variance

Variance monitoring schema for generalized CUSUM.

This module provides VarianceMonitoringSchema, which converts consecutive-observation differences into approximately standardized variance monitoring statistics.

class pysatl_cpd.algorithms.online.cusum.component.monitoring.variance.VarianceMonitoringSchema

Bases: IMonitoringSchema[UnivariateNumericArray, EstimatesGaussianMLE, UnivariateNumericArray]

Variance-change monitoring transform for univariate observations.

__init__()

Return type:

None

evaluate(observation, estimates)

Compute variance monitoring residual for one observation.

Uses consecutive-observation differences, standardised by the estimated variance, and applies a normalising transform.

Parameters:

• observation (ndarray[tuple[int], dtype[double]]) – New observation (must be dim=1).

• estimates (EstimatesGaussianMLE) – Estimates dict containing "mean" and "cov".

Returns:

Standardised variance-change residual.

Return type:

ndarray[tuple[int], dtype[double]]

Raises:

ValueError – If observation is not dim=1.

reset()

Clear the stored previous observation.

Return type:

None