深度学习笔记二：单层感知器学习与实践

单层感知器的理论知识上一节已经介绍过，这一节主要结合代码实现单层感知器的学习过程

这里我们用python3实现，用到numpy库

import numpy as np
import matplotlib.pyplot as plt




X = np.array([[1,3,3],[1,4,3],[1,1,1]])


Y = np.array([1,1,-1])


W = (np.random.random(3)-0.5)*2


print(W)
#学习率
lr = 0.11


n = 0


o = 0


def update():
    global X,Y,W,lr,n
    n += 1
    o = np.sign(np.dot(X,W.T)/X.shape[0])
    W_C = lr*((Y-o.T).dot(X))
    W = W + W_C


def predict(X):
    global W
    y = np.sign(np.dot(X,W.T))
    return y


for i in range(100):
    update()
    print(W)
    print(n)
    o = np.sign((np.dot(X,W.T))/X.shape[0])
    if(o == Y.T).all():
        print('finish')
        break
    
for e in X:
    print(predict(e))


x1 = [3,4]
y1 = [3,3]
x2 = [1]
y2 = [1]
xdata = np.linspace(0,5)


k = -W[1]/W[2]
d = -W[0]/W[2]


plt.figure()
plt.plot(xdata,xdata*k+d,'r')
plt.plot(x1,y1,'bo')
plt.plot(x2,y2,'yo')
plt.show()