The advantage of using matplotlib with NetworkX is the ease with which one can produce PDF, PNG, and SVG outputs.
import networkx as nx
import matplotlib.pyplot as plt
G1 = nx.Graph()
G.add_nodes_from(["a", "b"])
G.add_edge("a", "b")
nx.draw_networkx(G)
plt.savefig("graph.pdf")
plt.savefig("graph.png")
plt.savefig("graph.svg")
How can this code be modified to produce HTML5/Javascript Canvas code? Concrete code, however trivial, is more helpful than rooting for a given library. Sigmajs seems promising, if not for the scanty documentation.
The most straightforward way to achieve what you want is to simply export the graph as GEXF with Networkx using:
G = nx.path_graph(4)
nx.write_gexf(G, "test.gexf")
Then in your HTML you can use the sigma.js GEXF parser like so:
<div id="container">
<style>
#graph-container {
top: 0;
bottom: 0;
left: 0;
right: 0;
position: absolute;
}
</style>
<div id="graph-container"></div>
</div>
<script>
/**
* The plugin sigma.parsers.gexf can load and parse the GEXF graph file,
* and instantiate sigma when the graph is received.
*
* The object given as the second parameter is the base of the instance
* configuration object. The plugin will just add the "graph" key to it
* before the instanciation.
*/
sigma.parsers.gexf('data/arctic.gexf', { // path to graph here
container: 'graph-container'
});
</script>
If you want to look at a real live example, I suggest this blog post at the bottom.
DISCLAIMER: it's my blog ;)
