I am alone on this project and don't know anything with docker, and Synapse. I made a predictive model and spend more time to try to submit this model... I don't understand why it is not possible to have access to validation data. For me, this is just an other bias to avoid small team to participate.

Created by Bertrand Miannay bmiannay
Dear Bertrand, Please understand that the validation data is highly protected in this case. A few of the data providers mandated that we could not provide their data to participants (even de-identified data). Thus, this challenge would not be possible without using docker because we would not have a large enough validation dataset. You are correct that the use of docker does take participant efforts away from building models. The challenge organizers are working to make the process easier by providing better examples and more information. Please let us know if there are specific question or problems that we can address.
I can assure you the majority of the winning models in the recent three years are contributed by teams of single person, including those with docker. And almost exclusively by teams of no more than two. i hate docker too. i think that is another disaster after R. BUT if you have more specific questions, other participants or organizers can help you out

Why can we not have access to validation data ? page is loading…