ReLU after classifier

[ducha-aiki]

Hi, I have noticed that you put ReLU after classifier, which is not a common practice. Is there some reason for it?

layer {
  name: "conv10"
  type: "Convolution"
  bottom: "fire9/concat"
  top: "conv10"
  convolution_param {
    num_output: 1000
    kernel_size: 1
    weight_filler {
      type: "gaussian"
      mean: 0.0
      std: 0.01
    }
  }
}
layer {
  name: "relu_conv10"
  type: "ReLU"
  bottom: "conv10"
  top: "conv10"
}
layer {
  name: "pool10"
  type: "Pooling"
  bottom: "conv10"
  top: "pool10"
  pooling_param {
    pool: AVE
    global_pooling: true
  }
}

[forresti]

Hmm... no reason. Feel free to try inference (or retrain + inference) without this final ReLU.