Skip to main content

Vectara

Vectara 是一个用于构建 GenAI 应用的平台。它提供了一个易于使用的 API 用于文档索引和查询，该 API 由 Vectara 管理，并针对性能和准确性进行了优化。

¥Vectara is a platform for building GenAI applications. It provides an easy-to-use API for document indexing and querying that is managed by Vectara and is optimized for performance and accuracy.

你可以将 Vectara 用作 LangChain.js 的向量存储。

¥You can use Vectara as a vector store with LangChain.js.

👉 包含的嵌入​

¥👉 Embeddings Included

Vectara 内部使用自己的嵌入，因此你无需自己提供任何嵌入或调用其他服务来获取嵌入。

¥Vectara uses its own embeddings under the hood, so you don't have to provide any yourself or call another service to obtain embeddings.

这也意味着，如果你提供自己的嵌入，它们将没有任何作用。

¥This also means that if you provide your own embeddings, they'll be a no-op.

const store = await VectaraStore.fromTexts(
  ["hello world", "hi there"],
  [{ foo: "bar" }, { foo: "baz" }],
  // This won't have an effect. Provide a FakeEmbeddings instance instead for clarity.
  new OpenAIEmbeddings(),
  args
);

设置​

¥Setup

你需要：

¥You'll need to:

  • 创建一个 免费 Vectara 账户。

    ¥Create a free Vectara account.

  • 创建一个 corpus 来存储你的数据

    ¥Create a corpus to store your data

  • 创建一个具有 QueryService 和 IndexService 访问权限的 API 密钥，以便你可以访问此语料库。

    ¥Create an API key with QueryService and IndexService access so you can access this corpus

配置你的 .env 文件或提供参数以将 LangChain 连接到你的 Vectara 语料库：

¥Configure your .env file or provide args to connect LangChain to your Vectara corpus:

VECTARA_CUSTOMER_ID=your_customer_id
VECTARA_CORPUS_ID=your_corpus_id
VECTARA_API_KEY=your-vectara-api-key

请注意，你可以提供多个用逗号分隔的语料库 ID，以便一次查询多个语料库。例如：VECTARA_CORPUS_ID=3,8,9,43。要索引多个语料库，你需要为每个语料库创建一个单独的 VectaraStore 实例。

¥Note that you can provide multiple corpus IDs separated by commas for querying multiple corpora at once. For example: VECTARA_CORPUS_ID=3,8,9,43. For indexing multiple corpora, you'll need to create a separate VectaraStore instance for each corpus.

用法​

¥Usage

import { VectaraStore } from "@langchain/community/vectorstores/vectara";
import { VectaraSummaryRetriever } from "@langchain/community/retrievers/vectara_summary";
import { Document } from "@langchain/core/documents";

// Create the Vectara store.
const store = new VectaraStore({
  customerId: Number(process.env.VECTARA_CUSTOMER_ID),
  corpusId: Number(process.env.VECTARA_CORPUS_ID),
  apiKey: String(process.env.VECTARA_API_KEY),
  verbose: true,
});

// Add two documents with some metadata.
const doc_ids = await store.addDocuments([
  new Document({
    pageContent: "Do I dare to eat a peach?",
    metadata: {
      foo: "baz",
    },
  }),
  new Document({
    pageContent: "In the room the women come and go talking of Michelangelo",
    metadata: {
      foo: "bar",
    },
  }),
]);

// Perform a similarity search.
const resultsWithScore = await store.similaritySearchWithScore(
  "What were the women talking about?",
  1,
  {
    lambda: 0.025,
  }
);

// Print the results.
console.log(JSON.stringify(resultsWithScore, null, 2));
/*
[
  [
    {
      "pageContent": "In the room the women come and go talking of Michelangelo",
      "metadata": {
        "lang": "eng",
        "offset": "0",
        "len": "57",
        "foo": "bar"
      }
    },
    0.4678752
  ]
]
*/

const retriever = new VectaraSummaryRetriever({ vectara: store, topK: 3 });
const documents = await retriever.invoke("What were the women talking about?");

console.log(JSON.stringify(documents, null, 2));
/*
[
  {
    "pageContent": "<b>In the room the women come and go talking of Michelangelo</b>",
    "metadata": {
      "lang": "eng",
      "offset": "0",
      "len": "57",
      "foo": "bar"
    }
  },
  {
    "pageContent": "<b>In the room the women come and go talking of Michelangelo</b>",
    "metadata": {
      "lang": "eng",
      "offset": "0",
      "len": "57",
      "foo": "bar"
    }
  },
  {
    "pageContent": "<b>In the room the women come and go talking of Michelangelo</b>",
    "metadata": {
      "lang": "eng",
      "offset": "0",
      "len": "57",
      "foo": "bar"
    }
  }
]
*/

// Delete the documents.
await store.deleteDocuments(doc_ids);

API Reference:

请注意，lambda 是一个与 Vectara 混合搜索功能相关的参数，它在神经搜索和布尔/精确匹配之间进行权衡，如 此处 所述。我们建议将默认值设为 0.025，同时为高级用户提供自定义此值的方法。

¥Note that lambda is a parameter related to Vectara's hybrid search capbility, providing a tradeoff between neural search and boolean/exact match as described here. We recommend the value of 0.025 as a default, while providing a way for advanced users to customize this value if needed.

APIs​

Vectara 的 LangChain 向量存储使用 Vectara 的核心 API：

¥Vectara's LangChain vector store consumes Vectara's core APIs:

  • 索引 API 用于将文档存储在 Vectara 语料库中。

    ¥Indexing API for storing documents in a Vectara corpus.

  • 搜索 API 提供查询此数据的信息。此 API 支持混合搜索。

    ¥Search API for querying this data. This API supports hybrid search.

¥Related