feat(training): 支持 Label 预处理器

- DataPipeline 新增 label_processor_configs 参数
- 分离特征与 label 的预处理流程
- regression.py 添加 label 缩尾处理配置
- 调整学习率并更新排除因子列表

src/training/pipeline.py

@@ -40,6 +40,9 @@ class DataPipeline:
         filters: Optional[List[Any]] = None,
         stock_pool_filter_func: Optional[Callable] = None,
         stock_pool_required_columns: Optional[List[str]] = None,
+        label_processor_configs: Optional[
+            List[Tuple[Type[BaseProcessor], Dict[str, Any]]]
+        ] = None,
     ):
         """初始化数据流水线
 
@@ -50,6 +53,8 @@ class DataPipeline:
             filters: 类形式的过滤器列表（如 [STFilter]）
             stock_pool_filter_func: 函数形式的股票池筛选器
             stock_pool_required_columns: 股票池筛选所需的额外列
+            label_processor_configs: Label 数据处理器配置列表，格式与 processor_configs 相同
+                例如：[(Winsorizer, {"lower": 0.01, "upper": 0.99})] 用于对 label 进行缩尾处理
         """
         self.factor_manager = factor_manager
         self.processor_configs = processor_configs or []
@@ -57,6 +62,8 @@ class DataPipeline:
         self.stock_pool_filter_func = stock_pool_filter_func
         self.stock_pool_required_columns = stock_pool_required_columns or []
         self.fitted_processors: List[BaseProcessor] = []
+        self.label_processor_configs = label_processor_configs or []
+        self.fitted_label_processors: List[BaseProcessor] = []
 
     def prepare_data(
         self,
@@ -250,6 +257,7 @@ class DataPipeline:
         """预处理数据
 
         训练集使用 fit_transform，验证集和测试集使用 transform
+        同时支持对 label 进行 processor 处理
 
         Args:
             split_data: 划分后的数据字典
@@ -259,44 +267,83 @@ class DataPipeline:
         Returns:
             预处理后的数据字典
         """
-        if not self.processor_configs:
-            return split_data
+        label_name = split_data["train"]["y"].name
 
-        self.fitted_processors = []
+        # 处理特征
+        if self.processor_configs:
+            self.fitted_processors = []
 
-        # 实例化 processors（传入 feature_cols）
-        processors = []
-        for proc_class, proc_kwargs in self.processor_configs:
-            proc_kwargs_with_cols = {**proc_kwargs, "feature_cols": feature_cols}
-            processors.append(proc_class(**proc_kwargs_with_cols))
+            # 实例化 processors（传入 feature_cols）
+            processors = []
+            for proc_class, proc_kwargs in self.processor_configs:
+                proc_kwargs_with_cols = {**proc_kwargs, "feature_cols": feature_cols}
+                processors.append(proc_class(**proc_kwargs_with_cols))
 
-        # 训练集：fit_transform
-        if verbose:
-            print(f"  训练集预处理（fit_transform）...")
+            # 训练集：fit_transform
+            if verbose:
+                print(f"  训练集特征预处理（fit_transform）...")
 
-        train_data = split_data["train"]["raw_data"]
-        for processor in processors:
-            train_data = processor.fit_transform(train_data)
-            self.fitted_processors.append(processor)
+            train_data = split_data["train"]["raw_data"]
+            for processor in processors:
+                train_data = processor.fit_transform(train_data)
+                self.fitted_processors.append(processor)
 
-        # 更新训练集
-        split_data["train"]["raw_data"] = train_data
-        split_data["train"]["X"] = train_data.select(feature_cols)
-        split_data["train"]["y"] = train_data[split_data["train"]["y"].name]
+            # 更新训练集
+            split_data["train"]["raw_data"] = train_data
+            split_data["train"]["X"] = train_data.select(feature_cols)
+            split_data["train"]["y"] = train_data[label_name]
 
-        # 验证集和测试集：transform
-        for split_name in ["val", "test"]:
-            if split_name in split_data:
-                if verbose:
-                    print(f"  {split_name}集预处理（transform）...")
+            # 验证集和测试集：transform
+            for split_name in ["val", "test"]:
+                if split_name in split_data:
+                    if verbose:
+                        print(f"  {split_name}集特征预处理（transform）...")
 
-                split_df = split_data[split_name]["raw_data"]
-                for processor in self.fitted_processors:
-                    split_df = processor.transform(split_df)
+                    split_df = split_data[split_name]["raw_data"]
+                    for processor in self.fitted_processors:
+                        split_df = processor.transform(split_df)
 
-                split_data[split_name]["raw_data"] = split_df
-                split_data[split_name]["X"] = split_df.select(feature_cols)
-                split_data[split_name]["y"] = split_df[split_data[split_name]["y"].name]
+                    split_data[split_name]["raw_data"] = split_df
+                    split_data[split_name]["X"] = split_df.select(feature_cols)
+                    split_data[split_name]["y"] = split_df[label_name]
+
+        # 处理 label
+        if self.label_processor_configs:
+            self.fitted_label_processors = []
+
+            # 实例化 label processors（传入 label_name 作为 feature_cols）
+            label_processors = []
+            for proc_class, proc_kwargs in self.label_processor_configs:
+                proc_kwargs_with_label = {**proc_kwargs, "feature_cols": [label_name]}
+                label_processors.append(proc_class(**proc_kwargs_with_label))
+
+            # 训练集：fit_transform
+            if verbose:
+                print(f"  训练集 Label 预处理（fit_transform）...")
+
+            train_data = split_data["train"]["raw_data"]
+            for processor in label_processors:
+                train_data = processor.fit_transform(train_data)
+                self.fitted_label_processors.append(processor)
+
+            # 更新训练集
+            split_data["train"]["raw_data"] = train_data
+            split_data["train"]["X"] = train_data.select(feature_cols)
+            split_data["train"]["y"] = train_data[label_name]
+
+            # 验证集和测试集：transform
+            for split_name in ["val", "test"]:
+                if split_name in split_data:
+                    if verbose:
+                        print(f"  {split_name}集 Label 预处理（transform）...")
+
+                    split_df = split_data[split_name]["raw_data"]
+                    for processor in self.fitted_label_processors:
+                        split_df = processor.transform(split_df)
+
+                    split_data[split_name]["raw_data"] = split_df
+                    split_data[split_name]["X"] = split_df.select(feature_cols)
+                    split_data[split_name]["y"] = split_df[label_name]
 
         return split_data
 
@@ -307,3 +354,11 @@ class DataPipeline:
             已拟合的处理器列表（用于模型保存）
         """
         return self.fitted_processors
+
+    def get_fitted_label_processors(self) -> List[BaseProcessor]:
+        """获取已拟合的 Label 处理器列表
+
+        Returns:
+            已拟合的 Label 处理器列表（用于模型保存和预测时反转换）
+        """
+        return self.fitted_label_processors