Skip to content

Make validate_mask faster #84

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
lukas-lightly merged 1 commit into from
Jun 1, 2026
Merged

Make validate_mask faster #84

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
12 changes: 10 additions & 2 deletions src/labelformat/formats/semantic_segmentation/pascalvoc.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -377,8 +377,16 @@ class IDs not present in `valid_class_ids`.
f"mask (W,H)=({mw},{mh}) vs image (W,H)=({image_obj.width},{image_obj.height})"
)

uniques = np.unique(mask_np.astype(int))
unique_values = set(uniques)
# np.bincount is much faster than np.unique for finding unique non-negative integer classes.
# We fall back to np.unique if there are negative values or extremely large values
# (to prevent huge memory allocations in np.bincount).
if np.any(mask_np < 0) or mask_np.max() > 65535:
uniques = np.unique(mask_np)
unique_values = set(uniques)
else:
uniques = np.flatnonzero(np.bincount(mask_np.ravel()))
unique_values = set(uniques)

unknown_values = unique_values.difference(valid_class_ids)
if unknown_values:
raise ValueError(
Expand Down
Loading