Update app.py

app.py

@@ -8,8 +8,8 @@ import requests
 from bs4 import BeautifulSoup
 import pandas_ta as ta
 from pypfopt import EfficientFrontier, risk_models, expected_returns
-from prophet import Prophet
-from statsmodels.tsa.arima.model import ARIMA
+from typing import Dict
+import tensorflow as tf
 import warnings
 
 warnings.filterwarnings("ignore")
@@ -43,8 +43,6 @@ class EnhancedTechnicalAnalysis:
             if self.model:
                 preds = self.model.predict(self.data[['Close']].values)
                 self.data['DL_Prediction'] = preds.flatten()
-            else:
-                self.data['DL_Prediction'] = self._fallback_prediction()
             self.data['Volatility_Cluster'] = self._detect_volatility_clusters()
             return self.data.dropna(axis=1, how='all')
         except Exception as e:
@@ -55,34 +53,6 @@ class EnhancedTechnicalAnalysis:
         returns = self.data['Close'].pct_change()
         return (returns.rolling(20).std() > returns.std()).astype(int)
 
-    def _fallback_prediction(self) -> pd.Series:
-        # Fallback prediction using ARIMA
-        try:
-            model = ARIMA(self.data['Close'], order=(5, 1, 0))
-            model_fit = model.fit()
-            future_preds = model_fit.forecast(steps=5)
-            return pd.Series(future_preds, index=self.data.index[-5:])
-        except:
-            return pd.Series([self.data['Close'].iloc[-1]] * 5, index=self.data.index[-5:])
-
-    def predict_future_prices(self, n_days: int = 5) -> pd.DataFrame:
-        try:
-            if self.model is None:
-                st.warning("No deep learning model available for predictions. Using fallback method.")
-                preds = self._fallback_prediction()
-            else:
-                last_close = self.data['Close'].values[-1]
-                preds = []
-                for _ in range(n_days):
-                    pred = self.model.predict(np.array([[last_close]]))
-                    preds.append(pred[0][0])
-                    last_close = pred[0][0]  # Update last close for next prediction
-            future_dates = pd.date_range(start=self.data.index[-1] + timedelta(days=1), periods=n_days)
-            return pd.DataFrame({'Date': future_dates, 'Predicted_Close': preds})
-        except Exception as e:
-            st.error(f"Prediction failed: {str(e)}")
-            return pd.DataFrame()
-
 # Market Data Integrator
 class MarketDataIntegrator:
     """Enhanced market data integration with fixed market movers"""
@@ -93,6 +63,7 @@ class MarketDataIntegrator:
 
     def get_market_movers(self) -> Dict:
         try:
+            # Using Yahoo Finance API for more reliable data
             gainers = yf.Ticker("^GSPC").info.get('most_actively_traded', [])[:5]
             losers = yf.Ticker("^GSPC").info.get('most_actively_traded', [])[-5:]
             active = yf.Ticker("^GSPC").info.get('most_actively_traded', [])[:5]
@@ -231,7 +202,7 @@ class TradingAssistant:
 # Main Application
 def main():
     st.set_page_config(page_title="AI Trading Pro", layout="wide", page_icon="📈")
-    st.title("🚀 AI Trading Pro - Advanced Trading Solution")
+    st.title("🚀 AI Trading Pro - Enhanced Trading Solution")
     
     market_data = MarketDataIntegrator()
     trading_engine = TradingAssistant()
@@ -242,14 +213,12 @@ def main():
         asset_type = st.selectbox("Asset Class", list(trading_engine.asset_options.keys()))
         symbol = st.selectbox("Symbol", trading_engine.asset_options[asset_type])
         timeframe = st.select_slider("Analysis Period", options=["1D", "1W", "1M", "3M", "1Y"], value="1M")
-        prediction_days = st.slider("Prediction Horizon (Days)", 1, 30, 5)
-        capital = st.number_input("Initial Capital ($)", min_value=100, value=10000)
-        enable_portfolio = st.checkbox("Portfolio Optimization")
-        show_education = st.checkbox("Tutorial Mode")
-        risk_level = st.select_slider("Risk Profile", ["Low", "Medium", "High"])
-        contact_info = st.text("📞 Contact: +0335-xxxxxxx")
+        with st.expander("🔧 Advanced Tools"):
+            enable_portfolio = st.checkbox("Portfolio Optimization")
+            show_education = st.checkbox("Tutorial Mode")
+            risk_level = st.select_slider("Risk Profile", ["Low", "Medium", "High"])
 
-    tab_analysis, tab_prediction, tab_portfolio, tab_learn = st.tabs(["Analysis", "Prediction", "Portfolio", "Academy"])
+    tab_analysis, tab_portfolio, tab_learn = st.tabs(["Analysis", "Portfolio", "Academy"])
 
     with tab_analysis:
         if st.button("Run Analysis"):
@@ -259,6 +228,14 @@ def main():
                     if data.empty:
                         st.error("No data available for the selected symbol and timeframe.")
                         return
+                    analysis = trading_engine.generate_signal(symbol, data)
+                    movers = market_data.get_market_movers()
+                    sentiment = market_data.get_market_sentiment(symbol)
+                    col1, col2, col3, col4 = st.columns(4)
+                    col1.metric("Price", f"${data['Close'].iloc[-1]:.2f}")
+                    col2.metric("Signal", analysis['signal'])
+                    col3.metric("RSI", f"{analysis['details']['RSI']:.1f}")
+                    col4.metric("Sentiment", f"{sentiment:.2f}")
                     fig = go.Figure()
                     fig.add_trace(go.Candlestick(
                         x=data.index,
@@ -270,50 +247,21 @@ def main():
                     ))
                     fig.update_layout(height=600, xaxis_rangeslider_visible=False)
                     st.plotly_chart(fig, use_container_width=True)
-                    st.write("### Technical Indicators")
-                    analysis = EnhancedTechnicalAnalysis(data).calculate_all_indicators()
-                    st.dataframe(analysis.tail())
+                    with st.expander("📈 Market Movers"):
+                        cols = st.columns(3)
+                        cols[0].subheader("Top Gainers")
+                        cols[1].subheader("Top Losers")
+                        cols[2].subheader("Most Active")
+                        for i in range(5):
+                            if i < len(movers['gainers']):
+                                cols[0].write(f"{i+1}. {movers['gainers'][i]}")
+                            if i < len(movers['losers']):
+                                cols[1].write(f"{i+1}. {movers['losers'][i]}")
+                            if i < len(movers['active']):
+                                cols[2].write(f"{i+1}. {movers['active'][i]}")
                 except Exception as e:
                     st.error(f"Analysis failed: {str(e)}")
 
-    with tab_prediction:
-        if st.button("Generate Predictions"):
-            with st.spinner("Generating predictions..."):
-                try:
-                    data = yf.download(symbol, period="1y")
-                    if data.empty:
-                        st.error("No data available for the selected symbol.")
-                        return
-                    analyzer = EnhancedTechnicalAnalysis(data)
-                    future_prices = analyzer.predict_future_prices(prediction_days)
-                    if future_prices.empty:
-                        st.warning("No predictions available.")
-                        return
-                    fig = go.Figure()
-                    fig.add_trace(go.Scatter(
-                        x=data.index,
-                        y=data['Close'],
-                        mode='lines',
-                        name='Historical Prices'
-                    ))
-                    fig.add_trace(go.Scatter(
-                        x=future_prices['Date'],
-                        y=future_prices['Predicted_Close'],
-                        mode='lines',
-                        name='Predicted Prices'
-                    ))
-                    fig.update_layout(
-                        title="Historical and Predicted Prices",
-                        height=600,
-                        xaxis_title="Date",
-                        yaxis_title="Price"
-                    )
-                    st.plotly_chart(fig, use_container_width=True)
-                    st.write("### Future Price Predictions")
-                    st.dataframe(future_prices)
-                except Exception as e:
-                    st.error(f"Prediction failed: {str(e)}")
-
     with tab_portfolio:
         if enable_portfolio:
             st.subheader("⚖️ Portfolio Optimization")