A PyTorch tutorial book project with online readings and open-source code, focusing on practical deep learning applications and deployment. It has a sizable community, ongoing activity, and releases for PDF/EPUB. License is listed as CC BY-NC 4.0 in the README section.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
A written tutorial/book: PyTorch 实用教程（第二版）， covering PyTorch basics, applied projects in computer vision, NLP, and LLM deployment, with deployment frameworks and code structure for project-oriented learning. Online reading is open-source and free, and the accompanying code is open-source and free.
How it works
The book is structured in three parts: 上篇 (intro to PyTorch), 中篇 (industrial applications in CV, NLP, LLM), 下篇 (inference deployment). It includes practical projects across CV tasks (image classification, segmentation, object detection, tracking, GAN/Diffusion, image captioning, image retrieval), NLP tasks (text classification, translation, NER, QA, generation), and LLM deployment topics like ONNX/TensorRT, PTQ/QAT.
Getting started
Install/usage commands are not provided in the truncated README excerpt available here, but the project links include online reading and repository with code:
- Online reading: https://tingsongyu.github.io/PyTorch-Tutorial-2nd/
- Code: https://github.com/TingsongYu/PyTorch-Tutorial-2nd
Recent releases
Latest releases (from RELEASES):
- v1.0.0 v1.0.0-PDF-epub-发布 (2024-05-20)
- v1.0.0 (EPUB) (2025-12-20)
Traction
Stars: 4573 Forks: 487 Open issues: 1 Last push: 2026-08-04
Behind the repo
Linked project hosting: online reading site and GitHub repository; license listed in the README-based appendix as CC BY-NC 4.0.
Caveats
License: knowledge-sharing license CC BY-NC 4.0 International; open issues: 1. Creation date: 2021-12-16; language: Jupyter Notebook; last push: 2026-08-04.