audit nested YOLO labels without destructive rewrites

tests/test_audit_yolo_label_relationships.py

@@ -0,0 +1,45 @@
+from scripts.audit_yolo_label_relationships import Box, audit_boxes, classify_pair
+
+
+DEFAULTS = {
+    "near_duplicate_iou": 0.9,
+    "containment_threshold": 0.98,
+    "max_nested_area_ratio": 4.0,
+}
+
+
+def test_classifies_exact_near_and_possible_nested_relationships() -> None:
+    exact = Box(0, 0.5, 0.5, 0.2, 0.2)
+    near = Box(0, 0.501, 0.5, 0.2, 0.2)
+    outer = Box(0, 0.5, 0.5, 0.18, 0.18)
+    inner = Box(0, 0.5, 0.5, 0.1, 0.1)
+
+    assert classify_pair(exact, exact, **DEFAULTS) == ("exact_duplicate", 1.0)
+    assert classify_pair(exact, near, **DEFAULTS)[0] == "near_duplicate"
+    assert classify_pair(outer, inner, **DEFAULTS) == ("possible_nested", 1.0)
+
+
+def test_ignores_other_classes_and_non_overlapping_boxes() -> None:
+    first = Box(0, 0.2, 0.2, 0.1, 0.1)
+    other_class = Box(1, 0.2, 0.2, 0.1, 0.1)
+    distant = Box(0, 0.8, 0.8, 0.1, 0.1)
+
+    assert classify_pair(first, other_class, **DEFAULTS) is None
+    assert classify_pair(first, distant, **DEFAULTS) is None
+
+
+def test_audit_boxes_returns_pair_indices_and_coordinates() -> None:
+    first = Box(0, 0.5, 0.5, 0.2, 0.2)
+    second = Box(0, 0.5, 0.5, 0.2, 0.2)
+    relationships = audit_boxes([first, second], **DEFAULTS)
+
+    assert relationships == [
+        {
+            "relationship": "exact_duplicate",
+            "score": 1.0,
+            "first_index": 0,
+            "second_index": 1,
+            "first_box": [0, 0.5, 0.5, 0.2, 0.2],
+            "second_box": [0, 0.5, 0.5, 0.2, 0.2],
+        }
+    ]