fjxmlzn / infogan-cr Goto Github PK

can you please provide a pretrained model for celebA and testing codes

can you provide how to test on your own images for the celeba pretrained model and the results sizes are really small and hard to see the difference currently

在附录A.1对Remark2的证明中

我的理解是：根据原文对Q的假设，L_info可以分为两个部分。 实际上，InfoGAN仅优化了第二部分，因此最优解也会产生第一部分的偏差。
但是这样来说岂不是，产生这种偏差的原因就是，先验分布不一定是方差为0的独立高斯分布？

期待并感谢您的指正！

How well do you perform on dSprites using the mutual information gap (MIG)? I'd expect this to do well, so would love to see results.

Do you have a new version of this code, in Tensorflow 2 or in Pytorch?

Thanks in advance