[Biomod-commits] Error in BRT and GAM

[Damien Georges]

Dear Joaquin,

I guess this errors should come from 2 things :
     - You have very few points in your data set and models do not 
manage to converge
     - One (or more) of the explanatory variables you work with is a 
factor but is not declared properly.

Are you in one of this 2 case?

Best,
Damien

On 18/04/2013 12:23, Joaquin DL wrote:
> Dear all,
>
> I am having this error running BRT and GAM.
> How can I solve them?
>
> Thanks.
>
> Joaquín
>
> Model=Generalised Boosting Regression
> 	 500 maximum different trees and  5  Fold Cross-ValidationError in
> gbm.fit(x[i.train, , drop = FALSE][i, , drop = FALSE], y[i.train][i],
> :
>    The dataset size is too small or subsampling rate is too large:
> nTrain*bag.fraction <= n.minobsinnodeIn addition: Warning
> message:running command 'java' had status 1 Error in predict(model.bm,
> Data[, expl_var_names, drop = FALSE], on_0_1000 = TRUE) :
>    error in evaluating the argument 'object' in selecting a method for
> function 'predict': Error: object 'model.bm' not found
>
>     ! Note :  Pt.Test_AllData_RUN1_GBM failed!
>
> Model=GAM
> 	 GAM_mgcv algorithm chosen
> 	User defined control args building..
> 	Automatic formula generation...
> 	> GAM (mgcv) modelling...Error in
> smooth.construct.tp.smooth.spec(object, dk$data, dk$knots) :
>    A term has fewer unique covariate combinations than specified
> maximum degrees of freedomError in predict(model.bm, Data[,
> expl_var_names, drop = FALSE], on_0_1000 = TRUE) :
>    error in evaluating the argument 'object' in selecting a method for
> function 'predict': Error: object 'model.bm' not found
> _______________________________________________
> 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