Skip to content
Snippets Groups Projects
Commit 9debcb47 authored by Jakhes's avatar Jakhes
Browse files

Finishing logistic_regression tests

parent 3b3eba96
Branches
No related tags found
No related merge requests found
Show whitespace changes
Inline Side-by-side
src/methods/logistic_regression/logistic_regression.cpp
+ 76
− 46
View file @ 9debcb47
......@@ -18,6 +18,8 @@ using namespace mlpack::regression;
// Global Variable of the LogisticRegression object so it can be accessed from all functions
LogisticRegression logisticRegression;
int trainedDimension = 0;
// input: const MatType & predictors,
// const arma::Row< size_t > & responses,
......@@ -33,6 +35,13 @@ void initModelNoOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize,
{
// convert the Prolog array to arma::mat
mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
// check if labels fit the data
if (predictors.n_cols != responsesArrSize)
{
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return;
}
// convert the Prolog array to arma::rowvec
arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
......@@ -46,6 +55,7 @@ void initModelNoOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize,
raisePrologSystemExeption(e.what());
return;
}
trainedDimension = predictors.n_rows;
}
......@@ -60,25 +70,44 @@ void initModelNoOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize,
void initModelWithOptimizer(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum,
float *responsesArr, SP_integer responsesArrSize,
char const *optimizer,
double lambda)
double lambda, SP_integer maxIterations, double tolerance, double stepSize, SP_integer batchSize)
{
// convert the Prolog array to arma::mat
mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
if (predictors.n_cols != responsesArrSize)
{
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return;
}
// convert the Prolog array to arma::rowvec
arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
try
{
logisticRegression = LogisticRegression<>(predictors.n_cols, lambda);
if (strcmp(optimizer, "lbfgs") == 0)
{
logisticRegression.Train(predictors,responsesVector);
ens::L_BFGS lbfgsOpt;
lbfgsOpt.MaxIterations() = maxIterations;
lbfgsOpt.MinGradientNorm() = tolerance;
Log::Info << "Training model with L-BFGS optimizer." << endl;
// This will train the model.
logisticRegression.Train(predictors, responsesVector, lbfgsOpt);
}
else if (strcmp(optimizer, "psgd") == 0)
{
logisticRegression.Train(predictors,responsesVector, ens::ParallelSGD(100,64));
ens::SGD<> sgdOpt;
sgdOpt.MaxIterations() = maxIterations;
sgdOpt.Tolerance() = tolerance;
sgdOpt.StepSize() = stepSize;
sgdOpt.BatchSize() = batchSize;
Log::Info << "Training model with SGD optimizer." << endl;
// This will train the model.
logisticRegression.Train(predictors,responsesVector, sgdOpt);
}
else
{
......@@ -90,6 +119,7 @@ void initModelWithOptimizer(float *predictorsMatArr, SP_integer predictorsMatSiz
raisePrologSystemExeption(e.what());
return;
}
trainedDimension = predictors.n_rows;
}
......@@ -103,6 +133,11 @@ void initModelWithOptimizer(float *predictorsMatArr, SP_integer predictorsMatSiz
SP_integer classifyPoint(float *pointArr, SP_integer pointArrSize,
double decisionBoundary)
{
if (pointArrSize != trainedDimension)
{
raisePrologSystemExeption("The given Datapoint has a different Dimension than the trained Model!");
return 0;
}
// convert the Prolog array to arma::rowvec
rowvec pointVector = convertArrayToRowvec(pointArr, pointArrSize);
try
......@@ -130,8 +165,14 @@ void classifyMatrix(float *dataMatArr, SP_integer dataMatSize, SP_integer dataMa
float **probsMatArr, SP_integer *probsMatColNum, SP_integer *probsMatRowNum,
double decisionBoundary)
{
// convert the Prolog arrays to arma::mat
mat data = convertArrayToMat(dataMatArr, dataMatSize, dataMatRowNum);
if (data.n_rows != trainedDimension)
{
raisePrologSystemExeption("The given Datapoints have a different Dimension than the trained Model!");
return;
}
// get the ReturnVector
Row<size_t> labelsReturnVector;
......@@ -172,6 +213,17 @@ double computeAccuracy(float *predictorsMatArr, SP_integer predictorsMatSize, SP
{
// convert the Prolog array to arma::mat
mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
if (predictors.n_cols != responsesArrSize)
{
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return 0.0;
}
if (predictors.n_rows != trainedDimension)
{
raisePrologSystemExeption("The given Datapoints have a different Dimension than the trained Model!");
return 0.0;
}
// convert the Prolog array to arma::rowvec
arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
......@@ -199,57 +251,35 @@ double computeError(float *predictorsMatArr, SP_integer predictorsMatSize, SP_in
{
// convert the Prolog arrays to arma::mat
mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
// convert the Prolog array to arma::rowvec
arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
try
if (predictors.n_cols != responsesArrSize)
{
return logisticRegression.ComputeError(predictors, responsesVector);
raisePrologSystemExeption("The number of data points does not match the number of labels!");
return 0.0;
}
catch(const std::exception& e)
if (predictors.n_rows != trainedDimension)
{
raisePrologSystemExeption(e.what());
raisePrologSystemExeption("The given Datapoints have a different Dimension than the trained Model!");
return 0.0;
}
}
// input: const MatType & predictors,
// const arma::Row< size_t > & responses,
// OptimizerType & optimizer,
// CallbackTypes &&... callbacks
// output:
// description:
// Train the logistic_regression model on the given input data.
//
void train(float *predictorsMatArr, SP_integer predictorsMatSize, SP_integer predictorsMatRowNum,
float *responsesArr, SP_integer responsesArrSize,
char const *optimizer)
{
// convert the Prolog array to arma::mat
mat predictors = convertArrayToMat(predictorsMatArr, predictorsMatSize, predictorsMatRowNum);
// convert the Prolog array to arma::rowvec
arma::Row<size_t> responsesVector = convertArrayToVec(responsesArr, responsesArrSize);
double error = 0.0;
try
{
if (strcmp(optimizer, "lbfgs") == 0)
{
logisticRegression.Train(predictors,responsesVector);
}
else if (strcmp(optimizer, "psgd") == 0)
{
logisticRegression.Train(predictors,responsesVector, ens::ParallelSGD(100,64));
}
else
{
raisePrologDomainExeption(optimizer, 4, "The given Optimizer is unkown!", "initModelWithOptimizer");
}
error = logisticRegression.ComputeError(predictors, responsesVector);
}
catch(const std::exception& e)
{
raisePrologSystemExeption(e.what());
return;
return 0.0;
}
if (error != error)
{
raisePrologSystemExeption("Error contains nan Value!");
return 0.0;
}
return error;
}
\ No newline at end of file
src/methods/logistic_regression/logistic_regression.pl
+ 23
− 29
View file @ 9debcb47
:- module(logistic_regression, [ logistic_regression_initModelNoOptimizer/4,
logistic_regression_initModelWithOptimizer/5,
logistic_regression_initModelWithOptimizer/9,
logistic_regression_classifyPoint/3,
logistic_regression_classifyMatrix/6,
logistic_regression_computeAccuracy/5,
logistic_regression_computeError/4,
logistic_regression_train/4]).
logistic_regression_computeError/4]).
%% requirements of library(struct)
:- load_files(library(str_decl),
......@@ -34,6 +33,7 @@
%% Initialize the logistic_regression model without specifing a optimizer.
%%
logistic_regression_initModelNoOptimizer(DataList, DataRows, ResponsesList, Lambda) :-
Lambda >= 0.0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(ResponsesList, array(Ysize, Y)),
initModelNoOptimizerI(X, Xsize, Xrownum, Y, Ysize, Lambda).
......@@ -47,22 +47,31 @@ foreign(initModelNoOptimizer, c, initModelNoOptimizerI( +pointer(float_array), +
%% mat data,
%% vec responses,
%% string optimizer => "lbfgs", "psgd" => lbfgs,
%% float32 lambda => 0.0
%% float32 lambda => 0.0,
%% int maxIterations => 10000,
%% float32 tolerance => 1e-10,
%% float32 stepSize => 0.01,
%% int batchSize => 64
%%
%% --Output--
%%
%% --Description--
%% Initialize the logistic_regression model and specify the optimizer.
%%
logistic_regression_initModelWithOptimizer(DataList, DataRows, ResponsesList, Responses, Lambda) :-
logistic_regression_initModelWithOptimizer(DataList, DataRows, ResponsesList, Responses, Lambda, MaxIterations, Tolerance, StepSize, BatchSize) :-
Lambda >= 0.0,
MaxIterations >= 0,
Tolerance >= 0.0,
StepSize >= 0.0,
BatchSize > 0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(ResponsesList, array(Ysize, Y)),
initModelWithOptimizerI(X, Xsize, Xrownum, Y, Ysize, Responses, Lambda).
initModelWithOptimizerI(X, Xsize, Xrownum, Y, Ysize, Responses, Lambda, MaxIterations, Tolerance, StepSize, BatchSize).
foreign(initModelWithOptimizer, c, initModelWithOptimizerI( +pointer(float_array), +integer, +integer,
+pointer(float_array), +integer,
+string,
+float32)).
+float32, +integer, +float32, +float32, +integer)).
%% --Input--
......@@ -76,6 +85,8 @@ foreign(initModelWithOptimizer, c, initModelWithOptimizerI( +pointer(float_a
%% Classify the given point.
%%
logistic_regression_classifyPoint(PointList, DecisionBoundary, PredicLabel) :-
DecisionBoundary >= 0.0,
DecisionBoundary =< 1.0,
convert_list_to_float_array(PointList, array(Xsize, X)),
classifyPointI(X, Xsize, DecisionBoundary, PredicLabel).
......@@ -96,6 +107,8 @@ foreign(classifyPoint, c, classifyPointI( +pointer(float_array), +integer,
%% Classify the given points, returning the predicted labels for each point.
%%
logistic_regression_classifyMatrix(DataList, DataRows, PredictionList, ProbsList, ZCols, DecisionBoundary) :-
DecisionBoundary >= 0.0,
DecisionBoundary =< 1.0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrows, X)),
classifyMatrixI(X, Xsize, Xrows, Y, Ysize, Z, ZCols, ZRows, DecisionBoundary),
convert_float_array_to_list(Y, Ysize, PredictionList),
......@@ -119,6 +132,8 @@ foreign(classifyMatrix, c, classifyMatrixI( +pointer(float_array), +integer,
%% Compute the accuracy of the model on the given predictors and responses, using the given decision boundary.
%%
logistic_regression_computeAccuracy(DataList, DataRows, ResponsesList, DecisionBoundary, Accuracy) :-
DecisionBoundary >= 0.0,
DecisionBoundary =< 1.0,
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(ResponsesList, array(Ysize, Y)),
computeAccuracyI(X, Xsize, Xrownum, Y, Ysize, DecisionBoundary, Accuracy).
......@@ -149,33 +164,12 @@ foreign(computeError, c, computeErrorI( +pointer(float_array), +integer, +intege
[-float32])).
%% --Input--
%% mat data,
%% vec responses,
%% string optimizer => "lbfgs", "psgd" => lbfgs
%%
%% --Output--
%%
%% --Description--
%% Train the logistic_regression model on the given input data.
%%
logistic_regression_train(DataList, DataRows, ResponsesList, Optimizer) :-
convert_list_to_float_array(DataList, DataRows, array(Xsize, Xrownum, X)),
convert_list_to_float_array(ResponsesList, array(Ysize, Y)),
trainI(X, Xsize, Xrownum, Y, Ysize, Optimizer).
foreign(train, c, trainI( +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,
classifyMatrix,
computeAccuracy,
computeError,
train]).
computeError]).
:- load_foreign_resource(logistic_regression).
src/methods/logistic_regression/logistic_regression_test.pl
+ 241
− 17
View file @ 9debcb47
......@@ -6,38 +6,262 @@
:- use_module(logistic_regression).
:- use_module('../../helper_files/helper.pl').
reset_Model :-
logistic_regression_initModel(1,0,50,0.0001).
reset_Model_NoOptimizer :-
logistic_regression_initModelNoOptimizer([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).
reset_Model_WithOptimizer :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
%%
%% TESTING predicate predicate/10
%% TESTING predicate logistic_regression_initModelNoOptimizer/4
%%
:- begin_tests(predicate).
:- begin_tests(logistic_regression_initModelNoOptimizer).
%% 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(logistic_regression_InitModelNoOptimizer_Negative_Lambda, fail) :-
logistic_regression_initModelNoOptimizer([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).
test(logistic_regression_InitModelNoOptimizer_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_initModelNoOptimizer([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).
test(logistic_regression_InitModelNoOptimizer_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_initModelNoOptimizer([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).
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, _).
%% doesnt cause error
test(logistic_regression_InitModelNoOptimizer_Too_Many_Labelclasses) :-
logistic_regression_initModelNoOptimizer([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).
%% 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(logistic_regression_InitModelNoOptimizer_Normal_Use) :-
logistic_regression_initModelNoOptimizer([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).
test(testDescription4, [true(Error =:= 0.9797958971132711)]) :-
reset_Model_No_Train(perceptron),
test(logistic_regression_InitModelNoOptimizer_CSV_Input) :-
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).
logistic_regression_initModelNoOptimizer(Data, 4, [0,1,0,1,1,0,1,1,1,0], 0.0).
:- end_tests(logistic_regression_initModelNoOptimizer).
%%
%% TESTING predicate logistic_regression_initModelWithOptimizer/10
%%
:- begin_tests(logistic_regression_initModelWithOptimizer).
%% Failure Tests
test(logistic_regression_InitModelWithOptimizer_Wrong_Optimizer, [error(domain_error('The given Optimizer is unkown!' , wrongInput), _)]) :-
logistic_regression_initModelWithOptimizer([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], wrongInput, 0.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Negative_Lambda, fail) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, -1.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Negative_MaxIteration, fail) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, -1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Negative_Tolerance, fail) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, -1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Negative_StepSize, fail) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, -0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Negative_BatchSize, fail) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, -64).
test(logistic_regression_InitModelWithOptimizer_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
%% doesnt cause error
test(logistic_regression_InitModelWithOptimizer_Too_Many_Labelclasses) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
%% Successful Tests
test(logistic_regression_InitModelWithOptimizer_Normal_Use_LBFGS) :-
logistic_regression_initModelWithOptimizer([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], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_CSV_Input_LBFGS) :-
open('src/data_csv/iris2.csv', read, File),
take_csv_row(File, skipFirstRow,10, Data),
logistic_regression_initModelWithOptimizer(Data, 4, [0,1,0,1,1,0,1,1,1,0], lbfgs, 0.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_Normal_Use_PSGD) :-
logistic_regression_initModelWithOptimizer([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], psgd, 0.0, 1000, 1.0e-10, 0.01, 64).
test(logistic_regression_InitModelWithOptimizer_CSV_Input_PSGD) :-
open('src/data_csv/iris2.csv', read, File),
take_csv_row(File, skipFirstRow,10, Data),
logistic_regression_initModelWithOptimizer(Data, 4, [0,1,0,1,1,0,1,1,1,0], psgd, 0.0, 1000, 1.0e-10, 0.01, 64).
:- end_tests(logistic_regression_initModelWithOptimizer).
%%
%% TESTING predicate logistic_regression_classifyPoint/3
%%
:- begin_tests(logistic_regression_classifyPoint).
%% Failure Tests
test(logistic_regression_ClassifyPoint_Different_Dims, [error(_, system_error('The given Datapoint has a different Dimension than the trained Model!'))]) :-
reset_Model_WithOptimizer,
logistic_regression_classifyPoint([5.1,3.5,1.4,4.9], 0.5, _).
test(logistic_regression_ClassifyPoint_Bad_Boundary_Input, fail) :-
reset_Model_WithOptimizer,
logistic_regression_classifyPoint([5.1,3.5,1.4], -0.5, _),
logistic_regression_classifyPoint([5.1,3.5,1.4], 1.5, _).
%% Successful Tests
test(logistic_regression_ClassifyPoint_Normal_Use) :-
reset_Model_WithOptimizer,
logistic_regression_classifyPoint([5.1,3.5,1.4], 0.5, PredictLabel),
print('\nPredicted Label: '),
print(PredictLabel).
test(logistic_regression_ClassifyPoint_NoOptimizer) :-
reset_Model_NoOptimizer,
logistic_regression_classifyPoint([5.1,3.5,1.4], 0.5, PredictLabel),
print('\nPredicted Label: '),
print(PredictLabel).
:- end_tests(logistic_regression_classifyPoint).
%%
%% TESTING predicate logistic_regression_classifyMatrix/10
%%
:- begin_tests(logistic_regression_classifyMatrix).
%% Failure Tests
test(logistic_regression_ClassifyMatrix_Different_Dims, [error(_, system_error('The given Datapoints have a different Dimension than the trained Model!'))]) :-
reset_Model_WithOptimizer,
logistic_regression_classifyMatrix([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, _, _, _, 0.5).
test(logistic_regression_ClassifyMatrix_Bad_Boundary_Input, fail) :-
reset_Model_WithOptimizer,
logistic_regression_classifyMatrix([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, _, _, _, -0.5),
logistic_regression_classifyMatrix([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, _, _, _, 1.5).
%% Successful Tests
test(logistic_regression_ClassifyMatrix_Normal_Use) :-
reset_Model_WithOptimizer,
logistic_regression_classifyMatrix([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, PredictLabels, ProbsList, _, 0.5),
print('\nPredicted Labels: '),
print(PredictLabels),
print('\nProbabilities: '),
print(ProbsList).
test(logistic_regression_ClassifyMatrix_NoOptimizer) :-
reset_Model_NoOptimizer,
logistic_regression_classifyMatrix([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, PredictLabels, ProbsList, _, 0.5),
print('\nPredicted Labels: '),
print(PredictLabels),
print('\nProbabilities: '),
print(ProbsList).
:- end_tests(logistic_regression_classifyMatrix).
%%
%% TESTING predicate logistic_regression_computeAccuracy/5
%%
:- begin_tests(logistic_regression_computeAccuracy).
%% Failure Tests
test(logistic_regression_ComputeAccuracy_Different_Dims, [error(_, system_error('The given Datapoints have a different Dimension than the trained Model!'))]) :-
reset_Model_WithOptimizer,
logistic_regression_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], 0.5, _).
test(logistic_regression_ComputeAccuracy_Bad_Boundary_Input, fail) :-
reset_Model_WithOptimizer,
logistic_regression_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], -0.5, _),
logistic_regression_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], 1.5, _).
test(logistic_regression_ComputeAccuracy_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_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], 0.5, _).
test(logistic_regression_ComputeAccuracy_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_regression_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], 0.5, _).
%% doesnt cause error
test(logistic_regression_computeAccuracy_Too_Many_Labelclasses) :-
logistic_regression_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], 0.5, _).
%% Successful Tests
test(logistic_regression_ComputeAccuracy_Normal_Use) :-
reset_Model_WithOptimizer,
logistic_regression_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], 0.5, AccuracyList),
print('\nAccuracy: '),
print(AccuracyList).
test(logistic_regression_ComputeAccuracy_NoOptimizer) :-
reset_Model_NoOptimizer,
logistic_regression_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], 0.5, AccuracyList),
print('\nAccuracy: '),
print(AccuracyList).
:- end_tests(logistic_regression_computeAccuracy).
%%
%% TESTING predicate logistic_regression_computeError/4
%%
:- begin_tests(logistic_regression_computeError).
%% Failure Tests
test(logistic_regression_computeError_Different_Dims, [error(_, system_error('The given Datapoints have a different Dimension than the trained Model!'))]) :-
reset_Model_WithOptimizer,
logistic_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], 4, [0,1,0], _).
test(logistic_regression_computeError_Too_Few_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_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(logistic_regression_computeError_Too_Many_Labels, [error(_, system_error('The number of data points does not match the number of labels!'))]) :-
logistic_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(logistic_regression_computeError_Too_Many_Labelclasses, [error(_, system_error('Error contains nan Value!'))]) :-
logistic_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], _).
%% Successful Tests
test(logistic_regression_computeError_Normal_Use) :-
reset_Model_WithOptimizer,
logistic_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], ErrorList),
print('\nError: '),
print(ErrorList).
test(logistic_regression_computeError_NoOptimizer) :-
reset_Model_NoOptimizer,
logistic_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], ErrorList),
print('\nError: '),
print(ErrorList).
:- end_tests(predicate).
:- end_tests(logistic_regression_computeError).
run_logistic_regression_tests :-
run_tests.
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment