<< Click to Display Table of Contents >> Navigation:  Introduction > Architectures included in Distribution > Image Classification

VGG-16 is written for 2D data in TensorFlow, and contains 16 layers. It is dedicated to classification.

The number of dense units was reduced from 4096 to 1024. This architecture is available as Image Classification in the Create Learning Set dialog. The loss function used for this architecture is categorical cross entropy.

 

References:

https://neurohive.io/en/popular-networks/vgg16/