README.md

In geraudster/modelUtils: Utility function for testing predictive models

ModelUtils

An R package for manipulating predictive models.

Installation

You need the devtools package to be installed before, then in R:

devtools::install_github('geraudster/modelUtils')

Usage

Generate a model:

library(modelUtils)
data(iris)
inTrain <- createDataPartition(y=iris$Species, p=0.80, list=FALSE)
training <- na.omit(iris[inTrain,])
testing <- na.omit(iris[-inTrain,])
model <- testModel(Species ~ ., training, testing, 'Species', 'rpart')

Then to write the predictions:

submission <- writeSubmission(model, data.frame(id = rownames(testing), testing), 'id', 'Species', 'test')

Will create a file with:

id,Species
14,setosa
21,setosa
30,setosa
31,setosa
32,setosa
...

geraudster/modelUtils documentation built on May 17, 2019, 2:10 a.m.