October 2018
Intermediate to advanced
472 pages
10h 57m
English
The VGG16 model is a deep convolution neural network image classifier. The model uses a combination of Conv2D, MaxPooling2D, and Dense layers to form the final architecture, and the activation function that's used is ReLU. It accepts color images of shape 224*224*3, and is capable of predicting 1,000 classes. This means that the final Dense layer has 1,000 neurons, and it uses softmax activation to get scores for each class.
Read now
Unlock full access