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

Example 96 with DataTableSpec

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

the class AutoBinnerLearnNodeModel method configure.

/**
 * {@inheritDoc}
 */
@Override
protected PortObjectSpec[] configure(final PortObjectSpec[] inSpecs) throws InvalidSettingsException {
    DataTableSpec dataSpec = (DataTableSpec) inSpecs[0];
    AutoBinner binner = new AutoBinner(m_settings);
    return binner.getOutputSpec(dataSpec);
}

Example 97 with DataTableSpec

use of org.knime.core.data.DataTableSpec 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 98 with DataTableSpec

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

the class CategoryToNumberNodeModel method execute.

/**
 * {@inheritDoc}
 */
@Override
protected PortObject[] execute(final PortObject[] inObjects, final ExecutionContext exec) throws Exception {
    if (m_settings.getIncludedColumns().length == 0) {
        // nothing to convert, let's return the input table.
        setWarningMessage("No columns selected," + " returning input.");
    }
    BufferedDataTable inData = (BufferedDataTable) inObjects[0];
    DataTableSpec inSpec = (DataTableSpec) inObjects[0].getSpec();
    ColumnRearranger rearranger = createRearranger(inSpec);
    BufferedDataTable outTable = exec.createColumnRearrangeTable(inData, rearranger, exec);
    // the optional PMML in port (can be null)
    PMMLPortObject inPMMLPort = (PMMLPortObject) inObjects[1];
    PMMLPortObjectSpecCreator creator = new PMMLPortObjectSpecCreator(inPMMLPort, rearranger.createSpec());
    PMMLPortObject outPMMLPort = new PMMLPortObject(creator.createSpec(), inPMMLPort);
    for (CategoryToNumberCellFactory factory : m_factories) {
        PMMLMapValuesTranslator trans = new PMMLMapValuesTranslator(factory.getConfig(), new DerivedFieldMapper(inPMMLPort));
        outPMMLPort.addGlobalTransformations(trans.exportToTransDict());
    }
    return new PortObject[] { outTable, outPMMLPort };
}

Example 99 with DataTableSpec

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

the class CategoryToNumberNodeModel method configure.

/**
 * {@inheritDoc}
 */
@Override
protected PortObjectSpec[] configure(final PortObjectSpec[] inSpecs) throws InvalidSettingsException {
    DataTableSpec inSpec = (DataTableSpec) inSpecs[0];
    List<String> inputCols = new ArrayList<String>();
    for (DataColumnSpec column : inSpec) {
        if (column.getType().isCompatible(StringValue.class)) {
            inputCols.add(column.getName());
        }
    }
    // Auto configuration when included columns are not set
    if (null == m_settings.getIncludedColumns()) {
        // when there is no column with nominal data
        if (inputCols.isEmpty()) {
            throw new InvalidSettingsException("No column in " + "the input compatible to \"StringValue\".");
        } else {
            // auto-configure
            m_settings.setIncludedColumns(inputCols.toArray(new String[inputCols.size()]));
        }
    } else {
        if (!m_settings.getIncludeAll()) {
            if (m_settings.getIncludedColumns().length == 0) {
                setWarningMessage("No columns selected.");
            }
            List<String> included = new ArrayList<String>();
            included.addAll(Arrays.asList(m_settings.getIncludedColumns()));
            included.removeAll(inputCols);
            if (!included.isEmpty()) {
                // output first missing column
                throw new InvalidSettingsException("Missing column: " + included.get(0).toString());
            }
        } else {
            if (inputCols.isEmpty()) {
                throw new InvalidSettingsException("No column in " + "the input compatible to \"StringValue\".");
            } else {
                // automatically add all columns
                m_settings.setIncludedColumns(inputCols.toArray(new String[inputCols.size()]));
            }
        }
    }
    ColumnRearranger rearranger = createRearranger(inSpec);
    PMMLPortObjectSpec pmmlSpec = (PMMLPortObjectSpec) inSpecs[1];
    PMMLPortObjectSpecCreator pmmlSpecCreator = new PMMLPortObjectSpecCreator(pmmlSpec, inSpec);
    pmmlSpecCreator.addPreprocColNames(inputCols);
    return new PortObjectSpec[] { rearranger.createSpec(), pmmlSpecCreator.createSpec() };
}

Example 100 with DataTableSpec

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

the class TreeEnsembleClassificationPredictorNodeModel method createStreamableOperator.

/**
 * {@inheritDoc}
 */
@Override
public StreamableOperator createStreamableOperator(final PartitionInfo partitionInfo, final PortObjectSpec[] inSpecs) throws InvalidSettingsException {
    return new StreamableOperator() {

        @Override
        public void runFinal(final PortInput[] inputs, final PortOutput[] outputs, final ExecutionContext exec) throws Exception {
            TreeEnsembleModelPortObject model = (TreeEnsembleModelPortObject) ((PortObjectInput) inputs[0]).getPortObject();
            TreeEnsembleModelPortObjectSpec modelSpec = model.getSpec();
            DataTableSpec dataSpec = (DataTableSpec) inSpecs[1];
            final TreeEnsemblePredictor pred = new TreeEnsemblePredictor(modelSpec, model, dataSpec, m_configuration);
            ColumnRearranger rearranger = pred.getPredictionRearranger();
            StreamableFunction func = rearranger.createStreamableFunction(1, 0);
            func.runFinal(inputs, outputs, exec);
        }
    };
}