Mailman 3 November 2022 - scikit-learn

[ANN] scikit-learn 0.24.0rc1 is online!
by Olivier Grisel Jan. 19, 2024

Jan. 19, 2024

Please help us test the first release candidate for scikit-learn 0.24.0: pip install scikit-learn==0.24.0rc1 Changelog: https://scikit-learn.org/0.24/whats_new/v0.24.html In particular, if you maintain a project with a dependency on scikit-learn, please let us know about any regression. Feel free to also retweet the announcement to get more people to test it before the final release (probably in 1 week or 2): https://twitter.com/scikit_learn/status/1334562221498753026 Thanks to anybody who helped make this happen! -- Olivier

5 12

[ANNOUNCEMENT] scikit-learn 1.0 release
by Adrin June 30, 2023

June 30, 2023

Hi everyone, We're happy to announce the 1.0 release which you can install via pip or conda: pip install -U scikit-learn or conda install -c conda-forge scikit-learn You can read the release highlights under https://scikit-learn.org/stable/auto_examples/release_highlights/plot_relea… and the long list of the changes under https://scikit-learn.org/stable/whats_new/v1.0.html New major features include: mandatory keyword arguments in many places, Spline Transformers, Quantile Regressor, Feature Names Support, a more flexible plotting API, Online One-Class SVM, and much more! This version supports Python versions 3.7 to 3.9. A big thanks to all contributors for making this release possible. Regards, Adrin, On the behalf of the scikit-learn maintainer team.

13 19

[ANN] scikit-learn 1.1.3 is online!
by Guillaume Lemaître March 9, 2023

March 9, 2023

scikit-learn 1.1.3 is out on pypi.org and conda-forge! This bugfix release only includes fixes for compatibility with the latest SciPy release >= 1.9.2 and wheels for Python 3.11. Note that support for 32-bit Python on Windows has been dropped in this release. This is due to the fact that SciPy 1.9.2 also dropped the support for that platform. Windows users are advised to install the 64-bit version of Python instead. More details in the changelog: https://scikit-learn.org/dev/whats_new/v1.1.html#version-1-1-3 You can upgrade with pip as usual: pip install -U scikit-learn The conda-forge builds will be available shortly, which you can then install using: conda install -c conda-forge scikit-learn Thanks again to all the contributors. On behalf of the scikit-learn maintainer team. -- Guillaume Lemaitre Scikit-learn @ Inria Foundation https://glemaitre.github.io/

3 3

Deprecation of the 'sklearn' PyPI package starts December 1st
by Loïc Estève Nov. 28, 2022

Nov. 28, 2022

Hi everyone, "pip install scikit-learn" has always been the supported way to install scikit-learn with pip, but for some time, "pip install sklearn" has been working too. The scikit-learn team has decided to gradually deprecate the 'sklearn' PyPI package by doing a brownout over one year: - starting December 1st 2022, using 'sklearn' rather than 'scikit-learn' for pip commands will start to raise an error over some predefined time windows - these time windows durations will gradually increase over one year - on December 1st 2023, using 'sklearn' rather than 'scikit-learn' for pip commands will always raise an error More information about the 'sklearn' package deprecation is available at https://github.com/scikit-learn/sklearn-pypi-package. Any concerns about this, let us know by replying to this email or by creating an issue at https://github.com/scikit-learn/sklearn-pypi-package/issues/new. Cheers, Loïc

1 1

VotingRegressor with pretrained estimators from CV as input
by Fernando Marcos Wittmann Nov. 18, 2022

Nov. 18, 2022

Hello, I'm dealing with a problem without much data. As a solution, I'm training 10 estimators using a 10-Fold CV-Schema. Now, I wanted to persist those models. In order to avoid having to save 10 estimators, I was thinking about saving a single VotingRegressor with those pre-trained models or maybe writing a custom estimator to be used only for inference. Has anyone dealt with similar problems? TLDR: Looking for an easy way to store multiple models created using cross-validation as a single model. More context: I'll be using mlflow sklearn autolog, so it will be waaay more convenient to log only a single model instead of multiple models in each experiment. Thanks!

2 1

scikit-learn monthly developer meeting: Monday November 28, 2022
by Thomas J. Fan Nov. 18, 2022

Nov. 18, 2022

Dear all, The scikit-learn developer monthly meeting will take place on Monday November 28 at 15:00 UTC. - Video call link: https://meet.google.com/gmn-acub-mrr - Meeting notes / agenda: https://hackmd.io/0yokz72CTZSny8y3Re648Q - Local times: https://www.timeanddate.com/worldclock/meetingdetails.html?year=2022&month=… The goal of this meeting is to discuss ongoing development topics for the project. Everybody is welcome. As usual, please follow the code of conduct of the project: https://github.com/scikit-learn/scikit-learn/blob/main/CODE_OF_CONDUCT.md Regards, Thomas

1 0

ANN: New member in the Contributor Experience Team
by Guillaume Lemaître Nov. 17, 2022

Nov. 17, 2022

We are excited to welcome a new member to the Contributor Experience Team: - Tim Head: https://github.com/betatim Looking forward to furthering interactions within the scikit-learn community. On the behalf of the scikit-learn team. -- Guillaume Lemaitre Scikit-learn @ Inria Foundation https://glemaitre.github.io/

1 0

Inquiry on evolutionary/genetic algorithms
by Ellarizza Fredeluces Nov. 8, 2022

Nov. 8, 2022

Dear Scikit-learn enthusiasts and developers, I am new in optimization and machine learning in general. I am having a hard time. So I have an objective function y = a*x**3 + b*x**2 + c*x + d I have values for x and y and I would want to determine a,b, c,d using GA. I am confused with number of inputs and number of population. Should i match the population size with my number of inputs (x,y)? For example, I have 100 sets of x, y. Should I also have 100 as population size? Thank you. Sincerely, Ella

1 0

Pandas DataFrame output is now available for all sklearn transformers
by Reshama Shaikh Nov. 2, 2022

Nov. 2, 2022

Hello, Pandas DataFrame output is now available for all sklearn transformers (in dev version 1.2)! This will make running pipelines on data frames much easier, and provides better ways to track feature names. There is a 14-minute video with examples, some more information and some FAQs answered at the end [a]. This is one of the biggest improvements in scikit-learn in a long time and we'd love your feedback! Please try out the nightly built and give it a go. We'd love to hear both about whether this helps your use cases and any bugs you find. A special thanks to the maintainers: Thomas J. Fan, Guillaume LeMaitre, Christian Lorentzen ! [a] video https://youtu.be/J4KCu9WWDTo [b] example https://scikit-learn.org/dev/auto_examples/miscellaneous/plot_set_output.ht… [c] LinkedIn post https://www.linkedin.com/feed/update/urn:li:activity:6987027021608460289/?a… --- Reshama Shaikh she/her

1 1

VOTE SLEP 17
by Andreas Mueller Nov. 1, 2022

Nov. 1, 2022

Hey Everybody! SLEP 17 <https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep017…> (by Joel Nothman) introduces an __sklearn_clone__ protocol & method that allows estimators to overload what sklearn's clone function does. An implementation for this SLEP is available in PR 24568 <https://github.com/scikit-learn/scikit-learn/pull/24568> (by Thomas Fan). Feedback to both the SLEP and implementation are welcome! If you are a scikit-learn core developer, please cast your vote in this PR <https://github.com/scikit-learn/enhancement_proposals/pull/79>. . According to our governance model <https://scikit-learn.org/stable/governance.html#decision-making-process>, the vote will be open for a month (till December 1st), and the motion is accepted if 2/3 of the votes cast are in favor. Cheers, Andreas

1 0