[BUG] - Depracate the usage of Nested Tensors
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 9.3k
- Forks
- 4.4k
- Avg merge
- 1d 21h
- Merged PRs (30d)
- 4
Description
Add Link
intermediate_source/transformer_building_blocks.py
unstable_source/nestedtensor.py
https://github.com/pytorch/pytorch/issues/171814
Describe the bug
intermediate_source/transformer_building_blocks.py
We need to depracate the usage of NestedTensors in all tutorials and replace with varlen attention.
We put a warning to not use them but we should provide what should be used instead.
Describe your environment
- PyTorch 2.10
cc @albanD @jbschlosser
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Inspect intermediate_source/transformer_building_blocks.py and unstable_source/nestedtensor.py, then review PyTorch issue #171814 and the warning in PR #3726 for the intended varlen attention replacement. Update the tutorials so they no longer use NestedTensors and verify the affected examples remain consistent with the guidance.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- documentation
- Issue type
- Documentation
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100