This repository contains the relevant code to perform bayesian online changepoint detection at the linac3 source (at least in a very basic fashion).
It includes some notebooks to make clear how to use the algorithm and to produce some preliminary results, and hopefully it is a foundation that can be built upon.
It uses the linac3 module to load csv files and to provide some constants, however this can also be replaced by pandas. The input files are plain csv files as can be downloaded from CALS using Timber.