SGLReferenceGenerators

Generate a scale-free network based on the Barabasi-Albert preferential attachment model. More...

Generate a scale-free network based on the Barabasi-Albert preferential attachment model.

Generates an x-by-y sparse mesh.

This implementation is based on Sanders and Schulz [^1].

Vertices with lower vertex IDs are more likely to be the target for edges. Note that if add_reverse_edge is false, the outgoing edge distribution is constant, as each vertex will have exactly 'd' edges.

The returned view owns its underlying container.

Parameters

n	Number of vertices in the generated graph.
d	The number of outgoing edges to generate for each vertex

Returns

A view over the generated graph.

! https://i.imgur.com/2nwNn5l.png

Example

[^1] Scalable Generation of Scale-free Graphs. Peter Sanders, Christian Schulz. arXiv:1602.07106v1. [cs.DS] 23 Feb 2016. http://arxiv.org/pdf/1602.07106v1.pdf

The generated sparse mesh will contain x vertices in the horizontal direction and y vertices in the vertical direction. This is distributed n/p in the y-direction.

The returned view owns its underlying container.

Each edge is added with a probability of p. Thus, p = 1 generates a complete mesh, and p = 0.5 generates a mesh with 50% of the edges of a complete mesh.

Parameters

nx	Size of the x-dimension of the sparse mesh.
ny	Size of the y-dimension of the sparse mesh.
p	Probability of adding a particular edge to the mesh.
bidirectional	True to add back-edges in a directed graph, false for forward edges only.

Returns

A view over the generated graph.

Example