| weka.classifiers.RandomizableClassifier
 weka.classifiers.trees.SimpleCart
| SimpleCart | | public class SimpleCart extends RandomizableClassifier implements AdditionalMeasureProducer,TechnicalInformationHandler(Code) | |
Class implementing minimal cost-complexity pruning.
Note when dealing with missing values, use "fractional instances" method instead of surrogate split method.
For more information, see:
Leo Breiman, Jerome H. Friedman, Richard A. Olshen, Charles J. Stone (1984). Classification and Regression Trees. Wadsworth International Group, Belmont, California.
BibTeX:
@book{Breiman1984,
address = {Belmont, California},
author = {Leo Breiman and Jerome H. Friedman and Richard A. Olshen and Charles J. Stone},
publisher = {Wadsworth International Group},
title = {Classification and Regression Trees},
year = {1984}
}
Valid options are:
-S <num>
Random number seed.
(default 1)
-D
If set, classifier is run in debug mode and
may output additional info to the console
-M <min no>
The minimal number of instances at the terminal nodes.
(default 2)
-N <num folds>
The number of folds used in the minimal cost-complexity pruning.
(default 5)
-U
Don't use the minimal cost-complexity pruning.
(default yes).
-H
Don't use the heuristic method for binary split.
(default true).
-A
Use 1 SE rule to make pruning decision.
(default no).
-C
Percentage of training data size (0-1].
(default 1).
author:
Haijian Shi (hs69@cs.waikato.ac.nz)
version:
$Revision: 1.2 $ |
| Method Summary | |
| public void | buildClassifier(Instances data)
Build the classifier. | | public void | calculateAlphas()
Updates the alpha field for all nodes. | | protected double | computeGini(double[] dist, double total)
Compute and return gini index for a given distribution of a node. | | protected double | computeGiniGain(double[] parentDist, double[][] childDist)
Compute and return gini gain for given distributions of a node and its
successor nodes. | | protected double | computeSortedInfo(Instances data, int[][] sortedIndices, double[][] weights, double[] classProbs)
Compute sorted indices, weights and class probabilities for a given
dataset. | | public double[] | distributionForInstance(Instance instance)
Computes class probabilities for instance using the decision tree. | | public Enumeration | enumerateMeasures()
Return an enumeration of the measure names. | | protected void | fillInnerNodes(Vector nodeList)
Fills a list with all inner nodes in the tree. | | public Capabilities | getCapabilities()
Returns default capabilities of the classifier. | | public boolean | getHeuristic()
Get if use heuristic search for nominal attributes in multi-class problems. | | protected Vector | getInnerNodes()
Return a list of all inner nodes in the tree. | | public double | getMeasure(String additionalMeasureName)
Returns the value of the named measure. | | public double | getMinNumObj()
Get minimal number of instances at the terminal nodes. | | public int | getNumFoldsPruning()
Set number of folds in internal cross-validation. | | public String[] | getOptions()
Gets the current settings of the classifier. | | public double | getSizePer()
Get training set size. | | public TechnicalInformation | getTechnicalInformation()
Returns an instance of a TechnicalInformation object, containing
detailed information about the technical background of this class,
e.g., paper reference or book this class is based on. | | public boolean | getUseOneSE()
Get if use the 1SE rule to choose final model. | | public boolean | getUsePrune()
Get if use minimal cost-complexity pruning. | | public String | globalInfo()
Return a description suitable for displaying in the explorer/experimenter. | | public String | heuristicTipText()
| | public Enumeration | listOptions()
Returns an enumeration describing the available options. | | public static void | main(String[] args)
Main method. | | protected void | makeLeaf(Instances data)
Make the node leaf node. | | protected void | makeTree(Instances data, int totalInstances, int[][] sortedIndices, double[][] weights, double[] classProbs, double totalWeight, double minNumObj, boolean useHeuristic)
Make binary decision tree recursively. | | public double | measureTreeSize()
Return number of tree size. | | public String | minNumObjTipText()
| | public void | modelErrors()
Updates the numIncorrectModel field for all nodes when subtree (to be
pruned) is rooted. | | protected SimpleCart | nodeToPrune(Vector nodeList)
Find the node with minimal alpha value. | | protected String | nominalDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] giniGains, Instances data, boolean useHeuristic)
Compute distributions, proportions and total weights of two successor
nodes for a given nominal attribute. | | public String | numFoldsPruningTipText()
| | public int | numInnerNodes()
Method to count the number of inner nodes in the tree. | | public int | numLeaves()
Compute number of leaf nodes. | | public int | numNodes()
Compute size of the tree. | | protected double | numericDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] giniGains, Instances data)
Compute distributions, proportions and total weights of two successor
nodes for a given numeric attribute. | | public void | prune(double alpha)
Prunes the original tree using the CART pruning scheme, given a
cost-complexity parameter alpha. | | public int | prune(double[] alphas, double[] errors, Instances test)
Method for performing one fold in the cross-validation of minimal
cost-complexity pruning. | | public void | setHeuristic(boolean value)
Set if use heuristic search for nominal attributes in multi-class problems. | | public void | setMinNumObj(double value)
Set minimal number of instances at the terminal nodes. | | public void | setNumFoldsPruning(int value)
Set number of folds in internal cross-validation. | | public void | setOptions(String[] options)
Parses a given list of options. | | public void | setSizePer(double value)
Set training set size. | | public void | setUseOneSE(boolean value)
Set if use the 1SE rule to choose final model. | | public void | setUsePrune(boolean value)
Set if use minimal cost-complexity pruning. | | public String | sizePerTipText()
| | protected void | splitData(int[][][] subsetIndices, double[][][] subsetWeights, Attribute att, double splitPoint, String splitStr, int[][] sortedIndices, double[][] weights, Instances data)
Split data into two subsets and store sorted indices and weights for two
successor nodes. | | public String | toString()
Prints the decision tree using the protected toString method from below. | | protected String | toString(int level)
Outputs a tree at a certain level. | | public void | treeErrors()
Updates the numIncorrectTree field for all nodes. | | protected void | unprune()
Method to "unprune" the CART tree. | | public String | useOneSETipText()
| | public String | usePruneTipText()
|
| m_Alpha | | protected double m_Alpha(Code) | | Alpha-value (for pruning) at the node.
|
| m_Attribute | | protected Attribute m_Attribute(Code) | | Attribute used to split data.
|
| m_ClassAttribute | | protected Attribute m_ClassAttribute(Code) | | Class attriubte of data.
|
| m_ClassProbs | | protected double[] m_ClassProbs(Code) | | Class probabilities.
|
| m_ClassValue | | protected double m_ClassValue(Code) | | Class value if the node is leaf.
|
| m_Distribution | | protected double[] m_Distribution(Code) | | Distributions of leaf node (or temporary leaf node in minimal cost-complexity pruning)
|
| m_Heuristic | | protected boolean m_Heuristic(Code) | | If use huristic search for nominal attributes in multi-class problems (default true).
|
| m_Props | | protected double[] m_Props(Code) | | Proportion for each branch.
|
| m_Prune | | protected boolean m_Prune(Code) | | If use minimal cost-compexity pruning.
|
| m_SizePer | | protected double m_SizePer(Code) | | Training data size.
|
| m_SplitString | | protected String m_SplitString(Code) | | Split subset used to split data for nominal attributes.
|
| m_SplitValue | | protected double m_SplitValue(Code) | | Split point for a numeric attribute.
|
| m_UseOneSE | | protected boolean m_UseOneSE(Code) | | If use the 1SE rule to make final decision tree.
|
| m_isLeaf | | protected boolean m_isLeaf(Code) | | Indicate if the node is a leaf node.
|
| m_minNumObj | | protected double m_minNumObj(Code) | | Minimum number of instances in at the terminal nodes.
|
| m_numFoldsPruning | | protected int m_numFoldsPruning(Code) | | Number of folds for minimal cost-complexity pruning.
|
| m_numIncorrectModel | | protected double m_numIncorrectModel(Code) | | Number of training examples misclassified by the model (subtree rooted).
|
| m_numIncorrectTree | | protected double m_numIncorrectTree(Code) | | Number of training examples misclassified by the model (subtree not rooted).
|
| m_totalTrainInstances | | protected int m_totalTrainInstances(Code) | | Total number of instances used to build the classifier.
|
| buildClassifier | | public void buildClassifier(Instances data) throws Exception(Code) | | Build the classifier.
Parameters:
data - the training instances
throws:
Exception - if something goes wrong |
| calculateAlphas | | public void calculateAlphas() throws Exception(Code) | | Updates the alpha field for all nodes.
throws:
Exception - if something goes wrong |
| computeGini | | protected double computeGini(double[] dist, double total)(Code) | | Compute and return gini index for a given distribution of a node.
Parameters:
dist - class distributions
Parameters:
total - class distributions Gini index of the class distributions |
| computeGiniGain | | protected double computeGiniGain(double[] parentDist, double[][] childDist)(Code) | | Compute and return gini gain for given distributions of a node and its
successor nodes.
Parameters:
parentDist - class distributions of parent node
Parameters:
childDist - class distributions of successor nodes Gini gain computed |
| computeSortedInfo | | protected double computeSortedInfo(Instances data, int[][] sortedIndices, double[][] weights, double[] classProbs) throws Exception(Code) | | Compute sorted indices, weights and class probabilities for a given
dataset. Return total weights of the data at the node.
Parameters:
data - training data
Parameters:
sortedIndices - sorted indices of instances at the node
Parameters:
weights - weights of instances at the node
Parameters:
classProbs - class probabilities at the node total weights of instances at the node
throws:
Exception - if something goes wrong |
| distributionForInstance | | public double[] distributionForInstance(Instance instance) throws Exception(Code) | | Computes class probabilities for instance using the decision tree.
Parameters:
instance - the instance for which class probabilities is to be computed the class probabilities for the given instance
throws:
Exception - if something goes wrong |
| enumerateMeasures | | public Enumeration enumerateMeasures()(Code) | | Return an enumeration of the measure names.
an enumeration of the measure names |
| fillInnerNodes | | protected void fillInnerNodes(Vector nodeList)(Code) | | Fills a list with all inner nodes in the tree.
Parameters:
nodeList - the list to be filled |
| getCapabilities | | public Capabilities getCapabilities()(Code) | | Returns default capabilities of the classifier.
the capabilities of this classifier |
| getHeuristic | | public boolean getHeuristic()(Code) | | Get if use heuristic search for nominal attributes in multi-class problems.
if use heuristic search for nominal attributes in multi-class problems |
| getInnerNodes | | protected Vector getInnerNodes()(Code) | | Return a list of all inner nodes in the tree.
the list of all inner nodes |
| getMeasure | | public double getMeasure(String additionalMeasureName)(Code) | | Returns the value of the named measure.
Parameters:
additionalMeasureName - the name of the measure to query for its value the value of the named measure
throws:
IllegalArgumentException - if the named measure is not supported |
| getMinNumObj | | public double getMinNumObj()(Code) | | Get minimal number of instances at the terminal nodes.
minimal number of instances at the terminal nodes |
| getNumFoldsPruning | | public int getNumFoldsPruning()(Code) | | Set number of folds in internal cross-validation.
number of folds in internal cross-validation. |
| getOptions | | public String[] getOptions()(Code) | | Gets the current settings of the classifier.
the current setting of the classifier |
| getSizePer | | public double getSizePer()(Code) | | Get training set size.
training set size |
| getTechnicalInformation | | public TechnicalInformation getTechnicalInformation()(Code) | | Returns an instance of a TechnicalInformation object, containing
detailed information about the technical background of this class,
e.g., paper reference or book this class is based on.
the technical information about this class |
| getUseOneSE | | public boolean getUseOneSE()(Code) | | Get if use the 1SE rule to choose final model.
if use the 1SE rule to choose final model |
| getUsePrune | | public boolean getUsePrune()(Code) | | Get if use minimal cost-complexity pruning.
if use minimal cost-complexity pruning |
| globalInfo | | public String globalInfo()(Code) | | Return a description suitable for displaying in the explorer/experimenter.
a description suitable for displaying in the explorer/experimenter |
| heuristicTipText | | public String heuristicTipText()(Code) | | Returns the tip text for this property
tip text for this property suitable fordisplaying in the explorer/experimenter gui. |
| listOptions | | public Enumeration listOptions()(Code) | | Returns an enumeration describing the available options.
an enumeration of all the available options. |
| main | | public static void main(String[] args)(Code) | | Main method.
Parameters:
args - the options for the classifier |
| makeLeaf | | protected void makeLeaf(Instances data)(Code) | | Make the node leaf node.
Parameters:
data - trainging data |
| makeTree | | protected void makeTree(Instances data, int totalInstances, int[][] sortedIndices, double[][] weights, double[] classProbs, double totalWeight, double minNumObj, boolean useHeuristic) throws Exception(Code) | | Make binary decision tree recursively.
Parameters:
data - the training instances
Parameters:
totalInstances - total number of instances
Parameters:
sortedIndices - sorted indices of the instances
Parameters:
weights - weights of the instances
Parameters:
classProbs - class probabilities
Parameters:
totalWeight - total weight of instances
Parameters:
minNumObj - minimal number of instances at leaf nodes
Parameters:
useHeuristic - if use heuristic search for nominal attributes in multi-class problem
throws:
Exception - if something goes wrong |
| measureTreeSize | | public double measureTreeSize()(Code) | | Return number of tree size.
number of tree size |
| minNumObjTipText | | public String minNumObjTipText()(Code) | | Returns the tip text for this property
tip text for this property suitable fordisplaying in the explorer/experimenter gui |
| modelErrors | | public void modelErrors() throws Exception(Code) | | Updates the numIncorrectModel field for all nodes when subtree (to be
pruned) is rooted. This is needed for calculating the alpha-values.
throws:
Exception - if something goes wrong |
| nodeToPrune | | protected SimpleCart nodeToPrune(Vector nodeList)(Code) | | Find the node with minimal alpha value. If two nodes have the same alpha,
choose the one with more leave nodes.
Parameters:
nodeList - list of inner nodes the node to be pruned |
| nominalDistribution | | protected String nominalDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] giniGains, Instances data, boolean useHeuristic) throws Exception(Code) | | Compute distributions, proportions and total weights of two successor
nodes for a given nominal attribute.
Parameters:
props - proportions of each two branches for each attribute
Parameters:
dists - class distributions of two branches for each attribute
Parameters:
att - numeric att split on
Parameters:
sortedIndices - sorted indices of instances for the attirubte
Parameters:
weights - weights of instances for the attirbute
Parameters:
subsetWeights - total weight of two branches split based on the attribute
Parameters:
giniGains - Gini gains for each attribute
Parameters:
data - training instances
Parameters:
useHeuristic - if use heuristic search Gini gain for the given nominal attribute
throws:
Exception - if something goes wrong |
| numFoldsPruningTipText | | public String numFoldsPruningTipText()(Code) | | Returns the tip text for this property
tip text for this property suitable fordisplaying in the explorer/experimenter gui |
| numInnerNodes | | public int numInnerNodes()(Code) | | Method to count the number of inner nodes in the tree.
the number of inner nodes |
| numLeaves | | public int numLeaves()(Code) | | Compute number of leaf nodes.
number of leaf nodes |
| numNodes | | public int numNodes()(Code) | | Compute size of the tree.
size of the tree |
| numericDistribution | | protected double numericDistribution(double[][] props, double[][][] dists, Attribute att, int[] sortedIndices, double[] weights, double[][] subsetWeights, double[] giniGains, Instances data) throws Exception(Code) | | Compute distributions, proportions and total weights of two successor
nodes for a given numeric attribute.
Parameters:
props - proportions of each two branches for each attribute
Parameters:
dists - class distributions of two branches for each attribute
Parameters:
att - numeric att split on
Parameters:
sortedIndices - sorted indices of instances for the attirubte
Parameters:
weights - weights of instances for the attirbute
Parameters:
subsetWeights - total weight of two branches split based on the attribute
Parameters:
giniGains - Gini gains for each attribute
Parameters:
data - training instances Gini gain the given numeric attribute
throws:
Exception - if something goes wrong |
| prune | | public void prune(double alpha) throws Exception(Code) | | Prunes the original tree using the CART pruning scheme, given a
cost-complexity parameter alpha.
Parameters:
alpha - the cost-complexity parameter
throws:
Exception - if something goes wrong |
| prune | | public int prune(double[] alphas, double[] errors, Instances test) throws Exception(Code) | | Method for performing one fold in the cross-validation of minimal
cost-complexity pruning. Generates a sequence of alpha-values with error
estimates for the corresponding (partially pruned) trees, given the test
set of that fold.
Parameters:
alphas - array to hold the generated alpha-values
Parameters:
errors - array to hold the corresponding error estimates
Parameters:
test - test set of that fold (to obtain error estimates) the iteration of the pruning
throws:
Exception - if something goes wrong |
| setHeuristic | | public void setHeuristic(boolean value)(Code) | | Set if use heuristic search for nominal attributes in multi-class problems.
Parameters:
value - if use heuristic search for nominal attributes in multi-class problems |
| setMinNumObj | | public void setMinNumObj(double value)(Code) | | Set minimal number of instances at the terminal nodes.
Parameters:
value - minimal number of instances at the terminal nodes |
| setNumFoldsPruning | | public void setNumFoldsPruning(int value)(Code) | | Set number of folds in internal cross-validation.
Parameters:
value - number of folds in internal cross-validation. |
| setOptions | | public void setOptions(String[] options) throws Exception(Code) | | Parses a given list of options.
Valid options are:
-S <num>
Random number seed.
(default 1)
-D
If set, classifier is run in debug mode and
may output additional info to the console
-M <min no>
The minimal number of instances at the terminal nodes.
(default 2)
-N <num folds>
The number of folds used in the minimal cost-complexity pruning.
(default 5)
-U
Don't use the minimal cost-complexity pruning.
(default yes).
-H
Don't use the heuristic method for binary split.
(default true).
-A
Use 1 SE rule to make pruning decision.
(default no).
-C
Percentage of training data size (0-1].
(default 1).
Parameters:
options - the list of options as an array of strings
throws:
Exception - if an options is not supported |
| setSizePer | | public void setSizePer(double value)(Code) | | Set training set size.
Parameters:
value - training set size |
| setUseOneSE | | public void setUseOneSE(boolean value)(Code) | | Set if use the 1SE rule to choose final model.
Parameters:
value - if use the 1SE rule to choose final model |
| setUsePrune | | public void setUsePrune(boolean value)(Code) | | Set if use minimal cost-complexity pruning.
Parameters:
value - if use minimal cost-complexity pruning |
| sizePerTipText | | public String sizePerTipText()(Code) | | Returns the tip text for this property
tip text for this property suitable fordisplaying in the explorer/experimenter gui. |
| splitData | | protected void splitData(int[][][] subsetIndices, double[][][] subsetWeights, Attribute att, double splitPoint, String splitStr, int[][] sortedIndices, double[][] weights, Instances data) throws Exception(Code) | | Split data into two subsets and store sorted indices and weights for two
successor nodes.
Parameters:
subsetIndices - sorted indecis of instances for each attribute for two successor node
Parameters:
subsetWeights - weights of instances for each attribute for two successor node
Parameters:
att - attribute the split based on
Parameters:
splitPoint - split point the split based on if att is numeric
Parameters:
splitStr - split subset the split based on if att is nominal
Parameters:
sortedIndices - sorted indices of the instances to be split
Parameters:
weights - weights of the instances to bes split
Parameters:
data - training data
throws:
Exception - if something goes wrong |
| toString | | public String toString()(Code) | | Prints the decision tree using the protected toString method from below.
a textual description of the classifier |
| toString | | protected String toString(int level)(Code) | | Outputs a tree at a certain level.
Parameters:
level - the level at which the tree is to be printed a tree at a certain level |
| treeErrors | | public void treeErrors() throws Exception(Code) | | Updates the numIncorrectTree field for all nodes. This is needed for
calculating the alpha-values.
throws:
Exception - if something goes wrong |
| unprune | | protected void unprune()(Code) | | Method to "unprune" the CART tree. Sets all leaf-fields to false.
Faster than re-growing the tree because CART do not have to be fit again.
|
| useOneSETipText | | public String useOneSETipText()(Code) | | Returns the tip text for this property
tip text for this property suitable fordisplaying in the explorer/experimenter gui. |
| usePruneTipText | | public String usePruneTipText()(Code) | | Return the tip text for this property
tip text for this property suitable for displaying in the explorer/experimenter gui. |
| Fields inherited from weka.classifiers.RandomizableClassifier | protected int m_Seed(Code)(Java Doc)
|
|
|