classification

Classification recommendations metrics.

Functions

calc_classification_metrics(metrics, merged)	Calculate any classification metrics.
calc_confusions(merged, k)	Calculate some intermediate metrics from prepared data (it's a helper function).
make_confusions(reco, interactions, k)	Calculate some intermediate metrics from raw data (it's a helper function).

Classes

Accuracy(k[, debias_config])	Ratio of correctly recommended items among all items.
ClassificationMetric(k[, debias_config])	Classification metric base class.
F1Beta(k[, debias_config, beta])	Fbeta score for k first recommendations.
HitRate(k[, debias_config])	HitRate calculates the fraction of users for which the correct answer is included in the recommendation list.
MCC(k[, debias_config])	Matthew correlation coefficient calculates correlation between actual and predicted classification.
Precision(k[, debias_config, r_precision])	Ratio of relevant items among top-k recommended items.
Recall(k[, debias_config])	Ratio of relevant recommended items among all items user interacted with after recommendations were made.
SimpleClassificationMetric(k[, debias_config])	Simple classification metric base class.