DRecPy.Evaluation.Metrics package

DRecPy.Evaluation.Metrics.ranking module

class DRecPy.Evaluation.Metrics.ranking.AveragePrecision

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Average Precision at k.

class DRecPy.Evaluation.Metrics.ranking.DCG(strong_relevancy=True)

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Discounted Cumulative Gain at k.

Parameters:strong_relevancy – An optional boolean indicating which variant of the DCG is used. If set to True, usually results in smaller values than when it’s set to False. Default: True.

class DRecPy.Evaluation.Metrics.ranking.FScore(beta=1)

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

F-score at k.

Parameters:beta – An optional integer representing the weight of the recall value on the combined score. Beta > 1 favours recall over precision, while beta < 1 favours precision over recall. Default: 1.

class DRecPy.Evaluation.Metrics.ranking.HitRatio

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Hit Ratio at k.

class DRecPy.Evaluation.Metrics.ranking.NDCG(strong_relevancy=True)

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Discounted Cumulative Gain at k.

Parameters:strong_relevancy – An optional boolean indicating which variant of the DCG is used. If set to True, usually results in smaller values than when it’s set to False. Default: True.

class DRecPy.Evaluation.Metrics.ranking.Precision

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Precision at k.

class DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Bases: DRecPy.Evaluation.Metrics.metric_abc.MetricABC

class DRecPy.Evaluation.Metrics.ranking.Recall

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Recall at k.

class DRecPy.Evaluation.Metrics.ranking.ReciprocalRank

Bases: DRecPy.Evaluation.Metrics.ranking.RankingMetricABC

Reciprocal Rank at k.

DRecPy.Evaluation.Metrics.regression module

class DRecPy.Evaluation.Metrics.regression.MAE

Bases: DRecPy.Evaluation.Metrics.regression.PredictiveMetricABC

Mean absolute Error.

class DRecPy.Evaluation.Metrics.regression.MSE

Bases: DRecPy.Evaluation.Metrics.regression.PredictiveMetricABC

Mean Squared Error.

class DRecPy.Evaluation.Metrics.regression.PredictiveMetricABC

Bases: DRecPy.Evaluation.Metrics.metric_abc.MetricABC

class DRecPy.Evaluation.Metrics.regression.RMSE

Bases: DRecPy.Evaluation.Metrics.regression.PredictiveMetricABC

Root Mean Squared Error.