nvidia-docker VS Pytorch

Nvidia Docker on GitHub

Here's where I found discussion regarding this https://github.com/NVIDIA/nvidia-docker/issues/1671

I found the Faq, looks like Windows isn't supported which might indicate why I had this problem earlier. I might need to dual boot my machine if it won't work with WSL which I don't see mentioned in either page. WSL Cuda instructions found this I'll give it a try.

Arc has linux kernel support from 6.0, been using an A770 with tdarr for a few months. Super solid and no issues like the nvidia docker toolkit just losing the GPU. The workaround doesn't hold for long - https://github.com/NVIDIA/nvidia-docker/issues/1730 It is one of the reasons I went with Intel over waiting for a low end 40 series. The other was I that basically stole it for $199. So far Plex is the only thing that doesn't work with Arc and all the HW transcoding falls on the iGPU. Knowing how they prioritize things nobody wants, Arc support and AV1 transcoding will be added when 16th gen Intel CPUs are released.

Im trying to determine how to troubleshoot & resolve the HW transcoding, but based on my testing Im assuming its some change to the NVIDIA toolkit https://github.com/NVIDIA/nvidia-docker

NVIDIA Container Toolkit

So I am in the initial planning stages of setting up a new unraid server. Looking at picking up an SC846 24bay 4u chassis. I've got a Gigabyte Aorus mb with an AMD 5950x, 32gb of ddr4 (adding more as needed) and an nvidia 3070ti. I plan on getting an LSI 8i for the drives and leaves room for expansion server plans. My goal is to have plex setup via docker and utilize the gpu transcoding to offload the cpu work. I also want to setup vms or a vm server to essentially also have a "gaming server" mainly for me and the kids. This means down the road I would be adding another GPU to split up with other users. Im trying to allow for a max of 4 people while also still allowing plex to transcode as needed. Now I know there's other ways to do this but I dont feel like splitting this up into multiple systems unless I have to. So really just trying to see if this might be possible. My worry is that in order to make the gpu available to the plex docker I have to setup an nvidia container. https://github.com/NVIDIA/nvidia-docker

GraphModule compiles every instruction into low-level ATen operations.

8. TensorFlow and PyTorch: These frameworks support AI and machine learning integrations, allowing developers to build and deploy intelligent models and workflows. TensorFlow is widely used for deep learning applications, offering pre-trained models and extensive documentation. PyTorch provides flexibility and ease of use, making it ideal for research and experimentation. Both frameworks support neural network design, training, and deployment in production environments. Download TensorFlow here and Download PyTorch here.

Frameworks like TensorFlow and PyTorch can help you build and train models for various tasks, such as risk scoring, anomaly detection, and pattern recognition.

Python’s Growth in Data Work and AI: Python continues to lead because of its easy-to-read style and the huge number of libraries available for tasks from data work to artificial intelligence. Tools like TensorFlow and PyTorch make it a must-have. Whether you’re experienced or just starting, Python’s clear style makes it a good choice for diving into machine learning. Actionable Tip: If you’re new to Python, try projects that combine data with everyday problems. For example, build a simple recommendation system using Pandas and scikit-learn.

Basically this style of code - https://github.com/pytorch-labs/attention-gym/pull/84/files - has issues like this - https://github.com/pytorch/pytorch/pull/137452 https://github.com/pytorch/pytorch/issues/144511 https://github.com/pytorch/pytorch/issues/145869

For some higher level context, see https://pytorch.org/blog/flexattention/

The AI Service will be built using aiohttp (asynchronous Python web server) and integrates PyTorch, Hugging Face Transformers, numpy, pandas, and scikit-learn for financial data analysis.

Open source frameworks like PyTorch are already enabling Machine Learning breakthroughs because they’re living communities where great things happen through:

- Data Science and AI: TensorFlow, PyTorch and scikit-learn are only a few of the standard Python libraries. - Web Development: development of web-based applications is made simple by frameworks such as Flask as well as Django. - Prototyping: Python's ease of use lets you quickly iterate and testing concepts.

By chance, Tensorflow or PyTorch can work with pip packages from Nvidia.