K.T.

The problem didn't occur when I used the code which provided in "velo2cam_projection_detail.ipynb". Still don't know why, but at least I can get the proper result.

Hi @nothinglo, Thank you for your amazing work. Could you help us to download the code? The link seems to be broken. Any chance to fix it?

Hi, are you sure this is a bug? From my point of view, `the shortcut connection is just x`, so the line you were saying might be just the convolution...

I think you that have incompatible version of CUDA. It needs CUDA11 to work according to your message. Several ways to solve this issue: 1. Reinstall the correct CUDA version...

Hi, I am also wondering that. I think the implementation of adversarial loss is indeed missing.

@Jeffery-zhang-nfls Thank you for the contribution! Will look at this later.

Hi @Dannynis and @Jeffery-zhang-nfls, I am also wondering that...I thought this second step is only happening for updating the discriminator loss? Or am i misunderstanding as well?

Hi all, I thought the 2-step loss is only happening in discriminator based on paper, but the discriminator became too strong based on my experiments. Anyway, just for anyone who...

Have you try to specify the gpu device ?

Hi @zzulvmengke @Yongjiann you can find the link to pretrained model in Readme "pretrained" section. hope this would help!