Tensorflow 保存SavedModel:ValueError:至少有两个变量具有相同的名称

我有一个保存SavedModel的函数

在培训期间，每隔N个时期会不时调用此函数，以将当前保存的模型替换为较新的模型

def save_saved模型（sess、快照目录）：
示例=tf.get_集合（“示例”）[0]
label=tf.get_集合（“label”）[0]
softmax=tf.get_集合（“softmax”）[0]
logits=tf.get_集合（“logits”）[0]
savedmodel\u dir=快照\u dir+“/saved\u model”
rmtree（savedmodel\u dir，ignore\u errors=True）
tf.saved\u model.simple\u save(
sess，
savedmodel_目录，
输入={“示例”：示例}，
输出={“softmax”：softmax，“logits”：logits}，
)

在我将tensorflow升级到1.12之前，脚本一直有效

现在，第一次调用此函数时，它就工作了，随后的调用会导致附加的stacktrace，抱怨有两个变量名为

batch\u normalization/beta/Adagrad

：

ValueError: At least two variables have the same name:batch_normalization/beta/Adagrad

第一个困惑是这不是真的。这就是我在调用simple_save（）之前转储所有变量时看到的情况。

有

batch\u normalization/beta/Adagrad

和

batch\u normalization/beta/Adagrad\u 1

：这是正常的还是可疑的

让我困惑的是，我只是在保存模型。我不是在恢复模型。那么，在我第二次调用此函数之前，我的图是否已经损坏，或者保存模型是否会以某种方式修改会话

Pass your op to the equivalent parameter main_op instead.
Traceback (most recent call last):
  File "src/tictactoe/train.py", line 412, in <module>
    train(sess)
  File "src/tictactoe/train.py", line 312, in train
    write_snapshot(sess, saver, global_step_val, "champion")
  File "src/tictactoe/train.py", line 185, in write_snapshot
    save_savedmodel(sess, snapshot_dir)
  File "src/tictactoe/train.py", line 209, in save_savedmodel
    outputs={"softmax": softmax, "logits": logits},
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/simple_save.py", line 85, in simple_save
    clear_devices=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 488, in new_func
    return func(*args, **kwargs)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/builder_impl.py", line 415, in add_meta_graph_and_variables
    saver = self._maybe_create_saver(saver)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/builder_impl.py", line 272, in _maybe_create_saver
    allow_empty=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1102, in __init__
    self.build()
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1114, in build
    self._build(self._filename, build_save=True, build_restore=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1151, in _build
    build_save=build_save, build_restore=build_restore)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 773, in _build_internal
    saveables = self._ValidateAndSliceInputs(names_to_saveables)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 673, in _ValidateAndSliceInputs
    names_to_saveables = BaseSaverBuilder.OpListToDict(names_to_saveables)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 572, in OpListToDict
    name)
ValueError: At least two variables have the same name: batch_normalization/beta/Adagrad

将op传递给等效参数main_op。
回溯（最近一次呼叫最后一次）：
文件“src/tictactoe/train.py”，第412行，在
列车（sess）
列车中第312行的文件“src/tictactoe/train.py”
写入快照（sess、保存程序、全局步骤、冠军）
文件“src/tictactoe/train.py”，第185行，写入快照
保存\u保存的模型（sess、快照\u目录）
文件“src/tictactoe/train.py”，第209行，在save_savedmodel中
输出={“softmax”：softmax，“logits”：logits}，
文件“/usr/local/lib/python3.5/dist packages/tensorflow/python/saved_model/simple_save.py”，第85行，simple_save
清除（设备=真）
文件“/usr/local/lib/python3.5/dist packages/tensorflow/python/util/deprecation.py”，第488行，在new_func中
返回函数（*args，**kwargs）
文件“/usr/local/lib/python3.5/dist packages/tensorflow/python/saved_model/builder_impl.py”，第415行，在add_meta_graph_和_变量中
saver=self.\u可能\u创建\u saver（saver）
文件“/usr/local/lib/python3.5/dist packages/tensorflow/python/saved_model/builder_impl.py”，第272行，在“创建”保存程序中
允许（空=真）
文件“/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py”，第1102行，在__
self.build（）
文件“/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py”，第1114行，内部版本
self.\u build（self.\u文件名，build\u save=True，build\u restore=True）
文件“/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py”，第1151行，内部版本
生成保存=生成保存，生成还原=生成还原）
文件“/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py”，第773行，内部版本
saveables=self.\u验证DSLICEInputs（名称\u到\u saveables）
文件“/usr/local/lib/python3.5/dist packages/tensorflow/python/training/saver.py”，第673行，在_ValidateAndSliceInputs中
names_to_saveables=BaseSaverBuilder.OpListToDict（names_to_saveables）
文件“/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py”，第572行，在OpListToDict中
（姓名）
ValueError：至少有两个变量具有相同的名称：batch_normalization/beta/Adagrad

batch_normalization/beta/Adagrad/Initializer/Const
batch_normalization/beta/Adagrad
batch_normalization/beta/Adagrad/Assign
batch_normalization/beta/Adagrad/read
batch_normalization/beta/Adagrad/Initializer/Const_1
batch_normalization/beta/Adagrad_1
batch_normalization/beta/Adagrad/Assign_1
batch_normalization/beta/Adagrad/read_1

Pass your op to the equivalent parameter main_op instead.
Traceback (most recent call last):
  File "src/tictactoe/train.py", line 412, in <module>
    train(sess)
  File "src/tictactoe/train.py", line 312, in train
    write_snapshot(sess, saver, global_step_val, "champion")
  File "src/tictactoe/train.py", line 185, in write_snapshot
    save_savedmodel(sess, snapshot_dir)
  File "src/tictactoe/train.py", line 209, in save_savedmodel
    outputs={"softmax": softmax, "logits": logits},
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/simple_save.py", line 85, in simple_save
    clear_devices=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 488, in new_func
    return func(*args, **kwargs)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/builder_impl.py", line 415, in add_meta_graph_and_variables
    saver = self._maybe_create_saver(saver)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/saved_model/builder_impl.py", line 272, in _maybe_create_saver
    allow_empty=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1102, in __init__
    self.build()
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1114, in build
    self._build(self._filename, build_save=True, build_restore=True)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1151, in _build
    build_save=build_save, build_restore=build_restore)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 773, in _build_internal
    saveables = self._ValidateAndSliceInputs(names_to_saveables)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 673, in _ValidateAndSliceInputs
    names_to_saveables = BaseSaverBuilder.OpListToDict(names_to_saveables)
  File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 572, in OpListToDict
    name)
ValueError: At least two variables have the same name: batch_normalization/beta/Adagrad