cfzd

@jd926 如果出现这个问题，肯定是你编译my_interp的CUDA和你pytorch的CUDA不一样导致的。你的pytorch可能用的是通过conda安装的libcudatoolkit，而你编译my_interp的是你本地的cuda。你可以通过如下方式查看： - libcudatoolkit: `conda list | grep cuda` - 本地cuda: `ls -al /usr/local/cuda` 如果他们版本不一致，那么你需要安装一个和pytorch的libcudatoolkit版本一致的本地cuda编译my_interp

@jd926 I have replied in #14

As the problem occurred in the backward step, the network should have finished forward step. So I think the problem might be out of memory. The spatial attention method could...

@manmanCover Sorry, I have fixed the code. It's really confusing that I never met this problem and my experiment gains better performance. I think I need to release a benchmark...

@manmanCover I tested my implementation with moderate and aggressive memory settings. I found slightly unbalanced memory consumption, but neither of my GPUs is fully occupied. ``` +-----------------------------------------------------------------------------+ | NVIDIA-SMI 375.51...

@manmanCover 64x128 is a large feature size for spatial attention method. In this case, the "over-completed map" would have a feature size of [32385,64,128]. In my implementation, I always keep...

I did't find any description of sliding window in the paper and I didn't see any reason of using sliding window, as it is an attention module.

@zhaoyanchen123 可以贴完整的命令和错误吗？这个提示`DENSE_DEPTH`不存在，但你这个config里面有啊，而且https://github.com/cfzd/MonoGround/blob/05b3baf73228db21af0f857d58e767c6485dd880/config/defaults.py#L182 中也有啊

@kaixinbear It is a very common problem that the results of mono 3D detection are not stable since it is composed of many individual tasks. It is also the same...

@clw5180 This work is mainly tested on the KITTI dataset, and it is not fully tested on NuScenes. Frankly, I think the bad results on NuScenes are a technical problem,...