程序包 dev.langchain4j.store.embedding.mongodb

类 MongoDbEmbeddingStore

java.lang.Object

dev.langchain4j.store.embedding.mongodb.MongoDbEmbeddingStore

所有已实现的接口:

EmbeddingStore<TextSegment>

public class MongoDbEmbeddingStore
extends Object
implements EmbeddingStore<TextSegment>

Represents a MongoDB index as an embedding store.

More info to set up MongoDb as vectorDatabase.

tutorial how to use a knn-vector in MongoDB Atlas (great starting point).

If you are using a free tier, #createIndex = true might not be supported, so you will need to create an index manually. In your Atlas web console go to: DEPLOYMENT -> Database -> {your cluster} -> Atlas Search -> Create Index Search -> "JSON Editor" under "Atlas Search" -> Next -> Select your database in the left pane -> Insert the following JSON into the right pane (set "dimensions" and "metadata"->"fields" to desired values)

 {
   "mappings": {
     "dynamic": false,
     "fields": {
       "embedding": {
         "dimensions": 384,
         "similarity": "cosine",
         "type": "knnVector"
       },
       "metadata": {
         "dynamic": false,
         "fields": {
           "test-key": {
             "type": "token"
           }
         },
         "type": "document"
       }
     }
   }
 }
 

-> Next -> Create Search Index

嵌套类概要

嵌套类

修饰符和类型	类	说明
static class	MongoDbEmbeddingStore.Builder

构造器概要

构造器

构造器	说明
MongoDbEmbeddingStore(com.mongodb.client.MongoClient mongoClient, String databaseName, String collectionName, String indexName, Long maxResultRatio, com.mongodb.client.model.CreateCollectionOptions createCollectionOptions, org.bson.conversions.Bson filter, IndexMapping indexMapping, Boolean createIndex)

方法概要

修饰符和类型	方法	说明
String	add(Embedding embedding)
String	add(Embedding embedding, TextSegment textSegment)
void	add(String id, Embedding embedding)
List<String>	addAll(List<Embedding> embeddings)
List<String>	addAll(List<Embedding> embeddings, List<TextSegment> embedded)
static MongoDbEmbeddingStore.Builder	builder()
List<EmbeddingMatch<TextSegment>>	findRelevant(Embedding referenceEmbedding, int maxResults, double minScore)

从类继承的方法 java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

从接口继承的方法 dev.langchain4j.store.embedding.EmbeddingStore

findRelevant, findRelevant, findRelevant, remove, removeAll, removeAll, removeAll, search

构造器详细资料

MongoDbEmbeddingStore

public MongoDbEmbeddingStore(com.mongodb.client.MongoClient mongoClient,
 String databaseName,
 String collectionName,
 String indexName,
 Long maxResultRatio,
 com.mongodb.client.model.CreateCollectionOptions createCollectionOptions,
 org.bson.conversions.Bson filter,
 IndexMapping indexMapping,
 Boolean createIndex)

方法详细资料

builder

public static MongoDbEmbeddingStore.Builder builder()

add

public String add(Embedding embedding)

指定者:

add 在接口中 EmbeddingStore<TextSegment>

add

public void add(String id,
 Embedding embedding)

指定者:

add 在接口中 EmbeddingStore<TextSegment>

add

public String add(Embedding embedding,
 TextSegment textSegment)

指定者:

add 在接口中 EmbeddingStore<TextSegment>

addAll

public List<String> addAll(List<Embedding> embeddings)

指定者:

addAll 在接口中 EmbeddingStore<TextSegment>

addAll

public List<String> addAll(List<Embedding> embeddings,
 List<TextSegment> embedded)

指定者:

addAll 在接口中 EmbeddingStore<TextSegment>

findRelevant

public List<EmbeddingMatch<TextSegment>> findRelevant(Embedding referenceEmbedding,
 int maxResults,
 double minScore)

指定者:

findRelevant 在接口中 EmbeddingStore<TextSegment>