HF component update to not use fsspec components #159405
Closed
Conversation
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
Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors, which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159405
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit d3a9c53 with merge base 93da995 ( UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D78993550 |
Skylion007
reviewed
Jul 30, 2025
| ) | ||
| super().__init__( | ||
| path=path, | ||
| serialization_format=SerializationFormat.SAFETENSORS, |
There was a problem hiding this comment.
Safe open cannot use file like API of fsspec? That is surprising to me?
There was a problem hiding this comment.
Ya it requires a path unfortunately and doesn't take the file-like API.
Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors, which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/) cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta [ghstack-poisoned]
ankitageorge
added a commit
that referenced
this pull request
Aug 1, 2025
Pull Request resolved: #159405 Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors, which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. ghstack-source-id: 300108423 @exported-using-ghexport Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/)
|
This pull request was exported from Phabricator. Differential Revision: D78993550 |
Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors, which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/) cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta [ghstack-poisoned]
ankitageorge
added a commit
that referenced
this pull request
Aug 1, 2025
Pull Request resolved: #159405 Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors, which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. ghstack-source-id: 300168979 @exported-using-ghexport Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/)
|
This pull request was exported from Phabricator. Differential Revision: D78993550 |
saumishr
reviewed
Aug 6, 2025
| @@ -52,7 +52,7 @@ | |||
| __all__ = ["HuggingFaceStorageWriter", "HuggingFaceStorageReader"] | |||
|
|
|||
|
|
|||
| class HuggingFaceStorageWriter(FsspecWriter): | |||
| class HuggingFaceStorageWriter(FileSystemWriter): | |||
There was a problem hiding this comment.
@ankitageorge If any user needs the FSSpec based component, they should be able to write a wrapper extending the HuggingFaceStorageWriter. Could you please confirm if thats the case.
There was a problem hiding this comment.
ya if they write a class to inherit from both HuggingFaceStorageWriter and FsspecWriter, then it should work.
saumishr
reviewed
Aug 6, 2025
| @@ -52,7 +52,7 @@ | |||
| __all__ = ["HuggingFaceStorageWriter", "HuggingFaceStorageReader"] | |||
|
|
|||
|
|
|||
| class HuggingFaceStorageWriter(FsspecWriter): | |||
| class HuggingFaceStorageWriter(FileSystemWriter): | |||
There was a problem hiding this comment.
Will this be backward compatible? Its a very new feature so probably no major adoption yet but we enabled a few partners to start using the FSSpec component.
There was a problem hiding this comment.
The only usage I could find is torchtune, but considering it's being deprecated I think it is okay. The flow will just go back to users having to download files from HF hub before running, which was the old experience.
saumishr
approved these changes
Aug 7, 2025
Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors (30s vs 4s for load of 8b model), which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. Also, we can use the official methods provided by HF instead of relying on reading the metadata by bytes and loading it Differential Revision: [D78993550](https://our.internmc.facebook.com/intern/diff/D78993550/) cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D78993550 |
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
|
Starting merge as part of PR stack under #159681 |
|
Starting merge as part of PR stack under #159406 |
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 7, 2025
Reading the bytes and converting to tensors is much slower than using safe_open. For a 8B model across 8 ranks, took ~30s to load before this change and ~4s after. Differential Revision: [D78994259](https://our.internmc.facebook.com/intern/diff/D78994259/) Pull Request resolved: #159406 Approved by: https://github.com/saumishr ghstack dependencies: #159405
pytorchmergebot
pushed a commit
that referenced
this pull request
Aug 7, 2025
Get rid of the logic to read the metadata from the header of the safetensors file manually and use the functions as part of safe_open() to get the metadata. This is much cleaner and allows us to not rely on our own custom methods to get metadata, but use safetensors provided APIs Differential Revision: [D79460272](https://our.internmc.facebook.com/intern/diff/D79460272/) Pull Request resolved: #159681 Approved by: https://github.com/saumishr ghstack dependencies: #159405, #159406
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
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.
Stack from ghstack (oldest at bottom):
Update HF components to not inherit from fsspec components and instead use filesystem writer/reader. The reason is because there doesn't seem to be much of a need for fsspec, since users are using mounted storage. Using local storage will allow for performance improvements because we can take advantage of the safe_open API provided by HF safetensors (30s vs 4s for load of 8b model), which is signifcant performance wins over reading bytes and converting to tensors which is what we are doing now. Also, we can use the official methods provided by HF instead of relying on reading the metadata by bytes and loading it
Differential Revision: D78993550
cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta