2. Implementing DCGAN in Keras

Figure 4.2.1 shows DCGAN that is used to generate fake MNIST images:

Figure 4.2.1: A DCGAN model

DCGAN implements the following design principles:

• Use strides > 1, and a convolution instead of MaxPooling2D or UpSampling2D. With strides > 1, the CNN learns how to resize the feature maps.
• Avoid using Dense layers. Use CNN in all layers. The Dense layer is utilized only as the first layer of the generator to accept the z-vector. The output of the Dense layer is resized and becomes the input of the succeeding CNN layers.
• Use Batch Normalization (BN) to stabilize learning by normalizing the input to each layer to have zero mean and unit variance. There is no BN in the generator output layer and discriminator input layer. In the implementation example to be presented here, no batch normalization is used in the discriminator.
• Rectified Linear Unit (ReLU) is used in all layers of the generator except...