Skip to content

Rust library for of graph ensembles

License

Apache-2.0, MIT licenses found

Licenses found

Apache-2.0
LICENSE-APACHE
MIT
LICENSE-MIT
Notifications You must be signed in to change notification settings

Pardoxa/net_ensembles

Repository files navigation

Rust library for random graph ensembles

Crate Rust unit tests - master Docs

Minimal Rust version: 1.55.0

Implements simple sampling and monte carlo (or rather markov-) steps, that can be used to create a markov chain.

This is intended to be used for various different use cases. As such, you can easily define additional data that should be stored at each vertex.

Usage

Add this to your Cargo.toml:

[dependencies]
net_ensembles = "0.5"
# for feature "serde_support" (enabled by default) also use
serde = { version = "1.0", features = ["derive"] }

If you do not need serde support, add this instead:

[dependencies]
net_ensembles = { version = "0.5", default-features = false  }

Release Notes

See changelog. Note that savestates (created with serde) of v0.4 are incompatible with savestates generated by older versions and vise versa. Braking changes may or may not affect the loading of save states, you have to test that out yourself

currently implemented graph ensembles

  • Erdős-Rényi (x2)
  • small-world

work in progress

  • Barabási-Albert
  • Configuration Model
  • spacial networks

Note

On a 64 bit system drawing an usize consumes more randomness than on a 32 bit system, therefore ensembles drawn etc. are affected by the size of usize.

Graph

  • you can always visualize the current graph by creating a .dot file from it. There are different options for that, choose which one fits you best.

Implements measurable quantities

  • average degree
  • connected components
  • diameter
  • is_connected
  • leaf count
  • q_core
  • transitivity
  • biconnected component
  • vertex_load (closely related, often equal to betweeness)

Iterators

  • depth first search from index
  • breadth first search from index
  • over additional data

For each vertex

methods and more

  • degree
  • check adjacency with other nodes
  • access additional data

Iterators

  • iterate over indices stored in adjacency list

Documentation:

Notes

No warranties whatsoever, but since I am writing this library for my own scientific simulations, I do my best to avoid errors.

You can learn more about me and my research on my homepage.

If you notice any bugs, or want to request new features: do not hesitate to open a new issue on the repository.

vertices

  • The number of vertices has to be decided when creating a graph and cannot be changed later - at least for now.
  • I might add a method to add vertices if requested or I need it myself.

Due to implementation details, where I prioritize fast access of vertices, it is unlikely, that I will implement the option to remove vertices. If I do, it will likely be a relatively costly operation, so keep that in mind.

crates.io

  • I might move the sampling module into a different crate in the future. If I do, everything will likely be reexported at the same position as currently

License

Licensed under either of

at your option.

Contribution

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.