public void GetMultiLabelClassificationMetrics_Recall()
{
var data = new EvaluateItem <double>[]
{
new EvaluateItem <double>(new[] { 0.0, 1.0, 1.0 }, new[] { 0.0, 0.8, 0.6 }),
new EvaluateItem <double>(new[] { 1.0, 0.0, 1.0 }, new[] { 0.4, 0.9, 0.1 }),
new EvaluateItem <double>(new[] { 1.0, 0.0, 0.0 }, new[] { 0.7, 0.1, 0.7 }),
new EvaluateItem <double>(new[] { 1.0, 1.0, 1.0 }, new[] { 0.0, 0.8, 0.2 }),
};
var metrics = data.GetMultiLabelClassificationMetrics();
Assert.Equal(0.33, metrics.ClassesDistribution[0].Recall, 2); // 1/3=0.33
Assert.Equal(1, metrics.ClassesDistribution[1].Recall); // 2/2=1
Assert.Equal(0.33, metrics.ClassesDistribution[2].Recall, 1); // 1/3=0.33
}
public void GetMultiLabelClassificationMetrics_Precision()
{
var data = new EvaluateItem <double>[]
{
new EvaluateItem <double>(new[] { 0.0, 1.0, 1.0 }, new[] { 0.0, 0.8, 0.6 }),
new EvaluateItem <double>(new[] { 1.0, 0.0, 1.0 }, new[] { 0.4, 0.9, 0.1 }),
new EvaluateItem <double>(new[] { 1.0, 0.0, 0.0 }, new[] { 0.7, 0.1, 0.7 }),
new EvaluateItem <double>(new[] { 1.0, 1.0, 1.0 }, new[] { 0.0, 0.8, 0.2 }),
};
var metrics = data.GetMultiLabelClassificationMetrics();
Assert.Equal(1, metrics.ClassesDistribution[0].Precision); // 1/1=1
Assert.Equal(0.67, metrics.ClassesDistribution[1].Precision, 2); // 2/3=0.67
Assert.Equal(0.5, metrics.ClassesDistribution[2].Precision, 1); // 1/2=0.5
}