Examples with Attribute weka.core.Attribute used on opensource projects

Example 11 with Attribute

use of weka.core.Attribute in project lobcder by skoulouzis.

the class LDClustering method initAttributes.

private void initAttributes() throws ParseException, Exception {
    int index = 0;
    Attribute uidAttribute = new Attribute("uid", index++);
    // Declare a nominal attribute along with its values
    FastVector verbVector = new FastVector(Request.Method.values().length);
    for (Request.Method m : Request.Method.values()) {
        verbVector.addElement(m.code);
    }
    Attribute verbAttribute = new Attribute("verb", verbVector, index++);
    Attribute checksumAttribute = new Attribute("checksum", (FastVector) null, index++);
    Attribute contentTypeAttribute = new Attribute("contentType", (FastVector) null, index++);
    Attribute createDateAttribute = new Attribute("createDate", "yyyy-MM-dd HH:mm:ss", index++);
    Attribute locationPreferenceAttribute = new Attribute("locationPreference", (FastVector) null, index++);
    Attribute descriptionAttribute = new Attribute("description", (FastVector) null, index++);
    Attribute validationDateAttribute = new Attribute("validationDate", "yyyy-MM-dd HH:mm:ss", index++);
    Attribute lengthAttribute = new Attribute("length", index++);
    Attribute modifiedDateAttribute = new Attribute("modifiedDate", "yyyy-MM-dd HH:mm:ss", index++);
    Attribute pathAttribute = new Attribute("name", (FastVector) null, index++);
    Attribute parentRefAttribute = new Attribute("parentRef", index++);
    Attribute statusAttribute = new Attribute("status", (FastVector) null, index++);
    FastVector typeVector = new FastVector(3);
    typeVector.addElement(nl.uva.cs.lobcder.util.Constants.LOGICAL_DATA);
    typeVector.addElement(nl.uva.cs.lobcder.util.Constants.LOGICAL_FILE);
    typeVector.addElement(nl.uva.cs.lobcder.util.Constants.LOGICAL_FOLDER);
    Attribute typeAttribute = new Attribute("type", typeVector, index++);
    // Declare the class attribute along with its values
    FastVector supervisedVector = new FastVector(2);
    supervisedVector.addElement("true");
    supervisedVector.addElement("false");
    Attribute supervisedAttribute = new Attribute("supervised", supervisedVector, index++);
    Attribute ownerAttribute = new Attribute("owner", (FastVector) null, index++);
    // Declare the feature vector
    metdataAttributes = new FastVector();
    // 0
    metdataAttributes.addElement(uidAttribute);
    // 1
    metdataAttributes.addElement(verbAttribute);
    // 2
    metdataAttributes.addElement(checksumAttribute);
    // 3
    metdataAttributes.addElement(contentTypeAttribute);
    // 4
    metdataAttributes.addElement(createDateAttribute);
    // 5
    metdataAttributes.addElement(locationPreferenceAttribute);
    // 6
    metdataAttributes.addElement(descriptionAttribute);
    // 7
    metdataAttributes.addElement(validationDateAttribute);
    // 8
    metdataAttributes.addElement(lengthAttribute);
    // 9
    metdataAttributes.addElement(modifiedDateAttribute);
    // 10
    metdataAttributes.addElement(pathAttribute);
    // 11
    metdataAttributes.addElement(parentRefAttribute);
    // 12
    metdataAttributes.addElement(statusAttribute);
    // 13
    metdataAttributes.addElement(typeAttribute);
    // 14
    metdataAttributes.addElement(supervisedAttribute);
    // 15
    metdataAttributes.addElement(ownerAttribute);
}

Example 12 with Attribute

use of weka.core.Attribute in project dkpro-tc by dkpro.

the class WekaUtils method instanceListToArffFileMultiLabel.

/**
 * Converts a feature store to a list of instances. Multi-label case.
 *
 * @param outputFile
 *            the output file
 * @param instances
 *            the instances to convert
 * @param useDenseInstances
 *            dense features
 * @param useWeights
 *            use weights
 * @throws Exception
 *             in case of errors
 */
public static void instanceListToArffFileMultiLabel(File outputFile, List<Instance> instances, boolean useDenseInstances, boolean useWeights) throws Exception {
    // Filter preprocessingFilter = new ReplaceMissingValuesWithZeroFilter();
    AttributeStore attributeStore = WekaFeatureEncoder.getAttributeStore(instances);
    List<String> outcomes = new ArrayList<>();
    for (Instance i : instances) {
        outcomes.add(i.getOutcome());
    }
    List<Attribute> outcomeAttributes = createOutcomeAttributes(new ArrayList<String>(outcomes));
    // in Meka, class label attributes have to go on top
    for (Attribute attribute : outcomeAttributes) {
        attributeStore.addAttributeAtBegin(attribute.name(), attribute);
    }
    // for Meka-internal use
    Instances wekaInstances = new Instances(RELATION_NAME + ": -C " + outcomeAttributes.size() + " ", attributeStore.getAttributes(), instances.size());
    wekaInstances.setClassIndex(outcomeAttributes.size());
    if (!outputFile.exists()) {
        outputFile.mkdirs();
        outputFile.createNewFile();
    }
    ArffSaver saver = new ArffSaver();
    // preprocessingFilter.setInputFormat(wekaInstances);
    saver.setRetrieval(Saver.INCREMENTAL);
    saver.setFile(outputFile);
    saver.setCompressOutput(true);
    saver.setInstances(wekaInstances);
    for (int i = 0; i < instances.size(); i++) {
        Instance instance = instances.get(i);
        double[] featureValues = getFeatureValues(attributeStore, instance);
        // set class label values
        List<String> instanceOutcome = instance.getOutcomes();
        for (Attribute label : outcomeAttributes) {
            String labelname = label.name();
            featureValues[attributeStore.getAttributeOffset(labelname)] = instanceOutcome.contains(labelname.split(CLASS_ATTRIBUTE_PREFIX)[1]) ? 1.0d : 0.0d;
        }
        weka.core.Instance wekaInstance;
        if (useDenseInstances) {
            wekaInstance = new DenseInstance(1.0, featureValues);
        } else {
            wekaInstance = new SparseInstance(1.0, featureValues);
        }
        wekaInstance.setDataset(wekaInstances);
        Double instanceWeight = instance.getWeight();
        if (useWeights) {
            wekaInstance.setWeight(instanceWeight);
        }
        // preprocessingFilter.input(wekaInstance);
        // saver.writeIncremental(preprocessingFilter.output());
        saver.writeIncremental(wekaInstance);
    }
    // finishes the incremental saving process
    saver.writeIncremental(null);
}

Example 13 with Attribute

use of weka.core.Attribute in project dkpro-tc by dkpro.

the class WekaUtils method tcInstanceToMekaInstance.

/**
 * Converts a TC instance object into a Meka instance object, compatible with the given
 * attribute set and class labels.
 *
 * @param instance
 *            tc instance
 * @param trainingData
 *            training data
 * @param allClassLabels
 *            all labels
 * @return weka instance
 * @throws Exception
 *             in case of errors
 */
public static weka.core.Instance tcInstanceToMekaInstance(Instance instance, Instances trainingData, List<String> allClassLabels) throws Exception {
    AttributeStore attributeStore = new AttributeStore();
    List<Attribute> outcomeAttributes = createOutcomeAttributes(allClassLabels);
    // in Meka, class label attributes have to go on top
    for (Attribute attribute : outcomeAttributes) {
        attributeStore.addAttributeAtBegin(attribute.name(), attribute);
    }
    for (int i = outcomeAttributes.size(); i < trainingData.numAttributes(); i++) {
        attributeStore.addAttribute(trainingData.attribute(i).name(), trainingData.attribute(i));
    }
    double[] featureValues = getFeatureValues(attributeStore, instance);
    SparseInstance sparseInstance = new SparseInstance(1.0, featureValues);
    trainingData.setClassIndex(outcomeAttributes.size());
    sparseInstance.setDataset(trainingData);
    return sparseInstance;
}

Example 14 with Attribute

use of weka.core.Attribute in project dkpro-tc by dkpro.

the class WekaUtils method tcInstanceToWekaInstance.

/**
 * Converts a TC instance object into a Weka instance object, compatible with the given
 * attribute set and class labels.
 *
 * @param instance
 *            tc instance
 * @param trainingData
 *            training data
 * @param allClasses
 *            all classes
 * @param isRegressionExperiment
 *            is regression
 * @return weka instance
 * @throws Exception
 *             in case of errors
 */
public static weka.core.Instance tcInstanceToWekaInstance(Instance instance, Instances trainingData, List<String> allClasses, boolean isRegressionExperiment) throws Exception {
    AttributeStore attributeStore = new AttributeStore();
    // outcome attribute is last and will be ignored
    for (int i = 0; i < trainingData.numAttributes() - 1; i++) {
        attributeStore.addAttribute(trainingData.attribute(i).name(), trainingData.attribute(i));
    }
    // add outcome attribute
    Attribute outcomeAttribute = createOutcomeAttribute(allClasses, isRegressionExperiment);
    attributeStore.addAttribute(outcomeAttribute.name(), outcomeAttribute);
    double[] featureValues = getFeatureValues(attributeStore, instance);
    SparseInstance sparseInstance = new SparseInstance(1.0, featureValues);
    sparseInstance.setDataset(trainingData);
    return sparseInstance;
}

Example 15 with Attribute

use of weka.core.Attribute in project dkpro-tc by dkpro.

the class WekaUtils method featureSelectionMultilabel.

/**
 * Feature selection using Mulan.
 *
 * @param aContext
 *            Lab context
 * @param trainData
 *            training data
 * @param attributeEvaluator
 *            evaluator
 * @param labelTransformationMethod
 *            transformation method
 * @param numLabelsToKeep
 *            mapping
 * @return remove object
 * @throws TextClassificationException
 *             in case of errors
 */
public static Remove featureSelectionMultilabel(TaskContext aContext, Instances trainData, List<String> attributeEvaluator, String labelTransformationMethod, int numLabelsToKeep) throws TextClassificationException {
    // file to hold the results of attribute selection
    File fsResultsFile = getFile(aContext, TEST_TASK_OUTPUT_KEY, WekaTestTask.featureSelectionFile, AccessMode.READWRITE);
    // filter for reducing dimension of attributes
    Remove filterRemove = new Remove();
    try {
        MultiLabelInstances mulanInstances = convertMekaInstancesToMulanInstances(trainData);
        ASEvaluation eval = ASEvaluation.forName(attributeEvaluator.get(0), attributeEvaluator.subList(1, attributeEvaluator.size()).toArray(new String[0]));
        AttributeEvaluator attributeSelectionFilter;
        // is complicated due to missing commandline support of mulan):
        if (labelTransformationMethod.equals("LabelPowersetAttributeEvaluator")) {
            attributeSelectionFilter = new LabelPowersetAttributeEvaluator(eval, mulanInstances);
        } else if (labelTransformationMethod.equals("BinaryRelevanceAttributeEvaluator")) {
            attributeSelectionFilter = new BinaryRelevanceAttributeEvaluator(eval, mulanInstances, "max", "none", "rank");
        } else {
            throw new TextClassificationException("This Label Transformation Method is not supported.");
        }
        Ranker r = new Ranker();
        int[] result = r.search(attributeSelectionFilter, mulanInstances);
        // collect evaluation for *all* attributes and write to file
        StringBuffer evalFile = new StringBuffer();
        for (Attribute att : mulanInstances.getFeatureAttributes()) {
            evalFile.append(att.name() + ": " + attributeSelectionFilter.evaluateAttribute(att.index() - mulanInstances.getNumLabels()) + "\n");
        }
        FileUtils.writeStringToFile(fsResultsFile, evalFile.toString(), "utf-8");
        // create a filter to reduce the dimension of the attributes
        int[] toKeep = new int[numLabelsToKeep + mulanInstances.getNumLabels()];
        System.arraycopy(result, 0, toKeep, 0, numLabelsToKeep);
        int[] labelIndices = mulanInstances.getLabelIndices();
        System.arraycopy(labelIndices, 0, toKeep, numLabelsToKeep, mulanInstances.getNumLabels());
        filterRemove.setAttributeIndicesArray(toKeep);
        filterRemove.setInvertSelection(true);
        filterRemove.setInputFormat(mulanInstances.getDataSet());
    } catch (ArrayIndexOutOfBoundsException e) {
        // less attributes than we want => no filtering
        return null;
    } catch (Exception e) {
        throw new TextClassificationException(e);
    }
    return filterRemove;
}