Create PyTorch starter project and source repository

[michael-valdron]

Which area is this issue related to?

/kind task

/area registry

Issue Description

Create PyTorch starter project and source repository under https://github.com/devfile-samples

See example Tensorflow v2 starter project for reference.

Project Dependencies

• torch==2.3.0
• torchvision==0.18.0
• torchaudio==2.3.0

During both the runtime of the init command and the deployment container image build, everywhere this dependency is installed, it needs to pick one of the following index urls to use based on the chosen Compute Platform (CPU/GPU):

• CUDA 11.8 (Nvidia GPU) - https://download.pytorch.org/whl/cu118
• CUDA 12 (Nvidia GPU) - None
• CPU - https://download.pytorch.org/whl/cpu
• ROCm (AMD GPU) - https://download.pytorch.org/whl/rocm6.0

Acceptance Criteria

• Create starter project source repository
• Initialize PyTorch starter project that creates the MNIST model then saves and loads in
  • This lets someone override for loading in their own model to use
• Enable GPU if flag is set, more setup required in terms of how we install package dependencies

[github-actions]

This issue is stale because it has been open for 90 days with no activity. Remove stale label or comment or this will be closed in 60 days.

[github-actions]

This issue is stale because it has been open for 90 days with no activity. Remove stale label or comment or this will be closed in 120 days.