Orateur
Mlle
Camille Marini
Description
RAMP (Rapid Analytics and Model Prototyping) are collaborative data challenges, which aim at connecting data science to domain sciences, which is the main mission of the Paris-Saclay Center for Data Science.
The lifecycle of a RAMP is the following: A domain science expert brings data and an associated prediction problem, and a data scientist helps clean the data and formulate a machine learning problem. A RAMP event is then organized with 30-50 data scientists and domain scientists. During the RAMP, the participants submit a model generating predictive solutions. The models are trained on our back-end. The scores are displayed on a leaderboard. All participants have access to all code, and they are encouraged to look at and to reuse each other's solutions.
Past RAMP lead to a significant improvement over the baseline. Since the organizers have access to all the code, the result of the event is a fully functioning near-optimal prototype.
In this presentation, we are introducing the two tools we are building to run these collaborative data challenges:
- databoard: a platform where participants can submit their code (machine learning models), see code from other participants, and follow the leaderboard.
- datarun: a platform where submitted models are trained.
