53 Commits

Include renames

Author	SHA1	Message	Date
JensandClaude Opus 5	dd87a62e8f	report what an area selection actually measured ... Four ways a selection produced a confident number about a different area than the operator drew: Flood hazard divided the inundated cells by every cell in the drawn rectangle, including cells the VMM raster does not model at all. A selection reaching past the modelled extent therefore reported a diluted risk share, turning missing data into an implied absence of risk. Terrain, bathymetry and thematic raster already divided by valid cells; flood hazard was the outlier. It now reports the three populations separately, states model coverage next to the drawn area, and returns a null fraction rather than a zero when nothing was modelled. geometry_mask selects a cell when its centre falls inside the geometry, so a rectangle smaller than one cell — or one landing between four centres — selected nothing and the analysis returned zeros indistinguishable on screen from "we looked and there is nothing here". On a 100 m population raster a 40 m rectangle over a city block reported no inhabitants. Selection now falls back to the touched cells and says that it did, since the answer then covers more ground than was requested. rasterio.mask applies the same centre rule when cropping, so that call is widened too; the cells that count are still decided by the centre rule wherever it selects anything. The object count treated any feature touching the selection as whole, while intersection_area clipped it — two headline numbers on one panel describing different populations. The count stays whole-feature, which is what "objecten" means to an operator, but now reports how many the edge cuts and is marked an estimate when it does. The area_weighted_sum branch reuses that same count instead of issuing its own near-identical query. Partitioned selection de-duplicated the count on source_feature_id but returned the raw rows, so a building on a municipal boundary was counted once and drawn twice. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-08-22 14:33:19 +02:00
JensandClaude Opus 5	2b968b74cf	make QA scoring reproducible and threshold-independent ... The greedy IoU matcher gave a reference to whichever candidate was offered first. Row order decided that, and every detection in a run shares one transaction timestamp, so ordering by created_at left the assignment undefined: the same QA run over the same data produced different mean IoU, and the geometry shown to a reviewer as a false positive could be the better of two detections. Candidates are now ranked by confidence with feature identity as tiebreaker, which is also the COCO/PASCAL rule. A single precision/recall/F1 triple describes one operating point, so two models cannot be compared from it: a conservatively calibrated model looks worse at a low confidence cut and better at a high one without detecting anything differently. DetectionMetricsService adds the full curve, average precision and the threshold where F1 actually peaks. Also: - report the population the metrics were computed over, so matches + false_positives equals candidate_feature_count even under an area filter; raw dataset totals move to the _raw fields; - state whether candidates are axis-aligned boxes or footprint polygons. A box can never reach IoU 1 against a rotated building, so the strict score has a ceiling that has nothing to do with detection quality. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>	2026-08-22 14:31:21 +02:00
Jens	0209167cfd	fix(results): gate operational AI exports on QA	2026-08-09 11:31:51 +02:00
Jens	5b3c17b494	feat(provenance): govern source snapshots and data inputs	2026-08-01 23:46:17 +02:00
Jens	c76a746cd7	Update	2026-07-27 23:28:43 +02:00
Jens	9f05451f04	Preserve native orthophoto detail for training	2026-07-27 01:20:10 +02:00
Jens	ec6f2d9061	Add governed Wallonia and Brussels orthophotos	2026-07-26 22:02:10 +02:00
Jens	2be72fac58	feat: add governed nationwide AOI orchestration and CUDA enforcement	2026-07-26 05:23:33 +02:00
Jens	25b6f1ab39	feat: make spatial entry optional and authoritative	2026-07-26 03:39:50 +02:00
Codex	115f9850a7	feat: add operator landing and login	2026-07-22 20:10:21 +02:00
Codex	cee6cd05ae	feat: complete Wallonia land cover and terrain sources	2026-07-22 06:28:46 +02:00
Codex	c2101ea8f9	feat: complete governed Walloon coverage sources	2026-07-22 03:39:08 +02:00
Codex	0aff8e3b8c	feat(scope): make Belgium and North Sea operational default	2026-07-22 02:11:48 +02:00
Codex	20829f1a24	fix(map): make area analysis scale-aware	2026-07-21 21:59:56 +02:00
Codex	70c5e34ecf	Add governed SPW bathymetry analysis	2026-07-19 04:07:01 +02:00
Codex	50897c3473	Federate official Belgium data sources	2026-07-19 01:34:39 +02:00
Codex	0fae53a7de	Harden RC7 API response contracts	2026-07-18 05:07:35 +02:00
Codex	1e527bf810	Add Belgium and North Sea coverage foundation	2026-07-18 01:45:00 +02:00
Codex	9c402e0df2	Start autonomous Belgium and North Sea RC	2026-07-18 00:12:34 +02:00
Codex	c787fb2184	feat: operationalize Flemish land and nature themes	2026-07-17 22:38:25 +02:00
Codex	488da4cc83	Add bounded GRB map acquisition	2026-07-17 21:29:26 +02:00
Codex	47cde21e17	Add bounded terrain and flood analysis	2026-07-17 20:36:45 +02:00
Codex	4fd6c06f5c	Aggregate regional bathymetry partitions	2026-07-17 17:08:35 +02:00
Codex	132c4feed8	Add Flemish bathymetry partition workflow	2026-07-17 16:15:29 +02:00
Codex	5b4e18059a	Add governed bathymetry profile workflow	2026-07-17 15:25:00 +02:00
Codex	78a5f0b0b2	Close full operational audit findings	2026-07-17 14:19:02 +02:00
Codex	a29c787238	Complete operational map result workflow	2026-07-17 12:35:50 +02:00
Codex	da2371d662	Add governed Statbel edition probe	2026-07-16 22:14:48 +02:00
Codex	0a23e3484c	Add governed ALZ edition probe	2026-07-16 20:27:07 +02:00
Codex	eed6ee796e	Add governed GRB refresh workflow	2026-07-16 17:59:00 +02:00
Codex	595967f892	Add official source edition probes	2026-07-16 17:33:40 +02:00
Codex	1e0e7b6499	Add governed source freshness audit	2026-07-16 15:05:42 +02:00
Codex	45dc730e76	Complete regional raster exploration	2026-07-16 14:21:32 +02:00
Codex	035ec3b233	feat: add cross-domain Mol data profile	2026-07-16 03:06:36 +02:00
Codex	501f824257	Add governed VMM flood hazard scenarios	2026-07-15 22:14:12 +02:00
Codex	5d8b46ed60	Add governed DHMV terrain analysis	2026-07-15 18:13:05 +02:00
Codex	fb38eb3e91	feat: add governed hydrology and historical imagery	2026-07-15 12:17:45 +02:00
Codex	beacdf2560	feat: add source-grounded evolution and Ollama assistant	2026-07-15 06:45:56 +02:00
Codex	0baa9b069c	feat: add semantic GIS selection metrics	2026-07-15 06:07:34 +02:00
Codex	d22abe8e7b	feat: add measured detection review loop	2026-07-15 03:00:08 +02:00
Codex	daccd3869a	feat: add map-driven orthophoto analysis	2026-07-15 02:01:03 +02:00
Codex	616424e2a2	fix: query persisted work area geometry	2026-07-14 19:38:41 +02:00
Codex	c0943fe7d4	feat: add temporal Mol explorer	2026-07-14 15:19:42 +02:00
Codex	2958c25b6f	feat: make Mol explorer map-first	2026-07-14 13:21:53 +02:00
Codex	1689dce928	Promote focused small-building detector	2026-07-13 12:15:27 +02:00
Codex	6e2a8cbdc4	Add local model asset catalog	2026-07-06 20:59:03 +02:00
Codex	368c1ad738	Persist QA feature evidence	2026-06-25 02:57:50 +02:00
Codex	b92faafa74	Persist map selections as derived datasets	2026-06-25 02:25:49 +02:00
Codex	46fe3edfc5	Persist map area selection exports	2026-06-25 02:12:09 +02:00
Codex	851d7220df	Add map area selection extraction	2026-06-25 01:59:57 +02:00