Find answers from the community

Updated 11 months ago

Hi guys, I am using the simple fusion

At a glance

The community member is using the simple fusion retriever with Qdrant as the vector database, but is encountering an error when trying to retrieve nodes. The community members have shared a Colab notebook and discussed the issue, noting that the code works fine with the normal vector store index but not with the Qdrant vector store. They have also tried using the Weaviate index, which is working. The community members have identified the error being generated at a specific line in the code and suspect that the issue may be related to the version of Qdrant installed. They have also mentioned that Qdrant requires both a sync and async client, and that if using :memory:, those clients do not share data. The community members are unsure if this will be the same case when using their own Qdrant instance.

Useful resources
Hi guys, I am using the simple fusion retriever with Qdrant as vectorDB. While trying to run - nodes_with_scores = retriever.retrieve("How do I setup a chroma vector store?"), getting this error -
Attachment
Screenshot_2024-03-22_at_12.30.53_PM.png
W
P
L
17 comments
would you mind sharing the code?
its working fine when using the normal vectorstoreindex
but not working while using qdrant vector store
its even working with weaviate index
Also, you have added openAI key
still getting the same error
can you tried running it on your side?
@WhiteFang_Jr @dev_advocate any update on this?
@Logan M any comments on this issue?
qdrant is pretty janky with async

Set use_async=False, in the retriever
qdrant requires both a sync and async client

QdrantVectorStore(client=QdrantClient(..), aclient=AsyncQdrantClient(...))

But if using :memory: , those clients do not share data
will it be the same case if I am using my own qdrant instance?
at least then you can provide both a client and aclient, and they will access the same data
Add a reply
Sign up and join the conversation on Discord