[compiled autograd] support Tensor Subclasses in AOTBackward#144115
Closed
zou3519 wants to merge 15 commits intogh/zou3519/1112/basefrom
Closed
[compiled autograd] support Tensor Subclasses in AOTBackward#144115zou3519 wants to merge 15 commits intogh/zou3519/1112/basefrom
zou3519 wants to merge 15 commits intogh/zou3519/1112/basefrom
Conversation
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. We also change a DTensor constructor call to be DTensor.from_local. This is because Dynamo cannot handle raw DTensor constructor calls but is able to handle DTensor.from_local. Test Plan: - More compiled autograd x Tensor subclass tests pass now. The other failures need investigation but a lot of them are because Dynamo isn't able to trace DTensor internals that are now exposed (due to autograd.Function). - Existing tests [ghstack-poisoned]
This was referenced Jan 3, 2025
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/144115
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit b4dd186 with merge base 54e2f4b ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
zou3519
added a commit
that referenced
this pull request
Jan 3, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. We also change a DTensor constructor call to be DTensor.from_local. This is because Dynamo cannot handle raw DTensor constructor calls but is able to handle DTensor.from_local. Test Plan: - More compiled autograd x Tensor subclass tests pass now. The other failures need investigation but a lot of them are because Dynamo isn't able to trace DTensor internals that are now exposed (due to autograd.Function). - Existing tests ghstack-source-id: 3ab2945 Pull Request resolved: #144115
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. We also change a DTensor constructor call to be DTensor.from_local. This is because Dynamo cannot handle raw DTensor constructor calls but is able to handle DTensor.from_local. Test Plan: - More compiled autograd x Tensor subclass tests pass now. The other failures need investigation but a lot of them are because Dynamo isn't able to trace DTensor internals that are now exposed (due to autograd.Function). - Existing tests cc H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov xmfan [ghstack-poisoned]
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. NB: I tried to run this on the DTensor x compiled autograd tests, but it didn't improve anything. The main problem there was that the DTensor internal code now needs to be Dynamo-able and it is not (there are things like DTensor constructor calls and attribute accesses that are not yet Dynamo-able). Test Plan: - New basic test with TwoTensor - Existing tests [ghstack-poisoned]
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests [ghstack-poisoned]
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Jan 3, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests ghstack-source-id: 7660405 Pull Request resolved: #144115
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Jan 3, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests ghstack-source-id: 7a14d87 Pull Request resolved: #144115
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Jan 9, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests ghstack-source-id: 7d128cf Pull Request resolved: #144115
zou3519
added a commit
that referenced
this pull request
Jan 23, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests Pull Request resolved: #144115 Approved by: https://github.com/jansel, https://github.com/xmfan, https://github.com/bdhirsh ghstack dependencies: #143296, #143304, #143387, #143405, #143417
zou3519
added a commit
that referenced
this pull request
Jan 24, 2025
…144115)" This reverts commit 082c28c. Reverted #144115 on behalf of https://github.com/izaitsevfb due to breaking internal tests T213390054 ([comment](#143296 (comment))) [ghstack-poisoned]
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests cc H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov xmfan [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Jan 24, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests Pull Request resolved: #144115 Approved by: https://github.com/jansel, https://github.com/xmfan, https://github.com/bdhirsh ghstack dependencies: #143296, #143304, #143387, #143405, #143417 ghstack-source-id: b79847c
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests cc H-Huang awgu kwen2501 wanchaol fegin fduwjj wz337 wconstab d4l3k c-p-i-o voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov xmfan [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Jan 24, 2025
Compiled autograd's initial trace traces through the AOTBackward epilogue. The Tensor Subclass code is not traceable. This PR changes it so that when we see Tensor Subclass constructors, we proxy nodes for their construction into the graph. Test Plan: - New basic test with TwoTensor - Existing tests Pull Request resolved: #144115 Approved by: https://github.com/jansel, https://github.com/xmfan, https://github.com/bdhirsh ghstack dependencies: #143296, #143304, #143387, #143405, #143417 ghstack-source-id: 8876f26
facebook-github-bot
pushed a commit
that referenced
this pull request
Jan 24, 2025
Summary: This PR squashes together the following commits: #144115 #143417 #143405 #143387 #143304 #143296 This is a refactor of compiled autograd to use "functional autograd". The end goal is that it gets compiled autograd's initial capture to stop specializing on Tensor metadata, therefore allowing compiled autograd to better handle Tensor subclasses. For more information, please read the commit messages for each PR. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov xmfan bypass-github-export-checks failing CI test is bypassable on OSS but cannot be retried. Reviewed By: jansel, xmfan, bdhirsh Differential Revision: D68120850 Pulled By: zou3519
zou3519
added a commit
that referenced
this pull request
Jan 26, 2025
Summary: This PR squashes together the following commits: #144115 #143417 #143405 #143387 #143304 #143296 This is a refactor of compiled autograd to use "functional autograd". The end goal is that it gets compiled autograd's initial capture to stop specializing on Tensor metadata, therefore allowing compiled autograd to better handle Tensor subclasses. For more information, please read the commit messages for each PR. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov xmfan bypass-github-export-checks failing CI test is bypassable on OSS but cannot be retried. Reviewed By: jansel, xmfan, bdhirsh Differential Revision: D68120850 Pulled By: zou3519
pytorchmergebot
pushed a commit
that referenced
this pull request
Jan 27, 2025
This PR squashes together the following commits: #144115 #143417 #143405 #143387 #143304 #143296 This is a refactor of compiled autograd to use "functional autograd". The end goal is that it gets compiled autograd's initial capture to stop specializing on Tensor metadata, therefore allowing compiled autograd to better handle Tensor subclasses. For more information, please read the commit messages for each PR. Pull Request resolved: #144707 Approved by: https://github.com/bdhirsh, https://github.com/xmfan, https://github.com/jansel
nWEIdia
pushed a commit
to nWEIdia/pytorch
that referenced
this pull request
Jan 27, 2025
This PR squashes together the following commits: pytorch#144115 pytorch#143417 pytorch#143405 pytorch#143387 pytorch#143304 pytorch#143296 This is a refactor of compiled autograd to use "functional autograd". The end goal is that it gets compiled autograd's initial capture to stop specializing on Tensor metadata, therefore allowing compiled autograd to better handle Tensor subclasses. For more information, please read the commit messages for each PR. Pull Request resolved: pytorch#144707 Approved by: https://github.com/bdhirsh, https://github.com/xmfan, https://github.com/jansel
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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.
Stack from ghstack (oldest at bottom):
Compiled autograd's initial trace traces through the AOTBackward
epilogue. The Tensor Subclass code is not traceable. This PR changes it
so that when we see Tensor Subclass constructors, we proxy nodes for
their construction into the graph.
Test Plan:
cc @H-Huang @awgu @kwen2501 @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @c-p-i-o @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang @aakhundov @xmfan