consensusSubtypingTrain

apply several clustering methods and append results to a dataframe; usually will include one subject per row

consensusSubtypingTrain(
  dataToClust,
  featureNames,
  clustVec,
  ktrain,
  reorderingVariable,
  mvcl = "MVST",
  ksearch = 3,
  verbose = FALSE
)

Arguments

dataToClust

dataframe input that contains the relevant variables (may have others as well)

featureNames

names to use in the clustering

clustVec

names of the clustering methods to use

ktrain

the number of clusters

reorderingVariable

the name of the column to use to reorder the cluster names

mvcl

character prefix for the new cluster column names

ksearch

the cluster number(s) for clustering of the methods by concordance

verbose

boolean

Value

a list with newdata: dataframe with new variables attached; models contains the trained models; reorderers contains a dataframe for reordering cluster levels; quality measurements and clustering based on cluster concordance (adjusted rand index) are also returned.

Author

Avants BB

Examples

mydf = generateSubtyperData( 100 )