8.63 Evaluate the model

… and the rmse is a little high in this case which is properly because we didn’t tune any of the parameters.

augment(boost_fit, new_data = carseats_train) %>%
  rmse(truth = Sales, estimate = .pred)

## # A tibble: 1 × 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 rmse    standard     0.00162

augment(boost_fit, new_data = carseats_test) %>%
  rmse(truth = Sales, estimate = .pred)

## # A tibble: 1 × 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 rmse    standard        1.38

Training RMSE: 0.00162 Testing RMSE: 1.38 (definitively overfitting)