/**
* Initializes training. Runs through all data points in the training set and updates the weight
* vector whenever a classification error occurs.
*
* <p>Can be called multiple times.
*
* @param dataset the dataset to train on. Each column is treated as point.
* @param labelset the set of labels, one for each data point. If the cardinalities of data- and
* labelset do not match, a CardinalityException is thrown
*/
public void train(Vector labelset, Matrix dataset) throws TrainingException {
if (labelset.size() != dataset.columnSize()) {
throw new CardinalityException(labelset.size(), dataset.columnSize());
}
boolean converged = false;
int iteration = 0;
while (!converged) {
if (iteration > 1000) {
throw new TrainingException("Too many iterations needed to find hyperplane.");
}
converged = true;
int columnCount = dataset.columnSize();
for (int i = 0; i < columnCount; i++) {
Vector dataPoint = dataset.viewColumn(i);
log.debug("Training point: {}", dataPoint);
synchronized (this.model) {
boolean prediction = model.classify(dataPoint);
double label = labelset.get(i);
if (label <= 0 && prediction || label > 0 && !prediction) {
log.debug("updating");
converged = false;
update(label, dataPoint, this.model);
}
}
}
iteration++;
}
}
@Override
public double getScore(DocData Doc, QueryData query, RankingModel rModel) {
HashMap<String, Double> FeatureWeights = rModel.FeatureWeights;
TermFreqUtil.enableStemming = true;
List<String> termVector = new ArrayList<String>();
List<Double> queryVector =
Vectorizer.getQueryVector(
query.GetQueryString(),
termVector,
TermSmoothingType.LOGARITHM,
DocSmoothingType.IDF,
rModel);
Map<DocFeature, List<Double>> docVector =
Vectorizer.getDocumentVector(
Doc,
termVector,
TermSmoothingType.LOGARITHM,
DocSmoothingType.NONE,
NormalizationType.LENGTH,
rModel);
HashMap<String, Double> FeatureVariable = new HashMap<String, Double>();
double FeatureURL = VectorUtil.dot(queryVector, docVector.get(DocFeature.URL));
double FeatureTitle = VectorUtil.dot(queryVector, docVector.get(DocFeature.TITLE));
double FeatureBody = VectorUtil.dot(queryVector, docVector.get(DocFeature.BODY));
double FeatureHeaderL = VectorUtil.dot(queryVector, docVector.get(DocFeature.HEADER));
double FeatureAnchor = VectorUtil.dot(queryVector, docVector.get(DocFeature.ANCHOR));
double FirstPos = getFirstPos(Doc, query);
FeatureVariable.put("task4_W_url", FeatureURL);
FeatureVariable.put("task4_W_title", FeatureTitle);
FeatureVariable.put("task4_W_body", FeatureBody);
FeatureVariable.put("task4_W_header", FeatureHeaderL);
FeatureVariable.put("task4_W_anchor", FeatureAnchor);
FeatureVariable.put("task4_W_first_pos", FirstPos);
try {
return LinearModel.CalculateScore(FeatureVariable, FeatureWeights);
} catch (NoSuchFieldException e) {
e.printStackTrace();
}
return 0;
}
