Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support NestedTensor for XPU device #1141

Open
7 of 9 tasks
min-jean-cho opened this issue Dec 3, 2024 · 0 comments
Open
7 of 9 tasks

Support NestedTensor for XPU device #1141

min-jean-cho opened this issue Dec 3, 2024 · 0 comments
Assignees
@daisyden @fengyuan14 @xytintel @min-jean-cho

Comments

@min-jean-cho
Copy link
Contributor

min-jean-cho commented Dec 3, 2024 •
edited
Loading

🚀 The feature, motivation and pitch

Support NestedTensor for xpu device

Example

>>> nt = torch.nested.nested_tensor([]).to("xpu")
>>> nt
nested_tensor([

], device='xpu:0')
>>> nt.is_xpu
True

Alternatives

No response

Additional context

No response
@min-jean-cho min-jean-cho mentioned this issue Dec 10, 2024
Merged
This was referenced Jan 17, 2025
Closed
Merged
Merged
Merged
Merged
Closed
pytorchmergebot pushed a commit to pytorch/pytorch that referenced this issue Jan 24, 2025
@min-jean-cho @pytorchmergebot
[Intel GPU] Add TORCH_API macro to export symbol NestedTensor_to_mask…
7c314bf 
… for libtorch_xpu (#145467)

Part of intel/torch-xpu-ops#1141.

The `TORCH_API` macro is added to export the symbol `NestedTensor_to_mask`, which is needed by libtroch_xpu for `NestedTensor_softmax_dropout_xpu`.

Pull Request resolved: #145467
Approved by: https://github.com/guangyey, https://github.com/ezyang
github-merge-queue bot pushed a commit that referenced this issue Jan 26, 2025
@min-jean-cho
Add aten::_nested_tensor_softmax_with_shape (#1323)
b6786e3 
Part of #1141.

Depends on pytorch/pytorch#145467.

- `_nested_tensor_softmax_with_shape`
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@daisyden daisyden

@fengyuan14 fengyuan14

@xytintel xytintel

@min-jean-cho min-jean-cho

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
4 participants
@daisyden @fengyuan14 @xytintel @min-jean-cho