Generate a variable used to selected a training sample

make_train(n = 0.7, nr = NULL, blocks = NULL, seed = 1234)

Arguments

n: Number (or fraction) of observations to label as training
nr: Number of rows in the dataset
blocks: A vector to use for blocking or a data.frame from which to construct a blocking vector
seed: Random seed

Value

0/1 variables for filtering

Examples

make_train(.5, 10)
#>  [1] 1 1 0 0 0 1 0 1 1 0
make_train(.5, 10) %>% table()
#> .
#> 0 1 
#> 5 5 
make_train(100, 1000) %>% table()
#> .
#>   0   1 
#> 900 100 
make_train(.15, blocks = mtcars$vs) %>% table() / nrow(mtcars)
#> .
#>       0       1 
#> 0.84375 0.15625 
make_train(.10, blocks = iris$Species) %>% table() / nrow(iris)
#> .
#>   0   1 
#> 0.9 0.1 
make_train(.5, blocks = iris[, c("Petal.Width", "Species")]) %>% table()
#> .
#>  0  1 
#> 75 75