用python代码实现朴贝叶斯方法算法

以下是一个简单的朴素贝叶斯分类器的Python代码实现：

import numpy as np

class NaiveBayes:
    def __init__(self):
        self.classes = None
        self.class_prob = None
        self.feature_prob = None

    def fit(self, X, y):
        self.classes = np.unique(y)
        num_classes = len(self.classes)
        num_features = X.shape[1]

        self.class_prob = np.zeros(num_classes)
        self.feature_prob = np.zeros((num_classes, num_features))

        for i, c in enumerate(self.classes):
            X_c = X[c == y]
            self.class_prob[i] = X_c.shape[0] / X.shape[0]
            self.feature_prob[i, :] = X_c.sum(axis=0) / X_c.sum()

    def predict(self, X):
        y_pred = []
        for x in X:
            probabilities = []
            for i, c in enumerate(self.classes):
                probability = np.log(self.class_prob[i])
                for j in range(len(x)):
                    probability += np.log(self.feature_prob[i, j]) if x[j] == 1 else np.log(1 - self.feature_prob[i, j])
                probabilities.append(probability)
            idx = np.argmax(probabilities)
            y_pred.append(self.classes[idx])
        return y_pred

这个朴素贝叶斯分类器使用的是二进制特征，并假设特征之间条件独立。`fit`函数用于拟合模型，计算每个类别的先验概率和每个特征的条件概率。`predict`函数用于预测新数据点的类别。