crosier

Crosier CUSUM change-point function.

This module provides ChangepointFuncCrosierCUSUM, which applies a norm-based shrinkage update to the accumulated statistic.

class pysatl_cpd.algorithms.online.cusum.component.cpf.crosier.ChangepointFuncCrosierCUSUM(dim, delta=0.0)

Bases: ICusumChangepointFunc[UnivariateNumericArray]

Crosier-style CUSUM change-point statistic for vector observations.

Parameters:

• dim (int) – Observation dimensionality.

• delta (float) – Shrinkage/sensitivity parameter controlling statistic contraction. Default is 0.0.

__init__(dim, delta=0.0)

Parameters:

• dim (int)

• delta (float)

Return type:

None

update(observation)

Update Crosier CUSUM statistic with a new observation.

Applies norm-based shrinkage to the accumulated statistic.

Parameters:

observation (ndarray[tuple[int], dtype[double]]) – New monitoring-space observation vector.

Return type:

None

property value: float

Euclidean norm of the internal vector.

Return type:

float

Type:

Current Crosier CUSUM statistic

reset()

Reset internal accumulated statistic vector.

Return type:

None