Keras Deep Learning Cookbook

Generator

We create a sequential model with the following layers:

• Dense layer with an input of (self.latent_dim) and output of (*, 256 units )
• Leaky ReLU layer applies this function to incoming data
• Batch Normalization: normalizes the data

• Dense layer of 512: Layer with output of (*, 512 units)
• Leaky ReLU layer applies this function to incoming data
• Batch normalization
• Dense layer of (*, 1024)
• Leaky RELU 
• Batch normalization
• Dense layer of size (*, 256) with activation tanh
• Reshape back to img_shape
• Add some noise to the model of type shape=(self.latent_dim,):

def build_generator(self):    model = Sequential()    model.add(Dense(256, input_dim=self.latent_dim))    model.add(LeakyReLU(alpha=0.2))    model.add(BatchNormalization(momentum=0.8)) model.add(Dense( ...