-
Notifications
You must be signed in to change notification settings - Fork 492
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
Refactor gather_test_model to be hard coded directly #9227
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/9227
Note: Links to docs will display an error until the docs builds have been completed. ⏳ 26 Pending, 1 Unrelated FailureAs of commit af97aff with merge base 753da9a ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
b339b38
to
ef8d5d4
Compare
ef8d5d4
to
d57d66d
Compare
d57d66d
to
177794d
Compare
831a115
to
0a6fcd5
Compare
0a6fcd5
to
af97aff
Compare
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.
Are we able to specify specific runner per model using the matrix? We might need something like this to move the linux trunk test off gather_test_models.py e.g.
include:
- model: efficient_sam
backend: portable
runner: linux.4x.large
Yeah, you can use include and exclude logic directly in the matrix The example you provided works. |
One thing that I realized I had a bug. The include statements should define all the variables. |
Fixing #9227, It wasn't running efficient_sam, llama etc. Because I had to define all variables in the include statement. Since build-tool and runner are singletons, I just hard code it. Test Plan: Make sure trunk jobs have softmax, efficient_sam etc.
Fixing #9227 matrix.timeout is not populated. Just hard-code the value now. Also fix another cleanup DEMO_BACKEND_ID is not there anymore. Test Plan: Make sure there are 6 jobs (test-models-linux) in the pull.yml category.
Looks like I lied when I said #9227 was a no-op Adding back pull jobs for linux x86
The CI runner logic should directly be in the corresponding yml files.
Gather_test_models.py contains overly complicated python logic on what models, backends, runners to do, with many switch statements.
I think we should just hard code directly in job descriptions to simplify things and know exactly what jobs are running directly in the yml file.
This is a no-op PR