Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Duplicate #136 code and meet make errors #293

Open
lancelot-ch opened this issue May 30, 2022 · 16 comments
Open

Duplicate #136 code and meet make errors #293

lancelot-ch opened this issue May 30, 2022 · 16 comments

Comments

@lancelot-ch
Copy link

lancelot-ch commented May 30, 2022

I copied #136 codes and revised as Floop suggested that put 3 lines right below
Scope root = Scope::NewRootScope();
And I have many make errors.
If anyone could help I will be much appreciated.

I am running on the following conditions:
Freshly built Ubuntu 20.04LTS
CUDA 10.7
CuDnn 8.4.0
Bazel 5.1.1
GCC 9.4.0
make 4.2.1
tensoflow_cc build on 20-05-27, using tensorflow 2.9.0

I am using a PC with an Intel CPU, an Intel GPU and an NVidia T600.

I compiled tensorflow_cc with default CMakeLists file except for putting REQUIRE_CUDA ON, and make is successful.

Then I copied 136 codes:

#include "tensorflow/cc/client/client_session.h"
#include "tensorflow/cc/ops/standard_ops.h"
#include "tensorflow/core/framework/tensor.h"
#include "tensorflow/core/graph/default_device.h"

int main() {
using namespace tensorflow;
using namespace tensorflow::ops;
Scope root = Scope::NewRootScope();
GraphDef def;
TF_CHECK_OK(root.ToGraphDef(&def));

graph::SetDefaultDevice(false ? "/device:GPU:0" : "/cpu:0", &def);
// Matrix A = [3 2; -1 0]
auto A = Const(root, { {3.f, 2.f}, {-1.f, 0.f} });
// Vector b = [3 5]
auto b = Const(root, { {3.f, 5.f} });
// v = Ab^T
auto v = MatMul(root.WithOpName("v"), A, b, MatMul::TransposeB(true));

for (auto &node: *def.mutable_node()) {
node.set_device("/cpu:0");
std::cout << node.name() << " = '" << node.device() <<"'"<< std::endl;
}
std::cout << "=======================\n";

std::vector outputs;
ClientSession session(root);
// Run and fetch v
TF_CHECK_OK(session.Run({v}, &outputs));
// Expect outputs[0] == [19; -3]
LOG(INFO) << outputs[0].matrix();
return 0;
}

I have following make errors:

[ 50%] Building CXX object CMakeFiles/example.dir/example2.cpp.o
[100%] Linking CXX executable example
/usr/bin/ld: CMakeFiles/example.dir/example2.cpp.o: in function google::protobuf::Arena::AllocHook(std::type_info const*, unsigned long) const': example2.cpp:(.text._ZNK6google8protobuf5Arena9AllocHookEPKSt9type_infom[_ZNK6google8protobuf5Arena9AllocHookEPKSt9type_infom]+0x41): undefined reference to google::protobuf::Arena::OnArenaAllocation(std::type_info const*, unsigned long) const'
/usr/bin/ld: CMakeFiles/example.dir/example2.cpp.o: in function google::protobuf::internal::ArenaStringPtr::CreateInstance(google::protobuf::Arena*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const*)': example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x60): undefined reference to google::protobuf::internal::LogMessage::LogMessage(google::protobuf::LogLevel, char const*, int)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x78): undefined reference to google::protobuf::internal::LogMessage::operator<<(char const*)' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x8d): undefined reference to google::protobuf::internal::LogFinisher::operator=(google::protobuf::internal::LogMessage&)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x9d): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x221): undefined reference to google::protobuf::internal::ArenaImpl::AllocateAligned(unsigned long)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x240): undefined reference to google::protobuf::internal::ArenaImpl::AllocateAlignedAndAddCleanup(unsigned long, void (*)(void*))' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE[_ZN6google8protobuf8internal14ArenaStringPtr14CreateInstanceEPNS0_5ArenaEPKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE]+0x2d4): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()'
/usr/bin/ld: CMakeFiles/example.dir/example2.cpp.o: in function google::protobuf::internal::GetEmptyStringAlreadyInited[abi:cxx11]()': example2.cpp:(.text._ZN6google8protobuf8internal27GetEmptyStringAlreadyInitedB5cxx11Ev[_ZN6google8protobuf8internal27GetEmptyStringAlreadyInitedB5cxx11Ev]+0xb): undefined reference to google::protobuf::internal::fixed_address_empty_string[abi:cxx11]'
/usr/bin/ld: CMakeFiles/example.dir/example2.cpp.o: in function tensorflow::NodeDef::set_device(char const*)': example2.cpp:(.text._ZN10tensorflow7NodeDef10set_deviceEPKc[_ZN10tensorflow7NodeDef10set_deviceEPKc]+0x4b): undefined reference to google::protobuf::internal::LogMessage::LogMessage(google::protobuf::LogLevel, char const*, int)'
/usr/bin/ld: example2.cpp:(.text._ZN10tensorflow7NodeDef10set_deviceEPKc[_ZN10tensorflow7NodeDef10set_deviceEPKc]+0x63): undefined reference to google::protobuf::internal::LogMessage::operator<<(char const*)' /usr/bin/ld: example2.cpp:(.text._ZN10tensorflow7NodeDef10set_deviceEPKc[_ZN10tensorflow7NodeDef10set_deviceEPKc]+0x75): undefined reference to google::protobuf::internal::LogFinisher::operator=(google::protobuf::internal::LogMessage&)'
/usr/bin/ld: example2.cpp:(.text._ZN10tensorflow7NodeDef10set_deviceEPKc[_ZN10tensorflow7NodeDef10set_deviceEPKc]+0x85): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()' /usr/bin/ld: example2.cpp:(.text._ZN10tensorflow7NodeDef10set_deviceEPKc[_ZN10tensorflow7NodeDef10set_deviceEPKc]+0x11d): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()'
/usr/bin/ld: CMakeFiles/example.dir/example2.cpp.o: in function google::protobuf::RepeatedPtrField<tensorflow::NodeDef>::TypeHandler::Type* google::protobuf::internal::RepeatedPtrFieldBase::Mutable<google::protobuf::RepeatedPtrField<tensorflow::NodeDef>::TypeHandler>(int)': example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x49): undefined reference to google::protobuf::internal::LogMessage::LogMessage(google::protobuf::LogLevel, char const*, int)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x61): undefined reference to google::protobuf::internal::LogMessage::operator<<(char const*)' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x73): undefined reference to google::protobuf::internal::LogFinisher::operator=(google::protobuf::internal::LogMessage&)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x83): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0xb1): undefined reference to google::protobuf::internal::LogMessage::LogMessage(google::protobuf::LogLevel, char const*, int)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0xc9): undefined reference to google::protobuf::internal::LogMessage::operator<<(char const*)' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0xdb): undefined reference to google::protobuf::internal::LogFinisher::operator=(google::protobuf::internal::LogMessage&)'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0xeb): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()' /usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x12e): undefined reference to google::protobuf::internal::LogMessage::~LogMessage()'
/usr/bin/ld: example2.cpp:(.text._ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi[_ZN6google8protobuf8internal20RepeatedPtrFieldBase7MutableINS0_16RepeatedPtrFieldIN10tensorflow7NodeDefEE11TypeHandlerEEEPNT_4TypeEi]+0x150): undefined reference to `google::protobuf::internal::LogMessage::~LogMessage()'
collect2: error: ld returned 1 exit status
make[2]: *** [CMakeFiles/example.dir/build.make:87: example] Error 1
make[1]: *** [CMakeFiles/Makefile2:76: CMakeFiles/example.dir/all] Error 2

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 2, 2022

Can you try to link Protobuf to your program?

@lancelot-ch
Copy link
Author

lancelot-ch commented Jun 3, 2022

Thanks for the reply. Sorry that I don't know exactly how to link Protobuf.

I installed Protobuf from source and the folder is at ~/protobuf.3.9.0

The project CMakeLists.txt file is as below:

CMakeLists.txt

cmake_minimum_required(VERSION 3.16)
set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CXX_STANDARD_REQUIRED True)
find_package(TensorflowCC REQUIRED)
add_executable(test test.cpp)
target_include_directories(test PUBLIC
"${PROJECT_BINARY_DIR}"
)
add_subdirectory(abseil-cpp)
target_link_libraries(test TensorflowCC::TensorflowCC absl::strings)

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 3, 2022

See e.g., https://cmake.org/cmake/help/latest/module/FindProtobuf.html

find_package(Protobuf REQUIRED)
target_link_libraries(foo ${Protobuf_LIBRARIES})
target_include_directories(foo ${Protobuf_INCLUDE_DIRS})

@lancelot-ch
Copy link
Author

Thanks.

I tried include these 3 lines. And I get the following error:

$ cmake ..
CMake Error at CMakeLists.txt:20 (target_include_directories):
target_include_directories called with invalid arguments

And line 20 is :
target_include_directories(test ${Protobuf_INCLUDE_DIRS})

I am not sure what is wrong.

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 3, 2022

Please print the ${Protobuf_INCLUDE_DIRS} variable using message() call. Maybe there are some spaces and you have to wrap it in ".

@lancelot-ch
Copy link
Author

lancelot-ch commented Jun 3, 2022

Thanks.

I do see something weird.

cmake ..
/home/user/anaconda3/lib/libprotobuf.so-lpthread
/home/user/anaconda3/include
CMake Error at CMakeLists.txt:22 (target_include_directories):
target_include_directories called with invalid arguments

-- Configuring incomplete, errors occurred!

I recently installed Anaconda and I am wondering if that changed the path. Shall I just uninstall Anaconda and try?

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 3, 2022

You can try or you can try to change the PATH, but protobuf from anaconda is probably not the one you want. Anyway, the error you are seeing is most likely caused by something else, just add ${Protobuf_INCLUDE_DIRS} and ${Protobuf_LIBRARIES} to the target_include_directories and target_link_libraries directives that you already had.

@lancelot-ch
Copy link
Author

Thanks. Could I verify that I should install protobuf by myself(hopefully 3.9.0), and compile tensorflow_cc with

option(INSTALL_PROTOBUF "Install protobuf compatible with tensorflow version." OFF)

I am not sure if it is correct.

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 6, 2022 via email

@lancelot-ch
Copy link
Author

I tried but I failed.

I do with this way:

I revise tensorflow_cc/tensorflow_cc/CMakeLists.txt

option(INSTALL_PROTOBUF "Install protobuf compatible with tensorflow version." ON) #previously it is OFF

I run cmake:

~/tensorflow_cc/tensorflow_cc/build$ cmake -DLOCAL_RAM_RESOURCES=2048 -DLOCAL_CPU_RESOURCES=4 ..

I see below error messages:

-- The C compiler identification is GNU 9.4.0
-- The CXX compiler identification is GNU 9.4.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/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: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
Will build Protobuf from ''.
CMake Error at /usr/share/cmake-3.16/Modules/ExternalProject.cmake:2630 (message):
No download info given for 'protobuf-external' and its source directory:

/home/user/tensorflow_cc/tensorflow_cc/build/protobuf/src/protobuf-external

is not an existing non-empty directory. Please specify one of:

  • SOURCE_DIR with an existing non-empty directory
  • DOWNLOAD_COMMAND
  • URL
  • GIT_REPOSITORY
  • SVN_REPOSITORY
  • HG_REPOSITORY
  • CVS_REPOSITORY and CVS_MODULE
    Call Stack (most recent call first):
    /usr/share/cmake-3.16/Modules/ExternalProject.cmake:3236 (_ep_add_download_command)
    cmake/ProtobufExternal.cmake:17 (ExternalProject_Add)
    CMakeLists.txt:36 (include)

Seems that it suggests that I get an external Protobuf?

Many thanks.

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 8, 2022 via email

@lancelot-ch
Copy link
Author

I removed build folder and tried again. I got the same error.

Will build Protobuf from ''.
CMake Error at /usr/share/cmake-3.16/Modules/ExternalProject.cmake:2630 (message):
No download info given for 'protobuf-external' and its source directory:

/home/user/tensorflow_cc/tensorflow_cc/build/protobuf/src/protobuf-external

is not an existing non-empty directory. Please specify one of:

  • SOURCE_DIR with an existing non-empty directory
  • DOWNLOAD_COMMAND
  • URL
  • GIT_REPOSITORY
  • SVN_REPOSITORY
  • HG_REPOSITORY
  • CVS_REPOSITORY and CVS_MODULE
    Call Stack (most recent call first):
    /usr/share/cmake-3.16/Modules/ExternalProject.cmake:3236 (_ep_add_download_command)
    cmake/ProtobufExternal.cmake:17 (ExternalProject_Add)
    CMakeLists.txt:36 (include)

-- Configuring incomplete, errors occurred!
See also "/home/user/tensorflow_cc/tensorflow_cc/build/CMakeFiles/CMakeOutput.log".

@lancelot-ch
Copy link
Author

lancelot-ch commented Jun 9, 2022

Many thanks to FloopCZ that with Docker I can make the program. The results is as following:

root@9c79a0ff0c79:/home/test/build# ./test
2022-06-09 01:18:04.209457: I tensorflow/core/util/util.cc:169] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable TF_ENABLE_ONEDNN_OPTS=0.

2022-06-09 01:18:04.218837: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX512F AVX512_VNNI
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2022-06-09 01:18:05.116681: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.139072: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.139187: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.455365: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.455501: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.455588: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:975] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2022-06-09 01:18:05.455691: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1532] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 2636 MB memory: -> device: 0, name: NVIDIA T600 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 7.5
2022-06-09 01:18:05.456816: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:354] MLIR V1 optimization pass is not enabled

However, I realized that even if I change the device from GPU to CPU (graph::SetDefaultDevice("/cpu:0", &def);
), the result is always using GPU device. I do not quite understand.

And also I hope to get the source code installing done if possible. Thanks a lot.

@FloopCZ
Copy link
Owner

FloopCZ commented Jun 25, 2022

Hi, try to run

export CUDA_VISIBLE_DEVICES=""

before running your program to disable the GPU

@lancelot-ch
Copy link
Author

Hi, try to run

export CUDA_VISIBLE_DEVICES=""

before running your program to disable the GPU

Actually I am meant to use GPU to accelerate program running by using GPU. I have been successful to run program using tensorflow_cc on CPU. Anyway, I will keep trying.

@FloopCZ
Copy link
Owner

FloopCZ commented Jul 16, 2022

ok, I think I misunderstood your previous message:

However, I realized that even if I change the device from GPU to CPU (graph::SetDefaultDevice("/cpu:0", &def);
), the result is always using GPU device. I do not quite understand.

Is there anything else I can help with or should we close the issue?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants