【caffe】Layer解读之：Reduction

心已赠人 2022-05-17 12:14 207阅读 0赞

*  Layer type: Reduction
 *  头文件位置：./include/caffe/layers/reduction\_layer.hpp
 *  CPU 执行源文件位置: ./src/caffe/layers/reduction\_layer.cpp
 *  CUDA GPU 执行源文件位置: ./src/caffe/layers/reduction\_layer.cu
 *  Reduction层的功能：使用sum或mean等操作作用于输入blob按照给定参数规定的维度。（通俗的讲就是将输入的特征图按照给定的维度进行求和或求平均）。

##### 参数解读 #####

layer { name: "rnn_o_concat" type: "Concat" bottom: "o_1" bottom: "o_2" bottom: "o_3" top: "o" concat_param { axis: 0 }
    }
    layer { name: "o_pseudoloss" type: "Reduction" bottom: "o" top: "o_pseudoloss" loss_weight: 1 reduction_param{ axis: 0//default = 0 optional ReductionOp operation = SUM //default = SUM }
    }

根据Reduction的定义，我们知道就是将输入的特征图“o”的第一个维度的特征图都加起来。concat\_layer的用法见：[【caffe】Layer解读之：Concat][caffe_Layer_Concat]

##### 参数定义 #####

参数(ReductionParameter reduction\_param)  
定义位置 ./src/caffe/proto/caffe.proto:

// Message that stores parameters used by ReductionLayer
    message ReductionParameter {
      enum ReductionOp {
        SUM = 1;
        ASUM = 2;
        SUMSQ = 3;
        MEAN = 4;
      }
    
      optional ReductionOp operation = 1 [default = SUM]; // reduction operation
    
      // The first axis to reduce to a scalar -- may be negative to index from the
      // end (e.g., -1 for the last axis).
      // (Currently, only reduction along ALL "tail" axes is supported; reduction
      // of axis M through N, where N < num_axes - 1, is unsupported.)
      // Suppose we have an n-axis bottom Blob with shape:
      //     (d0, d1, d2, ..., d(m-1), dm, d(m+1), ..., d(n-1)).
      // If axis == m, the output Blob will have shape
      //     (d0, d1, d2, ..., d(m-1)),
      // and the ReductionOp operation is performed (d0 * d1 * d2 * ... * d(m-1))
      // times, each including (dm * d(m+1) * ... * d(n-1)) individual data.
      // If axis == 0 (the default), the output Blob always has the empty shape
      // (count 1), performing reduction across the entire input --
      // often useful for creating new loss functions.
      optional int32 axis = 2 [default = 0];
    
      optional float coeff = 3 [default = 1.0]; // coefficient for output
    }

##### 番外篇 #####

若想用Reduction解决如下问题：  
**我想把n\*c\*h\*w的blob变成n\*1\*h\*w（使用SUM），param该如何设定？**  
**答：ReductionLayer 干不了这事儿。因为它只支持从你指定的axis到tail axis为止的reduction操作。它不支持针对某个坐标轴独立做reduction，而是从某个坐标轴开始做到最后一个坐标轴。即无论你的指定哪个坐标轴，它都会默认reduce做到最后一个坐标轴的。n\*c\*h\*w, 你若设axis=1，它就变成n, 若你指定2，它就变成n\*c。**

正确解决方案（来自网络）：  
**答：你可以用convolutionLayer做你想做的事儿。如下，注意，kernel\_size，num\_output，weight的值都必须为1才能做到这个效果。**

convolution_param { num_output: 1 bias_term: false kernel_size: 1 weight_filler { type: 'constant' value: 1 } 
    }

[caffe_Layer_Concat]: https://blog.csdn.net/qiu931110/article/details/81584929