Add support for .to() for NestedTensor backends #87146

ani300 · 2022-10-17T22:25:42Z

Summary: This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140.

Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job.

Differential Revision: D40437585

pytorch-bot · 2022-10-17T22:25:45Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/87146

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit our office hours

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures, 1 Pending

As of commit 54edfa1:
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

facebook-github-bot · 2022-10-17T22:26:16Z

This pull request was exported from Phabricator. Differential Revision: D40437585

facebook-github-bot · 2022-10-17T22:37:07Z

This pull request was exported from Phabricator. Differential Revision: D40437585

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 fbshipit-source-id: 6f42b1dffa53d0945c19a92b0e087ea6abc978f0

aten/src/ATen/native/nested/NestedTensorMath.cpp

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 fbshipit-source-id: c15c94cbddad67224f3d133fa0bd4b653a66fb1e

facebook-github-bot · 2022-10-17T23:55:09Z

This pull request was exported from Phabricator. Differential Revision: D40437585

facebook-github-bot · 2022-10-18T13:56:16Z

This pull request was exported from Phabricator. Differential Revision: D40437585

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 fbshipit-source-id: 827ab5a96ce91a2b7b442607bc039d1a95e6c0b1

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 fbshipit-source-id: d59e0d83e04304033289a0fda099d157588c9e7d

facebook-github-bot · 2022-10-18T14:55:13Z

This pull request was exported from Phabricator. Differential Revision: D40437585

aten/src/ATen/native/nested/NestedTensorFactories.cpp

aten/src/ATen/native/nested/NestedTensorMath.cpp

aten/src/ATen/native/nested/NestedTensorMath.h

test/test_nestedtensor.py

drisspg

Few small comments but otherwise I think this is very close, great job

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 fbshipit-source-id: b54894943c605a619d900bc9cabebf5c0925e37b

facebook-github-bot · 2022-10-19T15:17:11Z

This pull request was exported from Phabricator. Differential Revision: D40437585

drisspg · 2022-10-19T19:29:09Z

aten/src/ATen/native/nested/NestedTensorMath.cpp

@@ -1,5 +1,7 @@
 #include <ATen/native/nested/NestedTensorMath.h>

+#include <c10/util/Optional.h>


Nit: do we need to add these headers? maybe we should have had the optional.h , but do we need the Copy.h header

drisspg

👍

Summary: Pull Request resolved: pytorch#87146 This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Reviewed By: drisspg Differential Revision: D40437585 fbshipit-source-id: b2b70ba4852c3b4ba4ae5d9ce796a37a902fb80a

facebook-github-bot · 2022-10-19T23:54:41Z

This pull request was exported from Phabricator. Differential Revision: D40437585

ani300 · 2022-10-19T23:58:38Z

@pytorchbot merge

pytorchmergebot · 2022-10-20T00:00:03Z

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

github-actions · 2022-10-20T03:47:23Z

Hey @ani300.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

Summary: This commit adds support for moving NestedTensors from CPU to GPU and back. The implementation includes requires implementing empty_like(), which is based on PR#83140. Test Plan: Added a new unit test based on the unit test for the main .to() implementation. All unit tests must pass, as well as every sandcastle job. Differential Revision: D40437585 Pull Request resolved: pytorch#87146 Approved by: https://github.com/drisspg

facebook-github-bot added the fb-exported label Oct 17, 2022

ani300 mentioned this pull request Oct 17, 2022

Add tensor.to()/tensor._to_copy() support to nested_tensor #87034

Closed

ani300 requested review from drisspg, cpuhrsch and mikekgfb October 17, 2022 22:28

ani300 force-pushed the export-D40437585 branch from 57066d5 to 357f62b Compare October 17, 2022 22:37

drisspg reviewed Oct 17, 2022

View reviewed changes

aten/src/ATen/native/nested/NestedTensorMath.cpp Outdated Show resolved Hide resolved

ani300 force-pushed the export-D40437585 branch from 357f62b to 64bba96 Compare October 17, 2022 23:55

ani300 force-pushed the export-D40437585 branch from 64bba96 to 1357cdd Compare October 18, 2022 13:56

ani300 force-pushed the export-D40437585 branch from 1357cdd to 238f326 Compare October 18, 2022 14:55

ani300 requested a review from drisspg October 18, 2022 20:24