Today's coding sprint was a bit more crowded, with some notable scipy hackers such as Ralph Gommers, Stefan van der Walt, David Cournapeau or Fernando Perez from Ipython joining in. On what got done: - We merged Jake's new BallTree code. This is a pure Cython implementation of a nearest-neighbor search similar to the KDTree class in scipy.spatial, but much faster. The code looks awesome and it's a big speedup compared to the older code. - Vlad is ready to merge hisdictionary learning code, something that should happen in the upcoming days. - Initial support for Python 3. scikit-learn should now at least build and import cleanly under Python 3. - some bugfixes in the Pipeline object and in docstrings. So this was the end of the scikit-learn sprint, but EuroScipy has just begun. See you tomorrow at the conference (follow the signs)!