Property Graph Index Visualization

Similar to the property_graph_basic notebook, in this notebook, we demonstrate an alternative visualization approach for the default SimplePropertyGraphStore

While the focus of the other notebook is querying the graph, this notebook focuses on the visualization aspect of what was created.

python

%pip install llama-index

Setup

python

import os
import urllib.request

os.environ["OPENAI_API_KEY"] = "sk-proj-..."

python

url = "https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/paul_graham/paul_graham_essay.txt"
filename = "data/paul_graham/paul_graham_essay.txt"
os.makedirs(os.path.dirname(filename), exist_ok=True)
urllib.request.urlretrieve(url, filename)

python

import nest_asyncio

nest_asyncio.apply()

python

from llama_index.core import SimpleDirectoryReader

python

documents = SimpleDirectoryReader("./data/paul_graham/").load_data()

Construction

python

from llama_index.core import PropertyGraphIndex
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.llms.openai import OpenAI

index = PropertyGraphIndex.from_documents(
    documents,
    llm=OpenAI(model="gpt-3.5-turbo", temperature=0.3),
    embed_model=OpenAIEmbedding(model_name="text-embedding-3-small"),
    show_progress=True,
)

Visualization

Let's explore what we created. Using the show_jupyter_graph() method to create our graph directly in the Jupyter cell!

Note that this only works in Jupyter environments.

python

index.property_graph_store.show_jupyter_graph()