Make encode() in wrapped model compatible with super encode() #337
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
Contributor
markstur
commented
Mar 14, 2024
markstur
commented
- Adding missing params
- Don't return unexpected tuple (with token count) unless asked
- Adding check to not use our params if given an unwrapped model
- Fixing some param position things
* Adding missing params * Don't return unexpected tuple (with token count) unless asked * Adding check to not use our params if given an unwrapped model * Fixing some param position things Signed-off-by: Mark Sturdevant <markstur@Marks-MacBook-Pro.local>
markstur
requested review from
alex-jw-brooks,
gkumbhat,
evaline-ju,
gabe-l-hart,
tharapalanivel and
Ssukriti
as code owners
evaline-ju
reviewed
Mar 14, 2024
|
|
||
| # Else... | ||
| # It's possible to init with a model that doesn't have the added kwargs. | ||
| # E.g. a SentenceTransformer or other transormer model. Remove those kwargs! |
There was a problem hiding this comment.
Suggested change
| # E.g. a SentenceTransformer or other transormer model. Remove those kwargs! | |
| # E.g. a SentenceTransformer or other transformer model. Remove those kwargs! |
| :param truncate_input_tokens: Truncation length for input tokens. | ||
| Truncation length for input tokens. | ||
| If less than zero, this truncation is left up to the tokenizer default (model max). | ||
| If zero or greater than the model's maximum, then this is used as a test | ||
| to see if truncation is needed. If needed is needed, an exception is thrown. | ||
| Otherwise, we take this usable truncation limit to truncate the input tokens. | ||
| :param return_token_count: If true, a tuple is returned to add the input token count. | ||
|
|
||
| :return: | ||
| A tuple of the embedding, as a numpy matrix, and the input_token_count int. |
There was a problem hiding this comment.
could this return be updated to reflect the return_token_count update?
There was a problem hiding this comment.
almost missed this again, but done!
| # This is not the normal use case but at least don't pass invalid kwargs, to encode() | ||
| # and don't return the unexpected tuple (adding token count). | ||
| del kwargs["truncate_input_tokens"] | ||
| del kwargs["return_token_count"] |
There was a problem hiding this comment.
might be good to have a small test for this case, in case an update could break it (again)?
There was a problem hiding this comment.
Yes and I came here to make the PR a draft because this isn't catching keyerrors (noticed while reviewing with Flavia)
There was a problem hiding this comment.
Yep and done. Definitely was a silly miss to not check AND not test. :(
…tion behavior. * First draft could KeyError when deleting kwargs that don't exist. Tests added. * Adding a config option so the desired default behavior can be either: - Throw an error if truncation is happening implicitly, or - Nah. Just let it go. First was requested, so trunction does not happen quietly. This is common behavoir for some models. The second is more aligned with SentenceTransformers and is probably necessary for standard tests to run without errors. Signed-off-by: Mark Sturdevant <mark.sturdevant@ibm.com>
markstur
force-pushed
the
compatible_encode
branch
from
b97e34f
to
c6ed5b2
Compare
evaline-ju
approved these changes
Mar 19, 2024
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.