VS2017编译GPU版Tensorflow(1.5.0 C++)

利用 Anaconda 创建虚拟环境（python3.6暂时不支持）

conda create -n tensorflow pip python=3.5
activate tensorflow
conda install numpy

用VS2017的 x64 Native Tools Command Prompt for VS 2017 下载tensorflow源码

（如果下载失败， git config --global http.postBuffer 5242880000）

git clone https://github.com/tensorflow/tensorflow.git v1.5.0
cd v1.5.0
git checkout tags/v1.5.0

创建目录

cd tensorflow\contrib\cmake
mkdir build
cd build

利用cmake生成项目

cmake .. -A x64 ^
-DCMAKE_BUILD_TYPE=Release ^
-DSWIG_EXECUTABLE=C:\Users\Zhumingde\Desktop\swigwin-3.0.12\swig.exe ^
-DPYTHON_EXECUTABLE=C:\Users\Zhumingde\AppData\Local\conda\conda\envs\tensorflow\python.exe ^
-DPYTHON_LIBRARIES=C:\Users\Zhumingde\AppData\Local\conda\conda\envs\tensorflow\libs\python35.lib ^
-Dtensorflow_ENABLE_GPU=ON ^
-DCUDNN_HOME="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0" ^
-Dtensorflow_BUILD_PYTHON_BINDINGS=OFF ^
-Dtensorflow_ENABLE_GRPC_SUPPORT=OFF ^
-Dtensorflow_BUILD_SHARED_LIB=ON ^
-DCUDA_HOST_COMPILER="C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC\bin/amd64/cl.exe"

1.5.0里的一行代码会让VS2017报错，将tensorflow\contrib\boosted_trees\lib\utils\sparse_column_iterable.cc里的

reference operator*() { return iter_->ix()(row_idx_, 0); }

替换为

reference operator*() const { return iter_->ix()(row_idx_, 0); }

然后就是愉快地编译了（或许要来两次……）

“C:\Program Files (x86)\Microsoft Visual Studio\2017\Professional\MSBuild\15.0\Bin\amd64\MSBuild.exe” ^
/m:1 ^
/p:CL_MPCount=1 ^
/p:Configuration=Release ^
/p:Platform=x64 ^
/p:PreferredToolArchitecture=x64 ALL_BUILD.vcxproj ^
/filelogger

安安静静干了三个小时别的事情，终于编完了

在生成的Release目录下可以看到tensorflow.lib以及tensorflow.dll，大功告成！

========================环境配置=====================

包含目录：

库目录

链接器-->输入加入tensorflow.lib

将tensorflow.dll从编译生成的release文件夹下拷贝到项目EXE同目录

测试代码：


#include <vector>
#include <eigen/Dense>
#include "tensorflow/core/public/session.h"
#include "tensorflow/cc/ops/standard_ops.h"
using namespace tensorflow;


GraphDef CreateGraphDef()
{
	Scope root = Scope::NewRootScope();

	auto X = ops::Placeholder(root.WithOpName("x"), DT_FLOAT,
		ops::Placeholder::Shape({ -1, 2 }));
	auto A = ops::Const(root, { { 3.f, 2.f },{ -1.f, 0.f } });

	auto Y = ops::MatMul(root.WithOpName("y"), A, X,
		ops::MatMul::TransposeB(true));

	GraphDef def;
	TF_CHECK_OK(root.ToGraphDef(&def));

	return def;
}

int main()
{
	GraphDef graph_def = CreateGraphDef();

	// Start up the session
	SessionOptions options;
	std::unique_ptr<Session> session(NewSession(options));
	TF_CHECK_OK(session->Create(graph_def));

	// Define some data.  This needs to be converted to an Eigen Tensor to be
	// fed into the placeholder.  Note that this will be broken up into two
	// separate vectors of length 2: [1, 2] and [3, 4], which will separately
	// be multiplied by the matrix.
	std::vector<float> data = { 1, 2, 3, 4 };
	auto mapped_X_ = Eigen::TensorMap<Eigen::Tensor<float, 2, Eigen::RowMajor>>
		(&data[0], 2, 2);
	auto eigen_X_ = Eigen::Tensor<float, 2, Eigen::RowMajor>(mapped_X_);

	Tensor X_(DT_FLOAT, TensorShape({ 2, 2 }));
	X_.tensor<float, 2>() = eigen_X_;

	std::vector<Tensor> outputs;
	TF_CHECK_OK(session->Run({ { "x", X_ } }, { "y" }, {}, &outputs));

	// Get the result and print it out
	Tensor Y_ = outputs[0];
	std::cout << Y_.tensor<float, 2>() << std::endl;

	session->Close();
	system("pause");
}

运行结果：

VS2017编译GPU版Tensorflow(1.5.0 C++)

========================环境配置=====================

猜你喜欢