#classification #dissimilarity-space #mmodule #seriation #divide-and-conquer

robinson_mmodules

This crate is a direct implementation of the algorithms presented in "MODULES IN ROBINSON SPACES" by MIKHAEL CARMONA, VICTOR CHEPOI, GUYSLAIN NAVES AND PASCAL PREA. It can determine if a given square matrice admit a compatible order in O(n²). If it does admit at least one, this crate will provide you with a permutation corresponding to one compatible order.

1 unstable release

0.1.3 Jul 7, 2023
0.1.2 Jul 6, 2023
0.1.1 Jul 6, 2023
0.1.0 Jul 6, 2023

#1504 in Algorithms

MIT license

22KB
422 lines

Robinson

This rust crate is a direct implementation of the algorithms presented in “MODULES IN ROBINSON SPACES” by MIKHAEL CARMONA, VICTOR CHEPOI, GUYSLAIN NAVES AND PASCAL PREA.

Given a dissimilarity space (~ what I call a DistanceMatrice), this algorithm can recognize if this dissimilarity space is Robinson.

Authors

Installation/Dependencies

You can add robinson_mmodules to your dependencies using

cargo add robinson_mmodules

in your project directory.

Or add Robinson to your dependencies directly in your cargo.toml file

[dependencies]
robinson_mmodules = "0.X.0"

Usage/Examples

extern crate robinson_mmodules;
use robinson_mmodules::Robin;

fn main() {
    let my_matrice = vec![
        vec![0, 1, 2, 4, 3],
        vec![0, 0, 1, 3, 1],
        vec![0, 0, 0, 2, 1],
        vec![0, 0, 0, 0, 2],
        vec![0, 0, 0, 0, 0],
    ];
    let mut robin = Robin::new(my_matrice);

    println!("{}", robin.resolve_robin()); //true if your matrice admit a compatible order (equivalent to if your dissimilarity space is Robinson), false otherwise. In this case, it's true.
    println!("{:?}", robin.solved_permut.clone().unwrap()); //[1, 2, 3, 5, 4] which is a valid permutation
    //That means that if we apply this permutation we will get a matrice that respect the Robinson property
    let robinson_distance_matrice = robin.dist.permut_matrice(&robin.solved_permut.unwrap());
    println!("{}", robinson_distance_matrice.is_robinson()); //True since this DistanceMatrice respect the Robinson property here
    robinson_distance_matrice.display_mat(); //As you can visualize here
}

You can also find a more in depth exemple in the main.rs and test.rs file of this github repository. If you want to try this one yourself you'll need to add the rand crate to your depencies rand = "0.8.5"

Documentation

Documentation

License

MIT

Dependencies

~120KB