What is proper version of torchvision for torch 2.0.0+nv23.5

I want test GPU is correctly work on pytorch
so i try run yolov5
but it dosen’t work

it said
‘RuntimeError: Couldn’t load custom C++ ops. This can happen if your PyTorch and torchvision versions are incompatible, or if you had errors while compiling torchvision from source. For further information on the compatible versions, check GitHub - pytorch/vision: Datasets, Transforms and Models specific to Computer Vision for the compatibility matrix. Please check your PyTorch version with torch.version and your torchvision version with torchvision.version and verify if they are compatible, and if not please reinstall torchvision so that it matches your PyTorch install.’

i try torchvision 0.15.2, 0.15.1, 0.15.0 because that link(git hub) say torch 2.0.0 is use 0.15
what is proper version for torch 2.0.0+nv23.5
or i reinstall torch older version?


Could you try v0.14.1?
Or you can use our PyTorch container which also uses TorchVision v0.14.1+ PyTorch v2.0.0.


I use python 3.8.10
jetpack 5.1.1[L4T 35.3.1]
visual studio code extension jupyter & python
According to pip list
now i’m using torch 2.0.0+nv23.5 and torchvision 0.14.1

but it still not working

i use script ‘pip install torchvision==0.14.1’ for install torchvision something wrong my script? or my environments?

@yys9905a that way will install torchvision without CUDA support, please uninstall it first and then see the Installation section of this post:

Or as Aasta suggested, you can run l4t-pytorch container if you are still having problems.

just follow Installation section of that post
It works
problem was using pip install…

Thank you!

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.