WARNING Could not run function search_knowledge_base(query=credit card welcome kit contents)
ERROR Retriever is not of type BaseRetriever: (VectorStoreRetriever(tags=[‘QdrantVectorStore’, ‘HuggingFaceEmbeddings’], vectorstore=<langchain_qdrant.qdrant.QdrantVectorStore object
at 0x7612692384c0>, search_kwargs={‘k’: 4, ‘filter’: Filter(should=None, min_should=None, must=[FieldCondition(key=‘metadata.language’, match=MatchValue(value=‘english’),
range=None, geo_bounding_box=None, geo_radius=None, geo_polygon=None, values_count=None)], must_not=None)}),)
Traceback (most recent call last):
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/phi/tools/function.py”, line 319, in execute
self.result = self.function.entrypoint(**entrypoint_args, **self.arguments)
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/pydantic/_internal/_validate_call.py”, line 38, in wrapper_function
return wrapper(*args, **kwargs)
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/pydantic/_internal/_validate_call.py”, line 111, in call
res = self.pydantic_validator.validate_python(pydantic_core.ArgsKwargs(args, kwargs))
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/phi/agent/agent.py”, line 2575, in search_knowledge_base
docs_from_knowledge = self.get_relevant_docs_from_knowledge(query=query)
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/phi/agent/agent.py”, line 1088, in get_relevant_docs_from_knowledge
relevant_docs: List[Document] = self.knowledge.search(query=query, num_documents=num_documents, **kwargs)
File “/home/sai-chand/miniconda3/envs/llm/lib/python3.10/site-packages/phi/knowledge/langchain.py”, line 42, in search
raise ValueError(f"Retriever is not of type BaseRetriever: {self.retriever}")
ValueError: Retriever is not of type BaseRetriever: (VectorStoreRetriever(tags=[‘QdrantVectorStore’, ‘HuggingFaceEmbeddings’],
vectorstore=<langchain_qdrant.qdrant.QdrantVectorStore object at 0x7612692384c0>, search_kwargs={‘k’: 4, ‘filter’: Filter(should=None, min_should=None,
must=[FieldCondition(key=‘metadata.language’, match=MatchValue(value=‘english’), range=None, geo_bounding_box=None, geo_radius=None, geo_polygon=None, values_count=None)],
must_not=None)}),)
i have a retriver as
retriever= db.as_retriever(
search_kwargs={
“k”: TARGET_SOURCE_CHUNKS,
“filter”: qdrant_models.Filter(
must=[
qdrant_models.FieldCondition(
key=“metadata.language”,
match=qdrant_models.MatchValue(value=language),
)
]
),
}
),
-*- Create a knowledge base from the vector store
knowledge_base = LangChainKnowledgeBase(retriever=retriever)