Parts of architecture | Components of each part | ||||||
---|---|---|---|---|---|---|---|
Layer’s name | Number of filters | Kernel size | Pool size | Activation function | Padding type | Dropout ratio | |
CNN | Convolution | 512 | 5 | Â | relu | Same | Â |
Dropout | Â | Â | Â | Â | Â | 0.3 | |
Average pooling | Â | Â | 3 | Â | Same | Â | |
Convolution | 256 | 3 | Â | relu | Same | Â | |
Dropout | Â | Â | Â | Â | Â | 0.3 | |
Convolution | 64 | 3 | Â | relu | Same | Â | |
Average pooling | Â | Â | 3 | Â | Same | Â | |
Convolution | 128 | 3 | Â | relu | Same | Â | |
Convolution | 256 | 5 | Â | relu | Same | Â | |
Dropout | Â | Â | Â | Â | Â | 0.3 | |
Convolution | 512 | 7 | Â | relu | Same | Â | |
Dropout | Â | Â | Â | Â | Â | 0.3 | |
Average pooling | Â | Â | 3 | Â | Same | Â |
 | Layer’s name | Number of units | Activation function | ||||
---|---|---|---|---|---|---|---|
LSTM | LSTM | Â | Â | 512 | Â | Â | tanh |
 | Layer’s name | Number of neurons | Activation function | ||||
---|---|---|---|---|---|---|---|
Fully connected network | Dense | Â | Â | 100 | Â | Â | relu |
Dense | Â | Â | 28 | Â | Â | relu | |
Dense | Â | Â | 64 | Â | Â | relu | |
Dense | Â | Â | 6 | Â | Â | softmax |