weka.core.WekaException Java Examples

/**
 * Validate a data split of train and validation data
 *
 * @param trainData Training data
 * @param valData Validation data
 * @throws WekaException Invalid validation split
 */
protected void validateSplit(Instances trainData, Instances valData)
    throws WekaException {
  if (earlyStopping.getValidationSetPercentage() < 10e-8) {
    // Use no validation set at all
    return;
  }
  int classIndex = trainData.classIndex();
  int valDataNumDinstinctClassValues = valData.numDistinctValues(classIndex);
  int trainDataNumDistinctClassValues =
      trainData.numDistinctValues(classIndex);
  if (trainData.numClasses() > 1
      && valDataNumDinstinctClassValues != trainDataNumDistinctClassValues) {
    throw new InvalidValidationPercentageException(
        "The validation data did not contain the same classes as the training data. "
            + "You should increase the validation percentage in the EarlyStopping configuration.");
  }
}
 

/**
 * Validate a data split of train and validation data
 *
 * @param trainData Training data
 * @param valData Validation data
 * @throws WekaException Invalid validation split
 */
protected void validateSplit(Instances trainData, Instances valData)
    throws WekaException {
  if (earlyStopping.getValidationSetPercentage() < 10e-8) {
    // Use no validation set at all
    return;
  }
  int classIndex = trainData.classIndex();
  int valDataNumDinstinctClassValues = valData.numDistinctValues(classIndex);
  int trainDataNumDistinctClassValues =
      trainData.numDistinctValues(classIndex);
  if (trainData.numClasses() > 1
      && valDataNumDinstinctClassValues != trainDataNumDistinctClassValues) {
    throw new InvalidValidationPercentageException(
        "The validation data did not contain the same classes as the training data. "
            + "You should increase the validation percentage in the EarlyStopping configuration.");
  }
}
 

/**
 * Prints the classification to the buffer.
 * 
 * @param classifier the classifier to use for printing the classification
 * @param inst the instance to print
 * @param index the index of the instance
 * @throws Exception if check fails or error occurs during printing of
 *           classification
 */
public void printClassification(Classifier classifier, Instance inst,
    int index) throws Exception {
  String error;

  if ((error = checkBasic()) != null)
    throw new WekaException(error);

  doPrintClassification(classifier, inst, index);
}
 

/**
 * Prints the classification to the buffer.
 * 
 * @param dist the distribution from classifier for the supplied instance
 * @param inst the instance to print
 * @param index the index of the instance
 * @throws Exception if check fails or error occurs during printing of
 *           classification
 */
public void printClassification(double[] dist, Instance inst, int index)
    throws Exception {
  String error;

  if ((error = checkBasic()) != null)
    throw new WekaException(error);

  doPrintClassification(dist, inst, index);
}
 

/**
  * Determines the output format based on the input format and returns 
  * this. In case the output format cannot be returned immediately, i.e.,
  * hasImmediateOutputFormat() returns false, then this method will called
  * from batchFinished() after the call of preprocess(Instances), in which,
  * e.g., statistics for the actual processing step can be gathered.
  *
  * @param inputFormat     the input format to base the output format on
  * @return                the output format
  * @throws Exception      in case the determination goes wrong
  */
 protected Instances determineOutputFormat(Instances inputFormat) throws Exception {
   int			i;
   int[]		indices;
   StringBuilder	order;
   Instances		output;
   
   m_AttributeIndices.setUpper(inputFormat.numAttributes() - 1);
   order   = new StringBuilder();
   indices = m_AttributeIndices.getSelection();
   if (indices.length == 0)
     throw new WekaException("No attributes defined as class attributes!");
   for (i = 0; i < indices.length; i++) {
     if (i > 0)
order.append(",");
     order.append("" + (indices[i]+1));
   }
   for (i = 0; i < inputFormat.numAttributes(); i++) {
     if (m_AttributeIndices.isInRange(i))
continue;
     order.append(",");
     order.append("" + (i+1));
   }
   m_Reorder.setAttributeIndices(order.toString());
   m_Reorder.setInputFormat(inputFormat);
   
   output = m_Reorder.getOutputFormat();
   output.setClassIndex(indices.length);
   output.setRelationName("-C " + indices.length);
   
   return output;
 }
 

/**
  * Generates the classifier.
  *
  * @param instances the instances to be used for building the classifier
  * @throws Exception if the classifier can't be built successfully
  */
 public void buildClassifier(Instances instances) 
   throws Exception {
   
   boolean noRule = true;

   // can classifier handle the data?
   getCapabilities().testWithFail(instances);

   // remove instances with missing class
   Instances data = new Instances(instances);
   data.deleteWithMissingClass();

   // only class? -> build ZeroR model
   if (data.numAttributes() == 1) {
     System.err.println(
  "Cannot build model (only class attribute present in data!), "
  + "using ZeroR model instead!");
     m_ZeroR = new weka.classifiers.rules.ZeroR();
     m_ZeroR.buildClassifier(data);
     return;
   }
   else {
     m_ZeroR = null;
   }
   
   // for each attribute ...
   Enumeration enu = instances.enumerateAttributes();
   while (enu.hasMoreElements()) {
     try {
OneRRule r = newRule((Attribute) enu.nextElement(), data);

// if this attribute is the best so far, replace the rule
if (noRule || r.m_correct > m_rule.m_correct) {
  m_rule = r;
}
noRule = false;
     } catch (Exception ex) {
     }
   }
   
   if (noRule)
     throw new WekaException("No attributes found to work with!");
 }
 

/**
 * Computes the distribution for a given instance.
 *
 * @param instance 		the instance for which distribution is computed
 * @return 			the distribution
 * @throws Exception 		if the distribution can't be computed successfully
 */
public double[] distributionForInstance (Instance instance) throws Exception {

  if (!getDoNotReplaceMissingValues()) {
    m_ReplaceMissingValues.input(instance);
    m_ReplaceMissingValues.batchFinished();
    instance = m_ReplaceMissingValues.output();
  }

  if (getConvertNominalToBinary() 
      && m_NominalToBinary != null) {
    m_NominalToBinary.input(instance);
    m_NominalToBinary.batchFinished();
    instance = m_NominalToBinary.output();
  }

  if (m_Filter != null) {
    m_Filter.input(instance);
    m_Filter.batchFinished();
    instance = m_Filter.output();
  }

  Object x = instanceToArray(instance);
  double v;
  double[] result = new double[instance.numClasses()];
  if (m_ProbabilityEstimates) {
    if (m_SVMType != SVMTYPE_L2_LR) {
      throw new WekaException("probability estimation is currently only " +
          "supported for L2-regularized logistic regression");
    }

    int[] labels = (int[])invokeMethod(m_Model, "getLabels", null, null);
    double[] prob_estimates = new double[instance.numClasses()];

    v = ((Integer) invokeMethod(
          Class.forName(CLASS_LINEAR).newInstance(),
          "predictProbability",
          new Class[]{
            Class.forName(CLASS_MODEL),
      Array.newInstance(Class.forName(CLASS_FEATURENODE), Array.getLength(x)).getClass(),
      Array.newInstance(Double.TYPE, prob_estimates.length).getClass()},
      new Object[]{ m_Model, x, prob_estimates})).doubleValue();

    // Return order of probabilities to canonical weka attribute order
    for (int k = 0; k < prob_estimates.length; k++) {
      result[labels[k]] = prob_estimates[k];
    }
  }
  else {
    v = ((Integer) invokeMethod(
          Class.forName(CLASS_LINEAR).newInstance(),
          "predict",
          new Class[]{
            Class.forName(CLASS_MODEL),
      Array.newInstance(Class.forName(CLASS_FEATURENODE), Array.getLength(x)).getClass()},
      new Object[]{
        m_Model,
      x})).doubleValue();

    assert (instance.classAttribute().isNominal());
    result[(int) v] = 1;
  }

  return result;
}
 

/**
 * Parses a given list of options. <p/>
 * 
 <!-- options-start -->
 * Valid options are: <p/>
 * 
 * <pre> -D
 *  Produce debugging output.
 *  (default no debugging output)</pre>
 * 
 * <pre> -E &lt;estimator&gt;
 *  The estimator can be one of the following:
 *   eb -- Empirical Bayes estimator for noraml mixture (default)
 *   nested -- Optimal nested model selector for normal mixture
 *   subset -- Optimal subset selector for normal mixture
 *   pace2 -- PACE2 for Chi-square mixture
 *   pace4 -- PACE4 for Chi-square mixture
 *   pace6 -- PACE6 for Chi-square mixture
 * 
 *   ols -- Ordinary least squares estimator
 *   aic -- AIC estimator
 *   bic -- BIC estimator
 *   ric -- RIC estimator
 *   olsc -- Ordinary least squares subset selector with a threshold</pre>
 * 
 * <pre> -S &lt;threshold value&gt;
 *  Threshold value for the OLSC estimator</pre>
 * 
 <!-- options-end -->
 * 
 * @param options the list of options as an array of strings
 * @throws Exception if an option is not supported
 */
public void setOptions(String[] options) throws Exception {
  
  setDebug(Utils.getFlag('D', options));

  String estimator = Utils.getOption('E', options);
  if ( estimator.equals("ols") ) paceEstimator = olsEstimator;
  else if ( estimator.equals("olsc") ) paceEstimator = olscEstimator;
  else if( estimator.equals("eb") || estimator.equals("") ) 
    paceEstimator = ebEstimator;
  else if ( estimator.equals("nested") ) paceEstimator = nestedEstimator;
  else if ( estimator.equals("subset") ) paceEstimator = subsetEstimator;
  else if ( estimator.equals("pace2") ) paceEstimator = pace2Estimator; 
  else if ( estimator.equals("pace4") ) paceEstimator = pace4Estimator;
  else if ( estimator.equals("pace6") ) paceEstimator = pace6Estimator;
  else if ( estimator.equals("aic") ) paceEstimator = aicEstimator;
  else if ( estimator.equals("bic") ) paceEstimator = bicEstimator;
  else if ( estimator.equals("ric") ) paceEstimator = ricEstimator;
  else throw new WekaException("unknown estimator " + estimator + 
		 " for -E option" );

  String string = Utils.getOption('S', options);
  if( ! string.equals("") ) olscThreshold = Double.parseDouble( string );
  
}
 

/**
 * Convert an arbitrary NDArray to Weka instances
 *
 * @param ndArray Input array
 * @param inputFormat Format to use for the instances
 * @param attributesPerLayer Hashmap of layer names and how many attributes there are per layer
 * @return Instances object
 * @throws WekaException Invalid input
 */
public static Instances ndArrayToInstances(INDArray ndArray, Instances inputFormat, Map<String, Long> attributesPerLayer) throws WekaException {
  int numInstances = (int) ndArray.size(0);
  long[] shape = ndArray.shape();
  int dims = shape.length;
  if (dims != 2) {
    throw new WekaException("Invalid input, NDArray shape needs to be two dimensional "
        + "but was " + Arrays.toString(shape));
  }

  long numAttributes = shape[1];
  int classI = -1;
  if (inputFormat != null) {
    classI = (int) (numAttributes - 1);
  }

  // Create the new attribute names
  ArrayList<Attribute> atts = new ArrayList<>();
  for (int i = 0; i < numAttributes; i++) {
    if (i == classI && inputFormat != null) {
      if (inputFormat.classAttribute().isNominal())
        atts.add(copyNominalAttribute(inputFormat.classAttribute()));
      else
        atts.add(new Attribute(inputFormat.classAttribute().name()));
    } else {
      atts.add(new Attribute(getAttributeName(attributesPerLayer, i)));
    }
  }

  // Actually create the instances from the values in the given NDArray
  Instances instances = new Instances("Transformed", atts, numInstances);
  instances.setClassIndex(classI);
  for (int i = 0; i < numInstances; i++) {
    INDArray row = ndArray.get(NDArrayIndex.point(i));
    double[] instanceVals = row.toDoubleVector();
    Instance inst = new DenseInstance(1.0, instanceVals);
    inst.setDataset(instances);
    instances.add(inst);
  }

  return instances;
}
 

/**
 * Convert an arbitrary NDArray to Weka instances
 *
 * @param ndArray Input array
 * @param inputFormat Format to use for the instances
 * @param attributesPerLayer Hashmap of layer names and how many attributes there are per layer
 * @return Instances object
 * @throws WekaException Invalid input
 */
public static Instances ndArrayToInstances(INDArray ndArray, Instances inputFormat, Map<String, Long> attributesPerLayer) throws WekaException {
  int numInstances = (int) ndArray.size(0);
  long[] shape = ndArray.shape();
  int dims = shape.length;
  if (dims != 2) {
    throw new WekaException("Invalid input, NDArray shape needs to be two dimensional "
        + "but was " + Arrays.toString(shape));
  }

  long numAttributes = shape[1];
  int classI = -1;
  if (inputFormat != null) {
    classI = (int) (numAttributes - 1);
  }

  // Create the new attribute names
  ArrayList<Attribute> atts = new ArrayList<>();
  for (int i = 0; i < numAttributes; i++) {
    if (i == classI && inputFormat != null) {
      if (inputFormat.classAttribute().isNominal())
        atts.add(copyNominalAttribute(inputFormat.classAttribute()));
      else
        atts.add(new Attribute(inputFormat.classAttribute().name()));
    } else {
      atts.add(new Attribute(getAttributeName(attributesPerLayer, i)));
    }
  }

  // Actually create the instances from the values in the given NDArray
  Instances instances = new Instances("Transformed", atts, numInstances);
  instances.setClassIndex(classI);
  for (int i = 0; i < numInstances; i++) {
    INDArray row = ndArray.get(NDArrayIndex.point(i));
    double[] instanceVals = row.toDoubleVector();
    Instance inst = new DenseInstance(1.0, instanceVals);
    inst.setDataset(instances);
    instances.add(inst);
  }

  return instances;
}