[Biomod-commits] Questions about BIOMOD

Wilfried Thuiller wilfried.thuiller at ujf-grenoble.fr
Wed Jul 22 07:01:58 CEST 2009


Dear Chris,
> We have two questions:
>
> 1) The first question about BIOMOD deals with the independent 
> explanatory and independent response inputs. Since we have the "truth" 
> in the shrub census data and are sampling from it, we would like to 
> compare the sample against the "truth." In other words, instead of 
> splitting our sample data into training and testing sets, the split 
> would be the sample and the "truth." I have been reading in the manual 
> about the independent inputs, and I am confused as to whether or not 
> this is where we would place the "truth" data. I have included an 
> example of the initial.state below:
>
> > Initial.State(Response = sample[,c(27, 28, 29, 30, 31,32,)], 
> Explanatory = sample[,15:26],
> + IndependantResponse = alldata[,c(27, 28, 29, 30, 31,32,)], 
> IndependentExplanatory = alldata[,15:26]
>
> so the object "sample" is a table of our sample of the census, and the 
> object "alldata" is the census dataset. Is this correct? If not, is 
> there a way to use the census data set to test the sampled data in the 
> models?
This is perfectly correct.
> 2) We are running multiple batches in a simulation and errors caused 
> by small sample sizes are causing the simulations to crash. We would 
> like to allow the modeling to stop but the simulation to loop to the 
> next larger sample size and continue. Is there a way to trap the 
> errors and prevent the modeling package from stopping all together?
Unfortunately R does not handle errors as far as I know, as you can do 
in C or Java.  I am curious to hear if someone has an idea.

One way to cope with this problem, is to have a filter based on 
prevalence. You first estimate the sample size of the species. If the 
species has a sample size smaller than 10 or 15 points, remove it before 
running BIOMOD.
Quite primitive but probably efficient.

Best
Wilfried



>
> Chris
>
> Christopher A. Walter
> Biological Science Technician
> USGS Leetown Science Center
> 11649 Leetown Road
> Kearneysville, WV 25430
> Phone: 304 724-4479
>
> "Its easy not to think when you're not told that much."
>    -Justin Pierre
> ------------------------------------------------------------------------
>
> _______________________________________________
> Biomod-commits mailing list
> Biomod-commits at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/biomod-commits
>   

-- 
--
Wilfried Thuiller
Laboratoire d'Ecologie Alpine, UMR-CNRS 5553
Université J. Fourier
BP 53, 38041 Grenoble Cedex 9, France
Tel: +33 (0)4 76 63 54 53
Fax: +33 (0)4 76 51 42 79

Email: wilfried.thuiller at ujf-grenoble.fr
Home page: http://www.will.chez-alice.fr
Website: http://www-leca.ujf-grenoble.fr/equipes/tde.htm

FP6 European MACIS project: http://www.macis-project.net
FP6 European EcoChange project: http://www.ecochange-project.eu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20090722/75fc4f87/attachment.htm 


More information about the Biomod-commits mailing list