yojeep / conditional-wgan-gp Goto Github PK
View Code? Open in Web Editor NEWThis project forked from u7javed/conditional-wgan-gp
Implementation of a Wasserstein Generative Adversarial Network with Gradient Penalty to enforce lipchitz constraint. The WGAN utilizes the wasserstein loss or critic as its loss function instead of the vanilla GAN loss. It has shown to perform better as is often used as a solution to mode collapse,