Table 1 Parameters for proposed transformer-based MI EEG Recognition model.
From: Deep temporal networks for EEG-based motor imagery recognition
Model setting | Â | Training setting | |
|---|---|---|---|
Number of transformers | 4 | Input shape | N \(\times \) 118 \(\times \) 1 |
Dimension of token | 118 \(\times \) 1 | Batch size | 200 |
Number of attention heads | 4 | Learning rate | 0.0001 |
| Â | Â | Training epochs | 278 |
Dataset size 196,000 \(\times \) 118 \(\times \) 1 | Â | Â | Â |
Total trainable parameters 44,106 | Â | Â | Â |
Training time epoch 44,106 | Â | Â | Â |
