Finalize the model
Once you have executed the tuning process and found the best-performing hyperparameters, there are only two last steps to finalize your model: plug the winners into the dummy model specification and re-train your model on the whole training set using this final specification.
This way, your final model will be trained on the whole training set using the optimal hyperparameters.
Pre-loaded in your workspace are tune_results
from the last exercise, the untuned specification, boost_spec
, and the training data, customers_train
.
This exercise is part of the course
Machine Learning with Tree-Based Models in R
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Select the final hyperparameters
best_params <- ___
best_params