Metrics

Performance metrics useful in chemoinformatics.

---

General metrics:

auroc_score	Area Under Receiver Operating Characteristic curve (AUROC / ROC AUC).
spearman_correlation	Spearman correlation.

Multioutput metrics:

multioutput_accuracy_score	Accuracy score for multioutput problems.
multioutput_auprc_score	Area Under Precision-Recall Curve (AUPRC / AUC PRC / average precision) score for multioutput problems.
multioutput_auroc_score	Area Under Receiver Operating Characteristic curve (AUROC / ROC AUC) score for multioutput problems.
multioutput_balanced_accuracy_score	Balanced accuracy (average recall) score for multioutput problems.
multioutput_cohen_kappa_score	Cohen's kappa score for multioutput problems.
multioutput_f1_score	F1 score for multioutput problems.
multioutput_matthews_corr_coef	Matthews Correlation Coefficient (MCC) for multioutput problems.
multioutput_mean_absolute_error	Mean absolute error (MAE) for multioutput problems.
multioutput_mean_squared_error	Mean squared error (MSE) for multioutput problems.
multioutput_precision_score	Precision score for multioutput problems.
multioutput_recall_score	Recall score for multioutput problems.
multioutput_root_mean_squared_error	Root mean squared error (RMSE) for multioutput problems.

Virtual screening metrics:

bedroc_score	Boltzmann-enhanced discrimination of ROC (BEDROC).
enrichment_factor	Enrichment factor (EF).
rie_score	Robust Initial Enhancement (RIE).

Utility functions:

extract_pos_proba

Extract positive class probabilities (y-score).