Uh oh!
There was an error while loading. Please reload this page.
[PyTorch] TE checkpoint pass-through logic fix - #782
Merged
Conversation
Member
There was a problem hiding this comment.
@denera I think by using root_module.modules() instead, we can do away with recursion as well as requiring 2 functions (has/is module) since it also returns self
CollaboratorAuthor
There was a problem hiding this comment.
Didn't realize torch.nn.Module.modules() recursed into all submodules in the network. Definitely makes things easier. Fixed in the latest commit.
…TE submodules Signed-off-by: Alp Dener <adener@nvidia.com>
deneraforce-pushed
the
te-torch-checkpoint-passthrough-fix
branch
from
April 16, 2024 16:53
2b1c4f3 to
8209eb2CompareSigned-off-by: Alp Dener <adener@nvidia.com>
pggPL pushed a commit
to pggPL/TransformerEngine
that referenced
this pull request
May 23, 2024
* changed TE checkpoint passthrough logic to also recursively look for TE submodules Signed-off-by: Alp Dener <adener@nvidia.com> * simplified search for TE modules in the checkpointed network Signed-off-by: Alp Dener <adener@nvidia.com> --------- Signed-off-by: Alp Dener <adener@nvidia.com> Signed-off-by: Pawel Gadzinski <pgadzinski@nvidia.com>
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 freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR makes the TE checkpoint recursively search for TE submodules in the checkpointed module and only passes pure-PyTorch modules through to the native PyTorch checkpoint function.
Fix for #776.