Conduct a tidy random forest analysis with hyperparameter tuning via tidymodels framework

Usage

tidy_forest(
  data,
  outcome_var,
  drop_vars = NULL,
  split_prop = 0.5,
  num_threads = 6,
  importance = "permutation",
  mode = "classification",
  num_trees = 500,
  levels = 5,
  best_metric = "accuracy",
  ...
)

Arguments

data: A data frame
outcome_var: A string representing outcome variable
drop_vars: A vector of strings representing variables to drop
split_prop: A number representing proportion of data to use for training
num_threads: A number representing number of threads to use
importance: A string representing importance method
mode: A string representing mode
num_trees: A number representing number of trees
levels: A number representing number of levels
best_metric: A string representing best metric
...: Additional arguments to pass to parsnip::rand_forest()

Value

A list of objects