Where communities thrive

  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
Repo info
  • 03:42
    PSSF23 commented #18888
  • 02:52
    chritter synchronize #21726
  • 02:48
    chritter commented #22148
  • 02:48
    chritter commented #22148
  • 02:46
    chritter synchronize #22148
  • 02:35
    scikit-learn-bot edited #22142
  • Jan 17 23:17
    cmarmo commented #22145
  • Jan 17 22:52
    thomasjpfan synchronize #22237
  • Jan 17 22:26
    cmarmo commented #22145
  • Jan 17 22:00
    thomasjpfan synchronize #22237
  • Jan 17 21:24
    SparklePigBang commented #21350
  • Jan 17 21:03
    JimQ312 commented #11647
  • Jan 17 20:15
    fangzhouli commented #22230
  • Jan 17 20:12
    thomasjpfan closed #11026
  • Jan 17 20:12
    thomasjpfan commented #11026
  • Jan 17 20:01
    thomasjpfan closed #7253
  • Jan 17 20:01
    thomasjpfan commented #7253
  • Jan 17 19:41
    thomasjpfan synchronize #22237
  • Jan 17 19:40
    github-actions[bot] labeled #22237
  • Jan 17 19:40
    thomasjpfan opened #22237
or is it only for classification ?
Guillaume Lemaitre
confusion matrix and derived metrics are only for classification
look at the regression metrics instead
Markus Löning
Hi all, how do you manage your changelog? Any useful tools or files that you could point me too?
Nicolas Hug
@mloning most PRs come with a changelog entry, e.g. https://github.com/scikit-learn/scikit-learn/pull/20727/files
Then we have scripts to process / check it which are described in https://scikit-learn.org/stable/developers/maintainer.html (we call the change log a "What's new")
Markus Löning
@NicolasHug thanks - that's very helpful!
Harsh Chauhan
A very good evening everyone!!
I am trying to setup scikit-learn project into my development env but encountered with an error when I try to run "pip install --verbose --no-build-isolation --editable ."
Can someone help me to fix this error?
Adrin Jalali
@RAVANv2 can only help if you tell us what the issue is :)
Harsh Chauhan
Hi @adrinjalali,
Actually, I solve it on my own :)
Harsh Chauhan
Hi all, I made one PR on issue #20754 but it failing the linting job of black and showing the error "##[error]Bash exited with code '1'". Can anyone tell me where I am doing wrong? I am totally a newbie in open-source but I m really enjoying :)
Adrin Jalali
You can see how to fix that here: https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist You need to run black on your code
Kiran Hipparagi
Hello Everyone, I am a newbie to open source with basic knowledge about scikit-learn and want to contribute to this repository. Can anyone tell me how should I start? I have basic knowledge about generating my first PR and solving good first issues. Now I want to fix bugs and do some code contributions but I am unable to understand the issues. Can anyone guide me?
Andreas Mueller
Hi @KiranHipparagi , thanks for wanting to contribute! Have you read the contributor's guide? I would suggest you look for issues tagged "good first issue". Many of those are multi-part issues where you can pick just a part to work on
this one might be good to get started: scikit-learn/scikit-learn#20308
Hello i am starting out this one project and some of the parts are really complicated . And i am happy to learn during the complicated parts. But i feel i need more people. If anyone is interested in working together and learning together feel free to dm me
my discord is Aura#5549
Reshama Shaikh

Deleted this message from the /dev/ channel. Copying and pasting here:

I am Bhavya Bhardwaj (https://github.com/Bhavya1705). I am a student of Electronics and Communication at Amrita Vishwa Vidyapeetham, India. My thanks to you and the team for sklearn. I have been try to make some contributions to the scikit-learn library - scikit-learn/scikit-learn#5516. I have made the code, and the necessary changes to the init file and test files, in addition to the _classification file. This is the links to my commits - scikit-learn/scikit-learn#20861, as you will see, there are many mistakes, that I have made, Any help that you can render to me would be much appreciated and would be a wonderful learning experience.
Thank You

Bhavya Bhardwaj
@reshamas Thank You, I have managed to solve the issue.

Hi. I am trying to develop my own Estimator based on TransformerMixin and BaseEstimator. To make sure I am doing things right I have added a test to my project :

import MyEstimator
from sklearn.utils.estimator_checks import check_estimator
def test () : 
    me = MyEstimator(**params)

If I run the test, I get the following error message :

AssertionError: The error message should contain one of the following patterns:
               0 feature\(s\) \(shape=\(\d*, 0\)\) while a minimum of \d* is required.

I don't understand how I am supposed to take care of that. I am even more surprise because my fit_transorm method uses self._validate_data at the beginning. I would expect that function to take care of case like these. Could someone help me with that issue ?

Roman Yurchak
@adriente: Could you please open a Github issue with full traceback and tag me (@rth) in?
Freddy Boulton
Hello! I opened this feature request a week ago. Just bumping it here in case it got lost: scikit-learn/scikit-learn#20890
Guillaume Lemaitre
@freddyaboulton I can assure you it is not lost :) I saw it but I did not look at it yet because we are kind of working on releasing 1.0. Once the release done, you might get some attention from core-devs
1 reply
Loki The Great
Morning all
I was kind of curious . I’ve been dragging my feet using torch and I was wondering does this lib offer anything over torch ? Maybe this is better suited for the low level scientist trying to learn theory ? Or is it just an alternative ?
Omg I’m trapped
Nicolas Hug
@makingglitches pytorch and scikit-learn operate at different levels of abstractions but simply put in terms of scope, pytorch is for deep-learning while scikit-learn is for the rest of ML that's not deep learning. So one might be better suited than the other, depending on the theory that you're interested in.
Siddhant Khare

[FEATURE REQUEST] Add GitHub Organisation README profile

Just found out this new GitHub feature on GitHub org.

Like this: https://twitter.com/vinzvinci/status/1438033675313025024

Andrew Knyazev
When a PR generates html doc, where to see it?
Adrin Jalali
@lobpcg do you have a PR number?
Andrew Knyazev
@adrinjalali #21148
Guillaume Lemaitre
When the documentation CIs are finished
there will be a "ci/circleci: doc artifact — Link to 0/doc/_changed.html " line
You can clicked on "Details"
It will redirect to an HTML page where you will have hyperlinks to each documentation page that has been generated
in PRs we only generate documentation pages where there is a modification
Andrew Knyazev
@glemaitre great, found it - thanks!
Pavel Yakovlev
Morning all
I want to know if scikit-learn 1.1 will be released in late 2021?
Adrin Jalali
There will be minor releases this year (1.0.1 for instance), but the next major release will be next year.
1 reply
Olivier Grisel
Hello everyone, we are having a live community office hour on discord. Feel free to join to discuss your PRs!
Julien Jerphanion
:boom: :+1:
has anyone tried to implement GAM's via sklearn pipeline's before?
I'm using scikit-learn 0.22.2.post1 and getting the following error
AttributeError: '_CalibratedClassifier' object has no attribute 'classes_'when I try to use predict_probaon calibrated classifier
Do you you know if this issue is related to the scikit-learn's version ?
Noah Wöhler
Hi, can I post a call for participants in an interview study on open source projects here? If any mod wants more details via DM first, then I'm happy to oblige :)
Adam Li

Is cohen kappa score and balanced accuracy score supposed to work w/ multiclass labels?

I have a 3-class classification and I'm trying to use cross_validate, but it returns nans for all my scores. I tested the problem by running cross_val_score on all scores individually and isolated it to those 2 metrics.

X = (100, 5)
y = (100, 3)
clf is a Random Forest Classifier

from sklearn.model_selection import cross_val_score

cross_val_score(clf, X, y, cv=5, scoring='balanced_accuracy')
Adrin Jalali
@razou could you please paste a fully reproducible piece of code?
@NoahWoehler_twitter we get quite a bit of these requests these days (which is a good thing, shows people are looking into issues). But it would help people decide if they want to spend time on it, if you give a tiny bit of intro on what it is. Also, feel free to send an email to the mailing list with that information if you want to reach more people.
Noah Wöhler
Sure, I wasn't sure whether this falls under advertising. We are looking for open source contributors who are willing to talk to us about how security and trust are handled within their projects' communities. This is the landing page with more info: https://research.teamusec.de/2021-interviews-oss/