Spring/spring-ai
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update Elasticsearch Store: Use KNN insettad of script_score

Browse Source 
- knn instead of script_score, removed initialization
 - only using normalized similarities, adjusted unit test
 - making l2norm's distances consistent with others
 - update dependency version and docs
 - upate autoconfigure ITs
This commit is contained in:
Laura Trotta
2024-04-12 17:42:13 +02:00
committed by Christian Tzolov
parent c9dd336ea3
commit 78f3797f8d
11 changed files with 184 additions and 203 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
spring-ai-docs/src/main/antora/modules/ROOT/pages/api/vectordbs/elasticsearch.adoc
View File

@@ -134,11 +134,18 @@ Properties starting with the `spring.ai.vectorstore.elasticsearch.*` prefix are
|`spring.ai.vectorstore.elasticsearch.index-name` | The name of the index to store the vectors. | spring-ai-document-index
|`spring.ai.vectorstore.elasticsearch.dimensions` | The number of dimensions in the vector. | 1536
|`spring.ai.vectorstore.elasticsearch.dense-vector-indexing` | Whether to use dense vector indexing. | true
|`spring.ai.vectorstore.elasticsearch.similarity` | The similarity function to use. | `cosine`
|`spring.ai.vectorstore.elasticsearch.initialize-schema`| whether to initialize the required schema | `false`
|===
The following similarity functions are available:
* cosine
* l2_norm
* dot_product
More details about each in the https://www.elastic.co/guide/en/elasticsearch/reference/master/dense-vector.html#dense-vector-params[Elasticsearch Documentation] on dense vectors.
== Metadata Filtering
You can leverage the generic, portable xref:api/vectordbs.adoc#metadata-filters[metadata filters] with Elasticsearch as well.
@@ -214,10 +221,11 @@ Read the link:https://www.elastic.co/guide/en/elasticsearch/client/java-api-clie
----
@Bean
public RestClient restClient() {
RestClientBuilder builder = RestClient.builder(new HttpHost("<host>", 9200, "http"));
Header[] defaultHeaders = new Header[] { new BasicHeader("Authorization", "Basic <encoded username and password>") };
builder.setDefaultHeaders(defaultHeaders);
return builder.build();
RestClient.builder(new HttpHost("<host>", 9200, "http"))
.setDefaultHeaders(new Header[]{
new BasicHeader("Authorization", "Basic <encoded username and password>")
})
.build();
}
----