public static double SvmGetSvrProbability(SvmModel model)
{
if ((model.param.svmType == SvmType.EpsilonSvr || model.param.svmType == SvmType.NuSvr) && model.probA != null)
{
return(model.probA[0]);
}
Info("Model doesn't contain information for SVR probability inference\n");
return(0);
}
public static int SvmCheckProbabilityModel(SvmModel model)
{
if (((model.param.svmType == SvmType.CSvc || model.param.svmType == SvmType.NuSvc) && model.probA != null &&
model.probB != null) ||
((model.param.svmType == SvmType.EpsilonSvr || model.param.svmType == SvmType.NuSvr) && model.probA != null))
{
return(1);
}
return(0);
}
public static void SvmGetLabels(SvmModel model, int[] label)
{
if (model.label != null)
{
for (int i = 0; i < model.nrClass; i++)
{
label[i] = model.label[i];
}
}
}
public static float SvmPredict(SvmModel model, BaseVector x)
{
int nrClass = model.nrClass;
double[] decValues;
if (model.param.svmType == SvmType.OneClass || model.param.svmType == SvmType.EpsilonSvr ||
model.param.svmType == SvmType.NuSvr)
{
decValues = new double[1];
}
else
{
decValues = new double[nrClass * (nrClass - 1) / 2];
}
float predResult = SvmPredictValues(model, x, decValues);
return(predResult);
}
public override ClassificationModel Train(BaseVector[] x, int[][] y, int ngroups, Parameters param, int nthreads,
Action<double> reportProgress)
{
string err = CheckInput(x, y, ngroups);
if (err != null){
throw new Exception(err);
}
ParameterWithSubParams<int> kernelParam = param.GetParamWithSubParams<int>("Kernel");
SvmParameter sp = new SvmParameter{
kernelFunction = KernelFunctions.GetKernelFunction(kernelParam.Value, kernelParam.GetSubParameters()),
svmType = SvmType.CSvc,
c = param.GetParam<double>("C").Value
};
bool[] invert;
SvmProblem[] problems = CreateProblems(x, y, ngroups, out invert);
SvmModel[] models = new SvmModel[problems.Length];
ThreadDistributor td = new ThreadDistributor(nthreads, models.Length,
i => { models[i] = SvmMain.SvmTrain(problems[i], sp); }, fractionDone => { reportProgress?.Invoke(fractionDone); });
td.Start();
return new SvmClassificationModel(models, invert);
}
public static float SvmPredictProbability(SvmModel model, BaseVector x, double[] probEstimates)
{
if ((model.param.svmType == SvmType.CSvc || model.param.svmType == SvmType.NuSvc) && model.probA != null &&
model.probB != null)
{
int nrClass = model.nrClass;
double[] decValues = new double[nrClass * (nrClass - 1) / 2];
SvmPredictValues(model, x, decValues);
const double minProb = 1e-7;
double[][] pairwiseProb = new double[nrClass][];
for (int m = 0; m < nrClass; m++)
{
pairwiseProb[m] = new double[nrClass];
}
int k = 0;
for (int i = 0; i < nrClass; i++)
{
for (int j = i + 1; j < nrClass; j++)
{
pairwiseProb[i][j] = Math.Min(Math.Max(SigmoidPredict(decValues[k], model.probA[k], model.probB[k]), minProb),
1 - minProb);
pairwiseProb[j][i] = 1 - pairwiseProb[i][j];
k++;
}
}
MulticlassProbability(nrClass, pairwiseProb, probEstimates);
int probMaxIdx = 0;
for (int i = 1; i < nrClass; i++)
{
if (probEstimates[i] > probEstimates[probMaxIdx])
{
probMaxIdx = i;
}
}
return(model.label[probMaxIdx]);
}
return(SvmPredict(model, x));
}
// Cross-validation decision values for probability estimates
internal static void SvmBinarySvcProbability(SvmProblem prob, SvmParameter param, double cp, double cn,
IList <double> probAb)
{
int i;
const int nrFold = 5;
int[] perm = new int[prob.Count];
double[] decValues = new double[prob.Count];
// random shuffle
for (i = 0; i < prob.Count; i++)
{
perm[i] = i;
}
for (i = 0; i < prob.Count; i++)
{
int j = i + rand.Next(prob.Count - i);
do
{
int _ = perm[i];
perm[i] = perm[j];
perm[j] = _;
} while (false);
}
for (i = 0; i < nrFold; i++)
{
int begin = i * prob.Count / nrFold;
int end = (i + 1) * prob.Count / nrFold;
int j;
int count = prob.Count - (end - begin);
SvmProblem subprob = new SvmProblem {
x = new BaseVector[count], y = new float[count]
};
int k = 0;
for (j = 0; j < begin; j++)
{
subprob.x[k] = prob.x[perm[j]];
subprob.y[k] = prob.y[perm[j]];
++k;
}
for (j = end; j < prob.Count; j++)
{
subprob.x[k] = prob.x[perm[j]];
subprob.y[k] = prob.y[perm[j]];
++k;
}
int pCount = 0, nCount = 0;
for (j = 0; j < k; j++)
{
if (subprob.y[j] > 0)
{
pCount++;
}
else
{
nCount++;
}
}
if (pCount == 0 && nCount == 0)
{
for (j = begin; j < end; j++)
{
decValues[perm[j]] = 0;
}
}
else if (pCount > 0 && nCount == 0)
{
for (j = begin; j < end; j++)
{
decValues[perm[j]] = 1;
}
}
else if (pCount == 0 && nCount > 0)
{
for (j = begin; j < end; j++)
{
decValues[perm[j]] = -1;
}
}
else
{
SvmParameter subparam = (SvmParameter)param.Clone();
subparam.probability = false;
subparam.c = 1.0;
subparam.nrWeight = 2;
subparam.weightLabel = new int[2];
subparam.weight = new double[2];
subparam.weightLabel[0] = +1;
subparam.weightLabel[1] = -1;
subparam.weight[0] = cp;
subparam.weight[1] = cn;
SvmModel submodel = SvmTrain(subprob, subparam);
for (j = begin; j < end; j++)
{
double[] decValue = new double[1];
SvmPredictValues(submodel, prob.x[perm[j]], decValue);
decValues[perm[j]] = decValue[0];
// ensure +1 -1 order; reason not using CV subroutine
decValues[perm[j]] *= submodel.label[0];
}
}
}
SigmoidTrain(prob.Count, decValues, prob.y, probAb);
}
public static float SvmPredictValues(SvmModel model, BaseVector x, double[] decValues)
{
if (model.l == 0)
{
return(float.NaN);
}
if (model.param.svmType == SvmType.OneClass || model.param.svmType == SvmType.EpsilonSvr ||
model.param.svmType == SvmType.NuSvr)
{
double[] svCoef = model.svCoef[0];
double sum = 0;
for (int i = 0; i < model.l; i++)
{
sum += svCoef[i] * KFunction(x, model.sv[i], model.param);
}
sum -= model.rho[0];
decValues[0] = sum;
if (model.param.svmType == SvmType.OneClass)
{
return(sum > 0 ? 1 : -1);
}
return((float)sum);
}
int nrClass = model.nrClass;
int l = model.l;
double[] kvalue = new double[l];
for (int i = 0; i < l; i++)
{
kvalue[i] = KFunction(x, model.sv[i], model.param);
}
int[] start = new int[nrClass];
start[0] = 0;
for (int i = 1; i < nrClass; i++)
{
start[i] = start[i - 1] + model.nSv[i - 1];
}
int[] vote = new int[nrClass];
for (int i = 0; i < nrClass; i++)
{
vote[i] = 0;
}
int p = 0;
for (int i = 0; i < nrClass; i++)
{
for (int j = i + 1; j < nrClass; j++)
{
double sum = 0;
int si = start[i];
int sj = start[j];
int ci = model.nSv[i];
int cj = model.nSv[j];
int k;
double[] coef1 = model.svCoef[j - 1];
double[] coef2 = model.svCoef[i];
for (k = 0; k < ci; k++)
{
sum += coef1[si + k] * kvalue[si + k];
}
for (k = 0; k < cj; k++)
{
sum += coef2[sj + k] * kvalue[sj + k];
}
sum -= model.rho[p];
decValues[p] = sum;
if (decValues[p] > 0)
{
++vote[i];
}
else
{
++vote[j];
}
p++;
}
}
int voteMaxIdx = 0;
for (int i = 1; i < nrClass; i++)
{
if (vote[i] > vote[voteMaxIdx])
{
voteMaxIdx = i;
}
}
return(model.label[voteMaxIdx]);
}
public static void SvmGetLabels(SvmModel model, int[] label)
{
if (model.label != null){
for (int i = 0; i < model.nrClass; i++){
label[i] = model.label[i];
}
}
}
public SvmClassificationModel(SvmModel[] models, bool[] invert)
{
this.models = models;
this.invert = invert;
}
//
// Interface functions
//
public static SvmModel SvmTrain(SvmProblem prob, SvmParameter param)
{
SvmModel model = new SvmModel{param = param};
if (param.svmType == SvmType.OneClass || param.svmType == SvmType.EpsilonSvr || param.svmType == SvmType.NuSvr){
// regression or one-class-svm
model.nrClass = 2;
model.label = null;
model.nSv = null;
model.probA = null;
model.probB = null;
model.svCoef = new double[1][];
if (param.probability && (param.svmType == SvmType.EpsilonSvr || param.svmType == SvmType.NuSvr)){
model.probA = new double[1];
model.probA[0] = SvmSvrProbability(prob, param);
}
DecisionFunction f = SvmTrainOne(prob, param, 0, 0);
model.rho = new double[1];
model.rho[0] = f.rho;
int nSv = 0;
int i;
for (i = 0; i < prob.Count; i++){
if (Math.Abs(f.alpha[i]) > 0){
++nSv;
}
}
model.l = nSv;
model.sv = new BaseVector[nSv];
model.svCoef[0] = new double[nSv];
int j = 0;
for (i = 0; i < prob.Count; i++){
if (Math.Abs(f.alpha[i]) > 0){
model.sv[j] = prob.x[i];
model.svCoef[0][j] = f.alpha[i];
++j;
}
}
} else{
// classification
int l = prob.Count;
int[] tmpNrClass = new int[1];
int[][] tmpLabel = new int[1][];
int[][] tmpStart = new int[1][];
int[][] tmpCount = new int[1][];
int[] perm = new int[l];
// group training data of the same class
SvmGroupClasses(prob, tmpNrClass, tmpLabel, tmpStart, tmpCount, perm);
int nrClass = tmpNrClass[0];
int[] label = tmpLabel[0];
int[] start = tmpStart[0];
int[] count = tmpCount[0];
if (nrClass == 1){
Info("WARNING: training data in only one class. See README for details.\n");
}
BaseVector[] x = new BaseVector[l];
int i;
for (i = 0; i < l; i++){
x[i] = prob.x[perm[i]];
}
// calculate weighted C
double[] weightedC = new double[nrClass];
for (i = 0; i < nrClass; i++){
weightedC[i] = param.c;
}
for (i = 0; i < param.nrWeight; i++){
int j;
for (j = 0; j < nrClass; j++){
if (param.weightLabel[i] == label[j]){
break;
}
}
if (j == nrClass){
Info("WARNING: class label " + param.weightLabel[i] + " specified in weight is not found\n");
} else{
weightedC[j] *= param.weight[i];
}
}
// train k*(k-1)/2 models
bool[] nonzero = new bool[l];
for (i = 0; i < l; i++){
nonzero[i] = false;
}
DecisionFunction[] f = new DecisionFunction[nrClass*(nrClass - 1)/2];
double[] probA = null, probB = null;
if (param.probability){
probA = new double[nrClass*(nrClass - 1)/2];
probB = new double[nrClass*(nrClass - 1)/2];
}
int p = 0;
for (i = 0; i < nrClass; i++){
for (int j = i + 1; j < nrClass; j++){
int si = start[i], sj = start[j];
int ci = count[i], cj = count[j];
int c = ci + cj;
SvmProblem subProb = new SvmProblem{x = new BaseVector[c], y = new float[c]};
int k;
for (k = 0; k < ci; k++){
subProb.x[k] = x[si + k];
subProb.y[k] = +1;
}
for (k = 0; k < cj; k++){
subProb.x[ci + k] = x[sj + k];
subProb.y[ci + k] = -1;
}
if (param.probability){
double[] probAb = new double[2];
SvmBinarySvcProbability(subProb, param, weightedC[i], weightedC[j], probAb);
probA[p] = probAb[0];
probB[p] = probAb[1];
}
f[p] = SvmTrainOne(subProb, param, weightedC[i], weightedC[j]);
for (k = 0; k < ci; k++){
if (!nonzero[si + k] && Math.Abs(f[p].alpha[k]) > 0){
nonzero[si + k] = true;
}
}
for (k = 0; k < cj; k++){
if (!nonzero[sj + k] && Math.Abs(f[p].alpha[ci + k]) > 0){
nonzero[sj + k] = true;
}
}
++p;
}
}
// build output
model.nrClass = nrClass;
model.label = new int[nrClass];
for (i = 0; i < nrClass; i++){
model.label[i] = label[i];
}
model.rho = new double[nrClass*(nrClass - 1)/2];
for (i = 0; i < nrClass*(nrClass - 1)/2; i++){
model.rho[i] = f[i].rho;
}
if (param.probability){
model.probA = new double[nrClass*(nrClass - 1)/2];
model.probB = new double[nrClass*(nrClass - 1)/2];
for (i = 0; i < nrClass*(nrClass - 1)/2; i++){
model.probA[i] = probA[i];
model.probB[i] = probB[i];
}
} else{
model.probA = null;
model.probB = null;
}
int nnz = 0;
int[] nzCount = new int[nrClass];
model.nSv = new int[nrClass];
for (i = 0; i < nrClass; i++){
int nSv = 0;
for (int j = 0; j < count[i]; j++){
if (nonzero[start[i] + j]){
++nSv;
++nnz;
}
}
model.nSv[i] = nSv;
nzCount[i] = nSv;
}
Info("Total nSV = " + nnz + "\n");
model.l = nnz;
model.sv = new BaseVector[nnz];
p = 0;
for (i = 0; i < l; i++){
if (nonzero[i]){
model.sv[p++] = x[i];
}
}
int[] nzStart = new int[nrClass];
nzStart[0] = 0;
for (i = 1; i < nrClass; i++){
nzStart[i] = nzStart[i - 1] + nzCount[i - 1];
}
model.svCoef = new double[nrClass - 1][];
for (i = 0; i < nrClass - 1; i++){
model.svCoef[i] = new double[nnz];
}
p = 0;
for (i = 0; i < nrClass; i++){
for (int j = i + 1; j < nrClass; j++){
// classifier (i,j): coefficients with
// i are in sv_coef[j-1][nz_start[i]...],
// j are in sv_coef[i][nz_start[j]...]
int si = start[i];
int sj = start[j];
int ci = count[i];
int cj = count[j];
int q = nzStart[i];
int k;
for (k = 0; k < ci; k++){
if (nonzero[si + k]){
model.svCoef[j - 1][q++] = f[p].alpha[k];
}
}
q = nzStart[j];
for (k = 0; k < cj; k++){
if (nonzero[sj + k]){
model.svCoef[i][q++] = f[p].alpha[ci + k];
}
}
++p;
}
}
}
return model;
}
public static float SvmPredictValues(SvmModel model, BaseVector x, double[] decValues)
{
if (model.l == 0){
return float.NaN;
}
if (model.param.svmType == SvmType.OneClass || model.param.svmType == SvmType.EpsilonSvr ||
model.param.svmType == SvmType.NuSvr){
double[] svCoef = model.svCoef[0];
double sum = 0;
for (int i = 0; i < model.l; i++){
sum += svCoef[i]*KFunction(x, model.sv[i], model.param);
}
sum -= model.rho[0];
decValues[0] = sum;
if (model.param.svmType == SvmType.OneClass){
return sum > 0 ? 1 : -1;
}
return (float) sum;
}
int nrClass = model.nrClass;
int l = model.l;
double[] kvalue = new double[l];
for (int i = 0; i < l; i++){
kvalue[i] = KFunction(x, model.sv[i], model.param);
}
int[] start = new int[nrClass];
start[0] = 0;
for (int i = 1; i < nrClass; i++){
start[i] = start[i - 1] + model.nSv[i - 1];
}
int[] vote = new int[nrClass];
for (int i = 0; i < nrClass; i++){
vote[i] = 0;
}
int p = 0;
for (int i = 0; i < nrClass; i++){
for (int j = i + 1; j < nrClass; j++){
double sum = 0;
int si = start[i];
int sj = start[j];
int ci = model.nSv[i];
int cj = model.nSv[j];
int k;
double[] coef1 = model.svCoef[j - 1];
double[] coef2 = model.svCoef[i];
for (k = 0; k < ci; k++){
sum += coef1[si + k]*kvalue[si + k];
}
for (k = 0; k < cj; k++){
sum += coef2[sj + k]*kvalue[sj + k];
}
sum -= model.rho[p];
decValues[p] = sum;
if (decValues[p] > 0){
++vote[i];
} else{
++vote[j];
}
p++;
}
}
int voteMaxIdx = 0;
for (int i = 1; i < nrClass; i++){
if (vote[i] > vote[voteMaxIdx]){
voteMaxIdx = i;
}
}
return model.label[voteMaxIdx];
}
public static float SvmPredictProbability(SvmModel model, BaseVector x, double[] probEstimates)
{
if ((model.param.svmType == SvmType.CSvc || model.param.svmType == SvmType.NuSvc) && model.probA != null &&
model.probB != null){
int nrClass = model.nrClass;
double[] decValues = new double[nrClass*(nrClass - 1)/2];
SvmPredictValues(model, x, decValues);
const double minProb = 1e-7;
double[][] pairwiseProb = new double[nrClass][];
for (int m = 0; m < nrClass; m++){
pairwiseProb[m] = new double[nrClass];
}
int k = 0;
for (int i = 0; i < nrClass; i++){
for (int j = i + 1; j < nrClass; j++){
pairwiseProb[i][j] = Math.Min(Math.Max(SigmoidPredict(decValues[k], model.probA[k], model.probB[k]), minProb),
1 - minProb);
pairwiseProb[j][i] = 1 - pairwiseProb[i][j];
k++;
}
}
MulticlassProbability(nrClass, pairwiseProb, probEstimates);
int probMaxIdx = 0;
for (int i = 1; i < nrClass; i++){
if (probEstimates[i] > probEstimates[probMaxIdx]){
probMaxIdx = i;
}
}
return model.label[probMaxIdx];
}
return SvmPredict(model, x);
}
public static float SvmPredict(SvmModel model, BaseVector x)
{
int nrClass = model.nrClass;
double[] decValues;
if (model.param.svmType == SvmType.OneClass || model.param.svmType == SvmType.EpsilonSvr ||
model.param.svmType == SvmType.NuSvr){
decValues = new double[1];
} else{
decValues = new double[nrClass*(nrClass - 1)/2];
}
float predResult = SvmPredictValues(model, x, decValues);
return predResult;
}
public static double SvmGetSvrProbability(SvmModel model)
{
if ((model.param.svmType == SvmType.EpsilonSvr || model.param.svmType == SvmType.NuSvr) && model.probA != null){
return model.probA[0];
}
Info("Model doesn't contain information for SVR probability inference\n");
return 0;
}
//
// Interface functions
//
public static SvmModel SvmTrain(SvmProblem prob, SvmParameter param)
{
SvmModel model = new SvmModel {
param = param
};
if (param.svmType == SvmType.OneClass || param.svmType == SvmType.EpsilonSvr || param.svmType == SvmType.NuSvr)
{
// regression or one-class-svm
model.nrClass = 2;
model.label = null;
model.nSv = null;
model.probA = null;
model.probB = null;
model.svCoef = new double[1][];
if (param.probability && (param.svmType == SvmType.EpsilonSvr || param.svmType == SvmType.NuSvr))
{
model.probA = new double[1];
model.probA[0] = SvmSvrProbability(prob, param);
}
DecisionFunction f = SvmTrainOne(prob, param, 0, 0);
model.rho = new double[1];
model.rho[0] = f.rho;
int nSv = 0;
int i;
for (i = 0; i < prob.Count; i++)
{
if (Math.Abs(f.alpha[i]) > 0)
{
++nSv;
}
}
model.l = nSv;
model.sv = new BaseVector[nSv];
model.svCoef[0] = new double[nSv];
int j = 0;
for (i = 0; i < prob.Count; i++)
{
if (Math.Abs(f.alpha[i]) > 0)
{
model.sv[j] = prob.x[i];
model.svCoef[0][j] = f.alpha[i];
++j;
}
}
}
else
{
// classification
int l = prob.Count;
int[] tmpNrClass = new int[1];
int[][] tmpLabel = new int[1][];
int[][] tmpStart = new int[1][];
int[][] tmpCount = new int[1][];
int[] perm = new int[l];
// group training data of the same class
SvmGroupClasses(prob, tmpNrClass, tmpLabel, tmpStart, tmpCount, perm);
int nrClass = tmpNrClass[0];
int[] label = tmpLabel[0];
int[] start = tmpStart[0];
int[] count = tmpCount[0];
if (nrClass == 1)
{
Info("WARNING: training data in only one class. See README for details.\n");
}
BaseVector[] x = new BaseVector[l];
int i;
for (i = 0; i < l; i++)
{
x[i] = prob.x[perm[i]];
}
// calculate weighted C
double[] weightedC = new double[nrClass];
for (i = 0; i < nrClass; i++)
{
weightedC[i] = param.c;
}
for (i = 0; i < param.nrWeight; i++)
{
int j;
for (j = 0; j < nrClass; j++)
{
if (param.weightLabel[i] == label[j])
{
break;
}
}
if (j == nrClass)
{
Info("WARNING: class label " + param.weightLabel[i] + " specified in weight is not found\n");
}
else
{
weightedC[j] *= param.weight[i];
}
}
// train k*(k-1)/2 models
bool[] nonzero = new bool[l];
for (i = 0; i < l; i++)
{
nonzero[i] = false;
}
DecisionFunction[] f = new DecisionFunction[nrClass * (nrClass - 1) / 2];
double[] probA = null, probB = null;
if (param.probability)
{
probA = new double[nrClass * (nrClass - 1) / 2];
probB = new double[nrClass * (nrClass - 1) / 2];
}
int p = 0;
for (i = 0; i < nrClass; i++)
{
for (int j = i + 1; j < nrClass; j++)
{
int si = start[i], sj = start[j];
int ci = count[i], cj = count[j];
int c = ci + cj;
SvmProblem subProb = new SvmProblem {
x = new BaseVector[c], y = new float[c]
};
int k;
for (k = 0; k < ci; k++)
{
subProb.x[k] = x[si + k];
subProb.y[k] = +1;
}
for (k = 0; k < cj; k++)
{
subProb.x[ci + k] = x[sj + k];
subProb.y[ci + k] = -1;
}
if (param.probability)
{
double[] probAb = new double[2];
SvmBinarySvcProbability(subProb, param, weightedC[i], weightedC[j], probAb);
probA[p] = probAb[0];
probB[p] = probAb[1];
}
f[p] = SvmTrainOne(subProb, param, weightedC[i], weightedC[j]);
for (k = 0; k < ci; k++)
{
if (!nonzero[si + k] && Math.Abs(f[p].alpha[k]) > 0)
{
nonzero[si + k] = true;
}
}
for (k = 0; k < cj; k++)
{
if (!nonzero[sj + k] && Math.Abs(f[p].alpha[ci + k]) > 0)
{
nonzero[sj + k] = true;
}
}
++p;
}
}
// build output
model.nrClass = nrClass;
model.label = new int[nrClass];
for (i = 0; i < nrClass; i++)
{
model.label[i] = label[i];
}
model.rho = new double[nrClass * (nrClass - 1) / 2];
for (i = 0; i < nrClass * (nrClass - 1) / 2; i++)
{
model.rho[i] = f[i].rho;
}
if (param.probability)
{
model.probA = new double[nrClass * (nrClass - 1) / 2];
model.probB = new double[nrClass * (nrClass - 1) / 2];
for (i = 0; i < nrClass * (nrClass - 1) / 2; i++)
{
model.probA[i] = probA[i];
model.probB[i] = probB[i];
}
}
else
{
model.probA = null;
model.probB = null;
}
int nnz = 0;
int[] nzCount = new int[nrClass];
model.nSv = new int[nrClass];
for (i = 0; i < nrClass; i++)
{
int nSv = 0;
for (int j = 0; j < count[i]; j++)
{
if (nonzero[start[i] + j])
{
++nSv;
++nnz;
}
}
model.nSv[i] = nSv;
nzCount[i] = nSv;
}
Info("Total nSV = " + nnz + "\n");
model.l = nnz;
model.sv = new BaseVector[nnz];
p = 0;
for (i = 0; i < l; i++)
{
if (nonzero[i])
{
model.sv[p++] = x[i];
}
}
int[] nzStart = new int[nrClass];
nzStart[0] = 0;
for (i = 1; i < nrClass; i++)
{
nzStart[i] = nzStart[i - 1] + nzCount[i - 1];
}
model.svCoef = new double[nrClass - 1][];
for (i = 0; i < nrClass - 1; i++)
{
model.svCoef[i] = new double[nnz];
}
p = 0;
for (i = 0; i < nrClass; i++)
{
for (int j = i + 1; j < nrClass; j++)
{
// classifier (i,j): coefficients with
// i are in sv_coef[j-1][nz_start[i]...],
// j are in sv_coef[i][nz_start[j]...]
int si = start[i];
int sj = start[j];
int ci = count[i];
int cj = count[j];
int q = nzStart[i];
int k;
for (k = 0; k < ci; k++)
{
if (nonzero[si + k])
{
model.svCoef[j - 1][q++] = f[p].alpha[k];
}
}
q = nzStart[j];
for (k = 0; k < cj; k++)
{
if (nonzero[sj + k])
{
model.svCoef[i][q++] = f[p].alpha[ci + k];
}
}
++p;
}
}
}
return(model);
}
// Stratified cross validation
public static void SvmCrossValidation(SvmProblem prob, SvmParameter param, int nrFold, double[] target)
{
int i;
int[] foldStart = new int[nrFold + 1];
int l = prob.Count;
int[] perm = new int[l];
// stratified cv may not give leave-one-out rate
// Each class to l folds -> some folds may have zero elements
if ((param.svmType == SvmType.CSvc || param.svmType == SvmType.NuSvc) && nrFold < l)
{
int[] tmpNrClass = new int[1];
int[][] tmpLabel = new int[1][];
int[][] tmpStart = new int[1][];
int[][] tmpCount = new int[1][];
SvmGroupClasses(prob, tmpNrClass, tmpLabel, tmpStart, tmpCount, perm);
int nrClass = tmpNrClass[0];
int[] start = tmpStart[0];
int[] count = tmpCount[0];
// random shuffle and then data grouped by fold using the array perm
int[] foldCount = new int[nrFold];
int[] index = new int[l];
for (i = 0; i < l; i++)
{
index[i] = perm[i];
}
for (int c = 0; c < nrClass; c++)
{
for (i = 0; i < count[c]; i++)
{
int j = i + rand.Next(count[c] - i);
do
{
int _ = index[start[c] + j];
index[start[c] + j] = index[start[c] + i];
index[start[c] + i] = _;
} while (false);
}
}
for (i = 0; i < nrFold; i++)
{
foldCount[i] = 0;
for (int c = 0; c < nrClass; c++)
{
foldCount[i] += (i + 1) * count[c] / nrFold - i * count[c] / nrFold;
}
}
foldStart[0] = 0;
for (i = 1; i <= nrFold; i++)
{
foldStart[i] = foldStart[i - 1] + foldCount[i - 1];
}
for (int c = 0; c < nrClass; c++)
{
for (i = 0; i < nrFold; i++)
{
int begin = start[c] + i * count[c] / nrFold;
int end = start[c] + (i + 1) * count[c] / nrFold;
for (int j = begin; j < end; j++)
{
perm[foldStart[i]] = index[j];
foldStart[i]++;
}
}
}
foldStart[0] = 0;
for (i = 1; i <= nrFold; i++)
{
foldStart[i] = foldStart[i - 1] + foldCount[i - 1];
}
}
else
{
for (i = 0; i < l; i++)
{
perm[i] = i;
}
for (i = 0; i < l; i++)
{
int j = i + rand.Next(l - i);
do
{
int _ = perm[i];
perm[i] = perm[j];
perm[j] = _;
} while (false);
}
for (i = 0; i <= nrFold; i++)
{
foldStart[i] = i * l / nrFold;
}
}
for (i = 0; i < nrFold; i++)
{
int begin = foldStart[i];
int end = foldStart[i + 1];
int j;
int count = l - (end - begin);
SvmProblem subprob = new SvmProblem {
x = new BaseVector[count], y = new float[count]
};
int k = 0;
for (j = 0; j < begin; j++)
{
subprob.x[k] = prob.x[perm[j]];
subprob.y[k] = prob.y[perm[j]];
++k;
}
for (j = end; j < l; j++)
{
subprob.x[k] = prob.x[perm[j]];
subprob.y[k] = prob.y[perm[j]];
++k;
}
SvmModel submodel = SvmTrain(subprob, param);
if (param.probability && (param.svmType == SvmType.CSvc || param.svmType == SvmType.NuSvc))
{
double[] probEstimates = new double[submodel.nrClass];
for (j = begin; j < end; j++)
{
target[perm[j]] = SvmPredictProbability(submodel, prob.x[perm[j]], probEstimates);
}
}
else
{
for (j = begin; j < end; j++)
{
target[perm[j]] = SvmPredict(submodel, prob.x[perm[j]]);
}
}
}
}
public static int SvmCheckProbabilityModel(SvmModel model)
{
if (((model.param.svmType == SvmType.CSvc || model.param.svmType == SvmType.NuSvc) && model.probA != null &&
model.probB != null) ||
((model.param.svmType == SvmType.EpsilonSvr || model.param.svmType == SvmType.NuSvr) && model.probA != null)){
return 1;
}
return 0;
}
