欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页

使用TensorFlow实现简单的线性拟合

程序员文章站 2022-07-13 11:28:55
...

本文使用TensorFlow实现最简单的线性回归模型。

线性拟合\(y=2.7x+0.6\),代码如下:

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

n = 201    # x点数
X = np.linspace(-1, 1, n)[:,np.newaxis]    # 等差数列构建X,[:,np.newaxis]这个是shape,这一行构建了一个n维列向量([1,n]的矩阵)
noise = np.random.normal(0, 0.5, X.shape)    # 噪声值,与X同型
Y = X*2.7 + 0.6 + noise    # Y

xs = tf.placeholder(tf.float32, [None, 1])    # 下面两行是占位符tf.placeholder(dtype, shape)
ys = tf.placeholder(tf.float32, [None, 1])

w = tf.Variable(1.1)    # 这两行是weight变量,bias变量,括号中是初始值
b = tf.Variable(0.2)

ypredict = tf.add(w*xs,b)    # 根据 w, b 产生的预测值

loss = tf.reduce_sum(tf.pow(ys-ypredict,2.0))/n    # 损失函数,tf.reduce_sum()按某一维度元素求和,默认为按列

optimizer = tf.train.GradientDescentOptimizer(0.01).minimize(loss)    # 梯度下降优化器,0.01学习率,最小化losss

init = tf.global_variables_initializer()    # 初始化所有变量

with tf.Session() as sess: 
    sess.run(init)    # 运行初始化 
    for i in range (1000):    # 迭代1000次 
        sess.run(optimizer, feed_dict = {xs:X,ys:Y})    # 运行优化器,梯度下降用到loss,计算loss需要xs, ys所以后面需要feed_dict 
        if i%50==0:    # 每隔50次迭代输出w,b,loss
                 # 下面sess.run(w),sess.run(b)里面没有feed_dict是因为打印w,b不需要xs,ys,而打印loss需要 
                 print ("w:",sess.run(w),"\t b:", sess.run(b), "\t loss:", sess.run(loss,feed_dict={xs:X,ys:Y})) 
        
    plt.plot(X,X*sess.run(w)+sess.run(b))    # 运行迭代之后绘制拟合曲线,这需要在sess里面运行是因为要用到w,b 
    plt.scatter(X,Y)    # 绘制被拟合数据(散点) 
    plt.show()    # 绘制图像
结果:
w: 1.1106868     b: 0.2086223    loss: 1.2682248
w: 1.5626049     b: 0.4772562    loss: 0.7024503
w: 1.8849733     b: 0.57508457   loss: 0.47280872
w: 2.1149294     b: 0.61071056   loss: 0.36368176
w: 2.278966      b: 0.6236845    loss: 0.30917725
w: 2.3959787     b: 0.6284093    loss: 0.2815788
w: 2.4794474     b: 0.6301298    loss: 0.26755357
w: 2.5389886     b: 0.63075644   loss: 0.26041925
w: 2.5814607     b: 0.6309848    loss: 0.2567894
w: 2.611758      b: 0.6310678    loss: 0.25494233
w: 2.6333694     b: 0.6310981    loss: 0.25400248
w: 2.6487865     b: 0.631109     loss: 0.2535242
w: 2.659784      b: 0.63111293   loss: 0.25328085
w: 2.6676288       b: 0.6311139    loss: 0.25315702
w: 2.6732242     b: 0.6311139    loss: 0.25309405
w: 2.6772156     b: 0.6311139    loss: 0.25306198
w: 2.6800632     b: 0.6311139    loss: 0.25304565
w: 2.6820953     b: 0.6311139    loss: 0.25303733
w: 2.6835444     b: 0.6311139    loss: 0.25303313
w: 2.684578      b: 0.6311139    loss: 0.25303096

使用TensorFlow实现简单的线性拟合