Open pesekon2 opened 2 years ago
It would be nice to compare results also for trainings done using different types of loss functions. It is also being stated that boundary-based losses or compound losses could be valuable for imbalanced data.
For more, exempli gratia: https://jeune-research.tistory.com/entry/Loss-Functions-for-Image-Segmentation-Distance-Based-Losses
It would be nice to compare results also for trainings done using different types of loss functions. It is also being stated that boundary-based losses or compound losses could be valuable for imbalanced data.
For more, exempli gratia: https://jeune-research.tistory.com/entry/Loss-Functions-for-Image-Segmentation-Distance-Based-Losses