How to create random graph where each node has at least 1 edge using Networkx

submited by

Style Pass

2024-03-30 00:00:03

Communities for your favorite technologies. Explore all Collectives

Ask questions, find answers and collaborate at work with Stack Overflow for Teams. Explore Teams

I've managed to create a random undirected weighted graph for testing with Dijkstra's algorithm, but how can I make it so each node has at least one edge that connects them to the graph?

This creates the graph well, and I managed to plot it, so I can actually see it, my problem is with the probability for edge creation. I don't want it so high that all nodes have the max amount of edges, but putting a low value can result in a node with 0 edges. Is there a way to make sure that each node has at least one edge?

However, you could tweak a little bit the approach used in nx.gnp_random_graph, so that instead of setting an edge among all possible edge combinations with a random probability, we add one edge for each node randomly, and then add the remaining edges with a probability p.