MNet_DeepCDR icon indicating copy to clipboard operation
MNet_DeepCDR copied to clipboard

Published 20 hours ago • verified publisher icon HzFu

Training details

Open song-cc opened this issue 3 years ago • 0 comments

I am using MNet to train the REFUGE dataset and test it on the REFUGE validation dataset, and I found that my Dice is only 0.89, but the result using the model you pre_trained is 0.93. I'd be grateful if you could provide details about REFUGE training. Including the number of training pictures, learning rate, etc.

song-cc avatar Jan 07 '22 05:01 song-cc