-
Notifications
You must be signed in to change notification settings - Fork 2.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[NeMo-UX] Use single instance of loss reductions in GPTModel #9801
Conversation
b66e35d
to
6306228
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Functionality works
TRAIN_LOSS_REDUCTION = MaskedTokenLossReduction() | ||
VALIDATION_LOSS_REDUCTION = MaskedTokenLossReduction(validation_step=True) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's comment that this is a temporary WAR.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Or remove the globals.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Updated.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM,
Alternatively we could decorate training_loss_reduction
& validation_loss_reduction
with @cache
, but I have no preference. Please proceed with the merge.
Thanks.
6d35351
to
537259b
Compare
Signed-off-by: Hemil Desai <[email protected]>
Signed-off-by: hemildesai <[email protected]>
Signed-off-by: Hemil Desai <[email protected]>
537259b
to
eded322
Compare
* Use single instance of loss reductions Signed-off-by: Hemil Desai <[email protected]> * Apply isort and black reformatting Signed-off-by: hemildesai <[email protected]> * Refactor Signed-off-by: Hemil Desai <[email protected]> --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: hemildesai <[email protected]>
* Use single instance of loss reductions Signed-off-by: Hemil Desai <[email protected]> * Apply isort and black reformatting Signed-off-by: hemildesai <[email protected]> * Refactor Signed-off-by: Hemil Desai <[email protected]> --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: hemildesai <[email protected]>
…9861) * Use single instance of loss reductions * Apply isort and black reformatting * Refactor --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: Hemil Desai <[email protected]> Co-authored-by: hemildesai <[email protected]>
…9801) (NVIDIA#9861) * Use single instance of loss reductions * Apply isort and black reformatting * Refactor --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: Hemil Desai <[email protected]> Co-authored-by: hemildesai <[email protected]> Signed-off-by: adityavavre <[email protected]>
…9861) * Use single instance of loss reductions * Apply isort and black reformatting * Refactor --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: Hemil Desai <[email protected]> Co-authored-by: hemildesai <[email protected]>
…9801) (NVIDIA#9861) * Use single instance of loss reductions * Apply isort and black reformatting * Refactor --------- Signed-off-by: Hemil Desai <[email protected]> Signed-off-by: hemildesai <[email protected]> Co-authored-by: Hemil Desai <[email protected]> Co-authored-by: hemildesai <[email protected]> Signed-off-by: Hainan Xu <[email protected]>
What does this PR do ?
Uses a singleton for
training_loss_reduction
andvalidation_loss_reduction
inGPTModel
.Collection: llm
Changelog
Usage
# Add a code snippet demonstrating how to use this
GitHub 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