If device="auto" and multiple GPUs are present, only select the first. #517
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
We're seeing poor prediction quality when multiple GPUs are used. Switch to only using multiple GPUs if explicitly enabled, while we debug.
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request addresses a bug with multi-GPU inference when device="auto" by correctly defaulting to a single GPU (cuda:0) as a temporary fix. The implementation change in infer_devices is simple and effective. The accompanying docstring updates in TabPFNClassifier and TabPFNRegressor clearly communicate this new behavior to users.
My main concern, which is critical, is that this change breaks an existing unit test. I've left a specific comment with a suggested fix to ensure the test suite passes. Please address this to maintain code quality and test coverage.
noahho
approved these changes
Sep 17, 2025
Collaborator
noahho
left a comment
noahho
left a comment
There was a problem hiding this comment.
LGTM barring geminis comment
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Outdated suggestions cannot be applied.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.
We're seeing prediction errors when multiple GPUs are used. Switch to only using multiple GPUs if explicitly enabled, while we debug.
Tested manually on a machine with 2 GPUs and a dataset exhibiting the issue:
Also, update the device docstring. Don't include multi-gpu inference for now, until we've fixed it.