From e45dce569a96a9ccc68f1f2c6b623a9101de5c2f Mon Sep 17 00:00:00 2001
From: Jakhes <dean.schmitz@schmitzbauer.de>
Date: Tue, 25 Oct 2022 13:56:40 +0200
Subject: [PATCH] Adding approx_kfn tests
---
src/methods/approx_kfn/approx_kfn.cpp | 175 +++++++++----
src/methods/approx_kfn/approx_kfn.pl | 84 +++++--
src/methods/approx_kfn/approx_kfn_test.pl | 294 ++++++++++++++++++++--
test_all.pl | 2 +-
4 files changed, 472 insertions(+), 83 deletions(-)
diff --git a/src/methods/approx_kfn/approx_kfn.cpp b/src/methods/approx_kfn/approx_kfn.cpp
index cc77858..5a11190 100644
--- a/src/methods/approx_kfn/approx_kfn.cpp
+++ b/src/methods/approx_kfn/approx_kfn.cpp
@@ -18,160 +18,241 @@ using namespace mlpack::neighbor;
// Global Variable of the DrusillaSelect object so it can be accessed from all functions
DrusillaSelect<mat> drusillaSelect = DrusillaSelect(1,1);
+bool isDrusillaTrained = false;
// Global Variable of the QDAFN object so it can be accessed from all functions
QDAFN<mat> qdafn = QDAFN(1,1);
+bool isQDAFNTrained = false;
+
-// TODO:
// input: const size_t l,
// const size_t m
// output:
// description:
+// Initiates the DrusillaSearch Model but doesnt train it.
+//
void initDrusillaModelNoTrain(SP_integer l, SP_integer m)
{
drusillaSelect = DrusillaSelect(l, m);
+ isDrusillaTrained = false;
}
-// TODO:
+
// input: const MatType & referenceSet,
// const size_t l,
// const size_t m
// output:
// description:
-void initDrusillaModelWithTrain(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum, SP_integer l, SP_integer m)
+// Initiates the DrusillaSearch Model and trains it with the given reference Set.
+//
+void initDrusillaModelWithTrain(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum,
+ SP_integer l, SP_integer m)
{
// convert the Prolog arrays to arma::mat
mat referenceSet = convertArrayToMat(referenceSetMatArr, referenceSetMatSize, referenceSetMatRowNum);
- drusillaSelect = DrusillaSelect(referenceSet, l, m);
+
+ try
+ {
+ drusillaSelect = DrusillaSelect(referenceSet, l, m);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ isDrusillaTrained = true;
}
-// TODO:
+
// input: const MatType & querySet,
// const size_t k,
// arma::Mat< size_t > & neighbors,
// arma::mat & distances
// output:
// description:
+// Run Search on the given Queryset with the Drusilla Search Policy.
+//
void searchDrusilla(float *querySetMatArr, SP_integer querySetMatSize, SP_integer querySetMatRowNum,
SP_integer k,
float **neighborsMatArr, SP_integer *neighborsMatColNum, SP_integer *neighborsMatRowNum,
float **distancesMatArr, SP_integer *distancesMatColNum, SP_integer *distancesMatRowNum)
{
+ if (!isDrusillaTrained)
+ {
+ raisePrologSystemExeption("The Model is not Trained!");
+ return;
+ }
+
// convert the Prolog arrays to arma::mat
mat querySet = convertArrayToMat(querySetMatArr, querySetMatSize, querySetMatRowNum);
// create the ReturnMat
Mat< size_t > neighborsReturnMat;
-
- // create the ReturnMat
mat distancesReturnMat;
- drusillaSelect.Search(querySet, k, neighborsReturnMat, distancesReturnMat);
-
- // return the Matrix dimensions
- *neighborsMatColNum = neighborsReturnMat.n_cols;
- *neighborsMatRowNum = neighborsReturnMat.n_rows;
-
- // return the Matrix as one long Array
- *neighborsMatArr = convertToArray(neighborsReturnMat);
+
+ try
+ {
+ drusillaSelect.Search(querySet, k, neighborsReturnMat, distancesReturnMat);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
- // return the Matrix dimensions
- *distancesMatColNum = distancesReturnMat.n_cols;
- *distancesMatRowNum = distancesReturnMat.n_rows;
- // return the Matrix as one long Array
- *distancesMatArr = convertToArray(distancesReturnMat);
+ // return the Matrices
+ returnMatrixInformation(neighborsReturnMat, neighborsMatArr, neighborsMatColNum, neighborsMatRowNum);
+ returnMatrixInformation(distancesReturnMat, distancesMatArr, distancesMatColNum, distancesMatRowNum);
}
-// TODO:
+
// input: const MatType & referenceSet,
// const size_t l = 0,
// const size_t m = 0
// output:
// description:
-void trainDrusilla(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum, SP_integer l, SP_integer m)
+// Trains the DrusillaSearch Model with the given reference Set.
+//
+void trainDrusilla(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum,
+ SP_integer l, SP_integer m)
{
// convert the Prolog arrays to arma::mat
mat referenceSet = convertArrayToMat(referenceSetMatArr, referenceSetMatSize, referenceSetMatRowNum);
- drusillaSelect.Train(referenceSet, l, m);
+
+ try
+ {
+ drusillaSelect.Train(referenceSet, l, m);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ isDrusillaTrained = true;
}
-// TODO:
+
// input: const size_t l,
// const size_t m
// output:
// description:
+// Initiates the QDAFNSearch Model but doesnt train it.
+//
void initQDAFNModelNoTrain(SP_integer l, SP_integer m)
{
qdafn = QDAFN(l, m);
+ isQDAFNTrained = false;
}
-// TODO:
+
// input: const MatType & referenceSet,
// const size_t l,
// const size_t m
// output:
// description:
-void initQDAFNModelWithTrain(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum, SP_integer l, SP_integer m)
+// Initiates the QDAFNSearch Model and trains it with the given reference Set.
+//
+void initQDAFNModelWithTrain(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum,
+ SP_integer l, SP_integer m)
{
// convert the Prolog arrays to arma::mat
mat referenceSet = convertArrayToMat(referenceSetMatArr, referenceSetMatSize, referenceSetMatRowNum);
- qdafn = QDAFN(referenceSet, l, m);
+
+ try
+ {
+ qdafn = QDAFN(referenceSet, l, m);
+ }
+ catch(const std::out_of_range& e)
+ {
+ raisePrologSystemExeption("l*m must be smaller than the number of points in the dataset.");
+ return;
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ isQDAFNTrained = true;
}
-// TODO:
+
// input: const MatType & querySet,
// const size_t k,
// arma::Mat< size_t > & neighbors,
// arma::mat & distances
// output:
// description:
+// Run Search on the given Queryset with the QDAFN Search Policy.
+//
void searchQDAFN(float *querySetMatArr, SP_integer querySetMatSize, SP_integer querySetMatRowNum,
SP_integer k,
float **neighborsMatArr, SP_integer *neighborsMatColNum, SP_integer *neighborsMatRowNum,
float **distancesMatArr, SP_integer *distancesMatColNum, SP_integer *distancesMatRowNum)
{
+ if (!isQDAFNTrained)
+ {
+ raisePrologSystemExeption("The Model is not Trained!");
+ return;
+ }
+
// convert the Prolog arrays to arma::mat
mat querySet = convertArrayToMat(querySetMatArr, querySetMatSize, querySetMatRowNum);
// create the ReturnMat
Mat< size_t > neighborsReturnMat;
-
- // create the ReturnMat
mat distancesReturnMat;
- qdafn.Search(querySet, k, neighborsReturnMat, distancesReturnMat);
-
- // return the Matrix dimensions
- *neighborsMatColNum = neighborsReturnMat.n_cols;
- *neighborsMatRowNum = neighborsReturnMat.n_rows;
-
- // return the Matrix as one long Array
- *neighborsMatArr = convertToArray(neighborsReturnMat);
-
- // return the Matrix dimensions
- *distancesMatColNum = distancesReturnMat.n_cols;
- *distancesMatRowNum = distancesReturnMat.n_rows;
+
+ try
+ {
+ qdafn.Search(querySet, k, neighborsReturnMat, distancesReturnMat);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
- // return the Matrix as one long Array
- *distancesMatArr = convertToArray(distancesReturnMat);
+ // return the Matrices
+ returnMatrixInformation(neighborsReturnMat, neighborsMatArr, neighborsMatColNum, neighborsMatRowNum);
+ returnMatrixInformation(distancesReturnMat, distancesMatArr, distancesMatColNum, distancesMatRowNum);
}
-// TODO:
+
// input: const MatType & referenceSet,
// const size_t l = 0,
// const size_t m = 0
// output:
// description:
-void trainQDAFN(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum, SP_integer l, SP_integer m)
+// Trains the QDAFNSearch Model with the given reference Set.
+//
+void trainQDAFN(float *referenceSetMatArr, SP_integer referenceSetMatSize, SP_integer referenceSetMatRowNum,
+ SP_integer l, SP_integer m)
{
// convert the Prolog arrays to arma::mat
mat referenceSet = convertArrayToMat(referenceSetMatArr, referenceSetMatSize, referenceSetMatRowNum);
- qdafn.Train(referenceSet, l, m);
+
+ try
+ {
+ qdafn.Train(referenceSet, l, m);
+ }
+ catch(const std::out_of_range& e)
+ {
+ raisePrologSystemExeption("l*m must be smaller than the number of points in the dataset.");
+ return;
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ isQDAFNTrained = true;
}
diff --git a/src/methods/approx_kfn/approx_kfn.pl b/src/methods/approx_kfn/approx_kfn.pl
index 78bbf17..6815eac 100644
--- a/src/methods/approx_kfn/approx_kfn.pl
+++ b/src/methods/approx_kfn/approx_kfn.pl
@@ -1,12 +1,12 @@
:- module(approx_kfn, [ initDrusillaModelNoTrain/2,
- initDrusillaModelWithTrain/5,
- searchDrusilla/10,
- trainDrusilla/5,
+ initDrusillaModelWithTrain/4,
+ searchDrusilla/7,
+ trainDrusilla/4,
initQDAFNModelNoTrain/2,
- initQDAFNModelWithTrain/5,
- searchQDAFN/10,
- trainQDAFN/5]).
+ initQDAFNModelWithTrain/4,
+ searchQDAFN/7,
+ trainQDAFN/4]).
%% requirements of library(struct)
:- load_files(library(str_decl),
@@ -34,7 +34,12 @@
%% Initiates the DrusillaSearch Model but doesnt train it.
%% trainDrusilla has to be used before searchDrusilla can be used.
%%
-foreign(initDrusillaModelNoTrain, c, initDrusillaModelNoTrain(+integer, +integer)).
+initDrusillaModelNoTrain(L, M) :-
+ L > 0,
+ M > 0,
+ initDrusillaModelNoTrainI(L, M).
+
+foreign(initDrusillaModelNoTrain, c, initDrusillaModelNoTrainI(+integer, +integer)).
%% --Input--
@@ -48,7 +53,13 @@ foreign(initDrusillaModelNoTrain, c, initDrusillaModelNoTrain(+integer, +integer
%% Initiates the DrusillaSearch Model and trains it with the given reference Set.
%% Afterwards searchDrusilla can be used.
%%
-foreign(initDrusillaModelWithTrain, c, initDrusillaModelWithTrain( +pointer(float_array), +integer, +integer,
+initDrusillaModelWithTrain(DataList, DataRows, L, M) :-
+ L > 0,
+ M > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ initDrusillaModelWithTrainI(X, Xsize, Xrownum, L, M).
+
+foreign(initDrusillaModelWithTrain, c, initDrusillaModelWithTrainI( +pointer(float_array), +integer, +integer,
+integer, +integer)).
@@ -61,9 +72,16 @@ foreign(initDrusillaModelWithTrain, c, initDrusillaModelWithTrain( +pointer
%% mat distances
%%
%% --Description--
-%%
+%% Run Search on the given Queryset with the Drusilla Search Policy.
%%
-foreign(searchDrusilla, c, searchDrusilla( +pointer(float_array), +integer, +integer,
+searchDrusilla(DataList, DataRows, K, NeighborsList, YCols, DistancesList, ZCols) :-
+ K > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ searchDrusillaI(X, Xsize, Xrownum, K, Y, YCols, YRows, Z, ZCols, ZRows),
+ convert_float_array_to_2d_list(Y, YCols, YRows, NeighborsList),
+ convert_float_array_to_2d_list(Z, ZCols, ZRows, DistancesList).
+
+foreign(searchDrusilla, c, searchDrusillaI( +pointer(float_array), +integer, +integer,
+integer,
-pointer(float_array), -integer, -integer,
-pointer(float_array), -integer, -integer)).
@@ -79,8 +97,14 @@ foreign(searchDrusilla, c, searchDrusilla( +pointer(float_array), +integer,
%% --Description--
%% Trains the DrusillaSearch Model with the given reference Set.
%%
-foreign(trainDrusilla, c, trainDrusilla(+pointer(float_array), +integer, +integer,
- +integer, +integer)).
+trainDrusilla(DataList, DataRows, L, M) :-
+ L > 0,
+ M > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ trainDrusillaI(X, Xsize, Xrownum, L, M).
+
+foreign(trainDrusilla, c, trainDrusillaI( +pointer(float_array), +integer, +integer,
+ +integer, +integer)).
%% --Input--
@@ -93,7 +117,12 @@ foreign(trainDrusilla, c, trainDrusilla(+pointer(float_array), +integer, +intege
%% Initiates the QDAFNSearch Model but doesnt train it.
%% trainQDAFN has to be used before searchQDAFN can be used.
%%
-foreign(initQDAFNModelNoTrain, c, initQDAFNModelNoTrain(+integer, +integer)).
+initQDAFNModelNoTrain(L, M) :-
+ L > 0,
+ M > 0,
+ initQDAFNModelNoTrainI(L, M).
+
+foreign(initQDAFNModelNoTrain, c, initQDAFNModelNoTrainI(+integer, +integer)).
%% --Input--
@@ -107,7 +136,13 @@ foreign(initQDAFNModelNoTrain, c, initQDAFNModelNoTrain(+integer, +integer)).
%% Initiates the QDAFNSearch Model and trains it with the given reference Set.
%% Afterwards searchQDAFN can be used.
%%
-foreign(initQDAFNModelWithTrain, c, initQDAFNModelWithTrain( +pointer(float_array), +integer, +integer,
+initQDAFNModelWithTrain(DataList, DataRows, L, M) :-
+ L > 0,
+ M > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ initQDAFNModelWithTrainI(X, Xsize, Xrownum, L, M).
+
+foreign(initQDAFNModelWithTrain, c, initQDAFNModelWithTrainI( +pointer(float_array), +integer, +integer,
+integer, +integer)).
@@ -120,14 +155,21 @@ foreign(initQDAFNModelWithTrain, c, initQDAFNModelWithTrain( +pointer(float_a
%% mat distances
%%
%% --Description--
-%%
+%% Run Search on the given Queryset with the QDAFN Search Policy.
%%
-foreign(searchQDAFN, c, searchQDAFN( +pointer(float_array), +integer, +integer,
+searchQDAFN(DataList, DataRows, K, NeighborsList, YCols, DistancesList, ZCols) :-
+ K > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ searchQDAFNI(X, Xsize, Xrownum, K, Y, YCols, YRows, Z, ZCols, ZRows),
+ convert_float_array_to_2d_list(Y, YCols, YRows, NeighborsList),
+ convert_float_array_to_2d_list(Z, ZCols, ZRows, DistancesList).
+
+foreign(searchQDAFN, c, searchQDAFNI( +pointer(float_array), +integer, +integer,
+integer,
-pointer(float_array), -integer, -integer,
-pointer(float_array), -integer, -integer)).
-%% TODO:
+
%% --Input--
%% mat referenceSet,
%% int l => 0,
@@ -138,7 +180,13 @@ foreign(searchQDAFN, c, searchQDAFN( +pointer(float_array), +integer, +intege
%% --Description--
%% Trains the QDAFNSearch Model with the given reference Set.
%%
-foreign(trainQDAFN, c, trainQDAFN( +pointer(float_array), +integer, +integer,
+trainQDAFN(DataList, DataRows, L, M) :-
+ L > 0,
+ M > 0,
+ convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
+ trainQDAFNI(X, Xsize, Xrownum, L, M).
+
+foreign(trainQDAFN, c, trainQDAFNI( +pointer(float_array), +integer, +integer,
+integer, +integer)).
diff --git a/src/methods/approx_kfn/approx_kfn_test.pl b/src/methods/approx_kfn/approx_kfn_test.pl
index b37a384..7b21249 100644
--- a/src/methods/approx_kfn/approx_kfn_test.pl
+++ b/src/methods/approx_kfn/approx_kfn_test.pl
@@ -6,38 +6,298 @@
:- use_module(approx_kfn).
:- use_module('../../helper_files/helper.pl').
-reset_Model :-
- initModel(1,0,50,0.0001).
+reset_DrusillaModel_NoTrain :-
+ initDrusillaModelNoTrain(2, 2).
+
+reset_DrusillaModel_WithTrain :-
+ initDrusillaModelWithTrain([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, 2, 2).
+
+reset_QDAFNModel_NoTrain :-
+ initQDAFNModelNoTrain(2, 2).
+
+reset_QDAFNModel_WithTrain :-
+ initQDAFNModelWithTrain([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, 2, 2).
+
+%%
+%% TESTING predicate initDrusillaModelNoTrain/2
+%%
+:- begin_tests(initDrusillaModelNoTrain).
+
+%% Failure Tests
+
+test(approx_KFN_InitDrusillaNoTrain_Negative_L, fail) :-
+ initDrusillaModelNoTrain(-5,5).
+
+test(approx_KFN_InitDrusillaNoTrain_Negative_M, fail) :-
+ initDrusillaModelNoTrain(5,-5).
+
+
+%% Successful Tests
+
+test(approx_KFN_InitDrusillaNoTrain_Normal) :-
+ initDrusillaModelNoTrain(5,5).
+
+test(approx_KFN_InitDrusillaNoTrain_Alternative) :-
+ initDrusillaModelNoTrain(10,3).
+
+:- end_tests(initDrusillaModelNoTrain).
+
+
+
+%%
+%% TESTING predicate initDrusillaModelWithTrain/4
+%%
+:- begin_tests(initDrusillaModelWithTrain).
+
+%% Failure Tests
+
+test(approx_KFN_InitDrusillaWithTrain_Negative_L, fail) :-
+ initDrusillaModelWithTrain([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, -5, 5).
+
+test(approx_KFN_InitDrusillaWithTrain_Negative_M, fail) :-
+ initDrusillaModelWithTrain([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, 5, -5).
+
+test(approx_KFN_InitDrusillaWithTrain_Normal,
+ [error(_,system_error('DrusillaSelect::Train(): l and m are too large! Choose smaller values. l*m must be smaller than the number of points in the dataset.'))]) :-
+ initDrusillaModelWithTrain([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, 5, 5).
+
+
+%% Successful Tests
+
+test(approx_KFN_InitDrusillaWithTrain_Normal) :-
+ initDrusillaModelWithTrain([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, 2, 2).
+
+test(approx_KFN_InitDrusillaWithTrain_Alternative) :-
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ initDrusillaModelWithTrain(Data, 4, 5, 2).
+
+:- end_tests(initDrusillaModelWithTrain).
+
+
%%
-%% TESTING predicate predicate/10
+%% TESTING predicate searchDrusilla/7
+%%
+:- begin_tests(searchDrusilla).
+
+%% Failure Tests
+
+test(approx_KFN_SearchDrusilla_Negative_K, fail) :-
+ reset_DrusillaModel_WithTrain,
+ searchDrusilla([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, -2, _, _, _, _).
+
+test(approx_KFN_SearchDrusilla_Diffrent_Query_Dims, [error(_,system_error('subtraction: incompatible matrix dimensions: 4x1 and 3x1'))]) :-
+ reset_DrusillaModel_WithTrain,
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ searchDrusilla(Data, 4, 2, _, _, _, _).
+
+test(approx_KFN_SearchDrusilla_Befor_Train, [error(_,system_error('The Model is not Trained!'))]) :-
+ reset_DrusillaModel_NoTrain,
+ searchDrusilla([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, 2, _, _, _, _).
+
+
+%% Successful Tests
+
+test(approx_KFN_SearchDrusilla_Normal) :-
+ reset_DrusillaModel_WithTrain,
+ searchDrusilla([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, 2, NeighborsList, _, DistancesList, _),
+ print('\nNeighbors: '),
+ print(NeighborsList),
+ print('\nDistances: '),
+ print(DistancesList).
+
+test(approx_KFN_SearchDrusilla_Alternative) :-
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ initDrusillaModelWithTrain(Data, 4, 2, 5),
+ searchDrusilla(Data, 4, 3, NeighborsList, _, DistancesList, _),
+ print('\nNeighbors: '),
+ print(NeighborsList),
+ print('\nDistances: '),
+ print(DistancesList).
+
+:- end_tests(searchDrusilla).
+
+
+
+%% TESTING predicate trainDrusilla/4
%%
-:- begin_tests(predicate).
+:- begin_tests(trainDrusilla).
%% 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(approx_KFN_TrainDrusilla_Negative_L, fail) :-
+ reset_DrusillaModel_NoTrain,
+ trainDrusilla([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, -5, 5).
+
+test(approx_KFN_TrainDrusilla_Negative_M, fail) :-
+ reset_DrusillaModel_NoTrain,
+ trainDrusilla([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, 5, -5).
+
+test(approx_KFN_TrainDrusilla_Too_Large_M_And_L,
+ [error(_,system_error('DrusillaSelect::Train(): l and m are too large! Choose smaller values. l*m must be smaller than the number of points in the dataset.'))]) :-
+ reset_DrusillaModel_NoTrain,
+ trainDrusilla([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, 5, 5).
+
+
+%% Successful Tests
+
+test(approx_KFN_TrainDrusilla_Normal) :-
+ reset_DrusillaModel_NoTrain,
+ trainDrusilla([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, 2, 1).
+
+test(approx_KFN_TrainDrusilla_Alternative) :-
+ reset_DrusillaModel_NoTrain,
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ trainDrusilla(Data, 4, 2, 3).
+
+:- end_tests(trainDrusilla).
+
+
+
+%%
+%% TESTING predicate initQDAFNModelNoTrain/2
+%%
+:- begin_tests(initQDAFNModelNoTrain).
+
+%% Failure Tests
+
+test(approx_KFN_InitQDAFNNoTrain_Negative_L, fail) :-
+ initQDAFNModelNoTrain(-5,5).
+
+test(approx_KFN_InitQDAFNNoTrain_Negative_M, fail) :-
+ initQDAFNModelNoTrain(5,-5).
+
+
+%% Successful Tests
+
+test(approx_KFN_InitQDAFNNoTrain_Normal) :-
+ initQDAFNModelNoTrain(5,5).
+
+test(approx_KFN_InitQDAFNNoTrain_Alternative) :-
+ initQDAFNModelNoTrain(10,3).
+
+:- end_tests(initQDAFNModelNoTrain).
+
+
+
+%%
+%% TESTING predicate initQDAFNModelWithTrain/4
+%%
+:- begin_tests(initQDAFNModelWithTrain).
+
+%% Failure Tests
+
+test(approx_KFN_InitQDAFNWithTrain_Negative_L, fail) :-
+ initQDAFNModelWithTrain([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, -5, 5).
+
+test(approx_KFN_InitQDAFNWithTrain_Negative_M, fail) :-
+ initQDAFNModelWithTrain([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, 5, -5).
+
+test(approx_KFN_InitQDAFNWithTrain_Too_Large_L_And_M, [error(_,system_error('l*m must be smaller than the number of points in the dataset.'))]) :-
+ initQDAFNModelWithTrain([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, 5, 5).
+
+
+%% Successful Tests
+
+test(approx_KFN_InitQDAFNWithTrain_Normal) :-
+ initQDAFNModelWithTrain([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, 1, 2).
+
+test(approx_KFN_InitQDAFNWithTrain_Alternative) :-
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ initQDAFNModelWithTrain(Data, 4, 5, 2).
+
+:- end_tests(initQDAFNModelWithTrain).
+
+
+
+%%
+%% TESTING predicate searchQDAFN/7
+%%
+:- begin_tests(searchQDAFN).
+
+%% Failure Tests
+
+test(approx_KFN_SearchQDAFN_Negative_K, fail) :-
+ reset_QDAFNModel_WithTrain,
+ searchQDAFN([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, -2, _, _, _, _).
+
+test(approx_KFN_SearchQDAFN_Diffrent_Query_Dims, [error(_,system_error('dot(): objects must have the same number of elements'))]) :-
+ reset_QDAFNModel_WithTrain,
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ searchQDAFN(Data, 4, 2, _, _, _, _).
+
+test(approx_KFN_SearchQDAFN_Befor_Train, [error(_,system_error('The Model is not Trained!'))]) :-
+ reset_QDAFNModel_NoTrain,
+ searchQDAFN([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, 2, _, _, _, _).
+
+test(approx_KFN_SearchQDAFN_Too_Large_K, [error(_,system_error('QDAFN::Search(): requested k is greater than value of m!'))]) :-
+ reset_QDAFNModel_WithTrain,
+ searchQDAFN([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, 3, _, _, _, _).
+
+
+%% Successful Tests
+
+test(approx_KFN_SearchQDAFN_Normal) :-
+ reset_QDAFNModel_WithTrain,
+ searchQDAFN([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, 2, NeighborsList, _, DistancesList, _),
+ print('\nNeighbors: '),
+ print(NeighborsList),
+ print('\nDistances: '),
+ print(DistancesList).
+
+test(approx_KFN_SearchQDAFN_Alternative) :-
+ open('src/data_csv/iris2.csv', read, File),
+ take_csv_row(File, skipFirstRow,10, Data),
+ initQDAFNModelWithTrain(Data, 4, 2, 3),
+ searchQDAFN(Data, 4, 2, NeighborsList, _, DistancesList, _),
+ print('\nNeighbors: '),
+ print(NeighborsList),
+ print('\nDistances: '),
+ print(DistancesList).
+
+:- end_tests(searchQDAFN).
+
+
+
+%% TESTING predicate trainQDAFN/4
+%%
+:- begin_tests(trainQDAFN).
+
+%% Failure Tests
+
+test(approx_KFN_TrainQDAFN_Negative_L, fail) :-
+ reset_QDAFNModel_NoTrain,
+ trainQDAFN([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, -5, 5).
+
+test(approx_KFN_TrainQDAFN_Negative_M, fail) :-
+ reset_QDAFNModel_NoTrain,
+ trainQDAFN([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, 5, -5).
-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(approx_KFN_TrainQDAFN_Too_Large_M_And_L,
+ [error(_,system_error('l*m must be smaller than the number of points in the dataset.'))]) :-
+ reset_QDAFNModel_NoTrain,
+ trainQDAFN([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, 5, 5).
%% 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(approx_KFN_TrainQDAFN_Normal) :-
+ reset_QDAFNModel_NoTrain,
+ trainQDAFN([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, 2, 1).
-test(testDescription4, [true(Error =:= 0.9797958971132711)]) :-
- reset_Model_No_Train(perceptron),
+test(approx_KFN_TrainQDAFN_Alternative) :-
+ reset_QDAFNModel_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).
+ trainQDAFN(Data, 4, 2, 3).
-:- end_tests(predicate).
+:- end_tests(trainQDAFN).
run_approx_kfn_tests :-
run_tests.
diff --git a/test_all.pl b/test_all.pl
index a7b0293..f23cb27 100644
--- a/test_all.pl
+++ b/test_all.pl
@@ -12,7 +12,7 @@
:- use_module('src/methods/adaboost/adaboost_test.pl').
-%%:- use_module('src/methods/approx_kfn/approx_kfn_test.pl').
+:- use_module('src/methods/approx_kfn/approx_kfn_test.pl').
%%:- use_module('src/methods/bayesian_linear_regression/bayesian_linear_regression_test.pl').
--
GitLab