DISAPPEARED13

I saw the relative question opened at #83 , and @richzhang said, the mean and std is for [-1, 1], I wanna know if our data do from grayscale to...

> I think you can scale the data to [-1, +1], and repeat to 3 channels, and keep the scaling layer Thanks for replying! and is it properly that I...

> I think just keep the parameters, but there's no 100% correctly solution. This network was pretrained on Imagenet classification of natural images, so there's no real guarentee it works...

Have you solve this problem? @Pixie412 , I've met the same bug with you!

And I've noticed that when running default nnUNet, GPU-Util always got 95%-100%, how to get the parameters to control this? Thanks a lot!

My plans is edited from nnUNet_trainerV2 and change the code by just adding a extra convolutional block(not very deep) in Generic_Unet, and 4 channels input, 13 classes. But for more...

api已经更新到V6了

@leaderj1001 I have the same problem, too. @xiaosa96 have you solved this problem?

Hi there, I notice this problem, too. As we known that paper mentioned just 2 kind of loss(class loss and consistency loss) to optimize, what's the situation that student model...

assert isinstance(input_size, (tuple, list)) and len(input_size) == 3