matlabica.java

wekaUT是 university texas austin 开发的基于weka的半指导学习(semi supervised learning)的分类器

   */
  private void dumpInstances(String tempFile) {
    try { 
      PrintWriter writer = new PrintWriter(new BufferedOutputStream(new FileOutputStream(tempFile)));
      for (int k = 0; k < m_numInstances; k++) {
	Instance instance = m_trainInstances.instance(k);
	for (int j = 0; j < m_numAttribs; j++) {
	  writer.print(instance.value(j) + " ");
	}
	writer.println();
      }
      writer.close();
    } catch (Exception e) {
      System.err.println("Could not create a temporary file for dumping the data matrix: " + e);
    }
  }

  /** Create matlab m-file for ICA
   * @param filename file where matlab script is created
   */
  public void prepareMatlab(String filename) {
    try{
      PrintWriter writer = new PrintWriter(new BufferedOutputStream(new FileOutputStream(filename)));
      writer.println("addpath /var/local;");
      writer.println("DATA = load('/var/local/ICAdataMatrix.txt');");
      if (m_loadEigenValuesFromFile == true) {
	writer.println("E = load('/var/local/PCAeigenValues.txt');");
      }
      if (m_loadEigenVectorsFromFile == true) {
	writer.println("V = load('/var/local/PCAeigenVectors.txt');");
      }
      writer.print("[IC, A, invA] = fastica(DATA'"); // take transpose of data matrix, to make each instance a column
      if (m_loadEigenValuesFromFile == true) {
	writer.print(",'pcaD',E");
      }
      if (m_loadEigenVectorsFromFile == true) {
	writer.print(",'pcaE',V");
      }
      writer.print(",'approach','" + TAGS_APPROACH[m_ICAapproach].getReadable() + "'");
      writer.print(",'g','" + TAGS_FUNCTION[m_ICAfunction].getReadable() + "'");
      writer.print(",'numOfIC'," + m_NumIndependentComponents);
      writer.println(",'displayMode','off','stabilization','on');");
      writer.println("[ICnumRows, ICnumCols] = size(IC);");
      writer.println("[AnumRows, AnumCols] = size(A);");
      writer.println("[invAnumRows, invAnumCols] = size(invA);\n");

      writer.println("save " + m_mixingMatrixFilename + " AnumRows AnumCols A -ASCII -DOUBLE");
      writer.println("save " + m_inverseMixingMatrixFilename + " invAnumRows invAnumCols invA -ASCII -DOUBLE");
      writer.println("save " + m_independentComponentsFilename + " ICnumRows ICnumCols IC -ASCII -DOUBLE");
      writer.close();
    } 
    catch (Exception e) {
      System.err.println("Could not create matlab file: " + e);
    }
  }


  /** Run matlab in command line with a given argument
   * @param inFile file to be input to Matlab
   * @param outFile file where results are stored
   */
  public static void runMatlab(String inFile, String outFile) {
    // call matlab to do the dirty work
    try {
      int exitValue;
      do {
	System.out.println("Starting to run matlab");
	Process proc = Runtime.getRuntime().exec("matlab -tty < " + inFile + " > " + outFile);
	exitValue = proc.waitFor();
	if (exitValue != 0) {
	  System.err.println("WARNING!!!!!  Matlab returned exit value 1, trying again in 5 mins!");
	  Thread.sleep(300000);
	}
      } while (exitValue != 0);
      System.out.println("End of running matlab, exitValue = " + exitValue);
    } 
    catch (Exception e) {
      System.err.println("Problems running matlab: " + e);
    }
  } 
      

  /**
   * Return a summary of the analysis
   * @return a summary of the analysis.
   */
  private String independentComponentsSummary() {
    StringBuffer result = new StringBuffer();
    double cumulative = 0.0;
    Instances output = null;
    int numVectors=0;

    try {
      output = setOutputFormat();
      numVectors = (output.classIndex() < 0) 
	? output.numAttributes()
	: output.numAttributes()-1;
    } catch (Exception ex) {
    }
    // Todo: Add IC summary to result string
    result.append("\nAttribute ranking filter:\n");
    result.append(m_eval.toString());
    return result.toString();
  }

  /**
   * Returns a description of this attribute transformer
   * @return a String describing this attribute transformer
   */
  public String toString() {
    if (m_independentComponents == null) {
      return "Independent components hasn't been built yet!";
    } else {
      return "\tIndependent Components Attribute Transformer\n\n"
	+independentComponentsSummary();
    }
  }

  /**
   * Return a matrix as a String
   * @param matrix that is decribed as a string
   * @return a String describing a matrix
   */
  private String matrixToString(double [][] matrix) {
    StringBuffer result = new StringBuffer();
    int last = matrix.length - 1;

    for (int i = 0; i <= last; i++) {
      for (int j = 0; j <= last; j++) {
	result.append(Utils.doubleToString(matrix[i][j],6,2)+" ");
	if (j == last) {
	  result.append('\n');
	}
      }
    }
    return result.toString();
  }

  /**
   * Convert a ic transformed instance back to the original space
   */
  private Instance convertInstanceToOriginal(Instance inst)
    throws Exception {
    double[] newVals = null;

    if (m_hasClass) {
      newVals = new double[m_numAttribs+1];
    } else {
      newVals = new double[m_numAttribs];
    }

    if (m_hasClass) {
      // class is always appended as the last attribute
      newVals[m_numAttribs] = inst.value(inst.numAttributes() - 1);
    }

    for (int i = 0; i < m_numAttribs; i++) {
      for (int j = 0; j < m_outputNumAtts - 1; j++) {
	newVals[i] += m_mixingMatrix[i][j] * inst.value(j);
      }
    }
    
    if (inst instanceof SparseInstance) {
      return new SparseInstance(inst.weight(), newVals);
    } else {
      return new Instance(inst.weight(), newVals);
    }      
  }

  /**
   * Transform an instance in original (unormalized) format. Convert back
   * to the original space if requested.
   * @param instance an instance in the original (unormalized) format
   * @return a transformed instance
   * @exception Exception if instance cant be transformed
   */
  public Instance convertInstance(Instance instance) throws Exception {

    if (m_independentComponents == null) {
      //      throw new Exception("convertInstance: Independent components not " +"built yet");
      System.out.println("WARNING!! Independent components could not be built, returning original data");
    }

    double[] newVals = new double[m_outputNumAtts];
    Instance tempInst = (Instance)instance.copy();
    if (!instance.equalHeaders(m_trainCopy.instance(0))) {
      throw new Exception("Can't convert instance: header's don't match: MatlabICA");
    }

    m_replaceMissingFilter.input(tempInst);
    m_replaceMissingFilter.batchFinished();
    tempInst = m_replaceMissingFilter.output();

    if (m_normalize) {
      m_normalizeFilter.input(tempInst);
      m_normalizeFilter.batchFinished();
      tempInst = m_normalizeFilter.output();
    }

    if (m_attributeFilter != null) {
      m_attributeFilter.input(tempInst);
      m_attributeFilter.batchFinished();
      tempInst = m_attributeFilter.output();
    }

    if (m_hasClass) {
       newVals[m_outputNumAtts - 1] = instance.value(instance.classIndex());
    }

    for (int i=0; i<m_outputNumAtts-1; i++) {
      for (int j = 0; j < m_numAttribs; j++) {
	newVals[i] += (m_inverseMixingMatrix[i][j] *  tempInst.value(j));
      }
    }
    
    if (!m_transBackToOriginal) {
      if (instance instanceof SparseInstance) {
      return new SparseInstance(instance.weight(), newVals);
      } else {
	return new Instance(instance.weight(), newVals);
      }      
    } else {
      if (instance instanceof SparseInstance) {
	return convertInstanceToOriginal(new SparseInstance(instance.weight(), 
							    newVals));
      } else {
	return convertInstanceToOriginal(new Instance(instance.weight(),
						      newVals));
      }
    }
  }

  /**
   * Set up the header for the IC->original space dataset
   */
  private Instances setOutputFormatOriginal() throws Exception {
    FastVector attributes = new FastVector();
    
    for (int i = 0; i < m_numAttribs; i++) {
      String att = m_trainInstances.attribute(i).name();
      attributes.addElement(new Attribute(att));
    }
    
    if (m_hasClass) {
      attributes.addElement(m_trainCopy.classAttribute().copy());
    }

    Instances outputFormat = 
      new Instances(m_trainCopy.relationName()+"->IC->original space",
		    attributes, 0);
    
    // set the class to be the last attribute if necessary
    if (m_hasClass) {
      outputFormat.setClassIndex(outputFormat.numAttributes()-1);
    }

    return outputFormat;
  }

  /**
   * Set the format for the transformed data
   * @return a set of empty Instances (header only) in the new format
   * @exception Exception if the output format can't be set
   */
  private Instances setOutputFormat() throws Exception {
    if (m_independentComponents == null) {
      return null;
    }

    double cumulative = 0.0;
    FastVector attributes = new FastVector();
     for (int i=0; i<m_inverseMixingMatrix.length; i++) {
       StringBuffer attName = new StringBuffer("ICAattribute" + i);
       attributes.addElement(new Attribute(attName.toString()));
     }
        
     if (m_hasClass) {
       attributes.addElement(m_trainCopy.classAttribute().copy());
     }

     Instances outputFormat = 
       new Instances(m_trainInstances.relationName()+"_independent components",
		     attributes, 0);

     // set the class to be the last attribute if necessary
     if (m_hasClass) {
       outputFormat.setClassIndex(outputFormat.numAttributes()-1);
     }
     
     m_outputNumAtts = outputFormat.numAttributes();
     System.out.println("m_outputNumAtts: " + m_outputNumAtts);
     return outputFormat;
  }


  /** Get a timestamp string as a weak uniqueid
   * @returns a timestamp string in the form "mmddhhmmssS"
   */     
  public static String getLogTimestamp() {
    Calendar cal = Calendar.getInstance(TimeZone.getDefault());
    String DATE_FORMAT = "MMddHHmmssS";
    java.text.SimpleDateFormat sdf = new java.text.SimpleDateFormat(DATE_FORMAT);
    
    sdf.setTimeZone(TimeZone.getDefault());          
    return (sdf.format(cal.getTime()));
  }
  
  /**
   * Main method for testing this class
   * @param argv should contain the command line arguments to the
   * evaluator/transformer (see AttributeSelection)
   */
  public static void main(String [] argv) {
    try {
      //      String name = "../../data/20newsgroups/different-100_fromCCS.arff";
      String name = "/u/ml/software/weka-latest/data/iris.arff";
      if (argv.length == 1) {
	name = argv[0];
      }
      else {
	System.err.println("No data filename given as argument, running on default file " + name);
      }
      
      Reader r = new BufferedReader(new FileReader(name));
      Instances data = new Instances(r);
      data.setClassIndex(data.numAttributes() - 1);

      MatlabICA mica = new MatlabICA();
      mica.setNumIndependentComponents(2);
      mica.buildEvaluator(data);
      mica.transformedData();
    }
    catch (Exception e) {
      e.printStackTrace();
      System.out.println(e.getMessage());
    }
  }  
}