Machine Learning for OpenCV 4 - Second Edition

Testing the model

To test the generalization performance of the model, we calculate the mean squared error on the test data:

In [24]: y_pred = linreg.predict(X_test)In [25]: metrics.mean_squared_error(y_test, y_pred)Out[25]: 14.995852876582541

We note that the mean squared error is a little lower on the test set than the training set. This is good news, as we care mostly about the test error. However, from these numbers it is really hard to understand how good the model really is. Perhaps it's better to plot the data:

In [26]: plt.figure(figsize=(10, 6))...      plt.plot(y_test, linewidth=3, label='ground truth')...      plt.plot(y_pred, linewidth=3, label='predicted')...      plt.legend(loc='best')...      plt.xlabel('test data points')... plt.ylabel('target ...