CodaLab Competitions: An Open Source Platform for Organizing Scientific Challenges
Authors: Adrien Pavao, Isabelle Guyon, Anne-Catherine Letournel, Dinh-Tuan Tran, Xavier Baro, Hugo Jair Escalante, Sergio Escalera, Tyler Thomas, Zhen Xu; Published in 24(198):1−6, 2023.
Abstract
CodaLab Competitions is a web-based open source platform that aims to assist data scientists and research teams in crowd-sourcing solutions for machine learning problems through the organization of competitions, also referred to as challenges or contests. This platform offers various useful features including multiple phases, code and result submissions, multi-score leaderboards, and running jobs within Docker containers. It is highly flexible and capable of handling large-scale experiments, allowing organizers to upload large datasets and provide their own CPU or GPU compute workers.
[abs]
[code]