[MPS] Add group_norm[fwd+backward] and mean_var (take 2)
#91190
Closed
Conversation
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
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to register python implementation in mps backend TODO: Fix MPS var implementation and add it to prims Fixes #88331 [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/91190
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit f7ce741: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Contributor
Author
|
I wonder if it'll be better done using |
malfet
changed the title
mean_vargroup_norm[fwd+backward] and mean_var
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to register python implementation in mps backend TODO: Fix MPS var implementation and add it to prims Fixes #88331 [ghstack-poisoned]
malfet
added a commit
that referenced
this pull request
Dec 20, 2022
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to register python implementation in mps backend TODO: Fix MPS var implementation and add it to prims Fixes #88331 ghstack-source-id: 1437e0e Pull Request resolved: #91190
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to register python implementation in mps backend TODO: Fix MPS var implementation and add it to prims Fixes #88331 [ghstack-poisoned]
malfet
added
ciflow/mps
albanD
reviewed
Dec 21, 2022
Collaborator
albanD
left a comment
albanD
left a comment
There was a problem hiding this comment.
Can you explain in more details what is the difference between torch.ops and torch._ops.ops ?
As far as I can tell they are exactly the same thing.
The new registration looks good.
albanD
approved these changes
Dec 21, 2022
Contributor
Author
|
@pytorchbot merge -f "MPS tests are passing" |
Collaborator
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Contributor
|
@pytorchbot revert -m "Broke test_correct_module_names because of underscore _ops" -c weird |
Collaborator
|
@pytorchbot successfully started a revert job. Check the current status here. |
Collaborator
|
@malfet your PR has been successfully reverted. |
pytorchmergebot
added a commit
that referenced
this pull request
Dec 21, 2022
This reverts commit 371716e. Reverted #91190 on behalf of https://github.com/kit1980 due to Broke test_correct_module_names because of underscore _ops
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to make them importable from `torch/backend/mps/__init__.py` as this alias is defined in https://github.com/pytorch/pytorch/blob/15af4b1ceea3b689354ad664675b930486804c8e/torch/__init__.py#L1095 is executed last during init process. Depends on #91203 Fixes #88331 [ghstack-poisoned]
malfet
changed the title
group_norm[fwd+backward] and mean_vargroup_norm[fwd+backward] and mean_var (take 2)
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to make them importable from `torch/backend/mps/__init__.py` as this alias is defined in https://github.com/pytorch/pytorch/blob/15af4b1ceea3b689354ad664675b930486804c8e/torch/__init__.py#L1095 is executed last during init process. Depends on #91203 Fixes #88331 [ghstack-poisoned]
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to make them importable from `torch/backend/mps/__init__.py` as this alias is defined in https://github.com/pytorch/pytorch/blob/15af4b1ceea3b689354ad664675b930486804c8e/torch/__init__.py#L1095 is executed last during init process. Add `__all__` to `torch/backends/mps/__init__.py` as well as alias all imports as private Add `TestNNMPS.test_group_norm_backward` that validates no NaNs are generated during the backward pass Fixes #88331 [ghstack-poisoned]
malfet
added a commit
that referenced
this pull request
Dec 22, 2022
Use Prims to implement group_norm, group_norm_backward and mean_var Use `torch._ops.ops` instead of `torch.ops` in numerous subpackages in order to be able to register python implementation in mps backend TODO: Fix MPS var implementation and add it to prims Fixes #88331 ghstack-source-id: 43325e0 Pull Request resolved: #91190
Contributor
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour 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 |
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.
Stack from ghstack (oldest at bottom):
group_norm[fwd+backward]andmean_var(take 2) #91190Use Prims to implement group_norm, group_norm_backward and mean_var
Use
torch._ops.opsinstead oftorch.opsin numerous subpackages inorder to be able to make them importable from
torch/backend/mps/__init__.pyas this alias is defined inpytorch/torch/__init__.py
Line 1095 in 15af4b1
is executed last during init process.
Add
__all__totorch/backends/mps/__init__.pyas well as alias all imports as privateAdd
TestNNMPS.test_group_norm_backwardthat validates no NaNs are generated during the backward passFixes #88331