Deep learning in Python - 2nd, expanded edition

Machine learning has made remarkable progress in recent years, evolving from almost unusable speech and image recognition to near-superhuman accuracy, and from programmes that couldn't beat a slightly experienced chess player to defeating world champions.
The progress in the development of learning programmes is driven by what is known as deep learning, which is a combination of theories and proven technical procedures that have enabled the development of a range of previously unfeasible applications. With its help, we can analyse and synthesise text and spoken word, translate from one language to another, recognise people or control self-driving cars.
This book will teach readers with an intermediate knowledge of Python how to design deep learning systems in this language using the Keras and TensorFlow libraries, which are used by most authors of winning systems from deep learning competitions. The explanation is based on intuitive explanations and practical examples. You will practise challenging concepts on applications in computer vision, natural language processing and generative models. This will give you the knowledge and practical skills to apply deep learning in your own projects.
The author of the book is François Chollet, the creator of the Keras library and an artificial intelligence researcher at Google.
Explanation of the basic principles of deep learning and advanced skills Creation of a deep learning system for computer vision, time series, text and generating your own creations (for example, images) How modern AI systems like ChatGPT work Description of the differences when running programmes on CPU, GPU and FPU Working with the Colaboratory web environment, which allows the use of GPU and FPU on a server