public void getLrModel(string modelPath, bool BuildModel = false)
{
using (System.IO.StreamReader sr = new System.IO.StreamReader(modelPath))
{
string mType = sr.ReadLine();
modelTypes m = (modelTypes)Enum.Parse(typeof(modelTypes), mType);
if (m != modelTypes.LogisticRegression)
{
System.Windows.Forms.MessageBox.Show("Model file specified is not a Logistic Regression Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
InTablePath = sr.ReadLine();
IndependentFieldNames = sr.ReadLine().Split(new char[] { ',' });
DependentFieldNames = sr.ReadLine().Split(new char[] { ',' });
ClassFieldNames = sr.ReadLine().Split(new char[] { ',' });
categories = sr.ReadLine().Split(new char[] { ',' });
nvars = System.Convert.ToInt32(sr.ReadLine());
ncat = System.Convert.ToInt32(sr.ReadLine());
loglikelihood = System.Convert.ToDouble(sr.ReadLine());
deviance = System.Convert.ToDouble(sr.ReadLine());
x2 = System.Convert.ToDouble(sr.ReadLine());
pv = System.Convert.ToDouble(sr.ReadLine());
minValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
maxValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
sumValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
coefficients = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
standarderror = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
waldstat = new Accord.Statistics.Testing.WaldTest[coefficients.Length];
double[] waldStatVl = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
double[] waldStatPVl = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
for (int i = 0; i < waldStatVl.Length; i++)
{
double wVl = waldStatVl[i];
Accord.Statistics.Testing.WaldTest wt = new Accord.Statistics.Testing.WaldTest(wVl);
waldstat[i] = wt;
}
sr.Close();
if (BuildModel)
{
buildModel();
}
}
}
private MultinomialLogisticRegression buildModel()
{
if (independent == null)
{
formatData();
}
mlr = new MultinomialLogisticRegression(nvars, ncat);
LowerBoundNewtonRaphson lbn = new LowerBoundNewtonRaphson(mlr);
do
{
delta = lbn.Run(independent, dependent);
iteration++;
} while (iteration < totit && delta > converg);
coefficients = mlr.Coefficients;
standarderror = new double[ncat - 1][];
waldstat = new double[ncat - 1][];
waldpvalue = new double[ncat - 1][];
for (int i = 0; i < coefficients.Length; i++)
{
double[] steArr = new double[nvars + 1];
double[] waldStatArr = new double[nvars + 1];
double[] waldPvalueArr = new double[nvars + 1];
for (int j = 0; j < nvars + 1; j++)
{
Accord.Statistics.Testing.WaldTest wt = mlr.GetWaldTest(i, j);
steArr[j] = wt.StandardError;
waldStatArr[j] = wt.Statistic;
waldPvalueArr[j] = wt.PValue;
}
waldstat[i] = waldStatArr;
waldpvalue[i] = waldPvalueArr;
standarderror[i] = steArr;
}
loglikelihood = mlr.GetLogLikelihood(independent, dependent);
deviance = mlr.GetDeviance(independent, dependent);
x2 = mlr.ChiSquare(independent, dependent).Statistic;
pv = mlr.ChiSquare(independent, dependent).PValue;
return(mlr);
}
private void buildModel()
{
if (independentVls == null)
{
getMatrix();
}
Accord.Statistics.Links.ILinkFunction lFunc = new Accord.Statistics.Links.IdentityLinkFunction();
switch (Link)
{
case LinkFunction.Absolute:
lFunc = new Accord.Statistics.Links.AbsoluteLinkFunction();
break;
case LinkFunction.Cauchit:
lFunc = new Accord.Statistics.Links.CauchitLinkFunction();
break;
case LinkFunction.Inverse:
lFunc = new Accord.Statistics.Links.InverseLinkFunction();
break;
case LinkFunction.InverseSquared:
lFunc = new Accord.Statistics.Links.InverseSquaredLinkFunction();
break;
case LinkFunction.Logit:
lFunc = new Accord.Statistics.Links.LogitLinkFunction();
break;
case LinkFunction.Log:
lFunc = new Accord.Statistics.Links.LogLinkFunction();
break;
case LinkFunction.LogLog:
lFunc = new Accord.Statistics.Links.LogLogLinkFunction();
break;
case LinkFunction.Probit:
lFunc = new Accord.Statistics.Links.ProbitLinkFunction();
break;
case LinkFunction.Sin:
lFunc = new Accord.Statistics.Links.SinLinkFunction();
break;
case LinkFunction.Threshold:
lFunc = new Accord.Statistics.Links.ThresholdLinkFunction();
break;
default:
break;
}
glm = new Accord.Statistics.Models.Regression.GeneralizedLinearRegression(lFunc, NumberOfVariables);
Accord.Statistics.Models.Regression.Fitting.IterativeReweightedLeastSquares isl = new Accord.Statistics.Models.Regression.Fitting.IterativeReweightedLeastSquares(glm);
delta = 0;
do
{
delta = isl.Run(independentVls, dependentVls);
iterations += 1;
}while (delta > converge && iterations < totiterations);
coefficients = glm.Coefficients;
stdError = glm.StandardErrors;
loglikelihood = glm.GetLogLikelihood(independentVls, dependentVls);
if (Double.IsNaN(loglikelihood))
{
loglikelihood = 0;
}
loglikelihoodratio = glm.GetLogLikelihoodRatio(independentVls, dependentVls, glm);
if (Double.IsNaN(loglikelihoodratio))
{
loglikelihoodratio = 0;
}
deviance = glm.GetDeviance(independentVls, dependentVls);
if (Double.IsNaN(deviance))
{
deviance = 0;
}
Accord.Statistics.Testing.ChiSquareTest chiTest = glm.ChiSquare(independentVls, dependentVls);
pv = chiTest.PValue;
if (Double.IsNaN(pv))
{
pv = 0;
}
chisqr = chiTest.Statistic;
if (double.IsNaN(chisqr))
{
chisqr = 0;
}
waldTestValues = new double[IndependentFieldNames.Length];
waldTestPValues = new double[waldTestValues.Length];
for (int i = 0; i < waldTestValues.Length; i++)
{
Accord.Statistics.Testing.WaldTest wTest = glm.GetWaldTest(i);
double wS = wTest.Statistic;
double wP = wTest.PValue;
//if (Double.IsNaN(wS)) wS = 0;
//if (Double.IsNaN(wP)) wP = 0;
waldTestValues[i] = wS;
waldTestPValues[i] = wP;
}
//if (stdError.Length != coefficients.Length) stdError = new double[coefficients.Length];
//for (int i = 0; i < stdError.Length; i++)
//{
// double vl = stdError[i];
// if (Double.IsNaN(vl)) stdError[i] = 0;
//}
}
public void getLrModel(string modelPath,bool BuildModel=false)
{
using (System.IO.StreamReader sr = new System.IO.StreamReader(modelPath))
{
string mType = sr.ReadLine();
modelTypes m = (modelTypes)Enum.Parse(typeof(modelTypes), mType);
if (m != modelTypes.LogisticRegression)
{
System.Windows.Forms.MessageBox.Show("Model file specified is not a Logistic Regression Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
InTablePath = sr.ReadLine();
IndependentFieldNames = sr.ReadLine().Split(new char[] { ',' });
DependentFieldNames = sr.ReadLine().Split(new char[] { ',' });
ClassFieldNames = sr.ReadLine().Split(new char[] { ',' });
categories = sr.ReadLine().Split(new char[] { ',' });
nvars = System.Convert.ToInt32(sr.ReadLine());
ncat = System.Convert.ToInt32(sr.ReadLine());
loglikelihood = System.Convert.ToDouble(sr.ReadLine());
deviance = System.Convert.ToDouble(sr.ReadLine());
x2 = System.Convert.ToDouble(sr.ReadLine());
pv = System.Convert.ToDouble(sr.ReadLine());
minValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
maxValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
sumValues = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
coefficients = (from string s in (sr.ReadLine().Split(new char[]{','})) select System.Convert.ToDouble(s)).ToArray();
standarderror = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
waldstat = new Accord.Statistics.Testing.WaldTest[coefficients.Length];
double[] waldStatVl = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
double[] waldStatPVl = (from string s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
for (int i = 0; i < waldStatVl.Length; i++)
{
double wVl = waldStatVl[i];
Accord.Statistics.Testing.WaldTest wt = new Accord.Statistics.Testing.WaldTest(wVl);
waldstat[i] = wt;
}
sr.Close();
if (BuildModel) buildModel();
}
}
