示例#1
0
 /**
  * Constructor. Creates a new instance of Spea2Fitness for a given <code>SolutionSet</code>.
  *
  * @param solutionSet The <code>SolutionSet</code>
  */
 public Spea2Fitness(SolutionSet solutionSet) {
   distance = distance_.distanceMatrix(solutionSet);
   solutionSet_ = solutionSet;
   for (int i = 0; i < solutionSet_.size(); i++) {
     solutionSet_.get(i).setLocation(i);
   } // for
 } // Spea2Fitness
示例#2
0
  /** Assigns fitness for all the solutions. */
  public void fitnessAssign() {
    double[] strength = new double[solutionSet_.size()];
    double[] rawFitness = new double[solutionSet_.size()];
    double kDistance;

    // Calculate the strength value
    // strength(i) = |{j | j <- SolutionSet and i dominate j}|
    for (int i = 0; i < solutionSet_.size(); i++) {
      for (int j = 0; j < solutionSet_.size(); j++) {
        if (dominance_.compare(solutionSet_.get(i), solutionSet_.get(j)) == -1) {
          strength[i] += 1.0;
        } // if
      } // for
    } // for

    // Calculate the raw fitness
    // rawFitness(i) = |{sum strenght(j) | j <- SolutionSet and j dominate i}|
    for (int i = 0; i < solutionSet_.size(); i++) {
      for (int j = 0; j < solutionSet_.size(); j++) {
        if (dominance_.compare(solutionSet_.get(i), solutionSet_.get(j)) == 1) {
          rawFitness[i] += strength[j];
        } // if
      } // for
    } // for

    // Add the distance to the k-th individual. In the reference paper of SPEA2,
    // k = sqrt(population.size()), but a value of k = 1 recommended. See
    // http://www.tik.ee.ethz.ch/pisa/selectors/spea2/spea2_documentation.txt
    int k = 1;
    for (int i = 0; i < distance.length; i++) {
      Arrays.sort(distance[i]);
      kDistance = 1.0 / (distance[i][k] + 2.0); // Calcule de D(i) distance
      // population.get(i).setFitness(rawFitness[i]);
      solutionSet_.get(i).setFitness(rawFitness[i] + kDistance);
    } // for
  } // fitnessAsign
示例#3
0
文件: Main.java 项目: ai-se/SPL
  /** @param args the command line arguments -- modelname, alg_name, evaluation_times, [runid] */
  public static void main(String[] args) throws Exception {

    try {
      String name = args[0];
      URL location = Main.class.getProtectionDomain().getCodeSource().getLocation();
      String loc = location.toString();
      String project_path =
          loc.substring(5, loc.lastIndexOf("SPL/")) + "SPL/"; // with '/' at the end
      String fm = project_path + "dimacs_data/" + name + ".dimacs";
      String augment = fm + ".augment";
      String dead = fm + ".dead";
      String mandatory = fm + ".mandatory";
      String seed = fm + ".richseed";
      String opfile = fm + ".sipop";

      Problem p = new ProductLineProblem(fm, augment, mandatory, dead, seed);
      //            Problem p = new ProductLineProblemNovelPrep(fm, augment, mandatory, dead, seed,
      // opfile);
      //            GroupedProblem.grouping((ProductLineProblem) p, 100); System.exit(0);
      Algorithm a;
      int evaluation_times = Integer.parseInt(args[2]);
      String alg_name = args[1];
      String runid = "";
      if (args.length >= 4) {
        runid = args[3];
      }

      switch (alg_name) {
        case "IBEA":
          a = new SPL_SettingsIBEA(p).configureICSE2013(evaluation_times);
          break;
        case "SIPIBEA":
          a = new SPL_SettingsIBEA(p).configureSIPIBEA(evaluation_times);
          break;
        case "SPEA2":
          a = new SPL_SettingsEMOs(p).configureSPEA2(evaluation_times);
          break;
        case "NSGA2":
          a = new SPL_SettingsEMOs(p).configureNSGA2(evaluation_times);
          break;
        case "IBEASEED":
          a = new SPL_SettingsIBEA(p).configureIBEASEED(evaluation_times);
          break;
        case "SATIBEA":
          // a = new SPL_SettingsIBEA(p).configureICSE15(1000, fm, ((ProductLineProblem)
          // p).getNumFeatures(),
          // ((ProductLineProblem) p).getConstraints());
          a =
              new SPL_SettingsIBEA(p)
                  .configureSATIBEA(
                      evaluation_times,
                      fm,
                      ((ProductLineProblem) p).getNumFeatures(),
                      ((ProductLineProblem) p).getConstraints());
          break;
        default:
          a = new SPL_SettingsIBEA(p).configureICSE2013(evaluation_times);
      }

      long start = System.currentTimeMillis();
      SolutionSet pop = a.execute();
      float total_time = (System.currentTimeMillis() - start) / 1000.0f;

      String file_tag =
          name + "_" + alg_name + '_' + evaluation_times / 1000 + "k_" + runid + ".txt";
      String file_path = project_path + "j_res/" + file_tag;
      File file = new File(file_path);

      if (!file.exists()) {
        file.createNewFile();
      }

      FileWriter fw = new FileWriter(file.getAbsoluteFile());
      BufferedWriter bw = new BufferedWriter(fw);

      for (int i = 0; i < pop.size(); i++) {
        Variable v = pop.get(i).getDecisionVariables()[0];
        bw.write((Binary) v + "\n");
        System.out.println("Conf" + (i + 1) + ": " + (Binary) v + " ");
      }

      bw.write("~~~\n");

      for (int i = 0; i < pop.size(); i++) {
        Variable v = pop.get(i).getDecisionVariables()[0];
        for (int j = 0; j < pop.get(i).getNumberOfObjectives(); j++) {
          bw.write(pop.get(i).getObjective(j) + " ");
          System.out.print(pop.get(i).getObjective(j) + " ");
        }
        bw.write("\n");
        System.out.println("");
      }

      System.out.println(total_time);
      bw.write("~~~\n" + total_time + "\n");

      bw.close();
      fw.close();

    } catch (Exception e) {
      e.printStackTrace();
    }
  }
示例#4
0
  /**
   * Gets 'size' elements from a population of more than 'size' elements using for this de
   * enviromentalSelection truncation
   *
   * @param size The number of elements to get.
   */
  public SolutionSet environmentalSelection(int size) {

    if (solutionSet_.size() < size) {
      size = solutionSet_.size();
    }

    // Create a new auxiliar population for no alter the original population
    SolutionSet aux = new SolutionSet(solutionSet_.size());

    int i = 0;
    while (i < solutionSet_.size()) {
      if (solutionSet_.get(i).getFitness() < 1.0) {
        aux.add(solutionSet_.get(i));
        solutionSet_.remove(i);
      } else {
        i++;
      } // if
    } // while

    if (aux.size() < size) {
      Comparator comparator = new FitnessComparator();
      solutionSet_.sort(comparator);
      int remain = size - aux.size();
      for (i = 0; i < remain; i++) {
        aux.add(solutionSet_.get(i));
      }
      return aux;
    } else if (aux.size() == size) {
      return aux;
    }

    double[][] distance = distance_.distanceMatrix(aux);
    List<List<DistanceNode>> distanceList = new LinkedList<List<DistanceNode>>();
    for (int pos = 0; pos < aux.size(); pos++) {
      aux.get(pos).setLocation(pos);
      List<DistanceNode> distanceNodeList = new ArrayList<DistanceNode>();
      for (int ref = 0; ref < aux.size(); ref++) {
        if (pos != ref) {
          distanceNodeList.add(new DistanceNode(distance[pos][ref], ref));
        } // if
      } // for
      distanceList.add(distanceNodeList);
    } // for

    for (int q = 0; q < distanceList.size(); q++) {
      Collections.sort(distanceList.get(q), distanceNodeComparator);
    } // for

    while (aux.size() > size) {
      double minDistance = Double.MAX_VALUE;
      int toRemove = 0;
      i = 0;
      Iterator<List<DistanceNode>> iterator = distanceList.iterator();
      while (iterator.hasNext()) {
        List<DistanceNode> dn = iterator.next();
        if (dn.get(0).getDistance() < minDistance) {
          toRemove = i;
          minDistance = dn.get(0).getDistance();
          // i y toRemove have the same distance to the first solution
        } else if (dn.get(0).getDistance() == minDistance) {
          int k = 0;
          while ((dn.get(k).getDistance() == distanceList.get(toRemove).get(k).getDistance())
              && k < (distanceList.get(i).size() - 1)) {
            k++;
          }

          if (dn.get(k).getDistance() < distanceList.get(toRemove).get(k).getDistance()) {
            toRemove = i;
          } // if
        } // if
        i++;
      } // while

      int tmp = aux.get(toRemove).getLocation();
      aux.remove(toRemove);
      distanceList.remove(toRemove);

      Iterator<List<DistanceNode>> externIterator = distanceList.iterator();
      while (externIterator.hasNext()) {
        Iterator<DistanceNode> interIterator = externIterator.next().iterator();
        while (interIterator.hasNext()) {
          if (interIterator.next().getReference() == tmp) {
            interIterator.remove();
            continue;
          } // if
        } // while
      } // while
    } // while
    return aux;
  } // environmentalSelection
  /**
   * @author Juanjo This method selects N solutions from a set M, where N <= M using the same method
   *     proposed by Qingfu Zhang, W. Liu, and Hui Li in the paper describing MOEA/D-DRA (CEC 09
   *     COMPTETITION) An example is giving in that paper for two objectives. If N = 100, then the
   *     best solutions attenting to the weights (0,1), (1/99,98/99), ...,(98/99,1/99), (1,0) are
   *     selected.
   *     <p>Using this method result in 101 solutions instead of 100. We will just compute 100 even
   *     distributed weights and used them. The result is the same
   *     <p>In case of more than two objectives the procedure is: 1- Select a solution at random 2-
   *     Select the solution from the population which have maximum distance to it (whithout
   *     considering the already included)
   * @param n: The number of solutions to return
   * @return A solution set containing those elements
   */
  SolutionSet finalSelection(int n) throws JMException {
    SolutionSet res = new SolutionSet(n);
    if (problem_.getNumberOfObjectives() == 2) { // subcase 1
      double[][] intern_lambda = new double[n][2];
      for (int i = 0; i < n; i++) {
        double a = 1.0 * i / (n - 1);
        intern_lambda[i][0] = a;
        intern_lambda[i][1] = 1 - a;
      } // for

      // we have now the weights, now select the best solution for each of them
      for (int i = 0; i < n; i++) {
        Solution current_best = population.get(0);
        int index = 0;
        double value = fitnessFunction(current_best, intern_lambda[i]);
        for (int j = 1; j < n; j++) {
          double aux =
              fitnessFunction(
                  population.get(j), intern_lambda[i]); // we are looking the best for the weight i
          if (aux < value) { // solution in position j is better!
            value = aux;
            current_best = population.get(j);
          }
        }
        res.add(new Solution(current_best));
      }

    } else { // general case (more than two objectives)

      Distance distance_utility = new Distance();
      int random_index = PseudoRandom.randInt(0, population.size() - 1);

      // create a list containing all the solutions but the selected one (only references to them)
      List<Solution> candidate = new LinkedList<Solution>();
      candidate.add(population.get(random_index));

      for (int i = 0; i < population.size(); i++) {
        if (i != random_index) {
          candidate.add(population.get(i));
        }
      } // for

      while (res.size() < n) {
        int index = 0;
        Solution selected = candidate.get(0); // it should be a next! (n <= population size!)
        double distance_value =
            distance_utility.distanceToSolutionSetInObjectiveSpace(selected, res);
        int i = 1;
        while (i < candidate.size()) {
          Solution next_candidate = candidate.get(i);
          double aux =
              distance_value =
                  distance_utility.distanceToSolutionSetInObjectiveSpace(next_candidate, res);
          if (aux > distance_value) {
            distance_value = aux;
            index = i;
          }
          i++;
        }

        // add the selected to res and remove from candidate list
        res.add(new Solution(candidate.remove(index)));
      } //
    }
    return res;
  }