Get Started: DCGAN for Fashion-MNIST - PyImageSearch

In this tutorial, we are implementing a Deep Convolutional GAN (DCGAN) with TensorFlow 2 / Keras, based on the paper, Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks (Radford et al., 2016). This was one of the earliest GAN papers and is typically what you'd read to get started with learning GANs. Before we get started, are you familiar with how GANs work? If not, be sure to look at my previous post, "Intro to GANs," for a high-level intuition of how GANs work in general. Each GAN has at least one generator and one discriminator.