training and validation Dice scores always 0 #1874
Unanswered
weixiaoxkq
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I am training an auxiliary labeling tool using a public open-fracture dataset and a Segmentation model. The training process runs normally and the loss decreasing, but both the training and validation Dice scores always 0 throughout the process.
I have uploaded my modified segmentation.py file, and the training logs.
My training dataset contains four label classes (excluding background), all saved in NIfTI (.nii.gz) format. I have already confirmed that the labels and images have exactly the same spacing, orientation, and other metadata.
Could someone help me check what might be causing this issue?
segmentation.py file
train log
Beta Was this translation helpful? Give feedback.
All reactions