MeanEigen3.h

Go to the documentation of this file.

/*
 * Copyright (c) The Shogun Machine Learning Toolbox
 * Written (w) 2016 Pan Deng
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 * 1. Redistributions of source code must retain the above copyright notice, this
 *    list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 * The views and conclusions contained in the software and documentation are those
 * of the authors and should not be interpreted as representing official policies,
 * either expressed or implied, of the Shogun Development Team.
 */

#ifndef MEAN_EIGEN_H_
#define MEAN_EIGEN_H_

#include <shogun/lib/config.h>
#include <shogun/lib/SGVector.h>
#include <shogun/lib/SGMatrix.h>
#include <shogun/mathematics/linalg/internal/implementation/Sum.h>

#include <shogun/mathematics/eigen3.h>
#include <numeric>

namespace shogun
{

namespace linalg
{

namespace implementation
{

#ifndef SWIG // SWIG should skip this part

template <typename inputType>
struct int2float
{
    using floatType = inputType;
};

template <>
struct int2float<int32_t>
{
    using floatType = float64_t;
};

template <>
struct int2float<int64_t>
{
    using floatType = float64_t;
};
#endif

template <enum Backend, class Matrix>
struct mean
{
    typedef typename Matrix::Scalar T;

    typedef typename int2float<T>::floatType ReturnType;

    static ReturnType compute(Matrix a);

    static ReturnType compute(Matrix a, bool no_diag=false);
};

template <enum Backend, class Matrix>
struct rowwise_mean
{
    typedef typename Matrix::Scalar T;

    typedef typename int2float<T>::floatType ReturnDataType;

    static SGVector<ReturnDataType> compute(SGMatrix<T> m, bool no_diag);

    static void compute(SGMatrix<T> mat, SGVector<ReturnDataType> result, bool no_diag);
};

template <class Matrix>
struct mean<Backend::EIGEN3, Matrix>
{
    typedef typename Matrix::Scalar T;

    typedef typename int2float<T>::floatType ReturnType;

    static ReturnType compute(SGVector<T> vec)
    {
        REQUIRE(vec.vlen > 0, "Vector cannot be empty!\n");
        return (vector_sum<Backend::EIGEN3, SGVector<T> >::compute(vec)
            / ReturnType(vec.vlen));
    }

        static ReturnType compute(SGMatrix<T> mat, bool no_diag=false)
    {
        REQUIRE(mat.num_rows > 0, "Matrix row number cannot be zero!\n");
        if (no_diag)
        {
            if (mat.num_rows > mat.num_cols)
            {
                return (sum<Backend::EIGEN3, SGMatrix<T> >::compute(mat, no_diag)
                    / ReturnType(mat.num_rows * mat.num_cols - mat.num_cols));
            }
            else
            {
                return (sum<Backend::EIGEN3, SGMatrix<T> >::compute(mat, no_diag)
                    / ReturnType(mat.num_rows * mat.num_cols - mat.num_rows));
            }
        }
        else
        {
            return (sum<Backend::EIGEN3, SGMatrix<T> >::compute(mat, no_diag)
                / ReturnType(mat.num_rows * mat.num_cols));
        }
    }
};


template <class Matrix>
struct rowwise_mean<Backend::EIGEN3, Matrix>
{
    typedef typename Matrix::Scalar T;

    typedef typename int2float<T>::floatType ReturnDataType;

    typedef Eigen::Matrix<T,Eigen::Dynamic, Eigen::Dynamic> MatrixXt;

    static SGVector<ReturnDataType> compute(SGMatrix<T> m, bool no_diag)
    {
        SGVector<ReturnDataType> result(m.num_rows);
        compute(m, result, no_diag);
        return result;
    }

    static void compute(SGMatrix<T> mat, SGVector<ReturnDataType> result, bool no_diag)
    {
        REQUIRE(mat.num_cols > 0, "Matrix column number cannot be zero!\n");

        Eigen::Map<MatrixXt> m = mat;
        SGVector<T> temp(m.rows());
        temp = rowwise_sum<Backend::EIGEN3, SGMatrix<T> >::compute(mat, no_diag);

        if (!no_diag)
            for (index_t i = 0; i < m.rows(); ++i)
                result[i] = temp[i] / ReturnDataType(m.cols());
        else if (m.rows() <= m.cols())
            for (index_t i = 0; i < m.rows(); ++i)
                result[i] = temp[i] / ReturnDataType(m.cols() - 1);
        else
        {
            for (index_t i = 0; i < m.cols(); ++i)
                result[i] = temp[i] / ReturnDataType(m.cols() - 1);
            for (index_t i = m.cols(); i < m.rows(); ++i)
                result[i] = temp[i] / ReturnDataType(m.cols());
        }
    }

};

}

}

}

#endif