# A Very Short Introduction to Inception Score(IS)

Generative Adversarial Networks or GANs for short were successful in generating high-quality images, videos, and audio. We have seen various use cases of GANs. Some of the popular GAN networks are BigGAN, StyleGAN, GameGAN, and PGGAN. They have widespread adoption in the industry and academia. But, evaluation of generated samples can be very tricky and prone to errors if done subjectively by humans. IS was introduced to overcome this problem.

# What is the Inception Score?

The Inception Score(IS) is an objective performance metric, used to evaluate the quality of generated images or synthetic images, generated by Generative Adversarial Networks(GANs). It measures how realistic and diverse the output images are. It can be used instead of subjective evaluation by humans. After FID(Frechlet Inception Distance), it is the second most important evaluation performance metric. I have written an article explaining FID and how to calculate it.

1. Quality (Goodness)— How good the generated images are — Low entropy with high predictability is required.

# Formula

The formula to calculate the Inception Score(IS) is

1. Marginal Probability Distribution — p(y)

# How to calculate the Inception Score?

1. Pass the generated images through the Inception model to get the conditional label distribution p(y|x)
2. Calculate the marginal probability distribution p(y)
3. Calculate the KL Divergence between p(y) and p(y|x)
4. Calculate the sum over classes and take the average of outputs over images
5. Finally, take the exponential of the averaged value.

# Limitations of the Inception Score

1. IS will be high if the network generates one image per class only, which is not a good representation of the classes.
2. The Inception Network is trained on the Imagenet dataset consists of 1000 classes only. The Inception score will be low if you train your GAN on a class outside of these 1000 classes.
3. It works on the square and relatively small size images —ex. 300x300

Co-founder - Mate Labs | Co-founder - Raven Protocol | Author - Generative Adversarial Networks Projects | Democratizing Artificial Intelligence

## More from Kailash Ahirwar

Co-founder - Mate Labs | Co-founder - Raven Protocol | Author - Generative Adversarial Networks Projects | Democratizing Artificial Intelligence