R/ToDummyVariables.R

In citronmeliss/predictionpackr: Predictionpackr

#' To dummy variables
#'
#' This function transforms factor variables to dummy variables.
#' @param study.sample Type. Description. Default/No default.
#' @export
ToDummyVariables <- function(study.sample, not.feature.variables = c("composite", "tc", "s")) {
  ## Error handling
  if (!is.data.frame(study.sample))
    stop("study.sample must be a data frame")
  ## Transform the factor feature variables to dummy variables
  features <- study.sample[, !(names(study.sample) %in% not.feature.variables)]
  dummies <- caret::dummyVars(formula = ~., data = features)
  dummy.features <- predict(dummies, newdata = features)
  study.sample <- cbind(dummy.features, study.sample[, not.feature.variables])
  return(study.sample)
}