Swift (iOS/macOS) for Data Science: Machine Learning, Neural Networks, and Data Analysis on Your Device

This book is the first practical guide in Russian that breaks the stereotype that Data Science is exclusively Python and Jupyter Notebooks. The author, Andrey Sokolov, an experienced iOS developer and machine learning specialist, proves that the Apple ecosystem provides powerful and elegant tools for working with data, building models, and deploying ML solutions directly on users' devices.

What is this book about?

You will learn how to use Swift — a modern, fast, and safe programming language — for the complete Data Science cycle: from data collection and cleaning to training neural networks and integrating ready-made models into applications. The book covers both fundamental concepts and advanced techniques, including:

• Data Science Basics with Swift: working with arrays, matrices, and statistical functions using the Accelerate framework and the Swift Numerics library.
• Core ML and Create ML: creating, training, and tuning machine learning models (classification, regression, clustering) without leaving Xcode.
• Deep Learning: building neural networks using TensorFlow Lite for Swift and integrating pre-trained models from PyTorch via Core ML Converters.
• Natural Language Processing (NLP): text analysis, sentiment analysis, and creating chatbots using the NaturalLanguage framework.
• Computer Vision: image recognition, object detection, and real-time video processing with Vision and Core Image.
• Data Visualization: building charts and dashboards with Swift Charts and PlotKit.
• Deployment: publishing ML applications on the App Store and optimizing models for resource-constrained devices.

Who is this book for?

• iOS/macOS developers who want to add Data Science and machine learning skills to their stack.
• Data professionals looking to master Swift for building high-performance native applications with ML functionality.
• Students and enthusiasts interested in modern approaches to data analysis outside the Python ecosystem.
• Anyone who wants to understand how ML models work on iPhone, Apple Watch, and Mac.

What will you learn?

• How to load, process, and visualize data using Swift.
• How to build and train a machine learning model from scratch in 15 minutes in a Playground.
• How to convert models from Python (TensorFlow, PyTorch) to Core ML format.
• How to ensure user privacy by performing ML inference locally on the device.
• How to create applications for face recognition, voice recognition, text analysis, and behavior prediction.

Each chapter contains practical examples, ready to use in Xcode, and links to a repository with source code. The book is written in an accessible language but assumes basic familiarity with Swift syntax. If you want to take your development skills to the next level and create truly "smart" applications —