R help - Feb 2008 - Analysis with spatstat and Kcross() requires to much memory

Hi

I am running analysis with Kcross from the package spatstat and I am getting
a message that R can  not allocate enough memory for a vector of 900MB.
R seems to be running towards the 2GB limit per process.

The dataset is not to big (ca 3000 points) but the mask for the points is
extremely irregular (a buffer around roads which have been sampled) and I
can do trhe analysis if I use a rectangular mask covering the whole area
(which is obviously not correct).

My question is: is there a way of splitting the dataset or changing the mask
son that I can do the analysis, as I don't want to move to a 64bit computer
(if not necessary).

Thanks,

Rainer

My computrer:

SuSE 10, 32 bit, 1GB ram, 1GB swap.

R:
> version
               _
platform       i686-pc-linux-gnu
arch           i686
os             linux-gnu
system         i686, linux-gnu
status
major          2
minor          6.2
year           2008
month          02
day            08
svn rev        44383
language       R
version.string R version 2.6.2 (2008-02-08)


-- 
Rainer M. Krug, Dipl. Phys. (Germany), MSc Conservation Biology (UCT)

Plant Conservation Unit Department of Botany
University of Cape Town
Rondebosch 7701
South Africa

	[[alternative HTML version deleted]]

Rainer M Krug <r.m.krug <at> gmail.com> writes:
> 
> Hi
> 
> I am running analysis with Kcross from the package spatstat and I am
getting
> a message that R can  not allocate enough memory for a vector of 900MB.
> R seems to be running towards the 2GB limit per process.
> 
> The dataset is not to big (ca 3000 points) but the mask for the points is
> extremely irregular (a buffer around roads which have been sampled) and I
> can do trhe analysis if I use a rectangular mask covering the whole area
> (which is obviously not correct).
Rainer,

Would it be possible for you to rasterise the road buffer (I'm assuming that
it
is a vector buffer), so that the window is an image mask rather than a polygon
window? My assumption would be that as.im() would get you there, but you'd
need
to watch the dimyx= argument.

It is also usually helpful to look at traceback() after such a failure, so that
you can see at which point in Kcross things fell apart - that is, which internal
function was being used. Then you can concentrate on finding a way of solving
that specific problem.

Roger
> 
> My question is: is there a way of splitting the dataset or changing the
mask
> son that I can do the analysis, as I don't want to move to a 64bit
computer
> (if not necessary).
> 
> Thanks,
> 
> Rainer
>