Machine-Learning
:zap:机器学习实战（Python3）：kNN、决策树、贝叶斯、逻辑回归、SVM、线性回归、树回归
A collection of Python implementations of classical machine learning algorithms, k-NN, decision trees, SVM, AdaBoost, and more, written as study notes accompanying a Chinese-language tutorial series.
This repository is a collection of machine learning algorithm implementations written in Python 3, created as study notes accompanying a Chinese-language tutorial series. Each chapter pairs a written explanation (published on the author's blog and platforms like CSDN and Zhihu) with runnable code examples that apply a specific algorithm to a concrete problem.
The algorithms covered span the core of classical machine learning. The k-nearest neighbors chapter includes examples for date matching and handwritten digit recognition. The decision tree chapter builds a classifier for loan prediction and eye prescription fitting. Naive Bayes is applied to comment filtering and news categorization. Logistic regression is used to predict horse mortality rates. The SVM chapter works through both a simplified and a full implementation of the SMO optimization method, then applies it to handwritten digit recognition using scikit-learn. AdaBoost is implemented from scratch on single-layer decision trees, with an additional example on a difficult dataset and a section on drawing ROC curves. The regression chapters cover standard linear regression, locally weighted regression, and stepwise regression applied to predicting abalone age and used Lego set prices. A final tree regression chapter is also included.
All code is in Python and uses standard scientific computing libraries. The material is primarily written for Chinese-speaking learners, but the code itself and a linked English README are available for others. Articles are published first on the author's personal website, with reposts to CSDN, Zhihu, and other aggregator platforms.
The repository does not contain a full course or video lectures directly, but the author links to a Bilibili channel and a WeChat public account where newer content is released. If you are learning machine learning concepts from scratch and read Chinese, this is a guided walkthrough of the classical algorithms with matching practice code. Non-Chinese readers can still use the Python files directly.
Where it fits
- Study how k-nearest neighbors, decision trees, and SVM algorithms work by reading and running the provided Python examples.
- Use the handwritten digit recognition example as a starting point for your own image classification project.
- Learn how AdaBoost is implemented from scratch on decision stumps, including how to draw an ROC curve.
- Adapt the naive Bayes text classifier to categorize your own text data by following the pattern in the comments example.