CODEWITHSUNDEEP
pythonmatplotlibnetworkx
jfeuerman
jfeuerman

Reputation: 177

Creating a directed, weighted network graph from weighted edgelist

I have the following edgelist which is stored in edgelist.txt:

soldier1, soldier2, 3
soldier1, soldier3, 1
soldier1, soldier4, 2
soldier2, soldier1, 2
soldier2, soldier3, 1
soldier2, soldier4, 3
soldier3, soldier1, 3
soldier3, soldier2, 2
soldier3, soldier4, 1
soldier4, soldier1, 2
soldier4, soldier2, 1
soldier4, soldier3, 3

To draw the graph, I'm using the following code:

import networkx as nx
import matplotlib.pyplot as plt

G = nx.read_weighted_edgelist(path='edgelist.txt', delimiter=',', create_using=nx.DiGraph())
nx.draw_networkx(G, with_labels=True, arrowsize=5, pos=nx.spring_layout(G))
limits=plt.axis('off')
plt.show()

However, it is creating a graph with double the amount of nodes (there should only be 4, but it is creating the graph with 8.) This is evident when you view the plot of the graph: Network Graph

Here is the graph info:

Name: 
Type: DiGraph
Number of nodes: 8
Number of edges: 12
Average in degree:   1.5000
Average out degree:   1.5000

The number of edges is correct but there are double the nodes. How do I fix this?

Upvotes: 2

Views: 644

Answers (1)

abc
abc

Reputation: 11929

Since you specify , as a delimiter, the whitespace character is considered part of the name.
So you'll end up with two nodes ' soldierN' and 'soldierN' for each node.

You can change your delimiter to consider also the whitespace, i.e., from ',' to ', '

>>> G = nx.read_weighted_edgelist(path='edgelist.txt', delimiter=', ', create_using=nx.DiGraph())
>>> G.nodes()
['soldier1', 'soldier2', 'soldier3', 'soldier4']

Upvotes: 3

Related Questions