search

hust-linyi / SC-Net

This is the official code of the paper "Nuclei Segmentation with Point Annotations from Pathology Images via Self-Supervised Learning and Co-Training"
17 stars 1 forks source link

About the dependency issue. #5

Open joonmy opened 9 months ago

joonmy commented 9 months ago

Hello, thank you for your work. I have a question. what is the recommended python version? I think there are some dependency issues when I run 'pip install -r requirement.txt'.

joonmy commented 9 months ago

Moreover, why model 2 is WNet while model 1 is just ResWNet?

image
joonmy commented 9 months ago

Why the code make H-component & voronoi diagram & cluster annotation only for train dataset? In your val() function, valid image also need those informatoin.

joonmy commented 9 months ago

Network's input channel is 1. However, in your code, the pretrained weight of ResNet has 3 channel. How could you use the pretrained weight?