画图代码参考的帖子:~~~~~~点这~~~~~~ 还有这个 ~~~~~~摁这~~~~~~
渲染分类边界参考这个帖子:~~~~~~~~戳这~~~~~~~
一个二分类任务,根据年龄和薪资估计其是否会买车,
数据集有两个特征,Age 和 Salary
标签是 Purchased ,两个取值(0或1)
横轴 Age, 纵轴 Salary,把这些离散的点画出来,并根据标签取值上色
环境 jupyter, python 3.7
# This tells Jupyter to set up Matplotlib so it uses Jupyter’s own backend.
%matplotlib inline
import matplotlib.pyplot as plt
# 根据训练集标签 y 的取值确定所画点的颜色
# y_train 是个 array,里面只有0,1两种取值
# X_train 也是 array 有两列,X_train[i,0] 代表第 i 行,第 0 列, X_train[i][0] 也一样
# scatter 参数中,lw:点宽,s:不知道,color:'r'是红色, marker:'^'代表三角形
for i in range(len(y_train)):
if y_train[i] == 0:
s1 = plt.scatter(X_train[i,0], X_train[i,1], s=50, lw=3, color = 'r')
elif y_train[i] == 1:
s2 = plt.scatter(X_train[i,0], X_train[i,1], s=50, lw=3, color = 'g',marker='^')
plt.xlabel('Age')
plt.ylabel('Estimated Salary')
plt.title(" Logistic Regression (Training Set)")
# ('0','1') 代表与 (s1, s2) 对应的 label
# 也可写成 ('Purchased', 'UnPurchased')
plt.legend((s1,s2),('0','1') ,loc = 'best')