reverend

Kernel Bayes' Rule implementation and extensions as part of my PhD.

Requirements

For the C++ Executables: * Boost 1.53 (may work with earlier versions) * Eigen3 * Nlopt (included in this source tree)

For the Python scripts: * Python2 * Numpy/Scipy * Matplotlib

Building

First you need to build cnpy. In the directory cnpy type:

$ cmake . $ make

Now, go into the cpp directory and build the executable in-source:

$ cmake . $ make

CMake will complain if the required libraries aren't installed.

Demo

Just run

$python demo_regression.py

In the regression folder! If all is well, you will be rewarded with pretty pictures.

Bugs

It is almost certain that this code will break on a Windows machine. I'll fix this fairly soon.