Migrate away from deprecated PyTorch API #2400
Open
Conversation
This file contains 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
|
Thank you @yushangdi for this great fix. |
|
Yes, we will need some work to upgrade our torch support to 2.5, due to numpy conflict from latest executorch 0.4
Will come back to this PR as soon as we resolve that |
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 13, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) XLA: one test case in pytorch/xla, PR to remove: pytorch/xla#8398 3) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Reviewed By: tugsbayasgalan Differential Revision: D64056353
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 13, 2024
Summary: as title There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) XLA: one test case in pytorch/xla, PR to remove: pytorch/xla#8398 3) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Differential Revision: D64056353
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 16, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) XLA: one test case in pytorch/xla, PR to remove: pytorch/xla#8398 3) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Reviewed By: tugsbayasgalan Differential Revision: D64056353
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Dec 16, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) XLA: one test case in pytorch/xla, PR to remove: pytorch/xla#8398 3) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Reviewed By: tugsbayasgalan Differential Revision: D64056353 Pull Request resolved: #143224 Approved by: https://github.com/tugsbayasgalan
yushangdi
added a commit
to yushangdi/pytorch
that referenced
this pull request
Dec 17, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() There's no more call sites to `capture_pre_autograd_graph`. Update XLA pin to include pytorch/xla#8398 Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Differential Revision: D67354440
pytorchmergebot
pushed a commit
to pytorch/pytorch
that referenced
this pull request
Dec 18, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() Update XLA pin to include pytorch/xla#8398 There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, guarded by version guard, PR to remove: apple/coremltools#2400 2) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Differential Revision: D67354440 Pull Request resolved: #143426 Approved by: https://github.com/gmagogsfm
aditew01
pushed a commit
to aditew01/pytorch
that referenced
this pull request
Dec 18, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, PR to remove: apple/coremltools#2400 2) XLA: one test case in pytorch/xla, PR to remove: pytorch/xla#8398 3) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Reviewed By: tugsbayasgalan Differential Revision: D64056353 Pull Request resolved: pytorch#143224 Approved by: https://github.com/tugsbayasgalan
aditew01
pushed a commit
to aditew01/pytorch
that referenced
this pull request
Dec 18, 2024
Summary: Delete the following API: - capture_pre_autograd_graph() - capture_pre_autograd_graph_using_training_ir() - gm_using_training_ir() Update XLA pin to include pytorch/xla#8398 There's no more call sites to `capture_pre_autograd_graph`. Except 1) two test cases in coreml, guarded by version guard, PR to remove: apple/coremltools#2400 2) a few call sites guarded by version guard (< 2.5.0) Test Plan: CI Differential Revision: D67354440 Pull Request resolved: pytorch#143426 Approved by: https://github.com/gmagogsfm
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Hi CoreML group,
The
capture_pre_autograd_graphAPI is deprecated and will be deleted soon. Please use theexport_for_trainingAPI instead. Theexport_for_trainingAPI is available since PyTorch 2.5.0.Thanks!
cc @shoumikhin