NVIDIA-NeMo / NVIDIA-NeMo/DataDesigner
Update health-checks to account for multi-modality defined in the data designer workflow
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 2.2k
- Forks
- 211
- Avg merge
- 2d 6h
- Merged PRs (30d)
- 40
Description
Priority Level
Low (Nice to have)
Is your feature request related to a problem? Please describe.
- Nemotron parse, a vlm, requires images to be provided in all requests. Our health-checks fail for this model because it doesn't send images in the health-check request
Describe the solution you'd like
- If a workflow has column configs with multi-modality context, for example ImageContext defined, when doing health-checks we should add a dummy multi-modal context in the request.
Describe alternatives you've considered
Workaround right now is to skip health check added in https://github.com/NVIDIA-NeMo/DataDesigner/pull/244
Additional context
No response
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
Start by tracing the health-check request path and the data designer workflow's column configuration handling. Check how ImageContext or other multi-modality context is represented, then verify that health checks for Nemotron parse receive dummy image context while existing text-only checks remain unchanged.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, testing-qa
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100