validate_RF: Compute accuracy metrics on test set for RF

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

Compute accuracy metrics on test set for RF

Description

Compute accuracy metrics on test set for RF

Usage

validate_RF(
  formula,
  data,
  rep = 10,
  Ncpu = 1,
  spatial = FALSE,
  seed = 123,
  method = "CV",
  return_fit = TRUE,
  ...
)

Arguments

formula	a model formula used to fit the data
data	the full dataset
rep	the number of cross validation replicates (default = 10)
Ncpu	the number of CPU cores to be used (default = 1)
spatial	either FALSE (default), or TRUE (to use matrix of distances as predictor)
seed	the seed used to control the reproducibility of the cross validation
method	either "CV" for cross-validation or "OOB" for directly using the out-of-bag observations generated when growing the forest
return_fit	whether to return the fit on all the data as attribute (default = FALSE)
...	additional parameters to be passed to ranger::ranger()

Value

a tibble with the CV/OOB replicates in row and accuracy metrics in columns

See Also

compute_metrics() for details on the function computing the values outputted here

Examples

validate_RF(staff_rangers_log ~ PA_area_log, data = data_test, method = "CV")
validate_RF(staff_rangers_log ~ PA_area_log, data = data_test,
                 rep = 1, num.trees = 10, method = "OOB")

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