Added new normalization options (#3)

* Updated language version to use latest Updated target frameworks to widen the range Added a bunch of new metrics such as R2, Std Deviation, Std Error, etc * Added normalization code structure and examples Added log normalization and decimal normalization Updated example code Cleaned up exceptions to include more info Added some code documentation * Added a ZScore Normalization option Added a MinMax Normalization option --------- Signed-off-by: Franklin Moormann <cheatcountry@gmail.com>

Author: ooples

src/AiDotNet.csproj

@@ -5,7 +5,7 @@
 <ImplicitUsings>enable</ImplicitUsings>
 <Nullable>enable</Nullable>
 <GeneratePackageOnBuild>True</GeneratePackageOnBuild>
-<Version>0.0.3-preview</Version>
+<Version>0.0.4-preview</Version>
 <Title>Ai for .Net</Title>
 <Description>This is a preview library that will eventually showcase the latest and greatest in ai breakthroughs and bring them to the .net community</Description>
 <Company>Ooples Finance</Company>
@@ -14,7 +14,7 @@
 <PackageProjectUrl>https://github.com/ooples/AiDotNet</PackageProjectUrl>
 <RepositoryType>git</RepositoryType>
 <RepositoryUrl>https://github.com/ooples/AiDotNet</RepositoryUrl>
-<PackageTags>ai; regression; machine learning; artificial; intelligence; machine; chatgpt; learning; algorithm; algo; chatgpt-4</PackageTags>
+<PackageTags>ai; ai.net; aidotnet; regression; machine learning; artificial; intelligence; machine; chatgpt; learning; algorithm; algo; chatgpt-4</PackageTags>
 <PackageLicenseExpression>Apache-2.0</PackageLicenseExpression>
 <PackageRequireLicenseAcceptance>True</PackageRequireLicenseAcceptance>
  <LangVersion>latest</LangVersion>

src/Metrics/Metrics.cs

@@ -1,4 +1,4 @@
-namespace AiDotNet;
+namespace AiDotNet.Metrics;
 
 public sealed class Metrics : IMetrics
 {
@@ -53,7 +53,7 @@ internal override double CalculateRootMeanSquaredError()
  internal override double CalculateR2()
  {
  double residualSumSquares = 0, totalSumSquares = 0;
- for (int i = 0; i < SampleSize; i++)
+ for (var i = 0; i < SampleSize; i++)
  {
  residualSumSquares += Math.Pow(OosActualValues[i] - OosPredictions[i], 2);
  totalSumSquares += Math.Pow(OosActualValues[i] - OosActualValuesAvg, 2);

src/Normalization/MinMaxNormalization.cs

@@ -0,0 +1,34 @@
+namespace AiDotNet.Normalization;
+
+internal class MinMaxNormalization : INormalization
+{
+ private double InputsMin { get; set; }
+ private double InputsMax { get; set; }
+
+ internal override double[] Normalize(double[] rawValues)
+ {
+ var normalizedValues = new double[rawValues.Length];
+ for (var i = 0; i < rawValues.Length; i++)
+ {
+ normalizedValues.SetValue(Math.Abs(InputsMax - InputsMin) > 0 ? (rawValues[i] - InputsMin) / (InputsMax - InputsMin) : double.NaN, i);
+ }
+
+ if (normalizedValues.Contains(double.NaN))
+ {
+ throw new ArgumentException("Normalized values can't contain NaN values. " +
+ "MinMax Normalization creates NaN values when the training data has all of the same values or invalid data.", nameof(rawValues));
+ }
+
+ return normalizedValues;
+ }
+
+ internal override (double[] trainingInputs, double[] trainingOutputs, double[] oosInputs, double[] oosOutputs) PrepareData(
+ double[] inputs, double[] outputs, int trainingSize)
+ {
+ var (trainingInputs, trainingOutputs, oosInputs, oosOutputs) = NormalizationHelper.SplitData(inputs, outputs, trainingSize);
+ InputsMin = trainingInputs.Min();
+ InputsMax = trainingInputs.Max();
+
+ return (Normalize(trainingInputs), Normalize(trainingOutputs), Normalize(oosInputs), Normalize(oosOutputs));
+ }
+}

src/Normalization/ZScoreNormalization.cs

@@ -0,0 +1,32 @@
+namespace AiDotNet.Normalization;
+
+internal class ZScoreNormalization : INormalization
+{
+ internal override double[] Normalize(double[] rawValues)
+ {
+ var normalizedValues = new double[rawValues.Length];
+
+ var rawValuesAvg = rawValues.Average();
+ var rawValuesStdDev = Math.Sqrt(rawValues.Select(x => Math.Pow(x - rawValuesAvg, 2)).Sum() / rawValues.Length);
+
+ for (var i = 0; i < rawValues.Length; i++)
+ {
+ normalizedValues.SetValue(rawValuesStdDev != 0 ? (rawValues[i] - rawValuesAvg) / rawValuesStdDev : double.NaN, i);
+ }
+
+ if (normalizedValues.Contains(double.NaN))
+ {
+ throw new ArgumentException("Normalized values can't contain NaN values. " +
+ "Z-Score Normalization creates NaN values when the training data has all of the same values or invalid data.", nameof(rawValues));
+ }
+
+ return normalizedValues;
+ }
+
+ internal override (double[] trainingInputs, double[] trainingOutputs, double[] oosInputs, double[] oosOutputs) PrepareData(double[] inputs, double[] outputs, int trainingSize)
+ {
+ var (trainingInputs, trainingOutputs, oosInputs, oosOutputs) = NormalizationHelper.SplitData(inputs, outputs, trainingSize);
+
+ return (Normalize(trainingInputs), Normalize(trainingOutputs), Normalize(oosInputs), Normalize(oosOutputs));
+ }
+}

src/Regression/SimpleRegression.cs

@@ -1,4 +1,4 @@
-namespace AiDotNet.Regression;
+namespace AiDotNet.Regression;
 
 public sealed class SimpleRegression : IRegression
 {
@@ -63,7 +63,7 @@ public SimpleRegression(double[] inputs, double[] outputs, double trainingPctSiz
  var (trainingInputs, trainingOutputs, oosInputs, oosOutputs) = PrepareData(inputs, outputs, trainingSize, normalization);
  Fit(trainingInputs, trainingOutputs);
  Predictions = Transform(oosInputs);
- Metrics = new Metrics(Predictions, oosOutputs, inputs.Rank);
+ Metrics = new Metrics.Metrics(Predictions, oosOutputs, inputs.Rank);
  }
 
  internal override void Fit(double[] x, double[] y)