Skip to content

list of matrices --> array

7 messages · arun, Rolf Turner, David Winsemius +2 more

#
i'm somehow embarrassed to even ask this, but is there any built-in
method for doing this:

my_list <- list()
my_list[[1]] <- matrix(1:20, ncol = 5)
my_list[[2]] <- matrix(20:1, ncol = 5)

now, knowing that these matrices are identical in dimension, i'd like
to unfold the list to a 2x4x5 (or some other permutation of the dim
sizes) array.
i know i can initialize the array, then loop through my_list to fill
the array, but somehow this seems inelegant.
i also know i can vectorize the matrices and unlist the list, then
build the array from that single vector, but this also seems inelegant
(and an easy place to introduce errors/bugs).

i can't seem to find any built-in that handles this already... but
maybe i just haven't looked hard enough :-/

cheers,

-m
#
FYI - this is my current method, but somehow i just don't like it ;-)

foo <- array(NA, dim = c(4,5,length(my_list)))
for(k in 1:length(my_list)) {
  foo[,,k] <- my_list[[k]]
}

-m
On Thu, Feb 14, 2013 at 1:03 AM, Murat Tasan <mmuurr at gmail.com> wrote:
#
Hi,

May be this helps:
library(plyr)
res<-aaply(laply(my_list, as.matrix),c(2,3),function(x) x)
attr(res,"dimnames")<- NULL
?identical(res,foo)
#[1] TRUE
A.K.




----- Original Message -----
From: Murat Tasan <mmuurr at gmail.com>
To: r-help at r-project.org
Cc: 
Sent: Thursday, February 14, 2013 1:13 AM
Subject: Re: [R] list of matrices --> array

FYI - this is my current method, but somehow i just don't like it ;-)

foo <- array(NA, dim = c(4,5,length(my_list)))
for(k in 1:length(my_list)) {
? foo[,,k] <- my_list[[k]]
}

-m
On Thu, Feb 14, 2013 at 1:03 AM, Murat Tasan <mmuurr at gmail.com> wrote:
______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
#
require(abind)
do.call(abind,c(my_list,list(along=0))) # Gives 2 x 4 x 5
do.call(abind,c(my_list,list(along=3))) # Gives 4 x 5 x 2

The latter seems more natural to me.

     cheers,

         Rolf Turner
On 02/14/2013 07:03 PM, Murat Tasan wrote:
#
On Feb 13, 2013, at 10:03 PM, Murat Tasan wrote:

            
I think the built-in function you seek is `simplify2array`:

# gives 4 x 5 x 2
 > simplify2array(my_list)
, , 1

      [,1] [,2] [,3] [,4] [,5]
[1,]    1    5    9   13   17
[2,]    2    6   10   14   18
[3,]    3    7   11   15   19
[4,]    4    8   12   16   20

, , 2

      [,1] [,2] [,3] [,4] [,5]
[1,]   20   16   12    8    4
[2,]   19   15   11    7    3
[3,]   18   14   10    6    2
[4,]   17   13    9    5    1

 > # 2 x 4 x 5
  aperm( simplify2array(my_list), c(3,1,2) )


  # 4 x 2 x 5
 > aperm( simplify2array(my_list), c(1,3,2) )
3 days later
#
abind() (from package 'abind') can take a list of arrays as its first argument, so in general, no need for do.call() with abind().

As another poster pointed out, simplify2array() can also be used; while abind() gives more options regarding which dimension is created and how dimension names are constructed.

 > x <- list(A=cbind(X=c(a=1,b=2,c=3,d=4),Y=5:8,Z=9:12), B=cbind(X=c(a=13,b=14,c=15,d=16),Y=17:20,Z=21:24))
$A
   X Y  Z
a 1 5  9
b 2 6 10
c 3 7 11
d 4 8 12

$B
    X  Y  Z
a 13 17 21
b 14 18 22
c 15 19 23
d 16 20 24

 >
 > dim(abind(x, along=3))
[1] 4 3 2
 > dim(abind(x, along=1.5))
[1] 4 2 3
 > dim(abind(x, along=0.5))
[1] 2 4 3
 > dim(abind(x, along=1, hier.names=T)) # construct rownames in a hierarchical manner A.a, A.b, etc
[1] 8 3
 > dim(abind(x, along=2, hier.names=T)) # construct colnames in a hierarchical manner
[1] 4 6
 > abind(x, along=2, hier.names=T)
   A.X A.Y A.Z B.X B.Y B.Z
a   1   5   9  13  17  21
b   2   6  10  14  18  22
c   3   7  11  15  19  23
d   4   8  12  16  20  24
 >
On 2/14/2013 3:53 AM, Rolf Turner wrote:
#
thanks to all!
didn't know about simplify2array, nor about the abind package.
they're exactly what i wanted.

cheers,

-m
On Sun, Feb 17, 2013 at 9:41 AM, Tony Plate <taplate at gmail.com> wrote: