grf (version 2.3.2)

tune_forest: Tune a forest

Description

Finds the optimal parameters to be used in training a forest.

Usage

tune_forest(
  data,
  nrow.X,
  ncol.X,
  args,
  tune.parameters,
  tune.parameters.defaults,
  tune.num.trees,
  tune.num.reps,
  tune.num.draws,
  train
)

Value

tuning output

Arguments

data

The data arguments (output from create_train_matrices) for the forest.

nrow.X

The number of observations.

ncol.X

The number of variables.

args

The remaining call arguments for the forest.

tune.parameters

The vector of parameter names to tune.

tune.parameters.defaults

The grf default values for the vector of parameter names to tune.

tune.num.trees

The number of trees in each 'mini forest' used to fit the tuning model.

tune.num.reps

The number of forests used to fit the tuning model.

tune.num.draws

The number of random parameter values considered when using the model to select the optimal parameters.

train

The grf forest training function.