mlcommons / mlcommons/inference
Disambiguating execution modes
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.6k
- Forks
- 650
- Avg merge
- 1d 22h
- Merged PRs (30d)
- 6
Description
What are the difference between the test and valid modes list on the documentation?
Since this is intended to be a reproducible bechmark suite, would it be possible to add clear instructions on which modes are to be run and the impact each step has? For instance, I cannot tell if the test execution mode is optional or if it is generating a target metric (as indicated by "record the estimated offline_target_qps" on the documentation)
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 with the linked MLPerf Inference documentation and compare the test and valid execution-mode sections. Trace how each mode is described in the repository’s benchmark documentation or entry points, then update the documentation so required steps, optional steps, and the effect on offline_target_qps are explicit.
Written by the indexing model from the issue text.
Assessment
- Domain
- documentation, machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100