global_step Variables are used to save the global training steps (global training step) Value of .

It's always moving average , This parameter is used when the learning rate changes .

use optimizer.minimize() Can be updated automatically global_step.

# -*-coding: utf-8-*- import tensorflow as tf import numpy as np 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, 0.9,
staircase=False) loss = tf.pow(w*x - y, 2) train_op =
tf.train.GradientDescentOptimizer(learning_rate).minimize(loss,
global_step=global_steps) with tf.Session() as sess:
sess.run(tf.global_variables_initializer()) for i in range(10):
sess.run(train_op, feed_dict={x: np.linspace(1, 2, 10).reshape([10, 1]), y:
np.linspace(1, 2, 10).reshape([10, 1])}) print sess.run(learning_rate) print
sess.run(global_steps)

Technology
©2019-2020 Toolsou All rights reserved,
It's unexpected Python Cherry tree (turtle The gorgeous style of Library ) Some East 14 Pay change 16 salary , Sincerity or routine ? Browser kernel ( understand )java Four functional interfaces ( a key , simple )HashMap Explain in detail html Writing about cherry trees , Writing about cherry trees os Simple use of module