[Biomod-commits] contribution of each algorithm to the ensemble

Frederico Mestre mestre.frederico at gmail.com
Fri Dec 16 17:12:17 CET 2011


Hi:

 

I'm trying to understand which is the contribution of each of the algorithms
to the final ensemble. I've selected them by using the qual.th, and weighted
them using weight.method and decay in a "global ensemble" (with all the
algorithms).

 

Is there any way to get a table with the contribution of each of the
algorithms?

 

I've been trying to do this by projecting the individual algorithms, in
order to get an ensemble of each one (Presence/absence and evaluation runs).
But using the same threshold in the qual.th in this "partial" ensembles as I
use in the "global" (with all the algorithms) I always get the message:

 

Error in if (sum(ProbData) != 0) { : 

  missing value where TRUE/FALSE needed  

 

Which I thing has something to do with none of the models in the projection
being above the specified threshold. 

I don't understand how is it possible that in a "global" ensemble, with all
the algorithms, I'm able to get a final result (which means that, at least
some of the algorithms produced models above the quality threshold
specified) and then in the "partial" ensembles, of each algorithms, I get
this error.

 

I don't know if what I mean is clear, Am I missing anything? What's the
problem?

 

Thanks, 

 

Frederico Mestre

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/biomod-commits/attachments/20111216/4a053fc6/attachment.htm>


More information about the Biomod-commits mailing list