Error recognizing my GPU

1

Hi everyone, I have started working with multiple GPUs (K80) but when I run torch.cuda.is_available() the output is False, and the following error message appears in another section of my code:

AssertionError: 
The NVIDIA driver on your system is too old (found version 9010).
Please update your GPU driver by downloading and installing a new
version from the URL: http://www.nvidia.com/Download/index.aspx
Alternatively, go to: https://pytorch.org to install
a PyTorch version that has been compiled with your version
of the CUDA driver.

I have already checked my CUDA version with the nvcc–version command, it is 10.1, the Pytorch version is 1.5.1+cu101, torchvision 0.6.1+cu101, these are the specifications of the GPUs:

image
image646×503 22.5 KB

Thanks in advance for your help.

2

I suspect that the problem is exactly what the error mentions, that the driver versions are pretty old. Check the nvidia docs, the minimum requirements for cuda 10.1 is atleast 418. And the message shows your version is 390.

3

Thank you so much, I will update it.