Implementation of ideas from the paper Generative Adversarial Networks
Replicating core ideas from the paper Generative Adversarial Networks
Generator seems to be working reasonably well given the model size. Learning rate set at 3e-4 with ADAM as optimizer
