ecrs 0.1.0-beta.4

Evolutionary computation tools & algorithms
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88

//! Implementations of goodness calculation.
//!
//! Looking at equation for probability of ant choosing some edge
//! ((Scholarpedia)[http://www.scholarpedia.org/article/Ant_colony_optimization#ConstructAntSolutions])
//! we can precalculate every possible value of numerator, and use it to speed up calculations.
//! In this library we refer to this precalculated value as goodness, and this module contains
//! trait [Goodness] that must be implemented for every goodness calculating object and
//! implementations of aforementioned trait.
use crate::aco::pheromone::Pheromone;
use crate::aco::FMatrix;

/// # Goodness
///
/// This trait must be implemented for goodness calculating struct.
pub trait Goodness<P: Pheromone> {
    /// Calculates goodness based on pheromone and its own internal state
    ///
    /// ## Arguments
    /// `pheromone` - Pheromone in matrix representation.
    fn apply(&mut self, pheromone: &P) -> P;
}

/// # Canonical Goodness
///
/// Implements [Goodness]. Provides goodness calculation based on Ant System equations
pub struct CanonicalGoodness {
    pub(in crate::aco) alpha: f64,
    pub(in crate::aco) beta: f64,
    pub(in crate::aco) heuristic: FMatrix,
}

impl CanonicalGoodness {
    /// Creates a new instance of [CanonicalGoodness].
    ///
    /// ## Arguments
    /// * `alpha` - importance of weights in edge choosing.
    /// * `beta` - importance of heuristic in edge choosing.
    /// * `heuristic` - Weighted graph in matrix representation.
    pub fn new(alpha: f64, beta: f64, heuristic: FMatrix) -> Self {
        Self {
            alpha,
            beta,
            heuristic,
        }
    }

    /// Creates a new instance of [CanonicalGoodness] with default values.
    ///
    /// ## Arguments
    /// * `solution_size` - number of vertices
    pub fn default(solution_size: usize) -> Self {
        let h = FMatrix::repeat(solution_size, solution_size, 1.0);
        Self::new(1.0, 1.0, h)
    }
}

impl Goodness<FMatrix> for CanonicalGoodness {
    fn apply(&mut self, pheromone: &FMatrix) -> FMatrix {
        let solution_size = pheromone.nrows();
        let iter = pheromone
            .iter()
            .zip(self.heuristic.iter())
            .map(|(p, h)| p.powf(self.alpha) * h.powf(self.beta));

        FMatrix::from_iterator(solution_size, solution_size, iter)
    }
}

#[cfg(test)]
mod tests {
    use crate::aco::goodness::{CanonicalGoodness, Goodness};
    use crate::aco::FMatrix;

    #[test]
    fn canonical_goodness_calculations_are_right() {
        let heuristic = FMatrix::from_vec(2, 2, vec![1.0, 2.0, 4.0, 8.0]);
        let alpha = 2.0;
        let beta = 3.0;
        let pheromone = FMatrix::from_vec(2, 2, vec![4.0, 2.0, 8.0, 0.5]);

        let goodness = vec![16.0, 32.0, 4096.0, 128.0];

        let mut g_op = CanonicalGoodness::new(alpha, beta, heuristic);
        for (a, b) in goodness.iter().zip(g_op.apply(&pheromone).iter()) {
            assert_eq!(a, b);
        }
    }
}