You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
The poor performances of the multi-labels BraTs dataset #138
Hi！I have trouble with multi-label segmentation. Could you tell me if I want to have a triple classes segmentation (background+label 1+label2), where and which parameters I need to change
Activity
princerice commentedon Oct 24, 2023
I would like to know your batch and the number of training steps
smallboy-code commentedon Oct 24, 2023
batch_size: 32 and training steps : 30000
princerice commentedon Oct 24, 2023
Thank you for your answer. My batch is set to 8 and I use A5000GPU. I would like to know the size of the GPU memory you use
smallboy-code commentedon Oct 24, 2023
I using the four A100 GPUs. And I get the normal mask without brain tissue boundary, but the Dice is lower, about 0.4. So do you have this issue?
princerice commentedon Oct 24, 2023
I trained 40,000 steps on a gpu and the results are not good and we may need to adjust and improve ourselves
princerice commentedon Oct 24, 2023
I'm also confused that the loss has leveled off but the results are far from satisfactory
smallboy-code commentedon Oct 24, 2023
Yes, and I want to know the clamp range in the final "process_xstart" function. I am use (0,3).
princerice commentedon Oct 24, 2023
This part is not clear to me yet, I am just trying to reproduce and learn the code
teapanda628 commentedon Dec 7, 2025
Hi！I have trouble with multi-label segmentation. Could you tell me if I want to have a triple classes segmentation (background+label 1+label2), where and which parameters I need to change