c

scalation.minima

GeneticAlgorithm

class GeneticAlgorithm extends AnyRef

The GeneticAlgorithm class performs local search to find minima of functions defined on integer vector domains (z^n).

minimize f(x) subject to g(x) <= 0, x in Z^n

Linear Supertypes
AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. GeneticAlgorithm
  2. AnyRef
  3. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. Protected

Instance Constructors

  1. new GeneticAlgorithm(f: (VectorI) => Double, x0: VectorI, vMax: Int = 100, g: (VectorI) => Double = null, maxStep: Int = 5)

    f

    the objective function to be minimize (f maps an integer vector to a double)

    x0

    the starting point for the search (seed for GA)

    g

    the constraint function to be satisfied, if any

    maxStep

    the maximum/starting step size (make larger for larger domains)

Type Members

  1. type Vec_Func = (VectorI, Double)

    Pair consisting of an integer vector and its functional value (a double)

Value Members

  1. def crossOver(): Unit

    For each individual in the population, cross it with some other individual.

  2. def fg(x: VectorI): Double

    The objective function f re-scaled by a weighted penalty, if constrained.

    The objective function f re-scaled by a weighted penalty, if constrained.

    x

    the coordinate values of the current point

  3. def fitnessCrossOver(): Unit

    For each individual in the population, cross it with some other individual.

    For each individual in the population, cross it with some other individual. Let the crossover be dependent of the fitness of the individual.

  4. def fittest: Vec_Func

    Find the fittest individual (smallest value of objective function).

  5. def genPopulation(): Unit

    Generate an initial population of individuals.

  6. def mutate(): Unit

    Randomly select individuals for mutation (change a value at one position).

  7. def printPopulation(): Unit

    Print the current population

  8. def solve(): Vec_Func

    Solve the minimization problem using a genetic algorithm.