R语言中查找可重复元素组合（m可大于向量长度）

Oh right, I’ve run into this exact issue before! The base R combn() function only handles combinations without replacement, which is why it throws that n < m error when you try to pick more elements than are in your vector. What you’re looking for are combinations with replacement—and there are a couple of easy ways to get that in R.

Using the gtools Package (Simplest Method) #

The gtools library has a dedicated combinations() function that supports replacement out of the box. Here's how to use it for your example:

First, install and load the package if you haven’t already:

install.packages("gtools")
library(gtools)

Then generate your desired combinations:

combinations(n = 2, r = 3, v = letters[1:2], repeats.allowed = TRUE)

This will output the unordered combinations you want (each row is a unique group):

[,1] [,2] [,3]
[1,] "a"  "a"  "a" 
[2,] "a"  "a"  "b" 
[3,] "a"  "b"  "b" 
[4,] "b"  "b"  "b" 

If you want the transposed version (matching your example’s column-wise display), just wrap it in t():

t(combinations(n = 2, r = 3, v = letters[1:2], repeats.allowed = TRUE))

Which gives:

[,1] [,2] [,3] [,4]
[1,] "a"  "a"  "a"  "b" 
[2,] "a"  "a"  "b"  "b" 
[3,] "a"  "b"  "b"  "b" 

Base R Alternative (No External Packages) #

If you’d rather avoid installing extra packages, you can use expand.grid() to generate all possible ordered tuples, then filter down to unique sorted combinations:

x <- letters[1:2]
m <- 3

# Generate every possible ordered tuple (including duplicates)
all_tuples <- expand.grid(rep(list(x), m))

# Sort each row to normalize combinations, then keep only unique ones
unique_combs <- unique(t(apply(all_tuples, 1, sort)))

# Transpose to match your preferred output format
t(unique_combs)

This will produce the exact same result as the gtools method.

Quick Note on the Math #

For combinations with replacement, the number of unique groups is calculated by the formula (n + m - 1 choose m), where n is the length of your vector and m is the number of elements to pick. For your example (n=2, m=3), that’s (2+3-1 choose 3) = 4 choose 3 = 4—which perfectly matches the 4 combinations you expected.

内容的提问来源于stack exchange，提问作者Jian