fafa-dl/lhy_machine_learning

This repository collects the course materials for Li Hongyi's machine learning courses from 2021, 2022, and 2023, all offered at National Taiwan University. The course is widely followed in Chinese-speaking communities and the materials here are shared with the original instructor's permission. The repo is maintained by a third-party contributor who also runs a Bilibili channel. What you will find here are links to lecture videos hosted on Bilibili, slide decks, and homework assignment files organized by topic and year. The 2023 course covers topics such as how ChatGPT works at a high level, the basics of how machines learn from examples, how AI systems generate text and images, large language models, speech foundation models, and diffusion-based image generation models. Earlier years cover related but somewhat different material, and all three years are included so learners can pick whichever version suits their background. The homework files are Jupyter Notebooks, which are interactive documents that mix written instructions with runnable code cells. You do not need to understand the code to browse the course materials, but completing the assignments does require running Python code, typically in a cloud environment like Google Colab. The maintainer also links to an online file folder (accessed through a Chinese social media account) that bundles all course resources in one place, useful if any direct links go down over time. The repository also points to three related projects by the same maintainer covering image augmentation tools, image classification backbones, and annotation file conversion utilities. This is primarily a resource hub for students following the Li Hongyi machine learning curriculum, not a software project you install or deploy. Its value is as a centralized, organized index of one of the most-watched free machine learning courses available in Mandarin.