MODULE REFERENCE

detectors

anomx.detection.base.detectors

Detector taxonomy.

Detectors convert continuous scores into decisions or anomaly events:

` Detector ├── StaticThresholdDetector fixed thresholds ├── AdaptiveThresholdDetector data-driven thresholds (quantiles, baselines) ├── StatisticalDetector hypothesis tests, control charts ├── ChangePointDetector distribution shift detection └── EventAggregationDetector grouping decisions into interval events `

AdaptiveThresholdDetector

AdaptiveThresholdDetector

Bases: Detector, ABC

Decide with a threshold derived from the observed score distribution.

Members

detector_familyattribute
detector_family = 'adaptive_threshold'

ChangePointDetector

ChangePointDetector

Bases: Detector, ABC

Decide by detecting distribution shifts in the score sequence.

Members

detector_familyattribute
detector_family = 'change_point'

Detector

Detector

Bases: BaseComponent, ABC

Convert continuous scores into decisions or anomaly events.

Members

component_typeattribute
component_type = 'detector'
detector_familyattribute
detector_family: ClassVar[str] = ''
capabilitiesattribute
capabilities: ClassVar[ComponentCapabilities] = ComponentCapabilities(requires_training=False)
detectmethod
detect(predictions: Any) -> Any

Return predictions with anomaly labels.

Signature parameters
ParameterTypeDefault
predictionsAnyrequired

Returns Any

EventAggregationDetector

EventAggregationDetector

Bases: Detector, ABC

Aggregate raw decisions into interval events with severity.

Members

detector_familyattribute
detector_family = 'event_aggregation'

StaticThresholdDetector

StaticThresholdDetector

Bases: Detector, ABC

Decide with a fixed, configured threshold.

Members

detector_familyattribute
detector_family = 'static_threshold'

StatisticalDetector

StatisticalDetector

Bases: Detector, ABC

Decide with statistical tests or control-chart rules.

Members

detector_familyattribute