Reproducible research in BCI has a long way to go…

  • Unavailable code
  • Exotic data format/language/toolboxes
  • Preprocessed data (including errors)

-> No benchmark of BCI algorithms -> Huge waste of time for everyone

MOABB aims to be the standard benchmark for any new paper

  • Comprehensive benchmark of popular BCI algorithm
  • Extensive list of freely available EEG datasets
  • Ranking algorithms with fair evaluations