pair<Emotion, float> EmoDetector::predictBestWinsOneVsAll(cv::Mat& frame) {
pair<Emotion, float> best(UNKNOWN, numeric_limits<float>::min());
if (detectors_ext.size() == 0) {
return make_pair(UNKNOWN, 0.0f);
}
for (map<string, pair<vector<Emotion>, Classifier*> >::iterator ii =
this->detectors_ext.begin(); ii != this->detectors_ext.end(); ++ii) {
if (ii->second.first.size() != 1) {
continue;
}
Emotion emo = ii->second.first[0];
Classifier* cl = ii->second.second;
float prediction = cl->predict(frame);
if (best.second < prediction) {
best.first = emo;
best.second = prediction;
}
}
return best;
}
pair<Emotion, float> EmoDetector::predictVotingOneVsAllExt(cv::Mat& frame) {
map<Emotion,float> votes;
if (detectors_ext.size() == 0) {
return make_pair(UNKNOWN, 0.0f);
}
votes.insert(make_pair(NEUTRAL, 0.f));
votes.insert(make_pair(CONTEMPT , 0.f));
votes.insert(make_pair(DISGUST, 0.f));
votes.insert(make_pair(SADNESS, 0.f));
votes.insert(make_pair(ANGER, 0.f));
votes.insert(make_pair(HAPPY, 0.f));
votes.insert(make_pair(FEAR, 0.f));
votes.insert(make_pair(UNKNOWN, 0.f));
for(map<string, pair<vector<Emotion>, Classifier*> >::iterator ii =
this->detectors_ext.begin(); ii != this->detectors_ext.end(); ++ii) {
vector<Emotion> emo = ii->second.first; // detected emotions
Classifier* cl = ii->second.second;
float prediction = cl->predict(frame);
for(vector<Emotion>::iterator emo_it = emo.begin(); emo_it != emo.end(); ++emo_it) {
map<Emotion, float>::iterator it = votes.find(*emo_it);
if (it == votes.end()) {
votes.insert(make_pair(*emo_it, prediction));
} else{
if (prediction > 0.5) {
it->second += 0.5; //1.0;
} else {
//for(map<Emotion,float>::iterator votes_it = votes.begin(); votes_it != votes.end(); ++votes_it) {
// vector<Emotion>::iterator e_it = find(emo.begin(), emo.end(), votes_it->first);
// if (e_it == emo.end()) {
// // if I dont find emotion in detected emotion
// votes_it->second+=1.0;
// }
//}
it->second -= 0.0;
}
}
}
}
pair<Emotion,float> max_pair = make_pair(UNKNOWN, numeric_limits<float>::min());
for( map<Emotion, float>::iterator ii = votes.begin(); ii != votes.end();
++ii) {
if (ii->second > max_pair.second) {
max_pair.first = ii->first;
max_pair.second = ii->second;
}
}
return max_pair;
}
pair<Emotion, float> EmoDetector::predictMayorityOneVsAll(cv::Mat& frame){
map<Emotion,float> votes;
if (detectors_ext.size() == 0) {
return make_pair(UNKNOWN, 0.0f);
}
for (map<string, pair<vector<Emotion>, Classifier*> >::iterator ii =
this->detectors_ext.begin(); ii != this->detectors_ext.end(); ++ii) {
if (ii->second.first.size() != 1) {
continue;
}
Emotion emo = ii->second.first[0];
Classifier* cl = ii->second.second;
float prediction = cl->predict(frame);
map<Emotion, float>::iterator it = votes.find(emo);
if (it == votes.end()) {
votes.insert(make_pair(emo, prediction));
} else {
it->second+=prediction;
}
}
pair<Emotion,float> max_pair = make_pair(UNKNOWN, numeric_limits<float>::min());
for( map<Emotion, float>::iterator ii=votes.begin(); ii!=votes.end(); ++ii){
if (ii->second > max_pair.second){
max_pair.first=ii->first;
max_pair.second=ii->second;
}
}
return max_pair;
}
int main(void) {
double data[][4] = {
{3.5, 5.3, 0.2, -1.2},
{4.4, 2.2, 0.3, 0.4},
{1.3, 0.5, 4.1, 3.2}
};
int labels[] = {1, 2, 3};
Classifier* logReg = new LogisticRegression();
logReg->epsilon = 1e-5;
logReg->feedData<4>(data, 3, 4);
logReg->feedLabels(labels, 3);
// Get elapsed time in seconds
tic();
logReg->train();
fprintf("Elapsed time: %.3f seconds.\n", toc());
fprintf("W:\n");
printMatrix(*logReg->W);
fprintf("b:\n");
printVector(logReg->b, 3);
// double** dataTest = data;
double dataTest[][4] = {
{3.5, 5.3, 0.2, -1.2},
{4.4, 2.2, 0.3, 0.4},
{1.3, 0.5, 4.1, 3.2}
};
fprintf("Ground truth:\n");
printMatrix(*logReg->Y);
fprintf("Predicted probability matrix:\n");
Matrix& Prob_pred = logReg->predictLabelScoreMatrix<4>(dataTest, 3, 4);
disp(Prob_pred);
fprintf("Predicted label matrix:\n");
Matrix& Y_pred = logReg->predictLabelMatrix<4>(dataTest, 3, 4);
printMatrix(Y_pred);
int* pred_labels = logReg->predict(dataTest, 3, 4);
Classifier::getAccuracy(pred_labels, labels, 3);
tic();
std::string trainDataFilePath = "heart_scale";
logReg = new LogisticRegression();
DataSet& dataSet = readDataSetFromFile(trainDataFilePath);
logReg->feedData(*dataSet.X);
logReg->feedLabels(dataSet.Y, dataSet.nExample);
logReg->train();
Matrix& XTest = *dataSet.X;
pred_labels = logReg->predict(XTest);
int nt = XTest.getRowDimension();
Classifier::getAccuracy(pred_labels, logReg->labels, nt);
fprintf("Elapsed time: %.3f seconds.\n", toc());
return EXIT_SUCCESS;
}
