Fixed optim update error with non-contiguous grads/params #1187
Conversation
This file contains 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
|
@Edenzzzz Thanks for the PR! This looks good to me! |
|
Just ran the optimizer test suite as well, everything passing. Thanks @Edenzzzz and @matthewdouglas for tracking this down and providing a fix 🙌🏻 🚀, this was really helpful! |
Titus-von-Koeller
merged commit a3f55ce
into
bitsandbytes-foundation:main
22 checks passed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Fixes #1185
Non-contiguous params/gradients resulting from
torch.chunkandall_gatheretc. are ubiquitous in distributed training frameworks such as ZeRO. This avoids update errors as the C++ kernels assume row-major inputs.