samyama_optimization/algorithms/
bwr.rs

1use crate::common::{Individual, OptimizationResult, Problem, SolverConfig};
2use ndarray::Array1;
3use rand::prelude::*;
4use rayon::prelude::*;
5
6pub struct BWRSolver {
7    pub config: SolverConfig,
8}
9
10impl BWRSolver {
11    pub fn new(config: SolverConfig) -> Self {
12        Self { config }
13    }
14
15    pub fn solve<P: Problem>(&self, problem: &P) -> OptimizationResult {
16        let mut rng = thread_rng();
17        let dim = problem.dim();
18        let (lower, upper) = problem.bounds();
19
20        let mut population: Vec<Individual> = (0..self.config.population_size)
21            .map(|_| {
22                let mut vars = Array1::zeros(dim);
23                for i in 0..dim {
24                    vars[i] = rng.gen_range(lower[i]..upper[i]);
25                }
26                let fitness = problem.fitness(&vars);
27                Individual::new(vars, fitness)
28            })
29            .collect();
30
31        let mut history = Vec::with_capacity(self.config.max_iterations);
32
33        for iter in 0..self.config.max_iterations {
34            if iter % 10 == 0 {
35                println!("BWR Solver: Iteration {}/{}", iter, self.config.max_iterations);
36            }
37            let (best_idx, worst_idx) = self.find_best_worst(&population);
38            let best_vars = population[best_idx].variables.clone();
39            let worst_vars = population[worst_idx].variables.clone();
40            let best_fitness = population[best_idx].fitness;
41
42            history.push(best_fitness);
43
44            population = population
45                .into_par_iter()
46                .map(|mut ind| {
47                    let mut local_rng = thread_rng();
48                    let mut new_vars = Array1::zeros(dim);
49
50                    let r1: f64 = local_rng.gen();
51                    let r2: f64 = local_rng.gen();
52                    let r3: f64 = local_rng.gen();
53                    let r4: f64 = local_rng.gen();
54                    let t: f64 = local_rng.gen_range(1..3) as f64;
55                    
56                    let mut rand_vars = Array1::zeros(dim);
57                    for j in 0..dim {
58                        rand_vars[j] = local_rng.gen_range(lower[j]..upper[j]);
59                    }
60
61                    if r4 > 0.5 {
62                        for j in 0..dim {
63                            let delta = r1 * (best_vars[j] - t * rand_vars[j]) - r2 * (worst_vars[j] - rand_vars[j]);
64                            new_vars[j] = (ind.variables[j] + delta).clamp(lower[j], upper[j]);
65                        }
66                    } else {
67                        for j in 0..dim {
68                            new_vars[j] = (upper[j] - (upper[j] - lower[j]) * r3).clamp(lower[j], upper[j]);
69                        }
70                    }
71
72                    let new_fitness = problem.fitness(&new_vars);
73                    if new_fitness < ind.fitness {
74                        ind.variables = new_vars;
75                        ind.fitness = new_fitness;
76                    }
77                    ind
78                })
79                .collect();
80        }
81
82        let (final_best_idx, _) = self.find_best_worst(&population);
83        let final_best = &population[final_best_idx];
84
85        OptimizationResult {
86            best_variables: final_best.variables.clone(),
87            best_fitness: final_best.fitness,
88            history,
89        }
90    }
91
92    fn find_best_worst(&self, population: &[Individual]) -> (usize, usize) {
93        let mut best_idx = 0;
94        let mut worst_idx = 0;
95        for (i, ind) in population.iter().enumerate() {
96            if ind.fitness < population[best_idx].fitness {
97                best_idx = i;
98            }
99            if ind.fitness > population[worst_idx].fitness {
100                worst_idx = i;
101            }
102        }
103        (best_idx, worst_idx)
104    }
105}
samyama_optimization/algorithms/bwr.rs

samyama_optimization/algorithms/
bwr.rs
