recordCounter.h

Go to the documentation of this file.

/****************************************************************************
 *   This file is part of the aGrUM/pyAgrum library.                        *
 *                                                                          *
 *   Copyright (c) 2005-2025 by                                             *
 *       - Pierre-Henri WUILLEMIN(_at_LIP6)                                 *
 *       - Christophe GONZALES(_at_AMU)                                     *
 *                                                                          *
 *   The aGrUM/pyAgrum library is free software; you can redistribute it    *
 *   and/or modify it under the terms of either :                           *
 *                                                                          *
 *    - the GNU Lesser General Public License as published by               *
 *      the Free Software Foundation, either version 3 of the License,      *
 *      or (at your option) any later version,                              *
 *    - the MIT license (MIT),                                              *
 *    - or both in dual license, as here.                                   *
 *                                                                          *
 *   (see https://agrum.gitlab.io/articles/dual-licenses-lgplv3mit.html)    *
 *                                                                          *
 *   This aGrUM/pyAgrum library is distributed in the hope that it will be  *
 *   useful, but WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,          *
 *   INCLUDING BUT NOT LIMITED TO THE WARRANTIES MERCHANTABILITY or FITNESS *
 *   FOR A PARTICULAR PURPOSE  AND NONINFRINGEMENT. IN NO EVENT SHALL THE   *
 *   AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER *
 *   LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,        *
 *   ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR  *
 *   OTHER DEALINGS IN THE SOFTWARE.                                        *
 *                                                                          *
 *   See LICENCES for more details.                                         *
 *                                                                          *
 *   SPDX-FileCopyrightText: Copyright 2005-2025                            *
 *       - Pierre-Henri WUILLEMIN(_at_LIP6)                                 *
 *       - Christophe GONZALES(_at_AMU)                                     *
 *   SPDX-License-Identifier: LGPL-3.0-or-later OR MIT                      *
 *                                                                          *
 *   Contact  : info_at_agrum_dot_org                                       *
 *   homepage : http://agrum.gitlab.io                                      *
 *   gitlab   : https://gitlab.com/agrumery/agrum                           *
 *                                                                          *
 ****************************************************************************/
 

#ifndef GUM_LEARNING_RECORD_COUNTER_H
#define GUM_LEARNING_RECORD_COUNTER_H
 
#include <sstream>
#include <string>
#include <utility>
#include <vector>
 
#include <agrum/agrum.h>
 
#include <agrum/base/core/threads/threadData.h>
#include <agrum/base/database/DBRowGeneratorParser.h>
#include <agrum/base/stattests/idCondSet.h>
 
namespace gum {
 
  namespace learning {

    class RecordCounter: public ThreadNumberManager {
      public:
      // ##########################################################################
      // ##########################################################################


      RecordCounter(const DBRowGeneratorParser&                                 parser,
                    const std::vector< std::pair< std::size_t, std::size_t > >& ranges,
                    const Bijection< NodeId, std::size_t >&                     nodeId2columns
                    = Bijection< NodeId, std::size_t >());


      RecordCounter(const DBRowGeneratorParser&             parser,
                    const Bijection< NodeId, std::size_t >& nodeId2columns
                    = Bijection< NodeId, std::size_t >());

      RecordCounter(const RecordCounter& from);

      RecordCounter(RecordCounter&& from);

      virtual RecordCounter* clone() const;

      virtual ~RecordCounter();

 
 
      // ##########################################################################
      // ##########################################################################


      RecordCounter& operator=(const RecordCounter& from);

      RecordCounter& operator=(RecordCounter&& from);

 
 
      // ##########################################################################
      // ##########################################################################


      void clear();


      virtual void setNumberOfThreads(Size nb);

      void setMinNbRowsPerThread(const std::size_t nb) const;

      std::size_t minNbRowsPerThread() const;


      const std::vector< double >& counts(const IdCondSet& ids,
                                          const bool       check_discrete_vars = false);


      void setRanges(const std::vector< std::pair< std::size_t, std::size_t > >& new_ranges);

      void clearRanges();

      const std::vector< std::pair< std::size_t, std::size_t > >& ranges() const;


      template < typename GUM_SCALAR >
      void setBayesNet(const BayesNet< GUM_SCALAR >& new_bn);


      const Bijection< NodeId, std::size_t >& nodeId2Columns() const;

      const DatabaseTable& database() const;

 
#ifndef DOXYGEN_SHOULD_SKIP_THIS
 
      private:
      // the parsers used by the threads
      std::vector< ThreadData< DBRowGeneratorParser > > _parsers_;
 
      // the set of ranges of the database's rows indices over which the user
      // wishes to perform the counting
      std::vector< std::pair< std::size_t, std::size_t > > _ranges_;
 
      // the ranges actually used by the threads: there is a hopefully clever
      // algorithm that split the rows ranges into another set of ranges that
      // are assigned to the threads. For instance, if the database has 1000
      // rows and there are 10 threads, each one will be passed a set of 100
      // rows. These sets are precisely what are stored in the field below
      mutable std::vector< std::pair< std::size_t, std::size_t > > _thread_ranges_;
 
      // the mapping from the NodeIds of the variables to the indices of the
      // columns in the database
      Bijection< NodeId, std::size_t > _nodeId2columns_;
 
      // the last database-parsed counting
      std::vector< double > _last_DB_counting_;
 
      // the ids of the nodes for the last database-parsed counting
      IdCondSet _last_DB_ids_;
 
      // the last counting deduced from  _last_DB_counting_
      std::vector< double > _last_nonDB_counting_;
 
      // the ids of the nodes of last counting deduced from  _last_DB_counting_
      IdCondSet _last_nonDB_ids_;
 
      // the min number of rows that a thread should process in a
      // multithreading context
      mutable std::size_t _min_nb_rows_per_thread_{512};
 
      // returns a mapping from the nodes ids to the columns of the database
      // for a given sequence of ids. This is especially convenient when
      //  _nodeId2columns_ is empty (which means that there is an identity mapping)
      HashTable< NodeId, std::size_t > _getNodeIds2Columns_(const IdCondSet& ids) const;

      std::vector< double >& _extractFromCountings_(const IdCondSet&             subset_ids,
                                                    const IdCondSet&             superset_ids,
                                                    const std::vector< double >& superset_vect);

      std::vector< double >& _countFromDatabase_(const IdCondSet& ids);


      void _checkRanges_(
          const std::vector< std::pair< std::size_t, std::size_t > >& new_ranges) const;


      void _checkDiscreteVariables_(const IdCondSet& ids) const;


      void _raiseCheckException_(const std::vector< std::string >& bad_vars) const;

      void _dispatchRangesToThreads_();
 
#endif /* DOXYGEN_SHOULD_SKIP_THIS */
    };
 
  } /* namespace learning */
 
} /* namespace gum */

#include <agrum/base/stattests/recordCounter_tpl.h>

#ifndef GUM_NO_INLINE
#  include <agrum/base/stattests/recordCounter_inl.h>
#endif /* GUM_NO_INLINE */
 
#endif /* GUM_LEARNING_RECORD_COUNTER_H */