【TensorFlow】关键字global_step使用（十）

global_step经常在滑动平均，学习速率变化的时候需要用到，这个参数在
tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_steps)里面有，系统会自动更新这个参数的值，从1开始。

请看实例：

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

x = tf.placeholder(tf.float32, shape=[None, 1], name='x')  
y = tf.placeholder(tf.float32, shape=[None, 1], name='y')  
w = tf.Variable(tf.constant(0.0))  

global_steps = tf.Variable(0, trainable=False)  
learning_rate = tf.train.exponential_decay(0.1, global_steps, 10, 2, staircase=False)  
loss = tf.pow(w*x-y, 2)  

train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_steps)  

with tf.Session() as sess:  
    sess.run(tf.initialize_all_variables())  
    for i in range(10):  
        sess.run(train_step, feed_dict={x:np.linspace(1,4,10).reshape([10,1]),  
            y:np.linspace(1,4,10).reshape([10,1])})  
        print('learning_rate:{}'.format(sess.run(learning_rate)))  
        print('global_steps:{}'.format(sess.run(global_steps)))  

输出：
learning_rate:0.10717733949422836
global_steps:1
learning_rate:0.11486983299255371
global_steps:2
learning_rate:0.1231144443154335
global_steps:3
learning_rate:0.13195079565048218
global_steps:4
learning_rate:0.1414213627576828
global_steps:5
learning_rate:0.15157166123390198
global_steps:6
learning_rate:0.16245047748088837
global_steps:7
learning_rate:0.17411011457443237
global_steps:8
learning_rate:0.18660660088062286
global_steps:9
learning_rate:0.20000000298023224
global_steps:10

学习速率第一次训练开始变化，global_steps每次自动加1