SigmoidFit.h

Go to the documentation of this file.

//===========================================================================
/*!
 * 
 *
 * \brief       Optimization of the SigmoidModel according to Platt, 1999.
 * 
 * 
 *
 * \author      T. Glasmachers, O.Krause
 * \date        2010
 *
 *
 * \par Copyright 1995-2017 Shark Development Team
 * 
 * <BR><HR>
 * This file is part of Shark.
 * <http://shark-ml.org/>
 * 
 * Shark is free software: you can redistribute it and/or modify
 * it under the terms of 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.
 * 
 * Shark 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 Lesser General Public License for more details.
 * 
 * You should have received a copy of the GNU Lesser General Public License
 * along with Shark.  If not, see <http://www.gnu.org/licenses/>.
 *
 */
//===========================================================================


#ifndef SHARK_ALGORITHMS_TRAINERS_SIGMOIDFIT_H
#define SHARK_ALGORITHMS_TRAINERS_SIGMOIDFIT_H

#include <shark/Core/DLLSupport.h>
#include <shark/Algorithms/Trainers//AbstractTrainer.h>
#include <shark/Models/SigmoidModel.h>

namespace shark{
    
    
//! \brief Optimizes the parameters of a sigmoid to fit a validation dataset via backpropagation on the negative log-likelihood.
//! 
//! \par
//! The SigmoidFitPlatt class implements a non-iterative optimizer,
//! despite the optimization task and optimizer being iterative in nature.
//! This class simply relies on a user-definable number of Rprop optimization steps
//! to adapt the sigmoid parameters.
//! 
class SigmoidFitRpropNLL: public AbstractTrainer<SigmoidModel, unsigned int>
{
public:
    SHARK_EXPORT_SYMBOL SigmoidFitRpropNLL( unsigned int iters = 100 );

    /// \brief From INameable: return the class name.
    std::string name() const
    { return "SigmoidFitRpropNLL"; }

    SHARK_EXPORT_SYMBOL void train(SigmoidModel& model, LabeledData<RealVector, unsigned int> const& dataset);
    
private:
    unsigned int m_iterations;
};


//!
//! \brief Optimizes the parameters of a sigmoid to fit a validation dataset via Platt's method.
//!
//! \par
//! The SigmoidFitPlatt class implements a non-iterative optimizer,
//! despite the optimization task and optimizer being iterative in nature.
//! The algorithm corresponds to the one suggested by John Platt in 
//! 1999, and is almost literally taken from <br> <i>Probabilistic Outputs for
//! Support Vector Machines and Comparisons to Regularized Likelihood Methods,
//! Advances in Large Margin Classifiers, pp. 61-74,
//! MIT Press, (1999).</i><br>
//! The full paper can be downloaded from<br>
//! <i>http://www.research.microsoft.com/~jplatt</i><br>
//! --- pseudo-code is given in the paper.
//!
class SigmoidFitPlatt: public AbstractTrainer<SigmoidModel, unsigned int>
{
public:
    SHARK_EXPORT_SYMBOL void train(SigmoidModel& model, LabeledData<RealVector, unsigned int> const& dataset);

    /// \brief From INameable: return the class name.
    std::string name() const
    { return "SigmoidFitPlatt"; }
};


}
#endif