/// <summary>
/// Learns a function based on the cumulative density function of a normal distribution parameterized by
/// a mean and variance as observed during fitting.
/// </summary>
/// <param name="input">The column containing the vectors to apply the normalization to.</param>
/// <param name="ensureZeroUntouched">If set to <c>false</c>, then the learned distributional parameters will be
/// adjusted in such a way as to ensure that the input 0 maps to the output 0.
/// This is valuable for the sake of sparsity preservation, if normalizing sparse vectors.</param>
/// <param name="useLog">If set to true then we transform over the logarithm of the values, rather
/// than just the raw values. If this is set to <c>true</c> then <paramref name="ensureZeroUntouched"/> is ignored.</param>
/// <param name="maximumExampleCount">When gathering statistics only look at most this many examples.</param>
/// <param name="onFit">A delegate called whenever the estimator is fit, with the learned mean and standard
/// deviation for all slots.</param>
/// <remarks>Note that the statistics gathering and normalization is done independently per slot of the
/// vector values.</remarks>
/// <returns>The normalized column.</returns>
public static NormVector <double> NormalizeByCumulativeDistribution(
this Vector <double> input, bool ensureZeroUntouched = NormalizingEstimator.Defaults.EnsureZeroUntouched,
bool useLog = false, long maximumExampleCount = NormalizingEstimator.Defaults.MaximumExampleCount,
OnFitCumulativeDistribution <ImmutableArray <double> > onFit = null)
{
return(NormalizeByMVCdfCore(input, ensureZeroUntouched, useLog, true, maximumExampleCount, CdfMapper(onFit)));
}
private static Action <IColumnFunction> CdfMapper <TData>(OnFitCumulativeDistribution <TData> onFit)
{
Contracts.AssertValueOrNull(onFit);
if (onFit == null)
{
return(null);
}
return(col =>
{
var aCol = (NormalizerTransformer.ICdfData <TData>)col;
onFit(aCol.Mean, aCol.Stddev);
});
}
private static Action <IColumnFunction> CdfMapper <TData>(OnFitCumulativeDistribution <TData> onFit)
{
Contracts.AssertValueOrNull(onFit);
if (onFit == null)
{
return(null);
}
return(col =>
{
var aCol = (NormalizingTransformer.CdfNormalizerModelParameters <TData>)col?.GetNormalizerModelParams();
onFit(aCol.Mean, aCol.Stddev);
});
}
/// <summary>
/// Learns a function based on the cumulative density function of a normal distribution parameterized by
/// a mean and variance as observed during fitting.
/// </summary>
/// <param name="input">The input column.</param>
/// <param name="fixZero">If set to <c>false</c>, then the learned distributional parameters will be
/// adjusted in such a way as to ensure that the input 0 maps to the output 0.
/// This is valuable for the sake of sparsity preservation, if normalizing sparse vectors.</param>
/// <param name="useLog">If set to true then we transform over the logarithm of the values, rather
/// than just the raw values. If this is set to <c>true</c> then <paramref name="fixZero"/> is ignored.</param>
/// <param name="maxTrainingExamples">When gathering statistics only look at most this many examples.</param>
/// <param name="onFit">A delegate called whenever the estimator is fit, with the learned mean and standard
/// deviation for all slots.</param>
/// <remarks>Note that the statistics gathering and normalization is done independently per slot of the
/// vector values.</remarks>
/// <returns>The normalized column.</returns>
public static NormVector <double> NormalizeByCumulativeDistribution(
this Vector <double> input, bool fixZero = FZ, bool useLog = false, long maxTrainingExamples = MaxTrain,
OnFitCumulativeDistribution <ImmutableArray <double> > onFit = null)
{
return(NormalizeByMVCdfCore(input, fixZero, useLog, true, maxTrainingExamples, CdfMapper(onFit)));
}
