Sure:

DDrSebastianK

Sure:
vector_store_pc = LanceDBVectorStore(uri=s3_uri)
index_pc = VectorStoreIndex.from_vector_store(vector_store_pc, service_context=service_context_pc, storage_context=storage_context_pc)
test = index_pc.as_retriever(similarity_top_k=5)
retriever = QueryFusionRetriever(
retrievers=[test],
llm=gpt4,
query_gen_prompt=query_gen_prompt_str,
num_queries=3,
mode="reciprocal_rerank",
use_async=True,
verbose=True,
)
query_engine = RetrieverQueryEngine(retriever=retriever, response_synthesizer=response_synthesizer_pc)
result = query_engine.query(query)

7 comments

kkaushal

put the uri in double quotes

DDrSebastianK

uri is defined as a string elsewhere: s3_uri="s3://myfolder/lancedb"

DDrSebastianK

Standalone retriever works both sync and async perfectly.

DDrSebastianK

@Logan M can you give me an idea where to look for the source of this error?

LLogan M

sorry, whats the error here?

DDrSebastianK

Did anyone face similar problems with lancedb? When using it in a simple query engine or a standalone retriever it works, but when I plug it into a query fusion retriever it just throws this error always.
/usr/local/lib/python3.10/dist-packages/llama_index/vector_stores/lancedb.py in query(self, query, **kwargs)
159 .limit(query.similarity_top_k)
160 .where(where)
--> 161 .nprobes(self.nprobes)
162 )
163

AttributeError: 'LanceEmptyQueryBuilder' object has no attribute 'nprobes'

LLogan M

ah I see. I have no idea why that error is happening... maybe related to your lancedb version? maybe they updated the package?

But the source code is here
https://github.com/run-llama/llama_index/blob/66ec4aa5ef3abfedf967be1c3184b06d300cc613/llama_index/vector_stores/lancedb.py#L157

Add a reply

Find answers from the community

Sure: