Interactive Graph Visualization

Table of Contents

• Introduction
• Usage
• Installation
• Customizable Parameters
• Feedback
• Contributors

Introduction

Interactive Graph Visualization (igviz) is a library to help visualize graphs interactively using Plotly. This library provides a customizable api for visualizing graphs in a neat, visually appealing plot. It keeps larger graphs much more clean by displaying minimal text information and highlights node properties and relationships using colour and size while providing the same text information when needed.

Usage

Example notebooks can be found here.

Basic

import networkx as nx
import igviz as ig

G = nx.random_geometric_graph(200, 0.125)
nx.set_node_attributes(G, 3, "prop")
nx.set_edge_attributes(G, 5, "edge_prop")

ig.plot(G)

The default plot colors and sizes the nodes by the Degree but it is configurable.

Configurations

ig.plot(
    G, # Your graph
    title="My Graph",
    size_method="static", # Makes node sizes the same
    color_method="##ffcccb", # Makes all the node colours black,
    node_text=["prop"], # Adds the 'prop' property to the hover text of the node
    annotation_text="Visualization made by <a href='https://github.com/Ashton-Sidhu/plotly-graph'>igviz</a> & plotly.", # Adds a text annotation to the graph
)

ig.plot(
    G,
    title="My Graph",
    size_method="prop", # Makes node sizes the size of the "prop" property
    color_method="prop", # Colors the nodes based off the "prop" property and a color scale,
    node_text=["prop"], # Adds the 'prop' property to the hover text of the node
)

ig.plot(
    G,
    node_label="prop", # Display the "prop" attribute as a label on the node
    node_label_position="top center", # Display the node label directly above the node
    edge_text=["edge_prop"], # Display the "edge_prop" attribute on hover over the edge
    edge_label="edge_prop", # Display the "edge_prop" attribute on the edge
    edge_label_position="bottom center", # Display the edge label below the edge
)

How to add your own custom sizing method and colour method

To add your own custom sizing and color method, just pass a list to the size_method and color_method.

color_list = []
sizing_list = []

for node in G.nodes():
    size_and_color = G.degree(node) * 3

    color_list.append(size_and_color)
    sizing_list.append(size_and_color)

ig.plot(
    G,
    title="My Graph",
    size_method=sizing_list, # Makes node sizes the size of the "prop" property
    color_method=color_list, # Colors the nodes based off the "prop" property and a color scale,
    node_text=["prop"], # Adds the 'prop' property to the hover text of the node
)

Applying layouts

All layouts are calculated through the pos property on each node. Networkx has built in layouts you can use and can invoke through igviz.

ig.plot(
    G,
    title="My Graph",
    layout="kamada",
)

To add your own pos property you can set it via the nx.set_node_attributes function.

pos_dict = {
    0: [1, 2], # X, Y coordinates for Node 0
    1: [1.5, 3], # X, Y coordinates for Node 1
    ...
}

nx.set_node_attributes(G, pos_dict, "pos")

ig.plot(
    G
)

Directed & Multi Graphs

Igviz also plots Directed and Multigraphs with no configuration chages. For Directed Graphs the arrows are shown from node to node. For Multi Graphs only one edge is shown and it is recommended to display edge properties via edge_label or edge_text to display the weights of all edges between 2 Multi Graph nodes.

Directed Graph

def createDiGraph():
    # Create a directed graph (digraph) object; i.e., a graph in which the edges
    # have a direction associated with them.
    G = nx.DiGraph()

    # Add nodes:
    nodes = ['A', 'B', 'C', 'D', 'E']
    G.add_nodes_from(nodes)

    # Add edges or links between the nodes:
    edges = [('A','B'), ('B','C'), ('B', 'D'), ('D', 'E')]
    G.add_edges_from(edges)
    return G

DG = createDiGraph()

ig.plot(DG, size_method="static")

Multi Graph

MG = nx.MultiGraph()
MG.add_weighted_edges_from([(1, 2, 0.5), (1, 2, 0.75), (2, 3, 0.5)])

ig.plot(
    MG,
    layout="spring",
    size_method="static",
    edge_text=["weight"],
    colorscale="Rainbow"
)

Installation

pip install igviz

Customizable Parameters

• title : Title of the graph, by default "Graph"

• layout : Layout of the nodes on the plot ("random", "circular", "kamada", "planar", "spring", "spectral", "spiral"}, optional).

• size_method : How to size the nodes., by default "degree"

• color_method : How to color the node., by default "degree"

• node_label : Node property to be shown on the node.

• node_label_position: Position of the node label.

• node_text : A list of node properties to display when hovering over the node.

• edge_label : Edge property to be shown on the edge.

• edge_label_position: Position of the edge label.

• edge_text : A list of edge properties to display when hovering over the edge.

• titlefont_size : Font size of the title, by default 16

• annotation_text : Graph annotation text

• colorscale : Scale of the color bar ('Greys', 'YlGnBu', 'Greens', 'YlOrRd', 'Bluered', 'RdBu', 'Reds', 'Blues', 'Picnic', 'Rainbow', 'Portland', 'Jet', 'Hot', 'Blackbody', 'Earth', 'Electric', 'Viridis')

• colorbar_title : Color bar axis title.

• node_opacity : Opacity of the nodes (1 - filled in, 0 completely transparent), by default 1

• arrow_size : Size of the arrow for Directed Graphs and MultiGraphs, by default 2.

Feedback

I appreciate any feedback so if you have any feature requests or issues make an issue with the appropriate tag or futhermore, send me an email at ashton.sidhu1994@gmail.com

Contributors

This project follows the all-contributors specification and is brought to you by these awesome contributors.