Skip to content
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.

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

Black image: The result of ddim_sample is black, while p_sample_loop can generate fine images. #136

Open
GuHuangAI opened this issue Dec 9, 2022 · 5 comments

Comments

@GuHuangAI
Copy link

p_sample_loop:
generate5
ddim_sample:
sample-1
What happened?

@GuHuangAI
Copy link
Author

I find the balck image is not caused by ddim_sample but by ema model, why does the ema model generate wrong image? I set the ema_decay to 0.995.

@BEbillionaireUSD
Copy link

I find the balck image is not caused by ddim_sample but by ema model, why does the ema model generate wrong image? I set the ema_decay to 0.995.

How did you save it? Did you close ema?

@BEbillionaireUSD
Copy link

I find the balck image is not caused by ddim_sample but by ema model, why does the ema model generate wrong image? I set the ema_decay to 0.995.

How did you save it? Did you close ema? I have the same problem. Could you please share your solution? Thanks:)

@GuHuangAI
Copy link
Author

@cherylLbt Yeah, i closed ema. You can replace self.ema.ema_model.sample by self.model.module.sample in denoising_diffusion.py.

@BEbillionaireUSD
Copy link

@cherylLbt Yeah, i closed ema. You can replace self.ema.ema_model.sample by self.model.module.sample in denoising_diffusion.py.

Many thanks for your reply!!🙏

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants