Exercise

Regression evaluation

The test_set and model objects that you have derived in the previous exercise are available in your environment.

It's useful to present the accuracy of predictions with one number. You can then easily compare several models and show the progress to your employer or future employer.

Root Mean Squared Error and Mean Absolute Error are widely used to evaluate the regression models. Recall that their formulas are:

\(RMSE = \sqrt{\frac{1}{n} \sum_{i=1}^{n}(y_i - \hat{y}_i)^2}\)

\(MAE = \frac{1}{n} \sum_{i=1}^{n} |y_i - \hat{y}_i|\)

Instructions 1/2

undefined XP

1
2

Assign the Hwt variable from the test set to y.
Calculate the predictions using the test set and assign them to y_hat.
Assign the number of rows of the test set to n.