i am using `QueryFusionRetriever` with

At a glance

The community member is using QueryFusionRetriever with CondensePlusContextChatEngine, where they have two retrievers: BM25Retriever and VectorStoreIndex.from_vector_store, and they are using langfuse for traces. The issue they are facing is that when using the CondensePlusContextChatEngine, the traces are not well segregated like for multiple retrievers, multiple queries, and fusion nodes, as they are in the index.as_chat_engine.

The comments provide an example of how to set up the QueryFusionRetriever with multiple retrievers, but they do not provide specific information on how to segregate traces for multiple retrievers, multiple queries, and fusion nodes when using the CondensePlusContextChatEngine. The community members are advised to refer to the official LlamaIndex documentation or the source code for more detailed information on this topic.

ppayload

i am using QueryFusionRetriever with CondensePlusContextChatEngine, where i am having 2 retrievers BM25Retriever and VectorStoreIndex.from_vector_store and using langfuse for traces. When using condense plus context chat engine the traces are not well segrated like for multiple retriever, multiple queries and then fusion nodes. Just like well speperated as in index.as_chat_engine

6 comments

ppayload

Attachments

ppayload

@kapa.ai i am using QueryFusionRetriever with CondensePlusContextChatEngine, where i am having 2 retrievers BM25Retriever and VectorStoreIndex.from_vector_store and using langfuse for traces. When using condense plus context chat engine the traces are not well segrated like for multiple retriever, multiple queries and then fusion nodes. Just like well speperated as in index.as_chat_engine

kkapa.ai

Let me check the knowledge base for you. One moment! :books:

kkapa.ai

----------------------
Based on the information provided in the knowledge sources, you can use the QueryFusionRetriever with multiple retrievers like BM25Retriever and VectorStoreIndex.from_vector_store. Here is an example of how you can set it up:

Plain Text

from llama_index.core import VectorStoreIndex, StorageContext
from llama_index.core.storage.docstore import SimpleDocumentStore
from llama_index.vector_stores.chroma import ChromaVectorStore
import chromadb

docstore = SimpleDocumentStore()
docstore.add_documents(nodes)

db = chromadb.PersistentClient(path="./chroma_db")
chroma_collection = db.get_or_create_collection("dense_vectors")
vector_store = ChromaVectorStore(chroma_collection=chroma_collection)

storage_context = StorageContext.from_defaults(
    docstore=docstore, vector_store=vector_store
)

index = VectorStoreIndex(nodes=nodes, storage_context=storage_context)

Plain Text

import nest_asyncio

nest_asyncio.apply()

from llama_index.core.retrievers import QueryFusionRetriever

retriever = QueryFusionRetriever(
    [
        index.as_retriever(similarity_top_k=2),
        BM25Retriever.from_defaults(
            docstore=index.docstore, similarity_top_k=2
        ),
    ],
    num_queries=1,
    use_async=True,
)

(Source: GitHub ())

For the CondensePlusContextChatEngine, you can initialize it from default parameters as shown below:

Plain Text

from llama_index.core.chat_engine import CondensePlusContextChatEngine

chat_engine = CondensePlusContextChatEngine.from_defaults(
    retriever=retriever,
    # other parameters as needed
)

(Source: LlamaIndex Docs ())

kkapa.ai

However, the knowledge sources do not provide specific information on how to segregate traces for multiple retrievers, multiple queries, and fusion nodes when using the CondensePlusContextChatEngine. It's recommended to refer to the official LlamaIndex documentation or the source code for more detailed information on this topic.
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------

ppayload

@Logan M @WhiteFang_Jr

Add a reply

Find answers from the community

i am using `QueryFusionRetriever` with