fengdu78/coursera-ml-andrewng-notes

This repository contains one person's Chinese-language personal study notes for Andrew Ng's Machine Learning course, originally offered by Stanford University on Coursera in 2014. The problem it solves is accessibility: Ng's course is taught in English, and the author translated the subtitles and wrote comprehensive notes in Chinese so that Mandarin-speaking learners could follow along more easily. The notes cover all 18 lecture sessions of the course across 10 weeks. Topics from supervised learning include linear regression with one and multiple variables, logistic regression, regularization, neural network representation and backpropagation learning, and support vector machines with kernel functions. Topics from unsupervised learning include K-means clustering, principal component analysis for dimensionality reduction, anomaly detection using Gaussian distributions, and collaborative filtering for recommender systems. The final section covers large-scale machine learning techniques including stochastic gradient descent, mini-batch gradient descent, online learning, and map-reduce parallelism. The course concludes with a photo OCR application example that ties together the pipeline concepts. The repository provides the notes in multiple formats: Word documents, Markdown files, HTML files (with mathematical formulas rendered as images for online viewing), the original PPT lecture slides, and bilingual Chinese-English subtitle files for the video lectures. Python code from the course exercises is also included. The HTML version can be read online at ai-start.com. You would use this repository if you are learning machine learning and prefer Chinese-language materials, or if you want a structured companion reference to Andrew Ng's foundational ML course. The repository contains no runnable application code, it is a documentation and educational resource.