Create Bayesian-Optimized AutoTuner for Random Forest

Creates an AutoTuner with ranger and omics-optimized search space.

Usage

make_autotuner_ranger(
  task,
  n_evals = NULL,
  inner_folds = 3,
  measure = "classif.auc",
  num_trees = 500L
)

Arguments

task

An mlr3 Task

n_evals

Number of Bayesian optimization iterations

inner_folds

Number of inner CV folds

measure

Performance measure

num_trees

Number of trees (fixed, usually not worth tuning)

Value

An AutoTuner object