> Many of the scikit-learn staff have excellent blogs, I never saw any of those featured.
maybe you should submit more of their posts :)
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.
forgot the link: http://scikit-learn.org/dev/auto_examples/plot_classifier_co...
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.