8.60 Evaluate the model

This model has similar performance compared to the bagging model.

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

## # A tibble: 1 × 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 rmse    standard       0.858

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

## # A tibble: 1 × 3
##   .metric .estimator .estimate
##   <chr>   <chr>          <dbl>
## 1 rmse    standard        1.36

Training RMSE: 0.858 Testing RMSE: 1.36 (still overfitting)

We can likewise plot the true value against the predicted value.

augment(rf_fit, new_data = carseats_test) %>% 
     ggplot(aes(Sales, .pred)) + 
     geom_abline() + 
     geom_point(alpha = 0.5)