Table 3 The structure of five layers (Conv. and FC.) for W-Net
From: WBC image classification and generative models based on convolutional neural network
Layers | Output size | Structure |
|---|---|---|
1st Conv. | 65,536 | 3 × 3 kernel, 1 stride, 16 filters |
|  |  | 2 × 2 max-pool, 2 strides, 0 pad |
2nd Conv. | 32,768 | 3 × 3 kernel, 1 stride, 32 filters |
|  |  | 2 × 2 max-pool, 2 strides, 0 pad |
3rd Conv. | 16,384 | 3 × 3 kernel, 1 stride, 64 filter |
|  |  | 2 × 2 max-pool, 2 strides, 0 pad |
1st FC. | 1024 | 1024 units |
2nd FC. | 5 | 5 units |