Table 4 The list of CNN forecast component parameters
From: Time-series analysis with smoothed Convolutional Neural Network
Category | Parameter | Value |
|---|---|---|
Convolutional layer | Type of convolutional | Conv1D |
The number of convolutional layers | 3 | |
The number of filters | 128 | |
The filter size | 2 | |
The activation function | ReLU | |
Pooling layer | Type of pooling | MaXPooling1D |
The number of pooling layer | 1 | |
Size of the pooling window | 2 | |
Flattent layer | The number of flatten layer | 1 |
Fully connected layer | The number of hidden layers | Lucas number |
The number of units or neuron | 100 | |
The Activation function output | ReLU | |
Loss function | MSE | |
Type of optimizer | Adam | |
The number of epochs | 10,000 | |
The batch size | 16 |