public void writeModel(string outPath)
{
if (kmeans == null)
{
buildModel();
}
outmodelpath = outPath;
using (System.IO.StreamWriter sw = new System.IO.StreamWriter(outmodelpath))
{
sw.WriteLine(dataPrepBase.modelTypes.StrataCovCorr.ToString());
sw.WriteLine(InPath);
sw.WriteLine(String.Join(",", VariableFieldNames));
sw.WriteLine(n.ToString());
sw.WriteLine(prop.ToString());
sw.WriteLine(k.ToString());
sw.WriteLine(String.Join(",", lbl.ToArray()));
KMeansClusterCollection gCol = kmeans.Clusters;
for (int i = 0; i < gCol.Count; i++)
{
KMeansCluster gClust = gCol[i];
sw.WriteLine(String.Join(",", (from double d in gClust.Mean select d.ToString()).ToArray()));
sw.WriteLine(String.Join(",", (from double d in gClust.Covariance select d.ToString()).ToArray()));
sw.WriteLine(gClust.Proportion.ToString());
}
sw.Close();
}
}
public byte[] Denoise(byte[] data, int width, int height, int padding, out int it)
{
it = 0;
byte[,] arr = Utils.ByteArr1DTo2D(data, width, height);
bool[,] vi = (bool[, ])Array.CreateInstance(typeof(bool), height, width);
for (int i = 0; i < height; ++i)
{
for (int j = 0; j < width; ++j)
{
vi[i, j] = false;
}
}
List <Point> points = new List <Point>();
List <int> areas = new List <int>();
for (int i = padding; i < height - padding; ++i)
{
for (int j = padding; j < width - padding; ++j)
{
if (vi[i, j])
{
continue;
}
if (arr[i, j] == backgroundColor)
{
continue;
}
int sum = 0;
dfsSearch(arr, vi, i, j, width, height, padding, ref sum);
points.Add(new Point(j, i));
areas.Add(sum);
}
}
if (points.Count > 1)
{
Vector2[] vs = areas.Select(i => new Vector2(i, 0)).ToArray();
it = 0;
KMeansCluster result = KMeansClustering.AnnealCluster(vs, 2, out it);
int bigcluster = (result[0].Centroid.ModSqr() > result[1].Centroid.ModSqr()) ? 0 : 1;
for (int i = 0; i < points.Count; ++i)
{
if (result.ClusterIndex[i] != bigcluster)
{
//dfsColor(arr, points[i].Y, points[i].X, width, height, 1);
dfsClear(arr, points[i].Y, points[i].X, width, height, 1);
}
}
}
else
{
dfsColor(arr, points[0].Y, points[0].X, width, height, 1);
}
byte[] bytes = Utils.ByteArr2DTo1D(arr);
return(bytes);
}
private void setKMeansCluster(System.IO.StreamReader sr)
{
KMeans kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
string[] covVlsStr = sr.ReadLine().Split(new char[] { ',' });
double p = System.Convert.ToDouble(sr.ReadLine());
double[,] cov = new double[VariableFieldNames.Length, VariableFieldNames.Length];
for (int j = 0; j < VariableFieldNames.Length; j++)
{
for (int l = 0; l < VariableFieldNames.Length; l++)
{
int indexVl = (j * VariableFieldNames.Length) + l;
cov[l, j] = System.Convert.ToDouble(covVlsStr[indexVl]);
}
}
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = cov;
kc.Proportion = p;
}
clusterCollection = kmeansColl;
model = kmeans;
}
public static void CompareStratMeansVar(KMeans km1, KMeans km2, out double[] meanPvalues, out double[] varPvalues)
{
meanPvalues = null;
varPvalues = null;
int nPv1 = km1.Clusters.Count;
int nPv2 = km2.Clusters.Count;
if (nPv1 != nPv2)
{
System.Windows.Forms.MessageBox.Show("Not the same number of strata! Models are not comparable!");
return;
}
meanPvalues = new double[nPv1];
varPvalues = new double[nPv2];
for (int i = 0; i < nPv1; i++)
{
KMeansCluster kmC1 = km1.Clusters[i];
KMeansCluster kmC2 = km2.Clusters[i];
double[] means1 = kmC1.Mean;
double[] means2 = kmC2.Mean;
double[,] cov1 = kmC1.Covariance;
double[,] cov2 = kmC2.Covariance;
double m, v;
PairedTTestPValues(means1, cov1, means2, cov2, out m, out v);
meanPvalues[i] = m;
varPvalues[i] = v;
}
}
private void kmeansReport(Forms.RunningProcess.frmRunningProcessDialog rd)
{
KMeansClusterCollection gCol = kmeans.Clusters;
for (int i = 0; i < gCol.Count; i++)
{
KMeansCluster gClust = gCol[i];
double[] mns = gClust.Mean;
double[,] cov = gClust.Covariance;
double[,] corr = getCorr(cov);
rd.addMessage("\n\nStratum " + Labels[i] + ":\nMeans: " + String.Join(", ", (from double d in mns select d.ToString()).ToArray()) + "\nCovariance:");
for (int j = 0; j < VariableFieldNames.Length; j++)
{
string[] covStrArr = new string[VariableFieldNames.Length];
for (int l = 0; l < covStrArr.Length; l++)
{
covStrArr[l] = cov[l, j].ToString();
}
rd.addMessage("\n" + String.Join(",", covStrArr));
}
rd.addMessage("\nCorr:");
for (int j = 0; j < VariableFieldNames.Length; j++)
{
string[] corrStrArr = new string[VariableFieldNames.Length];
for (int l = 0; l < corrStrArr.Length; l++)
{
corrStrArr[l] = corr[l, j].ToString();
}
rd.addMessage("\n" + String.Join(",", corrStrArr));
}
}
}
public static void CompareStratMeansVar(string StratModel1, string StratModel2, out List <string> labels, out double[] meanDiff, out double[] varDiff, out double[] meanPvalues, out double[] varPvalues)
{
meanPvalues = null;
varPvalues = null;
meanDiff = null;
varDiff = null;
labels = null;
dataPrepClusterKmean dpc1 = new dataPrepClusterKmean();
dpc1.buildModel(StratModel1);
KMeans km1 = (KMeans)dpc1.Model;
dataPrepClusterKmean dpc2 = new dataPrepClusterKmean();
dpc2.buildModel(StratModel2);
List <string> labels2 = dpc2.Labels;
KMeans km2 = (KMeans)dpc2.Model;
int nPv1 = km1.Clusters.Count;
int nPv2 = km2.Clusters.Count;
if (nPv1 != nPv2)
{
System.Windows.Forms.MessageBox.Show("Not the same number of strata! Models are not comparable!");
return;
}
labels = dpc1.Labels;
meanPvalues = new double[nPv1];
varPvalues = new double[nPv2];
meanDiff = new double[nPv1];
varDiff = new double[nPv2];
foreach (string l in labels)
{
int ind1 = labels.IndexOf(l);
int ind2 = labels2.IndexOf(l);
KMeansCluster kmC1 = km1.Clusters[ind1];
KMeansCluster kmC2 = km2.Clusters[ind2];
double[] means1 = kmC1.Mean;
double[] means2 = kmC2.Mean;
double[,] cov1 = kmC1.Covariance;
double[,] cov2 = kmC2.Covariance;
double[] meanDiffArr = new double[means1.Length];
double[] varDiffArr = new double[means1.Length];
for (int i = 0; i < means1.Length; i++)
{
meanDiffArr[i] = means1[i] - means2[i];
varDiffArr[i] = cov1[i, i] - cov2[i, i];
}
meanDiff[ind1] = meanDiffArr.Average();
varDiff[ind1] = varDiffArr.Average();
double m, v;
PairedTTestPValues(means1, cov1, means2, cov2, out m, out v);
meanPvalues[ind1] = m;
varPvalues[ind1] = v;
}
}
private void binaryReport(Forms.RunningProcess.frmRunningProcessDialog rd)
{
KMeansClusterCollection gCol = (KMeansClusterCollection)clusterCollection;
for (int i = 0; i < gCol.Count; i++)
{
KMeansCluster gClust = gCol[i];
double[] mns = gClust.Mean;
rd.addMessage("\n\nCluster " + Labels[i] + ":\nMeans: " + String.Join(", ", (from double d in mns select d.ToString()).ToArray()));
}
}
private void writeKmeansData(System.IO.StreamWriter sw)
{
KMeansClusterCollection gCol = (KMeansClusterCollection)clusterCollection;
for (int i = 0; i < gCol.Count; i++)
{
KMeansCluster gClust = gCol[i];
sw.WriteLine(String.Join(",", (from double d in gClust.Mean select d.ToString()).ToArray()));
sw.WriteLine(String.Join(",", (from double d in gClust.Covariance select d.ToString()).ToArray()));
sw.WriteLine(gClust.Proportion.ToString());
}
}
public double[] computeNew(object category)
{
string cat = category.ToString();
int catIndex = lbl.IndexOf(cat);
int np = ((VariableFieldNames.Length * VariableFieldNames.Length) - VariableFieldNames.Length) / 2;
double[] pValues = new double[np];
if (catIndex == -1)
{
return(pValues);
}
KMeansClusterCollection gCol = kmeans.Clusters;
KMeansCluster gClust = gCol[catIndex];
double[] mns = gClust.Mean;
double[] var = new double[mns.Length];
double nSample = gClust.Proportion * N;
double seAdjust = Math.Sqrt(2 * 1 / nSample);
for (int j = 0; j < mns.Length; j++)
{
var[j] = gClust.Covariance[j, j];
}
int cnt = 1;
int pCnt = 0;
for (int j = 0; j < mns.Length - 1; j++)
{
for (int k = cnt; k < mns.Length; k++)
{
double mD = mns[j] - mns[k];
double pSD = Math.Sqrt((var[j] + var[k]) / 2);
double se = pSD * seAdjust;
double tStat = mD / se;
Accord.Statistics.Distributions.Univariate.TDistribution tDist = new Accord.Statistics.Distributions.Univariate.TDistribution(2 * nSample - 2);
double cdf = tDist.DistributionFunction(tStat);
double pValue = 0;
if (tStat > 0)
{
pValue = (1 - cdf) * 2;
}
else
{
pValue = (cdf * 2);
}
pValues[pCnt] = pValue;
pCnt++;
}
cnt += 1;
}
return(pValues);
}
private void makeKMeans()
{
kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = meansLst[i];
double p = proportionsLst[i];
double[,] scov = scovLst[i];
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = scov;
kc.Proportion = p;
}
}
private void setBinaryCluster(System.IO.StreamReader sr)
{
BinarySplit bSplit = new BinarySplit(k);
KMeansClusterCollection kmeansColl = bSplit.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
sr.ReadLine();
double p = System.Convert.ToDouble(sr.ReadLine());
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Proportion = p;
}
clusterCollection = kmeansColl;
model = bSplit;
}
public void buildModel(string modelPath)
{
outmodelpath = modelPath;
using (System.IO.StreamReader sr = new System.IO.StreamReader(outmodelpath))
{
dataPrepBase.modelTypes mType = (dataPrepBase.modelTypes)Enum.Parse(typeof(dataPrepBase.modelTypes), sr.ReadLine());
if (mType != dataPrepBase.modelTypes.TTEST)
{
System.Windows.Forms.MessageBox.Show("Not a TTest Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
inpath = sr.ReadLine();
stratafld = sr.ReadLine();
VariableFieldNames = sr.ReadLine().Split(new char[] { ',' });
n = System.Convert.ToInt32(sr.ReadLine());
prop = System.Convert.ToDouble(sr.ReadLine());
k = System.Convert.ToInt32(sr.ReadLine());
lbl = sr.ReadLine().Split(new char[] { ',' }).ToList();
kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
string[] covVlsStr = sr.ReadLine().Split(new char[] { ',' });
double p = System.Convert.ToDouble(sr.ReadLine());
double[,] cov = new double[VariableFieldNames.Length, VariableFieldNames.Length];
for (int j = 0; j < VariableFieldNames.Length; j++)
{
for (int l = 0; l < VariableFieldNames.Length; l++)
{
int indexVl = (j * VariableFieldNames.Length) + l;
cov[l, j] = System.Convert.ToDouble(covVlsStr[indexVl]);
}
}
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = cov;
kc.Proportion = p;
}
sr.Close();
}
}
public void getReport()
{
if (kmeans == null)
{
buildModel();
}
Forms.RunningProcess.frmRunningProcessDialog rd = new Forms.RunningProcess.frmRunningProcessDialog(false);
rd.Text = "Cluster Results";
rd.TopLevel = true;
rd.pgbProcess.Visible = false;
rd.FormBorderStyle = System.Windows.Forms.FormBorderStyle.Sizable;
rd.addMessage("Input path = " + InPath);
rd.addMessage("Sample size = " + n.ToString() + " proportion of total records = " + prop.ToString());
rd.addMessage("Number of Cluster = " + k.ToString());
rd.addMessage("Labels = " + String.Join(", ", lbl.ToArray()));
rd.addMessage("Variables: " + String.Join(" ,", VariableFieldNames));
KMeansClusterCollection gCol = kmeans.Clusters;
for (int i = 0; i < gCol.Count; i++)
{
KMeansCluster gClust = gCol[i];
double[] mns = gClust.Mean;
double[,] cov = gClust.Covariance;
rd.addMessage("\n\nCluster " + Labels[i] + ":\nMeans: " + String.Join(", ", (from double d in mns select d.ToString()).ToArray()) + "\nCovariance:");
for (int j = 0; j < VariableFieldNames.Length; j++)
{
string[] covStrArr = new string[VariableFieldNames.Length];
for (int l = 0; l < covStrArr.Length; l++)
{
covStrArr[l] = cov[l, j].ToString();
}
rd.addMessage("\n" + String.Join(",", covStrArr));
}
}
rd.enableClose();
rd.Show();
}
private void setBinaryCluster(System.IO.StreamReader sr)
{
BinarySplit bSplit = new BinarySplit(k);
KMeansClusterCollection kmeansColl = bSplit.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
sr.ReadLine();
double p = System.Convert.ToDouble(sr.ReadLine());
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Proportion = p;
}
clusterCollection = kmeansColl;
model = bSplit;
}
public void getReport()
{
if (kmeans == null)
{
buildModel();
}
rd = new Forms.RunningProcess.frmRunningProcessDialog(false);
rd.Text = "Independent T-Test Results";
rd.TopLevel = true;
rd.pgbProcess.Visible = false;
rd.FormBorderStyle = System.Windows.Forms.FormBorderStyle.Sizable;
rd.addMessage("Input path = " + InPath);
rd.addMessage("Total Sample size = " + n.ToString());
rd.addMessage("\nLabel |Compare V1-V2 |N |Dif |T-Stat |P-Value ");
rd.addMessage("-".PadRight(83, '-'));
KMeansClusterCollection gCol = kmeans.Clusters;
for (int i = 0; i < gCol.Count; i++)
{
string l = getValue(Labels[i], 8);
KMeansCluster gClust = gCol[i];
double[] mns = gClust.Mean;
double[] var = new double[mns.Length];
double nSample = gClust.Proportion * N;
if (nSample <= 1)
{
continue;
}
double seAdjust = Math.Sqrt(2 * 1 / nSample);
for (int j = 0; j < mns.Length; j++)
{
var[j] = gClust.Covariance[j, j];
}
int cnt = 1;
for (int j = 0; j < mns.Length - 1; j++)
{
for (int k = cnt; k < mns.Length; k++)
{
string fN1 = getValue(VariableFieldNames[j], 12);
string fN2 = getValue(VariableFieldNames[k], 12);
double mD = mns[j] - mns[k];
double pSD = Math.Sqrt((var[j] + var[k]) / 2);
double se = pSD * seAdjust;
double tStat = mD / se;
Accord.Statistics.Distributions.Univariate.TDistribution tDist = new Accord.Statistics.Distributions.Univariate.TDistribution(2 * nSample - 2);
double cdf = tDist.DistributionFunction(tStat);
double pValue = 0;
if (tStat > 0)
{
pValue = (1 - cdf) * 2;
}
else
{
pValue = (cdf * 2);
}
string pValueS = pValue.ToString();
if (pValue < 0.0001)
{
pValueS = "p < 0.0001";
}
string ln = l + "|" + fN1 + "-" + fN2 + "| " + getValue(nSample.ToString(), 6) + " | " + getValue(mD.ToString(), 6) + " | " + getValue(tStat.ToString(), 6) + " | " + getValue(pValueS, 10);
rd.addMessage(ln);
}
cnt += 1;
}
}
rd.addMessage("-".PadRight(83, '-'));
rd.enableClose();
rd.Show();
}
private void setKMeansCluster(System.IO.StreamReader sr)
{
KMeans kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
string[] covVlsStr = sr.ReadLine().Split(new char[] { ',' });
double p = System.Convert.ToDouble(sr.ReadLine());
double[,] cov = new double[VariableFieldNames.Length, VariableFieldNames.Length];
for (int j = 0; j < VariableFieldNames.Length; j++)
{
for (int l = 0; l < VariableFieldNames.Length; l++)
{
int indexVl = (j * VariableFieldNames.Length) + l;
cov[l, j] = System.Convert.ToDouble(covVlsStr[indexVl]);
}
}
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = cov;
kc.Proportion = p;
}
clusterCollection = kmeansColl;
model = kmeans;
}
public void buildModel(string modelPath)
{
outmodelpath = modelPath;
using (System.IO.StreamReader sr = new System.IO.StreamReader(outmodelpath))
{
dataPrepBase.modelTypes mType = (dataPrepBase.modelTypes)Enum.Parse(typeof(dataPrepBase.modelTypes), sr.ReadLine());
if (mType != dataPrepBase.modelTypes.StrataCovCorr)
{
System.Windows.Forms.MessageBox.Show("Not a StrataCovCorr Model!!", "Error", System.Windows.Forms.MessageBoxButtons.OK, System.Windows.Forms.MessageBoxIcon.Error);
return;
}
inpath = sr.ReadLine();
VariableFieldNames = sr.ReadLine().Split(new char[] { ',' });
n = System.Convert.ToInt32(sr.ReadLine());
prop = System.Convert.ToDouble(sr.ReadLine());
k = System.Convert.ToInt32(sr.ReadLine());
lbl = sr.ReadLine().Split(new char[] { ',' }).ToList();
kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = (from s in (sr.ReadLine().Split(new char[] { ',' })) select System.Convert.ToDouble(s)).ToArray();
string[] covVlsStr = sr.ReadLine().Split(new char[] { ',' });
double p = System.Convert.ToDouble(sr.ReadLine());
double[,] cov = new double[VariableFieldNames.Length, VariableFieldNames.Length];
for (int j = 0; j < VariableFieldNames.Length; j++)
{
for (int l = 0; l < VariableFieldNames.Length; l++)
{
int indexVl = (j * VariableFieldNames.Length) + l;
cov[l, j] = System.Convert.ToDouble(covVlsStr[indexVl]);
}
}
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = cov;
kc.Proportion = p;
}
sr.Close();
}
}
private void makeKMeans()
{
kmeans = new KMeans(k);
KMeansClusterCollection kmeansColl = kmeans.Clusters;
for (int i = 0; i < k; i++)
{
double[] mns = meansLst[i];
double p = proportionsLst[i];
double[,] scov = scovLst[i];
KMeansCluster kc = new KMeansCluster(kmeansColl, i);
kc.Mean = mns;
kc.Covariance = scov;
kc.Proportion = p;
}
}
