Reformatting a semester of NTU's ML lectures for easy reference

What it does

This repo is a complete set of study notes for Professor Hung-yi Lee’s 2021 NTU Machine Learning course. It covers the full deep-learning syllabus—from CNNs and Transformers to GANs, BERT, and reinforcement learning—with each chapter cross-linked to the original YouTube lectures. The author repackages the professor’s slides and explanations into three synchronized formats: Notion pages, a GitHub Pages site, and downloadable PDFs.

The interesting bit

There is no code here; the project is pure information architecture. Turning a video lecture series into a consistently formatted, triple-hosted reference is an exercise in documentation that other students clearly value—it has pulled in 581 stars. It is a useful reminder that curation and formatting can be a standalone contribution.

Key highlights

• Covers 14 major ML topics, from basic concepts through network compression and lifelong learning
• Every chapter links directly back to the corresponding YouTube lecture videos
• Available as Notion docs, a browsable static site, and offline-ready PDFs
• Diagrams and figures pulled directly from the professor’s official lecture slides
• Author transparently flags that a few small sections remain incomplete

Caveats

• A few sections are explicitly marked as incomplete by the author
• All content is in Traditional Chinese, so English-only readers need not apply
• Notes reflect one student’s interpretation of the lectures, so occasional errors are possible

Verdict

Worth bookmarking if you are learning ML and read Traditional Chinese, or if you want a template for turning a lecture series into a multi-format knowledge base. Skip it if you are looking for executable code or English-language tutorials.