Gabor Csardi <[EMAIL PROTECTED]> wrote in news:[EMAIL PROTECTED]:
> On Sun, Apr 20, 2008 at 08:16:11PM +0000, David Winsemius wrote: >> Gabor Csardi <[EMAIL PROTECTED]> wrote in >> news:[EMAIL PROTECTED]: >> >> > Hmm, my understanding is different, >> > >> > m <- matrix(sample(10*10), ncol=10) >> > m2 <- rbind( m[1:5,], 1:10, m[6:10,] ) >> > m3 <- cbind( m[,1:8], 1:10, m[,9:10] ) >> >> I read the question the same way and, in response to the part of >> the question asking for no temporary matrix, offer this refinement >> on your suggestion: >> >> m <- rbind( m[1:5,], 1:10, m[6:10,] ) # row insertion or ... >> >> # not to be followed by, but rather instead column insertion .. >> m <- cbind( m[,1:8], 1:10, m[,9:10] ) > > There might be something wrong with my eyes, but where is the > refinement here? Your lines are literally the same as mines. There > is no temporary matrix here, m2 and m3 are the results, he wanted > either between row 5 and 6 _OR_ column 8 and 9. > > Oh, if you mean that we immediately put back the result into 'm', > then 1) it does not really matter, R will create a temporary matrix > internally anyway, Am I correct in assuming that after the creation of m by way of a temporary matrix that the temporary matrix would then be available for garbage collection, whereas if both m and m2 were created, there would be more memory occupied by the two objects? -- David Winsemius ______________________________________________ R-help@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.
