public static RegressionParameter[] GridRegression(Problem problem, Problem validation, Parameter parameter, List <double> CValues, List <double> GammaValues, List <double> PValues, string outputFile, int MaxBestCount, double IgnoreVolatility)
{
if (!string.IsNullOrEmpty(outputFile))
{
File.WriteAllText(outputFile, string.Empty);
}
List <RegressionParameter> list = new List <RegressionParameter>(CValues.Count * GammaValues.Count * PValues.Count);
for (int i = 0; i < CValues.Count; i++)
{
for (int j = 0; j < GammaValues.Count; j++)
{
for (int k = 0; k < PValues.Count; k++)
{
parameter.C = CValues[i];
parameter.Gamma = GammaValues[j];
parameter.P = PValues[k];
Model model = Training.Train(problem, parameter);
double[] array = Prediction.Predict(validation, model);
double num = (IgnoreVolatility > 0.0 && (Math.Abs(array[0] - array[1]) <= IgnoreVolatility || Math.Abs(array[1] - array[2]) <= IgnoreVolatility)) ? 1.7976931348623157E+308 : Prediction.ComputeMSE(validation, array, 0);
Console.Write("{0} {1} {2} {3}", parameter.C, parameter.Gamma, parameter.P, num);
if (!string.IsNullOrEmpty(outputFile))
{
File.AppendAllText(outputFile, string.Format("{0} {1} {2} {3}\n", parameter.C, parameter.Gamma, parameter.P, num));
}
RegressionParameter regressionParameter = new RegressionParameter(parameter.C, parameter.Gamma, parameter.P, num);
regressionParameter.ValidatePredictions = array;
list.Add(regressionParameter);
}
}
}
list.Sort();
if (MaxBestCount > 0)
{
int l;
for (l = Math.Min(MaxBestCount, list.Count); l < list.Count && l < MaxBestCount * 3 && list[l - 1].SortValue == list[l].SortValue; l++)
{
}
if (list.Count > l)
{
list.RemoveRange(l, list.Count - l);
}
list.TrimExcess();
}
return(list.ToArray());
}
public static void Train(params string[] args)
{
Parameter parameters = default(Parameter);
Problem problem = default(Problem);
bool flag = default(bool);
int nrfold = default(int);
string filename = default(string);
Training.parseCommandLine(args, out parameters, out problem, out flag, out nrfold, out filename);
if (flag)
{
Training.PerformCrossValidation(problem, parameters, nrfold);
}
else
{
Model.Write(filename, Training.Train(problem, parameters));
}
}
public static void Grid(Problem problem, Problem validation, Parameter parameters, List <double> CValues, List <double> GammaValues, string outputFile, out double C, out double Gamma)
{
C = 0.0;
Gamma = 0.0;
double num = -1.7976931348623157E+308;
if (!string.IsNullOrEmpty(outputFile))
{
File.WriteAllText(outputFile, string.Empty);
}
for (int i = 0; i < CValues.Count; i++)
{
for (int j = 0; j < GammaValues.Count; j++)
{
parameters.C = CValues[i];
parameters.Gamma = GammaValues[j];
Model model = Training.Train(problem, parameters);
double num2 = Prediction.Predict(validation, "tmp.txt", model, false, 1);
Console.Write("{0} {1} {2}", parameters.C, parameters.Gamma, num2);
if (!string.IsNullOrEmpty(outputFile))
{
File.AppendAllText(outputFile, string.Format("{0} {1} {2}\n", parameters.C, parameters.Gamma, num2));
}
if (num2 > num)
{
C = parameters.C;
Gamma = parameters.Gamma;
num = num2;
Console.WriteLine(" New Maximum!");
}
else
{
Console.WriteLine();
}
}
}
}