Welcome to the Fastchange documentation!¶
Fastchange is a change point detection library that is intended to be extendable, interpretable, and above all else, fast! We acheive this by adhering to the Scikit-learn API and using Numba, a just-in-time compiler that allows developers to write pure Python with NumPy that is then compiled to fast C code.
What's in Fastchange?¶
Fastchange currently implements some the most popular offline change point detection methods, including:
Segmentation methods
-
Single, exact change point:
fastchange.seg.AmocSeg
-
Multiple, approximate change points:
fastchange.seg.BinSeg
-
Multple, exact change points:
fastchange.seg.PeltSeg
Cost functions
-
Normal mean and/or variance change:
fastchange.costs.normal
-
Poisson mean+variance change:
fastchange.costs.poisson.PoissonMeanVarCost
-
Poisson mean+variance change:
fastchange.costs.poisson.PoissonMeanVarCost
-
Poisson mean+variance change:
fastchange.costs.poisson.PoissonMeanVarCost
-
Nonparametric cost:
fastchange.costs.emp.EmpiricalCost
Penalty functions
-
MBIC:
fastchange.penalties.mbic_penalty
-
BIC:
fastchange.penalties.bic_penalty
-
AIC:
fastchange.penalties.aic_penalty
-
Hannan-Quinn:
fastchange.penalties.hq_penalty