Learning TensorFlow.js: Powerful Machine Learning in JavaScript

by Gant Laborde

Given the demand for AI and the ubiquity of JavaScript, TensorFlow.js was inevitable. With this Google framework, seasoned AI veterans and web developers alike can help propel the future of AI-driven websites. In this guide, author Gant Laborde (Google Developer Expert in machine learning and the web) provides a hands-on end-to-end approach to TensorFlow.js fundamentals for a broad technical audience that includes data scientists, engineers, web developers, students, and researchers. You'll begin by working through some basic examples in TensorFlow.js before diving deeper into neural network architectures, DataFrames, TensorFlow Hub, model conversion, transfer learning, and more. Once you finish this book, you'll know how to build and deploy production-ready deep learning systems with TensorFlow.js.

Genres: Programming, Computer Science

338 pages, Paperback

Published June 15, 2021

Reviews

[Laura Uzcategui · Rating 5 out of 5]

I was a technical reviewer for Learning Tensorflow.js book, and all I can say is: it's a great book if you want to learn how to create, train and run your models using Tensorflow.js in browser.

Personally, I don't have a ton of experience with Mobile Development and using Gant book I was able to fly through the content and learn a bit more about web development world, not only that but I was able to grasp the basic knowledge I need in case I'm developing a Website and want to run ML models within its ecosystem.

The book will take you through a journey from understanding Tensorflow ecosystem and where you can run it, to a look of practical and walk through scenarios of running it within your own web apps.

[Yaro · Rating 5 out of 5]

This is a great book for frontend engineers to jump into the machine learning world. The book describes well how to run models in the browser, lists various places where to find pre-trained models and later briefly explains basics of learning new models with tf.js. 5/5.

[Ivan Koma · Rating 3 out of 5]

At first it's nice but github source code have a problems with relative path and because of that CORS problem (in last chapters npm examples disappear)
Maybe one good big example can be more fascinating then many smaller