View Neptune Graph Schema using Jupyter notebook

Question

Is there a way to view the schema of a graph in a Neptune cluster using Jupyter Notebook?

Like you would do a "select * from tablename limit 10" in an RDS using SQL, similarly is there a way to get a sense of the graph data through Jupyter Notebook?

Answer 1

It depends on how large your graph is as to how well this will perform but you can get a sense of the type of nodes and edges you have using something like the example below. From the tags you used I assume you are using Gremlin:

g.V().groupCount().by(label)
g.E().groupCount().by(label)

If you have a very large graph try putting something like limit(100000) before the groupCount step.

If you are using a programming language like Python (with gremlin python installed) then you will need to add a next() terminal step to the queries as in:

g.V().groupCount().by(label).next()
g.E().groupCount().by(label).next()

Having found the labels and distribution of the labels you could use one of them to explore some properties. Let's imagine there is a label called "person".

g.V().hasLabel('person').limit(10).valueMap().toList()

Remember with Gremlin property graphs vertices with the same label may not necessarily have all the same properties so it's good to look at more than one vertex to get a sense for that as well.

Answer 2

If you want to visualize your graph ie. want to view all the child nodes below a given root node, then you can use this below gremlin query in a jupyter notebook cell.

    %%gremlin -p v,e  
    g.V({label},{property},{value}) 
    .repeat(outE().inV()).until(outE().count().is(0)).path() 
    .by(valueMap({propertiesList})).by(label)

Change

1. {label} : vertex label
2. {property} : vertex prop to filter
3. {value} : value of prop
4. {propertiesList} : list of properties of vertex to be shown in graph