Update app.py

app.py

@@ -154,7 +154,19 @@ def generate_signals(data, news_sentiment):
     data = calculate_indicators(data)
     data.dropna(inplace=True)
 
-    features = data[['RSI', 'Stochastic', 'MA', 'CCI', 'MACD', 'news_sentiment']]
+    # Check if the required columns exist in the data DataFrame
+    required_cols = ['RSI', 'Stochastic', 'MA', 'CCI', 'MACD', 'news_sentiment']
+    if all(col in data.columns for col in required_cols):
+        features = data[required_cols]
+    else:
+        # Handle the case when the required columns are not present
+        features = pd.DataFrame(index=data.index)
+        for col in required_cols:
+            if col in data.columns:
+                features[col] = data[col]
+            else:
+                features[col] = pd.Series([None] * len(data), index=data.index)
+
     labels = ((data['RSI'] < 30) & (data['Stochastic'] < 20)).astype(int) - ((data['RSI'] > 70) & (data['Stochastic'] > 80)).astype(int)
 
     X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=0.2, random_state=42)
@@ -167,8 +179,8 @@ def generate_signals(data, news_sentiment):
 
     if 0.8 <= accuracy <= 1.0:
         latest_data = data.iloc[-1]
-        prediction = model.predict([latest_data[['RSI', 'Stochastic', 'MA', 'CCI', 'MACD', 'news_sentiment']].values])
-        confidence = model.predict_proba([latest_data[['RSI', 'Stochastic', 'MA', 'CCI', 'MACD', 'news_sentiment']].values])[0][abs(prediction[0])]
+        prediction = model.predict([latest_data[required_cols].values])
+        confidence = model.predict_proba([latest_data[required_cols].values])[0][abs(prediction[0])]
 
         if prediction[0] == 1:
             buy_signal = (latest_data.name, latest_data['Close'], latest_data['Close'] * 0.95, "High Risk", confidence)
@@ -177,10 +189,11 @@ def generate_signals(data, news_sentiment):
 
     # Ensure at least one signal is displayed
     if buy_signal is None and sell_signal is None:
-        if data['RSI'].iloc[-1] < 30 and data['Stochastic'].iloc[-1] < 20:
-            buy_signal = (data.index[-1], data['Close'].iloc[-1], data['Close'].iloc[-1] * 0.95, "Low Confidence", 0.5)
-        elif data['RSI'].iloc[-1] > 70 and data['Stochastic'].iloc[-1] > 80:
-            sell_signal = (data.index[-1], data['Close'].iloc[-1], data['Close'].iloc[-1] * 1.05, "Low Confidence", 0.5)
+        if 'RSI' in data.columns and 'Stochastic' in data.columns:
+            if data['RSI'].iloc[-1] < 30 and data['Stochastic'].iloc[-1] < 20:
+                buy_signal = (data.index[-1], data['Close'].iloc[-1], data['Close'].iloc[-1] * 0.95, "Low Confidence", 0.5)
+            elif data['RSI'].iloc[-1] > 70 and data['Stochastic'].iloc[-1] > 80:
+                sell_signal = (data.index[-1], data['Close'].iloc[-1], data['Close'].iloc[-1] * 1.05, "Low Confidence", 0.5)
 
     return buy_signal, sell_signal