Performance Summary
This is a summary view of model performance on the ML10M 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-bpr | random-best | 0.253 | 0.176 | 0.445 |
flexmf-logistic | hyperopt-best | 0.252 | 0.175 | 0.445 |
flexmf-logistic | optuna-best | 0.245 | 0.171 | 0.431 |
flexmf-bpr | optuna-best | 0.244 | 0.169 | 0.430 |
flexmf-bpr | hyperopt-best | 0.243 | 0.170 | 0.435 |
flexmf-logistic | random-best | 0.238 | 0.168 | 0.417 |
flexmf-warp | optuna-best | 0.237 | 0.165 | 0.417 |
uknn-implicit | hyperopt-best | 0.235 | 0.167 | 0.416 |
uknn-implicit | optuna-best | 0.235 | 0.167 | 0.416 |
uknn-implicit | random-best | 0.234 | 0.167 | 0.415 |
flexmf-warp | hyperopt-best | 0.232 | 0.161 | 0.410 |
flexmf-warp | random-best | 0.221 | 0.153 | 0.400 |
uknn-implicit | default | 0.219 | 0.161 | 0.392 |
als-implicit | optuna-best | 0.215 | 0.154 | 0.391 |
als-implicit | hyperopt-best | 0.214 | 0.159 | 0.386 |
als-implicit | random-best | 0.213 | 0.157 | 0.387 |
iknn-implicit | hyperopt-best | 0.200 | 0.144 | 0.365 |
iknn-implicit | random-best | 0.200 | 0.144 | 0.364 |
iknn-implicit | optuna-best | 0.200 | 0.144 | 0.364 |
flexmf-warp | default | 0.200 | 0.140 | 0.376 |
iknn-implicit | default | 0.185 | 0.141 | 0.339 |
flexmf-explicit | optuna-best | 0.183 | 0.121 | 0.379 |
flexmf-bpr | default | 0.183 | 0.126 | 0.334 |
flexmf-logistic | default | 0.180 | 0.124 | 0.337 |
popular | default | 0.175 | 0.119 | 0.337 |
als-implicit | default | 0.136 | 0.123 | 0.256 |
flexmf-explicit | random-best | 0.117 | 0.085 | 0.259 |
bias | random-best | 0.094 | 0.068 | 0.264 |
bias | optuna-best | 0.094 | 0.068 | 0.264 |
bias | hyperopt-best | 0.094 | 0.068 | 0.264 |
als-biased | default | 0.085 | 0.067 | 0.210 |
als-biased | optuna-best | 0.081 | 0.066 | 0.190 |
als-biased | hyperopt-best | 0.064 | 0.058 | 0.153 |
iknn-explicit | default | 0.040 | 0.046 | 0.094 |
iknn-explicit | random-best | 0.038 | 0.046 | 0.087 |
iknn-explicit | hyperopt-best | 0.038 | 0.046 | 0.087 |
iknn-explicit | optuna-best | 0.038 | 0.046 | 0.087 |
flexmf-explicit | hyperopt-best | 0.037 | 0.046 | 0.102 |
als-biased | random-best | 0.025 | 0.042 | 0.065 |
flexmf-explicit | default | 0.021 | 0.029 | 0.062 |
uknn-explicit | default | 0.016 | 0.032 | 0.045 |
uknn-explicit | hyperopt-best | 0.014 | 0.033 | 0.043 |
uknn-explicit | random-best | 0.014 | 0.033 | 0.042 |
uknn-explicit | optuna-best | 0.013 | 0.033 | 0.042 |
bias | default | 0.005 | 0.030 | 0.030 |
Rating Prediction
RMSE | MAE | ||
---|---|---|---|
model | variant | ||
als-biased | hyperopt-best | 0.749 | 0.596 |
als-biased | optuna-best | 0.753 | 0.598 |
als-biased | random-best | 0.755 | 0.600 |
flexmf-explicit | optuna-best | 0.757 | 0.602 |
iknn-explicit | hyperopt-best | 0.759 | 0.604 |
iknn-explicit | random-best | 0.759 | 0.604 |
iknn-explicit | optuna-best | 0.759 | 0.604 |
flexmf-explicit | hyperopt-best | 0.759 | 0.603 |
iknn-explicit | default | 0.760 | 0.605 |
flexmf-explicit | random-best | 0.761 | 0.605 |
als-biased | default | 0.762 | 0.608 |
flexmf-explicit | default | 0.777 | 0.618 |
uknn-explicit | random-best | 0.778 | 0.617 |
uknn-explicit | hyperopt-best | 0.778 | 0.618 |
uknn-explicit | optuna-best | 0.778 | 0.617 |
bias | optuna-best | 0.783 | 0.630 |
bias | hyperopt-best | 0.783 | 0.630 |
bias | random-best | 0.783 | 0.630 |
uknn-explicit | default | 0.784 | 0.622 |
bias | default | 0.791 | 0.635 |