rxDiscretize: Creates equal width or equal frequency discretized variables...

In RevoEnhancements/RevoEnhancements: Selected enhancements to Revolution Analytics RevoScaleR

Description

Creates equal width or equal frequency discretized variables from continuous variables.

Usage

rxDiscretize(formula = ~., data, type = c("width", "freq"), nBins = NULL,
  subscript = "disc", sep = "_", integerLabels = TRUE,
  finalTransform = TRUE)

Arguments

formula	Formula
data	xdf file
type	Determines whether the discretization produces equal width or equal freq results
nBins	Number of bins
subscript	Character vector. Gets appended to column names to indicate the discretized columns.
sep	Separator character between column names and subscript
integerLabels	???
finalTransform	??? #'

See Also

Other Data mining functions: rxMoments; rxSample2Df; rxTreeDiscretize; rxUnique

Examples

library(RevoScaleR)
discTransforms <- rxDiscretize(~ ., data = USArrests, integerLabels = FALSE)
discTransforms2 <- rxDiscretize(~ Assault+UrbanPop+Rape, data = USArrests, subscript = "", sep = "")

newData <- rxDataStep(inData = USArrests,
    transforms = discTransforms)
claimsXdf <- file.path(rxGetOption("sampleDataDir"),"claims.xdf")
# Equal Width
discTransforms <- rxDiscretize(~ cost, data = claimsXdf, type = "width", nBins = 1000, subscript = "disc", sep = "_")
x <- rxDataStep(inData = claimsXdf, transforms = discTransforms)

# Equal Freq
discTransforms <- rxDiscretize(~ cost, data = claimsXdf, type = "freq", nBins = 1000, subscript = "disc", sep = "_")
x <- rxDataStep(inData = claimsXdf, transforms = discTransforms)

RevoEnhancements/RevoEnhancements documentation built on May 9, 2019, 9:46 a.m.