VGGVariant
Enums · AiDotNet.Enums
Defines the available VGG network architecture variants.
For Beginners
VGG networks are named after the Visual Geometry Group that created them. The number in the name (e.g., VGG16) refers to the total number of weight layers in the network. For example, VGG16 has 13 convolutional layers and 3 fully connected layers, totaling 16 weight layers. These networks were groundbreaking because they showed that network depth is critical for good performance. Despite being older architectures, they remain popular for transfer learning and as baselines.
How It Works
VGG networks are a family of deep convolutional neural networks developed by the Visual Geometry Group at Oxford University. They are characterized by their use of small (3x3) convolution filters stacked in increasing depth, which allows them to learn complex features while keeping the number of parameters manageable.
Batch Normalization Variants: The “_BN” suffix indicates variants that include batch normalization layers after each convolutional layer. Batch normalization helps stabilize training and often allows for faster convergence and better final accuracy.
Fields
| Field | Summary |
|---|---|
VGGVariant VGG11 |
VGG-11: 11 weight layers (8 conv + 3 FC). |
VGGVariant VGG11_BN |
VGG-11 with batch normalization after each convolutional layer. |
VGGVariant VGG13 |
VGG-13: 13 weight layers (10 conv + 3 FC). |
VGGVariant VGG13_BN |
VGG-13 with batch normalization after each convolutional layer. |
VGGVariant VGG16 |
VGG-16: 16 weight layers (13 conv + 3 FC). |
VGGVariant VGG16_BN |
VGG-16 with batch normalization after each convolutional layer. |
VGGVariant VGG19 |
VGG-19: 19 weight layers (16 conv + 3 FC). |
VGGVariant VGG19_BN |
VGG-19 with batch normalization after each convolutional layer. |
int value__ |