Question

我写了一个张量板代码，当我跑步时，它说＆＃34;试图使用一个封闭的会话。＆＃34;有人告诉我如何解决这个问题吗？这是代码：

# coding=utf-8
from  color_1 import read_and_decode, get_batch, get_test_batch
import cv2
import os
import time
import math
import numpy as np
import tensorflow as tf

max_steps=3000
batch_size=128
log_dir='/home/vrview/tensorflow/example/char/tfrecords'
def variable_with_weight_loss(shape,stddev,w1):
   var=tf.Variable(tf.truncated_normal(shape,stddev=stddev))
   if w1 is not None:
     weight_loss=tf.multiply(tf.nn.l2_loss(var),w1,name='weight_loss')
     tf.add_to_collection('losses',weight_loss)                                  
   return var
with tf.name_scope('input'):
    image_holder=tf.placeholder(tf.float32,[batch_size,56,56,3],name='x-input')
    label_holder=tf.placeholder(tf.int32,[batch_size],name='y-input')
with tf.name_scope('input_reshape'):
    image_input=tf.reshape(image_holder,[-1,56,56,3])
    tf.summary.image('input',image_input,10)

def variable_summaries(var):
  with tf.name_scope('summaries'):
     mean=tf.reduce_mean(var)
     tf.summary.scalar('mean',mean)
     with tf.name_scope('stddev'):
        stddev=tf.sqrt(tf.reduce_mean(tf.square(var-mean)))
     tf.summary.scalar('stddev',stddev)
     tf.summary.scalar('max',tf.reduce_max(var))
     tf.summary.scalar('min',tf.reduce_min(var))
     tf.summary.histogram('histogram',var)
def layer1(layer_name):
    with tf.name_scope(layer_name):
        with tf.name_scope('weight1'):
            weight1=variable_with_weight_loss(shape=[5,5,3,64],stddev=5e-2,w1=0.0) 
            variable_summaries(weight1)
        kernel1=tf.nn.conv2d(image_holder,weight1,[1,1,1,1],padding='SAME')
        with tf.name_scope('biases'):
            bias1=tf.Variable(tf.constant(0.0,shape=[64])) 
            variable_summaries(bias1)
        conv1=tf.nn.relu(tf.nn.bias_add(kernel1,bias1))
        pool1=tf.nn.max_pool(conv1,ksize=[1,3,3,1],strides=[1,2,2,1],padding='SAME')
        norm1=tf.nn.lrn(pool1,4,bias=1.0,alpha=0.001/9.0,beta=0.75)
        return norm1
def full_layer1(pool,name):
    with tf.name_scope(name):
        reshape=tf.reshape(pool,[batch_size,-1])
        dim=reshape.get_shape()[1].value  
        with tf.name_scope('weight3'):
            weight3=variable_with_weight_loss(shape=[dim,384],stddev=0.04,w1=0.004)
            variable_summaries(weight3)
        with tf.name_scope('bias3'):
            bias3=tf.Variable(tf.constant(0.1,shape=[384]))
            variable_summaries(bias3)
        with tf.name_scope('Wx_plus'):
            preactivate1=tf.matmul(reshape,weight3)+bias3
            tf.summary.histogram('pre_activations1',preactivate1)
        activations=tf.nn.relu(preactivate1,name='activation')
        tf.summary.histogram('activations',activations)
        return activations
def full_layer2(local,name):
    with tf.name_scope('weight4'):
        weight4=variable_with_weight_loss(shape=[384,192],stddev=0.04,w1=0.004)
        variable_summaries(weight4)
    with tf.name_scope('bias4'):
        bias4=tf.Variable(tf.constant(0.1,shape=[192]))
        variable_summaries(bias4)
    with tf.name_scope('Wx_plus2'):
        preactivate2 =tf.matmul(local,weight4)+bias4
        tf.summary.histogram('pre_activations2', preactivate2)
    activations = tf.nn.relu(preactivate2, name='activation')
    tf.summary.histogram('activations', activations)
    return activations
def full_layer3(local,name):
    with tf.name_scope('weight5'):
        weight5=variable_with_weight_loss(shape=[192,10],stddev=1/192,w1=0.0)
        variable_summaries(weight5)
    with tf.name_scope('bias5'):
        bias5=tf.Variable(tf.constant(0.0,shape=[10]))
        variable_summaries(bias5)

    logits=tf.add(tf.matmul(local,weight5),bias5)
    return logits
def loss(logits,labels):
   labels=tf.cast(labels,tf.int64)
   with tf.name_scope('cross_entropy'):
       cross_entropy=tf.nn.sparse_softmax_cross_entropy_with_logits(
       logits=logits,labels=labels,name ='cross_entropy_per_example')
       cross_entropy_mean=tf.reduce_mean(cross_entropy,name='cross_entropy')
       tf.add_to_collection('losses', cross_entropy_mean)
   tf.summary.scalar('cross_entropy',cross_entropy_mean)
   return tf.add_n(tf.get_collection('losses'),name='total_loss') 
norm1=layer1('layer1')
pool2=layer2(norm1,'layer2')
act1=full_layer1(pool2,'full_layer1')
act2=full_layer2(act1,'full_layer2')
act3=full_layer3(act2,'full_layer3')

loss=loss(act3,label_holder)
with tf.name_scope('train'):
    train_op=tf.train.AdamOptimizer(1e-3).minimize(loss)
top_k_op=tf.nn.in_top_k(act3,label_holder,1)
def run():

    image, label = read_and_decode('train.tfrecords')
    batch_image, batch_label = get_batch(image, label, batch_size=128, crop_size=56) 

    test_image, test_label = read_and_decode('val.tfrecords')
    test_images, test_labels = get_test_batch(test_image, test_label, batch_size=128, crop_size=56)  # batch 生成测试

    def feed_dict(train):
        if train:
            x=image_batch
            y=label_batch
        else:
            x=img_batch
            y=lab_batch
        return {image_holder:x,label_holder:y}

    saver=tf.train.Saver()
    num_examples = 10000
    num_iter = int(math.ceil(num_examples / batch_size))
    true_count = 0
    total_sample_count = num_iter * batch_size

    init = tf.global_variables_initializer()
    with tf.Session() as sess:
        sess.run(init)
        merged = tf.summary.merge_all()
        train_writer = tf.summary.FileWriter(log_dir + '/train', sess.graph)
        test_writer = tf.summary.FileWriter(log_dir + '/test')
        coord = tf.train.Coordinator()
        threads = tf.train.start_queue_runners(coord=coord)

    for step in range(max_steps):
        start_time = time.time()
        image_batch, label_batch = sess.run([batch_image, batch_label])

        if step % 10 == 0:

            run_options = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE)
            run_metadata = tf.RunMetadata()
            img_batch, lab_batch = sess.run([test_images, test_labels])
            summary,loss_value=sess.run([merged,loss],feed_dict=feed_dict(True))
            train_writer.add_run_metadata(run_metadata,'step%03d' % step)
            train_writer.add_summary(summary, step)

            saver.save(sess,log_dir+"/model.ckpt",step)

    step = 0
    while step < num_iter:
        img_batch, lab_batch = sess.run([test_images, test_labels])
        predictions = sess.run([top_k_op], feed_dict=feed_dict(False))
        true_count += np.sum(predictions)
        step += 1
        test_writer.add_summary(predictions)
        print('Predict:', predictions)
    precision = true_count * 1.0 / total_sample_count
    print 'true_count:', true_count, 'total_sample_count:', total_sample_count
    print 'precision:', precision
    train_writer.close()
    test_writer.close()
if __name__=='__main__':
    run()

当我运行它时，它说＆＃34;尝试使用已关闭的会话。＆＃34;就像这样：

File "/home/vrview/tensorflow/example/char/tfrecords/color2_board.py", line 238, in <module>
    run()
  File "/home/vrview/tensorflow/example/char/tfrecords/color2_board.py", line 207, in run
    image_batch, label_batch = sess.run([batch_image, batch_label])
  File "/home/vrview/tensorflow/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 767, in run
    run_metadata_ptr)
  File "/home/vrview/tensorflow/local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 903, in _run
    raise RuntimeError('Attempted to use a closed Session.')
RuntimeError: Attempted to use a closed Session.

剂量任何人都知道发生了什么？非常感谢你的帮助！

Answer 1

使用sess的任何内容都应在with tf.Session() as sess内。您基本上只需缩进从for step in range(max_steps):到test_writer.close()

的所有内容

您尝试在sess.run([batch_image, batch_label])范围之外调用with tf.Session() as sess，一旦超出范围，就会自动关闭sess对象。

Answer 2

就我而言：

GetOrAddAsync

我删除了try: model.load("model.tflearn") except: model.fit(training, output, n_epoch=1000, batch_size=8, show_metric=True) model.save("model.tflearn")和try:，仅使用最后两行即可解决问题。

我该如何解决这个问题：＆＃34;运行时错误：尝试使用已关闭的会话。＆＃34;

2 个答案: