caret: Interfaces for caret package for data science pipelines.

In intubate: Interface to Popular R Functions for Data Science Pipelines

Description

Interfaces to caret functions that can be used in a pipeline implemented by magrittr.

Usage

ntbt_avNNet(data, ...)
ntbt_bagEarth(data, ...)
ntbt_bagFDA(data, ...)
ntbt_calibration(data, ...)
ntbt_dummyVars(data, ...)
ntbt_icr(data, ...)
ntbt_knn3(data, ...)
ntbt_lift(data, ...)
ntbt_pcaNNet(data, ...)
ntbt_sbf(data, ...)
ntbt_train(data, ...)

Arguments

data	data frame, tibble, list, ...
...	Other arguments passed to the corresponding interfaced function.

Details

Interfaces call their corresponding interfaced function.

Value

Object returned by interfaced function.

Author(s)

Roberto Bertolusso

Examples

## Not run: 
library(intubate)
library(magrittr)
library(caret)


## ntbt_avNNet: Neural Networks Using Model Averaging
## Not found example using formula interface, and I am
## completely ignorant to construct one.
data(BloodBrain)
BB <- list(bbbDescr, logBBB)

## Original function to interface
avNNet(bbbDescr, logBBB, size = 5, linout = TRUE, trace = FALSE)

## The interface puts data as first parameter
ntbt_avNNet(BB, bbbDescr, logBBB, size = 5, linout = TRUE, trace = FALSE)

## so it can be used easily in a pipeline.
BB %>%
  ntbt_avNNet(bbbDescr, logBBB, size = 5, linout = TRUE, trace = FALSE)


## ntbt_bagEarth: Bagged Earth

## Original function to interface
bagEarth(Volume ~ ., data = trees)

## The interface puts data as first parameter
ntbt_bagEarth(trees, Volume ~ .)

## so it can be used easily in a pipeline.
trees %>%
  ntbt_bagEarth(Volume ~ .)


## ntbt_bagFDA: Bagged FDA
library(mlbench)
library(earth)
data(Glass)

set.seed(36)
inTrain <- sample(1:dim(Glass)[1], 150)

trainData <- Glass[ inTrain, ]
testData  <- Glass[-inTrain, ]
## Original function to interface
## bagFDA(Type ~ ., trainData)   ## There is an error:
## Error in requireNamespaceQuietStop("mda") : package mda is required
##                               ## even when mda is installed
## For now all of this stays commented.

## The interface puts data as first parameter
## ntbt_bagFDA(trainData, Type ~ .)

## so it can be used easily in a pipeline.
## trainData %>%
##   ntbt_bagFDA(Type ~ .)


## ntbt_calibration: Probability Calibration Plot
data(mdrr)
mdrrDescr <- mdrrDescr[, -nearZeroVar(mdrrDescr)]
mdrrDescr <- mdrrDescr[, -findCorrelation(cor(mdrrDescr), .5)]
inTrain <- createDataPartition(mdrrClass)
trainX <- mdrrDescr[inTrain[[1]], ]
trainY <- mdrrClass[inTrain[[1]]]
testX <- mdrrDescr[-inTrain[[1]], ]
testY <- mdrrClass[-inTrain[[1]]]
library(MASS)
ldaFit <- lda(trainX, trainY)
qdaFit <- qda(trainX, trainY)
testProbs <- data.frame(obs = testY,
lda <- predict(ldaFit, testX)$posterior[,1],
qda <- predict(qdaFit, testX)$posterior[,1])

## Original function to interface
calPlotData <- calibration(obs ~ lda + qda, data = testProbs)
xyplot(calPlotData, auto.key = list(columns = 2))

## The interface puts data as first parameter
calPlotData <- ntbt_calibration(testProbs, obs ~ lda + qda)
xyplot(calPlotData, auto.key = list(columns = 2))

## so it can be used easily in a pipeline.
testProbs %>%
  ntbt_calibration(obs ~ lda + qda) %>%
  xyplot(auto.key = list(columns = 2))


## ntbt_dummyVars
when <- data.frame(time = c("afternoon", "night", "afternoon",
                            "morning", "morning", "morning",
                            "morning", "afternoon", "afternoon"),
                   day = c("Mon", "Mon", "Mon",
                           "Wed", "Wed", "Fri",
                           "Sat", "Sat", "Fri"))

levels(when$time) <- list(morning="morning",
                          afternoon="afternoon",
                          night="night")
levels(when$day) <- list(Mon="Mon", Tue="Tue", Wed="Wed", Thu="Thu",
                         Fri="Fri", Sat="Sat", Sun="Sun")

## Original function to interface
mainEffects <- dummyVars(~ day + time, data = when)
mainEffects
predict(mainEffects, when[1:3,])

## The interface puts data as first parameter
mainEffects <- ntbt_dummyVars(when, ~ day + time)
mainEffects
predict(mainEffects, when[1:3,])

## so it can be used easily in a pipeline.
when %>%
  ntbt_dummyVars(~ day + time) %>%
  predict(when[1:3,])


## ntbt_icr: Independent Component Regression
## Not found example using formula interface, and I am
## completely ignorant to construct one.
data(BloodBrain)
BB <- list(bbbDescr, logBBB)

## Original function to interface
icr(bbbDescr, logBBB, n.comp = 5)

## The interface puts data as first parameter
ntbt_icr(BB, bbbDescr, logBBB, n.comp = 5)

## so it can be used easily in a pipeline.
BB %>%
  ntbt_icr(bbbDescr, logBBB, n.comp = 5)


## ntbt_knn3: k-Nearest Neighbour Classification
## Original function to interface
knn3(Species ~ ., iris)

## The interface puts data as first parameter
ntbt_knn3(iris, Species ~ .)

## so it can be used easily in a pipeline.
iris %>%
  ntbt_knn3(Species ~ .)


## ntbt_lift: Lift Plot
set.seed(1)
simulated <- data.frame(obs = factor(rep(letters[1:2], each = 100)),
perfect = sort(runif(200), decreasing = TRUE),
random = runif(200))
## Original function to interface
lift1 <- lift(obs ~ random, data = simulated)
lift1
xyplot(lift1)

## The interface puts data as first parameter
lift1 <- ntbt_lift(simulated, obs ~ random)
lift1
xyplot(lift1)

## so it can be used easily in a pipeline.
simulated %>%
  ntbt_lift(obs ~ random) %>%
  xyplot()


## ntbt_pcaNNet: Neural Networks with a Principal Component Step
## Not found example using formula interface, and I am
## completely ignorant to construct one.
data(BloodBrain)
BB <- list(bbbDescr, logBBB)

## Original function to interface
pcaNNet(bbbDescr[, 1:10], logBBB, size = 5, linout = TRUE, trace = FALSE)

## The interface puts data as first parameter
ntbt_pcaNNet(BB, bbbDescr[, 1:10], logBBB, size = 5, linout = TRUE, trace = FALSE)

## so it can be used easily in a pipeline.
BB %>%
  ntbt_pcaNNet(bbbDescr[, 1:10], logBBB, size = 5, linout = TRUE, trace = FALSE)


## ntbt_sbf: Selection By Filtering (SBF)
## Not found example using formula interface, and I am
## completely ignorant to construct one.
data(BloodBrain)
BB <- list(bbbDescr, logBBB)

## Be prepared to wait...
## Original function to interface
sbf(bbbDescr, logBBB,
    sbfControl = sbfControl(functions = rfSBF,
                            verbose = FALSE, 
                            method = "cv"))

## The interface puts data as first parameter
ntbt_sbf(BB, bbbDescr, logBBB,
         sbfControl = sbfControl(functions = rfSBF,
                                 verbose = FALSE, 
                                 method = "cv"))

## so it can be used easily in a pipeline.
BB %>%
  ntbt_sbf(bbbDescr, logBBB,
           sbfControl = sbfControl(functions = rfSBF,
                                   verbose = FALSE, 
                                   method = "cv"))


## ntbt_train: Fit Predictive Models over Different Tuning Parameters
library(mlbench)
data(BostonHousing)

## Original function to interface
train(medv ~ . + rm:lstat, data = BostonHousing, method = "lm")

## The interface puts data as first parameter
ntbt_train(BostonHousing, medv ~ . + rm:lstat, method = "lm")

## so it can be used easily in a pipeline.
BostonHousing %>%
  ntbt_train(medv ~ . + rm:lstat, method = "lm")

## End(Not run)

intubate documentation built on May 2, 2019, 2:46 p.m.