correct overlapping checkpoint evaluation
GeoIntel release gates / Compile, test, contracts and builds (push) Canceled after 0s
GeoIntel release gates / Python and npm vulnerability policy (push) Canceled after 0s
GeoIntel release gates / GIS image, SBOM and container scan (push) Canceled after 0s

This commit is contained in:
Jens
2026-08-09 20:41:54 +02:00
parent 2de438b9cc
commit 48084799c9
11 changed files with 539 additions and 10 deletions
@@ -187,3 +187,32 @@ visible; it was not suppressed or relabelled as success.
This min-4px version is the preferred experimental successor to the legacy
min-3px corpus. It explicitly remains ineligible for governed training until
real human review and all release-contract evidence exist.
## Non-overlapping checkpoint re-evaluation
The 36-tile checkpoint matrix was traced to a 512 px validation corpus with
stride 256. It contained nine views each of Turnhout, Westerlo, Postel-bos and
Arendonk-heide. Visual inspection then exposed that all Arendonk-heide views
are blank/no-data imagery, not meaningful pure-background observations. The
historical claim of 18 background images is therefore corrected: nine were
blank no-data and nine represented Postel under overlap.
A training-disabled validation view now covers each 1024 px AOI with four
non-overlapping 512 px tiles. Four blank Arendonk representatives are excluded
with explicit reason codes. The resulting set contains eight positive tiles
from Turnhout/Westerlo, four real Postel background tiles and 2,551 labels.
Exact reconstruction finds 2,384 unique interior objects, 167 edge labels and
zero repeated interior objects. The view has empty train directories, a
checksum-bound `NO_TRAINING.json`, and the training wrapper rejects that marker.
On the Tower RTX 4080 SUPER, the active checkpoint measures precision
`0.542632`, recall `0.435317`, mAP50 `0.342034` and mAP50-95 `0.141316`.
The reviewedexp6 challenger measures `0.581715`, `0.462300`, `0.368364` and
`0.155318` respectively. At background confidence 0.15 the active checkpoint
has zero Postel detections and the challenger has one. At 0.25 both have zero.
Thus the refined evidence confirms the challenger's relative metric advantage
and its threshold sensitivity, but still does not authorize promotion.
Non-overlap is not overstated as statistical independence: the positive tiles
remain adjacent and come from only two AOIs, edge objects can remain split, and
there is only one real background AOI. The production model remains unchanged.