Replace custom op pad with aten op, post-export #4603
Merged
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/4603
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ab7e2f3 with merge base ce7f5a0 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Note: add custom op to disallow list so it doesn't get converted into edge-op before being replaced. Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
2df8a86 to
82fc55e
Compare
angelayi
approved these changes
Aug 9, 2024
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Note: add custom op to disallow list so it doesn't get converted into edge-op before being replaced. Reviewed By: angelayi Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
82fc55e to
9d7e740
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Note: add custom op to disallow list so it doesn't get converted into edge-op before being replaced. Reviewed By: angelayi Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
9d7e740 to
e6cc839
Compare
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Note: add custom op to disallow list so it doesn't get converted into edge-op before being replaced. Reviewed By: angelayi Differential Revision: D60941693
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
force-pushed
the
export-D60941693
branch
from
e6cc839 to
fc023fa
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
force-pushed
the
export-D60941693
branch
from
fc023fa to
640ec37
Compare
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Note: add custom op to disallow list so it doesn't get converted into edge-op before being replaced. Reviewed By: angelayi Differential Revision: D60941693
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
640ec37 to
927c328
Compare
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Reviewed By: larryliu0820 Differential Revision: D60941693
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
force-pushed
the
export-D60941693
branch
from
927c328 to
3101db2
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 9, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Reviewed By: larryliu0820 Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
3101db2 to
9488e8a
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
added a commit
to lucylq/executorch-1
that referenced
this pull request
Aug 10, 2024
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Reviewed By: larryliu0820 Differential Revision: D60941693
lucylq
force-pushed
the
export-D60941693
branch
from
9488e8a to
b9af02c
Compare
Summary: Pull Request resolved: pytorch#4603 In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727). Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad. Reviewed By: larryliu0820 Differential Revision: D60941693
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D60941693 |
lucylq
force-pushed
the
export-D60941693
branch
from
b9af02c to
ab7e2f3
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
Summary:
In the preprocess nn.Module, we use a custom op for pad. The aten pad cannot export due to dynamism (require these changes D60687727).
Because the custom pad and aten pad perform the same function, we can replace the custom op with the aten op post-export and avoid writing a custom C++ kernel for pad.
Differential Revision: D60941693