[Biomod-commits] use YWeights or not: nothing changes

Damien Georges damien.georges2 at gmail.com
Tue Feb 12 12:37:50 CET 2013


Hi Luigi,

Are you working with Pseudo-absences ?
Because in this case, weights are ignored and but some are created to 
fit a given prevalence (.5 by default).

Cheers,

Damien

On 12/02/2013 12:38, Luigi Maiorano wrote:
> Hi all
>
> I have a question about using Yweights in BIOMOD_Modeling. Basically if I set it to null, or if I give it values nothing changes (and when I say nothing, I am saying that the cell values for all models are identical!). Is there a bug here? any hint?
> I'm in R2.15.2 with Biomod2 2.0.0. The two command lines that I am using are:
> # RUN THE MODELS WITH WEIGHTS
> ftBiomodModelOut_W <- BIOMOD_Modeling(ftBiomodData, models=c('GLM', 'GBM', 'GAM', 'ANN', 'CTA', 'MARS', 'RF'), models.options=ftBiomodOption, NbRunEval=1, DataSplit=80, Yweights=ftWeight, VarImport=3,models.eval.meth=c('TSS', 'ROC'), SaveObj=TRUE, rescal.all.models=TRUE)
>
> #RUN THE MODELS WITHOUT WEIGHTS
> ftBiomodModelOut <- BIOMOD_Modeling(ftBiomodData, models=c('GLM', 'GBM', 'GAM', 'ANN', 'CTA', 'MARS', 'RF'), models.options=ftBiomodOption, NbRunEval=1, DataSplit=80, Yweights=NULL, VarImport=3,models.eval.meth=c('TSS', 'ROC'), SaveObj=TRUE, rescal.all.models=TRUE)
>
>
> thanks for the help
>
> Luigi
>
> -------------------------------------------------------------------------------
> Dr. Luigi Maiorano
> Dipartimento di Biologia e Biotecnologie "Charles Darwin"
> Università di Roma "La Sapienza"
> viale dell'Università 32
> 00185 Roma
> tel: +39 0649694262
> fax: +39 06491135
>
> Email: luigi.maiorano at uniroma1.it
>
>
>
> _______________________________________________
> 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



More information about the Biomod-commits mailing list