sc_classAssess: assess the classifiers performance based on validation data
In pcahan1/singleCellNet: cell identity from single cell RNA-Seq data

assess the classifiers performance based on validation data

1
2
3

sc_classAssess(stDat, washedDat, dLevel = "description1",
  dLevelSID = "sample_name", minCells = 40, dThresh = 0,
  propTrain = 0.25, nRand = 50, nTrees = 2000, resolution = 0.005)

`stDat`	sample table
`dThresh`	detection threshold
`propTrain`	the proportion of the training data desire
`nRand`	the number of random sample one wants to generate
`nTrees`	number of branches one would like to build on the random forest classifier
`expDat`	normalized expression matrix
`nimCells`	the minimal number of cells one would like to have in each cell type