tensorflow c++进行在线预估卡死hang住

最新推荐文章于 2023-04-28 14:30:07 发布
最新推荐文章于 2023-04-28 14:30:07 发布
阅读量2k 收藏
点赞数 1
分类专栏： tensorflow 文章标签： tensorflow c++在线预估
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/yupei881027/article/details/86501663
版权
一、 tensorflow 离线训练导出格式
  1. Tensorflow 离线训练的结果可以使用某种格式导出，如 SessionBundle、SavedModel、FreezedGraph，本次在使用SavedModel方式保持模型，进行在线预估。
二、在线预估代码
  1. 模型加载逻辑
bool TensorModel::LoadModel(const std::string& model_path) {
  tensorflow::SessionOptions session_options;
  tensorflow::RunOptions run_options;
  std::unordered_set<std::string> saved_model_tags;
  saved_model_tags.insert(tensorflow::kSavedModelTagServe);
  tensorflow::Status status = tensorflow::LoadSavedModel(session_options, run_options, model_path,
      saved_model_tags, &model_bundle_);
  if (!status.ok()) {         
    LOG(ERROR) << "LoadSavedModel Failed: " << status.ToString();
    return false;
  }
  return true;
}
  1. 在线预估逻辑，发现在预估的时候此处出现了hang住的请求，没有反应
tensorflow::Status status = model_bundle_.session->Run(tf_feature.input, {out_tname}, {}, &outputs);
  1. 具体堆栈
Thread 14 (Thread 0x7f738b7ce700 (LWP 7641)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 13 (Thread 0x7f738afcd700 (LWP 7642)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 12 (Thread 0x7f738a7cc700 (LWP 7643)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 11 (Thread 0x7f7389fcb700 (LWP 7644)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 10 (Thread 0x7f73897ca700 (LWP 7645)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 9 (Thread 0x7f7388fc9700 (LWP 7646)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 8 (Thread 0x7f73887c8700 (LWP 7647)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 7 (Thread 0x7f7387fc7700 (LWP 7648)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002ba4874 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WaitForWork(Eigen::EventCount::Waiter*, tensorflow::thread::EigenEnvironment::Task*) ()
#4  0x0000000002ba52a2 in Eigen::NonBlockingThreadPoolTempl<tensorflow::thread::EigenEnvironment>::WorkerLoop(int) ()
#5  0x0000000002ba4208 in std::_Function_handler<void (), tensorflow::thread::EigenEnvironment::CreateThread(std::function<void ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) ()
#6  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#7  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#8  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 6 (Thread 0x7f73b0f81700 (LWP 7655)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x000000000072a5e7 in gpr_cv_wait ()
#2  0x000000000071b1b4 in executor_thread(void*) ()
#3  0x000000000072a08d in thread_body(void*) ()
#4  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#5  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 5 (Thread 0x7f73b1782700 (LWP 7656)):
#0  0x00007f74888bfd42 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x000000000072a637 in gpr_cv_wait ()
#2  0x0000000000710165 in timer_thread(void*) ()
#3  0x000000000072a08d in thread_body(void*) ()
#4  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#5  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 4 (Thread 0x7f73b1f83700 (LWP 7659)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x0000000002d3bd64 in nsync::nsync_mu_semaphore_p_with_deadline(nsync::nsync_semaphore_s_*, timespec) ()
#4  0x0000000002d3c191 in nsync::nsync_sem_wait_with_cancel_(nsync::waiter*, timespec, nsync::nsync_note_s_*) ()
#5  0x0000000002d3aa14 in nsync::nsync_cv_wait_with_deadline_generic(nsync::nsync_cv_s_*, void*, void (*)(void*), void (*)(void*), timespec, nsync::nsync_note_s_*) ()
#6  0x0000000002d3af4d in nsync::nsync_cv_wait(nsync::nsync_cv_s_*, nsync::nsync_mu_s_*) ()
#7  0x00000000007b5fab in tensorflow::serving::BatchingSession::Run(tensorflow::RunOptions const&, std::vector<std::pair<std::string, tensorflow::Tensor>, std::allocator<std::pair<std::string, tensorflow::Tensor> > > const&, std::vector<std::string, std::allocator<std::string> > const&, std::vector<std::string, std::allocator<std::string> > const&, std::vector<tensorflow::Tensor, std::allocator<tensorflow::Tensor> >*, tensorflow::RunMetadata*) ()
#8  0x0000000000753678 in tensorflow::serving::(anonymous namespace)::SavedModelPredict(tensorflow::RunOptions const&, tensorflow::serving::ServerCore*, tensorflow::serving::PredictRequest const&, tensorflow::serving::PredictResponse*) ()
#9  0x0000000000756a66 in tensorflow::serving::TensorflowPredictor::Predict(tensorflow::RunOptions const&, tensorflow::serving::ServerCore*, tensorflow::serving::PredictRequest const&, tensorflow::serving::PredictResponse*) ()
#10 0x00000000004f0944 in (anonymous namespace)::PredictionServiceImpl::Predict(grpc::ServerContext*, tensorflow::serving::PredictRequest const*, tensorflow::serving::PredictResponse*) ()
#11 0x00000000006b3b96 in std::_Function_handler<grpc::Status (tensorflow::serving::PredictionService::Service*, grpc::ServerContext*, tensorflow::serving::PredictRequest const*, tensorflow::serving::PredictResponse*), std::_Mem_fn<grpc::Status (tensorflow::serving::PredictionService::Service::*)(grpc::ServerContext*, tensorflow::serving::PredictRequest const*, tensorflow::serving::PredictResponse*)> >::_M_invoke(std::_Any_data const&, tensorflow::serving::PredictionService::Service*, grpc::ServerContext*, tensorflow::serving::PredictRequest const*, tensorflow::serving::PredictResponse*) ()
#12 0x00000000006b8150 in grpc::internal::RpcMethodHandler<tensorflow::serving::PredictionService::Service, tensorflow::serving::PredictRequest, tensorflow::serving::PredictResponse>::RunHandler(grpc::internal::MethodHandler::HandlerParameter const&) ()
#13 0x00000000006c2839 in grpc::Server::SyncRequestThreadManager::DoWork(void*, bool) ()
#14 0x00000000006c6f07 in grpc::ThreadManager::MainWorkLoop() ()
#15 0x00000000006c7024 in grpc::ThreadManager::WorkerThread::Run() ()
#16 0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#17 0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#18 0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 3 (Thread 0x7f73b2784700 (LWP 7765)):
#0  0x00007f7487dc7183 in epoll_wait () from /lib64/libc.so.6
#1  0x0000000000724ca8 in pollset_work ()
#2  0x0000000000701aa6 in cq_next(grpc_completion_queue*, gpr_timespec, void*) ()
#3  0x00000000007022bb in grpc_completion_queue_next ()
#4  0x00000000006c7a11 in grpc::CompletionQueue::AsyncNextInternal(void**, bool*, gpr_timespec) ()
#5  0x00000000006bc999 in grpc::Server::SyncRequestThreadManager::PollForWork(void**, bool*) ()
#6  0x00000000006c6e41 in grpc::ThreadManager::MainWorkLoop() ()
#7  0x00000000006c7024 in grpc::ThreadManager::WorkerThread::Run() ()
#8  0x00007f748835e1e0 in std::(anonymous namespace)::execute_native_thread_routine (__p=<optimized out>) at ../../../.././libstdc++-v3/src/c++11/thread.cc:84
#9  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#10 0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 2 (Thread 0x7f740cff9700 (LWP 7766)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x000000000072a5e7 in gpr_cv_wait ()
#2  0x0000000000710165 in timer_thread(void*) ()
#3  0x000000000072a08d in thread_body(void*) ()
#4  0x00007f74888bbe25 in start_thread () from /lib64/libpthread.so.0
#5  0x00007f7487dc6bad in clone () from /lib64/libc.so.6
Thread 1 (Thread 0x7f74890ec8c0 (LWP 7422)):
#0  0x00007f74888bf995 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
#1  0x00007f748835a55c in __gthread_cond_wait (__mutex=<optimized out>, __cond=<optimized out>) at /tmp/gcc-4.9.2/x86_64-unknown-linux-gnu/libstdc++-v3/include/x86_64-unknown-linux-gnu/bits/gthr-default.h:864
#2  std::condition_variable::wait (this=<optimized out>, __lock=...) at ../../../.././libstdc++-v3/src/c++11/condition_variable.cc:52
#3  0x00000000006c1eaa in grpc::Server::Wait() ()
#4  0x0000000000420f45 in main ()
三、问题总结

由于hang住没有任何信息，尝试看堆栈，看源码，打日志都没有办法定位问题，感觉很无助，最开始一直怀疑是input的问题，于是尝试使用tf_serving进行加载模型预估也存在同样的问题。
最终定位还是离线在使用savedmodle模型进行保存模型方式不对，保存的网络图结构不对，没有保存正确的output图节点，导致预估找不到结果。

软件之地
关注
专栏目录
DRD：线程错误检测器
swartz_lubel的专栏
06-12 1212
目录 8.1。概观 8.1.1。多线程编程范例8.1.2。POSIX线程编程模型8.1.3。多线程编程问题8.1.4。数据竞赛检测 8.2。使用DRD 8.2.1。DRD命令行选项8.2.2。检测到的错误：数据竞赛8.2.3。检测到的错误：锁定争用8.2.4。检测到的错误：滥用POSIX线程API8.2.5。客户端请求8.2.6。调试C ++ 11程序8.2.7。调试GNOME程序8.2.
2023-03-24 duckdb-线程池与并行任务处理
技术之路
03-24 1003
duckdb对于物理计划的执行采用了并行处理, 对pipeline的处理很值得学习.本文以聚合为例进行分析。
Tensorflow在linux中运行的时候进入的一些坑及自己找到的一些办法
一名研究生工作的日常记录
03-24 7124
入坑 1：Linux 的批处理文件 .sh文件 命令的连接符问题 2：object dection 测试代码中，测试出来的图像不显示 3：在进行对Google官方代码修改的时候，如果进行对学习率 learning_rate 的改变 — 如果只是直接用tf.assign()赋值操作，则会报一下错误： `‘float’ object has no attribute ‘dtype’ `tensorf...
深入解析 TiFlash丨多并发下线程创建、释放的阻塞问题
weixin_42241611的博客
04-28 183
首先定义三种工作模式: wait、 work 、 workOnNewThreadwait: while 循环，等待work: while 循环，每次memcpy20次(每次memcpywhile循环，每次申请新的 thread，新 thread 内memcpy20次，join等待线程结束，重复这个过程。接下来按不同的工作模式组合去做实验。多线程开发中， 应尽量采用线程池、协程等手段来避免频繁的线程创建和释放。尽量在简单环境下复现问题， 以减少会对排查产生干扰的因素。
多并发下线程创建、释放的阻塞问题
TiDB 社区干货传送门
08-15 264
作者： TiDB社区小助手 原文来源： https://tidb.net/blog/63e15ce7 ...
Linux环境下编译的TensorFlow C++ API动态库文件.7z
04-16
1、TensorFlow C++ API的编译是线上使用TensorFlow C++ API调用预先训练好的模型完成预测项目的必经之路，但是TensorFlow C++ API的编译并不是这么顺利，期间遇到了很多错误，它对版本要求非常严格，版本不对应会...
C++调用tensorflow教程
08-26
C++调用TensorFlow教程 本教程旨在指导读者如何在C++中调用TensorFlow，以便在项目中更好地使用深度学习技术。由于绝大多数的Python API都是基于Python的，而C++的API不够完善，因此本教程的出现将对项目的开发产生...
Tensorflow 模型 C++部署实战2-高级部分
06-20
本课程是《Tensorflow 模型 C++部署实战》部分的续集，高级部分，主要讲述，如何进行新版本移植，加载Frozen pb,? 虚拟GPU， TF-TRT优化专题，对部署框架增加更多功能和性能优化。
tensorflow C++.rar
06-06
ubuntu 16.04环境下 搭建tensorflow C++接口所需要的所有匹配文件，包括： bazel0.20.0 + tensorflow 1.13.1 + protobuf 3.7.0 + eigen3 + abseil
如何编译TensorFlow C++ API/配置一个带TensorFlow C++ API 的Docker镜像（r1.14）
OddNo77的博客
08-14 1501
前言 对于大部分机器学习爱好者来说，TensorFlow（以下简写为TF）是一个非常好的Python开源机器学习框架。但对于一部分开发者而言，他们经常需要在Python环境下训练自己的模型，然后在C++环境下进行部署。这个部署、测试用的环境常常用Docker。 本文针对这样的需求，结合我自己在过程中踩过的坑、熬过的夜，记录一下如何搭建一个带有TensorFlow C++环境的Docker imag...
C++ 程序抛异常产生的 core 文件，无法显示正确的函数调用栈信息（备忘）
范安崇的博客
09-26 4128
问题 比如，如下程序： #include <thread> #include <string> #include <chrono> int f() { throw int(1); } int f2() { f(); } int main() { std::thread t2(f2); t2.j...
利用Wrap Shell Script定位Android Native内存泄漏
weixin_34220623的博客
02-21 998
前提条件 Android版本为8.0以上 环境配置 cd到/src/main目录下，新建shell目录，同时shell目录下配置与libs目录下相同平台的目录，如下app下的层级结构，可看到shell/lib下具有与libs下相同的平台目录结构 ── AndroidManifest.xml ├── java ├── libs │ ├── arm64-...
Android NDK——必知必会之Native线程操作及线程同步全面详解（六）
最美不过，心中有梦，身旁有你！
12-22 6025
欢迎使用Markdown编辑器写博客 本Markdown编辑器使用StackEdit修改而来，用它写博客，将会带来全新的体验哦： Markdown和扩展Markdown简洁的语法 代码块高亮 图片链接和图片上传 LaTex数学公式 UML序列图和流程图 离线写博客 导入导出Markdown文件 丰富的快捷键 快捷键 加粗 Ctrl + B 斜体 Ctrl + I...
多线程并行与退出
I_love_blog的博客
05-17 427
例1：进程与线程并行输出 #include #include #include #include //线程函数入口 void *thread_routine(void *arg) { printf("I am a thread, my thread id is : %d\n", pthread_self()); int i; for ( i = 0; i <
运行TensorFlow模型报错：Process finished with exit code 139 (interrupted by signal 11: SIGSEGV)
blingblingbing_的博客
12-03 2415
https://blingblingbing.wordpress.com/2020/12/03/tensorflow-process-finished-with-exit-code-139-interrupted-by-signal-11-sigsegv/ 模型搭建完毕，先用少量数据跑起来并没有报错，加大数据量运行，程序被强行终止，显示Process finished with exit code 139 (interrupted by signal 11: SIGSEGV) 导致该错误的原因众多：
tensorflow C++ 安装
最新发布
11-28
以下是在Linux系统下安装TensorFlow C++的步骤： 1.首先，确保已经安装了TensorFlow。如果没有安装，请按照官方文档进行安装。 2.下载TensorFlow C++的源代码。可以从GitHub上下载最新版本的源代码。 3.解压缩源代码文件，并进入解压缩后的目录。 4.创建一个名为“tensorflow”的目录，并将“cc”和“core”目录复制到该目录下。可以使用以下命令完成此操作： ```shell sudo mkdir -p /usr/local/include/tf/tensorflow sudo cp -r tensorflow/cc /usr/local/include/tf/tensorflow sudo cp -r tensorflow/core /usr/local/include/tf/tensorflow ``` 5.编译TensorFlow C++库。可以使用以下命令完成此操作： ```shell bazel build --config=opt //tensorflow:libtensorflow_cc.so ``` 6.将生成的库文件复制到/usr/local/lib目录下。可以使用以下命令完成此操作： ```shell sudo cp bazel-bin/tensorflow/libtensorflow_cc.so /usr/local/lib ``` 7.将TensorFlow C++头文件复制到/usr/local/include目录下。可以使用以下命令完成此操作： ```shell sudo cp -r tensorflow /usr/local/include/tf ``` 8.设置LD_LIBRARY_PATH环境变量，以便系统可以找到TensorFlow C++库文件。可以使用以下命令完成此操作： ```shell export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH ``` 9.现在，您可以使用TensorFlow C++库来开发自己的应用程序了。
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值