Presented By O’Reilly and Intel Nervana
Put AI to work
September 17-18, 2017: Training
September 18-20, 2017: Tutorials & Conference
San Francisco, CA

Training vision models with public transportation datasets

Mo Patel (Independent), Laura Froelich (Think Big Analytics, a Teradata Company)
1:30pm–5:00pm Monday, September 18, 2017
Verticals and applications
Location: Yosemite BC Level: Intermediate
Secondary topics:  Data and training, Transportation and autonomous vehicles
Average rating: **...
(2.00, 3 ratings)

Prerequisite Knowledge

  • An intermediate understanding of Python, linear algebra, and matrix operations
  • Basic familiarity with image processing

Materials or downloads needed in advance

  • A laptop with the ability to SSH into a server and a browser that can connect to Jupyter notebooks on servers installed (You will be provided GPU-powered servers with the KITTI dataset installed for training models.)

What you'll learn

  • Learn how to build and manage training datasets for computer vision
  • Understand how to train computer vision models using convolutional neural networks


Computer vision is a key component in the artificial intelligence revolution. Assisted by deep learning, object detection allows automotive applications to make key navigation, guidance, and driving decisions based on the field of view and the objects within it to avoid collisions and navigation errors. Laura Froelich and Mo Patel demonstrate how to train deep learning models for object detection using publicly available transportation datasets, focusing on building and training a convolutional neural network (CNN) using techniques from a combination of leading-edge research papers to achieve the best training results.

Topics include:

  • The architecture and implementation of state-of-the-art convolutional neural networks for object detection, such as SSD, YOLOv2, Faster R-CNN, and DeepMask
  • Annotation techniques and training dataset creation
  • Hands-on model training using TensorFlow
Photo of Mo Patel

Mo Patel


Mo Patel is an independent deep learning consultant advising individuals, startups, and enterprise clients on strategic and technical AI topics. Mo has successfully managed and executed data science projects with clients across several industries, including cable, auto manufacturing, medical device manufacturing, technology, and car insurance. Previously, he was practice director for AI and deep learning at Think Big Analytics, a Teradata Company, where he mentored and advised Think Big clients and provided guidance on ongoing deep learning projects; he was also a management consultant and a software engineer earlier in his career. A continuous learner, Mo conducts research on applications of deep learning, reinforcement learning, and graph analytics toward solving existing and novel business problems and brings a diversity of educational and hands-on expertise connecting business and technology. He holds an MBA, a master’s degree in computer science, and a bachelor’s degree in mathematics.

Photo of Laura Froelich

Laura Froelich

Think Big Analytics, a Teradata Company

Laura Froelich is a data scientist at Think Big Analytics, a Teradata Company, where she is dedicated to utilizing data to discover patterns and underlying structure to enable optimization of businesses and processes, particularly through deep learning methods. Previously, she was part of a research group investigating nonspecific effects of vaccines using survival analysis methods. Laura holds a PhD from the Technical University of Denmark. For her dissertation, Decomposition and Classification of Electroencephalography Data, Laura used unsupervised decomposition and supervised classification methods to research brain activity and developed rigorous, interpretable approaches to classifying tensor data.