weka.classifiers.trees.m5.RuleNode Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of weka-dev Show documentation
Show all versions of weka-dev Show documentation
The Waikato Environment for Knowledge Analysis (WEKA), a machine
learning workbench. This version represents the developer version, the
"bleeding edge" of development, you could say. New functionality gets added
to this version.
/*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see RuleNode instance.
*
* @param globalDev the global standard deviation of the class
* @param globalAbsDev the global absolute deviation of the class
* @param parent the parent of this node
*/
public RuleNode(double globalDev, double globalAbsDev, RuleNode parent) {
m_nodeModel = null;
m_right = null;
m_left = null;
m_parent = parent;
m_globalDeviation = globalDev;
m_globalAbsDeviation = globalAbsDev;
}
/**
* Build this node (find an attribute and split point)
*
* @param data the instances on which to build this node
* @throws Exception if an error occurs
*/
@Override
public void buildClassifier(Instances data) throws Exception {
m_rootMeanSquaredError = Double.MAX_VALUE;
// m_instances = new Instances(data);
m_instances = data;
m_classIndex = m_instances.classIndex();
m_numInstances = m_instances.numInstances();
m_numAttributes = m_instances.numAttributes();
m_nodeModel = null;
m_right = null;
m_left = null;
if ((m_numInstances < m_splitNum)
|| (Rule.stdDev(m_classIndex, m_instances) < (m_globalDeviation * m_devFraction))) {
m_isLeaf = true;
} else {
m_isLeaf = false;
}
split();
}
/**
* Classify an instance using this node. Recursively calls classifyInstance on
* child nodes.
*
* @param inst the instance to classify
* @return the prediction for this instance
* @throws Exception if an error occurs
*/
@Override
public double classifyInstance(Instance inst) throws Exception {
if (m_isLeaf) {
if (m_nodeModel == null) {
throw new Exception("Classifier has not been built correctly.");
}
return m_nodeModel.classifyInstance(inst);
}
if (inst.value(m_splitAtt) <= m_splitValue) {
return m_left.classifyInstance(inst);
} else {
return m_right.classifyInstance(inst);
}
}
/**
* Applies the m5 smoothing procedure to a prediction
*
* @param n number of instances in selected child of this node
* @param pred the prediction so far
* @param supportPred the prediction of the linear model at this node
* @return the current prediction smoothed with the prediction of the linear
* model at this node
* @throws Exception if an error occurs
*/
protected static double smoothingOriginal(double n, double pred,
double supportPred) throws Exception {
double smoothed;
smoothed = ((n * pred) + (SMOOTHING_CONSTANT * supportPred))
/ (n + SMOOTHING_CONSTANT);
return smoothed;
}
/**
* Finds an attribute and split point for this node
*
* @throws Exception if an error occurs
*/
public void split() throws Exception {
int i;
Instances leftSubset, rightSubset;
SplitEvaluate bestSplit, currentSplit;
boolean[] attsBelow;
if (!m_isLeaf) {
bestSplit = new YongSplitInfo(0, m_numInstances - 1, -1);
currentSplit = new YongSplitInfo(0, m_numInstances - 1, -1);
// find the best attribute to split on
for (i = 0; i < m_numAttributes; i++) {
if (i != m_classIndex) {
// sort the instances by this attribute
m_instances.sort(i);
currentSplit.attrSplit(i, m_instances);
if ((Math.abs(currentSplit.maxImpurity() - bestSplit.maxImpurity()) > 1.e-6)
&& (currentSplit.maxImpurity() > bestSplit.maxImpurity() + 1.e-6)) {
bestSplit = currentSplit.copy();
}
}
}
// cant find a good split or split point?
if (bestSplit.splitAttr() < 0 || bestSplit.position() < 1
|| bestSplit.position() > m_numInstances - 1) {
m_isLeaf = true;
} else {
m_splitAtt = bestSplit.splitAttr();
m_splitValue = bestSplit.splitValue();
leftSubset = new Instances(m_instances, m_numInstances);
rightSubset = new Instances(m_instances, m_numInstances);
for (i = 0; i < m_numInstances; i++) {
if (m_instances.instance(i).value(m_splitAtt) <= m_splitValue) {
leftSubset.add(m_instances.instance(i));
} else {
rightSubset.add(m_instances.instance(i));
}
}
leftSubset.compactify();
rightSubset.compactify();
// build left and right nodes
m_left = new RuleNode(m_globalDeviation, m_globalAbsDeviation, this);
m_left.setMinNumInstances(m_splitNum);
m_left.setRegressionTree(m_regressionTree);
m_left.setSaveInstances(m_saveInstances);
m_left.buildClassifier(leftSubset);
m_right = new RuleNode(m_globalDeviation, m_globalAbsDeviation, this);
m_right.setMinNumInstances(m_splitNum);
m_right.setRegressionTree(m_regressionTree);
m_right.setSaveInstances(m_saveInstances);
m_right.buildClassifier(rightSubset);
// now find out what attributes are tested in the left and right
// subtrees and use them to learn a linear model for this node
if (!m_regressionTree) {
attsBelow = attsTestedBelow();
attsBelow[m_classIndex] = true;
int count = 0, j;
for (j = 0; j < m_numAttributes; j++) {
if (attsBelow[j]) {
count++;
}
}
int[] indices = new int[count];
count = 0;
for (j = 0; j < m_numAttributes; j++) {
if (attsBelow[j] && (j != m_classIndex)) {
indices[count++] = j;
}
}
indices[count] = m_classIndex;
m_indices = indices;
} else {
m_indices = new int[1];
m_indices[0] = m_classIndex;
m_numParameters = 1;
}
}
}
if (m_isLeaf) {
int[] indices = new int[1];
indices[0] = m_classIndex;
m_indices = indices;
m_numParameters = 1;
// need to evaluate the model here if want correct stats for unpruned
// tree
}
}
/**
* Build a linear model for this node using those attributes specified in
* indices.
*
* @param indices an array of attribute indices to include in the linear model
* @throws Exception if something goes wrong
*/
private void buildLinearModel(int[] indices) throws Exception {
// copy the training instances and remove all but the tested
// attributes
Instances reducedInst = new Instances(m_instances);
Remove attributeFilter = new Remove();
attributeFilter.setInvertSelection(true);
attributeFilter.setAttributeIndicesArray(indices);
attributeFilter.setInputFormat(reducedInst);
reducedInst = Filter.useFilter(reducedInst, attributeFilter);
// build a linear regression for the training data using the
// tested attributes
LinearRegression temp = new LinearRegression();
temp.buildClassifier(reducedInst);
double[] lmCoeffs = temp.coefficients();
double[] coeffs = new double[m_instances.numAttributes()];
for (int i = 0; i < lmCoeffs.length - 1; i++) {
if (indices[i] != m_classIndex) {
coeffs[indices[i]] = lmCoeffs[i];
}
}
m_nodeModel = new PreConstructedLinearModel(coeffs,
lmCoeffs[lmCoeffs.length - 1]);
m_nodeModel.buildClassifier(m_instances);
}
/**
* Returns an array containing the indexes of attributes used in tests below
* this node
*
* @return an array of attribute indexes
*/
private boolean[] attsTestedBelow() {
boolean[] attsBelow = new boolean[m_numAttributes];
boolean[] attsBelowLeft = null;
boolean[] attsBelowRight = null;
if (m_right != null) {
attsBelowRight = m_right.attsTestedBelow();
}
if (m_left != null) {
attsBelowLeft = m_left.attsTestedBelow();
}
for (int i = 0; i < m_numAttributes; i++) {
if (attsBelowLeft != null) {
attsBelow[i] = (attsBelow[i] || attsBelowLeft[i]);
}
if (attsBelowRight != null) {
attsBelow[i] = (attsBelow[i] || attsBelowRight[i]);
}
}
if (!m_isLeaf) {
attsBelow[m_splitAtt] = true;
}
return attsBelow;
}
/**
* Sets the leaves' numbers
*
* @param leafCounter the number of leaves counted
* @return the number of the total leaves under the node
*/
public int numLeaves(int leafCounter) {
if (!m_isLeaf) {
// node
m_leafModelNum = 0;
if (m_left != null) {
leafCounter = m_left.numLeaves(leafCounter);
}
if (m_right != null) {
leafCounter = m_right.numLeaves(leafCounter);
}
} else {
// leaf
leafCounter++;
m_leafModelNum = leafCounter;
}
return leafCounter;
}
/**
* print the linear model at this node
*
* @return the linear model
*/
@Override
public String toString() {
return printNodeLinearModel();
}
/**
* print the linear model at this node
*
* @return the linear model at this node
*/
public String printNodeLinearModel() {
return m_nodeModel.toString();
}
/**
* print all leaf models
*
* @return the leaf models
*/
public String printLeafModels() {
StringBuffer text = new StringBuffer();
if (m_isLeaf) {
text.append("\nLM num: " + m_leafModelNum);
text.append(m_nodeModel.toString());
text.append("\n");
} else {
text.append(m_left.printLeafModels());
text.append(m_right.printLeafModels());
}
return text.toString();
}
/**
* Returns a description of this node (debugging purposes)
*
* @return a string describing this node
*/
public String nodeToString() {
StringBuffer text = new StringBuffer();
System.out.println("In to string");
text.append("Node:\n\tnum inst: " + m_numInstances);
if (m_isLeaf) {
text.append("\n\tleaf");
} else {
text.append("\tnode");
}
text.append("\n\tSplit att: " + m_instances.attribute(m_splitAtt).name());
text.append("\n\tSplit val: " + Utils.doubleToString(m_splitValue, 1, 3));
text.append("\n\tLM num: " + m_leafModelNum);
text.append("\n\tLinear model\n" + m_nodeModel.toString());
text.append("\n\n");
if (m_left != null) {
text.append(m_left.nodeToString());
}
if (m_right != null) {
text.append(m_right.nodeToString());
}
return text.toString();
}
/**
* Recursively builds a textual description of the tree
*
* @param level the level of this node
* @return string describing the tree
*/
public String treeToString(int level) {
int i;
StringBuffer text = new StringBuffer();
if (!m_isLeaf) {
text.append("\n");
for (i = 1; i <= level; i++) {
text.append("| ");
}
if (m_instances.attribute(m_splitAtt).name().charAt(0) != '[') {
text.append(m_instances.attribute(m_splitAtt).name() + " <= "
+ Utils.doubleToString(m_splitValue, 1, 3) + " : ");
} else {
text.append(m_instances.attribute(m_splitAtt).name() + " false : ");
}
if (m_left != null) {
text.append(m_left.treeToString(level + 1));
} else {
text.append("NULL\n");
}
for (i = 1; i <= level; i++) {
text.append("| ");
}
if (m_instances.attribute(m_splitAtt).name().charAt(0) != '[') {
text.append(m_instances.attribute(m_splitAtt).name() + " > "
+ Utils.doubleToString(m_splitValue, 1, 3) + " : ");
} else {
text.append(m_instances.attribute(m_splitAtt).name() + " true : ");
}
if (m_right != null) {
text.append(m_right.treeToString(level + 1));
} else {
text.append("NULL\n");
}
} else {
text.append("LM" + m_leafModelNum);
if (m_globalDeviation > 0.0) {
text.append(" ("
+ m_numInstances
+ "/"
+ Utils.doubleToString(
(100.0 * m_rootMeanSquaredError / m_globalDeviation), 1, 3)
+ "%)\n");
} else {
text.append(" (" + m_numInstances + ")\n");
}
}
return text.toString();
}
/**
* Traverses the tree and installs linear models at each node. This method
* must be called if pruning is not to be performed.
*
* @throws Exception if an error occurs
*/
public void installLinearModels() throws Exception {
Evaluation nodeModelEval;
if (m_isLeaf) {
buildLinearModel(m_indices);
} else {
if (m_left != null) {
m_left.installLinearModels();
}
if (m_right != null) {
m_right.installLinearModels();
}
buildLinearModel(m_indices);
}
nodeModelEval = new Evaluation(m_instances);
nodeModelEval.evaluateModel(m_nodeModel, m_instances);
m_rootMeanSquaredError = nodeModelEval.rootMeanSquaredError();
// save space
if (!m_saveInstances) {
m_instances = new Instances(m_instances, 0);
}
}
/**
*
* @throws Exception
*/
public void installSmoothedModels() throws Exception {
if (m_isLeaf) {
double[] coefficients = new double[m_numAttributes];
double intercept;
double[] coeffsUsedByLinearModel = m_nodeModel.coefficients();
RuleNode current = this;
// prime array with leaf node coefficients
for (int i = 0; i < coeffsUsedByLinearModel.length; i++) {
if (i != m_classIndex) {
coefficients[i] = coeffsUsedByLinearModel[i];
}
}
// intercept
intercept = m_nodeModel.intercept();
do {
if (current.m_parent != null) {
double n = current.m_numInstances;
// contribution of the model below
for (int i = 0; i < coefficients.length; i++) {
coefficients[i] = ((coefficients[i] * n) / (n + SMOOTHING_CONSTANT));
}
intercept = ((intercept * n) / (n + SMOOTHING_CONSTANT));
// contribution of this model
coeffsUsedByLinearModel = current.m_parent.getModel().coefficients();
for (int i = 0; i < coeffsUsedByLinearModel.length; i++) {
if (i != m_classIndex) {
// smooth in these coefficients (at this node)
coefficients[i] += ((SMOOTHING_CONSTANT * coeffsUsedByLinearModel[i]) / (n + SMOOTHING_CONSTANT));
}
}
// smooth in the intercept
intercept += ((SMOOTHING_CONSTANT * current.m_parent.getModel()
.intercept()) / (n + SMOOTHING_CONSTANT));
current = current.m_parent;
}
} while (current.m_parent != null);
m_nodeModel = new PreConstructedLinearModel(coefficients, intercept);
m_nodeModel.buildClassifier(m_instances);
}
if (m_left != null) {
m_left.installSmoothedModels();
}
if (m_right != null) {
m_right.installSmoothedModels();
}
}
/**
* Recursively prune the tree
*
* @throws Exception if an error occurs
*/
public void prune() throws Exception {
Evaluation nodeModelEval = null;
if (m_isLeaf) {
buildLinearModel(m_indices);
nodeModelEval = new Evaluation(m_instances);
// count the constant term as a paramter for a leaf
// Evaluate the model
nodeModelEval.evaluateModel(m_nodeModel, m_instances);
m_rootMeanSquaredError = nodeModelEval.rootMeanSquaredError();
} else {
// Prune the left and right subtrees
if (m_left != null) {
m_left.prune();
}
if (m_right != null) {
m_right.prune();
}
buildLinearModel(m_indices);
nodeModelEval = new Evaluation(m_instances);
double rmsModel;
double adjustedErrorModel;
nodeModelEval.evaluateModel(m_nodeModel, m_instances);
rmsModel = nodeModelEval.rootMeanSquaredError();
adjustedErrorModel = rmsModel
* pruningFactor(m_numInstances, m_nodeModel.numParameters() + 1);
// Evaluate this node (ie its left and right subtrees)
Evaluation nodeEval = new Evaluation(m_instances);
double rmsSubTree;
double adjustedErrorNode;
int l_params = 0, r_params = 0;
nodeEval.evaluateModel(this, m_instances);
rmsSubTree = nodeEval.rootMeanSquaredError();
if (m_left != null) {
l_params = m_left.numParameters();
}
if (m_right != null) {
r_params = m_right.numParameters();
}
adjustedErrorNode = rmsSubTree
* pruningFactor(m_numInstances, (l_params + r_params + 1));
if ((adjustedErrorModel <= adjustedErrorNode)
|| (adjustedErrorModel < (m_globalDeviation * 0.00001))) {
// Choose linear model for this node rather than subtree model
m_isLeaf = true;
m_right = null;
m_left = null;
m_numParameters = m_nodeModel.numParameters() + 1;
m_rootMeanSquaredError = rmsModel;
} else {
m_numParameters = (l_params + r_params + 1);
m_rootMeanSquaredError = rmsSubTree;
}
}
// save space
if (!m_saveInstances) {
m_instances = new Instances(m_instances, 0);
}
}
/**
* Compute the pruning factor
*
* @param num_instances number of instances
* @param num_params number of parameters in the model
* @return the pruning factor
*/
private double pruningFactor(int num_instances, int num_params) {
if (num_instances <= num_params) {
return 10.0; // Caution says Yong in his code
}
return ((num_instances + m_pruningMultiplier * num_params) / (num_instances - num_params));
}
/**
* Find the leaf with greatest coverage
*
* @param maxCoverage the greatest coverage found so far
* @param bestLeaf the leaf with the greatest coverage
*/
public void findBestLeaf(double[] maxCoverage, RuleNode[] bestLeaf) {
if (!m_isLeaf) {
if (m_left != null) {
m_left.findBestLeaf(maxCoverage, bestLeaf);
}
if (m_right != null) {
m_right.findBestLeaf(maxCoverage, bestLeaf);
}
} else {
if (m_numInstances > maxCoverage[0]) {
maxCoverage[0] = m_numInstances;
bestLeaf[0] = this;
}
}
}
/**
* Return a list containing all the leaves in the tree
*
* @param v a single element array containing a vector of leaves
*/
public void returnLeaves(ArrayList[] v) {
if (m_isLeaf) {
v[0].add(this);
} else {
if (m_left != null) {
m_left.returnLeaves(v);
}
if (m_right != null) {
m_right.returnLeaves(v);
}
}
}
/**
* Get the parent of this node
*
* @return the parent of this node
*/
public RuleNode parentNode() {
return m_parent;
}
/**
* Get the left child of this node
*
* @return the left child of this node
*/
public RuleNode leftNode() {
return m_left;
}
/**
* Get the right child of this node
*
* @return the right child of this node
*/
public RuleNode rightNode() {
return m_right;
}
/**
* Get the index of the splitting attribute for this node
*
* @return the index of the splitting attribute
*/
public int splitAtt() {
return m_splitAtt;
}
/**
* Get the split point for this node
*
* @return the split point for this node
*/
public double splitVal() {
return m_splitValue;
}
/**
* Get the number of linear models in the tree
*
* @return the number of linear models
*/
public int numberOfLinearModels() {
if (m_isLeaf) {
return 1;
} else {
return m_left.numberOfLinearModels() + m_right.numberOfLinearModels();
}
}
/**
* Return true if this node is a leaf
*
* @return true if this node is a leaf
*/
public boolean isLeaf() {
return m_isLeaf;
}
/**
* Get the root mean squared error at this node
*
* @return the root mean squared error
*/
protected double rootMeanSquaredError() {
return m_rootMeanSquaredError;
}
/**
* Get the linear model at this node
*
* @return the linear model at this node
*/
public PreConstructedLinearModel getModel() {
return m_nodeModel;
}
/**
* Return the number of instances that reach this node.
*
* @return the number of instances at this node.
*/
public int getNumInstances() {
return m_numInstances;
}
/**
* Get the number of parameters in the model at this node
*
* @return the number of parameters in the model at this node
*/
private int numParameters() {
return m_numParameters;
}
/**
* Get the value of regressionTree.
*
* @return Value of regressionTree.
*/
public boolean getRegressionTree() {
return m_regressionTree;
}
/**
* Set the minumum number of instances to allow at a leaf node
*
* @param minNum the minimum number of instances
*/
public void setMinNumInstances(double minNum) {
m_splitNum = minNum;
}
/**
* Get the minimum number of instances to allow at a leaf node
*
* @return a double value
*/
public double getMinNumInstances() {
return m_splitNum;
}
/**
* Set the value of regressionTree.
*
* @param newregressionTree Value to assign to regressionTree.
*/
public void setRegressionTree(boolean newregressionTree) {
m_regressionTree = newregressionTree;
}
/**
* Print all the linear models at the learf (debugging purposes)
*/
public void printAllModels() {
if (m_isLeaf) {
System.out.println(m_nodeModel.toString());
} else {
System.out.println(m_nodeModel.toString());
m_left.printAllModels();
m_right.printAllModels();
}
}
/**
* Assigns a unique identifier to each node in the tree
*
* @param lastID last id number used
* @return ID after processing child nodes
*/
protected int assignIDs(int lastID) {
int currLastID = lastID + 1;
m_id = currLastID;
if (m_left != null) {
currLastID = m_left.assignIDs(currLastID);
}
if (m_right != null) {
currLastID = m_right.assignIDs(currLastID);
}
return currLastID;
}
/**
* Assign a unique identifier to each node in the tree and then calls
* graphTree
*
* @param text a StringBuffer value
*/
public void graph(StringBuffer text) {
assignIDs(-1);
graphTree(text);
}
/**
* Return a dotty style string describing the tree
*
* @param text a StringBuffer value
*/
protected void graphTree(StringBuffer text) {
text.append("N"
+ m_id
+ (m_isLeaf ? " [label=\"LM " + m_leafModelNum : " [label=\""
+ Utils.backQuoteChars(m_instances.attribute(m_splitAtt).name()))
+ (m_isLeaf ? " ("
+ ((m_globalDeviation > 0.0) ? m_numInstances
+ "/"
+ Utils.doubleToString(
(100.0 * m_rootMeanSquaredError / m_globalDeviation), 1, 3) + "%)"
: m_numInstances + ")") + "\" shape=box style=filled " : "\"")
+ (m_saveInstances ? "data=\n" + m_instances + "\n,\n" : "") + "]\n");
if (m_left != null) {
text.append("N" + m_id + "->" + "N" + m_left.m_id + " [label=\"<="
+ Utils.doubleToString(m_splitValue, 1, 3) + "\"]\n");
m_left.graphTree(text);
}
if (m_right != null) {
text.append("N" + m_id + "->" + "N" + m_right.m_id + " [label=\">"
+ Utils.doubleToString(m_splitValue, 1, 3) + "\"]\n");
m_right.graphTree(text);
}
}
/**
* Set whether to save instances for visualization purposes. Default is to
* save memory.
*
* @param save a boolean value
*/
protected void setSaveInstances(boolean save) {
m_saveInstances = save;
}
/**
* Returns the revision string.
*
* @return the revision
*/
@Override
public String getRevision() {
return RevisionUtils.extract("$Revision: 10283 $");
}
}
© 2015 - 2024 Weber Informatics LLC | Privacy Policy