Re: [Gretl-devel] small issue with calling libsvm
On Fri, 1 Sep 2017, Riccardo (Jack) Lucchetti wrote:
On Fri, 1 Sep 2017, Artur T. wrote:
> Dear all,
>
> I am just experimenting with libsvm. [...]
I'm studying SVMs and I'll bergin experimenting as soon as I feel
a little bit more confident about what I'm doing.
Just one question so far: what is the rationale for having a svm()
function rather than a "svm" command?
That choice is debatable, I agree. However, it seemed to me that the
closest existing cousins to svm are loess and nadarwat, which are
both functions. Also, it seemed a little more convenient to have an
optional bundle-pointer argument for "extra" input/output than doing
something equivalent with command options or accessors.
Allin