pub struct SimulatedAnnealingParameters<T, M>where
T: Clone,
M: MutationOperator<T>,{
pub mutation_operator: M,
pub mutation_probability: f64,
pub initial_temperature: f64,
pub minimum_temperature: f64,
pub cooling_rate: f64,
pub termination_criteria: TerminationCriteria,
pub random_seed: Option<u64>,
/* private fields */
}Fields§
§mutation_operator: M§mutation_probability: f64§initial_temperature: f64§minimum_temperature: f64§cooling_rate: f64§termination_criteria: TerminationCriteria§random_seed: Option<u64>Implementations§
Source§impl<T, M> SimulatedAnnealingParameters<T, M>where
T: Clone,
M: MutationOperator<T>,
impl<T, M> SimulatedAnnealingParameters<T, M>where
T: Clone,
M: MutationOperator<T>,
Sourcepub fn new(
mutation_operator: M,
mutation_probability: f64,
initial_temperature: f64,
cooling_rate: f64,
termination_criteria: TerminationCriteria,
) -> Self
pub fn new( mutation_operator: M, mutation_probability: f64, initial_temperature: f64, cooling_rate: f64, termination_criteria: TerminationCriteria, ) -> Self
Examples found in repository?
examples/experiment_parallel_vs_sequential.rs (lines 59-65)
35fn run_experiment(parallel: bool, runs: usize) -> Result<(Duration, ExperimentReport), String> {
36 let problem = build_problem();
37
38 let hill_climbing_case = HillClimbingParameters::new(
39 BitFlipMutation::new(),
40 0.12,
41 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(180)]),
42 )
43 .with_seed(111);
44
45 // Keep GA internally sequential to focus the comparison on experiment-level parallelism.
46 let genetic_algorithm_case = GeneticAlgorithmParameters::new(
47 80,
48 0.90,
49 0.06,
50 SinglePointCrossover::new(),
51 BitFlipMutation::new(),
52 BinaryTournamentSelection::new(),
53 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(60)]),
54 )
55 .with_elite_size(1)
56 .with_seed(222)
57 .sequential();
58
59 let simulated_annealing_case = SimulatedAnnealingParameters::new(
60 BitFlipMutation::new(),
61 0.10,
62 45.0,
63 0.985,
64 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(220)]),
65 )
66 .with_seed(333);
67
68 let pso_case = PSOParameters::new(
69 50,
70 0.72,
71 1.49,
72 1.49,
73 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(120)]),
74 )
75 .with_velocity_clamp(4.0)
76 .with_seed(444);
77
78 let experiment = Experiment::new(problem)
79 .with_runs(runs)
80 .add_case(hill_climbing_case)
81 .add_case(genetic_algorithm_case)
82 .add_case(simulated_annealing_case)
83 .add_case(pso_case);
84
85 measure_result(|| {
86 if parallel {
87 experiment.with_parallel().execute()
88 } else {
89 experiment.sequential().execute()
90 }
91 })
92}More examples
examples/mono_objective_experiment.rs (lines 52-58)
13fn main() {
14 let problem = KnapsackBuilder::new()
15 .with_capacity(150.0)
16 .add_item(1.0, 2.0)
17 .add_item(1.0, 2.0)
18 .add_item(2.0, 6.0)
19 .add_item(2.0, 6.5)
20 .add_item(3.0, 7.0)
21 .add_item(10.0, 20.0)
22 .add_item(20.0, 30.0)
23 .add_item(30.0, 60.0)
24 .add_item(35.0, 65.0)
25 .add_item(45.0, 100.0)
26 .add_item(55.0, 120.0)
27 .add_item(75.0, 211.0)
28 .add_item(75.0, 211.0)
29 .add_item(80.0, 160.0)
30 .add_item(90.0, 301.0)
31 .add_item(150.0, 301.0)
32 .build();
33
34 let hill_climbing_case = HillClimbingParameters::new(
35 BitFlipMutation::new(),
36 0.12,
37 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(180)]),
38 );
39
40 let genetic_algorithm_case = GeneticAlgorithmParameters::new(
41 80,
42 0.90,
43 0.06,
44 SinglePointCrossover::new(),
45 BitFlipMutation::new(),
46 BinaryTournamentSelection::new(),
47 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(60)]),
48 )
49 .with_elite_size(1)
50 .with_threads(4);
51
52 let simulated_annealing_case = SimulatedAnnealingParameters::new(
53 BitFlipMutation::new(),
54 0.10,
55 45.0,
56 0.985,
57 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(220)]),
58 )
59 .with_seed(777);
60
61 let pso_case = PSOParameters::new(
62 50,
63 0.72,
64 1.49,
65 1.49,
66 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(120)]),
67 )
68 .with_velocity_clamp(4.0)
69 .with_seed(999);
70
71 let report = Experiment::new(problem)
72 .with_runs(24)
73 .add_case(hill_climbing_case)
74 .add_case(genetic_algorithm_case)
75 .add_case(simulated_annealing_case)
76 .add_case(pso_case)
77 .execute();
78
79 match report {
80 Ok(report) => println!("{}", report.to_text_table()),
81 Err(error) => eprintln!("Experiment execution failed: {}", error),
82 }
83}pub fn with_minimum_temperature(self, minimum_temperature: f64) -> Self
Sourcepub fn with_seed(self, seed: u64) -> Self
pub fn with_seed(self, seed: u64) -> Self
Examples found in repository?
examples/experiment_parallel_vs_sequential.rs (line 66)
35fn run_experiment(parallel: bool, runs: usize) -> Result<(Duration, ExperimentReport), String> {
36 let problem = build_problem();
37
38 let hill_climbing_case = HillClimbingParameters::new(
39 BitFlipMutation::new(),
40 0.12,
41 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(180)]),
42 )
43 .with_seed(111);
44
45 // Keep GA internally sequential to focus the comparison on experiment-level parallelism.
46 let genetic_algorithm_case = GeneticAlgorithmParameters::new(
47 80,
48 0.90,
49 0.06,
50 SinglePointCrossover::new(),
51 BitFlipMutation::new(),
52 BinaryTournamentSelection::new(),
53 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(60)]),
54 )
55 .with_elite_size(1)
56 .with_seed(222)
57 .sequential();
58
59 let simulated_annealing_case = SimulatedAnnealingParameters::new(
60 BitFlipMutation::new(),
61 0.10,
62 45.0,
63 0.985,
64 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(220)]),
65 )
66 .with_seed(333);
67
68 let pso_case = PSOParameters::new(
69 50,
70 0.72,
71 1.49,
72 1.49,
73 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(120)]),
74 )
75 .with_velocity_clamp(4.0)
76 .with_seed(444);
77
78 let experiment = Experiment::new(problem)
79 .with_runs(runs)
80 .add_case(hill_climbing_case)
81 .add_case(genetic_algorithm_case)
82 .add_case(simulated_annealing_case)
83 .add_case(pso_case);
84
85 measure_result(|| {
86 if parallel {
87 experiment.with_parallel().execute()
88 } else {
89 experiment.sequential().execute()
90 }
91 })
92}More examples
examples/mono_objective_experiment.rs (line 59)
13fn main() {
14 let problem = KnapsackBuilder::new()
15 .with_capacity(150.0)
16 .add_item(1.0, 2.0)
17 .add_item(1.0, 2.0)
18 .add_item(2.0, 6.0)
19 .add_item(2.0, 6.5)
20 .add_item(3.0, 7.0)
21 .add_item(10.0, 20.0)
22 .add_item(20.0, 30.0)
23 .add_item(30.0, 60.0)
24 .add_item(35.0, 65.0)
25 .add_item(45.0, 100.0)
26 .add_item(55.0, 120.0)
27 .add_item(75.0, 211.0)
28 .add_item(75.0, 211.0)
29 .add_item(80.0, 160.0)
30 .add_item(90.0, 301.0)
31 .add_item(150.0, 301.0)
32 .build();
33
34 let hill_climbing_case = HillClimbingParameters::new(
35 BitFlipMutation::new(),
36 0.12,
37 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(180)]),
38 );
39
40 let genetic_algorithm_case = GeneticAlgorithmParameters::new(
41 80,
42 0.90,
43 0.06,
44 SinglePointCrossover::new(),
45 BitFlipMutation::new(),
46 BinaryTournamentSelection::new(),
47 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(60)]),
48 )
49 .with_elite_size(1)
50 .with_threads(4);
51
52 let simulated_annealing_case = SimulatedAnnealingParameters::new(
53 BitFlipMutation::new(),
54 0.10,
55 45.0,
56 0.985,
57 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(220)]),
58 )
59 .with_seed(777);
60
61 let pso_case = PSOParameters::new(
62 50,
63 0.72,
64 1.49,
65 1.49,
66 TerminationCriteria::new(vec![TerminationCriterion::MaxIterations(120)]),
67 )
68 .with_velocity_clamp(4.0)
69 .with_seed(999);
70
71 let report = Experiment::new(problem)
72 .with_runs(24)
73 .add_case(hill_climbing_case)
74 .add_case(genetic_algorithm_case)
75 .add_case(simulated_annealing_case)
76 .add_case(pso_case)
77 .execute();
78
79 match report {
80 Ok(report) => println!("{}", report.to_text_table()),
81 Err(error) => eprintln!("Experiment execution failed: {}", error),
82 }
83}Trait Implementations§
Source§impl<T, M> Clone for SimulatedAnnealingParameters<T, M>
impl<T, M> Clone for SimulatedAnnealingParameters<T, M>
Source§fn clone(&self) -> SimulatedAnnealingParameters<T, M>
fn clone(&self) -> SimulatedAnnealingParameters<T, M>
Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreSource§impl<T, M, P> ExperimentalCase<T, f64, P> for SimulatedAnnealingParameters<T, M>
impl<T, M, P> ExperimentalCase<T, f64, P> for SimulatedAnnealingParameters<T, M>
fn algorithm_name(&self) -> &str
Source§fn parameters(&self) -> Vec<CaseParameter>
fn parameters(&self) -> Vec<CaseParameter>
Returns generic parameter key/value pairs for reporting.
Source§fn run(&self, problem: &P) -> Result<Box<dyn SolutionSet<T, f64>>, String>
fn run(&self, problem: &P) -> Result<Box<dyn SolutionSet<T, f64>>, String>
Creates and executes the algorithm with its configured parameters.
Source§fn parameters_as_text(&self) -> String
fn parameters_as_text(&self) -> String
Helper to print all parameters in a single textual line.
Auto Trait Implementations§
impl<T, M> Freeze for SimulatedAnnealingParameters<T, M>where
M: Freeze,
impl<T, M> RefUnwindSafe for SimulatedAnnealingParameters<T, M>where
M: RefUnwindSafe,
T: RefUnwindSafe,
impl<T, M> Send for SimulatedAnnealingParameters<T, M>
impl<T, M> Sync for SimulatedAnnealingParameters<T, M>
impl<T, M> Unpin for SimulatedAnnealingParameters<T, M>
impl<T, M> UnsafeUnpin for SimulatedAnnealingParameters<T, M>where
M: UnsafeUnpin,
impl<T, M> UnwindSafe for SimulatedAnnealingParameters<T, M>where
M: UnwindSafe,
T: UnwindSafe,
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more