CLabels* CKNN::classify_NN()
{
ASSERT(distance);
ASSERT(num_classes>0);
int32_t num_lab = distance->get_num_vec_rhs();
ASSERT(num_lab);
CLabels* output = new CLabels(num_lab);
float64_t* distances = new float64_t[num_train_labels];
ASSERT(distances);
SG_INFO("%d test examples\n", num_lab);
CSignal::clear_cancel();
// for each test example
for (int32_t i=0; i<num_lab && (!CSignal::cancel_computations()); i++)
{
SG_PROGRESS(i,0,num_lab);
// get distances from i-th test example to 0..num_train_labels-1 train examples
distances_lhs(distances,0,num_train_labels-1,i);
int32_t j;
// assuming 0th train examples as nearest to i-th test example
int32_t out_idx = 0;
float64_t min_dist = distances[0];
// searching for nearest neighbor by comparing distances
for (j=0; j<num_train_labels; j++)
{
if (distances[j]<min_dist)
{
min_dist = distances[j];
out_idx = j;
}
}
// label i-th test example with label of nearest neighbor with out_idx index
output->set_label(i,train_labels[out_idx]+min_label);
}
delete [] distances;
return output;
}
CLabels* CGaussianNaiveBayes::apply()
{
// init number of vectors
int32_t num_vectors = m_features->get_num_vectors();
// init result labels
CLabels* result = new CLabels(num_vectors);
// classify each example of data
SG_PROGRESS(0, 0, num_vectors);
for (int i = 0; i < num_vectors; i++)
{
result->set_label(i,apply(i));
SG_PROGRESS(i + 1, 0, num_vectors);
}
SG_DONE();
return result;
};
