ClassificationMetricResults

class ClassificationMetricResults(data)

Bases: MetricResults

Results from computing metrics.

Initialize the result wrapper.

Attributes Summary

metrics

Methods Summary

from_scores(y_true, y_score)	Return an instance of these metrics from a given set of true and scores.
get_metric(name)	Get the given metric from the results.

Attributes Documentation

Parameters:

data (Mapping) –

metrics: ClassVar[Mapping[str, Type[Metric]]] = {'accuracy_score': <class 'pykeen.metrics.classification.AccuracyScore'>, 'average_precision_score': <class 'pykeen.metrics.classification.AveragePrecisionScore'>, 'balanced_accuracy_score': <class 'pykeen.metrics.classification.BalancedAccuracyScore'>, 'diagnostic_odds_ratio': <class 'pykeen.metrics.classification.DiagnosticOddsRatio'>, 'f1_score': <class 'pykeen.metrics.classification.F1Score'>, 'false_discovery_rate': <class 'pykeen.metrics.classification.FalseDiscoveryRate'>, 'false_negative_rate': <class 'pykeen.metrics.classification.FalseNegativeRate'>, 'false_omission_rate': <class 'pykeen.metrics.classification.FalseOmissionRate'>, 'false_positive_rate': <class 'pykeen.metrics.classification.FalsePositiveRate'>, 'fowlkes_mallows_index': <class 'pykeen.metrics.classification.FowlkesMallowsIndex'>, 'informedness': <class 'pykeen.metrics.classification.Informedness'>, 'markedness': <class 'pykeen.metrics.classification.Markedness'>, 'matthews_correlation_coefficient': <class 'pykeen.metrics.classification.MatthewsCorrelationCoefficient'>, 'negative_likelihood_ratio': <class 'pykeen.metrics.classification.NegativeLikelihoodRatio'>, 'negative_predictive_value': <class 'pykeen.metrics.classification.NegativePredictiveValue'>, 'positive_likelihood_ratio': <class 'pykeen.metrics.classification.PositiveLikelihoodRatio'>, 'positive_predictive_value': <class 'pykeen.metrics.classification.PositivePredictiveValue'>, 'prevalence_threshold': <class 'pykeen.metrics.classification.PrevalenceThreshold'>, 'roc_auc_score': <class 'pykeen.metrics.classification.RocAucScore'>, 'threat_score': <class 'pykeen.metrics.classification.ThreatScore'>, 'true_negative_rate': <class 'pykeen.metrics.classification.TrueNegativeRate'>, 'true_positive_rate': <class 'pykeen.metrics.classification.TruePositiveRate'>}

Methods Documentation

classmethod from_scores(y_true, y_score)

Return an instance of these metrics from a given set of true and scores.

Parameters:

• y_true (ndarray) –

• y_score (ndarray) –

get_metric(name)

Get the given metric from the results.

Parameters:

name (str) – The name of the metric

Return type:

float

Returns:

The value for the metric