Improve AnalyticDB Vector Store implementation without affecting user #6086
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
wangxuqi
force-pushed
the
analyticdb_v2
branch
4 times, most recently
from
adeac0a to
2d3226f
Compare
… behavior: 1) Streamline the __post_init__ behavior by creating a single table with automatic indexing. 2) Update the add_texts API for document insertion. 3) Optimize similarity_search_with_score_by_vector to retrieve results directly from the table. 4) Implement _similarity_search_with_relevance_scores. 5) Add an embedding_dimension parameter to support different dimension embedding functions. Users can continue using the API as before.
bradcrossen
added a commit
to bradcrossen/langchain
that referenced
this pull request
Jun 29, 2023
Support for SQLAlchemy 1.3 was removed in version 0.0.203 by change langchain-ai#6086. Re-adding support.
rlancemartin
added a commit
that referenced
this pull request
Jun 29, 2023
Support for SQLAlchemy 1.3 was removed in version 0.0.203 by change #6086. Re-adding support. - Description: Imports SQLAlchemy Row at class creation time instead of at init to support SQLAlchemy <1.4. This is the only breaking change and was introduced in version 0.0.203 #6086. A similar change was merged before: #4647 - Dependencies: Reduces SQLAlchemy dependency to > 1.3 - Tag maintainer: @rlancemartin, @eyurtsev, @hwchase17, @wangxuqi --------- Co-authored-by: rlm <[email protected]>
vowelparrot
pushed a commit
that referenced
this pull request
Jul 4, 2023
Support for SQLAlchemy 1.3 was removed in version 0.0.203 by change #6086. Re-adding support. - Description: Imports SQLAlchemy Row at class creation time instead of at init to support SQLAlchemy <1.4. This is the only breaking change and was introduced in version 0.0.203 #6086. A similar change was merged before: #4647 - Dependencies: Reduces SQLAlchemy dependency to > 1.3 - Tag maintainer: @rlancemartin, @eyurtsev, @hwchase17, @wangxuqi --------- Co-authored-by: rlm <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Hi there:
As I implement the AnalyticDB VectorStore use two table to store the document before. It seems just use one table is a better way. So this commit is try to improve AnalyticDB VectorStore implementation without affecting user behavior:
1. Streamline the
post_initbehavior by creating a single table with vector indexing.2. Update the
add_textsAPI for document insertion.3. Optimize
similarity_search_with_score_by_vectorto retrieve results directly from the table.4. Implement
_similarity_search_with_relevance_scores.5. Add
embedding_dimensionparameter to support different dimension embedding functions.Users can continue using the API as before.
Test cases added before is enough to meet this commit.