R help - Nov 2009 - How to transform the Matrix into the way I want it ???

Hi, R users,

I'm trying to transform a matrix A into B (see below). Anyone knows how to
do it in R? Thanks.

Matrix A (zone to zone travel time)

 zone z1 z2 z3  z1 0 2.9 4.3  z2 2.9 0 2.5  z3 4.3 2.5 0

B:

from to time z1 z1 0 z1 z2 2.9 z1 z3 4.3 z2 z1 2.9 z2 z2 0 z2 z3 2.5 z3 z1
4.3 z3 z2 2.5 z3 z3 0

The real matrix I have is much larger, with more than 2000 zones. But I
think it should be the same thing if I can transform A into B.

Thanks.

Garry

	[[alternative HTML version deleted]]

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of Hongwei Dong
> Sent: Monday, November 09, 2009 2:24 PM
> To: R-help Forum
> Subject: [R] How to transform the Matrix into the way I want it ???
> 
> Hi, R users,
> 
> I'm trying to transform a matrix A into B (see below). Anyone 
> knows how to
> do it in R? Thanks.
> 
> Matrix A (zone to zone travel time)
> 
>  zone z1 z2 z3  z1 0 2.9 4.3  z2 2.9 0 2.5  z3 4.3 2.5 0
> 
> B:
> 
> from to time z1 z1 0 z1 z2 2.9 z1 z3 4.3 z2 z1 2.9 z2 z2 0 z2 
> z3 2.5 z3 z1
> 4.3 z3 z2 2.5 z3 z3 0
Try
   data.frame(From=rownames(A)[row(A)], To=colnames(A)[col(A)],
Time=as.vector(A))
You probably do not want a matrix because all of its elements
must have the same type, forcing Time to be character instead
of numeric.

E.g.,
> A<-rbind(Anacortes=c(Anacortes=0,Seattle=90,"Mount
Vernon"=20),
+          Seattle=c(Anacortes=80, Seattle=0, "Mount Vernon"=60),
+          "Mount Vernon"=c(Anacortes=20,Seattle=70,"Mount
Vernon"=0))
> data.frame(From=rownames(A)[row(A)], To=colnames(A)[col(A)],
Time=as.vector(A))
          From           To Time
1    Anacortes    Anacortes    0
2      Seattle    Anacortes   80
3 Mount Vernon    Anacortes   20
4    Anacortes      Seattle   90
5      Seattle      Seattle    0
6 Mount Vernon      Seattle   70
7    Anacortes Mount Vernon   20
8      Seattle Mount Vernon   60
9 Mount Vernon Mount Vernon    0

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com  
> 
> The real matrix I have is much larger, with more than 2000 
> zones. But I
> think it should be the same thing if I can transform A into B.
> 
> Thanks.
> 
> Garry
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> 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.
>

This is not an answer to your question, but I have used SparseM  
package to represent large travel time matrices efficiently.

?as.matrix.ssr

if the traveltime matrix is symmetric.

On 9 Nov 2009, at 5:24PM, Hongwei Dong wrote:
> Hi, R users,
>
> I'm trying to transform a matrix A into B (see below). Anyone knows  
> how to
> do it in R? Thanks.
>
> Matrix A (zone to zone travel time)
>
> zone z1 z2 z3  z1 0 2.9 4.3  z2 2.9 0 2.5  z3 4.3 2.5 0
>
> B:
>
> from to time z1 z1 0 z1 z2 2.9 z1 z3 4.3 z2 z1 2.9 z2 z2 0 z2 z3 2.5  
> z3 z1
> 4.3 z3 z2 2.5 z3 z3 0
>
> The real matrix I have is much larger, with more than 2000 zones.  
> But I
> think it should be the same thing if I can transform A into B.
>
> Thanks.
>
> Garry
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.

On Nov 9, 2009, at 5:24 PM, Hongwei Dong wrote:
> Hi, R users,
>
> I'm trying to transform a matrix A into B (see below). Anyone knows  
> how to
> do it in R? Thanks.
>
> Matrix A (zone to zone travel time)
>
> zone z1 z2 z3  z1 0 2.9 4.3  z2 2.9 0 2.5  z3 4.3 2.5 0
>
 > ztz <- read.table(textConnection(" z1 z2 z3 \n z1 0 2.9 4.3 \n z2
2.9 0 2.5 \n z3 4.3 2.5 0"), header=T)
 > ztz
     z1  z2  z3
z1 0.0 2.9 4.3
z2 2.9 0.0 2.5
z3 4.3 2.5 0.0
> B:
>
> from to time z1 z1 0 z1 z2 2.9 z1 z3 4.3 z2 z1 2.9 z2 z2 0 z2 z3 2.5  
> z3 z1
> 4.3 z3 z2 2.5 z3 z3 0
Now taking some liberties with coercion:

 > as.data.frame.table(ztz, responseName="Time")
   Var1 Var2 Time
1   z1   z1  0.0
2   z2   z1  2.9
3   z3   z1  4.3
4   z1   z2  2.9
5   z2   z2  0.0
6   z3   z2  2.5
7   z1   z3  4.3
8   z2   z3  2.5
9   z3   z3  0.0


Or perhaps:

 > data.frame(stack(as.data.frame(ztz)), From=colnames(ztz))
   values ind From
1    0.0  z1   z1
2    2.9  z1   z2
3    4.3  z1   z3
4    2.9  z2   z1
5    0.0  z2   z2
6    2.5  z2   z3
7    4.3  z3   z1
8    2.5  z3   z2
9    0.0  z3   z3
>
> The real matrix I have is much larger, with more than 2000 zones.  
> But I
> think it should be the same thing if I can transform A into B.
>
> Thanks.
>
> Garry
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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