为什么在常量张量的创建过程中使用 "CopyFrom"？
Why "CopyFrom" is used during the creation of the constant Tensor?
常量Tensor的创建过程中有如下line:
tensor_value.tensor.CopyFrom(
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape))
CopyFrom 创建新创建的 Tensor 原型的副本。然而，这看起来像是浪费资源来应对，因为根据文档，make_tensor_proto 创建了一个新对象。是不是更够了，就做下一步：
tensor_value.tensor =
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape)
这不应创建新对象，而且它也是 OneOf protobuf 字段的有效用法。
您不能将原型分配给原型的字段，如本文档中所述：https://developers.google.com/protocol-buffers/docs/reference/python-generated
You cannot assign a value to an embedded message field. Instead, assigning a value to any field within the child message implies setting the message field in the parent.
如果删除 CopyFrom，将出现以下错误：
AttributeError: Assignment not allowed to field "tensor" in protocol message object.
常量Tensor的创建过程中有如下line:
tensor_value.tensor.CopyFrom(
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape))
CopyFrom 创建新创建的 Tensor 原型的副本。然而，这看起来像是浪费资源来应对，因为根据文档，make_tensor_proto 创建了一个新对象。是不是更够了，就做下一步：
tensor_value.tensor =
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape)
这不应创建新对象，而且它也是 OneOf protobuf 字段的有效用法。
您不能将原型分配给原型的字段，如本文档中所述：https://developers.google.com/protocol-buffers/docs/reference/python-generated
You cannot assign a value to an embedded message field. Instead, assigning a value to any field within the child message implies setting the message field in the parent.
如果删除 CopyFrom，将出现以下错误：
AttributeError: Assignment not allowed to field "tensor" in protocol message object.