简单线性拟合

import tensorflow as tf
import numpy as np

x_data = np.random.rand(100)
y_data = x_data*0.6 + 0.8

# 定义变量
k = tf.Variable(tf.zeros([1, 1]))
b = tf.Variable(tf.zeros([1, 1]))
y = k * x_data + b

# 定义二次代价函数
loss = tf.reduce_mean(tf.square(y - y_data))
# 定义梯度下降优化器
optimizer = tf.train.GradientDescentOptimizer(0.2)
# 最小化训练
train_step = optimizer.minimize(loss)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for i in range(201):
        sess.run(train_step)
        if i % 10 == 0:
            print("step: "+str(i))

    print(sess.run([k, b]))