Training the networks

To train the 3D-GAN, perform the following steps:

  1. Start by specifying the values for the different hyperparameters required for the training, shown as follows:
gen_learning_rate = 0.0025dis_learning_rate = 0.00001beta = 0.5batch_size = 32z_size = 200DIR_PATH = 'Path to the 3DShapenets dataset directory'generated_volumes_dir = 'generated_volumes'log_dir = 'logs'
  1. Next, create and compile both of the networks, shown as follows:
# Create instancesgenerator = build_generator()discriminator = build_discriminator()# Specify optimizer gen_optimizer = Adam(lr=gen_learning_rate, beta_1=beta)dis_optimizer = Adam(lr=dis_learning_rate, beta_1=0.9)# Compile networksgenerator.compile(loss="binary_crossentropy", optimizer="adam" ...

Get Generative Adversarial Networks Projects now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.