d2l-ai/d2l-zh
《动手学深度学习》：面向中文读者、能运行、可讨论。中英文版被70多个国家的500多所大学用于教学。
How d2l-zh is put together
This repository is the Chinese-language edition of the 'Dive into Deep Learning' (D2L.ai) interactive textbook. It is overwhelmingly a documentation/content project: the bulk of the codebase is Markdown chapter files organized into per-topic directories (e.g., chapter_linear-networks, chapter_attention-mechanisms), each containing executable code blocks that are built and rendered by the d2l-book/Sphinx toolchain into HTML and PDF outputs. A small amount of Python exists for build-time post-processing of generated LaTeX (static/post_latex/main.py) before PDF compilation. There is no runtime application, server, datastore, or deployment surface; data flow goes from Markdown sources through notebook evaluation and Sphinx/LaTeX builds to published web/PDF artifacts.
Languages
Frameworks
Infrastructure
Major components
Chapter content (chapter_* directories)
Markdown source for each book chapter with embedded runnable deep-learning code blocks, each paired with an English '_origin.md' source version.
Front matter and appendix
Introduction, installation guide, and tools-for-deep-learning appendix pages including AWS/SageMaker/Jupyter usage instructions.
Build configuration
config.ini drives the d2l-book build (notebook evaluation, Sphinx HTML/PDF generation); setup.py provides the Python package entry point.
LaTeX post-processing (static/post_latex)
Python script that rewrites the Sphinx-generated .tex output: unnumbers preface/summary/exercise sections, converts Section refs of chapter index pages to Chapter refs.
Bibliography
d2l.bib holds all BibTeX references cited across chapters.
Editorial standards
STYLE_GUIDE.md and TERMINOLOGY.md define writing style and consistent Chinese translations of technical terms for contributors.
No digest this week
1 weeksWeek of
Dependencies and code review
Dependency advisories
Security Watch
No known advisories across 0 scanned dependencies.
No known advisories in the scanned dependencies.
Code review
No concrete code-level issues confirmed.
Get this report every week for your repos.
GitZoid learns each repo, reports what changed, and flags what needs attention. One flat price for the whole team.
$19 a month, flat · First 10 outputs free · No card required