my environmen setup as
NVIDIA-SMI 525.78.01 Driver Version: 525.78.01 CUDA Version: 12.0
torch 1.13.0a0+git7c98e70
cuda 11.7
os Ubuntu 20.04.6 LTS Debian 5.10.158-2 (2022-12-13) x86_64
run the instruction: python -c “import torch; print(torch.cuda.is_available())”, console output
/output/workspace/pytorch/torch/cuda/__init__.py:88: UserWarning: CUDA initialization: CUDA unknown error - this may be due to an incorrectly set up environment, e.g. changing env variable CUDA_VISIBLE_DEVICES after program start. Setting the available devices to be zero. (Triggered internally at /output/workspace/pytorch/c10/cuda/CUDAFunctions.cpp:109.)
return torch._C._cuda_getDeviceCount() > 0
...
a few days ago, that is normal but today suddenly unusable, very weird.