When I run the training shell as the description in README, I get the following error that CUDA runtime is not found.
I checked that torch is installed and GPU is available (torch.cuda.is_available() returns True).
I wonder if anyone else gets the same error, or if there is something that I might be doing wrong?
lr: 1e-4, dropout: 0.05 , seed: 0,
No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'
The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `huggingface-cli` if you want to set the git credential as well.
Token is valid (permission: read).
...
Hi,
When I run the training shell as the description in README, I get the following error that CUDA runtime is not found. I checked that torch is installed and GPU is available (torch.cuda.is_available() returns True). I wonder if anyone else gets the same error, or if there is something that I might be doing wrong?
Thanks for your work!