NIH | National Cancer Institute | NCI Wiki  

Error rendering macro 'rw-search'

null

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 13 Next »

The following documentation is available for the MedICI challenge management system.

CodaLab Competitions is a powerful open source framework for running competitions that involve result or code submission. You can either participate in an existing competition or host a new competition. Most competitions hosted on Codalab are machine learning (data science) competitions, but Codalab is NOT limited to this application domain. It can accommodate any problem for which a solution can be provided in the form of a zip archive containing a number of files to be evaluated quantitatively by a scoring program (provided by the organizers). The scoring program must return a numeric score, which is displayed on a leaderboard where the performances of participants are compared.

Release notes v1.5:

  • Participants: can join competition-specific teams.
  • Organizers: can enable/monitor competition-specific teams; can run competitions in their own docker; can launch their own compute workers (to process participants' submissions) without needing to run their own Codalab instance; can provide an ingestion program to receive/parse participants' submissions; can dump their competition bundle (to save changes made with the editor); can run their own instance in "single competition mode".
  • Developers: can run Codalab in dockers.
  • Default docker: if not special docker is specified, submissions are run in codalab/codalab-legacy the default docker, which can be found on Github.



  • No labels