public void ThetasPropertyAfterTrainingReturnsTheSameThetasAsTrainerReturns()
{
// arrange
var dataSetMock = new Mock <IDataSet <bool, double> >();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(2);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new[] { 0.0 }),
TrainingSample.Create(false, new[] { 1.0 }) });
const int FeaturesCount = 1;
const double RegularizationParameter = 0.0;
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(dataSetMock.Object);
var classifierTraining = new LogisticRegressionClassifierTraining(FeaturesCount, RegularizationParameter, dataSetMock.Object);
var expectedThetas = classifierTraining.Train();
// act
var thetas = classifier.Thetas;
// assert
Assert.IsTrue(expectedThetas.SequenceEqual(thetas));
}
public void ConstructorDoesNotThrowExceptionIfAllParametersIsGood()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(1, 0.0);
// assert
}
public void ConstructorDoesNotThrowExceptionIfAllParametersIsGood()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(1, 0.0);
// assert
}
public void ConstructorThrowsArgumentOutOfRangeExceptionIfRegularizationParameterIsSmallerThanZero()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(1, -1.0);
// assert
Assert.Fail();
}
public void ConstructorThrowsArgumentOutOfRangeExceptionIfFeaturesCountIsNotPositiveNumber()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(0, 0.0);
// assert
Assert.Fail();
}
public void ConstructorThrowsArgumentOutOfRangeExceptionIfFeaturesCountIsNotPositiveNumber()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(0, 0.0);
// assert
Assert.Fail();
}
public void ConstructorThrowsArgumentOutOfRangeExceptionIfRegularizationParameterIsSmallerThanZero()
{
// arrange
// act
var classifier = new LogisticRegressionClassifier(1, -1.0);
// assert
Assert.Fail();
}
public void TrainThrowsArgumentNullExceptionIfTrainingDataSetIsNull()
{
// arrange
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.Train(null);
// assert
Assert.Fail();
}
public void ThetasPropertyReturnsAllZerosIfClassifierIsNotTrained()
{
// arrange
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
var thetas = classifier.Thetas;
// assert
Assert.IsTrue(thetas.All(d => d == 0));
}
public void TrainDoesNotThrowExceptionOnValidDataSet()
{
// arrange
var dataSetMock = new Mock <IDataSet <bool, double> >();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.Train(dataSetMock.Object);
// assert
}
public void ClassifyThrowsArgumentNullExceptionIfAttributesIsNull()
{
// arrange
var dataSetMock = new Mock<IDataSet<bool, double>>();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
classifier.Train(dataSetMock.Object);
// act
classifier.Classify(null);
// assert
Assert.Fail();
}
public void GetCategoryProbabilityReturnsCorrectProbabilityForTestData1()
{
// arrange
const int FeaturesCount = 2;
const double RegularizationParameter = 0.0; // no regularization
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(this.dataSet1);
// act
var probability = classifier.GetCategoryProbability(true, new[] { 45.0, 85.0 });
// assert
Assert.AreEqual(0.77628055852165889, probability, Delta);
}
public void GetCategoryProbabilityThrowsInvalidOperationExceptionIfClassifierIsNotTrained()
{
// arrange
var dataSetMock = new Mock <IDataSet <bool, double> >();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.GetCategoryProbability(true, new double[1]);
// assert
Assert.Fail();
}
public void GetCategoryProbabilityThrowsArgumentNullExceptionIfAttributesIsNull()
{
// arrange
var dataSetMock = new Mock <IDataSet <bool, double> >();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
classifier.Train(dataSetMock.Object);
// act
classifier.GetCategoryProbability(true, null);
// assert
Assert.Fail();
}
public void ClassifyThrowsArgumentOutOfRangeExceptionIfAttributesCountIsNotEqualToFeaturesCount()
{
// arrange
var dataSetMock = new Mock <IDataSet <bool, double> >();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
classifier.Train(dataSetMock.Object);
// act
classifier.Classify(new double[0]);
// assert
Assert.Fail();
}
public void ClassifyAccuracyOnTestData1Is89Percents()
{
// arrange
const int FeaturesCount = 2;
const double RegularizationParameter = 0.0; // no regularization
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(this.dataSet1);
// act
var correctClassifications = this.dataSet1.GetData()
.Select(data => classifier.Classify(data.Attributes) == data.Category ? data.Count : 0)
.Sum();
var accuracyPercents = (double)correctClassifications * 100 / this.dataSet1.GetTrainingSamplesCount();
// assert
Assert.AreEqual(89.0, accuracyPercents, Delta);
}
public void ClassifyAccuracyOnTestData2UsingDegree6AndRegularization1Is83Point05Percents()
{
// arrange
const int Degree = 6;
var mappedFeaturesCount = DataHelper.TwoMappedFeaturesToDegreeCount(Degree);
const double RegularizationParameter = 1.0;
var mappedDataSet2 = DataHelper.MapTwoFeaturesDataSetToDegree(this.dataSet2, Degree);
var classifier = new LogisticRegressionClassifier(mappedFeaturesCount, RegularizationParameter);
classifier.Train(mappedDataSet2);
// act
var correctClassifications = mappedDataSet2.GetData()
.Select(data => classifier.Classify(data.Attributes) == data.Category ? data.Count : 0)
.Sum();
var accuracyPercents = (double)correctClassifications * 100 / mappedDataSet2.GetTrainingSamplesCount();
// assert
Assert.AreEqual(83.050847457627114, accuracyPercents, Delta);
}
public void GetCategoryProbabilityThrowsArgumentOutOfRangeExceptionIfAttributesCountIsNotEqualToFeaturesCount()
{
// arrange
var dataSetMock = new Mock<IDataSet<bool, double>>();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
classifier.Train(dataSetMock.Object);
// act
classifier.GetCategoryProbability(true, new double[0]);
// assert
Assert.Fail();
}
public void TrainDoesNotThrowExceptionOnValidDataSet()
{
// arrange
var dataSetMock = new Mock<IDataSet<bool, double>>();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.Train(dataSetMock.Object);
// assert
}
public void ThetasPropertyReturnsAllZerosIfClassifierIsNotTrained()
{
// arrange
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
var thetas = classifier.Thetas;
// assert
Assert.IsTrue(thetas.All(d => d == 0));
}
public void ThetasPropertyAfterTrainingReturnsTheSameThetasAsTrainerReturns()
{
// arrange
var dataSetMock = new Mock<IDataSet<bool, double>>();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(2);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new[] { 0.0 }),
TrainingSample.Create(false, new[] { 1.0 }) });
const int FeaturesCount = 1;
const double RegularizationParameter = 0.0;
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(dataSetMock.Object);
var classifierTraining = new LogisticRegressionClassifierTraining(FeaturesCount, RegularizationParameter, dataSetMock.Object);
var expectedThetas = classifierTraining.Train();
// act
var thetas = classifier.Thetas;
// assert
Assert.IsTrue(expectedThetas.SequenceEqual(thetas));
}
public void GetCategoryProbabilityReturnsCorrectProbabilityForTestData1()
{
// arrange
const int FeaturesCount = 2;
const double RegularizationParameter = 0.0; // no regularization
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(this.dataSet1);
// act
var probability = classifier.GetCategoryProbability(true, new[] { 45.0, 85.0 });
// assert
Assert.AreEqual(0.77628055852165889, probability, Delta);
}
public void ClassifyAccuracyOnTestData1Is89Percents()
{
// arrange
const int FeaturesCount = 2;
const double RegularizationParameter = 0.0; // no regularization
var classifier = new LogisticRegressionClassifier(FeaturesCount, RegularizationParameter);
classifier.Train(this.dataSet1);
// act
var correctClassifications = this.dataSet1.GetData()
.Select(data => classifier.Classify(data.Attributes) == data.Category ? data.Count : 0)
.Sum();
var accuracyPercents = (double)correctClassifications * 100 / this.dataSet1.GetTrainingSamplesCount();
// assert
Assert.AreEqual(89.0, accuracyPercents, Delta);
}
public void ClassifyAccuracyOnTestData2UsingDegree6AndRegularization1Is83Point05Percents()
{
// arrange
const int Degree = 6;
var mappedFeaturesCount = DataHelper.TwoMappedFeaturesToDegreeCount(Degree);
const double RegularizationParameter = 1.0;
var mappedDataSet2 = DataHelper.MapTwoFeaturesDataSetToDegree(this.dataSet2, Degree);
var classifier = new LogisticRegressionClassifier(mappedFeaturesCount, RegularizationParameter);
classifier.Train(mappedDataSet2);
// act
var correctClassifications = mappedDataSet2.GetData()
.Select(data => classifier.Classify(data.Attributes) == data.Category ? data.Count : 0)
.Sum();
var accuracyPercents = (double)correctClassifications * 100 / mappedDataSet2.GetTrainingSamplesCount();
// assert
Assert.AreEqual(83.050847457627114, accuracyPercents, Delta);
}
public void ClassifyThrowsInvalidOperationExceptionIfClassifierIsNotTrained()
{
// arrange
var dataSetMock = new Mock<IDataSet<bool, double>>();
dataSetMock.Setup(ds => ds.GetTrainingSamplesCount()).Returns(1);
dataSetMock.Setup(ds => ds.GetData()).Returns(new[] { TrainingSample.Create(true, new double[1]) });
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.Classify(new double[1]);
// assert
Assert.Fail();
}
public void TrainThrowsArgumentNullExceptionIfTrainingDataSetIsNull()
{
// arrange
var classifier = new LogisticRegressionClassifier(1, 0.0);
// act
classifier.Train(null);
// assert
Assert.Fail();
}
