A Platform for Hosting and Scoring Machine Learning Competitions

What can data science do for Mongolia? Robert Ritz moved from Texas to Ulaanbaatar and founded LETU Mongolia, where he now teaches business and data science. Robert’s building some impressive applications with Anvil, from a Mongolian speech recognition app to an app for hosting machine learning competitions at LETU. He’s even using Anvil to monitor Mongolian air quality and perform stock exchange analysis.

Robert was teaching his Machine Learning course at LETU and realised there wasn’t an existing platform that quite met his needs for hosting and scoring ML competitions. So he built one with Anvil, and had the basic functionality up and running in just 10 hours!

Students using the platform can:

  • Sign in with their G suite account (restricted to students on Robert’s course)
  • View problem sets
  • Download training data
  • Submit predictions
  • View a live, dynamic leaderboard that updates when new scores are calculated
LETU Machine Learning Competition

LETU Machine Learning Competition

I built a machine learning competition site in Anvil for a machine learning class I’m teaching this semester, and that’s been a really fun experience. The reason why I liked it was I wanted to show the students that you don’t have to be a big tech company to make an app that is really easy to use and really useful, and the students are getting a lot out of it and I really enjoyed making it. I was able to put it together in a day which was pretty amazing.

Robert Ritz

Founder, LETU Mongolia