Table 5 Partial hyperparameter settings for TUSZ dataset and SWEZ dataset.
From: Combining meta and ensemble learning to classify EEG for seizure detection
Hyper-parameter | TUSZ | SWEZ |
|---|---|---|
Policy type | Gaussian | Gaussian |
Gamma | 0.99 | 0.99 |
Replay memory size | 1100 | 1100 |
Mini-batch size | 256 | 256 |
Steps of gradient updates | 3500 | 2500 |
Steps of random actions | 500 | 500 |
State-size (2n) | 14 | 14 |
max_estimators (m) | 37 | 37 |
