public void BalancedBinaryClassificationRandomForestLearnerThrowsForWrongLabel(
double firstLabel,
double secondLabel,
double thirdLabel,
double fourthLabel)
{
var trainDataObservations = new F64Matrix(
new[]
{
1.0,
2.0,
3.0,
4.0,
5.0,
6.0,
7.0,
8.0,
},
4,
2);
var trainDataLabels = new[] { firstLabel, secondLabel, thirdLabel, fourthLabel };
var learner = new BalancedBinaryClassificationRandomForestLearner(numberOfThreads: 1);
Assert.Throws <NotSupportedException>(() => learner.Learn(trainDataObservations, trainDataLabels));
}
public void BalancedBinaryClassificationRandomForestLearnerThrowsAtNanInLabels()
{
var trainDataObservations = new F64Matrix(
new[]
{
1.0,
2.0,
3.0,
4.0,
},
2,
2);
var trainDataLabels = new[] { 0.0, double.NaN };
var learner = new BalancedBinaryClassificationRandomForestLearner(numberOfThreads: 1);
Assert.Throws <ArgumentOutOfRangeException>(() => learner.Learn(trainDataObservations, trainDataLabels));
}
public void BalancedBinaryClassificationRandomForestLearnerDoesNotThrowForCorrectLabel(double firstLabel, double secondLabel)
{
var trainDataObservations = new F64Matrix(
new[]
{
1.0,
2.0,
3.0,
4.0,
},
2,
2);
var trainDataLabels = new[] { firstLabel, secondLabel };
var learner = new BalancedBinaryClassificationRandomForestLearner(numberOfThreads: 1);
var model = learner.Learn(trainDataObservations, trainDataLabels);
model.Predict(trainDataObservations);
}
public void BalancedBinaryClassificationRandomForestLearnerDoesNotThrowAtNanInObservations()
{
var trainDataObservations = new F64Matrix(
new[]
{
1.0,
2.0,
double.NaN,
double.NaN,
},
2,
2);
var trainDataLabels = new[] { 0.0, 1.0 };
var learner = new BalancedBinaryClassificationRandomForestLearner(numberOfThreads: 1);
var model = learner.Learn(trainDataObservations, trainDataLabels);
model.Predict(trainDataObservations);
}
