Enable embedding caching on all vectorizers #320
Merged
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
tylerhutcherson
changed the title
abrookins
approved these changes
Apr 16, 2025
|
|
||
| try: | ||
| # Efficient batch cache lookup | ||
| cache_results = await self.cache.amget(texts=texts, model_name=self.model) |
There was a problem hiding this comment.
The whole amget / efficiency aspect of this work is nice. 🔥
|
|
||
| model: str | ||
| dtype: str = "float32" | ||
| dims: Optional[int] = None | ||
| dims: Annotated[Optional[int], Field(strict=True, gt=0)] = None | ||
| cache: Optional[EmbeddingsCache] = Field(default=None) |
There was a problem hiding this comment.
Passing a cache object in to get caching feels elegant
| async def _aembed(self, text: str, **kwargs) -> List[float]: | ||
| """Asynchronously generate a vector embedding for a single text. | ||
|
|
||
| Note: This implementation falls back to the synchronous version as |
There was a problem hiding this comment.
This could be worth logging when these methods are called.
tylerhutcherson
changed the title
abrookins
pushed a commit
that referenced
this pull request
May 1, 2025
Adds support to the `BaseVectorizer` class to have an optional `EmbeddingsCache` attached. - Refactored the subclass vectorizers to implement private embed methods and then let the base class handle the cache wrapper logic. - Fixed some circular imports. - Fixed async client handling in the cache subclasses (caught during testing). - Handle some typing checks and pydantic stuff related to private attrs and custom attrs. TODO in a separate PR: - Add embeddings caching to our testing suite (CI/CD speed up??) - Add embeddings caching to our SemanticRouter
abrookins
pushed a commit
that referenced
this pull request
May 1, 2025
Adds support to the `BaseVectorizer` class to have an optional `EmbeddingsCache` attached. - Refactored the subclass vectorizers to implement private embed methods and then let the base class handle the cache wrapper logic. - Fixed some circular imports. - Fixed async client handling in the cache subclasses (caught during testing). - Handle some typing checks and pydantic stuff related to private attrs and custom attrs. TODO in a separate PR: - Add embeddings caching to our testing suite (CI/CD speed up??) - Add embeddings caching to our SemanticRouter
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.
Adds support to the
BaseVectorizerclass to have an optionalEmbeddingsCacheattached.TODO in a separate PR: