KDEDefaultParams Struct Reference

KDEDefaultParams contains the default input parameter values for KDE. More...

Static Public Attributes
static constexpr double	absError = 0
	Absolute error tolerance. More...
static constexpr size_t	initialSampleSize = 100
	Initial sample size for Monte Carlo estimations. More...
static constexpr double	mcBreakCoef = 0.4
	Monte Carlo break coefficient. More...
static constexpr double	mcEntryCoef = 3
	Monte Carlo entry coefficient. More...
static constexpr double	mcProb = 0.95
	Probability of a Monte Carlo estimation to be bounded by the relative error tolerance. More...
static constexpr KDEMode	mode = KDEMode::DUAL_TREE_MODE
	KDE algorithm mode. More...
static constexpr bool	monteCarlo = false
	Whether to use Monte Carlo estimations when possible. More...
static constexpr double	relError = 0.05
	Relative error tolerance. More...

Detailed Description

KDEDefaultParams contains the default input parameter values for KDE.

Definition at line 32 of file kde.hpp.

Member Data Documentation

absError

constexpr double absError = 0

static

Absolute error tolerance.

Definition at line 38 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::AbsoluteError(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef().

initialSampleSize

constexpr size_t initialSampleSize = 100

static

Initial sample size for Monte Carlo estimations.

Definition at line 51 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCInitialSampleSize().

mcBreakCoef

constexpr double mcBreakCoef = 0.4

static

Monte Carlo break coefficient.

Definition at line 57 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef().

mcEntryCoef

constexpr double mcEntryCoef = 3

static

Monte Carlo entry coefficient.

Definition at line 54 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCEntryCoef().

mcProb

constexpr double mcProb = 0.95

static

Probability of a Monte Carlo estimation to be bounded by the relative error tolerance.

Definition at line 48 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCProb().

mode

constexpr KDEMode mode = KDEMode::DUAL_TREE_MODE

static

KDE algorithm mode.

Definition at line 41 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::Mode().

monteCarlo

constexpr bool monteCarlo = false

static

Whether to use Monte Carlo estimations when possible.

Definition at line 44 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MonteCarlo().

relError

constexpr double relError = 0.05

static

Relative error tolerance.

Definition at line 35 of file kde.hpp.

Referenced by KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::MCBreakCoef(), and KDE< KernelType, metric::EuclideanDistance, arma::mat, TreeType >::RelativeError().

---

The documentation for this struct was generated from the following file:

• /home/ryan/src/mlpack.org/_src/mlpack-git/src/mlpack/methods/kde/kde.hpp