python手写代码实现朴素贝叶斯算法

好的，下面是 Python 实现朴素贝叶斯算法的代码：

import numpy as np

class NaiveBayes:
    def __init__(self):
        self.classes = None
        self.class_probabilities = None
        self.mean = None
        self.variance = None
    
    def fit(self, X, y):
        # 获取不同的类别
        self.classes = np.unique(y)
        n_classes = len(self.classes)
        
        # 计算每个类别的先验概率
        self.class_probabilities = np.zeros(n_classes)
        for i in range(n_classes):
            self.class_probabilities[i] = np.sum(y == self.classes[i]) / len(y)
        
        # 计算每个特征在每个类别下的均值和方差
        n_features = X.shape[1]
        self.mean = np.zeros((n_classes, n_features))
        self.variance = np.zeros((n_classes, n_features))
        for i in range(n_classes):
            X_i = X[y == self.classes[i]]
            self.mean[i, :] = X_i.mean(axis=0)
            self.variance[i, :] = X_i.var(axis=0)
    
    def predict(self, X):
        # 计算每个类别下的后验概率
        posteriors = np.zeros((len(X), len(self.classes)))
        for i in range(len(self.classes)):
            class_prior = np.log(self.class_probabilities[i])
            class_mean = self.mean[i]
            class_variance = self.variance[i]
            numerator = np.exp(-((X - class_mean) ** 2) / (2 * class_variance))
            denominator = np.sqrt(2 * np.pi * class_variance)
            posterior = class_prior + np.sum(np.log(numerator / denominator), axis=1)
            posteriors[:, i] = posterior
        
        # 返回概率最大的类别
        return self.classes[np.argmax(posteriors, axis=1)]

其中 `fit()` 方法用于训练模型，`predict()` 方法用于预测新数据的类别。在 `fit()` 方法中，我们首先计算了每个类别的先验概率，然后计算了每个特征在每个类别下的均值和方差。在 `predict()` 方法中，我们计算了每个类别下的后验概率，并返回概率最大的类别。