FSDP enhancements and fixes #1753
Merged
Conversation
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
|
The documentation is not available anymore as the PR was closed or merged. |
pacman100
changed the title
pacman100
changed the title
pacman100
changed the title
sgugger
approved these changes
Jul 21, 2023
Collaborator
sgugger
left a comment
sgugger
left a comment
There was a problem hiding this comment.
Makes sense to me, thanks!
4 tasks
winglian
pushed a commit
to OpenAccess-AI-Collective/accelerate
that referenced
this pull request
Jul 25, 2023
* if the model is already an FSDP instance, remove the warning and prep overhead * allow usage of `_no_split_modules` to simplify UX when using FSDP * Update other.py * fixes
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Outdated suggestions cannot be applied.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.
What does this PR do?
transformer_layer_cls_to_wraptomodel._no_split_modules? transformers#24568 via fsdp fixes and enhancements transformers#24980extract_from_parallelhow unwraps FSDP models too but these will lead to FSDP issues if there are modules as part of the global FSDP unit as the parameters of those modules won't be gathered if one isn't calling global FSDP wrapper's forward call. This is wrt issue Validation Errors When Training ControlNet with FSDP diffusers#4037