-
Notifications
You must be signed in to change notification settings - Fork 28.5k
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
[DEV Testing] Issues with test_modeling_common
#35857
Comments
Hello @muellerzr . Do you already have a branch that you can share so we can run and check 🙏 ? |
@ydshieh kinda. It's my own testing branch but basically add that test to |
OK, taking a look! |
I think it's because
If restricting only to If you want to test for all kind of |
@ydshieh that seems to somewhat work, but then leads me to: ______________________________________________ FlavaImageModelTest.test_training_gradient_accumulation _______________________________________________
self = <tests.models.flava.test_modeling_flava.FlavaImageModelTest testMethod=test_training_gradient_accumulation>
def test_training_gradient_accumulation(self):
> if not self.model_tester.is_training:
E AttributeError: 'FlavaImageModelTester' object has no attribute 'is_training' |
Which yes, I could just do: if not getattr(self.model_tester, "is_training", False):
self.skipTest(reason="ModelTester is not configured to run training tests") But wanted to point out another difference in existing code |
Yeah, thanks. Not sure how
is passing, maybe all tests that use |
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
I was writing some new tests to help us fix GA down in the modeling code and propagate the fix manually, since everything is breaking constantly with each fix and taking the dive.
When writing a test for this, which should be limiting to all ForCausalLM models, I was finding some strange errors on models that realistically shouldn't be trying to run these tests.
One of the general failures:
Said test:
List of tests:
cc @ydshieh @ArthurZucker
The text was updated successfully, but these errors were encountered: