Class CreateVectorSearchIndexModelBase<TDocument>
Defines common parts of a vector index model using strongly-typed C# APIs.
public abstract class CreateVectorSearchIndexModelBase<TDocument> : CreateSearchIndexModel
Type Parameters
TDocument
- Inheritance
-
CreateVectorSearchIndexModelBase<TDocument>
- Derived
- Inherited Members
Constructors
- CreateVectorSearchIndexModelBase(FieldDefinition<TDocument>, string, params FieldDefinition<TDocument>[])
Initializes a new instance of the CreateVectorSearchIndexModelBase<TDocument> class for a vector index where the vector embeddings are created manually. The required options for VectorSimilarity and the number of vector dimensions are passed to the constructor.
Properties
- Dimensions
Number of vector dimensions that vector search enforces at index-time and query-time, or uses to build the embeddings for auto-embedding indexes.
- ExcludedStoredFields
The fields that must NOT be stored in the index. Use WithExcludedStoredFields(params FieldDefinition<TDocument>[]) or WithExcludedStoredFields(params FieldDefinition<TDocument>[]) to configure this.
- Field
The field containing the vectors to index.
- FilterFields
Fields that may be used as filters in the vector query.
- HnswMaxEdges
Maximum number of edges (or connections) that a node can have in the Hierarchical Navigable Small Worlds graph.
- HnswNumEdgeCandidates
Analogous to numCandidates at query-time, this parameter controls the maximum number of nodes to evaluate to find the closest neighbors to connect to a new node.
- IncludedStoredFields
The fields that must be stored in the index. Use WithIncludedStoredFields(params FieldDefinition<TDocument>[]) or WithIncludedStoredFields(params FieldDefinition<TDocument>[]) to configure this.
- Quantization
Type of automatic vector quantization for your vectors.