convolution.hpp

Go to the documentation of this file.

#ifndef MLPACK_METHODS_ANN_LAYER_CONVOLUTION_HPP
#define MLPACK_METHODS_ANN_LAYER_CONVOLUTION_HPP

#include <mlpack/prereqs.hpp>

#include <mlpack/methods/ann/convolution_rules/border_modes.hpp>
#include <mlpack/methods/ann/convolution_rules/naive_convolution.hpp>
#include <mlpack/methods/ann/convolution_rules/fft_convolution.hpp>
#include <mlpack/methods/ann/convolution_rules/svd_convolution.hpp>

#include "layer_types.hpp"
#include "padding.hpp"

namespace mlpack {
namespace ann  {

template <
    typename ForwardConvolutionRule = NaiveConvolution<ValidConvolution>,
    typename BackwardConvolutionRule = NaiveConvolution<FullConvolution>,
    typename GradientConvolutionRule = NaiveConvolution<ValidConvolution>,
    typename InputDataType = arma::mat,
    typename OutputDataType = arma::mat
>
class Convolution
{
 public:
  Convolution();

  Convolution(const size_t inSize,
              const size_t outSize,
              const size_t kernelWidth,
              const size_t kernelHeight,
              const size_t strideWidth = 1,
              const size_t strideHeight = 1,
              const size_t padW = 0,
              const size_t padH = 0,
              const size_t inputWidth = 0,
              const size_t inputHeight = 0,
              const std::string& paddingType = "None");

  Convolution(const size_t inSize,
              const size_t outSize,
              const size_t kernelWidth,
              const size_t kernelHeight,
              const size_t strideWidth,
              const size_t strideHeight,
              const std::tuple<size_t, size_t>& padW,
              const std::tuple<size_t, size_t>& padH,
              const size_t inputWidth = 0,
              const size_t inputHeight = 0,
              const std::string& paddingType = "None");

  /*
   * Set the weight and bias term.
   */
  void Reset();

  template<typename eT>
  void Forward(const arma::Mat<eT>&& input, arma::Mat<eT>&& output);

  template<typename eT>
  void Backward(const arma::Mat<eT>&& /* input */,
                arma::Mat<eT>&& gy,
                arma::Mat<eT>&& g);

  /*
   * Calculate the gradient using the output delta and the input activation.
   *
   * @param input The input parameter used for calculating the gradient.
   * @param error The calculated error.
   * @param gradient The calculated gradient.
   */
  template<typename eT>
  void Gradient(const arma::Mat<eT>&& /* input */,
                arma::Mat<eT>&& error,
                arma::Mat<eT>&& gradient);

  const OutputDataType& Parameters() const { return weights; }
  OutputDataType& Parameters() { return weights; }

  const InputDataType& InputParameter() const { return inputParameter; }
  InputDataType& InputParameter() { return inputParameter; }

  const OutputDataType& OutputParameter() const { return outputParameter; }
  OutputDataType& OutputParameter() { return outputParameter; }

  const OutputDataType& Delta() const { return delta; }
  OutputDataType& Delta() { return delta; }

  const OutputDataType& Gradient() const { return gradient; }
  OutputDataType& Gradient() { return gradient; }

  const size_t& InputWidth() const { return inputWidth; }
  size_t& InputWidth() { return inputWidth; }

  const size_t& InputHeight() const { return inputHeight; }
  size_t& InputHeight() { return inputHeight; }

  const size_t& OutputWidth() const { return outputWidth; }
  size_t& OutputWidth() { return outputWidth; }

  const size_t& OutputHeight() const { return outputHeight; }
  size_t& OutputHeight() { return outputHeight; }

  size_t InputSize() const { return inSize; }

  size_t OutputSize() const { return outSize; }

  size_t KernelWidth() const { return kernelWidth; }
  size_t& KernelWidth() { return kernelWidth; }

  size_t KernelHeight() const { return kernelHeight; }
  size_t& KernelHeight() { return kernelHeight; }

  size_t StrideWidth() const { return strideWidth; }
  size_t& StrideWidth() { return strideWidth; }

  size_t StrideHeight() const { return strideHeight; }
  size_t& StrideHeight() { return strideHeight; }

  size_t PadHTop() const { return padHTop; }
  size_t& PadHTop() { return padHTop; }

  size_t PadHBottom() const { return padHBottom; }
  size_t& PadHBottom() { return padHBottom; }

  size_t PadWLeft() const { return padWLeft; }
  size_t& PadWLeft() { return padWLeft; }

  size_t PadWRight() const { return padWRight; }
  size_t& PadWRight() { return padWRight; }

  arma::mat& Bias() { return bias; }

  template<typename Archive>
  void serialize(Archive& ar, const unsigned int /* version */);

 private:
  /*
   * Return the convolution output size.
   *
   * @param size The size of the input (row or column).
   * @param k The size of the filter (width or height).
   * @param s The stride size (x or y direction).
   * @param pSideOne The size of the padding (width or height) on one side.
   * @param pSideTwo The size of the padding (width or height) on another side.
   * @return The convolution output size.
   */
  size_t ConvOutSize(const size_t size,
                     const size_t k,
                     const size_t s,
                     const size_t pSideOne,
                     const size_t pSideTwo)
  {
    return std::floor(size + pSideOne + pSideTwo - k) / s + 1;
  }

  /*
   * Function to assign padding such that output size is same as input size.
   */
  void InitializeSamePadding();

  /*
   * Rotates a 3rd-order tensor counterclockwise by 180 degrees.
   *
   * @param input The input data to be rotated.
   * @param output The rotated output.
   */
  template<typename eT>
  void Rotate180(const arma::Cube<eT>& input, arma::Cube<eT>& output)
  {
    output = arma::Cube<eT>(input.n_rows, input.n_cols, input.n_slices);

    // * left-right flip, up-down flip */
    for (size_t s = 0; s < output.n_slices; s++)
      output.slice(s) = arma::fliplr(arma::flipud(input.slice(s)));
  }

  /*
   * Rotates a dense matrix counterclockwise by 180 degrees.
   *
   * @param input The input data to be rotated.
   * @param output The rotated output.
   */
  template<typename eT>
  void Rotate180(const arma::Mat<eT>& input, arma::Mat<eT>& output)
  {
    // * left-right flip, up-down flip */
    output = arma::fliplr(arma::flipud(input));
  }

  size_t inSize;

  size_t outSize;

  size_t batchSize;

  size_t kernelWidth;

  size_t kernelHeight;

  size_t strideWidth;

  size_t strideHeight;

  size_t padWLeft;

  size_t padWRight;

  size_t padHBottom;

  size_t padHTop;

  OutputDataType weights;

  arma::cube weight;

  arma::mat bias;

  size_t inputWidth;

  size_t inputHeight;

  size_t outputWidth;

  size_t outputHeight;

  arma::cube outputTemp;

  arma::cube inputTemp;

  arma::cube inputPaddedTemp;

  arma::cube gTemp;

  arma::cube gradientTemp;

  ann::Padding<> padding;

  OutputDataType delta;

  OutputDataType gradient;

  InputDataType inputParameter;

  OutputDataType outputParameter;
}; // class Convolution

} // namespace ann
} // namespace mlpack

namespace boost {
namespace serialization {

template<
    typename ForwardConvolutionRule,
    typename BackwardConvolutionRule,
    typename GradientConvolutionRule,
    typename InputDataType,
    typename OutputDataType
>
struct version<
    mlpack::ann::Convolution<ForwardConvolutionRule, BackwardConvolutionRule,
        GradientConvolutionRule, InputDataType, OutputDataType> >
{
  BOOST_STATIC_CONSTANT(int, value = 1);
};

} // namespace serialization
} // namespace boost

// Include implementation.
#include "convolution_impl.hpp"

#endif