- Series: For Dummies
- Paperback: 360 pages
- Publisher: For Dummies; 1 edition (April 3, 2018)
- Language: English
- ISBN-10: 1119466210
- ISBN-13: 978-1119466215
- Product Dimensions: 7.3 x 0.8 x 9.2 inches
- Shipping Weight: 1 pounds (View shipping rates and policies)
- Average Customer Review: 4 customer reviews
- Amazon Best Sellers Rank: #130,383 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
TensorFlow For Dummies 1st Edition
Use the Amazon App to scan ISBNs and compare prices.
See the Best Books of 2018 So Far
Looking for something great to read? Browse our editors' picks for the best books of the year so far in fiction, nonfiction, mysteries, children's books, and much more.
Frequently bought together
Customers who bought this item also bought
From the Back Cover
- Explore the underlying machine learning concepts
- Deploy TensorFlow applications to the Google Cloud Platform
- Learn TensorFlow modules and create a neural network
Discover the magic of machine learning
TensorFlow, Google's free toolset for machine learning, has a huge following among corporations, academics, and financial institutions. With the guidance of this book, you can jump on board, too! TensorFlow For Dummies tames this sometimes intimidating technology and explains, in simple steps, how to write TensorFlow applications. Along the way, you'll get familiar with the concepts that underlie machine learning and discover some of the ways to use it in language generation, image recognition, and much more.
- Write machine learning apps
- Work with TensorFlow modules
- Apply statistical regression
- Code distributed applications
- Analyze images and text
- Use deep neural networks
- Categorize data sets
- Build TensorFlow estimators
About the Author
Matthew Scarpino has been a programmer and engineer for more than 20 years. He has worked extensively with machine learning applications, especially those involving financial analysis, cognitive modeling, and image recognition. Matthew is a Google Certified Data Engineer and blogs about TensorFlow at tfblog.com.
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
Machine Learning is both an important advance in technology and a difficult subject to master. It has led to self-driving cars, better targeted advertising and media placement, and better opponents for world champion chess and go players (as well as interactive video game opponents).
It is a subject that requires some substantial knowledge of Linear Algebra (Vectors, Matrices and Tensors), Calculus of Optimization and Partial Differentiation, Statistics including various Regression Methods, and significant knowledge of programming (usually in Python), as well as certain DevOps functions to maintain compatibility of packages and GPU hardware. Mathew Scarpino has significant background both in the exertions necessary to make Machine Learning and Deep Neural Networks work, and in teaching this challenging subject as well as explaining it online (e.g.on Stack Overflow). This is one of the best introductions and workbooks for this challenging subject, but it would be a fallacy to believe that it can be mastered with just a high school background in math and exposure to computer science. Some things I have learned in the last 2-3 years of trying to apply Deep Learning with a MS Engineering in Computer Science and significant professional experience that can lessen the difficult speed bumps of mastering this subject include: 1. Reading the principles and watching online course and videos before attempting difficult programming tasks; 2. using a Framework like Keras (mentioned in the book and included with current versions of Tensorflow) to simplify building Neural Net Models; 3. Be willing to type examples from books without initially understanding and then asking the help oi more skilled programmers and domain experts online; 4. using Containers and Package Mangers like Docker to simplify the process of making multiple packages and hardware compatibility issues (like GPU support) automatic.
This is one of the best books I've seen for Tensorflow and as a general introduction to Machine Learning. With that said it can be a bit of a struggle if you lack part of the necessary and sometimes assumed background.
The introduction chapter of the book offers relevant and concise information on the history of machine learning and some of the more popular frameworks in use today, such as Torch, Theano, Caffe, and Keras. Of course, TensorFlow is the name of the game in this book, so without further ado the book guides the reader through the installation process.
On a Windows 10 PC, the provided installation went fairly smoothly when installing python, pip3, and TensorFlow. I did have to edit my path environment variable in order to get python and pip working without absolute pathnames on the command line, and the book did not mention this at all. And I should also note that the book explains how to install a prebuilt binary for TensorFlow, not how to build TensorFlow for your specific hardware. This could be a major impediment if you have a modern deep learning accelerator such as an nvidia P100 or Volta – you’ll definitely want to build your own to take advantage of the speed.
The book is very well paced. I appreciated the gentle introduction to TensorFlow structure and syntax – the reader is guided through creating tensors and operations, and is introduced to graphs and running sessions. These foundations provide solid footing when moving into the more complicated topics of neural networks, convolutional neural networks, and recurrent neural networks. The book wraps up with concepts in parallel computing, threads, clusters, and running on the Google Cloud Platform.
The code for the book is also available online and was easily accessed. I liked Scarpino’s dry humor throughout the book also, it broke up the monotony of the topic in a fun way that was not distracting.
The paper stock used in the book is a little disappointing, however. It feels very pulpy, almost like a newspaper. The font is also too small for comfortable reading over longer periods of time. And the complete lack of color doesn’t help anything either.
But overall, this book is a very competent and thoughtful introduction to the TensorFlow application. It would certainly help to have at least a basic understanding of machine learning, so I’m not so sure that a “dummy” would find this interesting. But the unfortunate title aside, I highly recommend this book to those interested in learning about TensorFlow.
The presentation and layout of the book is typical of other books in the series, but the extensive use of actual code in the text may be offputting for readers who are skimming for the concepts alone.