如果安装2个cuda,怎么调换

CUDA_BIN_PATH113 %CUDA_PATH113%\bin

CUDA_LIB_PATH113 %CUDA_PATH113%\lib\x64

CUDA_PATH_V11_3 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3

CUDA_PATH113 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3

CUDA_SDK_BIN_PATH113 %CUDA_SDK_PATH113%\bin\win64

CUDA_SDK_LIB_PATH113 %CUDA_SDK_PATH113%\common\lib\x64

CUDA_SDK_PATH113 C:\ProgramData\NVIDIA Corporation\CUDA Samples\v11.3

NVCUDASAMPLES_ROOT C:\ProgramData\NVIDIA Corporation\CUDA Samples\v11.3

NVCUDASAMPLES11_3_ROOT C:\ProgramData\NVIDIA Corporation\CUDA Samples\v11.3

把上面一条条的加到环境变量里:

如果安装2个cuda,怎么调换_第1张图片

别忘记了path里也加上:

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\bin

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\libnvvp

C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.3\bin

C:\ProgramData\NVIDIA Corporation\CUDA Samples\v11.3\bin\win64

%CUDA_LIB_PATH113%

%CUDA_SDK_LIB_PATH113%

%CUDA_SDK_BIN_PATH113%

%CUDA_BIN_PATH113%

如果安装2个cuda,怎么调换_第2张图片

记得把上面的都清空了,就可以用cmd查看nvcc -V:

如果安装2个cuda,怎么调换_第3张图片

终于是11.6 了

你可能感兴趣的:(python,深度学习,机器学习)