[R] Create a multi dimensional array from a data frame

[David Winsemius]

On Jan 21, 2010, at 6:06 PM, LCOG1 wrote:

>
> Hello all,
>   I have data from a csv that i need to coerce into a multi  
> dimensional
> array.  As you can see from my sample code i have yet to master  
> building
> efficient code.  I can sort of do what i want but its very  
> cumbersome code
> and i know there is a better way to do it, i am just not expereinced  
> enought
> yet.  So for the following:
>
> #Create Data
> h1i1a1<-sample(1:20, 5, replace=TRUE)
> h1i1a2<-sample(1:20, 5, replace=TRUE)
> h1i1a3<-sample(1:20, 5, replace=TRUE)
> h1i2a1<-sample(1:20, 5, replace=TRUE)
> h1i2a2<-sample(1:20, 5, replace=TRUE)
> h1i2a3<-sample(1:20, 5, replace=TRUE)
> h1i3a1<-sample(1:20, 5, replace=TRUE)
> h1i3a2<-sample(1:20, 5, replace=TRUE)
> h1i3a3<-sample(1:20, 5, replace=TRUE)
> h1i4a1<-sample(1:20, 5, replace=TRUE)
> h1i4a2<-sample(1:20, 5, replace=TRUE)
> h1i4a3<-sample(1:20, 5, replace=TRUE)
> h2i1a1<-sample(1:20, 5, replace=TRUE)
> h2i1a2<-sample(1:20, 5, replace=TRUE)
> h2i1a3<-sample(1:20, 5, replace=TRUE)
> h2i2a1<-sample(1:20, 5, replace=TRUE)
> h2i2a2<-sample(1:20, 5, replace=TRUE)
> h2i2a3<-sample(1:20, 5, replace=TRUE)
> h2i3a1<-sample(1:20, 5, replace=TRUE)
> h2i3a2<-sample(1:20, 5, replace=TRUE)
> h2i3a3<-sample(1:20, 5, replace=TRUE)
> h2i4a1<-sample(1:20, 5, replace=TRUE)
> h2i4a2<-sample(1:20, 5, replace=TRUE)
> h2i4a3<-sample(1:20, 5, replace=TRUE)
> h3i1a1<-sample(1:20, 5, replace=TRUE)
> h3i1a2<-sample(1:20, 5, replace=TRUE)
> h3i1a3<-sample(1:20, 5, replace=TRUE)
> h3i2a1<-sample(1:20, 5, replace=TRUE)
> h3i2a2<-sample(1:20, 5, replace=TRUE)
> h3i2a3<-sample(1:20, 5, replace=TRUE)
> h3i3a1<-sample(1:20, 5, replace=TRUE)
> h3i3a2<-sample(1:20, 5, replace=TRUE)
> h3i3a3<-sample(1:20, 5, replace=TRUE)
> #Put into data frame
> hia<- 
> data 
> .frame 
> (h1i1a1 
> ,h1i1a2 
> ,h1i1a3 
> ,h1i2a1 
> ,h1i2a2 
> ,h1i2a3 
> ,h1i3a1 
> ,h1i3a2 
> ,h1i3a3 
> ,h1i4a1 
> ,h1i4a2 
> ,h1i4a3 
> ,h2i1a1 
> ,h2i1a2 
> ,h2i1a3 
> ,h2i2a1 
> ,h2i2a2 
> ,h2i2a3 
> ,h2i3a1 
> ,h2i3a2 
> ,h2i3a3 
> ,h2i4a1 
> ,h2i4a2 
> ,h2i4a3 
> ,h3i1a1,h3i1a2,h3i1a3,h3i2a1,h3i2a2,h3i2a3,h3i3a1,h3i3a2,h3i3a3)

  H <- array(c(h1i1a1, h1i1a2, h1i1a3, h1i2a1, h1i2a2, h1i2a3, h1i3a1,  
h1i3a2, h1i3a3, h1i4a1, h1i4a2, h1i4a3, h2i1a1, h2i1a2, h2i1a3,  
h2i2a1, h2i2a2, h2i2a3, h2i3a1, h2i3a2, h2i3a3, h2i4a1, h2i4a2,  
h2i4a3, h3i1a1, h3i1a2, h3i1a3, h3i2a1, h3i2a2, h3i2a3, h3i3a1,  
h3i3a2, h3i3a3), c(5,3,3,3) , dimnames=list(NULL, paste("i", 1:3,  
sep=""), paste("h", 1:3, sep=""),paste("a", 1:3, sep="") ))
  H

Would have been a lot easier to do this:
H < array( sample(1:20, 5*3*3*3, replace=TRUE), c(5,3,3,3) )

You can rearrange the names and dimensions to suit your tastes.


>
> i would like to produce a multidimensional array that looks  
> something like
>
>      h1     h2     h3
> i1    X       X      X
> i2    X       X      X
> i3    X       X      X
>
> a = 1 , TAZ = 1:5 ->for all 5 TAZ values or each row
>
>      h1     h2     h3
> i1    X       X      X
> i2    X       X      X
> i3    X       X      X
>
> a = 2 , TAZ = i in 1:5
>
>      h1     h2     h3
> i1    X       X      X
> i2    X       X      X
> i3    X       X      X
>
> a = 3 , TAZ = i in 1:5
>
> Basically a  matrix for all possible combination, so 27 matrices  
> altogether.
> But i guess its more like an array of arrays?  Again i am not being  
> lazy, i
> have been working on this all day but my code, like my sample stuff  
> is very
> inefficient and ultimately not working well so im not bothering to  
> post it.
> I know this will be complicated so thank you in advance, seeing the  
> end code
> will teach me a great deal.  Thanks
>
> Cheers,
> JR
> -- 
> View this message in context: http://n4.nabble.com/Create-a-multi-dimensional-array-from-a-data-frame-tp1068487p1068487.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.

David Winsemius, MD
Heritage Laboratories
West Hartford, CT