With these changes the linking module is now able to apply multiple classifiers on the input data and merge the outputs via a combination of set intersection/union (to actually join the final predictions) and majority-voting/averaging (to remove duplicates).
With these changes the linking module is now able to apply multiple classifiers on the input data and merge the outputs via a combination of set intersection/union (to actually join the final predictions) and majority-voting/averaging (to remove duplicates).
For more details, and results see #350