GibbsSampling_tpl.h

Go to the documentation of this file.

/****************************************************************************
 *   This file is part of the aGrUM/pyAgrum library.                        *
 *                                                                          *
 *   Copyright (c) 2005-2026 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 2005-2026                            *
 *       - 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                           *
 *                                                                          *
 ****************************************************************************/
 
#pragma once
 

 
#include <agrum/BN/inference/GibbsSampling.h>
 
#define GIBBS_SAMPLING_DEFAULT_EPSILON          std::exp(-1.6)
#define GIBBS_SAMPLING_DEFAULT_MIN_EPSILON_RATE std::exp(-5)
#define GIBBS_SAMPLING_DEFAULT_BURNIN           300
 
#define GIBBS_SAMPLING_POURCENT_DRAWN_SAMPLE 50   // percent drawn
#define GIBBS_SAMPLING_DRAWN_AT_RANDOM       true
 
namespace gum {

  template < GUM_Numeric GUM_SCALAR >
  GibbsSampling< GUM_SCALAR >::GibbsSampling(const IBayesNet< GUM_SCALAR >* bn) :
      SamplingInference< GUM_SCALAR >(bn),
      GibbsOperator< GUM_SCALAR >(*bn,
                                  &this->hardEvidence(),
                                  1 + (bn->size() * GIBBS_SAMPLING_POURCENT_DRAWN_SAMPLE / 100),
                                  GIBBS_SAMPLING_DRAWN_AT_RANDOM) {
    GUM_CONSTRUCTOR(GibbsSampling);
 
    this->setEpsilon(GIBBS_SAMPLING_DEFAULT_EPSILON);
    this->setMinEpsilonRate(GIBBS_SAMPLING_DEFAULT_MIN_EPSILON_RATE);
    this->setBurnIn(GIBBS_SAMPLING_DEFAULT_BURNIN);
  }

  template < GUM_Numeric GUM_SCALAR >
  GibbsSampling< GUM_SCALAR >::~GibbsSampling() {
    GUM_DESTRUCTOR(GibbsSampling);
  }
 
  template < GUM_Numeric GUM_SCALAR >
  Instantiation GibbsSampling< GUM_SCALAR >::monteCarloSample_() {
    return GibbsOperator< GUM_SCALAR >::monteCarloSample();
  }
 
  template < GUM_Numeric GUM_SCALAR >
  Instantiation GibbsSampling< GUM_SCALAR >::burnIn_() {
    // we initialize the nodes with hard evidence
    // hypothesis : burnIn_ is called at the beginning of makeInference
    this->updateSamplingNodes_();
 
    gum::Instantiation Ip;
    if (this->burnIn() == 0) return Ip;
 
    GUM_SCALAR w = 1.0f;
    Ip           = monteCarloSample_();
    for (Size i = 1; i < this->burnIn(); i++)
      Ip = this->draw_(&w, Ip);
 
    return Ip;
  }

 
  template < GUM_Numeric GUM_SCALAR >
  Instantiation GibbsSampling< GUM_SCALAR >::draw_(GUM_SCALAR* w, Instantiation prev) {
    *w = 1.0;
    return GibbsOperator< GUM_SCALAR >::nextSample(prev);
  }
 
  template < GUM_Numeric GUM_SCALAR >
  void GibbsSampling< GUM_SCALAR >::setBurnIn(Size b) {
    this->burn_in_ = b;
  }
 
  template < GUM_Numeric GUM_SCALAR >
  Size GibbsSampling< GUM_SCALAR >::burnIn() const {
    return this->burn_in_;
  }
 
}   // namespace gum