Table 7 Configurations of WCNN experiments.
From: Research on improved convolutional wavelet neural network
No. | Parameter type | Parameter name | WCNN |
|---|---|---|---|
1 | 1st to 5th layers | First type of NN | WCPNN |
2 | 2nd, 4th layers | Activation function of convolutional layer | Wavelet |
3 | 1st layer | Dimension of the 1st layer | \(28\times 28\) |
4 | 2nd layer | Dimension of 1st convolutional layer | \(28\times 28\) |
5 | 3rd layer | Dimension of 1st pooling layer | \(24\times 24\) |
6 | 2nd, 3rd layers | Number of features | \(6\) |
7 | 4th layer | Dimension of 2nd convolutional layer | \(12\times 12\) |
8 | 5th layer | Dimension of 2nd 1 pooling layer | \(8\times 8\) |
9 | 4th,5 th layers | Number of features | \(12\) |
10 | − 1st, − 2nd, − 3rd layers | Second type of NN | FCNN |
11 | − 3rd layer | Dimension of -3rd input layer | 192 |
12 | − 2nd layer | Dimension of -2nd hidden layer | None |
13 | − 2nd layer | Activation function of hidden layer | None |
14 | − 1st layer | Dimension of -1st output layer | 10 |
15 | − 1st layer | Activation function of output layer | Sigmoid |
16 | Hyperparameters | Learning rate \(\eta\) | 0.1 |
17 | Hyperparameters | Coefficient of inertia \(\mathrm{\alpha }\) | None |
18 | Hyperparameters | \(max\_SPs\) | 10 |
19 | Hyperparameters | \(max\_ACs\) | 6000 |
20 | Hyperparameters | \(target\_err\) | 0.0000001 |
21 | Hyperparameters | \(BatchSize\) | 10 |
