OSError: CUDA is required to compile MMSkeleton!

[SKBL5694]

I found the same issue in #310 . In that issue, the suggestion is to choose the correct CUDA version according to the pytorch version and the example is the same as the guide. But I do follow the guide and I found the same error. My system is linux with ubuntu 16.04, and I use the command conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch. And I use nvcc -V to check my cuda version which is ''Cuda compilation tools, release 10.0, V10.0.130'' and the cudnn is 7.6.2. I think my environment is same as the suggestion, but it still raise the error: " No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda-10.0' Traceback (most recent call last): File "setup.py", line 197, in include_dirs=[np.get_include()]), File "setup.py", line 134, in make_cuda_ext raise EnvironmentError('CUDA is required to compile MMSkeleton!') OSError: CUDA is required to compile MMSkeleton! " I don't know how to deal with this, can you help me?

[SKBL5694]

@yysijie

[Liyato]

我也不知道，希望赐教

[rosegit-lab]

any updates?

[NANDISHSHAH]

Hi any update on this