RunTimeError

[ks322181]

Runtimerror: Attempting to deserialize objects on 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 storage to cpu

I think there is a problem with torch and torchvision version

Can anyone help me?
Thanks in Advance.

[yexing99]

I got the same error but fixed it by obj_detect_state_dict = torch.load(_config['tracktor']['obj_detect_model'],map_location=torch.device('cpu'))

[timmeinhardt]

This is indeed an issue of running the code on CPU and our models being trained on GPU. I just pushed a change which should resolve this issue and is similar to @yexing99 solution.

[timmeinhardt]

Can we close this issue?

[ks322181]

I am working on GPU system. Where i have cuda version 8. But still i am facing the same issue. Is it compatible with cuda 8. If not can you specific which cuda version it will support.

[timmeinhardt]

If you look at the requirements.txt you can see that our code currently runs with PyTorch 1.3 which in turn is compatible with CUDA 9.2 and 10.1.

[ks322181]

Thanks i will check with higher version of cuda.