PyTorch and torchvision versions are incompatible

[Donalcon]

🐛 Describe the bug

My project has a dependency that requires torch==1.13.0 (bytetracker). I have installed this version, as well as the versions of torchvision and torch audio compatible with it:

pip install torch==1.13.0+cpu torchvision==0.14.0+cpu torchaudio==0.13.0 --extra-index-url https://download.pytorch.org/whl/cpu

The above is backed up by the compatibility matrix in this repo's readme:

torch | torchvision | Python
-- | -- | --
main / nightly | main / nightly | >=3.8, <=3.11
2.0 | 0.15 | >=3.8, <=3.11
1.13 | 0.14 | >=3.7.2, <=3.10
1.12 | 0.13 | >=3.7, <=3.10

Why am I still receiving a compatibility error?

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 https://github.com/pytorch/vision#installation 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.

Note: I don't have cuda so cpu version is what I'm after

Versions

above^^

[pmeier]

Could you please run the environment collection script as prompted by the issue template?

[NicolasHug]

Do you just need torch or do you also need torchvision? Feel free to ignore both torchvision and torchaudio from the installation command if you just need torch.