Does sklearex make your sklearn machine learning model training fly fast?

One 、 introduction

scikit-learn As a classic machine learning framework , It has developed for more than ten years since its birth , It is simple and easy to use API Deeply loved by users （fit()、predict()、transform() etc. ）, Other machine learning frameworks will more or less learn from . But its computing speed has been widely criticized by users . be familiar with scikit-learn Your friends should know ,scikit-learn Some of the included in are based on joblib The effect of the operation acceleration function of the library is limited , Can't make full use of computing power .

Today I'd like to share with you a skill , It can help us without changing the original code , Get tens or even thousands of times scikit-learn The computing efficiency is improved .

Two 、 utilize sklearnex Speed up scikit-learn

In order to speed up the calculation , You just need to add sklearnex This library , Can help us to have Intel On the processor's device , The operation efficiency is greatly improved .

This newer Library , It's best to create a clean conda Experiment in virtual environment （ To avoid some dependent library versions from following Base Conflict in the environment , More unnecessary trouble ） All orders are as follows , We installed it by the way jupyterlab As IDE：

conda create -n sklearnex python=3.8
conda activate sklearnex

conda install jupyter
conda install nb_conda
pip install scikit-learn scikit-learn-intelex -i http://pypi.douban.com/simple --trusted-host pypi.douban.com

Copy

After the preparation of the experimental environment , stay jupyter notebook Write test code in to see how the acceleration effect , It's easy to use , Just import... Into the code scikit-learn Before the relevant functional modules , Just run the following code ：

from sklearnex import patch_sklearn, unpatch_sklearn
patch_sklearn()

Copy

After the acceleration mode is successfully turned on, the following information will be printed ：

The other thing to do is to put your original scikit-learn The code can be executed later , I did a simple test on my old ASUS notebook, which I usually study and write code .

With K-Means Clustering as an example , On a sample dataset with a sample size of 100000 , It only takes... After acceleration is turned on 46.84 Seconds to complete the training set , While using unpatch_sklearn() After the acceleration mode is forced off （ Be careful scikit-learn Relevant modules need Re import ）, The training time increases to 100.52 second , Means through sklearnex We got 2 Multiple times of operation speed .

give the result as follows ：

And according to the official statement , The stronger CPU A higher percentage of performance improvements can be achieved （ I'm the only one running here exampls Look at , old Intel CPU The acceleration is not enough ）, The following figure shows the official Intel Xeon Platinum 8275CL The performance improvement results after testing a series of algorithms under the processor , It can not only improve the training speed , It can also improve the speed of model reasoning and prediction , In some cases, it even reaches Thousands of times Performance improvement of ：

Officials also offer some ipynb Example ：https://github.com/intel/scikit-learn-intelex/tree/master/examples/notebooks

It shows that K-means、DBSCAN、 Random forests 、 Logical regression 、 Ridge return And other common models , Interested readers can consult and learn by themselves .

Besides , Can also accelerate sklearn stay GPU Performance on , The method of use is similar to ：

import numpy as np
import dpctl
from sklearnex import patch_sklearn, config_context
patch_sklearn()

from sklearn.cluster import DBSCAN

X = np.array([[1., 2.], [2., 2.], [2., 3.],
            [8., 7.], [8., 8.], [25., 80.]], dtype=np.float32)
with config_context(target_offload="gpu:0"):
    clustering = DBSCAN(eps=3, min_samples=2).fit(X)

Copy

Refer to the ：

• Use sklearnex Greatly accelerate scikit-learn operation
• One line of code accelerates sklearn Thousands of times