Class CouchbaseEmbeddingStore

java.lang.Object
dev.langchain4j.store.embedding.couchbase.CouchbaseEmbeddingStore
All Implemented Interfaces:
EmbeddingStore<TextSegment>

public class CouchbaseEmbeddingStore extends Object implements EmbeddingStore<TextSegment>
Represents a Couchbase index as an embedding store. Current implementation assumes the index uses the cosine distance metric.
Supports storing Metadata
  • Constructor Details

    • CouchbaseEmbeddingStore

      public CouchbaseEmbeddingStore(String clusterUrl, String username, String password, String bucketName, String scopeName, String collectionName, String searchIndexName, Integer dimensions)
      Creates an instance of CouchbaseEmbeddingStore
      Parameters:
      clusterUrl - the url of the couchbase cluster to be used by the store
      username - username on the cluster
      password - password on the cluster
      bucketName - name of a cluster bucket in which to store the embeddings
      scopeName - name of a scope in the bucket in which to store the embeddings
      collectionName - name of a collection in the scope in which to store the embeddings
      searchIndexName - name of the FTS index to be used for searching embeddings
      dimensions - expected size (dimensionality) of vectors that will be stored
    • CouchbaseEmbeddingStore

      public CouchbaseEmbeddingStore(String clusterUrl, String username, String password, String bucketName, String scopeName, String collectionName, String searchIndexName, Integer dimensions, Integer bucketTimeout)
      Creates an instance of CouchbaseEmbeddingStore
      Parameters:
      clusterUrl - the url of the couchbase cluster to be used by the store
      username - username on the cluster
      password - password on the cluster
      bucketName - name of a cluster bucket in which to store the embeddings
      scopeName - name of a scope in the bucket in which to store the embeddings
      collectionName - name of a collection in the scope in which to store the embeddings
      searchIndexName - name of the FTS index to be used for searching embeddings
      dimensions - expected size (dimensionality) of vectors that will be stored
      bucketTimeout - bucket connection timeout
  • Method Details