Prolog mlpack Library

Getting started

To use the main file with the connected BayesianLinearRegression functions in sicstus use:

compile(main).

while running sicstus in the repository or:

['/full_path_to_main/main'].

when running sicstus in a different directory.

Compile the main files with Makefile

To compile the main files just use make

if your full path to splfr is not /usr/local/sicstus4.7.1/bin/splfr

then you will have to exchange this Makefile line with your splfr path.

splfr=/usr/local/sicstus4.7.1/bin/splfr

Compile the main files manually

To compile the main files manually with splfr you need to use the following command in the repository folder.

The flags

-lmlpack (for mlpack intern libarys like armadillo)

-lstdc++ (for stdI/O and vector etc.)

--struct (for the sicstus struct library)

splfr -lmlpack -lstdc++ -cxx --struct main.pl main.cpp mlpack-3.4.2/src/mlpack/methods/bayesian_linear_regression/bayesian_linear_regression.cpp 

or if you dont have a command shortcut for splfr use

/path_to_splfr/splfr -lmlpack -lstdc++ -cxx --struct main.pl main.cpp mlpack-3.4.2/src/mlpack/methods/bayesian_linear_regression/bayesian_linear_regression.cpp 

Available functions

For a list of the functions the BayesianLinearRegression method has go to

Mlpack BayesianLinearRegression documentation

For Arrays i have added predicates for converting Prolog lists [] to the needed float_array type and when receiving an array also Predicates for converting it to a normal list.

A Matrix is just a long list with the added information how many row it has.

In both cases when using the array in a function the size must also be send, since after sending or receiving it you cant get the length of the array from itself.

for example:

train :-
        convert_list_to_float_array([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5],3, array(Xsize, Xrownum, X)),
        convert_list_to_float_array([0.2,0.2,0.2,0.2], array(Ysize, Y)),
        train(X,Xsize, Xrownum,Y, Ysize).

here the longer list representing a Matrix gets convertet into an array X with the row amount, and the smaller list into an array Y. And they get used in the foreign function from BayesianLinearRegression that takes an Matrix of points and a vector of targets to train the model.