scikit-learn

?? PyData Paris 2024 talk ?? An Update on the Latest scikit-learn Features ??Speakers: scikit-learn maintainers Stefanie Senger, PhD & Guillaume Lemaitre ??30-minutes ?? In this talk, we provide an update on the latest `scikit-learn` features that have been implemented in versions 1.4 and 1.5. We will particularly discuss the following features: ?? the metadata routing API allowing to pass metadata around estimators; ??the `TunedThresholdClassifierCV` allowing to tuned operational decision through custom metric; ??better support for categorical features and missing values; ??interoperability of array and dataframe. #python #machinelearning #datascience #opensource PyData NumFOCUS https://lnkd.in/ecC-2YYQ

?? PyData Global 2023 talk ?? Get the Best from Your scikit-learn Classifier ??Speaker: scikit-learn maintainer Guillaume Lemaitre ??30-minutes ?? When operating a classifier in a production setting (i.e. predictive phase), practitioners are interested in potentially two different outputs: a "hard" decision used to leverage a business decision or/and a "soft" decision to get a confidence score linked to each potential decision (e.g. usually related to class probabilities). Scikit-learn does not provide any flexibility to go from "soft" to "hard" predictions: it uses a cut-off point at a confidence score of 0.5 (or 0 when using decision_function) to get class labels. However, optimizing a classifier to get a confidence score close to the true probabilities (i.e. a calibrated classifier) does not guarantee to obtain accurate "hard" predictions using this heuristic. Reversely, training a classifier for an optimum "hard" prediction accuracy (with the cut-off constraint at 0.5) does not guarantee obtaining a calibrated classifier. In this talk, we will present a new scikit-learn meta-estimator allowing us to get the best of the two worlds: a calibrated classifier providing optimum "hard" predictions. This meta-estimator will land in a future version of scikit-learn. We will provide some insights regarding the way to obtain accurate probabilities and predictions and also illustrate how to use in practice this model on different use cases: cost-sensitive problems and imbalanced classification problems. #python #machinelearning #datascience #opensource PyData NumFOCUS https://lnkd.in/ewHFfDNy

?? PyData Global 2023 Tutorial ??Speaker: scikit-learn maintainer Olivier Grisel ?? Predictive survival analysis with scikit-learn, scikit-survival and lifelines ??This 90-minute tutorial will introduce how to train machine learning models for time-to-event prediction tasks (health care, predictive maintenance, marketing, insurance...) without introducing a bias from censored training (and evaluation) data. #python #machinelearning #datascience #opensource https://lnkd.in/dbh4FvTK

????Pushing Cython to its limits in scikit-learn Check out the PyData NYC 2024 presentation by maintainer Thomas J. Fan scikit-learn is a machine-learning library for Python that uses NumPy and SciPy for numerical operations. Scikit-learn has its own compiled code for performance-critical computation written in C, C++, and Cython. The library primarily focuses on Cython for compiled code because it is easy to use and approachable. In this talk, we dive into many techniques scikit-learn employs to utilize Cython fully. We will cover features like using the C++ standard library within Cython, fused types, code generation with the Tempita engine, and OpenMP for parallelization. #opensource #python #machinelearning #datascience https://lnkd.in/e27AhKUz

?? Calling all professional scikit-learn users! We’re developing professional support services and need your input. Imagine a central helpdesk to manage and discuss your challenges (confidentially!), direct input from core contributors, rapid technical guidance and best practices, early access to new features, and long-term support options—all in one place. Share your thoughts in our quick survey?to help shape these services for our entire community. ?? Complete the survey here ?? https://lnkd.in/gx6_Macc ???Plus, one lucky respondent will score our :probabl. #backpack, jam-packed with goodies—stickers, a sweatshirt, water bottle, candy, and more! ?? How to increase your chances for the giveaway: - Like this post?and?tag someone?who might also benefit from these services. - Each tag = an extra entry, so spread the word! #DataScience #MachineLearning #ScikitLearn #OpenSource

??scikit-learn is now on Bluesky! …. and our account is verified ??? Follow us here for updates on the #??1??, python, open source, machine learning library. https://lnkd.in/eR_j5CZT

scikit-learn is not only a library, it is an entire living data science ecosystem! There is a newbie in town — say hello to?skore!? From hyperparameter optimization to preprocessing tools and more, there are plenty of projects that work seamlessly alongside scikit-learn. Curious? Check out our Related Projects page (https://lnkd.in/evdj6rNR) and discover how to enhance your workflows! What is your favorite scikit-learn compatible project?

?? The Christmas release is out ?? ?? Release video below Discover scikit-learn 1.6 and its: ?? 2 major features & 34 features ?? 5 efficiency improvements & 21 enhancements ?? 14 API changes ?? 30 fixes ?? 160 contributors (thank you all!) More details in the changelog: https://lnkd.in/e5pui3ev You can upgrade with pip as usual: pip install -U scikit-learn Using conda-forge builds: conda install -c conda-forge scikit-learn Thanks to ?? Vincent D. Warmerdam for this release highlights video. https://lnkd.in/eRM_5rte #scikitlearn #Python #release #sklearn #software #ML #machinelearning #datavisualization #dataanalytics #data #dataanalysis #deeplearning #opensource #opensourcesoftware #opensourcecommunity

Talks from PyData Paris are now available! I’m excited to share our "Update on the Latest scikit-learn Features," presented alongside Guillaume Lemaitre. I cover advancements in the metadata routing API, while Guillaume introduces the new TunedThresholdClassifierCV and other quality improvements. Video link in the comments.

PyData Paris 2024 45-minute Keynote by Olivier Grisel: ?? Handling predictive uncertainty in machine learning Machine Learning practitioners build predictive models from "noisy" data resulting in uncertain predictions. But what does "noise" mean in a machine learning context? #python #machinelearning #datascience #opensource https://lnkd.in/ez2hn8dp