Skip to content

[tests] Fix failing float16 cuda tests #11835

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
merged 4 commits into from
Jul 1, 2025
Merged

[tests] Fix failing float16 cuda tests #11835

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d9c4f1e
update
a-r-r-o-w Jun 30, 2025
b1f2fd1
update
a-r-r-o-w Jun 30, 2025
e88ae2f
Merge branch 'main' into tests/fix-failing-float16-cuda
a-r-r-o-w Jun 30, 2025
2a8e0fe
Merge branch 'main' into tests/fix-failing-float16-cuda
sayakpaul Jul 1, 2025
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
10 changes: 6 additions & 4 deletions tests/pipelines/test_pipelines_common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1378,25 +1378,27 @@ def test_float16_inference(self, expected_max_diff=5e-2):
for component in pipe_fp16.components.values():
if hasattr(component, "set_default_attn_processor"):
component.set_default_attn_processor()

pipe_fp16.to(torch_device, torch.float16)
pipe_fp16.set_progress_bar_config(disable=None)

inputs = self.get_dummy_inputs(torch_device)
# Reset generator in case it is used inside dummy inputs
if "generator" in inputs:
inputs["generator"] = self.get_generator(0)

output = pipe(**inputs)[0]

fp16_inputs = self.get_dummy_inputs(torch_device)
# Reset generator in case it is used inside dummy inputs
if "generator" in fp16_inputs:
fp16_inputs["generator"] = self.get_generator(0)

output_fp16 = pipe_fp16(**fp16_inputs)[0]

if isinstance(output, torch.Tensor):
output = output.cpu()
output_fp16 = output_fp16.cpu()

max_diff = numpy_cosine_similarity_distance(output.flatten(), output_fp16.flatten())
assert max_diff < 1e-2
assert max_diff < expected_max_diff

@unittest.skipIf(torch_device not in ["cuda", "xpu"], reason="float16 requires CUDA or XPU")
@require_accelerator
Expand Down