Jens/geointel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

serve a promoted model at the post-processing it was gated on

Browse Source 
scripts/evaluate_belgium_building_candidate.py freezes its post-processing
before the protected test — NMS IoU and a containment threshold selected during
calibration, defaulting to 1.0. The runtime applied a hardcoded 0.85, so a
promoted candidate was served under suppression its evaluation never measured
and dropped detections the gate had counted. Neither report showed the
difference. That constant was mine, added without noticing the evaluation
pipeline already had a tuned value for the same concept.

Containment is now configuration, recorded on every run beside the duplicate
IoU threshold, so an operator can serve a candidate at the value it was gated
at and afterwards see which value produced a given score.

Two runs that post-processed differently produced different candidate sets from
the same model output, so the comparison endpoint refuses to rank them. Runs
recorded before those values were persisted carry none, and absence is not
treated as a difference.

The remaining gap is deliberate and documented rather than closed: the gate
scores the model on its protected test set, the workbench scores the whole
pipeline including coverage clipping and the tile-edge filter. A promoted
candidate will not reproduce its gate F1 exactly, and pretending otherwise
would be the worse answer.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Jens
2026-08-22 21:15:05 +02:00
co-authored by Claude Opus 5
parent 1a1a9af6e7
commit c8d32a4801
8 changed files with 207 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/API_CONTRACTS.md
+5 -2
View File
@@ -1891,8 +1891,11 @@ two readings is auditable rather than hidden.
Comparability is reported before any ranking. Runs over different source
rasters, scored against different references, without a proven inference
footprint, or covering a different evaluated population are not alternatives to
one another, and `comparability.blocking_reasons` names which of those applies.
footprint, covering a different evaluated population, or post-processed with a
different containment / duplicate-IoU threshold are not alternatives to one
another, and `comparability.blocking_reasons` names which of those applies.
Runs recorded before those thresholds were persisted carry none of them, and
absence is not treated as a difference.
The numbers are still returned — they are simply not a ranking.
Each run is scored through the same QA path the workbench uses, so a comparison