Machine Learning. Step-by-Step Guides

Introduction

The book «Machine Learning by Tutorials», released in 2021 by the Raywenderlich Tutorial Team, is a practical guide to implementing machine learning technologies in applications for the Apple ecosystem. Unlike most ML books focused on Python, this tutorial is built entirely on the Swift language and Apple frameworks — Core ML, Create ML, Vision, and Natural Language. It is 372 pages long, allowing it to cover both basics and advanced techniques without excessive theory.

What the book is about

The book is divided into 4 parts, each dedicated to a specific aspect of machine learning on the Apple platform:

• Part I: Introduction to Machine Learning — basic concepts: what supervised and unsupervised learning are, overfitting, quality metrics. Here, the reader also creates their first model using Create ML.
• Part II: Integrating Ready-Made Models — working with Core ML: loading pre-trained models (e.g., MobileNet, YOLO), image processing, text classification, object detection.
• Part III: Creating Your Own Models — training models on custom data using Create ML and Turi Create. Topics include image classification, regression, and recommendation systems.
• Part IV: Advanced Topics — neural networks, transfer learning, working with the Natural Language framework, integration with Vision, and deploying models on the device.

Who this book is for

The book is aimed at iOS/macOS developers who already have a grasp of Swift and Xcode basics but have no experience in machine learning. It does not require knowledge of Python, linear algebra, or statistics — all explanations are given at an intuitive level with a focus on practice. It will also be useful for those who want to quickly add ML features to an existing application without diving deep into mathematics.

What the reader will learn

• How to install and configure Create ML and Core ML in Xcode.
• How to classify images, text, and sounds using ready-made models.
• How to train your own model for object recognition, emotion detection, or recommendations.
• How to use Transfer Learning to fine-tune neural networks for your data.
• How to integrate a model into an application using SwiftUI and UIKit.
• How to process real-time video using Vision and Core ML.
• How to work with Natural Language for sentiment analysis, entity extraction, and language identification.
• How to optimize a model for on-device operation (compression, quantization).
• How to test and debug ML features in the simulator and on a real device.

Strengths

Practical focus. Each chapter contains a ready-made project that can be downloaded and run. The code is written in the Raywenderlich style — with detailed comments and step-by-step instructions. Relevance. The book uses the latest framework versions as of 2021 (iOS 14, Xcode 12), making it applicable to modern projects. Accessibility. Complex concepts (neural networks, gradient descent) are explained through analogies and visualizations, without formulas. Quality of presentation. Color screenshots, diagrams, and code listings all help to quickly understand the material.

Who it is recommended for

The book will be useful for:

• iOS developers looking to add "smart" features to their applications (face recognition, filters, recommendations).
• Students and beginners who want to learn ML through practice without delving into mathematics.
• Product managers and designers to understand the capabilities and limitations of on-device ML.
• All