[R] Grasper model error
Simon Wood
s.wood at bath.ac.uk
Wed Aug 16 11:07:38 CEST 2006
I'm not too familiar with using grasp, but the error is generated from the
`mgcv' package while setting up a GAM. The default smooths used for `s' terms
by mgcv::gam are thin plate regression splines based on an `eigen
approximation' to full thin plate splines. These have some nice properties
but become expensive to set up for data sets with more than a few thousand
points, in which case it is better to use a different basis for smooths of
one variable, e.g.
s(x,bs="cr")
or for smooths of several variables use either tensor product smooths, e.g.
te(x,z)
or use the `knots' argument to `gam' in the way shown at the end of the
examples in the `gam' help file.
Simon
On Tuesday 15 August 2006 19:29, Ken Nussear wrote:
> I tried this over a the grasp users yahoo group and got no
> response....So I wonder if anyone here knows about grasper
>
> I keep getting this error when trying to run a model.
>
>
> Error in smooth.construct.tp.smooth.spec(object, data, knots) :
> Too many knots for t.p.r.s term: see `gam.control' to increase limit,
> or use a
> different
> basis, or see large data set help for `gam'.
>
>
> I'm using R for OSX Gui Version 1.16 R-version 2.3.1, running grasper
> version
> 0.4-4
>
>
> My dataset has ~7500 rows of input, and 21 cols of input.
>
> Does anyone have an idea how to troubleshoot this?
>
> Thanks
>
>
> Ken
>
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html and provide commented, minimal,
> self-contained, reproducible code.
--
> Simon Wood, Mathematical Sciences, University of Bath, Bath, BA2 7AY UK
> +44 1225 386603 www.maths.bath.ac.uk/~sw283
More information about the R-help
mailing list