奇怪的问题，网上大部分解答都说Pytorch版本和cuda版本不符，但是通过
torch.cuda.is_available()
结果也是True，就很奇怪。搜了下github上的解答，解决问题
RuntimeError: CUDA error: no kernel image is available for execution on the driver, when use pytorch 1.7 on linux with RTX 3090 · Issue #49161 · pytorch/pytorch · GitHub Bug RuntimeError: CUDA error: no kernel image is available for execution on the driver when use Pytorch 1.7 on Linux with RTX 3090 + ubuntun 20 + GPU driver 455.45 + CUDA 11.0 I am a skilled user of pytorch-gpu, recently I purchased an...https://github.com/pytorch/pytorch/issues/49161原本的版本是 torch = 1.11，cuda = 11.6，降低到1.7就行了。。。凭啥呀。。。奇怪的bug
以下是安装1.7.1的代码
pip3 install torch==1.7.1+cu110 torchvision==0.8.2+cu110 -f https://download.pytorch.org/whl/torch_stable.html
关于这个问题可能的原因在另一个回答下有一个解释
Cuda error: no kernel image is available for execution on the device · Issue #31285 · pytorch/pytorch · GitHubhttps://github.com/pytorch/pytorch/issues/31285#issuecomment-787582949
