Update Gemma3VL model training scripts #15041
Merged
+423
−120
Conversation
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
genquan9
force-pushed
the
main
branch
2 times, most recently
from
1e2018f to
6475ad2
Compare
genquan9
force-pushed
the
main
branch
2 times, most recently
from
6f46cf8 to
07da129
Compare
hemildesai
previously approved these changes
Nov 14, 2025
Collaborator
hemildesai
left a comment
hemildesai
left a comment
There was a problem hiding this comment.
LGTM and the changes are all scoped to Gemma so shouldn't affect anything else.
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
genquan9
requested review from
chtruong814,
ko3n1g,
pablo-garay and
thomasdhc
as code owners
Signed-off-by: genquan9 <[email protected]>
Signed-off-by: genquan9 <[email protected]>
* Update MagpieTTS Signed-off-by: Jason <[email protected]> * allow None in dataset path Signed-off-by: Jason <[email protected]> * try to fix test by removing lhotse; fix yamls in fast dev run tests Signed-off-by: Jason <[email protected]> * increase zeroshot cer value; attempt to fix PO test; add back lhotse in parakeet inference to test segmentation fault Signed-off-by: Jason <[email protected]> * remove branch from test Signed-off-by: Jason <[email protected]> * use batch_size 1 Signed-off-by: Jason <[email protected]> * update GRPO test script Signed-off-by: Jason <[email protected]> * add use_lhotse as a param to transcribe; attempt to fix PO test again; attempt to catch error Signed-off-by: Jason <[email protected]> * fix tests Signed-off-by: Jason <[email protected]> * update rnnt transcribe; fix po test again Signed-off-by: Jason <[email protected]> * Apply suggestion from @XuesongYang Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Move FCD copyright text from TorchEval to top of file Signed-off-by: Fejgin, Roy <[email protected]> * Remove duplicate copyright text It is now at the top of the file. Signed-off-by: Fejgin, Roy <[email protected]> * Apply suggestion from @XuesongYang Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Apply suggestion from @XuesongYang Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Apply suggestion from @XuesongYang Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Apply suggestion from @XuesongYang Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> * Fix OnlinePO test: escape a special character in command line Signed-off-by: Fejgin, Roy <[email protected]> * Easier-to-read way to quote a special character in OnlinePO test Signed-off-by: Fejgin, Roy <[email protected]> * Work around ASR Lhotse issue ... and remove some debug code. Signed-off-by: Fejgin, Roy <[email protected]> * Remove FCD metric for now Signed-off-by: Fejgin, Roy <[email protected]> * Remove unused import Signed-off-by: Fejgin, Roy <[email protected]> * Update examples/tts/conf/magpietts/magpietts_lhotse.yaml Signed-off-by: Fejgin, Roy <[email protected]> Signed-off-by: Roy Fejgin <[email protected]> --------- Signed-off-by: Jason <[email protected]> Signed-off-by: Xuesong Yang <[email protected]> Signed-off-by: Fejgin, Roy <[email protected]> Signed-off-by: Roy Fejgin <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Co-authored-by: Fejgin, Roy <[email protected]> Signed-off-by: genquan9 <[email protected]>
…NeMo#15090) This reverts commit b557cfd. Signed-off-by: genquan9 <[email protected]>
…IA-NeMo#15091) * ASR Inference: load decoding params from config for RNN-T Signed-off-by: Vladimir Bataev <[email protected]> Signed-off-by: genquan9 <[email protected]>
Signed-off-by: genquan9 <[email protected]>
…NeMo#15090) This reverts commit b557cfd. Signed-off-by: genquan9 <[email protected]>
Signed-off-by: nithinraok <[email protected]> Signed-off-by: genquan9 <[email protected]>
Signed-off-by: Pablo Garay <[email protected]> Signed-off-by: genquan9 <[email protected]>
Signed-off-by: genquan9 <[email protected]>
Contributor
Author
|
i added missing headers for the new added files: 104d821 |
hemildesai
approved these changes
Nov 21, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
13 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.
Important
The
Update branchbutton must only be pressed in very rare occassions.An outdated branch is never blocking the merge of a PR.
Please reach out to the automation team before pressing that button.
What does this PR do ?
This PR is to fix Gemma3VL model training issues.
Collection: [Note which collection this PR will affect]
Changelog
Usage
# torchrun --nproc_per_node=1 ./scripts/vlm/gemma3vl_finetune.py --data_type=mockGitHub Actions CI
The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.
The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".
Before your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information