RandomNumberGenerator.hpp

Go to the documentation of this file.

/*
 
  This file is part of OpenFLUID software
  Copyright(c) 2007, INRA - Montpellier SupAgro
 
 
 == GNU General Public License Usage ==
 
  OpenFLUID 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.
 
  OpenFLUID 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 OpenFLUID. If not, see <http://www.gnu.org/licenses/>.
 
 
 == Other Usage ==
 
  Other Usage means a use of OpenFLUID that is inconsistent with the GPL
  license, and requires a written agreement between You and INRA.
  Licensees for Other Usage of OpenFLUID may use this file in accordance
  with the terms contained in the written agreement between You and INRA.
  
*/
 
 
/**
  @file RandomNumberGenerator.hpp
 
  @author Dorian GERARDIN <dorian.gerardin@inrae.fr>
  @author Alain DUPUY <contact@openfluid-project.org>
  @author David CREVOISIER <david.crevoisier@inrae.fr>
 */
 
 
#ifndef __OPENFLUID_TOOLS_RANDOMNUMBERGENERATOR_HPP__
#define __OPENFLUID_TOOLS_RANDOMNUMBERGENERATOR_HPP__
 
 
#include <string>
#include <iostream>
#include <random>
#include <vector>
 
#include <openfluid/ware/TypeDefs.hpp>
 
 
namespace openfluid { namespace tools {
 
 
/**
  @brief Class for management of random. It includes distribution utility functions.
 
  Sources:
  @li https://en.wikipedia.org/wiki/Random_number_generation
  @li https://en.wikipedia.org/wiki/Mersenne_Twister
 
 
  <I>Example : declaring an instance</I>
  @snippet wares/RNG.cpp rng_decl
 
  <I>Example : initializing with a custom seed</I>
  @snippet wares/RNG.cpp rng_initialize_seed
 
  <I>Example : initializing with a random seed</I>
  @snippet wares/RNG.cpp rng_initialize_random_seed
 
  <I>Example : getting a random generated number using normal distribution</I>
  @snippet wares/RNG.cpp rng_log
 
  <I>Example : getting randomized value using Bernoulli distribution</I>
  @snippet wares/RNG.cpp rng_bernoulli
*/
class OPENFLUID_API RandomNumberGenerator
{
  public:
 
    RandomNumberGenerator();
 
    ~RandomNumberGenerator() {}
 
    /**
      Set a seed for the generator
      @param[in] CustomSeed The seed to set. If negative, random seed will be used
    */
    void init(long int CustomSeed = -1);
 
    /**
      Returns the seed
      @return an uint64_t
    */
    uint64_t getSelectedSeed() const;
 
    /**
    * @brief Log normal distribution
    *
    * @tparam T type of value for distribution
    * @param N number of values to generate with distribution
    * @param Mean mean for log normal distribution
    * @param Sd standard deviation for log normal distribution
    * @return std::vector<T> vector with N randomized values using log normal distribution
    */
    template <typename T>
    std::vector<T> rlnorm(size_t N, double Mean, double Sd)
    {
      T Maxlnorm = std::exp(Mean - Sd * Sd);
      std::vector<T> vec(N, Maxlnorm);
      std::lognormal_distribution<T> Distribution(Mean, Sd);
 
      for (auto& v : vec)
      {
        v = Distribution(getGenerator());
      }
 
      return vec;
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Normal distribution
    *
    * @tparam T type of value for distribution
    * @param N number of values to generate with distribution
    * @param Mean mean for normal distribution
    * @param Sd standard deviation for normal distribution
    * @return std::vector<T> vector with N randomized values using normal distribution
    */
    template <typename T>
    std::vector<T> rnorm(size_t N, double Mean, double Sd)
    {
      std::vector<T> vec(N, Mean + Sd);
      std::normal_distribution<T> Distribution(Mean, Sd);
      for (auto& v : vec)
      {
        v = Distribution(getGenerator());
      }
 
      return vec;
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Uniform distribution
    *
    * @tparam T type of value for distribution
    * @param N number of values to generate with distribution
    * @param V1 min value for distribution
    * @param V2 max value for distribution
    * @return std::vector<T> vector with N randomized values using real uniform distribution
    */
    template <typename T>
    std::vector<T> runif(size_t N, T V1, T V2)
    {
      std::vector<T> vec(N, 0.5 * V1 + 0.5 * V2);
      if (V2 > V1)
      {
        std::uniform_real_distribution<T> Distribution(V1, V2);
        for (auto& v : vec)
        {
          v = Distribution(getGenerator());
        }
      }
      else 
      {
        throw openfluid::base::FrameworkException(OPENFLUID_CODE_LOCATION,"Max value < Min Value");
      }
 
      return vec;
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Uniform distribution
    *
    * @tparam T type of value for distribution
    * @param N number of values to generate with distribution
    * @param V1 min value for distribution
    * @param V2 max value for distribution
    * @return std::vector<T> vector with N randomized values using integer uniform distribution
    */
    template <typename T>
    std::vector<T> irunif(size_t N, T V1, T V2)
    {
      std::vector<T> vec(N, 0.5 * V1 + 0.5 * V2);
      if (V2 > V1)
      {
        std::uniform_int_distribution<T> Distribution(V1, V2);
        for (auto& v : vec)
        {
          v = Distribution(getGenerator());
        }
      }
      else 
      {
        throw openfluid::base::FrameworkException(OPENFLUID_CODE_LOCATION,"Max value < Min Value");
      }
 
      return vec;
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Log normal distribution
    *
    * @tparam T type of value for distribution
    * @param Mean mean for log normal distribution
    * @param Sd standard deviation for log normal distribution
    * @return Randomized value using log normal distribution
    */
    template <typename T>
    T rlnorm(double Mean, double Sd)
    {
      std::lognormal_distribution<T> Distribution(Mean, Sd);
      return Distribution(getGenerator());
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Normal distribution
    *
    * @tparam T type of value for distribution
    * @param Mean mean for normal distribution
    * @param Sd standard deviation for normal distribution
    * @return Randomized value using normal distribution
    */
    template <typename T>
    T rnorm(double Mean, double Sd)
    {
      std::normal_distribution<T> Distribution(Mean, Sd);
      return Distribution(getGenerator());
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Real uniform distribution
    *
    * @tparam T type of value for distribution
    * @param V1 min value for distribution
    * @param V2 max value for distribution
    * @return Randomized value using real uniform distribution
    */
    template <typename T>
    T runif(T V1, T V2)
    {
      if (V2 > V1)
      {
        std::uniform_real_distribution<T> Distribution(V1, V2);
        return Distribution(getGenerator());
      }
      else 
      {
        throw openfluid::base::FrameworkException(OPENFLUID_CODE_LOCATION,"Max value < Min Value");
      }
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Integer uniform distribution
    *
    * @tparam T type of value for distribution
    * @param V1 min value for distribution
    * @param V2 max value for distribution
    * @return Randomized value using integer uniform distribution
    */
    template <typename T>
    T irunif(T V1, T V2)
    {
      if (V2 > V1)
      {
        std::uniform_int_distribution<T> Distribution(V1, V2);
        return Distribution(getGenerator());
      }
      else 
      {
        throw openfluid::base::FrameworkException(OPENFLUID_CODE_LOCATION, "Max value < Min Value");
      }
    }
 
 
    // =====================================================================
    // =====================================================================
 
 
    /**
    * @brief Bernoulli distribution
    *
    * @param Probability probability of success
    * @return Randomized value using bernoulli distribution
    */
    bool bernoulli(double Probability)
    {
      std::bernoulli_distribution Distribution(Probability);
      return Distribution(getGenerator());
    }
 
  
  protected:
 
    /**
      The Mersenne Twister generator 
    */
    std::mt19937_64 m_Generator;
 
    uint64_t m_SelectedSeed;
 
    /**
      A boolean to check if we need to display the seed in logs 
    */
    bool m_DisplaySeed;
 
    /**
      A string to display the seed in logs as "Seed: ${m_SelectedSeed}"
    */
    const std::string m_DisplayNameInfo;
 
    std::mt19937_64& getGenerator();
 
    virtual void displaySeedInfo();
 
};
 
 
} } // namespaces
 
#endif /* __OPENFLUID_TOOLS_RANDOMNUMBERGENERATOR_HPP__ */