Net源文件
之前介绍了头文件,头文件中,主要定义了一些简单的内联函数以及声明一些成员函数,总的来说,Net实现的是网络中的层、层的输入输出变量、层的权重的统一调度(因为变量太多了)
#include <algorithm>
#include <map>
#include <set>
#include <string>
#include <utility>
#include <vector>
#include "hdf5.h"
#include "caffe/common.hpp"
#include "caffe/layer.hpp"
#include "caffe/net.hpp"
#include "caffe/parallel.hpp"
#include "caffe/proto/caffe.pb.h"
#include "caffe/util/hdf5.hpp"
#include "caffe/util/insert_splits.hpp"
#include "caffe/util/math_functions.hpp"
#include "caffe/util/upgrade_proto.hpp"
namespace caffe {
template <typename Dtype>
Net<Dtype>::Net(const NetParameter& param) {
Init(param);
}
template <typename Dtype>
Net<Dtype>::Net(const string& param_file, Phase phase,
const int level, const vector<string>* stages) {
NetParameter param;
ReadNetParamsFromTextFileOrDie(param_file, ¶m);
// Set phase, stages and level
//先说明,当前建立的网络是训练网络还是测试网络
param.mutable_state()->set_phase(phase);
if (stages != NULL) {
for (int i = 0; i < stages->size(); i++) {
param.mutable_state()->add_stage((*stages)[i]);
}
}
param.mutable_state()->set_level(level);
Init(param); // 这个Net类的关键成分,建立各种层,然后就是分配内存,归类指针(就是说,
//某些指针指向的数据块是干啥,具体来讲,一般是可训练参量,所有参量,输入参量,输出参量,网络的输入和输出)
}
//具体来看看怎么初始化的
template <typename Dtype>
void Net<Dtype>::Init(const NetParameter& in_param) {
// Set phase from the state.
phase_ = in_param.state().phase(); //设置当前阶段
// Filter layers based on their include/exclude rules and
// the current NetState.
NetParameter filtered_param;
FilterNet(in_param, &filtered_param);
LOG_IF(INFO, Caffe::root_solver())
<< "Initializing net from parameters: " << std::endl
<< filtered_param.DebugString();
// Create a copy of filtered_param with splits added where necessary.
NetParameter param;
InsertSplits(filtered_param, ¶m);
// Basically, build all the layers and set up their connections.
name_ = param.name(); //网络的名称
map<string, int> blob_name_to_idx;//blobs_的名字和索引的映射
set<string> available_blobs;
memory_used_ = 0;//当前内存的使用量
// For each layer, set up its input and output
bottom_vecs_.resize(param.layer_size()); //每一层的输入指针容器初始化
top_vecs_.resize(param.layer_size());//每一层输出的指针容器初始化
bottom_id_vecs_.resize(param.layer_size());//每一层的输入索引数初始化
param_id_vecs_.resize(param.layer_size());//网络参量索引初始化
top_id_vecs_.resize(param.layer_size());//每一层的输出索引数初始化
bottom_need_backward_.resize(param.layer_size());//每一层的输入是否需要反向传播初始化
//每一层所处阶段初始化,注意这是一个超大层的循环,循环建立每一层
for (int layer_id = 0; layer_id < param.layer_size(); ++layer_id) {
// Inherit phase from net if unset.
if (!param.layer(layer_id).has_phase()) {
param.mutable_layer(layer_id)->set_phase(phase_);
}
// Setup layer.
const LayerParameter& layer_param = param.layer(layer_id);
if (layer_param.propagate_down_size() > 0) {
CHECK_EQ(layer_param.propagate_down_size(),
layer_param.bottom_size())
<< "propagate_down param must be specified "
<< "either 0 or bottom_size times ";
}
//创建层以及对应的层名字并将其压入层和层名字指针容器,
layers_.push_back(LayerRegistry<Dtype>::CreateLayer(layer_param));
layer_names_.push_back(layer_param.name());
LOG_IF(INFO, Caffe::root_solver())
<< "Creating Layer " << layer_param.name();
//默认该层不反向传播
bool need_backward = false;
// Figure out this layer's input and output
//首先,每一层可能有多个输入和多个输出,这个for表示第layer_id层的输入,并将其
//加入到bottom_vecs_这个指针容器中,AppendBottom、AppendBottom、AppendParam为Net初始化的核心三个函数
for (int bottom_id = 0; bottom_id < layer_param.bottom_size();
++bottom_id) {
const int blob_id = AppendBottom(param, layer_id, bottom_id,
&available_blobs, &blob_name_to_idx);
// If a blob needs backward, this layer should provide it.
//如果bottom有一个输入需要反向传播,则该层就需要反向传播
need_backward |= blob_need_backward_[blob_id];
}
//添加了bottom之后,现在来添加top
//留一个悬念稍后揭晓,一个层的输出也就是下一个层的输入,top_vecs和bottom_vecs
//有一部分是指向相同的内存空间,caffe怎么处理的呢?
int num_top = layer_param.top_size();
for (int top_id = 0; top_id < num_top; ++top_id) {
AppendTop(param, layer_id, top_id, &available_blobs, &blob_name_to_idx);
// Collect Input layer tops as Net inputs.
// 如果当前层,是输入层,那么初始化输入指针
if (layer_param.type() == "Input") {
const int blob_id = blobs_.size() - 1;
net_input_blob_indices_.push_back(blob_id);
net_input_blobs_.push_back(blobs_[blob_id].get());
}
}
// If the layer specifies that AutoTopBlobs() -> true and the LayerParameter
// specified fewer than the required number (as specified by
// ExactNumTopBlobs() or MinTopBlobs()), allocate them here.
Layer<Dtype>* layer = layers_[layer_id].get();
//在caffe中,网络会自动给损失层分配一个Blob,即使在prototxt中并未说明
//这样做的目的就是保证网络反向传播的方便和计算的兼容
//简单说,就是损失层的输出,也是top的一个
if (layer->AutoTopBlobs()) {
const int needed_num_top =
std::max(layer->MinTopBlobs(), layer->ExactNumTopBlobs());
for (; num_top < needed_num_top; ++num_top) {
// Add "anonymous" top blobs -- do not modify available_blobs or
// blob_name_to_idx as we don't want these blobs to be usable as input
// to other layers.
AppendTop(param, layer_id, num_top, NULL, NULL);
}
}
// After this layer is connected, set it up.
//好了之前把,哪个指针指向哪个数据块(Blob)整明白了,
//但是记住的是这些数据块(Blob)并没有初始化其大小,以及形状,简单说,
//刚刚那些指针只是指向一个空数据块(Blob)
//接下来,就要调用每一层的Setup成员变量,初始化每一个数据块的形状大小了,分配内存空间了
layers_[layer_id]->SetUp(bottom_vecs_[layer_id], top_vecs_[layer_id]);
LOG_IF(INFO, Caffe::root_solver())
<< "Setting up " << layer_names_[layer_id];
//这个就是,分配损失层的权值了,
//正常的网络一般只有一个损害函数,但是也有多个损失函数的比如,Faster-RCNN人家可是有4个损失函数,因此,就需要为每一个损失函数定义一个权值
for (int top_id = 0; top_id < top_vecs_[layer_id].size(); ++top_id) {
if (blob_loss_weights_.size() <= top_id_vecs_[layer_id][top_id]) {
blob_loss_weights_.resize(top_id_vecs_[layer_id][top_id] + 1, Dtype(0));
}
blob_loss_weights_[top_id_vecs_[layer_id][top_id]] = layer->loss(top_id);
LOG_IF(INFO, Caffe::root_solver())
<< "Top shape: " << top_vecs_[layer_id][top_id]->shape_string();
if (layer->loss(top_id)) {
LOG_IF(INFO, Caffe::root_solver())
<< " with loss weight " << layer->loss(top_id);
}
//然后接下来结算一下,现在到底用了多少内存空间,累加的
memory_used_ += top_vecs_[layer_id][top_id]->count();
}
LOG_IF(INFO, Caffe::root_solver())
<< "Memory required for data: " << memory_used_ * sizeof(Dtype);
const int param_size = layer_param.param_size();
//当前层有多少个参量,注意这里的参量不是可学习的,只是layer中的成员变量blobs_
const int num_param_blobs = layers_[layer_id]->blobs().size();
CHECK_LE(param_size, num_param_blobs)
<< "Too many params specified for layer " << layer_param.name();
ParamSpec default_param_spec;
//然后,进行判定呀,如果该层的blobs_中有一个是需要反向传播的那么该层就需要反向传播
for (int param_id = 0; param_id < num_param_blobs; ++param_id) {
const ParamSpec* param_spec = (param_id < param_size) ?
&layer_param.param(param_id) : &default_param_spec;
const bool param_need_backward = param_spec->lr_mult() != 0;
need_backward |= param_need_backward;
layers_[layer_id]->set_param_propagate_down(param_id,
param_need_backward);
}
// 将net中的参数指针指向每一层的参数块,方便网络训练!
for (int param_id = 0; param_id < num_param_blobs; ++param_id) {
AppendParam(param, layer_id, param_id);
}
// Finally, set the backward flag
//最后设定方向传播标志位
layer_need_backward_.push_back(need_backward);
if (need_backward) {
for (int top_id = 0; top_id < top_id_vecs_[layer_id].size(); ++top_id) {
blob_need_backward_[top_id_vecs_[layer_id][top_id]] = true;
}
}
}
//以上便是每一个层的大循环,完成了大部分指针指向地址的分配,有利于在网络训练过程
//中,对可训练参数、每层的输入输出统一调度
// Go through the net backwards to determine which blobs contribute to the
// loss. We can skip backward computation for blobs that don't contribute
// to the loss.
// Also checks if all bottom blobs don't need backward computation (possible
// because the skip_propagate_down param) and so we can skip bacward
// computation for the entire layer
//以下代码,还是针对每一层是否反向传播,在上面的大循环里面,
//如果每一层如果有一个bottom需要反向传播,则该层需要反向传播
//如果每一层如果有一个blobs_需要反向传播,则该层需要反向传播
//以下的代码,主要说明,当某一层对损失函数没有贡献则可以跳过该层不进行反向传播
set<string> blobs_under_loss;
set<string> blobs_skip_backp;
//下面是两个嵌套循环,从最后一层向前遍历,然后对每一层的top进行遍历
//对于这一部分代码,有三点需要特别注意的:
//第一,在caffe中最后一层总是损失层,因此最开始layer_contributes_loss
//=ture,因此,将最后一层的bottom添加到blobs_under_loss这个set中;
//第二,最后一层的输入必定是损失值,即loss_weight != 0;
//第三,在代码中,大循环从最后一层到第一层,在小循环中,先遍历top, 而后遍历bottom,
//而下一层的bottom,就是上一层的top。
//总结:就是和损失层直接间接相连的都要被反向传播,因为layer_contributes_loss = true, 一直从损失到间接相连的第一层
for (int layer_id = layers_.size() - 1; layer_id >= 0; --layer_id) {
bool layer_contributes_loss = false;
bool layer_skip_propagate_down = true;
for (int top_id = 0; top_id < top_vecs_[layer_id].size(); ++top_id) {
//第layer_id层第top_id个输出的名字
const string& blob_name = blob_names_[top_id_vecs_[layer_id][top_id]];
//如果当前层的第top_id个输出的loss_weight != 0 或者 在
//blobs_under_loss容器中找到blob_name,
//则当前层的第top_id个输出对损失有贡献
if (layers_[layer_id]->loss(top_id) ||
(blobs_under_loss.find(blob_name) != blobs_under_loss.end())) {
layer_contributes_loss = true;
}
// 如果在blobs_skip_backp.find没有找到blob_name, 则当前层
//在反向传播时,不能被忽略
if (blobs_skip_backp.find(blob_name) == blobs_skip_backp.end()) {
layer_skip_propagate_down = false;
}
// 如果既对loss有贡献,又不能够被忽略,那就正常吧
if (layer_contributes_loss && !layer_skip_propagate_down)
break;
}
// If this layer can skip backward computation, also all his bottom blobs
// don't need backpropagation
// 如果当前层需要反向传播,但是当前层反向传播可以被忽略,那么
// 当前层就没必要反向传播了,所以呢?顺带把所有输入也不必要反向传播了
if (layer_need_backward_[layer_id] && layer_skip_propagate_down) {
layer_need_backward_[layer_id] = false;
for (int bottom_id = 0; bottom_id < bottom_vecs_[layer_id].size();
++bottom_id) {
bottom_need_backward_[layer_id][bottom_id] = false;
}
}
if (!layer_contributes_loss) { layer_need_backward_[layer_id] = false; }
if (Caffe::root_solver()) {
if (layer_need_backward_[layer_id]) {
LOG(INFO) << layer_names_[layer_id] << " needs backward computation.";
} else {
LOG(INFO) << layer_names_[layer_id]
<< " does not need backward computation.";
}
}
//遍历完top,就要来遍历bottom,如果top对损失有贡献,则他的bottom也是有贡献的
//此外,当前层的bottom也是下一层的top
for (int bottom_id = 0; bottom_id < bottom_vecs_[layer_id].size();
++bottom_id) {
if (layer_contributes_loss) {
const string& blob_name =
blob_names_[bottom_id_vecs_[layer_id][bottom_id]];
blobs_under_loss.insert(blob_name); //插入,当前数据块对损失有影响
} else {
bottom_need_backward_[layer_id][bottom_id] = false;
}
if (!bottom_need_backward_[layer_id][bottom_id]) {
const string& blob_name =
blob_names_[bottom_id_vecs_[layer_id][bottom_id]];
blobs_skip_backp.insert(blob_name);
}
}
}
// Handle force_backward if needed.
if (param.force_backward()) {
for (int layer_id = 0; layer_id < layers_.size(); ++layer_id) {
layer_need_backward_[layer_id] = true;
for (int bottom_id = 0;
bottom_id < bottom_need_backward_[layer_id].size(); ++bottom_id) {
bottom_need_backward_[layer_id][bottom_id] =
bottom_need_backward_[layer_id][bottom_id] ||
layers_[layer_id]->AllowForceBackward(bottom_id);
blob_need_backward_[bottom_id_vecs_[layer_id][bottom_id]] =
blob_need_backward_[bottom_id_vecs_[layer_id][bottom_id]] ||
bottom_need_backward_[layer_id][bottom_id];
}
for (int param_id = 0; param_id < layers_[layer_id]->blobs().size();
++param_id) {
layers_[layer_id]->set_param_propagate_down(param_id, true);
}
}
}
// In the end, all remaining blobs are considered output blobs.
//剩下的数据块就被认为是网络输出了
for (set<string>::iterator it = available_blobs.begin();
it != available_blobs.end(); ++it) {
LOG_IF(INFO, Caffe::root_solver())
<< "This network produces output " << *it;
net_output_blobs_.push_back(blobs_[blob_name_to_idx[*it]].get());
net_output_blob_indices_.push_back(blob_name_to_idx[*it]);
}
for (size_t blob_id = 0; blob_id < blob_names_.size(); ++blob_id) {
blob_names_index_[blob_names_[blob_id]] = blob_id;
}
for (size_t layer_id = 0; layer_id < layer_names_.size(); ++layer_id) {
layer_names_index_[layer_names_[layer_id]] = layer_id;
}
ShareWeights();
debug_info_ = param.debug_info();
LOG_IF(INFO, Caffe::root_solver()) << "Network initialization done.";
}
template <typename Dtype>
void Net<Dtype>::FilterNet(const NetParameter& param,
NetParameter* param_filtered) {
NetState net_state(param.state());
param_filtered->CopyFrom(param);
param_filtered->clear_layer();
for (int i = 0; i < param.layer_size(); ++i) {
const LayerParameter& layer_param = param.layer(i);
const string& layer_name = layer_param.name();
CHECK(layer_param.include_size() == 0 || layer_param.exclude_size() == 0)
<< "Specify either include rules or exclude rules; not both.";
// If no include rules are specified, the layer is included by default and
// only excluded if it meets one of the exclude rules.
bool layer_included = (layer_param.include_size() == 0);
for (int j = 0; layer_included && j < layer_param.exclude_size(); ++j) {
if (StateMeetsRule(net_state, layer_param.exclude(j), layer_name)) {
layer_included = false;
}
}
for (int j = 0; !layer_included && j < layer_param.include_size(); ++j) {
if (StateMeetsRule(net_state, layer_param.include(j), layer_name)) {
layer_included = true;
}
}
if (layer_included) {
param_filtered->add_layer()->CopyFrom(layer_param);
}
}
}
template <typename Dtype>
bool Net<Dtype>::StateMeetsRule(const NetState& state,
const NetStateRule& rule, const string& layer_name) {
// Check whether the rule is broken due to phase.
if (rule.has_phase()) {
if (rule.phase() != state.phase()) {
LOG_IF(INFO, Caffe::root_solver())
<< "The NetState phase (" << state.phase()
<< ") differed from the phase (" << rule.phase()
<< ") specified by a rule in layer " << layer_name;
return false;
}
}
// Check whether the rule is broken due to min level.
if (rule.has_min_level()) {
if (state.level() < rule.min_level()) {
LOG_IF(INFO, Caffe::root_solver())
<< "The NetState level (" << state.level()
<< ") is above the min_level (" << rule.min_level()
<< ") specified by a rule in layer " << layer_name;
return false;
}
}
// Check whether the rule is broken due to max level.
if (rule.has_max_level()) {
if (state.level() > rule.max_level()) {
LOG_IF(INFO, Caffe::root_solver())
<< "The NetState level (" << state.level()
<< ") is above the max_level (" << rule.max_level()
<< ") specified by a rule in layer " << layer_name;
return false;
}
}
// Check whether the rule is broken due to stage. The NetState must
// contain ALL of the rule's stages to meet it.
for (int i = 0; i < rule.stage_size(); ++i) {
// Check that the NetState contains the rule's ith stage.
bool has_stage = false;
for (int j = 0; !has_stage && j < state.stage_size(); ++j) {
if (rule.stage(i) == state.stage(j)) { has_stage = true; }
}
if (!has_stage) {
LOG_IF(INFO, Caffe::root_solver())
<< "The NetState did not contain stage '" << rule.stage(i)
<< "' specified by a rule in layer " << layer_name;
return false;
}
}
// Check whether the rule is broken due to not_stage. The NetState must
// contain NONE of the rule's not_stages to meet it.
for (int i = 0; i < rule.not_stage_size(); ++i) {
// Check that the NetState contains the rule's ith not_stage.
bool has_stage = false;
for (int j = 0; !has_stage && j < state.stage_size(); ++j) {
if (rule.not_stage(i) == state.stage(j)) { has_stage = true; }
}
if (has_stage) {
LOG_IF(INFO, Caffe::root_solver())
<< "The NetState contained a not_stage '" << rule.not_stage(i)
<< "' specified by a rule in layer " << layer_name;
return false;
}
}
return true;
}
// Helper for Net::Init: add a new top blob to the net.
//AppendTop则是将每层的输出,进行归类,同时创建指针
//这里创建的是每一层输入和输出的指针,以及开辟数据块的内存
//在Layer层构造函数,只是初始化每层的blob_,而输入输出当作参数并没有创建
//和开辟内存空间
//在Init函数中,首先调用的是AppendBottom然后是AppendTop
//在这里,我们先来理一下思路,
//第一,上一层的输出,是下一层的输入;
//第二,数据输入层只有Top没有Bottom
//第三,初始化是从第一层到损失层
//因此,首先遍历第一层时,并没有bottom,而后创建第一层的输出,
//当遍历第二层时,第一层的输出,也就是第二层的bottom,空间已经开辟好了,指针、名字
//索引直接按照要求分配就行...然后 第二层的top也就是第三层的bottom..依次循环下去
//创建整个网络的每一层输入输出
template <typename Dtype>
void Net<Dtype>::AppendTop(const NetParameter& param, const int layer_id,
const int top_id, set<string>* available_blobs,
map<string, int>* blob_name_to_idx) {
shared_ptr<LayerParameter> layer_param(
new LayerParameter(param.layer(layer_id)));
//若当前的top_id 大于该层的输出个数,则blob_name = (automatic)
//否则获取当前数据块的名字,后面通过名字找指针用
const string& blob_name = (layer_param->top_size() > top_id) ?
layer_param->top(top_id) : "(automatic)";
// Check if we are doing in-place computation
//判断该层的top 和bottom是否占用同样的内存块(也就是in-place机制)
//如果是in-place的话,就不需要重新创建(已经创建好了),只需把已有指针拷贝过来就行了
if (blob_name_to_idx && layer_param->bottom_size() > top_id &&
blob_name == layer_param->bottom(top_id)) {
// In-place computation
LOG_IF(INFO, Caffe::root_solver())
<< layer_param->name() << " -> " << blob_name << " (in-place)";
top_vecs_[layer_id].push_back(blobs_[(*blob_name_to_idx)[blob_name]].get());
top_id_vecs_[layer_id].push_back((*blob_name_to_idx)[blob_name]);
} else if (blob_name_to_idx &&
blob_name_to_idx->find(blob_name) != blob_name_to_idx->end()) {
// If we are not doing in-place computation but have duplicated blobs,
// raise an error.
LOG(FATAL) << "Top blob '" << blob_name
<< "' produced by multiple sources.";
} else {
// Normal output.
//正常的情况,就是 top的指针没有创建,bottom的指针已经创建好了
if (Caffe::root_solver()) {
LOG(INFO) << layer_param->name() << " -> " << blob_name;
}
//创建该层第top_id个输入,开辟一个容量为0的内存空间,然后将之存储在
//blob_中,以及存储该数据块的名字,名字和索引的映射等等,一系列的
//插入操作
shared_ptr<Blob<Dtype> > blob_pointer(new Blob<Dtype>());
const int blob_id = blobs_.size();
blobs_.push_back(blob_pointer);
blob_names_.push_back(blob_name);
blob_need_backward_.push_back(false);
if (blob_name_to_idx) { (*blob_name_to_idx)[blob_name] = blob_id; }
top_id_vecs_[layer_id].push_back(blob_id);
top_vecs_[layer_id].push_back(blob_pointer.get());
}
if (available_blobs) { available_blobs->insert(blob_name); }
}
// Helper for Net::Init: add a new bottom blob to the net.
template <typename Dtype>
int Net<Dtype>::AppendBottom(const NetParameter& param, const int layer_id,
const int bottom_id, set<string>* available_blobs,
map<string, int>* blob_name_to_idx) {
const LayerParameter& layer_param = param.layer(layer_id);
const string& blob_name = layer_param.bottom(bottom_id);
if (available_blobs->find(blob_name) == available_blobs->end()) {
LOG(FATAL) << "Unknown bottom blob '" << blob_name << "' (layer '"
<< layer_param.name() << "', bottom index " << bottom_id << ")";
}
const int blob_id = (*blob_name_to_idx)[blob_name];
LOG_IF(INFO, Caffe::root_solver())
<< layer_names_[layer_id] << " <- " << blob_name;
bottom_vecs_[layer_id].push_back(blobs_[blob_id].get());
bottom_id_vecs_[layer_id].push_back(blob_id);
available_blobs->erase(blob_name);
bool need_backward = blob_need_backward_[blob_id];
// Check if the backpropagation on bottom_id should be skipped
if (layer_param.propagate_down_size() > 0) {
need_backward = layer_param.propagate_down(bottom_id);
}
bottom_need_backward_[layer_id].push_back(need_backward);
return blob_id;
}
template <typename Dtype>
void Net<Dtype>::AppendParam(const NetParameter& param, const int layer_id,
const int param_id) {
const LayerParameter& layer_param = layers_[layer_id]->layer_param();
const int param_size = layer_param.param_size();
string param_name =
(param_size > param_id) ? layer_param.param(param_id).name() : "";
if (param_name.size()) {
param_display_names_.push_back(param_name);
} else {
ostringstream param_display_name;
param_display_name << param_id;
param_display_names_.push_back(param_display_name.str());
}
const int net_param_id = params_.size();
params_.push_back(layers_[layer_id]->blobs()[param_id]);
param_id_vecs_[layer_id].push_back(net_param_id);
param_layer_indices_.push_back(make_pair(layer_id, param_id));
ParamSpec default_param_spec;
const ParamSpec* param_spec = (layer_param.param_size() > param_id) ?
&layer_param.param(param_id) : &default_param_spec;
if (!param_size || !param_name.size() || (param_name.size() &&
param_names_index_.find(param_name) == param_names_index_.end())) {
// This layer "owns" this parameter blob -- it is either anonymous
// (i.e., not given a param_name) or explicitly given a name that we
// haven't already seen.
param_owners_.push_back(-1);
if (param_name.size()) {
param_names_index_[param_name] = net_param_id;
}
const int learnable_param_id = learnable_params_.size();
learnable_params_.push_back(params_[net_param_id].get());
learnable_param_ids_.push_back(learnable_param_id);
has_params_lr_.push_back(param_spec->has_lr_mult());
has_params_decay_.push_back(param_spec->has_decay_mult());
params_lr_.push_back(param_spec->lr_mult());
params_weight_decay_.push_back(param_spec->decay_mult());
} else {
// Named param blob with name we've seen before: share params
const int owner_net_param_id = param_names_index_[param_name];
param_owners_.push_back(owner_net_param_id);
const pair<int, int>& owner_index =
param_layer_indices_[owner_net_param_id];
const int owner_layer_id = owner_index.first;
const int owner_param_id = owner_index.second;
LOG_IF(INFO, Caffe::root_solver()) << "Sharing parameters '" << param_name
<< "' owned by "
<< "layer '" << layer_names_[owner_layer_id] << "', param "
<< "index " << owner_param_id;
Blob<Dtype>* this_blob = layers_[layer_id]->blobs()[param_id].get();
Blob<Dtype>* owner_blob =
layers_[owner_layer_id]->blobs()[owner_param_id].get();
const int param_size = layer_param.param_size();
if (param_size > param_id && (layer_param.param(param_id).share_mode() ==
ParamSpec_DimCheckMode_PERMISSIVE)) {
// Permissive dimension checking -- only check counts are the same.
CHECK_EQ(this_blob->count(), owner_blob->count())
<< "Cannot share param '" << param_name << "' owned by layer '"
<< layer_names_[owner_layer_id] << "' with layer '"
<< layer_names_[layer_id] << "'; count mismatch. Owner layer param "
<< "shape is " << owner_blob->shape_string() << "; sharing layer "
<< "shape is " << this_blob->shape_string();
} else {
// Strict dimension checking -- all dims must be the same.
CHECK(this_blob->shape() == owner_blob->shape())
<< "Cannot share param '" << param_name << "' owned by layer '"
<< layer_names_[owner_layer_id] << "' with layer '"
<< layer_names_[layer_id] << "'; shape mismatch. Owner layer param "
<< "shape is " << owner_blob->shape_string() << "; sharing layer "
<< "expects shape " << this_blob->shape_string();
}
const int learnable_param_id = learnable_param_ids_[owner_net_param_id];
learnable_param_ids_.push_back(learnable_param_id);
if (param_spec->has_lr_mult()) {
if (has_params_lr_[learnable_param_id]) {
CHECK_EQ(param_spec->lr_mult(), params_lr_[learnable_param_id])
<< "Shared param '" << param_name << "' has mismatched lr_mult.";
} else {
has_params_lr_[learnable_param_id] = true;
params_lr_[learnable_param_id] = param_spec->lr_mult();
}
}
if (param_spec->has_decay_mult()) {
if (has_params_decay_[learnable_param_id]) {
CHECK_EQ(param_spec->decay_mult(),
params_weight_decay_[learnable_param_id])
<< "Shared param '" << param_name << "' has mismatched decay_mult.";
} else {
has_params_decay_[learnable_param_id] = true;
params_weight_decay_[learnable_param_id] = param_spec->decay_mult();
}
}
}
}
template <typename Dtype>
Dtype Net<Dtype>::ForwardFromTo(int start, int end) {
CHECK_GE(start, 0);
CHECK_LT(end, layers_.size());
Dtype loss = 0;
for (int i = start; i <= end; ++i) {
for (int c = 0; c < before_forward_.size(); ++c) {
before_forward_[c]->run(i);
}
Dtype layer_loss = layers_[i]->Forward(bottom_vecs_[i], top_vecs_[i]);
loss += layer_loss;
if (debug_info_) { ForwardDebugInfo(i); }
for (int c = 0; c < after_forward_.size(); ++c) {
after_forward_[c]->run(i);
}
}
return loss;
}
template <typename Dtype>
Dtype Net<Dtype>::ForwardFrom(int start) {
return ForwardFromTo(start, layers_.size() - 1);
}
template <typename Dtype>
Dtype Net<Dtype>::ForwardTo(int end) {
return ForwardFromTo(0, end);
}
template <typename Dtype>
const vector<Blob<Dtype>*>& Net<Dtype>::Forward(Dtype* loss) {
if (loss != NULL) {
*loss = ForwardFromTo(0, layers_.size() - 1);
} else {
ForwardFromTo(0, layers_.size() - 1);
}
return net_output_blobs_;
}
template <typename Dtype>
const vector<Blob<Dtype>*>& Net<Dtype>::Forward(
const vector<Blob<Dtype>*> & bottom, Dtype* loss) {
LOG_EVERY_N(WARNING, 1000) << "DEPRECATED: Forward(bottom, loss) "
<< "will be removed in a future version. Use Forward(loss).";
// Copy bottom to net bottoms
for (int i = 0; i < bottom.size(); ++i) {
net_input_blobs_[i]->CopyFrom(*bottom[i]);
}
return Forward(loss);
}
template <typename Dtype>
void Net<Dtype>::BackwardFromTo(int start, int end) {
CHECK_GE(end, 0);
CHECK_LT(start, layers_.size());
for (int i = start; i >= end; --i) {
for (int c = 0; c < before_backward_.size(); ++c) {
before_backward_[c]->run(i);
}
if (layer_need_backward_[i]) {
layers_[i]->Backward(
top_vecs_[i], bottom_need_backward_[i], bottom_vecs_[i]);
if (debug_info_) { BackwardDebugInfo(i); }
}
for (int c = 0; c < after_backward_.size(); ++c) {
after_backward_[c]->run(i);
}
}
}
template <typename Dtype>
void Net<Dtype>::ForwardDebugInfo(const int layer_id) {
for (int top_id = 0; top_id < top_vecs_[layer_id].size(); ++top_id) {
const Blob<Dtype>& blob = *top_vecs_[layer_id][top_id];
const string& blob_name = blob_names_[top_id_vecs_[layer_id][top_id]];
const Dtype data_abs_val_mean = blob.asum_data() / blob.count();
LOG_IF(INFO, Caffe::root_solver())
<< " [Forward] "
<< "Layer " << layer_names_[layer_id]
<< ", top blob " << blob_name
<< " data: " << data_abs_val_mean;
}
for (int param_id = 0; param_id < layers_[layer_id]->blobs().size();
++param_id) {
const Blob<Dtype>& blob = *layers_[layer_id]->blobs()[param_id];
const int net_param_id = param_id_vecs_[layer_id][param_id];
const string& blob_name = param_display_names_[net_param_id];
const Dtype data_abs_val_mean = blob.asum_data() / blob.count();
LOG_IF(INFO, Caffe::root_solver())
<< " [Forward] "
<< "Layer " << layer_names_[layer_id]
<< ", param blob " << blob_name
<< " data: " << data_abs_val_mean;
}
}
template <typename Dtype>
void Net<Dtype>::BackwardDebugInfo(const int layer_id) {
const vector<Blob<Dtype>*>& bottom_vec = bottom_vecs_[layer_id];
for (int bottom_id = 0; bottom_id < bottom_vec.size(); ++bottom_id) {
if (!bottom_need_backward_[layer_id][bottom_id]) { continue; }
const Blob<Dtype>& blob = *bottom_vec[bottom_id];
const string& blob_name = blob_names_[bottom_id_vecs_[layer_id][bottom_id]];
const Dtype diff_abs_val_mean = blob.asum_diff() / blob.count();
LOG_IF(INFO, Caffe::root_solver())
<< " [Backward] "
<< "Layer " << layer_names_[layer_id]
<< ", bottom blob " << blob_name
<< " diff: " << diff_abs_val_mean;
}
for (int param_id = 0; param_id < layers_[layer_id]->blobs().size();
++param_id) {
if (!layers_[layer_id]->param_propagate_down(param_id)) { continue; }
const Blob<Dtype>& blob = *layers_[layer_id]->blobs()[param_id];
const Dtype diff_abs_val_mean = blob.asum_diff() / blob.count();
LOG_IF(INFO, Caffe::root_solver())
<< " [Backward] "
<< "Layer " << layer_names_[layer_id]
<< ", param blob " << param_id
<< " diff: " << diff_abs_val_mean;
}
}
template <typename Dtype>
void Net<Dtype>::UpdateDebugInfo(const int param_id) {
const Blob<Dtype>& blob = *params_[param_id];
const int param_owner = param_owners_[param_id];
const string& layer_name = layer_names_[param_layer_indices_[param_id].first];
const string& param_display_name = param_display_names_[param_id];
const Dtype diff_abs_val_mean = blob.asum_diff() / blob.count();
if (param_owner < 0) {
const Dtype data_abs_val_mean = blob.asum_data() / blob.count();
LOG_IF(INFO, Caffe::root_solver())
<< " [Update] Layer " << layer_name
<< ", param " << param_display_name
<< " data: " << data_abs_val_mean
<< "; diff: " << diff_abs_val_mean;
} else {
const string& owner_layer_name =
layer_names_[param_layer_indices_[param_owner].first];
LOG_IF(INFO, Caffe::root_solver())
<< " [Update] Layer " << layer_name
<< ", param blob " << param_display_name
<< " (owned by layer " << owner_layer_name << ", " << "param "
<< param_display_names_[param_owners_[param_id]] << ")"
<< " diff: " << diff_abs_val_mean;
}
}
template <typename Dtype>
void Net<Dtype>::ShareTrainedLayersWith(const Net* other) {
int num_source_layers = other->layers().size();
for (int i = 0; i < num_source_layers; ++i) {
Layer<Dtype>* source_layer = other->layers()[i].get();
const string& source_layer_name = other->layer_names()[i];
int target_layer_id = 0;
while (target_layer_id != layer_names_.size() &&
layer_names_[target_layer_id] != source_layer_name) {
++target_layer_id;
}
if (target_layer_id == layer_names_.size()) {
LOG(INFO) << "Ignoring source layer " << source_layer_name;
continue;
}
DLOG(INFO) << "Copying source layer " << source_layer_name;
vector<shared_ptr<Blob<Dtype> > >& target_blobs =
layers_[target_layer_id]->blobs();
CHECK_EQ(target_blobs.size(), source_layer->blobs().size())
<< "Incompatible number of blobs for layer " << source_layer_name;
for (int j = 0; j < target_blobs.size(); ++j) {
Blob<Dtype>* source_blob = source_layer->blobs()[j].get();
CHECK(target_blobs[j]->shape() == source_blob->shape())
<< "Cannot share param " << j << " weights from layer '"
<< source_layer_name << "'; shape mismatch. Source param shape is "
<< source_blob->shape_string() << "; target param shape is "
<< target_blobs[j]->shape_string();
target_blobs[j]->ShareData(*source_blob);
}
}
}
template <typename Dtype>
void Net<Dtype>::BackwardFrom(int start) {
BackwardFromTo(start, 0);
}
template <typename Dtype>
void Net<Dtype>::BackwardTo(int end) {
BackwardFromTo(layers_.size() - 1, end);
}
template <typename Dtype>
void Net<Dtype>::Backward() {
BackwardFromTo(layers_.size() - 1, 0);
if (debug_info_) {
Dtype asum_data = 0, asum_diff = 0, sumsq_data = 0, sumsq_diff = 0;
for (int i = 0; i < learnable_params_.size(); ++i) {
asum_data += learnable_params_[i]->asum_data();
asum_diff += learnable_params_[i]->asum_diff();
sumsq_data += learnable_params_[i]->sumsq_data();
sumsq_diff += learnable_params_[i]->sumsq_diff();
}
const Dtype l2norm_data = std::sqrt(sumsq_data);
const Dtype l2norm_diff = std::sqrt(sumsq_diff);
LOG(ERROR) << " [Backward] All net params (data, diff): "
<< "L1 norm = (" << asum_data << ", " << asum_diff << "); "
<< "L2 norm = (" << l2norm_data << ", " << l2norm_diff << ")";
}
}
template <typename Dtype>
void Net<Dtype>::Reshape() {
for (int i = 0; i < layers_.size(); ++i) {
layers_[i]->Reshape(bottom_vecs_[i], top_vecs_[i]);
}
}
template <typename Dtype>
void Net<Dtype>::CopyTrainedLayersFrom(const NetParameter& param) {
int num_source_layers = param.layer_size();
for (int i = 0; i < num_source_layers; ++i) {
const LayerParameter& source_layer = param.layer(i);
const string& source_layer_name = source_layer.name();
int target_layer_id = 0;
while (target_layer_id != layer_names_.size() &&
layer_names_[target_layer_id] != source_layer_name) {
++target_layer_id;
}
if (target_layer_id == layer_names_.size()) {
LOG(INFO) << "Ignoring source layer " << source_layer_name;
continue;
}
DLOG(INFO) << "Copying source layer " << source_layer_name;
vector<shared_ptr<Blob<Dtype> > >& target_blobs =
layers_[target_layer_id]->blobs();
CHECK_EQ(target_blobs.size(), source_layer.blobs_size())
<< "Incompatible number of blobs for layer " << source_layer_name;
for (int j = 0; j < target_blobs.size(); ++j) {
if (!target_blobs[j]->ShapeEquals(source_layer.blobs(j))) {
Blob<Dtype> source_blob;
const bool kReshape = true;
source_blob.FromProto(source_layer.blobs(j), kReshape);
LOG(FATAL) << "Cannot copy param " << j << " weights from layer '"
<< source_layer_name << "'; shape mismatch. Source param shape is "
<< source_blob.shape_string() << "; target param shape is "
<< target_blobs[j]->shape_string() << ". "
<< "To learn this layer's parameters from scratch rather than "
<< "copying from a saved net, rename the layer.";
}
const bool kReshape = false;
target_blobs[j]->FromProto(source_layer.blobs(j), kReshape);
}
}
}
template <typename Dtype>
void Net<Dtype>::CopyTrainedLayersFrom(const string trained_filename) {
if (H5Fis_hdf5(trained_filename.c_str())) {
CopyTrainedLayersFromHDF5(trained_filename);
} else {
CopyTrainedLayersFromBinaryProto(trained_filename);
}
}
template <typename Dtype>
void Net<Dtype>::CopyTrainedLayersFromBinaryProto(
const string trained_filename) {
NetParameter param;
ReadNetParamsFromBinaryFileOrDie(trained_filename, ¶m);
CopyTrainedLayersFrom(param);
}
template <typename Dtype>
void Net<Dtype>::CopyTrainedLayersFromHDF5(const string trained_filename) {
hid_t file_hid = H5Fopen(trained_filename.c_str(), H5F_ACC_RDONLY,
H5P_DEFAULT);
CHECK_GE(file_hid, 0) << "Couldn't open " << trained_filename;
hid_t data_hid = H5Gopen2(file_hid, "data", H5P_DEFAULT);
CHECK_GE(data_hid, 0) << "Error reading weights from " << trained_filename;
int num_layers = hdf5_get_num_links(data_hid);
for (int i = 0; i < num_layers; ++i) {
string source_layer_name = hdf5_get_name_by_idx(data_hid, i);
if (!layer_names_index_.count(source_layer_name)) {
LOG(INFO) << "Ignoring source layer " << source_layer_name;
continue;
}
int target_layer_id = layer_names_index_[source_layer_name];
DLOG(INFO) << "Copying source layer " << source_layer_name;
vector<shared_ptr<Blob<Dtype> > >& target_blobs =
layers_[target_layer_id]->blobs();
hid_t layer_hid = H5Gopen2(data_hid, source_layer_name.c_str(),
H5P_DEFAULT);
CHECK_GE(layer_hid, 0)
<< "Error reading weights from " << trained_filename;
// Check that source layer doesn't have more params than target layer
int num_source_params = hdf5_get_num_links(layer_hid);
CHECK_LE(num_source_params, target_blobs.size())
<< "Incompatible number of blobs for layer " << source_layer_name;
for (int j = 0; j < target_blobs.size(); ++j) {
ostringstream oss;
oss << j;
string dataset_name = oss.str();
int target_net_param_id = param_id_vecs_[target_layer_id][j];
if (!H5Lexists(layer_hid, dataset_name.c_str(), H5P_DEFAULT)) {
// Target param doesn't exist in source weights...
if (param_owners_[target_net_param_id] != -1) {
// ...but it's weight-shared in target, so that's fine.
continue;
} else {
LOG(FATAL) << "Incompatible number of blobs for layer "
<< source_layer_name;
}
}
hdf5_load_nd_dataset(layer_hid, dataset_name.c_str(), 0, kMaxBlobAxes,
target_blobs[j].get());
}
H5Gclose(layer_hid);
}
H5Gclose(data_hid);
H5Fclose(file_hid);
}
template <typename Dtype>
void Net<Dtype>::ToProto(NetParameter* param, bool write_diff) const {
param->Clear();
param->set_name(name_);
// Add bottom and top
DLOG(INFO) << "Serializing " << layers_.size() << " layers";
for (int i = 0; i < layers_.size(); ++i) {
LayerParameter* layer_param = param->add_layer();
layers_[i]->ToProto(layer_param, write_diff);
}
}
template <typename Dtype>
void Net<Dtype>::ToHDF5(const string& filename, bool write_diff) const {
hid_t file_hid = H5Fcreate(filename.c_str(), H5F_ACC_TRUNC, H5P_DEFAULT,
H5P_DEFAULT);
CHECK_GE(file_hid, 0)
<< "Couldn't open " << filename << " to save weights.";
hid_t data_hid = H5Gcreate2(file_hid, "data", H5P_DEFAULT, H5P_DEFAULT,
H5P_DEFAULT);
CHECK_GE(data_hid, 0) << "Error saving weights to " << filename << ".";
hid_t diff_hid = -1;
if (write_diff) {
diff_hid = H5Gcreate2(file_hid, "diff", H5P_DEFAULT, H5P_DEFAULT,
H5P_DEFAULT);
CHECK_GE(diff_hid, 0) << "Error saving weights to " << filename << ".";
}
for (int layer_id = 0; layer_id < layers_.size(); ++layer_id) {
const LayerParameter& layer_param = layers_[layer_id]->layer_param();
string layer_name = layer_param.name();
hid_t layer_data_hid = H5Gcreate2(data_hid, layer_name.c_str(),
H5P_DEFAULT, H5P_DEFAULT, H5P_DEFAULT);
CHECK_GE(layer_data_hid, 0)
<< "Error saving weights to " << filename << ".";
hid_t layer_diff_hid = -1;
if (write_diff) {
layer_diff_hid = H5Gcreate2(diff_hid, layer_name.c_str(),
H5P_DEFAULT, H5P_DEFAULT, H5P_DEFAULT);
CHECK_GE(layer_diff_hid, 0)
<< "Error saving weights to " << filename << ".";
}
int num_params = layers_[layer_id]->blobs().size();
for (int param_id = 0; param_id < num_params; ++param_id) {
ostringstream dataset_name;
dataset_name << param_id;
const int net_param_id = param_id_vecs_[layer_id][param_id];
if (param_owners_[net_param_id] == -1) {
// Only save params that own themselves
hdf5_save_nd_dataset<Dtype>(layer_data_hid, dataset_name.str(),
*params_[net_param_id]);
}
if (write_diff) {
// Write diffs regardless of weight-sharing
hdf5_save_nd_dataset<Dtype>(layer_diff_hid, dataset_name.str(),
*params_[net_param_id], true);
}
}
H5Gclose(layer_data_hid);
if (write_diff) {
H5Gclose(layer_diff_hid);
}
}
H5Gclose(data_hid);
if (write_diff) {
H5Gclose(diff_hid);
}
H5Fclose(file_hid);
}
template <typename Dtype>
void Net<Dtype>::Update() {
for (int i = 0; i < learnable_params_.size(); ++i) {
learnable_params_[i]->Update();
}
}
template <typename Dtype>
void Net<Dtype>::ClearParamDiffs() {
for (int i = 0; i < learnable_params_.size(); ++i) {
Blob<Dtype>* blob = learnable_params_[i];
switch (Caffe::mode()) {
case Caffe::CPU:
caffe_set(blob->count(), static_cast<Dtype>(0),
blob->mutable_cpu_diff());
break;
case Caffe::GPU:
#ifndef CPU_ONLY
caffe_gpu_set(blob->count(), static_cast<Dtype>(0),
blob->mutable_gpu_diff());
#else
NO_GPU;
#endif
break;
}
}
}
template <typename Dtype>
void Net<Dtype>::ShareWeights() {
for (int i = 0; i < params_.size(); ++i) {
if (param_owners_[i] < 0) { continue; }
params_[i]->ShareData(*params_[param_owners_[i]]);
params_[i]->ShareDiff(*params_[param_owners_[i]]);
}
}
template <typename Dtype>
bool Net<Dtype>::has_blob(const string& blob_name) const {
return blob_names_index_.find(blob_name) != blob_names_index_.end();
}
template <typename Dtype>
const shared_ptr<Blob<Dtype> > Net<Dtype>::blob_by_name(
const string& blob_name) const {
shared_ptr<Blob<Dtype> > blob_ptr;
if (has_blob(blob_name)) {
blob_ptr = blobs_[blob_names_index_.find(blob_name)->second];
} else {
blob_ptr.reset((Blob<Dtype>*)(NULL));
LOG(WARNING) << "Unknown blob name " << blob_name;
}
return blob_ptr;
}
template <typename Dtype>
bool Net<Dtype>::has_layer(const string& layer_name) const {
return layer_names_index_.find(layer_name) != layer_names_index_.end();
}
template <typename Dtype>
const shared_ptr<Layer<Dtype> > Net<Dtype>::layer_by_name(
const string& layer_name) const {
shared_ptr<Layer<Dtype> > layer_ptr;
if (has_layer(layer_name)) {
layer_ptr = layers_[layer_names_index_.find(layer_name)->second];
} else {
layer_ptr.reset((Layer<Dtype>*)(NULL));
LOG(WARNING) << "Unknown layer name " << layer_name;
}
return layer_ptr;
}
INSTANTIATE_CLASS(Net);
} // namespace caffe