BP神经网络

import numpy as np
import matplotlib.pyplot as plt
from math import sqrt
from sklearn import linear_model
import random


def create_data():
    X=np.array(
        [
        [1,0,0],
        [1,0,1],
        [1,1,0],
        [1,1,1]
            ])
    Y= np.array([[0,1,1,0]] )
    return X,Y

def sigmoid(x):
    return 1/(1+np.exp(-x))

def dsigmoid(x):
    return x*(1-x)

def update_ws(X,Y,W,V,lr=0.01):

    L1 =sigmoid(np.dot(X,V))##隐藏层输出，等于4*4
    L2 =sigmoid(np.dot(L1,W))##输出层输出，等于4*1
    
    
    L2_delta = (Y.T - L2)*dsigmoid(L2)
    L1_delta = L2_delta.dot(W.T)*dsigmoid(L1)

    W_C = lr*L1.T.dot(L2_delta)
    V_C = lr*X.T.dot(L1_delta)

    W = W + W_C
    V = V + V_C
    
    return W,V

def output(x):
    data=[]
    for d in x:
        if d[0]>=0.5:
            data.append(1)
        else:
            data.append(0)
    return data
    
def main():
    x_,y_=create_data()
    V= (np.random.random((3,4))*2-1)
    W= (np.random.random((4,1))*2-1)
    #print ("w = {},v={}".format(W,V))


    lr = 0.11
    
    for i in range(20000):
        W,V= update_ws(x_,y_,W,V,lr)

        if(i%500==0):
            L1 =sigmoid(np.dot(x_,V))
            L2 =sigmoid(np.dot(L1,W))
            print ("Error:",np.mean(np.abs(y_.T-L2)))

    L1 =sigmoid(np.dot(x_,V))
    L2 =sigmoid(np.dot(L1,W))
 
    print ("output={}.".format(output(L2)))



main()