中文 | English
--- # Machine Learning Systems: Design and Implementation An open-source book explaining the design principles and implementation experience of modern machine learning systems, covering the complete technology stack from programming interfaces and computational graphs to compilers and distributed training. **English version 1 (stable):** [openmlsys.github.io/html-en/](https://openmlsys.github.io/html-en/) **English version 2:** Under reconstruction. ## Table of Contents - [Target Audience](#target-audience) - [Content Overview](#content-overview) - [Build Guide](#build-guide) - [Contributing](#contributing) - [Community](#community) - [License](#license) ## Target Audience - **Students**: Those who have mastered machine learning fundamentals and want to deeply understand the design and implementation of modern ML systems. - **Researchers**: Those who need to develop custom operators or leverage distributed execution for large model development. - **Engineers**: Those responsible for building ML infrastructure and need to tune system performance or customize ML systems for business needs. ## Content Overview The book (2nd edition) consists of 9 chapters: | Chapter | Content | |---------|---------| | [Chapter 1: Introduction](v2/en_chapters/01_chapter_introduction/) | Overview of ML system architecture and technology stack | | [Chapter 2: Programming Interfaces and Computational Graphs](v2/en_chapters/02_chapter_programming_and_graph/) | Tensor abstraction, automatic differentiation, graph representation and execution | | [Chapter 3: AI Accelerators and Programming](v2/en_chapters/03_chapter_accelerator/) | GPU architecture and CUDA/Triton/CUTLASS programming models | | [Chapter 4: AI Compilers and Runtime Systems](v2/en_chapters/04_chapter_compiler_and_runtime/) | IR design, graph optimization, kernel generation, and runtime execution | | [Chapter 5: Data Processing Systems](v2/en_chapters/05_chapter_data_processing/) | Data loading, data pipelines, and distributed data processing | | [Chapter 6: Training Systems](v2/en_chapters/06_chapter_training_systems/) | Single-node and distributed training, parallelism strategies, and training optimization | | [Chapter 7: Model Serving](v2/en_chapters/07_chapter_model_serving/) | Inference optimization, online serving, and model management | | [Chapter 8: RL Systems](v2/en_chapters/08_chapter_rl_systems/) | Reinforcement learning pipelines, environment interaction, and RL system design | | [Chapter 9: Large-scale GPU Cluster Management](v2/en_chapters/09_chapter_gpu_cluster/) | GPU scheduling, resource management, and large-scale training infrastructure | ## Changelog | Date | Event | |------|-------| | 2022-01 | Project initialized; Chinese content writing begins | | 2022-05 | Extension chapters released (Federated Learning, RL Systems, Explainable AI) | | 2023-05 | Codebase adapted to MindSpore 2.0 | | 2026-03 | Bilingual (CN/EN) build architecture refactored; English version launched | ## Build Guide ### Prerequisites - curl - git - Python 3 ### Installation ```bash # Clone the repository git clone https://github.com/openmlsys/openmlsys-zh.git cd openmlsys-zh # Install Rust toolchain (Linux/macOS) curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh # Install mdbook cargo install mdbook ``` ### Build HTML ```bash sh build_mdbook_v2.sh # English output: .mdbook-v2/book # Chinese output: .mdbook-v2-zh/book ``` For more details, see the [Build Guide](CONTRIBUTING/info_en.md). ## Contributing We welcome all forms of contributions. For the full workflow, see the **[Contributing Guide](CONTRIBUTING/CONTRIBUTING_en.md)**. Before contributing, please read: - [Writing Style Guide](CONTRIBUTING/style_en.md) - [Terminology Guide](CONTRIBUTING/terminology.md) ## Community
Join our WeChat group by scanning the QR code