随着大数据时代的到来，信息检索技术在各个领域中扮演着越来越重要的角色。阿里云向量检索服务Milvus 版作为一款高性能的向量检索引擎，100%兼容开源Milvus，凭借其开箱即用、灵活扩展和全链路告警能力，成为企业大规模AI向量数据相似性检索服务的理想选择。其最新版本 2.5 在全文检索、关键词匹配以及混合检索（Hybrid Search）方面实现了显著的增强，在多模态检索、RAG等多场景中检索结果能够兼顾召回率与精确性。本文将详细介绍如何利用 Milvus 2.5 版本实现这些功能，并阐述其在RAG 应用的 Retrieve 阶段的最佳实践。
背景信息
Milvus 2.5 集成了高性能搜索引擎库 Tantivy，并内置 Sparse-BM25 算法，首次实现了原生全文检索功能。这一能力与现有的语义搜索功能完美互补，为用户提供更强大的检索体验。
-
内置分词器：无需额外预处理，通过内置分词器（Analyzer）与稀疏向量提取能力，Milvus 可直接接受文本输入，自动完成分词、停用词过滤与稀疏向量提取。
-
实时 BM25 统计：数据插入时动态更新词频（TF）与逆文档频率（IDF），确保搜索结果的实时性与准确性。
-
混合搜索性能增强：基于近似最近邻（ANN）算法的稀疏向量检索，性能远超传统关键词系统，支持亿级数据毫秒级响应，同时兼容与稠密向量的混合查询。
前提条件
使用限制
您可以执行以下命令来检查当前安装的版本。
pip3 show pymilvus
如果版本低于2.5，请使用以下命令更新。
pip3 install --upgrade pymilvus
操作流程
步骤一：安装依赖库
pip3 install pymilvus langchain dashscope
步骤二：数据准备
本文以 Milvus 官方文档作为示例，通过 LangChain SDK 切分文本，作为 Embedding 模型 text-embedding-v2 的输入，并将 Embedding 的结果和原始文本一起插入到 Milvus 中。
from langchain_community.document_loaders import WebBaseLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.embeddings import DashScopeEmbeddings
from pymilvus import MilvusClient, DataType, Function, FunctionType
dashscope_api_key = "<YOUR_DASHSCOPE_API_KEY>"
milvus_url = "<YOUR_MMILVUS_URL>"
user_name = "root"
password = "<YOUR_PASSWORD>"
collection_name = "milvus_overview"
dense_dim = 1536
loader = WebBaseLoader([
'https://raw.githubusercontent.com/milvus-io/milvus-docs/refs/heads/v2.5.x/site/en/about/overview.md'
])
docs = loader.load()
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=256)
# 使用LangChain将输入文档安照chunk_size切分
all_splits = text_splitter.split_documents(docs)
embeddings = DashScopeEmbeddings(
model="text-embedding-v2", dashscope_api_key=dashscope_api_key
)
text_contents = [doc.page_content for doc in all_splits]
vectors = embeddings.embed_documents(text_contents)
client = MilvusClient(
uri=f"http://{milvus_url}:19530",
token=f"{user_name}:{password}",
)
schema = MilvusClient.create_schema(
enable_dynamic_field=True,
)
analyzer_params = {
"type": "english"
}
# Add fields to schema
schema.add_field(field_name="id", datatype=DataType.INT64, is_primary=True, auto_id=True)
schema.add_field(field_name="text", datatype=DataType.VARCHAR, max_length=65535, enable_analyzer=True, analyzer_params=analyzer_params, enable_match=True)
schema.add_field(field_name="sparse_bm25", datatype=DataType.SPARSE_FLOAT_VECTOR)
schema.add_field(field_name="dense", datatype=DataType.FLOAT_VECTOR, dim=dense_dim)
bm25_function = Function(
name="bm25",
function_type=FunctionType.BM25,
input_field_names=["text"],
output_field_names="sparse_bm25",
)
schema.add_function(bm25_function)
index_params = client.prepare_index_params()
# Add indexes
index_params.add_index(
field_name="dense",
index_name="dense_index",
index_type="IVF_FLAT",
metric_type="IP",
params={"nlist": 128},
)
index_params.add_index(
field_name="sparse_bm25",
index_name="sparse_bm25_index",
index_type="SPARSE_WAND",
metric_type="BM25"
)
# Create collection
client.create_collection(
collection_name=collection_name,
schema=schema,
index_params=index_params
)
data = [
{"dense": vectors[idx], "text": doc}
for idx, doc in enumerate(text_contents)
]
# Insert data
res = client.insert(
collection_name=collection_name,
data=data
)
print(f"生成 {len(vectors)} 个向量，维度：{len(vectors[0])}")
本文示例涉及以下参数，请您根据实际环境替换。
![]()
该示例使用了Milvus 2.5最新的能力，通过创建 bm25_function 对象，Milvus就可以自动地将文本列转换为稀疏向量。
![]()
同样，在处理中文文档时，Milvus 2.5版本也支持指定相应的中文分析器。
重要
在Schema中完成Analyzer的设置后，该设置将对该Collections永久生效。如需设置新的Analyzer，则必须重新创建Collection。
# 定义分词器参数
analyzer_params = {
"type": "chinese" # 指定分词器类型为中文
}
# 添加文本字段到 Schema，并启用分词器
schema.add_field(
field_name="text", # 字段名称
datatype=DataType.VARCHAR, # 数据类型：字符串（VARCHAR）
max_length=65535, # 最大长度：65535 字符
enable_analyzer=True, # 启用分词器
analyzer_params=analyzer_params # 分词器参数
)
步骤三：全文检索
在 Milvus 2.5 版本中，您可以很方便地通过相关 API 使用最新的全文检索能力。代码示例如下所示。
from pymilvus import MilvusClient
# 创建Milvus Client。
client = MilvusClient(
uri="http://c-xxxx.milvus.aliyuncs.com:19530", # Milvus实例的公网地址。
token="<yourUsername>:<yourPassword>", # 登录Milvus实例的用户名和密码。
db_name="default" # 待连接的数据库名称，本文示例为默认的default。
)
search_params = {
'params': {'drop_ratio_search': 0.2},
}
full_text_search_res = client.search(
collection_name='milvus_overview',
data=['what makes milvus so fast?'],
anns_field='sparse_bm25',
limit=3,
search_params=search_params,
output_fields=["text"],
)
for hits in full_text_search_res:
for hit in hits:
print(hit)
print("\n")
"""
{'id': 456165042536597485, 'distance': 6.128782272338867, 'entity': {'text': '## What Makes Milvus so Fast？\n\nMilvus was designed from day one to be a highly efficient vector database system. In most cases, Milvus outperforms other vector databases by 2-5x (see the VectorDBBench results). This high performance is the result of several key design decisions:\n\n**Hardware-aware Optimization**: To accommodate Milvus in various hardware environments, we have optimized its performance specifically for many hardware architectures and platforms, including AVX512, SIMD, GPUs, and NVMe SSD.\n\n**Advanced Search Algorithms**: Milvus supports a wide range of in-memory and on-disk indexing/search algorithms, including IVF, HNSW, DiskANN, and more, all of which have been deeply optimized. Compared to popular implementations like FAISS and HNSWLib, Milvus delivers 30%-70% better performance.'}}
{'id': 456165042536597487, 'distance': 4.760214805603027, 'entity': {'text': "## What Makes Milvus so Scalable\n\nIn 2022, Milvus supported billion-scale vectors, and in 2023, it scaled up to tens of billions with consistent stability, powering large-scale scenarios for over 300 major enterprises, including Salesforce, PayPal, Shopee, Airbnb, eBay, NVIDIA, IBM, AT&T, LINE, ROBLOX, Inflection, etc.\n\nMilvus's cloud-native and highly decoupled system architecture ensures that the system can continuously expand as data grows:\n\n"}}
"""
步骤四：关键词匹配
关键词匹配是Milvus 2.5所提供的一项全新功能，该功能可以与向量相似性搜索相结合，从而缩小搜索范围并提高搜索性能。如果您希望使用关键词检索功能，则在定义模式时需要将enable_analyzer和enable_match 同时设置为True。
重要
开启 enable_match 会为该字段创建倒排索引，这将消耗额外的存储资源。
示例1：结合向量搜索的关键词匹配
在此代码示例片段中，我们使用过滤表达式限制搜索结果仅包含与指定词语 "query" 和 "node" 匹配的文档。之后，向量相似性搜索会在已过滤的文档子集上进行。
filter = "TEXT_MATCH(text, 'query') and TEXT_MATCH(text, 'node')"
text_match_res = client.search(
collection_name="milvus_overview",
anns_field="dense",
data=query_embeddings,
filter=filter,
search_params={"params": {"nprobe": 10}},
limit=2,
output_fields=["text"]
)
示例2：标量过滤查询
关键词匹配还可以用于查询操作中的标量过滤。通过在query() 中指定TEXT_MATCH 表达式，您可以检索与给定词语匹配的文档。在这个代码示例片段中，过滤表达式将搜索结果限制为仅包含与 "scalable" 或 "fast" 匹配的文档。
filter = "TEXT_MATCH(text, 'scalable fast')"
text_match_res = client.query(
collection_name="milvus_overview",
filter=filter,
output_fields=["text"]
)
步骤五：混合检索与RAG
结合向量搜索和全文检索，通过 RRF（Reciprocal Rank Fusion） 算法融合向量和文本检索结果，重新优化排序和权重分配，提升数据召回率和精确性。
![]()
代码示例如下所示。
from pymilvus import MilvusClient
from pymilvus import AnnSearchRequest, RRFRanker
from langchain_community.embeddings import DashScopeEmbeddings
from dashscope import Generation
# 创建Milvus Client。
client = MilvusClient(
uri="http://c-xxxx.milvus.aliyuncs.com:19530", # Milvus实例的公网地址。
token="<yourUsername>:<yourPassword>", # 登录Milvus实例的用户名和密码。
db_name="default" # 待连接的数据库名称，本文示例为默认的default。
)
collection_name = "milvus_overview"
# 替换为您的 DashScope API-KEY
dashscope_api_key = "<YOUR_DASHSCOPE_API_KEY>"
# 初始化 Embedding 模型
embeddings = DashScopeEmbeddings(
model="text-embedding-v2", # 使用text-embedding-v2模型。
dashscope_api_key=dashscope_api_key
)
# Define the query
query = "Why does Milvus run so scalable?"
# Embed the query and generate the corresponding vector representation
query_embeddings = embeddings.embed_documents([query])
# Set the top K result count
top_k = 5 # Get the top 5 docs related to the query
# Define the parameters for the dense vector search
search_params_dense = {
"metric_type": "IP",
"params": {"nprobe": 2}
}
# Create a dense vector search request
request_dense = AnnSearchRequest([query_embeddings[0]], "dense", search_params_dense, limit=top_k)
# Define the parameters for the BM25 text search
search_params_bm25 = {
"metric_type": "BM25"
}
# Create a BM25 text search request
request_bm25 = AnnSearchRequest([query], "sparse_bm25", search_params_bm25, limit=top_k)
# Combine the two requests
reqs = [request_dense, request_bm25]
# Initialize the RRF ranking algorithm
ranker = RRFRanker(100)
# Perform the hybrid search
hybrid_search_res = client.hybrid_search(
collection_name=collection_name,
reqs=reqs,
ranker=ranker,
limit=top_k,
output_fields=["text"]
)
# Extract the context from hybrid search results
context = []
print("Top K Results:")
for hits in hybrid_search_res: # Use the correct variable here
for hit in hits:
context.append(hit['entity']['text']) # Extract text content to the context list
print(hit['entity']['text']) # Output each retrieved document
# Define a function to get an answer based on the query and context
def getAnswer(query, context):
prompt = f'''Please answer my question based on the content within:
```
{context}
```
My question is: {query}.
'''
# Call the generation module to get an answer
rsp = Generation.call(model='qwen-turbo', prompt=prompt)
return rsp.output.text
# Get the answer
answer = getAnswer(query, context)
print(answer)
# Expected output excerpt
"""
Milvus is highly scalable due to its cloud-native and highly decoupled system architecture. This architecture allows the system to continuously expand as data grows. Additionally, Milvus supports three deployment modes that cover a wide...
"""
