diff --git a/src/methods/linear_SVM/linear_SVM.pl b/src/methods/linear_SVM/linear_SVM.pl
index 864abf15bc544ff25a46c097b45be277daf34ab0..996531079d10358bd08c1623dd55d7c3d1d0b2cc 100644
--- a/src/methods/linear_SVM/linear_SVM.pl
+++ b/src/methods/linear_SVM/linear_SVM.pl
@@ -37,6 +37,9 @@
%% Initializes the linear_svm model with the given data and trains it.
%%
initModelWithTrain(DataList, DataRows, LabelsList, NumClasses, Lambda, Delta, FitIntercept, Optimizer) :-
+ NumClasses >= 0,
+ Lambda >= 0.0,
+ Delta >= 0.0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(LabelsList, array(Ysize, Y)),
initModelWithTrainI(X, Xsize, Xrownum, Y, Ysize, NumClasses, Lambda, Delta, FitIntercept, Optimizer).
@@ -60,6 +63,9 @@ foreign(initModelWithTrain, c, initModelWithTrainI( +pointer(float_array), +
%% Initializes the linear_svm model with the given data but doesnt train it.
%%
initModelNoTrain(NumClasses, Lambda, Delta, FitIntercept) :-
+ NumClasses >= 0,
+ Lambda >= 0.0,
+ Delta >= 0.0,
initModelNoTrainI(NumClasses, Lambda, Delta, FitIntercept).
foreign(initModelNoTrain, c, initModelNoTrainI( +integer, +float32, +float32,
@@ -100,8 +106,8 @@ classifyPoint(DataList, Prediction) :-
convert_list_to_float_array(DataList, array(Xsize, X)),
classifyPointI(X, Xsize, Prediction).
-foreign(classifyPoint, c, classifyPointI(+pointer(float_array), +integer,
- [-integer])).
+foreign(classifyPoint, c, classifyPointI( +pointer(float_array), +integer,
+ [-integer])).
%% --Input--
@@ -120,7 +126,7 @@ computeAccuracy(DataList, DataRows, LabelsList, Accuracy) :-
convert_list_to_float_array(LabelsList, array(Ysize, Y)),
computeAccuracyI(X, Xsize, Xrownum, Y, Ysize, Accuracy).
-foreign(computeAccuracy, c, computeAccuracyI( +pointer(float_array), +integer, +integer,
+foreign(computeAccuracy, c, computeAccuracyI( +pointer(float_array), +integer, +integer,
+pointer(float_array), +integer,
[-float32])).
@@ -138,15 +144,16 @@ foreign(computeAccuracy, c, computeAccuracyI( +pointer(float_array), +integer
%% Train the Linear_svm model with the given training data.
%%
train(DataList, DataRows, LabelsList, NumClasses, Optimizer, ObjValue) :-
+ NumClasses >= 0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(LabelsList, array(Ysize, Y)),
trainI(X, Xsize, Xrownum, Y, Ysize, NumClasses, Optimizer, ObjValue).
-foreign(train, c, trainI(+pointer(float_array), +integer, +integer,
- +pointer(float_array), +integer,
- +integer,
- +string,
- [-float32])).
+foreign(train, c, trainI( +pointer(float_array), +integer, +integer,
+ +pointer(float_array), +integer,
+ +integer,
+ +string,
+ [-float32])).
%% Defines the functions that get connected from main.cpp
diff --git a/src/methods/linear_SVM/linear_SVM_test.pl b/src/methods/linear_SVM/linear_SVM_test.pl
index 2535dc83ec63581d605f973f2132df5e0464f205..9ddaa7301faeb1f4e184db6c66aee5a434badb79 100644
--- a/src/methods/linear_SVM/linear_SVM_test.pl
+++ b/src/methods/linear_SVM/linear_SVM_test.pl
@@ -6,50 +6,285 @@
:- use_module(linear_SVM).
:- use_module('../../helper_files/helper.pl').
-reset_Model :-
+reset_Model_NoTrain :-
initModelNoTrain(2, 0.0001, 1.0, 0).
+reset_Model_WithTrain :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs).
+
:- begin_tests(lists).
-%% 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, 1.4, 3.0, 2.1, 0.1],4, array(Xsize, Xrownum, X)),
- convert_list_to_float_array([1,1,0,0], array(Ysize, Y)),
- train(X,Xsize, Xrownum,Y, Ysize, 2, lbfgs, Result),
- print(Result).
-:- end_tests(lists).
+%%
+%% TESTING predicate initModelWithTrain/8
+%%
+:- begin_tests(initModelWithTrain).
+
+%% Failure Tests
+
+test(linear_SVM_InitModelWithTrain_Negative_NumClasses, fail) :-
+ initModelWithTrain([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, [0,1,0,1], -2, 0.0001, 1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Negative_Lambda, fail) :-
+ initModelWithTrain([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, [0,1,0,1], 2, -0.0001, 1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Negative_Delta, fail) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, -1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Wrong_Optimizer_Input, [error(domain_error('The given Optimizer is unkown!' , wrongInput), _)]) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, wrongInput).
+
+test(linear_SVM_InitModelWithTrain_Too_Few_Labels, [error(_,system_error('Error'))]) :-
+ initModelWithTrain([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, [0,1], 2, 0.0001, 1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Too_Many_Labels, [error(_,system_error('Error'))]) :-
+ initModelWithTrain([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, [0,1,0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Too_Many_LabelClasses, [error(_,system_error('Error'))]) :-
+ initModelWithTrain([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, [0,1,2,3], 2, 0.0001, 1.0, 0, lbfgs).
+
+
+%% Successful Tests
+
+test(linear_SVM_InitModelWithTrain_Normal_Use_LBFGS) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs).
+
+test(linear_SVM_InitModelWithTrain_Normal_Use_PSGD) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, psgd).
+
+test(linear_SVM_InitModelWithTrain_CSV_Inpupt_LBFGS) :-
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ initModelWithTrain(Data, 4, [0,1,0,1,1,0,1,1,1,0], 2, 0.000021, 2.0, 1, lbfgs).
+
+:- end_tests(initModelWithTrain).
+
+
+
+%%
+%% TESTING predicate initModelNoTrain/4
+%%
+:- begin_tests(initModelNoTrain).
+
+%% Failure Tests
+
+test(linear_SVM_InitModelNoTrain_Negative_NumClasses, fail) :-
+ initModelNoTrain(-2, 0.0001, 1.0, 0).
+
+test(linear_SVM_InitModelNoTrain_Negative_Lambda, fail) :-
+ initModelNoTrain(2, -0.0001, 1.0, 0).
+
+test(linear_SVM_InitModelNoTrain_Negative_Delta, fail) :-
+ initModelNoTrain(2, 0.0001, -1.0, 0).
+
+
+%% Successful Tests
+
+test(linear_SVM_InitModelNoTrain_Normal_Use) :-
+ initModelNoTrain(2, 0.0001, 1.0, 0).
+
+test(linear_SVM_InitModelNoTrain_Alternative) :-
+ initModelNoTrain(2, 0.042, 0.5, 1).
+
+:- end_tests(initModelNoTrain).
+
+
%%
-%% TESTING predicate predicate/10
+%% TESTING predicate classify/5
%%
-:- begin_tests(predicate).
+:- begin_tests(classify).
%% Failure Tests
-test(testDescription, [error(domain_error('expectation' , culprit), _)]) :-
- reset_Model_No_Train(perceptron),
- train([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, [0,0,0,0], 2, culprit, 50, 0.0001, _).
+test(linear_SVM_Classify_Before_Train, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ classify([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, _, _, _).
-test(testDescription2, [error(_,system_error('The values of the Label have to start at 0 and be >= 0 and < the given numClass!'))]) :-
- reset_Model_No_Train(perceptron),
- train([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, [0,1,0,2], 2, perceptron, 50, 0.0001, _).
+test(linear_SVM_Classify_Diffrent_Dims_Than_Train, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ classify([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5], 4, _, _, _).
%% Successful Tests
-test(testDescription3, [true(Error =:= 1)]) :-
- reset_Model_No_Train(perceptron),
- train([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, [0,0,0,0], 2, perceptron, 50, 0.0001, Error).
+test(linear_SVM_Classify_Normal_Use_LBFGS) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs),
+ classify([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, LabelsList, ScoresList, _),
+ print('\nLabels: '),
+ print(LabelsList),
+ print('\nScores: '),
+ print(ScoresList).
+
+test(linear_SVM_Classify_Normal_Use_PSGD) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, psgd),
+ classify([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, LabelsList, ScoresList, _),
+ print('\nLabels: '),
+ print(LabelsList),
+ print('\nScores: '),
+ print(ScoresList).
+
+test(linear_SVM_Classify_CSV_Input) :-
+ initModelWithTrain([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5], 4, [0,1,0], 2, 0.0001, 2.0, 1, lbfgs),
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ classify(Data, 4, LabelsList, ScoresList, _),
+ print('\nLabels: '),
+ print(LabelsList),
+ print('\nScores: '),
+ print(ScoresList).
+
+:- end_tests(classify).
+
+
+
+%%
+%% TESTING predicate classifyPoint/2
+%%
+:- begin_tests(classifyPoint).
+
+%% Failure Tests
+
+test(linear_SVM_ClassifyPoint_Before_Train, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ classifyPoint([5.1,3.5,1.4], _).
+
+test(linear_SVM_ClassifyPoint_Diffrent_Dims_Than_Train, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ classifyPoint([5.1,3.5,1.4,4.9], _).
+
+
+%% Successful Tests
+
+test(linear_SVM_ClassifyPoint_Normal_Use_LBFGS) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs),
+ classifyPoint([5.1,3.5,1.4], Label),
+ print('\nLabel: '),
+ print(Label).
+
+test(linear_SVM_ClassifyPoint_Normal_Use_PSGD) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, psgd),
+ classifyPoint([5.1,3.5,1.4], Label),
+ print('\nLabel: '),
+ print(Label).
+
+test(linear_SVM_ClassifyPoint_CSV_Input) :-
+ initModelWithTrain([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5], 4, [0,1,0], 2, 0.0001, 2.0, 1, lbfgs),
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,1, Data),
+ classifyPoint(Data, Label),
+ print('\nLabel: '),
+ print(Label).
+
+:- end_tests(classifyPoint).
+
+
+
+%%
+%% TESTING predicate computeAccuracy/4
+%%
+:- begin_tests(computeAccuracy).
+
+%% Failure Tests
+
+test(linear_SVM_ComputeAccuracy_Before_Train, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ computeAccuracy([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, [0,1,0,1], _).
+
+test(linear_SVM_ComputeAccuracy_Too_Few_Labels, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ computeAccuracy([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, [0,1], _).
+
+test(linear_SVM_ComputeAccuracy_Too_Many_Labels, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ computeAccuracy([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, [0,1,0,1,0,1], _).
+
+test(linear_SVM_ComputeAccuracy_Too_Many_LabelClasses, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ computeAccuracy([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, [0,1,2,3], _).
+
+test(linear_SVM_ComputeAccuracy_Wrong_Data_Dims, [error(_,system_error('Error'))]) :-
+ reset_Model_WithTrain,
+ computeAccuracy([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5], 4, [0,1,0], _).
+
+
+%% Successful Tests
+
+test(linear_SVM_ComputeAccuracy_Normal_Use_LBFGS) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, lbfgs),
+ computeAccuracy([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, [0,1,0,1], Accuracy),
+ print('\nAccuracy: '),
+ print(Accuracy).
+
+test(linear_SVM_ComputeAccuracy_Normal_Use_PSGD) :-
+ initModelWithTrain([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, [0,1,0,1], 2, 0.0001, 1.0, 0, psgd),
+ computeAccuracy([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, [0,1,0,1], Accuracy),
+ print('\nAccuracy: '),
+ print(Accuracy).
+
+test(linear_SVM_ComputeAccuracy_CSV_Input) :-
+ initModelWithTrain([5.1,3.5,1.4,4.9,3.0,1.4,4.7,3.2,1.3,4.6,3.1,1.5], 4, [0,1,0], 2, 0.0001, 2.0, 1, lbfgs),
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,1, Data),
+ computeAccuracy(Data, 4, [0,1,0,1,1,0,1,1,1,0], Accuracy),
+ print('\nAccuracy: '),
+ print(Accuracy).
+
+:- end_tests(computeAccuracy).
+
+
+
+%%
+%% TESTING predicate train/6
+%%
+:- begin_tests(train).
+
+%% Failure Tests
+
+test(linear_SVM_Train_Negative_NumClasses, fail) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,0,1], -2, lbfgs, _).
+
+test(linear_SVM_Train_Wrong_Optimizer_Input, [error(domain_error('The given Optimizer is unkown!' , wrongInput), _)]) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,0,1], 2, wrongInput, _).
+
+test(linear_SVM_Train_Too_Few_Labels, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ train([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, [0,1], 2, lbfgs, _).
+
+test(linear_SVM_Train_Too_Many_Labels, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,0,1,0,1], 2, lbfgs, _).
+
+test(linear_SVM_Train_Too_Many_LabelClasses, [error(_,system_error('Error'))]) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,2,3], 2, lbfgs, _).
+
+
+%% Successful Tests
+
+test(linear_SVM_Train_Normal_Use_LBFGS) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,0,1], 2, lbfgs, ObjectiveValue),
+ print('\nObjectiveValue: '),
+ print(ObjectiveValue).
+
+test(linear_SVM_Train_Normal_Use_PSGD) :-
+ reset_Model_NoTrain,
+ train([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, [0,1,0,1], 2, psgd, ObjectiveValue),
+ print('\nObjectiveValue: '),
+ print(ObjectiveValue).
-test(testDescription4, [true(Error =:= 0.9797958971132711)]) :-
- reset_Model_No_Train(perceptron),
+test(linear_SVM_Train_CSV_Inpupt_LBFGS) :-
+ reset_Model_NoTrain,
open('src/data_csv/iris2.csv', read, File),
take_csv_row(File, skipFirstRow,10, Data),
- train(Data, 4, [0,1,0,1,1,0,1,1,1,0], 2, perceptron, 50, 0.0001, Error).
+ train(Data, 4, [0,1,0,1,1,0,1,1,1,0], 2, lbfgs, ObjectiveValue),
+ print('\nObjectiveValue: '),
+ print(ObjectiveValue).
-:- end_tests(predicate).
+:- end_tests(train).
run_linear_SVM_tests :-
run_tests.