Examples with SplitCandidate org.knime.base.node.mine.treeensemble2.learner.SplitCandidate used on opensource projects

Example 26 with SplitCandidate

use of org.knime.base.node.mine.treeensemble2.learner.SplitCandidate in project knime-core by knime.

the class TreeNominalColumnDataTest method testCalcBestSplitRegressionBinary.

/**
 * Tests the method
 * {@link TreeNominalColumnData#calcBestSplitRegression(DataMemberships, RegressionPriors, TreeTargetNumericColumnData, RandomData)}
 * using binary splits
 *
 * @throws Exception
 */
@Test
public void testCalcBestSplitRegressionBinary() throws Exception {
    TreeEnsembleLearnerConfiguration config = new TreeEnsembleLearnerConfiguration(true);
    Pair<TreeNominalColumnData, TreeTargetNumericColumnData> tennisDataRegression = tennisDataRegression(config);
    TreeNominalColumnData columnData = tennisDataRegression.getFirst();
    TreeTargetNumericColumnData targetData = tennisDataRegression.getSecond();
    TreeData treeData = createTreeDataRegression(tennisDataRegression);
    double[] rowWeights = new double[SMALL_COLUMN_DATA.length];
    Arrays.fill(rowWeights, 1.0);
    IDataIndexManager indexManager = new DefaultDataIndexManager(treeData);
    DataMemberships dataMemberships = new RootDataMemberships(rowWeights, treeData, indexManager);
    RegressionPriors priors = targetData.getPriors(rowWeights, config);
    SplitCandidate splitCandidate = columnData.calcBestSplitRegression(dataMemberships, priors, targetData, null);
    assertNotNull(splitCandidate);
    assertThat(splitCandidate, instanceOf(NominalBinarySplitCandidate.class));
    assertTrue(splitCandidate.canColumnBeSplitFurther());
    assertEquals(32.9143, splitCandidate.getGainValue(), 0.0001);
    NominalBinarySplitCandidate binarySplitCandidate = (NominalBinarySplitCandidate) splitCandidate;
    TreeNodeNominalBinaryCondition[] childConditions = binarySplitCandidate.getChildConditions();
    assertEquals(2, childConditions.length);
    assertArrayEquals(new String[] { "R" }, childConditions[0].getValues());
    assertArrayEquals(new String[] { "R" }, childConditions[1].getValues());
    assertEquals(SetLogic.IS_NOT_IN, childConditions[0].getSetLogic());
    assertEquals(SetLogic.IS_IN, childConditions[1].getSetLogic());
}

Example 27 with SplitCandidate

use of org.knime.base.node.mine.treeensemble2.learner.SplitCandidate in project knime-core by knime.

the class TreeNominalColumnDataTest method testCalcBestSplitClassificationBinary.

/**
 * Tests the method
 * {@link TreeNominalColumnData#calcBestSplitClassification(DataMemberships, ClassificationPriors, TreeTargetNominalColumnData, RandomData)}
 * using binary splits.
 *
 * @throws Exception
 */
@Test
public void testCalcBestSplitClassificationBinary() throws Exception {
    final TreeEnsembleLearnerConfiguration config = createConfig(false);
    Pair<TreeNominalColumnData, TreeTargetNominalColumnData> tennisData = tennisData(config);
    TreeNominalColumnData columnData = tennisData.getFirst();
    TreeTargetNominalColumnData targetData = tennisData.getSecond();
    assertEquals(SplitCriterion.Gini, config.getSplitCriterion());
    double[] rowWeights = new double[SMALL_COLUMN_DATA.length];
    Arrays.fill(rowWeights, 1.0);
    TreeData tennisTreeData = tennisTreeData(config);
    IDataIndexManager indexManager = new DefaultDataIndexManager(tennisTreeData);
    DataMemberships dataMemberships = new RootDataMemberships(rowWeights, tennisTreeData, indexManager);
    ClassificationPriors priors = targetData.getDistribution(rowWeights, config);
    SplitCandidate splitCandidate = columnData.calcBestSplitClassification(dataMemberships, priors, targetData, null);
    assertNotNull(splitCandidate);
    assertThat(splitCandidate, instanceOf(NominalBinarySplitCandidate.class));
    assertTrue(splitCandidate.canColumnBeSplitFurther());
    // manually via libre office calc
    assertEquals(0.0689342404, splitCandidate.getGainValue(), 0.00001);
    NominalBinarySplitCandidate binSplitCandidate = (NominalBinarySplitCandidate) splitCandidate;
    TreeNodeNominalBinaryCondition[] childConditions = binSplitCandidate.getChildConditions();
    assertEquals(2, childConditions.length);
    assertArrayEquals(new String[] { "R" }, childConditions[0].getValues());
    assertArrayEquals(new String[] { "R" }, childConditions[1].getValues());
    assertEquals(SetLogic.IS_NOT_IN, childConditions[0].getSetLogic());
    assertEquals(SetLogic.IS_IN, childConditions[1].getSetLogic());
    BitSet inChild = columnData.updateChildMemberships(childConditions[0], dataMemberships);
    DataMemberships child1Memberships = dataMemberships.createChildMemberships(inChild);
    ClassificationPriors childTargetPriors = targetData.getDistribution(child1Memberships, config);
    SplitCandidate splitCandidateChild = columnData.calcBestSplitClassification(child1Memberships, childTargetPriors, targetData, null);
    assertNotNull(splitCandidateChild);
    assertThat(splitCandidateChild, instanceOf(NominalBinarySplitCandidate.class));
    // manually via libre office calc
    assertEquals(0.0086419753, splitCandidateChild.getGainValue(), 0.00001);
    inChild = columnData.updateChildMemberships(childConditions[1], dataMemberships);
    DataMemberships child2Memberships = dataMemberships.createChildMemberships(inChild);
    childTargetPriors = targetData.getDistribution(child2Memberships, config);
    splitCandidateChild = columnData.calcBestSplitClassification(child2Memberships, childTargetPriors, targetData, null);
    assertNull(splitCandidateChild);
}

Example 28 with SplitCandidate

use of org.knime.base.node.mine.treeensemble2.learner.SplitCandidate in project knime-core by knime.

the class TreeNominalColumnDataTest method testCalcBestSplitClassificationMultiwayXGBoostMissingValueHandling.

/**
 * This method tests the XGBoost missing value handling for classification in case of multiway splits.
 *
 * @throws Exception
 */
@Test
public void testCalcBestSplitClassificationMultiwayXGBoostMissingValueHandling() throws Exception {
    final TreeEnsembleLearnerConfiguration config = createConfig(false);
    config.setUseBinaryNominalSplits(false);
    config.setMissingValueHandling(MissingValueHandling.XGBoost);
    final TestDataGenerator dataGen = new TestDataGenerator(config);
    final RandomData rd = config.createRandomData();
    // test the case that there are no missing values in the training data
    final String noMissingCSV = "a, a, a, b, b, b, b, c, c";
    final String noMissingTarget = "A, B, B, C, C, C, B, A, B";
    TreeNominalColumnData dataCol = dataGen.createNominalAttributeColumn(noMissingCSV, "noMissings", 0);
    TreeTargetNominalColumnData targetCol = TestDataGenerator.createNominalTargetColumn(noMissingTarget);
    DataMemberships dataMem = createMockDataMemberships(targetCol.getNrRows());
    SplitCandidate split = dataCol.calcBestSplitClassification(dataMem, targetCol.getDistribution(dataMem, config), targetCol, rd);
    assertNotNull("There is a possible split.", split);
    assertEquals("Incorrect gain.", 0.216, split.getGainValue(), 1e-3);
    assertThat(split, instanceOf(NominalMultiwaySplitCandidate.class));
    NominalMultiwaySplitCandidate nomSplit = (NominalMultiwaySplitCandidate) split;
    assertTrue("No missing values in the column.", nomSplit.getMissedRows().isEmpty());
    TreeNodeNominalCondition[] conditions = nomSplit.getChildConditions();
    assertEquals("Wrong number of child conditions.", 3, conditions.length);
    assertEquals("Wrong value in child condition.", "a", conditions[0].getValue());
    assertEquals("Wrong value in child condition.", "b", conditions[1].getValue());
    assertEquals("Wrong value in child condition.", "c", conditions[2].getValue());
    assertFalse("Missing values should be sent to the majority child (i.e. b)", conditions[0].acceptsMissings());
    assertTrue("Missing values should be sent to the majority child (i.e. b)", conditions[1].acceptsMissings());
    assertFalse("Missing values should be sent to the majority child (i.e. b)", conditions[2].acceptsMissings());
    // test the case that there are missing values in the training data
    final String missingCSV = "a, a, a, b, b, b, b, c, c, ?";
    final String missingTarget = "A, B, B, C, C, C, B, A, B, C";
    dataCol = dataGen.createNominalAttributeColumn(missingCSV, "missings", 0);
    targetCol = TestDataGenerator.createNominalTargetColumn(missingTarget);
    dataMem = createMockDataMemberships(targetCol.getNrRows());
    split = dataCol.calcBestSplitClassification(dataMem, targetCol.getDistribution(dataMem, config), targetCol, rd);
    assertNotNull("There is a possible split.", split);
    assertEquals("Incorrect gain.", 0.2467, split.getGainValue(), 1e-3);
    assertThat(split, instanceOf(NominalMultiwaySplitCandidate.class));
    nomSplit = (NominalMultiwaySplitCandidate) split;
    assertTrue("Split should handle missing values.", nomSplit.getMissedRows().isEmpty());
    conditions = nomSplit.getChildConditions();
    assertEquals("Wrong number of child conditions.", 3, conditions.length);
    assertEquals("Wrong value in child condition.", "a", conditions[0].getValue());
    assertEquals("Wrong value in child condition.", "b", conditions[1].getValue());
    assertEquals("Wrong value in child condition.", "c", conditions[2].getValue());
    assertFalse("Missing values should be sent to b", conditions[0].acceptsMissings());
    assertTrue("Missing values should be sent to b", conditions[1].acceptsMissings());
    assertFalse("Missing values should be sent to b", conditions[2].acceptsMissings());
}

Example 29 with SplitCandidate

use of org.knime.base.node.mine.treeensemble2.learner.SplitCandidate in project knime-core by knime.

the class TreeNominalColumnDataTest method testCalcBestSplitClassificationBinaryPCA.

/**
 * Tests the method
 * {@link TreeNominalColumnData#calcBestSplitClassification(DataMemberships, ClassificationPriors, TreeTargetNominalColumnData, RandomData)}
 * using binary splits. In this test case the data has more than two classes and the used algorithm is therefore PCA
 * based.
 *
 * @throws Exception
 */
@Test
public void testCalcBestSplitClassificationBinaryPCA() throws Exception {
    TreeEnsembleLearnerConfiguration config = createConfig(false);
    Pair<TreeNominalColumnData, TreeTargetNominalColumnData> pcaData = createPCATestData(config);
    TreeNominalColumnData columnData = pcaData.getFirst();
    TreeTargetNominalColumnData targetData = pcaData.getSecond();
    TreeData treeData = createTreeData(pcaData);
    assertEquals(SplitCriterion.Gini, config.getSplitCriterion());
    double[] rowWeights = new double[targetData.getNrRows()];
    Arrays.fill(rowWeights, 1.0);
    IDataIndexManager indexManager = new DefaultDataIndexManager(treeData);
    DataMemberships dataMemberships = new RootDataMemberships(rowWeights, treeData, indexManager);
    ClassificationPriors priors = targetData.getDistribution(rowWeights, config);
    SplitCandidate splitCandidate = columnData.calcBestSplitClassification(dataMemberships, priors, targetData, null);
    assertNotNull(splitCandidate);
    assertThat(splitCandidate, instanceOf(NominalBinarySplitCandidate.class));
    assertTrue(splitCandidate.canColumnBeSplitFurther());
    assertEquals(0.0659, splitCandidate.getGainValue(), 0.0001);
    NominalBinarySplitCandidate binarySplitCandidate = (NominalBinarySplitCandidate) splitCandidate;
    TreeNodeNominalBinaryCondition[] childConditions = binarySplitCandidate.getChildConditions();
    assertEquals(2, childConditions.length);
    assertArrayEquals(new String[] { "E" }, childConditions[0].getValues());
    assertArrayEquals(new String[] { "E" }, childConditions[1].getValues());
    assertEquals(SetLogic.IS_NOT_IN, childConditions[0].getSetLogic());
    assertEquals(SetLogic.IS_IN, childConditions[1].getSetLogic());
}

Example 30 with SplitCandidate

use of org.knime.base.node.mine.treeensemble2.learner.SplitCandidate in project knime-core by knime.

the class TreeNumericColumnData method calcBestSplitRegression.

@Override
public SplitCandidate calcBestSplitRegression(final DataMemberships dataMemberships, final RegressionPriors targetPriors, final TreeTargetNumericColumnData targetColumn, final RandomData rd) {
    final TreeEnsembleLearnerConfiguration config = getConfiguration();
    final boolean useAverageSplitPoints = config.isUseAverageSplitPoints();
    final int minChildNodeSize = config.getMinChildSize();
    // get columnMemberships
    final ColumnMemberships columnMemberships = dataMemberships.getColumnMemberships(getMetaData().getAttributeIndex());
    final int lengthNonMissing = getLengthNonMissing();
    // missing value handling
    final boolean useXGBoostMissingValueHandling = config.getMissingValueHandling() == MissingValueHandling.XGBoost;
    // are there missing values in this column (complete column)
    boolean branchContainsMissingValues = containsMissingValues();
    boolean missingsGoLeft = true;
    double missingWeight = 0.0;
    double missingY = 0.0;
    // check if there are missing values in this rowsample
    if (branchContainsMissingValues) {
        columnMemberships.goToLast();
        while (columnMemberships.getIndexInColumn() >= lengthNonMissing) {
            missingWeight += columnMemberships.getRowWeight();
            missingY += targetColumn.getValueFor(columnMemberships.getOriginalIndex());
            if (!columnMemberships.previous()) {
                break;
            }
        }
        columnMemberships.reset();
        branchContainsMissingValues = missingWeight > 0.0;
    }
    final double ySumTotal = targetPriors.getYSum() - missingY;
    final double nrRecordsTotal = targetPriors.getNrRecords() - missingWeight;
    final double criterionTotal = useXGBoostMissingValueHandling ? (ySumTotal + missingY) * (ySumTotal + missingY) / (nrRecordsTotal + missingWeight) : ySumTotal * ySumTotal / nrRecordsTotal;
    double ySumLeft = 0.0;
    double nrRecordsLeft = 0.0;
    double ySumRight = ySumTotal;
    double nrRecordsRight = nrRecordsTotal;
    // all values in the current branch are missing
    if (nrRecordsRight == 0) {
        // it is impossible to determine a split
        return null;
    }
    double bestSplit = Double.NEGATIVE_INFINITY;
    double bestImprovement = 0.0;
    double lastSeenY = Double.NaN;
    double lastSeenValue = Double.NEGATIVE_INFINITY;
    double lastSeenWeight = -1.0;
    // compute the gain, keep the one that maximizes the split
    while (columnMemberships.next()) {
        final double weight = columnMemberships.getRowWeight();
        if (weight < EPSILON) {
            // ignore record: not in current branch or not in sample
            continue;
        } else if (Math.floor(weight) != weight) {
            throw new UnsupportedOperationException("weighted records (missing values?) not supported, " + "weight is " + weight);
        }
        final double value = getSorted(columnMemberships.getIndexInColumn());
        if (lastSeenWeight > 0.0) {
            ySumLeft += lastSeenWeight * lastSeenY;
            ySumRight -= lastSeenWeight * lastSeenY;
            nrRecordsLeft += lastSeenWeight;
            nrRecordsRight -= lastSeenWeight;
            if (nrRecordsLeft >= minChildNodeSize && nrRecordsRight >= minChildNodeSize && lastSeenValue < value) {
                boolean tempMissingsGoLeft = true;
                double childrenSquaredSum;
                if (branchContainsMissingValues && useXGBoostMissingValueHandling) {
                    final double[] tempChildrenSquaredSum = new double[2];
                    tempChildrenSquaredSum[0] = ((ySumLeft + missingY) * (ySumLeft + missingY) / (nrRecordsLeft + missingWeight)) + (ySumRight * ySumRight / nrRecordsRight);
                    tempChildrenSquaredSum[1] = (ySumLeft * ySumLeft / nrRecordsLeft) + ((ySumRight + missingY) * (ySumRight + missingY) / (nrRecordsRight + missingWeight));
                    if (tempChildrenSquaredSum[0] >= tempChildrenSquaredSum[1]) {
                        childrenSquaredSum = tempChildrenSquaredSum[0];
                        tempMissingsGoLeft = true;
                    } else {
                        childrenSquaredSum = tempChildrenSquaredSum[1];
                        tempMissingsGoLeft = false;
                    }
                } else {
                    childrenSquaredSum = (ySumLeft * ySumLeft / nrRecordsLeft) + (ySumRight * ySumRight / nrRecordsRight);
                }
                double criterion = childrenSquaredSum - criterionTotal;
                boolean randomTieBreaker = criterion == bestImprovement ? rd.nextInt(0, 1) == 1 : false;
                if (criterion > bestImprovement || randomTieBreaker) {
                    bestImprovement = criterion;
                    bestSplit = useAverageSplitPoints ? getCenter(lastSeenValue, value) : lastSeenValue;
                    // if there are no missing values go with majority
                    missingsGoLeft = branchContainsMissingValues ? tempMissingsGoLeft : nrRecordsLeft >= nrRecordsRight;
                }
            }
        }
        lastSeenY = targetColumn.getValueFor(columnMemberships.getOriginalIndex());
        lastSeenValue = value;
        lastSeenWeight = weight;
    }
    // + " but was " + lastSeenY * lastSeenWeight;
    if (bestImprovement > 0.0) {
        if (useXGBoostMissingValueHandling) {
            // return new NumericMissingSplitCandidate(this, bestSplit, bestImprovement, missingsGoLeft);
            return new NumericSplitCandidate(this, bestSplit, bestImprovement, new BitSet(), missingsGoLeft ? NumericSplitCandidate.MISSINGS_GO_LEFT : NumericSplitCandidate.MISSINGS_GO_RIGHT);
        }
        return new NumericSplitCandidate(this, bestSplit, bestImprovement, getMissedRows(columnMemberships), NumericSplitCandidate.NO_MISSINGS);
    } else {
        return null;
    }
}