Function perceptron_demo

examples/machine_learning/perceptron.cpp:73–118

Demo of one vs all logistic regression

// Demo of one vs all logistic regression
int perceptron_demo(bool console, int perc) {
    array train_images, train_targets;
    array test_images, test_targets;
    int num_train, num_test, num_classes;

    // Load mnist data
    float frac = (float)(perc) / 100.0;
    setup_mnist<true>(&num_classes, &num_train, &num_test, train_images,
                      test_images, train_targets, test_targets, frac);

    // Reshape images into feature vectors
    int feature_length = train_images.elements() / num_train;
    array train_feats  = moddims(train_images, feature_length, num_train).T();
    array test_feats   = moddims(test_images, feature_length, num_test).T();

    train_targets = train_targets.T();
    test_targets  = test_targets.T();

    // Add a bias that is always 1
    train_feats = join(1, constant(1, num_train, 1), train_feats);
    test_feats  = join(1, constant(1, num_test, 1), test_feats);

    // Train logistic regression parameters
    array Weights = train(train_feats, train_targets, 0.1, 0.01, 1000, true);

    // Predict the results
    array train_outputs = predict(train_feats, Weights);
    array test_outputs  = predict(test_feats, Weights);

    printf("Accuracy on training data: %2.2f\n",
           accuracy(train_outputs, train_targets));

    printf("Accuracy on testing data: %2.2f\n",
           accuracy(test_outputs, test_targets));

    benchmark_perceptron(train_feats, train_targets, test_feats);

    if (!console) {
        test_outputs = test_outputs.T();
        test_targets = test_targets.T();
        // Get 20 random test images.
        display_results<true>(test_images, test_outputs, test_targets, 20);
    }

    return 0;
}

int main(int argc, char **argv) {
    int device   = argc > 1 ? atoi(argv[1]) : 0;