Update app.py

app.py

@@ -280,6 +280,75 @@ def advanced_xgboost_predict(predict_days=5):
         latest_data = taiex_data.iloc[-1]
         
         # 建立特徵向量 (按照訓練數據記錄的順序)
+        # 先檢查每個技術指標是否存在並記錄狀態
+        tech_indicators_status = {}
+        
+        # RSI 檢查
+        rsi_value = latest_data['RSI'] if not pd.isna(latest_data['RSI']) else 50
+        tech_indicators_status['RSI'] = {
+            'value': rsi_value,
+            'is_real': not pd.isna(latest_data['RSI']),
+            'source': 'calculated' if not pd.isna(latest_data['RSI']) else 'default'
+        }
+        
+        # MACD 相關檢查
+        macd_value = latest_data['MACD'] if not pd.isna(latest_data['MACD']) else 0
+        macd_signal_value = latest_data['MACD_Signal'] if not pd.isna(latest_data['MACD_Signal']) else 0
+        macd_hist_value = latest_data['MACD_Histogram'] if not pd.isna(latest_data['MACD_Histogram']) else 0
+        
+        tech_indicators_status['MACD'] = {
+            'value': macd_value,
+            'is_real': not pd.isna(latest_data['MACD']),
+            'source': 'calculated' if not pd.isna(latest_data['MACD']) else 'default'
+        }
+        tech_indicators_status['MACD_Signal'] = {
+            'value': macd_signal_value,
+            'is_real': not pd.isna(latest_data['MACD_Signal']),
+            'source': 'calculated' if not pd.isna(latest_data['MACD_Signal']) else 'default'
+        }
+        tech_indicators_status['MACD_Histogram'] = {
+            'value': macd_hist_value,
+            'is_real': not pd.isna(latest_data['MACD_Histogram']),
+            'source': 'calculated' if not pd.isna(latest_data['MACD_Histogram']) else 'default'
+        }
+        
+        # KD 指標檢查
+        k_value = latest_data['K'] if not pd.isna(latest_data['K']) else 50
+        d_value = latest_data['D'] if not pd.isna(latest_data['D']) else 50
+        
+        tech_indicators_status['K'] = {
+            'value': k_value,
+            'is_real': not pd.isna(latest_data['K']),
+            'source': 'calculated' if not pd.isna(latest_data['K']) else 'default'
+        }
+        tech_indicators_status['D'] = {
+            'value': d_value,
+            'is_real': not pd.isna(latest_data['D']),
+            'source': 'calculated' if not pd.isna(latest_data['D']) else 'default'
+        }
+        
+        # DMI 指標檢查
+        plus_di_value = latest_data['+DI'] if not pd.isna(latest_data['+DI']) else 25
+        minus_di_value = latest_data['-DI'] if not pd.isna(latest_data['-DI']) else 25
+        adx_value = latest_data['ADX'] if not pd.isna(latest_data['ADX']) else 25
+        
+        tech_indicators_status['+DI'] = {
+            'value': plus_di_value,
+            'is_real': not pd.isna(latest_data['+DI']),
+            'source': 'calculated' if not pd.isna(latest_data['+DI']) else 'default'
+        }
+        tech_indicators_status['-DI'] = {
+            'value': minus_di_value,
+            'is_real': not pd.isna(latest_data['-DI']),
+            'source': 'calculated' if not pd.isna(latest_data['-DI']) else 'default'
+        }
+        tech_indicators_status['ADX'] = {
+            'value': adx_value,
+            'is_real': not pd.isna(latest_data['ADX']),
+            'source': 'calculated' if not pd.isna(latest_data['ADX']) else 'default'
+        }
+        
+        # 建立特徵向量
         features_list = [
             latest_data['Close'],  # close
             latest_data['Volume'],  # volume
@@ -290,15 +359,15 @@ def advanced_xgboost_predict(predict_days=5):
             us_market['S&P_500'],  # S&P_500
             us_market['TSM_ADR'],  # TSM_ADR
             sentiment_score_raw,  # NEWS (使用原始 sentiment_score_raw)
-            latest_data['RSI'] if not pd.isna(latest_data['RSI']) else 50,  # RSI
-            latest_data['MACD'] if not pd.isna(latest_data['MACD']) else 0,  # MACD
-            latest_data['MACD_Signal'] if not pd.isna(latest_data['MACD_Signal']) else 0,  # MACDsign
-            latest_data['MACD_Histogram'] if not pd.isna(latest_data['MACD_Histogram']) else 0,  # MACDvol
-            latest_data['K'] if not pd.isna(latest_data['K']) else 50,  # K
-            latest_data['D'] if not pd.isna(latest_data['D']) else 50,  # D
-            latest_data['+DI'] if not pd.isna(latest_data['+DI']) else 25,  # +DI
-            latest_data['-DI'] if not pd.isna(latest_data['-DI']) else 25,  # -DI
-            latest_data['ADX'] if not pd.isna(latest_data['ADX']) else 25,  # ADX
+            rsi_value,  # RSI
+            macd_value,  # MACD
+            macd_signal_value,  # MACDsign
+            macd_hist_value,  # MACDvol
+            k_value,  # K
+            d_value,  # D
+            plus_di_value,  # +DI
+            minus_di_value,  # -DI
+            adx_value,  # ADX
             15,    # business_climate (手動填入值)
             46.7   # PMI (手動填入值)
         ]
@@ -312,13 +381,60 @@ def advanced_xgboost_predict(predict_days=5):
         
         input_df = pd.DataFrame([features_list], columns=column_names)
         
-        print("特徵數據準備完成:")
-        print(f"- 收盤價: {features_list[0]:.2f}")
-        print(f"- 成交量: {features_list[1]:,.0f}")
-        print(f"- 匯率: {features_list[2]:.2f}")
-        print(f"- 道瓊指數: {features_list[3]:.2f}")
-        print(f"- 新聞情緒: {features_list[8]:.4f}")
-        print(f"- RSI: {features_list[9]:.2f}")
+        # 詳細的資料驗證日誌
+        print("=" * 50)
+        print("XGBoost 模型輸入特徵詳細檢查報告")
+        print("=" * 50)
+        
+        # 基本市場數據
+        print("📊 基本市場數據:")
+        print(f"  收盤價 (close): {latest_data['Close']:.2f}")
+        print(f"  成交量 (volume): {latest_data['Volume']:,.0f}")
+        print(f"  匯率 (rate): {exchange_rate:.4f}")
+        
+        # 美股指數
+        print("\n🇺🇸 美股指數數據:")
+        for key, value in us_market.items():
+            status = "✅ 正常" if value > 0 else "⚠️ 可能異常(=0)"
+            print(f"  {key}: {value:.2f} {status}")
+        
+        # 新聞情緒
+        print(f"\n📰 新聞情緒 (NEWS): {sentiment_score_raw:.6f}")
+        if sentiment_score_raw == 0:
+            print("  ⚠️ 新聞情緒分數為0，可能無新聞數據")
+        else:
+            print("  ✅ 新聞情緒分數正常")
+        
+        # 技術指標詳細狀態
+        print("\n📈 技術指標狀態:")
+        for indicator, status in tech_indicators_status.items():
+            status_symbol = "✅" if status['is_real'] else "⚠️"
+            source_info = "實際計算值" if status['is_real'] else "預設替代值"
+            print(f"  {indicator}: {status['value']:.4f} {status_symbol} ({source_info})")
+        
+        # 手動填入數據
+        print("\n🔧 手動填入數據:")
+        print(f"  business_climate: 15 ✅")
+        print(f"  PMI: 46.7 ✅")
+        
+        # 統計資料完整性
+        real_indicators = sum(1 for status in tech_indicators_status.values() if status['is_real'])
+        total_indicators = len(tech_indicators_status)
+        completeness = (real_indicators / total_indicators) * 100
+        
+        print(f"\n📋 技術指標完整性統計:")
+        print(f"  實際計算指標: {real_indicators}/{total_indicators} ({completeness:.1f}%)")
+        if completeness < 80:
+            print("  ⚠️ 警告：超過20%的技術指標使用預設值，可能影響預測準確性")
+        else:
+            print("  ✅ 技術指標完整性良好")
+        
+        # 顯示完整輸入向量
+        print(f"\n🔢 完整特徵向量 (共{len(features_list)}個特徵):")
+        for i, (name, value) in enumerate(zip(column_names, features_list)):
+            print(f"  [{i:2d}] {name:15s}: {value:10.4f}")
+        
+        print("=" * 50)
         
         # 進行預測
         predictions = xgb_model.predict('xgboost_model', input_df)