result_tracker_dupless.hpp

Go to the documentation of this file.

#ifndef FSTRESULTTRACKERDUPLESS_H
#define FSTRESULTTRACKERDUPLESS_H
 /* 
=========================================================================
Copyright and license:
  * FST4 software (with the exception of any bundled or externally linked
    third-party components) is Copyright (C) 2000-present Institute of
    Information Theory and Automation of the Czech Academy of Sciences (UTIA).
  * FST4 is free software: you can redistribute it and/or modify it under
    the terms of the GNU Affero General Public License as published by the
    Free Software Foundation, version 3. The full text of the license is in
    the file LICENSE in the distribution root and at
    https://www.gnu.org/licenses/agpl-3.0.html.
  * FST4 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 Affero General Public
    License for more details.
  * Closed-source use without the AGPL's copyleft obligations requires a
    commercial license from UTIA; in case of interest write to
    utia@utia.cas.cz. LICENSING.md in the distribution root explains both
    options, the contributor license agreement that every contribution
    requires, and the contacts (technical questions and pull requests:
    somol@utia.cas.cz, with "FST" in the subject line).
========================================================================== */
 
#include <memory>
#include <cassert>
#include <iostream>
#include <list>
#include "error.hpp"
#include "global.hpp"
#include "result_tracker.hpp"
 
/*============== Template parameter type naming conventions ==============
--------- Numeric types: -------------------------------------------------
DATATYPE - data sample values - usually real numbers (but may be integers
          in text processing etc.)
REALTYPE - must be real numbers - for representing intermediate results of 
          calculations like mean, covariance etc.
IDXTYPE - index values for enumeration of data samples - (nonnegative) integers, 
          extent depends on numbers of samples in data
DIMTYPE - index values for enumeration of features (dimensions), or classes (not 
          class sizes) - (nonnegative) integers, usually lower extent than IDXTYPE, 
          but be aware of expressions like _classes*_features*_features ! 
          in linearized representations of feature matrices for all classes
BINTYPE - feature selection marker type - represents ca. <10 different feature 
          states (selected, deselected, sel./desel. temporarily 1st nested loop, 2nd...)
RETURNTYPE - criterion value: real value, but may be extended in future to support 
          multiple values 
--------- Class types: ---------------------------------------------------
SUBSET       - class of class type Subset 
CLASSIFIER   - class implementing interface defined in abstract class Classifier 
EVALUATOR    - class implementing interface defined in abstract class Sequential_Step 
DISTANCE     - class implementing interface defined in abstract class Distance 
DATAACCESSOR - class implementing interface defined in abstract class Data_Accessor 
INTERVALCONTAINER - class of class type Data_Intervaller 
CONTAINER    - STL container of class type Data_Interval  
========================================================================== */
 
namespace FST {
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
class Result_Tracker_Dupless : public Clonable_Impl<Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>,Result_Tracker<RETURNTYPE,SUBSET>,CLONE_INHERIT,CLONE_INHERIT,CLONE_DEEP> {
public:
        typedef Result_Tracker<RETURNTYPE,SUBSET> parent; 
        typedef Clonable_Impl<Result_Tracker_Dupless,parent,CLONE_INHERIT,CLONE_INHERIT,CLONE_DEEP> parent_impl; 
        typedef typename parent::ResultRec ResultRec;     
        typedef typename parent::PResultRec PResultRec;   
        typedef std::shared_ptr<SUBSET> PSubset;          
        Result_Tracker_Dupless(const IDXTYPE capacity_limit=0):_margin(0),_restrict_to_d(0),_capacity_limit(capacity_limit) {notify("Result_Tracker_Dupless constructor.");}
        Result_Tracker_Dupless(const Result_Tracker_Dupless& rtd):parent_impl(rtd), _margin(rtd._margin), _restrict_to_d(rtd._restrict_to_d), _capacity_limit(rtd._capacity_limit) {notify("Result_Tracker_Dupless copy-constructor.");} 
        virtual ~Result_Tracker_Dupless() {notify("Result_Tracker_Dupless destructor.");}

        virtual bool add(const RETURNTYPE value, const PSubset sub) override;
        virtual void clear() override {results.clear();}
        virtual std::size_t size() const override {return results.size();}
        virtual std::size_t size(const RETURNTYPE margin) const; 
        virtual bool get_first(PResultRec &r) override; 
        virtual bool get_next(PResultRec &r) override; 
        virtual void join(Result_Tracker<RETURNTYPE,SUBSET> &src) override; 
 
                void set_inclusion_margin(const RETURNTYPE margin=0) {_margin=margin;} 
                void set_inclusion_cardinality_restriction(const DIMTYPE d) {_restrict_to_d=d;} 
        
 
 
        virtual std::ostream& print(std::ostream& os) const override {os << "Result_Tracker_Dupless(limit=" << _capacity_limit << ", margin=" << _margin << ") size " << size(); return os;}
protected:
        std::list<ResultRec> results; 
        RETURNTYPE _margin; 
        DIMTYPE _restrict_to_d; 
        const IDXTYPE _capacity_limit; 
 
public:
        typedef typename std::list<ResultRec>::const_iterator CONSTRESULTSITER; 
        typedef typename std::list<ResultRec>::iterator RESULTSITER; 
        RESULTSITER begin() {return results.begin();} 
        RESULTSITER end() {return results.end();} 
 
private:
        RESULTSITER getiter; 
};
 
 
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
bool Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>::add(const RETURNTYPE value, const PSubset sub)
{
        assert(_margin>=0);
        assert(sub);
        if(_restrict_to_d>0 && sub->get_d()!=_restrict_to_d) return false;
        ResultRec rr; rr.value=value; rr.sub.reset(new SUBSET(*sub));
        bool added=false;
        if(!results.empty()) 
        {
                RESULTSITER iter=results.begin();
                if(_margin>0 && iter->value-_margin>value)
                { // the new result is outside the _margin given by the so-far best result value. Will not be added
                } 
                else // sort-in the new result, avoiding duplicates + possibly trim the tail of list that falls below _margin, if the new result becomes the new best
                {
                        if(value>iter->value) {
                                results.push_front(rr); added=true;
                                while(_margin>0 && results.back().value<value-_margin) results.pop_back(); // trim the tail of results that fell below _margin
                                while(_capacity_limit>0 && results.size()>_capacity_limit) results.pop_back(); // if there is a limit, trim the tail so that the list size remains restricted
                        } else { // sort the new result in, trimming not necessary
                                bool dupl=false;
                                while(iter!=results.end() && (value<iter->value || (value==iter->value && sub->get_d()>iter->sub->get_d()))) iter++; // skip all results that rank better: higher value, or equal value with smaller cardinality
                                while(!dupl && iter!=results.end() && (value==iter->value && sub->get_d()==iter->sub->get_d())) if(sub->equivalent(*(iter->sub))) dupl=true; else iter++; // within the block of equal value and equal cardinality: is an equivalent subset already held?
                                if(!dupl) {
                                        results.insert(iter,rr); added=true;
                                        while(_capacity_limit>0 && results.size()>_capacity_limit) results.pop_back(); // if there is a limit, trim the tail so that the list size remains restricted
                                }
                        }
                }
        } else {
            results.push_back(rr); added=true;
        }
        return added;
}
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
std::size_t Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>::size(const RETURNTYPE margin) const
{
        assert(margin>=0);
        if(results.empty()) return 0;
        CONSTRESULTSITER iter=results.begin();
        std::size_t count=0;
        RETURNTYPE threshold=iter->value-margin;
        while(iter!=results.end() && iter->value>=threshold) {count++; iter++;}
        return count;
}
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
bool Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>::get_first(PResultRec &r)
{
        if(results.empty()) return false;
        getiter=results.begin();
        r=&(*getiter);
        ++getiter;
        return true;
}
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
bool Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>::get_next(PResultRec &r)
{
        if(getiter==results.end()) return false;
        r=&(*getiter);
        ++getiter;
        return true;
}
 
template<class RETURNTYPE, typename IDXTYPE, typename DIMTYPE, class SUBSET>
void Result_Tracker_Dupless<RETURNTYPE,IDXTYPE,DIMTYPE,SUBSET>::join(Result_Tracker<RETURNTYPE,SUBSET> &src)
{
        if(results.empty()) // simple copy
        {
                PResultRec pr;
                for(bool got=src.get_first(pr);got==true;got=src.get_next(pr)) results.push_back(*pr);
        }
        else // join
        {
                RESULTSITER iter=begin();
                PResultRec pr;
                bool got=src.get_first(pr);
                while(got)
                {
                        assert(pr->sub);
                        while(iter!=end() && (iter->value>pr->value || (iter->value==pr->value && iter->sub->get_d()<pr->sub->get_d()))) iter++;
                        bool dupl=false;
                        while(!dupl && iter!=end() && (iter->value==pr->value && iter->sub->get_d()==pr->sub->get_d())) if(pr->sub->equivalent(*(iter->sub))) dupl=true; else iter++;
                        if(!dupl) results.insert(iter,*pr);
                        got=src.get_next(pr);
                }
        }
        if(!results.empty()) { // make joined list coherent with _capacity_limit and _margin restrictions
                RETURNTYPE threshold=results.front().value-_margin;
                while(_capacity_limit>0 && results.size()>_capacity_limit) results.pop_back();
                while(_margin>0 && results.back().value<threshold) results.pop_back();
        }
}
 
} // namespace
#endif // FSTRESULTTRACKERDUPLESS_H ///:~