ZhengRun

Hello, Could you tell me where can I find the code for multiplying the DICE loss with the classification score that you mentioned in your paper? Thank you very much!...

Training curve volatile when using E-LPIPS as generator loss

2

I have greatly admired your work “Distilling Diffusion Models into Conditional GANs” and am currently experimenting with replacing the standard reconstruction loss of the generator with E-Latent LPIPS during distillation....

Hello, thank you very much for open-sourcing this project for us to learn from. I also came across Nitro（https://github.com/AMD-AGI/Nitro-1/blob/main/train.py） and noticed that in your training process, you first train the...

您好，代码GLMixBlock类中的self.norm1和self.norm2执行时会报错

1

 这里输入的x形状为[batch,c,h,w]，但是self.norm1是LayerNorm2d((64, 1, 1), eps=1e-05, elementwise_affine=True)，执行的时候在维度上会报错。self.norm2也是这个问题。