diff --git a/src/methods/linear_regression/linear_regression.cpp b/src/methods/linear_regression/linear_regression.cpp
index e71c0aa021478f52ce8aaf0a4b7d8ec376d9be7e..1042df5bb00751b6753625d6078a59476eafde5d 100644
--- a/src/methods/linear_regression/linear_regression.cpp
+++ b/src/methods/linear_regression/linear_regression.cpp
@@ -16,6 +16,7 @@ using namespace mlpack::regression;
// Global Variable of the LinearRegression object so it can be accessed from all functions
LinearRegression regressor;
+int trainingDimension = 0;
// input: const arma::mat &data,
// const arma::rowvec &responses,
@@ -27,19 +28,29 @@ void initModel(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowN
double lambda,
SP_integer bool_intercept)
{
- if(dataMatSize / dataMatRowNum != responsesArrSize)
- {
- cout << "Target dim doesnt fit to the Data dim" << endl;
- return;
- }
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+ // check if labels and weights fit the data
+ if (data.n_cols != responsesArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of labels!");
+ return;
+ }
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
- regressor = LinearRegression(data, responsesVector, lambda, bool_intercept);
+ try
+ {
+ regressor = LinearRegression(data, responsesVector, lambda, bool_intercept);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ trainingDimension = data.n_rows;
}
// input: const arma::mat &data,
@@ -54,19 +65,35 @@ void initModelWithWeights(float *dataMatArr, SP_integer dataMatSize, SP_integer
double lambda,
SP_integer bool_intercept)
{
- if(dataMatSize / dataMatRowNum != responsesArrSize)
- {
- cout << "Target dim doesnt fit to the Data dim" << endl;
- return;
- }
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+ // check if labels and weights fit the data
+ if (data.n_cols != responsesArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of labels!");
+ return;
+ }
+ if (data.n_cols != weightsArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of weights!");
+ return;
+ }
+
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
rowvec weightsVector = convertArrayToRowvec(weightsArr, weightsArrSize);
- regressor = LinearRegression(data, responsesVector, weightsVector, lambda, bool_intercept);
+ try
+ {
+ regressor = LinearRegression(data, responsesVector, weightsVector, lambda, bool_intercept);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
+ trainingDimension = data.n_rows;
}
// input: const arma::mat &points,
@@ -75,19 +102,29 @@ void initModelWithWeights(float *dataMatArr, SP_integer dataMatSize, SP_integer
double computeError(float *pointsMatArr, SP_integer pointsMatSize, SP_integer pointsMatRowNum,
float *responsesArr, SP_integer responsesArrSize)
{
- if(pointsMatSize / pointsMatRowNum != responsesArrSize)
- {
- cout << "Target dim doesnt fit to the Data dim" << endl;
- return 0;
- }
// convert the Prolog array to arma::mat
mat points = convertArrayToMat(pointsMatArr, pointsMatSize, pointsMatRowNum);
+ // check if labels fit the data
+ if (points.n_cols != responsesArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of labels!");
+ return 0.0;
+ }
+
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
// run the model function
- return regressor.ComputeError(points, responsesVector);
+ try
+ {
+ return regressor.ComputeError(points, responsesVector);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return 0.0;
+ }
}
// input:
@@ -95,7 +132,7 @@ double computeError(float *pointsMatArr, SP_integer pointsMatSize, SP_integer po
void parameters(float **bArr, SP_integer *bArrSize)
{
// create the ReturnVector
- rowvec bReturnVector = regressor.Parameters();
+ vec bReturnVector = regressor.Parameters();
// return the Vector
returnVectorInformation(bReturnVector, bArr, bArrSize);
@@ -106,7 +143,7 @@ void parameters(float **bArr, SP_integer *bArrSize)
void modifyParameters(float *newBArr, SP_integer newBArrSize)
{
// convert the Prolog array to arma::rowvec
- vec newBVector = convertArrayToRowvec(newBArr, newBArrSize);
+ vec newBVector = convertArrayToColvec(newBArr, newBArrSize);
// get parameter b vector ref
@@ -124,13 +161,27 @@ void predict(float *pointsMatArr, SP_integer pointsMatSize, SP_integer pointsMat
{
// convert the Prolog array to arma::mat
mat points = convertArrayToMat(pointsMatArr, pointsMatSize, pointsMatRowNum);
+ if (points.n_rows != trainingDimension)
+ {
+ raisePrologSystemExeption("The given Points have a diffrent Dimension than the trained Model!");
+ return;
+ }
+
// create the ReturnVector
rowvec predicReturnVector;
// run the predict function on the model
- regressor.Predict(points, predicReturnVector);
+ try
+ {
+ regressor.Predict(points, predicReturnVector);
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return;
+ }
// return the Vector
@@ -145,19 +196,30 @@ double train(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowNum
float *responsesArr, SP_integer responsesArrSize,
SP_integer bool_intercept)
{
- if(dataMatSize / dataMatRowNum != responsesArrSize)
- {
- cout << "Target dim doesnt fit to the Data dim" << endl;
- return 0;
- }
// convert the Prolog array to arma::mat
- mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+ mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+ // check if labels fit the data
+ if (data.n_cols != responsesArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of labels!");
+ return 0.0;
+ }
+
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
// run the model train function
- return regressor.Train(data, responsesVector, (bool_intercept == 1));
+ try
+ {
+ return regressor.Train(data, responsesVector, (bool_intercept == 1));
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return 0.0;
+ }
+ trainingDimension = data.n_rows;
}
// input: const arma::mat &data,
@@ -170,17 +232,33 @@ double trainWithWeights(float *dataMatArr, SP_integer dataMatSize, SP_integer da
float *weightsArr, SP_integer weightsArrSize,
SP_integer bool_intercept)
{
- if(dataMatSize / dataMatRowNum != responsesArrSize)
- {
- cout << "Target dim doesnt fit to the Data dim" << endl;
- return 0;
- }
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
+ // check if labels and weights fit the data
+ if (data.n_cols != responsesArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of labels!");
+ return 0.0;
+ }
+ if (data.n_cols != weightsArrSize)
+ {
+ raisePrologSystemExeption("The number of data points does not match the number of weights!");
+ return 0.0;
+ }
+
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
rowvec weightsVector = convertArrayToRowvec(weightsArr, weightsArrSize);
// run the model train function
- return regressor.Train(data, responsesVector, weightsVector, (bool_intercept == 1));
+ try
+ {
+ return regressor.Train(data, responsesVector, weightsVector, (bool_intercept == 1));
+ }
+ catch(const std::exception& e)
+ {
+ raisePrologSystemExeption(e.what());
+ return 0.0;
+ }
+ trainingDimension = data.n_rows;
}
diff --git a/src/methods/linear_regression/linear_regression_test.pl b/src/methods/linear_regression/linear_regression_test.pl
index f3b920ccda332eadbfaec09e83e70f268e4f36fa..507c2546697122a8889f6472968ce7b9dfc37a76 100644
--- a/src/methods/linear_regression/linear_regression_test.pl
+++ b/src/methods/linear_regression/linear_regression_test.pl
@@ -21,13 +21,14 @@ reset_Model_WithWeights :-
%% Failure Tests
-test(linear_regression_InitModel_Too_Few_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModel_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_initModel([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], 0.0, 1).
-test(linear_regression_InitModel_Too_Many_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModel_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_initModel([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], 0.0, 1).
-test(linear_regression_InitModel_Too_Many_Labelclasses, [error(_, system_error('Error'))]) :-
+%% doesnt cause error
+test(linear_regression_InitModel_Too_Many_Labelclasses) :-
linear_regression_initModel([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], 0.0, 1).
@@ -52,20 +53,21 @@ test(linear_regression_InitModel_CSV_Input) :-
%% Failure Tests
-test(linear_regression_InitModelWithWeights_Too_Few_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModelWithWeights_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_initModelWithWeights([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], [0.2,0.1,0.2,0.5], 0.0, 1).
-test(linear_regression_InitModelWithWeights_Too_Many_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModelWithWeights_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_initModelWithWeights([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], [0.2,0.1,0.2,0.5], 0.0, 1).
-test(linear_regression_InitModelWithWeights_Too_Many_Labelclasses, [error(_, system_error('Error'))]) :-
+%% doesnt cause error
+test(linear_regression_InitModelWithWeights_Too_Many_Labelclasses) :-
linear_regression_initModelWithWeights([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], [0.2,0.1,0.2,0.5], 0.0, 1).
-test(linear_regression_InitModelWithWeights_Too_Few_Weights, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModelWithWeights_Too_Few_Weights, [error(_, system_error('The number of data points does not match the number of weights!'))]) :-
linear_regression_initModelWithWeights([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], [0.2,0.5], 0.0, 1).
-test(linear_regression_InitModelWithWeights_Too_Many_Weights, [error(_, system_error('Error'))]) :-
+test(linear_regression_InitModelWithWeights_Too_Many_Weights, [error(_, system_error('The number of data points does not match the number of weights!'))]) :-
linear_regression_initModelWithWeights([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], [0.2,0.1,0.2,0.1,0.2,0.5], 0.0, 1).
@@ -90,15 +92,16 @@ test(linear_regression_InitModelWithWeights_CSV_Input) :-
%% Failure Tests
-test(linear_regression_ComputeError_Too_Few_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_ComputeError_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
reset_Model,
linear_regression_computeError([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_regression_ComputeError_Too_Many_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_ComputeError_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
reset_Model,
linear_regression_computeError([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_regression_ComputeError_Too_Many_Labelclasses, [error(_, system_error('Error'))]) :-
+%% doesnt cause error
+test(linear_regression_ComputeError_Too_Many_Labelclasses) :-
reset_Model,
linear_regression_computeError([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], _).
@@ -143,21 +146,21 @@ test(linear_regression_Parameters_Normal_Use) :-
%% Failure Tests
-test(linear_regression_ModifyParameters_Too_Few_Parameters, [error(_, system_error('Error'))]) :-
+test(linear_regression_ModifyParameters_Too_Few_Parameters) :-
reset_Model,
linear_regression_modifyParameters([3,4]),
linear_regression_parameters(Parameters),
print('\nParameters: '),
print(Parameters).
-test(linear_regression_ModifyParameters_Too_Many_Parameters, [error(_, system_error('Error'))]) :-
+test(linear_regression_ModifyParameters_Too_Many_Parameters) :-
reset_Model,
linear_regression_modifyParameters([1,2,1,2,3,4]),
linear_regression_parameters(Parameters),
print('\nParameters: '),
print(Parameters).
-test(linear_regression_ModifyParameters_Bad_Parameter_Value, [error(_, system_error('Error'))]) :-
+test(linear_regression_ModifyParameters_Bad_Parameter_Value) :-
reset_Model,
linear_regression_modifyParameters([1,2,-3,4]),
linear_regression_parameters(Parameters),
@@ -185,7 +188,7 @@ test(linear_regression_ModifyParameters_Normal_Use) :-
%% Failure Tests
-test(linear_regression_Predict_Diffrent_Dims_Than_Trained) :-
+test(linear_regression_Predict_Diffrent_Dims_Than_Trained, [error(_, system_error('The given Points have a diffrent Dimension than the trained Model!'))]) :-
reset_Model,
linear_regression_predict([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, _).
@@ -215,13 +218,14 @@ test(linear_regression_Predict_Normal_Use) :-
%% Failure Tests
-test(linear_regression_Train_Too_Few_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_Train_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_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], 1, _).
-test(linear_regression_Train_Too_Many_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_Train_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_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], 1, _).
-test(linear_regression_Train_Too_Many_Labelclasses, [error(_, system_error('Error'))]) :-
+%% doesnt cause error
+test(linear_regression_Train_Too_Many_Labelclasses) :-
linear_regression_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], 1, _).
@@ -250,20 +254,21 @@ test(linear_regression_Train_CSV_Input) :-
%% Failure Tests
-test(linear_regression_TrainWithWeights_Too_Few_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_TrainWithWeights_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_trainWithWeights([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], [0.2,0.1,0.2,0.5], 1 ,_).
-test(linear_regression_TrainWithWeights_Too_Many_Labels, [error(_, system_error('Error'))]) :-
+test(linear_regression_TrainWithWeights_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
linear_regression_trainWithWeights([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], [0.2,0.1,0.2,0.5], 1 ,_).
-test(linear_regression_TrainWithWeights_Too_Many_Labelclasses, [error(_, system_error('Error'))]) :-
+%% doesnt cause error
+test(linear_regression_TrainWithWeights_Too_Many_Labelclasses) :-
linear_regression_trainWithWeights([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], [0.2,0.1,0.2,0.5], 1 ,_).
-test(linear_regression_TrainWithWeights_Too_Few_Weights, [error(_, system_error('Error'))]) :-
+test(linear_regression_TrainWithWeights_Too_Few_Weights, [error(_, system_error('The number of data points does not match the number of weights!'))]) :-
linear_regression_trainWithWeights([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], [0.2,0.5], 1 ,_).
-test(linear_regression_TrainWithWeights_Too_Many_Weights, [error(_, system_error('Error'))]) :-
+test(linear_regression_TrainWithWeights_Too_Many_Weights, [error(_, system_error('The number of data points does not match the number of weights!'))]) :-
linear_regression_trainWithWeights([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], [0.2,0.1,0.2,0.1,0.2,0.5], 1 ,_).