惠州建网站公司_sem seo 区别_百度推广做二级域名_网站和网页的区别

docker pull chromadb/chroma

docker save -o chromadb.tar chromadb/chroma

scp /Users/本地目录/chromadb.tar root@远程机器ip:/远程机器路径

docker load -i chromadb.tar

docker run --rm --name chromadb -p 8001:8000 -v ./chromadb:/chroma/chroma --env-file ./.chroma_env chromadb/chroma

# 容器中数据是否持久化配置

IS_PERSISTENT=TRUE

# 容器中向量数据存储路径

chroma-data=/chroma/chroma/

# 服务端认证token

CHROMA_SERVER_AUTHN_CREDENTIALS=test-token

# 服务端认证方式

CHROMA_SERVER_AUTHN_PROVIDER=chromadb.auth.token_authn.TokenAuthenticationServerProvider

import chromadb

import chromadb.utils.embedding_functions as embedding_functions

import logging

import sys

from chromadb.config import Settings

from dotenv import load_dotenv

from datetime import datetime

# 使用日志记录查看查询和事件

# log_format = '%(asctime)s - %(levelname)s - %(message)s'

# logging.basicConfig(stream=sys.stdout, level=logging.INFO,format=log_format)

# 获取向量数据库客户端

chroma_client = chromadb.HttpClient(host='服务器ip',

                                    port=8001,

                                    settings=Settings(

                                            chroma_client_auth_provider="chromadb.auth.token_authn.TokenAuthClientProvider",

                                            chroma_client_auth_credentials="test-token")

                                    )

# 定义向量化函数（使用微软的azure）

openai_ef = embedding_functions.OpenAIEmbeddingFunction(

                api_key="",

                api_base="",

                api_type="azure",

                api_version="",

                model_name="embedding-3-small"

            )

# 创建数据集

# collection = chroma_client.create_collection(name="my_collection",

#                                              embedding_function=openai_ef,

#                                              metadata={

#                                                 "description": "my first Chroma collection",

#                                                 "created": str(datetime.now())

#                                              }  )

# 删除数据集

# chroma_client.delete_collection(name="my_collection")

# 获取数据集

collection = chroma_client.get_collection(name="my_collection",embedding_function=openai_ef)

# 向数据集中增加数据

# collection.add(

#     documents=[

#         "This is a document about pineapple",

#         "This is a document about hawai"

#     ],

#     metadatas=[{"chapter": "3", "verse": "16"}, {"chapter": "3", "verse": "5"}],

#     ids=["id1", "id2"]

# )

# 文本查询数据

# results = collection.query(

#     query_texts=["This is a query document about hawai"], # Chroma will embed this for you

#     n_results=1 # how many results to return

# )

# 文本查询并过滤

results = collection.query(

    query_texts=["This is a query document about hawai"],

    n_results=2,

    # metadatas 元数据过滤

    # where={"verse": "16"},

    # 文本内容过滤

    where_document={ "$contains":"pineapple"}

)

print(results)