[Biomod-commits] Re : pseudoabsences

[Bruno Lafourcade]

Hi again,

Mind that looping the pseudo-abs function will overwright the output if you are
not carefull. You need to change the name of the output during the loop, or
store it in a different object before the next step of the loop to keep track of it.

And yes you're right about the output, it enables to create a new dataframe with 
the ones and the selected zeros.

But I hope you realise that the Models() function has the pseudo-abs function
already inserted in it and deals with the storing issues. You have to use the
NbRepPA argument and the other arguments associated with it.

Doing a prior run and loading the database created in Initial.State() or setting  
NbRepPA to 1 will end up to the same result, unless you wish to run a different
pseudo-abs procedure for each species. To this regard, you do need to do a
prior run.

Tell me how it goes with the storing issues. Do you need precisions ?

 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 : Aaron Bueno-Cabrera <aaronbueno at hotmail.com>
A : Bruno Lafourcade <brunolafourcade at aol.com>
Envoyé le : Jeudi, 9 Septembre 2010 19:20
Sujet : RE: Re : [Biomod-commits] pseudoabsences


Hi Bruno,
no worries, I understand the delays and thanks for the answer;

I am gonna look carefully the right names the next time;

Regarding the pseudo.abs, basically you are saying that, if I have 50 species, then do I have to run 50 times the pseudo.abs function 
(or a loop for the same function) and then create just one large database containing all the sp' presences and pseudoabsences?; 
I want to do this procedure in order to run the models() function.

What I understood from the manual, is that after using this pseudo.abs function over my database (ceros and ones), 
I will have a new database containing ones(original presences) and new ceros (pesudoabs). Am i right?

Thanks for your support.
Regards
Aaron





To: aaronbueno at hotmail.com
Subject: Re : [Biomod-commits] pseudoabsences
Date: Tue, 7 Sep 2010 06:22:40 -0400
From: brunolafourcade at aol.com


 

 Hi Aaron,

Apologies for the very late answer.
Maybe you are not looking for the right name.
Look at what you have before and after using the function with ls().

Mind that it might overwright an existing object, so there will be no difference.
Also try to use the species.name argument. The default value is species.name= 'SpNoName'

About running the pseudo.abs function on several species at a time, you can only loop the job.
The function does only consider one species at a time.

Hopîng to solve your troubles.
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 : Aaron Bueno-Cabrera <aaronbueno at hotmail.com>
A : biomod-commits at lists.r-forge.r-project.org
Envoyé le : Jeudi, 19 Août 2010 21:00
Sujet : [Biomod-commits] pseudoabsences


Hello list members,

I am trying to run Biomod with several species, and predictors;
I don't know how to get the pseudoabsences for several species at the same time;
In fact, when I try to run the pseudo.abs command with just one species, I can't get the species.name object as the Biomod manual says (Page 75).  
Please anyone can send me some thoughts or suggestions to properly run the pseudo.abs function with several species at the sames time?
I want to use this objects in order to run the Models function.
Thanks a lot
Aaron
                       =
 
_______________________________________________

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/20100909/6199c607/attachment.htm>