Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

DICE loss

See original GitHub issue

It is one of the most used loss functions in semantic segmentation tasks. Unfortunately still not available in keras.loss.*. It has been asked many times, #3611, #13085, #9395, #10890.

(Not sure, if it’s fit here or keras.loss.*)

Issue Analytics

State:
Created a year ago
Comments:16 (10 by maintainers)

Top GitHub Comments

2reactions

gamenerd457commented, Apr 19, 2022

So should I make a pr here

2reactions

gamenerd457commented, Apr 13, 2022

Interested

Top Results From Across the Web

Understanding Dice Loss for Crisp Boundary Detection

Dice loss originates from Sørensen–Dice coefficient, which is a statistic developed in 1940s to gauge the similarity between two samples [Wikipedia].

Dice Loss Explained | Papers With Code

Introduced by Sudre et al. in Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations.

Dice-coefficient loss function vs cross-entropy - Cross Validated

One compelling reason for using cross-entropy over dice-coefficient or the similar IoU metric is that the gradients are nicer.

The Difference Between Dice and Dice Loss - PYCAD

The dice loss in a special case of the dice coefficient but physically they are the same. Read this arrticle for more details....

Loss Function Library - Keras & PyTorch | Kaggle

Combo loss is a combination of Dice Loss and a modified Cross-Entropy function that, like Tversky loss, has additional constants which penalise either...