用python写一个找好瓜的朴素贝叶斯代码

好的，以下是一个简单的朴素贝叶斯分类器的代码，用于判断西瓜是否好瓜。

import pandas as pd
import numpy as np

class NaiveBayes:
    
    def __init__(self):
        self.prior = {}
        self.conditional_probabilities = {}
        self.labels = []
        
    def fit(self, X_train, y_train):
        # 计算先验概率
        n_samples = len(y_train)
        for label in np.unique(y_train):
            self.prior[label] = len(y_train[y_train == label]) / n_samples
        
        # 计算条件概率
        self.labels = np.unique(y_train)
        for feature_idx in range(X_train.shape[1]):
            for label in self.labels:
                X_train_c = X_train[y_train == label]
                self.conditional_probabilities[(feature_idx, label)] = {}
                for feature_val in np.unique(X_train[:, feature_idx]):
                    self.conditional_probabilities[(feature_idx, label)][feature_val] = \
                        len(X_train_c[X_train_c[:, feature_idx] == feature_val]) / len(X_train_c)
                        
    def predict(self, X_test):
        y_pred = []
        for sample in X_test:
            probabilities = {}
            for label in self.labels:
                probabilities[label] = self.prior[label]
                for feature_idx, feature_val in enumerate(sample):
                    if (feature_idx, label) in self.conditional_probabilities:
                        probabilities[label] *= self.conditional_probabilities[(feature_idx, label)][feature_val]
                y_pred.append(max(probabilities, key=probabilities.get))
        return y_pred

这个朴素贝叶斯分类器的实现假设特征之间互相独立，因此它被称为朴素贝叶斯分类器。在这个例子中，我们使用了一个具有两个特征的西瓜数据集。您可以根据需要更改数据集和特征数量。