/// <param name="dataset"/>
/// <returns>a new GeneralDataset whose features and ids map exactly to those of this GeneralDataset. But labels are converted to be another set of labels</returns>
public virtual Edu.Stanford.Nlp.Classify.GeneralDataset <L2, F> MapDataset <L2>(Edu.Stanford.Nlp.Classify.GeneralDataset <L, F> dataset, IIndex <L2> newLabelIndex, IDictionary <L, L2> labelMapping, L2 defaultLabel)
{
Edu.Stanford.Nlp.Classify.GeneralDataset <L2, F> newDataset;
if (dataset is RVFDataset)
{
newDataset = new RVFDataset <L2, F>(this.featureIndex, newLabelIndex);
}
else
{
newDataset = new Dataset <L2, F>(this.featureIndex, newLabelIndex);
}
this.featureIndex.Lock();
this.labelIndex.Lock();
//System.out.println("inside mapDataset: dataset size:"+dataset.size());
for (int i = 0; i < dataset.Size(); i++)
{
//System.out.println("inside mapDataset: adding datum number"+i);
IDatum <L, F> d = dataset.GetDatum(i);
IDatum <L2, F> d2 = MapDatum(d, labelMapping, defaultLabel);
newDataset.Add(d2);
}
//System.out.println("old Dataset stats: numData:"+dataset.size()+" numfeatures:"+dataset.featureIndex().size()+" numlabels:"+dataset.labelIndex.size());
//System.out.println("new Dataset stats: numData:"+newDataset.size()+" numfeatures:"+newDataset.featureIndex().size()+" numlabels:"+newDataset.labelIndex.size());
//System.out.println("this dataset stats: numData:"+size()+" numfeatures:"+featureIndex().size()+" numlabels:"+labelIndex.size());
this.featureIndex.Unlock();
this.labelIndex.Unlock();
return(newDataset);
}
/// <param name="dataset"/>
/// <returns>
/// a new GeneralDataset whose features and ids map exactly to those of this GeneralDataset.
/// Useful when two Datasets are created independently and one wants to train a model on one dataset and test on the other. -Ramesh.
/// </returns>
public virtual Edu.Stanford.Nlp.Classify.GeneralDataset <L, F> MapDataset(Edu.Stanford.Nlp.Classify.GeneralDataset <L, F> dataset)
{
Edu.Stanford.Nlp.Classify.GeneralDataset <L, F> newDataset;
if (dataset is RVFDataset)
{
newDataset = new RVFDataset <L, F>(this.featureIndex, this.labelIndex);
}
else
{
newDataset = new Dataset <L, F>(this.featureIndex, this.labelIndex);
}
this.featureIndex.Lock();
this.labelIndex.Lock();
//System.out.println("inside mapDataset: dataset size:"+dataset.size());
for (int i = 0; i < dataset.Size(); i++)
{
//System.out.println("inside mapDataset: adding datum number"+i);
newDataset.Add(dataset.GetDatum(i));
}
//System.out.println("old Dataset stats: numData:"+dataset.size()+" numfeatures:"+dataset.featureIndex().size()+" numlabels:"+dataset.labelIndex.size());
//System.out.println("new Dataset stats: numData:"+newDataset.size()+" numfeatures:"+newDataset.featureIndex().size()+" numlabels:"+newDataset.labelIndex.size());
//System.out.println("this dataset stats: numData:"+size()+" numfeatures:"+featureIndex().size()+" numlabels:"+labelIndex.size());
this.featureIndex.Unlock();
this.labelIndex.Unlock();
return(newDataset);
}