Initial commit: cc-cursor 全链路量化研究平台

7 Sprints 全部完成: Sprint 0: 基础设施 (DataManager + MariaDB) Sprint 1: 因子引擎 (34因子/12分类) Sprint 2: VectorBT 回测 (5策略+截面) Sprint 3: Optuna 优化 (+Walk-Forward) Sprint 4: ML 模型 (LightGBM+CatBoost) Sprint 5: Qwen 情绪因子 (三源新闻+日期对齐) Sprint 6: Agent 系统 (4Agent+日报.md/.html) 生产加固 (15项): Tushare双源fallback, SSH自动恢复, pool_pre_ping, save_daily先删后插, load_dotenv绝对路径, 日报5d/20d修复, RiskAgent改上证指数, 昨日对比+数据截止, mac_report utf8mb4, CLAUDE-*.md 9条已知Bug, demo全参数化, djapi数据源归一化, indexDatas API修正 Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-07 15:59:05 +08:00
commit 271a9343a5
293 changed files with 59598 additions and 0 deletions
@@ -0,0 +1,74 @@
+"""
+Optuna 目标函数。
+
+将策略实例化 → 回测 → 提取指标，包装为 Optuna objective。
+"""
+
+import optuna
+import pandas as pd
+
+from backtest.base import BaseStrategy
+from backtest.vectorbt.engine import VectorBTEngine
+from optimizer.space import SearchSpace
+
+# 指标提取器：从 BacktestReport 取对应字段
+_METRIC_EXTRACTORS = {
+    "sharpe": lambda r: r.sharpe_ratio,
+    "cagr": lambda r: r.cagr,
+    "calmar": lambda r: r.calmar_ratio,
+    "total_return": lambda r: r.total_return,
+    "return_over_dd": lambda r: abs(r.total_return / r.max_drawdown) if r.max_drawdown != 0 else 0.0,
+    "win_rate": lambda r: r.win_rate,
+    "profit_factor": lambda r: r.profit_factor,
+}
+
+
+class Objective:
+    """
+    Optuna 目标函数（可调用）。
+
+    用法:
+        obj = Objective(SMACrossStrategy, sma_cross_space, price_df, factor_df, metric="sharpe")
+        study = optuna.create_study(direction="maximize")
+        study.optimize(obj, n_trials=100)
+    """
+
+    def __init__(
+        self,
+        strategy_class: type[BaseStrategy],
+        search_space: SearchSpace,
+        price_df: pd.DataFrame,
+        factor_df: pd.DataFrame | None = None,
+        bt_engine: VectorBTEngine | None = None,
+        metric: str = "sharpe",
+    ):
+        self.strategy_class = strategy_class
+        self.search_space = search_space
+        self.price_df = price_df
+        self.factor_df = factor_df if factor_df is not None else price_df
+        self.bt_engine = bt_engine or VectorBTEngine()
+        self.metric = metric
+        self._extractor = _METRIC_EXTRACTORS.get(metric)
+        if self._extractor is None:
+            raise ValueError(f"不支持的指标: '{metric}'。可选: {list(_METRIC_EXTRACTORS)}")
+
+    def __call__(self, trial: optuna.Trial) -> float:
+        params = self.search_space.suggest(trial)
+
+        try:
+            strategy = self.strategy_class(**params)
+        except TypeError:
+            # 过滤不匹配的参数
+            valid = {k: v for k, v in params.items()
+                     if k in self.strategy_class.__init__.__code__.co_varnames}
+            strategy = self.strategy_class(**valid)
+
+        report = self.bt_engine.run(strategy, self.price_df, self.factor_df)
+
+        value = self._extractor(report)  # type: ignore
+
+        # 无效值处理
+        if value is None or (isinstance(value, float) and (pd.isna(value) or value == float("inf"))):
+            return float("-inf")
+
+        return float(value)