Machine Learning and Rust: From Theory to Production Code

This book is a practical guide to using the Rust language in machine learning tasks. If you are already familiar with ML in Python (sklearn, PyTorch, TensorFlow) and want to transfer your skills to the Rust ecosystem, this edition is for you. The author, Alexey Vetrov, a senior engineer with 10 years of production experience, shows step by step how to build fast and safe ML systems.

What the book covers:

• Fundamentals of working with data in Rust: libraries ndarray, polars, csv.
• Implementation of classic algorithms (linear regression, kNN, decision trees) without frameworks.
• Using modern ML libraries: linfa, smartcore, burn.
• Inference and model serving: loading ONNX, TensorFlow SavedModel, working with tract.
• Performance optimization: parallelism (rayon), SIMD, caching, working with big data.
• Building production pipelines: error handling, logging, benchmarking.
• Deploying ML services: REST API with Actix-web, gRPC, WebAssembly for the browser.

Who this book is for:

• Data Scientists and ML engineers tired of slow Python pipelines.
• Rust developers who want to learn machine learning.
• Engineers working on embedded systems, edge devices, or real-time applications.
• Anyone who wants to write ML code that doesn't crash with memory errors and runs 10x faster.

What you will learn:

• How to load, clean, and transform data more efficiently than in pandas.
• How to train a model "from scratch" in Rust and compare its accuracy with Python counterparts.
• How to serialize a model and run it on a server without Python dependencies.
• How to use Rust's type system to prevent errors in ML pipelines.
• How to integrate a Rust module into an existing Python project via PyO3.

All code examples are available on GitHub and tested on real datasets. The book is written in clear language, with a focus on practice, without excessive theory. It is ideal for self-study and as a reference for production.

Dive into the world of fast, reliable, and modern machine learning in Rust!