iterative modeling with MLHO

iterative modeling with MLHO

Usage

mlho.it(
  dbmart,
  labels = labeldt,
  dems = NULL,
  test.sample = 30,
  MSMR.binarize = FALSE,
  MSMR.sparsity = 0.005,
  MSMR.jmi = TRUE,
  MSMR.topn = 200,
  MSMR.encounterLevel = FALSE,
  MSMR.valuesToMerge = FALSE,
  mlearn.save.model = FALSE,
  mlearn.note = "mlho_phewas run",
  mlearn.aoi = "demo",
  mlearn.cv = "cv",
  mlearn.nfold = 5,
  mlearn.calSHAP = FALSE,
  multicore = FALSE,
  preProc = TRUE,
  iterations = 5
)

Arguments

dbmart

dbmart table

labels

should be the labeldt table

dems

table containing the demographic variables

test.sample

put 20 if you want to use 20 percent for testing and 80 percent for training

MSMR.binarize

MSMR.lite parameter

MSMR.sparsity

MSMR.lite parameter

MSMR.jmi

MSMR.lite parameter

MSMR.topn

MSMR.lite parameter

MSMR.encounterLevel

MSMR.lite parameter

MSMR.valuesToMerge

MSMR.lite parameter

mlearn.save.model

mlearn parameter

mlearn.note

mlearn parameter

mlearn.aoi

mlearn parameter

mlearn.cv

mlearn parameter

mlearn.nfold

mlearn parameter

mlearn.calSHAP

mlearn parameter

multicore

if you want to parallelize the process

preProc

preprocessig on the train data or not

iterations

number of iterations you want. recommended at least 5. needs to be numeric