基于python的留一法+朴素贝叶斯分类2021-05-13

用python进行简单的朴素贝叶斯分类，这里是用性别、体重两个特征预测性别。新手，仅记录，欢迎指导。

import pandas as pd    #  导入Python的数据处理库pandas,相当于Python里的excel
import numpy as np    
data = pd.read_csv('biyelunwen.csv',encoding='gbk')

# 把data转成一个个的数组,不做这一步直接输data[:,0:2]，就会报错
A=np.array(data)
X=A[:,0:2]
y=A[:,2]

# 导入朴素高斯贝叶斯、LOO模块
from sklearn import naive_bayes
from sklearn.model_selection import LeaveOneOut
loo=LeaveOneOut()
clf=naive_bayes.GaussianNB()

right=0
for train_index,test_index in loo.split(data):
    X_train,X_test=X[train_index],X[test_index]
    y_train,y_test=y[train_index],y[test_index]
    
    clf.fit(X_train,y_train)
    y_test_pred = clf.predict(X_test)
    if y_test_pred==y_test:
        right=right+1
    else:
        right=right+0

# 用X.shape[0]输出X的行数，用X.shape[1]输出X的列数
acc=right/X.shape[0]
print('准确率={:.2f}%'.format(acc*100))

 最后结果：

准确率=80.00%