This repository was archived by the owner on Nov 22, 2022. It is now read-only.
removed bilstm dependence on seq_lengths #776
Closed
Conversation
This file contains 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
facebook-github-bot
added
the
CLA Signed
Summary: Pull Request resolved: facebookresearch#776 If seq_lengths is not provided as a parameter to BiLSTM and pack_sequence is disabled, then the BiLSTM representation should not use seq_lengths to initialize anything. This diff removes the dependence on seq_lengths in said scenarios by using the batch dimension of embedded_tokens to initialize zero hidden state. This seemingly minor change makes a dramatic difference when exporting models to caffe2 because seq_lengths will no longer be a part of the model graph. This can give both latency and memory savings during inference, albeit small. Differential Revision: D16229675 fbshipit-source-id: 9b64bad0391934b2819b26c97ca145e53a6da238
shreydesai
force-pushed
the
export-D16229675
branch
from
c6e16f3 to
84f4fc1
Compare
|
This pull request has been merged in 8a1a455. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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:
If seq_lengths is not provided as a parameter to BiLSTM and pack_sequence is disabled, then the BiLSTM representation should not use seq_lengths to initialize anything. This diff removes the dependence on seq_lengths in said scenarios by using the batch dimension of embedded_tokens to initialize zero hidden state.
This seemingly minor change makes a dramatic difference when exporting models to caffe2 because seq_lengths will no longer be a part of the model graph. This can give both latency and memory savings during inference, albeit small.
Differential Revision: D16229675