R › R help

Changing selected elements of an array

Classic

List

Threaded

5 messages Options

Øystein Godøy

Changing selected elements of an array

Hi!

I have a matrix defined on geographical positions (through) row and column
names. I need to change a number of elements in this matrix using the
information of a data.frame containing geographical positions and a number of
variables.

Changing the value of one specific element is easy, but changing on a number
of selected positions seems more difficult. When I use the geographical
positions of the data.frame as index, blobs are changed and not individual
elements.

How can I circumvent this feature of R?

E.g: in this situation I would like the downward diagonal to be changed, not
the square box...

> tmpmat <-
matrix(NA,nrow=9,ncol=10,dimnames=list(formatC(tmplat,format="f",digits=2),formatC(tmplon,format="f",digits=2)))
> tmpmat
8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
55.25 NA NA NA NA NA NA NA NA NA NA
56.75 NA NA NA NA NA NA NA NA NA NA
57.75 NA NA NA NA NA NA NA NA NA NA
58.00 NA NA NA NA NA NA NA NA NA NA
59.50 NA NA NA NA NA NA NA NA NA NA
62.75 NA NA NA NA NA NA NA NA NA NA
64.50 NA NA NA NA NA NA NA NA NA NA
66.25 NA NA NA NA NA NA NA NA NA NA
67.25 NA NA NA NA NA NA NA NA NA NA
> tmpmat[c("58.00","59.50","62.75"),c("15.00","15.75","18.00")] <- 300
> tmpmat
8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
55.25 NA NA NA NA NA NA NA NA NA NA
56.75 NA NA NA NA NA NA NA NA NA NA
57.75 NA NA NA NA NA NA NA NA NA NA
58.00 NA NA NA NA 300 300 300 NA NA NA
59.50 NA NA NA NA 300 300 300 NA NA NA
62.75 NA NA NA NA 300 300 300 NA NA NA
64.50 NA NA NA NA NA NA NA NA NA NA
66.25 NA NA NA NA NA NA NA NA NA NA
67.25 NA NA NA NA NA NA NA NA NA NA

While I wanted:
> tmpmat
8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
55.25 NA NA NA NA NA NA NA NA NA NA
56.75 NA NA NA NA NA NA NA NA NA NA
57.75 NA NA NA NA NA NA NA NA NA NA
58.00 NA NA NA NA 300 NA NA NA NA NA
59.50 NA NA NA NA NA 300 NA NA NA NA
62.75 NA NA NA NA NA NA 300 NA NA NA
64.50 NA NA NA NA NA NA NA NA NA NA
66.25 NA NA NA NA NA NA NA NA NA NA
67.25 NA NA NA NA NA NA NA NA NA NA

All the best
Øystein
--
Dr. Oystein Godoy
Norwegian Meteorological Institute
P.O.BOX 43, Blindern, N-0313 OSLO, Norway
Ph: (+47) 2296 3000 (switchb) 2296 3334 (direct line)
Fax:(+47) 2296 3050 Institute home page: http://met.no/

______________________________________________
[hidden email] 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.

Joshua Wiley-2

Re: Changing selected elements of an array

Hi Øystein,

You can do it by passing a matrix for indexing instead of two vectors.
Here's an example:

tmpmat <- matrix(NA, nrow = 10, ncol = 10,
dimnames = list(letters[1:10], LETTERS[1:10]))

tmpmat[cbind(c("d", "e", "f"), c("D", "E", "F"))] <- 100
tmpmat

The matrix is created using cbind() to columnwise bind the two vectors
together. Then it does what you want I think.

Hope this helps,

Josh

On Mon, May 21, 2012 at 7:15 AM, Øystein Godøy <[hidden email]> wrote:

> Hi!
>
> I have a matrix defined on geographical positions (through) row and column
> names. I need to change a number of elements in this matrix using the
> information of a data.frame containing geographical positions and a number of
> variables.
>
> Changing the value of one specific element is easy, but changing on a number
> of selected positions seems more difficult. When I use the geographical
> positions of the data.frame as index, blobs are changed and not individual
> elements.
>
> How can I circumvent this feature of R?
>
> E.g: in this situation I would like the downward diagonal to be changed, not
> the square box...
>
>> tmpmat <-
> matrix(NA,nrow=9,ncol=10,dimnames=list(formatC(tmplat,format="f",digits=2),formatC(tmplon,format="f",digits=2)))
>> tmpmat
> 8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
> 55.25 NA NA NA NA NA NA NA NA NA NA
> 56.75 NA NA NA NA NA NA NA NA NA NA
> 57.75 NA NA NA NA NA NA NA NA NA NA
> 58.00 NA NA NA NA NA NA NA NA NA NA
> 59.50 NA NA NA NA NA NA NA NA NA NA
> 62.75 NA NA NA NA NA NA NA NA NA NA
> 64.50 NA NA NA NA NA NA NA NA NA NA
> 66.25 NA NA NA NA NA NA NA NA NA NA
> 67.25 NA NA NA NA NA NA NA NA NA NA
>> tmpmat[c("58.00","59.50","62.75"),c("15.00","15.75","18.00")] <- 300
>> tmpmat
> 8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
> 55.25 NA NA NA NA NA NA NA NA NA NA
> 56.75 NA NA NA NA NA NA NA NA NA NA
> 57.75 NA NA NA NA NA NA NA NA NA NA
> 58.00 NA NA NA NA 300 300 300 NA NA NA
> 59.50 NA NA NA NA 300 300 300 NA NA NA
> 62.75 NA NA NA NA 300 300 300 NA NA NA
> 64.50 NA NA NA NA NA NA NA NA NA NA
> 66.25 NA NA NA NA NA NA NA NA NA NA
> 67.25 NA NA NA NA NA NA NA NA NA NA
>
> While I wanted:
>> tmpmat
> 8.00 9.50 9.75 14.25 15.00 15.75 18.00 20.50 21.75 25.25
> 55.25 NA NA NA NA NA NA NA NA NA NA
> 56.75 NA NA NA NA NA NA NA NA NA NA
> 57.75 NA NA NA NA NA NA NA NA NA NA
> 58.00 NA NA NA NA 300 NA NA NA NA NA
> 59.50 NA NA NA NA NA 300 NA NA NA NA
> 62.75 NA NA NA NA NA NA 300 NA NA NA
> 64.50 NA NA NA NA NA NA NA NA NA NA
> 66.25 NA NA NA NA NA NA NA NA NA NA
> 67.25 NA NA NA NA NA NA NA NA NA NA
>
> All the best
> Øystein
> --
> Dr. Oystein Godoy
> Norwegian Meteorological Institute
> P.O.BOX 43, Blindern, N-0313 OSLO, Norway
> Ph: (+47) 2296 3000 (switchb) 2296 3334 (direct line)
> Fax:(+47) 2296 3050 Institute home page: http://met.no/
>
> ______________________________________________
> [hidden email] 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.

--
Joshua Wiley
Ph.D. Student, Health Psychology
Programmer Analyst II, Statistical Consulting Group
University of California, Los Angeles
https://joshuawiley.com/
______________________________________________
[hidden email] 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.

Øystein Godøy

Re: Changing selected elements of an array

Hi Joshua,

Many thanks for your quick reply.

> You can do it by passing a matrix for indexing instead of two vectors.
> Here's an example:
>
> tmpmat <- matrix(NA, nrow = 10, ncol = 10,
> dimnames = list(letters[1:10], LETTERS[1:10]))
>
> tmpmat[cbind(c("d", "e", "f"), c("D", "E", "F"))] <- 100
> tmpmat
>
> The matrix is created using cbind() to columnwise bind the two vectors
> together. Then it does what you want I think.
>
> Hope this helps,

This looks interesting and is what I want, but I am not fully understanding
the output I receive. The input array has 100 elements while the resulting
vector after replacement is 106 elements long. I have tried to understand the
manual on this, but it is yet not obvious for me how I am supposed to handle
this output.

All the best
Øystein
--
Dr. Oystein Godoy
Norwegian Meteorological Institute
P.O.BOX 43, Blindern, N-0313 OSLO, Norway
Ph: (+47) 2296 3000 (switchb) 2296 3334 (direct line)
Fax:(+47) 2296 3050 Institute home page: http://met.no/

______________________________________________
[hidden email] 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.

Joshua Wiley-2

Re: Changing selected elements of an array

On Mon, May 21, 2012 at 9:27 AM, Øystein Godøy <[hidden email]> wrote:

> Hi Joshua,
>
> Many thanks for your quick reply.
>
>> You can do it by passing a matrix for indexing instead of two vectors.
>> Here's an example:
>>
>> tmpmat <- matrix(NA, nrow = 10, ncol = 10,
>> dimnames = list(letters[1:10], LETTERS[1:10]))
>>
>> tmpmat[cbind(c("d", "e", "f"), c("D", "E", "F"))] <- 100
>> tmpmat
>>
>> The matrix is created using cbind() to columnwise bind the two vectors
>> together. Then it does what you want I think.
>>
>> Hope this helps,
>
> This looks interesting and is what I want, but I am not fully understanding
> the output I receive. The input array has 100 elements while the resulting
> vector after replacement is 106 elements long. I have tried to understand the
> manual on this, but it is yet not obvious for me how I am supposed to handle
> this output.

I cannot reproduce this behavior. On my system, the output has 100 elements.

> All the best
> Øystein
> --
> Dr. Oystein Godoy
> Norwegian Meteorological Institute
> P.O.BOX 43, Blindern, N-0313 OSLO, Norway
> Ph: (+47) 2296 3000 (switchb) 2296 3334 (direct line)
> Fax:(+47) 2296 3050 Institute home page: http://met.no/

--
Joshua Wiley
Ph.D. Student, Health Psychology
Programmer Analyst II, Statistical Consulting Group
University of California, Los Angeles
https://joshuawiley.com/

______________________________________________
[hidden email] 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.

Øystein Godøy

Re: Changing selected elements of an array

Joshua Wiley wrote on 2012-05-21
...
> > This looks interesting and is what I want, but I am not fully
> > understanding the output I receive. The input array has 100 elements
> > while the resulting vector after replacement is 106 elements long. I
> > have tried to understand the manual on this, but it is yet not obvious
> > for me how I am supposed to handle this output.
>
> I cannot reproduce this behavior. On my system, the output has 100
> elements.
...
That is valuable information for me as your result is inline with what I
understand from the manual. I am using R version 2.10.1 (2009-12-14) through
the system implementation following Ubuntu Lucid. I have to look further into
the cause for this. I just doublechecked your sample code on another system
and got the same result there.

Thanks for your effort.

All the best
Øystein
--
Dr. Oystein Godoy
Norwegian Meteorological Institute
P.O.BOX 43, Blindern, N-0313 OSLO, Norway
Ph: (+47) 2296 3000 (switchb) 2296 3334 (direct line)
Fax:(+47) 2296 3050 Institute home page: http://met.no/

______________________________________________
[hidden email] 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.