Splitting Data into Train and Test using caret package in R

Splitting data in R using sample function and caret package

Data is split into Train and Test in R to train the model and evaluate the results.

There are multiple ways of doing this.

1. Splitting data using sample function

#load data into variable called mydata
mydata = read.csv('mydata.csv',header=T)
#setting seed so we get same data split each time
set.seed(100) #can provide any number for seed
nall = nrow(mydata) #total number of rows in data
ntrain = floor(0.7 * nall) # number of rows for train,70%
ntest = floor(0.3* nall) # number of rows for test, 30%
index = seq(1:nall)
trainIndex = sample(index, ntrain) #train data set
testIndex = index[-trainIndex]

train = mydata[trainIndex,]
test = mydata[testIndex,]

2. Splitting data using caret package

Data can be split in caret package based on the target variable, or y variable

For illustration, I am assuming target variable to be TARGET

#install caret package
#load package
trainIndex = createDataPartition(mydata$TARGET,
                       p=0.7, list=FALSE,times=1)

train = mydata[trainIndex,]
test = mydata[-trainIndex,]

There are more ways of splitting code. If you want to read more,
please refer to the following link