From aecd18f3f75a8324c92934fcf0920a859c57179e Mon Sep 17 00:00:00 2001 From: mhjensen Date: Thu, 24 Oct 2019 13:40:20 +0200 Subject: [PATCH] small change --- doc/pub/DimRed/html/._DimRed-bs000.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs001.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs002.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs003.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs004.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs005.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs006.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs007.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs008.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs009.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs010.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs011.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs012.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs013.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs014.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs015.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs016.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs017.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs018.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs019.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs020.html | 36 ++++---- doc/pub/DimRed/html/._DimRed-bs021.html | 35 ++++---- doc/pub/DimRed/html/._DimRed-bs022.html | 88 +++++++++++++------ doc/pub/DimRed/html/._DimRed-bs023.html | 78 ++++++++-------- doc/pub/DimRed/html/._DimRed-bs024.html | 70 +++++++-------- doc/pub/DimRed/html/._DimRed-bs025.html | 65 +++++--------- doc/pub/DimRed/html/._DimRed-bs026.html | 51 +++++------ doc/pub/DimRed/html/._DimRed-bs027.html | 59 ++++++++----- doc/pub/DimRed/html/._DimRed-bs028.html | 65 +++++--------- doc/pub/DimRed/html/._DimRed-bs029.html | 54 ++++++------ doc/pub/DimRed/html/DimRed-bs.html | 36 ++++---- doc/pub/DimRed/html/DimRed-reveal.html | 62 +++++++------ doc/pub/DimRed/html/DimRed-solarized.html | 78 ++++++++-------- doc/pub/DimRed/html/DimRed.html | 78 ++++++++-------- doc/pub/DimRed/ipynb/DimRed.ipynb | 75 ++++++++-------- doc/pub/DimRed/ipynb/ipynb-DimRed-src.tar.gz | Bin 191 -> 191 bytes doc/pub/DimRed/pdf/DimRed-minted.pdf | Bin 253145 -> 253120 bytes doc/src/DimRed/DimRed.do.txt | 39 ++++---- doc/src/DimRed/PCAexample.py | 54 +++++------- 39 files changed, 855 insertions(+), 888 deletions(-) diff --git a/doc/pub/DimRed/html/._DimRed-bs000.html b/doc/pub/DimRed/html/._DimRed-bs000.html index c3c8a51c9..396bd951a 100644 --- a/doc/pub/DimRed/html/._DimRed-bs000.html +++ b/doc/pub/DimRed/html/._DimRed-bs000.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -227,7 +225,7 @@ MathJax.Hub.Config({
  • 9
  • 10
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs001.html b/doc/pub/DimRed/html/._DimRed-bs001.html index ce342b3b1..d9043ef27 100644 --- a/doc/pub/DimRed/html/._DimRed-bs001.html +++ b/doc/pub/DimRed/html/._DimRed-bs001.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -227,7 +225,7 @@ data.
  • 10
  • 11
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs002.html b/doc/pub/DimRed/html/._DimRed-bs002.html index c537b36a4..cd7af24e6 100644 --- a/doc/pub/DimRed/html/._DimRed-bs002.html +++ b/doc/pub/DimRed/html/._DimRed-bs002.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -226,7 +224,7 @@ ensures that all features are exactly between \( 0 \) and \( 1 \). The
  • 11
  • 12
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs003.html b/doc/pub/DimRed/html/._DimRed-bs003.html index 11eadab08..8da86d618 100644 --- a/doc/pub/DimRed/html/._DimRed-bs003.html +++ b/doc/pub/DimRed/html/._DimRed-bs003.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -229,7 +227,7 @@ techniques.
  • 12
  • 13
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs004.html b/doc/pub/DimRed/html/._DimRed-bs004.html index 46e7ba32e..e8c8765cf 100644 --- a/doc/pub/DimRed/html/._DimRed-bs004.html +++ b/doc/pub/DimRed/html/._DimRed-bs004.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -304,7 +302,7 @@ svm.fit(X_train_scaled, y_train)
  • 13
  • 14
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs005.html b/doc/pub/DimRed/html/._DimRed-bs005.html index 6359652d4..4a94d8952 100644 --- a/doc/pub/DimRed/html/._DimRed-bs005.html +++ b/doc/pub/DimRed/html/._DimRed-bs005.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -254,7 +252,7 @@ svm.fit(X_train_scaled, y_train)
  • 14
  • 15
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs006.html b/doc/pub/DimRed/html/._DimRed-bs006.html index df43e01cf..6a971e73c 100644 --- a/doc/pub/DimRed/html/._DimRed-bs006.html +++ b/doc/pub/DimRed/html/._DimRed-bs006.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -233,7 +231,7 @@ logreg.fit(X_train_scaled, y_train)
  • 15
  • 16
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs007.html b/doc/pub/DimRed/html/._DimRed-bs007.html index e84fa6fa9..42caa2901 100644 --- a/doc/pub/DimRed/html/._DimRed-bs007.html +++ b/doc/pub/DimRed/html/._DimRed-bs007.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -288,7 +286,7 @@ applications.
  • 16
  • 17
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs008.html b/doc/pub/DimRed/html/._DimRed-bs008.html index d78dddb0d..e21955283 100644 --- a/doc/pub/DimRed/html/._DimRed-bs008.html +++ b/doc/pub/DimRed/html/._DimRed-bs008.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -218,7 +216,7 @@ We have a data set defined by a design/feature matrix \( \boldsymbol{X} \) (see
  • 17
  • 18
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs009.html b/doc/pub/DimRed/html/._DimRed-bs009.html index d7c080305..d59b62820 100644 --- a/doc/pub/DimRed/html/._DimRed-bs009.html +++ b/doc/pub/DimRed/html/._DimRed-bs009.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -267,7 +265,7 @@ In the above example this is the function we constructed using pandas.
  • 18
  • 19
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs010.html b/doc/pub/DimRed/html/._DimRed-bs010.html index 130708b5f..0499b414b 100644 --- a/doc/pub/DimRed/html/._DimRed-bs010.html +++ b/doc/pub/DimRed/html/._DimRed-bs010.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -262,7 +260,7 @@ $$
  • 19
  • 20
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs011.html b/doc/pub/DimRed/html/._DimRed-bs011.html index 371c75618..997227329 100644 --- a/doc/pub/DimRed/html/._DimRed-bs011.html +++ b/doc/pub/DimRed/html/._DimRed-bs011.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -250,7 +248,7 @@ C = np.c
  • 20
  • 21
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs012.html b/doc/pub/DimRed/html/._DimRed-bs012.html index 57633c696..2c5be19cc 100644 --- a/doc/pub/DimRed/html/._DimRed-bs012.html +++ b/doc/pub/DimRed/html/._DimRed-bs012.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -252,7 +250,7 @@ The above procedure with numpy can be made more compact if we use pand
  • 21
  • 22
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs013.html b/doc/pub/DimRed/html/._DimRed-bs013.html index 8f6909aab..14ec6ffb6 100644 --- a/doc/pub/DimRed/html/._DimRed-bs013.html +++ b/doc/pub/DimRed/html/._DimRed-bs013.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -234,7 +232,7 @@ We expand this model to the Franke function discussed above.
  • 22
  • 23
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs014.html b/doc/pub/DimRed/html/._DimRed-bs014.html index 0b53a02a9..a95f8dbbd 100644 --- a/doc/pub/DimRed/html/._DimRed-bs014.html +++ b/doc/pub/DimRed/html/._DimRed-bs014.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -270,7 +268,7 @@ matrix.
  • 23
  • 24
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs015.html b/doc/pub/DimRed/html/._DimRed-bs015.html index 990a16b7c..e66cf4e66 100644 --- a/doc/pub/DimRed/html/._DimRed-bs015.html +++ b/doc/pub/DimRed/html/._DimRed-bs015.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -250,7 +248,7 @@ It is easy to generalize this to a matrix \( \boldsymbol{X}\in {\mathbb{R}}^{n\t
  • 24
  • 25
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs016.html b/doc/pub/DimRed/html/._DimRed-bs016.html index cbb6d1706..3ed9f1660 100644 --- a/doc/pub/DimRed/html/._DimRed-bs016.html +++ b/doc/pub/DimRed/html/._DimRed-bs016.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -258,7 +256,7 @@ features/predictors.
  • 25
  • 26
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs017.html b/doc/pub/DimRed/html/._DimRed-bs017.html index 5233c395e..4ab4a0339 100644 --- a/doc/pub/DimRed/html/._DimRed-bs017.html +++ b/doc/pub/DimRed/html/._DimRed-bs017.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -241,7 +239,7 @@ After this we ask ourselves how do we prove the link between the maximum varianc
  • 26
  • 27
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs018.html b/doc/pub/DimRed/html/._DimRed-bs018.html index dcd6f46d7..c29620553 100644 --- a/doc/pub/DimRed/html/._DimRed-bs018.html +++ b/doc/pub/DimRed/html/._DimRed-bs018.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -228,7 +226,7 @@ The PCA theorem states that minimizing the above reconstruction error correspond
  • 27
  • 28
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs019.html b/doc/pub/DimRed/html/._DimRed-bs019.html index 363c14d79..63ee497dc 100644 --- a/doc/pub/DimRed/html/._DimRed-bs019.html +++ b/doc/pub/DimRed/html/._DimRed-bs019.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -230,7 +228,7 @@ where the vectors on the rhs are known.
  • 28
  • 29
  • ...
    • -
  • 31
    • +
  • 30
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs020.html b/doc/pub/DimRed/html/._DimRed-bs020.html index 1b841aa38..b41d42150 100644 --- a/doc/pub/DimRed/html/._DimRed-bs020.html +++ b/doc/pub/DimRed/html/._DimRed-bs020.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -244,8 +242,6 @@ We are almost there, we have obtained a relation between minimizing the reconstr
  • 28
  • 29
  • 30
    • -
  • ...
    • -
  • 31
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs021.html b/doc/pub/DimRed/html/._DimRed-bs021.html index 9fae4423f..5a04c8700 100644 --- a/doc/pub/DimRed/html/._DimRed-bs021.html +++ b/doc/pub/DimRed/html/._DimRed-bs021.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -256,7 +254,6 @@ chapter 12.4 and discussion therein.
  • 28
  • 29
  • 30
    • -
  • 31
  • »
    • diff --git a/doc/pub/DimRed/html/._DimRed-bs022.html b/doc/pub/DimRed/html/._DimRed-bs022.html index b106221a3..d613cfc50 100644 --- a/doc/pub/DimRed/html/._DimRed-bs022.html +++ b/doc/pub/DimRed/html/._DimRed-bs022.html @@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source ('Proof of the PCA Theorem', 2, None, '___sec18'), ('PCA Proof continued', 2, None, '___sec19'), ('The final step', 2, None, '___sec20'), - ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'), - ('Principal Component Analysis', 2, None, '___sec22'), - ('PCA and scikit-learn', 2, None, '___sec23'), - ('More on the PCA', 2, None, '___sec24'), - ('Incremental PCA', 2, None, '___sec25'), - ('Randomized PCA', 2, None, '___sec26'), - ('Kernel PCA', 2, None, '___sec27'), - ('LLE', 2, None, '___sec28'), - ('Other techniques', 2, None, '___sec29')]} + ('Principal Component Analysis', 2, None, '___sec21'), + ('PCA and scikit-learn', 2, None, '___sec22'), + ('More on the PCA', 2, None, '___sec23'), + ('Incremental PCA', 2, None, '___sec24'), + ('Randomized PCA', 2, None, '___sec25'), + ('Kernel PCA', 2, None, '___sec26'), + ('LLE', 2, None, '___sec27'), + ('Other techniques', 2, None, '___sec28')]} end of tocinfo --> @@ -160,15 +159,14 @@ MathJax.Hub.Config({
  • Proof of the PCA Theorem
  • PCA Proof continued
  • The final step
    • -
  • PCA and Scikit-Learn Functionality
    • -
  • Principal Component Analysis
    • -
  • PCA and scikit-learn
    • -
  • More on the PCA
    • -
  • Incremental PCA
    • -
  • Randomized PCA
    • -
  • Kernel PCA
    • -
  • LLE
    • -
  • Other techniques
    • +
  • Principal Component Analysis
    • +
  • PCA and scikit-learn
    • +
  • More on the PCA
    • +
  • Incremental PCA
    • +
  • Randomized PCA
    • +
  • Kernel PCA
    • +
  • LLE
    • +
  • Other techniques
    • @@ -184,17 +182,58 @@ MathJax.Hub.Config({ -

    PCA and Scikit-Learn Functionality

    +

    Principal Component Analysis

    +
    +
    +

    +Principal Component Analysis (PCA) is by far the most popular dimensionality reduction algorithm. +First it identifies the hyperplane that lies closest to the data, and then it projects the data onto it. +

    +The following Python code uses NumPy’s svd() function to obtain all the principal components of the +training set, then extracts the first two principal components. First we center the data using either pandas or our own code

    -

    # Now add PCA
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
+
    import numpy as np
+import pandas as pd
+from IPython.display import display
+np.random.seed(100)
+# setting up a 10 x 5 vanilla matrix 
+rows = 10
+cols = 5
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
+display(df)
 
-X_pca = pca.transform(X_train_scaled)
+# we center it ourselves
+X_centered = X - X.mean(axis=0)
+# Then check the difference between pandas and our own set up
+print(X_centered-df)
+#Now we do an SVD
+U, s, V = np.linalg.svd(X_centered)
+c1 = V.T[:, 0]
+c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
+
    
+
    
+PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
+the data for you. However, if you implement PCA yourself (as in the preceding example), or if you use other libraries, don’t
+forget to center the data first.
+
+
    
+Once you have identified all the principal components, you can reduce the dimensionality of the dataset
+down to \( d \) dimensions by projecting it onto the hyperplane defined by the first \( d \) principal components.
+Selecting this hyperplane ensures that the projection will preserve as much variance as possible. 
+
    
+
+
+
    W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
 
    
 
    
 
    
@@ -219,7 +258,6 @@ X_pca = pca.28
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs023.html b/doc/pub/DimRed/html/._DimRed-bs023.html
index cf2b2049f..51348ffde 100644
--- a/doc/pub/DimRed/html/._DimRed-bs023.html
+++ b/doc/pub/DimRed/html/._DimRed-bs023.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -182,41 +180,38 @@ MathJax.Hub.Config({
 
     
     
     
     
 
 
-
+
 
-
    Principal Component Analysis 
    
-
    
-
    
-
     
-Principal Component Analysis (PCA) is by far the most popular dimensionality reduction algorithm.
-First it identifies the hyperplane that lies closest to the data, and then it projects the data onto it.
+
    PCA and scikit-learn 
    
 
 
    
-The following Python code uses NumPy’s svd() function to obtain all the principal components of the
-training set, then extracts the first two principal components. First we center the data
+Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The
+following code applies PCA to reduce the dimensionality of the dataset down to two dimensions (note
+that it automatically takes care of centering the data):
 
    
 
 
-
    X_centered = X - X.mean(axis=0)
-U, s, V = np.linalg.svd(X_centered)
-c1 = V.T[:, 0]
-c2 = V.T[:, 1]
+
    #thereafter we do a PCA with Scikit-learn
+from sklearn.decomposition import PCA
+pca = PCA(n_components = 2)
+X2D = pca.fit_transform(X)
+print(X2D)
 
    
 
    
-PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
-the data for you. However, if you implement PCA yourself (as in the preceding example), or if you use other libraries, don’t
-forget to center the data first.
-
-
    
-Once you have identified all the principal components, you can reduce the dimensionality of the dataset
-down to \( d \) dimensions by projecting it onto the hyperplane defined by the first \( d \) principal components.
-Selecting this hyperplane ensures that the projection will preserve as much variance as possible. 
+After fitting the PCA transformer to the dataset, you can access the principal components using the
+components variable (note that it contains the PCs as horizontal vectors, so, for example, the first
+principal component is equal to 
 
    
 
 
-
    W2 = V.T[:, :2]
-X2D = X_centered.dot(W2)
+
    pca.components_.T[:, 0].
 
    
+
    
+Another very useful piece of information is the explained variance ratio of each principal component,
+available via the \( explained\_variance\_ratio \) variable. It indicates the proportion of the dataset’s
+variance that lies along the axis of each principal component. 
+More material to come here.
+
 
    
 
    
 
@@ -239,7 +234,6 @@ X2D = X_centered28
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs024.html b/doc/pub/DimRed/html/._DimRed-bs024.html
index 70af1ae21..c9ffe037b 100644
--- a/doc/pub/DimRed/html/._DimRed-bs024.html
+++ b/doc/pub/DimRed/html/._DimRed-bs024.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -182,36 +180,35 @@ MathJax.Hub.Config({
 
     
     
     
     
 
 
-
+
 
-
    PCA and scikit-learn 
    
+
    More on the PCA 
    
 
 
    
-Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The
-following code applies PCA to reduce the dimensionality of the dataset down to two dimensions (note
-that it automatically takes care of centering the data):
+Instead of arbitrarily choosing the number of dimensions to reduce down to, it is generally preferable to
+choose the number of dimensions that add up to a sufficiently large portion of the variance (e.g., 95%).
+Unless, of course, you are reducing dimensionality for data visualization — in that case you will
+generally want to reduce the dimensionality down to 2 or 3.
+The following code computes PCA without reducing dimensionality, then computes the minimum number
+of dimensions required to preserve 95% of the training set’s variance:
 
    
 
 
-
    from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-X2D = pca.fit_transform(X)
+
    pca = PCA()
+pca.fit(X)
+cumsum = np.cumsum(pca.explained_variance_ratio_)
+d = np.argmax(cumsum >= 0.95) + 1
 
    
 
    
-After fitting the PCA transformer to the dataset, you can access the principal components using the
-components variable (note that it contains the PCs as horizontal vectors, so, for example, the first
-principal component is equal to 
+You could then set \( n\_components=d \) and run PCA again. However, there is a much better option: instead
+of specifying the number of principal components you want to preserve, you can set \( n\_components \) to be
+a float between 0.0 and 1.0, indicating the ratio of variance you wish to preserve:
 
    
 
 
-
    pca.components_.T[:, 0]).
+
    pca = PCA(n_components=0.95)
+X_reduced = pca.fit_transform(X)
 
    
-
    
-Another very useful piece of information is the explained variance ratio of each principal component,
-available via the \( explained\_variance\_ratio \) variable. It indicates the proportion of the dataset’s
-variance that lies along the axis of each principal component. 
-More material to come here.
-
 
    
 
    
 
@@ -233,7 +230,6 @@ More material to come here.
   
  • 28
    • 
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs025.html b/doc/pub/DimRed/html/._DimRed-bs025.html
index 44a5f82f1..dc43d79f6 100644
--- a/doc/pub/DimRed/html/._DimRed-bs025.html
+++ b/doc/pub/DimRed/html/._DimRed-bs025.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -184,33 +182,15 @@ MathJax.Hub.Config({
 
 
 
-
    More on the PCA 
    
+
    Incremental PCA 
    
 
 
    
-Instead of arbitrarily choosing the number of dimensions to reduce down to, it is generally preferable to
-choose the number of dimensions that add up to a sufficiently large portion of the variance (e.g., 95%).
-Unless, of course, you are reducing dimensionality for data visualization — in that case you will
-generally want to reduce the dimensionality down to 2 or 3.
-The following code computes PCA without reducing dimensionality, then computes the minimum number
-of dimensions required to preserve 95% of the training set’s variance:
-
    
+One problem with the preceding implementation of PCA is that it requires the whole training set to fit in
+memory in order for the SVD algorithm to run. Fortunately, Incremental PCA (IPCA) algorithms have
+been developed: you can split the training set into mini-batches and feed an IPCA algorithm one minibatch
+at a time. This is useful for large training sets, and also to apply PCA online (i.e., on the fly, as new
+instances arrive).
 
-
-
    pca = PCA()
-pca.fit(X)
-cumsum = np.cumsum(pca.explained_variance_ratio_)
-d = np.argmax(cumsum >= 0.95) + 1
-
    
-
    
-You could then set \( n\_components=d \) and run PCA again. However, there is a much better option: instead
-of specifying the number of principal components you want to preserve, you can set \( n\_components \) to be
-a float between 0.0 and 1.0, indicating the ratio of variance you wish to preserve:
-
    
-
-
-
    pca = PCA(n_components=0.95)
-X_reduced = pca.fit_transform(X)
-
    
 
    
 
    
 
@@ -231,7 +211,6 @@ X_reduced = pca
   
  • 28
    • 
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs026.html b/doc/pub/DimRed/html/._DimRed-bs026.html
index 1dacaf321..a4e1b973d 100644
--- a/doc/pub/DimRed/html/._DimRed-bs026.html
+++ b/doc/pub/DimRed/html/._DimRed-bs026.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -184,14 +182,18 @@ MathJax.Hub.Config({
 
 
 
-
    Incremental PCA 
    
+
    Randomized PCA 
    
 
 
    
-One problem with the preceding implementation of PCA is that it requires the whole training set to fit in
-memory in order for the SVD algorithm to run. Fortunately, Incremental PCA (IPCA) algorithms have
-been developed: you can split the training set into mini-batches and feed an IPCA algorithm one minibatch
-at a time. This is useful for large training sets, and also to apply PCA online (i.e., on the fly, as new
-instances arrive).
+Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic
+algorithm that quickly finds an approximation of the first d principal components. Its computational
+complexity is \( O(m \times d^2)+O(d^3) \), instead of \( O(m \times n^2) + O(n^3) \), so it is dramatically faster than the
+previous algorithms when \( d \) is much smaller than \( n \).
+
+
    
+
    
+
    
+
 
 
    
 
    
@@ -212,7 +214,6 @@ instances arrive).
   
  • 28
    • 
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs027.html b/doc/pub/DimRed/html/._DimRed-bs027.html
index 0aeb436e1..20413ce9d 100644
--- a/doc/pub/DimRed/html/._DimRed-bs027.html
+++ b/doc/pub/DimRed/html/._DimRed-bs027.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -184,14 +182,28 @@ MathJax.Hub.Config({
 
 
 
-
    Randomized PCA 
    
+
    Kernel PCA 
    
+
    
+
    
+
     
 
 
    
-Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic
-algorithm that quickly finds an approximation of the first d principal components. Its computational
-complexity is \( O(m \times d^2)+O(d^3) \), instead of \( O(m \times n^2) + O(n^3) \), so it is dramatically faster than the
-previous algorithms when \( d \) is much smaller than \( n \).
+The kernel trick is a mathematical technique that implicitly maps instances into a
+very high-dimensional space (called the feature space), enabling nonlinear classification and regression
+with Support Vector Machines. Recall that a linear decision boundary in the high-dimensional feature
+space corresponds to a complex nonlinear decision boundary in the original space.
+It turns out that the same trick can be applied to PCA, making it possible to perform complex nonlinear
+projections for dimensionality reduction. This is called Kernel PCA (kPCA). It is often good at
+preserving clusters of instances after projection, or sometimes even unrolling datasets that lie close to a
+twisted manifold.
+For example, the following code uses Scikit-Learn’s KernelPCA class to perform kPCA with an
+
    
 
+
+
    from sklearn.decomposition import KernelPCA
+rbf_pca = KernelPCA(n_components = 2, kernel="rbf", gamma=0.04)
+X_reduced = rbf_pca.fit_transform(X)
+
    
 
    
 
    
 
    
@@ -215,7 +227,6 @@ previous algorithms when \( d \) is much smaller than \( n \).
   
  • 28
    • 
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs028.html b/doc/pub/DimRed/html/._DimRed-bs028.html
index d3de994a0..a221d679b 100644
--- a/doc/pub/DimRed/html/._DimRed-bs028.html
+++ b/doc/pub/DimRed/html/._DimRed-bs028.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -184,32 +182,14 @@ MathJax.Hub.Config({
 
 
 
-
    Kernel PCA 
    
-
    
-
    
-
     
+
    LLE 
    
 
 
    
-The kernel trick is a mathematical technique that implicitly maps instances into a
-very high-dimensional space (called the feature space), enabling nonlinear classification and regression
-with Support Vector Machines. Recall that a linear decision boundary in the high-dimensional feature
-space corresponds to a complex nonlinear decision boundary in the original space.
-It turns out that the same trick can be applied to PCA, making it possible to perform complex nonlinear
-projections for dimensionality reduction. This is called Kernel PCA (kPCA). It is often good at
-preserving clusters of instances after projection, or sometimes even unrolling datasets that lie close to a
-twisted manifold.
-For example, the following code uses Scikit-Learn’s KernelPCA class to perform kPCA with an
-
    
-
-
-
    from sklearn.decomposition import KernelPCA
-rbf_pca = KernelPCA(n_components = 2, kernel="rbf", gamma=0.04)
-X_reduced = rbf_pca.fit_transform(X)
-
    
-
    
-
    
-
    
-
+Locally Linear Embedding (LLE) is another very powerful nonlinear dimensionality reduction
+(NLDR) technique. It is a Manifold Learning technique that does not rely on projections like the previous
+algorithms. In a nutshell, LLE works by first measuring how each training instance linearly relates to its
+closest neighbors (c.n.), and then looking for a low-dimensional representation of the training set where
+these local relationships are best preserved (more details shortly).
 
 
    
 
    
@@ -228,7 +208,6 @@ X_reduced = rbf_pca28
   
  • 29
    • 
   
  • 30
    • 
-  
  • 31
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/._DimRed-bs029.html b/doc/pub/DimRed/html/._DimRed-bs029.html
index fd62370d4..b53f979ac 100644
--- a/doc/pub/DimRed/html/._DimRed-bs029.html
+++ b/doc/pub/DimRed/html/._DimRed-bs029.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -184,16 +182,22 @@ MathJax.Hub.Config({
 
 
 
-
    LLE 
    
+
    Other techniques 
    
 
 
    
-Locally Linear Embedding (LLE) is another very powerful nonlinear dimensionality reduction
-(NLDR) technique. It is a Manifold Learning technique that does not rely on projections like the previous
-algorithms. In a nutshell, LLE works by first measuring how each training instance linearly relates to its
-closest neighbors (c.n.), and then looking for a low-dimensional representation of the training set where
-these local relationships are best preserved (more details shortly).
+There are many other dimensionality reduction techniques, several of which are available in Scikit-Learn.
 
 
    
+Here are some of the most popular:
+
+
      
+
    •  Multidimensional Scaling (MDS) reduces dimensionality while trying to preserve the distances between the instances.
      • 
+
    •  Isomap creates a graph by connecting each instance to its nearest neighbors, then reduces dimensionality while trying to preserve the geodesic distances between the instances.
      • 
+
    •  t-Distributed Stochastic Neighbor Embedding (t-SNE) reduces dimensionality while trying to keep similar instances close and dissimilar instances apart. It is mostly used for visualization, in particular to visualize clusters of instances in high-dimensional space (e.g., to visualize the MNIST images in 2D).
      • 
+
    •  Linear Discriminant Analysis (LDA) is actually a classification algorithm, but during training it learns the most discriminative axes between the classes, and these axes can then be used to define a hyperplane onto which to project the data. The benefit is that the projection will keep classes as far apart as possible, so LDA is a good technique to reduce dimensionality before running another classification algorithm such as a Support Vector Machine (SVM) classifier discussed in the SVM lectures.
      • 
+
    
+
+
 
    
 
 
      
@@ -209,8 +213,6 @@ these local relationships are best preserved (more details shortly).
   
    • 28
      • 
   
    • 29
      • 
   
    • 30
      • 
-  
    • 31
      • 
-  
    • »
      • 
 
    
 
 
diff --git a/doc/pub/DimRed/html/DimRed-bs.html b/doc/pub/DimRed/html/DimRed-bs.html
index c3c8a51c9..396bd951a 100644
--- a/doc/pub/DimRed/html/DimRed-bs.html
+++ b/doc/pub/DimRed/html/DimRed-bs.html
@@ -93,15 +93,14 @@ Automatically generated HTML file from DocOnce source
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -160,15 +159,14 @@ MathJax.Hub.Config({
       
  • Proof of the PCA Theorem
    • 
       
  • PCA Proof continued
    • 
       
  • The final step
    • 
-      
  • PCA and Scikit-Learn Functionality
    • 
-      
  • Principal Component Analysis
    • 
-      
  • PCA and scikit-learn
    • 
-      
  • More on the PCA
    • 
-      
  • Incremental PCA
    • 
-      
  • Randomized PCA
    • 
-      
  • Kernel PCA
    • 
-      
  • LLE
    • 
-      
  • Other techniques
    • 
+      
  • Principal Component Analysis
    • 
+      
  • PCA and scikit-learn
    • 
+      
  • More on the PCA
    • 
+      
  • Incremental PCA
    • 
+      
  • Randomized PCA
    • 
+      
  • Kernel PCA
    • 
+      
  • LLE
    • 
+      
  • Other techniques
    • 
 
         
       
@@ -227,7 +225,7 @@ MathJax.Hub.Config({
   
  • 9
    • 
   
  • 10
    • 
   
  • ...
    • 
-  
  • 31
    • 
+  
  • 30
    • 
   
  • »
    • 
 
 
diff --git a/doc/pub/DimRed/html/DimRed-reveal.html b/doc/pub/DimRed/html/DimRed-reveal.html
index bb2c0b941..be7cae712 100644
--- a/doc/pub/DimRed/html/DimRed-reveal.html
+++ b/doc/pub/DimRed/html/DimRed-reveal.html
@@ -1148,23 +1148,7 @@ chapter 12.4 and discussion therein.
 
 
 
    
-
    PCA and Scikit-Learn Functionality 
    
-
-
    
-
-
-
    # Now add PCA
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
-
-X_pca = pca.transform(X_train_scaled)
-
    
-
    
-
-
-
    
-
    Principal Component Analysis 
    
+
    Principal Component Analysis 
    
 
    
 
 
    
@@ -1173,14 +1157,34 @@ First it identifies the hyperplane that lies closest to the data, and then it pr
 
 
    
 The following Python code uses NumPy’s svd() function to obtain all the principal components of the
-training set, then extracts the first two principal components. First we center the data
+training set, then extracts the first two principal components. First we center the data using either pandas or our own code
 
    
 
 
-
    X_centered = X - X.mean(axis=0)
+
    import numpy as np
+import pandas as pd
+from IPython.display import display
+np.random.seed(100)
+# setting up a 10 x 5 vanilla matrix 
+rows = 10
+cols = 5
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
+display(df)
+
+# we center it ourselves
+X_centered = X - X.mean(axis=0)
+# Then check the difference between pandas and our own set up
+print(X_centered-df)
+#Now we do an SVD
 U, s, V = np.linalg.svd(X_centered)
 c1 = V.T[:, 0]
 c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
 
    
 
    
 PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
@@ -1201,7 +1205,7 @@ X2D = X_centered.dot(W2)
 
 
 
    
-
    PCA and scikit-learn 
    
+
    PCA and scikit-learn 
    
 
 
    
 Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The
@@ -1210,9 +1214,11 @@ that it automatically takes care of centering the data):
 
    
 
 
-
    from sklearn.decomposition import PCA
+
    #thereafter we do a PCA with Scikit-learn
+from sklearn.decomposition import PCA
 pca = PCA(n_components = 2)
 X2D = pca.fit_transform(X)
+print(X2D)
 
    
 
    
 After fitting the PCA transformer to the dataset, you can access the principal components using the
@@ -1221,7 +1227,7 @@ principal component is equal to
 
    
 
 
-
    pca.components_.T[:, 0]).
+
    pca.components_.T[:, 0].
 
    
 
    
 Another very useful piece of information is the explained variance ratio of each principal component,
@@ -1232,7 +1238,7 @@ More material to come here.
 
 
 
    
-
    More on the PCA 
    
+
    More on the PCA 
    
 
 
    
 Instead of arbitrarily choosing the number of dimensions to reduce down to, it is generally preferable to
@@ -1263,7 +1269,7 @@ X_reduced = pca.fit_transform(X)
 
 
 
    
-
    Incremental PCA 
    
+
    Incremental PCA 
    
 
 
    
 One problem with the preceding implementation of PCA is that it requires the whole training set to fit in
@@ -1275,7 +1281,7 @@ instances arrive).
 
 
 
    
-
    Randomized PCA 
    
+
    Randomized PCA 
    
 
 
    
 Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic
@@ -1289,7 +1295,7 @@ previous algorithms when \( d \) is much smaller than \( n \).
 
 
 
    
-
    Kernel PCA 
    
+
    Kernel PCA 
    
 
    
 
 
    
@@ -1315,7 +1321,7 @@ X_reduced = rbf_pca.fit_transform(X)
 
 
 
    
-
    LLE 
    
+
    LLE 
    
 
 
    
 Locally Linear Embedding (LLE) is another very powerful nonlinear dimensionality reduction
@@ -1327,7 +1333,7 @@ these local relationships are best preserved (more details shortly).
 
 
 
    
-
    Other techniques 
    
+
    Other techniques 
    
 
 
    
 There are many other dimensionality reduction techniques, several of which are available in Scikit-Learn.
diff --git a/doc/pub/DimRed/html/DimRed-solarized.html b/doc/pub/DimRed/html/DimRed-solarized.html
index 4ec81ffb3..0925346bb 100644
--- a/doc/pub/DimRed/html/DimRed-solarized.html
+++ b/doc/pub/DimRed/html/DimRed-solarized.html
@@ -113,15 +113,14 @@ div { text-align: justify; text-justify: inter-word; }
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -1088,22 +1087,7 @@ chapter 12.4 and discussion therein.
 
    
 









    
 
-
    PCA and Scikit-Learn Functionality 
    
-
-
    
-
-
-
    # Now add PCA
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
-
-X_pca = pca.transform(X_train_scaled)
-
    
-
    
-









    
-
-
    Principal Component Analysis 
    
+
    Principal Component Analysis 
    
 
    
 
 
    
@@ -1112,14 +1096,34 @@ First it identifies the hyperplane that lies closest to the data, and then it pr
 
 
    
 The following Python code uses NumPy’s svd() function to obtain all the principal components of the
-training set, then extracts the first two principal components. First we center the data
+training set, then extracts the first two principal components. First we center the data using either pandas or our own code
 
    
 
 
-
    X_centered = X - X.mean(axis=0)
+
    import numpy as np
+import pandas as pd
+from IPython.display import display
+np.random.seed(100)
+# setting up a 10 x 5 vanilla matrix 
+rows = 10
+cols = 5
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
+display(df)
+
+# we center it ourselves
+X_centered = X - X.mean(axis=0)
+# Then check the difference between pandas and our own set up
+print(X_centered-df)
+#Now we do an SVD
 U, s, V = np.linalg.svd(X_centered)
 c1 = V.T[:, 0]
 c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
 
    
 
    
 PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
@@ -1139,7 +1143,7 @@ X2D = X_centered.dot(W2)
 
    
 
 
-
    PCA and scikit-learn 
    
+
    PCA and scikit-learn 
    
 
 
    
 Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The
@@ -1148,9 +1152,11 @@ that it automatically takes care of centering the data):
 
    
 
 
-
    from sklearn.decomposition import PCA
+
    #thereafter we do a PCA with Scikit-learn
+from sklearn.decomposition import PCA
 pca = PCA(n_components = 2)
 X2D = pca.fit_transform(X)
+print(X2D)
 
    
 
    
 After fitting the PCA transformer to the dataset, you can access the principal components using the
@@ -1159,7 +1165,7 @@ principal component is equal to
 
    
 
 
-
    pca.components_.T[:, 0]).
+
    pca.components_.T[:, 0].
 
    
 
    
 Another very useful piece of information is the explained variance ratio of each principal component,
@@ -1170,7 +1176,7 @@ More material to come here.
 
    
 









    
 
-
    More on the PCA 
    
+
    More on the PCA 
    
 
 
    
 Instead of arbitrarily choosing the number of dimensions to reduce down to, it is generally preferable to
@@ -1200,7 +1206,7 @@ X_reduced = pca.fit_transform(X)
 
    
 









    
 
-
    Incremental PCA 
    
+
    Incremental PCA 
    
 
 
    
 One problem with the preceding implementation of PCA is that it requires the whole training set to fit in
@@ -1212,7 +1218,7 @@ instances arrive).
 
    
 









    
 
-
    Randomized PCA 
    
+
    Randomized PCA 
    
 
 
    
 Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic
@@ -1227,7 +1233,7 @@ previous algorithms when \( d \) is much smaller than \( n \).
 
    
 









    
 
-
    Kernel PCA 
    
+
    Kernel PCA 
    
 
    
 
 
    
@@ -1256,7 +1262,7 @@ X_reduced = rbf_pca.fit_transform(X)
 
    
 









    
 
-
    LLE 
    
+
    LLE 
    
 
 
    
 Locally Linear Embedding (LLE) is another very powerful nonlinear dimensionality reduction
@@ -1268,7 +1274,7 @@ these local relationships are best preserved (more details shortly).
 
    
 









    
 
-
    Other techniques 
    
+
    Other techniques 
    
 
 
    
 There are many other dimensionality reduction techniques, several of which are available in Scikit-Learn.
diff --git a/doc/pub/DimRed/html/DimRed.html b/doc/pub/DimRed/html/DimRed.html
index 4d2c857e4..60f22749e 100644
--- a/doc/pub/DimRed/html/DimRed.html
+++ b/doc/pub/DimRed/html/DimRed.html
@@ -118,15 +118,14 @@ div { text-align: justify; text-justify: inter-word; }
               ('Proof of the PCA Theorem', 2, None, '___sec18'),
               ('PCA Proof continued', 2, None, '___sec19'),
               ('The final step', 2, None, '___sec20'),
-              ('PCA and Scikit-Learn Functionality', 2, None, '___sec21'),
-              ('Principal Component Analysis', 2, None, '___sec22'),
-              ('PCA and scikit-learn', 2, None, '___sec23'),
-              ('More on the PCA', 2, None, '___sec24'),
-              ('Incremental PCA', 2, None, '___sec25'),
-              ('Randomized PCA', 2, None, '___sec26'),
-              ('Kernel PCA', 2, None, '___sec27'),
-              ('LLE', 2, None, '___sec28'),
-              ('Other techniques', 2, None, '___sec29')]}
+              ('Principal Component Analysis', 2, None, '___sec21'),
+              ('PCA and scikit-learn', 2, None, '___sec22'),
+              ('More on the PCA', 2, None, '___sec23'),
+              ('Incremental PCA', 2, None, '___sec24'),
+              ('Randomized PCA', 2, None, '___sec25'),
+              ('Kernel PCA', 2, None, '___sec26'),
+              ('LLE', 2, None, '___sec27'),
+              ('Other techniques', 2, None, '___sec28')]}
 end of tocinfo -->
 
 
@@ -1093,22 +1092,7 @@ chapter 12.4 and discussion therein.
 
    
 









    
 
-
    PCA and Scikit-Learn Functionality 
    
-
-
    
-
-
-
    # Now add PCA
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
-
-X_pca = pca.transform(X_train_scaled)
-
    
-
    
-









    
-
-
    Principal Component Analysis 
    
+
    Principal Component Analysis 
    
 
    
 
 
    
@@ -1117,14 +1101,34 @@ First it identifies the hyperplane that lies closest to the data, and then it pr
 
 
    
 The following Python code uses NumPy’s svd() function to obtain all the principal components of the
-training set, then extracts the first two principal components. First we center the data
+training set, then extracts the first two principal components. First we center the data using either pandas or our own code
 
    
 
 
-
    X_centered = X - X.mean(axis=0)
+
    import numpy as np
+import pandas as pd
+from IPython.display import display
+np.random.seed(100)
+# setting up a 10 x 5 vanilla matrix 
+rows = 10
+cols = 5
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
+display(df)
+
+# we center it ourselves
+X_centered = X - X.mean(axis=0)
+# Then check the difference between pandas and our own set up
+print(X_centered-df)
+#Now we do an SVD
 U, s, V = np.linalg.svd(X_centered)
 c1 = V.T[:, 0]
 c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
 
    
 
    
 PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
@@ -1144,7 +1148,7 @@ X2D = X_centered
 
 
-
    PCA and scikit-learn 
    
+
    PCA and scikit-learn 
    
 
 
    
 Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we did before. The
@@ -1153,9 +1157,11 @@ that it automatically takes care of centering the data):
 
    
 
 
-
    from sklearn.decomposition import PCA
+
    #thereafter we do a PCA with Scikit-learn
+from sklearn.decomposition import PCA
 pca = PCA(n_components = 2)
 X2D = pca.fit_transform(X)
+print(X2D)
 
    
 
    
 After fitting the PCA transformer to the dataset, you can access the principal components using the
@@ -1164,7 +1170,7 @@ principal component is equal to
 
    
 
 
-
    pca.components_.T[:, 0]).
+
    pca.components_.T[:, 0].
 
    
 
    
 Another very useful piece of information is the explained variance ratio of each principal component,
@@ -1175,7 +1181,7 @@ More material to come here.
 
    
 









    
 
-
    More on the PCA 
    
+
    More on the PCA 
    
 
 
    
 Instead of arbitrarily choosing the number of dimensions to reduce down to, it is generally preferable to
@@ -1205,7 +1211,7 @@ X_reduced = pca
 
    
 









    
 
-
    Incremental PCA 
    
+
    Incremental PCA 
    
 
 
    
 One problem with the preceding implementation of PCA is that it requires the whole training set to fit in
@@ -1217,7 +1223,7 @@ instances arrive).
 
    
 









    
 
-
    Randomized PCA 
    
+
    Randomized PCA 
    
 
 
    
 Scikit-Learn offers yet another option to perform PCA, called Randomized PCA. This is a stochastic
@@ -1232,7 +1238,7 @@ previous algorithms when \( d \) is much smaller than \( n \).
 
    
 









    
 
-
    Kernel PCA 
    
+
    Kernel PCA 
    
 
    
 
 
    
@@ -1261,7 +1267,7 @@ X_reduced = rbf_pcaLLE 
+
    LLE 
    
 
 
    
 Locally Linear Embedding (LLE) is another very powerful nonlinear dimensionality reduction
@@ -1273,7 +1279,7 @@ these local relationships are best preserved (more details shortly).
 
    
 









    
 
-
    Other techniques 
    
+
    Other techniques 
    
 
 
    
 There are many other dimensionality reduction techniques, several of which are available in Scikit-Learn.
diff --git a/doc/pub/DimRed/ipynb/DimRed.ipynb b/doc/pub/DimRed/ipynb/DimRed.ipynb
index 312f09d3c..c76867fcf 100644
--- a/doc/pub/DimRed/ipynb/DimRed.ipynb
+++ b/doc/pub/DimRed/ipynb/DimRed.ipynb
@@ -1308,7 +1308,14 @@
     "\n",
     "\n",
     "\n",
-    "## PCA and Scikit-Learn Functionality"
+    "\n",
+    "\n",
+    "## Principal Component Analysis\n",
+    "Principal Component Analysis (PCA) is by far the most popular dimensionality reduction algorithm.\n",
+    "First it identifies the hyperplane that lies closest to the data, and then it projects the data onto it.\n",
+    "\n",
+    "The following Python code uses NumPy’s **svd()** function to obtain all the principal components of the\n",
+    "training set, then extracts the first two principal components. First we center the data using either **pandas** or our own code"
    ]
   },
   {
@@ -1319,38 +1326,30 @@
    },
    "outputs": [],
    "source": [
-    "# Now add PCA\n",
-    "from sklearn.decomposition import PCA\n",
-    "pca = PCA(n_components = 2)\n",
-    "pca.fit(X_train_scaled)\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "from IPython.display import display\n",
+    "np.random.seed(100)\n",
+    "# setting up a 10 x 5 vanilla matrix \n",
+    "rows = 10\n",
+    "cols = 5\n",
+    "X = np.random.randn(rows,cols)\n",
+    "df = pd.DataFrame(X)\n",
+    "# Pandas does the centering for us\n",
+    "df = df -df.mean()\n",
+    "display(df)\n",
     "\n",
-    "X_pca = pca.transform(X_train_scaled)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Principal Component Analysis\n",
-    "Principal Component Analysis (PCA) is by far the most popular dimensionality reduction algorithm.\n",
-    "First it identifies the hyperplane that lies closest to the data, and then it projects the data onto it.\n",
-    "\n",
-    "The following Python code uses NumPy’s **svd()** function to obtain all the principal components of the\n",
-    "training set, then extracts the first two principal components. First we center the data"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {
-    "collapsed": false
-   },
-   "outputs": [],
-   "source": [
+    "# we center it ourselves\n",
     "X_centered = X - X.mean(axis=0)\n",
+    "# Then check the difference between pandas and our own set up\n",
+    "print(X_centered-df)\n",
+    "#Now we do an SVD\n",
     "U, s, V = np.linalg.svd(X_centered)\n",
     "c1 = V.T[:, 0]\n",
-    "c2 = V.T[:, 1]"
+    "c2 = V.T[:, 1]\n",
+    "W2 = V.T[:, :2]\n",
+    "X2D = X_centered.dot(W2)\n",
+    "print(X2D)"
    ]
   },
   {
@@ -1368,7 +1367,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 12,
    "metadata": {
     "collapsed": false
    },
@@ -1392,15 +1391,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 13,
    "metadata": {
     "collapsed": false
    },
    "outputs": [],
    "source": [
+    "#thereafter we do a PCA with Scikit-learn\n",
     "from sklearn.decomposition import PCA\n",
     "pca = PCA(n_components = 2)\n",
-    "X2D = pca.fit_transform(X)"
+    "X2D = pca.fit_transform(X)\n",
+    "print(X2D)"
    ]
   },
   {
@@ -1414,13 +1415,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 14,
    "metadata": {
     "collapsed": false
    },
    "outputs": [],
    "source": [
-    "pca.components_.T[:, 0])."
+    "pca.components_.T[:, 0]."
    ]
   },
   {
@@ -1444,7 +1445,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 15,
    "metadata": {
     "collapsed": false
    },
@@ -1467,7 +1468,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 16,
    "metadata": {
     "collapsed": false
    },
@@ -1514,7 +1515,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 17,
    "metadata": {
     "collapsed": false
    },
diff --git a/doc/pub/DimRed/ipynb/ipynb-DimRed-src.tar.gz b/doc/pub/DimRed/ipynb/ipynb-DimRed-src.tar.gz
index 0fc16f2943f090385c00ed06c478d85118d5a686..3b0146384d3d443cbd22ca6ed4e6e5c15c42ddca 100644
GIT binary patch
literal 191
zcmV;w06_mAiwFQVjj>$-1MSaE3c@fD1>mlGia9|^nkKG=cHu&h;ssKY+Ne!xl7hXx
z{Q+GmZi)!`HjglmFjE%Wd|xE~?xWQpgjklrlp99pMCtOLVN3xNN;5L{7z2a}VQBuHia9|^(xk41cHu&h;ssKY+Ne!xl7hXx
zeSoeMH${Yeo1bBZVWup$`MyZ}-AAiI2xXMQl$(sriO5pVFlK;L#wa0_WEc=mQ$`{{
z>z(w{JJ0QSN^_RlN&R+i9INXOdzM$=nSbKgCfW~g*AfO6@&b$I_sUZ?u9A2=JnO8a~B1U#Zy6XF?KuN?F3tY!FM`rYndVcX1`J>+5Y2TQ_F&9Bx7o1DWw`+8HTsYInld`Hl=GQPR~d@T5GBprIq*vE3~(<=+z
zDK3X5U-No)8hBwwVIK
zH}`(DEZdsP{i#RJUBM}GsyNg(DLdkuVK^Csc;&iEh?E>nL3;Zvcq@vTX%cHXWE(Y_
zonda|WROW`VkSqIGJ;;jD|s0JE^#$oVlZwwUK2Bh
zpq`aLGsBW6m7KhqFVye@As*!yVUGIE*=HXftu+Xvf7ho9rCFE=7#)1zdk?BwlG(XU
zOu@c`eaL9e?JrsOAR{v0S(8caVz8CsAe|it9!8phmaV<}F&O4pmcl`YY_da21D_pw
zP6jy0Hfgm=tW9mebg<-(ShECRUt*Nt$AXyR)Mo-*FH3$HgIV_dOt^|+hqK^rJAPmq
zMI}tQ2&Qyiq*QQyET=1;|G}--Kti>J83;&3L11uo@i&N?VzCNCF`Co4>1OR1UxZ>9
zmBf#VoZ`6_Jh1vE=4N0{HtB+R+98OQKlGox+qIwlq=K6O!n!vy!Bis|ZgDJB1pR?b
zT1kjtELYE{xhAoH&_K+#tB8s+-F^qsbh)DTDG=Z&@kD!qZq01wQkm?;CTqD+Cg(gl
z=jtQ&f=N-chPbR}vOWLPXVPboE7nd(WjR*}_!w}{{}Q`=RmIX>V7w5Bw{=$ptF$nL
z|MJ(-$_T)u)`&V};{dXLL~*rua)mN2flLdVVTk2dQ;w&FCh84veh@oUVMZNv5hCb*
zudBtYA{GF*-*9&-_K;+mFyvKcNZGOJu3h%;W6p-A@
zF~;6|8qv8mVDA62n9>Q@UY`iYXwxAf6_$@+v3J76`;$+Vm-7TS3k{fabwt46WhcOV
zO@JsHRz1^S3z&{@!(sz~)@DvLVb3Sye|nkNTaripI~PO}zT3K7nUkOm`?GTfUlcEC
z5&$Z+J(I@oTp=REX+8%J&c7i<@3&a`f-0T7yZrM~g3KxzoaIKKN_IZ+i4w6v+~L;PmZNlrHN`(cL%V-#6NMnfXcchTqP`A?M-c~&
zOyg-e302X9cw*RDipJZ7aHDr`EKbo~f&<>T&u@yAKAz+2i=G=)(Z8L~W+erI6>I{X
z1l9LS&uWR0VzbpJT))v4aSR#|B;rdB;q^@l{5Gk;_Rz
z__?gCM%N0VV)4+xL_*A}F*n^IAw(MR!dxT9K{OY<6dY0c44zj_Ozw*3Xz}PW=K3`%|N;6P&orC9
zN`)3RG+xCQUqp!{cFKbb4`P!ZBS3^PnsgFQu&Eh0gq0sWueuE>3I1Xn_XGXBvV9jI
z2`62_kHiE9Y4jS9Hw2c**TU--7Y^k4lTOLLp~Kx`SzOf!9@{kbwpTtK&{Dem{OG@A
zhqo{S<;O`dzs<{Vz|}U6TZ+|a#uG9s;4tGaRVpeS?3an&^l~~MDO!0w3y@!4DGVKl
zrsm*$*x<;mCe-xN`~A~SLaT}d-Dr3g%o+hp1)PCO0zvjf`|VKUoTEfJ!PDy+?)MrJ
znYy=^(20yToRrXs8f-AJ5)_wwcrZVTl<3u_NUnnM#g+av)(d$GK>#pl4&9(
zGjHvNR7|xw0LyfZXpM6);
z$}Yp_0N9(0Q33E_}jdGpO@4nrSg
zrfs@1l{~U?pa2XnwEGjXO+CHn^0h0AwadxL3DLz%w4y{RT6)%SoBxL8CB7v1PL2h
z9A{B=o=%x2afI1fmTy3+UA7UVR+eelI$L2u@4Wr?{{?UyEv?^Xy~{GdpA|q0xetC+
zR*VR&iNwZ2&s66p*zUi%opI_xWuL+y3=sce;`mfo*)Nmm69YaYUxWx3CsiPV_7Vhw
zoQ6^mjsgBB>ZiRat!I*&cU*uUulG>3c+g?eAQnb&#b7Dl%tYC0qR!-(-dMst21f~+&Q8=-fs0G
z)~V19`#p^C{vwfH!iuu47yKYq4nx*;e&_C}m}|gkA`Ef&t113hddPPUUlW?mvL~z*
z_FxA?$sdPF>|Mf)HaKMZVP*8Na8A;R-cI|Hpkm0^>q{8~y;BzUiSG-v$
zCB1-bN^iODN<_R=EH9?sWG7;N5tisYn=Ksyb(9u+f1$5d_HNYyX?sG(Fqj)
z7RVb5d~+Rtx!P|J*vS!9_t_SOrWoZcDMisIA^jBEzUQG^&k{H#gK*iO46|)d@vmiT5!Xiz3$>EmruCL30>@JKUWI@k-pU&>TwANz6{YTN-KP&z&TSCEToS_B{k
zcu^k6Vc*^E2|NxeBZRaYhuBtJ_kY1)YcuBdp6Ol9R|catiHsbBKaI2JF<(^Auf;!E
zOJdg6pcEB~rKjmf&=WLUx?|Ea2(M3vxi0+K(HR&=A9$O_tgU|{qYDiKjbbT{DtKyf
zRy4nBZ}3U{U08PmXJ&8e;_7T>WC!;jbTGDt<6%!uBt!?};!JMvqyf}wYdZiZQ2p}_
z3eyc~Mvj}1_2CbHtdjpRZxk8RCYgW$WoWe>L#sSuA-bub)zn)~p_Y9@NGQK%d{URx
z^i!9^dX6@|EBAF@S?zVfNXk{LD2Iv>3gZ?l
z$8Lme?u)6`JFjp;hb(``Wjr4o
z!d86y>gzh9uBz(xSvf5RH&lU^DI0b>1FmPeY!t$|in>1M`8}xoa{5Zcco&SDmoNIFT}9oXmcU#UNTq(k;W764F-Pn8SB2{L
zSSf1m18{uAw0h&!20;XNv#J|g`AQpoTd93UcP?2*kYN!cs-~K9;@KpLlCdwo-czfr
zA5467%Em_i0T6F~@%Cliw_Ziqy#srfSu!6tD4KIZ5h+W(7m_YEZuZ0a4h}0=8rfXQ
zu27@lcKShnb27xbqT{*pT5|fW6pp{lP}pLtI+T=E(m~yeA}i)vdqQCGv~tJ3)1v{t
z)5FPDL4#o-Nsj)=MrfWA^Uy%TH2{;a2ogw(R@^#h3b1v~<0vAo@!p__WJea*#KBt-
z&5~?KUek5oY1w3`H@h?1h%(swWnQzf?N6XPtKn7@%FYBNx3Hh!dTD=9Lq@*hzTMN}
zpxh!pfyQ!fP+F79`+;NE!9|FFy0yM6-<#|KgvHJY`f=ZNTeL_YpNd8}?Hi`xw3|PU
zmtLkk21pX?eug{KdyYhMmJ6|S8P{n-vQJ8L!@NGrgzEGd<`MvK`!nr|;4^0e7|%xI
z`7s@gy8AqD*keB^U9HI
zx5V0n>6>SHF~eZ5Vqwo-UEc2YIBUmxwM|>x16EIfGp=Ih!0GAuiY$)kGW^~cI2L)O
zDj#a7S&~KGHy7D1-@kC5Vj73MT7~jGM1Bw>GVU*06S3{hiQvoP9Dz2u5nhqW3>PX1w2gYg5m_`!)Cmhw@_tTsZq}f>J9A8l7PyQ{wyUR0GJDNI_$#rY-
z09A4)|JS+ur9Io=_!$M?ROLH(O?6LAvqP%^|2y^dXNiyUB;b;es);ra9kq^Xzh
z>G!sh1s|acLO5!5PfjioPS5=TdcQ#^iP1tEn)C$3O?+=oPVz+JNF@C|B`n=1Z9dohvD&0uQ9w*iFs&YY5UFxkBV4~D?{KpzmbbgN0H%a!yke_6j;%-PNzDx>46fwT;(m4c|6=@q%{BMYUBDfBaF_&21%V3B>B*BBRv0WAiwI$Z&QS|
z`xW=r_KxD)#16L_C1sZh6c2QNtPGF95LV(rD*B;7iAknji7XC{Fvy7u2i4Q$w`+3V
zI5KAKp~ozS7xc~xi*B^$cH6-mVCC>Jz%6e3Pv<*+$cb74p%3!q$_*@Ifz3}&<2Xza
zwjR>C%FOaK3v}ZEL_MZ@OnaC}q?q5iw}Xo4=v;p!{S*(LBu0!ZtOuZ9QegLTcl60n
zQqdw$FZlq5xbt*kws#7_zwDs~kdfms4KtBBNy6MOx6R3C$Te&sMnLAQ0C$0aX^?NC
zUvv!o^E8`%NA>yyV(T{oONwQy??af{p&;hspmxo*_;-S)7;cxwpf)iXSA%?snjw*y
z0mrl=4zWH-^ze6aA1QOyb=6cu(KJ#25Q-
z^>ybF@Q5&BoA|3<1!e2ufQ42lB29VOiU4S4OW(a15!tNDpM8hC71rZE2lkpDcA+H!
zBBfa_0S*~%Pb@znp&BSbeeXT6zyo_kb(|KFbBvpeG(VN<58o5l7n1kHpaeREg9lLi!|9S|Ux3H`n3*#n)e-KB0VFHutRbX#
z-{pDHD7XZr8R~#8$ZJGbPO84l3LK^mYb)@`VL}7bEN0rN?-+5t3Jpnc_QZxUqg=YV
zedgX$#YID$`*5mOl`Q0!YL8)L4cm-CtAd@Iockvx*b=iPK-JWMLO6G0EobLemK!k4
zNs~Pj{ef`eKt3p*la1b8IG~!A82`=8kDLcLh-JB|CobrR8I@txBC@v!SHk4QQFy{;
z^mpxAbvy3lLKtd2y6G>yxm2PGN|k!fa7`d9c$yRvhS5>
zn_4bkeGAJWK#*%Oy7EYy+_@GGpGu0!BH++xmE*cLL`oj5I17d-GYp@$iCxTn@H1Za
z&#C%|4%*=jxti?ToQsYVIH4f7a>=v@J&w%tMZP39EBqggRK>#6#j3O{CcCDaqhcm5
zc!}-+M>iK
zBZZO;h59$A^=t5BytvW8B;hOCQXe@IDTl&+j0_bHiP9_ik3~AWtNEvIz~}t-n>hw=
z>SuuER0WAtNyg**dJY!mgW6E*O^tX8=w;9FEu3rvy={0
ztFq*|Oi!f)zP{W96h?!7!GVLh8~ys2J+&+}MNJTsBA*sn@^?%)D$OV31+8OOkt0P=
z)cL|5!~KdiD?Cond(la9+TRg^$9zR-MSz7llh_z8oan=29&x
z04UtZ8;1RM3S9RF9%34aH0;t_5<_o?Y3u>D3YK9X_pB%LRUeK~2eOUyE
zemP9<=c9y~%+mD43MNSMl`QPSyuPChkD$==Z30D0cxKW?ymCH&9)uE1mFKVZsbzzc
z8-NR?L>}SPEZHs)JM4ZUlTc3D{$CpSe@S2#W;POLlK+U|<|5C0Xi)BC&~o%<^?+(h
zQ08XPazblR)_)z0b~aQ{J==8UMsSSw5E-a>!ImOiFkyR;*c`;iux+-6h8=bcwXH)WOEWo<7A!I~SM$gY9E%gPcbwJC<;^
z#iU2%TP$1TV}x<`vOu^SXpp-9pR{lKUJd~Iav1Q3x_P(EO?V35I|cm9ncHEadj
zMd&rL;874!E^V-pdr9&TM5wW-SJIs#BQGJNED&f?uVdDj0a_-AN|44Y()^&I!(GWE
zY@k0OVGuS5^9<8KP+ySd%+NWLNw6b7DQ&eEIwGiIWn)zl1esN}5FyT6+AK+!2lb@^
zo80)7kj%V<&~UVS1nyNN@+XbF{89rrsLbhw(1$;7C8T>;q(|UV!HiVYScI{B)Zmn;
ztD)hV5Yjvs%*_T|quhgWF&^mA+)E-@M4IMeD2?zda4rK}LFg1(6e0bF7OYsMq@~ff
zgsn6f<*Ve*&7_hMYln>5f6tZl&<@uC2rGXXwvzD@LBi6>Vnj(KsALrgLkmEK!Uo^;
zJB)~s`DgTHaUO`=4-Wh5a~{lF-Zj2H_sIt@2pTeA0XeJ=PWm`2WxBs`3RhRre~b^SICz4z{Gzu7Sn=|U6)aEf$c
zC1vHrj}t#D1ThT>lbn=7R_nD??mcZATM>v}K6E`}jqTzqh)g`o1e5*p7t~
zGgP?=GqD~G=KN+3=3Leuz$qQA4!K-q={^6eS|?wH>DQHAv~3eM!@?0K$&hbxz{Zr>
zwQ>gHMP2HN3`J+5v8SIhfTgF$NU-dU%Crhluq_^K+Fok=$>iSt4$<)=A+=QBaW)l%
zi{6LgFdNQs>haCLj;kUX?DEF@6KNHGzjfo?fg;5(r~k#W*KwMD~j$xl}z*EThLAQYdrL<
z!-wQr-a>LahG#PVpwW}`8VD)uS5k@GXqu{0ebz@S@<*AYqMZe|E8NaamgHoGWpgg5%E)0G{Kr%5%~31XKDb-AdG>?Lm;3e0
z7u9>CR-(ItcRSa$-j98vT}i<0PCFbyB-ER$Km-DYVt6q&PjJO1z$l2eZpzbcJtfwC?X(27u(cc!0{TInopkRo|ha*xaq8#dfL0gCm-=_p--mGmrAQhfFwz-Zj2
z?2_mkN*#EE3~+%gDvag!kM*P37db@V&l`-Q{ylDgahJ1Mt^S3}=eo`SCc;_@m&oV)Lbwv+qwyNotKz1nAy;*wN
zr;!qUqgYV*j9LL)l!YAEk&#n$f45Q@lDuq5+!th&8QaUvHT}89(Ske#(Din+}&RA(hHk}wViE7?YnDb
z?VsPfTV<>G)Zn=vd~2@j!nt1e)20WB)1z~m)0C@hQtdQs`MO~#Yb7hCv4E7wlr$0^
z-~q;(N){F=81}Q5f_u?4OVX*_1W)Y2d(Vj(qA841{?Xz_gb;7B-2}0C(Nq6o?8w4GQJMNxW@yp($tL{c#SglC|8`A?%9-xfPkng7Ro15a4La;R^ED(w?0v
zkqvK#5O}RqRUA<46Cj3iJ5H}fr|nkPJ@>6C;gb2fTna2Uwc2kL9~~$A*%S%mT}cpd
z+5^;1Lt!O+Y66-lCXo|YthXysmn>q4gF(gosjONeGefApz*cHFnL(#Zy+7`OV+ByY&so~xo}1r
zyy3(|mTxzbn=+S7Mxw1E^aL*oEGx{M^t9NtYY17&Mde(L)+;GA;q#`zt;mF}@q&>%k>*T;)>1b8+
z+=}9S){$)1A~IT4+e%0M{iw1Ky8Nb+d!0b+dOy5=YRi&)#CLQuta|SCczbs7{N3|?
zxt{DJQ)I!<56JuH;co$_9{GNAc(8xZv>f#81BJJ8i5D=t6$^TV_e;>@U!Q>d{_m4`z^uN4VmLa*%~p0?y!*|vMEne8
z1xO}#n$`JsS?u=0#n<-lNBygvUO^P(SM*cWyJAnuylCtL3imG=gi$Id>Nqhf!DtgF
zQiU&WzB5W=@-HfXR1+s!N@LnTBe;HNI3D@fot(uqBidZ}^6?`Ga||P$FLuzr68g}6
zn(@~=06+hRAVmWx1Ckfgm|4g$Cbip9_}|9wgH((9FV|5%O4wj`sC0;dFN<&^t4f6k
zhx@Mlxdr?Jgjh;K@tjV*Xv+q;bFZGiao40T
zJH@B(I$?f%c=I-njV#{z8E5vFfz+H$L~RW!APuum|F^))8=ii#wWrWbSa`oHe9*z-
z>C4c?@q3Z~Mc=eNF`+7zMRYKyNR$6#toka$}-Z7@E+~mQzo;^>rHN0k8mP0VlZCIQu_$_V}
z;OY5xW=TD^?W<4s*O;AvI93o)*E^&V>ER6pAe`R$c(|Vm4#&?3o%*2)$s!J&Umv73
z=Y#5_-hc-h-#lZdPv=j-@A-T=zeww^+a(Tv0|!p=UQ*v8
z6ZZi9gUxZHCugs=26I_4l!R(7QO!otf8oAAf;-QDA10dgnVL
z+q9&ChDs>bX)Qq0dZ%(|89EqCe2VCh*9S6qC
z)%zOvp3&5hcJa=?FW=DAe)dlP508NTf4=q)mcf`UK-rT;0x;UC1i&TCll~b4fyLl+
zh@fo$u?^bI8o>$RK{@`*sXgFUu%KN3WwbGHJ~&XebcH$a2q+ddHf~n#^sPB?!vD?R
zbKr5H-05-i-~_mAa7+?b&MvMbY@Gj4&dlDz)slpjnT;d4@n1%#t)mL$M)#Yop97kf
zueC9z)l}8ilIlho!7ztR4Xp5Ih(X-&I*qxT8v;I8n+O}s2n_p{`Vyq9TNU!t6l4S8
z#Q8=2;vM7nsmT0=?^F|nHx#8S%KbjjP=05UrdZpA0;m<>F}@08n9mCvu~Eh_Zj@)qRipBFvdW-xB{y3U?zP{e44~`e_YqP?i46bBmG6Lta
zM)ujsidqN}lA!;tVvWhQWsG@-wq+W9wG-J|n6S+{iEjtzIyxB^o%t=d$Zj+;RL#$q
z^T!n5J^2Fg@DCmhQ?l17l$)Rps(FG`U7d)_$dZ$h8xZ*%LBg3sOt^zl?2QnBrkwCb
zU{KddW^m7r-yI!XE)jbU5k^oh8ImsGu;ecuMhn!L!f?2U3Ey&tu-{qiXXz|gk`cjh
zk==JS&nNi~#uwh-dk+t`qiO!341=1s`_~d}LEnIEhO1gL8|*l^HibP53CH@pU^iizS^|B%(qFhAXSF;R-c7DwO<552jp3ojPj3Ws6
zx;pI>CtINFLFio%9N~IuIebib+N;5hv<65vaS>A`TMCYb7IpJ&{kdaWq2~1&y65!<$AGp(Ey`ZoQr}%b%dT)OLpzl5
zX~w~(QEOK+r5k$nv!;FRi}uHvD1x4tM$S!O@
zQAHWqy@V|9-!~s81{f9M4D!hAQs^cM4|B{5(x0Fwn*+%?#HwX5E}UPuK2(!OZa9=b
zV3BWOzWsghD;=}LfB-rB&d%7fYI6~kHrRWD=(I_UdSth3-}weldn}~C#}7S%mPB~L
z$AJ)_rn1usW-e%E(cHc*;;;*jXn@#Z98G{p28N^0mV#(cN9}@cnINaOhM8|x7Rocc
z{;x~Q$ztZb0^}kH0<_Zo5D2h8(Y7eHQ?7e%f5Ip9&-0Q4LBunW=+JM$&E|a+WqM@x
zv!cGpJv|+4V8HVVtJv3EZgtsgHL89EzLYThaL_UEBrFddCe<%;hG5D;Bn8l9+)Oc3
zX052ousa;K1oY)>k7jI9QC@MZJD;#V*VZ~xvrKLFSZ*k!Oo?RR@&y4;e8SVYh&{rF
z6@x`j@uDKEsBfSfh)cEYDPZgK1i0a
    {HT95LnpzE6XBX&bH90vuiLk6z5(=7+jR
zNFu51aB{ksYxMgNf>XJA%K>VbOwk_Te7n0lf=HI}Ym^0uNXUZQ
zHTLm7$eo!M)Y})VQ=b~<_?OGbW_*o1OgUl2*%#Hykc~b4HKVIC^#HTlpB$d>BtAj|
z)ydJ#^ZMWgzk=&at^2M?6C&mJd>)SuuSJhZZBKKlY#`nkg_%5s23GT!83h}W3}OWp
z9l*A2W|6RSUFVhc%i~wq;5h?l{TMqugETN}Z`up~q`x`HiZ;hB6pYDr>rNWXZ3CyO
z>kTy54)H8&WoXX{)&b=Qi;0KX!tSp}2YctPp|&&n*bJ?hk9bTMkc0nH)`wl7<>eYFbm|NW)OKlNU`0A7^t7$JUm
zLpKwN)k5=WDKWxH&eQ+SebmrkkP{cWLD!cgc@taO^v{2Jd^988t!ix>a^Ds_ML+7c
z)S#=C(+?|RasSozdwNZ9$G`bwZs6)?QpkyWQR$vNl{2uN9#x7ao*^Qlz-V>ol|O<%
zwO@3}LaXVQWg>vZpgT8>bKjlRW5I(DHTkb@3nBp`(0;-l*ujmXQdzPYsYc7i8MX+1
z>`s~*D9ig(D)3rt`b~b5BOp=X$~^u@?#4C+!r4+SbArM)9QfIieH1g?ihN_Ja{Cq3
zbfWx~%Jdc1(-yO*Q_Qxhi@ndbaDw8oL`;S!mnnRS{U!hrcm1%=fjqf8Q-ZNG+p;7n
zS^=SHm)S2!sjBpoN%==(X%O}|8LbfAS*-$e>KCt-4-G>#r0J1xZ(HeKOx6|Dlrlrela#SHXcqf77kH%2@Wx45l$Ws76~41F;;F-W*#;cc0rQ=
z|A-#q|IsGw>}(|evRBi)Y+7R0LMnQ-dbABk&Y1s0#vaSY3Dh)B9KOi)I3@eF&tl#-UO%M!pj0)G3QYFvQ%%f~g{xhJ>
zJx<_9H#-a&X8$bS0g=8BCpWC#0kw*$h}ZWKc3{W7ol@1S&|Wgmlw&VD>(4tQb@TOn
z8uwr{W~CtxEM3frk_6;jSG_z%r~Ev-xWIb+fn5405S}IcCu(P6Xnex9uYMlSuiEb+zlm2lpE!qeG!IyF`)aNFaw4wVyYtyM{L2@=ospC2(!2Tn2@=P?3^{RvW
z$qEsxCp_7#cNS?IZ`B)6sRP)Q=H$7XQ&Q{F!k*>_4D}nL!HRZ8zJAy8vKn<`Q_Lhz
zN#Se|1xt;|T6!mxcvG1ZY+Qt(?oQTk8Ael}JeSs1_Pa9+qx^FXRIdSoG+u{tCa#+w
zS=&IzhNS}yycYT{lmGZ)JM+e98x;mT8_gD*Jx^E{AFr0P^Mm=+T`&x2<%%_oFsmhg
zQ&qk9oSUqKlEjvL&voON#3pYnUMkbn5y4|G1|s(HYzBR8HtCcL*w-6yFXfLF@)|oc
z*KX0gP8hiqBeGc#Q3U}PY^a(LIH11)&Thd!zHEL~n-hF_^%<7zekx;`C>@rDdl19;
z@a;w`aAvsO?|A0i6r7-@Lg9A+in&g@@#C7XN-P!7DaM)tjiyAN7bH%|qTN{#?bW1g
zV9sM21_jdyEgGelU8wGa+NO%y=*o4w>RpJj+YzLST~BF?yXAoNAhUK%{tTDD#u{@vXHM#h7pyiBjtRlQ0bO7>~?6
zvb#=>AwwxQ0;$#^F%CE3|B@P|0h)b2D~zNqA2`@QbscXk%%C!8#{EK!d1Xc#{VreZ
z>uuEN_(T)uZ0mu!LqcAAjX#%8SU1WuAf<>(=3PaHDK;}$!7Cgoer&LLPF$drgob=X$vN7t@EkV|C8<0fZ{CxwoXE1o6^_vTj=97fEMKtP*
znxhO5M-&hUbE-h^Er7PC&2}ogc{yC;d(rW<8~Z{b<0E%FeGR48X=q=VZA+(S6E~M`
zH~-nePw|3)%-AC7y^iPW-dE$z`cle%N&C0MgSLyrInTGXzQd5)5;aAUUo}L~hzndo
zj}OwAgoPP+5yrqn;dL3K3ET1#h)(pzTqQ3RVT_82XrqSk5l0rmc;
zMm-|=^^1#jd6`Z~tbtEWTeFACrE_MQ?Rg8u$(A5+Bb>l0{}t(dEvxfsv5+M~VH%x}
zb_Kdc_%1Cb$lDt8d-V=S>=#D^S=c2`I2H8F5MFEBZ>*{OKe;(i2-C2?X4J3o=-jRO
z81Be#g(Ax4L>j>}(N38xzYugdm1w7fX?fr7hc{+qv9@KAgkoM=+G~=TV7tP5W4`Wc
z3tK0EbFvVAOi#w``{2V}9-dZ|Dkt^NPP>PYspI#Llwk)Dl;I8~NPh#%OS?fc7$k>M
zIv9J-ekyBFeb=rOU4%@$9<&$;xbI#$lDgh{E)aO(pnp6I)cB=0EAgP`v~kThz^%0htN+zaBrfAUqP&2?-98Duq!q=U-xtQ@qQy}
zmI9vqX~%_E6%^=+J0^6L&Q>s7c1HjpLeaJsuAT0q)s>JFQrhg#t5nHTn*d^wuBI<4
zn)oN(5jw4lw!*Jq9Blj;`wCT_mVJz`k_2|N4GTpY9aHrX3^Qw-9@uHaz9lh7&D(Kc
z4O-m9$*u3txcSVfPHqyLO)x{_{1qgpdd$8)t&s|%r$4EvhlM@oJ{tU?R^2`TfP0!o
z4Py2#VO{1PU2%fOZa{?GgthAuiNC{I8Ee@(ittw-8-d
z?F22S!S^4Q_cYtBt0n~2YRZuYcF0nobc*NBPZui0{*cGg>WS4$xVLVXrzAQcT|Pni
z7kA9v&u#`FgQ&#S1hw_G4xqxka?%owVITh{p{04*Nk?>QCr>q}UqsFkJ~iSijFyXu
zbmmfrtOBfBIrnhaEAGDJHuebytJqSUpZYb5NgNy^8B+LLSuxVFeVVV8`+;c
z40u_3;44uCSnC*?xJ>(UlA2YhlA9QpW>5(U)cxtJdUw-kX?EJ3fsv2B?fzTZ*nDVP
z=IUKu?Uf1*XVssUA9jrAL&c&u)qr;})o3tL(^NM-N*3*=dE(K6FMZ@OI*{XI0kgMU
zX)g`84y^za5n`e^hJnwfY)`J^GX0zqL$tZ7B
zmCKym2f88YPmu>UddyUqrhXmCZH9k(RlQ)k?+tfmG5+i@@0m}eIj}dtvOfq$izpyp
z!78lC-IPV=#}g=hDfNqDEGOD7SrVD0Sqq^_9@Io5PR7Pu2<`C#
zY9_yRn3{v|Oyh3pJi$o=_z#a?HqX&lVs_o_l33xFC(u!h8J=uTd$BAjap&9P*>VIcR1$xK79YZVk`C)!8W8Y
zo|$>UTn(hsK6q&O%5lAk)wbf`suiY-s0LJRElmd5-lXtz$f()~&%KP3$nH4H`EF$g
z^D&cC*4!ExsUcq~-nuUOwIjF$>eq8cn3_5I${N{@Ltb<~9oFFkJIP8Ym{)~=4s`ms
zn1RK_MpN3a$Eo(P^ht6-3U~k3PFO`F3O;(xyl-yb-8J#BgE_?3xWH{ons)8?3nC^n
z<%!!}qU_fP>bwrhGfxT}V%}!llXt5jNtEmkP3gq)K7{Jv;#ZMdeJ35zyvHN4
zh{;^pJ(xwuVTBD^SSKCLZTrO+KE&uk&8Wcyd$S*ZdEbT|$&&l)KV%Le)x?=eyG>Y!
zJMuwYpAYLtES*8&rpRuv{_60aNC`hk?*^|4?Am70D1ZF~UT;tuL2MJ^K8E7SYI{Ai
z)ys~8W&u87PI3F2MlD_I_c{m7z1*!zpWili3RE+E4xe3bk7<$6U?_Hx*bxTfIyY4H
z?nL14O%I{1p-`h)?qCZp^YjW~9nUj|>DLuqwX}G6oJ7`N>qF@5P1X~%zwV_~+!1-h
zUkI)-W1EeDB3Z(-(8l}Vo=b}Sym>h&R3`|mCsfgG!BRv*X~&#)x?XsziFLnVG#<
zXJuIyO@%BCn}lp-96d=RiMpxJR>$~c>BetG;;-CY^ks9m6IqQ&Gg=T}
znLunlrWgH|tg3((oDVNojG*(5p&u?+lb(P@P8yaB#hJ<=;5*BGz>4HE0
z$$W_dgUvrSQI}OWFFJ01vvAq}gHDHf{Mbw~N9)ZjH;O^pmDaxrXfSCx)Si6<;@0kw
zHu_Q|%MMWr`?)i4B5hNmu&L=dL#N%`GXVG4O|r?%xgs
zaRk=it0`P4upo^r09B%V!jJ-&#B>Qj7;?xdY|V0Tel#qxl@T*LTPY=-E@$IOoEC=E0s~x=lUdJ
zv?$owry?VQg`>HfyQ_tX!~dO}Ol=YPxswx#vD5qiK+*%N^>pL++t35@jd#<3S`_Ou
zCJEqv@TdrZ{#iWt}bZa8S^v(~92U;i$dMLGhiI@4e~g_~b(mTHK``{TAp&QRB>;KzDc*2msH
z<@w7iKJhK)n^CR)rRJZRi0p4WeM2-%N5&_s9xHY-S2CA-4MLcS*=%Y(*4ww~d%#i%
zp>`~_a5*5gwz}j%iMn?m(!<^$Qdsn_Od@VxyWCg6bKzwx0lUO>v{6M
z1KY1N#tS_M2H8mCQ2#x-uSjG)D@U{#U$<4Fwh$LAiEDXoJbCi|Xc(>eV<^%RFpR8F
zL82=RG%23OA8lm&&1pGk!>QdL;^1J+6X)t4-@W+>+l&(}-G7^zbGy@|)y0qwv(ySdH8h|5i(_{uZ$0)UE=B!x
z#5yK1(zXx_{IQw;dpqw8G7I%2L;DJ
z?H>ZvSt%zIVN+BuI=slh+zW}5sq`bPN+k&0as$nqL>_wbZ;Ap)(^+yowi*cbwUun3
zB9WnG?68WK@HGZ>R%e`pJF*S=XmD%%ih<|WFC~zju3v`ME%L@FlB*CV|D3`VBRinH
z**YjEq*Iaf{I)(@`ijZJObC*!?>-?YLI~x6BzXSm;D*L?M~Fqa#_>GFfQZ+5h%sYz
z->7Zw#!~>3x+*7JS?qdR1G%Uoy@3umOOh)Mvkoi>VL}~fdkuk%d`@nvVil*22%*Z5
zl9MtO|to;%C`62Nt^VayV4L-AM
zDjnnbALbIt45a`hH=L`Mz-UvvM#7m^>-7;#?{PV&)D}V{%F}hC~rt
zlEoAQJ4AAO`-%E~72LiQlFZMG3lS^c6^%htoqsM133hVn^?XJQfjPjtn>ML%c1`IP
z=Q|sQmn+1XT|@oHH0LPs0+**pA{L=uoO;Zj9KMhCWR`m}Iawl?E>|MV7$x;}AquQpccxkioL(uTwyc25
z+QkZr&+m+x)mLj^9M}7fUwvhHYUO0x6
zj*`uv>iGqW!_q71<)R#-=P2EW7LkN~lNrs6_i)Lz@+p-SWm=Dp5LsB9~I9V%o9C}YsTPQ?^Zen+jQ`w5@qEMXh
z($q%yZFean*NS#GqB82W`uuni%4h5~jE;Q~`Gs=SA^C~Oc%FsYQH+041ohcnarXS^
z^{%b|RD%ab1+y1>No+KKArbXA@!et_on;5*0$)@XUmQ56ba_v+E~x>#g;~6L1qcA+Ru}?BN>)@$ni{_h?4Q#!popAp&1`SQC;?*%Bw&E
z72V}zUOTL<3e|~VSs(BU<{5HNRVHUT5p^fE=mp`-*+QQT;Ad1Qll)=E1Fwi_ayIs9
zRzO@nAo=?wXd-77IV`&f&R`~Dmwzw?u|TZzT;c8@mDPl%pJX@qJicaU@r^YdJo!%x
z>6v4P0rZ@#uDhD-R(EJb$}iR2J&AxcrpqJs;QRwP*J4@9ll(dy09t0
z*h+E^U(;4l<1FUO+TCDeKOGw(?+mH
z#yTxZ`rqotD97@FKJ(&AWM#jM!v`CY!juN$uWiepp};SL
ztBrNn9G}cXoEeX{-5!VWcgq^R7LL|%VQM)0e+(TUTj#OnRoR-jBMgT9FVHyr$It3Fk8VBxbLFs=Hg~lZ(
zX+lk8nBjSUeB8)9mRB4flS-|R#!o`~t->jchvZ^@(-ui@{V_BCbm3)t2J)(&>C83|
zV6~h7wY-09_dZbT?KHhU<`8G51_$q8V-4WZV@iR7NV?$<<&RqPP;TgY5-C!b3Xzaf
zxyfz$U`1LGjW6k0OEB_glWTXst_eplaMXL*2yv&pJl!j$wTg7|MwZ>Y3}D3$O^CQU
zA`%xa6oe^aKizAORrD3@0GdLGkREWyZqs?iG-eJgla*YdU^>zu*xu7@)N~QmZo8eh
z=~q#OW%BmOgD}Ide$s1`)~t1nO6(iE^H+G_B*SW~|BCfuaz9M;-pOm74K_;gF1LM|
zoT)No@-Uq2fN>Kf;Y)GozRn`Hp^q+Kp_~?ZnG9eUb#+u$kxMsiQ>9obffml`MQA_H
zn6E_ZnD1-%CF&wC^OP^@P29cqC&CnrT-~nOE}iDEi0LvGrnr@eJHGFQFi7NqRPEsg
zJEV^=DbaHKdU;y>2E#I>Fif9agusMg<7&+HDuo5-PY3sdz-rV9{7VDQo=%Pqj@5`;
z25$?_+4|q?Vh_&!KU`~>Kezx2IPd?0r)Y2*;{V;mZsn)|XGQ>F4}G;1p5dod5G?
zYqdCpnn3#I5Ck_k@Bf_JC-qld33+k+kF_866-zc&S}PVby<;z883|1!Hs
zBqGRBPx*c3_<|={N2TaDm$c^_L8v(8yyr76JZ9I+zYZ*2tZ(fxAgf
zM80AT*Jr5OaW`aW`pTMVQEHfxA2->sb>HEw%?=P9%#s#65*=WYc%qDh+NSmi$Z+eF
zQ>D?;--jXSfWj@5KZu~C7)hR>(x4Dv-uFLELv*afIsvrdJScP+bW^J|RKW^7=w4D>
z(oM97(&*E)OmXB1Vq?5om0*Jec7Y6NjmmonbVN(JR5kE;3}Rq61*qrjvhIQM5o!Z&
z@YA#>0EpgDW(9izfWc||0-ps%AFL%prRq$={E7(#PH9FfQFWKXufJi&722%~~v#B@=!K2Ui*G1x$LC08A6d^UqjUvmd<
zA^b97ufOS}0ErEf@Sg-Wnle<#9FD%+0^nY5udpB|=JE@QQ
z?mZhRGym9ml>aUDEZg<5-(@$5x}WSwb;Wvg?$~3*op>O<2WZtZV4~ebupC65?$|$l
z+1fL^*Gi&>s7-k2L?odhq{YyqE-nREsP%9pGgI()53Hp<$kDCfF?BQ9%Tc7
z78()a2QGC6w;)$`{ppv#)C1^W9l|JQ{q{VwmXO`U`6ZOL2vCb?;E4*rRy>>n4L(_KYY=2S*z1;bd*
zs#nEOt_9$fUDDfBC*hFN@?EFMCZdXVZwHjVG(((`rF#dm37^-D-peIzB;f&#jA_Et
zTPOZ4+=MD3qc**pdq6}Le>HAil0!dXH8d|MT(ztJoNj29@2Q5UZ8H*`c!crL4FVa~
zZtwdPc4_&3H>bGDqI2&wQ}`hK?XTT*BZfSuI&w?JsA*d-a4Jh-+t`F@WF;9;mmr|B
z`x&GRRIB{kf2FHaokPCl`3Dd93XO$#hz&ig&%lPP{gZ%%^?8)wP|s;TA7dG%tv9a-
zg-XH0;9PhAn1=s$tdPAUn@(vXcN!P(n7P}x$ipBju2W`j%hY>=*_-NBSqMjE(0>E0
z{1?|Fl!6w0bJmtN^`<;hoDbpYWzvGf1CFZ`HlQe#n3*
ziqu%Bu(vn2X8E`h2H>
zOD2i$FHxyjce)yY9I#GG5XSHN$!T>aP_mU042`TCb>>w%+d
zdVBW7C6RJNNa_}DbzC>lGok#k#=|zD+@%6K7iQEJ3GvmW0mlXi2b8Aqx(V#&9q*(@
zPU%YGS1s~oq?ob0C{%2t^7X%1WxOchw)l!!nTPbV8_$RE)8>lh6Mc2=5%S*d+N_Zg
zItj>D?!evN+;3fW>V9?r8xbpjCvU1xp&90xjRGbRVt=SHpVX2%La5$kCHD`6b?+t?aEZq0LDGsa2YcZxDk)k#Vm=)SqZm8NUKV7c6T)T
zh{yP1@kj7sZ(uC^qY24gGff32HjU`>kRo16Z>E>4$qB{NaZ)s(G;#ZfevGFe?{Q|2
z3M_AK279*9cJeaCwwYVd3>61DL7Y#nRgZdF|27bby{IXnDd<2L4Kmrm{G7LGpa7{-
zfUOB{#rCNyF?>P_!b^$-upJ-e8UFJ@){96|?nyy4TqU6`PJTIkA}7`X325I}U~u8I
zIq_?>iZyys#H+R(rAC(I<|73u6>r2MZKwLk%K&g_WUrj)@@$WjH<50r_Y@&sKa4>y
zVhQXWvp6C|A|{~9qMNjhvj>u^Wjsbf66@PG9dHW63KR4-0*k2=7E3eAOhpk($v)-5CV)L2>F@mrA4rmOkJ
z0Xpa8e&e>2B<*|2)iCuTt|e8W4?KX#G%)D2sgVD%TM-Y*-{c0|N_5jF)J}3(m;*@)
z7byKjrj^;I)sGG_Sd1PCgkX0GV$`R{tTIV7`Be{@%KsAbwCj}y=B{_!gJ6nf_34c|
zlBPLu}et!=+__PMy
z6(j?pvsnnsIi2z{6@`-8@!pPB4}d-}r*30H$}E!d_UfH;(qmHc3JEzbE(rGYN{Hw|
z(eFaEf@(*2a;`K7InuZDDJ*u%){mAs9|ppI%*uj~$4+$P6o_(O67;yeNxH%d7aBQq
zOJOl!TKH4kVmXBW*e~kul<+7->Gi~~TU=7P^`=JZCn0T7O~JMbdqcKQQ3F%7hs=N1
zbyA9aBHgHOx)~7}?j58N0W_i1IMP0?%>g(3lt)jBoYSxl@`;%A^aSB7(Xvw-LiMrk
zvY2y4f2=r@a!g>xhY4Hh8NI{Dez$duc##o
zL2QMj3JsItNvl#**={sq6+rgpWLma|bA>h81b0oB3qB%UWMVPl%S+Fa*&}s)$%1F*
zMYXX
zKT{H!n6rNgWbYF$d+aoY-XGOvw(KMtgO8|QB}cFG0t9u>reWE~VQZ=~4UMIvY#)w)
z2dRzh)b-a=xt@CMuLAXImM_~$yd(S)4t@3IF5V_u(tJhAm?&L!`RXhTr^sKnO7{w1
zar>yoEoE3KWo1I)>48bbugxZw0NHN
zWg2w?2CjW
zsC+E|2gtM^#g$%2+@m(KH?(zT%`kfAlX(OEzS^?$a`t%M5(~V`uj7iQ
zXwPr$1=rbfV<;EhGla~4$h(?{&BAwXs80F&VJm1b`WP|x$pYGE-T|ur`?LVIWXObOzvRESX
zSv!)WSAkybq-G9Lg?U?+^WPNkKCtzv+ye~*%L?9L#>OB!P2}tML$5o9RvNl36&04v
zr|acM!{_=Q{d}h@yiwcY$t)_b3Yn1wusKaIwh7(^j~QFI&c}9oXMkC(F^$c+*LWhgZB4by)!$
zNk0A^7XtNP{@|qk?4nm`m7}JsmmAG5#+lpmL`F>CjRwbRt?Gibfd}XNUj@T3qyPdq
zSGvUvWF$Qs&wrolxEa|vXxZ3k=@D3!oXn+6+%3rHqy#wF*!kJnIJo{-l8!SA8Ta2M
zVQUD2kbsK|fmPbZ)yIrIVg}0*sZJYw$6j;zg*jAwh_}y
z%>osfscgk$JnZ5ZeW-AuOkxe6w5nbtp&qqq{LL&oMkB7)qM>9g1);+7CDYY?zysA6?Z}L~
z)F2t2b++P(LcaB38<~8_qT_nkdqmGEijH6-&dBityZ1*QPMAPCK&|NC&WHan
zn=`K*bVn;B{f&P&0=SMB-NL#YeLS8vejB$A(Jlh8j9Z7b%*Vfei|DQ5U!q(LkQ4R8
z=L|rHI-^SB(ahVxQL?cqNr0Xso84V
zbo)@fY-jd-uL~`%O`V0T-l`G{Y9=>D1-n?xqXpoc3U-FORyWE5QpN
z_yY~aB~G$Q3W^!0;*pBYR*Xgu?Nlu=tC|?cc@S0{y;(tbfX|u4REq6y)6_g
zH+t(_=Qfs~*;}~Szs*{@L1OI53S^EoNyDegMY8XnhpZ(1WsWrAj%A4&N{<(BLe$SU
zoF3_na^!*+p&KOZ^a;9)5II?k{|)}7BAL%pnjgVF(HAj6`(+G{BDMMvEt`9rZ~LdT1(3VGJBwj~x_2?JQ1Xqr
z;-Qdue}fwoJs5(J;kh<2EXJ==rRrHCfc@Ltavw&RL-{vwDmT-ut%(TF
zHhdrJ2Wi>w)(uQJP$rpd)Ac~>OvT9Uy3u3g=ycOvGAGaJ^Vp;MS!)tGG_jUGBwZdE
zHJ*47Z701rL0pju-u{UbQZj^c{JWa$$Ir|oj>r1Z28R9Qh5tIBn<`6MIg+qjIDa(i
z9XM1zXK4uVbO2A}l7+xc+qqi3LQ!O5?S!`rZ$OWli9NRX<3>Ruaz{
z`GUxo2ft?
zd;7E^g-YJxl*N_4{;sMi&a<~#Qz~8hCmUxA^g8Qve9g7URbGzY8Lq^#s?yF7)*Y!TY7x>ukO?-SnyxUrAt%eu3)%HThIzrOf$lG-j^03@4dBEiljNyUx-N@;
zrbC^!aZ2@|`Ws^9?#tDarK&j7G^^b$G_zlm2DcO<8Bnw@yRkC5KLD~okMmRS@6kC6
zL}o9%U+q@ThVF=AD026`d92te#KXfoc#mh8%kAs=&EovI1H`7?G74_4>#UklS^VHGwY_!A+^@?MDygO*_1yaCyLR3Q?w{h7^S=4(aioH?PZKEiW
z=_TX_vjO!+4X^Sh3)}@5g+JAk`Fy(t(0Sit^iV*dAeW2`g+Z5
z*pGgQ*InFBrqPi;tJ@R-{#@;0zZUg$P1+qv?a>cAtn~iqx^#Sd-_#_EcGmounx=^#Vck#_G~bfUtH#^}7uyW&Erw^x`&TsfeozU}b{ohuDmpdJ9
z4Uz)LV~D`2VdG;##?H%wz^YBA&rZfk#{M6v?&ReDAIV9^My8IyD(7hF^dI8-e@Kc<
zUx-5-0AQEkl;QyJvGedtv+=S6_{90R_$8zyIJqSy_(aJ5|0RaV|DToN;^HU!A1gyV
z*Qt^sjy1e{9{{4Oq+=qhVWcDbrHQ54RXqZjg@!&nLdK?#5{_bDb1E9rnaR%3S<_uv
z)3p|t+{^#zWU}y?do(Jzow=UQX=(Y*(ddOa#s*n>F+@rUZ)>4-4MLJgZ6bZ@H?a<<
zI4I2X=0-EYy(palgDp&TLNZfM%LYDPKCALcts#2!;U~^hWS2rx(X!g1^i-i|l{Zi%
z78yrM*qWSVw~VZ9y&*WFA!?GhNUALvKQI8XJ_8(_q2eTX-AAOu{fuadLRd~BNOVtI
zwS5cYIsMX;*~l25eLT?OK=61$Ac~lEHJaAy88TRHPC957FzS
Aga7~l

diff --git a/doc/src/DimRed/DimRed.do.txt b/doc/src/DimRed/DimRed.do.txt
index 1ab6c39ac..163230d7f 100644
--- a/doc/src/DimRed/DimRed.do.txt
+++ b/doc/src/DimRed/DimRed.do.txt
@@ -877,19 +877,6 @@ chapter 12.4 and discussion therein.
 
 
 
-!split
-===== PCA and Scikit-Learn Functionality =====
-
-
-!bc pycod
-# Now add PCA
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
-
-X_pca = pca.transform(X_train_scaled)
-!ec
-
 
 
 !split
@@ -899,12 +886,32 @@ Principal Component Analysis (PCA) is by far the most popular dimensionality red
 First it identifies the hyperplane that lies closest to the data, and then it projects the data onto it.
 
 The following Python code uses NumPy’s _svd()_ function to obtain all the principal components of the
-training set, then extracts the first two principal components. First we center the data
+training set, then extracts the first two principal components. First we center the data using either _pandas_ or our own code
 !bc pycod
+import numpy as np
+import pandas as pd
+from IPython.display import display
+np.random.seed(100)
+# setting up a 10 x 5 vanilla matrix 
+rows = 10
+cols = 5
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
+display(df)
+
+# we center it ourselves
 X_centered = X - X.mean(axis=0)
+# Then check the difference between pandas and our own set up
+print(X_centered-df)
+#Now we do an SVD
 U, s, V = np.linalg.svd(X_centered)
 c1 = V.T[:, 0]
 c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
 !ec
 
 PCA assumes that the dataset is centered around the origin. Scikit-Learn’s PCA classes take care of centering
@@ -926,15 +933,17 @@ Scikit-Learn’s PCA class implements PCA using SVD decomposition just like we d
 following code applies PCA to reduce the dimensionality of the dataset down to two dimensions (note
 that it automatically takes care of centering the data):
 !bc pycod
+#thereafter we do a PCA with Scikit-learn
 from sklearn.decomposition import PCA
 pca = PCA(n_components = 2)
 X2D = pca.fit_transform(X)
+print(X2D)
 !ec
 After fitting the PCA transformer to the dataset, you can access the principal components using the
 components variable (note that it contains the PCs as horizontal vectors, so, for example, the first
 principal component is equal to 
 !bc pycod
-pca.components_.T[:, 0]).
+pca.components_.T[:, 0].
 !ec
 Another very useful piece of information is the explained variance ratio of each principal component,
 available via the $explained\_variance\_ratio$ variable. It indicates the proportion of the dataset’s
diff --git a/doc/src/DimRed/PCAexample.py b/doc/src/DimRed/PCAexample.py
index d1f9885a0..94cddcdb0 100644
--- a/doc/src/DimRed/PCAexample.py
+++ b/doc/src/DimRed/PCAexample.py
@@ -1,31 +1,3 @@
-from sklearn.decomposition import PCA
-pca = PCA(n_components = 2)
-pca.fit(X_train_scaled)
-
-X_pca = pca.transform(X_train_scaled)
-
-X_centered = X - X.mean(axis=0)
-U, s, V = np.linalg.svd(X_centered)
-c1 = V.T[:, 0]
-c2 = V.T[:, 1]
-
-W2 = V.T[:, :2]
-X2D = X_centered.dot(W2)
-
-pca = PCA(n_components = 2)
-X2D = pca.fit_transform(X)
-
-pca.components_.T[:, 0]).
-
-pca = PCA()
-pca.fit(X)
-cumsum = np.cumsum(pca.explained_variance_ratio_)
-d = np.argmax(cumsum >= 0.95) + 1
-
-pca = PCA(n_components=0.95)
-X_reduced = pca.fit_transform(X)
-
-
 import numpy as np
 import pandas as pd
 from IPython.display import display
@@ -33,11 +5,27 @@ np.random.seed(100)
 # setting up a 10 x 5 matrix
 rows = 10
 cols = 5
-a = np.random.randn(rows,cols)
-df = pd.DataFrame(a)
+X = np.random.randn(rows,cols)
+df = pd.DataFrame(X)
+# Pandas does the centering for us
+df = df -df.mean()
 display(df)
-print(df.mean())
-print(df.std())
-display(df**2)
 
+# we center it ourselves
+X_centered = X - X.mean(axis=0)
+print(X_centered-df)
+#Now we do an SVD
+U, s, V = np.linalg.svd(X_centered)
+c1 = V.T[:, 0]
+c2 = V.T[:, 1]
+W2 = V.T[:, :2]
+X2D = X_centered.dot(W2)
+print(X2D)
+#thereafter we do a PCA with Scikit-learn
+from sklearn.decomposition import PCA
+pca = PCA(n_components = 2)
+X2D = pca.fit_transform(X)
+print(X2D)
+
+print(pca.components_.T[:, 0])