Scikit, No Tears

Preface

This book is to teach students how to use scikit-learn. To follow along and execute the code samples, you will need Docker installed. The Docker container is located on Docker Hub. After you have installed Docker, you may run the container as follows.

docker run -it \
 -p 8888:8888 \
 oneoffcoder/book-scikit-intro

Note that this Docker container has Jupyter Lab running on port 8888. You may access Jupyter Lab at http://localhost:8888 when the Docker container is running.

The diagram below summarizes the scikit-learn workflow that the book follows: prepare data, fit models, validate them, and then branch out into visualization and supporting tools.

The chapters are not perfectly linear, but the diagram captures the main habit the book is trying to teach: model building is only one stage in a larger machine learning workflow.

Contents

• 1. Generating Data
  • 1.1. Generating data for regression
  • 1.2. Generating data for classification
  • 1.3. Generating data for multilabel classification
  • 1.4. Generating data for clustering
  • 1.5. Available datasets
• 2. Visualizing data
  • 2.1. Loading data
  • 2.2. Univariate visualization
  • 2.3. Bivariate visualization
  • 2.4. Categorical plots
  • 2.5. Multivariate visualization
  • 2.6. Visualizing missing data
• 3. Regression
  • 3.1. Generate data
  • 3.2. Types of regression
  • 3.3. Performance
• 4. Logistic Regression
  • 4.1. Generate data
  • 4.2. Types of logistic regression
  • 4.3. Performance
• 5. Poisson Regression
  • 5.1. Generate data
  • 5.2. Visualize y
  • 5.3. Poisson regression
  • 5.4. Histogram gradient boosting regression
• 6. Classification
  • 6.1. Generate data
  • 6.2. Types of classifiers
  • 6.3. Performance
• 7. Clustering
  • 7.1. Generate data
  • 7.2. Types of clustering
  • 7.3. Performance
• 8. Nearest Neighbors
  • 8.1. Ball tree
  • 8.2. KD tree
  • 8.3. Nearest centroid
  • 8.4. Nearest neighbor
• 9. Dimensionality Reduction
  • 9.1. Generate data
  • 9.2. Types of dimensionality reductions
  • 9.3. Performance
• 10. Pairwise Metrics
  • 10.1. Data
  • 10.2. Plotting function
  • 10.3. Real-valued vector space
  • 10.4. Integer-valued vector space
  • 10.5. Boolean-valued vector space
• 11. Data Preprocessing
  • 11.1. Binarizer
  • 11.2. K-bins discretizer
  • 11.3. Label binarizer
  • 11.4. Label encoder
  • 11.5. Multi-label binarizer
  • 11.6. Maximum absolute scaler
  • 11.7. Min-max scaler
  • 11.8. Normalizer
  • 11.9. One-hot encoder
  • 11.10. Ordinal encoder
  • 11.11. Polynomial features
  • 11.12. Power transformer
  • 11.13. Robust scaler
  • 11.14. Standard scaler
  • 11.15. Column transformation
• 12. Imputation
  • 12.1. Simple imputer
  • 12.2. Iterative imputer
  • 12.3. Missing indicator
• 13. Pipelines
  • 13.1. Regression pipeline
  • 13.2. Classification pipeline
• 14. Validation
  • 14.1. Splitting data
  • 14.2. K-fold cross validation example
• 15. Hyperparameter Tuning
  • 15.1. Create data
  • 15.2. Tuning Logistic Regression
  • 15.3. Tuning Random Forest
  • 15.4. Tuning with a pipeline
  • 15.5. Validation with tuning
  • 15.6. tune-sklearn
  • 15.7. Pipelines, column transformers, grid search
• 16. Feature Extraction
  • 16.1. Text Features
• 17. Feature Selection
  • 17.1. Univariate
  • 17.2. Models
  • 17.3. Sequential
  • 17.4. Recursive feature elimination
  • 17.5. Feature engineering with selection
• 18. Customized Estimators
  • 18.1. Basic estimator
  • 18.2. Basic regressor
  • 18.3. Basic classifier
  • 18.4. Basic transformer
  • 18.5. Custom estimator with pipeline

Plotting

• 1. Line Plot
• 2. Bar Plot
• 3. Heat Map
• 4. Distribution Plot
• 5. Box Plot
• 6. Area Plot
• 7. Pie Chart
• 8. Scatter Plot
• 9. Parallel Coordinates
• 10. Graph Plots
• 11. Mosiac Plots
• 12. Animations

Supporting APIs

• 1. Pandas
• 2. Dask
• 3. Widgets
• 4. Patsy
• 5. statsmodels
• 6. SHAP
• 7. imbalanced-learn
• 8. NLTK
• 9. gensim
• 10. PyOD, Anamoly Detection
• 11. combo
• 12. pyGAM
• 13. rpy2
• 14. MLflow
• 15. optuna
• 16. XGBoost
• 17. LightGBM

Misc

• 1. Tips

About

One-Off Coder is an educational, service and product company. Please visit us online to discover how we may help you achieve life-long success in your personal coding career or with your company’s business goals and objectives.

• Website

• Facebook

• Twitter

• Instagram

• YouTube

• LinkedIn

Copyright

This work is licensed under a Creative Commons Attribution 4.0 International License by One-Off Coder.



The full source code is available on GitHub.

Cite this book as follows.:

@misc{oneoffcoder_scikit_intro_2019,
title={Scikit, No Tears},
url={https://learn-scikit.oneoffcoder.com},
author={One-Off Coder},
year={2019},
month={Oct}}

Author

Jee Vang, Ph.D.

• Patreon