Tuning preparation
Tuning preparation is the foundation for tuning success. There are two main steps in preparing your tuning: marking hyperparameters using tune()
in the model specification and creating a grid of hyperparameters that is used in tuning.
You are going to execute these two fundamental steps of the tuning process in this exercise.
This exercise is part of the course
Machine Learning with Tree-Based Models in R
Exercise instructions
- Create a boosting specification with an
"xgboost"
engine for a classification model using 500 trees and mark the following parameters as tuning parameters:learn_rate
,tree_depth
, andsample_size
. Save the result asboost_spec
. - Build a regular tuning grid for the tuning parameters of
boost_spec
with three levels for each parameter.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create the specification with placeholders
boost_spec <- boost_tree(
trees = ___,
___,
___,
___) %>%
set_mode(___) %>%
set_engine(___)
# Create the tuning grid
tunegrid_boost <- ___(___,
levels = ___)
tunegrid_boost