Scikit-learn @ Inria Foundation

Fostering growth and sustainability of the reference machine-learning toolbox

with the help of our partners


Scikit-learn, a central tool in AI and data-science

Scikit-learn is a machine-learning library in Python. It is the engine that powers many applications of artifical intelligence and data science.

Scikit-learn is used on a regular basis by more than half a million of people in the world, with applications ranging from medical imaging to product recommendation.

More on scikit-learn


An open-source project

Scikit-learn is an open-source software (hosted on GitHub), under a license that facilitates commercial usage. It is developed by a world-wide community, gathering many different expertise on statistics, algorithms and software production.

The quality of scikit-learn, its algorithms, its interfaces, its documentation, are universally acclaimed. Its development follows a strict process to ensure this quality.


The foundation’s missions

The goal of the foundation is to enable maintaining scikit-learn’s high standards addressing new challenges.

The foundation employs central contributors to the project, to support scikit-learn’s community and to develop new ambitious features. The priorities of the foundation are set jointly by the community and its sponsors.




Technical Committee June 2, 2021 Agenda of the day 9.10 am - 10 am Presentation of the technical achievements and ongoing work by O. Grisel 10 am - 12 pm Feedback and exposition of each partner of the consortium 1 pm - 2 pm Collaborative drafting session of the updated roadmap 2 pm - 5 pm Afternoon discussions on Discord Attendees Scikit-learn @ Fondation Inria Alexandre Gramfort (Inria,…

How are the priorities of the consortium defined?

The scikit-learn Consortium @ Inria defines a roadmap every six to eight months during its Technical Committee. Previous roadmaps are available here. Why a roadmap? The members of the Consortium provide their financial support without any service counterpart. The definition of a development and general activities roadmap is an important step in building trust between the members of the Consortium. It represents our effort to focusing  together on…

Generalized Linear Models have landed in scikit-learn

While scikit-learn already had some Generalized Linear Models (GLM) implemented, e.g. LogisticRegression, other losses than mean squared error and log-loss were missing. As the world is almost (surely) never normally distributed, regression tasks might benefit a lot from the new PoissonRegressor, GammaRegressor and TweedieRegressor estimators: using those GLMs for positive, skewed data is much more appropriate than ordinary least squares and might lead to more adequate models. Starting…