derive cleaner min-4px YOLO corpus
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 19:51:17 +02:00
parent 736e773fb8
commit 2de438b9cc
9 changed files with 471 additions and 25 deletions
@@ -9,7 +9,9 @@ from pathlib import Path
ROOT = Path(__file__).resolve().parents[2]
def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(tmp_path: Path) -> None:
def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(
tmp_path: Path,
) -> None:
script_path = ROOT / "scripts" / "audit_operator_yolo_dataset_quality.py"
assert script_path.exists()
@@ -135,10 +137,12 @@ def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(tmp_path: Pat
text=True,
)
assert "Operator YOLO dataset quality audit passed" in result.stdout
assert "Operator YOLO dataset quality audit completed" in result.stdout
report = json.loads(
(output_dir / "operator_yolo_dataset_quality_audit.json").read_text(encoding="utf-8")
(output_dir / "operator_yolo_dataset_quality_audit.json").read_text(
encoding="utf-8"
)
)
assert report["status"] == "needs_attention"
assert report["sample_count"] == 3
@@ -150,7 +154,9 @@ def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(tmp_path: Pat
assert report["label_stats"]["parsed_label_count"] == 3
assert report["label_stats"]["invalid_label_count"] == 0
assert report["min_label_visible_ratio"] == 0.25
sample_by_slug = {sample["sample_slug"]: sample for sample in report["sample_summaries"]}
sample_by_slug = {
sample["sample_slug"]: sample for sample in report["sample_summaries"]
}
assert sample_by_slug["geel"]["parsed_label_count"] == 2
assert sample_by_slug["geel"]["invalid_label_count"] == 0
assert sample_by_slug["geel"]["small_box_share"] == 0.5
@@ -161,7 +167,9 @@ def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(tmp_path: Pat
]
assert sample_by_slug["turnhout"]["parsed_label_count"] == 1
assert sample_by_slug["turnhout"]["small_box_share"] == 0.0
assert sample_by_slug["turnhout"]["quality_warnings"] == ["median_box_area_below_gate"]
assert sample_by_slug["turnhout"]["quality_warnings"] == [
"median_box_area_below_gate"
]
assert sample_by_slug["postel_bos"]["parsed_label_count"] == 0
assert sample_by_slug["postel_bos"]["quality_warnings"] == []
@@ -173,7 +181,9 @@ def test_operator_yolo_dataset_quality_audit_reports_dataset_risks(tmp_path: Pat
assert "small_box_share_above_gate" in warning_codes
assert "positive_tiles_have_low_visual_variance" in warning_codes
markdown = (output_dir / "operator_yolo_dataset_quality_audit.md").read_text(encoding="utf-8")
markdown = (output_dir / "operator_yolo_dataset_quality_audit.md").read_text(
encoding="utf-8"
)
assert "Operator YOLO Dataset Quality Audit" in markdown
assert "Label Quality" in markdown
assert "Minimum visible label ratio" in markdown