Query Neo4j Knowledge Graphs with LlamaIndex
Neo4j Query Engine Pack: Creates a Neo4j query engine and executes its query function.
Why it matters
Leverage your Neo4j knowledge graph to answer complex questions by integrating with LlamaIndex. This pack provides flexible query engines for diverse data retrieval needs.
Outcomes
What it gets done
Connect to Neo4j databases with provided credentials.
Build various query engines: vector-based, keyword-based, hybrid, or raw vector retrieval.
Execute queries against your knowledge graph and retrieve relevant information.
Integrate with external data sources like Wikipedia for enriched context.
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/li-pack-packs-neo4j-query-engine | bash Steps
Steps in the chain
Download the Neo4jQueryEnginePack using llamaindex-cli with the command: llamaindex-cli download-llamapack Neo4jQueryEnginePack --download-dir ./neo4j_pack. Inspect the files at ./neo4j_pack and use them as a template for your own project.
Install required dependencies using pip install llama-index-readers-wikipedia to enable loading documents from Wikipedia.
Load documents using WikipediaReader. Example: loader = WikipediaReader(); docs = loader.load_data(pages=["Paleolithic diet"], auto_suggest=False). Print the number of loaded documents.
Load Neo4j connection parameters from credentials.json file. Extract username, password, url, and database fields from the JSON file.
Instantiate Neo4jQueryEnginePack with Neo4j credentials and loaded documents: neo4j_pack = Neo4jQueryEnginePack(username=username, password=password, url=url, database=database, docs=docs). Optionally specify query_engine_type parameter.
Run a query using the pack's run() function: response = neo4j_pack.run("Tell me about the benefits of paleo diet."). Alternatively, access query_engine directly and call query_engine.query("query_str").
Overview
Neo4j Query Engine Pack
What it does
This LlamaPack creates a Neo4j query engine and executes its `query` function. It supports multiple query engine types, including knowledge graph vector-based entity retrieval (default), knowledge graph keyword-based entity retrieval, knowledge graph hybrid entity retrieval, raw vector index retrieval, custom combo query engine (vector similarity + KG entity retrieval), `KnowledgeGraphQueryEngine`, and `KnowledgeGraphRAGRetriever`.
How it connects
Use this pack to create and execute a Neo4j query engine. It is suitable for scenarios requiring various knowledge graph querying strategies. Do not use this pack if you do not have a Neo4j database set up.
Source README
Description pending for li-pack-packs-neo4j-query-engine.
Step 1: Download Neo4jQueryEnginePack
Download the Neo4jQueryEnginePack using llamaindex-cli with the command: llamaindex-cli download-llamapack Neo4jQueryEnginePack --download-dir ./neo4j_pack. Inspect the files at ./neo4j_pack and use them as a template for your own project.
Step 2: Install dependencies
Install required dependencies using pip install llama-index-readers-wikipedia to enable loading documents from Wikipedia.
Step 3: Load documents
Load documents using WikipediaReader. Example: loader = WikipediaReader(); docs = loader.load_data(pages=["Paleolithic diet"], auto_suggest=False). Print the number of loaded documents.
Step 4: Get Neo4j credentials
Load Neo4j connection parameters from credentials.json file. Extract username, password, url, and database fields from the JSON file.
Step 5: Create the pack
Instantiate Neo4jQueryEnginePack with Neo4j credentials and loaded documents: neo4j_pack = Neo4jQueryEnginePack(username=username, password=password, url=url, database=database, docs=docs). Optionally specify query_engine_type parameter.
Step 6: Execute query
Run a query using the pack's run() function: response = neo4j_pack.run("Tell me about the benefits of paleo diet."). Alternatively, access query_engine directly and call query_engine.query("query_str").Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.