用python写一个寻找好瓜的朴素贝叶斯代码

朴素贝叶斯是一种常用的分类算法，在数据挖掘和机器学习领域得到了广泛应用。下面是一个用 Python 实现朴素贝叶斯算法来判断西瓜好坏的示例代码：

import pandas as pd
import numpy as np

class NaiveBayes:
    def __init__(self):
        self.prior = {}  # 存储先验概率
        self.conditional_probability = {}  # 存储条件概率

    def fit(self, X, y):
        self.X = X
        self.y = y
        self.classes = np.unique(y)  # 获取类别
        self.feature_names = X.columns.tolist()  # 获取特征名
        for c in self.classes:
            self.prior[c] = len(y[y == c]) / len(y)  # 计算先验概率
            self.conditional_probability[c] = {}  # 初始化条件概率字典
            for feature in self.feature_names:
                feature_values = np.unique(X[feature])  # 获取特征值
                for value in feature_values:
                    # 计算条件概率
                    self.conditional_probability[c][(feature, value)] = \
                        len(X[(y == c) & (X[feature] == value)]) / len(X[y == c])

    def predict(self, X):
        y_pred = []
        for index, row in X.iterrows():
            max_prob = -1
            max_class = None
            for c in self.classes:
                # 计算后验概率
                p = self.prior[c]
                for feature in self.feature_names:
                    value = row[feature]
                    p *= self.conditional_probability[c][(feature, value)]
                if p > max_prob:
                    max_prob = p
                    max_class = c
            y_pred.append(max_class)
        return y_pred

# 加载西瓜数据集
watermelon = pd.read_csv('watermelon.csv')
X = watermelon.drop(columns=['编号', '好瓜'])
y = watermelon['好瓜']

# 划分训练集和测试集
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# 训练模型
nb = NaiveBayes()
nb.fit(X_train, y_train)

# 预测测试集
y_pred = nb.predict(X_test)
print('测试集准确率:', sum(y_pred == y_test) / len(y_test))

参考资料：

- [使用 Python 实现朴素贝叶斯算法](https://zhuanlan.zhihu.com/p/33544836)