Building TensorFlow

build_tensorflow.txt

# Install Chocolatey
# Install CUDA Toolkit 10.0
# Install cuDNN
# Install Visual Studio Build Tools (C++)
# Include VC++ 2015.3 v14.00 (v140) toolset for desktop

# Install Miniconda
cinst miniconda3 --params "/AddToPath:1" -y

# Create Python environment
conda create -n tensorflow_build python=3.6 --yes
conda activate tensorflow_build

# Install packages
conda install six numpy wheel --yes
pip install keras_applications==1.0.6 --no-deps
pip install keras_preprocessing==1.0.5 --no-deps

# Install Bazel (installs MSYS2)
cinst bazel --version 0.19.0 -y

# In MSYS2 bash
# Synchronize packages,
# download a fresh copy of the master package database from the server
# upgrades all packages that are out-of-date
pacman -Syu

# Install patch and unzip
pacman -S patch unzip

# Install Git
cinst git --params "/SChannel" -y

# Download source
git clone https://github.com/tensorflow/tensorflow.git
cd tensorflow
git checkout v1.12.0

# Patch eigen
third_party:
Add eigen_half.patch file

tensorflow/workspace.bzl:
    ...
    tf_http_archive(
        name = "eigen_archive",
        build_file = clean_dep("//third_party:eigen.BUILD"),
        sha256 = "d956415d784fa4e42b6a2a45c32556d6aec9d0a3d8ef48baee2522ab762556a9",
        strip_prefix = "eigen-eigen-fd6845384b86",
        urls = [
            "https://mirror.bazel.build/bitbucket.org/eigen/eigen/get/fd6845384b86.tar.gz",
            "https://bitbucket.org/eigen/eigen/get/fd6845384b86.tar.gz",
        ],
        patch_file = clean_dep("//third_party:eigen_half.patch"),  # add this line
    )
    ...

# Configure
python configure.py

You have bazel 0.19.0 installed.
Please specify the location of python. [Default is C:\tools\miniconda3\envs\tensorflow_build\python.exe]:
Please input the desired Python library path to use.  Default is [C:\tools\miniconda3\envs\tensorflow_build\lib\site-packages]
Do you wish to build TensorFlow with Apache Ignite support? [Y/n]:
Do you wish to build TensorFlow with XLA JIT support? [y/N]:
Do you wish to build TensorFlow with ROCm support? [y/N]:
Do you wish to build TensorFlow with CUDA support? [y/N]: y
Please specify the CUDA SDK version you want to use. [Leave empty to default to CUDA 9.0]: 10.0
Please specify the location where CUDA 10.0 toolkit is installed. Refer to README.md for more details. [Default is C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.0]:
Please specify the cuDNN version you want to use. [Leave empty to default to cuDNN 7]:
Please specify the location where cuDNN 7 library is installed. Refer to README.md for more details. [Default is C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v10.0]:
Please note that each additional compute capability significantly increases your build time and binary size. [Default is: 3.5,7.0]: 6.1
Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is /arch:AVX]: /arch:AVX /arch:AVX2
Would you like to override eigen strong inline for some C++ compilation to reduce the compilation time? [Y/n]:

# Fix Bazel bug
.bazelrc:
import d:/code/python/tensorflow/tools/bazel.rc  # add this line
import d:/code/python/tensorflow/.tf_configure.bazelrc

# Set environment variables
set BAZEL_SH=C:\tools\msys64\usr\bin\bash.exe
set BAZEL_VC=C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC

# Build
bazel build -c opt --config=opt --verbose_failures --define=no_tensorflow_py_deps=true //tensorflow/tools/pip_package:build_pip_package
bazel-bin\tensorflow\tools\pip_package\build_pip_package tmp\tensorflow_pkg

# Install package
pip install tmp\tensorflow_pkg\tensorflow-1.12.0-cp36-cp36m-win_amd64.whl