Your function only works for the first dimensions (e.g. indices
indicating the positions in the first two dimensions in datacube), correct?
Otherwise it looks very handy! And certainly more elegent than my
function monster!
Jannis
On 08/04/2011 09:58 PM, R. Michael Weylandt wrote:
Hi Jannis,
Like Gene, I'm not entirely sure what your code is intended to do, but it
wasn't hard to adapt mine to at least print out the desired slice through
the higher-dimensional array:
cubeValues<- function(Insert, Destination, Location) {
Location = as.matrix(Location)
Location = array(Location,dim(Destination))
if (is.null(Insert)) {
Destination = round(Destination,3)
Destination[!Location] = NA
print(Destination)
return(invisible())
}
Destination[Location]<- Insert
return(Destination)
}
If Insert = NULL, it adopts a printing rather than value assigning behavior.
If you could indicate how you want the values when they come out, it's
pretty easy to adapt this to do whatever, but I can't just pull out
subarrays of arbitrary shape while keeping shape: e.g.,
x = matrix(1:4,2,2,byrow=T)
y = rbind(c(T,F),c(F,T))
is(x[y])
"vector"
If you just want the values in a vector, take this version of my code:
cubeValues<- function(Insert, Destination, Location) {
Location = as.matrix(Location)
Location = array(Location,dim(Destination))
if (is.null(Insert)) {
return(Destination[Location])
}
Destination[Location]<- Insert
return(Destination)
}
It sounds like you've got what you want, but hopefully this will be of some
use to anyone who stumbles across this and, like Gene& myself, doesn't
really get your code.
NB: I have not tested the provided code very much -- it relies on the
array() function to repeat Location as appropriate. If you know how R
repeats smaller arrays to make them fit big arrays, this should be fine for
you -- caveat code-or.
Michael Weylandt
PS -- The combinations() function of the gtools package might be of help to
you as well. We could get the entire example Gene got by
ans = combinations(1:4,2,repeats.allowed=T)
rbind(cbind(ans,4),cbind(ans,1))
and it's probably not hard to simplify the entire code as desired.
On Thu, Aug 4, 2011 at 6:33 AM, Jannis<bt_jan...@yahoo.de> wrote:
Thanks, Michael. I was, however, after a function I coul use for both
extracting and replacing subarrays. In case anybody else stumbles over this
problem, here is my solution. Its programming is most probably horribly
clumsy:
ind.datacube = function(
##title<< create logical index matrices for multidimensional datacubes
datacube ##<< array: datacube from which to extract the subparts
, logical.ind ##<< logical array: TRUE/FALSE index matrix for a subset of
the dimensions
## of datacube. The size of logical.ind`s dimesnions has
to match the
## sizes of the corresponding dimesnions in datacube.
, dims='auto' ##<< integer vector or 'auto' : indices of the dimensions
in datacube corresponding
## to the dimensions of logical.ind. If set to 'auto'
this matching is tried to
## be accomplished by comparing the sizes of the
dimensions of the two objects.
)
{
if (sum(logical.ind) == 0) {
stop('No TRUE value in index matrix!')
} else {
if (dims == 'auto')
{
if (is.null(dim(logical.ind)[1])) {
size.ind = length(logical.ind)
logical.ind = matrix(logical.ind,ncol=1)
} else {
size.ind = dim(logical.ind)
}
dims = match(size.ind, dim(datacube))
if (sum(duplicated(size.ind))> 0 || sum(duplicated(dims))> 0 )
stop('dimensions do not match unambigously. Supply dims
manually!')
}
dims.nonapply<- setdiff(1:length(dim(datacube)**),dims)
ind.matrix<- which(logical.ind, arr.ind = TRUE)
args.expand.grid<- list()
counter = 1
for (i in 1: length(dim(datacube)))
{
if (is.element(i,dims.nonapply)) {
args.expand.grid[[i]] = 1:dim(datacube)[dims.nonapply[**i]]
} else {
args.expand.grid[[i]] = ind.matrix[,counter]
counter = counter + 1
}
}
ind.all<- as.matrix(do.call(expand.grid, args.expand.grid))
ind.matrix<- ind.all[,order(c(dims.**nonapply,dims))]
}
##value<< integer index matrix which can be used to index datacube
ind.matrix
}
On 08/04/2011 12:12 AM, R. Michael Weylandt<michael.weyla...@gmail.com>
wrote:
This might be a little late: but how about this (slightly clumsy)
function:
putValues<- function(Insert, Destination, Location) {
Location = as.matrix(Location)
Location = array(Location,dim(**Destination))
Destination[Location]<- Insert
return(Destination)
}
It currently assumes that the location array lines up in dimension order,
but other than that seems to work pretty well. If you want, it shouldn't
be
hard to change it to take in a set of dimensions to arrange Location
along.
If you like any of the other suggested behaviors, you could put in a
is.null(Insert) option that returns the desired subset of values. I
haven't
tested it completely, but for a few sample inputs, it seems be do as
desired.
Michael
On Wed, Aug 3, 2011 at 5:00 PM, Jannis<bt_jan...@yahoo.de> wrote:
Thanks for all the replies!Unfortunately the solutions only work for
extracting subsets of the data (which was exactly what I was asking for)
and
not to replace subsets with other values. I used them, however, to
program a
rather akward function to do that. Seems I found one of the few aspects
where Matlab actually is slightly easier to use than R.
Thanks for your help!
Jannis
On 08/01/2011 05:50 PM, Gene Leynes wrote:
What do you think about this?
apply(data, 3, '[', indices)
On Mon, Aug 1, 2011 at 4:38 AM, Jannis<bt_jan...@yahoo.de> wrote:
Dear R community,
I have a general question regarding indexing in multidiemensional
arrays.
Imagine I have a three dimensional array and I only want to extract on
vector along a single dimension from it:
data<- array(rnorm(64),dim=c(4,4,4))
result<- data[1,1,]
If I want to extract more than one of these vectors, it would now
really
help me to supply a logical matrix of the size of the first two
dimensions:
indices<- matrix(FALSE,ncol=4,nrow=4)
indices[1,3]<- TRUE
indices[4,1]<- TRUE
result<- data[indices,]
This, however would give me an error. I am used to this kind of
indexing
from Matlab and was wonderingt whether there exists an easy way to do
this
in R without supplying complicated index matrices of all three
dimensions or
logical vectors of the size of the whole matrix?
The only way I could imagine would be to:
result<- data[rep(as.vector(indices),******times=4)]
but this seems rather complicated and also depends on the order of the
dimensions I want to extract.
I do not want R to copy Matlabs behaviour, I am just wondering whether
I
missed one concept of indexing in R?
Thanks a lot
Jannis
______________________________******________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/******listinfo/r-help<https://stat.ethz.ch/mailman/****listinfo/r-help>
<https://**stat.ethz.ch/mailman/****listinfo/r-help<https://stat.ethz.ch/mailman/**listinfo/r-help>
<https://stat.**ethz.ch/**mailman/listinfo/r-**help<http://ethz.ch/mailman/listinfo/r-**help>
<http**s://stat.ethz.ch/mailman/**listinfo/r-help<https://stat.ethz.ch/mailman/listinfo/r-help>
PLEASE do read the posting guide http://www.R-project.org/**
posting-guide.html<http://www.****R-project.org/posting-guide.**
**html<http://www.R-project.**org/posting-guide.html<http://www.R-project.org/posting-guide.html>
and provide commented, minimal, self-contained, reproducible code.
______________________________****________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/****listinfo/r-help<https://stat.ethz.ch/mailman/**listinfo/r-help>
<https://stat.**ethz.ch/mailman/listinfo/r-**help<https://stat.ethz.ch/mailman/listinfo/r-help>
PLEASE do read the posting guide http://www.R-project.org/**
posting-guide.html<http://www.**R-project.org/posting-guide.**html<http://www.R-project.org/posting-guide.html>
and provide commented, minimal, self-contained, reproducible code.
______________________________________________
R-help@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.
