On Wed, Mar 12, 2014 at 1:14 PM, Gael Varoquaux <
[email protected]> wrote:
> On Wed, Mar 12, 2014 at 08:30:10AM -0400, James Bergstra wrote:
> > I'd be thrilled if key sklearn people would be open a GSoC project
> looking at
> > using e.g. hyperopt to do model selection in a way that's smart for using
> > various cross-validation schemes and regularization paths (where
> available).
>
> Scikit-learn will not add a dependency, even optional. It's contrary to
> the vision of the project being a foundation project giving a base layer.
>
> What I would find interesting it getting a super simplified version of
> the basic hyperopt functionality in scikit-learn :). You know, the 80/20
> tradeoff.
>
>
I appreciate the 80/20 tradeoff in general, but in this case I'm not sure
what more could usefully be added to sklearn's model selection logic and
fit with the project vision.
Maybe you could add these things to another project (e.g. hyperopt,
hyperopt-sklearn) if you want to work on them.
------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
