deepcharles/ruptures

Health v2 · maintenance only

Adoption not part of the score

2096 stars · 201 forks observed · 2026-10-12

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

ruptures is a Python library for offline change point detection in non-stationary signals. It provides exact and approximate detection algorithms across parametric and non-parametric models with a consistent, modular, well-documented interface.

Use cases

• detect change points in a time series
• segment a non-stationary signal into regimes
• find structural breaks in sensor data
• split a signal where its statistical properties change
• analyze piecewise stationary signals in Python
• detect regime shifts in financial or physiological time series

When to choose

• you need offline (batch) change point detection in Python
• you want a choice of many detection algorithms and cost models with a unified API
• you need a well-documented, citable scientific library for signal segmentation

When to avoid

• you need online/streaming change point detection in real time
• you work outside Python or need deep-learning-based segmentation
• you only need simple anomaly or outlier detection rather than regime changes

Facets

library · maturity active

machine-learning data-science math data-science analytics python change-point-detection signal-processing segmentation time-series scientific-computing algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="deepcharles/ruptures")