regression_distribution.hpp

Go to the documentation of this file.

#ifndef MLPACK_CORE_DISTRIBUTIONS_REGRESSION_DISTRIBUTION_HPP
#define MLPACK_CORE_DISTRIBUTIONS_REGRESSION_DISTRIBUTION_HPP

#include <mlpack/prereqs.hpp>
#include <mlpack/core/dists/gaussian_distribution.hpp>
#include <mlpack/methods/linear_regression/linear_regression.hpp>

namespace mlpack {
namespace distribution {

class RegressionDistribution
{
 private:
  regression::LinearRegression rf;
  GaussianDistribution err;

 public:
  RegressionDistribution() { /* nothing to do */ }

  RegressionDistribution(const arma::mat& predictors,
                         const arma::vec& responses) :
      rf(regression::LinearRegression(predictors, responses))
  {
    err = GaussianDistribution(1);
    arma::mat cov(1, 1);
    cov(0, 0) = rf.ComputeError(predictors, responses);
    err.Covariance(std::move(cov));
  }

  template<typename Archive>
  void Serialize(Archive& ar, const unsigned int /* version */)
  {
    ar & data::CreateNVP(rf, "rf");
    ar & data::CreateNVP(err, "err");
  }

  const regression::LinearRegression& Rf() const { return rf; }
  regression::LinearRegression& Rf() { return rf; }

  const GaussianDistribution& Err() const { return err; }
  GaussianDistribution& Err() { return err; }

  void Train(const arma::mat& observations);

  void Train(const arma::mat& observations, const arma::vec& weights);

  double Probability(const arma::vec& observation) const;

  double LogProbability(const arma::vec& observation) const {
    return log(Probability(observation));
  }

  void Predict(const arma::mat& points, arma::vec& predictions) const;

  const arma::vec& Parameters() const { return rf.Parameters(); }

  size_t Dimensionality() const { return rf.Parameters().n_elem; }
};


} // namespace distribution
} // namespace mlpack

#endif