source: branches/DataPreprocessing/HeuristicLab.Problems.DataAnalysis.Transformations/3.3/ShiftStandardDistributionTransformation.cs @ 10841

using System.Collections.Generic;
using System.Linq;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Parameters;
using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;

namespace HeuristicLab.Problems.DataAnalysis.Transformations {
  [Item("ShiftStandardDistributionTransformation.cs", "f(x) = ((x - m_org) / s_org ) * s_tar + m_tar | Represents Transformation to unit standard deviation and additional linear transformation to a target Mean and Standard deviation")]
  public class ShiftStandardDistributionTransformation : Transformation<double> {
    protected const string OriginalMeanParameterName = "OriginalMean";
    protected const string OriginalStandardDeviationParameterName = "OriginalStandardDeviation";
    protected const string MeanParameterName = "Mean";
    protected const string StandardDeviationParameterName = "StandardDeviation";

    #region Parameters
    public IValueParameter<DoubleValue> OriginalMeanParameter {
      get { return (IValueParameter<DoubleValue>)Parameters[OriginalMeanParameterName]; }
    }
    public IValueParameter<DoubleValue> OriginalStandardDeviationParameter {
      get { return (IValueParameter<DoubleValue>)Parameters[OriginalStandardDeviationParameterName]; }
    }
    public IValueParameter<DoubleValue> MeanParameter {
      get { return (IValueParameter<DoubleValue>)Parameters[MeanParameterName]; }
    }
    public IValueParameter<DoubleValue> StandardDeviationParameter {
      get { return (IValueParameter<DoubleValue>)Parameters[StandardDeviationParameterName]; }
    }
    #endregion

    #region properties
    public double OriginalMean {
      get { return OriginalMeanParameter.Value.Value; }
      set { OriginalMeanParameter.Value.Value = value; }
    }
    public double OriginalStandardDeviation {
      get { return OriginalStandardDeviationParameter.Value.Value; }
      set { OriginalStandardDeviationParameter.Value.Value = value; }
    }
    public double Mean {
      get { return MeanParameter.Value.Value; }
    }
    public double StandardDeviation {
      get { return StandardDeviationParameter.Value.Value; }
    }
    #endregion

    [StorableConstructor]
    protected ShiftStandardDistributionTransformation(bool deserializing) : base(deserializing) { }
    protected ShiftStandardDistributionTransformation(Transformation<double> original, Cloner cloner)
      : base(original, cloner) {
    }
    public ShiftStandardDistributionTransformation(IEnumerable<string> allowedColumns)
      : base(allowedColumns) {
      Parameters.Add(new ValueParameter<DoubleValue>(OriginalMeanParameterName, "m_org | Mean value of the original data's deviation.", new DoubleValue()));
      Parameters.Add(new ValueParameter<DoubleValue>(OriginalStandardDeviationParameterName, "s_org | Standard deviation of the original data.", new DoubleValue()));
      OriginalMeanParameter.Hidden = true;
      OriginalStandardDeviationParameter.Hidden = true;
      Parameters.Add(new ValueParameter<DoubleValue>(MeanParameterName, "m_tar | Mean value for the target deviation.", new DoubleValue(0.0)));
      Parameters.Add(new ValueParameter<DoubleValue>(StandardDeviationParameterName, "s_tar | Standard deviation for the target data.", new DoubleValue(1.0)));
    }

    public override IDeepCloneable Clone(Cloner cloner) {
      return new ShiftStandardDistributionTransformation(this, cloner);
    }

    // http://en.wikipedia.org/wiki/Standard_deviation
    // http://www.statistics4u.info/fundstat_germ/ee_ztransform.html
    // https://www.uni-due.de/~bm0061/vorl12.pdf p5
    public override IEnumerable<double> Apply(IEnumerable<double> data) {
      ConfigureParameters(data);
      if (OriginalStandardDeviation == 0.0) {
        foreach (var e in data) {
          yield return e;
        }
        yield break;
      }

      foreach (var e in data) {
        double unitNormalDistributedValue = (e - OriginalMean) / OriginalStandardDeviation;
        yield return unitNormalDistributedValue * StandardDeviation + Mean;
      }
    }

    public override bool Check(IEnumerable<double> data, out string errorMsg) {
      ConfigureParameters(data);
      errorMsg = "";
      if (OriginalStandardDeviation == 0.0) {
        errorMsg = "Standard deviaton for the original data is 0.0, Transformation cannot be applied onto these values.";
        return false;
      }
      return true;
    }

    protected void ConfigureParameters(IEnumerable<double> data) {
      OriginalStandardDeviation = data.StandardDeviation();
      OriginalMean = data.Average();
    }
  }
}