aeon-toolkit/aeon

Health v2 · maintenance only

Adoption not part of the score

1438 stars · 316 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

aeon is a scikit-learn compatible Python toolkit for machine learning on time series, covering classification, regression, clustering, forecasting, anomaly detection, segmentation, distances, similarity search, and benchmarking. Many algorithms are implemented and maintained by the researchers who developed them, including state-of-the-art and deep learning methods.

Use cases

• classify time series with machine learning
• forecast future values of a time series
• detect anomalies in sensor data over time
• cluster similar time series
• segment a time series into regimes
• benchmark time series algorithms
• run deep learning models on time series data
• compute distances between time series

When to choose

• you need a scikit-learn compatible API for time series tasks
• you want access to state-of-the-art time series classification, clustering, or forecasting algorithms
• you need benchmarking tools for time series methods
• you want researcher-maintained implementations of published algorithms

When to avoid

• you only need simple classical forecasting like ARIMA or Prophet
• you work with tabular non-time-series data
• you need a non-Python language

Facets

library · maturity active

machine-learning deep-learning data-science benchmarking etl machine-learning data-science time-series artificial-intelligence python cross-platform time-series-classification time-series-clustering forecasting anomaly-detection segmentation similarity-search scikit-learn-compatible transformations

Member repositories

For agents

markdown · JSON · MCP: product_card(name="aeon-toolkit/aeon")