Deepo2 is now a series ofDocker images that

• allows you to quickly set up your deep learning research environment
• supports almost all commonly used deep learning frameworks:theano,tensorflow,sonnet,pytorch,keras,lasagne,mxnet,cntk,chainer,caffe,torch

and their Dockerfile generator that

• allows you to customize your own environment with Lego-like modules
• automatically resolves the dependencies for you

Table of contents

• Quick Start
  • Installation
  • Usage
• Customization
  • I hate all-in-one solution
  • Other python versions
  • Build your own customized image
• Comparison to Alternatives
• Contributing
• Licensing

Step 1. InstallDocker andnvidia-docker.

Step 2. Obtain the all-in-one image fromDocker Hub

docker pull ufoym/deepo

nvidia-docker run --rm ufoym/deepo nvidia-smi

This should work and enables Deepo to use the GPU from inside a docker container.
If this does not work, searchthe issues section on the nvidia-docker GitHub -- many solutions are already documented. To get an interactive shell to a container that will not be automatically deleted after you exit do

nvidia-docker run -it ufoym/deepo bash

If you want to share your data and configurations between the host (your machine or VM) and the container in which you are using Deepo, use the -v option, e.g.

nvidia-docker run -it -v /host/data:/data -v /host/config:/config ufoym/deepo bash

This will make/host/data from the host visible as/data in the container, and/host/config as/config. Such isolation reduces the chances of your containerized experiments overwriting or using wrong data.

Please note that some frameworks (e.g. PyTorch) use shared memory to share data between processes, so if multiprocessing is used the default shared memory segment size that container runs with is not enough, and you should increase shared memory size either with--ipc=host or--shm-size command line options tonvidia-docker run.

nvidia-docker run -it --ipc=host ufoym/deepo bash

You are now ready to begin your journey.

$ python

>>>importtensorflow>>>importsonnet>>>importtorch>>>importkeras>>>importmxnet>>>importcntk>>>importchainer>>>importtheano>>>importlasagne>>>importcaffe

$ caffe --version

caffe version 1.0.0

$ th

 │  ______             __   |  Torch7 │ /_  __/__  ________/ /   |  Scientific computing for Lua. │  / / / _ \/ __/ __/ _ \  |  Type ? for help │ /_/  \___/_/  \__/_//_/  |  https://github.com/torch │                          |  http://torch.ch │ │th>

Note thatdocker pull ufoym/deepo mentioned inQuick Start will give you a standard image containing all available deep learning frameworks. You can customize your own environment as well.

docker pull ufoym/deepo:tensorflow

docker pull ufoym/deepo:py27

docker pull ufoym/deepo:tensorflow-py27

Seehttps://hub.docker.com/r/ufoym/deepo/tags/ for a complete list of all available tags. These pre-built images are all built fromdocker/Dockerfile.* andcircle.yml. SeeHow to generatedocker/Dockerfile.* andcircle.yml if you are interested in how these files are generated.

git clone https://github.com/ufoym/deepo.gitcd deepo/generatorpip install -r requirements.txt

python generate.py Dockerfile pytorch lasagne

python generate.py Dockerfile pytorch lasagne python==3.6

docker build -t my/deepo.

Deepo is aDocker image with a full reproducible deep learning research environment. It contains most popular deep learning frameworks:theano,tensorflow,sonnet,pytorch,keras,
lasagne,mxnet,cntk,chainer,caffe,torch.

• Quick Start
  • Installation
  • Usage
• Comparison to Alternatives
• Licensing

Step 1. InstallDocker andnvidia-docker.

Step 2. Obtain the Deepo image

You can either directly download the image from Docker Hub, or build the image yourself.

Option 1: Get the image from Docker Hub (recommended)

docker pull ufoym/deepo

Option 2: Build the Docker image locally

git clone https://github.com/ufoym/deepo.gitcd deepo&& docker build -t ufoym/deepo.

Note that this may take several hours as it compiles a few libraries from scratch.

nvidia-docker run --rm ufoym/deepo nvidia-smi

This should work and enables Deepo to use the GPU from inside a docker container.
If this does not work, searchthe issues section on the nvidia-docker GitHub -- many solutions are already documented. To get an interactive shell to a container that will not be automatically deleted after you exit do

nvidia-docker run -it ufoym/deepo bash

If you want to share your data and configurations between the host (your machine or VM) and the container in which you are using Deepo, use the -v option, e.g.

nvidia-docker run -it -v /host/data:/data -v /host/config:/config ufoym/deepo bash

This will make/host/data from the host visible as/data in the container, and/host/config as/config. Such isolation reduces the chances of your containerized experiments overwriting or using wrong data.

You are now ready to begin your journey.

tensorflow

$ python

>>>importtensorflow>>>print(tensorflow.__name__,tensorflow.__version__)tensorflow1.3.0

sonnet

$ python

>>>importsonnet>>>print(sonnet.__name__,sonnet.__path__)sonnet ['/usr/local/lib/python3.5/dist-packages/sonnet']

pytorch

$ python

>>>importtorch>>>print(torch.__name__,torch.__version__)torch0.2.0_3

keras

$ python

>>>importkeras>>>print(keras.__name__,keras.__version__)keras2.0.8

mxnet

$ python

>>>importmxnet>>>print(mxnet.__name__,mxnet.__version__)mxnet0.11.0

cntk

$ python

>>>importcntk>>>print(cntk.__name__,cntk.__version__)cntk2.2

chainer

$ python

>>>importchainer>>>print(chainer.__name__,chainer.__version__)chainer3.0.0

theano

$ python

>>>importtheano>>>print(theano.__name__,theano.__version__)theano0.10.0beta4+14.gb6e3768

lasagne

$ python

>>>importlasagne>>>print(lasagne.__name__,lasagne.__version__)lasagne0.2.dev1

caffe

$ python

>>>importcaffe>>>print(caffe.__name__,caffe.__version__)caffe1.0.0

$ caffe --version

caffe version 1.0.0

torch

$ th

 │  ______             __   |  Torch7 │ /_  __/__  ________/ /   |  Scientific computing for Lua. │  / / / _ \/ __/ __/ _ \  |  Type ? for help │ /_/  \___/_/  \__/_//_/  |  https://github.com/torch │                          |  http://torch.ch │ │th>