#Hypergraphs are fun! 

Code is written in python 3.

I use tools such as

• IPython
• numpy
• networkx

##Installation

• clone the library
• create virtualenv
• install requirements
• run setup.py

$ python setup.py install

You can also use automated provisioning which would set up a virtual machine with all dependencies installed and ready to use interactive examples in IPython notebook server.

I write about first two on my blog...

IPython has a powerful notebook which is amazing for sharing code and insights and for interactive development.

##Links to notebooks:

###How can we represent hypergraphs in code? A very good representation is bipartite graph of nodes and hyperedges. Another representation is a graph with hyperedges as nodes connected if hyperedges have common nodes. Is it a good representation for our diffusion problem? Can we use it as a reference model?

• hypergraphs 1

###How different are hypergraphs from graphs with cliques? Cliques are sets of nodes in which every node is connected to every other node. They are a bit similar to hypergraphs, but how the nodes are connected is conceptually different from hypergraphs.

###How are hypergraphs different from graphs with cliques in diffusion simulation with markov chain?

• hypergraphs vs graphs of cliques - diffusion

###Hypergraph traversal with random walk

I developed two models. One is node based, second is edge based.

You can find an interactive notebook here:

• models, numerical solutions, interactive comparison