void diabetesTest(void)
{
Frontend fe;
if (!fe.loadDataFromFile("testfiles/diabietes2.train"))
{
std::cerr << "File not found." << std::endl;
return;
}
std::cout << "File succesfully loaded" << std::endl;
//FIXME create auto finder hidden layer size
z3ann ann(8, 2, 2, -1, 1);
ann.setAllLink(false);
ann.setSelfLink(false);
// ann.setRangeLink(0, 7, 8, 11, true);
// ann.setRangeLink(8, 11, 12, 13, true);
ann.setRangeLink(0, 7, 8, 9, true);
ann.setRangeLink(8, 9, 10, 11, true);
// I1
// ann.setLink(0, 13, false);
// ann.setLink(0, 14, false);
// ann.setLink(0, 4, true);
// ann.setAllLink(true);
ann.loadTrainingDataSet(fe.getData());
ann.solve();
std::cout << "========= C Backend Output ==========" << std::endl;
BackendC output(ann);
std::cout << output.getOutput();
}
int Application::Execute()
{
if (!LoadData(trainingImgs, Utils::TRAINING_IMAGES_PER_SUBJECT, true))
return Utils::EXIT_F;
if (!LoadData(validationImgs, Utils::VALIDATION_IMAGES_PER_SUBJECT, false))
return Utils::EXIT_F;
InitData(&trainingInputs, &trainingLabels, Utils::TRAINING_TOTAL_INPUT, trainingImgs);
InitData(&validationInputs, &validationLabels, Utils::VALIDATION_TOTAL_INPUT, validationImgs);
int hiddenLayerNeurons[] = { 5 };
NeuralNetwork ann(Utils::PIXEL_COUNT, Utils::HIDDEN_LAYER_COUNT, hiddenLayerNeurons, Utils::SUBJECT_COUNT);
printf(Utils::BEGIN_TRAINING);
ann.Train(trainingInputs, trainingLabels, Utils::TRAINING_TOTAL_INPUT);
printf(Utils::END_TRAINING);
ann.Save("ocr.ann");
for (int x = 0; x < Utils::VALIDATION_TOTAL_INPUT; x++)
{
for (int y = 0; y < Utils::SUBJECT_COUNT; y++)
validationLabels[x][y] = ann.GetGuess(validationInputs[x], y);
}
for (int x = 0; x < Utils::VALIDATION_TOTAL_INPUT; x++)
{
for (int y = 0; y < Utils::SUBJECT_COUNT; y++)
std::printf("%5.5f\n", validationLabels[x][y]);
std::printf("\n");
}
Clean();
return Utils::EXIT_S;
}
std::string* allt(int* idel, std::string ideligen) {
if (*idel > 3 || ideligen == "ifall")
return new std::string("igen");
int identifiering = *idel + 1;
std::string igenom("ihopkoppling");
int ihop = aning(&identifiering, igenom);
std::string* ikapp = new std::string("ilning");
int* illa = ann(identifiering, ikapp);
std::string* ilska = new std::string("implementering");
int* imperfekt = advent(identifiering, ilska);
std::string* inalles = new std::string("inblick");
int* inblandning = aftonsol(identifiering, inalles);
std::string* inbromsning = new std::string("indelning");
std::string indatagenerering = alltigenom(identifiering, inbromsning);
std::string indentering("indikering");
int indexering = aldrig(&identifiering, indentering);
std::string infix("informationsbit");
std::string* inflygning = alm(&identifiering, infix);
std::string ing("ingendera");
int ingalunda = andel(&identifiering, ing);
std::string* ingenstans = new std::string("ingetdera");
int* ingenting = allmoge(identifiering, ingenstans);
std::string* ingnidning = new std::string("initialisering");
return ingnidning;
} // allt
int* alltmera(int kryddnejlika, std::string* krympning) {
if (kryddnejlika > 3 || *krympning == "kugg")
return new int (29408);
int krypta = kryddnejlika + 1;
std::string* kula = new std::string("kulmen");
int* kullerbytta = ann(krypta, kula);
std::string* kulspruta = new std::string("kundmottagning");
int* kunde = allmoge(krypta, kulspruta);
std::string* kung = new std::string("kunnat");
int* kunna = anskaffning(krypta, kung);
std::string kunskapsutveckling("kupering");
int kupa = annanstans(&krypta, kunskapsutveckling);
std::string kurva("kvar");
std::string* kvadd = alltihop(&krypta, kurva);
std::string* kvarleva = new std::string("kvarvarande");
int* kvarn = anskaffning(krypta, kvarleva);
std::string* kvinna = new std::string("kyrka");
int* kyckling = adressering(krypta, kvinna);
std::string kyrkoherde("l");
int kyss = anslutningspropp(0, kyrkoherde);
std::string la("lada");
int labb = annanstans(&krypta, la);
int* laddning = new int(14289);
return laddning;
} // alltmera
// -----------------------------------------------------------------------------
//
// Purpose and Method:
// Inputs:
// Outputs:
// Dependencies:
// Restrictions and Caveats:
//
// -----------------------------------------------------------------------------
int
main
(
int argc,
char **argv
)
{
// Evaluate head pose
std::string ffd_config_file = "data/config_ffd.txt";
std::string headpose_config_file = "data/config_headpose.txt";
std::string face_cascade = "data/haarcascade_frontalface_alt.xml";
// Parse configuration files
ForestParam hp_param, mp_param;
if (!loadConfigFile(headpose_config_file, hp_param))
return EXIT_FAILURE;
if (!loadConfigFile(ffd_config_file, mp_param))
return EXIT_FAILURE;
// Loading images annotations
std::vector<FaceAnnotation> annotations;
if (!loadAnnotations(hp_param.image_path, annotations))
return EXIT_FAILURE;
// Separate annotations by head-pose classes
std::vector< std::vector<FaceAnnotation> > ann(NUM_HEADPOSE_CLASSES);
for (unsigned int i=0; i < annotations.size(); i++)
ann[annotations[i].pose+2].push_back(annotations[i]);
// Evaluate performance using 90% train and 10% test
std::vector< std::vector<FaceAnnotation> > test_ann(NUM_HEADPOSE_CLASSES);
for (unsigned int i=0; i < ann.size(); i++)
{
int num_train_imgs = static_cast<int>(ann[i].size() * TRAIN_IMAGES_PERCENTAGE);
test_ann[i].insert(test_ann[i].begin(), ann[i].begin()+num_train_imgs, ann[i].end());
}
FaceForestOptions ff_options;
ff_options.fd_option.path_face_cascade = face_cascade;
ff_options.hp_forest_param = hp_param;
ff_options.mp_forest_param = mp_param;
evalForest(ff_options, test_ann);
return EXIT_SUCCESS;
};
void Cluster::dropEvent(QDropEvent *event)
{
if (event->source() == 0) {
event->setDropAction(Qt::IgnoreAction);
}
else if (event->source() == this) {
event->setDropAction(Qt::IgnoreAction);
}
else if (event->mimeData()->hasFormat("Annotation")) {
QByteArray ba = event->mimeData()->data("Annotation");
Annotation ann(ba);
addAnnotation(ann);
event->acceptProposedAction();
}
}
// -----------------------------------------------------------------------------
//
// Purpose and Method:
// Inputs:
// Outputs:
// Dependencies:
// Restrictions and Caveats:
//
// -----------------------------------------------------------------------------
int
main
(
int argc,
char **argv
)
{
// Train feature points detector
std::string ffd_config_file = "data/config_ffd.txt";
// Parse configuration file
ForestParam mp_param;
if (!loadConfigFile(ffd_config_file, mp_param))
return EXIT_FAILURE;
// Loading images annotations
std::vector<FaceAnnotation> annotations;
if (!loadAnnotations(mp_param.image_path, annotations))
return EXIT_FAILURE;
// Separate annotations by head-pose classes
std::vector< std::vector<FaceAnnotation> > ann(NUM_HEADPOSE_CLASSES);
for (unsigned int i=0; i < annotations.size(); i++)
ann[annotations[i].pose+2].push_back(annotations[i]);
// Evaluate performance using 90% train and 10% test
std::vector< std::vector<FaceAnnotation> > train_ann(NUM_HEADPOSE_CLASSES);
for (unsigned int i=0; i < ann.size(); i++)
{
int num_train_imgs = static_cast<int>(ann[i].size() * TRAIN_IMAGES_PERCENTAGE);
train_ann[i].insert(train_ann[i].begin(), ann[i].begin(), ann[i].begin()+num_train_imgs);
}
// Train facial-feature forests
/*for (int i=0; i < NUM_HEADPOSE_CLASSES; i++)
for (int j=0; j < mp_param.ntrees; j++)
trainTree(mp_param, train_ann, i, j);*/
// Train facial-feature tree
int idx_forest = atoi(argv[1]);
int idx_tree = atoi(argv[2]);
trainTree(mp_param, train_ann, idx_forest, idx_tree);
return EXIT_SUCCESS;
};
int main() {
Simulator sim;
Announce ann("HEAR HEAR -- ");
sim.AddEvent( new AnnounceEvent(3.0,&ann,&Announce::DoWork) );
sim.AddEvent( new AnnounceEvent(2.0,&ann,&Announce::DoWork) );
sim.AddEvent( new AnnounceEvent(5.0,&ann,&Announce::DoWork) );
sim.AddEvent( new AnnounceEvent(1.0,&ann,&Announce::DoWork) );
sim.Start();
//slightly more interesting
Simulator sim2;
Repeater rep(&sim2,"I repeat");
sim2.AddEvent(
new RepeatingEvent(6.0,&rep,&Repeater::ContinueRepeating,10,0.3) );
sim2.AddEvent(
new RepeatingEvent(5.0,&rep,&Repeater::ContinueRepeating,5,0.8) );
sim2.Start();
}
int alltnog(int* lag, std::string lagning) {
if (*lag > 4 || lagning == "lagstiftning")
return 16702;
int lagring = *lag + 1;
std::string* lagtolkning = new std::string("landning");
int* lampa = aftonsol(lagring, lagtolkning);
std::string landsdel("lapp");
int landyta = aldrig(&lagring, landsdel);
std::string lathund("ledning");
int lax = annars(&lagring, lathund);
std::string* legat = new std::string("lek");
std::string legering = al(lagring, legat);
std::string lektionssal("levat");
int lera = allesammans(&lagring, lektionssal);
std::string* leve = new std::string("lika");
int* lie = ann(lagring, leve);
std::string* likafullt = new std::string("liksom");
int* likaledes = anhopning(lagring, likafullt);
std::string likt("limerick");
int lilja = allteftersom(&lagring, likt);
int limpa(11755);
return limpa;
} // alltnog
int main()
{
// diabetesTest();
// return 0;
std::cout << "+-------------------------------+" << std::endl;
std::cout << "| XOR example |" << std::endl;
std::cout << "+-------------------------------+" << std::endl;
std::cout << "| Set 1: Input{-1, -1} |" << std::endl;
std::cout << "| Set 2: Input{ 1, -1} |" << std::endl;
std::cout << "| Set 3: Input{-1, 1} |" << std::endl;
std::cout << "| Set 4: Input{ 1, 1} |" << std::endl;
std::cout << "+-------------------------------+" << std::endl;
std::cout << "| 0 -> -1 | activation : |" << std::endl;
std::cout << "| 1 -> 1 | -> Sum(Ii* Wij) > 0 |" << std::endl;
std::cout << "| | then 1 else -1 |" << std::endl;
std::cout << "+---------+---------------------+" << std::endl;
std::cout << "| o2 |" << std::endl;
std::cout << "| / \\ |" << std::endl;
std::cout << "| Input 1 -> o0 \\ |" << std::endl;
std::cout << "| \\ \\ |" << std::endl;
std::cout << "| 3o---o5 -> Output|" << std::endl;
std::cout << "| / / |" << std::endl;
std::cout << "| Input 2 -> o1 / |" << std::endl;
std::cout << "| \\ / |" << std::endl;
std::cout << "| o4 |" << std::endl;
std::cout << "+-------------------------------+" << std::endl << std::endl;
std::vector<trainingData> data;
data.resize(4);
data[0].input.resize(2);
data[1].input.resize(2);
data[2].input.resize(2);
data[3].input.resize(2);
data[0].output.resize(1);
data[1].output.resize(1);
data[2].output.resize(1);
data[3].output.resize(1);
data[0].input[0] = -1;
data[0].input[1] = -1;
data[0].output[0] = -1;
data[1].input[0] = 1;
data[1].input[1] = -1;
data[1].output[0] = 1;
data[2].input[0] = -1;
data[2].input[1] = 1;
data[2].output[0] = 1;
data[3].input[0] = 1;
data[3].input[1] = 1;
data[3].output[0] = -1;
z3ann ann(2, 3, 1, -1, 1);
ann.setAllLink(false);
// I1
ann.setLink(0, 2, true);
ann.setLink(0, 3, true);
ann.setLink(0, 4, true);
// I2
ann.setLink(1, 2, true);
ann.setLink(1, 3, true);
ann.setLink(1, 4, true);
// hidden to output layer
ann.setLink(2, 5, true);
ann.setLink(3, 5, true);
ann.setLink(4, 5, true);
// ann.setAllLink(true);
ann.loadTrainingDataSet(data);
ann.solve();
std::cout << "========= C Backend Output ==========" << std::endl;
BackendC output(ann);
std::cout << output.getOutput();
return 0;
}
rhomap::IMapView *rho_map_create(rho_param *p, rhomap::IDrawingDevice *device, int width, int height)
{
if (!p || p->type != RHO_PARAM_HASH)
rho_ruby_raise_argerror("Wrong input parameter (expect Hash)");
rho_param *provider = NULL;
rho_param *settings = NULL;
rho_param *annotations = NULL;
for (int i = 0, lim = p->v.hash->size; i < lim; ++i)
{
char *name = p->v.hash->name[i];
rho_param *value = p->v.hash->value[i];
if (!name || !value)
continue;
if (strcasecmp(name, "provider") == 0)
provider = value;
else if (strcasecmp(name, "settings") == 0)
settings = value;
else if (strcasecmp(name, "annotations") == 0)
annotations = value;
}
std::string providerId = "google";
if (provider)
{
if (provider->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'provider' value (expect String)");
providerId = provider->v.string;
}
std::transform(providerId.begin(), providerId.end(), providerId.begin(), &::tolower);
char *map_type = "roadmap";
bool use_center_radius = false;
double latitude = 0;
double longitude = 0;
double latitudeSpan = 0;
double longitudeSpan = 0;
char *center = NULL;
double radius = 0;
bool zoom_enabled = true;
bool scroll_enabled = true;
if (settings)
{
if (settings->type != RHO_PARAM_HASH)
rho_ruby_raise_argerror("Wrong 'settings' value (expect Hash)");
for (int i = 0, lim = settings->v.hash->size; i < lim; ++i)
{
char *name = settings->v.hash->name[i];
rho_param *value = settings->v.hash->value[i];
if (!name || !value)
continue;
if (strcasecmp(name, "map_type") == 0)
{
if (value->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'map_type' value (expect String)");
map_type = value->v.string;
if (strcasecmp(map_type,"standard") == 0) {
map_type = "roadmap";
}
}
else if (strcasecmp(name, "region") == 0)
{
if (value->type == RHO_PARAM_ARRAY)
{
if (value->v.array->size != 4)
rho_ruby_raise_argerror("'region' array should contain exactly 4 items");
rho_param *lat = value->v.array->value[0];
if (!lat) continue;
rho_param *lon = value->v.array->value[1];
if (!lon) continue;
rho_param *latSpan = value->v.array->value[2];
if (!latSpan) continue;
rho_param *lonSpan = value->v.array->value[3];
if (!lonSpan) continue;
latitude = lat->type == RHO_PARAM_STRING ? strtod(lat->v.string, NULL) : 0;
longitude = lon->type == RHO_PARAM_STRING ? strtod(lon->v.string, NULL) : 0;
latitudeSpan = latSpan->type == RHO_PARAM_STRING ? strtod(latSpan->v.string, NULL) : 0;
longitudeSpan = lonSpan->type == RHO_PARAM_STRING ? strtod(lonSpan->v.string, NULL) : 0;
use_center_radius = false;
}
else if (value->type == RHO_PARAM_HASH)
{
for (int j = 0, limm = value->v.hash->size; j < limm; ++j)
{
char *rname = value->v.hash->name[j];
rho_param *rvalue = value->v.hash->value[j];
if (!rname || !rvalue)
continue;
if (strcasecmp(rname, "center") == 0)
{
if (rvalue->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'center' value (expect String)");
center = rvalue->v.string;
}
else if (strcasecmp(rname, "radius") == 0)
{
if (rvalue->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'radius' value (expect String or Float)");
radius = strtod(rvalue->v.string, NULL);
}
}
use_center_radius = true;
}
else
rho_ruby_raise_argerror("Wrong 'region' value (expect Array or Hash");
}
else if (strcasecmp(name, "zoom_enabled") == 0)
{
if (value->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'zoom_enabled' value (expect boolean)");
zoom_enabled = strcasecmp(value->v.string, "true") == 0;
}
else if (strcasecmp(name, "scroll_enabled") == 0)
{
if (value->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong 'scroll_enabled' value (expect boolean)");
scroll_enabled = strcasecmp(value->v.string, "true") == 0;
}
}
}
typedef rhomap::Annotation ann_t;
typedef std::vector<ann_t> ann_list_t;
ann_list_t ann_list;
if (annotations)
{
if (annotations->type != RHO_PARAM_ARRAY)
rho_ruby_raise_argerror("Wrong 'annotations' value (expect Array)");
for (int i = 0, lim = annotations->v.array->size; i < lim; ++i)
{
rho_param *ann = annotations->v.array->value[i];
if (!ann)
continue;
if (ann->type != RHO_PARAM_HASH)
rho_ruby_raise_argerror("Wrong annotation value found (expect Hash)");
bool latitude_set = false;
double latitude = 0;
bool longitude_set = false;
double longitude = 0;
char const *address = "";
char const *title = "";
char const *subtitle = "";
char const *url = "";
char const *image = NULL;
int x_off = 0;
int y_off = 0;
for (int j = 0, limm = ann->v.hash->size; j < limm; ++j)
{
char *name = ann->v.hash->name[j];
rho_param *value = ann->v.hash->value[j];
if (!name || !value)
continue;
if (value->type != RHO_PARAM_STRING)
rho_ruby_raise_argerror("Wrong annotation value");
char *v = value->v.string;
if (strcasecmp(name, "latitude") == 0)
{
latitude = strtod(v, NULL);
latitude_set = true;
}
else if (strcasecmp(name, "longitude") == 0)
{
longitude = strtod(v, NULL);
longitude_set = true;
}
else if (strcasecmp(name, "street_address") == 0)
address = v;
else if (strcasecmp(name, "title") == 0)
title = v;
else if (strcasecmp(name, "subtitle") == 0)
subtitle = v;
else if (strcasecmp(name, "url") == 0)
url = v;
else if (strcasecmp(name, "image") == 0)
image = v;
else if (strcasecmp(name, "image_x_offset") == 0) {
x_off = (int)strtod(v, NULL);
}
else if (strcasecmp(name, "image_y_offset") == 0) {
y_off = (int)strtod(v, NULL);
}
}
if (latitude_set && longitude_set) {
ann_t ann(title, subtitle, latitude, longitude, url);
if (image != NULL) {
ann.setImageFileName(image, x_off, y_off);
}
ann_list.push_back(ann);
}
else {
ann_t ann(title, subtitle, address, url);
if (image != NULL) {
ann.setImageFileName(image, x_off, y_off);
}
ann_list.push_back(ann);
}
}
}
rhomap::IMapView *mapview = RHOMAPPROVIDER().createMapView(providerId, device);
if (!mapview)
return NULL;
mapview->setSize(width, height);
if (map_type)
mapview->setMapType(map_type);
if (use_center_radius)
{
mapview->moveTo(center);
mapview->setZoom(radius, radius);
}
else
{
mapview->moveTo(latitude, longitude);
mapview->setZoom(latitudeSpan, longitudeSpan);
}
mapview->setZoomEnabled(zoom_enabled);
mapview->setScrollEnabled(scroll_enabled);
for (ann_list_t::iterator it = ann_list.begin(), lim = ann_list.end(); it != lim; ++it)
mapview->addAnnotation(*it);
return mapview;
}
int main()
{
const int train_sample_count = 300;
bool is_regression = false;
const char* filename = "data/waveform.data";
int response_idx = 21;
CvMLData data;
CvTrainTestSplit spl( train_sample_count );
if(data.read_csv(filename) != 0)
{
printf("couldn't read %s\n", filename);
exit(0);
}
data.set_response_idx(response_idx);
data.change_var_type(response_idx, CV_VAR_CATEGORICAL);
data.set_train_test_split( &spl );
const CvMat* values = data.get_values();
const CvMat* response = data.get_responses();
const CvMat* missing = data.get_missing();
const CvMat* var_types = data.get_var_types();
const CvMat* train_sidx = data.get_train_sample_idx();
const CvMat* var_idx = data.get_var_idx();
CvMat*response_map;
CvMat*ordered_response = cv_preprocess_categories(response, var_idx, response->rows, &response_map, NULL);
int num_classes = response_map->cols;
CvDTree dtree;
printf("======DTREE=====\n");
CvDTreeParams cvd_params( 10, 1, 0, false, 16, 0, false, false, 0);
dtree.train( &data, cvd_params);
print_result( dtree.calc_error( &data, CV_TRAIN_ERROR), dtree.calc_error( &data, CV_TEST_ERROR ), dtree.get_var_importance() );
#if 0
/* boosted trees are only implemented for two classes */
printf("======BOOST=====\n");
CvBoost boost;
boost.train( &data, CvBoostParams(CvBoost::DISCRETE, 100, 0.95, 2, false, 0));
print_result( boost.calc_error( &data, CV_TRAIN_ERROR ), boost.calc_error( &data, CV_TEST_ERROR), 0 );
#endif
printf("======RTREES=====\n");
CvRTrees rtrees;
rtrees.train( &data, CvRTParams( 10, 2, 0, false, 16, 0, true, 0, 100, 0, CV_TERMCRIT_ITER ));
print_result( rtrees.calc_error( &data, CV_TRAIN_ERROR), rtrees.calc_error( &data, CV_TEST_ERROR ), rtrees.get_var_importance() );
printf("======ERTREES=====\n");
CvERTrees ertrees;
ertrees.train( &data, CvRTParams( 10, 2, 0, false, 16, 0, true, 0, 100, 0, CV_TERMCRIT_ITER ));
print_result( ertrees.calc_error( &data, CV_TRAIN_ERROR), ertrees.calc_error( &data, CV_TEST_ERROR ), ertrees.get_var_importance() );
printf("======GBTREES=====\n");
CvGBTrees gbtrees;
CvGBTreesParams gbparams;
gbparams.loss_function_type = CvGBTrees::DEVIANCE_LOSS; // classification, not regression
gbtrees.train( &data, gbparams);
//gbt_print_error(&gbtrees, values, response, response_idx, train_sidx);
print_result( gbtrees.calc_error( &data, CV_TRAIN_ERROR), gbtrees.calc_error( &data, CV_TEST_ERROR ), 0);
printf("======KNEAREST=====\n");
CvKNearest knearest;
//bool CvKNearest::train( const Mat& _train_data, const Mat& _responses,
// const Mat& _sample_idx, bool _is_regression,
// int _max_k, bool _update_base )
bool is_classifier = var_types->data.ptr[var_types->cols-1] == CV_VAR_CATEGORICAL;
assert(is_classifier);
int max_k = 10;
knearest.train(values, response, train_sidx, is_regression, max_k, false);
CvMat* new_response = cvCreateMat(response->rows, 1, values->type);
//print_types();
//const CvMat* train_sidx = data.get_train_sample_idx();
knearest.find_nearest(values, max_k, new_response, 0, 0, 0);
print_result(knearest_calc_error(values, response, new_response, train_sidx, is_regression, CV_TRAIN_ERROR),
knearest_calc_error(values, response, new_response, train_sidx, is_regression, CV_TEST_ERROR), 0);
printf("======== RBF SVM =======\n");
//printf("indexes: %d / %d, responses: %d\n", train_sidx->cols, var_idx->cols, values->rows);
CvMySVM svm1;
CvSVMParams params1 = CvSVMParams(CvSVM::C_SVC, CvSVM::RBF,
/*degree*/0, /*gamma*/1, /*coef0*/0, /*C*/1,
/*nu*/0, /*p*/0, /*class_weights*/0,
cvTermCriteria(CV_TERMCRIT_ITER+CV_TERMCRIT_EPS, 1000, FLT_EPSILON));
//svm1.train(values, response, train_sidx, var_idx, params1);
svm1.train_auto(values, response, var_idx, train_sidx, params1);
svm_print_error(&svm1, values, response, response_idx, train_sidx);
printf("======== Linear SVM =======\n");
CvMySVM svm2;
CvSVMParams params2 = CvSVMParams(CvSVM::C_SVC, CvSVM::LINEAR,
/*degree*/0, /*gamma*/1, /*coef0*/0, /*C*/1,
/*nu*/0, /*p*/0, /*class_weights*/0,
cvTermCriteria(CV_TERMCRIT_ITER+CV_TERMCRIT_EPS, 1000, FLT_EPSILON));
//svm2.train(values, response, train_sidx, var_idx, params2);
svm2.train_auto(values, response, var_idx, train_sidx, params2);
svm_print_error(&svm2, values, response, response_idx, train_sidx);
printf("======NEURONAL NETWORK=====\n");
int num_layers = 3;
CvMat layers = cvMat(1, num_layers, CV_32SC1, calloc(1, sizeof(double)*num_layers*1));
cvmSetI(&layers, 0, 0, values->cols-1);
cvmSetI(&layers, 0, 1, num_classes);
cvmSetI(&layers, 0, 2, num_classes);
CvANN_MLP ann(&layers, CvANN_MLP::SIGMOID_SYM, 0.0, 0.0);
CvANN_MLP_TrainParams ann_params;
//ann_params.train_method = CvANN_MLP_TrainParams::BACKPROP;
CvMat ann_response = cvmat_make_boolean_class_columns(response, num_classes);
CvMat values2 = cvmat_remove_column(values, response_idx);
ann.train(&values2, &ann_response, NULL, train_sidx, ann_params, 0x0000);
//ann.train(values, &ann_response, NULL, train_sidx, ann_params, 0x0000);
ann_print_error(&ann, values, num_classes, &ann_response, response, response_idx, train_sidx);
#if 0 /* slow */
printf("======== Polygonal SVM =======\n");
//printf("indexes: %d / %d, responses: %d\n", train_sidx->cols, var_idx->cols, values->rows);
CvMySVM svm3;
CvSVMParams params3 = CvSVMParams(CvSVM::C_SVC, CvSVM::POLY,
/*degree*/2, /*gamma*/1, /*coef0*/0, /*C*/1,
/*nu*/0, /*p*/0, /*class_weights*/0,
cvTermCriteria(CV_TERMCRIT_ITER+CV_TERMCRIT_EPS, 1000, FLT_EPSILON));
//svm3.train(values, response, train_sidx, var_idx, params3);
svm3.train_auto(values, response, var_idx, train_sidx, params3);
svm_print_error(&svm3, values, response, response_idx, train_sidx);
#endif
return 0;
}
JObject JClass::getAnnotation(QString className)
{
JClass ann(className);
return getAnnotation(ann);
}
