Skip to content

[Model][Gemma3] Simplify image input validation #18710

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 2 commits into from
May 27, 2025
Merged

[Model][Gemma3] Simplify image input validation #18710

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f930c68
[Model][Gemma3] Simplify image input validation
lgeiger May 26, 2025
3d0acfd
Address review comments
lgeiger May 26, 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
18 changes: 6 additions & 12 deletions vllm/model_executor/models/gemma3_mm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -504,18 +504,12 @@ def dtype(self):
return next(self.parameters()).dtype

def _validate_pixel_values(self, data: torch.Tensor) -> torch.Tensor:
h = w = self.config.vision_config.image_size
expected_dims = (3, h, w)

def _validate_shape(d: torch.Tensor):
if d.shape != expected_dims:
raise ValueError(
"The expected shape of pixel values per image per batch "
f"is {expected_dims}. You supplied {tuple(d.shape)}.")

for d in data:
_validate_shape(d)

image_size = self.config.vision_config.image_size
expected_dims = (3, image_size, image_size)
if data.shape[1:] != expected_dims:
raise ValueError(
"The expected shape of pixel values per image per batch is "
f"{expected_dims}. You supplied {tuple(data.shape)}.")
return data

def _parse_and_validate_image_input(
Expand Down