Visitar URL original
Skip tracklet stitching when no animals were detected by C-Achard · Pull Request #3504 · DeepLabCut/DeepLabCut · GitHub
Skip to content

Skip tracklet stitching when no animals were detected #3504

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

Draft
C-Achard wants to merge 21 commits into
base: dev
Choose a base branch
from
Draft

Skip tracklet stitching when no animals were detected #3504

Changes from 1 commit
Commits
Show all changes
21 commits
Select commit Hold shift + click to select a range
64a2f4e
fix _generate_output_data crash on empty predictions
deruyter92 Sep 8, 2026
510090e
Allow create_df_from_prediction() to handle an empty predictions list.
deruyter92 Sep 8, 2026
4b45fb9
Treat zero predictions as a distinct warning case in `video_inference()`
deruyter92 Sep 8, 2026
c78f439
Don't write dataframe for empty predictions
deruyter92 Sep 9, 2026
16a2977
update test_videos: pop `"PAFinds"` before comparison
deruyter92 Sep 9, 2026
f534de6
update test_videos: assert no DF is written for empty predictions
deruyter92 Sep 9, 2026
624b1b1
Return empty-detection inference results
C-Achard Sep 9, 2026
c57dd35
Use CTD predictions for video output
C-Achard Sep 9, 2026
4e4dab7
Handle empty prediction outputs gracefully
C-Achard Sep 9, 2026
42098ac
Reuse shared frame key width helper
C-Achard Sep 9, 2026
292d068
Clarify video inference warnings
C-Achard Sep 9, 2026
d5ba290
Go back to ValueError for empty predictions
C-Achard Sep 9, 2026
bff4fea
Remove NaN fallback for empty video predictions
C-Achard Sep 9, 2026
06486de
Fix CTD export guard for empty predictions
C-Achard Sep 9, 2026
e456aeb
Clarify empty CTD prediction handling
C-Achard Sep 9, 2026
a162439
Centralize test fakes
C-Achard Sep 9, 2026
11ff4fd
Add test for async with no images
C-Achard Sep 9, 2026
a210b69
Clarify async inference test comment
C-Achard Sep 9, 2026
c7a8c34
Use shared frame key width helper
C-Achard Sep 9, 2026
8c6311e
Skip empty tracklet stitching
C-Achard Sep 9, 2026
5c5c566
Add empty-tracklet regression tests
C-Achard Sep 9, 2026
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
Prev Previous commit
Next Next commit
Clarify async inference test comment 
Refines the test docstring in `test_runners_inference.py` to describe the async behavior more clearly by comparing it to the sequential path without implying a regression pin.
  • Loading branch information
@C-Achard
C-Achard committed Sep 9, 2026
commit a210b69e8e7f5ae800689178eafe8b7fa5c51020
2 changes: 1 addition & 1 deletion tests/pose_estimation_pytorch/runners/test_runners_inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,7 +172,7 @@ def test_async_inference_emits_images_without_detections(detections_per_image):
queue item. The async consumer therefore only emits it if results are drained
once more after the producer signals completion; without that final drain the
trailing undetected images are dropped, and with no detections at all the whole
result set comes back empty (issue #3485). Pinned against the sequential path,
result set comes back empty (issue #3485). Compared to the sequential path,
which has always handled this because it extracts after every image.
"""
images = _top_down_images(detections_per_image, h=8, w=8)
Expand Down
Loading