Unable use pytorch with cuda

Hey everyone,

I am new to PyTorch, any help will be greatly appreciated.

I am loading cuda, cudann and in the interactive node, I am trying torch.cuda.is_available()
and the output is False.

I am so confused, I haven’t done any GPU computing before. Hence, any help is greatly appreciated.

please ensure you request gpu resources: https://www.carc.usc.edu/user-information/user-guides/hpc-basics/running-jobs

or in short: add --gres=gpu:<gpu_type>:<number>
where <gpu_type> and <number> can be obtained with nodeinfo command

Usually <number> can be 1