flowForest: Train a flowForest model
In flowforest/flowforest: Random Forest for Flow Cytometry Results

Train a flowForest model

flowForest(
  x,
  y,
  trainMask = 1:x[2],
  testMask = (1:x[2])[!(1:x[2]) %in% trainMask],
  nOfTrees = 100,
  gateTries = 5,
  gateMode = "pseudosphere",
  asymmetric = TRUE,
  keepForest = TRUE,
  verbose = TRUE,
  relaxedGateAccept = FALSE
)

`x`	Flow cytometry measurements, in a form of an object of the class `tubeset`, as that created by the function `loadTubeset`.
`y`	Decision; must be a logical vector of a same length as the number of objects in `x`. Unknown decisions may be marked with `NA`; they can't be in the training set, though
`trainMask`	A vector of indices of objects in tubeset which are to be assumed to reside in the trainig set. By default the whole set
`testMask`	A vector of indices of objects in tubeset which are to be assumed to reside in the test set. By default those not in `trainMask`
`nOfTrees`	Number of trees to grow
`gateTries`	Number of gates to try at each split; this does not count gates rejected due to insufficient event fraction
`gateMode`	Mode of the gate generation; only `pseudosphere` is currently supported. See `makeRandomGate` for more information.
`asymmetric`	If `TRUE`, only gates which hold on average more events for the `TRUE` class will be taken into account. This is useful to force flowForest to focus on event populations characteristic for the `TRUE` class.
`keepForest`	Shall the forest structure be saved? This is required for importance and predicting new data, but requires more memory.
`verbose`	If `TRUE`, performance scores of `flowForest` counted on test set and OOB will be printed after growing each tree. Otherwise, only after building whole `flowForest`.
`relaxedGateAccept`	Accept gates with some event, not only at least one event for all objects.