/// <summary> Sets the format of the input instances.
///
/// </summary>
/// <param name="instanceInfo">an Instances object containing the input instance
/// structure (any instances contained in the object are ignored - only the
/// structure is required).
/// </param>
/// <returns> true if the outputFormat may be collected immediately
/// </returns>
/// <exception cref="Exception">if the inputFormat can't be set successfully
/// </exception>
public override bool setInputFormat(Instances instanceInfo)
{
base.setInputFormat(instanceInfo);
m_removeFilter = null;
return false;
}
public void EndTrainingSession()
{
Stream s = new MemoryStream ();
TextWriter tw = new StreamWriter (s);
AbstractBasicTextVector.WriteInstancesArff (tw, vectors, "c45recommender", tags, results);
tw.Flush ();
s.Position = 0;
Instances source = new Instances (new InputStreamReader (new InputStreamWrapper (s)));
tw.Close ();
s.Close ();
Instances[] derived = new Instances[this.not];
classifiers = new AbstractClassifier[this.not];
int[] args = new int[this.not - 1];
int l = source.numAttributes () - this.not;
for (int i = 0; i < this.not-1; i++) {
args [i] = i + l + 1;
}
for (int i = 0; i < this.not; i++) {
Remove rem = new Remove ();
rem.setAttributeIndicesArray (args);
rem.setInputFormat (source);
derived [i] = Filter.useFilter (source, rem);
classifiers [i] = GenerateClassifier ();
derived [i].setClassIndex (derived [i].numAttributes () - 1);
classifiers [i].buildClassifier (derived [i]);
if (i < this.not - 1) {
args [i] = l + i;
}
}
datasets = derived;
}
/// <summary> Signify that this batch of input to the filter is finished.
///
/// </summary>
/// <returns> true if there are instances pending output
/// </returns>
public override bool batchFinished()
{
if (getInputFormat() == null)
{
throw new System.SystemException("No input instance format defined");
}
if (m_removeFilter == null)
{
// establish attributes to remove from first batch
Instances toFilter = getInputFormat();
int[] attsToDelete = new int[toFilter.numAttributes()];
int numToDelete = 0;
for (int i = 0; i < toFilter.numAttributes(); i++)
{
if (i == toFilter.classIndex())
continue; // skip class
AttributeStats stats = toFilter.attributeStats(i);
if (stats.distinctCount < 2)
{
// remove constant attributes
attsToDelete[numToDelete++] = i;
}
else if (toFilter.attribute(i).Nominal)
{
// remove nominal attributes that vary too much
double variancePercent = (double) stats.distinctCount / (double) stats.totalCount * 100.0;
if (variancePercent > m_maxVariancePercentage)
{
attsToDelete[numToDelete++] = i;
}
}
}
int[] finalAttsToDelete = new int[numToDelete];
Array.Copy(attsToDelete, 0, finalAttsToDelete, 0, numToDelete);
m_removeFilter = new Remove();
m_removeFilter.SetAttributeIndicesArray(finalAttsToDelete);
m_removeFilter.set_InvertSelection(false);
m_removeFilter.setInputFormat(toFilter);
for (int i = 0; i < toFilter.numInstances(); i++)
{
m_removeFilter.input(toFilter.instance(i));
}
m_removeFilter.batchFinished();
Instance processed;
Instances outputDataset = m_removeFilter.getOutputFormat();
// restore old relation name to hide attribute filter stamp
outputDataset.RelationName = toFilter.relationName();
setOutputFormat(outputDataset);
while ((processed = m_removeFilter.output()) != null)
{
processed.Dataset = outputDataset;
push(processed);
}
}
flushInput();
m_NewBatch = true;
return (numPendingOutput() != 0);
}
