPjY1V1S9DzYt4>Loj85LaV_Nd{F)3G523S`!EYKBQkFbfMrrHQ-h&vZ%xPS=sZ6e_`Z+u4uRBRTZItgDt|v
z9g>$)bWWx}|7sV+rO1}}x+ccZaBX<{c4cmR0yzz@@OsgIs-SO3q(VU;CKDU|>4KVb
zg)ijQ29L^zfw%c
z%8g{MPL601Ro>JoE}*zYjr#g$USS+-?m8Xgr(X}{Md3?Dqwc}8vF|$_Q
z1=jw(ECMGbe|jNLn`xGKqH4RTht<$V4vu>-7$|uti80lB7dtL#N|yw8B1%;x@36C<
zKTzDwdQz|1@0>z`<8(BAlf@3d|C*|<)AOgNec*Ivm9OAw~B()CP=;Uf<`;#m`
zT>%0fAItoy*I9opwzV&sXG;mMbw|W_DFK>uX>7vo5Z^qW#iz42Uf_~@PYx^pVHNuq
zniuDPAe}!Syf`psY1gP>th*My=Bu0E<}C*P}go^!3)+cg8K)gT-HKYieH4AMA#XI@~f}sViZh
z8tmE>4^$PGZn##!($~KgIo%U0FG~uTC>pFEqfbKh978VEB&C!`dVJ;e)G!=Pltj4{x#61i{37L-cv)5_iwEm7DNV?aFtZo;nWZbh
z>5qm9%k0V-^fGwsnuwnGndgz0TdE9GdKz_Ol59obZ9alGcq+aF_+=>0~7@_GdVXhlc6Xlf3&(~RGeGZ
zHHs6Q;7;N0?(XgmK?f)gxAupq%TSa5f@;2s=q<(#)Wr~B)_cQBxs%h%j%
z@BLH(6`86AlbD608Bp30f-P8C`4Fi9GC&Z}1+26Hn0W(~fNrLm-cCR^0Ilg?
zK-JOJjmgZ^6|4q=tZYC)IC1!xKqbFEs1E6&T0nC8bruLQqM@xVvP#2)3A*HSXkWp9Be^S+;V+IG-aCdTY
zboswpBs4U&WEcUGV#=CQ0H8J_Kt@YL^Y>3pAP8K)6(c}d6RiK82W0BEh<+?@DXSUf#FnXTMi-IyI+
zteBnb|Ef>Z+Qt>&f9dFA2LQji0PTT)5#tWB0MqGa4g5R6?^OXP+L!}DuE5_;(vE*C
z9l%tAoxpOpe-{QO!tHlV`@hWru0Y^_q_H-2{mWNTRaFt-U}^($1A;BTM>&EEwA0TS*mF28dq{Y&NY?==4^UEC4enW4R}f4`~cf3KJ+$lcZFAKd(B
z+vbiSR~uJ1*S{kI0hTuQz~9xo{$4X1&|fkoF=aVvDGf~~MeyK(n3NpB?0}fvyxjh>
z{vA(DQjs6P%gPO4P3>KQ{x+6C@C%}^tEmSN;O62E^!NRz;y)208xO$3#@r1&>fr5y_*Zl}
zkfkGl?{6`~8#i<7zp4FOF6U+n
zo_{frl|2x=EPsi#es3vz@a%%mADiFD7l4V4ll#AP;ITBf0|8xK0i1uk0l`H4SH)oV
zf7c6Of02{XSC^G%_y8xWr-TVQTjxLD57nGX|z#{fr^f%%Lu!#RbJOCDnKZqB=BKZgL0a&E|
zjd)lAEYg1v8-PXT4`K(f$o@eb;D-JnP5_Jie;)+SuJ{Lmvn%~U;OxqO5IDQazY#Au
zyXqeV&Zqtdf%9qnLEwCve-OBLtv?8yPv_r=4~$IzAaHiGKM0)N{NIQRtTT7C2QT!$
zB%GYTB@PaM{MlH+C0hIgf*At-i2&~Vui^W{?r*)byQ%%ZB^=;Bz{_ZF>hO;!;7%?7
ze?WF{S(Y|`qH_F(9{))5+uYIJOQwSZ4DN2rk_A
z9}vu$-9I3h82f)faC!eIo(;^%pU7NbR}gq({6i0>!0}HpU}wkwP=lLs`lAL1b20_r
z9ri%WKP+;x{h!q3Kgb2|MiwWai;d%-e+A(L6X9eJzLoyO2TPp)tjBNU4s`u%KmXh2
zcOYkXM>pVq#%E&#^Y{;z*}$aziOdD=#1-gZ^PeNb_1gyM@ei)Kz(HJXy#AC3X2{jv
z)YbYQfx-Fzq~!v4>t^i&{KtZWS#k4p{D%WLhWkGtc!50r0l}4d{$rfL-d_KJf8g-m
z|A1iref|^&j^qP$`8(Bro@1)N@AAJcDb_y++y6b}{Dn2#TpaCyIyM&I2fTlnD4Dvs
z*mxPRf^Qf$upIpLuiuRRBLUT)%kdxH;^K~8zD%4PtN)XpTa>MeLf0a+Lp2ah4
z->_WlF}xPg3EBx4CfXt0UhlCIVoiI*7U`$o6-3|+%fmT!cVg~9#uke_#nA3#@!w%e
z(cU|+EomLV8$`LbBN0Im2T~H1Tgd=uQVbDsJTrTha8se*kE22+6}fh}2N%dqNYLbl
z9^yB=om_a}g
zpNH(wTFdZqGK?x<$tuq&AKEyq8I~hpv5k%JFt-HcKPE1yFVhb&xl5eHaSm=}zITs5
zZ{SwMwiO~17)^W^f5ph7xiPqVyfCV-i<+-Z@Pj1ri;*YY`sdp5k)gOIy6&%sRA|Z#
zitDv8+KbUo)5+IMaxZz=PaiP-xE{h?#S`Pll_2nea26`mZ(o(IYLY2&ovC*=^yYiY
z#PSIj++QkbBSdA1*R)HVAA8jZym!w;u!X5j7DQzfESp$=f9NAf8GfqnsHJ@609J-5
zE^i{@*2MG6&@%2}At6}K9EA)UKr8jH7}fgcgp1*OCacUv+2t6Ve<(gEP2N8vZ{1#76_Lyrp&drF
z&LUYDlm-Z8eJn(P;&6R+m&28?ki_gc$$2AN^uWQS%M|l^H8Cl?v6BRKyfsIeUwnEC
z0jsBIj0UfCV+^fT&e7j>IeWuZm14~tZa}zdf1qXN64sWPlyUL?nX#ecBi|~Y!>-t5
zFr>rWf4Me4bvVK7_6E|NDJZ`PCy|N2;W*v@G1GYDc>CScbZx#0PKYBRb@*AxMsf_V
zSr=DuRiNP`h<0;w&>`FM1hPqslYVCzEr{_goh?#(FDG}QUTOof_~CRSU9e)z82%bD
zsCq-jqhyQM?0}4s6hbubZCDjooTH&8`*PXhf3_iV9QtCGqySMfoyvkDwLyyf8=cCX
zL@Yo3rlTR((KtToT@($SUB87}uBZ*jv#ll#H@}kcL=<*PxoLWQHwWvWhPFA|kMrfe1AoDsYb!``btW(h!r4?HZ>7CIBb(Fk}y
ze@;upWaI5!|5WR_gyw&7D8l>{DG0|7wRVlN5z-)8U+qkD=wOg=lkckbF+hXrFpe&~
z+BXmDkkL;K!7})pGh;FhX>iQ*Q}i7|cH8Rc>!$H$UaL2Hw4P0rwq_Ti32X5kbogA3$Zz0jY-zkf153
zjcY{))5=^0+w&p|9tSARNV~H|%*4cl+ZqImY=cc_xqBuc2^g0Y=I062a5j+)#@!jY
zvo{3l+)RC@pdifSZnyRF%xY+DLti5Ne{wAGI;K>zP!bxJHW*ySwd1%Tf7O0x%uMgl
z)nt@9*O9IL^oirx|8<1Q81~3?4}$B1J%|;`!f_PRHTp|)rFz@M+*r>3(8T9{CVP@s
z(a$S1J6+_W!g!V0`3$OjZxM6;R{GN_H0wkOgwt
zu*!chA-9Vk`Jy9ba0o$&e`!TUE6;p?w|N;Vy3SMXzHE9VM^|AM$y8l1@BI%<$T6kHfJ6TF89B+OAo`dzdSnf(j{RIDjk9vwQtgJ#qZR|D
z9N9`iMFEBhw>XbB>JL3CflrHh928#Gc`CJT+x?uH)nBnnKPZrAf6SX`ZAdGN*DmBc
zJUfjpr#O~=yD+e}rfxV@GI;lP<|T?SDXxo?AMy>XY!JEjBZdxE6q>6+zI2qPdiDOi
zP`bXifV!Axbdu~3TiJp1E5+pDnV-;X{yi+yLkpy&&m*h2i1s#!?w^29Wvs_RfqcC10xIQmBlLkbT0fGGjqpy+j?#?RHl+O#6V#>4U
ztBkMG4&_OqIneR*!yPs$t!`SwJc`n+k77KtCLV9z;|I}q9j{3zhPU;Wd{D~Vvs>-w
zI793N5>D$QxZ3^rkx#R6PS-cj!+a_C0ATzibg9AUq2LfZaef`Eyf
zM95-7c8pUSsT14yqU;>7`mjv6yyuA^ths
zGD<}IE0GD@e--N1d3cT83;zaHz9tJwpf%z710USwTP)vC((|!qjVv97`BI@*-=$)g
z#sfd~zefLJM{$Q@YfyhoydbaX;l)Kh^@u5MY@eG{Lw&S6qr@K~r1+Ki!RX765ht_w&krthg3<=4pb36-JoDun_2W
zxwDfXB&p|NEBC<&jq>32DB|nH=E}w@RcEF*qMShKxK9>MOoAev4n)3b->BJhVYrvj
zC|zPMe@wgve^&fBsxgV=5?2|3U)e)ZYb#=;5nz^BfO?B*{(g*pnb0+{38HUJs!%P@
znY%Q^nImJ6jaTIpbta|>ke@n?@gut@Blc0qPb<#XZ0aQ6i^+#y$@k+$vBjjg^~c;$
z?u4bQQCr0E&$#;iOYzVmSL0af$}zdP!O@!6iK
ze~LM-rN9^5b{F-~F-XB&glzuVcdbL8=~e6UxSb?gVhz&5APYx>(;dh}zw|$dyffY&
zlt(Z7rY2r=2cp^S)PcO8S=hlvynXHzM8IK=R_Py==~gWr;OWS~?tD|s{rn5AiGhp0PQC9gA}uW4pQ_8)Yi0GZBe|c9`9Vb|@=81gOybLv-%xyv
zq3fSM9%)>;o>7)Eubhmj<|NzAZh!x^*K>NqAg!{ep?(AMnvJ>GNM>oDo^P{;f1O8~!s@-M3(KqdC=Um=>(wgX!56kB>S6
z5B&;V$XyqvtHm$E^Btb<<}m`m?^Tw_)HREOs)Ote7sV@h8F9OSDl8D}|#o$F0WNrEI;3ASvec
zRHugbw>#?slO;381(9fF_My}MZ|$3sL=6vk?YL5)=+pUe_JMOfe;a|D-a)v)!Z1P<
zLbq!j_9fbNq$V~%lqBUlL4C24)-*)62J}LK5v+deO2J$FkEDw~k9Ms}AW{6`OWT`Y
zCBJ?Fq!wno;faXtI5KLtHDvEq-5%iREn+Ub3uCD2J2E5-><{m<*WZyDN*)mATb}3}
zD;g(UGj?=6hOgUSe}gzVs)_eo@kvvry$L>LZs`#=+wh(};a#jxC1)!+6OA#OFTmL(o0zz5>a3bGdIGs-AE$)^_gAYsR6#z^qW?}l#yH2Wdl7@ov&&9Z%VRf5eM>sDOvCZ
zrj7fg!fh;Zf1)cU`m!8TJ>j{6lDE4QQ9m9IPl+PRQ`b3t5kIyjcut!BnnL~BMl*ky
zgo(VlhEy05KTewx#LE_p49bF|{;n@l`u3aNeZQ#{9e&ZadNEI6mVF*_QiGapbY#sd
zV$pZJYZIpA@k0wA;G5EuEV!jdJ!P~JGlm+=LYomvf45ms*+jwscLq(yB(~Prf}A)-
z>dH~MKjh3P7o%l$(8cxZD}vC>y%&_q*R*IEnhn~R%TJWyA!-r{uyu!|)^=Ry@Jv5C
z)f-#Wa}_siV-lq4+yJlJ*tR)vC=wbs8Jko!l#l~?=hty~>_2&B=o1J&lo75j~f9y1O{wXCY>qi`Ql>~{>G6v}!28>qraXYmq?{W1BhBqI{RvuOMgcZH^Qbo%Y
zGcfG@h5lZ=i@b0f;~B^8O2eo^4mAOmQ0851@`1@+$vdd9J2Z^$%l8bFhu-It2sE}w
z+s?#9M&t{C)i96}1=M?LHWv6>>}6-kz%PW$f8odu0ne!>;UY2lUFeoN^@c5ZoSl>?
zuWkjmnlF=!l_&4szq_2v;F2bl*DydRG2ZPhSC9Cidr^$U<_4YMbEg$v`J;J=2(i&Y*XNjHuYw$0Q7Kv
zZ$OV)yA&;~3nR=$rhtebAu~M{|0GotwOjQ^ksgX-lTK5luKc9O=S~~}wdo|QGGWu_
zEHPN2M&ApIQq%)+R+txCEy;1HhMalkfA-)bdE-m|Ix$-98p=i4?K#aE#N^8pmj35C?7HEb~c6e{I|X
zj5Ho`Wi%7{_zYJF`aLcP8;DS>g-fXP_3MShaV>)?ckcN1TGZ5pD20n4{Iut+4VcfniYy7wQE@70Zm8IiGzO4P5p{Dfm0_jMbU4351C$P
zH?e)g;CS)b&+9ZWRPCWIioB$8f5_TV!ql&)BwIe^JQCzn7M2Rf&FD9H&^qoMqbcb^
zXuYioE^{*FY2|&A_@Sa#ceN1-Wo>rZxQ~nRC%GS4FjNe_!5)z`K9Kh%?3X}VHCJoE2a6Ahf5_Sj3J8P^
zjczr_zrJ2nxnmEUatq2+YL~*kk8}<}*+X$IqWgB_pp1nCQ^M6m3{wZ$&zY^=OLEz?
zp*_l6Mz%C`K5Ry#*%GCXa9}sJC2+QizZQ-`B%Pf_Q|)Z~)Z{?)Z1wm3mlRC9#$4R3*&u1#t
zhMXTKb0@ym7DyF>A9KAM-9X`>b;z$0WoG@=3O8GO-yQ%LUW7%X{49Aq9=!r%l0~|G
zH)l9dUBv96!~Eire{6hhDI=4}FzVR7oaJN^EWVNBy*#;mtzDP}IeXul1ySzWUAkWfvS9LuRo-1qZ`>2AcfV)fAq~+bqDX=0Za{`Y5S6K
ztBX%w##k^y6oIa*>xuEZdb=KMaU8zML#(R%7{)gZf_~4WiM=rD6bmZ$gY)$-BnlQ}
zMLa14;N5HQGe|0az{9j#ZSfFU^f|k=IR93YG1waF@Qkbq;FU++!)z*79id
zdf7^opdOLZCI0
ztgspRyUhV=Q{2=$_@+#>W#H!z1+X@)eQVAP(*Ci9rLB`iaxUmX9?
zB}Au>r4(v;O8qa5f8|@bt-UhDY!~(
zVyreeKFW`LQ#e<9lALHenwUTks?{z^!`~cnhx(3e8+*O})cX|O4{11qmer>XCOZgb
zi}2%PT#H1RBh4sleR
z1Q8_M+QO-4HqaE9v#b7*N&sb4nb&gNkxG{6L)TEcvm6NtW^neeq6!f&Ctt|6*5+@3
z`ei)~%__*y>z-!yUt_uA46rTZ~(2%iL(7jW2AJmWoM?!s6lkrrUNVMRcI?6>cmzL0>{)W>=D`MSvvW}&s#
z_39I?(P@qiD!JidUj8YJafdwDM#m}%f7LqRh-96sN=%awvVKb?)MPj!;*k0Fpc@_N
zYNf&|RQkU-voG$r_9f7Q5o
zw}9vv84$#i|I`?2Ti)egL@lpuBEcT%^CgHKvvNV05t`wLCxr(&eldhCOx|>zs@sCW
zQl_bqA-e?B{r~zz5*9}0YjOAJR-IpvimHyU1Lm96Y%xUxi>{*Fx276}9{BN%B^OKe
z_+&h?$D4k86sp@pL>`Dr`u@5_e>NI;L5^kBus`FWBScF4YkXWdgVgZE9j;~|l3|i3
zF(*D8ySOMZwLx5lx~G&FO>`OpMV?4r?l^J_`yq;_BzSY-q3A~kob)d?4ikd!j5>>y
zSh7*J>?v`l&At=QzrvZ&av-rBdKg&m<{A~e82us^((bOUIr;2N=-yv_fBYI|nsnXY
z;2NyYU*hb
z##5pGCzdq2YM?niOY$#{0j-BpG{bPzD(D{e?Hz$
zajHhd{PrNU83uzgqxWTQJg?=;xYBZi+>ns0%LAUV*j+z2Rrx3aPLV~-O-GyBQnTvn
zfPR04n$~sbGe1Tz>x~@M+xo@&Gb8=FmiEz)B?1ZDs&}>r@Klo$YTM#Av-xg
z*@m};!;UvwR`7tK)8X1-6q{20Usrj}(SE&e%##gXB&5$x{mg;fG8@T3_f=kV7}OfH
zG2X4C6e*1;hMe<_+Xg`D{gI
zK%u?&iCzUAeS~<|e?vMO_IQfy8QUfQI6(o=xFbkgr4!E7?Ysh2U|;(4;!-J9L=9X%
z0eWUErOPKP3VK6fnfIIwr{gg{i>3*
za{7zwTE22cf7Bx1{j&1^5lEDpSo+FcuJ_!pim{NIFjiC?2}lJ#oOX|Gpm<-8%Q0Ii
zkn`*jRF-XGAO!gxYDJs=kNX!sv4@|t$XIW3mPG#p8B2iL!>-s>;lq60=RKdFV|cH>
z7|??mK~gOFEc?4Rc}Lhxo|Et4#k5Qm-lTiYZo-Vvf9)S{2B!DSLuTjr*KgFGAX%f;
zz^o3@up(fKInkaA1@l6{B~5s)DI8c-7^Zh8)>_N49NS_rX^?YF#=VK>^*8c2C+!sw
z7qVUR`l6=oAvCN!$+@aZukSVdvB&tK*=XoF=@aGsO%?4=$^PEM+r_XLJD7mMIygPn
zeZP}Af7kO=kK@COACYWceiajs{zW%Db^D0G19!p3#>-Q
zRZ?*6+&f?H%S#8GONFvq171Yw=x4b7S0bE`Wp`u-<==LVa*(;7$SstleO}=ppY;su
zEQ4Or56NjX`+gm-wQJXctO8x6oL%6w!f$m=zJTPe28khAxas!ohBG}l&1%T1
zYSK1~-97D_Y!abm7+Nq@$s3YI4{^zgqv>iU5+5X2^(riVR11#aYLELYT}}G!R-*+2-(j{p_9z@CMr_c{kQCb
ze*!y~AP?i}7k)46lD+FPg&n}hP1dK}o-Z{934v<9>nfaB{ghmbuCzqXA$2Ij5~}Z%QLI02IPg3k3nQ~c$K&ALiP+l96A0(
zkmyQ>OYDLI@GrbHQ;_O2q0rYaK)7~de_Hd>Ri;DK0rVwUancDcWsj&_=)xjs-lZxr
zZj8w&@as~q!8H6tYrlM3??mzH>yx*PM{Lol&yjdWcW8@kUu!Eu-1iA(_WE%ePgLga
zh5Dzl@GJCIv;TwnkxR9_^3zG0?E5mT8*J=$ua)9n@ZwMXvg}f7bY>
zVbqUd+K-z%id1n7H1gS#{=5qDZjM1^=qV@Yn8ERCWDgxmV}$zKM$Y6HC#IG3zFTj&
z^*90fpIAB8v+ouQ_b#c9R-GsMFftf5$YF+Nv-L
zHO_~$d15a?VVtzAloPx%{%y1o_{6
zO+X#9eiXS5i{(Z3I%%PA58F@6->Jy_rh~nzaeJJ#;79birKUDUhp{~Ul14Mm
z@kzq}jmobT^~B1pHfYxse|N~WoJQnICatqVI6C-X+{mCAT?$B2uNMbR3fz$B0bgSFq2k
zq*$-EJ>Nq|2LRke+Bo$MX*uUC@3*U7KP{15;K
z?<1O4@0enN^Vvi`jq##~DYT~KGFU>qr!KMy#3vXc_ih!XpzT9f-0bozyO!-}BeJ8z
zuhERXK5>KK7pm|m4-A_%Wo4Bl5^D(5j}xUC%`{2+P}2(rf5;RbHY>O<1hP}8r9HBW
zFso+^xLigmkuKU${A(O@r|2EDTo{Gj7)->`So!*XZTGv&)qz$0J8IlfqI?wS>fGsy
zQB)RgWTf+QMs(5L4Xk0WOSOS8m=v@=^zT{b3b%gEiBjKk@K!CkPN}EakbXUDk>R`#
znuwQVwKQI_e-Hk_#Bq-v)Wsb%6|X?8e|W>(B~$h0sNAdpO1DX4!k|0t$)wgiK@sCL
zqy!Ej4gK^Od74(6ea?`PZt!k`097WCdW?(vL-dv&7MIL;xJ_Btk>E?5oSx6zD%e@h#!o1nb-^Xy#87s?oXN|*hep%!Lm=9?V@jQ>2vb|O
zX45N7ZlB$q9F!LLp|p=g3pWWV7e(TM$M<
z{yIYVe{1$H}uOrm!^K>^nbIJ1c@mI}k
zuxiVwizqIHr<76b4|X&xx#^9*Ad>Z+!PYHV=8)WkBud@xG=sa2ney-evv2JXaU)26
ze_J3F_kAC@f@37-lvOeLQcS~m;(k3P%Zg1cA(m?DGf_h&h@I8EP##3uj>jq8xsApn
zTb*yp=7nnU1C{~sEd~sJPv+=)
zm|~|SQ9zx(nWWdnzT+?PTI*b$HzKx+f2(mbP9_%@>WvJ&VsDfQ@z`2ViaW7UdAO*@
zwLaSq1r`N9&o89DA$cTDbN}hfN%fGQtROWOM5k|8C0P8OO&{a&9a8e?G2>R_5Z>oR
zEAF2u=+i{|Y==BlE?*PGs`?=DjX0%|6l+jik)~XT;2hbsgsHKG4qm3~el#o`59M*sl{ZP(AUzy2gmQPBonCY^n?K)Dnn)%{w_tha2wnK2l=cs-IGhf9U$s^tdE~R=7gvy}w%sYPehQr(TTZ-|{0Pid8AbWMUBa
z4G(P{8%IfuR^G2ZGkK7k%?=ZX8Id($1d2eJ$=qG^r2zU5wo%g_s*}I{G
ztcP=Xn6b%=HzjBaBcL1HQGG1Rn;7Ks-A_QC66O=sODirX1iT;xe^d`o7_uv!${dY|
zDP=$w7m`8s>7p7vbRm!@BD_ieBlYknB8cm@9-BcK&D%~r{1;lKt0T6yyRCWIGuG}(
zS*5y2G%t2F?>9H`#`uOibhrqs$CuGl@<uJ8MR^rdjhzTj^*@+`%f!D^l8{^v
zMGe$_mZ7PnFJZNhf0~NQH-y`=Jh;HcEn#hiAs5F;Q9jf
zlm*pa7p{6%yB#Ol2CkqA?jM22BbCxP$u0++vD3ae+I8hIe-H@=q2U26SWL|2Fb?}g
zP(mQcWP(va*?XS%NDf9lBD(x<2WJP9{3scr3fWF+k&mlBzoet8W6xJvn6g7NDwPIk
zt$S7#$$A)>IUMKr#BdGEE?K_aILSe0>C=l$-lI^7E^D59k3+X!2B`_{sVzb_Xxdc9
z%GLE5B!ILFe;IpP>eykxAS~^S>-N%p>w$rHz=kJ#QDYIw;2xDx%|gioi%#U9q(L-p
zoU<@DUA#;}Th!38a$g-NiSWHeU&S{}f*I3r(mZ$l&}nc<
zPhUiYAj9OAXACA?^XSkWC`MFiUpfL5&DeeJ;%)NWcsM?Ib%^%n=&aHr
zokb=8ufHw42}oC*CZ$Mkw^J_iP(;q|41d23`7#l<=Bg2y5`$e{we0AA^Z=mHFggnqHev}l1yY$AX{?+#Z}Hl7B0y|xy!~1E
ze|VcdVfD~#-~P!
z8-~I64O^vaQGJ^C(1O`nvhNyR$9@qle-~%f@*uDvqQHO&#l1F5aCe$s-Zl_<$489W
z<5;6``hlc_1h{A$bep$k{^U!tHSgNl&xc*VOsxAmlH|b
zWZnFvUqeyHbvUn=&{EAB5Ui{jO5Z6QFyHjXpINo0?JNTipT(OcwFeUQHVs}GVvhZuGmq@j3Y|;g>$#ijW
zXQ%N;(+~D6VZlmabk|2*KQh$kf0fd2TNpcokUSny(CDTlifc#mf$fmj-U=UMs4A&Tr9$u9vTn1{SPXcAJ7&+~^_R^x6GxBZ3ZEHkgbv0-e<@IT?sZu>
zzHWX@5&y7DPfvv-vQ92<4FWzf3$y63aOBBevb5I`vTylCz%F>(Ka{z!Qw&#YDG%?Y
zv&4udJZm5?%1IQTNoSso9hc*=?b@NlmyCR~Yx7c$$BYdiuwLHh>mqjGw#J~QyXt4o
zewE8P#DdRen=@QY_-xbOf5&S}ZN*c;+uWcM6tbEe(mJ+A`EJQBwRJ5=SnA;tlYOwr
z@oe}esxAMzrD%2$sBJfNxcIP5B-?4K5X!kI2k1Wuai}ax7QAbw0w`1D2il`zMy_h!
z47S4)Dd+jKB5IoF!#$F3t(ke!WwiE1naAlTFD7!tDO!`(e&R65r0r{Jemh3>X?nNGc9MBer53S8ex}jKQ7u62QFNLu>lTVc
zS}kRkZ8TcSiNK=9?ZE4kT^~}|1srqkmscHYrV5JJ8BTLp@!%nven616znrg4{i(q8
z>1G6fNIhnwP9EvAe@+b6r`4$@kSy?Gn*o9XKQav(9uP{_3oJ%gAvnA{YMYUNciHlR
zA&kB3tIsdRaEGPt&g`E!b#h|?bXJ8?4l_SkadBL`f@j_1jsk<2a(tt-)nw^*;x)GB&e@U@3@ZWB
z!AfOlu+_n-qw0LegSMX}mbfED+KK%e*2gJTX^_5|;NikN@h@T1XYq%{_M#Gf?^VNE
z+kIoKQ^X}UKao464)ZWh
zkY>Tqfh+#jW21;O@oYt(t^mdkl1*rM3-`R7>itB7I&H%K7k9mdvEH_KHLQ!4X?rzX
z+Y=9Q0*&`x4tkT8iW_LrJpwpd(#YQ{?L9HGwk
zDaHljr`I;JLfZCN(1Uw;Jr{iXQPkaB{
ze-<_@SQ~BYEFn||1mX^;?BixmnYVcGr8+B`#_-}I^Vvrn&Vuy~prf$V(LEe7?(C6+
z23vZVDeDZryn9lW=J;P*w3Sm#2!fWlH^}cl<|^~YLom8^%EZ_b)Id74@Ym&a
znUr)V-IfZn%b%NNQs^0uU@3CN5Mpd;;40hUf0_xH
zvw;GG8YogEN<0Is8_!Y41abc337|)`_-Axa>$Cdk(=<=JF
zTP~0^%B#$s3ylhXg-VZ#R)P4<;^N+_VfrYXFox?KlQA?}1nJwhPM|Lsasj`dFyn}r
zBsG2BAM@+%*&dS`=JhDxJ7941e-mU#9Jz+vcuQOchxH^;RammZ@+;Ea@5lywdH%MI
zqfX)K$YPV~h=J}CP}d2RazdADl_#ii}gqI5%`>Byp_|r
zS;f8F^*`6`IrcW)qcZ_CM0DP{>i5g0T$9{|gnvrmVNZ$uSh&8ipp=DECf<)GNA{wHgC@csrfkbhByfXs|h
z2J03dN26x`F0A8wgFoUx>ehVb_of9ZDvrGJuwJCEiHwZSXy{F7jn_NB_9QJ}YG@68
zhA_tl1@EsQLP@9eGyTRP%7CZ=*t18n_6tJ1u~;Ad&ErlQJgjhLPdnM4pKLzLOxQfp
z1FeWHkEDnM`6)?Vjyf4N(SIr!
zWi`C*yiZ*LY*?MziTJ-DwjEN=33a+jQ2N(lFK@BO@H^r7CdbayM%}d@bth%UV4SoJ
z+CkUg-TH@JaD94X7cMhW7hw0E`DJxrUpw%vrRLd3So()|K_~b)Lx0q8F$&V|yC2Q-
zazSZ&sM_Q`X1h6<{7l!!9z1h{$`2J(EgCx>31mkn{w*yq0?^UZN!`#yV=vsP*6S=M
zHqIU2XN?UaNg)C7c%!~Uy3{z5U9ZRvkG=#EXRqht_H@NtWK8f>>m;Np`Od15Y)p0(
z)QFAjaaZhK^IMjgv43)mkp12cQ^=H!p71-aVYL7k)pLmuD62PbAyTG%T|ZZ32r1>y
zCRa*pO8NfUY%!oD!!t#L3<{TZ4G>hlk9z{Lls6b2r7PikHdaaY??9lT-o_^Hp&vSo
z%Ni7touUnSaXu7ai*-g7N%!EBRU!=ujNDgo+pbFkP<1F0m4BJKSZPZ=Q&&Gi5mlt$
z?^+e=toomR>tH684>E|hPseR
z2c1UF7HIU${C~KPrlFj%#gwPEbaE5Pivtu^$KK%{{|=3M!b}v0qqe{7;a{p}KAe|<`1vTc|;;XGa9t*Key)tdieP^?GM4t8!fOB2G!_)p?$CtcQ16PFn`!sXM$G8}Acgan#SPcm%^2ED5|_OowSLiKT%Adru8StPTyR-;W
z#NbnvF=Zc~#Ll
zmKi#bt2Z?ljTKSu=drrb)y=wN1{8~xrT%yjhNK4|N99Y*C|J3t+)TjX5gL`ZeJUpP
zcz;7vc-S^#jI;008lav6Gf}ogA%KFB94gmPo-l;=Q5tHz_;73bZ6$;n?41Pl!>J!~
zPH&O(NEP9rcF*xAhgAdm-S)EO$CSdYROnB&3v+g6E^@D7-QtIa_2%yl_ffIL1cR7|
z_6^xb5b`Ov-*fCU#QZ9j{)yEr2;F2^)PFqeOCp2W*kfk0m(p(#c-Nh!9gsK(RQ9+s
z$wtk!ermtEVHB!+7v)p%P;>52u^aDTY8
z6|C%unpGnI(hopYxN=7=yKK4%hwnAi#Hou|2D^-8H~jSIU2o}TNj~w!FG^R2<4Kgs
zm}`hkOzU0v;+l;u^y?nttr}b{p`Y$+KN-U|2QXue8;)IW;nSl4OuAnbcjL
zAbph(rY*z|EocgD#3w4$R(Oz7i+@2V=7^}uu8P*de9PBCadsUeA3fVU@2q*EKT-=oVAQE9p&;LXz9+UZi9i~W9
z>}hR#s&1F*JRO{!VY7f-A!(Z671@>WXtoFE#rf!){+kCzj`2KC(?G#`0zmwG>3MkLL=>Gd10KTYzQx_$l!J
zW9uBYL{Su7TefZ6wr$(CZGYRgYn5%=wr$&fx3gdC{6iYa7@5g5B{Oj>6!thJ>HdVw
zIQ(8+catPGUEojQNKA1wfsZm$(Z&4KZuYypbzbs9@dK13kspXip-vvRHYsFGlQ}as=v@tY>VK;F(s8P%G~Hpq
zl(~goc|tYUpvtvf3`{
zNjmE)PyJS8Ck(zaH=SEc4Gl}#=Pv9Yq_ZO~l4fwm$i~*CtEI^B(E>_%-A#I~keZe_
zv4v<1lJeK7GB#tqxwDtOYxLFFjl
zU)#sLGj>SuLldRCVne~2y-
zz1)f1qrOH}s()^$$Xk3aS*5mPBFU%?ZZG#xmAMq%*eOa3m9!etAlC4V$0zk+wdAXUqyD(2WG29>_y
z1@&Blm$lfkPp1HlQfnF(@3Yaw9BK`g7J|;Mo7o3zBpI#k($ohK_1l5YTt%i8GMJ|((MUuOjrg&nN-+YWvl6qZOf
zW`JzE?ZyqZlq=UG
znty*3I;3YzV>whOPo=X7I=`_$XK(AzmXtq$0*+)$>HLQ@OV#rBT~KQs)&$Ukvk7Zm
zX3SF(xC02^D3$FOWb#_s_BwGE_WWer{XpXp@*GcHc93qd|Mac;+ECD12k@sm970WX
zQ2-^CapldB4Qv2L^g}3JJ6NWiRtbePj(^!w=}V3wIPVvvE${>7=Mls2973(vlkAqS
zCQ|3rJGIHUJ-#dh<8@t5vPmBi^0L*yWv1l@5)!!9_^A?fNl43pPCfTX0*@2&-x
z$4wy=D~+qAG}Ph10{9dZxyt%k%4-0a9Nb;D0It&H8?n$f^VDo}5~*gIy^^jw7=Lh!
zUeJ4V&G;Tx7C@XeJKVt*VS)Fg%b_(|4UnfVKVt=$fR}Cb9WAUn3}Je2EEUHAcZ3yb
zBV#yK_RB)xnV38#|+{$<+Ja6w0#9!nLbv_Mrcufl!JUi
zK;f%x>iVxMvH;8m!48kMg^hRDL$s^6EMAAjE{ytB(VbLU~b
z=^2nV$CrI9X@2{J_TpOJ-GLcUu>_9x;YHnf2SEbHSG!zY?SJBRe8Zft?fmK>e+jaS
zYVRb@znHh*d5{KFH&OMdmw&=?MbhUsAeD^__3X-q(=H5W+erY(qDt4FGEFvhxHlk>
zxT}zb6$-o6mr!3VM4XnwNHLuON}5a8g};bmLiOR;tK%m^uWsV93huGq%^>zAo>9bJ
zTi2~1zp&u>uI?(K4%Tpy!FQQ51JVdKMO+pA1Fn{
zB%QGB-v-BNF&myA@|S_wVt`oKINaX;
zMv{|UJ8SIDmG)+b9eTvLr<(EB5LBOPa3~8!jYgx;y`qx*@;y}oMsC=F{O*_A9p3eC
z-G?!4eg;c<0Dm9LEYIHlsC(h+;a8QSm<^cFK~|yM+uX=xSnHkul;^SF^BIk!N^7`<
zK%@QLb0SuA$CPf3^>wTZ6j7s|ZjO)j)1A0~23gCtVIKc(mAwO32;l2muBz&`50!px
zy9}y-U`1>ySsBKJ%xqm*ZhEAWWUCWB#>RD9UE8RcDSvZNjJ44#j=(2Xdpf~V`x^9$
zO!ljlMPaNLXmJ~8ENXMA)^OLW`%B?dI*NQ-`uQ;eBEUZ$VHsaY9+*n+kjH$5Kt#e#
zX+m#6U0AJZWr1Ly4!@(qBsReq)Y9h;T~cqf^Uk-GdwXe}4_Z(G0+xOayhLvmk^9dG
zT|MOdJAcKKos~Jo?dQ|F(98AunMG=ipF*x_Hr)>~!3YlM;n42Xp3G3m>PKbxdXkq2
zgUp(i(2iknVc8=WyZ)2^K_!?B=^?q(3F8Q!~$JWd`^#0vPbLtx||BD}=N)
zVCmBYc!1#Uf`TZuUDwrCoqt~z8IPhx=_;CCt5sS_CIIs?2LrQq
zEuS%Ux`YE@`{vm=eEYRt*2e}6+
z+KS>;KCO4k_>;nZ6e!DbJ#@O3B@EU-19*Tr=k^!C$vC~%u$RjbbeWBOwBZYO2Sb?$
z@qfD|xp~u~X+(Rw-rG!bHc2H2tR^ZegL4fMXVtLi5uEYGC{jp1hqe~*55UyTGSCcD
zhqgyS#8?q+l$W=%{OQ!xj<1zr9mWs2QxoNWl4z>VEZPi!cA2uqh7ij_FqC_
z=3Ya)6$qNkt^h`tU>3ZMNb2SNbMpEjfPWv%ROXdYcGZ4g>{e%cd(
zf#Dd92c1L3FRF9>yrW}}=gW_z*8H|rx(?Ak2J6xIVI0Ae2GX+HfcN9Vg5svt?^JpzztVgoZQieyqG3q<_gq
zmtJbq{`N=B9aT~K(ps=Bk2;UYPU(F3Jhe5cZONfA8T?M1e=(?f2~UV0V_}6~F;J^Q
zBx6>WYIWa=V4462ML-B;#+CFazhk+{-_YGLoER}X&hAS%{E8;!0A<#>11tNcpF1~P
zEQJ%xJse?plIvlU?=a~63^&Yf%YPY@@P?{z(t17h+0T*^b~vtyT81LxjzK4OkXrY2
z*U{OpbrMp#)@zs-HZOePmj_K2wW&(yDkcb+qS$|ujBLm;$
zZ3NB}Fg(PhbSysf3kyX!eSiveC(WO>-E#|jNiU6$QqV^Gj4_{~!Hu;NHTSki7`oKm
z9LI9xO1u_eevj)deHkvfmWvafZ5i$J))QBY!8|sDK1W
z$oSW~ujXYrQDq#4>Zt1jWv7t@`HDiwvd(hIvR~NJfWrNUWxeO8ynn?%Tao9f&S1Qn
z{g8A4sg*!>DA}5#A<$Cg0JkvEBY$3?28i7X6=M7AzShBNG#Q79fC*B%lHahj=uCDE
z>WvY+{SB!z3!4OVkbnVv8e`8K_zSkrUGdPpRbqvljI$RlEm-9fw)I*K<8~w!xWb&pTeJ=DB;&BkYY3%Ldf3EV
zr%k_oN+;CjgrH~HSqR1G2hPS$qcADp*P085P*90Ikc7QiF(~>jcqY>7LM6VD(raWL
zJ_mJAh%iF6eZ4l7RxKmQLL&)-AJ3tt~`rg6sm6Sm1e7z?r?gGAX!p
zJ@oovP4GJ=h<{PX+sHc7QU+qn?JU(s!1Fs&5h84As+^waZ
z`4b81NPn&7oVYRfh_19lK+_J06z`_o00^+^F-8#8`DLb$^h0b2Nhq&z@CDE0)_Ht6
z33KOy-wYs;m|s;Y(0nRT+T^|xlM%`?n$DcXM>q9x3+I~x;UF$a>sab&$9M9gpBQ05
z_}CUpDDT@dcS}Wf10PK;8IMSfEpwUt%W5QTLx1ns4^Cc!dN{~Bj1u7Bx0<{v+eXwc
zl?AP^05?dhMYAC~>d&CPiS
zpG>#{GG&RFSH#bUHXc_}N!{}*k?bzzqovrgA;b>gx{7!?H90XRqMOV~@Fm}bA3<=e
zA%DW9UgdCxg6cRaR-RoODebGXX`eeR876R&y@t+cQm{UrQDcFk5s`ahfO2ASISh4b
z%VBi1P;zP{UsKUYN3{lG*uCb{%zd94ane8#N|#LJ>|waPzF8i!cPlR%&L@j#kHcw!b;CK4&B?)tqz_3|9YmpV2?mp|#_qQJM@9P(l@I)9X(GWMM(
zpT(hXf0FctBPqW*CyCp&4bWX87!Y8@l?YfAIShX`!{h4sIg-O_pv=+=qU$ctPk+q=
z=aYMmc$zy&3ug6I5_GExq&`ZeMB?MAYu4a)_KMmYfm(`YyDjZ#>2h<6G(x|*j+`#`
zUutyH{f;3T;9zZ|%H;xnbN<3Emdw1n^h7Zi0*HjhK6J`X?)k|_)NeO(Vg)t=bXkPd
zF|$4Y)qJ#fa*M*WE#3k_m399E^?&!dOYL1okt-#Oi9|@A#mnzCUk31qlHRCMq8l0D
zJjLWGN}(&=3fg03!CKw}-6G<9>Ll=OTFF40FYoIb?6l+;Ubie({GCj;Z@ED1YUx653-6
zG?=}=fn{=FhL)7i)s^DtszX41bv)e$1!OWe>;!jx1dir2v28fl^ZeVJ!_|8_Zjz0M
zjpAIk=}NlNR*lWUw|Iy2z2|w)R3)cVcU2V_s&8LI6qk#AF#=s>jTK{=Ax$wNT8dLFX{o818^s{0*T2nT
zw=ufh)Jp0!b}Q~+_U!1Wsp{pq?@`=>;U4(hPf>@^PM@_$F5uZZ$oVTU)0b$i{-48CIl0zdPLt>|)y|X=Wt9Tr`V|slTqf5KY!!yR&moXZ?TFqOpSUq
zAo$9hJ_m(wAv{u9985qDE84I1`552fwHz*trzKH>tdUf1U#=fmUD#Mr^>pVBMQ4qC
zpe+&WUd%cd9L-7nPxv6OG|$rNVGPBX0UFa)HI)FHoN8DdF^tFshl&YfdD(D+5>;v~
zesp0NCTJLK8GpDUTBK`kR4^D-R<^prDphcSH>yeNbhGbIfYNIk(@=2<|H0M}VF-3O
zg8prJl-)viws~dk=#U@~$H`);B^FBQvjR!e$E5oH{L1WXLJFOG6}hv0k-hY$C23mC
z<;G&qT4eE&!e34mnmo~gH!G*qaHSbfwd(PyWXeGES$}MP<^>AZY`hT!ZiTzS<=vgj
zI2|CTv1NxNoqDba*6CTr7C5Cq2aw~XHaVDL8W<|1hS5Bs(p-~PdN|*BKkcjAZOAiS
zO(`c9Zc@X4^9aSMgpZ#leY2)a?w04gG6$Ch>A^1D^xLT&$s4ntLr^lw7+d|r5;&#u
za)G@OMSpk_Stf0_IfM>49`*mV@
zFR#pb3bh$vUjo8hPk()#kW%>}a8h_Q4jO-1(=Ip~L_5e#)yN|-5TI-Ki
zEB9?z$XRPRFt(*13Hjz?$giX#-$J1XA>RER^U^U8XlU}P7m
zIa=w8=&KJFJZD8^|B?+8alqdICX;<-6C&JGB|gXV%tltM8djSx%nyeF*jm!}N&G1yXHYG=%hlDZ7LWN
z0-TE>X|F}4CF_25kc)yT)AHoJk}CnVWw2OxEWP2>b*sIANx19^ABNe7E2>6`UeuvK
z!=nDxn(JSU$jL`%Qt*5|W#EPs&lLMJfhtKXojKQ+$mg7|ctbkVI>tDTIx5|-KYztn
z1z!^_q5&T2^LA@KU#4hub14nxVg3#vP#a=NYE|@M}lyR
zINkzcG)t@DV>${CG)B215pb6J6iYSFTZq14jR~m>NNyMW=#~s1GNOyLd0my5%e{P;
zIXLCt!42uI?_{HG)NC_Md3Cxm`F{@ZQBYt`Qr#CRXLiQymFwYU2O2lB2#9(|7^5J*
zFv<53G6P7rO*a`6LhmDs!
z&OZ;0%J;l%vZ5A)nT@}9!)rezS$5rHgYLphfmgscY_pkkQuz3nn?IbI8b&>wh!y-ntvK1yNNr
z5h=o1{FCbm@^h1Ec~1QES~G`sP7nDufdhSoq875pAW@918`Vl@VNnIs25-sahd0YY
zC0G55C2<43B9(FW#Qd*RKxi(dOisGdRMEMx`RCV<_l<~_M(IiS(`~(nqHzapkXi+L
z@o4?shK&fB4IDCXRex$2Lo~-~WaIi;I>4?|(BpZ=^(cc@ya1iC!1D&nkib4AF((MbJl6iTew6Y
z;-U7Y+wMRZ3;QoUUBY~!DaZH?db>!(j~K_fFaOy>DvJo75<+&}W^>si=c`=(^E$nY
z)OmzwVMdeJcz^fJ1yK}Z2g2RrTBJO=>~_jFbx6D2x2TPY86JlBdX5zKE#Eq^aG0wl
z{1N{u+0t$cq;&{$cb9IA_c&hwxnSFmF4-Zh-y3oh{i2RQ-7rB7in0`&-jOEXQJ}nZMFt&
zg`d_d$TL(C8vj`C%Brvhjto3S=dvnG(U6bNN^GL70h9(xB@r2pk!~@#%KKXk>cK{<
zKA0=h-Xe9d(F0_?K)KhCuUi(IT(R1JT`Mk~bbYC^c0g|++un67^O^lj9u7lm1*7d)
zYzPE(R)1|>94Gq-0SFN#JZ(2EMfFgmB%Az^UxY!9(?*vmvO#GUlSzCc0Je$(*{+w;
z%XLihXhC}k^;I|u6e32*(fhWTkNn9(3X!IWIc`p`0PnM~IVXR&@N<7r&sLh+0H}bU
zWoI)n+Zq!(U7}9y{8J{I&DXdTX7sRPRN+;-LVqyvpa&%?aJ=%zgoCt!L(Nq6WjO$n
z?n`0HjI^8n{Tm#kOw5Z!Lo>RThM#jWNm{D3%CJ{^ffq%L4(FRp*k}%ute~|Wq+KyqN3dYGY>S#AZFQC?&3E(jrsC+WCknUs}
z4wwC?#gLYu;Z|VYCPw(QgdtuIB$16VujCe{h-q)?{oA|}KFF{!n(wFGP`p-zl?E#O
zEDpV!4f@fCQG(|rY>*XU5&x%{&HHE3V}BDFTq+VXuF?wI9`$`k7-x(DB}*UM*S9a5
zqDghE-dRxNn~+nMxD6kbKo!%c14@um9Mos~B`Zft&`o^k&A`3`?CAFVrGx6|ZP!n%
zeF`u-Wd!XcogcataZ#C4(myHJmu5FVl;6q=8p@Uz0173^wCD#q1N{9wiN>4zB7c6(
z<{XLi-T6Vth*N;9BFrcm{ZL%4H?lJ#S-?7|biqtQ?P|`A3{+9A&<~e6}F3^P_Jk4@!0qQodmU^K-?<5FCPU@Xh
zTZ=R3;q=0Ga|1Ms*S|F^M}M;@tajG2RQBDw=JJjj(<3(aV9@-Z8Tm5=cy6k4Z+rak
z1D2ntYg2D4Easd^pcXalaA{eb<^YQHsf=vzkAD`cR19lUYuiw|Iz~?xBv;FGjNS;p
z%$0K@O}1N(JAHEm40OY#Bs=Ly*#f#Lkangu$p~U4CCOFQ_zHxeoPWmIL6KABlY7f6
zOVgv}ZsB+8#{4uy3`o*LQn5I_?)3LPNR?)9i=EE5<0HGeEw?rh31nMCWmL`zqYbkZ
zJjn@aMgU`nm;PT!=`mdnj7BYmG~;w(^7I{Z&e;aV8WrJpUbbb@r=3qaD|95^D;2T@
zr!orVBr<`S^KD|9+<&e5-|!6R9nob6%K5FLU$;Gl|3z#DKz<{;B$?JSSk?m|rEL0I
z=aYbfXT`;Mb?uf{6-`%rxep}g7&(K@l+2*<-ip(%xx>%vYZhfoE2(0nxZtI>u2~x1
zL~zVeqM9nhd+RLKU?#$V3IZCQSAPiB_P*F4cwR|2RMrJ7
zFvkpuG*pBsy%yexiv99#B~?%n#k4xsfi07-65tY2fZDiH-RFiL
z21`umKL0ye-gFL6w?}=*fEEFRR$)PS!*se*tU|=XzVAg5vxa$;(OK+F@iN!}v|$$g
zgDF88*iA3V?SE@Wk07tifOe=txH~OlP(?f{PEka9Zh|A(XleJJ?<(=8
z^ie(L7=bsa#?)QM9O?Oncc?rSC84@GkNt|>PMhrr(9}78F7tpNxZ0|3W4s?V7W9W&
zOqe$j%2fP8#|kCC%7Q09{V1sMeBcAZR72qZAGLm~Ab*VClWwT=kC;k#-P>uBlP?K+
z`PhGdPM-Q~Fp&-nrvfcR8ExBy4jB#{6487EHjr5AK35Prx|#!haLQP@tR%!6S_R;;vx#kOFVIhhVo(
z?wq5_P?I&eb2b3K=MOps^r#^OK!yBWB5XI>kck}`elpHP-Cc%Fp4NqcKd8GUUw*#J
zdpT6HPWYYLwd{dgqE3csr_lL3ARiw5_t%5IdM}l3AQRN%!VwmG9$Q}0qxeU(9ZW?N
z#eWuOv#~wDIAXhfpv}J+4r$6&7AZWy`3m6{=1&v!kCWE#$Jy4Z=sGtG=rrr1
zqAhZ|;}_nv3ua;?f%dhgvR(|;o%c`*|-euhF&!qM0Ul-7cH0O28GjWK;fcEfX(
z_GJ>9DOLCQ(<@7!tE2m-ROPWj;}rwmM!!7(t>%m0vc5-_A^2rma_FibV4xYCT_+h9
zrIiWIDL7ZYC3SKNoN0LW%H4BkUG+A#V3nm?IEqer-(o27gZDl_LpR9C%6}d|*6v02
zL@IaINiI-MK!G+x&<+&P6&#*B7<~J24U4_7aKY@)+qLAZDH)Ke8jr_nN~f>&>3p<*
z8efY;uoJXU;Qh!}j^hT^Ss;=6Uxqcc{ViI?wVDQE37%CEj3GWNhyfwBd!bU)Aajug
zHgB1DBepoG`2;B{f*UTo_kYnjTIRrrWV-*#ygtTY4{q`O_JKV;kwmMg@C}fM!pCCz
zOkG;krOhk|A!mB81LaPg<~+c)raj_Fp1Vk5l;@S4C!iGe_g~Bk`q7k3p+f*)n^+qt
zOVn^O-{!&>z(i9f-48lOK-!G!8B_K8LHO?$GlM6QbQfM
z4!Y=M3K1KEU~t(^1Mlg99~+cbPFF#ovTxm0j{0ODR})0s=(0aOMAU#zbKHZ5K1uFu
z;PI`2X=!$3SQxBqhJSKLp(i_*Bc@e(W))(a9vG~+Dmza%toU3_osDt~=
zcvkFEJ6EtkNP|$Gue&*#lDB!45Duu$a>FYfsoZyZM2$zs^?wM>{^x>#Fup=ko;I~t
zCQF`!UT<8q>{1qx%EzD>ik_Z11$gD2o|JX=I9S@zbF-1WEne765*NB4edC|YFwK@}
zHzJ-x$&ni6hVl?i3T(2FTLiD9){`PX`Br34hwW1KhctlQg@Q*l+R)yW&N*k4&79-g
zcBC%2|0Pdl+<%=}!x$NJh0-sm@L8wq{dvYy^$kBM+Y2u{u%N`sG4X*1DMXOMWWBYz
zp>551|M^lDcYj}@5hi1HT`3HMMd9x7*9S7@%ne6Q_f1|W?z*YoZOi|)kAlFsv$5M-
zmPeg>=8I@9Q(6y|gcyPu*Wu9LD-W>eOB(iAAB*z>*?(;3j6{=`L88;w^H{8|P&HeH
z8AZ^hn{zpIhcvoCK|%ARzo3@MTMk#TsNraP&JmyW2FT1bq%rw69#bsbu!gKvffX&H
zx1|e^*H&SZTYGZNS=|H_wV|>AXf~(lky2)yK>^ny4Kidj`iGyxL;oW6^q0TRf)T~c
zc`B+C7=K!zdsqpRol<;DoD-so>WhKzwo(hnK!{t=4lR{TKQE?Lu0rQFXMJ=Cw-8tV)7@LI%AY9k1FDTdW5{zmNXX@W1uJ~8$Oa7}T`Vc@yp1Soc
zaugVM8}D0wvxG(6(I6?M=GOvfS8OAC57y_s5A!&p5dp1YWgFPUr0T9%QB>B~7|PR(
zD}U^lAfeIkDrQGXY_w1osoRFkS21o#Z{pUcfN*lC_>Y3X!mYVE#?t^bSKG>(Cet`i6_`jDMD+`0
zm`50;X}E*uh59{fgvAUv2DD$3(4*s&rhm}YE87^Va`jD;%6LHi!jCe9?i?9
zi1KrjkErFFg?p>}rkGT|@do_AOj4qca-v16b4xGPxCOTRaFY+EV?eHSW1FcF;F(;7
zBOTn`x2Fpwv0GWnI{-e;N+-R@hG#ZjnIbtR?#4-9Q#Dpb3P@y#VM=o;isS^Wy?@z&
z(>|(NMyz6FPBenBdMI=NDCz(|6new|8;}F$P{9faUGiP;DMnS+7aXLEEff!yVNMrL
zQ09&2QuU@)&`p)}CdD4j5D&V3m?kX_jArP{v1E3d=U|6wBMxGKAd~LnhsZ2SgqCWi
zSTK_lk4EPuApr}or&^HCq*Af_F@NiUJx0HK?x&ohf!0=d-0CI;A086GJfq5j38KwW
zy4io<=a?GDDB?)@SRyZV1Mhs9aKzVmmU8hqasud~A!$B0&_U2#?YCSCfY>62#k9AH
zrLNRMM7NWG)U0#A2PV>ZjVDb>Q(9=rPvM+i$4t3SZSonRUUdi6*hcZ4F9}$-Lj4M=zbMWN>`Unxf}K3dC{jmm^Jo>jFI&
zZ&n80g}(iei9J`jF~u&uQGgCJkZMt|A9Kn|)0Rd#rP`Tyy4uE}q<mM$EG{bzyU6a8ky
zSHK4~pyrnzKTB|h#uTV66Em#yq865TQj^yB=)D(r7m=?OBBZPpwtog4G&(M#QY~)%
zC=OCpC)jrjS#|pT=tPFQebw-I_i3fx=&x-V{*$(K>&=<~3uLhfovSqfrn&zz56#E6
zcpWdLEsN50m0kH#8$l9_fG1_7{*dzZa{=`VZZ_&H@n#H5{IQC&{HV<;;2(lETC6(s
zJ^A32sN5k1_Esag>wj6#h9=k$9&^I`xp@z4NOkYt_EUGl
zR5?bVtW2m-HG4<$IpYnEY(b
zwL`Y~=;h{yIcUW8n3`oQ85C01$mU6
zBwUsZJ^Rn*8X}9JIX(Jbl1-_Er~EN5v9FRcXIJ_xV5p?ov8RVd``5z-PalpwhIHq&
za4g6H+hD7z^;|LhHsI#I5x35l!4#~fGV(8OlTh)0N?a=Renk=}NKwnoQA3q|4(4Lk
zBvJN30Xzgt_W$A#Abax{Fjl^0}64F^SKq2ogDDLU0O$t=d|}Ah6>`#^f(5{
zvk%`1%Bl(!&+pmm6O;h!DFkbne1v&YDiXYT
z?D{_Y4}T#r)S`a#e7mC6q2nhK@PvA0;_(q)RyLJO{45ecOczil|Ee~C@RYtk-}A!t
zH12w<|8Q=w_jc(34t*J2fs;SeVghlKcjwX{=a=3_fGPUqn)WuW0zFFo{#|pSho^=Z
zKTE65f$jkismn&W<8Eg#wsX1)>4?%m;h+sakblI`D!EEqW~(rvW8q0Y{VEJ$+{;fO
zI2bT0sr1ZgQMOS^nF|x^LYsn}Od9|qLOO$I1{)6|85!*!1wewQ5^F_|KKZ!L(W!!v
zo+isVxPj>ikimRsS12C&Tg8}DCugpn-pMwa8chYH7T`lZ71rl@8X=euFf!)3OG~S>OrFg(#qRj&Q}LZaNy=^0
z$#?^%{hNPJDK~2SaiC3AB}KVCG4XzTcYoOZUP71UmH{_UXSN6>%_l#C?KGVvikDd*
zRkr+LUr!fs$$pjKXtGAg;O)7DLjXn&E1cU&qkv1Rb*iH*W%
zm`>bVbh5g2)CLDd)_nxkLMLw3h=0hurbPh`r>ieANQ-)%Kqc-{N`s%E_eiy?po
z5@I3RM=0U{C!t@66C8ytr$O;@Jq&4l3O1e=)u>F+%Ib-S#
zOQt@g)3xJj69{8_!UNUhPZA)l
z*Dy&bUgGQUqrGU66HL8nd6UzeR<^O)ISg9)d=rHn=lFrcL6}LLh7=S_YKL6qsev_6zUJL;dD+P<
zonB2f3!P-gM`z&D=9IY@I=zjxp`vj`%~{ud;S;w70uF-{$*&#&)5>bg7%XaTk>EoD-
zJ7#2V8-4M`ZbR#7@qe2wi=M0pBWr@H`q{ImM=fz7ts@6@e~-{$sUjQxH#Zc(1#)Ow
zgs1fdZ|SF8vT0PROunMbN*v6}NyaZpPYBBO!9u(<|J3}Z9mP0YQEh}5dPsP%3eLv
zJ;K}|!PGoVb&Hl2@kdiNJHH<9^&`HsbfR~{8K9BCIa)52H#MH*zrTvkRz88_vGQf8
zQb(_?c@)e^S$|8wRMeK1Y;a%BYi5x5+TsoukFh>Zji^n6@XIWJ@r|Ql>Cr3i01>^9
zIQN&wR-Stm-=LVVDvL{C>ih{Iec~;P7A8AvIMK-K!_;MatWmTvW4&N3W5UXm=umH2
zdN{23J)KSExJZW_cKYQbU3d`Zl?Nv~8ptVx*b%uf4B3
z?L$2Y<;O;ZRyHNUV!h%$cyfT*PT?|Ch$JB*WeVRB1C{#Zz&dx28kDYyk
zbk5Q|{bsHIZe(00`Zb}jM?Q<+kI_fQ&>^l$l7BioXT_7!_>xz6r;i*!8DK
zhVOJlb4jm`3L}w;ZuYR6G#yFtuFmAVaK$@MT!qV2DvDq*nhJbVmd_n>Q?4Pdx;(j{
z9o4-(=sR*sgW;g^&=|^h`Z%D-qG;HH2Ijw`zMgOOyl8edm%12R`h!DdxyihV;Q+5y
z{C~e*(~A$JSVSWp(?Y=);D%=V8}+1$&D_`+9zVXtDiI5=g?&X-96+-*I`|;LT@oa?
z4estTXmAVet^*A2KEXA(dw}5X5FCO#BuKCXdwl=@E#B_!?yXa&yB59ZzNf3M?AVQ&
zM&NIL7{8Us-iMi)Ma7}7D|G(^>Y=u316-oB-lJt1`J-qB;
zp3Gd=ntT#c9)h{pQ$6jXSf@)opOn`3+RPu=QKp!lE#1kjhakcbNkvZz!f4_9ewH5mB}cv7A=JQGnlw0W%%c8&JdSd1o0wym{L
zJ(kV;zO~42zdH~}n=Ypq0w-5RM_;y?pqPSKiipdX9xYnoebu1)$i48GAF|%tE4jO>VDxBvHWAGkoebq4!p9k)a}7VgqC5he1okMKJMD_Z
zPbp+CUz5HBg-Ma#TL%>=Qg#fBFW1ZpAa+)14C#?Qt!E%w%ie0KPB7VYa82ZQDk`}f$uD&~$pg2q_z6Bu==lIZypZGz$ax0dLTvwn-5jVG(B^!w
z>UP^kO9@!4%u+?T%gGfkzSf!^xbgU2N)(IJ7md97~TYtZwtO$Fq9`5leVlTgoV7t5u+^
znCVA*P5>DP--7tN$1VRho~t^=Uq;}Q$K-w3cvtibYoMXRr)3nkKIVrnwHMXeVHoqJ
z8P2@Wvf%e*9c~h&e&tEp7}Y`58v4yVa|ywc4Uo~De_Cc%zi%b(b+bkgI_9o2#h8C#
zrPrDEx?8BPTqKPnlF@b?6Y!IGs>p4ugq2~ut0k47MDE4iUOtZuvH2#LkiVDU#l95i
zsQiVq5tTQoklBBH-11IePbLU{SRYka=>=^kP)Hb_xJH4tuc*G}QOi;`|G+DLPD2Uk
z42HD8{b-WUj@4N=JMlxqQjaLs_B1nYq~C>;E?m{N5^?wXxi;jjb^qbLmeDG;SJQfk
zN!G7d%_-p#S!{buUi7qG8d1C2J6bV~AmkeqJl``@X%Tjq0=ydv6KkOP@&#DK*Si`2
z3rm!0^ulaeO~PB!xG3M;fwH1U<
zj=5#J!IfZ6N2>kVgo_BX6!%+Yb_&orw~?myTe9Ov7#m>X?e8HgH{QDFSM+e)26vJ2
zNS#NvfzlcuXj@Zts4q#deFL7&uane@upMg$GA4~^@p4O>t2zi4#ln$bT_R%bIJ;$Y
z?p0A#yS2i1`AUUe&;ccqRT2BplAqerP8M88EYDkR}p#xxC)wd$)C9=jYluN49e6a^^!gWLPadrE>M
zcMcdeE^fhl&1DLMi6#F8FbZm5fyKdivwGyupExkz-4V;vpde%zmmgd59?637??KGqV?#b~V4H3<*559M$Esy|~YKk;%Qe;=G^C0tXC%5`c*3@>6N?hHDBk|?DB!kWrxu(1vJr6
z?SdEK<%W54cy7-3#6(R*#TSyqlsUC2owTQ$4kVC?s-@U$Vd?~j`nSd-h38KBXQ?|%
z=*P>xw$f1s&jC}!M^lJkGzCiJz#@UdJn@s%yX|I}Rh1zi?F?bA!jx_Pz14T|uFxP8
zoG89EZ!@-jh4gP>tm$*FN_{B)NLT7x4x(XgMaXh;otoNA*a3a&*Jvj{4;GHNA_OCS
z8Lu6yf7i`P$f7ltvC{W{x@5Vqg8s1fA6b4B2#+GQ@
zpNXE(gI{UukM^uMdyIly3W{j*z72A)coM2-zv|m4GknAYy&yC(By0Dwy?KHI>8A=e
zD*i?WOptyX!3ys<^Hr(nQClsKp~hY#XFDM4AnfN&8MPlqApF?)1a?9=r6anrFVhlU
z!|0EeG|$m4{dfzRcpNG8&|ixU1*Y~3+Y>@SgVbR0i&KPQ!$y~n)raWufiJhlpIp9U
z_7dzf96|g3to9fg=;YaLBBK$5V;Efd{?rW;BF@+v=XOZXluq&=Sqj-FO=%4;dNLZf
z^pg*S;o$grxbRB<(KR1uFwuP0rvpaHMZf#PXB7j<2k+r?+JL~9!O>pFJu2Um
zxP~d?n-4628e~)AA!?sGcRQFYre{e!%TjBw(+U8?V}p
z9Q;_c_RQE+UH73JWdmX7>g8Y+=u&7$wPX9kq%QlW3-n#R`#8mVbJ!J$ui1>r&v`tsuvat
z-YCec97A5!4H0+H)EBavQexVDyI!jy>RgiZMXTK)y8m7*KZs`MzC$T@OA^$UA^(_L
z&TkjDjHiqX14yeSws8e{GZ3Gix$3nx@i7zK(xlNgwwtXTMCVW-HX3VbOf690>@T)&HsT~F*MxlomD
z{G9G?4I83u^w0YI{VNF$_15z#8)hmdi9V05s1LiTD8A=A#xH3|14ZXGqCe*-XSRC%
z%p%(C$7$9jn!@Q9bE-F4JqbF-4M#O?=5>1QB;tpqRb!Q09-<83Jo|W{$g^oJTa!q;
zk_JPlpFrkkA+B*)(w2uf`{Sq2#3k$rWs*!*tW@S%=vGda?jCMdW{&?s&gORLoFHCU
zNFf3~jMW+e*v3hNa3}*0g5d&?2;11K5i|r-;~fav(h3pK)l-A!3EP^l5ZJ9@B$Dt1
zZ4Y{gGM`~HN^elJ2hj!uyYqx6Y@5AB6vu&ig}p&~0MaKkm`L_pLOM2*4H^OmC+w9N
zDH55J1H>%=izPv#{4cB}K}tm6=7WimB9W2uaiX)!+Pb-WP=k1Q(b<))oUA=;sChW~
z`BDoQ86ht^V_18niY-b4|-dTmdsv%Lo$Z(yqPrVD+Z_QwnWA8
zxQl6q=qv>?Yk4-~7C9{L+
zjE~D*-;0l|gfwapv$}R7HDJy#|m-8cS00aIAyHDB98XWn3Ckb
z)1a$B
z#Xs<%L;Y0hQyb81@Uo!TE8I3*KC+EABoIbjTIk#gU93ca2KJKRm%iIs_|veA9c@yZ
zXq~+p@&EvB=S4TBsDJ7m)MfLR|BhC$=yY&{d>b}#X8lF4v2d$sU%M+{*y!@nlx6Om
zZ0GByTy|q7sWPf}s9MX6g}Chp^^X0!zJz8dxNAVaj`!EOEumad!eh)5SzYp)K18qX
zedDHs2P(%*O{!a2kDH_3ulnBtbQ0rmUU035pDM_*->yW6NHJnCDtTlvctxO<-a*Pj`ULFkZK8mp?m9eRX
z@)rFc)RoE%Yen>wC5bJXi&0LeXw*;q-qco=CLQ^{zvnhJdGyLJE`ParO+Z>J*$GZ8
z_ax@IS@M&D4{;I4#NuYE17*#UY!W3unn5#FOy(4;L^n_6I(z3(>B5F^2jtC62-}fM
z?ZL&Lu433hGL3R>7nCC6%Df|X#K@&Sf&aczHqG6DW2`JZjPNX%!M}e6mywWPCGF!b
z+t{vvrEkKgnMd)-Fr(vRoT*G^k+;qq6VcB@#m8VS|(o+f4P;7&gCg3{M!*
z)lEwPUs{-H{MQj0$pJ#Y@F36g&sh>t%XI_FJQ3vL3iZZu-$)1Hgvv&T1Lbh{*;p?I
z92fF-hdVs+0<#}!1Y#UXc@Gfh>z8$c
zXzq*qwT~bFrUkIIe2F1L%tQE$Ci^>ehf@Xd`jYP&UJ-m-9}5IpQ>I*nWoIWUsd@scrU
ztL&5_0+nTwb{oG21~R%Zj(kc%?2?Q*KPJTteoDIA;rx{C%@9fs29P9oOPWxPzFhJdJ(UCcMf$NpGDe>;_;G`+ih*K|
zIZmeYNEGS;%edbGl}P6`IdMj29>|M>u?*AUusML9Wn}``?$k~D$FUv&J8ii|g4$tp
zS#W$I?D4YWT3W6Fap~Y^X2K`NI*oyYq_Qa`Vz>Ns1s<#^0rS?Icq;CJaQRU~h9ESR5kCfr?GV#*eL;V2hC_;&6uBa<`WgCjBw
zEypRcWC@>;k13dEzvj;-$%x$CmZ2|9q4uKoP^M)<^lJt!&`?Gx%c(iNCpeH+npPsP
z?tPDi=NNQ7E7<150Xs;qELm*CqJ{9nL2wpg3FDIuN*IqYKfwLo+i-ln
zwzKo?Y4r5SAEkS5q?N3UQA2|;@})^2hjutxE46<2NL{IcS)oH)gJ*yogGwGxLB>a4
z;~P~2{uqu-NrB!B%0N6UA``<7Iv7@;A#l(C+`x7{isxNB?QVin>0J)MoUq`g48RVv
z7Zqz_gIsqIuuiWByBWx_+I|)pVKl>jnMI%$(TFJ`V`@u~KGbupFiCJ1OL6c>+{t#9
zXvpKPqMX_IIwho;-O|&@$EI|=TAG7JUCPb+jek={LnsPRiBoanR5nZ3Fxtlwycx+0
zW8k9g?I&aDm+Pq3q^9+jnU!29@lo>dsqfd7f@~Ztjnwb1)y$dt*V~F%Gb#6#Y7yH5
zpSMMp)CFW^%XJwqZMZD$?Q3S=3Y{?+35sDp&x$!DruluOGtLe*flz44iPhO@yiHUm
zOg#I0;^5KpWg2kv!7diCi0;R
zVqHw+FF1$66X0rDAXrT~t#&qs&~#kj$J6DflQ_+OO-C8!?*Fk*c7|x+>iA}7Kfn+}
zpnQ}4bOQT<)J{EjUq;=cmy&2;_0f+er~+7Ny$Gq)T{QF5!m6rUwDr_kTx>gCFBhz1
ztBlRpCAuvk)qbsc(T%T~u+vnjsyeNN%rhTaFEA9~SdNbMsDuhxv^b9n{299_Vek%BwcxNZOU-A_{l
zxjYY1j~%{tdC?&FTqM}bFmSXtS9RnES~CnLh0yTgDoAO(8Eh53e}CrJP24948Fq8!
ztU8N3PCT`~^?s+_gi(31nI|0d2dEy%YiRiNVxggS=z>GMWfmADLOJaO&@Q>%+;
zAY8BysA1VjM{YMBr*sL7uF?4L)?6u$ZY4%GRX0440sqZNIxz6*`_*(fd9Pz&HT`r?%p5@cKCL8l3N
zr_g0YyShF39CEFK`%&yJTydkZvd08_L}q0WQXKR0j3!{~CvrBSkUCcq;UXALxM9s-
z#jy8pJ=C9jP3Uo9@xjkH&_wR%#qtlCDSYDKquYZh-|S^bKgplDL_1N+X?C4L->cZNSNp5$ZIMdN^MuXk^Pu`&gy!AcIYl?RPl>U!Pn7*`DBLkk
z#YM_Sy8otP-k;zAVg$nuF#cMF`FdE%*aJEhdKkvx{ngjf~Wp+^9-uJ-c}D)NurKKXxdpl_TYCyXal=E
zdw5*l{A!|iSsxdYryASA7RecgVcmxqK%%Dfcw{!1ibS9GAjhj^FcsXkLG9|}lwzvH
z?Z;IQD%v}t!4n%V!CJL4T`_AFfjz8-%lomD8wVbNNDH~G?hsdH)w6F8)rje?bY|}f
z-TS{7ZlR@)@1B)rVg23}pg-9D-bIugZx&|!fvb_I3PxHm=TC9fiD
zPxpjz;umoWITLjjA0`HR5xH@gPW&isGjnmOi`$7uKV0v=;vEqmEGaf5
zWnG;Y_xTU9VBM^%`gig}l&@w#5{sLLhH`>N<(ic;Pa@sjvUbwL=t7u=jk3OTF)JFbe36v8UjHSZ_RAXBPCjd
zpaz@jt+_g052>}5I9q#l#0&R*+ZnJpM}Sc68AKCNT%Xvi9s8wqlYPx;UZcTbFoWy7
z0to%=4QN}uVtp}}8>+ROC4eLKr+RW^?q7fMp;s8#WWLt}feo+wMMN{wnd=bKJwqUz
zO!Gl-1iUVgnD{cl?$*si;8QcuFxj(x7lMXyCiSfl8lne$K%q?+hQr7qyXJQ3U1b#;
zNFc{lO3;a9wRQT80_6L^=K>;<`YRxeQyY)~rx$G+dW1PcV1tgU%GZd1!pbC=bIE^<
zfT#YA$N?ZD$B7_+tS*d*XNYiQjkCPEO9H&5u+%AgTk8-d3H4wzSUPQpn>u+M@;9tC
zqwvdenGfGHQGcPiKW2j0i)`)oB)}-L>Gu8kxum~!?}%ptVp2dPUeYy%Vi%H%biQfWAo@cOuZ$F>(c1>jG)-cicAjK{pV|gRNwqLB(-(C
zLi-@Dl0SL=wSP396H=;zb~o&jswV#}LvM*#;b_Ltet)nsrPC_-7(F@Hq@l;zs&NJG
zRHy)a5^;JyhrZGs!o|S_yI?`0fpBmep|gW+{j8|DIJnW-b*T+Fskx~+--4RX&K_?;
z4r&f+O>}kzr_auBjPt)tn%Y2EiicN5nwOV{pAW>#Ed%1_@^WyB
zQvcs6z_*S5tCR3@a!~(2o#ey(nCjazY52;$@}FYWPoAk+W%L+CkSZh4rkJ0noDmp>
z2xKja=5&<`09dH*Vzghu2f%GfdOM|kc|fgIUSWJqoMJawtm)9c&$0M?$Ih*!R){>)
zyOGgDGFCV>^|51;dW-t57nM8`>VkPUV4TM+9?H70jv{9o>Tjg(0b%a`F>RpFFTDm>
zFGR8;lvVhp9s(?HRA=y%r1I$07PIZ4)WPmGBZd~UEEK>dT<4vRy~#eFwZIdHQPjVS
zFeUa15f#|gc`f2&l%Dq$Fx`Dn=$RNv6%{xHUFl+jOH07ly3uweO&q^Lw^E7fgVHLg
z;~ogA{Y?B=sV)(85F1Ybsne91BM0))wy^R5ayu$)n~P^haI#}6aJf@=`=cRLpdEv6
zgYf8fRCpuIz=iC0FM@E0Oxu#0NC5Wl0Yo2ErAt&wcO+g@l05ec!MY@^bE5a8IEUdk
z$tuuHOu=_?7B=LWEv8G7O-lfoSwl|EK6#n3UyCJPOr3!>5ayOnKArTBzqyMQDSk8e
zq*eZ|V*RRme-BnP%1bw9_VjPP)TtP0!WQ&c5H>K-J-zJ`s98Ardo"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "from numpy import *\n",
     "from numpy.random import randint, randn\n",
diff --git a/doc/src/DecisionTrees/DecisionTrees.do.txt b/doc/src/DecisionTrees/DecisionTrees.do.txt
index c04e655e0..a734ad86b 100644
--- a/doc/src/DecisionTrees/DecisionTrees.do.txt
+++ b/doc/src/DecisionTrees/DecisionTrees.do.txt
@@ -35,6 +35,7 @@ given some assumptions, make predictions about the target feature value
 !split
 ===== A typical Decision Tree with its pertinent Jargon, Classification Problem =====
 
+Figure to come here.
 
 
 
@@ -462,6 +463,23 @@ os.system(cmd)
 !ec
 
 
+!split
+===== Algorithms for Setting up Decision Trees =====
+Two algorithms stand out in the set up of decision trees:
+o The CART (Classification And Regression Tree) algorithm for both classification and regression
+o The ID3 algorithm based on the computation of the information gain for classification
+
+We discuss both algorithms with applications here. The popular library -Scikit-Learn_ uses the CART algorithm. For classification problems you can use either the _gini_ index or the _entropy_ to split a tree in two branches.
+
+!split
+===== The CART algorithm for Classification =====
+
+
+!split
+===== The CART algorithm for Regression =====
+
+
+
 !split
 ===== Computing the Gini index =====
 
@@ -1127,6 +1145,24 @@ plt.show()
 
 However, by aggregating many decision trees, using methods like bagging, random forests, and boosting, the predictive performance of trees can be substantially improved. 
 
+
+!split
+===== From a Single Tree to Many Trees, that is meet the Jungle of Methods =====
+
+As stated above and seen in many of the examples discussed here about
+a single decision tree, we often end up overfitting our training
+data. This normally means that we have a high variance. Can we reduce
+the variance of a statistical learning method?
+
+This leads us to a set of different methods that can combine different
+machine learning algorithms or just use one of them to construct forests and jungles of trees, homogeneous ones or heterogenous ones. These methods are recognized by different names which we will try to explain here. These are
+o Votign classifiers
+o Bagging and Pasting
+o Random forests
+o Boosting methods
+
+We discuss these methods here.
+
 !split
 ===== Bagging =====
 
@@ -1426,9 +1462,19 @@ np.sum(y_pred == y_pred_rf) / len(y_pred)
 
 Example will be added here.
 
-!split
-===== Boosting: AdaBoost =====
 
+!split
+===== Boosting, a Bird'e Eye =====
+
+
+
+
+!split
+===== Adaptive boosting: AdaBoost, Basic Algorithm =====
+
+
+!split
+===== AdaBoost Examples =====
 !bc pycod
 from sklearn.ensemble import AdaBoostClassifier
 
@@ -1461,14 +1507,17 @@ for subplot, learning_rate in ((121, 1), (122, 0.5)):
 
 save_fig("boosting_plot")
 plt.show()
-
-
-
 !ec
 
 
 !split
-===== Gradient Boosting =====
+===== Gradient boosting: Basic Algorithm =====
+
+
+
+
+!split
+===== Gradient Boosting, Examples =====
 !bc pycod
 np.random.seed(42)
 X = np.random.rand(100, 1) - 0.5
@@ -1617,3 +1666,7 @@ for n_estimators in range(1, 120):
 print(gbrt.n_estimators)
 print("Minimum validation MSE:", min_val_error)
 !ec
+
+!split
+===== XGBoost: Extreme Gradient Boosting =====
+