Package smile.base.cart

Class DecisionNode

java.lang.Object
smile.base.cart.LeafNode
smile.base.cart.DecisionNode
All Implemented Interfaces:
Serializable, Node
public class DecisionNode extends LeafNode
A leaf node in decision tree.
See Also:

  • Constructor Details

    • DecisionNode

      public DecisionNode(int[] count)
      Constructor.
      Parameters:
      count - the number of node samples in each class.

  • Method Details

    • output

      public int output()
      Returns the predicted value.
      Returns:
      the predicted value.

    • count

      public int[] count()
      Returns the sample size in each class.
      Returns:
      the sample size in each class.

    • deviance

      public double deviance()
      Description copied from interface: Node
      Returns the deviance of node.
      Returns:
      the deviance of node.

    • dot

      public String dot(StructType schema, StructField response, int id)
      Description copied from interface: Node
      Returns the dot representation of node.
      Parameters:
      schema - the schema of data
      response - the schema of response variable
      id - node id
      Returns:
      the dot representation of node.

    • toString

      public int[] toString(StructType schema, StructField response, InternalNode parent, int depth, BigInteger id, List<String> lines)
      Description copied from interface: Node
      Adds the string representation (R's rpart format) to a collection.
      Parameters:
      schema - the schema of data
      response - the schema of response variable
      parent - the parent node
      depth - the depth of node in the tree. The root node is at depth 0.
      id - node id
      lines - the collection of node's string representation.
      Returns:
      the sample count of each class for decision tree; single element array [node size] for regression tree.

    • impurity

      public double impurity(SplitRule rule)
      Returns the impurity of node.
      Parameters:
      rule - the node split rule.
      Returns:
      the impurity of node

    • impurity

      public static double impurity(SplitRule rule, int size, int[] count)
      Returns the impurity of samples.
      Parameters:
      rule - the node split rule.
      size - the number of samples.
      count - the number of samples in each class.
      Returns:
      the impurity of node

    • equals

      public boolean equals(Object o)
      Overrides:
      equals in class Object

    • posteriori

      public double[] posteriori(double[] prob)
      Returns the class probability.
      Parameters:
      prob - the output variable of posteriori probabilities.
      Returns:
      the posteriori probabilities.

    • posteriori

      public static double[] posteriori(int[] count, double[] prob)
      Returns the class probability.
      Parameters:
      count - the input variable of the number of samples per class.
      prob - the output variable of posteriori probabilities.
      Returns:
      the posteriori probabilities.

    • deviance

      public static double deviance(int[] count, double[] prob)
      Returns the deviance of node.
      Parameters:
      count - the input variable of the number of samples per class.
      prob - the output variable of posteriori probabilities.
      Returns:
      the deviance of node.