No Coding Required: Training Models with Ludwig, Uber’s Open Source Deep Learning Toolbox
June 14, 2019 / GlobalMachine learning models perform a diversity of tasks at Uber, from improving our maps to streamlining chat communications and even preventing fraud.
In addition to serving a variety of use cases, it is important that we make machine learning as accessible as possible for experts and non-experts alike so it can improve areas across our business. In this spirit, we built Ludwig, an open source, deep learning toolbox built on top of TensorFlow that allows users to train and test machine learning models without writing code.
To explain how this powerful framework works, Piero Molino, Ludwig creator and Uber AI senior research scientist, discusses Ludwig’s origin story, common use cases, and how others can get started with the software:
Interested in learning more about Ludwig and AI at Uber? Check out the Ludwig repo and keep up-to-date with other projects Uber AI by subscribing to the Uber Engineering Newsletter!
Special thanks to Piero Molino, Wayne Cunningham, Stan Yee, Seamus Strahan-Malik, Deidre Locklear, Robert Brent Wilson, Blake Henderson, and Doug Rae for their contributions to this video.
Molly Vorwerck
Molly Vorwerck is the Eng Blog Lead and a senior program manager on Uber's Tech Brand Team, responsible for overseeing the company's technical narratives and content production. In a previous life, Molly worked in journalism and public relations. In her spare time, she enjoys scouring record stores for Elvis Presley records, reading and writing fiction, and watching The Great British Baking Show.
Posted by Molly Vorwerck
Related articles
Most popular
Lucene: Uber’s Search Platform Version Upgrade
Open Source and In-House: How Uber Optimizes LLM Training
Introducing Uber Shuttle for Events in the US
Horacio’s story: gaining mobility independence through innovative transportation solutions
Products
Company