Introduction to GANs with Python and TensorFlow
Introduction
Generative models are a family of AI architectures whose aim is to create data samples from scratch. They achieve this by capturing the data distributions of the type of things we want to generate.
These kind of models are being heavily researched, and there is a huge amount of hype around them. Just look at the chart that shows the numbers of papers published in the field over the past few years:
Since 2014, when the first paper on Generative Adversarial Networks was published, generative models are becoming incredibly powerful, and we are now able to generate hyper-realistic data samples for a wide range of distributions: images, videos, music, pieces of writing, etc.
Here are some examples of images generated by a GAN:
What are Generative Models?
The GANs Framework
The most successful framework proposed for generative models, at least over recent years, takes the name of Generative Adversarial Networks (GANs).
Simply put, a GAN is composed of two separate models, represented by neural networks: a generator G and a discriminator D. The goal of the discriminator is