本文主要解析caffe框架中原始碼檔案/src/caffe/blob.cpp，該檔案主要實現caffe的資料儲存與傳遞。

caffe中Blob類主要用來表示網路中的資料，包括訓練資料，網路各層自身的引數(包括權值、偏置以及它們的梯度)，網路之間傳遞的資料都是通過 Blob 來實現的，同時 Blob 資料也支援在 CPU 與 GPU 上儲存，能夠在兩者之間做同步。

下面是我看原始碼時，蒐集的註釋，以及對原始碼的理解

Blob.hpp：：：：：：：：：：：：：：：：

#ifndef CAFFE_BLOB_HPP_
#define CAFFE_BLOB_HPP_

#include <algorithm>
 

#include <string>
#include <vector>

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

const int kMaxBlobAxes = 32;

namespace caffe {

/**
 * @brief A wrapper around SyncedMemory holders serving as the basic
 *        computational unit through which Layer%s, Net%s, and Solver%s
 *        interact.
 *
 * TODO(dox): more thorough description.
 */
 

template <typename Dtype>
class Blob {
 public:
  Blob() //建構函式
       : data_(), diff_(), count_(0), capacity_(0) {}

  /// @brief Deprecated; use <code>Blob(const vector<int>& shape)</code>.
  explicit Blob(const int num, const int channels, const int height,
      const int width);
  explicit
 
 Blob(const vector<int>& shape);

  /// @brief Deprecated; use <code>Reshape(const vector<int>& shape)</code>.
  void Reshape(const int num, const int channels, const int height,
      const int width);
  /**
   * @brief Change the dimensions of the blob, allocating new memory if
   *        necessary.
   *
   * This function can be called both to create an initial allocation
   * of memory, and to adjust the dimensions of a top blob during Layer::Reshape
   * or Layer::Forward. When changing the size of blob, memory will only be
   * reallocated if sufficient memory does not already exist, and excess memory
   * will never be freed.
   *
   * Note that reshaping an input blob and immediately calling Net::Backward is
   * an error; either Net::Forward or Net::Reshape need to be called to
   * propagate the new input shape to higher layers.
   */
  void Reshape(const vector<int>& shape);
  void Reshape(const BlobShape& shape);
  void ReshapeLike(const Blob& other);
  inline string shape_string() const {
    ostringstream stream;    //輸出資料的維度,以空格分隔，最後輸出一維維度(total) 
    for (int i = 0; i < shape_.size(); ++i) {
      stream << shape_[i] << " ";
    }
    stream << "(" << count_ << ")";
    return stream.str();
  }

  //返回blob的shape
  inline const vector<int>& shape() const { return shape_; }
  /**
   * @brief Returns the dimension of the index-th axis (or the negative index-th
   *        axis from the end, if index is negative).
   *
   * @param index the axis index, which may be negative as it will be
   *        "canonicalized" using CanonicalAxisIndex.
   *        Dies on out of range index.
   */
   //獲取第index維的大小，返回某一維的尺寸
  inline int shape(int index) const {
    return shape_[CanonicalAxisIndex(index)];
  }
  //返回資料維度就維的個數
  inline int num_axes() const { return shape_.size(); }

  //返回資料的所有維度的乘積,即資料的個數  
  inline int count() const { return count_; }

  /**
   * @brief Compute the volume of a slice; i.e., the product of dimensions
   *        among a range of axes.
   *
   * @param start_axis The first axis to include in the slice.
   *
   * @param end_axis The first axis to exclude from the slice.
   */
   // 獲取某幾維資料的大小 
  inline int count(int start_axis, int end_axis) const {
    CHECK_LE(start_axis, end_axis);
    CHECK_GE(start_axis, 0);
    CHECK_GE(end_axis, 0);
    CHECK_LE(start_axis, num_axes());
    CHECK_LE(end_axis, num_axes());
    int count = 1;
    for (int i = start_axis; i < end_axis; ++i) {
      count *= shape(i);
    }
    return count;
  }
  /**
   * @brief Compute the volume of a slice spanning from a particular first
   *        axis to the final axis.
   *
   * @param start_axis The first axis to include in the slice.
   */
   // 給定的維度到最後的維度之間包含的資料個數  
  inline int count(int start_axis) const {
    return count(start_axis, num_axes());
  }

  /**
   * @brief Returns the 'canonical' version of a (usually) user-specified axis,
   *        allowing for negative indexing (e.g., -1 for the last axis).
   *
   * @param axis_index the axis index.
   *        If 0 <= index < num_axes(), return index.
   *        If -num_axes <= index <= -1, return (num_axes() - (-index)),
   *        e.g., the last axis index (num_axes() - 1) if index == -1,
   *        the second to last if index == -2, etc.
   *        Dies on out of range index.
   */
   // 支援負數維度索引，負數表示從後往前，返回的是正確的維度索引（相當於將負數索引進行的轉換） 
   // Blob的Index是可以從負座標開始讀的,標準化索引，主要是對引數索引進行標準化，以滿足要求，轉換座標軸索引[-N，N]為[0，N]  
  inline int CanonicalAxisIndex(int axis_index) const {
    // 判斷是否在範圍內[-numaxes, numaxes]  
    CHECK_GE(axis_index, -num_axes())
        << "axis " << axis_index << " out of range for " << num_axes()
        << "-D Blob with shape " << shape_string();
    CHECK_LT(axis_index, num_axes())
        << "axis " << axis_index << " out of range for " << num_axes()
        << "-D Blob with shape " << shape_string();
    if (axis_index < 0) {
      return axis_index + num_axes();
    }
    return axis_index;
  }

  /// @brief Deprecated legacy shape accessor num: use shape(0) instead.
  inline int num() const { return LegacyShape(0); }
  /// @brief Deprecated legacy shape accessor channels: use shape(1) instead.
  inline int channels() const { return LegacyShape(1); }
  /// @brief Deprecated legacy shape accessor height: use shape(2) instead.
  inline int height() const { return LegacyShape(2); }
  /// @brief Deprecated legacy shape accessor width: use shape(3) instead.
  inline int width() const { return LegacyShape(3); }
  // 檢查blob的維度個數是不是小於4，Blob中的4個維num,channel,height,width可以直接通過shape(0),shape(1),shape(2),shape(3)來訪問
  // 返回的是每維資料的大小,等同於shape()函式的功能s
  inline int LegacyShape(int index) const {
    CHECK_LE(num_axes(), 4)
        << "Cannot use legacy accessors on Blobs with > 4 axes.";
    CHECK_LT(index, 4);   // 檢查維度索引是不是小於4  
    CHECK_GE(index, -4);  // 檢查維度索引是不是大於-4
    if (index >= num_axes() || index < -num_axes()) {
      // Axis is out of range, but still in [0, 3] (or [-4, -1] for reverse
      // indexing) -- this special case simulates the one-padding used to fill
      // extraneous axes of legacy blobs.
      return 1;
    }
    return shape(index);
  }
  // 計算一維線性偏移量  
  inline int offset(const int n, const int c = 0, const int h = 0,
      const int w = 0) const {
    CHECK_GE(n, 0);        /*判斷輸入引數是否超過閾值*/
    CHECK_LE(n, num());
    CHECK_GE(channels(), 0);
    CHECK_LE(c, channels());
    CHECK_GE(height(), 0);
    CHECK_LE(h, height());
    CHECK_GE(width(), 0);
    CHECK_LE(w, width());
    return ((n * channels() + c) * height() + h) * width() + w;
  }

  // 計算一維線性偏移量,只不過引數用的是vector<int> 
  inline int offset(const vector<int>& indices) const {
    CHECK_LE(indices.size(), num_axes());
    int offset = 0;
    for (int i = 0; i < num_axes(); ++i) {
      offset *= shape(i);
      if (indices.size() > i) {
        CHECK_GE(indices[i], 0);
        CHECK_LT(indices[i], shape(i));
        offset += indices[i];
      }
    }
    return offset;
  }
  /**
   * @brief Copy from a source Blob.
   *
   * @param source the Blob to copy from
   * @param copy_diff if false, copy the data; if true, copy the diff
   * @param reshape if false, require this Blob to be pre-shaped to the shape
   *        of other (and die otherwise); if true, Reshape this Blob to other's
   *        shape if necessary
   * 從給定的blob進行復制，如果copy_diff=true則新的blob複製的是diff,
   * 如果reshape=true則改變新blob的形狀 
   */
  void CopyFrom(const Blob<Dtype>& source, bool copy_diff = false,
      bool reshape = false);

  // 獲取在記憶體下的資料(前向傳播所用的資料)  
  inline Dtype data_at(const int n, const int c, const int h,
      const int w) const {
    return cpu_data()[offset(n, c, h, w)];
  }

  // 獲取在記憶體下的後向傳播的資料 
  inline Dtype diff_at(const int n, const int c, const int h,
      const int w) const {
    return cpu_diff()[offset(n, c, h, w)];
  }

  // 獲取cpu記憶體中offset指定位置的前向傳播資料
  inline Dtype data_at(const vector<int>& index) const {
    return cpu_data()[offset(index)];
  }

  // 獲取cpu記憶體中offset指定位置的返向傳播資料
  inline Dtype diff_at(const vector<int>& index) const {
    return cpu_diff()[offset(index)];
  }

  // 返回前向傳播資料地址(前向傳播資料一般為影象本身資料)
  inline const shared_ptr<SyncedMemory>& data() const {
    CHECK(data_);
    return data_;
  }

  // 返回後向傳播資料地址(後向傳播資料一般為影象資料導數)
  inline const shared_ptr<SyncedMemory>& diff() const {
    CHECK(diff_);
    return diff_;
  }

  //記憶體資料的地址返回，資料清空等操作,詳見.cpp
  const Dtype* cpu_data() const;
  void set_cpu_data(Dtype* data);
  const int* gpu_shape() const;
  const Dtype* gpu_data() const;
  const Dtype* cpu_diff() const;
  const Dtype* gpu_diff() const;

  // 一些記憶體同步與處理的函式見SycedMem.cpp中具體定義
  Dtype* mutable_cpu_data();
  Dtype* mutable_gpu_data();
  Dtype* mutable_cpu_diff();
  Dtype* mutable_gpu_diff();

  // 資料更新，blob裡面的data部分減去diff部分
  void Update();

  // 從protobuf序列化檔案讀取blob物件  
  void FromProto(const BlobProto& proto, bool reshape = true);
  // 將物件序列化為protobuf檔案 
  void ToProto(BlobProto* proto, bool write_diff = false) const;

  /// @brief Compute the sum of absolute values (L1 norm) of the data.
  // 計算data的L1範數
  Dtype asum_data() const;
  /// @brief Compute the sum of absolute values (L1 norm) of the diff.
  // 計算diff的L1範數
  Dtype asum_diff() const;
  /// @brief Compute the sum of squares (L2 norm squared) of the data.
  // 計算data的L2範數
  Dtype sumsq_data() const;
  /// @brief Compute the sum of squares (L2 norm squared) of the diff.
  // 計算diff的L2範數
  Dtype sumsq_diff() const;

  /// @brief Scale the blob data by a constant factor.
  // 歸一化data資料
  void scale_data(Dtype scale_factor);
  /// @brief Scale the blob diff by a constant factor.
  // 歸一化diff資料
  void scale_diff(Dtype scale_factor);

  /**
   * @brief Set the data_ shared_ptr to point to the SyncedMemory holding the
   *        data_ of Blob other -- useful in Layer%s which simply perform a copy
   *        in their Forward pass.
   *
   * This deallocates the SyncedMemory holding this Blob's data_, as
   * shared_ptr calls its destructor when reset with the "=" operator.
   */
   // 與other共享data資料,把other的data資料指標傳給本blob
  void ShareData(const Blob& other);
  /**
   * @brief Set the diff_ shared_ptr to point to the SyncedMemory holding the
   *        diff_ of Blob other -- useful in Layer%s which simply perform a copy
   *        in their Forward pass.
   *
   * This deallocates the SyncedMemory holding this Blob's diff_, as
   * shared_ptr calls its destructor when reset with the "=" operator.
   */
   // 與other共享diff資料,把other的diff資料指標傳給本blob
  void ShareDiff(const Blob& other);
  // 判斷本blob與other形狀是否相等  
  bool ShapeEquals(const BlobProto& other);

 protected:/*shared_ptr屬於boost庫的智慧指標*/
    // 前向傳播的資料  
  shared_ptr<SyncedMemory> data_;
    // diff是反向傳播的資料即偏差  
  shared_ptr<SyncedMemory> diff_;
    // 舊的儲存Blob的形狀
  shared_ptr<SyncedMemory> shape_data_;
    // 新的儲存Blob的形狀
  vector<int> shape_;
    //資料的個數,也就是個數*通道數*高度*寬度  (實際資料的大小)
  int count_;
    //元素個數  (記憶體最大能儲存資料的大小)
  int capacity_;

  DISABLE_COPY_AND_ASSIGN(Blob);
};  // class Blob

}  // namespace caffe

#endif  // CAFFE_BLOB_HPP_

Blob.cpp：：：：：：：：：：：：：：：：

#include <climits>
#include <vector>

#include "caffe/blob.hpp"
#include "caffe/common.hpp"
#include "caffe/syncedmem.hpp"
#include "caffe/util/math_functions.hpp"

namespace caffe {

template <typename Dtype>  /*老的reshape方法，呼叫下面的新reshape*/
void Blob<Dtype>::Reshape(const int num, const int channels, const int height,
    const int width) {
  vector<int> shape(4);
  shape[0] = num;
  shape[1] = channels;
  shape[2] = height;
  shape[3] = width;
  Reshape(shape);
}

template <typename Dtype> /*新的reshape及其具體實現*/
void Blob<Dtype>::Reshape(const vector<int>& shape) {
  CHECK_LE(shape.size(), kMaxBlobAxes); //是否小於規定的最大BLOB的維度(32維)  
  count_ = 1;
  shape_.resize(shape.size()); //首先將大小設定為vector<int> shape_; 即新的形狀資料的大小  
  if (!shape_data_ || shape_data_->size() < shape.size() * sizeof(int)) {
    shape_data_.reset(new SyncedMemory(shape.size() * sizeof(int)));
  }
  int* shape_data = static_cast<int*>(shape_data_->mutable_cpu_data());
  for (int i = 0; i < shape.size(); ++i) {
    // 檢查形狀資料是否合法  
    CHECK_GE(shape[i], 0);
    if (count_ != 0) {
      CHECK_LE(shape[i], INT_MAX / count_) << "blob size exceeds INT_MAX";
    }
    // 計算資料個數  
    count_ *= shape[i];
    // 複製shape到新的和舊的形狀資料  
    shape_[i] = shape[i];
    shape_data[i] = shape[i];
  }
  // 判斷是否大於儲存的容量  
  if (count_ > capacity_) {
    capacity_ = count_;
    // 重新分配記憶體  
    data_.reset(new SyncedMemory(capacity_ * sizeof(Dtype)));
    diff_.reset(new SyncedMemory(capacity_ * sizeof(Dtype)));
  }
}

// 所謂的reshape實際上就僅僅是複製了shape的資料而已  
template <typename Dtype>
void Blob<Dtype>::Reshape(const BlobShape& shape) {
  CHECK_LE(shape.dim_size(), kMaxBlobAxes);// 維度是否小於32  
  vector<int> shape_vec(shape.dim_size());
  // 複製形狀資料  
  for (int i = 0; i < shape.dim_size(); ++i) {
    shape_vec[i] = shape.dim(i);
  }
  // 呼叫新的reshape函式  
  Reshape(shape_vec);
}

/*依照其他blob來修改當前blob的形狀*/
template <typename Dtype>
void Blob<Dtype>::ReshapeLike(const Blob<Dtype>& other) {
  Reshape(other.shape());
}

/*blob建構函式*/
template <typename Dtype>
Blob<Dtype>::Blob(const int num, const int channels, const int height,
    const int width)
  // capacity_ must be initialized before calling Reshape
  : capacity_(0) {
  Reshape(num, channels, height, width);  //先初始化容量為0，然後用reshape來分配記憶體了
}

/*blob建構函式*/
template <typename Dtype>
Blob<Dtype>::Blob(const vector<int>& shape)
  // capacity_ must be initialized before calling Reshape
  : capacity_(0) {
  Reshape(shape);
}

/*返回gpu中blob物件中資料的記憶體地址*/
template <typename Dtype> 
const int* Blob<Dtype>::gpu_shape() const {
  CHECK(shape_data_);
  return (const int*)shape_data_->gpu_data();
}

/*返回cpu中blob物件中資料的記憶體地址*/
template <typename Dtype>
const Dtype* Blob<Dtype>::cpu_data() const {
  CHECK(data_);
  return (const Dtype*)data_->cpu_data();
}

/*呼叫SyncedMemory的set_cpu_data函式來設定cpu的資料的記憶體地址,並清空資料*/
template <typename Dtype>
void Blob<Dtype>::set_cpu_data(Dtype* data) {
  CHECK(data);
  data_->set_cpu_data(data);
}

/*返回gpu中blob物件中資料的記憶體地址*/
template <typename Dtype>
const Dtype* Blob<Dtype>::gpu_data() const {
  CHECK(data_);
  return (const Dtype*)data_->gpu_data();
}

/*返回cpu中blob物件中資料的導數的記憶體地址*/
template <typename Dtype>
const Dtype* Blob<Dtype>::cpu_diff() const {
  CHECK(diff_);
  return (const Dtype*)diff_->cpu_data();
}

/*返回gpu中blob物件中資料的導數的記憶體地址*/
template <typename Dtype>
const Dtype* Blob<Dtype>::gpu_diff() const {
  CHECK(diff_);
  return (const Dtype*)diff_->gpu_data();
}

//呼叫SyncedMemory.cpp中的mutable_cpu_data()
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_cpu_data() {
  CHECK(data_);
  return static_cast<Dtype*>(data_->mutable_cpu_data());
}

//呼叫SyncedMemory.cpp中的mutable_gpu_data()
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_gpu_data() {
  CHECK(data_);
  return static_cast<Dtype*>(data_->mutable_gpu_data());
}

//呼叫SyncedMemory.cpp中的mutable_cpu_data()
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_cpu_diff() {
  CHECK(diff_);
  return static_cast<Dtype*>(diff_->mutable_cpu_data());
}

//呼叫SyncedMemory.cpp中的mutable_gpu_data()
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_gpu_diff() {
  CHECK(diff_);
  return static_cast<Dtype*>(diff_->mutable_gpu_data());
}

// 當前blob資料的指標指向其他blob的資料,以實現共享data
template <typename Dtype>
void Blob<Dtype>::ShareData(const Blob& other) {
  CHECK_EQ(count_, other.count());
  data_ = other.data();
}

// 當前blob資料的指標指向其他blob的資料,以實現共享diff
template <typename Dtype>
void Blob<Dtype>::ShareDiff(const Blob& other) {
  CHECK_EQ(count_, other.count());
  diff_ = other.diff();
}

// The "update" method is used for parameter blobs in a Net, which are stored
// as Blob<float> or Blob<double> -- hence we do not define it for
// Blob<int> or Blob<unsigned int>.
template <> void Blob<unsigned int>::Update() { NOT_IMPLEMENTED; }
template <> void Blob<int>::Update() { NOT_IMPLEMENTED; }

// Update是計算data=-1 * diff + data  
// 更新data_的資料，合併data與diff
template <typename Dtype>
void Blob<Dtype>::Update() {
  // We will perform update based on where the data is located.
  switch (data_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    // perform computation on CPU
    // axpby即alpha * x plus beta *y 這個含義,blas的函式命名真是見名知意
    // caffe_axpy計算的是Y=alpha * X + Y ，其中alpha=-1了這裡
    // 儲存的時候用到了mutable_cpu_data，防止其他執行緒訪問  
    caffe_axpy<Dtype>(count_, Dtype(-1),
        static_cast<const Dtype*>(diff_->cpu_data()),
        static_cast<Dtype*>(data_->mutable_cpu_data()));
    break;
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
    // perform computation on GPU
    caffe_gpu_axpy<Dtype>(count_, Dtype(-1),
        static_cast<const Dtype*>(diff_->gpu_data()),
        static_cast<Dtype*>(data_->mutable_gpu_data()));
#else
    NO_GPU;
#endif
    break;
  default:
    LOG(FATAL) << "Syncedmem not initialized.";
  }
}

template <> unsigned int Blob<unsigned int>::asum_data() const {
  NOT_IMPLEMENTED;
  return 0;
}

template <> int Blob<int>::asum_data() const {
  NOT_IMPLEMENTED;
  return 0;
}

// 計算data的L1範數  
// 呼叫math_function.hpp中的函式caffe_cpu_asum()和caffe_gpu_asum
// 實現求cpu_data或者gpu_data中每個元素絕對值的和
template <typename Dtype>
Dtype Blob<Dtype>::asum_data() const {
  if (!data_) { return 0; }
  switch (data_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    return caffe_cpu_asum(count_, cpu_data());
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
  {
    Dtype asum;
    caffe_gpu_asum(count_, gpu_data(), &asum);
    return asum;
  }
#else
    NO_GPU;
#endif
  case SyncedMemory::UNINITIALIZED:
    return 0;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << data_->head();
  }
  return 0;
}

template <> unsigned int Blob<unsigned int>::asum_diff() const {
  NOT_IMPLEMENTED;
  return 0;
}

template <> int Blob<int>::asum_diff() const {
  NOT_IMPLEMENTED;
  return 0;
}

// 計算diff的L1範數 
// 呼叫math_function.hpp中的函式caffe_cpu_asum()和caffe_gpu_asum
// 實現求cpu_diff或者gpu_diff中每個元素絕對值的和
template <typename Dtype>
Dtype Blob<Dtype>::asum_diff() const {
  if (!diff_) { return 0; }
  switch (diff_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    return caffe_cpu_asum(count_, cpu_diff());
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
  {
    Dtype asum;
    caffe_gpu_asum(count_, gpu_diff(), &asum);
    return asum;
  }
#else
    NO_GPU;
#endif
  case SyncedMemory::UNINITIALIZED:
    return 0;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << diff_->head();
  }
  return 0;
}

template <> unsigned int Blob<unsigned int>::sumsq_data() const {
  NOT_IMPLEMENTED;
  return 0;
}

template <> int Blob<int>::sumsq_data() const {
  NOT_IMPLEMENTED;
  return 0;
}

// 計算sum of square of data(L2範數)  
// 呼叫math_function.hpp中的中的函式caffe_cpu_dot(),caffe_cpu_strided_dot(),caffe_gpu_dot(), caffe_gpu_strided_dot()
// 實現求cpu_data或者gpu_data中每個元素絕對值的平方的和
template <typename Dtype>
Dtype Blob<Dtype>::sumsq_data() const {
  Dtype sumsq;
  const Dtype* data;
  if (!data_) { return 0; }
  switch (data_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    data = cpu_data();
    sumsq = caffe_cpu_dot(count_, data, data);
    break;
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
    data = gpu_data();
    caffe_gpu_dot(count_, data, data, &sumsq);
#else
    NO_GPU;
#endif
    break;
  case SyncedMemory::UNINITIALIZED:
    return 0;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << data_->head();
  }
  return sumsq;
}

template <> unsigned int Blob<unsigned int>::sumsq_diff() const {
  NOT_IMPLEMENTED;
  return 0;
}

template <> int Blob<int>::sumsq_diff() const {
  NOT_IMPLEMENTED;
  return 0;
}

// 計算sum of square of diff(L2範數)  
// 呼叫math_function.hpp中的中的函式caffe_cpu_dot(),caffe_cpu_strided_dot(),caffe_gpu_dot(), caffe_gpu_strided_dot()
// 實現求cpu_diff或者gpu_diff中每個元素絕對值的平方的和
template <typename Dtype>
Dtype Blob<Dtype>::sumsq_diff() const {
  Dtype sumsq;
  const Dtype* diff;
  if (!diff_) { return 0; }
  switch (diff_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    diff = cpu_diff();
    sumsq = caffe_cpu_dot(count_, diff, diff);
    break;
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
    diff = gpu_diff();
    caffe_gpu_dot(count_, diff, diff, &sumsq);
    break;
#else
    NO_GPU;
#endif
  case SyncedMemory::UNINITIALIZED:
    return 0;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << data_->head();
  }
  return sumsq;
}

template <> void Blob<unsigned int>::scale_data(unsigned int scale_factor) {
  NOT_IMPLEMENTED;
}

template <> void Blob<int>::scale_data(int scale_factor) {
  NOT_IMPLEMENTED;
}

// 將data部分乘以一個因子scale_factor  
template <typename Dtype>
void Blob<Dtype>::scale_data(Dtype scale_factor) {
  Dtype* data;
  if (!data_) { return; }
  switch (data_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    data = mutable_cpu_data();
    caffe_scal(count_, scale_factor, data);
    return;
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
    data = mutable_gpu_data();
    caffe_gpu_scal(count_, scale_factor, data);
    return;
#else
    NO_GPU;
#endif
  case SyncedMemory::UNINITIALIZED:
    return;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << data_->head();
  }
}

template <> void Blob<unsigned int>::scale_diff(unsigned int scale_factor) {
  NOT_IMPLEMENTED;
}

template <> void Blob<int>::scale_diff(int scale_factor) {
  NOT_IMPLEMENTED;
}

// 將diff部分乘以一個因子sacle_factor  
template <typename Dtype>
void Blob<Dtype>::scale_diff(Dtype scale_factor) {
  Dtype* diff;
  if (!diff_) { return; }
  switch (diff_->head()) {
  case SyncedMemory::HEAD_AT_CPU:
    diff = mutable_cpu_diff();
    caffe_scal(count_, scale_factor, diff);
    return;
  case SyncedMemory::HEAD_AT_GPU:
  case SyncedMemory::SYNCED:
#ifndef CPU_ONLY
    diff = mutable_gpu_diff();
    caffe_gpu_scal(count_, scale_factor, diff);
    return;
#else
    NO_GPU;
#endif
  case SyncedMemory::UNINITIALIZED:
    return;
  default:
    LOG(FATAL) << "Unknown SyncedMemory head state: " << diff_->head();
  }
}

//兩個blob的形狀是否一樣
template <typename Dtype>
bool Blob<Dtype>::ShapeEquals(const BlobProto& other) {
  if (other.has_num() || other.has_channels() ||
      other.has_height() || other.has_width()) {
    // Using deprecated 4D Blob dimensions --
    // shape is (num, channels, height, width).
    // Note: we do not use the normal Blob::num(), Blob::channels(), etc.
    // methods as these index from the beginning of the blob shape, where legacy
    // parameter blobs were indexed from the end of the blob shape (e.g., bias
    // Blob shape (1 x 1 x 1 x N), IP layer weight Blob shape (1 x 1 x M x N)).
    return shape_.size() <= 4 &&
           LegacyShape(-4) == other.num() &&
           LegacyShape(-3) == other.channels() &&
           LegacyShape(-2) == other.height() &&
           LegacyShape(-1) == other.width();
  }
  // 如果不是舊的blob則直接判斷  
  vector<int> other_shape(other.shape().dim_size());
  for (int i = 0; i < other.shape().dim_size(); ++i) {
    other_shape[i] = other.shape().dim(i);
  }
  return shape_ == other_shape;
}

// 從別的blob進行復制  
template <typename Dtype>
void Blob<Dtype>::CopyFrom(const Blob& source, bool copy_diff, bool reshape) {
  if (source.count() != count_ || source.shape() != shape_) {
    if (reshape) {
      ReshapeLike(source);//複製shape資料 
    } else {
      LOG(FATAL) << "Trying to copy blobs of different sizes.";
    }
  }
  switch (Caffe::mode()) {
  case Caffe::GPU:
    // GPU複製diff  
    if (copy_diff) {
      // 這都用 template <> void caffe_copy<float>(const int N, const float* X, float* Y) { cblas_scopy(N, X, 1, Y, 1); }  
      caffe_copy(count_, source.gpu_diff(),
          static_cast<Dtype*>(diff_->mutable_gpu_data()));
    } else {
      caffe_copy(count_, source.gpu_data(),
          static_cast<Dtype*>(data_->mutable_gpu_data()));
    }
    break;
    // CPU複製diff  
  case Caffe::CPU:
    if (copy_diff) {
      caffe_copy(count_, source.cpu_diff(),
          static_cast<Dtype*>(diff_->mutable_cpu_data()));
    } else {
      caffe_copy(count_, source.cpu_data(),
          static_cast<Dtype*>(data_->mutable_cpu_data()));
    }
    break;
  default:
    LOG(FATAL) << "Unknown caffe mode.";
  }
}

// 從定義在caffe.proto 中的一個message來複制資料
template <typename Dtype>
void Blob<Dtype>::FromProto(const BlobProto& proto, bool reshape) {
  if (reshape) {
    vector<int> shape;
    if (proto.has_num() || proto.has_channels() ||
        proto.has_height() || proto.has_width()) {
      // Using deprecated 4D Blob dimensions --
      // shape is (num, channels, height, width).
      // 如果是舊的blob直接轉換為新的blob中的shape資料  
      shape.resize(4);
      shape[0] = proto.num();
      shape[1] = proto.channels();
      shape[2] = proto.height();
      shape[3] = proto.width();
    } else {
      shape.resize(proto.shape().dim_size());
      for (int i = 0; i < proto.shape().dim_size(); ++i) {
        shape[i] = proto.shape().dim(i);
      }
    }
    Reshape(shape);// 複製shape資料到當前blob  
  } else {
    CHECK(ShapeEquals(proto)) << "shape mismatch (reshape not set)";
  }
  // copy data
  Dtype* data_vec = mutable_cpu_data();// 獲取當前的blob在記憶體上的資料指標，該指標是互斥的 
  if (proto.double_data_size() > 0) {
    CHECK_EQ(count_, proto.double_data_size());
    for (int i = 0; i < count_; ++i) {
      data_vec[i] = proto.double_data(i);
    }
  } else {
    CHECK_EQ(count_, proto.data_size());
    for (int i = 0; i < count_; ++i) {
      data_vec[i] = proto.data(i);
    }
  }
  if (proto.double_diff_size() > 0) {
    CHECK_EQ(count_, proto.double_diff_size());
    Dtype* diff_vec = mutable_cpu_diff();// 獲取當前的diff在記憶體上的資料指標，該指標是互斥的
    for (int i = 0; i < count_; ++i) {
      diff_vec[i] = proto.double_diff(i);
    }
  } else if (proto.diff_size() > 0) {
    CHECK_EQ(count_, proto.diff_size());
    Dtype* diff_vec = mutable_cpu_diff();
    for (int i = 0; i < count_; ++i) {
      diff_vec[i] = proto.diff(i);
    }
  }
}

//將資料寫到proto
template <>
void Blob<double>::ToProto(BlobProto* proto, bool write_diff) const {
  proto->clear_shape();
  for (int i = 0; i < shape_.size(); ++i) {
    proto->mutable_shape()->add_dim(shape_[i]);
  }
  proto->clear_double_data();
  proto->clear_double_diff();
  const double* data_vec = cpu_data();
  for (int i = 0; i < count_; ++i) {
    proto->add_double_data(data_vec[i]);//將data寫入proto  
  }
  if (write_diff) {
    const double* diff_vec = cpu_diff();
    for (int i = 0; i < count_; ++i) {
      proto->add_double_diff(diff_vec[i]);//將diff寫入proto  
    }
  }
}

template <>
void Blob<float>::ToProto(BlobProto* proto, bool write_diff) const {
  proto->clear_shape();
  for (int i = 0; i < shape_.size(); ++i) {
    proto->mutable_shape()->add_dim(shape_[i]);
  }
  proto->clear_data();
  proto->clear_diff();
  const float* data_vec = cpu_data();
  for (int i = 0; i < count_; ++i) {
    proto->add_data(data_vec[i]);
  }
  if (write_diff) {
    const float* diff_vec = cpu_diff();
    for (int i = 0; i < count_; ++i) {
      proto->add_diff(diff_vec[i]);
    }
  }
}

INSTANTIATE_CLASS(Blob);
template class Blob<int>;
template class Blob<unsigned int>;

}  // namespace caffe