public void RocAucRegressionMetric_Error_Not_Binary()
{
var targets = new double[] { 0, 1, 2 };
var probabilities = new double[0];
var sut = new RocAucRegressionMetric(1);
var actual = sut.Error(targets, probabilities);
}
public void RocAucRegressionMetric_Error_No_Error()
{
var targets = new double[] { 0, 1 };
var probabilities = new double[] { 0.0, 1 };
var sut = new RocAucRegressionMetric(1);
var actual = sut.Error(targets, probabilities);
Assert.AreEqual(0.0, actual);
}
public void RocAucRegressionMetric_Error()
{
var targets = new double[] { 1, 0, 1, 0, 0, 0, 1, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 0, 0, 1, 1 };
var probabilities = new double[] { 0.052380952, 0.020725389, 0.993377483, 0.020725389, 0.020725389, 0.111111111, 0.193377483, 0.793377483, 0.020725389, 0.012345679, 0.885860173, 0.714285714, 0.985860173, 0.020725389, 0.985860173, 0.993377483, 0.993377483, 0.954545455, 0.020725389, 0.020725389, 0.985860173, 0.985860173 };
var sut = new RocAucRegressionMetric(1);
var actual = sut.Error(targets, probabilities);
Assert.AreEqual(0.0085470085470086277, actual, 0.00001);
}
public void RocAucRegressionMetric_Error_Using_Mapping()
{
var targets = new double[] { 1, 0, 1, 0, 0, 0, 1, 3, 0, 0, 1, 1, 3, 0, 1, 1, 3, 2, 0, 0, 2, 1 };
var probabilities = new double[] { 0.052380952, 0.020725389, 0.993377483, 0.020725389, 0.020725389, 0.111111111, 0.193377483, 0.793377483, 0.020725389, 0.012345679, 0.885860173, 0.714285714, 0.985860173, 0.020725389, 0.985860173, 0.993377483, 0.993377483, 0.954545455, 0.020725389, 0.020725389, 0.985860173, 0.985860173 };
var sut = new RocAucRegressionMetric(1, new Dictionary <double, double> {
{ 0.0, 0.0 }, { 1.0, 1.0 }, { 2.0, 1.0 }, { 3.0, 1.0 }
});
var actual = sut.Error(targets, probabilities);
Assert.AreEqual(0.0085470085470086277, actual, 0.00001);
}
public void RocAucRegressionMetric_Error_Random()
{
var positives = Enumerable.Range(0, 800).Select(s => 1.0).ToList();
var negatives = Enumerable.Range(0, 800).Select(s => 0.0).ToList();
positives.AddRange(negatives);
var targets = positives.ToArray();
var random = new Random(42);
var probabilities = targets.Select(s => random.NextDouble()).ToArray();
var sut = new RocAucRegressionMetric(1);
var actual = sut.Error(targets, probabilities);
Assert.AreEqual(0.488959375, actual, 0.0001);
}
public void RocAucRegressionMetric_Error_Always_Positve()
{
var positives = Enumerable.Range(0, 800).Select(s => 1.0).ToList();
var negatives = Enumerable.Range(0, 200).Select(s => 0.0).ToList();
positives.AddRange(negatives);
var targets = positives.ToArray();
var probabilities = targets
.Select(s => 1.0)
.ToArray();
var sut = new RocAucRegressionMetric(1);
var actual = sut.Error(targets, probabilities);
Assert.AreEqual(0.5, actual, 0.0001);
}
