Struct VnBest 

pub struct VnBest {
    pub part_count: usize,
}

Steepest descent Vector-of-Numbers algorithm

This algorithm greedily moves weights from parts to parts in such a way that the balance gain is maximized on each move. See also its non-greedy version.

Example

use coupe::Partition as _;
use rand;

let part_count = 2;
let mut partition = [0; 4];
let weights = [4, 6, 2, 9];

coupe::Random { rng: rand::thread_rng(), part_count }
    .partition(&mut partition, ())
    .unwrap();
coupe::VnBest { part_count }
    .partition(&mut partition, weights)
    .unwrap();

Reference

Remi Barat. Load Balancing of Multi-physics Simulation by Multi-criteria Graph Partitioning. Other [cs.OH]. Université de Bordeaux, 2017. English. NNT : 2017BORD0961. tel-01713977

Fields

part_count: usize

Trait Implementations

impl<W> Partition<W> for VnBest

where W: IntoIterator, W::Item: VnBestWeight,

type Metadata = usize

Diagnostic data returned for a specific run of the algorithm.

type Error = Error

Error details, should the algorithm fail to run.

fn partition( &mut self, part_ids: &mut [usize], weights: W, ) -> Result<Self::Metadata, Self::Error>

Partition the given data and output the part ID of each element in part_ids.