Aeon is an open source toolkit designed for learning from time series, offering access to the latest machine learning algorithms and a range of classical techniques. The package aims to provide a single framework for algorithm comparison, employing efficient implementations using numba and interfaces with other time series packages. Aeon’s latest release is v0.3.0, and a suspension of the deprecation policy allows for faster development and maintenance during the forking transition period. A Python version of 3.8 or greater is required for installation, which is made simple via pip. The package also provides a comprehensive getting started guide, alongside examples of forecasting and classification. Questions may be directed to Aeon’s GitHub Issue Tracker or Slack platforms.
https://github.com/aeon-toolkit/aeon