[WiP] DDPO video WM trainer#16
Open
pmcurtin wants to merge 9 commits into
Open
Conversation
…diffusion model and sample fresh noise at each steps, also we estimate the variances for the schedule
|
to get rid of my vibe-fix of the log probs we might be able to make use of https://github.com/yifan123/flow_grpo/blob/main/flow_grpo/diffusers_patch/sd3_sde_with_logprob.py#L71 (thanks @fairy for digging this out!) |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
for now, no reward function as that will depend on the teacher model reward architecture. Will test with random rewards or something like that. Ignore the resources/ dir, it contains reference impl. of DDPO from the original ddpo-pytorch repo, and will be removed.