So I am trying to create a list of all combinations of 14 numbers that have been split into two separate groups. For example, the beginning of this list may look like:
| Group X | Group Y |
|---|---|
| 1 | 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 |
| 1, 2 | 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 |
| 1, 2, 3 | 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 |
And so forth. This list would consist of 16384 possible combinations of these 14 numbers when split into two groups. (2^14) Any ideas on how to do this in R?
You may try (for 2^3 cases)
x <- expand.grid(replicate(3, c(T,F), simplify = F))
apply(x, 1, function(x){list(group1 = c(1:3)[x], group2 = c(1:3)[!x])})
[[1]]
[[1]]$group1
[1] 1 2 3
[[1]]$group2
numeric(0)
[[2]]
[[2]]$group1
[1] 2 3
[[2]]$group2
[1] 1
[[3]]
[[3]]$group1
[1] 1 3
[[3]]$group2
[1] 2
[[4]]
[[4]]$group1
[1] 3
[[4]]$group2
[1] 1 2
[[5]]
[[5]]$group1
[1] 1 2
[[5]]$group2
[1] 3
[[6]]
[[6]]$group1
[1] 2
[[6]]$group2
[1] 1 3
[[7]]
[[7]]$group1
[1] 1
[[7]]$group2
[1] 2 3
[[8]]
[[8]]$group1
numeric(0)
[[8]]$group2
[1] 1 2 3
I believe that this code accomplishes something close to what you are requesting. The results are stored in the non-zero entries of the matrices "group1" and "group2" by corresponding row, so you may need to alter the code in order to render the results in a form that is useful for your purposes.
The output is shown as it was returned from the R console; the truncated data tables at the end illustrate the form of the output.
# Written in R version 4.1.1
require(data.table)
group1 = matrix(data = 0, nrow = 16384, ncol = 14, byrow = T)
group2 = matrix(data = 0, nrow = 16384, ncol = 14, byrow = T)
a = 1:14
count = 1
for(i in 0:14){
nf = data.frame(combn(a,i))
for(j in 1:dim(nf)[2]){
g1 = c(nf[,j])
if(length(g1) > 0){
group1[count,1:length(g1)] = g1
}
g2 = a[-g1]
if(length(g2) > 0){
group2[count,1:length(g2)] = g2
}
if(i == 0 & j == 1){
group2[count,1:14] = 1:14
}
count = count + 1
}}
data.table(group1)
# V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13 V14
# 1: 0 0 0 0 0 0 0 0 0 0 0 0 0 0
# 2: 1 0 0 0 0 0 0 0 0 0 0 0 0 0
# 3: 2 0 0 0 0 0 0 0 0 0 0 0 0 0
# 4: 3 0 0 0 0 0 0 0 0 0 0 0 0 0
# 5: 4 0 0 0 0 0 0 0 0 0 0 0 0 0
# ---
# 16380: 1 2 3 5 6 7 8 9 10 11 12 13 14 0
# 16381: 1 2 4 5 6 7 8 9 10 11 12 13 14 0
# 16382: 1 3 4 5 6 7 8 9 10 11 12 13 14 0
# 16383: 2 3 4 5 6 7 8 9 10 11 12 13 14 0
# 16384: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
data.table(group2)
# V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13 V14
# 1: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
# 2: 2 3 4 5 6 7 8 9 10 11 12 13 14 0
# 3: 1 3 4 5 6 7 8 9 10 11 12 13 14 0
# 4: 1 2 4 5 6 7 8 9 10 11 12 13 14 0
# 5: 1 2 3 5 6 7 8 9 10 11 12 13 14 0
# ---
# 16380: 4 0 0 0 0 0 0 0 0 0 0 0 0 0
# 16381: 3 0 0 0 0 0 0 0 0 0 0 0 0 0
# 16382: 2 0 0 0 0 0 0 0 0 0 0 0 0 0
# 16383: 1 0 0 0 0 0 0 0 0 0 0 0 0 0
# 16384: 0 0 0 0 0 0 0 0 0 0 0 0 0 0
print(paste("count = ",count - 1))
# [1] "count = 16384"
We can try combn like below
v <- 1:14
res <- do.call(
c,
lapply(
c(0, seq_along(v)),
function(k) {
combn(v, k, function(x) list(grp1 = x, grp2 = v[!v %in% x]),
simplify = FALSE
)
}
)
)
and we will see
> length(res)
[1] 16384
> head(res)
[[1]]
[[1]]$grp1
integer(0)
[[1]]$grp2
[1] 1 2 3 4 5 6 7 8 9 10 11 12 13 14
[[2]]
[[2]]$grp1
[1] 1
[[2]]$grp2
[1] 2 3 4 5 6 7 8 9 10 11 12 13 14
[[3]]
[[3]]$grp1
[1] 2
[[3]]$grp2
[1] 1 3 4 5 6 7 8 9 10 11 12 13 14
[[4]]
[[4]]$grp1
[1] 3
[[4]]$grp2
[1] 1 2 4 5 6 7 8 9 10 11 12 13 14
[[5]]
[[5]]$grp1
[1] 4
[[5]]$grp2
[1] 1 2 3 5 6 7 8 9 10 11 12 13 14
[[6]]
[[6]]$grp1
[1] 5
[[6]]$grp2
[1] 1 2 3 4 6 7 8 9 10 11 12 13 14