| 1 | #region License Information
|
|---|
| 2 | /* HeuristicLab
|
|---|
| 3 | * Copyright (C) 2002-2010 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 |
|
|---|
| 22 | using System.Collections.Generic;
|
|---|
| 23 | using System.Linq;
|
|---|
| 24 | using HeuristicLab.Analysis;
|
|---|
| 25 | using HeuristicLab.Common;
|
|---|
| 26 | using HeuristicLab.Core;
|
|---|
| 27 | using HeuristicLab.Data;
|
|---|
| 28 | using HeuristicLab.Encodings.SymbolicExpressionTreeEncoding;
|
|---|
| 29 | using HeuristicLab.Operators;
|
|---|
| 30 | using HeuristicLab.Optimization;
|
|---|
| 31 | using HeuristicLab.Parameters;
|
|---|
| 32 | using HeuristicLab.Persistence.Default.CompositeSerializers.Storable;
|
|---|
| 33 | using HeuristicLab.Problems.DataAnalysis.Evaluators;
|
|---|
| 34 | using HeuristicLab.Problems.DataAnalysis.Symbolic;
|
|---|
| 35 | using System;
|
|---|
| 36 |
|
|---|
| 37 | namespace HeuristicLab.Problems.DataAnalysis.Regression.Symbolic.Analyzers {
|
|---|
| 38 | [Item("OverfittingAnalyzer", "")]
|
|---|
| 39 | [StorableClass]
|
|---|
| 40 | public sealed class OverfittingAnalyzer : SingleSuccessorOperator, ISymbolicRegressionAnalyzer {
|
|---|
| 41 | private const string RandomParameterName = "Random";
|
|---|
| 42 | private const string SymbolicExpressionTreeParameterName = "SymbolicExpressionTree";
|
|---|
| 43 | private const string SymbolicExpressionTreeInterpreterParameterName = "SymbolicExpressionTreeInterpreter";
|
|---|
| 44 | private const string ProblemDataParameterName = "ProblemData";
|
|---|
| 45 | private const string ValidationSamplesStartParameterName = "SamplesStart";
|
|---|
| 46 | private const string ValidationSamplesEndParameterName = "SamplesEnd";
|
|---|
| 47 | private const string UpperEstimationLimitParameterName = "UpperEstimationLimit";
|
|---|
| 48 | private const string LowerEstimationLimitParameterName = "LowerEstimationLimit";
|
|---|
| 49 | private const string EvaluatorParameterName = "Evaluator";
|
|---|
| 50 | private const string MaximizationParameterName = "Maximization";
|
|---|
| 51 | private const string RelativeNumberOfEvaluatedSamplesParameterName = "RelativeNumberOfEvaluatedSamples";
|
|---|
| 52 |
|
|---|
| 53 | #region parameter properties
|
|---|
| 54 | public ILookupParameter<IRandom> RandomParameter {
|
|---|
| 55 | get { return (ILookupParameter<IRandom>)Parameters[RandomParameterName]; }
|
|---|
| 56 | }
|
|---|
| 57 | public ScopeTreeLookupParameter<SymbolicExpressionTree> SymbolicExpressionTreeParameter {
|
|---|
| 58 | get { return (ScopeTreeLookupParameter<SymbolicExpressionTree>)Parameters[SymbolicExpressionTreeParameterName]; }
|
|---|
| 59 | }
|
|---|
| 60 | public ScopeTreeLookupParameter<DoubleValue> QualityParameter {
|
|---|
| 61 | get { return (ScopeTreeLookupParameter<DoubleValue>)Parameters["Quality"]; }
|
|---|
| 62 | }
|
|---|
| 63 | public ScopeTreeLookupParameter<DoubleValue> ValidationQualityParameter {
|
|---|
| 64 | get { return (ScopeTreeLookupParameter<DoubleValue>)Parameters["ValidationQuality"]; }
|
|---|
| 65 | }
|
|---|
| 66 | public IValueLookupParameter<ISymbolicExpressionTreeInterpreter> SymbolicExpressionTreeInterpreterParameter {
|
|---|
| 67 | get { return (IValueLookupParameter<ISymbolicExpressionTreeInterpreter>)Parameters[SymbolicExpressionTreeInterpreterParameterName]; }
|
|---|
| 68 | }
|
|---|
| 69 | public ILookupParameter<ISymbolicRegressionEvaluator> EvaluatorParameter {
|
|---|
| 70 | get { return (ILookupParameter<ISymbolicRegressionEvaluator>)Parameters[EvaluatorParameterName]; }
|
|---|
| 71 | }
|
|---|
| 72 | public ILookupParameter<BoolValue> MaximizationParameter {
|
|---|
| 73 | get { return (ILookupParameter<BoolValue>)Parameters[MaximizationParameterName]; }
|
|---|
| 74 | }
|
|---|
| 75 | public IValueLookupParameter<DataAnalysisProblemData> ProblemDataParameter {
|
|---|
| 76 | get { return (IValueLookupParameter<DataAnalysisProblemData>)Parameters[ProblemDataParameterName]; }
|
|---|
| 77 | }
|
|---|
| 78 | public IValueLookupParameter<IntValue> ValidationSamplesStartParameter {
|
|---|
| 79 | get { return (IValueLookupParameter<IntValue>)Parameters[ValidationSamplesStartParameterName]; }
|
|---|
| 80 | }
|
|---|
| 81 | public IValueLookupParameter<IntValue> ValidationSamplesEndParameter {
|
|---|
| 82 | get { return (IValueLookupParameter<IntValue>)Parameters[ValidationSamplesEndParameterName]; }
|
|---|
| 83 | }
|
|---|
| 84 | public IValueParameter<PercentValue> RelativeNumberOfEvaluatedSamplesParameter {
|
|---|
| 85 | get { return (IValueParameter<PercentValue>)Parameters[RelativeNumberOfEvaluatedSamplesParameterName]; }
|
|---|
| 86 | }
|
|---|
| 87 |
|
|---|
| 88 | public IValueLookupParameter<DoubleValue> UpperEstimationLimitParameter {
|
|---|
| 89 | get { return (IValueLookupParameter<DoubleValue>)Parameters[UpperEstimationLimitParameterName]; }
|
|---|
| 90 | }
|
|---|
| 91 | public IValueLookupParameter<DoubleValue> LowerEstimationLimitParameter {
|
|---|
| 92 | get { return (IValueLookupParameter<DoubleValue>)Parameters[LowerEstimationLimitParameterName]; }
|
|---|
| 93 | }
|
|---|
| 94 | public ILookupParameter<PercentValue> RelativeValidationQualityParameter {
|
|---|
| 95 | get { return (ILookupParameter<PercentValue>)Parameters["RelativeValidationQuality"]; }
|
|---|
| 96 | }
|
|---|
| 97 | //public IValueLookupParameter<PercentValue> RelativeValidationQualityLowerLimitParameter {
|
|---|
| 98 | // get { return (IValueLookupParameter<PercentValue>)Parameters["RelativeValidationQualityLowerLimit"]; }
|
|---|
| 99 | //}
|
|---|
| 100 | //public IValueLookupParameter<PercentValue> RelativeValidationQualityUpperLimitParameter {
|
|---|
| 101 | // get { return (IValueLookupParameter<PercentValue>)Parameters["RelativeValidationQualityUpperLimit"]; }
|
|---|
| 102 | //}
|
|---|
| 103 | public ILookupParameter<DoubleValue> TrainingValidationQualityCorrelationParameter {
|
|---|
| 104 | get { return (ILookupParameter<DoubleValue>)Parameters["TrainingValidationCorrelation"]; }
|
|---|
| 105 | }
|
|---|
| 106 | public IValueLookupParameter<DoubleValue> CorrelationLimitParameter {
|
|---|
| 107 | get { return (IValueLookupParameter<DoubleValue>)Parameters["CorrelationLimit"]; }
|
|---|
| 108 | }
|
|---|
| 109 | public ILookupParameter<BoolValue> OverfittingParameter {
|
|---|
| 110 | get { return (ILookupParameter<BoolValue>)Parameters["Overfitting"]; }
|
|---|
| 111 | }
|
|---|
| 112 | public ILookupParameter<ResultCollection> ResultsParameter {
|
|---|
| 113 | get { return (ILookupParameter<ResultCollection>)Parameters["Results"]; }
|
|---|
| 114 | }
|
|---|
| 115 | public ILookupParameter<DoubleValue> InitialTrainingQualityParameter {
|
|---|
| 116 | get { return (ILookupParameter<DoubleValue>)Parameters["InitialTrainingQuality"]; }
|
|---|
| 117 | }
|
|---|
| 118 | public ILookupParameter<DoubleMatrix> TrainingAndValidationQualitiesParameter {
|
|---|
| 119 | get { return (ILookupParameter<DoubleMatrix>)Parameters["TrainingAndValidationQualities"]; }
|
|---|
| 120 | }
|
|---|
| 121 | public IValueLookupParameter<DoubleValue> PercentileParameter {
|
|---|
| 122 | get { return (IValueLookupParameter<DoubleValue>)Parameters["Percentile"]; }
|
|---|
| 123 | }
|
|---|
| 124 | #endregion
|
|---|
| 125 | #region properties
|
|---|
| 126 | public IRandom Random {
|
|---|
| 127 | get { return RandomParameter.ActualValue; }
|
|---|
| 128 | }
|
|---|
| 129 | public ItemArray<SymbolicExpressionTree> SymbolicExpressionTree {
|
|---|
| 130 | get { return SymbolicExpressionTreeParameter.ActualValue; }
|
|---|
| 131 | }
|
|---|
| 132 | public ISymbolicExpressionTreeInterpreter SymbolicExpressionTreeInterpreter {
|
|---|
| 133 | get { return SymbolicExpressionTreeInterpreterParameter.ActualValue; }
|
|---|
| 134 | }
|
|---|
| 135 | public ISymbolicRegressionEvaluator Evaluator {
|
|---|
| 136 | get { return EvaluatorParameter.ActualValue; }
|
|---|
| 137 | }
|
|---|
| 138 | public BoolValue Maximization {
|
|---|
| 139 | get { return MaximizationParameter.ActualValue; }
|
|---|
| 140 | }
|
|---|
| 141 | public DataAnalysisProblemData ProblemData {
|
|---|
| 142 | get { return ProblemDataParameter.ActualValue; }
|
|---|
| 143 | }
|
|---|
| 144 | public IntValue ValidiationSamplesStart {
|
|---|
| 145 | get { return ValidationSamplesStartParameter.ActualValue; }
|
|---|
| 146 | }
|
|---|
| 147 | public IntValue ValidationSamplesEnd {
|
|---|
| 148 | get { return ValidationSamplesEndParameter.ActualValue; }
|
|---|
| 149 | }
|
|---|
| 150 | public PercentValue RelativeNumberOfEvaluatedSamples {
|
|---|
| 151 | get { return RelativeNumberOfEvaluatedSamplesParameter.Value; }
|
|---|
| 152 | }
|
|---|
| 153 |
|
|---|
| 154 | public DoubleValue UpperEstimationLimit {
|
|---|
| 155 | get { return UpperEstimationLimitParameter.ActualValue; }
|
|---|
| 156 | }
|
|---|
| 157 | public DoubleValue LowerEstimationLimit {
|
|---|
| 158 | get { return LowerEstimationLimitParameter.ActualValue; }
|
|---|
| 159 | }
|
|---|
| 160 | #endregion
|
|---|
| 161 |
|
|---|
| 162 | public OverfittingAnalyzer()
|
|---|
| 163 | : base() {
|
|---|
| 164 | Parameters.Add(new LookupParameter<IRandom>(RandomParameterName, "The random generator to use."));
|
|---|
| 165 | Parameters.Add(new LookupParameter<ISymbolicRegressionEvaluator>(EvaluatorParameterName, "The evaluator which should be used to evaluate the solution on the validation set."));
|
|---|
| 166 | Parameters.Add(new ScopeTreeLookupParameter<SymbolicExpressionTree>(SymbolicExpressionTreeParameterName, "The symbolic expression trees to analyze."));
|
|---|
| 167 | Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("Quality"));
|
|---|
| 168 | Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("ValidationQuality"));
|
|---|
| 169 | Parameters.Add(new LookupParameter<BoolValue>(MaximizationParameterName, "The direction of optimization."));
|
|---|
| 170 | Parameters.Add(new ValueLookupParameter<ISymbolicExpressionTreeInterpreter>(SymbolicExpressionTreeInterpreterParameterName, "The interpreter that should be used for the analysis of symbolic expression trees."));
|
|---|
| 171 | Parameters.Add(new ValueLookupParameter<DataAnalysisProblemData>(ProblemDataParameterName, "The problem data for which the symbolic expression tree is a solution."));
|
|---|
| 172 | Parameters.Add(new ValueLookupParameter<IntValue>(ValidationSamplesStartParameterName, "The first index of the validation partition of the data set."));
|
|---|
| 173 | Parameters.Add(new ValueLookupParameter<IntValue>(ValidationSamplesEndParameterName, "The last index of the validation partition of the data set."));
|
|---|
| 174 | Parameters.Add(new ValueParameter<PercentValue>(RelativeNumberOfEvaluatedSamplesParameterName, "The relative number of samples of the dataset partition, which should be randomly chosen for evaluation between the start and end index.", new PercentValue(1)));
|
|---|
| 175 | Parameters.Add(new ValueLookupParameter<DoubleValue>(UpperEstimationLimitParameterName, "The upper estimation limit that was set for the evaluation of the symbolic expression trees."));
|
|---|
| 176 | Parameters.Add(new ValueLookupParameter<DoubleValue>(LowerEstimationLimitParameterName, "The lower estimation limit that was set for the evaluation of the symbolic expression trees."));
|
|---|
| 177 | Parameters.Add(new LookupParameter<PercentValue>("RelativeValidationQuality"));
|
|---|
| 178 | //Parameters.Add(new ValueLookupParameter<PercentValue>("RelativeValidationQualityUpperLimit", new PercentValue(0.05)));
|
|---|
| 179 | //Parameters.Add(new ValueLookupParameter<PercentValue>("RelativeValidationQualityLowerLimit", new PercentValue(-0.05)));
|
|---|
| 180 | Parameters.Add(new LookupParameter<DoubleValue>("TrainingValidationCorrelation"));
|
|---|
| 181 | Parameters.Add(new ValueLookupParameter<DoubleValue>("CorrelationLimit", new DoubleValue(0.65)));
|
|---|
| 182 | Parameters.Add(new LookupParameter<BoolValue>("Overfitting"));
|
|---|
| 183 | Parameters.Add(new LookupParameter<ResultCollection>("Results"));
|
|---|
| 184 | Parameters.Add(new LookupParameter<DoubleValue>("InitialTrainingQuality"));
|
|---|
| 185 | Parameters.Add(new LookupParameter<DoubleMatrix>("TrainingAndValidationQualities"));
|
|---|
| 186 | Parameters.Add(new ValueLookupParameter<DoubleValue>("Percentile", new DoubleValue(1)));
|
|---|
| 187 |
|
|---|
| 188 | }
|
|---|
| 189 |
|
|---|
| 190 | [StorableConstructor]
|
|---|
| 191 | private OverfittingAnalyzer(bool deserializing) : base(deserializing) { }
|
|---|
| 192 |
|
|---|
| 193 | [StorableHook(HookType.AfterDeserialization)]
|
|---|
| 194 | private void AfterDeserialization() {
|
|---|
| 195 | if (!Parameters.ContainsKey("InitialTrainingQuality")) {
|
|---|
| 196 | Parameters.Add(new LookupParameter<DoubleValue>("InitialTrainingQuality"));
|
|---|
| 197 | }
|
|---|
| 198 | //if (!Parameters.ContainsKey("RelativeValidationQualityUpperLimit")) {
|
|---|
| 199 | // Parameters.Add(new ValueLookupParameter<PercentValue>("RelativeValidationQualityUpperLimit", new PercentValue(0.05)));
|
|---|
| 200 | //}
|
|---|
| 201 | //if (!Parameters.ContainsKey("RelativeValidationQualityLowerLimit")) {
|
|---|
| 202 | // Parameters.Add(new ValueLookupParameter<PercentValue>("RelativeValidationQualityLowerLimit", new PercentValue(-0.05)));
|
|---|
| 203 | //}
|
|---|
| 204 | if (!Parameters.ContainsKey("TrainingAndValidationQualities")) {
|
|---|
| 205 | Parameters.Add(new LookupParameter<DoubleMatrix>("TrainingAndValidationQualities"));
|
|---|
| 206 | }
|
|---|
| 207 | if (!Parameters.ContainsKey("Percentile")) {
|
|---|
| 208 | Parameters.Add(new ValueLookupParameter<DoubleValue>("Percentile", new DoubleValue(1)));
|
|---|
| 209 | }
|
|---|
| 210 | if (!Parameters.ContainsKey("ValidationQuality")) {
|
|---|
| 211 | Parameters.Add(new ScopeTreeLookupParameter<DoubleValue>("ValidationQuality"));
|
|---|
| 212 | }
|
|---|
| 213 | }
|
|---|
| 214 |
|
|---|
| 215 | public override IOperation Apply() {
|
|---|
| 216 | var trees = SymbolicExpressionTree;
|
|---|
| 217 | ItemArray<DoubleValue> qualities = QualityParameter.ActualValue;
|
|---|
| 218 | ItemArray<DoubleValue> validationQualities = ValidationQualityParameter.ActualValue;
|
|---|
| 219 |
|
|---|
| 220 | //string targetVariable = ProblemData.TargetVariable.Value;
|
|---|
| 221 |
|
|---|
| 222 | //// select a random subset of rows in the validation set
|
|---|
| 223 | //int validationStart = ValidiationSamplesStart.Value;
|
|---|
| 224 | //int validationEnd = ValidationSamplesEnd.Value;
|
|---|
| 225 | //int seed = Random.Next();
|
|---|
| 226 | //int count = (int)((validationEnd - validationStart) * RelativeNumberOfEvaluatedSamples.Value);
|
|---|
| 227 | //if (count == 0) count = 1;
|
|---|
| 228 | //IEnumerable<int> rows = RandomEnumerable.SampleRandomNumbers(seed, validationStart, validationEnd, count);
|
|---|
| 229 |
|
|---|
| 230 | //double upperEstimationLimit = UpperEstimationLimit != null ? UpperEstimationLimit.Value : double.PositiveInfinity;
|
|---|
| 231 | //double lowerEstimationLimit = LowerEstimationLimit != null ? LowerEstimationLimit.Value : double.NegativeInfinity;
|
|---|
| 232 |
|
|---|
| 233 | //double bestQuality = Maximization.Value ? double.NegativeInfinity : double.PositiveInfinity;
|
|---|
| 234 | //SymbolicExpressionTree bestTree = null;
|
|---|
| 235 |
|
|---|
| 236 | //List<double> validationQualities = new List<double>();
|
|---|
| 237 | //foreach (var tree in trees) {
|
|---|
| 238 | // double quality = Evaluator.Evaluate(SymbolicExpressionTreeInterpreter, tree,
|
|---|
| 239 | // lowerEstimationLimit, upperEstimationLimit,
|
|---|
| 240 | // ProblemData.Dataset, targetVariable,
|
|---|
| 241 | // rows);
|
|---|
| 242 | // validationQualities.Add(quality);
|
|---|
| 243 | // //if ((Maximization.Value && quality > bestQuality) ||
|
|---|
| 244 | // // (!Maximization.Value && quality < bestQuality)) {
|
|---|
| 245 | // // bestQuality = quality;
|
|---|
| 246 | // // bestTree = tree;
|
|---|
| 247 | // //}
|
|---|
| 248 | //}
|
|---|
| 249 |
|
|---|
| 250 | //if (RelativeValidationQualityParameter.ActualValue == null) {
|
|---|
| 251 | // first call initialize the relative quality using the difference between average training and validation quality
|
|---|
| 252 | double avgTrainingQuality = qualities.Select(x => x.Value).Average();
|
|---|
| 253 | double avgValidationQuality = validationQualities.Select(x => x.Value).Average();
|
|---|
| 254 |
|
|---|
| 255 | if (Maximization.Value)
|
|---|
| 256 | RelativeValidationQualityParameter.ActualValue = new PercentValue(avgValidationQuality / avgTrainingQuality - 1);
|
|---|
| 257 | else {
|
|---|
| 258 | RelativeValidationQualityParameter.ActualValue = new PercentValue(avgTrainingQuality / avgValidationQuality - 1);
|
|---|
| 259 | }
|
|---|
| 260 | //}
|
|---|
| 261 |
|
|---|
| 262 | // best first (only for maximization
|
|---|
| 263 | var orderedDistinctPairs = (from index in Enumerable.Range(0, qualities.Length)
|
|---|
| 264 | select new { Training = qualities[index].Value, Validation = validationQualities[index].Value })
|
|---|
| 265 | .OrderBy(x => -x.Training)
|
|---|
| 266 | .ToList();
|
|---|
| 267 |
|
|---|
| 268 | int n = (int)Math.Round(PercentileParameter.ActualValue.Value * orderedDistinctPairs.Count);
|
|---|
| 269 |
|
|---|
| 270 | double[] validationArr = new double[n];
|
|---|
| 271 | double[] trainingArr = new double[n];
|
|---|
| 272 | //double[,] qualitiesArr = new double[n, 2];
|
|---|
| 273 | for (int i = 0; i < n; i++) {
|
|---|
| 274 | validationArr[i] = orderedDistinctPairs[i].Validation;
|
|---|
| 275 | trainingArr[i] = orderedDistinctPairs[i].Training;
|
|---|
| 276 |
|
|---|
| 277 | //qualitiesArr[i, 0] = trainingArr[i];
|
|---|
| 278 | //qualitiesArr[i, 1] = validationArr[i];
|
|---|
| 279 | }
|
|---|
| 280 | double r = alglib.correlation.spearmanrankcorrelation(trainingArr, validationArr, n);
|
|---|
| 281 | TrainingValidationQualityCorrelationParameter.ActualValue = new DoubleValue(r);
|
|---|
| 282 | if (InitialTrainingQualityParameter.ActualValue == null)
|
|---|
| 283 | InitialTrainingQualityParameter.ActualValue = new DoubleValue(avgValidationQuality);
|
|---|
| 284 | bool overfitting =
|
|---|
| 285 | avgTrainingQuality > InitialTrainingQualityParameter.ActualValue.Value && // better on training than in initial generation
|
|---|
| 286 | // RelativeValidationQualityParameter.ActualValue.Value < 0.0 && // validation quality is worse than training quality
|
|---|
| 287 | r < CorrelationLimitParameter.ActualValue.Value; // low correlation between training and validation quality
|
|---|
| 288 |
|
|---|
| 289 |
|
|---|
| 290 | OverfittingParameter.ActualValue = new BoolValue(overfitting);
|
|---|
| 291 | //TrainingAndValidationQualitiesParameter.ActualValue = new DoubleMatrix(qualitiesArr);
|
|---|
| 292 | return base.Apply();
|
|---|
| 293 | }
|
|---|
| 294 |
|
|---|
| 295 | [StorableHook(HookType.AfterDeserialization)]
|
|---|
| 296 | private void Initialize() { }
|
|---|
| 297 |
|
|---|
| 298 | private static void AddValue(DataTable table, double data, string name, string description) {
|
|---|
| 299 | DataRow row;
|
|---|
| 300 | table.Rows.TryGetValue(name, out row);
|
|---|
| 301 | if (row == null) {
|
|---|
| 302 | row = new DataRow(name, description);
|
|---|
| 303 | row.Values.Add(data);
|
|---|
| 304 | table.Rows.Add(row);
|
|---|
| 305 | } else {
|
|---|
| 306 | row.Values.Add(data);
|
|---|
| 307 | }
|
|---|
| 308 | }
|
|---|
| 309 | }
|
|---|
| 310 | }
|
|---|
