A simple explanation of the Inception score

The Inception score (IS) is a popular metric for judging the image outputs of Generative Adversarial Networks (GANs). A GAN is a network that learns how to generate (hopefully realistic looking) new unique images similar to its training data. Most papers about GANs use the IS to show their improvement versus the prior art. In this article, I'll explain how the score works

Read the full article

March 6, 2019