[R] Memory issue with svm modeling in R

[Jessica Streicher]

Well, i'm no expert on these topics, but if its 2.7 gig and R can maximally use 2gig, then the easiest solution would be giving R more memory. Did you read through help(memory.size) as the error suggested?

try calling memory.size(T) or memory.limit(3000) and see if it works.

I don't have any experience with either Rstudio or Amazon whatever. The local system seems to be windows so the above might work, don't know the other, you might need to change the memory limit at startup of the console if its not.

On 22.10.2012, at 10:18, Vignesh Prajapati wrote:

> Hello Jessica,
> 
> Thanks for inform this and  very sorry for inconvenience, Here I have attached two Files
> 1.  crash.png- For Issue with Amazon Instance
> 2.  localmachine_error.bmp - for Issue with local machine
> 
> Thanks
> 
> On Mon, Oct 22, 2012 at 1:42 PM, Jessica Streicher <j.streicher at micromata.de> wrote:
> Hello Vignesh, we did not get any attachments, maybe you could upload them somewhere?
> 
> On 19.10.2012, at 09:46, Vignesh Prajapati wrote:
> 
> > As I found the memory problem with local machine/micro instance(amazon) for
> > building SVM model in R on large dataset(2,01,478 rows with 11 variables),
> > then I have migrated our micro instance to large instance at Amazon. Still
> > I have memory issue with large amazon instance while developing R model for
> > this dataset due to large size. I have attached the snap of error with
> > local machine(localmachine_error.bmp) and amazon instance(crash.png ) with
> > this post.
> >
> > Issue on local Machine ::
> >
> > [image: enter image description here]
> >
> > Issue on Amazon large Instance ::
> >
> > [image: enter image description here]
> >
> > Can any one suggest me for the solution of this issue.?
> >
> > Thanks
> >
> > Vignesh
> >
> >       [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > 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.
> 
> 
> <crash.png><localmachine_error.bmp>