|
|
||
|---|---|---|
| .. | ||
| cmake | ||
| ie_bridges/python | ||
| include | ||
| samples | ||
| src | ||
| tests | ||
| thirdparty | ||
| CMakeLists.txt | ||
| README.md | ||
| install_dependencies.sh | ||
README.md
Build on Linux* Systems
The software was validated on:
- Ubuntu* 16.04 with default GCC* 5.4.0
- CentOS* 7.4 with default GCC* 4.8.5 (using clDNN library built separately with GCC* 5.2)
- Intel® Graphics Compute Runtime for OpenCL™ Driver package 18.28.11080.
Software Requirements
- CMake* 3.9 or higher
- GCC* 4.8 or higher to build the Inference Engine
- GCC* 5.2 or higher to build the Compute Library for Deep Neural Networks (clDNN library)
- OpenBLAS*
Build Steps
- Install OpenBLAS and other dependencies using the
install_dependencies.shscript in the project root folder. - Create a build folder:
mkdir build
- Inference Engine uses a CMake-based build system. In the created
builddirectory, runcmaketo fetch project dependencies and create Unix makefiles, then runmaketo build the project:
cmake -DCMAKE_BUILD_TYPE=Release ..
make -j16
You can use the following additional build options:
- Use
BLAS_INCLUDE_DIRSandBLAS_LIBRARIEScmake options to specify path to OpenBLAS headers and library, for example use the following options on CentOS*:-DBLAS_INCLUDE_DIRS=/usr/include/openblas -DBLAS_LIBRARIES=/usr/lib64/libopenblas.so.0 - To build clDNN from sources, please specify the
-DENABLE_CLDNN_BUILD=ONoption forcmake. By default pre-built version of the clDNN library is used. - To switch on/off the CPU and GPU plugins, use
cmakeoptions-DENABLE_MKL_DNN=ON/OFFand-DENABLE_CLDNN=ON/OFF.
Build on Windows* Systems:
The software was validated on:
- Microsoft* Windows* 10 with Visual Studio 2017 and Intel® C++ Compiler 2018 Update 3
- Intel® Graphics Driver for Windows* [24.20] driver package.
Software Requirements
- CMake* 3.9 or higher
- OpenBLAS* or
- Intel® C++ Compiler 18.0 to build the Inference Engine on Windows.
Build Steps
- Download and install Intel® C++ Compiler 18.0
- Install OpenBLAS:
- Download OpenBLAS*
- Unzip the downloaded package to a directory on your machine. In this document, this directory is referred to as
<OPENBLAS_DIR>.
- Create build directory:
mkdir build - In the
builddirectory, runcmaketo fetch project dependencies and generate a Visual Studio solution:
cd build
cmake -G "Visual Studio 15 2017 Win64" -T "Intel C++ Compiler 18.0" -DOS_FOLDER=ON ^
-DBLAS_INCLUDE_DIRS=<OPENBLAS_DIR>\include ^
-DBLAS_LIBRARIES=<OPENBLAS_DIR>\lib\libopenblas.dll.a ^
-DCMAKE_BUILD_TYPE=Release ^
-DICCLIB="C:\Program Files (x86)\IntelSWTools\compilers_and_libraries_2018\windows\compiler\lib" ..
- Build generated solution in Visual Studio 2017 or run
cmake --build .to build from the command line.
* Other names and brands may be claimed as the property of others.