AI Crash Course: A fun and hands-on introduction to machine learning, reinforcement learning, deep learning, and artificial intelligence with Python

I am super-excited for my dear friend Hadelin for publishing his first book! I love my copy - it's got fantastic explanations and hands-on examples of AI applications. If you've taken any of our courses on Deep Learning and Artificial Intelligence, you will find this book a fantastic supplement to your learning journey. And if you're new to Hadelin's teaching method, you're in for a treat! This book will hold your hand as you venture into the world of AI, Reinforcement Learning, Q-Learning, Deep Convolutional Q-Learning and more.This book is build around real-world use-cases of AI: from advertising and energy savings to warehouse optimization and self-driving cars. The business applications is what really makes this book stand out. However, what I, perhaps, appreciate the most is that Hadelin gets you started from scratch. Artificial Intelligence libraries have gotten so easy to use that even with minimal knowledge of Python you can write up a few dozen lines of code - and voila! Have a Deep Neural Network ready to go. In this book Hadelin makes sure to supply you with that minimal Python foundation to get you going.In short - you're going to love this book, learn a lot and feel accomplished after every chapter!- Kirill

So far this book has been excellent (I'm currently on Ch6). It is a quick read, explaining the concepts well. I hope to have a first read through done by the end of the day. The author seems to have strong understanding of AI fundamentals, but can still explain in a way that a novice can understand.But.....All the graphs are in grey scale / black and white. This makes it very difficult to distinguish between the yellow, green and purple plots on a graph. Or which plot corresponds to which title in the key.

In AI Crash Course, Hadelin has capitalized on his years of experience teaching thousands of learners across the globe by taking the best of his MOOC content and converting it into a concise and engaging book format. Even if you have little or no experience coding in Python, Hadelin provides you with just enough of the basics to successfully implement the hands-on projects included in the book. Coupled with the use of Google's Colab environment, even novices can successfully complete the most sophisticated models included, with the end result being a working, shareable portfolio of real-world AI applications.

I loved the examples on how to apply AI. Most of books focus on teaching the syntax but no how to apply it!!

I don't know what a "crash course" should contain. I would only say that this book is narrowly focused. For a full treatment of AI see AI A Modern Approach by Russell and Norvig. Books like The Encyclopedia of Artificial Intelligence will also give you some idea of the 100 or so active AI subfields.

I'm really enjoying this book and learning quite a bit. The descriptions are excellent, as are the encouragements to try to figure things out on your own before being given the answer/solution. I definitely recommend this book.

The book is written with very wrong audience.Many things are repeated several times for filling the book volume (one figure is used many times without appropriate info on the figure).The figures are in black and white, when the content addresses the color items!!Some Python code sections in the book have problems!Code download website is a trap??

This book is an introductory crash course that covers the four modern AI models: Thompson Sampling, Q-Learning, Deep Q-Learning, and Deep Convolutional Q-Learning. It starts with a brief introduction of those models followed by the list of AI-applicable industries. Just like his online courses, Hadelin makes sure to provide the basic tool set and guidance for those who do not have any prior knowledge of the AI modeling and/or Python libraries. For each topic, the book navigates the reader by using the same 3-step approach: (1) a short description of how the model works, (2) the basic math behind the model theory, and (3) implementation of the model using Python.The author also provides the invaluable lessons of how to architect the AI model in practice. When we are beginners of coding a model, we often get lost and/or end up with starting the part that might not be most efficient to begin with. In each use case, Hadelin explains how to formulate the business problem into the AI modeling framework in the most efficient way, all of which is elegantly explained in casual and plain English with passion and enthusiasm. In the actual implementation, most lines of code are also explained, so it is easy to follow how it works or why such a parameter is set.The book understandably omits certain more advanced topics. For example, when implementing an artificial neural network, either traditional or convolutional, it is essential to prevent overfitting from happening. Learning the parameter tuning and various methods to detect and avoid overfitting is apparently beyond the scope of this book; however, I think it should be worth mentioning the issue.Overall, I highly recommend this book for those who like to quickly jump into the AI modeling world.

I am surprised how this got published. On page 42 he is mentioning a formula for a distribution, before on page 48 (!) he tries to tell you what a distribution IS. He uses the word "magic", pretty much the last word I want to see in an AI book. He refers to COLORS in diagrams that are clearly printed in GRAYSCALE (e.g. p. 51). He starts with a Python course that leads nowhere because along the way, further Python concepts... and that is really how this feels, as an entire book: a series of scraps, chaotically scattered, in no logical sequence, full with deviations. The author has neither decided on the audience, nor on the content — is it for beginners, is it for experts, how much theoretical concepts, and when which (which mathematical, which algorithmic). I am sorry, but I cannot recommend this at all.

Beaucoup de techno ne sont plus d'actualité. Le code ne marche plus car certaines librairies ne sont plus disponibles.Dans l'ensemble, un livre intéressant. Mais il faut une bonne mise à jour.

Tout est dans le titre. Les exemples sont tout autant complexes que concrets et très bien expliqués.