Changeset 9363 for branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators

Timestamp:

04/16/13 13:13:41 (11 years ago)

Author:

spimming

Message:

#1888:

• Merged revisions from trunk

Location:

branches/OaaS

Files:

• . (modified) (2 props)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4 (modified) (1 prop)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionConstantOptimizationEvaluator.cs (modified) (8 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveEvaluator.cs (modified) (1 diff)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMaxAbsoluteErrorEvaluator.cs (modified) (4 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMeanAbsoluteErrorEvaluator.cs (modified) (4 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator.cs (modified) (4 diffs)
• HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.cs (modified) (3 diffs)

branches/OaaS

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll
 packages

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll
 packages

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4

@@ +1 @@
+*.user
+Plugin.cs
 bin
-*.user
-HeuristicLabProblemsDataAnalysisSymbolicRegressionPlugin.cs
 obj
-*.vs10x
-Plugin.cs

@@ +1 @@
+*.user
+Plugin.cs
 bin
-*.user
-HeuristicLabProblemsDataAnalysisSymbolicRegressionPlugin.cs
 obj
-*.vs10x
-Plugin.cs

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionConstantOptimizationEvaluator.cs

@@ -20 +20 @@
 #endregion
 
+using System;
 using System.Collections.Generic;
 using System.Linq;
+using AutoDiff;
 using HeuristicLab.Common;
 using HeuristicLab.Core;
@@ -37 +39 @@
     private const string ConstantOptimizationProbabilityParameterName = "ConstantOptimizationProbability";
     private const string ConstantOptimizationRowsPercentageParameterName = "ConstantOptimizationRowsPercentage";
-
-    private const string EvaluatedTreesResultName = "EvaluatedTrees";
-    private const string EvaluatedTreeNodesResultName = "EvaluatedTreeNodes";
-
-    public ILookupParameter<IntValue> EvaluatedTreesParameter {
-      get { return (ILookupParameter<IntValue>)Parameters[EvaluatedTreesResultName]; }
-    }
-    public ILookupParameter<IntValue> EvaluatedTreeNodesParameter {
-      get { return (ILookupParameter<IntValue>)Parameters[EvaluatedTreeNodesResultName]; }
-    }
+    private const string UpdateConstantsInTreeParameterName = "UpdateConstantsInSymbolicExpressionTree";
 
     public IFixedValueParameter<IntValue> ConstantOptimizationIterationsParameter {
@@ -60 +53 @@
       get { return (IFixedValueParameter<PercentValue>)Parameters[ConstantOptimizationRowsPercentageParameterName]; }
     }
+    public IFixedValueParameter<BoolValue> UpdateConstantsInTreeParameter {
+      get { return (IFixedValueParameter<BoolValue>)Parameters[UpdateConstantsInTreeParameterName]; }
+    }
 
     public IntValue ConstantOptimizationIterations {
@@ -72 +68 @@
     public PercentValue ConstantOptimizationRowsPercentage {
       get { return ConstantOptimizationRowsPercentageParameter.Value; }
+    }
+    public bool UpdateConstantsInTree {
+      get { return UpdateConstantsInTreeParameter.Value.Value; }
+      set { UpdateConstantsInTreeParameter.Value.Value = value; }
     }
 
@@ -85 +85 @@
     public SymbolicRegressionConstantOptimizationEvaluator()
       : base() {
-      Parameters.Add(new FixedValueParameter<IntValue>(ConstantOptimizationIterationsParameterName, "Determines how many iterations should be calculated while optimizing the constant of a symbolic expression tree (0 indicates other or default stopping criterion).", new IntValue(3), true));
+      Parameters.Add(new FixedValueParameter<IntValue>(ConstantOptimizationIterationsParameterName, "Determines how many iterations should be calculated while optimizing the constant of a symbolic expression tree (0 indicates other or default stopping criterion).", new IntValue(10), true));
       Parameters.Add(new FixedValueParameter<DoubleValue>(ConstantOptimizationImprovementParameterName, "Determines the relative improvement which must be achieved in the constant optimization to continue with it (0 indicates other or default stopping criterion).", new DoubleValue(0), true));
       Parameters.Add(new FixedValueParameter<PercentValue>(ConstantOptimizationProbabilityParameterName, "Determines the probability that the constants are optimized", new PercentValue(1), true));
       Parameters.Add(new FixedValueParameter<PercentValue>(ConstantOptimizationRowsPercentageParameterName, "Determines the percentage of the rows which should be used for constant optimization", new PercentValue(1), true));
-
-      Parameters.Add(new LookupParameter<IntValue>(EvaluatedTreesResultName));
-      Parameters.Add(new LookupParameter<IntValue>(EvaluatedTreeNodesResultName));
+      Parameters.Add(new FixedValueParameter<BoolValue>(UpdateConstantsInTreeParameterName, "Determines if the constants in the tree should be overwritten by the optimized constants.", new BoolValue(true)));
     }
 
@@ -98 +96 @@
     }
 
+    [StorableHook(HookType.AfterDeserialization)]
+    private void AfterDeserialization() {
+      if (!Parameters.ContainsKey(UpdateConstantsInTreeParameterName))
+        Parameters.Add(new FixedValueParameter<BoolValue>(UpdateConstantsInTreeParameterName, "Determines if the constants in the tree should be overwritten by the optimized constants.", new BoolValue(true)));
+    }
+
     public override IOperation Apply() {
-      AddResults();
-      int seed = RandomParameter.ActualValue.Next();
       var solution = SymbolicExpressionTreeParameter.ActualValue;
       double quality;
@@ -106 +108 @@
         IEnumerable<int> constantOptimizationRows = GenerateRowsToEvaluate(ConstantOptimizationRowsPercentage.Value);
         quality = OptimizeConstants(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, ProblemDataParameter.ActualValue,
-           constantOptimizationRows, ConstantOptimizationImprovement.Value, ConstantOptimizationIterations.Value, 0.001,
-           EstimationLimitsParameter.ActualValue.Upper, EstimationLimitsParameter.ActualValue.Lower,
-          EvaluatedTreesParameter.ActualValue, EvaluatedTreeNodesParameter.ActualValue);
+           constantOptimizationRows, ApplyLinearScalingParameter.ActualValue.Value, ConstantOptimizationIterations.Value,
+           EstimationLimitsParameter.ActualValue.Upper, EstimationLimitsParameter.ActualValue.Lower, UpdateConstantsInTree);
+
         if (ConstantOptimizationRowsPercentage.Value != RelativeNumberOfEvaluatedSamplesParameter.ActualValue.Value) {
           var evaluationRows = GenerateRowsToEvaluate();
-          quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows);
+          quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows, ApplyLinearScalingParameter.ActualValue.Value);
         }
       } else {
         var evaluationRows = GenerateRowsToEvaluate();
-        quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows);
+        quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, evaluationRows, ApplyLinearScalingParameter.ActualValue.Value);
       }
       QualityParameter.ActualValue = new DoubleValue(quality);
-      EvaluatedTreesParameter.ActualValue.Value += 1;
-      EvaluatedTreeNodesParameter.ActualValue.Value += solution.Length;
 
       if (Successor != null)
@@ -127 +127 @@
     }
 
-    private void AddResults() {
-      if (EvaluatedTreesParameter.ActualValue == null) {
-        var scope = ExecutionContext.Scope;
-        while (scope.Parent != null)
-          scope = scope.Parent;
-        scope.Variables.Add(new Core.Variable(EvaluatedTreesResultName, new IntValue()));
-      }
-      if (EvaluatedTreeNodesParameter.ActualValue == null) {
-        var scope = ExecutionContext.Scope;
-        while (scope.Parent != null)
-          scope = scope.Parent;
-        scope.Variables.Add(new Core.Variable(EvaluatedTreeNodesResultName, new IntValue()));
-      }
-    }
-
     public override double Evaluate(IExecutionContext context, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows) {
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
       EstimationLimitsParameter.ExecutionContext = context;
-
-      double r2 = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows);
+      ApplyLinearScalingParameter.ExecutionContext = context;
+
+      // Pearson R² evaluator is used on purpose instead of the const-opt evaluator, 
+      // because Evaluate() is used to get the quality of evolved models on 
+      // different partitions of the dataset (e.g., best validation model)
+      double r2 = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
       EstimationLimitsParameter.ExecutionContext = null;
+      ApplyLinearScalingParameter.ExecutionContext = null;
 
       return r2;
     }
 
+    #region derivations of functions
+    // create function factory for arctangent
+    private readonly Func<Term, UnaryFunc> arctan = UnaryFunc.Factory(
+      eval: Math.Atan,
+      diff: x => 1 / (1 + x * x));
+    private static readonly Func<Term, UnaryFunc> sin = UnaryFunc.Factory(
+      eval: Math.Sin,
+      diff: Math.Cos);
+    private static readonly Func<Term, UnaryFunc> cos = UnaryFunc.Factory(
+       eval: Math.Cos,
+       diff: x => -Math.Sin(x));
+    private static readonly Func<Term, UnaryFunc> tan = UnaryFunc.Factory(
+      eval: Math.Tan,
+      diff: x => 1 + Math.Tan(x) * Math.Tan(x));
+    private static readonly Func<Term, UnaryFunc> square = UnaryFunc.Factory(
+       eval: x => x * x,
+       diff: x => 2 * x);
+    private static readonly Func<Term, UnaryFunc> erf = UnaryFunc.Factory(
+      eval: alglib.errorfunction,
+      diff: x => 2.0 * Math.Exp(-(x * x)) / Math.Sqrt(Math.PI));
+    private static readonly Func<Term, UnaryFunc> norm = UnaryFunc.Factory(
+      eval: alglib.normaldistribution,
+      diff: x => -(Math.Exp(-(x * x)) * Math.Sqrt(Math.Exp(x * x)) * x) / Math.Sqrt(2 * Math.PI));
+    #endregion
+
+
     public static double OptimizeConstants(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree tree, IRegressionProblemData problemData,
-      IEnumerable<int> rows, double improvement, int iterations, double differentialStep, double upperEstimationLimit = double.MaxValue, double lowerEstimationLimit = double.MinValue, IntValue evaluatedTrees = null, IntValue evaluatedTreeNodes = null) {
+      IEnumerable<int> rows, bool applyLinearScaling, int maxIterations, double upperEstimationLimit = double.MaxValue, double lowerEstimationLimit = double.MinValue, bool updateConstantsInTree = true) {
+
+      List<AutoDiff.Variable> variables = new List<AutoDiff.Variable>();
+      List<AutoDiff.Variable> parameters = new List<AutoDiff.Variable>();
+      List<string> variableNames = new List<string>();
+
+      AutoDiff.Term func;
+      if (!TryTransformToAutoDiff(tree.Root.GetSubtree(0), variables, parameters, variableNames, out func))
+        throw new NotSupportedException("Could not optimize constants of symbolic expression tree due to not supported symbols used in the tree.");
+      if (variableNames.Count == 0) return 0.0;
+
+      AutoDiff.IParametricCompiledTerm compiledFunc = AutoDiff.TermUtils.Compile(func, variables.ToArray(), parameters.ToArray());
+
       List<SymbolicExpressionTreeTerminalNode> terminalNodes = tree.Root.IterateNodesPrefix().OfType<SymbolicExpressionTreeTerminalNode>().ToList();
-      double[] c = new double[terminalNodes.Count];
-      int treeLength = tree.Length;
-
-      //extract inital constants
-      for (int i = 0; i < terminalNodes.Count; i++) {
-        ConstantTreeNode constantTreeNode = terminalNodes[i] as ConstantTreeNode;
-        if (constantTreeNode != null) c[i] = constantTreeNode.Value;
-        VariableTreeNode variableTreeNode = terminalNodes[i] as VariableTreeNode;
-        if (variableTreeNode != null) c[i] = variableTreeNode.Weight;
-      }
-
-      double epsg = 0;
-      double epsf = improvement;
-      double epsx = 0;
-      int maxits = iterations;
-      double diffstep = differentialStep;
-
-      alglib.minlmstate state;
-      alglib.minlmreport report;
-
-      alglib.minlmcreatev(1, c, diffstep, out state);
-      alglib.minlmsetcond(state, epsg, epsf, epsx, maxits);
-      alglib.minlmoptimize(state, CreateCallBack(interpreter, tree, problemData, rows, upperEstimationLimit, lowerEstimationLimit, treeLength, evaluatedTrees, evaluatedTreeNodes), null, terminalNodes);
-      alglib.minlmresults(state, out c, out report);
-
-      for (int i = 0; i < c.Length; i++) {
-        ConstantTreeNode constantTreeNode = terminalNodes[i] as ConstantTreeNode;
-        if (constantTreeNode != null) constantTreeNode.Value = c[i];
-        VariableTreeNode variableTreeNode = terminalNodes[i] as VariableTreeNode;
-        if (variableTreeNode != null) variableTreeNode.Weight = c[i];
-      }
-
-      return (state.fi[0] - 1) * -1;
-    }
-
-    private static alglib.ndimensional_fvec CreateCallBack(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree tree, IRegressionProblemData problemData, IEnumerable<int> rows, double upperEstimationLimit, double lowerEstimationLimit, int treeLength, IntValue evaluatedTrees = null, IntValue evaluatedTreeNodes = null) {
-      return (double[] arg, double[] fi, object obj) => {
-        // update constants of tree
-        List<SymbolicExpressionTreeTerminalNode> terminalNodes = (List<SymbolicExpressionTreeTerminalNode>)obj;
-        for (int i = 0; i < terminalNodes.Count; i++) {
-          ConstantTreeNode constantTreeNode = terminalNodes[i] as ConstantTreeNode;
-          if (constantTreeNode != null) constantTreeNode.Value = arg[i];
-          VariableTreeNode variableTreeNode = terminalNodes[i] as VariableTreeNode;
-          if (variableTreeNode != null) variableTreeNode.Weight = arg[i];
-        }
-
-        double quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(interpreter, tree, lowerEstimationLimit, upperEstimationLimit, problemData, rows);
-
-        fi[0] = 1 - quality;
-        if (evaluatedTrees != null) evaluatedTrees.Value++;
-        if (evaluatedTreeNodes != null) evaluatedTreeNodes.Value += treeLength;
+      double[] c = new double[variables.Count];
+
+      {
+        c[0] = 0.0;
+        c[1] = 1.0;
+        //extract inital constants
+        int i = 2;
+        foreach (var node in terminalNodes) {
+          ConstantTreeNode constantTreeNode = node as ConstantTreeNode;
+          VariableTreeNode variableTreeNode = node as VariableTreeNode;
+          if (constantTreeNode != null)
+            c[i++] = constantTreeNode.Value;
+          else if (variableTreeNode != null)
+            c[i++] = variableTreeNode.Weight;
+        }
+      }
+      double[] originalConstants = (double[])c.Clone();
+      double originalQuality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(interpreter, tree, lowerEstimationLimit, upperEstimationLimit, problemData, rows, applyLinearScaling);
+
+      alglib.lsfitstate state;
+      alglib.lsfitreport rep;
+      int info;
+
+      Dataset ds = problemData.Dataset;
+      double[,] x = new double[rows.Count(), variableNames.Count];
+      int row = 0;
+      foreach (var r in rows) {
+        for (int col = 0; col < variableNames.Count; col++) {
+          x[row, col] = ds.GetDoubleValue(variableNames[col], r);
+        }
+        row++;
+      }
+      double[] y = ds.GetDoubleValues(problemData.TargetVariable, rows).ToArray();
+      int n = x.GetLength(0);
+      int m = x.GetLength(1);
+      int k = c.Length;
+
+      alglib.ndimensional_pfunc function_cx_1_func = CreatePFunc(compiledFunc);
+      alglib.ndimensional_pgrad function_cx_1_grad = CreatePGrad(compiledFunc);
+
+      try {
+        alglib.lsfitcreatefg(x, y, c, n, m, k, false, out state);
+        alglib.lsfitsetcond(state, 0.0, 0.0, maxIterations);
+        //alglib.lsfitsetgradientcheck(state, 0.001);
+        alglib.lsfitfit(state, function_cx_1_func, function_cx_1_grad, null, null);
+        alglib.lsfitresults(state, out info, out c, out rep);
+      }
+      catch (ArithmeticException) {
+        return originalQuality;
+      }
+      catch (alglib.alglibexception) {
+        return originalQuality;
+      }
+
+      //info == -7  => constant optimization failed due to wrong gradient
+      if (info != -7) UpdateConstants(tree, c.Skip(2).ToArray());
+      var quality = SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.Calculate(interpreter, tree, lowerEstimationLimit, upperEstimationLimit, problemData, rows, applyLinearScaling);
+
+      if (!updateConstantsInTree) UpdateConstants(tree, originalConstants.Skip(2).ToArray());
+      if (originalQuality - quality > 0.001 || double.IsNaN(quality)) {
+        UpdateConstants(tree, originalConstants.Skip(2).ToArray());
+        return originalQuality;
+      }
+      return quality;
+    }
+
+    private static void UpdateConstants(ISymbolicExpressionTree tree, double[] constants) {
+      int i = 0;
+      foreach (var node in tree.Root.IterateNodesPrefix().OfType<SymbolicExpressionTreeTerminalNode>()) {
+        ConstantTreeNode constantTreeNode = node as ConstantTreeNode;
+        VariableTreeNode variableTreeNode = node as VariableTreeNode;
+        if (constantTreeNode != null)
+          constantTreeNode.Value = constants[i++];
+        else if (variableTreeNode != null)
+          variableTreeNode.Weight = constants[i++];
+      }
+    }
+
+    private static alglib.ndimensional_pfunc CreatePFunc(AutoDiff.IParametricCompiledTerm compiledFunc) {
+      return (double[] c, double[] x, ref double func, object o) => {
+        func = compiledFunc.Evaluate(c, x);
       };
     }
 
+    private static alglib.ndimensional_pgrad CreatePGrad(AutoDiff.IParametricCompiledTerm compiledFunc) {
+      return (double[] c, double[] x, ref double func, double[] grad, object o) => {
+        var tupel = compiledFunc.Differentiate(c, x);
+        func = tupel.Item2;
+        Array.Copy(tupel.Item1, grad, grad.Length);
+      };
+    }
+
+    private static bool TryTransformToAutoDiff(ISymbolicExpressionTreeNode node, List<AutoDiff.Variable> variables, List<AutoDiff.Variable> parameters, List<string> variableNames, out AutoDiff.Term term) {
+      if (node.Symbol is Constant) {
+        var var = new AutoDiff.Variable();
+        variables.Add(var);
+        term = var;
+        return true;
+      }
+      if (node.Symbol is Variable) {
+        var varNode = node as VariableTreeNode;
+        var par = new AutoDiff.Variable();
+        parameters.Add(par);
+        variableNames.Add(varNode.VariableName);
+        var w = new AutoDiff.Variable();
+        variables.Add(w);
+        term = AutoDiff.TermBuilder.Product(w, par);
+        return true;
+      }
+      if (node.Symbol is Addition) {
+        List<AutoDiff.Term> terms = new List<Term>();
+        foreach (var subTree in node.Subtrees) {
+          AutoDiff.Term t;
+          if (!TryTransformToAutoDiff(subTree, variables, parameters, variableNames, out t)) {
+            term = null;
+            return false;
+          }
+          terms.Add(t);
+        }
+        term = AutoDiff.TermBuilder.Sum(terms);
+        return true;
+      }
+      if (node.Symbol is Subtraction) {
+        List<AutoDiff.Term> terms = new List<Term>();
+        for (int i = 0; i < node.SubtreeCount; i++) {
+          AutoDiff.Term t;
+          if (!TryTransformToAutoDiff(node.GetSubtree(i), variables, parameters, variableNames, out t)) {
+            term = null;
+            return false;
+          }
+          if (i > 0) t = -t;
+          terms.Add(t);
+        }
+        term = AutoDiff.TermBuilder.Sum(terms);
+        return true;
+      }
+      if (node.Symbol is Multiplication) {
+        AutoDiff.Term a, b;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out a) ||
+          !TryTransformToAutoDiff(node.GetSubtree(1), variables, parameters, variableNames, out b)) {
+          term = null;
+          return false;
+        } else {
+          List<AutoDiff.Term> factors = new List<Term>();
+          foreach (var subTree in node.Subtrees.Skip(2)) {
+            AutoDiff.Term f;
+            if (!TryTransformToAutoDiff(subTree, variables, parameters, variableNames, out f)) {
+              term = null;
+              return false;
+            }
+            factors.Add(f);
+          }
+          term = AutoDiff.TermBuilder.Product(a, b, factors.ToArray());
+          return true;
+        }
+      }
+      if (node.Symbol is Division) {
+        // only works for at least two subtrees
+        AutoDiff.Term a, b;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out a) ||
+          !TryTransformToAutoDiff(node.GetSubtree(1), variables, parameters, variableNames, out b)) {
+          term = null;
+          return false;
+        } else {
+          List<AutoDiff.Term> factors = new List<Term>();
+          foreach (var subTree in node.Subtrees.Skip(2)) {
+            AutoDiff.Term f;
+            if (!TryTransformToAutoDiff(subTree, variables, parameters, variableNames, out f)) {
+              term = null;
+              return false;
+            }
+            factors.Add(1.0 / f);
+          }
+          term = AutoDiff.TermBuilder.Product(a, 1.0 / b, factors.ToArray());
+          return true;
+        }
+      }
+      if (node.Symbol is Logarithm) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = AutoDiff.TermBuilder.Log(t);
+          return true;
+        }
+      }
+      if (node.Symbol is Exponential) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = AutoDiff.TermBuilder.Exp(t);
+          return true;
+        }
+      } if (node.Symbol is Sine) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = sin(t);
+          return true;
+        }
+      } if (node.Symbol is Cosine) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = cos(t);
+          return true;
+        }
+      } if (node.Symbol is Tangent) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = tan(t);
+          return true;
+        }
+      }
+      if (node.Symbol is Square) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = square(t);
+          return true;
+        }
+      } if (node.Symbol is Erf) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = erf(t);
+          return true;
+        }
+      } if (node.Symbol is Norm) {
+        AutoDiff.Term t;
+        if (!TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out t)) {
+          term = null;
+          return false;
+        } else {
+          term = norm(t);
+          return true;
+        }
+      }
+      if (node.Symbol is StartSymbol) {
+        var alpha = new AutoDiff.Variable();
+        var beta = new AutoDiff.Variable();
+        variables.Add(beta);
+        variables.Add(alpha);
+        AutoDiff.Term branchTerm;
+        if (TryTransformToAutoDiff(node.GetSubtree(0), variables, parameters, variableNames, out branchTerm)) {
+          term = branchTerm * alpha + beta;
+          return true;
+        } else {
+          term = null;
+          return false;
+        }
+      }
+      term = null;
+      return false;
+    }
+
+    public static bool CanOptimizeConstants(ISymbolicExpressionTree tree) {
+      var containsUnknownSymbol = (
+        from n in tree.Root.GetSubtree(0).IterateNodesPrefix()
+        where
+         !(n.Symbol is Variable) &&
+         !(n.Symbol is Constant) &&
+         !(n.Symbol is Addition) &&
+         !(n.Symbol is Subtraction) &&
+         !(n.Symbol is Multiplication) &&
+         !(n.Symbol is Division) &&
+         !(n.Symbol is Logarithm) &&
+         !(n.Symbol is Exponential) &&
+         !(n.Symbol is Sine) &&
+         !(n.Symbol is Cosine) &&
+         !(n.Symbol is Tangent) &&
+         !(n.Symbol is Square) &&
+         !(n.Symbol is Erf) &&
+         !(n.Symbol is Norm) &&
+         !(n.Symbol is StartSymbol)
+        select n).
+      Any();
+      return !containsUnknownSymbol;
+    }
   }
 }

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveEvaluator.cs

@@ -30 +30 @@
 namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
   [StorableClass]
-  public abstract class SymbolicRegressionSingleObjectiveEvaluator : SymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>, ISymbolicRegressionSingleObjectiveEvaluator {
-    private const string ApplyLinearScalingParameterName = "ApplyLinearScaling";
-    public IFixedValueParameter<BoolValue> ApplyLinearScalingParameter {
-      get { return (IFixedValueParameter<BoolValue>)Parameters[ApplyLinearScalingParameterName]; }
-    }
-    public bool ApplyLinearScaling {
-      get { return ApplyLinearScalingParameter.Value.Value; }
-      set { ApplyLinearScalingParameter.Value.Value = value; }
-    }
-
+  public abstract class SymbolicRegressionSingleObjectiveEvaluator : SymbolicDataAnalysisSingleObjectiveEvaluator<IRegressionProblemData>, ISymbolicRegressionSingleObjectiveEvaluator {  
     [StorableConstructor]
     protected SymbolicRegressionSingleObjectiveEvaluator(bool deserializing) : base(deserializing) { }
     protected SymbolicRegressionSingleObjectiveEvaluator(SymbolicRegressionSingleObjectiveEvaluator original, Cloner cloner) : base(original, cloner) { }
-    protected SymbolicRegressionSingleObjectiveEvaluator()
-      : base() {
-      Parameters.Add(new FixedValueParameter<BoolValue>(ApplyLinearScalingParameterName, "Flag that indicates if the individual should be linearly scaled before evaluating.", new BoolValue(true)));
-      ApplyLinearScalingParameter.Hidden = true;
-    }
-
-    [StorableHook(HookType.AfterDeserialization)]
-    private void AfterDeserialization() {
-      if (!Parameters.ContainsKey(ApplyLinearScalingParameterName)) {
-        Parameters.Add(new FixedValueParameter<BoolValue>(ApplyLinearScalingParameterName, "Flag that indicates if the individual should be linearly scaled before evaluating.", new BoolValue(false)));
-        ApplyLinearScalingParameter.Hidden = true;
-      }
-    }
-
-    [ThreadStatic]
-    private static double[] cache;
-
-    protected static void CalculateWithScaling(IEnumerable<double> targetValues, IEnumerable<double> estimatedValues,
-      double lowerEstimationLimit, double upperEstimationLimit,
-      IOnlineCalculator calculator, int maxRows) {
-      if (cache == null || cache.GetLength(0) < maxRows) {
-        cache = new double[maxRows];
-      }
-
-      //calculate linear scaling
-      //the static methods of the calculator could not be used as it performs a check if the enumerators have an equal amount of elements
-      //this is not true if the cache is used
-      int i = 0;
-      var linearScalingCalculator = new OnlineLinearScalingParameterCalculator();
-      var targetValuesEnumerator = targetValues.GetEnumerator();
-      var estimatedValuesEnumerator = estimatedValues.GetEnumerator();
-      while (targetValuesEnumerator.MoveNext() && estimatedValuesEnumerator.MoveNext()) {
-        double target = targetValuesEnumerator.Current;
-        double estimated = estimatedValuesEnumerator.Current;
-        cache[i] = estimated;
-        linearScalingCalculator.Add(estimated, target);
-        i++;
-      }
-      double alpha = linearScalingCalculator.Alpha;
-      double beta = linearScalingCalculator.Beta;
-
-      //calculate the quality by using the passed online calculator
-      targetValuesEnumerator = targetValues.GetEnumerator();
-      var scaledBoundedEstimatedValuesEnumerator = Enumerable.Range(0, i).Select(x => cache[x] * beta + alpha)
-        .LimitToRange(lowerEstimationLimit, upperEstimationLimit).GetEnumerator();
-
-      while (targetValuesEnumerator.MoveNext() & scaledBoundedEstimatedValuesEnumerator.MoveNext()) {
-        calculator.Add(targetValuesEnumerator.Current, scaledBoundedEstimatedValuesEnumerator.Current);
-      }
-    }
+    protected SymbolicRegressionSingleObjectiveEvaluator(): base() {}    
   }
 }

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMaxAbsoluteErrorEvaluator.cs

@@ -20 +20 @@
 #endregion
 
-using System;
 using System.Collections.Generic;
 using HeuristicLab.Common;
@@ -47 +46 @@
       IEnumerable<int> rows = GenerateRowsToEvaluate();
 
-      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScaling);
+      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScalingParameter.ActualValue.Value);
       QualityParameter.ActualValue = new DoubleValue(quality);
 
@@ -68 +67 @@
         mse = OnlineMaxAbsoluteErrorCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
       }
-      if (errorState != OnlineCalculatorError.None) return Double.NaN;
-      else return mse;
+      if (errorState != OnlineCalculatorError.None) return double.NaN;
+      return mse;
     }
 
@@ -75 +74 @@
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
       EstimationLimitsParameter.ExecutionContext = context;
+      ApplyLinearScalingParameter.ExecutionContext = context;
 
-      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScaling);
+      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
       EstimationLimitsParameter.ExecutionContext = null;
+      ApplyLinearScalingParameter.ExecutionContext = null;
 
       return mse;

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMeanAbsoluteErrorEvaluator.cs

@@ -20 +20 @@
 #endregion
 
-using System;
 using System.Collections.Generic;
 using HeuristicLab.Common;
@@ -47 +46 @@
       IEnumerable<int> rows = GenerateRowsToEvaluate();
 
-      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScaling);
+      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScalingParameter.ActualValue.Value);
       QualityParameter.ActualValue = new DoubleValue(quality);
 
@@ -58 +57 @@
       OnlineCalculatorError errorState;
 
-      double mse;
+      double mae;
       if (applyLinearScaling) {
         var maeCalculator = new OnlineMeanAbsoluteErrorCalculator();
         CalculateWithScaling(targetValues, estimatedValues, lowerEstimationLimit, upperEstimationLimit, maeCalculator, problemData.Dataset.Rows);
         errorState = maeCalculator.ErrorState;
-        mse = maeCalculator.MeanAbsoluteError;
+        mae = maeCalculator.MeanAbsoluteError;
       } else {
-        IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit,
-                                                                                  upperEstimationLimit);
-        mse = OnlineMeanSquaredErrorCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
+        IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
+        mae = OnlineMeanAbsoluteErrorCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
       }
-      if (errorState != OnlineCalculatorError.None) return Double.NaN;
-      else return mse;
+      if (errorState != OnlineCalculatorError.None) return double.NaN;
+      return mae;
     }
 
@@ -76 +74 @@
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
       EstimationLimitsParameter.ExecutionContext = context;
+      ApplyLinearScalingParameter.ExecutionContext = context;
 
-      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScaling);
+      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
       EstimationLimitsParameter.ExecutionContext = null;
+      ApplyLinearScalingParameter.ExecutionContext = null;
 
       return mse;

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectiveMeanSquaredErrorEvaluator.cs

@@ -20 +20 @@
 #endregion
 
-using System;
 using System.Collections.Generic;
 using HeuristicLab.Common;
@@ -47 +46 @@
       IEnumerable<int> rows = GenerateRowsToEvaluate();
 
-      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScaling);
+      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScalingParameter.ActualValue.Value);
       QualityParameter.ActualValue = new DoubleValue(quality);
 
@@ -68 +67 @@
         mse = OnlineMeanSquaredErrorCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
       }
-      if (errorState != OnlineCalculatorError.None) return Double.NaN;
-      else return mse;
+      if (errorState != OnlineCalculatorError.None) return double.NaN;
+      return mse;
     }
 
@@ -75 +74 @@
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
       EstimationLimitsParameter.ExecutionContext = context;
+      ApplyLinearScalingParameter.ExecutionContext = context;
 
-      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScaling);
+      double mse = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
       EstimationLimitsParameter.ExecutionContext = null;
+      ApplyLinearScalingParameter.ExecutionContext = null;
 
       return mse;

branches/OaaS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/SymbolicRegressionSingleObjectivePearsonRSquaredEvaluator.cs

@@ -48 +48 @@
       IEnumerable<int> rows = GenerateRowsToEvaluate();
 
-      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows);
+      double quality = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, solution, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, ProblemDataParameter.ActualValue, rows, ApplyLinearScalingParameter.ActualValue.Value);
       QualityParameter.ActualValue = new DoubleValue(quality);
 
@@ -54 +54 @@
     }
 
-    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows) {
+    public static double Calculate(ISymbolicDataAnalysisExpressionTreeInterpreter interpreter, ISymbolicExpressionTree solution, double lowerEstimationLimit, double upperEstimationLimit, IRegressionProblemData problemData, IEnumerable<int> rows, bool applyLinearScaling) {
       IEnumerable<double> estimatedValues = interpreter.GetSymbolicExpressionTreeValues(solution, problemData.Dataset, rows);
-      IEnumerable<double> originalValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
+      IEnumerable<double> targetValues = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows);
       OnlineCalculatorError errorState;
-      double r2 = OnlinePearsonsRSquaredCalculator.Calculate(estimatedValues, originalValues, out errorState);
-      if (errorState != OnlineCalculatorError.None) return 0.0;
-      else return r2;
+
+      double r2;
+      if (applyLinearScaling) {
+        var r2Calculator = new OnlinePearsonsRSquaredCalculator();
+        CalculateWithScaling(targetValues, estimatedValues, lowerEstimationLimit, upperEstimationLimit, r2Calculator, problemData.Dataset.Rows);
+        errorState = r2Calculator.ErrorState;
+        r2 = r2Calculator.RSquared;
+      } else {
+        IEnumerable<double> boundedEstimatedValues = estimatedValues.LimitToRange(lowerEstimationLimit, upperEstimationLimit);
+        r2 = OnlinePearsonsRSquaredCalculator.Calculate(targetValues, boundedEstimatedValues, out errorState);
+      }
+      if (errorState != OnlineCalculatorError.None) return double.NaN;
+      return r2;
     }
 
@@ -66 +76 @@
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = context;
       EstimationLimitsParameter.ExecutionContext = context;
+      ApplyLinearScalingParameter.ExecutionContext = context;
 
-      double r2 = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows);
+      double r2 = Calculate(SymbolicDataAnalysisTreeInterpreterParameter.ActualValue, tree, EstimationLimitsParameter.ActualValue.Lower, EstimationLimitsParameter.ActualValue.Upper, problemData, rows, ApplyLinearScalingParameter.ActualValue.Value);
 
       SymbolicDataAnalysisTreeInterpreterParameter.ExecutionContext = null;
       EstimationLimitsParameter.ExecutionContext = null;
+      ApplyLinearScalingParameter.ExecutionContext = null;
 
       return r2;

@@ -21 +21 @@
 protoc.exe
 _ReSharper.HeuristicLab 3.3 Tests
+Google.ProtocolBuffers-2.4.1.473.dll
 packages

@@ +1 @@
+*.user
+Plugin.cs
 bin
-*.user
-HeuristicLabProblemsDataAnalysisSymbolicRegressionPlugin.cs
 obj
-*.vs10x
-Plugin.cs