#8282nd of 118 in Learning and cookbooks

nndl-practice

PyTorch notebooks for a Chinese deep learning textbook, from tensors to LLM agents

Stars
3.6k
License
no license
Last commit
Sep 2026
Last release
Jun 2026
Language
Jupyter Notebook
Stars, last 29 weeks+242
25 Apr 2026: 3,330 stars2 May 2026: 3,330 stars9 May 2026: 3,331 stars16 May 2026: 3,333 stars23 May 2026: 3,334 stars30 May 2026: 3,341 stars6 Jun 2026: 3,355 stars13 Jun 2026: 3,366 stars20 Jun 2026: 3,372 stars27 Jun 2026: 3,378 stars4 Jul 2026: 3,384 stars11 Jul 2026: 3,394 stars18 Jul 2026: 3,401 stars25 Jul 2026: 3,410 stars1 Aug 2026: 3,427 stars8 Aug 2026: 3,434 stars15 Aug 2026: 3,441 stars22 Aug 2026: 3,449 stars29 Aug 2026: 3,452 stars5 Sept 2026: 3,469 stars12 Sept 2026: 3,475 stars19 Sept 2026: 3,493 stars26 Sept 2026: 3,507 stars3 Oct 2026: 3,516 stars10 Oct 2026: 3,526 stars17 Oct 2026: 3,540 stars24 Oct 2026: 3,551 stars31 Oct 2026: 3,558 stars7 Nov 2026: 3,564 stars14 Nov 2026: 3,572 stars

3,330 on 25 Apr 20263,572 on 14 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Companion code for 邱锡鹏's 神经网络与深度学习:案例与实践 (2nd edition). Ten chapters of Jupyter notebooks implement key components from scratch in PyTorch, covering linear models, CNNs, RNNs, attention and Transformers, GNNs, and nanoGPT with LoRA, SFT, DPO, ReAct and RAG. A small teaching framework (`pytorch/nndl/`) and per-chapter pytest sanity tests are included.

Who it is for: Chinese-reading students and engineers learning deep learning in PyTorch

Strengths

  • Ten chapters run from tensor basics to nanoGPT, LoRA, DPO, ReAct and RAG
  • Notebooks default to CPU; longer-training chapters have short configs
  • Per-chapter pytest sanity tests check operators, model shapes and key results
  • Reusable `nndl` package grows from RunnerV1 to V3 across chapters

Weaknesses

  • Text and notebooks are in Chinese only
  • No license file, so reuse terms are unclear
  • Second edition is pre-publication; code and book text may be out of sync
  • Chapters 6 and 8 need IMDB / LCQMC downloaded manually

What it needs

  • GPU optional
  • Needs Python 3.11+, PyTorch 2.7+, torchvision, Jupyter, pytest
  • Models: LeNet-5, ResNet, LSTM, Transformer, GCN

Also in Learning and cookbooks

See all 118
Also in Learning and cookbooks
RankProjectScore
1ai-engineering-from-scratchFree curriculum of 542 coding lessons on building AI systems from scratch79.4k starsMIT91 out of 100
2generative-ai-for-beginners21-lesson course on building generative AI apps in Python and TypeScript123.8k starsMIT83 out of 100
3rag_techniquesCollection of 42 runnable notebooks implementing RAG techniques30.4k starscustom license83 out of 100
#4cs249r_bookOpen textbook series on ML systems, with TinyTorch, labs and hardware kits30.2k starscustom license83 out of 100
#5agents-towards-productionJupyter tutorials for taking LLM agents from prototype to production21.9k starscustom license82 out of 100