/**
* makes a copy of this GABitSet
*
* @return a copy of the object
* @throws CloneNotSupportedException if something goes wrong
*/
public Object clone() throws CloneNotSupportedException {
GABitSet temp = new GABitSet();
temp.setObjective(this.getObjective());
temp.setFitness(this.getFitness());
temp.setChromosome((BitSet) (this.m_chromosome.clone()));
return temp;
// return super.clone();
}
/**
* evaluates an entire population. Population members are looked up in a hash table and if they
* are not found then they are evaluated using ASEvaluator.
*
* @param ASEvaluator the subset evaluator to use for evaluating population members
* @throws Exception if something goes wrong during evaluation
*/
private void evaluatePopulation(SubsetEvaluator ASEvaluator) throws Exception {
int i;
double merit;
for (i = 0; i < m_popSize; i++) {
// if its not in the lookup table then evaluate and insert
if (m_lookupTable.containsKey(m_population[i].getChromosome()) == false) {
merit = ASEvaluator.evaluateSubset(m_population[i].getChromosome());
m_population[i].setObjective(merit);
m_lookupTable.put(m_population[i].getChromosome(), m_population[i]);
} else {
GABitSet temp = m_lookupTable.get(m_population[i].getChromosome());
m_population[i].setObjective(temp.getObjective());
}
}
}
/**
* Searches the attribute subset space using a genetic algorithm.
*
* @param ASEval the attribute evaluator to guide the search
* @param data the training instances.
* @return an array (not necessarily ordered) of selected attribute indexes
* @throws Exception if the search can't be completed
*/
@Override
public int[] search(ASEvaluation ASEval, Instances data) throws Exception {
m_best = null;
m_generationReports = new StringBuffer();
if (!(ASEval instanceof SubsetEvaluator)) {
throw new Exception(ASEval.getClass().getName() + " is not a " + "Subset evaluator!");
}
if (ASEval instanceof UnsupervisedSubsetEvaluator) {
m_hasClass = false;
} else {
m_hasClass = true;
m_classIndex = data.classIndex();
}
SubsetEvaluator ASEvaluator = (SubsetEvaluator) ASEval;
m_numAttribs = data.numAttributes();
m_startRange.setUpper(m_numAttribs - 1);
if (!(getStartSet().equals(""))) {
m_starting = m_startRange.getSelection();
}
// initial random population
m_lookupTable = new Hashtable<BitSet, GABitSet>(m_lookupTableSize);
m_random = new Random(m_seed);
m_population = new GABitSet[m_popSize];
// set up random initial population
initPopulation();
evaluatePopulation(ASEvaluator);
populationStatistics();
scalePopulation();
checkBest();
m_generationReports.append(populationReport(0));
boolean converged;
for (int i = 1; i <= m_maxGenerations; i++) {
generation();
evaluatePopulation(ASEvaluator);
populationStatistics();
scalePopulation();
// find the best pop member and check for convergence
converged = checkBest();
if ((i == m_maxGenerations) || ((i % m_reportFrequency) == 0) || (converged == true)) {
m_generationReports.append(populationReport(i));
if (converged == true) {
break;
}
}
}
return attributeList(m_best.getChromosome());
}
/**
* checks to see if any population members in the current population are better than the best
* found so far. Also checks to see if the search has converged---that is there is no difference
* in fitness between the best and worse population member
*
* @return true is the search has converged
* @throws Exception if something goes wrong
*/
private boolean checkBest() throws Exception {
int i, count, lowestCount = m_numAttribs;
double b = -Double.MAX_VALUE;
GABitSet localbest = null;
BitSet temp;
boolean converged = false;
int oldcount = Integer.MAX_VALUE;
if (m_maxFitness - m_minFitness > 0) {
// find the best in this population
for (i = 0; i < m_popSize; i++) {
if (m_population[i].getObjective() > b) {
b = m_population[i].getObjective();
localbest = m_population[i];
oldcount = countFeatures(localbest.getChromosome());
} else if (Utils.eq(m_population[i].getObjective(), b)) {
// see if it contains fewer features
count = countFeatures(m_population[i].getChromosome());
if (count < oldcount) {
b = m_population[i].getObjective();
localbest = m_population[i];
oldcount = count;
}
}
}
} else {
// look for the smallest subset
for (i = 0; i < m_popSize; i++) {
temp = m_population[i].getChromosome();
count = countFeatures(temp);
;
if (count < lowestCount) {
lowestCount = count;
localbest = m_population[i];
b = localbest.getObjective();
}
}
converged = true;
}
// count the number of features in localbest
count = 0;
temp = localbest.getChromosome();
count = countFeatures(temp);
// compare to the best found so far
if (m_best == null) {
m_best = (GABitSet) localbest.clone();
m_bestFeatureCount = count;
} else if (b > m_best.getObjective()) {
m_best = (GABitSet) localbest.clone();
m_bestFeatureCount = count;
} else if (Utils.eq(m_best.getObjective(), b)) {
// see if the localbest has fewer features than the best so far
if (count < m_bestFeatureCount) {
m_best = (GABitSet) localbest.clone();
m_bestFeatureCount = count;
}
}
return converged;
}