Add support for other interesting learning frameworks and libraries

[johann-petrak]

This is just to keep track of potentially interesting candidates and their advantages and disadvantages.

• https://www.h2o.ai/
• https://github.com/JohnLangford/vowpal_wabbit (!!!)
• FACTORIE http://factorie.cs.umass.edu/ , https://github.com/factorie/factorie
• http://dlib.net/ Pro: usable license, fast linear ranking SVM, Python wrapper. Written in C++
• http://www.mlpack.org/ written in C++
• https://github.com/JohnLangford/vowpal_wabbit/wiki
• https://github.com/johann-petrak/arow_csc

Also the ones at http://lamda.nju.edu.cn/Data.ashx#code especially https://github.com/kingfengji/gcForest and https://github.com/xzhren/DeepForest

[johann-petrak]

For topic modelling using mallet representation / in memory data, this library may be interesting: https://github.com/lejon/PartiallyCollapsedLDA