From 2c8fdf957436214b409d767ce65c7c0be9719674 Mon Sep 17 00:00:00 2001 From: mhjensen Date: Tue, 5 Nov 2019 17:48:11 +0100 Subject: [PATCH] started to add material on gradient boosting --- .../html/._DecisionTrees-bs000.html | 21 ++-- .../html/._DecisionTrees-bs001.html | 19 +-- .../html/._DecisionTrees-bs002.html | 19 +-- .../html/._DecisionTrees-bs003.html | 19 +-- .../html/._DecisionTrees-bs004.html | 19 +-- .../html/._DecisionTrees-bs005.html | 19 +-- .../html/._DecisionTrees-bs006.html | 19 +-- .../html/._DecisionTrees-bs007.html | 19 +-- .../html/._DecisionTrees-bs008.html | 19 +-- .../html/._DecisionTrees-bs009.html | 19 +-- .../html/._DecisionTrees-bs010.html | 19 +-- .../html/._DecisionTrees-bs011.html | 19 +-- .../html/._DecisionTrees-bs012.html | 19 +-- .../html/._DecisionTrees-bs013.html | 19 +-- .../html/._DecisionTrees-bs014.html | 19 +-- .../html/._DecisionTrees-bs015.html | 19 +-- .../html/._DecisionTrees-bs016.html | 19 +-- .../html/._DecisionTrees-bs017.html | 19 +-- .../html/._DecisionTrees-bs018.html | 19 +-- .../html/._DecisionTrees-bs019.html | 19 +-- .../html/._DecisionTrees-bs020.html | 19 +-- .../html/._DecisionTrees-bs021.html | 19 +-- .../html/._DecisionTrees-bs022.html | 19 +-- .../html/._DecisionTrees-bs023.html | 19 +-- .../html/._DecisionTrees-bs024.html | 19 +-- .../html/._DecisionTrees-bs025.html | 19 +-- .../html/._DecisionTrees-bs026.html | 19 +-- .../html/._DecisionTrees-bs027.html | 19 +-- .../html/._DecisionTrees-bs028.html | 19 +-- .../html/._DecisionTrees-bs029.html | 19 +-- .../html/._DecisionTrees-bs030.html | 19 +-- .../html/._DecisionTrees-bs031.html | 19 +-- .../html/._DecisionTrees-bs032.html | 23 ++-- .../html/._DecisionTrees-bs033.html | 19 +-- .../html/._DecisionTrees-bs034.html | 19 +-- .../html/._DecisionTrees-bs035.html | 19 +-- .../html/._DecisionTrees-bs036.html | 19 +-- .../html/._DecisionTrees-bs037.html | 19 +-- .../html/._DecisionTrees-bs038.html | 19 +-- .../DecisionTrees/html/DecisionTrees-bs.html | 21 ++-- .../html/DecisionTrees-reveal.html | 103 ++++++++++++++-- .../html/DecisionTrees-solarized.html | 111 +++++++++++++++--- doc/pub/DecisionTrees/html/DecisionTrees.html | 111 +++++++++++++++--- .../DecisionTrees/ipynb/DecisionTrees.ipynb | 94 +++++++++++++-- .../ipynb/ipynb-DecisionTrees-src.tar.gz | Bin 213983 -> 213983 bytes .../pdf/DecisionTrees-minted.pdf | Bin 439743 -> 441329 bytes doc/src/DecisionTrees/DecisionTrees.do.txt | 86 ++++++++++++-- doc/src/DecisionTrees/rfcancer.py | 66 +++++++++++ 48 files changed, 993 insertions(+), 346 deletions(-) create mode 100644 doc/src/DecisionTrees/rfcancer.py diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html index 42e102612..f05dcbec4 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs000.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • @@ -242,7 +247,7 @@ MathJax.Hub.Config({
    [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University

    -

    Nov 4, 2019

    +

    Nov 5, 2019


    diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html index ecec6e052..9c06d6c60 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs001.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({

  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html index 10e742c0d..d88b8a470 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs002.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html index babc3f8e6..d5c66f891 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs003.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html index e27daa8f3..af914bbb3 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs004.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html index cd89bf89f..64c8d5688 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs005.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html index a633a013f..cbc9f6f66 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs006.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html index ec8fd5c4c..38bb1f918 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs007.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html index 5d7a0a02d..5520a7aa1 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs008.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html index 7e0d79c64..f2c1e07f2 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs009.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html index a90f7ef7f..ba9313b7f 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs010.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html index b468cbcc0..edb4eae11 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs011.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html index 5acbad162..9050aa322 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs012.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html index 743f757cd..318493518 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs013.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html index 990d9246e..cba54186a 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs014.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html index d8bed0045..313a7f044 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs015.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html index 39660ce4b..bdb273caf 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs016.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html index 01461e8b7..c7242f821 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs017.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html index 4ca9f7c88..86bc41c16 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs018.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html index 7d20eb124..b1e65e39f 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs019.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html index addf1619c..6a9d745ab 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs020.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html index 3ab39a28b..e3f573b4b 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs021.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html index d7341066e..0849a9b02 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs022.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html index 846eefc81..fb43399b8 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs023.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html index 6c328174c..9b22dcdc4 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs024.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html index ff5c15b78..2fcfd9fb9 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs025.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html index c42b05767..2fbaeef3c 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs026.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html index e68237480..9470bf605 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs027.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html index f17414fad..49351bbaa 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs028.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html index 01ded5581..d625f57f1 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs029.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html index 9f624b150..6130f1d64 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs030.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html index fabfcc25f..8b1675cc3 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs031.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html index adc5fb5d2..18213ab5c 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs032.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
  • Final regressor code
  • Pros and cons of trees, pros
  • Disadvantages
    • -
  • From a Single Tree to Many Trees, that is meet the Jungle of Methods
    • +
  • From a Single Tree to Many Trees, Meet the Jungle of Methods
  • Bagging
  • More bagging
  • Simple Voting Example, head or tail
    • @@ -196,8 +201,8 @@ MathJax.Hub.Config({
  • Now Bagging
  • Random forests
  • Random Forest Algorithm
    • -
  • A simple scikit-learn example
    • -
  • Then random forests
    • +
  • Random Forests Compared with other Methods on the Cancer Data
    • +
  • Compare Bagging on Trees with Random Forests
  • Feature Importance
  • Boosting, a Bird'e Eye
  • Adaptive boosting: AdaBoost, Basic Algorithm
    • @@ -223,7 +228,7 @@ MathJax.Hub.Config({ -

    From a Single Tree to Many Trees, that is meet the Jungle of Methods

    +

    From a Single Tree to Many Trees, Meet the Jungle of Methods

    As stated above and seen in many of the examples discussed here about @@ -236,7 +241,7 @@ 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. Voting classifiers
    3. Bagging and Pasting
    4. Random forests
    5. Boosting methods
      6. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html index 975c811e7..814e81304 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs033.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    6. Final regressor code
    7. Pros and cons of trees, pros
    8. Disadvantages
      9. -
    9. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      10. +
    10. From a Single Tree to Many Trees, Meet the Jungle of Methods
    11. Bagging
    12. More bagging
    13. Simple Voting Example, head or tail
      14. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    14. Now Bagging
    15. Random forests
    16. Random Forest Algorithm
      17. -
    17. A simple scikit-learn example
      18. -
    18. Then random forests
      19. +
    19. Random Forests Compared with other Methods on the Cancer Data
      20. +
    20. Compare Bagging on Trees with Random Forests
    21. Feature Importance
    22. Boosting, a Bird'e Eye
    23. Adaptive boosting: AdaBoost, Basic Algorithm
      24. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html index fdf6c10c4..fc1c21ca1 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs034.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    24. Final regressor code
    25. Pros and cons of trees, pros
    26. Disadvantages
      27. -
    27. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      28. +
    28. From a Single Tree to Many Trees, Meet the Jungle of Methods
    29. Bagging
    30. More bagging
    31. Simple Voting Example, head or tail
      32. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    32. Now Bagging
    33. Random forests
    34. Random Forest Algorithm
      35. -
    35. A simple scikit-learn example
      36. -
    36. Then random forests
      37. +
    37. Random Forests Compared with other Methods on the Cancer Data
      38. +
    38. Compare Bagging on Trees with Random Forests
    39. Feature Importance
    40. Boosting, a Bird'e Eye
    41. Adaptive boosting: AdaBoost, Basic Algorithm
      42. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html index 474dba6b1..cdc69bf8c 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs035.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    42. Final regressor code
    43. Pros and cons of trees, pros
    44. Disadvantages
      45. -
    45. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      46. +
    46. From a Single Tree to Many Trees, Meet the Jungle of Methods
    47. Bagging
    48. More bagging
    49. Simple Voting Example, head or tail
      50. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    50. Now Bagging
    51. Random forests
    52. Random Forest Algorithm
      53. -
    53. A simple scikit-learn example
      54. -
    54. Then random forests
      55. +
    55. Random Forests Compared with other Methods on the Cancer Data
      56. +
    56. Compare Bagging on Trees with Random Forests
    57. Feature Importance
    58. Boosting, a Bird'e Eye
    59. Adaptive boosting: AdaBoost, Basic Algorithm
      60. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html index 0f51a5012..d6a711693 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs036.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    60. Final regressor code
    61. Pros and cons of trees, pros
    62. Disadvantages
      63. -
    63. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      64. +
    64. From a Single Tree to Many Trees, Meet the Jungle of Methods
    65. Bagging
    66. More bagging
    67. Simple Voting Example, head or tail
      68. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    68. Now Bagging
    69. Random forests
    70. Random Forest Algorithm
      71. -
    71. A simple scikit-learn example
      72. -
    72. Then random forests
      73. +
    73. Random Forests Compared with other Methods on the Cancer Data
      74. +
    74. Compare Bagging on Trees with Random Forests
    75. Feature Importance
    76. Boosting, a Bird'e Eye
    77. Adaptive boosting: AdaBoost, Basic Algorithm
      78. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html index 32bcd8d58..31039557f 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs037.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    78. Final regressor code
    79. Pros and cons of trees, pros
    80. Disadvantages
      81. -
    81. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      82. +
    82. From a Single Tree to Many Trees, Meet the Jungle of Methods
    83. Bagging
    84. More bagging
    85. Simple Voting Example, head or tail
      86. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    86. Now Bagging
    87. Random forests
    88. Random Forest Algorithm
      89. -
    89. A simple scikit-learn example
      90. -
    90. Then random forests
      91. +
    91. Random Forests Compared with other Methods on the Cancer Data
      92. +
    92. Compare Bagging on Trees with Random Forests
    93. Feature Importance
    94. Boosting, a Bird'e Eye
    95. Adaptive boosting: AdaBoost, Basic Algorithm
      96. diff --git a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html index 33680f159..9acb85de1 100644 --- a/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html +++ b/doc/pub/DecisionTrees/html/._DecisionTrees-bs038.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    96. Final regressor code
    97. Pros and cons of trees, pros
    98. Disadvantages
      99. -
    99. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      100. +
    100. From a Single Tree to Many Trees, Meet the Jungle of Methods
    101. Bagging
    102. More bagging
    103. Simple Voting Example, head or tail
      104. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    104. Now Bagging
    105. Random forests
    106. Random Forest Algorithm
      107. -
    107. A simple scikit-learn example
      108. -
    108. Then random forests
      109. +
    109. Random Forests Compared with other Methods on the Cancer Data
      110. +
    110. Compare Bagging on Trees with Random Forests
    111. Feature Importance
    112. Boosting, a Bird'e Eye
    113. Adaptive boosting: AdaBoost, Basic Algorithm
      114. diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html index 42e102612..f05dcbec4 100644 --- a/doc/pub/DecisionTrees/html/DecisionTrees-bs.html +++ b/doc/pub/DecisionTrees/html/DecisionTrees-bs.html @@ -85,8 +85,7 @@ Automatically generated HTML file from DocOnce source ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -101,8 +100,14 @@ Automatically generated HTML file from DocOnce source ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -187,7 +192,7 @@ MathJax.Hub.Config({
    114. Final regressor code
    115. Pros and cons of trees, pros
    116. Disadvantages
      117. -
    117. From a Single Tree to Many Trees, that is meet the Jungle of Methods
      118. +
    118. From a Single Tree to Many Trees, Meet the Jungle of Methods
    119. Bagging
    120. More bagging
    121. Simple Voting Example, head or tail
      122. @@ -196,8 +201,8 @@ MathJax.Hub.Config({
    122. Now Bagging
    123. Random forests
    124. Random Forest Algorithm
      125. -
    125. A simple scikit-learn example
      126. -
    126. Then random forests
      127. +
    127. Random Forests Compared with other Methods on the Cancer Data
      128. +
    128. Compare Bagging on Trees with Random Forests
    129. Feature Importance
    130. Boosting, a Bird'e Eye
    131. Adaptive boosting: AdaBoost, Basic Algorithm
      132. @@ -242,7 +247,7 @@ MathJax.Hub.Config({
      [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University

      -

      Nov 4, 2019

      +

      Nov 5, 2019


      diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html b/doc/pub/DecisionTrees/html/DecisionTrees-reveal.html index af922e8f1..11ddd11e0 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 4, 2019

      +

      Nov 5, 2019


      @@ -1439,7 +1439,7 @@ However, by aggregating many decision trees, using methods like bagging, random

      -

      From a Single Tree to Many Trees, that is meet the Jungle of Methods

      +

      From a Single Tree to Many Trees, Meet the Jungle of Methods

      As stated above and seen in many of the examples discussed here about @@ -1452,7 +1452,7 @@ 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. Voting classifiers

      3. Bagging and Pasting

      4. Random forests

      5. Boosting methods
        6. @@ -1754,30 +1754,111 @@ this setting.

        Random Forest Algorithm

        The algorithm described here can be applied to both classification and regression problems. + +

        +We will grow of forest of say \( M \) trees. + +

          +

        1. For \( m=1:M \) we
          2. + +
            + +

          • Draw a bootstrap sample of from the training data organized in our \( \boldsymbol{X} \) matrix.
            • + +

          • We grow then a random forest tree \( T_m \) based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
            • + +
              + +

            1. we select \( m \le p \) varibales at random from the \( p \) predictors/features
              2. + +

            2. pick the best split point among the \( m \) features using either the CART algorithm or the ID3 for classification and create a new node
              3. + +

            3. split the node into daughter nodes
              4. +
            +

            +

          +

        2. Output then the ensemble of trees \( \{T_m\}_1^{M} \) and make predictions for either a regression type of problem or a classification type of problem.
          3. +
        -

        A simple scikit-learn example

        +

        Random Forests Compared with other Methods on the Cancer Data

        -

        from sklearn.ensemble import RandomForestClassifier
+
        import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
+
+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")
+#Instantiate the model with 500 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
 #Cross validation
-accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#<matplotlib.axes._subplots.AxesSubplot object at 0x7fe967d64490>
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()
        -

        Then random forests

        +

        Compare Bagging on Trees with Random Forests

        diff --git a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html index 56336a6f5..19099da6f 100644 --- a/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html +++ b/doc/pub/DecisionTrees/html/DecisionTrees-solarized.html @@ -105,8 +105,7 @@ div { text-align: justify; text-justify: inter-word; } ('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', + ('From a Single Tree to Many Trees, Meet the Jungle of Methods', 2, None, '___sec31'), @@ -121,8 +120,14 @@ div { text-align: justify; text-justify: inter-word; } ('Now Bagging', 2, None, '___sec37'), ('Random forests', 2, None, '___sec38'), ('Random Forest Algorithm', 2, None, '___sec39'), - ('A simple scikit-learn example', 2, None, '___sec40'), - ('Then random forests', 2, None, '___sec41'), + ('Random Forests Compared with other Methods on the Cancer Data', + 2, + None, + '___sec40'), + ('Compare Bagging on Trees with Random Forests', + 2, + None, + '___sec41'), ('Feature Importance', 2, None, '___sec42'), ("Boosting, a Bird'e Eye", 2, None, '___sec43'), ('Adaptive boosting: AdaBoost, Basic Algorithm', @@ -180,7 +185,7 @@ MathJax.Hub.Config({

        [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University

        -

        Nov 4, 2019

        +

        Nov 5, 2019












        @@ -1431,7 +1436,7 @@ However, by aggregating many decision trees, using methods like bagging, random











        -

        From a Single Tree to Many Trees, that is meet the Jungle of Methods

        +

        From a Single Tree to Many Trees, Meet the Jungle of Methods

        As stated above and seen in many of the examples discussed here about @@ -1444,7 +1449,7 @@ 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. Voting classifiers
        3. Bagging and Pasting
        4. Random forests
        5. Boosting methods
          6. @@ -1741,27 +1746,103 @@ this setting. The algorithm described here can be applied to both classification and regression problems.

          +We will grow of forest of say \( M \) trees. + +

            +
          1. For \( m=1:M \) we
            2. + +
              +
            • Draw a bootstrap sample of from the training data organized in our \( \boldsymbol{X} \) matrix.
              • +
            • We grow then a random forest tree \( T_m \) based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
              • + +
                +
              1. we select \( m \le p \) varibales at random from the \( p \) predictors/features
                2. +
              2. pick the best split point among the \( m \) features using either the CART algorithm or the ID3 for classification and create a new node
                3. +
              3. split the node into daughter nodes
                4. +
              + +
            + +
          2. Output then the ensemble of trees \( \{T_m\}_1^{M} \) and make predictions for either a regression type of problem or a classification type of problem.
            3. +
          +









          -

          A simple scikit-learn example

          +

          Random Forests Compared with other Methods on the Cancer Data

          -

          from sklearn.ensemble import RandomForestClassifier
+
          import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
+
+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")
+#Instantiate the model with 500 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
 #Cross validation
-accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#<matplotlib.axes._subplots.AxesSubplot object at 0x7fe967d64490>
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()
 
          
 
          
 









          
 
-
          Then random forests 
          
+
          Compare  Bagging on Trees with Random Forests  
          
 
          
 
 
diff --git a/doc/pub/DecisionTrees/html/DecisionTrees.html b/doc/pub/DecisionTrees/html/DecisionTrees.html
index c0ce80a23..1456bf337 100644
--- a/doc/pub/DecisionTrees/html/DecisionTrees.html
+++ b/doc/pub/DecisionTrees/html/DecisionTrees.html
@@ -110,8 +110,7 @@ div { text-align: justify; text-justify: inter-word; }
               ('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',
+              ('From a Single Tree to Many Trees, Meet the Jungle of Methods',
                2,
                None,
                '___sec31'),
@@ -126,8 +125,14 @@ div { text-align: justify; text-justify: inter-word; }
               ('Now Bagging', 2, None, '___sec37'),
               ('Random forests', 2, None, '___sec38'),
               ('Random Forest Algorithm', 2, None, '___sec39'),
-              ('A simple scikit-learn example', 2, None, '___sec40'),
-              ('Then random forests', 2, None, '___sec41'),
+              ('Random Forests Compared with other Methods on the Cancer Data',
+               2,
+               None,
+               '___sec40'),
+              ('Compare  Bagging on Trees with Random Forests',
+               2,
+               None,
+               '___sec41'),
               ('Feature Importance', 2, None, '___sec42'),
               ("Boosting, a Bird'e Eye", 2, None, '___sec43'),
               ('Adaptive boosting: AdaBoost, Basic Algorithm',
@@ -185,7 +190,7 @@ MathJax.Hub.Config({
 
          [2] Department of Physics and Astronomy and National Superconducting Cyclotron Laboratory, Michigan State University
          
 
          
 
          
-
          Nov 4, 2019
           
+
          Nov 5, 2019
           
 
          
 
          
 









          
@@ -1436,7 +1441,7 @@ However, by aggregating many decision trees, using methods like bagging, random
 
          
 









          
 
-
          From a Single Tree to Many Trees, that is meet the Jungle of Methods 
          
+
          From a Single Tree to Many Trees, Meet the Jungle of Methods 
          
 
 
          
 As stated above and seen in many of the examples discussed here about
@@ -1449,7 +1454,7 @@ 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.  Voting classifiers
            3. 
 
          3.  Bagging and Pasting
            4. 
 
          4.  Random forests
            5. 
 
          5.  Boosting methods
            6. 
@@ -1746,27 +1751,103 @@ this setting.
 The algorithm described here can be applied to both classification and regression problems.
 
 
            
+We will grow of forest of say \( M \) trees.
+
+
              
+
            1.  For \( m=1:M \) we
              2. 
+
+
                
+  
              •  Draw a bootstrap sample of from the training data organized in our \( \boldsymbol{X} \) matrix.
                • 
+  
              •  We grow then a random forest tree \( T_m \) based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
                • 
+
+
                  
+   
                1.  we select \( m \le p \) varibales at random from the \( p \) predictors/features
                  2. 
+   
                2.  pick the best split point among the \( m \) features using either the CART algorithm or the ID3 for classification and create a new node
                  3. 
+   
                3.  split the node into daughter nodes
                  4. 
+
                
+
+
              
+
+
            2.  Output then the ensemble of trees \( \{T_m\}_1^{M} \) and make predictions for either a regression type of problem or a classification type of problem.
              3.  
+
            
+
 









            
 
-
            A simple scikit-learn example 
            
+
            Random Forests Compared with other Methods on the Cancer Data 
            
 
            
 
 
-
            from sklearn.ensemble import RandomForestClassifier
+
            import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
+
+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")
+#Instantiate the model with 500 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
 #Cross validation
-accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#<matplotlib.axes._subplots.AxesSubplot object at 0x7fe967d64490>
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()
 
            
 
            
 









            
 
-
            Then random forests 
            
+
            Compare  Bagging on Trees with Random Forests  
            
 
            
 
 
diff --git a/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb b/doc/pub/DecisionTrees/ipynb/DecisionTrees.ipynb
index 7cf60d741..a8f87d819 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 4, 2019**\n",
+    "Date: **Nov 5, 2019**\n",
     "\n",
     "Copyright 1999-2019, Morten Hjorth-Jensen. Released under CC Attribution-NonCommercial 4.0 license\n",
     "\n",
@@ -1397,7 +1397,7 @@
     "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",
+    "## From a Single Tree to Many Trees, 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",
@@ -1406,7 +1406,7 @@
     "\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",
+    "1. Voting classifiers\n",
     "\n",
     "2. Bagging and Pasting\n",
     "\n",
@@ -1766,7 +1766,24 @@
     "## Random Forest Algorithm\n",
     "The algorithm described here can be applied to both classification and regression problems.\n",
     "\n",
-    "## A simple scikit-learn example"
+    "We will grow of forest of say $M$ trees.\n",
+    "1. For $m=1:M$ we\n",
+    "\n",
+    "  * Draw a bootstrap sample of from the training data organized in our $\\boldsymbol{X}$ matrix.\n",
+    "\n",
+    "  * We grow then a random forest tree $T_m$ based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached\n",
+    "\n",
+    "1. we select $m \\le p$ varibales at random from the $p$ predictors/features\n",
+    "\n",
+    "2. pick the best split point among the $m$ features using either the CART algorithm or the ID3 for classification and create a new node\n",
+    "\n",
+    "3. split the node into daughter nodes\n",
+    "\n",
+    "\n",
+    "\n",
+    "4. Output then the ensemble of trees $\\{T_m\\}_1^{M}$ and make predictions for either a regression type of problem or a classification type of problem. \n",
+    "\n",
+    "## Random Forests Compared with other Methods on the Cancer Data"
    ]
   },
   {
@@ -1777,23 +1794,79 @@
    },
    "outputs": [],
    "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "from sklearn.model_selection import  train_test_split \n",
+    "from sklearn.datasets import load_breast_cancer\n",
+    "from sklearn.svm import SVC\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.tree import DecisionTreeClassifier\n",
+    "\n",
+    "# Load the data\n",
+    "cancer = load_breast_cancer()\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)\n",
+    "print(X_train.shape)\n",
+    "print(X_test.shape)\n",
+    "# Logistic Regression\n",
+    "logreg = LogisticRegression(solver='lbfgs')\n",
+    "logreg.fit(X_train, y_train)\n",
+    "print(\"Test set accuracy with Logistic Regression: {:.2f}\".format(logreg.score(X_test,y_test)))\n",
+    "# Support vector machine\n",
+    "svm = SVC(gamma='auto', C=100)\n",
+    "svm.fit(X_train, y_train)\n",
+    "print(\"Test set accuracy with SVM: {:.2f}\".format(svm.score(X_test,y_test)))\n",
+    "# Decision Trees\n",
+    "deep_tree_clf = DecisionTreeClassifier(max_depth=None)\n",
+    "deep_tree_clf.fit(X_train, y_train)\n",
+    "print(\"Test set accuracy with Decision Trees: {:.2f}\".format(deep_tree_clf.score(X_test,y_test)))\n",
+    "#now scale the data\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "scaler = StandardScaler()\n",
+    "scaler.fit(X_train)\n",
+    "X_train_scaled = scaler.transform(X_train)\n",
+    "X_test_scaled = scaler.transform(X_test)\n",
+    "# Logistic Regression\n",
+    "logreg.fit(X_train_scaled, y_train)\n",
+    "print(\"Test set accuracy Logistic Regression with scaled data: {:.2f}\".format(logreg.score(X_test_scaled,y_test)))\n",
+    "# Support Vector Machine\n",
+    "svm.fit(X_train_scaled, y_train)\n",
+    "print(\"Test set accuracy SVM with scaled data: {:.2f}\".format(logreg.score(X_test_scaled,y_test)))\n",
+    "# Decision Trees\n",
+    "deep_tree_clf.fit(X_train_scaled, y_train)\n",
+    "print(\"Test set accuracy with Decision Trees and scaled data: {:.2f}\".format(deep_tree_clf.score(X_test_scaled,y_test)))\n",
+    "\n",
+    "\n",
     "from sklearn.ensemble import RandomForestClassifier\n",
     "from sklearn.preprocessing import LabelEncoder\n",
     "from sklearn.model_selection import cross_validate\n",
     "# Data set not specificied\n",
-    "X = dataset.XXX\n",
-    "Y = dataset.YYY\n",
-    "#Instantiate the model with 100 trees and entropy as splitting criteria\n",
-    "Random_Forest_model = RandomForestClassifier(n_estimators=100,criterion=\"entropy\")\n",
+    "#Instantiate the model with 500 trees and entropy as splitting criteria\n",
+    "Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion=\"entropy\")\n",
+    "Random_Forest_model.fit(X_train_scaled, y_train)\n",
     "#Cross validation\n",
-    "accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']"
+    "accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']\n",
+    "print(accuracy)\n",
+    "print(\"Test set accuracy with Random Forests and scaled data: {:.2f}\".format(Random_Forest_model.score(X_test_scaled,y_test)))\n",
+    "\n",
+    "\n",
+    "import scikitplot as skplt\n",
+    "y_pred = Random_Forest_model.predict(X_test_scaled)\n",
+    "skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)\n",
+    "#\n",
+    "plt.show()\n",
+    "y_probas = Random_Forest_model.predict_proba(X_test_scaled)\n",
+    "skplt.metrics.plot_roc(y_test, y_probas)\n",
+    "plt.show()\n",
+    "skplt.metrics.plot_cumulative_gain(y_test, y_probas)\n",
+    "plt.show()"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Then random forests"
+    "## Compare  Bagging on Trees with Random Forests"
    ]
   },
   {
@@ -1976,7 +2049,6 @@
     "\n",
     "\n",
     "\n",
-    "\n",
     "## Gradient Boosting, Examples"
    ]
   },
diff --git a/doc/pub/DecisionTrees/ipynb/ipynb-DecisionTrees-src.tar.gz b/doc/pub/DecisionTrees/ipynb/ipynb-DecisionTrees-src.tar.gz
index 39e57db23b7def1959b1a6f19cf4ed36889552ba..2a0930555a8f0f82918570c66a1f6905daff9b65 100644
GIT binary patch
delta 27
jcmccL&U?R|mtDS_gJI3`gN^L1?2N7KOk3HRZ#V-0jHU@O

delta 27
icmccL&U?R|mtDS_gW-?ofkyULcE(nArmgJEH=F^C76~E%

diff --git a/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf b/doc/pub/DecisionTrees/pdf/DecisionTrees-minted.pdf
index b20336b321c06673d5905b92fc9cb17be5a6312e..6d63acf38125e1d63a21d1d92b9c1c7dfbe3c0c3 100644
GIT binary patch
delta 22930
zcmZs>Q*fYNw00ZYwma(Bwr$(C^@bhWPC7})wvCQ$r(@g6`F8EvSO2+tYOTvPtJX7W
z%-(an+h@EwUNAsG8n@asb!^@gCWuZv^Ut^`1v-l`of(cOo!!BtN(?^}zY<9AtRkz1
zDa^4YvEik+2`N5(i4;bYBAL{$-`*Nx5su1OLf_s$r@XA?G)xDF{yC6`Po{g=#C5H{
z{@e5JJT|twJiU0HhTmCWV!1t;#<_^JlTL2i{@Yghr~DJhvQq6lKYnnkv`dT@Ypg-Q
zmiIo!r5UM=ZLj{bL)rhX`IIH{53N40l9QGr#I(eR+lpiGsUdsA@qz
zb|B;(JpIk&p+;S2ad@JfQ^nl6bw*dHDeID{Z#@J@TRe7t*7&sPf#Qn?Sp$iZK0vOzov6-A$g$9X75Iqf~Cmsm-5
z4ni)tcgO-L#W~2ndAh|56EBtNl|w2n>fL1Xr71tlDL(vyKGz?g8tVk>nTCCzuLSGkX^nCAp~Q1%0<
zjpAm1mUE2CH^fLVl-|x9b!Ch%?>tg64gj3Zh3=-pv4w$~{r$)eN{t06=--R}9^rQos^kE6pdgn`#=NZua
zTSdne)#bob_R6cfshKmuqnYtvN72umq48E`i11!yYA;Cnp;!iTe9l#T-JKwXI$3NP
z4D%NLbK~TKoe=jV*yM~$1sq{vWVJSqU0(%5Ta!CEit?T%3cVqF86|B7IfI(_Pyl|xU{yP0Mr{k)XCjp)Nu9r%m3gQX>
zG-XUF8>#0*S?gV?)&2lgJ4~v3*}@IqU(`MeF~iTTuA6w+EBep)v-Lt9z)2uaHhJgd
zLr~d6IjHa)f-=M8jvl6lbhkkbVG8$S`H>De9lGLp$RFWx`%&5aq-eQ%&>XOfvzO;j
zwA61PnX_MEcCQK0jS|1|)k|9NGu`iXW$j^EdrA&VwA<-6`O`F{yXcE_NA+FhqOg)#
zYf61~e}XM>LfDfRVx9X_6Yi*!B1F;nUn|BVM4N0AWY2Mb@#Q%s7l|A8>zULxQ1KyL
zb8NWC)2*9_DWjG4cCO{gUIEDE>hbQ&A4j29D`A$*=?9Y=k+i*;&&++c3a3Qdkp<3Y
zssGW|e@rPxLl=C%Dk(*VlH&8l;$uI$AKtnc?2Mvn9~(3s++yDGPuc_lhu=FqUB{Wk
z6^Sgyf{x->6&20gxjSqA2=+Y1^L}HOPR`mEA)#Pwp9b5osF+bb5CPCsccXNQHS@f!
z67gSd8-2>LSjKM`75`C=;-={Eg~Dwr?suMM9n>)W{OU&Y$`N}P3;m|=p=5cC$qu=#
zqE7#oHq%b?^*QgL%4`2(1oye`MG$7!ycNVFi874ZB`HW)kjM*o8#)-x{pEVODlq-;
z^E18hn=NN%N~`ti8jPS0C-DVHYf?I);HXwzS*CD3$P*+M_&~n!z0(hwgbtsC25p^$
z25U+LW9CkuYXzeQsMgp+@kODISH8ggKP6N
zspTDyS4yMQ7i}-L2wdu4sgoW*DrEldxkw=a9|;#$qWf)_8S5p|B)a8BG}Gp=ZAjq$KT6@3uHdb
zv8Qol6&J5wGWRSz*>haHJXziP#xqE17B!CEY*~9Ik_6aumFx!?QawcOM41*1e!|Fr
zxSINc&=9=-gG$VPamCAI?y^vF>pa1-@HlOVspFt?`x_Qu>wnAORm^5Gq9HbGb@e16r2
z3WDn8I1T8HjQX#r(W-RRCp#RWP0lsC8|@v}a^U#EQgWVu^vw=y`q5Ca
zqgJ2BbGjCKyyTPKemKKgl<{)9fYigq%uV6o`Pktd&26V_E1Iquex_JlaJBP9-Bdi5*MRvFoOT^H1HMG!t8abl6*~#eZBgzCWn(
z39YE5q3o_b*$6ckDRc+Iw$aAL`j$o$^}Fc+Wm>_<_6#CbA`F|=un6Aw2Z8ljuo`7i
zY!-@_f6}Lz=jU?!mTnj)lD4mTue*36EAk447YXS%Vg4>AAvw7Rab-Y6i1u-xiiyu?
z5);>{^K#zaj^D(_K(H|%v*Kif;6)gW`L5Ok{^b3mj`ed`#x&eyXkx5`tH9~;1c@~c
zAlHM6!WwA4w%QXB8L;Y8$HSBS-o`la^JZTxBD4{0j6L2C!KPQQf?K}2GL}ZZ_Qz(C
z%EN?446IZQ}Vi={dZCA7jVZ%~)X}UL-H6QFX
z=`X$Hg;xoii$~M^;&V|$1Pg2u+cK&$v7;ODXHmW>+#SZ4xSmb!8?|7dl+qw3#DL0Di5RfM}*yk;C;W
zHxi~hbCo^^8T
zdEyoMPioMdPFZ3C1FK|!eNvut+d!(m7|1I%S-<3HBM54!j&cL}&P)P~Yh-}&e0QI{j*TaS*Mn7Q^
zomkFs6^w?tvdSqVS)TmIwW_VM*A?8uL?Tq4t(M@55kKtLlC$)$Ydnt|Ue-*%!v!kQuS
zi@iNjBKZKX@_Jc|8lA!IqpTE=x^W=*V?!Y#VHptl8@pCQ%K2VDKF~mwGcf=?8t-m1UOb`%)Kgk(4rdVsD{q<&z7vORQU<5rgTYVtvAjhlzPZH
z|5)Dc78;X#dWm{hV(1OGTSs?)+5clWF`Y~gdej4*JK&`N5ITaBHhHzGr+_UHl}9#9
zDb_0gb?_%3x2PEAZ4?d9`9WC({RhJ@tTAjb=?$!~M0jPi%aRxp=LAVLc1*JU5v5i(O;hwJ0
zVHbMCEMrGrfIV-C4IVCZu8IpsH5)RdB0q~2F}((sQc!+2cJ`)AoH>8q{lOobq2Lg7
z;FX*n_ifMz%dv`N2H7wDx<&N5Bs`_5-XlCXPaSvHVReQP-u
z*UEnmOm573HxAzmVaZRbKUF{@_rYGAb^P~Y1cE>$&`-+?+p~w>`aac4XqF&yt_8!u
zfj2hxD_i2*ct5!NWRqCr)6-+av(~~I%<8a^QozAHdtm7O
zrs=(2^kBt5bCi!6W8OiK3g;J>=GF)lP2QCub)B)}%$Fa{P47KW`~
za0+fe*Mg$2Cq4c5wDuaol3G?kfa)~b`!*qGf|GN&@_^Rh^Id)NVB@lqlh~|CdoR;@}hY{`_(MpZ!kNV2YDQ`%g+r+G7RWsUv>U{%m9y2H+v
zJ?8P`<@Uqz3CW5OT~N(^itRjhoE#vDQ&_rH6Pzc0uesK@m(+}Uq?9&$=zW@fJSYjH
zFZk>J;PW~9BskXGVa~%WayZneUrF~CELZ~#by05rb-EBlYw`aKYE}6xRl6RoWAR?FPdYKwtp`gv?g+g{+fVCjV1%gAY*ju&xH4|
zRk*0B;%SL`tK{+4dK#=Ex>v}$e*wWKQiX6RxrZb6jGjdN^JnwZCBwObjABe0`0GNE
zb5IiE#IIY8m!mcSsJDUQ$4*0S@vJt8!Z08EbV~qS1>agC2PzeS?hA^COM}rWRhXB`
z5OGBFYr+8sytU=ew~%^4jAc`5VvCfz5#-`DHXriIja5e~8G{&^d>ei~
zQEb@}WOosjbI@Wexe|QW`xG?fvy3uM(d^bfW`X2C@A~yLr`OsR`fg}*%C|a(q_XoZ
zrunzhx7EibFPjq}FKlJIOd#=RvJGj-S?e~gDdE~FQZL_-w6*PEFFQR&LRNTYPZ+dhf@L$mSNAF
z>v^ljV~Qk#4OKj;ilbOYW!n)4nc$QUL`rhBM|43T!F?7uaC%0;fOlP&CHW;Cg%NL8
z?ni{(FDIP^3Vc<6Z!?44U|7;B301icU0~P;ZBd}|C>Vo{V%*a^mIdr&a6#RNu;~Ng
zn1s=bqgFuHUco~SI!%lm*l;{f;hog{&0sEmD%R_>?9;BkdH;Pjvd;7g=SW}%WOU|y
zy<+{O0}?<_)Srpu*mZt6X>;Kk!$|DZ>5StGLZ&^Elu(nZghy%Xc
zT4b(7f)w9U{q_6C(|!MH01BjPs>?pI^Lu9@14}{?&-P+f({MPN795`no>MT)&(_s~$`z>_HRV6$i&^er2<^|9FaN2$A9t0)D_
zfzQ1j>XoU$tVer_KJBdG$#OtO4S+$6ZUZZd+we}(x8RmmW!QQJ5n%>!&xkc{pipe5
zcQBX>`hs2DAj=FrN_8Ntt}LYL@LIEJL<)j_tVh1d7Hf;_&r7lnm?LI9@PMPH_Q
zcS!525469m1Wkkz?v^qZ-Q}Uz9lk-KD{zNz<={q3G2{uE2ss~gEFO6*3fD&h8EpgJ^)u~0
zeZs59>q~7+k*z4w7W%m8U(FHW_(K$JweKy(F4Z(7#MGf}I4`w)8|eMTcM)BS7j_pu
z+%WFk?X>zs8df=5bO(G*8=7!oViea*yTXTrmUKbXgyRs!Lr@alySI$E4Y1sa+$78;
z!Jhn1r-WMt`L{=^Q5-*1oKN!xg3eITbzeViQ?Jj}csOt$?xN?}=<=v@@&55Wu&4Xl
zLAP1CV22A_FZl1IylaUZi){_m`b)cQLH;mATxB?1HcpN^hI|UM5$O)Lv7QigkY2xK
zqBa4)#;1A9meO4R)nBYlUqCR};EE6nU#@_p<1i;S6Q0b6*L|q&1}?}sI>xu}{ReYY
zWA?H#^yyJ8wa+m8jP}R^U{{v_6@L~)k85n{2W>+FFLpg!(eusov!nKM*};KUH?8nD
z6upVedmiNry>%;o-9I0B0v!$MrSxx;KRbdo%`uz)rPN90yma?!W`NH&qHV|_g_)pW
zwTz`V$E=j_tk`?pZzMI(CgJAz;h1t&AqJ1w&Et+dWYupFjdNMJ+`XQMEGRZ<$iez-
zJp7324?F*EwNS8@F$P3TSGH9GGjqV1LPuGswe+N$BE%E*SwdH
z@{Y77DE%a}whR3iY9K=*qr}{;T+%?$$Xdra(Wmx0y&SFZlFCu2)$z25#LbslIxG{G
zh0k804Zl>4J&$g**cLHwwySgT2SDIt%B5fk>MsYM7YczL;jlUC3lE`@tY?;RTO28#
zM+vHh8I3d2K)zV~0s=V5lAp+(NjgxejL6_=6a7R1CEHDp1kO|1T2ytD5{2hUa91B$
zFHOHnXS1O2J~%XVE}SMnLNhV*U%hAx%@_-Uo=ujPjzQ3=^JH6VxwfFmwsuL^>jje`0Yzcd~jI{NTaUcNYS
zqd1TcQd~3wP@#|5b~OA%Bym3eY>j9ILxI9(C1E0QG_ir>f3#EFcj%Syi3LLi;D_P6$$yKC&X
z+(hFBn^j0pj8s$C+6dhqhpOh@3CD!4o%8_bap2sdN58dV-rg#
zd~=&NqkzSA4dle~VNWe_y_!B_nDQFEAE33|)~j!Lebw7;Kr18q>=>SJ&h@Vows1PJ}OrXUO;=*${@s{rAU52c(SyQ38mv*&)+k
z6{#$Kwd(J}D?M4=r`dG#>PioP-Q{QG6~39b&Oc$qd3Aa%wl3W%G8Wf7R_90?hfqLWlnfsyjE$ilTXM?{-QNmY+=i$-x=QhFHXW=t}H<$-^+P&Y9
z%Mneg8rray+-W1R5*-^&%$Rc^1>jwSB`aY@nj5oNlWtUG&5kO^Ad$_0?=rcTp&FuDv;h67H3Q$TP@Mt|%_cYjSgs38;q6bvT(P
z`xU%<;)Rf}J4F3ECX=DTY~5^USF$r6HlMS`Ii1T%IPy1FJ?6&5{&*&3&XIIn4m!fL
za5P)|Xi8aGRLHVVC*{^X)gL*n#
z;3DQHQARZ)v)r)UecV0&OTpC9E05+{WxqY9Fw-*T+h}fWYIY4Q(nU;>8YBm8>(J1k
z4ETK^#d4?3qtzlh0UnlQ{kd)*Wg
zLX1STEEhixk{y$MQ7G*85XvOwQ{1#>BoDD^2GHm_vex-^mg~j$I2(Fydd}1w#i8>+
zVGK5`7&SL;E4Uaw(qs>K3*($5VqY9X7z*8~zvQmQ(sM%tfT@xsJ)TGJr&n(Dq-qRMvaZi
zc)^8XyZ|vzfVA&HV68@%Nk*wKdTM`gNzhM^dAY)vvLB=br*)sfcr(z*DNINZnvV5~
zo!J30-OzY5G-ef|Qv@UWsmkdyKqdjAn=mBj>{F=K}F6MJEOM?UumOK5L)1GA48CCUg=LOTfAzxi74Kctt#kaLB#R2m}9;zz4W$^haa@wcw;+LZV8Ceck)J$BIgm%AFC8%3kYV+^4ELUi!uU3Z{)0S>qi2JggwzHxU0Uu4z$|X~H5OjV3
z=n^$v5CDmWdaB5=Qu2en)dQs*hn;nwzKq+G;}Ih2{)tqFKF&Ta*Io`(dp_5cxXp%B
zuEO_%by^fqL3bh;Fp=%k7-J$D8-0%?_GMD^*9nsLHPWIg#-gH^t?Pj>VrbFQQXP0$
z4|(g!|KwW#D`CKmq&FrO35I2cF2Z;TD1Zf;&~#Bnz$R_D-P
zEFQ{K-eyHcUC@B@0-^|MAlM}as0{To(i}Au-(XA2IBToCm|%S6Bx59|q~I{ZTZ|;#
zeUBaW{%FV?deA|KZPDLth8QAZW#$+4%&O$7ieD;X@&H6|oeBV{PZY+vcs7{mA)U2i
z8a5>XI9lipPdt4ve~^cC4n@@*@fy`#S`xg}BQMpcU&`^poYc@T@Wb+e>6~AqXC0;v
zo>(~pg^hkRc!So-yG>^+%<}Rkc8CD6_yEPzpLEMk7hU_j@Dly5f_6h6ym^TEes*dB
zy#!+m+~4WjBfrhBYw7gPd)z~7D{2O@EJVxh3nFZ_z`&-hA}_;!=H`URXD2{@DIQ_`
zLlQthSwsj7n;issiv9HsJc>l6!OtecX4X2<@3F$Oo(BmOalie3jHxe=wvyZ8Cen3`
zu4*QAZg;Nc#fTlmni-G`u6_pv^8jk(Wav5>4APP
z9*n{Gv|fUkONhy?+IJ9x1WV|2^}2G$6@hIAN|j(9!h8$@&?}I@8`8n)Q@TZ@EhU2B
zf69xQmZY)*(tm9RPK+FSVJIZP-1px~SL2UI3?%45$*@2@NLL&b+a>y~!pZ;k+hQ02
z>cTO1u)%`t8=w6EsK|j+fJkHrlq`%O=&mGsapPH$IX87uPSsg5%E;NkXW`>HXZSXz
zUJFI*8d{mgo)qD=y)G_Pp->SGE_WGd*XV?sUTs=DwD0ov{^4;RYRAr;NuHK|EOG$wZ5({_m$moc5
zp$byCPlx-jCml>6iqv-a@TM!07m-z~)#sx={&kK+df(grmGe6+zn~*j1#CsjZoR*e
zYQq6$J&dr<`+tx|H@flJoujrTjeX7^e}fLd`s=GZs(uk;A0{L!*@-pW@aAk?Af|%k
z{5C;LhL(Um*azDsxYQqjjn*4NZ+!K2gzXC}?8c4vrgk2U|N4=)MI7Mr&BZ7*h|;Qx
zwZqH`c?8QSfBAU8HOpntZ8!eZV-wz)wzmwFoQe=cwQi4A{@WuI)-BoB|Nb(HkMA4-
zP4ri;OHtWF_wN-W>347O*Hmx5G9c8y;w~I0-Ct*==_+9lIf2QN+$ed?us93TtwqY-
zUb6Q>+9~4WfB&_&zo9X~)X#M^>hXawCi@>qxG(uYv(m`nsCA~wic)ZrXZU;Daoz*^
z+)7h1(D-&}F6&Lvwvu64?9wP;Axz@3cE^yYMV5#LWO%0WQJGGR&?TLJ^w$YKK#_f
zD%wK)eV;Y=nPuElO22H5bOmp5)$9l(az>)6YMI^
zF&To0)Sm$t{ds)5Xrxc4+cmbeTJHb%G)b*Mg1yv_k;wQ=n~CK5nFI{83uJ1Vbz5u8
z(S)#s`1}u(L*V5KuQAKGyWk&>OF+%vQGV3dKvIah4aR;mN
ztJ}aRNQ(;shYBeAUB+Qku&AF7_7$3sGE2ttuvu}vna6^Ziucq
zhRh#=7?Fq~5s?U-(eW{jN`cU%B(Io$xmz!OOKk%agyZ=$cRp7EYhJ00uXvZQ#LIs>
zeZ4cgc7ERn*;GWo
zp}d75$*dQpos}A_EC5ioh3lmMpU36qZj~DaTORC=2NNgL{|-*Gx;J0S`v!w9uC|QFZHj&-uI%!~dr$0aF!Q|;4(ntLKd(59
zHdJD6D4jMrnRz;euT-HJ(r;>?@POl-X>Z^<|JqWZQSaG@f91KSm)vcAQbmL?))T`0
zhASzQ0;2{72N$1Qn&FLt0xKjbxeg?^yG{y>1U5BzkW@q@3j#oq9@ur(kwPnjks43Z
z1CgB#f#~23h6qIa4T1{hCQKkz@G=)zL~MkRlmH`$tb~RlP6Wcrmq@V}L7}oqGL0Jo
zK1@xM<%KK=HnJAhWGJd5V$wSw&;o_QB=mdpMnH@cA#kV=6ugWXgQ@>rDF#OfBOaVU
z7@J}lwvQJSW)^^|YglZlgK(pltDdAAE=U*qp#Z%MqJy=XuWC=AgMNLbU@RYcF;N;;cp^4{{$cd(
z-?Qc9MXcBNz2`QtJFD<+8TfP;sJ+LOXHd|y3;-(Blw&&_mI)AjptI;8{>lt>pCiz&
zjgD!XoT?D82seGcAPQ1+s0c$`K`D$IF!DOng-)@l4~w
zgP|p{D77_JEi%S0ef=?}A(wAYV|vHNA4dLejc2K*<|L4!0GnrlW=Qxw(L1J!WSY+d
z!n(A;)^6tW)T=lzT*ZMDCTZ$49k}QXXoE*J@Ca<$n=no%DsELUs(d!P~Lo%fr;hW
zzIP}oDKIpZVYeq_^?e!Zt|x&65g2>EsK-sq9IAYK=tdSKO(N@
z9yyDBGY9A8mNlCITdM52Di@tLkQB(j$2&6cnPF*nb89?5bD4ZwKjr~EI^Sv{d9SSb
zBA)QA>u6(N3A`1sgd^iIy%RLBO&ssag9$$D&>5*8+X+COzx8?Pvt=_h`=>1Ia@eu;
z&_AgD(#)JPyZ69t_W^lztH%lE_`>?-l8D}mB$$K}d$4<(m(AUMGoUZYdn1m!6d8TX
zL%0J|8d)C>saL^awJHs`^})&^-I>3A`tC_4Es&M@6Hdoc%i}mJ@N_*K0yjr{H_;cG
zESP>TLTRm*%`tpufS>z)*l@IMY#GbhNyVto;yB6;8hrB9Esw$PMlYA@f})V_#t}-1
z2-JZ1`^tuCUiu%$edW}KX2<_l7TN11&E4btDzw7
z=~gAf5m#v#?Ir_JCqtz&HYG7*G#;DVQ;2T~oI6h{gy7L5Yia1#5Zl!sMl6$#bdlWV
z8ZQNLqbL~i3XPX#oXohq<1?DZGlOh@x;v}2!T2e2EE;G1E3}Sq1F+cH7A11h>)55m!RYR|R98a2V>iTUBzN{F8w6A5>8Tq)C1W|-?7vtc~+7{j*;r1hwK
zauUf1K1?iI!WA}*-oQdI+o(U4nm32cL_GP*(Om<2ESu~|9%Ifkrl^%)5<8XcNFNt+#3L}?SX}BNMuc6J&v{#+FbXS{`aeG3`WFR30SuakDoUpz7$q5Q*OjP!`
zL3OxA1u4!yo2>?mlO!|#(>ZDhM9-2Hm2%+P&ZLeOoW}qy8DtPxg$q0=PZMH*w
zQT@HOFJ40lFV{mpFU>heyQabnAupdWm)+J)UUrA9x|{6sNBa>lsY!#`Us&!?rC{R$G)7oL`4x^J2(f
zKDr0QRqF;*&Y9Yn_t%Q$+PDFkbb|TEJ3XLBIDW;|harC~PqzGZ$s6oN^EGmv!s{$*
zry1tzR5()?*j!kA==SOKEfwIr#i`6YMdx@|0dyt#ZzSZ7bjM~2cI%sb2S?{5ckp$m
zEZ)ZExl{h?jo&!GOq(w&QZedGT2d`I!8>1%Ssrfu9^nsqDLq;FC%csX!7`PfW|Gfb%B}}TI8DtMCE5ECy@lGI_wi6;8G2|fK4C6fLrno{m
ztz8hZuJ13Byj53a+m4{glY3-s<8eS+PxtUzebngexp3OEI1SC{QKB1zm)tABvw}CU
z`>g0V+PP{VH9OF`~+?AtTSM8PQg@ZicnO=jw><2JWIT;?h
zOEvZ4iX
zbL8voeWX4*om1&%b9Yoav)$!dQii(f6Qs?4`)mKGBze(ko8f#SEibu6g*hTP8ATbu
zhbT4}yBUr0yaZ|Z9+Py(^{ePJOq+|VfQlZga9vlr>uuun)wKR!_Y?r*8Ly_{cP4b^
zcCF!Pky*!YJzW}lrR~ZQPy0xCR&o$GC7J6$j_ekWn(0#OrRt2?co5-TMDnt7R{BKb1F>x$@$h+7D
zP{s=hk;O}|MdByVx(*2?V&#A@Y6*6UEl}Y3xoZz8hy{^xY=GgU
z<7fUp^+K9F4CPRnZ-?J>3fWBnqLI}L-r59Tp(dIBFdb&j&iowc-JJar{RlE*
z2AJzt&0Dw1{{5v9F84&2euW!%WWRdK{yJFxM#X&;r|aH+`bk+zhXW7Mx{a0IUKfgf
zuURI{lvnL)jX5yxZg+AR_5QE03du3IbGT%6i;d>D+SNS-+V$d9s8>TCw;VIgc*a*uzuSB~}`@!8$J;l$tj*{?b=*O>#(cc?cpW>_CI)*C*#
z`@*hZd1$t+YleB1rTvS!X-Q#uLSym#tUqE_$=_AaVk-%Da&OJNQzhrsZY>?zSleTm
z-qL^BQvCr$+JDr*!&D=aL@hDow2Z%$7bexPx%wO$nC%os6E;?eCd4L1uOes06Xcq;
zt*0=cBDb^WNG)F4C&plsWn<|;c6Jl(m__@mu9!vW0u%Pah<0j^Q?p%Deq7{cx_dW9
zVS*G)r?l^}b8gYL{?UyOjY8T;lfqvPcbuAv&PV`)$}poJsRn1Ik-wY26!lLrBk$PR
zA4*fJZ&?=O36GF`-DrY2e$*ZL6=Mk4Ol*P#!UsuhnT;_~-DO+W7EUR-tcPjb$^kPy
zmd~nkRiO0UOB&<1fMGn(}I_DK#_ANw;rAGYM!FuVOXmyA4!djzX#~sn&?v
z8*G3H>*&5k-LD{uSyM>MN=nt1N#jS4`Y9A>iQn2Iu;T?7e+OeUTxZ}@OVQzvZi
z>xwRkZ^Jumau|bm9$+yun-|+0O-+1H5{KcpHG5D2_eUACJ#EIiz+#{mtMWxHanty5
zHO$!mBMSDKL0g2LW*l0XHV$V!gI<8rSOdS{C&Lmu&UfZ^=G&yl%D+0hlNZef>33iQ
zf5=vFJYD9!m=AXubw|q>{GwVrv0|GIdzmBl@T6BP<)c`+ES^BXHgI(tl@u0o-Pk)ZYjF?yr(X24PoRM2DTM3gj-c;m0XBpe2
z8qkb7h1Zd2H{5fnB^v%zgkc0G+-)&fa-8^%{qG+cGUMLO`d;y4sK}$ho_`&bo?9=_
z@FG?8ftetwKF5|S2gi(i{GP!G)*%YL!PgcVO@!Ll=Zif$j>%&=Ur_~mZ9*B2hmZ$)
zVtU!EF!Ie~kV8Bk|1hUY;4?r3S)>n@d^E*;-sU%mO3}(D%P%7%R#|8kc~Y80M9UcD
z#f~li<*)kqVCY+{`iW-cRUKLRv$L&r>XYF;j{PKOR&OS>Q%k&-twHXXI!vk3c$O!3
z>br*a={m0FErFJtmHoL?O|gPj@r`Z&8$ru@W<(K1@}~m
zocfCd>aX~pG)YjOFHOfh_on3+v5d5C*pu$$su8@>a6O8Thqq8r-S&hdHm9m}`yoBp
zlWZT4R~Q;2+b!f$`nu6n#1S<7LuMNxUVcY!z;G#axvMdKDpeO4n^^FWH4i{hKJh%C
zyqh3s6Z%sDwwH1DaW@M+9BiEPoqkDpjPCGn<8C~Xg2=e-pFC|vgI_MzXm;f_x|)@V
zjUrOev!%^UBixB44F)s_aalewo>qF{j@O}9QvVORF`YfZCM^EMM(*`=j(N1+Qs3N^
zPILah{Csc!;$*vl5O%kpq_R9ek*x#7$twUq4
zpcLgj29l#-Ew0WdqDL_~@hUcrz@EIvN<8xi?2~sW%HscW7oYbauv-HDiLza+{i{BW
z0}=-n+i?mAijX@akODd0<EC9Eb*hyRht^_I$Z`vRcwz|EX&L-XqTWI#P#ahE&efr
z^oA|;KM*uOyTUt%r55{4KIjYJGiR)Qb&nl~
z%$B_>*N+BV+(&!bq|>+MdkJnQMMiC(S}Zr$z=j1WwIyFY)czft?xUcN!6x%W-z7p+
z9;-Z-mTnWGhH{dPWf)oi1n{kENE|mb8Btj#c|!0dPlYHqOie~X*n69qD4A7&>iyiu
z@kevn(=><3-M*#4Yz|mg1k`$^alX!4#($rqm2|`Jt>NK?jHh?-Hy(aMx)$UKhnSBtn?t9^
zQ;Q8RTLnM9#o`gC{-xW2R|l6cBMnX#(`!slO>pOz*uOa>ze@`L2xzyglTlw=Dl>6^
zeysz~9GgnV^kL(FaJ<*^UFhP)B~|mCj6t_$8ln{OfJ(~lz8QnayD%rT^05pvK9o8#*A!u+CCs%%Ha&*Wyq*Y!%B)qIX$Xu
z@|3fd>m3cOULJHZE;30EM?V?CBRrpfrnRcl7=F7KZlEL_wL9mC37V4_!z#Ol8by$iW!my
zkwx4@Cit}BPKk8mX|cX9`!O=tR%4f4w(e-Uc!|AA2n2iGrzXFZg@D_Z=
zxb||UB?GRYg(`{aU=2Wmbx$|Hh_?*7idPaUM|a*8c0FsK7~FjnT)hD0b|*!1*nFvN78Y$!q`rvLL0
zBy1eaEdMJXrgXI(@ww3gR%%a1VX>6&1p@oQocXv%oRge$Y=ilz6-Jli?7S(wN7f8`
zAJv3|w}|ZfQWQb8j8let{2MQ?kka+6%1gfuf&S$muRYGx}5`(T-LgTFT)Bal0TN@Y#
zGtyT{%4W?bJF0)w36mTm)%!3?TKejVR~t6VZ!;XOniKK4;H$EMinN0*Z}`Y0Y1JN8
ziWEbDUROy&NeuIpD{Dwkm@r8X0e7DjfB;uOiFk^<)($KAtycI|uA+Zoo;Ot9H91PA
zQa7n|o{xf=g^FxT)TkalmH$g4Y6^3gRx>`>0>^HZjI6KtEndE*91v4Ucaq6KQ5yy%
zlKEbMW-9<9i=T|fk^P&J8GWYlehc~6{?E8S0JZ+
zcxWLG-0*B{QDx-QvE9_b%gMK^yakV8gA7UMBaCc7=ntYHLOn^iMYN}ML%y}if;$>J
zHPE+NsKgUoc-+h+@!=Ic`BNFZj4*F+kg82(ILD0EK)tr~m^qihwx7iFD#uN%dY`D%
zkHUh7vvaSg`2@P8s(|&RNK@l@xm>^JhdocrzsgF7aw@$y(d+O&=Igp8QNT*&{Z7J~
zt@P21=e~lYjv2t+ogO@NQ+S|6>A&6vALbFc(Y56$WTi=v-yZrh(xWJ=0PFc^iVtb9
zca^&BN$577$mz<&g%EFIQTQT0pZC8&ZhHm(ue^reUx%%Eo2s8Jwy7E99g6fMd|g0(
zr#2Cd6ng0!dF0F=T`?KI&<*iKIu>{-j_MxxdTPv}rzz@OOGv)Hh#-ru2OGMT1
z&CZL`EO-qw%UnuN+84Zv8raXCe{S1uQN>>cV<3;(7(#gR>yGa897?CZd}GJ00I{o=
z8|(Xh(VL|%I6*NG1rSnSv6;FvGJhh-S7WjKA}ztPmXra`oTFOeVY>w>o=@T+OcJX8q>}x
zA?&QskFoE>uSnHzgt|3pu_92b4Olnv?bO*d9~fPDz#!+vaGkYyMipIkBKq-}*edb{*TO?7_mMQX2Hk8hrm#NAzW@f-YUYY1rr(@%ZNX~Io3Sewx
z6(g7=yO(53A@NZ(kvCg-{>B@b0;*sr&TLCbQ;3OSpJJbaI>2f*!lC=PtOkaYOL=q^
z=;3jA12g9gvyUFYgese6WQGp;*in3uiPkeqp9D^L*;w{a#4lAkgLgS`cpOEukgwB%
zi*8by92$mWuJAr~;w+{n4d|0B(*z5;-Bk>dF-rrTbrE$zh{o9)`4A(@(n-}pOoKUy
zeX>Y#oZx(W3e25eA=aWD?}Xpu(Z)^P?buV@!M45X`Yl&8PCfI@nx)FGO7kj%o6IT>
zE{^XjI`*EZN8BU8ceMxZMIyeerb&F0bp3KF3PZ0`k+F^N6RvrGbYN4gqVaxyK
zh6O3;ZsT^6IWYf3$>&Xj1bF9*(uxVeTa6_V5xnT4TYzCLYW9dCrE1wX8Ms~3tX{hC
z{;c2SJW~Q2)8iea;JVX$ebd7h{MNvHPOfd93Q9|2R*a-(R^Z~Rw(znEkvCxskZ+=onGsO!I9`#Dk%vW1VQA#PKm8NxqehzN!XVpChi)LE~SM8EjOgwUQ)Sq#%>z
z9sB2Fp^e`Vd#
z4j5^Mg96Tln;f`I8SI{}wVMRW{LxbPwzK!2G~dyjI^0%+B7V4SpFR2_0kp_EEkM=M0wQvbI(Q~qx8!o%mo;KZ_jBlmr|QD+ob}z+wRwfw^S^t9?D@A
zazjc{820L@(W|7l8fhtc-1|`EL*e#bTPvOjHyplP`2hj_t2iqD>Kv2(mZCU964)m&
zT24nt$DRaRQEttE?8h=T&rd5+DJQ3a_J9)^;S8us_Jl*-^XW0(sO#
zk#vQCblZeN#Qe7WWOkp=h7{J1(4zbTN2;IJmeSC}^dgg0oCCVA<>y_Pa!P5sEQ~P>
z<{R&*n=Kk?Gt`2cleRuX-8+0{wqUF^y6*2_KRzK)vDX;=g2Kk?1AL*NXhKzbBXODC*EUJ#lI_^fd_tOi$O3sswACQdQ0@<$ENiHnfSiuwaSdqkT?M2{gq|C
ztPEDQIK@TT)}WHZky57R4j-;}639<7J0*vJI(7t(+Qhk1lm;SlP?YNOfGNI;f#f2P
zXg#hx9?cSVwRRoSawZ5pu$9x@Ef+!r8MnxIjW&ox#vQZb;>_~W>Nq&Wb$PSs2Bv8e
z8NmQNj<#x+BKoEn0#>r$))pP&ACo&anAlOnibp+Edvg~HK4tzkslRN4--s0Xj3aOR
zFB#If;_1`{k{HcylQiz9!#LOn^))Rst#fTL1^(jW=biP@#i$o)`I3C7U3TeQ;`-B9
ze=aPi)`c5moog~%w8v#F#*p6hhxCcr4pXAtSRSR5xA!rc-RgdWa;H8bp2=xfv_^%bBy%#0JudA
zGoo+t^_5ZJM}PxvUekO?BV?rOdm~nY(WFxMWxYCYFK1bcIzQu|Hl0JaZBMF&`un*n
zD^N#vJd3O9y(?5r%sIjIvF-VZCIKHm_tZ{Stgz>x5&FJhe6OL$*85_A%UwqkTc9ITckeW4s`Elx
zEG`*43*pGqqk5)}h`eZfc#;(I1{Jx8?Cr)sInf`}P*QjKKP8Yx%5PuxgzMqo3>4M~
zD(`C1PJy~L;O4_%>k-bSvz8Cwrj${q6vrBWy(hyiR{W7d*oCT(G2SpYLh-Jta@ZNs
zY~f>=lWMchJy9O(YVcI?P)M1&hmaNK;vN~X2T{T}SrCocsrm>r#JUqPd9jz~RJ&O&
zRhNWe=`TlZ&9Io;V?Qh6{)m|Am^%|e;<64+*#rf}gptRK^b5til-WNWdW$0$9#xIa
z2vJvn=GqXjH#@iI>Zx!er9XV?0Hu(-=
zjKG?89lbQqKZh3nT)$T6j|^Q};$CPtv{Z6P7NpiK(J$=Tfj9v%U^ow7Xm{EnWsW6T
zZF;z>7{y>)5q@GP5ACTL4lRn=GY_FoO584PAmk-@QOVh|Zg<}aYwz~bz~(YuV-`Tf
zzT&yn9rO&Uqe5mbeKDV7dpb9aF^-JocMPEAJ^L4qCx&h%e+R1h*OLnEMLO+_)Z{}C
z$eqW#_0LC2TV{6spPu%cf2O^9)O##5^qWWcO`Ey`n=r((n~PU4znY2pJzU)Aa~^DY
zp28-Zylq
z+D?YGD{z%EBdg_C=wCE%bg{kG`Az~eanznlA%%x4N6a`7$*QLgiZ}}IIM~nR=SNv1
zd!0a_nJQSeBX(0(_TE8->6o`;ey^#88Rl)`+%YmW*ZAPZP@rCMnO`gTLdyz#`}wmpc9T$
za*NY}=lkmf?)A>+GYc-07Fm&T3cf(h{l(zNuf@uqf}5W2*B+bI6HGb7+R>7Y>tS#3
zB&={8MZ9b_w|3XqxZaAElU$+8uVIGcuTxP%J$t9W85RA>PkT8-kTs4`?LKeG5JB2P
z2(Nh|8pZF#snJyH&>FjuBh)pT+;@JDzQ}8LBu=fZ&a~kSPQ{`Q92Hf|%g*M1fV~H*
zn4=&x(;SB7=$IwUU12c<~hhziW>QFRA@T#+Cp;SOUMfK}W`ZF~o
zZ|K>%?P<%n$v^%Q&md}5=dpda^F%`52Q}e=h}o2p2ZO&`;P#&z7p|Z#lG0mH2DG$V
z$lF!*{dRiARB6E*vGQ@dstb25Tvq#G+M;GHQ5sVXs8~0}SVVf^DMHp`7*>K3o%^|G
zVROiQb{?VTeSdYfU2!w9($>KVD2htHYe02|+PTq`ug0~7vS~#h(73aT812WTPN(^@
ztOi`2-!7eUFKexbbSa=V41U$wns|32eK@wVYb7L1uFJALa{Bq2Zt|a?jBexXxNbad
zHM^zFX6K#vq__FIEQG@AK$vURjgx_a5lcw;*cj3Ax6x{7wOlpDmIFU&GYXvheV!GR
z@{Lg0&Jq0bouZU5628`9!ac!M{A}aPv
z#C$leO-8P$9Z%V{ONWBp@sEF;n5z&fMz{R>#ii*Jb(M4Bx3W5kU9c!UwSgU{H@UJU
z3>H2(;I`Lr<=WPLoe6q>!HCqOy56_>C9Q;P{*C@ApjwKFecENo8s<{s0m0z9GyASw
z?9=5R5zS44A-VE=eRGN2M#
z=S~yn)T(B%sakVMB6tBg32h!Li6x#N%@T6~c37znqC5N*hZ(G757*T1es7p}74V~}
zbW$tP|A1XliD-##a_1hO-vs%2NLQX`=!DAW4Ti{e4;0)ZqqZWJeNwv{hjZtF9%S=m
zIub#QD>W|^;4#I9J>m<_7S(WwQ~(XmKdyZ3!yC)
z@pBn#JeMyx9*AAjou7wJAaM2K55$F5{Mp)vKOBy}@PR!`yTG}*__pr=V?H0OF$G{7
zG$H}BCJ3!*VRhm)tj3jJwhxLE>R|4tz)#yT){}$Rc=XGvB)^eya9a~ejY1NWx53o)wK(S?`y-t*OTD>WUA{^`DIu38CSfjtp%b#7fC8B
z!Gd31-o!JJe-4JrAK?(3+2S${7V;rmqY@S(0P)9(w;1HVDKGGy-+ETNkO}EtRYJ!%
z7@iddn1{GvB;F)0ShfDKj4+!m^AB|DHy_}layM?RfVya1Q69eRzzM$EnOJ}G`iy~LwscV{Qo0bS^L
z0y>5yeGq{>P<_}BYbr@>BZhn(o^^fS-f^AZEqZc2;ap6=U>C2b7I~_>qmW=|MVM25
z?fvJ&YVY=MRR|FYHJJ`AC#b5nG$={7LUwbv@mCJLZ8IND!v%Qk?X9f=t+^D9
z`J39>Qq3PB)k#i2MI6ty2CCcTHnFmZvGgaK@}RN-c~EcB-zecffrd1!l>c!-#4dYgx@nkI~T_C%@1-3wEM0CWaU_AFudawK);gOPw-fF6DUzhaDBr
z4gc1g*eh2g?G>15vp&}xs1}7N5$Ru!wrQSgz?RqCvPW1(kP%aRyt^*%KH`y{w*P)y
zYo{4J&i#n^Ld!B?#>91+K4M_}B;s;+ITkPf{jX&W7*@m<@N1?)f7G;W
zyO7^~=jXpFb`uG4yPTHjrb*)yDt^L`w&4FiI0g`y=YMkQ>ALmkcl46V^S9w9klDPB2YFt0R^
zz>BNQCjjD?m6Zd7WaQ-dr1*uzY5)H%crT6qMZtq#TK%^#&6FYvjngEHnDFwPpaCR%
zU?wTdbG`vC
zE{lD3A8#b~j-Fws4ZpM9o;%#ysE&|C(*fDHf*<4wT;@*S>ITt{&n*^XuzALX@@-IL
zpz2FuW?P$N!E{iyH$-yhf@F;Q@kO19OHnI6xF+eK%?CB*WdRAd25DB{oU{yq7lkI(
zD8GBqYXP(M+wbfM4=7tuxq?Pk7jX_H2}?;OULJoG2Cm_a#QY(hv>Z>S?VL;KWG3eB
zJ#x+1{Gn^+_(DjAeG~Ad?*Zl#wrFd4st=)kC)B8sFiBA%9C%P+9*TPiS5A!zdl2Sj_BeMxu;~Td6ily+xY~kQ)H95n_O0qbPsBsS9_0ew
z++HLL=77wIltkkV__db_YH!i#P(lR#%2v(D$Ea>k2ZU-
zm)|U!N14HJBXT2Z<;Q*_lUZaLZ5a)+lqI$vkAVyyHKA?9PDj$m)5jO2d`I@z{C7>^
zohzFiO^NhBXEo0VM!h2axL=V|D8Dk+
z91l|4iD=;?(nk062=HR~uwZ|1k#(r54Jmo-ijuXAohqL=$7QI+;_C5~LGQ1sI-}VOKE+
zKQ_hHz3?@}$E2!|!E90@lZh_y(UJ_eQNiZ_cBhzkvr$wv7@Ih^C5WEN^0G@BT=_oS
zb?-bfwZ6E%_?(5@T47~)x>~@!h_x0=>)o1et1m741LT`4cAj0l+tpjg#S67HL!(N&
zAEVKZRz|lqmK>7yziEBsiJc>~=2R1ocXx*(re)N9Yd3Q%Ue?v*xRJ_u1O}<+=V1Fm
z-a{}vPn~Jk^%sYvD7aQEui4}b1{*Ri8hW%qaJR>z6y?p%8ty55d65+!-_Tss92jVq
zK%x{p0Aip8Q;nx+KC@VKIbjpt!8J=}j-?>yXiRMa)z?m(XCwCx#!cmg&g3$tjnDSm
zLhJeby{4ks%wSH;!*(bc)y_rx0*
zlzX)S0Ga~NOy0aLrku||tnPJF+gsdD3>6Y@b@xxhc$jrzFSHb=Qus(Xar&effL#bM
znYu!QLmVV;*fo-uC>u$uAll}3o+#V?K{GtPUWv+=EiU>Z*t>gd(|S?>mbp*HcN--S
z0QjEr5ARPQ7Xa;Cn
z{Da}s98dNEd2*PfuF{`UxE>ssHPp69{%hCzz9Yq_G;x9#x)gD9&V=Fz6|8IhuLa&`fQ~YC
zO$SuReHXb$x54Hnu2k11rt{7Mx`N5c7FN*E9wb_K5XGTHT0#u&RZPvT5aoIaEO{i$
zR_=4X^!)7r&qT2Fv^!;Vem*3P9_}4~SsgpG7e%7lo;3=?AqROm9VP|in%`h2ofq48
z>`$Xu{l0PwH&wuhui{B@0BE2(0ga-264lRf|1m)5@HEarW#pO^S)cf|MoaIURa`k`(z%kZZnmgXa
zpdEkSdYR6n*+)M@^2*OJVb#lgzt@3nm|^=lF(}n~Yrw3kb4Y)|5BZ)1(BLk+oL*x_
zd-i&QC4E9LoEvDKO`!#U)K45JV|vhy^aj}`-vrrvG+KOdL&ZbtjPi9OyA@RbAzFK6
zw9eV3pMxQ%ll^w9?!r+Alq%HX-|EOgnfz>-Z0JxyJB)XBJD#I}jr!rs-J(+Oe*jRX7m?BB&k3=@)6|_?pMz
zzuvTa)L^ksJTEAnlaAnI=m|u?Zz%6~TxXm#)6stop!pOE{YVCXF$_^NyhmpSJT;Q1
ze#si?ruci@c2MW^e$#+^-VHYjs^2*9u*>uV)?`Gryd3w*tP))|zE*37B1HYdI|HI+ebf
z>nt-ZJN$$mg5-HD>NmgdUpvk>Lz3#X2xeCL!2DYkIe+Ci2(4X88mf@zlB7jK5O{2YSrsnkfTodJP8AiI#mY|hc;y*Z(dCT-!e>kpH
z=`+VXTYyy$@8oG)2Y~BEn5`1&fEkty-Y-Al$0nnCrf~&j
zbt7xkbe@k@`56noQ)vk;n+0sIcUAKy7Z;w#!G?dngqZ~%^mJ6ahaaSMB(xszsTIvE
z(;U<;6lH$I@{J$tF3{$dA579f5zx2|qk~rD*@ZRZv4*>Q=}3NYlK2m?FGu(`O@PVf
z41lR1AsbrtFF@3MBu2tErOc7h>iDR4abksyN4A%3DMK=cw5s>M`kBNxysAokfkLsc
z=`YU2-iqc=?K-~EU%o1hxlyQfzIPx=+l-RUNH_BN$3{7Wums5btysw!vR~JYu(%aa
z*BLi>U<2Da2(_)PY_REmb*HOAT*D`*Qo!S6F%F-D3$h8iGctpnY4~&;=z#S`y@>ku
zk-bOQyO~S0n=7Rw{nusZ>1$YS_L}*Gv5}>d{9G_BX+}pWXt4n?0tNCRl4Gu|dEy!C
z2qYal@wD^T{;`BGNN)QPa`X%ke3rqw&v0!3EvH--3A;ox1p;}vX1$wJ|AEXo0)RDw
zlx!##VbJY7!_kD;n2jVQjNwi&!$gNcFnZ>AZ1mptsN;2i#5yTcrrGDQ{xr;3s!VOpbpjP1KEhYt(a=kYdw
z;61Epp+Gv;pWOQXnp=cv!FUH{3rPJ-x_>z@zNYe`0Rgl|g320rB_Q`!9D0eB{;1O(
zkcS5{{58CGgp&n_(LX$PiliDi4_3Ia4I@r2Iz%mE^Q15R;Puo&Tl)xR5`2gE%28{#
z*Ghp#-0y2;T#RGYIn~bc!F)nAXWS)YW8)I}fFBOVt~WyT#CEL_#WBu>chd^?FQEut6h$we=CQ8UAkJGAgGir)WOveQG^+kv4V%Lo-~-CATc(x
z`^sm1-1mpK`#xIhFDOcrBVd5BwyW7-mxtqk30ABB=}NNC&okF^gc*AqSI<;=Xl=k;
zEdxWxwmUK@bBX=J#Nnay1H2ndd%B6Ac+e1ljZ
z+x}X3?!KIIW`b1Yn}zHMc0oWoa3l_Em_&Cnm(1wULDy3{o`a9^bqHK45GShpwF9n*|A}0
zGX%Q=o1zLUx?Rpl6aX6TCDva0DyOxl!PwRTIy>7=3u5J86QG+e|jb;AB+6TF;}EXaoQM!4voKKjmdFEL>2_1}iqNk~CY>BsFG
zibCq3ceEED<8}jmP8Z-R^u7+^VVtg_TD1DIEHv<04V>UVXZ}N9Jib8s-I6p
zRvC>(7}GQ=;pN-K57Wi*@<|>q6aC5_)KbV-o@x9Mm9|ptUs0EJwI4`RFNc=1j{`+7
zMTEHFI3}_{0UM=?ifivhy
z!_TUt?9y1cw0=3Vi6|vJ5m~2h5s5Sxgs?Wl3zE?CD^k%h$mk&2y)v0xMXBk|E<^D=
zceaDj{2X_qQMk~fMpAqO1l9V`VMxhp!JJzCp2VY;r^h8#qv5cG?+KVCcliwH;}&bq
zx!~hn0K4u&Uo-e&Z+b%mqcPsLVQT|1@Igyuyy%YipaJ9bEmFj~9yB@s+ycpi1E(mH
zaI2dIOuT)Qa942I-LM$Uc_xD;JAZKL=UNNK%6=yz^Ngqk&2xJl$<#~GHD+-xqaxi
z_lC9zZM9TJx?w>?!hKO@@hlS6yUodZc8|Zy#-=|xEYq_u#m-704PjZv9GtKG
zWe)OqR
z;ziNjmv*a0DvFn)5m&$iN&g49v*YkY^HIcC|B@x-;`HVc*$7{;!4pUS(=}rZwrxJU
z|6yyrr%JzG$^LmUeJn0rX_*n*?fu4+9=HGAhw2Zh?fUZ5SIUP@fj!EFJjdF2Yz6UO
z5ieb72=xu~rM%p|Z%pn(yW_UDLJ&U~5O%J#zzh&-K#sO-+!hx~?`z$>-HK4=f$kz0
zH<=L2xB&+5QWSLjbfDctE7FED88?5fVV^pkO6s<=2NPyg9Eo;Z9YL{IqqOJ4B|PHZ
z>nmrSU?d&2mT`&v&02&}_8HR{Y64V=migmwX0)_f;-~%tIezQPb&d)NzpJ`-U1;ad
zUSx9_puuhJrkp*IzkQ~ZX2?hr*bY6M@Uj$|t<8f)@f#%9{&J?3dttcpy0g+rRK0PF
zw?_P<@4NFl^oSrdfI{)8vJ
zZ=nGn&emB`#qW1GE&sC;B0EgrkQr>Zh1Jy!U@H(u{$y~lprj6#Eas8}o#U}tKa83Y
z%tHJ}4CajnTMZFLFmX_r+e;_xHct1C>e>nde8YoI@scBom?VsLz`3wJ-i!0uzX%yQs)S$Jml7kIsQwC2#jeR~KUB32H&A#I
zfP%ye2^)(F6+(gtKdl|wjOx=^n1A?oXXvSrdVVzR=A@KvF-<*)^%-O>q^l>JzRUn@
zX-I@tkT}hC;M%A|KJaSz#e9LwG8IF-#t_gl+^he$@?dWe`lysEXD*O;|KNllw9oJ9
zONmGRbTOd=Y)-5G3gR+g&33bvpp|zMV1C++nYl*7jG=Xws@412JFb2)%J17&{h_W2
zc_w<|oMYb;rnaoz!LOW)k0!ALOobqipXTWkEp}aBa#QK#X9zfvsncTuZLCME>a;#m
z1i?wyRq0|5GLwjh3#;|BtknNS)$x^txxCyN(1qHn`q1Uy1`h4yk5L0~3tk5TV5hiM
zy|fT7V8nKY32Ypq{H+2x)lWkIV8(tn^9qRuy+c59
z>|eg2p1Nq3R3>_U-;ysELrW^WT71A!8>LJOD7hGr`216D!!vD{=R?
z7>mZEV@0qz-&&KS_H<*zDcZt!FeGnuz)Wj+Xl&>Ad`CIaxCE0h(~Yl_t?LKEkqBU0
zU5wDC3gnB`mU)a50H+s91bAbIbD0RHn>9nK&I`a)yz)ga7MFT-B{7Ca1T9`h>o34`
z4-nh$M6D6|@dFCvSRLisBM2LEg2K5cBKmu_BrChG<}wqF2nBqdtBv6`3x2`vlL8UX8S7D#{Xq(f$E+?O*2etCXugo*)Q8Odxr
z59B$S%Q2&4&2QW2#C)Ql`>^CNe9U$VZizO8d0HgKhTLC&$9dl-&M>B!VnFOb<&Xso
z#FZ#w2s&5-9Q%wMM=}oD8^W%q98(5aCLYCO^gu(C_r*14*WPR>w~@F0gzL`V^vGuI
zFLGfwhCy$Dx}@#@1H>8wZ_8fHh>`5Wl+fpQzJdR5hFiPIRlg
zTt7}ShLYLj&P{r6>KxN^g1H^xaR;-BsqRAT*q5|f3f?<%Qoq36gJcH=HI~+4Td0PY
z{%M2-7;#6<2U)-Z<+75WNRieSA1ZHf`~_K>PQgGqlN=X@@1@LmT1jt>goq;@O=#2RbrP<=X!}!uVKriCb)P2*hQTz95oZgtlj-f85_Ywu@xMq<=v4y
zY8vCw^Bd06%;r-r4lMNaV3QGz?~ANJ~$-I0cdrWB=o4q=-4
z4s<9E97@5>E3KT*6vEwWXtfv^4Z;sKci>{ubkpCW$I^CgKZe462Wccv5=mv{S#0+Q
z7!$_l8__fop@PAV|2cA3T0KJ|RyLrzBTx2DHS
z>3V;lRFp{{_AMZK5%6Bb7bE`-UM1_>`%23lv9M>q88(LN{Tcoiah=kxwGBEuajA?t
z*O2-~EX-PDWp_$7#-hCa(9J!8_x~z~#I99BC+N#`6e-EF2$#5qUa163_ZMDNaIwaTJ4pca
z8ifa%Cox76Fe%0<>y==y`r#?_WPD?@718*dpUejki4Bh9Yh(IoB&!8RWy>Q0fcSRjPT;Wk)Q^cSs8%Qk>CcFU6aWoXl?k{kJpIjRAB^YWOVJpZFc-n40
z1Y8ULcBQU4hci2kUODeL~Ddf
zVQTSJdtdL_G#qszfwTgSZPZJ|(1G6Ux9?dzg$K0!uHpQ~IIOvFhabVcR8Pm_9UVvJ=)~Fu9y|;f&=F1c2kq&)7?cE;Y!)YU@P)
zEQ}M@evwj;q0LP5RQ-L<+2_pSp)x2i|KCy2Rx{P@jg!~fN1c8Qo9}Dv7IRA=)Bt;d
zh+>2%9B!jn(lk^QE}U@?rL2T01;2xt7WwE{>>DZxsb0jtg*r@Gd$mc0gkrLZ^6cxa
z4}8^xIEST3Edc+Y<3-v1hN~Mz{q%tlTZwb-8#1%t1}=}MVY5Ullc(se+d2KwgT`n@
z&YZ*ouoh?H-a8E8h*jIph`dg5Dif-_y++f`;<~$n@1du-e6azEFegXc$^$+NwLTS@
zxdbaLsq9b;yuMi{}34nefBUsGB`PNSPP1wWG_EUKm
z{O344QVz)&>@ALM`|-_XxCe_o@}KEM#PT5HWypx+RKHS3AfD<-Z`QR?D7%w8efb
zQz75DsptClY-7w1b^%|0u7_53h*}
zU`N#Kf+Hl>Hybu3o;VX~SK}KwpId|o0(^;bL0WjkQ2MGj)6dpFVSxNPsSUI18P7aL
z%Hu6RnL14om(Pp`fl<+ecm0I~g>nNKU_r=NQEdA~3ljr1ItBSMl{h8e1vlwt%sgE7
z%k1!c5`V>JGGN_OZ8aEW*eSGRp0anGd%;~EiBSd4g!V6Y@abV;*Q^d^0uX=y
zdmP?DS!b~NNb@^_{oA#uySLg$u!?CsP`RsVBiLOtq<7D~cU4))IGs9hSdZ`+r=OB`
zFznVqqZz?yTpgQ_YjN1rPZ&bB<&;C&s}ZVTqVE5{U00AfNfs|moi*fk}WB=Pz
z(4^-KE}G$FU(DItqF4X(Q;F2$lVaQ8mRHcTn^BYM*ymU0>Eg-R!^Nu)8W4j`-*3a2
z4clM~i4g_d=ADUSh7{NwZ_tuGPg;sO^@GCvlJ0M&>TtVnOPhD(tOVa5+F`9RcJ;0p
z5(RNkX*AS#ORV$912h^R$~#g^7h<$j5-KzzyvIs2C5F|CO$<~9xuLSSasW1c0uPc7
z1O{BEn@@GMQ@IZUt~70!4{&v;?T$cH<_U-TbCA#ak6=C$V-BrBpfo&AoD&vi>yY{8
z(_g#!XqbALlQ9O^0!Ex$eS%x;TfAjdEGCl5Gml7ra?bkUarTurb{U@vx}A_Wu6+ikLNlcC;^FvzQdLN1QkA7IW@B53_vU(N<&pxt_}pp
zhsM3v{E{uMGCL|U9_t|@Quy{q*Zj9ZVjTK64OPf>`j;}`JpK34;r#c4{HN~c(}$R6
za#r)RkLU8&`tWzmhTK&K)JY%Y<;;)$5di?-K|-cFu)-R_ZgUic`|E#d2${C-F;PnC
zKoS`osS>AxmSzE?1)zuKx;)nM&y=`$vZ9_wptB=8tCQsqVK5)e{E8*$ec
zNWSzl=x_(Bz5Qw$*Yu1=g%fe)O=D#j3+Tp!w$r@QE1_bsweAauE#QhUJ#d-dnaEQG35d*0#8Ss12Si7)2Dl7&fx*MlmO?gp
zPOw}7AA_`;cb*AI#u6%l1^&Y$#2`<=uQxVqP~|}B6@y3-<@uRAYMj@>%5ko_ZfFEQ);Xeg<$7nbX)UxUhBVVCuk&b!v{&}{g-g%rx;{-wOSi#vSUD?
za)Vq1asUDl21q^2>#`Z~m^0xEUkB@3kFhNoZF{%W@%-M)_(|RYME7M^01Lo*um+fy}uH{L@
z)T9t_M$RLr&BP|x)x@%LFI?)(OTUSK7wmSbvs;}w07^}jfphek=8&_RZxh~I4$l-J
zsxC3|l_v&t0n2DDepi#CImxf?kT-E8|LAMCX9G5CsR<^%39GM>rljLBw}+)lfxOzA
zBj9zmjzT@m&XKTAGX9>0Rc^NN8B%`@F}iC%?IAgg0E#%mKn1s1zWPDje;I-(?7L08pMm)6
zFlyC2v)YjCwlg{T?NH$#|48=Y=efV@9WV+GkUM*NP>tjzaVm^6eXXx_>DPz6C-m(b
z-y#Ns<<<&#$TvVk_=Vo7lBXdZIM5e;DNT6k30!q__P$}F7Q6SrnAw}UxH_8|+x?fI
zOdU+DVVJp@+BT*@mhnNExmnuk|AF9H{Fm6${!3_WM@pc<4*!)LX^gL+=xu<0P)V5o
zO0NGI-I)YcrUPbb`)|x20p!15T^%?WIwd;`45OTxy@jhK3lSR!3)}yxSc#baPb6aJ
zk7&EL#y@
z6Moe^j)9`xXdV&nE86$!=kw{mm$!GY?__ZKp8e{!;}TsT&`<$}C))_xA|+NCjHadn
z&BVkers~wG#s(ahkj4azuLt$t%D_`nlY^lEu%zQenUR#p>
zfu;c%g;2;4&B7#7rSA;^IgmzBDhpJI7D_Nk;bH$dQ40k-{j}LCJgo
zK;W66=A}}oszO`G-5CuPSjige%Oe{ZVn2wO!PJv-pK$rmoFc@)cZb+QThnMQkC@;n
zR$utDKoV4$fHFjSN6HvocKSSy{Nt10fB#+)823-WpVOK-6zAfKNM<1!oMn?_G85UtGODFaf*Q&O=*!;OR)Ul>srNUb
zf)5|EO`i*d7N$iyHZouae0c@%r3fc
zF?!6n1_%{a!;#sdVBVZGkroSr3;@3-W+FvN;lL^U6~uj&WTeLg8Gr&@gGK&m>i@z?
z@E$-D{PKRq)ywW9*eH{tY(UGu4TY6i9_$ELGv`JQd?&u8eQT0)Ed&CApr|w|Nn>;W
zGbCurktdy($aJSa@L=G`G<;Abw9{dZ6^j=EHZ}FoCz4MS9A85C6UGC?y237;shj3Rh1&lf{UNd@8RQtAQr
zLBVcx8GUF3l5-e)g3SRp`T&rETLGCrAQp-j{zx>k1s7a{eKmP7St{{*Te0!6>!7Q4Csz%M%M{{${J8@=kkIc!9G*e3&a&=bBS$A0x%y4I6r
zD7b^t?zA=`XB>#x>?S-tWTiKKWf!iR^ZI_e9_kcS^R
z6h*rBZEEktEC=~Mgl7vly}fJfuFKcYq8@|u&v*X!9<2cFz->EkxmmZl4vHHMg&bV+
zcUPRbPx-jJtxlDJir6h~J;z;Z4y{KtpL|OmRbM~GIs>-qSB0EkxklN7_ppoS!Qe%O
zr>Tigr0RoLT&(PF&-~-waU>GNyvuXRzz7CP`eF~wgWbQFx0dX_l(0&J=U`58so3om
za{a=-5FY?}72(|(zlI3lr<+y_?_@|gPr>zGCqZStr}i$|?;Klb!HKNLB`bN*vxgaW
zNso1Wrx^lvcX~U|^hZ@~7HN}1(Pw;m%mrkXK~-X2U0lc*2Tx=73BFx+^%`b*GsxQk
zBM;Z5uoktQri$CJUA*Pet+&LY?9wIA=F1k(VSRx3jOHb%w?_eqQgEx`uPEujJcG)JcmA8FE<{Vb1QWobbwORp1h1WlG8Mgfr}KW#c}`
zgvMF<=28jol4Kp;*$errUL0HJYO$57s#+g;Q~6FoUEI(cQ+!|8M&3~vKl)WRHzg0#
zV?4ls`+b#&^ZxAG45P2$R`BS$`}UbIH^u}BFKg>~e$GFY(0iM!tvyHUPfT79I*h1L0j^wfEh8W5WrB3W#njKk;(Qa?e(j(eNBssemiZ#7~7vAMJ@n_TT
znCV4{gko9PjpXNU^5l1LQkg|J7`q5oEES-NKhCeZ0SL{z^Xk@UhIQ+#z}>bvkKy7m
zF?K`#CQN80&8B0Ug&n^N+C7iqUg88ZrqQ_$7n&UmV;~5tkm5#{Vl3j#-7dPBh4?vC
zU2CRaI%CXQTtWhL^P5@zt?8(yCnV>dvhF40RYi;@2W~L)sXfeOtjq+liSW%D1QW2=
zJF&kxj#6JskLuy6hVi0W9(&ziv3dj2L$dOtZtpLLe$-})F+QZi
zE-CQ1qS}d_3Z-YHe=THt3X#lrbqTrT$pd06(%hBvt5TG!?)^^Y`oTrpm;70Pi%PMJ
zcN2eEaT)ZoPKq=5u)~S(s}<{!i3hOpPZZQXIi;HO+NoZWp!JAndUtNfkyOQzZa6uL
z=RIr*4Cmz*)*6Cn_HFS+>YH_5wv9M2iPoWQ&54T;HQkHZRDXzQM;hZTs~v{%$0}l)
zP-CXiy)t{o#*^yzCtV%c)~amZbSZdd+X;&=UPlEl!h+oaivc3T!*
zCE*4(^ag__Q>UGUm?TB!rccMn`b}URtkb0&_n^C3y~Q($F;eZcbr}4Qpb9WWn@Vcmro%X9k!B8^xN9a!T#OtP0kjsv$U;bgKM?L(-69=pEVU`%~8H)&AEAvkD}Yo5je
zN(h>Aa8@#i>w7(bD87qKuoF?Kavo?Knt36ywevlmOnAyWKQT?M8n;uq#_Ss1k($3p
z!S?$UPuioAJ!V>)&*D|Gu^{HuZ;S9r9k{A
z-UPSD(X6md_jc~_tzoU$Pz8=*`GUdgUoXB+s+K6}^yT>)((C6Nr8op~-U0LHn#&Gh
zUU!Xxj=M1vfL+eM&NX3a--nc|!nQ<5J>&1ph{H+sQvKt9K5&zD`6GlfoSucfjzoq*
z%SRaPIxAia;qqViGEFJuemeZMcWd}3Di&Cc!1{9|CFPEfyZSi2xA61wd<$V-_U&Vf
zb)`PRv=4pH=}uo-#&h}WT9`UMwA+ai@8G(duiz8Xpu(qk#6jtOPvA`2u9h@tnNK-L
z`@Bz^fi;GrM|c?&B_iTe`~(?V2ab!J2M2W*Bi+GS8xVl<%#%UI*W$}@%UajEAROLp
znLe+gfhR{XMj*uhBS4?M$uIbD!B~WuoaAd6JlamkZb;QZN_E@P!C7=^x_(1
zs&~pbGwAi!j|FC?#*vqQ7W(q|%BAT}I*M90PGh;Envg&#o0CySe?O^J7`C35CwEAS
zMM8TyBp)z?`rwyK=9=k_
z!2pF4iZ(`lq-Ni@oIi@d1l)_%Ao-}&r2gcc^kIXh&;a{fS^<9~madK~T`)6wC
zO!b9hFQA_3YdaIwVsfpq^#h#an<4pG;@4#36m$1}x1Hd#U|n^kivE@jOPM(Z5XkS=
zc}+>-oY~yWB4fPw1}
zV*teOJ7}xK-*MCos9)ZELk8D~Sw*^@R!Lnn4AcPMYvsT)m85$H9-K`Qq8$z5+A|nx
z-QD|Z6*7lOW1a=AcrCwfJoa8R>Dyz7X5}3sYWWuO6Rk}UOkA_q8gn;oq@~6u0J~*)
zL*!H)+lgD>K6ev#EwhDeo9>n-)mqyfu>pdd3FnQ)n)sRm>tajLqu_Qxw=5dKxH$m2rK#Iu)IFr#2zMB6hOIb8x&c_*xGy!rVzW)`EL*(ExPE
z&HaJgj8q3yhc^*A4qxv+rv&}BYVj}zhA!O0)If};fS55qZu-)8k>0sfWGi*zev5wX
z>%PjXYD+ve!P#CK-;JO;zv2eJ_29FIsk#UUwQ32eSYaZUK0!*hLwzVd!D@S7K1n6H
z#2=r#s&pBQzf92WDM_cGJ`rb4gn-XuSZ5ShRw_nn-EHCruZ4l9F%99
zGjiL9%r-94!b2P8YuT07#+8$nEe#OXJ@a7h5#PnUG{UPJL(?sVkVW)D6aZ>pq^4oM
zhiccspJ<`b#MA{$_F3_|IGcPk4dpHNFa~Bs1G)x-8hW?GgGB6Vd|ycb4S|MLd=hlo
zUtPXk9PfU%O|ZY}3Ezy>elkgn*+o{$Zx|+TV$>r*HulCg)0sha^lFhL3~-{;Z3JhQr4c
zb#kwMDqQV!e;hrlP`EWnP^~;7;FfDiV4n!Ti2fEs638mc9c6QZW+c3ew-$22`|7RJG%Er2v*BEellarP!N
zen%^(jhWVV@{+gI-DQL=Zkm+9OOsWxpXiujbMj+*=9cpeB474jsFu_f!TE>G1KC>u
zlS588>uzVgSlVrxzJAUk69qEPa;Xb6iMn9*c
zOP+gV;+d;(L#LD9KR`0FZU9e{H$$natVJI=XSqK5I)BzbjL=3GrF-MIe>3f6{M;em
zsU7j%JI^#8hU)EAX%-=tXHm|p7Mii+#Z2&wapQ0Al}fd*noa9s7L#Kka3=YP9UPc9
z1l!LezqR3p$Q9(LV2
zn+w;BvKj&+nws7Wmk@>_GZUA$?o1(>t5oT=i5RH$!9K-c0JWjgP&IywA@TTW3r;O`
zLMh8m74x70S0iQ3?J?)i(;Pdf`)f|?0|6a0_4OZ7Nx&+vmfCRD>n!9Ys9(OxQVa3>
z8fjf?oQh1$%7()mmS3F&_1?_{Z11!6LFwWy!R}15s^o658`IT0wXHsi72m<+
z8>@uKcjG{5%T7VYGOQ{6#p#l^O|w)5-K%)qpGpuP9U98)kC}f_Q*jWXgRh*y3}W~+
zY$nynS%6b%9cYUqTLe$A3R1^hb1(2{B{Efzl&3{z$Wak3JB*EKER)TflN7ww&eO_Z
z4on7hV?Ei(|rEqYtJ10Og&Y%u2ZJ^B0fY+#?I_oDN{2b!4*}iPB$$8s9ezF@xC0lya`3wWyE5cP+{s`CE?r&EXXERxR?YUwF
zJYVyd(ek!I6_^f=ef-uOC`tC0M%M>i>{;yu_Su<2nXD+`Fq#R{mhi%feumYNU8|Y;
z%&-!5r{o1q!{$AvPqI`1BHiZajp}1WwBI2QNTo7N)d(K)nU8k*J`4r$(!}45azH`q
zANw)pA%^?#yd%ehg5E!{X;JI0u4(Uj>rsn(W~+?DmLv0K_RlLZNL>IaKJ4BSBp2qA
z7}y7@lk|rai`fiF>{aa+Qz
zm*g$Po=ec#&r7Mf#X59KcZdA-2N>R6_XjLj#(i=V7T6{>Ny72M444oY23kI2-^1bw
z;WMMZXu2643EG1wxo^Y8W6PsmDH*7<0?BOtulgfQTfQ9+osu2=-(?K8et-ac4Js23
z)nRum<|QRh+d=ZCSC}ru!N3+a;ge`lA_hw%>*iv#qGfNmtw=o`DT_!CuFs9u=*O{7
z!X`|aC-kn_K77xrEQ*9T;}hegUcK?6e3g&=TR&aYQS1_gwAC~kFbuum
z?BANoN=%BwEu+@zbi`2umnK=J6$`ZW>!WgGgW|gPm=hl#JobLQf>(u_u#NYL!`HhPIj&?5lm1=B<5qrColYthp
z>ifP5cpJ(V6Ic<-AAu!8_JmpNM)(RZ6v6s4{9mP}`s1P&jW^*D#h5Zsz2#p>pI%*&
z|A*86m!`8Y{ckzH#|M)M17YH5<1Yb&k^o`iO8YO6v{_w(iSztN=5^%Y_(~wmEdK-P
zhi>4E79h<3dn)A)xR^NzGspkp^tK>T2y+e)X0EmiF9<(mV5a|(c)uu!BUlg?mNt_#
z2p?z=HqJDnRuD==ZjS#;>}*^d9RGit1vNlV*9Dhz(QmCTee8UT1@cu^-MWfys*NfV
z|Gk&tzo%5|5K~4?iYDQ8%^~lGIO0G)xu#~b5^E3$KL6A8D4&_E$gWncB&RaFOt5p(
zlO?6qvpGe<(>;aR6Couc^0T^BAdS{RrKG1esKIA5izZlVt~E0A^sY9NV6ImNW)4sw
zgD}@1gFRO0YhILUx4013NKfRkvSe;
z`j$is%Py7r9D$Wii(j<6R8K^KBVRK*4~4%W&>SBqAOWR80Td*~ip|KSm1PxEOj`yg
zFNMQvwqQbgX0pCCp>9Ck{5HueA`2)JLQhXIIYFh6t9Z~(VxZO>(9X=DfXB8r@5h+P
zysA3SXVgYda`NG^5MkP!oWGhdU)0>70hj-v0pSF96&Fd`XOG-}V%xTKwjc|`;gY3x
zPFW}pYQ(T>E`*`E#kMR;EaNBw_5fk%-x}ypl7d`EvZ^+h6o;pb=Vkv_EDcDX1mffh
zj{P%d>K$5FNNh@!LPA$~uAGXIk4@J<|CHB)W(73HgA)SP5Xdyf$B__&P{i_uf>gwz
zo?66K#?Q6K4MXHP44yetM4Px8&XWnHGjL2}~
zH2ql)W%_sbSotr8j4%VS5j{rhLAFF=wJVp#f|TIBS>o9nCjbq-*U*;-)%eXtYkc4f9`M`@h)C+=!n
zdHvFWV%Qy}mEj3bd{AK!&A+D>Mg7~74KKz}|5*`t^Oo<>48UVIng$pK`FeXZ|9Dzy
z7PJe;pRd;x#MW1+$^+D1P(nq~nJ9AMS-5F2R}D4Unyt)-w_
zAkDmn!?>fhHdqwy%BO2Jo;B?$ubW1;;;-E4yoO`Bqq#i^1eLA-LPI?V&
zmu_C@+uS_*(uM;`Ez9_OsVXabc|N__1wQrsvT?NuIRNkqSxIG(c}y!~SC*ja#GMA^
zCTwav_SCZQLHmQ^vM7P>?2D``qQ_ggRp-^W${2?Wm!cO|$5dp0d*oPY)`m7<-}A@0
z(R|_yYzOA|r2rbJL;@AW&ACa*_JY#Be{+pjz
zi9se*|0YVs9;KwUQL!){kncwNAOf8t7C;&?WpM!0wN=0r)hBP;W3oL
zLwALOYC^K%aaygV#?OT^gjWl$9kx7edAs1T4)}FU{Q=`%6Al(rK920A9IWK9NwTEt
ze7VyJW4>kf=ohdN7(!O;QQRn^G%T=I3HjX^KeK#!T*c?;pzR9wkEust#YSi(!wCo3
z;DuQa-kvMw)`>hN@-~-U$(hbQ@AX{>yH*!3bEmn^*Y=7QmTfgLbPR29jjp~O@9FW_
zMm1QU(}3NSc=SCx8g@vKm$Vajf_1m#191do6E>JoJ#L;5E0Ia+3OKEGZ?Wb$;2ZgL
z8Y&9RPQ!3E^1k$lpNrh)gyO$l|H4!KfF$nD;n674xg8mG5{(lJ?m3Jk!Ec4sWGYTf
z$k+Q^=EmOPqB7g|X*V1r*rIVhd6cgQ``S4NZ$*2IqYa92tPc)4%Q%eg=-H)xY)-mhK
z{P2p8d6o;rf6)rgLemEQLXzJ~3usJDZ`_>%TiU2FQ|GE;+t154rcK0FD${3trl?|E
z${XKT&yV=Nri(Fu+EbND>h2o9QfbiBcEHI+og6<_d!HQqYJlfOa=1(CfsH7SY&%gW
z4Tb{{qerU*tR2a5gsZwpI5S$L%6svDj6wg*W^-Qb2x}F!$cuus#Sl#sOIALpcyRaD
zOL@#Xr-rBkiDbpxY*#pfz)&NAf0gpm<@~*xYKO4(XGWA-VkUjEZ5L&Tt0g#J1N@sK
zr;h_6_R+&!*E6s|5cf^SmD#0S_1%N+FO!r2hRb5dId8{Etv7l3;oejKZ;;37PgEYL
zu#XP(iKB7Zq7+_D@EkZwqqyRi1Tl9Xby&5~=ATr1`cm*}Hve{fgK-bqOvCH>%W77x
zZ|q_@*e(dWcL+gBExPiRym@|CFh*c-JaAGFF_v!~2jh8jDa=z@*mo$G+9yP5P38&U
zn&SJaR|UMp$m_pD+Zx75vLEpwSoq8buD%h?t7!8gZw4aqm%19_m)ra0LqHuL$TV*J@&0}AI
zVlg0_K)3qqrdT-#1g9|~^yc&LNQd!Pe?_fiTQEZJ>3S2A4lH*Ey}qlS+x$cIKnnt$
z$G-E8n|usSl{2ex@eg#UkZJy2u`|z-teMG3-wv@(E`ogF5Mh&3p06yI5vKtb_Essv
z_@!rBi1Ehqt&w^7KWm*o5t9s!8@k7uqek7b&QzZ^+7U-Ie3)aKu9}cN{^D{CL`c-0
z#E^2&j)JqEr}_s06F1K0YU>zS&y5Lc`NpGYVroR%Dv|uw#t)pSRH8;gsesG^&kWt4
zmut%iv{ua)9k9!IVU|1%!UM+5d;9mO^JwBNv0pneaNwtPdxxIHpRFWfG`36vVl=%BXb$Xx+
zv3==*RfByW6spN3aIWh*r0ay=pC@2)V#_{t$SvoHTG=nG<&`ur6QxtVpcDSwhI+-#
zK3{S}kh}G(yFgIj>5G^7>Wq5Hw!8G!s?r$jhO6ca2|k{VXL)IE5tYebo$n?}Qj8=D
zcFug~mv#{1zZ~e5ZNCZBL@voHsBYQkkL50zQ)5qL?6`B6TgH(R@h8nqIN`kY{_apy
zs)ftt;e&%CqT)QODY>MV+DcUT3-B*0p=;+vokKP7j@(
zQl0~5YL
z<-s~{oei#1jk1WNSwV$^@C*~#+%>9b;tPE|JOUMuB
zArHJpLFd|G=V$$)3y)Io2aV{*qy*}m8d(DPbp0f%O}D`u^p9F}AO+aAgAIT*!*v-3Fd4>rQsMXI@O|JOUbk
z1JAYPU9aPhcaSA8!~qEEEn@%uPW?CP`EXg%L<@(>I;n%RpoRHHuWMgRIHnysz<97*
zy$JEUykmsgnG36@S!ZeV@H~9$Fe~_JB|6ag0E>=O+cyugakdur9klb
z-Qj!d+;;YL?XX8MTX-XnK}T625X#~lEf93rqf0rn5!vw=5t~q9xBm+LOKI*=sBt%a)Pm9m$BDGp
zK_vs_Wr<=5dm1LQYfc-c@NktUPifQ6&1DOiWupdyHE_g90<3bpiNC4}Hlsy9$bMCJ
zgf6G>g_CLqjrqSWw19f-pFi##_>pw@83GIrdB9s4*K$+tY8H4gmg>Q3OZLDq6E0cB
z-iZo)`-g(@x+zJf$|T!`)~hC(!=y*k$hx}|r5PPUpO<;^@*uUAJ2goP+V(e*
zy_2Y(9PQVwx*8dY?>$z(tLVND5Kf-3!kZPdEZG@0e|3x+dy=Nl80>b9Tj!I@jHFSA<|femMqzUS$;f&OEMeT1j%&ot^{M4SD#qo;X&
z(Py}RtKCbEs)e((+CU)F)v6ubzysIsTgd#zcfa=}V=8`<96q<-n~RuRdN#M?vr&EO
zR-T<`a=_(qP&f!00jq+O>B{=$%M{!I70_53C>-l`xsT%o8%VgRCSo+k0(Zy
zq}Sddar6+bFIw!Gz||6^KR#B$+_`jU@DX&3p`|aRs)fcAZ8|+u}_ec9aa~vI{
z$1A%(+M07ad9Cl*-hyv)``fh~+36I(lbjFG6L`j=1LsllN?`fT$QUAwV)+2fWv?#neS
zWy?yg=5KlXGn4nU*$IC)xongz8GXR!h`(z&ZE<$%F#e_XMt(+?;#8;md2jp|I(RiF
z8mZY`_DYZ+Z0aghWg;jxI`bBqB?7MnvWKntP!gnqDZ0c|SEYBALw~Ga%e6tP`mUZ@
z2k85zU6)d~d*A3pQHkE@45}bm#J{e-@xU}wI|No^!rvr^GAxZF0XHHU88_97tSJm^
zwAAbM6!j!y2eX~N*jBZ`#1n8ck?%69c0!IR-x_ZpGR|y$CHATqCDfVGzU8F{l#U1M
zC*LaGXnQD;65MVlk?^8j{3qy$L5{|xg>LYZ0rHQG(|_t!J)&vR@R-p4+aP
zYY08}HCbO6%zksc06lb%7q^)1SQYHWD*h3Ub=8(&>9G^c14>w
z5Ib1bMj_StnJcnXR6d0Ed_qXV^odJ!dhLl3A-&=0Gqh?DlNCYdd5>!k?R8mw_Y$sDABfYkot3Bjymv(loe1n;
zQ4>FwzZttW5fST~tWL$v=+8C{w>cRCOy9}eGLfwOEipd{+Z?v-es(xSs7F>}CFzxZ
zcO`7GmTZL%$}ag(i3iGo0jsmpKU9UK2MEk&s;;#+Vt0OQ-glg
zPcE(f?fzI6V9mM7%PY`Wk&^J>wti-5&U35rbQ{NU?Q!dIlb4e)P{G#AhSEfx+hP1B`h0z5`SpU_?(5aVcWTxr*<~6f60)~~U;apa)
zK=ELif^rM5`*x3$iUqc(#6X3m77cH4adyH*jTO6rz_>wmL)lqRHNAesw~$$NS2`a8
z)Gp6rM?c(4hGW}Gful49E-=)+^35$i)-M=3J3o+kDUcyVvgi8^iv`HwK~JU`W{kkSLr&_{BR<1gjWG
zltS2m)ku_8ob~TRJx50*tC$3YRfJWKLRihg((#Xy{Et#%HIfEN{@snw6hxm&DnC_J
z5Qm71izp~6fgs{fC7+5aL1bC~{}QS{7yXmnW(A8$vi`HC_b^sU=cVrbXM?ih>wY1A
zj>PZx1#f=eXCh@d2>kx#rtEWXLVa+@_l}vVp`T!;D%NrXY8vy*awUK;I8Yy%TQRO=
zwQIRr@46>j>kjZQC%ax_UF&Eh{Qsixr@7v&;t0mf48|A8-(BC@83~}Iv1+-Aq2O+&
z6(JosFpig?0RC#0Xd3^>Te6j#VsLxnBTm{S3QaKP7oHx^j5weP1-D2Ac^-`NyVa80
zks3AYmxTvAkGPF^h>#2UY0OR$Fg3Fb@%lRS44B{ZOhDh-Aytvrz6ij|j8Go?`GZ!p67IQ&BT4VK3tyBI
zkv6k35OO&^pbVf{s5ij`ueNX1D~W^pMKneK&@a+9=IN&w^1s-qCun`k@@{1RQlZvPt;~sN
zPbAtg)K#8tSPESV9SLF&|1V34)i=C#|1j)!u+X_9^?0NRWV998hcC;~?)E-?!zY~J
zsH_2xK5ekbH>!xP9OIY~VM8jcaV?5JU>+uxSE`T6T1`?`8j93uzUY_H7>phHHj?yth2Fowx~(&g{Cs0#Fc+aOsJVF5wPTZcC3RgwZakp|
wSc3Iwin(E0AnFC-SK6n~-L45Hs^f6~oye%6dQw_SaR>wiq2S_D)_F$pKXqY)uK)l5

diff --git a/doc/src/DecisionTrees/DecisionTrees.do.txt b/doc/src/DecisionTrees/DecisionTrees.do.txt
index dc02103a0..3ce2cee55 100644
--- a/doc/src/DecisionTrees/DecisionTrees.do.txt
+++ b/doc/src/DecisionTrees/DecisionTrees.do.txt
@@ -1148,7 +1148,7 @@ However, by aggregating many decision trees, using methods like bagging, random
 
 
 !split
-===== From a Single Tree to Many Trees, that is meet the Jungle of Methods =====
+===== From a Single Tree to Many Trees, 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
@@ -1157,7 +1157,7 @@ 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 Voting classifiers
 o Bagging and Pasting
 o Random forests
 o Boosting methods
@@ -1425,24 +1425,91 @@ this setting.
 ===== Random Forest Algorithm =====
 The algorithm described here can be applied to both classification and regression problems.
 
+We will grow of forest of say $M$ trees.
+o For $m=1:M$ we
+  * Draw a bootstrap sample of from the training data organized in our $\bm{X}$ matrix.
+  * We grow then a random forest tree $T_m$ based on the bootstrapped data by repeating the steps outlined till we reach the maximum node size is reached
+    o we select $m \le p$ varibales at random from the $p$ predictors/features
+    o pick the best split point among the $m$ features using either the CART algorithm or the ID3 for classification and create a new node
+    o split the node into daughter nodes
+o Output then the ensemble of trees $\{T_m\}_1^{M}$ and make predictions for either a regression type of problem or a classification type of problem. 
+
+
 !split
-===== A simple scikit-learn example =====
+===== Random Forests Compared with other Methods on the Cancer Data =====
 !bc pycod 
+import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
+
 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")
+#Instantiate the model with 500 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
 #Cross validation
-accuracy = cross_validate(Random_Forest_model,X,Y,cv=10)['test_score']
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()
+
 !ec
 
 
 !split
-===== Then random forests =====
+===== Compare  Bagging on Trees with Random Forests  =====
 !bc pycod
 bag_clf = BaggingClassifier(
     DecisionTreeClassifier(splitter="random", max_leaf_nodes=16, random_state=42),
@@ -1578,7 +1645,6 @@ plt.show()
 
 
 
-
 !split
 ===== Gradient Boosting, Examples =====
 !bc pycod
diff --git a/doc/src/DecisionTrees/rfcancer.py b/doc/src/DecisionTrees/rfcancer.py
new file mode 100644
index 000000000..3c82c5bd8
--- /dev/null
+++ b/doc/src/DecisionTrees/rfcancer.py
@@ -0,0 +1,66 @@
+import matplotlib.pyplot as plt
+import numpy as np
+from sklearn.model_selection import  train_test_split 
+from sklearn.datasets import load_breast_cancer
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.tree import DecisionTreeClassifier
+
+# Load the data
+cancer = load_breast_cancer()
+
+X_train, X_test, y_train, y_test = train_test_split(cancer.data,cancer.target,random_state=0)
+print(X_train.shape)
+print(X_test.shape)
+# Logistic Regression
+logreg = LogisticRegression(solver='lbfgs')
+logreg.fit(X_train, y_train)
+print("Test set accuracy with Logistic Regression: {:.2f}".format(logreg.score(X_test,y_test)))
+# Support vector machine
+svm = SVC(gamma='auto', C=100)
+svm.fit(X_train, y_train)
+print("Test set accuracy with SVM: {:.2f}".format(svm.score(X_test,y_test)))
+# Decision Trees
+deep_tree_clf = DecisionTreeClassifier(max_depth=None)
+deep_tree_clf.fit(X_train, y_train)
+print("Test set accuracy with Decision Trees: {:.2f}".format(deep_tree_clf.score(X_test,y_test)))
+#now scale the data
+from sklearn.preprocessing import StandardScaler
+scaler = StandardScaler()
+scaler.fit(X_train)
+X_train_scaled = scaler.transform(X_train)
+X_test_scaled = scaler.transform(X_test)
+# Logistic Regression
+logreg.fit(X_train_scaled, y_train)
+print("Test set accuracy Logistic Regression with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Support Vector Machine
+svm.fit(X_train_scaled, y_train)
+print("Test set accuracy SVM with scaled data: {:.2f}".format(logreg.score(X_test_scaled,y_test)))
+# Decision Trees
+deep_tree_clf.fit(X_train_scaled, y_train)
+print("Test set accuracy with Decision Trees and scaled data: {:.2f}".format(deep_tree_clf.score(X_test_scaled,y_test)))
+
+
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.preprocessing import LabelEncoder
+from sklearn.model_selection import cross_validate
+# Data set not specificied
+#Instantiate the model with 100 trees and entropy as splitting criteria
+Random_Forest_model = RandomForestClassifier(n_estimators=500,criterion="entropy")
+Random_Forest_model.fit(X_train_scaled, y_train)
+#Cross validation
+accuracy = cross_validate(Random_Forest_model,X_test_scaled,y_test,cv=10)['test_score']
+print(accuracy)
+print("Test set accuracy with Random Forests and scaled data: {:.2f}".format(Random_Forest_model.score(X_test_scaled,y_test)))
+
+
+import scikitplot as skplt
+y_pred = Random_Forest_model.predict(X_test_scaled)
+skplt.metrics.plot_confusion_matrix(y_test, y_pred, normalize=True)
+#
+plt.show()
+y_probas = Random_Forest_model.predict_proba(X_test_scaled)
+skplt.metrics.plot_roc(y_test, y_probas)
+plt.show()
+skplt.metrics.plot_cumulative_gain(y_test, y_probas)
+plt.show()