Have some problem with cuda

torch.package / torch::deploy
1

I am running cuda with pytorch 1.12 and gpu is not configured . i installed pytorch with cuda driver

conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.6 -c pytorch -c conda-forge

image
image698×252 15.8 KB

2

Based on your output I guess you’ve installed the CPU-only binary as it works for me:

conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.6 -c pytorch -c conda-forge
...
The following NEW packages will be INSTALLED:
...
  pytorch            pytorch/linux-64::pytorch-1.12.1-py3.8_cuda11.6_cudnn8.3.2_0
  pytorch-mutex      pytorch/noarch::pytorch-mutex-1.0-cuda
  torchaudio         pytorch/linux-64::torchaudio-0.12.1-py38_cu116
  torchvision        pytorch/linux-64::torchvision-0.13.1-py38_cu116

Output:

>>> import torch
>>> torch.__version__
'1.12.1'
>>> torch.cuda.is_available()
True
>>> torch.cuda.get_arch_list()
['sm_37', 'sm_50', 'sm_60', 'sm_61', 'sm_70', 'sm_75', 'sm_80', 'sm_86', 'compute_37']

So double check your install log and make sure the right version is picked.

3

what is the difference between your code and mine

image
image1220×198 19.4 KB

I think that I installed cuda version