wrswoR

The goal of wrswoR is to provide faster implementations of weighted random sampling without replacement in R.

Installation

You can install the released version of wrswoR from CRAN with:

install.packages("wrswoR")

And the development version from GitHub with:

# install.packages("pak")
# pak::pak("krlmlr/wrswoR")
install.packages("wrswoR", repos = c('https://krlmlr.r-universe.dev', 'https://cloud.r-project.org'))

Example

The functions in this package are a drop-in replacement to sample.int(n, size, replace = FALSE, prob = prob). With large n, sample.int() becomes too slow to be practical, unlike the functions in this package.

library(wrswoR)
set.seed(20200726)

sample_int_crank(20, 10, 1:20)
#>  [1]  8 18 14 17 11 15 10  4 13  5