tensorflow学习（二）直线拟合

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

num_point = 1000
vet = []

for i in range(num_point):
    x1 = np.random.normal(0.0,0.7)
    y1 = x1*0.2 + 0.3 + np.random.normal(0.0,0.03)
    vet.append(([x1, y1]))

xdata = [i[0] for i in vet]
ydata = [i[1] for i in vet]




#初始化权重w和偏差b，w的值可以为随机数 b的值可以为0
w = tf.Variable(tf.random_uniform([1],-1.0,1.0),name ='w')
b = tf.Variable(tf.zeros([1],name = 'b'))
y = w * xdata + b
#计算实际值和预测值之间的差值 以均方差作为损失函数
loss = tf.reduce_mean(tf.square(y-ydata),name = 'loss')
#采用梯度下降法进行优化
optimizer = tf.train.GradientDescentOptimizer(0.5).minimize(loss)

init = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init)
    print('w=',sess.run(w),'b = ',sess.run(b),'loss=', sess.run(loss))
    for step in range(20):
        sess.run(optimizer)
        print('w=', sess.run(w), 'b = ', sess.run(b), 'loss=', sess.run(loss))
    plt.scatter(xdata, ydata, c='r')
    plt.plot(xdata, sess.run(w) * xdata + sess.run(b))
    plt.show()