mirror of
https://github.com/opensearch-project/neural-search.git
synced 2026-11-07 07:57:25 +00:00
Neural search transforms text into vectors and facilitates vector search both at ingestion time and at search time.
- Java 97.7%
- C++ 1.5%
- Shell 0.4%
- Python 0.3%
- CMake 0.1%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
9fb76e8693* Match the k-NN explanation by shape in explain ITs, Lucene 10.6 changed its wording
Lucene 10.6.0 (apache/lucene#16271) rewords DocAndScoreQuery.explain() for a
k-NN match from "within top N docs" to "Within top N doc(s)", and OpenSearch
core moved to 10.6.0 in opensearch-project/OpenSearch#23475. Since then the
scheduled main build fails HybridQueryExplainIT (two tests asserting the exact
old text) and HybridQueryFusedModeExplainIT (a contains("docs") check that
"doc(s)" does not satisfy).
The text belongs to Lucene, so assert the shape of that leaf instead: a k-NN
explanation within the top N candidates, in either wording. TestUtils gets the
pattern plus isKnnWithinTopDocsExplanation and assertKnnWithinTopDocsExplanation,
which still checks N. Both wordings are accepted so the same tests hold on 3.9,
which builds against Lucene 10.5.1.
Signed-off-by: Martin Gaievski <gaievski@amazon.com>
* Add CHANGELOG entry for #2077
Signed-off-by: Martin Gaievski <gaievski@amazon.com>
---------
Signed-off-by: Martin Gaievski <gaievski@amazon.com>
|
||
| .github | ||
| .idea | ||
| formatter | ||
| gradle/wrapper | ||
| jni | ||
| qa | ||
| release-notes | ||
| scripts | ||
| src | ||
| .gitignore | ||
| .gitmodules | ||
| .whitesource | ||
| ADMINS.md | ||
| build.gradle | ||
| CHANGELOG.md | ||
| CODE_OF_CONDUCT.md | ||
| codecov.yml | ||
| CONTRIBUTING.md | ||
| DEVELOPER_GUIDE.md | ||
| gradle.properties | ||
| gradlew | ||
| gradlew.bat | ||
| LICENSE | ||
| lombok.config | ||
| MAINTAINERS.md | ||
| NOTICE | ||
| README.md | ||
| RELEASING.md | ||
| repositories.gradle | ||
| SECURITY.md | ||
| settings.gradle | ||
| TRIAGING.md | ||
OpenSearch Neural Search
OpenSearch Neural Search is an OpenSearch plugin that adds dense neural retrieval into the OpenSearch ecosystem. The plugin provides the capability for indexing documents and doing neural search on the indexed documents.
Project Resources
- Project Website
- Downloads.
- Documentation
- Need help? Try Forums
- Project Principles
- Contributing to OpenSearch
- Maintainer Responsibilities
- Release Management
- Admin Responsibilities
- Security
- Code of Conduct
- License
- Copyright
Code of Conduct
This project has adopted the Amazon Open Source Code of Conduct. For more information see the Code of Conduct FAQ, or contact opensource-codeofconduct@amazon.com with any additional questions or comments.
License
This project is licensed under the Apache v2.0 License.
Copyright
Copyright OpenSearch Contributors. See NOTICE for details.