Generate synthetic faces using Generative Adversarial Network (GAN). In this project, here defined and trained a DCGAN on a dataset of faces. The goal is to get a generator network to generate new images of faces that look real but don't exist in reality.