cuda problem when tying to inference

[yalyzak]

File "inference.py", line 285, in
main(opt)
File "inference.py", line 273, in main
detect3d(
File "inference.py", line 71, in detect3d
regressor = regressor_factorymodel_select.cuda()
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\nn\modules\module.py", line 491, in cuda
return self._apply(lambda t: t.cuda(device))
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\nn\modules\module.py", line 387, in _apply
module._apply(fn)
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\nn\modules\module.py", line 387, in _apply
module._apply(fn)
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\nn\modules\module.py", line 409, in _apply
param_applied = fn(param)
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\nn\modules\module.py", line 491, in
return self.apply(lambda t: t.cuda(device))
File "C:\Users\yalyz\miniconda3\envs\yolo3d\lib\site-packages\torch\cuda_init.py", line 164, in _lazy_init
raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled
i tried it on a computer with and without cude and still faced the same issue

[TowerPro]

it seems the saved weight file(*.pkl) saves cuda device in the dict, you can ignore the device so can select device in other computers