caffe代码阅读4：DataTransformer以及io的实现细节-2016.3.16

一、DataTransformer的作用简介

该类主要负责对数据进行变换，将Datum、const vector<Datum>、cv::Mat&、vector<cv::Mat> 、Blob<Dtype>*类型的数据变换到目标大小的blob。

此外还负责对对上述类型的数据推断其shape。

该类用到了TransformationParameter。

其在caffe.proto的定义为

// Message that stores parameters used to apply transformation
// to the data layer's data
message TransformationParameter {
  // For data pre-processing, we can do simple scaling and subtracting the
  // data mean, if provided. Note that the mean subtraction is always carried
  // out before scaling.
  optional float scale = 1 [default = 1];
  // Specify if we want to randomly mirror data.
  optional bool mirror = 2 [default = false];
  // Specify if we would like to randomly crop an image.
  optional uint32 crop_size = 3 [default = 0];
  // mean_file and mean_value cannot be specified at the same time
  optional string mean_file = 4;
  // if specified can be repeated once (would substract it from all the channels)
  // or can be repeated the same number of times as channels
  // (would subtract them from the corresponding channel)
  repeated float mean_value = 5;
  // Force the decoded image to have 3 color channels.
  optional bool force_color = 6 [default = false];
  // Force the decoded image to have 1 color channels.
  optional bool force_gray = 7 [default = false];
}

二、DataTransformer类的详细介绍

1）构造函数

  // 构造函数

  explicit DataTransformer(const TransformationParameter& param, Phase phase);
  virtual ~DataTransformer() {}

2）成员变量

  // 变换所使用的参数
  TransformationParameter param_;
  // 随机数生成器的种子
  shared_ptr<Caffe::RNG> rng_;
  // 是训练还是测试？
  Phase phase_;
  // 数据均值 blob
  Blob<Dtype> data_mean_;
  // 数据均值blob的容器
  vector<Dtype> mean_values_;

3）成员函数

 // 初始化随机数生成器，因为在对数据进行变换的时候有可能用到，比如说打乱数据的输入顺序
  void InitRand();
   // 对Datum的数据进行变换，放入到transformed_blob中
  void Transform(const Datum& datum, Blob<Dtype>* transformed_blob);
   // 对Datum容器的数据进行变换翻入到transformed_blob
  void Transform(const vector<Datum> & datum_vector,
                Blob<Dtype>* transformed_blob);
   // 如果定义OpenCV还可能对mat容器数据类型的数据进行变换
  void Transform(const vector<cv::Mat> & mat_vector,
                Blob<Dtype>* transformed_blob);
   // 将opencv读取的单个图像转换到blob中去
  void Transform(const cv::Mat& cv_img, Blob<Dtype>* transformed_blob);
   // 将输入的blob进行变换，可能是取出blob的中的一部分数据到新的blob
  void Transform(Blob<Dtype>* input_blob, Blob<Dtype>* transformed_blob);
   // 根据Datum获取blob的形状
  vector<int> InferBlobShape(const Datum& datum);
   // 根据Datum容器获取blob的形状
  vector<int> InferBlobShape(const vector<Datum> & datum_vector);
  // 根据Mat容器获取blob的形状
  vector<int> InferBlobShape(const vector<cv::Mat> & mat_vector);
  // 根据Mat获取blob的形状
  vector<int> InferBlobShape(const cv::Mat& cv_img);
// 生成从0到n-1的服从均匀分布的随机数,要求继承他的都必须实现如何生成随机数
  virtual int Rand(int n);
  // 将给定的Datum进行转换
  void Transform(const Datum& datum, Dtype* transformed_data);

4）具体函数的实现：

首先是构造函数

在介绍构造函数之前不得不先贴出BlobShape和、BlobProto这两个结构体的在caffe.proto中的定义。
message BlobShape {
  repeated int64 dim = 1 [packed = true]; //blob的形状
}

message BlobProto {
  optional BlobShape shape = 7;
  repeated float data = 5 [packed = true]; // 前向传播的数据
  repeated float diff = 6 [packed = true]; // 反向传播的数据
  repeated double double_data = 8 [packed = true]; // double类型的前向传播的数据
  repeated double double_diff = 9 [packed = true]; // 依次类推

  // 4D dimensions -- deprecated.  Use "shape" instead.
  // 下面是为了兼容
  optional int32 num = 1 [default = 0];
  optional int32 channels = 2 [default = 0];
  optional int32 height = 3 [default = 0];
  optional int32 width = 4 [default = 0];
}


template<typename Dtype>
DataTransformer<Dtype>::DataTransformer(const TransformationParameter& param,
    Phase phase)
    : param_(param), phase_(phase) {
  // check if we want to use mean_file
  if (param_.has_mean_file()) {
    CHECK_EQ(param_.mean_value_size(), 0) <<
      "Cannot specify mean_file and mean_value at the same time";
    const string& mean_file = param.mean_file();
    if (Caffe::root_solver()) {
      LOG(INFO) << "Loading mean file from: " << mean_file;
    }
    BlobProto blob_proto;
    ReadProtoFromBinaryFileOrDie(mean_file.c_str(), &blob_proto);
    data_mean_.FromProto(blob_proto);
  }
  // check if we want to use mean_value
  if (param_.mean_value_size() > 0) {
    CHECK(param_.has_mean_file() == false) <<
      "Cannot specify mean_file and mean_value at the same time";
    for (int c = 0; c < param_.mean_value_size(); ++c) {
      mean_values_.push_back(param_.mean_value(c));
    }
  }
}

具体的实现如下：

#ifdef USE_OPENCV
#include <opencv2/core/core.hpp>
#endif  // USE_OPENCV

#include <string>
#include <vector>

#include "caffe/data_transformer.hpp"
#include "caffe/util/io.hpp"
#include "caffe/util/math_functions.hpp"
#include "caffe/util/rng.hpp"

namespace caffe {
// 构造函数
template<typename Dtype>
DataTransformer<Dtype>::DataTransformer(const TransformationParameter& param,
    Phase phase)
    : param_(param), phase_(phase) {
  // check if we want to use mean_file
  // 判断是否有平均值文件
  if (param_.has_mean_file()) {
    CHECK_EQ(param_.mean_value_size(), 0) <<
      "Cannot specify mean_file and mean_value at the same time";
    // 平均值文件的路径
    const string& mean_file = param.mean_file();
    if (Caffe::root_solver()) {
      LOG(INFO) << "Loading mean file from: " << mean_file;
    }
    BlobProto blob_proto;
    ReadProtoFromBinaryFileOrDie(mean_file.c_str(), &blob_proto);
    data_mean_.FromProto(blob_proto);
  }
  // check if we want to use mean_value
  if (param_.mean_value_size() > 0) {
    CHECK(param_.has_mean_file() == false) <<
      "Cannot specify mean_file and mean_value at the same time";
    for (int c = 0; c < param_.mean_value_size(); ++c) {
      mean_values_.push_back(param_.mean_value(c));
    }
  }
}

template<typename Dtype>
void DataTransformer<Dtype>::Transform(const Datum& datum,
                                       Dtype* transformed_data) {
  // 参考TransformationParameter的定义
  const string& data = datum.data();
  const int datum_channels = datum.channels();//数据的channel
  const int datum_height = datum.height();//数据的行数
  const int datum_width = datum.width();// 数据的列数

  const int crop_size = param_.crop_size();// crop大小
  const Dtype scale = param_.scale();// 缩放比例
  const bool do_mirror = param_.mirror() && Rand(2);// 该参数用于在镜像位置对数据处理
  const bool has_mean_file = param_.has_mean_file();// 是否有均值文件
  const bool has_uint8 = data.size() > 0;// 数据是否为uint8还是float类型的
  const bool has_mean_values = mean_values_.size() > 0;// 是否有每个channel的均值

  // 检查合法性
  CHECK_GT(datum_channels, 0);
  CHECK_GE(datum_height, crop_size);
  CHECK_GE(datum_width, crop_size);

  Dtype* mean = NULL;
  if (has_mean_file) {// 检查mean_file是否与数据的参数一致
    CHECK_EQ(datum_channels, data_mean_.channels());
    CHECK_EQ(datum_height, data_mean_.height());
    CHECK_EQ(datum_width, data_mean_.width());
    mean = data_mean_.mutable_cpu_data();
  }
  if (has_mean_values) {
    CHECK(mean_values_.size() == 1 || mean_values_.size() == datum_channels) <<
     "Specify either 1 mean_value or as many as channels: " << datum_channels;
    if (datum_channels > 1 && mean_values_.size() == 1) {
      // Replicate the mean_value for simplicity
      for (int c = 1; c < datum_channels; ++c) {
        mean_values_.push_back(mean_values_[0]);
      }
    }
  }

  int height = datum_height;
  int width = datum_width;

  // 根据是否需要crop来生成h_off和w_off
  int h_off = 0;
  int w_off = 0;
  if (crop_size) {// 如果crop_size不为0
    height = crop_size;
    width = crop_size;
    // We only do random crop when we do training.
    // 在训练的时候随机crop图像块,这里需要自己实现Rand这个函数来确定是如何随机的
    if (phase_ == TRAIN) {
      h_off = Rand(datum_height - crop_size + 1);// 产生从0到datum_height - crop_size的随机数
      w_off = Rand(datum_width - crop_size + 1);
    } else {// 测试的时候不用随机，取图像的中心
      h_off = (datum_height - crop_size) / 2;
      w_off = (datum_width - crop_size) / 2;
    }
  }

  // 对数据进行变换，主要是将原来的像素值减去均值，然后乘以scale这么一个操作
  // 如果需要crop则最终转换的Blob的大小即为crop*crop
  // 如果不是，则最终的Blob大小即为datum_height*datum_width
  Dtype datum_element;
  int top_index, data_index;
  for (int c = 0; c < datum_channels; ++c) {
    for (int h = 0; h < height; ++h) {
      for (int w = 0; w < width; ++w) {
        data_index = (c * datum_height + h_off + h) * datum_width + w_off + w;// 获取数据的索引
        if (do_mirror) {// 是否需要在镜像位置转换
          top_index = (c * height + h) * width + (width - 1 - w);//在宽这个坐标上做文章，来实现镜像
        } else {//
          top_index = (c * height + h) * width + w;
        }
        if (has_uint8) {// 数据如果是uint8则进行转换
          datum_element =
            static_cast<Dtype>(static_cast<uint8_t>(data[data_index]));
        } else {// 否则就是float
          datum_element = datum.float_data(data_index);
        }
        if (has_mean_file) {// 如果有mean_file,则原来的像素值减去均值，然后乘以scale
          transformed_data[top_index] =
            (datum_element - mean[data_index]) * scale;
        } else {
          if (has_mean_values) {// 否则减去该channel的均值（每个channel有其一个均值），然后乘以scale
            transformed_data[top_index] =
              (datum_element - mean_values_[c]) * scale;
          } else {// 否则如果没有均值那么就直接乘以scale即可
            transformed_data[top_index] = datum_element * scale;
          }
        }
      }
    }
  }
}


template<typename Dtype>
void DataTransformer<Dtype>::Transform(const Datum& datum,
                                       Blob<Dtype>* transformed_blob) {
  // If datum is encoded, decoded and transform the cv::image.
  if (datum.encoded()) {//  检查是否编码了，如果是则解码
#ifdef USE_OPENCV
    // 先检查是不是两个属性都设置， 如果是则说明参数设置有误
    CHECK(!(param_.force_color() && param_.force_gray()))
        << "cannot set both force_color and force_gray";
    cv::Mat cv_img;
    if (param_.force_color() || param_.force_gray()) {
        // 如果强制彩色或者强制灰度图像一个成立则使用DecodeDatumToCVMat解码
    // If force_color then decode in color otherwise decode in gray.
      cv_img = DecodeDatumToCVMat(datum, param_.force_color());
    } else {// 否则使用DecodeDatumToCVMatNative解码
      cv_img = DecodeDatumToCVMatNative(datum);
    }
    // Transform the cv::image into blob.
    // 变换
    return Transform(cv_img, transformed_blob);
#else
    LOG(FATAL) << "Encoded datum requires OpenCV; compile with USE_OPENCV.";
#endif  // USE_OPENCV
  } else {// 如果没有编码则，检查force_color和force_gray是否设置，如果设置则不合法，因为该选项只适合于编码后的数据
    if (param_.force_color() || param_.force_gray()) {
      LOG(ERROR) << "force_color and force_gray only for encoded datum";
    }
  }

  const int crop_size = param_.crop_size();
  const int datum_channels = datum.channels();
  const int datum_height = datum.height();
  const int datum_width = datum.width();

  // Check dimensions.
  const int channels = transformed_blob->channels();
  const int height = transformed_blob->height();
  const int width = transformed_blob->width();
  const int num = transformed_blob->num();

  CHECK_EQ(channels, datum_channels);
  CHECK_LE(height, datum_height);
  CHECK_LE(width, datum_width);
  CHECK_GE(num, 1);

  if (crop_size) {
    CHECK_EQ(crop_size, height);
    CHECK_EQ(crop_size, width);
  } else {
    CHECK_EQ(datum_height, height);
    CHECK_EQ(datum_width, width);
  }
  // 继续变换数据
  Dtype* transformed_data = transformed_blob->mutable_cpu_data();
  Transform(datum, transformed_data);
}

template<typename Dtype>
void DataTransformer<Dtype>::Transform(const vector<Datum> & datum_vector,
                                       Blob<Dtype>* transformed_blob) {
  const int datum_num = datum_vector.size();
  // 变换到的目标blob的形状
  const int num = transformed_blob->num();
  const int channels = transformed_blob->channels();
  const int height = transformed_blob->height();
  const int width = transformed_blob->width();

  CHECK_GT(datum_num, 0) << "There is no datum to add";
  CHECK_LE(datum_num, num) <<
    "The size of datum_vector must be no greater than transformed_blob->num()";
  // 新建一个uni_blob,里面只有一个batch
  Blob<Dtype> uni_blob(1, channels, height, width);
  for (int item_id = 0; item_id < datum_num; ++item_id) {
    int offset = transformed_blob->offset(item_id);
    uni_blob.set_cpu_data(transformed_blob->mutable_cpu_data() + offset);
    Transform(datum_vector[item_id], &uni_blob);
  }
}

#ifdef USE_OPENCV
template<typename Dtype>
void DataTransformer<Dtype>::Transform(const vector<cv::Mat> & mat_vector,
                                       Blob<Dtype>* transformed_blob) {
  // 获取mat的参数
  const int mat_num = mat_vector.size();
  const int num = transformed_blob->num();
  const int channels = transformed_blob->channels();
  const int height = transformed_blob->height();
  const int width = transformed_blob->width();

  CHECK_GT(mat_num, 0) << "There is no MAT to add";
  CHECK_EQ(mat_num, num) <<
    "The size of mat_vector must be equals to transformed_blob->num()";
  //  同上
  Blob<Dtype> uni_blob(1, channels, height, width);
  for (int item_id = 0; item_id < mat_num; ++item_id) {
    int offset = transformed_blob->offset(item_id);
    uni_blob.set_cpu_data(transformed_blob->mutable_cpu_data() + offset);
    Transform(mat_vector[item_id], &uni_blob);
  }
}

// 如果是图像的话，需要减去均值乘以scale，判断是不是需要做镜像处理
// 逻辑与前面类似
template<typename Dtype>
void DataTransformer<Dtype>::Transform(const cv::Mat& cv_img,
                                       Blob<Dtype>* transformed_blob) {
  const int crop_size = param_.crop_size();
  const int img_channels = cv_img.channels();
  const int img_height = cv_img.rows;
  const int img_width = cv_img.cols;

  // Check dimensions.
  const int channels = transformed_blob->channels();
  const int height = transformed_blob->height();
  const int width = transformed_blob->width();
  const int num = transformed_blob->num();

  CHECK_EQ(channels, img_channels);
  CHECK_LE(height, img_height);
  CHECK_LE(width, img_width);
  CHECK_GE(num, 1);

  CHECK(cv_img.depth() == CV_8U) << "Image data type must be unsigned byte";

  const Dtype scale = param_.scale();
  const bool do_mirror = param_.mirror() && Rand(2);
  const bool has_mean_file = param_.has_mean_file();
  const bool has_mean_values = mean_values_.size() > 0;

  CHECK_GT(img_channels, 0);
  CHECK_GE(img_height, crop_size);
  CHECK_GE(img_width, crop_size);

  Dtype* mean = NULL;
  if (has_mean_file) {
    CHECK_EQ(img_channels, data_mean_.channels());
    CHECK_EQ(img_height, data_mean_.height());
    CHECK_EQ(img_width, data_mean_.width());
    mean = data_mean_.mutable_cpu_data();
  }
  if (has_mean_values) {
    CHECK(mean_values_.size() == 1 || mean_values_.size() == img_channels) <<
     "Specify either 1 mean_value or as many as channels: " << img_channels;
    if (img_channels > 1 && mean_values_.size() == 1) {
      // Replicate the mean_value for simplicity
      for (int c = 1; c < img_channels; ++c) {
        mean_values_.push_back(mean_values_[0]);
      }
    }
  }

  int h_off = 0;
  int w_off = 0;
  cv::Mat cv_cropped_img = cv_img;
  if (crop_size) {
    CHECK_EQ(crop_size, height);
    CHECK_EQ(crop_size, width);
    // We only do random crop when we do training.
    if (phase_ == TRAIN) {
      h_off = Rand(img_height - crop_size + 1);
      w_off = Rand(img_width - crop_size + 1);
    } else {
      h_off = (img_height - crop_size) / 2;
      w_off = (img_width - crop_size) / 2;
    }
    cv::Rect roi(w_off, h_off, crop_size, crop_size);
    cv_cropped_img = cv_img(roi);
  } else {
    CHECK_EQ(img_height, height);
    CHECK_EQ(img_width, width);
  }

  CHECK(cv_cropped_img.data);

  Dtype* transformed_data = transformed_blob->mutable_cpu_data();
  int top_index;
  for (int h = 0; h < height; ++h) {
    const uchar* ptr = cv_cropped_img.ptr<uchar>(h);
    int img_index = 0;
    for (int w = 0; w < width; ++w) {
      for (int c = 0; c < img_channels; ++c) {
        if (do_mirror) {
          top_index = (c * height + h) * width + (width - 1 - w);
        } else {
          top_index = (c * height + h) * width + w;
        }
        // int top_index = (c * height + h) * width + w;
        Dtype pixel = static_cast<Dtype>(ptr[img_index++]);
        if (has_mean_file) {
          int mean_index = (c * img_height + h_off + h) * img_width + w_off + w;
          transformed_data[top_index] =
            (pixel - mean[mean_index]) * scale;
        } else {
          if (has_mean_values) {
            transformed_data[top_index] =
              (pixel - mean_values_[c]) * scale;
          } else {
            transformed_data[top_index] = pixel * scale;
          }
        }
      }
    }
  }
}
#endif  // USE_OPENCV

template<typename Dtype>
void DataTransformer<Dtype>::Transform(Blob<Dtype>* input_blob,
                                       Blob<Dtype>* transformed_blob) {
  const int crop_size = param_.crop_size();
  const int input_num = input_blob->num();
  const int input_channels = input_blob->channels();
  const int input_height = input_blob->height();
  const int input_width = input_blob->width();

  if (transformed_blob->count() == 0) {
    // Initialize transformed_blob with the right shape.
    if (crop_size) {
      transformed_blob->Reshape(input_num, input_channels,
                                crop_size, crop_size);
    } else {
      transformed_blob->Reshape(input_num, input_channels,
                                input_height, input_width);
    }
  }

  const int num = transformed_blob->num();
  const int channels = transformed_blob->channels();
  const int height = transformed_blob->height();
  const int width = transformed_blob->width();
  const int size = transformed_blob->count();

  CHECK_LE(input_num, num);
  CHECK_EQ(input_channels, channels);
  CHECK_GE(input_height, height);
  CHECK_GE(input_width, width);


  const Dtype scale = param_.scale();
  const bool do_mirror = param_.mirror() && Rand(2);
  const bool has_mean_file = param_.has_mean_file();
  const bool has_mean_values = mean_values_.size() > 0;

  int h_off = 0;
  int w_off = 0;
  if (crop_size) {
    CHECK_EQ(crop_size, height);
    CHECK_EQ(crop_size, width);
    // We only do random crop when we do training.
    if (phase_ == TRAIN) {
      h_off = Rand(input_height - crop_size + 1);
      w_off = Rand(input_width - crop_size + 1);
    } else {
      h_off = (input_height - crop_size) / 2;
      w_off = (input_width - crop_size) / 2;
    }
  } else {
    CHECK_EQ(input_height, height);
    CHECK_EQ(input_width, width);
  }

  // 如果有均值文件则
  Dtype* input_data = input_blob->mutable_cpu_data();
  if (has_mean_file) {
    CHECK_EQ(input_channels, data_mean_.channels());
    CHECK_EQ(input_height, data_mean_.height());
    CHECK_EQ(input_width, data_mean_.width());
    for (int n = 0; n < input_num; ++n) {
      int offset = input_blob->offset(n);
      /*
         template <typename Dtype>
       void caffe_sub(const int N, const Dtype* a, const Dtype* b, Dtype* y);
       math_function中定义的caffe_sub目的是矩阵相减input_data（以offset开始的矩阵） = input_data（以offset开始的矩阵） - data_mean_
    */
      caffe_sub(data_mean_.count(), input_data + offset,
            data_mean_.cpu_data(), input_data + offset);
    }
  }
  // 如果每个channel有均值则
  if (has_mean_values) {
    CHECK(mean_values_.size() == 1 || mean_values_.size() == input_channels) <<
     "Specify either 1 mean_value or as many as channels: " << input_channels;
    if (mean_values_.size() == 1) {
      caffe_add_scalar(input_blob->count(), -(mean_values_[0]), input_data);
    } else {
      for (int n = 0; n < input_num; ++n) {
        for (int c = 0; c < input_channels; ++c) {
          int offset = input_blob->offset(n, c);
          // 给nput_data[offset]地址开始的每一个元素加上一个-mean_values_[c]
          caffe_add_scalar(input_height * input_width, -(mean_values_[c]),
            input_data + offset);
        }
      }
    }
  }

  // 如果啥均值都没有则直接复制
  Dtype* transformed_data = transformed_blob->mutable_cpu_data();

  for (int n = 0; n < input_num; ++n) {
    int top_index_n = n * channels;
    int data_index_n = n * channels;
    for (int c = 0; c < channels; ++c) {
      int top_index_c = (top_index_n + c) * height;
      int data_index_c = (data_index_n + c) * input_height + h_off;
      for (int h = 0; h < height; ++h) {
        int top_index_h = (top_index_c + h) * width;
        int data_index_h = (data_index_c + h) * input_width + w_off;
        if (do_mirror) {
          int top_index_w = top_index_h + width - 1;
          for (int w = 0; w < width; ++w) {
            transformed_data[top_index_w-w] = input_data[data_index_h + w];
          }
        } else {
          for (int w = 0; w < width; ++w) {
            transformed_data[top_index_h + w] = input_data[data_index_h + w];
          }
        }
      }
    }
  }
  if (scale != Dtype(1)) {
    DLOG(INFO) << "Scale: " << scale;
    caffe_scal(size, scale, transformed_data);
  }
}

template<typename Dtype>
vector<int> DataTransformer<Dtype>::InferBlobShape(const Datum& datum) {
  if (datum.encoded()) {
#ifdef USE_OPENCV // 如果使用OpenCV则可以用先转换为CVMat，然后在推断blob的形状
    CHECK(!(param_.force_color() && param_.force_gray()))
        << "cannot set both force_color and force_gray";
    cv::Mat cv_img;
    if (param_.force_color() || param_.force_gray()) {
    // If force_color then decode in color otherwise decode in gray.
      cv_img = DecodeDatumToCVMat(datum, param_.force_color());
    } else {
      cv_img = DecodeDatumToCVMatNative(datum);
    }
    // InferBlobShape using the cv::image.
    return InferBlobShape(cv_img);
#else
    LOG(FATAL) << "Encoded datum requires OpenCV; compile with USE_OPENCV.";
#endif  // USE_OPENCV
  }

  // 否则直接粗暴地从datum里面获取形状的数据
  const int crop_size = param_.crop_size();
  const int datum_channels = datum.channels();
  const int datum_height = datum.height();
  const int datum_width = datum.width();
  // Check dimensions.
  CHECK_GT(datum_channels, 0);
  CHECK_GE(datum_height, crop_size);
  CHECK_GE(datum_width, crop_size);
  // Build BlobShape.
  vector<int> shape(4);
  shape[0] = 1;
  shape[1] = datum_channels;
  shape[2] = (crop_size)? crop_size: datum_height;
  shape[3] = (crop_size)? crop_size: datum_width;
  return shape;
}

template<typename Dtype>
vector<int> DataTransformer<Dtype>::InferBlobShape(
    const vector<Datum> & datum_vector) {
  const int num = datum_vector.size();
  CHECK_GT(num, 0) << "There is no datum to in the vector";
  // Use first datum in the vector to InferBlobShape.
  // 使用第一个来进行推断
  vector<int> shape = InferBlobShape(datum_vector[0]);
  // Adjust num to the size of the vector.
  shape[0] = num;
  return shape;
}

#ifdef USE_OPENCV
// 如果使用OpenCV
// 使用CVMat中的信息来推断形状
template<typename Dtype>
vector<int> DataTransformer<Dtype>::InferBlobShape(const cv::Mat& cv_img) {
  const int crop_size = param_.crop_size();
  const int img_channels = cv_img.channels();
  const int img_height = cv_img.rows;
  const int img_width = cv_img.cols;
  // Check dimensions.
  CHECK_GT(img_channels, 0);
  CHECK_GE(img_height, crop_size);
  CHECK_GE(img_width, crop_size);
  // Build BlobShape.
  vector<int> shape(4);
  shape[0] = 1;
  shape[1] = img_channels;
  shape[2] = (crop_size)? crop_size: img_height;
  shape[3] = (crop_size)? crop_size: img_width;
  return shape;
}

template<typename Dtype>
vector<int> DataTransformer<Dtype>::InferBlobShape(
    const vector<cv::Mat> & mat_vector) {
  const int num = mat_vector.size();
  CHECK_GT(num, 0) << "There is no cv_img to in the vector";
  // Use first cv_img in the vector to InferBlobShape.
  // 使用第一个来推断
  vector<int> shape = InferBlobShape(mat_vector[0]);
  // Adjust num to the size of the vector.
  shape[0] = num;
  return shape;
}
#endif  // USE_OPENCV

// 初始化随机数种子
template <typename Dtype>
void DataTransformer<Dtype>::InitRand() {
  // 要么需要镜像要么训练阶段和需要crop同时满足的情况下才初始化随机数种子
  const bool needs_rand = param_.mirror() ||
      (phase_ == TRAIN && param_.crop_size());
  if (needs_rand) {
    const unsigned int rng_seed = caffe_rng_rand();// 获得随机数种子（通过熵池或者时间生成种子）
    rng_.reset(new Caffe::RNG(rng_seed));//初始化随机数种子并实例化随机数生成器
  } else {
    rng_.reset();//否则随机数生成器设置为空
  }
}

// 产生从0到n的随机数
template <typename Dtype>
int DataTransformer<Dtype>::Rand(int n) {
  CHECK(rng_);
  CHECK_GT(n, 0);
  caffe::rng_t* rng =
      static_cast<caffe::rng_t*>(rng_->generator());
  return ((*rng)() % n);
}

INSTANTIATE_CLASS(DataTransformer);
/*
初始化类的宏定义是这样的，前面有讲过，这里再给出来
#define INSTANTIATE_CLASS(classname) \
  char gInstantiationGuard##classname; \
  template class classname<float>; \
  template class classname<double>
*/

}  // namespace caffe

三、与DataTransformer类相关类的介绍

（1）io的介绍

首先给出io中定义的各个函数的含义：

#ifndef CAFFE_UTIL_IO_H_
#define CAFFE_UTIL_IO_H_

#include <unistd.h>
#include <string>

#include "google/protobuf/message.h"

#include "caffe/blob.hpp"
#include "caffe/common.hpp"
#include "caffe/proto/caffe.pb.h"

namespace caffe {

using ::google::protobuf::Message;
// 内联函数，创建临时文件
inline void MakeTempFilename(string* temp_filename) {
  temp_filename->clear();
  *temp_filename = "/tmp/caffe_test.XXXXXX";
  char* temp_filename_cstr = new char[temp_filename->size() + 1];
  // NOLINT_NEXT_LINE(runtime/printf)
  strcpy(temp_filename_cstr, temp_filename->c_str());
  int fd = mkstemp(temp_filename_cstr);
  CHECK_GE(fd, 0) << "Failed to open a temporary file at: " << *temp_filename;
  close(fd);
  *temp_filename = temp_filename_cstr;
  delete[] temp_filename_cstr;
}

// 内联函数，创建临时目录
inline void MakeTempDir(string* temp_dirname) {
  temp_dirname->clear();
  *temp_dirname = "/tmp/caffe_test.XXXXXX";
  char* temp_dirname_cstr = new char[temp_dirname->size() + 1];
  // NOLINT_NEXT_LINE(runtime/printf)
  strcpy(temp_dirname_cstr, temp_dirname->c_str());
  char* mkdtemp_result = mkdtemp(temp_dirname_cstr);
  CHECK(mkdtemp_result != NULL)
      << "Failed to create a temporary directory at: " << *temp_dirname;
  *temp_dirname = temp_dirname_cstr;
  delete[] temp_dirname_cstr;
}
// 从txt读取proto的定义
bool ReadProtoFromTextFile(const char* filename, Message* proto);

// 从text读取proto的定义
inline bool ReadProtoFromTextFile(const string& filename, Message* proto) {
  return ReadProtoFromTextFile(filename.c_str(), proto);
}
// 从text读取proto的定义,只是增加了检查而已
inline void ReadProtoFromTextFileOrDie(const char* filename, Message* proto) {
  CHECK(ReadProtoFromTextFile(filename, proto));
}
// 从text读取proto的定义,只是增加了检查而已
inline void ReadProtoFromTextFileOrDie(const string& filename, Message* proto) {
  ReadProtoFromTextFileOrDie(filename.c_str(), proto);
}
// 将proto写入到txt文件
void WriteProtoToTextFile(const Message& proto, const char* filename);
inline void WriteProtoToTextFile(const Message& proto, const string& filename) {
  WriteProtoToTextFile(proto, filename.c_str());
}
// 从bin读取proto的定义
bool ReadProtoFromBinaryFile(const char* filename, Message* proto);
// 从bin读取proto的定义
inline bool ReadProtoFromBinaryFile(const string& filename, Message* proto) {
  return ReadProtoFromBinaryFile(filename.c_str(), proto);
}
// 从bin读取proto的定义,只是增加了检查而已
inline void ReadProtoFromBinaryFileOrDie(const char* filename, Message* proto) {
  CHECK(ReadProtoFromBinaryFile(filename, proto));
}
// 从bin读取proto的定义,只是增加了检查而已
inline void ReadProtoFromBinaryFileOrDie(const string& filename,
                                         Message* proto) {
  ReadProtoFromBinaryFileOrDie(filename.c_str(), proto);
}

// 将proto写入到bin文件
void WriteProtoToBinaryFile(const Message& proto, const char* filename);
// 内联函数，将proto写入到bin文件
inline void WriteProtoToBinaryFile(
    const Message& proto, const string& filename) {
  WriteProtoToBinaryFile(proto, filename.c_str());
}
// 从文件读取数据到Datum
bool ReadFileToDatum(const string& filename, const int label, Datum* datum);
// 内联函数，从文件读取数据到Datum
inline bool ReadFileToDatum(const string& filename, Datum* datum) {
  return ReadFileToDatum(filename, -1, datum);
}

// 从图像文件读取数据到Datum
bool ReadImageToDatum(const string& filename, const int label,
    const int height, const int width, const bool is_color,
    const std::string & encoding, Datum* datum);
// 内联函数，从图像文件（彩色还是黑白？）读取数据到Datum，指定图像大小
inline bool ReadImageToDatum(const string& filename, const int label,
    const int height, const int width, const bool is_color, Datum* datum) {
  return ReadImageToDatum(filename, label, height, width, is_color,
                          "", datum);
}
// 内联函数，从彩色图像文件读取数据到Datum，指定图像大小
inline bool ReadImageToDatum(const string& filename, const int label,
    const int height, const int width, Datum* datum) {
  return ReadImageToDatum(filename, label, height, width, true, datum);
}
// 内联函数，从图像文件（彩色还是黑白？）读取数据到Datum，自动获取图像大小
inline bool ReadImageToDatum(const string& filename, const int label,
    const bool is_color, Datum* datum) {
  return ReadImageToDatum(filename, label, 0, 0, is_color, datum);
}
// 内联函数，从彩色图像文件读取数据到Datum，自动获取图像大小
inline bool ReadImageToDatum(const string& filename, const int label,
    Datum* datum) {
  return ReadImageToDatum(filename, label, 0, 0, true, datum);
}
// 内联函数，从彩色图像文件读取数据到Datum，自动获取图像大小，指定编码格式
inline bool ReadImageToDatum(const string& filename, const int label,
    const std::string & encoding, Datum* datum) {
  return ReadImageToDatum(filename, label, 0, 0, true, encoding, datum);
}
// 对Datum进行解码
bool DecodeDatumNative(Datum* datum);
// 对彩色图像的Datum进行解码
bool DecodeDatum(Datum* datum, bool is_color);

#ifdef USE_OPENCV
// 将图像读取到CVMat，指定图像大小，是否彩色
cv::Mat ReadImageToCVMat(const string& filename,
    const int height, const int width, const bool is_color);
// 将图像读取到CVMat，指定图像大小
cv::Mat ReadImageToCVMat(const string& filename,
    const int height, const int width);
// 将图像读取到CVMat，指定是否彩色
cv::Mat ReadImageToCVMat(const string& filename,
    const bool is_color);
// 将图像读取到CVMat
cv::Mat ReadImageToCVMat(const string& filename);
// 将Datum解码为为CVMat
cv::Mat DecodeDatumToCVMatNative(const Datum& datum);
// 将彩色图像的Datum解码为为CVMat
cv::Mat DecodeDatumToCVMat(const Datum& datum, bool is_color);
// 将CVMat转换为Datum
void CVMatToDatum(const cv::Mat& cv_img, Datum* datum);
#endif  // USE_OPENCV

}  // namespace caffe

#endif   // CAFFE_UTIL_IO_H_

接下来给出io中的具体的实现的注释

#include <fcntl.h>
#include <google/protobuf/io/coded_stream.h>
#include <google/protobuf/io/zero_copy_stream_impl.h>
#include <google/protobuf/text_format.h>
#include <opencv2/core/core.hpp>
#ifdef USE_OPENCV
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/highgui/highgui_c.h>
#include <opencv2/imgproc/imgproc.hpp>
#endif  // USE_OPENCV
#include <stdint.h>

#include <algorithm>
#include <fstream>  // NOLINT(readability/streams)
#include <string>
#include <vector>

#include "caffe/common.hpp"
#include "caffe/proto/caffe.pb.h"
#include "caffe/util/io.hpp"

const int kProtoReadBytesLimit = INT_MAX;  // Max size of 2 GB minus 1 byte.

namespace caffe {

using google::protobuf::io::FileInputStream;
using google::protobuf::io::FileOutputStream;
using google::protobuf::io::ZeroCopyInputStream;
using google::protobuf::io::CodedInputStream;
using google::protobuf::io::ZeroCopyOutputStream;
using google::protobuf::io::CodedOutputStream;
using google::protobuf::Message;

// 从文件读取Proto的txt文件
bool ReadProtoFromTextFile(const char* filename, Message* proto) {
  int fd = open(filename, O_RDONLY);
  CHECK_NE(fd, -1) << "File not found: " << filename;
  FileInputStream* input = new FileInputStream(fd);
  // 注意如何使用protobuf去读取
  bool success = google::protobuf::TextFormat::Parse(input, proto);
  delete input;
  close(fd);
  return success;
}

// 将proto写入到txt文件
void WriteProtoToTextFile(const Message& proto, const char* filename) {
  int fd = open(filename, O_WRONLY | O_CREAT | O_TRUNC, 0644);
  FileOutputStream* output = new FileOutputStream(fd);
  // 注意如何写入
  CHECK(google::protobuf::TextFormat::Print(proto, output));
  delete output;
  close(fd);
}

// 从bin读取proto的定义
bool ReadProtoFromBinaryFile(const char* filename, Message* proto) {
  int fd = open(filename, O_RDONLY);
  CHECK_NE(fd, -1) << "File not found: " << filename;
  ZeroCopyInputStream* raw_input = new FileInputStream(fd);
  //  解码流com.google.protobuf.CodedInputStream
  CodedInputStream* coded_input = new CodedInputStream(raw_input);
  coded_input->SetTotalBytesLimit(kProtoReadBytesLimit, 536870912);

  bool success = proto->ParseFromCodedStream(coded_input);

  delete coded_input;
  delete raw_input;
  close(fd);
  return success;
}

// 将proto写入到bin文件
void WriteProtoToBinaryFile(const Message& proto, const char* filename) {
  fstream output(filename, ios::out | ios::trunc | ios::binary);
  CHECK(proto.SerializeToOstream(&output));
}

#ifdef USE_OPENCV
// 将图像读取到CVMat，指定图像大小，是否彩色
cv::Mat ReadImageToCVMat(const string& filename,
    const int height, const int width, const bool is_color) {
  cv::Mat cv_img;
  int cv_read_flag = (is_color ? CV_LOAD_IMAGE_COLOR :
    CV_LOAD_IMAGE_GRAYSCALE);
  cv::Mat cv_img_origin = cv::imread(filename, cv_read_flag);
  if (!cv_img_origin.data) {
    LOG(ERROR) << "Could not open or find file " << filename;
    return cv_img_origin;
  }
  if (height > 0 && width > 0) {
    cv::resize(cv_img_origin, cv_img, cv::Size(width, height));
  } else {
    cv_img = cv_img_origin;
  }
  return cv_img;
}

cv::Mat ReadImageToCVMat(const string& filename,
    const int height, const int width) {
  return ReadImageToCVMat(filename, height, width, true);
}

cv::Mat ReadImageToCVMat(const string& filename,
    const bool is_color) {
  return ReadImageToCVMat(filename, 0, 0, is_color);
}

cv::Mat ReadImageToCVMat(const string& filename) {
  return ReadImageToCVMat(filename, 0, 0, true);
}

// Do the file extension and encoding match?
// 看看是不是jpg还是jpeg的图像
static bool matchExt(const std::string & fn,
                     std::string en) {
  size_t p = fn.rfind('.');
  std::string ext = p != fn.npos ? fn.substr(p) : fn;
  std::transform(ext.begin(), ext.end(), ext.begin(), ::tolower);
  std::transform(en.begin(), en.end(), en.begin(), ::tolower);
  if ( ext == en )
    return true;
  if ( en == "jpg" && ext == "jpeg" )
    return true;
  return false;
}
// 从图像文件读取数据到Datum
bool ReadImageToDatum(const string& filename, const int label,
    const int height, const int width, const bool is_color,
    const std::string & encoding, Datum* datum) {
  cv::Mat cv_img = ReadImageToCVMat(filename, height, width, is_color);
  if (cv_img.data) {
    if (encoding.size()) {
      if ( (cv_img.channels() == 3) == is_color && !height && !width &&
          matchExt(filename, encoding) )
        return ReadFileToDatum(filename, label, datum);
      std::vector<uchar> buf;
      // 对数据解码
      cv::imencode("."+encoding, cv_img, buf);
      datum->set_data(std::string(reinterpret_cast<char*>(&buf[0]),
                      buf.size()));
      // 数据标签
      datum->set_label(label);
      // 是否被编码
      datum->set_encoded(true);
      return true;
    }
    CVMatToDatum(cv_img, datum);
    datum->set_label(label);
    return true;
  } else {
    return false;
  }
}
#endif  // USE_OPENCV
// 从文件读取数据到Datum
bool ReadFileToDatum(const string& filename, const int label,
    Datum* datum) {
  std::streampos size;

  fstream file(filename.c_str(), ios::in|ios::binary|ios::ate);
  if (file.is_open()) {
    size = file.tellg();
    std::string buffer(size, ' ');
    file.seekg(0, ios::beg);
    file.read(&buffer[0], size);
    file.close();
    datum->set_data(buffer);
    datum->set_label(label);
    datum->set_encoded(true);
    return true;
  } else {
    return false;
  }
}

#ifdef USE_OPENCV
// 直接编码数据的Datum到CVMat
cv::Mat DecodeDatumToCVMatNative(const Datum& datum) {
  cv::Mat cv_img;
  CHECK(datum.encoded()) << "Datum not encoded";
  const string& data = datum.data();
  std::vector<char> vec_data(data.c_str(), data.c_str() + data.size());
  cv_img = cv::imdecode(vec_data, -1);//flag=-1
  if (!cv_img.data) {
    LOG(ERROR) << "Could not decode datum ";
  }
  return cv_img;
}

// 直接编码彩色或者非彩色Datum到CVMat
cv::Mat DecodeDatumToCVMat(const Datum& datum, bool is_color) {
  cv::Mat cv_img;
  CHECK(datum.encoded()) << "Datum not encoded";
  const string& data = datum.data();
  std::vector<char> vec_data(data.c_str(), data.c_str() + data.size());
  int cv_read_flag = (is_color ? CV_LOAD_IMAGE_COLOR :
    CV_LOAD_IMAGE_GRAYSCALE);
  cv_img = cv::imdecode(vec_data, cv_read_flag);// flag为用户指定的
  if (!cv_img.data) {
    LOG(ERROR) << "Could not decode datum ";
  }
  return cv_img;
}

// If Datum is encoded will decoded using DecodeDatumToCVMat and CVMatToDatum
// If Datum is not encoded will do nothing
bool DecodeDatumNative(Datum* datum) {
  if (datum->encoded()) {
    cv::Mat cv_img = DecodeDatumToCVMatNative((*datum));
    CVMatToDatum(cv_img, datum);
    return true;
  } else {
    return false;
  }
}

// 将Datum进行解码
bool DecodeDatum(Datum* datum, bool is_color) {
  if (datum->encoded()) {
    cv::Mat cv_img = DecodeDatumToCVMat((*datum), is_color);
    CVMatToDatum(cv_img, datum);
    return true;
  } else {
    return false;
  }
}

// 将CVMat转换到Datum
void CVMatToDatum(const cv::Mat& cv_img, Datum* datum) {
  CHECK(cv_img.depth() == CV_8U) << "Image data type must be unsigned byte";
  datum->set_channels(cv_img.channels());
  datum->set_height(cv_img.rows);
  datum->set_width(cv_img.cols);
  datum->clear_data();
  datum->clear_float_data();
  datum->set_encoded(false);
  int datum_channels = datum->channels();
  int datum_height = datum->height();
  int datum_width = datum->width();
  int datum_size = datum_channels * datum_height * datum_width;
  std::string buffer(datum_size, ' ');
  for (int h = 0; h < datum_height; ++h) {
    const uchar* ptr = cv_img.ptr<uchar>(h);
    int img_index = 0;
    for (int w = 0; w < datum_width; ++w) {
      for (int c = 0; c < datum_channels; ++c) {
        int datum_index = (c * datum_height + h) * datum_width + w;
        buffer[datum_index] = static_cast<char>(ptr[img_index++]);
      }
    }
  }
  datum->set_data(buffer);
}
#endif  // USE_OPENCV
}  // namespace caffe

四、总结

总结起来就是，DataTransformer所作的工作实际上就是crop数据，让数据减去均值，以及缩放数据。

然后就是根据数据来推断形状。此外还介绍了io的内容，里面包含了创建临时文件临时目录操作，以及从txt文件以及bin文件读取proto数据或者写入proto的数据到txt或者bin文件。

参考：

[1]你可能需要了解关于cv::imencode和 cv::imdecode函数的flag的含义

http://docs.opencv.org/2.4/modules/highgui/doc/reading_and_writing_images_and_video.html