Caffe框架源码剖析(5)—卷积层ConvolutionLayer

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/tianrolin/article/details/52571548

ConvolutionLayer是BaseConvolutionLayer的子类，功能较为简单。类中不包含成员变量，仅包含几个虚函数的实现。

conv_layer.hpp头文件的定义如下：

template <typename Dtype>
class ConvolutionLayer : public BaseConvolutionLayer<Dtype> {
 public:
  explicit ConvolutionLayer(const LayerParameter& param)
      : BaseConvolutionLayer<Dtype>(param) {}
  virtual inline const char* type() const { return "Convolution"; }

 protected:
  // CPU正向传导虚函数
  virtual void Forward_cpu(const vector<Blob<Dtype>*>& bottom,
      const vector<Blob<Dtype>*>& top);
  // GPU正向传导虚函数
  virtual void Forward_gpu(const vector<Blob<Dtype>*>& bottom,
      const vector<Blob<Dtype>*>& top);
  // CPU反向传导虚函数
  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
  // GPU反向传导虚函数
  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);
  // 标明是卷积过程（反卷积层则return true）
  virtual inline bool reverse_dimensions() { return false; }
  // 计算卷积后输出尺寸虚函数
  virtual void compute_output_shape();
};

conv_layer.cpp定义如下，

// 计算卷积后输出尺寸（虚函数，被父类中的Reshape函数调用）
template <typename Dtype>
void ConvolutionLayer<Dtype>::compute_output_shape() {
  const int* kernel_shape_data = this->kernel_shape_.cpu_data();
  const int* stride_data = this->stride_.cpu_data();
  const int* pad_data = this->pad_.cpu_data();
  const int* dilation_data = this->dilation_.cpu_data();
  this->output_shape_.clear();
  for (int i = 0; i < this->num_spatial_axes_; ++i) {
    // i + 1 to skip channel axis
    const int input_dim = this->input_shape(i + 1);
    const int kernel_extent = dilation_data[i] * (kernel_shape_data[i] - 1) + 1;
    const int output_dim = (input_dim + 2 * pad_data[i] - kernel_extent)
        / stride_data[i] + 1;
    this->output_shape_.push_back(output_dim);
  }
}

// CPU正向传导
template <typename Dtype>
void ConvolutionLayer<Dtype>::Forward_cpu(const vector<Blob<Dtype>*>& bottom,
      const vector<Blob<Dtype>*>& top) {
  // blobs_[0]中存放的是卷积核
  const Dtype* weight = this->blobs_[0]->cpu_data();
  for (int i = 0; i < bottom.size(); ++i) {
    const Dtype* bottom_data = bottom[i]->cpu_data();
    Dtype* top_data = top[i]->mutable_cpu_data();
    for (int n = 0; n < this->num_; ++n) {
      // 调用父类函数进行卷积
      this->forward_cpu_gemm(bottom_data + n * this->bottom_dim_, weight,
          top_data + n * this->top_dim_);
      // 如果含有偏置项，则加上偏置项
      if (this->bias_term_) {
        // blobs_[1]中存放的是偏置项
        const Dtype* bias = this->blobs_[1]->cpu_data();
        this->forward_cpu_bias(top_data + n * this->top_dim_, bias);
      }
    }
  }
}

// CPU反向传导
template <typename Dtype>
void ConvolutionLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top,
      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom) {
  const Dtype* weight = this->blobs_[0]->cpu_data();
  Dtype* weight_diff = this->blobs_[0]->mutable_cpu_diff();
  for (int i = 0; i < top.size(); ++i) {
    const Dtype* top_diff = top[i]->cpu_diff();
    const Dtype* bottom_data = bottom[i]->cpu_data();
    Dtype* bottom_diff = bottom[i]->mutable_cpu_diff();
    // 计算偏置的偏导
    if (this->bias_term_ && this->param_propagate_down_[1]) {
      Dtype* bias_diff = this->blobs_[1]->mutable_cpu_diff();
      for (int n = 0; n < this->num_; ++n) {
        this->backward_cpu_bias(bias_diff, top_diff + n * this->top_dim_);
      }
    }
    if (this->param_propagate_down_[0] || propagate_down[i]) {
      for (int n = 0; n < this->num_; ++n) {
        // bottom_data * top_diff -> weight_diff
        if (this->param_propagate_down_[0]) {
          this->weight_cpu_gemm(bottom_data + n * this->bottom_dim_,
              top_diff + n * this->top_dim_, weight_diff);
        }
        // top_diff * weight -> bottom_diff
        if (propagate_down[i]) {
          this->backward_cpu_gemm(top_diff + n * this->top_dim_, weight,
              bottom_diff + n * this->bottom_dim_);
        }
      }
    }
  }
}

// 如果CPU_ONLY模式则禁止Forward_gpu和Backward_gpu函数
#ifdef CPU_ONLY
STUB_GPU(ConvolutionLayer);
#endif