[R] Create a data frame of all possible unique combinations of factors

[David Winsemius]

On Jul 5, 2011, at 7:16 PM, Q wrote:

> Hello,
>
> I'm trying to create a data frame where each row has a unique  
> combination of
> factors.
> I start with a vector of species like so:
>
>> 1> test <- c("A","B","C","D")
>> 1> test
>> [1] "A" "B" "C" "D"
>>
> To get all species combinations I have used expand.grid like this:
>
>> 1> pairs <- expand.grid(test,test)
>
>> 1> pairs
>
>>   Var1 Var2
>
>> 1     A    A
>> 2     B    A
>> 3     C    A
>> 4     D    A
>> 5     A    B
>

snipped
> Now I want to select only the unique pairs, which I have tried to do  
> with
> the function "unique":
>
>
>
>> 1> unique(pairs)

  You want the duplicated function or more precisely its negation, ...  
and you need to sort within rows if yoy want (b,a) to look like (a,b).

>>
>
> , but that doesn't do anything... I guess because it considers A,B  
> to be
> different from B,A.  The data frame I would like to end up with  
> should look
> like this.

 > pairs[!duplicated(t(apply(pairs, 1, sort))), ]
    Var1 Var2
1     A    A
2     B    A
3     C    A
4     D    A
6     B    B
7     C    B
8     D    B
11    C    C
12    D    C
16    D    D

The t( ) is needed to get the row-wise arrangement restore after apply  
transposed it.

> 	[[alternative HTML version deleted]]
Nabble allows posting in plain text. You should.

>
PLease --->
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html

-- 

David Winsemius, MD
West Hartford, CT