This repository has been archived by the owner on Nov 22, 2022. It is now read-only.
Dimension mismatch after setting max sequence length #1154
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
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D18566684 |
…h#1154) Summary: Pull Request resolved: facebookresearch#1154 TokenTensorizer and ByteTokenTensorizer has difference way of handling max sequence length. Usually this won't cause any problem if the model does not use the two tensorizers to process inputs and targets. But for the smart keyboard, it uses TokenTensorizer to process labels and ByteTokenTensorizer to process text inputs. This will cause dimension mismatch if the sentence is longer than the max sequence length. ``` TokenTokenizer: len(<EOS> + tokens + <BOS>) <= max sequence length ByteTokenTensorizer: len(tokens) <= max sequence length ``` This diff is to change the way ByteTokenTensorizer truncate text to max sequence length the same as TokenTokenize. Reviewed By: psuzhanhy Differential Revision: D18566684 fbshipit-source-id: c5f6e0668c383bdc8eec1cf108466b356cd9adcb
FanW123
force-pushed
the
export-D18566684
branch
from
69bfb1e to
e097852
Compare
|
This pull request was exported from Phabricator. Differential Revision: D18566684 |
|
This pull request has been merged in 98e6761. |
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Outdated suggestions cannot be applied.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:
TokenTensorizer and ByteTokenTensorizer has difference way of handling max sequence length. Usually this won't cause any problem if the model does not use the two tensorizers to process inputs and targets.
But for the smart keyboard, it uses TokenTensorizer to process labels and ByteTokenTensorizer to process text inputs. This will cause dimension mismatch if the sentence is longer than the max sequence length.
This diff is to change the way ByteTokenTensorizer truncate text to max sequence length the same as TokenTokenize.
Reviewed By: psuzhanhy
Differential Revision: D18566684