Update app.py

app.py

@@ -86,22 +86,22 @@ def predict_future_prices(ticker, periods=1825):
     pred_lstm = scaler.inverse_transform(np.array(pred_lstm).reshape(-1, 1))
     future_lstm = pd.DataFrame({'ds': future_dates[:len(pred_lstm)], 'yhat': pred_lstm.flatten()})
     
-
-    # XGBoost 모델 생성 및 학습
+  # XGBoost 모델 생성 및 학습
     model_xgb = XGBRegressor(n_estimators=100, learning_rate=0.1)
-    model_xgb.fit(X.values.reshape(-1, 1), y)
-    future_xgb = pd.DataFrame({'ds': future_dates, 'yhat': model_xgb.predict(X_future.values.reshape(-1, 1))})
+    model_xgb.fit(X.reshape(-1, 1), y)
+    future_xgb = pd.DataFrame({'ds': future_dates, 'yhat': model_xgb.predict(X_future)})
     
     # SVR 모델 생성 및 학습
     model_svr = SVR(kernel='rbf', C=1e3, gamma=0.1)
-    model_svr.fit(X.values.reshape(-1, 1), y)
-    future_svr = pd.DataFrame({'ds': future_dates, 'yhat': model_svr.predict(X_future.values.reshape(-1, 1))})
+    model_svr.fit(X.reshape(-1, 1), y)
+    future_svr = pd.DataFrame({'ds': future_dates, 'yhat': model_svr.predict(X_future)})
     
     # Bayesian Regression 모델 생성 및 학습
     model_bayes = BayesianRidge()
-    model_bayes.fit(X.values.reshape(-1, 1), y)
-    future_bayes = pd.DataFrame({'ds': future_dates, 'yhat': model_bayes.predict(X_future.values.reshape(-1, 1))})
+    model_bayes.fit(X.reshape(-1, 1), y)
+    future_bayes = pd.DataFrame({'ds': future_dates, 'yhat': model_bayes.predict(X_future)})
     
+
     # 예측 결과 그래프 생성
     forecast_prophet['ds'] = forecast_prophet['ds'].dt.strftime('%Y-%m-%d')
     fig = go.Figure()
@@ -116,6 +116,7 @@ def predict_future_prices(ticker, periods=1825):
     
     return fig, forecast_prophet[['ds', 'yhat', 'yhat_lower', 'yhat_upper']], future_lr[['ds', 'yhat']], future_arima[['ds', 'yhat']], future_lstm[['ds', 'yhat']], future_xgb[['ds', 'yhat']], future_svr[['ds', 'yhat']], future_bayes[['ds', 'yhat']]
 
+    
 # Gradio 인터페이스 설정 및 실행
 with gr.Blocks() as app:
     with gr.Row():