在本示例中，我们将演示如何组合来自多个查询和多个索引的检索结果。

检索到的节点将根据这篇论文中演示的 Reciprocal Rerank Fusion 算法重新排序。它提供了一种高效的方法来重排检索结果，而无需过多计算或依赖外部模型。

完整致谢 GitHub 上的 @Raduaschl，感谢其示例实现。

%pip install llama-index-llms-openai
%pip install llama-index-retrievers-bm25

import os
import openai

os.environ["OPENAI_API_KEY"] = "sk-..."
openai.api_key = os.environ["OPENAI_API_KEY"]

设置

如果你在 Colab 中打开这个 Notebook，可能需要先安装 LlamaIndex 🦙。

下载数据

!mkdir -p 'data/paul_graham/'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'

--2024-02-12 17:59:58--  https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/paul_graham/paul_graham_essay.txt
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 2606:50c0:8003::154, 2606:50c0:8001::154, 2606:50c0:8002::154, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|2606:50c0:8003::154|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 75042 (73K) [text/plain]
Saving to: ‘data/paul_graham/paul_graham_essay.txt’

data/paul_graham/pa 100%[===================>]  73.28K   327KB/s    in 0.2s

2024-02-12 17:59:59 (327 KB/s) - ‘data/paul_graham/paul_graham_essay.txt’ saved [75042/75042]

from llama_index.core import SimpleDirectoryReader

documents = SimpleDirectoryReader("./data/paul_graham/").load_data()

接下来，我们将在文档之上设置一个向量索引。

from llama_index.core import VectorStoreIndex
from llama_index.core.node_parser import SentenceSplitter

splitter = SentenceSplitter(chunk_size=256)

index = VectorStoreIndex.from_documents(documents, transformations=[splitter])

创建混合融合检索器

在这一步中，我们将索引与基于 BM25 的检索器融合。这将使我们能够同时捕获输入查询中的语义关系和关键词。

由于这两个检索器都会计算分数，我们可以使用倒数重排算法对节点重新排序，而无需使用额外模型或过多计算。

此设置还将查询 4 次：一次使用你的原始查询，并生成另外 3 个查询。

默认情况下，它使用以下提示词生成额外查询：

QUERY_GEN_PROMPT = (
    "You are a helpful assistant that generates multiple search queries based on a "
    "single input query. Generate {num_queries} search queries, one on each line, "
    "related to the following input query:\n"
    "Query: {query}\n"
    "Queries:\n"
)

首先，我们创建检索器。每个检索器将检索前 2 个最相似的节点：

from llama_index.retrievers.bm25 import BM25Retriever

vector_retriever = index.as_retriever(similarity_top_k=2)

bm25_retriever = BM25Retriever.from_defaults(
    docstore=index.docstore, similarity_top_k=2
)

接下来，我们可以创建融合检索器，它将从检索器返回的 4 个节点中返回前 2 个最相似的节点：

from llama_index.core.retrievers import QueryFusionRetriever

retriever = QueryFusionRetriever(
    [vector_retriever, bm25_retriever],
    similarity_top_k=2,
    num_queries=4,  # set this to 1 to disable query generation
    mode="reciprocal_rerank",
    use_async=True,
    verbose=True,
    # query_gen_prompt="...",  # we could override the query generation prompt here
)

# apply nested async to run in a notebook
import nest_asyncio

nest_asyncio.apply()

nodes_with_scores = retriever.retrieve(
    "What happened at Interleafe and Viaweb?"
)

Generated queries:
1. What were the major events or milestones in the history of Interleafe and Viaweb?
2. Can you provide a timeline of the key developments and achievements of Interleafe and Viaweb?
3. What were the successes and failures of Interleafe and Viaweb as companies?

for node in nodes_with_scores:
    print(f"Score: {node.score:.2f} - {node.text}...\n-----\n")

Score: 0.03 - The UI was horrible, but it proved you could build a whole store through the browser, without any client software or typing anything into the command line on the server.

Now we felt like we were really onto something. I had visions of a whole new generation of software working this way. You wouldn't need versions, or ports, or any of that crap. At Interleaf there had been a whole group called Release Engineering that seemed to be at least as big as the group that actually wrote the software. Now you could just update the software right on the server.

We started a new company we called Viaweb, after the fact that our software worked via the web, and we got $10,000 in seed funding from Idelle's husband Julian. In return for that and doing the initial legal work and giving us business advice, we gave him 10% of the company. Ten years later this deal became the model for Y Combinator's. We knew founders needed something like this, because we'd needed it ourselves....
-----

Score: 0.03 - Now we felt like we were really onto something. I had visions of a whole new generation of software working this way. You wouldn't need versions, or ports, or any of that crap. At Interleaf there had been a whole group called Release Engineering that seemed to be at least as big as the group that actually wrote the software. Now you could just update the software right on the server.

We started a new company we called Viaweb, after the fact that our software worked via the web, and we got $10,000 in seed funding from Idelle's husband Julian. In return for that and doing the initial legal work and giving us business advice, we gave him 10% of the company. Ten years later this deal became the model for Y Combinator's. We knew founders needed something like this, because we'd needed it ourselves.

At this stage I had a negative net worth, because the thousand dollars or so I had in the bank was more than counterbalanced by what I owed the government in taxes. (Had I diligently set aside the proper proportion of the money I'd made consulting for Interleaf?...
-----

如我们所见，返回的两个节点都正确地提到了 Viaweb 和 Interleaf！

在查询引擎中使用！

现在，我们可以将检索器接入查询引擎，以合成自然语言响应。

from llama_index.core.query_engine import RetrieverQueryEngine

query_engine = RetrieverQueryEngine.from_args(retriever)

response = query_engine.query("What happened at Interleafe and Viaweb?")

Generated queries:
1. What were the major events or milestones in the history of Interleafe and Viaweb?
2. Can you provide a timeline of the key developments and achievements of Interleafe and Viaweb?
3. What were the outcomes or impacts of Interleafe and Viaweb on the respective industries they operated in?

from llama_index.core.response.notebook_utils import display_response

display_response(response)

最终响应： 在 Interleaf，有一个名为 Release Engineering 的团队，其规模与真正编写软件的团队一样大。这表明当时非常重视管理软件的版本和端口。然而，在 Viaweb，创始人意识到他们可以直接在服务器上更新软件，从而无需版本和端口。他们创立了 Viaweb，一家构建通过 Web 运行的软件的公司。他们获得了 10,000 美元的种子资金，并将公司 10% 的股份给了 Julian，后者提供了资金和商业建议。这笔交易后来成为 Y Combinator 的模式。