Fix: allow Union[str, dict, None] fields like deepspeed to be passed via CLI#39625
Open
JH-debug wants to merge 2 commits intohuggingface:mainfrom
Open
Fix: allow Union[str, dict, None] fields like deepspeed to be passed via CLI#39625JH-debug wants to merge 2 commits intohuggingface:mainfrom
JH-debug wants to merge 2 commits intohuggingface:mainfrom
Conversation
Member
|
Makes sense, but please check the CI! It seems like a lot of tests are broken, so you might have to iterate on this to get it to pass. Also, you can resolve the quality issues with |
This was referenced Apr 29, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR fixes an issue in
HfArgumentParserthat causes aValueErrorwhen parsing fields of type Union[str, dict, None] from the command line, such as thedeepspeedargument in TrainingArguments.The original parser rejected this valid type signature, even though it is commonly used to support both JSON config files (via path strings) and dictionary input. This PR adds a specific check to treat such fields as str when parsed via CLI, while still allowing JSON/dict parsing from file-based or programmatic inputs.
Motivation & Context
When passing
--deepspeed path/to/ds_config.jsonfrom the CLI, the parser raised the following error:ValueError: Only Union[X, NoneType] (i.e., Optional[X]) is allowed for Union ...This was because
Union[str, dict, None]didn't meet the parser’s strict rule of "only one type allowed" in unions. However, this pattern is standard in transformers for flexible inputs (e.g., both string path and dict configs).This patch modifies HfArgumentParser._parse_dataclass_field() to detect this specific case and allow it.
Fixes
Closes a CLI parsing limitation for common use cases involving fields like:
deepspeed: Optional[Union[str, dict]] = field(default=None, ...)