mlcommons / training_policies

Issues related to MLPerf™ training policies, including rules and suggested changes
https://mlcommons.org/en/groups/training
Apache License 2.0
93 stars 66 forks source link

Improve reference contribution doc #532

Closed ShriyaPalsamudram closed 9 months ago

ShriyaPalsamudram commented 10 months ago

Add sections to explain how to choose target accuracy, eval frequency, number of submission runs and some things to keep in mind while generating RCPs for the first time during reference development.

Use learnings from previously added benchmarks to help make similar decisions in the future

github-actions[bot] commented 10 months ago

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅