周志华机器学习西瓜书第三章习题3.3编程实现对率回归

# coding&＃61;utf-8
import tensorflow as tf
from numpy import *
import numpy as np
import matplotlib.pyplot as pltdata &＃61; mat([[0.697,0.460,1],[0.774,0.376,1],[0.634,0.264,1],[0.608,0.318,1],[0.556,0.215,1],[0.403,0.237,1],[0.481,0.149,1],[0.437,0.211,1],[0.666,0.091,0],[0.243,0.267,0],[0.245,0.057,0],[0.343,0.099,0],[0.639,0.161,0],[0.657,0.198,0],[0.360,0.370,0],[0.593,0.042,0],[0.719,0.103,0]])x_train&＃61;data[:,0:2].astype(float32)
y_train&＃61;data[:,2].astype(float32)
y_train &＃61; mat(y_train)
w&＃61;tf.Variable(tf.zeros([2,1]))
b&＃61;tf.Variable(tf.zeros([1,1]))
y&＃61;1/(1&＃43;tf.exp(-(tf.matmul(x_train,w)&＃43;b)))
loss &＃61; tf.reduce_mean(- y_train.reshape(-1, 1) * tf.log(y) - (1 - y_train.reshape(-1, 1)) * tf.log(1 - y))
train&＃61;tf.train.GradientDescentOptimizer(0.5).minimize(loss)
init&＃61;tf.global_variables_initializer()
sess&＃61;tf.Session()
sess.run(init)
for i in range(1000):sess.run(train)
print(i,sess.run(w).flatten(),sess.run(b).flatten())

#画图
X0 &＃61; array(data[:8])#选择训练集中为好瓜的样本
X1 &＃61; array(data[8:])#选择训练集中为坏瓜的样本
ax &＃61; plt.subplot(111)
ax.scatter(X0[:,0],X0[:,1],c&＃61;&＃39;g&＃39;,label&＃61;&＃39;&＃43;&＃39;)#选择好瓜样本的第一列和第二列 画出散点图
ax.scatter(X1[:,0],X1[:,1],c&＃61;&＃39;b&＃39;,label&＃61;&＃39;-&＃39;)#选择坏瓜样本的第一列和第二列 画出散点图
x1 &＃61; arange(-0.9,1.0,0.1)
y1 &＃61; (-b-w[0]*x1)/w[1]
plt.sca(ax)
plt.plot(x1,sess.run(y1).flatten())
plt.legend()#设置图例
plt.show()