finetune_RF: Fine tune the parameters of a random forest

In courtiol/rangeRinPA: Protected Area Personnel and Ranger Numbers are Insufficient to Deliver Global Expectations

Fine tune the parameters of a random forest

Description

Fine tune the parameters of a random forest

Usage

finetune_RF(
  values_to_try,
  param_to_tune,
  formula,
  data,
  rep = 10,
  Ncpu = 1,
  target = "staff_rangers_log",
  fn = mean,
  ...
)

finetune_RF_grid(grid, formula, data, rep = 10, Ncpu = 1, fn = mean, ...)

Arguments

values_to_try	the parameter value to try
param_to_tune	the quoted name of the parameter to study
formula	the formula for the forest
data	the full dataset
rep	the number of cross validation replicates (default = 10)
Ncpu	the number of CPU cores to be used (default = 1)
target	the name of the response variable
fn	the function for aggregating all metrics but p-values
...	additional parameters to be passed to ranger::ranger()
grid	a data frame defining the combination of parameters to try

Value

a list with a data frame and a plot

Functions

• finetune_RF_grid(): Fine tune the parameters of a random forest all at once

courtiol/rangeRinPA documentation built on Sept. 29, 2022, 9:54 a.m.