private IClassificationProblemData CreateDefaultProblem(int rows, int columns)
{
List <string> allowedInputVariables = Enumerable.Range(0, columns - 1).Select(x => "x" + x.ToString()).ToList();
string targetVariable = "y";
var variableNames = allowedInputVariables.Union(targetVariable.ToEnumerable());
double[,] variableValues = new double[rows, columns];
FastRandom random = new FastRandom(12345);
int len0 = variableValues.GetLength(0);
int len1 = variableValues.GetLength(1);
for (int i = 0; i < len0; i++)
{
for (int j = 0; j < len1; j++)
{
if (j == len1 - 1)
{
variableValues[i, j] = (j + i) % 2;
}
else
{
variableValues[i, j] = random.Next(1, 100);
}
}
}
Dataset dataset = new Dataset(variableNames, variableValues);
var ret = new ClassificationProblemData(dataset, allowedInputVariables, targetVariable);
ret.SetClassName(0, "NOK");
ret.SetClassName(1, "OK");
return(ret);
}
private IClassificationProblemData CreateDefaultProblem()
{
List <string> allowedInputVariables = new List <string>()
{
"x1", "x2", "x3", "x4", "x5"
};
string targetVariable = "y";
var variableNames = allowedInputVariables.Union(targetVariable.ToEnumerable());
double[,] variableValues = new double[100, variableNames.Count()];
FastRandom random = new FastRandom(12345);
int len0 = variableValues.GetLength(0);
int len1 = variableValues.GetLength(1);
for (int i = 0; i < len0; i++)
{
for (int j = 0; j < len1; j++)
{
if (j == len1 - 1)
{
variableValues[i, j] = (j + i) % 2;
}
else
{
variableValues[i, j] = random.Next(1, 100);
}
}
}
Dataset dataset = new Dataset(variableNames, variableValues);
var ret = new ClassificationProblemData(dataset, allowedInputVariables, targetVariable);
ret.SetClassName(0, "NOK");
ret.SetClassName(1, "OK");
return(ret);
}
