Issue of running predict.py on cpu only machine

[huyu11]

Hi Team,,

Thanks for development this nice tool. I keep getting this following error when running the example data.

predict.py 355 <module>
main(args)

predict.py 273 main
model = EnsembleMunisModel(checkpoints)

model.py 248 __init__
model = MunisModel.load_from_checkpoint(checkpoint)

module.py 1531 load_from_checkpoint
loaded = _load_from_checkpoint(

saving.py 60 _load_from_checkpoint
checkpoint = pl_load(checkpoint_path, map_location=map_location)

cloud_io.py 51 _load
return torch.load(f, map_location=map_location)  # type: ignore[arg-type]

serialization.py 1462 load
return _load(

serialization.py 1964 _load
result = unpickler.load()

_weights_only_unpickler.py 512 load
self.append(self.persistent_load(pid))

serialization.py 1928 persistent_load
typed_storage = load_tensor(

serialization.py 1900 load_tensor
wrap_storage=restore_location(storage, location),

serialization.py 693 default_restore_location
result = fn(storage, location)

serialization.py 631 _deserialize
device = _validate_device(location, backend_name)

serialization.py 600 _validate_device
raise RuntimeError(

RuntimeError:
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.

I understand it is because of the cpu only machine. However, could you recommend how to solve this issue?

Thanks,
Yu