Python 如何使用tf-keras将tfrecord转换为numpy

我将数据存储在TFrecords中，并尝试使用Tensorflow分布式策略和keras模型处理它们

起初我编写了如下代码， （根据关于如何在keras模型中使用tfrecords的说明：）

但是在我运行代码后，它给出了错误：

Traceback (most recent call last):
  File "main2.py", line 108, in <module>
    target_tensors=[train_label])
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/training/tracking/base.py", line 457, in _method_wrapper
    result = method(self, *args, **kwargs)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/keras/engine/training.py", line 259, in compile
    weighted_metrics)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/keras/engine/training.py", line 1569, in _validate_compile_param_for_distribution_strategy
    raise ValueError('target_tensors is not supported with '
ValueError: target_tensors is not supported with tf.distribute.Strategy.

并得到以下错误：

2019-07-08 04:23:22.307255: W tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Feature: image (data type: string) is required but could not be found.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/keras/backend/tensorflow_backend.py", line 673, in eval
    return to_dense(x).eval(session=get_session())
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 731, in eval
    return _eval_using_default_session(self, feed_dict, self.graph, session)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 5579, in _eval_using_default_session
    return session.run(tensors, feed_dict)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 950, in run
    run_metadata_ptr)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1173, in _run
    feed_dict_tensor, options, run_metadata)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1350, in _do_run
    run_metadata)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1370, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Feature: image (data type: string) is required but could not be found.
     [[{{node ParseSingleExample/ParseSingleExample}}]]
     [[IteratorGetNext]]

2019-07-08 04:23:22.307255:W tensorflow/core/framework/op_kernel.cc:1502]op_REQUIRES在示例_解析_ops时失败。cc:240:无效参数：需要功能：图像（数据类型：字符串），但找不到。
回溯（最近一次呼叫最后一次）：
文件“”，第1行，在
文件“/ma1gpu07_nfsv4/zhangxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/keras/backend/tensorflow_backend.py”，第673行，评估
返回到_dense（x）.eval（session=get_session（））
文件“/ma1gpu07_nfsv4/zhungxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/framework/ops.py”，第731行，在eval中
使用默认会话返回评估会话（self、feed、dict、self.graph、session）
文件“/ma1gpu07\u nfsv4/zhangxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/framework/ops.py”，第5579行，使用默认会话进行评估
返回会话。运行（张量、提要）
文件“/ma1gpu07_nfsv4/zhungxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/client/session.py”，第950行，正在运行
运行_元数据_ptr）
文件“/ma1gpu07_nfsv4/zhungxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/client/session.py”，第1173行，正在运行
feed_dict_tensor、options、run_元数据）
文件“/ma1gpu07_nfsv4/zhungxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/client/session.py”，第1350行，在运行中
运行（元数据）
文件“/ma1gpu07_nfsv4/zhungxy/anaconda3/envs/dlipy2/lib/python2.7/site packages/tensorflow/python/client/session.py”，第1370行，在
提升类型（e）（节点定义、操作、消息）
tensorflow.python.framework.errors\u impl.InvalidArgumentError:Feature:image（数据类型：string）是必需的，但找不到。
[{{node ParseSingleExample/ParseSingleExample}}]]
[[IteratorGetNext]]

有人知道我如何解决这个问题吗？ 非常感谢

from keras import backend as K
K.eval(train_image)

2019-07-08 04:23:22.307255: W tensorflow/core/framework/op_kernel.cc:1502] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Feature: image (data type: string) is required but could not be found.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/keras/backend/tensorflow_backend.py", line 673, in eval
    return to_dense(x).eval(session=get_session())
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 731, in eval
    return _eval_using_default_session(self, feed_dict, self.graph, session)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 5579, in _eval_using_default_session
    return session.run(tensors, feed_dict)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 950, in run
    run_metadata_ptr)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1173, in _run
    feed_dict_tensor, options, run_metadata)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1350, in _do_run
    run_metadata)
  File "/ma1gpu07_nfsv4/zhuangxy/anaconda3/envs/dlipy2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1370, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Feature: image (data type: string) is required but could not be found.
     [[{{node ParseSingleExample/ParseSingleExample}}]]
     [[IteratorGetNext]]