Update app.py

app.py

@@ -19,7 +19,7 @@ def get_stock_data(ticker, start_date, end_date):
 
 # Function to preprocess data
 def preprocess_data(data):
-    data = data.fillna(method='ffill')  # Replace this with `data.ffill()` if needed
+    data = data.ffill()
     scaler = MinMaxScaler()
     scaled_data = scaler.fit_transform(data)
     scaled_data = pd.DataFrame(scaled_data, columns=data.columns)
@@ -41,7 +41,7 @@ def create_features(data):
         return rsi
 
     data['RSI'] = calculate_rsi(data, 14)
-    data = data.fillna(method='ffill')  # Replace this with `data.ffill()` if needed
+    data = data.ffill()
     data = data.fillna(0)  # Fill any remaining NaN values
     return data
 
@@ -65,8 +65,8 @@ class LSTMModel(nn.Module):
         self.fc = nn.Linear(hidden_size, output_size)
 
     def forward(self, x):
-        h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)
-        c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(device)
+        h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(x.device)
+        c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(x.device)
         out, _ = self.lstm(x, (h0, c0))
         out = self.fc(out[:, -1, :])
         return out
@@ -128,27 +128,48 @@ def predict_stock(ticker, start_date, end_date):
         # Inverse transform only the 'Close' column
         predicted_stock_price = scaler.inverse_transform(predicted_df)[:, 3]
 
-        # Plot actual vs. predicted stock prices
-        plt.figure(figsize=(14, 5))
-        plt.plot(stock_data['Close'].values, label='Actual Stock Price')
-        plt.plot(range(seq_length, len(predicted_stock_price) + seq_length), predicted_stock_price, label='Predicted Stock Price')
-        plt.legend()
-        plt.show()
+        # Calculate the latest actual price and predicted next price
+        latest_actual_price = stock_data['Close'].values[-1]
+        predicted_next_price = predicted_stock_price[-1]
 
-        return predicted_stock_price[-1]
+        # Determine whether to buy or sell the stock
+        recommendation = "Buy" if predicted_next_price > latest_actual_price else "Sell"
+        
+        # Calculate performance summary
+        three_years_ago_price = stock_data['Close'].values[0]
+        price_change = latest_actual_price - three_years_ago_price
+        percentage_change = (price_change / three_years_ago_price) * 100
+
+        return latest_actual_price, predicted_next_price, recommendation, price_change, percentage_change
 
     except ValueError as e:
-        return str(e)
+        return str(e), '', '', '', ''
     except Exception as e:
-        return f"An error occurred: {str(e)}"
+        return f"An error occurred: {str(e)}", '', '', '', ''
 
 # Gradio Interface
+tickers = ['AAPL', 'GOOGL', 'MSFT', 'AMZN', 'TSLA', 'FB', 'NFLX', 'NVDA', 'BRK.B', 'JPM']
 inputs = [
-    gr.Textbox(label="Stock Ticker"),
-    gr.Textbox(label="Start Date (YYYY-MM-DD)"),
-    gr.Textbox(label="End Date (YYYY-MM-DD)")
+    gr.Dropdown(choices=tickers, label="Stock Ticker"),
+    gr.Textbox(label="Start Date (YYYY-MM-DD)", type="text"),
+    gr.Textbox(label="End Date (YYYY-MM-DD)", type="text")
 ]
 
-outputs = gr.Textbox(label="Predicted Next Closing Price")
+outputs = [
+    gr.Textbox(label="Latest Actual Price"),
+    gr.Textbox(label="Predicted Next Closing Price"),
+    gr.Textbox(label="Recommendation"),
+    gr.Textbox(label="Price Change"),
+    gr.Textbox(label="Percentage Change")
+]
 
-gr.Interface(fn=predict_stock, inputs=inputs, outputs=outputs, title="Stock Price Prediction").launch()
+gr.Interface(
+    fn=predict_stock,
+    inputs=inputs,
+    outputs=outputs,
+    title="Stock Price Prediction",
+    examples=[
+        ["AAPL", "2020-01-01", "2023-01-01"],
+        ["GOOGL", "2020-01-01", "2023-01-01"]
+    ]
+).launch()