search

lzx1413 / PytorchSSD

pytorch version of SSD and it's enhanced methods such as RFBSSD,FSSD and RefineDet
MIT License
708 stars 237 forks source link

运行train_test.py时AssertionError: Invalid device id #50

Open Eileen2014 opened 5 years ago

Eileen2014 commented 5 years ago

Traceback (most recent call last): File "train_test.py", line 183, in net = torch.nn.DataParallel(net, device_ids=list(range(args.ngpu))) File "/home/cv2018/anaconda3/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 102, in init _check_balance(self.device_ids) File "/home/cv2018/anaconda3/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 17, in _check_balance dev_props = [torch.cuda.get_device_properties(i) for i in device_ids] File "/home/cv2018/anaconda3/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 17, in dev_props = [torch.cuda.get_device_properties(i) for i in device_ids] File "/home/cv2018/anaconda3/lib/python3.6/site-packages/torch/cuda/init.py", line 292, in get_device_properties raise AssertionError("Invalid device id") AssertionError: Invalid device id

rw1995 commented 5 years ago

你把refinedet——train的关于gpu的default改一下，改成你的GPU——ID

yosunpeng commented 4 years ago

How much gpus did you want to use? Please note that argumentation, '--ngpu' in train_test.py. First, add 'CUDA_VISIBEL_DIVICES=x1' (x1 is the gpu id) before 'python'
Second, add '--ngpu x'(x is the amount of gpus) behind your running shell command
Here is an example:

CUDA_VISIBLE_DEVICES=0  python train_test.py -d VOC -v FSSD_vgg -s 300 --ngpu 1

Try it, hopes working for you.