Software and data

Scikit-SpLearn

Scikit-SpLearn is a python toolbox implementing spectral learning algorithms for weighted automata. The toolbox is compliant with the well-known scikit-learn machine learning platform. In particular, all data tools of sk-learn, like cross-validation and gridsearch, can be used with scikit-splearn.

This was done with the help of the Archimède Labex software development team in a professional setting (unit testing, subversionning with git, PEP8 code, user manual, etc.)

PAutomaC Benchmark

PAutomaC is an on-line competition we organized in 2012 on learning probability distribution from sequential data. It provides 50 data sets that have been extensively used as benchmark since the end of the competition.

SPiCe Benchmark

SPiCe is an on-line competition we organized in 2016 on guessing the next symbol in a sequence. It provides 15 data sets from various fileds (biology, NLP, software, etc.)