-
Notifications
You must be signed in to change notification settings - Fork 26.9k
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
VLM Generate: tag test_static_cache_matches_dynamic
as flaky
#33630
VLM Generate: tag test_static_cache_matches_dynamic
as flaky
#33630
Conversation
@@ -4747,7 +4748,7 @@ def test_static_cache_matches_dynamic(self): | |||
output_logits=True, | |||
return_dict_in_generate=True, | |||
) | |||
self.assertTrue(torch.allclose(dynamic_out.logits[0], static_out.logits[0], rtol=1e-3, atol=1e-3)) | |||
self.assertTrue(torch.allclose(dynamic_out.logits[0], static_out.logits[0], rtol=1e-3, atol=1e-4)) |
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.
restores the original precision
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
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.
Thanks for flagging it, will definitely check
(cc @LysandreJik -- can we fix this band-aid for the time being? It is a common failure in our CI :) ) |
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.
yes sure!
What does this PR do?
test_static_cache_matches_dynamic
is VERY flaky on some VLMs. Tags it as flaky to avoid breaking CI while it is being investigated (cc @zucchini-nlp)This test was enabled recently on VLMs on #33533