LaplacianInferenceMethod.h

Go to the documentation of this file.

/*
 * This program 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.
 *
 * Written (W) 2013 Roman Votyakov
 * Copyright (C) 2012 Jacob Walker
 * Copyright (C) 2013 Roman Votyakov
 *
 * Code adapted from Gaussian Process Machine Learning Toolbox
 * http://www.gaussianprocess.org/gpml/code/matlab/doc/
 */

#ifndef CLAPLACIANINFERENCEMETHOD_H_
#define CLAPLACIANINFERENCEMETHOD_H_

#include <shogun/lib/config.h>

#ifdef HAVE_EIGEN3

#include <shogun/machine/gp/InferenceMethod.h>

namespace shogun
{

class CLaplacianInferenceMethod: public CInferenceMethod
{
public:
    CLaplacianInferenceMethod();

    CLaplacianInferenceMethod(CKernel* kernel, CFeatures* features,
            CMeanFunction* mean, CLabels* labels, CLikelihoodModel* model);

    virtual ~CLaplacianInferenceMethod();

    virtual EInferenceType get_inference_type() const { return INF_LAPLACIAN; }

    virtual const char* get_name() const { return "LaplacianInferenceMethod"; }

    virtual float64_t get_negative_log_marginal_likelihood();

    virtual SGVector<float64_t> get_alpha();

    virtual SGMatrix<float64_t> get_cholesky();

    virtual SGVector<float64_t> get_diagonal_vector();

    virtual SGVector<float64_t> get_posterior_mean();

    virtual SGMatrix<float64_t> get_posterior_covariance();

    virtual float64_t get_newton_tolerance() { return m_tolerance; }

    virtual void set_newton_tolerance(float64_t tol) { m_tolerance=tol; }

    virtual int32_t get_newton_iterations() { return m_iter; }

    virtual void set_newton_iterations(int32_t iter) { m_iter=iter; }

    virtual float64_t get_minimization_tolerance() { return m_opt_tolerance; }

    virtual void set_minimization_tolerance(float64_t tol) { m_opt_tolerance=tol; }

    virtual float64_t get_minimization_max() { return m_opt_max; }

    virtual void set_minimization_max(float64_t max) { m_opt_max=max; }

    virtual bool supports_regression() const
    {
        check_members();
        return m_model->supports_regression();
    }

    virtual bool supports_binary() const
    {
        check_members();
        return m_model->supports_binary();
    }

    virtual void update();

protected:
    virtual void update_alpha();

    virtual void update_chol();

    virtual void update_approx_cov();

    virtual void update_deriv();

    virtual SGVector<float64_t> get_derivative_wrt_inference_method(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_likelihood_model(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_kernel(
            const TParameter* param);

    virtual SGVector<float64_t> get_derivative_wrt_mean(
            const TParameter* param);

private:
    void init();

private:
    float64_t m_tolerance;

    index_t m_iter;

    float64_t m_opt_tolerance;

    float64_t m_opt_max;

    SGVector<float64_t> m_mu;

    SGMatrix<float64_t> m_Sigma;

    SGVector<float64_t> W;

    SGVector<float64_t> sW;

    SGVector<float64_t> dlp;

    SGVector<float64_t> d2lp;

    SGVector<float64_t> d3lp;

    SGVector<float64_t> m_dfhat;

    SGMatrix<float64_t> m_Z;

    SGVector<float64_t> m_g;
};
}
#endif /* HAVE_EIGEN3 */
#endif /* CLAPLACIANINFERENCEMETHOD_H_ */