【caffe源码研究】第二章：使用篇(3) : C++接口

在caffe提供的样例里有一个C++接口的范例，整理一下如下。
首先是预测的 Classifier 类，定义构造函数Classifier和预测函数Classify。

#include <caffe/caffe.hpp>
#ifdef USE_OPENCV
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#endif  // USE_OPENCV
#include <algorithm>
#include <iosfwd>
#include <memory>
#include <string>
#include <utility>
#include <vector>

#ifdef USE_OPENCV
using namespace caffe;  // NOLINT(build/namespaces)
using std::string;

/* Pair (label, confidence) representing a prediction. */
typedef std::pair<string, float> Prediction;

class Classifier {
 public:
  Classifier(const string& model_file,
             const string& trained_file,
             const string& mean_file,
             const string& label_file);

  std::vector<Prediction> Classify(const cv::Mat& img, int N = 5);

 private:
  void SetMean(const string& mean_file);

  std::vector<float> Predict(const cv::Mat& img);

  void WrapInputLayer(std::vector<cv::Mat>* input_channels);

  void Preprocess(const cv::Mat& img,
                  std::vector<cv::Mat>* input_channels);

 private:
  shared_ptr<Net<float> > net_;
  cv::Size input_geometry_;
  int num_channels_;
  cv::Mat mean_;
  std::vector<string> labels_;
};

定义了一个Pair (label, confidence)来表示预测结果，标签+可信度。

类成员函数的具体实现：

Classifier::Classifier(const string& model_file,
                       const string& trained_file,
                       const string& mean_file,
                       const string& label_file) {
#ifdef CPU_ONLY
  Caffe::set_mode(Caffe::CPU);
#else
  Caffe::set_mode(Caffe::GPU);
#endif

  /* Load the network. */
  net_.reset(new Net<float>(model_file, TEST));
  net_->CopyTrainedLayersFrom(trained_file);

  CHECK_EQ(net_->num_inputs(), 1) << "Network should have exactly one input.";
  CHECK_EQ(net_->num_outputs(), 1) << "Network should have exactly one output.";

  Blob<float>* input_layer = net_->input_blobs()[0];
  num_channels_ = input_layer->channels();
  CHECK(num_channels_ == 3 || num_channels_ == 1)
    << "Input layer should have 1 or 3 channels.";
  input_geometry_ = cv::Size(input_layer->width(), input_layer->height());

  /* Load the binaryproto mean file. */
  SetMean(mean_file);

  /* Load labels. */
  std::ifstream labels(label_file.c_str());
  CHECK(labels) << "Unable to open labels file " << label_file;
  string line;
  while (std::getline(labels, line))
    labels_.push_back(string(line));

  Blob<float>* output_layer = net_->output_blobs()[0];
  CHECK_EQ(labels_.size(), output_layer->channels())
    << "Number of labels is different from the output layer dimension.";
}

static bool PairCompare(const std::pair<float, int>& lhs,
                        const std::pair<float, int>& rhs) {
  return lhs.first > rhs.first;
}

/* Return the indices of the top N values of vector v. */
static std::vector<int> Argmax(const std::vector<float>& v, int N) {
  std::vector<std::pair<float, int> > pairs;
  for (size_t i = 0; i < v.size(); ++i)
    pairs.push_back(std::make_pair(v[i], static_cast<int>(i)));
  std::partial_sort(pairs.begin(), pairs.begin() + N, pairs.end(), PairCompare);

  std::vector<int> result;
  for (int i = 0; i < N; ++i)
    result.push_back(pairs[i].second);
  return result;
}

/* Return the top N predictions. */
std::vector<Prediction> Classifier::Classify(const cv::Mat& img, int N) {
  std::vector<float> output = Predict(img);

  N = std::min<int>(labels_.size(), N);
  std::vector<int> maxN = Argmax(output, N);
  std::vector<Prediction> predictions;
  for (int i = 0; i < N; ++i) {
    int idx = maxN[i];
    predictions.push_back(std::make_pair(labels_[idx], output[idx]));
  }

  return predictions;
}

/* Load the mean file in binaryproto format. */
void Classifier::SetMean(const string& mean_file) {
  BlobProto blob_proto;
  ReadProtoFromBinaryFileOrDie(mean_file.c_str(), &blob_proto);

  /* Convert from BlobProto to Blob<float> */
  Blob<float> mean_blob;
  mean_blob.FromProto(blob_proto);
  CHECK_EQ(mean_blob.channels(), num_channels_)
    << "Number of channels of mean file doesn't match input layer.";

  /* The format of the mean file is planar 32-bit float BGR or grayscale. */
  std::vector<cv::Mat> channels;
  float* data = mean_blob.mutable_cpu_data();
  for (int i = 0; i < num_channels_; ++i) {
    /* Extract an individual channel. */
    cv::Mat channel(mean_blob.height(), mean_blob.width(), CV_32FC1, data);
    channels.push_back(channel);
    data += mean_blob.height() * mean_blob.width();
  }

  /* Merge the separate channels into a single image. */
  cv::Mat mean;
  cv::merge(channels, mean);

  /* Compute the global mean pixel value and create a mean image
   * filled with this value. */
  cv::Scalar channel_mean = cv::mean(mean);
  mean_ = cv::Mat(input_geometry_, mean.type(), channel_mean);
}

std::vector<float> Classifier::Predict(const cv::Mat& img) {
  Blob<float>* input_layer = net_->input_blobs()[0];
  input_layer->Reshape(1, num_channels_,
                       input_geometry_.height, input_geometry_.width);
  /* Forward dimension change to all layers. */
  net_->Reshape();

  std::vector<cv::Mat> input_channels;
  WrapInputLayer(&input_channels);

  Preprocess(img, &input_channels);

  net_->Forward();

  /* Copy the output layer to a std::vector */
  Blob<float>* output_layer = net_->output_blobs()[0];
  const float* begin = output_layer->cpu_data();
  const float* end = begin + output_layer->channels();
  return std::vector<float>(begin, end);
}

/* Wrap the input layer of the network in separate cv::Mat objects
 * (one per channel). This way we save one memcpy operation and we
 * don't need to rely on cudaMemcpy2D. The last preprocessing
 * operation will write the separate channels directly to the input
 * layer. */
void Classifier::WrapInputLayer(std::vector<cv::Mat>* input_channels) {
  Blob<float>* input_layer = net_->input_blobs()[0];

  int width = input_layer->width();
  int height = input_layer->height();
  float* input_data = input_layer->mutable_cpu_data();
  for (int i = 0; i < input_layer->channels(); ++i) {
    cv::Mat channel(height, width, CV_32FC1, input_data);
    input_channels->push_back(channel);
    input_data += width * height;
  }
}

void Classifier::Preprocess(const cv::Mat& img,
                            std::vector<cv::Mat>* input_channels) {
  /* Convert the input image to the input image format of the network. */
  cv::Mat sample;
  if (img.channels() == 3 && num_channels_ == 1)
    cv::cvtColor(img, sample, cv::COLOR_BGR2GRAY);
  else if (img.channels() == 4 && num_channels_ == 1)
    cv::cvtColor(img, sample, cv::COLOR_BGRA2GRAY);
  else if (img.channels() == 4 && num_channels_ == 3)
    cv::cvtColor(img, sample, cv::COLOR_BGRA2BGR);
  else if (img.channels() == 1 && num_channels_ == 3)
    cv::cvtColor(img, sample, cv::COLOR_GRAY2BGR);
  else
    sample = img;

  cv::Mat sample_resized;
  if (sample.size() != input_geometry_)
    cv::resize(sample, sample_resized, input_geometry_);
  else
    sample_resized = sample;

  cv::Mat sample_float;
  if (num_channels_ == 3)
    sample_resized.convertTo(sample_float, CV_32FC3);
  else
    sample_resized.convertTo(sample_float, CV_32FC1);

  cv::Mat sample_normalized;
  cv::subtract(sample_float, mean_, sample_normalized);

  /* This operation will write the separate BGR planes directly to the
   * input layer of the network because it is wrapped by the cv::Mat
   * objects in input_channels. */
  cv::split(sample_normalized, *input_channels);

  CHECK(reinterpret_cast<float*>(input_channels->at(0).data)
        == net_->input_blobs()[0]->cpu_data())
    << "Input channels are not wrapping the input layer of the network.";
}

调用的主函数

int main(int argc, char** argv) {
  if (argc != 6) {
    std::cerr << "Usage: " << argv[0]
              << " deploy.prototxt network.caffemodel"
              << " mean.binaryproto labels.txt img.jpg" << std::endl;
    return 1;
  }

  ::google::InitGoogleLogging(argv[0]);

  string model_file   = argv[1];
  string trained_file = argv[2];
  string mean_file    = argv[3];
  string label_file   = argv[4];
  Classifier classifier(model_file, trained_file, mean_file, label_file);

  string file = argv[5];

  std::cout << "---------- Prediction for "
            << file << " ----------" << std::endl;

  cv::Mat img = cv::imread(file, -1);
  CHECK(!img.empty()) << "Unable to decode image " << file;
  std::vector<Prediction> predictions = classifier.Classify(img);

  /* Print the top N predictions. */
  for (size_t i = 0; i < predictions.size(); ++i) {
    Prediction p = predictions[i];
    std::cout << std::fixed << std::setprecision(4) << p.second << " - \""
              << p.first << "\"" << std::endl;
  }
}

从主函数开始分析，可以看出使用的方法是

./build/classification deploy.prototxt network.caffemodel mean.binaryproto labels.txt img.jpg

其中

• deploy.prototxt：预测的网络结构
• network.caffemodel：模型文件
• mean.binaryproto：均值文件
• labels.txt：标签文件
• img.jpg：图像路径

其中labels.txt就是把应该具有的标签列举出来，举个例子，如果是数字识别，一共一个类，则labels.txt如下

0
1
2
3
4
5
6
7
8
9

主函数比较简单，构造了一个classifier类，调用Classify函数，返回一个vector<Prediction>类型的predictions。predictions中的每一项是一个pairs类型，第一个数据是类别数，第二个是可信度。

比较核心的代码是

net_.reset(new Net<float>(model_file, TEST));
net_->CopyTrainedLayersFrom(trained_file);

net_->Forward();

/* Copy the output layer to a std::vector */
Blob<float>* output_layer = net_->output_blobs()[0];

暂时不分析其他具体代码。看一下如何在自己的C++具体项目里调用caffe。
第一种方法是直接在这个代码基础上改成自己的内容，然后重新编译caffe。但这种方式太笨拙。第二种方式是调用生成的libcaffe.so库。

写好自己的代码。例如predict.cpp之后，写一个Cmake文件 CMakeLists.txt

cmake_minimum_required (VERSION 2.8)

project (predict)

add_executable(predict predict.cpp)
include_directories ( 
        /home/nmtest/fangjin/caffe/include
        /home/work/cuda-8.0/include/
        /usr/local/include
        /home/nmtest/fangjin/opt/include/
        ~/.jumbo/include/
        ~/.jumbo/opt/gcc46/include/
        )

target_link_libraries(predict
        /home/nmtest/fangjin/caffe/build/lib/libcaffe.so
        ~/.jumbo/lib/libopencv_highgui.so
        ~/.jumbo/lib/libopencv_core.so
        ~/.jumbo/lib/libopencv_imgproc.so
        ~/.jumbo/lib/libglog.so
        ~/.jumbo/lib/libgflags.so
        /home/nmtest/fangjin/opt/lib/libboost_system.so)

新建一个build文件夹，将文件编译到build下。进入build文件夹，对上一级目录生成Makefile文件

cmake ..

结果如下

[fangjin@scom build]$ cmake ..
-- The C compiler identification is GNU 4.6.3
-- The CXX compiler identification is GNU 4.6.3
-- Check for working C compiler: /home/nmtest/.jumbo/opt/gcc46/bin/cc
-- Check for working C compiler: /home/nmtest/.jumbo/opt/gcc46/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /home/nmtest/.jumbo/opt/gcc46/bin/c++
-- Check for working CXX compiler: /home/nmtest/.jumbo/opt/gcc46/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Configuring done
-- Generating done
-- Build files have been written to: /home/nmtest/fangjin/hackthson/predict/build

生成文件列表如下

-rw-rw-r-- 1 nmtest nmtest 12033 Dec 23 10:39 CMakeCache.txt
drwxrwxr-x 5 nmtest nmtest  4096 Dec 23 10:39 CMakeFiles
-rw-rw-r-- 1 nmtest nmtest  1393 Dec 23 10:39 cmake_install.cmake
-rw-rw-r-- 1 nmtest nmtest  4910 Dec 23 10:39 Makefile

编译

[fangjin@com build]$ make
/usr/bin/make64 MAC=64
Scanning dependencies of target predict
[100%] Building CXX object CMakeFiles/predict.dir/predict.cpp.o
Linking CXX executable predict
[100%] Built target predict

提示[100%] Built target predict则生成成功。

写预测脚本

#!/usr/bin/env sh
set -e
APP=/home/users/fangjin/caffe/examples/cpp_classification/build/classification 

${APP} lenet_deploy.prototxt lenet_iter_10000.caffemodel image_mean.binaryproto labels.txt testData/9/9-7-ZZSHAHGX00.jpg

执行结果

[fangjin@com number_data]$ sh cpp_calssify.sh 
---------- Prediction for testData/9/9-7-ZZSHAHGX00.jpg ----------
1.0000 - "9"
0.0000 - "1"
0.0000 - "3"
0.0000 - "4"
0.0000 - "0"

分类结果9，正确。