this looks pretty awesome - especially the dataset abstraction is pretty
neat - would be great if we could merge this into scikit-learn.

btw: what kind of truncated gradient algorithm does lightning use for L1
penalized SGD? As far as I can see its not the one that's currently used in
SGDClassifier...

best,
 Peter


2013/5/2 Andreas Mueller <amuel...@ais.uni-bonn.de>

> On 05/01/2013 03:32 PM, Mathieu Blondel wrote:
> > My library for large-scale linear classification / regression,
> > lightning, already supports multiclass logistic regression (by SGD and
> > by coordinate descent). It also includes many other goodies, including
> > my upcoming paper on sparse multiclass classification. For all my
> > Cython code, I used a dataset abstraction which allows to support both
> > dense and sparse data in a single implementation (and without memory
> > copy). I will blog about lightning later when my paper is up.
> >
> > https://github.com/mblondel/lightning
> Cool.
> How does the performance differ from the sklearn implementation?
> And when will we see the PR ;)
>
> Cheers,
> Andy
>
>
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Scikit-learn-general mailing list
> Scikit-learn-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>



-- 
Peter Prettenhofer
------------------------------------------------------------------------------
Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
Get 100% visibility into your production application - at no cost.
Code-level diagnostics for performance bottlenecks with <2% overhead
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap1
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general

Reply via email to