source: branches/2457_ExpertSystem/WalkExporter/Standardizer.cs @ 18235

using System.Collections.Generic;
using System.Linq;
using HeuristicLab.Common;

namespace WalkExporter {
  public class InstancesStandardizer {
    private IDictionary<string, (double Avg, double Stdev)> normalization;

    public IEnumerable<double> GetMeans() {
      return normalization.Select(x => x.Value.Item1);
    }
    public IEnumerable<double> GetStdevs() {
      return normalization.Select(x => x.Value.Item2);
    }

    private InstancesStandardizer(IList<ProblemInstanceDescriptor> instances, IEnumerable<string> features) {
      normalization = instances
        .SelectMany(x => x.Features)
        .GroupBy(x => x.Key)
        .ToDictionary(x => x.Key,
                      x => {
                        var values = x.Select(y => y.GetNumericValue());
                        return (Avg: values.Average(), Stdev: values.StandardDeviation());
                      });
    }

    public static InstancesStandardizer Create(IList<ProblemInstanceDescriptor> instances, IEnumerable<string> features) {
      return new InstancesStandardizer(instances, features);
      
    }
    public static InstancesStandardizer CreateAndApply(IList<ProblemInstanceDescriptor> instances, IEnumerable<string> features) {
      var standardizer = Create(instances, features);
      standardizer.Apply(instances);
      return standardizer;
    }

    public void Apply(IList<ProblemInstanceDescriptor> instances) {
      for (var i = 0; i < instances.Count; i++) {
        var inst = instances[i];
        for (var x = 0; x < inst.Features.Count; x++) {
          var feat = inst.Features[x];
          feat.SetNumericValue((inst.Features[x].GetNumericValue() - normalization[feat.Key].Avg) / normalization[feat.Key].Stdev);
        }
      }
    }
  }
}