Table 1.
Model | Hyper-parameter description | |
---|---|---|
Hyper-parameter | Value | |
CNN-LSTM | LSTM layer | 100 units |
LSTM layer activation function | Tanh | |
Epoch | 100 | |
Batch size | 64 | |
Max-pooling layer size | 2 | |
Pooling layer activation function | Relu | |
Pooling layer padding | Same | |
Loss function | MSE | |
Optimizer | ADAM | |
Learning rate | 0.001 | |
Convolutional layer size (32 filters) | 2 | |
Convolutional layer size (64 filters) | 2 | |
Convolution layer activation function | tanh | |
Convolution layer padding | Same |