6

I have a list as c4_leaves = [56,78,90,112]. I'm trying to create a complete graph using these elements in c4_leaves as nodes. Here's what I've tried:

V_ex = c4_leaves
G_ex = nx.Graph() 
G_ex.add_nodes_from(V_ex)
G_ex = nx.complete_graph(4)

for u,v in G_ex.edges():
    G_ex[u][v]['distance'] = distance(points33, u, v)

And then the minimum spanning tree of the above graph as:

T_ex = nx.minimum_spanning_tree(G_ex, weight='distance')
F_ex = list(T_ex.edges())

When I draw G_ex, it gives me the correct graph, but when I print details of the minimum spanning tree, it shows that T_ex.nodes() = [0,1,2,3,56,78,90,112].

Can someone show me the mistake I'm making?

wovano
  • 4,543
  • 5
  • 22
  • 49
ccc
  • 574
  • 2
  • 9
  • 22

3 Answers3

6

Instead of using complete_graph, which generates a new complete graph with other nodes, create the desired graph as follows:

import itertools
import networkx as nx

c4_leaves = [56,78,90,112]
G_ex = nx.Graph()
G_ex.add_nodes_from(c4_leaves)
G_ex.add_edges_from(itertools.combinations(c4_leaves, 2))

In the case of directed graphs use:

G_ex.add_edges_from(itertools.permutations(c4_leaves, 2))
zohar.kom
  • 1,765
  • 3
  • 12
  • 28
1

It is an old question. However, I am still putting my two cents in it. I was faced with the same issue. I am not sure what exactly was blocking point for the actual question but I will write down what I did.

So, I want to create a complete graph with four nodes (56,78,90, and 112). I have a list. I looked up the definition of complete_graph And here is what I saw

Signature: nx.complete_graph(n, create_using=None)
Docstring:
Return the complete graph `K_n` with n nodes.

Parameters
----------
n : int or iterable container of nodes
    If n is an integer, nodes are from range(n).
    If n is a container of nodes, those nodes appear in the graph.
create_using : NetworkX graph constructor, optional (default=nx.Graph)
   Graph type to create. If graph instance, then cleared before populated.

Examples
--------
>>> G = nx.complete_graph(9)
>>> len(G)
9
>>> G.size()
36
>>> G = nx.complete_graph(range(11, 14))
>>> list(G.nodes())
[11, 12, 13]
>>> G = nx.complete_graph(4, nx.DiGraph())
>>> G.is_directed()
True

Which means it can take an iterator. In the same spirit, I tried the following code

In [6]: l = [56,78,90,112]

In [7]: G = nx.complete_graph(l)

In [8]: G.edges(data=True)
Out[8]: EdgeDataView([(56, 78, {}), (56, 90, {}), (56, 112, {}), (78, 90, {}), (78, 112, {}), (90, 112, {})])
    
In [10]: G.nodes(data=True)
Out[10]: NodeDataView({56: {}, 78: {}, 90: {}, 112: {}})

So, there you have it, a complete graph built out of a list.

I hope that this answers the question.

SRC
  • 2,123
  • 3
  • 31
  • 44
0

The command G_ex = nx.complete_graph(4) creates a complete graph G whose nodes are 0, 1, 2, and 3. You then add more to G, but it has those nodes.

Joel
  • 22,598
  • 6
  • 69
  • 93