Drawing large graph with graphviz

drawgraphgraphviz

I have here generated a large .dot file of my facebook friends' graph with fb-map. It has 287 nodes and almost 2000 edges. I'm using dot and neato to generate a .png image, using the overlap="orthoyx" paramethers, but it doesn't give a nice effect. Too many overlapped edges.

Do you know any set of options to manage such huge graphs and draw them in an eye-readable way with Graphviz?

Best Answer

The igraph library layout algorithms seem to cope much better with very large graphs than dot does. It even includes an implementation of Kamada-Kawai and Fruchterman-Reingold layout algorithms.

There are up-to-date python bindings available, not to mention nice documentation and tutorials.

Related Solutions

Python – How to change the size of figures drawn with Matplotlib

figure tells you the call signature:

from matplotlib.pyplot import figure

figure(figsize=(8, 6), dpi=80)

figure(figsize=(1,1)) would create an inch-by-inch image, which would be 80-by-80 pixels unless you also give a different dpi argument.

Windows – Graphviz: How to go from .dot to a graph

type: dot -Tps filename.dot -o outfile.ps

If you want to use the dot renderer. There are alternatives like neato and twopi. If graphiz isn't in your path, figure out where it is installed and run it from there.

You can change the output format by varying the value after -T and choosing an appropriate filename extension after -o.

If you're using windows, check out the installed tool called GVEdit, it makes the whole process slightly easier.

Go look at the graphviz site in the section called "User's Guides" for more detail on how to use the tools:

http://www.graphviz.org/documentation/

(See page 27 for output formatting for the dot command, for instance)

http://www.graphviz.org/pdf/dotguide.pdf

Best Answer

Related Solutions

Python – How to change the size of figures drawn with Matplotlib

Windows – Graphviz: How to go from .dot to a graph

Related Topic