回答

前提：有docker运行环境和openai的有效key

1、安装qdrant

docker pull ainize/adrant

2、运行qdrant

docker run -p 6333:6333 \ -v $(pwd)/path/to/data:/qdrant/storage \ qdrant/qdrant

 3、python客户端

pip install qdrant_client

pip install openai

4、创建数据库test

from qdrant_client import QdrantClient
from qdrant_client.http.models import Distance, VectorParams
 
client = QdrantClient("127.0.0.1", port=6333)
client.recreate_collection(
    collection_name='test',
    vectors_config=VectorParams(size=1536, distance=Distance.COSINE),
)

数据文件db.txt

小明是一个小数生
小明爱游泳，经常去游泳馆
小明的妈妈很爱干净，家里一尘不染

python运行文件

from qdrant_client import QdrantClient
from qdrant_client.http.models import Distance, VectorParams
import openai, numpy as np


openai.api_key = "apikey"

def getembedding(txt):
    resp = openai.Embedding.create(
    input=[txt],
    engine="text-embedding-3-small")
    return resp['data'][0]['embedding']
    


client = QdrantClient("127.0.0.1", port=6333)
with open('db.txt', 'r', encoding='utf-8') as f:
    for index, line in enumerate(f.readlines()):
        print(line)
        embedding = getembedding(line)
        client.upsert(
            collection_name='test',
            wait=True,
            points=[
                PointStruct(id=index+1, vector=embedding, payload={"text": line}),
            ],
        )

from qdrant_client import QdrantClient
from qdrant_client.http.models import Distance, VectorParams
import openai, numpy as np


openai.api_key = "apikey"

def getembedding(txt):
    resp = openai.Embedding.create(
    input=[txt],
    engine="text-embedding-3-small")
    return resp['data'][0]['embedding']
    

client = QdrantClient("127.0.0.1", port=6333)
question = '小明的妈妈有什么习惯？'
search_results = client.search(
    collection_name='kb',
    query_vector=getembedding(question),
    limit=3,
    search_params={"exact": False, "hnsw_ef": 128}
)

https://qdrant.tech/documentation/embeddings/openai/