Initial public release

backend/tests/test_belgium_training_iteration_assessment.py

@@ -0,0 +1,37 @@
+from __future__ import annotations
+
+import importlib.util
+from pathlib import Path
+
+import pytest
+
+
+SCRIPT = Path(__file__).parents[2] / "scripts" / "assess_belgium_building_training_iteration.py"
+SPEC = importlib.util.spec_from_file_location("iteration_assessment", SCRIPT)
+assert SPEC and SPEC.loader
+MODULE = importlib.util.module_from_spec(SPEC)
+SPEC.loader.exec_module(MODULE)
+
+
+def test_calibration_selection_prefers_worst_region_then_aggregate() -> None:
+    report = {
+        "sweeps": [
+            {"threshold": 0.1, "pure_empty_false_positives": 0, "aggregate": {"f1": 0.8}, "regions": {"a": {"f1": 0.2}}},
+            {"threshold": 0.2, "pure_empty_false_positives": 0, "aggregate": {"f1": 0.6}, "regions": {"a": {"f1": 0.5}}},
+        ]
+    }
+    assert MODULE.select_calibration_threshold(report)["threshold"] == 0.2
+
+
+def test_threshold_lookup_is_exact() -> None:
+    report = {"sweeps": [{"threshold": 0.25, "aggregate": {}}]}
+    assert MODULE.find_threshold(report, 0.25)["threshold"] == 0.25
+
+
+def test_release_assessment_rejects_changed_inference_configuration() -> None:
+    calibration = {field: None for field in MODULE.INFERENCE_CONFIG_FIELDS}
+    calibration.update({"model": "/models/candidate.pt", "nms_iou": 0.3, "containment_nms": 0.95})
+    test = dict(calibration)
+    test["nms_iou"] = 0.4
+    with pytest.raises(ValueError, match="nms_iou"):
+        MODULE.assert_same_inference_config(calibration, test, "test")