cn_splitMakeAssess: make classifiers and assess

In pcahan1/CellNet: Assess and improve cell fate engineering with network biology

make classifiers and assess

Description

split the data by prop, make classifiers, assess them on held out data

Usage

cn_splitMakeAssess(stDat, expDat, grns, prop = 0.5,
  dLevel = "description1", dLevelStudy = "exp_id",
  dLevelSID = "sample_id")

Arguments

stDat	sample table
expDat	expression data
grns	result of running cn_grnDoRock
prop	numeric 0<prop<1 what proportion of data to train classifiers on
dLevel	sample table column to use as response value in classifier (e.g. cell type)
dLevelStudy	column name to indicate experiment or study id
dLevelSID	column to indicate sample

Value

list if classifiers, classifier result as applied to held out data, ROCs

Examples

assessmentResults<-cn_splitMakeAssess(stTrain, expTrain, ctGRNs, prop=.5, dLevel="description1")

pcahan1/CellNet documentation built on May 18, 2023, 4:58 p.m.