Open Chuncheuk opened 2 years ago
The pre-packaged full installer should work out of box.
If you are running from source code, make sure you install the correct pytorch version, as mentioned in the doc
pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html
Check if your onnxruntime is gpu version?
The graphics card is1660ti,CUDA version is 11.3 and has correct output nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2021 NVIDIA Corporation Built on Sun_Mar_21_19:24:09_Pacific_Daylight_Time_2021 Cuda compilation tools, release 11.3, V11.3.58 Build cuda_11.3.r11.3/compiler.29745058_0
After the GPU conversion is checked, the GPU occupation of Python is always 0. I want to ask why and how to make it operate on the GPU