21K stars for Chinese deep-learning crib notes

What it does

This repo bundles Jupyter Notebook companions for four Chinese deep-learning video series: a PyTorch primer (“土堆”), Li Mu’s Dive into Deep Learning, Andrew Ng’s deep-learning course, and a nascent LLM-Agent track (“大飞”). Notes are numbered 100–122, 200–268, 300–354, and 400–409 respectively. A Baidu Pan link holds shared datasets. The author also runs WeChat study groups and openly solicits résumés for referral to a long list of Chinese tech and healthcare firms.

The interesting bit

The project treats GitHub as a combined CMS, file locker, and recruiting billboard. The README is roughly half study notes, half job-market hustle—complete with “沾沾喜气” (borrow some luck) screenshots of successful placements. It’s a bluntly pragmatic take on what an open-source education repo can monetize.

Key highlights

• ~21.7K stars suggest strong demand for Chinese-language DL study materials
• Notebooks are meant to be downloaded locally; GitHub’s renderer mangles images and formulas
• Explicitly requires Anaconda’s Jupyter, not PyCharm’s, for correct image rendering
• Includes a table-of-contents plugin recommendation for navigation
• Job-referral network spans 50+ companies from Alibaba to Siemens Healthineers

Caveats

• The 500-series Agent notes are marked “预计明年上线” (expected next year) and currently show only placeholder text
• One video link for the 400-series Agent course is blank in the README
• Heavy WeChat-centric workflow: QR codes for groups, personal contact for dataset link rot, and direct-messaging for “resume roasting”

Verdict

Worth a bookmark if you’re a Mandarin-speaking self-learner working through those specific video courses and want structured, numbered notes. Skip it if you expect a standalone textbook or an English-language resource; the social and recruiting layers are tightly coupled to the Chinese job market.