在make_initializable_iterator和make_one_shot_iterator之间的tensorflow数据集API差异

Question

我想知道的区别make_initializable_iterator和make_one_shot_iterator.
1. Tensorflow文件说A "one-shot" iterator does not currently support re-initialization.这究竟是什么意思？
2.以下2个片段是否相同？
使用make_initializable_iterator

iterator = data_ds.make_initializable_iterator()
data_iter = iterator.get_next()
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for e in range(1, epoch+1):
    sess.run(iterator.initializer)
    while True:
        try:
            x_train, y_train = sess.run([data_iter])
            _, cost = sess.run([train_op, loss_op], feed_dict={X: x_train,
                                                               Y: y_train})
        except tf.errors.OutOfRangeError:   
            break
sess.close()

使用 make_one_shot_iterator

iterator = data_ds.make_one_shot_iterator()
data_iter = iterator.get_next()
sess = tf.Session()
sess.run(tf.global_variables_initializer())
for e in range(1, epoch+1):
    while True:
        try:
            x_train, y_train = sess.run([data_iter])
            _, cost = sess.run([train_op, loss_op], feed_dict={X: x_train,
                                                               Y: y_train})
        except tf.errors.OutOfRangeError:   
            break
sess.close()

Answer 1

假设您要使用相同的代码进行培训和验证.您可能希望使用相同的迭代器,但初始化为指向不同的数据集; 类似以下内容:

def _make_batch_iterator(filenames):
    dataset = tf.data.TFRecordDataset(filenames)
    ...
    return dataset.make_initializable_iterator()


filenames = tf.placeholder(tf.string, shape=[None])
iterator = _make_batch_iterator(filenames)

with tf.Session() as sess:
    for epoch in range(num_epochs):

        # Initialize iterator with training data
        sess.run(iterator.initializer,
                 feed_dict={filenames: ['training.tfrecord']})

        _train_model(...)

        # Re-initialize iterator with validation data
        sess.run(iterator.initializer,
                 feed_dict={filenames: ['validation.tfrecord']})

        _validate_model(...)

使用一次性迭代器,您无法像这样重新初始化它.