Re: [Scikit-learn-general] [ANN] Scikit-learn 0.17 released

Hurray! Thank you Andy for shepperding this release! Gaël On Thu, Nov 05, 2015 at 07:45:18PM -0500, Andreas Mueller wrote: > Hey everybody. > I'm happy to announce the release of scikit-learn 0.17. > A big thank you to everybody who contributed. > Highlights of the release include: > - Latent

[Scikit-learn-general] [ANN] Scikit-learn 0.17 released

Hey everybody. I'm happy to announce the release of scikit-learn 0.17. A big thank you to everybody who contributed. Highlights of the release include: - Latent Dirichlet Allocation via Online Variational Inference - A Stochastic Average Gradient solver for LogisticRegression and RidgeRegressio

Re: [Scikit-learn-general] Comparisons of classifiers

Comparing VW and liblinear seems pretty meaningless (and calling liblinear Python is also odd). It's clear that there are faster gbm packages (and it seems random forests for some settings of the parameters) We recently had some improvements to the trees and it would be interesting to benchmark

Re: [Scikit-learn-general] Comparisons of classifiers

On Thu, Nov 05, 2015 at 07:05:11AM +, Raphael C wrote: > https://github.com/szilard/benchm-ml > The upshot is that in some cases it seems that the scikit-learn > versions have room for improvement. The various main lessons that I can see from those results are: * Linear models (aka LogisticR