How to multiply two feature map pixel by pixel in caffe?

Question

When I convert onnx model to caffe mode, there is a problem. I want to adding two feature map pixel by pixel, like matrix shape A is (3x80x80x2) and B is (3x80x80x2). And A is from caffe layer output, but B is a constant matrix (from learned param in onnx). I consider use Eltwise layer in caffe but the layer must have two input blob. Then I use Bias layer and set the only 1 input blob, because if Bias layer have only 1 input, it will use filler blob to adding.

Bias layer have filler parameter, how can I fit my constant matrix data to the filler blob using pycaffe?

my prototxt like this:

layer {
  name: "bias"
  type: "Bias"
  bottom: "A"
  top: "bias_output"
}

my code like this:

layer = net.params['bias'][0].data[...] = B 

raise some error: shape not match, error: B shape is (3x80x80x2), net.params['bias'][0].data shape is (3,)

or other ideas? How can I solve this problem?

in caffe src prototxt, BiasParameter define in below.

message BiasParameter {
  // The first axis of bottom[0] (the first input Blob) along which to apply
  // bottom[1] (the second input Blob).  May be negative to index from the end
  // (e.g., -1 for the last axis).
  //
  // For example, if bottom[0] is 4D with shape 100x3x40x60, the output
  // top[0] will have the same shape, and bottom[1] may have any of the
  // following shapes (for the given value of axis):
  //    (axis == 0 == -4) 100; 100x3; 100x3x40; 100x3x40x60
  //    (axis == 1 == -3)          3;     3x40;     3x40x60
  //    (axis == 2 == -2)                   40;       40x60
  //    (axis == 3 == -1)                                60
  // Furthermore, bottom[1] may have the empty shape (regardless of the value of
  // "axis") -- a scalar bias.
  optional int32 axis = 1 [default = 1];

  // (num_axes is ignored unless just one bottom is given and the bias is
  // a learned parameter of the layer.  Otherwise, num_axes is determined by the
  // number of axes by the second bottom.)
  // The number of axes of the input (bottom[0]) covered by the bias
  // parameter, or -1 to cover all axes of bottom[0] starting from `axis`.
  // Set num_axes := 0, to add a zero-axis Blob: a scalar.
  optional int32 num_axes = 2 [default = 1];

  // (filler is ignored unless just one bottom is given and the bias is
  // a learned parameter of the layer.)
  // The initialization for the learned bias parameter.
  // Default is the zero (0) initialization, resulting in the BiasLayer
  // initially performing the identity operation.
  optional FillerParameter filler = 3;
}

and