News for hackers. Stuff that matters.
Re:Why use SVM by t3kcit on Wednesday December 26, 2012 @08:52AM (#4969196)
I'm a bit disappointed that such an unreflective post makes it to hacker news. Many of the scikit-learn staff have excellent blogs, I never saw any of those featured.

Btw, the sklearn docs have an example that I think explains different classifiers way better and also hints at a very important point: non-linear decision boundaries are mostly important for low-dimensional dense data and not helpful for text data for example.

Re:Why use SVM by riffraff on Wednesday December 26, 2012 @09:57AM (#4969381)
> Many of the scikit-learn staff have excellent blogs, I never saw any of those featured.

maybe you should submit more of their posts :)

Re:Why use SVM by achompas on Wednesday December 26, 2012 @01:13PM (#4970194)
I've seen them on here. They're excellent technical posts that probably don't interest many of the main-page viewers here.

Here's Jake Vanderplas on SVD algorithms for sparse matrices:

http://jakevdp.github.com/blog/2012/12/19/sparse-svds-in-pyt...

Excellent content.

Re:Why use SVM by t3kcit on Wednesday December 26, 2012 @08:52AM (#4969197)