Hands-On Machine Learning for Algorithmic Trading

Defining the generator network

The generator network is slightly shallower but has more than twice as many parameters:

def build_generator():    model = Sequential([        Dense(128 * 7 * 7, activation='relu', input_dim=latent_dim),        Reshape((7, 7, 128)),        UpSampling2D(),        Conv2D(128, kernel_size=3, padding='same'),        BatchNormalization(momentum=0.8),        Activation('relu'),        UpSampling2D(),        Conv2D(64, kernel_size=3, padding='same'),        BatchNormalization(momentum=0.8),        Activation('relu'),        Conv2D(channels, kernel_size=3, padding='same'),        Activation('tanh')])    model.summary()    noise = Input(shape=(latent_dim,))    img = model(noise)    return Model(noise, img)