[fix] Add trailing slash to local directory for uploads to existing cloud directory with fsspec#1614
Merged
Conversation
…oud directory with fsspec Signed-off-by: SumanthRH <sumanthrh99@gmail.com>
Contributor
There was a problem hiding this comment.
Code Review
This pull request modifies the upload_directory function in skyrl/backends/skyrl_train/utils/io/io.py to ensure the local path ends with a trailing slash, preventing fsspec from creating unwanted subdirectories during synchronization. A review comment points out that the manual string concatenation used is not cross-platform and suggests using os.path.join to ensure the trailing separator is handled correctly on all operating systems.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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?
Fixes an issue with checkpoint upload format with LoRA. With LoRA, we make multiple calls to
io.local_work_dir, which in turn callsio.upload_directory. Each time, we write checkpoint files to a temporary file and then upload them to directory in cloud storage. Currently, subsequent calls toio.upload_directoryleads to nesting of ckpt contents:The root cause is that with
fsspec, while uploading files in a directorysrcto an existing directorydstwith fsspec, we need to ensure that the file path ends in a trailing slash. otherwise, fsspec will create a subdirectorydst/srcinstead of directly syncing contents ofsrcinto the rootdstdirectory