Title: | Murty's Algorithm for k-Best Assignments |
---|---|
Description: | Calculates k-best solutions and costs for an assignment problem following the method outlined in Murty (1968) <doi:10.1287/opre.16.3.682>. |
Authors: | Aljaz Jelenko [aut, cre] |
Maintainer: | Aljaz Jelenko <[email protected]> |
License: | MIT + file LICENSE |
Version: | 0.3.1 |
Built: | 2025-02-27 03:15:49 UTC |
Source: | https://github.com/arg0naut91/murty |
Find k-best assignments for a given matrix (returns both solved matrices and costs).
get_k_best( mat, k_best = NULL, algo = "hungarian", by_rank = FALSE, objective = "min", proxy_Inf = 10000000L )
get_k_best( mat, k_best = NULL, algo = "hungarian", by_rank = FALSE, objective = "min", proxy_Inf = 10000000L )
mat |
Square matrix (N x N) in which values represent the weights. |
k_best |
How many best scenarios should be returned. If by_rank = TRUE, this equals best ranks. |
algo |
Algorithm to be used, either 'lp' or 'hungarian'; defaults to 'hungarian'. |
by_rank |
Should the solutions with same cost be counted as one and stored in a sublist? Defaults to FALSE. |
objective |
Should the cost be minimized ('min') or maximized ('max')? Defaults to 'min'. |
proxy_Inf |
What should be considered as a proxy for Inf? Defaults to 10e06; if objective = 'max' the sign is automatically reversed. |
A list with solutions and costs (objective values).
set.seed(1) mat <- matrix(sample.int(15, 10*10, TRUE), 10, 10) get_k_best(mat, 3)
set.seed(1) mat <- matrix(sample.int(15, 10*10, TRUE), 10, 10) get_k_best(mat, 3)