这是继上一篇文章 “Elasticsearch:使用 Open AI 和 Langchain 的 RAG - Retrieval Augmented Generation (一)” 的续篇。在这篇文章中,我主要来讲述 ElasticVectorSearch 的使用。
我们的设置和之前的那篇文章是一样的,只不过,在这里我们使用 ElasticVectorSearch 而不是 ElasticKnnSearch。
创建应用并展示
安装包
#!pip3 install langchain
导入包
from dotenv import load_dotenv
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import ElasticKnnSearch
from langchain.text_splitter import CharacterTextSplitter
from urllib.request import urlopen
import os, jsonload_dotenv()openai_api_key=os.getenv('OPENAI_API_KEY')
elastic_user=os.getenv('ES_USER')
elastic_password=os.getenv('ES_PASSWORD')
elastic_endpoint=os.getenv("ES_ENDPOINT")
elastic_index_name='elastic-vector-search'
将文档分成段落
import json# Load data into a JSON object
with open('workplace-docs.json') as f:workplace_docs = json.load(f)print(f"Successfully loaded {len(workplace_docs)} documents")metadata = []
content = []for doc in workplace_docs:content.append(doc["content"])metadata.append({"name": doc["name"],"summary": doc["summary"],"rolePermissions":doc["rolePermissions"]})text_splitter = CharacterTextSplitter(chunk_size=100, chunk_overlap=0)
docs = text_splitter.create_documents(content, metadatas=metadata)
把数据写入到 Elasticsearch
embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)url = f"https://{elastic_user}:{elastic_password}@{elastic_endpoint}:9200"ssl_verify = {"verify_certs": True,"basic_auth": (elastic_user, elastic_password),"ca_certs": "./http_ca.crt"
}es = ElasticVectorSearch.from_documents( docs,embedding = embeddings, elasticsearch_url = url, index_name = elastic_index_name, ssl_verify = ssl_verify)
如上所示,ElasticVectorSearch 在未来的发布中将被移除。
运行完上面的代码后,我们可以到 Kibana 中进行查看:
展示结果
def showResults(output):print("Total results: ", len(output))for index in range(len(output)):print(output[index])
Similarity / Vector Search (KNN Search)
query = "work from home policy"
result = es.similarity_search(query=query)showResults(result)
我们上面实现的代码可以在地址 https://github.com/liu-xiao-guo/semantic_search_es/blob/main/ElasticVectorSearch.ipynb 进行下载。