module change_detection.tornado.rddm

Reactive Drift Detection Method.

The source code was adopted from tornado, please cite:

The Tornado Framework By Ali Pesaranghader University of Ottawa, Ontario, Canada E-mail: apesaran -at- uottawa -dot- ca / alipsgh -at- gmail -dot- com

Original Paper: Barros, Roberto, et al. "RDDM: Reactive drift detection method." Published in: Expert Systems with Applications. Elsevier, 2017. URL: https://www.sciencedirect.com/science/article/pii/S0957417417305614

Copyright (C) 2022 Johannes Haug.

class RDDM

RDDM change detector.

method RDDM.__init__

__init__(
    min_instance: int = 129,
    warning_level: float = 1.773,
    drift_level: float = 2.258,
    max_size_concept: int = 40000,
    min_size_stable_concept: int = 7000,
    warn_limit: int = 1400,
    reset_after_drift: bool = False
)

Inits the change detector.

Args:

  • min_instance: Todo (left unspecified by the Tornado library).
  • warning_level: Todo (left unspecified by the Tornado library).
  • drift_level: Todo (left unspecified by the Tornado library).
  • max_size_concept: Todo (left unspecified by the Tornado library).
  • min_size_stable_concept: Todo (left unspecified by the Tornado library).
  • warn_limit: Todo (left unspecified by the Tornado library).
  • reset_after_drift: A boolean indicating if the change detector will be reset after a drift was detected.

method RDDM.detect_change

detect_change() → bool

Detects global concept drift.

Returns:

  • bool: True, if a concept drift was detected, False otherwise.

method RDDM.detect_partial_change

detect_partial_change() → Tuple[bool, list]

Detects partial concept drift.

Notes:

RDDM does not detect partial change.

method RDDM.detect_warning_zone

detect_warning_zone() → bool

Detects a warning zone.

Returns:

  • bool: True, if the change detector has detected a warning zone, False otherwise.

method RDDM.partial_fit

partial_fit(pr_scores: List[bool])

Updates the change detector.

Args:

  • pr_scores: A boolean vector indicating correct predictions. 'True' values indicate that the prediction by the online learner was correct, otherwise the vector contains 'False'.

method RDDM.reset

reset()

Resets the change detector.

This file was automatically generated via lazydocs.