【caffe源码】Blob和syncedmem
在caffe中，blob是最基础的一个数据块，那么其实现是怎样的呢？
首先可以看见，blob.hpp只有三个头文件，分别是
#include "caffe/common.hpp"
#include "caffe/proto/caffe.pb.h"
#include "caffe/syncedmem.hpp"
commom，caffe的proto，和syncedmem。
首先
const int kMaxBlobAxes = 32;
说明了最大维度是32。
代码中也提到Blob是A wrapper around SyncedMemory holders serving as the basic computational unit.
整个头文件中，就只有一个blob的class。
首先是一个默认构造函数
public:
Blob()
: data_(), diff_(), count_(0), capacity_(0) {}
然后是指定了shape的构造函数，注意到使用了vector来表示每个维度的大小。
explicit Blob(const vector<int>& shape);
接下来是reshape和一些查询各个维度和总数的函数，和我加入在代码中的中文注释。
/**
* @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;
for (int i = 0; i < shape_.size(); ++i) {
stream << shape_[i] << " ";
}
stream << "(" << count_ << ")";
return stream.str();
}
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.
*/
inline int shape(int index) const {
return shape_[CanonicalAxisIndex(index)];
}
inline int num_axes() const { return shape_.size(); }
inline int count() const { return count_; }
//计算某些维度上的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.
*/
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.
*/
inline int count(int start_axis) const {
return count(start_axis, num_axes());
}
//获取canonical的axis，允许负数。
/**
* @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 " << 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;
}
//废弃的shape获取
/// @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);
}
//给出axis后，计算偏置
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;
}
//拷贝一个blob
/**
* @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
*/
void CopyFrom(const Blob<Dtype>& source, bool copy_diff = false,
bool reshape = false);
//获取位置数据和指针。
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_;
}
const Dtype* cpu_data() const;
void set_cpu_data(Dtype* data);
const int* gpu_shape() const;
const Dtype* gpu_data() const;
void set_gpu_data(Dtype* data);
const Dtype* cpu_diff() const;
const Dtype* gpu_diff() const;
Dtype* mutable_cpu_data();
Dtype* mutable_gpu_data();
Dtype* mutable_cpu_diff();
Dtype* mutable_gpu_diff();
void Update();
void FromProto(const BlobProto& proto, bool reshape = true);
void ToProto(BlobProto* proto, bool write_diff = false) const;
下面是获取L1,L2范数。乘一个常数。共享数据。
/// @brief Compute the sum of absolute values (L1 norm) of the data.
Dtype asum_data() const;
/// @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;
/// @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);
/// @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);
/**
* @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);
最后，看到有成员变量。
protected:
shared_ptr<SyncedMemory> data_;
shared_ptr<SyncedMemory> diff_;
shared_ptr<SyncedMemory> shape_data_;
vector<int> shape_;
int count_;
int capacity_;
可以看出，这里的指针都是SyncedMemory类型，那么SyncedMemory是什么呢？
我们可以来看看SyncedMemory.hpp。
首先是CaffeMallocHost分配空间。CaffeFreeHost来Free。如果用了MKL，则使用。
注释说的很清楚了，在多卡训练时，使用cudaMallocHost可以提高性能。
// If CUDA is available and in GPU mode, host memory will be allocated pinned,
// using cudaMallocHost. It avoids dynamic pinning for transfers (DMA).
// The improvement in performance seems negligible in the single GPU case,
// but might be more significant for parallel training. Most importantly,
// it improved stability for large models on many GPUs.
inline void CaffeMallocHost(void** ptr, size_t size, bool* use_cuda) {
#ifndef CPU_ONLY
if (Caffe::mode() == Caffe::GPU) {
CUDA_CHECK(cudaMallocHost(ptr, size));
*use_cuda = true;
return;
}
#endif
#ifdef USE_MKL
*ptr = mkl_malloc(size ? size:1, 64);
#else
*ptr = malloc(size);
#endif
*use_cuda = false;
CHECK(*ptr) << "host allocation of size " << size << " failed";
}
inline void CaffeFreeHost(void* ptr, bool use_cuda) {
#ifndef CPU_ONLY
if (use_cuda) {
CUDA_CHECK(cudaFreeHost(ptr));
return;
}
#endif
#ifdef USE_MKL
mkl_free(ptr);
#else
free(ptr);
#endif
}
下面是SyncedMemory类，此类用于在cpu和gpu之间同步内存，也就是我们的显存和内存。
public:
SyncedMemory();
explicit SyncedMemory(size_t size);
~SyncedMemory();
const void* cpu_data();
void set_cpu_data(void* data);
const void* gpu_data();
void set_gpu_data(void* data);
void* mutable_cpu_data();
void* mutable_gpu_data();
enum SyncedHead { UNINITIALIZED, HEAD_AT_CPU, HEAD_AT_GPU, SYNCED };
SyncedHead head() { return head_; }
size_t size() { return size_; }
#ifndef CPU_ONLY
void async_gpu_push(const cudaStream_t& stream);
#endif
private:
void check_device();
void to_cpu();
void to_gpu();
void* cpu_ptr_;
void* gpu_ptr_;
size_t size_;
SyncedHead head_;
bool own_cpu_data_;
bool cpu_malloc_use_cuda_;
bool own_gpu_data_;
int device_;
可以看到获取方式有const和mutable两种，以及同步状态有UNINITIALIZED, HEAD_AT_CPU, HEAD_AT_GPU, SYNCED四种。head_则是表示了当前的状态。
那么我们来看看实现。
首先是构造函数和析构函数
SyncedMemory::SyncedMemory()
: cpu_ptr_(NULL), gpu_ptr_(NULL), size_(0), head_(UNINITIALIZED),
own_cpu_data_(false), cpu_malloc_use_cuda_(false), own_gpu_data_(false) {
#ifndef CPU_ONLY
#ifdef DEBUG
CUDA_CHECK(cudaGetDevice(&device_));
#endif
#endif
}
SyncedMemory::SyncedMemory(size_t size)
: cpu_ptr_(NULL), gpu_ptr_(NULL), size_(size), head_(UNINITIALIZED),
own_cpu_data_(false), cpu_malloc_use_cuda_(false), own_gpu_data_(false) {
#ifndef CPU_ONLY
#ifdef DEBUG
CUDA_CHECK(cudaGetDevice(&device_));
#endif
#endif
}
SyncedMemory::~SyncedMemory() {
check_device();
if (cpu_ptr_ && own_cpu_data_) {
CaffeFreeHost(cpu_ptr_, cpu_malloc_use_cuda_);
}
#ifndef CPU_ONLY
if (gpu_ptr_ && own_gpu_data_) {
CUDA_CHECK(cudaFree(gpu_ptr_));
}
#endif // CPU_ONLY
}
CPU和GPU的转移函数。
inline void SyncedMemory::to_cpu() {
check_device();
switch (head_) {
case UNINITIALIZED:
CaffeMallocHost(&cpu_ptr_, size_, &cpu_malloc_use_cuda_);
caffe_memset(size_, 0, cpu_ptr_);
head_ = HEAD_AT_CPU;
own_cpu_data_ = true;
break;
case HEAD_AT_GPU:
#ifndef CPU_ONLY
if (cpu_ptr_ == NULL) {
CaffeMallocHost(&cpu_ptr_, size_, &cpu_malloc_use_cuda_);
own_cpu_data_ = true;
}
caffe_gpu_memcpy(size_, gpu_ptr_, cpu_ptr_);
head_ = SYNCED;
#else
NO_GPU;
#endif
break;
case HEAD_AT_CPU:
case SYNCED:
break;
}
}
inline void SyncedMemory::to_gpu() {
check_device();
#ifndef CPU_ONLY
switch (head_) {
case UNINITIALIZED:
CUDA_CHECK(cudaMalloc(&gpu_ptr_, size_));
caffe_gpu_memset(size_, 0, gpu_ptr_);
head_ = HEAD_AT_GPU;
own_gpu_data_ = true;
break;
case HEAD_AT_CPU:
if (gpu_ptr_ == NULL) {
CUDA_CHECK(cudaMalloc(&gpu_ptr_, size_));
own_gpu_data_ = true;
}
caffe_gpu_memcpy(size_, cpu_ptr_, gpu_ptr_);
head_ = SYNCED;
break;
case HEAD_AT_GPU:
case SYNCED:
break;
}
#else
NO_GPU;
#endif
}
获取，重置数据。
const void* SyncedMemory::cpu_data() {
check_device();
to_cpu();
return (const void*)cpu_ptr_;
}
void SyncedMemory::set_cpu_data(void* data) {
check_device();
CHECK(data);
if (own_cpu_data_) {
CaffeFreeHost(cpu_ptr_, cpu_malloc_use_cuda_);
}
cpu_ptr_ = data;
head_ = HEAD_AT_CPU;
own_cpu_data_ = false;
}
const void* SyncedMemory::gpu_data() {
check_device();
#ifndef CPU_ONLY
to_gpu();
return (const void*)gpu_ptr_;
#else
NO_GPU;
return NULL;
#endif
}
void SyncedMemory::set_gpu_data(void* data) {
check_device();
#ifndef CPU_ONLY
CHECK(data);
if (own_gpu_data_) {
CUDA_CHECK(cudaFree(gpu_ptr_));
}
gpu_ptr_ = data;
head_ = HEAD_AT_GPU;
own_gpu_data_ = false;
#else
NO_GPU;
#endif
}
void* SyncedMemory::mutable_cpu_data() {
check_device();
to_cpu();
head_ = HEAD_AT_CPU;
return cpu_ptr_;
}
void* SyncedMemory::mutable_gpu_data() {
check_device();
#ifndef CPU_ONLY
to_gpu();
head_ = HEAD_AT_GPU;
return gpu_ptr_;
#else
NO_GPU;
return NULL;
#endif
}
GPU推数据，查询GPU。
#ifndef CPU_ONLY
void SyncedMemory::async_gpu_push(const cudaStream_t& stream) {
check_device();
CHECK(head_ == HEAD_AT_CPU);
if (gpu_ptr_ == NULL) {
CUDA_CHECK(cudaMalloc(&gpu_ptr_, size_));
own_gpu_data_ = true;
}
const cudaMemcpyKind put = cudaMemcpyHostToDevice;
CUDA_CHECK(cudaMemcpyAsync(gpu_ptr_, cpu_ptr_, size_, put, stream));
// Assume caller will synchronize on the stream before use
head_ = SYNCED;
}
#endif
void SyncedMemory::check_device() {
#ifndef CPU_ONLY
#ifdef DEBUG
int device;
cudaGetDevice(&device);
CHECK(device == device_);
if (gpu_ptr_ && own_gpu_data_) {
cudaPointerAttributes attributes;
CUDA_CHECK(cudaPointerGetAttributes(&attributes, gpu_ptr_));
CHECK(attributes.device == device_);
}
#endif
#endif
}
好了，syncedmem看完，现在来看一看blob的实现吧。
首先是一个Reshape函数，将传入的参数转为一个vector或传入vector来进行reshape。
注意到初始化blob也是通过reshape实现的，空间不足时会自动分配空间。
template <typename Dtype>
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>
void Blob<Dtype>::Reshape(const vector<int>& shape) {
CHECK_LE(shape.size(), kMaxBlobAxes);
count_ = 1;
shape_.resize(shape.size());
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) {
CHECK_GE(shape[i], 0);
if (count_ != 0) {
CHECK_LE(shape[i], INT_MAX / count_) << "blob size exceeds INT_MAX";
}
count_ *= shape[i];
shape_[i] = shape[i];
shape_data[i] = shape[i];
}
if (count_ > capacity_) {
capacity_ = count_;
data_.reset(new SyncedMemory(capacity_ * sizeof(Dtype)));
diff_.reset(new SyncedMemory(capacity_ * sizeof(Dtype)));
}
}
template <typename Dtype>
void Blob<Dtype>::Reshape(const BlobShape& shape) {
CHECK_LE(shape.dim_size(), kMaxBlobAxes);
vector<int> shape_vec(shape.dim_size());
for (int i = 0; i < shape.dim_size(); ++i) {
shape_vec[i] = shape.dim(i);
}
Reshape(shape_vec);
}
template <typename Dtype>
void Blob<Dtype>::ReshapeLike(const Blob<Dtype>& other) {
Reshape(other.shape());
}
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);
}
template <typename Dtype>
Blob<Dtype>::Blob(const vector<int>& shape)
// capacity_ must be initialized before calling Reshape
: capacity_(0) {
Reshape(shape);
}
获取shape，数据的函数实现。
template <typename Dtype>
const int* Blob<Dtype>::gpu_shape() const {
CHECK(shape_data_);
return (const int*)shape_data_->gpu_data();
}
template <typename Dtype>
const Dtype* Blob<Dtype>::cpu_data() const {
CHECK(data_);
return (const Dtype*)data_->cpu_data();
}
template <typename Dtype>
void Blob<Dtype>::set_cpu_data(Dtype* data) {
CHECK(data);
// Make sure CPU and GPU sizes remain equal
size_t size = count_ * sizeof(Dtype);
if (data_->size() != size) {
data_.reset(new SyncedMemory(size));
diff_.reset(new SyncedMemory(size));
}
data_->set_cpu_data(data);
}
template <typename Dtype>
const Dtype* Blob<Dtype>::gpu_data() const {
CHECK(data_);
return (const Dtype*)data_->gpu_data();
}
template <typename Dtype>
void Blob<Dtype>::set_gpu_data(Dtype* data) {
CHECK(data);
// Make sure CPU and GPU sizes remain equal
size_t size = count_ * sizeof(Dtype);
if (data_->size() != size) {
data_.reset(new SyncedMemory(size));
diff_.reset(new SyncedMemory(size));
}
data_->set_gpu_data(data);
}
template <typename Dtype>
const Dtype* Blob<Dtype>::cpu_diff() const {
CHECK(diff_);
return (const Dtype*)diff_->cpu_data();
}
template <typename Dtype>
const Dtype* Blob<Dtype>::gpu_diff() const {
CHECK(diff_);
return (const Dtype*)diff_->gpu_data();
}
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_cpu_data() {
CHECK(data_);
return static_cast<Dtype*>(data_->mutable_cpu_data());
}
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_gpu_data() {
CHECK(data_);
return static_cast<Dtype*>(data_->mutable_gpu_data());
}
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_cpu_diff() {
CHECK(diff_);
return static_cast<Dtype*>(diff_->mutable_cpu_data());
}
template <typename Dtype>
Dtype* Blob<Dtype>::mutable_gpu_diff() {
CHECK(diff_);
return static_cast<Dtype*>(diff_->mutable_gpu_data());
}
template <typename Dtype>
void Blob<Dtype>::ShareData(const Blob& other) {
CHECK_EQ(count_, other.count());
data_ = other.data();
}
template <typename Dtype>
void Blob<Dtype>::ShareDiff(const Blob& other) {
CHECK_EQ(count_, other.count());
diff_ = other.diff();
}
更新blob在两端的数据
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
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.";
}
}
求范数的代码，考虑了GPU的实现。
template <> unsigned int Blob<unsigned int>::asum_data() const {
NOT_IMPLEMENTED;
return 0;
}
template <> int Blob<int>::asum_data() const {
NOT_IMPLEMENTED;
return 0;
}
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;
}
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;
}
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;
}
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;
}
Scala 数据的实现。
template <> void Blob<unsigned int>::scale_data(unsigned int scale_factor) {
NOT_IMPLEMENTED;
}
template <> void Blob<int>::scale_data(int scale_factor) {
NOT_IMPLEMENTED;
}
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;
}
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();
}
}
判断类型是否一致及拷贝数据。
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();
}
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;
}
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);
} else {
LOG(FATAL) << "Trying to copy blobs of different sizes.";
}
}
switch (Caffe::mode()) {
case Caffe::GPU:
if (copy_diff) {
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;
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.";
}
}
Proto的序列化和反序列化。
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).
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);
} else {
CHECK(ShapeEquals(proto)) << "shape mismatch (reshape not set)";
}
// copy data
Dtype* data_vec = mutable_cpu_data();
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();
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);
}
}
}
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]);
}
if (write_diff) {
const double* diff_vec = cpu_diff();
for (int i = 0; i < count_; ++i) {
proto->add_double_diff(diff_vec[i]);
}
}
}
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]);
}
}
}