Examples with DoubleValue org.knime.core.data.DoubleValue used on opensource projects

Example 16 with DoubleValue

use of org.knime.core.data.DoubleValue in project knime-core by knime.

the class AutoBinner method calcDomainBoundsIfNeccessary.

/**
 * Determines the per column min/max values of the given data if not already present in the domain.
 *
 * @param data the data
 * @param exec the execution context
 * @param recalcValuesFor The columns
 * @return The data with extended domain information
 * @throws InvalidSettingsException ...
 * @throws CanceledExecutionException ...
 */
public BufferedDataTable calcDomainBoundsIfNeccessary(final BufferedDataTable data, final ExecutionContext exec, final List<String> recalcValuesFor) throws InvalidSettingsException, CanceledExecutionException {
    if (null == recalcValuesFor || recalcValuesFor.isEmpty()) {
        return data;
    }
    List<Integer> valuesI = new ArrayList<Integer>();
    for (String colName : recalcValuesFor) {
        DataColumnSpec colSpec = data.getDataTableSpec().getColumnSpec(colName);
        if (!colSpec.getType().isCompatible(DoubleValue.class)) {
            throw new InvalidSettingsException("Can only process numeric " + "data. The column \"" + colSpec.getName() + "\" is not numeric.");
        }
        if (recalcValuesFor.contains(colName) && !colSpec.getDomain().hasBounds()) {
            valuesI.add(data.getDataTableSpec().findColumnIndex(colName));
        }
    }
    if (valuesI.isEmpty()) {
        return data;
    }
    Map<Integer, Double> min = new HashMap<Integer, Double>();
    Map<Integer, Double> max = new HashMap<Integer, Double>();
    for (int col : valuesI) {
        min.put(col, Double.MAX_VALUE);
        max.put(col, Double.MIN_VALUE);
    }
    int c = 0;
    for (DataRow row : data) {
        c++;
        exec.checkCanceled();
        exec.setProgress(c / (double) data.getRowCount());
        for (int col : valuesI) {
            double val = ((DoubleValue) row.getCell(col)).getDoubleValue();
            if (min.get(col) > val) {
                min.put(col, val);
            }
            if (max.get(col) < val) {
                min.put(col, val);
            }
        }
    }
    List<DataColumnSpec> newColSpecList = new ArrayList<DataColumnSpec>();
    int cc = 0;
    for (DataColumnSpec columnSpec : data.getDataTableSpec()) {
        if (recalcValuesFor.contains(columnSpec.getName())) {
            DataColumnSpecCreator specCreator = new DataColumnSpecCreator(columnSpec);
            DataColumnDomainCreator domainCreator = new DataColumnDomainCreator(new DoubleCell(min.get(cc)), new DoubleCell(max.get(cc)));
            specCreator.setDomain(domainCreator.createDomain());
            DataColumnSpec newColSpec = specCreator.createSpec();
            newColSpecList.add(newColSpec);
        } else {
            newColSpecList.add(columnSpec);
        }
        cc++;
    }
    DataTableSpec spec = new DataTableSpec(newColSpecList.toArray(new DataColumnSpec[0]));
    BufferedDataTable newDataTable = exec.createSpecReplacerTable(data, spec);
    return newDataTable;
}

Example 17 with DoubleValue

use of org.knime.core.data.DoubleValue in project knime-core by knime.

the class EnrichmentPlotterModel method execute.

/**
 * {@inheritDoc}
 */
@Override
protected BufferedDataTable[] execute(final BufferedDataTable[] inData, final ExecutionContext exec) throws Exception {
    final double rowCount = inData[0].size();
    final BufferedDataContainer areaOutCont = exec.createDataContainer(AREA_OUT_SPEC);
    final BufferedDataContainer discrateOutCont = exec.createDataContainer(DISCRATE_OUT_SPEC);
    for (int i = 0; i < m_settings.getCurveCount(); i++) {
        final ExecutionMonitor sexec = exec.createSubProgress(1.0 / m_settings.getCurveCount());
        exec.setMessage("Generating curve " + (i + 1));
        final Curve c = m_settings.getCurve(i);
        final Helper[] curve = new Helper[KnowsRowCountTable.checkRowCount(inData[0].size())];
        final int sortIndex = inData[0].getDataTableSpec().findColumnIndex(c.getSortColumn());
        final int actIndex = inData[0].getDataTableSpec().findColumnIndex(c.getActivityColumn());
        int k = 0, maxK = 0;
        for (DataRow row : inData[0]) {
            DataCell c1 = row.getCell(sortIndex);
            DataCell c2 = row.getCell(actIndex);
            if (k++ % 100 == 0) {
                sexec.checkCanceled();
                sexec.setProgress(k / rowCount);
            }
            if (c1.isMissing()) {
                continue;
            } else {
                curve[maxK] = new Helper(((DoubleValue) c1).getDoubleValue(), c2);
            }
            maxK++;
        }
        Arrays.sort(curve, 0, maxK);
        if (c.isSortDescending()) {
            for (int j = 0; j < maxK / 2; j++) {
                Helper h = curve[j];
                curve[j] = curve[maxK - j - 1];
                curve[maxK - j - 1] = h;
            }
        }
        // this is for down-sampling so that the view is faster;
        // plotting >100,000 points takes quite a long time
        final int size = Math.min(MAX_RESOLUTION, maxK);
        final double downSampleRate = maxK / (double) size;
        final double[] xValues = new double[size + 1];
        final double[] yValues = new double[size + 1];
        xValues[0] = 0;
        yValues[0] = 0;
        int lastK = 0;
        double y = 0, area = 0;
        int nextHitRatePoint = 0;
        final double[] hitRateValues = new double[DISCRATE_POINTS.length];
        final HashMap<DataCell, MutableInteger> clusters = new HashMap<DataCell, MutableInteger>();
        for (k = 1; k <= maxK; k++) {
            final Helper h = curve[k - 1];
            if (m_settings.plotMode() == PlotMode.PlotSum) {
                y += ((DoubleValue) h.b).getDoubleValue();
            } else if (m_settings.plotMode() == PlotMode.PlotHits) {
                if (!h.b.isMissing() && (((DoubleValue) h.b).getDoubleValue() >= m_settings.hitThreshold())) {
                    y++;
                }
            } else if (!h.b.isMissing()) {
                MutableInteger count = clusters.get(h.b);
                if (count == null) {
                    count = new MutableInteger(0);
                    clusters.put(h.b, count);
                }
                if (count.inc() == m_settings.minClusterMembers()) {
                    y++;
                }
            }
            area += y / maxK;
            if ((int) (k / downSampleRate) >= lastK + 1) {
                lastK++;
                xValues[lastK] = k;
                yValues[lastK] = y;
            }
            if ((nextHitRatePoint < DISCRATE_POINTS.length) && (k == (int) Math.floor(maxK * DISCRATE_POINTS[nextHitRatePoint] / 100))) {
                hitRateValues[nextHitRatePoint] = y;
                nextHitRatePoint++;
            }
        }
        xValues[xValues.length - 1] = maxK;
        yValues[yValues.length - 1] = y;
        area /= y;
        m_curves.add(new EnrichmentPlot(c.getSortColumn() + " vs " + c.getActivityColumn(), xValues, yValues, area));
        areaOutCont.addRowToTable(new DefaultRow(new RowKey(c.toString()), new DoubleCell(area)));
        for (int j = 0; j < hitRateValues.length; j++) {
            hitRateValues[j] /= y;
        }
        discrateOutCont.addRowToTable(new DefaultRow(new RowKey(c.toString()), hitRateValues));
    }
    areaOutCont.close();
    discrateOutCont.close();
    return new BufferedDataTable[] { areaOutCont.getTable(), discrateOutCont.getTable() };
}

Example 18 with DoubleValue

use of org.knime.core.data.DoubleValue in project knime-core by knime.

the class PolyRegLearnerNodeModel method getCellFactory.

private CellFactory getCellFactory(final int dependentIndex) {
    final int degree = m_settings.getDegree();
    return new CellFactory() {

        @Override
        public DataCell[] getCells(final DataRow row) {
            double sum = m_betas[0];
            int betaCount = 1;
            double y = 0;
            for (int col = 0; col < row.getNumCells(); col++) {
                if ((col != dependentIndex) && m_colSelected[col]) {
                    final double value = ((DoubleValue) row.getCell(col)).getDoubleValue();
                    double poly = 1;
                    for (int d = 1; d <= degree; d++) {
                        poly *= value;
                        sum += m_betas[betaCount++] * poly;
                    }
                } else if (col == dependentIndex) {
                    y = ((DoubleValue) row.getCell(col)).getDoubleValue();
                }
            }
            double err = Math.abs(sum - y);
            m_squaredError += err * err;
            return new DataCell[] { new DoubleCell(sum), new DoubleCell(err) };
        }

        @Override
        public DataColumnSpec[] getColumnSpecs() {
            DataColumnSpecCreator crea = new DataColumnSpecCreator("PolyReg prediction", DoubleCell.TYPE);
            DataColumnSpec col1 = crea.createSpec();
            crea = new DataColumnSpecCreator("Prediction Error", DoubleCell.TYPE);
            DataColumnSpec col2 = crea.createSpec();
            return new DataColumnSpec[] { col1, col2 };
        }

        @Override
        public void setProgress(final int curRowNr, final int rowCount, final RowKey lastKey, final ExecutionMonitor execMon) {
        // do nothing
        }
    };
}

Example 19 with DoubleValue

use of org.knime.core.data.DoubleValue in project knime-core by knime.

the class RegressionPredictorNodeModel method createRearranger.

private ColumnRearranger createRearranger(final DataTableSpec inSpec, final PMMLPortObjectSpec regModelSpec, final PMMLRegressionTranslator regModel) throws InvalidSettingsException {
    if (regModelSpec == null) {
        throw new InvalidSettingsException("No input");
    }
    // exclude last (response column)
    String targetCol = "Response";
    for (String s : regModelSpec.getTargetFields()) {
        targetCol = s;
        break;
    }
    final List<String> learnFields;
    if (regModel != null) {
        RegressionTable regTable = regModel.getRegressionTable();
        learnFields = new ArrayList<String>();
        for (NumericPredictor p : regTable.getVariables()) {
            learnFields.add(p.getName());
        }
    } else {
        learnFields = new ArrayList<String>(regModelSpec.getLearningFields());
    }
    final int[] colIndices = new int[learnFields.size()];
    int k = 0;
    for (String learnCol : learnFields) {
        int index = inSpec.findColumnIndex(learnCol);
        if (index < 0) {
            throw new InvalidSettingsException("Missing column for " + "regressor variable : \"" + learnCol + "\"");
        }
        DataColumnSpec regressor = inSpec.getColumnSpec(index);
        String name = regressor.getName();
        DataColumnSpec col = inSpec.getColumnSpec(index);
        if (!col.getType().isCompatible(DoubleValue.class)) {
            throw new InvalidSettingsException("Incompatible type of " + "column \"" + name + "\": " + col.getType());
        }
        colIndices[k++] = index;
    }
    // try to use some smart naming scheme for the append column
    String oldName = targetCol;
    if (inSpec.containsName(oldName) && !oldName.toLowerCase().endsWith("(prediction)")) {
        oldName = oldName + " (prediction)";
    }
    String newColName = DataTableSpec.getUniqueColumnName(inSpec, oldName);
    DataColumnSpec newCol = new DataColumnSpecCreator(newColName, DoubleCell.TYPE).createSpec();
    SingleCellFactory fac = new SingleCellFactory(newCol) {

        @Override
        public DataCell getCell(final DataRow row) {
            RegressionTable t = regModel.getRegressionTable();
            int j = 0;
            double result = t.getIntercept();
            for (NumericPredictor p : t.getVariables()) {
                DataCell c = row.getCell(colIndices[j++]);
                if (c.isMissing()) {
                    return DataType.getMissingCell();
                }
                double v = ((DoubleValue) c).getDoubleValue();
                if (p.getExponent() != 1) {
                    v = Math.pow(v, p.getExponent());
                }
                result += p.getCoefficient() * v;
            }
            return new DoubleCell(result);
        }
    };
    ColumnRearranger c = new ColumnRearranger(inSpec);
    c.append(fac);
    return c;
}

Example 20 with DoubleValue

use of org.knime.core.data.DoubleValue in project knime-core by knime.

the class AbstractTreeEnsembleModel method createNominalNumericPredictorRecord.

private PredictorRecord createNominalNumericPredictorRecord(final DataRow filterRow, final DataTableSpec trainSpec) {
    final int nrCols = trainSpec.getNumColumns();
    Map<String, Object> valueMap = new LinkedHashMap<String, Object>((int) (nrCols / 0.75 + 1.0));
    for (int i = 0; i < nrCols; i++) {
        DataColumnSpec col = trainSpec.getColumnSpec(i);
        String colName = col.getName();
        DataType colType = col.getType();
        DataCell cell = filterRow.getCell(i);
        if (cell.isMissing()) {
            valueMap.put(colName, PredictorRecord.NULL);
        } else if (colType.isCompatible(NominalValue.class)) {
            TreeNominalColumnMetaData nomColMeta = (TreeNominalColumnMetaData) m_metaData.getAttributeMetaData(i);
            NominalValueRepresentation[] nomVals = nomColMeta.getValues();
            int assignedInteger = -1;
            String val = cell.toString();
            // find assignedInteger of value
            for (NominalValueRepresentation nomVal : nomVals) {
                if (nomVal.getNominalValue().equals(val)) {
                    assignedInteger = nomVal.getAssignedInteger();
                    break;
                }
            }
            // the value is not known to the model
            if (assignedInteger == -1) {
                // treat as missing value
                valueMap.put(colName, PredictorRecord.NULL);
            } else {
                valueMap.put(colName, Integer.valueOf(assignedInteger));
            }
        } else if (colType.isCompatible(DoubleValue.class)) {
            double val = ((DoubleValue) cell).getDoubleValue();
            if (Double.isNaN(val)) {
                // make sure that NaNs are treated as missing values
                // bug AP-7169
                valueMap.put(colName, PredictorRecord.NULL);
            } else {
                valueMap.put(colName, val);
            }
        } else {
            throw new IllegalStateException("Expected nominal or numeric column type for column \"" + colName + "\" but got \"" + colType + "\"");
        }
    }
    return new PredictorRecord(valueMap);
}