@bot /style

Style fixes have been applied. View the workflow run here.

@gapatron Thank you for the PR. The changes look good to me!

Could you run make fix-copies and push the relevant changes? Also, could you share some generation results and comparisons for anyone reviewing the PR to have something to look at immediately?

cc @asomoza

Of course! I am working on a background infilling problem for astronomical images. The Laplace scheduler gives me more consistent results across configurations, in general. Please take a look at the images below, which contain a significant region masked and then infilled with various trained models. All of them are taken from an evaluation batch, at epoch 25.

Laplace Schedule
epsilon-prediction

x-prediction

v-prediction

Cosine Schedule
epsilon-prediction

x-prediction

v-prediction

Linear Schedule
epsilon-prediction

x-prediction

v-prediction

Laplace scheduler, empirically, is the more robust and the leas likely to blow up without clip sampling. Additionally, for large masks the Laplace scheduler consistently provides me with the smoother, most natural infills. I attach the numerical results of these experiments, in case they are helpful.

Results
results.csv

P.S.: @a-r-r-o-w a ,@yiyixuxu is this okay? I tried pushing it from my Windows, which gave me some issues, so I had to force-revert to a previous version of the commit.

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

ping?

Hi all — just following up here!

I’ve reset the PR to its clean state (Laplace scheduler only), provided generation results, and included comparisons showing consistent improvements in stability and quality across configurations.

Please let me know if there’s anything else I should address or tweak... happy to follow up. Would love to see this scheduler added, especially given its robustness across loss types and masking setups. I've personally found more success using it for training inpainting models than other schedules, and it would be convenient to just have it added instead of writing it down every time I update diffusers.

cc @a-r-r-o-w @yiyixuxu

gentle ping @yiyixuxu @sayakpaul @DN6