从张量流数据集迭代器获取输入（文件名）

Question

我正在使用 tensorflow 数据集来训练模型。数据集获取文件名列表以在会话期间读取它们，我想将文件名与图像一起获取。更详细地说，我有这样的东西：

filenames = tf.constant(["/var/data/image1.jpg", "/var/data/image2.jpg", ...])
labels = tf.constant([0, 37, ...])
dataset = tf.data.Dataset.from_tensor_slices((filenames, labels))
dataset.shuffle()

def _parse_function(filename, label):
  image_string = tf.read_file(filename)
  image_decoded = tf.image.decode_jpeg(image_string)
  image_resized = tf.image.resize_images(image_decoded, [28, 28])
  return image_resized, label

dataset = dataset.map(_parse_function)
iterator = dataset.make_one_shot_iterator()
X, Y = iterator.get_next()

sess = tf.Session()
sess.run(iterator.initializer)
while True:
  sess.run(X) #Here I want the element from filenames being used for X

我以为 iterator 中可以包含此信息，但找不到。

Answer 1

您只需要在数据集中保留文件名和图像数据：

filenames = tf.constant(["/var/data/image1.jpg", "/var/data/image2.jpg", ...])
labels = tf.constant([0, 37, ...])
dataset = tf.data.Dataset.from_tensor_slices((filenames, labels))
dataset.shuffle()

def _parse_function(filename, label):
  image_string = tf.read_file(filename)
  image_decoded = tf.image.decode_jpeg(image_string)
  image_resized = tf.image.resize_images(image_decoded, [28, 28])
  return filename, image_resized, label

dataset = dataset.map(_parse_function)
iterator = dataset.make_one_shot_iterator()
F, X, Y = iterator.get_next()

sess = tf.Session()
sess.run(iterator.initializer)
while True:
  sess.run(F, X)

从张量流数据集迭代器获取输入（文件名）

Get input (filenames) from tensorflow dataset iterators

python

iterator

tensorflow

tensorflow-datasets