Deep Learning for Computer Vision with SAS. Robert Blanchard
capabilities with regard to deep learning. This book introduces the basics of deep learning with a focus on computer vision. The book details and demonstrates how to build computer vision models using SAS software. Both the “art” and science behind model building is covered.
Is This Book for You?
The general audience for this book should be either SAS or Python programmers with knowledge of traditional machine learning methods.
What Should You Know about the Examples?
This book includes tutorials for you to follow to gain hands-on experience with SAS.
Software Used to Develop the Book’s Content
To follow along with the demos in this book, you will need the following software:
• SAS Viya (VDMML)
• SAS Studio
• Python
Example Code and Data
You can access the example code and data for this book by linking to its author page at https://support.sas.com/blanchard or on GitHub at https://github.com/sassoftware.
We Want to Hear from You
SAS Press books are written by SAS Users for SAS Users. We welcome your participation in their development and your feedback on SAS Press books that you are using. Please visit sas.com/books to do the following:
● Sign up to review a book
● Recommend a topic
● Request information on how to become a SAS Press author
● Provide feedback on a book
Do you have questions about a SAS Press book that you are reading? Contact the author through [email protected] or https://support.sas.com/author_feedback.
SAS has many resources to help you find answers and expand your knowledge. If you need additional help, see our list of resources: sas.com/books.
Learn more about this author by visiting his author page https://support.sas.com/blanchard. There you can download free book excerpts, access example code and data, read the latest reviews, get updates, and more.
About The Author
Robert Blanchard is a Senior Data Scientist at SAS where he builds end-to-end artificial intelligence applications. He also researches, consults, and teaches machine learning with an emphasis on deep learning and computer vision for SAS. Robert has authored several professional courses on topics including neural networks, deep learning, and optimization modeling. Before joining SAS, Robert worked under the Senior Vice Provost at North Carolina State University, where he built models pertaining to student success, faculty development, and resource management. While working at North Carolina State University, Robert also started a private analytics company that focused on predicting future home sales. Prior to working in academia, Robert was a member of the research and development group on the Workforce Optimization team at Travelers Insurance. His models at Travelers focused on forecasting and optimizing resources. Robert graduated with a master’s degree in Business Analytics and Project Management from the University of Connecticut and a master’s degree in Applied and Resource Economics from East Carolina University.
Learn more about this author by visiting his author page https://support.sas.com/blanchard. There you can download free book excerpts, access example code and data, read the latest reviews, get updates, and more.
Chapter 1: Introduction to Deep Learning
Introduction to Neural Networks
Introduction to ADAM Optimization
Batch Normalization with Mini-Batches
Traditional Neural Networks versus Deep Learning
Building a Deep Neural Network
Training a Deep Learning CAS Action Model
Demonstration 1: Loading and Modeling Data with Traditional Neural Network Methods
Demonstration 2: Building and Training Deep Learning Neural Networks Using CASL Code
Introduction to Neural Networks
Artificial neural networks mimic key aspects of the brain, in particular, the brain’s ability to learn from experience. In order to understand artificial neural networks, we first must understand some key concepts of biological neural networks, in other words, our own biological brains.
A biological brain has many features that would be desirable in artificial systems, such as the ability to learn or adapt easily to new environments. For example, imagine you arrive at a city in a country that you have never visited. You don’t know the culture or the language. Given enough time, you will learn the culture and familiarize yourself with the language. You will know the location of streets, restaurants, and museums.
The brain is also highly parallel and therefore very fast. It is not equivalent to one processor, but instead it is equivalent to a multitude of millions of processors, all running in parallel. Biological brains can also deal with information that is fuzzy, probabilistic, noisy, or inconsistent, all while being robust, fault tolerant, and relatively small. Although inspired by cognitive science (in particular, neurophysiology), neural networks largely draw their methods from statistical physics (Hertz et al. 1991). There are dozens, if not hundreds, of neural network algorithms.
Biological Neurons
In order to imitate neurons in artificial systems, first their mechanisms needed to be understood. There is still much to be learned, but the key functional aspects of neurons, and even small systems (networks) of neurons, are now known.
Neurons are the fundamental units of cognition, and they are