将参数从训练传递到推理图

为了不将优化器和渐变节点带入推理环境,我正在尝试创建两个版本的图形 - 一个带有训练节点,另一个没有.

并且想法是用于tensorflow.train.Saver将变量从训练图形版本传递到推理图形版本.

所以我尝试了以下内容:

# Create training graph
trainingGraph = tf.Graph()
with (trainingGraph.as_default()):
  trainOp, lossOp = self.CreateTrainingGraph()
  trainInitOp = tf.initialize_variables(tf.all_variables(), "init_variables")

  # Add saver op
  self.saverOp = tf.train.Saver()

# Create inference graph
inferenceGraph = tf.Graph()
with (inferenceGraph.as_default()):
  self.CreateInferenceGraph()

  # Add saver op, compatible with training saver
  tf.train.Saver(saver_def=self.saverOp.as_saver_def())

sess.run([model.saverOp._restore_op_name],
         { model.saverOp._filename_tensor_name : "Params/data.pb" })

 returned a result with an error set