Tensorflow 更改图层的名称'；张量信息？

我有一个带有输入层

bar

和输出层

output\u层

的模型。

saved\u model\u cli

的输出给出了

signature_def['serving_default']:
  The given SavedModel SignatureDef contains the following input(s):
    inputs['bar'] tensor_info:
        dtype: DT_INT32
        shape: (-1, 1)
        name: serving_default_bar:0
  The given SavedModel SignatureDef contains the following output(s):
    outputs['output_layer'] tensor_info:
        dtype: DT_FLOAT
        shape: (-1, 6)
        name: StatefulPartitionedCall:0
  Method name is: tensorflow/serving/predict

有没有办法将

outputs['output\u layer']tensor\u info

中的名称从

StatefulPartitionedCall

更改为自定义名称