Wei Jiang comments

Repositories
Issues
Comments

Results 5 comments of


                                            Wei Jiang

CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

I met the same problem but I didn't have the priviliage to upgrade g++ and CUDA. Instead, I solve it by adding `extra_cflags=["-std=c++14"], extra_cuda_cflags=["-std=c++14"]` in every call of `torch.utils.cpp_extension.load`. Besides,...

CUDA driver error

Same problem while running ```dev_scripts/dtu_test_inf/inftest_scan1.sh```

PyCUDA ERROR: The context stack was not empty upon module cleanup

same problem

failed to run inference with pretrained weights

Hi @songlin @jiangxf0929 @zhao-yiqun @Riser6 , I've met the same issue. Have you solved it? Thanks.

『报错』AttributeError: 'jittor_core.Flags' object has no attribute 'cuda_archs'

可以试一下删除 jittor 的临时文件夹 ``` rm -r ~/.cache/jittor ``` 然后再使用。 You may delete jittor's cache ``` rm -r ~/.cache/jittor ``` and try again.