This is the result after applying trained svm..i don't know why i got support vectors as 0's and why i got uncompressed support vectors. Can someone help me?
<?xml version="1.0"?>
-<opencv_storage>
-<opencv_ml_svm>
<format>3</format>
<svmType>C_SVC</svmType>
-<kernel>
<type>LINEAR</type>
</kernel>
<C>1.</C>
-<term_criteria>
<epsilon>9.9999999999999995e-07</epsilon>
<iterations>1000</iterations>
</term_criteria>
<var_count>8</var_count>
<class_count>2</class_count>
-<class_labels type_id="opencv-matrix">
<rows>2</rows>
<cols>1</cols>
<dt>i</dt>
<data> -1 1</data>
</class_labels>
<sv_total>1</sv_total>
-<support_vectors>
<_> 0. 0. 0. 0. 0. 0. 0. 0.</_>
</support_vectors>
<uncompressed_sv_total>2</uncompressed_sv_total>
-<uncompressed_support_vectors>
<_> 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00</_>
<_> 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00 1.00024951e+00</_>
</uncompressed_support_vectors>
-<decision_functions>
-<_>
<sv_count>1</sv_count>
<rho>1.</rho>
<alpha> 1.</alpha>
</_>
</decision_functions>
</opencv_ml_svm>
</opencv_storage>Here is my code:
#include <opencv2/core.hpp>
#include <opencv2/opencv.hpp>
#include <opencv2/imgproc.hpp>
#include "opencv2/imgcodecs.hpp"
#include <opencv2/highgui.hpp>
#include <opencv2/ml.hpp>
#include <iostream>
#include <fstream>
#include<string.h>
#define POS 8
#define NEG 28
#define TOTAL 36
#define DESCRIPT 2937060
using namespace std;
using namespace cv;
using namespace cv::ml;
float trainingData[TOTAL][DESCRIPT];
int labels[TOTAL];
void set_labels() {
int i;
for (i = 0; i < TOTAL; i++) {
if (i < POS) {
labels[i] = 1; //positive
}
else {
labels[i] = -1; //negative
}
}
return;
}
int main(int, char**)
{
FileStorage fpd("Positive_image.xml", FileStorage::READ);
FileStorage fnd("Negative_image.xml", FileStorage::READ);
// Set up training data
vector <float> train_D, pos_D, neg_D;
int k = 0;
//POSITIVE images
for (int i = 1; i < POS + 1; i++) {
stringstream a;
a << i;
fpd["Descriptors" + a.str()] >> pos_D;
for (int j = 0; j < pos_D.size(); j++) {
train_D.push_back(pos_D[j]);
}
}
//negative images
for (int i = 1; i < NEG + 1; i++) {
stringstream a;
a << i;
fpd["Descriptors" + a.str()] >> neg_D;
for (int j = 0; j < neg_D.size(); j++) {
train_D.push_back(neg_D[j]);
}
}
//total number of samples
for (int i = 0; i < TOTAL; i++) {
for (int j = 0; j < DESCRIPT; j++) {
trainingData[i][j] = train_D[k];
k++;
}
}
Mat trainingDataMat(TOTAL, DESCRIPT, CV_32FC1, trainingData);//training data matrix
Mat labelsMat(TOTAL, 1, CV_32SC1, labels);//label matrix
// Train the SVM
Ptr<cv::ml::SVM> svm = cv::ml::SVM::create();
//set up SVM parameters
svm->setType(SVM::C_SVC);
svm->setKernel(SVM::LINEAR);
svm->setTermCriteria(TermCriteria(TermCriteria::MAX_ITER, 100, 1e-6));
cout << "Training SVM" << endl;
svm->train(trainingDataMat, ROW_SAMPLE, labelsMat);
cout << "Saving Trained SVM xml ..." << endl;
svm->save("SVM.xml");//saving trained svm file
//Classification data
vector< float> descriptorsValues;
//vector to Mat
Mat fm = Mat(descriptorsValues);
Ptr<SVM> svm2 = Algorithm::load<SVM>("SVM.xml");//loading SVM file
std::cout << "Model Loaded" << std::endl;
float result = svm->predict(fm);
cout << "Predict value: " << result << endl;
return(0);
}