我们使用一个三层的小网络来,模拟函数y = x^3+b函数
 import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt #训练数据
x_data = np.linspace(-6.0,6.0,30)[:,np.newaxis]
y_data = np.power(x_data,3) + 0.7
#验证数据
t_data = np.linspace(-20.0,20.0,40)[:,np.newaxis]
ty_data = np.power(t_data,3) + 0.7
#占位符
x = tf.placeholder(tf.float32,[None,1])
y = tf.placeholder(tf.float32,[None,1]) #network
#--layer one--
l_w_1 = tf.Variable(tf.random_normal([1,10]))
l_b_1 = tf.Variable(tf.zeros([1,10]))
l_fcn_1 = tf.matmul(x, l_w_1) + l_b_1
relu_1 = tf.nn.relu(l_fcn_1)
#---layer two----
l_w_2 = tf.Variable(tf.random_normal([10,20]))
l_b_2 = tf.Variable(tf.zeros([1,20]))
l_fcn_2 = tf.matmul(relu_1, l_w_2) + l_b_2
relu_2 = tf.nn.relu(l_fcn_2) #---output---
l_w_3 = tf.Variable(tf.random_normal([20,1]))
l_b_3 = tf.Variable(tf.zeros([1,1]))
l_fcn_3 = tf.matmul(relu_2, l_w_3) + l_b_3
#relu_3 = tf.tanh(l_fcn_3)
# init
init = tf.global_variables_initializer()
#定义 loss func
loss = tf.reduce_mean(tf.square(y-l_fcn_3))
learn_rate =0.001
train_step = tf.train.GradientDescentOptimizer(learn_rate).minimize(loss) with tf.Session() as sess:
sess.run(init);
for epoch in range(20):
for step in range(5000):
sess.run(train_step,feed_dict={x:x_data,y:y_data})
y_pred = sess.run(l_fcn_3,feed_dict={x:t_data})
print sess.run(l_fcn_3,feed_dict={x:[[10.]]})
plt.figure()
plt.scatter(t_data,ty_data)
plt.plot(t_data,y_pred,'r-')
plt.show()
[[ 533.45062256]]
 
aaarticlea/png;base64,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" alt="" />

使用神经网络来拟合函数y = x^3 +b的更多相关文章

  1. matlab最小二乘法数据拟合函数详解

    定义: 最小二乘法(又称最小平方法)是一种数学优化技术.它通过最小化误差的平方和寻找数据的最佳函数匹配.利用最小二乘法可 以简便地求得未知的数据,并使得这些求得的数据与实际数据之间误差的平方和为最小. ...

  2. matlab-非线性拟合函数lsqcurvefit的使用和初值选取

    所解决问题: 我们知道我们的表达式是y=A+B*exp(-x.^2)-C./log(x), 而且现在我们手里面有x与y对应的一大把数据. 我们需要根据x, y的值找出最佳的A.B.C值.则我们现在借助 ...

  3. cftool拟合&函数逼近

    cftool拟合&函数逼近 cftool 真是神奇,之前我们搞的一些线性拟合解方程,多项式拟合,函数拟合求参数啊,等等. 已经超级多了,为啥还得搞一个cftool拟合啊?而且毫无数学理论. 如 ...

  4. matlab的拟合函数polyfit()函数

    matlab的多项式拟合: polyfit()函数 功能:在最小二乘法意义之上,求解Y关于X的最佳的N次多项式函数. clc;clear; close all; x=[ ]; y=[2.7 7.4 2 ...

  5. 函数 y=x^x的分析

    关于函数 y=xx的分析: 由图像得,y在负无穷大到0图像处处不连续,故y的定义域为(0,正无穷大): 故该函数不就是y=e^(lnxx)吗? 1.定义域:我们变形一下,y=e^(xlnx),显然是0 ...

  6. 2、函数y=f(x)

    /* Note:Your choice is C IDE */ #include "stdio.h" /* 3.函数y=f(x)可表示为: */ void main() { int ...

  7. 2017年全国卷3的21题与2018年全国卷3的21题命题背景是同一个函数$y=\frac{2x}{\ln(x+1)}$(再次瞎谈)

    2017年四川高考数学(全国卷3)理科21题第1问 已知函数\(f(x)=x-1-a\ln x\) (1)若\(f(x)\geqslant 0\),求\(a\)的值\(.\) 该不等式等价于$a\ln ...

  8. 神经网络可以拟合任意函数的视觉证明A visual proof that neural nets can compute any function

    One of the most striking facts about neural networks is that they can compute any function at all. T ...

  9. C# + Matlab 实现计件工时基于三层BP神经网络的拟合--真实项目

    工序工时由该工序的工艺参数决定,有了工时后乘以固定因子就是计件工资.一般参考本地小时工资以及同类小时工资并考虑作业的风险等因素给出固定因子 采用的VS2010 , Matlab2015a 64,  开 ...

随机推荐

  1. 关于postgres中的一些宏的tips

    Postgresql作为C语言开发的代码,其中大量的运用了一些宏的操作. 因此理解这些宏很重要,然而有时候这些宏总让人很费解. 作为一个经常翻翻postgresql源码的小白,在这里做一个记录吧,方便 ...

  2. 初识Avro

    Avro是Hadoop生态圈的一部分,由Hadoop的创始人Doug Cutting牵头开发,当前最新版本1.8.2.Avro是一个数据序列化系统,设计用于支持大批量数据交换的应用.它的主要特点有: ...

  3. ES6初体验

    开始学习ES6,打算走全栈这条路了,废话不多说,开始吧. 首先安装node环境,去node官网上面下载node最新版本的,我用的系统是window10,所以我只需要下一步下一步就行了,安装完成后打开c ...

  4. 挑战App Store,微信通过“跳一跳”秀了一下“小程序”的肌肉

    2017年即将结束的时候,微信放了一个大招.随着最新的微信v6.6.1版本更新,基于小程序的"小游戏"板块正式上线.微信上首发的这款"小游戏"叫"跳一 ...

  5. golang 如何验证struct字段的数据格式

    本文同时发表在https://github.com/zhangyachen/zhangyachen.github.io/issues/125 假设我们有如下结构体: type User struct ...

  6. Java中堆与栈

    简单的说:Java把内存划分成两种:一种是栈内存,一种是堆内存. 1:什么是堆内存: 堆内存是是Java内存中的一种,它的作用是用于存储Java中的对象和数组,当我们new一个对象或者创建一个数组的时 ...

  7. JavaScript Array 对象方法 以及 如何区分javascript中的toString()、toLocaleString()、valueOf()方法

    1.concat() 2.join() 3.pop() 4.push() 5.reverse() 6.shift() 7.unshift() 8.slice() 9.sort() 10.splice( ...

  8. Centos 7 防火墙firewalld命令

    今天自己在Hyper-v下搭建三台Linux服务器集群,用于学习ELKstack(即大数据日志解决技术栈Elasticsearch,Logstash,Kibana的简称),下载的Linux版本为cen ...

  9. C#语言和SQL Server第十章笔记

    第十章 :使用关键字模糊查询 笔记 一:使用关键字 :LIKE  BETWEEN  IN进行模糊查询 通配符:  一类字符,代替一个或多个真正的字符 与LIKE关键字一起使用 通配符: 解释 实例 符 ...

  10. python简单爬虫技术

    项目中遇到这个只是点,捣鼓了半天最后没用上,但是大概对爬虫技术有了些许了解 要先 比如: #抓取网页代码 import urllib2 import json url_data = urllib2.u ...