cudaGetDevice

[albrono]

Hello,
I have installed the offline package of deepmd v2.0.0 (CUDA10.1) on an HCP, but I get the below error when I run the "dp train" command:
RuntimeError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version
The GPU is:
TITAN X (Pascal), computeCapability: 6.1
The Cuda version is:
Cuda compilation tools, release 10.0, V10.0.130
How can I solve the issue? ("I cannot update the CUDA driver on the cluster")
Thanks,

[amcadmus]

just as the error message said " CUDA driver version is insufficient for CUDA runtime version ". plz upgrade the driver to match the cuda runtime