feat: 리밸런싱 엔진 V1 + GAS 버그 수정 (2026-06-13)

주요 변경: - tools/build_rebalance_engine_v1.py: REBALANCE_ENGINE_V1 신규 * account_snapshot 직접 합산(_build_snap_position_map) → 소수주 분리 행 병합 * 레짐 소스 macro.REGIME_PRELIM 최우선 (GAS 와 동일) - src/gas_adapter_parts/gdf_06_rebalance.gs: runRebalanceSheet_() 신규 * Logger.log / getSpreadsheet_() 로 run_all 연동 수정 - src/gas_adapter_parts/gdc_01_fetch_fundamentals.gs * _mergePositionRecord_(): 소수주 중복 행 합산 신규 * parseInt → parseFloat (qty, availQty) - src/gas_adapter_parts/gdf_01_price_metrics.gs * 미보유 종목 SELL_READY → WATCH_EXIT_SIGNAL - spec/41_release_dag.yaml: build_rebalance_sheet 노드 추가 (step_count 63) - spec/51_formula_lifecycle_registry.yaml: REBALANCE_ENGINE_V1 등록 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-13 13:20:14 +09:00
commit ee3e799de1
1474 changed files with 176087 additions and 0 deletions
@@ -0,0 +1,69 @@
+from __future__ import annotations
+
+import argparse
+import json
+from datetime import datetime, timezone
+from statistics import mean
+
+from v7_hardening_common import ROOT, TEMP, load_json, save_json
+
+
+DEFAULT_OUT = TEMP / "confidence_calibration_v2.json"
+
+
+def _label_rank(label: str) -> int:
+    order = {
+        "BEARISH": 0,
+        "NEUTRAL": 1,
+        "BULLISH": 2,
+        "STRONG_BULLISH": 3,
+    }
+    return order.get(str(label).upper(), 1)
+
+
+def main() -> int:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--out", default=str(DEFAULT_OUT))
+    args = ap.parse_args()
+
+    exposure = load_json(TEMP / "imputed_data_exposure_gate_v2.json")
+    pred = load_json(TEMP / "prediction_accuracy_harness_v2.json")
+    conf = load_json(TEMP / "portfolio_alpha_confidence_per_ticker_v1.json")
+    rows = [r for r in conf.get("rows", []) if isinstance(r, dict)]
+
+    monotonic = True
+    if rows:
+        ordered = sorted(rows, key=lambda r: float(r.get("pac_score") or 0.0))
+        label_scores = [_label_rank(r.get("pac_label") or "") for r in ordered]
+        monotonic = all(a <= b for a, b in zip(label_scores, label_scores[1:]))
+
+    high_conf_low_evidence = sum(
+        1 for r in rows
+        if float(r.get("pac_score") or 0.0) >= 80.0 and str(r.get("fundamental_grade") or "").upper() in {"D", "F", "E"}
+    )
+
+    result = {
+        "formula_id": "CONFIDENCE_CALIBRATION_V2",
+        "generated_at": datetime.now(timezone.utc).isoformat(),
+        "confidence_cap_basis_score": float(exposure.get("raw_confidence_cap_basis") or exposure.get("effective_confidence_honest") or 0.0),
+        "effective_confidence_honest": float(exposure.get("effective_confidence_honest") or 0.0),
+        "confidence_cap_gap_pct": round(float(exposure.get("confidence_cap_inflation_gap") or 0.0), 1),
+        "calibration_brier_score_improving": str(pred.get("calibration_state") or "").upper() == "CALIBRATED",
+        "confidence_bucket_monotonicity": "PASS" if monotonic else "FAIL",
+        "high_confidence_low_evidence_count": high_conf_low_evidence,
+        "calibration_state": pred.get("calibration_state"),
+        "portfolio_label_diversity": int(conf.get("label_diversity") or 0),
+        "portfolio_score_mean": round(mean(float(r.get("pac_score") or 0.0) for r in rows), 2) if rows else 0.0,
+        "source_paths": [
+            "Temp/imputed_data_exposure_gate_v2.json",
+            "Temp/prediction_accuracy_harness_v2.json",
+            "Temp/portfolio_alpha_confidence_per_ticker_v1.json",
+        ],
+    }
+    save_json(args.out, result)
+    print(json.dumps(result, ensure_ascii=False, indent=2))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())