[prev in list] [next in list] [prev in thread] [next in thread] 

List:       r-sig-geo
Subject:    Re: [R-sig-Geo] [R-sig-eco] Error in run GAM (Biomod 2)
From:       Lara Silva <lara.sfp.silva () gmail ! com>
Date:       2019-03-18 13:04:22
Message-ID: CALN9TEQ7GdMWXCcRSYmsj0ZEYZpcnzpauMTC=uZ=f0WBKdv8vg () mail ! gmail ! com
[Download RAW message or body]

Thanks for the information.

Drew Tyre <atyre2@unl.edu> escreveu no dia segunda, 18/03/2019 Ã (s) 11:48:

> I don't think the number of presences is the problem.
> "A term has fewer unique covariate combinations than specified maximum
> degrees of freedom"
> One of your covariates has a small number of unique values.
> lapply(your_data_frame, function(x)length(unique(x)))
> will print out the number of unique values in each variable. Values < 30
> could be problematic -- the docs are a bit ambiguous on the default values
> of k, the maximum degrees of freedom for a smooth term. Another approach to
> diagnose the problem is to set k to a low value, like 5, and see if that
> makes the problem go away. I'm not familiar with biomod2, but I think doing
> something like
> MO <- BIOMOD_ModelingOptions(GAM  = list(k = 5))
> BIOMOD_Modeling( ..., models.options = MO, ...)
> 
> should do it. Note that setting k to a low value causes other problems, I
> would use mcgv::gam.check() to ensure k is large enough, but I don't know
> how to do that with biomod2.
> 
> hth
> 
> --
> Drew Tyre
> 
> School of Natural Resources
> University of Nebraska-Lincoln
> 416 Hardin Hall, East Campus
> 3310 Holdrege Street
> Lincoln, NE 68583-0974
> 
> phone: +1 402 472 4054
> fax: +1 402 472 2946
> email: atyre2@unl.edu
> http://snr.unl.edu/tyre
> http://drewtyre.rbind.io
> The point is that our true nature is not some ideal that we have to
> live up to. It's who we are right now, and that's what we can make friends
> with and celebrate.
> Excerpted from: Awakening Loving-Kindness by Pema Chödrön
> 
> 
> On 3/18/19, 6:28 AM, "R-sig-ecology on behalf of Lara Silva" <
> r-sig-ecology-bounces@r-project.org on behalf of lara.sfp.silva@gmail.com>
> wrote:
> 
> Hello
> 
> I am trying to run several algorithms in biomod 2 (GLM, GAM, ANN, SRE)
> but
> I received the following menssage.
> 
> Model=GAM
> GAM_mgcv algorithm chosen
> 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 freedom
> Error in predict(model.bm, Data[, expl_var_names, drop = FALSE],
> on_0_1000
> = TRUE) :
> object 'model.bm' not found
> 
> *** inherits(g.pred,'try-error')
> ! Note :  Lactuca.global_AllData_RUN1_GAM failed!
> 
> I have 19 presences and 1019 pseudo-absences ...
> It is because the low number of presences?
> 
> Thanks
> 
> Lara
> 
> [[alternative HTML version deleted]]
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology@r-project.org
> 
> https://urldefense.proofpoint.com/v2/url?u=https-3A__stat.ethz.ch_mailman_listinfo_r \
> -2Dsig-2Decology&d=DwICAg&c=Cu5g146wZdoqVuKpTNsYHeFX_rg6kWhlkLF8Eft-wwo&r=aLEaSryyUc \
> ERVqcKVZl7lQ&m=2aNX9-kGSZ1EivodV2ZAVeJGY76pX4TqGpx2QXqsx00&s=-syyYDDESQAcvQx2Khl1KE_jbj3yBD2UqgKcFE4mBpw&e=
>  
> 
> 

	[[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo


[prev in list] [next in list] [prev in thread] [next in thread]