Take a look at this: x <- matrix(1:9, 3) x[2,2] <- Inf x[3,1] <- NA
rowMeans(x * is.finite(x), na.rm = TRUE) I at first thought you could simply use x[is.finite(x)] but that looses the matrix-ness of it so instead, we use the fact that 0 * Inf = NaN which gets killed by na.rm = TRUE. Also see my notes inline below. Best Michael On Sat, Jun 9, 2012 at 6:00 AM, Trying To learn again <[email protected]> wrote: > Hi all, > > I have a csv matrix "KT.csv" and it has Inf and NA Incidentally, this probably signals some confusion -- there's no such thing as a "csv matrix" in R. If you read in a csv you will usually (always?) get a data.frame which is most definitely not a matrix. > > I want to calculate the mean of each row so I use > > rowMeans(KT,na.rm = TRUE) but with this Inf cannot be omminted. This was your best guess -- see above. > > I´m trying to use before running rowMeans(KT,na.rm = TRUE) > > KT<-range(KT,finite=TRUE) This only replaces KT by its min and max loosing your data entirely. Not at all what you were looking for. > > but it doesn´t works... > > Do you know a simple way to ommit Inf en the calculations? > > Many thanks. > > I have tried also to replace Inf by NA bt also I dind´n t get it.... > > I feel....like a mule... > > ______________________________________________ > [email protected] 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. > ______________________________________________ [email protected] 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.
