Installing locally and following instructions exactly with virtual envirnoment

[anumerico]

Still getting errors, thats the current one:

RuntimeError: TemporalEncoderDecoder: TemporalViTEncoder: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

[CarlosGomes98]

Indeed, for running on CPU I merged a new PR that should fix this. You should be able to pull again from main and this should be fixed.

[anumerico]

Im installing NVIDIA drivers, CUDA and so on to see if it fixes the issue..
As far as I have gotten:

RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

[anumerico]

Cool I saw what you did @CarlosGomes98 , I did git pull and hopefully it works with non CUDA installations, now the issue is I managed to get CUDA but is an old GPU that is not supported:
Found GPU0 NVIDIA GeForce **** which is of cuda capability 3.5.
PyTorch no longer supports this GPU because it is too old.
The minimum cuda capability supported by this library is 3.7.
How can I disable GPU to test your new CPU only code?

[anumerico]

That sorted the issue:
https://stackoverflow.com/questions/53266350/how-to-tell-pytorch-to-not-use-the-gpu

Just tested your new code for CPU only cases, it works.
And if the case is deprecated GPU or not supported just run this command in the shell before running the training command:

export CUDA_VISIBLE_DEVICES=""