Re: [R] replicate matrix blocks different numbers of times into new matrix

From: Ralph S. <ruffel1_at_hotmail.com>
Date: Sun, 20 Jul 2008 04:39:42 +0000

yes, this works, thank yo. very smart solution!

-R


> Date: Sat, 19 Jul 2008 19:55:58 -0400
> From: jholtman_at_gmail.com
> To: ruffel1_at_hotmail.com
> Subject: Re: [R] replicate matrix blocks different numbers of times into new matrix
> CC: r-help_at_r-project.org
> 
> Will this do it:
> 

>> # determine the row numbers of each of the factors
>> x.row <- split(seq(nrow(x)), x[,1])
>> # process the data and replicate the rows
>> result <- lapply(seq_along(x.row), function(.fact){
> +     x[rep(x.row[[.fact]], n[.fact]),]
> + })

>> do.call(rbind, result)
> [,1] [,2] > [1,] 1 1 > [2,] 1 3 > [3,] 1 1 > [4,] 1 3 > [5,] 2 4 > [6,] 2 4 >> > > > On Sat, Jul 19, 2008 at 7:17 PM, Ralph S. wrote: >>

>> Actually not quite - my mistake, since I oversimplified the problem I have.
>>

>> Here is a more realistic x matrix (plus some additional information):
>>

>> # the data
>> x<-matrix(c(1,1,2,1,3,4),3,2)
>>

>> # number of factors
>> n_f<-2
>>

>> # number of rows taken by each factor
>> f_length <- c(2,1)
>>

>> # number of repetitions I want for the first and second factor
>> # actually, always the same factor of expansion: both factors are to be replicated n times
>> n<-c(2,2)
>>

>> I want something like
>>

>> [,1] [,2]
>> [1,] 1 1
>> [2,] 1 3
>> [3,] 1 1
>> [4,] 1 3
>> [5,] 2 4
>> [6,] 2 4
>>

>> but it is only easy to get
>> [,1] [,2]
>> [1,] 1 1
>> [2,] 1 1
>> [3,] 1 3
>> [4,] 1 3
>> [5,] 2 4
>> [6,] 2 4
>>

>> I am not sure about the first target matrix.
>>

>> I could loop through each level of the factor, use a "which(x[,1]==f[k]" to get the row indices for each factor f[k] and then replicate those indices n times and append them to the result for the previous level of the factor. This does not seem efficient, given that I actually have a large matrix with more than 600 factors.
>>

>> Sorry for the initial misspecification - any ideas how I could solve my problem?
>>

>> Best,
>>

>> Ralph
>>

>> ----------------------------------------
>>> Date: Sat, 19 Jul 2008 21:39:25 +0200 >>> From: p.dalgaard_at_biostat.ku.dk >>> To: ruffel1_at_hotmail.com >>> CC: r-help_at_r-project.org >>> Subject: Re: [R] replicate matrix blocks different numbers of times into new matrix >>> >>> Ralph S. wrote: >>>> Hi, >>>> >>>> I am trying to replicate blocks of a matrix (defined by factors) into another matrix, but an unequal, consecutive number of times for each factor. >>>> >>>> I need to find an elegant and fast way to do this, so loops will not work. >>>> >>>> An example of what I am trying to do is the following: >>>> >>>> # the data - first column entries are both data and the two factors >>>> x<-matrix(c(1,2,3,4),2,2) >>>> >>>>> x >>>>> >>>> [,1] [,2] >>>> [1,] 1 3 >>>> [2,] 2 4 >>>> >>>> # the number of repetitions of the first and second factor >>>> n<-c(1,3) >>>> >>>> This is what I want as output: >>>> >>>> [,1] [,2] >>>> [1,] 1 3 >>>> [2,] 2 4 >>>> [3,] 2 4 >>>> [4,] 2 4 >>>> >>>> >>>> Any ideas how to get there? I have tried using tapply with combination of rep, but this does not work (I need 1 and then 3 replications). >>>>

>> [[elided Hotmail spam]]
>>>>
>>> Will this do?
>>>
>>> x[rep(1:2,n),]
>>>
>>> --
>>> O__ ---- Peter Dalgaard ster Farimagsgade 5, Entr.B
>>> c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
>>> (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
>>> ~~~~~~~~~~ - (p.dalgaard_at_biostat.ku.dk) FAX: (+45) 35327907
>>>
>>

>> _________________________________________________
>> _________________________________________________________________
>> >>

>> family_safety_072008
>> ______________________________________________
>> 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.
>>
> 
> 
> 
> -- 
> Jim Holtman
> Cincinnati, OH
> +1 513 646 9390
> 
> What is the problem you are trying to solve?

_________________________________________________________________


enger2_072008



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. Received on Sun 20 Jul 2008 - 04:44:22 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Sun 20 Jul 2008 - 08:31:50 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive