Yes Python Anyone who knows the language knows Python Simple , Rich third-party libraries , It can be used in artificial intelligence 、 machine learning 、 Web crawler 、web Development and other fields , And the editor of this article wants to introduce the most famous , It's also a library for machine learning , It is Scikit-learn.

Scikit-learn Is aimed at Python Free software machine learning library for programming languages , There are various classifications 、 Regression and clustering algorithms , Including support vector machines 、 Random forests 、 Gradient rise ,K Mean and DBSCAN, And it aims to communicate with Python The library of numerical Science Numpy and Scipy.

Scikit-learn Project start Scikit.learn, This is a David Cournapeau Of Google Summer of Code project . Its name comes from the fact that it is Scikit The concept of , It is Scikit Independent development and distributed third-party extension of , The original code base was later rewritten by other developers .2010 Annual fee Pedregosa, Gail Varoquaux, Alexander Gramfort and Vincent Michelle, , The National Institute of information and automation, coureau, France , The French , Put the project leader and make the IPO on February 1 2010 In all kinds of scikits in ,scikit-learn as well as scikit-image stay 2012 year 11 The month is described as well maintained and popular . Scikit-learn yes GitHub One of the most popular machine learning libraries in the world .

Scikit-learn Mainly used Python Compiling , And it's widely used Numpy High performance linear algebra and array operations . Besides , use cython Write some core algorithms to improve performance , Support vector machines are made up of LIBSVM Of cython Wrapper implementation ; Similar packaging of logistic regression and linear support vector machine LIBLINEAR. In this case , May not be able to use Python Extend these methods .

Scikit-learn And many other Python Libraries are well integrated , such as matplotlib and plotly For drawing 、numpy For array vectorization 、pandas Data frame 、scipy etc. .