[Question]: Find top-K nodes with a given node

[MikeChu-GitHub]

Question Validation

• [X] I have searched both the documentation and discord for an answer.

Question

Hi, I build an index and have it persisted in local disk. I'm able to see the embeddings in default_vector_store.json. How do I find top-K nodes with a specific node? I know I can find top-K nodes with a query. However, is it possible to achieve this with a node id instead of a query?

[dosubot[bot]]

🤖

Hey there, @MikeChu-GitHub!👋 I'm Dosu, a friendly bot here to lend a hand while you're waiting for a human maintainer. I'm all about helping solve bugs, answering your questions, and guiding you on your way to becoming a fantastic contributor. Don't worry, I'm taking a look at your issue about finding top-K nodes with a specific node id, and I'll get back to you with a full answer shortly. Hang tight!

[ravi03071991]

@MikeChu-GitHub You can probably get text with the help of node id and use that text as query in your retriever.

[MikeChu-GitHub]

Yeah, but it will require the node content to be embedded. I'm trying to avoid this(and the fee) since this is already available in the index.

[dosubot[bot]]

🤖

Based on the current functionality of the LlamaIndex, there isn't a direct method to find the top-K nodes with a specific node id without using a query. However, you can achieve this by using the get_networkx_graph method to get a networkx representation of the graph structure and then use networkx's built-in methods to find the top-K nodes based on some criteria.

Here is an example of how you can do this:

import networkx as nx

# Get the networkx graph from the LlamaIndex
graph = llama_index.get_networkx_graph(limit=1000)  # Adjust the limit as needed

# Define your node id
node_id = "your_node_id"

# Check if the node exists in the graph
if node_id in graph:
    # Get the neighbors of the node
    neighbors = list(graph.neighbors(node_id))

    # If you want the top-K nodes based on degree centrality
    centrality = nx.degree_centrality(graph)
    top_k_nodes = sorted([(node, centrality[node]) for node in neighbors], key=lambda x: x[1], reverse=True)[:K]
else:
    print(f"Node {node_id} does not exist in the graph.")

In this example, top_k_nodes will be a list of tuples where the first element is the node id and the second element is its degree centrality. The list is sorted in descending order of degree centrality, so the first element in the list is the node with the highest degree centrality.

Please note that this example assumes that you want the top-K nodes based on degree centrality. If you want the top-K nodes based on some other criteria, you will need to adjust the code accordingly. Also, the get_networkx_graph method requires the networkx library to be installed. If it's not installed, you can install it with pip install networkx.

You can find more information about the get_networkx_graph method in the llama_index/indices/knowledge_graph/base.py file.

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

Sources

• llama_index/indices/knowledge_graph/base.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.

[ravi03071991]

You could probably use get_top_k_embeddings function by passing the node embedding as query embedding and embedding ids as nodes ids, embeddings as node embeddings and accordingly use it.