i have problem

[wangyuxuan-123]

when i run this code,it show Attempting to deserialize object on CUDA device 1 but torch.cuda.device_count() is 1. Please use torch.load with map_location to map your storages to an existing device.Do you know how to solve this ?

[wangyuxuan-123]

i write map_location='cuda:0',it doesn't work.

[EloyRoura]

I had a similar issue, first I modified the serilization.py class to hardcoding the cuda device I wanted to use.
Now, the workarround I found was to add map_location={'cuda:1': 'cuda:0'} when loading