Chapter 22 Split data into train and test subsets

Here you can find several simple approaches to split data into train and test subset to fit and to test parameters of your model. We want to take 0.8 of our initial data to train our model.
Data: datasets::iris.

1. First approach is to create a vector containing randomly selected row ids and to apply this vector to split data.

inTrain = sample(nrow(iris), nrow(iris)*0.8)

# split data
train = iris[inTrain, ]
test = iris[-inTrain, ]

2. The same idea to split data as before using caret package.
   The advantage is that createDataPartition function allows to split data many times and use these subsets to estimate parameters of our model.

library(caret)
trainIndex <- createDataPartition(iris$Species, p=.8,
                                  list = FALSE,        # if FALSE - create a vector/matrix, if TRUE - create a list
                                  times = 1)           # how many subsets
# split data
train <- iris[trainIndex, ]
test <- iris[-trainIndex, ]

3. Another approch is to create a logical vecotor containing randomly distributed true/false and apply this vector to subset data.

inTrain = sample(c(TRUE, FALSE), nrow(iris), replace = T, prob = c(0.8,0.2))

# select data
train = iris[inTrain, ]
test = iris[!inTrain, ]

4. Using caTools.

library(caTools)
inTrain = sample.split(iris, SplitRatio = .8)
train = subset(iris, inTrain == TRUE)
test  = subset(iris, inTrain == FALSE)

5. Using dplyr

library(dplyr)
iris$id <- 1:nrow(iris)
train <- iris %>% dplyr::sample_frac(.8)
test  <- dplyr::anti_join(iris, train, by = 'id')