Dear organizers, Would you let us know the evaluation metrics ? Regards, Satoshi

Created by SKJP
Dear Satoshi, The metric for the ranking will be the application-dependent balanced accuracy with metric-based aggregation. For more information, you could refer to the publication of our previous challenge ([MISAW Preprint](https://arxiv.org/abs/2103.13111)) or the corresponding evaluation script ([MISAW evaluation script](syn25325779)). Please note that for task 3 and 5, we will also compute the IoU for the segmentation. This metric will be only used for information, not for the ranking.

Evaluation metrics page is loading…