Update sample_encoder.py #14547
Open
Update sample_encoder.py #14547
+5
−0
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
More than one initial tokens are mismatched between `tokens` and `answer_tokens` due to tokenizer's additional prefix space. This leads to additional warnings with labels not computed, inconsistence in data training. This is addressed by turning off the prefix space. Signed-off-by: Shiva Rama Krishna Parvatham <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What does this PR do ?
We have observed more than one initial tokens are mismatched between
tokensandanswer_tokensdue to tokenizer's additional prefix space.This leads to additional warnings with labels not computed, inconsistence in data preprocessing.
This is addressed by turning off the prefix space.
For Instance, consider the below
![]()
.......![]()
ASSISTANT: cute denim shift dress in blue acid wash
tokensdecoded:USER: What is in the photo?
answer_tokensdecoded:cute denim shift dress in blue acid wash
while
tokenshave [..........., '_cute', '_den', 'im', '_shift', '_dress',........]but
answer_tokenshave [ 'c', 'ute', '_den', 'im', '_shift', '_dress',......]which is a mismatch leads incomplete in compute labels
Collection: [multimodal]