about loss function

[LeeCASC]

In this paper, the author said the loss function is perfectly crossentropy, but when I modify the MSELoss to CrossEntropy Loss. The code do not work. How to modify the code better? Thank you!

[elepp]

use criterion_segment = torch.nn.BCEWithLogitsLoss() and take away nn.ReLU(inplace=True) in models.py at self.layer5 = nn.Sequential( nn.Conv2d(1024, 1, 1), nn.ReLU(inplace=True) )