数学之路(3)-机器学习(3)-机器学习算法-神经网络[16]

我们调用第三方的神经网络python组件继续进行更复杂的函数拟合，这次拟合一个比sin函数更复杂的函数f(x)=sin(x)*0.5+cos(x)*0.5

python代码如下

 

#!/usr/bin/env python

#-*- coding: utf-8 -*-

#bp ann 函数拟合sin*0.5+cos*0.5

import neurolab as nl

import numpy as np

import matplotlib.pyplot as plt

isdebug=False



#x和d样本初始化

train_x =[]

d=[]

samplescount=1000

myrndsmp=np.random.rand(samplescount)

for yb_i in xrange(0,samplescount):

    train_x.append([myrndsmp[yb_i]*4*np.pi-2*np.pi])

for yb_i in xrange(0,samplescount):

    d.append(np.sin(train_x[yb_i])*0.5+np.cos(train_x[yb_i])*0.5)



myinput=np.array(train_x)   

mytarget=np.array(d)



bpnet = nl.net.newff([[-2*np.pi, 2*np.pi]], [5, 1])

err = bpnet.train(myinput, mytarget, epochs=800, show=100, goal=0.02)



simd=[]

for xn in xrange(0,len(train_x)):

#        print "====================="

#        print u"样本：%f=> "%(train_x[xn][0])

        simd.append(bpnet.sim([train_x[xn]])[0][0])

#        print simd[xn]

#        print u"--正确目标值--"

#        print d[xn]

#        print "====================="        



temp_x=[]

temp_y=simd

temp_d=[]

i=0

for mysamp in train_x:

     temp_x.append(mysamp[0])

     temp_d.append(d[i][0])

     i+=1

                 

x_max=max(temp_x)

x_min=min(temp_x)

y_max=max(max(temp_y),max(d))+0.2

y_min=min(min(temp_y),min(d))-0.2

    

plt.xlabel(u"x")

plt.xlim(x_min, x_max)

plt.ylabel(u"y")

plt.ylim(y_min, y_max)

plt.title("http://blog.csdn.net/myhaspl" )

lp_x1 = temp_x

lp_x2 = temp_y

lp_d = temp_d

plt.plot(lp_x1, lp_x2, 'r*')

plt.plot(lp_x1,lp_d,'b*')

plt.show()

>>> runfile(r'I:\book_prog\ann_bpnhsincos1.py', wdir=r'I:\book_prog')
Epoch: 100; Error: 0.528978849953;
Epoch: 200; Error: 0.33336612138;
Epoch: 300; Error: 0.253337487331;
Epoch: 400; Error: 0.20472927421;
Epoch: 500; Error: 0.202153963051;
Epoch: 600; Error: 0.19900731385;
Epoch: 700; Error: 0.197426245762;
Epoch: 800; Error: 0.193607559472;
The maximum number of train epochs is reached
>>> 

 

拟合效果为：

 

本博客所有内容是原创，如果转载请注明来源

http://blog.csdn.net/u010255642