[18693] in s-news-athena
Re: [S] Sorting matrix elements, keeping track of original positions
daemon@ATHENA.MIT.EDU (Kim Elmore)
Wed Jul 21 16:51:46 1999
Date: Wed, 21 Jul 1999 15:46:37 -0500 (CDT)
From: Kim Elmore <elmore@nssl.noaa.gov>
To: S-News mail list <s-news@wubios.wustl.edu>
In-Reply-To: <Pine.HPP.3.95.990721163237.18605U-100000@jove.mps.ohio-state.edu>
Message-Id: <Pine.GSO.4.10.9907211539420.4946-100000@ev.nssl.noaa.gov>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
In *record* time, I recieved excellent replies, all of
which, in one way or another, do what I want. I'm amazed by the pool of
very bright people who so willingly share their knowledge. Thanks so
much.
I summarize them below, but the first is very convenient for may
problem of the moment:
On Wed, 21 Jul 1999, Joseph S. Verducci wrote:
> On second thought, perhaps you want
>
> positioner _ function(a){
> ro _ sapply(sort(a),function(x) row(a)[a==x])
> co _ sapply(sort(a),function(x) col(a)[a==x])
> cbind(ro,co)}
>
> which gives the row and column of the sorted data
>
> > a
> [,1] [,2] [,3]
> [1,] 27 24 29
> [2,] 21 22 28
> [3,] 23 26 25
>
> > positioner(a)
> ro co
> [1,] 2 1
> [2,] 2 2
> [3,] 3 1
> [4,] 1 2
> [5,] 3 3
> [6,] 3 2
> [7,] 1 1
> [8,] 2 3
> [9,] 1 3
From byers@research.att.com Wed Jul 21 15:41:40 1999
## Try this,
x <- matrix(rnorm(80), ncol=8, nrow=10)
out <- cbind( c(x), c(row(x)), c(col(x)))
out[order(out[,1]),]
### Simon Byers, AT&T Labs.
From: Joseph S. Verducci <jsv@stat.ohio-state.edu>
Is
matrix(rank(a),nrow(a),ncol(a))
what you want?
Example:
> a
[,1] [,2] [,3]
[1,] 27 24 29
[2,] 21 22 28
[3,] 23 26 25
> sort(a)
[1] 21 22 23 24 25 26 27 28 29
> matrix(rank(a),nrow(a),ncol(a))
[,1] [,2] [,3]
[1,] 7 4 9
[2,] 1 2 8
[3,] 3 6 5
From: Jim Hughes <hughes@diamond.cfas.washington.edu>
Kim -
Does this do what you want?
ord <- order(mat)
mat[ord] # orders the elements of the matrix mat
cbind(row(mat)[ord],col(mat)[ord]) # gives the row, col index of each
# ordered element in the original matrix
Jim
Kim Elmore, [N5OP, PP ASMEL/Glider 2232456]
"All of Meteorology is divided into three parts: Yes, No and Maybe. The
greatest of these is Maybe." -- The original Latin appears to be garbled.
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
