You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The SplitToSequence refactor added a special case for element_size when it's a string, but this stops the computation of the A offset on line 510 from advancing as the offset is zero. So it copies the same string into each sequence element. This PR removes that check and adds a test for the string split behaviour.
I'm not sure why the special case on line 456 was added in the original PR (#18594), but all the tests pass in the new PR so presumably it isn't necessary.
Motivation and Context
SplitToSequence didn't work on string tensors, fixes#19726.
/azp run Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline,Linux QNN CI Pipeline,MacOS CI Pipeline,ONNX Runtime Web CI Pipeline,Windows ARM64 QNN CI Pipeline
/azp run Windows CPU CI Pipeline,Windows GPU CI Pipeline,Windows GPU TensorRT CI Pipeline,Windows x64 QNN CI Pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,onnxruntime-binary-size-checks-ci-pipeline
The Windows GPU TensorRT failure says that it couldn't find a CUDA device in tests other than the one I added, and they don't appear at first glance to use SplitToSequence either. Maybe it hit some dodgy hardware?
/azp run ONNX Runtime Web CI Pipeline,orttraining-amd-gpu-ci-pipeline,Big Models,Android CI Pipeline,iOS CI Pipeline,ONNX Runtime React Native CI Pipeline
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
The SplitToSequence refactor added a special case for element_size when it's a string, but this stops the computation of the
Aoffset on line 510 from advancing as the offset is zero. So it copies the same string into each sequence element. This PR removes that check and adds a test for the string split behaviour.I'm not sure why the special case on line 456 was added in the original PR (#18594), but all the tests pass in the new PR so presumably it isn't necessary.
Motivation and Context
SplitToSequence didn't work on string tensors, fixes #19726.
cc @pranavsharma