himashi92 / duo-segnet Goto Github PK

Where are $$k_s$$ and $$k_c$$, cannot find in your codes.

Hi, thank you a lot for your contribution.

I cannot reproduce the results shown in your paper on spleen dataset.
Under 5%, 10% and 50% labeled setting, the final training output does not make sense:

How many epochs do you train to get these results?

Thanks for you contributions.
I notice that in nuclei dataset, one image have multiple masks for each nuclei. However, when training, in your codes, it seems that each image has only one mask. I wonder if you combine all masks into a single one or there are some procedures I neglected.