| 40 | // ---------------------------------------------------------------------------------------- |
| 41 | |
| 42 | int main() |
| 43 | { |
| 44 | // In this example, we have three types of samples: class 0, 1, or 2. That is, each of |
| 45 | // our sample vectors falls into one of three classes. To keep this example very |
| 46 | // simple, each sample vector is zero everywhere except at one place. The non-zero |
| 47 | // dimension of each vector determines the class of the vector. So for example, the |
| 48 | // first element of samples has a class of 1 because samples[0](1) is the only non-zero |
| 49 | // element of samples[0]. |
| 50 | sample_type samp(3); |
| 51 | std::vector<sample_type> samples; |
| 52 | samp = 0,2,0; samples.push_back(samp); |
| 53 | samp = 1,0,0; samples.push_back(samp); |
| 54 | samp = 0,4,0; samples.push_back(samp); |
| 55 | samp = 0,0,3; samples.push_back(samp); |
| 56 | // Since we want to use a machine learning method to learn a 3-class classifier we need |
| 57 | // to record the labels of our samples. Here samples[i] has a class label of labels[i]. |
| 58 | std::vector<int> labels; |
| 59 | labels.push_back(1); |
| 60 | labels.push_back(0); |
| 61 | labels.push_back(1); |
| 62 | labels.push_back(2); |
| 63 | |
| 64 | |
| 65 | // Now that we have some training data we can tell the structural SVM to learn the |
| 66 | // parameters of our 3-class classifier model. The details of this will be explained |
| 67 | // later. For now, just note that it finds the weights (i.e. a vector of real valued |
| 68 | // parameters) such that predict_label(weights, sample) always returns the correct |
| 69 | // label for a sample vector. |
| 70 | column_vector weights = train_three_class_classifier(samples, labels); |
| 71 | |
| 72 | // Print the weights and then evaluate predict_label() on each of our training samples. |
| 73 | // Note that the correct label is predicted for each sample. |
| 74 | cout << weights << endl; |
| 75 | for (unsigned long i = 0; i < samples.size(); ++i) |
| 76 | cout << "predicted label for sample["<<i<<"]: " << predict_label(weights, samples[i]) << endl; |
| 77 | } |
| 78 | |
| 79 | // ---------------------------------------------------------------------------------------- |
| 80 |
nothing calls this directly
no test coverage detected