Method SVM_predict_values

svm.go:2258–2340

(model *SVM_Model, x []SVM_Node, dec_values []float64)

}

func (this *SVM) SVM_predict_values(model *SVM_Model, x []SVM_Node, dec_values []float64) float64 {
	var i int
	var rst float64
	if model.param.svm_type == ONE_CLASS ||
		model.param.svm_type == EPSILON_SVR ||
		model.param.svm_type == NU_SVR {
		sv_coef := model.sv_coef[0]
		sum := float64(0)
		for i = 0; i < model.l; i++ {
			sum += sv_coef[i] * k_function(x, model.SV[i], model.param)
		}
		sum -= model.rho[0]
		dec_values[0] = sum

		if model.param.svm_type == ONE_CLASS {
			if sum > 0 {
				rst = 1
			} else {
				rst = -1
			}
		} else {
			rst = sum
		}
	} else {
		nr_class := model.nr_class
		l := model.l

		kvalue := make([]float64, l)
		for i = 0; i < l; i++ {
			kvalue[i] = k_function(x, model.SV[i], model.param)
		}

		start := make([]int, nr_class)
		start[0] = 0
		for i = 1; i < nr_class; i++ {
			start[i] = start[i-1] + model.nSV[i-1]
		}

		vote := make([]int, nr_class)
		for i = 0; i < nr_class; i++ {
			vote[i] = 0
		}

		p := 0
		for i = 0; i < nr_class; i++ {
			for j := i + 1; j < nr_class; j++ {
				sum := float64(0)
				si := start[i]
				sj := start[j]
				ci := model.nSV[i]
				cj := model.nSV[j]

				var k int
				coef1 := model.sv_coef[j-1]
				coef2 := model.sv_coef[i]
				for k = 0; k < ci; k++ {
					sum += coef1[si+k] * kvalue[si+k]
				}