source: trunk/yat/statistics/AUC.cc @ 1703

// $Id: AUC.cc 1487 2008-09-10 08:41:36Z jari $

/*
  Copyright (C) 2004, 2005 Peter Johansson
  Copyright (C) 2006, 2007 Jari Häkkinen, Peter Johansson
  Copyright (C) 2008 Peter Johansson

  This file is part of the yat library, http://dev.thep.lu.se/yat

  The yat library 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.

  The yat library 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 yat. If not, see <http://www.gnu.org/licenses/>.
*/

#include "AUC.h"
#include "yat/classifier/DataLookupWeighted1D.h"
#include "yat/classifier/Target.h"
#include "yat/utility/stl_utility.h"
#include "yat/utility/VectorBase.h"

#include <cassert>
#include <cmath>
#include <utility>
#include <map>

namespace theplu {
namespace yat {
namespace statistics {  

  AUC::AUC(bool absolute)
    : Score(absolute)
  {
  }

  double AUC::score(const classifier::Target& target, 
                    const utility::VectorBase& value) const
  {
    assert(target.size()==value.size());
    // key data, pair<target, weight>
    std::multimap<double, std::pair<bool, double> > m;
    for (unsigned int i=0; i<target.size(); i++)
      m.insert(std::make_pair(value(i), 
                              std::make_pair(target.binary(i),1.0)));
        
    return score(m);
  }



  double AUC::score(const classifier::Target& target, 
                    const classifier::DataLookupWeighted1D& value) const
  {
    assert(target.size()==value.size());
    // key data, pair<target, weight>
    std::multimap<double, std::pair<bool, double> > m;
    for (unsigned int i=0; i<target.size(); i++)
      if (value.weight(i))
        m.insert(std::make_pair(value.data(i), 
                                std::make_pair(target.binary(i), 
                                               value.weight(i))));
        
    return score(m);
  }


  double AUC::score(const classifier::Target& target, 
                    const utility::VectorBase& value, 
                    const utility::VectorBase& weight) const
  {
    assert(target.size()==value.size());
    assert(target.size()==weight.size());
    // key data, pair<target, weight>
    std::multimap<double, std::pair<bool, double> > m;
    for (unsigned int i=0; i<target.size(); i++)
      if (weight(i))
        m.insert(std::make_pair(value(i), 
                                std::make_pair(target.binary(i), weight(i))));
        
    return score(m);
  }


  double AUC::score(const MultiMap& m) const
  {
    double area=0;
    double cumsum_pos_w=0;
    double cumsum_neg_w=0;
    double cache_cumsum_pos_w=0;
    double cache_cumsum_neg_w=0;
    typedef MultiMap::const_iterator iter;

    for (iter i=m.begin(); i!=m.end(); ++i) {
      if (i->second.first){
        cache_cumsum_pos_w += i->second.second;
      }
      else
        cache_cumsum_neg_w += i->second.second;

      iter n(i);
      ++n;
      // last entry OR entry is smaller than next one
      if ( n==m.end() || i->first < n->first) {
        if (i->second.first)
          area += cache_cumsum_pos_w * ( cumsum_neg_w + cache_cumsum_neg_w/2 );
        cumsum_pos_w += cache_cumsum_pos_w;
        cache_cumsum_pos_w = 0;
        cumsum_neg_w += cache_cumsum_neg_w;
        cache_cumsum_neg_w = 0;
      }
    }
    // max area is cumsum_neg_w * cumsum_pos_w
    area/=(cumsum_neg_w*cumsum_pos_w);
    
    if (area<0.5 && absolute_)
      return 1.0-area;
    return area;
}

}}} // of namespace statistics, yat, and theplu