[Data] Fix bug where you can't return objects and array from UDF #45287
Conversation
This file contains 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
Signed-off-by: Balaji Veeramani <balaji@anyscale.com>
Signed-off-by: Balaji Veeramani <balaji@anyscale.com>
bveeramani
requested review from
ericl,
scv119,
c21,
amogkam,
scottjlee,
raulchen,
stephanie-wang and
omatthew98
as code owners
bveeramani
commented
May 13, 2024
| @@ -185,22 +185,14 @@ def from_bytes(cls, data: bytes) -> "ArrowBlockAccessor": | |||
|
|
|||
| @staticmethod | |||
| def numpy_to_block( | |||
| batch: Union[np.ndarray, Dict[str, np.ndarray], Dict[str, list]], | |||
There was a problem hiding this comment.
We don't directly pass ndarrays to this method anywhere in the codebase.
c21
approved these changes
May 13, 2024
This was referenced May 29, 2024
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Outdated suggestions cannot be applied.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.
Why are these changes needed?
If you try to return arbitrary objects and arrays from a UDF, Ray Data errors complaining that "per-column arrays must each be 1-dimensional." The issue is that when Ray Data falls back to pandas, it doesn't use the tensor extension type. This PR fixes the bug.
Related issue number
Fixes #45235
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.