Python implementation of GAN
A basic GAN (generative adversarial network) is implemented for learning and testing. Currently it is based on the MNIST dataset and creates synthetic handwritten digits. In a later step this will be expanded to create synthetic 3d models.
For execution run the jupyter notebook.
It is based on this website.