@Test
public void testFactorNACalc() {
FactorExtraTrees et = getFET(10, 5, false);
FactorExtraTrees etw = getFET(10, 5, true);
double gini;
gini = 1 - (0.4 * 0.4 + 0.6 * 0.6);
assertEquals(gini, et.get1NaNScore(AbstractTrees.seq(10)), 1e-6);
gini = 1 - (Math.pow(3.0 / 9.0, 2) + Math.pow(6.0 / 9.0, 2));
assertEquals(gini, et.get1NaNScore(AbstractTrees.seq(9)), 1e-6);
// testing NaN counts:
CutResult cr;
cr = new CutResult();
et.calculateCutScore(AbstractTrees.seq(9), 2, 0.5, cr);
assertEquals(3.0, cr.nanWeigth, 1e-6);
cr = new CutResult();
et.calculateCutScore(AbstractTrees.seq(9), 1, 0.5, cr);
assertEquals(0.0, cr.nanWeigth, 1e-6);
// testing weights:
cr = new CutResult();
etw.calculateCutScore(AbstractTrees.seq(9), 2, 0.5, cr);
assertEquals(1.5, cr.nanWeigth, 1e-6);
}
@Test
public void testRegressionNACalc() {
ExtraTrees et = getET(10, 5, false);
ExtraTrees etw = getET(10, 5, true);
double var, mean;
mean = 4.0 / 10.0;
var = 0.6 * Math.pow(mean, 2) + 0.4 * Math.pow(1 - mean, 2);
assertEquals(var, et.get1NaNScore(AbstractTrees.seq(10)), 1e-6);
mean = 3.0 / 9.0;
var = 6 / 9.0 * Math.pow(mean, 2) + 3 / 9.0 * Math.pow(1 - mean, 2);
assertEquals(var, et.get1NaNScore(AbstractTrees.seq(9)), 1e-6);
// testing NaN counts:
CutResult cr;
cr = new CutResult();
et.calculateCutScore(AbstractTrees.seq(9), 2, 0.5, cr);
assertEquals(3.0, cr.nanWeigth, 1e-6);
cr = new CutResult();
et.calculateCutScore(AbstractTrees.seq(9), 1, 0.5, cr);
assertEquals(0.0, cr.nanWeigth, 1e-6);
// testing weights:
cr = new CutResult();
etw.calculateCutScore(AbstractTrees.seq(9), 2, 0.5, cr);
assertEquals(1.5, cr.nanWeigth, 1e-6);
}
@Test
public void testRegressionNALearn() {
int ndim = 5;
ExtraTrees et = getET(100, ndim, false);
ExtraTrees etw = getET(100, ndim, true);
et.learnTrees(3, 3, 5);
etw.learnTrees(3, 3, 5);
double[] x = new double[ndim];
for (int i = 0; i < x.length; i++) {
x[i] = Double.NaN;
}
double[] val;
val = et.getValues(new Matrix(x, 1, ndim));
assertTrue(Double.isNaN(val[0]));
val = etw.getValues(new Matrix(x, 1, ndim));
assertTrue(Double.isNaN(val[0]));
// checking if getRange works with NaN
double[] col2 = ((Matrix) et.input).getCol(2);
double[] range2 = AbstractTrees.getRange(col2);
assertEquals(0.5, range2[0], 1e-6);
assertEquals(0.5, range2[1], 1e-6);
}
