public HistogramAggregator(int nClasses, HistogramData data)
{
// bins_ = new int[nClasses];
dataHandle_ = data.GetMemory();
sampleCount_ = 0;
}
public void Aggregate(IDataPointCollection data, int index, Object userData)
{
DataPointCollection concreteData = (DataPointCollection)(data);
HistogramData histogramData = (HistogramData)(userData);
histogramData.Increment(dataHandle_, concreteData.GetIntegerLabel((int)index));
// bins_[concreteData.GetIntegerLabel((int)index)]++;
sampleCount_ += 1;
}
//public int FindTallestBinIndex()
//{
// int maxCount = bins_[0];
// int tallestBinIndex = 0;
// for (int i = 1; i < bins_.Length; i++)
// {
// if (bins_[i] > maxCount)
// {
// maxCount = bins_[i];
// tallestBinIndex = i;
// }
// }
// return tallestBinIndex;
//}
#region IStatisticsAggregator implementation
public void Clear(Object userData)
{
HistogramData histogramData = (HistogramData)(userData);
histogramData.Clear(dataHandle_);
//for (int b = 0; b < bins_.Length; b++)
// bins_[b] = 0;
sampleCount_ = 0;
}
public void Aggregate(HistogramAggregator aggregator, Object userData)
{
// System.Diagnostics.Debug.Assert(aggregator.bins_.Length == bins_.Length);
HistogramData histogramData = (HistogramData)(userData);
histogramData.Aggregate(dataHandle_, aggregator.dataHandle_);
//for (int b = 0; b < bins_.Length; b++)
// bins_[b] += aggregator.bins_[b];
sampleCount_ += aggregator.sampleCount_;
}
public double Entropy(HistogramData data)
{
//if (sampleCount_ == 0)
// return 0.0;
//double result = 0.0;
//for (int b = 0; b < bins_.Length; b++)
//{
// double p = (double)bins_[b] / (double)sampleCount_;
// result -= p == 0.0 ? 0.0 : p * Math.Log(p, 2.0);
//}
//return result;
return(data.Entropy(dataHandle_, sampleCount_));
}
public float GetProbability(int classIndex, HistogramData data)
{
// return (float)(bins_[classIndex]) / sampleCount_;
return((float)(data.GetBin(dataHandle_, classIndex) / sampleCount_));
}