Welcome to scikit-multiflow’s website.

scikit-multiflow is inspired by MOA, the most popular open source framework for machine learning for data streams, and MEKA, an open source implementation of methods for multi-label learning. scikit-multiflow is also inspired on scikit-learn, the most popular framework for machine learning in Python. Following the SciKits philosophy, scikit-multiflow is an open source machine learning framework for multi-output/multi-label and stream data.

skmultiflow_plot

scikit-multiflow is implemented in Python given its increasing popularity in the Machine Learning community. It complements scikit-learn, whose primary focus is batch learning, expanding the set of Machine Learning tools on this platform. In its current state, scikit-multiflow contains stream generators, stream classifiers for multi-output/multi-target, change detectors and evaluation methods.

scikit-multiflow is being developed by Télécom ParisTech and École Polytechnique.

Installation

Installation

Install scikit-multiflow on your machine.

Read More

User Guide

User Guide

Read more about scikit-multiflow core elements and how to start using it.

Read More

Documentation

Documentation

scikit-multiflow package documentation.

Read More

Community

Users

A mailing list is available for users to learn, teach and ask questions: scikit-multiflow users

Contributing

As an open source project, we welcome contributions from the community. Please refer to the GitHub Repository for further information.

Citing scikit-multiflow

If you want to cite scikit-multiflow in a publication, please refer to the following paper:

Scikit-Multiflow: A Multi-outputStreaming Framework. Jacob Montiel, Jesse Read, Albert Bifet, Talel Abdessalem; Journal of Machine Learning Research, 19(72):1−5, 2018 | Bibtex