Conversation
|
Any reason this has not been merged? As context this also fixed an issue we have by saving a distributed checkpoint on a 8x16 and then loading it in other topology. |
Collaborator
Author
|
@alanwaketan Do you have a sec to review? I'll also add @JackCaoG |
Collaborator
|
does this change needs to go into 2.4 release(branch cut was yesterday)? |
JackCaoG
approved these changes
Jun 11, 2024
Collaborator
Author
|
Yes, we should cherry-pick this. I'll open a PR. |
jonb377
added a commit
that referenced
this pull request
Jun 11, 2024
This was referenced Jun 11, 2024
alanwaketan
reviewed
Jun 12, 2024
Collaborator
alanwaketan
left a comment
There was a problem hiding this comment.
LGTM. Sorry for missing this PR.
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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In create_read_items_for_chunk_list, the ReadItem's
dest_offsetsare set to index into the local shard. Despite this, we performed an translation from global to local indices. This caused issues restoring checkpoints with padded tensors, i.e. where the global tensor size is not evenly divisible by the shard size.We should instead rely on
dest_offsetsdirectly to narrow the local shards.