Model Compilation and Training
While compiling a model we provide these three essential parameters:
Python3
model.compile(optimizer='adam', loss='mean_squared_error') history = model.fit(x_train, y_train, epochs=10) |
Output:
Progress of model training epoch by epoch
For predicting we require testing data, so we first create the testing data and then proceed with the model prediction.
Python3
test_data = scaled_data[training - 60:, :] x_test = [] y_test = dataset[training:, :] for i in range(60, len(test_data)): x_test.append(test_data[i-60:i, 0]) x_test = np.array(x_test) x_test = np.reshape(x_test, (x_test.shape[0], x_test.shape[1], 1)) # predict the testing data predictions = model.predict(x_test) predictions = scaler.inverse_transform(predictions) # evaluation metrics mse = np.mean(((predictions - y_test) ** 2)) print("MSE", mse) print("RMSE", np.sqrt(mse)) |
Output:
2/2 [==============================] - 1s 13ms/step MSE 46.06080444818086 RMSE 6.786811066191607
Now that we have predicted the testing data, let us visualize the final results.
Python3
train = apple[:training] test = apple[training:] test['Predictions'] = predictions plt.figure(figsize=(10, 8)) plt.plot(train['Date'], train['Close']) plt.plot(test['Date'], test[['Close', 'Predictions']]) plt.title('Apple Stock Close Price') plt.xlabel('Date') plt.ylabel("Close") plt.legend(['Train', 'Test', 'Predictions']) |
Output:
Prediction for Stock Prices of Apple
Stock Price Prediction Project using TensorFlow
In this article, we shall build a Stock Price Prediction project using TensorFlow. Stock Market price analysis is a Timeseries approach and can be performed using a Recurrent Neural Network. To implement this we shall Tensorflow. Tensorflow is an open-source Python framework, famously known for its Deep Learning and Machine Learning functionalities. Building Neural Networks becomes easy by writing just a few lines of Tensorflow code.
Contact Us