Skip to content

Fix channel-first indices buffer for distance_transform_edt (return_indices=True) #8656 #8657

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

Open
alexanderjaus wants to merge 3 commits into
base: dev
Choose a base branch
from
Open

Fix channel-first indices buffer for distance_transform_edt (return_indices=True) #8656 #8657

Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2d5eb24
Fix EDT indices allocation for channel-first tensors on CPU/GPU
alexanderjaus Dec 11, 2025
258a472
Apply codestyle fixes
alexanderjaus Dec 11, 2025
838e347
DCO Remediation Commit for alexanderjaus <[email protected]>
alexanderjaus Dec 13, 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
4 changes: 2 additions & 2 deletions monai/transforms/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2498,7 +2498,7 @@ def distance_transform_edt(
if return_indices:
dtype = torch.int32
if indices is None:
indices = torch.zeros((img.dim(),) + img.shape, dtype=dtype) # type: ignore
indices = torch.zeros((img.shape[0],) + (img.dim() - 1,) + img.shape[1:], dtype=dtype) # type: ignore
else:
if not isinstance(indices, torch.Tensor) and indices.device != img.device:
raise TypeError("indices must be a torch.Tensor on the same device as img")
Expand Down Expand Up @@ -2532,7 +2532,7 @@ def distance_transform_edt(
raise TypeError("distances must be a numpy.ndarray of dtype float64")
if return_indices:
if indices is None:
indices = np.zeros((img_.ndim,) + img_.shape, dtype=np.int32)
indices = np.zeros((img_.shape[0],) + (img_.ndim - 1,) + img_.shape[1:], dtype=np.int32)
else:
if not isinstance(indices, np.ndarray):
raise TypeError("indices must be a numpy.ndarray")
Expand Down
Loading