Gate filtered YOLO candidate and harden provenance

CHANGELOG.md

@@ -7,6 +7,15 @@
 
 # Changelog
 
+## Sprint 169 Filtered YOLO candidate gate and operator hardening (2026-07-12)
+
+- Trained and fully gated inactive `geointel-building-yolov8s-aoi1024cleanpx12vis035lowvar512e50-pt` against seven positive AOIs and split pure-empty/sparse-context backgrounds.
+- Rejected the candidate for default promotion: best mean positive F1 was approximately `0.154`, below the `0.25` gate; threshold `0.05` also produced one pure-empty background detection.
+- Preserved the current active `aoi1024bg512r3e50` model and all runtime defaults.
+- Added SHA256 provenance to future YOLO training summaries.
+- Improved long project/dataset/AOI readability with matching tooltips and compact two-line readiness values.
+- No API contract, migration, provider fetching, fake output, model download or automatic model activation changed.
+
 ## Sprint 164 Tower AI deploy env hardening (2026-07-11)
 
 - Fixed Tower deploy automation so `scripts/deploy_tower.sh` and `scripts/deploy_tower.ps1` source the remote `.env` before building the all-in-one image.

backend/tests/test_sprint129_operator_yolo_training_dataset.py

@@ -70,5 +70,9 @@ def test_operator_yolo_train_smoke_script_contract() -> None:
     assert "DejaVuSans.ttf" in script
     assert "Arial.ttf" in script
     assert "training_summary.json" in script
+    assert '"dataset_yaml_sha256"' in script
+    assert '"dataset_summary_sha256"' in script
+    assert '"base_model_sha256"' in script
+    assert '"trained_model_sha256"' in script
     assert "download" not in script.lower()
     assert "fixture_mode" not in script

backend/tests/test_sprint169_long_context_name_readability.py

@@ -0,0 +1,22 @@
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[2]
+
+
+def test_long_context_names_remain_compact_and_inspectable() -> None:
+    app = (ROOT / "frontend" / "src" / "App.tsx").read_text(encoding="utf-8")
+    status = (
+        ROOT / "frontend" / "src" / "components" / "WorkbenchStatusStrip.tsx"
+    ).read_text(encoding="utf-8")
+    css = (ROOT / "frontend" / "src" / "styles" / "app.css").read_text(
+        encoding="utf-8"
+    )
+
+    assert "const projectContextLabel = selectedProject?.name ?? 'No project'" in app
+    assert "const datasetContextLabel = selectedDataset?.name ?? (datasets.length > 0 ? 'Select dataset' : 'No dataset')" in app
+    assert "<strong title={projectContextLabel}>{projectContextLabel}</strong>" in app
+    assert "<strong title={datasetContextLabel}>{datasetContextLabel}</strong>" in app
+    assert "<strong title={item.value}>{item.value}</strong>" in status
+    assert ".status-tile > strong" in css
+    assert "-webkit-line-clamp: 2;" in css

docs/CODEX_EXECUTION_LOG.md

@@ -1,3 +1,32 @@
+## Sprint 169 Filtered YOLO candidate gate and operator hardening (2026-07-12)
+
+Changed:
+- Trained inactive local model asset `geointel-building-yolov8s-aoi1024cleanpx12vis035lowvar512e50-pt` from the visually audited AOI1024 dataset after low-variance negative filtering.
+- Added SHA256 provenance fields to future `training_summary.json` output for `dataset.yaml`, the dataset summary, the local base model and the copied trained model.
+- Kept long workbench context names compact with matching native tooltips and clamped readiness values to two lines.
+- Archived the training, positive-AOI, split-background and promotion evidence under `artifacts/model-review/aoi1024cleanpx12vis035lowvar512e50` locally and matching Tower artifact directories.
+
+Runtime evidence:
+- Training completed for 50 CPU epochs with YOLOv8s, image size `512`, batch `4`; best-model validation ended at precision `0.454`, recall `0.491`, mAP50 `0.368` and mAP50-95 `0.149`.
+- Model catalog SHA256: `75345767b51cc66692a9d2c2cd6577b7feecaf5971c8762365e8b610b9dfde8e`; catalog status `available`, `active=false`, `will_download_models=false`.
+- Local model-load preflight passed with Torch `2.13.0` and Ultralytics `8.4.92`; CUDA is unavailable and no model download occurred.
+- Seven-AOI persisted QA matrix produced 28 runs. Best single result was Westerlo at threshold `0.15`, F1 `0.3002114164904862`.
+- Mean positive F1 was `0.153943` at threshold `0.05`, `0.153872` at `0.15`, `0.128891` at `0.25` and `0.099011` at `0.35`.
+- Strict pure-empty background evidence was clean at thresholds `0.15`, `0.25` and `0.35`; threshold `0.05` produced one Postel-bos detection.
+- Split-aware promotion report recommended `none`: every threshold failed the positive mean-F1 gate, and `0.05` also failed background false-positive pressure.
+- The existing active `geointel-building-yolov8s-aoi1024bg512r3e50-pt` remains materially stronger at its promoted `0.35` profile with mean F1 `0.320866`; no model default or `.env` value was changed.
+
+Tested:
+- Red step: `python -m pytest backend/tests/test_sprint169_long_context_name_readability.py -q` failed before context tooltips and readiness clamping were present.
+- `python -m pytest backend/tests/test_sprint169_long_context_name_readability.py backend/tests/test_sprint22_workbench_status_strip.py backend/tests/test_sprint161_widescreen_workbench.py -q` (`5 passed`).
+- Red step: `python -m pytest backend/tests/test_sprint129_operator_yolo_training_dataset.py::test_operator_yolo_train_smoke_script_contract -q` failed before training hashes were recorded.
+- `python -m pytest backend/tests/test_sprint129_operator_yolo_training_dataset.py -q` (`3 passed`).
+- `bash -n scripts/train_operator_yolo_detector.sh` and frontend typecheck passed.
+- Full readiness: `bash scripts/run_readiness_check.sh` (`461 passed`; one Alembic head; frontend typecheck/build and shell syntax gates passed).
+
+Next:
+- Do not retrain the same architecture blindly. Inspect per-AOI false-negative evidence and improve label geometry/class balance or add targeted positive samples for the weakest AOIs before the next candidate.
+
 ## Sprint 164 Tower AI deploy env hardening (2026-07-11)
 
 Changed:

docs/TODO.md

@@ -133,7 +133,9 @@ This file now starts with the current implementation status. Older preparation/b
 - [x] Add deterministic visual YOLO label QA contact sheets before spending more CPU on another training run.
 - [x] Filter no-data/low-variance pure-empty negative tiles from operator YOLO exports before the next training run.
 - [x] Regenerate the AOI1024 cleanpx YOLO dataset with low-variance negative filtering and rerun visual contact-sheet QA before training.
-- [ ] Train one inactive candidate from the filtered AOI1024 cleanpx YOLO dataset and gate it through the positive-AOI plus split-background promotion workflow.
+- [x] Train one inactive candidate from the filtered AOI1024 cleanpx YOLO dataset and gate it through the positive-AOI plus split-background promotion workflow; reject it because mean positive F1 remains below gate.
+- [x] Add deterministic dataset/base/trained-model SHA256 provenance to future operator training summaries.
+- [ ] Review per-AOI false-negative evidence for the weakest AOIs and improve positive sample/label geometry coverage before another training candidate.
 - [ ] Apply promoted V1 default building detector only after explicit operator review of the emitted `.env` updates, followed by rebuild/restart and browser/runtime smoke.
 
 ## Sprint 8 status

frontend/src/App.tsx

@@ -631,6 +631,10 @@ function App(): JSX.Element {
       target: 'exports',
     },
   ]
+  const projectContextLabel = selectedProject?.name ?? 'No project'
+  const areaContextLabel = selectedArea?.name ?? (areas.length > 0 ? 'Select area' : 'No AOI')
+  const datasetContextLabel = selectedDataset?.name ?? (datasets.length > 0 ? 'Select dataset' : 'No dataset')
+  const layerContextLabel = mapFeatureCollection ? `${mapFeatureCount} features` : 'No active layer'
 
   return (
     <div className="app-shell workbench-shell">
@@ -645,19 +649,19 @@ function App(): JSX.Element {
         <div className="context-bar" aria-label="Active workbench context">
           <div>
             <span>Project</span>
-            <strong>{selectedProject?.name ?? 'No project'}</strong>
+            <strong title={projectContextLabel}>{projectContextLabel}</strong>
           </div>
           <div>
             <span>AOI</span>
-            <strong>{selectedArea?.name ?? (areas.length > 0 ? 'Select area' : 'No AOI')}</strong>
+            <strong title={areaContextLabel}>{areaContextLabel}</strong>
           </div>
           <div>
             <span>Dataset</span>
-            <strong>{selectedDataset?.name ?? (datasets.length > 0 ? 'Select dataset' : 'No dataset')}</strong>
+            <strong title={datasetContextLabel}>{datasetContextLabel}</strong>
           </div>
           <div>
             <span>Layer</span>
-            <strong>{mapFeatureCollection ? `${mapFeatureCount} features` : 'No active layer'}</strong>
+            <strong title={layerContextLabel}>{layerContextLabel}</strong>
           </div>
         </div>
       </header>

frontend/src/components/WorkbenchStatusStrip.tsx

@@ -122,7 +122,7 @@ export function WorkbenchStatusStrip({
               <span>{item.label}</span>
               <span className="status-pill">{item.state}</span>
             </div>
-            <strong>{item.value}</strong>
+            <strong title={item.value}>{item.value}</strong>
             <p>{item.detail}</p>
           </div>
         ))}

frontend/src/styles/app.css

@@ -4554,6 +4554,14 @@ section {
   line-height: 1.2;
 }
 
+.status-tile > strong {
+  display: -webkit-box;
+  overflow: hidden;
+  overflow-wrap: anywhere;
+  -webkit-box-orient: vertical;
+  -webkit-line-clamp: 2;
+}
+
 .status-tile p {
   margin-top: 0.18rem;
   font-size: 0.72rem;

scripts/README.md

@@ -337,7 +337,9 @@ docker exec \
 
 The training wrapper is intentionally outside the product UI. It runs
 Ultralytics from the existing runtime, copies the best trained artifact to
-`TRAIN_MODEL_OUTPUT_PATH` and writes `training_summary.json`. Afterward, treat
+`TRAIN_MODEL_OUTPUT_PATH` and writes `training_summary.json`. The summary records
+SHA256 provenance for `dataset.yaml`, the available YOLO dataset summary, the
+local base model and the copied trained model. Afterward, treat
 the resulting `.pt` file like any other local model asset: verify preflight,
 run the real-data matrix and compare persisted QA/QC metrics before activating
 it as a useful default.

scripts/train_operator_yolo_detector.sh

@@ -82,12 +82,21 @@ mkdir -p "${TRAIN_OUTPUT_DIR}" "$(dirname "${TRAIN_MODEL_OUTPUT_PATH}")"
 "${PYTHON_BIN}" - <<'PY'
 from __future__ import annotations
 
+import hashlib
 import json
 import os
 import shutil
 from pathlib import Path
 
 
+def sha256_file(path: Path) -> str:
+    digest = hashlib.sha256()
+    with path.open("rb") as handle:
+        for chunk in iter(lambda: handle.read(1024 * 1024), b""):
+            digest.update(chunk)
+    return digest.hexdigest()
+
+
 def seed_ultralytics_font() -> None:
     font_candidates = [
         Path("/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf"),
@@ -141,6 +150,17 @@ if not best_path.exists():
     raise SystemExit(f"Expected trained model artifact was not created: {best_path}")
 
 shutil.copy2(best_path, trained_model_output_path)
+dataset_summary_path = next(
+    (
+        candidate
+        for candidate in (
+            dataset_yaml.parent / "yolo_tile_dataset_summary.json",
+            dataset_yaml.parent / "yolo_dataset_summary.json",
+        )
+        if candidate.is_file()
+    ),
+    None,
+)
 summary = {
     "status": "ok",
     "dataset_yaml": str(dataset_yaml),
@@ -149,6 +169,11 @@ summary = {
     "train_run_name": run_name,
     "trained_model_path": str(trained_model_output_path),
     "best_artifact_path": str(best_path),
+    "dataset_yaml_sha256": sha256_file(dataset_yaml),
+    "dataset_summary_path": str(dataset_summary_path) if dataset_summary_path else None,
+    "dataset_summary_sha256": sha256_file(dataset_summary_path) if dataset_summary_path else None,
+    "base_model_sha256": sha256_file(base_model_path),
+    "trained_model_sha256": sha256_file(trained_model_output_path),
     "epochs": epochs,
     "image_size": image_size,
     "batch_size": batch_size,