220 - What is the best loss function for semantic segmentation?

Описание к видео 220 - What is the best loss function for semantic segmentation?

IoU and Binary Cross-Entropy are good loss functions for binary semantic segmentation. but Focal loss may be better.

Focal loss is good for multiclass classification where some classes are easy and other difficult to classify. ​It is just an extension of the cross-entropy loss.​ It down-weights easy classes and focuses training on hard to classify classes.​ In summary, focal loss turns the model’s attention towards the difficult to classify examples.​

Комментарии

Информация по комментариям в разработке

Похожие видео