Deep Learning bible: translated by the internet, for charity
A volunteer Chinese translation of the canonical Deep Learning textbook that uses open source as a proofreading engine and a publishing experiment.

What it does
This repository holds the community-translated Chinese edition of the canonical Deep Learning textbook. Four volunteers divided the twenty chapters among themselves, then opened the LaTeX source to GitHub contributors for proofreading and corrections. The output is a downloadable PDF the maintainers describe as at least 90% readable and accurate, intended as a companion to—not a replacement for—the original English text.
The interesting bit
The project treats open source less as software development and more as distributed editorial quality control. The team also donated their estimated ¥20,000 print royalty to a rural education charity, betting that freely available PDFs can actually support rather than undermine physical book sales.
Key highlights
- Twenty chapters plus references, parceled across four core translators and a long tail of volunteer proofreaders who filed issues and PRs
- PDF generated from open LaTeX source, with a helper script that converts chapters to Markdown for web reading
- Figures and bibliography files deliberately excluded from the repo to respect copyright, meaning the source builds an incomplete but text-complete edition
- Proceeds from the printed version (published by Posts & Telecom Press) donated to 杉树公益 for high-school student support in Guizhou
- The v0.5-beta PDF is frozen; ongoing fixes land in source only, while the print edition is considered the authoritative release
Caveats
- No figures or bib files in the repository, so you will need the original book or the print edition for the full visual and citation context
- The maintainers acknowledge they are not professional translators, and some variance in readability across chapters persists
- Because of copyright constraints, the initial PDF release is no longer updated; readers who want the latest corrections must track the source or buy the paperback
Verdict
Chinese-speaking readers looking for a free, searchable reference to the Deep Learning fundamentals should start here, but serious researchers and anyone who needs the original figures should still keep the English edition close at hand.
Frequently asked
- What is exacity/deeplearningbook-chinese?
- A volunteer Chinese translation of the canonical Deep Learning textbook that uses open source as a proofreading engine and a publishing experiment.
- Is deeplearningbook-chinese open source?
- Yes — exacity/deeplearningbook-chinese is an open-source project tracked on heatdrop.
- What language is deeplearningbook-chinese written in?
- exacity/deeplearningbook-chinese is primarily written in TeX.
- How popular is deeplearningbook-chinese?
- exacity/deeplearningbook-chinese has 37.3k stars on GitHub.
- Where can I find deeplearningbook-chinese?
- exacity/deeplearningbook-chinese is on GitHub at https://github.com/exacity/deeplearningbook-chinese.