Fix PT2 compliant opcheck tests #3404
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
|
This pull request was exported from Phabricator. Differential Revision: D66346179 |
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
spcyppt
force-pushed
the
export-D66346179
branch
from
8b73378 to
a882610
Compare
spcyppt
added a commit
to spcyppt/FBGEMM
that referenced
this pull request
Nov 22, 2024
Summary: X-link: facebookresearch/FBGEMM#492 Add decorator to skip PT2 compliant opcheck tests. ``` AssertionError: op 'fbgemm::split_embedding_codegen_lookup_rowwise_adagrad_function' was tagged with torch.Tag.pt2_compliant_tag but it failed some of the generated opcheck tests (['BackwardAdagradGlobalWeightDecay.test_faketensor__test_backward_adagrad_global_weight_decay', ``` This is due to faketensor test being added to the failure_dict. Faketensor tests fail because it expect all tensors to be on the same device, whereas we set learning_rate_tensor to be on CPU to avoid D->H sync as it will be converted to float for the kernel. Reviewed By: q10 Differential Revision: D66346179
|
This pull request was exported from Phabricator. Differential Revision: D66346179 |
spcyppt
added a commit
to spcyppt/FBGEMM
that referenced
this pull request
Nov 22, 2024
Summary: X-link: facebookresearch/FBGEMM#492 Add decorator to skip PT2 compliant opcheck tests. ``` AssertionError: op 'fbgemm::split_embedding_codegen_lookup_rowwise_adagrad_function' was tagged with torch.Tag.pt2_compliant_tag but it failed some of the generated opcheck tests (['BackwardAdagradGlobalWeightDecay.test_faketensor__test_backward_adagrad_global_weight_decay', ``` This is due to faketensor test being added to the failure_dict. Faketensor tests fail because it expect all tensors to be on the same device, whereas we set learning_rate_tensor to be on CPU to avoid D->H sync as it will be converted to float for the kernel. Reviewed By: q10 Differential Revision: D66346179
Summary: X-link: facebookresearch/FBGEMM#492 Add decorator to skip PT2 compliant opcheck tests. ``` AssertionError: op 'fbgemm::split_embedding_codegen_lookup_rowwise_adagrad_function' was tagged with torch.Tag.pt2_compliant_tag but it failed some of the generated opcheck tests (['BackwardAdagradGlobalWeightDecay.test_faketensor__test_backward_adagrad_global_weight_decay', ``` This is due to faketensor test being added to the failure_dict. Faketensor tests fail because it expect all tensors to be on the same device, whereas we set learning_rate_tensor to be on CPU to avoid D->H sync as it will be converted to float for the kernel. Reviewed By: q10 Differential Revision: D66346179
spcyppt
force-pushed
the
export-D66346179
branch
from
a882610 to
6a3a99e
Compare
|
This pull request was exported from Phabricator. Differential Revision: D66346179 |
|
This pull request has been merged in ed15cca. |
q10
pushed a commit
to q10/FBGEMM
that referenced
this pull request
Apr 10, 2025
Summary: Pull Request resolved: facebookresearch/FBGEMM#492 X-link: pytorch#3404 Add decorator to skip PT2 compliant opcheck tests. ``` AssertionError: op 'fbgemm::split_embedding_codegen_lookup_rowwise_adagrad_function' was tagged with torch.Tag.pt2_compliant_tag but it failed some of the generated opcheck tests (['BackwardAdagradGlobalWeightDecay.test_faketensor__test_backward_adagrad_global_weight_decay', ``` This is due to faketensor test being added to the failure_dict. Faketensor tests fail because it expect all tensors to be on the same device, whereas we set learning_rate_tensor to be on CPU to avoid D->H sync as it will be converted to float for the kernel. Reviewed By: q10 Differential Revision: D66346179 fbshipit-source-id: 8b0784071a9845a33ba3a75ab8a486d727a92672
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.
Summary:
Add decorator to skip PT2 compliant opcheck tests.
This is due to faketensor test being added to the failure_dict.
Faketensor tests fail because it expect all tensors to be on the same device, whereas we set learning_rate_tensor to be on CPU to avoid D->H sync as it will be converted to float for the kernel.
Reviewed By: q10
Differential Revision: D66346179