Max Zhaoshuo Li 李赵硕
Max Zhaoshuo Li 李赵硕
The user warning is because of a recent torch upgrade that changes the behavior of `//`. When an operation like `b=a//2` is used, this warning is thrown. You can find...
@EhrazImam I am sorry that I don't have any experience with SOM. I am afraid I can't offer any help here. Have you tried with any other PCs/GPUs that you...
The provided code is supervised yes. Though you can train using proxy cues such as photometric loss. But it doesn't work as well as supervised.
Hi @Comet2dh This is not overfitting but very likely a problem of normalization. - Have you seen #49? I did a recent PR to fix this issue. - Are you...
This is odd. Do you mind sending me your checkpoint so I can take a look?
Hi @Comet2dh When I inference with your checkpoint, the result is much better than what you have. I used both pytorch 1.5.1, 1.6.0 and 1.9.0. They gave me the same...
Hi @rebecca0011 Your folde structure seems right to me. Looking at the log the **occlusion data** was not loaded properly. Do you mind checking?
Hi @yszhuu, yes, as stated in the paper, our code currently only supports batch size of 1. Changing the code of reshaping operations are needed to support larger batch size.
Hi @EhrazImam, I am sorry that I can't help with this issue. Maybe you can look elsewhere to see if Jetson Xavier has some specific set of steps to install...
Hi @mc171819, this is a very good question. The idea behind self-attention is that there will inherently be some differences between left and right images, due to both geometric factor...