# Installing PyTorch into a virtual Python environment

**URL:** https://discourse.nixos.org/t/installing-pytorch-into-a-virtual-python-environment/34720
**Category:** Uncategorized
**Created:** [October 28, 2023, 9:34am UTC](https://discourse.nixos.org/t/installing-pytorch-into-a-virtual-python-environment/34720 "2023-10-28T09:34:38Z")
**Posts on this page:** 1
**Showing post:** 2

<div class="post-metadata">

### Author: ![SergeK](https://discourse.nixos.org/user_avatar/discourse.nixos.org/sergek/32/2972_2.png) [@SergeK](https://discourse.nixos.org/u/SergeK)
#### Post date: [October 28, 2023, 10:19am UTC](https://discourse.nixos.org/t/installing-pytorch-into-a-virtual-python-environment/34720/2 "2023-10-28T10:19:35Z")

</div>

> `.venv/bin/pip install numpy torch`  
> `.venv/bin/python -c "import torch; print(torch.cuda.is_available())"`  
> `False`

PyTorch from PyPi doesn’t know where to find the cuda driver, you need to tell tell it explicitly e.g. like so: `LD_LIBRARY_PATH=/run/opengl-driver/lib python -c "import torch; ..."`

_This is because torch from PyPi is built specifically for FHS operating systems, i.e. systems that drop their libraries in the global `/usr/lib` and expose a global `/etc/ld.so.conf` file which tells them extra library search paths, including the location of the cuda driver (`libcuda.so`). If you want a better experience using prebuilt programs built with this assumption, you might also want to look into `programs.nix-ld.enable`._

> Interestingly the global one (the one that does work) doesn’t have any cuda things listed in its `Requires`.

This is because pytorch on pypi only recently started declaring as python dependencies at the dist-info level. They used to just copy cuda native libraries into their own wheels instead. For our source-build this is irrelevant, because we don’t event want to be using the python-packaged cuda, we want to link to our own `cudaPackages`, which has a bunch of patchelf hardening applied and is (or at least should be in the long run, we’re still not doing too great) split into independent outputs to reduce runtime closure sizes

---

_[View the full topic](https://discourse.nixos.org/t/installing-pytorch-into-a-virtual-python-environment/34720)._