Obsidian
Probabilistic Geophysical Joint Inversion
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Pages
kernel.cpp File Reference

Contains the implementation of the squared exponetial kernel. More...

#include "kernel.hpp"
#include <iostream>
Include dependency graph for kernel.cpp:

Namespaces

 obsidian
 
 obsidian::world
 

Functions

Eigen::MatrixXd obsidian::world::sqExp (const Eigen::MatrixXd &x1, const Eigen::MatrixXd &x2, const Eigen::VectorXd &lengthScale, bool noisy=false)
 Compute a covariance matrix using the square exponential function. More...
 
Eigen::MatrixXd obsidian::world::sqExp2d (const Eigen::Matrix< double, 2,-1 > &x1, const Eigen::Matrix< double, 2,-1 > &x2, const Eigen::Vector2d &lengthScale, bool noisy=false)
 Compute a covariance matrix using the squared exponential function. This function is faster than sqExp, but only applies when the dimensions of the points is 2 (which is true for control points and query points). More...
 
Eigen::Vector2d obsidian::world::autoLengthScale (const std::pair< double, double > &x1x2, const std::pair< double, double > &y1y2, uint resx, uint resy)
 Get an appropriate length scale given the grid resolution and a smoothing factor. More...
 

Detailed Description

Contains the implementation of the squared exponetial kernel.

Author
Alistair Reid
Darren Shen
Date
2014
License:
Affero General Public License version 3 or later