Performance Summary

This is a summary view of model performance on the ML32M data set.

Top-N Recommendation Accuracy

Rating Prediction Accuracy

Resource Consumption

Time

Memory

Power

Leaderboard Table

This provides quick numeric access to the model results by mean metric. Note, however, that simple means can be misleading!

Top-N Recommendation

		RBP	NDCG	RecipRank
model	variant
flexmf-logistic	optuna-best	0.181	0.139	0.348
flexmf-bpr	optuna-best	0.172	0.132	0.334
als-implicit	optuna-best	0.170	0.138	0.327
flexmf-warp	optuna-best	0.168	0.130	0.329
iknn-implicit	optuna-best	0.151	0.115	0.298
iknn-implicit	default	0.142	0.113	0.287
als-implicit	default	0.132	0.121	0.261
popular	default	0.119	0.087	0.265
flexmf-warp	default	0.106	0.084	0.225
flexmf-explicit	optuna-best	0.092	0.068	0.220
flexmf-logistic	default	0.083	0.065	0.201
bias	optuna-best	0.074	0.052	0.209
flexmf-bpr	default	0.040	0.044	0.109
als-biased	optuna-best	0.037	0.034	0.110
als-biased	default	0.036	0.032	0.110
iknn-explicit	optuna-best	0.007	0.010	0.024
iknn-explicit	default	0.005	0.005	0.018
flexmf-explicit	default	0.003	0.003	0.012
bias	default	0.000	0.000	0.001

Rating Prediction

		RMSE	MAE
model	variant
als-biased	optuna-best	0.798	0.648
als-biased	default	0.805	0.655
iknn-explicit	optuna-best	0.816	0.656
iknn-explicit	default	0.816	0.656
flexmf-explicit	optuna-best	0.822	0.669
bias	optuna-best	0.829	0.680
flexmf-explicit	default	0.829	0.672
bias	default	0.850	0.691