jianpengz / conresnet Goto Github PK

In your code ConResNet/utils/loss.py

class BCELossBoud(nn.Module):
    ...
   def forward(self, predict, target):
         predict = torch.softmax(predict, 1)

in forward section, I think we should add predict = torch.softmax(predict, 1)

Hi,thanks for your sharing code.It is a very impressive work!

Can you please add the preprocessing file of the pancreas dataset?
I tried to modify the preprocessing file of the Brats dataset, but I encountered some problems.

Thank you !

Hello, can you upload the training code about NIH pancreas data set?

Hi!

thanks for sharing the code! it is a wonderful work! But i am wondering have you guys tested the effect of WS? what is you baseline results without WS?

Thank you!

Hi, Impressive job!

Please Can you include link to the data and image_id_list for training purposes. Thanks!

In the Data Preparation step, what does the image_id_list correspond to? For example, in the 2018 training set under HGG, a folder is named Brats18_2013_2_1. So is the first entry in image_id_list "2013_2_1"? Also, do we retain the same folder structure that comes with the downloaded data (~\MICCAI_BraTS_2018_Data_Training\HGG..." or we get rid of the HGG and LGG sub-directories and combine all training data in a separate directory?

During my training in another task, I found that the Dice improved slowly.
May I ask about the change of loss and verification set dice during the model training both in BraTS and Pancreas-CT?
Thank you!