代码梳理
.
├── build -> .build_release # 存放编译结果
├── cmake
│ ├── External
│ ├── Modules
│ └── Templates
├── data # 存放原始数据及其获取脚本
│ ├── cifar10
│ ├── ilsvrc12
│ └── mnist
├── distribute # 编译后，用于迁移的发布包存放处
│ ├── bin
│ └── lib
├── docker # 便于迁移
│ ├── cpu
│ └── gpu
├── docs
│ ├── images
│ ├── _layouts
│ ├── stylesheets
│ └── tutorial
│ ├── fig
│ └── layers
├── examples # 简单例程
│ ├── cifar10
│ ├── cpp_classification
│ ├── feature_extraction
│ ├── finetune_flickr_style
│ ├── finetune_pascal_detection
│ ├── hdf5_classification
│ ├── imagenet
│ ├── images
│ ├── mnist
│ │ ├── mnist_test_lmdb
│ │ └── mnist_train_lmdb
│ ├── net_surgery
│ ├── pycaffe
│ │ └── layers
│ ├── siamese
│ └── web_demo
│ └── templates
├── include #Caffe头文件集中营（重要）
│ └── caffe
│ ├── layers
│ ├── test
│ └── util
├── matlab
│ ├── +caffe
│ │ ├── imagenet
│ │ ├── private
│ │ └── +test
│ ├── demo
│ └── hdf5creation
├── models
│ ├── bvlc_alexnet
│ ├── bvlc_googlenet
│ ├── bvlc_reference_caffenet
│ ├── bvlc_reference_rcnn_ilsvrc13
│ └── finetune_flickr_style
├── python
│ └── caffe
│ ├── imagenet
│ ├── proto
│ ├── __pycache__
│ └── test
├── scripts
│ └── travis
├── src # Caffe 源码 （重要）
│ ├── caffe
│ │ ├── layers
│ │ ├── proto # proto描述文件，从这学习数据结构
│ │ ├── solvers
│ │ ├── test
│ │ │ └── test_data
│ │ └── util
│ └── gtest
└── tools # 工具源码 （重要）
└── extra
然后就是阅读源代码，bula,bula,bula~,enjoy!
数据结构
几大基本概念
- blob：提供统一的存储器接口，持有一批图像或其他数据、权值、权重更新值，类似于Torch/Theano/TensorFlow中的Tensor；Caffe使用称为Blob的4维数组用于存储和交换数据。
维数解释：Blob的四维分别表示为(width_,height_,channels,num_) ，num_在视频流技术中表示为第几帧。
基本用法：Blob是一个模板类，需制定模板参数。
caffe.cpp相关代码阅读：
// Specifies the shape (dimensions) of a Blob.
// 该结构是对Blob形状信息的描述
message BlobShape {
repeated int64 dim = 1 [packed = true];
//只包含类型值为int16的，表示Blob4个维度的大小。packed表示这些值是紧密排布的，没有空洞。
}
//该结构是对Blob在磁盘中序列化之后形态的描述.
message BlobProto {
optional BlobShape shape = 7;
//可选，包含一个BlobShape对象.
repeated float data = 5 [packed = true];
//包含若干浮点数，存储数据或权值，元素数目有shape或（n,c,h,w）决定，内部精密排布。
repeated float diff = 6 [packed = true];
//包含若干浮点数，存储增量信息，维度与data一样。
repeated double double_data = 8 [packed = true];
//与data并列，但模型为double.
repeated double double_diff = 9 [packed = true];
//与diff并列，但模型为double.
// 4D dimensions -- deprecated. Use "shape" instead.
//以下为维度信息，新版本caffe推荐使用shape
optional int32 num = 1 [default = 0];
optional int32 channels = 2 [default = 0];
optional int32 height = 3 [default = 0];
optional int32 width = 4 [default = 0];
}
Blob作为一个模板类，封装了SyncedMemory类，作为基本计算单元服务Layer、Net、Slover等，源码阅读：
#ifndef CAFFE_BLOB_HPP_
#define CAFFE_BLOB_HPP_
#include <algorithm>
#include <string>
#include <vector>
#include "caffe/common.hpp"
#include "caffe/proto/caffe.pb.h" / /～由proto生成的头文件，
声明了BlobProto、 BlobShape等遵循caffe.proto协议的数据结构
#include "caffe/syncedmem.hpp" //～CPU/GPU共享内存类， 用于数据同步
const int kMaxBlobAxes = 32; //～Blob最大维数目
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 { //～类申明
public: // 默认构造函数
Blob()
: data_(), diff_(), count_(0), capacity_(0) {}
//～显示构造函数，避免隐式类型数据转换
/// @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.
*/
//～变形函数，根据输入参数重新设置当前Blob形状，必要时从新分配内存
void Reshape(const vector<int>& shape);
void Reshape(const BlobShape& shape);
void ReshapeLike(const Blob& other);
//～得到Blob形状字符串用于打印log，类似“Top shape: 100 1 28 28 (78400)”
inline string shape_string() const {
ostringstream stream;
for (int i = 0; i < shape_.size(); ++i) {
stream << shape_[i] << " ";
}
stream << "(" << count_ << ")";
return stream.str(); //~ 返回Blob形状
}
inline const vector<int>& shape() const { return shape_; } //～返回某一维度的尺寸
/**
* @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.
*/
inline int shape(int index) const {
return shape_[CanonicalAxisIndex(index)]; //～返回维度数目
}
inline int num_axes() const { return shape_.size(); } //～返回Blob中元素总数
inline int count() const { return count_; } //～返回Blob中某几维子集的总数
/**
* @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.
*/
inline int count(int start_axis, int end_axis) const {
CHECK_LE(start_axis, end_axis);//～start_axis<=end_axis
CHECK_GE(start_axis, 0); //~start_axis>=0
CHECK_GE(end_axis, 0); //~end_axis>=0
CHECK_LE(start_axis, num_axes()); //~start_axis<=总维度数
CHECK_LE(end_axis, num_axes()); //～end_axis<=总维度数
int count = 1;
for (int i = start_axis; i < end_axis; ++i) {
count *= shape(i);
}
return count;
}//~从某一维度开始，元素总数
/**
* @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.
*/
inline int count(int start_axis) const {
return count(start_axis, num_axes());
} //～转换[-N,N)->[0,N)
/**
* @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.
*/
inline int CanonicalAxisIndex(int axis_index) const {
CHECK_GE(axis_index, -num_axes()) //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()) //保证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();
//负索引表示从后向前访问，-1表示最后一个元素，普通索引值为N-1；同理，-2=>N-2,...
}
return axis_index;
}
//获取形状某一维的尺寸
/// @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); }
inline int LegacyShape(int index) const {
CHECK_LE(num_axes(), 4)
<< "Cannot use legacy accessors on Blobs with > 4 axes.";
CHECK_LT(index, 4);
CHECK_GE(index, -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);
}
//计算偏移量
inline int offset(const int n, const int c = 0, const int h = 0,
const int w = 0) const {
CHECK_GE(n, 0);
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;
}
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
*/
//按值拷贝Blob到当前的Blob
void CopyFrom(const Blob<Dtype>& source, bool copy_diff = false,
bool reshape = false);
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)];
}
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)];
}
inline Dtype data_at(const vector<int>& index) const {
return cpu_data()[offset(index)];
}
inline Dtype diff_at(const vector<int>& index) const {
return cpu_diff()[offset(index)];
}
inline const shared_ptr<SyncedMemory>& data() const {
CHECK(data_);
return data_;
}
inline const shared_ptr<SyncedMemory>& diff() const {
CHECK(diff_);
return diff_;
}
//只读访问cpu data
const Dtype* cpu_data() const;
//设置cpu data
void set_cpu_data(Dtype* data);
//只读访问gpu data
const int* gpu_shape() const;
const Dtype* gpu_data() const;
void set_gpu_data(Dtype* data);
//只读访问cpu diff
const Dtype* cpu_diff() const;
const Dtype* gpu_diff() const;
//读写访问cpu data
Dtype* mutable_cpu_data();
Dtype* mutable_gpu_data();
Dtype* mutable_cpu_diff();
Dtype* mutable_gpu_diff();
void Update();//Blob更新运算，可简单理解为data与diff的merge的过程
void FromProto(const BlobProto& proto, bool reshape = true);//反序列化函数，从BlobProto中恢复一个Blob对象。
void ToProto(BlobProto* proto, bool write_diff = false) const;//序列化函数，将内存中的Blob对象保存到BlobProto中。
/// @brief Compute the sum of absolute values (L1 norm) of the data.
Dtype asum_data() const;//计算data的L1范数
/// @brief Compute the sum of absolute values (L1 norm) of the diff.
Dtype asum_diff() const;
/// @brief Compute the sum of squares (L2 norm squared) of the data.
Dtype sumsq_data() const;//计算data的L2范数
/// @brief Compute the sum of squares (L2 norm squared) of the diff.
Dtype sumsq_diff() const;
/// @brief Scale the blob data by a constant factor.
void scale_data(Dtype scale_factor);//data乘上一个标量
/// @brief Scale the blob diff by a constant factor.
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.
*/
void ShareData(const Blob& other);//共享其他的Blob的data_
/**
* @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.
*/
void ShareDiff(const Blob& other);//共享其他的Blob的diff_
bool ShapeEquals(const BlobProto& other);
protected:
shared_ptr<SyncedMemory> data_;//存放指向data的指针
shared_ptr<SyncedMemory> diff_;
shared_ptr<SyncedMemory> shape_data_;
vector<int> shape_;//形状信息
int count_;//存放有效元素数目信息
int capacity_;//存放Blob容器的容量信息
DISABLE_COPY_AND_ASSIGN(Blob);//禁止拷贝构造函数、赋值运算符重载
}; // class Blob
} // namespace caffe
#endif // CAFFE_BLOB_HPP_