Interface INeuralQuery

A neural query.

Inherited Members
Namespace: OpenSearch.Client
Assembly: OpenSearch.Client.dll
Syntax
public interface INeuralQuery : IFieldNameQuery, IQuery

Properties

| Edit this page View Source

ExpandNestedDocs

When true, retrieves scores for all nested field documents within each parent document. Used with nested queries.

Declaration
[DataMember(Name = "expand_nested_docs")]
bool? ExpandNestedDocs { get; set; }
Property Value
Type Description
bool?
| Edit this page View Source

Filter

A query used to reduce the number of documents considered.

Declaration
[DataMember(Name = "filter")]
IQueryContainer Filter { get; set; }
Property Value
Type Description
IQueryContainer
| Edit this page View Source

K

The number of results the k-NN search returns. Only one of K, MaxDistance, or MinScore can be specified; if none is set, the server default is k = 10.

Declaration
[DataMember(Name = "k")]
int? K { get; set; }
Property Value
Type Description
int?
| Edit this page View Source

MaxDistance

The maximum physical vector space distance required in order for a neighbor to be considered a hit (radial search). Only one of K, MaxDistance, or MinScore can be specified.

Declaration
[DataMember(Name = "max_distance")]
float? MaxDistance { get; set; }
Property Value
Type Description
float?
| Edit this page View Source

MethodParameters

Additional engine-specific parameters that control the approximate k-NN search at query time, such as ef_search (HNSW) or nprobes (IVF).

Declaration
[DataMember(Name = "method_parameters")]
IDictionary<string, object> MethodParameters { get; set; }
Property Value
Type Description
IDictionary<string, object>
| Edit this page View Source

MinScore

The minimum similarity score required in order for a neighbor to be considered a hit (radial search). Only one of K, MaxDistance, or MinScore can be specified.

Declaration
[DataMember(Name = "min_score")]
float? MinScore { get; set; }
Property Value
Type Description
float?
| Edit this page View Source

ModelId

The ID of the model that will be used in the embedding interface. The model must be indexed in OpenSearch before it can be used in Neural Search.

Declaration
[DataMember(Name = "model_id")]
string ModelId { get; set; }
Property Value
Type Description
string
| Edit this page View Source

QueryImage

A Base64-encoded string that corresponds to the query image from which to generate vector embeddings. You must specify at least one of QueryText or QueryImage.

Declaration
[DataMember(Name = "query_image")]
string QueryImage { get; set; }
Property Value
Type Description
string
| Edit this page View Source

QueryText

The query text from which to produce queries. You must specify at least one of QueryText or QueryImage.

Declaration
[DataMember(Name = "query_text")]
string QueryText { get; set; }
Property Value
Type Description
string
| Edit this page View Source

QueryTokens

A raw sparse vector, expressed as tokens and their weights. Used as an alternative to QueryText for direct vector input.

Declaration
[DataMember(Name = "query_tokens")]
IDictionary<string, float> QueryTokens { get; set; }
Property Value
Type Description
IDictionary<string, float>
| Edit this page View Source

Rescore

Controls the rescoring of approximate k-NN search results using full-precision vectors. Pass a bool to enable or disable rescoring, or a KnnQueryRescoreContext to configure it.

Declaration
[DataMember(Name = "rescore")]
Union<bool, KnnQueryRescoreContext> Rescore { get; set; }
Property Value
Type Description
Union<bool, KnnQueryRescoreContext>
| Edit this page View Source

SemanticFieldSearchAnalyzer

An analyzer for tokenizing QueryText when using a sparse encoding model. Cannot be used together with ModelId.

Declaration
[DataMember(Name = "semantic_field_search_analyzer")]
string SemanticFieldSearchAnalyzer { get; set; }
Property Value
Type Description
string

Extension Methods