Table 3.
Characteristic features of adopted pretrained networks.
| Model/network | Number of layers | Learnable parameters (in millions) | Input size of the image |
|---|---|---|---|
| AlexNet | 8 | 60.0 | 227 × 227 |
| VGG-16 | 16 | 137.0 | 224 × 224 |
| GoogLeNet | 22 | 7.1 | 224 × 224 |
| ResNet-50 | 50 | 25.7 | 224 × 224 |