StyleGAN - AshokBhat/ml GitHub Wiki About A GAN for producing an unlimited number of (often convincing) portraits of fake human faces. Introduced by Nvidia researchers in December 2018. Depends on Nvidia's CUDA software, GPUs and on TensorFlow An example picture (not a real person) See also GAN