NetworkX – Network Analysis in Python

I recommend to complement with the recent Reddit thread in &#x2F;r&#x2F;Python [1].<p>[1] &quot;What are the best libraries to work with graphs?&quot; <a href="https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;Python&#x2F;comments&#x2F;185xexg&#x2F;what_are_the_best_libraries_to_work_with_graphs&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;Python&#x2F;comments&#x2F;185xexg&#x2F;what_are_th...</a>

Been a few years since I put NetworkX through its paces, but the several times I have tried it, found remarkably weak support for graph layout and display. NetworkX analytic routines may be strong, but attractively displaying graph-structured problems remarkably more interactive and attractive via d3.js, GraphViz, etc. At least for my problems, communicating graph structures, and having nodes and edges that represent different kinds of things…these are basic requirements, not optional frills.

I found the documentation for networkx much better than the one from igraph[1] (at least the Python version). However, for community detection algorithms graph-tool[2] is better (it also uses a different class of models than the standard in literature)<p>[1] <a href="https:&#x2F;&#x2F;igraph.org" rel="nofollow noreferrer">https:&#x2F;&#x2F;igraph.org</a> [2] <a href="https:&#x2F;&#x2F;graph-tool.skewed.de" rel="nofollow noreferrer">https:&#x2F;&#x2F;graph-tool.skewed.de</a>

NetworkX let me whip up a useful shortest path routing proof of concept from telco data in a few hours. I was impressed with myself, but all glory goes to NetworkX !

Related:<p><i>NetworkX 3.0 - create, manipulate, and study complex networks in Python</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=34321135">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=34321135</a> - Jan 2023 (55 comments)

If you need more speed (but less features), check out networkit - <a href="https:&#x2F;&#x2F;networkit.github.io&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;networkit.github.io&#x2F;</a>

Love it. Using it for load balancing of substation transformers. The grid is a graph.

NetworkX has helped me with so many Advent of Code questions. Some where it was warranted, some where I made it work anyway because I like it!

See also <a href="https:&#x2F;&#x2F;github.com&#x2F;Qiskit&#x2F;rustworkx">https:&#x2F;&#x2F;github.com&#x2F;Qiskit&#x2F;rustworkx</a> – a general purpose graph library for Python written in Rust to take advantage of the performance and safety that Rust provides.<p>&gt; Rustworkx was originally called retworkx and was created initially to be a replacement for qiskit&#x27;s previous (and current) NetworkX usage (hence the original name). The project was originally started to build a faster directed graph to use as the underlying data structure for the DAG at the center of qiskit-terra&#x27;s transpiler. However, since it&#x27;s initial introduction the project has grown substantially and now covers all applications that need to work with graphs which includes Qiskit.

NetworkX&#x27;s documentation is the source of one of the most surreal but true sentences ever written (with just a slight alteration) [0]:<p><pre><code> A lobster is a tree that reduces to a caterpillar when pruning all leaves. </code></pre> [0] <a href="https:&#x2F;&#x2F;networkx.org&#x2F;documentation&#x2F;latest&#x2F;reference&#x2F;generated&#x2F;networkx.generators.random_graphs.random_lobster.html" rel="nofollow noreferrer">https:&#x2F;&#x2F;networkx.org&#x2F;documentation&#x2F;latest&#x2F;reference&#x2F;generate...</a>

I&#x27;ve used recently the networkx algorithm to find Hamiltonian cycles in a graph, in order to generate a Secret Santa with constraints (couples don&#x27;t send gifts to each other, and people don&#x27;t give to the same person as last year), it works great even though the problem is NP-complete, since my number of participants is very low.<p>I&#x27;ve tried the same in Rust with petgraph which resembles networkx, but it doesn&#x27;t have the algorithm for Hamiltonian built in and I couldn&#x27;t wrap my head around the DFS&#x2F;BFS visitor pattern, but I&#x27;ll continue this some day.

Igraph and cugraph, and graph tool are far superior for a wide variety of reasons

Happy user of NetworkX since 2009!

Back in the days I’ve discovered NetworkX and Gephi in a Coursera course and was really surprised about how simple it managed to represent visually such a hard problem (I’ve never been able to find this course again it started with Erdos number that’s the only thing I remember)

On a related note, is there a popular option nowadays for a solid FOSS graph visualization library in the browser.<p>Years back I used zoomcharts which was really good but had an expensive licence for on premise usage per client.

Very useful library. Representing large networks with graphs is super useful.

used this throughout my ML on graphs grad course this sem, it’s a really neat library

The &quot;N&quot; in HN stands for &quot;news&quot;. Is there anything new about networkx to talk about?