AttributeError: 模块“tensorflow._api.v1.data”没有属性“AUTOTUNE”

如何解决AttributeError: 模块“tensorflow._api.v1.data”没有属性“AUTOTUNE”

我是一名大学生。我正在使用 EfficientDet 模型完成我的毕业工作。 首先，我使用 Efficientdet 模型来微调我的自定义数据集。

python = 3.7
tensorflow = 2.5
tensorflow-gpu = 1.15
cuda = 11.0

这些是我安装的软件包的版本，

错误信息是

   =====> Starting training,epoch: 1.
WARNING:tensorflow:From /home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_core/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.
Instructions for updating:
Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.
W0527 01:25:16.531095 140153861850944 deprecation.py:323] From /home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_core/python/training/training_util.py:236: Variable.initialized_value (from tensorflow.python.ops.variables) is deprecated and will be removed in a future version.
Instructions for updating:
Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.
Traceback (most recent call last):
  File "main.py",line 407,in <module>
    app.run(main)
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/absl/app.py",line 303,in run
    _run_main(main,args)
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/absl/app.py",line 251,in _run_main
    sys.exit(main(argv))
  File "main.py",line 400,in main
    run_train_and_eval(e)
  File "main.py",line 384,in run_train_and_eval
    max_steps=e * FLAGS.num_examples_per_epoch // FLAGS.train_batch_size)
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/estimator.py",line 370,in train
    loss = self._train_model(input_fn,hooks,saving_listeners)
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/estimator.py",line 1161,in _train_model
    return self._train_model_default(input_fn,line 1188,in _train_model_default
    input_fn,ModeKeys.TRAIN))
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/estimator.py",line 1025,in _get_features_and_labels_from_input_fn
    self._call_input_fn(input_fn,mode))
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/estimator.py",line 1116,in _call_input_fn
    return input_fn(**kwargs)
  File "/home/ais-public/ms_ys/Liberty/automl/efficientdet/DataLoader.py",line 431,in __call__
    _prefetch_dataset,num_parallel_calls=tf.data.AUTOTUNE)
  File "/home/ais-public/anaconda3/envs/peace/lib/python3.7/site-packages/tensorflow_core/python/util/module_wrapper.py",line 193,in __getattr__
    attr = getattr(self._tfmw_wrapped_module,name)
AttributeError: module 'tensorflow._api.v1.data' has no attribute 'AUTOTUNE'

我找不到“tensorflow/_api/v1/data”文件夹，我不知道为什么会出现这个错误。 请告诉我如何解决这个错误。谢谢。

解决方法

尝试找到这个文件和行：

文件“/home/ais-public/ms_ys/Liberty/automl/efficientdet/dataloader.py”，第 >431 行，调用 _prefetch_dataset,num_parallel_calls=tf.data.AUTOTUNE)

并将 tf.data.AUTOTUNE 更改为 tf.data.experimental.AUTOTUNE。 您有一个较旧的 tensorflow-gpu 版本。据我所知，AUTOTUNE 不再是实验性的。

或者尝试只使用 tensorflow 2.5，以避免版本冲突。