drwnRandomForest.h

/******************************************************************************
** DARWIN: A FRAMEWORK FOR MACHINE LEARNING RESEARCH AND DEVELOPMENT
** Distributed under the terms of the BSD license (see the LICENSE file)
** Copyright (c) 2007-2015, Stephen Gould
** All rights reserved.
**
******************************************************************************
** FILENAME:    drwnRandomForest.h
** AUTHOR(S):   Stephen Gould <stephen.gould@anu.edu.au>
**
*****************************************************************************/

#pragma once

#include <cstdlib>
#include <vector>

#include "drwnBase.h"
#include "drwnClassifier.h"
#include "drwnDecisionTree.h"

using namespace std;

// drwnRandomForest ----------------------------------------------------------

class drwnRandomForest : public drwnClassifier {
 public:
    // default training parameters
    static int NUM_TREES;    
    static int MAX_DEPTH;    
    static int MAX_FEATURES; 

 protected:
    // actual training parameters
    int _numTrees;                 
    int _maxDepth;                 
    int _maxFeatures;              

    vector<drwnDecisionTree *> _forest;
    vector<double> _alphas;

 public:
    drwnRandomForest();
    drwnRandomForest(unsigned n, unsigned k = 2);
    drwnRandomForest(const drwnRandomForest &c);
    ~drwnRandomForest();

    // access functions
    virtual const char *type() const { return "drwnRandomForest"; }
    virtual drwnRandomForest *clone() const { return new drwnRandomForest(*this); }

    // initialization
    virtual void initialize(unsigned n, unsigned k = 2);

    // i/o
    virtual bool save(drwnXMLNode& node) const;
    virtual bool load(drwnXMLNode& node);

    // training
    using drwnClassifier::train;
    virtual double train(const drwnClassifierDataset& dataset);

    // evaluation (log-probability)
    using drwnClassifier::getClassScores;
    virtual void getClassScores(const vector<double>& features,
        vector<double>& outputScores) const;
};