Skip to content

VLM Generate: tag test_static_cache_matches_dynamic as flaky #33630

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.

Sign up for GitHub

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

Merged
gante merged 1 commit into from
Oct 3, 2024
Merged

VLM Generate: tag test_static_cache_matches_dynamic as flaky #33630

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion tests/test_modeling_common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4711,6 +4711,7 @@ def test_custom_4d_attention_mask(self):
normalized_1 = F.softmax(out_shared_prefix_last_tokens)
torch.testing.assert_close(normalized_0, normalized_1, rtol=1e-3, atol=1e-4)

@is_flaky(max_attempts=10) # TODO @raushan: this test is VERY flaky on some VLMs, like paligemma
def test_static_cache_matches_dynamic(self):
"""
Tests that generating with static cache give almost same results as with dynamic cache.
Expand Down Expand Up @@ -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))
Copy link
Contributor Author

@gante gante Sep 20, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

restores the original precision


# For now, Let's focus only on GPU for `torch.compile`
@slow
Expand Down