caffe网络在多线程中无法使用GPU的解决方案-cpp-caffe-net-run-in-multiple-threads

37次阅读

共计 3594 个字符,预计需要花费 9 分钟才能阅读完成。

本文首发于个人博客 https://kezunlin.me/post/8d877e63/,欢迎阅读!

cpp caffe net run in multiple threads

Guide

set_mode

Caffe fails to use GPU in a new thread???
see here

    
the `Caffe::mode_` variable that controls this is thread-local,
so ensure you’re calling `caffe.set_mode_gpu()` in each thread
before running any Caffe functions. That should solve your issue.

Caffe set_mode GPU 在多线程下失效
在 main thread 中设置 GPU 模式,在 worker thread 中调用网络进行检测,GPU 模式不起效,默认仍然使用 CPU 模式,所以速度很慢,和 GPU 相比慢了
10 倍左右。解决方案:在子线程中 set_mode, 然后调用网络进行检测。(1) 创建网络在 main thread。static 网络存储在全局静态数据区。worker thread 可以直接使用。(2) 在 worker thread 中检测,需要在子线程中 set_mode, 然后调用网络进行检测。结论:(1)caffe 的 set_mode 所在的线程必须和使用 nets 进行 forward 的线程相同。否则默认使用 CPU 模式,速度会很慢。(2)caffe 的 nets 初始化可以在 main thread 也可以在 worker thread。

code example

#include <iostream>
#include <string>
#include <thread>

#include <gtest/gtest.h>
#include <glog/logging.h>

#include <boost/date_time/posix_time/posix_time.hpp>  

// opencv
#include <opencv2/core.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/imgproc.hpp>

using namespace std;

#include "algorithm/algorithm.h"
using namespace kezunlin::algorithm;

#pragma region net-demo

void topwire_demo(bool run_in_worker_thread)
{if (run_in_worker_thread) {CaffeApi::set_mode(true, 0, 1234);// set in worker thread-1, use GPU-0
    }
    
    // do net detect 
    // ...
}

void railway_demo(bool run_in_worker_thread)
{if (run_in_worker_thread) {CaffeApi::set_mode(true, 0, 1234);// set in worker thread-1, use GPU-0
    }
    
    // do net detect 
    // ...
}

void sidewall_demo(bool run_in_worker_thread)
{if (run_in_worker_thread) {CaffeApi::set_mode(true, 0, 1234);// set in worker thread-1, use GPU-0
    }

    // do net detect 
    // ...
}

void lockcatch_demo(bool run_in_worker_thread)
{if (run_in_worker_thread) {CaffeApi::set_mode(true, 0, 1234);// set in worker thread-1, use GPU-0
    }

    // do net detect 
    // ...
}

#pragma endregion


#pragma region worker-thread-demo

void worker_thread_topwire_demo(bool run_in_worker_thread)
{std::thread thr(topwire_demo, run_in_worker_thread);
    thr.join();}

void worker_thread_railway_demo(bool run_in_worker_thread)
{std::thread thr(railway_demo, run_in_worker_thread);
    thr.join();}

void worker_thread_sidewall_demo(bool run_in_worker_thread)
{std::thread thr(sidewall_demo, run_in_worker_thread);
    thr.join();}

void worker_thread_lockcatch_demo(bool run_in_worker_thread)
{std::thread thr(lockcatch_demo, run_in_worker_thread);
    thr.join();}

#pragma endregion

enum DETECT_TYPE {
    SET_IN_MAIN_DETECT_IN_MAIN, // 主线程 set_mode,主线程检测,40ms 左右,使用 GPU
    SET_IN_WORKER_DETECT_IN_WORKER, // 子线程 set_mode,子线程检测,40ms 左右,使用 GPU
    SET_IN_MAIN_DETECT_IN_WORKER // 主线程 set_mode,子线程检测,400ms 左右,慢了 10 倍左右,没有使用 GPU
};

void thread_demo()
{
    DETECT_TYPE detect_type = SET_IN_MAIN_DETECT_IN_MAIN;
    detect_type = SET_IN_WORKER_DETECT_IN_WORKER;
    detect_type = SET_IN_MAIN_DETECT_IN_WORKER;

    init_algorithm_api();

    switch (detect_type)
    {
    case SET_IN_MAIN_DETECT_IN_MAIN:
        topwire_demo(false);
        railway_demo(false);
        sidewall_demo(false);
        lockcatch_demo(false);
        break;
    case SET_IN_WORKER_DETECT_IN_WORKER:
        worker_thread_topwire_demo(true);
        worker_thread_railway_demo(true);
        worker_thread_sidewall_demo(true);
        worker_thread_lockcatch_demo(true);
        break;
    case SET_IN_MAIN_DETECT_IN_WORKER:
        worker_thread_topwire_demo(false);
        worker_thread_railway_demo(false);
        worker_thread_sidewall_demo(false);
        worker_thread_lockcatch_demo(false);
        break;
    default:
        break;
    }

    free_algorithm_api();}

void test_algorithm_api()
{thread_demo();
}

TEST(algorithn_test, test_algorithm_api) {test_algorithm_api();
}
  • SET_IN_MAIN_DETECT_IN_MAIN, // 主线程 set_mode,主线程检测,40ms 左右,使用 GPU
  • SET_IN_WORKER_DETECT_IN_WORKER, // 子线程 set_mode,子线程检测,40ms 左右,使用 GPU
  • SET_IN_MAIN_DETECT_IN_WORKER // 主线程 set_mode,子线程检测,400ms 左右,慢了 10 倍左右,没有使用 GPU

Reference

  • Caffe fails to use GPU in a new thread

History

  • 20180712: created.

Copyright

  • Post author: kezunlin
  • Post link: https://kezunlin.me/post/8d877e63/
  • Copyright Notice: All articles in this blog are licensed under CC BY-NC-SA 3.0 unless stating additionally.

正文完
 0