Update README.md

Author: JordiCorbilla

README.md

@@ -580,25 +580,79 @@ Now it is time to prepare our testing data and send it through our deep-learning
 First, we need to import the test data using the same approach we used for the training data using the time steps:
 
 ```python
-    # Testing Data Transformation
-    x_test = []
-    y_test = []
-    for i in range(time_steps, test_scaled.shape[0]):
-        x_test.append(test_scaled[i - time_steps:i])
-        y_test.append(test_scaled[i, 0])
-
-    x_test, y_test = np.array(x_test), np.array(y_test)
-    x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 1))
+def infer_data():
+    print(tf.version.VERSION)
+    inference_folder = os.path.join(os.getcwd(), RUN_FOLDER)
+    stock = StockPrediction(STOCK_TICKER, STOCK_START_DATE, STOCK_VALIDATION_DATE, inference_folder, GITHUB_URL, EPOCHS, TIME_STEPS, TOKEN, BATCH_SIZE)
+
+    data = StockData(stock)
+
+    (x_train, y_train), (x_test, y_test), (training_data, test_data) = data.download_transform_to_numpy(TIME_STEPS, inference_folder)
+    min_max = data.get_min_max()
+
+    # load future data
+    print('Latest Stock Price')
+    latest_close_price = test_data.Close.iloc[-1]
+    latest_date = test_data[-1:]['Close'].idxmin()
+    print(latest_close_price)
+    print('Latest Date')
+    print(latest_date)
+
+    tomorrow_date = latest_date + timedelta(1)
+    # Specify the next 300 days
+    next_year = latest_date + timedelta(TIME_STEPS * 100)
+
+    print('Future Date')
+    print(tomorrow_date)
+
+    print('Future Timespan Date')
+    print(next_year)
+
+    x_test, y_test, test_data = data.generate_future_data(TIME_STEPS, min_max, tomorrow_date, next_year, latest_close_price)
+
+    # Check if the future data is not empty
+    if x_test.shape[0] > 0:
+        # load the weights from our best model
+        model = tf.keras.models.load_model(os.path.join(inference_folder, 'model_weights.h5'))
+        model.summary()
+
+        # perform a prediction
+        test_predictions_baseline = model.predict(x_test)
+        test_predictions_baseline = min_max.inverse_transform(test_predictions_baseline)
+        test_predictions_baseline = pd.DataFrame(test_predictions_baseline, columns=['Predicted_Price'])
+
+        # Combine the predicted values with dates from the test data
+        predicted_dates = pd.date_range(start=test_data.index[0], periods=len(test_predictions_baseline))
+        test_predictions_baseline['Date'] = predicted_dates
+        
+        # Reset the index for proper concatenation
+        test_data.reset_index(inplace=True)
+        
+        # Concatenate the test_data and predicted data
+        combined_data = pd.concat([test_data, test_predictions_baseline], ignore_index=True)
+        
+        # Plotting predictions
+        plt.figure(figsize=(14, 5))
+        plt.plot(combined_data['Date'], combined_data.Close, color='green', label='Simulated [' + STOCK_TICKER + '] price')
+        plt.plot(combined_data['Date'], combined_data['Predicted_Price'], color='red', label='Predicted [' + STOCK_TICKER + '] price')
+        plt.xlabel('Time')
+        plt.ylabel('Price [USD]')
+        plt.legend()
+        plt.title('Simulated vs Predicted Prices')
+        plt.savefig(os.path.join(inference_folder, STOCK_TICKER + '_future_comparison.png'))
+        plt.show()
+    else:
+        print("Error: Future data is empty.")
+
+start_date = pd.to_datetime('2017-01-01')
+end_date = datetime.today()
+duration = end_date - start_date
+STOCK_VALIDATION_DATE = start_date + 0.8 * duration
 ```
 
-Now we can call the predict method which will allow us to generate the stock prediction based on the training done over the training data. As a result, we will generate a csv file that contains the result of the prediction and also a chart that shows what's the real vs the estimation. 
+Now we can call the `infer_data` method which will allow us to generate the stock prediction based on the training done over the training data. As a result, we will generate a csv file that contains the result of the prediction and also a chart that shows what's the real vs the estimation. 
 
-![](https://github.com/JordiCorbilla/stock-prediction-deep-neural-learning/raw/master/Alphabet%20Inc_prediction.png)
-
-With the validation loss and validation MSE metrics:
-
-![](https://github.com/JordiCorbilla/stock-prediction-deep-neural-learning/raw/master/loss.png)
-![](https://github.com/JordiCorbilla/stock-prediction-deep-neural-learning/raw/master/MSE.png)
+![image](https://github.com/JordiCorbilla/stock-prediction-deep-neural-learning/assets/7347994/36ce113b-acf4-4245-bf1e-771e77aeb0a1)
 
 # 4) Usage