Fix multivalidation - #3388
Merged
Merged
Conversation
Signed-off-by: rprenger <rprenger@nvidia.com>
… into fix_multivalidation Signed-off-by: rprenger <rprenger@nvidia.com>
RPrenger
added
the
Expert Review
Member
|
/ok to test 2c4221b |
asolergi-nv
reviewed
Mar 3, 2026
| if split == Split.valid and args.full_validation: | ||
| batch_sampler = MegatronPretrainingSampler( | ||
| # Use specialized sampler for full validation that handles small datasets | ||
| batch_sampler = MegatronFullValidationSampler( |
Contributor
There was a problem hiding this comment.
Should we only use MegatronFullValidationSampler when required? (When using small datasets & full_validation)
Otherwise keep MegatronPretrainingSampler
Note we can access len(dataset) & DP size
Contributor
Author
There was a problem hiding this comment.
No, we need it whenever we're doing full_validation. The wording of the comment is a little confusing though. Should I change it?
Phlip79
removed
the
Expert Review
Contributor
Author
|
/ok to test 66028ab |
…erators The previous change ran the cross-DP all-reduce of eval_iters unconditionally and dropped the per-dataloader _get_iterator calls, which (1) failed with `TypeError: object of type 'list_iterator' has no len()` when dataloader_type="external" and (2) left valid_data_iterators set to None for non-full-validation runs. Gate the all-reduce on args.full_validation and restore the _get_iterator calls for both multiple_validation_sets and single-set paths, using "cyclic" iteration only under full_validation so short DP ranks loop up to the MAX-reduced eval_iters.
Contributor
Author
|
/ok to test 0fcfa93 |
…alidation data loader
Contributor
Author
|
/ok to test b4855d3 |
Contributor
Author
|
/ok to test 72d6cf1 |
asolergi-nv
approved these changes
Apr 23, 2026
jaredcasper
approved these changes
Apr 24, 2026
Contributor
|
🔄 Merge queue validation started! You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/24910443901 |
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Apr 24, 2026
Contributor
|
🔄 Merge queue validation started! You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/24938930211 |
yangbofun
pushed a commit
to xlm-research/Megatron-LM
that referenced
this pull request
May 22, 2026
Signed-off-by: rprenger <rprenger@nvidia.com>
yhgalaxy
pushed a commit
to yhgalaxy/Megatron-LM
that referenced
this pull request
Jun 17, 2026
Signed-off-by: rprenger <rprenger@nvidia.com> Signed-off-by: yhgalaxy <yhgalaxy@outlook.com>
jon-barker
pushed a commit
to jon-barker/Megatron-LM
that referenced
this pull request
Jul 10, 2026
Signed-off-by: rprenger <rprenger@nvidia.com> Signed-off-by: Jon Barker <jbarker@aws-cmh-slurm-1-vscode-02.cm.cluster>
terminator123
pushed a commit
to 021ai/Megatron-LM
that referenced
this pull request
Aug 3, 2026
Signed-off-by: rprenger <rprenger@nvidia.com>
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
5 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.
What does this PR do ?
This fixes a bug where using multiple validation sets hangs training if the validation sets are small compared to data parallel size.
Contribution process
flowchart LR A[Pre-checks] --> B[PR Tests] subgraph Code Review/Approval C1[Expert Review] --> C2[Final Review] end B --> C1 C2 --> D[Merge]Pre-checks
Core 0.8)Code review
The following process is enforced via the CODEOWNERS file for changes into
megatron/core. For changes outside ofmegatron/core, it is up to the PR author whether or not to tag the Final Reviewer team.For MRs into `main` branch
Feel free to message or comment the @mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!
(Step 1): Add PR label
Expert Review(Step 2): Collect the expert reviewers reviews
Expert Reviewlabel when your PR is ready for review.Final Review might get declined if these requirements are not fulfilled.
(Step 3): Final Review
Final Reviewlabel(Optional Step 4): Cherry-pick into release branch
If this PR also needs to be merged into
core_r*release branches, after this PR has been merged, selectCherry-pickto open a new PR into the release branch.For MRs into `dev` branch
The proposed review process for `dev` branch is under active discussion.MRs are mergable after one approval by either
eharper@nvidia.comorzijiey@nvidia.com.Merging your PR
Any member of core-adlr and
core-nemowill be able to merge your PR.