逻辑回归

通过逻辑回归方式，对于二次边界的数据集进行分类

---

---

前言

数据集为二维的数据样本，带标签的数据分为两类，之间的划分边界是一个多项式边界，通过逻辑回归建立模型，通过系数转变来还原二次边界。

---

一、导包

主要是通过机器学习的LogisticRegression模型进行分类预测。

import pandas as pd
import numpy as np
from matplotlib import pyplot as plt
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

二、数据导入并进行可视化分析

ata = pd.read_csv('task2_data.csv')
#print(data.head())

y = data.loc[:,'y']
mask = y==1

x = data.drop(['y'],axis=1)

# fig = plt.figure()
# abnormal = plt.scatter(data.loc[:,'pay1'][mask],data.loc[:,'pay2'][mask])
# normal = plt.scatter(data.loc[:,'pay1'][~mask],data.loc[:,'pay2'][~mask])
# plt.xlabel('pay1')
# plt.ylabel('pay2')
# plt.title('pay1-pay2')
# plt.legend((abnormal,normal),('abnormal','normal'))
# plt.show()

三、数据处理

x1 = data.loc[:,'pay1']
x2 = data.loc[:,'pay2']
x1_2 = x1*x1
x2_2 = x2*x2
x1_x2 = x1*x2
x_new = {'x1':x1,'x2':x2,'x3':x1_2,'x4':x2_2,'x5':x1_x2}
x_new = pd.DataFrame(x_new)

四、模型建立与预测

model = LogisticRegression()
model.fit(x_new,y)
y_predict = model.predict(x_new)
accuracy = accuracy_score(y,y_predict)
#print(accuracy)

五、系数提取、边界还原

tanta0 = model.intercept_[0]
tanta1 = model.coef_[0][0]
tanta2 = model.coef_[0][1]
tanta3 = model.coef_[0][2]
tanta4 = model.coef_[0][3]
tanta5 = model.coef_[0][4]

x1_new = x1.sort_values()
a = tanta4
b = tanta2+tanta5*x1_new
c = tanta0+tanta1*x1_new+tanta3*x1_new*x1_new

y_new = (-b+np.sqrt(b*b-4*a*c))/(2*a)

六、结果边界可视化

fig = plt.figure()
abnormal = plt.scatter(data.loc[:,'pay1'][mask],data.loc[:,'pay2'][mask])
normal = plt.scatter(data.loc[:,'pay1'][~mask],data.loc[:,'pay2'][~mask])
plt.plot(x1_new,y_new)
plt.xlabel('pay1')
plt.ylabel('pay2')
plt.title('pay1-pay2')
plt.legend((abnormal,normal),('abnormal','normal'))
plt.savefig('result.png')
plt.show()

---

总结

学习了利用逻辑回归搭建二分类模型，实现了多项式边界的分类任务。