PosgreSQL hybrid search #958
Open
+141
−40
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
Include: - Text search index - Hybrid Search configuration - Vector or hybrid search
|
@microsoft-github-policy-service agree |
- Index name should be different per table. - Add missing filter
dluc
reviewed
Jan 8, 2025
dluc
reviewed
Jan 8, 2025
| @@ -175,7 +183,8 @@ public async Task CreateTableAsync( | |||
| {this._colContent} TEXT DEFAULT '' NOT NULL, | |||
| {this._colPayload} JSONB DEFAULT '{{}}'::JSONB NOT NULL | |||
| ); | |||
| CREATE INDEX IF NOT EXISTS idx_tags ON {tableName} USING GIN({this._colTags}); | |||
| CREATE INDEX IF NOT EXISTS {tableName}_idx_tags ON {tableName} USING GIN({this._colTags}); | |||
| CREATE INDEX IF NOT EXISTS {tableName}_idx_content ON {tableName} USING GIN(to_tsvector('english',{this._colContent})); | |||
There was a problem hiding this comment.
Should the language be configurable?
dluc
reviewed
Jan 8, 2025
dluc
reviewed
Jan 8, 2025
dluc
reviewed
Jan 8, 2025
dluc
reviewed
Jan 8, 2025
Comment on lines
+473
to
+475
| FROM {tableName}, plainto_tsquery('english', @query) query | ||
| WHERE {filterSqlHybridText} AND to_tsvector('english', {this._colContent}) @@ query | ||
| ORDER BY ts_rank_cd(to_tsvector('english', {this._colContent}), query) DESC |
There was a problem hiding this comment.
ditto: language should be configurable
dluc
reviewed
Jan 8, 2025
dluc
reviewed
Jan 8, 2025
dluc
requested changes
Jan 8, 2025
|
Looks like the PR got stale, with some unsolved errors and comments. We might have to archive it unless someone can kindly complete the task. |
I will resolve the comments this weekend. |
Add parametrization to text search language dictionary and parametrization of the Reciprocal Ranked Fusion "k-nearest neighbor" to score results of Hybrid Search
SignalRT
commented
Jan 30, 2025
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.
This PR Includes:
Motivation and Context (Why the change? What's the scenario?)
Vector search do not provide the best results in an important number of scenarios. Hybrid search provides better results.
High level description (Approach, Design)
This change includes a new parameter for activate hybrid search on PostgreSQL extension. This parameters defaults to the previous implementation (vector search).
The search will use vector search or hybrid search depending on the parameter.