[R] remove columns containing all zeros (or other value)

[Marc Schwartz]

Careful:

x <- matrix(c(1, 5, 3, 2, 1, 4, -1, 0, 1),
            ncol = 3, nrow = 3)

> x
     [,1] [,2] [,3]
[1,]    1    2   -1
[2,]    5    1    0
[3,]    3    4    1

> x[, colSums(x) != 0]
     [,1] [,2]
[1,]    1    2
[2,]    5    1
[3,]    3    4


Not quite the result wanted...  :-)

Try this:

> x[, colSums(x == 0) != nrow(x)]
     [,1] [,2]
[1,]    1    2
[2,]    5    1
[3,]    3    4


x <- matrix(c(1, 5, 3, 2, 1, 4, -1, 0, 1),
              ncol = 3, nrow = 3)

> x[, colSums(x == 0) != nrow(x)]
     [,1] [,2] [,3]
[1,]    1    2   -1
[2,]    5    1    0
[3,]    3    4    1



HTH,

Marc Schwartz


on 01/14/2009 04:29 PM Gustavo Carvalho wrote:
> Sorry for the double post, but this is probably faster:
> 
> x[, colSums(x) != 0]
> 
> On Wed, Jan 14, 2009 at 8:22 PM, Gustavo Carvalho
> <gustavo.bio+R at gmail.com> wrote:
>> You can also try this:
>>
>> x[,-(which(colSums(x) == 0))]
>>
>> Cheers,
>>
>> Gustavo.
>>
>> On Wed, Jan 14, 2009 at 8:01 PM, Anthony Dick <adick at uchicago.edu> wrote:
>>> Hello-
>>>
>>> I would like to remove the columns of a matrix that contain all zeros. For
>>> example, from
>>> x<-matrix(c(1,5,3,2,1,4,0,0,0), ncol=3,nrow=3)
>>>
>>> I would like to remove the third column. However, because this is in a loop
>>> I need a way to first determine which columns are all zeros, and only then
>>> remove them. I.e., I don't know which column of x contains all zeros until
>>> after x is created.
>>>
>>> Thanks!
>>>
>>> Anthony
>>>