[Biomod-commits] Re : Error in ensemble forecasting

[Bruno Lafourcade]

Hi Samiya,

So apparently the function doesn't recognize any model for which to calculate some weights.
In other words, all your models are considered to have a score of 0 (or rather NA as I'd rather think).

I guess it might come from the 'weight.method' that you choose, did you run it in Models ?
Else, can you check the values in 4th column of the Evaluation.results of your method to see
what you have there.

Bruno.


 

-------
Bruno Lafourcade
Statistical tools engineer

Laboratoire d'Ecologie Alpine, bureau 308
CNRS - UMR 5553, 2233 rue de la piscine
38400 Saint Martin d'Hères
-------
 

 

-----E-mail d'origine-----
De : Samiya Selim <bs09sas at leeds.ac.uk>
A : biomod-commits at r-forge.wu-wien.ac.at
Envoyé le : Jeudi, 25 Mars 2010 14:40
Sujet : Re: [Biomod-commits] Error in ensemble forecasting




----- Forwarded message from bs09sas at leeds.ac.uk -----
     Date: Sun, 21 Mar 2010 19:09:00 +0000
     From: Samiya Selim <bs09sas at leeds.ac.uk>
Reply-To: Samiya Selim <bs09sas at leeds.ac.uk>
  Subject: Re: [Biomod-commits] (no subject)
       To: Wilfried Thuiller <wilfried.thuiller at ujf-grenoble.fr>
       Cc: biomod-commits at r-forge.wu-wien.ac.at

Hi Wilfried,

Thanks so much for the help. I did manage to fix the Projection
problem I was having(now I have a new projection problem)

And now I am also having problems with the actual
Ensemble Forecasting. This is the error I am getting:

Ensemble.Forecasting(Proj.name= "area", weight.method='Roc', PCA.median=T,
+ binary=T, bin.method='Roc', Test=F, decay=1.6, repetition.models=T)
Sp1
Error in rep(1, sum(wk != 0)) : invalid 'times' argument

Is it because GBM is included again when it says repetition.models=T?

I tried putting only the ones that are tru, but I got errors for that as well.

What does the line mean - Error in rep(1, sum(wk != 0)) : invalid
'times' argument?

Thanks again for any help.

Best wishes,
Samiya

Quoting Wilfried Thuiller <wilfried.thuiller at ujf-grenoble.fr>:

> Hi there,
>>
>>> Projection(Proj=area[,4:8],Proj.name='area',GLM=T,GAM=T,
>> + CTA=T,ANN=T,SRE=T,Perc025=T,Perc05=F,MDA=T,MARS=T,RF=T,
>> + BinRoc=T,BinKappa=T,BinTSS=T,FiltRoc=T,FiltKappa=T,FiltTSS=T)
>>
>>
>> The following models can not be used to render projections :  GBM
>>  they have not been trained in Models()
>
> Note that in the Projection function, all models are set to TRUE by   
>  default. Here GBM is not set up at FALSE whereas you have not run it.
> It just create a warnings but this is not related to the error you    
> have bellow.
>
>>
>> Sp1
>> Error in eval(expr, envir, enclos) : object 'Variable1' not found
>
> You must ensure you have the same variable names in 'area' than in    
> the data you use to calibrate the models.
>
> colnames(DataBIOMOD)
> colnames(area)
>
> The environmental variables must have the same labels.
>
> Let me know if the problem is not here.
>
> All the best
> Wilfried
>
>
>
>
>
>>
>> I have tried playing around with the data, using different data, still
>> keep getting the same.
>>
>> Any suggestions would be much appreciated!
>> Samiya
>> _______________________________________________
>> 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
>
> --------------------------
> Dr. Wilfried Thuiller
> Laboratoire d'Ecologie Alpine, UMR CNRS 5553
> Université Joseph Fourier
> BP53, 38041 Grenoble cedex 9, France
> tel: +33 (0)4 76 51 44 97
> 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
>
>
>
>
>
>


_______________________________________________
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


----- End forwarded message -----

_______________________________________________
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/20100325/d41c636a/attachment.htm