Welcome

This repository contains the notebooks for the "Machine learning for data streams with scikit-multiflow" tutorial for the IJCAI-PRICAI 2020 conference.

Announcement

In November 2020, scikit-muiltiflow and creme, the two most popular open-source libraries for stream learning merged into a new library named river. Accordingly, the notebooks are based on river, the title of the tutorial was not changed as the conference's program was already announced.

Content

• demo-river.ipynb - Provides an overview of river, how to use it and extend/create functionality.
• demo-visualization-hv.ipynb - An advanced example for visualizing model performance in real time with holoviews.

Additional resources:

• Tutorial webpage
• River webpage: https://riverml.xyz
  • Documentation
  • Examples
• GitHub repository: https://github.com/online-ml/river/
  • Follow development
  • Interact with the development team and the project's community (Discussions)
  • Report issues and bugs
• River: machine learning for streaming data in Python (arXiv)