Add Kaggle kernels to documentation (#993)

* Add Kaggle kernels

* Change year in docs

* Update build script

Author: outoftardis

doc/build-doc.sh

@@ -15,9 +15,17 @@
 # limitations under the License.
 #===============================================================================
 
+SAMPLES_DIR=sources/samples
+
+# remove the samples folder if it exists
+if [ -d "$SAMPLES_DIR" ]; then rm -Rf $SAMPLES_DIR; fi
+
+# create a samples folder
+mkdir $SAMPLES_DIR
+
 # copy jupyter notebooks
 cd ..
-cp examples/notebooks/*.ipynb doc/sources/samples
+cp examples/notebooks/*.ipynb doc/$SAMPLES_DIR
 
 # build the documentation
 cd doc

doc/sources/acceleration.rst

@@ -36,3 +36,12 @@ Configurations:
 
 - HW: c5.24xlarge AWS EC2 Instance using an Intel Xeon Platinum 8275CL with 2 sockets and 24 cores per socket
 - SW: scikit-learn version 0.24.2, scikit-learn-intelex version 2021.2.3, Python 3.8
+
+Kaggle Kernels
+**************
+
+Check out `Introduction to scikit-learn-intelex
+<https://www.kaggle.com/code/lordozvlad/introduction-to-scikit-learn-intelex/notebook>`_,
+a Kaggle notebook that summarizes the speedup you can achieve with |intelex|.
+
+The acceleration is measured for a variety of machine learning workflows and Kaggle datasets.

doc/sources/conf.py

@@ -37,13 +37,13 @@
 # -- Project information -----------------------------------------------------
 
 project = 'Intel(R) Extension for Scikit-learn*'
-copyright = '2021, Intel'
+copyright = '2022, Intel'
 author = 'Intel'
 
 # The short X.Y version
 version = '2021'
 # The full version, including alpha/beta/rc tags
-release = '2021.5'
+release = '2021.6'
 
 
 # -- General configuration ---------------------------------------------------

doc/sources/kaggle.rst

@@ -19,6 +19,17 @@
 Kaggle Kernels
 --------------------
 
+Check out Kaggle notebooks created by |intelex| developers.
+
+.. rubric:: Acceleration
+
+`Introduction to scikit-learn-intelex <https://www.kaggle.com/code/lordozvlad/introduction-to-scikit-learn-intelex/notebook>`_
+provides a summary of the speedup you can achieve with |intelex|.
+
+.. rubric:: Machine Learning Workflows
+
+Browse this chapter to find Kaggle kernels that use scikit-learn-intelex for a specific type of a machine learning task.
+
 Kaggle kernels that use scikit-learn and |intelex|:
 
 .. toctree::

doc/sources/kaggle/automl.rst

@@ -26,6 +26,10 @@
 .. |automl_with_intelex_titanic| replace:: AutoML Binary Classification (Gradient Boosting, Random Forest, kNN) using AutoGluon with |intelex|
 .. _automl_with_intelex_titanic: https://www.kaggle.com/lordozvlad/titanic-automl-with-intel-extension-for-sklearn/notebook
 
+.. |automl_with_intelex_tps_jan| replace:: AutoML Binary Classification (Random Forest, SVR, Blending) using PyCaret with |intelex|
+.. _automl_with_intelex_tps_jan: https://www.kaggle.com/code/lordozvlad/tps-jan-fast-pycaret-with-scikit-learn-intelex/notebook
+
+
 Kaggle Kernels that use AutoML and |intelex|
 --------------------------------------------
 
@@ -55,4 +59,8 @@ The following Kaggle kernels show how to patch autoML framewokrs with |intelex|.
    * - |automl_with_intelex_tps_nov|_
 
        **Data:** [TPS Nov 2021] Synthetic spam emails data
-     - Identify spam emails via features extraced from the email
+     - Identify spam emails via features extraced from the email
+   * - |automl_with_intelex_tps_jan|_
+
+       **Data:** [TPS Jan 2022] Fictional Sales data
+     - Predict the corresponding item sales for each date-country-store-item combination

doc/sources/kaggle/classification.rst

@@ -45,7 +45,18 @@ Binary Classification
        - search for optimal parameters using Optuna
        - training and prediction using scikit-learn-intelex
        - performance comparison to scikit-learn
+   * - `Feature Importance in Random Forest for Binary Classification
+       <https://www.kaggle.com/code/lordozvlad/fast-feature-importance-using-scikit-learn-intelex/notebook>`_
+     
+       **Data:** [TPS Nov 2021] Synthetic spam emails data
 
+     - Identify spam emails via features extraced from the email
+     - 
+      
+       - reducing DataFrame memory usage
+       - computing feature importance with ELI5 and the default scikit-learn permutation importance
+       - training using scikit-learn-intelex
+       - performance comparison to scikit-learn
    * - `Random Forest for Binary Classification
        <https://www.kaggle.com/andreyrus/tps-apr-rf-with-intel-extension-for-scikit-learn>`_
 
@@ -80,7 +91,6 @@ Binary Classification
        - training and prediction using scikit-learn-intelex
        - performance comparison to scikit-learn
 
-
 MultiClass Classification
 +++++++++++++++++++++++++
 
@@ -115,7 +125,7 @@ MultiClass Classification
    * - `Stacking Classifer with Logistic Regression, kNN, Random Forest, and Quantile Transformer
        <https://www.kaggle.com/owerbat/tps-jun-fast-stacking-with-scikit-learn-intelex>`_
 
-       **Data:** [TPS Jun 2021] synthetic eCommerce data
+       **Data:** [TPS Jun 2021] Synthetic eCommerce data
      - Predict the category of an eCommerce product
      - 
 
@@ -125,6 +135,36 @@ MultiClass Classification
        - searching for optimal parameters for the stacking classifier
        - training and prediction using scikit-learn-intelex
        - performance comparison to scikit-learn
+   * - `Support Vector Classification (SVC) for MultiClass Classification
+       <https://www.kaggle.com/code/alexeykolobyanin/tps-dec-svc-with-sklearnex-20x-speedup>`_
+       
+       **Data:** [TPS Dec 2021] Synthetic Forest Cover Type data
+     - Predict the forest cover type
+     - 
+       - data preprocessing
+       - training and prediction using scikit-learn-intelex
+       - performance comparison to scikit-learn
+   * - `Feature Importance in Random Forest for MultiClass Classification
+       <https://www.kaggle.com/code/lordozvlad/tps-dec-fast-feature-importance-with-sklearnex>`_
+     
+       **Data:** [TPS Dec 2021] Synthetic Forest Cover Type data
+
+     - Predict the forest cover type
+     - 
+      
+       - reducing DataFrame memory usage
+       - computing feature importance with ELI5
+       - training and prediction using scikit-learn-intelex
+       - performance comparison to scikit-learn
+   * - `k-Nearest Neighbors (kNN) for MultiClass Classification
+       <https://www.kaggle.com/code/alexeykolobyanin/tps-feb-knn-with-sklearnex-13x-speedup>`_
+       
+       **Data:** [TPS Feb 2022] Bacteria DNA
+     - Predict bacteria species based on repeated lossy measurements of DNA snippets
+     - 
+       - data preprocessing
+       - training and prediction using scikit-learn-intelex
+       - performance comparison to scikit-learn
 
 Classification Tasks in Computer Vision
 +++++++++++++++++++++++++++++++++++++++
@@ -190,4 +230,15 @@ Classification Tasks in Natural Language Processing
        - feature extraction using TfidfVectorizer
        - training and prediction using scikit-learn-intelex
        - performance comparison to scikit-learn
-
+   * - `Support Vector Classification (SVC) for Binary Classification with Sparse Data (NLP task)
+       <https://www.kaggle.com/code/alex97andreev/fast-svm-for-sparse-data-from-nlp-problem>`_
+       
+       **Data:** Stack Overflow questions
+     - Predict the binary quality rating for Stack Overflow questions
+     - 
+     
+       - data preprocessing
+       - TF-IDF calculation
+       - search for optimal paramters using Optuna
+       - training and prediction using scikit-learn-intelex
+       - performance comparison to scikit-learn

doc/sources/kaggle/regression.rst

@@ -93,6 +93,17 @@ Using a Single Regressor
        - search for optimal parameters using Optuna
        - training and prediction using scikit-learn-intelex
        - performance comparison to scikit-learn
+   * - `Random Forest Regression with Feature Importance Computation
+       <https://www.kaggle.com/code/lordozvlad/tps-mar-fast-workflow-using-scikit-learn-intelex>`_
+
+       **Data:** [TPS Mar 2022] Spatio-temporal traffic data
+     - Forecast twelve-hours of traffic flow in a major U.S. metropolitan area
+     -
+
+       - feature engineering
+       - computing feature importance with ELI5
+       - training and prediction using scikit-learn-intelex
+       - performance comparison to scikit-learn
    * - `Ridge Regression
        <https://www.kaggle.com/alexeykolobyanin/tps-sep-ridge-with-sklearn-intelex-2x-speedup>`_