New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Question about the normalization of the input data for ddpm. #307

Open

MingCongSu opened this issue Apr 13, 2024 · 0 comments

MingCongSu commented Apr 13, 2024 •

edited

Just want to ask two naive questions as I am new to the diffusion model. 😅

I noticed that in README.md, it says that the data is normalized from 0 to 1.

denoising-diffusion-pytorch/README.md

Line 45 in d0c68fc

    
           training_images = torch.rand(8, 3, 128, 128) # images are normalized from 0 to 1

Why do we have to conduct this normalization on data? If we didn't do that, would DDPM still work properly?
If the input is latent like LDM does, do I have to make sure the latent space is also from 0 to 1?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment