Skip to content
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

loss mask aware final layer applicaiton #7275

Merged
merged 14 commits into from
Aug 22, 2023
Merged

Conversation

arendu
Copy link
Collaborator

@arendu arendu commented Aug 18, 2023

What does this PR do ?

save memory during training by loss mask selection for final output layer + softmax.

Collection: [NLP]

Changelog

  • Add specific line by line info of high level changes in this PR.

Usage

  • You can potentially add a usage example below
# Add a code snippet demonstrating how to use this 

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?
  • Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
    • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

  • New Feature
  • Bugfix
  • Documentation

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

  • Related to # (issue)

@github-actions github-actions bot added the NLP label Aug 18, 2023
self.tokenizer.ids_to_text(t[l.item() :][: data_cfg.get('tokens_to_generate')])
for t, l in zip(output['token_ids'], batch['context_lengths'])
]
if data_cfg.get("write_predictions_to_file", False) or data_cfg.metric.name != 'loss':
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a merge from another PR which is going to be merged soon.

Copy link
Collaborator

@stevehuang52 stevehuang52 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have tested the code locally and it worked. Thanks for the effort~!

@arendu arendu merged commit 6c06c82 into main Aug 22, 2023
15 checks passed
@arendu arendu deleted the adithyare/loss_mask_mem_save branch August 22, 2023 21:17
styagi130 pushed a commit to styagi130/NeMo that referenced this pull request Aug 23, 2023
* loss mask for final output and softmax

Signed-off-by: arendu <[email protected]>

* bs2 working

Signed-off-by: arendu <[email protected]>

* Fix skip generation

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add metric condition

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* encoder_input is none check

Signed-off-by: arendu <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Signed-off-by: arendu <[email protected]>
Signed-off-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Signed-off-by: Siddharth Tyagi <[email protected]>

Signed-off-by: Siddharth Tyagi <[email protected]>
dorotat-nv pushed a commit to dorotat-nv/NeMo that referenced this pull request Aug 24, 2023
* loss mask for final output and softmax

Signed-off-by: arendu <[email protected]>

* bs2 working

Signed-off-by: arendu <[email protected]>

* Fix skip generation

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add metric condition

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* encoder_input is none check

Signed-off-by: arendu <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Signed-off-by: arendu <[email protected]>
Signed-off-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Signed-off-by: dorotat <[email protected]>
rohitrango pushed a commit to rohitrango/NeMo that referenced this pull request Jun 25, 2024
* loss mask for final output and softmax

Signed-off-by: arendu <[email protected]>

* bs2 working

Signed-off-by: arendu <[email protected]>

* Fix skip generation

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Add metric condition

Signed-off-by: Cheng-Ping Hsieh <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* encoder_input is none check

Signed-off-by: arendu <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Signed-off-by: arendu <[email protected]>
Signed-off-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: Cheng-Ping Hsieh <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants