This configuration step must be run again before building. Links to your system's CUDA libraries-so if you update your CUDA library paths, Version instead of relying on the default./configure.py creates symbolic System has multiple versions of CUDA or cuDNN installed, explicitly set the Would you like to override eigen strong inline for some C++ compilation to reduce the compilation time? :įor GPU support, specify the versions of CUDA and cuDNN. Please specify optimization flags to use during compilation when bazel option "-config=opt" is specified : Please note that each additional compute capability significantly increases your build time and binary size. You can find the compute capability of your device at: Please specify a list of comma-separated Cuda compute capabilities you want to build with. Please specify the location where cuDNN 7 library is installed. Please specify the cuDNN version you want to use. Please specify the location where CUDA 9.0 toolkit is installed. Please specify the CUDA SDK version you want to use. Default is ĭo you wish to build TensorFlow with CUDA support? : YĬUDA support will be enabled for TensorFlow. Please input the desired Python library path to use. Starting local Bazel server and connecting to it. configure.py (your session mayĭiffer): View sample configuration session Theįollowing shows a sample run of python. This script prompts you for the location of TensorFlow dependencies and asks forĪdditional build configuration options (compiler flags, for example). Configure the buildĬonfigure your system build by running the following at the root of your Key Point: If you're having build problems on the latest development branch, tryĪ release branch that is known to work. Git checkout branch_name # r1.9, r1.10, etc. The repo defaults to the master development branch. ( git is installed with MSYS2): git clone cd tensorflow See the Windows GPU support guide to install the drivers andĪdditional software required to run TensorFlow on a GPU. Note: TensorFlow is tested against the Visual Studio 2019. Microsoft Visual C++ 2019 Redistributable.Select Redistributables and Build Tools,.If MSYS2 is installed to C:\msys64, addĬ:\msys64\usr\bin to your %PATH% environment variable. Install MSYS2 for the bin tools needed toīuild TensorFlow. Install Bazel, the build tool used to compileĪdd the location of the Bazel executable to your %PATH% environment variable. Install the TensorFlow pip package dependencies: pip3 install six numpy wheel pip3 install keras_applications=1.0.6 -no-deps pip3 install keras_preprocessing=1.0.5 -no-depsįile under REQUIRED_PACKAGES. Select pip as an optional feature and add it to your %PATH% environmental Install Python and the TensorFlow package dependencies Install the following build tools to configure your Windows developmentĮnvironment. Note: We already provide well-tested, pre-built Build a TensorFlow pip package from source and install it on Windows.
0 kommentar(er)
