source: branches/3107_LearningALPS/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/Creators/NodeImpactsReinitializationStrategyController.cs @ 17917

using System;
using System.Collections.Generic;
using System.Linq;

using HEAL.Attic;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
using HeuristicLab.Operators;
using HeuristicLab.Optimization;
using HeuristicLab.Parameters;

namespace HeuristicLab.Problems.DataAnalysis.Symbolic.Regression {
  [Item("NodeImpactsReinitializationStrategyController", "")]
  [StorableType("C015C8C3-283D-4F51-A582-367587596709")]
  public class NodeImpactsReinitializationStrategyController : InstrumentedOperator, IReinitializationStrategyController {
    private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
    private const string SymbolicExpressionTreeGrammarParameterName = "SymbolicExpressionTreeGrammar";
    private const string SymbolicDataAnalysisTreeInterpreterParameterName = "SymbolicDataAnalysisTreeInterpreter";
    private const string ProblemDataParameterName = "ProblemData";
    private const string EstimationLimitsParameterName = "EstimationLimits";

    #region Parameter Properties
    public IScopeTreeLookupParameter<ISymbolicExpressionTree> SymbolicExpressionTreeParameter {
      get { return (IScopeTreeLookupParameter<ISymbolicExpressionTree>)Parameters[SymbolicExpressionTreeParameterName]; }
    }

    public IValueLookupParameter<ISymbolicExpressionGrammar> SymbolicExpressionTreeGrammarParameter {
      get { return (IValueLookupParameter<ISymbolicExpressionGrammar>)Parameters[SymbolicExpressionTreeGrammarParameterName]; }
    }

    public ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter> SymbolicDataAnalysisTreeInterpreterParameter {
      get { return (ILookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>)Parameters[SymbolicDataAnalysisTreeInterpreterParameterName]; }
    }

    public ILookupParameter<IRegressionProblemData> ProblemDataParameter {
      get { return (ILookupParameter<IRegressionProblemData>)Parameters[ProblemDataParameterName]; }
    }

    public IValueLookupParameter<DoubleLimit> EstimationLimitsParameter {
      get { return (IValueLookupParameter<DoubleLimit>)Parameters[EstimationLimitsParameterName]; }
    }
    #endregion

    #region Constructors
    public NodeImpactsReinitializationStrategyController() {
      Parameters.Add(new ScopeTreeLookupParameter<ISymbolicExpressionTree>(SymbolicExpressionTreeParameterName, "The symbolic expression tree whose length should be calculated."));
      Parameters.Add(new ValueLookupParameter<ISymbolicExpressionGrammar>(SymbolicExpressionTreeGrammarParameterName, "The tree grammar that defines the correct syntax of symbolic expression trees that should be created."));
      Parameters.Add(new LookupParameter<ISymbolicDataAnalysisExpressionTreeInterpreter>(SymbolicDataAnalysisTreeInterpreterParameterName, "The symbolic data analysis tree interpreter for the symbolic expression tree."));
      Parameters.Add(new LookupParameter<IRegressionProblemData>(ProblemDataParameterName, "The problem data for the symbolic regression solution."));
      Parameters.Add(new ValueLookupParameter<DoubleLimit>(EstimationLimitsParameterName, "The lower and upper limit for the estimated values produced by the symbolic regression model."));
    }

    private NodeImpactsReinitializationStrategyController(NodeImpactsReinitializationStrategyController original, Cloner cloner)
      : base(original, cloner) { }
    public override IDeepCloneable Clone(Cloner cloner) {
      return new NodeImpactsReinitializationStrategyController(this, cloner);
    }

    [StorableConstructor]
    private NodeImpactsReinitializationStrategyController(StorableConstructorFlag _) : base(_) { }
    [StorableHook(HookType.AfterDeserialization)]
    private void AfterDeserialization() {
    }
    #endregion


    public override IOperation InstrumentedApply() {
      var trees = SymbolicExpressionTreeParameter.ActualValue;
      var grammar = SymbolicExpressionTreeGrammarParameter.ActualValue;
      var pd = ProblemDataParameter.ActualValue;
      var interpreter = SymbolicDataAnalysisTreeInterpreterParameter.ActualValue;

      if (interpreter == null) {
        interpreter = new SymbolicDataAnalysisExpressionTreeBatchInterpreter();
      }

      var estimationLimits = EstimationLimitsParameter.ActualValue;
      var ds = ((Dataset)pd.Dataset).ToModifiable();

      var symbolImpacts = new Dictionary<string, double>();
      var symbolCounts = new Dictionary<string, int>();

      var impactValuesCalculator = new SymbolicRegressionSolutionImpactValuesCalculator();

      foreach (var tree in trees) {
        var model = new SymbolicRegressionModel(pd.TargetVariable, tree, interpreter, estimationLimits.Lower, estimationLimits.Upper);
        var root = tree.Root;
        while (root.Symbol is ProgramRootSymbol || root.Symbol is StartSymbol) {
          root = root.GetSubtree(0);
        }

        foreach(var node in root.IterateNodesPrefix().Where(x => x.SubtreeCount > 0)) {
          impactValuesCalculator.CalculateImpactAndReplacementValues(model, node, pd, pd.TrainingIndices, out double impactValue, out double replacementValue, out double newQuality);

          var name = node.Symbol.Name;
          if (symbolCounts.TryGetValue(name, out int count)) {
            symbolCounts[name] = count + 1;
            symbolImpacts[name] += impactValue;
          } else {
            symbolCounts[name] = 1;
            symbolImpacts[name] = impactValue;
          }
        }
      }

      foreach(var symbol in grammar.AllowedSymbols) {
        if (symbolImpacts.TryGetValue(symbol.Name, out double impact)) {
          var f = Math.Max(0, impact / symbolCounts[symbol.Name]); // do something clever here
          symbol.InitialFrequency = f;
        }
      }

      return base.InstrumentedApply();
    }
  }
}