source: trunk/yat/classifier/ConsensusInputRanker.cc @ 2119

// $Id$

/*
  Copyright (C) 2004 Jari Häkkinen, Peter Johansson
  Copyright (C) 2005 Peter Johansson
  Copyright (C) 2006 Jari Häkkinen, Peter Johansson, Markus Ringnér
  Copyright (C) 2007, 2008 Jari Häkkinen, 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 "ConsensusInputRanker.h"
#include "InputRanker.h"
#include "IRRetrieve.h"
#include "MatrixLookup.h"
#include "MatrixLookupWeighted.h"
#include "Sampler.h"
#include "Target.h"
#include "yat/statistics/Score.h"
#include "yat/statistics/utility.h"
#include "yat/statistics/VectorFunction.h"
#include "yat/utility/stl_utility.h"

#include <cassert>
#include <functional>
#include <iostream>
#include <utility>
#include <vector>
#include <cmath>

namespace theplu {
namespace yat {
namespace classifier {  

  ConsensusInputRanker::ConsensusInputRanker(const IRRetrieve& retriever,
                                             const statistics::VectorFunction&
                                             vf)
    : retriever_(retriever), vec_func_(vf)
  {
  }


  void ConsensusInputRanker::add(const Sampler& sampler, 
                                 const MatrixLookup& data, 
                                 const statistics::Score& score)
  {
    assert(sampler.size());
    assert(id_.empty() || id_.size()==data.rows());
    input_rankers_.reserve(sampler.size()+input_rankers_.size());
    id_.resize(data.rows());
    rank_.resize(data.rows());
    for (size_t i=0; i<sampler.size(); ++i){
      input_rankers_.push_back(InputRanker(MatrixLookup(data,sampler.training_index(i), false), 
                                           sampler.training_target(i), 
                                           score));       
    }
    update();
  }

  void ConsensusInputRanker::add(const Sampler& sampler, 
                                 const MatrixLookupWeighted& data, 
                                 const statistics::Score& score)
  {
    assert(id_.empty() || id_.size()==data.rows());
    id_.resize(data.rows());
    rank_.resize(data.rows());
    for (size_t i=0; i<sampler.size(); ++i){
      input_rankers_.push_back(InputRanker(MatrixLookupWeighted(data,sampler.training_index(i), false), 
                                           sampler.training_target(i), 
                                           score));       
    }
    update();
  }

  void ConsensusInputRanker::add(const InputRanker& ir)
  {
    assert(id_.empty() || id_.size()==ir.id().size());
    input_rankers_.push_back(ir);
  }

  size_t ConsensusInputRanker::id(size_t i) const
  {
    assert(i<id_.size());
    return id_[i];
  }

  const InputRanker& ConsensusInputRanker::input_ranker(size_t i) const
  {
    assert(i<input_rankers_.size());
    return input_rankers_[i];
  }

  size_t ConsensusInputRanker::rank(size_t i) const
  {
    return rank_[i];
  }


  void ConsensusInputRanker::reserve(size_t n)
  {
    input_rankers_.reserve(n);
  }


  void ConsensusInputRanker::update(void)
  {
    // Sorting with respect to VectorFunction(info) where info is a
    // vector and each element contains infomation retrieved with
    // retriever_ from each InputRanker
    std::vector<std::pair<double,size_t> > cons_rank;
    cons_rank.reserve(id_.size());
    for (size_t i=0; i<id_.size(); i++){ 
      std::vector<double> scores;
      scores.reserve(input_rankers_.size());
      for (size_t j=0; j<input_rankers_.size(); j++) {
        scores.push_back(retriever_(input_rankers_[j],i));
      }
      cons_rank.push_back(std::make_pair(vec_func_(scores), i));
    }
    
    //sort cons_rank and assign id_ and rank_
    sort(cons_rank.begin(), cons_rank.end(),
         std::greater<std::pair<double, size_t> >()); 
         
    for (size_t i=0; i<cons_rank.size(); i++){
      assert(i<id_.size());
      id_[i]=cons_rank[i].second;
      assert(id_[i]<rank_.size());
      rank_[id_[i]]=i;
    }
  }


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