AliceAria

Thank you for sharing your codes. I found that there is no weight clipping in the paper Least Squares Generative Adversarial Networks. The weight clipping is used in the paper...

Does the is_score correct?

1

I test the is-score for the cifar-10 dataset with the code and I get the score 8.6(+-0.09). While the score for cifar-10 is 11.24 in papers.