Update app.py

app.py

@@ -20,7 +20,7 @@ def stock_prediction_app(ticker, start_date, end_date):
         return "No data available for the selected date range.", None
     
     # Prepare the data for LSTM model
-    df_close = stock_data[['Close']]  # Use only the 'Close' column for prediction
+    df_close = stock_data[['Close', 'Open', 'High', 'Low', 'Volume']]
     scaler = MinMaxScaler(feature_range=(0, 1))
     scaled_data = scaler.fit_transform(df_close)
 
@@ -28,19 +28,21 @@ def stock_prediction_app(ticker, start_date, end_date):
     def create_dataset(data, time_step=60):
         X_train, y_train = [], []
         for i in range(len(data)-time_step-1):
-            X_train.append(data[i:(i+time_step), 0])
-            y_train.append(data[i + time_step, 0])
+            X_train.append(data[i:(i+time_step), :])
+            y_train.append(data[i + time_step, 0])  # Predicting the 'Close' price
         return np.array(X_train), np.array(y_train)
     
     X_train, y_train = create_dataset(scaled_data)
 
     # Reshape the data for LSTM [samples, time steps, features]
-    X_train = X_train.reshape(X_train.shape[0], X_train.shape[1], 1)
+    X_train = X_train.reshape(X_train.shape[0], X_train.shape[1], X_train.shape[2])  # (samples, time steps, features)
     
-    # Define LSTM model
+    # Define LSTM model with dropout for regularization
     lstm_model = tf.keras.Sequential([
-        tf.keras.layers.LSTM(50, return_sequences=True, input_shape=(X_train.shape[1], 1)),
-        tf.keras.layers.LSTM(50, return_sequences=False),
+        tf.keras.layers.LSTM(100, return_sequences=True, input_shape=(X_train.shape[1], X_train.shape[2])),
+        tf.keras.layers.Dropout(0.2),
+        tf.keras.layers.LSTM(100, return_sequences=False),
+        tf.keras.layers.Dropout(0.2),
         tf.keras.layers.Dense(25),
         tf.keras.layers.Dense(1)
     ])
@@ -48,16 +50,16 @@ def stock_prediction_app(ticker, start_date, end_date):
     # Compile the model
     lstm_model.compile(optimizer='adam', loss='mean_squared_error')
     
-    # Train the model
-    lstm_model.fit(X_train, y_train, batch_size=1, epochs=1, verbose=0)
+    # Train the model with more epochs
+    lstm_model.fit(X_train, y_train, batch_size=32, epochs=50, verbose=0)
 
     # Predict on the same data (just for demonstration)
     predictions = lstm_model.predict(X_train)
-    predictions = scaler.inverse_transform(predictions)  # Convert back to original scale
+    predictions = scaler.inverse_transform(np.concatenate((predictions, np.zeros((predictions.shape[0], scaled_data.shape[1]-1))), axis=1))[:, 0]  # Convert back to original scale
 
     # Create a plot to show predictions
     plt.figure(figsize=(10, 5))
-    plt.plot(df_close.values, label='Actual Stock Price', color='blue')
+    plt.plot(df_close['Close'].values, label='Actual Stock Price', color='blue')
     plt.plot(predictions, label='Predicted Stock Price', color='orange')
     plt.title(f'{ticker} Stock Price Prediction')
     plt.xlabel('Days')