scikit-learn

With PyTorch Conference Europe kicking off today, it felt like a good moment to take stock of where open source tooling for data science and AI stands, and where it's headed next. To do just that, I sat down with Gael Varoquaux, our CSO at :probabl., and Matt White, the Global CTO of AI at The Linux Foundation, and picked their brains about the trends and priorities shaping their respective ecosystems. 🗣️ We covered a lot of ground, including how tools like scikit-learn are democratizing ML for millions of daily users; enterprise demand for openness wherever AI intersects with strategic control considerations; and the opportunities at the intersection of open source ML tooling and agentic AI. 🗞 Read or listen to my interview with Gael and Matt here: https://lnkd.in/dFTCSMpn

Onwards to GPU: Update on array API adoption in scikit-learn A blog post by our core developer Lucy L. on work to add open-ended compute engines, including GPU support, to scikit-learn https://lnkd.in/egB9tsTN Exciting times!

🏆 scikit-learn: A Core Component of 2025's Winning Machine Learning Solutions! 🏆 According to the newly released 2025 State of Machine Learning Competitions report by ML Contests, scikit-learn is a "Core" Python package in the toolkits of competition-winning data scientists, specifically highlighted for its enduring utility in models, transforms, and metrics🥇 While deep learning and massive compute budgets often grab the headlines, robust, reliable, and efficient machine learning fundamentals never go out of style. Read the full 2025 report here: https://lnkd.in/eeSUiq4E #MachineLearning #DataScience #Python #ScikitLearn #OpenSource #Kaggle #AI #MLContests

Do you still fit before you split? Put your score in comments. :probabl. is offering free certifications.

I’m thrilled to announce that I’m stepping up as :probabl.’s CSO (Chief Science Officer) to supercharge scikit-learn and its ecosystem, pursuing my dreams of tools that help go from data to impact. scikit-learn is central to data-scientists’ work. It has grown over more than a decade, supported by volunteers’ time, donations, and grant funding, with a central role of Inria. With :probabl.’s recent seed funding, we have a unique opportunity to accelerate scikit-learn’s development. Our analysis is that to build best on scikit-learn, enterprises need dedicated tooling and partners. Part of scikit-learn’s success has always been to nurture an ecosystem, for instance via its simple API that has become a standard. :probabl. is not only consolidating scikit-learn, but also this ecosystem: the skops project, to put scikit-learn based models in production, the skrub project, that facilitates data preparation. We have an amazing team at :probabl.. Many old-time scikit-learn contributors (Adrin J., Guillaume Lemaitre, Jérémie Du Boisberranger -not on Linkedin-, Loïc Estève, Olivier Grisel) are joined by new contributors (David Arturo Amor Quiroz, François Goupil, Stefanie Senger, and more to come). Working directly with businesses gives us an acute understanding of where the ecosystem can be improved, and I profoundly enjoy working with our top management, François MÉRO and Yann Lechelle, whose intimate understanding of business is very complementary to my background, and our broader :probabl. team, bringing a score of new skills and thinking. As CSO at :probabl., my role will be to nourish our development strategy with understanding of machine learning, data science and open source. Making sure that scikit-learn and its ecosystem are enterprise ready, will bring resources for scikit-learn’s sustainability, enabling its ecosystem to grow into a standard-setting platform for the industry.

Scikit-learn is on The Linux Foundation Open Source Insights board, alongside with many other central projects. What an honor to contribute to the open-source tissue that cements the world!

A scikit-learn blog article on our on-going improvements of the user experience (UX), in particular on making it easy for the user to understand scikit-learn models during development, especially while using tools like jupyter notebooks. https://lnkd.in/ecD2FScp

A new version of scikit-learn is out! And I'm proud that my PR #32100 is among the key highlights of this release 😁 (faster decision trees when using the absolute error). I'd like to thank the reviewers once again for their very thorough reviews! Thanks to them (and thanks to the tests too!), I'm confident that I delivered reliable and maintainable code. The PR: https://lnkd.in/eeEahkv9 The technical report: https://lnkd.in/eWXcSNFv You’ll find a detailed analysis of the different algorithms I considered to address the efficiency challenges of fitting decision trees with the absolute error.