Tair
Tair is a cloud native in-memory database service developed by
Alibaba Cloud. It provides rich data models and enterprise-grade capabilities to support your real-time online scenarios while maintaining full compatibility with open-sourceRedis.Tairalso introduces persistent memory-optimized instances that are based on the new non-volatile memory (NVM) storage medium.
本笔记本展示了如何使用与 Tair 向量数据库相关的功能。
您需要使用 pip install -qU langchain-community 安装 langchain-community 才能使用此集成
运行时，您应该有一个正在运行的 Tair 实例。
from langchain_community.embeddings.fake import FakeEmbeddings
from langchain_community.vectorstores import Tair
from langchain_text_splitters import CharacterTextSplitter
from langchain_community.document_loaders import TextLoader
loader = TextLoader("../../how_to/state_of_the_union.txt")
documents = loader.load()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
docs = text_splitter.split_documents(documents)
embeddings = FakeEmbeddings(size=128)
API 参考：TextLoader
使用 TAIR_URL 环境变量连接到 Tair
export TAIR_URL="redis://{username}:{password}@{tair_address}:{tair_port}"
或关键字参数 tair_url。
然后将文档和嵌入存储到Tair中。
tair_url = "redis://localhost:6379"
# drop first if index already exists
Tair.drop_index(tair_url=tair_url)
vector_store = Tair.from_documents(docs, embeddings, tair_url=tair_url)
查询相似文档。
query = "What did the president say about Ketanji Brown Jackson"
docs = vector_store.similarity_search(query)
docs[0]
Tair 混合搜索索引构建
# drop first if index already exists
Tair.drop_index(tair_url=tair_url)
vector_store = Tair.from_documents(
docs, embeddings, tair_url=tair_url, index_params={"lexical_algorithm": "bm25"}
)
Tair 混合搜索
query = "What did the president say about Ketanji Brown Jackson"
# hybrid_ratio: 0.5 hybrid search, 0.9999 vector search, 0.0001 text search
kwargs = {"TEXT": query, "hybrid_ratio": 0.5}
docs = vector_store.similarity_search(query, **kwargs)
docs[0]