How to handle ImageScaler operator of ONNX?

Questions
1

ImageScaler: (https://github.com/onnx/onnx/blob/master/docs/Operators.md)
Scale and bias the input image. Bias values are stored in the same ordering as the image pixel format.

Attributes
bias : list of floats
Bias applied to each channel, same size as C.
scale : float
(float, default 1.0) the scale to apply.

In our current implementation framework of ONNX, we would like to handle like this:

class Scale(OnnxOpConverter):

@classmethod
def _impl_v1(cls, inputs, attr, params):
    scale = float(attr.get('scale', 1.0))
    scaled_sym = inputs[0] * scale

Then how do we handle lists of bias values? We can not use scaled_sym + attr.get(‘bias’), because attr.get(‘bias’) is type of tuple, which is not symbol or number type. We also can not use _sym.Variable(attr.get(‘bias’)), because we can not create here, all Symbol Variable should be all created in the:

 for i in graph.input:
        # from onnx v0.2, GraphProto.input has type ValueInfoProto,
        #  and the name is 'i.name'
        i_name = self._parse_value_proto(i)
        if i_name in self._params:
            # i is a param instead of input
            name_param = 'param_{}'.format(self._num_param)
            self._num_param += 1
            self._params[name_param] = self._params.pop(i_name)
            self._nodes[name_param] = _sym.Variable(
                name=name_param, shape=self._params[name_param].shape)
            self._renames[i_name] = name_param
        else:
            name_input = 'input_{}'.format(self._num_input)
            self._num_input += 1
            self._nodes[name_input] = _sym.Variable(name=name_input)
            self._renames[i_name] = name_input
    # construct nodes, nodes are stored as directed acyclic graph

Many thanks in advance.