error in train id ?

[Anislar]

File "train.py", line 40, in cc_trainer = Trainer(loading_data,cfg_data,pwd) File "C:\Users\xx\Desktop\C-3-Framework-python3.x\trainer.py", line 25, in init self.net = CrowdCounter(cfg.GPU_ID,self.net_name).cuda() File "C:\Users\xx\Desktop\C-3-Framework-python3.x\models\CC.py", line 29, in init self.CCN = torch.nn.DataParallel(self.CCN, device_ids=gpus).cuda() File "C:\Users\xx\Anaconda3.1\envs\opencv\lib\site-packages\torch\nn\parallel\data_parallel.py", line 133, in init _check_balance(self.device_ids) File "C:\Users\xx\Anaconda3.1\envs\opencv\lib\site-packages\torch\nn\parallel\data_parallel.py", line 19, in check_balance dev_props = [torch.cuda.get_device_properties(i) for i in device_ids] File "C:\Users\xx\Anaconda3.1\envs\opencv\lib\site-packages\torch\nn\parallel\data_parallel.py", line 19, in dev_props = [torch.cuda.get_device_properties(i) for i in device_ids] File "C:\Users\xx\Anaconda3.1\envs\opencv\lib\site-packages\torch\cuda_init.py", line 326, in get_device_properties raise AssertionError("Invalid device id") AssertionError: Invalid device id

what can'i do ? if i use SHHA

[BJTUJia]

It looks like trouble with your GPU setting, not the training ids. The simplest fix is to set your GPU id to 0.