
Find scikit-learn here, on our most active platforms: • LinkedIn: linkedin.com/company/7986535… • Bluesky: bsky.app/profile/scikit-lear… • Mastodon: fosstodon.org/@sklearn

Easy-to-use and general-purpose machine learning in Python

Find scikit-learn here, on our most active platforms: • LinkedIn: linkedin.com/company/7986535… • Bluesky: bsky.app/profile/scikit-lear… • Mastodon: fosstodon.org/@sklearn

You can upgrade with pip as usual: pip install -U scikit-learn Using the conda-forge builds: conda install -c conda-forge scikit-learn


❄️ The Christmas release is out ❄️ Discover scikit-learn 1.6 and its: 🟢 2 major features & 34 features 🔵 5 efficiency improvements & 21 enhancements 🟡 14 API changes 🔴 30 fixes 👥 160 contributors youtu.be/7wiHChpwJe8

scikit-learn 1.5.2 is out! It fixes several regressions introduced in version 1.5 More details in the changelog: scikit-learn.org/dev/whats_n… You can upgrade as usual: pip install -U scikit-learn conda install -c conda-forge scikit-learn Thanks to all the contributors!

📝 A survey results report and anonymized dataset will be publicly released by the end of 2024.

🌎 We have the survey available in these languages: Arabic, English, French, Japanese, Mandarin, Portuguese, Spanish. ⏱️ The survey will take about 15 minutes of your time and close on September 27th, 2024.

The 2024 scikit-learn user survey is out! Please join this structured dialogue with the scikit-learn team to better guide and prioritize decision-making about the development of the project: docs.google.com/forms/d/e/1F…

Scikit-learn has been awarded a CZI-EOSS grant for cycle 6! 🎉 With this funding, we'll improve tools for predictive model evaluation and inspection. Read all about it here: blog.scikit-learn.org/fundin… 📝 A big thank you to @chanzuckerberg and @wellcometrust! #OpenScience

Thank you very much to the 136 people who contributed to this release!

Contributors can find the updated instructions on how to build scikit-learn from source if interested: scikit-learn.org/dev/develop…

We switched the default build tool from setuptools to meson: it's no longer needed to manually rebuild when changing source files: the editable mode of meson-python detects changes automatically and rebuilds what's necessary when importing sklearn.

Along with the new release, we updated the website look and feel to improve navigation in the doc based on the pydata sphinx theme (with dark mode)! It comes with near-instant filtering of public API by keyword on this page: scikit-learn.org/stable/api/… Thanks Charlie XIAO!

You can quickly walk through this new feature and other release highlights here: scikit-learn.org/stable/auto… and the full changelog at: scikit-learn.org/stable/what…

This is especially powerful to implement cost-sensitive learning, in particular when used in conjunction of per-individual side-metadata used in the cost function, thanks to our meta-data routing infrastructure.

Among the new features, this new release introduces the TunedThresholdClassifierCV class to adjust the decision threshold of any binary classifier to assign custom costs or gains to true/false positives/negatives.