Update app.py

app.py

@@ -933,112 +933,80 @@ def advanced_xgboost_predict(predict_days=5):
         
         taiex_data = calculate_technical_indicators(taiex_data)
         taiex_data = calculate_new_features(taiex_data)
+
+        
         
-        # 強制重新獲取美股數據
+        # 新增 volume_weighted_return 計算
+        if 'return_t-1' in taiex_data.columns and 'Volume' in taiex_data.columns:
+             taiex_data['volume_weighted_return'] = abs(taiex_data['return_t-1']) * taiex_data['Volume']
+        else:
+             taiex_data['volume_weighted_return'] = 0
+
         print("正在獲取美股數據...")
         us_market_data = get_us_market_data()
         
-        # 獲取新聞情緒分數
         try:
             if predictor is not None:
                 sentiment_score_raw = predictor.get_news_index()
-                if sentiment_score_raw is None:
-                    sentiment_score_raw = 0
+                if sentiment_score_raw is None: sentiment_score_raw = 0
                 print(f"新聞情緒分數: {sentiment_score_raw}")
             else:
                 sentiment_score_raw = 0
         except:
             sentiment_score_raw = 0
         
-        # 準備特徵數據
         latest_data = taiex_data.iloc[-1]
-        yesterday_close = latest_data['Close']
-
-        # 特徵列表保持不變
-        new_feature_columns = [
-            'return_t-1',
-            'return_t-5',
-            'MA5_close',
-            'volatility_5d',
-            'volume_ratio_5d',
-            'MACD_diff',
-            'MACDvol',
-            'RSI_14',
-            'ADX',
-            'volume_weighted_return'
-            
-        ]
-        
-        # 獲取美股報酬率
+
+        # 【【修改點】】定義完整的特徵列表
+        feature_columns_map = {
+            'close': latest_data.get('Close', 0),
+            'return_t-1': latest_data.get('return_t-1', 0),
+            'return_t-5': latest_data.get('return_t-5', 0),
+            'MA5_close': latest_data.get('MA5_close', 0),
+            'volatility_5d': latest_data.get('volatility_5d', 0),
+            'volume_ratio_5d': latest_data.get('volume_ratio_5d', 0),
+            'MACD_diff': latest_data.get('MACD_diff', 0),
+            'NEWS': sentiment_score_raw,
+            'MACDvol': latest_data.get('MACDvol', 0),
+            'RSI_14': latest_data.get('RSI', 0), # 注意: app.py中計算的欄位名為 'RSI'
+            'ADX': latest_data.get('ADX', 0),
+            'volume_weighted_return': latest_data.get('volume_weighted_return', 0)
+        }
+
         dji_return = 0
         sox_return = 0
-        
         try:
             dji_data = get_stock_data('^DJI', '5d')
             if not dji_data.empty and len(dji_data) >= 2:
                 dji_return = (dji_data['Close'].iloc[-1] / dji_data['Close'].iloc[-2] - 1)
-                print(f"道瓊報酬率: {dji_return:.4f}")
-        except:
-            pass
-            
+        except: pass
         try:
             sox_data = get_stock_data('^SOX', '5d') 
             if not sox_data.empty and len(sox_data) >= 2:
                 sox_return = (sox_data['Close'].iloc[-1] / sox_data['Close'].iloc[-2] - 1)
-                print(f"費半報酬率: {sox_return:.4f}")
-        except:
-            pass
-        
-        # 建立特徵向量
-        features_list = []
-        feature_names = []
-        
-        for feature in new_feature_columns:
-            if feature in latest_data.index:
-                value = latest_data[feature]
-                if pd.isna(value):
-                    if 'return' in feature: default_value = 0.0
-                    elif 'MA' in feature: default_value = latest_data['Close'] if not pd.isna(latest_data['Close']) else 100
-                    elif 'volatility' in feature: default_value = 0.02
-                    elif 'volume_ratio' in feature: default_value = 1.0
-                    elif 'MACD' in feature: default_value = 0.0
-                    else: default_value = 0.0
-                    
-                    features_list.append(default_value)
-                else:
-                    features_list.append(value)
-                
-                feature_names.append(feature)
+        except: pass
 
-        # 添加其他特徵
-        features_list.extend([dji_return, sox_return, yesterday_close, sentiment_score_raw])
-        feature_names.extend(['dji_return_t-1', 'sox_return_t-1', 'close', 'NEWS'])
-        
-        # 轉換為 DataFrame
-        input_df = pd.DataFrame([features_list], columns=feature_names)
+        feature_columns_map['dji_return_t-1'] = dji_return
+        feature_columns_map['sox_return_t-1'] = sox_return
+
+        # 處理可能的 NaN 值
+        for key, value in feature_columns_map.items():
+            if pd.isna(value):
+                feature_columns_map[key] = 0
+
+        input_df = pd.DataFrame([feature_columns_map])
         
-        print(f"特徵向量: {[f'{f:.4f}' for f in features_list[:5]]}...")  # 只顯示前5個
-        # 🔍 新增這段：完整印出本次預測輸入��料
         print("\n=== 📊 本次預測輸入特徵 DataFrame ===")
         print(input_df)
-        print("=== ✅ 檢查以上特徵是否每次都有變 ===\n")
+        print("======================================\n")
         
-        # 進行預測
         predictions = xgb_model.predict('xgboost_model', input_df)
         
         if predictions is None:
             return None
         
-        # 處理預測結果
-        pred_mapping = {
-            1: 'Change_pct_t1_pred',
-            5: 'Change_pct_t5_pred',
-            10: 'Change_pct_t10_pred', 
-            20: 'Change_pct_t20_pred'
-        }
-        
-        available_days = [1, 5, 10, 20]
-        closest_day = min(available_days, key=lambda x: abs(x - predict_days))
+        pred_mapping = {1: 'Change_pct_t1_pred', 5: 'Change_pct_t5_pred', 10: 'Change_pct_t10_pred', 20: 'Change_pct_t20_pred'}
+        closest_day = min(pred_mapping.keys(), key=lambda x: abs(x - predict_days))
         pred_key = pred_mapping[closest_day]
         
         predicted_change_pct = predictions[pred_key]
@@ -1047,11 +1015,7 @@ def advanced_xgboost_predict(predict_days=5):
         
         print(f"XGBoost預測結果 - 漲幅: {predicted_change_pct:+.2f}%, 時間: {datetime.now().strftime('%H:%M:%S')}")
         
-        return {
-            'predicted_price': predicted_price,
-            'change_pct': predicted_change_pct,
-            'confidence': 0.75
-        }
+        return {'predicted_price': predicted_price, 'change_pct': predicted_change_pct, 'confidence': 0.75}
         
     except Exception as e:
         print(f"XGBoost 預測錯誤: {e}")
@@ -1146,7 +1110,15 @@ def calculate_technical_indicators(df):
     df['-DI'] = (df['-DM'].ewm(com=13, adjust=False).mean() / df['TR'].ewm(com=13, adjust=False).mean()) * 100
     df['DX'] = abs(df['+DI'] - df['-DI']) / (df['+DI'] + df['-DI']) * 100
     df['ADX'] = df['DX'].ewm(com=13, adjust=False).mean()
-    
+
+    # 【【新增】】計算成交量MACD (MACDvol)
+    if 'Volume' in df.columns and not df['Volume'].isnull().all():
+        exp1_vol = df['Volume'].ewm(span=12, adjust=False).mean()
+        exp2_vol = df['Volume'].ewm(span=26, adjust=False).mean()
+        df['MACDvol'] = exp1_vol - exp2_vol
+    else:
+        df['MACDvol'] = 0 # 如果沒有成交量數據，則設為0
+        
     return df
 
 def calculate_volume_profile(df, num_bins=50):