sparseinstance.java

一个数据挖掘软件ALPHAMINERR的整个过程的JAVA版源代码

      }
    } else {
      if (value != 0) {
	double[] tempValues = new double[m_AttValues.length + 1];
	int[] tempIndices = new int[m_Indices.length + 1];
	System.arraycopy(m_AttValues, 0, tempValues, 0, index + 1);
	System.arraycopy(m_Indices, 0, tempIndices, 0, index + 1);
	tempIndices[index + 1] = attIndex;
	tempValues[index + 1] = value;
	System.arraycopy(m_AttValues, index + 1, tempValues, index + 2, 
			 m_AttValues.length - index - 1);
	System.arraycopy(m_Indices, index + 1, tempIndices, index + 2, 
			 m_Indices.length - index - 1);
	m_AttValues = tempValues;
	m_Indices = tempIndices;
      }
    }
  }

  /**
   * Sets a specific value in the instance to the given value 
   * (internal floating-point format). Performs a deep copy
   * of the vector of attribute values before the value is set.
   *
   * @param indexOfIndex the index of the attribute's index 
   * @param value the new attribute value (If the corresponding
   * attribute is nominal (or a string) then this is the new value's
   * index as a double).  
   */
  public void setValueSparse(int indexOfIndex, double value) {

    if (value != 0) {
      double[] tempValues = new double[m_AttValues.length];
      System.arraycopy(m_AttValues, 0, tempValues, 0, m_AttValues.length);
      m_AttValues = tempValues;
      m_AttValues[indexOfIndex] = value;
    } else {
      double[] tempValues = new double[m_AttValues.length - 1];
      int[] tempIndices = new int[m_Indices.length - 1];
      System.arraycopy(m_AttValues, 0, tempValues, 0, indexOfIndex);
      System.arraycopy(m_Indices, 0, tempIndices, 0, indexOfIndex);
      System.arraycopy(m_AttValues, indexOfIndex + 1, tempValues, indexOfIndex, 
		       m_AttValues.length - indexOfIndex - 1);
      System.arraycopy(m_Indices, indexOfIndex + 1, tempIndices, indexOfIndex, 
		       m_Indices.length - indexOfIndex - 1);
      m_AttValues = tempValues;
      m_Indices = tempIndices;
    }
  }

  /**
   * Returns the values of each attribute as an array of doubles.
   *
   * @return an array containing all the instance attribute values
   */
  public double[] toDoubleArray() {

    double[] newValues = new double[m_NumAttributes];
    for (int i = 0; i < m_AttValues.length; i++) {
      newValues[m_Indices[i]] = m_AttValues[i];
    }
    return newValues;
  }

  /**
   * Returns the description of one instance in sparse format. 
   * If the instance doesn't have access to a dataset, it returns the 
   * internal floating-point values. Quotes string values that contain 
   * whitespace characters.
   *
   * @return the instance's description as a string
   */
  public String toString() {

    StringBuffer text = new StringBuffer();
    
    text.append('{');
    for (int i = 0; i < m_Indices.length; i++) {
      if (i > 0) text.append(",");
      if (isMissingValue(m_AttValues[i])) {
	text.append(m_Indices[i] + " ?");
      } else {
	if (m_Dataset == null) {
	  text.append(m_Indices[i] + " " + 
		      Utils.doubleToString(m_AttValues[i],6));
	} else {
	  if (m_Dataset.attribute(m_Indices[i]).isNominal() || 
	      m_Dataset.attribute(m_Indices[i]).isString()) {
	    try {
	      text.append(m_Indices[i] + " " +
			  Utils.quote(m_Dataset.attribute(m_Indices[i]).
				      value((int)valueSparse(i))));
	    } catch (Exception e) {
              e.printStackTrace();
              System.err.println(new Instances(m_Dataset, 0));
              System.err.println("Att:" + m_Indices[i] + " Val:" + valueSparse(i));
	      throw new Error("This should never happen!");
	    }
	  } else {
	    text.append(m_Indices[i] + " " +
			Utils.doubleToString(m_AttValues[i],6));
	  }
	}
      }
    }
    text.append('}');
    return text.toString();
  }

  /**
   * Returns an instance's attribute value in internal format.
   *
   * @param attIndex the attribute's index
   * @return the specified value as a double (If the corresponding
   * attribute is nominal (or a string) then it returns the value's index as a 
   * double).
   */
  public double value(int attIndex) {

    int index = locateIndex(attIndex);
    if ((index >= 0) && (m_Indices[index] == attIndex)) {
      return m_AttValues[index];
    } else {
      return 0.0;
    }
  }  

  /**
   * Deletes an attribute at the given position (0 to 
   * numAttributes() - 1).
   *
   * @param pos the attribute's position
   */
  void forceDeleteAttributeAt(int position) {

    int index = locateIndex(position);

    m_NumAttributes--;
    if ((index >= 0) && (m_Indices[index] == position)) {
      int[] tempIndices = new int[m_Indices.length - 1];
      double[] tempValues = new double[m_AttValues.length - 1];
      System.arraycopy(m_Indices, 0, tempIndices, 0, index);
      System.arraycopy(m_AttValues, 0, tempValues, 0, index);
      for (int i = index; i < m_Indices.length - 1; i++) {
	tempIndices[i] = m_Indices[i + 1] - 1;
	tempValues[i] = m_AttValues[i + 1];
      }
      m_Indices = tempIndices;
      m_AttValues = tempValues;
    } else {
      int[] tempIndices = new int[m_Indices.length];
      double[] tempValues = new double[m_AttValues.length];
      System.arraycopy(m_Indices, 0, tempIndices, 0, index + 1);
      System.arraycopy(m_AttValues, 0, tempValues, 0, index + 1);
      for (int i = index + 1; i < m_Indices.length; i++) {
	tempIndices[i] = m_Indices[i] - 1;
	tempValues[i] = m_AttValues[i];
      }
      m_Indices = tempIndices;
      m_AttValues = tempValues;
    }
  }

  /**
   * Inserts an attribute at the given position
   * (0 to numAttributes()) and sets its value to be missing. 
   *
   * @param pos the attribute's position
   */
  void forceInsertAttributeAt(int position)  {

    int index = locateIndex(position);

    m_NumAttributes++;
    if ((index >= 0) && (m_Indices[index] == position)) {
      int[] tempIndices = new int[m_Indices.length + 1];
      double[] tempValues = new double[m_AttValues.length + 1];
      System.arraycopy(m_Indices, 0, tempIndices, 0, index);
      System.arraycopy(m_AttValues, 0, tempValues, 0, index);
      tempIndices[index] = position;
      tempValues[index] = MISSING_VALUE;
      for (int i = index; i < m_Indices.length; i++) {
	tempIndices[i + 1] = m_Indices[i] + 1;
	tempValues[i + 1] = m_AttValues[i];
      }
      m_Indices = tempIndices;
      m_AttValues = tempValues;
    } else {
      int[] tempIndices = new int[m_Indices.length + 1];
      double[] tempValues = new double[m_AttValues.length + 1];
      System.arraycopy(m_Indices, 0, tempIndices, 0, index + 1);
      System.arraycopy(m_AttValues, 0, tempValues, 0, index + 1);
      tempIndices[index + 1] = position;
      tempValues[index + 1] = MISSING_VALUE;
      for (int i = index + 1; i < m_Indices.length; i++) {
	tempIndices[i + 1] = m_Indices[i] + 1;
	tempValues[i + 1] = m_AttValues[i];
      }
      m_Indices = tempIndices;
      m_AttValues = tempValues;
    }
  }

  /**
   * Main method for testing this class.
   */
  public static void main(String[] options) {

    try {

      // Create numeric attributes "length" and "weight"
      Attribute length = new Attribute("length");
      Attribute weight = new Attribute("weight");
      
      // Create vector to hold nominal values "first", "second", "third" 
      FastVector my_nominal_values = new FastVector(3); 
      my_nominal_values.addElement("first"); 
      my_nominal_values.addElement("second"); 
      my_nominal_values.addElement("third"); 
      
      // Create nominal attribute "position" 
      Attribute position = new Attribute("position", my_nominal_values);
      
      // Create vector of the above attributes 
      FastVector attributes = new FastVector(3);
      attributes.addElement(length);
      attributes.addElement(weight);
      attributes.addElement(position);
      
      // Create the empty dataset "race" with above attributes
      Instances race = new Instances("race", attributes, 0);
      
      // Make position the class attribute
      race.setClassIndex(position.index());
      
      // Create empty instance with three attribute values
      SparseInstance inst = new SparseInstance(3);
      
      // Set instance's values for the attributes "length", "weight", and "position"
      inst.setValue(length, 5.3);
      inst.setValue(weight, 300);
      inst.setValue(position, "first");
      
      // Set instance's dataset to be the dataset "race"
      inst.setDataset(race);
      
      // Print the instance
      System.out.println("The instance: " + inst);
      
      // Print the first attribute
      System.out.println("First attribute: " + inst.attribute(0));
      
      // Print the class attribute
      System.out.println("Class attribute: " + inst.classAttribute());
      
      // Print the class index
      System.out.println("Class index: " + inst.classIndex());
      
      // Say if class is missing
      System.out.println("Class is missing: " + inst.classIsMissing());
      
      // Print the instance's class value in internal format
      System.out.println("Class value (internal format): " + inst.classValue());
      
      // Print a shallow copy of this instance
      SparseInstance copy = (SparseInstance) inst.copy();
      System.out.println("Shallow copy: " + copy);
      
      // Set dataset for shallow copy
      copy.setDataset(inst.dataset());
      System.out.println("Shallow copy with dataset set: " + copy);

      // Print out all values in internal format
      System.out.print("All stored values in internal format: ");
      for (int i = 0; i < inst.numValues(); i++) {
	if (i > 0) {
	  System.out.print(",");
	}
	System.out.print(inst.valueSparse(i));
      }
      System.out.println();

      // Set all values to zero
      System.out.print("All values set to zero: ");
      while (inst.numValues() > 0) {
	inst.setValueSparse(0, 0);
      }
      for (int i = 0; i < inst.numValues(); i++) {
	if (i > 0) {
	  System.out.print(",");
	}
	System.out.print(inst.valueSparse(i));
      }
      System.out.println();

      // Set all values to one
      System.out.print("All values set to one: ");
      for (int i = 0; i < inst.numAttributes(); i++) {
	inst.setValue(i, 1);
      }
      for (int i = 0; i < inst.numValues(); i++) {
	if (i > 0) {
	  System.out.print(",");
	}
	System.out.print(inst.valueSparse(i));
      }
      System.out.println();

      // Unset dataset for copy, delete first attribute, and insert it again
      copy.setDataset(null);
      copy.deleteAttributeAt(0);
      copy.insertAttributeAt(0);
      copy.setDataset(inst.dataset());
      System.out.println("Copy with first attribute deleted and inserted: " + copy); 

      // Same for second attribute
      copy.setDataset(null);
      copy.deleteAttributeAt(1);
      copy.insertAttributeAt(1);
      copy.setDataset(inst.dataset());
      System.out.println("Copy with second attribute deleted and inserted: " + copy); 

      // Same for last attribute
      copy.setDataset(null);
      copy.deleteAttributeAt(2);
      copy.insertAttributeAt(2);
      copy.setDataset(inst.dataset());
      System.out.println("Copy with third attribute deleted and inserted: " + copy); 
      
      // Enumerate attributes (leaving out the class attribute)
      System.out.println("Enumerating attributes (leaving out class):");
      Enumeration em = inst.emerateAttributes();
      while (em.hasMoreElements()) {
	Attribute att = (Attribute) em.nextElement();
	System.out.println(att);
      }
      
      // Headers are equivalent?
      System.out.println("Header of original and copy equivalent: " +
			 inst.equalHeaders(copy));

      // Test for missing values
      System.out.println("Length of copy missing: " + copy.isMissing(length));
      System.out.println("Weight of copy missing: " + copy.isMissing(weight.index()));
      System.out.println("Length of copy missing: " + 
			 Instance.isMissingValue(copy.value(length)));
      System.out.println("Missing value coded as: " + Instance.missingValue());

      // Prints number of attributes and classes
      System.out.println("Number of attributes: " + copy.numAttributes());
      System.out.println("Number of classes: " + copy.numClasses());

      // Replace missing values
      double[] meansAndModes = {2, 3, 0};
      copy.replaceMissingValues(meansAndModes);
      System.out.println("Copy with missing value replaced: " + copy);

      // Setting and getting values and weights
      copy.setClassMissing();
      System.out.println("Copy with missing class: " + copy);
      copy.setClassValue(0);
      System.out.println("Copy with class value set to first value: " + copy);
      copy.setClassValue("third");
      System.out.println("Copy with class value set to \"third\": " + copy);
      copy.setMissing(1);
      System.out.println("Copy with second attribute set to be missing: " + copy);
      copy.setMissing(length);
      System.out.println("Copy with length set to be missing: " + copy);
      copy.setValue(0, 0);
      System.out.println("Copy with first attribute set to 0: " + copy);
      copy.setValue(weight, 1);
      System.out.println("Copy with weight attribute set to 1: " + copy);
      copy.setValue(position, "second");
      System.out.println("Copy with position set to \"second\": " + copy);
      copy.setValue(2, "first");
      System.out.println("Copy with last attribute set to \"first\": " + copy);
      System.out.println("Current weight of instance copy: " + copy.weight());
      copy.setWeight(2);
      System.out.println("Current weight of instance copy (set to 2): " + copy.weight());
      System.out.println("Last value of copy: " + copy.toString(2));
      System.out.println("Value of position for copy: " + copy.toString(position));
      System.out.println("Last value of copy (internal format): " + copy.value(2));
      System.out.println("Value of position for copy (internal format): " + 
			 copy.value(position));
    } catch (Exception e) {
      e.printStackTrace();
    }
  }
}