Where: Hall B
Start Time: 09:00
End Time: 17:00
Date: Monday, May 20, 2019 from 9:00 am to 5:00 pm
Location: The Embedded Vision Summit, Santa Clara, California
When you sign up for this training, remember to also sign up for the Embedded Vision Summit. The Summit gives you access to two additional days of learning after your TensorFlow training. You’ll learn from 90+ presentations, see 100+ demos and have the opportunity to talk with 1,200 fellow innovators building or enabling vision in their products.
Are you an engineer who wants to design intelligent computer vision systems that learn from complex or large-scale datasets? Would you like to build your visual AI development skills?
Get the hands-on knowledge you need to develop deep learning computer vision applications—both on embedded systems and in the cloud—with TensorFlow, today’s most popular framework for deep learning. (As an added bonus, this class will cover the latest developments that will be released in the forthcoming TensorFlow 2.0 release—our instructor is Google Developer Expert and thus has early access to TensorFlow 2.0!)
The class covers deep learning for computer vision applications using TensorFlow 2.0. We assume that:
- You know the basics of deep learning algorithms and concepts for computer vision, including convolutional neural networks.
- You know the basics of the Python programming language
- You do not know TensorFlow or TensorFlow 2.0
To make sure you’re up to speed on Python, please review sections 1-5 of this online Python tutorial before class: https://docs.python.org/3/tutorial/.
The class runs 9 am to 5 pm with lunch and breaks. The class is a combination of lecture and lab exercises using Jupyter Notebooks. We cover the following topics:
- Introduction to AI, machine learning, and TensorFlow 2.0
- Neural networks in TensorFlow
- Linear regression
- Shallow image recognition
- Deep image recognition
- Convolutional neural networks
- Data set creation and augmentation
- Off-the-shelf network architectures; transfer learning
- Object detection
- TensorFlow Lite
You will need to bring a laptop with wifi and Chrome or Firefox browser. All the lab work will be done in the cloud using Jupyter Notebooks, so you won’t need to install any software on your laptop.
About the Instructor:
Doug Perry, a Google Developer Expert in TensorFlow, is an experienced hardware engineer who has worked in the field of artificial intelligence since 2007. Recent examples of some of his projects include an AI-based sports application for golf and baseball swing classification and a consumer application using facial recognition technology to categorize facial features. He also architected a quantized neural network accelerator using FPGAs that showed a 100X speedup relative to GPU based architectures. His most recent project is a TensorFlow model for financial options trading. He the author of numerous books on hardware design and holds several patents related to hardware instrumentation and debugging.