如何使用 tf.MonitoredTrainingSession 在训练和验证数据集之间切换？

Question

我想在 tensorflow 数据集 API 中使用 feedable 迭代器设计，因此我可以在一些训练步骤后切换到验证数据。但是如果我切换到验证数据，它将结束整个会话。

下面的代码演示了我想要做什么：

import tensorflow as tf


graph = tf.Graph()
with graph.as_default():
    training_ds = tf.data.Dataset.range(32).batch(4)
    validation_ds = tf.data.Dataset.range(8).batch(4)

    handle = tf.placeholder(tf.string, shape=[])
    iterator = tf.data.Iterator.from_string_handle(
        handle, training_ds.output_types, training_ds.output_shapes)
    next_element = iterator.get_next()

    training_iterator = training_ds.make_initializable_iterator()
    validation_iterator = validation_ds.make_initializable_iterator()


with graph.as_default():

    with tf.train.MonitoredTrainingSession() as sess:
        training_handle = sess.run(training_iterator.string_handle())
        validation_handle = sess.run(validation_iterator.string_handle())
        sess.run(training_iterator.initializer)
        count_training = 0
        while not sess.should_stop():
            x = sess.run(next_element, feed_dict={handle: training_handle})
            count_training += 1
            print('{} [training] {}'.format(count_training, x.shape))
            # print(x)

            # we do periodic validation
            if count_training % 4 == 0:
                sess.run(validation_iterator.initializer)
                count_validation = 0
                while not sess.should_stop():
                    y = sess.run(next_element, feed_dict={handle: validation_handle})
                    count_validation += 1
                    print('  {} [validation] {}'.format(count_validation, y.shape))
                    # print(y)

训练数据有32个元素，4个batch，所以得到8个batch 我们每 4 个步骤进行一次验证，所以我预计：

#  1 [training]
# 2 [training]
# 3 [training]
# 4 [training]
#      1 [validation]
#      2 [validation]
# 5 [training]
# 6 [training]
# 7 [training]
# 8 [training]
#      1 [validation]
#      2 [validation]

但在第一次验证完成后停止:

# 1 [training]
# 2 [training]
# 3 [training]
# 4 [training]
#      1 [validation]
#      2 [validation]

那么，如何在tf.MonitoredTrainingSession中使用这个feedable迭代器呢？

Answer 1

我建议在验证数据集末尾捕获 tf.errors.OutOfRangeError（您也可以在官方 API 中检查 the processing multiple epochs section 以使用 repeat数据集）：

while not sess.should_stop():
    x = sess.run(next_element, feed_dict={handle: training_handle})
    count_training += 1
    print('{} [training] {}'.format(count_training, x.shape))

    # we do periodic validation
    if count_training % 4 == 0:
        sess.run(validation_iterator.initializer)
        count_validation = 0
        while True:
            try:
                y = sess.run(next_element, feed_dict={handle: validation_handle})
                count_validation += 1
                print('  {} [validation] {}'.format(count_validation, y.shape))
            except tf.errors.OutOfRangeError:
                break

这段代码打印：

1 [training] (4,)  
2 [training] (4,)  
3 [training] (4,)  
4 [training] (4,)  
  1 [validation] (4,)  
  2 [validation] (4,)  
5 [training] (4,)
6 [training] (4,)
7 [training] (4,)
8 [training] (4,)
  1 [validation] (4,)
  2 [validation] (4,)

如何使用 tf.MonitoredTrainingSession 在训练和验证数据集之间切换？

How to switch between training and validation dataset with tf.MonitoredTrainingSession?

dataset

tensorflow

tensorflow-datasets

tensorflow-estimator