This is a Beta version for the code of our ICML 2014's paper : Large margin metric learning for constrained partitioning problems.
The implementation there permits to run the change point detection metric learning. The MATLAB script testGlobal.m gives an instance of how this code is meant to be used, generating a toy example which can be used as a proof of concept. 
The optimization procedure implemented is a stochastic subgradient method. It is straightforward to implement fancier techniques.

A new version is expected to be released soon.

This code is distributed ``as is'' and without ANY warranty.

R. Lajugie
