意外需要为图中不相关的占位符提供输入值

Question

https://gist.github.com/Wermarter/466e9585579ef65927fa934fe4e0ffd4 在这里，我尝试使用 TFLearn 在 Tensorflow 中实现 Variational AutoEncoder。

我在 self.training_model.session 的一个大图中构建了用于训练、编码和生成的计算。 self.generating_model 和 self.recognition_model 与 self.training_model.

共享同一个会话

当我运行 generating_model 生成 MNIST 2D Latent space 时一切顺利。但是当我运行 self.recognition_model 对给定的 input_data 进行编码时出现错误，它要求我给属于 self.training_model 的 self.train_data 输入值。

这是完整的错误：

Traceback (most recent call last):
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1039, in _do_call
    return fn(*args)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1021, in _run_fn
    status, run_metadata)
  File "/home/wermarter/anaconda3/lib/python3.5/contextlib.py", line 66, in __exit__
    next(self.gen)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/errors_impl.py", line 466, in raise_exception_on_not_ok_status
    pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'train_data/X' with dtype float
     [[Node: train_data/X = Placeholder[dtype=DT_FLOAT, shape=[], _device="/job:localhost/replica:0/task:0/gpu:0"]()]]
     [[Node: add_5/_47 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_8_add_5", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/wermarter/Desktop/vae.py", line 178, in <module>
    main()
  File "/home/wermarter/Desktop/vae.py", line 172, in main
    vae.img_transition(trainX[4], trainX[100])
  File "/home/wermarter/Desktop/vae.py", line 130, in img_transition
    enc_A = self.encode(A)[0]
  File "/home/wermarter/Desktop/vae.py", line 121, in encode
    return self.recognition_model.predict({self.input_data: input_data})
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tflearn/models/dnn.py", line 257, in predict
    return self.predictor.predict(feed_dict)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tflearn/helpers/evaluator.py", line 69, in predict
    return self.session.run(self.tensors[0], feed_dict=feed_dict)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 778, in run
    run_metadata_ptr)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 982, in _run
    feed_dict_string, options, run_metadata)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1032, in _do_run
    target_list, options, run_metadata)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1052, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'train_data/X' with dtype float
     [[Node: train_data/X = Placeholder[dtype=DT_FLOAT, shape=[], _device="/job:localhost/replica:0/task:0/gpu:0"]()]]
     [[Node: add_5/_47 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_8_add_5", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

Caused by op 'train_data/X', defined at:
  File "/home/wermarter/Desktop/vae.py", line 178, in <module>
    main()
  File "/home/wermarter/Desktop/vae.py", line 169, in main
    vae = VAE()
  File "/home/wermarter/Desktop/vae.py", line 28, in __init__
    self._build_training_model()
  File "/home/wermarter/Desktop/vae.py", line 78, in _build_training_model
    self.train_data = tflearn.input_data(shape=[None, *self.img_shape], name='train_data')
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tflearn/layers/core.py", line 81, in input_data
    placeholder = tf.placeholder(shape=shape, dtype=dtype, name="X")
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/ops/array_ops.py", line 1507, in placeholder
    name=name)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/ops/gen_array_ops.py", line 1997, in _placeholder
    name=name)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/op_def_library.py", line 768, in apply_op
    op_def=op_def)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 2336, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "/home/wermarter/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 1228, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'train_data/X' with dtype float
     [[Node: train_data/X = Placeholder[dtype=DT_FLOAT, shape=[], _device="/job:localhost/replica:0/task:0/gpu:0"]()]]
     [[Node: add_5/_47 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_8_add_5", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

Answer 1

这是特定于代码的错误。我的 self.recognition_model 实际上是 link 通过 self._sample_z() 中的 self.curr_batch_size 编辑到占位符 self.train_data。我的解决办法是重新linkself.curr_batch_size到self.input_data.

的大小

就是这样。快乐编码

意外需要为图中不相关的占位符提供输入值

Unexpectedly needed to give input value to an irrelevant placeholder in the graph

graph

tensorflow

tflearn