n_distinct() counts the number of unique/distinct combinations in a set
of one or more vectors. It's a faster and more concise equivalent to
nrow(unique(data.frame(...))).
Arguments
- ...
Unnamed vectors. If multiple vectors are supplied, then they should have the same length.
- na.rm
If
TRUE, exclude missing observations from the count. If there are multiple vectors in..., an observation will be excluded if any of the values are missing.
Examples
x <- c(1, 1, 2, 2, 2)
n_distinct(x)
#> [1] 2
y <- c(3, 3, NA, 3, 3)
n_distinct(y)
#> [1] 2
n_distinct(y, na.rm = TRUE)
#> [1] 1
# Pairs (1, 3), (2, 3), and (2, NA) are distinct
n_distinct(x, y)
#> [1] 3
# (2, NA) is dropped, leaving 2 distinct combinations
n_distinct(x, y, na.rm = TRUE)
#> [1] 2
# Also works with data frames
n_distinct(data.frame(x, y))
#> [1] 3