Table 4.
FDC-3 Model Hyper-Parameters.
| Layer (Type) | Output Shape | Param # |
|---|---|---|
| Conv2D | (None, 128, 128, 32) | 320 |
| BatchNormalization | (None, 128, 128, 32) | 128 |
| Conv2D | (None, 126, 126, 32) | 9248 |
| BatchNormalization | (None, 126, 126, 32) | 128 |
| MaxPooling2D | (None, 63, 63, 32) | 0 |
| Dropout | (None, 63, 63, 32) | 0 |
| Conv2D | (None, 63, 63, 32) | 18,496 |
| BatchNormalization | (None, 63, 63, 32) | 256 |
| Conv2D | (None, 61, 61, 64) | 36,928 |
| BatchNormalization | (None, 61, 61, 64) | 256 |
| MaxPooling2D | (None, 31, 31, 64) | 0 |
| Dropout | (None, 31, 31, 64) | 0 |
| Conv2D | (None, 31, 31, 64) | 36,928 |
| BatchNormalization | (None, 31, 31, 64) | 256 |
| Conv2D | (None, 29, 29, 64) | 36,928 |
| BatchNormalization | (None, 29, 29, 64) | 256 |
| MaxPooling2D | (None, 15, 15, 64) | 0 |
| Dropout | (None, 15, 15, 64) | 0 |
| Conv2D | (None, 15, 15, 64) | 73,856 |
| BatchNormalization | (None, 15, 15, 64) | 512 |
| Conv2D | (None, 13, 13, 128) | 147,584 |
| BatchNormalization | (None, 13, 13, 128) | 512 |
| MaxPooling2D | (None, 7, 7, 128) | 0 |
| Dropout | (None, 7, 7, 128) | 0 |
| GlobalAveragePooling | (None, 128) | 0 |
| Dense | (None, 6) | 1290 |
| Activation | (None, 6) | 0 |
| Total Parameters: 363,366 | ||
| Trainable Parameters: 362,214 | ||
| Non-Trainable Parameters: 1152 | ||