当前位置:  开发笔记 > 人工智能 > 正文

TensorFLow用Saver保存和恢复变量

这篇文章主要为大家详细介绍了TensorFLow用Saver保存和恢复变量,具有一定的参考价值,感兴趣的小伙伴们可以参考一下

本文为大家分享了TensorFLow用Saver保存和恢复变量的具体代码,供大家参考,具体内容如下

建立文件tensor_save.py, 保存变量v1,v2的tensor到checkpoint files中,名称分别设置为v3,v4。

import tensorflow as tf

# Create some variables.
v1 = tf.Variable(3, name="v1")
v2 = tf.Variable(4, name="v2")

# Create model
y=tf.add(v1,v2)

# Add an op to initialize the variables.
init_op = tf.initialize_all_variables()

# Add ops to save and restore all the variables.
saver = tf.train.Saver({'v3':v1,'v4':v2})

# Later, launch the model, initialize the variables, do some work, save the
# variables to disk.
with tf.Session() as sess:
 sess.run(init_op)
 print("v1 = ", v1.eval())
 print("v2 = ", v2.eval())
 # Save the variables to disk.
 save_path = saver.save(sess, "f:/tmp/model.ckpt")
 print ("Model saved in file: ", save_path)

建立文件tensor_restror.py, 将checkpoint files中名称分别为v3,v4的tensor分别恢复到变量v3,v4中。

import tensorflow as tf

# Create some variables.
v3 = tf.Variable(0, name="v3")
v4 = tf.Variable(0, name="v4")

# Create model
y=tf.mul(v3,v4)

# Add ops to save and restore all the variables.
saver = tf.train.Saver()

# Later, launch the model, use the saver to restore variables from disk, and
# do some work with the model.
with tf.Session() as sess:
 # Restore variables from disk.
 saver.restore(sess, "f:/tmp/model.ckpt")
 print ("Model restored.")
 print ("v3 = ", v3.eval())
 print ("v4 = ", v4.eval())
 print ("y = ",sess.run(y))

以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持。

推荐阅读
我我檬檬我我186
这个屌丝很懒,什么也没留下!
DevBox开发工具箱 | 专业的在线开发工具网站    京公网安备 11010802040832号  |  京ICP备19059560号-6
Copyright © 1998 - 2020 DevBox.CN. All Rights Reserved devBox.cn 开发工具箱 版权所有