[R] ffnet problem
v.demartino2 at virgilio.it
Wed Jun 2 17:51:39 CEST 2004
On Tuesday 01 June 2004 12:24, Adrian Trapletti wrote:
> >Given that those files are dated 27/06/2000, they are unlikely to work
> > with the current version of R. (R has changed quite a bit since 2000!!)
> In fact, I stopped to support ffnet in 2000, and I would be very
> surprised if it would still run under the newer versions of R.
> >Don't know why/how the code depends on NR codes, as one can find better
> >quality code in many instances. It might not be too difficult to get rid
> > of the dependency on NR codes.
> The underlying c++ code of ffnet has been programmed around 1996 at a
> time when the capabilities of R to do multidimensional optimization
> through an API where rather limited. Hence my decision at that time to
> base the code on Numerical Recipes. But from today's perspective, it
> would be much better to call the optimizers from optim() through the API.
> >BTW, why not use nnet in the VR bundle? (Obviously I have not looked at
> >ffnet at all...)
> I also suggest you use nnet.
Thanks to the many who answered clarifying the point.
I was just having a go at nnet when I knew about ffnet and wanted only to have
an alternative, one more option. You know open source and free sofware
usually offers this opportunity!
More information about the R-help