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
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183

use crate::engines::alterers::alter::Alter;
use crate::engines::codex::Codex;
use crate::engines::engine::Engine;
use crate::engines::genetic_engine_params::GeneticEngineParams;
use crate::engines::genome::genes::gene::Gene;
use crate::engines::genome::population::Population;
use crate::engines::optimize::Optimize;
use crate::engines::schema::timer::Timer;
use crate::engines::score::Score;

use super::engine_context::EngineContext;
use super::genome::phenotype::Phenotype;
use super::selectors::selector::Select;

pub struct GeneticEngine<G, A, T>
where
    G: Gene<G, A>
{
    pub params: GeneticEngineParams<G, A, T>,
}

impl<G, A, T> GeneticEngine<G, A, T>
where
    G: Gene<G, A>
{

    pub fn new(params: GeneticEngineParams<G, A, T>) -> Self {
        GeneticEngine { params }
    }

    pub fn from_codex(codex: Codex<G, A, T>) -> GeneticEngineParams<G, A, T> {
        GeneticEngineParams::new().codex(codex)
    }

    pub fn evaluate(&self, handle: &mut EngineContext<G, A, T>) {
        let codex = self.codex();
        let fitness_fn = self.fitness_fn();
        let optimize = self.optimize();

        for idx in 0..handle.population.len() {
            let individual = handle.population.get_mut(idx);
            if !individual.score().is_some() {
                let decoded = codex.decode(individual.genotype());
                let score = fitness_fn(&decoded);

                individual.set_score(Some(score));
            }
        }

        optimize.sort(&mut handle.population);
    }

    pub fn select_survivors(&self, population: &Population<G, A>) -> Population<G, A> {
        let selector = self.survivor_selector();
        let count = self.survivor_count();

        selector.select(population, count)
    }

    pub fn select_offspring(&self, population: &Population<G, A>) -> Population<G, A> {
        let selector = self.offspring_selector();
        let count = self.offspring_count();

        selector.select(population, count)
    }

    pub fn alter(&self, population: &mut Population<G, A>, generation: i32) {
        let alterer = self.alterer();
        let optimize = self.optimize();

        alterer.alter(population, optimize, generation);
    }

    pub fn filter(&self, population: &mut Population<G, A>, generation: i32) {
        let max_age = self.params.max_age;
        let codex = self.codex();

        for individual in population.iter_mut() {
            if individual.age(generation) > max_age {
                *individual = Phenotype::from_genotype(codex.encode(), generation);
            } else if !individual.genotype().is_valid() {
                *individual = Phenotype::from_genotype(codex.encode(), generation);
            }
        }
    }

    pub fn recombine(
        &self,
        handle: &mut EngineContext<G, A, T>,
        survivors: Population<G, A>,
        offspring: Population<G, A>,
    ) {
        handle.population = survivors
            .into_iter()
            .chain(offspring.into_iter())
            .collect::<Population<G, A>>();
    }

    pub fn audit(&self, output: &mut EngineContext<G, A, T>) {
        let codex = self.codex();

        if !output.population.is_sorted {
            self.optimize().sort(&mut output.population);
        }

        output.best = codex.decode(&output.population.get(0).genotype());
        output.index += 1;
    }

    pub fn survivor_selector(&self) -> &impl Select<G, A> {
        &self.params.survivor_selector
    }

    pub fn offspring_selector(&self) -> &impl Select<G, A> {
        &self.params.offspring_selector
    }

    pub fn alterer(&self) -> &impl Alter<G, A> {
        self.params.alterer.as_ref().unwrap()
    }

    pub fn codex(&self) -> &Codex<G, A, T> {
        self.params.codex.as_ref().unwrap()
    }

    pub fn fitness_fn(&self) -> &impl Fn(&T) -> Score {
        self.params.fitness_fn.as_ref().unwrap()
    }

    pub fn population(&self) -> &Population<G, A> {
        self.params.population.as_ref().unwrap()
    }

    pub fn optimize(&self) -> &Optimize {
        &self.params.optimize
    }

    pub fn survivor_count(&self) -> usize {
        self.params.population_size - self.offspring_count()
    }

    pub fn offspring_count(&self) -> usize {
        (self.params.population_size as f32 * self.params.offspring_fraction) as usize
    }
}

impl<G: Gene<G, A>, A, T: Clone> Engine<G, A, T> for GeneticEngine<G, A, T> {
    fn fit<F: Fn(&EngineContext<G, A, T>) -> bool>(&self, limit: F) -> EngineContext<G, A, T> {
        let mut ctx = self.start();

        loop {
            self.evaluate(&mut ctx);

            let mut survivors = self.select_survivors(&ctx.population);
            let mut offspring = self.select_offspring(&ctx.population);

            self.alter(&mut offspring, ctx.index);

            self.filter(&mut survivors, ctx.index);
            self.filter(&mut offspring, ctx.index);

            self.recombine(&mut ctx, survivors, offspring);

            self.evaluate(&mut ctx);
            self.audit(&mut ctx);

            if limit(&ctx) {
                break self.stop(&mut ctx)
            }
        }
    }

    fn start(&self) -> EngineContext<G, A, T> {
        let population = self.population();

        EngineContext {
            population: population.clone(),
            best: self.codex().decode(&population.get(0).genotype()),
            index: 0,
            timer: Timer::new(),
        }
    }
}