scikit-learn: machine learning in Python

Easy-to-use and general-purpose machine learning in Python

scikit-learn is a Python module integrating classic machine learning algorithms in the tightly-knit world of scientific Python packages (numpy, scipy, matplotlib).

It aims to provide simple and efficient solutions to learning problems that are accessible to everybody and reusable in various contexts: machine-learning as a versatile tool for science and engineering.

Features:

• Solid: Supervised learning: Support Vector Machines, Generalized Linear Models.
• Work in progress: Unsupervised learning: Clustering, Gaussian mixture models, Manifold learning, ICA, Gaussian Processes, matrix factorization
• Planned: Gaussian graphical models

License:

Open source, commercially usable: BSD license (3 clause)

Note

This document describes scikit-learn 0.9. For other versions and printable format, see Documentation resources.

User Guide

• 1. Installing scikit-learn
  • 1.1. Installing an official release
  • 1.2. Third party distributions of scikit-learn
  • 1.3. Bleeding Edge
  • 1.4. Testing
• 2. Getting started: an introduction to machine learning with scikit-learn
  • 2.1. Machine learning: the problem setting
  • 2.2. Loading an example dataset
  • 2.3. Learning and Predicting
  • 2.4. Model persistence
• 3. Supervised learning
  • 3.1. Generalized Linear Models
  • 3.2. Support Vector Machines
  • 3.3. Stochastic Gradient Descent
  • 3.4. Nearest Neighbors
  • 3.5. Gaussian Processes
  • 3.6. Partial Least Squares
  • 3.7. Naive Bayes
  • 3.8. Multiclass algorithms
  • 3.9. Feature selection
• 4. Unsupervised learning
  • 4.1. Gaussian mixture models
  • 4.2. Manifold learning
  • 4.3. Clustering
  • 4.4. Decomposing signals in components (matrix factorization problems)
  • 4.5. Covariance estimation
  • 4.6. Hidden Markov Models
• 5. Model Selection
  • 5.1. Cross-Validation
  • 5.2. Grid Search
• 6. Dataset loading utilities
  • 6.1. Toy datasets
  • 6.2. Sample images
  • 6.3. Sample generators
  • 6.4. Datasets in svmlight / libsvm format
  • 6.5. The Olivetti faces dataset
  • 6.6. The 20 newsgroups text dataset
  • 6.7. Downloading datasets from the mldata.org repository
  • 6.8. The Labeled Faces in the Wild face recognition dataset
• 7. Preprocessing data
  • 7.1. Standardization or Mean Removal and Variance Scaling
  • 7.2. Normalization
  • 7.3. Binarization
• 8. Feature extraction
  • 8.1. Text feature extraction
  • 8.2. Image feature extraction
• 9. Class Reference
  • 9.1. Support Vector Machines
  • 9.2. Generalized Linear Models
  • 9.3. Naive Bayes
  • 9.4. Nearest Neighbors
  • 9.5. Gaussian Mixture Models
  • 9.6. Hidden Markov Models
  • 9.7. Clustering
  • 9.8. Metrics
  • 9.9. Covariance Estimators
  • 9.10. Signal Decomposition
  • 9.11. Linear Discriminant Analysis
  • 9.12. Partial Least Squares
  • 9.13. Cross Validation
  • 9.14. Grid Search
  • 9.15. Feature Selection
  • 9.16. Feature Extraction
  • 9.17. Preprocessing and normalization
  • 9.18. Manifold learning
  • 9.19. Datasets
  • 9.20. Pipeline
  • 9.21. Utilities

Example Gallery

• Examples
  • General examples
  • Examples based on real world datasets
  • Clustering
  • Covariance estimation
  • Decomposition
  • Gaussian Process for Machine Learning
  • Generalized Linear Models
  • Manifold learning
  • Gaussian Mixture Models
  • Nearest Neighbors
  • Support Vector Machines

Development

• Contributing
  • Submitting a bug report
  • Retrieving the latest code
  • Contributing code
  • Other ways to contribute
  • Coding guidelines
  • APIs of scikit learn objects
• How to optimize for speed
  • Python, Cython or C/C++?
  • Profiling Python code
  • Performance tips for the Cython developer
  • Profiling compiled extensions
  • Multi-core parallelism using joblib.Parallel
  • A sample algorithmic trick: warm restarts for cross validation
• About us
  • History
  • People
  • Citing the scikit-learn
  • Funding