向量化及存储

@Bean
public EmbeddingModel embeddingModel() {
    return OpenAiEmbeddingModel.builder()
            .apiKey(System.getenv("DASHSCOPE_KEY"))
            .modelName("text-embedding-v3")
            .baseUrl("https://dashscope.aliyuncs.com/compatible-mode/v1")
            .build();
}

Embedding queryEmbedding = embeddingModel.embed("host报错请检查环境").content();

docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant

<dependency>
   <groupId>dev.langchain4j</groupId>
   <artifactId>langchain4j-qdrant</artifactId>
</dependency>

@Bean
public QdrantClient qdrantClient() {
    QdrantGrpcClient.Builder grpcClientBuilder = QdrantGrpcClient.newBuilder("127.0.0.1", 6334, false);
    return new QdrantClient(grpcClientBuilder.build());
}

var vectorParams = Collections.VectorParams.newBuilder()
        .setDistance(Collections.Distance.Cosine)
        .setSize(1024)
        .build();
qdrantClient.createCollectionAsync("testv", vectorParams);

@Bean
public EmbeddingStore<TextSegment> embeddingStore() {
    return QdrantEmbeddingStore.builder()
            .host("127.0.0.1")
            .port(6334)
            .collectionName("testv")
            .build();
}

TextSegment segment1 = TextSegment.from("浏览器报错 404，请检测您输入的路径是否正确");
segment1.metadata().put("author", "冷冷");
Embedding embedding1 = embeddingModel.embed(segment1).content();
embeddingStore.add(embedding1, segment1);

Embedding queryEmbedding = embeddingModel.embed("404 是哪里的问题？").content();
EmbeddingSearchRequest embeddingSearchRequest = EmbeddingSearchRequest.builder()
        .queryEmbedding(queryEmbedding)
        .maxResults(1)
        .build();
EmbeddingSearchResult<TextSegment> searchResult = embeddingStore.search(embeddingSearchRequest);
System.out.println(searchResult.matches().get(0).embedded().text());

EmbeddingSearchRequest embeddingSearchRequest = EmbeddingSearchRequest.builder()
        .queryEmbedding(queryEmbedding)
        .filter(metadataKey("author").isEqualTo("冷冷"))
        .maxResults(1)
        .build();

List<EmbeddingMatch<TextSegment>> relevant = searchResult.matches();
EmbeddingMatch<TextSegment> embeddingMatch = relevant.get(0);

System.out.println(embeddingMatch.score());  // 输出匹配分数
System.out.println(embeddingMatch.embedded().text());  // 输出匹配文本