Correct SNR weighted loss in v-prediction case by only adding 1 to SNR on the denominator #6307
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
should also be able to fix sample prediction_type (x-pred) |
examples/kandinsky2_2/text_to_image/train_text_to_image_decoder.py
Outdated
Show resolved
Hide resolved
sayakpaul
left a comment
sayakpaul
left a comment
There was a problem hiding this comment.
Thanks for your PR.
https://github.com/huggingface/diffusers/pull/6307/files#r1436118465 is my primary question.
|
done |
|
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. |
|
Will merge after the CI is green. |
|
https://github.com/huggingface/diffusers/tree/main/examples/dreambooth was missed with this change |
|
@siegelaaron94 would you like to create a PR? |
…R on the denominator (huggingface#6307) * fix minsnr implementation for v-prediction case * format code * always compute snr when snr_gamma is specified --------- Co-authored-by: Sayak Paul <[email protected]>
5 participants
What does this PR do?
Fixes #5654 (issue)
Before submitting
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.
@bghira