fix: 修复 auto_trade 推理预处理逻辑，新增 create_labels 版对比测试脚本

auto_trade_polymarket.py: 新增 preprocess_for_inference 方法，推理时不调用
create_labels（避免 df[:-1] 砍掉输入K线），与 test 脚本逻辑一致。
test_dlinear_jan_2026_create_labels.py: 使用 create_labels 方式的 DLinear 测试
脚本，用于对比两种预处理方式的准确率差异。

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

auto_trade_polymarket.py

@@ -123,24 +123,44 @@ class ClassificationPreprocessor:
         df = self.add_features(df)
         df = self.create_labels(df)
         df = df.dropna().reset_index(drop=True)
-        
-        feature_cols = ['open', 'high', 'low', 'close', 'volume', 
+
+        feature_cols = ['open', 'high', 'low', 'close', 'volume',
                        'rsi', 'macd', 'macd_hist', 'macd_signal']
-        
+
         if fit:
             scaled_features = self.scaler.fit_transform(df[feature_cols])
             self.is_fitted = True
             self.feature_columns = feature_cols
         else:
             scaled_features = self.scaler.transform(df[feature_cols])
-        
+
         X, y = [], []
         for i in range(len(scaled_features) - self.lookback):
             X.append(scaled_features[i:i + self.lookback])
             y.append(df['label'].iloc[i + self.lookback])
-        
+
         return np.array(X), np.array(y), df
 
+    def preprocess_for_inference(self, df: pd.DataFrame):
+        """推理专用预处理：不调用create_labels，整个df一次性处理
+
+        与test_dlinear_jan_2026.py逻辑一致：
+        加特征 -> dropna -> scaler.transform -> 构建窗口
+        """
+        df = self.add_features(df)
+        df = df.dropna().reset_index(drop=True)
+
+        feature_cols = ['open', 'high', 'low', 'close', 'volume',
+                       'rsi', 'macd', 'macd_hist', 'macd_signal']
+
+        scaled_features = self.scaler.transform(df[feature_cols])
+
+        X = []
+        for i in range(len(scaled_features) - self.lookback):
+            X.append(scaled_features[i:i + self.lookback])
+
+        return np.array(X), df
+
 
 class PolymarketBTCTrader:
     """Polymarket BTC Hourly交易器"""
@@ -616,11 +636,19 @@ class PolymarketBTCTrader:
         logger.info(f"  当前K线时间: {current_time} (预测目标)")
         logger.info(f"  当前收盘价: ${current_close:,.2f} (实际结果)")
         
-        # 用除最后一根外的数据进行预测
+        # 用除最后一根外的数据进行预测（无create_labels，与test逻辑一致）
         df_for_prediction = df.iloc[:-1]
-        X, _, _ = self.preprocessor.preprocess(df_for_prediction, fit=False)
+        X, processed_df = self.preprocessor.preprocess_for_inference(df_for_prediction)
+
+        # 打印模型输入的最后一根K线数据
+        last_input_row = df_for_prediction.iloc[-1]
+        logger.info(f"  📌 模型输入末根K线: {last_input_row['timestamp']} "
+                     f"O:{last_input_row['open']:.2f} H:{last_input_row['high']:.2f} "
+                     f"L:{last_input_row['low']:.2f} C:{last_input_row['close']:.2f} "
+                     f"V:{last_input_row['volume']:.0f}")
+
         X_last = torch.FloatTensor(X[-1:]).to(self.device)
-        
+
         with torch.no_grad():
             logits = self.model(X_last).cpu().numpy()
             prob = 1 / (1 + np.exp(-logits))
@@ -632,14 +660,6 @@ class PolymarketBTCTrader:
         logger.info(f"🤖 DLinear预测: {direction}")
         logger.info(f"   置信度: {confidence:.2%}")
         
-        # 显示预测是否正确 (用于验证)
-        actual_up = current_close > prev_close
-        if prediction == 1:
-            correct = actual_up
-        else:
-            correct = not actual_up
-        logger.info(f"   实际涨跌: {'📈 涨' if actual_up else '📉 跌'} ({'✅ 正确' if correct else '❌ 错误'})")
-        
         return prediction, probability
     
     def run_once(self) -> Dict:
@@ -784,9 +804,9 @@ class PolymarketBTCTrader:
         logger.info(f"  上一K线: {prev_time} 收盘价: ${prev_close:,.2f}")
         logger.info(f"  当前K线: {current_time} 收盘价: ${current_close:,.2f}")
         
-        # 进行预测
+        # 进行预测（无create_labels，与test逻辑一致）
         df_for_prediction = df.iloc[:-1]
-        X, _, _ = self.preprocessor.preprocess(df_for_prediction, fit=False)
+        X, _ = self.preprocessor.preprocess_for_inference(df_for_prediction)
         X_last = torch.FloatTensor(X[-1:]).to(self.device)
         
         with torch.no_grad():