Nieme is a machine learning library for large-scale classification, regression and ranking. It relies on the framework of energy-based models which unifies several learning algorithms. This framework also unifies batch and stochastic learning which are both seen as energy minimization problems. Nieme is released under the GPL license. It is efficiently implemented in C++ and works on Linux, MacOS and Windows. Interfaces are available for C++, Java and Python. (french/français)

Energy Based Machines	Nieme Explorer
A unifying framework for learning machines	A user interface for visualizing and debugging
Composite Vectors	Multiple Language Bindings
Structure into the vectors	C++, Java, Python and more
Infinite Attribute Space	Introspection
New learning features at any time	Fast implementation with introspective features

This project has been funded with support from the French National Research Agency ANR-05-101 - S0604153 W.