Adding logistic_regression

src/methods/logistic_regression/Makefile

+splfr=/usr/local/sicstus4.7.1/bin/splfr
+
+METHOD_NAME=logistic_regression
+
+$(METHOD_NAME).so: $(METHOD_NAME).pl $(METHOD_NAME).cpp
+	$(splfr) -larmadillo -fopenmp -lmlpack -lstdc++ -cxx --struct $(METHOD_NAME).pl $(METHOD_NAME).cpp ../../helper_files/helper.cpp
+clean:
+	rm $(METHOD_NAME).so

src/methods/logistic_regression/logistic_regression.cpp

+#include <sicstus/sicstus.h>
+/* ex_glue.h is generated by splfr from the foreign/[2,3] facts.
+   Always include the glue header in your foreign resource code.
+*/
+#include "logistic_regression_glue.h"
+#include <mlpack/methods/logistic_regression/logistic_regression.hpp>
+#include <mlpack/core.hpp>
+
+// including helper functions for converting between arma structures and arrays
+#include "../../helper_files/helper.hpp"
+
+// some of the most used namespaces
+using namespace arma;
+using namespace mlpack;
+using namespace std;
+using namespace mlpack::regression;
+
+// Global Variable of the LogisticRegression object so it can be accessed from all functions
+LogisticRegression logisticRegression;
+
+// TODO: 
+// input: 	const MatType & 				predictors,
+//			const arma::Row< size_t > & 	responses,
+//			const arma::rowvec & 			initialPoint,
+//			const double 					lambda = 0
+// output: 
+// description: 
+void initModelNoOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum, float *responsesArr, SP_integer responsesArrSize, double lambda)
+{
+	// convert the Prolog arrays to arma::mat
+	mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
+	// convert the Prolog arrays to arma::rowvec
+	arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
+
+	logisticRegression = LogisticRegression(predictors, responsesVector, lambda);
+}
+
+// TODO: 
+// input: 	const MatType & 				predictors,
+//			const arma::Row< size_t > & 	responses,
+//			OptimizerType & 				optimizer,
+//			const double 					lambda 
+// output: 
+// description: 
+void initModelWithOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum, float *responsesArr, SP_integer responsesArrSize, char const *optimizer, double lambda)
+{
+	// convert the Prolog arrays to arma::mat
+	mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
+	// convert the Prolog arrays to arma::rowvec
+	arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
+
+	logisticRegression = LogisticRegression<>(predictors.n_cols, lambda);
+	
+	if (strcmp(optimizer, "lbfgs") == 0)
+	{
+		logisticRegression.Train(predictors,responsesVector);
+	}
+	else if (strcmp(optimizer, "psgd") == 0)
+	{
+		logisticRegression.Train(predictors,responsesVector, ens::ParallelSGD(100,64));
+	}
+	else
+	{
+		cout << "Wrong optimizer input" << endl;
+	}
+}
+
+// TODO: 
+// input: 	const VecType & 	point,
+//			const double 		decisionBoundary = 0.5
+//
+// output: 	size_t				predicted point label
+// description: 
+SP_integer classifyPoint(float *pointArr, SP_integer pointArrSize, double decisionBoundary)
+{
+	// convert the Prolog arrays to arma::rowvec
+	rowvec pointVector = convertArrayToRowvec(pointArr, pointArrSize);
+	return logisticRegression.Classify(pointVector, decisionBoundary);
+}
+
+// TODO: 
+// input: 	const MatType & 		dataset,
+//			arma::Row< size_t > & 	labels <-,
+//			const double 			decisionBoundary = 0.5
+// output: 
+// description: 
+void classifyLabels(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowNum, float **labelsArr, SP_integer *labelsArrSize, double decisionBoundary)
+{
+	// convert the Prolog arrays to arma::mat
+	mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+
+	// get the ReturnVector
+	Row<size_t> labelsReturnVector;
+
+	logisticRegression.Classify(data, labelsReturnVector, decisionBoundary);
+
+	// return the Vector lenght
+	*labelsArrSize = labelsReturnVector.n_elem;
+
+	// return the Vector as Array
+	*labelsArr = convertToArray(labelsReturnVector);
+}
+
+// TODO: 
+// input: 	const MatType & 	dataset,
+//			arma::mat & 		probabilities <-
+// output: 
+// description: 
+void classifyProbs(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowNum, float **probsMatArr, SP_integer *probsMatColNum, SP_integer *probsMatRowNum)
+{
+	// convert the Prolog arrays to arma::mat
+	mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+
+	// get the ReturnMat
+	mat probsReturnMat;
+
+	logisticRegression.Classify(data, probsReturnMat);
+
+	// return the Matrix dimensions
+	*probsMatColNum = probsReturnMat.n_cols;
+	*probsMatRowNum = probsReturnMat.n_rows;
+
+	// return the Matrix as one long Array
+	*probsMatArr = convertToArray(probsReturnMat);
+}
+
+// TODO: 
+// input: 	const MatType & 				predictors,
+//			const arma::Row< size_t > & 	responses,
+//			const double 					decisionBoundary = 0.5
+//
+// output: 	double							accuracy
+// description: 
+double computeAccuracy(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum, float *responsesArr, SP_integer responsesArrSize, double decisionBoundary)
+{
+	// convert the Prolog arrays to arma::mat
+	mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
+	// convert the Prolog arrays to arma::rowvec
+	arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
+
+	return logisticRegression.ComputeAccuracy(predictors, responsesVector, decisionBoundary);
+}
+
+// TODO: 
+// input: 	const MatType & 				predictors,
+//			const arma::Row< size_t > & 	responses
+//
+// output: 	double							error
+// description: 
+double computeError(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum, float *responsesArr, SP_integer responsesArrSize)
+{
+	// convert the Prolog arrays to arma::mat
+	mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
+	// convert the Prolog arrays to arma::rowvec
+	arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
+
+	return logisticRegression.ComputeError(predictors, responsesVector);
+}
+
+// TODO: 
+// input: 	const MatType & 				predictors,
+//			const arma::Row< size_t > & 	responses,
+//			OptimizerType & 				optimizer,
+//			CallbackTypes &&... 			callbacks
+// output: 
+// description: 
+void train(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum, float *responsesArr, SP_integer responsesArrSize, char const *optimizer)
+{
+	// convert the Prolog arrays to arma::mat
+	mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
+	// convert the Prolog arrays to arma::rowvec
+	arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
+
+	if (strcmp(optimizer, "lbfgs") == 0)
+	{
+		logisticRegression.Train(predictors,responsesVector);
+	}
+	else if (strcmp(optimizer, "psgd") == 0)
+	{
+		logisticRegression.Train(predictors,responsesVector, ens::ParallelSGD(100,64));
+	}
+	else
+	{
+		cout << "Wrong optimizer input" << endl;
+	}
+}
\ No newline at end of file

src/methods/logistic_regression/logistic_regression.pl

+:- module(logistic_regression, [   initModelNoOptimizer/6,
+                                   initModelWithOptimizer/7,
+                                   classifyPoint/4,
+                                   classifyLabels/6,
+                                   classifyProbs/6,
+                                   computeAccuracy/7,
+                                   computeError/6,
+                                   train/6]).
+
+%% requirements of library(struct)
+:- load_files(library(str_decl),
+              [when(compile_time), if(changed)]).
+
+%% needed for using the array type
+:- use_module(library(structs)).
+:- use_module('../../helper_files/helper.pl').
+
+%% type definitions for the float array
+:- foreign_type
+       float32          = float_32,
+       float_array      = array(float32).
+
+%% definitions for the connected function
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(initModelNoOptimizer, c, initModelNoOptimizer(+pointer(float_array), +integer, +integer, +pointer(float_array), +integer, +float32)).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(initModelWithOptimizer, c, initModelWithOptimizer(+pointer(float_array), +integer, +integer, +pointer(float_array), +integer, +string, +float32)).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(classifyPoint, c, classifyPoint(+pointer(float_array), +integer, +float32, [-integer])).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(classifyLabels, c, classifyLabels(+pointer(float_array), +integer, +integer, -pointer(float_array), -integer, +float32)).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(classifyProbs, c, classifyProbs(+pointer(float_array), +integer, +integer, -pointer(float_array), -integer, -integer)).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(computeAccuracy, c, computeAccuracy(+pointer(float_array), +integer, +integer, +pointer(float_array), +integer, +float32, [-float32])).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(computeError, c, computeError(+pointer(float_array), +integer, +integer, +pointer(float_array), +integer, [-float32])).
+
+%% TODO: 
+%% input: 
+%% output: 
+%% description: 
+foreign(train, c, train(+pointer(float_array), +integer, +integer, +pointer(float_array), +integer, +string)).
+
+
+%% Defines the functions that get connected from main.cpp
+foreign_resource(logistic_regression, [   initModelNoOptimizer,
+                                          initModelWithOptimizer,
+                                          classifyPoint,
+                                          classifyLabels,
+                                          classifyProbs,
+                                          computeAccuracy,
+                                          computeError,
+                                          train]).
+
+:- load_foreign_resource(logistic_regression).
\ No newline at end of file

src/methods/logistic_regression/logistic_regression_test.pl

+:- use_module(library(plunit)).
+
+:- use_module(logistic_regression).
+:- use_module('../../helper_files/helper.pl').
+
+reset_Model :-
+        initModel(1,0,50,0.0001).
+
+:- begin_tests(lists).
+
+%% alpha tests
+test(alpha_std_init) :-
+        reset_Model,
+        alpha(0).
+test(alpha_wrong_input, fail) :-
+        reset_Model,
+        alpha(1).
+test(alpha_after_train, A =:= 9223372036854775808) :-
+        reset_Model,
+        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),
+        alpha(A).
+
+%% train tests
+test(correct_train) :-
+        reset_Model,
+        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).
+test(false_train, fail) :-
+        reset_Model,
+        convert_list_to_float_array([],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).
+test(false_train2, fail) :-
+        reset_Model,
+        convert_list_to_float_array([],0, 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).
+test(false_train3, fail) :-
+        reset_Model,
+        convert_list_to_float_array([1,2],0, 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).
+test(false_train3, fail) :-
+        reset_Model,
+        convert_list_to_float_array([1,2,44,3],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).
+test(false_train4) :-
+        reset_Model,
+        convert_list_to_float_array([1,2,44,3],2, 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).
+:- end_tests(lists).
\ No newline at end of file