[R] Re : Memory in R.

[Gavin Simpson]

On Wed, 2008-04-02 at 16:46 +0200, R RR wrote:
> Dear R users,
> Many thanks for your answers.
> I've made much progress since my last posting.
> 
> I have now the followuing problem. I've run the GAM model
> 
> mygam <- gam(Y ~ factor(year)
> + m1.q02 + m1.q05y + m1.q05y2 + m1.q06 + m4b.q05 + m4b.q052 + m5a.q01
> + depratio + depratio2 + residence10y + urbrur + factor(prefect)
> + m1.q02_ps
> + m1.q05y_ps
> + m1.q05y2_ps
> + m1.q06_ps
> + m4b.q05_ps
> + m4b.q052_ps
> + m5a.q01_ps
> + depratio_ps
> + depratio2_ps
> + residence10y_ps
> + urbrur_ps
>                   +factor(hhid), data=cwp2)
> 
> and obtained the following error code:

No idea if this will help, but seeing as there doesn't appear to be any
smooth terms in that model (s() or lo() depending on whether this is
mgcv:::gam or gam:::gam), and you are fitting a Gaussian model (no
family so default used), why not fit this using lm? You might find this
takes less memory than the way gam handles making the model matrix.

HTH

G

> 
> Erreur : impossible d'allouer un vecteur de taille 236.3 Mo
> (in english: cannot allocate a 236.3 Mo vector memory).
> 
> I have 7237 observations in my data.
> 
> Is there any way to increase the memory to fit the model?
> 
> Best regards.
> 
> Amadou DIALLO
> 
> 
> 
> 2008/4/1, Liviu Andronic <landronimirc at gmail.com>:
> > Hello Amadou,
> >
> > On Mon, Mar 31, 2008 at 4:53 PM, R RR <rstat.diallo at gmail.com> wrote:
> > >  1 - I want to increase memory, but I can't find
> > >  the right way. E.g: in stata, you just type "set memory 1g".
> > >  When I try to load huge datasets, R crashes.
> >
> > You will find this recent thread [1] interesting. You'd also want to
> > check packages filehash, ff and sqldf.
> > Regards,
> > Liviu
> >
> > [1] http://www.nabble.com/How-to-read-HUGE-data-sets--tt15729830.html
> >
> 
> ______________________________________________
> 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.
-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%