© 2021 Strange Loop
Deep learning has the capacity to take in rich, high dimensional data and produce insights that can create totally new mobile experiences for developers. However, the constraints of network availability and latency limit what kinds of work can be done in the mobile application space and vastly increase the cost to developers. We have recently developed a customer facing mobile application that leverages real-time computer vision models and will share our experiences of moving multiple deep learning models from the server onto the client. In this presentation, we dive into technical solutions for porting custom architectures for various vision tasks and how to serialize them from Python to binary assets, while avoiding common issues such as unsupported hardware instructions. We also discuss the theory and practice of quantizing models, model fusion, and storing tensors in last memory format for optimization. We conclude by demonstrating how to benchmark the performance of client-side models for various devices and operating systems.
Tyler Kirby is the Principal Data Scientist at UniGroup where he focuses on dynamic pricing and computer vision. He has degrees in classics and computer science, and is an active contributor to research in digital philology. Tyler is also the proud father of two miniature dachshunds.
Shane is a St. Louis based machine learning engineer who used to be a penetration tester and is primarily interested in the intersection of machine learning and information security. Currently Shane works at UniGroup as their Director of Artificial Intelligence, using deep learning to create computer vision solutions for the moving industry. He has potentially read too many William Gibson novels.