用c++写一个鸢尾花k近邻算法

在C++中，使用鸢尾花数据集（Iris dataset）实现K近邻（K-Nearest Neighbors, KNN）算法的基本步骤如下：

1. **导入所需库**：

#include <iostream>
#include <fstream>
#include <vector>
#include <string>
#include <cmath>
#include <algorithm>

2. **读取数据**：

struct IrisSpecies {
    double sepal_length;
    double sepal_width;
    double petal_length;
    double petal_width;
    std::string species; // 标签，如"Iris-setosa"等
};

std::vector<IrisSpecies> load_data(const std::string& filename) {
    std::ifstream file(filename);
    std::vector<IrisSpecies> data;
    if (file.is_open()) {
        double row[4];
        while (file >> row[0] >> row[1] >> row[2] >> row[3]) {
            data.push_back({row[0], row[1], row[2], row[3], "Iris-" + std::to_string(row[3])});
        }
        file.close();
    } else {
        std::cerr << "Error opening file." << std::endl;
    }
    return data;
}

3. **计算距离函数**（这里用欧氏距离为例）：

double euclidean_distance(IrisSpecies a, IrisSpecies b) {
    double sum = 0.0;
    for (int i = 0; i < 4; ++i)
        sum += pow(a[i] - b[i], 2);
    return sqrt(sum);
}

4. **KNN分类**：

std::string knn_predict(const std::vector<IrisSpecies>& train_data, const IrisSpecies& test_data, int k) {
    std::vector<std::pair<double, std::string>> distances(train_data.size(), {std::numeric_limits<double>::max(), ""});

    // 计算测试样本与所有训练样本的距离并排序
    for (const auto& d : train_data) {
        double dist = euclidean_distance(test_data, d);
        distances.emplace_back(dist, d.species);
        std::sort(distances.begin() + 1, distances.end());
    }

    // 返回前k个最近邻居中出现最多的类别
    std::string most_common = "";
    int max_count = 0;
    for (int i = 0; i < k; ++i) {
        int count = std::count_if(distances.begin(), distances.begin() + k, [d = distances[i].second](const auto& p) { return p.second == d; });
        if (count > max_count) {
            max_count = count;
            most_common = d;
        }
    }

    return most_common;
}

5. **主程序示例**：

int main() {
    std::vector<IrisSpecies> train_data = load_data("iris.data");
    // ... (实际应用中划分训练集和测试集)

    IrisSpecies test_data; // 假设已经填充好测试样本
    int k = 3; // 指定K值
    std::string prediction = knn_predict(train_data, test_data, k);
    std::cout << "Predicted species: " << prediction << std::endl;

    return 0;
}