milesial / Pytorch-UNet

PyTorch implementation of the U-Net for image semantic segmentation with high quality images
GNU General Public License v3.0
9.32k stars 2.52k forks source link

Predicted output is black and Dice is 0 or unchanged #507

Open allenazz opened 3 months ago

allenazz commented 3 months ago

You can replace the optimizer in train.py,use Adam. such as optimizer =optim.adam.Adam(model.parameters(), lr=learning_rate, weight_decay=weight_decay)