R/environmentSetup.R

In tkolisnik/Rf2pval: A comprehensive approach to genomic analysis using scikit-learn's Random Forest models and rank-based feature reduction in R

#' Centralized Setup for Python Environment
#'
#' This function sets up the Python environment for the package. It imports essential Python
#' libraries using the `reticulate` package. This setup is used across various functions in the
#' package that require Python interoperability. This package depends on Python 3.9.18 being installed and the conda environment being loaded.
#'
#' @details
#' The function imports the following Python libraries:
#' \itemize{
#'   \item \code{sklearn}: Comprehensive machine learning library in Python.
#'   \item \code{numpy}: Fundamental package for scientific computing in Python.
#'   \item \code{shap}: Library for SHAP (SHapley Additive exPlanations) values, which explain the output of machine learning models.
#' }
#' These libraries are essential for various machine learning tasks and analyses within the package.
#' This centralized setup ensures consistency and reduces redundancy.
#'
#' @return
#' A list containing imported Python modules: \code{sklearn}, \code{numpy}, and \code{shap}.
#'
#' @examples
#' python_pkgs <- setup_python_pkgs()
#' # You can now use python_pkgs$sklearn, python_pkgs$numpy, and python_pkgs$shap in other functions
#' @import reticulate
#' @export
setup_python_pkgs <- function() {
  require(reticulate)
  sklearn <- reticulate::import("sklearn", convert = FALSE)
  numpy <- reticulate::import("numpy", convert = FALSE)
  shap <- reticulate::import("shap", convert = FALSE)
  return(list(sklearn = sklearn, numpy = numpy, shap = shap))
}