Evaluate the ensemble
So far, so good. But how good exactly? Prove your model evaluation skills by cross-validating your out-of-sample AUC!
The specification boost_spec and the customers_train tibble are still loaded.
This exercise is part of the course
Machine Learning with Tree-Based Models in R
Exercise instructions
- Create five CV folds of your training set and save them as
folds. - Fit and evaluate a model that predicts
still_customerfor every fold, using your specification, all predictor variables, and the AUC metric. - Collect the metrics of
cv_resultsand check the mean AUC.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
set.seed(99)
# Create CV folds
folds <- ___
# Fit and evaluate models for all folds
cv_results <- fit_resamples(___,
___,
resamples = ___,
___)
# Collect cross-validated metrics
___