import tensorflow as tf
import numpy as np
#使用numpy构造100个随机点
x_data=np.random.rand(100)
y_data=x_data*0.5+0.6
#构造一个线性模型
b=tf.Variable(0.)
k=tf.Variable(0.)
y=x_data*b+k
#二次代价函数
loss=tf.reduce_mean(tf.square(y_data-y))
#定义一个梯度下降法来进行训练的优化器
optimizer=tf.train.GradientDescentOptimizer(0.2)
#最小化代价函数
train=optimizer.minimize(loss)
#初始化变量
init=tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init)
for i in range(300):
sess.run(train)
if i ==0:
print(i,[b.eval(),k.eval()])
0 [0.18192242, 0.33778045]
20 [0.41750893, 0.6432164]
40 [0.45200694, 0.6251434]
60 [0.47207776, 0.6146284]
80 [0.48375493, 0.60851073]
100 [0.49054867, 0.6049515]
120 [0.49450126, 0.6028808]
140 [0.49680084, 0.60167605]
160 [0.49813876, 0.6009751]
180 [0.49891713, 0.60056734]
200 [0.49936998, 0.60033005]
220 [0.49963346, 0.60019207]
240 [0.49978673, 0.6001117]
260 [0.4998759, 0.600065]
280 [0.49992776, 0.6000379]
