Changeset 14237 for branches/symbreg-factors-2650/HeuristicLab.Algorithms.DataAnalysis/3.4/Linear/AlglibUtil.cs
- Timestamp:
- 08/05/16 14:25:28 (8 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
branches/symbreg-factors-2650/HeuristicLab.Algorithms.DataAnalysis/3.4/Linear/AlglibUtil.cs
r14185 r14237 20 20 #endregion 21 21 22 using System; 22 23 using System.Collections.Generic; 23 24 using System.Linq; … … 27 28 public static class AlglibUtil { 28 29 public static double[,] PrepareInputMatrix(IDataset dataset, IEnumerable<string> variables, IEnumerable<int> rows) { 29 List<string> variablesList = variables.ToList(); 30 // check input variables. Only double variables are allowed. 31 var invalidInputs = 32 variables.Where(name => !dataset.VariableHasType<double>(name)); 33 if (invalidInputs.Any()) 34 throw new NotSupportedException("Unsupported inputs: " + string.Join(", ", invalidInputs)); 35 30 36 List<int> rowsList = rows.ToList(); 31 32 double[,] matrix = new double[rowsList.Count, variablesList.Count]; 37 double[,] matrix = new double[rowsList.Count, variables.Count()]; 33 38 34 39 int col = 0; … … 45 50 return matrix; 46 51 } 52 47 53 public static double[,] PrepareAndScaleInputMatrix(IDataset dataset, IEnumerable<string> variables, IEnumerable<int> rows, Scaling scaling) { 54 // check input variables. Only double variables are allowed. 55 var invalidInputs = 56 variables.Where(name => !dataset.VariableHasType<double>(name)); 57 if (invalidInputs.Any()) 58 throw new NotSupportedException("Unsupported inputs: " + string.Join(", ", invalidInputs)); 59 48 60 List<string> variablesList = variables.ToList(); 49 61 List<int> rowsList = rows.ToList(); … … 64 76 return matrix; 65 77 } 78 79 /// <summary> 80 /// Prepares a binary data matrix from a number of factors and specified factor values 81 /// </summary> 82 /// <param name="dataset">A dataset that contains the variable values</param> 83 /// <param name="factorVariables">An enumerable of categorical variables (factors). For each variable an enumerable of values must be specified.</param> 84 /// <param name="rows">An enumerable of row indices for the dataset</param> 85 /// <returns></returns> 86 /// <remarks>Factor variables (categorical variables) are split up into multiple binary variables one for each specified value.</remarks> 87 public static double[,] PrepareInputMatrix( 88 IDataset dataset, 89 IEnumerable<KeyValuePair<string, IEnumerable<string>>> factorVariables, 90 IEnumerable<int> rows) { 91 // check input variables. Only string variables are allowed. 92 var invalidInputs = 93 factorVariables.Select(kvp => kvp.Key).Where(name => !dataset.VariableHasType<string>(name)); 94 if (invalidInputs.Any()) 95 throw new NotSupportedException("Unsupported inputs: " + string.Join(", ", invalidInputs)); 96 97 int numBinaryColumns = factorVariables.Sum(kvp => kvp.Value.Count()); 98 99 List<int> rowsList = rows.ToList(); 100 double[,] matrix = new double[rowsList.Count, numBinaryColumns]; 101 102 int col = 0; 103 foreach (var kvp in factorVariables) { 104 var varName = kvp.Key; 105 var cats = kvp.Value; 106 var catCount = cats.Count(); 107 if (catCount == 0) continue; 108 foreach (var cat in cats) { 109 var values = dataset.GetStringValues(varName, rows); 110 int row = 0; 111 foreach (var value in values) { 112 matrix[row, col] = value == cat ? 1 : 0; 113 row++; 114 } 115 col++; 116 } 117 } 118 return matrix; 119 } 66 120 } 67 121 }
Note: See TracChangeset
for help on using the changeset viewer.