ML1M Performance Summary
This is a summary view of model performance on the ML1M 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 | |||
uknn-implicit | random-best | 0.108 | 0.285 | 0.356 |
uknn-implicit | hyperopt-best | 0.108 | 0.285 | 0.356 |
als-implicit | hyperopt-best | 0.108 | 0.298 | 0.340 |
als-implicit | random-best | 0.105 | 0.293 | 0.328 |
uknn-implicit | default | 0.104 | 0.280 | 0.333 |
flexmf-logistic | hyperopt-best | 0.101 | 0.281 | 0.345 |
flexmf-bpr | hyperopt-best | 0.101 | 0.281 | 0.342 |
flexmf-bpr | random-best | 0.093 | 0.261 | 0.320 |
flexmf-logistic | random-best | 0.093 | 0.264 | 0.316 |
implicit-bpr | hyperopt-best | 0.089 | 0.247 | 0.307 |
flexmf-warp | random-best | 0.086 | 0.246 | 0.302 |
implicit-bpr | random-best | 0.082 | 0.230 | 0.286 |
implicit-bpr | default | 0.081 | 0.231 | 0.281 |
flexmf-warp | hyperopt-best | 0.074 | 0.215 | 0.265 |
iknn-implicit | random-best | 0.073 | 0.198 | 0.265 |
iknn-implicit | hyperopt-best | 0.073 | 0.198 | 0.265 |
als-implicit | default | 0.070 | 0.231 | 0.213 |
iknn-implicit | default | 0.068 | 0.188 | 0.250 |
flexmf-bpr | default | 0.043 | 0.145 | 0.159 |
flexmf-logistic | default | 0.043 | 0.139 | 0.160 |
popular | default | 0.040 | 0.129 | 0.149 |
flexmf-warp | default | 0.033 | 0.117 | 0.123 |
flexmf-explicit | hyperopt-best | 0.027 | 0.087 | 0.108 |
flexmf-explicit | random-best | 0.014 | 0.054 | 0.055 |
als-biased | default | 0.014 | 0.054 | 0.054 |
bias | hyperopt-best | 0.010 | 0.046 | 0.036 |
uknn-explicit | random-best | 0.010 | 0.048 | 0.042 |
bias | random-best | 0.010 | 0.045 | 0.035 |
iknn-explicit | hyperopt-best | 0.007 | 0.040 | 0.027 |
iknn-explicit | random-best | 0.007 | 0.040 | 0.027 |
flexmf-explicit | default | 0.006 | 0.032 | 0.027 |
iknn-explicit | default | 0.006 | 0.040 | 0.026 |
uknn-explicit | hyperopt-best | 0.006 | 0.040 | 0.027 |
uknn-explicit | default | 0.003 | 0.030 | 0.017 |
als-biased | random-best | 0.003 | 0.032 | 0.017 |
als-biased | hyperopt-best | 0.002 | 0.032 | 0.015 |
bias | default | 0.001 | 0.029 | 0.011 |
Rating Prediction
RMSE | MAE | ||
---|---|---|---|
model | variant | ||
iknn-explicit | hyperopt-best | 0.823 | 0.687 |
iknn-explicit | random-best | 0.823 | 0.687 |
iknn-explicit | default | 0.824 | 0.688 |
als-biased | hyperopt-best | 0.826 | 0.690 |
als-biased | default | 0.833 | 0.699 |
als-biased | random-best | 0.833 | 0.696 |
flexmf-explicit | hyperopt-best | 0.836 | 0.706 |
flexmf-explicit | random-best | 0.837 | 0.702 |
uknn-explicit | hyperopt-best | 0.859 | 0.717 |
uknn-explicit | random-best | 0.859 | 0.717 |
flexmf-explicit | default | 0.863 | 0.719 |
uknn-explicit | default | 0.864 | 0.721 |
bias | hyperopt-best | 0.871 | 0.736 |
bias | random-best | 0.871 | 0.735 |
bias | default | 0.873 | 0.736 |