跳至主要内容

WeaviateStore

Weaviate 是一个开源向量数据库，它存储对象和向量，允许将向量搜索与结构化过滤相结合。LangChain 通过 weaviate-ts-client 包连接到 Weaviate，weaviate-ts-client 包是 Weaviate 的官方 Typescript 客户端。

本指南简要概述了 Weaviate 向量存储 的入门方法。有关所有 WeaviateStore 功能和配置的详细文档，请查看 API 参考。

概述​

集成详细信息​

类包PY 支持最新包
WeaviateStore@langchain/weaviate✅NPM - Version

设置​

要使用 Weaviate 向量存储，您需要设置一个 Weaviate 实例并安装 @langchain/weaviate 集成包。您还应该安装 weaviate-ts-client 包以初始化一个客户端连接到您的实例，如果要为索引文档分配 ID，还要安装 uuid 包。

本指南还将使用 OpenAI 嵌入，这需要您安装 @langchain/openai 集成包。您也可以使用 其他支持的嵌入模型。

提示

查看 此部分以获取有关安装集成包的一般说明。

yarn add @langchain/weaviate @langchain/core weaviate-ts-client uuid @langchain/openai

您需要在本地或服务器上运行 Weaviate。查看 Weaviate 文档 以了解更多信息。

凭据​

设置好实例后，请设置以下环境变量

// http or https
process.env.WEAVIATE_SCHEME = "";
// If running locally, include port e.g. "localhost:8080"
process.env.WEAVIATE_HOST = "YOUR_HOSTNAME";
// Optional, for cloud deployments
process.env.WEAVIATE_API_KEY = "YOUR_API_KEY";

如果您在本指南中使用 OpenAI 嵌入，还需要设置 OpenAI 密钥

process.env.OPENAI_API_KEY = "YOUR_API_KEY";

如果您希望自动跟踪模型调用，还可以通过取消注释以下内容来设置 LangSmith API 密钥

// process.env.LANGCHAIN_TRACING_V2="true"
// process.env.LANGCHAIN_API_KEY="your-api-key"

实例化​

import { WeaviateStore } from "@langchain/weaviate";
import { OpenAIEmbeddings } from "@langchain/openai";

import weaviate from "weaviate-ts-client";
// import { ApiKey } from "weaviate-ts-client"

const embeddings = new OpenAIEmbeddings({
  model: "text-embedding-3-small",
});

// The Weaviate SDK has an issue with types
const weaviateClient = (weaviate as any).client({
  scheme: process.env.WEAVIATE_SCHEME ?? "http",
  host: process.env.WEAVIATE_HOST ?? "localhost",
  // If necessary
  // apiKey: new ApiKey(process.env.WEAVIATE_API_KEY ?? "default"),
});

const vectorStore = new WeaviateStore(embeddings, {
  client: weaviateClient,
  // Must start with a capital letter
  indexName: "Langchainjs_test",
  // Default value
  textKey: "text",
  // Any keys you intend to set as metadata
  metadataKeys: ["source"],
});

管理向量存储​

向向量存储中添加项目​

注意： 如果您希望将 ID 与索引文档关联，则它们必须是 UUID。

import type { Document } from "@langchain/core/documents";
import { v4 as uuidv4 } from "uuid";

const document1: Document = {
  pageContent: "The powerhouse of the cell is the mitochondria",
  metadata: { source: "https://example.com" },
};

const document2: Document = {
  pageContent: "Buildings are made out of brick",
  metadata: { source: "https://example.com" },
};

const document3: Document = {
  pageContent: "Mitochondria are made out of lipids",
  metadata: { source: "https://example.com" },
};

const document4: Document = {
  pageContent: "The 2024 Olympics are in Paris",
  metadata: { source: "https://example.com" },
};

const documents = [document1, document2, document3, document4];
const uuids = [uuidv4(), uuidv4(), uuidv4(), uuidv4()];

await vectorStore.addDocuments(documents, { ids: uuids });
[
  '610f9b92-9bee-473f-a4db-8f2ca6e3442d',
  '995160fa-441e-41a0-b476-cf3785518a0d',
  '0cdbe6d4-0df8-4f99-9b67-184009fee9a2',
  '18a8211c-0649-467b-a7c5-50ebb4b9ca9d'
]

从向量存储中删除项目​

您可以通过 ID 删除，方法是传递 filter 参数

await vectorStore.delete({ ids: [uuids[3]] });

查询向量存储​

创建向量存储并添加相关文档后，您很可能希望在链或代理运行期间查询它。

直接查询​

执行简单的相似度搜索可以按如下方式完成

const filter = {
  where: {
    operator: "Equal" as const,
    path: ["source"],
    valueText: "https://example.com",
  },
};

const similaritySearchResults = await vectorStore.similaritySearch(
  "biology",
  2,
  filter
);

for (const doc of similaritySearchResults) {
  console.log(`* ${doc.pageContent} [${JSON.stringify(doc.metadata, null)}]`);
}
* The powerhouse of the cell is the mitochondria [{"source":"https://example.com"}]
* Mitochondria are made out of lipids [{"source":"https://example.com"}]

查看 此页面 以了解有关 Weaviate 过滤语法的信息。

如果您希望执行相似度搜索并接收相应的得分，您可以运行

const similaritySearchWithScoreResults =
  await vectorStore.similaritySearchWithScore("biology", 2, filter);

for (const [doc, score] of similaritySearchWithScoreResults) {
  console.log(
    `* [SIM=${score.toFixed(3)}] ${doc.pageContent} [${JSON.stringify(
      doc.metadata
    )}]`
  );
}
* [SIM=0.835] The powerhouse of the cell is the mitochondria [{"source":"https://example.com"}]
* [SIM=0.852] Mitochondria are made out of lipids [{"source":"https://example.com"}]

通过转换为检索器查询​

您还可以将向量存储转换为 检索器，以便更轻松地在链中使用。

const retriever = vectorStore.asRetriever({
  // Optional filter
  filter: filter,
  k: 2,
});
await retriever.invoke("biology");
[
  Document {
    pageContent: 'The powerhouse of the cell is the mitochondria',
    metadata: { source: 'https://example.com' },
    id: undefined
  },
  Document {
    pageContent: 'Mitochondria are made out of lipids',
    metadata: { source: 'https://example.com' },
    id: undefined
  }
]

用于检索增强生成的使用​

有关如何将此向量存储用于检索增强生成 (RAG) 的指南，请查看以下部分

API 参考​

有关所有 WeaviateStore 功能和配置的详细文档，请查看 API 参考。

此页面是否有帮助？


您也可以留下详细的反馈 在 GitHub 上.