prabhuomkar / prabhuomkar/pytorch-cpp
[feature] pytorch DDP with C++ (utilizing CUDA / NCCL)
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 2.1k
- Forks
- 273
- Avg merge
- 1h 53m
- Merged PRs (30d)
- 1
Description
Is your feature request related to a problem? Please describe.
In my experience the same code (for training) is faster with pure C++ than python. My workflows utilize distributed compute for training, therefore such a solution would be awesome.
Describe the solution you'd like
A similar software like torch.distributed but for C++
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no files, tests, or entry points. Start by surveying the repository for its existing C++ PyTorch structure and any distributed-training support, then define the scope of a C++ equivalent to torch.distributed using CUDA/NCCL. Done should mean a documented, working C++ distributed-training solution with a clear validation path.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, pytorch
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100