scikit-learn (sklearn) official documentation Chinese version

logo

sklearn 0.21.3 Chinese documentation

sklearn 0.21.3 Chinese example

sklearn English official website

introduce

sklearn (scikit-learn) is a machine learning tool based on Python language

Simple and efficient data mining and data analysis tools
Can be reused by everyone in a variety of environments
Built on NumPy, SciPy and matplotlib
Open source, available for commercial use - BSD license

Organization Building [Website]

GitHub Pages (foreign): https://sklearn.apachecn.org
Gitee Pages (domestic): https://apachecn.gitee.io/sklearn-doc-zh

Third-party webmaster [website]

Address A: xxx (Welcome to leave a message, we will improve it)

Other supplements

Official Github
EPUB download address
ApacheCN translation and proofreading part-time group 713436582

download

Docker

 docker pull apachecn0/sklearn-doc-zh
docker run -tid -p :80 apachecn0/sklearn-doc-zh
# 访问 http://localhost:{port} 查看文档

PYPI

 pip install sklearn-doc-zh
sklearn-doc-zh 
# 访问 http://localhost:{port} 查看文档

NPM

 npm install -g sklearn-doc-zh
sklearn-doc-zh 
# 访问 http://localhost:{port} 查看文档

Install scikit-learn
User Guide
- 1. Supervised learning
  - 1.1. Generalized linear model
  - 1.2. Linear and quadratic discriminant analysis
  - 1.3. Kernel Ridge Regression
  - 1.4. Support vector machine
  - 1.5. Stochastic Gradient Descent
  - 1.6. Nearest neighbor
  - 1.7. Gaussian process
  - 1.8. Cross decomposition
  - 1.9. Naive Bayes
  - 1.10. Decision tree
  - 1.11. Integrated approach
  - 1.12. Multi-class and multi-label algorithms
  - 1.13. Feature selection
  - 1.14. Semi-supervised learning
  - 1.15. Equation Regression
  - 1.16. Probabilistic calibration
  - 1.17. Neural network model (supervised)
- 2. Unsupervised learning
  - 2.1. Gaussian mixture model
  - 2.2. Manifold learning
  - 2.3. Clustering
  - 2.4. Biclustering
  - 2.5. Decomposing signals into components (matrix factorization problem)
  - 2.6. Covariance estimation
  - 2.7. Novelty and outlier detection
  - 2.8. Density estimation
  - 2.9. Neural network model (unsupervised)
- 3. Model selection and evaluation
  - 3.1. Cross-validation: Evaluating estimator performance
  - 3.2. Adjusting the hyperparameters of the estimator
  - 3.3. Model evaluation: quantifying the quality of predictions
  - 3.4. Model persistence
  - 3.5. Validation Curve: Plot scores to evaluate the model
- 4. Inspection
  - 4.1. Partial dependency graph
- 5. Dataset conversion
  - 5.1. Pipeline and FeatureUnion: combined evaluators
  - 5.2. Feature extraction
  - 5.3 Preprocessing data
  - 5.4 Missing value imputation
  - 5.5. Unsupervised dimensionality reduction
  - 5.6. Random projection
  - 5.7. Kernel Approximation
  - 5.8. Pairs of matrices, categories and kernel functions
  - 5.9. Conversion of prediction target ( y )
- 6. Dataset loading tool
  - 6.1. Common Dataset API
  - 6.2. Toy Dataset
  - 6.3 Real-world data sets
  - 6.4. Sample generator
  - 6.5. Loading other datasets
- 7. Calculate using scikit-learn
  - 7.1. Strategies for large-scale computing: larger amounts of data
  - 7.2. Computational performance
  - 7.3. Parallelism, resource management and configuration
Tutorial
- Introduction to machine learning using scikit-learn
- Statistical learning tutorial on scientific data processing
  - Machine Learning: Settings and Prediction Objects in scikit-learn
  - Supervised learning: predicting output variables from high-dimensional observations
  - Model selection: choosing estimators and their parameters
  - Unsupervised learning: the search for data representation
  - put them together
  - Ask for help
- Process text data
- Choose the right estimator (estimator.md)
- External resources, videos and talks
API reference
FAQ
Time axis

Historical version

scikit-learn (sklearn) 0.19 official document Chinese version
scikit-learn (sklearn) 0.18 official document Chinese version

How to compile and use historical versions:

Unzip the 0.19.x.zip folder
Copy the image resources of master/img to 0.19.x
For the normal compilation process of gitbook, you can use sh run_website.sh

Contribution Guide

In order to continuously improve the quality of translation, we have launched [translation, proofreading, and note-taking activities] and opened multiple proofreading projects. Contributors can receive a reward of 2 to 4 yuan per thousand words after proofreading a chapter. For ongoing proofreading activities, please see the activity list. For more details, please contact Feilong (Q562826179, V: wizardforcel).

DOCX: Initiative for Open Sharing of Research Records

We actively respond to the Open Source Initiative for Research (DOCX). Nowadays open source is not just open source, but also includes data sets, models, tutorials and experimental records. We are also exploring other categories of open source solutions and protocols.

I hope everyone will understand this initiative, combine it with your own interests, and do something within your ability. Everyone's small contribution, when gathered together, is the entire open source ecosystem.

Project leader

Format: GitHub + QQ

The first issue (2017-09-29)

@Nayi Mo smile
@moment
@小瑶

Second issue (2019-06-29)

@N!no：1352899627
@mahaoyang：992635910
@loopyme: 3322728009
Feilong: 562826179
Moment: 529815144

-- Requirements from the person in charge: (Welcome to contribute to sklearn 中文版本)

Love open source and like to show off
Use sklearn for a long time (at least 0.5 years) + submit Pull Requests>=3
Able to have time to optimize page bugs and user issues in a timely manner
Trial period: 2 months
Welcome to contact: 529815144

Contributor

【0.19.X】Contributor list

Suggestions and feedback

File an issue on our apachecn/pytorch-doc-zh github.
Send an email to Email: [email protected] .
Just contact the group owner/administrator in our QQ group-Search: communication method.

project agreement

Recently, many people have contacted us about content licensing issues!
Open source means that knowledge should focus on dissemination and iteration (rather than prohibiting others from reprinting)
Otherwise, if you open source it on GitHub and then say you are not allowed to reprint it, you must be sick!
Commercialization is prohibited, comply with the protocol specifications, and please note the source of the address. Key points: No need to send us an email to apply.
Projects without an agreement under the ApacheCN account will be regarded as CC BY-NC-SA 4.0.

Kind tips:

For those who want to make a copy and update it themselves
I also had this experience, but this passion could not last for a few months before I became discouraged!
Not only is your hard work wasted, but it is also wasted that more people will see your translation results! it is a pity! What do you think?
My personal suggestion is: fork -> pull requests to https://github.com/apachecn/sklearn-doc-zh
So why choose ApacheCN ?
Because when we do translation, we feel happy and pretentious, which is relatively pure!
If you like it, you can participate/even be responsible for this project without any restrictions on academic qualifications or background.