RuntimeError: Attempted to use a closed Session.

在练习tensorflow时，如下代码：

with tf.Session() as sess:init_op = tf.global_variables_initializer()sess.run(init_op)print(sess.run(w1)) print(sess.run(w2))#设定训练的轮数
STEPS = 5000
for i in range(STEPS):#每次选取一个样本进行训练start = (i * batch_size) % dataset_sizeend = min(start + batch_size,dataset_size)#通过选取的样本训练神经网络并更新参数sess.run(train_step,feed_dict = {x:X[start:end],y_:Y[start:end]})if i % 1000 == 0:#每隔一段时间计算在所有数据上的交叉熵并输出total_cross_entropy = sess.run(cross_entropy,feed_dict={x:X,y_:Y})print('After %d training step(s),cross_entropy on all data is %g ' % (i,total_cross_entropy))print(sess.run(w1))
print(sess.run(w2))

运行结果会出现Attempted to use a closed Session报错提示，原因是因为‘STEPS=5000’这步出现了问题。如果“STEPS=5000”与“with tf.Session() as sess1:”对齐就会有这样的提示。

解决办法：把“STEPS=5000”及以下的code整体向后移。

正确完整代码如下：

import tensorflow as tffrom numpy.random import RandomState
batch_size = 8w1 = tf.Variable(tf.random_normal([2,3],stddev = 1,seed=1))
w2 = tf.Variable(tf.random_normal([3,1],stddev = 1,seed=1))x = tf.placeholder(tf.float32,shape=(None,2),name='x-input')
y_= tf.placeholder(tf.float32,shape=(None,1),name='y-input')#定义神经网络前向传播过程
a = tf.matmul(x,w1)
y = tf.matmul(a,w2)#定义损失函数和前向传播算法
cross_entropy = -tf.reduce_mean(y_*tf.log(tf.clip_by_value(y,1e-10,1.0)))
train_step = tf.train.AdamOptimizer(0.001).minimize(cross_entropy)#通过随机数生成一个模拟数据集
rdm = RandomState(1)
dataset_size = 128
X = rdm.rand(dataset_size,2)#定义规则给出样本标签
Y = [[int(x1 + x2 < 1)] for (x1,x2) in X ]#创建一个会话来运行程序
with tf.Session() as sess:init_op = tf.global_variables_initializer()sess.run(init_op)print(sess.run(w1)) print(sess.run(w2))#设定训练的轮数 STEPS = 5000for i in range(STEPS):#每次选取一个样本进行训练start = (i * batch_size) % dataset_sizeend = min(start + batch_size,dataset_size)#通过选取的样本训练神经网络并更新参数sess.run(train_step,feed_dict = {x:X[start:end],y_:Y[start:end]})if i % 1000 == 0:#每隔一段时间计算在所有数据上的交叉熵并输出total_cross_entropy = sess.run(cross_entropy,feed_dict={x:X,y_:Y})print('After %d training step(s),cross_entropy on all data is %g ' % (i,total_cross_entropy))print(sess.run(w1))print(sess.run(w2))

实际上是with机制帮你关闭了文件f，这样可以避免有些人读写文件后忘记关闭文件。所以with代码块里的内容运行完了，你的sess就自动由with机制帮你关闭了，因此再次使用sess就会报错。解决办法就是把代码块向后缩进，实际上就是指，所有与sess相关的操作都必须放在with代码块下，如果需要在外面再次使用，那你就需要重新开启一个sess

RuntimeError: Attempted to use a closed Session.相关推荐

MySQL5.7 Attempted to use a closed or broken resource pool 处理过程
1,tomcat启动报错不能获取JDBC连接 mysql版本:5.6.10 tomcat版本:7.0 jdk版本:1.7.0_55 开发人员强调说是以前经历过重启下mysql数据库就好了,不过我不 ...
log4j:ERROR Attempted to append to closed appender named [null].
最近用Velocity总是出现以下消息 log4j:ERROR Attempted to append to closed appender named [null]. 试了很多方法未果,最后到Vel ...
这个bug很简单-log4j:ERROR Attempted to append to closed appender named
最近调试log4j的时候,一直遇到这个错误: bug: log4j:ERROR Attempted to append to closed appender named 一直未找到原因,看了几篇博客后 ...
log4j:ERROR Attempted to append to closed appender named 的问题
一.log4j源码解析最近出现一个问题,弄得我不得不把log4j的源代码读了一篇. 如果自己不想写代码,可以下载此代码:https://github.com/lwwcl1314/atlantis ...
log4j源码解析及一个log4j:ERROR Attempted to append to closed appender named 的问题
一.log4j源码解析最近出现一个问题,弄得我不得不把log4j的源代码读了一篇. 如果自己不想写代码,可以下载此代码:https://github.com/lwwcl1314/atlantis/t ...
tensorflow 运行报错 runtime error use a closed session
https://blog.csdn.net/weixin_39390732/article/details/76474294
3-Tensorflow-demo_1-Graph和Session
import tensorflow as tf import numpy as np""" 本节内容:1.Graph图的创建.2.使用tf.constant 和 tf.a ...
TensorFlow调试常见问题（pycharm）
1. RuntimeError: Attempted to use a closed Session. 在pycharm下调用tensorflow库时,运行出现以下问题: RuntimeError: ...
模拟数据集上训练神经网络，网络解决二分类问题练习
#2018-06-24 395218 June Sunday the 25 week, the 175 day SZ ''' 模拟数据集上训练神经网络,网络解决二分类问题.'''import tens ...

RuntimeError: Attempted to use a closed Session.

RuntimeError: Attempted to use a closed Session.相关推荐

最新文章

热门文章