4 Commits

Include renames

Author	SHA1	Message	Date
NuklearRabbit	7fbb42c9ef	test: align source contracts with modular workbench	2026-08-29 01:52:08 +02:00
JensandClaude Opus 5	2cd2c49389	calibrate a confidence threshold from one inference pass ... Threshold calibration ran the model over every tile once per threshold — three GPU passes to compare 0.50, 0.25 and 0.15 on a hundred-tile raster. The answer is already in a single run at the lowest value: detections above a higher cut are a subset of it, and duplicate suppression walks candidates in descending confidence, so a lower-confidence box can never displace a higher-confidence one. The kept set above any cut is identical whichever threshold the run used, which is what makes one pass sufficient rather than merely cheaper. QA now takes calibration_thresholds and reads each operating point off the same precision/recall walk it already performs, marking the F1-optimal cut. The lab runs inference once and fills its table from the sweep. The contract test asserted the per-threshold loop by name, pinning the waste it was meant to describe. It now states what calibration owes an operator: a row per requested threshold, from one run. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-08-22 19:37:19 +02:00
Codex	78a5f0b0b2	Close full operational audit findings	2026-07-17 14:19:02 +02:00
Codex	c0bebd609b	Add guided detection calibration runner	2026-07-08 13:11:02 +02:00