tsseg.algorithms.pelt package

PELT — Pruned Exact Linear Time change point detection.

Description

PELT solves the penalised change point optimisation problem exactly by pruning the search space with a dynamic-programming rule. Under mild conditions on the change point distribution, the average complexity is \(O(C\,K\,n)\) — linear in the number of samples — where K is the number of change points and C the cost function complexity. In practice, "l2" cost models are significantly faster than linear or autoregressive ones.

Key tuning levers:

penalty — higher values produce fewer change points.
min_size — minimum distance between change points.
jump — grid step for admissible positions.

Type: change point detection
Supervision: fully unsupervised
Scope: univariate and multivariate
Complexity: \(O(C\,K\,n)\) average

Parameters

Name	Type	Default	Description
`model`	str	`"l2"`	Ruptures cost model.
`min_size`	int	`2`	Minimum segment length.
`jump`	int	`5`	Grid step for candidate positions.
`penalty`	float	`10.0`	Penalty threshold for the PELT stopping criterion.
`cost_params`	dict / None	`None`	Extra keyword arguments for the cost factory.
`axis`	int	`0`	Time axis.

Usage

from tsseg.algorithms import PeltDetector

detector = PeltDetector(model="l2", penalty=10)
labels = detector.fit_predict(X)

Implementation: Vendored from ruptures v1.1.8. BSD 2-Clause.

Reference: Killick, Fearnhead & Eckley (2012), Optimal detection of changepoints with a linear computational cost, JASA.

tsseg.algorithms.pelt package

Description

Parameters

Usage

Submodules

tsseg.algorithms.pelt.detector module

Module contents