NAME¶

mlpack::regression::LinearRegression - A simple linear regression algorithm using ordinary least squares.

SYNOPSIS¶

Public Member Functions¶

LinearRegression (const arma::mat &predictors, const arma::vec &responses, const double lambda=0) Creates the model. LinearRegression (const std::string &filename) Initialize the model from a file. LinearRegression (const LinearRegression &linearRegression) Copy constructor. LinearRegression () Empty constructor. double ComputeError (const arma::mat &points, const arma::vec &responses) const Calculate the L2 squared error on the given predictors and responses using this linear regression model. double Lambda () const Return the Tikhonov regularization parameter for ridge regression. double & Lambda () Modify the Tikhonov regularization parameter for ridge regression. const arma::vec & Parameters () const Return the parameters (the b vector). arma::vec & Parameters () Modify the parameters (the b vector). void Predict (const arma::mat &points, arma::vec &predictions) const Calculate y_i for each data point in points. std::string ToString () const

Private Attributes¶

double lambda The Tikhonov regularization parameter for ridge regression (0 for linear regression). arma::vec parameters The calculated B.

Detailed Description¶

A simple linear regression algorithm using ordinary least squares. Optionally, this class can perform ridge regression, if the lambda parameter is set to a number greater than zero. Definition at line 35 of file linear_regression.hpp.

Constructor & Destructor Documentation¶

mlpack::regression::LinearRegression::LinearRegression (const arma::mat &predictors, const arma::vec &responses, const doublelambda = 0)¶

Creates the model. Parameters:

mlpack::regression::LinearRegression::LinearRegression (const std::string &filename)¶

Initialize the model from a file. Parameters:

mlpack::regression::LinearRegression::LinearRegression (const LinearRegression &linearRegression)¶

Copy constructor. Parameters:

mlpack::regression::LinearRegression::LinearRegression () [inline]¶

Empty constructor. Definition at line 65 of file linear_regression.hpp.

Member Function Documentation¶

double mlpack::regression::LinearRegression::ComputeError (const arma::mat &points, const arma::vec &responses) const¶

Calculate the L2 squared error on the given predictors and responses using this linear regression model. This calculation returns where $ y $ is the responses vector, $ X $ is the matrix of predictors, and $ B $ is the parameters of the trained linear regression model. As this number decreases to 0, the linear regression fit is better. Parameters:

double mlpack::regression::LinearRegression::Lambda () const [inline]¶

Return the Tikhonov regularization parameter for ridge regression. Definition at line 101 of file linear_regression.hpp. References lambda.

double& mlpack::regression::LinearRegression::Lambda () [inline]¶

Modify the Tikhonov regularization parameter for ridge regression. Definition at line 103 of file linear_regression.hpp. References lambda.

const arma::vec& mlpack::regression::LinearRegression::Parameters () const [inline]¶

Return the parameters (the b vector). Definition at line 96 of file linear_regression.hpp. References parameters.

arma::vec& mlpack::regression::LinearRegression::Parameters () [inline]¶

Modify the parameters (the b vector). Definition at line 98 of file linear_regression.hpp. References parameters.

void mlpack::regression::LinearRegression::Predict (const arma::mat &points, arma::vec &predictions) const¶

Calculate y_i for each data point in points. Parameters:

std::string mlpack::regression::LinearRegression::ToString () const¶

Member Data Documentation¶

double mlpack::regression::LinearRegression::lambda [private]¶

The Tikhonov regularization parameter for ridge regression (0 for linear regression). Definition at line 119 of file linear_regression.hpp. Referenced by Lambda().

arma::vec mlpack::regression::LinearRegression::parameters [private]¶

The calculated B. Initialized and filled by constructor to hold the least squares solution. Definition at line 113 of file linear_regression.hpp. Referenced by Parameters().

Author¶

Generated automatically by Doxygen for MLPACK from the source code.