metrics

Metrics calculation tools (`rectools.metrics`).

Tools for fast and convenient calculation of different recommendation metrics.

Metrics

metrics.Precision metrics.Recall metrics.MAP metrics.NDCG metrics.MRR metrics.HitRate metrics.PartialAUC metrics.PAP metrics.F1Beta metrics.Accuracy metrics.MCC metrics.MeanInvUserFreq metrics.IntraListDiversity metrics.AvgRecPopularity metrics.Serendipity metrics.Intersection metrics.SufficientReco metrics.UnrepeatedReco metrics.CoveredUsers

Tools

metrics.calc_metrics - calculate a set of metrics efficiently metrics.PairwiseDistanceCalculator metrics.PairwiseHammingDistanceCalculator metrics.SparsePairwiseHammingDistanceCalculator

Modules

`rectools.metrics.auc`	ROC AUC based ranking recommendations metrics.
`rectools.metrics.base`	Base metric module.
`rectools.metrics.classification`	Classification recommendations metrics.
`rectools.metrics.distances`	Distance metrics.
`rectools.metrics.diversity`	Diversity metrics.
`rectools.metrics.dq`	Recommendations data quality metrics.
`rectools.metrics.intersection`
`rectools.metrics.novelty`	Novelty metrics.
`rectools.metrics.popularity`	Popularity metrics.
`rectools.metrics.ranking`	Ranking recommendations metrics.
`rectools.metrics.scoring`	Metrics calculation module.
`rectools.metrics.serendipity`	Serendipity is designed to find balance between novelty and relevance.