noreset_trace

No-reset detection trace wrapper.

class pysatl_cpd.benchmark.online.noreset.detector.noreset_trace.NoResetDetectionTrace(*, detected_change_points=<factory>, detector_description, threshold=None, processing_time, detection_function, forced_change_points=<factory>, signal_change_points=<factory>, skip_periods=<factory>, learning_periods=<factory>, algorithm_states)

Bases: OnlineDetectionTrace, Generic

Trace derived from an infinite-threshold run for no-reset benchmarking.

Parameters:

• detected_change_points (Sequence[int])

• detector_description (ChangePointDetectorDescription)

• threshold (Number | None)

• processing_time (UnivariateNumericArray)

• detection_function (UnivariateNumericArray)

• forced_change_points (list[int])

• signal_change_points (list[int])

• skip_periods (list[tuple[int, int]])

• learning_periods (list[tuple[int, int]])

• algorithm_states (list[StateT | None])

classmethod from_inf_trace(source_trace, detected_change_points, threshold)

Build a no-reset trace from an infinite-threshold source trace.

Copies detection function values and algorithm states from the source while injecting the detected change points and threshold determined by a policy. Processing time is left empty.

Parameters:

• source_trace (OnlineDetectionTrace[TypeVar(StateT, bound= OnlineAlgorithmState)]) – Trace produced by an infinite-threshold run.

• detected_change_points (list[int]) – Change-point indices selected by the policy.

• threshold (float) – Threshold value that was applied during policy evaluation.

Returns:

A trace ready for classification or ARL analysis.

Return type:

NoResetDetectionTrace[TypeVar(StateT, bound= OnlineAlgorithmState)]