2

I am trying to create a graph where number of nodes is more than 400 which is not visualizing in a proper way. 

import networkx as nx 
G=nx.MultiGraph() 
G.add_nodes_from(freqItemSet) 
G.add_edges_from([edge]) 
plt.figure(figsize=(10000,10000)) 
nx.draw(G)

I was using the above code. Is there any other way to visualize in a proper way.

Thank you

Aric
  • 24,511
  • 5
  • 78
  • 77
user3064366
  • 1,509
  • 2
  • 11
  • 15
  • 2
    Try AT&T graphviz. It's wonderful. http://www.graphviz.org/ But you might need to rethink your idea about "proper". – duffymo Feb 20 '14 at 13:25
  • That's great if you can use GraphViz, but @duffymo makes a good point about a 'proper' graph. At some point, without effective zooming in/out, you are going to hit limitations with any graphing software. You can check out [this large graphviz example](http://www.graphviz.org/content/softmaint) or [this large example from networkx](http://networkx.github.io/documentation/latest/examples/drawing/random_geometric_graph.html) to get an idea of graphing large networks – StephenH Feb 20 '14 at 13:36
  • [And here is a networkx example](http://networkx.github.io/documentation/latest/examples/drawing/lanl_routes.html) that actually even uses graphviz. – StephenH Feb 20 '14 at 13:37

1 Answers1

0

You can use a tool called "Gephi" which is free and has good UI. You can export the file from networkx as "*.gexf" format

szues
  • 11
  • 1