xychelsea / tensorflow-docker

Docker Containers for TensorFlow in Anaconda 3

TensorFlow in Anaconda

TensorFlow is an open source platform for machine learning. It provides tools, libraries and community resources for researcher and developers to build and deploy machine learning applications. Anaconda is an open data science platform based on Python 3. This container installs TensorFlow through the conda command with Anaconda in the /usr/local/anaconda directory. The default user, anaconda runs a Tini shell /usr/bin/tini, and comes preloaded with the conda command in the environment $PATH. Additional versions with NVIDIA/CUDA support and Jupyter Notebooks tags are available.

NVIDIA/CUDA GPU-enabled Containers

Two flavors provide an NVIDIA GPU-enabled container with TensorFlow pre-installed through Anaconda.

Getting the containers

Vanilla TensorFlow

The base container, based on the xychelsea/anaconda3:latest Anaconda 3 container stack (xychelsea/anaconda3:latest) running Tini shell. For the container with a /usr/bin/tini entry point, use:

docker pull xychelsea/tensorflow:latest

With Jupyter Notebooks server pre-installed, pull with:

docker pull xychelsea/tensorflow:latest-jupyter

TensorFlow with NVIDIA/CUDA GPU support

Modified versions of nvidia/cuda:latest container, with support for NVIDIA/CUDA graphical processing units through the Tini shell. For the container with a /usr/bin/tini entry point:

docker pull xychelsea/tensorflow:latest-gpu

With Jupyter Notebooks server pre-installed, pull with:

docker pull xychelsea/tensorflow:latest-gpu-jupyter

Running the containers

Vanilla TensorFlow

docker run --rm -it xychelsea/tensorflow:latest

With Jupyter Notebooks server pre-installed, run with:

docker run --rm -it -d -p 8888:8888 xychelsea/tensorflow:latest-jupyter

TensorFlow with NVIDIA/CUDA GPU support

docker run --gpus all --rm -it xychelsea/tensorflow:latest-gpu /bin/bash

With Jupyter Notebooks server pre-installed, run with:

docker run --gpus all --rm -it -d -p 8888:8888 xychelsea/tensorflow:latest-gpu-jupyter

Building the containers

To build either a GPU-enabled container or without GPUs, use the tensorflow-docker GitHub repository.

git clone git://github.com/xychelsea/tensorflow-docker.git

Vanilla TensorFlow

The base container, based on the xychelsea/anaconda3:latest Anaconda 3 container stack (xychelsea/anaconda3:latest) running Tini shell:

docker build -t xychelsea/tensorflow:latest -f Dockerfile .

With Jupyter Notebooks server pre-installed, build with:

docker build -t xychelsea/tensorflow:latest-jupyter -f Dockerfile.jupyter .

TensorFlow with NVIDIA/CUDA GPU support

docker build -t xychelsea/tensorflow3:latest-gpu -f Dockerfile.nvidia .

With Jupyter Notebooks server pre-installed, build with:

docker build -t xychelsea/tensorflow:latest-gpu-jupyter -f Dockerfile.nvidia-jupyter .

References

TensorFlow

NVIDIA CUDA container

Anaconda 3