-
Notifications
You must be signed in to change notification settings - Fork 354
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Clean up lowering APIs #3690
Clean up lowering APIs #3690
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/3690
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 453dd5e with merge base 705ac96 (): This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This pull request was exported from Phabricator. Differential Revision: D57579738 |
expo_program, compile_config=EdgeCompileConfig(_check_ir_validity=False) | ||
expo_program, | ||
compile_config=EdgeCompileConfig( | ||
_check_ir_validity=False, _skip_dim_order=True |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's this added _skip_dim_order
kwarg for?
inputs: Tuple[Any, ...], | ||
dump_graphs: bool = False, | ||
) -> EdgeProgramManager: | ||
edge_program_manager = export_to_edge(model, inputs) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should this just take in an EdgeProgramManager that separately had export_to_edge
called so that the APIs can be combined?
to_edge_ops_count = get_ops_count(to_edge_gm) | ||
jarvis_ops_count = get_ops_count(jarvis_gm) | ||
|
||
# De-duplicate the "<op>" and "<op>_copy" ops |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why remove this part?
Summary: Changes some types in `export_to_edge`, add an `export_to_cadence` function, and refactor some calls. Mostly preemptive work before adding a lot of AoT passes. Reviewed By: dulinriley, zonglinpengmeta Differential Revision: D57579738
09e215d
to
453dd5e
Compare
This pull request was exported from Phabricator. Differential Revision: D57579738 |
This pull request has been merged in ed2da4f. |
Summary: Pull Request resolved: pytorch#3690 Changes some types in `export_to_edge`, add an `export_to_cadence` function, and refactor some calls. Mostly preemptive work before adding a lot of AoT passes. Reviewed By: dulinriley, zonglinpengmeta Differential Revision: D57579738 fbshipit-source-id: 8c7975488636bd946894f7f7e62da7e76eb29c65
Summary: Changes some types in
export_to_edge
, add anexport_to_cadence
function, and refactor some calls. Mostly preemptive work before adding a lot of AoT passes.Differential Revision: D57579738