KANGXI123456

Results 2 comments of KANGXI123456

getting NaN loss

Which dataset are you running on? I have similar issues on CelebA_align, and learn_lr=1e-5. I am getting inf loss. on v0.27.4 > [12f95b3](https://github.com/lucidrains/denoising-diffusion-pytorch/commit/12f95b33d8a2f44ffb48ec1079083ec634c4728f) seems to have fixed this issue for...

The result is too noisy

> Excuse me, can you tell me where to add the resblocks? Thanks you very much