Scikit-learn

Scikit-learn (formerly scikits.learn and also known as sklearn) is a free and open-source machine learning library for the Python programming language. It helps computers learn from data by using different methods such as classification, regression, and clustering.

Some of these methods include support-vector machines, random forests, gradient boosting, k-means, and DBSCAN.

One of the great things about scikit-learn is that it works well with other Python tools like NumPy and SciPy. This makes it easier for people to build and test machine learning models.

Scikit-learn is supported by NumFOCUS, a group that helps keep important open-source projects running. Because it is free and open to everyone, scikit-learn has become popular among scientists, teachers, and students who want to explore how machines can learn and make decisions.

Overview

The scikit-learn project started as a summer job by French data scientist David Cournapeau. Later, a group of researchers from France led the project and released the first public version in 2010. Today, scikit-learn is one of the most popular tools for machine learning. It is used all over the world to help computers learn from data and make predictions and decisions.

Features

Scikit-learn gives you many tools to work with data. It has many machine learning methods and ways to get data ready to use.

The library also has useful functions for common jobs, like splitting data into parts for testing and checking your work. It helps you follow a clear path from the beginning to the end when you work with data.

Examples

Here is a simple example of how to use a random forest classifier with scikit-learn. You start by importing the RandomForestClassifier and making one. Then, you give it some data (called features) and labels (called classes) to teach it. Finally, you use the .fit() method to train it with your data.

>>> from sklearn.ensemble import RandomForestClassifier
>>> classifier = RandomForestClassifier(random_state=0)
>>> X = [[ 1,  2,  3],  # 2 samples, 3 features
...      [11, 12, 13]]
>>> y = [0, 1]  # classes of each sample
>>> classifier.fit(X, y)
RandomForestClassifier(random_state=0)

Implementation

Scikit-learn is mostly written in Python and uses NumPy for math and data tasks. Some parts are written in Cython to make them faster.

It works well with many Python tools, like Matplotlib and plotly for making graphs, Pandas for handling data tables, and SciPy for scientific calculations.

History

Scikit-learn started as a Google Summer of Code project in 2007 by David Cournapeau. Soon after, Matthieu Brucher joined and used it for his school work. In 2010, INRIA, the French Institute for Research in Computer Science and Automation, helped with the project. The first public version, v0.1 beta, came out in January 2010.

The project’s first stable version, 1.0.0, was released on September 24, 2021. The latest version, 1.8, arrived on December 10, 2025. This version added support for using PyTorch and CuPy arrays for faster computing, plus some bug fixes and new features.

Applications

Scikit-learn helps computers learn from data in many fields. Companies like AXA and J.P. Morgan use it to make better choices about insurance and banking. Places such as Booking.com use it to suggest hotels and find fake bookings.

Other companies, like Spotify and Evernote, use scikit-learn to make recommendations and handle lots of information. Schools, such as Télécom ParisTech, teach students how to use it to learn about machine learning.

Awards

Scikit-learn has won awards for its work in machine learning and open science. In 2019, it got the Inria-French Academy of Sciences-Dassault Systèmes Innovation Prize. In 2022, it received the Open Science Award for Open Source Research Software for its quality, many contributors, and good documentation.