Skip to content

Run out of memory

2 messages · Hongli Li, Brian Ripley

Original
Hongli Li
# 
I am trying to use R to cluster 7129 samples, my data set is a 7129 x 38 matrix,  when I trying to get the distance matrix using function dist( ), the memory exhausted, and I tried to set the memory when I run R by 
R --vsize=250M --nsize=1000k

no matter what I set for vsize the result is the same, it says:

Error: heap memory (256000kb) exhausted [need 198498Kb more]
                                                                        (this doesn't change)

What can I do to solve this?

Thank you  very much!

Hongli Li
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://stat.ethz.ch/pipermail/r-help/attachments/20010402/21156492/attachment.html
Brian Ripley
#
On Mon, 2 Apr 2001, Hongli Li wrote:

            

      
      
      
    

        
matrix,  when I trying to get the distance matrix using function dist( ), the
memory exhausted, and I tried to set the memory when I run R by

            

      
      
      
    

        
That's not the way to do it in recent versions of R.

            

      
      
      
    

        
doesn't change)

            

      
      
      
    

        
Upgrade your R and your computer's RAM?   To store the result needs
200Mb on my calculation, and then you need to do something with it.
It worked in about 5 minutes and 400Mb on a 1Gb 600MHz Linux machine.

However, what are you then going to do with this?  Hierarchical clustering
on 7000 points is not very sensible, and kmeans does not need a distance
matrix.