Caffe解讀（三）

Caffe學習（三）

Caffe 梳理

• layer.hpp中：Forward和Backward對應前向計算和反向更新，輸入統一都是bottom，輸出爲top，其中Backward裏面有個propagate_down參數，用來表示該Layer是否反向傳播梯度。在Forward和Backward的具體實現裏，會根據Caffe::mode()進行對應的操作，即使用cpu或者gpu進行計算，兩個都實現了對應的接口Forward_cpu、Forward_gpu和Backward_cpu、Backward_gpu，這些接口都是virtual，具體還是要根據layer的類型進行對應的計算（注意：有些layer並沒有GPU計算的實現，所以封裝時加入了CPU的計算作爲後備）。另外，還實現了ToProto的接口，將Layer的參數寫入到protocol buffer文件中。
• 在梳理Net.cpp代碼後，我們將詳細說明propagate_down的具體使用方法，以及在finetune中決定是否反傳的各個參數的區別

Net-Init

Net.cpp-Init函數

• 代碼梳理

#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, &param);
  // Set phase, stages and level
  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);
}

template <typename Dtype>
void Net<Dtype>::Init(const NetParameter& in_param) {
  // Set phase from the state.
  phase_ = in_param.state().phase();
  // Filter layers based on their include/exclude rules and
  // the current NetState.
  // FilterNet
  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.
  // 第二次處理Net 的param，作用：對於底層一個輸出blob對應多個上層的情況，要再加入Split 層，
  // 形成新的網絡，這麼做的主要原因是多個層反傳給該blob的梯度需要累加
  // 如Label需傳入accuracy 層和loss 層，就會在這裏插入一層，也就是caffe 的log 中打印出來的
  // 一個新的層label_mnist_1_split 層，爲該層創建兩個top blob 分別爲：
  // Label_mnist_1_split_0 和Label_mnist_1_split_1 分別傳入accuracy 層和loss
  NetParameter param;
  InsertSplits(filtered_param, &param);
  // Basically, build all the layers and set up their connections.
  // 下面是層及層間blob的創建
  name_ = param.name();
  map<string, int> blob_name_to_idx;
  set<string> available_blobs;
  memory_used_ = 0;
  // For each layer, set up its input and output
  bottom_vecs_.resize(param.layer_size());// 存每一層的輸入(bottom)blob指針
  top_vecs_.resize(param.layer_size());// 存每一層輸出(top)的blob指針  
  bottom_id_vecs_.resize(param.layer_size());// 存每一層輸入(bottom)blob的id  
  param_id_vecs_.resize(param.layer_size());// 存每一層參數blob的id
  top_id_vecs_.resize(param.layer_size());// 存每一層輸出(top)的blob的id  
  bottom_need_backward_.resize(param.layer_size());//該blob是需要返回的bool值 

  //for循環對每一層處理 
  for (int layer_id = 0; layer_id < param.layer_size(); ++layer_id) {
    // Inherit phase from net if unset.
    // 如果當前層沒有設置phase，則將當前層phase設置爲網絡net的phase
    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);//當前層的參數
    // 檢查LayerParameter 類型propagate_down 成員的個數是否達標
    // 也就是我們在prototxt 中設置的“propagate_down: ××”，他的個數要麼是0（也就是不寫）
    // 要麼等於bottom blob 的個數，代表當前layer的梯度是否反向傳播（後續有詳細解釋）
    // indicating whether to compute the diff of each param blob. 
    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 ";
    }
    // 創建一個具體的層，並壓入到layers_ 中
    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();
    bool need_backward = false;

    // Figure out this layer's input and output
    // 分別處理輸入（bottom blob）和輸出（top blob）
    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);
      // AppendBottom() 函數爲該層創建bottom blob
      // If a blob needs backward, this layer should provide it.
      need_backward |= blob_need_backward_[blob_id];
    }
    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);
      // 創建top blob壓入top_vecs_
      // Collect Input layer tops as Net inputs.
      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();
    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.
    // Set_Up中爲AppendTop() 中創建的Blob 分配內存空間
    layers_[layer_id]->SetUp(bottom_vecs_[layer_id], top_vecs_[layer_id]);
    LOG_IF(INFO, Caffe::root_solver())
        << "Setting up " << layer_names_[layer_id];

    // 每次循環都會更新
    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_,每次遍歷一個layer的時候，都會resize blob_loss_weights_,   
      //然後調用模板類layer 的loss 函數返回loss_weight
      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);
      }
      memory_used_ += top_vecs_[layer_id][top_id]->count();// 計算所需內存
    }
    LOG_IF(INFO, Caffe::root_solver())
        << "Memory required for data: " << memory_used_ * sizeof(Dtype);

    // 對每層的param blob的處理
    const int param_size = layer_param.param_size();
    // param_size 是Layermeter 類型對象layer_param 中ParamSpec param 成員的個數,
    // 是層內blob_ 的數量，即該層有幾個權重參數（每個blob內有一個參數）
    // 例如；cov層和IP層都有兩個參數對應w和b
    const int num_param_blobs = layers_[layer_id]->blobs().size();
    // num_param_blobs是一個Layer中learnable parameter blob的個數，
    // param_size <= num_param_blobs   
    CHECK_LE(param_size, num_param_blobs)
        << "Too many params specified for layer " << layer_param.name();
    ParamSpec default_param_spec;
    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;
      // 這裏說明了如果在prototxt 中將lr置爲0，即關掉，該層參數便不再更新
      need_backward |= param_need_backward;
      // 由param_need_backward 來決定need_backward 是否爲真，  
      // 並且，只要有一次遍歷使得need_backward 爲真，則這個for循環結束後，need_backward 也爲真 
      layers_[layer_id]->set_param_propagate_down(param_id,
                                                  param_need_backward);
    }
    // 添加parameter blob,如果當前layer沒有parameter blob(num_param_blobs==0), 
    // 比如ReLU，那麼就不進入循環，不添加parameter blob     
    // AppendParam 只是執行爲當前layer 添加parameter blob 的相關工作， 
    // 並不會修改與backward的相關屬性  
    for (int param_id = 0; param_id < num_param_blobs; ++param_id) {
      AppendParam(param, layer_id, param_id);
      // 將param blob以及blob的id添加到params_,param_id_vecs_等 
    }
    // Finally, set the backward flag
    // 之前在AppendTop 函數中，在遍歷當前層的每一個top blob的時候 
    // 都已將一個false（默認值）壓入向量blob_need_backward_中 
    // 下面如果這個layer need backward，則會更新blob_need_backward_
    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;
      }
    }
  }
// 至此，各個層都已被創建並啓動，下面部分是按照反向順序修正backward設置？

  // 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

  // 需要注意的是，上述代碼中關於backward設置的部分，是按照前向的順序設置的，
  // 而下面的代碼是按反向順序修正前向設置的結果。
  // 一個layer是否需要backward computation，主要依據兩個方面：
  // (1)該layer的top blob 是否參與loss的計算；
  // (2)該layer的bottom blob 是否需要backward computation，
  //    比如Data層一般就不需要backward computation
  set<string> blobs_under_loss;
  set<string> blobs_skip_backp;
  // 反向
  for (int layer_id = layers_.size() - 1; layer_id >= 0; --layer_id) {
    bool layer_contributes_loss = false;// 這就是上面所說的標誌layer是否參與loss計算
    bool layer_skip_propagate_down = true;//標誌layer是否需要 “跳過backward compute”
    for (int top_id = 0; top_id < top_vecs_[layer_id].size(); ++top_id) {
      const string& blob_name = blob_names_[top_id_vecs_[layer_id][top_id]];
      if (layers_[layer_id]->loss(top_id) ||
          (blobs_under_loss.find(blob_name) != blobs_under_loss.end())) {
        layer_contributes_loss = true;
      }
      if (blobs_skip_backp.find(blob_name) == blobs_skip_backp.end()) {
        layer_skip_propagate_down = false;
      }
      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
    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;
        // 所含項標誌整個網絡所有網絡層的bottom blob 是否需要backward
      }
    }
    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.";
      }
    }
    // 修正前向設置的結果
    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);// 添加到blobs_under_loss
      } 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);// 添加到blobs_skip_backp
      }
    }
  }
  // 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.
  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) {
  // 作用：把模型參數文件中不符合當前階段規則的層去掉
  // 如Test階段只用網絡的前向，需要將設置爲phase：Train的層去掉
  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();
    // include 和 exclude 分別代表什麼？
    // Rules controlling whether and when a layer is included in the network,  
    // based on the current NetState.  You may specify a non-zero number of rules  
    // to include OR exclude, but not both.  If no include or exclude rules are  
    // specified, the layer is always included.  If the current NetState meets  
    // ANY (i.e., one or more) of the specified rules, the layer is  
    // included/excluded.  
    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);// 默認true
    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;// 如果不包含（exclude），只要meet一個exclude rules就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;// 如果包含（include），只要meet一個include rules就true
      }
    }
    if (layer_included) { //如果是true（包含），就添加進去
      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.
// 此函數爲該層創建top blob，該函數真正的new的一個blob的對象
// 並將top blob 的指針壓入到top_vecs_中
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)));
  const string& blob_name = (layer_param->top_size() > top_id) ?
      layer_param->top(top_id) : "(automatic)";
  // Check if we are doing in-place computation
  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.
    if (Caffe::root_solver()) {
      LOG(INFO) << layer_param->name() << " -> " << blob_name;
    }
    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.
// AppendBottom()函數爲該層創建bottom blob，由於網絡是堆疊而成，
// 即：當前層的輸入bottom blob是前一層的輸出top blob，因此此函數並沒沒有真正的創建blob，
// 只是在將前一層的指針壓入到了bottom_vecs_中
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;
  }
}

// 功能：從Other網絡複製某些層 
// 步驟：對Other網絡的第i層（源層）： 
// 1. 定義一個Layer的指針指向第i層 
// 2. 讀取第i層（源層）的名字 
// 3. 找通過名字來找目標層如果沒找到，即target_layer_id == layer_names_.size()則忽略Other的第i層，
//    即Other的第i層不需要share給網絡 
// 4. 如果找到了，即other的第i層需要share給網絡，則把目標層的所有blob讀到target_blobs中 
//    1判斷目標層和源層的blob數量是否相等 
//    2判斷每個blob大小是否相等 
//    3調用ShareData函數把源層的blob賦給目標層的blob 
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);
}

// 對整個網絡進行反向傳播
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]);
  }
}

// 功能：和ShareTrainedLayersWith一樣 
// 步驟：不同的是調用FromProto函數把源層的blob賦給目標層的blob
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 (trained_filename.size() >= 3 &&
      trained_filename.compare(trained_filename.size() - 3, 3, ".h5") == 0) {
    CopyTrainedLayersFromHDF5(trained_filename);
  } else {
    CopyTrainedLayersFromBinaryProto(trained_filename);
  }
}

template <typename Dtype>
void Net<Dtype>::CopyTrainedLayersFromBinaryProto(
    const string trained_filename) {
  NetParameter param;
  ReadNetParamsFromBinaryFileOrDie(trained_filename, &param);
  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);
}

// 功能：把網絡的參數存入prototxt中 ？

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);
}

// 更新params_中blob的值
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

propagate_down具體使用方法及決定是否反傳的各個參數的區別

假設我們有4個卷積層A->B->C->D

propagate_down

• 我們希望C層的參數不改變，以及C前面的A、B層的參數也不改變，這種情況也就是D層的梯度不往前反向傳播到D層的輸入blob（也就是C層的輸出blob 沒有得到梯度），我們可以通過設置D層的propagate_down爲false來做到 propagate_down的數量與輸入blob的數量相同，假如你某個層有2個輸入blob，那麼你應該在該layer的Param裏面寫上兩行：
  propagate_down : 0 # 第1個輸入blob不會得到反向傳播的梯度 propagate_down : 0 # 第2個輸入blob不會得到反向傳播的梯度
  這樣，你這個layer的梯度就不會反向傳播，前面的所有layer的參數也就不會改變了

• propagate_down的設置使得前面的所有層都不能得到反傳梯度，即前面所有層的參數都不更新

• 具體使用方法

 layer{
         name: "conv2_3x3"
         type: "ConvolutionData"
         bottom: "pool1_3x3"
         top: "conv2_3x3"
         param{
                 lr_mult: 1
                 decay_mult: 1
         }
         propagate_down: 0
         convolution_param {
                 num_output: 64
                 bias_term: false
                 pad: 1
                 kernel_size: 3
                 stride: 1
                 weight_filler {
                         type: "xavier"
                 }       
         }
 }

lr_mult

• 我們希望C層的參數不會改變，但是C前面的A、B層的參數需要改變，這種情況，只是固定了C層的參數，C層得到的梯度依然會反向傳播給前面的B層。只需要將對應的參數blob的學習率調整爲0：在layer裏面加上 “ param { lr_mult: 0 }” 就可以了
• lr_mult的設置只是使得該層的學習率爲零該層參數不更新，梯度依然會反向傳播回去
• 具體使用方法

 layer{
    name: "ip1"
    type: "InnerProduct"
    bottom: "pool3"
    top: "ip1"
    param{ # 對應第1個參數blob的配置，也就是全連接層的參數矩陣的配置
        lr_mult: 0 # 學習率爲0
        decay_mult: 0
    }
    param{ # 對應第2個參數blob的配置，也就是全連接層的偏置項的配置
        lr_mult: 0 # 學習率爲0
        decay_mult: 0
    }
    inner_product_param {
        num_output: 2
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

參考
https://www.zhihu.com/question/35754716
http://blog.csdn.net/langb2014/article/details/50987593
http://blog.csdn.net/qq_16055159/article/details/45057297