adabag: Interfaces for adabag package for data science pipelines.

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

Description

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

Usage

ntbt_autoprune(data, ...)
# ntbt_bagging(data, ...)  ## Already defined in ipred
ntbt_bagging.cv(data, ...)
ntbt_boosting(data, ...)
ntbt_boosting.cv(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(adabag)


## ntbt_autoprune: Builds automatically a pruned tree of class rpart
## Original function to interface
autoprune(Species ~ ., data = iris)

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

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

## ntbt_bagging: Applies the Bagging algorithm to a data set
library(rpart)
data(iris)

## Original function to interface
bagging(Species ~ ., data = iris, mfinal = 10)

## The interface puts data as first parameter
ntbt_bagging(iris, Species ~ ., mfinal = 10)

## so it can be used easily in a pipeline.
iris %>%
  ntbt_bagging(Species ~ ., mfinal = 10)



## Original function to interface
iris.baggingcv <- bagging.cv(Species ~ ., v = 2, data = iris, mfinal = 10,
                             control = rpart.control(cp = 0.01))
iris.baggingcv[-1]

## The interface puts data as first parameter
iris.baggingcv <- ntbt_bagging.cv(iris, Species ~ ., v = 2, mfinal = 10,
                                  control = rpart.control(cp = 0.01))
iris.baggingcv[-1]

## so it can be used easily in a pipeline.
iris.baggingcv <- iris %>%
  ntbt_bagging.cv(Species ~ ., v = 2, mfinal = 10,
                  control = rpart.control(cp = 0.01))
iris.baggingcv[-1]


## ntbt_boosting: Applies the AdaBoost.M1 and SAMME algorithms to a data set
## Original function to interface
boosting(Species ~ ., data = iris, boos = TRUE, mfinal = 5)

## The interface puts data as first parameter
ntbt_boosting(iris, Species ~ ., boos = TRUE, mfinal = 5)

## so it can be used easily in a pipeline.
iris %>%
  ntbt_boosting(Species ~ ., boos = TRUE, mfinal = 5)


## ntbt_boosting.cv: Runs v-fold cross validation with AdaBoost.M1 or SAMME
## Original function to interface
iris.boostcv <- boosting.cv(Species ~ ., v = 2, data = iris, mfinal = 10, 
                            control = rpart.control(cp = 0.01))
iris.boostcv[-1]

## The interface puts data as first parameter
iris.boostcv <- ntbt_boosting.cv(iris, Species ~ ., v = 2, mfinal = 10, 
                                 control = rpart.control(cp = 0.01))
iris.boostcv[-1]

## so it can be used easily in a pipeline.
iris.boostcv <- iris %>%
  ntbt_boosting.cv(Species ~ ., v = 2, mfinal = 10, 
                   control = rpart.control(cp = 0.01))
iris.boostcv[-1]

## End(Not run)

rbertolusso/intubate documentation built on May 27, 2019, 3 a.m.