Make lowest precision, parallelism, MBS, and config_filename mandatory for v6.1 - #472
Merged
pavanky merged 2 commits intoAug 21, 2026
Conversation
Require the linear/attn/comm precision and TP/PP/CP/EP keys in every training 6.1 result log so the checker can enforce issue #878. Signed-off-by: ShriyaRishab <spalsamudram@nvidia.com>
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
4 tasks
These disclosure keys were still optional; make them EXACTLY_ONE so every result log identifies the micro-batch size and matching config. Signed-off-by: ShriyaRishab <spalsamudram@nvidia.com>
ShriyaRishab
changed the title
carualejandro
approved these changes
Aug 21, 2026
pavanky
approved these changes
Aug 21, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
lowest_numerical_precision_in_{linear,attn,comm},{tensor,pipeline,context,expert}_parallelism,micro_batch_size, andconfig_filenamerequired (EXACTLY_ONE) intraining_6.1.0/common.yamlso every Closed and Open result log must report them.1.Fixes mlcommons/training#878
Companion policy PR: mlcommons/training_policies#592
Test plan
training_6.0.0/common.yamlstill marks these keysOPTIONAL= 1, and a non-emptyconfig_filenameand confirm it passes these checks< 1, ormicro_batch_size < 1fails the valueCHECK