Open pprp opened 4 years ago
I also get this error. Any solution?
same error . Have you solve it ? python3.6 pytorch0.4.1 cuda10.0
I haven't solved the problem yet.😧
Refer to the following link https://github.com/xingyizhou/CenterNet/issues/184 But it cannot solved the problem。
For now, the workaround is setting "torch.backends.cudnn.enabled = False". However, this would lower the running performance.
@LiberiFatali, in which file should we change that setting?
@LiberiFatali, in which file should we change that setting?
You can set that right after "import torch"
Try “torch.backends.cudnn.deterministic = True” rather than "torch.backends.cudnn.enabled = False"
Thank you @lxfhfut I have set : import torch torch.backends.cudnn.deterministic = True in the demo_centernet_deepsort.py and run successfully.
Thank you @lxfhfut I have set : import torch torch.backends.cudnn.deterministic = True in the demo_centernet_deepsort.py and run successfully.
CenterNet/src/lib/detectors/ctdet.py
Thank u for your excellent work! I encounter a problem as shown below.
The error is out of memory but I use
nvidia-smi
to check the memory and find the memory is enough.GPU: gtx 2080Ti 11G