Have a look at ruptures package for changepoint detection

[MichaelClerx]

@fcooper8472 found this: https://github.com/deepcharles/ruptures

"ruptures: change point detection in Python."

The documentation links to a survey of changepoint detection methods by the same authors (https://arxiv.org/abs/1801.00718) which seems quite comprehensive, and mentions the authors of the R-package in the citations

[MichaelClerx]

Might be changepoint repo: https://github.com/rkillick/changepoint/

@fergus found: https://github.com/deepcharles/ruptures
ruptures: change point detection in Python. Contribute to deepcharles/ruptures development by creating an account on GitHub.
michael 12:26 PM
The ruptures thing looks quite good. References PELT (whatever that is) which is also mentioned in the R version. And there's an arxiv paper https://arxiv.org/abs/1801.00718
arXiv.orgarXiv.org
Selective review of offline change point detection methods
This article presents a selective survey of algorithms for the offline detection of multiple change points in multivariate time series. A general yet structuring methodological strategy is adopted...
fergus 12:26 PM
Cool!
michael 12:29 PM
Arxiv paper is actually a massive survey of changepoint algorithms, which seem to be implemented in the python package. Cites the author of the R package twice
Feeling good about this 😄

[MichaelClerx]

@abhidg is now implementing this as part of #780