Experimental Setup: PGGANs

Discover how to implement the progressive growing methodology of GANs, covering training techniques and supporting functions.

In this lesson, we are going to implement training functions that are compatible with the PGGAN methodology. These training and logging functions will be used to train the discriminator and the generator, which will also be implemented in this lesson.

We will also implement helper functions in order to make it easier to compute our loss function based on the WGAN-GP.

Training

Our training method consists of the following two parts:

  • Initializing the required variables and compiling models

  • The training loop itself

We will start by looking at the training layout:

Get hands-on with 1200+ tech skills courses.