Table 2 Independent residual network architecture
From: Research on epileptic EEG recognition based on improved residual networks of 1-D CNN and indRNN
Layer | Hidden layer | Related parameters (filters, kernels, stride) | ||
|---|---|---|---|---|
BLOCK1 | Conv1D+BN+LeakyReLU | 64 | 8 | 1 |
Conv1D+BN+LeakyReLU | 64 | 5 | 2 | |
Conv1D+BN | 64 | 3 | 1 | |
Conv1D+BN | 64 | 1 | 1 | |
Add | – | – | – | |
LeakyReLU | – | – | – | |
BLOCK2 | Conv1D+BN+LeakyReLU | 128 | 8 | 1 |
Conv1D+BN+LeakyReLU | 128 | 5 | 2 | |
Conv1D+BN | 128 | 3 | 1 | |
Conv1D+BN | 128 | 1 | 1 | |
Add | – | – | – | |
LeakyReLU | – | – | – | |
BLOCK3 | Conv1D+BN+LeakyReLU | 64 | 8 | 1 |
Conv1D+BN+LeakyReLU | 64 | 5 | 2 | |
Conv1D+BN | 64 | 3 | 1 | |
Add | – | – | – | |
LeakyReLU | – | – | – | |
GlobalAveragePooling1D | – | 2 | – | |
indRNN+BN | 128 | |||
indRNN+BN | 128 | |||
Fully connected | 256 | |||
Softmax | n_class | |||