Vald

Vald 是一个高度可扩展的分布式快速近似最近邻 (ANN) 密集向量搜索引擎。

此笔记本展示了如何使用与 Vald 数据库相关的功能。

要运行此笔记本，您需要一个正在运行的 Vald 集群。有关更多信息，请查看 入门。

请参阅 安装说明。

%pip install --upgrade --quiet  vald-client-python langchain-community

基本示例

from langchain_community.document_loaders import TextLoader
from langchain_community.vectorstores import Vald
from langchain_huggingface import HuggingFaceEmbeddings
from langchain_text_splitters import CharacterTextSplitter

raw_documents = TextLoader("state_of_the_union.txt").load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
documents = text_splitter.split_documents(raw_documents)
model_name = "sentence-transformers/all-mpnet-base-v2"
embeddings = HuggingFaceEmbeddings(model_name=model_name)
db = Vald.from_documents(documents, embeddings, host="localhost", port=8080)

API 参考：TextLoader | Vald | HuggingFaceEmbeddings | CharacterTextSplitter

query = "What did the president say about Ketanji Brown Jackson"
docs = db.similarity_search(query)
docs[0].page_content

按向量进行相似性搜索

embedding_vector = embeddings.embed_query(query)
docs = db.similarity_search_by_vector(embedding_vector)
docs[0].page_content

带分数的相似性搜索

docs_and_scores = db.similarity_search_with_score(query)
docs_and_scores[0]

最大边际相关性搜索 (MMR)

除了在检索器对象中使用相似性搜索外，您还可以使用 mmr 作为检索器。

retriever = db.as_retriever(search_type="mmr")
retriever.invoke(query)

或者直接使用 max_marginal_relevance_search

db.max_marginal_relevance_search(query, k=2, fetch_k=10)

使用安全连接的示例

为了运行此笔记本，需要运行一个具有安全连接的 Vald 集群。

以下是一个使用 Athenz 身份验证的 Vald 集群配置示例。

ingress(TLS) -> authorization-proxy(检查 grpc 元数据中的 athenz-role-auth) -> vald-lb-gateway

import grpc

with open("test_root_cacert.crt", "rb") as root:
    credentials = grpc.ssl_channel_credentials(root_certificates=root.read())

# Refresh is required for server use
with open(".ztoken", "rb") as ztoken:
    token = ztoken.read().strip()

metadata = [(b"athenz-role-auth", token)]

from langchain_community.document_loaders import TextLoader
from langchain_community.vectorstores import Vald
from langchain_huggingface import HuggingFaceEmbeddings
from langchain_text_splitters import CharacterTextSplitter

raw_documents = TextLoader("state_of_the_union.txt").load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
documents = text_splitter.split_documents(raw_documents)
model_name = "sentence-transformers/all-mpnet-base-v2"
embeddings = HuggingFaceEmbeddings(model_name=model_name)

db = Vald.from_documents(
    documents,
    embeddings,
    host="localhost",
    port=443,
    grpc_use_secure=True,
    grpc_credentials=credentials,
    grpc_metadata=metadata,
)

API 参考：TextLoader | Vald | HuggingFaceEmbeddings | CharacterTextSplitter

query = "What did the president say about Ketanji Brown Jackson"
docs = db.similarity_search(query, grpc_metadata=metadata)
docs[0].page_content

按向量进行相似性搜索

embedding_vector = embeddings.embed_query(query)
docs = db.similarity_search_by_vector(embedding_vector, grpc_metadata=metadata)
docs[0].page_content

带分数的相似性搜索

docs_and_scores = db.similarity_search_with_score(query, grpc_metadata=metadata)
docs_and_scores[0]

最大边际相关性搜索 (MMR)

retriever = db.as_retriever(
    search_kwargs={"search_type": "mmr", "grpc_metadata": metadata}
)
retriever.invoke(query, grpc_metadata=metadata)

或者

db.max_marginal_relevance_search(query, k=2, fetch_k=10, grpc_metadata=metadata)

相关

• 向量存储 概念指南
• 向量存储 操作指南