Create scorer.py

scorer.py

@@ -0,0 +1,49 @@
+from dataclasses import dataclass
+from typing import Dict, Any, List
+import re
+
+REQ = [
+    "combination_coherence_score",
+    "dominant_response_basin",
+    "cross_system_alignment_vector",
+    "early_incoherence_flags",
+    "net_functional_shift",
+]
+
+@dataclass
+class ScoreResult:
+    score: float
+    details: Dict[str, Any]
+
+def _f(p: str, key: str):
+    m = re.search(rf"{key}\s*[:=]\s*(0\.\d+|1\.0)\b", p)
+    return float(m.group(1)) if m else None
+
+def score(sample: Dict[str, Any], prediction: str) -> ScoreResult:
+    p = (prediction or "").lower()
+    words_ok = len(p.split()) <= 950
+    hits = sum(1 for k in REQ if k in p)
+
+    coh = _f(p, "combination_coherence_score")
+    num_ok = int(coh is not None and 0 <= coh <= 1)
+
+    basin_ok = int("dominant_response_basin" in p)
+    vec_ok = int("cross_system_alignment_vector" in p and ":" in p)
+    flags_ok = int("early_incoherence_flags" in p)
+    shift_ok = int("net_functional_shift" in p)
+
+    raw = (
+        0.18 * int(words_ok) +
+        0.42 * (hits / len(REQ)) +
+        0.18 * num_ok +
+        0.08 * basin_ok +
+        0.07 * vec_ok +
+        0.03 * flags_ok +
+        0.04 * shift_ok
+    )
+    return ScoreResult(score=min(1.0, raw), details={"id": sample.get("id"), "hits": hits})
+
+def aggregate(results: List[ScoreResult]) -> Dict[str, Any]:
+    if not results:
+        return {"mean": 0.0, "n": 0}
+    return {"mean": sum(r.score for r in results) / len(results), "n": len(results)}