R/TuneResult.R

In Najah-lshanableh/R-data-mining2: Machine Learning in R

#' @title Result of tuning.
#'
#' @description
#' Container for results of hyperparameter tuning.
#' Contains the obtained point in search space, its performance values
#' and the optimization path which lead there.
#'
#' Object members:
#' \describe{
#' \item{learner [\code{\link{Learner}}]}{Learner that was optimized.}
#' \item{control [\code{\link{TuneControl}}]}{Control object from tuning.}
#' \item{x [\code{list}]}{Named list of hyperparameter values identified as optimal.
#'   Note that when you have trafos on some of your params, \code{x} will always be
#'   on the TRANSFORMED scale so you directly use it.}
#' \item{y [\code{numeric}]}{Performance values for optimal \code{x}.}
#' \item{threshold [\code{numeric}]}{Vector of finally found and used thresholds
#'   if \code{tune.threshold} was enabled in \code{\link{TuneControl}}, otherwise not present and
#'   hence \code{NULL}.}
#' \item{opt.path [\code{\link[ParamHelpers]{OptPath}}]}{Optimization path which lead to \code{x}.
#'   Note that when you have trafos on some of your params, the opt.path always contains the
#'   UNTRANSFORMED values on the original scale. You can simply call \code{trafoOptPath(opt.path)} to
#'   transform them, or, \code{as.data.frame{trafoOptPath(opt.path)}}.
#'   If mlr option \code{on.error.dump} is \code{TRUE}, \code{OptPath} will have a \code{.dump} object
#'   in its \code{extra} column which contains error dump traces from failed optimization evaluations.
#'   It can be accessed by \code{getOptPathEl(opt.path)$extra$.dump}.}
#' }
#' @name TuneResult
#' @rdname TuneResult
NULL
makeTuneResult = function(learner, control, x, y, threshold, opt.path, ...) {
  makeOptResult(learner, control, x, y, threshold, opt.path, "TuneResult", ...)
}

makeTuneResultFromOptPath = function(learner, par.set, measures, control, opt.path) {
  i = getOptPathBestIndex(opt.path, measureAggrName(measures[[1]]), ties = "random")
  e = getOptPathEl(opt.path, i)
  x = trafoValue(par.set, e$x)
  x = removeMissingValues(x)
  threshold = if (control$tune.threshold) getThresholdFromOptPath(opt.path, i) else NULL
  makeTuneResult(learner, control, x, e$y, threshold, opt.path)
}


#'@export
print.TuneResult = function(x, ...) {
  catf("Tune result:")
  catf("Op. pars: %s", paramValueToString(x$opt.path$par.set, x$x))
  if (!is.null(x$threshold))
    catf("Threshold: %s", collapse(sprintf("%2.2f", x$threshold)))
  catf("%s", perfsToString(x$y))
}