Getting the embedding results from Python

Computing embedding is usually not the end of the story. Once you have managed to get a vector representation of nodes, you can continue your machine learning task, such as node classification, with your favorite package (this can be scikit-learn to use decision trees or support vectors, for instance). We won't go into the details of such analysis in this book, but here is how to retrieve the computed vectors from Neo4j.

First, we need to create a Neo4j driver (check Chapter 8, Using Graph-Based Features in Machine Learning, if you don't know about the Neo4j Python driver):

from neo4j import GraphDatabase
driver = GraphDatabase.driver("bolt://localhost:7687", auth=("neo4j", "<YOUR_PASSWORD>"))

Then, we can run an embedding procedure and get the results in a DataFrame:

import pandas as pd

with driver.session() as session:
    result = session.run(
        "CALL gds.alpha.node2vec('proj_graph...