remove UtilsBuilder load, use torch (un)flatten ops#3728
Merged
jeffra merged 7 commits intodeepspeedai:masterfrom Jun 14, 2023
Merged
remove UtilsBuilder load, use torch (un)flatten ops#3728jeffra merged 7 commits intodeepspeedai:masterfrom
jeffra merged 7 commits intodeepspeedai:masterfrom
Conversation
inkcherry
requested review from
jeffra,
mrwyattii,
samyam and
tjruwase
as code owners
Collaborator
|
Thank you @inkcherry! This is a great reminder, I see that torch has updated their (un)flatten implementations pytorch/pytorch#58006. I also did a benchmark just now and see no difference between versions. I've pushed a commit to remove |
jeffra
approved these changes
Jun 14, 2023
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Torch (un)flatten currently has no performance issues, we have tested it on the CUDA device. Skip compiling unnecessary util.
It will not depend on the deepspeed compile process for many training processes. This may also be friendly to other device integration.