my problem is : i have a neural network U-Net, but to do the segmentation on my sparse annotation, i need to implement the loss function for the imbalanced classes so the article says, that there is a method where they apply 2 approaches : instance based balancing, mini-batch based balancing

i would love to understand more about these 2 approaches and how to implement them in my neural network ??

can anyone help please ?

PS: i'm a beginner, but i'm so motivated and really want to understand how it works !!

  • $\begingroup$ You may want to look at another loss function that handle unbalanced dataset such as the focal loss. $\endgroup$
    – Lelouch
    May 11 at 14:23


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.