[Biomod-commits] Projection Function in BIOMOD version 1.1-7.03

Damien Georges damien.georges2 at gmail.com
Tue Aug 28 12:32:38 CEST 2012

Hi  Matthew,

The bug you reported is quite mysterious. The loadMethod function is not 
a BIOMOD one.. and runing a similar example on my computer (Linux) not 
crashes but the BinaryTransformation function has changed recently so 
make sure to work with the last version of BIOMOD.

Even if it's not really the best solution, I guess you can remove the 
error by not doing binary transformation of your projections (setting 
BinRoc=F, BinTSS=F, BinKappa=F in projection function)

That being said, for your future analyses, I strongly recommend you to 
work with biomod2 package (BIOMOD realease).



On 10/08/2012 21:31, Matthew Moskwik wrote:
> I'm using BIOMOD version 1.1-7.03 in R 2.15.1 on a Linux cluster.  
> I've written an R script which builds models and then uses the 
> Projection function to project onto several future climate surfaces.  
> When the script reaches the Projection function, I receive the 
> following error:
> Error in BinaryTransformation(g[, a, Nrep, jj], 
> as.numeric(Evaluation.results.Roc[[i]][a, :
> could not find function "loadMethod"
> Calls: Projection -> BinaryTransformation
> Execution halted
> I've run the same script on a local Windows machine with the same 
> version of R and BIOMOD with no problems.  However, if I use the 
> previous BIOMOD version 1.1-7.02 on the Linux cluster, I do not 
> receive this error (the script completes without any errors).  Thus, 
> the error only appears when running the most current version of BIOMOD 
> on the Linux cluster.  I should also add that the installed R packages 
> on my Windows machine are identical to those installed on the Linux 
> cluster.  Any ideas what might be causing this problem?   Thanks.
> I wanted to use the most recent version of BIOMOD, as it calculates 
> evaluation metrics (Roc, TSS, Kappa) for independent data for each 
> replicate, whereas version 1.1-7.02 does not.
> Any help is appreciated.  Thanks.
> Matthew Moskwik
> _______________________________________________
> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20120828/ff910798/attachment.html>

More information about the Biomod-commits mailing list