hoeffding_categorical_split.hpp
Go to the documentation of this file.
1 
13 #ifndef MLPACK_METHODS_HOEFFDING_TREES_HOEFFDING_CATEGORICAL_SPLIT_HPP
14 #define MLPACK_METHODS_HOEFFDING_TREES_HOEFFDING_CATEGORICAL_SPLIT_HPP
15 
16 #include <mlpack/prereqs.hpp>
18 
19 namespace mlpack {
20 namespace tree {
21 
43 template<typename FitnessFunction>
45 {
46  public:
49 
57  HoeffdingCategoricalSplit(const size_t numCategories = 0,
58  const size_t numClasses = 0);
59 
66  HoeffdingCategoricalSplit(const size_t numCategories,
67  const size_t numClasses,
68  const HoeffdingCategoricalSplit& other);
69 
76  template<typename eT>
77  void Train(eT value, const size_t label);
78 
89  void EvaluateFitnessFunction(double& bestFitness, double& secondBestFitness)
90  const;
91 
93  size_t NumChildren() const { return sufficientStatistics.n_cols; }
94 
102  void Split(arma::Col<size_t>& childMajorities, SplitInfo& splitInfo);
103 
105  size_t MajorityClass() const;
107  double MajorityProbability() const;
108 
110  template<typename Archive>
111  void serialize(Archive& ar, const uint32_t /* version */)
112  {
113  ar(CEREAL_NVP(sufficientStatistics));
114  }
115 
116  private:
120  arma::Mat<size_t> sufficientStatistics;
121 };
122 
123 } // namespace tree
124 } // namespace mlpack
125 
126 // Include implementation.
127 #include "hoeffding_categorical_split_impl.hpp"
128 
129 #endif
void Split(arma::Col< size_t > &childMajorities, SplitInfo &splitInfo)
Gather the information for a split: get the labels of the child majorities, and initialize the SplitI...
Linear algebra utility functions, generally performed on matrices or vectors.
The core includes that mlpack expects; standard C++ includes and Armadillo.
void Train(eT value, const size_t label)
Train on the given value with the given label.
double MajorityProbability() const
Get the probability of the majority class given the points seen so far.
HoeffdingCategoricalSplit(const size_t numCategories=0, const size_t numClasses=0)
Create the HoeffdingCategoricalSplit given a number of categories for this dimension and a number of ...
void serialize(Archive &ar, const uint32_t)
Serialize the categorical split.
CategoricalSplitInfo SplitInfo
The type of split information required by the HoeffdingCategoricalSplit.
This is the standard Hoeffding-bound categorical feature proposed in the paper below: ...
size_t NumChildren() const
Return the number of children, if the node were to split.
size_t MajorityClass() const
Get the majority class seen so far.
void EvaluateFitnessFunction(double &bestFitness, double &secondBestFitness) const
Given the points seen so far, evaluate the fitness function, returning the gain for the best possible...