Qdrant官方快速入门和教程简化版
说明：
- 首次发表日期：2024-08-28
- Qdrant官方文档：https://qdrant.tech/documentation/
关于
阅读Qdrant一小部分的官方文档，并使用中文简化记录下，更多请阅读官方文档。
使用Docker本地部署Qdrant
docker pull qdrant/qdrant
docker run -d -p 6333:6333 -p 6334:6334 \
-v $(pwd)/qdrant_storage:/qdrant/storage:z \
qdrant/qdrant
默认配置下，所有的数据存储在./qdrant_storage。
快速入门
安装qdrant-client包（python）：
pip install qdrant-client
初始化客户端：
from qdrant_client import QdrantClient
client = QdrantClient(url="http://localhost:6333")
所有的向量数据（vector data）都存储在Qdrant Collection上。创建一个名为test_collection的collection，该collection使用dot product作为比较向量的指标。
from qdrant_client.models import Distance, VectorParams
client.create_collection(
collection_name="test_collection",
vectors_config=VectorParams(size=4, distance=Distance.DOT),
)
添加带payload的向量。payload是与向量相关联的数据。
from qdrant_client.models import PointStruct
operation_info = client.upsert(
collection_name="test_collection",
wait=True,
points=[
PointStruct(id=1, vector=[0.05, 0.61, 0.76, 0.74], payload={"city": "Berlin"}),
PointStruct(id=2, vector=[0.19, 0.81, 0.75, 0.11], payload={"city": "London"}),
PointStruct(id=3, vector=[0.36, 0.55, 0.47, 0.94], payload={"city": "Moscow"}),
PointStruct(id=4, vector=[0.18, 0.01, 0.85, 0.80], payload={"city": "New York"}),
PointStruct(id=5, vector=[0.24, 0.18, 0.22, 0.44], payload={"city": "Beijing"}),
PointStruct(id=6, vector=[0.35, 0.08, 0.11, 0.44], payload={"city": "Mumbai"}),
]
)
print(operation_info)
运行一个查询：
search_result = client.query_points(
collection_name="test_collection", query=[0.2, 0.1, 0.9, 0.7], limit=3
).points
print(search_result)
输出：
[
{
"id": 4,
"version": 0,
"score": 1.362,
"payload": null,
"vector": null
},
{
"id": 1,
"version": 0,
"score": 1.273,
"payload": null,
"vector": null
},
{
"id": 3,
"version": 0,
"score": 1.208,
"payload": null,
"vector": null
}
]
添加一个过滤器：
from qdrant_client.models import Filter, FieldCondition, MatchValue
search_result = client.query_points(
collection_name="test_collection",
query=[0.2, 0.1, 0.9, 0.7],
query_filter=Filter(
must=[FieldCondition(key="city", match=MatchValue(value="London"))]
),
with_payload=True,
limit=3,
).points
print(search_result)
输出：
[
{
"id": 2,
"version": 0,
"score": 0.871,
"payload": {
"city": "London"
},
"vector": null
}
]
教程
语义搜索入门
安装依赖：
pip install sentence-transformers
导入模块：
from qdrant_client import models, QdrantClient
from sentence_transformers import SentenceTransformer
使用all-MiniLM-L6-v2编码器作为embedding模型，embedding模型可以将raw data转化为embeddings）
encoder = SentenceTransformer("all-MiniLM-L6-v2")
添加数据集：
documents = [
{
"name": "The Time Machine",
"description": "A man travels through time and witnesses the evolution of humanity.",
"author": "H.G. Wells",
"year": 1895,
},
{
"name": "Ender's Game",
"description": "A young boy is trained to become a military leader in a war against an alien race.",
"author": "Orson Scott Card",
"year": 1985,
},
{
"name": "Brave New World",
"description": "A dystopian society where people are genetically engineered and conditioned to conform to a strict social hierarchy.",
"author": "Aldous Huxley",
"year": 1932,
},
{
"name": "The Hitchhiker's Guide to the Galaxy",
"description": "A comedic science fiction series following the misadventures of an unwitting human and his alien friend.",
"author": "Douglas Adams",
"year": 1979,
},
{
"name": "Dune",
"description": "A desert planet is the site of political intrigue and power struggles.",
"author": "Frank Herbert",
"year": 1965,
},
{
"name": "Foundation",
"description": "A mathematician develops a science to predict the future of humanity and works to save civilization from collapse.",
"author": "Isaac Asimov",
"year": 1951,
},
{
"name": "Snow Crash",
"description": "A futuristic world where the internet has evolved into a virtual reality metaverse.",
"author": "Neal Stephenson",
"year": 1992,
},
{
"name": "Neuromancer",
"description": "A hacker is hired to pull off a near-impossible hack and gets pulled into a web of intrigue.",
"author": "William Gibson",
"year": 1984,
},
{
"name": "The War of the Worlds",
"description": "A Martian invasion of Earth throws humanity into chaos.",
"author": "H.G. Wells",
"year": 1898,
},
{
"name": "The Hunger Games",
"description": "A dystopian society where teenagers are forced to fight to the death in a televised spectacle.",
"author": "Suzanne Collins",
"year": 2008,
},
{
"name": "The Andromeda Strain",
"description": "A deadly virus from outer space threatens to wipe out humanity.",
"author": "Michael Crichton",
"year": 1969,
},
{
"name": "The Left Hand of Darkness",
"description": "A human ambassador is sent to a planet where the inhabitants are genderless and can change gender at will.",
"author": "Ursula K. Le Guin",
"year": 1969,
},
{
"name": "The Three-Body Problem",
"description": "Humans encounter an alien civilization that lives in a dying system.",
"author": "Liu Cixin",
"year": 2008,
},
]
将embedding数据存储在内存中：
client = QdrantClient(":memory:")
创建一个collection：
client.create_collection(
collection_name="my_books",
vectors_config=models.VectorParams(
size=encoder.get_sentence_embedding_dimension(), # Vector size is defined by used model
distance=models.Distance.COSINE,
),
)
上传数据：
client.upload_points(
collection_name="my_books",
points=[
models.PointStruct(
id=idx, vector=encoder.encode(doc["description"]).tolist(), payload=doc
)
for idx, doc in enumerate(documents)
],
)
问一个问题：
hits = client.query_points(
collection_name="my_books",
query=encoder.encode("alien invasion").tolist(),
limit=3,
).points
for hit in hits:
print(hit.payload, "score:", hit.score)
输出：
{'name': 'The War of the Worlds', 'description': 'A Martian invasion of Earth throws humanity into chaos.', 'author': 'H.G. Wells', 'year': 1898} score: 0.570093257022374
{'name': "The Hitchhiker's Guide to the Galaxy", 'description': 'A comedic science fiction series following the misadventures of an unwitting human and his alien friend.', 'author': 'Douglas Adams', 'year': 1979} score: 0.5040468703143637
{'name': 'The Three-Body Problem', 'description': 'Humans encounter an alien civilization that lives in a dying system.', 'author': 'Liu Cixin', 'year': 2008} score: 0.45902943411768216
过滤以便缩窄查询：
hits = client.query_points(
collection_name="my_books",
query=encoder.encode("alien invasion").tolist(),
query_filter=models.Filter(
must=[models.FieldCondition(key="year", range=models.Range(gte=2000))]
),
limit=1,
).points
for hit in hits:
print(hit.payload, "score:", hit.score)
输出：
{'name': 'The Three-Body Problem', 'description': 'Humans encounter an alien civilization that lives in a dying system.', 'author': 'Liu Cixin', 'year': 2008} score: 0.45902943411768216
简单的神经搜索
下载样本数据集：
wget https://storage.googleapis.com/generall-shared-data/startups_demo.json
安装SentenceTransformer等依赖库：
pip install sentence-transformers numpy pandas tqdm
导入模块：
from sentence_transformers import SentenceTransformer
import numpy as np
import json
import pandas as pd
from tqdm.notebook import tqdm
创建sentence encoder：
model = SentenceTransformer(
"all-MiniLM-L6-v2", device="cuda"
) # or device="cpu" if you don't have a GPU
读取数据：
df = pd.read_json("./startups_demo.json", lines=True)
为每一个description创建embedding向量。encode内部会将输入切分为一个个batch，以便提高处理速度。
vectors = model.encode(
[row.alt + ". " + row.description for row in df.itertuples()],
show_progress_bar=True,
)
vectors.shape
# > (40474, 384)
保存为npy文件：
np.save("startup_vectors.npy", vectors, allow_pickle=False)
启动docker服务
docker pull qdrant/qdrant
docker run -p 6333:6333 \
-v $(pwd)/qdrant_storage:/qdrant/storage \
qdrant/qdrant
创建Qdrant客户端
# Import client library
from qdrant_client import QdrantClient
from qdrant_client.models import VectorParams, Distance
client = QdrantClient("http://localhost:6333")
创建collection，其中384是embedding模型（all-MiniLM-L6-v2）的输出维度。
if not client.collection_exists("startups"):
client.create_collection(
collection_name="startups",
vectors_config=VectorParams(size=384, distance=Distance.COSINE),
)
加载数据
fd = open("./startups_demo.json")
# payload is now an iterator over startup data
payload = map(json.loads, fd)
# Load all vectors into memory, numpy array works as iterable for itself.
# Other option would be to use Mmap, if you don't want to load all data into RAM
vectors = np.load("./startup_vectors.npy")
上传数据到Qdrant
client.upload_collection(
collection_name="startups",
vectors=vectors,
payload=payload,
ids=None, # Vector ids will be assigned automatically
batch_size=256, # How many vectors will be uploaded in a single request?
)
创建neural_searcher.py文件：
from qdrant_client import QdrantClient
from sentence_transformers import SentenceTransformer
class NeuralSearcher:
def __init__(self, collection_name):
self.collection_name = collection_name
# Initialize encoder model
self.model = SentenceTransformer("all-MiniLM-L6-v2", device="cpu")
# initializa Qdrant client
self.qdrant_client = QdrantClient("http://localhost:6333")
def search(self, text:str):
# Convert text query into vector
vector = self.model.encode(text).tolist()
# Use `vector` for search for closet vectors in the collection
search_result = self.qdrant_client.search(
collection_name=self.collection_name,
query_vector=vector,
query_filter=None, # If you don't want any filters for now
limit=5, # 5 the most closet results is enough
)
# `search_result` contains found vector ids with similarity scores along with stored payload
# In this function you are interested in payload only
payloads = [hit.payload for hit in search_result]
return payloads
使用FastAPI部署：
pip install fastapi uvicorn
from qdrant_client import QdrantClient
from qdrant_client.models import Filter
from sentence_transformers import SentenceTransformer
class NeuralSearcher:
def __init__(self, collection_name):
self.collection_name = collection_name
# Initialize encoder model
self.model = SentenceTransformer("all-MiniLM-L6-v2", device="cpu")
# initializa Qdrant client
self.qdrant_client = QdrantClient("http://localhost:6333")
def search(self, text:str):
# Convert text query into vector
vector = self.model.encode(text).tolist()
# Use `vector` for search for closet vectors in the collection
search_result = self.qdrant_client.search(
collection_name=self.collection_name,
query_vector=vector,
query_filter=None, # If you don't want any filters for now
limit=5, # 5 the most closet results is enough
)
# `search_result` contains found vector ids with similarity scores along with stored payload
# In this function you are interested in payload only
payloads = [hit.payload for hit in search_result]
return payloads
def search_in_berlin(self, text:str):
# Convert text query into vector
vector = self.model.encode(text).tolist()
city_of_interest = "Berlin"
# Define a filter for cities
city_filter = Filter(**{
"must": [{
"key": "city", # Store city information in a field of the same name
"match": { # This condition checks if payload field has the requested value
"value": city_of_interest
}
}]
})
# Use `vector` for search for closet vectors in the collection
search_result = self.qdrant_client.query_points(
collection_name=self.collection_name,
query=vector,
query_filter=city_filter,
limit=5,
).points
# `search_result` contains found vector ids with similarity scores along with stored payload
# In this function you are interested in payload only
payloads = [hit.payload for hit in search_result]
return payloads
from fastapi import FastAPI
app = FastAPI()
# Create a neural searcher instance
neural_searcher = NeuralSearcher(collection_name="startups")
@app.get("/api/search")
def search_startup(q: str):
return {"result": neural_searcher.search(text=q)}
@app.get("/api/search_in_berlin")
def search_startup_filter(q: str):
return {"result": neural_searcher.search_in_berlin(text=q)}
if __name__ == "__main__":
import uvicorn
uvicorn.run(app, host="0.0.0.0", port=8001)
如果是在jupyter notebook中运行，则需要添加
import nest_asyncio
nest_asyncio.apply()
安装nest_asyncio：
pip install nest_asyncio
异步使用Qdrant
Qdrant原生支持async
from qdrant_client import models
import qdrant_client
import asyncio
async def main():
client = qdrant_client.AsyncQdrantClient("localhost")
# Create a collection
await client.create_collection(
collection_name="my_collection",
vectors_config=models.VectorParams(size=4, distance=models.Distance.COSINE),
)
# Insert a vector
await client.upsert(
collection_name="my_collection",
points=[
models.PointStruct(
id="5c56c793-69f3-4fbf-87e6-c4bf54c28c26",
payload={
"color": "red",
},
vector=[0.9, 0.1, 0.1, 0.5],
),
],
)
# Search for nearest neighbors
points = await client.query_points(
collection_name="my_collection",
query=[0.9, 0.1, 0.1, 0.5],
limit=2,
).points
# Your async code using AsyncQdrantClient might be put here
# ...
asyncio.run(main())
917
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
