Bootstrap random forest analysis

This function creates one or more random forest models from a spreadsheet of MLST profiles. In many situations, you would want to make many random forest models and make predictions from each one downstream.

Usage

bootstrapping(
  input,
  output,
  ncores = 1L,
  bootstrap_reps = 1L,
  loci_start_with = "LMO",
  my_seed = 23L
)

Arguments

input

(character) The input file path for MLST profile data in csv or csv.gz format.

output

(character) The output directory for random forest models.

ncores

(integer, default: 1L) The number of cores to use for parallel processing.

bootstrap_reps

(integer, default: 1L) The number of bootstrap replicates.

loci_start_with

(character, default: "LMO") The prefix for loci names.

my_seed

(integer, default: 23L) The seed for reproducibility.

Value

my_filenames A list of random forest filenames

Examples

if (FALSE) {
# Example usage:
model_filenames <- bootstrapping(
                      input = "tests/testthat/isolates_original_plus_new_dec_1_2021.csv.gz",
                      output = "results/",
                      ncores = 4, 
                      bootstrap_reps = 100)
}