obsidian::distrib Namespace Reference

Classes
struct	MultiGaussian
	Represents a multivariate Gaussian distribution. More...

Functions
double	logGaussian (const Eigen::VectorXd &x, const Eigen::VectorXd &mu, const Eigen::VectorXd &sigma)
	Compute the log Gaussian distribution PDF with a diagonal covariance matrix. More...
MultiGaussian	coupledGaussianBlock (const Eigen::MatrixXd &mean, double coupledSD, double decoupledSD)
double	logPDF (const Eigen::MatrixXd &theta, const MultiGaussian &input, const Eigen::MatrixXd &thetaMin, const Eigen::MatrixXd &thetaMax)
	Compute the log PDF of a multivariate Gaussian distribution.
double	logPDF (const Eigen::VectorXd &theta, const MultiGaussian &input, const Eigen::VectorXd &thetaMin, const Eigen::VectorXd &thetaMax)
	Compute the log PDF of a multivariate Gaussian distribution.
double	uniformLogPDF (const Eigen::MatrixXd &theta, const MultiGaussian &input, const Eigen::MatrixXd &thetaMins, const Eigen::MatrixXd &thetaMaxs)
Eigen::MatrixXd	drawValues (const MultiGaussian &input, std::mt19937 &gen)
	Draw a sample from a multivariate Gaussian distribution.
Eigen::MatrixXd	drawUniformValues (const Eigen::MatrixXd &min, const Eigen::MatrixXd &max, std::mt19937 &gen)
std::vector< Eigen::MatrixXd >	drawFrom (const std::vector< distrib::MultiGaussian > &prior, std::mt19937 &gen, const std::vector< Eigen::MatrixXd > &mins, const std::vector< Eigen::MatrixXd > &maxs, const std::vector< bool > &uniformFlags)
	Draw a sample from a multivariate Gaussian distribution.
std::vector< Eigen::VectorXd >	drawVectorFrom (const std::vector< distrib::MultiGaussian > &prior, std::mt19937 &gen, const std::vector< Eigen::VectorXd > &mins, const std::vector< Eigen::VectorXd > &maxs)
	Draw a sample from a multivariate Gaussian distribution.

Detailed Description

Namespace for probability distribution related functions.

Function Documentation

double obsidian::distrib::logGaussian	(	const Eigen::VectorXd &	x,
		const Eigen::VectorXd &	mu,
		const Eigen::VectorXd &	sigma
	)

Compute the log Gaussian distribution PDF with a diagonal covariance matrix.

Parameters

x	The vector to evaluate the distribution at.
mu	The mean vector containing the means of each dimension.
sigma	The diagonal of the covariance matrix.