Table 2 The detailed Resnet parameters
From: Detection method of absence seizures based on Resnet and bidirectional GRU
layer | The name of network layer | Activation function | Output size | Convolution kernel size | Filters | parameters |
|---|---|---|---|---|---|---|
input | – | 64 × 256 × 19 | – | – | ||
1 | conv2d_1 | RELU | 64 × 256 × 32 | 3 × 3 | 32 | 5472 |
2 | conv2d_2 | – | 64 × 256 × 32 | 3 × 3 | 32 | 9216 |
3 | conv2d | – | 64 × 256 × 32 | 1 × 1 | 32 | 640 |
4 | Add+RELU | RELU | 64 × 256 × 32 | – | – | – |
5 | Max_pooling | – | 64 × 128 × 32 | – | – | – |
6 | conv2d_3 | – | 64 × 128 × 64 | 1 × 1 | 64 | 2112 |
7 | conv2d_4 | RELU | 64 × 128 × 64 | 3 × 3 | 64 | 18,432 |
8 | conv2d_5 | – | 64 × 128 × 64 | 3 × 3 | 64 | 36,864 |
9 | Add+RELU | RELU | 64 × 128 × 64 | – | – | – |
10 | Max_pooling | – | 64 × 64 × 64 | – | – | – |
11 | conv2d_6 | – | 64 × 64 × 128 | 1 × 1 | 128 | 8320 |
12 | conv2d_7 | RELU | 64 × 64 × 128 | 3 × 3 | 128 | 73,728 |
13 | conv2d_8 | – | 64 × 64 × 128 | 3 × 3 | 128 | 147,456 |
14 | Add+RELU | RELU | 64 × 64 × 128 | – | – | – |
15 | Max_pooling | – | 64 × 32 × 128 | – | – | – |
16 | conv2d_9 | – | 64 × 32 × 64 | 1 × 1 | 128 | 8256 |
17 | conv2d_10 | RELU | 64 × 32 × 64 | 3 × 3 | 128 | 73,728 |
18 | conv2d_11 | – | 64 × 32 × 64 | 3 × 3 | 128 | 36,864 |
19 | Add+RELU | RELU | 64 × 32 × 64 | – | – | – |
20 | Max_pooling | – | 64 × 16 × 64 | – | – | – |