深度学习笔记（1）：caffe 添加新层 attention LSTM layer和LSTM layer代码精读

总结一下最近的工作：LSTM layer 代码，caffe 加入新层 Attention LSTM layer

LSTM layer

关键代码如下，可以参考图1进行阅读，图一来自博客

namespace caffe {

template 
void LSTMLayer::RecurrentInputBlobNames(vector* names) const {
  names->resize(2);
  (*names)[0] = "h_0";
  (*names)[1] = "c_0";   //定义h_0,c_0 的输入
}                      

template 
void LSTMLayer::RecurrentOutputBlobNames(vector* names) const {
  names->resize(2);
  (*names)[0] = "h_" + this->int_to_str(this->T_);
  (*names)[1] = "c_T";  // 定义输出，不同时刻的h_t
}

template 
void LSTMLayer::OutputBlobNames(vector* names) const {
  names->resize(1);
  (*names)[0] = "h";  // 最终输出h
}

template 
void LSTMLayer::FillUnrolledNet(NetParameter* net_param) const {
  const int num_output = this->layer_param_.recurrent_param().num_output();
  CHECK_GT(num_output, 0) << "num_output must be positive";
  const FillerParameter& weight_filler =
      this->layer_param_.recurrent_param().weight_filler();
  const FillerParameter& bias_filler =
      this->layer_param_.recurrent_param().bias_filler(); // 权重W和偏差Bias

  // Add generic LayerParameter's (without bottoms/tops) of layer types we'll
  // use to save redundant code.
  LayerParameter hidden_param;
  hidden_param.set_type("InnerProduct");
  hidden_param.mutable_inner_product_param()->set_num_output(num_output * 4);
  hidden_param.mutable_inner_product_param()->set_bias_term(false);
  hidden_param.mutable_inner_product_param()->set_axis(2);
  hidden_param.mutable_inner_product_param()->
      mutable_weight_filler()->CopyFrom(weight_filler);

  LayerParameter biased_hidden_param(hidden_param);
  biased_hidden_param.mutable_inner_product_param()->set_bias_term(true);
  biased_hidden_param.mutable_inner_product_param()->
      mutable_bias_filler()->CopyFrom(bias_filler);

  LayerParameter sum_param;
  sum_param.set_type("Eltwise");
  sum_param.mutable_eltwise_param()->set_operation(
      EltwiseParameter_EltwiseOp_SUM);

  LayerParameter slice_param;
  slice_param.set_type("Slice");
  slice_param.mutable_slice_param()->set_axis(0);

  LayerParameter split_param;
  split_param.set_type("Split");

  BlobShape input_shape;
  input_shape.add_dim(1);  //      eg:1x3x256,256是lstm层设置的，3为buffersize或则通道数
  input_shape.add_dim(this->N_);
  input_shape.add_dim(num_output);

  net_param->add_input("c_0");
  net_param->add_input_shape()->CopyFrom(input_shape);

  net_param->add_input("h_0");
  net_param->add_input_shape()->CopyFrom(input_shape);

  LayerParameter* cont_slice_param = net_param->add_layer();
  cont_slice_param->CopyFrom(slice_param);
  cont_slice_param->set_name("cont_slice");
  cont_slice_param->add_bottom("cont");
  cont_slice_param->mutable_slice_param()->set_axis(1);  // cont 为0或则1，slice

  // Add layer to transform all timesteps of x to the hidden state dimension.
  //     W_xc_x = W_xc * x + b_c
  {
    LayerParameter* x_transform_param = net_param->add_layer();
    x_transform_param->CopyFrom(biased_hidden_param);
    x_transform_param->set_name("x_transform");
    x_transform_param->add_param()->set_name("W_xc");
    x_transform_param->add_param()->set_name("b_c");
    x_transform_param->add_bottom("x");
    x_transform_param->add_top("W_xc_x");//全连接层，W x X+ b ,实际上对应着下图的维度的变化，最后一维必须要变成1024
  }                                  //这样才能和4个gate对应，因为W x X+ b 包含了4个gate，可以看做[[],[],[],[]].因此维度需要为1024,！

  LayerParameter* x_slice_param = net_param->add_layer();
  x_slice_param->CopyFrom(slice_param);
  x_slice_param->add_bottom("W_xc_x");
  x_slice_param->set_name("W_xc_x_slice");  //维度变换完成后，需要slice                                  //举例，一个视频，10帧，则一帧一帧的输入，这样应该好理解一点，参考下图
  LayerParameter output_concat_layer;
  output_concat_layer.set_name("h_concat");
  output_concat_layer.set_type("Concat");
  output_concat_layer.add_top("h");
  output_concat_layer.mutable_concat_param()->set_axis(0); //lstm每一个T都输出h_t,至于你取哪一个或者都去，取决于你的prototxt代码的编写

  for (int t = 1; t <= this->T_; ++t) {    // 开始实现 平铺 LSTM 层，可以看 这里
    string tm1s = this->int_to_str(t - 1);
    string ts = this->int_to_str(t);

    cont_slice_param->add_top("cont_" + ts);
    x_slice_param->add_top("W_xc_x_" + ts);  

    // Add layers to flush the hidden state when beginning a new
    // sequence, as indicated by cont_t.
    //     h_conted_{t-1} := cont_t * h_{t-1}
    //
    // Normally, cont_t is binary (i.e., 0 or 1), so:
    //     h_conted_{t-1} := h_{t-1} if cont_t == 1
    //                       0   otherwise
    {
      LayerParameter* cont_h_param = net_param->add_layer();
      cont_h_param->CopyFrom(sum_param);
      cont_h_param->mutable_eltwise_param()->set_coeff_blob(true);
      cont_h_param->set_name("h_conted_" + tm1s);
      cont_h_param->add_bottom("h_" + tm1s);
      cont_h_param->add_bottom("cont_" + ts);
      cont_h_param->add_top("h_conted_" + tm1s); // h_conted 的产生，这个根据cont_slice 决定，与cont_slice 相乘

    }

    // Add layer to compute
    //     W_hc_h_{t-1} := W_hc * h_conted_{t-1}
    {
      LayerParameter* w_param = net_param->add_layer();
      w_param->CopyFrom(hidden_param);
      w_param->set_name("transform_" + ts);
      w_param->add_param()->set_name("W_hc");
      w_param->add_bottom("h_conted_" + tm1s);
      w_param->add_top("W_hc_h_" + tm1s);
      w_param->mutable_inner_product_param()->set_axis(2); // 计算 W X H
    }

    // Add the outputs of the linear transformations to compute the gate input.
    //     gate_input_t := W_hc * h_conted_{t-1} + W_xc * x_t + b_c
    //                   = W_hc_h_{t-1} + W_xc_x_t + b_c
    {
      LayerParameter* input_sum_layer = net_param->add_layer();
      input_sum_layer->CopyFrom(sum_param);
      input_sum_layer->set_name("gate_input_" + ts);
      input_sum_layer->add_bottom("W_hc_h_" + tm1s);
      input_sum_layer->add_bottom("W_xc_x_" + ts);
      if (this->static_input_) {
        input_sum_layer->add_bottom("W_xc_x_static");
      } 
      input_sum_layer->add_top("gate_input_" + ts);     //根据公式，计算4个gate的输出
    } 

    // Add LSTMUnit layer to compute the cell & hidden vectors c_t and h_t.
    // Inputs: c_{t-1}, gate_input_t = (i_t, f_t, o_t, g_t), cont_t
    // Outputs: c_t, h_t
    //     [ i_t' ]
    //     [ f_t' ] := gate_input_t
    //     [ o_t' ]
    //     [ g_t' ]
    //         i_t := \sigmoid[i_t']
    //         f_t := \sigmoid[f_t']
    //         o_t := \sigmoid[o_t']
    //         g_t := \tanh[g_t']
    //         c_t := cont_t * (f_t .* c_{t-1}) + (i_t .* g_t)
    //         h_t := o_t .* \tanh[c_t]
    {
      LayerParameter* lstm_unit_param = net_param->add_layer();
      lstm_unit_param->set_type("LSTMUnit");
      lstm_unit_param->add_bottom("c_" + tm1s);
      lstm_unit_param->add_bottom("gate_input_" + ts);
      lstm_unit_param->add_bottom("cont_" + ts);
      lstm_unit_param->add_top("c_" + ts);
      lstm_unit_param->add_top("h_" + ts);
      lstm_unit_param->set_name("unit_" + ts);  //4 个 gate 输入值进行不同函数的非线性变换
    }
    output_concat_layer.add_bottom("h_" + ts);  // 中间隐藏层的状态输出
  }  // for (int t = 1; t <= this->T_; ++t)

  {
    LayerParameter* c_T_copy_param = net_param->add_layer();
    c_T_copy_param->CopyFrom(split_param);
    c_T_copy_param->add_bottom("c_" + this->int_to_str(this->T_));
    c_T_copy_param->add_top("c_T");            
  }
  net_param->add_layer()->CopyFrom(output_concat_layer); //split：将C_T复制几份，分别给不同的layer，这些上层layer共享这个blob
}

INSTANTIATE_CLASS(LSTMLayer);
REGISTER_LAYER_CLASS(LSTM);  //LSTM layer 的注册

}  // namespace caffe

   上面是LSTMlayer的代码分析，整体流程比较清晰，但是由于Attention Model 的提出，很多模型需要attention LSTM，而caffe 没有alstm layer，因此博主自己写了个ALSTM layer，由于篇幅比较长，换下一篇写。