geneticsearch.java

一个数据挖掘软件ALPHAMINERR的整个过程的JAVA版源代码

					while (bSelected[iSelected2]) {
						iSelected2 = (iSelected2 + 1) % getDescendantPopulationSize();
					}
					if (descendantPopulation[iSelected2].getScore() > descendantPopulation[iSelected].getScore()) {
						iSelected = iSelected2;
					}
				} else {
					// find best scoring network in population
					while (bSelected[iSelected]) {
						iSelected++;
					}
					double fScore = descendantPopulation[iSelected].getScore();
					for (int j = 0; j < getDescendantPopulationSize(); j++) {
						if (!bSelected[j] && descendantPopulation[j].getScore() > fScore) {
							fScore = descendantPopulation[j].getScore();
							iSelected = j;
						}
					}
				}
				population[i] = descendantPopulation[iSelected];
				bSelected[iSelected] = true;
			}
        }
        
        // restore current network to best network
		copyParentSets(bayesNet, bestBayesNet);
		
		// free up memory
		bestBayesNet = null;
    } // search


	/** copyParentSets copies parent sets of source to dest BayesNet
	 * @param dest: destination network
	 * @param source: source network
	 */
	void copyParentSets(BayesNet dest, BayesNet source) {
		int nNodes = source.getNrOfNodes();
		// clear parent set first
		for (int iNode = 0; iNode < nNodes; iNode++) {
			dest.getParentSet(iNode).copy(source.getParentSet(iNode));
		}		
	} // CopyParentSets

    /**
    * @return number of runs
    */
    public int getRuns() {
        return m_nRuns;
    } // getRuns

    /**
     * Sets the number of runs
     * @param nRuns The number of runs to set
     */
    public void setRuns(int nRuns) {
        m_nRuns = nRuns;
    } // setRuns

	/**
	 * Returns an enumeration describing the available options.
	 *
	 * @return an enumeration of all the available options.
	 */
	public Enumeration listOptions() {
		Vector newVector = new Vector(7);

		newVector.addElement(new Option("\tPopulation size\n", "L", 1, "-L <integer>"));
		newVector.addElement(new Option("\tDescendant population size\n", "A", 1, "-A <integer>"));
		newVector.addElement(new Option("\tNumber of runs\n", "U", 1, "-U <integer>"));
		newVector.addElement(new Option("\tUse mutation.\n\t(default true)", "M", 0, "-M"));
		newVector.addElement(new Option("\tUse cross-over.\n\t(default true)", "C", 0, "-C"));
		newVector.addElement(new Option("\tUse tournament selection (true) or maximum subpopulatin (false).\n\t(default false)", "O", 0, "-O"));
		newVector.addElement(new Option("\tRandom number seed\n", "R", 1, "-R <seed>"));

		Enumeration em = super.listOptions();
		while (em.hasMoreElements()) {
			newVector.addElement(em.nextElement());
		}
		return newVector.elements();
	} // listOptions

	/**
	 * Parses a given list of options. Valid options are:<p>
	 *
	 * For other options see search algorithm.
	 *
	 * @param options the list of options as an array of strings
	 * @exception Exception if an option is not supported
	 */
	public void setOptions(String[] options) throws Exception {
		String sPopulationSize = Utils.getOption('L', options);
		if (sPopulationSize.length() != 0) {
			setPopulationSize(Integer.parseInt(sPopulationSize));
		}
		String sDescendantPopulationSize = Utils.getOption('A', options);
		if (sDescendantPopulationSize.length() != 0) {
			setDescendantPopulationSize(Integer.parseInt(sDescendantPopulationSize));
		}
		String sRuns = Utils.getOption('U', options);
		if (sRuns.length() != 0) {
			setRuns(Integer.parseInt(sRuns));
		}
		String sSeed = Utils.getOption('R', options);
		if (sSeed.length() != 0) {
			setSeed(Integer.parseInt(sSeed));
		}
		setUseMutation(Utils.getFlag('M', options));
		setUseCrossOver(Utils.getFlag('C', options));
		setUseTournamentSelection(Utils.getFlag('O', options));
		
		super.setOptions(options);
	} // setOptions

	/**
	 * Gets the current settings of the search algorithm.
	 *
	 * @return an array of strings suitable for passing to setOptions
	 */
	public String[] getOptions() {
		String[] superOptions = super.getOptions();
		String[] options = new String[11 + superOptions.length];
		int current = 0;
		
		options[current++] = "-L";
		options[current++] = "" + getPopulationSize();

		options[current++] = "-A";
		options[current++] = "" + getDescendantPopulationSize();

		options[current++] = "-U";
		options[current++] = "" + getRuns();

		options[current++] = "-R";
		options[current++] = "" + getSeed();

		if (getUseMutation()) {
		  options[current++] = "-M";
		}
		if (getUseCrossOver()) {
		  options[current++] = "-C";
		}
		if (getUseTournamentSelection()) {
		  options[current++] = "-O";
		}

		// insert options from parent class
		for (int iOption = 0; iOption < superOptions.length; iOption++) {
			options[current++] = superOptions[iOption];
		}

		// Fill up rest with empty strings, not nulls!
		while (current < options.length) {
			options[current++] = "";
		}
		return options;
	} // getOptions

	/**
	 * @return whether cross-over is used
	 */
	public boolean getUseCrossOver() {
		return m_bUseCrossOver;
	}

	/**
	 * @return whether mutation is used
	 */
	public boolean getUseMutation() {
		return m_bUseMutation;
	}

	/**
	 * @return descendant population size
	 */
	public int getDescendantPopulationSize() {
		return m_nDescendantPopulationSize;
	}

	/**
	 * @return population size
	 */
	public int getPopulationSize() {
		return m_nPopulationSize;
	}

	/**
	 * @param bUseCrossOver: sets whether cross-over is used
	 */
	public void setUseCrossOver(boolean bUseCrossOver) {
		m_bUseCrossOver = bUseCrossOver;
	}

	/**
	 * @param bUseMutation: sets whether mutation is used
	 */
	public void setUseMutation(boolean bUseMutation) {
		m_bUseMutation = bUseMutation;
	}

	/**
	 * @return whether Tournament Selection (true) or Maximum Sub-Population (false) should be used
	 */
	public boolean getUseTournamentSelection() {
		return m_bUseTournamentSelection;
	}

	/**
	 * @param bUseTournamentSelection: sets whether Tournament Selection or Maximum Sub-Population should be used
	 */
	public void setUseTournamentSelection(boolean bUseTournamentSelection) {
		m_bUseTournamentSelection = bUseTournamentSelection;
	}

	/**
	 * @param iDescendantPopulationSize: sets descendant population size
	 */
	public void setDescendantPopulationSize(int iDescendantPopulationSize) {
		m_nDescendantPopulationSize = iDescendantPopulationSize;
	}

	/**
	 * @param iPopulationSize: sets population size
	 */
	public void setPopulationSize(int iPopulationSize) {
		m_nPopulationSize = iPopulationSize;
	}

	/**
	* @return random number seed
	*/
	public int getSeed() {
		return m_nSeed;
	} // getSeed

	/**
	 * Sets the random number seed
	 * @param nSeed The number of the seed to set
	 */
	public void setSeed(int nSeed) {
		m_nSeed = nSeed;
	} // setSeed

	/**
	 * This will return a string describing the classifier.
	 * @return The string.
	 */
	public String globalInfo() {
		return "This Bayes Network learning algorithm uses genetic search for finding a well scoring " +
		"Bayes network structure. Genetic search works by having a population of Bayes network structures " +
		"and allow them to mutate and apply cross over to get offspring. The best network structure " +
		"found during the process is returned.";
	} // globalInfo
	
	/**
	 * @return a string to describe the Runs option.
	 */
	public String runsTipText() {
	  return "Sets the number of generations of Bayes network structure populations.";
	} // runsTipText
	
	/**
	 * @return a string to describe the Seed option.
	 */
	public String seedTipText() {
	  return "Initialization value for random number generator." +
	  " Setting the seed allows replicability of experiments.";
	} // seedTipText

	/**
	 * @return a string to describe the Population Size option.
	 */
	public String populationSizeTipText() {
	  return "Sets the size of the population of network structures that is selected each generation.";
	} // populationSizeTipText

	/**
	 * @return a string to describe the Descendant Population Size option.
	 */
	public String descendantPopulationSizeTipText() {
	  return "Sets the size of the population of descendants that is created each generation.";
	} // descendantPopulationSizeTipText

	/**
	 * @return a string to describe the Use Mutation option.
	 */
	public String useMutationTipText() {
		return "Determines whether mutation is allowed. Mutation flips a bit in the bit " +
			"representation of the network structure. At least one of mutation or cross-over " +
			"should be used.";
	} // useMutationTipText

	/**
	 * @return a string to describe the Use Cross-Over option.
	 */
	public String useCrossOverTipText() {
		return "Determines whether cross-over is allowed. Cross over combined the bit " +
			"representations of network structure by taking a random first k bits of one" +
			"and adding the remainder of the other. At least one of mutation or cross-over " +
			"should be used.";
	} // useCrossOverTipText

	/**
	 * @return a string to describe the Use Tournament Selection option.
	 */
	public String useTournamentSelectionTipText() {
		return "Determines the method of selecting a population. When set to true, tournament " +
			"selection is used (pick two at random and the highest is allowed to continue). " +
			"When set to false, the top scoring network structures are selected.";
	} // useTournamentSelectionTipText
} // GeneticSearch