plMonteCarlo: Monte Carlo Cross-Validation
In exprso: Rapid Deployment of Machine Learning Algorithms

Description Usage Arguments Details Value Examples

Perform Monte Carlo style cross-validation.

1 2	plMonteCarlo(array, B = 10, ctrlSS, ctrlFS = NULL, ctrlGS, ctrlMS = NULL, save = FALSE)

`array`	Specifies the `ExprsArray` object to undergo cross-validation.
`B`	A numeric scalar. The number of times to `split` the data.
`ctrlSS`	Arguments handled by `ctrlSplitSet`.
`ctrlFS`	A list of arguments handled by `ctrlFeatureSelect`.
`ctrlGS`	Arguments handled by `ctrlGridSearch`.
`ctrlMS`	Arguments handled by `ctrlModSet`. Optional.
`save`	A logical scalar. Toggles whether to save randomly split training and validation sets.

Analogous to how plGrid manages multiple build and predict tasks, one can think of plMonteCarlo as managing multiple pl tasks.

Specifically, plMonteCarlo will call the provided split function (via ctrlSS) some B times, perform all feature selection tasks (listed via ctrlFS) on each split of the data, and execute the pl function (via ctrlGS) using the bootstrapped set.

To perform multiple feature selection tasks, supply a list of multiple ctrlFeatureSelect argument wrappers to ctrlFS. To reduce the results of plMonteCarlo to a single performance metric, you can feed the returned ExprsPipeline object through the helper function calcMonteCarlo.

When embedding another plMonteCarlo or plNested call within this function (i.e., via ctrlGS), outer-fold model performance will force aucSkip = TRUE and plotSkip = TRUE.

An ExprsPipeline-class object.

## Not run: 
require(golubEsets)
data(Golub_Merge)
array <- arrayEset(Golub_Merge, colBy = "ALL.AML", include = list("ALL", "AML"))
array <- modFilter(array, 20, 16000, 500, 5) # pre-filter Golub ala Deb 2003
array <- modTransform(array) # lg transform
array <- modNormalize(array, c(1, 2)) # normalize gene and subject vectors
ss <- ctrlSplitSet(func = "splitStratify", percent.include = 67, colBy = NULL)
fs <- list(ctrlFeatureSelect(func = "fsStats", top = 0, how = "t.test"),
           ctrlFeatureSelect(func = "fsPrcomp", top = 50))
gs <- ctrlGridSearch(func = "plGrid", how = "buildSVM", top = c(2, 3, 4), fold = 10,
                     kernel = c("linear", "radial"), cost = 10^(-3:3), gamma = 10^(-3:3))
boot <- plMonteCarlo(array, B = 3, ctrlSS = ss, ctrlFS = fs, ctrlGS = gs)

## End(Not run)