Update app.py

app.py

@@ -724,19 +724,26 @@ class RiskAnalyzer:
         return covariance / market_variance if market_variance != 0 else 0
 # ========================== 風險管理模組 END ==========================
 
-def get_stock_data(symbol, period='1y'):
-    """獲取股票資料"""
+def get_stock_data(symbol, period):
+    """
+    從 yfinance 獲取股票數據
+    
+    Args:
+        symbol: 股票代碼
+        period: 數據時間段（例如 '1mo', '3mo', '1y'）
+    
+    Returns:
+        pd.DataFrame: 股票數據
+    """
     try:
-        stock = yf.Ticker(symbol)
-        data = stock.history(period=period)
-        if data.empty and symbol == 'TXF=F':
-            stock = yf.Ticker('0050.TW')
-            data = stock.history(period=period)
-            if data.empty:
-                stock = yf.Ticker('^TWII')
-                data = stock.history(period=period)
+        ticker = yf.Ticker(symbol)
+        data = ticker.history(period=period)
+        if data.empty:
+            print(f"警告：無法獲取 {symbol} 的數據")
+            return pd.DataFrame()
         return data
-    except:
+    except Exception as e:
+        print(f"獲取 {symbol} 數據時發生錯誤：{e}")
         return pd.DataFrame()
 
 def get_us_market_data():
@@ -805,49 +812,71 @@ def simple_statistical_predict(data, predict_days=5):
         'confidence': max(0.6, 1 - volatility * 2)
     }
 
-def calculate_new_features(df):
+def calculate_features(stock_data, dji_data=None, sox_data=None, news_score=0.1):
     """
-    計算新的技術指標特徵 - 針對新特徵需求
+    從股票數據計算 XGBoostModel 所需的特徵
+    
+    Args:
+        stock_data: 來自 yfinance 的股票數據
+        dji_data: 道瓊指數數據
+        sox_data: 費半指數數據
+        news_score: 新聞情緒分數（預設 0.1）
+    
+    Returns:
+        pd.DataFrame: 包含特徵的 DataFrame
     """
-    if df.empty:
-        return df
+    df = stock_data.copy()
     
-    # 1. return_t-1 – 前一日報酬率
+    # 1. close - 前一日收盤價
+    df['close'] = df['Close']
+    
+    # 2. return_t-1 - 前一日報酬率
     df['return_t-1'] = df['Close'].pct_change()
     
-    # 2. return_t-5 – 過去 5 日累積報酬率
+    # 3. return_t-5 - 過去 5 日累積報酬率
     df['return_t-5'] = (df['Close'] / df['Close'].shift(5) - 1)
     
-    # 3. MA5_close – 5 日移動平均價
+    # 4. MA5_close - 5 日移動平均價
     df['MA5_close'] = df['Close'].rolling(window=5).mean()
     
-    # 4. MA20_close – 20 日移動平均價  
-    df['MA20_close'] = df['Close'].rolling(window=20).mean()
-    
-    # 5. volatility_5d – 5 日報酬標準差（短期波動）
+    # 5. volatility_5d - 5 日報酬標準差
     df['volatility_5d'] = df['return_t-1'].rolling(window=5).std()
     
-    # 6. volume_ratio_5d – 今日成交量 ÷ 5 日均量
+    # 6. volume_ratio_5d - 今日成交量 ÷ 5 日均量
     df['volume_5d_avg'] = df['Volume'].rolling(window=5).mean()
     df['volume_ratio_5d'] = df['Volume'] / df['volume_5d_avg']
     
-    # 7. RSI_14 – 14 日 RSI 指標
-    delta = df['Close'].diff()
-    gain = (delta.where(delta > 0, 0)).rolling(window=14).mean()
-    loss = (-delta.where(delta < 0, 0)).rolling(window=14).mean()
-    rs = gain / loss
-    df['RSI_14'] = 100 - (100 / (1 + rs))
-    
-    # 8. MACD_diff – MACD - signal（趨勢強弱）
+    # 7. MACD_diff - MACD - signal
     exp1 = df['Close'].ewm(span=12).mean()
     exp2 = df['Close'].ewm(span=26).mean()
     macd_line = exp1 - exp2
     signal_line = macd_line.ewm(span=9).mean()
     df['MACD_diff'] = macd_line - signal_line
     
-    # 移除輔助欄位
-    if 'volume_5d_avg' in df.columns:
-        df = df.drop('volume_5d_avg', axis=1)
+    # 8. dji_return_t-1 - 前一日道瓊指數報酬率
+    if dji_data is not None and not dji_data.empty:
+        df['dji_return_t-1'] = dji_data['Close'].pct_change()
+    else:
+        df['dji_return_t-1'] = 0.0
+    
+    # 9. sox_return_t-1 - 前一日費半指數報酬率
+    if sox_data is not None and not sox_data.empty:
+        df['sox_return_t-1'] = sox_data['Close'].pct_change()
+    else:
+        df['sox_return_t-1'] = 0.0
+    
+    # 10. NEWS - 新聞情緒分數
+    df['NEWS'] = news_score
+    
+    # 選擇 XGBoostModel 所需的特徵
+    feature_columns = [
+        'close', 'return_t-1', 'return_t-5', 'MA5_close', 'volatility_5d',
+        'volume_ratio_5d', 'MACD_diff', 'dji_return_t-1', 'sox_return_t-1', 'NEWS'
+    ]
+    df = df[feature_columns]
+    
+    # 移除 NaN 值
+    df = df.fillna(method='ffill').fillna(0)
     
     return df
 
@@ -1062,23 +1091,43 @@ def advanced_xgboost_predict(predict_days=5):
         traceback.print_exc()
         return None
 
-def get_prediction(data, predict_days=5):
+def get_prediction(stock_data, days):
     """
-    【【模型預測控制器】】
-    根據 USE_ADVANCED_MODEL 的設定，呼叫對應的預測模型。
-    """
-    if USE_ADVANCED_MODEL:
-        print(f"模式: 進階XGBoost模型 | 預測天期: {predict_days}天")
-        prediction = advanced_xgboost_predict(predict_days)
-        # 如果進階模型預測失敗，則自動降級使用簡易模型
-        if prediction is not None:
-            return prediction
-        else:
-            print("進階模型預測失敗，自動降級為簡易統計模型。")
+    為給定的股票數據生成預測
     
-    # 預設或降級時執行簡易模型
-    print(f"模式: 簡易統計模型 | 預測天期: {predict_days}天")
-    return simple_statistical_predict(data, predict_days)
+    Args:
+        stock_data: 股價歷史資料
+        days: 預測天數 (1, 5, 10, 20)
+    
+    Returns:
+        float: 預測的漲幅百分比
+    """
+    try:
+        # 獲取道瓊和費半數據
+        dji_data = get_stock_data('^DJI', '1mo')
+        sox_data = get_stock_data('^SOX', '1mo')
+        
+        # 計算特徵（使用最新一天數據）
+        features_df = calculate_features(stock_data, dji_data, sox_data, news_score=0.1)
+        if features_df.empty:
+            return None
+            
+        # 只取最新一天的特徵
+        latest_features = features_df.tail(1)
+        
+        # 初始化 XGBoost 模型
+        model = XGBoostModel()
+        
+        # 進行預測
+        predictions = model.predict('xgboost_model', latest_features)
+        if predictions is None:
+            return None
+            
+        target_key = f'Change_pct_t{days}_pred'
+        return predictions.get(target_key, 0.0)
+    except Exception as e:
+        print(f"預測失敗：{e}")
+        return None
 
 def calculate_technical_indicators(df):
     """計算技術指標"""