NumericalDriftMetrics Classe

Notes

Il s’agit d’une classe expérimentale qui peut changer à tout moment. Pour plus d’informations, consultez https://aka.ms/azuremlexperimental.

Héritage

azure.ai.ml.entities._mixins.RestTranslatableMixin

NumericalDriftMetrics

Constructeur

NumericalDriftMetrics(*, jensen_shannon_distance: float = None, normalized_wasserstein_distance: float = None, population_stability_index: float = None, two_sample_kolmogorov_smirnov_test: float = None, metric: str | None = None, metric_threshold: float | None = None)

Méthodes

defaults
get_name_and_threshold

defaults

defaults() -> NumericalDriftMetrics

get_name_and_threshold

get_name_and_threshold()