I'm quite new to GANs and I am trying to use a Wasserstein GAN as an augmentation technique. I found this article https://www.sciencedirect.com/science/article/pii/S2095809918301127, and would like to replicate their method of evaluating the GAN. The method is shown in the figure.

In the article they write that they extract the generated samples that fooled the discriminator and use these to train a classifier. They also say that they use a Wasserstein GAN. Does anyone know how it is possible to extract samples that fooled the discriminator, since for a Wasserstein GAN the critic (discriminator) only puts a rating and not a label on the generated data?

enter image description here


1 Answer 1


My suggestion is to compare the critics rating for the real data and the generated data. Compute the mean of the ratings on real data to get some kind of threshold. If your critic is designed so that a high rating indicates a real sample, then generated data with ratings greater than the mean of the ratings of the real data could be viewed as good enough to fool the critic at that point in training.

This might be a too harsh criteria, but you will at least not accept poorly generated data.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .