Implementing Advanced Fusion Retriever with Storage Context

[bhavya-giri]

I want to try Advance Fusion Retriever with Milvus Vector Store with already indexed data
https://docs.llamaindex.ai/en/stable/examples/low_level/fusion_retriever.html

[dosubot[bot]]

🤖

Hello @bhavya-giri! 👋 I'm Dosu, your friendly bot assistant here. I'm designed to help you with bugs, answer your queries, and guide you on your journey to becoming a contributor while we wait for a human maintainer. Always ready to lend a helping hand! 🤖 ✋

To implement the Advanced Fusion Retriever with the Milvus Vector Store using already indexed data in LlamaIndex, you would need to create an instance of the MilvusVectorStore class and use its query method to retrieve the most similar nodes to a given query. Here is an example of how you can do this:

from llama_index.vector_stores.milvus import MilvusVectorStore
from llama_index.vector_stores.types import VectorStoreQuery

# Create an instance of the MilvusVectorStore
milvus_store = MilvusVectorStore(
    uri="http://localhost:19530",  # replace with your Milvus instance URI
    collection_name="llamalection",  # replace with your collection name
    dim=128,  # replace with the dimension of your embeddings
    similarity_metric="IP",  # replace with your similarity metric
    overwrite=False,  # set to True if you want to overwrite existing collection
)

# Create a VectorStoreQuery
query = VectorStoreQuery(
    query_embedding=[0.1, 0.2, 0.3, ...],  # replace with your query embedding
    similarity_top_k=10,  # replace with the number of similar nodes you want to retrieve
)

# Query the MilvusVectorStore
result = milvus_store.query(query)

# The result is a VectorStoreQueryResult object
# You can access the most similar nodes, their similarities, and their ids with:
nodes = result.nodes
similarities = result.similarities
ids = result.ids

Please replace the placeholders in the code with your actual values. The query_embedding should be a list of floats representing the embedding of your query. The similarity_top_k is the number of most similar nodes you want to retrieve. The uri, collection_name, dim, and similarity_metric should match the configuration of your Milvus instance and your collection.

Please note that this code assumes that you have already indexed your data in the Milvus Vector Store. If you haven't, you would need to use the add method of the MilvusVectorStore to add your data to the store before you can query it. You can find more details about this in the llama_index/vector_stores/milvus.py file.

The Advanced Fusion Retriever in the context of LlamaIndex is a class named QueryFusionRetriever that extends the BaseRetriever class. It is designed to handle multiple retrievers and apply a fusion method to the results they return. The QueryFusionRetriever interacts with the Milvus Vector Store through the retrievers passed to it during initialization. These retrievers could be any class that extends the BaseRetriever class, including those that interact with the Milvus Vector Store. You can find more details about this in the llama_index/retrievers/fusion_retriever.py file.

I hope this helps! If you have any other questions, feel free to ask.

Best,
Dosu

Sources

• llama_index/vector_stores/milvus.py
• llama_index/retrievers/fusion_retriever.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.