Add method variants in dynamic shim to use core tensor/evalue/result types #7834
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/7834
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 095e826 with merge base 6332238 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Feb 23, 2025
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: digantdesai Differential Revision: D67650320
GregoryComer
force-pushed
the
export-D67650320
branch
from
44e688f to
1d1da4c
Compare
GregoryComer
requested review from
dbort,
JacobSzwejbka and
lucylq
as code owners
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
kirklandsign
approved these changes
Feb 24, 2025
GregoryComer
force-pushed
the
export-D67650320
branch
from
1d1da4c to
cddeb9c
Compare
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Mar 26, 2025
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: digantdesai Differential Revision: D67650320
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Mar 26, 2025
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: digantdesai Differential Revision: D67650320
GregoryComer
force-pushed
the
export-D67650320
branch
from
cddeb9c to
1a2cd42
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
|
@GregoryComer has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
GregoryComer
force-pushed
the
export-D67650320
branch
from
1a2cd42 to
25ae583
Compare
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Mar 26, 2025
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: metascroy Differential Revision: D67650320 Pulled By: GregoryComer
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
GregoryComer
added a commit
to GregoryComer/executorch
that referenced
this pull request
Mar 26, 2025
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: metascroy Differential Revision: D67650320 Pulled By: GregoryComer
GregoryComer
force-pushed
the
export-D67650320
branch
from
25ae583 to
b6033c9
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
…types (pytorch#7834) Summary: This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this: 1) Wrapper types are limited in functionality compared to the core types. 2) We need to add more functionality to the wrappers as we support more models, leading to duplicated effort. 3) API surface is inconsistent with ET documentation, which uses the core types. This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead. Reviewed By: metascroy Differential Revision: D67650320 Pulled By: GregoryComer
GregoryComer
force-pushed
the
export-D67650320
branch
from
b6033c9 to
095e826
Compare
|
This pull request was exported from Phabricator. Differential Revision: D67650320 |
kirklandsign
pushed a commit
that referenced
this pull request
Apr 11, 2025
…types Differential Revision: D67650320 Pull Request resolved: #7834
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:
This change adds forward and execute methods to the dynamic shim interface that make use of the core executorch tensor, evalue, and result types. The intent is to allow users to use the "real" ET types, rather than TensorWrapper/EValueWrapper. There are several reasons to do this:
This change should not introduce significant binary size changes, as these new methods are not currently used and are thus optimized out in release builds. When updating call sites to use the new methods and core ET types, impact is <= 5kB (see diffs in this stack). The ET core runtime is designed to be small, so directly including these types should not carry significant size overhead.
Differential Revision: D67650320