R/validate.R

In jr-packages/jrPredictive: Predictive Analytics by Jumping Rivers

#' A function for validating models for submission during the predictive analytics course
#' 
#' @param model A model resulting from a call to train
#' @return A model of type train
#' @export
validate = function(model = NULL){
  if(is.null(model)) stop("No model provided")
  if(!(inherits(model,"train"))) stop("The model is not an object returned by train")
  
  ## Don't mess up global environment
  env = new.env()
  data(FuelEconomy, package = "AppliedPredictiveModeling", envir = env)
  cars2011 = env$cars2011
  
  test = tryCatch(predict(model, cars2011[1,]), error = function(e) e)
  if(inherits(test, "error")) stop("Your model can not successfully predict the test evaluation")
  message("Success, mark your model using the mark function")
  class(model) = c(class(model),"validated")
  model
}