Keras Deep Learning Cookbook

Generator

Creating a generator consists of implementing two layers of convolutional layers with leaky ReLU activations:

• Dense layer with output of (128*7*7)
• Reshape it to (7,7,128)
• Upsample 2D (upsampling refers to a technique that will upsamples an image to a higher resolution; 2D means upsample two 2D images)
• Convolution 2D with an output of 128 filters
• Batch normalization
• RELU activation
• Upsample 2D
• Convolution 2D with an output of 64 filters
• Batch normalization
• RELU activation
• Convolution 2D with an output of three filters
• Last activation of tanh:

def build_generator(self):    model = Sequential()    model.add(Dense(128 * 7 * 7, activation="relu", input_dim=self.latent_dim))    model.add(Reshape((7, 7, 128)))    model.add(UpSampling2D()) model.add(Conv2D( ...