Artificial Intelligence with Python Cookbook: Practical recipes for next-generation deep learning and neural networks using TensorFlow and PyTorch

English | 2020 | ISBN: 978-1789133967 | 286 Pages | PDF, EPUB | 68 MB

Work through practical recipes to learn how to automate complex machine learning and deep learning problems using Python
With artificial intelligence systems, we can develop goal-driven agents to automate problem-solving. This involves predicting and classifying the available data and training agents to execute tasks successfully. This book will help you to solve complex AI problems using practical recipes.
The AI with Python book starts by showing you how to install Python and its essential packages and then takes you through the fundamentals of data loading and exploration of datasets. You’ll learn how to build probabilistic models and work with heuristic search techniques. You’ll also understand how to use deep learning techniques to perform optical character recognition and build models for videos, speech-to-text, and gender recognition. As you advance, the book also covers segmentation techniques, reinforcement learning, neural networks, and genetic programming with the help of independent and insightful recipes. You’ll discover AI use cases in industries such as healthcare and insurance and explore techniques such as constraint optimization, reinforcement learning, and online learning. Finally, the book covers examples of deploying models to production.
By the end of this book, you’ll be able to identify an AI approach for solving business problems, implement and test it, and deploy it as a service.
What you will learn

• Implement data preprocessing steps and optimize model hyperparameters
• Work with large amounts of data using distributed and parallel computing techniques
• Get to grips with representational learning from images using InfoGAN
• Delve into deep probabilistic modeling with a Bayesian network
• Create your own artwork using adversarial neural networks
• Understand a model’s key performance characteristics and bring solutions to production as APIs
• Go from proof to production covering data loading and visualization as well as modeling and deployment as microservices