@Test
public void testFindOptimumForRosenbrockFunction() throws Exception {
MultiparamDoubleFunction function =
individual -> {
double[] x = individual.getValue();
double sum = 0;
for (int i = 0; i < x.length - 1; i++) {
sum += 100 * Math.pow(x[i + 1] - Math.pow(x[i], 2), 2) + Math.pow(x[i] - 1, 2);
}
return sum;
};
DifferentialEvolution ev =
new DifferentialEvolution(
100000,
function,
new MultiparamDoubleIndividual(15, 4, 7, 20, 5),
new MultiparamDoubleIndividual(10, 6, 2, 88, 2),
new MultiparamDoubleIndividual(6, 3, 5, 1, 0),
new MultiparamDoubleIndividual(0, 0, 1, 3, 0),
new MultiparamDoubleIndividual(-2, -1, 0, 5, 6),
new MultiparamDoubleIndividual(0, 1, 2, 3, 4));
printMultipartResult(((MultiparamDoubleIndividual) ev.findOptimum()));
assertAllParamsCloseToZero(((MultiparamDoubleIndividual) ev.findOptimum()));
}
@Test
public void testFindOptimumForSphereFunction() throws Exception {
final int n = 5;
DoubleIndividualFunction function =
individual -> {
double sum = 0;
for (int i = 0; i < n; i++) {
sum += individual.getValue() * individual.getValue();
}
return sum;
};
DifferentialEvolution ev =
new DifferentialEvolution(
100000,
function,
new DoubleIndividual(15),
new DoubleIndividual(10),
new DoubleIndividual(6),
new DoubleIndividual(12),
new DoubleIndividual(-2),
new DoubleIndividual(-6));
assertEquals(0, ((DoubleIndividual) ev.findOptimum()).getValue(), 0.2);
}
@Test
public void testFindOptimumForSphereFunction2() throws Exception {
MultiparamDoubleFunction function =
individual -> {
double sum = 0;
for (int i = 0; i < individual.getValue().length; i++) {
sum += individual.getValue()[i] * individual.getValue()[i];
}
return sum;
};
DifferentialEvolution ev =
new DifferentialEvolution(
100000,
function,
new MultiparamDoubleIndividual(15, 4, 7, 20, 5),
new MultiparamDoubleIndividual(10, 6, 2, 88, 2),
new MultiparamDoubleIndividual(6, 3, 5, 1, 0),
new MultiparamDoubleIndividual(0, 0, 1, 3, 0),
new MultiparamDoubleIndividual(-2, -1, 0, 5, 6),
new MultiparamDoubleIndividual(0, 1, 2, 3, 4));
printMultipartResult(((MultiparamDoubleIndividual) ev.findOptimum()));
assertAllParamsCloseToZero(((MultiparamDoubleIndividual) ev.findOptimum()));
}
/**
* The actual optimization routine It finds a minimum close to vector x when the absolute
* tolerance for each parameter is specified.
*
* @param f multivariate function
* @param xvec initial guesses for the minimum (contains the location of the minimum on return)
* @param tolfx absolute tolerance of function value
* @param tolx absolute tolerance of each parameter
*/
public void optimize(MultivariateFunction f, double[] xvec, double tolfx, double tolx) {
if (optimiser_ == null || xvec.length != currentNumberOfParameters_) {
optimiser_ = new DifferentialEvolution(xvec.length);
this.currentNumberOfParameters_ = xvec.length;
}
optimiser_.optimize(f, xvec, tolfx, tolx);
}
@Test
public void testFindOptimum() throws Exception {
DoubleIndividualFunction function = i -> i.getValue() * i.getValue();
DifferentialEvolution ev =
new DifferentialEvolution(
100,
function,
new DoubleIndividual(1555),
new DoubleIndividual(1200),
new DoubleIndividual(2),
new DoubleIndividual(0.2),
new DoubleIndividual(-2.6),
new DoubleIndividual(4));
System.out.println(((DoubleIndividual) ev.findOptimum()).getValue());
assertEquals(0, ((DoubleIndividual) ev.findOptimum()).getValue(), 0.2);
}