OpenCV SVM train()がうまく機能しない

前提・実現したいこと

OpenCVでSVMを利用するため，サンプルコードを実行して動作を勉強したいと考えています．

発生している問題・エラーメッセージ

古いコードだったので色々修正を加えて動作する段階までこぎつけたものの，train()の部分がうまく動作してくれません．
分類結果の出力の形式に問題があると言っているように見えるのですが，具体的にどこを修正すれば良いかがわかりませんでした．

OpenCV(4.5.3) Error: Bad argument (in the case of classification problem the responses must be categorical; either specify varType when creating TrainData, or pass integer responses) in cv::ml::SVMImpl::train, file C:\build\master_winpack-build-win64-vc15\opencv\modules\ml\src\svm.cpp, line 1630
OpenCV: terminate handler is called! The last OpenCV error is:
OpenCV(4.5.3) Error: Bad argument (in the case of classification problem the responses must be categorical; either specify varType when creating TrainData, or pass integer responses) in cv::ml::SVMImpl::train, file C:\build\master_winpack-build-win64-vc15\opencv\modules\ml\src\svm.cpp, line 1630

該当のソースコード

Cpp
#include <iostream>
#include <fstream>
#include <opencv2/opencv.hpp>
#include <opencv2/core/core.hpp>
#include <opencv2/features2d/features2d.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/xfeatures2d/nonfree.hpp>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/ml/ml.hpp>
#include <stdio.h>
#include <stdlib.h>
#include <fstream>
#include <iostream>


#pragma warning(disable:4996)

using namespace cv;
using namespace std;

const int WINDOW_HEIGHT = 512;
const int WINDOW_WIDTH = 512;
const int TRAINING_DATA_NUM = 200;

inline int labelOf(double x, double y) {
    // return (x - 0.5) * (x - 0.5) + (y - 0.5) * (y - 0.5) <= 0.1 ? 1 : -1;
    // return y >= sin(10 * x) ? 1 : -1;
    return y >= 10 * x * x ? 1 : -1;
}

int main() {
    // Data for visual representation
    Mat image = Mat::zeros(WINDOW_HEIGHT, WINDOW_WIDTH, CV_8UC3);

    // Set up training data
    Mat trainingData(TRAINING_DATA_NUM, 2, CV_32FC1);
    Mat labels(TRAINING_DATA_NUM, 1, CV_32FC1);

    for (int i = 0; i < TRAINING_DATA_NUM; i++) {
        float x = 1. * rand() / RAND_MAX;
        float y = 1. * rand() / RAND_MAX;

        trainingData.at<float>(i, 0) = x;
        trainingData.at<float>(i, 1) = y;

        labels.at<float>(i, 0) = labelOf(x, y);
    }

    
    // Set up SVM's parameters
    Ptr<ml::SVM> params = ml::SVM::create();
    
    
    params->setType(ml::SVM::C_SVC);
    params->setKernel(ml::SVM::RBF);
    params->setDegree(3);
    params->setGamma(3);
    //params->setCoef0(0);
    params->setC(0.1);
    //params->setP(0.1);
    //params->setNu(0.5);
    params->setTermCriteria(TermCriteria(TermCriteria::MAX_ITER, 100, 1e-6));
    
    // Train the SVM
    params->train(trainingData, ml::ROW_SAMPLE, labels);
    

    // Show the decision regions given by the SVM
    Vec3b red(0, 0, 128), blue(128, 0, 0);
    for (int i = 0; i < image.rows; ++i) {
        for (int j = 0; j < image.cols; ++j)
        {
            Mat sampleMat =
                (Mat_<float>(1, 2) << 1. * j / WINDOW_HEIGHT, 1. * i / WINDOW_WIDTH);
            float response = params->predict(sampleMat);
            if (response == 1)
                image.at<Vec3b>(i, j) = red;
            else if (response == -1)
                image.at<Vec3b>(i, j) = blue;
        }
    }

    // Show the training data
    for (int i = 0; i < TRAINING_DATA_NUM; i++) {

        if (labels.at<float>(i, 0) == 1)
            circle(image, Point(trainingData.at<float>(i, 0) * WINDOW_HEIGHT, trainingData.at<float>(i, 1) * WINDOW_WIDTH), 3, Scalar(0, 0, 255), -1, 8);
        else
            circle(image, Point(trainingData.at<float>(i, 0) * WINDOW_HEIGHT, trainingData.at<float>(i, 1) * WINDOW_WIDTH), 3, Scalar(255, 0, 0), -1, 8);
    }

    // Show support vectors
    imshow("Kernel SVM", image);

    waitKey(0);
}