Skip to content
Snippets Groups Projects
Commit a42fdf0d authored by Jakhes's avatar Jakhes
Browse files

Updating lars tests

parent 5f09d318
Branches
No related tags found
No related merge requests found
Show whitespace changes
Inline Side-by-side
src/methods/lars/lars.cpp
+ 102
− 18
View file @ a42fdf0d
......@@ -17,6 +17,7 @@ using namespace mlpack::regression;
// Global Variable of the Lars regressor object so it can be accessed from all functions
LARS regressor;
bool isModelTrained = false;
// input: const bool useCholesky,
// const double lambda1,
......@@ -27,6 +28,7 @@ void initModelNoDataNoGram(SP_integer useCholesky,
double lambda1, double lambda2, double tol)
{
regressor = new LARS((useCholesky == 1), lambda1, lambda2, tol);
isModelTrained = false;
}
// input: const bool useCholesky,
......@@ -42,6 +44,7 @@ void initModelNoDataWithGram(SP_integer useCholesky,
mat gramMatrix = convertArrayToMat(gramArr, gramSize, gramRowNum);
regressor = new LARS((useCholesky == 1), gramMatrix, lambda1, lambda2, tol);
isModelTrained = false;
}
// input: const arma::mat & dataMatrix,
......@@ -58,19 +61,30 @@ void initModelWithDataNoGram(float *dataMatArr, SP_integer dataMatSize, SP_integ
SP_integer useCholesky,
double lambda1, double lambda2, double tol)
{
if(dataMatSize / dataMatRowNum != responsesArrSize)
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
// check if labels fit the data
if (data.n_cols != responsesArrSize)
{
cout << "Target dim doesnt fit to the Data dim" << endl;
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return;
}
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
try
{
regressor = new LARS(data, responsesVector, (transposeData == 1), (useCholesky == 1), lambda1, lambda2, tol);
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return;
}
isModelTrained = true;
}
// input: const arma::mat & dataMatrix,
// const arma::rowvec & responses,
......@@ -88,21 +102,32 @@ void initModelWithDataWithGram(float *dataMatArr, SP_integer dataMatSize, SP_int
float *gramMatArr, SP_integer gramMatSize, SP_integer gramMatRowNum,
double lambda1, double lambda2, double tol)
{
if(dataMatSize / dataMatRowNum != responsesArrSize)
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
// check if labels fit the data
if (data.n_cols != responsesArrSize)
{
cout << "Target dim doesnt fit to the Data dim" << endl;
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return;
}
// convert the Prolog array to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
// convert the Prolog array to arma::rowvec
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
// convert the Prolog array to arma::mat
mat gram = convertArrayToMat(gramMatArr, gramMatSize, gramMatRowNum);
try
{
regressor = new LARS(data, responsesVector, (transposeData == 1), (useCholesky == 1), gram, lambda1, lambda2, tol);
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return;
}
isModelTrained = true;
}
// input:
// output: const std::vector<size_t>&
......@@ -120,8 +145,14 @@ void activeSet(float **activeSetArr, SP_integer *activeSetSize)
// output: arma::vec&
void beta(float **betaArr, SP_integer *betaArrSize)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return;
}
// create the ReturnVector
rowvec betaReturnVector = regressor.Beta();
vec betaReturnVector = regressor.Beta();
// return the Vector
returnVectorInformation(betaReturnVector, betaArr, betaArrSize);
......@@ -131,6 +162,11 @@ void beta(float **betaArr, SP_integer *betaArrSize)
// output: std::vector<arma::vec>&
void betaPath(float **betaPathArr, SP_integer *betaPathColNum, SP_integer *betaPathRowNum)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return;
}
// get the betaPath matrix
vector<vec> matrix = regressor.BetaPath();
......@@ -150,6 +186,11 @@ void betaPath(float **betaPathArr, SP_integer *betaPathColNum, SP_integer *betaP
double computeError(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowNum,
float *responsesArr, SP_integer responsesArrSize, SP_integer rowMajor)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return 0.0;
}
if(dataMatSize / dataMatRowNum != responsesArrSize)
{
cout << "Target dim doesnt fit to the Data dim" << endl;
......@@ -161,13 +202,28 @@ double computeError(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMa
rowvec responsesVector = convertArrayToRowvec(responsesArr, responsesArrSize);
// run the model function
return regressor.ComputeError(data, responsesVector, (rowMajor == 1));
double result;
try
{
result = regressor.ComputeError(data, responsesVector, (rowMajor == 1));
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return 0.0;
}
return result;
}
// input:
// output: std::vector<double>&
void lambdaPath(float **lambdaPathArr, SP_integer *lambdaPathSize)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return;
}
std::vector<double> lambdaPathVec = regressor.LambdaPath();
// give back the sizes and the converted results as arrays
......@@ -181,6 +237,11 @@ void lambdaPath(float **lambdaPathArr, SP_integer *lambdaPathSize)
// output: arma::mat& upper triangular cholesky factor
void matUtriCholFactor(float **factorMatArr, SP_integer *factorMatColNum, SP_integer *factorMatRowNum)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return;
}
// create the ReturnMat
mat factorReturnMat = regressor.MatUtriCholFactor();
......@@ -194,13 +255,26 @@ void matUtriCholFactor(float **factorMatArr, SP_integer *factorMatColNum, SP_int
// output:
void predict(float *pointsMatArr, SP_integer pointsMatSize, SP_integer pointsMatRowNum, float **predicArr, SP_integer *predicArrSize, SP_integer rowMajor)
{
if (!isModelTrained)
{
raisePrologSystemExeption("The Model is not trained!");
return;
}
// convert the Prolog array to arma::mat
mat points = convertArrayToMat(pointsMatArr, pointsMatSize, pointsMatRowNum);
// create the ReturnVector
rowvec predicReturnVector;
try
{
regressor.Predict(points, predicReturnVector, (rowMajor == 1));
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return;
}
// return the Vector
......@@ -233,10 +307,20 @@ double train(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMatRowNum
// run the model function
double error = regressor.Train(data, responsesVector, betaReturnVector, (transposeData == 1));
double error;
try
{
error = regressor.Train(data, responsesVector, betaReturnVector, (transposeData == 1));
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return 0.0;
}
// return the Vector
returnVectorInformation(betaReturnVector, betaArr, betaArrSize);
isModelTrained = true;
return error;
}
src/methods/lars/lars_test.pl
+ 33
− 32
View file @ a42fdf0d
......@@ -73,20 +73,21 @@ test(lars_InitModelWithDataNoGram_Negative_Tolerance, fail) :-
lars_initModelWithDataNoGram([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, 0, 0.0, 0.0, -1.0e-16).
test(lars_InitModelWithDataNoGram_Too_Few_Labels, [error(_,system_error('Error'))]) :-
lars_initModelWithDataNoGram([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, 0.0, 0.0, -1.0e-16).
test(lars_InitModelWithDataNoGram_Too_Few_Labels, [error(_,system_error('The number of data points does not match the number of labels!'))]) :-
lars_initModelWithDataNoGram([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, 0.0, 0.0, 1.0e-16).
test(lars_InitModelWithDataNoGram_Too_Many_Labels, [error(_,system_error('Error'))]) :-
lars_initModelWithDataNoGram([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, 0.0, 0.0, -1.0e-16).
test(lars_InitModelWithDataNoGram_Too_Many_Labels, [error(_,system_error('The number of data points does not match the number of labels!'))]) :-
lars_initModelWithDataNoGram([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, 0.0, 0.0, 1.0e-16).
test(lars_InitModelWithDataNoGram_Too_Many_Labelclasses, [error(_,system_error('Error'))]) :-
lars_initModelWithDataNoGram([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, 0.0, 0.0, -1.0e-16).
%% doesnt cause error
test(lars_InitModelWithDataNoGram_Too_Many_Labelclasses) :-
lars_initModelWithDataNoGram([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, 0, 0.0, 0.0, 1.0e-16).
%% Successful Tests
test(lars_InitModelWithDataNoGram_Normal_Use) :-
lars_initModelWithDataNoGram([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, 0, 0.0, 0.0, 1.0e-16).
lars_initModelWithDataNoGram([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], 1, 0, 0.0, 0.0, 1.0e-16).
test(lars_InitModelWithDataNoGram_Alternative_Use) :-
lars_initModelWithDataNoGram([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], 1, 1, 1.0, -1.0, 1.0e-16).
......@@ -106,24 +107,24 @@ test(lars_InitModelWithDataWithGram_Negative_Tolerance, fail) :-
lars_initModelWithDataWithGram([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, 0, [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, -1.0e-16).
test(lars_InitModelWithDataWithGram_Too_Few_Labels, [error(_,system_error('Error'))]) :-
lars_initModelWithDataWithGram([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, [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, -1.0e-16).
test(lars_InitModelWithDataWithGram_Too_Few_Labels, [error(_,system_error('The number of data points does not match the number of labels!'))]) :-
lars_initModelWithDataWithGram([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, [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, 1.0e-16).
test(lars_InitModelWithDataWithGram_Too_Many_Labels, [error(_,system_error('Error'))]) :-
lars_initModelWithDataWithGram([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, [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, -1.0e-16).
test(lars_InitModelWithDataWithGram_Too_Many_Labels, [error(_,system_error('The number of data points does not match the number of labels!'))]) :-
lars_initModelWithDataWithGram([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, [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, 1.0e-16).
test(lars_InitModelWithDataWithGram_Too_Many_Labelclasses, [error(_,system_error('Error'))]) :-
lars_initModelWithDataWithGram([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, [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, -1.0e-16).
lars_initModelWithDataWithGram([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, 0, [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, 1.0e-16).
test(lars_InitModelWithDataWithGram_Diffrent_Dimensions, [error(_,system_error('Error'))]) :-
lars_initModelWithDataWithGram([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, 0, [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, 0.0, 0.0, -1.0e-16).
lars_initModelWithDataWithGram([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], 1, 0, [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, 0.0, 0.0, 1.0e-16).
%% Successful Tests
test(lars_InitModelWithDataWithGram_Normal_Use) :-
lars_initModelWithDataWithGram([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, 0, [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, 1.0e-16).
lars_initModelWithDataWithGram([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], 1, 0, [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, 1.0e-16).
test(lars_InitModelWithDataWithGram_Alternative_Use) :-
lars_initModelWithDataWithGram([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], 1, 0, [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, 1.0, -1.0, 1.0e-16).
......@@ -139,7 +140,7 @@ test(lars_InitModelWithDataWithGram_Alternative_Use) :-
%% Failure Tests
test(lars_ActiveSet_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_ActiveSet_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_activeSet(_).
......@@ -163,18 +164,18 @@ test(lars_ActiveSet_Normal_Use) :-
%% Failure Tests
test(lars_Beta_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_Beta_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_beta(_).
%% Successful Tests
test(lars_Beta_Normal_Use) :-
reset_Model_WithTrain,
lars_beta(BetaList),
print('\nBeta : '),
print(BetaList).
%%test(lars_Beta_Normal_Use) :-
%% reset_Model_WithTrain,
%% lars_beta(BetaList),
%% print('\nBeta : '),
%% print(BetaList).
:- end_tests(lars_beta).
......@@ -187,7 +188,7 @@ test(lars_Beta_Normal_Use) :-
%% Failure Tests
test(lars_BetaPath_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_BetaPath_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_betaPath(_, _).
......@@ -211,7 +212,7 @@ test(lars_BetaPath_Normal_Use) :-
%% Failure Tests
test(lars_ComputeError_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_ComputeError_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_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, _).
......@@ -262,7 +263,7 @@ test(lars_ComputeError_CSV_Input) :-
%% Failure Tests
test(lars_LambdaPath_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_LambdaPath_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_lambdaPath(_).
......@@ -286,7 +287,7 @@ test(lars_LambdaPath_Normal_Use) :-
%% Failure Tests
test(lars_MatUtriCholFactor_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_MatUtriCholFactor_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_matUtriCholFactor(_, _).
......@@ -310,21 +311,21 @@ test(lars_MatUtriCholFactor_Normal_Use) :-
%% Failure Tests
test(lars_Predict_Before_Train, [error(_,system_error('Error'))]) :-
test(lars_Predict_Before_Train, [error(_,system_error('The Model is not trained!'))]) :-
reset_Model_NoTrain,
lars_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], 3, 0, _).
lars_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], 3, 1, _).
test(lars_Predict_Diffrent_Dims, [error(_,system_error('Error'))]) :-
reset_Model_WithTrain,
lars_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, 0, _).
lars_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, 1, _).
%% Successful Tests
test(lars_Predict_Normal_Use) :-
reset_Model_WithTrain,
lars_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], 3, 0, PredictList),
lars_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], 3, 1, PredictList),
print('\nPrediction: '),
print(PredictList).
......@@ -338,7 +339,7 @@ test(lars_Predict_CSV_Input) :-
reset_Model_WithTrain,
open('src/data_csv/iris2.csv', read, File),
take_csv_row(File, skipFirstRow,10, Data),
lars_predict(Data, 4, 0, PredictList),
lars_predict(Data, 4, 1, PredictList),
print('\nPrediction: '),
print(PredictList).
......@@ -410,4 +411,4 @@ test(lars_Train_CSV_Input) :-
:- end_tests(lars_train).
run_lars_tests :-
run_tests.
\ No newline at end of file
run_tests(lars_predict).
\ No newline at end of file
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment