source: branches/DataPreprocessing/HeuristicLab.Problems.DataAnalysis/3.4/Implementation/Regression/RegressionProblemData.cs @ 11070

Last change on this file since 11070 was 11070, checked in by mkommend, 5 years ago

#2206: Clean up of data preprocessing code (removed unused code, used more appropriate collections, hiding of the backtransform button of symreg models).

File size: 8.7 KB
Line 
1#region License Information
2/* HeuristicLab
3 * Copyright (C) 2002-2013 Heuristic and Evolutionary Algorithms Laboratory (HEAL)
4 *
5 * This file is part of HeuristicLab.
6 *
7 * HeuristicLab is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation, either version 3 of the License, or
10 * (at your option) any later version.
11 *
12 * HeuristicLab is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License
18 * along with HeuristicLab. If not, see <http://www.gnu.org/licenses/>.
19 */
20#endregion
21
22using System;
23using System.Collections.Generic;
24using System.Linq;
25using HeuristicLab.Common;
26using HeuristicLab.Core;
27using HeuristicLab.Data;
28using HeuristicLab.Parameters;
29using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
30
31namespace HeuristicLab.Problems.DataAnalysis {
32  [StorableClass]
33  [Item("RegressionProblemData", "Represents an item containing all data defining a regression problem.")]
34  public class RegressionProblemData : DataAnalysisProblemData, IRegressionProblemData, IStorableContent {
35    protected const string TargetVariableParameterName = "TargetVariable";
36    public string Filename { get; set; }
37
38    #region default data
39    private static double[,] kozaF1 = new double[,] {
40          {2.017885919, -1.449165046},
41          {1.30060506,  -1.344523885},
42          {1.147134798, -1.317989331},
43          {0.877182504, -1.266142284},
44          {0.852562452, -1.261020794},
45          {0.431095788, -1.158793317},
46          {0.112586002, -1.050908405},
47          {0.04594507,  -1.021989402},
48          {0.042572879, -1.020438113},
49          {-0.074027291,  -0.959859562},
50          {-0.109178553,  -0.938094706},
51          {-0.259721109,  -0.803635355},
52          {-0.272991057,  -0.387519561},
53          {-0.161978191,  -0.193611001},
54          {-0.102489983,  -0.114215349},
55          {-0.01469968, -0.014918985},
56          {-0.008863365,  -0.008942626},
57          {0.026751057, 0.026054094},
58          {0.166922436, 0.14309643},
59          {0.176953808, 0.1504144},
60          {0.190233418, 0.159916534},
61          {0.199800708, 0.166635331},
62          {0.261502822, 0.207600348},
63          {0.30182879,  0.232370249},
64          {0.83763905,  0.468046718}
65    };
66    private static readonly Dataset defaultDataset;
67    private static readonly IEnumerable<string> defaultAllowedInputVariables;
68    private static readonly string defaultTargetVariable;
69
70    private static readonly RegressionProblemData emptyProblemData;
71    public static RegressionProblemData EmptyProblemData {
72      get { return emptyProblemData; }
73    }
74
75    static RegressionProblemData() {
76      defaultDataset = new Dataset(new string[] { "y", "x" }, kozaF1);
77      defaultDataset.Name = "Fourth-order Polynomial Function Benchmark Dataset";
78      defaultDataset.Description = "f(x) = x^4 + x^3 + x^2 + x^1";
79      defaultAllowedInputVariables = new List<string>() { "x" };
80      defaultTargetVariable = "y";
81
82      var problemData = new RegressionProblemData();
83      problemData.Parameters.Clear();
84      problemData.Name = "Empty Regression ProblemData";
85      problemData.Description = "This ProblemData acts as place holder before the correct problem data is loaded.";
86      problemData.isEmpty = true;
87
88      problemData.Parameters.Add(new FixedValueParameter<Dataset>(DatasetParameterName, "", new Dataset()));
89      problemData.Parameters.Add(new FixedValueParameter<ReadOnlyCheckedItemList<StringValue>>(InputVariablesParameterName, ""));
90      problemData.Parameters.Add(new FixedValueParameter<IntRange>(TrainingPartitionParameterName, "", (IntRange)new IntRange(0, 0).AsReadOnly()));
91      problemData.Parameters.Add(new FixedValueParameter<IntRange>(TestPartitionParameterName, "", (IntRange)new IntRange(0, 0).AsReadOnly()));
92      problemData.Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>()));
93      emptyProblemData = problemData;
94    }
95    #endregion
96
97    public IConstrainedValueParameter<StringValue> TargetVariableParameter {
98      get { return (IConstrainedValueParameter<StringValue>)Parameters[TargetVariableParameterName]; }
99    }
100    public string TargetVariable {
101      get { return TargetVariableParameter.Value.Value; }
102      set {
103        if (value == null) throw new ArgumentNullException("targetVariable", "The provided value for the targetVariable is null.");
104        if (value == TargetVariable) return;
105
106        var matchingParameterValue = TargetVariableParameter.ValidValues.FirstOrDefault(v => v.Value == value);
107        if (matchingParameterValue == null) throw new ArgumentException("The provided value is not valid as the targetVariable.", "targetVariable");
108        TargetVariableParameter.Value = matchingParameterValue;
109      }
110    }
111
112    [StorableConstructor]
113    protected RegressionProblemData(bool deserializing) : base(deserializing) { }
114    [StorableHook(HookType.AfterDeserialization)]
115    private void AfterDeserialization() {
116      RegisterParameterEvents();
117    }
118
119    protected RegressionProblemData(RegressionProblemData original, Cloner cloner)
120      : base(original, cloner) {
121      RegisterParameterEvents();
122    }
123    public override IDeepCloneable Clone(Cloner cloner) {
124      if (this == emptyProblemData) return emptyProblemData;
125      return new RegressionProblemData(this, cloner);
126    }
127
128    public RegressionProblemData()
129      : this(defaultDataset, defaultAllowedInputVariables, defaultTargetVariable) {
130    }
131    public RegressionProblemData(IRegressionProblemData regressionProblemData)
132      : this(regressionProblemData.Dataset, regressionProblemData.AllowedInputVariables, regressionProblemData.TargetVariable) {
133      TrainingPartition.Start = regressionProblemData.TrainingPartition.Start;
134      TrainingPartition.End = regressionProblemData.TrainingPartition.End;
135      TestPartition.Start = regressionProblemData.TestPartition.Start;
136      TestPartition.End = regressionProblemData.TestPartition.End;
137    }
138
139    public RegressionProblemData(Dataset dataset, IEnumerable<string> allowedInputVariables, string targetVariable, IEnumerable<ITransformation> transformations = null)
140      : base(dataset, allowedInputVariables, transformations ?? Enumerable.Empty<ITransformation>()) {
141      var variables = InputVariables.Select(x => x.AsReadOnly()).ToList();
142      Parameters.Add(new ConstrainedValueParameter<StringValue>(TargetVariableParameterName, new ItemSet<StringValue>(variables), variables.Where(x => x.Value == targetVariable).First()));
143      RegisterParameterEvents();
144    }
145
146    private void RegisterParameterEvents() {
147      TargetVariableParameter.ValueChanged += new EventHandler(TargetVariableParameter_ValueChanged);
148    }
149    private void TargetVariableParameter_ValueChanged(object sender, EventArgs e) {
150      OnChanged();
151    }
152
153    protected override bool IsProblemDataCompatible(IDataAnalysisProblemData problemData, out string errorMessage) {
154      if (problemData == null) throw new ArgumentNullException("problemData", "The provided problemData is null.");
155      IRegressionProblemData regressionProblemData = problemData as IRegressionProblemData;
156      if (regressionProblemData == null)
157        throw new ArgumentException("The problem data is not a regression problem data. Instead a " + problemData.GetType().GetPrettyName() + " was provided.", "problemData");
158
159      var returnValue = base.IsProblemDataCompatible(problemData, out errorMessage);
160      //check targetVariable
161      if (problemData.InputVariables.All(var => var.Value != TargetVariable)) {
162        errorMessage = string.Format("The target variable {0} is not present in the new problem data.", TargetVariable)
163                       + Environment.NewLine + errorMessage;
164        return false;
165      }
166      return returnValue;
167    }
168
169    public override void AdjustProblemDataProperties(IDataAnalysisProblemData problemData) {
170      if (problemData == null) throw new ArgumentNullException("problemData", "The provided problemData is null.");
171      RegressionProblemData regressionProblemData = problemData as RegressionProblemData;
172      if (regressionProblemData == null)
173        throw new ArgumentException("The problem data is not a regression problem data. Instead a " + problemData.GetType().GetPrettyName() + " was provided.", "problemData");
174
175      base.AdjustProblemDataProperties(problemData);
176      TargetVariable = regressionProblemData.TargetVariable;
177    }
178  }
179}
Note: See TracBrowser for help on using the repository browser.