-
Type:
Improvement
-
Resolution: Unresolved
-
Priority:
Unknown
-
None
-
Affects Version/s: None
-
Component/s: AI/ML
-
None
Context
We currently have two separate methods to generate vector search aggregation pipelines for use in our AI frameworks: vector_search_stage and autoembedding_vector_search_stage. Both generate similar pipelines, the only difference being that one is for auto-embedded vector search indexes and one is for externally-embedded indexes. That difference is not large enough to require two separate methods.
Definition of done
Combine vector_search_stage and autoembedding_vector_search_stage into a single updated vector_search_stage function that serves both types of search pipeline.
Pitfalls
All existing usages of both functions must continue to work unchanged.
- is blocked by
-
INTPYTHON-909 Migrate LangChain retrievers to pymongo-search-utils
-
- Blocked
-