Add real data detection QA workflow smoke
GeoIntel CI / docs-smoke (push) Has been cancelled
GeoIntel CI / contract-smoke (push) Has been cancelled

This commit is contained in:
Codex
2026-07-07 00:19:23 +02:00
parent e30e7c4f34
commit f0a58011fe
9 changed files with 641 additions and 0 deletions
+8
View File
@@ -7,6 +7,14 @@
# Changelog
## Sprint 121 Real data detection and QA workflow smoke (2026-07-07)
- Added `scripts/verify_real_data_detection_qa_workflow.sh` for operator-provided GeoTIFF/reference-vector validation against a live runtime.
- The smoke uploads a real raster source dataset and real reference building vector, validates GIS metadata, tiles the raster, selects a mounted local model asset, runs configured YOLO detection, runs persisted detection QA/QC and exports detection GeoJSON.
- Registered the script in the readiness gate as a syntax check so normal development remains green without real local imagery or model files.
- Documented exact Tower usage in `scripts/README.md`, `backend/README.md`, `docs/AI_PIPELINES.md` and `docs/TODO.md`.
- The script refuses missing files, unsupported formats, demo workflow seeding, fixture detections, live provider fetching and model downloads.
## Sprint 120 Model asset detection workflow smoke (2026-07-06)
- Added `scripts/verify_model_asset_detection_workflow.sh` for live Docker/Tower validation of the configured-YOLO path with a selected local model asset.
+18
View File
@@ -352,6 +352,24 @@ Detection GeoJSON output. It does not download weights or inject detector
fixtures. A zero detection result is still a valid runtime smoke outcome on the
synthetic demo raster.
To validate the configured building model on operator-provided GIS data, mount
or copy a real georeferenced raster and a real reference-building GeoJSON onto
the runtime host, then run:
```bash
REAL_RASTER_PATH=/mnt/user/appdata/geointel/data/orthophoto.tif \
REAL_REFERENCE_VECTOR_PATH=/mnt/user/appdata/geointel/data/reference-buildings.geojson \
bash scripts/verify_real_data_detection_qa_workflow.sh http://192.168.10.150:1202
```
This smoke refuses missing/unsupported inputs, uploads the raster and reference
dataset through the normal dataset service, generates raster tiles, selects a
local model asset, runs configured YOLO detection, compares persisted
detections against persisted `vector_features`, persists QA/QC rows and exports
the detection GeoJSON. It never seeds demo detections, enables fixture mode,
fetches live providers or downloads model weights. Current V1 upload support is
limited to GeoTIFF-style rasters and GeoJSON/JSON reference vectors.
### Run backend
```bash
@@ -0,0 +1,33 @@
from pathlib import Path
ROOT = Path(__file__).resolve().parents[2]
def test_real_data_detection_qa_smoke_requires_operator_inputs_and_checks_full_chain() -> None:
script_path = ROOT / "scripts" / "verify_real_data_detection_qa_workflow.sh"
readiness = (ROOT / "scripts" / "run_readiness_check.sh").read_text(encoding="utf-8")
assert script_path.exists()
script = script_path.read_text(encoding="utf-8")
assert "bash -n scripts/verify_real_data_detection_qa_workflow.sh" in readiness
assert "REAL_RASTER_PATH" in script
assert "REAL_REFERENCE_VECTOR_PATH" in script
assert "usage()" in script
assert "/api/v1/projects" in script
assert "/datasets/upload" in script
assert "dataset_role=reference" in script
assert "reference_layer_name=buildings" in script
assert "/raster/inspect" in script
assert "/raster/tile" in script
assert "/api/v1/detection/model-assets" in script
assert "/api/v1/detection/yolo/preflight" in script
assert "/api/v1/detection/run" in script
assert "/qa/reference" in script
assert "/api/v1/exports/geojson" in script
assert "Response is not a canonical GeoIntel data envelope" in script
assert "No local model assets are available" in script
assert "demo/workflow" not in script
assert "fixture_mode" not in script
assert "Fixture detections" not in script
+28
View File
@@ -137,6 +137,34 @@ detector fixtures or download weights. A zero detection count is acceptable on
the synthetic demo raster; production usefulness still requires validation on
real georeferenced orthophotos and reference vectors.
### Real-data detection and QA validation
The real operational validation path uses operator-provided files rather than
demo fixtures:
```bash
REAL_RASTER_PATH=/mnt/user/appdata/geointel/data/orthophoto.tif \
REAL_REFERENCE_VECTOR_PATH=/mnt/user/appdata/geointel/data/reference-buildings.geojson \
bash scripts/verify_real_data_detection_qa_workflow.sh http://192.168.10.150:1202
```
The script verifies the full persisted chain:
- source raster upload with CRS and bounds metadata;
- reference building vector upload as `dataset_role=reference`;
- raster inspect and tile manifest generation;
- local model asset selection and read-only YOLO preflight;
- configured-YOLO detection run through Job, AnalysisRun and Detection rows;
- detection GeoJSON generated from persisted geometry;
- detection QA against persisted reference `vector_features` with persisted
`QualityCheck` and `Metric` rows;
- detection run GeoJSON export.
It refuses to run without a real GeoTIFF-style raster and GeoJSON/JSON reference
vector. It does not seed demo data, use `fixture_mode`, fetch live providers or
download model weights. A zero detection count is valid as runtime evidence but
does not prove the model is useful for the target imagery.
### Sprint 8C detection visualization and QA status
Sprint 8C makes persisted detections reviewable:
+27
View File
@@ -1,3 +1,30 @@
## Sprint 121 Real data detection and QA workflow smoke (2026-07-07)
Changed:
- Added `scripts/verify_real_data_detection_qa_workflow.sh` for live-runtime validation with operator-provided real GIS inputs.
- The smoke creates a project, uploads a real GeoTIFF-style raster as a source dataset, uploads a real reference-building GeoJSON/JSON as `dataset_role=reference`, validates CRS/bounds/features, tiles the raster, selects a mounted local model asset, runs read-only YOLO preflight, runs configured YOLO detection, runs detection QA against persisted `vector_features`, and exports the detection run GeoJSON.
- Registered the smoke in `scripts/run_readiness_check.sh` as a syntax check only, so normal readiness does not require real orthophotos, reference vectors, optional AI dependencies or model files.
- Documented Tower usage and limitations in `scripts/README.md`, `backend/README.md`, `docs/AI_PIPELINES.md`, `docs/TODO.md` and `CHANGELOG.md`.
Validation:
- RED: `python -m pytest backend/tests/test_sprint121_real_data_detection_qa_smoke.py -q` failed while `scripts/verify_real_data_detection_qa_workflow.sh` did not exist.
- `python -m pytest backend/tests/test_sprint121_real_data_detection_qa_smoke.py -q` passed: 1 test.
- `bash -n scripts/verify_real_data_detection_qa_workflow.sh` passed.
- `bash scripts/verify_real_data_detection_qa_workflow.sh --help` passed and printed required `REAL_RASTER_PATH` and `REAL_REFERENCE_VECTOR_PATH` usage.
- `python -m compileall backend/app` passed.
- `python scripts/smoke_docs.py` passed.
- `bash scripts/run_readiness_check.sh` passed: 385 backend tests, Alembic head check, frontend typecheck, frontend production build and shell syntax checks.
- `cd backend && python -m alembic upgrade head --sql` passed.
- Missing-input guard passed: `bash scripts/verify_real_data_detection_qa_workflow.sh http://localhost:1202` returned exit code 2 and printed usage.
- Local `docker compose config` could not run in this Windows Codex environment because the `docker` command is not installed.
Limitations:
- The full real-data smoke was not executed in this Codex workspace because no operator-provided real GeoTIFF and reference GeoJSON were found locally.
- The script enforces real inputs and never seeds demo data, enables fixture detections, fetches live GRB/OSM/Sentinel data or downloads model weights.
Next recommended pass:
- Place a target orthophoto/GeoTIFF and matching reference-building GeoJSON under the Tower appdata path and run `REAL_RASTER_PATH=... REAL_REFERENCE_VECTOR_PATH=... bash scripts/verify_real_data_detection_qa_workflow.sh http://192.168.10.150:1202`.
## Sprint 120 Model asset detection workflow smoke (2026-07-06)
Changed:
+1
View File
@@ -91,6 +91,7 @@ This file now starts with the current implementation status. Older preparation/b
- [x] Add QA/QC and Exports usability layout pass with calmer evidence review and handoff artifact scanning.
- [x] Add AI Labs Detection/Segmentation hierarchy and result density polish.
- [x] Add Export/System handoff hierarchy and provider registry density polish.
- [x] Add operator-provided real raster/reference detection + QA workflow smoke.
- [ ] Validate the configured building model on a real georeferenced Kempen orthophoto/GeoTIFF with persisted reference vectors and QA/QC metrics.
## Sprint 8 status
+21
View File
@@ -151,6 +151,27 @@ count is allowed because the demo raster is a synthetic runtime fixture; the
script validates the operational path and provenance, not production model
quality. The main readiness gate checks this script's syntax only.
Verify the full operator-provided raster/reference detection and QA path:
```bash
REAL_RASTER_PATH=/mnt/user/appdata/geointel/data/orthophoto.tif \
REAL_REFERENCE_VECTOR_PATH=/mnt/user/appdata/geointel/data/reference-buildings.geojson \
bash scripts/verify_real_data_detection_qa_workflow.sh http://192.168.10.150:1202
```
The real-data smoke is intentionally mutating and refuses to run without
operator-supplied files. Current V1 upload support expects a georeferenced
`.tif`, `.tiff` or `.geotiff` raster and a `.geojson` or `.json` reference
building vector. The script creates a project, uploads the raster as a source
dataset, uploads the vector as a `reference` dataset, validates raster/vector
metadata, tiles the raster, selects a mounted local model asset, verifies
read-only YOLO preflight, runs configured YOLO detection, runs detection QA
against persisted `vector_features`, and exports the detection run as GeoJSON.
It does not seed demo data, enable fixture detections, fetch external data or
download model weights. A zero detection count is accepted operationally, but
must be interpreted as model/data quality evidence rather than as a successful
building extraction result.
Docker images install only the GIS runtime by default. To build a local/Tower
image with PyTorch/Ultralytics available for the configured-YOLO preflight and
runtime path, set:
+1
View File
@@ -54,6 +54,7 @@ bash -n scripts/verify_demo_export_workflow.sh
bash -n scripts/verify_demo_raster_workflow.sh
bash -n scripts/verify_ai_handoff_interactions.sh
bash -n scripts/verify_model_asset_detection_workflow.sh
bash -n scripts/verify_real_data_detection_qa_workflow.sh
bash -n scripts/verify_workbench_default_state.sh
bash -n scripts/verify_workbench_interactions.sh
bash -n scripts/verify_gis_runtime.sh
@@ -0,0 +1,504 @@
#!/usr/bin/env bash
set -euo pipefail
usage() {
cat >&2 <<'EOF'
Usage:
REAL_RASTER_PATH=/path/to/orthophoto.tif \
REAL_REFERENCE_VECTOR_PATH=/path/to/reference-buildings.geojson \
bash scripts/verify_real_data_detection_qa_workflow.sh [base_url]
or:
bash scripts/verify_real_data_detection_qa_workflow.sh [base_url] /path/to/orthophoto.tif /path/to/reference-buildings.geojson
Required inputs:
REAL_RASTER_PATH Georeferenced .tif/.tiff/.geotiff raster.
REAL_REFERENCE_VECTOR_PATH EPSG-aware .geojson/.json reference vector with building polygons.
Optional environment:
REAL_PROJECT_NAME Project name for the validation run.
REAL_MODEL_ASSET_ID Specific /api/v1/detection/model-assets id to use.
REAL_TILE_SIZE Raster tile size, default 640.
REAL_TILE_OVERLAP Raster tile overlap, default 64.
REAL_CONFIDENCE_THRESHOLD Detection confidence threshold, default 0.5.
REAL_IOU_THRESHOLD QA IoU threshold, default 0.5.
EOF
}
BASE_URL="${1:-${GE_INTEL_BASE_URL:-http://localhost:1202}}"
REAL_RASTER_PATH="${2:-${REAL_RASTER_PATH:-}}"
REAL_REFERENCE_VECTOR_PATH="${3:-${REAL_REFERENCE_VECTOR_PATH:-}}"
REAL_PROJECT_NAME="${REAL_PROJECT_NAME:-GeoIntel Real Data Validation}"
REAL_TILE_SIZE="${REAL_TILE_SIZE:-640}"
REAL_TILE_OVERLAP="${REAL_TILE_OVERLAP:-64}"
REAL_CONFIDENCE_THRESHOLD="${REAL_CONFIDENCE_THRESHOLD:-0.5}"
REAL_IOU_THRESHOLD="${REAL_IOU_THRESHOLD:-0.5}"
REAL_MODEL_ASSET_ID="${REAL_MODEL_ASSET_ID:-}"
TMP_DIR="$(mktemp -d)"
trap 'rm -rf "${TMP_DIR}"' EXIT
if [ "${BASE_URL}" = "-h" ] || [ "${BASE_URL}" = "--help" ]; then
usage
exit 0
fi
if [ -z "${REAL_RASTER_PATH}" ] || [ -z "${REAL_REFERENCE_VECTOR_PATH}" ]; then
usage
exit 2
fi
if [ ! -f "${REAL_RASTER_PATH}" ]; then
echo "REAL_RASTER_PATH does not point to a readable file: ${REAL_RASTER_PATH}" >&2
exit 2
fi
if [ ! -f "${REAL_REFERENCE_VECTOR_PATH}" ]; then
echo "REAL_REFERENCE_VECTOR_PATH does not point to a readable file: ${REAL_REFERENCE_VECTOR_PATH}" >&2
exit 2
fi
case "${REAL_RASTER_PATH,,}" in
*.tif|*.tiff|*.geotiff) ;;
*)
echo "REAL_RASTER_PATH must be a .tif, .tiff or .geotiff file for the current V1 raster upload flow." >&2
exit 2
;;
esac
case "${REAL_REFERENCE_VECTOR_PATH,,}" in
*.geojson|*.json) ;;
*)
echo "REAL_REFERENCE_VECTOR_PATH must be .geojson or .json for the current V1 vector upload flow." >&2
exit 2
;;
esac
if ! command -v curl >/dev/null 2>&1; then
echo "curl is required for real data detection/QA workflow verification" >&2
exit 1
fi
if [ -n "${PYTHON_BIN:-}" ]; then
PYTHON_BIN="${PYTHON_BIN}"
else
PYTHON_BIN=""
for candidate in python3 python.exe python; do
if command -v "${candidate}" >/dev/null 2>&1 && "${candidate}" -c "import json, sys" >/dev/null 2>&1; then
PYTHON_BIN="${candidate}"
break
fi
done
fi
if [ -z "${PYTHON_BIN}" ]; then
echo "A Python interpreter is required for JSON parsing" >&2
exit 1
fi
json_field() {
local file_path="$1"
local expression="$2"
"${PYTHON_BIN}" - "$file_path" "$expression" <<'PY'
import json
import sys
path, expression = sys.argv[1], sys.argv[2]
with open(path, "r", encoding="utf-8") as handle:
payload = json.load(handle)
value = payload
for part in expression.split("."):
if part:
value = value[part]
print(value)
PY
}
require_json_data() {
local file_path="$1"
"${PYTHON_BIN}" - "$file_path" <<'PY'
import json
import sys
with open(sys.argv[1], "r", encoding="utf-8") as handle:
payload = json.load(handle)
if "data" not in payload:
raise SystemExit("Response is not a canonical GeoIntel data envelope")
PY
}
echo "== GeoIntel real data detection + QA workflow verification =="
echo "Base URL: ${BASE_URL}"
echo "Raster: ${REAL_RASTER_PATH}"
echo "Reference vector: ${REAL_REFERENCE_VECTOR_PATH}"
"${PYTHON_BIN}" - "${TMP_DIR}/project_request.json" "${REAL_PROJECT_NAME}" <<'PY'
import json
import sys
from datetime import datetime, timezone
path, project_name = sys.argv[1], sys.argv[2]
stamp = datetime.now(timezone.utc).strftime("%Y%m%dT%H%M%SZ")
payload = {
"name": f"{project_name} {stamp}",
"description": "Operator-provided real data validation: raster upload, reference vector upload, configured YOLO detection, QA/QC and GeoJSON export.",
"region": "Kempen",
}
with open(path, "w", encoding="utf-8") as handle:
json.dump(payload, handle)
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/projects" \
-H "Content-Type: application/json" \
--data-binary "@${TMP_DIR}/project_request.json" > "${TMP_DIR}/project.json"
require_json_data "${TMP_DIR}/project.json"
project_id="$(json_field "${TMP_DIR}/project.json" "data.id")"
if [ -z "${project_id}" ] || [ "${project_id}" = "None" ] || [ "${project_id}" = "null" ]; then
echo "Project creation did not return a project id" >&2
exit 1
fi
curl -fsS -X POST "${BASE_URL%/}/api/v1/projects/${project_id}/datasets/upload" \
-F "file=@${REAL_RASTER_PATH}" \
-F "dataset_type=raster" \
-F "source=user_upload" \
-F "dataset_role=source" \
-F "source_name=manual" \
-F 'source_metadata_json={"validation_workflow":"real_data_detection_qa","input_kind":"orthophoto"}' \
-F 'provenance_metadata_json={"operator_supplied":true,"no_external_fetch":true}' \
> "${TMP_DIR}/raster_upload.json"
require_json_data "${TMP_DIR}/raster_upload.json"
raster_dataset_id="$(json_field "${TMP_DIR}/raster_upload.json" "data.id")"
"${PYTHON_BIN}" - "${TMP_DIR}/raster_upload.json" <<'PY'
import json
import sys
with open(sys.argv[1], "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("dataset_type") != "raster":
raise SystemExit(f"Uploaded raster returned wrong dataset_type: {data.get('dataset_type')}")
if data.get("status") != "ready":
raise SystemExit(f"Uploaded raster is not ready: {data.get('status')}")
if not data.get("crs"):
raise SystemExit("Uploaded raster does not expose CRS metadata; refusing geospatial AI validation")
if not data.get("bounds_json"):
raise SystemExit("Uploaded raster does not expose bounds metadata; refusing geospatial AI validation")
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/projects/${project_id}/datasets/upload" \
-F "file=@${REAL_REFERENCE_VECTOR_PATH}" \
-F "dataset_type=vector" \
-F "source=user_upload" \
-F "dataset_role=reference" \
-F "source_name=manual" \
-F "reference_layer_name=buildings" \
-F 'source_metadata_json={"validation_workflow":"real_data_detection_qa","input_kind":"reference_buildings"}' \
-F 'provenance_metadata_json={"operator_supplied":true,"no_external_fetch":true}' \
> "${TMP_DIR}/reference_upload.json"
require_json_data "${TMP_DIR}/reference_upload.json"
reference_dataset_id="$(json_field "${TMP_DIR}/reference_upload.json" "data.id")"
"${PYTHON_BIN}" - "${TMP_DIR}/reference_upload.json" <<'PY'
import json
import sys
with open(sys.argv[1], "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("dataset_type") != "vector":
raise SystemExit(f"Uploaded reference returned wrong dataset_type: {data.get('dataset_type')}")
if data.get("dataset_role") != "reference":
raise SystemExit("Uploaded reference did not persist dataset_role=reference")
if data.get("reference_layer_name") != "buildings":
raise SystemExit("Uploaded reference did not persist reference_layer_name=buildings")
if data.get("status") != "ready":
raise SystemExit(f"Uploaded reference is not ready: {data.get('status')}")
if int(data.get("feature_count") or 0) < 1:
raise SystemExit("Uploaded reference has no persisted features; QA would be meaningless")
PY
curl -fsS "${BASE_URL%/}/api/v1/projects/${project_id}/datasets/${raster_dataset_id}/raster/inspect" > "${TMP_DIR}/raster_inspect.json"
require_json_data "${TMP_DIR}/raster_inspect.json"
"${PYTHON_BIN}" - "${TMP_DIR}/raster_inspect.json" "${raster_dataset_id}" <<'PY'
import json
import sys
path, raster_id = sys.argv[1], sys.argv[2]
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("dataset_id") != raster_id:
raise SystemExit("Raster inspect returned the wrong dataset id")
if data.get("ready") is not True:
raise SystemExit("Raster inspect did not report ready=true")
metadata = data.get("metadata") or {}
if not metadata.get("crs"):
raise SystemExit("Raster inspect metadata has no CRS")
if int(metadata.get("width") or 0) < 1 or int(metadata.get("height") or 0) < 1:
raise SystemExit("Raster inspect metadata has invalid dimensions")
PY
curl -fsS "${BASE_URL%/}/api/v1/projects/${project_id}/datasets/${reference_dataset_id}/vector/summary" > "${TMP_DIR}/reference_summary.json"
require_json_data "${TMP_DIR}/reference_summary.json"
"${PYTHON_BIN}" - "${TMP_DIR}/reference_summary.json" <<'PY'
import json
import sys
with open(sys.argv[1], "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if int(data.get("feature_count") or 0) < 1:
raise SystemExit("Reference vector summary reports no features")
geometry_types = set(data.get("geometry_types") or [])
if not geometry_types.intersection({"Polygon", "MultiPolygon"}):
raise SystemExit(f"Reference vector does not expose polygonal building geometries: {sorted(geometry_types)}")
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/projects/${project_id}/datasets/${raster_dataset_id}/raster/tile" \
-H "Content-Type: application/json" \
-d "{\"tile_size\":${REAL_TILE_SIZE},\"overlap\":${REAL_TILE_OVERLAP},\"output_name\":\"real_data_detection_tiles\"}" > "${TMP_DIR}/tile.json"
require_json_data "${TMP_DIR}/tile.json"
manifest_path="$(json_field "${TMP_DIR}/tile.json" "data.result_json.manifest_path")"
if [ -z "${manifest_path}" ] || [ "${manifest_path}" = "None" ] || [ "${manifest_path}" = "null" ]; then
echo "Raster tile response did not include a manifest_path" >&2
exit 1
fi
"${PYTHON_BIN}" - "${TMP_DIR}/tile.json" "${raster_dataset_id}" <<'PY'
import json
import sys
path, raster_id = sys.argv[1], sys.argv[2]
with open(path, "r", encoding="utf-8") as handle:
job = json.load(handle)["data"]
if job.get("job_type") != "raster.tile":
raise SystemExit(f"Unexpected tile job type: {job.get('job_type')}")
if job.get("status") not in {"success", "completed"}:
raise SystemExit(f"Raster tile job did not complete: {job.get('status')}")
result = job.get("result_json") or {}
if result.get("dataset_id") != raster_id:
raise SystemExit("Raster tile result returned the wrong dataset id")
if result.get("ready") is not True:
raise SystemExit("Raster tile result did not report ready=true")
manifest = result.get("manifest") or {}
tiles = manifest.get("tiles") or []
if not tiles:
raise SystemExit("Raster tile manifest has no tiles")
if manifest.get("source_dataset_id") != raster_id:
raise SystemExit("Raster tile manifest source_dataset_id drifted")
PY
curl -fsS "${BASE_URL%/}/api/v1/detection/model-assets" > "${TMP_DIR}/model_assets.json"
require_json_data "${TMP_DIR}/model_assets.json"
model_asset_id="$("${PYTHON_BIN}" - "${TMP_DIR}/model_assets.json" "${REAL_MODEL_ASSET_ID}" <<'PY'
import json
import sys
path, requested = sys.argv[1], sys.argv[2]
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
items = data.get("items") or []
if not items:
raise SystemExit("No local model assets are available. Mount a local .pt/.onnx/.engine file before running this smoke.")
for item in items:
if item.get("will_download_models") is not False:
raise SystemExit("Model asset catalog must never report automatic model downloads")
if requested:
matches = [item for item in items if item.get("model_asset_id") == requested]
if not matches:
raise SystemExit(f"Requested REAL_MODEL_ASSET_ID was not found: {requested}")
selected = matches[0]
else:
selected = next((item for item in items if item.get("active")), items[0])
print(selected["model_asset_id"])
PY
)"
curl -fsS -G "${BASE_URL%/}/api/v1/detection/yolo/preflight" \
--data-urlencode "tile_manifest_path=${manifest_path}" \
--data-urlencode "model_asset_id=${model_asset_id}" > "${TMP_DIR}/preflight.json"
require_json_data "${TMP_DIR}/preflight.json"
"${PYTHON_BIN}" - "${TMP_DIR}/preflight.json" "${model_asset_id}" <<'PY'
import json
import sys
path, expected_model_asset_id = sys.argv[1], sys.argv[2]
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("model_asset_id") != expected_model_asset_id:
raise SystemExit("YOLO preflight did not use the selected model_asset_id")
if data.get("will_download_models") is not False:
raise SystemExit("YOLO preflight must never download model weights")
if data.get("will_run_inference") is not False:
raise SystemExit("YOLO preflight must remain read-only")
if data.get("status") != "ready":
raise SystemExit(f"YOLO preflight is not ready: {data.get('status')} {data.get('message')}")
checks = data.get("checks") or {}
if checks.get("manifest_valid") is not True:
raise SystemExit("YOLO preflight did not validate the raster tile manifest")
if checks.get("model_file_exists") is not True:
raise SystemExit("YOLO preflight did not confirm the local model file")
PY
"${PYTHON_BIN}" - "${TMP_DIR}/run_request.json" "${project_id}" "${raster_dataset_id}" "${model_asset_id}" "${manifest_path}" "${REAL_CONFIDENCE_THRESHOLD}" <<'PY'
import json
import sys
path, project_id, dataset_id, model_asset_id, tile_manifest_path, confidence = sys.argv[1:7]
payload = {
"project_id": project_id,
"dataset_id": dataset_id,
"model_id": "yolo-configured",
"model_asset_id": model_asset_id,
"confidence_threshold": float(confidence),
"class_filter": ["building"],
"tile_manifest_path": tile_manifest_path,
"parameters_json": {
"workflow": "real_data_detection_qa",
"operator_supplied_inputs": True,
},
}
with open(path, "w", encoding="utf-8") as handle:
json.dump(payload, handle)
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/detection/run" \
-H "Content-Type: application/json" \
--data-binary "@${TMP_DIR}/run_request.json" > "${TMP_DIR}/detection_run.json"
require_json_data "${TMP_DIR}/detection_run.json"
"${PYTHON_BIN}" - "${TMP_DIR}/detection_run.json" <<'PY'
import json
import sys
with open(sys.argv[1], "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("model_id") != "yolo-configured":
raise SystemExit("Detection run did not use yolo-configured")
if not data.get("analysis_run_id") or not data.get("job_id"):
raise SystemExit("Detection run did not return persisted run/job ids")
if data.get("status") != "success":
raise SystemExit(f"Detection run failed: {data.get('error_code')} {data.get('message')}")
if int(data.get("detection_count") or 0) < 0:
raise SystemExit("Detection count cannot be negative")
PY
analysis_run_id="$(json_field "${TMP_DIR}/detection_run.json" "data.analysis_run_id")"
detection_count="$(json_field "${TMP_DIR}/detection_run.json" "data.detection_count")"
curl -fsS "${BASE_URL%/}/api/v1/detection/runs/${analysis_run_id}/detections" > "${TMP_DIR}/detections.json"
require_json_data "${TMP_DIR}/detections.json"
"${PYTHON_BIN}" - "${TMP_DIR}/detections.json" "${detection_count}" <<'PY'
import json
import sys
path, expected_count = sys.argv[1], int(sys.argv[2])
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
items = data.get("items") or []
if int(data.get("total") or 0) != expected_count:
raise SystemExit("Detection list total does not match run detection_count")
if len(items) != expected_count:
raise SystemExit("Detection list item count does not match run detection_count")
for item in items:
if not item.get("source_tile_path"):
raise SystemExit("Persisted detection is missing source_tile_path provenance")
PY
curl -fsS "${BASE_URL%/}/api/v1/detection/runs/${analysis_run_id}/geojson" > "${TMP_DIR}/detection_geojson.json"
require_json_data "${TMP_DIR}/detection_geojson.json"
"${PYTHON_BIN}" - "${TMP_DIR}/detection_geojson.json" "${detection_count}" <<'PY'
import json
import sys
path, expected_count = sys.argv[1], int(sys.argv[2])
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("type") != "FeatureCollection":
raise SystemExit("Detection GeoJSON response is not a FeatureCollection")
if len(data.get("features") or []) != expected_count:
raise SystemExit("Detection GeoJSON feature count does not match detection_count")
PY
"${PYTHON_BIN}" - "${TMP_DIR}/qa_request.json" "${reference_dataset_id}" "${REAL_IOU_THRESHOLD}" "${REAL_CONFIDENCE_THRESHOLD}" <<'PY'
import json
import sys
path, reference_dataset_id, iou_threshold, confidence = sys.argv[1:5]
payload = {
"reference_dataset_id": reference_dataset_id,
"iou_threshold": float(iou_threshold),
"class_name": "building",
"min_confidence": float(confidence),
}
with open(path, "w", encoding="utf-8") as handle:
json.dump(payload, handle)
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/detection/runs/${analysis_run_id}/qa/reference" \
-H "Content-Type: application/json" \
--data-binary "@${TMP_DIR}/qa_request.json" > "${TMP_DIR}/qa.json"
require_json_data "${TMP_DIR}/qa.json"
"${PYTHON_BIN}" - "${TMP_DIR}/qa.json" "${analysis_run_id}" "${reference_dataset_id}" <<'PY'
import json
import sys
path, analysis_run_id, reference_dataset_id = sys.argv[1:4]
with open(path, "r", encoding="utf-8") as handle:
data = json.load(handle)["data"]
if data.get("analysis_run_id") != analysis_run_id:
raise SystemExit("Detection QA returned the wrong analysis_run_id")
if data.get("reference_dataset_id") != reference_dataset_id:
raise SystemExit("Detection QA returned the wrong reference_dataset_id")
if not data.get("quality_check_id"):
raise SystemExit("Detection QA did not persist a quality_check_id")
if int(data.get("reference_feature_count") or 0) < 1:
raise SystemExit("Detection QA reference feature count is empty")
for key in ("matches", "false_positives", "false_negatives"):
if int(data.get(key) or 0) < 0:
raise SystemExit(f"Detection QA returned a negative {key}")
PY
quality_check_id="$(json_field "${TMP_DIR}/qa.json" "data.quality_check_id")"
"${PYTHON_BIN}" - "${TMP_DIR}/export_request.json" "${analysis_run_id}" <<'PY'
import json
import sys
path, analysis_run_id = sys.argv[1], sys.argv[2]
payload = {
"export_kind": "detection_run",
"analysis_run_id": analysis_run_id,
"name": "real-data-detection-run",
}
with open(path, "w", encoding="utf-8") as handle:
json.dump(payload, handle)
PY
curl -fsS -X POST "${BASE_URL%/}/api/v1/exports/geojson" \
-H "Content-Type: application/json" \
--data-binary "@${TMP_DIR}/export_request.json" > "${TMP_DIR}/export.json"
require_json_data "${TMP_DIR}/export.json"
export_id="$(json_field "${TMP_DIR}/export.json" "data.export_id")"
if [ -z "${export_id}" ] || [ "${export_id}" = "None" ] || [ "${export_id}" = "null" ]; then
echo "Detection GeoJSON export did not return an export_id" >&2
exit 1
fi
curl -fsS "${BASE_URL%/}/api/v1/exports/${export_id}/content" > "${TMP_DIR}/export_content.json"
require_json_data "${TMP_DIR}/export_content.json"
"${PYTHON_BIN}" - "${TMP_DIR}/export_content.json" "${detection_count}" <<'PY'
import json
import sys
path, expected_count = sys.argv[1], int(sys.argv[2])
with open(path, "r", encoding="utf-8") as handle:
content = json.load(handle)["data"]["content"]
if content.get("type") != "FeatureCollection":
raise SystemExit("Detection export content is not a FeatureCollection")
if len(content.get("features") or []) != expected_count:
raise SystemExit("Detection export feature count does not match detection_count")
PY
echo "Real data detection + QA workflow verification passed"
echo "Project: ${project_id}"
echo "Raster dataset: ${raster_dataset_id}"
echo "Reference dataset: ${reference_dataset_id}"
echo "Model asset: ${model_asset_id}"
echo "Manifest: ${manifest_path}"
echo "Analysis run: ${analysis_run_id}"
echo "Detections: ${detection_count}"
echo "Quality check: ${quality_check_id}"
echo "Detection export: ${export_id}"