Free cookie consent management tool by TermsFeed Policy Generator

# Changeset 8238 for trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis

Ignore:
Timestamp:
07/05/12 16:55:26 (12 years ago)
Message:

Location:
trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3
Files:
14 edited
3 copied

Unmodified
Removed
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/HeuristicLab.Problems.Instances.DataAnalysis-3.3.csproj

 r8226
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionEight.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionNine : ArtificialRegressionDataDescriptor { public class KeijzerFunctionEight : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 9 f(x) = sqrt(x)"; } } public override string Name { get { return "Keijzer 8 f(x) = sqrt(x)"; } } public override string Description { get {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionEleven.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionTwelve : ArtificialRegressionDataDescriptor { public class KeijzerFunctionEleven : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 12 f(x, y) = xy + sin((x - 1)(y - 1))"; } } public override string Name { get { return "Keijzer 11 f(x, y) = xy + sin((x - 1)(y - 1))"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x, y) = xy + sin((x - 1)(y - 1))" + Environment.NewLine + "range(train): 20 Training cases x,y = rnd(-3, 3)" + Environment.NewLine + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the intercal [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x, y) = xy + sin((x - 1)(y - 1))" + Environment.NewLine + "range(train): 20 Training cases x,y = rnd(-3, 3)" + Environment.NewLine + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the interval [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; } } protected override int TrainingPartitionEnd { get { return 20; } } protected override int TestPartitionStart { get { return 2500; } } protected override int TestPartitionEnd { get { return 5000; } } protected override int TestPartitionEnd { get { return 2600; } } protected override List> GenerateValues() { List> data = new List>(); for (int i = 0; i < AllowedInputVariables.Count(); i++) { data.Add(ValueGenerator.GenerateUniformDistributedValues(5000, -3, 3).ToList()); data.Add(ValueGenerator.GenerateUniformDistributedValues(5020, -3, 3).ToList()); }
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionFifteen.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionSixteen : ArtificialRegressionDataDescriptor { public class KeijzerFunctionFifteen : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 16 f(x, y) = x^3 / 5 + y^3 / 2 - y - x"; } } public override string Name { get { return "Keijzer 15 f(x, y) = x³ / 5 + y³ / 2 - y - x"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x, y) = x^3 / 5 + y^3 / 2 - y - x" + Environment.NewLine + "Function: f(x, y) = x³ / 5 + y³ / 2 - y - x" + Environment.NewLine + "range(train): 20 Training cases x,y = rnd(-3, 3)" + Environment.NewLine + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the intercal [-3, 3] (not ca. 360000 as described) " + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the interval [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; } protected override int TrainingPartitionEnd { get { return 20; } } protected override int TestPartitionStart { get { return 2500; } } protected override int TestPartitionEnd { get { return 5000; } } protected override int TestPartitionEnd { get { return 2600; } } protected override List> GenerateValues() {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionFive.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionSix : ArtificialRegressionDataDescriptor { public class KeijzerFunctionFive : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 6 f(x) = (30 * x * z) / ((x - 10)  * y^2)"; } } public override string Name { get { return "Keijzer 5 f(x) = (30 * x * z) / ((x - 10)  * y²)"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = (30 * x * z) / ((x - 10)  * y^2)" + Environment.NewLine + "Function: f(x) = (30 * x * z) / ((x - 10)  * y²)" + Environment.NewLine + "range(train): 1000 points x,z = rnd(-1, 1), y = rnd(1, 2)" + Environment.NewLine + "range(test): 10000 points x,z = rnd(-1, 1), y = rnd(1, 2)" + Environment.NewLine
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionFour.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFive : ArtificialRegressionDataDescriptor { public class KeijzerFunctionFour : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 5 f(x) = x ^ 3  * exp(-x) * cos(x) * sin(x) * (sin(x) ^ 2 * cos(x) - 1)"; } } public override string Name { get { return "Keijzer 4 f(x) = x³  * exp(-x) * cos(x) * sin(x) * (sin(x)² * cos(x) - 1)"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = x ^ 3  * exp(-x) * cos(x) * sin(x) * (sin(x) ^ 2 * cos(x) - 1)" + Environment.NewLine + "Function: f(x) = x³  * exp(-x) * cos(x) * sin(x) * (sin(x)² * cos(x) - 1)" + Environment.NewLine + "range(train): x = [0:0.05:10]" + Environment.NewLine + "range(test): x = [0.05:0.05:10.05]" + Environment.NewLine
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionFourteen.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFifteen : ArtificialRegressionDataDescriptor { public class KeijzerFunctionFourteen : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 15 f(x, y) = 8 / (2 + x^2 + y^2)"; } } public override string Name { get { return "Keijzer 14 f(x, y) = 8 / (2 + x² + y²)"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x, y) = 8 / (2 + x^2 + y^2)" + Environment.NewLine + "Function: f(x, y) = 8 / (2 + x² + y²)" + Environment.NewLine + "range(train): 20 Train cases x,y = rnd(-3, 3)" + Environment.NewLine + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the intercal [-3, 3] (not ca. 360000 as described) " + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the interval [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; } protected override int TrainingPartitionEnd { get { return 20; } } protected override int TestPartitionStart { get { return 2500; } } protected override int TestPartitionEnd { get { return 5000; } } protected override int TestPartitionEnd { get { return 2600; } } protected override List> GenerateValues() {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionNine.cs

 r8226 public class KeijzerFunctionNine : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 9 f(x) = sqrt(x)"; } } public override string Name { get { return "Keijzer 9 f(x) = arcsinh(x)  i.e. ln(x + sqrt(x² + 1))"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = sqrt(x)" + Environment.NewLine + "Function: f(x) = arcsinh(x)  i.e. ln(x + sqrt(x² + 1))" + Environment.NewLine + "range(train): x = [0:1:100]" + Environment.NewLine + "range(test): x = [0:0.1:100]" + Environment.NewLine protected override string[] AllowedInputVariables { get { return new string[] { "X" }; } } protected override int TrainingPartitionStart { get { return 0; } } protected override int TrainingPartitionEnd { get { return 101; } } protected override int TestPartitionStart { get { return 101; } } protected override int TestPartitionEnd { get { return 1102; } } protected override int TrainingPartitionEnd { get { return 100; } } protected override int TestPartitionStart { get { return 100; } } protected override int TestPartitionEnd { get { return 1100; } } protected override List> GenerateValues() { for (int i = 0; i < data[0].Count; i++) { x = data[0][i]; results.Add(Math.Sqrt(x)); results.Add(Math.Log(x + Math.Sqrt(x*x + 1))); } data.Add(results);
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionOne.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFour : ArtificialRegressionDataDescriptor { public class KeijzerFunctionOne : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 4 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Name { get { return "Keijzer 1 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Description { get {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionSeven.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionEight : ArtificialRegressionDataDescriptor { public class KeijzerFunctionSeven : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 8 f(x) = log(x)"; } } public override string Name { get { return "Keijzer 7 f(x) = ln(x)"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = log(x)" + Environment.NewLine + "range(train): x = [0:1:100]" + Environment.NewLine + "range(test): x = [0:0.1:100]" + Environment.NewLine + "Function: f(x) = ln(x)" + Environment.NewLine + "range(train): x = [1:1:100]" + Environment.NewLine + "range(test): x = [1:0.1:100]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: The problem starts with 1 to avoid log(0), which is minus infinity!";
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionSix.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionSeven : ArtificialRegressionDataDescriptor { public class KeijzerFunctionSix : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 7 f(x) = Sum(1 / i) From 1 to X"; } } public override string Name { get { return "Keijzer 6 f(x) = Sum(1 / i) From 1 to X"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = (30 * x * y) / ((x - 10)  * y^2)" + Environment.NewLine + "Function: f(x) = Sum(1 / i) From 1 to X" + Environment.NewLine + "range(train): x = [1:1:50]" + Environment.NewLine + "range(test): x = [1:1:120]" + Environment.NewLine
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionTen.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionEleven : ArtificialRegressionDataDescriptor { public class KeijzerFunctionTen : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 11 f(x, y) = x ^ y"; } } public override string Name { get { return "Keijzer 10 f(x, y) = x ^ y"; } } public override string Description { get { protected override int TrainingPartitionEnd { get { return 100; } } protected override int TestPartitionStart { get { return 100; } } protected override int TestPartitionEnd { get { return 10301; } } protected override int TestPartitionEnd { get { return 10100; } } protected override List> GenerateValues() {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionThirteen.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFourteen : ArtificialRegressionDataDescriptor { public class KeijzerFunctionThirteen : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 14 f(x, y) = 6 * sin(x) * cos(y)"; } } public override string Name { get { return "Keijzer 13 f(x, y) = 6 * sin(x) * cos(y)"; } } public override string Description { get { + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the intercal [-3, 3] (not ca. 360000 as described) " + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the interval [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; } protected override int TrainingPartitionEnd { get { return 20; } } protected override int TestPartitionStart { get { return 2500; } } protected override int TestPartitionEnd { get { return 5000; } } protected override int TestPartitionEnd { get { return 2600; } } protected override List> GenerateValues() {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionThree.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFour : ArtificialRegressionDataDescriptor { public class KeijzerFunctionThree : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 4 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Name { get { return "Keijzer 3 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Description { get { + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = 0.3 * x *sin(2 * PI * x)" + Environment.NewLine + "range(train): x = [-1:0.1:1]" + Environment.NewLine + "range(test): x = [-1:0.001:1]" + Environment.NewLine + "range(train): x = [-3:0.1:3]" + Environment.NewLine + "range(test): x = [-3:0.001:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)"; } protected override string[] AllowedInputVariables { get { return new string[] { "X" }; } } protected override int TrainingPartitionStart { get { return 0; } } protected override int TrainingPartitionEnd { get { return 21; } } protected override int TestPartitionStart { get { return 21; } } protected override int TestPartitionEnd { get { return 2022; } } protected override int TrainingPartitionEnd { get { return 61; } } protected override int TestPartitionStart { get { return 61; } } protected override int TestPartitionEnd { get { return 6062; } } protected override List> GenerateValues() { List> data = new List>(); data.Add(ValueGenerator.GenerateSteps(-1, 1, 0.1).ToList()); data[0].AddRange(ValueGenerator.GenerateSteps(-1, 1, 0.001)); data.Add(ValueGenerator.GenerateSteps(-3, 3, 0.1).ToList()); data[0].AddRange(ValueGenerator.GenerateSteps(-3, 3, 0.001)); double x;
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionTwelve.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionThirteen : ArtificialRegressionDataDescriptor { public class KeijzerFunctionTwelve : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 13 f(x, y) = x^4 - x^3 + y^2 / 2 - y"; } } public override string Name { get { return "Keijzer 12 f(x, y) = x^4 - x³ + y² / 2 - y"; } } public override string Description { get { return "Paper: Improving Symbolic Regression with Interval Arithmetic and Linear Scaling" + Environment.NewLine + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x, y) = x^4 - x^3 + y^2 / 2 - y" + Environment.NewLine + "Function: f(x, y) = x^4 - x³ + y² / 2 - y" + Environment.NewLine + "range(train): 20 Training cases x,y = rnd(-3, 3)" + Environment.NewLine + "range(test): x,y = [-3:0.01:3]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)" + Environment.NewLine + Environment.NewLine + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the intercal [-3, 3] (not ca. 360000 as described) " + "Note: Test partition has been adjusted to only 100 random uniformly distributed test cases in the interval [-3, 3] (not ca. 360000 as described) " + ", but 5000 cases are created"; } protected override int TrainingPartitionEnd { get { return 20; } } protected override int TestPartitionStart { get { return 2500; } } protected override int TestPartitionEnd { get { return 5000; } } protected override int TestPartitionEnd { get { return 2600; } } protected override List> GenerateValues() {
• ## trunk/sources/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Regression/Keijzer/KeijzerFunctionTwo.cs

 r8226 namespace HeuristicLab.Problems.Instances.DataAnalysis { public class KeijzerFunctionFour : ArtificialRegressionDataDescriptor { public class KeijzerFunctionTwo : ArtificialRegressionDataDescriptor { public override string Name { get { return "Keijzer 4 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Name { get { return "Keijzer 2 f(x) = 0.3 * x *sin(2 * PI * x)"; } } public override string Description { get { + "Authors: Maarten Keijzer" + Environment.NewLine + "Function: f(x) = 0.3 * x *sin(2 * PI * x)" + Environment.NewLine + "range(train): x = [-1:0.1:1]" + Environment.NewLine + "range(test): x = [-1:0.001:1]" + Environment.NewLine + "range(train): x = [-2:0.1:2]" + Environment.NewLine + "range(test): x = [-2:0.001:2]" + Environment.NewLine + "Function Set: x + y, x * y, 1/x, -x, sqrt(x)"; } protected override string[] AllowedInputVariables { get { return new string[] { "X" }; } } protected override int TrainingPartitionStart { get { return 0; } } protected override int TrainingPartitionEnd { get { return 21; } } protected override int TestPartitionStart { get { return 21; } } protected override int TestPartitionEnd { get { return 2022; } } protected override int TrainingPartitionEnd { get { return 41; } } protected override int TestPartitionStart { get { return 41; } } protected override int TestPartitionEnd { get { return 4042; } } protected override List> GenerateValues() { List> data = new List>(); data.Add(ValueGenerator.GenerateSteps(-1, 1, 0.1).ToList()); data[0].AddRange(ValueGenerator.GenerateSteps(-1, 1, 0.001)); data.Add(ValueGenerator.GenerateSteps(-2, 2, 0.1).ToList()); data[0].AddRange(ValueGenerator.GenerateSteps(-2, 2, 0.001)); double x;