Find answers from the community

Updated 10 months ago

Opensearch

vector_store = PineconeVectorStore(pinecone.Index("quickstart"))
index = VectorStoreIndex.from_vector_store(vector_store=vector_store)

Is there anything like this for Opensearch
L
N
W
12 comments
I referred to the mentioned link and I am able to view the embeddings present in opensearch index.
Can you please help me how do I query the opensearch index which already has embeddings stored in it
The same way you had above

index =VectorStoreIndex.from_vector_store(vector_store)
(Assuming llamaindex initially created the index)
Is it possible to query just by passing the opensearch index instead of connecting to opensearchclient repeatedly for querying
Similar to the 1st code I have sent, only pinecone index is being passed and no connectivity
I mean, it still has to connect to pinecone in order to query? Not sure what you mean
Once we connect to opensearch we can create our index and query. But every time I connect to opensearch the embeddings are getting created and then the indexing and querying happens. So my question is how to pass the index in opensearch directly as it already has embeddings generated instead of embeddings getting generated every time I connect to opensearch.
I just want to prevent opensearch from creating embeddings everytime I connect as my index in opensearch already have embeddings
Are you passing documents while creating the index?
If so then it will create embeddings for the passed documents.
Add a reply
Sign up and join the conversation on Discord