cvGrid: Grid Search via K-fold Cross Validation
In SkadiEye/ITRlearn: Individualized Treatment Rule Learning

Description Usage Arguments Value See Also

Grid Search for hyper-parameters in a reg or wcls model via K-fold Cross Validation.

cvGrid(trainObj, k.fold = 10, gsCtrl, cutoff = 0.5, criteria = c("acc",
  "sens", "spec", "auc", "ppv", "npv", "mse", "w.acc", "a.acc", "val.f",
  NA)[ifelse("RegObj" %in% class(trainObj), 7, 4)], weight = rep(1,
  length(trainObj@X)[1]), inclusion = rep(TRUE, length(trainObj@X)[1]),
  print.fold = FALSE, progress = TRUE, all.models = FALSE)

`trainObj`	A `RegObj` or `wClsObj` containing the training set.
`k.fold`	The number of folds in the cross validation.
`gsCtrl`	A `list` of arguments that specifies a regression or classification method, handled by `gsCtrlPanel`.
`cutoff`	A `numeric` value used as a cutoff to calculate accuracy when the outcome is bianry.
`criteria`	A criteria to decide for the best tuning parameters. If NA, then the best model will not be picked.
`weight`	A `numeric` vector indicating the weight of all samples. Used for the 'w.acc' criteria.
`inclusion`	A `logical` vector indicating whether a sample shoule be included in the performance evaluation.
`print.fold`	A `logical` value indicating whether each fold should be reported while model fitting.
`progress`	A `logical` value indicating whether a progress bar would be used to show the progress.
`all.models`	A `logical` value indicating whether to save all fitted models.