ClaSPSegmentation

ClaSP (Classification Score Profile) Segmentation.

Quickstart

from sktime.detection.clasp import ClaSPSegmentation

estimator = ClaSPSegmentation(period_length=10, n_cps=1, exclusion_radius=0.05)

Parameters(3)

period_lengthint, default = 10

size of window for sliding, based on the period length of the data

n_cpsint, default = 1

the number of change points to search

exclusion_radiusint

Exclusion Radius for change points to be non-trivial matches

Examples

>>> from sktime.detection.clasp import ClaSPSegmentation
>>> from sktime.detection.clasp import find_dominant_window_sizes
>>> from sktime.datasets import load_gun_point_segmentation
>>> X, true_period_size, cps = load_gun_point_segmentation ()
>>> dominant_period_size = find_dominant_window_sizes (X)
>>> clasp = ClaSPSegmentation (dominant_period_size, n_cps = 1)
>>> found_cps = clasp. fit_predict (X)
>>> profiles = clasp. profiles
>>> scores = clasp. scores