This code implements the GPstruct learning algorithm:
- in the chain case, as presented in our TPAMI paper, to appear (currently an arXiv preprint) by S Bratieres, N Quadrianto, Z Ghahramani
- in the grid case, as presented in our ICML 2014 paper by S Bratieres, N Quadrianto, S Nowozin, Z Ghaharamani
An Ipython notebook showing how the source code works can be found here on nbviewer.ipython.org.
Another notebook demonstrates the NLP experiments reported in our TPAMI paper.
Note that this code is research code: it will evolve and backwards compatibility is not expected.
Feel free to contact the authors:
- Sébastien Bratières: sb358 and then the at sign, then cam and also ac.uk
- Novi Quadrianto: letter n followed by dot then quadrianto and the at, and also sussex.ac.uk