Smile – Statistical Machine Intelligence and Learning Engine

haifengl.github.io

78 points·based2··7 comments

This page has a lot of thinly veiled blackhat SEO. I suppose that doesn't say anything about the project's technical merits, but it really puts me off. There are three separate paragraphs that consist of lists of technical terms and nothing else.

> Smile is the most exciting project on Github today!

> Smile covers every aspect of machine learning.

> Smile provides hundreds advanced algorithm with clean interface.

Lots of exclamation points and kind of vague yet sloppy writing... if I didn't know any better I'd say it's bordering on snake oil.

This is supposed to be faster than XGBoost? I'm skeptical, but I'd like to know the specifics of the benchmarks and maybe an outline of the code / reasons why. It was not benchmarked by the same person who did https://github.com/szilard/benchm-ml

Vowpal Wabbit is IO limited. Meaning that there's no way it is slower than anything else on a single machine. On multiple machines it glides faster than light.

So, the benchmark is probably incorrect for VW.

http://xyclade.github.io/MachineLearning/

Looks like a mammoth project given the amount of contributors. However, what is more missing in the face of h2o, dl4j and mllib would be a practical scientific computing / compute graph based library for Scala.

Looks very interesting to me. Obviously an enormous effort on behalf of the dev. I'm looking forward to trying it.

Reply on news.ycombinator.com