Generate offspring by crossover

Real-coded mode uses simulated binary crossover (SBX); binary mode uses uniform crossover. When only one parent is available the offspring are exact copies of that parent.

Usage

crossover_cpp(
  selected_parents,
  offspring_size,
  crossover_rate = 0.9,
  eta_c = 20,
  representation = "real"
)

Arguments

selected_parents

Numeric matrix of selected individuals.

offspring_size

Number of offspring to generate.

crossover_rate

Probability of crossover per offspring (default: 0.9).

eta_c

SBX distribution index (default: 20.0).

representation

Either "real" or "binary".

Value

Numeric matrix of offspring.

References

Deb K, Agrawal RB (1995). "Simulated binary crossover for continuous search space." Complex Systems, 9(2), 115-148.

See also

initialize_population_cpp(), evaluate_fitness_cpp(), selection_cpp(), mutation_cpp(), replacement_cpp(), bioga_main_cpp()

Examples

genomic_data <- matrix(rnorm(100), nrow = 10, ncol = 10)
population <- BioGA::initialize_population_cpp(genomic_data, population_size = 5)
fitness <- BioGA::evaluate_fitness_cpp(genomic_data, population, c(1.0, 0.5))
parents <- BioGA::selection_cpp(population, fitness, num_parents = 2)
BioGA::crossover_cpp(parents, offspring_size = 2)
#>           [,1]       [,2]      [,3]      [,4]       [,5]       [,6]      [,7]
#> [1,] 0.8606541 -0.8968653 0.1844905 0.4581388 -0.4174473 -0.9783576 0.2208403
#> [2,] 0.8606541 -0.8968653 0.1844905 0.4581388 -0.4174473 -0.9783576 0.2208403
#>           [,8]       [,9]    [,10]
#> [1,] 0.3136316 -0.1592713 0.684092
#> [2,] 0.3136316 -0.1592713 0.684092