Weaviate中BM25及各类搜索函数无法返回预期结果求助

问题描述 #

已在Weaviate中创建集合，并通过LlamaIndex将若干文档导入Weaviate数据库。使用默认搜索时始终返回错误文档；测试BM25搜索时，即便复制目标文档中的完整短语，高分结果仍为其他文档。

服务器配置信息 #

• Weaviate服务器版本：1.24.10
• 部署方式：Docker
• LlamaIndex版本：0.10.42

文档准备 #

• 目标文档：某IEEE论文PDF（本地存储），另有20份文档一同导入用于检索测试。

Python配置信息 #

导入语句 #

# Weaviate相关
import weaviate
from weaviate.classes.config import Configure, VectorDistances, Property, DataType
from weaviate.util import generate_uuid5
from weaviate.query import MetadataQuery

# LlamaIndex相关
from llama_index.core import SimpleDirectoryReader, VectorStoreIndex, Settings, StorageContext
from llama_index.vector_stores.weaviate import WeaviateVectorStore
from llama_index.core.node_parser import SentenceSplitter

基于Weaviate创建索引代码 #

# 创建Weaviate集合
def create_collection(client, collection_name):
  client.collections.create(
    collection_name,
    vectorizer_config=Configure.Vectorizer.text2vec_transformers(),
    vector_index_config=Configure.VectorIndex.hnsw(distance_metric=VectorDistances.COSINE),
    reranker_config=Configure.Reranker.transformers(),
    inverted_index_config=Configure.inverted_index(
      bm25_b=0.7,
      bm25_k1=1.25,
      index_null_state=True,
      index_property_length=True,
      index_timestamps=True
    ),
  )
 
# 使用LlamaIndex创建索引
def create_weaviate_index(client, index_name, doc_folder):
  create_collection(client, index_name)
  vector_store = WeaviateVectorStore(weaviate_client=client, index_name=index_name, text_key="content")
  storage_context = StorageContext.from_defaults(vector_store=vector_store)
  index = VectorStoreIndex.from_documents([], storage_context=storage_context)
  documents = SimpleDirectoryReader(input_dir=doc_folder)
  nodes = node_parser.get_nodes_from_documents(documents)
  index.insert_nodes(nodes)
  return index

client = weaviate.connect_to_local()
index_name = "LlamaIndex"
doc_folder = "/path/to/doc_folder"
create_weaviate_index(client, index_name, doc_folder)

文档查询测试 #

使用LlamaIndex查询 #

query_engine = index.as_query_engine()
question = "EMA was created in 2001 to?" # 取自目标文档中的片段
response = query_engine.query(question)
print(response)

for node in response.source_nodes:
  print(node.metadata) # 未返回包含该片段的目标文档

使用Weaviate混合搜索（alpha=0） #

collection = client.collections.get("LlamaIndex")
question = "EMA was created in 2001 to?" # 取自目标文档中的片段
query_vector = embed_model.get_query_embedding(question)

response = collection.query.hybrid(
  query=question,
  vector=query_vector,
  limit=5,
  alpha=0,

  return_metadata=MetadataQuery(
    distance=True,
    certainty=True,
    score=True,
    explain_score=True
  )
)

for obj in response.objects:
  print(f"METADATA: {obj.metadata}") # 未返回包含该片段的目标文档

使用Weaviate BM25搜索 #

collection = client.collections.get("LlamaIndex")
question = "EMA was created in 2001 to?" # 取自目标文档中的片段
response = collection.query.bm25(
  query=question,
  limit=5,

  return_metadata=MetadataQuery(
    distance=True,
    certainty=True,
    score=True,
    explain_score=True
  )
)

for obj in response.objects:
  print(f"METADATA: {obj.metadata}") # 未返回包含该片段的目标文档

使用Weaviate近文本搜索 #

collection = client.collections.get("LlamaIndex")
question = "EMA was created in 2001 to?" # 取自目标文档中的片段
response = collection.query.near_text(
  query=question,
  limit=5,

  return_metadata=MetadataQuery(
    distance=True,
    certainty=True,
    score=True,
    explain_score=True
  )
)

for obj in response.objects:
  print(f"METADATA: {obj.metadata}") # 未返回包含该片段的目标文档

内容的提问来源于stack exchange，提问作者SoftwearEnginear