search

jeffreyyihuang / two-stream-action-recognition

Using two stream architecture to implement a classic action recognition method on UCF101 dataset
MIT License
855 stars 252 forks source link

no CUDA-capable device is detected (only for motion network) #52

Closed duygusar closed 5 years ago

duygusar commented 5 years ago

I was able to run the spatial network, but when I try to run motion.network I get this error (which is super weird because obviously I run it on the same exact system/environment/terminal):

two-stream-action-recognition-master/venv/local/lib/python2.7/site-packages/torchvision/transforms/transforms.py:208: UserWarning: The use of the transforms.Scale transform is deprecated, please use transforms.Resize instead. "please use transforms.Resize instead.") ==> Training data : 573 videos (20, 224, 224) ==> Validation data : 4294 frames (20, 224, 224) ==> Build model and setup loss and optimizer THCudaCheck FAIL file=/pytorch/aten/src/THC/THCGeneral.cpp line=74 error=38 : no CUDA-capable device is detected Traceback (most recent call last): File "motion_cnn.py", line 264, in main() File "motion_cnn.py", line 67, in main model.run() File "motion_cnn.py", line 112, in run self.build_model() File "motion_cnn.py", line 86, in build_model self.model = resnet101(pretrained= True, channel=self.channel).cuda() File "/media/d/DATA_2/two-stream-action-recognition-master/venv/local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 258, in cuda return self._apply(lambda t: t.cuda(device)) File "/media/d/DATA_2/two-stream-action-recognition-master/venv/local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 185, in _apply module._apply(fn) File "/media/d/DATA_2/two-stream-action-recognition-master/venv/local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 191, in _apply param.data = fn(param.data) File "/media/d/DATA_2/two-stream-action-recognition-master/venv/local/lib/python2.7/site-packages/torch/nn/modules/module.py", line 258, in return self._apply(lambda t: t.cuda(device)) RuntimeError: cuda runtime error (38) : no CUDA-capable device is detected at /pytorch/aten/src/THC/THCGeneral.cpp:74

I have only one GPU so I don't understand why this is happening, I tried to set the GPU with CUDA_VISIBLE_DEVICES=0 but it didn't work (export CUDA_VISIBLE_DEVICES=0 didn't work either)

duygusar commented 5 years ago

I just noticed os.environ["CUDA_VISIBLE_DEVICES"] = "1" !! My bad!