Forums - ERROR: CAFFE LAYER OF TYPE SCALE NOT PRECEEDED BY BATCHNORM

Dear Sir or Madam,

I've encountered a problem when converting my caffemodel into dlc file. Basically, the error message is flagged out as below:

ERROR_CAFFE_LAYER_OF_TYPE_SCALE_NOT_PRECEEDED_BY_BATCHNORM: Cannot resolve Scale layer conv1_1/negative as it is not preceded by a BatchNorm layer into which it can be folded.

For your information, the concerned snippet of my prototxt file is pasted below:

---------------------------------
layer {
  name: "data"
  type: "Input"
  top: "data"
  input_param: {
    shape: {
      dim: 1
      dim: 3
      dim: 720
      dim: 1280
    }
  }
}
layer {
  name: "conv1_1/conv"
  type: "Convolution"
  bottom: "data"
  top: "conv1_1/conv"
  convolution_param {
    num_output: 16
    bias_term: false
    pad: 3
    kernel_size: 7
    stride: 2
  }
}
layer {
  name: "conv1_1/bn"
  type: "BatchNorm"
  bottom: "conv1_1/conv"
  top: "conv1_1/conv"
  batch_norm_param {
    use_global_stats: true
  }
}
layer {
  name: "conv1_1/negative"
  type: "Scale"
  bottom: "conv1_1/conv"
  top: "conv1_1/negative"
  scale_param {
    num_axes: 1
    bias_term: false
  }
}
layer {
  name: "conv1_1/concat"
  type: "Concat"
  bottom: "conv1_1/conv"
  bottom: "conv1_1/negative"
  top: "conv1_1/concat"
}
layer {
  name: "conv1_1/scale"
  type: "Scale"
  bottom: "conv1_1/concat"
  top: "conv1_1/concat"
  scale_param {
    num_axes: 1
    bias_term: true
  }
}

------------------------

To clarigy, BatchNorm layer conv1_1/bn outputs conv1_1/conv, which is passed into a Scale layer conv1_1/negative to get the negative value of  conv1_1/conv, the negative values conv1_1/negative are then concatenated with conv1_1/conv, and the combined blob is then feed into a Scale layer conv1_1/scale to accomplish the batchNorm+scale task.

Would you please share your thoughts on how to avoid this error message?

Cheers,