# aeon-toolkit/aeon

A toolkit for time series  machine learning and deep learning

Repository: https://github.com/aeon-toolkit/aeon
Canonical: https://ross.abutalabs.com/products/aeon
Homepage: https://aeon-toolkit.org/
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: data-mining, data-science, machine-learning, scikit-learn, time-series, time-series-analysis, time-series-anomaly-detection, time-series-classification, time-series-clustering, time-series-regression, time-series-segmentation, forecasting, deep-learning, neural-network, aeon, ai, artificial-intelligence
Last push: 2026-08-26T19:16:33+00:00

## Health v2 (maintenance only)
Score: 91/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 78, longevity 96
- inputs: {"age_days": 1352, "days_push": 7, "days_rel": 65, "gap_med": 89.0, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1438, forks 316 (observed 2026-08-28T04:04:43.956206+00:00)

## What it is
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
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, data-science, benchmarking, etl
- domain: machine-learning, data-science, time-series, artificial-intelligence
- platform: python, cross-platform
- tags: time-series-classification, time-series-clustering, forecasting, anomaly-detection, segmentation, similarity-search, scikit-learn-compatible, transformations

## Member repositories
- aeon-toolkit/aeon (main) score 91

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:43.956206+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:36:39.063508+00:00, confidence not recorded.
  - readme: https://github.com/aeon-toolkit/aeon (fetched 2026-08-28T04:04:43.956206+00:00, sha ff1389d8ab71)
  - homepage: https://aeon-toolkit.org/ (fetched 2026-08-29T11:47:15.403607+00:00, sha 3a70d4c02642)
  - registry_pypi: https://pypi.org/pypi/aeon/json (fetched 2026-08-29T11:47:15.412803+00:00, sha 1925d91678a2)
- Data as of 2026-08-30T08:39:29.467469+00:00.