Quantile#

class oats.models.rule_based.quantile.QuantileModel(window: int = 10, quantile: float = 0.98)[source]#

Bases: Model

Quantile Model

Rule-based model where we label timestep as anomalous if current observation is larger than past window onbservations at quantile.

Parameters:

  • window (int, optional) – rolling window size to feed into the predictor. Defaults to 10.

  • quantile (float, optional) – threshold level for anomaly labelling. Defaults to 0.98.

fit(*args, **kwargs)[source]#
get_scores(data)[source]#