/// <inheritdoc />
public double Transform(double[] data)
{
if (double.IsNaN(data[_fieldIdx]))
{
throw new InvalidOperationException($"Invalid data value at input field index {_fieldIdx} (NaN).");
}
_lastValues.Enqueue(data[_fieldIdx], true);
BasicStat stat = new BasicStat();
for (int i = 0; i < _lastValues.Count; i++)
{
stat.AddSample(_lastValues.GetElementAt(i, true));
}
return(stat.Get(_cfg.OutputFigure));
}
/// <summary>
/// Updates the predictors provider.
/// </summary>
/// <param name="activation">The current value of the activation.</param>
/// <param name="normalizedActivation">The current value of the activation normalized between 0 and 1.</param>
/// <param name="spike">Indicates whether the neuron is currently firing.</param>
public void Update(double activation, double normalizedActivation, bool spike)
{
//Continuous statistics
double activationDiff = activation - _lastActivation;
_activationStat?.AddSample(activation);
_activationDiffStat?.AddSample(activationDiff);
//Data windows
_activationMDW?.AddSample(activation);
_firingMDW?.Enqueue(spike ? (byte)1 : (byte)0);
//Predictors
foreach (IPredictor predictor in _predictorCollection)
{
predictor.Update(activation, normalizedActivation, spike);
}
//Last activation
_lastActivation = activation;
_lastNormalizedActivation = normalizedActivation;
_lastSpike = spike;
return;
}
/// <summary>
/// Computes the rescaled range.
/// </summary>
public double Compute()
{
double rescaledRange = 0;
if (_valueCollection.Count > 0)
{
BasicStat devStat = new BasicStat();
Interval cumulRange = new Interval();
double mean = _sum / _valueCollection.Count;
double cumulDeviation = 0;
for (int i = 0; i < _valueCollection.Count; i++)
{
devStat.AddSample(_valueCollection[i] - mean);
cumulDeviation += _valueCollection[i] - mean;
cumulRange.Adjust(cumulDeviation);
}
if (devStat.StdDev != 0)
{
rescaledRange = cumulRange.Span / devStat.StdDev;
}
}
return(rescaledRange);
}
/// <summary>
/// Checks the predictors and sets the general enabling/disabling switches.
/// </summary>
/// <param name="predictorsCollection">The collection of predictors.</param>
private void InitOutputFeaturesGeneralSwitches(List <double[]> predictorsCollection)
{
//Allocate general switches
OutputFeatureGeneralSwitchCollection = new bool[PredictorDescriptorCollection.Count];
//Init general predictor switches to false
OutputFeatureGeneralSwitchCollection.Populate(false);
//Compute statistics on predictors
Tuple <int, double>[] predictorValueSpanCollection = new Tuple <int, double> [PredictorDescriptorCollection.Count];
Parallel.For(0, PredictorDescriptorCollection.Count, i =>
{
BasicStat stat = new BasicStat();
for (int row = 0; row < predictorsCollection.Count; row++)
{
stat.AddSample(predictorsCollection[row][i]);
}
//Use predictor's value span as a differentiator
predictorValueSpanCollection[i] = new Tuple <int, double>(i, stat.Span);
});
//Sort collected predictor differentiators
Array.Sort(predictorValueSpanCollection, ComparePredictors);
//Enable predictors
int numOfPredictorsToBeRejected = (int)(Math.Round(PredictorDescriptorCollection.Count * _preprocessorCfg.PredictorsReductionRatio));
int firstIndexToBeRejected = predictorValueSpanCollection.Length - numOfPredictorsToBeRejected;
NumOfActivePredictors = 0;
for (int i = 0; i < predictorValueSpanCollection.Length; i++)
{
if (predictorValueSpanCollection[i].Item2 > _preprocessorCfg.PredictorValueMinSpan && i < firstIndexToBeRejected)
{
//Enable predictor
OutputFeatureGeneralSwitchCollection[predictorValueSpanCollection[i].Item1] = true;
++NumOfActivePredictors;
}
}
return;
}