ConvolutionalFeatureMap.h

Go to the documentation of this file.

/*
 * Copyright (c) 2014, Shogun Toolbox Foundation
 * 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.
 *
 * 3. Neither the name of the copyright holder nor the names of its
 * contributors may be used to endorse or promote products derived from this
 * software without specific prior written permission.
 *
 * 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 HOLDER 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.
 *
 * Written (W) 2014 Khaled Nasr
 */

#ifndef __CONVOLUTIONALFEATUREMAP_H__
#define __CONVOLUTIONALFEATUREMAP_H__

#include <shogun/lib/common.h>
#include <shogun/neuralnets/NeuralLayer.h>

namespace shogun
{

enum EConvMapActivationFunction
{
    CMAF_IDENTITY = 0,

    CMAF_LOGISTIC = 1,

    CMAF_RECTIFIED_LINEAR = 2
};

template <class T> class SGVector;
template <class T> class SGMatrix;
class CDynamicObjectArray;

class CConvolutionalFeatureMap
{
public:
    CConvolutionalFeatureMap(int32_t input_width, int32_t input_height,
            int32_t radius_x, int32_t radius_y,
            int32_t stride_x=1, int32_t stride_y=1,
            int32_t index=0,
            EConvMapActivationFunction function = CMAF_IDENTITY,
            ENLAutoencoderPosition autoencoder_position = NLAP_NONE);

    void compute_activations(SGVector<float64_t> parameters,
            CDynamicObjectArray* layers,
            SGVector<int32_t> input_indices,
            SGMatrix<float64_t> activations);

    void compute_gradients(SGVector<float64_t> parameters,
            SGMatrix<float64_t> activations,
            SGMatrix<float64_t> activation_gradients,
            CDynamicObjectArray* layers,
            SGVector<int32_t> input_indices,
            SGVector<float64_t> parameter_gradients);

    void pool_activations(SGMatrix<float64_t> activations,
            int32_t pooling_width,
            int32_t pooling_height,
            SGMatrix<float64_t> pooled_activations,
            SGMatrix<float64_t> max_indices);

protected:
    void convolve(SGMatrix<float64_t> inputs,
            SGMatrix<float64_t> weights,
            SGMatrix<float64_t> outputs,
            bool flip,
            bool reset_output,
            int32_t inputs_row_offset,
            int32_t outputs_row_offset);

    void compute_weight_gradients(SGMatrix<float64_t> inputs,
            SGMatrix<float64_t> local_gradients,
            SGMatrix<float64_t> weight_gradients,
            int32_t inputs_row_offset,
            int32_t local_gradients_row_offset);

protected:
    int32_t m_input_width;

    int32_t m_input_height;

    int32_t m_radius_x;

    int32_t m_radius_y;

    int32_t m_stride_x;

    int32_t m_stride_y;

    int32_t m_index;

    EConvMapActivationFunction m_activation_function;

    int32_t m_output_width;

    int32_t m_output_height;

    int32_t m_input_num_neurons;

    int32_t m_output_num_neurons;

    int32_t m_row_offset;

    int32_t m_filter_width;

    int32_t m_filter_height;

    ENLAutoencoderPosition m_autoencoder_position;
};

}
#endif