How to set CUDA location?

[frenzybiscuit]

Which version of recognize are you using?

8.1.1

Enabled Modes

Object recognition, Face recognition, Video recognition, Music recognition

TensorFlow mode

GPU mode

Downstream App

Files App

Which Nextcloud version do you have installed?

30.0.2

Which Operating system do you have installed?

Fedora 41

Which database are you running Nextcloud on?

Postgresql 16

Which Docker container are you using to run Nextcloud? (if applicable)

None

How much RAM does your server have?

128GB

What processor Architecture does your CPU have?

x86_64

Describe the Bug

I am trying to use GPU mode with a Nvidia 2080 GPU. the install is bare metal. Everything appears to be green/work up until CUDA.

I am using fedora with selinux in permissive mode.

Current error:

Successfully loaded libtensorflow in Node.js, but couldn't load GPU. Make sure CUDA Toolkit and cuDNN are installed and accessible, or turn off GPU mode.

When using CUDA, I have to specify the path location in .bashrc for my regular users. I'm assuming this is the issue with Nextcloud/Recognize. If I do not specify this in the .bashrc cuda refuses to work/compile.

The path should be:

export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

Is there a way to set this in the recognize app?

Expected Behavior

works as expected

To Reproduce

N/A

Debug log

N/A