(4)李航《统计学习方法》基于Python实现——朴素贝叶斯

1&＃xff1a;高斯模型下的朴素贝叶斯

import numpy as np
import pandas as pdfrom sklearn.datasets import load_iris
from sklearn.model_selection import train_test_splitfrom collections import Counter
import math# 特征假设是高斯分布# datadef create_data():iris &＃61; load_iris()df &＃61; pd.DataFrame(iris.data,columns &＃61; iris.feature_names)df[&＃39;label&＃39;] &＃61; iris.targetdf.columns &＃61; [&＃39;sepal length&＃39;, &＃39;sepal width&＃39;, &＃39;petal length&＃39;, &＃39;petal width&＃39;, &＃39;label&＃39;]data &＃61; np.array(df.iloc[:100,:])return data[:,:-1],data[:,-1]X,y &＃61; create_data()
X_train, X_test, y_train, y_test &＃61; train_test_split(X,y, test_size&＃61;0.3)X_test[0], y_test[0]


class NaiveBayes:def __init__(self):self.model &＃61; None# 数学期望
# &＃64;staticmethoddef mean(X):return sum(X) / float(len(X))# 标准差&＃xff08;方差&＃xff09;def stdev(self, X):avg &＃61; self.mean(X)return math.sqrt(sum([pow(x-avg, 2) for x in X]) / float(len(X)))# 概率密度函数def gaussian_probability(self, x, mean, stdev):exponent &＃61; math.exp(-(math.pow(x-mean,2)/(2*math.pow(stdev,2))))return (1 / (math.sqrt(2*math.pi) * stdev)) * exponent# 处理X_traindef summarize(self, train_data):summaries &＃61; [(self.mean(i), self.stdev(i)) for i in zip(*train_data)]return summaries# 分类别求出数学期望和标准差def fit(self, X, y):labels &＃61; list(set(y))data &＃61; {label:[] for label in labels}for f, label in zip(X, y):data[label].append(f)self.model &＃61; {label: self.summarize(value) for label, value in data.items()}return &＃39;gaussianNB train done!&＃39;# 计算概率def calculate_probabilities(self, input_data):# summaries:{0.0: [(5.0, 0.37),(3.42, 0.40)], 1.0: [(5.8, 0.449),(2.7, 0.27)]}# input_data:[1.1, 2.2]probabilities &＃61; {}for label, value in self.model.items():probabilities[label] &＃61; 1for i in range(len(value)):mean, stdev &＃61; value[i]probabilities[label] *&＃61; self.gaussian_probability(input_data[i], mean, stdev)return probabilities# 类别def predict(self, X_test):# {0.0: 2.9680340789325763e-27, 1.0: 3.5749783019849535e-26}label &＃61; sorted(self.calculate_probabilities(X_test).items(), key&＃61;lambda x: x[-1])[-1][0]return labeldef score(self, X_test, y_test):right &＃61; 0for X, y in zip(X_test, y_test):label &＃61; self.predict(X)if label &＃61;&＃61; y:right &＃43;&＃61; 1return right / float(len(X_test))


主函数

model &＃61; NaiveBayes()
model.fit(X_train, y_train)print(model.predict([4.4, 3.2, 1.3, 0.2]))model.score(X_test, y_test)


2&＃xff1a;sklearn贝叶斯实现

from sklearn.naive_bayes import GaussianNB
clf &＃61; GaussianNB()
clf.fit(X_train, y_train)clf.score(X_test, y_test)
clf.predict([[4.4, 3.2, 1.3, 0.2]])