Intuitive StarGAN Implemented in Tensorflow 2.3

Follow the full discussion on Reddit.
Lately, a lot of my work has been involving StarGAN, so I wanted to create an easy to read implementation of the architecture that functions, can be freely used, and most importantly that explains what's going on! [StarGAN](https://arxiv.org/abs/1711.09020) is a class translation model that uses a single generator to translate freely between some number of classes. It is an advanced GAN, and although outdated, shares some important ideas. My implementation documentation is aimed at those who have a familiarity with machine learning. Please let me know any feedback you might have, this is my first time sharing a project!