autointent.metrics.scoring

Scoring metrics for multiclass and multilabel classification tasks.

Attributes

logger

Classes

ScoringMetricFn

Protocol for scoring metrics.

Functions

ignore_oos(func)	Ignore OOS in metrics calculation (decorator).
scoring_log_likelihood(labels, scores[, eps])	Calculate log likelihood score for multiclass and multilabel cases.
scoring_roc_auc(labels, scores)	Calculate ROC AUC score for multiclass and multilabel cases.
scoring_accuracy(labels, scores)	Calculate accuracy for multiclass and multilabel classification.
scoring_f1(labels, scores)	Calculate F1 score for multiclass and multilabel classification.
scoring_precision(labels, scores)	Calculate precision for multiclass and multilabel classification.
scoring_recall(labels, scores)	Calculate recall for multiclass and multilabel classification.
scoring_hit_rate(labels, scores)	Calculate hit rate for multilabel classification.
scoring_neg_coverage(labels, scores)	Calculate negative coverage for multilabel classification.
scoring_neg_ranking_loss(labels, scores)	Calculate negative ranking loss for multilabel classification.
scoring_map(labels, scores)	Calculate mean average precision (MAP) score for multilabel classification.

Module Contents

autointent.metrics.scoring.logger