source: branches/2679_HeuristicLab.GoalSeekingProblem/HeuristicLab.GoalSeekingProblem/3.4/GoalSeekingUtil.cs @ 17607

#region License Information
/* HeuristicLab
 * Copyright (C) 2002-2016 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
 *
 * This file is part of HeuristicLab.
 *
 * HeuristicLab is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * HeuristicLab is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
 */
#endregion

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using HeuristicLab.Collections;
using HeuristicLab.Common;
using HeuristicLab.Core;
using HeuristicLab.Data;
using HeuristicLab.Encodings.RealVectorEncoding;
using HeuristicLab.Problems.DataAnalysis;

namespace HeuristicLab.GoalSeeking {
  internal static class GoalSeekingUtil {
    internal static void UpdateTargets(ICheckedItemList<GoalParameter> goals, IEnumerable<IRegressionModel> models, EventHandler changedEventHandler) {
      var targets = models.Select(x => x.TargetVariable).Distinct().OrderBy(x => x);
      var dict = goals.ToDictionary(x => x.Name, x => x);
      goals.Clear();
      foreach (var t in targets) {
        GoalParameter gp;
        if (dict.ContainsKey(t)) {
          gp = dict[t];
        } else {
          var goal = double.NaN;
          var weight = 1.0;
          var variance = double.NaN;
          var step = 1e-6;
          gp = new GoalParameter(t, goal, weight, variance, step, active: false); // new goals are not active by default as they need to be configured by the user
          gp.Changed += changedEventHandler;
        }
        goals.Add(gp, gp.Active);
      }
    }

    internal static void UpdateInputs(ICheckedItemList<InputParameter> inputs, IEnumerable<IRegressionModel> models, EventHandler changedEventHandler) {
      var variables = models.SelectMany(x => x.VariablesUsedForPrediction).Distinct();
      var dict = inputs.ToDictionary(x => x.Name, x => x); // save old parameter settings
      inputs.Clear();
      foreach (var v in variables) {
        InputParameter ip;
        if (dict.ContainsKey(v)) {
          ip = dict[v];
        } else {
          ip = new InputParameter(v, double.NaN, -1, 1, 1e-6, active: false); // new inputs are not active by default. the user needs to configure them first
          ip.Changed += changedEventHandler;
        }
        inputs.Add(ip, ip.Active);
      }
    }

    internal static RealVectorEncoding CreateEncoding(IEnumerable<InputParameter> inputParameters) {
      var encoding = new RealVectorEncoding(inputParameters.Count());
      encoding.Bounds = new DoubleMatrix(encoding.Length, 2); // only two columns: min and max
      encoding.Bounds.RowNames = inputParameters.Select(x => x.Name);
      encoding.Bounds.ColumnNames = new[] { "Min.", "Max." };

      int i = 0;
      foreach (var parameter in inputParameters) {
        encoding.Bounds[i, 0] = parameter.Min;
        encoding.Bounds[i, 1] = parameter.Max;
        ++i;
      }
      return encoding;
    }

    internal static void UpdateEncoding(RealVectorEncoding encoding, IEnumerable<InputParameter> inputParameters) {
      encoding.Length = inputParameters.Count();
      encoding.Bounds = new DoubleMatrix(encoding.Length, 2);
      encoding.Bounds.RowNames = inputParameters.Select(x => x.Name);
      encoding.Bounds.ColumnNames = new[] { "Min.", "Max." };

      int i = 0;
      foreach (var parameter in inputParameters) {
        encoding.Bounds[i, 0] = parameter.Min;
        encoding.Bounds[i, 1] = parameter.Max;
        ++i;
      }
    }

    internal static void ValidateGoalsAndInputs(IEnumerable<GoalParameter> goals, IEnumerable<InputParameter> inputs) {
      var notConfiguredGoals = goals.Where(x => x.Active && (double.IsNaN(x.Variance) || double.IsNaN(x.Goal))).ToList();
      var notConfiguredInputs = inputs.Where(x => x.Active && (double.IsNaN(x.Min) || double.IsNaN(x.Max) || double.IsNaN(x.Value))).ToList();
      StringBuilder sb = null;
      if (notConfiguredGoals.Any()) {
        sb = new StringBuilder();
        sb.AppendLine("The following GoalParameters are not configured.");
        foreach (var goal in notConfiguredGoals) {
          sb.AppendLine(goal.Name);
        }
      }
      if (notConfiguredInputs.Any()) {
        if (sb == null)
          sb = new StringBuilder();
        sb.AppendLine("The following InpuTParameters are not configured.");
        foreach (var input in notConfiguredInputs) {
          sb.AppendLine(input.Name);
        }
      }
      if (sb != null)
        throw new InvalidOperationException(sb.ToString());
    }

    internal static void Configure(IEnumerable<GoalParameter> goals, IEnumerable<InputParameter> inputs, IRegressionProblemData problemData, int row) {
      var inputMap = inputs.ToDictionary(x => x.Name, x => x);
      var goalMap = goals.ToDictionary(x => x.Name, x => x);

      foreach (var variable in problemData.Dataset.DoubleVariables) {
        if (inputMap.ContainsKey(variable)) {
          var values = problemData.Dataset.GetReadOnlyDoubleValues(variable);
          var input = inputMap[variable];
          input.Value = values[row];
          double min = values[0], max = values[0];
          foreach (var v in values) {
            if (min > v) min = v;
            if (max < v) max = v;
          }
          input.Min = min;
          input.Max = max;
        } else if (goalMap.ContainsKey(variable)) {
          var values = problemData.Dataset.GetReadOnlyDoubleValues(variable);
          var goal = goalMap[variable];
          goal.Variance = values.Variance();
          goal.Goal = values[row];
        }
      }
    }

    internal static void Goals_CheckedItemsChanged(object sender, CollectionItemsChangedEventArgs<IndexedItem<GoalParameter>> args) {
      var goals = (CheckedItemList<GoalParameter>)sender;
      foreach (var item in args.Items.Select(x => x.Value)) {
        item.Active = goals.ItemChecked(item);
      }
    }

    internal static void Inputs_CheckedItemsChanged(object sender, CollectionItemsChangedEventArgs<IndexedItem<InputParameter>> args) {
      var inputs = (CheckedItemList<InputParameter>)sender;
      foreach (var item in args.Items.Select(x => x.Value)) {
        item.Active = inputs.ItemChecked(item);
      }
    }

    internal static void RaiseEvent(object sender, EventHandler handler) {
      if (handler == null) return;
      handler(sender, EventArgs.Empty);
    }
  }
}