【统计学习方法】 第2章 感知机代码实现（二）

import numpy as npimport randomimport matplotlib.pyplot as plt
# 定义一个函数def random_point_nearby_line(weight , bias , size = 10):    x_point = np.linspace(-1, 1, size)[:,np.newaxis]    noise = np.random.normal(0, 0.5, x_point.shape)    y_point = weight * x_point + bias + noise    input_arr = np.hstack((x_point, y_point))    return input_arr

# 直线的真正参数real_weight = 2real_bias = 1size = 100

# 生成输入的数据input_point = random_point_nearby_line(real_weight, real_bias, size)
# 给数据打标签，在直线之上还是直线之下，above=1,below=-1label = np.sign(input_point[:,1] - (input_point[:,0] * real_weight + real_bias)).reshape((size, 1))

# 初始化参数weight = [0, 0]bias = 0learning_rate = 0.1train_num = 1000

train_data = list(zip(input_point, label))

for i in range(train_num):    train = random.choice(train_data)    [x1,x2],y_label = train;    y_predict = np.sign(weight[0]*x1 + weight[1]*x2 + bias)    # print("train data:x:(%d, %d) y:%d ==>y_predict:%d" %(x1,x2,y_label,y_predict))    if y_label*y_predict<=0:        weight[0] = weight[0] + learning_rate*y_label*x1        weight[1] = weight[1] + learning_rate*y_label*x2        bias = bias + learning_rate*y_label        print("update weight and bias:")        print(weight[0], weight[1], bias)print("stop training :")print(weight[0], weight[1], bias)

stop training :[-1.28282828] [0.61316304] [-0.6]

for i in range(len(input_point)):
    if label[i] == 1:        plt.plot(input_point[i][0], input_point[i][1], 'ro')    else:        plt.plot(input_point[i][0], input_point[i][1], 'bo')#plt.plot()x_1 = []x_2 = []for i in range(-1,3):    x_1.append(i)    x_2.append((-weight[0]*i-bias)/weight[1])plt.plot(x_1,x_2)plt.show()