Add support for distributed Metrics Visualization
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 63
- Forks
- 8
- Avg merge
- 1h 57m
- Merged PRs (30d)
- 13
Description
The README includes a sample visualization which showcases live loss metrics for a given pytorch worker on ray.
Implement a more general interface for the live publishing of pytorch metrics on raydar, which can help users understand the quality/progress of their distributed ML workflows on Ray in real time. A wishlist of supported features might include:
- Real time loss metrics as distributed training is happening
- Real time custom metrics per-model as distributed training is happening
- Support from within the pytorch api on ray
Contributor guide
No contributing guide indexed for this repository
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
Start with the README sample visualization for live loss metrics from a PyTorch worker on Ray. Clarify the interface needed for live loss and custom per-model metrics, including how it should work from the PyTorch API on Ray. Done means distributed ML workflows can publish and view those metrics in real time.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- distributed-systems, machine-learning, observability
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100