// Evolve implements the inner loop of the evolutionary algorithm.
// The population calls the Evolve method of each genome, in parallel. Then,
// each receiver returns a value to replace it in the next generation. A global
// selector object synchronises replacement among the parallel parents.
func (ack *ackley) Evolve(suitors ...evo.Genome) evo.Genome {
for i := 0; i < 7; i++ {
// Creation:
// We create the child genome from recycled memory when we can.
var child ackley
child.gene = vectors.Get().(real.Vector)
child.steps = vectors.Get().(real.Vector)
// Crossover:
// Select two parents at random.
// Uniform crossover of object parameters.
// Arithmetic crossover of strategy parameters.
mom := suitors[rand.Intn(len(suitors))].(*ackley)
dad := suitors[rand.Intn(len(suitors))].(*ackley)
real.UniformX(child.gene, mom.gene, dad.gene)
real.ArithX(1, child.steps, mom.steps, dad.steps)
// Mutation: Evolution Strategy
// Lognormal scaling of strategy parameters.
// Gausian perturbation of object parameters.
child.steps.Adapt()
child.steps.LowBound(precision)
child.gene.Step(child.steps)
child.gene.HighBound(bounds)
child.gene.LowBound(-bounds)
// Replacement: (40,280)
// Each child is added to the global selection pool.
selector.Put(&child)
}
// Finally, block until all parallel calls have added their children to the
// selection pool and return one of the selected replacements.
return selector.Get()
}
func TestUniformX(t *testing.T) {
mom := real.Random(8, 1)
dad := real.Random(8, 1)
child := make([]float64, 8)
real.UniformX(child, mom, dad)
for i := range child {
if child[i] != mom[i] && child[i] != dad[i] {
t.Fail()
}
}
}
