[R] index to select rows of a large matrix

[tsunhin wong]

Dear all,

I found the answer:
intersect()

- John

On Thu, May 21, 2009 at 7:32 PM, tsunhin wong <thjwong at gmail.com> wrote:
> Dear Jim,
> Thanks for your suggestion.
>
> I have a follow-up question for fellow R Users that have followed this thread:
>
> *I used to create two lists by some very flexible criteria to compare
> from "iData" and pass the two lists to *ANOTHER FUNCTION* that further
> decompose the two lists and do some case by case analysis:
>
> comp.lst[[1]] <- subset(iData,(Session==1 & TrialList==2 & BadTrial==0
> & Correct==1 & Critical.BT>0))
> comp.lst[[2]] <- subset(iData,(Session==2 & TrialList==2 & BadTrial==0
> & Correct==1 & Critical.BT>0))
>
> *The function will then do some subject by subject processing:
>
> tmplist <- comp.lst[[1]]
> subject <- unique(tmplist$Subject)
> Psubjmatrix <- NULL
> for(j in 1:length(subj)) { ##divide the list and process per subject in the list
>        subjtriallist <- subset(tmplist,(Subject==subj[j])) ##divide
> the list and process per subject in the list
>        print(paste("Start working on Subject ",subj[j],sep=""),quote=FALSE)
>        ##The sub-divided list is passed to another function to
> extract data from corresponding object
>        Ptrialmatrix <- SeekObject(subjtriallist,seconds,samplesize,sdmax);
>        ##add rowmean to subject pupil matrix
>        Psubjmatrix <- cbind(Psubjmatrix, rowMeans(Ptrialmatrix,
> na.rm=TRUE, dims=1))
> }
>
> *But now, I have a pre-processed the objects and cooked up a
> 1500x20000 matrix, and there is no need to get these objects one by
> one for extraction, I may just do the following:
>
> Ptrialmatrix <- PMasterMatrix[subjindex,,drop=FALSE]
>
> *However, on passing the index to the function, I cannot further
> sub-divide the list index by Subject.
>
> I have been thinking of passing the criteria to within the function.
> But... what smarter thing I can do to make it works?
>
> - John
>
>
> On Thu, May 21, 2009 at 10:33 AM, jim holtman <jholtman at gmail.com> wrote:
>> Assuming that you get the list of indices into iData for the criteria, then
>> you can use that to get the appropriate rows:
>>
>> indx <- which(iData >5)  # or whatever your criteria is
>> DataSeq[indx,, drop=FALSE]  # gives you a subset matrix of just the rows you
>> are interested in.
>>
>> On Thu, May 21, 2009 at 10:10 AM, tsunhin wong <thjwong at gmail.com> wrote:
>>>
>>> Dear R Users,
>>>
>>> I have created a 1500 x 20000 data frame - DataSeq. Each of the 1500
>>> rows represents a data sequence.
>>> I have another data frame iData that stores the information of these
>>> 1500 data sequences in the same order, for example, condition, gender,
>>> etc.
>>>
>>> If I use "subset" to select certain groups within iData according to
>>> some criteria that I have set, e.g. condition, gender
>>> Then how can I used the retrieved subset of iData to point to and
>>> retrieve corresponding rows in the DataSeq data.frame for
>>> manipulations and analysis?
>>>
>>> I hope some of you can give me some idea!
>>> Thank you very much!!!
>>>
>>> - John
>>>
>>> ______________________________________________
>>> 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 that you are trying to solve?
>>
>