Skip to content

Latest commit

 

History

History
56 lines (41 loc) · 2.97 KB

File metadata and controls

56 lines (41 loc) · 2.97 KB

Production-scale validation

This case study records a private six-class run without publishing images, model weights, local paths or dataset names. It shows that the review gate was exercised on production-scale evidence rather than only on the public synthetic fixture.

Anonymized production validation summary

Run profile

Item Result
Dataset images 29,071
Specialist Teachers 6
Image-model inference passes 174,426
Raw prediction evidence rows 99,696
Review rows 30,183
Render failures 0
Source labels modified No
Human decisions completed 0 at snapshot time

The scan used CUDA FP16 inference at 832 pixels with batch=32, one Teacher resident on GPU at a time and all six classes evaluated over train, validation and test splits.

Decision routing

Terminal group Rows Share of review queue Handling
Train missing candidates 18,120 60.03% Human add/reject decision
Evaluation missing candidates 7,951 26.34% Held from training writes by default
Same-target ambiguity 3,688 12.22% Replace or reject; never append a second box
Cross-class conflict 424 1.41% Human semantic decision

Before the review queue was created, the engine suppressed 60,563 already-labelled predictions, 2,050 model duplicates and 6,900 below-review predictions. These 69,513 rows remain in the audit trail but do not consume reviewer time.

Workload distribution

Class Review rows Share
person 25,066 83.05%
helmet 1,823 6.04%
smoking 1,265 4.19%
vest 1,031 3.42%
slipper 886 2.94%
tractor 112 0.37%

The person Teacher dominates the queue. This is useful operational evidence: a technically correct exhaustive queue can still be economically poor. The next review cycle should allocate a class-aware budget or calibrate the person policy from audited decisions instead of silently raising one global threshold.

Qualitative findings

  • Helmet, vest and slipper samples contain visually credible missing annotations.
  • High-confidence smoking evidence includes hard negatives such as microphones near the mouth. This demonstrates why confidence is evidence strength, not write authority.
  • Same-target extent disagreements are common enough (3,688 rows) that IoU-only append logic would create harmful duplicate GT boxes.
  • Validation and test candidates account for 7,951 rows. Holding them by default prevents remediation from silently moving the evaluation target.
  • The package rendered every review image successfully, so it can be handed to a non-GPU annotation team as a portable review artifact.

Reproducibility boundary

Only aggregate, anonymized evidence is published. The private run keeps its original manifest, policies, raw CSV streams and reviewer package locally. No private image, label, weight, hostname or filesystem path is committed to this repository.