Module vgg 

VGG-16 model implementation.

VGG-16 is a convolutional neural network architecture. It consists of 13 convolutional layers followed by 3 fully connected layers.

Key characteristics:

• Conv layers with 3x3 filters
• Max pooling after every 2-3 conv layers
• Three fully connected layers of 4096, 4096, 1000 units
• ReLU activation and dropout

References:

• Very Deep Convolutional Networks for Large-Scale Image Recognition

Structs

Vgg

Enums

Models