noreset_detector

No-reset online change-point detector.

Runs the same loop as OnlineResetDetector but does not call OnlineAlgorithm.reset() when a changepoint is declared (the detection statistic is not restarted), while still applying skip periods and run-length rules from the solver.

class pysatl_cpd.benchmark.online.noreset.detector.noreset_detector.NoResetOnlineDetector(algorithm, *, collect_states=True, data_transformer=None, bisegment_cut=BisegmentCut(left_trim=0, right_trim=0))[source]

Bases: OnlineDetector, Generic

Online detector that does not reset the algorithm after a changepoint.

For the reset loop, see OnlineResetDetector; the only difference is that the post-detection OnlineAlgorithm.reset() call is omitted.

Parameters:

algorithm (OnlineAlgorithm[TypeVar(DataT), TypeVar(ConfigurationT, bound= OnlineAlgorithmConfiguration), TypeVar(StateT, bound= OnlineAlgorithmState)]) – Algorithm instance that drives the per-step detection logic.
collect_states (bool) – Whether to retain algorithm states at every step (default True).
data_transformer (Optional[IDataTransformer[Any, Any]]) – Optional transformer applied to incoming data before processing.
bisegment_cut (BisegmentCut) – Optional provider crop margins (left/right in sample points).

__init__(algorithm, *, collect_states=True, data_transformer=None, bisegment_cut=BisegmentCut(left_trim=0, right_trim=0))[source]

Parameters:

algorithm (OnlineAlgorithm[DataT, ConfigurationT, StateT])
collect_states (bool)
data_transformer (IDataTransformer[Any, Any] | None)
bisegment_cut (BisegmentCut)

Return type:

None

property description: ChangePointDetectorDescription: Description identifying this detector as a no-reset online detector.

property left_trim: int: Left trim in sample points used by detector-level crop.

property right_trim: int: Right trim in sample points used by detector-level crop.

clone()[source]

Create an independent copy of this detector.

Reconstructs the detector by calling recreate() on the underlying algorithm, ensuring parallel workers each have an isolated instance.

Returns:: A new detector instance with identical configuration.
Return type:: NoResetOnlineDetector[TypeVar(DataT), TypeVar(ConfigurationT, bound= OnlineAlgorithmConfiguration), TypeVar(StateT, bound= OnlineAlgorithmState)]