学习率：表示了每次更新参数的幅度大小。学习率过大，会导致待优化的参数在最小值附近波动，不收敛；学习率过小，会导致待优化的参数收敛缓慢。
在训练过程中，参数的更新相纸损失函数梯度下降的方向。参数的更新公式为：
假设损失函数为 loss = (w + 1)2。梯度是损失函数 loss 的导数为 ∇=2w+2。如参数初值为 5，学习率为 0.2，则参数和损失函数更新如下：
1 次 参数 w： 5 - 0.2 * (2 * 5 + 2) = 2.6
2 次 参数 w： 2.6 - 0.2 * (2 * 2.6 + 2) = 1.16
3 次 参数 w： 1.16 – 0.2 * (2 * 1.16 + 2) = 0.296
4 次 参数 w： 0.296
损失函数 loss = (w + 1)2 的图像为：
由图可知，损失函数loss的最小值会在（-1,0）处得到，此时损失函数的倒数为0，得到最终参数w = -1。代码如下
#coding:utf-8
#设损失函数 loss=(w+1)^2，令w初始值为5。反向传播就是求最优w，即最小loss对应的w值
import tensorflow as tf
w = tf.Variable(tf.constant(5, dtype=tf.float32))
loss = tf.square(w+1)
train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss)
with tf.Session() as sess:
init_op = tf.global_variables_initializer()
sess.run(init_op)
for i in range(40):
sess.run(train_step)
w_val = sess.run(w)
loss_val = sess.run(loss)
print "After %s steps: w is %f, loss is %f." %(i, w_val, loss_val)
运行结果如下：
After 35 steps: w is -1.000000, loss is 0.000000.
After 36 steps: w is -1.000000, loss is 0.000000.
After 37 steps: w is -1.000000, loss is 0.000000.
After 38 steps: w is -1.000000, loss is 0.000000.
After 39 steps: w is -1.000000, loss is 0.000000.
若上述代码学习率设置为1，则运行结果如下：
After 35 steps: w is 5.000000, loss is 36.000000.
After 36 steps: w is -7.000000, loss is 36.000000.
After 37 steps: w is 5.000000, loss is 36.000000.
After 38 steps: w is -7.000000, loss is 36.000000.
After 39 steps: w is 5.000000, loss is 36.000000.
若学习率设置为0.001，则运行结果如下：
After 35 steps: w is 4.582785, loss is 31.167484.
After 36 steps: w is 4.571619, loss is 31.042938.
After 37 steps: w is 4.560476, loss is 30.918892.
After 38 steps: w is 4.549355, loss is 30.795341.
After 39 steps: w is 4.538256, loss is 30.672281.指数衰减学习率：随着训练轮数变化而动态更新
学习率计算公式如下：
用 Tensorflow 的函数表示为：
global_step = tf.Variable(0, trainable=False)
learning_rate=tf.train.exponential_decay(LEARNING_RATE_BASE,global_step,LEARNING_RATE_STEP,LEARNING_RATE_DECAY,staircase=True/False)
其中， LEARNING_RATE_BASE 为学习率初始值， LEARNING_RATE_DECAY 为学习率衰减率,global_step 记录了当前训练轮数，为不可训练型参数。学习率 learning_rate 更新频率为输入数据集总样本数除以每次喂入样本数。若 staircase 设置为 True 时，表示 global_step/learning rate step 取整数，学习率阶梯型衰减；若 staircase 设置为 false 时，学习率会是一条平滑下降的曲线。
例如：在本例中，模型训练过程不设定固定的学习率，使用指数衰减学习率进行训练。其中，学习率初值设置为 0.1，学习率衰减率设置为 0.99， BATCH_SIZE 设置为 1。
#coding:utf-8
#设损失函数 loss=(w+1)^2，令w初始值为10。反向传播就是求最优w，即求最小loss对应的w值
#使用指数衰减的学习率，在迭代初期得到较高的下降速度，可以在较小的训练轮数下取得更有效的收敛速度。
import tensorflow as tf
LEARNING_RATE_BASE = 0.1 #最初学习率
LEARNING_RATE_DECAY = 0.99 #学习率衰减率
LEARNING_RATE_STEP = 1 #喂入多少轮BATCH_SIZE后，更新一此学习率，一般为：总样本数/BATCH_SIZE
#运行了几轮BATCH_SIZE的计数器，初始值为0，设为不被训练
global_step = tf.Variable(0, trainable=False)
#定义指数下降学习率
learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE, global_step, LEARNING_RATE_STEP, LEARNING_RATE_DECAY, staircase=True)
w = tf.Variable(tf.constant(5, dtype=tf.float32))
loss = tf.square(w+1)
train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step = global_step)
with tf.Session() as sess:
init_op = tf.global_variables_initializer()
sess.run(init_op)
for i in range(40):
sess.run(train_step)
learning_rate_val = sess.run(learning_rate)
global_step_val = sess.run(global_step)
w_val = sess.run(w)
loss_val = sess.run(loss)
print "After %s steps: gloabal_step is %f, w is %f, learning rate is %f, loss is %f" %(i, global_step_val, w_val, learning_rate_val, loss_val
运行结果如下：
After 35 steps: gloabal_step is 36.000000, w is -0.992297, learning rate is 0.069641, loss is 0.000059
After 36 steps: gloabal_step is 37.000000, w is -0.993369, learning rate is 0.068945, loss is 0.000044
After 37 steps: gloabal_step is 38.000000, w is -0.994284, learning rate is 0.068255, loss is 0.000033
After 38 steps: gloabal_step is 39.000000, w is -0.995064, learning rate is 0.067573, loss is 0.000024
After 39 steps: gloabal_step is 40.000000, w is -0.995731, learning rate is 0.066897, loss is 0.000018
由结果可以看出，随着训练轮数增加学习率在不断减小
