用c++写一个判断西瓜好坏的决策树代码，并附上注释和训练数据

好的，下面是C++代码实现：

#include <iostream>
using namespace std;

// 定义西瓜的属性结构体
struct Watermelon {
    int color; // 颜色，0代表青绿，1代表乌黑，2代表浅白
    float root; // 根蒂，0代表蜷缩，1代表稍蜷，2代表硬挺
    float sound; // 敲声，0代表沉闷，1代表浊响，2代表清脆
    int good; // 是否好瓜，0代表否，1代表是
};

// 定义训练数据
Watermelon train_data[] = {
    {0, 0, 0, 0},
    {0, 0, 1, 0},
    {1, 0, 1, 1},
    {2, 0, 1, 1},
    {2, 1, 0, 1},
    {2, 2, 1, 0},
    {1, 1, 0, 1},
    {0, 2, 1, 0},
    {0, 1, 0, 1},
    {2, 1, 1, 1},
    {0, 1, 1, 1},
    {1, 0, 0, 0},
    {1, 1, 1, 1},
    {2, 1, 0, 0}
};

// 计算数据集中好瓜和坏瓜的个数
void count_good_bad(Watermelon *data, int len, int &good, int &bad) {
    good = 0;
    bad = 0;
    for (int i = 0; i < len; i++) {
        if (data[i].good == 1) {
            good++;
        } else {
            bad++;
        }
    }
}

// 计算数据集中给定属性的某个值的好瓜和坏瓜的个数
void count_good_bad_by_attr(Watermelon *data, int len, int attr, float value, int &good, int &bad) {
    good = 0;
    bad = 0;
    for (int i = 0; i < len; i++) {
        if (data[i].good == 1 && data[i].color == value) {
            good++;
        } else if (data[i].good == 0 && data[i].color == value) {
            bad++;
        }
    }
}

// 计算数据集中给定两个属性的某个值的好瓜和坏瓜的个数
void count_good_bad_by_attr(Watermelon *data, int len, int attr1, float value1, int attr2, float value2, int &good, int &bad) {
    good = 0;
    bad = 0;
    for (int i = 0; i < len; i++) {
        if (data[i].good == 1 && data[i].color == value1 && data[i].root == value2) {
            good++;
        } else if (data[i].good == 0 && data[i].color == value1 && data[i].root == value2) {
            bad++;
        }
    }
}

// 计算数据集中给定三个属性的某个值的好瓜和坏瓜的个数
void count_good_bad_by_attr(Watermelon *data, int len, int attr1, float value1, int attr2, float value2, int attr3, float value3, int &good, int &bad) {
    good = 0;
    bad = 0;
    for (int i = 0; i < len; i++) {
        if (data[i].good == 1 && data[i].color == value1 && data[i].root == value2 && data[i].sound == value3) {
            good++;
        } else if (data[i].good == 0 && data[i].color == value1 && data[i].root == value2 && data[i].sound == value3) {
            bad++;
        }
    }
}

// 训练决策树
void train_decision_tree(Watermelon *data, int len, int depth) {
    int good, bad;
    count_good_bad(data, len, good, bad);
    if (good == 0 || bad == 0) {
        if (good == 0) {
            cout << "这是个坏瓜" << endl;
        } else {
            cout << "这是个好瓜" << endl;
        }
        return;
    }
    if (depth == 0) {
        if (good > bad) {
            cout << "这是个好瓜" << endl;
        } else {
            cout << "这是个坏瓜" << endl;
        }
        return;
    }
    // 选择最优属性
    float info_gain, max_info_gain = -1;
    int best_attr = -1;
    for (int i = 0; i < 3; i++) {
        int num_values;
        float values[3];
        if (i == 0) {
            num_values = 3;
            values[0] = 0;
            values[1] = 1;
            values[2] = 2;
        } else {
            num_values = 2;
            values[0] = 0;
            values[1] = 1;
        }
        for (int j = 0; j < num_values; j++) {
            int temp_good, temp_bad;
            if (i == 0) {
                count_good_bad_by_attr(data, len, i, values[j], temp_good, temp_bad);
            } else if (i == 1) {
                count_good_bad_by_attr(data, len, i, values[j], 2, 0, temp_good, temp_bad);
            } else {
                count_good_bad_by_attr(data, len, i, values[j], 0, 0, 1, 1, temp_good, temp_bad);
            }
            float p_good = (float)temp_good / (float)(temp_good + temp_bad);
            float p_bad = (float)temp_bad / (float)(temp_good + temp_bad);
            float temp_info_gain = -p_good * log2(p_good) - p_bad * log2(p_bad);
            if (temp_info_gain > max_info_gain) {
                max_info_gain = temp_info_gain;
                best_attr = i;
            }
        }
    }
    // 根据最优属性划分子数据集并递归训练
    if (best_attr == 0) {
        int num_values = 3;
        float values[3];
        values[0] = 0;
        values[1] = 1;
        values[2] = 2;
        for (int j = 0; j < num_values; j++) {
            int temp_good, temp_bad;
            count_good_bad_by_attr(data, len, best_attr, values[j], temp_good, temp_bad);
            cout << "如果颜色是" << values[j] << "：";
            if (temp_good == 0 || temp_bad == 0) {
                if (temp_good == 0) {
                    cout << "这是个坏瓜" << endl;
                } else {
                    cout << "这是个好瓜" << endl;
                }
            } else {
                Watermelon *sub_data = new Watermelon[temp_good + temp_bad];
                int index = 0;
                for (int k = 0; k < len; k++) {
                    if (data[k].color == values[j]) {
                        sub_data[index++] = data[k];
                    }
                }
                train_decision_tree(sub_data, index, depth - 1);
            }
        }
    } else if (best_attr == 1) {
        int num_values = 2;
        float values[2];
        values[0] = 0;
        values[1] = 1;
        for (int j = 0; j < num_values; j++) {
            int temp_good, temp_bad;
            count_good_bad_by_attr(data, len, best_attr, values[j], 2, 0, temp_good, temp_bad);
            cout << "如果根蒂是" << values[j] << "：";
            if (temp_good == 0 || temp_bad == 0) {
                if (temp_good == 0) {
                    cout << "这是个坏瓜" << endl;
                } else {
                    cout << "这是个好瓜" << endl;
                }
            } else {
                Watermelon *sub_data = new Watermelon[temp_good + temp_bad];
                int index = 0;
                for (int k = 0; k < len; k++) {
                    if (data[k].root == values[j]) {
                        sub_data[index++] = data[k];
                    }
                }
                train_decision_tree(sub_data, index, depth - 1);
            }
        }
    } else {
        int num_values = 2;
        float values[2];
        values[0] = 0;
        values[1] = 1;
        for (int j = 0; j < num_values; j++) {
            int temp_good, temp_bad;
            count_good_bad_by_attr(data, len, best_attr, values[j], 0, 0, 1, 1, temp_good, temp_bad);
            cout << "如果敲声是" << values[j] << "：";
            if (temp_good == 0 || temp_bad == 0) {
                if (temp_good == 0) {
                    cout << "这是个坏瓜" << endl;
                } else {
                    cout << "这是个好瓜" << endl;
                }
            } else {
                Watermelon *sub_data = new Watermelon[temp_good + temp_bad];
                int index = 0;
                for (int k = 0; k < len; k++) {
                    if (data[k].sound == values[j]) {
                        sub_data[index++] = data[k];
                    }
                }
                train_decision_tree(sub_data, index, depth - 1);
            }
        }
    }
}

int main() {
    train_decision_tree(train_data, 14, 3);
    return 0;
}

训练数据如下：

颜色 根蒂 敲声 是否好瓜
0     0    0    0
0     0    1    0
1     0    1    1
2     0    1    1
2     1    0    1
2     2    1    0
1     1    0    1
0     2    1    0
0     1    0    1
2     1    1    1
0     1    1    1
1     0    0    0
1     1    1    1
2     1    0    0

其中，颜色属性的值为0代表青绿，1代表乌黑，2代表浅白；根蒂属性的值为0代表蜷缩，1代表稍蜷，2代表硬挺；敲声属性的值为0代表沉闷，1代表浊响，2代表清脆；是否好瓜属性的值为0代表否，1代表是。