[R] hclust error
ramasamy at cancer.org.uk
Mon Jul 19 17:56:07 CEST 2004
Just out of curiosity, have you tried Mike Eisen's software Cluster
(http://rana.lbl.gov/EisenSoftware.htm). Try version 2.12 instead as
version 2.20 appears to give wrong results.
I never tried with 32000 rows, but when I tried a 10000 row last year it
seem to produce the results in a few minutes whereas R took either too
long or ran out of memory. Let us know if it works.
I assume you are trying to cluster the rows/genes here in which case the
number of columns/samples will not matter in your case.
On Mon, 2004-07-19 at 16:14, Liaw, Andy wrote:
> This has been asked on R-help several times in the past.
> The `dist' object for the data of that size would be:
> > 32513*32512/2
>  528531328
> To store a vector of that size in double precision (8 bytes), you'd need:
> > 32513*32512/2*8/1024^2
>  4032.374
> I.e., that's nearly 4 GB. Is your computer capable of doing that?
> > From: Herman, David (NIH/NIMH)
> > Hello,
> > I'm trying to do a cluster analysis on a large
> > data set. I
> > tried it out with a smaller one first, but I got this error:
> > > hc<-hclust(dist(x),"ave")
> > Error: cannot allocate vector of size 4129151 Kb
> > The data sample used (i.e. "x") is a numerical data set of
> > size 32513 by 31
> > Does anyone know how I can do this analysis? Is R capable of
> > this data
> > size?
> > Ultimately I wanna do an analysis on a dataset of size 32513 by 220
> > Thanks!
> > dave
> > [[alternative HTML version deleted]]
> > ______________________________________________
> > R-help at stat.math.ethz.ch mailing list
> > https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide!
> > http://www.R-project.org/posting-guide.html
> R-help at stat.math.ethz.ch mailing list
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
More information about the R-help