apply output layer pruning #5426
Closed
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/5426
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 20d7986 with merge base ad95e46 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D62143905 |
iseeyuan
commented
Sep 17, 2024
|
This pull request was exported from Phabricator. Differential Revision: D62143905 |
iseeyuan
force-pushed
the
export-D62143905
branch
from
3695692 to
20ea3af
Compare
iseeyuan
pushed a commit
to iseeyuan/executorch-1
that referenced
this pull request
Sep 17, 2024
Summary: Pull Request resolved: pytorch#5426 Apply output layer pruning if we are using a model trained with a large output vocabulary to use as a classification task to output only smaller set of vocabulary. The output interface is ensured to be the same as unpruned model. e.g., if the last linear layer has 2048 x 128k shape, and we trained the model to output only 20 output vocab, then we can prune away the last layer to have a shape of 2048 x 20. But we still expand the 1,20 output shape to 1,128k so that the app consuming the model outputs don't need to change. Reviewed By: iseeyuan Differential Revision: D62143905
tarun292
approved these changes
Sep 17, 2024
|
This pull request was exported from Phabricator. Differential Revision: D62143905 |
iseeyuan
pushed a commit
to iseeyuan/executorch-1
that referenced
this pull request
Sep 17, 2024
Summary: Pull Request resolved: pytorch#5426 Apply output layer pruning if we are using a model trained with a large output vocabulary to use as a classification task to output only smaller set of vocabulary. The output interface is ensured to be the same as unpruned model. e.g., if the last linear layer has 2048 x 128k shape, and we trained the model to output only 20 output vocab, then we can prune away the last layer to have a shape of 2048 x 20. But we still expand the 1,20 output shape to 1,128k so that the app consuming the model outputs don't need to change. Reviewed By: tarun292, iseeyuan Differential Revision: D62143905
iseeyuan
force-pushed
the
export-D62143905
branch
from
20ea3af to
dd6f355
Compare
Summary: Pull Request resolved: pytorch#5426 Apply output layer pruning if we are using a model trained with a large output vocabulary to use as a classification task to output only smaller set of vocabulary. The output interface is ensured to be the same as unpruned model. e.g., if the last linear layer has 2048 x 128k shape, and we trained the model to output only 20 output vocab, then we can prune away the last layer to have a shape of 2048 x 20. But we still expand the 1,20 output shape to 1,128k so that the app consuming the model outputs don't need to change. Reviewed By: tarun292, iseeyuan Differential Revision: D62143905
|
This pull request was exported from Phabricator. Differential Revision: D62143905 |
iseeyuan
force-pushed
the
export-D62143905
branch
from
dd6f355 to
20d7986
Compare
|
This pull request has been merged in 2afcd96. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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:
Apply output layer pruning if we are using a model trained with a large output vocabulary to use as a classification task to output only smaller set of vocabulary. The output interface is ensured to be the same as unpruned model.
e.g., if the last linear layer has 2048 x 128k shape, and we trained the model to output only 20 output vocab, then we can prune away the last layer to have a shape of 2048 x 20. But we still expand the 1,20 output shape to 1,128k so that the app consuming the model outputs don't need to change.
Differential Revision: D62143905