Ordering from tensorflow queues - tensorflow

Question

I am trying to understand queues in better detail. Using the code below I expect that since I am not shuffling the alphabetic list the output collection will be in alphabetic order. This seems to be the case for all but the initial epoch. Am I misunderstanding something?
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import time
import tensorflow as tf
import numpy as np
import string
# Basic model parameters as external flags.
flags = tf.app.flags
FLAGS = flags.FLAGS
flags.DEFINE_float('learning_rate', 0.01, 'Initial learning rate.')
flags.DEFINE_integer('num_epochs', 2, 'Number of epochs to run trainer.')
flags.DEFINE_integer('hidden1', 128, 'Number of units in hidden layer 1.')
flags.DEFINE_integer('hidden2', 32, 'Number of units in hidden layer 2.')
flags.DEFINE_integer('batch_size', 100, 'Batch size. '
'Must divide evenly into the dataset sizes.')
flags.DEFINE_string('train_dir', '/tmp/data',
'Directory to put the training data.')
flags.DEFINE_boolean('fake_data', False, 'If true, uses fake data '
'for unit testing.')
def run_training():
# Tell TensorFlow that the model will be built into the default Graph.
with tf.Graph().as_default():
with tf.name_scope('input'):
# Input data
images_initializer = tf.placeholder(
dtype=tf.int64,
shape=[52,1])
input_images = tf.Variable(
images_initializer, trainable=False, collections=[])
image = tf.train.slice_input_producer(
[input_images], num_epochs=2)
images = tf.train.batch(
[image], batch_size=1)
alph_initializer = tf.placeholder(
dtype=tf.string,
shape=[26,1])
input_alph = tf.Variable(
alph_initializer, trainable=False, collections=[])
alph = tf.train.slice_input_producer(
[input_alph], shuffle=False, capacity=26)
alphs = tf.train.batch(
[alph], batch_size=1)
my_list = np.array(list(range(0,52))).reshape(52,1)
my_list_val = np.array(list(string.ascii_lowercase)).reshape(26,1)
# Create the op for initializing variables.
init_op = tf.initialize_all_variables()
# Create a session for running Ops on the Graph.
sess = tf.Session()
# Run the Op to initialize the variables.
sess.run(init_op)
sess.run(input_images.initializer,
feed_dict={images_initializer: my_list})
sess.run(input_alph.initializer,
feed_dict={alph_initializer: my_list_val})
sess.run(tf.local_variables_initializer())
sess.run(tf.global_variables_initializer())
# Start input enqueue threads.
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=sess, coord=coord)
# And then after everything is built, start the training loop.
collection = []
try:
step = 0
while not coord.should_stop():
start_time = time.time()
# Run one step of the model.
integer = sess.run(image)
#print("Integer val", integer)
char = sess.run(alph)
collection.append(char[0][0])
print("String val", char)
duration = time.time() - start_time
except tf.errors.OutOfRangeError:
print('Saving')
print('Done training for %d epochs, %d steps.' % (FLAGS.num_epochs, step))
finally:
# When done, ask the threads to stop.
coord.request_stop()
print(str(collection))
# Wait for threads to finish.
coord.join(threads)
sess.close()
def main(_):
run_training()
if __name__ == '__main__':
tf.app.run()

Answer 1

Changing the above to the below clears up my confusion
try:
step = 0
while not coord.should_stop():
start_time = time.time()
# Run one step of the model.
integer = sess.run(images)
#print("Integer val", integer)
char = sess.run(alphs)
collection.append(char[0][0])
print("String val", char)
duration = time.time() - start_time
except tf.errors.OutOfRangeError:
print('Saving')
print('Done training for %d epochs, %d steps.' % (FLAGS.num_epochs, step))
finally:
# When done, ask the threads to stop.
coord.request_stop()
print(str(collection))