iterVarSelRF: iterVarSelRF

In USEPA/LakeTrophicModelling: Package to reproduce Hollister et al. (2015) Modeling Lake Trophic State: A Randrom Forest Approach.

Description

This function iterates the 'varselRF' function to capture variability in the possibly selected variables. This process forms the basis of the analysis conducted in Hollister, Kreakie, and Milsted (2014). We us the proportion of iterations that a given variable is selected as a variable selection metric. A parallel implementation is possible using 'snowfall'. User only need specify the number of cores to use. Function relies on knitr,snowfall,parallel, and varSelRF

Usage

iterVarSelRF(
  indVar,
  depVar,
  numRun = 10,
  numCore = 1,
  outStr = tempfile(tmpdir = getwd()),
  ntree = 10,
  ntreeIterat = 5,
  vars.drop.frac = NULL,
  vars.drop.num = 1,
  time = FALSE
)

Arguments

indVar	independent variables
depVar	dependent variable
numRun	number of iterations. default is 10
numCore	number of cores to split runs across. defualt is 1
outStr	string to append to output files. Defaults to a tmp file
time	logical true will output overall time for the run.

Examples

data(LakeTrophicModelling)
x1<-iterVarSelRF(ltmData[predictors_gis],ltmData$TS_CHLA_4,10,ntree=10,ntreeIterat=5,
vars.drop.frac=NULL,vars.drop.num=1,time=TRUE)
x2<-iterVarSelRF(ltmData[predictors_gis],ltmData$TS_CHLA_4,10,4,ntree=10,ntreeIterat=5,
vars.drop.frac=NULL,vars.drop.num=1,time=TRUE)

USEPA/LakeTrophicModelling documentation built on Oct. 15, 2020, 4:13 p.m.