tf11. 保存变量

关键代码

第一步
tf.Variable
第二步
saver = tf.train.Saver()
第三步
saver.save或者saver.restore

保存变量

import sys
print(sys.version)
'''
3.5.3 |Continuum Analytics, Inc.| (default, May 15 2017, 10:43:23) [MSC v.1900 64 bit (AMD64)]
'''
import tensorflow as tf
import numpy as np

# Save to file
#remember to define the same dtype and shape when restore
W = tf.Variable([[1,2,3],[3,4,5]], dtype=tf.float32, name='weights')
b = tf.Variable([[1,2,3]], dtype=tf.float32, name='biases')

# init= tf.initialize_all_variables() # tf 马上就要废弃这种写法
# 替换成下面的写法:
init = tf.global_variables_initializer()
saver = tf.train.Saver()

with tf.Session() as sess:
    sess.run(init)
    save_path = saver.save(sess, "save/save_net.ckpt")
    print("Save to path: ", save_path)

"""    
Save to path:  my_net/save_net.ckpt
"""

提取变量

# 先建立 W, b 的容器
W = tf.Variable(np.arange(6).reshape((2, 3)), dtype=tf.float32, name="weights")
b = tf.Variable(np.arange(3).reshape((1, 3)), dtype=tf.float32, name="biases")

# 这里不需要初始化步骤 init= tf.initialize_all_variables()

saver = tf.train.Saver()
with tf.Session() as sess:
    # 提取变量
    saver.restore(sess, "save/save_net.ckpt")
    print("weights:", sess.run(W))
    print("biases:", sess.run(b))

"""
weights: [[ 1.  2.  3.]
          [ 3.  4.  5.]]
biases: [[ 1.  2.  3.]]
"""

在变量很多的情况下，每个变量都加name很麻烦，可以用下面这种形式

保存
with tf.variable_scope("regression"):
    W = tf.Variable([[1,2,3],[3,4,5]], dtype=tf.float32)
    b = tf.Variable([[1,2,3]], dtype=tf.float32)
提取
with tf.variable_scope("regression"):
    W = tf.Variable(np.arange(6).reshape((2, 3)), dtype=tf.float32)
    b = tf.Variable(np.arange(3).reshape((1, 3)), dtype=tf.float32)

疑问？

def regression(x):
    W = tf.Variable(tf.zeros([784, 10]), dtype=tf.float32)
    b = tf.Variable(tf.zeros([10]), dtype=tf.float32)
    y = tf.nn.softmax(tf.matmul(x, W) + b)
    return y, [W, b]

恢复变量

with tf.variable_scope("regression"):
    y1, variables = model.regression(x)
saver = tf.train.Saver(variables)

在恢复变量时，w和b必须指定dtype或者name，不然报错
但是下面这种情况就不用指定

    def weight_variable(shape):
        initial = tf.truncated_normal(shape, stddev=0.1)
        return tf.Variable(initial)

    def bias_variable(shape):
        initial = tf.constant(0.1, shape=shape)
        return tf.Variable(initial)