Docker images to compile TensorFlow yourself.
Tensorflow only provide a limited set of build and it can be challenging to compile yourself on certain configuration. With this Dockerfile
, you should be able to compile TensorFlow on any Linux platform that run Docker.
Compilation images are provided for Ubuntu 18.10, Ubuntu 16.04, CentOS 7.4 and CentOS 6.6.
docker
docker-compose
- Clone this repository:
git clone https://github.com/hadim/docker-tensorflow-builder.git
- Edit the
build.sh
file to modify TensorFlow compilation parameters. Then launch the build:
LINUX_DISTRO="ubuntu-16.04"
# or LINUX_DISTRO="ubuntu-18.10"
# or LINUX_DISTRO="centos-7.4"
# or LINUX_DISTRO="centos-6.6"
cd "tensorflow/$LINUX_DISTRO"
# Set env variables
export PYTHON_VERSION=3.6
export TF_VERSION_GIT_TAG=v1.13.1
export BAZEL_VERSION=0.19
export USE_GPU=0
# Build the Docker image
docker-compose build
# Start the compilation
docker-compose run tf
# You can also do:
# docker-compose run tf bash
# bash build.sh
- Edit the
build.sh
file to modify TensorFlow compilation parameters. Then launch the build:
LINUX_DISTRO="ubuntu-16.04"
# or LINUX_DISTRO="ubuntu-18.10"
# or LINUX_DISTRO="centos-7.4"
# or LINUX_DISTRO="centos-6.6"
cd "tensorflow/$LINUX_DISTRO"
# Set env variables
export PYTHON_VERSION=3.6
export TF_VERSION_GIT_TAG=v1.13.1
export BAZEL_VERSION=0.19
export USE_GPU=1
export CUDA_VERSION=10.0
export CUDNN_VERSION=7.5
export NCCL_VERSION=2.4
# Build the Docker image
docker-compose build
# Start the compilation
docker-compose run tf
# You can also do:
# docker-compose run tf bash
# bash build.sh
- Refer to tested build configurations to know which
BAZEL_VERSION
you need. - Be patient, the compilation can be long.
- Enjoy your Python wheels in the
wheels/
folder. - Don't forget to remove the container to free the space after the build:
docker-compose rm --force
.
Tensorflow | Python | Distribution | Bazel | CUDA | cuDNN | NCCL | Comment |
---|---|---|---|---|---|---|---|
v2.0.0-alpha0 | 3.6 | Ubuntu 18.10 | 0.20 | 10.0 | 7.5 | 2.4 | seg fault error |
v2.0.0-alpha0 | 3.6 | Ubuntu 18.10 | 0.20 | - | - | - | OK |
v2.0.0-alpha0 | 3.6 | Ubuntu 16.04 | 0.20 | 10.0 | 7.5 | 2.4 | TODO |
v2.0.0-alpha0 | 3.6 | Ubuntu 16.04 | 0.20 | - | - | - | TODO |
1.9.0 | 3.6 | Ubuntu 16.04 | - | - | 0.19 | - | OK |
1.9.0 | 3.6 | Ubuntu 16.04 | 9.0 | 0.19 | 7.1 | - | OK |
1.9.0 | 3.6 | Ubuntu 16.04 | 9.1 | 0.19 | 7.1 | - | OK |
1.9.0 | 3.6 | Ubuntu 16.04 | 9.2 | 0.19 | 7.1 | - | OK |
1.9.0 | 3.6 | CentOS 6.6 | - | - | 0.19 | - | OK |
1.9.0 | 3.6 | CentOS 6.6 | 9.0 | 0.19 | 7.1 | - | OK |
1.9.0 | 3.6 | CentOS 6.6 | 9.1 | 0.19 | 7.1 | - | OK |
1.9.0 | 3.6 | CentOS 6.6 | 9.2 | 0.19 | 7.1 | - | OK |
- Hadrien Mary hadrien.mary@gmail.com
- Dimitri Gerin dimitri.gerin@gmail.com
- Clone this repository:
git clone https://github.com/hadim/docker-tensorflow-builder.git
- Clone tensorflow project inside docker-tensorflow-builder (example with https://github.com/Dimitri1/tensorflow-quant)
cd docker-tensorflow-builder
git clone https://github.com/Dimitri1/tensorflow-quant tensorflow
- build
bash build.sh
MIT License. See LICENSE.