AI applicationsTools and integrationsLlamaIndex

LlamaIndex

LlamaIndex is a data framework for connecting your own data to large language models (LLMs). With the Memgraph graph store, LlamaIndex builds a knowledge graph in Memgraph from unstructured data and answers questions about it in natural language. You can also follow the example in the LlamaIndex docs.

Installation

Install LlamaIndex and the Memgraph graph store:

pip install llama-index llama-index-graph-stores-memgraph

Environment setup

You need a running Memgraph instance. To start one, see Getting started.

Create the graph store with the connection details of your database:

from llama_index.graph_stores.memgraph import MemgraphPropertyGraphStore
 
username = ""  # Enter your Memgraph username (default "")
password = ""  # Enter your Memgraph password (default "")
url = ""  # Specify the connection URL, e.g., 'bolt://localhost:7687'
 
graph_store = MemgraphPropertyGraphStore(
    username=username,
    password=password,
    url=url,
)

The example uses OpenAI, so set your OpenAI API key:

import os
os.environ["OPENAI_API_KEY"] = "YOUR_OPENAI_API_KEY"

Dataset

The example uses this text about Charles Darwin, saved in the ./data/charles_darwin/charles.txt file:

Charles Robert Darwin was an English naturalist, geologist, and biologist,
widely known for his contributions to evolutionary biology. His proposition that
all species of life have descended from a common ancestor is now generally
accepted and considered a fundamental scientific concept. In a joint publication
with Alfred Russel Wallace, he introduced his scientific theory that this
branching pattern of evolution resulted from a process he called natural
selection, in which the struggle for existence has a similar effect to the
artificial selection involved in selective breeding. Darwin has been described
as one of the most influential figures in human history and was honoured by
burial in Westminster Abbey.
from llama_index.core import SimpleDirectoryReader
 
documents = SimpleDirectoryReader("./data/charles_darwin/").load_data()

The documents variable now holds the text, ready for building the graph.

Graph construction

LlamaIndex has several graph constructors. This example uses SchemaLLMPathExtractor, which can use a schema you define or let the LLM choose the entities.

from llama_index.core import PropertyGraphIndex
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.llms.openai import OpenAI
from llama_index.core.indices.property_graph import SchemaLLMPathExtractor
 
index = PropertyGraphIndex.from_documents(
    documents,
    embed_model=OpenAIEmbedding(model_name="text-embedding-ada-002"),
    kg_extractors=[
        SchemaLLMPathExtractor(
            llm=OpenAI(model="gpt-4", temperature=0.0),
        )
    ],
    property_graph_store=graph_store,
    show_progress=True,
)

The text becomes a graph in Memgraph:

A paragraph about Charles Robert Darwin is turned by PropertyGraphIndex.from_documents() into a graph in Memgraph, with Darwin connected to scientific theory and natural selection, and a chunk node that mentions them

Querying

LlamaIndex can combine several retrievers to find nodes and paths in the graph. If you don’t choose any, it uses LLMSynonymRetriever and VectorContextRetriever. The Memgraph graph store creates a vector index on the entity embeddings, and VectorContextRetriever queries it with Memgraph’s vector search.

query_engine = index.as_query_engine(include_text=True)
 
response = query_engine.query("Who did Charles Robert Darwin collaborate with?")
print(str(response))

LLMSynonymRetriever finds the nodes that match the question’s keywords and returns their neighborhood:

LLMSynonymRetriever generates keywords and synonyms, finds nodes by exact keyword match, returns their one-hop neighborhood and turns the result into the answer Alfred Russel Wallace

Demos

LangChainLightRAG