[feature] Change the docker base image to nvidia/cuda, add docker-compose.yml

[LucienShui]

Hi @utkuozdemir, thanks for your great work!

I noticed #110 #112 talking about how docker accessing the gpu, I thought using nvidia/cuda as the base image might be a good choice for less configuration to run in the docker, to I made this PR.

I added a file named docker-compose.yml and updated the document so that everyone can make everything fine and easy.

And also remote accessing from arm64 like rapberrypi-3b or later would still work because nvidia/cuda support arm64. (but not tested because I don't have such a device)

Feel free if there is any question or something wrong. :)

Best regards!

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• I have read the CONTRIBUTING document.
• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have added tests to cover my changes.
• All new and existing tests passed.

[utkuozdemir]

Thanks for the contribution - Just I am not sure if cuda's license allows us to use it as a base image - Nvidia's licenses are tricky. Need to look into that. If you have any findings, I'd appreciate if you share them here.

[felixsmli]

The plain Ubuntu 24.04 image can access nvidia-smi if you have installed NVIDIA Container Toolkit (which is the suggested way for using Docker and NVIDIA GPUs), as shown in this simple example found from NVIDIA's documentation

sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi

So I guess you can avoid nvidia/cuda and just update the example to:

 docker run -d -p 9835:9835 --runtime=nvidia --gpus all utkuozdemir/nvidia_gpu_exporter:1.2.0

Which is much simpler. But of course this won't work with machines that have Docker but cannot install the NVIDIA Container Toolkit.