Dear Bjørn,
Yes, teams may upload and use additional files with their entries, such as pretrained models for transfer learning.
However, teams must not use pretrained models or other uploaded files to attempt to subvert the requirements that (1) teams provide working training code that learns from the provided training data and (2) we are able to run the training code with the provided computational resources within the allotted time. For example, if we were to run a team's training code on two (very) different training sets, then the training process should learn from the data produce different trained models. Here are related links with more details:
https://groups.google.com/g/physionet-challenges/c/Ui0cj7Z4B6M/m/CXSeSPvyBgAJhttps://moody-challenge.physionet.org/faq/#external-dataTeams do not need to upload the data for pretraining, but they should thoroughly document the contents of the databases that they used for pretraining.
Best,
Matt
(On behalf of the Challenge team.)
Please post questions and comments in the forum. However, if your question reveals information about your entry, then please email info at
physionetchallenge.org. We may post parts of our reply publicly if we feel that all Challengers should benefit from it. We will not answer emails about the Challenge to any other address. This email is maintained by a group. Please do not email us individually.