I am the principal author of the following software:

MANET: MArkov NEtwork learning in Python

MANET: MArkow NEtwork learning in Python. The library allows learning of Markov Network classifiers with arbitrary complex graph of label interactions. It supports both supervised learning as well as learning from examples with missing labels.

Statistical Pattern Recognition toolbox for Matlab

STPRtool implements a selection of statistical pattern recognition methods described in the monograph M.I. Schlesinger, V. Hlavac: Ten lectures on the statistical and structural pattern recognition, Kluwer Academic Publishers, 2002

LIBQP

LIBQP is an open source library implementing Quadratic Programming solvers for tasks frequently appearing ML algorithms.

LIBOCAS

LIBOCAS is an open-source library implementing the Optimized Cutting Plane algorithm for training linear SVM classifiers. The library is written in C and it comes with interface to Matlab.

I have been also contributing to the following projects:

Landmark detection

CLandmark is an open-source library implementing real-time facial landmark detector based on Deformanble Part Models learned from example by Structured Output SVMs.

Shogun

Shogun machine learning toolbox provides a wide range of methods. It has been developed by a large entusiastic team of people lead by Soeren Sonnenburg.