我们将使用GPT-4的推理能力来推断元数据过滤器。根据您的使用情况,"gpt-3.5-turbo"
也可以使用。
本指南演示了如何在LlamaIndex中使用Weaviate执行自动检索。
Weaviate向量数据库支持一组元数据过滤器,除了用于语义搜索的查询字符串。给定自然语言查询,我们首先使用大型语言模型(LLM)推断一组元数据过滤器,以及要传递给向量数据库的正确查询字符串(也可以为空)。然后针对向量数据库执行整个查询包。
这允许进行比top-k语义搜索更动态、更具表现力的检索形式。对于给定查询的相关上下文可能只需要在元数据标签上进行过滤,或者需要在过滤集合内进行过滤+语义搜索的联合组合,或者只需要原始语义搜索。
我们尝试运行一些示例数据。请注意元数据过滤器是如何被推断出来的 - 这有助于更精确地检索!
如果您在Colab上打开此笔记本,您可能需要安装LlamaIndex 🦙。
%pip install llama-index-vector-stores-weaviate
!pip install llama-index weaviate-client
# 从Weaviate向量数据库进行自动检索
本指南演示了如何在[LlamaIndex](https://weaviate.io/)中使用Weaviate执行**自动检索**。
Weaviate向量数据库支持一组[元数据过滤器](https://weaviate.io/developers/weaviate/search/filters),除了用于语义搜索的查询字符串。给定自然语言查询,我们首先使用大型语言模型(LLM)推断一组元数据过滤器,以及要传递给向量数据库的正确查询字符串(也可以为空)。然后针对向量数据库执行整个查询包。
这允许进行比top-k语义搜索更动态、更具表现力的检索形式。对于给定查询的相关上下文可能只需要在元数据标签上进行过滤,或者需要在过滤集合内进行过滤+语义搜索的联合组合,或者只需要原始语义搜索。
我们将使用GPT-4的推理能力来推断元数据过滤器。根据您的使用情况,"gpt-3.5-turbo"
也可以使用。
# 设置OpenAIimport osimport getpassimport openaios.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API Key:")openai.api_key = os.environ["OPENAI_API_KEY"]
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.llms.openai import OpenAI
from llama_index.core.settings import Settings
Settings.llm = OpenAI(model="gpt-4")
Settings.embed_model = OpenAIEmbedding()
这个笔记本使用嵌入模式中的Weaviate,该模式支持Linux和macOS。
如果你更喜欢尝试Weaviate的完全托管服务,Weaviate Cloud Services (WCS),你可以取消注释中的代码。
import weaviatefrom weaviate.embedded import EmbeddedOptions# 连接到嵌入模式的Weaviate客户端client = weaviate.connect_to_embedded()# 如果要使用Weaviate云服务而不是嵌入模式,请启用此代码。"""import weaviate# 云端cluster_url = ""api_key = ""client = weaviate.connect_to_wcs(cluster_url=cluster_url, auth_credentials=weaviate.auth.AuthApiKey(api_key), )# 本地# client = weaviate.connect_to_local()"""
我们向向量数据库中插入一些包含文本块的示例节点。请注意,每个 TextNode
不仅包含文本,还包含元数据,例如 category
和 country
。这些元数据字段将被转换/存储在底层的向量数据库中。
from llama_index.core.schema import TextNode
nodes = [
TextNode(
text=(
"Michael Jordan is a retired professional basketball player,"
" widely regarded as one of the greatest basketball players of all"
" time."
),
metadata={
"category": "Sports",
"country": "United States",
},
),
TextNode(
text=(
"Angelina Jolie is an American actress, filmmaker, and"
" humanitarian. She has received numerous awards for her acting"
" and is known for her philanthropic work."
),
metadata={
"category": "Entertainment",
"country": "United States",
},
),
TextNode(
text=(
"Elon Musk is a business magnate, industrial designer, and"
" engineer. He is the founder, CEO, and lead designer of SpaceX,"
" Tesla, Inc., Neuralink, and The Boring Company."
),
metadata={
"category": "Business",
"country": "United States",
},
),
TextNode(
text=(
"Rihanna is a Barbadian singer, actress, and businesswoman. She"
" has achieved significant success in the music industry and is"
" known for her versatile musical style."
),
metadata={
"category": "Music",
"country": "Barbados",
},
),
TextNode(
text=(
"Cristiano Ronaldo is a Portuguese professional footballer who is"
" considered one of the greatest football players of all time. He"
" has won numerous awards and set multiple records during his"
" career."
),
metadata={
"category": "Sports",
"country": "Portugal",
},
),
]
在这里,我们将数据加载到向量存储中。如上所述,每个节点的文本和元数据都将转换为Weaviate中的相应表示。我们现在可以从Weaviate对这些数据运行语义查询,还可以进行元数据过滤。
from llama_index.core import VectorStoreIndex, StorageContext
from llama_index.vector_stores.weaviate import WeaviateVectorStore
vector_store = WeaviateVectorStore(
weaviate_client=client, index_name="LlamaIndex_filter"
)
storage_context = StorageContext.from_defaults(vector_store=vector_store)
index = VectorStoreIndex(nodes, storage_context=storage_context)
VectorIndexAutoRetriever
¶我们定义了核心的 VectorIndexAutoRetriever
模块。该模块接收 VectorStoreInfo
,其中包含向量存储集合的结构化描述以及其支持的元数据过滤器。然后,这些信息将在自动检索提示中使用,LLM 将推断元数据过滤器。
from llama_index.core.retrievers import VectorIndexAutoRetriever
from llama_index.core.vector_stores.types import MetadataInfo, VectorStoreInfo
vector_store_info = VectorStoreInfo(
content_info="brief biography of celebrities",
metadata_info=[
MetadataInfo(
name="category",
type="str",
description=(
"Category of the celebrity, one of [Sports, Entertainment,"
" Business, Music]"
),
),
MetadataInfo(
name="country",
type="str",
description=(
"Country of the celebrity, one of [United States, Barbados,"
" Portugal]"
),
),
],
)
retriever = VectorIndexAutoRetriever(
index, vector_store_info=vector_store_info
)
我们尝试运行一些示例数据。请注意元数据过滤器是如何被推断出来的 - 这有助于更精确地检索!
response = retriever.retrieve("Tell me about celebrities from United States")
print(response[0])
response = retriever.retrieve(
"Tell me about Sports celebrities from United States"
)
print(response[0])