triangular_kernel.hpp

Go to the documentation of this file.

#ifndef MLPACK_CORE_KERNELS_TRIANGULAR_KERNEL_HPP
#define MLPACK_CORE_KERNELS_TRIANGULAR_KERNEL_HPP

#include <mlpack/prereqs.hpp>
#include <mlpack/core/metrics/lmetric.hpp>

namespace mlpack {
namespace kernel {

class TriangularKernel
{
 public:
  TriangularKernel(const double bandwidth = 1.0) : bandwidth(bandwidth) { }

  template<typename VecTypeA, typename VecTypeB>
  double Evaluate(const VecTypeA& a, const VecTypeB& b) const
  {
    return std::max(0.0, (1 - metric::EuclideanDistance::Evaluate(a, b) /
        bandwidth));
  }

  double Evaluate(const double distance) const
  {
    return std::max(0.0, (1 - distance) / bandwidth);
  }

  double Gradient(const double distance) const
  {
    if (distance < 1)
    {
      return -1.0 / bandwidth;
    }
    else if (distance > 1)
    {
      return 0;
    }
    else
    {
      return arma::datum::nan;
    }
  }

  double Bandwidth() const { return bandwidth; }
  double& Bandwidth() { return bandwidth; }

  template<typename Archive>
  void serialize(Archive& ar, const uint32_t /* version */)
  {
    ar(CEREAL_NVP(bandwidth));
  }

 private:
  double bandwidth;
};

template<>
class KernelTraits<TriangularKernel>
{
 public:
  static const bool IsNormalized = true;
  static const bool UsesSquaredDistance = false;
};

} // namespace kernel
} // namespace mlpack

#endif