Files
geointel/docs
JensandClaude Opus 5 f6eced1b94 rank model variants on average precision, and say when they are not comparable
The workbench ranks model variants by a stored F1, each measured at that
variant's own confidence threshold. A conservatively calibrated detector then
looks worse than a liberal one without detecting anything differently: the
number says as much about the threshold as about the model. POST
/detection/runs/compare ranks on average precision instead, which describes the
whole ranking a model produced, and keeps each run's own-threshold F1 visible
next to it so the difference between the two readings is auditable.

Comparability comes before the 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 no metric makes them so. The report names which of those applies and still
returns the numbers — they are simply not a ranking.

Each run is scored through the same QA path the workbench uses, so a comparison
and the persisted quality checks cannot drift apart.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-22 19:47:14 +02:00
..
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-07-18 08:33:40 +02:00
2026-06-16 23:36:32 +02:00
2026-06-16 23:36:32 +02:00
2026-08-21 22:35:17 +02:00
2026-06-16 23:36:32 +02:00

GeoIntel Docs Index

Start with these active release controls:

  • 00-start/START_HERE.md
  • RC_SCOPE_FREEZE_BELGIUM_NORTH_SEA.md
  • RC_ROADMAP_BELGIUM_NORTH_SEA.md
  • RELEASE_RUNBOOK.md
  • KNOWN_LIMITATIONS.md
  • DEFINITION_OF_DONE.md

The milestone and sprint documents below are retained as historical design and implementation evidence. TODO.md is an implementation archive, not the active release board.

Product foundation

  • PRODUCT_VISION.md
  • PRODUCT_BLUEPRINT.md
  • SPECIFICATION_FREEZE_M0.md
  • V1_SCOPE_FREEZE.md
  • M0_HANDOFF_SUMMARY.md

Data and processing

  • DATASET_STRATEGY.md
  • DATA_CATALOG.md
  • DATA_SPECIFICATION.md
  • DATA_SOURCES.md
  • RASTER_OPERATIONS_SPEC.md
  • VECTOR_OPERATIONS_SPEC.md
  • DETECTION_PIPELINE_SPEC.md
  • SEGMENTATION_PIPELINE_SPEC.md
  • CHANGE_DETECTION_SPEC.md
  • ANALYSIS_SPECIFICATIONS.md
  • QA_QC_SPECIFICATION.md
  • FIXTURE_STRATEGY.md

Architecture

  • ARCHITECTURE.md
  • SERVICE_ARCHITECTURE.md
  • DATABASE_SCHEMA.md
  • API_SPECIFICATION.md
  • STORAGE_ARCHITECTURE.md
  • REPOSITORY_CONVENTIONS.md

UI and modules

  • MODULES.md
  • UI_UX_SPEC.md
  • UI_PAGE_SPECIFICATIONS.md
  • COMPONENT_BREAKDOWN.md

Historical execution

  • CODEX_MASTER_PROMPT.md
  • CODEX_EXECUTION_PLAN.md
  • CODEX_BUILD_PLAN.md
  • CODEX_EXECUTION_LOG.md
  • ACCEPTANCE_CRITERIA.md
  • TODO.md (historical implementation archive)
  • IMPLEMENTATION_BACKLOG.md
  • DEVELOPMENT_RULES.md
  • AGENTS.md