Join us at Snowflake Summit June 26-29 in Las Vegas!

Visualize Graphs in Notebooks

Pete Vilter and Mary McGrath

11 November 2021

less than a minute read

We are excited to announce that you can now use Graphviz to visualize directed and undirected graphs in RAI notebooks. Just provide node and edge relations, and the notebook will render the graph.

You can use attributes to determine the style and layout of the graph. Subgraphs allow finer grained control.


A module containing the following:

  • :node: a unary relation of identifiers (usually strings or integers) to use as node ids.
  • :edge: a binary relation of edges as (from, to) pairs of node ids.

Here is an example, using a small family tree to generate a graph. We have defined the nodes and edges as below.

def nodes = "grandparent"

def edges = {
  ("grandparent", "parent A");
  ("grandparent", "parent B");

module attributes
    def node = "fontname", "Helvetica"

module subgraphs["parents"]
    def node = {"parent A"; "parent B"}
    def edge = "parent B", "child"

    module attribute
        def node = {("color", "lightgrey"); ("style", "filled")}
        def graph = "bgcolor", "lightblue"

module subgraphs["children"]
    def parent = "parents"
    def node = "child"

module graph
    def node =  nodes
    def edge = edges
    def attribute = attributes
    def subgraph = subgraphs

def output = graphviz[graph]

When run in the RAI notebook, the following graph is generated:

Please refer to our Visualization Library (display) for further information.

Related Posts

Get Early Access

Join our community, keep up to date with the latest developments in our monthly newsletter, and get early access to RelationalAI.