Changing the number of hyperparameters to tune
When we examine the model object closely, we can see that caret
already did some automatic hyperparameter tuning for us: train
automatically creates a grid of tuning parameters. By default, if p
is the number of tuning parameters, the grid size is 3^p. But we can also specify the number of different values to try for each hyperparameter.
The data has again been preloaded as bc_train_data
. The libraries caret
and tictoc
have also been preloaded.
This exercise is part of the course
Hyperparameter Tuning in R
Exercise instructions
- Test four different values for each hyperparameter with automatic tuning in
caret
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Set seed.
set.seed(42)
# Start timer.
tic()
# Train model.
gbm_model <- train(diagnosis ~ .,
data = bc_train_data,
method = "gbm",
trControl = trainControl(method = "repeatedcv", number = 5, repeats = 3),
verbose = FALSE,
___)
# Stop timer.
toc()